LPT1 Encodes a Membrane-bound O-Acyltransferase Involved in the Acylation of Lysophospholipids in the Yeast Saccharomyces cerevisiae*

Phospholipids are major components of cellular membranes that participate in a range of cellular processes. Phosphatidic acid (PA) is a key molecule in the phospholipid biosynthetic pathway. In Saccharomyces cerevisiae, SLC1 has been identified as the gene encoding lysophosphatidic acid acyltransferase, which catalyzes PA synthesis. However, despite the importance of PA, disruption of SLC1 does not affect cell viability (Nagiec, M. M., Wells, G. B., Lester, R. L., and Dickson, R. C. (1993) J. Biol. Chem. 268, 22156–22163). We originally aimed to identify the acetyl-CoA:lyso platelet-activating factor acetyltransferase (lysoPAF AT) gene in yeast. Screening of a complete set of yeast deletion clones (4741 homozygous diploid clones) revealed a single mutant strain, YOR175c, with a defect in lysoPAF AT activity. YOR175c has been predicted to be a member of the membrane-bound O-acyltransferase superfamily, and we designated the gene LPT1. An Lpt1-green fluorescent protein fusion protein localized at the endoplasmic reticulum. Other than lysoPAF AT activity, Lpt1 catalyzed acyltransferase activity with a wide variety of lysophospholipids as acceptors, including lysophosphatidic acid, lysophosphatidylcholine, lysophosphatidylethanolamine, lysophosphatidylglycerol, lysophosphatidylinositol, and lysophosphatidylserine. A liquid chromatography-mass spectrometry analysis indicated that lysophosphatidylcholine and lysophosphatidylethanolamine accumulated in the Δlpt1 mutant strain. Although the Δlpt1 mutant strain did not show other detectable defects, the Δlpt1 Δslc1 double mutant strain had a synthetic lethal phenotype. These results indicate that, in concert with Slc1, Lpt1 plays a central role in PA biosynthesis, which is essential for cell viability.

The cell membrane is a semipermeable lipid bilayer found in all living cells that physically separates the cytoplasm of the cell from the extracellular environment. Glycerophospholipids and sphingophospholipids are the major components of most cell membranes. Phosphatidic acid (PA) 2 is a key intermediate in the biosynthesis of glycerophospholipids. PA is synthesized by two major de novo biosynthetic pathways that utilize either glycerol 3-phosphate (G-3-P) or dihydroxyacetone phosphate (DHAP) as precursors (1,2). G-3-P is acylated by G-3-P acyltransferase at the sn-1 position to form lysophosphatidic acid (LPA). DHAP is acylated at the sn-1 position by DHAP acyltransferase to produce 1-acyl-DHAP, which is reduced by 1-acyl-DHAP reductase to form LPA. LPA produced by these two different pathways is further acylated by LPA acyltransferase in the sn-2 position to yield PA.
In mammals, several LPA acyltransferase genes have been cloned, and their gene products have been characterized (3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13). In yeast, SLC1, originally obtained as a gene suppressing a defect in the biosynthesis of the sphingolipid long chain base (1,14), was identified as an LPA acyltransferase gene (15). Glycerophospholipids, including phosphatidylserine (PS), phosphatidylethanolamine (PE), and phosphatidylcholine (PC), are synthesized from PA through the cytidinediphosphodiacylglycerol pathway. Cytidinediphosphodiacylglycerol is also used as a precursor for the synthesis of phosphatidylinositol and cardiolipin. Alternatively, PE and PC are also synthesized via the CDP-ethanolamine and CDP-choline, respectively (Kennedy pathway) (1,16). Glycerophospholipids can also be generated by remodeling (also known as the Lands' cycle (17)(18)(19), in which the rapid turnover of the sn-2 acyl moiety of glycero-phospholipids is carried out by phospholipases and lysophospholipid acyltransferases, but the specific enzyme involved in the remodeling of glycerophospholipids had not been identified. Very recently, a gene for mouse lysophosphatidylcholine (LPC) acyltransferase was cloned by two different research groups and designated LPCAT1 (20,21). LPCAT1 is highly expressed in alveolar type II cells and is thought to be involved in the production of surfactant lipids. A related gene, LPCAT2, which encodes an enzyme having both acetyl-CoA:lyso-platelet-activating factor acetyltransferase (lysoPAF AT) activity and LPC acyltransferase activity, has also been reported (22).
