at"

anumw...

.mm; m. .

u. is
“3.“...

Lam».

 

 

.9an ..
A ‘. .2.

 

 

 

”mmmwawflfi .awE.

has
V:

c: «all! .

$1.-

‘ I. 1!
£349.“. a...
Jul-.9

 

'1v

 

 

g) .‘.t'

F‘-

"i'rvln:

~ld.|'

. :l? 4:...1}

      

.l..a.I.
.3 .

{IV 1 . .1153
n

V, .
. 1...“.u‘ .2 . .aa..-nf»v WW2 .4. ‘ .aw, w #340...
:«wTaaqafiuuamfi ‘. ‘ ram—33. \3

. I...

 

.. ‘ a?

. :3

 

.z.

 

 

.T.‘

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5108 K:IProlecc&Pres/CIRCIDateDue.indd

 

Role of the plastid envelope membrane in
integrating the plastid into cellular metabolic
networks

By

Andrea Brautigam

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Genetics

2008

Abstract

Role of the plastid envelope membrane in integrating the plastid into cellular metabolic
networks

by Andrea Brautigam

Plastids are the defining organelle of the Archaeplastida which include all land
plants. They can differentiate into several subtypes in a tissue dependent manner, for
example, brightly colored chromoplasts in bell pepper fruits, green chloroplasts in leaves,
or colorless proplastids in meristems, and each subtype is adapted to the cell it resides in.
All plastid types of land plants are separated from the cytosol by two membranes, the
inner envelope and the outer envelope membrane, and metabolites and signals have to
cross these envelope membranes to connect to the remainder of the cell. This work is
focused on integrating the plastids into the metabolic network of their respective cells;
the adaptations of envelope membranes of different plastid subtypes were analyzed by
comparative proteomics.

Initially, envelope proteomics of plants without a sequenced genome or
transcriptome was established with the garden pea Pisum sativum as the new model. A
novel sequencing technology, pyrosequencing, was used to create a transcriptome
database which is suitable for proteomics applications. Data generated with pea leaf
chloroplasts served as the template to which other differentiated plastid types were

compared both qualitatively and semi-quantitatively.

The qualitative comparison of chloroplast and proplastid envelopes revealed
specific adaptations of the proplastid envelope to its role in meristematic cells:
Proplastids serve as cellular factories for amino acids, fatty acids and nucleotides for the
proliferating cells. The transport protein complement is geared to import precursor
metabolites and export products from the heterotrophic plastid. Chloroplast envelopes of
pea were also compared to those of maize mesophyll cells in a semi-quantitative manner.
Maize plants employ a specific subtype of photosynthesis called C4 photosynthesis
which causes immense metabolite fluxes across the chloroplast envelope. The
comparison revealed quantitative changes in the envelope protein composition indicating
that the flux is accommodated by increased amounts of transport protein. It also revealed
new candidate transport proteins for metabolite fluxes of C4 chloroplasts. One of these
candidate transport proteins was characterized in more detail in the model plant
Arabidopsis thaliana and it likely is a monocarboxylate transporter employed in
photorespiration in C3 plants. Finally, proteome samples of the plastid associated
membranes, the inner and the outer envelope of chloroplast and endomembrane system
were investigated. GFP fiision protein analyses of candidate outer envelope residents
demonstrated that the outer envelope is a dynamic system capable of producing
extensions of the envelope membrane as well as vesicles. Plastid associated membranes
are hypothesized to be part of the autophagy system for chloroplasts since they contain
stromal proteins in addition to envelope and endomembrane system residents. The

hypothesis is supported by GFP fusion protein analysis.

DEDICATED TO MY FAMILY

iv

ACKNOWLEDGEMENTS

I would like to thank Dr. Andreas Weber for allowing me to work on exiting projects in

his lab and for sharing his passion for original research.

I would also like to thank my guidance committee members, Dr. Kenneth Keegstra, Dr.

Christoph Benning and Dr. Yair Shachar—Hill for advice and encouragement.

I would also like to thank my colleagues in the ‘Weber Lab’, past and present, my fellow
students who became colleagues and then friends and my family for their continuing

support.

Special thanks to the directors and administrator of the PhD program in genetics at

Michigan State University.

1 >2 -
~ .

)"‘|"' - .
-... 1.. ,

. f..

Table of contents

List of tables ..................................................................................................................... viii

List of figures ..................................................................................................................... ix

Chapter 1

Introduction and literature review ........................................................................................ 1
Plastid differentiation ....................................................................................................... 2
Biosynthetic pathways of plastids .................................................................................... 4
Plastid architecture ........................................................................................................... 5
Project summary and rationale ......................................................................................... 6
References ...................................................................................................................... l 1

Chapter 2

Transport processes — connecting the reactions of C4 photosynthesis .............................. 15
Abbreviations ................................................................................................................. l 6
Abstract .......................................................................................................................... 17
Introduction .................................................................................................................... l9
Intercellular fluxes .......................................................................................................... 22
Transport processes in the NADP malic enzyme type ................................................... 24
Transport processes in the NAD malic enzyme type ..................................................... 38
Transport processes in the PEP-CK type ....................................................................... 44
Transport processes in single cell C4 organisms ........................................................... 48
Future prospects ............................................................................................................. 49
References ...................................................................................................................... 53

Chapter 3

Comparison of the use of a species-specific database generated by pyrosequencing with

databases from related species for proteome analysis of pea chloroplast envelopes ......... 63
Abstract .......................................................................................................................... 64
Introduction .................................................................................................................... 65
Material and Methods ..................................................................................................... 67
Results ............................................................................................................................ 72
Discussion ...................................................................................................................... 82
References ...................................................................................................................... 93

Chapter 4

Proplastid specific protein patterns revealed by cauliflower curd plastid proteomics ....... 97
Abstract .......................................................................................................................... 98
Introduction .................................................................................................................... 99
Material and Methods ................................................................................................... 102
Results and Discussion ................................................................................................. 104
Conclusion .................................................................................................................... l 20

vi

.‘9 ;; "T‘."I‘.‘ .
'. zines-haul" 5L.-‘ili‘ o": "

References .................................................................................................................... 122

Chapter 5
Comparative proteomics of chloroplast envelopes from C3 and C4 plants reveals specific
adaptations of the plastid envelope to C4 photosynthesis and candidate proteins required

for maintaining C4 metabolite fluxes .............................................................................. 128
Abstract ........................................................................................................................ 129
Introduction .................................................................................................................. 130
Material and Methods ................................................................................................... 134
Results .......................................................................................................................... 138
Discussion .................................................................................................................... 147
Conclusion .................................................................................................................... l 61
References .................................................................................................................... I 63

Chapter 6

Identification and characterization of a metabolite transport protein involved in

photorespiration: the chloroplastic glycerate glycolate carrier Mepl .............................. 170
Introduction .................................................................................................................. 1 7 1
Material and Methods ................................................................................................... 173
Results .......................................................................................................................... 175
Discussion .................................................................................................................... 186
References .................................................................................................................... 194

Chapter 7

Dynamics of the membrane systems surrounding the chloroplast stroma revealed by

proteomics and GFP fusion proteins ................................................................................ 199
Abstract ........................................................................................................................ 200
Introduction .................................................................................................................. 20]
Material and Methods ................................................................................................... 204
Results .......................................................................................................................... 207
Discussion .................................................................................................................... 221
References .................................................................................................................... 232

Chapter 8

Conclusion and Future Perspectives ................................................................................ 237
References .................................................................................................................... 244

Appendix

High—throughput colorimetric method for the parallel assay of glyoxylic acid and

ammonium in a single extract .......................................................................................... 247
Introduction .................................................................................................................. 248
Method ......................................................................................................................... 250
Results and Discussion ................................................................................................. 251
References .................................................................................................................... 257

vii

LIST OF TABLES

Table 2-1: Summary of transport processes necessary in NADP-ME plants .................... 25
Table 2-2: Summary of transport processes necessary in NAD-ME plants ...................... 40
Table 2-3: Summary of transport processes necessary in PEP-CK plants ........................ 46
Table 3-1: Summary of protein identifications with different databases ........................... 74

Table 4-1: Comparison of membrane protein identifications in proplastid and chloroplast
envelope ........................................................................................................................... 104

Table 4-2: Transport proteins of the proplastid envelope and their function .................. 106

Table 6-1: Co-expression coefficients for genes involved in photorespiration and for the
new candidate gene Mepl ................................................................................................ 176

Table 7-1: Known localizations of proteins assigned to the different fractions .............. 213

Table 7-2: GF P localization results for selected outer envelope and PLAM proteins ..... 216

viii

LIST OF FIGURES

Figure 2-1: Scheme outlining the reactions of C4 photosynthesis and the connecting

transport processes for the NADP malic enzyme type of C4 photosynthesis .................... 24

Figure 2-2: Scheme outlining the reactions of C4 photosynthesis and the connecting
transport processes for the NAD malic enzyme type of C4 photosynthesis ...................... 38

Figure 2-3: Scheme outlining the reactions of C4 photosynthesis and the connecting
transport processes for the PEP-CK type of C4 photosynthesis ........................................ 44

Figure 3-1. The length distribution of contigs (assembled transcripts) in databases used
for proteome analyses ........................................................................................................ 72

Figure 3-2. The number of spectral counts detected for a protein is correlated with the
rate of identification ........................................................................................................... 76

Figure 3-3. The number of predicted membrane spanning helices for a protein is not
correlated with the rate of identification ............................................................................ 78

Figure 3-5. The non-species specific databases yield a significant amount of mis-
identifications and multiple identifications ........................................................................ 81

Figure 3-6. Non-species specific databases limit the discovery of new proteins and the
interpretation of known proteins in the pea chloroplast envelope proteome sample ......... 89

Figure 4-1: The pr0plastid envelope enables efficient exchange of metabolites needed for
amino acid biosynthesis ................................................................................................... 108

Figure 4-2: Expression of the protein import complex components varies between leaves
and the shoot apex ............................................................................................................ 113

Figure 4-3: Overall proteome contents in plastid envelopes ........................................... 116

Figure 4-81: RNA accumulation patterns between shoot apex and leaves for orthologous

genes in Arabidopsis ........................................................................................................ 120
Figure 5-1: Schematic representation of central carbon metabolism and associated

transport processes in C3 chloroplasts and C4 PCA type ................................................ 131
Figure 5-2: The envelope proteomes are similar when analyzed qualitatively ............... 140

ix

Figure 5-3: Quantitatively the envelope proteomes differ in selected proteins ............... 142

Figure 5-4: For the extremes and selected unchanged proteins out of Figure 5-3, detailed
results were plotted .......................................................................................................... 144

Figure 5-Sl: For selected proteins, mRNA accumulation patterns in different tissues were

analyzed ........................................................................................................................... l 55
Figure 6-1: GUS expression pattern of a P(Mepl )::GUS fusion ..................................... 178
Figure 6-2: Isolation of a Mepl insertion mutant ............................................................ 179
Figure 6-3: Metabolite accumulation in the Mepl mutant compared to wild type ......... 182
Figure 6-4: Time course of metabolite accumulation in the Mepl insertion ................... 185
Figure 6-5: Mepl is localized in the chloroplast envelope .............................................. 186
Figure 6-6: Schematic representation of photorespiration ............................................... 190

Figure 7-1: Protein distribution across the fractions plotted as number of proteins
determined for each membrane system including targeting predictions from TargetP ...207

Figure 7-2: Distribution of the spectral count between the proteome samples for proteins
of the protein import complex .......................................................................................... 210

Figure 7-3: GFP localization of fusion proteins after initial (A, C, E and G) and
prolonged (B, D, F, H) transient expression in tobacco leaves ........................................ 228

Figure 7-4: GFP expression pattern of Hsp60-like protein after three days of expression
in tobacco leaves .............................................................................................................. 221

Figure 1 (Appendix): Absorption spectra ........................................................................ 249

Figure 2 (Appendix): Comparison of standard curves for glyoxylate (A) and ammonia (B)
determination ................................................................................................................... 252

Chapter 1

Introduction and literature review

 

Plants are the primary source of food for humankind and are increasingly
considered for biofuel production. Since the arable land area is decreasing due to
destruction of arable land for example by salt accumulation, climate change, or
urbanization, supplying the human population with food is one of the challenges of the
let century. Understanding the fundamental processes in plant cells will aid the
development of crops with higher yields or altered nutrient composition as well as
engineering of more efficient crops both for biofuel and food production. I focus my
attention on plastids since they are involved in either synthesis, storage, or both, of all
major agricultural products which are derived from assimilated carbon, carbohydrates,
fatty acids, or amino acids.

Plastid differentiation

A plant contains several different types of plastids which can be classified
according to their morphology and physiological role. In each plant, plastid
differentiation starts from undifferentiated proplastids of embryonic and meristematic
tissues (Hoober, 2006). The proplastids of meristematic tissues differentiate into
chloroplasts, the best studied plastid type, upon exposure to light. Chloroplasts are the
most abundant plastid type in most photosynthesizing plants and harness the energy of
light to fix inorganic carbon. If light is withheld experimentally or by soil cover in the
field, a developmental program causes proplastid differentiation into etioplasts defined by
their developmental state and particular ultrastructure which subsequently develop into
chloroplasts if light exposure eventually occurs (Holm and Deng 1999). Chloroplasts are

the only plastid type which is autotrophic for organic carbon, energy and reducing

equivalents whereas all other plastid types depend on the import of precursor metabolites,
energy, and reducing power. Chloroplast themselves can differentiate even further into
two subtypes in plant species which employ C4 photosynthesis to fix inorganic carbon
(Hall et al. 1998; Cribb et a1. 2001). Both proplastids and chloroplasts can differentiate
into chromoplasts. Chromoplasts are defined based on their color as yellow or reddish
plastids which contain enhanced carotenoid levels (Kirk and Tilney-Bassett 1978). The
differentiation of proplastids into chromoplasts has been studied in the natural
cauliflower mutant Orange (Crisp et al. 1975) which is caused by a mutation in a DnaJ
cysteine rich domain protein targeted to colorless plastids (Lu et al. 2006). Although the
exact mechanism of chromoplast differentiation remains unresolved it likely involves a
protein identified as promoting chromoplast development in bell pepper (Hugueney et al.
1995) which is upregulated in the Orange mutant of cauliflower (Lu et al. 2006). Upon
differentiation of chloroplasts into chromoplasts extensive vesiculation of the envelope
occurs (Hugueney et al. 1995). Leucoplasts are defined as colorless mature plastids and
include amyloplasts in roots. These plastids have the capability to green upon light
exposure which induces the formation of thylakoids (Ljubicic et al. 1998). Plastids in
storage organs are generally white heterotrophic plastids which either store starch (in
plants which rely on starch as the reserve material in the developing seedling) or produce
fatty acids for triglyceride production (e.g. in Brassica napus seeds) or amino acids (for
plants which mainly store reserves as proteins). Aside from any specific roles depending
on the plastid subtype, e. g. color production and display in chromoplasts or providing
rapidly dividing small plastids ready to differentiate in proplastids (Kirk and Tilney-

Bassett 1978), all functional plastids can provide their cells with essential metabolites.

 

 

Biosynthetic pathways of plastids

All plastids are semiautonomous organelles and cannot be generated de novo but
only by division of plastids. They entered the plant lineage in a single endosymbiosis
event and have since lost most of their gene content to the nucleus. However most of the
anabolic pathways remain localized in the plastids of plant cells. A possible reason for
retaining a large number of pathways inside the plastid may be that plastids provide a
subcompartment with different reductant and ATP concentrations compared to the
remainder of the cell (Heineke et al. 1991).

Chloroplasts are the only autotrophic plastid type as they can transform light
energy into chemical energy. The electron transfer chain located within an internal
membrane system, the thylakoids, generates energy in the form of ATP and reducing
power in the form of NADPH. Both are used to fix inorganic carbon and nitrogen into
organic molecules. Energy and reducing power generated and stored in organic molecules
is the basis for the anabolic reactions of plants.

Most of the amino acids produced in plants are produced in the plastids. Branched
chain amino acids are synthesized in plastids whereas turnover and breakdown is
restricted to the mitochondria and the peroxisomes (Binder et al. 2007). Aromatic amino
acids are also exclusively synthesized in the plastid via a common intermediate
chorismate (Herrmann and Weaver 1999). Chorismate and the aromatic amino acids are
precursors for many products of the secondary metabolism such as phylloquinones
(Shimada et al. 2005) and— via phenylalanine - salicylic acid, flavonoids and lignins
(Weaver and Herrmann 1997). Arginine synthesis is also considered to be a plastidic

pathway (Slocum 2005) as is lysine synthesis (Hudson et al. 2005). A number of essential

plant metabolites synthesized in the plastid are derived from 5-ribosyl l-pyrophosphate.
The amino acids tryptophan (Zhao and Last 1995) and histidine (Stepansky and Leustek
2006) are synthesized in the plastid as are purines and pyrimidines with the exception of
one reduction step in pyrimidine biosynthesis (Zrenner et al. 2006). Fatty acids are also
mainly synthesized in the plastids (Rawsthorne 2002) and serve as precursors for the
synthesis of a number of membrane lipids (Benning et al. 2006). The production of a
large range of very different metabolites in plastids raises the question how the plastids
efficiently import the necessary precursors and export the products. For many
agriculturally important products, the elucidation of their biosynthetic pathways has
initially been focused on studying the soluble components rather than the transport
processes connecting the metabolic network.
Plastid architecture

Plastids are derived from an ancient endosymbiotic event. They can only multiply
by division of existing plastids and contain a rudimentary genome as well as transcription
and translation machinery. Most of the plastidic proteins are synthesized on cytosolic
ribosomes and imported posttranslationally (Leister, 2003). Plastids contain an aqueous
phase, the stroma, which is separated from the cytosol by two membranes, the outer
envelope membrane which faces the cytosol and the inner envelope membrane which
faces the stroma. The stroma contains an additional membrane system, which is well
developed in chloroplasts and called thylakoids or a set of vesicles in most other plastid
types. The outer envelope is believed to be relatively permeable for solutes smaller than
lOkDa (Weber et al. 2005; Weber and Fischer 2007) by way of a number of porins with

broad substrate specificity (Pohlmeyer et al. 1997; Pohlmeyer et al. 1998; Bolter et al.

1999; Goetze et al. 2006). Proteins targeted to the plastids posttranslationally pass the
outer envelope through an elaborate protein translocation system called Translocon outer
envelope of chloroplast or Toc complex and pass the inner envelope through the
Translocon inner envelope of chloroplasts or Tic complex (Schnell 2000; Jarvis and
Robinson 2004; 8011 and Schleiff 2004). Small solutes pass the inner envelope by way of
a number of specific metabolite transport proteins (Weber 2004; Weber et al. 2005). The
only plastid envelope which has been analyzed comprehensively is the chloroplast
envelope of dicot plants such as spinach and Arabidopsis (Ferro et al. 2002; Miras et al.
2002; Ferro et al. 2003; Froehlich et al. 2003). Studies of other differentiated plastid
types have been focused on the soluble proteome although occasionally envelope proteins
have also been identified (Baginsky et al. 2004; Kleffmann et al. 2004; von Zychlinski et
al. 2005; Siddique et al. 2006; Kleffmann et al. 2007).
Project summary and rationale

The long term goal of studying plastid envelopes is to tie the plastid into the
different regulatory and metabolic networks of the cell since the ultimate barrier of both
metabolites and signaling molecules which originate in the stroma or at the thylakoids are
the inner and outer envelope membranes. Evidence based on proteome projects focused
on either the thylakoid proteome of chloroplasts, the soluble proteome of chloroplasts or
the entire plastid has indicated that, in addition to housekeeping proteins, the plastid
carries a protein complement adapted to its specific role (Baginsky et al. 2004;
Kleffmann et al. 2004; von Zychlinski et al. 2005; Siddique et al. 2006; Kleffmann et al.
2007). Proteome analysis has mostly been focused on the soluble or thylakoid proteome

of different plastid types (Majeran et al. 2005; Majeran et al. 2008) since both are easier

to access and analyze than the envelope proteome. Envelope proteomes have only been
analyzed in chloroplasts of C3 plants. In this work, I report about dynamic envelope
membranes with respect to both protein content and protein localization. A new reference
chloroplast envelope of Pisum sativum chloroplasts is introduced and compared
qualitatively to a proplastid envelope yielding new insights in proplastid metabolism. The
reference chloroplast proteome is also compared semi-quantitatively to the envelope of
C4 chloroplasts which reveals that protein amount is adjusted depending on the metabolic
fluxes required for each plastid type (for a detailed description of C4 photosynthesis,
please see chapter 2). The characterization of accumulation patterns for known metabolite
transport proteins of the chloroplast envelope allows the determination of several
candidate transport proteins for known metabolite fluxes. One of these candidate
transport proteins named Mepl was characterized in more detail. Finally, Pisum sativum
chloroplast envelopes and associated membrane as well as a light microsome fraction are
used to gain insight in the dynamics of protein localization in the membrane systems
surrounding chloroplasts.

Chapter 2 reviews the current status of knowledge about transport processes in C4
photosynthesis. C4 photosynthesis requires two different chloroplast types to achieve a
more efficient photosynthesis compared to that of C3 plants and has been targeted in
bioengineering efforts to achieve higher yields, for example in rice plants. The review
identifies the transport proteins of one of the understudied fields in C4 research which
may hamper engineering efforts.

Chapter 3 reports the establishment of a novel model for plastid envelope

proteomics, Pisum sativum. Prior to this study, chloroplast proteomics was mainly

performed with species that already had a sequenced genome (e.g. Arabidopsis or rice) or
transcriptome (e.g. maize) available. Pea as a model for plastid proteomics has several
advantages over the models used previously, because large quantities of high quality
isolation material can be grown in a short time. There are isolation protocols not only for
leaf chloroplasts but also for root leucoplasts available. Moreover, the outer and the inner
envelope of pea chloroplasts can be separated efficiently. Although it has been generally
assumed that cross species identification is sufficient for organellar proteomics (Baginsky
et al. 2004; Siddique et al. 2006), I identify cross species comparisons as a major obstacle
to proteome analysis and analyze the parameters limiting it. In addition, I determine that a
species-specific transcriptome database suitable for organellar proteome analysis can be
generated by pyrosequencing.

In chapter 4, I report the first envelope proteome analysis of proplastids and
determine the metabolic role of proplastids based on their envelope protein content.
Proplastid envelopes have never been studied since material of meristems is limited in
model organism such as Arabidopsis or rice. Electron micrographs of proplastids show
that they contain a rudimentary internal membrane system of unknown composition and
frequently a small starch granule (Kirk and Tilney-Bassett 1978; Journet and Douce
1985). Proteome analysis of BY-2 plastids which represent a meristem-like system with
undifferentiated plastids indicates that these plastids are enriched in enzymes for amino
acid biosynthesis (Baginsky et al. 2004). To address proplastid biology, I analyze
proplastid envelopes by proteomics and compare them to chloroplast envelopes. The
proteins identified are different from those in pea envelopes and point towards the

proplastid as an active cellular factory which produces amino acids, fatty acids and

nucleotide precursors for the remainder of the proliferating cell. They contain not only a
specific subset of transport proteins but different subsets of thylakoid proteins and
components of the protein import complex compared to leaf chloroplast envelopes. The
surprising identification of the triosephosphate phosphate translocator at least partially
redefines the role of the glucose-6-phoshate phosphate translocator and the oxidative
branch of the pentosephosphate pathway in heterotrophic plastids.

Chapter 5 reports the identification of transport protein amount as one of the ways
to control metabolite traffic across the inner and outer envelope. C4 photosynthesis is a
highly compartmented process which involves chloroplasts in two different cell types. It
causes high volumes of metabolite traffic in excess of those observed in C3 plants. When
this work was started, the adaptations of the chloroplast envelopes to the increased
volume of traffic were unknown. To address this question, the first detailed semi-
quantitative picture of a C3 chloroplast envelope was produced and compared to the
chloroplast envelope of a C4 plant. The comparison indicated major changes in quantity
of membrane transport proteins of both the inner and the outer envelope and provided the
first indication that membrane flux is regulated at least in part by the amount of available
transport protein. It also yielded a number of candidate transport proteins for metabolite
flux across the envelope of C4 chloroplasts.

Chapter 6 describes the analysis of a membrane protein identified as one of the
candidate proteins in chapter 5. At the onset of the work nothing was known about the
function of the protein and I hypothesized that a detailed analysis of the Arabidopsis
homologue with the tools of molecular biology and reverse genetics may have the power

to gain insights into the function of the protein in vivo. To this end, I characterized the

expression pattern and confirmed the localization. 1 isolated an insertion mutant and
characterized it as a null mutant which can be complemented. The null mutant displayed
a visible phenotype which is predated by a unique biochemical phenotype indicating the
Arabidopsis mutant has limited capacity to transport glycerate and glycolate. All attempts
to characterize the protein biochemically failed.

Chapter 7 reports the study of the proteomes of not only mixed envelope
membranes but also of inner and outer envelope membranes and chloroplast associated
membranes in comparison with an ER enriched microsome fraction. The protein
composition of outer envelopes has only been studied by electrophoretic methods (Cline
et al. 1981) and by a limited proteome analysis without a dedicated sequence database
(Schleiff et al. 2003). Recently, a new class of membrane associated with chloroplasts
and labeled by a ER OF P marker protein has been identified (Andersson et al. 2007). The
comparative proteome analysis was designed to both identify new residents of chloroplast
envelopes and localize them to one of the membrane systems. To address this question, I
collaborated with Henrik Tjellstroem and Anna Stina Sandelius. The comparative
proteome analysis reveals unexpected fluidity of the membrane systems and defines the
proteins content of a novel organelle, the plastid associated membranes. GFP fusion
proteins support the new hypothesis that the protein localization in the membrane systems

of chloroplasts is more dynamic than previously assumed.

10

References

Andersson MX, Goksor M, Sandelius AS (2007) Optical manipulation reveals strong
attracting forces at membrane contact sites between endoplasmic reticulum and
chloroplasts J Biol Chem 282, 1170-1174

Baginsky S, Siddique A, Gruissem W (2004) Proteome analysis of tobacco bright yellow-
2 (BY-2) cell culture plastids as a model for undifferentiated heterotrophic plastids J
Proteome Res 3, 1128-1137

Benning C, Xu CC, Awai K (2006) Non-vesicular and vesicular lipid trafficking
involving plastids Curr Opin Plant Biol 9, 241-247

Binder S, Knill T, Schuster J (2007) Branched-chain amino acid metabolism in higher
plants Physiol Plant 129, 68-78

Bolter B, 8011 J, Hill K, Hemmler R, Wagner R (1999) A rectifying ATP-regulated solute
channel in the chloroplastic outer envelope from pea EMBO J 18, 5505-5516

Cline K, Andrews J, Mersey B, Newcomb EH, Keegstra K (1981) Separation And
Characterization Of Inner And Outer Envelope Membranes Of Pea-Chloroplasts Proc
Natl Acad Sci USA 78, 3595-3599

Cribb L, Hall LN, Langdale JA (2001) Four mutant alleles elucidate the role of the G2
protein in the development of C-4 and C-3 photosynthesizing maize tissues Genetics 159,
787-797

Crisp P, Walkey DGA, Bellman E, Roberts E (1975) Mutation Affecting Curd Color In
Cauliflower (Brassica-Oleracea L Var Botrytis Dc) Euphytica 24, 173-176

Ferro M, Salvi D, Brugiere S, Miras S, Kowalski S, Louwagie M, Garin J, Joyard J,
Rolland N (2003) Proteomics of the chloroplast envelope membranes from Arabidopsis
thaliana Mol Cell Proteomics 2, 325-345

Ferro M, Salvi D, Riviere-Rolland H, Vermat T, Seigneurin-Bemy D, Grunwald D, Garin
J, Joyard J, Rolland N (2002) Integral membrane proteins of the chloroplast envelope:
identification and subcellular localization of new transporters Proc Natl Acad Sci U S A
99, 11487-92.

Froehlich JE, Wilkerson CG, Ray WK, McAndrew RS, Osteryoung KW, Gage DA,
Phinney BS (2003) Proteomic study of the Arabidopsis thaliana chloroplastic envelope
membrane utilizing alternatives to traditional two-dimensional electrophoresis J
Proteome Res 2, 413-25

11

er—u’fl

 

Goetze TA, Philippar K, Ilkavets I, 8011 J, Wagner R (2006) OEP37 Is a New Member of
the Chloroplast Outer Membrane Ion Channels J Biol Chem 281, 17989-17998

Hall LN, Rossini L, Cribb L, Langdale JA (1998) GOLDEN 2: A novel transcriptional
regulator of cellular differentiation in the maize leaf Plant Cell 10, 925-936

Heineke D, Riens B, Grosse H, Hoferichter P, Peter U, Flugge UI, Heldt HW (1991)
Redox Transfer Across The Inner Chloroplast Envelope Membrane Plant Physiol 95,
1131-1137

Herrmann KM, Weaver LM (1999) The shikimate pathway Annu Rev Plant Physiol Plant
Mol Biol 50, 473-503

Holm M, Deng XW (1999) Structural organization and interactions of COPI, a light-
regulated developmental switch Plant Mol Biol 41, 151-158

Hudson AO, Bless C, Macedo P, Chatterjee SP, Singh BK, Gilvarg C, Leustek T (2005)
Biosynthesis of lysine in plants: evidence for a variant of the known bacterial pathways
Biochim Biophys Acta-Gen Subj 1721, 27-36

Hugueney P, Bouvier F, Badillo A, Dharlingue A, Kuntz M, Camara B (1995)
Identification Of A Plastid Protein Involved In Vesicle Fusion And/Or Membrane-Protein
Translocation Proc Natl Acad Sci U S A 92, 5630-5634

Jarvis P, Robinson C (2004) Mechanisms of protein import and routing in chloroplasts
Curr Biol 14, R1064-R1077

Journet EP, Douce R (1985) Enzymic Capacities Of Purified Cauliflower Bud Plastids
For Lipid-Synthesis And Carbohydrate-Metabolism Plant Physiol 79, 458-467

Kirk JTO, Tilney-Bassett RAE (1978) 'The plastids: Their chemistry, structure, growth
arid inheritance.’ (Elsevier: Amsterdam/Oxford)

K1 effinann T, Russenberger D, von Zychlinski A, Christopher W, Sjolander K, Gruissem
W, Baginsky S (2004) The Arabidopsis thaliana chloroplast proteome reveals pathway
abundance and novel protein functions Curr Biol 14, 354-362

I(leffinann T, von Zychlinski A, Russenberger D, Hirsch-Hoffmann M, Gehrig P,
P 1"-11 Ssem W, Baginsky S (2007) Proteome dynamics during plastid differentiation in rice
lant Physiol 143, 912-923

.17‘5 ubicic JM, Wrischer M, Ljubesic N (1998) Formation of the photosynthetic apparatus

171—1 plastids during greening of potato microtubers Plant Physiology And Biochemistry 36,
4 '7-752

12

 

Lu S, Van Eck J, Zhou X, Lopez AB, O'Halloran DM, Cosman KM, Conlin BJ, Paolillo
DJ, Garvin DF, Vrebalov J, Kochian LV, Kupper H, Earle ED, Cao J, Li L (2006) The
cauliflower or gene encodes a DnaJ cysteine-rich domain-containing protein that
mediates high levels of beta-carotene accumulation Plant Cell 18, 3594-3605

Majeran W, Cai Y, Sun Q, van Wijk KJ (2005) Functional differentiation of bundle
sheath and mesophyll maize chloroplasts determined by comparative proteomics Plant
Cell 17, 3111-3140

Majeran W, Zybailov B, Ytterberg AJ, Dunsmore J, Sun Q, van Wijk KJ (2008)
Consequences of C4 differentiation for chloroplast membrane proteomes in maize
mesophyll and bundle sheath cells Mol Cell Proteomics 10.1074/mcp.M800016-
MCP200, M800016-MCP200

Miras S, Salvi D, Ferro M, Grunwald D, Garin J, Joyard J, Rolland N (2002) Non-
canonical transit peptide for import into the chloroplast J Biol Chem 277, 47770-47778

Pohlmeyer K, Soll J, Grimm R, Hill K, Wagner R (1998) A High-Conductance Solute
Channel in the Chloroplastic Outer Envelope from Pea Plant Cell 10, 1207-1216

Pohlmeyer K, 8011 J, Steinkamp T, Hinnah S, Wagner R (1997) Isolation and
Characterization of an amino acid-selective channel protein present in the chloroplastic
outer envelope membrane Proc Natl Acad Sci U S A 94, 9504-9509

Rawsthome S (2002) Carbon flux and fatty acid synthesis in plants Prog Lipid Res 41 ,
1 82-196

Schleiff E, Eichacker LA, Eckart K, Becker T, Mirus O, Stahl T, 8011 J (2003) Prediction
Of the plant beta-barrel proteome: A case study of the chloroplast outer envelope Protein
Sci 12, 748-759

'Schnell DJ (2000) Functions and origins of the chloroplast protein-import machinery. In
M Olecular Trafficking' pp. 47-59)

Shimada H, Ohno R, Shibata M, Ikegami 1, Onai K, Ohto M, Takarniya K (2005)
Ir1€i<>tivation and deficiency of core proteins of photosystems I and II caused by genetical
phy lloquinone and plastoquinone deficiency but retained lamellar structure in a T-DNA
mutant of Arabidopsis Plant J 41, 627-637

S iC-iCiique MA, Grossmann J, Gruissem W, Baginsky S (2006) Proteome analysis of bell
Ipepper (Capsicum annum L.) chromoplasts
lant Cell Physiol, 33-39

S10

P 1 Cum RD (2005) Genes, enzymes and regulation of arginine biosynthesis in plants

atlt Physiology And Biochemistry 43, 729-745

13

1 8011 J, Schleiff E (2004) Protein import into chloroplasts Nature reviews Molecular cell
biology 5, 198-208

Stepansky A, Leustek T (2006) Histidine biosynthesis in plants Amino Acids 30, 127-142

 

von Zychlinski A, Kleffmann T, Krishnamurthy N, Sjolander K, Baginsky S, Gruissem
W (2005) Proteome analysis of the rice etioplast - Metabolic and regulatory networks and
novel protein functions Mol Cell Proteomics 4, 1072-1084

I Weaver LM, Hemnann KM (1997) Dynamics of the shikimate pathway in plants Trends
. Plant Sci 2, 346-351

Weber APM (2004) Solute transporters as connecting elements between cytosol and
plastid stroma Curr Opin Plant Biol 7, 247-253

Weber APM, Fischer K (2007) Making the connections - The crucial role of metabolite
transporters at the interface between chloroplast and cytosol F EBS Lett 581, 2215-2222

Weber APM, Schwacke R, F lugge UI (2005) Solute transporters of the plastid envelope
membrane Ann Rev Plant Biol 56, 133-164

Zhao J, Last RL (1995) Immunological Characterization and Chloroplast Localization of
the Tryptophan Biosynthetic Enzymes of the Flowering Plant Arabidopsis thaliana J Biol
Chem 270, 6081-6087

Zrenner R, Stitt M, Sonnewald U, Boldt R (2006) Pyrimidine and purine biosynthesis and
degradation in plants Ann Rev Plant Biol 57, 805-836

 

14

 

Chapter 2

Transport processes — connecting the reactions of C4 photosynthesis

This chapter has been accepted for publication in the series Advances of Photosynthesis
and Respiration: Andrea Brautigam and Andreas P. M. Weber: Transport processes -
Connecting the reactions of C4 photosynthesis. (2009) Advances in Photosynthesis and

Respiration. Series Editor R. Govindjee, Springer

15

 

 

Abbreviations

2-OG
3-PGA
AAC
DHAP
DIC
DiT
DTC
MCF
MDH
NAD-ME
NADP-ME
OAA

PCA

2-oxoglutarate

3-phosphoglycerate

ATP ADP carrier

dihydroxyacetone phosphate

dicarboxylate carrier

dicarboxylate translocator

di- and tricarboxylate carrier

mitochondrial carrier family

malate dehydrogenase

NAD-malic enzyme

NADP-malic enzyme

oxaloacetate

photosynthetic carbon assimilation
photosynthetic carbon reduction
phosphoenolpyruvate

phosphoenolpyruvate carboxylase
phosphoenolpyruvate phosphate translocator
PEP phosphate dikinase

PEP carboxykinase

ribulosebisphosphate carboxylase oxygenase

triosephosphate phosphate translocator

16

Abstract

The C4 cycle requires immense metabolite fluxes. The spatial separation of initial
carbon fixation by phosphoenolpyruvate carboxylase and entry in the photosynthetic

carbon reduction cycle through Rubisco requires metabolites to shuttle not only between

_.——\

cells but also across intracellular membranes. C4 photosynthesis is a highly

compartmentalized process. Atmospheric C02 is fixed into C4 acids (photosynthetic
carbon assimilation, PCA) in one domain and C4 acids donate CO2 to Rubisco in another
domain (photosynthetic carbon reduction, PCR). PCA occurs in chloroplasts and cytosol;
PCR occurs in chloroplasts, and depending on the subtype of C4 photosynthesis may
involve the mitochondria and cytosol. lntercellular transport likely occurs symplastically
but the intracellular transport processes across the organellar membranes are at least in
l part mediated by specific transport proteins. These transport processes are of particular
interest because metabolites have to be transported at the rate of carbon assimilation;
each carbon which is shuttled as a C4 acid necessitates distinct transport processes as
does the C3 acid which returns to recycle the initial carbon acceptor. Currently, it is not
fillly understood how the organellar membranes accommodate the high volume and
Velocity of the necessary flux.
This chapter will review the different types of C4 cycle reactions and the transport
D recesses required for each sub-type based on the localization of the enzymes involved in
the C4 cycle. For each transport process the current knowledge about the transport
l31‘Qteins involved is stated in detail including discussion of candidate transport proteins

=lb-laracterized in C3 systems. Finally, novel strategies for identifying and characterizing

17

 

—.—-—..—-—,_

 

molecular candidates for transport proteins and their importance for engineering a C4

cycle in C3 crop plants are described.

18

\
l
l

 

Introduction

One of the highest steady state fluxes of metabolites across organellar membranes
known to date originates from the process of C4 photosynthesis. In C4 photosynthesis,
the production of the primary carbon acceptor phosphoenolpyruvate (PEP) and fixation
of atmospheric C02 by phosphoenolpyruvate carboxylase (PEPC) occurs in the
photosynthetic carbon assimilation (PCA) compartment. In the photosynthetic carbon
reduction (PCR) compartment C02 is released from C4 acids and enters the Calvin Cycle
(reviewed in Hatch (1987) and Edwards et al. (20013)). In most C4 species, the PCA and
PCR compartments are located in two distinct cell types, with PCA cells being leaf
mesophyll tissue and PCR cells surrounding the veins as a bundle sheath. This
arrangement is called Kranz anatomy. Metabolite transport between the cells is not
completely understood but it likely involves symplastic connections (Craig and
Goodchild 1977; Hattersley and Browning 1981; Sowinski et al. 2008). In contrast,

intracellular transport requires specific transport proteins in the membranes surrounding
the organelles since the reactions of the C4 cycle occur in three subcellular
e()mpartments, the cytosol, the chloroplasts, and the mitochondria (Hatch 1987). Both
cl‘lloroplasts and mitochondria are separated from the cytosol by two membranes. In
cl'Illoroplasts, the outer envelope membrane which faces the cytosol is relatively
Ibgl'meable to small metabolites because it contains a number of porins with broad
S"‘~llostrate specificities (Pohlmeyer et al. 1997; Pohlmeyer et al. 1998; Bolter et al. 1999;
QQetze et al. 2006; Murcha et al. 2007). The inner envelope membrane between the
it‘termembrane space and the stroma is the diffusion barrier and contains specific

transport proteins (Weber 2004; Weber et al. 2005; Weber and Fischer 2007). In

19

mitochondria, the outer membrane is also relatively permeable because of a number of
porins (Benz 1994) and the inner membrane represents the selectivity filter. Most specific
transport proteins residing in the inner envelopes catalyze the movement of two
molecules, operating either in counter-exchange or co-transport mode (Weber et al.
2005). When the transport of a solute is coupled to that of another ion or solute it is not
only possible to avoid creating electrochemical gradients but the solute can even be
transported against its concentration gradient at the expense of the gradient of the second
molecule.

For each carbon atom that is assimilated and reduced in C4 species, the C4 cycle
performs at least one full turn, and the metabolites involved in C4 photosynthesis are
transported across different organellar membranes at the rate of carbon assimilation
(Figure 2-1). Overcycling to enrich CO2 within the PCR tissue, which has been estimated
at 10% to 40% of the apparent rate of CO2 assimilation (Henderson et al. 1992; Laisk

and Edwards 2000; Kubasek et al. 2007), firrther increases the load on the metabolite
transport systems. Compared to the most abundant transport protein in C3 chloroplasts,
tlie triosephosphate phosphate translocator (TPT), each metabolite transport protein

i Ilvolved in the C4 cycle has to carry at least three times more cargo per unit of time. In
Zea mays, a NADP malic enzyme (NADP-ME) type C4 plant, for each carbon
assimilated, at least four transport processes across the PCA chloroplast envelope and at
l $ast three transport processes across the PCR chloroplast envelope are required (for a
Q1etailed description of transport processes in all C4 photosynthesis subtypes see below).
FOr the three carbon atoms contained in one molecule of triosephosphate, this adds up to

at least 21 transport processes. In a C3 plant, however, for three carbon atoms

20

 

assimilated, at most one transport process across the chloroplast envelope, (i.e., the
export of one molecule of triosephosphate from the chloroplast) is required. It is currently
not fully understood how the at least twentyfold higher metabolite flux across the
organellar membranes of C4 plants is mastered, especially since a number of transport
proteins that carry only minor fluxes in C3 plants are likely playing a more prominent
role in C4 photosynthesis because they have to carry a major flux in these plants. In
addition to the metabolites which are shuttled to support CO2 assimilation, other pathway
intermediates also have to be shuttled between PCR and PCA chloroplasts. Chloroplasts
in PCR tissues of C4 plants frequently have reduced water splitting activity to minimize
the oxygen concentration around Rubisco (reviewed in Walker and Edwards (1983);
Meierhoff and Westhoff 1993). In consequence, reactions such as the reduction of Calvin
Cycle intermediates, fatty acid synthesis, and nitrogen or sulfur reduction depend on the
reducing power generated in chloroplasts of the PCA tissue. Since pyridine or pyrimidine
nucleotides cannot be directly exported or imported into chloroplasts, reduction power
can be either shuttled in the form of oxidized and reduced metabolite pairs, such as 3-
phosphoglycerate (3-PGA) and triosephosphate, or the intermediates of the pathways
themselves will have to be shuttled between PCR and PCA type tissues. To our
knowledge, the C4 cycle causes one of the highest steady state fluxes of metabolites
occurring across any organellar envelope.

Despite the high load of transported metabolites, C4 photosynthesis likely did not
require the invention of novel transport proteins. C4 photosynthesis independently
evolved at least 50 times (Sage 2004; Muhaidat et al. 2007), making it highly unlikely

that the invention of metabolite transport proteins occurred each time in a convergent

21

 

 

 

manner. Rather, as is the case for the soluble enzymes involved in C4 photosynthesis
(Matsuoka et al. 2001; Sage 2004) pre-existing transport proteins may have been
recruited for C4 photosynthesis by increasing their abundance and/or changing their
expression patterns. Possibly, minor transport substrates of C3 plant transporters have
become major transport substrates of the corresponding C4 plant transporter. Therefore,
whenever possible, evidence from C3 plants is considered when transport proteins for C4

species are discussed.
lntercellular fluxes

All C4 plants that rely on two different cell types to enrich C02 in the vicinity of
RubisCO need to transport the metabolites of the C4cycle between these cells types. The
two photosynthetic cell types involved are connected by an unusually high number of
plasmodesmata (Craig and Goodchild 1977; Evert et al. 1977; Hattersley and Browning
1981). Based on the frequency of plasmodesmata and the diffusion surface area it was
calculated that concentration gradients of 10 mM are necessary to drive efficient transport
of metabolites by diffusion (Hatch 1987). Indeed, concentration gradients of 5-10 mM for
many metabolites involved in the C4 cycle appear to be present in C4 species (Leegood
1985; Stitt and Heldt 1985; Furbank and Hatch 1987). The transport of metabolites
between cells is thus dependent on Brownian motion which is limited by cytoplasmic
viscosity and temperature (Leegood and Edwards 1996). The concentration gradient for
pyruvate, however, which was experimentally determined between whole mesophyll and
bundle sheath cells in Zea mays, is not steep enough to drive efficient diffusion (F lfigge et
al. 1985; Stitt and Heldt 1985). It was concluded that the sequestration of metabolites into

subcellular compartments such as chloroplasts may provide the driving force for

22

 

 

intercellular transport (Fliigge et al. 1985). Studies of plasmodesmatal architecture
(Overall et al. 1982; Ding et al. 1992; Roberts and Oparka 2003) have also questioned the
assumptions of the simple diffusion model (Sowinski et al. 2008). In particular, the
plasmodesmata contain internal structures, most notably the desmotubule, which
connects the endoplasmatic reticulum of the adjacent cells and limits the cross section
available for diffusion. Plasmodesmata themselves have different size exclusion limits for
molecules that can traverse them depending on the cells they connect (Erwee and
Goodwin 1985). Recent estimates for the concentration gradients needed to drive
metabolite transport are three orders of magnitude higher than previously assumed
(Sowinski et al. 2008). Sowinski et al. (2008) propose alternative transport mechanisms
such as diffusion through the desmotubule or vesicle transport of metabolites but the
membrane continuity between the endoplasmatic reticulum and chloroplasts or
mitochondria required for either of the alternatives proposed has never been conclusively
demonstrated. Although symplastic transport by diffusion remains the accepted model for
intercellular transport (Sowinski et al. 2008), it is currently not known how the steep

metabolite gradients needed to drive diffusion are generated.

23

 

 

 

Transport processes in the NADP malic enzyme type

 

 

 

   

 

 

 

r r a
c malate malate \ c
Q \\
. 0AA \\
0AA \ malate /\’malate
Pi 9. F’i
PEP . PEP mvate ruvate
' ,I’l py \/py
pyruvate pyruvate '
.
H+l ‘ H+l
+ +
Na Na
3—PGA 3-PGA <---- 3-PGA 3-PGA
.
1 6 _______ rp
L J

 

 

Figure 2-1: Scheme outlining the reactions of C4 photosynthesis and the connecting
transport processes for the NADP malic enzyme type of C4 photosynthesis in both the PCA
(left) and the PCR (right) compartment; the cell wall is represented by a grey bar; intracellular
transport by arrows; intercellular transport by dashed arrows; metabolic reactions by broad
arrows; C chloroplast, for additional abbreviations see text;

Operation of the C4 cycle in the NADP-ME type involves the cytosol and
chloroplasts of the PCA tissue and the chloroplasts of PCR tissue (Figure 2-1). This type
of the C4cycle necessitates at least seven metabolite transport steps for each molecule of
C02 assimilated (Figure 2-1). The CO2 acceptor PEP is first generated within
chloroplasts of the PCA tissue and then exported to the cytosol by a
phosphoenolpyruvate/phosphate translocator (PPT). Oxaloacetate (OAA) is produced
from PEP and C02 in the carboxylation reaction in the cytosol of PCA cells and then
imported into the chloroplasts of these cells. There it is reduced to malate and then
exported to cytosol again. The import and export of these C4 acids are assumed to occur

through a single transport protein, the OAA/malate exchanger (Hatch et al. 1984b;

24

 

 

Taniguchi et al. 2004). The reduction of OAA to malate enables the C4 acid to carry not
only C02 but also one reducing equivalent to the PCR tissue. After import into the PCR
type chloroplasts, malate is oxidatively decarboxylated to yield pyruvate, NADPH, and
CO2, the latter two entering the photosynthetic carbon reduction (or Calvin) cycle.
Pyruvate is exported back out of the PCR chloroplasts and returns to the PCA tissue
where it is imported back into the chloroplasts to regenerate the C02 acceptor PEP
(reviewed in Hatch (1987)). The transport processes are summarized in Table 2-1. The
best studied C4 species are of the NADP—ME subtype and they include species from the
genera Zea, Sorghum, and F laveria.

Table 2-1: Summary of transport processes necessary in NADP-ME plants, for abbreviations

 

see text
domain substrate co- name evidence references
substrate
biochemica'and ”121.11? 5.381333?
PCA PEP Pi PPT moltecular, supported B rantigam et al. (2008);
Y ”0‘60““cs Majeran et al. (2008)
OAT biochemical Hatch et al. (1984a)
- - Weber et al. (1995);
PCA OAA malate D'T bllocqemlcal an: d Taniguchi et al. (2002,
' "‘0 5°" 8” sum.” 8 2004); Renne et al.
y proteomics (2003)
inferred from
PCR malate ? ? localization of soluble -
proteins
inferred from
PCR pyruvate ? ? localization of soluble -
proteins
+ . . Flilgge et al. (1995);
H ? biochemical .
PC A pyruvate + Aokr et al. (1992)
Na ? biochemical Aoki et al. (1992)
- - Day and Hatch (1981);
PC A mSEChlgmcal anged Rumpho und Edwards
b “ rotesdiiii); (1985); Brautigam et al.
3-PGA DHAP TPT y p (2008)
inferred from
PCR localization of soluble Majeran et al. (2008)

proteins

 

25

 

 

 

PEP export from PCA type chloroplasts

The primary C02 acceptor PEP is regenerated from pyruvate by pyruvate phosphate
dikinase (PPDK). This enzyme is localized in PCA type chloroplasts, both in single-cell
and Kranz-type C4 photosynthesis, necessitating the export of PEP across the chloroplast
envelope (Figure 2-1) (Hatch 1987; Voznesenskaya et al. 2001; Voznesenskaya et al.
2002). It was initially characterized biochemically using isolated PCA chloroplasts of Zea
mays and Digitaria sanguinalis (Huber and Edwards 1977b; Day and Hatch 1981). Both
inorganic phosphate (Pi) and 3-PGA but not pyruvate applied externally lead to export of
PEP from chloroplasts in vitro in D. sanguinalis (Huber and Edwards 1977b). The
transport protein has high affinity for Pi with an apparent KM of 200uM and an equally
high affinity for PEP, determined as the apparent Ki=450rrM for inhibition of Pi transport
by PEP. Likewise, PEP formation of intact chloroplasts in the light could be stimulated
by external application of Pi but not pyruvate, and presumably Pi serves both for export
of PEP and as a substrate for ATP formation needed as a substrate for PPDK (Huber and
Edwards 1977b). In Zea mays, the exchange of PEP and 3-PGA with PEP, 3-PGA, and Pi
but not pyruvate or malate and the exchange of 3-PGA for dihydroxyacetonephosphate
(DHAP) across intact chloroplast envelopes was demonstrated (Day and Hatch 1981). 3-
PGA could exchange with DHAP and the exchange of PEP for DHAP was not tested.
From the biochemical data it was concluded that all substrates are exchanged by the same
phosphate translocator protein. The first transporter capable of exchanging
phosphorylated sugars for phosphate was identified at the molecular level as a

triosephosphate phosphate translocator (TPT) in C3 chloroplasts of spinach (Flugge and

Heldt 1984). It accepts triosephosphates, 3-PGA, and P; but not PEP as substrates. Based

26

 

 

 

on earlier biochemical evidence with intact chloroplasts (Day and Hatch 1981; Rumpho
and Edwards 1984; Rumpho and Edwards 1985), tests were performed to determine
whether the TPTs of C4 plants, unlike those of the C3 plant spinach, were capable of
mediating PEP transport. The TPT from the C4 species Flaveria trinervia and Zea mays
both also accepted PEP in contrast to the spinach phosphate translocator, which has a low
affinity for PEP (Fischer et al. 1994). However, the comparison of transport rates of PEP
by intact chloroplasts of C4 species revealed substantial differences between the transport
capabilities of intact chloroplasts compared to the transport characteristics of C4TPTs
(Fischer et al. 1994). A phosphate translocator specific to PEP transport in exchange for
Pi, the phosphoenolpyruvate phosphate translocator (PPT), was identified on the
molecular level from several tissues and species, including maize endosperm, and was
characterized in detail from cauliflower bud envelopes (Fischer et al. 1997). When
heterologously expressed, this protein has a high affinity for PEP (Ki=300uM for
inhibition of Pi transport) whereas 3-PGA and triosephosphates are only poorly bound
and transported (Ki=8,000 and K1=4,600uM, respectively). RNA gel blot analysis showed
that the PPT identified in maize endosperm is expressed at high levels only in non-green
tissues (Fischer et al. 1997). Phosphate translocators are evolutionary ancient proteins,
and the split into the PPT and TPT subfamilies of transporters occurred already in the
common ancestor of the red and the green lineage (Weber et al. 2006). Consequentially,
PPTs are present in all plant and algal genomes sequenced to date. The genome of the C3
plant Arabidopsis thaliana harbors two PPT genes. The AtPPTl, like the protein from
spinach, accepts both PEP and 2-phosphoglycerate whereas AtPPT2 accepts PEP rather

than 2-phosphoglycerate (Knappe et al. 2003). In C3 plants, the PPT imports PEP into

27

 

 

chloroplasts as a substrate for the shikimate pathway since chloroplasts lack the activities
of enolase and phosphoglyceromutase that are required to produce PEP from 3-
phosphoglycerate; therefore PPT provides the substrate for aromatic amino acid
biosynthesis (Knappe et al. 2003; Voll et al. 2003). The import of PEP into chloroplasts is
a minor flux in C3 plants compared with the export of assimilates, for example, mediated
by TPT. PPT activity is also required in bundle sheath chloroplasts of C3 plants to
generate a signal that is required for proper mesophyll development (Streatfield et al.
1999; Voll et al. 2003). Recently, a PPT protein abundant in mesophyll chloroplast
envelopes and expressed leaf specifically was identified in maize (Bréiutigam et al.
2008a) and shown to be mesophyll specific (Majeran et al. 2008) but its activity has not
been tested biochemically. Compared to C3 chloroplasts envelopes, this PPT is vastly
increased in abundance indicating that the high PEP export rates are maintained at least in
part by increasing the amount of transport protein present at the envelope (Brautigam et
al. 2008a). Possibly, based on the results by Huber and Edwards (1977) outlined above,
the C4 PPT from Zea mays can also exchange 3-PGA for PEP, which is not the case for
any of the C3 PPTs characterized to date.

