gig“ A

w >
hL-M-

Lu

.3:

kiwi}? ‘

1‘ '2?!”

836%"

9.4m

4:3:
«0‘. g'

I n... rm

4
--

NW.
wrfl‘wr5; - .
. .m- A.

.1
A

.;

-
u

up»

w

..'.T'

 

.'

(E?
4.:
it

{Gui \‘u '_.
my? ,
W '

z’l‘ ' ‘
£sz

 

QOfiM
swat/010

This is to certify that the
dissertation entitled

FATTY ACID METABOLISM IN HIGHER PLANTS:
PROBING THE MECHANISM OF FATTY ACID TRANSPORT
ACROSS THE CHLOROPLAST ENVELOPE

presented by

Abraham Jeong-Kyu Koo

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Plant Biology

6235 mm,

Major Professor’s Signature

Date

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

- _ —.-,_. _-___ fl N‘mA— +.—_.__._ A

' LIERARV '
Michigan State

University

 

 

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDuo.965-p.15

FATTY ACID METABOLISM IN HIGHER PLANTS:
PROBING THE MECHANISM OF FATTY ACID TRANSPORT ACROSS THE
CHLOROPLAST ENVELOPE
By

Abraham J eong-Kyu Koo

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Plant Biology

2004

ABSTRACT
FATTY ACID METABOLISM IN HIGHER PLANTS:

PROBING THE MECHANISM OF FATTY ACID TRANSPORT ACROSS THE
CHLOROPLAST ENVELOPE

By
Abraham Jeong-Kyu Koo

Fatty acids (FA’s) play a variety of crucial roles in living organisms. Despite the
important roles in various aspects of life, FA transport remains one of the most poorly
understood areas of FA metabolism. The main objective of this thesis is to begin to
determine the mechanism(s) of FA transport in plants. Lipid metabolism within the cell
requires massive movements of acyl chains between intracellular organelles. One such
example is the export of FA from the plastid. De novo fatty acid synthesis (FAS) in
plants occurs almost exclusively in the plastids. The free fatty acid (FF A) released at the
final step of FAS by hydrolysis of acyl-acyl carrier protein (ACP) crosses the double
envelope membrane to be re-esterified to acyl-coenzyme A in the cytosol. The maximum
half life for FF A in this export pool in pea and spinach leaves was determined to be S l 5.
Kinetic labeling experiments with isolated pea chloroplasts indicated that the measured
long-chain acyl CoA synthesis (LACS) reaction using freely diffusing bulk FF A substrate
cannot account for the efficient LACS reaction that must occur at the very low in vivo
concentrations of FFA. The “nascent” FF A was also shown to be protected from binding
to extraplastidial bovine serum albumin before being converted to the long-chain acyl
CoA. These data discount a free diffusion model of FFA export and indicate a channeled

delivery of FFA to the LACS.

Plant cells are capable of incorporating, elongating and desaturating exogenously
provided short- and medium-chain FA’s. The existence of a pathway in the plastid for
the esterification of exogenous FFA to ACP has only been inferred from past studies. In
this thesis the plastid was demonstrated to be the major site for the elongation of the
exogenous FFA into C16 and C18 FA’s. An Arabidopsis T-DNA insertion mutant,

aae15, was isolated through a reverse genetics approach, and this mutant displayed as

. . . 14
much as 80% reduction In elongation of exogenous [1- C]laurate to C16 and C18 FA’s.

The gene (AAE15 (acyl activating enzyme15)) disrupted in the mutant is a homolog of
long-chain acyl-CoA synthetases, is represented by 11 expressed sequence tags (EST’s)
and contains a plastid targeting sequence. The findings indicate the existence of a novel
pathway of FFA metabolism in the plastid.

It is likely that specific membrane proteins are involved in lipid traffic across the
envelope. Despite the importance and complexity of the plastid envelope, only a small
fraction of integral envelope proteins have been purified or characterized. Through a
bioinformatics strategy 541 proteins were selected from the Arabidopsis thaliana nuclear
genome as potential candidates of the plastid inner envelope membrane proteins (AtPEM
candidates). A web based database was developed with detailed information about the
predicted proteins, including their expression profiles based on EST databases. Twenty
ATP Binding Cassette transporter proteins were selected as candidates for plastid
envelope lipid transporters and 11 T-DNA insertion mutants were analyzed for FA
compositions and contents. The knock-outs of At1g70610 and At5g64940 consistently
showed reduced oil contents (about 50-70% of wild type) in seeds and therefore are

candidates for further analysis as potential lipid transporters.

The LORD is my shepherd, I shall not be in want.

Psalm 23 :1

T 0
My grandfather &
Joseph Hang (PhD candidate) &

Hester Hughes (PhD candidate)

iv

ACKNOWLEDGMENTS

The most valuable gain during my 5 years of graduate study is the people. First I thank
my major professor John Ohlrogge for providing me with an opportunity to be a part of
this truly outstanding research community. I enjoyed working under his advice and
learned a great deal from his knowledge, wisdom, and insights. He also taught me about
the priorities of life. I appreciate his open mindedness (allowing any kind of idea),
family-like lab management, encouragement to think and gain knowledge about many
different aspects about plant lipid science, and all the supports he has provided me with.
I cannot imagine having had a better person to serve as my primary advisor. I want to
acknowledge Mike Pollard for his major contribution to this thesis. I especially
appreciate him for spending hours and hours of his precious time to teach me how to
interpret meaningless-looking kinetic data and also about lipid biochemistry. Everybody
benefits from his brilliant chemistry knowledge. I am grateful for having Ken Keegstra,
Christoph Benning and Kathy Osteryoung as my thesis committee members. Just to have
their names among my committee is my honor but I also feel privileged to have had
chance to discuss my work with them formally and privately. I thank for their discussion
and advise to make my thesis a competitive one. I thank our department chair Richard
Triemer and the director of graduate studies Frank Ewers for writing recommendations

for the ‘Dissertation Fellowship’.

I want to thank the current and the past members of Ohlrogge Lab who have made my

time in the lab invaluable and unforgettable. I acknowledge hereby that any

accomplishment that I have made in this thesis is through a collective effort of these
people. Our big brother Bao was always in the lab (regardless when I came) to challenge
everybody to work hard. Bao’s practical advices on the research direction and the know-
how that he shared helped me greatly. I’m also grateful for the time he shared at IM
sports to teach me how to work out. I thank my buddies Gustavo, Rodrigo and Fred. We
had many stimulating and interesting scientific discussions. I’ll remember the many crazy
ideas which we never tried. I also thank Mi Chung (for introducing the Ohlrogge lab to
me while I was in Korea), David Schultz, Sergei, Linda (for throwing baby shower party
for Joseph in her house), Quino, Sari, Jay, David Pan (for giving me his bike when he
left), Sarah, Anne Milcamps, Anne Jones, Tomas, Joerg, Katrin, Guiyun, Younghua,
Ajay, Vandana, Greg, Phil (hopefully there will be another graduate student in the lab
soon), Kathy, Young-hee, Tomas Edwards (for timing me with the kinetic assays and for
the peas), Naima, Troy, Swati and others I may have forgotten here. I have had a great
fortune to collaborate with you and I acknowledge that my life has been greatly enriched
by sharing it with outstanding individuals like you. I also acknowledge Uwe Rossbach,
John Guenther, Rob Halgren, Curtis Wilkerson, John Froehlich, Jeff Landgraf, Andreas
Weber, David Oliver, Martin Fulda, and John Browse for their contributions to this

thesis.

I would like to thank my parents for their love and support. I feel in debt to my in-law
parents for helping us out each time when our two babies were born. I would like to
thank my family in Jesus, Peter Yoon, Ruth Yoon, Luke Lee and Rebekah Lee for their

kind care and supports. I thank Martin Skaates, the Hughes family and many other

vi

friends in Lansing who made my life in America rewarding. There is not enough time or

space to truly express my gratitude. Thank you all.

Finally without my family it may have been impossible for me to study in a foreign
country especially in a place where winter lasts for 6 months! My love and gratitude
goes to my wonderful wife and coworker Hannah. Today is our 5‘h year anniversary.
Thank you for your love, prayer, patience and humility. I thank my two awesome babies,
Joseph and Faith for being the pride and joy of my life. I give my deepest thanks to my

Lord, God who has blessed me throughout these years and who is always with me.

vii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................ xii
LIST OF FIGURES ......................................................................................................... xiii
CHAPTER 1 ....................................................................................................................... 1
INTRODUCTION .......................................................................................................... 1
Fatty Acid Metabolism in Higher Plants: the underlying questions about FA
movements .............................................................................................................. 3
The majority of the de novo fatty acid synthesis product must be exported from
the plastid ............................................................................................................ 3
Plant membrane lipid biosynthesis requires extensive exchange of lipids
between organelles .............................................................................................. 6
Fatty acid derived storage and structural lipid syntheses .................................. 10
Membrane Transport of Free Fatty Acids: The Debate over Passive Diffusion
versus Facilitated Transport .................................................................................. 12
Diffusional model ............................................................................................. 14
Facilitated transport model: kinetic evidence for protein-mediated transport .. 18
Candidates of Lipid Transporters Identified in Various Organisms ..................... 21
Unesterified FA transporters ............................................................................. 22
Escherichia coli F adL and FadD .................................................................. 22
Yeast Fatlp and Faalp, Faa4p ...................................................................... 24
Mammalian FABPpm ................................................................................... 25
Mammalian FAT/CD36 ................................................................................ 26
FATP family conserved from mycobacterium to mammals ......................... 26
Other mammalian FA binding proteins ........................................................ 27
Activated FA transporters ................................................................................. 28
Mitochondrial camitine system ..................................................................... 28
Peroxisomal LCFA membrane transporters .................................................. 29
Membrane lipid transporters ............................................................................. 30
Human erythrocyte scramblase ..................................................................... 31
Aminophospholipid translocase .................................................................... 32
Multidrug Resistance Proteins ...................................................................... 33
Interorganellor phospholipids transport via contact zone ............................. 34
Lipid transporters in plants ............................................................................... 35
LITERATURE CITED ............................................................................................. 39
CHAPTER 2 ..................................................................................................................... 56
ON THE EXPORT OF FATTY ACIDS FROM THE CHLOROPLAST .................... 56

viii

ABSTRACT .............................................................................................................. 56

INTRODUCTION .................................................................................................... 58
RESULTS ................................................................................................................. 61
The total in vivo pool of FFA in leaves is very small and implies a very short
half life for the FFA pool involved in export from the chloroplast .................. 61
Time Course for Acetate Incorporation into Fatty Acyl Products by Isolated
Pea Chloroplasts ................................................................................................ 67
The Long Chain Acyl-CoA Synthesis (LACS) Reaction in the Dark Utilizing in
situ Generated F F A ........................................................................................... 71
Evidence for Distinct Kinetic Pools of FFA in Chloroplast Assays ................. 72
The Dependence of the LACS Reaction in the Dark on the Concentration of in
situ Generated FF A ........................................................................................... 78
Accessibility of Chloroplast F FA Products to Sequestration by BSA in the Dark
........................................................................................................................... 78
BSA Competes with LACS for the Bulk in situ FF A Pool but Is Not As
Effective for the Nascent FFA Pool .................................................................. 79
DISCUSSION ........................................................................................................... 84
The Endogenous Free Fatty Acid Pool ............................................................. 84
F FA Pools in Incubations with Intact Chloroplasts .......................................... 87
LACS Activity in Isolated Chloroplasts and a Comparison with Leaf Tissue
Labeling Experiments Suggest a Channeled FF A Export Pool ........................ 90
Possible Mechanisms for Fatty Acid Export .................................................... 93
MATERIALS AND METHODS .............................................................................. 97
Materials ........................................................................................................... 97
In vivo Incubations ........................................................................................... 97
Chloroplast Preparation and Assays ................................................................. 98
Lipid Analysis ................................................................................................... 99
LITERATURE CITED ........................................................................................... 103
CHAPTER 3 ................................................................................................................... 108
SITES AND MECHANISMS FOR ACTIVATION AND ELONGATION OF
EXOGENOUS FATTY ACIDS ................................................................................. 108
ABSTRACT ............................................................................................................ 108
INTRODUCTION .................................................................................................. 1 10
RESULTS ............................................................................................................... 112
Exogenous [1-I4C]FFA metabolism by Arabidopsis leaves .......................... 112
Effect of plastidial FAS inhibition on exogenous laurate elongation by leaves
......................................................................................................................... 117
Effect of cytosolic FA elongation inhibition on exogenous laurate elongation by
detached Arabidopsis leaves ........................................................................... 122
Contribution of recycled acetate units on the labeled products of exogenous FA
elongation by detached Arabidopsis leaves is small ....................................... 125
Exogenous laurate elongation by isolated pea chloroplasts ............................ 127
How do exogenous FA’s get activated in the plastid? .................................... 132

ix

14 . . .
Intermediates of exogenous [l- C]laurate elongatlon 1n Isolated pea

chloroplasts ..................................................................................................... 134
Laurate elongation activity in Arabidopsis mutants disrupted in expression of
candidate genes for plastid acyl activation ..................................................... 136
DISCUSSION ......................................................................................................... 145
Exogenous [1- 14C]FFA metabolism in Arabidopsis leaves ........................... 145
The plastid is the major site of exogenous MCFA elongation into C16 and C18
FA’s ................................................................................................................. 147
Identification of acyl activating enzyme in the plastid through reverse genetics
approach .......................................................................................................... 150
What is the in vivo function of FA activation/elongation in the plastid? ....... 153
MATERIALS AND METHODS ............................................................................ 156
Materials ......................................................................................................... 1 56
Homozygous T-DNA Insertion Mutant Isolation ........................................... 157
Arabidopsis detached leaf incubations ............................................................ 158
Chloroplast preparations and assays ............................................................... 158
Lipid analysis .................................................................................................. 160
Perrnanganate-periodate oxidation .................................................................. 161
Acyl-ACP intermediate analysis ..................................................................... 161
LITERATURE CITED ........................................................................................... 163
CHAPTER 4 ................................................................................................................... 167

THE PREDICTED CANDIDATES OF ARABIDOPSIS PLASTID INNER
ENVELOPE MEMBRANE PROTEINS AND THEIR EXPRESSION PROFILES:

Search for lipid transporters of Arabidopsis plastid envelope .................................... 167
ABSTRACT ............................................................................................................ 167
INTRODUCTION .................................................................................................. 169
RESULTS AND DISCUSSION ............................................................................. 173

Part I. The Predicted Candidates of Arabidopsis thaliana Plastid Inner Envelope
Membrane Proteins and Their Expression Profiles ............................................ 173
Plastid Envelope Membrane Protein Candidates ............................................ 173
What are they? ................................................................................................ 178
Characteristic Features of Plastid Envelope Candidates ................................. 186
Digital mRNA Expression Profile .................................................................. 187
Microarray Analysis ........................................................................................ 199
Towards the Function of Unknown Candidates .............................................. 204
Part II. Candidates for Lipid Transporters of the Plastid Envelope ................... 204
Plastid targeted ABC transporters ................................................................... 204

Phenotypic Observations and the Fatty Acid Composition/Contents in the KO
plants of the Plastid ABC transporter candidates ........................................... 206
CONCLUSIONS ..................................................................................................... 210
MATERIALS AND METHODS ............................................................................ 213
Establishment of a Database for Plastid Envelope Protein Candidates .......... 213

Classification by Function .............................................................................. 214

Digital mRNA Expression Profiling ............................................................... 214
Microarray Data .............................................................................................. 216
T-DNA Insertion Mutants ............................................................................... 216
Homozygous T-DNA Insertion Mutant Isolation ........................................... 217

Fatty Acid Composition and Content Measurements by Gas Chromatography
......................................................................................................................... 2 1 7
LITERATURE CITED ........................................................................................... 219
Chapter 5 ......................................................................................................................... 225
CONCLUSIONS AND FUTURE RESEARCH PERSPECTIVES ........................... 225

Three models for the mechanism of fatty acid export from the plastid and

directions for future studies ............................................................................ 226
A novel fatty acyl activating enzyme in the plastid ........................................ 234
More distantly related projects from the plastid envelope proteomics ........... 237
Future perspectives on fatty acid transport research in plant .......................... 239
LITERATURE CITED ........................................................................................... 243

xi

LIST OF TABLES

Table 1. Functions of fatty acid derived lipids in higher plants. ........................................ 2
Table 2. Summary of identified lipid transporter candidates. .......................................... 23

Table 3. Distribution of radioactivity among the major glycerolipids following 3 hours of

14
incubation of Arabidopsis leaves with various [1- C] substrates ......................... 113

Table 4. Distribution of radioactivity among the total fatty acids following 3 hours of

. . . . l4
incubation of Arabrdopsrs leaves wrth various [1- C] substrates ......................... 116

. . . . 14 .
Table 5. Label distribution among the ox1dat1ve cleavage fragments of C-labeled olerc

. . . . . 14
acid of Arabidop51s leaves Incubated wrth drfferent [1- C] labeled FA substrates.
................................................................................................................................. 127

Table 6. Summary of gene candidates of exogenous FFA activation in the plastid tested
in this study. ............................................................................................................ 139

Table 7. AtPEM candidates involved in protein import and glycerolipid metabolism. . 184

Table 8. The AtPEM candidates represented in the chloroplast envelope proteins
identified through protein mass spectrometry by Ferro et al. (2002, 2003) and by
Froehlich et al. (2003). ............................................................................................ 191

Table 9. Summary of ESTs of AtPEM candidates by tissue types ................................. 193

Table 10. Summary of AtPEM candidates with tissue-specific transcript abundances. 195

Table 11. Tissue-specific transcript abundance comparison between the plastid envelope,
thylakoid and stroma localized proteins. ................................................................ 200

Table 12. The Arabidopsis T-DNA insertion mutants for the predicted plastid ABC
transporters .............................................................................................................. 206

Table 13. Percent wild type fatty acid compositions in the leaves and seeds of the
homozygous T-DNA knock-out plants of the plastid ABC transporter candidates.211

Table 14. Homozygous knock-outs of plastid ABC transporter candidates that showed
reduced oil content in seeds compared to the wild type plants. .............................. 212

xii

LIST OF FIGURES

Figure 1. Schemetic of F AS in the plastid. ........................................................................ 4
Figure 2. Schematic overview of the glycerolipid biosynthetic pathways in Arabidopsis 7

Figure 3. Schematic for fatty acid export from the site of de novo fatty acid synthesis in
the chloroplast stroma to the cytosol. ....................................................................... 59

Figure 4. Time course for acetate incorporation into acyl lipids by pea leaves .............. 63

Figure 5. Time course for acetate incorporation into fatty acyl products by pea

chloroplasts. .............................................................................................................. 68
Figure 6. Depletion of FFA pool after transition from light to dark. ............................... 73
Figure 7. F FA concentration dependence of the LACS reaction. ................................... 80
Figure 8. Sequestration of F FA by binding to BSA ........................................................ 82
Figure 9. Competition between LACS and BSA for in situ generated F FA .................... 83

Figure 10. Competition between LACS and BSA for “nascent FFA” in fatty acid
synthesis assays ......................................................................................................... 85

. . . l4 . . .
Figure 11. Time course Incorporatlon of [ l - C]laurate Into glycerolipids by detached
Arabidopsis leaf. ..................................................................................................... 114

. . 14 . .
Figure 12. Time course elongation of [l- C]laurate substrate by detached Arabrdopsrs
leaf. .......................................................................................................................... 118

. . 14 . .
figure 13. Cerulemn effect on elongation of [1- C]laurate by detached Arabrdopsrs
leaves ....................................................................................................................... 123

. l4 . .
Figure 14. Haloxyfop effect on elongation of [1- C]laurate by detached Arabrdopsrs
leaf. .......................................................................................................................... 126

. 14 .
Figure 15. Elongation of exogenous [1- C]laurate by isolated pea chloroplasts. ...... 129

Figure 16. Time course light synthesis and dark catabolism of acyl-ACP by isolated pea
chloroplasts. ............................................................................................................ 1 3 5

xiii

14 . . .
Figure 17. Exogenous [1- C]laurate elongation by wrld type and mutant lines of
candidates for plastidial acyl activating enzymes. .................................................. 141

. 14 14
Figure 18. Elongation of [1- C]acetate and [1- C]MCF A substrates by detached
leaves of wild type and single/double knock outs of AAE15 and AAE16 ............... 144

Figure 19. Phylogenetic comparison ofAAE15 and other LACS genes ........................ 155

Figure 20. Summary scheme of plastid envelope protein prediction from Arabidopsis
thaliana nuclear genome. ........................................................................................ 175

Figure 21. Functional classification of the Arabidopsis plastid envelope protein (AtPEM)

candidates. ............................................................................................................... 181
Figure 22. Characteristics of AtPEM candidates. .......................................................... 188
Figure 23. Digital differential display analysis. ............................................................. 197
Figure 24. Tissue-specific microarray analysis of AtPEM candidates ........................... 202

xiv

CHAPTER 1

INTRODUCTION

Fatty acids play a variety of roles in living organisms (see Table 1). Fatty acids are the
core components of all biological membranes. Fatty acids are a crucial energy source for
animals especially in heart muscles, skeletal muscles, liver and adipose tissue. In plants
fatty acids are the most important means to store carbon and energy for the growth of the
next generation. Fatty acids are also the most abundant form of reduced carbon chains
available from nature and have diverse uses ranging from food to industrial feedstocks.
Current world vegetable oil production is estimated at 87 million metric tons with an
approximate market value of 40 billion US. dollars (Gunstone, 2001). As much as 25%
of human caloric intake in developed countries is derived from plant fatty acids (Broun et
a1, 1999; Thelen and Ohlrogge, 2002). Plant fatty acids are also the major ingredients of
nonfood products such as soaps, detergents, lubricants, biofuels, cosmetics, and paints

(Ohlrogge, 1994).

The basic metabolic pathways leading to the production of fatty acids (FA’s) and their
major derivatives have been well studied (reviewed below) and are known to occur by
similar mechanisms in all organisms. However, there is one major area that has failed to
be explored in lipid research and that area is the various aspects of fatty acid movement.

Comparatively more effort has been put into understanding fatty acid transport issues in
animal systems largely due to their important medical implications. However, until

recently there has essentially been no major research done in plants on this topic. The

work described in this dissertation is one of the few pioneering studies dealing with fatty

acid transport in plants (Footitt et al., 2002; Xu et al., 2003). The main objective of this

thesis is to begin to explore the mechanism(s) of fatty acid transport in plants. I begin

this thesis with the following goals:

1. Understand the general topic and issues of fatty acid movement.

2. Examine and define research areas to be explored in plants in terms of FA transport.

3. Gain new information on at least one major aspect of fatty acid transport and develop

strategies for the future research directions.

Table 1. Functions of fatty acid derived lipids in higher plants.

 

Function FA derived lipids

references

 

Membrane structural glycerolipids, sphingolipids
components

Storage compounds triacylglycerols, waxes

Surface protection cutin, suberin, surface waxes
and water proofing

Signaling jasmonate, oxylipins,
diacylglycerol,
phosphatidylinositol 4,5-
bisphosphate, phosphatidic acid

Posttranslational myristic acid, palmitic acid,
protein modification phosphatidylinositol, ceramide

Browse and Somerville,
1994

Ohlrogge and Browse, 1995

Kunst and Samuels, 2003

Vick and Zimmerman, 1984;
Farmer and Ryan, 1990;
Alexandre and Lassalles,
1992; Cote and Crain, 1993;
Munnik, 2001

Towler et al., 1988;
McIlhinney, 1990;
Milligan et al., 1995

 

Chapter I is organized to give an overview of 1) FA metabolism in plants with an
emphasis on the subject of FA movements, 2) the debate over FA membrane transport in
animal and other cells and 3) the major lipid transporter candidates identified in various
organisms. I devote a relatively large part of chapter I to describe studies in other (non-
plant) organisms because so little information is available for plants.

Fatty Acid Metabolism in Higher Plants: the underlying questions about FA
movements

Unlike animal cells, each plant cell is autonomous with respect to fatty acid synthesis,
and there is believed to be no extensive exchange of fatty acids or lipids between cells
and between distant parts of plant tissues. However, lipid metabolism within the cell
demands movement of fatty acids or lipids from one organelle to another. Transport
systems control the flux of metabolites between organelles, and are often highly
regulated. This may be the case for fatty acids, representing as they do the key building
blocks for membrane lipid biogenesis.

The majority of the de novo fatty acid synthesis product must be exported from the
plastid

In plants, de novo fatty acid synthesis (FAS) occurs almost exclusively in the plastid
(Ohlrogge et al., 1979). FAS is a continued cycle of reactions that result in growth of
fatty acyl chain by 2 carbon units (Figure 1). The source of the 2 carbon units are derived
from acetyl-CoA which becomes committed to F AS by the action of acetyl-CoA
carboxylase (ACCase) which catalyses the ATP dependent carboxylation of acetyl-CoA
to yield malonyl-CoA (Ohlrogge and Browse, 1995). Acyl carrier protein (ACP) is used

as a scaffold of the growing acyl chain and used as a vehicle to move FA chains between

sequential biosynthetic steps. Except for the primary condensation reaction between
malonyl-ACP and acety-CoA, the elongation reaction proceeds by condensation between
the growing chain of acyl-ACP and malonyl-ACP. The condensation reaction is
catalyzed by a class of enzymes called B-ketoacyl-ACP synthases (KAS). Generally in
higher plants these cycles continue seven or eight rounds forming 16 or 18 carbon
saturated acyl-ACP’s. The elongation reaction stops when the growing FA chain is either
hydrolyzed to a free fatty acid molecule or alternatively transferred to another acceptor
molecules (glycerol-3-phosphate, lysophosphatidic acid) to form lyso-/phosphatidic acid

for plastidial glycerolipid synthesis.

Problem

Since the plastid is the primary and almost the only source of fatty acid precursors for the
entire cell, the flux of acyl chains through the plastid envelope should be large. In
addition fatty acid synthesis and utilization must be well coordinated and regulated in
order to meet the different demands for FA’s in/at various cell types and developmental
stages (Ohlrogge and Jaworski, 1997). It is not known how the plastid synthesized fatty
acids get exported, nor how transport is regulated. The current model only proposes that
the released free fatty acid crosses the double membrane of plastid envelope to the outer
envelope where it is reactivated to acyl-CoA for utilization in cytosolic lipid synthesis.
The basis of this model comes from the observations that the acyl-ACP thioesterase is a
stromal activity (Shine et al., 1977; Doerrnann et al., 1995; Voelker et al., 1997), while
the acyl-CoA synthetase is associated with the outer envelope (Block et al., 1983;

Andrews and Keegstra, 1983). In addition free fatty acids are end-products of fatty acid

synthesis in chloroplast assays, unless ATP and CoA are added, in which case acyl-CoA

becomes a major product (Roughan and Slack, 1982). The involvement of a hydrolytic
. . . . . . l8 .

step in this process was conclusrvely demonstrated by a study usrng m vzvo O labeling

technique (Pollard and Ohlrogge, 1999). Chapter 2 of this thesis addresses the question

whether the export of free fatty acid involves free physical diffusion or involves a

channeled system.

Figure l. Schemetic of F AS in the plastid.

The thickness of arrows does not necessarily indicate the flux.

 

 

 

    

_ ER pathway
Plastid Envelope 1
Malonyl-CoA Free Fatty Acids
AL
co: I 13:1 ACP Plastld
Acetyl-CoA pathway
MalonyI-ACP N"18:0 ACP
:16: O ACP
co Acyl-ACP
CONDENSATION
Cycles until
166/186 FA
REDUCTION
REDUCTION
DEHYDRATION

Plant membrane lipid biosynthesis requires extensive exchange of lipids between
organelles

Membrane glycerolipids have F A’s attached to the sn-1 and sn-2 positions of the glycerol
backbone and a polar head group in the sn-3 position. The combination of nonpolar acyl
chains with polar head groups leads to the amphipathic physical properties of these
molecules, which are essential for the formation of membrane bilayers (Nagle et al.,
2000). The acyl chains usually are composed of 16 and 18 hydrocarbon chains with one

to three cis double bonds in plants.

Two distinct pathways for glycerolipid synthesis exist in higher plants (Figure 2). Based
on their similarity to the bacterial pathway and animal pathway they have been termed
the prokaryotic pathway and the eukaryotic pathway respectively (Roughan and Slack,
1982). The prokaryotic pathway takes place within the chloroplast whereas the
eukaryotic pathway involves export of free fatty acid from the plastid and glycerolipid
synthesis in the endoplasmic reticulum (ER). Some of these ER synthesized lipids return
back to the chloroplast for the synthesis of plastidial glycerolipids. The eukaryotic
pathway is the principal route of glycerolipid synthesis in all nonphotosynthetic tissues as
well as in the photosynthetic tissues of 18:3 plants. In leaves of 16:3 plants such as
Arabidopsis and spinach both pathways contribute about equally to the synthesis of leaf

glycerolipids (Somerville and Browse, 1991).

Problem
According to the two-pathway model, a considerable amount of lipid should travel back

to plastids from the ER. In addition, evidence from several Arabidopsis mutants

Figure 2. Schematic overview of the glycerolipid biosynthetic pathways in

Arabidopsis (adapted from Browse and Somerville, 1991).

Abbreviations for the lipid structures: G3P, glycerol-3-phosphate; LPA, l-acyl-glycerol-
3-phosphate; PA, phosphatidic acid; DAG, diacylglycerol; PG, phosphatidylglycerol;
PG-P, phosphatidylglycerol-3-phosphate; MGDG, monogalactosyldiacylglycerol;
DGDG, digalactosyldiacylglycerol; SL, sulfoquinovosyldiacylglycerol; PC,

phosphatidylcholine; PI, phosphatidylinositol; PE, phosphatidylethanolamine.

 

\ 9535a: ouch—335v 0.54.5

 

        

 

 

             

 

 

 

  

 

 

 

 
  

 

 

30: 60:
n9 09 n9 n9 n9 39 99 99 09 n9 99 99 to: 99
IL L LIL llrIIL IIPIIIL LIL ILIL
.v D b .P b. 1.. 1.. b b P 2e
.a 4. ._. I d .33.
99 «9 o9 «9 «9 «9 :9. «9 [Iql
I I'll. L
c. p p c. an...
n .i d. i as: 4. 4 Be
:0. to. 20: to. nod 00—
I'lrl. IL
I. e00 «NOS
mo<wu0w
_ 5.0: .-r l-l
«.9 99 «.9 «"9 «.9 «”9 ed. .9 99 .9 09 79 99 :9
IL IGIOIIUEIbII- IrIL LrIL‘llIIII I-mmlr.
.5 /§ one: .5 e80 one:
6.9— . / \ \2
«.9 «ILIL9 o9 .9 0.9 :9 . -ommb.m8§do<_§\
tattoo I1 I \

 

5. 0: .o. 0: 5. 0: 5. 0:
p.0w '0'

5.0:

00pn0w a“; w:0wp0p «:0—p0w
I A:LI‘.I.LIL‘I
040

. . .

ma 040100 I Ga .3

 
 

39:52— ouotaxamv 23:0..me oiwfinoazm.

indicates that lipid exchange between the ER and the plastid is somewhat reversible
(Miquel and Browse, 1992; Browse et al., 1993). Mutants deficient in ER desaturases
still contain polyunsaturated fatty acids in extraplastid membranes which suggests that
they were derived from the plastid. Also, several lines of evidence indicate the existence
of regulatory mechanisms that coordinate the activity of the two pathways. For example,
the act] mutant of Arabidopsis is deficient in activity of the first enzyme of the
prokaryotic pathway and thus has severely reduced flux through the prokaryotic pathway
(Kunst et al., 1988). However the mutant plant compensates the deficiency by increasing
the flux through the eukaryotic pathway. What are the signals that communicate between
the ER and the plastid to coordinate these pathways? What is the system that the plant
cells use to move hydrophobic lipid molecules between each membrane compartments?
How does the DAG moiety coming from ER cross the outer membrane of the plastid
envelope and reach the inner membrane where MGDG synthesis occurs? Vesicular
transport between the ER and the plastid has not been observed. Once lipid-transfer
proteins were thought to be the intracellular carriers of lipid molecules, but it has been
shown that their main location is outside the plasma membrane in the cell wall space of

epidermal cells (Bernhard et al., 1991; Sterk et al., 1991; Thoma et al., 1993).

The plastid envelope is clearly an intermediary site for exchange of both fatty acids and
lipids. The major fluxes are believed to involve fatty acid e_xpgr_t from plastids and
glycerolipid ir_npo_rt from the ER, although movements in the opposite direction may
occur in some cases (Chapter 3; Browse et al., 1993). A large part of lipid metabolism

also occurs at the inner and outer membranes of plastid. It is likely that specific

membrane proteins mediate the exchange of lipid molecules between ER and plastid.
The transfer of lipid moieties between the outer envelope membrane and the inner
envelope membrane of the plastid also is likely to be mediated by specific proteins.

A genome wide survey of putative Arabidopsis plastid inner membrane proteins through
bioinformatics approaches is presented in chapter 4. Among the candidates identified
were ATP binding cassette transporters, many of which have been implicated as lipid
transporters in other systems (see review below). Initial efforts toward identification of
potential plastid envelope lipid transporters through reverse genetics approach are

described within the chapter.

Fatty acid derived storage and structural lipid syntheses

In seeds, fatty acids are precursors of triacylglycerol (TAG) biosynthesis, the major sink
of carbon in oilseeds. Some of the steps involved in triacylglycerol synthesis and many
aspects of subcellular compartmentation are the same as lipid synthesis in leaves (Stymne
and Stobart, 1987; Browse and Somerville, 1991; Miquel and Browse ,1994). As with
the eukaryotic pathway, PA for TAG synthesis is made in the ER and the
dephosphorylation of resulting PA forms DAG. In a TAG synthetic pathway called
Kennedy pathway, a third FA is transferred to the sn-3 position of DAG, catalyzed by
diacylglycerol acyltransferase (DAGAT) (Ohlrogge and Browse, 1995). Alternatively
DAG can be formed from PC. PC participates in TAG formation by either donating it’s
entire DAG moiety (Slack et al., 1983) or providing acyl moieties to the acyl-CoA pool

(Stymne and Stobart, 1987). TAG can also be synthesized in an acyl-CoA-independent

10

mechanism, which uses phospholipids as acyl donors and DAG as acceptor catalyzed by

phospholipid-diacylglycerol acyltransferase (PDAT) (Dahlqvist et al., 2000).

Relatively less is known about the syntheses of structural lipids such as wax, cutin and
suberin. The aliphatic components of the cuticular wax are synthesized in the epidermal
cells from the saturated very long-chain fatty acids (VLCFA, 20-34 carbons). The C16
and C18 FA’s synthesized in the plastid are elongated into the VLCFA by extra-plastidial
membrane-associated multienzyme complexes, known as the fatty acid elongases (FAE)
(Fehling and Mukherjee, 1991; Kunst and Samuels, 2003). The VLCFA’s are further
modified to produce the primary alcohols, aldehydes, alkanes, secondary alcohols, and
ketones, which contribute to the wax formation (Cheesbrough and Kolattukudy, 1984).
The principal constituents of cutins and suberins are also derived from the saturated and
unsaturated FA’s that are hydroxylated, epoxidated, and crosslinked to form a network

(Kolattukudy, 2001).

Problem

Free fatty acids are toxic to the cell and could cause deleterious effects on membrane
integrity. Therefore the cell must control the accumulation of free fatty acids during the
synthesis and degradation of storage and structural lipids. During the germination of
seeds, TAG stored in the oil body needs to be immobilized to be used to provide carbon
and chemical energy. Catabolism of fatty acids occurs exclusively in the glyoxysome in
plants. Therefore, once the fatty acid gets liberated from the glycerol backbone of the

TAG, they should be funneled into the glyoxysome to be oxidized. Other than the

11

microscopic observation that during germination, the oil body and glyoxysome come in
close physical contact which is hypothesized to facilitate the transfer of fatty acids, not
much is known about this process (Mollenhauer et al., 1970; Vigil et al., 1970; Wanner et
al., 1982). As will be reviewed later, a membrane transporter is thought to be involved in
the internalization of fatty acids into the glyoxysome. How the hydrophobic building
blocks of structural lipids are moved from the ER to the site of polymerization and
eventually out of the plasma membrane is also currently not known (Kunst and Samuels,

2003)

A set of experiments dealing with exogenously provided fatty acid metabolism within the
cell are presented in Chapter 3. Plant cells possess a system to accommodate excessive
amounts of fiee fatty acids by incorporating them into their endogenous lipids. A stromal
localized protein is involved in the activation of medium chain fatty acids for integration

into the regular plastid FAS machinery.

Membrane Transport of Free Fatty Acids: The Debate over Passive Diffusion versus

Facilitated Transport

The studies of membrane transport of unesterified FA’s reviewed in this section are
mostly adapted from animal system where release and uptake of FFA’s to/from plasma
by different types of cells is a routine process. Although plant cells are conceptually
different in respect to exchanging FFA between cells, they require massive FFA

exchange between subcellular compartments as reviewed above. FA movement across

12

biological membrane is essential to the physiology of all living organisms and thus may

have conserved aspects across the bacterial, plant and animal kingdoms.

Dietary triglycerides digested by gastric and pancreatic enzymes are emulsified with bile
salts in the digestive tract and are internalized by intestinal epithelial cells (Frohnert and
Bemlohr, 2000). Free fatty acids are released from circulating TAG’s by endothelial
lipoprotein lipases for uptake by multiple tissues including adipose, skeletal and heart
muscles. In the resting state, FA’s are the heart’s primary fuel source. Fuel reserves in
the plasma are essentially negligible and well-oxygenated heart muscle cells receive FA
via the plasma from adipocytes in distal sites (Neely and Morgan, 1974). Adipose TAG’s
are the single largest energy depot, enough to maintain life for about three months (Cahill
Jr., 1970). FA must leave the adipocytes, cross the capillary endothelium, travel through

the blood stream, and eventually enter the cytosol of the heart cell (Hamilton, 1998).

The amphipathic nature of F FA’s has given rise to a major and long-standing debate over
whether they move across cell membranes by passive diffusion, or by facilitated transport
mechanism. Some investigators have proposed that the passive diffusion of FA is slow
and thus is the rate-limiting step for FA transport in model membranes (Kleinfeld et al.,
1995; Kleinfeld, 2000). In direct contrast others argue that simple diffusion by lipid
physical chemistry is fast and sufficient to match the physiological requirements (Kamp
et al., 1995; Hamilton, 1998; Hamilton and Kamp, 1999; Kamp et al., 2003). Some
investigators have suggested transporter involvement under low concentrations of FFA

and passive diffusion at higher levels of FFA (Abumrad et al., 1981) or vise versa;

13

passive diffusion under basal condition and transporters that increase the uptake above

the basal level (Schaffer and Lodish, 1994).

Diffusional model

The physical chemistry studies of FA transport involve use of model membrane vesicles.
Model vesicles are mostly a single lipid bilayer encapsulating aqueous buffer. Small
unilamellar vesicles (SUV’s) have around 250 A diameter and are prepared by
sonication. The large unilamellar vesicles (LUV’s) have a lower surface curvature and
have a higher variability in diameter (usually < 500 nm). LUV’s can entrap smaller
vesicles. The membrane may be made with phospholipids of choice with or without
proteins (Hamilton, 1998). In classical uptake studies, membrane vesicles are incubated
with FA’s in a solution containing albumin and initial uptake rates are evaluated by

measuring radiolabled FA’s bound to or trapped in the vesicles (Hamilton and Kamp,

. 13 .
1999; Stump et al., 2001). Alternatrvely, C NMR spectrosc0py was used to monitor

the FA binding to albumin or to phospholipids by characteristic chemical shifts of the
carboxyl carbon (Hamilton and Cistola, 1986). Fluorescence was also used to monitor
FA transfer from a donor to an acceptor (Noy and Zakim, 1985; Kamp and Hamilton,
1993). The transbilayer movement of FA was monitored by detecting the change in pH
inside the vesicles using a fluorescent pH probe (pyranin) (Kamp and Hamilton, 1992;

Kamp et al., 1993).

14

The diffusional model views passive diffusion of FA through the lipid bilayer as the
central mechanism of transport of FA into and out of cells and predicts that in general M
metabolism regulates transport of FA rather than the reverse (Hamilton, 1998). Three
minimal steps are involved in membrane FA transport by passive diffusion model; 1)
adsorption from extracellular media (albumin) to the lipid bilayer, 2) transmembrane flip-

flop, 3) desorption from lipid bilayer to the cytosol.

Various studies have demonstrated the preferential partitioning of long-chain FA’s into

membranes as opposed to water. The partition coefficient (IOICatelmembrane/
[oleate]water) of olerc acrd 1n model phospholipids vesrcles was 0.5 X 10 (Richrerr et al.,
1995; Kamp et al., 2003). The calculated rate constant (Keq = kofp’kon) of pahnitate to

7 -1 . . .
dimyristoyl PC SUV was 2.4 X 10 5 (Hamilton, 1998). If albumin rs a donor the

partitioning of FA between albumin and the membrane will be affected by factors such as

the FA to albumin ratio, the type of FA, the relative amounts of membrane and albumin,

. . 13
the membrane composmon, the pH, and temperature. C-NMR spectroscopy

demonstrated a reversible transfer of albumin bound oleate to phospholipid bilayers with

exchange times in the range of 0.1 s < t1/2 < min (Hamilton and Cistola, 1986).

Membrane permeability coefficients for short- and medium-chain FA’s (up to 12

carbons) are high and increases with chain length (Antonenko et al., 1993; Zakim, 1996).

15

The un-ionized FA move across the bilayer rapidly (tl/Z < 1 s, Kamp and Hamilton,

1992) compared to the anion. However, if FA is present in a memberane exclusively as
anions, FA diffiision across membrane must be slow and will have to depend on anion
transporters. Several investigators have shown that the apparent pK of the FA in SUV is
about 7.5, close to physiological pH (Ptak et al., 1980; Hamilton and Cistola, 1986).
About 50% of the FA population in the phospholipid bilyer is expected to be in un-
ionized form in the physiological pH range. Thus it is possible that FA could diffuse
across the lipid bilayer via ther unionized form (Hamilton, 1998). There seems to be
disagreement between the measurements of LCFA’s in larger size model vesicles.
Keinfeld et al. (1998) reported an order of magnitude decrease in the rate of flip-flop in
PC/cholesterol vesicles when the diameter was raised from 100 nm (LUV’s) to 200 nm
(giant unilamellar vesicles). In pancreatic B-cells, the rate of acidification upon addition
of LCFA was slower than for LUV’s by several orders of magnitude (Hamilton et al.,
1994). Kamp et al. (2003) used both vesicles of the plasma membrane

(protein/phospholipids ratio, 1:1 (w/w), 100-500nm diameter) and intact adipocytes.

. . . . 6 .
Thelr reported partrtron coefficrents of 1x 10 for plasma membrane vesrcle and 1.4 X

106 for adipocytes were similar to that of protein free model membranes. Oleic acid flip-

flop across the plasma membrane vesicle and adipocyte cell was measured by the change

in fluorescent pH probes in the vesicle and was fast (t1 /2 < 5 3). Both in LUV’s and in

cells, the lack of dependence of the rate of pH change on FA chain length or structure

(FA dimmers or alkylamines) suggested a non-protein mediated mechanism.

16

Desorption of FA from a phospholipid surface is slower than flip—flop and dependent on
the FA chain length and the level of unsaturation. The spontaneous desorption of typical

FA from 3 PL interface when evaluated from SUV to acceptor SUV was reported to be

fast (t1/2 < 1 s) but was slower for very long-chain FA’s (t1/2 about 30 min for C24)

(Massey, 1997; Zang et al., 1996). These studies support the conclusion that the physical
properties of FA movements across the membranes are rapid and do not require protein-

mediated transport.

However, there are several caveats to the above arguments that may weaken such a
simple conclusion. First, the studies that show fast movement of FA either in model
vesicles or cells do not rule out the possibility that alternate or additional mechanisms
exist particularly in cells with high FA uptake (Kleinfeld et al., 1997). Second, the
spontaneous movement of FA’s in these model membrane vesicles may not necessarily
be reflected in biological system where extracellular components such as the collagen
matrix, non-lipid components in the membrane and the unstirred water layer affects the
physical properties of FA interaction with membranes (Richieri et al., 1995; Kleinfeld,
2000). Third, the internal acidification upon FA entry into the vesicle which was used to

probe the FA flip-flop may be attributed to an indirect effect in cases with cell
. . . 4 .

metabolism (Abumrad et al., 1998). The volume In LUV’s Is 10 times less than that of

an average cell (20pm) and the changes in pH measured in larger volume compartments

will be smaller. The flip-flop of FA in biological membranes may be much slower,

where the different curvature, lipid composition, and the presence of membrane proteins

might slow down FA diffusion (Abumrad et al., 1998; Berk and Stump, 1999; Schaffer

l7

et al., 2002). Fourth, it turns out that the ratio between albumin and FA, especially the
unbound FA concentration, is a very important factor in determining kinetics. Many of
the physical chemistry experiments to measure FA diffusion into the membrane vesicles
were carried out at very high FA to albumin ratios (Abumrad et al., 1998). Fifth, the
relative contribution of the protonated and ionized forms to FA uptake remain uncertain.
Sixth, although there is agreement that short-chain FA can diffuse very quickly across the
membrane, reported measurements of long-chain FA’s differ considerably (Kleinfeld and
Storch, 1993; Kleinfeld et al., 1997). Seventh, flip-flop of nonionized FA’s within the
membrane and subsequent ionization are not necessarily equivalent to in vivo transport of

FA for metabolism (Schaffer, 2002).

Facilitated transport model: kinetic eviden ce for protein-mediated transport

One of the key arguments for facilitated FA transport came from the saturation kinetics
observed for FA uptake into isolated rat adipocytes (Abumrad et al., 1981, 1984). The
uptake of FA tracers was carried out at physiologic albumin concentration. At each time
point, uptake was stopped by addition of ice-cold solutions containing 0.1% BSA and
phloretin. Phloretin inhibits further influx or efflux of FA whereas BSA removes any
FA’s that adhere to the cell surface. When FA and albumin are both present in the
extracelluar medium, the kinetics of uptake follows the molar ratio of FA to albumin
rather than the total FA concentration (Sorrentino et al., 1989; Spector et al., 1965).
Initial rates of LCF A transport into isolated adipocytes were a function of unbound FA

concentration in the medium, and demonstrated to have both saturable and nonsaturable

18

components. The saturable components are observed at low (< 3.0) LCF A to BSA ratios
while the kinetics exhibits nonsaturable components with the increase of the ratio. The

ratio in normal human serum is about 0.74 (Richieri and Keinfeld, 1995). The measured

Km from transport was 7nM for oleate which is in the range of circulating concentrations

of unbound F A’s (Abumrad et al., 1998). The saturable component is specific for LCF A’s
but not for medium-chain fatty acids (MCFA’s). Saturation characteristics were also
observed later with hepatocytes, cardiac myocytes (Stremmel and Berk, 1986), CacoZ
human intestinal cells (Kim and Storch, 1992), myocytes (Stremmel, 1988; Luiken et al.,
1997), and giant sarcolemmal membrane vesicles from rat hindlim skeletal muscles

(Bonen, 1998).