In this study, we searched for lysoPAF AT genes in yeast by measuring the enzyme activity of each strain in a complete set of yeast deletion clones. By this brute force approach, we identified a novel lysoPAF AT gene that did not show any homology with LPCAT2, and which we designated LPT1 (lysoPAF AT 1). LPT1 gene products possess lysophospholipid acyltransferase activity with a wide variety of substrate acceptors, including LPC, LPE, LPA, LPI, LPS, and LPG. We also characterized in detail the enzymatic activity and cellular role of Lpt1.

EXPERIMENTAL PROCEDURES
Strains and Media-The complete set of Yeast Deletion Clones (Homozygous Diploid) and the parental strain BY4743 were purchased from Invitrogen. Other yeast strains used in this study are shown in Table 1. Strains were grown on either YPD medium (1% yeast extract, 2% Bacto-peptone, and 2% glucose), synthetic dextrose (0.67% yeast nitrogen base without amino acids (Difco), 2% glucose, and appropriate supplements) or synthetic complete (SC) medium (SD medium, with dropout powder) prepared as previously described (23). Selection for integration of deletion cassettes containing loxP-kanMX-loxP was performed on YPD plates containing 200 g/ml G418.
The ORF of the PLB1 gene was amplified by PCR using primers (5Ј-AGTCGACATGAAGTTGCAGAGTTTGTTGG-3Ј and 5Ј-TGCGGCCGCCAATTAGACCGAAGACGGCACT-AATG-3Ј) to introduce a SalI site before the first ATG codon and a NotI site just before the stop codon, and the ORF was ligated into the SmaI site of plasmid pUC19 to form PLB1/ pUC19. A SalI-NotI fragment containing the PLB1 ORF was inserted into YEGAp (25) to form PLB1/YEGAp, in which the PLB1 gene was expressed under the control of the glyceraldehyde-3-phosphate dehydrogenase promoter.
Enzyme Preparation for Screening-Each strain of the complete set of Yeast Deletion Clones were grown in 5 ml of YPD medium containing 200 g/ml G418 at 30°C overnight. Cells were collected by centrifugation and suspended in 2 volumes of 50 mM Tris-HCl buffer (pH 8) containing 1 mM 2-mercaptoethanol and 2 mM phenylmethylsulfonyl fluoride. The cells were disrupted with glass beads by vortexing at 4°C, and then subjected to an enzyme assay as described below, except that the reaction was performed at 30°C.
Preparation of Microsomal Fractions-Yeast cells were suspended in 2 volumes of 50 mM Tris-HCl buffer (pH 7.5) containing 5 mM 2-mercaptoethanol and 2 mM phenylmethylsulfonyl fluoride, and disrupted with glass beads by vortexing at 4°C. The cell debris was removed by centrifugation at 18,800 ϫ g for 15 min at 4°C to obtain the crude enzyme fraction. The microsomal fraction was pelleted by centrifugation of the crude enzyme fraction at 100,000 ϫ g for 60 min at 4°C, and the pellet was resuspended in a small amount of the same buffer. Alternatively, cells were suspended in 5 volumes of spheroplasting buffer (20 mM Tris-HCl (pH 7.5), 1.2 M sorbitol, 50 mM potassium acetate, and 1 mM 2-mercaptoethanol); Zymolyase 100T (Seikagaku, Japan) was added to a concentration of 1 mg/ml, and the cells were incubated at 30°C for 10 -30 min. The spheroplasts were washed with spheroplasting buffer and lysed in ice-cold lysis buffer (20 mM Tris-HCl (pH 7.5), 100 mM sorbitol, 50 mM potassium acetate, 1 mM 2-mercaptoethanol, and 2 mM phenylmethylsulfonyl fluoride) using a Teflon homogenizer. The cell lysate was further subjected to centrifugation and ultracentrifugation to obtain the microsome fraction as described above.
