CHLOROPLAST LIPID IMPORT AND METABOLISM, AND THEIR EFFECT ON
CHLOROPLAST STABILITY
By
Anna Hurlock

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Biochemistry and Molecular Biology - Doctor of Philosophy
2017

ABSTRACT
CHLOROPLAST LIPID IMPORT AND METABOLISM, AND THEIR EFFECT ON
CHLOROPLAST STABILITY
By
Anna Hurlock
Much is already known about plant lipid metabolism and the basic enzymes required for
lipid assembly. However, lipid exchange between subcellular compartments remains a
challenging area, which this thesis helps to address. Currently, it is known that the chloroplast
synthesizes fatty acids that are subsequently used by lipid biosynthetic pathways in both the
chloroplast and the endoplasmic reticulum. Fatty acid export to the endoplasmic reticulum (ER)
and lipid transport from the ER to the chloroplast are both known to occur, but the participating
membrane proteins and transporters have proven to be difficult to work with due to their
hydrophobic nature. Many in vivo, genetic, and in vitro techniques have been employed to carry
out these studies, but the field has reached a point at which new creative techniques need to be
developed in order to answer the remaining questions regarding the nature of transported lipid
substrates and the mechanisms of intermembrane lipid transfer. In this thesis, both in vivo and
in vitro methods were developed in order to study lipid transport, acyl editing, and other aspects
of lipid metabolism for which our knowledge is incomplete. An in vivo method using a "tag and
track" approach introducing an ER-located fatty acid desaturase into Arabidopsis, which is
distinct from any of the native fatty acid desaturases, led to a subtle modification of the acyl
group composition of phosphatidylcholine. This "tag and track" approach proved useful in
studying lipid trafficking with high sensitivity in different lipid mutant backgrounds affected in lipid
trafficking and assembly. It clearly demonstrated the import of lipid precursors from the ER for
phosphatidylglycerol biosynthesis and extensive acyl editing of chloroplast lipids as described in
Chapter 2.

In vitro methods to study lipid exchange between different membranes using isolated
chloroplasts and envelope membranes from different lipid trafficking mutants with either radioor fluorescently labeled lipids as described in Chapter 3, yielded promising results, further
suggesting candidates for lipid molecular species that might be transported from the ER to the
chloroplast. However, there were many caveats with the interpretations of the results using
these methods that will require further consideration. One example is the differing stability of
isolated chloroplasts from different lipid mutants which affected the interpretation of the results.
On the other hand, the discovery of differences in the stability of isolated chloroplasts for
different lipid mutants, as described in Chapter 4, provides a new avenue to study the specific
physiological effects of altered chloroplast lipid composition.
As a whole, the work presented here not only yielded new information about plant lipid
metabolism, but it also provides a new tool set for further studies of lipid trafficking and lipid
metabolism in general.

ACKNOWLEDGMENTS

I came to Michigan State University in 2011 to pursue my PhD in Biochemistry and
Molecular Biology because my undergraduate advisor, Clint Chapple, had told me MSU is the
best place to study plant biochemistry with the leading scientists in the field and a huge number
of resources. He had told me who some of the best professors to rotate with would be, and
among those people was Christoph Benning.
The first time I met Christoph at a conference as an undergraduate, I found him very
serious and intimidating. I thought there was no way I would fit into his lab, but I went ahead and
did my first rotation with Christoph taking on a challenging rotation project studying
interorganellar lipid transport using radiolabeled lipids. This was my first experience with
radioactivity, and the experiments were lengthy, delicate, and frustrating. Christoph trusted in
my abilities allowing me to work on this project as a beginning student, and for that, I am
thankful. I moved onto my next rotation thinking the lab environment and project in Christoph's
lab were going to be too intense for my permanent lab as I was worn out from my first rotation,
but for some reason, I just could not get that pesky plant lipid transport project out of my head. I
was hooked on plant lipid metabolism, and I missed being a part of the Benning lab and all of
the hard-working people that provided extra support and advice. I joined Christoph's lab not only
because of the project, but also because his mentorship style allows for independence and
fosters an environment where everyone works as equals providing different perspectives to
each person's project. Christoph deserves a special thanks for accepting me into his lab and
supporting me through all of the ups and downs of my project.
I would also like to thank my committee members for providing extra support and
guidance. John Ohlrogge has been a huge help with all of my radiolabeling assays and with
interpreting my data. As a plant lipid expert, he has been a resource of knowledge that is not
easily pulled up in a literature search, and my depth of knowledge has grown more thanks to
v

him. I also appreciate that he agreed to be on my committee despite being so close to
retirement when I started graduate school. Tom Sharkey has provided me with suggestions and
resources for experiments, and he even taught me how to use an oxygen electrode for
measurement of oxygen evolution from chloroplasts which he had built himself. Gregg Howe
has challenged me to think from a different perspective with his comments and questions, and I
also had a great rotation experience in his lab. Finally, I appreciate that Shelagh Fergason-Miller
stepped in to fill a vacancy on my committee so late in my career as a graduate student.

vi

TABLE OF CONTENTS

LIST OF TABLES..................................................................................................................... viii
LIST OF FIGURES .................................................................................................................... ix
KEY TO ABBREVIATIONS ......................................................................................................... x
CHAPTER 1 Introduction: Lipid Trafficking in Plant Cells - a Literature Review........................... 1
Abstract ............................................................................................................................... 2
Introduction .......................................................................................................................... 2
Chloroplast Export ............................................................................................................... 4
ER to Chloroplast Envelope and Inter-Envelope Transport ............................................... 10
Inner Envelope to Thylakoid Transport ............................................................................. 14
Transport to the Vacuole ................................................................................................... 16
Lipid Trafficking Related to Lipid Droplets ......................................................................... 18
Transport to the Cuticle ..................................................................................................... 20
Transport to the Symbiosome ........................................................................................... 22
Sphingolipid Trafficking ..................................................................................................... 23
Concluding Remarks ......................................................................................................... 25
REFERENCES ......................................................................................................................... 26
CHAPTER 2 In Vivo Lipid “Tag and Track” Approach Shows Acyl Editing of Plastid Lipids and
Chloroplast Import of Phosphatidylglycerol Precursors in Arabidopsis thaliana ......................... 41
Abstract ............................................................................................................................. 42
Introduction ........................................................................................................................ 42
Results .............................................................................................................................. 45
Discussion ......................................................................................................................... 55
Materials and Methods ...................................................................................................... 58
REFERENCES ......................................................................................................................... 65
CHAPTER 3 Lipid Composition Affects the Stability of Isolated Chloroplasts ............................ 70
Abstract ............................................................................................................................. 71
Introduction ........................................................................................................................ 71
Results .............................................................................................................................. 73
Discussion ......................................................................................................................... 80
Materials and Methods ...................................................................................................... 84
REFERENCES ......................................................................................................................... 87
CHAPTER 4 Development of In Vitro Lipid Transfer Assays for Determining the Lipid Molecular
Species Transferred from the Endoplasmic Reticulum to the Chloroplast ................................. 90
Abstract ............................................................................................................................. 91
Introduction ........................................................................................................................ 91
Results .............................................................................................................................. 93
Discussion ....................................................................................................................... 102
Materials and Methods .................................................................................................... 105
REFERENCES ....................................................................................................................... 110

vii

CHAPTER 5 Conclusions ....................................................................................................... 114
REFERENCES ....................................................................................................................... 120

viii

LIST OF TABLES

Table 4.1 Summary of Radiolabeled Lipid Transfer Assays. ..................................................... 96

ix

LIST OF FIGURES

Figure 1.1 Generation of the Acyl-CoA Pool for ER Lipid Synthesis in Chlamydomonas
reinhardtii and Arabidopsis thaliana ............................................................................................ 7
Figure 1.2 Subcellular Locations of Lipid Synthesis and Transport Proteins for the Major
Chloroplast Lipids ....................................................................................................................... 9
Figure 2.1 Determination of Δ6D Subcellular Location .............................................................. 46
Figure 2.2 GLA and SDA Content of Each Independent Line .................................................... 47
Figure 2.3 Phenotypes of Transgenic Plants ............................................................................. 49
Figure 2.4 16C to 18C Ratio...................................................................................................... 51
Figure 2.5 Distribution of Δ6 Acyl Groups Among Polar Glycerolipids ....................................... 52
Figure 2.6 Ion Chromatograms for MGDG ................................................................................ 54
Figure 2.7 14C-Acetate Labeling ................................................................................................ 56
Figure 3.1 Measurement of Properties of Isolated Chloroplasts ................................................ 74
Figure 3.2 Polar Lipid Analysis .................................................................................................. 76
Figure 3.3 Acyl Group Profiles for Polar Lipids .......................................................................... 79
Figure 3.4 Lipid Related Ratios and Double Bond Index ........................................................... 81
Figure 4.1 14C Labeled ER Microsome to Intact Chloroplast Lipid Transfer ............................... 94
Figure 4.2 Fluorescent Lipid Transport Assay Model................................................................. 99
Figure 4.3 NBD-Lipid to Pea Envelope Transport.................................................................... 101
Figure 4.4 NBD-Lipid to Arabidopsis Envelope Transport ....................................................... 102

x

KEY TO ABBREVIATIONS

PA

Phosphatidic Acid

PC

Phosphatidylcholine

PG

Phosphatidylglycerol

PI

Phosphatidylinositol

PE

Phosphatidylethanolamine

PS

Phosphatidylserine

MGDG

Monogalactosyldiacylglycerol

DGDG

Digalactosyldiacylglycerol

TGDG

Trigalactosyldiacylglycerol

SQDG

Sulfoquinovosyldiacylglycerol

GLA

γ-Linolenic Acid

SDA

Stearidonic Acid

xi

CHAPTER 1
Introduction: Lipid Trafficking in Plant Cells - a Literature Review1

1

This review was published: Hurlock AK, Roston RL, Wang K, & Benning C (2014) Lipid
trafficking in plant cells. Traffic 15(9):915-932. The sections which were relevant to this review
were decided by me with guidance from Christoph Benning and Rebecca Roston. Rebecca
Roston contributed the sections "Inner Envelope to Thylakoid Transport" and "Transport to the
Vacuole." Kun Wang contributed the sections "Chloroplast Export" and "Lipid Trafficking
Related to Lipid Droplets." I contributed the rest of the sections and all of the figures as well as
performed extensive editing of the other sections. I also added some extra information into the
review since publication in order to cover advances in knowledge after publication.
1

Abstract
Plant cells contain unique organelles such as chloroplasts with an extensive photosynthetic
membrane. In addition, specialized epidermal cells produce an extracellular cuticle composed
primarily of lipids, and storage cells accumulate large amounts of storage lipids. As lipid
assembly is associated only with discrete membranes or organelles, there is a need for
extensive lipid trafficking within plant cells, more so in specialized cells and sometimes also in
response to changing environmental conditions such as phosphate deprivation. Due to the
complexity of plant lipid metabolism and the inherent recalcitrance of membrane lipid
transporters to molecular characterization, the mechanisms of lipid transport within plant cells
are far from fully understood. Recently, several new proteins have been implicated in different
aspects of plant lipid trafficking. While these proteins provide only first insights into limited
aspects of lipid transport phenomena in plant cells, they represent exciting opportunities for
further studies.

Introduction
Plant cells have many membranes that are generally comparable to those in animal cells
including the plasma, mitochondrial, nuclear, and peroxisomal membranes. In addition, plant
cells contain unique membrane-bound compartments such as the chloroplast, vacuole, and
symbiosome (1) and have other unique cellular structures composed of lipids, e.g. the cuticle of
epidermal cells. As the key photosynthetic organelle in plants, one focus of plant lipid research
has been on the chloroplast. It is surrounded by the outer and the inner envelope membranes
and encompasses one of the most extensive membrane systems found in nature, the
photosynthetic membrane organized into thylakoids. This membrane is unique in its lipid
composition

with

the

two

galactolipids,

monogalactosyldiacylglycerol

(MGDG)

and

digalactosyldiacylglycerol (DGDG), being predominant (2, 3). Unlike animal cells which
synthesize fatty acids in the cytoplasm and assemble glycerolipids primarily in the endoplasmic
2

reticulum (ER), golgi apparatus and mitochondria (4), plant cells synthesize fatty acids in the
stroma of the chloroplast and assemble glycerolipids mainly by two pathways, the
prokaryotic/plastid pathway in the chloroplast envelope membranes and the eukaryotic pathway
predominantly in the ER (5). It should be noted that while lipid metabolism occurs in plastids
other than chloroplasts, for example chromoplasts in fruits and flowers or leucoplasts in storage
tissues, most research on lipid trafficking has focused on chloroplasts.
The origin of specific glycerolipids from either the ER or the chloroplast pathways in
plants can be determined due to the specificity of the respective acyltransferases (6, 7). The ER
localized acyltransferase preferentially transfers 18 carbon fatty acids to the sn-2 (carbon 2)
position of the glycerol backbone while the chloroplast localized acyltransferase preferentially
transfers 16 carbon fatty acids to the sn-2 position of the glycerol backbone (6, 7). Because of
this difference in enzyme specificity, it has been well established that plants use both
chloroplast and ER derived lipids as precursors in the assembly of chloroplast-specific
galactolipids (7-9). Different plants utilize each pathway for the synthesis of chloroplast lipids to
different extents (10); Arabidopsis uses each pathway nearly equally while many other plants,
e.g. grasses, nearly exclusively use the ER pathway (8, 9). Based on lipid acyl composition
alone, the unicellular green alga, Chlamydomonas, has been reported to use only the plastid
pathway for the synthesis of chloroplast lipids (11); however, the specificity of the different
acyltransferases involved has not yet been determined and may not be the same as for seed
plants.
Certainly in all seed plants, the presence of ER derived chloroplast lipids requires the
transport of ER lipid precursors to the chloroplast. Lipids, being hydrophobic molecules, must be
trafficked from their place of synthesis to their destination whether that is another membrane or
the extracellular space in the case of the cuticle. Because this is a review focused on plant cells,
lipid trafficking between organelles largely similar to those found in animal cells will not be
covered. This includes trafficking to the plasma membrane, mitochondria, and peroxisome,
3

which are covered in recent reviews on these topics (12-15). There are several possible modes
of lipid movement within cells which will be mentioned throughout such as vesicular movement,
transport by soluble proteins, flipping across bilayer membranes, and movement at membrane
contact sites. Covering the basic mechanistic principles of different modes of lipid movement
goes beyond this review, and the reader is referred for further details to the following
publications (14, 16-19).

Chloroplast Export
Fatty acids (FAs) synthesized in chloroplasts need to be exported to the ER for the biosynthesis
of phospholipids and the major storage lipid, triacylglycerol (TAG), and to provide lipid
precursors for export to the extracellular cuticle. FA synthase is a multi-enzyme complex that
catalyzes a series of condensation and elongation steps to produce C16 and C18 acyl chains
attached to an acyl carrier protein (ACP) (5). A soluble Δ9-desaturase converts 18:0-ACP
(carbons: double bonds) to 18:1-ACP (5). The 16:0, 18:0, and 18:1 acyl-ACPs are hydrolyzed
by two thioesterases designated FATA and FATB, and it is generally assumed that the free FAs
have to be activated to acyl-CoA at some step during export to the ER (5, 20-23). This reaction
is thought to be catalyzed by long-chain acyl-CoA synthases (LACSs) (24, 25). The Arabidopsis
genome contains nine LACS genes, and only LACS9 is known to be localized in the chloroplast
envelope (26, 27). Labeling experiments using

14

C-18:1 or

14

C-16:0 with chloroplasts isolated

from the lacs9 null mutant showed a reduction in measurable LACS activity suggesting LACS9
contributes much of this activity in chloroplasts (26). However, the mutant plant was
indistinguishable from wild type in growth and morphology as well as lipid and FA composition,
which suggests that functionally redundant orthologs of LACS9 exist (26). It is also possible that
the conversion of FA to acyl-CoA is not a rate-limiting step in FA export. LACS8 shows high
sequence similarity with LACS9, but it is associated with the ER (28). No changes in seed FA
content or growth phenotype were detected in the double knock-out of lacs8lacs9, but lipids in
4

other tissues have not been analyzed (28). Like LACS8, LACS1 is also ER localized, but it is
primarily involved in providing FAs for cuticle lipid synthesis (28, 29). However, while none of
the single mutants have seed FA phenotypes, both the double lacs1lacs9 and triple
lacs1lacs8lacs9 knock-out mutants have similarly decreased seed FAs suggesting LACS1 and
LACS9 may have overlapping functions (28). These LACS proteins warrant further study in nonseed tissues, although it seems that one or more chloroplast-associated LACS involved in FA
export remain to be identified.
Other than the LACS proteins, another protein has been implicated in FA export from
chloroplasts, Fatty Acid Export 1 (FAX1) (30). FAX1 is an inner envelope protein, and mutants
have decreased ER derived lipid species and increased plastid derived lipids (30). This mutant
phenotype combined with the fact that FAX1 can complement fatty acid transport in yeast
mutants makes FAX1 the protein with the most concrete evidence for being directly involved in
FA export from chloroplasts (30); however, due to its inner envelope location, it is likely other
proteins are still involved in chloroplast FA export.
ATP-binding cassette (ABC) transporter proteins can transfer lipophilic compounds
through membranes, and it is possible that this class of transporters is involved in acyl export
from the chloroplast or in the import of acyl derivatives into the ER (31). For example when
AtABCA9, an ER-localized ABC transporter in Arabidopsis, was knocked-out (32), the seed
TAG amount was decreased while overexpression of AtABCA9 increased the TAG amount (32).
Labeling experiments with

14

C-acetate,

14

C-18:1, or

14

C-18:1-CoA suggested that AtABCA9

plays a role in supplying free FAs or acyl-CoA substrates for TAG biosynthesis in the ER (32).
Although the specific substrate for AtABCA9 is unknown, analysis of the mutant indirectly
suggested that acyl trafficking to the ER might be limiting TAG biosynthesis in this mutant (32).
Another ABC transporter, AtABCD2/AtPMP1, which is plastid localized, was hypothesized to be
involved in chloroplast FA export based on its similarity to AtABCD1, which is a peroxisomal
localized ABC transporter that is required for FA import into peroxisomes (31, 33, 34). However,
5

the function of AtABCD2 has not been experimentally verified, and the transport of hormone
precursors, carotenoids, or tocopherols by this protein have also been proposed (31, 33).
During a global phenotyping screen to determine functions of chloroplast proteins, no strong
phenotypes were detected in two characterized T-DNA insertion lines fully disrupted in
AtABCD2 (35) making it unlikely that this protein is responsible for major fluxes of acyl groups
from the plastid to the ER.
Acyl-CoA-binding proteins (ACBPs) have also been suggested to play a role in FA
export from plastids (36, 37). Studies of yeast and mammalian ACBPs provide evidence for
ACBPs acting as acyl-CoA transporters (38, 39). Many of the ACBPs in plants have been
implicated in stress response, but others such as BnACBP, AtACBP4, and AtACBP5 may be
involved in transfer of acyl-CoAs from the chloroplast to the ER (37, 40, 41). An ACBP from
Brassica napus, BnACBP, enhances acyl exchange between phosphatidylcholine (PC) and the
acyl-CoA pool, particularly with 18:1-CoA (41). Two of the 6 Arabidopsis ACBPs, ACBP4 and
ACBP5, are localized to the cytosol and are of particular interest as they have high binding
affinity for 18:1 acyl-CoA in vitro, can also bind PC, and have increased expression in response
to light (36, 40, 42, 43). Analyses of T-DNA lines of ACBP5 have yet to be published, but lipid
analysis of the T-DNA acbp4 mutant showed reduced levels of membrane lipids, including
MGDG, DGDG, PC, phosphatidylethanolamine (PE), and phosphatidylinositol (PI) in leaves
(42). The decreased ER phospholipid levels and perhaps the galactolipids derived from ER
precursors suggest a possible role of ACBP4 in the transport of acyl-CoA from plastids to the
ER, but further FA analysis of the MGDG and DGDG would need to be done in this mutant to
support this hypothesis (36, 42, 44). Also, there is some evidence supporting the role of ACBP4
in stress response instead as it has been shown to co-express under stress conditions, colocalize with, and bind to AtEBP, an ethylene-responsive element binding protein (45).
Direct membrane contact sites between the ER and the chloroplast outer envelope could
provide possible conduits for lipid transfer (46). Rapid labeling studies using
6

14

C-acetate or

14

C-

glycerol indicated that FAs de novo synthesized in the chloroplast are first incorporated into PC,
which is present in the ER as well as in the outer envelope membrane of chloroplasts (47),
before entering the cytosolic acyl–CoA pool. Lyso-PC acyltransferase activity likely involved in
this acyl exchange process is associated with isolated chloroplasts (48, 49). Thus, PC might
serve as the first intermediate for FAs exported from the chloroplast, and extensive acylexchange through chloroplast envelope PC into cytosolic acyl-CoAs might provide the acyl
groups needed for the assembly of phospholipids at the ER (Figure 1.1). Phosphatidylglycerol
(PG) may also be a source of acyl groups for synthesis of other lipids. A chloroplast specific
lipase, Plastid Lipase 1 (PLIP1), has been shown to hydrolyze acyl groups from PG, which are
then shuttled through PC before being used for seed oil biosynthesis (50). While this process is

Figure 1.1 Generation of the Acyl-CoA Pool for ER Lipid Synthesis in Chlamydomonas reinhardtii and
Arabidopsis thaliana. Hypotheses for how FAs enter the acyl-CoA pool are presented. A long standing
hypothesis has been that FAs are released from acyl-ACP and converted to acyl-CoAs at the oEM as shown for
both organisms. In Arabidopsis, a high rate of acyl exchange between PC and the acyl-CoA pool has been
observed. As PC is also located in the oEM as well as in the ER, it is likely acyl-exchange may occur at both the
ER and the oEM. In Chlamydomonas, which lacks PC, there is evidence of newly synthesized FAs being
incorporated into MGDG first, and then hydrolyzed from MGDG by the enzyme PGD1 and used for TAG synthesis
in the ER. The models presented are simplified and may not be exclusionary with regard to plants, as a possible
role of MGDG in acyl exchange and FA export still needs to be investigated. It should be noted that TAG
biosynthesis can also occur at the chloroplast envelope in both organisms but is not shown here. ER, endoplasmic
reticulum; oEM, outer envelope membrane; iEM, inner envelope membrane; FAS, fatty acid synthase; FA, fatty
acid; G3P, glycerol 3-phosphate; acyl-CoA, acyl-coenzyme A; PL, phospholipid; TAG, triacylglycerol; L-PA, lysophosphatidic acid; PA, phosphatidic acid; DAG, diacylglycerol; L-MGDG, lyso-monogalactosyldiacylglycerol;
MGDG, monogalactosyldiacylglycerol; L-PC, lyso-phosphatidylcholine; PC, phosphatidylcholine.

7

most active in the seeds, it does not mean a similar process could not be occurring in leaf tissue
by way of one of the many other uncharacterized lipases.
A similar mechanism of using a membrane lipid intermediate for FA export from the
chloroplast to the ER has been proposed for Chlamydomonas which lacks PC (51) (Figure 1.1).
The MGDG lipase, Plastid Galactolipid Degradation 1 (PGD1), is involved in the
production of TAG from de novo synthesized FAs exported from the chloroplast when N is
limited (51). PGD1 specifically releases the newly incorporated FA from the sn-2 position of the
glycerol backbone of MGDG before it gets further desaturated, thereby making it available for
TAG biosynthesis either at the ER or the outer chloroplast envelopes (51). A lyso-MGDG
acyltransferase has not yet been identified, but together with PGD1, it could provide a metabolic
cycle shuttling de novo synthesized FAs from the inside of the plastid to the outer leaflet of the
outer chloroplast envelope. PGD1 is thought to have access to its substrate at the outer
envelope (Figure 1.2) as it does not have a predicted plastid transit peptide, but its exact
location has not yet been verified. This hypothesis implies that acyl groups attached to lipids - in
this case MGDG - are passed between and through the envelope membranes. In thylakoid lipid
biosynthesis, it is also necessary to transfer galactolipids from their site of synthesis on the
intermembrane face of the inner and the outer membrane face of the outer envelope back to the
thylakoids inside plastids. Whether this process is related to FA export and how galactolipids
move between the plastid membranes remains to be defined.
In plants, lipid metabolism and trafficking can be greatly altered under nutrient stress.
For example, under normal growth conditions, DGDG is exclusively localized in chloroplasts
(52). However, when the availability of phosphate is limited, DGDG biosynthesis is elevated
partially due to the transcriptional induction of MGDG synthases, MGD2 and 3, and DGDG
synthase, DGD2 (53, 54). To maintain the proper function of membranes, DGDG is transported
from the chloroplast envelopes to the extraplastidic membranes where DGDG replaces
phospholipids (52, 55-57). While the transporters which are responsible for DGDG export from
8

Figure 1.2 Subcellular Locations of Lipid Synthesis and Transport Proteins for the Major Chloroplast
Lipids. The subcellular locations of the chloroplast envelope associated enzymes for final assembly of the major
chloroplast lipids are depicted along with known lipid transport proteins. Presumed movement of lipid precursors
for the synthesis of major thylakoid lipids between and through the envelope membranes is indicated as
necessitated by the location of the respective enzymes. All proteins represented have been studied in
Arabidopsis except for PGD1 which was discovered in Chlamydomonas but has orthologs in Arabidopsis. ER,
endoplasmic reticulum; oEM, outer envelope membrane; iEM, inner envelope membrane; G3P, glycerol 3phosphate; ACP, acyl carrier protein; L-PA, lyso-phosphatidic acid; PA, phosphatidic acid; DAG, diacylglycerol;
SQDG, sulfoquinovosyldiacylglycerol; L-MGDG, lyso-monogalactosyldiacylglycerol; MGDG, monogalactosyldiacylglycerol; DGDG, digalactosyldiacylglycerol

the plastid have not yet been reported, several putative lipid binding or transport proteins with
unknown function are predicted to be present in Arabidopsis chloroplasts (33). Interestingly,
when the bacterial lipid, glucosylgalactosyldiacylglycerol, was introduced into Arabidopsis
DGDG-deficient mutants, dgd1 and dgd1dgd2, the accumulation of this lipid in extraplastidic
membranes was observed, which indicated the transport mechanism does not discriminate the
glucose and galactose in the lipid head group in vivo (58).
Characterization of the Arabidopsis fad3 mutant, provides another example for the
possible export of desaturated lipids from chloroplast (59). The fad3 mutant is deficient in the
ER 18:2 desaturase activity, but in

14

C-acetate labeling experiments, the fad3 mutant showed

delayed and reduced accumulation of 18:3 in PC and PE, suggesting that plastid desaturated
18:3 FAs can find their way into ER lipids (59). However, in this study the possibility that another
18:2 desaturase exists in the ER or that the fad3 point mutant allele is leaky could not be ruled
out. FAD7 and FAD8 are the two chloroplast localized isoforms of fad3 (60, 61). When
9

comparing the lipid phenotype of two triple mutants fad3-2 fad7-2 fad8 and fad3-2 fad7-1 fad8,
the latter of which carries a leaky fad7 allele, higher 18:3 amounts in PC and PE were found in
the fad7 leaky triple mutant (62). This indicates that ER phospholipids containing18:3 acyl
chains are dependent on the precursors from the chloroplast, supporting the hypothesis that
18:3 FAs can transfer from the chloroplast to the ER. However, the mechanism of this export
and the lipid molecules transported remain to be determined.