A second argument comes from competition experiments where competition for uptake
was observed between different substrates. Competition was demonstrated among long-
chain FA’s but not short-chain FA’s with different cells including adipocytes (Abumrad
et al., 1984; Storch et al., 1991), hepatocytes (Sorrentino et al., 1996), and an intestinal
cell line (Trotter et al., 1996). In a study with a cultured cell line of adipocytes, a
fluorescent FA analogue that can not be metabolized in the cell was used as competitor
(in order to avoid the possibility of different demand of FA’s in cell metabolism being the

cause of observed competition) but competition was still observed (Storch et al., 1991).

The third evidence for protein mediated transport is derived from studies that used
protein modification reagents to block the FA transfer process. Transport was inhibited

by phloretin, pentachlorophenol, diisosalicylate, quercetin, diisothiodisulfonic acid,

19

sulfosuccinimidyl derivatives of oleate and myristate, photoaffinity labeling reagent 11-
azistearate derivative and in some cases with proteases such as pronase and trypsin in a
variety of cells tested including hepatocytes, cardiac myocytes, CacoZ intestinal cells,
myocytes, giant sarcolemmal membrane vesicles, rat renal basolateral membrane vesicles
(Abumrad et al., 1998; Frohnert and Bemlohr, 2000). Many of the reagents were known
anion transport inhibitors which may suggest that the transported fraction of FA’s is in

ionized form.

It is possible that both diffusional and facilitated mechanisms are operative in the
movement of FA’s across the membranes with the relative contribution being different at
different physiological conditions, with different FA needs for different metabolic
processes, or with different FA substrates. Studies of hepatocytes and adipocytes,
conducted over a wide range of FA to BSA ratio specifically identified both saturable and
nonsaturable uptake components (Stump et al., 1992; Stump et al., 2001). At the ratio
ranging from 0.5 to 3, more than 90% of uptake was via the saturable pathway. As the

ratio increased, the linear process took over. Within the physiological range of 0.5-3, rate

. -1
constants for saturable transmembrane Influx were 2.9 s and were 10 to 30 fold faster

than those for linear uptake (k = 0.1 — 0.26 5-1, t1 /2 = 2.7 - 6.6) (Stump et al., 2001). Cell-

specific changes in the Vmax for saturable FA uptake occur in human disease states
including obesity, type 2 diabetes, alcoholic liver disease, and cardiomyopathy (Zhou et
al., 1992; Berk et al., 1997, 1999; Stump et al., 2001). In addition a syndrome of recurrent
episodes of severe liver failure in children was attributed to a defect in LCFA membrane

transport (Odaib et al., 1998).

20

Some of the counter arguments to the facilitated model include the following. First the
observation of saturation kinetics is not sufficient evidence for protein transport and may
reflect partitioning of FA between membrane and albumin (Kamp et al., 1993).
Secondly, reagents used to modify proteins to show inhibition of FA uptake may interfere
with intracellular lipid metabolism or may alter the membrane structures that resulted in
the physical partitioning of FA (Kamp et al., 2003). Thirdly, most of the studies
supporting protein-mediated transport were conducted with the transmembrane step not
being isolated from adsorption and desorption steps. Fourth, although FA uptake was
affected by expression or suppression of FA transport related proteins, the actual
mechanism of action has not been established and thus it is hard to know whether they
are directly involved in transport per se or are involved in the overall uptake process

indirectly (Hamilton, 1998).

Candidates of Lipid Transporters Identified in Various Organisms

Proteins that are thought to be involved in the various aspects of lipid movement have
been accumulated over many years in different organisms using different experimental
strategies. Some candidate proteins are supported with more convincing functional data
while others lack evidence. Thus the level of confidence may vary between the
candidates. In order to assist better judgment on such candidates short details are

provided in this section. Table 2 summarizes the candidates reviewed in this section.

21

They are organized into 4 main groups; 1) unesterified FA transporters, 2) activated FA

transports, 3) membrane lipid transporters, and 4) plant lipid transporters.

Unesterified FA transporters

Escherichia coli FadL and F adD

E. coli can be grown under condition where de novo LCF A biosynthesis is blocked by the
FAS inhibitor cerulenin and providing LCFA’s in the media as the sole carbon source.
This feature makes genetic screening possible for mutants that can not grow under this
condition. The outer membrane of E .coli is composed of an external layer of
lipopolysaccharide and an internal layer of phospholipids associated with a layer of
peptidoglycan. The lipopolysaccharide is refractory towards hydrophobic compounds
such as LCFA’s. The outer membrane and inner membrane are separated by a
periplasmic space. Transport of LCFA across these barriers is necessary under the above
screening condition. The screen led to identification and mapping of the fadL gene,
encoding the long-chain fatty acid transport protein FadL (Nunn and Simons, 1978). The
F adL protein has binding specificity toward long-chain fatty acids, and the carboxylate of
the fatty acid is essential for ligand binding (Black, 1990; Black and Zhang, 1995). FadL
spans the outer membrane and is composed of two discrete domains; the amino terminal
domain is exposed at the surface making it protease sensitive and is involved in ligand
binding, and the carboxyl terminal domain is embedded within the membrane and is
thought to form LCFA specific channel. Two allelic mutations in the transmembrane

region keep FadL in an open conformation (Kumar and Black, 1991, 1993).

22

Table 2. Summary of identified lipid transporter candidates.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- Example match in
Classification Name Organrsm/ . _ 1
organelle Arabrdopsrs
5 genes for Tsp (E <
FadL, FadD, Tsp, Hi/FA bacteria 2 , ,
cotransporter 9e-7) , AMP binding
protein family for F adD
. AMP binding protein
U .f d Fatlp, Faalp, Faa4p fungi family
nesterr 1e . 7 genes
FA transporters FABPpm 3mm“ (E < e-115)
FAT/CD36 animal No significant 3
F ATP family animal 9MP binding protein
amily
Caveolin, 56 kDa renal FA . . .
. . . animal No Significant
binding protein
Carnitine system; CPT-I, . h . mitochrfmdrrlal tsubstrate
. C ACT CPT-II mitoc ondrra carrier arm y or
Activated FA ’ CACT
transporters long-chain acyl-CoA . PXAl, ABC transporter
transporters; Patlp, Pat2p, peroxrsome famil
ALDp y
Scramblase animal NO Significant
Aminophospholipid . AL A1, haloacid
translocase; Rh protein, PS- .
. . dehalogenase-lrke
stimulated Mg2+ ATPase, animal h drolase famil
Membrane chromaffin granule ATPase y y
lipid II
transporters Multidrug Resistance
Proteins; Mdr2, MDRl, animal, .
MDR3, MRP1, ABCR, bacteria ABC “WWI“ fa‘mly
ABCl, MsbA
Interorganellor contact fun i WD-40 repeat family
zones (MAML MET3O g
Plant lipid PXAl VIPPl TGDl
’ ’ ’ l t N/A
transporters ALA] p an

 

Blast search was performed using ‘The Arabidopsis Information Resource (TAIR)

glast’ program (URL http://www.arabidopsis.org/Blast/) using peptide sequences.

The ‘Expect value (E)’ is from the TAIR blast and is a parameter that describes the
number of hits one can "expect" to see just by chance when searching a database of a

particular size.

E > 0.01 is considered insignificant in this table but no ‘defined’ E value for

insignificant homology exists.

23

 

The LCFA transport is shown to be shock-sensitive, which supports the existence of a
periplasmic component, although no specific periplasmic FA binding proteins were
identified. The periplasmic protease Tsp is required for optimal levels of transport and is

thus believed to be peripherally involved (Azizan and Black, 1994). An oleate binding

. . . + .
protein rs postulated to be an inner membrane H /FA cotransporter but remains to be

proved (Kameda et al., 1987).

Another gene that was identified fi'om the genetic screening is fadD (Black et al., 1992).
FadD belongs to the family of long-chain fatty acyl-CoA synthetases which catalyzes the
formation of fatty acyl-CoA from FFA. FadD is routinely isolated from the cytosolic
fraction although there is evidence that it becomes membrane-bound in response to

specific physiological conditions (Mangroo and Gerber, 1993).

Yeast Fatlg and Faalp, Faa4p

The LCFA uptake by yeast follows saturable kinetics suggesting involvement of
facilitated transport (Kohlwein and Paltauf, 1983). Disruption of the FA T1 gene in yeast
resulted in impaired uptake and growth on LCFA (Faergeman et al., 1997). Deletion
mutants can not grow on media containing cerulenin and LCFA. Furthermore the
mutants are impaired in accumulation of the fluorescent FA analogue BODIPY-3823,
myristate, palmitate and oleate. Fatlp belongs to the family of adenylate forming
enzymes which contain ATP/AMP binding domain but lacks the long-chain fatty acyl-
CoA synthetase signature (Faergeman et al., 1997). Fatlp has high similarity to the

murine fatty acid transport protein FATP identified from 3T3-L1 adipocytes (Schaffer

24

and Lodish, 1994). Fatlp is predicted to span the membrane four times, with the
ATP/AMP binding site residing on the cytoplasmic side. Upon entry the LCFA’s are
activated to CoA thioesters by Faalp and F aa4p (Johnson et al., 1994). Strains carrying
deletions in FAA] and FAA4 are phenotypically similar to the FA T1 deletion mutants. It
has been hypothesized that there is a mechanistic linkage between the import and
activation of LCFA, and that one or both of Faalp or F aa4p are the components of fatty

acid transport system in yeast (Faergeman et al., 2001).

Mammalian FABPpm

FABPpm was purified from solubilized rat liver plasma membranes and jejunal
microvillous membranes by oleate-agarose affinity chromatography (Stremmel et al.,
1986). FABPpm is expressed on the plasma membrane of liver, adipose tissue, cardiac
muscle, intestine, vascular endothelium and internal membranes where FA uptake/export
is highly active (Schwieterman et al., 1988; Sorrentino et al., 1988). FABPpm and
mitochondrial aspartate aminotransferase (mAspAT) are shown to be identical proteins
(Stump et al., 1993). When mAspAT was expressed on the surface of fibroblasts the
oleate uptake was increased (Isola et al., 1993) and antibodies to mAspAT/FABPpm
reduced FA uptake in various cells (Stremmel and Theilmann, 1986; Zhou et al.,
1992,1995). Consistent with the postulated function of fatty acid uptake, FABPpm
expression was found to be upregulated during adipocyte differentiation (Dutta-Roy et
al., 1996). The FABPpm expression was also found to change in adipocytes of diabetic

rats (Berk et al., 1997), during endurance training (Kiens et al., 1997) and fasting

25

(Turcotte et al., 1997). The mechanism by which this globular protein is targeted and

tethered to the plasma membrane is unclear.

Mammalian FAT/CD36

Fatty acid translocase (FAT, 88kDa) was identified by covalent labeling with

radiolabeled inhibitors of the FA transport step such as 4,4’-diisothiocyanostilbene-2,2’

disulfonate (DIDS), sulfosuccinimidyl myristate (SSM) or sulfo-N-succinimidyl oleate
($80) from adipocytes (Harmon and Abumrad, 1993). FAT has sequence similarity to
the human platelet integral membrane glycoprotein CD36 that functions as a multiligand
scavenger receptor (Abumrad et al., 1993). CD36 is found in tissues with a high
metabolic capacity for FA. Expression of CD36 in fibroblasts which normally lack this
protein induces appearance of a saturable, high affinity, phloretin-sensitive component of
FA uptake (Ibrahimi et al., 1996). CD36 expression is also increased in the muscle of
diabetic animals, in mice fed with a high fat diet, and is regulated by change in FA
utilization (Abumrad et al., 1998). CD36 overexpressing mice have decreased blood
levels of TAG and FA’s (Stremmel, 1988), whereas CD36 knockout mice have increased
fasting levels of FFA and TAG in serum and reduced uptake of FFA in isolated
adipocytes (Ailhaud, 1993). In humans, CD36 deficiency has been reported to be linked

with heart muscle hypertrophy or dysfunction (Shaffer, 2001).

FATP family conserved from mycobacterium to mammals

A member of the FATP family called mus musculus fatty acid transport protein 1

(mmFATPl) was first identified using a technique called expression cloning (Schaffer

26

and Lodish, 1994). An adipocyte cDNA library was expressed in COS7 cells and the
cells were screened for the internalization of a fluorescent fatty acid analog. FATPl is a
646 amino acid integral plasma membrane protein with a long hydr0phobic amino-
terrninal region that faces the extracellular space. FATPl has two domains that have
sequence similarities to long-chain acyl—CoA synthetases and very long-chain acyl-CoA
synthetase (Uchiyama et al., 1996; Schaffer, 2001). Hirsch et al. (1998) reported the
presence of 4 other murine homologues and the human homologues of all 5 members
plus a sixth member. F ATP homologues were also present in C. elegans, M. tuberculosis

and Drosophila melanogaster (Hirsch et al., 1998).

Transfection experiments of mmFATPl, mmFATP2, mmFATPS into 3T3 fibroblasts and
COS cells, resulted in increased FA and FA analogue uptake in the transfected cells
(Schaffer and Lodish, 1994; Hirsch et al., 1996). Non-mammalian homologues also
exhibited FA transport activity when transfected to COS cells and E. coli (Schaffer,
2001). The expression of FATPl increases during adipose development (Man et al.,
1996). A Human homologue of FATP4 (hsFATP4) shows polarized distribution in
enterocytes, with highest levels on the luminal side of the cell, indicating a possible role
in dietary FA uptake from intestinal lumen. Antisense depletion of FATP4 significantly
reduces LCFA import in murine enterocytes (Stahl et al., 1999). Insulin down regulated
whereas fasting upregulated mmFATPl mRNA expression in mouse adipose tissue (Man

et al., 1996).

Other mammalian FA binding groteins

27

Caveolin, a 22kDa integral membrane protein, was identified by using a photoreactive
FA analogue from 3T3-L1 preadipocytes (Trigatti et al., 1991, 1993). Caveolin showed
high affinity for FA’s, and mRNA expression was increased during adipocyte
differentiation. Later it was reported to be a structural component of caveolae, the flask-
shaped invaginations of the cell membrane which play a role in endo- and exocytosis
(Mangroo et al., 1995; Trigatii et al., 1999). The functional aspect of caveolin in FA
transport is not clearly distinguished from its role in endocytosis (Frohner and Bemlohr,

2000).

The 56 kDa renal FA binding protein was purified to homogeneity from rat renal
basolateral and cardiac myocyte membranes by a series of purification steps including
oleate-Sepharose 4B affinity chromatography (Fujii et al., 1987). However, no follow-up
experiments showing its role in FA transport have been reported (Hui and Bemlohr,

1997).

Activated FA transporters

Mitochondrial carnitine system

FA oxidation in animal cells takes place in peroxisomes and in mitochondria matrix. The
FA entry across the double membrane of mitochondria is achieved by the well-studied
camitine system. The mitochondrial camitine system consists of at least 3 enzyems;
camitine palmitoyltransferase I (CPT-I), camitinezacyl camitine translocase (CACT) and

camitine palmitoyltransferase II (CPT-II) (Kemer and Hoppel, 2000). FA’s from cytosol

28

are first activated to acyl-CoA by a long-chain acyl-CoA synthetase located on the
mitochondria outer membrane (Kemer and Hoppel, 2000). Long-chain acyl groups are
transferred to free camitine to form long-chain acyl-carnitine, catalyzed by the outer
membrane-localized CPT-I (McGarry and Brown, 1997). Long—chain acyl-camitines are
then translocated into the mitochondrial matrix in an exchange reaction catalyzed by an
integral inner membrane protein, CACT (Indiveri et al., 1990, 1995). Within the matrix
the long-chain acyl-camitines are reconverted to long-chain acyl-CoA by CPT-II. This
enzyme is anchored to the inner membrane as a peripheral membrane protein (Bieber,

1988).

CPT-I represents a key regulatory site controlling the flux through B-oxidation. CPT-I is
inhibited by malonyl-CoA but under starvation and diabetic conditions becomes less
sensitive to malonyl-CoA, allowing maximal FA oxidation (McGarry and Foster, 1979;
Drynan et al., 1996). CPT-I shows developmental stage-specific expression and is
subject to transcriptional controls by exogenous FA treatments, fasting, insulin. The
posttranslational modification-dependent interaction with cytoskeletal components are
also implicated in the regulation of CPT-l activity (Thumelin et al., 1994; Park et al.,

1995; Chatelain et al., 1996; Guzman et al., 1994; Velasco et al., 1998).

Peroxisomal LCFA membrane transporters

In yeast, FA oxidation takes place exclusively in peroxisomes (Hettema and Tabak,
2000). The LCFA’s are activated to the long-chain acyl-CoA on the cytosolic side of the

peroxisome membrane (Johnson et al., 1994; Hettema et al., 1996). Activated long-chain

29

fatty acyl-CoA’s are transported across the peroxisomal membrane by peroxisomal ABC
half transporters, Patlp (also known as an2p) and Pat2p (also know as anlp). Patlp
and Pat2p are shown to form a heteromeric complex (Shani and Valle, 1996). Patlp and
Pat2p are induced upon growth on oleate and are necessary for normal grth on
LCFA’s as a sole carbon source (Hettema et al., 1996; Shani and Valle, 1996). The

PA T I/PA T2 deletion mutants have considerably reduced B-oxidation activity.

Four different peroxisomal ABC half-transporters have been identified in mammals and
some of them are shown to be partially redundant (Braiterrnan et al., 1998). X-Linked
adrenoleukodystrophy patients accumulate VLCFA’s due to a defect in a peroxisomal
ABC transporter (ALDp), and it is likely that this transporter is involved in transport of

VLCFA’s across the peroxisomal membrane (Mosser et al., 1993, 1994).

Membrane lipid transporters

The asymmetric distribution of lipids across the bilayer is a fundamental property of most
biological membranes (Bretscher, 1973; Rothman and Lenard, 1977; Op den Kamp,
1979). Disruption of the asymmetry is part of the normal blood clotting mechanism but
in other cases results in defects in signal transduction pathways, defects in macrophage
recognition of senescing cells, and change in activities of cytoplasmic proteins. A
number of human diseases including Scott syndrome (Bevers et al., 1992), sickle cell
disease (Lubin et al., 1981), stroke (Ciavatti et al., 1978) and diabetes (Wali et al., 1988)

are related to problems in maintaining the homeostasis of this asymmetry.

30

Transbilayer lipid asymmetry is established during biosynthesis of each class of lipid

(van Meer, 1998). Once asymmetry is generated, spontaneous transbilayer lipid flip-flOp

is slow (t1 /2= hours to days) and is inversely proportional to fatty acyl chain length and

hydrophobicity (Kornberg and McConnell, 1971; Middelkoop et al., 1986). Protein-lipid
interactions partly help in preventing certain lipids from traversing to the other side of the
bilayer thus contribute to the maintenance of the asymmetry (Cohen et al., 1988), but a
specialized set of transporters are reported (as reviewed below) to play important roles in

the transbilayer movement of membranelipids.

Human erythrocyte scramblase

The scramblase plays a key role in the stimulation of plasma clotting factors by platelets
and in the recognition of senescent and apoptotic cells by macrophages (Bevers et al.,
1982; Conner et al., 1994). Human erythrocyte and platelet scrarnblases were purified

and subsequently cloned (Bassé et al., 1996; Comfiirius et al., 1996; Zhou et al., 1997).

. . . . . . . 2+ .
Thrs protern catalyzes the extemalrzatron of PS 1n response to nsmg Ca concentratlons

during cell activation which results in the lipid-randomizations. This protein is an
integral membrane protein with potential palmitoylation sites and a proline-rich
cytoplasmic N-terminus (Zhao et al., 1998; Zhou et al., 1998). The size (35kDa) and
predicted membrane topology lead to a speculation that this protein is not a monomeric
transporter but may form either a functional oligomer or act as a subunit of a larger

complex (Daleke and Lyles, 2000).

31

Aminophospholipid translocase

Energy-dependent, cytofacially directed PS and PE transport activity was reported and

confirmed in human erythrocytes (Seigneuret and Devaux, 1984; Loh and Huestis, 1993).

. . .. 2+ . .
This activity requires ATP and 1S sensrtrve to Ca , vanadate, N-ethylmalermrde, and

exhibits specificity for PS. Some purified proteins, including human erythrocyte Rh

. 2+
protein, human erythrocyte PS-stlmulated Mg ATPase, and chromaffin granule

ATPase 11 fit into this category and are thus candidates of aminophopholipid translocase.

The Rh protein was found by its selective reactivity with a radiolabeled sulflrydryl
reagent and with PS photoaffinity probe (Connor and Schroit, 1988). However, direct

evidence is still lacking that this protein is indeed an aminophospholipid translocase

2+
(Schroit and Zwaal, 1991). The PS-stimulated Mg -ATPase was partially purified from

human erythrocyte following the biochemical properties of aminophospholipid

. . 2+ . .
translocase, namely, activation by PS, and the Mg -ATPase activrty (Morrot et al.,

1990; Zimmerman and Daleke, 1993; Auland et al., 1994). The partially purified samples
were able to catalyze ATP-dependent movement of PS and PE in the reconstitution

experiments (Auland et al., 1994). The bovine chromaffin granule ATPaseII is

. . . . 2+ .
functionally srmrlar to the erythrocyte PS-strmulated Mg -ATPase, and was purified

and cloned (Moriyama and Nelson, 1988; Tang et al., 1996). This protein contains P-
type ATPase consensus sequences with 10 transmembrane spanning domains. DrsZp

from yeast when mutated shows aminophospholipid transport defect phenotype and has

32

sequence similarity to ATPaseII suggesting the possible role of ATPaseII in the

aminophospholipid transport (Tang et al., 1996).

Multidrug Resistance Proteins

The phosphatidylcholine (PC) translocator was first found in mouse null mutant of Mer
gene. The homozygous knockout mouse can not secrete PC into bile, and the absence of
PC in bile resulted in mild liver disease in mice. Deficiency in human Mer homologue
MDR3 (sometimes called MDR2), leads to a serious liver disease requiring
transplantation (Smit et al., 1993; Schinkel et al., 1997). Both Mdr2 of mouse and human
MDR3 encode P-glycoproteins (ng) which belong to the ATP binding cassette (ABC)
transporter family. The direct evidence of Mer involvement in PC transport was
demonstrated in a yeast mutant transfected with an Mdr2 cDNA construct (Ruetz and
Gros, 1994). The Mdr2 protein was able to catalyze the direct translocation of a PC
analogue and displayed high specificities for phospholipid analogues with a PC head
group. MDRl is also able to translocate lipid analogues. However, they show specificity

toward a wide range of molecules and their function as lipid transporters is still unclear.

Another set of ABC transporters known as multidrug resistance associated proteins
(MRP’s) are also able to transport lipid analogues (Cole et al., 1992; Kool et al., 1997).
MRPl is an organic anion transporter with high activity toward glutathione conjugated
compounds. The physiological evidence of MRP1 involvement in lipid metabolism is still

lacking (van Helvoort et al., 1996).

33

ABCR (or Rim protein), a distant relative of ng, was found to be defective in age-
related macular degeneration called Stargardt’s macular dystrophy (Allikmets et al.,
1997). ABCR appears to function as an outwardly-directed flippase for N—retinylidene-

PE.

ABCl, another distant relative of ngs, is linked with atherosclerosis (Young and
Fielding, 1999). This protein is one of the largest ABC transporters known with 2201
amino acids. When mutated it results in familial high density lipoprotein deficiency and
when is completely absent rapid degradation of high density lipoproteins in plasma
membrane leads to a disease known as Tangier disease (Rust et al., 1999). Further
research indicated that ABCl is responsible for cholesterol efflux from cells (Borst et al.,

2000)

MsbA is a bacterial Mdr homolog (Karow and Georgopoulous, 1993). Lipid A, a hexa-
acylated disaccharide of glucosamine, is the hydrophobic anchor of lipopolysaccharide
that covers the outer membrane of Gram-negative bacteria. The enzymes that make lipid
A are located in the cytoplasm or inner membrane. MsbA is demonstrated to be an
essential component of lipid A transport across the inner membrane and possibly to the
outer membrane (Polissi and Georgopoulos, 1996; Zhou et al., 1998; Doerrler et al.,

2001)

Interorganellor phospholipids transport via contact zone

34

PS can serve as a precursor for PE and PC synthesis in eukaryotes. PS is synthesized in
the ER but PS decarboxylase is found in mitochondria and Golgi-vacuole (Voelker,
1989). PB synthesized by PS decarboxylase then returns to ER where methylation occurs
to form PC. A specialized zone of ER called mitochondria-association membrane
(MAM) was implied to be involved in these exchanges of phospholipids (Vance, 1990;
Gaigg et al., 1995). Close association between Golgi and ER, and ER and plasma
membrane was also observed (March et al., 2001; Pichler et al., 2001). The MAM
fraction is found to be enriched in PS synthase. Direct evidence that MAM or other
contact zones are involved in interogranllor transfer of lipids is still lacking (Voelker,
2003). Genetic analyses in yeast lead to the finding of mutants (pstAI) defective in
mitochondrial PS transport (Schumacher et al., 2002). The MET 30 gene complemented
the deficiency in the mutant. MET 30 encodes a subunit of a multicomponent ubiquitin
ligase. There are increasing evidences of ubiquitination role in membrane trafficking

regulation (Hicke, 2001).

Lipid transporters in plants

There is almost no physical biochemistry work documented in plants concerning

unesterified FA transport. However, some proteins that are implicated to be involved in

several aspects of lipid transport in plant cells are beginning to be discovered, mostly

through genetic analyses.

35

PM] (also discovered independently and named PED3 or C T S by separate groups) is an
Arabidopsis version of the yeast peroxisomal long-chain acyl-CoA transporters Patlp and
Pat2p, as well as the human peroxisomal ABC transporter (ALDP) (Zolman et al., 2001;
Footitt et al., 2002; Hayashi et al., 2002). The pxaI mutant displays B-oxidation defect
phenotypes. In addition this mutant was found to accumulate very long-chain acyl-CoA’s
during the germination on sucrose supplied media. This and other data suggested that the
VLCFA’s are transported into the peroxisomes as activated acyl-CoA, consistent with the

observations made in yeast and mammals (Hettema et al., 1996; Mannaerts et al., 1982).

Thylakoid membrane components are synthesized in the inner and outer membrane of
chloroplasts. Vesiculation of the inner membrane seems to be the carrier of membrane
elements for thylakoid membrane biogenesis and maintenance (Morré et al., 1991; Kroll
et al., 2001). A pea 37 kDa protein, IM30, was found both in the chloroplast envelope
and in the thylakoid, and was hypothesized to be a candidate for the transfer of
galactolipids from the envelope to thylakoid (Li et al., 1994). Lack of vesicle formation
was paralleled by the inhibition of thylakoid formation in an Arabidopsis mutant
disrupted in the Arabidopsis homolog of IM30, the vesicle-inducing proteins in plastids 1
(VIPPI) (Kroll et al., 2001). Deletion of the VIPPI homolog in cyanobacteria also
resulted in a complete loss of thylakoid formation (Westphal et al., 2001). The precise
role of VIPPl in the cycle of vesicle budding, migration, and firsion remains to be

explored.

36

Xu et al. (2003) reported the finding of a mutant (trigalactosyldiacylglycerol1 (tgd1))
defective in the biosynthesis of ER-derived thylakoid lipids. The ER pathway for the
glycerolipid synthesis involves the return of DAG moieties from the ER to the
chloroplast for synthesis of galactolipids and sulfolipids. When pulse labeled, the tgd]
plant had reduced flux through the ER pathway of thylakoid lipid biosynthesis. The
cloned gene encoded a permease-like protein which was localized to the chloroplast outer

membrane, and was proposed to be a component of a lipid transfer complex.

Aminophospholipid ATPasel (ALAI) in Arabidopsis is a member of at least 11 P-type
ATPases and is homologous with yeast DRS2 and bovine ATPaseII aminophospholipid
translocases (Tang et al., 1996; Ding et al., 2000; Gomes et al., 2000). ALAI
complements the drs2 phenotype of deficiency in PS internalization, and when
overexpressed increased translocation of aminophospholipids in reconstituted yeast
membrane vesicles (Gomes et al., 2000). One of the physiological functions of this
putative aminophopholipid translocation seems to be linked to cold adaptation of plants

since the Arabidopsis ALA 1 antisense plants showed chilling sensitivity.

Very long-chain fatty acyl derivatives of wax components have to be transported from
their site of elongation and modification, the ER to the plasma membrane. After crossing
the hydrophobic plasma membrane, they have to partition to the aqueous apoplast.
Finally, the hydrophobic wax components must move through the hydrophilic cell wall
matrix to reach the cuticle (Kunst and Samuels, 2003). Unpublished works by Samuels

AL (Department of Botany, University of British Columbia, 6270 University Boulevard,

37

Vancouver, BC, Canada V6T lZ4) suggested the involvement of a member of the ABC

transporter family in the export of wax components from the cytosol.

The above review of lipid transport research in various organisms gives a general
overview of the underlying issues and previous research on this topic. Although plant
lipid metabolism in many ways differs from that of animal or unicellular organisms, the
firndamental requirement for fatty acid movement is common for all eukaryotic and many
prokaryotic organisms. The fact that the newly identified lipid transporters in plants have
their counterpart genes in other organisms shows that at least some members of the

transport system are of ancient origin and conserved throughout biology.

In this thesis three major aspects of fatty acid transport and metabolism are presented.
First, Chapter 2 presents various kinetic labeling data to examine whether the export of
FFA from plastids occurs via free physical diffusion or is mediated by a more complex
system. Second, in Chapter 3 the question of movement of FA in the other direction, that
is into the chloroplast is described and the compartmentation and the mechanism of
exogenous FFA elongation activity in a plant cell is characterized. Third, Chapter 4
presents bioinformatics and reverse genetics approaches in search for plastid envelope
lipid transporters. Finally, in Chapter 5, perspectives and directions for future research in

this area are discussed.

38

LITERATURE CITED

Abumrad NA, el-Maghrabi MR, Amri EZ, Lopez E and Grimaldi PA (1993)
Cloning of a rat adipocyte membrane protein implicated in binding or transport of long-
chain fatty acids that is induced during preadipocyte differentiation. J. Biol. Chem. 268:
17665-17668.

Abumrad NA, Park JH and Park CR (1984) Permeation of long-chain fatty acid into
adipocytes. Kinetics, specificity, and evidence for involvement of a membrane protein. J.
Biol. Chem. 259: 8945-8953.

Abumrad NA, Perkins RC, Park JH and Park CR (1981) Mechanism of long chain
fatty acid permeation in the isolated adipocyte. J Biol Chem. 256: 9183-9191.

Ailhaud G (1993) Regulation of gene expression by fatty acids in the adipose cell.
Prostaglandins Leukot. Essent. Fatty Acids 48: 89-90.

Alexandre J and Lassalles JP (1992) Intracellular Ca2+ release by InsP3 in plants and
effects of buffers on Ca2+diffusion. Philos. trans. R. Soc. Lond. Biol. Sci. 338: 53-61.

Allikmets R, Singh N, Sun H, Shroyer NF, Hutchinson A, Chidambaram A,
Gerrard B, Baird L, Stauffer D, Pfeiffer A, Rattner A, Smallwood P, Li Y,
Anderson KL, Lewis RA, Nathans J, Leppert M, Dean M and Lupski JR (1997) A
photoreceptor cell-specific ATP-binding transporter gene (ABCR) is mutated in recessive
Stargardt macular dystrophy. Nat. Genet. 15: 236—245.

Antonenko YN, Denisov GA and Pohl P (1993) Weak acid transport across bilayer
lipid membrane in the presence of buffers. Theoretical and experimental pH profiles in
the unstirred layers. Biophys. J . 64: 1701-1710.

Auland ME, Roufogalis BD, Devaux PF and Zachowski A (1994) Reconstitution of
ATP-dependent aminophospholipid translocation in proteoliposomes. Proc. Natl. Acad.
Sci. USA 91: 10938—10942.

Azizan A and Black PN ( 1994) Use of transposon TnphoA to identify genes for cell
envelope proteins of Escherichia coli required for long-chain fatty acid transport: the

periplasmic protein Tsp potentiates long-chain fatty acid transport. J. Bacteriol. 176:
6653-6662.

Baillie AGS, Coburn CT, and Abumrad NA (1996) Reversible binding of long-chain
fatty acids to purified FAT, the adipose CD36 homolog. J. Membr. Biol. 153: 75-81.

Bassé F, Stout JG, Sims PJ and Wiedmer T (1996) Isolation of an erythrocyte

membrane protein that mediates Ca2+-dependent transbilayer movement of phospholipid.
J. Biol. Chem. 271: 17205—17210.

39

Berk PD and Stump DD (1999) Mechanisms of cellular uptake of long chain free fatty
acids. Mol. Cell. Biochem. 192: 17-31.

Berk PD, Zhou SL, Kiang CL, Stump DD and Bradbury MW (1999) Selective up-
regulation of fatty acid uptake by adipocytes characterizes both genetic and diet-induced
obesity in rodents. J. Biol. Chem. 274: 28626-28631.

Berk PD, Zhou SL, Kiang CL, Stump D, Bradbury M and Isola LM (1997) Uptake
of long chain free fatty acids is selectively upregulated in adipocytes of Zucker rats with
genetic obesity and non-insulin dependent diabetes mellitus. J. Biol. Chem. 272: 8830-
8835.

Bernhard WR, Thoma S, Botella J and Somerville CR (1991) Isolation of a cDNA
clone for spinach lipid transfer protein and evidence that the protein is synthesized by the
secretory pathway. Plant Physiol. 95: 164-170.

Bevers EM, Comfurius P, van Rijn JL, Hemker HC and Zwaal RF (1982)
Generation of prothrombin-converting activity and the exposure of phosphatidylserine at
the outer surface of platelets. Eur. J. Biochem. 122: 429—436.

Bevers EM, Wiedmer T, Comfurius P, Shattil SJ, Weiss HJ, Zwaal RF and Sims PJ
(1992) Defective Ca(2+)-induced microvesiculation and deficient expression of
procoagulant activity in erythrocytes from a patient with a bleeding disorder: a study of
the red blood cells of Scott syndrome. Blood 79: 380—388.

Bieber LL (1988) Carnitine. Annu. Rev. Biochem. 57: 261—283.

Black PN (1990) Characterization of F adL-specific fatty acid binding in Escherichia coli.
Biochim. Biophys. Acta. 1046: 97-105.

Black PN, DiRusso CC, Metzger AK, and Heimert TL (1992) Cloning, sequencing,
and expression of the fadD gene of Escherichia coli encoding acyl coenzyme A
synthetase. J. Biol. Chem. 267: 25513-25520.

Black PN and Zhang Q (1995) Evidence that Hisl 10 of the protein FadL in the outer
membrane of Escherichia coli is involved in the binding and uptake of long-chain fatty
acids: possible role of this residue in carboxylate binding. Biochem. J. 310: 389-394.

Bonen A, Luiken FP, Liu S, Dyck DJ, Kiens B, Kristiansen S, Turcotte LP, Van der
Vusse GJ and Glatz JF (1998) Palmitate transport and fatty acid transporters in red and
white muscles. Am. J. Physiol. 275:E471-E478.

Borst P, Zelcer N and van Helvoort A (2000) ABC transporters in lipid transport.
Biochim. Biophys. Acta. 1486:128-144.

40

Braiterman LT, Zheng S, Watkins PA, Geraghty MT, Johnson G, McGuinness MC,
Moser AB and Smith KB (1998) Suppression of peroxisomal membrane protein defects
by peroxisomal ATP binding cassette (ABC) proteins. Hum. Mol. Genet. 7 :239—247.

Bretscher MS (1973) Membrane structure: some general principles. Science 181: 622—
629.

Broun P, Gettner S and Somerville C ( 1999) Genetic engineering of plant lipids. Annu.
Rev. Nutr. 19: 197—216.

Browse J, McConn M, James D Jr and Miquel M (1993) Mutants of Arabidopsis
deficient in the synthesis of alpha-linolenate. Biochemical and genetic characterization of
the endoplasmic reticulum linoleoyl desaturase. J. Biol. Chem. 268: 16345-16351.

Browse J and Somerville CR (1991) Glycerolipid metabolism: Biochemistry and
regulation. Annu. Rev. Plant Physiol. Plant Mol. Biol. 42: 467-506.

Browse J and Somerville CR ( 1994) Glycerolipids. In Arabidopsis. Meyerowitz EM,
and Somerville CR ed. (Cold Spring Harbor Laboratory Press, New York) pp. 881-912.

Cahill Jr. GF (1970) Starvation in man. N. Engl. J. Med. 282:.668-675.

Chatelain F, Kohl C, Esser V, McGarry JD, Girard J and Pegorier JP (1984) Cyclic
AMP and fatty acids increase camitine pahnitoyltransferase I gene transcription in
cultured fetal rat hepatocytes. Eur. J. Biochem. 235: 789—798.

Cheesbrough TM and Kolattukudy PE (1984) Alkane biosynthesis by decarbonylation
of aldehydes catalyzed by a particulate preparation from Pisum sativum. Proc. Natl.
Acad. Sci. USA 81: 6613-6617.

Ciavatti M, Michel G and Dechavanne M (1978) Platelet phospholipid in stroke. Clin.
Chim. Acta. 84: 347—351.

Clementz T, Bednarski JJ and Raetz CRH (1996) Function of the htrB High
Temperature Requirement Gene of Escherichia coli in the Acylation of Lipid A: HtrB
CATALYZED INCORPORATION OF LAURATE. J. Biol. Chem. 271: 12095-12102.

Cohen AM, Liu SC, Lawler J, Derick L, Palek J (1988) Identification of the protein
4.1 binding site to phosphatidylserine vesicles. Biochemistry 27: 614-619.

Cole SPC, Bhardwaj G, Gerlach JH, Mackie JE, Grant CE, Almquist KC, Stewart

AJ, Kurz EU, Duncan AMV and Deeley GR (1992) Overexpression of a transporter
gene in a multidrug-resistant human lung cancer cell line. Science 258: 650—1654.

41

Comfurius P, Williamson P, Smeets EF, Schlegel RA, Bevers EM and Zwaal RF
(1996) Reconstitution of phospholipid scramblase activity from human blood platelets.
Biochemistry 35: 7631—7634.

Connor J, Pak CC and Schroit AJ (1994) Exposure of phosphatidylserine in the outer
leaflet of human red blood cells. Relationship to cell density, cell age, and clearance by
mononuclear cells. J. Biol. Chem. 269: 2399—2404.

Connor J and Schroit AJ (1988) Transbilayer movement of phosphatidylserine in
erythrocytes: inhibition of transport and preferential labeling of a 31,000-dalton protein
by sulfliydryl reactive reagents. Biochemistry 27 : 848—85 1.

Cote GG and Crain RC (1993) Biochemistry of phosphoinositides. Annu. Rev. Plant
Physiol. Plant Mol. Biol. 44: 333-356.

Dahlqvist A, Stahl U, Lenman M, Banas A, Lee M, Sandager L, Ronne H, Stymne S.
Phospholipid:diacylglycerol acyltransferase: an enzyme that catalyzes the

acyl-CoA-independent formation of triacylglycerol in yeast and plants.

Proc Natl Acad Sci U S A. 2000 Jun 6;97(12):6487-92.

Daleke DL and Lyles JV (2000) Identification and purification of aminophospholipid
flippases. Biochim. Biophys. Acta. 1486: 108-127.

Ding J, Wu Z, Crider BP, Ma Y, Li X, Slaughter C, Gong L, and Xie XS (2000)
Identification and firnctional expression of four isoforrns of ATPase II, the putative
aminophospholipid translocase: Effect of isoform variation on the ATPase activity and
phospholipid specificity. J. Biol. Chem. 275: 23378-23386.

Doermann P, Voelker TA and Ohlrogge JB (1995) Cloning and expression in
Escherichia coli of a novel thioesterase from Arabidopsis thaliana specific for long-chain
acyl-acyl carrier proteins. Arch. Biochem. Biophys. 316: 612-618.

Doerrler WT, Reedy MC and Raetz CR (2001) An Escherichia coli mutant defective in
lipid export. J. Biol. Chem. 276:11461- 11464.

Drynan L, Quant PA and Zammit VA (1996) The role of changes in the sensitivity of
hepatic mitochondrial overt camitine palmitoyltransferase in determining the onset of the
ketosis of starvation in the rat. Biochem. J. 318: 767—770.

Dutta-Roy AK, Gordon MJ, Campbell L and Crosbie L (1996) Arachidonic acid
uptake by human platelets is mediated by CD36. Platelets. 7: 291-295.

Faergeman NJ, DiRusso CC, Elberger A, Knudsen J, Black PN (1997) Disruption of

the Saccharomyces cerevisiae homologue to the murine fatty acid transport protein
impairs uptake and growth on long-chain fatty acids. J. Biol. Chem. 272: 8531-8538.

42

Faergeman NJ, Black PN, Zhao XD, Knudsen J, DiRusso CC (2001) The Acyl-CoA
synthetases encoded within FAAl and F AA4 in Saccharomyces cerevisiae fimction as
components of the fatty acid transport system linking import, activation, and intracellular
Utilization. J. Biol. Chem. 276: 37051-37059.

Farmer EE and Ryan CA (1990) Interplant communication: Airborne methyl jasmonate
induces synthesis of proteinase inhibitors in plant leaves. Proc. Natl. Acad. Sci. USA 87:
7713-7716.

Fehling E, Mukherjee KD (1991) Acyl-CoA elongase from a higher plant (Lunaria
annua): metabolic intermediates of very-long-chain acyl-CoA products and substrate
specificity. Biochim. Biophys. Acta. 1082: 239-246.

Footitt S, Slocombe SP, Larner V, Kurup S, Wu Y, Larson T, Graham 1, Baker A,
Holdsworth M (2002) Control of germination and lipid mobilization by COMATOSE,
the Arabidopsis homologue of human ALDP. EMBO J. 21: 2912-2922.

Frentzen M (1993) Acyltransferases and triacylglycerols. In Lipid Metabolism in Plants,
TS Moore Jr. ed. (Boca Raton, FL: CRC Press), pp. 195-231.

Frohnert BI and Bernlohr DA (2000) Regulation of fatty acid transporters in
mammalian cells. Prog. Lipid Res. 39: 83-107.

Fujii S, Kawaguchi H and Yasuda H (1987) Isolation and partial characterization of an
amphiphilic 56-kDa fatty acid binding protein from rat renal basolateral membrane. J.
Biochem. (Tokyo) 101: 679—684.

Fujii S, Kawaguchi H and Yasuda H (1987) Purification of high affinity fatty acid
receptors in rat myocardial sarcolemmal membranes. Lipids 22: 544—546.

Gaigg B, Simbeni R, Hrastnik C, Paltauf F and Daum G (1995) Characterization of a
microsomal subfraction associated with mitochondria of the yeast, Saccharomyces
cerevisiae. Involvement in synthesis and import of phospholipids into mitochondria.

Biochim. Biophys. Acta. 1234: 214—220.

Gerhardt B (1992) Fatty acid degradation in plants. Prog Lipid Res 31: 417-446.

Gerber GE, Mangroo D and Trigatti BL (1993) Identification of high affinity
membrane-bound fatty acid-binding proteins using a photoreactive fatty acid. Mol. Cell
Biochem. 123: 39—44.

Gomes E, Jakobsen MK, Axelsen KB, Geisler M and Palmgren MG (2000) Chilling
tolerance in Arabidopsis involves ALAl, a member of a new family of putative

aminophospholipid translocases. Plant Cell 12: 2441-2454.

Gunstone FD (2001) Soybeans pace boost in oilseed production. Inform. 11: 1287—1289.

43

Guzman M, Kolodziej MP, Caldwell A, Corstorphine CG and Zammit V (1994)
Evidence against direct involvement of phosphorylation in the activation of camitine
palmitoyltransferase by okadaic acid in rat hepatocytes. Biochem. J. 300: 693—699.

Hamilton JA (1998) Fatty acid transport: difficult or easy? J. Lipid Res. 39: 467-481.

Hamilton JA and Kamp F (1999) How are free fatty acids transported in membranes? Is
it by proteins or by free diffusion through the lipids? Diabetes 48: 2255-2269.

Hamilton JA and Cistola DP (1986) Transfer of oleic acid between albumin and
phospholipid vesicles. Proc. Natl. Acad. Sci. USA. 83: 82-86.

Hamilton JA, Civelek VN, Kamp F, Tornheim K, and Corkey BE (1994) Changes in
internal pH caused by movement of fatty acids into and out of clonal pancreatic beta-cells
(HIT). J. Biol. Chem. 269: 20852-20856.

Harmon CM and Abumrad NA (1993) Binding sulfosuccinimidyl fatty acids to
adipocyte membrane proteins: isolation and amino-terminal sequence of an 88-kD protein
implicated in transport of long-chain fatty acids. J. Membr. Biol. 133: 43-49.

Hayashi M, Nito K, Takei—Hoshi 1R, Yagi M, Kondo M, Suenaga A, Yamaya T and
Nishimura M (2002) Ped3p is a peroxisomal ATP-binding cassette transporter that
might supply substrates for fatty acid B-oxidation. Plant Cell Physiol. 43: 1—11.

Hettema EH and Tabak HF (2000) Transport of fatty acids and metabolites across the
peroxisomal membrane. Biochim. Biophys. Acta. 1486: 18-27.

Hettema EH, Van Roermund CWT, Distel B, Van den Berg M, Vilela C, Rodrigues-
Pousada C, Wanders RJA and Tabak HF (1996) The ABC transporter proteins Patl
and Pat2 are required for import of long-chain fatty acids into peroxisomes of
Saccharomyces cerevisiae. EMBO J. 15 : 3813-3822.

Hicke L (2001) A new ticket for entry into budding vesicles—ubiquitin. Cell 106: 527—
530.

Hirsch D, Stahl A and! Lodish HF (1998) A family of fatty acid transporters conserved
from mycobacterium to man. Proc. Natl. Acad. Sci. USA. 95: 8625-8629.

Hui TY and Bernlohr DA (1997) Fatty acid transporters in animal cells. Front. Biosci.
2: d222-231.

Ibrahimi A, Sfeir Z, Magharaie H, Amri EZ, Grimaldi PA and Abumrad NA (1996)

Expression of the CD36 homolog (FAT) in fibroblast cells: effects on fatty acid transport.
Proc. Natl. Acad. Sci. USA. 93:2646-2651.

44

Indiveri C, Tonazzi A, Giangregorio N and Palmieri F (1995) Probing the active site
of the reconstituted camitine carrier from rat liver mitochondria with sulthydryl reagents.
Eur. J. Biochem. 228: 271—278.

Indiveri C, Tonazzi A and Palnrieri F (1990) Identification and purification of the
camitine carrier from rat liver mitochondria. Biochim. Biophys. Acta. 1020: 81—86.

Isola LM, Zhou SL, Kiang CL, Stump DD, Bradbury MW and Berk PD (1995) 3T3
fibroblasts transfected with a cDNA for mitochondrial aspartate aminotransferase express
plasma membrane fatty acid-binding protein and saturable fatty acid uptake. Proc. Natl.
Acad. Sci. USA 92: 9866-9870.

Johnson DR, Knoll LL, Levin DE and Gordon J] (1994) Saccharomyces cerevisiae
contains four fatty acid activation (FAA) genes: As assessment of their role in regulating
protein N-myristoylation and cellular lipid metabolism. J Cell Biol. 127:751-762.

Kameda K, Suzuki LK and Imai Y (1987) Transport of fatty acid is obligatory coupled
with HJr entry in pher0plasts of Escherichia coliK-12. biochem. Int. 14: 227-234.

Kamp F, Guo W, Souto R, Pilch PF, Corkey BE and Hamilton JA (2003) Rapid flip-
flop of oleic acid across the plasma membrane of adipocytes. J. Biol. Chem. 278: 7988-
7995.