We performed centrifugation at 18,800 ϫ g before ultracentrifugation to avoid the degradation of Lpt1-HA. When we examined the enzyme properties of samples obtained at lower centrifugation speeds before ultracentrifugation, the reaction was not linear, and quick enzyme degradation was observed during the assay even in the presence of protease inhibitors. We found that the microsome fraction prepared after performing centrifugation at 18,800 ϫ g was more stable and performed better in the enzyme assay even if the procedure resulted in the removal of a large amount of nuclei and nucleus-attached ER.
To investigate the substrate specificities of this enzyme, various lysophospholipids and acyl-CoAs (Sigma or Avanti Polar Lipids, Alabaster AL) were used in place of either lysoPAF or acetyl-CoA. The protein concentration was determined by the method of Bradford (29).
Western Blot Analysis-Microsome fraction (2 g of total protein) was electrophoresed on a 10% SDS-PAGE and then electroblotted onto polyvinylidene difluoride membranes. The polyvinylidene difluoride membranes were blocked with TBS-T (Tris-buffered saline, pH 7.6, 0.1% Tween 20) containing 5% skim milk for 1 h at room temperature and then incubated with 0.1 g/ml of 12CA5 monoclonal antibody or GFP polyclonal antibody for 1 h at room temperature. After washing with TBS-T, the membranes were incubated with a 1:5000 dilution of anti-mouse IgG peroxidase-linked whole antibody (from sheep) (Amersham Biosciences) or goat anti-rabbit IgG-horseradish peroxidase for 1 h at room temperature. After washing with TBS-T, HA-or GST-tagged protein was detected by the Super Signal West Pico detection reagent (Pierce).
Platelet Aggregation Assay-Total lipids were extracted by the method of Bligh and Dyer and applied to an alumina column. The column was washed with chloroform, and the PAF fraction was eluted with chloroform-methanol (1:1). The eluted fraction was further loaded onto a paper fiber-made TLC sheet (Chromato sheet, Wako) and developed with chloroform/ methanol/water (65:25:4). The area containing PAF was cut off, and the PAF fraction was eluted with chloroform/methanol (1:1), dried under N 2 gas, and dissolved in 100 l of bovine serum albumin/saline. Washed rabbit platelets were prepared by the method of Pincard et al. (30). Tyrode's solution (160 l, pH 7.2, containing 1.3 mM Ca 2ϩ ) was mixed with 40 l of washed rabbit platelets and preincubated at 37°C for 1 min. Then 10 l of sample lipid was added. Aggregation was assayed by monitoring the change in light transmittance with a Hematracer PAT-4A (Nikko, Japan).
Confocal Microscopy-The ⌬lpt1 mutant strain transformed with LPT1-GFP/YEp(rop) was grown in SC(Trp Ϫ ) liquid medium to early log phase, and the cells were analyzed using a confocal microscope (FV1000, Olympus) equipped with a 100 ϫ oil-immersion objective.
Liquid Chromatography/Mass Spectrometric Analysis-LC/MS analysis was performed in the positive mode on an LCMS-2010 mass spectrometer (Shimadzu, Kyoto, Japan) equipped with an electrospray ion source. High performance liquid chromatography separation was carried out on a normalphase column (Develosil 60, 2.0 ϫ 150 mm, Nomura Chemicals, Seto, Japan). The column was eluted with solvent A (ace-tonitrile-methanol (2:1) containing 0.1% ammonium formate (pH 6.4)) and solvent B (methanol-H 2 O (2:1) containing 0.1% ammonium formate (pH 6.4)) at a flow rate of 0.2 ml/min. The gradient separation was carried out using the following conditions: isocratic elution with 100% solvent A for 5 min followed by a linear gradient of 100% to 70% solvent A over 40 min. Spectra were obtained over a mass range from m/z 400 -1200 with a scan time of 1 s.

