JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M704063200 on October 15, 2007

J. Biol. Chem., Vol. 282, Issue 49, 35945-35953, December 7, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
282/49/35945    most recent
M704063200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lu, B.
Right arrow Articles by Benning, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lu, B.
Right arrow Articles by Benning, C.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

A Small ATPase Protein of Arabidopsis, TGD3, Involved in Chloroplast Lipid Import*Formula

Binbin Lu{ddagger}§, Changcheng Xu{ddagger}, Koichiro Awai{ddagger}1, A. Daniel Jones{ddagger}, and Christoph Benning{ddagger}2

From the {ddagger}Department of Biochemistry and Molecular Biology, the Department of Chemistry, and the §United States Department of Energy Plant Research Laboratory, Michigan State University, East Lansing, Michigan 48824-1312

Received for publication, May 16, 2007 , and in revised form, September 21, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Polar lipid trafficking is essential in eukaryotic cells as membranes of lipid assembly are often distinct from final destination membranes. A striking example is the biogenesis of the photosynthetic membranes (thylakoids) in plastids of plants. Lipid biosynthetic enzymes at the endoplasmic reticulum and the inner and outer plastid envelope membranes are involved. This compartmentalization requires extensive lipid trafficking. Mutants of Arabidopsis are available that are disrupted in the incorporation of endoplasmic reticulum-derived lipid precursors into thylakoid lipids. Two proteins affected in two of these mutants, trigalactosyldiacylglycerol 1 (TGD1) and TGD2, encode the permease and substrate binding component, respectively, of a proposed lipid translocator at the inner chloroplast envelope membrane. Here we describe a third protein of Arabidopsis, TGD3, a small ATPase proposed to be part of this translocator. As in the tgd1 and tgd2 mutants, triacylglycerols and trigalactolipids accumulate in a tgd3 mutant carrying a T-DNA insertion just 5' of the TGD3 coding region. The TGD3 protein shows basal ATPase activity and is localized inside the chloroplast beyond the inner chloroplast envelope membrane. Proteins orthologous to TGD1, -2, and -3 are predicted to be present in Gram-bacteria, and the respective genes are organized in operons suggesting a common biochemical role for the gene products. Based on the current analysis, it is hypothesized that TGD3 is the missing ATPase component of a lipid transporter involving TGD1 and TGD2 required for the biosynthesis of ER-derived thylakoid lipids in Arabidopsis.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Interorganellar non-vesicular lipid trafficking in plant cells is required for the biogenesis of the characteristic plant cell organelle, the chloroplast (1). The predominant polar lipids found in the photosynthetic membranes (thylakoids) of chloroplasts are the galactoglycerolipids monogalactosyldiacylglycerol (MGDG)3 and digalactosyldiacylglycerol (DGDG) (2). Their final assembly from diacylglycerol and UDP-galactose requires two lipid galactosyltransferases, which in the model plant Arabidopsis are designated MGD1 and DGD1 (3). Of these two enzymes MGD1 is associated with the outside of the inner chloroplast envelope membrane (4) and is responsible for the bulk of MGDG formation (5), whereas DGD1 is located on the outside of the outer chloroplast envelope membrane (6) and is required for the bulk of DGDG biosynthesis (7). The diacylglycerol moiety of the galactoglycerolipids is either synthesized de novo in the plastid or it is assembled at the endoplasmic reticulum (ER) and needs to be imported into the plastid (8). Understanding this complexity of subcellular organization of glycerolipid biosynthesis in plants and the underlying lipid trafficking phenomena still poses some of the biggest challenges of modern plant biochemistry.

The recent isolation and characterization of genetic suppressors of the DGDG-deficient dgd1 mutant of Arabidopsis has provided the first components, TGD1 and TGD2, of a potential lipid transport complex associated with the inner chloroplast envelope membrane in Arabidopsis (4, 9, 10). The TGD1 protein is similar to the permease component of bacterial ATP binding cassette (ABC) transporters, and the Arabidopsis tgd1 mutant shows a complex lipid phenotype; that is, the accumulation of oligogalactolipids and triacylglycerols, a disruption of the assembly of galactolipids from ER-derived precursors, and the elevation of phosphatidic acid (PtdOH) levels (4, 10). The name of the mutants, tgd (trigalactosyldiacylglycerol), refers to the characteristic accumulation of oligogalactolipids. Presumably, this is a secondary phenotype due to the activation of a processive galactosyl transferase of unknown function, for which the gene has not yet been identified. The accumulation of PtdOH and reduced incorporation of labeled PtdOH into galactoglycerolipids by isolated tgd1 chloroplasts led to the suggestion that TGD1 is involved in the transport of PtdOH through the inner chloroplast envelope membrane (4). The TGD2 protein is similar to membrane-tethered substrate-binding proteins associated with bacterial ABC transporters (9). Although not as extensively characterized as the tgd1 mutant, the Arabidopsis tgd1 and tgd2 mutants have a nearly identical lipid phenotype in all aspects tested. Moreover, the recombinant TGD2 protein lacking the single membrane-spanning domain specifically binds PtdOH (9). The nearly identical phenotypes of the two mutants and the organization of bacterial orthologs of TGD1 and TGD2 in operons (Fig. 1A) strongly suggest that both are involved in a common biochemical or cell biological process. Moreover, the specific binding of PtdOH to TGD2 corroborates the hypothesis that the two proteins are components of a PtdOH transporter at the inner chloroplast envelope.