ER to Chloroplast Envelope and Inter-Envelope Transport
The chloroplast has the ability to synthesize the major lipid components of its membranes, the
galactolipids, MGDG and DGDG, and the two anionic lipids, sulfoquinovosyldiacylglycerol
(SQDG) and PG (5, 53, 63). Positional analysis of acyl chains of the major plastid lipids (see
Introduction of this chapter) suggests that in all plants a fraction of these lipids is derived from
ER-synthesized precursors, although this proportion varies between different plant species (9).
In addition, PC, which the chloroplast cannot assemble, is a significant component of the
chloroplast outer envelope membrane (64, 65). Thus, the presence of ER-derived thylakoid
lipids and PC in the chloroplast requires the transport of ER lipids to this organelle. Because of
the localization of the enzymes required for the synthesis of the major chloroplast lipids in the
two envelope membranes, both the ER-derived and chloroplast-derived lipids used for the
synthesis of the major chloroplast lipids must be transported between the envelopes (Figure
1.2). ER-derived lipids must first be transported to the outer envelope and then to the outer
leaflet of the inner envelope where the predominant MGDG synthase, MGD1, is localized or to
the inner leaflet of the inner envelope where the SQDG synthase, SQD2, is localized (66-68).
Lipids originating in the chloroplast are first synthesized by the transfer of acyl-chains to
glycerol-3-phosphate to make phosphatidic acid (PA) at the inner leaflet of the inner envelope
where PA can then be further metabolized to PG or SQDG (5). Like the ER lipids, the
chloroplast derived lipid precursors must be transported to the outer leaflet of the inner envelope
10

for MGDG synthesis, and then MGDG made from both ER and plastid precursors must be
transported to the outer leaflet of the outer envelope where the DGDG synthase, DGD1, is
localized (5, 54, 69). It should also be noted that MGD2 and 3 are not considered here because,
as mentioned previously, they are induced under nutrient deprivation (63). The exact
mechanisms of transport of lipid precursors and lipids from the ER to the chloroplast and
between envelopes are still not fully understood, but recent progress has led to a refined
hypothesis.
A first step towards gaining a comprehensive understanding of ER to chloroplast lipid
transport has been the discovery and characterization of the TriGalactosylDiacylglycerol (TGD)
proteins of Arabidopsis. The TGDs were identified in a mutant screen for suppressors of dgd1, a
mutant deficient in DGDG synthesis, and the tgd mutants were named for their production of
oligogalactolipids such as trigalactosyldiacylglycerol (70). The tgd mutants present a range of
other phenotypes including an increase in the 16-to-18 carbon ratio in total FA composition
consistent with a decrease in ER-derived membrane lipids (70-73). Further analysis of
chloroplast-specific lipids showed that the total fatty acid change in the 16-to-18 carbon ratio is
due to decreases in galactolipids derived from ER precursors (70-75). Other tgd mutant
phenotypes include reductions in growth, fertility, germination, and formation of photosynthetic
membranes, as well as lethality of null-alleles of specific TGD genes (70-75). Taken together,
these phenotypes implicate the TGDs in ER to chloroplast transport and demonstrate the
importance of ER-derived chloroplast lipids to the formation of photosynthetic membranes and
the overall health and viability of plants.
Five TGDs have been identified and extensively studied. TGD1, TGD2, and TGD3 form
an ABC transporter complex in the inner envelope membrane, while TGD4 forms a homodimer
in the outer envelope membrane, and TGD5 is a small protein which can associate with the
other TGD proteins (76-80). TGD1 is predicted to be the permease of the ABC transporter and
is present in 2 copies per complex (70, 78). TGD3 is a confirmed ATPase and is also present in
11

2 copies per complex (72, 78). TGD2 is the substrate binding domain of the complex and is
known to bind PA (73, 78, 79, 81). TGD2 makes the TGD1,2,3 complex unique among ABC
transporters as it is present in 8-12 copies per complex, while most substrate binding domains
of bacterial ABC transporters are present in 1-2 copies if present at all (31, 78). TGD4 is a
predicted β-barrel protein which, like TGD2, binds PA (76, 77). TGD5 was shown to interact with
the other TGD proteins through co-immunoprecipitation when heterologously expressed in
tobacco, but it is still unclear how TGD5 mechanistically works with the other TGD proteins (80).
TGD4 and the TGD1,2,3 complex do not form any stable interactions (78), so if TGD5 helps to
connect TGD4 and TGD1,2,3, it may be a transient interaction.
Although two TGD proteins bind PA, there is no direct evidence demonstrating PA as the
transported lipid species of either the TGD4 or the TGD1,2,3 complexes. PA seems to be a
reasonable substrate for transport as it can support multiple membrane conformations and is a
non-bilayer forming lipid, so it may require less energy to be removed from membranes (82).
Several other candidates for the transported lipid species have been suggested including
diacylglycerol (DAG), PC, and lyso-PC (3). DAG is considered because, like PA, it is a nonbilayer forming lipid, but it also has the unique ability to flip between leaflets of a bilayer at
biologically relevant rates (83). PC and lyso-PC have both been proposed due to the inability of
chloroplasts to produce PC, but this hypothesis would only be relevant for TGD4 as PC is only
present in the outer envelope membrane (3, 64, 65).
Despite extensive forward genetic screening approaches that mostly yielded a sizable
number of additional alleles of the TGD proteins, the TGD proteins are the only proteins thus far
directly implicated in ER to chloroplast lipid transport (79). It is hypothesized that they work by
forming a lipid conduit through the outer and inner envelope membranes facilitating transport
(84). TGD2 has the ability to disrupt membranes in vitro which provides the basis for the
hypothesis that the TGD1,2,3 complex works by removing lipids from the outer envelope
membrane and inserting them into the inner envelope membrane, although there is no direct
12

evidence that the TGD1,2,3 complex contacts the outer envelope membrane in vivo (79). While
TGD4 is associated with the outer envelope membrane, a portion cofractionates with ER
markers and hence TGD4 may contact the ER membrane (76). The possibility that the TGD
complexes act as flippases moving lipids across leaflets within the same bilayer must also be
considered. Although DAG can flip between bilayers at a reasonable rate, flippases may be
required to facilitate this process to keep up with flux demands. In fact, DAG pools at different
chloroplast membrane leaflets can be distinguished in vivo (66) suggesting that DAG does not
rapidly equilibrate between the different leaflets.
Related hypotheses discussed in the field for both ER to chloroplast envelope and interenvelope transport involve PLastid Associated Membranes (PLAMs) and hemifusion. PLAMs
involve stable membrane contact between the ER and the chloroplast and have been observed
by microscopy and optical manipulation (85-89). While information on PLAMs is still scarce, they
are likely similar to Mitochondria Associated Membranes (MAMs), which have been extensively
studied and reviewed (90, 91). MAMs are known to be involved in lipid transport between the
ER and mitochondria (92, 93), so it is hypothesized that PLAMs may be involved in lipid
transport between the ER and chloroplasts. Transorganellar complementation studies suggest
there can be considerable transport of diverse hydrophobic compounds, e.g. tocopherol
precursors, between the ER and chloroplast envelopes (94). To accommodate the perceived
lack of specific transporters for the number of molecules that are likely exchanged between the
chloroplast and the ER, it has been suggested that intermembrane hemifusions of the outer
leaflets of the ER and outer envelope membranes and of the inner leaflet of the outer envelope
with the outer leaflet of the inner envelope membrane provide a mechanism for lipohilic
compound exchange between membranes without transporters (94, 95). However, hemifusion
of membranes has yet to be directly observed in plant cells, although hemifused membranes
have been shown to form in vivo during specific cellular events in other organisms (i.e. vesicle
fusion, egg fertilization) (96-98). A confirmation of the hemifusion hypothesis will require
13

identifying proteins that one might predict are involved in initiating and stabilizing membrane
hemifusions and in preserving the unique protein and lipid compositions of the membranes
involved in hemifusion. A protein complex which tethers ER to mitochondria has been identified
in yeast (99), and similar complexes must be involved if contact sites between the ER and the
plastid facilitate the exchange of membrane lipids and other membrane soluble small molecules.

Inner Envelope to Thylakoid Transport
As proplastids from meristematic cells differentiate into chloroplasts, thylakoid membranes
proliferate greatly. When mature, thylakoids are the major membrane system of the leaf cell and
form the boundary of a continuous aqueous compartment called the thylakoid lumen. The lipids
required to make the thylakoids are themselves made in the chloroplast envelopes (see above,
84). Accordingly, the thylakoids are assumed to derive directly from the chloroplast inner
envelope membrane, an idea which is further supported by the similarity of their respective lipid
compositions (100).
Three non-exclusive hypotheses for the mechanism of thylakoid expansion and
maintenance have been proposed. The first is that thylakoid lamellae derive from an
invagination of the inner envelope membrane which expands until it fuses with other such
invaginations and is then separated from the inner envelope. Observations of young
chloroplasts by transmission electron microscopy clearly show long inner envelope
invaginations protruding into the stroma (101). This hypothesis is favored for early development
of thylakoids but has not been observed in mature chloroplasts and may be unlikely as a
mechanism for mature thylakoid maintenance because ATP generation from photosynthesis
requires a stable, low pH in the lumen. The second hypothesis states that a vesicle transport
system exists between the inner envelope and thylakoid membranes. Vesicle transport is
supported by the observation of stromal vesicles in multiple plant species under cool conditions
which slow vesicle fusion (102), by assay (103-105), and by recent molecular biology studies
14

described in detail below. Finally, thylakoid membranes may be maintained by non-vesicular
lipid transport. MGDG was observed to have high rates of transport compared to other thylakoid
lipids, in spite of the fact that the steady-state composition of both the inner envelope and
thylakoid membranes is similar (106). This may indicate that it is moved to the thylakoids
preferentially through a separate soluble lipid transport system.
To date, few gene products have been described as direct effectors of thylakoid
biogenesis. For the most part, these have been identified because the respective mutants lack
normal thylakoid membranes, though confirmation that their direct effect is on the membrane
itself has been tenuous. Examples include THylakoid Formation 1 (THF1), which was later
found to have a direct effect on photosystem II (107), and Plastid Fusion and/or Translocation
Factor (PFTF), which was shown to be a FtsH protease (108). It seems likely in these cases
that the proteins in question are not also directly involved in thylakoid formation. Candidates
more likely to have direct roles have also been identified. First, a protein named VesicleInducing Protein in Plastids 1 (VIPP1) was shown to be required to form cold-induced vesicles
at the inner envelope membrane (109) and was localized to both the inner envelope and
thylakoid membranes (110). Direct evidence that VIPP1 induces vesicles is still lacking,
(reviewed in 111), and there is conflicting evidence from its homolog in Synechocystis which
suggests that aberrant thylakoid structure in knockouts may be due to a direct effect on
photosynthetic ability (112). Likewise, Snow-White Leaf1 (SWL1) has been described as
involved in thylakoid biogenesis in rice (113). It was shown to be important for development of
chloroplasts from proplastids, but not for amyloplasts or etioplasts, indicating a specific role in
thylakoid biogenesis; however, its direct function has not yet been shown. Alternative
approaches to forward genetic screens include bioinformatics, which have identified multiple
chloroplast proteins similar to the proteins required for COPII vesicle trafficking (114, 115). Each
of these studies has fallen short of identifying a chloroplast homolog of every protein required
for cytosolic COPII trafficking, though the chloroplast vesicle system may well be a simpler one.
15

There has been some confirmation of the COPII hypothesis using reverse genetic tools; these
studies showed that cpSAR1A is in the chloroplast and essential for viability (116) and identified
cpRABA5e as a chloroplast located Rab GTPase which may be involved in trafficking of stromal
vesicles (117). Both studies located the respective proteins as distributed between the stroma
and thylakoid membranes and suggest they are found on stromal vesicles. At this point,
suggesting the mechanism of vesicle transport would be premature, though it is worth pointing
out that no proteins have been suggested for the alternative thylakoid biogenesis hypotheses.
Thus, the vesicle hypothesis is the most substantiated, though all require further verification.
Once lipids have arrived at the thylakoid membrane, they are maintained asymmetrically
between the inner and outer leaflets. It has been shown both through multiple methods and in
several species that MGDG and PG are more populous in the stromal leaflet, while DGDG and
SQDG are more prominent in the lumenal leaflet of the membrane (118, 119). The asymmetry is
present at the earliest testable stage of thylakoid development (119), raising the hypothesis that
asymmetry precedes thylakoid integration of lipids and thus may be a result of passive
maintenance. However, rates of MGDG and DGDG incorporation into thylakoid membranes
were observed to differ considerably from their steady-state levels (106), discrediting the
passive maintenance hypothesis. Currently, the mechanism by which lipid asymmetry is
maintained, like the mechanisms by which thylakoid membranes are built, remains at the
hypothesis stage.

Transport to the Vacuole
In plants, vacuoles have developed some functions dissimilar from vacuoles of most motile
species. In seeds, a specific type of vacuole stores protein for the embryo, while in most other
tissues, a “lytic” vacuole is the largest cellular structure. Lytic vacuole functions include
maintaining turgor pressure, providing pH homeostasis, detoxifying the cytosol, degrading
waste, sequestering ions and participating in multiple signaling pathways (120). The unique
16

membrane composition of the vacuole (121) is known to affect its function. In plants capable of
withstanding high salt stress, vacuolar membrane fatty acids and overall protein levels change
to make a less permeable membrane (122). Similarly, during phosphate stress, the vacuolar
membrane is extensively remodeled, partially through the action of a specific phospholipase D
(PLD2) (123). Finally, the lipid composition of the vacuolar membrane is likely to directly affect
its morphology. It was shown that disruption or overexpression of genes encoding enzymes that
modify the number of phosphates on PI affect both vacuolar shape and vesicle traffic destined
for the vacuole (124).
Lipid transport to support lytic vacuolar membrane composition is likely to occur by
vesicle transport through the endomembrane system (120). Vacuole membrane lipids are
synthesized in the ER, though the diversity of their vacuole delivery methods has only recently
been appreciated. Vacuole development occurs during cell differentiation, when a tubular
“provacuolar” network found in undifferentiated meristematic cells fuses and expands to form
the large central vacuole (125). Recent developments show that vacuole biogenesis begins in a
specific area of the ER (125), shedding doubt on an earlier hypothesis that provacuole
membranes originate from fusions of vesicles from the trans-Golgi network (126). Similarly for
mature vacuoles, we now know that transport of sterols and some proteins occurs
independently of a functioning Golgi (125, 127-129), though the Golgi network and
multivesicular body are essential for delivery of others (reviewed in 130). Maintenance of the
vacuole membrane has not been well studied. It was shown to have only slight membrane
asymmetry in sycamore (131), which could have originated from asymmetry inherent in its
progenitor membranes. Vesicular transport to the vacuole has been studied primarily as a
means to identify targeting methods of proteins, thus whether additional mechanisms of lipid
transport may occur is unknown.

17

Lipid Trafficking Related to Lipid Droplets
Lipid droplets (LDs) are dynamic organelles containing lipid storage compounds, typically TAGs.
LDs are surrounded by a phospholipid monolayer into which diverse proteins are inserted.
Although LDs were initially regarded as inert storage depots, they have drawn extensive
attention since the discovery that disturbances in LD metabolism are at the basis of many
human diseases and that LD formation holds potential for biofuel production in plants. LDs are
observed in nearly all cell types in plants, and they have been shown to play important roles in
stress responses, pathogen resistance, and hormone metabolism, as well as in anther
development (132). Here the focus will be on lipid metabolism/trafficking during LD biogenesis
and other trafficking roles in which LDs may be involved. LD turnover and FA transport to the
peroxisome will not be covered as it is discussed in the peroxisomal review mentioned
previously (13).
Several models have been proposed for LD biogenesis in eukaryotic cells (133). In plant
seeds, the sequestration of TAG during biogenesis of LDs is believed to occur at specialized
domains of the ER similar to proposals for animal LD biogenesis (133-135). Evidence derives
from ultrastructural studies showing an intimate association of the ER and LDs (136, 137). In
addition, the biosynthesis of TAGs in the LD core and of the major LD membrane lipids, PC and
PE, is accomplished by ER associated enzymes (138). Accordingly, it is proposed that during
seed LD biogenesis, specific LD membrane proteins such as DAG AcylTransferase (DGAT) and
LD associated proteins, generally called perilipins, i.e. oleosins in plants, first aggregate in ER
specific domains (132, 135). DGAT is one of the enzymes responsible for the final step of TAG
biosynthesis (138), and oleosin is the most abundant membrane protein found in plant seed LDs
(132). In plants, TAG biosynthesis can occur by the conventional Kennedy Pathway or by acylCoA-independent mechanisms involving Phospholipid : DAG AcylTransferase (PDAT) as
reviewed in (138). Oleosin stabilizes the LDs and also controls the size of the LD by preventing
fusion between LDs (139). Therefore, after oleosins are recruited to the ER domains, they
18

stabilize the ER domains and support the budding process (132, 140). Localized DAG formation
is particularly important to LD biogenesis as shown in yeast and mammalian cells. Yeast
mutants impaired in punctate DAG synthesis, but with no changes in total DAG amounts,
showed a decrease in LD number, while mammalian cells with ER enriched in DAG showed
increased recruitment of perilipins (134, 141, 142). Despite what is known in other eukaryotes,
the mechanism of LD biogenesis in plant non-seed tissues is poorly understood. Non-seed
tissues are known to accumulate LDs in response to certain stresses such as freezing (132,
143). One protein known to provide DAG for LD biogenesis under freezing conditions is
Sensitive to Freezing 2 (SFR2) which removes MGDG from the chloroplast outer envelope by
forming oligogalactolipids and DAG (143). Since oleosins are absent, those ER domains that
are specific for TAG sequestration during LD biogenesis in seeds may not exist. Therefore, an
oleosin-independent model has been proposed (132). In this model, LD biogenesis is thought to
be solely dependent on the coalescence of small LDs forming, e.g., in the chloroplast envelopes
(132), but how these small LDs precisely fuse with other LDs instead of other organelles and
how LD size is controlled remain elusive.
Another class of perilipins are lipid-droplet associated proteins (LDAPs) identifed in the
mesocarp of avocado (Persea americana) (144). The protein level of LDAPs correlates well with
the fatty acid level of the TAG core, indicating this LDAP protein may be directly involved in the
accumulation of TAG for LD biogenesis or the regulation of TAG metabolism (144). Similarly, a
protein called major lipid droplet protein (MLDP) may have perilipin function in Chlamydomonas
cells that accumulate TAG following nutrient deprivation (145). When MLDP was knocked down
by RNA interference, increased LD size was observed (145). This supports the hypothesis that
perilipins may have roles in the control of LD size and likely in LD movement and fusion with
other LDs. Recruitment of enzymes for TAG synthesis or degradation to the LD may be another
function of plant lipid droplet specific proteins. For example, introducing a lipid droplet specific
protein from the unicellular algae Nannochloropsis into an Arabidopsis oleosin mutant reverted
19

LD size to be normal but did not rescue biochemical phenotypes (146), perhaps because the
algal LD protein lacked the specificity to recruit Arabidopsis proteins to the LD.
Recent studies have revealed that LDs may be more complex organelles than originally
thought (147, 148). One of the proposed functions for LDs is to transport neutral lipids and
phospholipids between different organelles (148). Proteomic and genomic studies of
mammalian and yeast LDs show that the LD membranes are rich in vesicular traffic related
proteins, such as the Rabs, SNAREs and Arf (147, 149). In plants, COP (coatomer protein)
vesicular traffic proteins have also been found in LD membranes, and it has been suggested
that these proteins may be involved in LD biogenesis (150). In Drosophila and mammalian cells,
cytoskeleton-dependent LD trafficking has been found, and the fusion between LDs is highly
related to microtubule activities (148). Moreover, intimate contacts between LDs and other
organelles have been observed with different microscopy techniques in various cell types, and
the biological relevance of these contacts has been discussed as reviewed in (148). Even
though the mechanism and regulation of this general transport remains unknown, this evidence
suggests the size and location of LDs is able to be controlled and likely to be well regulated. In
addition, it has been shown that LDs can deliver a group of 1,3-β-glucanases to the
plasmodesma channel connecting two plant cells, and it is proposed that this LD-dependent
transport mechanism could be an alternative pathway to transport cargoes to adjacent cells and
even cell-cell signal transduction (151, 152).

Transport to the Cuticle
Plant epidermal cells have a high demand for lipid transport as they synthesize the outer cuticle
at the surface of the plant which serves as a protective layer against the environment. The
cuticle is known to play an important role in resistance to pathogens, and thus, plants impaired
in cuticle formation have been identified not only by visually reduced waxes but also by
increased susceptibility to certain pathogens (153, 154). In epidermal cells, FA export from the
20

chloroplast to the ER is extensive to support the synthesis of waxes (usually very long chain FA
derivatives) and cutin precursors (e.g. hydroxy fatty acids and dicarboxylic acids) at the ER.
These cuticle components need to be transported from the ER to the plasma membrane and
then across the plasma membrane through the cell wall in order to form the cuticle where the
waxes crystallize and the cutin monomers polymerize to cutin (155). While monoacylglycerol
cutin precursors may be sufficiently soluble so that they may not require the use of a
transporter, waxes and other precursors likely require a transport system (155). Thus far, there
is little to no biochemical or molecular evidence for a mechanism of cuticular lipid precursor
trafficking from the ER to the cuticle, but several plasma membrane localized ABC transporters
have been genetically characterized and implicated in cutin precursor and wax transport across
the plasma membrane. Plants with mutated forms of ABCG12/WBC12/CER5 have a cuticle with
reduced wax components, in particular C29 alkanes (154, 156). ABCG12 is hypothesized to
function as a wax transporter because wax lacking in the cuticle accumulates within the cell
(154, 156). When ABCG11/WBC11 is mutated or knocked-out, plants have reduced levels of
both cuticular waxes and cutin (156, 157). Like with abcg12 mutants, waxes were found to
accumulate in the cells of abcg11 mutants (156, 157). The double mutants of ABCG11 and
ABCG12 do not completely deplete the cuticular wax load, and the two mutations do not have
additive effects (157). This suggests the involvement of other transporters as well as a possible
interaction between ABCG11 and ABCG12, which are both half transporters with similar mutant
phenotypes (156). ABCG12 has been confirmed to require the formation of a heterodimer with
ABCG11 in order to be trafficked to the plasma membrane while ABCG11 can form a
homodimer and does not require ABCG12 for proper localization to the plasma membrane
(158). This dimerization pattern and the mutant phenotypes suggest the ABCG11/ABCG12
heterodimer is responsible for transport of cuticular waxes while the ABCG11 homodimer is
responsible for transport of cutin precursors, but this has yet to be confirmed at the biochemical
level. Other ABC transporters with genetic evidence of involvement in cuticular lipid transport
21

include ABCG13 and ABCG32 (159, 160). Mutants carrying abcg13 alleles have reduced levels
of cutin monomers, but this phenotype is restricted to flower petals (159). The knockout of
ABCG32 (permeable cuticle1, pec1) has reduced levels of cutin monomers, but the cuticle is
only structurally impaired in the layer closest to the cell wall suggesting ABCG32 is only
involved in transport to this layer (160). Homologs of ABCG32 have been identified in both rice
and barley with mutants of each showing similar phenotypes as that of Arabidopsis suggesting a
conserved function across plant species (161). There are several other ABC transporters
encoded by genes with high expression in the epidermal cells that may be involved in cuticular
lipid transport, but they have yet to be investigated (156).
Another group of proteins implicated in cuticular lipid transport include the Lipid Transfer
Proteins (LTPs). By definition, LTPs are small (~9 kDa) proteins which have the ability to bind
and transfer lipids in vitro (162). Historically, LTPs were proposed to be involved in intracellular
lipid transfer, but more recently LTPs are thought to more likely mediate cuticular lipid transport
as a large number of LTPs have been localized to the plasma membrane, cell wall, and surface
wax (163-167). Mutants carrying mutations in two GPI-anchored LTPs, LTPG1 and LTPG2,
have been shown to have reduced stem cuticular wax with further wax reduction in the double
mutant over either single mutant (168-170). Both LTPs were chosen for study based on the high
expression of the respective genes in epidermal tissue (169-171). LTPG1 has been confirmed
to have the ability to bind lipids, and it has been localized to the extracellular space of epidermal
cells (169, 171). While this evidence supports a role for LTPG1 and LTPG2 in cuticular lipid
transport, direct biochemical evidence is still lacking.

Transport to the Symbiosome
One membrane unique to a subset of plants is the symbiosome formed around symbiotic
bacteria in root nodules of legumes. The symbiosome is derived from the plasma membrane of
the plant, but its lipid composition is modified by incorporation of considerable amounts of
22

DGDG which is produced in the chloroplast (172-174). As bacteria replicate, more symbiosome
membrane must be formed which requires the transport of plant lipids (172, 173). Lipid transport
to the symbiosome may occur by both vesicular and non-vesicular mechanisms. It is not known
how vesicles are specifically targeted to the symbiosome rather than the plasma membrane or
vacuoles, which both share similar membrane properties to the symbiosome (172, 173).
SNAREs are proteins well known for their involvement in directing vesicles to target membranes
(175).There is some evidence that symbiosome specific SNARE proteins, such as MtSYP132,
may be involved in vesicular trafficking to the symbiosome, but this hypothesis is based only on
bioinformatics and immunolocalization to the symbiosome (176).
For non-vesicular lipid transport to the symbiosome, LTPs are thought to be major
contributors. One LTP implicated in lipid transport to the symbiosome, AsE246, was discovered
in Chinese milk vetch (177). AsE246 gene expression is specific to root nodules, the protein has
been localized to the symbiosome membrane and binds several membrane lipids in vitro and
DGDG in vivo (177). Both overexpression and knockdown lines for AsE246 were analyzed. The
overexpression lines showed increased nodule formation while the knockdown lines had
impaired symbiosome development with lower lipid levels in the nodules (177). Another LTP
specifically expressed in nodules of Medicago, MtN5, was identified as increasing nodulation
and decreasing nodulation in overexpression and knockdown lines, respectively, and it was also
shown to bind lyso-lipids in vitro (178). These findings suggest that AsE246 and MtN5 are
involved in lipid transport to the symbiosome.

Sphingolipid Trafficking
Most of this review has focused on the trafficking of glycerolipids, but as sphingolipids have
many important roles in plant cells (e.g. protein trafficking, pathogen defense, signaling,
structure, etc.), current knowledge about their trafficking is relevant to plant cell biology.
Sphingolipid synthesis takes place in both the ER and the golgi with ceramide and
23

glucosylceramide (GlcCer) synthesis in the ER and glycosyl inositolphosphoceramide (GIPC)
synthesis in the golgi (179). Because of this compartmentalization of synthesis, ceramide must
be transported from the ER to the golgi. This transport may occur by vesicular or non-vesicular
means. In mammals, this ER to golgi ceramide transport occurs non-vesicularly by way of
ceramide transfer protein (CERT), however, no CERT-like proteins have been identified in
plants (179-181). Plant sphingolipids are most abundant in the plasma and vacuolar
membranes, and this means sphingolipids must be transported from both the ER and golgi in
order to accumulate in these other membranes (179, 182). This transport could also occur by
vesicular or non-vesicular means. A major hypothesis in the field is that much of the transport
occurs through the vesicular pathway. The vesicular pathway has been well-studied in yeast
and mammals, but much information is lacking in plants. The reader is referred to the following
papers for information on vesicular trafficking in plants and how it may differ from that in yeast
and mammals (183, 184). There is evidence sphingolipids are involved in targeting vesicular
trafficking by forming distinct microdomains, but the exact mechanisms of microdomain sorting
and specific membrane and protein targeting remain elusive (185-190). Thus far, only two
proteins have been implicated in sphingolipid transport in plants. The first protein, Accelerated
Cell Death 11 (ACD11), is known to cause programmed cell death when absent and was
originally characterized as having sphingosine transfer activity in vitro, but the rates were low
and sphingosine can move through an aqueous phase on its own (191). ACD11 is unable to
bind either ceramide or glycosphingolipids (191). ACD11 is localized to the cytosol, but it is able
to interact with membrane associated proteins BPA1, PRA7, and PRA8 in vitro and BPA1 in
vivo (192). BPA1 is of unknown function, but PRA7 and PRA8 may have roles in ER to golgi
trafficking based on their relation to mammalian proteins localized to the golgi (192). Recently
ACD11 was shown to be a ceramide-1-phosphate (C1P) transfer protein by in vitro binding
assays, crystallization, and accumulation of C1P in the acd11 mutant (193). As C1P is usually
kept at low levels under normal conditions and is involved in the regulation of programmed cell
24

death, it is unlikely ACD11 has a role in normal trafficking of sphingolipids to destination
membranes (193). The second protein implicated in sphingolipid transfer in plants is Glycolipid
Transfer Protein 1 (GLTP1). GLTP1 has the ability to transfer glycosphingolipids in vitro, but
there is no genetic or in vivo data available (194). Another GLTP, At1g21360, is predicted to
have possible sphingolipid transfer abilities, but no work has been published on this protein
(194). It is possible these GLTPs could have roles in the trafficking of sphingolipids in plants.

Concluding Remarks
Plant cells have many unique requirements for lipid trafficking, and the underlying mechanisms
are just beginning to be uncovered. Few transporters have been confirmed, and non-vesicular
trafficking may account for much of the transport. Some of the major questions remaining
include the following: What is the mechanism for transporting FAs to the ER? How are lipids
exported from the chloroplast to other organelles? Which lipid(s) is/are transported from the ER
to the chloroplast for galactolipid synthesis? How are thylakoid membranes assembled from
envelope derived lipids and maintained? How are lipids trafficked specifically to other structures
such as the vacuole, cuticle, and symbiosome? What role do lipid droplets play in lipid
trafficking? How are the bulk of sphingolipids transported to their destination membranes? In
pursuing these questions, lipid transfer phenomena in plants cells provide rich opportunities for
discovery of novel mechanisms and proteins catalyzing lipid transport.

25

REFERENCES

26

REFERENCES

1.

Oldroyd GE, Murray JD, Poole PS, & Downie JA (2011) The rules of engagement in the
legume-rhizobial symbiosis. Annu Rev Genet 45:119-144.

2.

Dormann P & Benning C (2002) Galactolipids rule in seed plants. Trends Plant Sci
7(3):112-118.

3.

Andersson MX & Dörmann P (2009) Chloroplast membrane lipid biosynthesis and
transport. The Chloroplast, Plant Cell Monographs, eds Sandelius A & Aronsson H
(Springer Berlin Heidelberg), Vol 13, pp 125-158.

4.

Fagone P & Jackowski S (2009) Membrane phospholipid synthesis and endoplasmic
reticulum function. J Lipid Res 50 Suppl:S311-316.

5.

Li-Beisson Y, et al. (2013) Acyl-lipid metabolism. in The Arabidopsis book / American
Society of Plant Biologists, p e0161.

6.

Frentzen M, Heinz E, McKeon TA, & Stumpf PK (1983) Specificities and selectivities of
glycerol-3-phosphate acyltransferase and monoacylglycerol-3-phosphate
acyltransferase from pea and spinach chloroplasts. Eur J Biochem 129(3):629-636.

7.

Frentzen M (1986) Biosynthesis and desaturation of the different diacylglycerol moieties
in higher plants. J Plant Physiol 124(3-4):193-209.

8.

Browse J, Warwick N, Somerville CR, & Slack CR (1986) Fluxes through the prokaryotic
and eukaryotic pathways of lipid synthesis in the '16:3' plant Arabidopsis thaliana.
Biochem J 235(1):25-31.