Oxaloacetate and Malate exchange in PCA type chloroplasts

After PEP is exported from the PCA chloroplast, it is carboxylated to OAA by PEP
carboxylase in the cytosol. Since carbon is shuttled to PCR cells in the form of malate in
NADP-ME type C4 plants, OAA and malate need to be exchanged across the chloroplast
envelope. Isolated chloroplasts from both C3 and C4 plants produce malate and evolve
oxygen when externally supplied with OAA (Heber 1974; Anderson and House 1979;

Day and Hatch 1981). In maize and spinach chloroplasts a high velocity OAA transporter

28

 

 

was characterized biochemically both by OAA-dependent oxygen evolution and
radiolabeled precursor uptake studies (Hatch et al. 1984a). In Zea mays PCA
chloroplasts, this transport protein binds OAA with an apparent KM between 53 and
71uM, depending on the method used for measuring, and the corresponding apparent
inhibitory binding constants for malate are K1=7,300uM and Kr=7,500uM. In spinach
chloroplasts, the KM(OAA) is 9uM and Ki(malate) is 1,400uM. The high affinity for
OAA compared to malate is crucial if, like in Zea mays, the malate pools are 10-100
times greater than those of OAA (Hatch et al. 1984a). The reaction equilibrium constant
of malate dehydrogenase favors the reaction in the direction of malate, rather than OAA
(Hatch et al. 1984a). Finally, the velocity of OAA transport in Zea mays is sufficiently
high to supply the C4 cycle metabolites and orders of magnitude higher than the velocity
of transport in C3 chloroplasts of spinach (Hatch et al. 1984a). In C3 plants, the
OAA/malate exchanger is hypothesized to be involved in a reducing equivalent shuttle
(Scheibe 2004; Scheibe et al. 2005). An OAA/malate exchanger working in concert with
malate dehydrogenases in the chloroplasts, the cytosol and the peroxisomes can balance
reducing power throughout the cell. The molecular identity of the OAA malate transport
protein has not been unequivocally established in either C3 plants or C4 plants. A small
family of transport proteins capable of transporting dicarboxylates (dicarboxylate
translocators, DiTs) was biochemically identified in spinach chloroplasts (Woo et al.
1987) and later identified at the molecular level (Weber et al. 1995). The protein initially
characterized transports malate, fumarate, succinate, and 2-oxoglutarate (2-OG), but not
glutamate and was named DiTl (Weber et al. 1995). Its transport characteristics fit well

with the model of two translocators which, working in concert, import 2-OG into the

29

chloroplast and export glutamate while the counter-substrate malate only cycles (Woo et
al. 1987). A second translocator, called DiT2, was also identified at the molecular level
and characterized (Taniguchi et al. 2002; Weber and Flugge 2002; Renne et al. 2003).
Proteins of the DiT2 family of transporters have a high affinity for glutamate and
aspartate in addition to 2-OG (Taniguchi et al. 2002; Renne et al. 2003). In C3 plants, a
knock down and a knock out for one member of each family have been analyzed. In
tobacco, a knockdown of the sole representative of the DiTl family of transport proteins
causes a photorespiratory phenotype with dramatic metabolic changes in precursors and
products of ammonia fixation as well as decreases in photosynthesis rate and sugar pools
(Schneidereit et al. 2006). In Arabidopsis, the knockout of one of the two members of the
DiT2 family results in a photorespiratory phenotype (Somerville and Ogren 1983; Renne
et al. 2003). Changes in OAA and malate pools or redox status have not been reported
(Renne et al. 2003; Schneidereit et al. 2006). All DiTs also transport OAA in vitro. The
affinity for OAA of both, C3 DiTl and C4 DiTl, is similar (Taniguchi et al. 2002;
Taniguchi et al. 2004). For the Arabidopsis protein the apparent KM for malate is 700uM
and the corresponding apparent Ki for OAA is 70uM (Taniguchi et al. 2002). For the
maize protein the KM for malate is 610uM and the corresponding Ki for OAA is 90 uM
(Taniguchi et al. 2004). It was proposed that DiTl is capable of exchanging OAA and
malate in vivo (Taniguchi et al. 2004), although the kinetic constants of recombinant
reconstituted DiTs are different from those determined with transport experiments with
intact isolated chloroplasts (Hatch et al. 1984a). The discrepancies between transport
experiments with intact chloroplasts and isolated proteins have yet to be resolved.

Proteorrric analysis of mesophyll and bundle sheath chloroplast membranes also indicates

30

DiTl may indeed be the OAA/malate exchanger. A DiTl homologue (called OMT in
maize) was enriched in mesophyll compared to bundle sheath chloroplast membranes
(Supplemental material (Majeran et al. 2008)).The proteome analysis resolves
controversial evidence for the expression pattern of this DiTl family protein which is
reported with different expression patterns in Zea mays and Sorghum bicolor (Renne et
al. 2003; Taniguchi et al. 2004; Sawers et al. 2007). Despite mounting evidence, the
differences between in vitro and in vivo data are unresolved and therefore it remains to be
determined whether DiTl exchanges OAA and malate in either C3 or C4 chloroplasts in
vivo (Table 2-1).

Malate import into PCR chloroplasts

Malate is produced in PCA cells from OAA, and it moves to PCR cells by diffusion
through plasmodesmata. The decarboxylation enzyme, NADP-ME, is localized in
chloroplasts and malate is imported into PCR chloroplasts (Figure 2-1). The transport
capacities of chloroplasts in PCR tissues are less well understood compared to
chloroplasts in PCA tissues. Most species with Kranz-type C4 photosynthesis have re-
enforced cell walls around the PCR cells, likely to prevent C02 from leaking. These cells
are therefore less amenable to isolation and characterization in comparison to PCA cells.
In Zea mays, malate import was studied as a function of CO2 fixation and as a function of
pyruvate generation (Boag and Jenkins 1985). Malate import was shown to be affected by
application of aspartate (Boag and Jenkins 1985) or aspartate and glutamate (Boag and
Jenkins 1986). Application of aspartate lowered the binding constant and increased the
maximal velocity of transport. The only candidate transport proteins currently known of

being capable of transporting malate into chloroplasts are the DiTs. DiT2 family proteins

31

from Zea mays (Taniguchi et al. 2004) and Sorghum bicolor (Renne et al. 2003) are
expressed at a higher level in the bundle sheath. A recent proteomic analysis identified a
DiT2 family member, ZmDCT2/3, as bundle sheath specific (Majeran et al. 2008).
However, DiTs are antiporters that exchange a dicarboxylic acid for malate (Weber et al.
1995; Taniguchi et al. 2002; Renne et al. 2003; Taniguchi et al. 2004). Therefore DiTs
cannot catalyze the net import of a C4acid in a uniport mode. The decarboxylation of
malate produces the three carbon monocarboxylate pyruvate. Pyruvate has been tested as
a counter-substrate for malate transport for a DiT2 protein from F Iaveria bidentis, a
NADP-ME species. It was shown that exchange of pyruvate with malate was negligible
in vitro (Renne et al. 2003). It is therefore unlikely that DiT2 proteins catalyze the
exchange of malate and pyruvate across PCR chloroplast envelopes (Renne et al. 2003).
It thus remains currently unknown how the net transport of C4acids across the chloroplast
envelopes of PCR cells is achieved.

Pyruvate Export from PCR chloroplasts

After import and decarboxylation of malate in PCR chloroplasts of NADP—ME
species, the resulting pyruvate is exported to allow the recycling of the initial carbon
acceptor in the PCA compartment. Pyruvate export (Figure 2-1) from PCR type
chloroplasts has only been studied in a NAD malic enzyme species, Panicum miliaceum,
in which the majority of pyruvate is generated in mitochondria. Pyruvate uptake into
those PCR chloroplasts was markedly different from pyruvate uptake into PCA
chloroplasts both in speed and binding affinity, and was not affected by light (Ohnishi
and Kanai 1987b). In NADP-ME species, pyruvate transport across the PCR type

chlorOplast envelope has not been studied (Flugge et al. 1985; Ohnishi and Kanai 1987a;

32

Aoki et al. 1992). Possibly, the high concentration of pyruvate that is generated by malate
decarboxylation drives the export of pyruvate out of the PCR chloroplasts. It remains to
be determined whether the export of pyruvate from the PCR chloroplasts is mediated by
the same or by a different transport protein than the one which imports pyruvate into
PCA chloroplasts.

Pyruvate import into PCA type chloroplasts

Finally, the pyruvate produced in the PCR compartment needs to be recycled to the
CO2 acceptor PEP in the PCA compartment (Figure 2-1). The overall gradient for
pyruvate between PCA and PCR tissue opposes the actual direction of transport (Stitt and
Heldt 1985), thus leading to the hypothesis that pyruvate is actively transported to
sequester it in PCA chloroplasts, hence displacing it from the equilibrium (Flugge et al.
1985). In C4 plants pyruvate import was characterized as a slow process in the dark as a
pyruvate anion symport in Digitaria sanguinalis (Huber and Edwards 1977a). Later,
light-driven active pyruvate transport was characterized in Zea mays (Fliigge et al. 1985)
and Panicum miliaceum (Ohnishi and Kanai 1987b). In Zea mays, light driven pyruvate
transport is dependent on proteins, and it is inhibited by protonophores (F lfigge et al.
1985). Transport can be initiated in the dark by applying a pH gradient between the
stroma of isolated chloroplasts and the external medium in vitro (Aoki et al. 1992). A
second, sodium-dependent mode of active pyruvate import into PCA chloroplasts was
discovered in Panicum miliaceum (Ohnishi and Kanai 1987a). Several other species, such
as Urochloa panicoides and Panicum maximum, but not Zea mays, also exhibit sodium-
dependent pyruvate transport (Ohnishi et a1. 1990). A systematic evaluation of more than

forty C4species revealed no correlation between the mode of pyruvate transport (i.e.

33

sodium- or proton-dependent) and biochemical C4 subtype (Aoki et al. 1992). In all
species tested, light driven pyruvate uptake into PCA type chloroplasts could be
mimicked either by a sodium or by a proton gradient in vitro but not by both for any
given species (Aoki et al. 1992).

In C3 plants, pyruvate transport was analyzed biochemically with isolated pea
chloroplasts. Pyruvate uptake followed saturation kinetics at low external concentrations
and linear kinetics at higher pyruvate concentrations. Based on these results carrier
mediated transport was proposed at low substrate concentrations and transport by
diffusion at high concentrations (Proudlove and Thurman 1981). Pyruvate transport into
the chloroplast of C3 species may be relevant when serving as a substrate for branched
chain amino acid biosynthesis or isoprenoid production (Singh and Shaner 1995;
Schwender et al. 1996) or for fatty acid production in certain tissues but not in seeds
(Andre and Benning 2007; Andre et al. 2007). At the molecular level, the protein or
proteins which catalyze pyruvate import remain unknown both in C3 and in C4 species.
Possibly, independent evolution of C4 photosynthesis recruited two different types of
pyruvate transport proteins, using either a sodium or a proton gradient as the driving
force. Passive diffusion of pyruvate across the chloroplast envelope appears unlikely
since the metabolite gradient for pyruvate between PCA and PCR tissue requires active
pyruvate import into PCA chloroplasts for C4 photosynthesis, and pyruvate transport is
dependent on intact proteins in the chloroplast envelope (Flugge et al. 1985). Establishing
either sodium or proton gradients to drive pyruvate transport across the chloroplast
envelope membrane of PCA cells requires the input of energy. To date, energy

requirements for driving metabolite transport have not been included in the overall ATP

34

balance required for driving the C4 biochemical CO2 pump. Molecular identification and
biochemical characterization of the respective transporters will be required to address this
question.
The knowledge about transport proteins involved in the C4cycle of NADP-ME
plants is summarized in Table 2-1.
Auxiliary transport processes
In addition to the C4cycle metabolites, pyruvate, PEP, OAA, and malate, several
other metabolites are also moved between PCA and PCR cells and compartments. The
C4cycle serves to enrich the PCR tissue in C02 and creates an environment that
minimizes photorespiration at the site of Rubisco. To further reduce photorespiration, in
some species of the NADP malic enzyme subtype, the partial pressure of O2 is
minimized. PCR tissue chloroplasts have minimal photosystem II activity (Leegood et al.
1983; Walker and Edwards 1983; Rumpho et al. 1987; Meierhoff and Westhoff 1993)
and consequentially minimal production of reducing equivalents through linear electron
transport. To supply the reactions in PCR tissue with reducing power, a reducing
equivalent shuttle is hypothesized to operate. This shuttle transports 3-PGA from PCR
chloroplasts (Figure 2-1) to PCA chloroplasts where 3-PGA is reduced to triosephosphate
and moved back to the PCR tissue. It has been proposed that in Zea mays, the PCR cycle
is compartmented between PCA and PCR chloroplasts with part of the Calvin cycle
relegated to the PCA chloroplasts (Majeran et al. 2005). The shuttling of 3-PGA across
two chloroplast envelopes into the PCA chloroplasts and the return of triosephosphate to
the PCR chloroplasts requires the concerted action of two triosephosphate phosphate

translocators. High activity of a transport protein exchanging 3-PGA, triosephosphates,

35

and P; has been demonstrated in PCA chloroplasts of Zea mays (Day and Hatch 1981;
Rumpho and Edwards 1985; Rumpho et al. 1987). In C3 plants the TPT is the most
abundant transport protein of the chloroplast envelope and it serves to export
triosephosphate from the chloroplast while inorganic phosphate is imported (Fli‘rgge and
Heldt 1984; F lfigge and Heldt 1991). A knock out in TPT causes massive accumulation
of transitory starch but no dramatic visible phenotype since the chloroplasts are able to
export assimilated carbon during the night in the form of glucose and maltose (Schneider
et al. 2002). The transport characteristics determined in vitro reveal that the TPT of
higher plants is also capable of exchanging 3-PGA for triosephosphates. It has been
proposed that TPT functions both as a reducing equivalent shuttle and as the exporter of
carbon in C4PCR chloroplasts (Flugge and Heldt 1991). It is unknown whether the
inverse direction of 3-PGA transport in comparison to C3 plants (i.e., export of 3-PGA
and import of DHAP) requires specific adaptations of the TPT protein, especially since
the stromal pH likely causes 3-PGA to have three negative charges.

Both enriching C02 and minimizing 02 in the vicinity of Rubisco reduces
photorespiration in C4 plants, yet it is not absent. Moreover, the separation of the
photorespiratory pathway into different tissues has been hypothesized to predate a true
C4cycle and indeed be an evolutionary intermediate of true C4 photosynthesis (reviewed
in Bauwe, 2009). It has been demonstrated that the final step of the photorespiratory
pathway, the phosphorylation of glycerate to 3-PGA by glycerate kinase, is localized
exclusively in PCA chloroplasts of different C4 plants belonging to all three subtypes
(U suda and Edwards 1980a). For maize, the localization was independently confirmed by

proteomics (Majeran et a1. 2005). In Panicum cappilare, it was demonstrated that the

36

PCR tissue photosynthetically produces glycerate which is metabolized in PCA tissue
(Usuda and Edwards 1980b). Photorespiration itself is a highly compartmentalized
process involving not only the chloroplasts but also peroxisomes and mitochondria
(Bauwe, 2008). Even in C3 plants, the transport proteins involved are unknown at the
molecular level although biochemical evidence for glycolate and glycerate transport
through the same transport protein (Howitz and McCarty 1986; Howitz and McCarty
1991; Young and McCarty 1993). as well as glycine and serine import into mitochondria
(Yu et al. 1983)exists. It is currently unknown whether the transport of photorespiratory
intermediates in C4 plants occurs through the same transport proteins as in C3 plants
especially since in C4 plants the export of glycolate from the chloroplast and the import
of glycerate into chloroplasts are located in different tissues.

Enzymes involved in lipid biosynthesis, nitrogen fixation, tetrapyrrol and
isoprenoid biosynthesis accumulate preferentially in mesophyll chloroplasts of Z. mays
whereas enzymes for sulfirr import accumulate preferentially in bundle sheath
chloroplasts (Majeran et al. 2005). Presumably, the preferential localization of pathways
necessitates transport of the pathway products to the other compartment but specific
adaptations of transport proteins are not known to date.

Finally, all metabolites entering the chloroplasts do not only need to cross the
specificity barrier, the inner envelope, but also the outer envelope. It has recently been
demonstrated that specific outer envelope porins are increased in abundance in Z. mays
chloroplast envelopes compared to Pisum sativum chloroplast envelopes (Brautigam et al.
2008a). The outer envelope porins OEP37 and OEP24 were increased in abundance and

OEP21 was decreased. Apparently, the increased amount of metabolite exchange across

37

the outer envelope is mastered by an increase in specific outer envelope porins
demonstrating the importance of the outer envelope for total metabolite traffic to the

chloroplast (Brautigarn et al. 2008a).

Transport processes in the NAD malic enzyme type

 

 

r W

I \‘\“> Asp /—\Asp

OAA 11 fl

 

 

 

Ala < —————— Ala ma'a‘e

pyruvate l I

pyruvate pyruvate N. pyruvate
t t

 

   

 

 

Figure 2-2: Scheme outlining the reactions of C4 photosynthesis and the connecting
transport processes for the NAD malic enzyme type of C4 photosynthesis in both the PCA
(left) and the PCR (right) compartment; the cell wall is represented by a grey bar; intracellular
transport by arrows; intercellular transport by dashed arrows; metabolic reactions by broad
arrows; C chloroplast, M mitochondrion; reactions processes which are not resolved are
represented by open arrows; for additional abbreviations see text;

The operation of the C4cycle in the NAD malic enzyme (NAD-ME) subtype of
C4photosynthesis is compartmentalized between the PCA chloroplasts and cytosol and
the PCR mitochondria (Figure 2-2). As is the case in the NADP—ME subtype, the initial
carbon acceptor PEP is exported from the chloroplasts and carboxylated in the cytosol.
The carboxylation product is OAA, which is subsequently transaminated in the cytosol to
yield aspartate. The C4amino acid aspartate then moves to the PCR tissue. The

compartmentation of the subsequent transamination to OAA is unresolved (Figure 2-2).

38

Aspartate may either be imported into the mitochondria and transaminated in the
mitochondrial matrix. Alternatively, it may be transaminated to OAA in the cytosol
before it is imported. OAA is reduced to malate, probably within the mitochondria (Hatch
1987; Edwards et al. 2001a), although the localization of the corresponding malate
dehydrogenase activity has not been demonstrated. Malate is subsequently
decarboxylated to yield pyruvate, C02, and NADH. The resulting CO2 enters the PCR
cycle where it is fixed by Rubisco. There are no specific transport proteins for either C02
or HCO3‘ at either the mitochondrial or the chloroplast envelopes known to date.
Pyruvate is exported from mitochondria and transaminated to alanine in the cytosol. If the
transamination of aspartate to OAA is localized in the mitochondria (Figure 2-2), a 2-
oxoglutarate (2-OG) glutamate shuttle is required to connect the transamination of
alanine in the cytosol with the one of aspartate in the mitochondria via the amino-donor/ -
acceptor pair, glutamate and 2-OG. If the transaminations of aspartate to OAA and
pyruvate to alanine both occur in the cytosol (Figure 2-2), they can either be directly
coupled or connected by the amino-donor/ -acceptor pair 2-OG and glutamate. Alanine
moves back to the PCA cytosol where it is converted to pyruvate in a transamination
reaction, and pyruvate is imported back into PCA chloroplasts where the carbon acceptor
PEP is regenerated by the action of pyruvate phosphate dikinase (reviewed in Hatch
( 1987)). In addition to this major route, there is a minor pathway in which malate is
formed in and exported from PCA chloroplasts and decarboxylated in PCR mitochondria
but this pathway accounts for less than 10% of the total C4 acids moved (Kagawa and

Hatch 1975; Hatch et al. 1988). This C4 subtype has also evolved multiple times

39

independently and includes species from the genera Amaranthus, Cleome, Digitaria, and

Atriplex (Sage 2004; Muhaidat et al. 2007).

Table 2-2: Summary of transport processes necessary in NAD-ME plants, for abbreviations

 

see text
domain substrate co- name evidence references
substrate
. - Huber and Edwards
PCA PEP p- PPT biochemical and (1977b); Fischer et al.
1 molecular (1997)
inferred from
aspartate ? ? presumed
localization of
PCR .
soluble proteins
OAA p- DIC mOIeCUIa’ '" C3 Palmieri et al. (2008)
I plants
inferred from
PCR pyruvate ? ? localization of -
soluble proteins
H+ ? biochemical Aoki et al. (1992)
PCA pyruvate
Na+ ? biochemical Aoki et al. (1992)

 

PEP export from PCA chloroplasts
Like in NADP-ME plants the primary carbon acceptor PEP is recycled in PCA
chloroplasts by PPDK (Hatch 1987). PEP export is thought to be catalyzed by a member
of the same transport protein class, the PPTs, in NAD-ME plants. A detailed description
can be found in IIIA.

Dicarboxylate transport in PCR mitochondria
PEP is carboxylated and the resulting OAA transaminated to aspartate in the cytosol of
PCA cells. Aspartate diffuses into the PCR domain. Unlike in NADP-ME plants, in
NAD-ME plants CO2 is liberated from malate in the mitochondria (Figure 2-2). It
remains unclear whether aspartate or OAA is the C4acid imported into mitochondria

(Figure 2-2) but like in chloroplasts of NADP-ME plants a net import and not exchange

40

of C4 acids is necessary. Generally, it is assumed that aspartate is the imported C4 acid in
NAD-ME plants (Hatch 1987; Hatch et al. 1988; Edwards et al. 2001b; Hatch 2002)
although the metabolite fluxes across the mitochondrial membrane have not been
established for any NAD-ME species. In C3 plants, metabolite transport of malate, OAA,
aspartate, 2-OG, and glutamate across the mitochondrial membrane has been
demonstrated (Desantis et al. 1976; Zoglowek et al. 1988; Hanning et al. 1999) and
several proteins catalyzing di- and tricarboxylate exchange have been characterized
biochemically and at the molecular level in C3 plants (Laloi 1999; Picault et al. 2002;
Picault et al. 2004; Palmieri et al. 2008). The di- and tricarboxylate carrier (DTC)
transports a broad spectrum of organic acids. While 2-OG, malate, succinate, OAA,
citrate, isocitrate, and sulfate are transported at high levels by DTC from both tobacco
and Arabidopsis, both transport proteins do not accept pyruvate or glutamate as a
substrate (Picault et al. 2002). DTC depends on a strict counter exchange of two
molecules with each other (Picault et al. 2002).The malate transport protein initially
identified as the malate transporter from Panicum miliaceum (Taniguchi and Sugiyama
1996) is an ortholog of the DTCs characterized from the C3 plants Arabidopsis and
tobacco (Picault et al. 2002). Although the spatial and temporal expression pattern of this
transporter agrees well with a role in C4photosynthesis (Taniguchi and Sugiyama 1997),
the substrate specificity of DTC makes it an unlikely candidate for a major role in
C4photosynthesis. The dicarboxylate carrier (DIC) transports a similar spectrum of
dicarboxylic acids. DICs from the C3 plant Arabidopsis transport 2-OG, OAA, succinate,
malate, phosphate, and sulfate (Palmieri et al. 2008). Like DTCs, DICs do not accept

three carbon organic acids such as PEP and pyruvate, or amino acids such as glutamate or

41

aspartate (Palmieri et al. 2008). In contrast to DTC, the DIC is capable of transporting
inorganic anions such as phosphate and DIC can therefore catalyze the net import of a C4
acid into the mitochondrial matrix by exchanging phosphate for a dicarboxylic acid
(Palmieri et al. 2008). The transport creates a phosphate imbalance which could be
compensated by a phosphate/proton transporter in the mitochondrial inner membrane
(McIntosh and Oliver 1994). Neither DTC nor DIC accepts glutamate or aspartate as a
substrate. Based on the transport specificities of the proteins known to transport
dicarboxylates, we posit that OAA is the main metabolite imported into mitochondria in
NAD-ME plants.
Pyruvate Export from PCR mitochondria

Like the import of the C4 acid into mitochondria for decarboxylation, the export of the
resulting C3 acid, pyruvate, has not been characterized biochemically in C4 plants of the
NAD-ME type (Figure 2-2). In C3 plants, pyruvate is imported into rather than exported
from mitochondria since it is one of the substrates for mitochondrial respiration. Isolated
mitochondria of the C3 plant pea import pyruvate with saturation kinetics (Proudlove and
Moore 1982) and the transport depends on a pH gradient, as protonophores efficiently
inhibit transport (Proudlove and Moore 1982). The molecular identity of the
mitochondrial pyruvate carrier from plants and other eukaryotes is unknown. Free
diffusion of protonated pyruvic acid is an alternative to carrier mediated transport. Small
organic acids including pyruvic acid can diffuse through biomembranes (Baker and
Vandam 1974; Proudlove and Thurman 1981; Benning 1986). Possibly, the irreversible
decarboxylation of malate creates sufficient amounts of pyruvate within C4 PCR

mitochondria to drive export by free diffusion of pyruvate, which is further promoted by

42

active uptake of pyruvate into C4PCA chloroplasts (Fliigge et al. 1985). Alternatively,
pyruvate might be transaminated to alanine in the mitochondria already and the neutral
amino acid may be exported since neutral amino acids are known to permeate the
mitochondrial membrane without specific transport mechanisms (Halling et al. 1973;
Wiskich 1977).
Pyruvate import into PCA chloroplasts

After pyruvate is exported from PCR mitochondria and has moved to the PCA
compartment, it reenters the PCA chloroplasts to serve as the substrate for PEP

regeneration. For a detailed discussion of the transport protein involved, see earlier.

In addition to the metabolites involved in the C4 cycle of NAD-ME plants
(summarized in Table 2-2) several additional metabolites likely have to be moved. The
liberation of C02 in the mitochondria and the localization of Rubisco in the chloroplasts
possibly results in the need for an inorganic carbon transport protein at either or both
organellar membranes. Unlike in NADP-ME plants (Majeran et al. 2008), knowledge
about the spatial distribution of reactions like nitrogen and sulfur fixation or isoprenoid
biosynthesis in NAD-ME species is limited and consequently, no specific hypothesis

about additional transport proteins can be put forward.

43

Transport processes in the PEP carboxykinase (PEP-CK) type

 

 

r W

 

 
  

 

 

   

 

 

. M
malate PI Pi
mama-44“
' 7" malate malate
OAA Asp ------- -> Asp
I I pyruvat
OAA OAA ATP ATP
pi P- m ADP: 8:14:39
PEP <- ------- PEP
Pij: :84 Ala *- ------- Atla
pyruvate pyruvate pyruvate <—
u L J

 

 

 

Figure 2-3: Scheme outlining the reactions of C4 photosynthesis and the connecting
transport processes for the PEP-CK type of C4 photosynthesis in both the PCA (left) and the
PCR (right) compartment; bold print represents the C4 cycle of PEP-CK, normal print represents
the C4 cycle of NAD-ME; the cell wall is represented by a grey bar; intracellular transport is
represent by arrows; intercellular transport by dashed arrows; metabolic reactions by broad
arrows; C chloroplast, M mitochondrion, for additional abbreviations see text;

The core reactions of the C4 cycle involving PEP-CK as the decarboxylation
enzyme require the least amount of transport processes (Figure 2-3, black part) (Hatch
1987). In this C4 cycle, the initial carbon acceptor PEP is in part recycled in the cytosol
of the PCR tissue. In the PCA cytosol, PEP is carboxylated to OAA, which is
subsequently transaminated to aspartate. This C4amino acid moves to the PCR tissue
where another transamination takes place and the resulting OAA is decarboxylated by
PEP-CK in the cytosol. The C02 produced enters the PCR cycle of the chloroplasts
possibly by free diffusion; whereas PEP moves back to the PCA tissue where it can again
serve as the primary carbon acceptor. The cycling metabolites aspartate and PEP

transport one aminogroup to PCR cells during each turn of the cycle. The activity of PEP-

44

CK in the cytosol requires ATP, which is produced in the mitochondria through NADH
oxidation and exported from the mitochondria in counter-exchange with ADP (Hatch
1987; Hatch et al. 1988) (Figure 2-3). It has been demonstrated that PEP-CK is not the
sole decarboxylation enzyme in PEP-CK type plants. NAD malic enzyme is proposed to
contribute about equally to C4 photosynthesis (Figure 2-3, grey part) (Bumell and Hatch
1988a; Bumell and Hatch 1988b; Hatch et al. 1988). Malate is probably the C4acid for
this part of the cycle. It is probably produced in PCA chloroplasts and exported in a
manner similar to that in NADP-ME plants (Figure 2-2and 2-3). After diffusing to the
PCR compartment, malate enters the mitochondria for decarboxylation. The C3 acid
produced through NAD malic enzyme in the mitochondria is exported, either as pyruvate
or possibly after being transaminated to alanine, and re-imported as pyruvate into PCA
chloroplasts to serve as the substrate for recycling the primary carbon acceptor PEP. The
cycling metabolites malate and alanine transport one amino group to the PCA tissue in
each cycle and therefore balance the amino group moved by aspartate and PEP if both
pathways contribute about 50% of the total decarboxylation reactions. The C4 cycle
involving PEP-CK does not rely on intracellular transport except for the provision of
ATP, but the C4 cycle involving the NAD-ME in mitochondria relies heavily on

intracellular transport.

45

Table 2-3: Summary of transport processes necessary in PEP-CK plants, for abbreviations

see text

domain

PCA

PCA

PCR

PC R

PC R

PCR

PCA

substrate

PEP

OAA

malate

pyruvate

ATP

ATP

pyruvate

CO-

substrate

Pi

malate

ADP

ADP

Na

name

PPT

OAT

DiT

DIC

NTT
(chloroplasts)

AAC
(mitochondria)

PEP export from PCA chloroplasts

evidence

inferred from
NADP-ME and
NAD-ME plants
inferred from
NADP-ME and
NAD-ME plants
inferred from
NADP-ME and
NAD-ME plants
molecular in C3
plants
inferred from
localization of
soluble proteins

molecular in C3
plants

molecular in C3
plants

inferred from
NADP-ME and
NAD-ME plants

inferred from
NADP-ME and
NAD-ME plants

references

Palmieri et al. (2008)

Neuhaus et al. (1997)

Emmermann et al.
(1991); Winning et al.
(1991)

PEP export is reduced in PEP-CK plants by at least half (Bumell and Hatch 1988b;

Bumell and Hatch 1988a; Hatch et al. 1988) compared to other C4 species since part of

the PEP is recycled from the cytosol of PCR tissue where it is generated by PEP-CK.

PEP recycled in the PCA chloroplasts by PPDK is likely exported through the PPT

described earlier.

46

Oxaloacetate malate exchange in PCA chloroplasts

Up to 50% of the fixed carbon is transported to the PCR tissue as malate (Bumell
and Hatch 1988b; Bumell and Hatch 1988a; Hatch et al. 1988) which is probably
produced in PCA chloroplast from OAA. The candidate transport protein is described in
detail earlier.

Malate import into PCR mitochondria

Like in NAD-ME plants part of the C02 liberated in PCR cells is produced in the
mitochondria by NAD dependent malic enzyme (Bumell and Hatch 1988b; Bumell and
Hatch 1988a; Hatch et al. 1988). Since malate is the C4 acid transported to PCR cells no
transamination reactions are required and malate can directly be imported in counter-
exchange with an inorganic anion such as phosphate or sulfate. For a detailed analysis of
the transport process, see earlier.

ATP/ADP translocation to supply PEP-CK with ATP

The import of malate may not only serve to supply NAD-ME with its substrate but
also to drive the tricarboxylic acid cycle which produces ATP for the cell. The ATP
demand of PCR cells in PEP-CK plants is high since PEP-CK uses one molecule of ATP
to liberate CO2 and thus uses ATP at the rate of carbon fixation. The mitochondrial
ATP/ADP carrier mediates the exchange of ATP and ADP across the mitochondrial
membrane (Emmermann et al. 1991; Winning et al. 1991). Alternatively, ATP may be
derived from cyclic electron flow and transported by the plastidic ATP/ADP carrier to the
cytosol (Neuhaus et al. 1997; Weber 2004).

The transport processes in PEP-CK plants are summarized in Table 2-3.

47

Transport processes in single cell C4metabolism

Plants which operate the C4 cycle within one cell do not require intercellular
transport through plasmodesmata. In these plants, the initial carbon fixation by PEPC and
the subsequent decarboxylation and refixation by Rubisco are spatially separated within
one cell, which underscores the importance of intracellular transport processes necessary
for enriching C02 at the site of Rubisco. Examples include Bienertia cycloptera, Suada
aralocaspica (formerly Borszczowia aralocaspica) and Hydrilla verticillata (Magnin et
al. 1997; Voznesenskaya et al. 2001; Voznesenskaya et al. 2002). In B. cycloptera, the
PCA compartment consists of grana-deficient chloroplasts in the cytoplasm around the
periphery of the cell which is devoid of mitochondria. This PCA compartment is
separated by the vacuole from the central PCR compartment, which contains
mitochondria and typical granal chloroplasts. Both compartments are connected by
cytoplasmic channels, which cross the vacuole. B. cycloptera is a NAD malic enzyme
type plant (Voznesenskaya et al. 2002) and requires all intracellular transport steps
described for typical NAD malic enzyme type plants with Kranz anatomy. In S.
aralocaspica, the PCA and PCR compartments localize to the proximal and distal ends of
the same cell, respectively, and are separated by a large vacuole. Again, mitochondria are
localized to the PCR compartment only and, like B. cycloptera, S. aralocaspica is a NAD
malic enzyme species (Voznesenskaya et al. 2001). H. verticillata is a facultative C4
species in which C4 photosynthesis of the NADP malic enzyme type is induced under
C02 limiting conditions (Magnin et al. 1997). Although single cell C4 plants do not need
to transfer metabolites symplastically between different cells, they have to move

metabolites efficiently from the PCA domain to the PCR domain and back. It is currently

48

not known whether metabolite flow is assisted by specialized structures within the cell.

The intracellular transport requirements are the same as in C4 plants with Kranz anatomy.
Future prospects

Discovering the molecular identity of C4adapted transport proteins

For virtually all transport proteins the molecular identity in C4 species is unknown
although there are a ntunber of good candidates which can be inferred from proteins
characterized in C3 systems. For example, ever since it was discovered in Zea mays
endosperm tissue and characterized from cauliflower, PPT has been assumed to be the
exporter of the primary carbon acceptor PEP. A PPT isoforrn expressed at moderate to
high levels in C4 leaves was recently discovered by proteomic analysis of maize
mesophyll chloroplast envelope membranes (Bréiutigam et al., 2008). Where candidate
transport proteins have been identified it is far from clear how the high metabolite flow
across the envelope is sustained by the transport protein. Possibly, the C4 isoforms of
transport proteins have altered kinetic characteristics compared to C3 homologues or the
increased flow is simply accomplished by increased amounts of transport proteins,
thereby increasing the Vmax of transport.

Most of the biochemical characterizations of organelle transport proteins have been
limited to one C4 species, Zea mays. New tools relying on collecting data from non-
model species in a high throughput manner may allow studies with the goal of
discovering and characterizing the remaining transport proteins. Transcriptomics
approaches, if they involve de novo sequencing of cDNAs, show great promise in both
generating sequence information for downstream applications (Brautigam et al. 2008b) as

well as generating quantitative information on transcript abundance (Weber et al. 2007).

49

After sequence information has been generated, proteomics of organellar membranes and
the soluble proteomes can reveal specific adaptations of C4 organelles to sustain transport
capacity (Majeran et al. 2005; Brautigam et al. 2008a; Majeran et al. 2008). The
comparison of data from different C4 species should provide information on differences
and similarities in transport properties of specific subtypes of C4 photosynthesis. Since
the different C4subtypes employ different enzymes as well as different transport proteins,
the identification and characterization of molecular candidate proteins for many of the
metabolite fluxes is likely possible. In this sense, maybe the question of which model
system best serves to study the C4 syndrome, maize, Flaveria, or Cleome (Brown et al.
2005), could be best answered with Zea mays, F laveria species and Cleome, possibly also
including in addition a PEP-CK C4photosynthesis plant and one or several single-cell C4
species. Detailed knowledge about the transport proteins and their specific adaptations
will assist in engineering C4 photosynthesis-like metabolism in crop plants, irrespective
as to whether the engineering aims to generate single-cell or multiple cell
C4photosynthesis. The discovery and characterization of novel metabolite transport
proteins may also inform studies of metabolite transport in C3 species. For example, the
identification of a pyruvate transporter may help to understand the contributions of
pyruvate to fatty acid synthesis during seed filling and to branched chain amino acid
biosynthesis, fatty acid biosynthesis, and isoprenoid biosynthesis in leaf tissues.

Prospects for engineering C4photosynthesis into C3 crop species

Metabolite flow across the membrane has been shown to be limiting for metabolic
pathways (Hausler et al. 2000; Zhang et al. 2008).For example, antisense repression of

TPT in tobacco leads to over-accumulation of transitory starch in chloroplasts during the

50

day since the export of triosephosphate is compromised (Hausler et al. 2000). A
simultaneous increase of glucose-6-phosphate and ATP import into potato tuber
amyloplasts by overexpression of both, the glucose-6-phosphate and ATP translocators
increases sink strength and yield in potatoes because carbon can be relocated to the
amyloplasts with greater efficiency (Zhang et al. 2008). Both examples illustrate how
metabolite flow through a pathway can be restricted or increased by decreasing and
increasing metabolite flow across membranes. Numerous studies aiming to engineer or
alter metabolic pathways have uncovered substrate availability as one of the major
limitations of pathway engineering (reviewed in Kunze et al. (2002)). Since metabolite
flow across membranes during C4 photosynthesis exceeds all metabolite flows across
membranes known in C3 plants, increasing membrane transport capacity will play a key
role in engineering C4photosynthesis in C3 crops. The over-expression of PPDK alone in
C3 species such as Arabidopsis, potato, and rice has minimal physiological impacts
despite over-expression levels of up to 40-fold have been achieved in rice (Fukayama et
al. 2001; reviewed in Matsuoka et al. (2001) and in Miyao (2003)). It has been discussed
that the limited impact of PPDK overexpression is due to free reversibility of the reaction
(Bumell and Hatch 1985) in combination with low activities of inorganic
pyrophosphatase and adenylate kinase in C3 plants (Matsuoka et al. 2001).However, the
limitation may well be both the lack of pyruvate import as well as PEP export from the
chloroplast, which limit substrate availability and cause product accumulation. To
substantially increase the metabolite flow through PPDK both pyruvate import and PEP
export need to be increased since both pyruvate and PEP transport are minor activities in

C3 plants and the chloroplast envelope is not adapted to allow major changes without

51

genetic modifications. Recently, single cell C4 metabolism has been considered an
alternative model system for engineering C3 plants to perform C4 photosynthesis since
its creation would avoid major alterations to leaf architecture and cell-to-cell
connectivity. Intracellular transport, however, is as crucial to single cell C4

photosynthesis as it is to C4 photosynthesis in species with Kranz anatomy.

Acknowledgements

Work in the author’s lab is supported by grants of the National Science Foundation
(USA) and the German Research Foundation (DFG, Germany). AB is grateful to the
Barnett-Rosenberg Foundation and the Deutsche Studienstiftung for financial support.

The authors wish to thank Dr. R. Sage and two anonymous reviewers for helpful

comments on improving the manuscript.

52

References

Anderson J W, House CM (1979) Polarographic Study Of Oxaloacetate Reduction By
Isolated Pea-Chloroplasts Plant Physiol 64, 105 8-1063

Andre C, Benning C (2007) Arabidopsis seedlings deficient in a plastidic pyruvate kinase
are unable to utilize seed storage compounds for germination and establishment Plant
Physiol 145, 1670-1680

Andre C, Froehlich J E, Moll MR, Benning C (2007) A heteromeric plastidic pyruvate
kinase complex involved in seed oil biosynthesis in Arabidopsis Plant Cell 19, 2006-2022

Aoki N, Ohnishi J, Kanai R (1992) 2 Different Mechanisms For Transport Of Pyruvate
Into Mesophyll Chloroplasts Of C-4 Plants-A Comparative-Study Plant Cell Physiol 33,
805-809

Bakker EP, Vandam K (1974) Movement Of Monocarboxylic Acids Across Phospholipid
Membranes - Evidence For An Exchange Diffusion Between Pyruvate And Other
Monocarboxylate Ions Biochim Biophys Acta 339, 285-289

Benning C (1986) Evidence Supporting A Model Of Voltage-Dependent Uptake Of
Auxin Into Cucurbita Vesicles Planta 169, 228-237

Benz R (1994) Permeation Of Hydrophilic Solutes Through Mitochondrial Outer
Membranes - Review On Mitochondrial Porins Biochim Biophys Acta 197, 167-196

Boag S, Jenkins CLD (1985) C02 Assimilation And Malate Decarboxylation By Isolated
Bundle Sheath Chloroplasts From Zea-Mays Plant Physiol 79, 165-170

Boag S, Jenkins CLD (1986) The Involvement Of Aspartate And Glutamate In The
Decarboxylation Of Malate By Isolated Bundle Sheath Chloroplasts From Zea-Mays
Plant Physiol 81, 115-119

Bolter B, 8011 J, Hill K, Hemmler R, Wagner R (1999) A rectifying ATP-regulated solute
channel in the chloroplastic outer envelope from pea EMBO J 18, 5505-5516

Bréiutigam A, Hofmann-Benning S, Weber APM (2008a) Comparative Proteomics of
Chloroplast Envelopes from C3 and C4 Plants Reveals Specific Adaptations of the

Plastid Envelope to C4 Photosynthesis and Candidate Proteins Required for Maintaining
C4 Metabolite Fluxes Plant Physiol 148, 568-579

Brautigam A, R.P. S, Whitten D, Wilkerson CG, Carr KM, Froehlich JE, Weber APM
(2008b) Comparison of the use of a species-specific database generated by

53

pyrosequencing with databases from related species for proteome analysis of pea
chloroplast envelopes J Biotechnol 136, 44-53

Brown NJ, Parsley K, Hibberd JM (2005) The futured C-4 research - Maize, Flaveria or
Cleome? Trends Plant Sci 10, 215-221

Bumell JN, Hatch MD (1985) Light Dark Modulation Of Leaf Pyruvate, P(I) Dikinase
Trends Biochem Sci 10, 288-291

Bumell JN, Hatch MD (1988a) Photosynthesis In Phosphoenolpyruvate Carboxykinase-
Type-C4 Plants - Pathways Of C-4 Acid Decarboxylation In Bundle Sheath-Cells Of
Urochloa-Panicoides Arch Biochem Biophys 260, 187-199

Bumell JN, Hatch MD (1988b) Photosynthesis In Phosphoenolpyruvate Carboxykinase-
Type-C4 Plants - Photosynthetic Activities Of Isolated Bundle Sheath-Cells From
Urochloa-Panicoides Arch Biochem Biophys 260, 177-186

Craig S, Goodchild DJ (1977) Leaf Ultrastructure Of Triodia-Irritans - C4 Grass
Possessing An Unusual Arrangement Of Photosynthetic Tissues Aust J Bot 25, 277-290

Day DA, Hatch MD (1981) Transport Of 3-Phosphoglyceric Acid, Phosphoenolpyruvate,
And Inorganic-Phosphate In Maize Mesophyll Chloroplasts, And The Effect Of 3-
Phosphoglyceric Acid On Malate And Phosphoenolpyruvate Production Arch Biochem
Biophys 211, 743-749

Desantis A, Arrigoni O, Palmieri F (1976) Carrier-Mediated Transport Of Metabolites In
Purified Bean Mitochondria Plant Cell Physiol 17, 1221-1233

Ding B, Turgeon R, Parthasarathy MV (1992) Substructure Of Freeze-Substituted
Plasmodesmata Protoplasma 169, 28-41

Edwards GE, Franceschi VR, Ku MSB, Voznesenskaya EV, Pyankov VI, Andreo CS

(20013) Compartmentation of photosynthesis in cells and tissues of C4 plants J Exp Bot
52, 577-590

Edwards GE, Furbank RT, Hatch MD, Osmond CB (2001b) What Does It Take to Be
C4? Lessons from the Evolution of C4 Photosynthesis Plant Physiol 125, 46-49

Emmermann M, Braun HP, Schmitz UK (1991) The ADP ATP Translocator From Potato
Has A Long Amino-Terminal Extension Curr Genet 20, 405-410

Erwee MG, Goodwin PB (1985) Symplast Domains In Extrastellar Tissues Of Egeria-
Densa Planch Planta 163, 9-19

54

Evert RF, Eschrich W, Heyser W (1977) Distribution And Structure Of Plasmodesmata
In Mesophyll And Bundle-Sheath Cells Of Zea-Mays-L Planta 136, 77-89

Fischer K, Arbinger B, Kammerer B, Busch C, Brink S, Wallmeier H, Sauer N,
Eckerskom C, Flugge UI (1994) Cloning and in-Vivo Expression of Functional Triose
Phosphate/Phosphate Translocators from C-3-Plants and C-4-Plants - Evidence for the
Putative Participation of Specific Amino-Acid-Residues in the Recognition of
Phosphoenolpyruvate Plant J 5, 215-226

Fischer K, Kammerer B, Gutensohn M, Arbinger B, Weber A, Hausler RE, Fli'rgge U]
(1997) A new class of plastidic phosphate translocators: A putative link between primary

and secondary metabolism by the phosphoenolpyruvate/phosphate antiporter Plant Cell 9,
453-462

F liigge UI, Heldt HW (1984) The Phosphate-Triose Phosphate-Phosphoglycerate
Translocator of the Chloroplast Trends Biochem Sci 9, 530-533

F liigge UI, Heldt HW (1991) Metabolite Translocators of the Chloroplast Envelope Annu
Rev Plant Physiol Plant Mol Biol 42, 129-144

Fliigge UI, Stitt M, Heldt HW (1985) Light-Driven uptake of Pyruvate into mesophyll
Chloroplasts from maize FEBS Lett 183, 335-339

Fukayarna H, Tsuchida H, Agarie S, Nomura M, Onodera H, Ono K, Lee BH, Hirose S,
Toki S, Ku MSB, Makino A, Matsuoka M, Miyao M (2001) Significant accumulation of

C-4-specific pyruvate,orthophosphate dikinase in a C-3 plant, rice Plant Physiol 127,
1136-1146

Furbank RT, Hatch MD (1987) Mechanism Of C-4 Photosynthesis - The Size And
Composition Of The Inorganic Carbon Pool In Bundle Sheath-Cells Plant Physiol 85,
958-964

Goetze TA, Philippar K, Ilkavets I, 8011 J, Wagner R (2006) OEP37 Is a New Member of
the Chloroplast Outer Membrane Ion Channels J Biol Chem 281, 17989-17998

Hailing PJ, Brand MD, Chappell J B (1973) Permeability Of Mitochondria To Neutral
Amino-Acids FEBS Lett 34, 169-171

Hanning I, Baumgarten K, Schott K, Heldt HW (1999) Oxaloacetate transport into plant
mitochondria Plant Physiol 119, 1025-1031

Hatch M (2002) C4 photosynthesis: discovery and resolution Photosyn Res 73, 251-256

Hatch M, Droscher L, Fliigge UI, Heldt HW (1984a) A specific Translocator for
Oxaloacetate Transport in Chloroplasts F EBS Lett 178, 15-19

55

Hatch MD (1987) C-4 Photosynthesis - a Unique Blend of Modified Biochemistry,
Anatomy and Ultrastructure Biochim Biophys Acta 895, 81-106

Hatch MD, Agostino A, Bumell JN (1988) Photosynthesis In Phosphoenolpyruvate
Carboxykinase-Type C-4 Plants - Activity And Role Of Mitochondria In Bundle Sheath-
Cells Arch Biochem Biophys 261, 357-367

Hatch MD, Droscher L, F lfigge UI, Heldt HW (1984b) A Specific Translocator for
Oxaloacetate Transport in Chloroplasts FEBS Lett 178, 15-19

Hattersley PW, Browning AJ (1981) Occurrence Of The Suberized Lamella In Leaves Of
Grasses Of Different Photosynthetic Types]. In Parenchymatous Bundle Sheaths And
Pcr (Kranz) Sheaths Protoplasma 109, 371-401

Hausler RE, Schlieben NH, Flugge UI (2000) Control of carbon partitioning and
photosynthesis by the triose phosphate/phosphate translocator in transgenic tobacco
plants (Nicotiana tabacum). 11. Assessment of control coefficients of the triose
phosphate/phosphate translocator Planta 210, 383-390

Heber U (1974) Metabolite Exchange Between Chloroplasts And Cytoplasm Annu Rev
Plant Physiol Plant Mol Biol 25, 393-421

Henderson SA, Voncaemmerer S, Farquhar GD (1992) Short-Term Measurements Of
Carbon Isotope Discrimination In Several C4 Species Aust J Plant Physiol 19, 263-285

Howitz KT, McCarty RE (1986) D-Glycerate Transport By The Pea Chloroplast
Glycolate Carrier - Studies On [1-C-14] D-Glycerate Uptake And D-Glycerate
Dependent O-2 Evolution Plant Physiol 80, 390-395

Howitz KT, McCarty RE (1991) Solubilization, Partial-Purification, And Reconstitution
Of The Glycolate Glycerate Transporter From Chloroplast Inner Envelope Membranes
Plant Physiol 96, 1060-1069

Huber SC, Edwards GE (1977a) Transport in C4 Mesophyll Chloroplasts -
Characterization of Pyruvate Carrier Biochim Biophys Acta 462, 583-602

Huber SC, Edwards GE (1977b) Transport in C4 Mesophyll Chloroplasts - Evidence for
an Exchange of Inorganic-Phosphate and Phosphoenolpyruvate Biochim Biophys Acta
462, 603-612

Kagawa T, Hatch MD (1975) Mitochondria As A Site Of C4 Acid Decarboxylation In
C4-Pathway Photosynthesis Arch Biochem Biophys 167, 687-696

56

Knappe S, Lottgert T, Schneider A, Vol] L, F li‘rgge UI, Fischer K (2003) Characterization
of two functional phosphoenolpyruvate/phosphate translocator (PPT) genes in
Arabidopsis-AtPPTl may be involved in the provision of signals for correct mesophyll
development Plant J 36, 411-420

Kubasek J, Setlik J, Dwyer S, Santrucek J (2007) Light and grth temperature alter
carbon isotope discrimination and estimated bundle sheath leakiness in C-4 grasses and
dicots Photosyn Res 91, 47-58

Kunze R, Frommer WB, Fliigge UI (2002) Metabolic engineering of plants: The role of
membrane transport Metab Eng 4, 57-66

Laisk A, Edwards GE (2000) A mathematical model of C—4 photosynthesis: The
mechanism of concentrating CO2 in NADP-malic enzyme type species Photosyn Res 66,
199-224