Figure 1
View larger version (40K):
[in this window]
[in a new window]

  		FIGURE 1.
Isolation of the TGD3 gene by reverse genetics. A, linear representation of the Arabidopsis TGD homologs in different bacterial species. The number designations (1-3) correspond to open reading frames presumed to encode homologs of Arabidopsis TGD1, TGD2, and TGD3 proteins, respectively. Letters refer to the bacterial open reading frame designations in the respective operons. Species and GenBankTM protein accession numbers for TGD3 orthologs: E. coli (ZP_00728201), Pseudomonas putida (ZP_01714441), Xylella fastidiosa (NP_779845), Synechococcus sp. JA-2-3B'a (YP_477326), Arabidopsis thaliana (NP_564850). B, an alignment of the TGD sequence with various ABC-transporter associated ATPases for which crystal structures are available: HlyB (PDB code 1MT0), MalK (PDB code 1G29), MJ1267 (PDB code 1G9X), and HisP (PDB code 1B0U). Open boxes mark conserved residues, and black boxes indicate identical residues. The relative positions of the main functional sites are assigned, and phenylalanine residue Phe94 of TGD3, which was mutated to alanine, corresponds to the A-loop conserved aromatic residue as indicated.

 
Unlike other eukaryotic ABC transporters or the MsbA lipid transporter from Escherichia coli (11), the TGD1 protein lacks an ATP binding domain. However, the Arabidopsis genome contains 26 genes encoding putative small ATPases with either one or two nucleotide binding domains (12). We hypothesized that one of these could encode a small ATPase potentially associated with the TGD1 protein. Here we describe the identification and biochemical characterization of TGD3, the small ATPase protein proposed to be associated with TGD1 and TGD2.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Plant Materials and Growth Conditions—Arabidopsis thaliana wild type and tgd1 and tgd2 mutant plants (9, 10) were of the ecotype Columbia-2 (Col-2). The following T-DNA insertion lines were made available by the Salk Institute Genome Analysis Laboratory (SIGnAL) and were obtained from the Arabidopsis Biological Resource Center at Ohio State University: SALK_027270 At4g33460, SALK_129048 At5g60790, SALK_029696 At5g09930, SALK_113472 At5g64840, SALK_116866 At5g14100, SALK_055996 At1g32500, and SALK_040335 At1g65410. Homozygous mutant lines were screened and used for all experiments. Seeds were surface-sterilized and grown on 0.8% (w/v) agar-solidified Murashige-Skoog medium (13) supplemented with 1% (w/v) sucrose for 5 days. Wild-type and mutant seedlings of similar developmental stage were then transferred to a fresh Murashige-Skoog agar plate with 1% sucrose and grown for three additional weeks before lipid analysis.

Genotyping and Complementation—The TGD3 open reading frame was isolated by reverse transcriptase-PCR (standard conditions) from mRNA preparations using RNeasy and Omniscript kits (Qiagen, Valencia, CA) with the primers 5'-ACGGTACCATGCTTTCGTTATCATGCTC-3' and 5'-CTGGTACCCTAGTATCTGATTGGTCCAT-3'. The PCR products were ligated into pGEM-Teasy (Promega, Madison, WI) and sequenced. The resulting plasmids were digested with KpnI and inserted into pCAMBIA1300 (4) followed by transformation into Agrobacterium. Plants were transformed by the floraldip method (14) and screened for resistance to hygromycin B (25 µg/ml) on agar-solidified Murashige-Skoog medium. For an estimation of TGD3 transcript levels by semiquantitative reverse transcriptase-PCR, RNA was isolated from Col-2 wild-type and tgd3 mutant plants, and reverse transcription was done using oligo(dT)12-18 primers (Invitrogen). The TGD3-specific primers mentioned above were used for the PCR reaction. The abundance of ubiquitin (UBQ10) was tested for control purposes. The following UBQ10-specific primers were used: 5'-TCAATTCTCTCTACCGTGATCAAGATGCA-3' and 5'-GTGTCAGAACTCTCCACCTCAAGAGTA-3'.

For genotyping during the complementation analysis, DNA was prepared from 4-week-old plants, and the following primers were used: T-DNA left border primer (LB), 5'-GCGTGGACCGCTTGCTGCAAC-3'; P1, 5'-AACTGATTGGGACGAGTATC-3'; P2, 5'-GCTATGCAACAGCAAGAGAC-3'. The PCR conditions were 94 °C 3 min followed by 35 cycles at 94 °C for 0.5 min, 54 °C for 0.5 min, and 72 °C for 0.5 min followed by a final step of 5 min at 72 °C.

Lipid Assays—Lipids were extracted from 4-week-old Col-2 wild-type and tgd3 mutant seedlings grown on Murashige-Skoog medium with 1% sucrose. Fatty acid methyl esters were prepared as previously described (15) and quantified by gas chromatography according to Rossak et al. (16). Polar lipids were analyzed on activated ammonium sulfate-impregnated (15) silica gel TLC plates (Si250PA; Mallinckrodt Baker) developed with chloroform/methanol/acetic acid/water (85/25/10/4, v/v). Neutral lipids were separated on untreated TLC plates developed with petroleum ether/ether/acetic acid (80/20/1, v/v). Lipids were visualized by brief exposure to iodine vapor or staining with {alpha}-naphthol to detect glycolipids (17). The fatty acid compositions at the sn-2 position of individual lipids were determined using Rhizopus lipase (Sigma) digestion according to Siebertz and Heinz (18) with modifications as described in Miquel et al. (19). Fatty acid methyl esters were formed from lyso-lipids, and the fatty acid methyl esters were quantified by gas chromatography.

Mass Spectrometry—Lipid extracts were characterized using liquid chromatography/mass spectrometry (MS) on a Waters LCT Premier time-of-flight mass spectrometer equipped with Shimadzu LC-20AD pumps and SIL-5000 autosampler. Extracts were analyzed using V-mode operation and electrospray ionization in both positive and negative ion modes similar to described protocols (20) except that, in place of MS/MS spectra, mass spectra were acquired in alternating acquisition functions at low (15 V) and high (75 V) potentials on aperture one to generate spectra with and without in-source collision-induced dissociation. The latter conditions allow for production of characteristic fragment ions that support structure assignments. High performance liquid chromatography separations were performed using a Restek Allure C18 column (1 x 150 mm) and a ternary gradient based on (a) 10 mM aqueous ammonium acetate, (b) methanol, and (c) 2-propanol.