9.

Mongrand S, Bessoule JJ, Cabantous F, & Cassagne C (1998) The C-16 : 3/C-18 : 3
fatty acid balance in photosynthetic tissues from 468 plant species. Phytochemistry
49(4):1049-1064.

10.

Heinz E & Roughan PG (1983) Similarities and differences in lipid metabolism of
chloroplasts isolated from 18:3 and 16:3 plants. Plant Physiol 72(2):273-279.

11.

Giroud C, Gerber A, & Eichenberger W (1988) Lipids of Chlamydomonas reinhardtii analysis of molecular species and intracellular site(s) of biosynthesis. Plant Cell Physiol
29(4):587-595.

12.

Tatsuta T, Scharwey M, & Langer T (2014) Mitochondrial lipid trafficking. Trends Cell
Biol 24(1):44-52.

13.

Hu J, et al. (2012) Plant peroxisomes: biogenesis and function. Plant Cell 24(6):22792303.

14.

van Meer G, Voelker DR, & Feigenson GW (2008) Membrane lipids: where they are and
how they behave. Nat Rev Mol Cell Biol 9(2):112-124.

27

15.

Zachowski A (2012) New insights in membrane lipid asymmetry in animal and plant cells
Transmembrane Dynamics of Lipids, ed Devaux PF, Herrmann, A. (John Wiley & Sons,
Inc., Hoboken, New Jersey), pp 47-64.

16.

Raggers RJ, Pomorski T, Holthuis JC, Kalin N, & van Meer G (2000) Lipid traffic: the
ABC of transbilayer movement. Traffic 1(3):226-234.

17.

Baldridge RD & Graham TR (2013) Two-gate mechanism for phospholipid selection and
transport by type IV P-type ATPases. Proc Natl Acad Sci USA 110(5):E358-367.

18.

Sprong H, van der Sluijs P, & van Meer G (2001) How proteins move lipids and lipids
move proteins. Nat Rev Mol Cell Biol 2(7):504-513.

19.

Holthuis JC & Levine TP (2005) Lipid traffic: floppy drives and a superhighway. Nat Rev
Mol Cell Biol 6(3):209-220.

20.

Dormann P, Voelker TA, & 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(1):612-618.

21.

Salas JJ & Ohlrogge JB (2002) Characterization of substrate specificity of plant FatA
and FatB acyl-ACP thioesterases. Arch Biochem Biophys 403(1):25-34.

22.

Bonaventure G, Salas JJ, Pollard MR, & Ohlrogge JB (2003) Disruption of the FATB
gene in Arabidopsis demonstrates an essential role of saturated fatty acids in plant
growth. Plant Cell 15(4):1020-1033.

23.

Pollard M & Ohlrogge J (1999) Testing models of fatty acid transfer and lipid synthesis in
spinach leaf using in vivo oxygen-18 labeling. Plant Physiol 121(4):1217-1226.

24.

Roughan PG & Slack CR (1977) Long-chain acyl-coenzyme A synthetase activity of
spinach chloroplasts is concentrated in the envelope. Biochem J 162(2):457-459.

25.

Joyard J & Stumpf PK (1981) Synthesis of long-chain acyl-CoA in chloroplast envelope
membranes. Plant Physiol 67(2):250-256.

26.

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(4):1700-1709.

27.

Shockey JM, Fulda MS, & Browse JA (2002) Arabidopsis contains nine long-chain acylcoenzyme a synthetase genes that participate in fatty acid and glycerolipid metabolism.
Plant Physiol 129(4):1710-1722.

28.

Zhao LF, Katavic V, Li FL, Haughn GW, & Kunst L (2010) Insertional mutant analysis
reveals that long-chain acyl-CoA synthetase 1 (LACS1), but not LACS8, functionally
overlaps with LACS9 in Arabidopsis seed oil biosynthesis. Plant J 64(6):1048-1058.

29.

Lu S, et al. (2009) Arabidopsis CER8 encodes LONG-CHAIN ACYL-COA
SYNTHETASE 1 (LACS1) that has overlapping functions with LACS2 in plant wax and
cutin synthesis. Plant J 59(4):553-564.
28

30.

Li N, et al. (2015) FAX1, a novel membrane protein mediating plastid fatty acid export.
PLoS Biology 13(2):e1002053.

31.

Roston RL, Hurlock AK, & Benning C (2014) Plastidic ABC Proteins. Plant ABC
Transporters, Signaling and Communication in Plants, ed Geisler M (Springer, Cham),
Vol 22, pp 103-136.

32.

Kim S, et al. (2013) AtABCA9 transporter supplies fatty acids for lipid synthesis to the
endoplasmic reticulum. Proc Natl Acad Sci USA 110(2):773-778.

33.

Jouhet J, Marechal E, & Block MA (2007) Glycerolipid transfer for the building of
membranes in plant cells. Prog Lipid Res 46(1):37-55.

34.

Baker A, Graham IA, Holdsworth M, Smith SM, & Theodoulou FL (2006) Chewing the
fat: beta-oxidation in signalling and development. Trends Plant Sci 11(3):124-132.

35.

Ajjawi I, Lu Y, Savage LJ, Bell SM, & Last RL (2010) Large-scale reverse genetics in
Arabidopsis: case studies from the Chloroplast 2010 Project. Plant Physiol 152(2):529540.

36.

Xiao S & Chye ML (2009) An Arabidopsis family of six acyl-CoA-binding proteins has
three cytosolic members. Plant Physiol Biochem 47(6):479-484.

37.

Xiao S & Chye ML (2011) New roles for acyl-CoA-binding proteins (ACBPs) in plant
development, stress responses and lipid metabolism. Prog Lipid Res 50(2):141-151.

38.

Schjerling CK, et al. (1996) Disruption of the gene encoding the acyl-CoA-binding protein
(ACB1) perturbs acyl-CoA metabolism in Saccharomyces cerevisiae. J Biol Chem
271(37):22514-22521.

39.

Faergeman NJ & Knudsen J (1997) Role of long-chain fatty acyl-CoA esters in the
regulation of metabolism and in cell signalling. Biochem J 323:1-12.

40.

Xiao S, Chen QF, & Chye ML (2009) Light-regulated Arabidopsis ACBP4 and ACBP5
encode cytosolic acyl-CoA-binding proteins that bind phosphatidylcholine and oleoylCoA ester. Plant Physiol Biochem 47(10):926-933.

41.

Yurchenko OP, et al. (2009) A 10-kDa acyl-CoA-binding protein (ACBP) from Brassica
napus enhances acyl exchange between acyl-CoA and phosphatidylcholine. Plant
Biotech J 7(7):602-610.

42.

Xiao S, Li HY, Zhang JP, Chan SW, & Chye ML (2008) Arabidopsis acyl-CoA-binding
proteins ACBP4 and ACBP5 are subcellularly localized to the cytosol and ACBP4
depletion affects membrane lipid composition. Plant Mol Biol 68(6):571-583.

43.

Leung KC, Li HY, Mishra G, & Chye ML (2004) ACBP4 and ACBP5, novel Arabidopsis
acyl-CoA-binding proteins with kelch motifs that bind oleoyl-CoA. Plant Mol Biol
55(2):297-309.

44.

Yurchenko OP & Weselake RJ (2011) Involvement of low molecular mass soluble acylCoA-binding protein in seed oil biosynthesis. N Biotechnol 28(2):97-109.
29

45.

Li HY, Xiao S, & Chye ML (2008) Ethylene- and pathogen-inducible Arabidopsis acylCoA-binding protein 4 interacts with an ethylene-responsive element binding protein. J
Exp Bot 59(14):3997-4006.

46.

Wang Z & Benning C (2012) Chloroplast lipid synthesis and lipid trafficking through ERplastid membrane contact sites. Biochem Sos Trans 40(2):457-463.

47.

Bates PD, Ohlrogge JB, & Pollard M (2007) Incorporation of newly synthesized fatty
acids into cytosolic glycerolipids in pea leaves occurs via acyl editing. J Biol Chem
282(43):31206-31216.

48.

Mongrand S, Cassagne C, & Bessoule JJ (2000) Import of lyso-phosphatidylcholine into
chloroplasts likely at the origin of eukaryotic plastidial lipids. Plant Physiol 122(3):845852.

49.

Andersson MX, Kjellberg JM, & Sandelius AS (2004) The involvement of cytosolic
lipases in converting phosphatidyl choline to substrate for galactolipid synthesis in the
chloroplast envelope. Biochim Biophys Acta 1684:46-53.

50.

Wang K, Froehlich JE, Zienkiewicz A, Hersh HL, & Benning C (2017) A Plastid
Phosphatidylglycerol Lipase Contributes to the Export of Acyl Groups from Plastids for
Seed Oil Biosynthesis. Plant Cell 29(7):1678-1696.

51.

Li XB, et al. (2012) A galactoglycerolipid lipase is required for triacylglycerol
accumulation and survival following nitrogen deprivation in Chlamydomonas reinhardtii.
Plant Cell 24(11):4670-4686.

52.

Hartel H, Dormann P, & Benning C (2000) DGD1-independent biosynthesis of
extraplastidic galactolipids after phosphate deprivation in Arabidopsis. Proc Natl Acad
Sci USA 97(19):10649-10654.

53.

Awai K, et al. (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 USA
98(19):10960-10965.

54.

Kelly AA, Froehlich JE, & Dormann P (2003) Disruption of the two
digalactosyldiacylglycerol synthase genes DGD1 and DGD2 in Arabidopsis reveals the
existence of an additional enzyme of galactolipid synthesis. Plant Cell 15(11):2694-2706.

55.

Andersson MX, Stridh MH, Larsson KE, Liljenberg C, & Sandelius AS (2003) Phosphatedeficient oat replaces a major portion of the plasma membrane phospholipids with the
galactolipid digalactosyldiacylglycerol. FEBS Lett 537(1-3):128-132.

56.

Jouhet J, et al. (2004) Phosphate deprivation induces transfer of DGDG galactolipid from
chloroplast to mitochondria. J Cell Biol 167(5):863-874.

57.

Andersson MX, Larsson KE, Tjellstrom H, Liljenberg C, & Sandelius AS (2005) The
plasma membrane and the tonoplast as major targets for phospholipid- to-glycolipid
replacement and stimulation of phospholipases in the plasma membrane. J Biol Chem
280(30):27578-27586.
30

58.

Holzl G, et al. (2009) The role of diglycosyl lipids in photosynthesis and membrane lipid
homeostasis in Arabidopsis. Plant Physiol 150(3):1147-1159.

59.

Browse J, McConn M, James D, Jr., & 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(22):16345-16351.

60.

Browse J, McCourt P, & Somerville C (1986) A mutant of Arabidopsis deficient in c(18:3)
and c(16:3) leaf lipids. Plant Physiol 81(3):859-864.

61.

McConn M, Hugly S, Browse J, & Somerville C (1994) A mutation at the Fad8 locus of
Arabidopsis identifies a 2nd chloroplast omega-3 desaturase. Plant Physiol 106(4):16091614.

62.

McConn M & Browse J (1996) The critical requirement for linolenic acid is pollen
development, not photosynthesis, in an Arabidopsis mutant. Plant Cell 8(3):403-416.

63.

Benning C & Ohta H (2005) Three enzyme systems for galactoglycerolipid biosynthesis
are coordinately regulated in plants. J Biol Chem 280(4):2397-2400.

64.

Dorne AJ, Joyard J, Block MA, & Douce R (1985) Localization of phosphatidylcholine in
outer envelope membrane of spinach chloroplasts. J Cell Biol 100(5):1690-1697.

65.

Bessoule JJ, Testet E, & Cassagne C (1995) Synthesis of phosphatidylcholine in the
chloroplast envelope after import of lysophosphatidylcholine from endoplasmic reticulum
membranes. Eur J Biochem 228(2):490-497.

66.

Muthan B, Roston RL, Froehlich JE, & Benning C (2013) Probing Arabidopsis
chloroplast diacylglycerol pools by selectively targeting bacterial diacylglycerol kinase to
suborganellar membranes. Plant Physiol 163(1):61-74.

67.

Miege C, et al. (1999) Biochemical and topological properties of type A MGDG synthase,
a spinach chloroplast envelope enzyme catalyzing the synthesis of both prokaryotic and
eukaryotic MGDG. Eur J Biochem 265(3):990-1001.

68.

Tietje C & Heinz E (1998) Uridine-diphospho-sulfoquinovose: diacylglycerol
sulfoquinovosyltransferase activity is concentrated in the inner membrane of chloroplast
envelopes. Planta 206(1):72-78.

69.

Froehlich JE, Benning C, & Dormann P (2001) The digalactosyldiacylglycerol (DGDG)
synthase DGD1 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(34):31806-31812.

70.

Xu C, Fan J, Riekhof W, Froehlich JE, & Benning C (2003) A permease-like protein
involved in ER to thylakoid lipid transfer in Arabidopsis. EMBO J 22(10):2370-2379.

71.

Xu C, Fan J, Cornish AJ, & Benning C (2008) Lipid trafficking between the endoplasmic
reticulum and the plastid in Arabidopsis requires the extraplastidic TGD4 protein. Plant
Cell 20(8):2190-2204.
31

72.

Lu B, Xu C, Awai K, Jones AD, & Benning C (2007) A small ATPase protein of
Arabidopsis, TGD3, involved in chloroplast lipid import. J Biol Chem 282(49):3594535953.

73.

Awai K, Xu C, Tamot B, & Benning C (2006) A phosphatidic acid-binding protein of the
chloroplast inner envelope membrane involved in lipid trafficking. Proc Natl Acad Sci
USA 103(28):10817-10822.

74.

Xu CC, Moellering ER, Muthan B, Fan JL, & Benning C (2010) Lipid transport mediated
by Arabidopsis TGD proteins is unidirectional from the endoplasmic reticulum to the
plastid. Plant Cell Physiol 51(6):1019-1028.

75.

Xu C, Fan J, Froehlich JE, Awai K, & Benning C (2005) Mutation of the TGD1
chloroplast envelope protein affects phosphatidate metabolism in Arabidopsis. Plant Cell
17(11):3094-3110.

76.

Wang Z, Xu C, & Benning C (2012) TGD4 involved in endoplasmic reticulum-tochloroplast lipid trafficking is a phosphatidic acid binding protein. Plant J 70(4):614-623.

77.

Wang Z, Anderson NS, & Benning C (2013) The phosphatidic acid binding site of the
Arabidopsis Trigalactosyldiacylglycerol 4 (TGD4) protein required for lipid import into
chloroplasts. J Biol Chem 288(7):4763-4771.

78.

Roston RL, Gao J, Murcha MW, Whelan J, & Benning C (2012) TGD1, -2, and -3
proteins involved in lipid trafficking form ATP-binding cassette (ABC) transporter with
multiple substrate-binding proteins. J Biol Chem 287(25):21406-21415.

79.

Roston R, Gao J, Xu C, & Benning C (2011) Arabidopsis chloroplast lipid transport
protein TGD2 disrupts membranes and is part of a large complex. Plant J 66(5):759-769.

80.

Fan J, Zhai Z, Yan C, & Xu C (2015) Arabidopsis
TRIGALACTOSYLDIACYLGLYCEROL5 Interacts with TGD1, TGD2, and TGD4 to
Facilitate Lipid Transfer from the Endoplasmic Reticulum to Plastids. Plant Cell
27(10):2941-2955.

81.

Lu BB & Benning C (2009) A 25-Amino Acid Sequence of the Arabidopsis TGD2 Protein
Is Sufficient for Specific Binding of Phosphatidic Acid. J Biol Chem 284(26):1742017427.

82.

Testerink C & Munnik T (2011) Molecular, cellular, and physiological responses to
phosphatidic acid formation in plants. J Exp Bot 62(7):2349-2361.

83.

Hamilton JA, Bhamidipati SP, Kodali DR, & Small DM (1991) The interfacial
conformation and transbilayer movement of diacylglycerols in phospholipid bilayers. J
Biol Chem 266(2):1177-1186.

84.

Benning C (2009) Mechanisms of lipid transport involved in organelle biogenesis in plant
cells. Annu Rev Cell Del Biol 25:71-91.

85.

Whatley JM, Mclean B, & Juniper BE (1991) Continuity of chloroplast and endoplasmic
reticulum membranes in Phaseolus vulgaris. New Phytol 117(2):209-217.
32

86.

Mclean B, Whatley JM, & Juniper BE (1988) Continuity of chloroplast and endoplasmic
reticulum membranes in Chara and Equisetum. New Phytol 109(1):59-65.

87.

Wooding FBP & Northcot.Dh (1965) Association of endoplasmic reticulum and plastids
in Acer and Pinus. Am J Bot 52(5):526-531.

88.

Tan XL, et al. (2011) A Brassica napus lipase locates at the membrane contact sites
involved in chloroplast development. PLoS One 6(10):e26831.

89.

Andersson MX, Goksor M, & Sandelius AS (2007) Optical manipulation reveals strong
attracting forces at membrane contact sites between endoplasmic reticulum and
chloroplasts. J Biol Chem 282(2):1170-1174.

90.

Fujimoto M & Hayashi T (2011) New insights into the role of mitochondria-associated
endoplasmic reticulum membrane. Int Rev Cell Mol Biol 292:73-117.

91.

Raturi A & Simmen T (2013) Where the endoplasmic reticulum and the mitochondrion tie
the knot: the mitochondria-associated membrane (MAM). BBA Mol Cell Res
1833(1):213-224.

92.

Vance JE & Shiao YJ (1996) Intracellular trafficking of phospholipids: import of
phosphatidylserine into mitochondria. Anticancer Res 16(3B):1333-1339.

93.

Shiao YJ, Lupo G, & Vance JE (1995) Evidence that phosphatidylserine is imported into
mitochondria via a mitochondria-associated membrane and that the majority of
mitochondrial phosphatidylethanolamine is derived from decarboxylation of
phosphatidylserine. J Biol Chem 270(19):11190-11198.

94.

Mehrshahi P, et al. (2013) Transorganellar complementation redefines the biochemical
continuity of endoplasmic reticulum and chloroplasts. Proc Natl Acad Sci USA
110(29):12126-12131.

95.

Mehrshahi P, Johnny C, & Dellapenna D (2014) Redefining the metabolic continuity of
chloroplasts and ER. Trends Plant Sci Advance Online
Publication:10.1016/j.tplants.2014.1002.1013.

96.

Chernomordik LV & Kozlov MM (2005) Membrane hemifusion: crossing a chasm in two
leaps. Cell 123(3):375-382.

97.

Chernomordik LV & Kozlov MM (2008) Mechanics of membrane fusion. Nat Struct Mol
Biol 15(7):675-683.

98.

Hernandez JM, et al. (2012) Membrane fusion intermediates via directional and full
assembly of the SNARE complex. Science 336(6088):1581-1584.

99.

Kornmann B, et al. (2009) An ER-mitochondria tethering complex revealed by a
synthetic biology screen. Science 325(5939):477-481.

100.

Marechal E, Block MA, Dorne AJ, Douce R, & Joyard J (1997) Lipid synthesis and
metabolism in the plastid envelope. Physiol Plant 100(1):65-77.

33

101.

Muhlethaler K & Freywyssling A (1959) Entwicklung und struktur der proplastiden. J
Biophys Biochem Cytol 6(3):507-512.

102.

Morre DJ, Sellden G, Sundqvist C, & Sandelius AS (1991) Stromal low temperature
compartment derived from the inner membrane of the chloroplast envelope. Plant
Physiol 97(4):1558-1564.

103.

Andersson MX, Kjellberg JM, & Sandelius AS (2001) Chloroplast biogenesis: regulation
of lipid transport to the thylakoid in chloroplasts isolated from expanding and fully
expanded leaves of pea. Plant Physiol 127(1):184-193.

104.

Rantfors M, Evertsson I, Kjellberg JM, & Sandelius AS (2000) Intraplastidial lipid
trafficking: regulation of galactolipid release from isolated chloroplast envelope. Physiol
Plant 110(2):262-270.

105.

Westphal S, Soll J, & Vothknecht UC (2001) A vesicle transport system inside
chloroplasts. FEBS Lett 506(3):257-261.

106.

Rawyler A, Meylan M, & Siegenthaler PA (1992) Galactolipid export from envelope to
thylakoid membranes in intact chloroplasts: I. characterization and involvement in
thylakoid lipid asymmetry. Biochim Biophys Acta 1104(2):331-341.

107.

Keren N, Ohkawa H, Welsh EA, Liberton M, & Pakrasi HB (2005) Psb29, a conserved
22-kD protein, functions in the biogenesis of photosystem II complexes in Synechocystis
and Arabidopsis. Plant Cell 17(10):2768-2781.

108.

Hugueney P, et al. (1995) Identification of a plastid protein involved in vesicle fusion
and/or membrane protein translocation. Proc Natl Acad Sci USA 92(12):5630-5634.

109.

Kroll D, et al. (2001) VIPP1, a nuclear gene of Arabidopsis thaliana essential for
thylakoid membrane formation. Proc Natl Acad Sci USA 98(7):4238-4242.

110.

Li HM, Kaneko Y, & Keegstra K (1994) Molecular cloning of a chloroplastic protein
associated with both the envelope and thylakoid membranes. Plant Mol Biol 25(4):619632.

111.

Vothknecht UC, Otters S, Hennig R, & Schneider D (2012) Vipp1: a very important
protein in plastids?! J Exp Bot 63(4):1699-1712.

112.

Gao H & Xu X (2009) Depletion of Vipp1 in Synechocystis sp. PCC 6803 affects
photosynthetic activity before the loss of thylakoid membranes. FEMS Microbiol Lett
292(1):63-70.

113.

Hayashi-Tsugane M, et al. (2014) A mutable albino allele in rice reveals that formation of
thylakoid membranes requires the SNOW-WHITE LEAF1 gene. Plant Cell Physiol
55(1):3-15.

114.

Andersson MX & Sandelius AS (2004) A chloroplast-localized vesicular transport
system: a bio-informatics approach. BMC Genomics 5(1):40.

34

115.

Khan NZ, Lindquist E, & Aronsson H (2013) New putative chloroplast vesicle transport
components and cargo proteins revealed using a bioinformatics approach: an
arabidopsis model. PLoS One 8(4):e59898.

116.

Brandizzi F (2011) Is there a COPII-mediated membrane traffic in chloroplasts? Traffic
12(1):9-11.

117.

Karim S, et al. (2014) A novel chloroplast localized Rab GTPase protein CPRabA5e is
involved in stress, development, thylakoid biogenesis and vesicle transport in
Arabidopsis. Plant Mol Biol 84(6):675-692.

118.

Gounaris K, Barber J, & Harwood JL (1986) The thylakoid membranes of higher plant
chloroplasts. Biochem J 237(2):313-326.

119.

Rawyler A, et al. (1987) The transmembrane distribution of galactolipids in chloroplast
thylakoids is universal in a wide variety of temperate climate plants. Photosynth Res
11(1):3-13.

120.

Viotti C (2014) ER and vacuoles: never been closer. Front Plant Sci 5(20):1-7.

121.

Yoshida S & Uemura M (1986) Lipid composition of plasma membranes and tonoplasts
isolated from etiolated seedlings of mung bean (Vigna radiata L.). Plant Physiol
82(3):807-812.

122.

Leach RP, Wheeler KP, Flowers TJ, & Yeo AR (1990) Molecular markers for ion
compartmentation in cells of higher plants 2: lipid composition of the tonoplast of the
halophyte Suaeda maritima (L) Dum. J Exp Bot 41(230):1089-1094.

123.

Yamaryo Y, Dubots E, Albrieux C, Baldan B, & Block MA (2008) Phosphate availability
affects the tonoplast localization of PLDzeta2, an Arabidopsis thaliana phospholipase D.
FEBS Lett 582(5):685-690.

124.

Nováková P, et al. (2014) SAC phosphoinositide phosphatases at the tonoplast mediate
vacuolar function in Arabidopsis. Proc Natl Acad Sci USA 111(7):2818-2823.

125.

Viotti C, et al. (2013) The endoplasmic reticulum is the main membrane source for
biogenesis of the lytic vacuole in Arabidopsis. Plant Cell 25(9):3434-3449.

126.

Taiz L (1992) The plant vacuole. J Exp Biol 172(Pt 1):113-122.

127.

Feraru E, et al. (2010) The AP-3 beta adaptin mediates the biogenesis and function of
lytic vacuoles in Arabidopsis. Plant Cell 22(8):2812-2824.

128.

Batistic O, Waadt R, Steinhorst L, Held K, & Kudla J (2010) CBL-mediated targeting of
CIPKs facilitates the decoding of calcium signals emanating from distinct cellular stores.
Plant J 61(2):211-222.

129.

De Marchis F, Bellucci M, & Pompa A (2013) Traffic of human alpha-mannosidase in
plant cells suggests the presence of a new endoplasmic reticulum-to-vacuole pathway
without involving the golgi complex. Plant Physiol 161(4):1769-1782.

35

130.

Pedrazzini E, Komarova NY, Rentsch D, & Vitale A (2013) Traffic routes and signals for
the tonoplast. Traffic 14(6):622-628.

131.

Tavernier E & Pugin A (1995) Transbilayer distribution of phosphatidylcholine and
phosphatidylethanolamine in the vacuolar membrane of Acer pseudoplatanus cells.
Biochimie 77(3):174-181.

132.

Chapman KD, Dyer JM, & Mullen RT (2012) Biogenesis and functions of lipid droplets in
plants: thematic review series: lipid droplet synthesis and metabolism: from yeast to
man. J Lipid Res 53(2):215-226.

133.

Walther TC & Farese RV, Jr. (2012) Lipid droplets and cellular lipid metabolism. Annu
Rev Biochem 81:687-714.

134.

Brasaemle DL & Wolins NE (2012) Packaging of fat: an evolving model of lipid droplet
assembly and expansion. J Biol Chem 287(4):2273-2279.

135.

Murphy DJ (2012) The dynamic roles of intracellular lipid droplets: from archaea to
mammals. Protoplasma 249(3):541-585.

136.

Frey-Wyssling A, Grieshaber E, & Mühlethaler K (1963) Origin of spherosomes in plant
cells. J Ultrastruct Res 8(5–6):506-516.

137.

Wanner G & Theimer RR (1978) Membranous appendices of spherosomes
(oleosomes): possible role in fat utilization in germinating oil seeds. Planta 140(2):163169.

138.

Chapman KD & Ohlrogge JB (2012) Compartmentation of triacylglycerol accumulation in
plants. J Biol Chem 287(4):2288-2294.

139.

Siloto RM, et al. (2006) The accumulation of oleosins determines the size of seed
oilbodies in Arabidopsis. Plant Cell 18(8):1961-1974.

140.

Murphy DJ & Vance J (1999) Mechanisms of lipid-body formation. Trends Biochem Sci
24(3):109-115.

141.

Adeyo O, et al. (2011) The yeast lipin orthologue Pah1p is important for biogenesis of
lipid droplets. J Cell Biol 192(6):1043-1055.

142.

Skinner JR, et al. (2009) Diacylglycerol enrichment of endoplasmic reticulum or lipid
droplets recruits Perilipin 3/TIP47 during lipid storage and mobilization. J Biol Chem
284(45):30941-30948.

143.

Moellering ER, Muthan B, & Benning C (2010) Freezing tolerance in plants requires lipid
remodeling at the outer chloroplast membrane. Science 330(6001):226-228.

144.

Gidda SK, et al. (2013) Lipid droplet-associated proteins (LDAPs) are involved in the
compartmentalization of lipophilic compounds in plant cells. Plant Signal Behav 8(11).

36

145.

Moellering ER & Benning C (2010) RNA interference silencing of a major lipid droplet
protein affects lipid droplet size in Chlamydomonas reinhardtii. Eukaryot Cell 9(1):97106.

146.

Vieler A, Brubaker SB, Vick B, & Benning C (2012) A lipid droplet protein of
Nannochloropsis with functions partially analogous to plant oleosins. Plant Physiol
158(4):1562-1569.

147.

Yang L, et al. (2012) The proteomics of lipid droplets: structure, dynamics, and functions
of the organelle conserved from bacteria to humans. J Lipid Res 53(7):1245-1253.

148.

Zehmer JK, et al. (2009) A role for lipid droplets in inter-membrane lipid traffic.
Proteomics 9(4):914-921.

149.

Murphy DJ, et al. (1991) A class of amphipathic proteins associated with lipid storage
bodies in plants: possible similarities with animal serum apolipoproteins. Biochim
Biophys Acta 1088(1):86-94.

150.

Kalantari F, Bergeron JJM, & Nilsson T (2010) Biogenesis of lipid droplets - how cells
get fatter. Mol Membr Biol 27(8):462-468.

151.

Paul LK, Rinne PLH, & Van Der Schoot C (2014) Refurbishing the plasmodesmal
chamber: a role for lipid bodies? Front Plant Sci 5(40):1-10.

152.

van der Schoot C, Paul LK, Paul SB, & Rinne PL (2011) Plant lipid bodies and cell-cell
signaling: a new role for an old organelle? Plant Signal Behav 6(11):1732-1738.

153.

Sohal AK, et al. (1999) Cauliflower Mosaic Virus infection stimulates lipid transfer protein
gene expression in Arabidopsis. J Exp Bot 50(341):1727-1733.

154.

Pighin JA, et al. (2004) Plant cuticular lipid export requires an ABC transporter. Science
306(5696):702-704.

155.

Yeats TH & Rose JK (2013) The formation and function of plant cuticles. Plant Physiol
163(1):5-20.

156.

Bird DA (2008) The role of ABC transporters in cuticular lipid secretion. Plant Sci
174(6):563-569.

157.

Bird D, et al. (2007) Characterization of Arabidopsis ABCG11/WBC11, an ATP binding
cassette (ABC) transporter that is required for cuticular lipid secretion. Plant J 52(3):485498.

158.

McFarlane HE, Shin JJ, Bird DA, & Samuels AL (2010) Arabidopsis ABCG transporters,
which are required for export of diverse cuticular lipids, dimerize in different
combinations. Plant Cell 22(9):3066-3075.

159.

Panikashvili D, Shi JX, Schreiber L, & Aharoni A (2011) The Arabidopsis ABCG13
transporter is required for flower cuticle secretion and patterning of the petal epidermis.
New Phytol 190(1):113-124.

37

160.

Bessire M, et al. (2011) A member of the PLEIOTROPIC DRUG RESISTANCE family of
ATP Binding Cassette transporters is required for the formation of a functional cuticle in
Arabidopsis. Plant Cell 23(5):1958-1970.

161.

Chen GX, et al. (2011) An ATP-binding cassette subfamily G full transporter is essential
for the retention of leaf water in both wild barley and rice. Proc Natl Acad Sci USA
108(30):12354-12359.

162.

Yeats TH & Rose JK (2008) The biochemistry and biology of extracellular plant lipidtransfer proteins (LTPs). Protein Sci 17(2):191-198.

163.

Pyee J, Yu H, & Kolattukudy PE (1994) Identification of a lipid transfer protein as the
major protein in the surface wax of broccoli (Brassica oleracea) leaves. Arch Biochem
Biophys 311(2):460-468.

164.

Pyee J & Kolattukudy PE (1995) The gene for the major cuticular wax-associated protein
and three homologous genes from broccoli (Brassica oleracea) and their expression
patterns. Plant J 7(1):49-59.

165.

Thoma S, et al. (1994) Tissue-specific expression of a gene encoding a cell walllocalized lipid transfer protein from Arabidopsis. Plant Physiol 105(1):35-45.

166.

Sterk P, Booij H, Schellekens GA, Vankammen A, & Devries SC (1991) Cell-specific
expression of the carrot EP2 lipid transfer protein gene. Plant Cell 3(9):907-921.

167.

Suh MC, et al. (2005) Cuticular lipid composition, surface structure, and gene expression
in Arabidopsis stem epidermis. Plant Physiol 139(4):1649-1665.

168.

Lee SB, et al. (2009) Disruption of glycosylphosphatidylinositol-anchored lipid transfer
protein gene altered cuticular lipid composition, increased plastoglobules, and enhanced
susceptibility to infection by the fungal pathogen Alternaria brassicicola. Plant Physiol
150(1):42-54.

169.

Debono A, et al. (2009) Arabidopsis LTPG is a glycosylphosphatidylinositol-anchored
lipid transfer protein required for export of lipids to the plant surface. Plant Cell
21(4):1230-1238.

170.

Kim H, et al. (2012) Characterization of glycosylphosphatidylinositol-anchored lipid
transfer protein 2 (LTPG2) and overlapping function between LTPG/LTPG1 and LTPG2
in cuticular wax export or accumulation in Arabidopsis thaliana. Plant Cell Physiol
53(8):1391-1403.

171.

Ambrose C, DeBono A, & Wasteneys G (2013) Cell geometry guides the dynamic
targeting of apoplastic GPI-linked lipid transfer protein to cell wall elements and cell
borders in Arabidopsis thaliana. PLoS One 8(11):e81215.

172.

Verma DP & Hong Z (1996) Biogenesis of the peribacteroid membrane in root nodules.
Trends Microbiol 4(9):364-368.

38

173.

Jones KM, Kobayashi H, Davies BW, Taga ME, & Walker GC (2007) How rhizobial
symbionts invade plants: the Sinorhizobium-Medicago model. Nat Rev Microbiol
5(8):619-633.

174.

Gaude N, et al. (2004) The galactolipid digalactosyldiacylglycerol accumulates in the
peribacteroid membrane of nitrogen-fixing nodules of soybean and lotus. J Biol Chem
279(33):34624-34630.

175.

Sanderfoot AA & Raikhel NV (1999) The specificity of vesicle trafficking: coat proteins
and SNAREs. Plant Cell 11(4):629-642.

176.

Catalano CM, Czymmek KJ, Gann JG, & Sherrier DJ (2007) Medicago truncatula
syntaxin SYP132 defines the symbiosome membrane and infection droplet membrane in
root nodules. Planta 225(3):541-550.

177.

Lei L, et al. (2014) A nodule-specific lipid transfer protein AsE246 participates in
transport of plant-synthesized lipids to symbiosome membrane and is essential for
nodule organogenesis in Chinese Milk Vetch. Plant Physiol 164(2):1045-1058.

178.

Pii Y, et al. (2009) The Medicago truncatula N5 gene encoding a root-specific lipid
transfer protein is required for the symbiotic interaction with Sinorhizobium meliloti. Mol
Plant Microbe In 22(12):1577-1587.

179.

Chen M, Cahoon EB, Saucedo-Garcia M, Plasencia J, & Gavilanes-Ruiz M (2009) Plant
sphingolipids: structure, synthesis and function. Lipids in Photosynthesis: Essential and
Regulatory Functions, eds Wada H & Murata N (Springer Netherlands), Vol 30, pp 77115.

180.

Hanada K, Kumagai K, Tomishige N, & Kawano M (2007) CERT and intracellular
trafficking of ceramide. Biochim Biophys Acta 1771(6):644-653.

181.

Kudo N, et al. (2008) Structural basis for specific lipid recognition by CERT responsible
for nonvesicular trafficking of ceramide. Proc Natl Acad Sci USA 105(2):488-493.

182.

Markham JE, Li J, Cahoon EB, & Jaworski JG (2006) Separation and identification of
major plant sphingolipid classes from leaves. J Biol Chem 281(32):22684-22694.

183.

Ueda T & Nakano A (2002) Vesicular traffic: an integral part of plant life. Curr Opin Plant
Biol 5(6):513-517.

184.

Nebenfuhr A (2002) Vesicle traffic in the endomembrane system: a tale of COPs, Rabs
and SNAREs. Curr Opin Plant Biol 5(6):507-512.

185.

Yang H, et al. (2013) Sterols and sphingolipids differentially function in trafficking of the
Arabidopsis ABCB19 auxin transporter. Plant J 74(1):37-47.

186.

Markham JE, et al. (2011) Sphingolipids containing very-long-chain fatty acids define a
secretory pathway for specific polar plasma membrane protein targeting in Arabidopsis.
Plant Cell 23(6):2362-2378.

39

187.

Klemm RW, et al. (2009) Segregation of sphingolipids and sterols during formation of
secretory vesicles at the trans-golgi network. J Cell Biol 185(4):601-612.

188.

Laloi M, et al. (2007) Insights into the role of specific lipids in the formation and delivery
of lipid microdomains to the plasma membrane of plant cells. Plant Physiol 143(1):461472.

189.

Breslow DK & Weissman JS (2010) Membranes in balance: mechanisms of sphingolipid
homeostasis. Mol Cell 40(2):267-279.

190.

Lingwood D & Simons K (2010) Lipid rafts as a membrane-organizing principle. Science
327(5961):46-50.

191.

Brodersen P, et al. (2002) Knockout of Arabidopsis ACCELERATED-CELL-DEATH11
encoding a sphingosine transfer protein causes activation of programmed cell death and
defense. Gene Dev 16(4):490-502.

192.

Petersen NHT, et al. (2009) Identification of proteins interacting with Arabidopsis
ACD11. J Plant Physiol 166(6):661-666.

193.

Simanshu Dhirendra K, et al. (2014) Arabidopsis Accelerated Cell Death 11, ACD11, Is
a ceramide-1-phosphate transfer protein and intermediary regulator of phytoceramide
levels. Cell Rep 6(2):388-399.

194.

West G, et al. (2008) Identification of a glycosphingolipid transfer protein GLTP1 in
Arabidopsis thaliana. FEBS J 275(13):3421-3437.

40

CHAPTER 2
In Vivo Lipid “Tag and Track” Approach Shows Acyl Editing of Plastid Lipids and
Chloroplast Import of Phosphatidylglycerol Precursors in Arabidopsis thaliana1

1

This work has been reviewed and is under revision for publication in The Plant Journal with
coauthors Kun Wang, Patrick Horn, and Christoph Benning. Kun Wang took the confocal
picutures and performed the mass spectrometry after I prepared the samples. Patrick Horn
contributed to the interpretation of results and writing strategy.
41

Abstract
In plant lipid metabolism, the synthesis of many intermediates or end products often appears
overdetermined with multiple synthesis pathways acting in parallel. Lipid metabolism is also
dynamic with interorganelle transport, turnover, and remodeling of lipids. To explore this
complexity in vivo, we developed an in vivo lipid “tag and track” method. Essentially, we probed
lipid metabolism in Arabidopsis thaliana by heterologously expressing a coding sequence for a
fatty acid desaturase from Physcomittrella patens (Δ6D). Δ6D places a double bond after the 6th
carbon from the carboxyl end of an acyl group attached to phosphatidylcholine at its sn-2
glyceryl position providing a subtle, but easily trackable modification of the glycerolipid.
Phosphatidylcholine is a central intermediate in plant lipid metabolism as it is modified and
converted to precursors for other lipids throughout the plant cell. Taking advantage of the
exclusive location of Δ6D in the endoplasmic reticulum (ER) and its known substrate specificity
for one of the two acyl groups on phosphatidylcholine, we were able to “tag and track” the
distribution of lipids within multiple compartments and their remodeling in transgenic lines of
different genetic backgrounds. Key findings were the presence of ER-derived precursors in
plastid phosphatidylglycerol and prevalent acyl editing of thylakoid lipids derived from multiple
pathways. We expect that this "tag and track" method will serve as a tool to address several
unresolved aspects of plant lipid metabolism, such as the nature and interaction of different
subcellular glycerolipid pools during plant development or in response to adverse conditions.

Introduction
Plant membrane lipids not only provide the structural basis for cell boundaries and sub-cellular
compartmentalization, but they are also involved in signaling and hormone production (1-5), are
integral to the photosynthetic machinery (6-11), and provide energy storage (12). Plant cells
have a unique lipid composition coinciding with specialized organelles, chloroplasts, to carry out
photosynthesis. While most cellular membranes are composed primarily of phospholipids and
42

sphingolipids, chloroplast membranes are lacking sphingolipids and many classes of
phospholipids,

and

they

are

primarily

composed

of

glycoglycerolipids,

with

monogalactosyldiacylglycerol (MGDG) and digalactosyldiacylglycerol (DGDG) being the most
abundant followed by the sulfolipid, sulfoquinovosyldiacylglycerol (SQDG). The outer envelope
membrane of the chloroplast does contain phosphatidylcholine (PC) (13), but the only
phospholipid abundantly found in the photosynthetic membranes, the thylakoids, is
phosphatidylglycerol (PG).
Complementing the unique chloroplast lipid composition, plants stand out amongst
eukaryotes in their synthesis of chloroplast lipids through the cooperation of two assembly
pathways. Fatty acid synthesis takes place in the stroma of chloroplasts, but chloroplast
glycerolipid biosynthesis occurs both in chloroplasts (plastid pathway) and in the endoplasmic
reticulum (ER pathway). Products from each of these pathways have historically been
distinguished due to the specificity of the respective acyltransferases catalyzing the second step
of glycerolipid assembly, the transfer of an acyl group to sn-1-acylglyceryl sn-3-phosphate (lysophosphatidic acid) (14, 15). The plastid acyltransferase prefers to add 16 carbon (16C) acyl
groups to the sn-2 position of the glyceryl backbone, while the ER acyltransferase prefers to add
18 carbon (18C) acyl groups to the same position (14, 15). Based on this distinction, it has been
posited that nearly 50% of chloroplast lipids of Arabidopsis thaliana are derived from ER
precursors with 18C fatty acids in the sn-2 position (16). This fraction varies among species with
plants, such as Arabidopsis, using both chloroplast and ER pathways, therefore, being
designated as 16:3 plants (17, 18). In contrast, plants designated as 18:3 plants almost entirely
use the ER pathway. However, plastid PG has almost exclusively a 16C fatty acid in the sn-2
position and is, therefore, presumed to be made nearly exclusively by the plastid pathway in
both 16:3 and 18:3 plants (16-18). The use of ER precursors to produce chloroplast lipids also
implies the presence of ER-to-chloroplast lipid transport. While key proteins involved in this

43

transport have been identified, TriGalactosylDiacylglycerol (TGD) 1-5, the transport mechanism
and the transported lipid species remain to be determined (19-26).
Because membrane lipids are important to so many different cellular processes,
extensive efforts have been made to gain a comprehensive understanding of plant lipid
metabolism. However, lipid metabolism has proven to be complex with not only multiple
pathways and transport between organelles but also with acyl editing by way of transacylases,
lipases, and acyltransferases. Acyl editing, in which de novo synthesized fatty acids are
transferred from the acyl-CoA pool to PC, has been characterized by in vivo kinetic radiolabeling
for PC of pea leaves (27) and may occur at the ER or plastid envelope membrane (28). The acyl
groups undergo desaturation on PC, and the now polyunsaturated fatty acids are returned to the
acyl-CoA pool in leaves (27). This acyl editing process actually receives the majority of the flux
from the de novo synthesized acyl-CoA pool derived from chloroplast fatty acid export (27).
Because there are many predicted plastid lipases of unknown function that could be involved in
acyl editing, it is reasonable to consider acyl editing of plastid lipids as part of the plastid
pathway, as has already been hinted at by others (27-29). However, the existence of extensive
acyl editing within the chloroplast could affect the currently accepted quantitative indicator for
identifying lipid precursor origin, the 16C versus 18C occupation in the plastid lipid sn-2 position.
It also raises new questions regarding the origin of plastid PG, which based on the abovementioned criterion, appears to be derived from the plastid pathway to a larger extent than any
other lipid.
Thus far, methods involving radiolabeling and positional analysis of lipids have been the
most informative to determine flux through the different pathways and the transport between
organelles, but conclusions based on these data rely on the sn-2 position 16C versus 18C
standard for distinguishing pathway origin. Here, in vivo lipid tagging was employed using
heterologous production of an ER-located Δ6 desaturase from Physcomitrella patens (Δ6D) in
Arabidopsis. Arabidopsis lacks a Δ6 desaturase, and this approach led to the formation of
44

uniquely traceable Δ6 fatty acyl groups as they appear in different lipids. This Δ6D is well
characterized with a substrate preference for the sn-2 position of PC (30, 31). Expression of the
Δ6D-encoding cDNA in Arabidopsis resulted in lipids with γ-linolenic acid (GLA, 18:3Δ6,9,12;
carbon number: double bond number, Δ double bond position from the carboxyl terminus) and
stearidonic acid (SDA, 18:4Δ6,9,12,15). Tracking of GLA and SDA in the wild type (WT) and
different, well-characterized lipid mutant backgrounds validated this “tag and track” approach as
an in vivo method for detailed study of plant lipid metabolism. Furthermore, it enabled us to
directly test the current standard of using the sn-2 position occupation by 16C versus 18C in
lipid molecular species for the extrapolation of the origin of chloroplast lipids.

Results
Δ6D Is External to the Chloroplast when Heterologously Produced.
In Physcomitrella patens, the Δ6 desaturase is predicted to be located in the ER based on the
presence of a cytochrome B5 domain which is considered indicative of ER fatty acid
desaturases (30). For this reason, the Δ6D was also predicted to be in the ER when
heterologously produced. For the “tag and track” approach to give conclusive results, it was
critical to demonstrate that Δ6D is not targeted to the chloroplast. Therefore, the subcellular
location of Δ6D was determined by cellular fractionation and subsequent immunoblot analysis of
Arabidopsis plants expressing a Δ6D-encoding sequence in the pEarleyGate203 vector
producing the recombinant enzyme with an N-terminal Myc tag (Myc-Δ6D) (32). Protein extract
from Saccharomyces cerevisiae containing Myc-Δ6D was used as a control, and when mixed
with ER protein from non-transgenic plants, the signal for the yeast Myc-Δ6D faded substantially
but remained detectable (Figure 2.1A). The Myc-Δ6D full-length protein and a presumed
degradation product were similarly detectable in two independent ER extracts from transgenic
plants, but both were absent from the transgenic chloroplast sample and the non-transgenic ER
sample (Figure 2.1A). Markers for the chloroplast fraction (oxygen evolving complex 33, OE33)
45

Figure 2.1 Determination of Δ6D Subcellular Location. (A) The portion of Coomassie Brilliant Blue (CBB)
stained gel where Myc-Δ6D runs is shown. The immunoblots for OE33 (33kDa) and BIP (74kDa) show the
extent of cross-contamination in each organelle preparation. The immunoblot for Myc-Δ6D (62 kDa) shows
presence of the recombinant protein in the ER but not the chloroplast. The arrows point to the Myc-Δ6D and a
presumed degradation product, which is only present in the transgenic ER samples. (B) Confocal images of
Δ6D-YFP (top) and empty vector (EV, bottom) are shown with YFP detection (green), chlorophyll detection (red),
and merged images (from left to right). Scale bars = 5μm. WT, wild-type.

and the ER (binding immunoglobulin protein, BIP) were included to indicate purity of the
fractions. Taken together, these results were consistent with an exclusive extra-plastidic location
of the recombinant protein.
For independent corroboration of subcellular location, the Δ6D cDNA was inserted into
the pEarleyGate101 vector (32) leading to a C-terminal YFP fusion protein (Δ6D-YFP). This
vector as well as the pEarlyGate101 empty vector (EV) were transiently expressed in tobacco
(Nicotiana benthamiana), and the tissue was examined by confocal laser microscopy. The YFP
signal in the tissue producing Δ6D-YFP showed a webbing-like pattern (similar to the pattern
usually associated with the ER) with some solid structures resembling ER cisternae (Figure
2.1B). In no instance did the YFP signal overlap with the chlorophyll auto-fluorescence of
chloroplasts, and the tissue with pEarleyGate101 EV showed no YFP signal (Figure 2.1B).

46

Figure 2.2 GLA and SDA Content of Each Independent Line. (A) The spread of GLA and SDA levels as a
mole percentage of total acyl groups between individual plants of each line in each genetic background is
shown. 12 ≤ n ≤58. The Δ6D wild-type (WT) and Δ6D fad3-2 plants are of the T2 generation, and the Δ6D ats1
and Δ6D tgd1-1 plants are of the F3 generation. (B) The ratio of GLA to SDA is shown for each background
using only data from individuals producing > 0.5 mol% SDA. Student's t-test was performed for each background
against the WT, n > 33, * p-value < 0.005. GLA, γ-linolenic acid; SDA, stearidonic acid. Error bars denote SD.

Based on these two lines of independent evidence, we concluded that the Δ6D-YFP was
exclusively located outside the chloroplast, primarily associated with the reticular structure of the
ER as previously predicted (30).

Plants Producing Δ6D Accumulate GLA and SDA.
Potential Arabidopsis Myc-Δ6D-producing T1 plants were screened for Basta resistance, and
survivors were subsequently rescreened by PCR. Positive transformants were then subjected to
total fatty acid methylester analysis. Four transgenic lines were isolated that produced high
levels of GLA and SDA in the WT background (Figure 2.2A). The best line in the WT
background produced 10-18 mol% GLA of total acyl groups and 5-10 mol% SDA (Figure 2.2A).
Progeny of the plants with the highest levels of GLA and SDA were investigated in further detail.
Myc-Δ6D was also produced in three different, well-characterized lipid metabolism
mutants, fad3-2, ats1-1, and tgd1-1, to explore the potential of and to provide proof-of-concept
for the lipid “tag and track" approach based on the Δ6 acyl group redistribution from PC into
different cellular lipids. The fad3-2 mutant is lacking the ER Δ15 desaturase (33) leaving only
47

the plastid Δ15 desaturases. Therefore, any SDA detected in fad3-2 must have received the Δ6
double bond in the ER and the Δ15 double bond in the chloroplast. The dual location of Δ15
desaturases accounts for SDA biosynthesis in the ER and the plastid in WT. However, with the
exclusive origin of GLA in the ER of transgenic lines, we hypothesized that the SDA distribution
between ER and chloroplast lipids as well as the GLA-to-SDA ratio are diagnostic of changes in
the relative contribution of the ER and plastid pathways in the different mutants. Accordingly,
eliminating the ER Δ15 desaturase in the fad3-2 mutant allowed us to verify this hypothesis.
The ats1-1 mutant is deficient in the plastid pathway for lipid assembly as it is strongly
reduced or even lacking the enzyme activity which transfers the first acyl group from an acyl
carrier protein to glycerol-3-phosphate (34, 35). We expected that when Myc-Δ6D is present in
this mutant background, GLA and SDA content in chloroplast lipids should be enhanced due to
the increased ER contribution to plastid lipid assembly compared to the WT background. On the
contrary, the tgd1-1 mutant is impaired in ER-to-chloroplast lipid transport (26), which decreases
the ER contribution to chloroplast lipid assembly and, therefore, should decrease GLA and SDA
content in chloroplast lipids in the presence of Myc-Δ6D.
To develop these lines, the Myc-Δ6D construct was introduced directly into both the WT
and fad3-2, and good producers in the WT background were crossed with the other two
mutants, ats1-1 and tgd1-1. Four independent lines, which were each capable of segregating
plants with high levels of GLA and SDA, were chosen for each genetic background and
subjected to detailed lipid analysis (Figure 2.2A). The levels of GLA and SDA produced were
somewhat variable between individuals within each line, and while only data for the T2 and F3
generations are shown (Figure 2.2A), this trend carried into future generations. Because of this
variability, total GLA and SDA levels were routinely determined for each plant before carrying
out analyses that were more detailed.

48

Figure 2.3 Phenotypes of Transgenic Plants. (A) The fresh weight of whole rosettes (20-day-old plants) is
plotted against the mole percentage of Δ6 acyl groups in total lipid is shown. Samples are binned with the upper
limit of each category shown on the x-axis, n.d. = no data (B) Morphology of non-transgenic plants and 2 lines of
transgenic plants. For each line of transgenic plants, a high producer (HP) of Δ6 acyl groups (>15 mol%) and a
low producer (LP) of Δ6 acyl groups (<1 mol%) are shown.

Total lipid analysis in each of the different mutant backgrounds revealed different relative
levels of GLA and SDA. The Myc-Δ6D fad3-2 plants accumulated less SDA than the Myc-Δ6D
WT plants resulting in a 2 to 3 fold higher GLA-to-SDA ratio (Figure 2.2B) as would be expected
for a mutant without a functioning ER Δ15 desaturase, the same location as for the Δ6
desaturase. The GLA-to-SDA ratio was also slightly increased in the Myc-Δ6D ats1-1 and MycΔ6D tgd1-1 lines (Figure 2.2B). While these mutants introduce a disruption in lipid fluxes
between the ER and the plastid in different ways, it seems likely that the reduced availability of
GLA substrate to the Δ15 desaturases in either the ER or the plastid is reducing overall GLA
conversion to SDA in both mutants leading to the observed increase in GLA-to-SDA ratios.

49

Plants Producing Myc-Δ6D Have a Subtle Secondary Phenotype.
The fresh weight and combined relative fraction (mol%) of Δ6-containing acyl groups, GLA and
SDA, were measured in 3-week-old plants for four independent Myc-Δ6D-producing lines in
each background, WT, fad3-2, ats1-1, and tgd1-1. The plants were binned based on the mol%
fraction of Δ6 acyl groups (Figure 2.3A). The fresh weight of the plants remained within the
same range for the <5 mol% category plants as the Δ6 acyl group mol% fraction increased, but
Myc-Δ6D-producing plants in the WT and fad3-2 backgrounds reached a weight threshold when
Δ6 acyl levels reached close to 25 mol%. Therefore, relatively high levels of GLA and SDA
and/or the corresponding reduction in other acyl groups had a small but discernible effect on
plant growth.
Some sets of plants showed a more perceptible growth phenotype after reaching the
reproductive stage with less or sometimes delayed growth as well as early yellowing of leaves
(Figure 2.3B). This phenotype was not always seen and may have been related to the
respective plants’ hypersensitivity to minor environmental fluctuations experienced in the growth
chambers. In the following, all lipid analyses were conducted on young plants that were
indistinguishable in their growth and appearance from WT.
The presence of GLA and SDA acyl groups was also accompanied by a subtle
secondary lipid phenotype. Myc-Δ6D-producing plants in WT, fad3-2, and ats1-1 backgrounds
had an increased 16C to 18C ratio in some polar lipids (Figure 2.4). This observation could
indicate that the TGD1 complex discriminates slightly against the import of Δ6 acyl group
containing molecular species leading to the observed increase in16C/18C ratio. While this
increase was statistically significant, it was too small to affect any conclusions drawn from GLA
and SDA distribution results discussed below.

50

Figure 2.4 16C to 18C Ratio. The ratio of 16C to 18C acyl groups is represented for total lipid and each polar
lipid for each genetic background, both transgenic and non-transgenic. The key is included in the figure. Student's
t-test was performed comparing Δ6D transgenic to nontransgenic, n = 9-10, * p-value < 0.05, ** p-value < 0.005.
Error bars denote SD. MGDG, monogalactosyldiacylglycerol; PG, phosphatidylglycerol; DGDG,
digalactosyldiacylglycerol;
SQDG,
sulfoquinovosyldiacylglycerol;
PE+PI,
phosphatidylethanolamine+
phosphatidylinositol; PC, phosphatidylcholine; WT, wild-type.

Δ6 Acyl Groups Are Present in Plastid PG.
The distribution of Δ6 acyl groups among polar lipid classes was determined in leaf tissues from
transgenic plants of each genetic background (Figure 2.5A). In general, Myc-Δ6D-producing
plants in each of the tested genetic backgrounds showed the predicted distribution patterns of
Δ6 acyl groups providing proof-of-concept for the described lipid “tag and track” method. The
pattern in Myc-Δ6D fad3-2 was very similar to that of Myc-Δ6D in WT (Figure 2.5A), with the
primary predicted phenotype being the increased GLA to SDA ratio as discussed above (Figure
2.2B). Myc-Δ6D tgd1-1 plants had lower levels of Δ6 acyl groups in the major chloroplast lipids,
MGDG and DGDG, and elevated Δ6 acyl group levels in ER-associated lipids, PC and
phosphatidylethanolamine (PE) and phosphatidylinositol (PI), due to decreased ER-tochloroplast lipid transport (Figure 2.5A). This result also indicates that the TGD complex
transports Δ6 acyl group-containing lipid species from the ER to the inside of the chloroplast
even though at perhaps reduced efficiency as already mentioned above. The Myc-Δ6D ats1-1
51

Figure 2.5 Distribution of Δ6 Acyl Groups Among Polar Glycerolipids. (A) The distribution of Δ6 acyl groups
among the different polar glycerolipids is shown as a % of the total Δ6 acyl groups in all of the polar lipids. (B) The
% of Δ6 acyl groups in each position of the glycerol backbone is shown for each lipid in the Δ6D transgenic wildtype (WT) background. Student's t-test was performed comparing each lipid to whole leaf PC, n = 3-7 for isolated
chloroplast samples, and n = 3-4 for whole leaf samples, * p-value < 0.05. (C) The 18:4 containing MGDG species
are shown as relative % of total MGDG ion counts. 34:7 (18:4/16:3), 36:7 (18:4/18:3), 36:8 (18:4/18:4), n = 8.
Abbreviations as in the Figure 3 legend. All error bars denote SD.

plants had elevated levels of Δ6 acyl groups in MGDG compared to the Myc-Δ6D WT plants
and decreased levels in PC and PI+PE because the ER pathway was compensating for the loss
of the plastid pathway in these plants. The Δ6 acyl group level of DGDG was not affected in the
Myc-Δ6D ats1-1 plants (Figure 2.5A) consistent with the previous observation that DGDG is
almost exclusively produced from ER precursors (16).
Each of the four genetic backgrounds tested showed the presence of Δ6 acyl groups in
PG, and the mol% fraction of the Δ6 acyl groups in PG increased in the Myc-Δ6D ats1-1
background, which is deficient in the plastid pathway (Figure 2.5A inset) and in which the import
of ER-derived lipid species is increased. Lipids were also analyzed from isolated chloroplasts in

52

order to observe the presence of Δ6 acyl groups specifically in plastid PG, and Δ6 acyl group
presence was corroborated (Figure 2.5B).

Δ6 Acyl Groups in Thylakoid Lipids Are Evenly Distributed Between the Glyceryl sn-1
and sn-2 Positions.
Based on the substrate preference of the Δ6D for the acyl group at the sn-2 position of PC (30,
31), Δ6 acyl groups were expected to be predominantly present in the sn-2 position of
chloroplast lipids, assuming that the chloroplast-imported lipid species containing Δ6 acyl
groups is derived without remodeling from PC assembled at the ER. Positional analysis of PC
from both whole leaf tissue and isolated chloroplasts showed a positional distribution of Δ6 acyl
groups with approximately 90% being in the sn-2 position (Figure 2.5B). However, the four
thylakoid lipids differed from PC with Δ6 acyl groups being nearly equally abundant in the sn-1
and sn-2 positions (Figure 2.5B).
Phosphatidic acid (PA) is a crucial intermediate of lipid assembly and a proposed lipid
species imported into plastids (36). Therefore, PA was isolated from whole leaf tissues and
isolated chloroplasts and subjected to positional analysis. PA from whole leaf tissues showed a
similar profile to PC with the majority of Δ6 acyl groups in the sn-2 position, while PA from
isolated chloroplasts had an altered distribution of Δ6 acyl groups in the sn-1 and sn-2 positions
more similar to that of the chloroplast lipids (Figure 2.5B). This result suggests that the observed
plastid PA is not directly or exclusively derived from extraplastidic PC, and that at least a
fraction of PA either imported or assembled in the plastid is likely remodeled by acyl editing.

SDA and 16:3 Are Found in the Same MGDG Molecule.
The acyl group compositional analysis of bulk lipids shown above does not provide information
on the presence of different acyl groups within a single molecule. We hypothesized that direct
evidence for acyl editing will be immediately apparent if exclusively ER- and plastid-derived acyl
53

Figure 2.6 Ion Chromatograms for MGDG. The fragmented and the parent ion chromatograms used to identify
34:7 MGDG are shown.

54

groups are found in the same chloroplast lipid molecule. Therefore, liquid chromatography mass
spectrometry (LC/MS) performed on lipids extracted from transgenic plants with high levels of
Δ6 acyl groups was employed to determine if 18:4 (SDA), derived from the ER, can be found in
the same MGDG molecule as 16:3, an acyl group made exclusively in the chloroplast by the
action of FAD5 (37). Focusing on the most abundant chloroplast lipid, 2% of MGDG was
determined to be a 34:7 molecular species containing one 16:3 and one 18:4 acyl group each
(Figure 2.5C & 2.6). As argued above, identification of this molecular species of MGDG provides
strong supporting evidence for acyl editing of an exclusive chloroplast lipid with ER-derived acyl
groups.

The Effect of Myc-Δ6D on the Metabolism of Specific Lipids.
A pulse-chase labeling experiment using whole leaves and

14

C-acetate was conducted in order

to determine if the rates of synthesis and turnover of specific polar lipids are affected by the
presence of 15-18 mol% Δ6 acyl groups in Δ6D transgenic plants. Both the non-transgenic and
transgenic plants showed similar patterns of

14

C accumulation in each of the lipids during the

pulse and the chase phases of the experiment (Figure 2.7A). However, the transgenic plants did
incorporate approximately two-fold higher total levels of

14

C-acetate per FW in polar lipids than

non-transgenic plants (Figure 2.7B) suggesting an increase in the rate of fatty acid biosynthesis
in the transgenic plants affecting all lipids. The amount of lipid per FW only slightly increased in
transgenic plants (Figure 2.7C) suggesting lipid turnover was likely increased in transgenic
plants to account for the two-fold higher radiolabeling.

Discussion
Differentiation of the two synthesis pathways for chloroplast lipids based on the specificities of
the respective acyltransferases resulting in the glyceryl sn-2 position 16C/18C diagnostic rule
has long been the accepted paradigm. For example, this distinction between lipid species
55

14

14

Figure 2.7 C-Acetate Labeling. (A) The results of a pulse-chase experiment with C-acetate are shown for
polar lipids in both transgenic and non-transgenic plants in the WT background. (B) Levels of radiolabel
incorporated per mg of tissue fresh weight (F.W.) after a 40 min pulse are shown for transgenic and nontransgenic plants in the WT background. (C) Acyl amount (pg) per mg of tissue F.W. is shown for the same
samples used in (B). For (B) and (C), Student's t-test was performed between Δ6D transgenic and nontransgenic, n = 14, * p < 0.05. All error bars denote SD. PA, phosphatidic acid; other lipid abbreviations are as in
the Figure 3 legend.

derived from the two pathways led to the conclusion that the majority of plastid PG is of plastid
origin (16), despite the apparent contradiction that the ats1-1 mutant disrupted in the plastid
pathway retains 75% of WT plastid PG levels (34, 35). The question was whether the ats1-1
mutant had residual ATS1 acyl transferase activity in the plastid. Based on glycerol labeling of
isolated ats1-1 chloroplasts, Xu and coworkers concluded that PA and PG are still being
synthesized in chloroplasts of the ats1-1 mutant, even though the recombinant mutant protein
had very little residual activity (35). Furthermore, when labeled palmitate was fed to leaves, its
incorporation was similar in the ats1-1 mutant and the WT leading to the conclusion that the ER
pathway is likely not responsible for maintaining the observed high plastid PG levels in the ats11 mutant (35). However, the 16C acyl group palmitate may not be the right fatty acid substrate
for detecting lipid precursor assembly at the ER and subsequent lipid import into plastids,
because most chloroplast-imported lipid species contain only 18C acyl groups. Our current
observation of an increase in Δ6 acyl groups in PG in the ats1-1 mutant background points

56

towards the incorporation of a fraction of acyl groups from the ER pathway into chloroplast PG
(Figure 2.5A inset). Based on comparison of the ratio of Δ6 acyl groups-to-total acyl groups
between MGDG and PG, we estimate this fraction for PG to be approximately 18% in the WT,
although we cannot rule out an additional ATS1-independent assembly pathway for PA and PG
as suggested by Xu and coworkers (35). Thus based on our analysis, which is independent of
the 16C/18C rule, a substantial fraction of PG is derived from the ER. It should be noted that the
majority of plastid PG contains an unusual C16 fatty acid at its sn-2, 16:1Δ3t, making it the target
of a specific lipase, PLIP1 (29). The partial ER-origin of plastid PG in Arabidopsis is likely
obscured using traditional approaches by specific acyl group remodeling of chloroplast PG but
is clearly apparent in the Myc-Δ6D transgenic lines using the “tag and track” approach.
The lipid “tag and track” approach also provided strong evidence for acyl editing of
thylakoid lipids. The observed distribution of Δ6 acyl groups in PG, DGDG, MGDG, and SQDG
with nearly equal distribution of Δ6 acyl groups between the sn-1 and sn-2 positions, while 90%
of the Δ6 acyl groups in PC occupy the sn-2 position (Figure 2.5B), provided one piece of
evidence. A second piece of evidence is the observation of SDA and 16:3 acyl groups in the
same MGDG molecule as the first acyl group originates in the ER and the second in the plastid
(Figure 2.5C). Acyl editing has been described for PC at the ER (27) or the plastid envelope
(28), for plastid MGDG in Chlamydomonas (38), and recently for plastid PG during seed
development (29). Each of these reactions are proposed to contribute to the export of acyl
groups from the plastid. It should be noted that acyl editing within the chloroplast might be
increased in Δ6D-producing plants triggered by the presence of unusual fatty acids in the
thylakoid lipids. In fact, the increased incorporation of labeled acetate in transgenic plants
(Figure 2.7B) in parallel with only slight increases in total acyl content (Figure 2.7C) may be an
indication of higher lipid turnover in the transgenic lines compared to WT. Nevertheless, the lipid
“tag and track” approach described here provides strong evidence for extensive acyl editing of

57

thylakoid lipids in the transgenic lines, which likely also occurs in WT leaves, but perhaps to a
lower extent.
The observed change in acyl group composition between PC in the ER or the outer
plastid envelope and the thylakoid lipids might suggest that PC is not the source of the ER
substrate for plastid lipid synthesis. PA has been proposed as the lipid species imported from
the ER based on its specific binding to proteins of the TGD complex involved in chloroplast lipid
import (39, 40). However, the positional analysis of PA was not conclusive because of its
differing positional profiles between whole leaf PA and PA of isolated chloroplasts (Figure 2.5B).
In both cases, we have to assume that the imported lipid species is modified by acyl editing, and
this could be true for PC or PA derived from PC at the outer envelope leaving the question of
the nature of the imported lipid species still unanswered.
Based on the evidence provided by the lipid “tag and track” approach shown here, we
propose a mechanism that includes a combination of direct backbone import and acyl editing
within the chloroplast. Assuming that this mechanism based on Δ6D transgenic plants also
applies to WT plants, one might call into question whether the glyceryl sn-2 position 16C/18C
diagnostic rule for identifying relative fluxes through the ER and plastid pathways still holds true
in view of the observed extensive chloroplast acyl editing in the transgenic lines. Evidence as
outlined above suggests that plastid PG does not entirely follow this rule. However, the ”tag and
track” approach shown here does not allow us to assess fully the extent of chloroplast acyl
editing in WT plants, which could be much lower than in the transgenic plants that have lipids
with unusual fatty acids that could trigger an acyl editing or repair response.

Materials and Methods
Plant Growth and Selection.
Arabidopsis (Arabidopsis thaliana) were grown under a 16h light/8h dark cycle at 22°C/20°C
(light/dark) with a light intensity of 100-120 μmol m-2s-1. The T1 generation of Δ6D plants was
58

screened by spraying 2-3 week-old soil grown plants with 0.01% Basta solution (BioWorld 1%
stock, Cat#30632006). Subsequent screening was done by PCR with internal primers for Δ6D:
forward - GAA GTT GTC GGG TAT GTG, reverse - CAT CCC ATT GTG GCT AAG. T2, T3, F3,
and F4 populations were screened for stable transgenic lines by selection on MS medium (41) +
1% sucrose containing 7mg/L glufosinate ammonium (Sigma Cat# 45520) and subsequent PCR
screening with the aforementioned primers. Arabidopsis for organelle isolation were grown to 34 weeks old on 150x15mm petri plates containing MS medium + 1% sucrose with approximately
30 mg of seed per plate.
Nicotiniana benthamiana were grown under a 16h light/8h dark cycle at a constant 25°C
with a light intensity of 100-120 μmol m-2s-1.

Δ6D Plasmid Construction.
Δ6D coding sequence (30) was amplified from the p35-d6Physco plasmid provided by Dr.
Simone Zäuner using primers compatible for use in the pENTR D-TOPO (Thermo Fisher) entry
vector and subsequent cloning into both pEarleyGate101 and pEarleyGate203 (32) destination
vectors. The same forward primer was used for amplification intended for both pEarleyGate
destination vectors: CAC CAT GGT ATT CGC GGG CGG T. A reverse primer containing a stop
codon was used for pEarleyGate203: TTA ACT GGT GGT AGC ATG CTG CT, and a reverse
primer with no stop codon was used for pEarleyGate101: ACT GGT GGT AGC ATG CTG CTC
T. For yeast expression, a N-terminal Myc tag and a BamHI site were added by PCR using the
forward primer: GAC GGA TCC AAA ATG GAA CAG AAG TTG ATT TCC GAA GAA GAC CTC
ATG GTA TTC GCG GGC, and an ApaI site was added with the reverse primer: TTA GGG
CCC TTA ACT GGT GGT AGC ATG. BamHI and ApaI restriction enzymes were used to clone
the PCR product into the yeast expression vector, pESC-His and create Δ6D pESC-His.

59

Arabidopsis Transgenic Lines.
Arabidopsis Columbia-2 (Col-2) and fad3-2 plants were transformed with Δ6D pEarleyGate203
using the floral dip method as described in (42). Four lines from each background were chosen
for further analysis based on total GLA and SDA content. The four lines in the WT Col-2
background were also crossed with the ats1-1 and tgd1-1 mutants. Four lines from independent
crosses in each of these mutant backgrounds were also chosen for further analysis based on
total GLA and SDA content.

Nicotiniana benthamiana Transient Expression.
Transient expression of Δ6D pEarleyGate101 was performed according to (43).

Recombinant Protein Production in Saccharomyces cerevisiae.
Δ6D pESC-His and pESC-His empty vector were introduced into S. cerevisiae (InvSc1) using
the Frozen EZ Yeast Transformation II kit from Zymo. Yeast cultures were grown in SD-His
selection medium to OD600 = 0.4-0.6, and the cells were then pelleted and resuspended in the
same volume of YPG induction medium. Cells were harvested at 0, 4, 8 and 24 h after induction
and used for protein extraction as described in (44).

Confocal Microscopy.
Confocal imaging was performed on 4-6 week-old tobacco leaves 36 hours post infiltration of
agrobacterium culture using a Nikon A1Rsi confocal laser scanning microscope. YFP fusions
were excited at 514 nm, and the emissions were detected with a 530-600 nm band pass filter.
Chlorophylls were excited at 647 nm, and the emissions were detected with a 660 nm long pass
filter. Images were merged and pseudocolored using Nikon NIS Elements AR software (version
4.30.01).

60

Organelle Isolation and Protein Preparation.
3-5 plates of 3-4 week old seedlings (details above) were used for each chloroplast and ER
isolation. Chloroplast isolation was carried out as described in (22), while ER and protein
extraction were carried out as described here. Protein was extracted from the isolated
chloroplasts by adding 200 μL of extraction buffer (0.1 M Tris-HCl pH6.8, 1% SDS, 15%
glycerol, 5% 2-mercaptoethanol) to approximately 250 μg chlorophyll equivalents of pelleted
chloroplasts. The chloroplast suspension was heated to 95°C for 10 min and re-pelleted in a
microfuge at 14000 x g for 5 min. The supernatant was transferred to a new tube and mixed 1:1
with 2X SDS-PAGE sample buffer (0.1 M Tris-HCl pH6.8, 4% SDS, 20% glycerol, 10% 2mercaptoethanol, 0.05% bromophenol blue) before loading on a protein gel. ER isolation was
done by grinding seedlings and 20 mL of ER grinding buffer (10 mM 2-mercaptoethanol, 5 mM
EDTA, 10mM Tris-HCl pH7.5) plus 2 cOmplete, mini, EDTA-free protease inhibitor tablets
(Roche) into a paste with a mortar and pestle on ice. The whole procedure was carried out on
ice or at 4°C. Solid tissue was filtered out through a double layer of Miracloth, and the liquid was
centrifuged at 3000 x g for 5 min. The supernatant was centrifuged again at 12,000 x g for 10
min and then again at 100,000 x g for 10 min. The pellet was resuspended in 300 μL of ER
grinding buffer and loaded onto a step gradient of 5 mL 25% sucrose in ER grinding buffer and
5 mL 18% sucrose in ER grinding buffer and then centrifuged at 100,000 x g for 1 h. The
interface was collected into a new tube, diluted with 10 mL of ER grinding buffer, and
centrifuged at 100,000 x g for 10 min. The final pellet was resuspended in 100 μL of import
buffer (330 mM sorbitol, 50 mM HEPES pH 8.0-KOH) and mixed 1:1 with 2X SDS-PAGE
sample buffer before loading on a protein gel.

SDS-PAGE and Immuno Blotting.
Protein samples (15 μg plant protein extracts or 20 μL of 32-fold diluted yeast protein extract)
were run on 10% Mini-Protean TGX 10 well, 50μL gels (BioRad cat# 4561034). Gels were either
61

subjected to Coomassie Brilliant Blue stain or used for transfer to polyvinylidene fluoride
membrane (90 V for 50 min at 4°C). Blocking of membranes was done in 5% milk in TBST (20
mM Tris-HCl pH 7.5, 150 mM NaCl, 0.05% TWEEN 20) while all antibody incubations were
done in 5% BSA in TBST (filter sterilized). The primary antibody used for a chloroplast marker
was rabbit αOE33 (Agrisera Cat#AS14 2824), 1:5000. The primary antibody for an ER marker
was rabbit αBIP (Santa Cruz Biotech Cat#sc33757, lot#H1806), 1:200. The primary antibody for
detection of Myc-Δ6D was rabbit αMyc (Cell Signaling Cat#2278, Lot#5), 1:1000. The same
secondary antibody was used in all cases, goat-anti-rabbit HRP, 1:10000. Visualization
substrate used for αOE33 and αBIP membranes was Clarity Western ECL substrate (BioRad
Cat# 1705060). A more sensitive substrate was needed for detection of Myc-Δ6D, so
SuperSignal West Dura substrate (Thermo Fisher Cat#34075) was used.

Lipid Analysis.
Whole tissue lipid analysis was done by subjecting leaf tissue to fatty acid methyl ester (FAME)
preparation as described in (45). Lipid extraction, TLC separation, FAME preparation for scrape
silica, and GC analysis were also carried out as described in (45). Modifications to the
referenced protocol are noted as follows. The FAME preparation for whole tissue lipid analysis
was started with whole leaf tissue rather than scraped silica. The 80°C incubation step for whole
leaf FAME preparation was extended to 45 min. For TLC separation of individual lipids, Sil G-25
20x20 TLC plates (Macherey Nagal Cat#809013) were treated with 0.15M ammonium sulfate
solution at least 2 days prior to use and baked at 250°C for 2.5 h directly before use.

Positional Analysis.
Positional analysis was carried out by first extracting lipids from both isolated chloroplasts and
whole leaf tissue and running them on a TLC as described above for all lipids except PA. The
TLC solvent system of chloroform:methanol:concentrated ammonium hydroxide (65:25:5 v/v/v)
62

was used for PA separation. Silica from areas with the desired lipids was scraped into tubes and
re-extracted with 2 mL of methanol:chloroform (2:1 v/v) and then 1 mL of 0.2 M H3PO4, 1 M KCl.
The organic layer was filtered through glass wool and dried under a nitrogen gas stream. Lipids
were resuspended in 300 μL PBS (phosphate-buffered saline), 1% Triton X-100 and sonicated
3 x 10 s at power level 0.5. 10 μL of 4 mg/mL Rhizopus lipase stock solution in PBS was added
to the lipid suspension and incubated at room temperature for 1 h. Lipids were extracted with
1.1 mL methanol:chloroform (2:1 v/v) and then 400 μL 0.2 M H3PO4, 1 M KCl. The organic layer
was concentrated under a nitrogen gas stream and run on an ammonium sulfate treated TLC.
MGDG, DGDG, PG, SQDG, and PA were run with acetone:toluene:water (91:30:7 v/v/v)
solvent, and PC was run with chloroform:methanol:acetic acid:water (60:30:8:5 v/v/v/v).

Lipid preparation and LC/MS measurement.
Total lipids were extracted from 100 mg (fresh weight) leaf tissue high Δ6 acyl producing plants
with 1 mL of an extraction buffer composed of methyl-tertiary-butyl-ether:methanol (3:1 v/v).
Samples were shaken vigorously for 5 min, and 500 μL of water:methanol (3:1 v/v) was added
and well mixed. Phases were completely separated by 5 min centrifugation at 15,000xg. The
upper phase of the separation was isolated and dried by nitrogen gas. The concentrated lipids
were finally resuspended in 100 µl acetonitrile:isopropanol (7:3 v/v) and diluted 30-fold for
LC/MS measurement.
The LC/MS measurement method was modified from (46). Lipid samples were analyzed
using a Waters Xevo G2-XS Q-TOF mass spectrometer interfaced with a Waters Acquity binary
solvent manager and Waters 2777c autosampler. 10 µl samples were injected onto an Acquity
UPLC CSH C18 column (100 x 2.1 741 mm, Waters Corp) held at 55˚C. The mobile phases
consist of 0.1% acetic acid and 10 mM NH4OAc in water (Solvent A) and methanol/isopropanol
(7:3, v/v containing 0.1% acetic acid and 10 mM NH4OAc ) (Solvent B). A gradient of mobile
phase was applied in a 20-min program with a flow rate of 0.4 ml / min. The gradient profile was
63

performed as follows: hold for one min after injection at 45% A / 55% B, followed by a oneminute linear ramp to 25% A / 75% B and another two-minute linear ramp to 11% A / 89% B.
Then, a 13-minute linear gradient was applied to further decrease to 1% A/ 99% B and
equilibrate for one minute, followed by a two-minute isocratic period at 45% A / 55% B. LC
separated analytes were ionized by negative ion mode electrospray ionization and mass spectra
were acquired using an MSe method in continuum mode over m/z 50 to 1500 to provide data
under non-fragmenting and fragmenting conditions (collision energy ramp from 20-80 V).
Processing and alignment of chromatograms, peak identification and integration were
performed using Progenesis QI (version 2.2). MGDG peaks were filtered out from all peaks
based on their m/z and their expected retention time. The acyl composition of the MGDG peaks
was determined by identification of the fragmented fatty acid ions (the m/z value of SDA is
275.21). The ion intensities of SDA containing MGDG species were normalized to the ion
intensity of all identified MGDG species.

Radiolabeling.
The leaf pulse-chase was performed as described in (29) with adjustments noted as follows. MS
salts and vitamins were added to the incubation medium, and the medium was filter-sterilized
before use or addition of 14C-acetate. Incorporation levels per mg of tissue were determined in a
separate experiment following the same general protocol as for the pulse-chase, but the only
time point taken was the 40 min pulse time point. 10 μL of the lipid extract was also directly
used for scintillation counting instead of running on TLC for individual lipid separation.

64

REFERENCES

65

REFERENCES

1.

Kooijman EE & Burger KN (2009) Biophysics and function of phosphatidic acid: a
molecular perspective. Biochim Biophys Acta 1791(9):881-888.

2.

Zhang W, et al. (2003) The oleate-stimulated phospholipase D, PLDdelta, and
phosphatidic acid decrease H2O2-induced cell death in Arabidopsis. Plant Cell
15(10):2285-2295.

3.

van der Schoot C, Paul LK, Paul SB, & Rinne PL (2011) Plant lipid bodies and cell-cell
signaling: a new role for an old organelle? Plant Signal Behav 6(11):1732-1738.

4.

Feussner I & Wasternack C (2002) The lipoxygenase pathway. Annu Rev Plant Biol
53:275-297.

5.

Janda M, et al. (2013) Phosphoglycerolipids are master players in plant hormone signal
transduction. Plant Cell Reports 32(6):839-851.

6.

Stroebel D, Choquet Y, Popot JL, & Picot D (2003) An atypical haem in the cytochrome
b(6)f complex. Nature 426(6965):413-418.

7.

Jordan P, et al. (2001) Three-dimensional structure of cyanobacterial photosystem I at
2.5 A resolution. Nature 411(6840):909-917.

8.

Liu Z, et al. (2004) Crystal structure of spinach major light-harvesting complex at 2.72 A
resolution. Nature 428(6980):287-292.

9.

Loll B, Kern J, Saenger W, Zouni A, & Biesiadka J (2005) Towards complete cofactor
arrangement in the 3.0 A resolution structure of photosystem II. Nature 438(7070):10401044.

10.

Jones MR (2007) Lipids in photosynthetic reaction centres: structural roles and
functional holes. Prog Lipid Res 46(1):56-87.

11.

Holzl G, et al. (2009) The role of diglycosyl lipids in photosynthesis and membrane lipid
homeostasis in Arabidopsis. Plant Physiol 150(3):1147-1159.

12.

Chapman KD, Dyer JM, & Mullen RT (2012) Biogenesis and functions of lipid droplets in
plants: thematic review series: lipid droplet synthesis and metabolism: from yeast to
man. J Lipid Res 53(2):215-226.

13.

Dorne AJ, Joyard J, Block MA, & Douce R (1985) Localization of phosphatidylcholine in
outer envelope membrane of spinach chloroplasts. J Cell Biol 100(5):1690-1697.

14.

Frentzen M (1986) Biosynthesis and Desaturation of the Different Diacylglycerol
Moieties in Higher-Plants. J Plant Physiol 124(3-4):193-209.

15.

Frentzen M, Heinz E, McKeon TA, & Stumpf PK (1983) Specificities and selectivities of
glycerol-3-phosphate
acyltransferase
and
monoacylglycerol-3-phosphate
acyltransferase from pea and spinach chloroplasts. Eur J Biochem 129(3):629-636.
66

16.

Browse J, Warwick N, Somerville CR, & Slack CR (1986) Fluxes through the prokaryotic
and eukaryotic pathways of lipid synthesis in the '16:3' plant Arabidopsis thaliana.
Biochem J 235(1):25-31.

17.

Mongrand S, Bessoule JJ, Cabantous F, & Cassagne C (1998) The C-16 : 3/C-18 : 3
fatty acid balance in photosynthetic tissues from 468 plant species. Phytochemistry
49(4):1049-1064.

18.

Heinz E & Roughan PG (1983) Similarities and differences in lipid metabolism of
chloroplasts isolated from 18:3 and 16:3 plants. Plant Physiol 72(2):273-279.

19.

Awai K, Xu C, Tamot B, & Benning C (2006) A phosphatidic acid-binding protein of the
chloroplast inner envelope membrane involved in lipid trafficking. Proc Natl Acad Sci U S
A 103(28):10817-10822.

20.

Fan
J,
Zhai
Z,
Yan
C,
&
Xu
C
(2015)
Arabidopsis
TRIGALACTOSYLDIACYLGLYCEROL5 Interacts with TGD1, TGD2, and TGD4 to
Facilitate Lipid Transfer from the Endoplasmic Reticulum to Plastids. Plant Cell
27(10):2941-2955.

21.

Lu B, Xu C, Awai K, Jones AD, & Benning C (2007) A small ATPase protein of
Arabidopsis, TGD3, involved in chloroplast lipid import. J Biol Chem 282(49):3594535953.

22.

Roston R, Gao J, Xu C, & Benning C (2011) Arabidopsis chloroplast lipid transport
protein TGD2 disrupts membranes and is part of a large complex. Plant J 66(5):759-769.

23.

Roston RL, Gao J, Murcha MW, Whelan J, & Benning C (2012) TGD1, -2, and -3
proteins involved in lipid trafficking form ATP-binding cassette (ABC) transporter with
multiple substrate-binding proteins. J Biol Chem 287(25):21406-21415.

24.

Xu C, Fan J, Cornish AJ, & Benning C (2008) Lipid trafficking between the endoplasmic
reticulum and the plastid in Arabidopsis requires the extraplastidic TGD4 protein. Plant
Cell 20(8):2190-2204.

25.

Xu CC, Moellering ER, Muthan B, Fan JL, & Benning C (2010) Lipid transport mediated
by Arabidopsis TGD proteins is unidirectional from the endoplasmic reticulum to the
plastid. Plant Cell Physiol 51(6):1019-1028.

26.

Xu C, Fan J, Riekhof W, Froehlich JE, & Benning C (2003) A permease-like protein
involved in ER to thylakoid lipid transfer in Arabidopsis. EMBO J 22(10):2370-2379.

27.

Bates PD, Ohlrogge JB, & Pollard M (2007) Incorporation of newly synthesized fatty
acids into cytosolic glycerolipids in pea leaves occurs via acyl editing. J Biol Chem
282(43):31206-31216.

28.

Tjellstrom H, Yang Z, Allen DK, & Ohlrogge JB (2012) Rapid kinetic labeling of
Arabidopsis cell suspension cultures: implications for models of lipid export from
plastids. Plant Physiol 158(2):601-611.

67

29.

Wang K, Froehlich JE, Zienkiewicz A, Hersh HL, & Benning C (2017) A Plastid
Phosphatidylglycerol Lipase Contributes to the Export of Acyl Groups from Plastids for
Seed Oil Biosynthesis. Plant Cell:doi: 10.1105/tpc.1117.00397.

30.

Girke T, Schmidt H, Zahringer U, Reski R, & Heinz E (1998) Identification of a novel
delta 6-acyl-group desaturase by targeted gene disruption in Physcomitrella patens.
Plant J 15(1):39-48.

31.

Domergue F, et al. (2003) Acyl carriers used as substrates by the desaturases and
elongases involved in very long-chain polyunsaturated fatty acids biosynthesis
reconstituted in yeast. J Biol Chem 278(37):35115-35126.

32.

Earley KW, et al. (2006) Gateway-compatible vectors for plant functional genomics and
proteomics. Plant J 45(4):616-629.

33.

Shah S, Xin Z, & Browse J (1997) Overexpression of the FAD3 desaturase gene in a
mutant of Arabidopsis. Plant Physiol 114(4):1533-1539.

34.

Kunst L, Browse J, & Somerville C (1988) Altered regulation of lipid biosynthesis in a
mutant of Arabidopsis deficient in chloroplast glycerol-3-phosphate acyltransferase
activity. Proc Natl Acad Sci USA 85(12):4143-4147.

35.

Xu C, Yu B, Cornish AJ, Froehlich JE, & Benning C (2006) Phosphatidylglycerol
biosynthesis in chloroplasts of Arabidopsis mutants deficient in acyl-ACP glycerol-3phosphate acyltransferase. Plant J 47(2):296-309.

36.

Hurlock AK, Roston RL, Wang K, & Benning C (2014) Lipid trafficking in plant cells.
Traffic 15(9):915-932.

37.

Kunst L, Browse J, & Somerville C (1989) A mutant of Arabidopsis deficient in
desaturation of palmitic Acid in leaf lipids. Plant Physiol 90(3):943-947.

38.

Li X, et al. (2012) A Galactoglycerolipid Lipase Is Required for Triacylglycerol
Accumulation and Survival Following Nitrogen Deprivation in Chlamydomonas
reinhardtii. Plant Cell 24(11):4670-4686.

39.

Lu BB & Benning C (2009) A 25-Amino Acid Sequence of the Arabidopsis TGD2 Protein
Is Sufficient for Specific Binding of Phosphatidic Acid. J Biol Chem 284(26):1742017427.

40.

Wang Z, Xu C, & Benning C (2012) TGD4 involved in endoplasmic reticulum-tochloroplast lipid trafficking is a phosphatidic acid binding protein. Plant J 70(4):614-623.

41.

Murashige T & Skoog F (1962) A Revised Medium for Rapid Growth and Bio Assays
with Tobacco Tissue Cultures. Physiol Plantarum 15(3):473-497.

42.

Clough SJ & Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated
transformation of Arabidopsis thaliana. Plant J 16(6):735-743.

68

43.

Ma W, et al. (2015) Deletion of a C-terminal intrinsically disordered region of
WRINKLED1 affects its stability and enhances oil accumulation in Arabidopsis. Plant J
83(5):864-874.

44.

Kushnirov VV (2000) Rapid and reliable protein extraction from yeast. Yeast 16(9):857860.

45.

Wang Z, Benning, C (2011) Arabidopsis thaliana Polar Glycerolipid Profiling by Thin
Layer Chromatography (TLC) Coupled with Gas-Liquid Chromatography (GLC). J Vis
Exp 49:e2518.

46.

Burgos A, et al. (2011) Analysis of short-term changes in the Arabidopsis thaliana
glycerolipidome in response to temperature and light. Plant J 66(4):656-668.

69

CHAPTER 3
Lipid Composition Affects the Stability of Isolated Chloroplasts

70

Abstract
Chloroplasts are often isolated to allow for studies which are free of the influence of all other
components. Here I investigate the effect of lipid composition on the stability of isolated
chloroplasts during extended experiments by comparing chloroplasts of wild-type Arabidopsis to
those of three lipid metabolism mutants, sfr2-3, tgd1-1, and sfr2-3 tgd1-1. The results suggest
both the presence of oligogalactolipids and the double bond index of the acyl groups have an
effect on isolated chloroplast stability, but the double bond index seems to have a stronger and
dominant effect.

Introduction
Chloroplasts are the complex organelles which house the machinery required to carry out
photosynthesis. They have a unique double membrane as an envelope, the outer and the inner
envelope membrane, enclosing the extensive thylakoid membranes, which contain the
photosynthetic protein complexes embedded in lipids. In addition to photosynthesis,
chloroplasts also carry out much of the lipid metabolism within the plant cell (1). Fatty acids are
produced in the chloroplast and are either used within the organelle for glycerolipid biosynthesis
or are exported for glycerolipid biosynthesis in the endoplasmic reticulum (ER) (1, 2). The
glycerolipid precursors produced in both the chloroplast and the ER serve in the assembly of
glycerolipids with galactose sugar head groups which are exclusive to the chloroplast under
normal growth conditions (3).
The combination of the presence of specialized chloroplast lipids and the unique
membrane arrangement allow each subchloroplast membrane to have a different lipid
composition. The outer envelope, which is the most permeable membrane (4), is the only
chloroplast membrane which contains phosphatidylcholine (PC) above levels of contamination
from ER membranes during isolation (5, 6). PC is actually located specifically in the outer leaflet
of the outer envelope membrane but makes up as much as 40% of the total lipid composition of
71

the outer envelope membrane (5, 7). Both envelopes as well as the thylakoid membranes are
composed of substantial quantities of the galactolipids, monogalactosyldiacylglycerol (MGDG)
and digalactosyldiacylglycerol (DGDG), with all three membranes containing close to 30%
DGDG and levels of MGDG increasing from close to 10% in the outer envelope to up to 50% in
the inner envelope or 65% in the thylakoid membranes (7). Because chloroplasts are largely
lacking in phospholipids abundantly found in extraplastidic membranes, phosphatidylinositol
(PI), phosphatidylethanolamine (PE), and phosphatidylserine (PS), the only other major lipids
found in the chloroplast are the sulfolipid, sulfoquinovosyldiacylglycerol (SQDG), and
phosphatidylglycerol (PG) (7). SQDG and PG occur in the range of 5-15% in each chloroplast
membrane, and are slightly more abundant in the thylakoid membranes (7).
Because of the unique biochemistry which takes place within the chloroplasts, these
organelles are often isolated for further study of everything from photosynthesis and protein
import to lipid metabolism. During my studies done with isolated chloroplasts from Arabidopsis
mutants disrupted in lipid metabolism, tgd1-1 and sfr2-3, I observed that chloroplasts with
different lipid composition were more or less stable over time when compared to those isolated
from wild-type (WT) plants. Aside from practical experimental implications, the further study of
this phenomenon may provide novel insights into the physiological function of specific
chloroplast lipids. The tgd1-1 mutant caries a leaky allele and is impaired in ER to chloroplast
lipid transport which is known to affect the ratio of 16C to 18C fatty acids. It was named for its
constitutive production of oligogalactolipids, specifically trigalactosyldiacylglycerol (8, 9). The
production of oligogalactolipids is due to the activation of SFR2 (Sensitive to Freezing 2), which
removes galactose head groups from MGDG and repeatedly adds them to other MGDG
moieties to make oligogalactolipids of which hexagalactosyldiacylglycerol is currently known as
the oligogalactolipid with the most galactosyl residues in its head group (10, 11). While the tgd11 mutant has a constitutively activated SFR2 for not yet known reasons (8, 9), chloroplasts
isolated from WT plants also have SFR2 activated (12). The sfr2-3 mutant lacks functional
72

SFR2 and, therefore, cannot produce oligogalactolipids even after chloroplast isolation. Thus
studying these mutants, I hoped to address the question of whether oligogalactolipids stabilize
chloroplasts.
Once isolated, chloroplasts remain intact for a limited amount of time before breaking.
For this reason, experiments requiring isolated chloroplasts to maintain the integrity of their
membranes for the duration of the experiment are limited to short time spans of usually less
than 1-2 hrs. Depending on the type of experiment being conducted, the chloroplasts of different
mutants may need to be compared to those of WT. In the case of lipid metabolism mutants,
there is a question of whether the isolated chloroplasts of plants with altered lipid composition
remain intact for the same amount of time as those of wild type. Different levels of chloroplast
breakage could potentially alter the conditions of experiments by leaking stromal contents into
the buffer system as well as giving proteins access to compounds in membranes with which
they do not normally come in contact. Overall, the aim of this study was to identify the
differences in chloroplast intactness over time between isolated chloroplasts of Arabidopsis with
different lipid compositions using specific mutants and to correlate these differences with the
lipid phenotype through detailed lipid analysis of isolated chloroplasts. I expected to gain novel
insights into the function of specific lipids and to develop a novel system to study the
physiological function of chloroplast membrane lipids.

Results
Isolated Chloroplasts With Altered Lipid Composition and Metabolism Differ in Stability.
In order to determine if the experimental conditions for isolated chloroplasts with altered lipid
composition are similar to those of wild type, the percentage of chloroplasts which remain intact
after a 2 h room temperature incubation was determined for WT, tgd1-1, sfr2-3, and sfr2-3 tgd11 chloroplasts (Figure 3.1A). As noted above, tgd1-1 plants are defective in ER to chloroplast
lipid transport causing an increase in 16C to 18C ratio as well as production of TGDG and other
73

Figure 3.1 Measurement of Properties of Isolated Chloroplasts. (A) The percent of chloroplasts which
remained intact after a 2 h incubation period is shown for chloroplasts from WT and each mutant. WT, n = 41;
sfr2-3, n = 22; tgd1-1, n = 44; sfr2-3 tgd1-1, n = 4. (B) The average area of the isolated chloroplasts is shown.
185 ≤ n ≤ 222. (C) The MDA equivalents measured for WT and tgd1-1 are shown for 0 and 1 hour after initial
isolation. MDA, malondialdehyde. n = 3. All error bars denote SD. Student's t-test was performed comparing
mutant to wild-type (WT) results, * p < 0.05.

oligogalactolipids (8, 9). The sfr2-3 mutant is known to be more similar to WT in its lipid
composition other than the inability to produce oligogalactolipids under inducing conditions (11,
13), and the sfr2-3 tgd1-1 mutant was included to distinguish between chloroplast stability
differences caused by oligogalactolipid production or the lipid compositional changes specific to
the tgd1-1 mutant. Chloroplasts were aliquoted into equal chlorophyll equivalents for the
experiment. After 2 h at room temperature in the light, chloroplasts remaining intact were reisolated on Percoll cushions, and their chlorophyll was extracted and measured. The chlorophyll
content of re-isolated chloroplasts was compared to the chlorophyll content of aliquots of the
same type, which had not been re-isolated after the 2 h incubation in order to determine the
percentage of chloroplasts which remained intact. Approximately 25% of the WT chloroplasts
were intact after 2 h compared to less than 15% of sfr2-3 chloroplasts (Figure 3.1A). In contrast,
approximately 30% of the tgd1-1 and sfr2-3 tgd1-1 mutant chloroplasts could be re-isolated after
the incubation period (Figure 3.1A).

WT and tgd1-1 Mutant Chloroplasts Show No Differences in Size or Production of Lipid
Oxidation Products.
Other than lipid composition, presumably differences in size of the chloroplasts or amounts of
reactive oxygen species (ROS) produced could affect the stability of chloroplasts in solution. In
order to check the size of the isolated chloroplasts, they were extracted from WT, tgd1-1, and
74

sfr2-3 plants and examined under a light microscope. A series of pictures were captured for
chloroplasts of each type, and ImageJ was used to determine the apparent area projections of
individual chloroplasts. Based on this measure, no apparent size differences were seen
between the WT and mutant chloroplasts (Figure 3.1B). For the ROS measurement, a TBARS
(thiobarbituric acid reactive substances) assay was used, which measures the malondialdehyde
(MDA) content (14). MDA is formed by some lipid peroxidation products, so this method is a
general estimate of lipid oxidation levels (14, 15). Only WT and tgd1-1 chloroplasts were used
for this assay as representative samples. Results showed no differences in MDA levels between
WT and mutant chloroplasts (Figure 3.1C).

Polar Lipid Amounts Change in the Chloroplasts from Different Mutant Backgrounds
While analysis of the polar lipids in whole leaves for the WT, sfr2-3, tgd1-1, and sfr2-3 tgd1-1
plants has been published (8, 9, 11, 13, 16), these analyses needed to be repeated on isolated
chloroplasts in order to determine how chloroplast specific differences in lipid composition might
affect chloroplast intactness. Chloroplast amounts used for lipid measurement were aliquoted
based on chlorophyll equivalents, and the lipids were extracted and used for polar lipid analysis.
When the amounts of polar lipid species were analyzed as a percent of the total polar lipid
amount, not many statistical differences were seen between WT and the three mutants, but the
tgd1-1 and sfr2-3 tgd1-1 mutants did show small increases in the percentage of PC or
oligogalactolipids (Figure 3.2A). It should be noted that the solvent, which was used for polar
lipid separation, could not separate the oligogalactolipids of higher order than TGDG from PC
(Figure 3.2D). Therefore, the apparent increase in PC could also indicate an increase in
oligogalactolipids, although, this would not be expected to be the case in the sfr2-3 tgd1-1
chloroplasts as they lack functional SFR2 protein. The solvent system also could not always
resolve TGDG from PE and PI (Figure 3.2D). Therefore, all three lipids were consistently
isolated together. The amount of lipid detected for the PE+PI+TGDG spot appeared to be
75

Figure 3.2 Polar Lipid Analysis. The individual polar lipid amounts are represented as a percent of total polar
lipid. (A) and as pmol per 200 μg chlorophyll (B) n = 3. (C) TGDG and other oligogalactolipid amounts are show
as fmol per 200 μg chlorophyll, n = 3. (D) Polar lipid separation on TLC for chloroplasts 0 and 1 hr after isolation is
shown with both iodine (all lipid) and α-naphthol (glyclolipid) stains. 1, PA, phosphatidic acid; 2, MGDG,
monogalactosyldiacylglycerol; 3, PG, phosphatidylglycerol; 4, DGDG, diglactosyldiacylglycerol; 5, SQDG,
sulfoquinovosyldiacylglycerol; 6, PE+PI+TGDG, phosphatidylethanolamine+phosphatidylinsitol+trigalactosyldiacylglycerol; 7, PC+Oligogalactolipid, phosphatidylcholine; Oligo, oligogalactolipid. Error bars denote SD.
Student's t-test was performed comparing mutant to wild-type (WT) results, * p < 0.05, ** p < 0.005.

76

increased in the tgd1-1 mutant based on visual examination of the TLC plates (Figure 3.2D) as
would be expected with constitutive production of TGDG, but with n = 3, the apparent difference
was not statistically significant (Figure 3.2A).
The polar lipid amounts were also represented as absolute amounts normalized to the
chlorophyll equivalents used for analysis (Figure 3.2B). While no differences in polar lipid
amounts were detected between WT and sfr2-3 chloroplasts, the tgd1-1 and sfr2-3 tgd1-1
chloroplasts showed increased total polar lipids (relative to chlorophyll) due to increases in
MGDG, PG, and PC+oligogalactolipid (Figure 3.2B). As for the representation as relative mol%
fractions, the PE+PI+TGDG visually seemed to be increased in the tgd1-1 mutant background
(Figure 3.2D), but again, this difference was not statistically significant (Figure 3.2B). This
increase in absolute amounts of polar lipids for tgd1-1 and sfr2-3 tgd1-1 chloroplasts is opposite
of the results obtained previously for leaf tissues based on dry weight (16). The tgd1-1 mutant
plants are known to be slightly paler than WT plants (8, 9), so this increase is probably an
indication of decreased pigment in the chloroplasts rather than increased lipid since these
results are normalized to chlorophyll equivalents. It should be noted that while the tgd1-1 plants
appear pale, previously performed ultrastructural studies showed no differences in the
appearance or abundance of the thylakoid membranes (8).
In order to get a better measure of TGDG and the other oligogalactolipids, a different
solvent system was used to resolve and analyze them separately. The sfr2-3 chloroplasts
showed a small base level of TGDG and higher order oligogalactolipids, which was probably
due to slight overlap with PA on the TLC plate (Figure 3.2C). The WT chloroplasts had
detectable amounts above the sfr2-3 background levels of TGDG but not higher order
oligogalactolipids (Figure 3.2C), which was expected based on the known activation of SFR2
upon isolation of chloroplasts (12). The tgd1-1 chloroplasts had significantly increased levels of
both TGDG and higher order oligogalactolipids compared to WT levels as expected with
constitutively active SFR2 (Figure 3.2C).
77

The appearance of lyso-lipids was also examined on the polar lipid TLC plates. WT, sfr23, and tgd1-1 chloroplast lipids were extracted from chloroplast aliquots immediately after
isolation as well as after a 1 hr room temperature incubation. It was expected that chloroplasts
which retained lower levels of intactness over time, like those of the sfr2-3 mutant, may show
signs of lipid degradation, but no lyso-lipids were detectable for any of the types of chloroplasts
before or after the 1 h incubation (Figure 3.2D).

The Fatty Acid Ratios in Polar Lipids Change as Expected for the Different Mutant
Backgrounds.
Detailed analysis of the mole percentages of different fatty acids in each polar lipid was
determined for WT, sfr2-3, tgd1-1, and sfr2-3 tgd1-1 isolated chloroplasts (Figure 3.3). No
differences were detected in the fatty acid composition of any lipids between the WT and sf2-3
chloroplasts (Figure 3.3). This result for isolated chloroplasts is consistent with the published
results analyzing lipids of whole leaves under normal growth conditions (11, 13). The tgd1-1 and
sfr2-3 tgd1-1 mutant chloroplasts showed decreases in 18:3 (carbon #: double bond #) and
increases in 18:1Δ9 (Δ# = double bond position from the carboxyl end) in all lipids except
PE+PI+TGDG in the tgd1-1 background did not show the decrease in 18:3 (Figure 3.3). PG,
DGDG, and PC+oligogalactolipids also had increased 16:3 levels in the tgd1-1 and sfr2-3 tgd11 chloroplasts, while 16:3 was decreased in MGDG (Figure 3.3). Also in these two types of
chloroplasts, both MGDG and DGDG showed increased 16:0, 16:1, and 16:2 levels, while
PC+oligogalactolipids had increased 16:1 and 16:2 but decreased 16:0 levels (Figure 3.3). All of
these fatty acid changes were consistent with what is known about the tgd1-1 mutant and
indicated that the lipid phenotype of sfr2-3 tgd1-1 replicates that of tgd1-1 in all aspects but
oligogalactolipid production (8, 9).

78

Figure 3.3 Acyl Group Profiles for Polar Lipids. The fractions of different acyl groups are shown as mole
fractions for each polar lipid. PA, phosphatidic acid; MGDG, monogalactosyldiacylglycerol; PG,
phosphatidylglycerol; DGDG, diglactosyldiacylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PE+PI+TGDG,
phosphatidylethanolamine+phosphatidylinsitol+trigalactosyldiacylglycerol; PC+Oligogalactolipid, phosphatidylcholine. Error bars denote SD and n = 3. Student's t-test was performed comparing mutant to wild-type (WT)
results, * p < 0.05, ** p < 0.005.

79

Chloroplasts from plants with the tgd1-1 Mutation Have Changes in Double Bond Index
as well as 16C to 18C and MGDG to DGDG Ratios
In order to get a more complete view of the chloroplast lipid composition differences between
WT, sfr2-3, tgd1-1, and sfr2-3 tgd1-1 chloroplasts, the double bond index (DBI) and both 16C to
18C and MGDG to DGDG ratios were calculated. The DBI represents the level of unsaturation
of the acyl groups with a lower number meaning fewer double bonds and, therefore, less
unsaturation. The DBI of all polar lipids was the same between WT and sfr2-3, but tgd1-1 and
sfr2-3 tgd1-1 both had similar decreases in the DBI of MGDG, PG, DGDG, and SQDG (Figure
3.4A). No changes were seen in either 16C to 18C or MGDG to DGDG ratio between WT and
sfr2-3 chloroplasts (Figure 3.4B&C). However, both the tgd1-1 and sfr2-3 tgd1-1 chloroplasts
had increased 16C to 18C ratios for MGDG, DGDG, and SQDG (Figure 3.4B), which is
consistent with published data for tgd1-1 (17). The MGDG to DGDG ratio was significantly
increased in the tgd1-1 chloroplasts, but the sfr2-3 tgd1-1 chloroplasts did not show a
statistically significant increase in this ratio with only n = 3 (Figure 3.4C). It should be noted that
the MGDG to DGDG ratio was neither statistically significant from WT nor tgd1-1.

Discussion
Assays done with isolated chloroplasts require much care to assure comparable conditions
between samples as well as consideration as to how the conditions may change throughout the
duration of the experiment. Here, isolated chloroplasts with different lipid compositions were
shown to have different levels of stability over time (Figure 3.1A). Because size of the
chloroplasts (Figure 3.1B) and levels of the lipid oxidation product, malondialdehyde (Figure
3.1C), were unchanged between the different types of chloroplasts and no lipid degradation
products were detected (Figure 3.2D), it is likely that the differences in lipid composition are the
direct cause of the stability differences. However, because the method of re-isolation on a
Percoll cushion cannot distinguish between chloroplasts truly remaining intact and chloroplasts
80

Figure 3.4 Lipid Related Ratios and Double Bond Index. (A) The 16C to 18C acyl composition of each polar
lipid is shown for each type of chloroplast, n = 3. (B) The double bond index is shown for each polar lipid of each
type of chloroplast, n = 3. (C) The MGDG to DGDG ratio is shown for each type of chloroplast, 3 ≤ n ≤ 6. PA,
phosphatidic acid; MGDG, monogalactosyldiacylglycerol; PG, phosphatidylglycerol; DGDG, diglactosyldiacylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PE+PI+TGDG, phosphatidylethanolamine+phosphatidylinsitol+trigalactosyldiacylglycerol; PC+Oligogalactolipid, phosphatidylcholine. Error bars denote SD. Student's ttest was performed comparing mutant to wild-type (WT) results, * p < 0.05, ** p < 0.005.

breaking to release stromal contents but then resealing, the stability differences measured here
could also be differences in the ability of the membranes to reseal following damage. In order to
distinguish between these two possibilities a different test for chloroplast intactness can be

81

used. For example, a test which measures for enzyme activity which is normally associated with
the stroma of chloroplasts, such as 6-phosphogluconate dehydrogenase, in the buffer (18),
might provide information about stroma leakage as a consequence of chloroplast rupture
followed by resealing.
Even without further study using a different intactness test, the current results indicate
that the differences in lipid composition of the chloroplasts are causing the chloroplasts to act
differently once isolated. The only lipid change between WT and sfr2-3 chloroplasts is the
reduced amount of TGDG (Figure 3.2C). Since the sfr2-3 chloroplasts are more prone to
breakage than those of WT (Figure 3.1A), TGDG and higher order oligogalactolipids must be
able to confer stability to the membrane whether that be prevention of breakage or the ability to
reseal. SFR2 is activated under freezing conditions, which can disrupt cellular membranes (11),
so it is understandable that oligogalactolipids would serve to protect the intactness of the
membrane.
In contrast to the sfr2-3 chloroplasts, chloroplasts from tgd1-1 and sfr2-3 tgd1-1 show
increased stability once isolated when compared to WT chloroplasts (Figure 3.1A). Because the
sfr2-3 tgd1-1 chloroplasts resemble the phenotype of tgd1-1 rather than that of sfr2-3,
something other than the presence of oligogalactolipids must affect the stability of isolated
chloroplasts, and it must be able to completely compensate for lack of oligogalactolipids. Both
tgd1-1 and sfr2-3 tgd1-1 chloroplasts have decreased DBI in certain lipids (Figure 3.4A), an
increased 16C to 18C ratio (Figure 3.4B), and increased ratios of lipid to chlorophyll (Figure
3.2B). The tgd1-1 chloroplasts also show an increased MGDG:DGDG ratio, while the results for
sfr2-3 tgd1-1 were inconclusive (Figure 3.4B). Since there is no indication of decreased
amounts of membrane damaging lipid oxidation in tgd1-1 (Figure 3.1C), it is unlikely the stability
is due to less chlorophyll and photosynthetic activity. Also, MGDG is a non-bilayer forming lipid,
so an increased MGDG to DGDG ratio would more likely destabilize a membrane than stabilize
it. This leaves the decreased DBI and increased 16C to 18C ratio as the most likely reasons for
82

increased chloroplast stability following isolation. Study of the effect of unsaturation levels on
the propensity for MGDG to form the hexagonal-II phase rather than a bilayer has shown that
decreased levels of unsaturation result in reduced formation of the hexagonal-II phase (19). The
degree of unsaturation has this effect on MGDG because fewer double bonds allow for the acyl
chains to take up less space and help to balance the smaller head group to form more of a
cylindrical than a cone shape. The levels of unsaturation are not known to have an effect on
bilayer formation for PG, DGDG, or SQDG (19). The decreased DBI in MGDG for tgd1-1 and
sfr2-3 tgd1-1 chloroplasts indicates a lower degree of unsaturation (Figure 3.4A). Therefore,
MGDG would presumably be less likely to form a hexagonal-II phase thereby decreasing the
probability of chloroplast disruption during phase transition of the membrane. It is also possible
that an increase in shorter acyl chains, like 16C, may have the same effect on MGDG because
this may also decrease the amount of space occupied by the acyl chains. In order to further
investigate how DBI and 16C to 18C ratio affect chloroplast membrane stability, the next step
would be to run tests on chloroplasts isolated from ats1-1 mutants which have a decreased 16C
to 18C ratio but a similar DBI of MGDG compared to WT (20, 21). Chloroplasts of the ats1-1
mutant would be expected to have reduced stability compared to those of WT if the 16C to 18C
ratio is at least partially responsible for the increased stability of tgd1-1 and sfr2-3 tgd1-1
isolated chloroplasts or no change in stability, if DBI is the only factor responsible.
This work on isolated chloroplasts speaks to how important lipid composition can be to
the physical characteristics of membranes, but additional work needs to be done in order to
substantiate the conclusions. Moving forward, data should be collected for chloroplasts from
ats1-1 plants for all of the experiments already presented here. It is also important to perform a
different type of intactness test, such as measurement for 6-phosphogluconic acid
dehydrogenase activity release into buffer, in order to distinguish between actual resistance to
breakage and the ability to reseal a broken membrane. Having these data in hand, it should be
possible to determine the relative importance of the different factors discussed above for the
83

integrity of isolated chloroplasts and extrapolate these results towards the behavior of
chloroplasts in intact plants exposed to different abiotic stresses that could affect membrane
integrity.

Materials and Methods
Plant Growth and Genotyping
Arabidopsis (Arabidopsis thaliana) were grown on petri plates containing MS medium as
described for organelle isolation in Chapter 2. The genotypes of all of the mutant plants used for
seed stock were also confirmed with established methods for both sfr2-3 and tgd1-1 mutations
(9, 11). This confirmation of genotype was needed because plants with the tgd1-1 mutation in
particular are prone to contamination due to poor pollen quality, so each plant from which seeds
are collected and used must be checked to ensure homozygosity.

Chloroplast Isolation
Chloroplast isolation was carried out as described in Chapter 2.

Chloroplast Intactness
Isolated chloroplasts were aliquoted into 50 μg chlorophyll equivalents and pelleted by
centrifugation at 700 x g for 5 min at 4°C. The pellets were resuspended in 50 μL of import
buffer (330 mM sorbitol, 50 mM HEPES, pH 8.0 KOH) for 1 μg chlorophyll equivalents/μL import
buffer. The chloroplast aliquots were incubated at room temperature in the light for 2 h. After
incubation, half of the aliquots were centrifuged at 5000 rpm in a microfuge for 5 min at room
temperature, and the pellet was extracted with 800 μL of methanol:chloroform (2:1 v/v). These
aliquots were used to determine the absorbance at 652 nm, a measurement for chlorophyll, for
the total amount of chloroplasts in each aliquot. The other half of the aliquots were loaded onto
3 mL 40% Percoll cushions in import buffer and centrifuged at 1500 x g for 5 min at 4°C in order
84

to pellet only the chloroplasts which remained intact for the duration of the incubation. Each
pellet was resuspended in 1 mL of import buffer, centrifuged at 5000 rpm in a microfuge for 5
min at room temperature, and extracted with 800 μL of methanol:chloroform (2:1 v/v). The
absorbance at 652 nm was measured for the re-isolated chloroplast extracts and taken as a
percent of the total absorbance for the chloroplasts which were not re-isolated. This percent was
considered the percent of chloroplasts which remained intact.

Measurement of Chloroplast Area
Isolated chloroplasts were centrifuged at 700 x g for 5 min at 4°C, and the pellets were
resuspended in import buffer (same as above) to a concentration of 1 mg chlorophyll
equivalents per mL import buffer. Images were taken using a Leica DMi8 inverted microscope
and HCX PL S-APO 63x/1.30 oil objective. The area of individual chloroplasts was then
determined from the images using ImageJ software for analysis.

Measurement of Reactive Oxygen Species by TBARS
Isolated chloroplasts were aliquoted into 50 μg chlorophyll equivalents and pelleted by
centrifugation at 700 x g for 5 min at 4°C. Half of the chloroplast aliquots were resuspended in
50 μL import buffer and incubated for 1 h at room temperature in the light. The other aliquots
were immediately resuspended in 1 mL of either TCA solution (4.06% trichloroacetic acid in
water) as the negative control or TBA solution (0.31% thiobarbituric acid in TCA solution) for
measurement of thiobarbituric acid reactive substances (TBARS). The chloroplasts subjected to
1 h incubation were re-pelleted and resuspended in either TCA or TBA solution after the
incubation step. The suspensions were heated at 95°C for 15 min along with 2 tubes of TBA
solution to use as blanks for the spectrophotometric measurements. The samples were allowed
to cool completely and then centrifuged at maxium speed in a microfuge for 3 min at room

85

temperature. The absorbances for each sample were taken at 440 nm, 532 nm, and 600 nm.
The MDA equivalents were calculated as in (15).

Lipid Analysis
Isolated chloroplasts were aliquoted into 200 μg chlorophyll equivalents (or 100 μg chlorophyll
equivalents for thin layer chromatography (TLC) separation only meant for staining and
visualization) and pelleted by centrifugation at 2000 x g in a microfuge for 3 min at room
temperature. Each pellet was then extracted with 800 μL methanol:chloroform:formic acid
(2:1:0.1 v/v/v) followed by 400 μL of 0.2 M H3PO4, 1 M KCl. The organic layer was transferred to
a new tube and dried under a N2 gas stream. The TLC plates used were the same as described
in Chapter 2. Samples were first resuspended in 25 μL chloroform and loaded onto TLC
followed by a second and third resuspensions and loading of 15 μL and 10 μL, respectively, in
order to ensure loading of the entire sample. Two different solvent systems were used for lipid
separation. The solvent used for general polar lipid separation was acetone:toluene:water
(91:30:7 v/v/v), and the solvent used for better separation of the oligogalactolipids was
chloroform:methanol:concentrated ammonium hydroxide (50:25:6 v/v/v). Fatty acid methyl ester
(FAME) production and GC analysis as well as iodine and α-naphthol staining were carried as
described in (22).

86

REFERENCES

87

REFERENCES

1.

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

2.

Li-Beisson Y, et al. (2013) Acyl-lipid metabolism. Arabidopsis Book 11:e0161.

3.

Dormann P & Benning C (2002) Galactolipids rule in seed plants. Trends Plant Sci
7(3):112-118.

4.

Heldt HW & Sauer F (1971) The inner membrane of the chloroplast envelope as the site
of specific metabolite transport. Biochim Biophys Acta 234(1):83-91.

5.

Dorne AJ, Joyard J, Block MA, & Douce R (1985) Localization of phosphatidylcholine in
outer envelope membrane of spinach chloroplasts. J Cell Biol 100(5):1690-1697.

6.

Dorne AJ, Joyard J, & Douce R (1990) Do thylakoids really contain phosphatidylcholine?
Proc Natl Acad Sci U S A 87(1):71-74.

7.

Andersson MX & Dörmann P (2009) Chloroplast Membrane Lipid Biosynthesis and
Transport. The Chloroplast, Plant Cell Monographs, eds Sandelius A & Aronsson H
(Springer Berlin Heidelberg), Vol 13, pp 125-158.

8.

Xu C, Fan J, Froehlich JE, Awai K, & Benning C (2005) Mutation of the TGD1
chloroplast envelope protein affects phosphatidate metabolism in Arabidopsis. Plant Cell
17(11):3094-3110.

9.

Xu C, Fan J, Riekhof W, Froehlich JE, & Benning C (2003) A permease-like protein
involved in ER to thylakoid lipid transfer in Arabidopsis. EMBO J 22(10):2370-2379.

10.

Roston RL, Wang K, Kuhn LA, & Benning C (2014) Structural determinants allowing
transferase activity in SENSITIVE TO FREEZING 2, classified as a family I glycosyl
hydrolase. J Biol Chem 289(38):26089-26106.

11.

Moellering ER, Muthan B, & Benning C (2010) Freezing tolerance in plants requires lipid
remodeling at the outer chloroplast membrane. Science 330(6001):226-228.

12.

Wintermans JFGM, van Besouw A, & Bögemann G (1981) Galactolipid formation in
chloroplast envelopes II. isolation-induced changes in galactolipid composition of
envelopes. Biochim Biophys Acta 663(1):99-107.

13.

Barnes AC, Benning C, & Roston RL (2016) Chloroplast membrane remodeling during
freezing stress is accompanied by cytoplasmic acidification activating SENSITIVE TO
FREEZING2. Plant Physiol 171(3):2140-2149.

14.

Janero DR (1990) Malondialdehyde and thiobarbituric acid-reactivity as diagnostic
indices of lipid peroxidation and peroxidative tissue injury. Free Radical Biol. Med.
9(6):515-540.

88

15.

Hodges DM, DeLong JM, Forney CF, & Prange RK (1999) Improving the thiobarbituric
acid-reactive-substances assay for estimating lipid peroxidation in plant tissues
containing anthocyanin and other interfering compounds. Planta 207(4):604-611.

16.

Fan J, Yan C, Roston R, Shanklin J, & Xu C (2014) Arabidopsis lipins, PDAT1
acyltransferase, and SDP1 triacylglycerol lipase synergistically direct fatty acids toward
beta-oxidation, thereby maintaining membrane lipid homeostasis. Plant Cell
26(10):4119-4134.

17.

Yang Y, Zienkiewicz A, Lavell A, & Benning C (2017) Co-evolution of Domain
Interactions in the Chloroplast TGD1, 2, 3 Lipid Transfer Complex Specific to
Brassicaceae and Poaceae Plants. Plant Cell.

18.

Journet EP & Douce R (1985) Enzymic capacities of purified cauliflower bud plastids for
lipid synthesis and carbohydrate metabolism. Plant Physiol 79(2):458-467.

19.

Webb MS & Green BR (1991) Biochemical and biophysical properties of thylakoid acyl
lipids. Biochimica et Biophysica Acta (BBA) - Bioenergetics 1060(2):133-158.

20.

Kunst L, Browse J, & Somerville C (1988) Altered regulation of lipid biosynthesis in a
mutant of Arabidopsis deficient in chloroplast glycerol-3-phosphate acyltransferase
activity. Proc Natl Acad Sci U S A 85(12):4143-4147.

21.

Xu C, Yu B, Cornish AJ, Froehlich JE, & Benning C (2006) Phosphatidylglycerol
biosynthesis in chloroplasts of Arabidopsis mutants deficient in acyl-ACP glycerol-3phosphate acyltransferase. Plant J 47(2):296-309.

22.

Wang Z, Benning, C (2011) Arabidopsis thaliana Polar Glycerolipid Profiling by Thin
Layer Chromatography (TLC) Coupled with Gas-Liquid Chromatography (GLC). J. Vis.
Exp. 49. 10.3791/2518.

89

CHAPTER 4
Development of In Vitro Lipid Transfer Assays for Determining the Lipid Molecular
Species Transferred from the Endoplasmic Reticulum to the Chloroplast

90

Abstract
Because plants use both ER and chloroplast based lipid synthesis pathways for the production
of chloroplast lipids, a lipid transport mechanism is needed between the two organelles. The
TriGalactosylDiacylglycerol (TGD) proteins have been identified to fulfill this need, but the
identity of the transported lipid species remains elusive. Here, in vitro transport assays are
developed in an effort to determine the identity of this transported lipid species. Many variations
of radiolabeled donor liposome representing the ER membrane to acceptor membranes
represented

by

isolated

chloroplasts

were

tested.

Labeled

phosphatidic

acid,

phosphatidylcholine, and phosphatidylethanolamine + phosphatidylinositol were all transferred
at a decreased rate to tgd mutant chloroplasts compared to those of wild type. Assays
measuring the transport of lipids tagged with a fluorophore to isolated chloroplast envelopes
were also explored. Fluorescent studies showed the TGD1,2,3 complex had a slight preference
for phosphatidic acid, while the TGD4 dimer had a slight preference for phosphatidylcholine.
Both radiolabel and fluorescent assays yielded promising results, but unexpected control results
in many cases questioned whether the in vitro data observed truly reflect the situation in vivo.

Introduction
Chloroplasts are most well-known for housing the photosynthetic machinery which can utilize
solar energy for carbon fixation. These organelles use an extensive membrane system, the
thylakoids, providing a scaffold for the photosynthetic proteins and the thylakoid lipids that are
also found as integral components of the protein complexes (1-6). Plants use two biosynthetic
pathways in order to produce all of the membrane lipids needed for the chloroplast. One
pathway, the plastid pathway, is located in the chloroplast, which is also the site of fatty acid
synthesis for plant cells. The other pathway, the endoplasmic reticulum (ER) pathway, uses fatty
acids exported from the chloroplast to then make membrane lipids, which are exported from the
ER to other organelles including the chloroplast. Membrane lipids require transport mechanisms
91

to transverse the aqueous space between organelles or membrane contact sites between
different organelles which generally still have up to 30 nm of aqueous space in between (7).
The only identified transport mechanism to move lipids from the ER to the chloroplast is
facilitated by TriGalactosylDiacylglycerol (TGD) proteins (8-14). Five TGD proteins have been
identified in genetic screens (8-15). The tgd mutants all show impaired ER to chloroplast lipid
transfer, which can be deduced by an increased 16 carbon to 18 carbon fatty acid ratio
(16C:18C) in these mutants (14). This fatty acid ratio is diagnostic for the relative contribution of
the two pathways because the different substrate specificities of the plastid and ER based acyl
transferases, result in the presence of 16C fatty acids in the sn-2 glyceryl position of plastid
derived lipids whereas 18C fatty acids reside in the sn-2 position of ER derived lipids (16, 17).
The TGD proteins have been well characterized in regards to mutant phenotypes and
complex formation. Plants with defective or missing TGD proteins show a distinctive increase in
the 16C:18C ratio as well as growth and reproductive phenotypes, in addition to the
accumulation of oligogalactolipids (which gave rise to their name) (8-11, 14, 18). TGD4 is a βbarrel protein, which forms a homodimer in the outer envelope membrane of the chloroplast and
has the ability to bind phosphatidic acid (PA) (11, 19, 20). TGD1, 2, and 3 form an ABCtransporter in the inner envelope membrane of the chloroplast with TGD1 and 3 being present in
2 copies each and TGD2 in 8-12 copies (13). TGD1 is the predicted permease while TGD3 has
been confirmed as the ATPase (10, 14). TGD2 is the substrate-binding component which has
been shown to both disrupt membranes and bind PA, and it likely spans the intermembrane
space forming a lipid conduit (8, 15, 18, 21). TGD5 is a small (9.2 kDa) protein which has been
shown to associate with both TGD4 and TGD2 proteins despite the lack of any detectable
association between the actual TGD4 and TGD1,2,3 complexes (9). While these studies provide
detailed characterization of the five TGD proteins, it is still unclear what lipid or lipids the TGD
complexes transport. TGD2 and TGD4 both have the ability to bind PA, but it has not been
shown that either complex can actually transport PA. It is possible that the binding of PA is just
92

a mechanism for membrane contact or disruption to facilitate the transfer of specific lipid
species.
Hypotheses for the transported lipid species between the ER and chloroplast include not
only PA but also phosphatidylcholine (PC), lyso-PC (L-PC), and diacylglycerol (DAG) (22). It is
also possible that more than one lipid species is transferred from the ER to the chloroplast. PC
has long been considered because it is present in the outer leaflet of the outer envelope of the
chloroplast, but the chloroplast lacks the required machinery to produce PC independently (22,
23). L-PC is a candidate for the same reason as PC, but its biophysical properties may allow it
to move more readily between membranes (24-26). DAG is a good candidate because not only
is it the precursor for galactolipid synthesis, but it also has the ability to flip between leaflets of a
membrane at biologically relevant rates (27, 28). This study focused on developing methods to
determine the transported lipid species from the ER to the chloroplast by way of the TGD
proteins.

Results
Isolated Chloroplasts from tgd1-1 Mutants Receive Less Radiolabeled ER Lipids than
Wild-type Chloroplasts.
In order to closely mimic the natural environment for ER to chloroplast lipid transport in vitro,
radiolabeled ER microsomes were incubated with nonlabeled isolated chloroplasts from both
wild type (WT) and the most severe, but still viable tgd mutant carrying the tgd1-1 allele. To
obtain labeled ER, Pisum sativum (pea) plants were incubated with 14C-acetate. 14C-acetate is a
lipid precursor that is incorporated almost exclusively into fatty acyl chains of all glycerolipid
classes. Pea plants were used because they could readily incorporate the label into lipids, and
ER isolation from pea had a higher yield than that from Arabidopsis. The labeled ER
microsomes were incubated with isolated chloroplasts from WT and tgd1-1 plants for 30 min.
Chloroplasts which remained intact throughout the duration of the incubation period were re93

14

Figure 4.1 C Labeled ER Microsome to Intact Chloroplast Lipid Transfer. The % of radiolabel transferred
14
from ER microsomes extracted from C-acetate labeled pea plants to isolated Arabidopsis chloroplasts from
both wild-type (WT) and lipid transport mutant, tgd1-1. (A) The percentage is represented as the percent of
radioactivity from each individual labeled lipid in the ER microsomes. (B) The percentage is represented as the
percent of the total amount of radioactivity in the ER microsomes regardless of lipid class. Student's t-test was
performed comparing WT to tgd1-1 for each lipid, n = 10, * p-value < 0.05, ** p-value < 0.005, error bars denote
SD. PA, phosphatidic acid; MGDG, monogalactosyldiacylglycerol; PG, phosphatidylglycerol; DGDG,
digalactosyldiacylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PE+PI, phosphatidylethanolamine+phosphatidylinositol; PC, phosphatidylcholine.

isolated for lipid analysis, and results were normalized based on chlorophyll equivalents. The
amount of

14

C label measured in each polar lipid from the chloroplasts was compared to the

amount of the

14

C label in each corresponding lipid from the ER microsomes (Figure 4.1A) as

well as the total 14C label in the ER microsomes (Figure 4.1B). The tgd1-1 chloroplasts received
significantly less label in PA and PC than the wild-type chloroplasts, but they also had less label
in phosphatidylethanolamine (PE) and phosphatidylinositol (PI). The PE+PI fraction was
considered a control as chloroplasts lack these phospholipids, and therefore, they should not be
transported at all. Because labeled PE+PI represented the second highest fraction of the total
radioactivity in the chloroplasts (as it is in the ER microsomes alone) and was lower in the tgd11 mutant chloroplasts, the radiolabel detected in the chloroplasts may be a result of
contamination from the microsomes. It is possible that the microsomes are less prone to
“sticking” to tgd mutant chloroplasts because of the presence of trigalactosyldiacylglycerol
94

(TGDG) which is already known to cause differences in chloroplast properties (see Chapter 3).
Alternatively, these lipids are actually transferred in this in vitro system, which may lack
specificity.

Variations of the Radiolabeling Assay Did Not Produce Conclusive Results.
One caveat to using ER microsomes extracted from labeled plants as the donor for the transport
assay with isolated chloroplasts was that the microsomes were always contaminated with
labeled chloroplast lipids such as MGDG and DGDG. With several lipids being labeled and
given the highly active metabolism of intact chloroplasts, it is difficult to distinguish direct
transport of a specific lipid from its production from a transported precursor. Because of these
experimental hurdles, several variations of the in vitro transport experiment were pursued (Table
1).
In one variation, commercially available

14

C-PA and

14

C-PC were simultaneously

incorporated into ER microsomes extracted from pea or WT Arabidopsis plants. The two labeled
lipids were incorporated simultaneously, so I could determine if one substrate was preferred
over the other when in direct competition. These donor microsomes from pea or Arabidopsis
were then incubated with pea or WT Arabidopsis isolated chloroplasts, respectively, in the
presence or absence of ATP. Because the TGD1,2,3 complex is an ABC transporter with an
ATPase subunit, it was expected the labeled transported lipid species would not be as abundant
in the acceptor chloroplasts in the absence of ATP. However, the results showed no significant
difference in the amounts of

14

C-PA or 14C-PC found in the acceptor chloroplasts with or without

ATP (Table 1). This was done with both pea and Arabidopsis chloroplasts because, while
Arabidopsis has its TGD proteins well characterized, pea, as an 18:3 plant (29), is almost
completely reliant on the ER pathway for the synthesis of chloroplast lipids. With this
dependence on lipid flux coming into the chloroplast from the ER, pea chloroplasts were
expected to show a more robust result than the Arabidopsis chloroplasts.
95

Donor

Acceptor

Variable

Above
Background?

Significant
Difference?

Pea Microsomes

Col-2 or tgd1-1
chloroplasts

N/A

Yes

Yes

C-PA and
14
C-PC

WT Microsomes

Col-2 chloroplasts

+/- ATP

Yes

No

C-PA and
14
C-PC

Pea Microsomes

Pea chloroplasts

+/-ATP

Yes

No

C-PA or
C-PC

WT Microsomes

WT or tgd3-1
chloroplasts

+/- UDP-Gal

Yes*

No

C-Acetate

Pea Microsomes

Pea chloroplasts

+/- cytosol

Yes*

No

P-H3PO4

Pea Microsomes

WT or tgd1-1
chloroplasts

N/A

No

No

14

C-Acetate

Liposomes from
Extracted Pea
Lipids

WT, tgd1-1, sfr2-3
or sfr2-3 tgd1-1
chloroplasts

N/A

Yes*

No

14

C-Acetate

Liposomes from
Extracted Pea
Lipids

Pea chloroplasts

pH 7.3 or 8.0

Yes

No

Label
14

C-Acetate

14

14

14

14

14

32

Table 4.1 Summary of Radiolabeled Lipid Transport Assays. * Indicates only 50% of the repeats were above
background levels. In order to determine if there was significant difference of p < 0.05, Student's t-test was performed
between WT and mutant and/or the variable condition (e.g. +/-ATP) when applicable. WT, wild type Arabidopsis.

Another experimental variation used Arabidopsis ER microsomes as mentioned above,
but the microsomes were either labeled with

14

C-PA or

14

C-PC, but not both concomitantly. This

was done in order to simplify the interpretation of the experimental results. By using either
PA or

14

C-

14

C-PC individually, I could distinguish which lipid is the donor of the label present in any

other radiolabeled lipids. The acceptor chloroplasts were isolated from either WT or tgd3-1
mutant chloroplasts with the expectation that the tgd3-1 mutant chloroplasts would accept less
of one of the lipids, therefore, identifying it as the transported lipid species. While tgd3-1 is a
weak mutant compared to other tgd mutants, it was chosen for this set of experiments due to
availability of seeds at the time. The stronger tgd mutant allele, tgd1-1, was used for most of the
other experiments. Different concentrations of UDP-galactose (UDP-Gal) were also added to
the incubation buffer. The addition of UDP-Gal was expected to stimulate the synthesis of

96

galactolipids, MGDG or DGDG, from the transported precursor, and in turn, result in more total
label above background being detected in the acceptor chloroplasts. The results with neither
14

C-PA nor

14

C-PC showed significant differences between amounts of label detected in WT or

tgd3-1 chloroplasts nor with varying amounts of UDP-Gal (Table 1).
Previously, it was shown that when fed with labeled PA, WT chloroplasts produced more
labeled MGDG when a cytosolic protein fraction was included in the buffer (30). Two more
studies also suggested cytosolic factors might be involved in either the transport or metabolism
of the lipid precursor from the ER (31, 32). One paper described that a cytosolic phospholipid
transport protein from spinach could facilitate transport of PC from liposomes to chloroplasts
(32), although this protein was later shown to be a cell wall protein rather than cytosolic (33).
The other paper concluded cytosolic phospholipases are needed for chloroplasts to produce
MGDG from labeled PC transported from liposomes (31). Based on the evidence from the first
and third paper mentioned (30, 31), I hypothesized adding cytosol to these radiolabeling assays
might increase the transfer and hence, the amount of radiolabel detected in the acceptor
chloroplasts. A cytosolic protein fraction was extracted from pea leaf tissue and either included
or excluded from incubation buffer in an assay using ER microsomes from

14

C-acetate labeled

pea as described above as donor and unlabeled isolated pea chloroplasts as acceptor. The
results of this assay showed no difference in amount of labeled lipid received by chloroplasts
with or without the cytosolic fraction (Table 1).
It was thought that one of the reasons no statistically significant differences between
mutant and WT were obtained with these different experimental variations was the overall low
levels of label being detected in the acceptor chloroplasts. Some repeats of the experiments
yielded results less than background levels of radioactivity while others were barely above
background. In order to circumvent this issue, 32P-H3PO4 was used to label pea plants. Because
32

P is a high energy β emitter, it was expected to be more easily detectable compared to the low

energy β emitter,

14

C. Another reason for using
97

32

P was to only detect phospholipids and

simplify the results. ER microsomes were extracted from the labeled pea plants and used as the
32

donor with WT or tgd1-1 mutant chloroplast acceptors. The level of

P label incorporated into

lipids was much lower than expected possibly due to seedlings already having a large unlabeled
phosphate pool, so label could not be detected in the acceptor chloroplasts above background
levels (Table 1).
32

Because the commercially available labeled lipids and the

P labeling did not work as

expected, another variation was attempted in order to get statistically significant results from a
more simplified donor than the ER microsomes extracted from
version of the experiment, total lipids were extracted from

14

C labeled pea plants. In this

14

C-acetate labeled pea plants and

separated by thin layer chromatography (TLC). Just the phospholipids, which would be found in
the ER (i.e. PA, PC, PI, and PE), were extracted from the TLC plate and used to make synthetic
donor liposomes. Those donor liposomes were incubated with WT or tgd1-1 chloroplasts.
Chloroplasts from mutants unable to produce TGDG, sfr2-3 and sfr2-3 tgd1-1, were also
included as acceptors in this experiment in order to mitigate any effects caused by the stability
differences between WT and tgd1-1 (see Chapter 3). While some of the experimental repeats
resulted in acceptor label levels above background, no statistically significant differences were
seen between the different mutant and WT acceptor chloroplasts (Table 1).
The last variation of the labeling experiment was done in consideration of cytosolic pH.
All of the previous assays were conducted in pH 8.0 incubation buffer as is standard for protein
import assays with isolated chloroplasts. However, the pH of the cytosol has been reported to
be close to 7.3 (34). For this assay, liposomes were produced from lipids extracted from

14

C-

acetate labeled pea plants as described above. The liposomes were incubated with isolated pea
chloroplasts in either pH 8.0 or pH 7.3 incubation buffer. Results showed no statistically
significant difference in the amount of label received by chloroplasts in the two different pH
buffers (Table 1).

98

Figure 4.2 Fluorescent Lipid Transport Assay Model. A representative scheme is shown depicting how the
transport of lipids tagged with an NBD fluorophore are transported from liposomes containing a quencher to
isolated chloroplast envelopes which are the location of known lipid transporter proteins. When the NBD-labeled
lipids are transported to the envelopes, an increase in fluorescence is expected due to removal from the
quencher. NBD, 6-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)hexanoic acid; N-Rh-PE, N-(lissamine rhodamine
B sulfonyl)phosphatidylethanolamine.

Transport of Fluorescently Labeled Lipids Can Be Measured in Real-Time.
Because results for transport assays using radiolabel as the mode of detection were below or
just above background levels and were inconclusive in identifying the transported lipid species
between the ER and chloroplast, a different method was pursued. Commercially available lipids
of interest (PA and PC) which were tagged with the fluorescent molecule, 6-(N-(7-nitrobenz-2oxa-1,3-diazol-4-yl)amino)hexanoic acid (NBD), were used in donor liposomes which included
unlabeled

phospholipids

as

well

as

a

quencher,

N-(lissamine

rhodamine

B

sulfonyl)phosphatidylethanolamine (N-Rh-PE). NBD-PS was also included as a control since PS
is not found in chloroplasts and, therefore, should not be transported. The donor liposomes
were always composed 25% each of some form of PA, PC, PS, and N-Rh-PE. The PA, PC, and
PS were either dioleoyl or NBD-labeled; when one lipid was NBD-labeled, the other two would
be dioleoyl. For this assay, the NBD fluorophore was quenched while in the donor liposomes,
and lipid transport to an acceptor (away from the quencher) was detected as an increase in
fluorescence over time (Figure 4.2).

99

Chloroplast envelope membranes were extracted from pea plants or WT, tgd1-1, and
tgd4-1 Arabidopsis plants and used as the acceptors in the assay. Envelope membranes were
used rather than intact chloroplasts for two reasons. One, using envelopes prevents any issues
with chlorophyll or other pigments found in thylakoids from interfering with excitation or detection
during the experiment. Two, the use of envelopes makes it easier to separate the action of the
TGD1,2,3 complex from that of TGD4 if they are isolated in such a way as to form separate
inner and outer envelope vesicles. The TGD1,2,3 complex and TGD4 work in the same
transport pathway, but TGD4 is located in the outer envelope membrane (12, 20), while
TGD1,2,3 is located in the inner envelope membrane (8, 13). While the two complexes do not
necessarily transport the same lipid species, the TGD1,2,3 complex presumably relies on TGD4
to transport its own substrate before TGD1,2,3 can access any substrate. This makes the use of
intact chloroplasts for this type of transport assay complicated. Using envelopes eliminates this
complexity by putting the TGD1,2,3 complex in direct contact with the donor membranes.
For experiments using pea chloroplasts as acceptors, the ATP and protein dependence
of the lipid transport were tested. The ATP dependence is a measure of TGD1,2,3 complex
function because of TGD3 being an ATPase, while the protein dependence is a measure of both
TGD1,2,3 complex and TGD4 function in that it will distinguish between protein-mediated
transport and some other nonspecific mechanism. The ATP dependence was tested by
inclusion or omission of ATP in the incubation buffer as well as by inclusion or omission of
apyrase, which depletes the ATP included in the incubation buffer. The transport of both NBDPA and NBD-PC were reduced by omission of ATP while the transport of NBD-PS was
unaffected by the absence of ATP (Figure 4.3A). The test of ATP dependence by inclusion of
apyrase was only attempted with NBD-PA, and this experiment confirmed the results of the
previous ATP dependence experiments (Figure 4.3B). While the NBD-PA result is consistent
with what is known about the TGD1,2,3 complex, the NBD-PC result is unexpected. TGD1,2,3 is
not expected to transport PC as PC is only found in the outer leaflet of the outer envelope of the
100

Figure 4.3 NBD-Lipid to Pea Envelope Transport. The percentage of the total fluorescence transported from
donor liposomes to acceptor envelopes as represented by the increase in fluorescence after a 200 second
incubation with envelopes. The percentage of increase in fluorescence was calculated using the amount of
flourescence after addition of Triton X-100 as the total since the detergent separates the NBD from the
quencher. The label on the x-axis indicates which lipid in the donor liposomes was labeled with NBD. (A) The
percentage of total NBD-lipid transported is shown in the presence (white) and absence (grey) of ATP. PA, n =
9; PC and PS, n = 4. (B) The percentage of total transported NBD-PA is shown without (white) or with (grey)
addition of apyrase. n = 3. (C) The percentage of total transported NBD-lipid is shown without (white) or with
(grey) ProteinaseK. PA, n = 7; PC and PC, n = 3. Student's t-test comparing the conditions for each lipid tested.
* p < 0.05, ** p < 0.005. Error bars denote SD. PA, phosphatidic acid; PC, phosphatidylcholine; PS,
phosphatidylserine.

chloroplast, and, therefore, the TGD1,2,3, complex would not be expected to have access to PC
in vivo (23). The result of NBD-PS being transported at all is also unexpected because
chloroplasts do not naturally contain any PS. Protein dependence of NBD-lipid transport was
tested by inclusion or omission of a protease, ProteinaseK. The transport of all tested lipid
species was decreased when donor liposomes were incubated with protease-treated envelopes
(Figure 4.3C). This was an important control experiment to confirm that the measured transport
was protein-mediated and not an artifact from using NBD-lipids which may possess altered
biophysical properties from native lipid species.
For fluorescently labeled lipid transport assays using Arabidopsis chloroplast envelopes
as acceptors, envelopes from WT, tgd1-1, and tgd4-1 plants were used. The tgd1-1 mutant
allele was chosen to represent the TGD1,2,3 complex because it has the most severe of all of
the tgd mutant phenotypes and may, therefore, yield the most obvious results. The tgd4-1
mutant allele was chosen to represent the TGD4 complex because while it may not have the
most severe phenotype of all the tgd4 mutant alleles, it is possible to maintain a homozygous
seed stock for this allele. Homozygous knock-out lines of TGD4 are sterile (11), and for the
purposes of extracting chloroplast envelopes, the availability of homozygous seed for tgd4-1
101

Figure 4.4 NBD-Lipid to Arabidopsis Envelope Transport. (A-C) The increase in fluorescence over time after
addition of acceptor envelopes to donor liposomes is shown for three NBD-labeled lipids: (A) PA, (B) PC, and (C)
PS. Each shows 2 replicates for wild-type (WT, blues) and tgd1-1 mutant (greens) collected during the same
experimental repeat. (D) The results of all experimental repeats taken as the % of total fluorescence after 200
seconds of incubation with envelopes are shown for WT, tgd1-1, and tgd4-1. Student's t-test was performed
comparing mutant to wild-type results from the same experiments. 10 ≤ n ≤ 21, * p < 0.05, ** p < 0.005. Error
bars denote SD. PA, phosphatidic acid; PC, phosphatidylcholine; PS, phosphatidylserine.

was crucial. Both tgd1-1 and tgd4-1 showed less increase in fluorescence over time than the
wild-type for NBD-PA and NBD-PC while no differences were seen with NBD-PS (Figure 4.4).
The difference with NBD-PA appeared to be greater and statistically significant for tgd1-1, and
the difference with NBD-PC appeared to be greater and statistically significant for tgd4-1 (Figure
4.4D). While this could suggest PA as the preferred substrate for TGD1,2,3 and PC for TGD4,
the results were often too variable within even a single experiment (Figure 4.4A-C) to draw firm
conclusions.

Discussion
As an approach to identify which lipid structure(s) is/are transported from the ER to the
chloroplast both radiolabeling and fluorescent studies were used with isolated chloroplasts or
chloroplast envelopes, respectivel,. A previous study used radiolabeled PC and isolated pea
chloroplasts to show that PC can be transferred to chloroplasts in a time- and protein-dependent
102

manner (35). The results of the radiolabel assays with Arabidopsis chloroplasts for the current
study corroborated these previous results. The methods used by the previous paper had similar
issues as the current study with obtaining results above background levels when trying to look
at multiple simultaneously labeled phospholipids. The previous study also recorded transport of
PE as was observed in our results (Figure 4.1) (35). Neither the published results nor this study
could show ATP dependent transport using radiolabeled lipids and isolated chloroplasts which
would be indicative of transport by the TGD1,2,3 complex (35). Two older studies also looked
specifically at the transfer of labeled PC to intact chloroplasts and isolated envelopes (31, 32).
Miquel et al. concluded that while labeled PC can be transferred to both envelope vesicles and
intact chloroplasts with help of a cytosolic factor, the label remains on the outer leaflet of the
outer envelope membrane (32). Andersson et al. then proceeded to show cytosolic
phospholipases are needed in order to produce MGDG from transported PC, but this paper also
had problems with PI and PE being transported along with PC (31). Taken together, based on
the work presented here and previous publications, the use of radiolabeled liposomes incubated
with isolated chloroplasts may not be a sufficiently robust system for measuring details of the
ER to chloroplast lipid transport. While I was able to show differences in PA and PC transport in
WT and tgd1-1 mutant chloroplasts, the amount of radiolabeled PE detected in chloroplasts was
also different between WT and mutant (Figure 4.1). This result in combination with difficulties
obtaining levels of radioactivity above background for transferred lipids make the interpretations
of results of this assay difficult. Perhaps the background issues could be resolved by longer
incubation times as commonly used in other lipid labeling studies, but isolated chloroplasts do
not remain intact long enough during longer incubation periods and were, therefore avoided. It
should also be noted that the results using intact isolated chloroplast are complicated by the
presence of two different transport complexes in the inner and outer envelopes and the fact that
these complexes may or may not transport the same lipid species. Using the tgd4-1 mutant in
conjunction with the tgd1-1 mutant may have been able to help minimize this complication, but
103

the phenotype of tgd4-1 is very weak. This weak phenotype and the problems with getting
results above background levels did not allow me to draw solid conclusions using this approach.
The fluorescent lipid transport assays seemed to provide more reliable results than the
radiolabeling experiments despite the lipid acyl chains being tagged with a bulky NBD
fluorophore. Many studies have been done investigating the biological relevance of using NBD
lipids. While many studies have found the NBD-lipids to act normally within the membrane and
with known protein interactions, others have found the presence of NBD can affect the fluidity
and architecture of membranes (36). For the assays performed here, the NBD-lipids were able
to form stable donor liposomes and levels of transport from the liposomes to acceptor envelope
membranes were low enough that the integrity of acceptor membranes should be maintained for
the short duration of the experiment. For these fluorescent lipid transport assays, there were no
problems with background levels, transport could be measured in real time rather than with end
point analysis, and the two TGD complexes could be distinguished by use of different tgd
mutants. The problem of the substrate for TGD1,2,3 having to first travel through the outer
envelope membrane was also resolved by using envelopes which were isolated in such a way
to get a mixture of vesicles made only of either inner or outer envelopes (Figure 4.2). While the
results of this experiment point towards the transported lipid species being PC for TGD4 and PA
for TGD1,2,3, the use of PS as a control complicates matters. The results show that while PS is
not differentially transported by WT and mutant or influenced by the presence or absence of
ATP, its transport is still protein-dependent (Figures 4.3 and 4.4). The PS transport could be due
to promiscuity of the transport proteins in vitro and limits the interpretations of the rest of the
results obtained by this method.
The studies described here yielded results that illustrate the challenges in identifying a
transported lipid species for the TGD complexes., Clearly, additional methods will need to be
explored for further investigation into the transported lipid species. A full reconstitution of the
complex in the test tube or molecular structures of different forms of the complex may be
104

required to ultimately reveal the nature of the lipid molecules that are transported from the ER to
chloroplasts.