RESULTS
Identification of lysoPAF AT in the Yeast S. cerevisiae-We previously reported that Saccharomyces cerevisiae produce platelet-activating factor (PAF) (31). To identify the gene for lysoPAF AT, we screened a complete set of homozygous diploid yeast deletion clones as described under "Experimental Procedures." Using [ 14 C]acetyl-CoA and lysoPAF, we detected lysoPAF AT activity in both crude cell-free extracts and microsome fractions (100,000 ϫ g pellets) of yeast. In the 4741 strains screened, we found one clone that did not produce [ 14 C]PAF. This clone lacked ORF YOR175c, an as-yet-unidentified gene encoding a putative membrane-bound O-acyltransferase. We designated this gene LPT1. We deleted the corresponding gene in several yeast strains with different genetic backgrounds. All strains tested possessed lysoPAF AT activity, and when the LPT1 gene was deleted, the activity completely disappeared (Fig. 1A). In contrast, the enzyme activity increased when the LPT1 gene was introduced with a multicopy vector (Fig. 1A). We also prepared microsome fractions from the ⌬lpt1 strain transformed with the LPT1 overexpression vector or a control vector and used these fractions for PAF synthesis with lysoPAF and acetyl-CoA as the substrates. The lipids were extracted from the enzyme reaction products and analyzed by LC/MS. Synthesized PAF was identified by comparing their retention time in the column and m/z 524.4 with those of standard C16: PAF. The reaction with microsomes obtained from the ⌬lpt1 strain transformed with the LPT1 overexpression vector produced a time-dependent increase of C16:PAF production; this increase was not observed in the ⌬lpt1 strain harboring the control vector (data not shown), suggesting that LPT1 is required for PAF synthesis. The lipid fractions were further purified, and the PAF fractions were subjected to a platelet aggregation assay. The PAF fraction from the LPT1-overexpressing strain showed strong platelet aggregation activity, which was hardly observed when the PAF fraction from the ⌬lpt1 strain with the control vector was used (Fig. 1B) or when lysoPAF was omitted from the reaction mixture. The platelet aggregation activity was also strongly inhibited in the presence of 10 Ϫ6 M WEB2086, a PAF antagonist (Fig. 1B). From these results we identified LPT1 as a lysoPAF AT gene in the yeast S. cerevisiae.
A search of databases for homologues of yeast Lpt1 identified several MBOATs, including SPBC16A3.10 in the fission yeast Schizosaccharomyces pombe, OACT1 and -2 in both human and mouse (Fig. 2), and C54G7.2 and C08F8.4 in Caenorhabditis elegans. The predicted amino acid sequence of LPT1 shared 27.5% identity with OACT1 (per 459 aa) and 28.8% identity with OACT2 (per 458 aa). It was predicted that a conserved histidine residue within a long hydrophobic region might be a candidate for the active-site residue (32). The corresponding histidine residue was found at position of 382 in Lpt1 (Fig. 2). To test whether His-382 is involved in the catalytic center, we constructed an HA-tagged mutant Lpt1 expression vector in which His-382 was replaced by asparagine. Although we confirmed the expression of both native and mutant Lpt1-HA in the microsomal fraction of the ⌬lpt1 mutant strain by Western blot analysis, we did not detect lysoPAF AT activity in the mutant enzyme fraction (see Fig. 6, B and C). From this result, we concluded that His-382 is critical for the catalytic reaction and is most likely an active-site residue.  NOVEMBER 23, 2007 • VOLUME 282 • NUMBER 47

JOURNAL OF BIOLOGICAL CHEMISTRY 34291
Substrate Selectivity of Lpt1-We tried to solubilize the enzyme from the microsomal fraction using various surfactants, including CHAPS and octylglucoside, but the enzyme activity was considerably decreased by solubilization (data not shown). We therefore used the microsome fraction from the Lpt1-overproducing strain to characterize the enzyme. Although the enzyme reaction was performed at 30°C during the screening, we noticed that the enzyme activity decreased rapidly at this reaction temperature because the Lpt1 protein degraded even in the presence of various protease inhibitors (data not shown). For these reasons, the reaction was performed at 0 -15°C in the buffer as described under "Experimental Procedures." In this condition, the reaction was linear for the first 15 min (data not shown).