Subcellular Localization Assays—For the generation of a green fluorescent protein fusion construct, the entire coding region of At1g65410 was inserted into the KpnI site of pCAMBIA1300GFP (4) using primers 5'-ACGGTACCATGCTTTCGTTATCATGCTC-3' and 5'-CTGGTACCGTATCTGATTGGTCCAT-3'. Transient expression of the construct in tobacco and microscopy of the resulting transgenic leaves was performed as previously described (4). For the in vitro import experiments using pea chloroplasts the same conditions and treatments were used as published for the localization of the MGD1 protein (4).

Recombinant MBP-{Delta}TGD3 Protein Production and Purification—The coding sequence for TGD3 leading to a truncated protein missing 46 N-terminal amino acids (designated as {Delta}TGD3) was PCR-amplified. As template, the pGEM-Teasy plasmid derivative mentioned above was used in combination with primers 5'-GGATCCTGCATAGCTCCACCTCAGAAC-3' and 5'-GTCGACTAGTATCTGATTGGTCCATCG-3'. The fragment was inserted into pGEM-Teasy (Promega) and sequenced. This plasmid served as template for F94A mutagenesis by using the forward/reverse primers and the QuikChange site-directed mutagenesis kit from Stratagene (La Jolla, CA). The forward and reverse primers for F94A are 5'-GTAGAGATGTCTATAAATCGGCGGGGGAGAAACATATCTTG-3' and 5'-CAAGATATGTTTCTCCCCCGCCGATTTATAGACATCTCTAC-3'. The underlined sequences are codons for the mutated residue (alanine). After digestion with BamHI and SalI, either the wild type {Delta}TGD3 or the mutant {Delta}TGD3F94A fragment was inserted 3' of the E. coli malE coding sequence into the pMalc2x vector (New England Biolabs, Ipswich, MA) to generate the maltose-binding protein fusion proteins MBP-{Delta}TGD3 or MBP-{Delta}TGD3F94A. For recombinant protein production, a 500-ml culture was grown at 37 °C, inoculated with 2 ml of an overnight culture. The protein was induced with 0.1 mM isopropyl-β-D-thiogalactopyranoside at an A600 of 0.4-0.6 and incubated at 28 °C for 4 h. The cells were then collected by centrifugation and lysed by sonication in the presence of 1 mg/ml lysozyme. Crude lysate was centrifuged at 18,000 x g for 20 min, and the soluble fraction was loaded onto an amylose column (New England Biolabs). Fusion proteins were purified according to manufacturer's instructions. Protein concentration was determined according to Bradford (21) using bovine serum albumin as a standard. Protein purity was verified by SDS-PAGE. After purification, samples were stored at 4 °C for a few weeks without significant loss of activity.

Assay for ATPase Activity—Purified MBP-{Delta}TGD3 protein was assayed for ATPase activity at a protein concentration of ~0.5-8 µM in 50-µl reaction mixtures containing assay buffer (40 mM Tris-HCl, pH 7.6,150 mM NaCl, 2 mM MgCl2). The MBP-{Delta}TGD3 protein was diluted in assay buffer to a final volume of 25 µl. To initiate the reaction, 25 µl of ATP in the assay buffer ranging from 0 to 2 mM (diluted from a 10 mM stock made in the same assay buffer) were added, giving a final ATP concentration of 0 to 1 mM. ATPase reactions were incubated at 37 °C for the indicated time and stopped by the addition of 50 µl 12% SDS. Released Pi was quantified by a colorimetric method, as described by Chifflet et al. (22) using potassium phosphate as the standard. After the addition of SDS, 100 µl of a solution containing equal volumes of 12% ascorbic acid in 1 M HCl and 2% ammonium molybdate were added, and the samples were incubated at room temperature for 5 min. This was followed by the addition of 150 µl of a solution composed of 2% sodium citrate, 2% sodium meta-arsenite, and 2% acetic acid. After 10 min of incubation at 37 °C, absorbance was measured at 850 nm. The assay was linear in the range of 0-25 nmol of Pi. The activity was corrected for nonenzymatic release of Pi from ATP and for Pi contamination by subtracting absorbance readings obtained in the absence of protein and ATP, respectively. A vector expressing only MBP was used to control for endogenous ATPase activity from E. coli. The kinetic constants were determined by nonlinear least squares fitting to the equation Vo = Vmax[S]/(Km + [S]) using Microcal Origin 7.0 (Microcal Software, Northampton, MA).


Figure 2
View larger version (35K):
[in this window]
[in a new window]

  		FIGURE 2.
Identification of the TGD3 locus (At1g65410). A, structure for the tgd3 mutant allele corresponding to Salk_040335 line. The coding region of At1g65410 is shown as black boxes. T-DNA insertion position and primers (P1, P2, and left T-DNA border primer LB) used for PCR amplification are indicated. B, semiquantitative reverse transcriptase-PCR analysis of TGD3 mRNA levels in wild-type (Col-2) and tgd3 plants. The expression of ubiquitin (UBQ10) in the same samples was used as a control. C, genotypic analysis at the TGD3 locus. Wild-type (Col-2), tgd3 mutant, and two tgd3 homozygous complementation lines expressing the TGD3 cDNA (cTGD3/tgd3) are shown. Primers were the same as in A, and diagnostic DNA fragments are shown with their respective lengths in base pairs. D, lipid phenotype of the different plant lines in C.A section of the thin-layer chromatogram stained for glycolipids is shown.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Identification of TGD3 by Genomics and Reverse Genetics—Bacterial genes involved in the same biochemical process are commonly linked in operons. This fact has been successfully exploited in genome annotation or the identification of individual genes (23). Accordingly, the location of TGD1 and TGD2 encoding homologs in bacterial operons (Fig. 1A) adjacent to putative ABC proteins suggests that these are the ABC components of the respective transport complex. Based on comparative genomics, it seemed likely that Arabidopsis has an ABC protein associated with the TGD1/2 transporter.