Materials and Methods
Plant Growth
Arabidopsis (Arabidopsis thaliana) were grown on petri plates containing MS medium as
described for organelle isolation in Chapter 2. Pea (Pisum sativum "Little Marvel") plants were
grown in vermiculite at room temperature under a 16h light/8h dark cycle.

Organelle Isolation
Chloroplast and ER isolation from Arabidopsis were carried out as described in Chapter 2. For
organelle isolation from pea the following adjustments were made. Two-week-old pea seedlings
were harvested for both ER and chloroplast extraction, and the ER extraction was carried out as
for Arabidopsis. For pea chloroplast isolation, the pelleting steps were carried out at 1500 x g,
and the Percoll gradient step was carried out at 8000 x g. Chloroplast envelope extraction was
carried out with the same procedures for both Arabidopsis and pea. The method for envelope
extraction was developed combining techniques as described in (13, 37, 38), and all steps were
done at 4°C. The chloroplasts were initially resuspended in hypertonic 0.6 M sucrose in TE
buffer (10 mM tricine pH 7.5 KOH, 2 mM EDTA) in order to separate the outer and inner
envelope membranes before breakage. The chloroplasts were then broken by 10 strokes with a
Dounce homogenizer. This method was chosen over the common freeze-thaw method in order
to ensure the integrity of protein complexes such as the TGD complex. The thylakoids were
removed by three 5 min centrifugations at 1500 x g after each of which the supernatant was
maintained. The supernatant was then subjected to a 30 min centrifugation at 100,000 x g in
order to obtain a crude envelope membrane pellet. The pellet was resuspended in 100 μL 0.2 M
sucrose in TE buffer and loaded onto a step sucrose gradient composed of 750 μL 0.45M
105

sucrose in TE buffer and 750 μL 1 M sucrose in TE buffer. The gradient was centrifuged at
100,000 x g for 2 hours, and the faint yellow interface was collected into a new tube and diluted
with 1 mL TE buffer. The final envelope membrane pellet was obtained by centrifuging the
suspension at 100,000 x g for 1 hour, and the pellet was resuspended in incubation buffer (330
mM sorbitol, 50 mM HEPES pH 8.0 KOH, 10 mM KCl, 2.5 mM MgCl 2, 1 mM ATP) for use in
transport assays.

Isolation of Cytosol Fraction
The cytosol fraction was isolated from pea seedlings according to (39) using 10 mL of
homogenization buffer instead of 50 mL.

Variations of Incubation Buffer for Both Radiolabeled and Flourescent Lipid Transport
Assays
The standard incubation buffer used in the transport assays was 330 mM sorbitol, 50 mM
HEPES pH 8.0 KOH, 10 mM KCl, 2.5 mM MgCl2, 1 mM ATP. Some assays used pH 7.3 buffer,
but all of the buffers with a variable ingredient were at pH 8.0. When cytosol fraction was used,
5 μL of the cytosol was added in place of 5 μL of water when preparing the buffer. When UDPGal was added, it was done so at either 2 mM, 6 mM, or 20 mM concentrations in order to
determine the effect of increasing concentrations. Apyrase was used in the incubation buffer at
0.04 U/μL, and ProteinaseK was used at 200 μg/mL.

Donor Vesicle Preparation for Radiolabeled Lipid Transport Assays
American Radiolabeled Chemicals, Inc. was the supplier for all radioactive material. For half of
the assays using

14

C-acetate labeled ER microsomes, ten 7-day-old pea seedlings were cut at

the stem while under water, and each seedling was placed in 900 μL water with 100 μL of 57
mCi/mmol acetic acid [1,2-14C] sodium salt in ethanol (10 μCi) for 8 h. The labeled seedlings
106

were then mixed with the leaves of 10-15 unlabeled pea seedling and used for ER microsome
preparation as described above. The labeled seedlings were mixed with unlabeled seedlings for
ER microsome preparation in order to increase total yield of ER. For the other half of assays
using

14

C-acetate labeled ER microsomes, cut leaves from 7-day-old pea seedlings were

incubated in 5 mL of 20 mM MES pH5.5, 0.01% Tween20 containing 250 μCi of 50mCi/mmol
acetic acid sodium salt [1-14C] for 1.75 hours with light and shaking. The leaf slices were then
combined with leaves from 30 unlabeled seedlings and used for ER microsome preparation as
described above.
To make ER microsomes labeled with commercially available

14

C-phospholipids, ER

microsomes extracted from either pea or Arabidopsis seedlings were centrifuged at 100,000 x g
for 10 min at 4°C. The pellet was resuspended in 1 mL of the desired incubation buffer, and 0.1
μCi of phosphatidic acid, L-α-dioleoyl [2-oleoyl-1-14C] sodium salt and/or phosphatidylcholine, Lα-dioleoyl [dioleoyl-1-14C] per μg of protein equivalents were added. The suspension was
sonicated 3 x 10 seconds at power level 1.0 (3 W), and then centrifuged at 100,000 x g for 10
min at 4°C. The supernatant was removed and subjected to scintillation counting in order to
determine the success of label incorporation. The pellet was resuspended in 1 mL of incubation
buffer to wash and then centrifuged at 100,000 x g for 10 min at 4°C. This final pellet was
resuspended in the desired variation of incubation buffer at 0.5 μg protein equivalents per μL of
buffer.
To prepare microsomes with

32

P-labeled lipids, cut leaves from 7-day-old pea seedlings

were incubated in 5 mL of 20 mM MES pH 5.5, 0.01% Tween20 containing 1000 μCi of
150mCi/mL orthophosphoric acid [32P] for 1.75 hours with light and shaking. The leaf slices
were then combined with leaves from 30 unlabeled seedlings and used for ER microsome
preparation as described above.
For assays using liposomes prepared from lipids extracted from

14

C-acetate labeled pea

seedlings, cut leaves from 7-day-old pea seedlings were incubated in 500 μL of 20mM MES pH
107

5.5, 0.01% Tween20 containing 250 μCi of 57 mCi/mmol acetic acid sodium salt [1-14C] for 2
hours with light and shaking. All excess liquid was then removed from the leaf slices, and 900
μL of methanol:chloroform:formic acid (2:1:0.1 v/v) was added and vortexed in order to extract
lipids. The lipid extract was moved to a fresh tube, and 450 μL of 0.2 M H3PO4, 1 M KCl was
added. The mixture was vortexed and centrifuged at max speed in a microfuge for 1 min. The
bottom organic layer was removed to a glass tube and dried under a nitrogen gas stream. The
lipids were then resuspended in chloroform for loading on a TLC for polar lipid separation with
acetone:toluene:water (91:30:7 v/v/v). PA, PE+PI, and PC were then scraped from the plate and
extracted as described for the lipids used for positional analysis in Chapter 2. Once extracted,
the lipids were dried under a nitrogen gas stream and placed under vacuum for 30-40 min in
order to remove any residual solvent. Then 1 mL of the desired incubation buffer was added,
the lipids were sonicated with a probe 3 x 10 seconds at power level 1.0 (3 W).

Radiolabeled Lipid Transport Assays
Each assay which included some form of labeled ER microsomes used 100 μL of labeled ER
(50 μg protein equivalents) and 50 μL of chloroplasts (50 μg chlorophyll equivalents) all in
incubation buffer. The samples were incubated at room temperature in the light for 30 min. After
30 min, the suspension was loaded onto a 1 mL 40% Percoll cushion in import buffer (330 mM
sorbitol, 50 mM HEPES pH8.0 KOH) and centrifuged in a 4° microfuge at 3000 rpm for 5 min.
Each pellet was extracted with 800 μL 2 methanol: 1 chloroform and vortexed. The samples
were centrifuged in a microfuge at max speed for 1 min, and the extract was transferred to a
cuvette for measuring absorbance at 652 nm. Chlorophyll equivalents were determined with the
same equation used for the chloroplast extraction protocol, and equal chlorophyll equivalents for
each sample were transferred to new tubes. The equivalent of the solvent volume of 0.2 M
H3PO4, 1 M KCl was added to each tube. Samples of 100 μL of labeled ER (50 μg protein
equivalents) were also extracted with 800 μL 2 methanol: 1 chloroform and then 600 μL 0.2 M
108

H3PO4, 1 M KCl to use as controls for total amounts of radiolabel available for transport. All
samples were vortexed and centrifuged at max speed in a microfuge for 1 min. The bottom
organic layer was moved to a fresh tube for each sample. The samples were dried down and
resuspended in chloroform for loading onto a TLC plate. Polar lipids were separated as
described above. PA, MGDG, PG, DGDG, SQDG, PE+PI, and PC bands were scraped into
scintillation vials, 1 mL of scintillation fluid was added to each vial, and the levels of radioactivity
for each lipid were counted.

Fluorescent Lipid Transport Assays
Donor liposomes were a mix of 25% each of N-Rh-PE, dioleoyl-PA or NBD-PA, dioleoyl-PC or
NBD-PC, and dioleoyl-PS or NBD-PS. Only one type of lipid was labeled with NBD at a time.
The liposomes were prepared by drying 10 nmol of total lipid under a nitrogen gas stream and
placing them under vacuum for 30-45 min. The lipids were then solubilized in 30 μL of 100%
ethanol and injected into 1 mL of cold incubation buffer. The liposomes as well as the acceptor
chloroplast envelope membranes were kept on ice until measurement. The chloroplast envelope
membranes were used at 25 μg protein equivalents per mL of incubation buffer. The fluorescent
quality of the liposomes was tested by recording an emission spectra for 200μL of each set of
quenched liposomes at excitation 450 nm and emission 485 - 585 nm followed by a second
emission spectra with 20 μL of 10% Triton X-100 added to break the liposomes for maximum
separation for quencher and, therefore, maximum fluorescence. Once quality of the liposomes
was assured, time-based measurements were taken at excitation 450 nm and emission 530 nm.
For each sample, 200 μL of liposomes were measured for 100 seconds to establish a baseline
fluorescence value. Then 20 μL of envelope membranes were carefully added with mixing by
pipetting up and down 5 times, and measurements were taken for another 200 seconds. 20 μL
of 10% Triton X-100 were then added to break the liposomes and to provide a total fluorescence
value after another 120 seconds.
109

REFERENCES

110

REFERENCES

1.

Stroebel D, Choquet Y, Popot JL, & Picot D (2003) An atypical haem in the cytochrome
b(6)f complex. Nature 426(6965):413-418.

2.

Jordan P, et al. (2001) Three-dimensional structure of cyanobacterial photosystem I at
2.5 A resolution. Nature 411(6840):909-917.

3.

Liu Z, et al. (2004) Crystal structure of spinach major light-harvesting complex at 2.72 A
resolution. Nature 428(6980):287-292.

4.

Loll B, Kern J, Saenger W, Zouni A, & Biesiadka J (2005) Towards complete cofactor
arrangement in the 3.0 A resolution structure of photosystem II. Nature 438(7070):10401044.

5.

Jones MR (2007) Lipids in photosynthetic reaction centres: structural roles and
functional holes. Prog Lipid Res 46(1):56-87.

6.

Holzl G, et al. (2009) The role of diglycosyl lipids in photosynthesis and membrane lipid
homeostasis in Arabidopsis. Plant Physiol 150(3):1147-1159.

7.

Prinz WA (2014) Bridging the gap: Membrane contact sites in signaling, metabolism, and
organelle dynamics. J Cell Biol 205(6):759-769.

8.

Awai K, Xu C, Tamot B, & Benning C (2006) A phosphatidic acid-binding protein of the
chloroplast inner envelope membrane involved in lipid trafficking. Proc Natl Acad Sci U S
A 103(28):10817-10822.

9.

Fan
J,
Zhai
Z,
Yan
C,
&
Xu
C
(2015)
Arabidopsis
TRIGALACTOSYLDIACYLGLYCEROL5 Interacts with TGD1, TGD2, and TGD4 to
Facilitate Lipid Transfer from the Endoplasmic Reticulum to Plastids. Plant Cell.

10.

Lu B, Xu C, Awai K, Jones AD, & Benning C (2007) A small ATPase protein of
Arabidopsis, TGD3, involved in chloroplast lipid import. J Biol Chem 282(49):3594535953.

11.

Xu C, Fan J, Cornish AJ, & Benning C (2008) Lipid trafficking between the endoplasmic
reticulum and the plastid in Arabidopsis requires the extraplastidic TGD4 protein. Plant
Cell 20(8):2190-2204.

12.

Xu CC, Moellering ER, Muthan B, Fan JL, & Benning C (2010) Lipid transport mediated
by Arabidopsis TGD proteins is unidirectional from the endoplasmic reticulum to the
plastid. Plant Cell Physiol 51(6):1019-1028.

13.

Roston RL, Gao J, Murcha MW, Whelan J, & Benning C (2012) TGD1, -2, and -3
proteins involved in lipid trafficking form ATP-binding cassette (ABC) transporter with
multiple substrate-binding proteins. J Biol Chem 287(25):21406-21415.

111

14.

Xu C, Fan J, Riekhof W, Froehlich JE, & Benning C (2003) A permease-like protein
involved in ER to thylakoid lipid transfer in Arabidopsis. EMBO J 22(10):2370-2379.

15.

Roston R, Gao J, Xu C, & Benning C (2011) Arabidopsis chloroplast lipid transport
protein TGD2 disrupts membranes and is part of a large complex. Plant J 66(5):759-769.

16.

Frentzen M (1986) Biosynthesis and Desaturation of the Different Diacylglycerol
Moieties in Higher-Plants. J Plant Physiol 124(3-4):193-209.

17.

Frentzen M, Heinz E, McKeon TA, & Stumpf PK (1983) Specificities and selectivities of
glycerol-3-phosphate
acyltransferase
and
monoacylglycerol-3-phosphate
acyltransferase from pea and spinach chloroplasts. Eur J Biochem 129(3):629-636.

18.

Hurlock AK, Roston RL, Wang K, & Benning C (2014) Lipid trafficking in plant cells.
Traffic 15(9):915-932.

19.

Wang Z, Anderson NS, & Benning C (2013) The phosphatidic acid binding site of the
Arabidopsis Trigalactosyldiacylglycerol 4 (TGD4) protein required for lipid import into
chloroplasts. J Biol Chem 288(7):4763-4771.

20.

Wang Z, Xu C, & Benning C (2012) TGD4 involved in endoplasmic reticulum-tochloroplast lipid trafficking is a phosphatidic acid binding protein. Plant J 70(4):614-623.

21.

Lu BB & Benning C (2009) A 25-Amino Acid Sequence of the Arabidopsis TGD2 Protein
Is Sufficient for Specific Binding of Phosphatidic Acid. J Biol Chem 284(26):1742017427.

22.

Andersson MX & Dörmann P (2009) Chloroplast Membrane Lipid Biosynthesis and
Transport. The Chloroplast, Plant Cell Monographs, eds Sandelius A & Aronsson H
(Springer Berlin Heidelberg), Vol 13, pp 125-158.

23.

Dorne AJ, Joyard J, Block MA, & Douce R (1985) Localization of phosphatidylcholine in
outer envelope membrane of spinach chloroplasts. J Cell Biol 100(5):1690-1697.

24.

Mongrand S, Bessoule JJ, & Cassagne C (1997) A re-examination in vivo of the
phosphatidylcholine-galactolipid metabolic relationship during plant lipid biosynthesis.
Biochem J 327 ( Pt 3):853-858.

25.

Mongrand S, Cassagne C, & Bessoule JJ (2000) Import of lyso-phosphatidylcholine into
chloroplasts likely at the origin of eukaryotic plastidial lipids. Plant Physiol 122(3):845852.

26.

Bessoule JJ, Testet E, & Cassagne C (1995) Synthesis of phosphatidylcholine in the
chloroplast envelope after import of lysophosphatidylcholine from endoplasmic reticulum
membranes. Eur J Biochem 228(2):490-497.

27.

Williams JP, Imperial V, Khan MU, & Hodson JN (2000) The role of phosphatidylcholine
in fatty acid exchange and desaturation in Brassica napus L. leaves. Biochem J 349(Pt
1):127-133.

112

28.

Hamilton JA, Bhamidipati SP, Kodali DR, & Small DM (1991) The interfacial
conformation and transbilayer movement of diacylglycerols in phospholipid bilayers. J
Biol Chem 266(2):1177-1186.

29.

Mongrand S, Bessoule JJ, Cabantous F, & Cassagne C (1998) The C-16 : 3/C-18 : 3
fatty acid balance in photosynthetic tissues from 468 plant species. Phytochemistry
49(4):1049-1064.

30.

Xu C, Fan J, Froehlich JE, Awai K, & Benning C (2005) Mutation of the TGD1
chloroplast envelope protein affects phosphatidate metabolism in Arabidopsis. Plant Cell
17(11):3094-3110.

31.

Andersson MX, Kjellberg JM, & Sandelius AS (2004) The involvement of cytosolic
lipases in converting phosphatidyl choline to substrate for galactolipid synthesis in the
chloroplast envelope. Biochim Biophys Acta 1684(1-3):46-53.

32.

Miquel M, et al. (1988) Protein-Mediated Transfer of Phosphatidylcholine from
Liposomes to Spinach Chloroplast Envelope Membranes. Biochim Biophys Acta
937(2):219-228.

33.

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

34.

Shen J, et al. (2013) Organelle pH in the Arabidopsis Endomembrane System. Mol Plant
6(5):1419-1437.

35.

Yin C, Andersson MX, Zhang H, & Aronsson H (2015) Phosphatidylcholine is transferred
from chemically-defined liposomes to chloroplasts through proteins of the chloroplast
outer envelope membrane. FEBS Lett 589(1):177-181.

36.

Haldar S & Chattopadhyay A (2013) Application of NBD-Labeled Lipids in Membrane
and Cell Biology. Fluorescent Methods to Study Biological Membranes, eds Mély Y &
Duportail G (Springer Berlin Heidelberg, Berlin, Heidelberg), pp 37-50.

37.

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 U S A 78(6):3595-3599.

38.

Keegstra K & Yousif AE (1986) Isolation and Characterization of Chloroplast Envelope
Membranes. Method Enzymol 118:316-325.

39.

Andersson MX, Kjellberg JM, & Sandelius AS (2001) Chloroplast biogenesis. Regulation
of lipid transport to the thylakoid in chloroplasts isolated from expanding and fully
expanded leaves of pea. Plant Physiol 127(1):184-193.

113

CHAPTER 5
Conclusions

114

Understanding plant lipid metabolism is not only important for gaining a comprehensive
picture of plant biology and biochemistry ranging from cellular structure and signaling to energy
storage and conversion, but it is also necessary in order to engineer plants for better nutrition
and biofuel production. The effort to advance our understanding of such a complex process
requires a combination of many approaches. The work presented here combines both in vivo
and in vitro methods in order to investigate some of the areas of plant lipid metabolism in which
our knowledge is incomplete, including lipid transport between the endoplasmic reticulum (ER)
and the chloroplast, acyl editing, and how lipid composition affects membrane stability.
The in vivo studies made use of a unique type of desaturase, a Δ6 desaturase, which
Arabidopsis is normally lacking. A Δ6 desaturase incorporates a double bond at the 6th carbon
position from the carboxyl end of the acyl group. Arabidopsis is known to have Δ9, Δ12, and
Δ15 desaturases, so the addition of a Δ6 desaturase was expected to add a subtle but readily
detectable tag to membrane lipids, which I anticipated would only minimally affect the natural
composition or function of the respective membrane or the overall balance of lipid metabolism.
A well-characterized Δ6 desaturase from Physcomitrella patens (Δ6D) was chosen because of
its location within the cell, the ER, and its known substrate specificity for the sn-2 position of the
glycerol backbone of phosphatidylcholine (PC). This location and specificity provided a unique
starting position for a "tag and track" approach as most in vivo radiolabel approaches start
labeling at fatty acid synthesis within the chloroplast or with pre-labeled fatty acids. With the
goal of probing lipid transport and metabolism with increased sensitivity, several mutants of lipid
metabolism were used for the "tag and track" approach described in Chapter 2. While the lipid
phenotypes of these plants verified key findings of previous studies of plant lipid metabolism in
both wild type (WT) and mutant plants validating the method, several of the results were
unexpected and provided novel insights into the complexities of plant lipid metabolism. Since
phosphatidylglycerol (PG) is thought to be made nearly exclusively from plastid precursors (1), it
was interesting to detect Δ6 acyl groups, which had to have originated in the ER, in PG at much
115

higher levels than expected and in both the sn-1 and sn-2 positions (Figure 2.5A&B). Another
striking observation was that the ratio of the Δ6 acyl groups in the sn-1 and sn-2 positions
changed between PC in the ER and the envelope membranes and genuine thylakoid lipids
(Figure 2.4B). This result combined with confirmation of an ER-derived Δ6 acyl group on the
same lipid molecule as a plastid derived 16:3 (carbon #: double bond #) acyl group provided
evidence for extensive acyl editing of chloroplast lipids (Figures 2.5C & 2.6). Knowing that acyl
editing occurs both outside (2) and inside the chloroplast (Chapter 2 and (3)) enriches our view
of lipid metabolism, and it can help us better explain results and formulate new hypotheses for
other studies involving these pathways.
New discoveries related to lipid metabolism may add complexity to our current models
as seen from the results described in Chapter 2, but they will get us closer to fully understanding
the dynamic processes governing plant lipid metabolism. Currently, designing experiments for
study of plant lipid metabolism can be difficult due to many unknown factors which can affect the
results. Experiments involving isolated chloroplasts are designed in order to remove some of
these unknown factors because there is only one organelle to consider. It provides the
opportunity to study chloroplasts in a controlled environment and focus on how its metabolism
functions under the chosen conditions. Unfortunately, even with this seemingly well-controlled
system, some variability is still to be expected. Chapter 3 focused on how differences in lipid
composition could affect these types of experiments and shows apparent pitfalls of studying lipid
metabolism with chloroplasts isolated from different lipid mutants. Because fatty acid synthesis
and one of the two major glycerolipid synthesis pathways occur in the chloroplast, it is not
uncommon to do experiments, which compare isolated chloroplasts from WT and lipid
metabolism mutant plants. If the distinct lipid composition of those mutant chloroplasts can alter
the conditions of these experiments, the design of experiments and interpretation of the results
must be adjusted accordingly. I found that both the presence of oligogalactolipids and the
double bond index and/or 16C to 18C acyl chain length ratio seemed to have an effect on either
116

isolated chloroplasts' resistance to breakage or the ability to reseal the membrane (Figures
3.1A, 3.2C, 3.4A&B). Differences in chloroplast breakage or the ability to reseal the membrane
could affect the composition of the buffer and substrate accessibility of chloroplast proteins. For
example, a broken chloroplast could give a stromal protein access to the contents of the outer
envelope or an outer envelope protein access to the contents of the thylakoid membrane.
Therefore, with the stability differences shown between isolated chloroplasts of different lipid
composition, these types of possibilities need to be taken into consideration when interpreting
the results. Although some of these aspects of studies with isolated chloroplast are concerning,
they provide a novel way to systematically analyze the role of membrane lipid composition on
chloroplast stability that should be pursued further in future studies.
Although the results described in Chapter 3 point out potential issues with comparing
isolated chloroplasts with different lipid compositions, an isolated chloroplast experimental
system could still be useful in determining the transported lipid species from the ER to the
chloroplast as long as the results are interpreted with careful consideration of the Chapter 3
findings. With this in mind, in vitro experiments using donor liposomes or ER microsomes and
acceptor whole chloroplasts or envelopes from WT and lipid transport mutants were undertaken
in order to identify the transported lipid species between the ER and the chloroplast. In
experiments with whole chloroplasts as acceptors, I tested different variations of radiolabeled
donor liposomes or ER microsomes. These experiments showed potential transport of both
phosphatidic acid (PA) and PC (Figure 4.1A), but the amounts of radiolabel detected were
barely above background questioning the efficacy of this assay under the different conditions
tested. The results were further complicated by the fact that the phosphatidylethanolamine+
phosphitidylinositol (PE+PI) fraction also showed possible transport (Figure 4.1A). These are
lipids typically absent from chloroplasts, thereby questioning the specificity of transport
observed in this system.

117

The experiments which used fluorescently labeled donor liposomes with chloroplast
envelope acceptors showed similar results to the radiolabeling experiments with PA and PC. In
particular, reduced transport of these two lipids to the mutant envelopes compared to WT were
promising (Figure 4.4D). However, these results were ultimately inconclusive because
phosphatidylserine (PS), a lipid not known to move to the chloroplast, also showed proteindependent transport to envelopes; although, there were no differences between mutant and
wild-type chloroplasts (Figures 4.3C & 4.4D). Based on these data, I could not rule out that the
lipid transport systems in the chloroplast may lack specificity in vitro resulting in experiments
which do not reflect the situation in the cell, possibly, because other critical factors are missing.
Summarizing the results of these experiments, my data are consistent with the hypothesis that
PC is first transported to the outer envelope by the TGD4 dimer and that PA is then transported
to the inner envelope by the TGD1,2,3 complex. This would require the conversion of PC to PA
by a phospholipase D, which has not yet been identified. Thus, aside from further development
of a robust in vitro transport assay, one key experiment to support this hypothesis would be to
identify a phospholipase D in the outer envelope membrane of chloroplasts. Given the large
number of predicted lipases in the plant genome (4), identifying such a lipase may not be an
easy task, particularly if there are redundant enzymes .
All of the methods presented here have the potential for continued use in the field of
plant lipid research. While the in vitro methods require further optimization, the results are
promising. Perhaps with more study of the donor liposomes and better understanding of their
biophysical properties over time, the transport assays would show solidly conclusive results.
The in vivo "tag and track" method is ready to use as is. It will have value when used in other
mutant backgrounds of interest, and it may be able to help researchers answer questions that
cannot quite be addressed with other approaches such as radiolabeling.
To ultimately understand the nature of the lipid species transported by the TGD complex
and its mechanistic function, structural analysis of the protein complex with trapped
118

intermediates might be necessary in the future. The TDG complexes are relatively stable during
isolation (5, 6), and cryo-electron microscopy techniques combined with cross-linking mass
spectrometry are particularly applicable to large complexes (7). These techniques have
advanced to a level that they may allow us to address this problem in the near future (7).

119

REFERENCES

120

REFERENCES

1.

Browse J, Warwick N, Somerville CR, & Slack CR (1986) Fluxes through the prokaryotic
and eukaryotic pathways of lipid synthesis in the '16:3' plant Arabidopsis thaliana.
Biochem J 235(1):25-31.

2.

Bates PD, Ohlrogge JB, & Pollard M (2007) Incorporation of newly synthesized fatty
acids into cytosolic glycerolipids in pea leaves occurs via acyl editing. J Biol Chem
282(43):31206-31216.

3.

Tjellstrom H, Yang Z, Allen DK, & Ohlrogge JB (2012) Rapid kinetic labeling of
Arabidopsis cell suspension cultures: implications for models of lipid export from
plastids. Plant Physiol 158(2):601-611.

4.

Wang K, Froehlich JE, Zienkiewicz A, Hersh HL, & Benning C (2017) A Plastid
Phosphatidylglycerol Lipase Contributes to the Export of Acyl Groups from Plastids for
Seed Oil Biosynthesis. Plant Cell.

5.

Roston RL, Gao J, Murcha MW, Whelan J, & Benning C (2012) TGD1, -2, and -3
proteins involved in lipid trafficking form ATP-binding cassette (ABC) transporter with
multiple substrate-binding proteins. J Biol Chem 287(25):21406-21415.

6.

Wang Z, Anderson NS, & Benning C (2013) The phosphatidic acid binding site of the
Arabidopsis Trigalactosyldiacylglycerol 4 (TGD4) protein required for lipid import into
chloroplasts. J Biol Chem 288(7):4763-4771.

7.

Schmidt C & Urlaub H (2017) Combining cryo-electron microscopy (cryo-EM) and crosslinking mass spectrometry (CX-MS) for structural elucidation of large protein assemblies.
Curr. Opin. Struct. Biol. 46(Supplement C):157-168.

121