To examine whether Lpt1 uses other acyl-CoAs as substrates, lysoPAF AT activity was measured in the presence of various acyl-CoAs. lysoPAF AT activity was dramatically decreased in the presence of propionyl-CoA (C3:0) and butyryl-CoA (C4:0) and was not detected in the presence of palmitoyl-CoA (C16:0) or arachidonoyl-CoA (C20:4) (Fig. 1C). These results suggest that Lpt1 also catalyzes acyltransferase activity. So we first examined the substrate specificity of Lpt1 using [1-14 C]linoleoyl-CoA (C18:2) as an acyl donor with a variety of lysophospholipid acceptors (Fig. 3). The microsomal fraction from the ⌬lpt1 mutant strain transformed with the LPT1 overexpression vector showed acyltransferase activity with various lysophospholipids as acceptors, including LPC, LPG, LPA, LPE, LPI, and LPS (Fig. 3). Lpt1 preferred an acyl residue to an alkyl residue at the sn-1 position of lysophospholipids. Because the reaction products were hardly detectable when microsomes from the ⌬lpt1 mutant strain with control vector were used, LPT1 would appear to be mainly responsible for both acetyl-and acyltransferase activity in the microsomal fraction in the yeast S. cerevisiae.
Kinetics of Lpt1-We examined the kinetic properties of Lpt1 by measuring acetyl-or acyltransferase activity in the microsomal fraction derived from ⌬lpt1 strains transformed by the LPT1 overexpression vector or control vector. We used  Cellular Localization of Lpt1-GFP-To clarify the cellular localization of Lpt1, we constructed an Lpt1-GFP fusion protein expression vector and expressed it in the ⌬lpt1 strain under the control of the LPT1 promoter. Transformed cells were grown in SC(Trp Ϫ ) medium, and the enzyme distribution at the exponential growth phase was examined by using confocal microscopy. Fluorescence from Lpt1-GFP was observed around the nucleus, most likely in the ER, and on the cell surface (Fig. 6A). Lpt1-GFP was functional and no degradation was observed (Fig. 6, B and C).
Lpt1 Plays a Critical Role in PA Synthesis in Concert with Slc1-Slc1 is an LPA acyltransferase present in lipid particles of S. cerevisiae (15). Because the majority of Lpt1 LPA acyltransferase activity was in the ER, we thought there might be a functional relationship between the LPT1 and SLC1 genes. We further explored the potential relationship between these genes by investigating whether null alleles of LPT1 and SLC1 genes interact genetically. To determine the phenotype of ⌬lpt1 ⌬slc1 double mutants, a diploid strain heterozygous for LPT1 and SLC1 deletion alleles was sporulated, and tetrads were dissected. Although strains containing single ⌬lpt1 or ⌬slc1 mutations grew normally, strains containing both ⌬lpt1 and ⌬slc1 mutations did not grow (Fig. 7, top). The synthetic lethal phenotype found in the ⌬lpt1 ⌬slc1 double mutant was complemented by the LPT1 expression vector (Fig. 7, bottom). These results indicate that the two different enzymes, Lpt1 and Slc1, are functionally related in PA synthesis.