An extensive and possibly complete inventory of transport ATPases and non-intrinsic ABC proteins potentially associated with membrane transporters in Arabidopsis has been published (12) listing 26 soluble nucleotide binding domain proteins. Because TGD1 and TGD2 are associated with the inner chloroplast membrane, which is homologous to the bacterial cell membrane, we expected that the respective ABC protein was localized on the inside of the inner chloroplast membrane requiring an N-terminal chloroplast targeting sequence. Nine of the above-mentioned ABC proteins were predicted to be targeted to the plastid using ChloroP (24): At1g32500, At1g65410, At3g10670, At4g04770, At4g33460, At5g09930, At5g14100, At5g60790, At5g64840. For seven (listed under "Experimental Procedures") of these candidate genes, T-DNA inactivation lines (25) were available at the time and were selfed and selected for homozygosity. The homozygous mutant lines were tested for the presence of trigalactosyldiacylglycerol, which is diagnostic for mutants with impaired TGD1 or TGD2 function. Only one inactivation line (SALK_040335) carrying a T-DNA 5' of the presumed ATG start codon of Arabidopsis gene At1g65410 (Fig. 2A) showed accumulation of the oligogalactolipid (Figs. 2D and 3B). This mutant was clearly leaky as the abundance of the respective mRNA estimated by semiquantitative reverse transcriptase-PCR was only partially reduced in the T-DNA line (Fig. 2B). A completely disrupted allele was not available (see also below).

To confirm that the mutant phenotype was not due to secondary mutations, a wild-type cDNA was expressed in the homozygous T-DNA line disrupted in At1g65410. Two transgenic homozygous T-DNA lines lacking the accumulation of the oligogalactolipid are shown in Fig. 2, C and D. The restoration of the wild-type phenotype in the homozygous T-DNA lines by the corresponding wild-type cDNA confirmed that the accumulation of oligogalactolipids in the T-DNA lines was due to the disruption of the At1g65410 locus. From here on, the SALK_040335 line will be referred to as tgd3 mutant line (carrying the here characterized tgd3-1 mutant allele). The At1g65410-encoded protein was previously designated Non-intrinsic ABC Protein 11 (NAP11) (12) and will be referred to here as trigalactosyldiacylglycerol 3 (TGD3) fitting more closely with its function. Comparing the amino acid sequence of TGD3 with proteins encoded by bacterial genomes using BLAST (26), ABC protein-encoding open reading frames adjacent to putative TGD1/2 homologs provided some of the best hits, with putative TGD3 homologs from Cyanobacteria, e.g. Synechococcus (Fig. 1A), showing 52% identity and the E. coli homolog 32% identity.

The tgd3 Mutant Shares a Complex Lipid Phenotype with Other tgd Mutants—It is important to note that all three tgd mutant alleles (tgd1-1, tgd2-1, tgd3-1) analyzed thus far and discussed here are leaky, leading to attenuated phenotypes. More severe impairment of the system as previously shown for TGD1 causes embryo-lethality (4) making fully gene-disrupted mutants currently inaccessible to analysis. If the TGD3 protein indeed is part of a lipid transport complex along with TGD1 and TGD2, mutant lipid phenotypes for all three loci should be similar. Therefore, leaf lipid extracts from the tgd1, tgd2, and the tgd3 lines were compared by thin-layer chromatography of neutral (Fig. 3A) and polar lipids (Fig. 3B). The tgd3 mutant extracts contained a lipid co-migrating with triacylglycerol (Fig. 3A), previously identified in the tgd1 mutant (4) and shown also to accumulate in tgd2 mutant extracts (9). In addition, polar lipid extracts of the tgd3 mutant contained a new lipid staining with {alpha}-naphthol, which is diagnostic for the presence of hexoses (Fig. 3B). This lipid co-chromatographed with authentic trigalactosyldiacylglycerol (TGDG) accumulating in tgd1 and tgd2.


Figure 3
View larger version (81K):
[in this window]
[in a new window]

  		FIGURE 3.
Lipid phenotype of the tgd3 mutant compared with the tgd1, tgd2 mutant, and Col-2 wild type. Total lipids were extracted from 4-week-old seedlings grown on Murashige-Skoog agar plates containing 1% sucrose and separated by thin-layer chromatography. A, thin-layer chromatogram of neutral lipids. Lipids were visualized by exposure to iodine vapor. B, thin-layer chromatogram of polar lipids. Lipids were visualized by sugar-specific {alpha}-naphthol staining. O, origin; PIG, pigments; TAG, triacylglycerol.

 


Figure 4
View larger version (23K):
[in this window]
[in a new window]

  		FIGURE 4.
Liquid chromatography/MS analyses of lipid extracts of tgd3 mutant and Col-2 wild type Arabidopsis leaves. Negative mode electrospray extracted ion chromatograms (XIC) for m/z 1129.6, the [M+acetate]- ion for TGDG 34:6, for extracts of tgd3 mutant (A) and Col-2 wild type (B) leaves. The mass spectrum at retention time 21.32 min for the tgd3 mutant (C) shows additional peaks corresponding to [M+Cl]- and [M+formate]- (unlabeled). Positive mode electrospray extracted ion chromatograms over the range of m/z 890-905 reflect the abundance of triacylglycerols with 54 carbons and various numbers of double bonds for tgd3 mutant (D) and Col-2 wild type (E) leaves. The summed mass spectrum over the peak eluting at 27.87 min (F) shows [M+NH4]+ peaks corresponding to triacylglycerols with 54 carbons and from 3-8 double bonds.