Kamp F and Hamilton JA (1992) pH gradients across phospholipid membranes caused
by fast flip-flop of un-ionized fatty acids. Proc. Natl. Acad. Sci. USA. 89: 11367-11370.

Kamp F and Hamilton JA (1993) Movement of fatty acids, fatty-acid analogues and
bile acids across phospholipid bilayers. Biochemistry 32:11074-11086.

Kamp F, Hamilton JA and Westerhoff HV ( 1993) Movement of fatty acids, fatty acid
analogues, and bile acids across phospholipid bilayers. Biochemistry 32, 11074-11086.

Kamp F, Zakim D, Zhang F, Noy N, Hamilton JA (1995) Fatty acid flip-flop in
phospholipid bilayers is extremely fast. Biochemistry 34: 11928-11937.

Karow M, Georgopoulos C (1993) The essential Escherichia coli msbA gene, a
multicopy suppressor of null mutations in the htrB gene, is related to the universally
conserved family of ATP-dependent translocators. Mol. Microbiol. 7: 69-79.

Kiens B, Kristiansen S, Jensen P, Richter EA and Turcotte LP (1997) Membrane
associated fatty acid binding protein (FABPpm) in human skeletal muscle is increased by
endurance training. Biochem. Biophys. Res. Commun. 231: 463-465.

Kim HK and Storch J (1992) Mechanism of free fatty acid transfer from rat heart fatty

acid-binding protein to phospholipid membranes. Evidence for a collisional process. J.
Biol. Chem. 267: 20051-20056.

45

 

 

 

KleinfeldAM(2000) Lipid phase fatty acid flip-flop, is it fast enough for cellular
transport? J. Membr. Biol. 175: 79-86.

Kleinfeld AM, Chu P and Romero C ( 1997) Transport of long-chain native fatty acids
across lipid bilayer membranes indicates that transbilayer flip-flop is rate limiting.
Biochemistry 36: 14146-14158.

Kleinfeld AM and Storch J (1993) Transfer of long-chain fluorescent fatty acids
between small and large unilamellar vesicles. Biochemistry 32: 2053-2061.

Kleinfeld A, Storms S and Watts M (1998) Transport of long-chain native fatty acids
across human erythrocyte ghost membranes. Biochemistry 37: 8011-8019.

Kohlwein DS and Paltauf F (1983) Uptake of fatty acids by the yeasts Saccharomyces
uvarum and Saccharomycopsis lipolytica. Biochim. Biophys. Acta. 792: 310-317.

Kolattukudy PE (2001) Polyesters in higher plants. Adv. Biochem. Eng. Biotechnol. 71:
1-49.

Kool M, De Haas M, Scheffer GL, Scheper RJ, Van Eijk MJT, Juijn JA, Baas F and
Borst P (1997) Analysis of expression of cMOAT (MRP2), MRP3, MRP4, and MRP5,
homologs of the multidrug resistance-associated protein gene (MRPl), in human cancer
cell lines. Cancer Res. 57: 3537—3547.

Kornberg RD and McConnell HM (1971) Inside-outside transitions of phospholipids in
vesicle membranes. Biochemistry 10: 1111—1120.

Kroll D, Meierhoff K, Bechtold N, Kinoshita M, Westphal S, Vothknecht UC, 8011 J
and Westhoff P (2001) VIPPl, a nuclear gene of Arabidopsis thaliana essential for
thylakoid membrane formation. Proc. Natl. Acad. Sci. USA. 98: 4238-4242.

Kumar GB and Black PN (1991) Linker mutagenesis of a bacterial fatty acid transport
protein. Identification of domains with fimctional importance. J. Biol. Chem. 266: 1348-
1353.

Kumar GB and Black PN (1993) Bacterial long-chain fatty acid transport. Identification
of amino acid residues within the outer membrane protein F adL required for activity. J.
Biol. Chem. 268: 15469-15476.

Kunst L, Browse J and Somerville C (1988) Altered regulation of lipid biosynthesis in
a mutant of Arabidopsis deficient in chloroplast glycerol phosphate acyltransferase
activity. Proc. Natl. Acad. Sci. USA. 85: 4143-4147.

Kunst L and Samuels AL (2003) Biosynthesis and secretion of plant cuticular wax.
Prog. Lipid Res. 42: 51-80.

46

 

Li H, Kaneko Y and Keegstra K (1994) Molecular cloning of a chloroplastic protein
associated with both the envelope and thylakoid membranes. Plant Mol. Biol. 25: 619-
632.

Loh RK and Huestis WH (1993) Human erythrocyte membrane lipid asymmetry:
transbilayer distribution of rapidly diffirsing phosphatidylserines. Biochemistry 32:
11722—11726.

Lubin B, Chin D, Bastacky J, Roelofsen B and Van Deenen LL (1981) Abnormalities
in membrane phospholipid organization in sickled erythrocytes. J. Clin. Invest. 67: 1643—
1649.

Luiken JJFP, van Nieuwenhoven FA, America G, van der Vusse GJ and Glatz JFC
(1997) Uptake and metabolism of palmitate by isolated cardiac myocytes from adult rats:
involvement of sarcolemmal proteins. J. Lipid Res. 38: 745-758.

Man MZ, Hui TY, Schaffer JE, Lodish HF and Bernlohr DA (1996) Mol. Endocrinol.
10: 1021-1028.

Mangroo D and Gerber GE (1993) Fatty acid uptake in Escherichia coli: Regulation by
recruitment of fatty acyl CoA synthetase to the plasma membrane. Biochem. Cell. Biol.
71: 431-443.

Mangroo D, Trigatti BL and Gerber GE (1995) Membrane permeation and
intracellular trafficking of long chain fatty acids: insights from Escherichia coli and 3T3-
L1 adipocytes Biochem. Cell Biol. 73: 223—234.

Mannaerts GP, van Veldhoven P, van Broekhoven A, Vandebroek G and Debeer LJ
(1982) Evidence that peroxisomal acyl-CoA synthetase is located at the cytoplasmic side
of the peroxisomal membrane. Biochem. J. 204: 17—23.

Marsh BJ, Mastronarde DN, Buttle KF, Howell KE and McIntosh JR (2001)
Organellar relationships in the Golgi region of the pancreatic beta cell line, HIT-T15,
visualized by high resolution electron tomography. Proc. Natl. Acad. Sci. USA. 98:
2399—2406.

Massey JB, Bick DH and Pownall HJ (1997) Spontaneous transfer of monoacyl
amphiphiles between lipid and protein surfaces. Biophys. J. 72, 1732-1743.

McIlhinney RAJ (1990) The fats of life: the importance and function of protein
acylation. TIBS 15: 387-391.

Milligan G, Parenti M and Maggee A1 (1995) The dynamic role of palmitylation in
signal transduction. TIBS 20: 181-186.

47

Miquel M and Browse J (1992) Arabidopsis mutants deficient in polyunsaturated fatty
acid synthesis. Biochemical and genetic characterization of a plant oleoyl-
phosphatidylcholine desaturase. J. Biol. Chem. 267: 1502-1509.

Miquel MF and Browse JA (1994) High-Oleate Oilseeds Fail to Develop at Low
Temperature. Plant Physiol. 106: 421-427.

McGarry JD and Brown NF (1997) The mitochondrial camitine palmitoyltransferase
system. Eur. J. Biochem. 244: 1—14.

McGarry JD and Foster DW (1979) In support of the roles of malonyl-CoA and
camitine acyltransferase I in the regulation of hepatic fatty acid oxidation and
ketogenesis. J. Biol. Chem. 254: 8163—8168.

Middelkoop E, Lubin BH, Op den Kamp JA and Roelofsen B (1986) F lip-flop rates
of individual molecular species of phosphatidylcholine in the human red cell membrane.
Biochim. BiOphys. Acta. 855: 421—424.

Mollenhauer HH and Toten C (1970) Plant Physiol. 46: 794-799.

Moriyama Y and Nelson N ( 1988) Purification and properties of a vanadate- and N-
ethylmaleimide-sensitive ATPase from chromaffin granule membranes. J. Biol. Chem.
263: 8521—8527.

Morré DJ, Selldén G, Sundqvist C and Sandelius AS (1991) Stromal low-temperature
compartment derived from the inner membrane of the chloroplast envelope. Plant
Physiol. 97: 1558-1564.

Morrot G, Zachowski A and Devaux PF (1990) Partial purification and
characterization of the human erythrocyte Mg2+ -ATPase A candidate
aminophospholipid translocase. FEBS Lett. 266: 29—32.

Mosser J, Douar AM, Sarde CO, Kioschis P, Feil R, Moser H, Poustka AM, Mandel
JM and Aubourg P (1993) Putative X-linked adrenoleukodystrophy gene shares
unexpected homology with ABC transporters. Nature 361: 726—730.

Mosser J, Lutz Y, Stoeckel ME, Sarde CO, Kretz C, Douar AM, Lopez J, Aubourg
P and Mandel JL (1994) The gene responsible for adrenoleukodystrophy encodes a

peroxisomal membrane protein. Hum. Mol. Genet. 3: 265—271.

Munnik T (2001) Phosphatidic acid: an emerging plant lipid second messenger. Trends
Plant Sci. 6: 227-233.

Nagle JF and Tristram—Nagle S (2000) Structure of lipid bilayers. Biochim. Biophys.
Acta. 1469: 159-195.

48

Neely JR and Morgan HE (1974) Relationship between carbohydrate and lipid
metabolism and the energy balance of heart muscle. Annu. Rev. Physiol. 36: 413-459.

Noy N and Zakim D (1985) Fatty acids bound to unilamellar vesicles as substrates for
microsomal acyl-CoA ligase. Biochemistry 24: 3521-3525.

Nunn WD and Simons RW (1978) Transport of long-chain fatty acids by Escherichia
coli: mapping and characterization of mutants in the fadL gene. Proc. Natl. Acad. Sci.
USA. 75: 3377-3381.

Odaib AA, Shneider BL, Bennett MJ, Pober BL, Reyes-Mujica M, Friedman AL,
Suchy FL and Rinaldo P (1998) A defect in the transport of long chain free fatty acids
associated with acute liver failure. N. Engl. J. Med. 339: 1752-1757.

Op den Kamp JAF (1979) Lipid asymmetry in membranes Annu. Rev. Biochem. 48:
47—71.

Ohlrogge JB (1994) Design of new plant products: Engineering of fatty acid
metabolism. Plant Physiol. 104: 821—826.

Ohlrogge J and Browse J (1995) Lipid Biosynthesis. Plant Cell 7: 957-970.

Ohlrogge JB, Kuhn DN and Stumpf PK (1979) Subcellular localization of acyl carrier
protein in leaf protoplasts of Spinacia olearcea. Proc. Natl. Acad. Sci. USA. 76: 1194-
1 198.

Park EA, Mynatt RL, Cook GD and Kashfi K (1995) Insulin regulates enzyme
activity, malonyl-CoA sensitivity and mRNA abundance of hepatic camitine
palmitoyltransferase-I. Biochem. J. 310: 853—858.

Pichler H, Gaigg B, Hrastnik C, Achleitner G, Kohlwein SD, Zellnig G, Perktold A
and Daum G (2001) A subfraction of the yeast endoplasmic reticulum associates with
the plasma membrane and has a high capacity to synthesize lipids. Eur. J. Biochem. 268:
2351—2361.

Polissi A and Georgopoulos C ( 1996) Mutational analysis and properties of the msbA
gene of Escherichia coli, coding for an essential ABC family transporter. Mol. Microbiol.
20: 1221-1233.

Pollard M and Ohlrogge J (1999) Testing models of fatty acid transfer and lipid
synthesis in spinach leaf using in vivo oxygen-18 labeling. Plant Physiol. 121: 1217-
1226.

Ptak M, Egret-Charier M, Sanson A and Bouloussa O (1980) A NMR study of the

ionization of fatty acids, fatty amines and N-acylamino acids incorporated in
phosphatidylcholine vesicles. Biochim. Biophys. Acta. 600: 387-397.

49

Raetz CRH (1990) Biochemistry of endotoxins. Annu. Rev. Biochem. 59: 129-170.

Richieri G and Kleinfeld A (1995) Unbound free fatty acid levels in human serum. J.
Lipid Res. 36: 229-240.

Richieri GV, Ogata RT and Kleinfeld AM (1995) Thermodynamics of fatty acid
binding to fatty acid-binding proteins and fatty acid partition between water and
membranes measured using the fluorescent probe ADIFAB. J. Biol. Chem. 270: 15076-
15084.

Rothman JE and Lenard J (1977) Membrane asymmetry. Science 195: 743—753.

Roughan PG and Slack CR (1982) Cellular organization of glycerolipid metabolism.
Annu. Rev. Plant Physiol. 33: 97-132.

Ruetz S and Gros P (1994) Phosphatidylcholine translocase: a physiological role for the
mer gene. Cell 77: 1071—1083.

Rust S, Rosier M, Funke H, Real J, Amoura Z, Piette J C, Deleuze JF, Brewer HB,
Duverger N, Denéfle P and Assmann G (1999) Tangier disease is caused by mutations
in the gene encoding ATP-binding cassette transporter 1. Nat. Genet. 22: 352-355.

Schaffer JE (2002) Fatty acid transport: the roads taken. Am. J. Physiol. 282: E239-
E246.

Schaffer JE and Lodish HF (1994) Expression cloning and characterization of a novel
adipocyte long chain fatty acid transport protein. Cell 79: 427-436.

Schinkel AH, Mayer U, Wagenaar E, Mol CAAM, Van Deemter L, Smit JJM, Van
der Valk MA, Voordouw AC, Spits H, Van Tellingen O, Zijlmans JMJM, Fibbe WE
and Borst P (1997) Normal viability and altered pharmacokinetics in mice lacking mdrl-
type (drug-transporting) P-glycoproteins. Proc. Natl. Acad. Sci. USA 94: 4028—4033.

Schroit AJ and Zwaal RF ( 1991) Transbilayer movement of phospholipids in red cell
and platelet membranes. Biochim. Biophys. Acta. 1071: 313—329.

Schumacher MM, Choi JY and Voelker D (2002) Phosphatidylserine transport to the
mitochondria is regulated by ubiquitination. J. Biol. Chem. 277: 5103 3—51042.

Schwieterman W, Sorrentino D, Potter lBJ, Rand J, Kiang CL, Stump D and Berk
PD (1988) Uptake of oleate by isolated rat adipocytes is mediated by a 40-kDa plasma
membrane fatty acid binding protein closely related to that in liver and gut. Proc. Natl.
Acad. Sci. USA. 85: 359-363.

50

 

Seigneuret M and Devaux PF (1984) ATP-dependent asymmetric distribution of spin-
labeled phospholipids in the erythrocyte membrane: relation to shape changes. Proc. Natl.
Acad. Sci. USA. 81: 3751-3755.

Shani N and Valle D (1996) A Saccharomyces cerevisiae homolog of the human
adrenoleukodystrophy transporter is a heterodimer of two half ATP-binding cassette
transporters. Proc. Natl. Acad. Sci. USA. 93: 11901—11906.

Shine WE, Mancha M, Stumpf PK (1976) Fat metabolism in higher plants: the function
of acyl thioesterases in the metabolism of acyl-coenzyme A and acyl-acyl carrier
proteins. Arch. Biochem. Biophys. 172: 110-116.

Smit JJM, Schinkel AH, Oude Elferink RPJ, Groen AK, Wagenaar E, Van Deemter
L, Mol CAAM, Ottenhof R, Van der Lugt NMT, Van Roon M, Van der Valk MA,
Offerhaus GJA, Berns AJM and Borst P (1993) Homozygous disruption of the murine
mer P-glycoprotein gene leads to a complete absence of phospholipid from bile and to
liver disease. Cell 75: 451—462.

Somerville C and Browse J (1991) Plant lipids, metabolism mutants, and membranes.
Science 252: 80-87.

Sorrentino D, Robinson RD, Kiang CL and Berk PD (1989) At physiologic albumin
oleate concentrations oleate uptake by isolated hepatocytes, cardiac myocytes and
adipocytes is a saturable function of the unbound oleate concentration. J. Clin. Invest. 84:
1324-1333

Sorrentino D, Stump D, Potter BJ, Robinson RB, White R, Kiang CL and Berk PD
(1988) Oleate uptake by cardiac myocytes is carrier mediated and involves a 40-kD
plasma membrane fatty acid binding protein similar to that in liver, adipose tissue, and
gut. J. Clin. Invest. 82: 928-935.

Sorrentino D, Stump DD, Van Ness K, Simard A, Schwab AJ, Zhou SL, Goresky
CA and Berk PD (1996) Oleate uptake by isolated hepatocytes and the perfused rat liver
is competitively inhibited by palmitate. Am. J. Physiol. 270: G3 85-G392.

Spector AA, Steinberg D and Tanaka A (1965) Uptake of free fatty acids by Ehrlich
ascites tumor cells. J. Biol. Chem. 240: 1032-1041.

Stahl A, Hirsch DJ, Gimeno RE, Punreddy S, Ge P, Watson N, Patel S, Kotler M,
Raimondi A, Tartaglia LA and Lodish HF (1999) Identification of the major intestinal
fatty acid transport protein. Mol. Cell 4: 299-308.

Sterk P, Booij H, Schellekens GA, Van Kammen A and De Vries SC (1991) Cell-
specific expression of the carrot EP2 lipid transfer protein gene. Plant Cell 3: 907-921.

51

Storch J, Lechene C and Kleinfeld AM (1991) Direct determination of free fatty acid
transport across the adipocyte plasma membrane using quantitative fluorescence
microscopy. J. Biol. Chem. 266: 13473-13476.

Stremmel W (1988) Fatty acid uptake by isolated rat heart myocytes represents a carrier-
mediated transport process. J. Clin. Invest. 81: 844-852.

Stremmel W and Berk PD (1986) Hepatocellular influx of [14C]oleate reflects
membrane transport rather than intracellular metabolism or binding. Proc. Natl. Acad.
Sci. USA. 83: 3086-3090.

Stremmel W, Strohmeyer G and Berk PD (1986) Hepatocellular uptake of oleate is
energy dependent, sodium linked, and inhibited by an antibody to a hepatocyte plasma
membrane fatty acid binding protein. Proc. Natl. Acad. Sci. USA. 83: 3584-3588.

Stremmel W and Theilmann L (1986) Selective inhibition of long-chain fatty acid
uptake in short-term cultured rat hepatocytes by an antibody to the rat liver plasma
membrane fatty acid-binding protein. Biochim. Biophys. Acta. 877: 191-197.

Stump DD, Nunes RM, Sorrentino D, Isola LM and Berk PD (1992) Characteristics
of oleate binding to liver plasma membranes and its uptake by isolated hepatocytes. J.
Hepatol. 16: 304-315.

Stump DD, Fan X and Berk PD (2001) Oleic acid uptake and binding by rat adipocytes
define dual pathways for cellular fatty acid uptake. J. Lipid Res. 42: 509-520.

Stump DD, Zhou SL and Berk PD (1993) Comparison of plasma membrane FABP and
mitochondrial isoforrn of aspartate aminotransferase from rat liver. Am. J. Physiol.
Gastrointest. Liver Physiol. 265: G894-G902.

Stymne S and Stobart AK (1987) Triacylglycerol biosynthesis. In The biochemistry of
plants. A comprehensive treatise. PK Stumpf and Conn EE ed.(Academic Press, New
York) Vol.9 pp.l75-214.

Tang X, Halleck MS, Schlegel RA and Williamson P (1996) A subfamily of P-type
ATPases with aminophospholipid transporting activity. Science 272: 1495—1497.

Thelen JJ and Ohlrogge JB (2001) Metabolic engineering of fatty acid biosynthesis in
plants. Metab. Eng.4: 12-21.

Thoma S, Kaneko Y and Somerville C (1993) A non-specific lipid transfer protein
from Arabidopsis is a cell wall protein. Plant J. 3: 427-436.

Thumelin S, Esser V, Charvy D, Kolodziej M, Zammit VA, McGarry D, Girard J

and Pegorier JP (1994) Expression of liver camitine palmitoyltransferase I and 11 genes
during development in the rat. Biochem. J. 300: 583—5 87.

52

Towler DA, Gordon J], Adams SP and Glaser L (1988) The biology and enzymology
of eukaryotic protein acylation. Annu. Rev. Biochem. 57: 69-99.

Trigatti BL, Anderson RG and Gerber GE (1999) Identification of Caveolin-1 as a
Fatty Acid Binding Protein. Biochem. Biophys. Res. Commun. 255: 34—39.

Trigatti BL, Mangroo D and Gerber GE (1991) Photoaffinity labeling and fatty acid
permeation in 3T3-L1 adipocytes J. Biol. Chem. 266: 22621—22625.

Trotter PJ, Ho SY and Storch J (1996) Fatty acid uptake by Caco-2 human intestinal
cells. J. Lipid Res. 37: 336-346.

Turcotte LP, Srivastava AK and Chiasson JL (1997) Fasting increases plasma
membrane fatty acid-binding protein (FABPpm) in red skeletal muscle. Mol. Cell.
Biochem. 166: 153-158.

Uchiyama A, Aoyama T, Kamijo K, Uchida Y, Kondo N, Orii T, and Hashimoto T
(1996) Molecular cloning of cDNA encoding rat very long-chain acyl-CoA synthetase. J.
Biol. Chem. 271: 30360-30365.

Vance JE (1990) Phospholipid synthesis in a membrane fraction associated with
mitochondria. J. Biol. Chem. 265: 7248—7256.

van Helvoort A, Smith AJ, Sprong H, Fritzsche I, Schinkel AH, Borst P and van
Meer G (1996) MDRl P-glycoprotein is a lipid translocase of broad specificity, while
MDR3 P-glycoprotein specifically translocates phosphatidylcholine. Cell 87 : 507—517.

van Meer G Lipids of the Golgi membrane. Trends Cell Biol. 8: 29—33.
Velasco G, Geelen MJH, Gomez del Pulgar T and Guzman M (1998) Malonyl-CoA-
independent acute control of hepatic camitine pahnitoyltransferase I activity. J. Biol.

Chem. 273: 21497—21504.

Vick BA and Zimmerman DC (1984) Biosynthesis of jasmonic acid by several plant
species. Plant Physiol. 75:458-461.

Vigil EL (1970) Cytochemical and developmental changes in microbodies
(glyoxysomes) and related organelles of castor bean endosperrn. J. Cell Biol. 46: 435-

454.

Voelker DR (1989) Reconstitution of phosphatidylserine import into rat liver
mitochondria. J. Biol. Chem. 264: 8019—8025.

53

 

Voelker TA, Jones A, Cranmer AM, Davies HM and Knutzon DS (1997) Broad-
range and binary-range acyl-acyl-carrier protein thioesterases suggest an alternative
mechanism for medium-chain production in seeds. Plant Physiol. 114: 669-677.

Wali RK, Jaffe S, Kumar D and Kalra VK (1988) Alterations in organization of
phospholipids in erythrocytes as factor in adherence to endothelial cells in diabetes
mellitus. Diabetes 37 : 104—1 11.

Westphal S, Heins L, $01] J and Vothknecht UC (2001) Vippl deletion mutant of
Synechocystis: A connection between bacterial phage shock and thylakoid biogenesis?
Proc. Natl. Acad. Sci. USA. 98: 4243-4248.

Xu C, Fan J, Riekhof W, Froehlich JE and Benning C (2003) A permease-like protein
involved in ER to thylakoid lipid transfer in Arabidopsis. EMBO J. 22: 2370-2379.

Young SG and Fielding CJ (1999) The ABCs of cholesterol efflux. Nat. Genet. 22:
316—318.

Zakim D (1996) Fatty acids enter cells by simple diffusion. (1996) Proc. Soc. Exp. Biol.
Med. 212: 5-14.

Zhang F, Kamp F and Hamilton JA (1996) Dissociation of long and very long chain
fatty acids from phospholipid bilayers. Biochemistry 35: 16055-16060.

Zhao J, Zhou Q, Wiedmer T and Sims PJ (1998) Palmitoylation of phospholipid
scramblase is required for normal fimction in promoting Ca2+-activated transbilayer
movement of membrane phospholipids. Biochemistry 37: 6361—6366.

Zhou SL, Stump D, Kiang CL, Isola LM and Berk PD (1995) Mitochondrial aspartate
aminotransferase expressed on the surface of 3T3-L1 adipocytes mediates saturable fatty
acid uptake. Proc. Soc. Exp. Biol. Med. 208: 263—270.

Zhou SL, Stump D, Sorrentino D, Potter BJ and Berk PD (1992) Adipocyte
differentiation of 3T3-L1 cells involves augmented expression of a 43 kDa plasma
membrane fatty acid binding protein. J. Biol. Chem. 267 : 14456-14461.

Zhou Q, Sims PJ and Wiedmer T (1998) Identity of a conserved motif in phospholipid
scramblase that is required for Ca2+-accelerated transbilayer movement of membrane
phospholipids. Biochemistry 37 : 2356—2360.

Zhou Q, Zhao J, Stout JG, Luhm RA, Wiedmer T and Sims PJ (1997) Molecular
cloning of human plasma membrane phospholipid scramblase. A protein mediating
transbilayer movement of plasma membrane phospholipids. J. Biol. Chem. 272: 18240—
18244.

54

Zhou Z, White KA, Polissi A, Georgopoulos C and Raetz CRH (1998) Function of
Escherichia coli MsbA, an Essential ABC Family Transporter, in Lipid A and
Phospholipid Biosynthesis. J. Biol. Chem. 273: 12466-12467.

Zimmerman ML and Daleke DL (1993) Regulation of a candidate aminophospholipid-
transporting ATPase by lipid. Biochemistry 32: 12257-12263.

Zolman BK, Silva ID, Bartel B (2001) The Arabidopsis pxal mutant is defective in an

ATP-binding cassette transporter-like protein required for peroxisomal fatty acid beta-
oxidation. Plant Physiol. 127: 1266-1278.

55

 

CHAPTER 2
ON THE EXPORT OF FATTY ACIDS FROM THE CHLOROPLAST

ABSTRACT

The model for export of fatty acids from plastids proposes that the acyl-ACP (acyl carrier
protein) product of de nava fatty acid synthesis is hydrolyzed in the stroma by acyl-ACP
thioesterases and the free fatty acid (FFA) released is then transferred to the outer
envelope of the plastid where it is reactivated to acyl-CoA for utilization in cytosolic
glycerolipid synthesis. Experiments were performed to assess whether the delivery of
nascent FFA from the stroma for long chain acyl-CoA synthesis (LACS) occurs via
simple diffusion or a more complex mechanism. The flux through the in viva FFA pool
was estimated using kinetic labeling experiments with spinach and pea leaves. The
maximum half life for FFA in the export pool was 5 l 3. Isolated pea chloroplasts
incubated in the light with [14C]acetate gave a linear accumulation of FFA. When
CoASH and ATP were present there was also a linear accumulation of acyl-CoA
thioesters (plus derived polar lipids), with no measurable lag phase (< 30 3), indicating
that the FFA pool supplying LACS rapidly reached steady state. The LACS reaction was
also measured independently in the dark afler in situ generated FFA had accumulated
yielding estimates of LACS substrate-velocity relationships. Based on these experiments
the LACS reaction with in situ generated FFA as substrate is only about 3% of the LACS
activity required in viva at the very low concentrations of the F FA export pool calculated
from the in viva experiment. Furthermore, bovine serum albumin rapidly removed in situ

generated FFA from chloroplasts, but could not compete effectively for “nascent” FFA

56

substrates of LACS. Together the data suggest a locally channeled pool of exported FFA

that is closely linked to LACS.

57

INTRODUCTION

De nova fatty acid biosynthesis in plants occurs mainly in the plastid (Ohlrogge et al.,
1979). The immediate product, acyl-ACP thioester, may be used directly by plastid-
localized acyltransferases for the synthesis of “prokaryotic” lipids that are assembled
within the chloroplast. Alternatively the acyl moiety of the acyl-ACP can be exported to
the cytoplasm where it is primarily incorporated into glycerolipids at the endoplasmic
reticulum (eukaryotic pathway) (Roughan and Slack, 1982; Somerville and Browse,
1991). In leaves of plants such as Arabidopsis and spinach (16:3 plants) both pathways
contribute about equally to the synthesis of leaf glycerolipids (Somerville and Browse,
1991), whereas in non-photosynthetic tissue of all plants and in leaves of 18:3 plants such
as pea the eukaryotic pathway is the major pathway for glycerolipid synthesis (Ohlrogge

and Browse, 1995).

The current model for export of fatty acids produced by de novo synthesis within the
plastid proposes that acyl-ACP hydrolysis occurs in the plastid stroma by the action of
acyl-ACP thioesterases, possibly at the inner leaflet of the inner envelope, and that the
free fatty acid (FFA) released is transferred to the outer envelope of the plastid where it is
reactivated to acyl-CoA for utilization in cytosolic glycerolipid synthesis (Figure 3). The
chemical principal of this model was first proposed by Shine et al. (Shine et al., 1976).
As subsequent experiments showed that ACP-dependent fatty acid synthesis was almost
entirely a chloroplast function (Ohlrogge et al., 1979) this implied that the soluble acyl-

ACP thioesterase was a stromal enzyme activity. Long-chain acyl-CoA synthesis

58

Figure 3. Schematic for fatty acid export from the site of de novo fatty acid synthesis

in the chloroplast stroma to the cytosol.

Two simplified models are shown. The lower route is for simple diffusion of FFA and

upper route is via a protein mediated export mechanism. RCO: long chain fatty acyl

group, ACP: acyl carrier protein

    

    
  
  
 
 

  
  

       
 
 
 

    

 

Stroma Inner Periplasmic Outer Cytosol
Envelope Space Envelope
RCO.SACP RCO.SCOA
//”'T——T‘\ CoASH.
/ RCOOH ATP
T 5953153 .. B \TRANSPORTERZ / /
RC DOH RCOOH RC 0 OH RCO 0 H
u \ / U \S'“
/ Fast If”
RCOOH RCOOH . .

 

 

 

    

59

(LACS) activity is associated with the chloroplast envelope (Roughan and Slack, 1977;
Joyard and Stumpf, 1981) and more specifically with the outer envelope (Andrews and
Keegstra, 1983; Block et al., 1983). In addition, the major product of fatty acid synthesis
in isolated chloroplasts is F FA, unless ATP and CoASH are both present, when acyl-CoA
becomes a major end product (Roughan et al., 1979; Roughan and Slack, 1981). In viva

confirmation that F FA is indeed an intermediate in fatty acid export from the plastid was

18 .
accomplished by the use of O labeling experiments (Pollard and Ohlrogge, 1999).

There are several unknowns when considering the details of a fatty acid export process.
First, it is unclear if the acyl-ACP thioesterases associate with the inner leaflet of the
inner membrane of the envelope, thus releasing the acyl group to the inner leaflet (as
shown in Figure 3). Second, the vectorial nature of the long chain acyl-CoA synthesis
(LACS) reaction is unknown. The outer envelope contains pores which can
accommodate molecules of up to 8-10 kDa (Flugge and Benz, 1984). Thus metabolites
such as ATP, CoASH and acyl-CoAs may diffuse into the interrnembrane space between
the inner and outer envelope. It is uncertain whether soluble acyl-CoA binding proteins,
which have a MW of 10—11 kDa (Engeseth et al., 1996), will be able to move as freely.
We do not know whether the plastid long chain acyl-CoA synthetase can bind FFA at the
inner leaflet of the outer envelope, then transfer F FA to the outer leaflet for activation to
acyl-CoA, or whether the transfer of FFA across the outer envelope is accomplished by
other means, or if the active site of the long chain acyl-CoA synthetase is at the inner

leaflet of the outer envelope. We note by way of analogy that pairs of proteins, one of

60

which is always an acyl-CoA synthetase, have been implicated in the vectorial import of
fatty acids in E. coli (DiRusso et al., 1999), in yeast (Zou et al., 2003) and in animal cells
(Schaffer and Lodish, 1994; Coe et al., 1999). Third, there appears to be no direct
connections between the inner and the outer envelope of the plastid, although there may
be some “contact points” (Douce and Joyard, 1990). Thus at the minimum we would
expect that FFA will have to flip-flop across the inner envelope membrane, and then
possibly disassociate into the periplasmic space to reach the long chain acyl-CoA
synthetase at the outer envelope (Figure 3 lower route). Given these unknowns a major
question is whether the movement of F FA out of chloroplasts is a transporter-facilitated
process or involves free diffusion of the FFA intermediates. This is a debate which has
parallels in other systems where FFA transport is required (Abumrad et al., 1998;
Hamilton, 1998; Kleinfeld et al., 1997; Kamp et al., 2003). We report a variety of
experiments designed to examine the question of FFA export from the plastid, using

assays with intact leaves and chloroplasts.

RESULTS

The total in vivo pool of FFA in leaves is very small and implies a very short half life
for the F FA pool involved in export from the chloroplast

Although it is generally known that labeled FFA are minor products when leaf tissue is
incubated with acetate, the amount of FF A has not previously been quantified. We have
shown that F FA are intermediates in fatty acyl group export from the chloroplast (Pollard

and Ohlrogge, 1999) and to better understand this export process it was important to

61

quantify this FFA pool. Figure 4 presents a time course for acetate incorporation into

total fatty acids and into FFA by pea leaves over a 20 min period.

With expanding pea leaves a linear rate of lipid synthesis was established within 2 min
03 i gure 4A). At all time points the labeled total lipid extract from pea leaves contains 80-
85% labeled fatty acids, mainly as phosphatidylcholine. Presumably the short lag phase
indicates the time to reach the steady state balance between transport processes and pool
filling in the biosynthetic utilization of acetate. To measure the F FA pool the leaf assays

were quenched while still in the presence of substrate in the light (Figure 48). Control

. . . . 14 . .
extractions sprking unlabeled tissue wrth [ C]ole1c acrd gave > 90% recovery of label.

In the experiment in Figure 4 the labeled FFA pool at 2 min was 4.3% of total labeled
lipids and contained essentially only palmitic and oleic acids. In pea most of the fatty
acids synthesized in the chloroplast are exported (ca. 90%). The turnover time of FFA
during plastid export for pea can be estimated as follows. Let the steady state rate of lipid
synthesis be 100 units per min, of which approximately 85 units are fatty acids.
Therefore 76.5 units (90% of 85%) must pass through the FFA transfer pool for export to
the eukaryotic pathway. At the two minute time point the steady state level of labeled
FFA in the transfer pool has been reached, and because the steady state rate extrapolates
back to the x-axis at l min (Figure 4A) the total amount of lipids synthesized is
equivalent to about 100 units. Thus 100 units x 4.3% of labeled FFA are present (4.3

units). At two minutes we have a flux of 76.5 units per min through a transfer pool of 4.3

units. This gives a turnover time of 60 x 4.3/76.5 = 3.4 s and a tug value of 2.35 s (t1/2 =

62

Figure 4. Time course for acetate incorporation into acyl lipids by pea leaves.

. . . . . l4
Pea leaf trssue was Incubated wrth non-saturating concentranons of [ C]acetate (0.105

mM). The data points are the average of triplicate determinations and error bars
represent the range of values. Figure 4A shows acetate incorporation into total lipids by
halved pea leaves over time. At each time point 80-85% of the label in total lipids was
present as labeled fatty acid. Figure 4B shows acetate incorporation into the labeled FF A

in the same experiment, and is measured as pmoles acetate incorporated/mg chlorophyll.

63

25

 

Time (min)

 

 

 

 

 

 

 

 

0 5 0 5 0
2 4| 1

=_>:ao._o_:o 953.055
023... .80... SE sown—093:. oumuoo<

64

 

 

 

 

 

 

 

 

 

 

200

a o 0

w m 5

2350.820 953.055
<0". 35 539393.... 2303‘

25

20

15

Time (min)

65

turnover time x 0.693). A repeat of the time course with pea leaves gave the onset of

steady state labeling within 1 min, while the labeled FFA pool at 2 min was 2.05 i 0.26%

(4 determinations), giving a t1 /2 value of 1.7 s. A similar time course and FF A pool

analysis experiment was performed with spinach leaf strips (data not shown). A turnover

time 0.96 s and a tug value of 0.66 s were determined.

These calculations for pea and spinach leaves represent estimated values for the FFA
transfer pool turnover times that are an upper limit. The observed FFA pool may include
the transient pool of FFA being exported from the plastid, FFA released from labeled
acyl thioester pools during the heat quench step, plus FFA derived from other sources.
The kinetics of labeled FFA appearance in the time course for pea leaves (Figure 4B)
give us an indication that much of the labeled FFA pool is not a transfer pool. If the
observed FFA pool were all the transfer pool then this pool would saturate by 2 minutes.
Instead the pool continues to grow substantially. We speculate that some of the F FA pool
actually derives from fatty acid synthesis at the damaged margins of the leaf strips in the
assay. Given the approximations in the calculations it seems reasonable to state that the
t1 /2 for FFA in any chloroplast export pool is about 1 s, or less, and possibly much less.
Having established this limiting in viva pool size we then examined the detailed
relationship between products of fatty acid synthesis in isolated, intact pea chloroplasts.
In pea, compared to spinach, much more of the fatty acid produced is exported from the

chloroplast.

66

Time Course for Acetate Incorporation into Fatty Acyl Products by Isolated Pea
Chloroplasts

The synthesis of fatty acids from acetate by intact, illuminated chloroplasts can produce a
variety of acyl products (Heinz and Roughan, 1983; Gardiner et al., 1984), depending on
the plant species from which the chloroplasts are isolated and on the cofactors in the
chloroplast incubation medium. In a minimal chloroplast incubation medium (no
CoASH, glycerol-3-phosphate, nor UDP-galactose) the major product is free fatty acid
(FFA). Smaller amounts of 1,2-diacylglycerol (DAG) (ca. 5-20%) and polar lipids (PL)
(ca. 10-20%) have also been observed. When CoASH and ATP are added, up to 60% of

the label may be found as acyl-CoA (Roughan and Slack, 1981).

The incorporation of acetate into acyl lipids by pea chloroplasts is shown in Figure 5. In

. . . . . l4 . . .
mrmmal chloroplast Incubation medium the [ C]acetate 1ncorporatron Into total fatty

acids was 15.5 pmoles fatty acid/s/mg chlorophyll (Figure 5A), but when CoASH and
ATP were present this rate increased 1.4-fold to 21.5 pmoles fatty acid/s/mg chlorophyll
(Figure 5B). In the minimal medium the major product was FFA (about 90% of total
label). When CoASH and ATP were present up to 80% of the label was found as acyl-
CoA plus polar lipids (PL). Figure SB shows that over a 35 min period in the light with
ATP and CoASH total fatty acid synthesis was linear, and that the appearance of the
individual products, namely FFA (ca. 25%), acyl-CoA (ca. 40%) and total PL (ca. 40%)
were also linear, with negligible lag phase detected. The labeled PL fraction was
composed of about 60% phosphatidylcholine and 10% phosphatidylglycerol. To confirm

the lack of a lag phase earlier time points were taken (Figure 5C). Any lag phase

67

Figure 5. Time course for acetate incorporation into fatty acyl products by pea

chloroplasts.

In Figure 5A pea chloroplasts (40ug chlorophyll) were incubated under illumination in
. 14

0.2 ml basal medium wrth 2mM [ C]acetate. For B and C lmM ATP and 0.5mM

CoASH were also added to the assay medium. The fractions assayed were total lipids

(I), FF A (El), acyl-CoA (O), and total polar lipids (o).

 

 

 

 

 

500
A

400

5%

*‘Q.

(U

a§300

e—

05

20)

BE

5:3 200

8.8.

<:
100-

 

 

 

4O

 

Time (min)

68

 

 

 

 

 

 

500

 

 

O
4
to
l .0
no
'
pa
. O
I o ._.D 3
, ..
z ..
o
a H.— \II
, .. .m
I .. m
v U 0 II\
, .. 2 e
, .. .m
/ .. T
o I
o—
v
—
'a
O
o... 1.
”—
oo
—
—
o
a.
, 0
O 0 0 0 0
0 O 0 O
4 3 2 1

2339.020 mEEQoEE
cozmaoeooc. 2900a.

69

 

O O O O O

m. m w m 5
23:80.5 9:00.085
20.30008

0
u.

D---—----_

   

2O 25

15

Time (min)

 

7O

required for the establishment of the steady state condition of product accumulation in the
presence of CoASH plus ATP was less than 30 s. In Figure 5B or 5C if the bulk FFA
pool were a precursor of acyl-CoA, which is subsequently used for PL synthesis, a
measurable lag would occur in acyl-CoA plus PL accumulation. Thus there was no

observable precursor-product kinetic relationship between labeled FFA and acyl-CoA.
This implies that the F F A concentration to run the LACS reaction at steady state rate has
been reached rapidly (< 30 s) and that the bulk labeled F FA pool accumulating in Figure
5B or 5C is not a precursor for the LACS reaction. The linear rates in Figure 5B or 5C
further imply that neither the bulk F F A nor acyl-CoA are acting as feedback inhibitors for
fatty acid synthesis by isolated chloroplasts. Our lag phase estimates are currently
limited to > 10 s by the gentle mixing required for delicate organelles like chloroplasts,
for pool filling from acetate through to end products of fatty acid synthesis, and quench

times.

The Long Chain Acyl-CoA Synthesis (LA CS) Reaction in the Dark Utilizing in situ
Generated F FA

When FFA is exported from the plastid it is converted to acyl-CoA for utilization in
cytosolic reactions of lipid biosynthesis. The LACS reaction has previously been studied
after preparation of chloroplast envelope vesicles and addition of FFA substrate (Joyard
and Stumpf, 1981). In this study we investigated the LACS reaction in intact pea
chloroplasts with in situ generated FFA. Figure 6 presents the effect of switching off the
light after 15 min of incubation of chloroplasts in media containing no CoASH during the
light period but added at the onset of the dark period, or in media containing CoASH

during both light and dark periods. As observed by many researchers (Browse et al.,

71

1981; Roughan et al., 1980) total fatty acid synthesis was completely halted in the dark.
What is noteworthy is the very different behavior of the individual acyl pools during the
dark period. In the absence of CoASH the major fatty acid product in the light was F FA
(Figure 6A). When CoASH (0.5mM) was added at the transition to darkness there was a
very rapid conversion of FFA to acyl-CoA and its metabolites (PL). The initial rate of
FFA depletion (from 15 to 20 min) was at least twice the rate of total fatty acid synthesis
in the light. By contrast, in Figure 6B the initial rate of FFA depletion on transition to
darkness was less than 30% that of the rate of fatty acid synthesis in the light. The

combined acyl-CoA and PL accumulation had essentially stopped in the dark period.

Because the LACS reaction with in situ generated FFA was very fast in Figure 6A this
experiment was repeated with additional measurements immediately after the light to
dark transition (Figure 6C). There were variations between five independent experiments
but upon the dark transition with coincident addition of CoASH, about 25-50%, and
sometimes >50% of the FFA label was removed by the LACS reaction in the first 30 5.
Thus the LACS reaction in chloroplasts has the capacity to run at a rate over 10 times that
of fatty acid synthesis. However, the LACS reaction did not rapidly go to completion.
Inspection of Figure 6C shows that disappearance of FFA does not follow simple first
order kinetics. There are at least two components to the disappearance of FFA; a rapid
rate of depletion accounting for at least 60-70% of the labeled FFA but sometimes up to

80-85% depending on experiment, and a slow decay for the remainder.

Evidence for Distinct Kinetic Pools of F FA in Chloroplast Assays

72

Figure 6. Depletion of FFA pool after transition from light to dark.

14
Pea chloroplasts (40 pg chlorophyll) were incubated with 2mM [ C]acetate under

illumination in 0.2 ml basal medium containing 1 mM ATP for 15 minutes. After 15
minutes the lights were removed and the assay continued. Panel A and C: CoASH
(0.5mM) was added at the moment of transition into dark. Panel B: CoASH (0.5mM) was
added from the beginning of incubation in the light. The data points are average of 3
independent determinations. In panel A and B the fractions assayed were total lipids (I),
FFA (Cl), acyl-CoA (O), and total polar lipids (O). Panel C shows only the reaction after
the CoASH addition on transfer to the dark, and the fast and slow rates of FFA depletion
are indicated. In panel C the fractions assayed were FFA (D), acyl-CoA (O), and acyl-

CoA plus polar lipids (A ).

73

 

 

 

 

 

 

 

 

 

 

 

i

 

 

 

_ u
_ -
_ ._
_ ..
a __ ._.
A _ _.
.m 0. _. ..
aC _ ..
D + . .—
P . .
T o o .D 0
A . ~. 2
— )
1-..». .m
................ // m
a”. 0 0
" m
, _ .rl.
. _
II —
on m- w
, _
a... , _
_
UA . n
_
_
, _
_
_
_
. . r O
0 0 0 0 0 0
6 4 3 2 1

23209020 mE\mo_oE5
500.6900... 0.0.09...

74

 

 

Dark
ATP + CoASH ‘

 

 

Light
ATP + CoASH

 

 

 

 

 

 

 

 

150

O 0
0 5

1

2.2.0225 9500.225
cow—05900:. 0.9000.

 

 

20 30 40

Time (min)

10

75

 

 

 

 

 

 

-
_
-
-
_
_
_
—
—
—
—
—
—
a
—
a
w
a
a
—

2359020 9900695
5.559005 0.0.00.0.

 

400 600 800

Time (s)

200

 

 

76

Based on Figure 6A we know the LACS reaction is highly active in the dark and

therefore the slow rate of LACS in Figure 6B appears somewhat paradoxical, especially
since in Figure 6A the amount of labeled FFA at the dark transition is 4.7 nmoles/mg
chlorophyll while in Figure 6B it is still 3.5 nmoles/mg chlorophyll. This paradox is also
seen in Figure 5B with the linear accumulation of F FA up to 12 nmoles/mg chlorophyll
in the presence of ATP and CoASH, and is best explained by multiple FFA pools. The
results shown in Figures 5 and 6 reveal three kinetically distinct pools of FFA products.
In all of these pools F FA is assumed to be associated with the chloroplasts and
partitioned largely or completely into membranes. These pools are (A) the very small
FFA pool supplying the LACS reaction in the light in the presence of CoASH and ATP
(Figure 5B and 5C), (B) a bulk FFA pool that builds up if the LACS reaction cannot run
in the light because of the absence of CoASH but which is accessible to LACS when
CoASH and ATP are present (Figure 5A), and (C) a bulk FFA pool (about 20 -30% of
total label) that is not readily accessible to LACS reaction either in the light or the dark in
the presence of ATP and CoASH. If the appearance of acyl-CoA in assays run in the
presence of ATP and CoASH in the light is indeed catalyzed by the same enzyme that
catalyzes the LACS reaction in the dark, then bulk pool (B) will provide substrate for
pool (A). Thus in the light in the presence of ATP and CoASH, newly synthesized FFA
rapidly moves through pool (A), or accumulates in a linear fashion in pool (C) (Figure 5B
and 5C). Stopping fatty acid synthesis by placing the assay in the dark does not cause
much additional synthesis of acyl-CoA and PL because pool (B) is absent (Figure 6B). In
the absence of CoASH during fatty acid synthesis in the light, both pool (B) and pool (C)

accumulate in a linear fashion, and when the assay is placed in the dark with CoASH,

77

pool (B) is available for a rapid LACS reaction (Figure 6A and 6C). The physical nature
of these distinct FFA pools becomes an important question. Presumably in viva only
pool (A) is present. Pool (B) is artefactual in that in viva the LACS reaction is not
cofactor limited, but as we will discuss later serves as a useful model if F FA is allowed to
diffuse freely within the organelle. The LACS reaction of pool (B) F FA would be a mass
action driven activation of FFA. Pool (C) is an artefact, perhaps as a consequence of a

population of damaged chloroplasts.