A Growth Defect Caused by the Overexpression of PLB1 Is Partially Suppressed by Coexpression of LPT1-Our enzymatic characterization of Lpt1 showed that the enzyme has acyltransferase activity with a wide variety of substrate lysophospholipids. As phospholipases catalyze the reverse reaction of lysophospholipid acyltransferase, we examined the relationship between Lpt1 and phospholipases. We found that overexpression of PLB1, a gene encoding the major phospholipase in yeast (36), caused similar growth delays in both wild-type and ⌬lpt1 mutant strains. However, when LPT1 was coexpressed with PLB1 in the ⌬lpt1 mutant strain, the growth-delay phenotype was partially suppressed (Fig. 8). The suppression was probably only partial, because Plb1 releases fatty acids from both the sn-1 and sn-2 positions of glycerophospholipids, whereas Lpt1 catalyzes acetyl and acyl transfer reactions only at the sn-2 position of lysophospholipids. We also examined the effect of another phospholipase that localizes to the ER, NTE1 (37), on cell growth. However, we did not detect a growth-delay following NTE1 overexpression (data not shown).
The Phospholipid Contents Varied in the ⌬lpt1 Mutant Strain-To test whether a lack of Lpt1 affects the phospholipid contents, we extracted total lipids from the wild-type and ⌬lpt1 mutant strain and subjected them to an LC/MS analysis as described under "Experimental Procedure." Phospholipids were identified by their retention time in the column and m/z as described previously (38). Using this method, we detected several kinds of glycerophospholipids, including PC, LPC, PE, and LPE. In the ⌬lpt1 strain, although the level of PC and PE (for example, m/z 758. 5

Lpt1, a Novel Lysophospholipid Acyltransferase in Yeast
type ( Fig. 9 and supplemental Fig. S1), thus further supporting that Lpt1 has acyltransferase activity with a wide variety of lysophospholipids.
Although we clearly showed that Lpt1 has PAF producing ability (Fig. 1), PAF could not be detected in the lipid fraction of LPT1-overexpressing cells by LC/MS analysis. These results indicate that the amount of PAF might be too small to be detected in the yeast cell, although we cannot exclude the possibility of PAF production in yeast.

DISCUSSION
PAF is a phospholipid mediator that activates a G proteincoupled receptor. PAF has potent biological activities, includ-

Lpt1, a Novel Lysophospholipid Acyltransferase in Yeast
ing platelet activation, airway constriction, and hypotension (39). In mammals, PAF can be synthesized by two biosynthetic pathways (40). One is called the remodeling pathway, in which an acetyl group of acetyl-CoA is transferred to an lysoPAF. In this pathway, lysoPAF is released from the membrane lipids by the action of phospholipase A2, and then lysoPAF AT catalyzes an acetylation reaction. The other is called the de novo pathway, which involves the conversion of 1-O-alkyl-2-acetylglycerol and CDP-choline to PAF.
PAF has been extensively studied because its strong enzyme activity is associated with various pathologies, including allergy and inflammation. However, it has been impossible to identify lysoPAF AT, a key enzyme in PAF synthesis, because of problems encountered in its purification (27,41,42). We previously found that S. cerevisiae produces PAF (31), which prompted us to start the screening the yeast deletion library to identify lysoPAF AT. This approach will not only permit us to examine the role of PAF and lysoPAF AT in yeast but will also provide insights into the identity of mammalian lysoPAF AT.
In this study, we identified a gene (LPT1) encoding an enzyme that has lysoPAF AT activity as well as lysophospholipid acyltransferase activity. To our surprise, Lpt1 did not show any homology with LPCAT2, a recently identified orthologue of LPC acyltransferase that has both lysoPAF AT and lysoPC acyltransferase activity (22), suggesting that Lpt1 is a novel lysoPAF AT.
LPT1 encodes a protein that is classified as an MBOAT. The MBOAT superfamily, which is conserved from yeast to human (32), consists of a large and diverse group of membrane associated acyltransferases. In yeast, 5 MBOATs have been found, ARE1, ARE2, GUP1, GUP2, and YOR175C (LPT1) (32,35). In this study, we found that Lpt1, the fifth MBOAT in yeast S. cerevisiae, catalyzes the transfer of acyl residues to a wide variety of substrate acceptors, including LPA, LPC, LPI, LPS, LPG, and LPE (Fig. 10).

Lpt1, a Novel Lysophospholipid Acyltransferase in Yeast
Inactivation of both GAT1 and GAT2/SCT1 resulted in a synthetic lethal phenotype, suggesting that these two genes cooperatively play a central role in the first step of the PA synthesis pathway. However, despite the importance of PA, disruption of SLC1 does not cause any defect in cell growth, indicating that PA biosynthesis may proceed through an alternative pathway (15). Indeed, the ⌬lpt1 ⌬slc1 double mutant showed a synthetic lethal phenotype like the ⌬gat1 ⌬gat2/⌬sct1 double mutant. Thus, we conclude that the concerted action of Lpt1 and Slc1 is indispensable for PA synthesis in vivo and for cell growth (Fig. 10). We could hardly detect acyltransferase activity with various lysophospholipid acceptors using [1-14 C]linoleoyl-CoA in the ⌬lpt1 mutant (Fig. 3), however, this does not exclude the possibility that different acyltransferases with overlapping functions could be found in the different assay conditions or different growth conditions. Indeed, 1-acyl-LPA acyltransferase activity was detected in the ⌬lpt1 mutant when [1-14 C]arachidonoyl-CoA was used (supplemental Fig. S2). The activity found in the ⌬lpt1 mutant might be catalyzed by Slc1, another LPA acyltransferase. We also examined the effect of the SLC1 deletion on lysophospholipid acyltransferase activity. The ⌬slc1 mutant showed slightly reduced acyltransferase activity compared with the wild-type strain and dramatically reduced 1-acyl LPA acyltransferase activity (supplemental Fig. S2). This result indicates that Slc1 mainly catalyzes LPA acyltransferase activity and that it may also be involved in various lysophospholipid acyltransferase reactions.
Tafazzin is a protein with some similarities to LPAAT. Mutation of the tafazzin gene is associated with diseases, including Barth syndrome (43). In yeast, an orthologue of tafazzin gene has been identified as TAZ1, whereas the role of Taz1 in yeast is controversial. It has been reported that Taz1 has LPC acyltransferase activity in the mitochondrial fraction (44), and others have hypothesized that Taz1 acts as a monolyso-CL acyltransferase (56). Although we examined lysophospholipid acyltransferase activity using the microsome fraction from the ⌬taz1 mutant, we could not detect any difference in enzyme activity between the wild-type and the ⌬taz1 mutant (supplemental Fig. S2).
Glycerophospholipid structure exhibits a high degree of heterogeneity. Saturated fatty acyl groups are predominated in the sn-1 position, whereas unsaturated fatty acids are commonly found at the sn-2 position. It is thought that unsaturated fatty acid incorporation at the sn-2 position does not occur through a novo synthetic pathway but through fatty acid remodeling (57). Our finding that Lpt1 catalyzes lysophospholipid acyltransferase activity with a strong preference for unsaturated medium-chain acyl-CoAs as acyl donors (Fig. 4B) may indicate the possible involvement of Lpt1 in glycerophospholipid heterogenicity in yeast.
Addendum-After submission of our manuscript, three independent reports identifying LPT1 (YOR175c) appeared. Dr. Voelker's group identified YOR175c as the major lysophosphatidylethanolamine acyltransferase-encoding gene and named it ALE1 (58). Dr. Conzelmann's group identified the same gene as a second 1-acylsn-glycerol-3-phosphate acyltransferase and named it SLC4, which has partially redundant function with SLC1, a previously identified LPA acyltransferase (59). Dr. Oelkers's group also identified the same gene as the lysophospholipid acyltransferase-encoding gene and named it LPT1 (60).