 
Further confirmation of the identities of TGDG and triacylglycerol lipids was provided by liquid chromatography/MS analyses of leaf extracts of Col-2 wild type and tgd3 mutants (Fig. 4, A-C). Extracted ion chromatograms generated in negative ion mode yielded a strong peak at m/z 1129.6 for the tgd3 mutant corresponding to [M+acetate]- for TGDG 34:6, which was the dominant TGDG lipid detected, but minimal amounts in the wild-type extracts. The in-source collision-induced dissociation spectra supported this assignment in the form of a fragment ion at m/z 277 corresponding to the C18:3 fatty acid anion.

Triacylglycerols were characterized using positive ion electrospray ionization, displaying peaks at m/z 892.7 to 902.8 corresponding to [M+NH4]+ of triacylglycerols with 54 carbons and a total number of double bonds ranging from 3 to 8 (Fig. 4F). Extracted ion chromatograms for signals in the range of m/z 890-905 showed about 5-fold greater signal in the tgd3 mutant than the wild type (Fig. 4, D and E). Ion abundances for both TGDG and triacylglycerol lipid peaks were consistent with quantitative analysis of fatty acid methyl esters by gas chromatography coupled to flame ionization detection.

The overall lipid composition of the tgd3 mutant differed from that of the wild type as shown in Fig. 5A. Most notably, the relative amounts of the major chloroplast lipids MGDG and DGDG were decreased, whereas the presumed precursor of galactolipids derived from the ER pathway, phosphatidylcholine (8), was more abundant. Analysis of the fatty acid composition of MGDG and DGDG revealed distinct changes in the fatty acid profiles in tgd3 as compared with the wild type (Fig. 5, B and C). Most notably, fatty acids were generally more saturated, and 18:3 fatty acid content was reduced. Moreover, the tgd3 fatty acid profile changes were very similar to those observed for the tgd1 and tgd2 mutants (9).

In general, the tgd mutants impaired in the ER pathway have an increased 16-carbon-to-18-carbon fatty acid ratio (9) in their galactolipids. This is particularly visible for DGDG, which is to a large extent derived from the ER pathway (27). Accordingly, the 16-carbon-to-18-carbon ratio for the digalactolipid increased from 0.29 to 0.76 in the tgd3 mutant (Fig. 5C). This phenomenon is due to the substrate specificities of the different acyltransferases in the plastid and the ER leading to 18-carbon fatty acids at the sn-1 position and 16-carbon fatty acids at the sn-2 position of the diacylglycerol backbone for plastid-derived lipids. Those lipids derived from the ER pathway carry 18-carbon fatty acids in both positions (27). Positional analysis using Rhizopus lipase (supplemental Fig. 1) confirmed an increase in 16-carbon fatty acids in the sn-2 position of MGDG and DGDG of the tgd3 mutant.


Figure 5
View larger version (15K):
[in this window]
[in a new window]

  		FIGURE 5.
Polar lipid composition and fatty acid content in the tgd3 mutant. A, 4-week-old Col-2 wild type and tgd3 mutant seedlings grown on the same Murashige-Skoog agar plate were analyzed. Five replicates were averaged, and the S.E. are shown. A, polar lipid composition (relative mol%) determined by quantification of fatty acid methyl esters derived from individual lipids. Fatty acid composition of MGDG (B) and DGDG (C). Fatty acids are indicated with the number of carbons:number of double bonds. PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphatidylinositol; SQDG, sulfoquinovosyldiacylglycerol.

 
Overall, tgd3 has a complex lipid phenotype very similar to tgd1 and tgd2 that is consistent with an impairment of the ER pathway of galactolipid biosynthesis. This similarity in phenotypes also suggests that the TGD1, TGD2, and TGD3 proteins are involved in the same biochemical process.

The TGD3 Protein Is Imported into the Chloroplast—The TGD3 sequence has a predicted chloroplast transit peptide. To verify the localization of the protein, the cDNA was C-terminally fused with a green fluorescent protein and transiently expressed in tobacco. The result shown in Fig. 6, A-C, shows the green fluorescent protein fluorescence associated with chloroplasts, which are identified by their chlorophyll fluorescence.

To further narrow the location of TGD3, in vitro translated labeled TGD3 protein was incubated with isolated pea chloroplasts. Two proteases were used to determine the sub-chloroplast localization; thermolysin, a large protease that can only digest cytosol-exposed domains of outer envelope membrane proteins unless detergent such as Triton-X-100 is added (28), and trypsin, which can penetrate the outer envelope membrane but not the inner (29). The result of this analysis is shown in Fig. 6D. The TGD3 preprotein was shortened upon import consistent with the removal of a transit peptide. Moreover, the mature TGD3 protein was resistant to thermolysin and trypsin treatment but was digested by thermolysin in the presence of detergent. These results were consistent with import of TGD3 into the chloroplast beyond the inner chloroplast envelope membrane. Based on these results, TGD3 would be available for association with the TGD1·TGD2 complex at the stroma side of the inner envelope membrane.