The Dependence of the LA CS Reaction in the Dark on the Concentration of in situ
Generated F FA

The rapid depletion of F FA by the LACS reaction seen in Figure 6C will require a series
of FFA transport steps (membrane flip-flops and association and disassociation steps) to
deliver the FFA to the site of the LACS reaction. We do not know the distribution of
FFA within the chloroplast, or the orientation of the LACS active site in the outer
envelope, so we cannot speculate on the complexity of these steps. However, if we
approximate the FFA fast depletion reaction to a single (averaged) time constant for
exponential decay of pool (B) then that time constant is approximately 30 s, or sometimes
less. The concentration dependence of the FFA depletion is shown in Figure 7, using a
30 second dark LACS reaction assay. At the levels of F FA we allowed to accumulate in
the light preincubation period, the rate of the dark LACS reaction was essentially

proportional to F FA concentration and thus not saturated with respect to FFA substrate.

Accessibility of Chloroplast F FA Products to Sequestration by BSA in the Dark

78

BSA is well known for its ability to bind FF A (Parks et al., 1983; Hamilton et al., 1984).

We have used this FFA-binding capacity in competition experiments with chloroplasts.

. 14 . .
Chloroplasts that were pro-loaded wrth in situ generated [ C]FFA were Incubated wrth

different concentrations of defatted BSA for 5 min in the dark and centrifuged to separate
the chloroplast pellet from the media. Radiolabeled lipids in the media and in the
chloroplast pellet were assayed. At the beginning of the assay 80-85% of the total
labeled product was FFA and >85% of the total labeled lipid was associated with the
chloroplast. However when the chloroplasts containing labeled FFA were mixed with
increasing concentrations of BSA, increasing amounts of FFA were found in the
supernatant. This redistribution saturated at about 80% of the label released into the

medium at a BSA concentration of 0.3 mg/ml (4.5 pM).

The concentration of labeled FFA produced in situ in the assay was 2.5 uM. Next the
rate of FFA transfer from the chloroplasts to the BSA was examined by varying the co-
incubation time at 0.5mg/ml BSA (Figure 8). Over 90% of the labeled FFA partitioned
into the media bound to the BSA within 45 s. This was the quickest practical time to
conduct transfers and centrifugation. Thus the tug for F F A sequestration by BSA is about
10 s or less. Presumably the FF AzBSA binding reaction is mediated via BSA binding to
the outer leaflet of the outer envelope and requires F FA to be present in the outer leaflet

and to move through to the outer leaflet rapidly from other sites in the chloroplast.

BSA Competes with LA CS for the Bulk in situ FFA Pool but Is Not As Effective for the
Nascent F FA Pool

79

Figure 7. FFA concentration dependence of the LACS reaction.

. . 14 . . . .
Isolated pea chloroplasts were incubated With [ C]acetate under Illumination in 0.2 ml

basal medium containing 1 mM ATP for varying times in the absence of CoASH and
then lights were removed. One set of assays was quenched at this point to determine the
FFA pool accumulated. To a second set of assays CoASH (0.5mM) was added at the
moment of transition into dark. After another 30 seconds in the dark the reaction was
stopped by freezing immediately with liquid nitrogen and the FFA assayed to assess its

depletion by the LACS reaction.

6007'“ I- r- - 777777 . ...... . L L L, L

00
O
O

N
O
0

Rate of Acyl-CoA Synthesis
(pmoles/s/mg chlorophyll)

_L
O
O

 

 

0 5 10 15 20 25

FFA concentration (nmoles/mg chlorophyll)

80

Because BSA can rapidly and almost completely remove FFA generated in situ by
chloroplasts (Figure 8), experiments were conducted to measure the rate of the LACS
reaction with FFA substrate bound to BSA. FFA was first generated in situ in the
presence or absence of BSA for 15 min, then CoASH was added in the dark (Figure 9).

Comparing the initial rates for the LACS reaction the control reaction utilizing FFA
bound to the chloroplast was at about 20 times faster than the rate for FFA substrate
sequestered in the medium bound to BSA. This difference may be greater still if the
results are normalized to the same assay concentration of F FA generated in situ. The data
from Figure 9 indicate that the release of FFA to the chloroplast by BSA is a relatively

slow step.

The results of a continuous labeling assay under light in the presence and absence of BSA
are shown in Figure 10. When ATP, CoASH and BSA are present from the beginning of
the assay, LACS and BSA are expected to compete for de novo synthesized “nascent”
FFA from the export pool (A). If the F FA diffuses out to the outer envelope, BSA may
have a chance to bind a fraction of the F FA and thereby increase the proportion of FFA in
the total product. As observed in a previous study (Johnson et al., 2002) BSA stimulated
total fatty acid synthesis activity 1.6 fold (average of 4 different experiments, data not
shown). With the addition of BSA the acyl-CoA product increased from 26% to 37% and
that of PL decreased from 52% to 41% while FFA level remained unchanged at 22%.
The fact that BSA was not able to increase the proportion of FFA in the products again

argues against the idea that there was a bulk freely-diffusing pool of FFA during the

81

Figure 8. Sequestration of FFA by binding to BSA.

. 14
Isolated pea chloroplasts were first loaded wrth in situ generated [ C] FFA by

. . . . . . . 14 . . . . .
incubation in minimal media With [ C]acetate for 15 min in light. Upon transrtion to the

dark BSA (0.5 mg/ml) was added to the assay medium which was then centrifuged
(16,000x g) for 15 s at various time points. Supernatant was aspirated and collected and
the transfer reaction was terminated by quenching both the supernatant and pellet
immediately with liquid nitrogen. The fractions assayed were FFA (I), acyl-CoA (O),

and polar lipids (1:1).

 

25000

20000 i

1 5000

10000 *

Label in Chloroplast Pellet (dpm)

 

 

 

 

O 100 200 300 400
Time (s)

82

 

 

 

Figure 9. Competition between LACS and BSA for in situ generated FFA.

. 14 .
Isolated pea chloroplasts were first loaded With in situ generated [ C] labeled fatty acids

. .' . . . . . . 14 .
by incubation in minimal media containing [ C]acetate as substrate Without CoASH but

in the presence (I) or absence (1:1) of 0.5 mg/ml BSA for 15 min in light. Upon transition
to dark, CoASH (0.5mM) was added. Reactions were terminated after further
incubations with CoASH for various times. Label in FFA was measured. The initial rate
in the presence of BSA was 4050 dpm/min, and in the absence of BSA (dotted line)

77000 dpm/min.

 

1 00000

+ BSA
80000

60000 I

 

40000

Labeled FFA (dpm)

20000

 

 

 

 

 

 

Time (min)

83

process of F F A export from the chloroplast and suggests that LACS has “prior” access to

exported FFA before it diffuses out to the outer leaflet of the outer envelope membrane.

DISCUSSION
The Endogenous Free Fatty Acid Pool

The principal products of de nova fatty acid synthesis exported from most chloroplasts

are palmitate and oleate (Ohlrogge and Browse, 1995). In our previous study of leaf fatty

. 18
acid synthesrs from O-acetate (Pollard and Ohlrogge, 1999), the fatty acyl export from

the plastid was demonstrated to involve a hydrolytic step to produce a free carboxylate
anion prior to the fatty acid reactivation for incorporation into eukaryotic lipids. This
was an expected result but had not been previously demonstrated in vivo. The
confirmation makes consideration of FFA transport mechanisms in plastids a relevant
physiological problem. An immediate question to address is the size of this FFA

transport pool. In all carefully conducted labeling studies of plant tissues only a very

. . . . . . 14
small radiolabeled FFA fraction is observed in experiments wrth [ C]acetate. From our

kinetic analysis in this study we estimate that the tj/z for FFA in any plastid export pool is

515.

At this point this upper limit placed on the t1 /2 value for FFA is not sufficient to rule out

any mechanistic possibility. At one extreme it is instructive to consider the stoichiometry

84

Figure 10. Competition between LACS and BSA for “nascent FFA” in fatty acid

synthesis assays.

. . 14 . . . .
Isolated pea chloroplasts were incubated With [ C]acetate under illumination in a basal

incubation medium containing lmM ATP and 0.5 mM CoASH, in the presence or
absence of 0.5 mg/ml BSA. Samples were collected at 5, 10 and 20 minute time points
and the percentage of total labeled lipids present in FFA, acyl-CoA and polar lipid
fractions was analyzed. Values represent the means of 15 measurements from 4
independent experiments, with standard deviations shown. Student t-test comparing
control and BSA added samples showed that differences are significant for acyl-CoA*

(p< 0.001) and PL** (p<0.01) but are insignificant for FFA*** (p > 0.5).

60 .1. 7 7 7 7 . .7 -7 .._7 _ 7 ___7 . 7 __777 _

icy—ATFICoAs’H — 7 l
I **

50 i I ATP, CoASH.BSA I

 

h
O

*

l
l
l
i
l
l
l

30

 

H

20

10

[14C] Lipid composition (%)

  

 

I
l
l
l
l
i
i

 

 

 

0 , .

 

acyl-CoA FFA

85

for any channeled, protein-mediated transport process. Heber and Heldt (1981) have
estimated that the inner envelope protein concentration is 200—300 ug/mg chlorophyll.
The endogenous rate of fatty acid export is about 15-55 pmoles/s/mg chlorophyll (see
below). Assuming a transport protein of 60 kDa, with a 1:1 proteinzFFA binding

stoichiometry and a 50% occupancy of binding sites, then if the average transfer time is
about one second, the transfer protein concentration is 2-6 rig/mg chlorophyll, or about
0.5-3% of the inner envelope protein. Thus the time constant is consistent with the
possibility of a channeled transport system. Should there be a FFA transporter protein
(passive or active) in the inner envelope, it is probable given the 2-10 mm periplasmic
space between the two envelopes that it may be in direct contact with the acyl-ACP
thioesterase at the stromal interface of the inner envelope and with the LACS protein on

the outer enve10pe (Figure 3 upper route).

At the other extreme, consider the process of free diffusion from the plastid. The
possible steps are shown in Figure 3 lower route. A debate between a facilitated process
and free diffusion for F F A uptake by various types of animal cells has persisted for many

years (Abumrad et al., 1998; Hamilton, 1998; Kamp et al., 2003; Kleinfeld et al., 1998),

and some points are instructive to the issues of FFA export from the plastid. Very fast

rates of F FA flip-flop (t1 /2 325 ms) are observed for small (0.025 mm diameter) and large

(0.1 pm diameter) unilamellar PC vesicles, (Hamilton, 1998; Kamp et al., 1993; Kamp et

al., 1995). However, Kleinfeld et al. (1997) pointed out that there is an inverse

relationship between t1/2 for FFA flip-flop and the radius of curvature of the liposomes

86

studied. In giant unilamellar vesicles (>0.2 um diameter) t1/2 values of 0.7-7 5 were

measured for FFA flip-flop, values that were later found to be similar for erythrocyte
ghosts (Kleinfeld et al., 1998). Chloroplasts typically have diameters of 1-4 pm. An
important point is that there is no direct physicochemical measurements for FFA flip-flop
or binding and desorption from liposomes composed primarily of galactolipids, such as to
mirror the composition of the chloroplast inner envelope or the inner leaflet of the outer
envelope. In the absence of such measurements it is difficult to speculate whether the
physical rates of FFA flip-flop and membrane desorption might be adequate to account
for the rate of FF A export from the plastid, or for any movement of FFA into and from
thylakoids. Even with a series of physical measurements the chloroplast is a complex
organelle to model when compared to liposomes and erythrocyte ghosts, in that the
envelope is a double membrane. In addition within each chloroplast there is the massive
endomembrane system of thylakoids that accounts for >90% of the organelle’s membrane
and which might sequester F FA. Furthermore, in the light the pH of the stroma is about
8.5 whereas the pH of the cytosol is about 7.5. Thus freely diffusing FFA will preferably
partition to stromal facing leaflets in chloroplast membranes, that is, to those of the inner

envelope and thylakoids.

F FA Pools in Incubations with Intact Chloroplasts

In assays with spinach chloroplasts Roughan et al. (1980) noted that when chloroplasts
were incubated in the light for 20 min in basal medium and then transferred to the dark

for 20 min with CoASH and ATP, the FFA labeled product was reduced by half with the

87

 

concomitant appearance of acyl-CoA as the major product (or PC when microsomes were
also added). These assays were not run as kinetic assays, since their purpose was product
identification. The implication in this and other papers and reviews is that the acyl—CoA

formed upon addition of ATP plus CoASH is derived from the “bulk” FFA pool (Johnson

et al., 2002). Because our kinetic analysis of in vivo leaf assays indicated that the t1 /2 for

FFA in any plastid export pool is S l 5 this very small pool is inconsistent with the large
FFA pools seen in assays with isolated chloroplasts, and particularly with their being
functional transport intermediates. We have investigated this apparent contradiction and
showed there is not in fact an inconsistency if the total FFA produced in chloroplast

assays is described as the sum of several distinct FFA pools.

The results from our kinetic assays with intact chloroplasts (Figure 5 — 10) are interpreted
in terms of three fates (or pools) for FFA. A very small FFA pool (A) that provides

“nascent” FFA for LACS is implicated by the lack of precursor-product kinetics between

total FFA and acyl-CoA, as shown in Figure 5B. This pool size (i.e. t1 /2 value) is too

small to measure accurately but we postulate it is analogous to the in viva situation. The
LACS reaction is not feedback inhibited by continued accumulation of acyl-CoA. The
bulk accumulating F FA (Figure 5A) can be best described as a combination of two pools,
neither of which represents a physiological situation. One pool (B) is a LACS-accessible
pool which fills when the LACS reaction cannot function because of lack of cofactors.

The other pool (C) is present in all assays, represents 20 to 40% of the total labeled fatty

88

acids and is not rapidly accessible to LACS. Because Pool C is likely a function of

damaged chloroplasts, it is not discussed fiirther.

At saturating concentrations BSA rapidly removes F F A from the chloroplast (ti/2 < 10 s)
(Figure 8). This allowed us to test it as a competitor against the putatively channeled
LACS reaction, since the time constant for FFA removal from the chloroplast by BSA
was equal to or less than that for the independent LACS reaction. In addition, once FFA
is bound to BSA, the LACS reaction is reduced at least 20 fold (Figure 9). When fatty
acid synthesis occurs simultaneously in the presence of ATP, CoASH and BSA, BSA
does not inhibit the appearance of acyl-CoA products by sequestration of the exported
FFA (Figure 10). A simple explanation can be advanced for this result. Since we believe
the BSA binding of FFA is a physical transfer process it is likely that the rate of removal
of FFA from chloroplasts by BSA is proportional to the bulk FFA concentration, which
initially is about 2.5 uM (Figure 8). In intact chloroplasts BSA cannot penetrate the outer
envelope, and therefore must load FFA from the outer leaflet of the outer envelope or
from the very low amounts of FFA partitioned into the medium. We can estimate pool
(A), the pool of FFA required to sustain the channeled LACS reaction, as filling within
30 s or 1 s (using the in vitro or in viva estimates), which gives the bulk concentration of
FFA product in the chloroplast assay as <35 nM or <1 nM. With either calculation, on
simple kinetic grounds BSA should not effectively “intercept” FFA in pool (A). That is,
the lack of effect by BSA on products of continuous chloroplast assays containing ATP

and CoASH (Figure 10) is consistent with our pool (A) interpretation.

89

LA CS Activity in Isolated Chloroplasts and a Comparison with Leaf Tissue Labeling
Experiments Suggest a Channeled F FA Export Pool

The independent LACS reaction was measured in isolated intact chloroplasts by first
loading chlor0plasts in situ with FFA accumulated from de nova fatty acid synthesis in
the light after which the LACS reaction was run in the dark by adding CoASH (Figure
6A, 6C). The data in Figure 7 show that the initial LACS rate for a 15 min loading
period is not saturated with respect to FFA (it is saturated with respect to ATP and

CoASH) and therefore the depletion of pool (B) can be very approximately described by

a first order rate equation k(lacs) -[FFA]B. Because in situ-generated FFA can be rapidly

removed by BSA (Figure 8) pool (B) must be freely diffusible. We also know it is
mostly associated with the chloroplasts (Figure 8) unless the reaction mixture contains
FFA scavengers. However, we cannot be certain that FFA in pool (B) has partitioned
into the thylakoid membranes. Thus the intra-organelle location of FFA pool (B) may be
complex and FFA depletion in the dark by LACS may reflect complex kinetic
parameters, that is, various FFA diffusion rates (lateral diffusion, flip-flop, partitioning

between membrane and media), LACS enzymatic properties, possible LACS feedback

inhibition and the possibility of multiple LACS enzymes with different Km values etc.

As we assert below that pool (A) represents a channeled system, pool (B) may be
considered analogous to what might happen in viva if there was no channeling (eg. in a

mutant).

90

The endogenous rate of fatty acid synthesis in leaves is of the order of 1-2 umole C2
units/h/mg chlorophyll (Pollard and Ohlrogge, 1999; Browse et al., 1981; Bao et al.,
2000). This translates to about 15-55 pmoles C18 fatty acids/s/mg chlorophyll for plastid
export, assuming that 50-90% of newly synthesized fatty acids are exported. In our pea
chloroplast assays with exogenous acetate (Figure SB) acyl-CoA synthesis (acyl-CoA

plus PL) reached a steady state rate of 17.5 pmoles C18 fatty acids/s/mg chlorophyll in

less than 30 s. This value is lower than the LACS Vmax measured by Joyard and Stumpf

(1981). These authors characterized the LACS reaction in envelope membrane vesicles
isolated from spinach leaf chloroplasts. Assuming that the protein to chlorophyll ratio is
20:1 and that envelope protein accounts for about 1% of total plastid protein their Vmax
for 18:1-CoA formation was 85-140 pmoles/s/mg chlorophyll. In Figure 6C the total rate
of fatty acid synthesis is 16.7 pmoles fatty acid/s/mg chlorophyll during the light period,
and when the dark reaction is run the initial LACS rate is 220 pmoles fatty acid/s/mg
chlorophyll. This is 13-fold greater than the rate of fatty acid synthesis. The maximum
rate of the dark LACS reaction is estimated from Figure 7 as 670 pmoles fatty acid/s/mg
chlorophyll. We estimate that our assay concentration of FFA to drive the LACS
reaction of intact pea leaf chloroplasts at half maximum velocity as 3 pM, which is well
below the 200 1.1M Km reported by Joyard and Stumpf (1981) for oleic acid activation by
isolated chloroplast envelopes from spinach. However comparing bulk F FA
concentrations between assay systems for envelope vesicles and chloroplasts causes
inconsistencies, particularly with respect to the ratio of FFA to membrane lipids etc. The

numerical comparison should not be taken literally, but the comparison, albeit for

91

different plant species, does suggest that the LACS enzyme has a much greater affinity

for FFA than measured by studies with isolated envelope vesicles.

A more significant comparison is to assess the LACS rates predicted by Figure 7 when
using estimates of the pool size of FFA available for the reaction. For the chloroplast
continuous labeling assays in the light the steady state rate (Figure 5B, C) is established
within 30 5. At this time point the assay concentration of FFA is 0.17 nmoles/mg
chlorophyll (Figure 5B, C), which, according to Figure 7 should sustain a LACS rate of
4.6 pmoles/s/mg chlorophyll. However, the steady state rate of acyl-CoA production of
17.5 pmoles/s/mg chlorophyll has been established. Clearly, LACS operating on a freely

diffusing FFA pool within the chloroplast can not explain the onset of the steady state

condition. Furthermore, if we take the t1 /2 value of one second, derived from the in vivo

experiments, as the time to fill the transfer FFA pool then the in viva FF A transfer pools
are about 15-55 pmoles C18 fatty acids/mg chlorophyll. These FFA concentrations
would support 0.4-1.4 pmoles C18 fatty acids/s/mg chlorophyll according to the
concentration dependence curve for chloroplast LACS activity shown in Figure 7. The
discrepancy between acyl-CoA synthesis rates in chloroplasts when fatty acid synthesis is
run in the presence of CoASH and ATP (Figure 5B and 5C) and the independently
measured LCAS reaction (Figure 7) of 3.8-fold or greater increases to 37-fold for the
comparison involving estimates of in viva FFA transfer pools. This may in turn be
interpreted as 37-fold (or greater, as we are discussing upper limits) concentrated
localization of FFA in the export pool in viva (Figure 4) compared to freely diffiising

condition (Figure 7). We believe that this leaf tissue to chloroplast comparison is valid

92

because (1) the chloroplasts are synthesizing fatty acids at rates corresponding to in viva
rates of fatty acid synthesis, (2) because the major products from chloroplast assays in the
presence of ATP and CoASH are acyl-CoA or acyl-CoA-derived products, (3) because
higher maximum velocity and lower FFA affinity for LACS activity are measured with

intact organelles loaded with FFA in situ rather than with vesicles challenged with

exogenous FFA, and (4) because the t1/2 value to establish the steady state in isolated

chloroplasts could easily be an order of magnitude or more lower if rapid kinetic

techniques were available for organelles.

Possible Mechanisms for Fatty Acid Export

The analysis of our data presented above allow us to discount the simple free diffusion-
based model for FFA export from the plastid, as shown in Figure 3. There are three
generic models (and combinations thereof) to replace the simple diffusion model, namely
( 1) intervening transfer proteins or (2) a facilitated diffusion process that would mediate
the export of fatty acyl groups between the stromal acyl-ACP thioesterase and the outer
envelope long chain acyl-CoA synthase, or even (3) a cryptic inner envelope LACS.

These models [i.e. for pool (A) in the intact chloroplast assays] are briefly discussed.

For facilitated diffusion without any intervening protein FFA must be deposited in the
envelope leaflet where it is most readily used by the LACS enzyme. The limiting in viva
FFA transfer pool size of one second calculates to about 0.1 to 0.01 mole% FFA if

deposited in a single envelope leaflet lipid phase. In this leaflet the LACS enzyme has a

93

very high affinity for FF A (low Km), so that very little FFA has time to move elsewhere.

When the LACS reaction is made inoperative in isolated chloroplasts by lack of CoASH
substrate then FFA diffusion to other chloroplast membranes creates pool (B). Little is
known about the actual localization, topography and/or vectorial nature of acyl-ACP
thioesterases and long-chain acyl-CoA synthetases. Plastid fatty acid synthesis is likely a
channeled process (Roughan and Ohlrogge, 1996) associated in viva with envelope
membranes as has been observed for acetyl-CoA carboxylase (Thelen and Ohlrogge,
2002) and as suggested by association of acyl ACP thioesterase with membrane fractions
when imported into the pea chloroplast by in vitro reconstitution experiments
(Unpublished observation communicated by Dr. John Froehlich, MSU-DOE Plant
Research Laboratory). There are at least nine LACS genes in Arabidopsis (Shockey et
al., 2002). One of them, LACS9 (At1g77590), is demonstrated to be localized to the
enve10pe of the plastid (it is not shown whether LACS9 is located on the outer or inner
envelope membrane) and was shown to be responsible for the major in vitro LACS
activity on the plastid enve10pe (Schnurr et al., 2002). This gene product may therefore
belong to the class responsible for the outer envelope LACS activity described previously
(Andrews and Keegstra, 1983; Block et al., 1983). To have facilitated FFA diffiision
without any additional polypeptides it seems necessary to propose that the fatty acyl ACP
thioesterase binds to the inner envelope, possibly causing a localized membrane
perturbation which can facilitate FFA flip-flop and loads the hydrolyzed product directly
onto the inner envelope. Also, the outer envelope long-chain acyl-CoA synthetase would
have a domain that would extend across periplasmic space and interact with the inner

envelope in a way that would facilitate F FA transfer.

94

Conceptually it is easier to envisage a protein-mediated channel (Figure 3). There is
evidence from bacterial, yeast and mammalian systems that pairs of proteins are often
involved in the vectorial import of fatty acids across biomembranes. In E. coli an outer
envelope localized protein (FadL) was shown to be necessary for the saturable binding
and permeation of exogenous long chain fatty acids (Mangroo and Gerber GE, 1992;
Kumar and Black, 1993; DiRusso and Black, 1999). The corresponding acyl-CoA
synthetase resides in the inner envelope. In yeast both the FATP protein and the fatty
acyl-CoA synthetases of the plasma membrane are interacting components of the fatty
acid import machinery (Zou et al., 2003). In animal cells FATP and acyl-CoA
synthetases, both plasma membrane proteins, are required for optimum fatty acid uptake
(Schaffer and Lodish, 1994). FATP is an AMP-binding domain protein with homology
to the very long chain acyl-CoA synthases (Coe et al., 1999). By contrast, activated long
chain fatty acid import across the peroxisomal membrane in yeast is mediated by an ATP
binding cassette (ABC) transporter (Hettema et al., 1996; Hettema and Tabak, 2000). The
data we have obtained are consistent with such an intermediate protein between the acyl-
ACP thioesterase and LACS. It is apparent that in the absence of the cofactors for the
LACS reaction, nascent FFA (pool (A)) is released (to become pool (3)). This release
may appear rapid on the time scale of the assays, but could be quite slow relative to the
time taken to channel F FA across the plastid envelope to the LACS protein in viva. If
there is only one LACS enzyme then the transport system would also have a finite rate of
loading from non-channeled FFA to be able to independently function with freely

diffusing FFA. Possible gene candidates for fatty acid transporters might include

95

At4g14070, At3g23790 and the ABC transporters associated with the plastid envelope

(K00 and Ohlrogge, 2002).

The final generic model proposes a latent inner envelope LACS enzyme. Although
LACS activity was localized to the outer envelope membrane (Andrews and Keegstra,
1983; Block et al., 1983) we can not definitively rule out the possibility of an inner
envelope LACS that is intimately linked to FF A export from the inner leaflet of the inner
envelope via an acyl-ACP thioesterase complex, but whose activity is cryptic in a simple
LACS assay. Among the possible Arabidopsis gene candidates for “cryptic” LACS
might be LACS8 (At2g04350) (Schnurr et al., 2002) or At4g14070 and At3g23790
(AMP-binding proteins with predicted plastid leader sequences). Disruption of a
transporter or cryptic LACS might result in an increased F FA pool in vivo, and if it has a
time constant for removal of the order of seconds, analogous to the removal of pool (B)
in chloroplast assays by the LACS dark reaction, then the rapid in viva labeling technique

might be useful for the discovery of Arabidopsis lines mutated in these gene products.

In conclusion, we believe that this study defines a set of parameters needed to more fully
understand the molecular mechanism of free fatty acid transfer from the plastid. There
are several possibilities, as described above, and, given the absence of physical-chemical
data on FFA transfer steps for galactolipid-rich membranes and the complexity
introduced by the double membrane envelope of plastids, these will not be easy

experimental systems. However, the possibility of using reverse genetics in Arabidopsis

96

should open up tractable new avenues to combine with the traditional ways of studying

such problems.

MATERIALS AND METHODS

Materials

Pea seeds (Pisum sativum L. cv Little marvel) were germinated on moist vermiculite. For
chloroplast preparations the peas were grown for 8-10 days in the growth chamber at
25°C and an 8 h photoperiod with occasional watering. For the in viva experiments the
peas were grown for 12 days in the growth chamber watering with nutrient solution.
Young pea leaves were removed that had yet to open. These were cut in half laterally,

immediately placed in assay medium and preincubated as described below. [1-

14 . . . . .
C]Acetic acid (57.2 Ci/mol) was purchased from American Radiolabeled Chemicals

Inc. (St. Louis, MO).
In vivo Incubations

Leaf assays with halved pea leaves were run in 5.0 ml medium containing 25 mM

NaMES buffer, pH 5.7, with 0.3-0.4 g fresh weight of tissue, at 28°C and under strong

. . . 2
illumination (180 pmol photons/m /s). Tween-20 was added as a wetting agent to a

. . . . . . . 14
concentration of 0.01% w/v. The reaction was initiated With sodium [1- C]acetate

97

solution. Time course assays were run with 0.03-0.06 mCi of labeled acetate (0.105-0.21
mM). These assays were terminated by partial removal of the medium and quenching

with hot isopropanol while the tissue was under illumination.
Chloroplast Preparation and Assays

Intact chloroplasts were isolated on a continuous Percoll gradient (Bruce et al., 1994).
All procedures were conducted at 0-4 °C. About 10-20g of 8-day old pea seedlings were

homogenized in 100 ml of semi-frozen homogenization buffer (50 mM HEPES, pH 8.0,

330 mM sorbitol, 0.1% BSA, 1 mM MgC12, 1 mM MnClz, 2 mM EDTA) using a

polytron (PT 10/35) and filtered through 2 layers of miracloth (CALBIOCHEM). Crude
chloroplasts were collected by centrifugation at 1200 xg max for 5 min. These were
purified by centrifugation through pre-made gradients of Percoll generated by
centrifugation of 30 ml of 50% Percoll in homogenization buffer for 30 min at 40,000x g.
Crude chloroplasts (2-3 ml) were centrifuged through the pre-made gradients at 13,000
xg max for 5 min in an HB-4 rotor without brake. The intact chloroplast band formed
near the bottom of the gradient was recovered and washed with resuspension buffer
containing 50 mM HEPES, pH 8.0 and 330 mM sorbitol. The resulting isolated
chloroplasts can incorporate acetate into fatty acids at rates corresponding to in viva rates
of fatty acid synthesis, import and process in vitro translated protein precursors, sustain
high rates of photosynthesis (Keegstra et al., 1986), and give about 90% intactness as

determined by phase contrast light microscopy and the reduction of ferricyanide before

98

and after an osmotic shock (Lilley et al., 1975; Douce and Joyard, 1982). Quantification

of chlorophyll was performed as described by Amon (1949).

Chloroplast incubations to synthesize fatty acids were performed largely according to the

methods described by Roughan (Roughan, 1987). Chloroplasts (40-50 pg chlorophyll)

2 . . .
were incubated under illumination (180 pmol photons/m /s) With shaking at 25°C in a

medium (200 pl) containing 0.33 M sorbitol, 25 mM HEPES-NaOH, pH 8.0, 10 mM

KHCO}, 2 mM Na3EDTA, 1 mM MgC12, 1 mM MnClz, 0.5 mM K2HPO4, and 2 mM

. . . 14 . .
acetate containing 2 00 of [ C]acetate. Where noted in the text assays also contained 1

mM ATP, 0.5 mM CoA, and/or 0.5 mg/ml BSA added at specific times. An aliquot was
removed before the illumination and used to measure the actual radioactivity added.
Reactions were terminated by placing the incubation mixture into liquid nitrogen and

were kept in -80°C freezer until lipid analysis.

In the BSAzFFA binding assays, chloroplasts were separated from the BSA in the
medium by centrifugation at 13,000 rpm (Sorvall Biofuge pico) for 15 s. The supernatant
was aspirated from the chloroplast pellet and frozen in liquid nitrogen. Fatty acids were

extracted from the supernatant and the chloroplast pellet separately.

Lipid Analysis

99

After quenching the tissue by heating in isopropanol at 80-90 °C for 5 min lipid

. l4 . . . .
extraction from the [ C]acetate fed leaf strips was carried out usrng hexane-isopropanol

(Hara and Radin, 1978). An aliquot of the lipid extract was suspended in 4:1
acetone/water and the OD at 652nm measured to determine chlorophyll (Amon, 1949)
another aliquot was assayed by liquid scintillation counting to determine radioactivity
incorporated. Suspension of the lipid residue in organic solvent and re-evaporation in the
presence of a few drops of glacial acetic acid was crucial to drive off residual acetate
substrate contamination to give accurate radioactivity determinations. Lipid classes and
fatty acid methyl esters were analyzed by thin layer chromatography (TLC) and isolated
by preparative TLC. TLC plates were scanned for radioactivity using a Packard Instant
Irnager (Canberra Instruments), both to quantify radioactivity and to locate the
appropriate bands for recovery from preparative TLC plates. For pea leaf extracts a non-
saponifable band (4-6% of total label) co-migrated very close to FFA on TLC, so the
samples were first treated with ethereal diazomethane and then fatty acid methyl esters
determined to measure the free fatty acid pool. Neutral lipids were analyzed on silica
plates developed half way with hexane/diethyl ether/acetic acid 70:30:l (v/v/v) and then
fully with hexane/diethyl ether/acetic acid l80:20:l (v/v/v). For analysis of individual
labeled fatty acids the methyl ester fraction was isolated by preparative TLC and
separated using either C18-reversed phase TLC, developing half way and then fully with
acetonitrile/methanol/water 130:70:1 (v/v/v), or by argentation-TLC using plates
impregnated with a 5% silver nitrate in acetonitrile solution, dried. and activated, and

developed with hexane-diethyl ether mixtures.

100

The method for analysis of radiolabeled products from chloroplast assays was adapted
from Roughan (Roughan, 1987). Samples were saponified by heating in 80-90°C for 60
min in 0.8 ml of 10% KOH (w/v) in methanol per 0.2ml reaction mixture. After cooling,
acidification and extraction with hexane and then evaporation of the hexane under
nitrogen the total radioactivity recovered was determined by liquid scintillation counting.

Since there are negligible non-saponifiable products this radioactivity is a measure of the

. . l4 . . . . .
incorporation of [ C]acetate into total fatty acids. Lipid class analysrs was performed

by mixing 0.2 ml of reaction mixture with 2 ml of chloroform/methanol 1:1 (v/v) and

0.72 ml of 0.2M H3PO4 in 2 M KCl. The lower organic phase was collected and the

upper aqueous methanol phase was back-extracted with hexane (4 ml). The combined
organic extracts were evaporated to dryness and an aliquot assayed for radioactivity by
liquid scintillation counting. Neutral and polar lipids were separated by TLC using K6
silica plates (Whatman, Clifton, PA) developed in hexane/diethyl ether/ glacial acetic acid
80/20/1 (v/v/v). To determine the polar lipid composition, the polar lipid band was
recovered and analyzed on K6 silica TLC plates impregnated with 0.15 M ammonium
sulphate and activated at 120 °C for 3 h, and developed in acetone/toluene/water 91/30/8
(v/v/v) (Kahn and Williams, 1977). The aqueous methanol phase contains acyl-CoA as
the major saponifiable lipid. Over 80% of the label was shown to co-migrate with long
chain acyl-CoA standard when extracted by butan-l-ol and analyzed by TLC (K6 plate
developed in butan-l-ol/acetic acid/water 50:20:25 (v/v/v)). Routinely the aqueous phase
was saponified, acidified, extracted with hexane and the extractable lipids (i.e. free fatty

acids) assayed for radioactivity (1 l, 32). Control extractions with chloroplasts spiked

101

with known amounts of radiolabeled oleic acid or oleoyl-CoA yielded essentially
complete recovery of oleic acid and > 85% recovery of oleoyl-CoA either in the presence

or absence of BSA.

102

LITERATURE CITED

Abumrad N, Harmon C and Ibrahimi A (1998) Membrane transport of long-chain
fatty acids: evidence for a facilitated process. J. Lipid Res. 39: 2309-2318.

Andrews J and Keegstra K (1983) Acyl-CoA synthetase is located in the outer
membrane and acyl-CoA thioesterase in the inner membrane of pea chloroplast
enve10pes. Plant Physiol. 72: 735-740.

Arnon D1 (1949) Copper enzymes in isolated chloroplasts. Polyphenol oxidase in Beta
vulgaris. Plant Physiol. 24: 1-15.

Block MA, Dome A-J, Joyard J and Douce R (1983) The acyl-CoA synthetase and
acyl-CoA thioesterase are located on the outer and inner membrane of the chloroplast
envelope, respectively. FEBS Lett. 153: 377-381.

Browse XM, Focke M, Pollard MR and Ohlrogge J (2000) Understanding in viva
carbon precursor supply for fatty acid synthesis in leaf tissue. Plant J. 22: 39-50.

Browse J, Roughan PG and Slack CR (1981) Light control of fatty acid synthesis and
diurnal fluctuations of fatty acid composition in leaves. Biochem. J. 196: 347-354.

Bruce BD, Perry S, Froehlich J and Keegstra K (1994) In vitro import of proteins into
chloroplasts. In Plant Molecular Biology Manual (Gelvin, S. B., and Schilperoot, R. A.,
eds) 2nd Ed., Section J1, pp. 1-15, Kluwer Academic Publishers, London.

Coe NR, Smith AJ, Frohnert BI, Watkins PA and Bernlohr DA (1999) The fatty acid
transport protein (FATPl) is a very long chain acyl-CoA synthetase. J Biol. Chem. 274:
36300-36304.

DiRusso CC and Black PN (1999) Long-chain fatty acid transport in bacteria and yeast.
Paradigms for defining the mechanism underlying this protein-mediated process. Mol.
Cell Biochem. 192: 41-52.

DiRusso CC, Black PN and Weimar JD (1999) Molecular inroads into the regulation
and metabolism of fatty acids, lessons from bacteria. Prog. Lipid Res. 38: 129-197.

Douce R and Joyard J ( 1982) Purification of the chloroplast envelope. In Methods in
Chloroplast Molecular Biology (Edelman, M., Hallick, RB, and Chua, N.H., eds.) pp.
239-256, Elsevier Biomedical, New York.

Douce R and Joyard J (1990) Biochemistry and function of the plastid envelope. Annu.
Rev. Cell Biol. 6: 173-216.

103

Engeseth NJ, Pacovsky R, Newman T and Ohlrogge JB (1996) Characterization of an
acyl-CoA-binding protein from Arabidopsis thaliana. Arch. Biochem. Biophys. 331: 55-
62.

Flugge UI and Benz R (1984) Pore-forming activity in the outer membrane of the
chloroplast envelope. FEBS Lett. 169: 85-89.

Gardiner SE, Heinz E and Roughan PG (1984) Rates and products of long-chain fatty

14 . .
acid synthesis from [1- C]acetate 1n chloroplasts isolated from leaves of 16:3 and 18:3
plants. Plant Physiol. 74: 890-896.

Hamilton JA (1998) Fatty acid transport — difficult or easy? J. Lipid Res. 39: 467-481.

Hamilton JA, Cistola DP, Morrisett JD, Sparrow JT and Small DM (1984)
Interactions of myristic acid with bovine serum albumin: a 13C NMR study. Proc. Natl.
Acad. Sci. USA. 81: 3718-3722.

Hara A and Radin NS (1978) Lipid extraction of tissues with a low toxicity solvent.
Anal. Biochem. 90: 420-426.

Heber U and Heldt HW (1981) The chloroplast envelope: structure, function and role in
leaf metabolism. Annu. Rev. Plant Physiol. 32: 139-168.

Heinz E and Roughan PG (1983) Similarities and differences in lipid metabolism of
chloroplasts isolated from 18:3 and 16:3 plants. Plant Physiol. 72: 273-279.

Hettema EH and Tabak HF (2000) Transport of fatty acids and metabolites across the
peroxisomal membrane. Biochim. Biophys. Acta.1486: 18-27.

Hettema EH, van Roermund CW, Distel B, van den Berg M, Vilela C, Rodrigues-
Pousada C, Wanders RJ and Tabak HF (1996) The ABC transporter proteins Patl

and Pat2 are required for import of long-chain fatty acids into peroxisomes of
Saccharomyces cerevisiae. EMBO J. 15: 3813-3822.

Johnson PE, Rawsthorne S and Hills MJ (2002) Export of acyl chains from plastids
isolated from embryos of Brassica napus (L.). Planta 215: 515-517 .

Joyard J and Stumpf PK (1981) Synthesis of long-chain acyl-CoA in chloroplast
envelope membranes. Plant Physiol. 67 : 250-256.

Kahn MU and Williams JP (1977) Improved thin layer chromatographic method for the
separation of major phospholipids and glycolipids from plant lipid extracts and

phosphatidylglycerol and bis(monoacylglycerol)phosphate from animal lipid extracts. J.
Chromatogr. 140: 179-185.

104

Kamp F, Guo W, Souto R, Pilch PF, Corkey BE and Hamilton JE (2003) Rapid flip-
flop of oleic acid across the plasma membrane of adipocytes. J. Biol. Chem. 278: 7988-
7995.

Kamp F and Hamilton JA (1993) Movement of fatty acids, fatty acid analogues, and
bile acids across the phospholipid bilayer. Biochemistry 32: 11074-11086.

Kamp F, Zakim D, Zhang F, Noy N and Hamilton JA (1995) Fatty acid flip-flop in
phospholipid bilayers is extremely fast. Biochemistry 34: 11928-11937.

Keegstra K and Yousif AE (1986) Isolation and characterization of chloroplast
envelope membranes. Methods Enzyrnol. 118: 316-325.

Kleinfeld AM, Chu P and Romero C (1997) Transport of long-chain native fatty acids
across lipid bilayer membranes indicates that transbilayer flip-flop is rate limiting.
Biochemistry 36: 14140-14158.

Kleinfeld AM, Storms S and Watts M (1998) Transport of long-chain native fatty acids
across human erythrocyte ghost membranes. Biochemistry 37 : 8011-8019.

Koo AJ and Ohlrogge JB (2002) The predicted candidates of Arabidopsis plastid inner
envelope membrane proteins and their expression profiles. Plant Physiol. 130: 823-36.

Kumar GB and Black PN (1993) Bacterial long-chain fatty acid transport. Identification
of amino acid residues within the outer membrane protein FadL required for activity. J.
Biol. Chem. 268: 15469-15476.

Lilley R McC, Fitzgerald MP, Rienits KG and Walker DA (1975) Criteria of
intactness and the photosynthetic activity of spinach chloroplast preparations. New
Phytol. 75: 1-10.

Mangroo D and Gerber GE (1992) Photoaffinity labeling of fatty acid-binding proteins
involved in long chain fatty acid transport in Escherichia coli. J. Biol. Chem. 267: 17095-
17101.

Ohlrogge J and Browse J (1995) Lipid biosynthesis. Plant Cell 7: 957-970.

Ohlrogge JB, Kuhn DN and Stumpf PK (1979) Subcellular localization of acyl carrier
protein in leaf protoplasts of Spinacia oleracea. Proc. Natl. Acad. Sci. USA 76: 1194-
1198.

Parks J S, Cistola DP, Small DM and Hamilton JA (1983) Interactions of the carboxyl

group of oleic acid with bovine serum albumin: a 13C NMR study. J Biol. Chem. 258:
9262-9269.

105

Pollard M and Ohlrogge J (1999) Testing models of fatty acid transfer and lipid
synthesis in spinach leaf using in viva oxygen-18 labeling. Plant Physiol. 121: 1217-
1226.

Roughan PG (1987) Long-chain fatty acid synthesis and utilization by isolated
chloroplasts. Methods Enzymol. 148: 327-337.

Roughan PG, Holland R and Slack CR (1979) On the control of long-chain-fatty acid
synthesis in isolated intact spinach (Spinacia oleracea) chloroplasts. Biochem. J. 184:
193-202.

Roughan PG, Holland R and Slack CR (1980) The role of chloroplasts and microsomal

. . . 14 .
fractions in polar-lipid synthesrs from [1- C]acetate by cell-free preparations from
spinach (Spinacia oleracea) leaves. Biochem. J. 188: 17-24.

Roughan PG, Kagawa T and Beevers H (1980) On the light dependency of fatty acid
synthesis by isolated spinach chloroplasts. Plant Sci. Lett. 18: 221-228.

Roughan PG and Ohlrogge JB (1996) Evidence that isolated chloroplasts contain an
integrated lipid-synthesizing assembly that channels acetate into long-chain fatty acids.
Plant Physiol. 110: 1239-1247.

Roughan PG and Slack CR (1977) Long-chain acyl-coenzyme A synthetase activity of
spinach chloroplasts is concentrated in the envelope. Biochem. J. 162: 457-459.

Roughan PG and Slack CR (1981) Acyl-CoA may be a neglected product in studies of
fatty acid synthesis by isolated chloroplasts. FEBS Lett. 135: 182-186.

Roughan PG and Slack CR (1982) Celluar-organization of glycerolipid metabolism.
Annu Rev Plant Physiol Plant Mol Biol 33: 97-132.

Schaffer JE and Lodish HF (1994) Expression cloning and characterization of a novel
adipocyte long chain fatty acid transport protein. Cell 79: 427-436.

Schnurr JA, Shockey JM, De Boer GJ and 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.

Shine WE, Mancha MI and Stumpf PK (1976) Fat metabolism in higher plants. The
function of acyl thioesterases in the metabolism of acyl-coenzymes A and acyl-acyl
carrier proteins. Arch. Biochem. Biophys. 172: 110-116.

Shockey JM, Fulda MS and Browse JA (2002) Arabidopsis contains nine long-chain

acyl-coenzyrne a synthetase genes that participate in Fatty Acid and glycerolipid
metabolism. Plant Physiol. 129: 1710-1722.

106

 

Somerville C and Browse J (1991) Plant lipids: metabolism, mutants and membranes.
Science 252: 80-87.

Thelen JJ and Ohlrogge JB (2002) The multisubunit acetyl-CoA carboxylase is
strongly associated with the chloroplast envelope through non-ionic interactions to the
carboxyltransferase subunits. Arch. Biochem. Biophys. 400: 245-257.

Zou Z, Tong F, Faergeman NJ, Borsting C, Black PN and DiRusso CC (2003)
Vectorial acylation in Saccharomyces cerevisiae. Fatlp and fatty acyl-CoA synthetase
are interacting components of a fatty acid import complex. J. Biol. Chem. 278: 16414-
16422.

107

CHAPTER 3

SITES AND MECHANISMS FOR ACTIVATION AND ELONGATION OF
EXOGENOUS FATTY ACIDS

ABSTRACT

Plant cells are capable of incorporating, elongating and desaturating exogenously
provided short- and medium-chain free fatty acids (FFA’s). However the subcellular

sites and mechanisms for these processes are not currently known. When detached

. . . . 14
Arabidopsrs leaves or isolated pea chloroplasts were incubated With [1- C]FFA’s ($16-

carbon), label appeared in more elongated and desaturated species of fatty acids (FA’s).
The rate of laurate elongation was 4.9 nmoles /h/mg chlorophyll with intact leaves and 22
nmoles/h/mg chlorophyll with isolated chloroplasts. The elongation by isolated
chloroplasts was light dependent, stimulated 1.4 fold by lmM ATP and was independent
of exogenous coenzyme A. The elongation was 60 % inhibited by 10 uM cerulenin, an
inhibitor of plastidial ketoacyl-ACP synthetase I/II (KAS UK). In contrast, haloxyfop, an

inhibitor of cytosolic acetyl-CoA carboxylase, inhibited only elongation into very-long

14
chain FA’s (2 C20) but not the synthesis of C-unsaturated C18 or C16 FA’s. These

data indicate that the elongation of exogenous medium-chain FA’s into C16 and C18
FA’s by Arabidopsis leaves is mainly the function of chloroplasts, most likely through

the common de nova FA synthesis machineries. The measured pool size of acyl-ACP

. 14 .
during [1- C] laurate elongation by pea chloroplasts was 2.3 uM. In contrast, there was

108

less than 0.1 nM (or < 1/ 1000 acyl-ACP concentration) of labeled acyl-CoA indicating
that almost no detectable acyl-CoA intermediate was formed during the elongation

process. A T-DNA insertion mutant, for a gene (AAE15) that is predicted to encode an

14
acyl activating enzyme showed 80% reduction of [1- C]laurate elongation into C16 and

C18 FA’s. AAE15 has sequence similarity to long-chain acyl-CoA synthetases (LACS’s)

and is predicted to be targeted to the plastid.