Laloi M (1999) Plant mitochondrial carriers: an overview Cell Mol Life Sci 56, 918-944

Leegood RC (1985) The lntercellular Compartmentation of Metabolites in Leaves of
Zea-Mays-L Planta 164, 163-171

Leegood RC, Crowther D, Walker DA, Hind G (1983) Energetics Of Photosynthesis In
Zea-Mays. 1. Studies Of The Flash-Induced Electrochromic Shift And Fluorescence
Induction In Bundle Sheath-Cells Biochim Biophys Acta 722, 116-126

Leegood RC, Edwards GE (1996) 'Photosynthesis and the Environment.’ Kluwer
Academic Publishers: Dordrecht

Magnin NC, Cooley BA, Reiskind JB, Bowes G (1997) Regulation and localization of
key enzymes during the induction of Kranz-less, C-4-type photosynthesis in Hydrilla
verticillata Plant Physiol 115, 1681-1689

Majeran W, Cai Y, Sun Q, van Wijk KJ (2005) Functional differentiation of bundle
sheath and mesophyll maize chloroplasts determined by comparative proteomics Plant
Cell 17, 3111-3140

Majeran W, Zybailov B, Ytterberg AJ, Dunsmore J, Sun Q, van Wijk KJ (2008)
Consequences of C4 differentiation for chloroplast membrane proteomes in maize

mesophyll and bundle sheath cells Mol Cell Proteomics 7, 1609-1638

Matsuoka M, Furbank RT, H. F, M. M (2001) Molecular Engeneering of C4
Photosynthesis Annu Rev Plant Physiol Plant Mol Biol 52, 297-314

McIntosh CA, Oliver DJ (1994) The Phosphate Transporter From Pea Mitochondria -
Isolation And Characterization In Proteolipid Vesicles Plant Physiol 105, 47-52

57

Meierhoff K, Westhoff P (1993) Differential Biogenesis Of Photosystem-Ii In Mesophyll
And Bundle-Sheath Cells Of Monocotyledonous Nadp-Malic Enzyme-Type C-4 Plants -
The Nonstoichiometric Abundance Of The Subunits Of Photosystem-Ii In The Bundle-
Sheath Chloroplasts And The Translational Activity Of The Plastome-Encoded Genes
Planta 191, 23-33

Miyao M (2003) Molecular evolution and genetic engineering of C4
photosynthetic enzymes J Exp Bot 54, 179-189

Muhaidat R, Sage RF, Dengler NG (2007) Diversity of Kranz anatomy and biochemistry
in C-4 eudicots Am J Bot 94, 362-381

Murcha MW, Elhafez D, Lister R, Tonti-Filippini J, Baumgartner M, Philippar K, Carrie
C, Mokranjac D, Soll J, Whelan J (2007) Characterization of the Preprotein and Amino
Acid Transporter Gene Family in Arabidopsis Plant Physiol 143, 199-212

Neuhaus HE, Thom E, Mohlmann T, Steup M, Karnpfenkel K (1997) Characterization of
a novel eukaryotic ATP/ADP translocator located in the plastid envelope of Arabidopsis
thaliana L Plant J 11, 73-82

Ohnishi J, Flfigge UI, Heldt HW, Kanai R (1990) Involvement Of Na+ In Active Uptake
Of Pyruvate In Mesophyll Chloroplasts Of Some C4 Plants - Na+/Pyruvate Cotransport
Plant Physiol 94, 950-959

Ohnishi JI, Kanai R (1987a) Na-(+)-Induced Uptake Of Pyruvate Into Mesophyll
Chloroplasts Of A C-4 Plant, Panicum-Miliaceum FEBS Lett 219, 347-350

Ohnishi J I, Kanai R (1987b) Pyruvate Uptake By Mesophyll And Bundle Sheath
Chloroplasts Of A C-4 Plant, Panicum-Miliaceum L Plant Cell Physiol 28, 1-10

Overall RL, Wolfe J, Gunning BES (1982) Inter-Cellular Communication In Azolla
Roots. 1. Ultrastructure Of Plasmodesmata Protoplasma 111, 134-150

Palmieri L, Picault N, Arrigoni R, Besin E, Palmieri F, Hodges M (2008) Molecular
identification of three Arabidopsis thaliana mitochondrial dicarboxylate carrier isoforms:
organ distribution, bacterial expression, reconstitution into liposomes and functional
characterization Biochem J 410, 621-629

Picault N, Hodges M, Paimieri L, Palmieri F (2004) The growing family of mitochondrial
carriers in Arabidopsis Trends Plant Sci 9, 138-146

Picault N, Palmieri L, Pisano I, Hodges M, Palmieri F (2002) Identification of a novel
transporter for dicarboxylates and tricarboxylates in plant mitochondria - Bacterial

58

expression, reconstitution, functional characterization, and tissue distribution J Biol
Chem 277, 24204-24211

Pohlmeyer K, 8011 J, Grimm R, Hill K, Wagner R (1998) A High-Conductance Solute
Channel in the Chloroplastic Outer Envelope from Pea Plant Cell 10, 1207-1216

Pohlmeyer K, Soll J, Steinkamp T, Hinnah S, Wagner R (1997) Isolation and
characterization of an amino acid-selective channel protein present in the chloroplastic
outer envelope membrane Proc Natl Acad Sci U S A 94, 9504-9509

Proudlove MO, Moore AL (1982) Movement Of Amino-Acids Into Isolated Plant-
Mitochondria FEBS Lett 147, 26-30

Proudlove MO, Thurman DA (1981) The Uptake of 2-Oxoglutarate and Pyruvate by
Isolated Pea-Chloroplasts New Phytol 88, 255-264

Renne P, Dressen U, Hebbeker U, Hille D, F liigge UI, Westhoff P, Weber APM (2003)
The Arabidopsis mutant dot is deficient in the plastidic glutamate/malate translocator
DiT2 Plant J 35, 316-331

Roberts AG, Oparka KJ (2003) Plasmodesmata and the control of symplastic transport
Plant, Cell Environ 26, 103-124

Rumpho ME, Edwards GE (1984) Inhibition Of 3-Phosphoglycerate-Dependent O-2
Evolution By Phosphoenolpyruvate In C-4 Mesophyll Chloroplasts Of Digitaria-
Sanguinalis (L) Scop Plant Physiol 76, 711-718

Rumpho ME, Edwards GE (1985) Characterization Of "4,4'-Diisothiocyano-2,2'-
Disulfonic Acid Stilbene Inhibition Of 3-Phosphoglycerate-Dependent 0-2 Evolution In
Isolated-Chloroplasts - Evidence For A Common Binding-Site On The O4 Phosphate
Translocator For 3-Phosphoglycerate, Phosphoenolpyruvate, And Inorganic-Phosphate
Plant Physiol 78, 537-544

Rumpho ME, Wessinger ME, Edwards GE (1987) Influence Of Organic-Phosphates On
3-Phosphoglycerate Dependent O-2 Evolution In C-3 And C-4 Mesophyll Chloroplasts
Plant Cell Physiol 28, 805-813

Sage RF (2004) The evolution of C-4 photosynthesis New Phytol 161, 341-3 70

Sawers RJ H, Liu P, Anufrikova K, Hwang JTG, Brutnell TP (2007) A multi-treatrnent
experimental system to examine photosynthetic differentiation in the maiZe leaf BMC

genomics 8:12

Scheibe R (2004) Malate valves to balance cellular energy supply Physiol Plant 120, 21-
26

59

Scheibe R, Backhausen J E, Emmerlich V, Holtgrefe S (2005) Strategies to maintain
redox homeostasis during photosynthesis under changing conditions J Exp Bot 56, 1481-
1489

Schneider A, Hausler RE, Kolukisaoglu U, Kunze R, van der Graaff E, Schwacke R,
Catoni E, Desimone M, Fliigge U] (2002) An Arabidopsis thaliana knock-out mutant of
the chloroplast triose phosphate/phosphate translocator is severely compromised only
when starch synthesis, but not starch mobilisation is abolished Plant J 32, 685-699

Schneidereit J, Hausler RE, Fiene G, Kaiser WM, Weber APM (2006) Antisense
repression reveals a crucial role of the plastidic 2-oxoglutarate/malate translocator DiTl
at the interface between carbon and nitrogen metabolism Plant J 45, 206-224

Schwender J, Seemann M, Lichtenthaler HK, Rohmer M (1996) Biosynthesis of
isoprenoids (carotenoids, sterols, prenyl side-chains of chlorophylls and plastoquinone)
via a novel pyruvate/glyceraldehyde 3-phosphate non-mevalonate pathway in the green
alga Scenedesmus obliquus Biochem J 316, 73-80.

Singh BK, Shaner DL (1995) Biosynthesis of Branched-Chain Amino-Acids - from Test-
Tube to Field Plant Cell 7, 935-944

Somerville SC, Ogren WL (1983) An Arabidopsis-Thaliana Mutant Defective in
Chloroplast Dicarboxylate Transport Proc Natl Acad Sci USA 80, 1290-1294

Sowinski P, Szczepanik J, Minchin PEH (2008) On the mechanism of C4photosynthesis
intermediate exchange between Kranz mesophyll and bundle sheath cells in grasses J Exp
Bot 59, 1137-1147

Stitt M, Heldt HW (1985) Generation and Maintenance of Concentration Gradients
between the Mesophyll and Bundle Sheath in Maize Leaves Biochim Biophys Acta 808,
400-414

Streatfield SJ, Weber A, Kinsman EA, Hausler RE, Li JM, Post-Beittenmiller D, Kaiser
WM, Pyke KA, Fli‘rgge UI, Chory J (1999) The phosphoenolpyruvate/phosphate
translocator is required for phenolic metabolism, palisade cell development, and plastid-
dependent nuclear gene expression Plant Cell 11, 1609-1621

Taniguchi M, Sugiyama T (1996) Isolation, characterization and expression of cDNA
clones encoding a mitochondrial malate translocator from Panicum miliaceum L Plant
Mol Biol 30, 51-64

Taniguchi M, Sugiyama T (1997) The expression of 2-oxoglutarate/malate translocator in
the bundle-sheath Mitochondria of Panicum miliaceum, a NAD-malic enzyme-type C-4
plant, is regulated by light and development Plant Physiol 114, 285-293

60

Taniguchi M, Taniguchi Y, Kawasaki M, Takeda S, Kato T, Sato S, Tahata S, Miyake H,
Sugiyama T (2002) Identifying and characterizing plastidic 2-oxoglutarate/malate and
dicarboxylate transporters in Arabidopsis thaliana Plant Cell Physiol 43, 706-717

Taniguchi Y, Nagasaki J, Kawasaki M, Miyake H, Sugiyama T, Taniguchi M (2004)
Differentiation of dicarboxylate transporters in mesophyll and bundle sheath chloroplasts
of maize Plant Cell Physiol 45, 187-200

Usuda H, Edwards GE (1980a) Localization Of Glycerate Kinase And Some Enzymes
For Sucrose Synthesis In C-3 And C-4 Plants Plant Physiol 65, 1017-1022

Usuda H, Edwards GE (1980b) Photosynthetic Formation Of Glycerate In Isolated
Bundle Sheath-Cells And Its Metabolism In Mesophyll-Cells Of The O4 Plant Panicum-
Capillare L Aust J Plant Physiol 7, 655-662

Voll LM, Hausler RE, Hecker R, Weber APM, Weissenbock G, Fiene G, Waffenschmidt
S, Flfigge UI (2003) The phenotype of the Arabidopsis cuel mutant is not simply caused
by a general restriction of the shikimate pathway Plant J 36, 301ff

Voznesenskaya EV, F ranceschi VR, Kiirats O, Artyusheva EG, Freitag H, Edwards GE

(2002) Proof of C4photosynthesis without Kranz anatomy in Bienertia cycloptera
(Chenopodiaceae) Plant J 31, 649-62

Voznesenskaya EV, Franceschi VR, Kiirats O, Freitag H, Edwards GE (2001) Kranz
anatomy is not essential for terrestrial C4plant photosynthesis Nature 414, 543-6

Walker DA, Edwards GE (1983) 'C3, C4: mechanisms, and cellular and environmental
regulation, of photosynthesis.’ Blackwell Scientific publications: Oxford

Weber A, Menzlaff E, Arbinger B, Gutensohn M, Eckerskom C, Flugge UI (1995) The
2-oxoglutarate/malate translocator of chloroplast envelope membranes: molecular
cloning of a transporter containing a 12-helix motif and expression of the functional
protein in yeast cells Biochemistry 34, 2621-7.

Weber APM (2004) Solute transporters as connecting elements between cytosol and
plastid stroma Curr Opin Plant Biol 7, 247-253

Weber APM, Fischer K (2007) Making the connections - The crucial role of metabolite
transporters at the interface between chloroplast and cytosol F EBS Lett 581, 2215-2222

Weber APM, Fliigge UI (2002) Interaction of cytosolic and plastidic nitrogen metabolism
in plants J Exp Bot 53, 865-874

61

Weber APM, Linka M, Bhattacharya D (2006) Single, ancient origin of a plastid
metabolite translocator family in Plantae from an endomembrane-derived ancestor
Eukaryot Cell 5, 609-612

Weber APM, Schwacke R, Flugge U1 (2005) Solute transporters of the plastid envelope
membrane Ann Rev Plant Biol 56, 133-164

Weber APM, Weber KL, Carr K, Wilkerson C, Ohlrogge JB (2007) Sampling the
Arabidopsis Transcriptome with Massively Parallel Pyrosequencing Plant Physiol 144,
32-42

Winning BM, Day CD, Sarah CJ, Leaver CJ (1991) Nucleotide-Sequence Of 2 Cdnas
Encoding The Adenine-Nucleotide Translocator From Zea-Mays L Plant Mol Biol 17,
305-307

Wiskich JT (1977) Mitochondrial Metabolite Transport Annu Rev Plant Physiol Plant
Mol Biol 28, 45-69

Woo KC, Flugge UI, Heldt HW (1987) A 2-Translocator Model for the Transport of 2-
Oxoglutarate and Glutamate in Chloroplasts During Ammonia Assimilation in the Light
Plant Physiol 84, 624-632

Young XK, McCarty RE (1993) Assay Of Proton-Coupled Glycolate And D-Glycerate
Transport Into Chloroplast Inner Envelope Membrane-Vesicles By Stopped-F low
Fluorescence Plant Physiol 101, 793-799

Yu C, Claybrook DL, Huang AHC (1983) Transport Of Glycine, Serine, And Proline
Into Spinach Leaf Mitochondria Arch Biochem Biophys 227, 180-187

Zhang L, Hausler RE, Greiten C, Hajirezaei M, Haferkarnp I, Neuhaus HE, Flugge U-I,
Ludewig F (2008) Overriding the co-limiting import of carbon and energy into tuber

amyloplasts increases the starch content and yield of transgenic potato plants Plant
Biotechnol J, 6, 453-464

Zoglowek C, Kromer S, Heldt HW (1988) Oxaloacetate And Malate Transport By Plant-
Mitochondria Plant Physiol 87, 109-115

62

Chapter 3
Comparison of the use of a species-specific database generated by pyrosequencing
with databases from related species for proteome analysis of pea chloroplast

envelopes

This chapter has been previously published in the Journal of Biotechnology: Briiutigam
A, Shrestha RP, Whitten D, Wilkerson CG, Carr KM, Froehlich JE and Weber APM
(2008) Comparison of the use of a species-specific database generated by pyrosequencing
with databases from related species for proteome analysis of pea chloroplast envelopes
Journal of Biotechnology 136, 44-53

I would like to acknowledge Kevin Carr for contribution of the pea data to Figure 3-1.

63

Abstract

Proteomics is a valuable tool for establishing and comparing the protein content
of defined tissues, cell types, or subcellular structures. Its use in non-model species is
currently limited because the identification of peptides critically depends on sequence
databases. In this study, a low-quality preliminary cDNA database for the non-model
species Pisum sativum created by a small number of massive parallel pyrosequencing
(MPSS) runs was compared to comprehensive cDNA databases from Medicago
truncatula and Arabidopsis thaliana created by Sanger sequencing. Each database was
used to identify proteins from a pea leaf chloroplast envelope preparation. It is shown that
the pea database identified more proteins with higher accuracy, although the sequence
quality was low and the sequence contigs were short compared to databases from model
species. Although the number of identified proteins in non-species specific databases
could potentially be increased by lowering the threshold for successful protein
identifications, this strategy markedly increases the number of wrongly identified
proteins. The identification rate with non-species specific databases correlated with
spectral abundance but not with the predicted membrane helix content, and strong
conservation is necessary but not sufficient for protein identification with a non-species
specific database. It is concluded that massively-parallel sequencing of cDNAs

substantially increases the power of proteomics in non-model species.

64

Introduction

The first sequenced plant model species, Arabidopsis thaliana (thale cress), was
chosen not only for its relatively small genome, but also for its small size and rapid life
cycle that make it amenable to genetics (Meinke et al., 1998; Somerville and Somerville,
1999; TAGI, 2000). In addition, a large collection of mutants is available, including
sequence-indexed insertion mutants that facilitate both forward and reverse genetics
approaches (Jander et al., 2002; Parinov and Sundaresan, 2000). However, due to its
small size, the presence of a range of secondary metabolites, and the lack of established
protocols for isolation of subcellular organelles, Arabidopsis does not always represent
the ideal model system for, e.g., organellar proteomics. Proteomics is as a valuable tool
for establishing the protein complement of cells and subcellular structures (Baginsky and
Gruissem, 2006; Baginsky et al., 2004; Dunkley et al., 2006; Heazlewood et al., 2004; Ito
et al., 2007; Kleffmann et al., 2006; Lilley and Dupree, 2006; Peltier et al., 2004; von
Zychlinski et al., 2005; Ytterberg et al., 2006), especially since the prediction capability
of bioinformatics approaches proved insufficient for large scale annotation of organelle
proteomes (Jarvis, 2004; Millar et al., 2006; Reyes-Prieto et al., 2007). In addition,
multiple targeting of proteins has been documented frequently (Duchene et al., 2005;
Millar et al., 2006; Taira et al., 2004) and recently also non-canonical targeting of
proteins to, e.g., chloroplast via the secretory system (Miras et al., 2002; Miras et al.,
2007; Radhamony and Theg, 2006; Villarejo et al., 2005). In contrast to Arabidopsis, the
garden pea (Pisum sativum), is excellently suited for organelle isolation and biochemical
studies of enzymes and established protocols for organelle isolation are available in the

literature (e.g., Corpas et al., 1999; Miflin and Beevers, 1974; Tobin, 1996).

65

m— 1

Unfortunately, little is known about the power of proteomics in non-model species for
which no extensive sequence database is available. Current peptide identification
technology relies on the generation of ideal mass spectra from theoretical libraries. A
sequence database is translated in six frames and the resulting protein sequences are in
silico digested with trypsin. The resulting peptides are used to calculate an ideal mass
spectrum. If an observed spectrum matches a theoretically predicted spectrum with a
certain probability the corresponding peptide is called “identified”. This method of
identification demands a perfect sequence match between the sample peptide and the
database peptide, although some programs, such as implementations of the Xl-Tandem
software (Craig and Beavis, 2004), allow the inclusion of single amino acid mismatches.
Allowing more than one mismatch increases the error rate and the time required for the
search. With increasing evolutionary distance, perfect matches become less likely even
between highly conserved proteins, in particular since conservative changes, such as
aspartate to glutamate will already cause a spectral mismatch. In contrast, low quality
databases, such as the one discussed in this communication, limit the identification of
peptides either by sequencing and assembling errors causing amino acid changes in
predicted peptides or by not providing enough peptide coverage for correct identifications
due to short contigs. De novo sequencing of peptides is considered too slow and limited
by computing time for high throughput applications (Baginsky and Gruissem, 2006;
Pevtsov et al., 2006). Currently, the identification of proteins from non-model species
with limited sequence coverage frequently relies on databases generated from closely

related species (Schmidt et al., 2007) or indeed all sequences that are available in public

66

databases (Taylor et al., 2005) although this method will especially limit the
identification of less conserved proteins.

It is has been recently proposed to use massively-parallel pyrosequencing to fully
explore the potential of proteomics in non-model species such as pea (Weber et al.,
2007). In this study, we systematically assessed the potential and limitations of
massively-parallel pyrosequencing to support proteomics applications. To this end, we
compared proteomics based on a low-coverage transcriptome sequence database of the
garden pea consisting of many short sequence contigs with frequent frame shift errors
with a conventionally created and fairly comprehensive cDNA database of a closely
related model species (Medicago truncatula), and with a high-quality, virtually error-free
database generated from a completely sequenced model species (Arabidopsis thaliana).
We established the limitations of each database and we tested how the degree of
conservation, the abundance of mass spectra generated from a particular protein, and the
number of transmembrane domains influences the odds for successful protein
identification using a non-species specific database. Finally we discuss the consequences

of interpreting the proteomics sample based on the different database results.

Material and Methods
Massively-parallel pyrosequencing and generation of sequence databases

Three different databases were generated for proteome analyses. For the
generation of the pea transcriptome database, one non-normalized and several normalized
libraries were generated and sequenced using massively-parallel pyrosequencing

technology (Margulies et al., 2005). The preparation of cDNA libraries was conducted as

67

described previously (Weber et al., 2007), with the exception that some libraries were
normalized to decrease the proportion of highly abundant transcripts. To this end, 1 pg of
double-stranded cDNA was normalized using a commercial kit (Trimmer-kit, Evrogen,
Moscow, Russia) that is based on Kamchatka crab duplex-specific nuclease (Zhulidov et
al., 2004). Following normalization, the cDNA populations were amplified by PCR and 3
pg of the resulting PCR-amplified cDNA was used per sequencing reaction. One cDNA
library was generated from leaves and one from hypocotyls. Five preparations generated
from the leaf library (four normalized, one non-normalized) and one preparation from the
hypocotyl library (normalized) were sequenced using a Roche/454 G820 instrument. This
technology delivered an average read length of 100 nucleotides. Two additional libraries
from etiolated and de-etiolated (4h light) leaves were sequenced on a half plate each,
using a GS F LX instrument, which allowed for average read lengths of 250 nucleotides.
The preliminary pea database used in the reported work was assembled from
approximately 2 million pyrosequencing reads with a total of about 230 million
nucleotides. The pyrosequencing reads were combined with publicly available pea cDNA
sequences from Genbank and the IPK Gatersleben EST database. All sequences, except
the full-length and partial mRNA sequences from GenBank, were subjected to QA using
the Sequean EST trimming and validation tool cDNAs and ESTs were clustered and
assembled using the TGI Clustering Tools (TGICL) in a multi stage pipeline. The
sequence reads were loosely clustered with a modified version of megablast (Zhang et al.,
2000) and subsequently the clusters were submitted to the CAP3 sequence assembling
program (Huang and Madan, 1999). All programs are available at

http://compbiodfci.harvard.edu/tgi/software. 1,570,251 reads (80% of the total) were

 

68

assembled into 135,250 contigs. For the proteome analysis, only contigs longer than 100
base pairs were used. The pea cDNA database is currently undergoing further
development and a detailed description of the pea cDNA sequence database and its
assembly will be reported in an upcoming manuscript (R.P. Shrestha et al., in
preparation). The Medicago database was based on the M truncatula gene index of
tentative consensus sequences assembled by TIGR, currently maintained at the Dana-
Faber Cancer Institute and Harvard Medical School of Health
(http://compbiodfci.harvard.edu/tgi/cgi-bin/tgi/gimain.pl‘.’gudb=medicago). For the
Arabidopsis database the latest Arabidopsis proteome annotation from the TAIR7
(www.mabidopsisorg) genome release was used.
Processing of proteomics samples

As a proteome sample, chloroplast envelope membranes were isolated from 10-14
days old pea plants as described previously (Douce and Joyard, 1979; Keegstra and
Yousif, 1986). Envelope membrane samples (approx. 100 ug of protein) were mixed with
SDS—PAGE loading buffer, incubated for 20 minutes on a reaction tube shaker at 15° C,
and subsequently separated by 12.5% SDS-PAGE. After staining with Coomassie
Brilliant Blue, each gel lane was cut into ten equally-sized slices. Proteins contained in
the gel slices were subjected to tryptic cleavage as described by Shevchenko et al.
(Shevchenko et al., 1996). Extracted peptides were automatically loaded onto a Waters
Symmetry C18 peptide trap (5 um, 180 um x 20mm) at a flow rate of 4 uL/min in 2%
Acetonitrile/0.1%Formic Acid for 5 minutes by a Waters nanoAcquity Sample Manager.
The peptides were eluted onto a Waters BEH C18 nanoAcquity column (1.7 um, 100 um

x 100mm) and eluted over 90 minutes using a Waters nanoAcquity UPLC into a

69

ThermoElectron LTQ-FTICR mass spectrometer with a flow rate of 300 nL/min (Buffer
A = 99.9% Water/0.1% F ormic Acid, Buffer B = 99.9% Acetonitrile/0.1% Formic Acid:
gradient of 5% B to 40% B from 0 to 63 minutes, 40% B to 90% B from 63 to 71 minutes
and 5% B from 71 to 90 minutes). The top ten ions of each survey scan, which were
taken at a resolution of 50,000, were subjected to automated low energy collision induced
dissociation. The resulting MS/MS spectra were converted to a peak list using BioWorks
Browser v 3.2. This empirical mass spectra library of 27,443 queries was compared to
three databases with sequences from Arabidopsis thaliana, Medicago truncatula and
Pisum sativum, respectively, using the Mascot searching algorithm, v 2.2
(www.matrixscience.com). Carbarnidomethyl cysteine was set as a fixed peptide
modification and oxidation of methionine was allowed. Up to two missed tryptic sites
were allowed. The peptide tolerance was set to +/-10ppm and the MS/MS tolerance to
0.8kDa.
Analysis of proteomics data

The Mascot output was loaded into the program Scaffold

(wwwprotcomesoftware.com) which calculates protein identification probabilities based

 

on PeptideProphet and ProteinProphet (Keller et al., 2002; Nesvizhskii et al., 2003). The
protein identification threshold was set to 99% probability for all databases and at least
two unique peptides for each protein (“stringent criteria”) or to 99% probability for at
least one database with lesser probabilities allowed for the other databases on the same
protein and one unique peptide for each protein (“relaxed criteria”). For further analysis
the data was exported to MS Excel. For each protein identified, the sequences in both the

pea database as well as the Medicago database were mapped to the Arabidopsis proteome

70

using BlastX (Altschul et al., 1997) and the Arabidopsis gene identifier (AG1) of the
closest homologue was noted. Based on its closest Arabidopsis homologue, the protein
was annotated and the predicted number of membrane spanning helices was retrieved
from TAIR. In cases where the AGls obtained from the three different identification
strategies did not match, the identifications were manually inspected and priority was
given to the identification with the highest probability score. For identical probability
scores the highest number of unique peptides mapped to a protein was declared the
correct identification. When multiple sequences were matched to the same spectra,
identifications were either called “mistaken” if none of the matches were identical to the
protein identified correctly in the other databases or “multiple” if one of the proteins was
identical to the correct match determined as outlined earlier. The complete list of proteins
was collapsed into a non-redundant list based on the AGIs with sequences yielding the

same hit in the BlastX hit being summed as the same protein.

71

Results

Properties of the cDNA sequence databases

 

 

 

  
 

>5kb * - Pea database
- Medicago Datbase
_ - Arabidopsis database

'0 4kb-5kb

.9

‘E

8 3kb-4kb

‘5

5

a 2kb-3kb

5

" 1kb-2kb .-.

300-1kb '

 

 

0 5000 1 0000 1 5000 20000 25000 30000
Number of contigs

Figure 3-1. The length distribution of contigs (assembled transcripts) in databases used for
proteome analyses.

The characteristics of a low-coverage pea cDNA database generated by limited
pyrosequencing was compared to two databases generated with conventional sequencing
technology. The pea cDNA sequence database contained more than 31,000 relatively
short contigs. About 29,000 contigs were between 300 and 1,000 nucleotides in length,
about twice as many in this length category as in the Medicago
(http://compbiodl‘ci.harvardcdu/tgi/cui-bin/tgi/gimain.pl‘?gudb=mcdicago) and an
Arabidopsis transcriptome databases (httpz/l’compbiodfci.harvardedu/Lgi/cgi-

bin/tgi/gimain.pl‘?gudb=arab) (Figure 3-1). In contrast, contigs between one and two kb

72

were massively underrepresented (1697 contigs) in the pea database in comparison to the
Medicago database (6,224 long contigs) and the Arabidopsis database (13,033 long
contigs) and contigs longer than 2000 nucleotides were almost absent from the pea
database (Figure 3-1). In addition to having shorter contigs, many pea contigs, when
translated in all six frames, apparently contained frame shifts that are likely due to
assembly or base calling errors (see Supplementary File of pea contigs).

The Medicago database consisted of the tentative consensus sequences (TCs)
assembled from EST projects from M truncatula. Medicago is being developed as a
model legume species and is hypothesized to serve as a research template for the garden
pea. The Medicago TCs were comprised of about 36,000 unique sequences. About one
half of these sequences were tentative consensus sequences that represent contigs
consisting of several ESTs and the other half were singletons. The length distribution of
the Medicagos TCs was more similar to that of the Arabidopsis transcriptome database
and the sequence quality judged from open reading frame annotation software was high

(http://compbiodfci.harvard.edu/tgi/cgi-bin/tgi/gimainpl?gudb=mcdicago). While both

 

the Arabidopsis and Medicago databases contain similar numbers of cDNA sequences
that are shorter than lkb, the Medicago database contains only half as many sequences
between one and two kb, and only one tenth for transcripts longer than two kb (Figure 3-
1). The Arabidopsis proteome derived from the TAIR7 build of the completely sequenced
genome served as an example for a completely sequenced genome. The capability of the
new pea database to reliably identify proteins despite its obvious shortcomings was

compared with the capabilities of the databases of M. truncatula and A. thaliana.

73

Performance of the databases in proteomics

Table 3-1: Summary of protein identifications with different databases

 

 

 

stringent criteria relaxed criteria
Database
total avg. total avg.
number of number of number of number number
identifications number 0f spectra identifications of of spectra
spectra
per ID spectra per ID
combined 255 8222 32 283 8892 31
Pisum 221 5012 23 255 5139 20
sativum
Medicago 125 1977 16 203 2198 1 1
truncatula
Arabidopsis 82 1233 15 165 1555 9
thaliana

 

A pea chloroplast envelope membrane proteome sample was analyzed using the
sequence databases described above. To identify the most advantageous database and
program parameters for protein identification, we employed a factorial approach. An
empirical library of uninterpreted mass spectra was generated from a single proteomics
experiment of pea chloroplast envelope membranes. This particular sample was chosen
because it represents a relatively minor share of the total cellular proteome and because it
contains a large number of highly hydrophobic membrane proteins and thus represents a
challenging target for protein identification. In addition, several proteomic studies of
chloroplast envelope membranes from a range of plant species have been published, thus
providing good references for comparison. All three databases described above were
matched to the empirical spectra library using either stringent or relaxed criteria. For both

criteria, the largest number of proteins could be identified with the pea database, followed

74

by Medicago and Arabidopsis. Applying stringent criteria, a total of 8,222 spectra were
matched to 255 non-redundant proteins using a combination of all three databases. A list
of all identifications can be found in Supplementary Table 3-1 and a non-redundant list of
identified proteins in Supplementary Table 3-2. Under stringent conditions the pea
database allowed matches of 5,012 spectra against 221 non-redundant proteins (86% of
the total). On average each protein was identified by 23 spectra with up to 362 spectra per
protein (Table 3-1). The Medicago database yielded 1,977 matched spectra on 125
proteins (49% of the total). With only 32% or 82 proteins the Arabidopsis database
allowed the fewest matches. With relaxed criteria the total number of proteins identified
using a combination of all three databases increased to 283 and 8,892 spectra were
matched. In the pea database, the relaxed criteria allowed for 5,139 matching spectra, but
a large number of proteins that were identified with only a few spectra caused the average
number of spectra per protein to drop to 20. The Medicago TC database yielded 2,198
spectra mapping to 203 proteins, almost 72% of the total. The A. thaliana database
yielded the lowest number of identified proteins, with only 165 proteins or 58% of the
total (Table 3-1).

There were 36 proteins which could not be identified with the pea database but
yielded significant identifications with at least one of the other databases. Eight of these
proteins are encoded on an organelle genome (Supplemental Table 3-2). Six proteins
were very similar to closely related proteins, which were present in the pea database and
were identified. For nine of these proteins corresponding sequences were absent from the

pea database. The remaining twelve proteins remained unidentified with the pea database

75

although corresponding sequences were contained in the database, as indicated by Blast
searches against the database.
Abundance of mass spectra is positively correlated with correct protein

identification in non-species specific databases

 

 

 

 

  

   

100
- in Medicago database

8 80 _ - In Arabidopsis database
3
i
a 60 ‘
8
"6
g 40 -
c
0
2
El 20 .

0 _

dkfa\oo a (doc va\co
«09 9° 93 «0,49 ‘59 do ‘5

abundance

Figure 3-2. The number of spectral counts detected for a protein is correlated with the rate
of identification. Proteins in the pea database were grouped by spectral abundance and the
percentage of proteins also identified with the non-species specific databases was plotted. The
following absolute numbers equal 100%: >100 spectral counts eight proteins, 50-99 spectral
counts 14 proteins, 20-49 spectral counts 44 proteins, 10-19 spectral counts 43 proteins and <10
spectral counts 111 proteins.

With both non-species specific databases 51% or 68% of the proteins could not be

identified when stringent criteria were applied. We next investigated whether there are

76

protein characteristics that might serve as predictors of identification probability in the
non-species specific databases. Although the correlation between the spectral count and
the abundance of a protein in not absolute, the spectral count has been used successfully
to estimate protein abundance (Liu et al., 2004; Lu et al., 2007; Zybailov et al., 2005).
The proteins in the pea database were arranged into five groups based on the abundance
of their matching mass spectra. More than half of the protein identifications using the pea
database were based on two to ten spectra, whereas only 3.6% of the proteins were
identified with more than one hundred spectra each. For each abundance class, the
percentage of proteins that could also be identified with the Medicago and the
Arabidopsis databases were plotted (Figure 3-2). The abundance of spectra as determined
by the spectral count obtained with the pea database correlated with the identification rate
with non-species specific databases (R(squared)>0.95). With the Medicago database
almost all of the proteins represented by more than 100 mass spectra could be identified
whereas only 38% of the proteins with a spectral count below 10 were identified. With
the Arabidopsis database about two thirds of the proteins with a high spectral count were
identified and the identification rate dropped to 17% for proteins that had spectral counts

of less than 10 (Figure 3-2).

77

Protein hydrophobicity is a poor predictor for correct protein identification

 

 

 

 

100
- in Medicago database
3 80 - m Arabidopsis database
3
i
a 60 '1
8
“6
§ 40 -
c
§
S 20 -
0 _

   

1 2 3 4-6 7-13 beta sheets

Number of predicted transmembrane helices

Figure 3-3. The number of predicted membrane spanning helices for a protein is not
correlated with the rate of identification. Proteins in the pea database were grouped by
predicted membrane spanning helices and the percentage of proteins also identified with the non-
species specific databases was plotted. The following absolute numbers equal 100%: no predicted
transmembrane helices lll proteins, one predicted transmembrane helix 37 proteins, two
predicted transmembrane helices 24 proteins, three predicted transmembrane helices seven
proteins, four to six predicted transmembrane helices 26 proteins, seven to thirteen predicted
transmembrane helices ten proteins, predicted beta sheets five proteins

We also tested whether the presence of transmembrane helices adversely affected
the probability that a protein was identified using a non-species specific database as has
been proposed by (Eichacker et al., 2004). Proteins that were identified using the pea
database were grouped according to the number of predicted membrane spanning helices
(none, one, two, three, four to six, more than seven) and the presence of beta sheets. More

than half of the proteins identified with the pea database were not predicted to contain

78

transmembrane helices and the other groups contained 2.3-16.8% of the proteins. For
each group the percentage of proteins that could also be identified with the non-species
specific databases was plotted (Figure 3-3). Unlike protein abundance as deduced from
spectral counts, the number of predicted transmembrane helices is of less predictive value
for identification rate since no correlation was observed between the number of
transmembrane domains and the probability for identification using a non-species
specific database. However, among the proteins with more than seven predicted
membrane spanning helices only one (Medicago database) or none (Arabidopsis
database) was identified out of ten proteins identified with the pea database.

Analysis of abundant non-identified and of wrongly identified proteins

Although proteins with more matching spectra in the pea database had a higher
probability to be also identified with the non-species specific databases, many of the
fifteen proteins with the highest number of spectra could only be identified with the
species-specific database (Table 3-2). The Medicago database did not allow the
identification of the small subunit of RubisCO and a chaperonin, and a hydroperoxide
lyase and the triosephosphate phosphate translocator (TPT) did not pass the stringent
threshold for identification probability (p<0.01). With the Arabidopsis database both the
hydroperoxide lyase and the small subunit of RubisCO remained unidentified whereas
the TPT, two components of the protein import complex, Toc159 and ToC34, a carbonic
anhydrase and an unknown protein did not pass the probability threshold (Table 3-2). The
sequences of TPT, which was not identified with the Medicago and the Arabidopsis
databases, and of malate dehydrogenase (NAD-MDH), which was identified with all

three databases, were aligned and the tryptic fragments were determined. PsTPT is highly

79

similar to both AtTPT (77.5% identity and 14.5% similarity) and MtTPT (92.8% identity
and 14.5% similarity). The same is true for PsNAD-MDH and AtNAD-MDH (73%
identity and 18.6% similarity) and PsNAD-MDH and MtNAD-MDH (91.7% identity and
3.9% similarity). Despite the high degree of sequence identity AtTPT shares only one and
MtTPT only six tryptic peptides with the pea protein. The TPT peptides from the
empirical spectra library were mapped onto the aligned sequences. All but one of the TPT
peptides that generated mass spectra contained at least one amino acid exchange with
respect to Medicago or Arabidopsis. In contrast, 16 peptides of the Medicago NAD-MDH
exactly matched the pea peptides. For thirteen of those 16 theoretically predicted peptides
mass spectra were detected experimentally, thus permitting reliable identification of the
protein using the Medicago database. Only four tryptic peptides of the Arabidopsis NAD-
MDH matched the pea protein sequence, but since for two of them mass spectra were

experimentally detected the protein could still be reliably identified.

80

 

Arabidopsis, relaxed

 

Medlcago ' relaxed - correct identification

- mis-ldentification
Pea, relaxed - multipleidentlficatlons

 

Arabidopsis, stringent

Medicago , stringent

Pea, stringent

 

 

l I I T I

0 20 40 60 80 100

Percentage of all proteins identified in a database

Figure 3-5. The non-species specific databases yield a significant amount of mis-
identifications and multiple identifications

When the successful protein identifications were compared between the three
databases, several non-matching identifications were revealed (Figure 3-5). Manual
inspection of the peptide sequences, the peptide probability scores, and the protein
coverage indicated that one of the identifications is likely correct. With stringent criteria
none of the identifications with the pea database appeared to be invalid compared to the
non-species specific databases. Using relaxed criteria, two proteins were erroneously
identified. Peptides for the large subunit of RubisCO were mistakenly annotated as part
of a RubisCO-like protein of unknown function encoded on the mitochondrial genome. A
second misidentification is of the root glutamate synthase (GLU2) instead of the leaf
glutamate synthase (GLUl). Using the Medicago database, 24 proteins were erroneously

identified with relaxed criteria and in three cases the peptide with matching spectra is part

81

of more than one protein and therefore yields multiple identifications of proteins only one
of which is correct. Sixteen of the misidentified proteins are closely related to the correct
protein and eight identifications corresponded to completely different protein. Unlike
with the pea database, even under stringent conditions, fifteen misidentifications
persisted. With the Arabidopsis database 29 proteins were mistakenly identified when
relaxed criteria were used and eight multiple identifications existed. More stringent
criteria still resulted in five misidentifications and in one peptide, which identified
multiple proteins, but all persisting misidentifications identify closely related proteins

(for details see Supplementary Table 3-2).

Discussion

The results presented in this paper indicate that the prospects for identifying
proteins from a species with limited sequence resources by proteomics can be massively
increased by generating a species-specific transcriptome database by MPSS, even if the
resulting database is of low quality, compared to sequence databases generated by
conventional sequencing. When non-species specific databases are used, the odds for
protein discovery are limited, and the probability to identify a protein can be predicted by
its abundance but not by its content of membrane spanning helices. Strong sequence
conservation is necessary but not sufficient to identify a protein with a non-species
specific database. Especially when the identification criteria are relaxed to allow
imperfect matches and therefore more protein identifications with non-species specific

databases, the identifications are more prone to erroneous identifications.

82

Our initial expectation was that both short contig length and the relatively high
rate of sequencing errors that are characteristic to low-coverage MPSS projects would
severely limit the prospects for successful protein identification. Unexpectedly, however,
the pea database developed in this study was superior to the tested non-species specific
databases with regard to the rate of protein discovery and the quality of identifications
despite being of low quality and low coverage, as compared to conventional databases.
Not only more proteins were identified but also the average number of spectra mapping
to each protein was higher (Table 3-1). The advantage of species specificity clearly
outweighs the quality issues of the database. Only 10-14% of the total proteins identified
remained unidentified with this database. Detailed analysis of these non-identified
proteins revealed that organelle-encoded proteins are overrepresented (22%) among those
not identified with the pea database, as compared to the overall proportion of identified
proteins that are encoded in organelles (4%) (Supplemental Table 3-2). Since the mRNA-
isolation protocol used in this study included two consecutive rounds of poly-A+
purification and because reverse translation of mRNA into first strand cDNA was primed
by oligo-dT, it is likely that transcripts from organellar genomes are underrepresented in
the sample, as reported previously (Weber et al., 2007). For nuclear encoded proteins,
sequences could not be identified for nine proteins in the pea transcriptome database and
for seven additional proteins the corresponding nucleotide sequences were either short or
fragmented into multiple unassembled short contigs, thus demonstrating how an
unfinished MPSS-generated sequence database limits proteomic identification technology
if the sequence contigs are too short. Incidentally, most of the contigs in the pea database

are shorter than one kb with the majority being shorter than 400 nucleotides (Figure 3-1).

83

This translates into a stretch of approximately 140 amino acids, given no 5’ or 3’-UTRs
are contained in the sequence. Depending on how the tryptic fragments map onto the
short sequence, the identification probability can probably be too low to identify proteins
reliably. This problem is currently being addressed by generating additional sequence
coverage and new assembling methods for the short and midrange sequence reads
obtained by the MPPS technology. The recently released GS FLX instrument produces
sequence reads that are 2.5-times longer as the sequence reads of the GS 20. In addition,
preliminary tests showed that including a second, less stringent clustering step in the
assembly pipeline might serve to produce longer contigs that are more suitable for
proteomics applications although this may aggravate the problem of frame shift errors
during assembly. In conclusion, the analysis of the proteome sample with several
databases served to determine the quality of the sequence database for proteomics. Since
it was less likely for proteins to be identified with non-species specific databases the
presence of such identifications indicated that the species-specific database could still be
improved.

Considering the multitude of short contigs (Figure 3-1) and the presence of
sequencing errors that resulted in frame shifts (Supplementary File) it was surprising that,
with both databases from model species, fewer proteins were identified than with the
novel pea database (Table 3-1). The Arabidopsis database represents a completely
sequenced genome that is well annotated and presumably complete. Although the
evolutionary split between the genome of P. sativum and A. thaliana occurred about 100
million years ago (Wikstrom et al., 2001), we hypothesized that the degree of

conservation might be sufficient to identify most proteins, albeit with a lower peptide

84

count, especially given that the database is complete. Surprisingly, the Arabidopsis
database only allowed the identification of 32% of the proteins relative to all databases
combined. Notably, in a previous chloroplast envelope proteomics study that used
spinach chloroplasts and mostly non-species specific sequences for identification only 50
proteins (25% compared to this study and 15% compared to (Froehlich et al., 2003)) were
identified in total, amongst them 21 membrane proteins with more than 4 membrane
helices (Ferro et al., 2002). The phosphate translocators that were identified in this
previous study were limited to those for which species-specific sequences were available
in public databases at that time. In addition to the Arabidopsis database a more closely
related model species database was tested for its performance in a proteomics application.
M truncatula and P. sativum are close relatives in the subfamily Papilionoideae which
separated about 25 million years ago (Lavin et al., 2005), which is equivalent to the
evolutionary distance between Arabidopsis and the Brassica species (Yang et al., 1999),
and both microsynteny and macrosynteny between the genomes have been demonstrated
(Gualtieri et al., 2002; Kalo et al., 2004). Medicago is being developed as the model
legume species, also serving as a research template for pea. Since the evolutionary
distance between Medicago and pea is smaller than between Arabidopsis and pea, we
hypothesized that the Medicago database would allow for the identification of more
proteins. Indeed, the Medicago database allowed the detection of half of the proteins
under stringent conditions, although the average number of spectra detected for each
protein was only 50% of those obtained with the pea database. Although the number of
sequence contigs longer than lkb is significantly lower in the Medicago database than in

the Arabidopsis database (Figure 3-1), indicating that a large portion of the available

85

sequences do not represent full length cDNAs, the Medicago database is more successful
than the Arabidopsis database in identifying chloroplast envelope proteins (compare to
Table 3-1). Based on the quality of the databases and the number of proteins that could be
identified, we conclude that evolutionary distance imposes a higher penalty for protein
identification than does the quality of the sequence database. This underscores the
inability of current peptide identification software for proteomics applications to tolerate
amino acid mismatches between the theoretically and experimentally generated spectra
especially when stringent identification criteria are applied.

We hypothesized that relaxed identification criteria may allow imperfect matches
for peptides and might thus enhance the identification rate for both the Medicago and the
Arabidopsis database. Relative to the total number of proteins identified, 58% of proteins
could be identified with relaxed criteria compared to 32% with stringent criteria with the
Arabidopsis database and 72% compared to 49% with the Medicago database.
Unfortunately, the relaxed identification criteria also lead to a high number of mistakenly
identified proteins and the identification of multiple family members that shared a single
conserved peptide. In conclusion, although allowing protein identifications with a single
peptide match and low scoring peptides could increase the coverage, the high number of
mistaken and multiple identifications make this strategy inadvisable for increasing the
number of protein identifications.

The experiment clearly established the value of a species-specific database, even
if it was of low quality (Figure 3-1), for proteomics identification technologies. Since
many projects will have to rely on sequence databases of related species, we tested

several parameters to determine what limits the identification. Theoretically, for any

86

successful identification under stringent criteria, a protein must yield at the least two
fragmented peptides whose spectra match theoretical spectra from the library. Frequently
a higher number of unique peptides are identified for more abundant proteins and the
likelihood, that a least two completely conserved peptides are present among them,
increases. Accordingly, the abundance of spectra and the identification rate were indeed
closely correlated (Figure 3-2). The correlation was not absolute though; we found that
extremely abundant proteins, such as the TPT, were only identified with the pea database
(Table 3-2) whereas about one third (Medicago database) to one fifth (Arabidopsis
database) of low abundance proteins could still be identified with a non-species specific
database (Figure 3-2). Conserved structural features of proteins may also limit
identification. Proteins with comparatively high numbers of membrane spanning helices
and underrepresented recognition sites for trypsin are harder to identify than soluble
proteins since there are frequently less spectra available for matching to the respective
database (Eichacker et al., 2004). When the proteins identified in this study were grouped
according to their predicted membrane helix content and the ratio of proteins identified
with the non-species specific databases relative to the pea database was plotted, however,
no correlation was observed (Figure 3-3). Possibly, the sample size with two databases
and a limited number of membrane proteins was too small to reveal a correlation. Since
highly hydrophobic proteins with more than 7 transmembrane helices, such as the TPT
(Weber et al., 2005), could not be identified with the non-species specific databases, any
analysis of a membrane proteome will likely critically depend on the availability of a

species-specific sequence database.

87

Finally, we studied several of the abundant proteins that were identified in all
three databases and the proteins that remained unidentified in the Medicago and the
Arabidopsis database. As an example for an abundant membrane protein, TPT, which
remained unidentified in the non-species specific databases, was compared with the
soluble protein NAD-MDH that was identified with all three databases. Although the
TPT orthologues are slightly more conserved than the NAD-MDH orthologues, the mass
spectra generated from TPT could not be mapped onto the sequences provided by the
Medicago and Arabidopsis databases. For Arabidopsis, only a single tryptic peptide is
completely conserved between the corresponding pea and Arabidopsis proteins; hence the
protein could not be identified with confidence. Although the Medicago sequence shares
six tryptic peptides with the pea sequence, only one of these peptides was matched to the
empirical spectra library, which is not sufficient for a significant identification with
stringent criteria. Two of the conserved peptides are longer than 23 amino acids and two
are very short, which might be the reason why spectra corresponding to these peptides
were not experimentally detected. NAD-MDH could be identified with all databases
since matching spectra were generated from the protein. A high degree of conservation is
necessary but not sufficient for a successful identification of a protein with a non-species

specific database since a single amino acid exchange will preclude a peptide match.

88

Pea database

Protein Import Complex

..........

 

O
O
O
s O
e +28 unknown membrane proteins
.I
.,,_
° I
a", . _
I, . .
......... n .' ‘ s
.. o ..
' . ......... ......10'0“ """

Arabidopsis database

Protein Import Complex

   

®.00000I00o0..
Figure 3-6. Non-species specific databases limit the discovery of new proteins and the
interpretation of known proteins in the pea chloroplast envelope proteome sample.

 

89

The reduced information caused by using a non-species specific database hampers
the interpretation of the data. To visualize this fact, proteins identified with each database
were drawn in identical positions onto a schematic representation of a chloroplast. The
likely mitochondrial and endomembrane system contaminants were also drawn on the
corresponding structures (Figure 3-6). As expected from the reduced number of proteins
identified with the Arabidopsis and Medicago databases, the total number of membrane
proteins and contaminating proteins is also lower. The Arabidopsis database identified no
protein likely to reside in the endomembrane system. Hence one would conclude that
there is no contamination from this source, although both the Medicago and the pea
database reveal that there is at least one protein present that is believed to reside in the
endomembrane system. In the extreme case of the Arabidopsis database used to analyze
the pea chloroplast envelope proteome, none of the transport proteins catalyzing the
major metabolite fluxes across the envelope (see Weber, 2004; Weber and Fischer, 2007,
for recent reviews) could be identified. With the pea database all transporters for
phosphorylated sugars that are predicted to reside in the inner envelope of chloroplasts
were identified: the triosephosphate, the phosphoenolpyruvate, and the
pentosephosphate/phosphate translocators (Flugge et al., 2003; Weber et al., 2004)
whereas only the pentose phosphate translocator could be identified with the Medicago
database. The two translocator system for importing 2-oxoglutarate and exporting
glutamate could also be identified only with the pea database (Renne' et al., 2003;
Reumann and Weber, 2006; Schneidereit et al., 2006; Weber and Flugge, 2002; Weber et
al., 1995). A plastidic ADP/ATP translocator (Mohlmann et al., 1998; Neuhaus et al.,

1997; Reiser et al., 2004) could be identified with all three databases as well as several

90

members of the mitochondrial carrier family that have previously been shown to be
targeted to chloroplasts (Bedhomme et al., 2005; Bouvier et al., 2006; Picault et al.,
2004). At least one member of the potassium proton exchanger family (Maser et al.,
2001) was identified with all databases but a magnesium transporter (Li et al., 2001) and
a putative mechanosensitive channel (Haswell and Meyerowitz, 2006) were only seen in
the pea database. Two ABC transporters, PAAl (Shikanai et al., 2003) and HMAl
(Seigneurin-Bemy et al., 2006), which import copper and possibly other metal ions into
the chloroplast, could be identified with the pea database but not with either of the non-
species specific databases. Three other ABC type transporters were sufficiently
conserved for identification with all three databases. For the import apparatus, all
canonical components that are known to date (Gutensohn et al., 2006) were identified
with the pea database except for Tic21 (Teng et al., 2006), which was recently assigned
the function of an iron transporter (Duy et al., 2007). With the Medicago database the
majority of import complex components could be identified but with the Arabidopsis
database only four components were identified. Of the plastid division machinery, PDV2
and MinE were found with the pea database (Glynn et al., 2007; Miyagishima et al.,
2006), whereas another component, Are 6 that has been identified previously (Froehlich
et al., 2003), was not identified, but was also only present as a highly fragmented
sequence in the pea database. Several likely membrane associated proteins involved in
fatty acid and membrane lipid metabolism were identified. For fatty acid synthesis and
modification, acetyl co-enzyme A carboxylase, a fatty acid desaturase (McConn et al.,
1994) and a long chain fatty acid coenzyme A ligase (Schnurr et al., 2002) were

identified as well as proteins involved in lipid metabolism such as 1,2-diacylglycerol 3-

91

beta-galactosyltransferase for the synthesis of galacto- and UDP-sulfoquinovosezDAG
sulfoquinovosyltransferase for synthesis of sulfolipids (SQD2) (Jarvis et al., 2000; Yu et
al., 2002). Most proteins were identified with the pea database but SQD2 was only
identified with the non-species specific databases. SQD2 was likely not identified
because its sequence is fragmented into several short contigs in the pea database (for a
complete list of identified proteins and the capabilities of each database please refer to
Supplementary Table 3-1). Based on the analysis we conclude that any proteome analysis
relying on protein identifications based on a non-species specific database is limited in its
conclusions about the presence of proteins and that generating a species specific database

even if it is of low quality can massively enhance protein discovery.