Recombinant TGD3 Protein Has ATPase Activity—The TGD3 protein contains motifs characteristic for other small ATPases associated with bacterial membrane transporters (30, 31) designated Walker-A, Walker-B, A-loop, and Q-loop (Fig. 1B). To test the biochemical function of TGD3, an N-terminal-truncated cDNA missing the coding region for the first 46 amino acids representing the predicted transit peptide was fused to the C terminus-encoding end of the E. coli maltose-binding protein open reading frame and expressed in E. coli.Of several expression systems tested, this was the only providing soluble TGD3 protein. Purification of the recombinant protein designated MBP-{Delta}TGD3 on an amylose column is shown in Fig. 7A. The fusion protein was highly purified (estimated >90%) and migrated as a 78-kDa protein. The fusion protein in the E3 fraction was assayed for its ability to hydrolyze ATP. For control purposes, maltose-binding protein by itself or a mutant cDNA giving rise to an F94A point mutant of TGD3 fused in the same manner as the truncated wild type cDNA were included (MBP-{Delta}TGD3F94A). It should be noted that the F94A mutation did not affect protein expression or protein purification. Replacing the A-loop residue homologous to Phe94 in TGD3 in bacterial orthologs with a nucleophilic cysteine residue was previously shown to greatly decrease the ATP-dependent transport activities (32). Under standard assay conditions as defined under "Experimental Procedures," the reaction was linear for more than 40 min (Fig. 7B). Furthermore, the reaction velocity increased linearly with MBP-{Delta}TGD3 protein concentration in the range tested (Fig. 7C). Varying the concentration of ATP, a classic Michaelis-Menten hyperbolic substrate velocity function was observed for the MBP-{Delta}TGD3 protein (Fig. 7D). Estimated kinetic constants based on this experiment were 151.7 µM for the Km and 0.026 µmol min-1mg-1 for Vmax. These values are in the same range as previously reported for ATP hydrolysis by other ABC transporters (33). The MBP-{Delta}TGD3F94A mutant protein had less than 20% residual activity insufficient for accurate determination of the kinetic constants (Fig. 7D). The maltose-binding protein had no activity.


Figure 6
View larger version (38K):
[in this window]
[in a new window]

  		FIGURE 6.
Subcellular localization of TGD3. A-C, transient expression of the TGD3-encoding cDNA fused to the N terminus of the green fluorescent protein coding sequence. Confocal images are shown with red chlorophyll fluorescence of chloroplasts (A), green fluorescence specific for the TGD3-green fluorescent protein fusion protein (B) and an overlay of A and B. The scale bar equals 8 µm. D, import of in vitro translated TGD3 protein into isolated pea chloroplasts. An autoradiogram of the protein gel is shown. The sensitivity of the imported protein to thermolysin (TL), trypsin (TR), and to thermolysin in the presence of Triton X-100 (TX) was tested. The type of plastid treatment is indicated by +/-. The translation product mixture (TP) containing the TGD3 preprotein (pP) is shown as well. The mature protein (mP) after removal of the transit peptide is indicated.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
The previous identification of TGD1 and TGD2 proteins and the characterization of the respective mutants (4, 9, 10) provided reasonable evidence suggesting that these two proteins are involved in PtdOH transfer possibly from the outer envelope through the inner envelope membrane of chloroplasts in Arabidopsis. Typically, ABC-type lipid transporters such as the biochemically well characterized MsbA protein from E. coli (34, 35) consist of a dimer of a bifunctional protein, with a membrane-spanning permease domain and a second ATP binding domain. In the case of the TGD1/TGD2 transporter, TGD1 represents the permease domain only, and TGD2 is similar to a membrane-tethered substrate binding protein like those associated with other bacterial ABC transporters. The TGD3 (NAP11) protein of Arabidopsis described here appears to be the missing ATP binding component that would be required for a functional TGD1·TGD2·TGD3 complex as proposed in Fig. 8.

Several lines of indirect evidence are consistent with the current hypothesis for the composition of the PtdOH transporter of the inner chloroplast envelope membrane and the involvement of the TGD3 (NAP11) protein; 1) most Gram-bacteria and mycobacteria contain potential orthologs of the three Arabidopsis TGD proteins including the newly described TGD3 protein. Their respective genes are typically linked in operons consistent with their function in the same biochemical process (Fig. 1A). 2) Of the 26 genes of Arabidopsis encoding small ATP-binding proteins without membrane-spanning domains (12), 9 were predicted to be targeted to the chloroplasts. Of the seven genes tested, only the tgd3 mutant corresponding to the gene encoding TGD3 (NAP11) showed a phenotype identical to the previously characterized tgd1 and tgd2 mutants. 3) The TGD3 (NAP11) protein was localized to the stroma of chloroplasts and showed basic ATPase activity in vitro when expressed as a maltose-binding protein fusion.

With the presumed ATPase component of the TGD transport complex identified, reconstitution and direct proof of biochemical transport activity should in principle be possible. However, our current efforts have been hampered by the fact that the production of detergent-soluble, functional full-length TGD1 and TGD2 proteins in E. coli has not yet been feasible. Even the TGD3 protein described here could only be functionally produced fused to maltose-binding protein. The difficulties in demonstrating lipid transport by integral membrane complexes are also evident from the fact that even for the well studied MsbA protein no direct evidence of lipid transfer across the membrane is available. Indirect evidence for lipid transport by MsbA is currently based on the respective mutant phenotype (34) and the observation that specific lipids stimulate the ATPase activity of MsbA in vitro (35). Testing the possible stimulation of ATPase activity of TGD3 was not considered a viable approach, because unlike the MsbA protein, the TGD3 protein lacks the permease domain that would directly interact with the lipid substrates. As for the TGD complex of Arabidopsis, the precise function of MsbA still remains under debate, especially since a mutant of Neisseria meningitis lacking the MsbA ortholog remains capable of exporting phospholipid building blocks of the outer membrane (36). Unfortunately, technical barriers currently prevent us from directly and unambiguously demonstrating the lipid transfer activities of MsbA-like or TGD complex-like lipid transporters.