109

 

INTRODUCTION

It has been observed that plant tissues are capable of incorporating exogenously provided
fatty acids into their endogenous lipids. Depending on their length and level of

unsaturation, the exogenous fatty acids undergo different metabolic fates. Exogenous [1-

14C]palmitic and stearic acid go through the eukaryotic pathway and initially are

incorporated into cytosolic glycerolipids mainly at the sn-1 position of
14 .
phosphatidylcholine (PC) and sn-l and sn-3 of triacylglycerol (TAG). [1- C]oleic acid

metabolism is similar except that it has less positional preference between sn-l and sn-2
(Roughan et al., 1987). On the other hand short and medium chain fatty acids of chain
length up to 12 carbons have been generally considered to serve as substrates for fatty
acid synthesis (Hitchcock and Nichols, 1971). Labeled acetic, octanoic, decanoic,
dodecanoic, tetra decanoic acted as precursors for oleic, linoleic and linolenic acids by
isolated leaves of Ricinus communis, cabbage, lettuce, Brussels sprouts and chicory
(James, 1962). Barley seedlings and leaf slices of spinach converted 8 - 14 carbon free
fatty acid precursors into longer chain fatty acids (Hawke and Stumpf, 1965; Kannangara
CG et al., 1973). Furthermore labeled acetate and laurate were elongated, desaturated
and assembled into both prokaryotic and eukaryotic molecular species in Arabidopsis

leaves (Norman and St. John, 1986).

Although these elongations of substrates shorter than 16-carbons (C16) were suggested to

occur via fatty acid synthesis (FAS) enzymes in the plastid, direct evidence for this

110

suggestion has not been presented. Nor is the mechanism of how these substrates get
activated prior to their involvement in condensation reactions known. De nova FAS in
plants involves condensation reactions between malonyl-ACP and growing chain of acyl-
ACP. Mattoo et al (1989) identified an in viva acylated protein following incubation with
exogenous radiolabeled palmitate as acyl-carrier protein in Spirodela oligorrhiza. They
hypothesized that the exogenous palmitate was first activated to an acyl-CoA derivative
and then was transesterified to form acyl-ACP by a side reaction of the condensing

enzyme 3-ketoacyl-ACP synthetase (KAS). In their study cell-free extracts fi'om

14 . . l4 .
Spirodela oligorrhiza formed [ C]palmitoyl-ACP when incubated With [ C]palmitoyl-

CoA and ACP. Cerulenin an inhibitor of fatty acid condensing enzyme 3-ketoacy1-ACP
synthetase I/II (KASl/II) inhibited in viva formation of acyl-ACP from palmitate but only
partially inhibited the formation of acyl-ACP from acyl-CoA in the cell free extract.
However, even if this is true with other higher plants, activating exogenous fatty acid into
CoA esters in the stroma where transacylation and elongation presumably occurs remains

questionable because to date there are no enzymes described that catalyze this reaction.

In this paper we characterize in more detail the elongation of exogenous fatty acids by
Arabidopsis leaves and isolated pea chloroplasts. Inhibitors of plastid fatty acid
elongation and cytosolic malonyl-CoA formation were used to provide information on
the subcellular sites of elongation. We have then used this system to test antisense and
knock out plants in candidate genes for acyl activating enzymes. These experiments have
allowed us to identify an Arabidopsis gene product that is likely involved in activation of

exogenous fatty acids in the chloroplast.

lll

RESULTS

14
Exogenous [I- CIF FA metabolism by Arabidopsis leaves

14
When detached leaves of Arabidopsis were incubated with C labeled FFA’s in the

light, the label appeared in essentially all classes of glycerolipids in 3 hours. Table 3
presents the distribution of radioactivity among the major glycerolipids after 3 hours of
incubation with short to long chain fatty acid substrates. In 3 h of incubation, shorter
fatty acid substrates (3 10 carbons) were incorporated preferentially into plastidial lipids
(monogalatosyldiacylglycerol (MGDG) ~24%, phosphatidylglycerol (PG) ~10%)
whereas longer substrates (2 14-carbons) were greatly preferred by cytosolic lipids (PC
~30%). Laurie acid showed an intermediate level of distribution between MGDG (14%)
and PC (22%). The distribution of label with acetate indicates that over the 3 h
incubation time labeling did not reach equilibrium in order to reflect in viva glycerolipid
compositions of Arabidopsis leaves. A previous study showed that when palmitic and
stearic acids were fed to spinach leaves they followed eukaryotic pathway being
incorporated into PC at first and subsequently were transferred into plastidial

galactolipids (Roughan et al., 1987).

. . . . . 14
Figure 11. displays the time course labeling pattern With [1- C]laurate. The label

initially went rapidly into PC and TAG. The incorporation into PC, PG and MGDG

increased approximately linearly until 90 min. In contrast, the rapid incorporation into

112

Table 3. Distribution of radioactivity among the major glycerolipids following 3

14
hours of incubation of Arabidopsis leaves with various [1- C] substrates

 

14
Exogenous [l- C] FFA substrates

 

 

 

2:0 8:0 10:0 12:0 14:0 16:0

Percent total radioactivity in the lipid extract (%)
TAG 1- - - 2.9 3.7 1.1
FFA 5.2 8.2 5 6.4 9 10
DAG 1 - 1.3 2.3 2.9 3.2
MGDG 23.9 24 25.6 14.2 5.6 2.8
PG 10.4 10.8 9.8 5.8 4.5 3.9
DGDG 1.9 2.5 2.7 1.8 1.2 1.2
PI,PS,SL 2.4 2.6 3.2 3.3 3.6 4.2
PE 1.9 2 2.7 3.6 4 6.1
PC 14.4 10.4 16.8 22.2 31.4 31.4
Sum 61.1 61.9 67.3 62.5 65.9 63.9

 

Note. Detached Arabidopsis leaves were incubated in 25mM NaMES (pH5.7) buffer

. l4 . . . . . .
contaimng [1- C] labeled substrates indicated in the table under illummation at 25°C for

3 hours.

1<l%.

113

14
Figure 11. Time course incorporation of [1- C]laurate into glycerolipids by

detached Arabidopsis leaf.
Detached Arabidopsis leaves (20-30 days old) were incubated in 25mM NaMES (pH5.7)

. . 14 . . .
buffer containing 10 pM [l- C]laurate under illumination at 25°C for 2.5 h. Samples

were removed from light and were frozen at designated time points. Averages of 3

different extractions are plotted with error bars (:tSD).

 

 

 

 

- 25 7 A * #9 777777777 6
= l
> i .—
E. l l E
O - -r
.5 . l. 5 g-
— 2
e l l a
0i A PC I 1:.
2 1.5 I FFA I 3
g I ‘- - l- l ’5
c: l 3
E l l" I 3
O I 2

l 0
e" a l E
8 l , 2
.s i g
2 i l :'
9 ] :=
=
.‘E l

 

time (min)

114

TAG ended by 30 min after which label in TAG decreased presumably through transfer

into other lipids or to degradation.

Analysis of total lipid extracts by transmethylation revealed the elongation and
desaturation pattern of each fatty acid substrates (Table 4). In all cases, label was found
in fatty acids longer than the substrates but there was an inverse relationship between
substrate chain length and efficiency for use in elongation and desaturation. With S C10

FA substrates over 95% of the radioactivity in the fatty acid extract was found to be

14 . 14
elongated or desaturated after 3 h of incubation. For [1- C]lauric acid and [1- C]

myristic acid 65% and 37% of the label in the lipid extract was elongated respectively.
Palmitate was the least suitable substrate for elongation (13%) and especially for
desaturation. Reduced desaturation in the labeled products was also observed as the
length of substrates used grew longer. The elongation reaction with little desaturation
might suggest direct elongation of the MCFA substrates by extraplastidial FA elongase

system some of which might be involved in wax synthesis in the epidermal cells.

The rate of exogenous laurate incorporation into longer fatty acids calculated at 20 min is
about 4.9 nmoles C12/h/mg chlorophyll (Figure 12A). This translates to about 11 nmoles
C2/h/mg chlorophyll (C14 20%, C16 50%, C18 28%, C20 0.64%, C22 0.64%, C24 1.3%,
C26 0.64%). For comparison, with IOuM acetate the incorporation rates into fatty acid
measured in spinach leaf and isolated spinach chloroplast are about 10—100 nmoles/h/mg
chlorophyll (Roughan et al., 1978; Pollard and Ohlrogge, 1999). Although it is difficult

to make direct comparison between incorporation of different substrates especially when

115

Table 4. Distribution of radioactivity among the total fatty acids following 3 hours

14
of incubation of Arabidopsis leaves with various [1- C] substrates

 

14
Exogenous [1- C] substrates

 

 

 

2:0 8:0 10:0 12:0 14:0 16:0
FAME
classes Percent total radioactivity in the lipid extract (%)
10:0 -2 - 1.3 - - -
12:0 0.9 1.0 0.6 32.3 - 1.3
14:0 0.6 1.3 1.7 8.2 61.1 0.9
16:0 24.4 30.3 33.2 25.5 22.7 75.]
16:1 2.7 3.7 3.0 1.8 0.8 0.2
16:2 3.4 5.0 3.8 2.5 0.8 0.3
16:3 2.6 4.6 4.7 2.8 1.1 -
18:0 3.9 3.1 3.3 2.5 2.7 6.8
18:1 28.7 19.4 20.3 10.0 3.1 1.1
18:2 13.7 11.2 11.3 5.8 1.9 0.3
18:3 9.0 8.4 11.5 4.9 1.7 -
2 20:0 3.8 5.9 4.3 2.1 2.6 6.3
Sum of
elongated 94.0 93.4 97.3 66.4 37.2 13.8
products

 

Note. Detached Arabidopsis leaves were incubated in 25mM NaMES (pH5.7) buffer

. . 14 . . . . . .
containing [1- C] labeled substrates indicated in the table under illumination at 25°C for

3 hours. At the end of the incubation, buffers were removed and leaves were washed with

water and blotted briefly on filter paper. The leaves were frozen with liquid nitrogen and

stored in -80°C until lipid extraction.

1 . . .
percentage of label in the fatty ac1ds longer than applied substrates.

2 < 0.2%.

they have markedly different uptake efficiencies, solubility, enzyme specificities and
cellular toxicity effects, nevertheless, the laurate elongation rate seems to be similar or

slightly lower than fatty acid synthesis rate from acetate.

. 14 . . . . . .
The time course of elongation of [1- C]laurate into indiVIdual FA’s (Figure 12B, minor

components are not shown) showed that the highest proportion of labeled fatty acid
species were palmitate followed by oleate and myristate. The steady state synthesis rates
of 14:0, 16:0, 16:1, 18:0 and 18:1 were maintained for about 30 min or less whereas more
elongated and desaturated species steadily increased until 90 min or more. Although
there was still laurate in the lipid extract at the 30 min time point, it is possible that
availability of substrates for elongation became limiting by 30 min. The more elongated
(2 C20 FA’S) and desaturated species (16:2/16:3, 18:2/18:3) continued to accumulate at

the expense of shorter and less desaturated precursors.
Effect of plastidial FAS inhibition an exogenous laurate elongation by leaves

It was previously assumed that the elongation of exogenously fed short and medium
chain fatty acids takes place in the chloroplast where de novo fatty acid synthesis takes
place (Hitchcock and Nichols, 1971; Norman and John, 1986; Thompson Jr. et al., 1986;
Roughan et al., 1987). However, no direct evidence for this assumption has been
presented and a mechanism to activate fatty acids to ACP in the chloroplast has not been
described. Plants also have the ability to elongated FA’s by a membrane bound ER

system which might provide an alternative to plastids. In order to investigate the

117

14
Figure 12. Time course elongation of [1- C]laurate substrate by detached

Arabidopsis leaf.
Detached Arabidopsis leaves (20—30 days old) were incubated in 25mM NaMES (pH5.7)

. . 14 . . . .
buffer containing 10uM [l- C]laurate under illumination at 25°C for 90 min. Samples

were withdrawn from light at various time points and individual fatty acids were

analyzed by TLC following transmethylation (see Material and Method). A. Overall [1-

C]laurate incorporations into > C12 FA’s are plotted. B. Elongation into indiVIdual

F A’s is shown. Averages of 3 different extractions are plotted with error bars (:tSD).

118

 

 

 

 

23:00.25 9: . 00.085
20:00:20 20.5.0.5e ..- E

 

100

 

time (min)

119

 

 

 

 

 

 

 

 

I 0
.12
o 1. .m "c
00°- 8 1 In >—
1 I 6
1 0 m 1
1 6
1
. . . .
B 6. 4. 2
0 O 0 O

A..>_.ao..o.:o mE . 0225..
9.0.3.893... 80.50.53- 5

 

100

80

60

4O

20

time (min)

120

cytosolic contribution to laurate elongation, cerulenin, an inhibitor of plastidial KASI/II
enzymes, was used. If the cytosolic contribution is major then we would expect laurate

to be elongated despite inhibition of plastidial fatty acid synthesis.

. . . . . 14 14
Detached Arab1dopsrs leaves were incubated w1th e1ther [1- C]acetate or [1- C]laurate

under illumination for l h in the presence of various concentrations of cerulenin. Figure

13A displays the effect of cerulenin concentration on laurate elongation by Arabidopsis

14 . . . . .
leaves. [1- C]acetate Incorporation into fatty aCldS was used as control of effectlveness

of cerulenin. Acetate incorporation was inhibited about 85% at SOuM of cerulenin and
95% at ZOOuM. Laurate elongation was inhibited with an almost identical dose
dependence. Two conclusions can be made from this experiment. First, the major
activity of laurate elongation into C16 and C18 F A’s is most likely associated with KAS
U11 and is localized to the plastid. Second, laurate apparently is a poor substrate for direct
elongation by cytosolic fatty acid elongation enzymes. Figure 13B shows cerulenin
effect on laurate incorporation into glycerolipids. Cerulenin inhibition of elongation had
almost no effect on laurate incorporation into PC and TAG. Since laurate was not able to
elongate in the presence of cerulenin, incorporation into PC and TAG most likely was
mediated by acyl transferases that had activity on laurate as substrates. On the other hand
label appearance in MGDG, PG and DGDG was greatly reduced by the inhibition of

elongation.

121

Effect of cytosolic FA elongation inhibition on exogenous laurate elongation by

detached Arabidopsis leaves

Acetyl-CoA carboxylase (ACCase) catalyzes activation of C02 and its transfer to acetyl-
CoA to form malonyl-CoA, the carbon donor of fatty acid synthesis. There are 2 types of
ACCase found in dicots with different sensitivity to herbicides such as haloxyfop. The
heteromeric ACCase in plastids that provides malonyl-CoA for de novo FAS is
insensitive, whereas the cytosolic homodimeric ACCase, that supplies malonyl-CoA for

fatty acid elongation is inhibited by haloxyfop (Bao et a1, 1998).

. . . . 14 . .
When detached leaves of Arab1dops1s were incubated With [1- C]laurate and Increasmg

amounts of haloxyfop, elongation into 2 C20 FA’s was selectively inhibited (60-70%)
whereas there was almost no effect on overall synthesis of FA’s up to C18 (Figure 14).
These data demonstrate that the initial elongation of laurate into palmitic and oleic acid is
not inhibited by haloxyfop and supports the conclusion from the cerulenin experiments
that this process occurs mainly in plastids. However, synthesis of 14:0 and 18:0 was
moderately reduced by haloxyfop. An interpretation consistent with this result is that a
portion of stearate synthesis from exogenous laurate occurs after palmitate is exported
from plastids. Furthermore, these data suggest that part of the 14:0 FA products are

derived from a haloxyfop sensitive entity, either cytosol or mitochondria.

122

- 14
Figure 13. Cerulenin effect on elongation of [1- C]laurate by detached

Arabidopsis leaves.

. . . . . 14 14
Detached Arabidop51s leaves were incubated With either [1- C]acetate or [1- C]laurate

as described previously with increasing concentrations of cerulenin. A. Cerulenin effect

on acetate incorporation into FA’s and laurate elongation. B. Cerulenin effect on [1-

14 . . . . . .
C]laurate incorporation into major glycerolipids.

A.
1201* -- *'** *m-———
100 i .
+ laurate elongation rate (> C12 FAs)
- - -a- - - acetate incorporation rate I
T:
b
i:
o l
o l
°\O

 

 

0 50 100 1 50 200 250

cerulenin concentration (pm

123

250

N
O
O

_3
01
O

100

°/o control laurate incorporation

01
O

 

i-—'—-T¢W§ 1'

i+oeoo l!
|---x---PG l!

 

 

50 100

150

cerulenin concentration (uM)

124

Contribution of recycled acetate units on the labeled products of exogenous FA

elongation by detached Arabidopsis leaves is small

The interpretation of the labeling patterns presented so far could be confused if the added

14 . . . . l4 .
[l- C]fatty ac1ds were SUbjCCt to B-ox1dation and C products were recycled into de
. . . . . . . . 14 . .
novo FAS. To evaluate this pOSSIbility we determined the distribution of C in the oleic

. 14 . . . .
ac1d product after [1- C]-fatty ac1d and acetate labeling. Ox1dative cleavage at the

double bond position of oleic acid methyl ester creates nonanoic acid (C9A) and 1,0)-

. . 14 . . . .
nonane-diow monomethyl ester fragments (C9AE). If C in 01610 18 derived from [1-
14 . . . . .

C]-acetate, the theoretical distribution of label between the two 9 carbon fragments 18

14 . . . . . l4 .
0.8. If C in oleic 18 derived only from intact [1- C] FA precursors and there is no

contribution of recycled acetate units, either C9AE or C9A will be labeled depending on

the length of precursor used.

Table 5 presents the distribution of label among the two oxidative fragments of 14C-

labeled oleic acid after Arabidopsis leaves were incubated with different [1-14C]FA
l4 . . . l4

substrates. C-oleic ac1d from leaves labeled With [1- C] acetate showed close to the

theoretical (0.8) C9A/C9AE ratio of 0.9. When [1-14C] labeled C10 through C14 FA’s

125

14
Figure 14. Haloxyfop effect on elongation of [l- C]laurate by detached

Arabidopsis leaf.

. . l4 . .
Detached Arabidopsis leaves were incubated With [1- C]laurate as described preViously

with various concentrations of cerulenin for 30 min under illumination. Samples were
withdrawn from light at various time points and individual fatty acids were analyzed by

TLC following transmethylation.

120 ~~~~~~~+~ ++~A4~

S 18 carbon FAs

 
   

100
a
E
g 804
.2
'0 ~__ _ .
E 60 9 .,¢:::::::::" X18-O
g "114:0
E
8 40
o\° 2 20 carbon FAs

20

0 . - ._ - __-_ . ,A ,‘
O 50 100 150 200 250

Herbicide concentration (uM)

126

14 . . .
were used as substrates almost all C label was recovered in the carboxylic half of oleic
. . . . l4 . . . . .
ac1d indicating recycled C-acetate was not a major contributor of label in the oleic ac1d.

[1-14C]Octanoic acid substrate labeled mostly the methyl end of oleic acid (83%).

These results are consistent with an early study made with spinach leaf slices labeled
with 6:0, 8:0, 10:0, 12:0 FA’s that showed little B-oxidative degradation and re-synthesis

when examined by reductive ozonolysis of the C18 unsaturated fatty acids (Kannanagara

et al., 1973).
Exogenous laurate elongation by isolated pea chloroplasts

Isolated pea chloroplasts (40iig chlorophyll) were incubated under illumination for 20

14 . . . .
min with IOuM [l- C]laurate (0.2 140). Figure 15A displays the time course of [1-

C]laurate incorporation into elongated F A’s. The overall rate of elongation into > C12

FA determined at the first 10 min was 22 nmoles C12/h/mg chlorophyll. This is about 4
times faster than that was measured in the intact leaf. The difference may reflect

difficulties for exogenous laurate to penetrate the leaf cuticle and leaf cells to reach the
plastid, cytotoxicity of unesterified laurate, competition by endogenous substrates, or

perhaps feedback regulation in the intact leaf compared to isolated chloroplasts.

127

14
Table 5. Label distribution among the oxidative cleavage fragments of C-labeled

14
oleic acid of Arabidopsis leaves incubated with different [1- C] labeled FA

substrates.

 

l4
[1- C] substrates 2:0 8:0 10:0 12:0 14:0

 

% total radioactivity in each oxidative cleavage fragments
C9A 37.2 82.8 4.4 3.4 1.6
C9AE 40.4 14.5 96.4 95.2 88.4
ratio between the oxidative cleavage fragments

C9A / C9AE 0.9 5.7 0.04 V 0.04 0.02

 

Note. Oleic FAME was recovered and cleaved at the double bond by permanganate-
periodate oxidation (Christie, 2003). The resulting nonanoic acid (C9A) and l,(0-nonane-

dioic monomethyl ester fragments (C9AE) were examined by TLC.

128

14
Figure 15. Elongation of exogenous [1- C]laurate by isolated pea chloroplasts.

14 .
A. Time course of [l- C]laurate elongation. Isolated pea chloroplasts (40pg

chlorophyll) were incubated under illumination in a medium described in the ‘Material

. . l4 .
and Methods’ section w1th lmM ATP and lOpM [1- C]laurate (0.2 uCi). Each data

point represents average value of 2 independent lipid extractions and analyses. B.

14 .
Characteristics of [1- C]laurate elongation. Isolated pea chloroplasts (80pg

chlorophyll) were incubated either in the presence or absence of light with lOuM [l-

14 . . . .
C]laurate (0.2 uCi) as substrate for 20 min. Indicated amounts of ATP, cerulenin,

unlabeled acetate were included in the media. Mean of 3 different extractions were used

to plot the graph.

129

16 0
18:1
14 0

 

 

 

 

snow—500.055 :03039005 0.0.50.

25

20

15

10

time (min)

130

laurate elongation rate (nmoleslhlmgChl)

 

12 . - - -- - - - - w- - A, A- .z_...____- - +,

 

 

1O

 

0.%.

Dark

1'

.....

 

 

 

 

 

 

 

 

Light

131

 

Light

(1011M
cerulenin)

 

 

 

 

.4.”

 

 

 

 

Light

(20mM
acetate)

Several characteristics of laurate elongation by isolated pea chloroplasts are shown in
Figure 15B. No detectable elongation was observed in the dark. Although the elongation
was not dependent on exogenously provided ATP, 1mM ATP stimulated the elongation
about 1.4 fold in the light as observed also for fatty acid synthesis from acetate (Koo et
al., 2004). Low concentrations (1-2 mM) of unlabeled acetate increased laurate

incorporation into elongated FA’s (data not shown). However 20 mM unlabeled acetate

. . . 14 .
inhibited the incorporation of exogenous [l- C]laurate into elongated FA’s to less than

. 14
18% of control (without unlabeled acetate). The elongation of [1- C]laurate was

sensitive to cerulenin and was inhibited in a manner that is similar to the inhibition of [1-

14 . . . . . .
C]acetate incorporation into fatty aCids by cerulenin. Overall elongation of laurate was

reduced 60% with IOuM cerulenin (Figure 15B) and almost completely with SOuM
cerulenin (data not shown). These experiments with isolated chloroplast indicate that the
chloroplast has the ability to elongate exogenous MCFA through the de novo FAS

system.
How do exogenous FA ’s get activated in the plastid?

The data presented so far indicate that the majority of the exogenous MCF A elongation
activities are associated with plastid de novo FAS. But how do the exogenous FA’s get
activated prior to their use in FAS in the plastid? The existence of a pathway in plants for
esterification of exogenous FA to ACP has only been inferred in past studies

(Kannangara et al., 1973; Terzaghi, 1986; Thompson Jr. et al., 1987). Mattoo et al.

132

(1989) identified an acylated soluble protein as ACP following the incubation of
Spriodela plant culture with radiolabeled palmitate. They hypothesized that exogenous
FA was first activated to the acyl-CoA derivative and subsequently transesterified to ACP
via a side reaction of the FAS condensing enzyme, (Alberts et al., 1972). In support of
this hypothesis cell free extracts from Spirodela were able to catalyze the formation of
palmitoyl-ACP from palmitoyl-CoA whereas the cell free system could not esterify free
palmitic acid to ACP. Even if we suppose this idea is true we still need either an enzyme
to catalyze formation of long chain acyl—CoA in the stroma / inner membrane of plastid
where FAS occurs or a transport system to introduce long chain acyl-CoA formed at the
cytosolic side of plastid double membrane into the site of FAS. An alternative possibility
might be direct formation of acyl-ACP from free exogenous FA’s in the stroma / inner
membrane. Acyl-ACP synthetase (EC 6.2.1.20) has been isolated from bacteria and yeast
(Rock and Cronan Jr., 1981; Fice et al., 1993; Jackowski et al., 1994; Gangar et al.,
2001). Lessire and Cassagne (1979) also reported the direct synthesis of stearoyl-ACP
from stearate in a microsomal fraction from Allium porrum epidermal cells although at a

much slower rate compared to acyl—CoA formation (about 1%).

We used two different approaches to understand the activation process of exogenous F FA

in the plastid. First we examined acyl-ACP and acyl-CoA pools formed during the

. 14 . .
elongation of exogenous [1- C]laurate in the isolated pea chloroplasts. Second, we

tested several Arabidopsis mutants/transgenics disrupted in gene candidates that might be

involved in the acyl activation processes.

133

14
Intermediates of exogenous [1 - C]laurate elongation in isolated pea chloroplasts

The procedure of Mancha et al. (1975) was used to isolate the acyl-ACP fraction from

. . 14 14 . .
pea chloroplasts incubated With [1- C]laurate or [1- C]acetate. Lipids were first

removed by extraction with organic solvent and proteins were subsequently precipitated
in the presence of ammonium sulfate and chloroform / methanol. The acyl esters in the
pellets were then transmethylated and separated from any FFA contaminants by TLC
(Mancha et al., 1975; $011 and Roughan, 1982). In order to verify that the recovered
acyl-ACP fraction did contain acyl-ACP molecules, the time course labeling pattern of
acyl-ACP fractions in the light and in the dark were examined as shown in Figure 16.
Acetate incorporation into total lipid increased linearly in the light. Upon removal of
light, acetate incorporation stopped immediately and resumed with re-illumination. The
labeling pattern of acyl-ACP fi'action with both acetate and laurate showed typical
labeling pattern of intermediates with high turn over rates. Within about 1 min of
labeling in the light, acyl-ACP fraction reached saturation and did not increase further in

the light. In the dark, label disappeared rapidly and completely. The estimated pool size

14
of [1- C]acyl-ACP was 0.5-2.3uM assuming a chloroplast volume of 47 uL/mg

chlorophyll (Wirtz et al., 1980). The reported total ACP molecule content in spinach
chloroplast was 8uM (Ohlrogge et al., 1979) and total acyl-ACP in the acetate labeled

spinach chloroplast was estimated at SuM (S011 and Roughan, 1982).

134

Figure 16. Time course light synthesis and dark catabolism of acyl-ACP by isolated

pea chloroplasts.
Isolated pea chloroplasts (90ug chlorophyll) were incubated under illumination and in the

. . . . 14
dark in a basal medium (Materials and Methods) With either 0.2mM [1- C]acetate (5

14
uCi) or 13 uM [l- C]laurate (0.36 uCi) as substrates. Total lipids as well as acyl-

ACP’s were extracted following direct quench at each time points. Acetate incorporation

14 14
into total lipid (A) follows left axis scale. [1- C]Acetate (Cl) and [1- C]laurate (I)

incorporation into acyl-ACP fractions are depicted following right axis scale.

 

 

 

 

 

 

 

 
  
 
  
    

45 - . -~ -- , . i 0.25
. Light ’ ‘ . "Dark ; Light 1‘

40 g I
l t tal Ii ids 4"

35: n 9.“ 0 P " 0.2
i .

30 '

i
251

l

20;

   

acetate fed

(nmoles/mg chlorophyll)

acetate incorporation into total lipid
("Aqdmomo Btu/septum
dOV'IKOV ow! uogeiodioau! eieiisqns

A, lau rate fed

 

time (min)

135

To detect radiolabeled acyl-CoA in these incubations, several different extraction
methods were employed. We tested (i) direct separation of the reaction mixture by TLC
(butanol/acetic acid/water developing system) following rapid quench and solubilization
With chloroform/methanol (1/ 1), (ii) a modified version of acyl-CoA separation
procedure developed by Mancha et al.(1975), and (iii) butanol extraction of aqueous

phase after lipid extraction (Bligh and Dyer,1959) followed by TLC separation. With

14
each method, recovery of [1- C]-acyl-CoA standards was at least 60%, but we could

detect less than 0.1 nM (or < 1/ 1000 of acyl-ACP concentration) of labeled acyl-CoA in

14 . . .
the isolated pea chloroplasts labeled With [1- C]laurate. These results indicate that, in

contrast to acyl-ACP formation, almost no detectable acyl-CoA intermediate was formed

during the elongation process.

Laurate elongation activity in Arabidopsis mutants disrupted in expression of

candidate genes for plastid acyl activation

To further investigate the activation of laurate to ACP by chloroplasts we took a
bioinformatics approach to select potential Arabidopsis genes that might be involved in
MCFA activation in the plastid (Table 6). A superfamily containing 63 different genes
encoding acyl-activating enzymes (AAE) in Arabidopsis genome has been described
(Shockey et al., 2003). Among those were 3 members of acetyl-CoA synthetases (ACS 's)
and 11 members of long chain acyl-CoA synthetases (LACS’s). LACS ’s were

characterized in detail previously for their ability to complement a LACS-deficient strain

136

of yeast and for the in vitro LACS enzyme activity in the cell free lysates of yeast or E.
coli cells overexpressing the candidate LA CS genes (Shockey et al., 2002). Nine of these
candidates were found to have LACS enzyme activity. One of those 9 members, LA CS9
(Atlg77590), was demonstrated to be localized to the envelope of the plastid (it is not
known whether LA CS9 is located on the outer or inner envelope membrane) and was
shown to be responsible for the major in vitro LACS activity on the plastid envelope

(Schnurr et al., 2002).

Because laurate elongation was associated mainly with plastids, we selected candidates
that were reported to be plastidial or had plastid targeting sequence as predicted by Target
P (Emanuelsson et al., 2000, http://www.cbs.dtu.dk/services/TargetP/). One of above 3
ACS’s and 4 of 11 LACS’s had predicted plastidial targeting sequences with reliability
classes ranging from 1 through 5 (Table 6). At5g36880 was reported to be a plastidial
ACS in a previous study (Behal et al., 2002). In vitro chloroplast import assays of LA CS8
(At2g04350) in the studies by Schnurr et a1 (2002) was reported to be positive. AAE15
(At4g14070) and AAE16 (At3 g23 790) had sequence elements leading to categorization as
LACS 's along with other 9 LA CS members before. The predicted proteins have a high
reliability class for plastidial targeting. Shockey et a1. (2002) reported that both AAE15
and AAE16 could not complement the mutant phenotype of a yeast mutant deficient in
LACS (C14-C18 FA’s were tested) and neither had activity in in vitro assays using cell-
free lysates of yeast expressing AAE15 and AAE16 and using oleic acid as substrate.
Overexpression strains of E. coli also lacked LACS activity for 14:0, 16:0, 16:1, 18:0,

18:1, 18:2, 18:3 and 20:1 FA substrates (Shockey et al., 2002). In a study of fatty acid

137

export from plastids, Koo et al. (2004) raised the possibility of AAE15 and AAE16 as
“cryptic” LACS’s that are perhaps linked to FFA substrates coming out from the stroma

upon release from acyl—ACP thioesterase.

To test the possible involvement of these candidates in exogenous FFA activation in the

plastid we took a reverse genetics approach. The idea is to test the ability of Arabidopsis

. . . . 14
mutants impaired in those selected gene candidates to elongate exogenous [l- C]laurate

substrate. ACS gene silenced lines were kindly sent to us for test by Dr. David J. Oliver
from Iowa State University (Behal et al., 2002). These T2 seeds had about 15% to 30%
of wildtype ACS activity (Oliver DJ, personal communication). T-DNA insertion mutant
lines for LACS8, AAE15 and AAE16 were searched from the Sequence-Indexed Library
of Insertion Mutations generated by the Salk Institute Genome Analysis Laboratory using
Arabidopsis Gene Mapping Tool (http://signal.salk.edu/cgi-bin/tdnaexpress).
Homozygous KO lines were selected among the T3 generations by PCR screening for

both the presence of T-DNA insertion and the absence of an intact gene. The lines tested

14 . . . .
for [l- C]laurate elongation actiVity were T4 generation homozygous knock-outs. The
. . th . . .
insertion was at the 8 intron for LACSS KO (lacs8, ABRC stock identifier SALK_

. d .
136060) and at the 7th intron and 2n exon respectively for AAE15 KO (aae15, SALK_

139663) and AAE16 KO lines (aae16, SALK_067859).

138

Table 6. Summary of gene candidates of exogenous FFA activation in the plastid

 

 

tested in this study.
1 2
subcelluar TM
gene MIPS description Locus mutant lines
localization domain
ACS Acetyl CoA synthetase At5g36880 C (5) 1 Antisense
Putative acyl-CoA
LACS8 At2g04350 C (5) l SALK_136060
synthetase
AAE15 AMP-binding protein At4gl4070 C (2) none SALK_139663
AMP—binding protein,
AAE16 At3g23790 C (1) none SALK_067859
putative

 

1 . .
Target P (web server at http://www.cbs.dtu.dk/serv1ces/TargetP/) was used to predict

subcelluar localization. Reliability class (RC) values are shown in parenthesis.

2
TMHMM v2.0 (web server at http://www.cbs.dtu.dk/services/TMHMM-2.0/ ) was

used for or helical transmembrane domain prediction.

139

Detached leaves from ACS gene silenced lines, lacs8, aae15/aae16 and wild type

. . . . . . 14
(ecotype Columbia) ArabldOpSIS were incubated for 20 min With [1- C]laurate.

Displayed in Figure 17 are an image of radiolabeled FAME classes separated on silver
nitrate impregnated TLC plate (A) and a corresponding graph presents the laurate
elongation rates (B). Bands corresponding to 18:1, 16:1, 18:2, 16:2 and 18:3/16:3 (orders
from top to bottom) are greatly reduced in aae15/aae16. There was about 80% reduction
in overall elongation of laurate into both C16 and C18 FA’s. No significant changes

were observed with ACS antisensed lines and lacs8.

Elongation into 14:0 FA was apparently not affected. It has been noted previously
(Figure 13A and Figure 14) with inhibitors of cytosolic and plastidial FAS that at least
some portion of the elongation into 14:0 was contributed by both haloxyfop sensitive (i.e.
ER, mitochondrial) and cerulenin sensitive (i.e. plastid, mitochondrial) entities. The fact
that elongation into 14:0 FA was not blocked in the plant that were knocked out in
elongation activity of exogenous laurate into C16 and C18 FA might suggest that when
plastidial elongation pathway gets blocked cells may redirect the flux through ER or

mitochondrial pathways.

In order to find out which gene among AAE15 and AAE16 was responsible for the defect
in exogenous MCFA elongation activity, single knock-outs were tested for elongation

activity (Figure 18). Since it is also possible that there could be substrate specificities

among the two gene products, we tried 4 different substrates, [1-14C]labeled 2:0, 8:0,

140

14
Figure 17. Exogenous [1- C]laurate elongation by wild type and mutant lines of

candidates for plastidial acyl activating enzymes.
Detached leaves of each lines were incubated in 25mM NaMES (pH5.7) buffer

. . . . 14 . . . .
containing indicated [1- C]FFA substrates under illumination for 20 min. A. Image of
14 . . .
[1- C]FAME classes separated on Silver nitrate impregnated TLC plate. B. [1-

14 . . . .
C]Laurate incorporation rate into major F A’s.

 

A Iacs8 ACS Wt aaet5/aae16
silenced Col

.,

 

 

 

 

   

saturated '

 

16:1

18:2 m m «9. MM,

. . 11.4) J. p; ‘.;_ . . -- i. ‘t'f‘ ‘1 1. ~ '_
163/18.3 m I“ Wm m

141

 

 

w

 

 

laurate elongation rate (pmoles/mgChI/h)

2500 __, , fl , f 7 f - A- - - _---.---,----- ,. , _. _.‘
l “—_—' " *f'j‘
1 “8088 it
2000 ‘ aACS silenced 1i

   

1

1

i Dwtcoi.
1500-1 1

l
[aae15/16 1‘

 

  

1000.:
1

500;
1

o.

14 carbon 16 carbon 18 carbon

fatty acyl products

142

l4 . .
12:0 and 14:0. [1- C]Acetate was used as a control to test whether the disruption of

MCFA elongation was due to some defect in FAS machinery other than activation
process. There were no differences in acetate incorporation rate into FA’s in both single

knock-outs (aae15, aae16) and double knock-out (aaeI5/aae16) compared to wild type

14 . .
(WS). Elongation of [1- C]8:O into both C16 and C18 FA’s was conSIStently reduced to

10% or less of Wild type elongation rate, in the aae15 and aae15/aae16 plants.

Elongation activity in aae15 and aae15/aae16 was reduced to about 20% wild type with

14 . .
[l- C]12:0 substrate. Both aae15 and aae15/aae16 retained about 60% of elongation

. . 14 . .
activity into C16 FA With [1- C]l4:0 substrate. The observation that elongation was

reduced to a lesser degree with longer substrates is consistent with the idea that some
portion of elongation activity was coming from extra plastidial origin and that it had
preferences for longer substrates. With all substrates tested there was no apparent change
in the aae16 plant. We conclude that disruption of the AAE15 gene product was

responsible for defective elongation of exogenous MCFA by Arabidopsis leaves.

143

14 14
Figure 18. Elongation of [1- C]acetate and [1- C]MCFA substrates by detached

leaves of wild type and single/double knock outs of AAE15 and AAE16.

Detached leaves of each lines were incubated in 25mM NaMES (pH5.7) buffer

. l4 . . . . . .
containing indicated [1— C]substrates under illumination for 20 min. Ind1v1dual fatty

acids were analyzed by TLC following extraction and transmethylation.

‘ TTTTT "516071301
'18 carbonl
1

     

1
1
1
1
1
1

incorporation rate (nm oleslm gChllh)

 

aaeI5/aae1 6

          

aae15 aae16
g 12:0
E - -- if 7 7 4-
3,401 916 carbon 1
TE” 301 '18 carbon 1
% ' ;
5
$2.01
51.01
€0.01 L‘
3

WS aae15 aaeIS/aae16

aae16

   
  

fiehfifl
'18 carbon

 

1
1
1

aae16 aae15/aae16

.0

elongation rate (nmoles/mgChl/h)

g 14:0

E 2.01 ,7_. A 7 ' *—*
Ea ‘ '16 carbon 1
\ I

E 1.51 18 carbon 1
o 1
E ‘ 1
E10 1
.s as 1
a 1
g 0.0 .

" WS aae15 aael 6 aae15/aae16

144

DISCUSSION

14
Exogenous [1- CIF FA metabolism in Arabidopsis leaves

Metabolism of exogenously provided fatty acids by plants has been studied in the past
with emphases on their incorporation into endogenous lipids (James, 1962; Hawke and
Stumpf, 1965; Kannangara CG et al., 1973; Norman and St. John, 1986; Roughan et al.,
1987). The incorporated fatty acyl moieties were often found elongated and desaturated
as well. In this study we have tried to dissect further the compartmentation and
metabolism of exogenous FA’s in Arabidopsis detached leaves with an emphasis on
understanding where and how the exogenous MCFA gets elongated. We used
Arabidopsis leaves as a model system in order to take advantage of genomic information

and mutants that could guide reverse genetic strategies.

Fatty acid molecules that are entering from outside the cells might be expected to
partition into virtually all endomembrane systems. However we are focusing on
organelles that are known to be involved actively in fatty acid metabolism. The main
subcellular fractions to consider in terms of fatty acid metabolism in plant leaves are
endoplasmic reticulum, plastids and peroxisomes. In addition biochemical studies of pea
leaf mitochondria indicated that mitochondria are capable of de novo FAS beginning
from malonate (Wada et al., 1997) although the activity was much lower (< 1%) than

plastidial F AS.

145

The endogenous substrates for glycerolipid acyltransferases are mainly 18:1, 16:0 and
18:0. The exogenous FA may either be used directly for acyltransfer reactions or first
elongated and then be used. Incorporation of exogenous FA’s into PC and TAG
preferred longer chain (2 l4 carbons) FA substrates whereas shorter fatty acid substrates
(5 10 carbons) were preferentially incorporated into the plastidial lipids (Table 3). It may
appear that plastidial glycerolipids accepted shorter substrates as their substrates.

However in reality it was the opposite. The time course labeling pattern of lipids by

14 . . . .
exogenous [l- C]laurate (Figure 11) indicated that the direct use of exogenous laurate

in PC and TAG synthesis was considerable. Moreover this reaction was insensitive to
cerulenin inhibition of de novo FAS (Figure 13B). On the other hand laurate
incorporation into MGDG, DGDG and PG were severely affected by cerulenin treatment
suggesting they must be elongated to appropriate chain lengths prior to their fiirther
metabolism by the plastid. The poor incorporation of 14:0 and 16:0 into plasitidial lipids
may be due to either a preference by cytosolic enzymes, resulting in these fatty acids
being sequestered before they reach the plastid or alternatively their poor usage in the
plastid elongation reactions. Over 95% of the radioactivity in the fatty acid extract was
found to be elongated after 3 h of incubation with 5 C10 FA substrates (Table 4). The
percent elongation decreased inversely with an increase in chain length with 2 C12 FA’s.
In addition when palmitate was fed to isolated pea chloroplasts, there was no obvious
synthesis of oleic acid after 30 min although most of the palmitate was in free available

form (data not shown).

146

The TAG content of leaves is usually minimal (<1% of lipid). However, when leaves
were fed exogenous laurate, over 15% of label appeared in TAG at 30 min. This
incorporation was transient and after 30 min, the label in TAG started to decrease While
incorporation into other glycerolipids continued. This redistribution does not seem to be
due to the depletion of FFA substrates because about 25% of the total label was still
recovered as FFA at 30 min and the decreased label in the TAG can not account for the
increased label in PC, MGDG and other major glycerolipids which continued over 90
min. Rapid incorporation of exogenous FFA into TAG and transfer to other destination

could be general response of cells to deal with excess F FA’s in leaves.
The plastid is the major site of exogenous M CFA elongation into C16 and C18 FA ’s

Stearoyl ACP desaturases (ECl.14.19.2) which introduces the first double bond in 18:0-
ACP reside in the chloroplast. Desaturation of palmitate at A3 and A7 in higher plants
also occurs in the chloroplasts (Browse et al., 1985; Kunst etal., 1989). Thus the level of

desaturation may serve as a marker for exogenous FA entry into the chloroplast. The

. 14 14 . .
saturated to unsaturated ratio of C products formed from [1- C]octan01c and decanOic
. . . . 14 . .
ac1d treated samples were slightly higher (0.8) than With [1- C]acetate (0.6). This ratio

14
increased 5 fold With 11- C]myristate and up to 16 fold with [1-14C]palmitate (Table 4).

James (1962) also observed that although palmitate and stearate were incorporated into
leaf lipids they were not able to act as precursors for oleic, linoleic and linolenic acids in

leaves of Ricinus Communis after 3 to 5 h incubation. However desaturation of 16:0 and

147

18:0 into 16:3 and 18:3 was observed in spinach leaves with longer incubation time (> 8
h) (Thompson Jr. et al., 1986; Roughan et al., 1987). Although it was not shown
explicitly that this was not through incorporation of acetate unit generated by B-

oxidation, the authors argued that there being minimal amount of labeled C16 fatty acids

. . l4 . . . . . . .
in leaves treated w1th [ C]stearate is indirect eVidence that contribution by FA recycling

was small (Roughan et al., 1987). Elongation of exogenous F A’s with little desaturation
might suggest direct elongation of the MCFA substrates by extraplastidial FA elongase
system some of which might be involved in wax synthesis in the epidermal cells (Kunst

and Samuels, 2003).

14 . . . . .
Isolated pea chloroplasts were able to elongate [1- C]laurate into myristic, palmitic, and

oleic acids at rates faster than measured with the intact leaves demonstrating their ability
to incorporate exogenous MCFA into de novo FAS machinery. However, this doesn’t
exclude the possible contribution of other sources in exogenous MCFA elongation in
Whole leaf system. We used inhibitors to examine the mechanism and subcellular site of
MCFA elongation in the whole leaf. Cytosolic FA elongation requires malonyl-CoA
production by eukaryotic ACCase. In addition there is no mitochondrial isoform of
ACCase found in Arabidopsis, suggesting it is dependent on cytosolic production of
malonate. The eukaryotic homodimeric ACCase is sensitive to several herbicides that
have no effect on the multisubunit ACCase (Burton et al., 1987,1991; Ohlrogge and
Browse, 1995). On the other hand the 3-ketoacyl—ACP inhibitor, cerulenin is expected to

inhibit the plastidial and mitochondrial elongation system, allowing only cytosolic

148

elongation to operate. About 85% of acetate incorporation into fatty acids by
Arabidopsis whole leaf was blocked by 5011M cerulenin and 95% by ZOOuM cerulenin.
Cerulenin was slightly less (5%) effective on exogenous laurate elongation than on
acetate incorporation (Figure 13A). Although natural substrates for the cytosolic
elongation system are 16:0, 18:1 and 18:0 exported from plastid, our data suggested that
5-10% of the laurate elongation may occur through direct elongation of laurate by a
cytosolic elongation system. However over 90% of the elongation activity comes either
from plastid or mitochondria. The blockage in the elongation into very long chain fatty

acids was a secondary effect of reduced 16:0, 18:0, and 18:1 production. When detached

. . . . 14 . . .
leaves of ArabidopSis were incubated With [1- C]laurate w1th increasmg amounts of

haloxyfop, laurate elongation into very long chain (2 C20) FA’s were selectively
inhibited. Laurate elongation into C 16, C18 FA’s especially into the unsaturated series
was increased or not affected. The elongation not being sensitive to haloxyfop provides
further evidence that the plastid is the major site of MCFA elongation into C16/C18

FA’s.

Perhaps the most critical evidence of the plastid being the major source of MCFA
elongation into C16/C18 FA’s came from the knock-out plant impaired in a gene
predicted to be targeted to the plastid. In aae15, elongation activity was reduced down to
10-20% of the wild type. Together these data confirm the plastid as the major source of

MCFA elongation into C16/C18 FA’s.

149

A possibility of an additional contribution by mitochondria may be suggested at least

with the accumulating pool of 14:0. Myristic acid is usually a minor component in the

. . . . 14
leaves/isolated chloroplasts incubations With [1- C]acetate. However, an unusually

. 14
high portion of label was found in 14:0 FA when [1- C]laurate was used as substrate

(Table 4, Figure 15). This activity was sensitive to KAS inhibition by cerulenin. Unlike
unsaturated C16 and C18 F A’s, accumulation of 14:0 was affected, although not as much
as elongation into very long chain FA’s, by the cytosolic ACCase inhibitor, haloxyfop.
Since Arabidopsis mitochondria don’t possess their own ACCase they are expected to
rely on cytosolic production of malonate for FAS and thus thought to be sensitive to
haloxyfop treatment. Mitochondrial FAS system is also sensitive to cerulenin (Wada et

al., 1997). Laurate elongation into 14:0 was not affected in AAE15 knock-out plants

. . . 14 . . .
further supporting a separate origin for the C-l4:0 accumulating in intact leaves. The

14:0 accumulation in isolated chloroplast might therefore be due to contamination of
different fractions in the isolated plastids preparation. It is also possible that FA
synthesis system in damaged chloroplast may contribute to certain extent where coupling
between elongating products and FAS system is less tight thereby allowing growing

chain of intermediates to escape prematurely.

Identification of acyl activating enzyme in the plastid through reverse genetics

approach

150

The above analysis of subcellular sites of elongation does not identify the crucial step
needed to activate acyl chains before their elongation. All elongation steps in de novo
FAS involve condensation between malonyl-ACP and a growing chain of acyl-ACP and
therefore exogenous FFA must be activated prior to involvement in any further FAS
reactions. Likewise the desaturation of exogenous palmitate and stearate observed in the
prolonged incubation of spinach leaves that was mentioned earlier (Thompson Jr. et al.,
1986; Roughan et al., 1987), presumably requires activation of free FA substrates in the
plastid since the enzymes that introduces the first double bond to pahnitoyl ACP or
stearoyl ACP resides in the plastid (Browse et al., 1985; Kunst et al., 1989) unless there’s

palmitoyl-, stearoyl-CoA desaturase in the cytosol. Acyl-CoA synthesis in isolated

. l4 .
chloroplasts incubated w1th [ C]acetate lS absolutely dependent on exogenous ATP and
coenzyme A as cofactors either in the dark or in the light (Koo et al., 2004). However,
exogenous MCFA elongation in isolated chloroplasts is independent of exogenous ATP
or coenzyme A (Figure 15B). This implies an activation reaction that is separate from the
described LACS activity of long chain fatty acids in chloroplasts. There is no described

fatty acyl-activating enzyme in the stroma; neither acyl-CoA synthetase nor acyl-ACP

synthetase. The Arabidopsis mutant that had T-DNA insertion in AAE15 gene was

. . . . . 14
normal in fatty ac1d syntheSIS act1v1ty when [1- C]acetate was used as substrate

showing that it possessed intact FAS enzymes (Figure 18). However elongation of

14
exogenous [l- C]MCFA was severely reduced. AAE15 (At4g14070) gene has

similarity to other known LACS’s and contains sequence elements characteristic for

LACS’s including the classical AMP binding domain and additional linker domain

151

(Shockey et al., 2002) but is distinguished from the other 9 members of LACS family
because it was not able to complement the mutant phenotype of yeast deficient in LACS

nor had LA CS activity in the cell-free lysates of both yeast and E .coli over expressing the

gene (Shockey et al., 2002).

Both the subcellular localization predictors (both ChloroP and TargetP) and the
biochemical data presented in this paper indicate AAE15 is a plastidial enzyme. AAE15
was also represented in the list of recently identified Arabidopsis plastid envelope
proteins through mass spectrometry based proteomics approach (Ferro et al., 2003).
There is a disagreement concerning predicted on helical transmembrane domains between
different predictors. TMHMM v2.0 (Krogh et al., 2001) and TopPred 2 (Claros and von
Heijne, 1994) did not detect any obvious or helical transmembrane segments whereas
TMpred (Hofmann and Stoffel, 1993) and DAS (Cserzo et al., 1997) predicted 1
transmembrane domain region. However, it isn’t clear yet exactly how AAE15 is
involved in exogenous FA elongation. As mentioned above elongation in the isolated
chloroplast occurs without an extra supply of cofactors. As shown in Figure 16 there are
at least two possible ways of converting exogenous FFA into acyl-ACP. One is via
forming acyl-CoA intermediate first and then transacylating it into acyl-ACP. Bacterial
FAS condensing enzyme was reported to have acyl-CoAzACP transacylase activity as a

side reaction (Alberts et al., 1972). Plastidial KAS might also be able to catalyze this

14
step. Mattoo et al., (1989) reported formation of [ C]palmitoyl-ACP when a cell-free

extract of Spirodela oligorrhiza was incubated with [14C]palmitoyl-CoA and ACP.

152

Cerulenin inhibited in vivo formation of acyl-ACP from palmitate supporting the idea of
KAS involvement in this step but only partially inhibited the formation of acyl-ACP from
acyl-CoA in the cell free extract indicating alternative ways to this reaction end. A
second possible route of acyl—ACP formation is the direct esterification of free fatty acid
with ACP. Acyl-ACP synthetase (EC 6.2.1.20) has been isolated from bacteria and yeast
(Rock and Cronan Jr, 1981; Fice et al., 1993; Jackowski et al., 1994; Gangar et al., 2001)
but not in plant. Lessire and Cassagne (1979) reported the direct synthesis of stearoyl-
ACP from stearate in a microsomal fraction from Allium porrum epidermal cells although
at a much slower rate compared to acyl-CoA formation (about 1%). Our analysis of

intermediates formed during laurated elongation favor the direct formation. When

. . 14 .
isolated pea chloroplasts were incubated With [1- C]laurate, and elongation was
. 14 .
allowed, rapid formation of [l- C]acyl-ACP was detected (0.5-2.3 uM)(Figure 16).

14
However, with three methods we failed to find detectable levels of [1- C]acyl—CoA

intermediates. If there were indeed acyl-CoA intermediates the pool size is expected to
be at least one thousand fold lower than acyl-ACP concentration in the stroma. However,
we cannot rule out a very small pool of acyl-CoA is an intermediate in the formation of

the acyl-ACP.
What is the in vivo function of FA activation/elongation in the plastid?

Exogenous fatty acids are important energy sources for unicelluar organisms and animal

cells. However intercellular transport and exchange of lipid molecules is minimal among

153

plant tissues. Thus activation and elongation of exogenous FA may look irrelevant to
‘normal’ physiological function of plants. Indeed aae15 does not show any visible
phenotypes under normal grth conditions in the growth chamber. However, the gene
product of AAE15 may become important under certain conditions. AAE15 has 11 EST’s
in the public EST libraries and thus is expressed at levels similar to many enzymes of
lipid metabolism. Based on the reduced elongation of 8:0, 12:0, and 14:0 in aae15
(Figure 18), the substrate specificity of AAE15 toward medium chain fatty acids may be
broad. This suggests one hypothetical function of AAE15 might be membrane editing
and proof reading of de novo FAS. Fatty acid elongation during de novo FAS in the
plastid halts when the growing chain of fatty acid gets hydrolyzed from ACP by action of
fatty acyl-ACP thioesterase. If the shorter fatty acids escape FAS system prematurely and
gets released into the membrane, there may need to be mechanisms to recapture them for
return to the regular process. Also there could be occasions other than biosynthetic error
Where potentially harmful free fatty acids can get released from the membrane, including
stress conditions and mechanical wounding. In this case the released FFA’s are expected
to be the long-chain fatty acids which do not need to be elongated. Perhaps AAE15 may
act as both acyl—ACP synthetase and acyltransferase. The E. coli AAS gene product was
shown to possess 2 enzymatic activities that work in concerted manner; first, to activate
exogenous free fatty acid into acyl-ACP and second, to transfer the acyl group to 2-
acylglycerophosphoethanolamine to synthesize phosphatidylethanolamine. It is possible
that AAE15 protein recycles the released FFA’s and lysoglycerolipids under stress

conditions. Nine out of 11 AAE15 EST’s are derived from libraries that were constructed

154

Figure 19. Phylogenetic comparison of AAE15 and other LA CS genes.

The deduced full-length amino acid sequences of the genes that contained both the
AMPBP signature motif and the linker domain were aligned by T-Coffee (Notredarne et
al., 2000, web server at http://www.ch.embnet.org/sofiware/TCoffeehtml) and was
displayed as an unrooted nearest neighbor phylogenetic tree using the TreeView program

(Page et al.,1996).

AAE1 5 AAE16
cyano BAB75301
LACSZ
LACSS
cyano BAC08853
LACS4
LAC83 LACSS
LACS1
LACSB

LACS? LACSG

155

 

after dehydration and cold treatments. The expression pattern of AAE15 in these
conditions may give us clue to the in vivo function.

One other interesting characteristic of AAE15 is that it’s closest homolog other than
AAE16 and one of Brassica LACS ’s (accession: CAA96521, Fulda et al., 1997) was the
long chain fatty acid CoA ligase from cyanobacteria Nostoc sp. PCC 7120 (accession:
BAB75301) (Figure 19). Their expected location in the chloroplast and their similarity to

cyanobacterial LA CS makes it interesting topic to study in respect to their origin.

MATERIALS AND METHODS

Materials

Wild-type Arabidopsis thaliana (ecotype Columbia (Col) was used in all detached leaf
incubation experiments unless otherwise indicated as Wassilewskija (WS).

ACS gene silenced lines were kindly provided for us to test by Dr. David J. Oliver from
Iowa State University (Behal et al., 2002). T-DNA insertion mutant lines for LACS8
(At2g04350, seed stock identifier SALK_136060), AAE15 (At4g14070, SALK_139663)
and AAE16 (At3g23790, SALK_O67859) were searched from the Sequence-Indexed
Library of Insertion Mutations generated by the Salk Institute Genome Analysis
Laboratory using Arabidopsis Gene Mapping Tool (http://signal.salk.edu/cgi-
bin/tdnaexpress) and the seed stocks were ordered from Arabidopsis Biological Resource
Center (ABRC). The double knock-out line of AAE15 and AAE16 (aae15/aae16) was

kindly provided by Dr. John Browse (Institute of Biological Chemistry, Washington State

156

University) and Dr Martin Fulda (Department for Plant Biochemistry, Albrecht-vou-

Haller-Institute for Plant Sciences, Georg-August-University).

All Arabidopsis plants were grown at 80 to 100 umol/mZ/s and 22°C under an 18-h-
light/6-h—dark photoperiod. Pea seeds (Pisum sativum L. cv Little marvel) were grown

for 8-10 days in the growth chamber at 25°C and an 8 h photoperiod with occasional

14 . . . 14 . . .
watering. [1— C]Acetic ac1d (57.2 Ci/mol), [l- C]ocatan01c ac1d (SSmCi/mmol), [1-
14 . . . 14 . . .
C]decanorc ac1d (SSmCi/mmol), [ C]dodecan01c ac1d(55mC1/mmol), [1-

14 . . . 14 . . .
C]tetradecanOic ac1d(55mC1/mmol), [l- C]hexadecan01c ac1d(55mC1/mmol) were

purchased from American Radiolabeled Chemicals Inc. (St. Louis, MO). Cerulenin (2,3-
epoxy-4-oxo-7,lO-dodecadienamide) was purchased from Sigma and was used as SOmM
stock in absolute ethanol. The herbicide haloxyfop was a gift from DowElanco

(Indianapolis, IN) and was used as lOOmM stock in 100% methanol.
Homozygous T -DNA Insertion Mutant Isolation

T-DNA (5.5-kb) insertions into the LACS8, AAE15 and AAE16 genes were identified by
PCR screening for both the presence of T-DNA insertion and the absence of intact gene.
The gene-specific primers used for the screening of insertions into the LACS8 gene were
5'-CTGATTCGCAACGCTTCATC-3' (forward) and 5'-CTATAGCTGATGGAAGTGG-
3' (reverse), and for AAE15 were 5'-ATCATCACTGGTCACACAGG-3' (forward) and

5'-AAGGCAGCAATTACTCTCGC-3' (reverse), and for AAE16 were 5'-

157

AACCTCTTATCTCCTCAGCC-3' (forward) and S'-CATCGACAGGAGTTATCAGC-
3' (reverse). The T-DNA— specific primer matching the left end of the T-DNA was 5'

TGGTTCACGTAGTGGGCCATCG 3'.
Arabidopsis detached leaf incubations

Twenty to thirty days old detached Arabidopsis leaves (0.2g fresh weight/assay) were
incubated in 25mM NaMES (pH5.7) buffer containing 0.01% w/v amount of tween-20 as
wetting agent under illumination (180 pmol photons/mz/s) at 25°C for designated length
of times. For cerulenin and haloxyfop inhibition assays, the leaves were pre-incubated in

above media containing various concentrations of inhibitors for 20 — 30 min. The

. . . . 14 . .
reactions were initiated by adding 0.1—1 uCi’s of [l- C] labeled fatty aCld substrates in

. . . l4
5-10uM concentration (1mM unlabeled acetate was added in case With [1- C]acetate

substrate) as indicated in the Figure legends. At the end of each incubations, buffer was
removed and leaves were washed with excessive amount of water and blotted briefly on
to the filter paper. The leaves were frozen with liquid nitrogen and stored in -80°C until

lipid extraction.
Chloroplast preparations and assays

Intact chloroplasts were isolated on a continuous Percoll gradient (Bruce et al., 1994).

All procedures were conducted at 0-4 °C. About 10-20 g of 8-day old pea seedlings were

158

homogenized in 100 mL of semi-frozen homogenization buffer (50 mM HEPES, pH 8.0,

330 mM sorbitol, 0.1% BSA, 1 mM MgC12, 1 mM MnClz, 2 mM EDTA) using a

polytron (PT 10/35) and filtered through 2 layers of miracloth (CALBIOCHEM). Crude
chloroplasts were collected by centrifugation at 1200 xg max for 5 min. These were
purified by centrifugation through pre-made gradients of Percoll generated by
centrifugation of 30 mL of 50% Percoll in homogenization buffer for 30 min at 40,000x g.
Crude chloroplasts (2-3 mL) were centrifuged through the pre-made gradients at 13,000
xg max for 5 min in an HB-4 rotor without brake. The intact chloroplast band formed
near the bottom of the gradient was recovered and washed with resuspension buffer
containing 50 mM HEPES, pH 8.0 and 330 mM sorbitol. The resulting isolated
chloroplasts can incorporate acetate into fatty acids at rates corresponding to in vivo rates
of fatty acid synthesis, import and process in vitro translated protein precursors.

Quantification of chlorophyll was performed as described by Amon (1949).

Chloroplast incubations to synthesize fatty acids were performed largely according to the
methods described by Roughan (1987). Chloroplasts (80 pg chlorophyll) were incubated
for designated amount of time under illumination (180 umol photons/mz/s) with shaking

at 25°C in a medium (200 -500ul) containing 0.33 M sorbitol, 25 mM HEPES-NaOH, pH

8.0, 10 mM KHCO3, 2 mM Na3EDTA, 1 mM MgC12, 1 mM MnClz, 0.5 mM K2HP04,

. . 14 . 1
and With either 0.2mM [1- C]acetate (5 11.0) or 10 11M [1— 4C]laurate (0.1 uCi). Where

noted in the text, assays also contained 1 mM ATP, designated concentration of cerulenin

and unlabeled sodium acetate. An aliquot was removed before the illumination and used

159

 

to measure the actual radioactivity added. Reactions were terminated by placing the
incubation mixture into liquid nitrogen and were kept in -80°C freezer until lipid

analyses.

Lipid analysis

Total lipids from labeled leaf samples were extracted according to the method of Hara
and Radin (1978). After quenching the tissue by heating in isopropanol at 80-90 °C for 5
min lipid extraction was carried out using hexane-isopropanol. An aliquot of the lipid
extract was suspended in 4:1 acetone/water and the OD at 652nm measured to determine
chlorophyll (Amon, 1949); another aliquot was assayed by liquid scintillation counting to
determine radioactivity incorporated. TAG, FFA, DAG were separated on regular K6
TLC using hexane / diethylether / acetic acid (v/v/v=80/20/l). Other glycerol lipids were
separated by developing on ammonium impregnated K6 TLC with acetone: toluene:
water (v/v/v=9l/30/8). For individual fatty acyl group analysis, lipid extract was first
transmethylated by heating at 90°C for l h in 0.1 mL of toluene and 1 mL of 10% (v/v)
boron trichloride:methanol (Sigma). After acidification with aqueous acetic acid, fatty
acid methyl esters were extracted two times with hexane. Fatty acid methyl esters
(FAME’s) were separated by argentation-TLC using plates impregnated with a 7.5%
silver nitrate in acetonitrile solution, dried and activated, and developed with toluene at -
20°C (Morris et al., 1967). Saturated FAME’s were recovered and separated second time
on C18-reversed phase TLC with acetonitrile/methanol/water 13027021 (v/v/v).

Hydrogenation of unsaturated fatty acids were done by stirring samples in hydrogen gas

160

saturated air for 3 h in the presence of platinum(IV) oxide (PtOz, Sigma) in methanol.

Radioactivity distribution on the TLC plate was located and quantified with an Instant

Imager (Packard Instrument Co., Meriden, CT).

Permanganate-periodate oxidation

In order to chemically cleave oleic acid at the double bond position permanganate-
periodate oxidation (also called von Rudloff oxidative cleavage, Christie, 2003) was
carried out. Oleic FAME band from aragentation-TLC was scraped into test tubes and

eluted with 6mL of hexanezethyl ether (2:1, v/v) and dried. The samples were

redissolved in lmL of 3.9mM KZCO3 in t-butanol/water (3/2) solution. After a thorough

vortex, 1 mL of second solution containing (3.3mM KMnO4, 102mM NaIO4) was added

and left for l h in room temperature with occasional vortex. Samples were acidified by
few drops of 5M aqueous HZSO4 and then sulphurous acid (H2803) solution was added
until yellow-green color persisted. The aqueous phase was extracted 2 times with 3mL

diethyl ether and water was removed through anhydrous NaZSO4. After carefully

evaporating to dryness the resulting nonanoic acid (C9A) and 1,(0-nonane-dioic
monomethyl ester fragments (C9AE) were separated by K6 silica TLC using hexane:
diethyl ether: acetic acid (90:10:1, v/v/v) system. Percent total radioactivity in each lane

was quantified using the Instant Imager.

Acyl-A CP intermediate analysis

161

Modified version of procedure described by Mancha et a1. (1975) was used to isolate
acyl—ACP. Frozen samples were rapidly quenched on a heating block (> 90°C) with
addition of hot isopropanol (equal volume with reaction) for 10 minutes. Lipids were
extracted 2 times with 3 mL hexane. Precipitation of acyl-ACP fraction was carried out
by addition of SOuL saturated ammonium sulfate solution and 8 mL of chloroform /
methanol (1/2, v/v) at room temperature for 20 minutes. Pellet was collected by
centrifugation (2,600 g for 5 minutes) and was washed again with 5 mL chloroform /

methanol (1/2, WV), 1% dimethoxypropane. The combined supernatant (13mL) was

partitioned against 7 mL of 0.2 M H3PO4 in 2M KCl and the chloroform layer was

recovered. After evaporation under N2 gas, acyl lipids were transmethylated with BF 3 in

methanol and FAME were extracted twice with 3mL hexane and was combined with
lipid extracts collected at the initial hexane extractions. Radioactivity of an aliquot was
counted by liquid scintillation counter (Beckman Coulter Inc.) to determine acetate

incorporation into the total lipids.

The pellet fraction was resuspended in 2mL of 0.5M NaOCH3 (in methanol) and

transmethylation was allowed to proceed at room temperature for 20 minutes. Two mL’s
of water and lipid carriers were added and FAME was extracted three times with 2.5mL
hexane. An aliquot was counted with liquid scintillation counter. FAME was separated
from unesterified FA’s by K6 TLC (hexane/diethyl ether/acetic acid=90/10/l). Percent

total radioactivity in each lane was quantified using the Instant Imager.

162

LITERATURE CITED

Alberts AW, Bell RM and Vagelos PR (1972) Acyl carrier protein. XV. Studies of -
ketoacyl-acyl carrier protein synthetase. J. Biol. Chem. 247 : 3190-3198.

Arnon DI ( 1949) Copper enzymes in isolated chloroplasts. Polyphenol oxidase in Beta
vulgaris. Plant Physiol. 24: 1-15.

Bao X, Pollard M and Ohlrogge JB (1998) The biosynthesis of erucic acid in
developing embryos of Brassica rapa. Plant Physiol. 118: 183-190.

Behal RH, Lin M, Back S and Oliver DJ (2002) Role of acetyl-coenzyme A synthetase
in leaves of Arabidopsis thaliana. Arch. Biochem. Biophys. 402: 259-267.

Bligh EG and Dyer WJ (1959) A rapid method of total lipid extraction and purification.
Can. J. Med. Sci. 37: 911-917.

Browse JA, McCourt PJ and Somerville CR (1985) A mutant of Arabidopsis lacking a
chloroplast-specific lipid. Science 227:763-765.

Bruce BD, Perry S, Froehlich J and Keegstra K (1994) In vitro import of proteins into
chloroplasts. In Plant Molecular Biology Manual (Gelvin, S. B., and Schilperoot, R. A.,
eds) 2nd Ed., Section J 1, pp. 1-15, Kluwer Academic Publishers, London.

Burton JD, Gronwald JW, Keith RA, Somers DA, Gengenbach BG and Wyse DL
(1991) Kinetics of inhibition of plant acetyl coenzyme A carboxylase by sethoxydim and
haloxyfop. Pestic. Biochem. Physiol. 39: 100-109.

Burton JD, Gronwald JW, Somers DA, Connelly JA, Gengenbach BG and Wyse DL
(1987) Inhibition of plant acetyl-coenzyme A carboxylase by the herbicides sethoxydim
and haloxyfop. Biochem. Biophys. Res. Commun. 148: 1039-1044.

Christie WW (2003) Isolation of fatty acids and identification by spectroscopic and
chemical degradative techniques. In Lipid Analysis (Christie, W. W., eds) 3rd Ed., pp.
255-297, The Oily Press, Bridgwater, England.

Claros MG and von Heijne G (1994) TopPred II: An improved software for membrane
protein structure predictions. CABIOS. 10: 685-686.

Cserzo M, Wallin E, Simon 1, von Heijne G and Elofsson A (1997) Prediction of

transmembrane alpha-helices in procariotic membrane proteins: the Dense Alignment
Surface method. Prot. Eng. 10: 673-676.

163

Emanuelsson O,Nielsen H,Brunak S, and von Heijne G (2000) Predicting subcellular
localization of proteins based on their N-terminal amino acid sequence. J. Mol. Biol. 300:
1005-1016.

Fice D, Shen Z and Byers DM (1993) Purification and characterization of fatty acyl-
acyl carrier protein synthetase from Vibrio harveyi. J. Bacteriol. 175:1865-1870.

Fulda M, Heinz E and Walter FP (1997) Brassica napus cDNAs encoding fatty acyl-
CoA synthetase. Plant Mol. Biol. 33:911-922.

Gangar A, Karande AA and Rajasekharan R (2001) Purification and characterization
of acyl-acyl carrier protein synthetase from oleaginous yeast and its role in triacylglycerol
biosynthesis. Biochem J. 360: 471-479.

Hara A and Radio NS (1978) Lipid extraction of tissues with a low toxicity solvent.
Anal. Biochem. 90: 420-426.

Hawke JC and Stumpf PK (1965) Fat metabolism in higher plants. 28. The biosynthesis
of saturated and unsaturated fatty acids by preparations from barley seedlings.
J. Biol. Chem. 240: 4746-4752.

Hitchcock C and Nichols NW (1971) Plant Lipid Biochemistry. Academic Press, New
York, pp 134-140.

Hofmann K and Stoffel W (1993) TMbase - A database of membrane spanning proteins
segments. Biol. Chem. Hoppe-Seyler 374: 166.

Jackowski S, Jackson PD and Rock CO (1994) Sequence and function of the aas gene
in Escherichia coli. J. Biol. Chem. 269: 2921-2928.

James AT (1962) The biosynthesis of unsaturated fatty acids in isolated plant leaves.
Biochim. Biophys. Acta. 57: 167-169.

Kannangara CG, Jacobson BS and Stumpf PK (1973) In vivo biosynthesis of -
linolenic acid in plants. Biochem. Biophys. Res. Commun. 52: 648-55.

Koo AJ, Ohlrogge JB and Pollard M (2004) On the export of fatty acids from the
chloroplast. in press J. Biol. Chem.

Krogh A, Larsson B, van Heijne G and Sonnhammer ELL (2001) Predicting

transmembrane protein topology with a hidden Markov model: Application to complete
genomes. J. Mol. Biol. 305: 567-580.

Kunst L, Browse J and Somerville C (1989) A mutant of Arabidopsis deficient in
desaturation of palmitic acid in leaf lipids. Plant Physiol. 90: 943-947.

164

Kunst L and Samuels AL (2003) Biosynthesis and secretion of plant cuticular wax.
Prog. Lipid Res. 42: 51-80.

Lessire R and Cassagne C (1979) Direct synthesis of stearoyl-acyl carrier protein from
stearate by microsomes from allium porrum epidermal cells. Plant Sci. Lett. 14: 43-48.

Mancha M, Stokes GB and Stumpf PK (1975) Fat metabolism in higher plants. The
determination of acyl-acyl carrier protein and acyl coenzyme A in a complex lipid
mixture. Anal. Biochem. 68: 600-608.

Mattoo AK, Callahan FE, Mehta RA and Ohlrogge JB (1989) Rapid in vivo acylation
of acyl carrier protein with exogenous fatty acids in Spirodela oligorrhiza. Plant Physiol.
89: 707-711.

Norman HA and St. John JB (1986) Metabolism of unsaturated
monogalactosyldiacylglycerol molecular species in Arbidopsis thaliana reveals different
sites and substrates for linolenic acid synthesis. Plant Physiol. 81: 731-736.

Notredame C, Higgins DG and Heringa J (2000) T-Coffee: a novel method for fast and
accurate multiple sequence alignment. J. Mol. Biol. 302: 205—217.

Ohlrogge JB and Browse J (1995) Lipid Biosynthesis. Plant Cell 7: 957-970.

Ohlrogge JB, Kuhn DN and Stumpf PK (1979) Subcellular localization of acyl carrier
protein in leaf protoplasts of Spinacia oleracea. Proc. Natl. Acad. Sci. U S A. 76: 1194-
1198.

Page R (1996) TREEVIEW: an application to display phylogenetic trees on personal
computers. Comp. Appl. Biosci. 12: 357-358.

Pollard M and Ohlrogge JB (1999) Testing models of fatty acid transfer and lipid
synthesis in spinach leaf using in vivo oxygen-18 labeling. Plant Physiol. 121: 1217-
1226.

Rock CO and Cronan JE Jr. (1981) Acyl-acyl carrier protein synthetase from
Escherichia coli. Methods Enzymol. 71: 163-168.

Roughan PG (1987) Long-chain fatty acid synthesis and utilization by isolated
chloroplasts. Methods Enzymol. 148: 327-337.

Roughan PG, Holland R and Slack CR (1978) Acetate is the preferred substrate for
long-chain fatty acid synthesis in isolated spinach chloroplasts. Biochem. J. 184: 565-
569.

Roughan PG, Thompson GA Jr. and Cho SH (1987) Metabolism of exogenous long-
chain fatty acids by spinach leaves. Arch. Biochem. Biophys. 259: 481-496.

165

Schnurr JA, Shockey JM, de Boer GJ and Browse JA (2002) Fatty acid export from
the chloroplast. Molecular characterization of a major plastidial acyl-coenzyme A
synthetase fi'om Arabidopsis. Plant Physiol. 129: 1700-1709.

Shockey JM, Fulda MS and Browse JA (2002) Arabidopsis contains nine long-chain
acyl-coenzyme a synthetase genes that participate in fatty acid and glycerolipid
metabolism. Plant Physiol. 129: 1710-1722.

Shockey JM, Fulda MS and Browse J (2003) Arabidopsis contains a large superfamily
of acyl-activating enzymes. Phylogenetic and biochemical analysis reveals a new class of
acyl-coenzyme a synthetases. Plant Physiol. 132: 1065-1076.

8011 J and Roughan G (1982) Acyl-acyl carrier protein pool sizes during steady-state
fatty acid synthesis by isolated spinach chloroplasts. FEBS Lett. 146: 189-192.

Terzaghi WB (1986) A system for manipulating the membrane fatty acid compositon of
soybean cell cultures by adding tween-fatty acid esters to their growth medium. Plant
Physiol. 82:771-779.

Thompson GA Jr., Roughan PG, BRowse JA, Slack CR, and Gardiner SE (1986)
Spinach leaves desaturate exogenous [14C] pahnitate to hexadecatrienoate: Evidence that
de novo glycerolipid synthesis in chloroplasts can utilize free fatty acids imported from
other cellular compartments. Plant Physiol. 82: 357-362.

Wada H, Shintani D and Ohlrogge J (1997) Why do mitochondria synthesize fatty
acids? Evidence for involvement in lipoic acid production. Proc. Natl. Acad. Sci. U S A.
94: 1591-1596.

Wirtz W, Stitt M and Heldt HW (1980) Enzmic determination of metabolites in the
subcellular compartments of spinach protoplasts. Plant Physiol. 66: 187-193.

166

CHAPTER 4
THE PREDICTED CANDIDATES OF ARABIDOPSIS PLASTID INNER
ENVELOPE MEMBRANE PROTEINS AND THEIR EXPRESSION PROFILES:
Search for lipid transporters of Arabidopsis plastid envelope
ABSTRACT
Plastid envelope proteins from the Arabidopsis thaliana nuclear genome were predicted
using computational methods. Selection criteria were: first, to find proteins with NHZ-
terminal plastid targeting peptides from all annotated open reading frames from
Arabidopsis thaliana, second, to search for proteins with membrane spanning domains
among the predicted plastidial targeted proteins, and third, to subtract known thylakoid
membrane proteins. 541 proteins were selected as potential candidates of the Arabidopsis
thaliana plastid inner envelope membrane proteins (AtPEM candidates). Only 34% (183)
of the AtPEM candidates could be assigned to putative functions based on sequence
similarity to proteins of known fiinction (compared to the 69% function assignment of the
total predicted proteins in the genome). Of the 183 candidates with assigned functions,
40% were classified in the category of ‘transport facilitation’ indicating that this
collection is highly enriched in membrane transporters. Information on the predicted
proteins, tissue-expression data from expressed sequence tags (EST’s) and microarrays,
and publicly available T-DNA insertion lines were collected. The data set complements
proteomic based efforts in the increased detection of integral membrane proteins, low
abundance proteins, or those not expressed in tissues selected for proteomic analysis.
Digital northern analysis of EST’s suggested that the transcript levels of most AtPEM

candidates were relatively constant among different tissues in contrast to stroma and the

167

thylakoid proteins. However, both digital northern and microarray analysis identified a

number of AtPEM candidates with tissue-specific expression patterns.

Fatty acid (FA) metabolism in a plant cell requires the coordination and the exchange of
FA and F A-derived lipids between the stroma and the extraplastidial organelles across the
double membrane of the plastid envelope. It is likely that specific membrane proteins are
involved in these processes. Of the more than 120 members of the ATP Binding Cassette
(ABC) transporter family reported in the Arabidopsis genome, 20 possess plastid
targeting sequences and transmembrane domains. These were selected as candidates for
plastid envelope lipid transporters. T-DNA insertion mutants for 11 selected candidates
were obtained and analyzed for FA compositions in leaves and seeds and for lipid
contents of seeds. There were no major changes in fatty acid composition in any of these
mutants. However, insertions in two genes (Atlg70610, At5g64940) consistently showed

reduced oil contents (about 60-70% of wild type) in seeds.

168

INTRODUCTION

The plastid envelope double-membrane separates lipid metabolism within the plastid
from extraplastidial lipid metabolism. There is a considerable flux of lipid exchange and
intermixing between the plastid and the endoplasmic reticulum (Ohlrogge and Browse,
1995) and this flux involves lipid transport across the plastid envelope membranes. In
this chapter bioinformatics and reverse genetics approaches are described that were taken
in efforts to search for plastid envelope lipid transporters. This search consisted of two
major parts. First, a genome wide analysis was made to create an inventory of
Arabidopsis plastid inner membrane proteins and their expression patterns. Second,
candidates for possible lipid transporters were selected and a preliminary characterization

of T-DNA insertion mutants of those candidates was conducted.

Plastids exist in a wide range of differential forms, including proplastids, chloroplasts,
etioplasts, amyloplasts, leucoplasts, and chromoplasts, depending on the developmental
stage and fiinction of the plant cells in which they reside. To a large extent, the types of
plastids that are carried by cells determine the metabolic function and products of the

particular plant tissue (Kirk and Tilney—Bassett, 1978).

One constant feature among the various types of plastids is the double membrane
envelope structure that surrounds the organelle. The envelope, especially the inner
envelope, effectively separates plastid metabolism from that of the cytosol. At the same

time, almost all carbon and a major flux of other metabolites, various polypeptides, and

169

signals must move through the envelope to coordinate and integrate metabolism with the
entire cell. Plastid envelopes contain protein transport machinery (Schnell, 1995; Cline
and Henry, 1996; Heins et al., 1998), and are a major site for membrane biogenesis.
Metabolite transporters from chloroplast and/or nongreen plastids accommodate the
requirements of the different photosynthetic or heterotrophic tissues (Kammerer et al.,
1998; Neuhaus and Emes, 2000). Membrane constituents (glycerolipids, pigments,
prenquuinones) are synthesized on the envelope membrane as well as the porphyrin ring
and phytyl chain used in chlorophyll synthesis, and the enzymes required for chlorophyll
breakdown in senescing plastids (for review see Joyard et al., 1998). It is also
hypothesized that fatty acids that are synthesized within the stroma are exported through
a channeled system on the envelope (Pollard and Ohlrogge, 1999; Koo et al., 2004). The
acyl group modification of lipids such as desaturation takes place on the inner envelope
and lipid-derived signals are produced on the envelope membranes (Miquel and Browse,

1992)

Despite the importance and complexity of plastid envelope, only a small fraction of
envelope proteins have been purified or characterized (Joyard et al., 1998). Recently,
several groups have initiated proteomic studies to identify the constituents of plant
subcellular organelles and membranes (Santoni et al.,1998; Sazuka et al., 1999;
Seigneurin-Bemy et al., 1999; Ferro et al., 2000; Peltier et al., 2000; Schubert et al.,
2002). Most of these studies have been based on 1-dimensional or 2-dimensional
electrophoretic methods followed by mass spectrometric analysis of peptides derived

from fractionated proteins. The continued development and improvement of these

170

methods resulted in the identification of hundreds of proteins, particularly from
organisms with sequenced genomes. Nevertheless, there are some limitations in current
proteomic technologies. In particular, many membrane proteins are found in very low
abundance and/or expressed only in certain tissues, making comprehensive samples
difficult to obtain. In addition, many integral membrane proteins remain difficult to
solubilize and often do not resolve well by gel electrophoresis or isoelectric focusing
(Adessi et al., 1997; Santoni et al., 2000). Finally, trypsin, most frequently used to digest
proteins prior to mass spectrometric analysis has relatively few sites in the more
hydrophobic integral membrane proteins. Together, these limitations make it likely that a
significant fraction of the integral proteins present in membranes will remain difficult to

detect by most proteomic technologies.

One of the first steps in understanding the function of a gene is to determine the
subcellular localization of the gene product (Somerville and Dangl, 2000). To augment
the proteomic approaches described above, an additional opportunity for the discovery of
putative localization of proteins is now possible due to complete sequencing of the
Arabidopsis thaliana genome (The Arabidopsis Genome Initiative, 2000). To determine
subcellular localization of plastidial proteins one can make use of certain features of
plastid-targeted proteins. The plastid genome itself encodes only about 120 single copy
genes and therefore must rely on proteins from the nucleus. The protein constituents that
are encoded in the nuclear genome are synthesized in the cytoplasm as higher molecular
weight precursors containing an amino-terminal transit peptide which is cleaved

following entry into the plastid (Cline and Henry, 1996). These pre-sequences of nuclear-

171

encoded plastid proteins, even though not strictly conserved, share some common
features that can be used to predict localization using computer algorithms (Emanuelsson
et al., 1999). This in combination with transmembrane a-helical domain characteristics,
namely charge bias and hydrophobicity (Sonnhammer et al., 1998; Krogh et al., 2001),
make it possible to identify candidates for plastid membrane proteins. Considering the
nature of predictors and the selection criteria used (discussed in the Result and
Discussion) the collection described in this paper is expected to contain mostly inner

envelope integral membrane proteins.

Evidence is emerging that some lipid transporters belong to the ABC transporter class of
membrane proteins. For example, the peroxisomal long-chain fatty acid transporters in
yeast (Patlp/ Path, Hettema and Tabak, 2000), animal (ALDp, Mosser et al., 1993,
1994) and plant (PXAl, PED3, CTS) (Zolman et al., 2001; Footitt et al., 2002; Hayashi et
al., 2002) are ABC transporters. The phosphatidylcholine translocators (mouse Mdr2,
human MDR3), the N-retinlidine PE flippase (ABCR) and the cholesterol transporter
(ABCl) in mammals are P-glycoproteins belong to the ABC transporter family (Ruetz
and Gros, 1994; Allikmets et al., 1997; Borst et al., 2000). The E.coli ‘lipid A’
transporter in the inner membrane (msbA) also belongs to the multidrug resistant protein
subfamily of ABC transporter superfamily (Zhou et al., 1998). The Arabidopsis TGDl,
involved in the lipid transfer from ER to plastid is similar to an ABC domain-lacking
permease half-molecules of multipartite bacterial ABC transporter complexes (Xu et al.,
2003). The Arabidopsis ABC transporter family consists of more than 120 members

(Sanchez-Femandez et al., 2001). In this study I present preliminary characterization of

172

T-DNA knockouts of several ABC transporters that are predicted as plastid inner

membrane proteins.

RESULTS AND DISCUSSION

Part I. The Predicted Candidates of Arabidopsis thaliana Plastid Inner Envelope

Membrane Proteins and Their Expression Profiles

Plastid Envelope Membrane Protein Candidates

The strategy to select candidates of plastid envelope integral membrane proteins from
Arabidopsis nuclear genome sequences by computational methods and the results are

summarized in Figure 20.

The first step was to predict plastid targeted proteins from the entire Arabidopsis genomic
sequences (See Methods and Materials for Arabidopsis genomic sequence retrieval). This
was done using TargetP (http://www.cbs.dtu.dk/services/TargetP/). TargetP is considered
the best subcellular location predictor that is publicly available (Emanuelsson and von
Heijne, 2001; Bannai et al., 2002; Peltier et al., 2002) and was also chosen by the
Arabidopsis Genome Initiative to analyze the recently finished genome sequence of
Arabidopsis thaliana (The Arabidopsis Genome Initiative, 2000). Protein targeting to the
plastid usually relies on its amino terminal pre—sequence (Cline and Henry, 1996) and

TargetP, a neural network-based tool, is trained to recognize those signals. Overall

173

success rate on test sets analyzed by the developers was 85% (Emanuelsson et al., 2000).
The sensitivity for chloroplast-targeted proteins was 85% while the specificity was lower
(69% and 84% on two different test sets). This means that more false positives are
expected than the false negatives. In order to increase the specificity, cut off restraints can
be applied. However, in order to include a maximum number of possible candidates we
instead chose default decision rule (Winner takes all) and designated in our database the
‘reliability classes’ from 1 to 5 (Emanuelsson et al., 2000). In total 3,665 out of 25,552
(14.3%) proteins encoded by the Arabidopsis nuclear genome were predicted to be
plastidial which is similar to the reported value of 3,574 from the Arabidopsis Genome

Initiative (The Arabidopsis Genome Initiative, 2000).

These 3,665 predicted plastid targeted proteins were next analyzed for transmembrane 0t-
helical domains (Figure 20). There are several web-based predictors available (Moller et
al., 2001). A recent study compared the performance of 14 different methods, against 883
membrane spanning regions of biochemically characterized proteins, and concluded that
TMHMM is the most accurate method for transmembrane segment detection (Moller et
al., 2001). Subsequently, HMMTOP also has been shown to have similar performances
(Moller et al., 2002). Most recent methods for prediction of transmembrane helices rely
on hydrophobicity and charge bias of the transmembrane regions. There were variations
in the accuracy of overall correct topology prediction (number and location and

orientation of transmembrane regions), however correct prediction of the presence of

174

Figure 20. Summary scheme of plastid envelope protein prediction from Arabidopsis

thaliana nuclear genome.

The graph displays the number of proteins following each selection steps. Subcellular
localization was predicted using TargetP v1.01 web server. Transmembrane region
prediction was done by TMHMM v2.0 web server. cTP: chloroplast transit peptide. RC:
TargetP reliability class. TM: transmembrane domain. See Materials and Methods for

web site addresses for the predictors.

 

       
     
     

 

 

 
  
    
   

 
  
   
 
    

 
  
   
   

        

 

  

  

25,552
. . 3 25000 1 : :_ ‘Ii
ArabldOpSlS 5 20000_
gemme 0’ .
25,552 genes “5 15000 "
a 10000- 3.555
'2 (14% nuclear) 562
g 5000 (15% plastid) 541
Genes with cTP 0 ' . . '
Target P v1.01 nuclear plastid plastid plastid
"° cum genome targeted membrane envelope
- lastid
lastid P'aSt'd p
taprgeted: b membrane , WWII?”
3.665 genes Genes with TM “new" 552 '"tegrtai 5“
domains TMHMM v2 0 Pmteins '°m°"° pro ems
.. ' known
cutoff probability less than .
0.5 remove genes with TM thylakOId TaretP RC
proteins RC 1. 88 genes

domain within the predicted
cTP region tag genes with
TM domains near the cTP
cleavage site

RC2: 107 genes
RC3: 121 genes
RC4: 101 genes
R05: 124 genes

175

transmembrane segments per se was overall 95% percent in a comparison of 3 best
predictors (Sonnhammer et al., 1998). TMHMM is reported to discriminate between
soluble and membrane proteins with both specificity and sensitivity better than 99%
(Krogh et al., 2001). TMHMM, like Target P is also a neural network based program but
built with an architecture that corresponds more closely to the biological system (Hidden
Markov Model). The probability of a predicted integral protein to be a true membrane
protein is positively correlated with the expected number of amino acid residues within
the transmembrane segments and also with the expected number of transmembrane
segments in the protein. In other words the more amino acid residues per predicted
transmembrane segments and the more predicted number of transmembrane segments per
gene protein, the more likely that it is a truly integral protein. In order to include as many
integral proteins as possible we selected all proteins that have at least one predicted
transmembrane domain regardless of amino acid residue numbers. One of the advantages
of TMHMM, is that it provides plots of probabilities for prediction throughout the
analyzed proteins. In our study, proteins that were overall not predicted to be
transmembrane proteins yet contained weak or-helical domains with probabilities over
50% were also included in the database as potential membrane proteins with their
probabilities. Because the accuracy of TMHMM prediction drops when signal peptides
are present (Krogh et al., 2001), transmembrane predictions within the predicted targeting
peptides (according to the TargetP cleavage site predictions) were removed and those
domains which were predicted close to the targeting peptides were designated in the

database as ‘N-term’. After processing as described above, out of 3,665 predicted

176

plastidial proteins 562 (about 15%) contained potential transmembrane (ii-helical domains

(Figure 20) and thus are considered as plastid membrane protein candidates.

The third step was to eliminate the thylakoid integral membrane proteins. Although there
are suborganelle location predictors such as PSORT the overall accuracy of such
programs is not high enough (about 50%) to use for discrimination between the thylakoid
and the envelope localized proteins (Nakai et al., 1999). It is estimated that there are at
least 200 proteins in the lurnenal space and in the periphery of the thylakoid membrane
(Peltier et al., 2000). Four multisubunit protein complexes with 75 to 100 peptides
involved in photosynthesis, comprise the majority of the thylakoid membrane proteome
(Peltier et al., 2000). These proteins are relatively well studied and many of them are
known to be encoded by the plastid genome which contains about 90 protein encoding
genes (Sugita and Sugiura, 1996). If the known nuclear encoded thylakoid membrane
proteins are removed from our candidate list of 562, the remaining candidates are
expected to be highly enriched in envelope proteins. After manual removal of 21 known
thylakoid associated proteins and their homologs, 541 proteins remained as potential

candidates for Arabidopsis thaliana Plastid Envelope Membrane proteins (AtPEM).

It is important to note that these are predicted to be integral membrane protein candidates
and the candidate list will contain omissions and inclusions of various types. First, as
mentioned above, the specificity of Target P prediction is lower than the sensitivity.
Because we did not employ any specificity cut off for Target P prediction and used low

stringency for transmembrane prediction (transmembrane domain probability > 50%),

177

these 541 candidates may contain many false positives. If instead, the Target P
specificity is set to > 0.95 and proteins with low probabilities of transmembrane domain
are excluded, about 250 candidates remain and these represent a higher reliability core
set. These higher reliability candidates are color coded on our website. Secondly, there
will be many membrane proteins that do not have transmembrane domains but are
associated with the envelope through different types of interactions such as B-sheet
conformation, polyisoprenylation, acylation, glycolipid anchors, protein-protein
interaction mediated associations, etc. Thirdly, most outer membrane localized proteins
appear to insert directly into the outer envelope membrane without going through the
general import apparatus (Cline and Henry, 1996; Fulgosi and 8011, 2001; Schleiff and
Klosgen, 2001). Therefore, the 541 AtPEM candidates will under represent the outer
membrane proteins. However, the protein composition of outer membranes is much
simpler than that of inner membrane. The lipid to protein ratio of outer membrane is
about 3 times higher than that of inner membrane and the inner envelope membrane,
which contains various metabolite translocators, is the main permeability barrier for
solutes although there is growing evidence suggesting that the outer envelope may
contain regulated ion channels (Cline et al., 1981; Block et al., 1983; Douce and Joyard,
1990; Pohlmeyer et al., 1998; Flugge, 2000; Bolter and 8011, 2001). Thus, we consider it

likely that AtPEM candidates represent the majority of integral plastid envelope proteins.

What are they?

178

Figure 21A indicates the functional identification status for the selected 541 AtPEM
candidates based on the short descriptions in the annotation databases (MIPS and
GenBank). The proteins with identified function (identified) or that have homology with
known proteins (X-like, putative-X) comprised only about 29% of the 541 whereas 71%
of the candidate proteins could not be assigned to any known function (putative,
unknown, hypothetical). This is a very low characterized status compared to the overall
69% functional assignment of the whole genome (The Arabidopsis Genome Initiative,
2000) and reflects the past difficulties in studying membrane proteins (Wilkins et al.,

1998; Seigrreurin-Bemy et al., 1999).

More detailed functional classification of the AtPEM candidates was done according to
the automatically derived functional categories from MIPS Arabidopsis thaliana Data
Base using Protein Extraction Description and Analysis Tool (PEDANT) web server
(http://pedant.gsf.de/, Frishman etal., 2001). Among the 541 candidates, 183 (34%) were
found in the PEDANT database and were classified into 17 classes and 89 subclasses.
Figure 21B shows the number of AtPEM candidates found in each category. The largest
number of AtPEM candidates (39% of 183 classified predicted proteins) fell into the class
of ‘transport facilitation’. Thus membrane transporters are highly enriched in the selected
AtPEM candidates. Figure 21C shows the sub-classification of ‘transport facilitation’ and
‘metabolism’ classes. The 3 major subclasses for ‘transport facilitation’ were ‘ion
transporters’, ‘C-compound and carbohydrate transporters’ and ‘ABC (ATP binding

cassette) transporters’.

179

There were also several candidates involved in protein translocation into the plastids.
None of the outer membrane components of the protein translocon were found among
AtPEM candidates as was expected (many outer membrane proteins lack obvious plastid
targeting sequences as discussed above) whereas 5 Arabidopsis homologs to the known
pea translocon at the inner membrane of chloroplast (Tic)110, Tic20, Tic40, TicSS
(J ackson-Constan and Keegstra, 2001) were present among the candidates (Table 7). Two
Arabidopsis homologs of Tic22 were predicted to be targeted to the plastids by the
TargetP but did not have obvious transmembrane spanning domain (by TMHMM
prediction). Tic22 is thought to be peripherally associated with the inner envelope

membrane facing the intennembrane space (J ackson-Constan and Keegstra, 2001).