Acknowledgements

We thank Shari Tjugum-Holland and Jeff Landgraff of the Michigan State University
Research Technology Support Facility for assistance with RNA and DNA analysis and
DNA sequencing. This work was supported by a Strategic Partnership Grant (Next
Generation Sequencing Center) of the Michigan State University Foundation (to
A.P.M.W), NSF-grants lOB-0548610 (to A.P.M.W) and MCB-0519740 (to A.P.M.W),
and by an Arabidopsis Emotional Genomics Network (WE 2231/4-1) award of the
Deutsche F orschungsgemeinschaft (to A.P.M.W).

Supplementary Material

Table 83-1 Redundant list of all proteins identified in the experiment

Table 83-2 Non-redundant list of all proteins identfified in the experiment

92

References

Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J ., Zhang, 2., Miller, W., Lipman,
DJ. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database
search programs. Nucleic Acids Res. 25, 3389-3402

Bedhomme, M., Hoffmann, M., McCarthy, E.A., Gambonnet, B., Moran, R.G., Rebeille,
F., Ravanel, S. (2005) Folate metabolism in plants - An Arabidopsis homolog of the

mammalian mitochondrial folate transporter mediates folate import into chloroplasts. J.
Biol. Chem. 280, 34823-34831

Bouvier, F ., Linka, N., Isner, J .C., Mutterer, J ., Weber, A.P.M., Camara, B. (2006)
Arabidopsis SAMTl defines a plastid transporter regulating plastid biogenesis and plant
development. Plant Cell 18, 3088-3105

Craig, R., Beavis, RC. (2004) TANDEM: matching proteins with tandem mass spectra.
Bioinforrnatics 20, 1466-1467

Douce, R., Joyard, J. (1979) Isolation and Properties of the Envelope of Spinach
Chloroplasts. In Reid, E, ed, Plant Organelles. Ellis Horwood Publishers, pp 47-59

Buy, D., Wanner, G., Meda, A.R., von Wiren, N., 8011, J ., Philippar, K. (2007) PICl, an
Ancient Permease in Arabidopsis Chloroplasts, Mediates Iron Transport. Plant Cell 19,
986-1006

Eichacker, L.A., Granvogl, B., Mirus, O., Muller, B.C., Miess, C., Schleiff, E. (2004)
Hiding behind Hydrophobicity: Transmembrane Segments In Mass Spectrometry. J. Biol.
Chem. 279, 50915-50922

Ferro, M., Salvi, D., Riviere-Rolland, H., Verrnat, T., Seigneurin-Bemy, D., Grunwald,
D., Garin, J ., Joyard, J ., Rolland, N. (2002) Integral membrane proteins of the chloroplast
envelope: identification and subcellular localization of new transporters. Proc. Natl.
Acad. Sci. U. S. A. 99, 11487-11492.

F lugge, U.I., Hausler, R.E., Ludewig, F., Fischer, K. (2003) Functional genomics of
phosphate antiport systems of plastids. Physiol. Plant. 118, 475-482

Froehlich, J .E., Wilkerson, C.G., Ray, W.K., McAndrew, R.S., Osteryoung, K.W., Gage,
D.A., Phinney, BS. (2003) Proteomic study of the Arabidopsis thaliana chloroplastic

envelope membrane utilizing alternatives to traditional two-dimensional electrophoresis.
J. Proteome Res. 2, 413-425

Glynn, J .M., Miyagishima, S.Y., Yoder, D.W., Osteryoung, K.W., Vitha, S. (2007)
Chloroplast Division. Traffic 8, 451-461

93

Gualtieri, G., Kulikova, O., Limpens, E., Kim, D.J., Cook, D.R., Bisseling, T., Geurts, R.
(2002) Microsynteny between pea and Medicago truncatula in the SYM2 region. Plant
Mol. Biol. 50, 225-235

Gutensohn, M., Fan, E., Frielingsdorf, S., Harmer, P., Hou, B., Hust, B., Klosgen, RB.
(2006) Too, Tic, Tat et al.: structure and function of protein transport machineries in
chloroplasts. J. Plant Physiol. 163, 333-347

Huang, X.Q., Madan, A. (1999) CAP3: A DNA sequence assembly program. Genome
Res. 9, 868-877

Jarvis, P., Dorrnann, P., Peto, C.A., Lutes, J ., Benning, C., Chory, J. (2000) Galactolipid
deficiency and abnormal chloroplast development in the Arabidopsis MGD synthase 1
mutant. Proc. Natl. Acad. Sci. U. S. A. 97, 8175-8179

Kalo, P., Seres, A., Taylor, S.A., Jakab, J ., Kevei, Z., Kereszt, A., Endre, G., Ellis,
T.H.N., Kiss, GB. (2004) Comparative mapping between Medicago sativa and Pisum
sativum. Mol. Genet. Genomics 272, 235-246

Keegstra, K., Yousif, AB (1986) Isolation and Characterization of Chloroplast Envelope
Membranes. Meth. Enzymol. 118, 316-325

Keller, A., Nesvizhskii, A.I., Kolker, E., Aebersold, R. (2002) Empirical statistical model
to estimate the accuracy of peptide identifications made by MS/MS and database search.
Anal. Chem. 74, 5383-5392

Lavin, M., Herendeen, P.S., Wojciechowski, M.F. (2005) Evolutionary Rates Analysis of
Leguminosae Implicates a Rapid Diversification of Lineages during the Tertiary. Syst.
Biol. 54, 575 — 594

Liu, H., Sadygov, R.G., Yates, J .R. (2004) A Model for Random Sampling and
Estimation of Relative Protein Abundance in Shotgun Proteomics. Anal. Chem. 76, 4193-
4201

Lu, P., Vogel, C., Wang, R., Yao, X., Marcotte, EM. (2007) Absolute protein expression
profiling estimates the relative contributions of transcriptional and translational
regulation. Nature Biotechnol. 25, 117-124

Margulies, M., Egholm, M., Altman, W.E., Attiya, S., Bader, J .S., Bemben, L.A., Berka,
J., Braverrnan, M.S., Chen, Y.-J., Chen, Z., Dewell, S.B., Du, L., Fierro, J.M., Gomes,
X.V., Godwin, B.C., He, W., Helgesen, S., Ho, C.H., Irzyk, G.P., Jando, S.C., Alenquer,
M.L.I., Jarvie, T.P., Jirage, K.B., Kim, J.-B., Knight, J.R., Lanza, J.R., Leamon, J.H.,
Lefkowitz, S.M., Lei, M., Li, J ., Lohman, K.L., Lu, H., Makhijani, V.B., McDade, K.E.,
McKenna, M.P., Myers, E.W., Nickerson, E., Nobile, J .R., Plant, R., Puc, B.P., Ronan,
M.T., Roth, G.T., Sarkis, G.J., Simons, J .F., Simpson, J .W., Srinivasan, M., Tartaro,
K.R., Tomasz, A., Vogt, K.A., Volkmer, G.A., Wang, S.H., Wang, Y., Weiner, M.P., Yu,

94

P., Begley, R.F., Rothberg, J .M. (2005) Genome sequencing in microfabricated high-
density picolitre reactors. Nature 437, 376-3 80

Maser, P., Thomine, S., Schroeder, J .I., Ward, J .M., Hirschi, K., Sze, H., Talke, I.N.,
Amtmann, A., Maathuis, F.J.M., Sanders, D., Harper, J .F., Tchieu, J ., Gribskov, M.,
Persans, M.W., Salt, D.E., Kim, S.A., Guerinot, ML. (2001) Phylogenetic Relationships
within Cation Transporter Families of Arabidopsis. Plant Physiol. 126, 1646-1667

McConn, M., Hugly, 8., Browse, J ., Somerville, C. (1994) A Mutation at the fad8 Locus
of Arabidopsis Identifies a Second Chloroplast [omega]-3 Desaturase. Plant Physiol. 106,
1609-1614

Miyagishima, S.Y., F roehlich, J .E., Osteryoung, K.W. (2006) PDVl and PDV2 mediate
recruitment of the dynamin-related protein ARCS to the plastid division site. Plant Cell
18, 2517-2530

Nesvizhskii, A.I., Keller, A., Kolker, E., Aebersold, R. (2003) A statistical model for
identifying proteins by tandem mass spectrometry. Anal. Chem. 75, 4646-4658

Picault, N., Hodges, M., Paimieri, L., Palmieri, F. (2004) The growing family of
mitochondrial carriers in Arabidopsis. Trends Plant Sci. 9, 138-146

Schmidt, U.G., Endler, A., Schelbert, S., Brunner, A., Schnell, M., Neuhaus, H.E., Marty-
Mazars, D., Marty, F., Baginsky, S., Martinoia, E. (2007) Novel tonoplast transporters

identified using a proteomic approach with vacuoles isolated from cauliflower buds. Plant
Physiol. 145, 216-229

Schnurr, J .A., Shockey, J .M., de Boer, G.-J., Browse, J .A. (2002) Fatty Acid Export from
the Chloroplast. Molecular Characterization of a Major Plastidial Acyl-Coenzyme A
Synthetase from Arabidopsis. Plant Physiol. 129, 1700-1709

Seigneurin-Bemy, D., Gravot, A., Auroy, P., Mazard, C., Kraut, A., Finazzi, G.,
Grunwald, D., Rappaport, F., Vavasseur, A., Joyard, J ., Richaud, P., Rolland, N. (2006)
HMAl , a New Cu-ATPase of the Chloro plast Envelope, Is Essential for Growth under
Adverse Light Conditions. J. Biol. Chem. 281, 2882-2892

Shevchenko, A., Wilm, M., Vorrn, 0., Mann, M. (1996) Mass Spectrometric Sequencing
of Proteins from Silver-Stained Polyacrylamide Gels. Anal. Chem. 68, 850-858

Shikanai, T., Muller-Moule, P., Munekage, Y., Niyogi, K.K., Pilon, M. (2003) PAAl, a
P-Type ATPase of Arabidopsis, Functions in Copper Transport in Chloroplasts. Plant
Cell 15, 1333-1346

Taylor, N.L., Heazlewood, J .L., Day, D.A., Millar, AH. (2005) Differential Impact of

Environmental Stresses on the Pea Mitochondrial Proteome. Mol. Cell. Proteomics 4,
1122-1133

95

Teng, Y.S., Su, Y.S., Chen, L.J., Lee, Y.J., Hwang, 1., Li, HM. (2006) Tic21 Is an
Essential Translocon Component for Protein Translocation across the Chloroplast Inner
Envelope Membrane. Plant Cell 18, 2247-2257

Weber, A.P.M., Schneidereit, J ., Voll, L.M. (2004) Using mutants to probe the in vivo
function of plastid envelope membrane metabolite transporters. J. Exp. Bot. 55, 123 l-
1244

Weber, A.P.M., Schwacke, R., Flugge, U]. (2005) Solute transporters of the plastid
envelope membrane. Annu. Rev. Plant Biol. 56, 133-164

Weber, A.P.M., Weber, K.L., Carr, K., Wilkerson, C., Ohlrogge, J.B. (2007) Sampling
the Arabidopsis Transcriptome with Massively Parallel Pyrosequencing. Plant Physiol.
144, 32-42

Wikstrdm, N., Savolainen, V., Chase, M.W. (2001) Evolution of the angiosperms:
calibrating the family tree. Proc. R. Soc. Lond., B, Biol. Sci. 268, 2211-2220

Yang, Y.W., Lai, K.N., Tai, P.Y., Li, W.H. (1999) Rates of nucleotide substitution in
angiosperrn mitochondrial DNA sequences and dates of divergence between Brassica and
other angiosperrn lineages. J. Mol. Evol. 48, 597-604

Yu, B., Xu, C.C., Benning, C. (2002) Arabidopsis disrupted in SQD2 encoding sulfolipid
synthase is impaired in phosphate-limited growth. Proc. Natl. Acad. Sci. U. S. A. 99,
5732-5737

Zhang, Z., Schwartz, 8., Wagner, L., Miller, W. (2000) A greedy algorithm for aligning
DNA sequences. J. Comp. Biol. 7, 203-214

Zhulidov, P.A., Bogdanova, E.A., Shcheglov, A.S., Vagner, L.L., Khaspekov, G.L.,
Kozhemyako, V.B., Matz, M.V., Meleshkevitch, E., Moroz, L.L., Lukyanov, S.A.,
Shagin, DA. (2004) Simple cDNA normalization using kamchatka crab duplex-specific
nuclease. Nucleic Acids Res. 32, e37

Zybailov, 8., Coleman, M.K., Florens, L., Washburn, MP. (2005) Correlation of
Relative Abundance Ratios Derived from Peptide Ion Chromatograms and Spectrum
Counting for Quantitative Proteomic Analysis Using Stable Isotope Labeling. Anal.
Chem. 77, 6218-6224

96

Chapter 4
Proplastid specific protein patterns revealed by cauliflower curd plastid envelope

proteomics

Andrea Bréiutigam and Andreas P. M. Weber

This chapter has been submitted to the Journal of Experimental Botany for publication

97

Abstract

Proplastids are undifferentiated plastids of meristematic tissues. Although
undifferentiated, they provide the meristematic cells with plastid synthesized amino acids
for protein synthesis, with fatty acids for membrane lipid production and purines and
pyrimidines for DNA and RNA synthesis, and. Both the precursor metabolites as well as
the products need to cross the plastid envelope. While proplastids are metabolically
highly active, they do depend on supply with reductants and energy from the remainder
of the cell. They only contain a rudimentary internal membrane system. Little is known
about proplastid membrane composition since meristematic tissues are generally limited
to the shoot apex and the root tip and therefore do not provide sufficient material for
organellar proteome analysis. To overcome this difficulty, cauliflower curd tissue was
chosen as a source for isolation of proplastid envelope membranes. The proplastid
envelope has a specific membrane protein composition different from those of
chloroplasts which is geared to importing precursor metabolites and exporting product
metabolites for the rapidly dividing cell. Moreover, the protein import complex has a
different composition and the internal membrane system is remarkably different from
chloroplast thylakoids. Cauliflower curd tissue is an excellent model system for the study
of plastid differentiation since not only white proplastidic cauliflower but also varieties

producing chromoplasts as well as green plastids are known.

98

Introduction

The apical and lateral meristems of plants serve as reservoirs for undifferentiated
stem cells that permit indeterminate plant growth. In all meristems only very few cells are
actual stem cells while the remainder of the meristem consists of cells which will
continue to divide for a finite number of divisions. Meristematic cells are usually
surrounded by a thin primary cell wall and they lack a central vacuole and differentiated
features. Since meristems do not contain photosynthetically active chloroplasts but
undifferentiated proplastids, they are sink tissues which rely on the source tissues for
supply with reduced carbon. Although undifferentiated, proplastids support cell growth
and division by providing building blocks for the cells, branched chain and aromatic
amino acids, fatty acids and lipids as well as nucleotide precursors. Proplastids can
differentiate into all plastid subtypes, such as chloroplasts, chromoplasts, and leukoplasts.
Since meristematic tissues are usually small, they have not been amenable to analysis by
methods requiring large amounts of tissue such as proteomics. In contrast, cauliflower
(Brassica oleracea ssp. botrytis) is an excellent model for proteomics of meristematic
tissues. The head of cauliflower, named the ‘curd’, represents a highly branched
abnormally proliferated meristem before floral transition (Kieffer et al., 1998). The
branch primordia only grow for a short time until they themselves become apical
meristems and produce additional branch primordia (Kieffer et al., 1998). This phenotype
has been genetically at least partially tied to floral identity gene mutations (Smith and
King, 2000) although most of the phenotypic variability of B. oleracea remains
unexplained (Labate et al., 2006). Only much later in development do these abnormally

proliferating meristems initiate floral development (Kieffer et al., 1998). Before initiation

99

of floral development, the curd remains an excellent source of meristematic or near
meristematic tissues and organelles. The plastids of cauliflower curd tissue (Joumet and
Douce, 1985) resemble proplastids of other meristematic tissues (summarized in (Kirk
and Tilney-Bassett, 1978). These plastids within the meristematic tissue appear to be
functional proplastids as they are able to undergo differentiation into chloroplasts as
demonstrated by ‘greening’ of the curds if the cauliflower is not properly covered either
by its own leaves or foil (commercial vegetable production guide, Oregon State
University, 2004). Cauliflower curd proplastids can also differentiate into chromoplasts
(Crisp et al., 1975; Lu et al., 2006).

Like the differentiated plastid types, proplastids are surrounded by two envelope
membranes, an outer and an inner envelope. Within these boundaries a protein-rich
matrix, the stroma, and a rudimentary internal membrane system are contained (J oumet
and Douce, 1985). The internal membrane system will differentiate into a prolamellar
body in etioplasts or into a thylakoid system in chloroplasts. Although the phloem stream
is capable of supplying a number of metabolites to the meristem, the only model for
proplastids, plastids of BY-2 cells, are enriched in enzymes for amino acid biosynthesis
indicating a metabolically active plastid (Baginsky et al., 2004). Biochemical analysis of
cauliflower curd proplastids have revealed that they rely on import of carbon as glucose-
6-phosphate and that store carbon as starch (Joumet and Douce, 1985; Emes and
Neuhaus, 1997; Neuhaus and Emes, 2000). A large supply of carbon is needed not only
to prepare the proplastid for differentiation but also for anabolic reactions such as fatty
acid or amino acid biosyntheses. Organic nitrogen is also delivered to the meristem by

the phloem but it is unknown how proplastids take up organic nitrogen (Emes and

100

Neuhaus, 1997; Neuhaus and Emes, 2000). Since plastids and proplastids are the site of
biosynthesis for branched chain and aromatic amino acids as well as lysine, arginine and
histidine, a high rate of nitrogen uptake into the proplastids is necessary to provide the
dividing and growing cell with sufficient amounts of amino acids. Since the proplastid is
the source for a number of essential metabolites for the dividing and growing
meristematic cells, efficient exchange of metabolites across the envelope necessitates the
presence of metabolite transport proteins. Whole tobacco BY-2 plastids have been
analyzed as a model system for undifferentiated heterotrophic plastids (Baginsky et al.,
2004)

In this work the focus is on the envelope proteome of cauliflower curd proplastids
to understand the adaptations of the envelope to proplastid-specific metabolism. The
transport protein complement indicates that proplastids are active cell factories which
imports precursor metabolites and exports products such as amino acids, fatty acids and
nucleotides. The proplastids not only contain a distinct set of transport proteins but also a
subset of protein import complex components as well as a distinct subset of thylakoid
resident proteins very unlike the thylakoid contamination found in chloroplast envelope
proteomics. The RNA accumulation pattern of proteins identified in this work is shifted
towards shoot apex expressed genes compared to proteins identified in leaf chloroplast
envelope proteomics. In summary, the proplastid envelope is specifically adapted to both

produce building blocks for the cell as well as differentiate upon external cues.

101

Material and Methods

Isolation and treatment of membranes

Cauliflower curd plastids were isolated according to (Joumet and Douce, 1985).
The plastids were lysed and plastid envelopes were prepared as described in (Douce and
Joyard, 1979; Keegstra and Yousif, 1986). The envelope membranes were separated into
chloroform/methanol (1:1, v/v) soluble and insoluble fiactions by diluting a resuspended
pellet of envelope membranes with ten volumes of solvent. After incubation on ice for
twenty minutes, phases were separated by centrifugation (20,000g; 20 min). The
chloroform-soluble fraction was then transferred to a fresh tube whereas the insoluble
fraction was recovered as a pellet after removal of the supernatant. Both fractions were
dried, washed with hexane to remove lipids, carefully dried again at room temperature
and redissolved in SDS-PAGE loading buffer. All fractions were separated by 12.5%
SDS-PAGE
Protein identification

Gels were briefly stained with Coomassie Brilliant Blue and then cut into ten
equally sized slices. Proteins within each slice were modified and digested with trypsin as
described by (Shevchenko et al., 1996). After extraction, the peptides were loaded by a
Waters nanoAcquity Sample Manager onto a Waters Symmetry C18 peptide trap (5 um,
180 um x 20mm) at a flow rate of 4 uL/min in 2% Acetonitrile/0.1%Formic Acid for 5
minutes. Separation of peptides occurred on a Waters BEH C18 nanoAcquity column
(1.7 um, 100 mm x 100mm) for 90 minutes and the peptides were injected into a
ThermoElectron LTQ-FTICR mass spectrometer with a flow rate of 300 nL/min (Buffer

A = 99.9% Water/0.1% Formic Acid, Buffer B = 99.9% Acetonitrile/0.1% Formic Acid:

102

gradient of 5% B to 40% B from 0 to 63 minutes, 40% B to 90% B from 63 to 71 minutes
and 5% B from 71 to 90 minutes). The top ten ions of each survey scan (resolution
50,000) were automatically dissociated by low energy collision. The resulting MS/MS
spectra were converted to a peak list by the BioWorks Browser version v3.2. Peptides
were identified by comparing all mass spectra libraries to a sequence database from
Arabidopsis thaliana, TAIR 8, (Swarbreck et al., 2008) using the Mascot search

algorithm v 2.2 (www.matrixsciencecom). Oxidation of methionine was permitted and

 

carbamidomethylcysteine was set as fixed peptide modifications. Two missed tryptic
sites were allowed and the MS/MS tolerance was set to 0.8kDa and the peptide tolerance
to +/- 10ppm. The results of the Mascot search algorithm were loaded into Scaffold® and
analyzed with peptide and protein prophet algorithms (Keller et al., 2002). Parameters
were set to 99% confidence for protein identification, 95% confidence for peptide
identification and at least two peptides identified for each protein. When multiple
matches were reported, the matches were analyzed manually and lower scoring matches
were discarded. For proteins discussed in detail, multiple matches to different members
of protein families were excluded by protein alignment and for plastid transport proteins
as well as import complex components, no identical tryptic peptides were identified in the
alignment and the experiment. Results were exported to MS Excel for further analysis.
Protein analysis

Annotation and classification of proteins was assigned based on information in
TAIR (Swarbreck et al., 2008), ARAMEMNON (Schwacke et al., 2003), and manual
curation of the literature. The predicted location, the number of transmembrane helices

and the prediction of betafold structures was retrieved from TAIR. Comparison to data

103

from other proteome projects was conducted using plprot (Kleffmann et al., 2006). For
comparison, data from pea chloroplast envelopes described in (Briiutigam et al., 20083),
which were produced exactly as described for cauliflower proplastid envelopes, was
extracted from the public repository PRIDE and reanalyzed according to the same
specifications (Supplemental Table 2)
Gene Expression Analysis

For each protein, the RNA expression data was extracted from the AtGenExpress
Developmental series (Schmid et al., 2005). The shoot apex expression level was
calculated as the arithmetic mean from experiments ATGE_6, ATGE_8, ATGE_29 and
ATGE_46 to ATGE_52. The leaf expression level was calculated as the average
expression from experiments ATGE_IO to ATGE_21. The ratio of expression was
calculated as the log(2) of the quotient expression(apex) divided by expression(shoot)

and compared to data from pea chloroplast envelope proteins (Supplemental Table 2).

Results and Discussion

Overall membrane proteome analysis

Table 4-1: Comparison of membrane protein identifications in proplastid and chloroplast
envelope; data for chloroplast envelopes extracted from (Brautigam et al., 2008b) and reanalyzed

 

known putative
membrane
total . transport transport
proteins . .
proteins proteins
pmp'asnd 174 56 16 7
envelope
chloroplast 321 110 16 21
envelope

 

104

Proteomic analysis of cauliflower curd plastid envelope membranes identified 174
distinct proteins (Table 4-S1). This number is significantly lower in comparison to what
has been reported for maize or pea chloroplast envelope membranes (Brautigam et al.,
2008a) (Table 4-1). Using a non-species specific database significantly lowers the
identification probability especially for proteins expressed at a low level (Brautigam et
al., 2008b) but the drawback of lower identification probability is superseded by superior
tissue availability. The proplastid envelope sample allowed the identification of sixteen
known and seven putative membrane proteins, 9.2% and 4% of the total proteins
identified, whereas the chloroplast envelope proteome sample allowed only the detection
of 5% and 6.5% respectively. The overall proportion one third of proteins with predicted
transmembrane helices is similar between both proteome samples. It was concluded that

the proplastid envelope proteome sample is suitable for transport protein analysis.

105

 

 

29.535 w
285:? Ea: mcmBEmE moon >55 - ca.- wcmenEoE 305 e w
. . m. d
85 2535 . u 0
e85:?. <zm ucm <20 $3250 528:3 ... 535%. SEEtE a M .W
. . m
285qu m m
m_mw5c>m <Zm new <20 mwum_>cmum - 09500.3: Hogan m. m.
mauoBEoc m w
£85585 539a «Bum oEEm - v.28 oEEm - 3.1m 0 M
A
3853:.
305:3 2a: mcmBEmE Bum 22529.03 - v.9. 2.2389: $388 09.
w m 3
35.888 95359 :53. m w RENEW ”...:mwB m m
. U
c_3m>uoEoc oc_co_5mE .m.
> o
mcozomm. mcEacm: _>5mE _>mocmcm-m _>mocmum-m 5:02 5 E u_2<m w
285:5. new 9.5% 322: 38233on - «to .m
£8553 Eon 05:5 .822: 395333 comet: 3590 who m.
A3923 w
ES: 385:? Bum >53 3923 we.
A3933 E9: m_mw5c>m $9383 -.ocmozamoca 282me 558 .73 .W
Eon 05:5 593 umcocmen w
>25;qu 3.25%.: em.
.08 mumgamoca $2.23 32393 4.. - fimhxa w
20582 2.035 5... “RE a: 58cc :2 w.
2268. 2.895 __m ao< a: >996 .E.z w
m_mm5c>m Em.m>_:cm c5262 Smemozmocamocaé 32305835 .258 .5532: r: W
£3553 555323 concave. 3m 30 385 - - 39o m m co .8 s
.6353 mumsamocommoucoa : ... . a m D _ v. e E0
£355..“ 533.0. 829.05 aéé 33297. .853 E0
2955
$62, an a
5 293 o toamcmb toamcmb E36 3395 5:03:85

m2zonfioE 8 2685 comment 5.55.5 Bozo—o 815: .8583 2552:: he... ...—a 2.29:3 Ewan—no...— 95 ..o 2:89:— toA—meauh. "N4. 039—.

106

Transport proteins

Compared to the membrane transport proteins identified in chloroplast envelope
proteome samples (F erro et al. 2003; Froehlich et al. 2003) different membrane transport
proteins are identified in the proplastid envelope proteome sample (Tables 4-Sl and 4-
82). Generally, one can divide the transport proteins in the proplastid envelope proteome
sample in those that import precursor metabolites for synthesis as well as those that
provide reduction power and energy and those which export products to the cytosol
(Table 4-2).
Transport proteins for carbon skeleton import

The phosphate translocators act as importers in proplastids (Table 4-2). Among
the phosphate translocators, two isoforms of the glucose-6-phosphate phosphate
translocator (GPT), the phosphoenolpyruvate phosphate translocator (PPT) and the
triosephosphate phosphate translocator (TPT) were identified in the proteome sample
(Table 4-Sl). Multiple matches of the same peptide to different family family members
were excluded by aligning the proteins and checking for identical tryptic peptides. The
phosphate translocator proteins are distinct from each other. The GPTs are not present in
sufficient quantities in leaves of either Arabidopsis (Ferro et al., 2003; Froehlich et al.,
2003), Pisum sativum (Brautigam et al., 2008a; Brautigam et al., 2008b) or Zea mays
(Brautigam et al., 2008a) to be detected by chloroplast envelope proteomics but at least
one isoform can be detected in tobacco BY-2 cell plastids (Baginsky et al., 2004). In
heterotrophic plastids, this transporter can supply the plastid with G6P for starch
biosynthesis as well as reduction power (Emes and Neuhaus, 1997; Kammerer et al.,

1998; Neuhaus and Emes, 2000) and carbon skeletons through the oxidative pentose
107

phosphate pathway. The GPT exchanges glucose-6-phosphate for either phosphate or
triosephosphate in a strict counter-exchange mode (Kammerer et al., 1998). Since two
GPT isoforms are identified (Table 4-Sl), it is possible that, in vivo, GPT isoforms may
have higher affinity for one of the counter substrates. Reduction power is generated in the
oxidative branch of the pentosephosphate pathway (OPPP) by oxidizing G6P to
gluconolactone which subsequently is converted into triosephosphate by a series of
reactions. The intermediates of this pathway serve as carbon skeletons for anabolic
reactions or the final product can serve as the counter-exchange substrate for the import
of G6P.

Transport proteins for amino acid production

amino

 
  
 

OEP16-Iik

.. 3
s V amino groups
........... pcarbon skeletons

        
 

 

   

    

, enemy amino acid
‘7 syntheslsli amino
7’ ....................... GAP-DH ......................... . reducing power acids

Figure 4-1: The proplastid envelope enables efficient exchange of metabolites needed for
amino acid biosynthesis; the transport proteins are printed italic, TPT triosephosphate phosphate
translocator, PHT phosphate transporter, NTT ATP ADP exchanger, GPT glucose-6-phosphate
phosphate translocator, PPT phosphoenolpyruvate phosphate translocator, DiT dicarboxylate
transporter, OEP outer envelope porin; enzymatic reactions are printed in boxes, Per pyruvate
kinase PPP pentosephosphate pathway, GAP-DH glyceraldehyde dehydrogenase complex;

108

The phosphoenolpyruvate phosphate translocator (PPT) is also present in the
proplastid envelope proteome sample (Table 4-2, Figure 4-1). The proplastid is the sole
cellular source of aromatic amino acids for the rapidly growing and dividing cells
(Neuhaus and Emes, 2000; Baginsky et al., 2004). Phosphoenolpyruvate transported by
PPT is a direct precursor of aromatic amino acids (Knappe et al., 2003; Voll et al., 2003)
together with erythrose-4-phosphate produced in the pentosephosphate pathway. PEP is
also the source for pyruvate needed for branched chain amino acid biosynthesis as the
plastids contain substantial pyruvate kinase activity (Andre and Benning, 2007; Andre et
al., 2007). In addition to branched chain and aromatic amino acids the plastids are also
the sole source for arginine and lysine as well as histidine and threonine. The amino acids
not only require carbon, but also amino groups for their production. The dicarboxylate
carriers DiTl and DiT2 are both identified (Table 4-Sl, Figure 4-1). Ditl can exchange
2-oxoglutarate for malate whereas DiT2 is more specific for glutamate malate exchange
(Weber et al., 1995; Taniguchi et al., 2002; Renne et al., 2003; Schneidereit et al., 2006;
Weber and Fischer, 2007). Working in concert these proteins are suited to import
glutamate as a source for organic nitrogen and to export 2-oxoglutarate for further
metabolic reactions, for example in mitochondria (Figure 4-1). Glutamate is the major
amino acid transported in the phloem (Weibull and Melin, 1990; Lohaus and Moellers,
2000). The identification of both dicarboxylate carriers solves the question how nitrogen

is imported into heterotrophic proplastids (Figure 4-1) (N euhaus and Emes, 2000).

109

Transport proteins for energy, reducing power and anabolic precursors

TPT has also been identified with a high spectral count (Table 4-2, Table 4-Sl).
This protein has been considered the hallmark transport protein of chloroplasts (Flugge
and Heldt, 1984; Schneider et al., 2002; F lugge et al., 2003; Weber et al., 2005) and has
been assumed to be absent from heterotrophic plastids (Flugge et al., 2003). In
chloroplasts, TPT exports the main product of the reductive pentosephosphate pathway,
triosephosphate, into the cytosol in exchange for inorganic phosphate. In vitro
characteristics indicate that TPT can also exchange 3-PGA for triosephosphate
(Loddenkotter et al., 1993). Since cauliflower proplastids lack a functional photosynthetic
thylakoid membrane system (Table 4-Sl) (Joumet and Douce, 1985), they are unlikely to
produce large amounts of triosephosphate for export through photosynthetic carbon
assimilation. Possibly, the TPT acts as a reduction equivalent shuttle, as has been
proposed for TPT in chloroplasts of C4plants (Bréiutigarn et al., 2008a) and for C3 leaves
(Heineke et al., 1991). In this case, a GAP-DH in the chloroplast produces NADPH by
oxidizing triosephosphate to 3-PGA which can be exported to the cytosol again (Figure
4-1). The resulting reducing power can be used for anabolic reactions such as amino acid
synthesis (Figure 4-1).

Many if not all of the anabolic reactions of starch, amino acid and nucleotide
synthesis require energy in form of ATP. Since cauliflower curd proplastids lack
chlorophyll (Joumet and Douce, 1985) they are unlikely to produce substantial amounts
of ATP by photophosphorylation. Two ATP ADP exchangers (NTTs) are detected with
high absolute spectral counts among the transport proteins (Table 4-Sl, Figure 4-1).

These proteins provide the proplastids with energy by importing ATP in counter-
110

exchange for ADP. The consumption of ATP produces inorganic phosphate which is not
transported by NTTs. This phosphate imbalance may be countered by the putative
phosphate transport protein PHT3.1 of proplastids (Figure 4-1). This putative phosphate
transport protein is also present in BY2 cell plastids as well as chloroplast envelopes
(Baginsky et al., 2004; Kleffmann et al., 2004; Bréiutigam et al., 2008a).

The proplastid envelope also harbors a S-adenosylmethionine (SAM) carrier
which provides the plastid with SAM for a multitude of methylation reactions (Bouvier et
al., 2006).The main biochemical effect of loss of SAM in the plastids is decreased leaf
chlorophyll content leading to stunted growth (Bouvier et al., 2006) but it likely also
effects meristematic tissues directly since it is present on proplastid envelopes.

Transport proteins for export of products

Although the plastids are the sole source of a number of amino acids, the amino
acid exporter at the inner envelope has not been characterized at the molecular level.
However, the transport pore through the outer envelope is known, the outer envelope
porin OEP16 (Pohlmeyer et al., 1997). OEP 16 can transport a broad spectrum of amino
acids but excludes sugars and sugar phosphates (Pohlmeyer et al., 1997) and it is one of
the proteins with the highest spectral count in the proteome sample (Table 4-Sl). Two
OEP16 homologues, AGIl and AG12 (Murcha et al., 2007), are also identified with a
high spectral count (Table 4-Sl). At least one of these proteins is localized in the inner
envelope (Brautigam et al. submitted) and is considered a candidate protein for amino
acid transport at the inner envelope (Figure 4-1).

Apart from amino acids for protein biosynthesis, the cytosol also depends on

purins as well as pyrimidins for DNA synthesis since, for both pathways, the majority of

111

the enzymes is localized in the plastids (Zrenner et al., 2006). Purin derivatives, more
specifically, adenylates can be exported through an adenylate uniporter, Brittle] , in dicots
(Kirchberger et al., 2008). Brittlel of Arabidopsis is mainly expressed in the roots with
the most intense expression being in the root tip in the region of the meristem. GUS
activity was not readily visible in the shoot meristem (Kirchberger et al., 2008) although
publicly available expression data (Schmid et al., 2005) visualized in the eFP browser
(Winter et al., 2007) indicates strong expression in the shoot apex confirming the
proplastid envelope proteome results (Table 4-Sl). For pyrimidine biosynthesis, dihydro-
orotate has to be exported to be reduced to orotic acid at the mitochondrial membrane;
the transport protein is no known (Zrenner et al., 2006). The exporter for the pyrimidine
pathway intermediates is also unknown. Possibly, one of the membrane proteins of
unknown function (Table 4-1) is either of the unknown transport proteins involved in
pyrimidine metabolism. None of the proteins is similar to the characterized
purine/pyrimidine transport proteins residing outside of the plastid (Gillissen et al., 2000).
In addition to nucleotide precursors for DNA and RNA synthesis and amino acids for
protein biosynthesis, fatty acids and membrane lipids are needed to sustain rapid cell
proliferation. Fatty acids and membrane lipids are produced in a complex metabolic
network localized in multiple compartments (Benning et al., 2006). Within the proteome
sample, one of three components of the TGD transporter which imports phosphatidic acid
from the ER (Awai et al., 2006) was identified as well as a long acyl chain CoA synthase

implicated in fatty acid transport (Schnurr et al., 2002) (Table 4-2).

112

The protein import complex

 

 

 

 

 

 

 

 

 

 

,2 1.0

'5

o

.r:

i”,

c

.9. F—

3 0.5. 7

Q.

i x

9 §

; §§§®

g 0.0 e m

7'5

.9

33

e

3

E -0.5-

E

”6

o

:57

O

- -1-0 I I I I I I I I I 1 I I I I I—
Q'Lgbt\\\\\0‘b,\\$9\3 ’b‘D’L
(L «'5 a . \N '5 Q «‘3 X 9 ‘3 Q)

«00 «00 «010664 C918 «\0 «0° {\C’vl' ‘62.,‘00 (\o‘fl’ («P9639 (‘0 ‘<‘°

Figure 4-2: Expression of the protein import complex components varies between leaves
and the shoot apex; white bars, identified in proplastids only, black bars, identified in
chloroplasts only, hatched bars, identified in both samples; for each protein, the gene expression
of the corresponding Arabidopsis gene was extracted from the AtGenExpress database; the
expression values were calculated as loglO of the ratio between expression in the shoot apex and
expression in the leaves, i.e. positive values indicate a higher expression in the shoot apex;

The protein import complex components identified in proplastid envelopes
include all three import receptors, Toc120, Toc132 and Toc159, two smaller GTPases
ToC33 and ToC34 and the main channel through the outer envelope Toc75-III as well as

Toc75-V (Figure 4-2). Of the inner envelope complex, Tie] 10, Tic20, Tic40 and Tic22 as

113

well as the import chaperone Hsp93 (also called ClpC) were found (Figure 4-2). In
contrast, in chloroplast envelope proteomics, Toc120 and Toc132 have not been
identified so far (Ferro et al., 2003; Froehlich et al., 2003; Kleffmann et al., 2004;
Kleffmann et al., 2007; Bréiutigam et al., 2008a; Brautigam et al., 2008b). For the inner
envelope complex, additional components, Tic32, Tic55 and Tic62, can be identified in
chloroplast proteomics (Table 4-82) e.g.(Bréiutigam et al., 2008b). The expression pattern
of the protein import complex components was analyzed in detail and compared to the
identification of components in chloroplast and proplastid proteomics. Expression values
for the genes were extracted from publicly available microarray data in Arabidopsis and
the ratio of shoot apex to leaf expression was calculated and normalized by logarithmic
transformation. The proteins with the highest shoot apex/leaf expression ratio, the
alternative import receptors Toc120 and Tocl32, are both identified only in the envelope
proteome sample of cauliflower (Figure 4-2). In contrast, the proteins with the lowest
ratio, i.e. the proteins expressed more highly in the leaf, are the import complex
components Tic55 and Tic62, which both could only be identified in pea leaf envelopes
(Figure 4-2, Table 4-82). Toc33 identified in proplastids only and Tic32, which was only
identified in chloroplast envelopes, is of approximately equal expression between both
tissues. Toc120 and Toc132 are alternative import receptors to Toc159 (Jackson-Constan
and Keegstra, 2001; Ivanova et al., 2004; Kubis et al., 2004). If both are knocked out,
plants are severely compromised in their ability to establish growth, contain almost no
chlorophyll and have damaged chloroplasts (Kubis et al., 2004). It has been proposed that
Toc132 and Toc120 are essential for the of import housekeeping proteins into plastids

whereas Toc159 is specialized for the import of proteins involved in photosynthesis

114

(Ivanova et al., 2004; Kubis et al., 2004; Kessler and Schnell, 2006), although the
distinction of imported products is not absolute (Kubis et al., 2004). Since to date neither
Toc132 nor Toc120 have been identified by proteomics in either chloroplast or
chloroplast envelope studies, it may be concluded that the protein levels of both proteins
are rather low in leaves, precluding identification by proteomics, whereas their higher
expression in meristematic tissue (Figure 4-2) allows identification. Indeed based on their
spectral abundance, Toc159, Tocl32 and TochO appear to be equally expressed (Table
4-Sl). Toc75, TicllO, Tic20, Tic22 and Tic40 are identified in proplastids as well as
chloroplast envelopes in line with their housekeeping function (Figure 4-2). The members
of the redox sensing complex however, Tic32, Tic62 and Tic55 cannot be identified in
proplastid envelope proteomics. This may either be due to a lower expression level for
example of Tic55 and Tic62 (Figure 4-2) or to a lack of membrane association of any of
the components. Possibly, the undifferentiated proplastid without a completely developed
thylakoid system is not yet poised to sense and relay the redox status mostly generated by

photosynthesis.

115

Overall proteome analysis

 
   
  
  
 
 
  

endomembrane
. . s stem
mitochondrial y
cytosol
unclassified
nuclear
chaperone
division
9 n zyme
translation .
transporter
putative / porin
signalling
redox

Figure 4-3: Overall proteome contents in plastid envelopes; all proteins were annotated and
classified (Tables4- SI and 4-82), (A) proplastid envelope proteome, (B) chloroplast envelope
proteome

116

Figure 4-3 (cont’d)

endomembrane
system

  
   
   

B mitochondrial

chaperone
. division

unclassified

  

enzyme

  
 
 

translation protein import

tra nsporter

putative protease

transporter redox

signalling

After establishing patterns of the transport proteins for metabolites and protein
import complex in the proplastid envelope, we compared the remaining proteins
identified in proplastid envelopes (Table 4-Sl) with those identified in chloroplast
envelopes (Table 4-S2).

117

The annotation and classification of proteins identified revealed approximately
similar proportions of enzymes in both proplastid and chloroplast envelopes (Figure 4-3A
and 4-3B). However, the types of enzymes are different. The proplastid envelope sample
does not contain sufficient amounts of enzymes of the PPP including Rubisco to be
identified by proteomics (Table 4-Sl) in contrast to the chloroplast envelope sample
which contains Rubisco as a dominant protein and seven enzymes of the PPP (Table 4-
82). Both envelope samples contain enzymes for pigment biosynthesis but enzymes for
tocopherol biosynthesis can only be identified in chloroplast envelopes. The proplastid
sample does contain two peroxiredoxins (Dietz et al., 2002; Horling et al., 2003) and
sixteen proteins known to reside in thylakoids in chloroplasts (Peltier et al., 2000; Peltier
et al., 2002; Friso et al., 2004; Peltier et al., 2004) (Figure 4-3A). The thylakoid resident
proteins identified represent only a subset of known thylakoid proteins (Table 4-83). In
the cauliflower proteome, all soluble subunits of the ATP synthase are identified, alpha,
beta, gamma, delta, epsilon and F. In addition only very few others were found, PsbP and
PsbS, two photosystem II subunits, two FtsH proteases, FtsH] and FtsH8, a CAAX type
protease and a cytochrome b6f complex subunit, cyt6 (Table 4-S3). In contrast, in other
envelope projects the spectrum of thylakoid proteins identified with the envelope is much
broader and also includes many subunits of the photosystems and light harvesting
complex proteins (Table 4-S2) (see for example (Brautigam et al., 2008a)). Proteins of
the thylakoid membrane system always appear in envelope proteome analysis (Ferro et
al., 2003; Froehlich et al., 2003; Brautigam et al., 2008a; Brautigam et al., 2008b). The
only proteins abundant enough to be identified by proteomics in cauliflower proplastid

envelopes are components of the ATPase, two PS II components and no PS I

118

components. One of the two PSII components that can be detected are involved in redox
protection. PsbS plays a key role in non-photochemical quenching (Li et al., 2000)
whereas PsbP is involved in PS 11 assembly and stability (Ishihara et al., 2007). Both
proteins accumulate in etioplasts compared to chloroplasts (Kanervo et al., 2008). The
absence of enzymes of the PPP and the limitation of thylakoid resident proteins confirm
the absence of carbon fixation in proplastids already indicated by the absence of a
developed thylakoid system and the absence of chlorophyll (Joumet and Douce, 1985).
On the other hand, the proplastids appear poised to differentiate; they are equipped to
both dissipate the proton gradient and handle the redox stress generated during
photosynthetic electron transfer.

The annotation and classification of all identified proteins (Figure 4-3A) also
revealed that about 20% of the proteins can be assigned to a location other than the
plastid, such as the mitochondria, the endomembrane system, the cytosol or the nucleus.
Hence, the proportion of putative contaminants in cauliflower curd envelope membranes
is approximately two-fold higher as compared to pea envelope proteome samples (Figure
4-3B). It has been noted earlier that cauliflower curd plastids carry a substantial
mitochondrial contamination, as determined by marker enzyme assays (Joumet and
Douce, 1985), which is confirmed by our proteomics study. Putative contaminants
identified in cauliflower bud plastids differ from those that are routinely found in leaf
chloroplasts preparations, such as glycine decarboxylase complex subunits (Brautigam et
al., 2008a), which is absent from the proplastid preparation. In addition to the proteins
typically encountered in plastid isolations, such as histones and abundant proteins of the

mitochondrial respiratory chain and ATPases, the cauliflower proteome sample contained

119

unusual contaminants, e. g. a prohibitin and a Mcm4-like protein (Table 4-Sl) not
identified in other plastid envelope proteome. Both of these are highly expressed in the
shoot apex (Table 4-Sl), which may explain their presence in this sample. The total
expression analysis of the proteins identified in proplastid and chloroplast envelopes also

showed a shift towards shoot apex expressed genes (Figure 4-Sl).

 

 

 

 

 

 

 

2. proplastid envelope chloroplast envelope
proteins proteins
To
a 3 1.
a i: I
1‘ 9 . *. _ , _
'6 8 ° ' i i i
A o
E E I
3 ° '1‘
.2‘ -—

 

Figure 4—Sl: RNA accumulation patterns between shoot apex and leaves for orthologous
genes in Arabidopsis visualized by a whisker plot; for each orthologue of both proplastid and

chloroplast identified proteins, the logo of the expression ratio between shoot apex and leaf was
calculated, i.e. higher values indicate a shift of expression towards the shoot apex

Conclusion

Proteomics of the proplastid membrane system identified proplastid-specific
patterns: The transport proteins of the envelopes are specifically adapted to supply the
proliferating cells of the meristems with the necessary building blocks and the proteins
import complex is geared to import proteins involved in metabolism rather than
photosynthesis. Overall proteome analysis indicates proplastids are not only equipped to

support cellular growth but also adapted to differentiate upon light exposure. Cauliflower

120

curd will be well suited to study plastid differentiation. Untreated cauliflower curd
contains proplastids, differentiation into chloroplasts can be induced by exposure to light
and differentiation into chromoplasts can be studied in the orange variety of cauliflower

allowing the study of three different plastid types in the same system.
Supplementary material

Supplementary table 4-1: Proteins identified in proplastid envelopes from cauliflower
curd tissue

Supplementary table 4-2: Proteins identified in chloroplast envelopes from pea leaves
(Brautigam et al., 2008a) were extracted from PRIDE and reanalyzed

Supplementary table 4-3: Thylakoid proteins identified in envelopes of proplastids and

chloroplasts
Acknowledgements

We would like to acknowledge expert help from the Michigan State University Research

Technology facility.

121

References

Andre, C and Benning, C. 2007. Arabidopsis seedlings deficient in a plastidic pyruvate
kinase are unable to utilize seed storage compounds for germination and establishment.
Plant Physiology 145, 1670-1680.

Andre, C, F roehlich, JE, Moll, MR and Benning, C. 2007. A heteromeric plastidic
pyruvate kinase complex involved in seed oil biosynthesis in Arabidopsis. Plant Cell 19,
2006-2022.

Awai, K, Xu, CC, Tamot, B and Benning, C. 2006. A phosphatidic acid-binding protein
of the chloroplast inner envelope membrane involved in lipid trafficking. Proceedings of
the National Academy of Sciences of the United States of America 103, 10817-10822.

Baginsky, S, Siddique, A and Gruissem, W. 2004. Proteome analysis of tobacco bright
yellow-2 (BY-2) cell culture plastids as a model for undifferentiated heterotrophic
plastids. Journal of Proteome Research 3, 1128-1137.

Benning, C, Xu, CC and Awai, K. 2006. Non-vesicular and vesicular lipid trafficking
involving plastids. Current Opinion in Plant Biology 9, 241-247.

Bouvier, F, Linka, N, Isner, JC, Mutterer, J, Weber, APM and Camara, B. 2006.
Arabidopsis SAMTI defines a plastid transporter regulating plastid biogenesis and plant
development. Plant Cell 18, 3088-3105.

Brautigam, A, Hofmann-Benning, S and Weber, APM. 2008a. Comparative Proteomics
of Chloroplast Envelopes from C3 and C4 Plants Reveals Specific Adaptations of the
Plastid Envelope to C4 Photosynthesis and Candidate Proteins Required for Maintaining
C4 Metabolite Fluxes. Plant Physiology 148, 568-579.

Brautigam, A, Shrestha, RP, Whitten, D, Wilkerson, CG, Carr, KM, Froehlich, J E and
Weber, APM. 2008b. Comparison of the use of a species-specific database generated by
pyrosequencing with databases from related species for proteome analysis of pea
chloroplast envelopes. Journal of Biotechnology 136, 44-53.

Crisp, P, Walkey, DGA, Bellman, E and Roberts, E. 1975. Mutation Affecting Curd
Color In Cauliflower (Brassica-Oleracea L Var Botrytis Dc). Euphytica 24, 173-176.

Dietz, KJ, Horling, F, Konig, J and Baier, M. 2002. The function of the chloroplast 2-
cysteine peroxiredoxin in peroxide detoxification and its regulation. Journal of

Experimental Botany 53, 1321-1329.

Douce, R and Joyard, J. 1979. Isolation and Properties of the Envelope of Spinach
Chloroplasts: Ellis Horwood.

122

Emes, MJ and Neuhaus, HE. 1997. Metabolism and transport in non-photosynthetic
plastids. Journal of Experimental Botany 48, 1995-2005.

Ferro, M, Salvi, D, Brugiere, S, Miras, S, Kowalski, S, Louwagie, M, Garin, J, Joyard, J
and Rolland, N. 2003. Proteomics of the chloroplast envelope membranes from
Arabidopsis thaliana. Molecular & Cellular Proteomics 2, 325-345.

F lugge, UI, Hausler, RE, Ludewig, F and Fischer, K. 2003. Functional genomics of
phosphate antiport systems of plastids. Physiologia Plantarum 118, 475-482.

Flugge, UI and Heldt, HW. 1984. The Phosphate-Triose Phosphate-Phosphoglycerate
Translocator of the Chloroplast. Trends in Biochemical Sciences 9, 530-533.

Friso, G, Giacomelli, L, Ytterberg, AJ, Peltier, JB, Rudella, A, Sun, Q and van Wijk, KJ.
2004. In-depth analysis of the thylakoid membrane proteome of Arabidopsis thaliana

chloroplasts: New proteins, new fimctions, and a plastid proteome database. Plant Cell
16, 478-499.

Froehlich, JE, Wilkerson, CG, Ray, WK, McAndrew, RS, Osteryoung, KW, Gage, DA
and Phinney, BS. 2003. Proteomic study of the Arabidopsis thaliana chloroplastic

envelope membrane utilizing alternatives to traditional two-dimensional electrophoresis.
Journal of Proteome Research 2, 413-25.

Gillissen, B, Burkle, L, Andre, B, Kuhn, C, Rentsch, D, Brandl, B and Frommer, WB.
2000. A new family of high-affinity transporters for adenine, cytosine, and purine
derivatives in arabidopsis. Plant Cell 12, 291-300.

Heineke, D, Riens, B, Grosse, H, Hoferichter, P, Peter, U, Flugge, UI and Heldt, HW.
1991. Redox Transfer Across The Inner Chloroplast Envelope Membrane. Plant
Physiology 95, 1131-1137.

Horling, F, Lamkemeyer, P, Konig, J, F inkemeier, I, Kandlbinder, A, Baier, M and Dietz,
KJ. 2003. Divergent light-, ascorbate-, and oxidative stress-dependent regulation of

expression of the peroxiredoxin gene family in Arabidopsis. Plant Physiology 131, 317-
325.

Ishihara, S, Takabayashi, A, Ido, K, Endo, T, Ifuku, K and Sato, F. 2007. Distinct
functions for the two PsbP-like proteins PPLI and PPL2 in the chloroplast thylakoid
lumen of arabidopsis. Plant Physiology 145, 668-679.

Ivanova, Y, Smith, MD, Chen, KH and Schnell, DJ. 2004. Members of the Toc159

import receptor family represent distinct pathways for protein targeting to plastids.
Molecular Biology of the Cell 15, 3379-3392.

123

Jackson-Constan, D and Keegstra, K. 2001. Arabidopsis genes encoding components of
the chloroplastic protein import apparatus. Plant Physiology 125, 1567-1576.

Joumet, EP and Douce, R. 1985. Enzymic Capacities Of Purified Cauliflower Bud
Plastids For Lipid-Synthesis And Carbohydrate-Metabolism. Plant Physiology 79, 45 8—
467.

Kammerer, B, Fischer, K, Hilpert, B, Schubert, S, Gutensohn, M, Weber, A and Flugge,
UI. 1998. Molecular characterization of a carbon transporter in plastids from
heterotrophic tissues: the glucose 6-phosphate/phosphate antiporter. Plant Cell 10, 105-
17.

Kanervo, E, Singh, M, Suorsa, M, Paakkarinen, V, Aro, E, Battchikova, N and Aro, EM.
2008. Expression of protein complexes and individual proteins upon transition of
etioplasts to chloroplasts in pea (Pisum sativum). Plant And Cell Physiology 49, 396-410.

Keegstra, K and Yousif, AE. 1986. Isolation And Characterization Of Chloroplast
Envelope Membranes. Methods in Enzymology 118, 316-325.

Keller, A, Nesvizhskii, AI, Kolker, E and Aebersold, R. 2002. Empirical statistical model
to estimate the accuracy of peptide identifications made by MS/MS and database search.
Analytical Chemistry 74, 5383-5392.

Kessler, F and Schnell, DJ. 2006. The function and diversity of plastid protein import
pathways: A multilane GTPase highway into plastids. Traffic 7, 248-257.

Kieffer, M, Fuller, MP and Jellings, AJ. 1998. Explaining curd and spear geometry in
broccoli, cauliflower and 'romanesco': quantitative variation in activity of primary
meristems. Planta 206, 34-43.

Kirchberger, S, Tjaden, J H and Neuhaus, E. 2008. Characterization of the Arabidopsis
Brittlel transport protein and impact of reduced activity on plant metabolism. The Plant
Journal 56, 51-63.

Kirk, JTO and Tilney-Bassett, RAE. 1978. The plastids: Their chemistry, structure,
growth and inheritance. Amsterdam/Oxford: Elsevier.

Kleffmann, T, Hirsch-Hoffmann, M, Gruissem, W and Baginsky, S. 2006. plprot: A
comprehensive proteome database for different plastid types. Plant And Cell Physiology
47, 432-436.

Kleffmann, T, Russenberger, D, von Zychlinski, A, Christopher, W, Sjolander, K,

Gruissem, W and Baginsky, S. 2004. The Arabidopsis thaliana chloroplast proteome
reveals pathway abundance and novel protein functions. Current Biology 14, 354-362.

124

Kleffmann, T, von Zychlinski, A, Russenberger, D, Hirsch-Hoffmann, M, Gehrig, P,
Gruissem, W and Baginsky, S. 2007. Proteome dynamics during plastid differentiation in
rice. Plant Physiology 143, 912-923.

Knappe, S, Lottgert, T, Schneider, A, Vol], L, F lugge, UI and Fischer, K. 2003.
Characterization of two functional phosphoenolpyruvate/phosphate translocator (PPT)
genes in Arabidopsis-AtPPTl may be involved in the provision of signals for correct
mesophyll development. Plant Journal 36, 411-420.

Kubis, S, Patel, R, Combe, J, Bedard, J, Kovacheva, S, Lilley, K, Biehl, A, Leister, D,
Rios, G, Koncz, C and Jarvis, P. 2004. Functional specialization amongst the Arabidopsis
Toc159 family of chloroplast protein import receptors. Plant Cell 16, 2059-2077.

Labate, JA, Robertson, LD, Baldo, AM and Bjorkman, T. 2006. Inflorescence identity

gene alleles are poor predictors of inflorescence type in broccoli and cauliflower. Journal
Of The American Society For Horticultural Science 131, 667-673.

Li, XP, Bjorkman, O, Shih, C, Grossman, AR, Rosenquist, M, Jansson, S and Niyogi,
KK. 2000. A pigment-binding protein essential for regulation of photosynthetic light
harvesting. Nature 403, 391-395.

Loddenkotter, B, Kammerer, B, Fischer, K and Flugge, U]. 1993. Expression of the
Functional Mature Chloroplast Triose Phosphate Translocator in Yeast Internal
Membranes and Purification of the Histidine-Tagged Protein by a Single Metal-Affinity
Chromatography Step. Proceedings of the National Academy of Sciences of the United
States of America 90, 2155-2159.

Lohaus, G and Moellers, C. 2000. Phloem transport of amino acids in two Brassica napus

L. genotypes and one B-carinata genotype in relation to their seed protein content. Planta
211, 833-840.

Lu, S, Van Eck, J, Zhou, X, Lopez, AB, O'Halloran, DM, Cosman, KM, Conlin, BJ,
Paolillo, DJ, Garvin, DF, Vrebalov, J, Kochian, LV, Kupper, H, Earle, ED, Cao, J and Li,
L. 2006. The cauliflower or gene encodes a Dual cysteine-rich domain-containing protein
that mediates high levels of beta-carotene accumulation. Plant Cell 18, 3594-3605.

Murcha, MW, Elhafez, D, Lister, R, Tonti-Filippini, J, Baumgartner, M, Philippar, K,
Carrie, C, Mokranjac, D, 8011, J and Whelan, J. 2007. Characterization of the Preprotein

and Amino Acid Transporter Gene Family in Arabidopsis. Plant Physiology 143, 199-
212.

Neuhaus, HE and Emes, MJ. 2000. Nonphotosynthetic metabolism in plastids. Annual
review of plant physiology and plant molecular biology 51, 111-140.

Peltier, JB, Emanuelsson, O, Kalume, DE, Ytterberg, J, Friso, G, Rudella, A, Liberles,
DA, Soderberg, L, Roepstorff, P, von Heijne, G and van Wijk, KJ. 2002. Central

125

functions of the lumenal and peripheral thylakoid proteome of Arabidopsis determined by
experimentation and genome-wide prediction. Plant Cell 14, 211-36.

Peltier, JB, F riso, G, Kalume, DE, Roepstorff, P, Nilsson, F, Adamska, I and van Wijk,
KJ. 2000. Proteomics of the chloroplast: systematic identification and targeting analysis
of lumenal and peripheral thylakoid proteins. Plant Cell 12, 319-41.

Peltier, J B, Ytterberg, AJ, Sun, Q and van Wijk, KJ. 2004. New functions of the
thylakoid membrane proteome of Arabidopsis thaliana revealed by a simple, fast, and

versatile fractionation strategy. Journal of Biological Chemistry 279, 49367-493 83.

Pohlmeyer, K, 8011, J, Steinkamp, T, Hinnah, S and Wagner, R. 1997. Isolation and
characterization of an amino acid-selective channel protein present in the chloroplastic

outer envelope membrane. Proceedings of the National Academy of Sciences of the
United States of America 94, 9504-9509.

Renne, P, Dressen, U, Hebbeker, U, Hille, D, Flugge, UI, Westhoff, P and Weber, APM.
2003. The Arabidopsis mutant dct is deficient in the plastidic glutamate/malate
translocator DiT2. Plant Journal 35, 316-331.

Schmid, M, Davison, TS, Henz, SR, Pape, UJ, Demar, M, Vingron, M, Scholkopf, B,
Weigel, D and Lohmann, JU. 2005. A gene expression map of Arabidopsis thaliana
development. Nature Genetics 37, 501-506.

Schneider, A, Hausler, RE, Kolukisaoglu, U, Kunze, R, van der Graaff, E, Schwacke, R,
Catoni, E, Desimone, M and Flugge, UI. 2002. An Arabidopsis thaliana knock-out
mutant of the chloroplast triose phosphate/phosphate translocator is severely

compromised only when starch synthesis, but not starch mobilisation is abolished. Plant
Journal 32, 685-699.

Schneidereit, J, Hausler, RE, F iene, G, Kaiser, WM and Weber, APM. 2006. Antisense
repression reveals a crucial role of the plastidic 2-oxoglutarate/malate translocator DiTl
at the interface between carbon and nitrogen metabolism. Plant Journal 45, 206-224.

Schnurr, JA, Shockey, JM, de Boer, G-J and Browse, J A. 2002. Fatty Acid Export from
the Chloroplast. Molecular Characterization of a Major Plastidial Acyl-Coenzyme A
Synthetase from Arabidopsis. Plant Physiology 129, 1700-1709.

Schwacke, R, Schneider, A, van der Graaff, E, Fischer, K, Catoni, E, Desimone, M,
Frommer, WB, Flugge, U1 and Kunze, R. 2003. ARAMEMNON, a novel database for
Arabidopsis integral membrane proteins. Plant Physiology 131, 16-26.

Shevchenko, A, Wilm, M, Vorm, O and Mann, M. 1996. Mass Spectrometric Sequencing
of Proteins from Silver-Stained Polyacrylamide Gels. Analytical Chemistry 68, 850-858.

126

Smith, LB and King, GJ. 2000. The distribution of BoCAL-a alleles in Brassica oleracea
is consistent with a genetic model for curd development and domestication of the
cauliflower. Molecular Breeding 6, 603-613.

Swarbreck, D, Wilks, C, Larnesch, P, Berardini, TZ, Garcia-Hemandez, M, Foerster, H,
Li, D, Meyer, T, Muller, R, Ploetz, L, Radenbaugh, A, Singh, S, Swing, V, Tissier, C,
Zhang, P and Huala, E. 2008. The Arabidopsis Information Resource (TAIR): gene
structure and function annotation. Nucleic Acids Research 36, D1009-D1014.

Taniguchi, M, Taniguchi, Y, Kawasaki, M, Takeda, S, Kato, T, Sato, S, Tahata, S,
Miyake, H and Sugiyama, T. 2002. Identifying and characterizing plastidic 2-
oxoglutarate/malate and dicarboxylate transporters in Arabidopsis thaliana. Plant And
Cell Physiology 43, 706-717.

Voll, LM, Hausler, RE, Hecker, R, Weber, APM, Weissenbock, G, Fiene, G,
Waffenschmidt, S and F lfigge, UI. 2003. The phenotype of the Arabidopsis cuel mutant
is not simply caused by a general restriction of the shikimate pathway. Plant Journal 36,

301ff.

Weber, A, Menzlaff, E, Arbinger, B, Gutensohn, M, Eckerskom, C and Flugge, UI. 1995.
The 2-oxoglutarate/malate translocator of chloroplast envelope membranes: molecular
cloning of a transporter containing a 12-helix motif and expression of the functional
protein in yeast cells. Biochemistry 34, 2621-7.

Weber, APM and Fischer, K. 2007. Making the connections - The crucial role of
metabolite transporters at the interface between chloroplast and cytosol. F EBS Letters
581, 2215-2222.

Weber, APM, Schwacke, R and Flugge, UI. 2005. Solute transporters of the plastid
envelope membrane. Annual Review of Plant Biology 56, 133-164.

Weibull, J and Melin, G. 1990. Free Amino-Acid Content Of Phloem Sap From Brassica
Plants In Relation To Performance Of Lipaphis-Erysimi (Hemiptera, Aphididae). Annals
Of Applied Biology 116, 417-423.

Winter, D, Vinegar, B, Nahal, H, Ammar, R, Wilson, GV and Provart, NJ. 2007. An
Electronic Fluorescent Pictograph Browser for Exploring and Analyzing Large-Scale

Biological Data Sets. PLoS ONE 2, e718.

Zrenner, R, Stitt, M, Sonnewald, U and Boldt, R. 2006. Pyrimidine and purine
biosynthesis and degradation in plants. Annual Review of Plant Biology 57, 805-836.

127

Chapter 5
Comparative proteomics of chloroplasts envelopes from C3 and C4 plants reveals
specific adaptations of the plastid envelope to C4 photosynthesis and candidate

proteins required for maintaining C4 metabolite fluxes

This chapter has been previously published in Plant Physiology: Bréiutigam A, Hofmann-
Benning S and Weber APM (2008) Comparative Proteomics of Chloroplast Envelopes
from C3 and C4 Plants Reveals Specific Adaptations of the Plastid Envelope to C4
Photosynthesis and Candidate Proteins Required for Maintaining C4 Metabolite Fluxes
Plant Physiol 148, 568-579

I would like to acknowledge Dr. Susanne Hoffmann-Benning for assistance in
proteomics.

128

Abstract

C4 plants have up to ten-fold higher apparent C02 assimilation rates than the
most productive C3 plants. This requires higher fluxes of metabolic intermediates across
the chloroplast envelope membranes of C4 plants in comparison to that of C3 plants. In
particular, the fluxes of metabolites involved in the biochemical inorganic carbon pump
of C4 plants, such as malate, pyruvate, oxaloacetate, and phosphoenolpyruvate must be
considerably higher in C4 plants because they exceed the apparent rate of photosynthetic
C02 assimilation whereas they represent relatively minor fluxes in C3 plants. While the
enzymatic steps involved in the C4 biochemical inorganic carbon pump have been
studied in much detail, little is known about the metabolite transporters in the envelope
membranes of C4 chloroplasts. In this study, we have used comparative proteomics of
chloroplast envelope membranes fi'om the C3 plant Pisum sativum and mesophyll cell
chloroplast envelopes from the C4 plant Zea mays to analyze the adaptation of the
mesophyll cell chloroplast envelope proteome to the requirements of C4 photosynthesis.
We show that C3 and C4-type chloroplasts have qualitatively similar but quantitatively
very different chloroplast envelope membrane proteomes. In particular, translocators
involved in the transport of triose phosphate and phosphoenolpyruvate as well as two
outer envelope porins are much more abundant in C4 plants. Several putative transport
proteins have been identified that are highly abundant in C4 plants, but relatively minor
in C3 envelopes. These represent prime candidates for the transport of C4 photosynthetic

intermediates, such as pyruvate, oxaloacetate, and malate.

129

Introduction

C4 photosynthesis allows fast biomass accumulation with high nitrogen and water
use efficiency (Leegood and Edwards, 1996; Sage, 2004) and is a desired trait to increase
productivity of crop plants (Matsuoka et al., 1998). To facilitate C4 photosynthesis, in
maize, a C4 plant of the NADP-malic enzyme type, the primary fixation and the
reduction of carbon are spatially separated between two different cell types. Primary
carbon fixation occurs in the mesophyll cells (Hatch, 1987). The mesophyll surrounds the
bundle sheath cells (BSC) where C02 is enriched around Rubisco and the reduction of
carbon takes place. The chloroplasts of mesophyll and bundle sheath tissues are adapted
to their respective roles (Slack et al., 1969; Edwards et al., 2001; Majeran et al., 2005). In
addition to carbon fixation and reduction several other pathways, such as nitrogen
reduction and assimilation, are partitioned between mesophyll and bundle sheath
chloroplasts (Renne, 2003; Majeran et al., 2005) and the adaptation of the soluble
chloroplast proteome to C4 photosynthesis has been studied in considerable detail
(Majeran et al., 2005). In maize, initial carbon assimilation in the mesophyll cell
cytoplasm is accomplished by PEP carboxylase (PEPC), yielding oxaloacetate (OAA).
OAA is then imported into the chloroplasts where it is reduced to malate, and
subsequently exported to the cytosol again. After diffusion into BSC, malate is
decarboxylated in the chloroplasts, yielding CO2, NADPH, and pyruvate. While C02
and NADPH enter the Calvin cycle in BSC, pyruvate is returned to the mesophyll, where
it is imported into the chloroplasts and converted to PEP by phosphoenolpyruvate
phosphate dikinase (PPDK), thus regenerating the primary C02 acceptor, which is

exported to the cytosol to enter a new round of C02 assimilation (Figure 5-1). In maize,
130

carbon fixation is optimized beyond simply concentrating C02 in the vicinity of Rubisco.
The bundle sheath chloroplasts have limited photosystem II activity (Meierhoff and
Westhoff, 1993) and produce less 02, which further reduces the oxygenation reaction of
Rubisco. However, absence of PSII activity prevents operation of linear electron
transport, limiting the production of reduction equivalents in the bundle sheath. Since
C02 assimilation in the Calvin cycle requires NADPH, this necessitates shuttling of
reduction equivalents between mesophyll and bundle sheath by a 3-phosphoglycerate (3-
PGA)/triosephosphate (TrioseP) shuttle (Figure 5-1). Despite the detailed knowledge
about the soluble proteins involved in and necessary for C4 photosynthesis, the
adaptation of integral and associated membrane proteins remains largely unknown. In this
work, we focus on analyzing the quantitative and qualitative differences between

chloroplast envelope membranes of C3 and C4 plants.

 

 

F ca chloropla§\ f c4 PCA chloroplast
pyruvate Ci) tar-:9

... Pl OAgtzmalate

 

 

 

 

Triose? H Pl PEP/Q OAA V
pyruvate maiate

3'32 ‘3‘ 1602

a ‘6'
from PCR tissue to PCR tissue

 

Figure 5-1: Schematic representation of central carbon metabolism and associated
transport processes in C3 chloroplasts and C4 PCA type chloroplasts: in C3 chloroplasts, for
three carbons fixed, at most one transport process is required; in C4 PCA type chloroplasts, for
three carbons fixed, at least twelve transport processes are required, for abbreviations see text

131

The plastids of green plants are separated from the cytosol by two membranes.
Metabolite transport across the outer envelope is controlled by substrate-specific pore-
forming proteins (Pohlmeyer et al., 1997; Pohlmeyer et al., 1998; Bolter et al., 1999;
Goetze et al., 2006). Solute transport across the inner envelope membrane is catalyzed by
a large range of specific metabolite transporters (Weber, 2004; Weber et al., 2005; Weber
and Fischer, 2007), some of which are capable of transporting metabolites against a
concentration gradient. The spatial separation of initial carbon fixation and subsequent
reduction in C4 plants requires a very high metabolite flow across the chloroplast
envelope of both mesophyll and bundle sheath chloroplasts that exceeds the apparent rate
of carbon assimilation (Laisk and Edwards, 2000). Pea fixes about 17 micromole carbon
per square meter leaf area and second (Grodzinski et al., 1998) and requires at most one
transport process for three carbons fixed. Maize fixes about 27 micromoles carbon per
square meter and second (Grodzinski et al., 1998) and requires at least four transport
processes (Figure 5-1) for each carbon fixed. Consequently, the total metabolite transport
rate across the chloroplast envelopes in C4 plants exceeds the one in C3 plants by a factor
of at least eighteen. High velocity transport of all four metabolites involved in core C4
photosynthesis across the mesophyll chloroplast envelope has been demonstrated using
isolated chloroplasts (Huber and Edwards, 1977; Hatch et al., 1984; Flugge et al., 1985;
Aoki et al., 1992). While most transport proteins involved in core C4 photosynthesis in
mesophyll chloroplasts have not yet been unequivocally identified at the molecular level,
good candidates exist for PEP export, triose phosphate shuttling, and for
oxaloacetate/malate transport. The molecular nature of the pyruvate transporter, however,

is unknown. Likewise, it is unknown whether pyruvate transport across the mesophyll

132

and the bundle sheath chloroplast envelope is mediated by the same transport protein or
different transport proteins. Adaptations of additional membrane proteins as a
consequence of the spatial separation of photosynthesis, similar to what has been
demonstrated for soluble proteins, are unknown. Since increasing the capacity for
metabolite transport across the chloroplast envelope membrane is likely a key adaptation
to C4 photosynthesis (Edwards et al., 2001), engineering efforts for introducing C4
photosynthesis in a C3 crop plant will likely critically depend on engineering not only the
C4 pathway but also metabolite flux. 3

In this work, the protein complements of envelopes membranes of C3 chloroplasts
and C4 mesophyll chloroplasts are analyzed qualitatively and semi-quantitatively. We
hypothesized that analyzing chloroplasts with different modes of photosynthesis, such as
the C3 and C4-types of carbon dioxide assimilation, will reveal the adaptations of the
chloroplast envelope proteome to increased metabolite flow. Unfortunately, routine
methods are not available to compare membrane proteins of different species
quantitatively or even semi-quantitatively. Membrane proteins are not amenable to two
dimensional gel electrophoresis since extremely hydrophobic proteins, such as metabolite
transporters, do not focus in the first dimension (Choe et al., 2005) and quantitative
comparison relying on identical peptides such as affinity tagging are also not applicable
since there is considerable evolutionary distance between the C3 model pea and the C4
model maize. A direct quantification method, the total spectral count of proteins (the
number of mass spectra which map to one protein) has been used to compare and even
quantify proteins on a large scale (Liu et al., 2004; Lu et al., 2007) (Majeran et al., 2008).

This method has recently also been applied to yeast membrane proteins, with. results

133

comparable to SILAC (Zybailov et al., 2005). We applied this strategy to compare the
relative abundance of proteins in the chloroplast envelopes of C3 and C4 plants. We
demonstrate that the massive metabolite fluxes across the chloroplast envelope required
for maintaining the high photosynthetic rates of C4 plants are associated with significant
increases of the relative abundance of several metabolite transporters, thus pinpointing

apparent bottlenecks in metabolite flux across the chloroplast envelope membrane.
Material and Methods

Preparation of chloroplast envelope protein samples

Chloroplast envelope membranes were isolated from pea (Pisum sativum var.
Little Marvel) plants as described previously (Douce and Joyard, 1979; Keegstra and
Yousif, 1986) and from maize (Zea mays) plants grown on field sites. Briefly, fully
expanded maize leaves were harvested, stored on ice, and cut into small pieces using
razor blades. The leaves were homogenized in a Waring Blender and the resulting slurry
was filtered through several layers of miracloth to remove the bundle sheath strands.
Chloroplasts and chloroplast envelopes were isolated as described in (Douce and Joyard,
1979; Keegstra and Yousif, 1986). Envelope membranes were diluted in ten volumes of
ice cold 1:1 (v/v) chloroform/methanol and stored on ice for twenty minutes. Insoluble
proteins were sedimented by centrifugation at 20,000 g for 20 minutes (“pellet fraction”)
and both the protein pellet and the soluble fraction were dried and washed with hexane to
remove residual membrane lipids. Envelope membrane samples and fractionated samples
were mixed with SDS—PAGE loading buffer, incubated for 20 minutes on a reaction tube

shaker at 15° C, and subsequently separated by 12.5% SDS-PAGE.

134

Protein identification

After staining with Coomassie Brilliant Blue, each gel lane was cut into ten
equally-sized slices. Proteins contained in the gel slices were subjected to tryptic
cleavage as described by Shevchenko (Shevchenko et al., 1996). Peptides were extracted
and loaded onto a Waters Symmetry C18 peptide trap (5 pm, 180 um x 20mm) at a flow
rate of 4 uL/min in 2% Acetonitrile/0.1%Formic Acid for 5 minutes, using a Waters
nanoAcquity Sample Manager. Using a Waters nanoAcquity UPLC system, the peptides
were separated on a Waters BEH C18 nanoAcquity column (1.7 pm, 100 um x 100mm)
over 90 minutes and fed into a ThermoElectron LTQ-FTICR mass spectrometer with a
flow rate of 300 nL/min (Buffer A = 99.9% Water/0.1% Formic Acid, Buffer B = 99.9%
Acetonitrile/0.1% Formic Acid: gradient of 5% B to 40% B from 0 to 63 minutes, 40% B
to 90% B from 63 to 71 minutes and 5% B from 71 to 90 minutes). Survey scans were
taken at a resolution of 50,000 and the top ten ions were dissociated by automated low
energy collision. The BioWorks Browser version v3.2 was used for converting the
resulting MS/MS spectra to a peak list.
All mass spectra libraries were compared to sequence databases from Pisum sativum
(Brautigam et al., 2008) and Zea mays
(ftp://occams.dfci.harvard.edu/pub/bio/tgi/data/Zea_mays), respectively, using the
Mascot search algorithm, v 2.2 (\xl'wwmatrixscicncecom). Carbamidomethyl cysteine
was set as a fixed peptide modification and oxidation of methionine was permitted. Up to
two missed tryptic sites were allowed. The peptide tolerance was set to +/-10ppm and the

MS/MS tolerance to 0.8kDa.

135

Protein annotation and bioinformatics

Mascot Results were analyzed with using an implementation of the peptide and
protein prophet algorithms (Keller et al., 2002) (Scaffold®) with parameters set to 99%
confidence for protein identification, requiring at least two unique peptides for each
protein and 95% confidence for all peptides counted. Where Scaffold reported multiple
proteins identified for the same peptides, each match was manually inspected and low
scoring matches were discarded. The results were then exported to MS Excel for further
analysis. Cross-identifications in previous chloroplast proteomics projects was
determined using BlastX against the plprot database (Kleffmann et al., 2006) and the
corresponding Arabidopsis proteins were identified by BlastX (Altschul et al., 1997) in
TAIR. Functional annotation and classification presented in this manuscript was deduced
from information in TAIR (Swarbreck et al., 2008), ARAMEMNON (Schwacke et al.,
2003), membranetransportorg (Ren et al., 2007), and manual curation of the pertinent
literature. The predicted location and the number of transmembrane helices were
retrieved from ARAMEMNON. Tentative consensus sequences that identified the same
Arabidopsis protein were aligned. If the sequences were identical, one of the
identifications was discarded. If the sequences overlapped only partially, the annotations
were unified and the number of peptides summed up to generate a list of non-redundant
identifications (Tables S-Sl and 5-S2). Before semi-quantitative analysis the spectral
counts in each fraction were corrected for loading. The original data can be downloaded

from PRIDE (http://wwwebi.ac.uk/pride/) (Martens et al., 2005).

136

Semi-quantitative analysis of protein abundance

The semi-quantitative analysis of protein abundance was based on the spectral
count, i.e., the number of mass spectra mapping to a given protein in a single experiment.
In the first experiment for each envelope preparation all proteins in the sample were
separated by SDS-PAGE and identified by LC-ESI-MS/MS without prior fractionation
(“whole envelopes”). In a second experiment, the proteins were first fractionated into a
chloroform/methanol soluble and an insoluble fraction. Proteins from both fractions were
then separated by SDS-PAGE and subsequently identified by LC-ESI-MS/MS. The
spectral counts for each protein in both fractions were summed up to yield the “sum”
fraction. For all four experiments the spectral count for each protein was normalized to
the total number of spectra within the experiment (“percentage of the total spectral
count”) (Table S-Sl and 5-82). The robustness of the semi-quantitative analysis was
tested by introducing a number of disturbances into the experiment: omitting all proteins
with a spectral count lower than 10 spectra identified, and by including and excluding
putative extraplastidial contaminations. The results were robust.
Accession numbers: All proteomics data reported in this manuscript have been

submitted to the PRIDE data repository (htgx/Avxmnebi.ac.uk/_pride/) (Martens et al.,

 

2005) and can be downloaded from PRIDE using the following PRIDE experiment
accession numbers:

Maize C4 PCA type chloroplast envelopes - pellet fraction' (PRIDE Experiment Accession #
3370)

Maize C4 PCA type chloroplast envelopes - soluble fraction' (PRIDE Experiment Accession #
3371)

Maize C4 PCA type chloroplast envelopes - whole' (PRIDE Experiment Accession # 3372)

Pea C3 type chloroplast envelopes - pellet fraction' (PRIDE Experiment Accession # 3376)

Pea C3 type chloroplast envelopes - soluble fraction' (PRIDE Experiment Accession # 3377)

Pea C3 type chloroplast envelopes - whole' (PRIDE Experiment Accession # 3378)

137

Results

Envelope proteome coverage and purity

The pea chloroplast envelope proteome was chosen to represent the envelope
proteome of a C3 chloroplast. The proteins of the protein import complex found in our
study were compared to those identified in earlier efforts (Froehlich et al., 2001; Ferro et
al., 2003; Froehlich et al., 2003). We were able to identify all import complex
components found by Froehlich et al. (2003) with the exception of Toc33 and we
identified three additional import complex components that were not previously found. In
comparison to Ferro et al. (2003) we also identified two additional components while
Toc33 was again missing. With regard to a major metabolite transport protein family, the
phosphate translocators (Knappe et al., 2003), we identified all members predicted to be
present in the inner envelope of C3 chloroplasts (Eicks et al., 2002; Flugge et al., 2003).
Based on this data and further analysis (not shown), we conclude that the proteome of pea
chloroplast envelope has similar qualitative composition as the proteomes from other C3
plants analyzed previously.

The maize chloroplast isolation protocol applied in this study was optimized for
isolation of maize mesophyll chloroplasts and thus C4 mesophyll chloroplast envelope
membranes. Based on the virtual absence of Rubisco and complete absence of malic
enzyme, two markers for bundle sheath chloroplasts, and the relative abundance of
mesophyll marker enzymes, such as PPDK and PEPC, the maize chloroplast envelope
samples indeed represents a highly mesophyll enriched preparation (Table 5-Sl).

For each of the envelope proteome samples, the relative spectral abundance likely

resulting from extraplastidial sources such as mitochondria, endomembrane system,
138

cytosol, and nucleus were determined. The level of contamination based on this measure
was low; for the samples from maize it was below 2.2% and for the samples from pea
below 5.2%. In maize, no mitochondrial contamination was detected and extraplastidial
proteins were mostly residents of the cytosol and the endomembrane system. In pea, the
main contaminant was mitochondrial proteins. The complete list of extraplastidial
proteins can be extracted from Tables S-Sl and 5-82. Relative abundance comparisons
were performed with and without removing the contaminations from the samples and the
results were robust and therefore independent of the level and source of the
contamination. We concluded that the samples are suitable for comparing a C3 type with

a C4 mesophyll chloroplast envelope.

139

The envelope proteomes of C3 and C4 chloroplasts are qualitatively similar

 

    

 

 

 

 

 

 

 

 

 

 

 

A 100
E — C4 mesophyll envelope
:3, 8O , [:21 C3 envelope
:9
w
.E
.9. 60 ~
9
O.
E
3 4o -
d)
m .
é ..
8 20 a it
'03
n. ..
30“)“ 3650“ c‘o‘I‘s e\\ce$ Sveexs dad o\ub\e
@363 :ge\\ee‘§eaefia “as“ embtage “”03 e\'\9°\d ”(323m ‘5
the” e‘“ n at
\0 (ed
“Ema
'3 100
'O
O)
E
g 80 .
39
(I)
.E
.9 60 -
9
O.
E
'5 40 ~
0)
O)
‘3
8 204
'n‘:
n.
0 _ t”- at” \J ed e
\35 “d“ d“ awe“ 60““ “0‘1
o‘t‘\°‘°9\ \00“0 9010 o“! 99
s
targeted to

Figure 5-2: The envelope proteomes are similar when analyzed qualitatively. The percentage
of proteins within one proteome is plotted. They are similar with regard to their physico-chemical
properties (solubility in organic solvents and presence of membrane attachment structures) (A)
and with regard to their predicted targeting and percentage of novel proteins (B).

140

In the C4 mesophyll chloroplast envelope proteome, 231 non-redundant proteins
were identified and, in the pea chloroplast envelope proteome, 322 non-redundant
proteins were identified. Taken together, 420 unique proteins were identified of which
368 (87.6%) are traditional chloroplast residents. In both samples a similar percentage of
proteins are soluble or insoluble in chloroform/methanol, respectively, with about one
third of all proteins being soluble in organic solvents to at least some degree (Figure 5-
2A). Likewise, a similar share of proteins could be detected in both fractions (Figure 5-
2A). The fraction soluble in organic solvents contains a number of proteins with high
membrane helix content but neither the hydrophobicity index nor the number of predicted
membrane helices were strongly correlated with the solubility in organic solvent (data not
shown).

Little more than one half of the proteins in both samples contain recognizable
structures for membrane attachment (Figure 5-2A). Most of these proteins have predicted
alpha helices which can span a membrane, some have demonstrated or predicted beta
sheets, and very few are predicted to be anchored to the lipid bilayer by prenylation. The
other half of the proteins has no obvious domains for membrane attachment or insertion.

The proteins in both envelope preparations are also very similar when their
bioinforrnatically generated targeting predictions are compared. Most of the proteins
identified in both the C4 mesophyll and the C3 envelope proteome sample possess a
canonical target peptide for the protein import complex of chloroplasts (Figure 5-2B)
(Emanuelsson et al., 1999; Emanuelsson et al., 2000; Schwacke et al., 2003). Less than
ten percent are predicted being targeted to the mitochondria (Figure S-ZB). For a

surprisingly large group, no targeting signal can be identified within the N-terminus of

141

the protein sequence and a number of proteins have strong bioinformatics support for
targeting to the secretory pathway (Figure 5-2B). These proteins include well-known
residents of the chloroplast envelope such as Toc64 and T00159 as well as two outer
envelope porins, OEP21 and OEP24.

Both envelope proteome samples yielded a comparable proportion of proteins not
previously identified by proteomics, with 58 novel proteins from the C4 maize mesophyll
envelope proteome samples and 69 novel proteins from the C3 type pea envelope
proteome sample.

Proteins with similar relative abundance between samples

10

 

 

 

 

Difference in relative abundance [°/o]

 

 

 

.4 l l I T I l l
0 50 100 150 200 250 300 350
Proteins

Figure 5-3: Quantitatively the envelope proteomes differ in selected proteins. The relative
abundance in C3-type envelopes was subtracted from the relative abundance in C4 PCA-type

envelopes and the difference was plotted for each protein; dotted lines represent differences of +/-
0.5%

142

To visualize the compositional differences between C3 and C4 envelope
membrane proteomes, the differences between the percentages of the total spectral count
for each protein (spectral count percentage in pea was subtracted from the spectral count
percentage in maize) were plotted. Proteins that were identified in only a subset of the
experiments were set to zero in the remaining experiments. Plotting the difference
between the relative spectral abundance in C4 mesophyll and C3 chloroplast envelopes
revealed that the majority of proteins do not differ more than 0.5% in their relative
spectral abundance (Figure 5-3). This large group can be broken down into smaller
groups of proteins. Each group differs for the reason it belongs to the proteins which are
similar and selected examples are shown in Figure 5-4A. In a first group, both relative
spectral counts are high or intermediate, as is the case for proteins of the protein import
complex components (Figure 5-4A). As examples, part of the inner envelope pore,
Tic110, and two outer envelope components, T0034 and Toc64, are shown (Figure 5-4A).
A long acyl chain CoA synthase and a protein of unknown function also belong to this
group of proteins with high relative spectral abundance in all samples. A fourth import
complex component, Toc159, inexplicably was reduced in one of two replicate
experiments for C4 PCA type envelopes (Figure 5-4A). Proteins that showed large
variance between replicate experiments were considered unreliable and were therefore
not further considered. The putative glucose transporter pGlcT, a putative ATP-
dependent transporter, and an enzyme of chlorophyll biosynthesis are of intermediate
relative abundance in both samples. There are also proteins, which have a low absolute
spectral count in one or both samples and therefore do not generate large differences,

such as the transcription factor CIL or two proteins of unknown function (Figure 5-4A
143

and Tables 5-Sl and 5-S2). A complete list of proteins with similar relative spectral

abundance can be found in Supplementary Table 5-S3.

A

 

 

- C4 mesophyll envelope, whole
1:: C4 mesophyll envelope, sum
- C3 envelope, whole

E C3 envelope, sum

 

 

 

Percentage of the total spectral count [%]

 

 

 

      

’\ \o“ «“7. 59 cu.“ 3 $0 «59
$68“ wao C Wfdda 1 0c. gags?”

of”)
Figure 5-4: For the extremes and selected unchanged proteins out of Figure 5-3 detailed
results were plotted; selected proteins that do not change significantly in relative abundance
(A); the twelve proteins which are lowest in relative abundance in C4-PCA type envelopes
compared to C3-type envelopes (B); the twelve proteins which are highest in relative abundance
in C4-PCA type envelopes compared to C3-type envelopes(C);

144

Figure 5—4 (cont’d)

Percentage of the total spectral count [%]

C

Percentage of the total spectral count [%]

4

 

w
1

N
1

—I
J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1\€;a\\5%,g0\" Faetpomowfafwdafi cwgwasefieige 6°
“P‘ “Saw 00a

 

.3
O
l

on
1

 

 

 

w92oei’31 1? va‘ WT cc“ 032301715 @393 we" s‘T‘ 990*

145

 

 

 

 

Proteins with different relative spectral abundance in C4 mesophyll envelopes
compared to C3 type envelopes

Among the proteins with markedly decreased relative spectral abundance, only
four out of twelve contained membrane-spanning helices, whereas of the proteins with
increased abundance all but one were integral membrane proteins. Most of the proteins
that were underrepresented in C4 mesophyll envelopes could not be detected at all in
either of the replicate C4 experiments. The twelve proteins with the highest relative
decreases are plotted in Figure 5-4B. There are four proteins involved in carbon fixation
for PCR cycle, Rubisco large and small subunits, the Rubisco activase, and a carbonic
anhydrase, with Rubisco large subunit showing the highest relative decrease. In addition
there are three protein associated with the protein import complex, Tic55, the
ferredoxinzNADP reductase, and the import chaperone Hsp93/ClpC (8011 and Schleiff,
2004). There are also three enzymes, VTE3, a methyltransferase involved in vitamin E
and plastoquinone biosynthesis (Cheng et al., 2003), protochlorophyllide reductase
(Beale, 1999), and a NAD-dependent malate dehydrogenase. Finally, there are two
proteins of unknown function one of which is a putative ATP-dependent transporter.

The proteins that occupy a larger percentage of the spectral count in maize have
high amplitudes of up to 9%, whereas proteins that occupy a larger percentage in the pea
envelope have a lower amplitude of up to 3% (Figure 5-3). Most of the following proteins
which show major relative increases in maize belong to the classes of known and putative
transport proteins, except for PPDK, the enzyme required for regenerating the C02
acceptor PEP (Figure 5-4C). The known transport proteins are two phosphate

translocators, phosphoenolpyruvate phosphate translocator (PPT) (Fischer et al., 1997)
146

and triosephosphate phosphate translocator (TPT) (Flugge and Heldt, 1984), two
dicarboxylate translocators (DiTs), DCTl and DCT2/3, and the ATP/ADP translocator
(NTTl) (Neuhaus et al., 1997). There are also two outer envelope proteins (OEPs),
OEP24 and OEP37 (Pohlmeyer et al., 1998; Goetze et al., 2006),. Finally, we identified
four proteins of unknown function in this group (Mesophyll envelope protein 1-4). Mepl
is predicted to have twelve membrane spanning helices and Mep2 is predicted to have a
single membrane spanning helix. Mep3 and Mep4 are paralogues, which map to the same
Arabidopsis ortholog and both are predicted to have four membrane spanning helices. Of
the twelve proteins with the highest difference in spectral count compared to C3
envelopes, ten have differential protein accumulation patterns between mesophyll and
bundle sheath (Majeran et al., 2008). Eight accumulate to a higher level in mesophyll
membranes and two accumulate to a higher level in bundle sheath membranes (Table 5-
S4).

Discussion

Pea was chosen to represent C3 plants because it has served as a model for C3
chloroplasts for a long time and high purity chloroplast envelopes can be isolated with
relative ease. Maize was chosen to represent C4 plants since most of the biochemical
work on transport proteins has been published for maize chloroplasts compared to other
C4 models (Huber and Edwards, 1977a, 1977b; Hatch et al., 1984; Ohnishi et al., 1990;
Aoki et al., 1992).We established that the C3 envelope proteome from pea is comparable
to earlier envelope proteomes prepared from the C3 plant Arabidopsis and confirmed the
presence and absence, respectively, of marker proteins of C4 photosynthesis for the C4

mesophyll envelope sample such as PPDK, PEPC, and Rubisco. The level of
147

contamination was at most 5.2%. In the samples isolated from pea, the biggest
contributors were mitochondrial outer envelope proteins such as porins. It is well known
that, to foster metabolite exchange for photorespiration, chloroplasts of C3 plants are
closely associated with mitochondria and peroxisomes (Schumann et al., 2007), thus
explaining mitochondria being the major source of contaminating proteins. In the maize
sample mitochondrial contaminants were virtually absent. The contaminant with the
highest relative spectral count was PEPC, a cytosolic enzyme that is required for initial
C02 fixation. Considering that the isolation protocols for pea and maize are almost
identical, the marked difference in extraplastidial contaminants may result from the
altered requirements in organelle association. Mesophyll chloroplasts do not photorespire
since Rubisco is virtually absent and these chloroplasts therefore do not require a close
association of chloroplasts, mitochondria and peroxisomes. Both envelope preparations
also contain a nLunber of proteins identified in previous thylakoid proteome studies
(Tables 5-Sl and 5-82) (Peltier et al., 2000; Peltier et al., 2002; Peltier et al.,
2004).Currently it remains unknown whether these proteins are trapped en route to the
internal membrane system or whether they are a contamination introduced during the
envelope isolation (Ferro et al., 2002; Froehlich et al., 2003).Relative abundance
comparisons were performed with and without removing the contaminations from the
samples and the results were robust and therefore independent of the level and source of
the contamination.

We analyzed whether detailed qualitative comparisons were possible. Solid
judgments about the significance of presence or absence of proteins require proteomics to

be saturated to avoid false negative calls. To determine whether the proteome

148

identifications in either sample were saturated or whether a substantial number of
proteins remained unidentified, the well understood pathways of glycolipid biosynthesis
were analyzed. They provide a number of housekeeping proteins that are expected to be
identified in envelope proteomics studies if saturation was reached, such as two enzymes
necessary for sulfolipid biosynthesis and two known enzymes and a three-partite
transport protein involved in galactolipid biosynthesis (Benning et al., 2006). As in
earlier efforts, the envelope proteomes in this study only identify a subset of proteins in
each pathway indicating that saturating coverage of the chloroplast envelope proteome
remains to be achieved. It remains to be determined whether enzymes of glycolipid
biosynthesis are difficult to detect with mass spectrometry or whether they are of too low
absolute abundance. When additional replicates of the pea envelope proteome were
tested, we also observed that enzymes with a low absolute spectral count disappeared and
the enzyme catalyzing the next step appeared from replicate samples (data not shown).
This may indicate that when proteins with a low number of spectra are analyzed, small
variations during peptide separation, ionization, and detection may have determined
whether or not they are present in any given sample. As a consequence, the envelope
proteomes and not single proteins were the basis for the qualitative comparison.

We identified 231 and 322 non-redundant proteins in the C4 and C3 chloroplast
envelopes, respectively. The higher number of proteins identified in the pea sample likely
results from two reasons: (i) the total envelope sample from C4 mesophyll envelopes
yielded a lower total spectral count with fewer proteins identified (Table 5-Sl) although
the relative abundances for each protein remained similar (Figure 5-4). Many proteins

with a low absolute spectral count in the other experiments might have escaped detection.

149

(ii) The C4 mesophyll envelope sample contains some proteins with a high relative
spectral count compared to the C3 envelope sample, with up to 9% difference in relative
abundance (Figure 5-3). The peptides belonging to these proteins may have suppressed
peptides of lesser abundance during ionization or detection in the mass spectrometer. The
pre-fractionation by organic solvent extraction permitted the detection of additional
proteins that could not be detected in a whole envelope preparation as many of the
proteins yielding high relative spectral counts fractionated into the organic solvent
soluble fraction, thus removing the main source for ion suppression. Yet total coverage
did not reach the level obtained with C3 envelopes.

The analysis of the physico-chemical properties revealed that the C4 mesophyll
and the C3 envelope proteome are remarkably similar. The fractionation pattern into
soluble and insoluble in organic solvent was reproducible, as was the proportion of
integral membrane proteins (Figure 5-2A). Little more than one half of the proteins in
both samples contain recognizable structures for membrane attachment. In both envelope
proteomes this group of proteins included a number of proteins for which a close
association with the membrane has previously been demonstrated, such as the membrane
lipid synthesizing and modifying enzymes (Jarvis et al., 2000; Froehlich et al., 2001;
Sanda et al., 2001; Yu et al., 2002). It cannot be excluded that the remaining seemingly
soluble proteins also are closely associated with the chloroplast envelope, similar to what
has been demonstrated for glycolytic enzymes at the mitochondrial membranes (Graham
etaL,2007)

The proteins in both envelope preparations are also very similar when their

bioinforrnatically generated targeting predictions are compared. About half of the

150

proteins identified in both envelope proteome sample possess a chloroplast target peptide
for the protein import complex (Figure 5-2B) (Emanuelsson et al., 1999; Emanuelsson et
al., 2000; Schwacke et al., 2003), as was expected based on earlier results (Ferro et al.,
2002; Froehlich et al., 2003). Some of the proteins that are predicted possessing a
mitochondrial target peptide might be erroneously annotated as mitochondrial proteins by
the prediction algorithm; a well documented case in point is the most abundant
metabolite transport protein on the C3 envelope, the TPT, which is predicted to be
targeted to the mitochondria. Alternatively, it may be due to contamination of the
envelope preparation with true mitochondrial proteins. This, however, is unlikely at least
for the C4 mesophyll envelope sample since no bonafide mitochondrial proteins could be
identified in this preparation. For a surprisingly large group, no targeting signal can be
identified within the N-terminus of the protein sequence and a number of proteins have
strong bioinformatics support for targeting to the secretory pathway (Figure 5-2B),
including well established plastid residents. Especially for proteins of the import complex
it has been established that not all of them require the canonical import machinery. Many
of the proteins without classical chloroplast targeting peptides have been identified in
multiple independent plastid proteomics studies (Tables 5-S1 and 5-82). The proteins
identified in this and other studies might represent candidates for novel protein import
pathways, as have recently been reported for a carbonic anhydrase (Villarejo et al., 2005)
and outer envelope proteins (Bae et al., 2008).

Both envelope proteome samples yielded a comparable proportion of proteins not
previously identified in plastid proteome projects. Some of the novel identifications may

be due to the instrumentation used in our study, since the ultra high pressure HPLC

15]

coupled to the FT-ICR is capable of protein identification with very high resolution.
Some proteins may have been identified since the sample was fractionated prior to
proteome analysis and some proteins, especially from the maize envelope sample, may
have been identified since the chloroplast envelope is adapted to C4 photosynthesis and
C4 chloroplast envelopes have not yet been analyzed by proteomics.
A semi-quantitative view of the envelope proteomes

For several reasons a semi-quantitative approach was needed to understand the
differences between a C4 mesophyll and a C3 type chloroplast envelope. As pointed out
earlier, qualitative analysis is hampered by unsaturated proteome identification and hence
some uncertainty is associated with the identification of proteins with low absolute
spectral counts. The proteome sample from maize was compared to previous proteome
samples and more than 70% of the proteins identified in maize have been previously
found in the plastid proteomes from other species (Figure 5-2B), indicating a large
portion of the plastid envelope proteome is shared between different plastid species.
Based on these results and on the adaptations of soluble proteins to C4 photosynthesis,
we hypothesized that the differences between the C3 and the C4 chloroplast envelope are
quantitative rather than qualitative. Unfortunately, no quantitative tools for comparing
proteomes of different species are available. To overcome this limitation, we introduced
percentage of the total spectral count as a measure for quantitative composition of the
envelope proteome. This percentage is normalized to the total number of spectra
identified within one single experiment, similar to the normalization procedures used for
interpretation of RNA hybridization experiments. This method enables comparisons

between evolutionary distant species. It is based on the assumption that orthologous

152

proteins from different species have similar physico-chemical properties and thus behave
similarly throughout separation and identification when contained in similar samples,
such as chloroplast envelopes. Although the percentage of total spectral counts is not an
absolute measure of protein abundance, it is capable of captru'ing the relative contribution
of a protein to the total, which enables comparison of non-related samples. The
compositional differences between C3 and C4 envelope membrane proteomes were
visualized by plotting the differences between the percentages of the total spectral count
for each protein (spectral count percentage in pea was subtracted from the spectral count
percentage in maize). We chose to compare the difference in relative abundance over the
fold-change between the samples. Fold changes are likely a good measure if the proteins
to be compared have high absolute spectral counts, which would allow a wide range of
comparable values. In contrast, the envelope samples mainly consist of proteins of up to
ten absolute spectral counts each (Tables 5-S1 and 5-S2) similar to results reported earlier
(Brautigam et al., 2008). A comparison based on fold-changes would lead to many
proteins of low absolute spectral count to be erroneously identified as differentially
expressed between C3 and C4 mesophyll envelopes and would fail to identify the protein
with the second highest difference in relative spectral count, the TPT (data not shown).
We thus restricted analysis to the proteins with the highest relative change in expression,
which yielded comparable results in both experiments in each sample.
Marker Enzymes

A number of the proteins that were reduced or absent in the C4 sample are
associated with functions that are expected to be absent from C4 mesophyll chloroplasts,

such as the Rubisco large and small subunits, Rubisco activase, and carbonic anhydrase

153

for photosynthetic carbon reduction (Figure 5-4B). Since C4 mesophyll tissue has
strongly reduced or absent Rubisco activity, the enzyme itself and its activase are also
reduced. In mesophyll cells, a carbonic anhydrase, which quickly equilibrates C02 and
hydrogen carbonate, is needed in the cytosol for PEPC rather than in the chloroplast. The
only soluble protein, which is massively increased in the C4 mesophyll envelope
samples, is PPDK (Figure 5-4C). The detection of this soluble enzyme, which occupies a
large percentage of the spectral count within the C4 mesophyll envelope proteome
sample, may result from its high abundance resulting from its involvement in C4
photosynthesis and/or a close association with the membranes. Likely, it is absent from
the pea sample because in contrast to C4 plants it represents a minor plastidic and
cytosolic protein (Parsley and Hibberd, 2006) in C3 plants.
Proteins of the Protein Import Complex

At least two of the proteins which form the protein import complex seem to be
housekeeping proteins, Tic110 and Toc75, which have a high relative spectral abundance
in both samples. They form the pore in the inner and the outer envelope (S011 and
Schleiff, 2004). The import receptor Toc159 was excluded from analysis since its relative
abundance varied considerably between the biological replicates conducted on the C3
envelope. Two protein import complex components, Tic55 and ClpC/Hsp93, the import
chaperone or protease subunit, were found among the proteins with a lower relative
spectra abundance in C4 mesophyll envelopes. The remaining proteins, which are
believed to be involved in redox regulation of protein import, Tic32 and Tic62 (Kuchler
et al., 2002), were identified in the C3 envelope sample and could not be identified from

the C4 mesophyll envelope. Taken together with the results of import complex
154

components, the two proteins involved in reduction equivalent synthesis and balancing,
the FNR (ferredoxin:NADP+ reductase) and a NAD-dependent malate dehydrogenase
may indicate a different mode of redox-dependent import in C3 as suggested by (Kuchler
et al., 2002) compared to C4 mesophyll envelopes. This difference may be explained by
the spatial separation of reduction equivalent production between mesophyll and bundle
sheath chloroplasts, which may result in a change of redox status regulation.

Transport proteins

 

7 growth stage 65 growth stage 12

tissue root leaf male flower female flower green leaf etiolated leaf

number

ofcycles 22 25 30 22 25 30 22 25 30 22 25 30 22 25 30 22 25 30
actin . ...- - a - i ‘
PPT_I Err. . a I! M -1195-
PPT_h ”“‘ £3- - £3 :21? E a, -
Mepl - a. a —* -
Mep3 C. 3 7b.; 1 . ii

Figure S-Sl: For selected proteins, mRNA accumulation patterns in different tissues were
analyzed

At the beginning of the light period tissue from roots, leaves, male and female flowers (dehusked)
was harvested from maize stage 65. Total RNA was prepared and 0.5ug total RNA was reverse
transcribed with Superscript II] (lnvitrogen) according to the manufacturer’s instructions. The
cDNA for each gene was amplified with GoTaq (Promega) according to the manufacturer’s
instructions. Primers were chosen to span 300bp (actin fwdzaagtacccgattgagcatgg
rev:acctgaccatcaggcatctc, PPT_low fwdzCTTCCAGTCAAGGAATGTGCT
revaTTGCCCACTGAATGTGTGA, PPT_high fwdeGCCTCTGGTACCTGTTCAA
rev:CAGAAAAGAAAGGCTCCATAGC, Mepl fwdzTCGGCATGTTCTGTGTCTTC
rev:GATGGTGCAAGAAGCTCTGTC and Mep3 fwdzTGCAGAGGATTGGAGCTTTI‘
rev:AAGCACATTGCACTCAGCAG) with the resulting PCR product spanning an intron.
Amplifications were carried out with 22, 25 and 30 cycles respectively and an extension time of
30 seconds. PPT_l (TC327380) is highly expressed in roots as demonstrated by Fischer et al.
(1997) and female flowers and expressed more intensively in etiolated tissue compared to green
leaves. In contrast PPT_h (TC349162) is mainly expressed in leaves as are the candidate genes
Mepl (TC319539) and Mep3 (TC343640) and their expression is very low in etiolated tissue.

155

Phosphate translocators The transport protein TPT is one of the proteins
with the highest relative spectral abundance in both envelope samples but it is two-fold
more abundant in C4 mesophyll than in C3 type envelopes (Figure 5-4C). TPT is the
most abundant envelope transport protein in C3 chloroplast envelopes because it carries
the major flux of carbon out of the C3 chloroplast during the day. In C4 mesophyll
chloroplasts, the carbon fixation by Rubisco occurs in the bundle sheath plastids and
carbon export can therefore not be the reason for the high relative abundance of TPT.
However, since the bundle sheath chloroplasts are deficient in reduction equivalents due
to limited photosystem II activity (Meierhoff and Westhoff, 1993), the reduction of 3-
phosphoglycerate to triosephosphate occurs in mesophyll chloroplasts (Majeran et al.,
2005). Since one exchange of the reduced for the oxidized form is necessary for each
carbon fixed by Rubisco, the flux through the C4 TPT is at least threefold as compared to
the C3 TPT, where only one exchange is required for three fixed carbon units for export
(Figure 5-1). Compared to the C3 TPT, the C4 TPT exchanges 3-PGA rather than
phosphate for triosephosphate (Figure 5-1) and may thus be specifically adapted to its
new role. Interestingly, this protein is reported to be more abundant in C4 mesophyll
chloroplast membranes compared to bundle sheath chloroplast membranes (Majeran et
al., 2008) although TPT has to also export carbon from bundle sheath chloroplasts. In C4
PCA-type chloroplasts, phosphoenolpyruvate (PEP) has to be exported from the
chloroplast with a rate slightly exceeding the rate of carbon fixation (Laisk and Edwards,
2000). In C3 chloroplasts, PEP transport is a minor flux and the PEP phosphate

translocator (PPT) was initially identified in maize endosperm and characterized from

156

cauliflower buds (Fischer et al., 1997). This PPT identified in maize is highly expressed
in roots and the female flower and not higher expressed in green leaves compared to
etiolated tissue (Figure 5-Sl). Orthologues of this PPT are expressed in leaf tissue of the
C3 plant Arabidopsis thaliana (Knappe et al., 2003; Vol] et al., 2003), albeit at low
levels. The maize mesophyll chloroplast envelope samples contain an isoform of PPT
that is among the three most abundant proteins in this sample (Tables 5-S1 and Figure 5-
4C), while in pea PPT belongs to the low-abundance group of proteins. In contrast to the
PPT identified earlier this PPT is expressed highly in leaves, barely detectable in roots
and higher expressed in green than in etiolated leaves (Figure 5-Sl).The massive flux of
PEP required for CO2 fixation is mediated by a specific PPT in the C4 leaf and it can
only be maintained by increasing the amount of PPT in the envelopes, as compared to the
envelope of the C3 species pea. This PPT protein is reported to be mesophyll specific
(Majeran et al., 2008) in accordance with a role in C4 photosynthesis.

The demands for two of the four high volume fluxes necessary for C4
photosynthesis (triosephosphate vs. 3-phosphoglycerate and PEP vs. Pi) are thus
accommodated by increased amounts of the respective transport proteins and hence
increased Vmax. The pentose phosphate translocator XPT could only be identified in the
C3 envelope sample and a GPT was not detected in either experiment.

Dicarboxylate translocators The envelope proteome of C4 mesophyll
chloroplasts contains a higher percentage of proteins from the dicarboxylate translocator
(DiT) family (Weber et al., 1995). Proteins of both the glutamate/malate exchanger type,
DiT2 (Taniguchi et al., 2002; Renne et al., 2003), called DCTl and DCT2/3 in maize

(Taniguchi et al., 2004), and of the 2-oxoglutarate/malate exchanger family, DiTl
157

(Weber et al., 1995), called OMT in maize (Taniguchi et al., 2004), are enriched. These
transport proteins connect cytosolic and plastidic nitrogen metabolism through a two-
translocator mechanism in C3 plants (Woo et al., 1987; Weber and F liigge, 2002; Renne
et al., 2003) with DiTl and DiT2 also playing a major role in photorespiration (Taniguchi
et al., 2002; Renne et al., 2003; Schneidereit et al., 2006). Their function in C4
chloroplasts, such as those of maize mesophyll cells, is less well understood. There is
controversial evidence with respect to their mRNA accumulation patterns in mesophyll
cells (Renne et al., 2003; Taniguchi et al., 2004; Sawers et al., 2007). The protein
accumulation pattern indicates higher expression of OMT and DCTl in mesophyll
chloroplast membranes and higher expression of DCT2/3 in bundle sheath chloroplast
membranes (Majeran et al., 2008) (Table 5-S4) The DiT family members have been
proposed to play a role in central nitrogen metabolism (Renne et al., 2003) and, for OMT
of the DiTl family, to be the oxaloacetate/malate shuttle that is needed for core C4
photosynthesis (Figure 5-1) (Taniguchi et al., 2004). Currently, the in viva function of
DiTs in C4 plants remains unclear, although their higher abundance in C4 compared to
C3 may suggest C4 photosynthesis causes higher fluxes of their cargo metabolites.
Especially the role of the additional DiT2 family member DCT2/3 present in the bundle
sheath of maize has not been elucidated as the connection of nitrogen metabolism only
requires two translocators and the function as an oxaloacetate/malate shuttle has only
been proposed for OMT of the DiTl family. In both samples, all members of the
respective DiT families were identified.

Outer envelope porins The higher metabolite flux across the inner

envelope of C4 chloroplast necessitates a comparably high flow through the outer

158

envelope. Two outer envelope porins, OEP24 and OEP37, both occupy a larger
percentage of the spectral count in C4 mesophyll envelopes. In vitro, OEP24 transports
triosephosphates and dicarboxylates and is thus perfectly suited to accommodate the
metabolite fluxes needed for core C4 photosynthesis (Pohlmeyer et al., 1998). OEP37 has
been shown to transport inorganic cations in vitro but in vivo substrates have not yet been
established since the corresponding knockout mutant in the C3 plant Arabidopsis does
not display an apparent phenotype (Goetze et al., 2006). In contrast, the regulated outer
envelope porin, OEP21 (Bolter et al., 1999), is reduced in relative abundance (Table 5-
S3), although not among the top twelve reduced proteins. It may be reduced since the
regulation, which allows fine-tuning of the metabolite flow across the C3 envelope by the
supply of ATP, 3-phosphoglycerate, and triosephosphate, may hinder metabolite
exchange under C4 conditions. OEP16, an outer envelope protein which also forms a
channel through the membrane, likely transports amino acids (Pohlmeyer et al., 1997). In
contrast to the other outer envelope porins identified, this protein does not differ in
relative spectral abundance between C3 and C4 PCA-type chloroplasts. The adaptations
of the outer envelope proteins apparently reflect the changes in metabolite flux,
indicating that the flux across the outer envelope might be limited and regulated by its
proteins.

Other transport proteins Although metabolite transport proteins appeared to
be generally increased in C4 mesophyll chloroplast envelope over C3-type envelopes,
two proteins with unknown function, such as a putative ABC type transport protein and a
protein of unknown function predicted being anchored to the membrane were absent from

the C4 sample. In addition to the phosphate and dicarboxylate translocators, the C4
159

mesophyll envelopes contain more ATP/ADP translocator protein as compared to C3
envelopes from pea. Mesophyll chloroplasts have a high demand for ATP since the
regeneration of the primary C02 acceptor, PEP, from pyruvate requires two ATP for
each reaction. Since mesophyll chloroplasts are the source of reduction equivalents for
both mesophyll and bundle sheath chloroplasts, cyclic electron transport may be limited
in favor of linear electron transport, thus reducing the availability of ATP in the
chloroplast stroma. This limitation could be overcome by importing ATP from other
sources into PCA-type chloroplasts. Within the group of proteins that are more abundant
in maize mesophyll compared to pea C3 chloroplast envelopes are also four proteins of
unknown function. Of these proteins one has one, three have four, and one has twelve
predicted transmembrane helices. This study, complemented by data from (Majeran et al.,
2008) allows us to posit hypothesis about proteins catalyzing additional C4 metabolite
fluxes. Mepl, a protein with twelve predicted transmembrane helices, is enriched in C4
mesophyll compared to C3 envelopes and its protein accumulates evenly between
mesophyll and bundle sheath. Moreover, its mRNA accumulates mainly in green leaves
(Figure 5-Sl). This pattern of expression fits to the pyruvate transport protein which
carries a higher load in C4 plants compared to C3 plants and is needed in both mesophyll
and bundle sheath chloroplasts. Mep3 and Mep4, a pair of closely related proteins with
four predicted transmembrane helices of which one accumulated predominantly in the
bundle sheath and the other in mesophyll tissue, are also candidates for the pyruvate
transporter since they are both elevated in C4 compared to C3. In contrast, Mep2
accumulates mainly in the mesophyll and hence is a candidate for an oxaloacetate/malate

shuttle if that function is not fulfilled by a member of the DiT family. The protein
160

sequence and predicted structure of all candidate proteins is unrelated to any
characterized proteins.

Apart from being strong candidates for catalyzing metabolite fluxes across the
maize mesophyll chloroplast envelope, which are increased to transfer core C4
photosynthesis metabolites, proteins of unknown function may carry fluxes which are
increased as a byproduct of the C4 syndrome. For example, sulfur metabolism seems to
be differentially localized in C4 chloroplasts between mesophyll and bundle sheath
(Majeran et al., 2005) and therefore may require abundant transfer proteins. A
comparison of bundle sheath with C3 chloroplast envelope membranes may be necessary

to identify a candidate for the malate importer of bundle sheath chloroplasts.
Conclusions

The comparison of the C4 mesophyll and C3 chloroplast envelopes proteomes has
revealed differences beyond the expected changes in metabolite transport proteins needed
to support core C4 photosynthesis including major changes in the outer envelope. The
molecular nature of the phosphate translocators involved in C4 photosynthesis was
established and a number of candidate proteins for the additional fluxes were identified.
Similar to what is observed during the transition from C3 to CAM metabolism in
Mesembryanthemum crystallinum (Hausler et al., 2000), the abundance of chloroplast
envelope membrane transporters is adjusted to meet the high metabolic flux rates
demanded by C4 photosynthesis. To date, metabolite transport proteins have not been
included in efforts to reengineer C4 photosynthesis. This analysis points to a greater role
of the chloroplast outer and inner envelope membranes at least in mesophyll tissue for

establishing the C4 carbon concentrating mechanism than previously assumed.
161

Limitations in metabolite exchange across the chloroplast envelope may well have

hampered efforts to establish C4 photosynthesis in C3 crop plants.

Supplemental Material

Supplemental Data Table S-Sl. Proteins identified in C4 PCA-type maize chloroplast
envelope membranes. The table lists proteins identified, number of spectra mapping to
each maize accession number, annotation, classification, number of membrane spanning
domains, targeting prediction, and previous identifications in other proteomics studies.
Supplemental Data Table S-SZ. Proteins identified in C3-type pea chloroplast envelope
membranes. The table lists proteins identified, number of spectra mapping to each maize
accession number, annotation, classification, ntunber of membrane spanning domains,
targeting prediction, and previous identifications in other proteomics studies.
Supplemental Data Table 5-S3. Percentage of total spectral counts for each protein

identified in C4 PCA-type and C3-type chloroplasts of maize and pea, respectively.

Acknowledgements

We are grateful to the MSU Proteomics Core Facility for help with proteomic
identification of chloroplast envelope membrane proteins. We thank the Bioinforrnatics
Core of the MSU Research Technology Support Facility for assistance with database

generation, sequence annotation, and data mining.

162

References

Altschul SF, Madden TL, Schaffer AA, Zhang JH, Zhang Z, Miller W, Lipman DJ (1997)
Gapped BLAST and PSI-BLAST: a new generation of protein database search programs.
Nucleic Acids Res 25: 3389-3402

Aoki N, Ohnishi J, Kanai R (1992) 2 Different Mechanisms For Transport Of Pyruvate
Into Mesophyll Chloroplasts Of C-4 Plants-A Comparative-Study. Plant Cell Physiol 33:
805-809

Bae W, Lee YJ, Kim DH, Lee J, Kim S, Sohn EJ, Hwang I (2008) AKr2A-mediated
import of chloroplast outer membrane proteins is essential for chloroplast biogenesis.
Nature Cell Biol 10: 220-U101

Beale SI (1999) Enzymes of chlorophyll biosynthesis. Photosyn Res 60: 43

Benning C, Xu CC, Awai K (2006) Non-vesicular and vesicular lipid trafficking
involving plastids. Curr Opin Plant Biol 9: 241-247

Bolter B, Soll J, Hill K, Hemmler R, Wagner R (1999) A rectifying ATP-regulated solute
channel in the chloroplastic outer envelope from pea. EMBO J 18: 5505-5516

Brautigam A, Shresta RP, Whitten D, Wilkerson CG, Carr KM, Froehlich JE, Weber
APM (2008) Comparison of the use of a species-specific database generated by

pyrosequencing with databases from related species for proteome analysis of pea
chloroplast envelopes. J Biotechnol doi:10.1016/j.jbiotec.2008.02.007

Cheng ZG, Sattler S, Maeda H, Sakuragi Y, Bryant DA, DellaPenna D (2003) Highly
divergent methyltransferases catalyze a conserved reaction in tocopherol and

plastoquinone synthesis in cyanobacteria and photosynthetic eukaryotes. Plant Cell 15:
2343-23 56

Choe LH, Aggarwal K, Franck Z, Lee KH (2005) A comparison of the consistency of
proteome quantitation using two-dimensional electrophoresis and shotgun isobaric
tagging in Escherichia coli cells. Electrophoresis 26: 243 7-2449

Douce R, Joyard J (1979) Isolation and Properties of the Envelope of Spinach
Chloroplasts. In E Reid, ed, Plant Organelles. Ellis Horwood Publishers, pp 47-59

Edwards GE, Furbank RT, Hatch MD, Osmond CB (2001) What Does It Take to Be C4?
Lessons from the Evolution of C4 Photosynthesis. Plant Physiol 125: 46-49

Eicks M, Maurino V, Knappe S, Flugge UI, Fischer K (2002) The plastidic pentose
phosphate translocator represents a link between the cytosolic and the plastidic pentose
phosphate pathways in plants. Plant Physiol 128: 512-522

163

Emanuelsson O, Nielsen H, Brunak S, von Heijne G (2000) Predicting subcellular
localization of proteins based on their N-terminal amino acid sequence. J Mol Biol 300:
1004-1016

Emanuelsson O, Nielsen H, von Heijne G (1999) ChloroP, a neural network-based
method for predicting chloroplast transit peptides and their cleavage sites. Prot Sci 8:
978-984

Ferro M, Salvi D, Brugiere S, Miras S, Kowalski S, Louwagie M, Garin J, Joyard J,
Rolland N (2003) Proteomics of the chloroplast envelope membranes from Arabidopsis
thaliana. Mol Cell Prot 2: 325-345

F erro M, Salvi D, Riviere-Rolland H, Verrnat T, Seigneurin-Bemy D, Grunwald D, Garin
J, J oyard J, Rolland N (2002) Integral membrane proteins of the chloroplast envelope:
identification and subcellular localization of new transporters. Proc Natl Acad Sci U S A
99: 11487-11492.

Fischer K, Kammerer B, Gutensohn M, Arbinger B, Weber A, Hausler RE, F liigge UI
(1997) A new class of plastidic phosphate translocators: A putative link between primary

and secondary metabolism by the phosphoenolpyruvate/phosphate antiporter. Plant Cell
9: 453-462

F liigge UI, Hausler RE, Ludewig F, Fischer K (2003) Functional genomics of phosphate
antiport systems of plastids. Physiol Plant 118: 475-482

Fliigge UI, Heldt HW (1984) The Phosphate-Triose Phosphate-Phosphoglycerate
Translocator of the Chloroplast. Trends Biochem Sci 9: 530-533

F liigge UI, Stitt M, Heldt HW (1985) Light-Driven uptake of Pyruvate into mesophyll
Chloroplasts from maize. F EBS Lett 183: 335-339

Froehlich J E, Benning C, Dormann P (2001) The digalactosyldiacylglycerol (DGDG)
synthase DGDl is inserted into the outer envelope membrane of chloroplasts in a manner
independent of the general import pathway and does not depend on direct interaction with

monogalactosyldiacylglycerol synthase for DGDG biosynthesis. J Biol Chem 276:
31806-31812.

Froehlich J E, Wilkerson CG, Ray WK, McAndrew RS, Osteryoung KW, Gage DA,
Phinney BS (2003) Proteomic study of the Arabidopsis thaliana chloroplastic envelope

membrane utilizing alternatives to traditional two-dimensional electrophoresis. J Prot Res
2: 413-425

Goetze TA, Philippar K, Ilkavets 1, S011 J, Wagner R (2006) OEP37 Is a New Member of
the Chloroplast Outer Membrane Ion Channels. J Biol Chem 281: 17989-17998

164

Graham J WA, Williams TCR, Morgan M, Femie AR, Ratcliffe RG, Sweetlove LJ (2007)
Glycolytic enzymes associate dynamically with mitochondria in response to respiratory
demand and support substrate channeling. Plant Cell 19: 3723-3738

Grodzinski B, J iao JR, Leonardos ED (1998) Estimating photosynthesis and concurrent
export rates in C—3 and C-4 species at ambient and elevated CO2. Plant Physioll 17: 207-
215

Hatch MD (1987) C-4 Photosynthesis - a Unique Blend of Modified Biochemistry,
Anatomy and Ultrastructure. Biochim Biophys Acta 895: 81-106

Hatch MD, Droscher L, F liigge UI, Heldt HW (1984) A Specific Translocator for
Oxaloacetate Transport in Chloroplasts. FEBS Lett 178: 15-19

Hausler RE, Baur B, Scharte J, Teichmann T, Eicks M, Fischer KL, Fliigge UI, Schubert
S, Weber A, Fischer K (2000) Plastidic metabolite transporters and their physiological

functions in the inducible crassulacean acid metabolism plant Mesembryanthemum
crystallinum. Plant J 24: 285-296

Huber SC, Edwards GE (1977) Transport in C4 Mesophyll Chloroplasts -
Characterization of Pyruvate Carrier. Biochim Biophys Acta 462: 583-602

Huber SC, Edwards GE (1977) Transport in C4 Mesophyll Chloroplasts - Evidence for
an Exchange of Inorganic-Phosphate and Phosphoenolpyruvate. Biochim Biophys Acta
462: 603 -612

Jarvis P, Dormann P, Peto CA, Lutes J, Benning C, Chory J (2000) Galactolipid
deficiency and abnormal chloroplast development in the Arabidopsis MGD synthase 1
mutant. Proc Natl Acad Sci U S A 97: 8175-8179

Keegstra K, Yousif AE (1986) Isolation and Characterization of Chloroplast Envelope
Membranes. Meth Enzymol 118: 316-325

Keller A, Nesvizhskii AI, Kolker E, Aebersold R (2002) Empirical statistical model to
estimate the accuracy of peptide identifications made by MS/MS and database search.
Anal Chem 74: 5383-5392

Kleffmann T, Hirsch-Hoffmann M, Gruissem W, Baginsky S (2006) plprot: A
comprehensive proteome database for different plastid types. Plant Cell Physiol 47: 43 2-
436

Knappe S, Fliigge UI, Fischer K (2003) Analysis of the Plastidic phosphate translocator
Gene Family in Arabidopsis and Identification of New phosphate translocator-

165

Homologous Transporters, Classified by Their Putative Substrate-Binding Site. Plant
Physiol 131: 1178-1190

Knappe S, Lottgert T, Schneider A, Voll L, Fliigge UI, Fischer K (2003) Characterization
of two functional phosphoenolpyruvate/phosphate translocator (PPT) genes in
Arabidopsis-AtPPTl may be involved in the provision of signals for correct mesophyll
development. Plant J 36: 411-420

Kuchler M, Decker S, Hormann F, 8011 J, Heins L (2002) Protein import into chloroplasts
involves redox-regulated proteins. EMBO 21: 6136-6145

Laisk A, Edwards GE (2000) A mathematical model of C-4 photosynthesis: The
mechanism of concentrating C02 in NADP-malic enzyme type species. Photosyn Res 66:
199-224

Leegood RC, Edwards GE (1996) Photosynthesis and the Environment, Vol 5. Kluwer
Academic Publishers, Dordrecht

Liu H, Sadygov RG, Yates JR (2004) A Model for Random Sampling and Estimation of
Relative Protein Abundance in Shotgun Proteomics. Anal Chem 76: 4193-4201

Lu P, Vogel C, Wang R, Yao X, Marcotte EM (2007) Absolute protein expression
profiling estimates the relative contributions of transcriptional and translational
regulation. Nature Biotechnol 25: 117-124

Majeran W, Cai Y, Sun Q, van Wijk KJ (2005) Functional differentiation of bundle
sheath and mesophyll maize chloroplasts determined by comparative proteomics. Plant
Cell 17: 3111-3140

Majeran W, Zybailov B, Ytterberg AJ, Dunsmore J, Sun Q, van Wijk KJ (2008)
Consequences of C4 differentiation for chloroplast membrane proteomes in maize
mesophyll and bundle sheath cells. Mol Cell Prot 10.1074/mcp.M800016-MCP200:
M800016-MCP800200

Martens L, Hermjakob H, Jones P, Adarnski M, Taylor C, States D, Gevaert K,
Vandekerckhove J, preiler R (2005) PRIDE: The proteomics identifications database.
Proteomics 5: 3537-3545

Matsuoka M, Nomura M, Agarie S, Miyao-Tokutomi M, Ku MSB (1998) Evolution of
C4 photosynthetic genes and overexpression of maize C4 genes in rice. J Plant Res 111:
333-337

Meierhoff K, Westhoff P (1993) Differential Biogenesis Of Photosystem-II In Mesophyll
And Bundle-Sheath Cells Of Monocotyledonous NADP-Malic Enzyme-Type C-4 Plants
- The Nonstoichiometric Abundance Of The Subunits Of Photosystem-II In The Bundle-

166

Sheath Chloroplasts And The Translational Activity Of The Plastome-Encoded Genes.
Planta 191: 23-33

Neuhaus HE, Thom E, Mohlmann T, Steup M, Kampfenkel K (1997) Characterization of
a novel eukaryotic ATP/ADP translocator located in the plastid envelope of Arabidopsis
thaliana L. Plant J 11: 73-82

Ohnishi J, Flugge UI, Heldt HW, Kanai R (1990) Involvement of Na+ in Active Uptake
of Pyruvate in Mesophyll Chloroplasts of Some C4 Plants - Na+/Pyruvate Cotransport.
Plant Physiol 94: 950-959

Parsley K, Hibberd JM (2006) The Arabidopsis PPDK gene is transcribed from two
promoters to produce differentially expressed transcripts responsible for cytosolic and
plastidic proteins. Plant Mol Biol 62: 339-349

Peltier JB, Emanuelsson O, Kalume DE, Ytterberg J, F riso G, Rudella A, Liberles DA,
Soderberg L, Roepstorff P, von Heijne G, van Wijk KJ (2002) Central functions of the
lumenal and peripheral thylakoid proteome of Arabidopsis determined by
experimentation and genome-wide prediction. Plant Cell 14: 211-236.

Peltier JB, Friso G, Kalume DE, Roepstorff P, Nilsson F, Adamska I, van Wijk KJ (2000)
Proteomics of the chloroplast: systematic identification and targeting analysis of lumenal
and peripheral thylakoid proteins. Plant Cell 12: 319-341.

Peltier J B, Ytterberg AJ, Sun Q, van Wijk KJ (2004) New functions of the thylakoid
membrane proteome of Arabidopsis thaliana revealed by a simple, fast, and versatile
fractionation strategy. J Biol Chem 279: 49367—49383Pohlmeyer K, 8011 J, Grimm R,
Hill K, Wagner R (1998) A High-Conductance Solute Channel in the Chloroplastic Outer
Envelope from Pea. Plant Cell 10: 1207-1216

Pohlmeyer K, 8011 J, Steinkamp T, Hinnah S, Wagner R (1997) Isolation and
characterization of an amino acid-selective channel protein present in the chloroplastic
outer envelope membrane. Proc Natl Acad Sci U S A 94: 9504-9509

Ren QH, Chen KX, Paulsen IT (2007) TransportDB: a comprehensive database resource
for cytoplasmic membrane transport systems and outer membrane channels. Nucleic
Acids Res 35: D274-D279

Renne’ P, Dressen U, Hebbeker U, Hille D, Fliigge UI, Westhoff P, Weber APM (2003)
The Arabidopsis mutant dc! is deficient in the plastidic glutamate/malate translocator

DiT2. Plant J 35: 316-331

Sage RF (2004) The evolution of C-4 photosynthesis. New Phytologist 161: 341-3 70

167

Sanda S, Leustek T, Theisen MJ, Garavito RM, Benning C (2001) Recombinant
Arabidopsis SQDI converts UDP-glucose and sulfite to the sulfolipid head group
precursor UDP-sulfoquinovose in vitro. J Biol Chem 276: 3941-3946.

Sawers RJH, Liu P, Anufrikova K, Hwang JTG, Brutnell TP (2007) A multi-treatment
experimental system to examine photosynthetic differentiation in the maize leaf. BMC
Genomics 8:12

Schneidereit J, Hausler RE, F iene G, Kaiser WM, Weber APM (2006) Antisense
repression reveals a crucial role of the plastidic 2-oxoglutarate/malate translocator DiTl
at the interface between carbon and nitrogen metabolism. Plant J 45: 206-224

Schumann U, Prestele J, O'Geen H, Brueggeman R, Wanner G, Gietl C (2007)
Requirement of the C3HC4zinc RING finger of the Arabidopsis PEXlO for

photorespiration and leaf peroxisome contact with chloroplasts. Proc Natl Acad Sci U S
A 104: 1069-1074

Schwacke R, Schneider A, van der Graaff E, Fischer K, Catoni E, Desimone M, Frommer
WB, Flugge UI, Kunze R (2003) ARAMEMNON, a novel database for Arabidopsis
integral membrane proteins. Plant Physiol 131: 16-26

Shevchenko A, Wilm M, Vorrn O, Mann M (1996) Mass Spectrometric Sequencing of
Proteins from Silver-Stained Polyacrylamide Gels. Anal Chem 68: 850-858

Slack CR, Hatch MD, Goodchild J (1969) Distribution of Enzymes in Mesophyll and
Parenchyma-Sheath Chloroplasts of Maize Leaves in Relation to C4-Dicarboxylic Acid
Pathway of Photosynthesis. Biochem J 114: 489

8011 J, Schleiff E (2004) Protein import into chloroplasts. Nature Rev Mol Cell Biol 5:
198

Swarbreck D, Wilks C, Lamesch P, Berardini TZ, Garcia-Hemandez M, Foerster H, Li
D, Meyer T, Muller R, Ploetz L, Radenbaugh A, Singh S, Swing V, Tissier C, Zhang P,
Huala E (2008) The Arabidopsis Information Resource (TAIR): gene structure and
function annotation. Nucleic Acids Res 36: D1009-D1014

Taniguchi M, Taniguchi Y, Kawasaki M, Takeda S, Kato T, Sato S, Tahata S, Miyake H,
Sugiyama T (2002) Identifying and characterizing plastidic 2-oxoglutarate/malate and
dicarboxylate transporters in Arabidopsis thaliana. Plant Cell Physiol 43: 706-717

Taniguchi Y, Nagasaki J, Kawasaki M, Miyake H, Sugiyama T, Taniguchi M (2004)
Differentiation of dicarboxylate transporters in mesophyll and bundle sheath chloroplasts
of maize. Plant Cell Physiol 45: 187-200

168

Voll LM, Hausler RE, Hecker R, Weber APM, Weissenbock G, Fiene G, Waffenschmidt
S, Fliigge UI (2003) The phenotype of the Arabidopsis cue] mutant is not simply caused
by a general restriction of the shikimate pathway. Plant J 36: 301ff

Weber A, Menzlaff E, Arbinger B, Gutensohn M, Eckerskom C, Fliigge UI (1995) The
2-oxoglutarate/malate translocator of chloroplast envelope membranes: molecular
cloning of a transporter containing a 12-helix motif and expression of the functional
protein in yeast cells. Biochemistry 34: 2621-2627.

Weber APM (2004) Solute transporters as connecting elements between cytosol and
plastid stroma. Curr Opin Plant Biol 7: 247-253

Weber APM, Fischer K (2007) Making the connections - The crucial role of metabolite
transporters at the interface between chloroplast and cytosol. FEBS Lett 581: 2215-2222

Weber A, F liigge U] (2002) Interaction of cytosolic and plastidic nitrogen metabolism in
plants. J Exp Bot 53: 865-874

Weber APM, Schwacke R, Fliigge UI (2005) Solute transporters of the plastid envelope
membrane. Ann Rev Plant Biol 56: 133-164

Woo KC, Flugge UI, Heldt HW (1987) A 2-Translocator Model for the Transport of 2-
Oxoglutarate and Glutamate in Chloroplasts During Ammonia Assimilation in the Light.
Plant Physiol 84: 624-632

Yu B, Xu CC, Benning C (2002) Arabidopsis disrupted in SQD2 encoding sulfolipid
synthase is impaired in phosphate-limited growth. Proc Natl Acad Sci U S A 99: 5732-
5737

Zybailov B, Coleman MK, Florens L, Washbum MP (2005) Correlation of Relative
Abundance Ratios Derived from Peptide Ion Chromatograms and Spectrum Counting for
Quantitative Proteomic Analysis Using Stable Isotope Labeling. Anal Chem 77: 6218-
6224

169

Chapter 6
Identification and characterization of a metabolite transport protein involved in

photorespiration: the chloroplastidic glycerate glycolate carrier Mepl

Andrea Bréiutigam and Andreas Weber

170

Introduction

The photorespiratory pathway is the refixation of phosphoglycolate produced by
the oxygenation reaction of Rubisco (Ogren, 1984). The active center of Rubisco enables
not only the carboxylation of ribulose-1,5-bisphosphate to form two molecules of 3-
phosphoglycerate (3-PGA) but also an oxygenation reaction with 02 which produces one
molecule of 3-PGA and one of phosphoglycolate (Schneider et al., 1992; Kellogg and
Juliano 1997). The side reaction is caused by the stereochemistry of the active center
(Wingler et al., 2000). Therefore a recycling process for carbon lost to phosphoglycolate
has evolved which is called the photorespiratory pathway. This pathway is a highly
compartmented process and uses enzymes located in chloroplasts, peroxisomes and
mitochondria (Ogren, 1984). Phosphoglycolate is dephosphorylated in the chloroplasts
(Somerville and Ogren, 1979b) and glycolate leaves the chloroplasts and enters the
peroxisomes (Ogren, 1984). A candidate transport protein was characterized
biochemically (Young and McCarty 1993). The glycolate glycerate carrier transports
glycolate and glycerate depending on a pH gradient across the membrane (Young and
McCarty, 1993). The transport is not a strict counter-exchange but the presence of the
second substrate on the opposite side of the membrane stimulates transport (Howitz and
McCarty, 1986; Howitz and McCarty, 1991). After glycolate is imported into
peroxisomes it is oxidized to glyoxylate and transaminated to glycine with either serine
or glutamate as the donor of the amino group (Somerville and Ogren, 1980; Ogren,
1984). Glycine is moved to the mitochondria where it is decarboxylated by a multi-
subunit complex to finally form serine (Somerville and Ogren, 1981; Somerville and

Ogren, 1982). At this point photorespiration is coupled to one-carbon metabolism since
171

the decarboxylation produces methyl-THF (Collakova et al., 2008). Serine is exported
from the mitochondria and reenters the peroxisomes where it is dearninated to
hydroxypyruvate (Somerville and Ogren, 1980). Photorespiration is coupled to ammonia
fixation in the chloroplasts (Weber, 2004; Linka and Weber, 2005). The exchange of 2-
oxoglutarate and glutamate across the chloroplast envelope is necessary for efficient
photorespiration probably to balance the amount of amino donors and —acceptors as well
as free ammonia (Weber, 2004; Linka and Weber, 2005). Hydroxypyruvate is reduced to
glycerate and has to be re-imported into chloroplasts where it can be phosphorylated to 3-
PGA and reenter photosynthetic carbon reduction (Ogren, 1984). The reentry may be
catalyzed by the same protein which mediates glycolate export and is also driven by a pH
gradient (Howitz and McCarty, 1986; Howitz and McCarty, 1991; Young and McCarty,
1993). The compartmentation of the enzymes in different compartments as well as the
connection to ammonia fixation necessitates a large number of transport steps. Although
the last enzyme involved in photorespiration has been identified recently (Boldt et al.
2005) only the transport proteins for 2-oxoglutarate and glutamate which are only
peripherally associated with photorespiration have been identified at the molecular level
(Somerville and Ogren 1983; Weber et al. 1995; Renne et al. 2003). No transport protein
for core photorespiratory metabolites has been identified at the molecular level.
Co-expression of proteins involved in the same pathway is a well known concept
in bacterial operons which recently has been applied to eukaryotes (e.g. (Reumann and
Weber, 2006)). The operon structure of many genes in procaryotes which leads to
coordinated gene expression has enabled researches to identify additional genes is a

pathway once the first member of an operon was characterized (Overbeek et al., 2005).

172

Although eukaryotic genes are generally not spatially arranged into clusters according to
metabolic pathways, the expression of genes involved in one pathway is frequently
regulated coordinately (Horan et al., 2008).

In this work a transport protein for photorespiratory metabolites at the chloroplast
envelope was identified. We describe the identification of a candidate transport protein
for photorespiratory metabolites based on publicly available microarray data. The
intracellular localization of the candidate transporter was determined and both mutant
analysis as well as localization allowed us to pinpoint the molecular function. It was

attempted to demonstrate biochemical function.
Materials and Methods

Co-expression analysis

The photorespiratory genes were extracted from TAIR based on the predicted or
annotated function and confirmed through literature review whenever possible. For each
gene, publicly available microarray data was queried through the CSB.DB (Steinhauser et
al. 2004). The lists of genes were manually screened for membrane proteins by
comparing Arabidopsis genome identification codes with ARAMEMNON (Schwacke et
al. 2003). The membrane proteins were compiled into lists and retested when additional
databases became available. Finally, Pearson and Spearman correlation coefficients were
extracted from the plant expression database (Horan et al. 2008) for photorespiratory
genes and Mepl. In parallel, microarray datasets were assembled and clustered
hierarchically according to (Reumann and Weber 2006) to determine how

photorespiratory genes cluster compared to the complete genome.

173

Molecular Biology

The -1.5kb upstream gene region of Mepl (Atlg32080) was amplified and cloned
directionally into pENTR-SD-TOPO. The insert was mobilized into pMDCl63 with LR
clonase (Invitrogen) and the resulting construct was transferred to Arabidopsis by
A grobacterium tumefaciens mediated transformation with the floral dip method (Bent,
2000). Transformed plants were selected; the T3 generation was rescreened on
hygromycin and subjected to staining at suitable developmental stages. Staining was
carried out as described in (Knappe et al. 2003). Plant were fixed with 3.7%
formaldehyde in 80% ethanol, destained, and maintained in 80% ethanol.

A C-terminal GF P fusion was created by amplifying the coding region of Mepl
and cloning the product into pDONR22l. The insert was mobilized into pMDC83 and
transiently expressed in tobacco leaves according to (Waadt and Kudla 2007).

The insertion line was isolated and verified according to the recommendations
from the SALK center. Total RNA was prepared according to (Knappe et al. 2003),
cDNA synthesis and PCR amplification of targets with the S8111 Polymerase (NEB) was
carried out according to the manufacturer’s instructions with 35 amplification cycles. The
insertion mutant line was maintained at elevated CO2 concentrations for improved
growth and seed set.

Metabolite analysis

Plants were grown with a 12h light, 12 hour dark cycle at 3000ppm CO2 with
100uE light intensity for four to six weeks and watered as needed with half strength
Hoagland’s solution. For metabolite analysis, the complete rosette was harvested, shock

frozen in liquid nitrogen and ground to a fine powder before being aliquoted under liquid

174

nitrogen. For metabolite analysis, approximately 100mg of tissue (fresh weight) was
extracted with 700ul methanol and 700ul chloroform. The chloroform fraction was
discarded. For LC-MS/MS analysis, the methanolic extract was analyzed according to
(Lu et al., 2008) except that standards were also prepared in methanol. For GC-MS
analysis, the methanolic extract was dried and derivatized according to (Fiehn et al.,
2001). Derivatized metabolites were separated for 30 minutes on HP5 columns.
Metabolites were identified by co-elution with standards and by fractionation pattern and

quantified with external standard curves of complex standards.
Results

Identification of candidate transport proteins for photorespiratory metabolite flux
Co-expression of genes has been used successfully to identify genes involved in
the same pathway (Reumann and Weber, 2006). The enzymes known to be involved in
the photorespiratory pathway were used to query publicly available co-expression
databases. These genes belong to a large cluster of 100 to 150 genes depending on the
clustering method which includes genes involved in photorespiration, photosynthesis and
chloroplast function (Horan et al., 2008). The co-regulation coefficients for both the
Spearman (S) and the Pearson (P) frmctions between the photorespiratory enzymes were
annotated and a large number of genes involved in photorespiration are indeed co-

expressed (Table 6-1).

175

 

 

dd. a adv aad dad dad add dad adv dad dad dad add Ems. dadaaasa. 8.22
dde adv adv adv adv adv adv adv adv adv adv v.20 ddadda:< 82: 29820
- - . . . . . . . . . . on moamdosuo.
dd V 8 d S d a av a dv 3 d a d aa a dd d aa d Ea: 28 :< 9935592:
- - - . . . . . . . . . m cameodmcmdifioe
dd F «a d a dv a d aa a dd d 3 a ad a aa a Era ddaad 3< axed? dam
- - - - . . . . . . . . a £205
dd . aa d «a d a dv a dv a dv «a d «a d .540 dad: :< -p aaaaxoeaaad do

- - - - - dd; cad adv adv adv adv dad drone ddvaaaz<
- - - - - - dde adv adv adv «ad dad £30 dadaaadé. 5299: one
- - - - - - - 84 3d dad dad add Ego dddaaaa< 5293. one
- - - - - - - - . . . . mandodmcmboEEm
dd V dd d 8 d «a d F50 ddaaaadz adasxozaéo
- - - - .. - - - - . . . oamdodacgosEm
dd d aa d 3 d fio< ddad dag aaasxodadam
- - - - - - - - - - ddd add :80 83.33 aaaada 22820
- - - - - - - - - - - ddd Eden. daadaaai 892308 on:

co m

Eds. v20 Ea: :zzm E50 290 £90 590 £00 :3 :80 Eden. damn“ _o< aefidw

ooflEon cod—Ea odd ..oVa 5:5 20:22.8 £305 a_ E30580 5:22.50 7.3 of .23 58.3%
2: >._:o :82 3% odd—:38 3d: of .8 can 539.3383...— E 329:: dozen .5.— 353503 uaEhaoam 538.3560 "To «Sun.

176

Phosphoglycolate phosphatase is co-expressed with all genes for the pathway
except the L-subunit of GDC and glycerate kinase (Table 6-1). For the enzymes where
several isoforms of the enzymes exist, only one is strongly co-expressed with other
photorespiratory genes. Of the five isoforms encoding enzymes for glycolate oxidation
only AtGOXl is strongly correlated (AtGOX2 is indistinguishable due to the probe on
the ATHl chip) whereas the other four candidate genes show no co-expression with other
photorespiratory genes (Table 6-1 and data not shown). Likewise only GGTl but not
GGT2, only one isoform each of the subunits of GDC and only the SHM isoform is co-
expressed. Afier co-expression between the enzymes involved in photorespiration was
established, proteins of unknown fimction which have a co-regulation coefficient of at
least 0.9 in either the Pearson or the Spearman correlation were tested for the presence of
predicted membrane spanning helices. One protein of unknown function (Mepl) was
both co-expressed with genes involved in photorespiration and had twelve predicted
membrane spanning helices.

Mepl is expressed in green tissues only

After the co-expression analysis revealed AtMepl to be a candidate for a transport
protein involved in photorespiration, the expression pattern was tested experimentally. A
promoter::GUS fusion expressed stably in Columbia-0 background was stained only in
tissues and organs containing chloroplasts in all developmental stages tested. A total of
eight lines were analyzed 'with seven lines showing identical patterns and no GUS
activity visible in the eighth line. In the seedling stage, both cotyledons were stained

strongly, with the hypocotyl stained weaker and the primary root remained white.

177

 

Figure 6-1: GUS expression pattern of a P(Mepl)::GUS fusion - this image is presented in
color

In adult plants of the vegetative stage, all rosette leaves stained although the major
veins of leaves and the center of the rosette around the meristem show less GUS activity
(Figure 6-1D). As in the seedling stage, the roots remain white although small patches of
GUS activity were occasionally observed in single plants. Within the flower, GUS
activity is confined to green tissues (Figure 6-1A and B). The sepals but not the petals or
the stamen show blue staining. In siliques the GUS staining is confined to the tip and the

base of the siliques in various stages of ripening but the green silique walls showed no

178

GUS activity (Figure 6-lC). This expression pattern agrees with data extracted from the
AtGenExpress database of publicly available microarrays where the highest expression is
detected in rosette leaves and the stem with moderate expression in flowers and very low

expression in roots, pollen and siliques (Figure 6-1A and B).

The Mep1 insertion mutant shows a visible phenotype in ambient air

 

 

 

 

A
T-DNA
. ITTE
—m r l-‘i l—l:l-l:l—E:l—l:l—l:l—l:l:l—
B Mep1 actin
+RT -RT +RT -RT

 

 

 

Figure 6—2: Isolation of a Mep1 insertion mutant; Structure of the Mepl gene (Atlg32080) and
insertion site of the SALK line (A); Mepl specific transcripts are undetectable in the insertion

mutant of Mepl mepI-I (B); the insertion mutant mepl-I shows a distinct phenotype (C); this
image is presented in color

179

Figure 6-2 (cont’d)

C mep1-1 Col-0

 

To elucidate the in planta function of Mep1 an insertion line in the gene was
isolated from the SALK collection. Although the SALK insertion is localized in an intron
of the gene (Figure 6-2A), no mRNA was detected in saturated RT-PCR (Figure 6-2B).
The plants show a strong visible phenotype if grown in ambient air with at least 14 hour
light periods (Figure 6-2C). Leaves start to yellow in the regions between the veins and
finally bleach out but stay green along the veins. Young leaves are not affected. It was
observed that the phenotype is conditional both based on the CO2 content of the
surrounding air and the light intensity and light period. The Visible phenotype is
completely suppressed by supplying the plants with at least 3000ppm C02 independent
of the light conditions. In ambient air, increasing the light intensity or increasing the day
length aggravates the phenotype which is always visible at day lengths greater than 14
hours with a light intensity greater than 80 uE after at least five days of exposure. Once
the phenotype occurs, it cannot be reversed by decreasing the light intensity or increasing

the carbon concentration. Newly emerging leaves however do not display the phenotype
180

if kept under non-inducing conditions. The visible phenotype of mep] -1 can be
complemented by introducing a 3.5kb genomic DNA fragment which contains the full
length gene and its native promoter.
A biochemical phenotype is observed even under non-inducing conditions

To understand the underlying biochemical condition of the visible phenotype and
the basis of suppression in air enriched in CO2, steady state metabolite concentrations
were determined. Three time points were chosen. Since the phenotype is aggravated by
increasing light duration, all samples were taken in the evening based on the hypothesis
that metabolites accumulate with increasing day length. Initial measurements were taken
of plants grown in CO2 enriched air. Plants were shifted to ambient air at 80 uE and after
two days (no phenotype visible) and after five days (at the onset of phenotype) whole
rosettes were harvested. The steady state concentrations of amino acids were determined
from liquid methanolic extracts, organic acids and simple sugars were measured from

dried and derivatized metabolite extracts.

181

 

 

A Glycolate B Giyoxylate

 

 

 

 

 

 

 

 

 

 

 

0d 2d 5d 0d 2d 5d

Figure 6-3: Metabolite accumulation in the Mepl mutant compared to wild type on a shift
from 3000ppm C02 to ambient air; metabolites were measured for plants kept at high C02 and
after two and five day shift to ambient air; all values are measured in umol/ g fresh weight except
hydroxypyruvate which is shown as arbitrary units; glycolate (A), glyoxylate (B), glycine (C),
serine (D), hydroxypyruvate (E) and glycerate (F);

182

 

 

 

 

 

All amino acids were measured as well as the following organic acids: glycolate,
hydroxypyruvate, glycerate, organic acids of the TCA cycle except for oxaloacetate. The
five metabolites showing the biggest change are those involved in photorespiration
(Figure 6-3). Glycerate and glycolate accumulate to high levels already under high CO2
conditions (Figure 6-3A and F). Glycolate accumulates eight fold and increases to sixteen
fold when shifted to ambient air. Glycerate accumulates twenty fold initially and
increases to forty fold when shifted. Three other metabolites, glycine, serine and
hydroxypyruvate, are rather similar to wild type as long as plants are kept in high C02 air
but start to accumulate to higher levels than wild type once plants are shifted (Figure 6-
3C, D and E). Glyoxylate is not significantly different (Figure 6-3B). The highest
accumulation of metabolites is measured two days afier the shift and metabolite
concentrations start to decline at the five day time point. To gain a more dynamic picture
of the metabolite accumulation pattern and to exclude that the accumulation of
metabolites measured was the result of a pleiotropic effect, a detailed time course of
plants grown under high CO2 conditions and of plants shifted to ambient air was
measured. Both glycolate and glycerate accumulate in a light dependent pattern already
under high CO2 conditions (Figure 6-4 A and E). Glycolate levels at night are
comparable to wild type levels and rise in the course of the day to the highest point in the
evening (Figure 6-4 A). Glycerate levels remain at a high level even during the night but
rise during the day (Figure 6-4 E). As demonstrated in before, glycine, serine and
hydroxypyruvate are similar to wild type levels when plants are incubated in high C02.

Serine is the only one of the photorespiratory metabolite which displays changes in wild

183

type upon shift to ambient air; the levels approximately double. Glycine, serine and
hydroxypyruvate levels increase upon shifting the plants but hydroxypyruvate and serine
do not display daylight dependent accumulation but continue to accumulate during the
initial measurement during the night before they fall, and start accumulating again on the
second day of ambient conditions. All plants show no visible phenotype during the
experiment and only start to develop yellow leaf areas on day five of the shift to ambient

air.

184

 

 

 

 

O-INw-hO'IOiN
l

O—L—r'l

 

 

 

 

 

g!“
on

0.6 -
0.4 -

0.2 1 "

ol

0 24 48 0 24 48

Figure 6-4: Time course of metabolite accumulation in the Mep1 insertion line compared to
wild type upon shifi from 3000ppm (until 24 hours) to ambient air (30 hours till end); all values
are measured in umol/ g fresh weight except hydroxypyruvate which is shown as arbitrary units;
glycolate (A), glycine (B), serine (C), hydroxypyruvate (D) and glycerate (E)

 

 

 

 

 

 

 

 

 

 

185

Mepl is localized to the chloroplast envelope

 

Figure 6-5: Mep1 is localized in the chloroplast envelope; fluorescence of GFP (A); chlorophyll auto-
fluorescence (B) and the overlay of both (C); this image is in color

Mep1 was localized as a C-terminal GFP fusions protein. The fusion of Mep1 was
expressed transiently in tobacco leaves under a 3SS-CaMV promoter. The fluorescence of
GFP forms circles (Figure 6-5A) which surround the chlorophyll auto-fluorescence
(Figure 6-5B). The fluorescent circle of GFP is not a complete circle since several
openings remain. Prolonged overexpression leads to accumulation of fluorescent protein

in the cytosol.

Discussion

None of the transport proteins involved in transport of core photorespiratory
metabolites has been characterized at the molecular level. Due to the high level of
compartmentation, several transport processes are needed at the organellar membranes; at
the chloroplast envelope glycolate has to be exported while glycerate is imported, the
peroxisomal membrane is crossed by not only glycolate and glycerate but also glycine

and serine, and the mitochondria need to import two molecules of glycine while only one

186

serine is exported (Ogren, 1984). We report the identification of a protein with twelve
predicted transmembrane helices which is co-expressed over a wide range of conditions
with other photorespiratory genes (Table 6-1). The correlation coefficient is similar to the
ones of other photorespiratory genes except glycerate kinase which is not co-expressed at
all. All other photorespiratory genes correlate well with each other and with Mepl.
However, the photorespiratory genes are part of a larger cluster comprising many genes
involved in photosynthesis (Horan et al., 2008) (Weber APM, unpublished). Co-
expression analysis was successfully used to identify a candidate transport protein for
photorespiratory metabolites. The experimentally determined expression pattern of Mepl
agrees well with expression data gathered in different Affymetrix experiments (Figure 6-
1) (Schmid et al., 2005; Winter et al., 2007). The gene is exclusively expressed in green
tissues with the strongest expression in leaves (Figure 6-1) (Schmid et al., 2005; Winter
et al., 2007).

An insertion mutant line was isolated as a tool to analyze the metabolic
consequences if the transport protein is absent from the cell. The visible phenotype is
unlike that of known photorespiratory mutants which are frequently lethal when grown
under ambient air (Somerville and Ogren, 1979b; Somerville and Ogren, 1979a;
Somerville and Ogren, 1981). The visible phenotype of mepl -1 was compared to the one
of shm] -1 which is impaired in the mitochondrial production of serine during
photorespiration (Voll et al., 2006). ShmI-I yellows under ambient air but dies before
developing bleached leaf regions. But when grown under high light conditions, shmI -1
mimics the phenotype of mep] -1 , thus demonstrating that the visible phenotype of

photorespiratory mutants is variable with varying light conditions. Comparison of the

187

visible phenotype with the expression pattern of the gene showed that the bleached
intercoastal regions do not correspond to the expression domains of the protein (Figure 6-
1) indicating that the bleaching is a secondary effect of the mutation. After establishing
that mepI -1 is a likely photorespiratory mutant, a detailed phenotypic analysis was
initiated. To avoid pleiotropic effects caused by the yellowing and bleaching of the leaf
regions and the concomitant slower growth, plants were kept under non-inducing
conditions and the steady state metabolite content was analyzed after two and five days of
shift to ambient air. Although a photorespiratory phenotype was suspected, the metabolite
analysis was unbiased covering all amino acids, all TCA cycle organic acids except
oxaloacetate and the photorespiratory intermediates glycolate, hydroxypyruvate and
glycerate, and the sugars glucose, fructose, sucrose and maltose. The suppression of
pleiotropic phenotypes was successful. Of more than forty metabolites, major changes
were recorded only for those metabolites which are intermediates of one pathway, the
photorespiratory pathway (Figures 7-3 and 7-4). The detailed time course allowed
resolving the accumulation kinetics for all metabolites and again, the biggest changes
were recorded for photorespiratory intermediates (Figure 6-4). Notably, the biggest
change is recorded in the two metabolites which have to cross the chloroplast envelope,
the glycolate which needs to be exported from the chloroplast and the glycerate which
needs to be re-imported from the cytosol to be converted to 3-phosphoglycerate. The
analysis of transiently expressed proteinzzGFP fusions indicated that Mep1 is localized to
the plastid envelope (Figure 6-5). This finding is supported by localization programs such
as TargetP which assign a canonical target peptide for the chloroplast at the protein N-

tenninus and also by various chloroplast envelope proteome projects in which Mepl is

188

identified as a chloroplast enve10pe protein (Ferro et al., 2003; Froehlich et al., 2003;
Brautigam et al., 2008). The other intermediates involved in photorespiration only
accumulate after the flux through the photorespiratory pathway is increased by lowering
the C02 concentration which has been shown also for other photorespiratory mutants
(Boldt et al., 2005). The steady state levels of metabolites measured do not reflect the
flux through the pathway, they only indicated the pool size of a metabolite at a given time
point. In absolute measure, the largest metabolite accmnulation by far is in the glycerate
pool during the day (Figures 7-3 and 7-4). The main pathway to metabolize
photorespiratory glycerate in the leaf is the chloroplast localized glycerate kinase. The
insertion mutant in glycerate kinase also shows glycerate accumulation in ambient air
(Boldt et al., 2005). The time course of metabolite accumulation indicates that the
decreasing pool size of glycerate is accompanied by increasing pool sizes of
hydroxypyruvate and serine during the night in ambient air (Figure 6-4) which may
indicate that glycerate is metabolized by extraplastidial enzymes such as
hydroxypyruvate reductase and that the hydroxypyruvate pool is equilibrated with the
serine pool by transamination reactions. Thus the increasing pool sizes of
photorespiratory metabolites other than glycolate and glycerate may be caused by a block

in glycerate metabolism.

189

 

 

 

r 3—PGA phosphoglycolate N
ADP H20
ATP P.

K plastid glycerate glycol ate J

 

 

( glycerate glycol ate \
+

 

 

 

 

 

NAD 02
NADH "202
hydroxypyruvate glyoxylate
aminoacid aminoacid
'd moacid
axoacr
. serine glycine J
\ peroxrsome
r serine glycine
THF NAD+THF
NADH NH
THF 3
H20 glycine 5,10 methylene (:02
mitochondrion

 

Figure 6-6: Schematic representation of photorespiration.

At first glance, it is counterintuitive why both the initial intermediate leaving the
chloroplast, glycolate, and the final intermediate entering the chloroplast, glycerate,
accumulate in the map] -1 mutant. However with a pathway as compartmented as
photorespiration not only the linear pathway but also its compartmentation has to be

considered. Mepl is a chloroplast envelope localized protein which is only expressed in

190

green tissues. McCarty and coworkers described biochemically a protein which is capable
of transporting glycolate and glycerate and which does not depend on stochiometric
exchange but rather on a pH gradient across the envelope membrane (Howitz and
McCarty, 1986; Howitz and McCarty, 1991; Young and McCarty, 1993). If this protein
were mutated, both the export of glycolate as well as the import of glycerate would be
disturbed. With many transport protein mutant, mild phenotypes are reported. Unlike
enzymes, most transport proteins have comparatively broad substrate specificities but
display high affinity for their in vivo substrates. However, if a mutation causes high
metabolite accumulation, a transport protein with low affinity may now be able to
transport a substrate it did not under wild type conditions. In addition, small acids are
known to cross the envelope independently of transport proteins. Hence, glycolate
accumulating in the chloroplast due to the absence of Mepl may accumulate only to the
level at which free diffusion or another transport protein moves it out of the chloroplast.
Once glycolate is available in the cytosol, photorespiration can occur as in wild type until
glycerate is produced in peroxisomes. It can no longer be imported to the chloroplast
where the final enzyme involved in photorespiration in localized (Boldt et al., 2005).
Unlike glycolate, glycerate accumulates to higher levels in the cell which only decline to
a defined limit during the night before accumulating again once flux through
photorespiration resumes. The overnight decline in glycerate is accompanied in a rise in
its precursors, hydroxypyruvate and serine, which may reflect on the pathway which
glycerate takes during the night. The combined evidence of expression pattern, protein
localization and phenotype analysis points to Mepl as the glycolate glycerate carrier of

the chloroplast envelope. To conclusively proof its function, a biochemical

191

characterization is necessary. Four different systems to characterize the protein
biochemically were tested. The availability of mutant and wild type plants enables a
transport analysis on whole chloroplasts. Previous work in wheat and pea indicates that
chloroplasts evolve oxygen if fed with glycerate since the 3-phosphoglycerate formed by
glycerate kinase can be reduced to triosephosphate thus consuming reducing power and
resupplying linear electron transfer with NADP (Heber et al., 1974). Unfortunately, the
assay requires fine tuning of ATP, phosphate, and glycerate concentration supplied to the
chloroplasts during the import assay (Heber et al., 1974). Isolation of metabolite import
competent chloroplasts was successful as indicated by oxygen evolution of chloroplasts
fed with 3-PGA but no conditions could be established where wild type chloroplasts
evolved oxygen when fed with glycerate. It was also attempted to express the protein
heterologously in different systems for reconstitution assays in artificial liposomes. Many
transport proteins were characterized after expression in Saccharomyces cerevisiae
(Loddenkotter et al., 1993; Fischer et al., 1994; Weber et al., 1995; Kammerer et al.,
1998; Eicks et al., 2002). For Mep1 neither expression with a C-terminal nor an N-
terminal histidine tag proved successful despite optimizing the translation initiation.
Changing the 5’ end of the construct as suggested in (Klepek et al., 2005) to improve
RNA stability and/or translation initiation did not improve expression. Expression in a
bacterial system was also attempted since membrane transport proteins have also been
reconstituted successfully from bacterial protein extracts (Fierrnonte et al., 2001; Picault
et al., 2002; Palmieri et al., 2008). Both C-terminal and N-terminal fusion proteins which
did or did not carry a T7 tag for improving expression did not lead to detectable amounts

of protein. Expression conditions were varied for length of induction and growing

192

temperature without any success. Autoinduction has been reported as a possible strategy
to overcome protein toxicity in heterologous expression (Studier, 2005). Autoinduction
was tested instead of induction by IPTG without any success. Novel bacterial strains
optimized for membrane protein production (Miroux and Walker, 1996) did not lead to
protein expression either. Finally, an in vitro protein expression system was established
based on protocols from (Schwarz et al. 2007). In this system, transcription is initiated
from a T7 promoter by T7 polymerase and translation is carried out with an Escherichia
coli extract. Amino acids and nucleotides are supplied to the transcription/translation
assay by dialysis (Schwarz et a1. 2007). This system has been used successfully to
express membrane proteins which had proved recalcitrant to expression before
(summarized in Schwarz et al., 2007). Three different membrane proteins were
successfully expressed but expression of Mep1 with different tags for detection lead to no
detectable protein expression and instead frequently caused the complete assay mixture to
precipitate from solution. It has not been possible to analyze Mep1 biochemically since
all attempts to produce protein for reconstitution remain unsuccessful. Possibly, the high
hydrophobicity of Mep1, a protein with twelve transmembrane helices and virtually no
hydrophilic loops causes the protein to aggregate in non-native expression systems. Plant
cells may express chaperones to keep highly hydrophobic proteins in solution prior to

insertion into the membranes.

193

References

Bent AF (2000) Arabidopsis in planta transformation. Uses, mechanisms, and prospects
for transformation of other species Plant Physiol 124, 1540-7.

Boldt R, Edner C, Kolukisaoglu U, Hagemann M, Weckwerth W, Wienkoop S,
Morgenthal K, Bauwe H (2005) D-GLYCERATE 3-KINASE, the last unknown enzyme

in the photorespiratory cycle in Arabidopsis, belongs to a novel kinase family Plant Cell
17, 2413-2420

Brautigam A, Hofmann-Benning S, Weber APM (2008) Comparative Proteomics of
Chloroplast Envelopes from C3 and C4 Plants Reveals Specific Adaptations of the
Plastid Envelope to C4 Photosynthesis and Candidate Proteins Required for Maintaining
C4 Metabolite Fluxes Plant Physiol 148, 568-579

Collakova E, Goyer A, Naponelli V, Krassovskaya 1, Gregory J F, Hanson AD, Shachar-
Hill Y (2008) Arabidopsis lO-formyl tetrahydrofolate deformylases are essential for
photorespiration Plant Cell 20, 1818-1832

Eicks M, Maurino V, Knappe S, Flugge UI, Fischer K (2002) The plastidic pentose
phosphate translocator represents a link between the cytosolic and the plastidic pentose
phosphate pathways in plants Plant Physiol 128, 512-522

Ferro M, Salvi D, Brugiere S, Miras S, Kowalski S, Louwagie M, Garin J, Joyard J,
Rolland N (2003) Proteomics of the chloroplast envelope membranes from Arabidopsis
thaliana Mol Cell Proteomics 2, 325-345

F iehn O, Kopka J, Dorrnann P, Altmann T, Trethewey RN, Willmitzer L (2001)
Metabolite profiling for plant functional genomics Nature biotechnol 19, 173-173

Fiermonte G, Dolce V, Palmieri L, Ventura M, Runswick MJ, Palmieri F, Walker JE
(2001) Identification of the human mitochondrial oxodicarboxylate carrier - Bacterial

expression, reconstitution functional characterization, tissue distribution, and
chromosomal location J Biol Chem 276, 8225-8230

Fischer K, Arbinger B, Kammerer B, Busch C, Brink S, Wallmeier H, Sauer N,
Eckerskom C, Flugge UI (1994) Cloning and in-Vivo Expression of Functional Triose
Phosphate/Phosphate Translocators from C-3-Plants and C-4-Plants - Evidence for the
Putative Participation of Specific Amino-Acid-Residues in the Recognition of
Phosphoenolpyruvate Plant J 5, 215-226

Froehlich JE, Wilkerson CG, Ray WK, McAndrew RS, Osteryoung KW, Gage DA,
Phinney BS (2003) Proteomic study of the Arabidopsis thaliana chloroplastic envelope
membrane utilizing alternatives to traditional two-dimensional electrophoresis J
Proteome Res 2, 413-25

194

Heber U, Kirk MR, Gimmler H, Schafer G (1974) Uptake And Reduction Of Glycerate
By Isolated-Chloroplasts Planta 120, 31-46

Horan K, Jang C, Bailey-Serres J, Mittler R, Shelton C, Harper JF, Zhu JK, Cushman JC,
Gollery M, Girke T (2008) Annotating genes of known and unknown function by large-
scale coexpression analysis Plant Physiol 147, 41-57

Howitz KT, McCarty RE (1986) D-Glycerate Transport By The Pea Chloroplast
Glycolate Carrier - Studies On [l-C-14] D-Glycerate Uptake And D-Glycerate
Dependent O-2 Evolution Plant Physiol 80, 390-395

Howitz KT, McCarty RE (1991) Solubilization, Partial-Purification, And Reconstitution
Of The Glycolate Glycerate Transporter From Chloroplast Inner Envelope Membranes
Plant Physiol 96, 1060-1069

Kammerer B, Fischer K, Hilpert B, Schubert S, Gutensohn M, Weber A, Flugge UI
(1998) Molecular characterization of a carbon transporter in plastids from heterotrophic
tissues: the glucose 6-phosphate/phosphate antiporter Plant Cell 10, 105-1 7.

Kellogg EA, Juliano ND (1997) The structure and function of RuBisCO and their
implications for systematic studies Am J Bot 84, 413-428

Klepek YS, Geiger D, Stadler R, Klebl F, Landouar—Arsivaud L, Lemoine R, Hedrich R,
Sauer N (2005) Arabidopsis POLYOL TRANSPORTERS, a new member of the
monosaccharide transporter-like superfamily, mediates H+-symport of numerous
substrates, including myo-inositol, glycerol, and ribosele Plant Cell 17, 204-218

Knappe S, Lottgert T, Schneider A, Vol] L, Flugge UI, Fischer K (2003) Characterization
of two functional phosphoenolpyruvate/phosphate translocator (PPT) genes in

Arabidopsis-AtPPTl may be involved in the provision of signals for correct mesophyll
development Plant J 36, 411-420

Linka M, Weber APM (2005) Shuffling armnonia between mitochondria and plastids
during photorespiration Trends Plant Sci 10, 461-465

Loddenkotter B, Kammerer B, Fischer K, Flugge UI (1993) Expression of the Functional
Mature Chloroplast Triose Phosphate Translocator in Yeast Internal Membranes and

Purification of the Histidine-Tagged Protein by a Single Metal-Affinity Chromatography
Step Proc Natl Acad Sci U S A 90, 2155-2159

Lu Y, Savage LJ, Ajjawi I, Imre KM, Yoder DW, Benning C, DellaPenna D, Ohlrogge

J B, Osteryoung KW, Weber AP, Wilkerson CG, Last RL (2008) New connections across
pathways and cellular processes: Industrialized mutant screening reveals novel
associations between diverse phenotypes in Arabidopsis Plant Physiol 146, 1482-1500

195

Miroux B, Walker J E (1996) Over-production of proteins in Escherichia coli: Mutant

hosts that allow synthesis of some membrane proteins and globular proteins at high levels
J Mol Biol 260, 289-298

Ogren WL (1984) Photorespiration - Pathways, Regulation, And Modification Annu Rev
Plant Physiol Plant Mol Biol 35, 415-442

Overbeek R, Begley T, Butler RM, Choudhuri JV, Chuang HY, Cohoon M, de Crecy-
Lagard V, Diaz N, Disz T, Edwards R, Fonstein M, Frank ED, Gerdes S, Glass EM,
Goesmann A, Hanson A, Iwata-Reuyl D, Jensen R, Jarnshidi N, Krause L, Kubal M,
Larsen N, Linke B, McHardy AC, Meyer F, Neuweger H, Olsen G, Olson R, Osterrnan
A, Portnoy V, Pusch GD, Rodionov DA, Ruckert C, Steiner J, Stevens R, Thiele I,
Vassieva O, Ye Y, Zagnitko O, Vonstein V (2005) The subsystems approach to genome
annotation and its use in the project to annotate 1000 genomes Nucl Acids Res 33, 5691-
5702

Palmieri L, Picault N, Arrigoni R, Besin E, Palmieri F, Hodges M (2008) Molecular
identification of three Arabidopsis thaliana mitochondrial dicarboxylate carrier isoforms:
organ distribution, bacterial expression, reconstitution into liposomes and functional
characterization Biochem J 410, 621-629

Picault N, Palmieri L, Pisano I, Hodges M, Palmieri F (2002) Identification of a novel
transporter for dicarboxylates and tricarboxylates in plant mitochondria - Bacterial

expression, reconstitution, functional characterization, and tissue distribution J Biol
Chem 277, 24204-24211

Renne P, Dressen U, Hebbeker U, Hille D, F lugge UI, Westhoff P, Weber APM (2003)
The Arabidopsis mutant dot is deficient in the plastidic glutamate/malate translocator
DiT2 Plant J 35, 316-331

Reumarm S, Weber APM (2006) Plant peroxisomes respire in the light: Some gaps of the
photorespiratory C-2 cycle have become filled - Others remain Biochimica Et Biophysica
Acta-Molecular Cell Research 1763, 1496-1510

Schmid M, Davison TS, Henz SR, Pape UJ, Demar M, Vingron M, Scholkopf B, Weigel
D, Lohmann JU (2005) A gene expression map of Arabidopsis thaliana development
Nature Genet 37, 501-506

Schneider G, Lindqvist Y, Branden C1 (1992) Rubisco - Structure And Mechanism Annu
Rev Biophys Biomolec Struct 21, 119-143

Schwacke R, Schneider A, van der Graaff E, Fischer K, Catoni E, Desimone M, Frommer
WB, Flugge UI, Kunze R (2003) ARAMEMNON, a novel database for Arabidopsis
integral membrane proteins Plant Physiol 131, 16-26

196

Schwarz D, Klammt C, Koglin A, Lohr F, Schneider B, Dotsch V, Bernhard F (2007)

Preparative scale cell-free expression systems: New tools for the large scale preparation
of integral membrane proteins for functional and structural studies Methods 41, 355-369

Somerville CR, Ogren WL (1979a) Gas-Exchange Analysis of a Photosynthesis-
Photorespiration Mutant of Arabidopsis-Thaliana Plant Physiol 63, 152-152

Somerville CR, Ogren WL (1979b) Phosphoglycolate Phosphatase-Deficient Mutant of
Arabidopsis Nature 280, 833-836

Somerville CR, Ogren WL (1980) Photo-Respiration Mutants of Arabidopsis-Thaliana
Deficient in Serine—Glyoxylate Aminotransferase Activity Proc Natl Acad Sci USA 77,
2684-2687

Somerville CR, Ogren WL (1981) Photorespiration-Deficient Mutants of Arabidopsis-

Thaliana Lacking Mitochondrial Serine Transhydroxymethylase Activity Plant Physiol
67, 666-671

Somerville CR, Ogren WL (1982) Mutants of the Cruciferous Plant Arabidopsis-Thaliana
Lacking Glycine Decarboxylase Activity Biochem J 202, 373-3 80

Somerville SC, Ogren WL (1983) An Arabidopsis-Thaliana Mutant Defective in
Chloroplast Dicarboxylate Transport Proc Natl Acad Sci USA 80, 1290-1294

Steinhauser D, Usadel B, Luedemann A, Thimm O, Kopka J (2004) CSB.DB: a
comprehensive systems-biology database Bioinformatics 20, 3647-3651

Studier FW (2005) Protein production by auto-induction in high-density shaking cultures
Protein Expr Purif 41, 207-234

Vol] LM, Jamai A, Renne P, Voll H, McClung CR, Weber APM (2006) The
Photorespiratory Arabidopsis shml Mutant Is Deficient in SHMl
Plant Physiol 140, 59-66

Waadt R, Kudla J (2007) Cold Spring Harbor Protocols

Weber A, Menzlaff E, Arbinger B, Gutensohn M, Eckerskom C, Flugge UI (1995) The
2-oxoglutarate/malate translocator of chloroplast envelope membranes: molecular
cloning of a transporter containing a 12-helix motif and expression of the functional

protein in yeast cells Biochemistry 34, 2621-7.

Weber APM (2004) Solute transporters as connecting elements between cytosol and
plastid stroma Curr Opin Plant Biol 7, 247-253

197

Wingler A, Lea PJ, Quick WP, Leegood RC (2000) Photorespiration: metabolic pathways
and their role in stress protection Philos Trans R Soc Lond, B, Biol Sci 355, 1517-1529

Winter D, Vinegar B, Nahal H, Ammar R, Wilson GV, Provart NJ (2007) An Electronic
Fluorescent Pictograph Browser for Exploring and Analyzing Large-Scale Biological
Data Sets PLoS ONE 2, e718

Young XK, McCarty RE (1993) Assay Of Proton—Coupled Glycolate And D-Glycerate

Transport Into Chloroplast Inner Envelope Membrane-Vesicles By Stopped-Flow
Fluorescence Plant Physiol 101, 793-799

198

Chapter 7
Dynamics of the membrane systems surrounding the chloroplast stroma revealed by

proteomics and GFP fusion proteins

Andrea Brautigamm, Frederique K. H. Breuers], Henrik Tjellstrtims, John Froehlich4,

John Ohlrogge3, Anna Stina Sandelius5 and Andreas P. M. Weber"3

I would like to acknowledge Henrik Tjellstrom and Anna Stina Sandelius for preparation
of the plastid associated membrane and ER enriched microsome proteome samples. I
would also like to acknowledge Frederique K. H. Breuers for cloning and infiltrating the

constructs used for protein localization by transient expression tobacco plants.

199

Abstract

The soluble stroma of plastids is surrounded by a two membrane systems, the
outer and the inner envelope. Traditionally, the plastid has been considered completely
separated from the remainder of the cell but recent evidence has pointed to tight
connections between the chloroplasts and the remainder of the cell. We have developed
Pisum sativum as a model system for organellar proteomics which now allows us to
isolate and compare not only the inner and outer envelopes, but also the newly discovered
plastid associated membranes (PLAMs) and a light microsome fraction representing the
ER. Comparative proteomics showed expected distribution of marker proteins and also
allowed identification of putative markers for PLAMs and revealed remarkable dynamics
of outer envelope proteins. The comparative proteome data constitutes a new resource for
differential localization data for proteins of the chloroplast envelope. Transient
expression of GFP fusions of the candidate PLAM proteins supported PLAM localization
for two membrane bound and one soluble protein. Analysis of GFP fusions of outer
envelope proteins over time showed induction of stromule and vesicle formation with
increasing protein load on the system only for outer envelope proteins but not for inner
envelope proteins thereby identifying a tool for systematic analysis of stromule formation
and function. No proteomic evidence for a role of PLAMs in lipid transfer between the
ER and the chloroplasts was found. However, the data led to a new hypothesis that
PLAMs may represent budding autophagosomes at the tip of stromules and at the

chloroplast envelope.

200

Introduction

Chloroplasts are the defining organelles of the Archaeplastida (Adl et al., 2005).
Frequently, the chloroplast has been viewed as a distinct entity, separated from the
remainder of the cell by the membrane system surrounding the chloroplast, the inner and
the outer envelope membranes. The inner envelope is the true barrier to solute import and
metabolite and ion exchange which are mediated by a number of specific transport
proteins (Weber, 2004; Weber et al., 2005; Weber and Fischer, 2007). The outer envelope
is relatively permeable to small solutes by way of a number of porins with a broad
metabolite transport spectrum, but impermeable to molecules larger than lOkDa such as
most proteins (Weber, 2004; Weber et al., 2005; Weber and Fischer, 2007). Most proteins
are imported through a multi-protein complex spanning both the outer and the inner
envelope (Toc and Tie complex). Cargo proteins are addressed to a receptor at the outer
envelope by a chloroplast target peptide (cTP) and are transferred to the stroma through
two protein channels, Toc75 in the outer and Tic110 in the inner envelope (Jarvis and
Robinson, 2004; 8011 and Schleiff, 2004; Kessler and Schnell, 2006). Membrane proteins
of the inner envelope are also imported through this system and inserted either during or
after transit (Li and Schnell, 2006; Tripp et al., 2007). In addition to plastid envelope
proteomics (Brautigam et al., 2008a; Briiutigam et al., 2008b) and detailed targeting
analysis (Villarejo et al., 2005; Nanjo et al., 2006) have revealed a number of chloroplast
resident proteins which do not possess a canonical cTP and enter the plastid by other
ways. For example, two soluble enzymes, a carbonic anhydrase and a nucleotide
pyrophosphatase reach the chloroplast stroma through the endomembrane system

(Villarejo et al., 2005; Nanjo et al., 2006), but the targeting from the Golgi compartment
201

to the chloroplast remains unresolved. Two outer envelope proteins, OEP64 (formerly
called Toc64) and OEP7, are both recognized by a protein receptor, AKR2A, and inserted
into the envelope independent of the Toc and Tic complex proteins (Bae et al., 2008).
The binding motif, hydrophobic amino acids arranged in an alpha helix followed by a
stretch of charged amino acids (Lee et al., 2001; Bae et al., 2008), is similar to the
canonical target peptide for the secretory pathway which also consists of a stretch of
hydrophobic amino acids (F ourrier et al., 2008). Only two cargo proteins for this pathway
are known to date, hence no prediction programs are available for the binding motif.
Although the envelope membranes of plastids have often been viewed as static
barriers, they are indeed a remarkably dynamic system. The envelopes can form long
structures protruding from the body of the chloroplast (Gray et al., 2001). These
‘stromules’ are lined by both the inner and the outer envelope and contain stroma but no
thylakoids (Gray et al., 2001; Hanson and Sattarzadeh, 2008). They can extend and
retract and even shed their tip, resulting in a stroma filled vesicle (Hanson and
Sattarzadeh, 2008). Their function is currently not known (Hanson and Sattarzadeh,
2008). The envelope can form vesicles both to the inside as well as to the outside. Plant
viruses can induce the formation of invaginations and eventually vesicles at the envelope
which harbor the virus replication machinery (Hatta et al., 1973; Prod'homme et al.,
2003). Vesicles towards the cytosol are formed when chloroplast proteins are destined for
degradation in the vacuole (Ishida et al., 2008) but the mechanism underlying the
formation of these vesicles is not understood. Finally, advanced microscopy techniques
have revealed that a fraction of the ER is in close contact with the chloroplasts and that

even during chloroplast isolation a fraction of the ER remains firmly attached to the

202

chloroplasts (Andersson et al., 2007). This fraction has been named plastid associated
membranes and it has been hypothesized that this may represent a contact site between
the endomembrane system and the plastids (Benning et al., 2006; Andersson et al., 2007).
The production of fatty acids and lipids is dependent on a close and efficient interaction
of two different compartments, the ER and the chloroplast (Benning et al., 2006).
Essentially all acyl chains are produced in plastids and in most plants 85% or more of
these acyl chains must cross the envelope before assembly into glycerolipids.
Furthermore, glycerolipids can be exchanged between both compartments in both
directions (Schnurr et al., 2002; Xu et al., 2005; Awai et al., 2006; Benning et al., 2006;
Lu et al., 2007) and indeed in terms of lipid traffic, the plastid envelope may be the most
active membrane found in biology.

A protocol for separation of outer and inner envelope membranes has been
developed for leaf chloroplasts of Pisum sativum (Cline et al., 1981). The isolation of
outer envelope membranes of suitable quality for proteomics requires large amounts of
leaf material making the adaptation of the protocol for the model plant Arabidopsis
impractical. In this work, we made use of a novel transcriptome database for Pisum
sativum specifically developed to enable organellar proteomics (Brautigam et al.,
2008b).The different membrane systems surrounding the chloroplast, the inner and outer
envelope, plastid associated membranes (PLAMs) and an ER enriched microsome
fraction, were defined by comparative proteomics and the results were integrated with
bioinforrnatic targeting information. Both the outer envelope membranes as well as the
plastid associated membranes are considerably less static than previously assumed.

Intense protein dynamics through the membrane systems are confirmed by GFP fusion

203

proteins. Finally, the distribution of proteins involved in fatty acid and membrane lipid

biosynthesis as well as novel candidate transport proteins are discussed.

Material and Methods

Preparation and analysis of proteome samples

Four different proteome samples from 10-14 days old pea plants were isolated.
Chloroplasts from pea leaves were isolated as described previously (Douce and Joyard,
1979; Keegstra and Yousif, 1986). Inner and outer envelopes were separated as described
by (Cline et al. 1981), plastid associated membranes and ER enriched microsomes were
isolated as described by (Andersson et al., 2007).Membrane samples were mixed with
SDS—PAGE loading buffer and proteins were dissolved by shaking at 15° C for 20
minutes. Proteins were separated by SDS-PAGE and each gel lane was cut into ten
equally-sized slices. The proteins of each slice were digested with trypsin and modified
according to (Shevchenko et al., 1996) and loaded automatically onto a Waters Symmetry
C18 peptide trap (5 pm, 180 um x 20mm) at a flow rate of 4 uL/min in 2%
Acetonitrile/0.1%Formic Acid for 5 minutes by a Waters nanoAcquity Sample Manager.
The peptides were separated on a Waters BEH C18 nanoAcquity column (1.7 pm, 100
um x 100mm) with a 90 minute gradient using a Waters nanoAcquity UPLC and eluted
into a ThermoElectron LTQ-FTICR mass spectrometer with a flow rate of 300 nL/min
(Buffer A = 99.9% Water/0.1% Formic Acid, Buffer B = 99.9% Acetonitrile/0.1%
Formic Acid: gradient of 5% B to 40% B from 0 to 63 minutes, 40% B to 90% B from 63
to 71 minutes and 5% B from 71 to 90 minutes). Survey scans at a resolution of 50,000

were used to identify the top ten ions of each survey scan which were then subjected to
204

automated low energy collision induced dissociation. A BioWorks Browser v3.2
converted the MS/MS spectra to peaks lists which were compared to a Pisum sativum

cDNA database using the Mascot searching algorithm, v 2.2 (www.matrixsciencecom).

 

Carbamidomethyl cysteine was set as a fixed peptide modification and oxidation of
methionine was allowed. Up to two missed tryptic sites were allowed. The peptide
tolerance was set to +/-10ppm and the MS/MS tolerance to 0.8kDa. 'Mascot results were
imported into Scaffold® for analysis.
Data analysis

Protein identifications were called if at least two peptides were identified of which
each had to be identified itself with 95% probability and the total protein identification
probability was 99%. Identifications of the mixed envelope sample were imported from
PRIDE (Brautigam et al., 2008a) and added into the analysis. For all matches, the closest
Arabidopsis Blast Hit was retrieved and both annotation and categorization is based on
the Arabidopsis protein. Targeting information, structural analysis and prediction of
transmembrane helices were retrieved with the bulk data retrieval tool at TAIR

(wwwarabidopsisorg). A protein was assigned to a membrane system based on its

 

absolute spectral count observed in the proteome samples. Having twice as many counts
in one or two categories over the others merited designation with three different
confidence classes [class 1: ratio >twofold, absolute count >10; class 2: ratio >twofold,
absolute count >5; 3 trend] with class 4 referring to identification in three or more

membrane systems.

205

GFP fusion protein construction and analysis

For selected proteins, the localization was tested using C-terminal GFP fusion
constructs. The cDNA of interest was amplified from RNA isolated from different
developmental stages of Arabidopsis (young seedling, adult leaves or senescing leaves),
using oligonucleotide primers compatible with the GATEWAY BP reaction system. The
forward primer contained an optimized KOZAK consensus to ensure efficient translation.
After BP reaction, the vectors were isolated and the insert was mobilized into pMDC83
((Curtis and Grossniklaus, 2003). GFP was observed after transient expression in tobacco
leaves (Waadt and Kudla, 2006) and isolation of protoplasts after one, two, three and four
days of leaf infiltration with a Pascal confocal microscope. GFP fluorescence was
isolated with a bandpass filter BP505-550nm; chloroplast fluorescence was filtered with a
longpass filter LP650 and imaged after excitation with an argon laser at 458 and 488nm

with 5% intensity. Pictures represent a single optical slice.

206

Results

The overall proteome analysis reveals a distinct protein localization pattern.

 

 

 

 

 

 

 

200 J — chloroplast
mitochondrion
C: secreted
g [:1 no targeting signal, cytosolic
% 150 + undefined
a
“5
E 100 ~
E
D
2
50 -
0 _
e“qe\0v
«\V: inferred localization

Figure 7-1: Protein distribution across the fractions plotted as number of proteins
determined for each membrane system including targeting predictions from TargetP; IE
inner envelope, OE outer envelope, PLAM plastid associated membrane; multiple localization
cannot be determined because the protein distributes evenly;

Proteomics identified a total of 522 proteins with at least two peptides each
(peptide probability, p<0.05; protein probability p<0.01) (Table 7-Sl). All but six could
be mapped onto an ortholog in the Arabidopsis proteome (Table 7-Sl). The localization
of each protein was inferred by comparing its absolute spectral abundance in each sample
(for details see material and methods). Not all proteins could be assigned to one of the

samples. A number of proteins were equally abundant in two adjacent membrane systems

207

(i.e., some of the IE/OE localized proteins could not unequivocally be sorted into either
of the categories but were identified in both samples). Twenty proteins were identified
with similar abundance in all samples. The number of protein identifications was not
evenly distributed. Both the inner envelope and the ER enriched microsome fraction
harbored more proteins than the outer envelope and the PLAM samples (Figure 7-1) with
203 proteins in the inner envelope and 117 proteins in the ER-enriched microsomes. For
each of the proteins the localization was predicted from its sequence by TargetP and
plotted onto the overall protein distribution (Figure 7-1). Many of the proteins only
identified in mixed envelopes (Figure 7-1) have canonical target peptides for the Tic and
Toc complex, as do most of the proteins categorized to the inner envelope fraction, and
are identified as chloroplast proteins by prediction programs (Emanuelsson et al., 1999;
Emanuelsson et al., 2000). In these categories, very few are predicted to be targeted to the
endomembrane system or the cytosol. 26% and 28%, respectively, cannot be sorted by
bioinformatics at all and are classified as undefined. These proteins present conflicting
evidence for targeting at their N-terminus and can therefore not be predicted by the
program (Figure 7-1). The 39 proteins assigned to the outer envelope fraction present a
very different prediction pattern. Proteins appear to have very different N-termini most of
which cannot be predicted by the program, resulting in an ‘undefined’ prediction. The 32
proteins of the PLAM fraction present a picture rather like the inner envelope proteins.
The second category with well defined bioinformatics predictions is the fraction of
proteins residing in the ER based on their selective enrichment in the ER-enriched
microsome fraction. This category contains most of the proteins also predicted to be

targeted to the secretory pathway though over half of the proteins cannot be assigned to a

208

location by the prediction program. In addition to a few proteins predicted to enter the
chloroplasts, some of the proteins are predicted to localize to the mitochondria. Many of
these are proteins that have already been localized to the mitochondria by various
methods indicating that the ER-enriched microsomal fraction but none of the others also
contains bonafide mitochondrial proteins (Table 7-Sl).
Known residents of the inner and outer envelope

To analyze whether the preparation of the proteome samples had succeeded in
enriching the distinct fractions, the protein import complex (Tic and Toc complex) was
analyzed. It has been studied in great detail in the past three decades and the localization
of the resident proteins has been well established. The outer envelope components are
denoted by Toc for translocon outer envelope of chloroplasts and a number representing
their molecular weight at first analysis and the inner envelope components are denoted as
Tic for translocon inner envelope of chloroplasts and a number. All proteins residing in
the outer envelope based on prior knowledge have their highest spectral abundance in the

outer envelope proteome sample (Figure 7-2).

209

>

 

100 ~ ‘ 1

 

 

 

 

 

 

 

 

 

     

 

 

 

 

E _ IE 3
§ [:1 CE g
_ 80 PLAM 2
e * :m a
+- 4
o r
o a
5i i
- I
5 6°
-°- 3
a: 3
e 4o -
8
0’ i
to r
E 20 - g
8 2
'03 Z
a. 2
0 2
0‘0 .\\ 0 5 1.17)“ ,\\\
{\c fl 9(0) \ «010 T‘ ‘0“ $05 «(0'3" alfio’fi'
B
100
2:5 - IE ‘1
_ or: 5;
§ 80 . PLAM 3
g -' [:1 'ER' 3
§ 60 - g.
a 40 - 3
3 r
C
9.3 2o -
w a i
a '12 i
'2 i nd
0 _ .. . _;f .3 .
‘13 N“ - 03A ’\ 59 «'5 37' «’29
«0013' 0016; 0016‘ 0016 Ciro «0553006 00 .300

Figure 7-2: Distribution of the spectral count between the proteome samples for proteins of
the protein import complex; (A) Proteins of the inner envelope complex; (B) proteins of the
outer envelope complex; IE inner envelope sample, OE outer envelope sample, PLAM plastid
associated membranes sample, ‘ER’ light microsome fraction, nd not detected

210

They include the import receptor Toc159, the two Toc75 isoforms one of which is
the protein import channel through the outer envelope. Both Toc64 and ToC34 are also
most abundant in the outer envelope sample. The alternative import receptor Toc132 not
only localized to the outer envelope but has been identified in a chloroplast proteome
sample for the first time. Virtually all of the Tocs can also be found in the PLAM
proteome sample with high spectral abundance but not in the ER enriched microsome
fraction. In contrast, the inner envelope components distribute mostly to the inner
envelope proteome sample. Ticl 10, the presumed channel through the inner envelope, as
well as Tic 40 and Tic55 are mainly in the inner envelope (Figure 7-2). One Tic22
isoform presumably localized to the interrnembrane space, but associated with the
transport complex at the inner envelope is also identified in the inner envelope fraction
but the other Tic22 isoform is exclusively localized in the PLAM fraction (Figure 7-2). In
addition to the proteins of the import complex, known and putative transport proteins and
porins were checked. Members of the mitochondrial carrier family, the putative and
known metabolite transport proteins, and most ATP dependent transport proteins
localized to the inner envelope (Table 7-Sl). Only WBC7, an ATP dependent transport
protein, localized to the outer envelope. All the porins localize to the outer envelope
(OEP37, OEP24) or to the PLAMs (OEP24) but a protein related to OEP16 localized to
the inner envelope.

Proteins of the outer envelope and the plastid associated membranes

Thirty-seven proteins were assigned to the outer envelope, 8 were assigned to

both, the outer envelope and the PLAMs, 32 localized to the PLAM, and 13 were

assigned to both the PLAM and the ER (Figure 7-1). The PLAM fraction contains the
211

largest nrunber of proteins that are predicted to be soluble whereas the fraction of proteins
shared between the PLAMs and the ER contains the largest proportion of membrane
protein (Table 7-Sl). In the outer envelope, the proteins with the highest spectral count
are those involved in transport. All known proteins of the Toc complex except Toc120
and all known outer envelope porins for metabolite transport were identified. The only
putative transport protein localized to the outer envelope is the ABC transporter WBC7.
In addition there are a number of proteins with putative functions in fatty acid and lipid
metabolism as well as proteins containing predicted protein-protein interaction domains.
Proteins with a metabolic role other than lipid metabolism include an ascorbate
peroxidase, a glysosyl hydrolase whose target is unknown, hexokinase and a number of
putative enzymes (Table 7-S2). One protein of unknown function consists of only beta
strands (Table 7-S1), but overall no trend towards preferential accumulation of beta-fold
proteins was identified in either of the outer envelope associated fractions (Table 7-S1).
After it became apparent that many fractions also contain known stroma proteins, the
known localizations of proteins (endomembrane system, envelopes, stroma, localization
not classified and protein of unknown function, see Table 7-1) were determined. Whereas
the inner and the outer envelope contain less than ten and less than fourteen percent of
stromal proteins, the fraction of proteins seemingly divided between these two
compartments contains mostly stromal proteins and proteins of unknown fimction (Table

7-1).

212

Table 7-1: Known localizations of proteins assigned to the different fractions; localization
was not classified if it is not known or known to be outside the three criteria (e.g. mitochondrial
for proteins assigned to the ER fraction)

percentage of proteins in each fraction

 

 

classical
localization lE lE/OE OE OEIPLAM PLAM PLAM/ER ER
endomembrane
system 0.5 0.0 2.8 12.5 6.5 69.2 39.6
envelopes 16.8 8.1 25.0 37.5 9.7 0.0 1.1
stroma 9.9 35.1 13.9 37.5 41.9 0.0 0.0
localization not
classified 38.1 32.4 27.8 0.0 16.1 15.4 20.9
pmtei" °f ””kmw" 34.7 24.3 30.6 12.5 25.8 15.4 33.5

funcfion

 

Seven proteins are shared between the OE and the PLAM samples, among them
four enzymes (Table 7-S3). There is a glucosidase known to localize to the OE (Fourrier
et al., 2008), a putative transaldolase, a monodehydroascorbate reductase known to
localize to peroxisomes (Lisenbee et al., 2005; Eastrnond, 2007), and a putative subunit
of the pyruvate dehydrogenase complex. In addition to the enzymes there are also two
isoforms of the outer envelope porin OEP16 (Pohlmeyer et al., 1997) and a protein of
unknown function carrying a prefoldin domain.

The proteins mainly localized in the PLAMs consist mostly of known stromal
proteins and proteins of unknown function (Figure 7-3). There are plastidic ribosomal
proteins (eight) and proteins of unknown function (eight). In addition to proteins involved
in protein biosynthesis there are also three proteins involved in protein folding, two
Hsp60-like proteins and a chaperonin. Three putative enzymes, two different pyruvate
dehydrogenase complex subunits and a dephospho-CoA kinase also localize exclusively
to the PLAMs. The mitochondrial import protein Tom-20 is part of this group as is one of

the Tic22 isoforms identified. Finally, there is a group of proteins with diverse functions,
213

the outer envelope porin 24, a putative steroid binding protein, Chupl, a protein involved
in chloroplast positioning, a subunit of plasmamembrane ATPase, cytochrome b5, two
ferritins, and a eukaryotic ribosomal subunit. Many proteins of the PLAM fraction have
canonical cTPs (Figure 7-1).

The third group of proteins is shared between the ER-enriched microsomal
fraction and the PLAMS. Notably, both stromal and envelope proteins are absent from
this fraction (Table 7-2). There are four ATPase subunits, three proteins involved in
protein folding, and two proteins of unknown function. A VDAC-like porin, a
strictosidine synthase-like protein and another protein involved in red light signaling are
also present in this group. Compared to the proteins assigned to the outer envelope (14
predicted to have transmembrane helices, 23 predicted to be soluble), PLAMS are
enriched in soluble proteins (27 to five with predicted transmembrane helices). PLAMS
contain a mixture of proteins traditionally considered plastid like proteins and of proteins
traditionally considered ER proteins as well as a number of proteins of unknown function
(Figure 7-1, Table 7-1).

Within the ER fraction, known residents of the endomembrane system, in
particular proteins involved in protein folding, are identified. Some of them are
exclusively identified in the ER- enriched microsome fraction whereas most are shared
with the PLAMS. Only one of the proteins, a calnexin is also identified in the chloroplast
envelopes.

GF P localization data
The dynamic localization of a number of proteins localized to the outer envelope

or the PLAMS was investigated by fusion to GFP and transient expression in tobacco

214

leaves. The transient expression driven by the cauliflower mosaic virus 35S promoter
allowed monitoring of the cellular distribution of the GF P signal with increasing protein
load on the system with time and hence a dynamic distribution of fusion proteins.

The outer envelope protein Toc64 known to be an outer envelope protein served
as the control for the outer envelope; likewise Mepl as the inner envelope control.
MDAR (peroxisome), TGD4 (ER) and hexokinase 1 (mitochondrion) were included
since they have been reported to localize to compartments other than the chloroplast by
previous publications (Damari-Weissler et al., 2007; Eastmond, 2007; Xu et al., 2008).
The Tic22 isoform, the EF hand protein, and the kinesin—like protein were included to
determine whether they may serve as markers for the PLAMS in future. Two heatshock
proteins were evaluated since they are expected to be soluble but were identified in the
membrane samples and six proteins of unknown function were tested for their

localization.

215

 

359:5»

 

.23“. .2. 22.8 9. .5288 a? 89.20 +5.9. 08832
.2<..n. 2. 38m 2. ...-umo: E0 25%?
mo 9. 22.0 2. 9...-th at 3.8; 3939..
m. 2. 28m .25 8.2.0 a: cook 882% 9.8 08832
....Sn. 2 8.0.8., 8.83 E 520:. 3.78%... “Po 23 En...
5.8.088
. . 336:3.
23amo > $90 7.98.0.5 38.... 5.9. 8389....
Em: .303 22.889353055
>
5.52: > 890 98m 38m masofimu so... .8920 +5.9. 25892
mo > $90 9228...... 9.22.8.5 F 82.3.92 89.20 +5.9. on Emma...
59:8
8.9.3:. > 8.3.0 .836 .825 $85. 5.; 5205 8.8.8 9.9. 8839.,
m0 2 890 $389.5 new new. 3.9.8.35. 5:829. 5505...: 3920 ten; owmwommz.
mo >_ 330 35895 new 9.8 $3879: .2833: Bow 33 3920 +56: 80:93
mo_o_mo> .o cozoca
mo ... 98.0 8.3595 as... 95. ac.m+mo.o.$> 5.65:: ..o 5.29.. 8920 +5.9. 0558....
2.3.8
5.5“. ... $3 8.3526 2m 8.x ac_m+mm_o.$> “m. 5.; 532... £29.... 8 82832
m0 =. $90 2. 8.2m? 88.. 2.53 E0 38.39....
mo_o_mo> .0
mo .. 88 8.3596 Em 9E 9...... 3 $0 . 899.0 2.9. 2233,
m_ _ $20 6330 new 9.2 QB 5203 30.95 55.. n.._.o omommmz<
£32m 538
95205 539230
:0 comma comm—“Man 39565 cofimfios countomoo GMMMWW... _0<
co=m~__moo_ :0. S. moo .otm co_um~__moo._ . . .
85.... .. ._ ..

.Eomoa tong £865 BEBE <Nm¥< mo >=oE wczowhfi cowasrxfi;
.23:qu 695 328826 Pro .8522: oEocow m_mgoc5m.< _O< ”2:89:— E<Jm E... 2.2023 58.5 $83.8 ..o. 338.. 555.32 .EU "NA. 933—.

216

The fluorescence pattern of fifteen fusion proteins showed four distinctive
chloroplast fluorescence patterns when expressed transiently in tobacco leaf cells for 1-4
days. Class I proteins initially localized in a circle surrounding the chlorophyll auto-
fluorescence (Figure 7-3A) and prolonged overexpression of the protein results in
unspecific accumulation in the cytosol (Figure 7-3B). This pattern is displayed by the
inner envelope control protein, Mep1 (Table 7-2). Proteins of class II initially display a
ring shaped fluorescence surrounding the auto-fluorescence of chlorophyll (Figure 7-3C).
OEP64 is the sole member of this class (Table 7-2). Its fluorescence pattern eventually
displays extensive stromule formation (Figure 7-3 D). Frequently, instead of stromules, an
abundance of vesicles of 1-1.5um diameter is visible (Figure 7-3D). Proteins of the third
class label a vesicle membrane with distinct openings with a weaker label in a circle

surrounding the chlorophyll auto-fluorescence

217

 

Figure 7-3: OF P localization of fusion proteins after initial (A, C, E and G) and prolonged
(B, D, F, H) transient expression in tobacco leaves; bar size is Sum for panels A to G, for panel
H 2um; for each picture: lefi panel chlorophyll auto-fluorescence, middle panel GFP fluorescence
and right panel overlay; localization patterns of the inner envelope marker Mepl (A, B), the outer
envelope marker OEP64 (C, D), a protein of unknown function with EF hand domains (E, F) and
an ABC transporter of unknown function. WBC7 (G, H); this image is in color

218

 

Figure 7-3E). Prolonged expression will result in stronger labeling of the circle and
eventually formation of stromules (Figure 7-3F). A protein with EF hands, a putative
lipase and a small protein without known domains display this pattern (Table 7-3).
Proteins in class IV also label initially a circle around the chlorophyll auto-fluorescence
but this circle frequently has loops (Figure 7-3G). Prolonged expression will result in
stromule formation (Figure 7-3H) and occasionally vesiculation (as visualized for the
Class 11 protein OEP64, Figure 7-3D, lower panel). The formation of stromules and
extensive vesicles are observed for all putative outer envelope proteins but not for Mepl.
Class V encompasses all extraplastidial localizations. The kinesin-like protein localized
to the cytosol, the hexokinase localized to small dots of a fixed size whereas the
monodehydroascorbate reductase and the sterolmethyltransferase localize to spots of

varying size.

220

A B

C D
Figure 7-4: GFP expression pattern of Hsp60-like protein after three days of expression in
tobacco leaves; (A) OF P fluorescence is strongest in extension of the chloroplast envelopes and

vesicles, (B) detail of A, vesicle forming at the envelope, (C) GFP expression is visible in a
vesicle forming at the tip of a stromule, (D) detail of (C); this image is in color

 

 

Transiently expressed GFP fusion proteins of Hsp60-like co-localizes with
chlorophyll auto-fluorescence but is particularly visible in vesicles attached to the
chloroplast envelope (Figure 7-4A and B) and forming at the end of stromules (Figure 7-
4C and D).

Discussion
Targeting Predictions for Envelope proteins

To understand the network of proteins residing in the membrane system
surrounding the plastids, the inner envelope, the outer envelope, and the PLAMs, we
analyzed four separate proteome samples, one preparation enriched in inner envelopes,

one enriched in outer envelopes, an acid wash of intact chloroplasts to obtain a PLAM

221

sample, and an ER—enriched microsome fraction. The inner envelope fraction was
considerably richer in proteins than the outer envelope, as had been suggested previously
based on gel electrophoretic analysis (Cline et al., 1981; Wernerwashburne et al., 1983).
This is supported by mixed envelope samples (Ferro et al., 2003; Froehlich et al., 2003;
Brautigam et al., 2008a), which mainly contain proteins of the inner envelope with very
few outer envelope proteins present (Table 7-S1). Bulk data retrieval from TAIR which
uses mainly the TargetP program (Emanuelsson et al., 2000) for assignment identifies
about two thirds of the proteins assigned to the inner envelope as chloroplast residents
based on their N-terminal cTP (Figure 7-1) confirming that most of the proteins of the
inner envelope reach it by a conventional targeting mechanism through the protein import
complex. Of the remainder of the proteins, most have conflicting targeting information at
their N-terminus resulting in ‘undefined’ targeting. Much of this conflicting targeting
information can be resolved by manual inspection of the targeting results, which
frequently identify the mitochondrion as the alternative target (data not shown). This kind
of targeting is frequent for inner envelope proteins with the most prominent case being
the most abundant inner envelope protein, the triosephosphate phosphate translocator.
There are also two percent of the proteins which appear to have a targeting signal for the
secretory pathway as has been demonstrated in other plastid envelope proteome projects
of mixed envelopes. In total chloroplast proteomics, the number of proteins with
canonical target peptide was higher (86%) and fewer proteins with target peptides to the
secretory pathway (0.6%) were identified (Zybailov et al., 2008). This may reflect more
diverse protein targeting pathways of inner envelope proteins compared to stromal

proteins. The pattern of targeting predictions for the proteins assigned to the outer

222

envelope is very different from the pattern of proteins of proteins assigned to the inner
envelope. It contains a large proportion of proteins with predicted target peptides for the
secretory pathway and many proteins with ‘undefined’ targeting as well as proteins with
canonical cTPs. Manual inspection of the targeting for the proteins with ‘undefined’
localization by bulk data retrieval with the help of ARAMEMNON reveals that many of
these proteins carry weak signals for the secretory pathway (Table 7-Sl). Unlike the inner
envelope proteins, most of the known outer envelope proteins such as members of the
Toc complex do not contain a cTP ($011 and Schleiff, 2004) except for Toc75-III which is
targeted by a cTP and imported through the classical protein import complex (Inoue and
Potter, 2004). The different N-terminal motifs among the proteins of the outer envelope
may indicate that there is a multitude of targeting pathways. Apart from the targeting
mechanism for Toc75, the targeting mechanism for two additional outer envelope
proteins is known. They are inserted with the help of the protein AKRZA and the
recognition motif at the N-terminus consists of a stretch of hydrophobic amino acids
followed by positively charged amino acids (Bae et al., 2008). This pattern may be
misrecognized as a weak signal for the secretory pathway. Many proteins of the outer
envelope fraction hence represent putative additional cargo for the AKR2A mediated
pathway. For example, LACS9 and WBC7 both contain a predicted transmembrane helix
at the very N-terminus which is followed by charged residues (Table 7-2). Two soluble
proteins which reach the chloroplast stroma through the endomembrane system are
known to date showing a connection between the Golgi and the chloroplast (V illarejo et
al., 2005; Nanjo et al., 2006). Possibly there are also membrane proteins targeted via the

endomembrane system. The targeting predictions hint that proteins can take a multitude

223

of pathways to enter the outer envelope membrane. Therefore, proteome analysis
followed by GF P fusion protein analysis as conducted here represents the only method to
reliably determine the outer envelope proteome. The ER enriched microsome fraction
consists mostly of proteins with a canonical target peptide for the secretory pathway or of
undefined targeting. Only very few proteins in this fraction carry an N-terminal extension
recognized by prediction programs to target the chloroplast or the mitochondrion.
Although the ER enriched microsome fraction represents the proteome sample with the
most mitochondrial proteins, the total amount of contamination is very low, emphasizing
the purity of the preparations.
Protein Import Complex

With the different proteome samples at hand, not only large scale overall
assignment are possible but also detailed localization analysis of proteins with known and
putative function as well as analysis of isoforms within gene families. The strength of this
method becomes obvious when the protein import complex is considered (Figure 7-2).
Eight different proteins of the Toc complex were identified; the main import channel
Toc75-III, three isoforms of Toc75-V, the import receptors ToclS9 and Toc132 and
ToC34 as well as Toc64. All proteins have their highest absolute spectral count in the
outer envelope sample but they are also isolated in the inner envelope and PLAM but not
the ER-enriched microsome fraction. Cross contamination of inner and outer envelope
samples with each other has been observed frequently (Cline et al., 1981; Schleiff et al.,
2003) but the strong presence of import complex components in the PLAM sample
remains unexplained. For the Tic complex, the localization is almost exclusively to the

inner envelope and, to a lesser degree, to the outer envelope sample with two exceptions.

224

Two isoforms of Ticl 10, Tic40, one isoform of Tic22, Tic20, Tic62 and Tic55 are clearly
inner envelope proteins although Tic22 is described as only peripherally associated with
the inner envelope.Tic22-III is exclusively localized in the PLAMS. Transient expression
of Tic22-III OF P fusion proteins yielded a fluorescence pattern confirming localization to
structures distinct of the chloroplast. The function of Tic22 in the envelope is currently
unknown making it impossible to speculate about the function of its isoform in the
PLAMS. TiC32 displays a similar pattern of inner/outer envelope sample localization but
amazingly also is present in substantial amounts in the ER enriched microsome fraction.
Although the import complex largely distributes between the proteome samples as
expected, the strong presence of Toc complex components as well as Tic22-III in the
PLAMS and the unusual pattern of TiC32 posit new questions.
Purity of the proteome samples

Tagging a number of proteins with GFP allowed us to not only observe dynamics
of proteins within the envelope but also identify proteins which localize to other
organelles if expressed as GF P fusion proteins. The kinesin-like protein localized to the
cytosol (Table 7-2) and thus may have been peripherally associated with the outer
envelope, the PLAMS and the ER-enriched microsomes, and thus been identified in all
three proteome samples. For other proteins, the identification in the outer envelope or the
PLAMS is less easy to explain. Several proteins localized to defined spots (Table 7-2)
within the cytosol which may represent localization to mitochondria, peroxisomes or the
Golgi. The identification of these proteins in proteomics is probably due to contamination

of the preparation with these organelles which however may also represent a tight

225

association in vivo (Brautigam et al., 2008a). Once extraplastidal localization was
confirmed the proteins were removed from the analysis.
Trafficking of proteins to and from the plastid

The isolation and initial characterization of vesicles attached to the chloroplasts
that can be labeled with an ER marker protein (Andersson et al., 2007) prompted us to
analyze the proteome of this new vesicle class. In comparison to the envelope fractions
and the ER fraction, the PLAM fraction contains proteins with a different targeting
pattern, mostly proteins with a canonical TP indicating that the PLAM fraction is not a
subfraction of either the ER-enriched microsomes or the outer envelope. The composition
also argues against an artificial fraction produced during isolation of the fraction (Table
7-S1). In contrast, PLAMS appear to be a compartment assembled with proteins from
both the ER as well as the chloroplast with a large contribution of outer envelope
proteins. Not only the targeting information of the proteins from the outer envelope and
the PLAM fractions is different, the PLAM fraction also contains a large portion of
soluble stromal proteins. Both the pattern of targeting predictions as well as the protein
content of the PLAMS marks them as a hybrid compartment of the chloroplast and the
ER. Lipid analysis also indicates that PLAMS contain a mixture of lipids typical for
membranes of the endomembrane system and of the chloroplast membranes (Andersson
et al., 2007). Several of the proteins identified in the outer envelope and the PLAMS were
tested for their localization in vivo with transient expression of the respective GFP fusion
proteins (Table 7-2).

The analyses of targeting for GFP fusion proteins revealed unexpected patterns

with protein localizations that changed depending on the time of observation (Figure 7-

226

3E and F). Two of the GFP fusion proteins localize to vesicles with bright fluorescence
and comparatively dim fluorescence is observed in the envelope initially (Table 7-2). The
proteins are both of unknown function, one protein has EF hand domains indicating the
potential to bind calcium, the other one is a protein of unknown function. Prolonged
expression leads to localization to the envelope in a circle surrounding the chlorophyll
auto-fluorescence in addition to the vesicles (Figure 7-3E and F). The temporal pattern of
expression which progresses from vesicles to the envelope suggests the proteins mainly
localize to vesicles in vivo. The localization to vesicles is supported by the proteome
analysis which mainly localized the protein to the PLAM fraction (Table 7-2). To our
knowledge, this is the first demonstration of a membrane protein localizing to vesicles
associated with the chloroplast and the envelope both by proteomics and by GFP fusion
proteins. Both proteins progress from labeling vesicles and weakly labeling the envelope
to a strong labeling of the chloroplast envelope in addition to the vesicles. This labeling
pattern may indicate that the vesicles in vivo localization is in the membranes of the
vesicles but the proteins pass through the envelope on their way to the vesicles since
increased protein load leads to partial retention of the GFP signal in the envelope and in
stromules forming at the envelope. Neither of the proteins producing this pattern has a
canonical cTP (Table 7-2). A third protein mainly localizing to vesicles with dimmer
GF P fluorescence in the envelope and the stroma is a Hsp60 (Figure 7-4). The protein is
localized to the PLAMS by proteomics and it has a canonical cTP (Table 7-2). The
protein can indeed be detected in the stroma (Figure 7-4). However the strongest
fluorescent is visible in nascent vesicles forming at the tip of stromules (Figure 7-43) and

at the envelope (Figure 7-4D). Two proteins which were mainly detected in the PLAM
227

proteome sample were confirmed to localize to vesicles associated with chloroplasts by
GFP fusions. The third protein was detected mainly in outer envelope samples also
localizes to PLAMS when expressed transiently as a C-terminal GFP fusion. All three
proteins can serve as markers for future studies of this new vesicle class inside plant cells.
A number of putative outer envelope resident proteins were also investigated. These
proteins, which initially localize to vesicles or to the envelope (Table 7-3) with the
exception of the inner envelope marker protein Mep1, finally can be observed in either
stromules or vesicles (e.g. Figure 7-3D). The formation of vesicles out of stromules is a
well known phenomenon and has been described as ‘tip-shedding’ (Gray et al., 2001)
although the function of either stromules or vesicles remains unknown (Gray et al., 2001;
Hanson and Sattarzadeh, 2008). Formation of stromules can only be observed by
prolonged overexpression and therefore increased protein load but never during initial
stages of the experiment (Figure 7-3). In contrast to the known and putative outer
envelope protein fusions which reside in defined structures such as stromules or vesicles,
the inner envelope marker protein fusion accumulates as diffuse material in the cytosol
which may represent proteins not yet imported into the envelope (Figure 7-3B).
Formation of stromules upon overexpression of another inner envelope protein, the
triosephosphate phosphate translocator, has not been reported (Gray et al., 2001). The
difference in localization to the inner or outer envelope, respectively, seems to define the
pattern upon prolonged expression but additional experimentation with more inner
envelope resident proteins will be used to support the hypothesis. The overexpression of
outer envelope proteins thus revealed a novel protein localization pattern of unknown

significance. The vesicles labeled by outer envelope GFP fusion proteins are of l-1.5um

228

in size (Figure 7-3), the same size observed for autophagosomes in plant cells (Bassham,
2007; Ishida et al., 2008). Although the production of vesicles has not been observed
directly, they start to appear at the same time when stromules appear in the cells.
Remarkably, the PLAMS are richer in outer envelope proteins than in inner envelope
proteins (Table 7-Sl), a pattern which is particularly evident when components of the
protein import complex are considered (Figure 7-2). The correlation of these two
phenomena, comparative proteomics reveals PLAMS are rich in outer envelope
membranes and overexpression of outer envelope proteins leads to formation of vesicles,
prompted us to speculate that PLAMS may be budding autophagosomes. It is currently
not known how the different membranes systems are arranged within the PLAMS,
whether these are truly mixed membranes or whether they represent envelope vesicles
engulfed by ER-type membrane. It has recently been demonstrated that, if the proton
ATPase in unable to acidify compartments, GFP tagged Rubisco and other chloroplast
proteins can be detected in autophagy bodies in the vacuole apart from the plastid (Ishida
et al., 2008). Moreover, an autophagy protein (ATG8e) colocalizes with stroma filled
protuberances from the chloroplast leading to a model of quadruple membrane bound
autophagosomes being delivered from the chloroplast to the vacuole for degradation of
stromal proteins (Ishida et al., 2008). The PLAM proteome sample does contain several
proton ATPase subunits needed for acidification (Table 7-Sl). Several proteins assigned
to the PLAM fraction are known stromal residents, most notably ribosomal proteins,
ferritins and the large subunit of Rubisco which is encoded on the plastid genome as well
as a transaldolase. It has been suggested previously that PLAMS represent the contact

sites for lipid transfer between the ER and the chloroplasts (Benning et al., 2006;
229

Andersson et al., 2007) but neither the localization of GFP fusion proteins nor the
proteome analysis confirmed the localization of lipid metabolism enzymes or transport
proteins specifically in the PLAMS.

We posit that PLAMS (Andersson et al., 2007) represent budding autophagosomes
as proposed by (Ishida et al., 2008) based on the following evidence: (a) PLAMS contain
a large number of stromal soluble proteins as expected for autophagosomes (b) the
protein composition of PLAMS is a hybrid of the endomembrane system and the
chloroplast as proposed in the model for autophagosome formation (c) increased protein
load of outer envelope but not inner envelope proteins induces the formation of stromules
and vesicles (d) proteomics of PLAMS indicate a high proportion of outer envelope
proteins in the sample.

This hypothesis is currently being tested by colocalization experiments of
stromules and vesicles formed after prolonged expression of outer envelope proteins with
markers for autophagosomes such as monodansylcadaverine (Contento et al., 2005),
lysotracker (Bassham, 2007) and ATG8, an autophagosomal marker protein (Contento et
al., 2005). In addition, co-localization studies of different combinations of proteins which
label proteins observed in early or late vesicles are designed to understand whether these
vesicles are of the same type. Since cells with excess amounts of stromules and vesicles
can now be produced efficiently through the prolonged overexpression of defined
proteins, electron microscopic reconstructions are possible. With electron microscopy,

the membrane arrangement within the vesicles may possibly be resolved.

230

Acknowledgements

We would like to thank the Michigan State University Research Technology Support
Facility for expert proteome analysis.

Supplemental Material

Supplemental Table 7-1: All proteins which were identified in the proteome experiments
Supplemental Table 7-2: Proteins of the outer envelope fractions

Supplemental Table 7-3: Proteins of the plastid associated membrane fractions

23]

References

Adl SM, Simpson AGB, Farmer MA, Andersen RA, Anderson OR, Barta JR, Bowser SS,
Brugerolle G, Fensome RA, Fredericq S, James TY, Karpov S, Kugrens P, Krug J, Lane
CE, Lewis LA, Lodge J, Lynn DH, Mann DG, McCourt RM, Mendoza L, Moestrup O,
Mozley-Standridge SE, Nerad TA, Shearer CA, Smimov AV, Spiegel F W, Taylor M
(2005) The new higher level classification of eukaryotes with emphasis on the taxonomy
of protists J Eukaryot Microbiol 52, 399-451

Andersson MX, Goksor M, Sandelius AS (2007) Optical manipulation reveals strong
attracting forces at membrane contact sites between endoplasmic reticulum and
chloroplasts J Biol Chem 282, 1170-1174

Awai K, Xu CC, Tamot B, Benning C (2006) A phosphatidic acid-binding protein of the
chloroplast inner envelope membrane involved in lipid trafficking Proc Natl Acad Sci U
S A103,10817-10822

Bae W, Lee YJ, Kim DH, Lee J, Kim S, Sohn EJ, Hwang I (2008) AKr2A-mediated
import of chloroplast outer membrane proteins is essential for chloroplast biogenesis Nat
Cell Biol 10, 220-U101

Bassham DC (2007) Plant autophagy-more than a starvation response Curr Opin Plant
Biol 10, 587-593

Benning C, Xu CC, Awai K (2006) Non-vesicular and vesicular lipid trafficking
involving plastids Curr Opin Plant Biol 9, 241-247

Brautigam A, Hofinann-Benning S, Weber APM (2008a) Comparative Proteomics of
Chloroplast Envelopes from C3 and C4 Plants Reveals Specific Adaptations of the
Plastid Envelope to C4 Photosynthesis and Candidate Proteins Required for Maintaining
C4 Metabolite Fluxes Plant Physiol 148, 568-579

Briiutigam A, Shrestha RP, Whitten D, Wilkerson CG, Carr KM, Froehlich JE, Weber
APM (2008b) Comparison of the use of a species-specific database generated by
pyrosequencing with databases from related species for proteome analysis of pea
chloroplast envelopes Journal of Biotechnology 136, 44-53

Cline K, Andrews J, Mersey B, Newcomb EH, Keegstra K (1981) Separation And
Characterization Of Inner And Outer Envelope Membranes Of Pea-Chloroplasts Proc
Natl Acad Sci USA 78, 3595-3599

Contento AL, Xiong Y, Bassham DC (2005) Visualization of autophagy in Arabidopsis

using the fluorescent dye monodansylcadaverine and a GFP-AtATG8e fusion protein
Plant J 42, 598-608

232

Curtis MD, Grossniklaus U (2003) A Gateway Cloning Vector Set for High-Throughput
Functional Analysis of Genes in Planta Plant Physiol 133, 462-469

Damari-Weissler H, Ginzburg A, Gidoni D, Mett A, Krassovskaya I, Weber APM,
Belausov E, Granot D (2007) Spinach SoHXKl is a mitochondria-associated hexokinase
Planta 226, 1053-1058

Douce R, Joyard J (1979) 'Isolation and Properties of the Envelope of Spinach
Chloroplasts.‘ (Ellis Horwood)

Eastmond PJ (2007) MONODEHYROASCORBATE REDUCTASE4 is required for
seed storage oil hydrolysis and postgerrninative growth in Arabidopsis Plant Cell 19,
1376-1387

Emanuelsson O, Nielsen H, Brunak S, von Heijne G (2000) Predicting subcellular
localization of proteins based on their N-terminal amino acid sequence J Mol Biol 300,
1004-16

Emanuelsson O, Nielsen H, von Heijne G (1999) ChloroP, a neural network-based
method for predicting chloroplast transit peptides and their cleavage sites Protein Sci 8,
978-84

Ferro M, Salvi D, Brugiere S, Miras S, Kowalski S, Louwagie M, Garin J, Joyard J,
Rolland N (2003) Proteomics of the chloroplast envelope membranes from Arabidopsis
thaliana Mol Cell Proteomics 2, 325-345

F ourrier N, Bedard J, Lopez-Juez E, Barbrook A, Bowyer J, Jarvis P, Warren G, Thorlby
G (2008) A role for SENSITIVE TO FREEZING2 in protecting chloroplasts against
freeze-induced damage in Arabidopsis Plant J 55, 734-745

Froehlich JE, Wilkerson CG, Ray WK, McAndrew RS, Osteryoung KW, Gage DA,
Phinney BS (2003) Proteomic study of the Arabidopsis thaliana chloroplastic envelope
membrane utilizing alternatives to traditional two-dimensional electrophoresis J
Proteome Res 2, 413-25

Gray J C, Sullivan JA, Hibberd JM, Hansen MR (2001) Stromules: Mobile protrusions
and interconnections between plastids Plant Biology 3, 223-233

Hanson MR, Sattarzadeh A (2008) Dynamic morphology of plastids and stromules in
angiosperm plants Plant Cell Environ 31, 646-657

Hatta T, Bullivan.S, Matthews RE (1973) Fine-Structure Of Vesicles Induced In

Chloroplasts Of Chinese Cabbage Leaves By Infection With Turnip Yellow Mosaic-
Virus Journal Of General Virology 20, 37-50

233

Inoue K, Potter D (2004) The chloroplastic protein translocation channel Toc75 and its
paralog OEP80 represent two distinct protein families and are targeted to the
chloroplastic outer envelope by different mechanisms Plant J 39, 354-365

Ishida H, Yoshimoto K, Izumi M, Reisen D, Yano Y, Makino A, Ohsumi Y, Hanson MR,
Mae T (2008) Mobilization of Rubisco and Stroma-Localized Fluorescent Proteins of

Chloroplasts to the Vacuole by an ATG Gene-Dependent Autophagic Process Plant
Physiol 148, 142-155

Jarvis P, Robinson C (2004) Mechanisms of protein import and routing in chloroplasts
Curr Biol 14, R1064-R1077

Keegstra K, Yousif AE (1986) Isolation And Characterization Of Chloroplast Envelope
Membranes Meth Enzymol 118, 316-325

Kessler F, Schnell DJ (2006) The function and diversity of plastid protein import
pathways: A multilane GTPase highway into plastids Traffic 7, 248-257

Lee YJ, Kim DH, Kim YW, Hwang l (2001) Identification of a signal that distinguishes
between the chloroplast outer envelope membrane and the endomembrane system in vivo
Plant Cell 13, 2175-2190

Li M, Schnell DJ (2006) Reconstitution of protein targeting to the inner envelope
membrane of chloroplasts J Cell Biol 175, 249-259

Lisenbee CS, Lingard MJ, Trelease RN (2005) Arabidopsis peroxisomes possess
functionally redundant membrane and matrix isoforms of monodehydroascorbate
reductase Plant J 43, 900-914

Lu BB, Xu CC, Awai K, Jones AD, Benning C (2007) A small ATPase protein of
Arabidopsis, TGD3, involved in chloroplast lipid import J Biol Chem 282, 35945-35953

Nanjo Y, Oka H, Ikarashi N, Kaneko K, Kitajima A, Mitsui T, Munoz FJ, Rodriguez-
Lopez M, Baroja-Fernandez E, Pozueta-Romero J (2006) Rice plastidial N-glycosylated
nucleotide pyrophosphatase/phosphodiesterase is transported from the ER-Golgi to the
chloroplast through the secretory pathway Plant Cell 18, 25 82-2592

Pohlmeyer K, Soll J, Steinkamp T, Hinnah S, Wagner R (1997) Isolation and
characterization of an amino acid-selective channel protein present in the chloroplastic
outer envelope membrane Proc Natl Acad Sci U S A 94, 9504-9509

Prod'homme D, Jakubiec A, Tournier V, Drugeon G, Jupin l (2003) Targeting of the

Turnip Yellow Mosaic Virus 66K replication protein to the chloroplast envelope is
mediated by the 140K protein Journal Of Virology 77, 9124-9135

234

Schleiff E, Eichacker LA, Eckart K, Becker T, Mirus O, Stahl T, $011 I (2003) Prediction
of the plant beta-barrel proteome: A case study of the chloroplast outer envelope Protein
Sci 12, 748-759

Schnurr JA, Shockey J M, de Boer G-J, Browse JA (2002) Fatty Acid Export from the
Chloroplast. Molecular Characterization of a Major Plastidial Acyl-Coenzyme A
Synthetase from Arabidopsis Plant Physiol 129, 1700-1709

Shevchenko A, Wilm M, Vorm 0, Mann M (1996) Mass Spectrometric Sequencing of
Proteins from Silver-Stained Polyacrylamide Gels Anal Chem 68, 850-858

Soll J, Schleiff E (2004) Protein import into chloroplasts Nature reviews Molecular cell
biology 5, 198-208

Tripp J, Inoue K, Keegstra K, F roehlich J E (2007) A novel serine/proline-rich domain in
combination with a transmembrane domain is required for the insertion of AtTic40 into
the inner envelope membrane of chloroplasts Plant J 52, 824-838

Villarejo A, Buren S, Larsson S, Dejardin A, Monne M, Rudhe C, Karlsson J, Jansson S,
Lerouge P, Rolland N, von Heijne G, Grebe M, Bako L, Samuelsson G (2005) Evidence
for a protein transported through the secretory pathway en route to the higher plant
chloroplast Nat Cell Biol 7, 1224

Weber APM (2004) Solute transporters as connecting elements between cytosol and
plastid stroma Curr Opin Plant Biol 7, 247-253

Weber APM, Fischer K (2007) Making the connections - The crucial role of metabolite
transporters at the interface between chloroplast and cytosol FEBS Lett 581, 2215-2222

Weber APM, Schwacke R, F lugge UI (2005) Solute transporters of the plastid envelope
membrane Ann Rev Plant Biol 56, 133-164

Wemerwashbume M, Cline K, Keegstra K (1983) Analysis Of Pea Chloroplast Inner
And Outer Envelope Membrane-Proteins By Two-Dimensional Gel-Electrophoresis And
Their Comparison With Stromal Proteins Plant Physiol 73, 569-575

Xu CC, Fan J, Froehlich J E, Awai K, Benning C (2005) Mutation of the TGDl
chloroplast envelope protein affects phosphatidate metabolism in Arabidopsis Plant Cell
17, 3094-3110

Xu CC, Fan JL, Cornish AJ, Benning C (2008) Lipid trafficking between the

endoplasmic reticulum and the plastid in Arabidopsis requires the extraplastidic TGD4
protein Plant Cell 20, 2190-2204

235

Zybailov B, Rutschow H, Friso G, Rudella A, Emanuelsson 0, Sun Q, van Wijk KJ
(2008) Sorting Signals, N-Terminal Modifications and Abundance of the Chloroplast
Proteome PLoS ONE 3, e1994

236

 

 

Chapter 8

Conclusions and Future Perspectives

237

 

The long term goal of studying plastids is to fully understand the plastid localized
biological processes as well their connection to the remainder of the cell. Plastids are
metabolic factories and provide the cell with many precursor molecules for maintenance,
growth and differentiation. They are the site of starch synthesis in higher plants, they
produce most of the fatty acids in plant cells, and they are the source of branched chain
and aromatic amino acids for protein biosynthesis. Human nutrition critically depends on
plants for carbohydrates, proteins and fatty acids as well as vitamins. Either plants serve
directly as food or are fed to livestock which is in turn consumed by humankind. The
increasing world population and the decreasing area of arable land pose unique
challenges to feeding the world and feeding the world’s hunger for non-fossil energy.
Understanding the metabolite factory of the plant cell, the plastid, may enable us to
engineer plants with increased yield for both food and biofuel production or increased
nutritional value. Although the metabolic pathways residing in plastids are fairly well
understood, there are major limitations in understanding the connection of plastids to the
remainder of the cell. We decided to gain insight into plastid biology and differentiation
by focusing proteome studies on the envelope which separates the soluble stroma from
the remainder of the cell and serves as the interface between the stroma and the cytosol.
Initial comprehensive proteome studies of plastids have been focused mainly on green
photosynthetic chloroplasts. The internal membrane system which contains the electron
transfer chain, the thylakoids, the stroma and also the envelopes of chloroplasts have
been studied in detail by proteomics (Ferro et al., 2000; Peltier et al., 2000; Ferro et al.,

2002; Peltier et al., 2002; Ferro et al., 2003; Froehlich et al., 2003; Baginsky and
238

Gruissem, 2004; F riso et al., 2004; Kleffmann et al., 2004; Sun et al., 2004; Baginsky et
al., 2005; Zybailov et al., 2008). Studies of other plastid subtypes have been focused on
whole plastids (Baginsky et al., 2004; von Zychlinski et al., 2005; Siddique et al., 2006;
Kleffmann et al., 2007) which limits the discovery of proteins integral to or associated
with the envelope membranes. The goal of my work was to explore new model systems
to analyze differentiated plastid types in comparison with classical chloroplasts and to
provide detailed insights into envelope biology by using comparative instead of
qualitative proteome analysis.

In chapter 3, I report on using an organism without a sequenced genome for
proteome analysis. In cases like this, the proteome analysis has been based on sequence
information from related species (Baginsky et al., 2004). This new analysis identifies the
limitations of this approach for proteomics which are especially severe for proteins of
low abundance. This paper not only reports the limitations but also demonstrates a
possible strategy to overcome them, the generation of a suitable database by
pyrosequencing. Alternative strategies to increase the detection power on non-species
specific databases are shown to lead to increased numbers of false positives and thus are
not suitable. The generation of new model organism for proteomics beyond the plants
with sequenced genomes is important because most of the models were chosen either
since they are crop plants or because they are amenable to genetics but never because
they are suitable for biochemistry. For the new model Pisum sativum, protocols for
isolation of both leaf chloroplasts and root amyloplasts (Borchert et al., 1989; Borchert et
al., 1993) are in place. For pea chloroplasts, protocols to subfractionate the chloroplast

envelope into the inner and outer envelope have been described (Cline et al., 1981).

239

Mitochondria can also be efficiently purified from pea roots, stems and leaves
(Bourguignon et al., 1992; Zancani et al., 1995). Pea plants also allow the purification of
endomembrane system components in sufficient quantity for biochemical analysis
(Hellgren et al., 1993). The new database enables comparative proteomics of organelles
in different tissues of the same plant. The pea chloroplast envelope proteome dataset
containing at least semi-quantitative information also serves as a template to which other
plastid types can be compared.

In chapter 4, the first analysis of the envelope proteome of a non-green plastid is
reported. The primary role of proplastids was considered to act as a reservoir for plastid
development rather than being metabolically active themselves (Wise, 2006). The
analysis of proplastid-like plastids in tobacco cell cultures had already revealed highly
active plastids which contain many proteins for amino acid synthesis (Baginsky et al.,
2004) but meristematic proplastids were never analyzed in vivo since meristematic tissue
has been limited prior to the introduction of cauliflower as the tissue source. The analysis
of the envelope proteome revealed a distinct set of envelope proteins indicating
meristematic proplastids are highly active cell factories for amino acids and nucleotides.
The presence of a reduction equivalent shuttle at the envelope leads to a reconsideration
of the role of glucose-6-phosphate import and the two translocator model could be
adapted to proplastids. In addition to candidates for uncharacterized fluxes, the proteome
analysis also revealed the adaptation of the protein import complex in proplastids and a
unique subset of thylakoid proteins. The successful analysis of meristematic plastids in a
Brassicaceae species opens the possibility to study chromoplasts, proplastids and

chloroplasts in the same model plant.

240

 

In chapter 5, I report on changes in protein abundance in different chloroplast
types rather than on differences in the protein pattern as reported for the comparison of
chloroplasts and proplastids. The analysis reveals that increasing the transport protein
amount in the inner envelope is one of the major adaptations for increasing the metabolite
flow across a membrane to accommodate a biochemical pathway like C4 photosynthesis.
Surprisingly, the respective outer envelope porins were also increased. This pattern has
implication both for engineering efforts of pathways and for C4 photosynthesis research.
Although the plastid envelope has evolved to handle the metabolic flow required for its
respective role and thus exerts no control over the flow through the pathway, major
changes in metabolite flow requires evolutionary adjustments at the plastid envelope.
Bioengineering of bulk flow in biochemical pathways thus may require not only the
presence of soluble enzymes but also changes in transport protein content if different
compartments are involved. Since at least the known transport proteins involved in C4
photosynthesis accommodate the increased metabolite flow by increased abundance, new
transport protein candidates for defined flows may be identified based on this feature.
The analysis demonstrates the power of comparative ‘omics’ approaches which are
suitable to identify new components involved in pathways with high metabolite flow
without introducing bias into the experiments by predetermined expectations.

Chapter 6 reports the analysis of a transport protein candidate.

In chapter 7d, the newly generated pea sequence database was used to study
chloroplast envelopes in more detail. Since the envelope contains only a small fraction of
the cellular protein, analysis by proteomics was limited with small model species like

Arabidopsis. Using pea allowed the separation and analysis of inner and outer envelope

241

membranes as well as plastid associated membranes and a light microsomal fraction. The
outer envelope has received little attention compared to the inner envelope membrane of
plastids although it represents the ultimate barrier to the cytosol and therefore also the
interface to the remainder of the cell. At the onset of this work, little was known about
the dynamics of envelope membranes other than the fact that stromules, stroma filled
protuberances, can extend from the envelope in some cell types. The recent discovery of
a close association with the endomembrane system (plastid associated membranes or
PLAMS) prompted a comparative proteome analysis with the goal to define the new
compartment PLAM and to discover new proteins of the envelope as well as localize
known proteins to either the outer or the inner envelope. The comparative proteome data
already revealed considerable protein dynamics between the membrane systems and
marked the PLAMS as a hybrid compartment of the ER and the chloroplast with a large
contribution of outer envelope proteins. The dynamic localization patterns were analyzed
in vivo by protein GFP fusions transiently expressed in tobacco cells. The GFP fusion
proteins confirmed the presence of chloroplast associated vesicles and three proteins, two
membrane bound and one soluble, are primarily associated with these vesicles. The
analysis of known and putative outer envelope proteins revealed new fluorescence
patterns based on the protein load on the system. After prolonged expression, many of the
outer envelope protein labeled the envelope, stromules and vesicles, a behavior which has
never been observed with either stoma or inner envelope targeted proteins. Considering
the OF P data in context of the proteome data lead to a new testable hypothesis about
PLAMs. They are budding autophagosomes and their formation can be induced by

overexpression of both PLAM and outer envelope proteins. The identification of both an

242

inducible system and of resident proteins allows testing several hypothesis about PLAMS.
Their colocalization with autophagy markers such as monodansylcadaverine and
AtATG8e is being tested. We will also start analyzing the budding of vesicles with co-
localization experiments to test whether all proteins that label vesicular structures indeed
colocalize and to determine a time-resolved pattern of protein localization during the
formation of a vesicle. This new approach will yield insights into protein transport out of
the plastid and highlight the dynamic nature of the envelope membranes.

In summary, this work has highlighted dynamic features of the envelope and
introduced novel models for the study of plastids by proteomics. The plastid envelope
turned out to be dynamic in regard to the protein composition; the proplastid envelope
has a qualitatively different pattern compared to the chloroplast envelope. When
comparing two chloroplast types, the differences were based on protein abundance rather
than protein presence and therefore only revealed by semi-quantitative comparison. In all
cases, the envelopes reflected the metabolic roles of the respective plastids. The
envelopes of chloroplasts themselves are also dynamic. Traditionally considered static
with regard to membrane architecture of the envelope membranes, both proteomics and
GFP fusion protein analysis reveal vesicles with distinct composition and varying

localization patterns of proteins based on their protein load in the system.

243

References

Baginsky S, Gruissem W (2004) Chloroplast proteomics: potentials and challenges J Exp
Bot 55, 1213-1220

Baginsky S, Kleffmann T, von Zychlinski A, Gruissem W (2005) Analysis of shotgun
proteomics and RNA profiling data from Arabidopsis thaliana chloroplasts J Proteome
Res 4, 63 7-640

Baginsky S, Siddique A, Gruissem W (2004) Proteome analysis of tobacco bright yellow-
2 (BY-2) cell culture plastids as a model for undifferentiated heterotrophic plastids J
Proteome Res 3, 1128-1137

Borchert S, Grosse H, Heldt HW (1989) Specific Transport Of Inorganic-Phosphate,
Glucose-6-Phosphate, Dihydroxyacetone Phosphate And 3-Phosphoglycerate Into
Amyloplasts From Pea Roots FEBS Lett 253, 183-186

Borchert S, Harborth J, Schunemann D, Hoferichter P, Heldt HW (1993) Studies Of The
Enzymatic Capacities And Transport-Properties Of Pea Root Plastids Plant Physiol 101,
303-312

Bourguignon J, Macherel D, Neuburger M, Douce R (1992) Isolation, Characterization,
And Sequence-Analysis Of A Cdna Clone Encoding L-Protein, The Dihydrolipoamide
Dehydrogenase Component Of The Glycine Cleavage System From Pea-Leaf
Mitochondria Eur J Biochem 204, 865-873

Cline K, Andrews J, Mersey B, Newcomb EH, Keegstra K (1981) Separation And
Characterization Of Inner And Outer Envelope Membranes Of Pea-Chloroplasts Proc
Natl Acad Sci USA 78, 3595-3599

Ferro M, Salvi D, Brugiere S, Miras S, Kowalski S, Louwagie M, Garin J, Joyard J,
Rolland N (2003) Proteomics of the chloroplast envelope membranes from Arabidopsis
thaliana Mol Cell Proteomics 2, 325-345

Ferro M, Salvi D, Riviere-Rolland H, Vermat T, Seigneurin-Bemy D, Grunwald D, Garin
J, J oyard J, Rolland N (2002) Integral membrane proteins of the chloroplast envelope:
identification and subcellular localization of new transporters Proc Natl Acad Sci U S A
99, 11487-92.

F erro M, Seigneurin-Bemy D, Rolland N, Chapel A, Salvi D, Garin J, Joyard J (2000)

Organic solvent extraction as a versatile procedure to identify hydrophobic chloroplast
membrane proteins Electrophoresis 21, 3517-26.

244

Friso G, Giacomelli L, Ytterberg AJ, Peltier JB, Rudella A, Sun Q, van Wijk KJ (2004)
In-depth analysis of the thylakoid membrane proteome of Arabidopsis thaliana

chloroplasts: New proteins, new functions, and a plastid proteome database Plant Cell 16,
478-499

Froehlich J E, Wilkerson CG, Ray WK, McAndrew RS, Osteryoung KW, Gage DA,
Phinney BS (2003) Proteomic study of the Arabidopsis thaliana chloroplastic envelope

membrane utilizing alternatives to traditional two-dimensional electrophoresis J
Proteome Res 2, 413-25

Hellgren L, Morre DJ, Sellden G, Sandelius AS (1993) Isolation Of A Putative Vesicular
Intermediate In The Cell-Free Transfer Of Membrane From Transitional Endoplasmic-
Reticulum To The Golgi-Apparatus Of Etiolated Seedlings Of Garden Pea J Exp Bot 44,
197-205

Kleffmann T, Russenberger D, von Zychlinski A, Christopher W, Sjolander K, Gruissem
W, Baginsky S (2004) The Arabidopsis thaliana chloroplast proteome reveals pathway
abundance and novel protein functions Curr Biol 14, 354-362

Kleffmann T, von Zychlinski A, Russenberger D, Hirsch-Hoffmann M, Gehrig P,
Gruissem W, Baginsky S (2007) Proteome dynamics during plastid differentiation in rice
Plant Physiol 143, 912-923

Peltier J B, Emanuelsson O, Kalume DE, Ytterberg J, Friso G, Rudella A, Liberles DA,
Soderberg L, Roepstorff P, von Heijne G, van Wijk KJ (2002) Central functions of the
lumenal and peripheral thylakoid proteome of Arabidopsis determined by
experimentation and genome-wide prediction Plant Cell 14, 211-36.

Peltier JB, Friso G, Kalume DE, Roepstorff P, Nilsson F, Adamska I, van Wijk KJ (2000)
Proteomics of the chloroplast: systematic identification and targeting analysis of lumenal
and peripheral thylakoid proteins Plant Cell 12, 319-41.

Siddique MA, Grossmann J, Gruissem W, Baginsky S (2006) Proteome analysis of bell
pepper (Capsicum annum L.) chromoplasts
Plant Cell Physiol, 33-39

Sun Q, Emanuelsson O, van Wijk KJ (2004) Analysis of curated and predicted plastid
subproteomes of Arabidopsis. Subcellular compartmentalization leads to distinctive
proteome properties Plant Physiol 135, 723-734

von Zychlinski A, Kleffmann T, Krishnamurthy N, Sjolander K, Baginsky S, Gruissem
W (2005) Proteome analysis of the rice etioplast - Metabolic and regulatory networks and
novel protein functions Mol Cell Proteomics 4, 1072-1084

Wise R (2006) 'The diversity of plastid form and function.‘ (Springer: Amsterdam)

245

Zancani M, Macri F, Peruffo ADB, Vianello A (1995) Isolation Of The Catalytic Subunit
Of A Membrane-Bound H+-Pyrophosphatase From Pea Stem Mitochondria Eur J
Biochem 228, 138-143

Zybailov B, Rutschow H, Friso G, Rudella A, Emanuelsson 0, Sun Q, van Wijk KJ
(2008) Sorting Signals, N-Terrninal Modifications and Abundance of the Chloroplast
Proteome PLoS ONE 3, e1994

246

Appendix
High-throughput colorimetric method for the parallel assay of glyoxylic acid and

ammonium in a single extract
This chapter has been previously published in Analytical Biochemistry: Bréiutigam A*,
Gagneul D* and Weber APM (2007) High-throughput colorimetric method for the
parallel assay of glyoxylic acid and ammonium in a single extract Anal Biochem 362,

151-3

* These authors contributed equally to the work.

I would like to acknowledge Dr. David Gagneul for collaboration on developing the

extract and contributing figure ZB

247

Introduction

Glyoxylate and ammonia are signature metabolites of a major plant metabolic
pathway, the C2 oxidative carbon cycle, also known as photorespiration. This complex
salvage pathway recycles most of the carbon that is lost from the Calvin cycle in the form
of phosphoglycolate as a consequence of the oxygenation reaction of RubisCO and it
constitutes a significant metabolic flux in photosynthesizing leaves of C3 plants [1].
Mutations in enzymes of this pathway are usually lethal, underlining the importance of
this salvage pathway [2-4]. Although all enzymes and some of the metabolite transporters
involved in this highly-compartmentalized pathway have been identified [5-9],
information about the majority of the transport proteins and the processes regulating the
pathway is still missing [1, 4], possibly because mutations in the corresponding genes
cause subtle metabolic phenotypes. Thus, a high throughput method to quantify
metabolites known to accumulate in photorespiratory mutants, glyoxylate and
ammonium, is desirable. Glyoxylate is also a pathway intermediate of the glyoxylate
cycle that is involved in making carbon moieties from fatty acids available to
carbohydrate biosynthesis by gluconeogenesis. This pathway is important during seed
germination and seedling establishment of oil seed plants and possibly during senescence
but does not play a significant role in photosynthesizing leaves. While primary nitrogen
assimilation by nitrate and nitrite reductase and the phenylpropanoid pathway are also
sources of ammonium in planta [10], photorespiration contributes at least an order of
magnitude more ammonium than these pathways [11, 12], making it a suitable marker for

the photorespiratory pathway.

248

Here we present improved versions of two colorimetric methods for the
quantification of ammonia and glyoxylate that makes these procedures suitable for high

throughput and sensitive quantification of these signature metabolites in plant tissues.

 

 

2.0 5
ii

1.5 -§;
6 i: a
‘c’ 1'
m E: A C
e 10 « -~
0
m
.D
<

0.5 -

    
 

 

 

 

0 a" - .. .7..°.':-.".".'. . . . . .
300 400 500 600 700 800 900
Wavelength (nm)

Figure 1: Absorption spectra of a crude acidic extract of Arabidopsis leaves (a), of a partially

purified acidic extract of Arabidopsis leaves (b), of the formazan of glyoxylic acidic (c), and of
the indophenol derivative of ammonia (d) as described in the text.

Gas chromatography is not suitable for the determination of glyoxylate,
presumably because it is unstable and non-enzymatically decomposes when coming into
contact with oxygen. Glyoxylate has been measured using a colorimetric method [5, 13,

14]. In this method, a l,5-diphenylformazancarboxylic acid derivative of glyoxylic acid is

249

produced through mild oxidation of a glyoxylic acid hydrazone in the presence of excess
phenylhydrazine. The formazan of glyoxylic acid shows maximal absorption at a
wavelength of 520 nm (Figure 1). The quantification of ammonium (based on the
Berthelot reaction [15, 16]) from the same plant extract as glyoxylate has been described
previously in tobacco [5]. Ammonia reacts with hypochlorite and forms a
monochlorarnine, which reacts with phenol to an indophenol derivative that has deep blue
colour and displays maximal absorbance at 620 nm (Figure 1). Although both methods
have been used previously to determine glyoxylate and ammonia contents in plant tissues
[5, 17], it is hampered by a number of interfering compounds that quench the signal. In
addition, many plant tissues accumulate anthocyanines, especially under stress
conditions, which absorb light at similar wavelengths and therefore introduce noise to the
assays. To address these problems and to render the assay suitable for high-throughput
analysis, we have developed a new protocol optimized for use in a 96-well format that
allows the colorimetric quantification of glyoxylate and ammonia from a single plant

extract.

Method

Plant tissue is harvested and snap-frozen in liquid nitrogen, and ground to a fine
powder under liquid nitrogen. The modified extraction procedure is as follows: 100 mg of
powder is extracted in 1 ml of 100 mM HCl, and 500 pl of chloroform is subsequently
added. The samples are rotated for 15 minutes at 4°C and the phases are separated by
centrifugation (12,000xg, 10 minutes, 8°C). The aqueous phase is transferred to a fresh

tube containing 50 mg of acid washed activated charcoal, thoroughly mixed, and
250

 

 

centrifuged again (20,000xg, 5 minutes, 8°C). For glyoxylate quantification, 450 pl of the
supernatant is mixed with 50 ul of a freshly prepared 1% (v/v) solution of
phenylhydrazine in 100 mM HCl and the sample is incubated at 95°C for two minutes
and cooled on ice for 6 minutes. The sample is acidified by adding 250 pl concentrated
HCl and divided into three aliquots of 225 pl each in cavities of a 96 well plate: Two
technical replicates are oxidized with 25 ul of a solution of 1.6% (w/v) K3Fe(CN)6 in
water and one aliquot serves as control with 25 ul water added. Colour formation is
measured exactly 6 minutes after addition of K3Fe(CN)6 solution since the reaction
reaches a plateau after 4 minutes and the coloured reaction product slowly decomposes
with time afterwards.

For ammonia quantification the supernatant obtained after charcoal treatment is
diluted 1:1 (v/v) in 100 mM HCI..20 ul of this solution is mixed with 100 pl of a 1%
(w/v) phenol 0.005% (w/v) sodium-nitroprusside solution in water (“reagent 1”), and
100 pl of a 1% (v/v) sodium-hypochlorite 0.5% (w/v) sodium hydroxide solution in water
(“reagent II”) are added. The samples are incubated at 37°C for 30 minutes and light

absorption at 620 nm is measured.

Results and Discussion

To develop the glyoxylate assay we recapitulated the original procedure of
Dekker and Maitra [13] with varying concentrations of sodium glyoxylate in water. 10 [.11
sodium glyoxylate solution was added to 500 pl 0.1% phenylhydrazine in 100 mM HCl
as previously described, heated, cooled on ice and acidified with concentrated HCl and

oxidized to observe colour formation. We then tested, the parameters of the analysis to
251

determine the optimal incubation temperature (95°C), incubation time (2 minutes),
cooling time (6 minutes) and time between addition of oxidizing agent and measurement
(6 minutes) that still allowed the processing of 32 samples per experiment. Changing
these parameters leads to changed absorption values but still produces linear standard

curves. The standard curve was linear from 0.25-200 uM sodium glyoxylate.

 

 

 

 

 

A 1
C
0.8 -
a
g 0.6 - b
O
N
2?.
§ 04 -
'8'-
O
3
< 0.2
0 7 l I I
0 10 20 30 40

pM glyoxylate in sample

Figure 2: Comparison of standard curves for glyoxylate (A) and ammonia (B) determination. The
external standard curve was prepared in 100 mM HCl (a). The internal standard curves were prepared in
crude acidic extract of Arabidopsis leaves and were either measured (b) or partially purified and measured
(c). The glyoxylate and ammonia contents were determined as described in the text. For all curves,
R(squared) > 0.99.

252

Figure 2 (cont’d)

 

 

 

 

B 1
0.9«
0.3-

,5: 0.7l

g o...

.5 °-5

§ 0.4 -

E 0.3-

<
02-
0.1

o

 

NH“+ (nmoleslassay)

We then tested the suitability of the Dekker and Maitra method [13] for plant
extracts. Plant tissue was extracted in 0.1% phenylhydrazine in 100 mM HCl and spiked
with varying concentrations of sodium glyoxylate. The debris in spiked extracts was
precipitated by centrifugation and the supernatant was boiled, cooled, and acidified, as
described above. Upon acidification the remaining soluble proteins denatured and turned
the sample turbid necessitating an additional centrifugation to clear the extract. One
aliquot served as blank and two aliquots were oxidized as described above. When this
internal standard curve was compared to the external standard curve prepared without the
addition of plant extract, we observed a marked difference in slope between both curves.
This indicates the presence of one or more compounds in the plant extract that quench the
signal (Figure 2A). Plant tissues also frequently contain varying amounts of
anthocyanines, which might obscure the specific signal because they absorb light at a

253

similar wavelength as the formazan (Figure 1). To remove lipophilic compounds such as
chlorophylls and proteins from the extract we added chloroform to powdered plant tissue
extracted with 100 mM HCl (derived from [18]). Although after centrifugation proteins
and chlorophylls were quantitatively removed from the aqueous phase, the slope of the
resulting standard curve remained identical to the initial experiment. We therefore tested
several commercially available resins for removal of the interfering compounds.
Amberlite XAD7 quantitatively removed anthocyanines from the extract as described in
[19] without absorbing glyoxylic acid, but the interfering compound or compounds
remained, as judged from the slopes of internal and external standard calibration curves.
Finally, we tested whether acid washed activated charcoal would absorb the interfering
compound without impacting the recovery of glyoxylic acid. In 100 mM HCI, charcoal
did not absorb glyoxylic acid but efficiently absorbs either phenylhydrazine or the
reaction product of glyoxylic acid and phenylhydrazine, glyoxylic acid hydrazone.
Therefore the extraction of glyoxylic acid and the conjugation to phenylhydrazine had to
be temporally separated by crude extraction in 100 mM HCl, extract purification, and
subsequent reaction of the glyoxylic acid with phenylhydrazine.

When plant tissue is extracted with 100 mM HCl and chloroform and purified
with activated charcoal, both anthocyanines and the unknown interfering compounds are
efficiently removed without absorbing glyoxylic acid since the slope of the internal
standard curve fits perfectly to the slope of the external standard curve (Figure 2A). The
shift between both curves represents the glyoxylic acid content in the plant tissue.

To develop the ammonium assay for plant tissue we first tested the Berthelot

method with standard solutions of varying concentrations of ammonium in 100 mM HCl.

254

The method of Weatherburn [15] was adapted to the 96 well format. 100ul reagent I was
mixed with 20 [.11 sample and 100 pl reagent II. The samples were incubated at 37°C for
30 minutes and measured at 620 nm. The resulting standard curve was linear in the range
0.01-3 mM ammonium. Next we extracted finely ground plant tissue in 100 mM HCl and
spiked with varying concentrations of ammonium. Although the internal standard curve
was linear in the range 0.01-1 mM ammonium, its slope was markedly different from the
external standard curve, indicating the presence of one or more interfering compounds
(Figure ZB). To remove the unknown interfering compounds, plant tissue was extracted
with 100 mM HCl, spiked with varying concentrations of ammonia, and centrifuged at
14,000xg to remove cell debris. The supernatant was incubated with acid washed
activated charcoal, as decribed above. To 100 pl reagentl 20 ul purified diluted extract
was added followed by 100 pl reagent II. The slope of the resulting internal standard
curve fitted perfectly to the slope of the internal standard curve. Finally we ensured that
the additional purification step with chloroform needed for glyoxylate determination did
not interfere with the determination of ammonium (Figure 23). It has been reported
previously that the Berthelot method is not suitable for determining ammonia
concentrations in plant extracts [20]. Interfering compounds result in a difference in slope
of the internal standard curve compared to an external standard as shown in Figure 2. To
assess whether different amino acid concentrations in the extract interfere with the
quantification of ammonium, we doubled and quadrupled the extract concentration in the
essay. A twice-concentrated extract will not result in a changed slope whereas
quadrupling the extract concentration introduces sufficient interfering compounds to

change the slope. The protocol presented in [20] requires elaborate instrumentation

255

 

whereas the novel extraction and purification protocol presented here allows the reliable
and efficient quantification of ammonium with a spectrophotometer. However, if this
new protocol is to be adapted for different tissues or plant species, we recommend

optimizing the extract concentration to minimize interference as described above.

In summary, we have developed a novel purification protocol for acidic plant
extracts in which interfering compounds and co-absorbing compounds were
quantitatively removed by chloroform and charcoal extraction. This new method allows
the simultaneous determination of two signature metabolites of photorespiration,
glyoxylate and ammonium, from the same plant extract. The method has also been
adapted to the high-throughput 96-well format and can potentially be robotized, thus

making it suitable for mutant screens.
Acknowledgements

This work was supported by NSF-awards IOB-0548610, MCB-O6l8335, and MCB-

0519740.

256

References

1.

A.P.M. Weber, Synthesis, Export and Partitioning of the End Products of
Photosynthesis, in: R. R. Wise, J. K. Hoober (Eds.) The Structure and Function of
Plastids, Springer, Dordrecht, 2006, pp. 273-292.

C.R. Somerville, An early Arabidopsis demonstration. Resolving a few issues
concerning photorespiration, Plant Physiol. 125 (2001) 20-24.

A. Wingler, P.J. Lea, W.P. Quick, R.C. Leegood, Photorespiration: metabolic
pathways and their role in stress protection, Philos. Trans. R. Soc. Lond. Ser. B-
Biol. Sci. 355 (2000) 1517-1529.

S. Reumann, A.P.M. Weber, Plant peroxisomes respire in the light: Some gaps of
the photorespiratory C2 cycle have become filled - others remain, Biochim.
Biophys. Acta 1763 (2006) http://dx.doi.org/IO.1016/j.bbamcr.2006.09.008.

J. Schneidereit, R.E. Hausler, G. Fiene, W.M. Kaiser, A.P.M. Weber, Antisense
repression reveals a crucial role of the plastidic 2-oxoglutarate/malate translocator
DiTl at the interface between carbon and nitrogen metabolism, Plant J. 45 (2006)
206-224.

R. Boldt, C. Edner, U. Kolukisaoglu, M. Hagemann, W. Weckwerth, S.
Wienkoop, K. Morgenthal, H. Bauwe, D-glycerate 3-kinase, the last unknown

enzyme in the photorespiratory cycle in Arabidopsis, belongs to a novel kinase
family, Plant Cell 17 (2005) 2413-2420.

L.M. Voll, A. Jamai, P. Renne, H. Voll, C.R. McClung, A.P.M. Weber, The
photorespiratory Arabidopsis shml mutant is deficient in SHM], Plant Physiol.
140 (2006) 59-66.

P. Renne, U. DreBen, U. Hebbeker, D. Hille, U.I. Flugge, P. Westhoff, A.P.M.
Weber, The Arabidopsis mutant dot is deficient in the plastidic glutamate/malate
translocator DiT2, Plant J. 35 (2003) 316-331.

M. Linka, A.P.M. Weber, Shuffling ammonia between mitochondria and plastids
during photorespiration, Trends Plant Sci. 10 (2005) 461-465.

257

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

C. Fritz, N. Palacios-Rojas, R. Feil, M. Stitt, Regulation of secondary metabolism
by the carbon-nitrogen status in tobacco: nitrate inhibits large sectors of
phenylpropanoid metabolism, Plant J. 46 (2006) 533-548.

A.J. Keys, I.F. Bird, M.J. Cornelius, P.J. Lea, R.M. Wallsgrove, B.J. Miflin,
Photorespiratory Nitrogen Cycle, Nature 275 (1978) 741-743.

H. Bauwe, U. Kolukisaoglu, Genetic manipulation of glycine decarboxylation, J.
Exp. Bot. 54 (2003) 1523-1535.

E.E. Dekker, U. Maitra, Conversion of Gamma-Hydroxyglutamate to Glyoxylate
and Alanine - Purification and Properties of Enzyme System, J. Biol. Chem. 237
(1962) 2218-2227.

D.N. Kramer, N. Klein, R.A. Baselice, Quantitative Determination of Glyoxylic
Acid, Anal. Chem. 31 (1959) 250-252.

M.W. Weatherburn, Phenol-Hypochlorite Reaction for Determination of
Ammonia, Anal. Chem. 39 (1967) 971.

O]. Patton, S.R. Crouch, Spectrophotometric and Kinetics Investigation of
Berthelot Reaction for Determination of Ammonia, Anal. Chem. 49 (1977) 464-
469.

A. Wingler, P.J. Lea, R.C. Leegood, Photorespiratory metabolism of glyoxylate
and formate in glycine-accumulating mutants of barley and Amaranthus edulis,
Planta 207 (1999) 518-526.

W. Weckwerth, K. Wenzel, O. Fiehn, Process for the integrated extraction,
identification, and quantification of metabolites, proteins and RNA to reveal their
co-regulation in biochemical networks, Proteomics 4 (2004) 78-83.

K. Takeda, S. Tominaga, The Anthocyanin in Blue Flowers of Centaurea cyanus,
Bot. Mag. Tokyo 96 (1983) 359-363.

258

20. S. Husted, C.A. Hebbem, M. Mattsson, J .K. Schjoerring, A critical experimental
evaluation of methods for determination of NH4Jr in plant tissue, xylem sap and
apoplastic fluid, Physiol. Plant. 109 (2000) 167-179.

259

File List of Supplementary Data

Excel File containing two sheets as a supplemental file to chapter 3
Supplement to Chapter 3.xls 110KB

Supplemental table: Proteins identified with the three databases

Excel F ile containing three sheets as a supplemental file to chapter 4
Supplement to Chapter 4.xls 21 lKB

Supplementary table 1: Proteins identified in proplastid envelopes from cauliflower curd
tissue

Supplementary table 2: Proteins identified in chloroplast envelopes from pea leaves
(Brautigam et al., 2008a) were extracted from PRIDE and reanalyzed

Supplementary table 3: Thylakoid proteins identified in envelopes of proplastids and
chloroplasts

Excel File containing three sheets as a supplemental file to chapter 5
Supplement to Chapter 5.xls 224KB

Supplemental Table 5-S1. Proteins identified in C4 PCA-type maize chloroplast envelope
membranes. The table lists proteins identified, number of spectra mapping to each maize
accession number, annotation, classification, number of membrane spanning domains,
targeting prediction, and previous identifications in other proteomics studies.
Supplemental Table 5-82. Proteins identified in C3-type pea chloroplast envelope
membranes. The table lists proteins identified, number of spectra mapping to each maize
accession number, annotation, classification, number of membrane spanning domains,
targeting prediction, and previous identifications in other proteomics studies.
Supplemental Table 5-S3. Percentage of total spectral counts for each protein identified
in C4 PCA-type and C3-type chloroplasts of maize and pea, respectively.

Excel File containing three sheets as a supplemental file to chapter 7
Supplement to Chapter 7 217KB

Supplemental Table 7-1: All proteins which were identified in the proteome experiments
Supplemental Table 7-2: Proteins of the outer envelope fractions
Supplemental Table 7-3: Proteins of the plastid associated membrane fractions