Figure 7
View larger version (44K):
[in this window]
[in a new window]

  		FIGURE 7.
Purification and ATPase activity of recombinant TGD3 protein. The TGD3 protein is N-terminal-truncated lacking the transit peptide and fused to the C terminus of maltose-binding protein designated as MBP-{Delta}TGD3. A, purification of MBP-{Delta}TGD3 in E. coli. Analysis of 10 µl of total protein from flow-through (F), washing fraction (W), and a series of different elution fractions (1-7) by SDS-PAGE is shown. Proteins were visualized by Coomassie Blue staining. The E3-fraction contained ~5 µg of protein. B, time-course of ATPase activity. The assay was conducted at a protein concentration of ~8 µM and final ATP concentration of 1 mM at 37 °C. Buffer solutions containing no protein were included as background control. Maltose-binding protein treated identical to MBP-{Delta}TGD3 was included to test for endogenous ATPase activity. Relative activity is shown as absorbance readings at 850 nm. C, production of inorganic phosphate as a function of protein concentration at 1 mM ATP. Reactions were performed at 37 °C with an incubation time of 30 min. Absorbance readings were corrected for ATP spontaneous hydrolysis. Three replicates were averaged, and the S.E. are shown. D, ATPase activity of MBP-{Delta}TGD3 and MBP-{Delta}TGD3F94A. The phenylalanine at position 94 (A-loop) was mutated to an alanine residue by site-directed mutagenesis. ATPase assays were performed at various ATP concentrations as indicated. Three replicates were averaged, and the S.E. are shown.

 


Figure 8
View larger version (8K):
[in this window]
[in a new window]

  		FIGURE 8.
Model of the proposed TGD lipid transporter complex. A dimeric composition is shown with numbers corresponding to the TGD1, TGD2, and TGD3 proteins. Outer (oE) and inner chloroplast envelope membranes (iE) are indicated.

 
Despite these difficulties, we are now reasonably certain about the localization of the TGD complex in the inner chloroplast membrane. Because the localization data for TGD1 or TGD2 left open questions, the clear result for TGD3 presented here provides crucial evidence. For example, TGD1 was partially thermolysin-sensitive in chloroplast import experiments (4, 10), and therefore, it seemed possible that at least subpopulations of TGD1 might be localized in the inner and outer chloroplast envelope membranes. Alternatively, the TGD1 protein could be present in a fusion zone between the two membranes. Likewise, TGD2 wild-type protein was resistant to trypsin, but the TGD2-green fluorescent protein fusion protein was not (9). This result could be interpreted as the presence of TGD2 in localized domains between the inner and the outer envelope membrane, which might be inaccessible to trypsin. On the other hand, the evidence for TGD3 presented here is unambiguous; TGD3 is imported into the chloroplast beyond the inner envelope membrane and processed during in vitro import experiments (Fig. 6B). The only way TGD3 could interact with TGD1 and TGD2 as suggested by the identical phenotypes for the respective mutants and the organization of predicted bacterial orthologs is if TGD1 and TGD2 are intrinsic proteins of the inner chloroplast envelope membrane as shown in Fig. 8, thereby confirming previous conclusions drawn about TGD1 and TGD2 localization.

Demonstrating a direct interaction of the three TGD proteins in the inner chloroplast envelope has proven difficult in our hands using available in vivo techniques. It will require either the isolation of the native complex or its reconstitution from recombinant proteins, both of which have not yet been accomplished. However, our current data clearly suggest that TGD3 is the missing ATPase subunit of this transporter, bringing us a step closer to the actual reconstitution of a core transporter possibly consisting of the TGD1 permease and the TGD3 ATP-binding protein.


   FOOTNOTES
 
* This work was supported in part by United States National Science Foundation Grant MCB 0453858 and United States Department of Energy Grant DE-FG02-98ER20305. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

Formula The on-line version of this article (available at http://www.jbc.org) contains supplemental Fig. S1. Back

1 Current address: Graduate School of Science and Engineering, Saitama University, 255 Shimo-Okubo, Sakura-Ku, Saitama City, Saitama 338-8570, Japan. Back

2 To whom correspondence should be addressed: Dept. of Biochemistry and Molecular Biology, Michigan State University, Biochemistry 215, East Lansing, MI 48824-1319; Tel.: 517-355-1609; Fax: 517-355-9334; E-Mail: benning{at}msu.edu.

3 The abbreviations used are: Col-2, Columbia-2; NAP11, non-intrinsic ABC protein 11; TGDG, trigalactosyldiacylglycerol; MBP, maltose-binding protein; DGDG, digalactosyldiacylglycerol; MGDG, monogalactosyldiacylglycerol; ER, endoplasmic reticulum; PtdOH, phosphatidic acid; ABC, ATP binding cassette; MS, mass spectrometry. Back