In agreement with the known role of plastid envelopes in lipid metabolism, ‘lipid, fatty-
acid and isoprenoid metabolism’ related proteins represented the largest sub-set among
the ‘metabolism’ class. Table 7 presents the result of a more detailed survey of proteins
predicted or known to be involved in plant lipid metabolism (search was carried out using
our database of predicted lipid related genes in Arabidopsis thaliana (Mekhedov et al.,
2001; Beisson et al., 2003)). Seven proteins potentially responsible for known enzyrnetic
reactions to produce glycerol lipids in the plastid envelope were found among AtPEM
candidates (Table 7, for review see Ohlrogge and Browse, 1995). Arabidopsis

homologs of plastidial 2-lysophosphatidic acid acyltransferase (EC 2.3.1.51) which
mediates acyl group transfer from acyl-ACP (acyl carrier protein) to sn-2 position of

lyso-phosphatidic acid (l-acyl-sn-glycerol 3-phosphate) to produce phosphatidic acid

180

Figure 21. Functional classification of the Arabidopsis plastid envelope protein

(AtPEM) candidates.

(A) Overall functional annotation status based on the short descriptions of each
candidates from MIPS (chromosome 1,3,5) and from GenBank (chromosome 2,4). X:
any protein/gene/enzyme name. (B) Detailed functional classification using automatically
derived functional categories from PEDANT web server (see Materials and Methods).
(C) Subclassification of the proteins classified as ‘transport facilitation’ and ‘metabolism’

by PEDANT.

. unknown
hypothetical 20%

26%

   
  
 

identified
7%

putative-X
1 1 %
putative

25% X-like

11%

181

 

 

 

 

 

 

 

 

 

 

L

0 O O
6 4

:53 2.39:.

041mm

Um<mr0.uz.mz...

4x>zmom=ujoz

.umo...m_z m<z._._..mm.m
omrrcgm Omo>z.N>flOz

.umOAmE 0mm._._z>..._Oz

._._»>zm.u0m>m_.m
mrmZmZAm. <.m>.. >20

Ow0>z_m.<_ mpmo..u_o
pmoamim

.Oz_0 IOZmOmdymE

omrrcfyw 435200.»...
mzmmo<

25:60:25

om....c.u>.~ 002:2.o>._..Oz
w_oz>r Hm>zm0cojoz

Omrrcfpm w.00mzmm_m

4m>zmp0md .u>o._._...>..._Oz

OmE. wmmocm.om1mzmm.
om_... Um>..._.. >20 >0m.zo

om.._. 0m0<<...... om.... D.<_m_oz

>20 02> 324135

classification

182

C.

 

 

Class Subclass no. of genes

TRANSPORT FA CILI TA T ION
ion transporters 24
C-compound and carbohydrate transporters 22
ABC transporters 16
other transporter facilitators 15
Channels/pores 9
amino-acid transporters 8
drug transporters 6
purine and pyrimidine transporters 6
Transport ATPases 2
peptide-transporters l

METABOLISM
lipid, fatty-acid and isoprenoid metabolism 8
nucleotide metabolism 6
C-compound and carbohydrate metabolism 5

metabolism of vitamins, cofactors, and

prosthetic groups 5
nitrogen and sulfur metabolism 5
phosphate metabolism 4
amino acid metabolism 3
secondary metabolism 3

 

183

Table 7. AtPEM candidates involved in protein import and glycerolipid metabolism.

 

 

Protein / Enzyme Name locus ID Description
plastid envelope protein import
a
machinery
Tic 110 Atl g06950 Chloroplast inner envelope protein, putative
Tic 55 At2g24820 putative Rieske iron-sulfur protein
Tie 20 Atl g04940 unknown protein
A t 4g033 20 putative chloroplast protein import
component
Tic 40 At5g16620 translocon Tic40-like protein
glycerolipid biosynthetic enzymes
Plastidial cytidine-5'-diphosphate . .
diacylglycerol synthase At4g26770 putative CDP-diacylglycerol synthetase
A t3g 60 620 phosphatidate cytidylyltransferase - like
protein
Plastidial A t2 39290 putative CDP-diacylglycerol--glycerol-3-
hosphatidylglycerophosphate synthase g phosphate 3-phosphatidyltransferase
Monogalactosyldiacylglycerol monogalactosyldiacylglycerol synthase-like
At4g3 1780 .
synthase protein
Plastidial oleate desaturase (FAD6) At 4g3 09 5 0 pgggpplast omega-6 fatty ac1d desaturase
Plastidial linoleate desaturase A t 5 05580 temperature-sensitive omega-3 fatty acid
(FAD7/FAD8) g desaturase, chloroplast precursor
At3gl 1 170 omega-3 fatty ac1d desaturase, chloroplast
precursor
c
ABC transporters
Full Molecule ABC Transporters Atl g15210 putative ABC transporter
Atlg15520 ABC transporter, putative
Atl g59870 ABC transporter, putative
At1g66950 ABC transporter, putative
At3g16340 putative ABC transporter
At3g21250 unknown protein
Half-Molecule ABC Transporters Atl g70610 putative ABC transporter
Atl g54350 ABC transporter family protein
At2g39l90 putative ABC transporter
At2g28070 putative ABC transporter
At3 g25 620 membrane transporter, putative
At3g52310 ABC transporter, putative

 

184

Table 7. continued

 

Protein / Enzyme Name locus ID Description

 

At5 g03910 ABC transporter -like protein
At5g19410 membrane transporter - like protein
At5g52860 ABC transporter-like protein

At5 g64940 ABC transporter-like

 

a Arabidopsis homologs of the pea plastid protein import machinery were according to
the published data by J ackson-Constan and Keegstra (2001).
Arabidopsis glycerolipid biosynthetic enzyme candidates were searched using our

database of lipid metabolism related genes in Arabidopsis thaliana (Mekhedov et al.,
2001; Beisson et al., 2003).

C Search for the ABC transporters was done using the complete inventory of ABC

protein superfamily in Arabidopsis from the website at http://www.arabidopsisabc.net/,

Sanchez-Femandez et al., 2001).

(1,2-diacyl-sn-glycerol 3-phosphate) was missing. All three Arabidopsis homologs of this
enzyme were predicted to a different subcellular location (‘mitochondria’ and ‘others’)
by TargetP. Digalactosyldiacylglycerol synthase (EC 2.4.1.184) which was known to
localize on the outer envelope (Froehlich et al., 2001) was not among the AtPEM
candidates. Although it had plastid transit peptide recognized by TargetP, no membrane

spanning region was detected by TMHMM.

Surprisingly there were several candidates that are similar to proteins involved in cell
wall biogenesis and modifications, such as arabinogalactan-protein homolog, cellulose

synthase catalytic subunit -like protein, pectinesterase - like protein, pectin

185

methylesterase-like protein, putative xyloglucan fucosyltransferase, etc. These
unexpected relationships between the plastid envelope and predicted proteins annotated
with functions that are generally considered non-plastidial may reflect inaccurate
prediction by TargetP software. Alternatively, because these putative functions are
assigned based solely upon sequence similarities, an equally likely possibility is that the
annotations point toward related, but previously un-described functions in the plastid.
This incongruence therefore represents an example where bioinformatics analysis can

direct attention toward potential novel functions in the plastid envelope.

Characteristic Features of Plastid Envelope Candidates

The distribution of the peptide length of AtPEM candidates is shown in Figure 22A. 80%
of the candidates were smaller than 600 amino acids in length. The average length of the
peptides was 411 amino acids which is slightly smaller than the average peptide length of
434 amino acids for the nuclear genome (The Arabidopsis Genome Initiative, 2000). As
indicated in Figure 22B, 63% of the AtPEM candidates contained 2 or less membrane
spanning domains whereas there were 40 candidates that had 10 or more membrane
spanning domains. The membrane spanning domains of proteins with only a few such
domains might serve as an anchor to the envelope or to tether the peripheral proteins.
Interestingly 70% of candidates with 10 or more membrane spanning domains were
classified as ‘transport facilitation’. Due to their high hydrophobicity these potential
transporters would be particularly difficult to display on 2 dimensional electrophoretic

gels of purified plastid envelopes.

186

AtPEM candidates provide an opportunity to compare predicted plastid envelope
proteome with the proteome of other organisms, i.e., Synechocystis. The Synechocystis
genome contains about 3,167 protein encoding genes which is roughly the number
predicted to be plastidial in Arabidopsis. Highly similar proteins for about 32% (175) of
AtPEM candidates were in the Synechocystis genome (data not shown) compared to 44%
and 47% of thylakoid and stromal proteins, respectively. Thus, the boundary envelop of
the plastid may have a more “mixed” evolutionary origin than the internal components.

We also compared our list of candidates to proteins identified by mass spectrometric
analysis of Arabidopsis chloroplast envelope preparations (Froehlich et al., 2003; F erro et
al., 2002, 2003). In Table 8 are listed 37 AtPEM candidates that are represented in the
proteins identified through proteomics approach by both groups. The proteomics
approach identified many non-integral membrane proteins and outer membrane proteins
not selected by our approach. Thus, the bioinforrnatic identifications reported in this
study clearly provide novel information on a set of inner envelope integral membrane

proteins that are not easily characterized by proteomics.

Digital mRNA Expression Profile

In addition to the subcellular localization, indirect information on cellular or

developmental function can be obtained from spatial and temporal expression patterns of

genes. Gene chips, microarrays, Expressed Sequence Tags (ESTs), and Serial Analysis of

187

Figure 22. Characteristics of AtPEM candidates.

(A) Peptide length distribution. L: number of amino acid residues. Plastid targeting
sequences are removed based on TargetP cleavage site predictions. (B) Distribution of
the number of membrane spanning domains. TMHMM v2.0 web server was used for the
membrane spanning domain prediction. prob<1, proteins with spanning domain

probability less than 1.

 

 

 

 

 

 

 

 

 

 

 

 

L>1500E
A E
g 900
'6 800 E3 I
(U 71'.
g 700 ”a 1
5 500 . “ ..
2 400 ..................................................................
:3 300 ' _ - - - -1: ---------------- :13-
3- 200- t? “ -4 , ~~
a.

 

 

 

 

0 20 40 60 80 100 120

number of proteins

188

 

 

w.
a...
F
. ,.
.

_
.2.
I.

 

 

 

   

  

.
....fl..

11.2 ms. .4 5. 6 7...... g9 w. m,
>
5082. 0:05508055 .o .353:

 

 

 

 

pr0b<1

200

150

100

50

number of proteins
189

Gene Expression (SAGE) all provide useful means to study mRNA expression profiles

(Bouchez and Hofte, 1998).

A large pr0portion (68%) of AtPEM candidates have at least 1 EST in the GenBank
dbEST (113,330 ESTs as of January 4, 2002, Table 9) and 87% (470/541) were
represented by The Institution of Genomic Research (TIGR) Tentative Consensus (TC)
sequences (163,752 TCs as of May 23, 2001, http://www.tigr.org/tdb/agi/ ,Quackenbush
et al., 2000, see Materials and Methods. Thus the majority of the AtPEM candidates are

transcriptionally active.

The abundance of ESTs sequenced from different cDNA libraries can provide an estimate
of relative transcript abundance provided a number of conditions are met (Audio and
Claverie, 1997). To obtain tissue-specific expression profiles of AtPEM candidates,
110,000 ESTs deposited in GenBank and sequenced from 55 different cDNA libraries
were grouped into 8 ‘library pools’ according to the source tissues from which the cDNA
libraries were derived (Table 9, Beisson et al., 2003). We then surveyed the abundance of
ESTs within each library pool for AtPEM candidates after normalization by dividing the

number of AtPEM candidate ESTs in the given ‘library pools’ by the total number of
ESTs of that ‘library pool’. The relative abundances of ESTs for AtPEM candidates in
each tissue were similar, ranging from 1.4 to 2.5 % of the total ESTs, except for the
‘flowers’ pool which showed the highest abundance at 4.2% of ESTs. Although these
aggregate EST frequencies varied little, a number of individual proteins were found to

have more distinct tissue-specific expression patterns. In order to study this in detail,

190

Table 8. The AtPEM candidates represented in the chloroplast envelope proteins
identified through protein mass spectrometry by Ferro et al. (2002, 2003) and by

Froehlich et al. (2003).

 

 

locus 11) 21:21:50 a TMb probabilityc ESTs Annotation

Atlg06950 189 2 l 5 Ticl 10 homologue

At 1 g20830 349 l 1 l hypothetical protein

Atl g3 2080 512 12 l 2 unknown protein

At 1 g42960 1 68 l l 8 hypothetical protein

Atl g6 70 80 220 4 l 2 hypothetical protein

At 1 g78620 3 3 3 3 l 9 hypothetical protein

A t2g2 4820 539 2 0. 8 ACOO6585_25 putative Rieske
iron-sulfur protein

At2g34460 280 0.5 6 unlmown protein

A t2g3 85 5 0 3 27 3 1 3 putative non-green plastid inner
envelope membrane protein
putative CDP-diacylglycerol--

At2g3 9290 296 1 1 3 glycerol-3-phosphate 3-
phosphatidyltransferase

At2g47840 208 4 l 1 hypothetical protein

A t3gl 1 170 446 3 1 14 omega-3 fatty ac1d desaturase,
chloroplast precursor

At3 g203 20 3 81 1 l 5 unknown protein

At3g47060 802 0.8 3 FtsH metalloprotease - like protein

At3g51870 381 2 l 4 putative carrier protein

At3 g5 72 80 239 3 1 hypothetical protein

At3 g605 90 329 5 1 4 putative protein

At3g61870 272 4 1 7 putative protein

A t3g 63 4 1 0 3 3 8 1 1 44 putative chloroplast inner

191

envelope protein

Table 8. continued.

 

 

locus 11) 21:21:10 a TMb probabilityc ESTs Annotation

At4g25650 548 l 3 l 8 Putative protein

A t 4g3 09 5 O 448 2 4 1 1 1 chloroplast omega-6 fatty ac1d
desaturase

A t 4g3 1 7 8 0 533 2 1 1 4 monogalactosyldiacyl glycerol
synthase

At4g39460 330 4 3 l l 1 mitochondrial carrier-like protein

At5g01500 415 3 0.7 4 putative protein
2-oxoglutarate/malate

At5g12860 557 l 13 l 21 translocator precursor —like
protein

At5g13720 262 2 3 l 3 unknown protein

At5g16150 560 5 10 l 2 sugar transporter-like protein

At5g16620 447 2 0.5 10 translocon Tic40-like protein

At5 g1 7520 347 l 7 l 2 root cap 1 (RCPl)

At5g19750 288 2 2 1 2 putative protein

At5g22790 433 1 l l l putative protein

At5 g23 890 946 3 l l 5 unknown protein

A t 5g333 2 0 408 1 6 1 38 phosphate/phosphoenolpyruvate
translocator precursor

At5 g5 2540 461 1 10 1 8 putative protein

At5g59250 579 1 9 l 3 D'xy'.°°°'H+ Sympmer ' "ke
protein

At5g62720 243 3 4 1 9 putative protein

A t5g 64290 5 63 2 9 1 7 2-oxoglutarate/malate

translocator

 

‘Reliability Class’ of the subcellular localization prediction provided by the

TargetP web server.

number of spanning region predicted by TMHMM web server.

c probability of membrane spanning domain provided by TMHMM web server.

192

Table 9. Summary of ESTs of AtPEM candidates by tissue types.

 

Tissue Specific Library Pools a

 

Cell
Flowers Leaves Mixed Roots Seedlings Seeds Siliques SUM
suspension
Library size 986 7566 3774 42855 18350 6058 10678 12842 103109
b
no. OfESTS 14 321 91 1061 319 153 158 232 2349
percent total
c 1.4 4.2 2.4 2.5 1.7 2.5 1.5 1.8 2.3
ESTs
no. of
. _ 152 89
candidates With 11(2) 99(18) 78(14) 249 (46) 75 (14) 127 (24)365 (68)
d (28) (16)
ESTs

 

a EST libraries were grouped into 8 ‘library pools’ according to the source tissues where

the cDNA library was derived from (Beisson et al., 2003).

b . . . .
Number of ESTs corresponding to the AtPEM candidates found in each library pool.
C Percent total EST 3 in the library.

d . . . .
Number of AtPEM candidates that has at least one EST in given library pool. Percent

total number of AtPEM candidates are within the parentheses.

193

tissue-specific expression patterns of individual candidates were analyzed using statistical
equations developed by Audio and Claverie (http://igs-server.cnrs-mrs.fr) which
differentiate between random EST sampling fluctuations versus significant change inEST
frequencies. The number of ESTs corresponding to a given gene (AtPEM candidate)
found in each of three tissue-specific library pools (‘flowers’, ‘roots’, or ‘seeds’) was
compared with that in the reference ‘mixed’ pool and the results are presented in Table
10. A total of 21 AtPEM candidates (6% of 365 candidates that had at least 1 EST)
displayed tissue-specific expressions when p<0.005 (p, a probability of the compared
EST abundances being different by chance) was applied. Figure 23 visualizes the digital
expression profile of 21 AtPEM candidates that showed tissue-specific expression pattern
at p<0.005. Among the differentially expressed genes were glucose-6-
phosphate/phosphate translocator (GPT) and phosphoenol pyruvate/phosphate
translocator (PPT). GPT was previously shown to be highly expressed in developing
maize kernel and potato tuber (Kammerer et al., 1998) and the ESTs of GPT were
abundant in Arabidopsis developing seed EST database (White et al., 2000). In
agreement with these observations, GPT had higher EST abundance in ‘seeds’ library
pool than the reference at p<0.004. Similarly, for PPT, biochemical analysis and mRNA
blotting results indicated a high expression of PPT in non-green tissue (Kammerer et al.,
1998) especially roots (Fischer et al., 1997) and in our EST analysis, PPT expression was
highest in the ‘roots’ (p<0.003). Thus, the comparisons in Table 10 agree with previous
biochemical characterizations and support the validity of the digital expression profile
approach. Many of the proteins shown in Table 10 are of unknown function, including

several with high representation in specific tissues.

194

Table 10. Summary of AtPEM candidates with tissue-specific transcript

abundances.

Number of ESTs in the ‘flowers’, ‘roots’, ’seeds’ pools was compared with that in the
‘mixed’ pool. Candidates with probabilities of EST frequency differences being by

chance less than 0.005 are grouped according to the tissue specificity of the transcript

abundances (shown in the left shaded column). a F: Flowers, R: Roots, S: Seeds, M:

Mixed. Probability calculations were according to Audic and Claverie, 1997 (see

Materials and Methods).

195

 

 

 

 

    
 
  
  
  
 

 

   
  
  
  
  
 
  
 
 

 

 
  
   
  
 
 
  

 

Number of ESTs Probabilities a
Locus ID Descriptions
F R S M F/M R/M SIM
Atl g083 80 unknown protein 19 l 0 10 0.2 0.2
: At1g20340 hypothetical protein 21 5 O 41 0.004 0.001
; Atlg30380 hypothetical protein 20 1 0 12 0.1 0.2
' At2g06520 hypothetical protein 76 3 l 36
; At2g15290 unknown protein 4 l 2 2
' Atlg74730 unlmown protein 3 l 0 1
At2g26500 unknown protein 6 0 O 7
putative ABC
; Atlg15210 transporter 0 10 0 2
At2g40380 unknown protein 0 6 0 0
At3g24550 ”me?“ kmase’ 1 9 0 2
- putative
‘ At4g01750 hypothetical protein 0 6 0 O
At5gl3390 putative protein 0 7 0 1
cellulose synthase
2 At5g16910 catalytic subimit - 0 7 l l
‘ like protein
phosphate/phosphoe
2 At5g33320 “°l'pymva'e 0 16 6 12
translocator
precursor
' putative cellulose
‘ At3g03050 synthase catalytic 0 8 l 3
subunit
Ca2+/H+-
_ At3g51860 exchanging protein- 0 l 4 0
like
j Atl g099 60 putative sucrose/11+ O 0 3 0
symporter
' Atlg61800 hypothetical protein 0 0 3 0
At4g07990 hypothetical protein 0 0 3 0
At5g52860 ABC "anipmer' 0 0 3 0
like protein
glucose-6—
At5g54800 phosphate/phosphate 0 0 4 1
translocator
1 EST library size 7566 18350 10678 42855

 

 

196

 

Figure 23. Digital differential display analysis.

AtPEM candidates with statistically different levels (p<0.005, where p is probability of
difference being by chance) of EST frequencies are displayed. Number of ESTs in the
'Flowers', 'Roots' or 'Seeds' pools was compared with that in the reference 'mixed' library
pool. The bars indicate the abundances of the ESTs (EST frequency axis) corresponding

to the proteins (proteins axis) in each library pools (tissue axis).

  
 
  
  
  
 
  
  
 
 
  

 

   

[\7 35
76-. 1‘ 130
, 1 " 1 25 :1”
20 g
. 1 C
O
, \110 ‘<
O

 
    

aseuix uiaimd ’

ioieoolsueii aieAnMd
[ouaoudsoud/eieudsoud

m
_m
m
m
.m‘.
.111

uieiOJd
'+FV+ZEO

      

Jeuodw/is
uiaiOJd aAiieind ‘ ,

SHEIMO'H '
81008
80338
umouxun
+H/esmons eAiieind
uieimd umouxun

uieimd |eoiieuiodltu
6u16ueuoxa

uiaimd leoiiauiodfiu

uiaiOJd umouxun

ugaiOJd leoiiauiod/lu
1.:

.9.

£2.

3

(n

197

The bioinformatics predictions of suborganellar localization together with tissue-specific
expression provides initial clues that may help in discovering the functions of these
proteins. However, it should be noted in such analysis that the normalization procedures
used in the construction of the cDNA libraries for EST projects can undermine the
statistical analysis and even with the high probability such as p<0.005 used in Table 10,
there could still be false positives (Audio and Claverie, 1997). Therefore, the digital
expression profile analysis should be considered as an initial step to find possible
candidates for differentially expressed genes from a large collection of data and needs to

be verified by experiments such as northern blots, RT-PCR (Reverse Trascription —

Polymerase Chain Reaction) or western blots, etc.

Plastids undergo massive changes during differentiation into their various forms
(chloroplast, chromoplast, amyloplast, leucoplast, etioplast, etc.) and need to import
different spectra of proteins according to their changed biochemical properties. However
the structure of the concentric pair of envelope membranes remains constant (Douce and
J oyard, 1990; J oyard et al., 1998). It is yet unknown the extent to which the proteome of
the envelope will also remain unchanged. The low number (21) of candidates
differentially expressed between the different tissues may indicate that the envelope
proteome candidates are relatively constitutive in terms of transcript abundance across
the different tissues. Furthermore, when the tissue-specific EST distributions of AtPEM
candidates and that of 39 thylakoid localized proteins and 1,802 stromal proteins were
compared, AtPEM candidates showed the least changes (Table 11). Whereas only 8% of
the AtPEM candidates were tissue-specific at p<0.05, 69% of the analyzed thylakoid

localized proteins showed specific EST tissue distributions. The stromal proteins also

198

displayed a higher level (19% at p<0.05) of differential expression than the AtPEM
candidates. Thus based on EST frequencies from different cDNA libraries, the transcript
abundance for the envelope membrane proteins appears relatively constant compared to

that of the stromal or thylakoid proteins.

Microarray Analysis

In order to further investigate the tissue-specific gene expression pattern of AtPEM
candidates, we analyzed publicly available cDNA microarray data from the Stanford
Microarray Database (SMD, http://genome-www.stanford.edu/microarray, Sherlock et
al., 2001) and from the Arabidopsis developing seed array website at
http://www.bpp.msu.edu/Seed/SeedArray.htm (Girke et al., 2000; Ruuska et al., 2002).
Expression profiles of 230 AtPEM candidates for about 110 different microarray
experiments were found in SMD public domain (as of March 12, 2002). We specifically
examined tissue comparison data which were publicly available for comparisons between
flower, leaf, or root versus the whole plant for 147 AtPEM candidates and for seed versus
leaf or seed versus plantlets for 59 AtPEM candidates. Figure 24 presents a summary of

genes which displayed more than 2 fold changes in at least 2 different tissue comparisons
and the pattern of which agreed in all combinations (i.e. At1g29390 expression is higher
in leaf compared to that in flower and higher in leaf also when compared to that in root).

Twenty two candidates showed higher levels of transcript abundance in leaf when

199

Table 11. Tissue-specific transcript abundance comparison between the plastid

envelope, thylakoid and stroma localized proteins.

 

 

envelope thylakoid stromal
number of predicted proteins analyzed a 541 39 1802
b
number of ESTs 2349 1961 17970

. . . . . 0
genes w1th altered EST tissue distribution

at probability < 0.05 8% (42) 69% (27) 19% (342)
at probability < 0.01 2% (13) 51% (20) 4% (71)

 

a Predicted envelope localized proteins were from AtPEM candidates. Thylakoid proteins

were selected from known thylakoid proteins and their homologs mostly involved in
photosynthesis. Stromal localized proteins were randomly chosen from the predicted
plastidial proteins (by TargetP) subtracted by those with membrane spanning domains

(predicted by TMHMM) and known thylakoid proteins.
b . .
Number of ESTs corresponding to the analyzed genes. The ESTs for thylakmd and

stromal proteins were from TIGR Gene Index Tentative Consensus.

C Percent total number of genes analyzed that showed difference in EST abundance (at

probability of difference being by chance less than 0.05 and 0.01) in pair wise
comparisons between ‘seeds’, ‘leaves’, ‘flowers’, ‘roots’ library pools. The actual

number of genes is parenthesized. Probability calculations were according to Audio and

Claverie (1997).

200

compared to that of root whereas only 3 showed higher levels in root than in leaf. Twenty
one of these 22 candidates (expressed higher in leaf than in root) also had higher
transcript abundance in leaf than that in flower whereas only 3 candidates showed higher
expression in flower than in leaf. Comparison between the seed and the leaf also showed
higher levels of transcript abundance in leaf (4 higher in seed, 23 higher in leaf).
Although it is well known that thylakoid and stromal chloroplast proteins are very
abundant in leaves, very little comparative information is available on the envelop
proteins. The data of Figure 24 suggest that transcripts for many plastid envelope proteins

are also more abundant in leaves than from heterotrophic tissues.

Six genes selected by the digital expression profile analysis in Table 10 as tissue-specific
were also present in the microarray data shown in Figure 24. Although the digital
expression profiles used mixed tissues as a reference and microarray data compared
individual tissues or plantlets, data for all six genes agree at least partially between the
two types of analysis. However the cross validity between the statistical EST analysis and
the microarray analysis should be further characterized by larger sets of genes which
have data analyzed by both methods, combined with additional experiments (northern

blots, RT-PCR, etc.) to verify results.

Changes in transcript abundance in seeds during 5 to 13 days after flowering were
recently reported by Ruuska et al. (2002). Twenty-six AtPEM candidates changed more
than two fold during this period (data not shown). The expression of many genes

involved in seed storage protein, starch and lipid biosynthesis, also change during these

201

Figure 24. Tissue-specific microarray analysis of AtPEM candidates.

AtPEM candidates with > 2 fold tissue—specific expression in at least 2 different tissue
comparisons are shown clustered according to the tissue specificity (indicated with the
brackets and labeled). The intensity of green and red colors represents the relative level
of transcripts (refer to the color keys at the bottom for the actual ratios) in tissues under
comparison (the tissues under comparison are indicated above the image and colored with
representative colors). The data for flower, leaf, root comparisons were from SMD
(http://genome-www.stanford.edu/microarray) and data for seed versus leaf and seed
versus plantlet were from the Arabidopsis developing seed array (Girke et al., 2000;
Ruuska et al., 2002). The graphic was generated by Tree View software (Eisen et al.,

1998).

202

flower vs root

9w
leaf vs flower

leaf vs root

flower
I 1‘1

leaf

 

Color Intensity
L092 (Red/Green)

 

 

 

AtPEM genes with > 2 fold
tissue-specific
expression based on
AFGC microarray

 

At3949220
At5904770
At5912860
At1912760
At1921060
At1929390
At1930380
At1952220
At1956200
At1975460
At2906520
At2926500
At2929670
M391 1810
At3915780
At3926580
At3926710
At3950680
At3956010
At5902160
At5913720
At5920110
At5923240
At5937360
At1980830
At5g16910
At5917640

    

reen vs red

seed

leaf I plantlet

 

leaf vs seed

plantlet vs seed

 

 

 

AtPEM genes with > 2
fold tissue-specific
expression based on
Seed microarray

 

At2934460
At3953580
At4914690
At1920500
At4925650
At4929490
At5911300
At5919540
At5904990
At5906130
At2942220
At2946550
At1954320
At1904940
At1964350
At2931040
At2915290
At1910830
M391 1 170
At4930950
At1976100
At2906520
At5923040
At4go1 150
At2935260
At1954780
At2946820

 

_.'~.~v s
1.9.4., u v
.

5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 O -0.5 -1 -1.5 -2 -2.5 -3 -3.5 -4 -4.5 -5

203

 

stages (Ruuska et al.,2002) and therefore these 26 AtPEM candidates may be involved in

the role of plastids in seed filling and development.

Towards the Function of Unknown Candidates

Gene knock-outs often can provide key information to link genes of unknown function to a
phenotype. Large collections of T-DNA insertion mutants are publicly available and when the
DNA flanking the T-DNA insertion sites are sequenced and aligned with the Arabidopsis genome

sequence, provide mutants useful for studying gene function. We searched the Sequence-
Indexed Library of Insertion Mutations generated by the Salk Institute Genome Analysis
Laboratory (http://signal.salk.edu/tabout.html; containing 32,758 T-DNA sequences as of
March 8, 2002). In total 388 insertion lines corresponding to 217 AtPEM candidates
(approximately 40% of AtPEM candidates) were found. The

‘Insertion 1135’ can be used to search and to order the mutant line seed stocks from the
Arabidopsis Biological Resource Center (ABRC,http://www.biosci.ohiostate.edu
/~plantbio/Facilities/abrc/abrchomehtm) at Ohio State University. These ‘Insertion 1135’
as well as the accession numbers to search the publicly available microarray data in SMD
(http://genome-www.stanford.edu/microarray) were deposited on our web site

(http://www.plantbiology.msu.edu/PlastidEnvelopeO.

Part 11. Candidates for Lipid Transporters of the Plastid Envelope

Plastid targeted ABC transporters

204

The ATP-binding cassette (ABC) protein super family is one of the largest protein
families known in the Arabidopsis genome with more than 120 members (Sanchez-
Ferdandez et al., 2001). The overexpression of some ABC proteins has conferred
multidrug resistance phenotypes in mammals (Gottesman and Pastan, 1993). Some of the
known fimctions of ABC transporters include i) plasma membrane efflux pumps for the
amphipathic cations, ii) translocation of cationic phospholipids between membrane
bilayer leaflets, iii) glutathione S-conjugate pumps of organic anions, iv) transport of
anticancer drugs, pigments, iron-sulfer complex, peptides and long-chain acyl-CoA
substrates (See Sanchez-Fernandez et al., 2001 for review). Therefore we hypothesized
that some of the ABC transporters in Arabidopsis might be involved in the

hydrophobic/amphipathic lipid movements across the plastid double-membrane envelope.

There were 16 potential ABC transporters among the AtPEM candidates (Table 7). In
order to include as many candidates as possible for the reverse genetics screening, 4
additional ABC transporters (At5g61700, At3g47750, At5g61690 and At5g61740) were
included as potential ABC transporters in the plastid envelope. Target P predicted those
4 proteins as mitochondrial targeted whereas another subcellular localization predictor,
Predator (http://www.inra.fi/predotar/) predicted them to be either dual targeted or
plastidial. Among these 20 candidates, T—DNA insertion mutant lines were available for
11 candidates from the Salk Institute Genome Analysis Laboratory (http://signal.salk.

edu/tabout.html, Table 12). The T-DNA (5.5-kb) insertions into these 11 candidates were
identified by PCR screening. Homozygous insertion lines for 9 candidates were

established which showed no PCR bands with gene specific primers but showed bands

205

with a gene specific primer and a T-DNA lefi border primer set. The identities of the
amplified bands were confirmed by sequencing for some of those lines (SALK_013945,
SALK_000578, SALK_015438, SALK_016209) (data not shown). No homozygous
knock-outs were found from the seed stocks SALK_008761 (Atl g15210) and SALK_

056267 (At3g25620) (Table 12).

Phenotypic Observations and the Fatty Acid Composition/Contents in the K 0 plants of

the Plastid ABC transporter candidates

What would be the expected phenotype if the lipid transporters on the plastid envelope
are disrupted? Many Arabidopsis lipid mutants do not display dramatic growth
phenotypes because Arabidopsis possesses two different pathways for the synthesis of
membrane glycerolipids, namely, prokaryotic and eukaryotic pathways (Browse and
Somerville, 1994). When one pathway is blocked, the cell redirects fatty acid flux
through the other pathway to compensate the losses. For example, act] mutants have a
defect in the acyl-ACP sn-glycerol-3-phosphate acyltransferase, the first enzyme of the
prokaryotic pathway but the plant do not look different from the wild type plants because

the deficiency is compensated for by increase in the eukaryotic synthesis of chloroplast
lipids (Kunst et al., 1988). Partial deficiency in the export of certain fatty acids resulted
in reduced growth and abnormality in seeds in plants knocked out in the FA TB (acyl-acyl
carrier protein thioesterase B) gene (Bonaventure et al., 2003). However, the major cause
for the phenotype was attributed to the lack of palmitate not the reduced amount of the

fatty acid exported from the chloroplast. If fatty acid export from the plastid is mediated

206

Table 12. The Arabidopsis T-DNA insertion mutants for the predicted plastid ABC
transporters.

 

 

Name Locus ID aStock ID insertion Gene specific primers for PCR screen
ID 1 Atl g15210 SALK_008761 Exon 5'—CGTGTGGAACAGGTTACCTC-3'
5'-CTGGTGATGAGGACTTACGG-3'
ID 2 Atlg15520 SALK_01394S Exon 5'-G'ITCCAGG‘I'l'I'I‘CATCGAGA-3'
5'-TATTCTGTGGACATAGCGAGAG-3'
ID 5 At1g59870 SALK_000578 Exon 5'-CTAAATGGGGTCATCGTTATFG-3'
5'-CCATAACGAGTACCAACACCTT-B'
ID 11 Atlg70610 SALK_015438 Intron 5'-TGGATGTCAAGTGGCTI‘AGG-3'
5'-TGGAGAGTGAGGGTGATGAA-3'
ID 13 At2g39190 SALK_016209 Intron 5'-CCACTGTAGGTGAACAAAGGG-3'
5'-ACCGAATGAAGAATCTTCCTTG-3'
ID 21 At3g25620 SALK_056267 Exon 5'-CGCGTTGCI‘CA'I‘TTACAAGA-3'
5'-CCACAACAAAGCCACTCATG-3'
ID 23 At5g03910 SALK_052673 Exon 5'-TGGAAAAAG'I‘TGCI‘GGAGAAG-3'
5'-TCAAAAAGGCATAACTAGGGAA-
3'
ID 27 At5g64940 SALK_045739 Intron 5'-CTTGATCCTTGTACCCTTGC-3'
5'-TAGTTGCAGGTGTCTGAGAACA~3'
ID 32 At3g47750 SALK_041452 L300-5' 5'-TGG"ITI‘ACCTCGT1‘ATCITCGT-3'
5'-TGGGAAGATCTGAGTGAATCC-3'
ID 26 At5g52860 SALK_015052 Exon 5'-CCACACAAGACTCGCTCAAG-3'
5'-GGCAAGAGAGCTCAAGAAGAG-3'
ID 71 Atlg54350 SALK_003891 Exon 5'-GAGTCTCAGCAGCGTGATGTT-3'
5'-GAGATAGCGCTGCAGGTAAT-3'
SALK_069087 Exon 5'~GAGTCTCAGCAGCGTGATGTT-3'

5'—ATCACACTGCTTAACGAGGC-3'

 

Note. The lines with homozygous T—DNA insertion found were indicated by bold cases.
a seed stock identification numbers from the Arabidopsis Gene Mapping Tool (http://

signal.salk.edu/cgi-bin/tdnaexpress) and fiom the Arabidopsis Biological Resource

Center.

the posuion of T-DNA insertion from the ArabIdOPSlS Gene Mapping Tool.

207

by proteins then disruption of such proteins may cause increased flux through the
prokaryotic pathway which may result in increased 16 carbon fatty acids in the overall

fatty acid composition. Alternatively, growth impairment might occur if free fatty acids
were to accumulate or if the defect reduces the export of the saturated or other fatty acids
which play a crucial role for plant growth. It may also be possible to detect the effect of
blockage in the return of lipids from the ER pathway through fatty acid composition
analysis by examining the signatures typical for each pathways (lipids derived fi'om the
prokaryotic pathway contain 18 carbon fatty acids at the sn-1 position and 16 carbon fatty
acids at the sn-2 position whereas ER derived lipids contain 18 carbon fatty acids in both
positions). A mutation in the recently discovered permease-like protein (T GD] ), which
was proposed to be involved in the transfer of diacylglycerol moiety from the ER to the
plastid, showed increased prokaryotic pathway-derived galactolipids (Xu et al., 2003).
This mutation however did not cause decrease in the overall amount of galactolipids nor
did it show significant growth phenotype, again showing the remarkable flexibility of the
plant using the alternative pathways. The enzymes responsible for the MGDG and the
DGDG synthesis (MGDI and DGDI respectively) are localized specifically to the inner
membrane and to the outer membrane of the plastid envelope (Awai et al., 2001;
Froehlich et al., 2001). The MGDG synthesized in the inner membrane needs to be
transported to the outer membrane for DGDG synthesis. The synthesized DGDG should
be transported back to the inner membrane and eventually transferred to the thylakoid
membrane. A more drastic effect is expected similar to that in the dgd] mutant if any of

these processes are disrupted for DGDG availability in the photosynthetic membrane.

208

None of the selected 9 homozygous insertion mutants showed any visual grth
phenotype. Initially, when the 132 (11 candidates X 12 plants per each candidate) T3
seeds from ABRC stock center were grown, many showed various abnormalities
including slow growth, very late bolting, smaller plants etc. However, these phenotypes
did not correlate well with the homozygosity of the insertion events. Furthermore, when
seeds collected from these plants were re-grown (next generation), most of the abnormal
phenotypes were abolished. The fatty acid composition of leaves and seeds of the 9
homozygous T-DNA knock-out plants are shown in Table 13. When compared to the
samples from the wild-type plants there was no marked change in the fatty acid
compositions. The oil content in the seeds was also analyzed and lines that showed
marked changes are presented in Table 14. The homozygous knock-outs of At1g70610
and At5g64940 displayed reduction in the seed oil contents (60-70% of the wild type
seed oil content). When the seeds from the progeny (T4 generation) of these mutants
were analyzed, they consistently showed reduced oil contents in seeds (ID 11-3-1-2, ID
27-8-1-1, ID 27-8-1-3 in Table 14). These results are still preliminary (because the
purpose of this initial analysis was to select more interesting lines among the large

number of plants) and needs more accurate statistical measurements to confirm the

results.

209

CONCLUSIONS

In this study we attempted to predict integral plastid envelope proteins from the
Arabidopsis thaliana nuclear genome using computational methods (Part I) and from
these candidates, putative plastidial ABC transporters were selected and studied by
reverse genetics approach in an effort to identify proteins that could potentially be
involved in the lipid transport across the envelope membranes (Part 11). So far there has
been no established inventory of plastid envelope proteins. This work is one of the first
published (K00 and Ohlrogge, 2002) works attempting to create a database for the entire

plastid envelope proteome from Arabidopsis thaliana.

As the word ‘candidates’ implies the results of our study are not definitive due to the
nature of prediction software and due to the possibility of errors in protein annotations in
the genome databases (van Wijk, 2001). Although it is likely that at least 10% of the
AtPEM candidates represent incorrect predictions, it is reasonable to assume that the
selected candidates represent a large portion of the real envelope proteome.

In addition to a broader interest in plastid metabolism, my specific interest was to identify
putative lipid transporters in the plastid envelope. The road to the actual identification
and characterization of a novel lipid transporter by these approaches is still long.
However, the initiated analysis of putative ABC transporters of the plastid membrane

lays a foundation towards future attempts to include all the possible ABC transporter

210

Table 13. Percent wild type fatty acid compositions in the leaves and seeds of the

homozygous T-DNA knock-out plants of the plastid ABC transporter candidates.

 

 

 

leaves seeds
FAME class a C16 C18 C16 C18 C20 C22

ID 21 96.4 101.8 93.3 92.3 93.2 104.2
ID 2-1-1 - - 110.2 105.4 86.9 82.7
ID 53 93.9 103.8 102.8 102.4 106.5 95.6
ID 11b 96.1 102.0 96.6 96.3 98.5 101.6
ID 11-3-1-2 - - 114.9 108.1 85.8 -

ID 135 100.0 99.6 101.5 102.0 107.0 95.0
ID 235 89.6 105.5 99.5 98.5 102.1 105.6
ID 26 b - - 99.8 100.6 97.9 100.6

b

ID 27 92.1 104.3 94.6 94.6 99.6 112.9
ID 27-8 - — 109.3 106.2 87.4 44.9
ID 32 b 91.1 107.3 95.1 97.7 103.0 104.3
ID 433 - - 97.3 92.6 89.9 108.3
ID 43-3-1-1 - - 104.3 100.9 97.4 97.3
ID 71 b - - 103.5 100.7 98.0 100.4

 

Note. The name of each knock-out lines follows Table 12. Tissue samples from 3-5
independent wild type plants were used for the composition analyses and the average

values are uses as references.
a fatty acid methyl ester species in the total lipid extract classified according to the
number of carbons.

values of these lines are average values of samples from at least 3 independent

progenies.

211

Table 14. Homozygous knock-outs of plastid ABC transporter candidates that

showed reduced oil content in seeds compared to the wild type plants.

 

 

 

oil content % wild typea
Locus ID Name Homozygous KO lines
(Hg/20 seeds) oil content
b
At1g70610 ID 11-3 SALK_015438-3 66.5 55.17
b
ID 11-5 SALK_015438-5 71.3 59.16
ID 11-3-1-2 SALK_015438-3-1-2 66.3 54.83
15
At5g64940 ID 27-8 SALK_045739-8 89.1 73.95
ID 27-8-1-1 SALK_045739-8-l-1 72.4 59.86
ID 27-8-1-3 SALK_O45739-8-1-3 71.9 59.49

 

a Oil contents of seeds from 3-5 independent wild type plants were used as references.

average of 2 independent extractions.

212

candidates (only roughly half were analyzed so far) in the search for lipid transporters

and towards the bigger plan of assigning fimction to all genes of Arabidopsis.

MATERIALS AND METHODS

Establishment of a Database for Plastid Envelope Protein Candidates

The results of TargetP prediction for Arabidopsis chromosome 2 and chromosome 4 were
from the web server at http://www.cbs.dtu.dk/services/TargetP/predictions/pred.html.
The sequences for these two chromosomes were from The Institution of Genomic
Research (TIGR) and the European Union Arabidopsis Sequencing Consortium (as of
January 7, 2000). Additional sequence retrieval was from the National Center for
Biotechnology Information WCBI) Batch Entrez web-server (http://www.ncbi.nlm.nih.
gov:80/Entrez/batch.html). Chromosome 1, 3, and 5 sequences were downloaded from
the Munich Information Center for Protein Sequences (MIPS) Arabidopsis thaliana
Database fip site (fip:l/ftpmips.gs£de/cress/, as of June 2001). Subcellular localization
was predicted using TargetP v1.01 from the Center for Biological Sequence Analysis
(http://www.cbs.dtu.dk/services/TargetP/). No cutoff was applied but instead ‘Reliability
Class’ (RC) values from 1 to 5 were designated for each predicted proteins. Proteins with
plastid transit peptides were then evaluated for membrane spanning domains using the

TMHMM v2.0 web server at http://www.cbs.dtu.dk/services/TMHMM-2.0/.

213

Proteins that were not strongly predicted to be transmembrane proteins by the program
yet contained weak or-helical domains with probabilities over 50% were also included in
the database as possible membrane proteins. Proteins with transmembrane or-helical
domains within the range of possible plastid targeting sequences (plastid targeting
sequence prediction was according to the TargetP cleavage site prediction) were removed
and in cases where the domain located close to but not within the predicted cleavage site
were marked ‘N-term’ in order to reduce false prediction of hydrophobic targeting
sequences as membrane spanning domain. Proteins that are known to locate in thylakoid

were removed manually.

Classification by Function

Functional classification was based on the MIPS Arabidopsis thaliana Database
automatically derived functional categories. The catalogue was downloaded from the
Protein Extraction Description and Analysis Tool (PEDANT) web server
(http://pedant.gsf.de/). The catalogue of genes for plant glycerolipid biosynthesis was
from the web server at http://www.canr.msu.edu//lgc/ (Mekhedov et al., 2000). An
updated inventory of lipid metabolism related genes was available at our regional
database (Beisson et al., 2003). The full inventory of Arabidopsis ABC proteins was
downloaded from the website at http://www.arabidopsisabc.net/ (Sanchez-Femandez et

al., 2001) and was queried for AtPEM candidates.

Digital mRNA Expression Profiling

214

A set of all public Arabidopsis ESTs was obtained through a Structured Query Language
(SQL) query of our in house ‘SeqStore’ database which contained 103,109 EST
sequences from GenBank Expressed Sequence Tag database (dbEST, http://www.ncbi.
nlm.nih.gov/dbEST/index.html). The EST sequences were used as queries (BLASTN
version 2.2.1) against the target database of all predicted transcripts from Arabidopsis
genome (ftp://f’cp.tigr.org/pub/data/a_thaliana/athl/, January 10, 2002). Then the plastid
envelope protein candidates were queried against this resulting database using common
chromosomal locus identifiers (i.e. At2g39040). Tentative Consensus (TC) sequences
assembled from ESTs and Expressed Transcript (ET) sequences were retrieved from
TIGR Arabidopsis Gene Index (http://www.tigr.org/tdb/agi/). In order to match TCs with
the chromosomal locus identifiers, TCs retrieved were mapped to all the predicted
transcripts of Arabidopsis genome by Blast alignments (BLASTN version 2.2.1). TCs
corresponding to the plastid envelope protein candidates were selected by chromosomal

locus identifiers.

To identify possible differential expression of the AtPEM candidates, the relative
frequencies of ESTs between tissue-specific library pools were compared (55 different
EST libraries were grouped into 8 tissue—specific ‘library pools’ according to the source
tissues from which the library was derived (Beisson et al., 2003). The influence of
random fluctuations and sampling size was considered statistically to discern the
reliability of the digital expression profiling (Audio and Claverie, 1997). For this

analysis, software was downloaded from http://igs-server.cnrs-mrs.fr. EST analysis for

215

the thylakoid and stromal proteins was done using 39 selected known thylakoid localized
proteins, most of which are related to photosynthesis, and 1,802 stromal proteins which
were chosen randomly from plastidial proteins predicted by TargetP and subtracting
those with transmembrane domains and obvious thylakoid proteins. ESTs for the

thylakoid and stroma localized proteins were from TIGR Arabidopsis Gene Index TCs.

Microarray Data

Public microarray data for AtPEM candidates were searched fiom a local database that
contained most of the microarray data downloaded from Stanford Microarray Database
(as of March 1, 2002, Sherlock et al., 2001, ) and was courtesy of Rodrigo Gutierrez. The
duplicates in the tissue comparison data sets (SMD experiment identifiers: 7197, 7199,
7200, 7201, 7203, 7205) were averaged and divided each other in a way to give pair wise
comparisons between flower, leaf and root tissues. Arabidopsis developing seed array
data was from local database and are available at http://www.bpp.msu.edu/Seed/
SeedArray.htm (Girke et al., 2000; Ruuska et al., 2002). The cluster image (Figure 24)
was generated by Tree View software downloaded from the website at http://rana.lbl.gov/

(Eisen et al., 1998).

T -DNA Insertion Mutants

T-DNA insertion mutant lines were searched from the Sequence-Indexed Library of
Insertion Mutations generated by the Salk Institute Genome Analysis Laboratory using

Arabidopsis Gene Mapping Tool (http://signal.salk.edu/cgi-bin/tdnaexpress).

216

Information on AtPEM candidates and their various attributes including tissue-specific
EST frequencies, accession numbers to search SMD, T-DNA insertion mutant stock
numbers from Salk Institute Genome Analysis Laboratory can be downloaded from our

web site at http://www.plantbiology.msu.edu/PlastidEnvelope/.

Homozygous T -DNA Insertion Mutant Isolation

T-DNA insertion mutant lines for plastid ABC transporters were searched by Arabidopsis
Gene Mapping Tool (http://signal.salk.edu/cgi-bin/tdnaexpress) and the mutant seed
stocks were obtained through the Arabidopsis Biological Resource Center (ABRC,
http://www.biosci.ohiostate.edu/~plantbio/Facilities/abrc/abrchome.htm) at Ohio State
University. Homozygous KO lines were selected among the T3 generations by PCR
screening for both the presence of T-DNA insertion and the absence of an intact gene.
The gene-specific primers used for the screening of insertions are shown in Table 12. The
T-DNA—specific primer matching the lefl end of the T-DNA was 5' TGGTTCACGT

AGTGGGCCATCG 3'. Seeds from the selected homozygous knock-outs (T4 generation)
were grown again and the second round screening was carried out to confirm the

homozygosity.

Fatty Acid Composition and Content Measurements by Gas Chromatography

About 0.1 g fresh weight leaves or 20-30 seeds were heated at 90°C for 1h 30 min in 0.3

mL of toluene and 1 mL of 5% (v/v) H2804 with heptadecanoic acid (17:0) as an internal

217

standard. After the samples were cool, 1.5 mL 0.9% (w/v) NaCl was added and fatty acid

methyl esters were extracted 3 times with 2 mL of hexane. Samples were evaporated

under a stream of N2 and redissolved in 300 1.1L hexane. The fatty acid methyl esters

were analyzed by GC with a flame ionization detector (GC-FID) on a DB-23 capillary

column (J &W Scientific, Folsom, CA).

218

LITERATURE CITED

Adessi C, Miege C, Albrieux C, Rabilloud T (1997) Two-dimensional electrophoresis
of membrane proteins: a current challenge for immobilized pH gradients. Electrophoresis
18:127-135.

Allikmets R, Singh N, Sun H, Shroyer NF, Hutchinson A, Chidambaram A,
Gerrard B, Baird L, Stauffer D, Pfeiffer A, Rattner A, Smallwood P, Li Y,
Anderson KL, Lewis RA, Nathans J, Leppert M, Dean M and Lupski JR (1997) A
photoreceptor cell-specific ATP-binding transporter gene (ABCR) is mutated in recessive
Stargardt macular dystrophy. Nat. Genet. 15: 236—245.

Audie S, Claverie JM (1997) The significance of digital gene expression profiles.
Genome Res. 7: 986-995.

Awai K, Marechal E, Block MA, Brun D, Masuda T, Shimada H, Takamiya K,
Ohta H, Joyard J (2001) Two types of MGDG synthase genes, found widely in both
16:3 and 18:3 plants, differentially mediate galactolipid syntheses in photosynthetic and
nonphotosynthetic tissues in Arabidopsis thaliana. Proc. Natl. Acad. Sci. U S A. 98:
10960-10965.

Bannai H, Tamada Y, Maruyama O, Nakai K, Miyano S (2002) Extensive feature
detection of N-terminal protein sorting signals. Bioinforrnatics 18: 298-305.

Beisson F, Koo AJ, Ruuska S, Schwender J, Pollard M, Thelen JJ, Paddock T, Salas
JJ, Savage L, Milcamps A, Mhaske VB, Cho Y, Ohlrogge JB (2003) Arabidopsis
genes involved in acyl lipid metabolism. A 2003 census of the candidates, a study of the

distribution of expressed sequence tags in organs, and a web-based database.
Plant Physiol. 132: 681-697.

Block MA, Dorne AJ, Joyard J, Douce R (1983) Preparation and characterization of
membrane fractions enriched in outer and inner envelope membranes from spinach

chloroplasts. I. Electrophoretic and immunochemical analyses. J. Biol. Chem. 258:
13273-13280.

Bolter B, 8011 J (2001) Ion channels in the outer membranes of chloroplasts and
mitochondria: open doors or regulated gates? EMBO J. 20: 935-940.

Bonaventure G, Salas JJ, Pollard MR and Ohlrogge JB (2003) Disruption of the
FATB gene in Arabidopsis demonstrates an essential role of saturated fatty acids in plant
growth. Plant Cell 15: 1020-1033.

Borst P, Zelcer N and van Helvoort A (2000) ABC transporters in lipid transport.
Biochim. Biophys. Acta. 1486:128-144.

219

Bouchez D, Hofte H (1998) Functional Genomics in Plants. Plant Physiol. 118: 725-732.

Browse J and Somerville CR (1994) Glycerolipids. In Arabidopsis. Meyerowitz EM,
and Somerville CR ed. (Cold Spring Harbor Laboratory Press, New York) pp. 881-912.

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.

Cline K, Henry R (1996) Import and routing of nuclear-encoded chloroplast proteins.
Annu. Rev. Cell Dev. Biol. 12: 1-26.

Douce R, Joyard J (1990) Biochemistry and function of the plastid envelope. Annu.
Rev. Cell Biol. 6: 173-216.

Eisen MB, Spellman PT, Brown PO, Botstein D (1998) Cluster analysis and display of
genome-wide expression patterns. Proc. Natl. Acad. Sci. U S A. 95: 14863-14868.

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:
1005-1016.

Emanuelsson 0, von Heijne G (2001) Prediction of organellar targeting signals.
Biochim. Biophys. Acta. 1541: 114-119.

Ferro M, Seigneurin-Berny 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: 3517-3526.

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 22325-45.

Ferro M, Salvi D, Riviere-Rolland H, Vermat T, Seigneurin-Berny 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.

Fischer K, Kammerer B, Gutensohn M, Arbinger B, Weber A, Hausler RE, Flugge
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.

Frishman D and Argos P (1996) Incorporation of long-distance interactions into a
secondary structure prediction algorithm. Protein Engineering 9: 133-142.

220

Flugge U] (2000) Transport in and out of plastids: does the outer envelope membrane
control the flow? Trends Plant Sci. 5:135-137.

Footitt S, Slocombe SP, Larner V, Kurup S, Wu Y, Larson T, Graham 1, Baker A,
Holdsworth M (2002) Control of germination and lipid mobilization by COMATOSE,
the Arabidopsis homologue of human ALDP. EMBO J. 21: 2912-2922.

Frishman D, Albermann K, Hani J, Heumann K, Metanomski A, Zollner A, Mewes
HW (2001) Functional and structural genomics using PEDANT. Bioinforrnatics 17: 44-
57.

Froehlich JE, 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 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: 413-425.

Fulgosi H, Soll J (2001) A gateway to chloroplast-protein translocation and beyond. J
Plant Physiol. 158: 273-284.

Girke T, Todd J, Ruuska S, White J, Benning C, Ohlrogge J (2000) Microarray
analysis of developing Arabidopsis seeds. Plant Physiol. 124:1570-81.

Gottesman MM and Pastan l (1993) Biochemistry of multidrug resistance mediated by
the multidrug transporter. Annu. Rev. Biochem. 62: 385-427.

Hayashi M, Nito K, Takei-Hoshi R, Yagi M, Kondo M, Suenaga A, Yamaya T and
Nishimura M (2002) Ped3p is a peroxisomal ATP-binding cassette transporter that
might supply substrates for fatty acid B—oxidation. Plant Cell Physiol. 43: 1—1 1.

Heins L, Collinson I, 8011 J (1998) The protein translocation apparatus of chloroplast
envelopes. Trends Plant Sci. 3: 56-61.

Hettema EH, Tabak HF (2000) Transport of fatty acids and metabolites across the
peroxisomal membrane. Biochim. Biophys. Acta. 1486:18-27.

J ackson-Constan D, Keegstra K (2001) Arabidopsis genes encoding components of the
chloroplastic protein import apparatus. Plant Physiol. 125: 1567-1576.

Joyard J, Teyssier E, Miege C, Berny-Seigneurin D, Marechal E, Block MA, Dorne
AJ, Rolland N, Ajlani G, Douce R (1998) The biochemical machinery of plastid
envelope membranes. Plant Physiol. 118: 715-723.

221

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: 105-117.

Kirk JTO, Tilney-Bassett RAE (1978) The plastids: their chemistry, structure, growth
and inheritance. Elsevier Press, Amsterdam.

Koo AJ, Ohlrogge JB and Pollard M (2004) On the export of fatty acids from the
chloroplast. J. Biol. Chem. in press.

Krogh A, Larsson B, van Heijne G, Sonnhammer EL (2001) Predicting
transmembrane protein topology with a hidden Markov model: application to complete
genomes. J. Mol. Biol. 305: 567-580.

Kunst L, Browse J and Somerville C (1988) Altered regulation of lipid biosynthesis in
a mutant of Arabidopsis deficient in chloroplast glycerol phosphate acyltransferase
activity. Proc. Natl. Acad. Sci. USA. 85: 4143-4147.

Mekhedov S, de Ilarduya OM, Ohlrogge J (2000) Towards a functional catalog of the
plant genome: A survey of genes for lipid biosynthesis. Plant Physiol. 122:389-401.

Miquel M, Browse J (1992) Arabidopsis mutants deficient in polyunsaturated fatty acid
synthesis. Biochemical and genetic characterization of a plant oleoyl-phosphatidylcholine
desaturase. J. Biol. Chem. 267 : 1502-1509.

Moller S, Croning MD, preiler R (2001) Evaluation of methods for the prediction of
membrane spanning regions. Bioinforrnatics 17: 646-653.

Moller S, Croning MD, preiler R (2002) Evaluation of methods for the prediction of
membrane spanning regions. Bioinfonnatics 18: 218.

Mosser J, Douar AM, Sarde CO, Kioschis P, Feil R, Moser H, Poustka AM, Mandel
JM and Aubourg P (1993) Putative X-linked adrenoleukodystrophy gene shares
unexpected homology with ABC transporters. Nature 361: 726—730.

Nakai K, Horton P (1999) PSORT: a program for detecting sorting signals in proteins
and predicting their subcellular localization. Trends Biochem. Sci. 24: 34-36.

Neuhaus HE, Emes MJ (2000) Nonphotosynthetic metabolism in plastids. Annu Rev
Plant Physiol. Plant Mol. Biol. 51: 111-140.

Ohlrogge J, Browse J (1995) Lipid biosynthesis. Plant Cell 7: 957-970.

222

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 12:3 19-341 .

Peltier JB, 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: 211-236.

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.

Pollard M, Ohlrogge JB (1999) Testing models of fatty acid transfer and lipid synthesis
in spinach leaf using in vivo oxygen-18 labeling. Plant Physiol. 121: 1217-1226.

Quackenbush J, Liang F, Holt 1, Pertea G, Upton J (2000) The TIGR gene indices:
reconstruction and representation of expressed gene sequences. Nucleic Acids Res.
28:141-145.

Ruetz S, Gros P (1994) Phosphatidylcholine translocase: a physiological role for the
mdr2 gene. Cell. 77: 1071-1081.

Ruuska SA, Girke T, Benning C, Ohlrogge JB (2002) Contrapuntal networks of gene
expression during Arabidopsis seed filling. Plant Cell 14: 1191-1206.

Santoni V, Molloy M, Rabilloud T (2000) Membrane proteins and proteomics: Un
amour impossible? Electrophoresis 21: 1054-1070.

Santoni V, Rouquie D, Doumas P, Mansion M, Boutry M, Degand H, Dupree P,
Packman L, Sherrier J, Prime T, Bauw G, Posada E, Rouze P, Dehais P, Sahnoun I,
Barlier I, Rossignol M (1998) Use of a proteome strategy for tagging proteins present at
the plasma membrane. Plant J. 16: 633-641.

Sanchez-Fernandez R, Davies TG, Coleman JO, Rea PA (2001) The Arabidopsis
thaliana ABC protein superfamily, a complete inventory. J. Biol. Chem. 276: 30231-
30244.

Sazuka T, Yamaguchi M, Ohara O (1999) CyanoZDbase updated: linkage of 234
protein spots to corresponding genes through N-terrninal microsequencing.

Electrophoresis 20: 2160-2171.

Schleiff E, Klosgen RB (2001) Without a little help from 'my' fiiends: direct insertion of
proteins into chloroplast membranes? Biochim. Biophys. Acta. 1541: 22-33.

Schnell DJ (1995) Shedding light on the chloroplast protein import machinery. Cell 83:
521-524.

223

Schubert M, Petersson UA, Haas BJ, Funk C, Schroder WP, Kieselbach T (2002)
Proteome map of the chloroplast lumen of Arabidopsis thaliana. J. Biol. Chem. 277:
8354-8365.

Seigneurin-Berny D, Rolland N, Garin J, Joyard J (1999) Technical Advance:
Differential extraction of hydrophobic proteins from chloroplast envelope membranes: a
subcellular-specific proteomic approach to identify rare intrinsic membrane proteins.
Plant J. 19: 217-228.

Sherlock G, Hernandez-Boussard T, Kasarskis A, Binkley G, Matese JC, Dwight
SS, KaIOper M, Weng S, Jin H, Ball CA, Eisen MB, Spellman PT, Brown PO,
Botstein D, Cherry JM (2001) The Stanford Microarray Database. Nucleic Acids Res.
29: 152-155.

Somerville C, Dangl J (2000) Plant Biology in 2010. Science 290: 2077-2078.

Sonnhammer EL, von Heijne G, Krogh A (1998) A hidden Markov model for
predicting transmembrane helices in protein sequences. Proc. Int. Conf. Intell. Syst. Mol.
Biol. 6: 175-82.

Sugita M, Sugiura M (1996) Regulation of gene expression in chloroplasts of higher
plants. Plant Mol. Biol. 32:315-26.

The Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the
flowering plant Arabidopsis thaliana. Nature 408: 796-813.

van Wijk KJ (2001) Challenges and prospects of plant proteomics. Plant Physiol. 126:
501-508.

White JA, Todd J, Newman T, Focks N, Girke T, de Ilarduya OM, Jaworski JG,
Ohlrogge JB, Benning C (2000) A new set of Arabidopsis expressed sequence tags from
developing seeds. The metabolic pathway from carbohydrates to seed oil. Plant Physiol.

124: 1582-1594.

Wilkins MR, Gasteiger E, Sanchez JC, Bairoch A, Hochstrasser DF (1998) Two-
dimensional gel electrophoresis for proteome projects: the effects of protein
hydrophobicity and copy number. Electrophoresis 19:1501-1505.

Xu C, Fan J, Riekhof W, Froehlich JE and Benning C (2003) A permease-like protein
involved in ER to thylakoid lipid transfer in Arabidopsis. EMBO J. 22: 2370-23 79.

Zhou Z, White KA, Polissi A, Georgopoulos C, Raetz CR (1998) Function of

Escherichia coli MsbA, an essential ABC family transporter, in lipid A and phospholipid
biosynthesis. J. Biol. Chem. 273: 12466-12475.

224

CHAPTER 5

CONCLUSIONS AND FUTURE RESEARCH PERSPECTIVES

Understanding how free fatty acids are exported from the plastid has been one long-
standing missing link in basic knowledge about plant lipid biochemistry. In most reviews
of plant lipid metabolism the export process has usually been drawn on biochemical
pathway diagrams without much elaboration as a continuous arrow from the site of de
novo fatty acid synthesis (FAS), the stroma, to the endoplasmic reticulum (ER)
membranes. In my work on fatty acid export presented in chapter 11 of this thesis (also
published in Koo et al., 2004), a new focus was applied to FA export and the process was
discussed in much more details, raising questions that were not so deeply asked in the
past. The data in Chapter II define and measure parameters that are needed to better
understand the molecular mechanisms of fatty acid export from the plastid in plants. The
data have suggested a more complicated mechanism than just a simple physical diffusion
model where the free fatty acid (FFA) flip-flops across the inner envelope membrane,
dissociates into the intermembrane space and finally repartitions into the outer envelope
membrane where the long-chain acyl-CoA synthetase (LACS) captures the FFA and thus
drives the net flow of the FFA from the stroma to the cytosol. A major conclusion of
Chapter II is that the data discount this simple free diffusion-based model for FFA export

from the plastid.

225

Three models for the mechanism of fatty acid export from the plastid and directions for

future studies

Three generic models (and combinations thereof) can be proposed as alternatives to the
simple diffusion model and are based on our current knowledge about the plastid
envelope and the enzymes related to the fatty acid metabolism. These models are: (1) a
non-protein-mediated facilitated-diffusion (channeling) between the stromal acyl-ACP
thioesterase and the outer envelope long-chain acyl-CoA synthase, (2) a cryptic inner
envelope LACS, and (3) intervening transfer proteins. These models are explained below
together with suggested directions for future studies to characterize and discriminate

between them.

Non-protein-mediated facilitated-diffiision begins with the preferential partitioning or
loading of newly synthesized free fatty acid onto the inner envelope membrane. The
exact sub-plastidial organization of the members of the FAS enzymes in vivo is not
known although some researchers believe it to be formed into a supra-molecular complex
in the stroma. The individual FAS enzymes are soluble enzymes but the observation that
the plastid retains FAS activity even when the isolated plastid is swollen to release up to
50% of its soluble proteins into the extraplastidial media make us believe that they are
formed into an enzyme complex (Roughan and Ohlrogge, 1996). The multisubunit
acetyl-CoA carboxylase is found to be strongly associated with the chloroplast envelope
through non-ionic interactions which supports the FAS super-complex association with

the envelope (Thelen and Ohlrogge, 2002). The final enzyme in the series of fatty-acyl

226

chain elongation reaction of FAS before the export from the plastid is acyl-ACP
thioesterase. The acyl-ACP thioesterase hydrolyses the growing chain of fatty acids from
acyl carrier proteins. Unpublished studies suggest at least the FatB form of this enzyme
is associated with the envelope membrane when imported into the pea chloroplast by in
vitro reconstitution experiments (communicated by Dr. John Froehlich, MSU-DOE Plant
Research Laboratory). The acyl-ACP thioesterase association with the inner membrane
should be characterized more carefully since it can partly explain the uni-directional
movement of the free fatty acid towards exit from the plastid by loading the FFA directly

on to the membrane.

Second, fiuther studies about the plastid envelope localized long-chain acyl-CoA
synthesis (LACS) enzymes are critical to gaining insights into the mechanism of fatty
acid export for all 3 models. From the data (chapter II) we learned that LACS has a)
excess capacity (may run at least 10 times faster than the de novo F AS rate) to handle the

entire flux of free fatty acid coming out from the site of synthesis, b) has very low in vivo

Km (the long-chain acyl-CoA production reached steady state at very low amount (bulk

concentration) of free fatty acid), and c) also has priority for binding FFA over the
extraplastidial proteins such as BSA. If the LACS can access FFA substrates supplied
directly from the inner membrane then at least most of our observations regarding the
efficient in vivo pool (‘pool A’) can be explained without invoking other transport
proteins. At this point it is not clear how many or which are the major players of
envelope LACS that is/are involved in fatty acid export. ‘There are 4 candidate genes for

the plastid LACS proteins in the Arabidopsis genome. The null mutants of these genes in

227

combination with localization studies should identify the major LACS’s. So far knock-
out plants for the suggested major plastid envelope LACS gene showed no growth
phenotype nor was there any compositional change in the lipids that might suggest
compromised export of acyl groups from the chloroplasts (Schnurr et al., 2002). In fact
the single T-DNA insertion knock-outs for all 4 candidates displayed no visible
phenotypes (data not shown) making it very important to create multiple knock-outs.
According to the current two-pathway scheme for the glycerolipid synthesis, disruption
of all envelope LACS proteins should cause major problems in the synthesis of ER
derived lipids and theoretically be lethal unless the other pathway can compensate for the
loss. A similar phenotype could be expected fiom mutants that are blocked in the fatty
acid export pathway from the plastid (for example, by knock-out of all acyl-ACP

thioesterase genes).

Once a LACS involved in fatty acid export is identified, experiments to find out the
precise location of the protein, whether on the inner envelope membrane facing the
intennembrane space or on the outer envelope facing the intermembrane space/cytosol,
should be pursued. This will provide important clues for the non-protein-mediated
facilitated diffusion model and the ‘cryptic’ LACS model. Protease protection assays
with either the in vitro plastid-targeted radiolabeled-LACS protein or the chloroplasts
with endogenous LACS using western blot as a detection method can be used (antibody
specific for this LACS should be raised and since LACS genes are not very well
conserved in sequence this shouldn’t be difficult). In addition using protease in

combination with the kinetic analysis of acyl-CoA synthesis with the isolated

228

chloroplasts may also give useful information about the contribution by different
LACS’s. However, if these multiple knock-outs show no effect on the ER-pathway, there
will need to be a paradigm change in our current scheme about fatty acid delivery from
the plastid to the ER and about the involvement of the known plastid LACS’s in this

process.

Third, several different approaches were/can be taken in order to test if a protein
mediated transport mechanism (model 3) is involved in FA export. However, these
approaches require more knowledge about the fatty acid export to be accumulated before
taking major risks, such as forward genetic screening or biochemical purification, to
identify the protein component of the export system. In this thesis I tried to lay the
foundations on which the future works can be built upon. As described briefly in chapter
1, I first began to collect information about lipid transporter candidates that have been
identified in various other organisms. Through this study I learned that at least the long-
chain acyl-CoA synthetase family and the ABC transporter family were consistently
involved in various aspects of lipid transport. This led me to focus on these proteins in
Arabidopsis. Eventually, as more and more lipid transporters get to be identified, the
sequence information from these proteins can be used to identify common lipid binding
motifs. This is especially important because according to my blast search attempts of the
animal- or bacterial-lipid proteins against the Arabidopsis genome sequences, the
majority did not have obvious orthologs in Arabidopsis. Making an inventory of this

review work about the identified lipid transporter candidates in all organisms should help

229

to organize our efforts in the future. As far as I know there is no such gene catalog

available yet that focuses on lipid transport.

Next I again took a more general approach to increase our understanding about the plastid
envelope proteome using bioinformatics tools. The plastid envelope is the site of various
biochemical reactions, including many of lipid metabolism and also serves as the
interface/barrier for many different cellular metabolisms which should be coordinated.
Since membrane proteins are harder to study, much less was known about the
hydrophobic integral membrane proteins of the envelope (only 34% of the AtPEM
candidates could be assigned to putative functions). The completed genome sequence of
Arabidopsis opened up a new opportunity to tackle this problem. I used various
computational methods to predict the integral plastid envelope membrane proteins. Other
information such as number of deduced amino acids, number of transmembrane domains,
expected cleavage sites of presequences, putative functions, number of expressed
sequence tags (EST’s) and their source libraries, microarray data, and the publicly
available T-DNA insertion mutants, etc., were combined and were made into an intemet
based data base (AtPEM) available for the public (http://www.plantbiology.msu.edu/
PlastidEnvelope/index.htrn). The AtPEM became one of the first published databases for

plastid envelope membrane proteins (Koo and Ohlrogge, 2002).

A reverse genetics approach was applied next to search for lipid transporters among the
AtPEM candidates. The approach was designed not only to look for proteins involved in
FFA export but also to look for proteins involved in the import of the ER pathway

derived lipids. The disruption in one of the two glycerolipid synthetic pathways is

230

expected to accompany the changes in the fatty acid composition as was exemplified by
many mutants in these pathways (Browse and Somerville, 1994). The tng mutant, which
was implied to be disrupted in the transfer of diacylglycerol moiety from the ER to the
plastid, also showed changes in the fatty acid composition, i.e., increase in the
prokaryotic pathway-derived galactolipids (Xu et al., 2003). Twenty ABC transporter
family member genes were selected as such candidates and 11 available T-DNA insertion
mutants were obtained. Among these 11, homozygous insertion mutants were
established for 9 of them. These were screened for visual phenotypic changes and for
fatty acid composition/content changes. In the initial screening of fatty acid composition,
changes were not observed, which might have been expected for the disruption in one of
the two ways of glycerolipid synthesis. However, two knockout lines were found to
have reduced oil content in their seeds. This result must be verified through statistical
analyses to evaluate the significance of the changes and also through complementation
experiment. In addition the time course of the oil accumulation dining seed development
should give more convincing results concerning abnormality in the oil content. In order
to understand whether the effect of the gene disruption was related to the influence on
lipid metabolism, fatty acid (pulse-) labeling experiments should be conducted.
Ultimately the transgenic plants over-expressing these genes can be tested to see their

ability to bring changes in oil content in seeds.

In addition, only 11 of 20 of the putative plastid ABC transporters have been examined

and the other 9 should be considered when the T-DNA knock-outs become available.

There are some additional characteristics to be examined with all the candidate knock-

231

outs in addition to the changes in fatty acid compositions and oil contents. First, the
disruption in FA export may result in increased FFA export pool size. Short pulse
labeling of FA’s by radiolabeled acetate could help to measure the changes in the F FA
pool size among those mutants. Second, the mutants with decreased DGDG contents and
increased MGDG contents could potentially identify a MGDG translocator between the
inner membrane and the outer membrane. The mutant plant with this type of lipid
composition is expected to show a visible growth phenotype similar to that observed in
the dgdl mutant. Perhaps the involvement of proteins in this process can be tested in a
separate experiment using various inhibitors of the ATPases, the P-glycoproteins and the
multidrug resistance proteins that are implied to be involved in lipid transport (i.e.
vanadate, verapamil, colchicines and vinblastine). Simple assays can be developed first
to test the effects of these drugs on the synthesis of DGDG from the radiolabeled UDP-
galactose. If the MGDG synthesis is not affected yet the DGDG synthesis gets inhibited
by those drugs this may indicate the involvement of those drug-sensitive proteins in the
translocation of MGDG from the inner membrane to the outer membrane. However the

indirect effects of those drugs on other pathways in plants should be taken into account.

If a putative glycerolipid transporter is identified through the T-DNA knock-out strategy,
then in vitro lipid transport assays in artificial-membrane vesicles can be used to directly
demonstrate its function. One example of such assays was developed by Stephan Ruetz
and Philippe Gros (1994) and was used to demonstrate the function of the Mdr2 protein
as a PC translocase. First, in order to prepare the artificial-membrane vesicles with the

expressed protein in same polarity (the expressed protein in these vesicles all face

232

outside), a yeast mutant, defective in the last step of the vesicular fusion pathway was
used. This mutant accumulates large amounts of unfused secretory vesicles at the
nonpermissive temperature. Then they used the internalization of the fluorescent PC
analog into the artificial-membrane vesicles in the presence/absence of the expressed
Mdr2 protein to demonstrate its function as a PC translocase. The asymmetric
distribution of the fluorescent PC in the outer and the inner leaflets of the lipid bilayer
was measured by fluorescence spectroscopy. Following the addition of a membrane
impermeable, fluorescent-PC reducing agent (dithionite) the internalized fluorescent PC
was deduced from the residual fluorescence. Similar assay have been used in plants by
Gomes et al.(2000) to demonstrate the role of an Arabidopsis P-type ATPase, ALA], in
the internalization of phosphatidylserine in reconstituted yeast membrane vesicles. These
assays in combination with the use of ATP, Mg2+ and various inhibitors of ABC
transporter will provide more direct evidence of the putative protein being a genuine

phospholipid transporter.

One possible biochemical strategy to look for potential fatty acid transporters would use
antibodies to proteins such as the acyl-ACP thioesterases or the LACS’s. The protein
cross-linking reagents can be used to cross-link the interacting proteins with the acyl-
ACP thioesterases or the LACS’s (Akita et al., 1997; Chou et al., 2003; Snyder et al.,
2000). Following solublization to remove membrane lipids, the cross-linked protein
complex can be immuno-precipitated with the antibodies raised against the acyl-ACP
thioestrases or the LACS’s. Alternatively, the cross-linked protein complex can be

separated on a native polyacrylamide gel and detected by the western blot. Protein mass-

233

spectrometry can be used to identify the peptides in the complex. Antibodies raised
against newly identified polypeptides can be tested for their ability to inhibit fatty acid
uptake. In addition, the expression pattern of such proteins in those plant tissues that
demand high fatty acid export from the plastid such as, in the rapidly expanding leaves or
in the developing seeds at the stage of oil accumulation, may provide additional
evidences that the protein is involved in fatty acid transport. At this point we don’t have
clear speculations about the effect of disrupting such components in plant. But perhaps
the reagents that were used in the animal systems to block FA transport such as various
anion transporter inhibitors, fatty acid analogues, photoaffmity labeling reagents, etc.
(Abumrad et al., 1998), could be tried in isolated chloroplasts first for their effect on the
fatty acid export. This type of inhibition studies although no guarantee for the success

may give us information about the effects of disrupting fatty acid export from the plastid.

A novel fatty acyl activating enzyme in the plastid

In chapter 111 the plastid was demonstrated to be the major site for the elongation of the
exogenous MCFA’s into the C16 and C18 FA’s by i) measuring the saturated to the
unsaturated fatty acid ratio among the labeled products in the leaf fed with the
radiolabeled MCFA’s, ii) using inhibitors of fatty acid synthesis in the plastid and in the
ER, iii) testing the ability of isolated chloroplast to elongate the exogenous MCFA and
iv) using a mutant plant impaired in an acyl activating gene predicted to be targeted to the
plastid. The fatty acid synthesis pathway in the mutant plant was intact, as demonstrated

by using acetate as substrate for fatty acid synthesis, but the mutant plant was unable to

234

elongate the exogenous short- and medium- chain fatty acids. The gene (AAE15)
disrupted in the mutant has sequence similarity to other known long-chain acyl-CoA
synthetases and in fact was shown previously by another group to have no LACS activity
in the heterologous expression systems. Measurement of the acyl-ACP and the acyl-CoA
intermediate levels raises the possibility that the AAE15 gene product is an acyl-ACP
synthetase (AAS), an enzyme that catalyzes the direct esterification of the free fatty acid
to an acyl carrier protein (ACP). The acyl-ACP synthetases (EC 6.2.1.20) have been
isolated from bacteria and yeast (Rock and Cronan Jr, 1981; Fice et al., 1993; J ackowski
et al., 1994; Gangar et al., 2001) but have not yet been demonstrated in plants. In my
most recent assays using isolated pea chloroplast homogenates to test for in vitro AAS
activity, an exogenous ACP-dependent acyl-ACP formation was observed (data not
presented) further supporting the existence of AAS activity in the plastid. Similar assays
should be carried out with chloroplasts isolated from both the wild type and the aae15
knock-out Arabidopsis to relate the observed in vitro AAS activity with the AAE15
enzyme. More directly the recombinant AAE15, expressed/purified heterologously in E.
coli and in the yeast, should be tested for its ability to catalyze the in vitro AAS reaction.
If the heterologous system gives positive results then more detailed enzyme

characterization, including the substrate specificity test, should be followed.

The focus of future experiments on this topic should be to reveal the in vivo function of
AAE15. The aae15 knock-out plant is not visibly distinguishable from the wild type
plant under normal grth conditions. However more detailed characterization of this

mutant should be carried out both under normal conditions and stress conditions. Nine

235

out of 11 EST’s of AAE15 are derived from libraries that were constructed after
dehydration and the cold treatments. The gene expression pattern of AAE15 in the wild
type plant and the phenotypic analyses of the aae15 plants in these conditions may give
us clues to the in vivo function. In addition the tissue specific expression pattern of
AAE15 should guide us where to focus in these analyses. One of the hypothetical
functions of AAE15 proposed in the discussion of chapter HI was involvement in
membrane editing and/or a proof reading function of de novo FAS. It is possible that
some of the growing chain of fatty acids during FAS get released prematurely. In
addition there are many uncharacterized lipase gene homologs in the Arabidopsis genome
(Beisson et al., 2003) which could potentially be involved in the release of free fatty
acids. It is possible that some of those lipases may become active under certain
occasions, i.e. abiotic stresses, mechanical wounding, herbivor attacks, senescence, etc. to
release free fatty acids. Perhaps plant cells use AAE15 in a mechanism to
recapture/recycle those released free fatty acid into the regular glycerol lipid synthesis.
To test this hypothesis the fatty acid compositional changes in the aae15- ko plants in
various conditions should be analyzed by GC and/or GC-MS. In addition it would be
interesting to examine whether there is any change of the AAE15 expression in the
medium-chain acyl-ACP thioesterase-overexpressing plants which releases large amount

of medium-chain fatty acids prematurely (Eccleston and Ohlrogge, 1998)

Perhaps the acyl-ACP formation and the elongation of exogenous free fatty acid by the

isolated chloroplasts can be used to conduct in vitro kinetic analyses to study the import

process. The utilization of the exogenous fatty acids was inversely proportional to the

236

fatty acid chain length and thus the long-chain fatty acids were poor substrates for the
elongation by isolated chloroplasts. This was not due to substrate depletion by prior
usage in the acyltransferase reactions before reaching the stroma, since no exogenous
coenzyme A was provided in the assay and the majority of the long-chain fatty acid
substrates stayed in free form at the end of the assay. Therefore possibly the substrate
specificity of the pea/spinach homolog of AAE15 for shorter fatty acids determines the
relative ability of fatty acids to be elongated. However if it turns out that the AAE15 gene
product does have activity towards long-chain fatty acids such as palmitate then we could
hypothesize that the import of the long-chain fatty acid was the problem. The free
diffusion of the short- and the medium-chain fatty acids across the membrane seems to be
very fast and thus may not require special transport system but the free diffusion rate
(flip-flop) of the long-chain fatty acids is still questionable (as reviewed in chapter I).
The observed inefficient elongation of the long-chain fatty acid could also be interpreted
as the long-chain fatty acid export system not being functional in the reverse direction,

that is towards the stroma.

More distantly related projects from the plastid envelope proteomics

The work described in chapter IV is not only directed towards the search for lipid
transporters but also is aimed for a larger interest in plastid envelope proteins. Plastids
draw the attention of plant biologists in large part because of their defining roles in
establishing the character of the plant cell. Besides the known important roles of the

envelope in various metabolisms, the envelope proteins may hold one key to the

237

understanding of coordinated control between the plastid and the rest of the cell. The
plant research community is moving toward the goal to assign function to each gene
products and in order to facilitate the accomplishment of this goal, global analyses of
whole genomes are encouraged. Bioinforrnatics, genome wide expression profiling,
proteomics, and reverse genetics are examples of such approach. Assigning the
subcellular localization is one of the important steps towards the understanding of the
function of any protein (Somerville and Dangl, 2000). Therefore the work of predicting
the plastid envelope proteins has a broader perspective for current and future directions of

plant science.

The AtPEM database can be used as a template to expand into a more comprehensive
plastid envelope functional proteome database by incorporating rapidly accumulating
information such as experimentally confirmed envelope proteins, proteins identified by
mass spectroscopy based proteomics approach, T-DNA insertion mutant phenotypes, etc.
Also the database can be used as a model to create similar databases for the agriculturally
important crops that are being sequenced such as rice and canola. In the future plastid
envelope specific gene chips can be designed to study the global expression of envelope

proteins.

The selected candidates in the AtPEM database can also be used in an alternative
approach to complement the ‘proteomics’ approach to identify novel envelope membrane
proteins. The candidates can be analyzed for their targeting to the chloroplasts by in vitro

reconstitution of import into the chloroplast and their partitioning to the envelope

238

membranes by fractionation studies. This approach can be especially powerful in
identifying highly-hydrophobic and low-abundance proteins. Above all it can be used to
circumvent the difficulties of preparing ‘pure’ plastid envelope membranes from various
nonphotosynthetic tissues. For example, it is challenging to isolate plastids from
Arabidopsis developing seeds and to prepare sufficient plastid envelope membrane
samples for a proteomics project (besides the issue of purity). In a work not described in
this thesis, I have selected 89 putative developing seed-plastid envelope proteins among
the AtPEM by selecting those with EST’s from the developing seed library. Among them
were proteins with EST’s exclusively from the developing seeds. Some proteins showed
more than 2 fold changes during the seed filling according to the deve10ping seed
microarray data (these proteins may have important fianction in the seed filling). Also
some of them had only 1 or 2 total EST’s suggesting that they are expressed at very low
level thus will be difficult to be identified by standard proteomics approach. Gene
specific primers for 66 candidates were designed to amplify the cDNA templates by the
reverse transcriptase polymerase chain reactions (RT-PCR). The primers were specially
designed to contain necessary sequences for the direct use of the amplified cDNA in in
vitro transcription/translation reactions and subsequently for the plastid targeting
experiments. Although time did not allow these experiments to be completed, the
approach is clear-cut for delving deeper into characterization of plastid envelope proteins

that are specific to seeds.

Future perspectives on fatty acid transport research in plant

239

Membranes compartmentalize different biosynthetic pathways and the proteins that
transport various metabolites across them are potentially important regulatory sites
(Flugge, 2000). This might be true for fatty acid synthesis as well. Long-chain acyl
CoA’s had inhibitory effects on plastidial glucose-6-phosphate transporter and on the
adenylated transporter, reducing carbon flux from either glucose-6-phosphate or acetate
into long-chain fatty acids in plastids isolated from oilseed rape embryos and from pea
roots (Fox et al., 2000; 2001). These inhibitory effects were reversed by the addition of
proteins that bind long-chain acyl CoA’s implying a mechanism for the regulation of
fatty acid synthesis by the ratio between the acyl-CoA binding proteins and acyl-CoA’s.
Feedback inhibition of fatty acid synthesis was also observed in tobacco suspension cells
and the data were interpreted as acetyl-CoA carboxylase being central in the feedback
regulation although the exact mechanism is still obscure (Shintani and Ohrogge, 1995).
BSA had a positive effect on the overall acetate incorporation rate into the fatty acids by
isolated chloroplasts (data not shown). This stimulatory effect of BSA was also observed
by Fox et al. (2000) but they explained it through BSA acting as an acyl-CoA binding
protein to remove the inhibitory effect of long-chain acyl CoA’s on the adenylate
transporter (Fox et al., 2000). However the stimulatory effect of BSA was observed even
in the absence of coenzyme A or/and ATP (unpublished data). Perhaps the BSA acted
positively to the overall acetate incorporation rate by sequestering unesterified fatty acids
which posed inhibitory effects on the FAS. These observations suggest that the fatty acid
export system may potentially be an important regulatory point. The regulatory aspect of
the fatty acid transport is connected with the possibility of genetic engineering and thus

warrents future attentions and further investigations.

240

Further study about the molecular mechanism of fatty acid transport also calls for
increased knowledge about plastid structure in general. Although much is known about
the plastid double membrane envelope, the molecular resolution of the fine structure
remains unresolved. For example, there appear to be no direct connections between the
inner and the outer membrane of the plastid envelope although some “contact points”
have been observed (Douce and Joyard, 1990). These contact points are mainly thought
to be involved in peptide import from the cytosol but it is not known whether these kind
of structures are also responsible for the transfer of fatty acid or other molecules between
the two membranes. A number of proteins that form the plastid division machineries are
arranged concertedly in both the inner and the outer membrane envelope during plastid
division. The complexity of this machinery which is beginning to be understood
(reviewed in Osteryoung and Nunnari, 2003), in a way predicts other undiscovered
mechanistic organizations of the envelope. Another example of such specialized
structures on the envelope is the cytoskeletal fiamework that allows the photorelocation
movement of the plastids. Plastids are not static but able to migrate or be arranged into
specific orientations in response to the environmental changes, namely, the availability of
lights of different wavelengths (reviewed in Wada et al., 2003). The plastid envelope
seems to be more dynamic than it has been considered to be in the past. Green
fluorescent protein (GF P) imaging techniques have revealed a striking image of long thin
tube like structures (called stromules) emanating from the plastid surface (Kohler et al.,
1997; Kohler and Hanson, 2000). The stromules can stretch into distant areas in the cells

sometimes interconnecting plastids. The stroma localized proteins were observed to flow

241

through the stromules. Although the function of the stromules is not known they may be
involved in intracellular coordination. The increased surface area of the plastid envelope
caused by the stromule will affect the free fatty acid diffusion kinetics in those tissues
where stromules are observed. Finally, specified zones in the ER that interact with other
organelles such as mitochondria, Golgi and plasma membrane were implicated in the
interorganellor exchange of lipid molecules in animals and yeasts (V oelker et al., 2003).
It is possible that such structure may exist between the plastid and the ER that mediates
exchange of certain lipids. These yet to be discovered aspects of plastid biology are
essential for complete understanding of fatty acid export from the plastid and for the

study of the lipid trafficking in general.

In conclusion, after very little research for decades, lipid transport in plants is now
beginning to be studied. This challenging task may uncover many hidden aspects of the
lipid biochemistry in plants. The continued accumulation of vast shared information
among the science research community will accelerate our search for the lipid
transporters in plants. As has occurred so far in other organisms, combining genetic
(mutant) analyses with the proper biochemical analyses will continue to make major

breakthroughs in this area.

242

LITERATURE CITED

Abumrad N, Harmon C, Ibrahimi A (1998) Membrane transport of long-chain fatty
acids: evidence for a facilitated process. J. Lipid Res. 39: 2309-2318.

Akita M, Nielsen E and Keegstra K (1997) Identification of protein transport
complexes in the chloroplastic envelope membranes via chemical cross-linking. J. Cell
Biol. 136: 983—994.

Beisson F, Koo AJ, Ruuska S, Schwender J, Pollard M, Thelen JJ, Paddock T, Salas
JJ, Savage L, Milcamps A, Mhaske VB, Cho Y and Ohlrogge JB (2003) Arabidopsis
genes involved in acyl lipid metabolism. A 2003 census of the candidates, a study of the
distribution of expressed sequence tags in organs, and a web-based database. Plant

Physiol. 132: 681-697.

Browse J and Somerville CR (1994) Glycerolipids. In Arabidopsis. Meyerowitz EM,
and Somerville CR ed. (Cold Spring Harbor Laboratory Press, New York) pp. 881-912.

Chou ML, Fitzpatrick LM, Tu SL, Budziszewski G, Potter-Lewis S, Akita M, Levin
JZ, Keegstra K, and Li HM (2003) Tic40, a membrane-anchored co-chaperone
homolog in the chloroplast protein translocon. EMBO J. 22: 2970-2980.

Douce R, Joyard J (1990) Biochemistry and function of the plastid envelope. Annu.
Rev. Cell. Biol. 6: 173-216.

Eccleston VS, Ohlrogge JB (1998) Expression of lauroyl-acyl carrier protein
thioesterase in brassica napus seeds induces pathways for both fatty acid oxidation and
biosynthesis and implies a set point for triacylglycerol accumulation. Plant Cell 10: 613-
622.

Fice D, Shen Z and Byers DM (1993) Purification and characterization of fatty acyl-
acyl carrier protein synthetase from Vibrio harveyi. J. Bacteriol. 175: 1865-1870.

Fox SR, Hill LM, Rawsthorne S, Hills MJ (2000) Inhibition of the glucose-6-phosphate
transporter in oilseed rape (Brassica napus L.) plastids by acyl-CoA thioesters reduces
fatty acid synthesis. Biochem J. 352: 525-32.

Fox SR, Rawsthorne S, Hills MJ (2001) Fatty acid synthesis in pea root plastids is
inhibited by the action of long-chain acyl- coenzyme as on metabolite transporters.
Plant Physiol. 126: 1259-1265.

Gangar A, Karande AA and Rajasekharan R (2001) Purification and characterization
of acyl-acyl carrier protein synthetase from oleaginous yeast and its role in triacylglycerol
biosynthesis. Biochem J. 360: 471-479.

243

Flugge UI (2000) Transport in and out of plastids: does the outer envelope membrane
control the flow? Trends Plant Sci. 5: 135-137.

Gomes E, Jakobsen MK, Axelsen KB, Geisler M, Palmgren MG (2000) Chilling
tolerance in Arabidopsis involves ALAl, a member of a new family of putative
aminophospholipid translocases. Plant Cell 12: 2441-2454.

Jackowski S, Jackson PD and Rock CO (1994) Sequence and function of the aas gene
in Escherichia coli. J. Biol. Chem. 269: 2921-2928.

Kohler RH, Cao J, Zipfel WR, Webb WW, Hanson MR (1997) Exchange of protein
molecules through connections between higher plant plastids. Science 276: 2039-2042.

Kohler RH, Hanson MR (2000) Plastid tubules of higher plants are tissue-specific and
developmentally regulated. J. Cell Sci. 113: 81-89.

Koo AJ, Ohlrogge JB (2002) The predicted candidates of Arabidopsis plastid inner
envelope membrane proteins and their expression profiles. Plant Physiol. 130: 823-836.

Koo AJ, Ohlrogge JB, Pollard M (2004) On the export of fatty acids from the
chloroplast. J. Biol. Chem. Feb 5 [Epub. ahead of print]

Osteryoung KW, Nunnari J (2003) The division of endosymbiotic organelles. Science.
302: 1698-1704.

Rock CO and Cronan JE Jr. (1981) Acyl-acyl carrier protein synthetase from
Escherichia coli. Methods Enzymol. 71: 163-168.

Roughan PG, Ohlrogge JB (1996) Evidence that isolated chloroplasts contain an
integrated lipid-synthesizing assembly that channels acetate into long-chain fatty acids.
Plant Physiol. 110: 1239-1247.

Ruetz S and Gros P (1994) Phosphatidylcholine translocase: a physiological role for the
mdr2 gene. Cell 77: 1071—1083.

Schnurr JA, Shockey JM, de Boer GJ, 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.

Shintani DK and Ohlrogge JB (1995) Feedback inhibition of fatty acid synthesis in
tobacco suspension cells. Plant J. 7 :577-587.

Snyder WB, Koller A, Choy AJ, and Subramani S (2000) The peroxin Pexl9p

interacts with multiple, integral membrane proteins at the peroxisomal membrane.
J. Cell Biol. 149: 1171-1178.

244

Somerville C, Dangl J (2000) Plant biology in 2010. Science 290: 2077-2078.
Thelen JJ, Ohlrogge JB (2002) The multisubunit acetyl-CoA carboxylase is strongly
associated with the chloroplast envelope through non-ionic interactions to the

carboxyltransferase subunits. Arch. Biochem. Biophys. 400: 245-257.

Voelker DR (2003) New perspectives on the regulation of intennembrane
glycerophospholipid traffic. J. Lipid Res. 44: 441-449.

Wada M, Kagawa T, Sato Y (2003) Chloroplast movement. Annu. Rev. Plant Physiol.
Plant Mol. Biol. 54: 455-468.

Xu C, Fan J, Riekhof W, Froehlich JE and Benning C (2003) A permease-like protein
involved in ER to thylakoid lipid transfer in Arabidopsis. EMBO J. 22: 2370-2379.

245