   ACKNOWLEDGMENTS
 
We thank John Froehlich from the Michigan State University Department of Energy Plant Research Laboratory for helpful advice on the chloroplast import experiments. In addition, we thank Michael R. Garavito from the Michigan State University Department of Biochemistry and Molecular Biology for advice on protein expression and purification.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Benning, C., Xu, C., and Awai, K. (2006) Curr. Opin. Plant Biol. 9, 241-247[CrossRef][Medline] [Order article via Infotrieve]
  2. Dörmann, P., and Benning, C. (2002) Trends Plant Sci. 7, 112-118[CrossRef][Medline] [Order article via Infotrieve]
  3. Benning, C., and Ohta, H. (2005) J. Biol. Chem. 280, 2397-2400[Abstract/Free Full Text]
  4. Xu, C., Fan, J., Froehlich, J. E., Awai, K., and Benning, C. (2005) Plant Cell 17, 3094-3110[Abstract/Free Full Text]
  5. Jarvis, P., Dörmann, P., Peto, C. A., Lutes, J., Benning, C., and Chory, J. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 8175-8179[Abstract/Free Full Text]
  6. Froehlich, J. E., Benning, C., and Dörmann, P. (2001) J. Biol. Chem. 276, 31806-31812[Abstract/Free Full Text]
  7. Dörmann, P., Balbo, I., and Benning, C. (1999) Science 284, 2181-2184[Abstract/Free Full Text]
  8. Roughan, P. G., and Slack, C. R. (1982) Annu. Rev. Plant Physiol. 33, 97-132
  9. Awai, K., Xu, C., Tamot, B., and Benning, C. (2006) Proc. Natl. Acad. Sci. U. S. A. 103, 10817-10822[Abstract/Free Full Text]
  10. Xu, C., Fan, J., Riekhof, W., Froehlich, J. E., and Benning, C. (2003) EMBO J. 22, 2370-2379[CrossRef][Medline] [Order article via Infotrieve]
  11. Doerrler, W. T. (2006) Mol. Microbiol. 60, 542-552[CrossRef][Medline] [Order article via Infotrieve]
  12. Sanchez-Fernandez, R., Davies, T. G., Coleman, J. O., and Rea, P. A. (2001) J. Biol. Chem. 276, 30231-30244[Abstract/Free Full Text]
  13. Murashige, T., and Skoog, F. (1962) Physiol. Plant. 15, 473-497[CrossRef]
  14. Clough, S. J., and Bent, A. F. (1998) Plant J. 16, 735-743[CrossRef][Medline] [Order article via Infotrieve]
  15. Dörmann, P., Hoffmann-Benning, S., Balbo, I., and Benning, C. (1995) Plant Cell 7, 1801-1810[Abstract]
  16. Rossak, M., Schäfer, A., Xu, N., Gage, D. A., and Benning, C. (1997) Arch. Biochem. Biophys. 340, 219-230[CrossRef][Medline] [Order article via Infotrieve]
  17. Benning, C., Huang, Z. H., and Gage, D. A. (1995) Arch. Biochem. Biophys. 317, 103-111[CrossRef][Medline] [Order article via Infotrieve]
  18. Siebertz, H. P., and Heinz, E. (1977) Z. Naturforsch. 32, 193-205
  19. Miquel, M., Cassagne, C., and Browse, J. (1998) Plant Physiol. 117, 923-930[Abstract/Free Full Text]
  20. Houjou, T., Yamatani, K., Imagawa, M., Shimizu, T., and Taguchi, R. (2005) Rapid Commun. Mass Spectrom. 19, 654-666[CrossRef][Medline] [Order article via Infotrieve]
  21. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254[CrossRef][Medline] [Order article via Infotrieve]
  22. Chifflet, S., Torriglia, A., Chiesa, R., and Tolosa, S. (1988) Anal. Biochem. 168, 1-4[CrossRef][Medline] [Order article via Infotrieve]
  23. Osterman, A., and Overbeek, R. (2003) Curr. Opin. Chem. Biol. 7, 238-251[CrossRef][Medline] [Order article via Infotrieve]
  24. Emanuelsson, O., Nielsen, H., and von Heijne, G. (1999) Protein Sci. 8, 978-984[Abstract]
  25. Alonso, J. M., Stepanova, A. N., Leisse, T. J., Kim, C. J., Chen, H., Shinn, P., Stevenson, D. K., Zimmerman, J., Barajas, P., Cheuk, R., Gadrinab, C., Heller, C., Jeske, A., Koesema, E., Meyers, C. C., Parker, H., Prednis, L., Ansari, Y., Choy, N., Deen, H., Geralt, M., Hazari, N., Hom, E., Karnes, M., Mulholland, C., Ndubaku, R., Schmidt, I., Guzman, P., Aguilar-Henonin, L., Schmid, M., Weigel, D., Carter, D. E., Marchand, T., Risseeuw, E., Brogden, D., Zeko, A., Crosby, W. L., Berry, C. C., and Ecker, J. R. (2003) Science 301, 653-657[Abstract/Free Full Text]
  26. Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D. J. (1997) Nucleic Acids Res. 25, 3389-3402[Abstract/Free Full Text]
  27. Heinz, E., and Roughan, G. (1983) Plant Physiol. 72, 273-279[Abstract/Free Full Text]
  28. Cline, K., Werner-Washburne, M., Andrews, J., and Keegstra, K. (1984) Plant Physiol. 75, 675-678[Abstract/Free Full Text]
  29. Jackson, D. T., Froehlich, J. E., and Keegstra, K. (1998) J. Biol. Chem. 273, 16583-16588[Abstract/Free Full Text]
  30. Ambudkar, S. V., Kim, I. W., Xia, D., and Sauna, Z. E. (2006) FEBS Lett. 580, 1049-1055[CrossRef][Medline] [Order article via Infotrieve]
  31. Davidson, A. L., and Chen, J. (2004) Annu. Rev. Biochem. 73, 241-268[CrossRef][Medline] [Order article via Infotrieve]
  32. Zhao, Q., and Chang, X. B. (2004) J. Biol. Chem. 279, 48505-48512[Abstract/Free Full Text]
  33. Schneider, E., and Hunke, S. (1998) FEMS Microbiol. Rev. 22, 1-20[CrossRef][Medline] [Order article via Infotrieve]
  34. Doerrler, W. T., Reedy, M. C., and Raetz, C. R. (2001) J. Biol. Chem. 276, 11461-11464[Abstract/Free Full Text]
  35. Doerrler, W. T., and Raetz, C. R. (2002) J. Biol. Chem. 277, 36697-36705[Abstract/Free Full Text]
  36. Tefsen, B., Bos, M. P., Beckers, F., Tommassen, J., and de Cock, H. (2005) J. Biol. Chem. 280, 35961-35966[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Plant CellHome page
C. Xu, J. Fan, A. J. Cornish, and C. Benning
Lipid Trafficking between the Endoplasmic Reticulum and the Plastid in Arabidopsis Requires the Extraplastidic TGD4 Protein
PLANT CELL, August 1, 2008; 20(8): 2190 - 2204.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
282/49/35945    most recent
M704063200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lu, B.
Right arrow Articles by Benning, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lu, B.
Right arrow Articles by Benning, C.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS