Enzymatic activities of the human AGPAT isoform 3 and isoform 5: localization of AGPAT5 to mitochondria

Abstract

The enzyme 1-acylglycerol-3-phosphate-O-acyltransferase (AGPAT) converts lysophosphatidic acid (LPA) to phosphatidic acid (PA). In this study, we show enzymatic properties, tissue distribution, and subcellular localization of human AGPAT3 and AGPAT5. In cells overexpressing these isoforms, the proteins were detected in the nuclear envelope and the endoplasmic reticulum. AGPAT5-GFP fusion protein was localized in the mitochondria of both Chinese hamster ovary and human epithelial cervical cancer cells. Using lysates of AD293 cells infected with AGPAT3 and AGPAT5 recombinant adenovirus, we show that AGPAT3 and AGPAT5 proteins have AGPAT activity. Both the isoforms have similar apparent V(max) of 6.35 and 2.42 nmol/min/mg protein, respectively, for similar LPA. The difference between the two isoforms is in their use of additional lysophospholipids. AGPAT3 shows significant esterification of lysophosphatidylinositol (LPI) in the presence of C20:4 fatty acid, whereas AGPAT5 demonstrates significant acyltransferase activity toward lysophosphatidylethanolamine (LPE) in the presence of C18:1 fatty acid. The AGPAT3 mRNA is ubiquitously expressed in human tissues with several-fold differences in the expression pattern compared with the closely related AGPAT4. In summary, we show that in the presence of different fatty acids, AGPAT3 and AGPAT5 prefer different lysophospholipids as acyl acceptors. More importantly, localization of overexpressed AGPAT5 (this study) as well as GPAT1 and 2 (previous studies) in mitochondria supports the idea that the mitochondria might be capable of synthesizing some of their own glycerophospholipids.

Fig. 1.

Localization of AGPAT3-GFP and AGPAT5-GFP to endo-membranes in cultured cells. A, B: CHO cells overexpressing AGPAT3-GFP were fixed in methanol and incubated with antibody sec61β specific for endoplasmic reticulum and lamin A/C specific for nuclear lamina and imaged for green and red fluorescence using fluorescence microscopy. Shown are representative images for AGPAT3 (green fluorescence), sec61β (red fluorescence), DAPI (blue fluorescence), colocalization channel (yellow fluorescence), and the merged image. C, D: Fluorescence images for AGPAT5-GFP expressed in CHO cells. Cells for sec61β and lamin A/C were processed as above. E: The AGPAT5-GFP expressing cells were incubated with MitoTracker Red dye, fixed in 4% paraformaldehyde, and imaged as before. Shown for each image is a single z-stack image, where the x and y axes show the z-stacks. Scale bar: 5 µm for larger images and 2 µm for higher magnification images. AGPAT, 1-acylglycerol-3-phosphate-O-acyltransferase; CHO, Chinese hamster ovary.

Fig. 1.

Localization of AGPAT3-GFP and AGPAT5-GFP to endo-membranes in cultured cells. A, B: CHO cells overexpressing AGPAT3-GFP were fixed in methanol and incubated with antibody sec61β specific for endoplasmic reticulum and lamin A/C specific for nuclear lamina and imaged for green and red fluorescence using fluorescence microscopy. Shown are representative images for AGPAT3 (green fluorescence), sec61β (red fluorescence), DAPI (blue fluorescence), colocalization channel (yellow fluorescence), and the merged image. C, D: Fluorescence images for AGPAT5-GFP expressed in CHO cells. Cells for sec61β and lamin A/C were processed as above. E: The AGPAT5-GFP expressing cells were incubated with MitoTracker Red dye, fixed in 4% paraformaldehyde, and imaged as before. Shown for each image is a single z-stack image, where the x and y axes show the z-stacks. Scale bar: 5 µm for larger images and 2 µm for higher magnification images. AGPAT, 1-acylglycerol-3-phosphate-O-acyltransferase; CHO, Chinese hamster ovary.

Fig. 2.

Enzymatic activity and Western blot of wild-type human V5-AGPAT3 and V5-AGPAT5 expressed in AD293 cells. A: The Western blot for the recombinant AGPAT3 and AGPAT5 proteins from the whole-cell lysate as probed with V5-antibody. Lysates from cells infected with LacZ were loaded as a negative control. The same blot was stripped and reprobed with full form GAPDH to demonstrate protein loading. B: The Western blot shows the presence of AGPAT5 but absence of AGPAT3 in the mitochondrial fraction. The blot was also probed for the mitochondria-specific protein prohibitin. C: The Western blot for the presence of AGPAT3 and AGPAT5 in the microsomal fraction of the cell. The blot was also reprobed with calnexin, specific for the microsomes. D: The AGPAT activity in whole-cell lysate for AGPAT3 as determined by conversion of ³H-LPA to ³H-PA in the presence of oleoyl-CoA and expressed as product (³H-PA) formed nmol per min per mg protein. The LPA to PA conversion by recombinant β-galactosidase adenovirus was used as a control. E: The enzymatic activity for AGPAT5. Not shown is the conversion of substrate in the absence of enzyme. Each bar represents mean ± SD from two independent experiments carried out in triplicate. The P value is shown above the bars. AGPAT, 1-acylglycerol-3-phosphate-O-acyltransferase; LacZ, recombinant adenovirus β-galactosidase.

Fig. 3.

Specificity of AGPAT3 and AGPAT5 for various LPAs. A: Shows LPA specificity of recombinant human AGPAT3 expressed in AD293 cells. Specificity of human recombinant AGPAT3 for various species of sn-1-lysophosphatidic acid acceptors was determined using radioactive oleoyl-CoA and arachidonoyl-CoA as donors. Activities are expressed as product (¹⁴C-PA) formed nmol per min per mg protein. All enzymatic activities were determined in two independent experiments in triplicate. Bar represents mean ± SD. The P values are shown above the bars. B: Shows LPA specificity of recombinant human AGPAT5 expressed in AD293 cells. The LPA specificity of human recombinant AGPAT5 was determined and expressed as for AGPAT3. All enzymatic activities were determined in two independent experiments in triplicate. Each bar represents mean ± SD. The P values are shown above the bars. AGPAT, 1-acylglycerol-3-phosphate-O-acyltransferase; LPA, lysophosphatidic acid.

Fig. 4.

Acyl-CoA specificity of recombinant human AGPAT3 and AGPAT5 expressed in AD293 cells. Specificity of human recombinant AGPAT3 (A) and AGPAT5 (B) for acyl-CoA donors was determined using sn-1-oleoyl-lysophosphatidic acid as an acceptor and various short-, medium-, and long-chain fatty acyl-CoA as donors. The enzymatic activity is compared with C18:1 (100%) (shown on top of the bar). Percent activity is marked only for those acyl CoA species which have least 20% of the activity compared with C18:1. All enzymatic activities were determined in two independent experiments in triplicate. Each bar represents mean ± SD. The P values are shown above the bars. AGPAT, 1-acylglycerol-3-phosphate-O-acyltransferase.

Fig. 5.

Specificity for acyltransferase activity of AGPAT3 and AGPAT5 for lysophospholipids LPA, LPC, LPG, LPI, LPE, and LPS. A: The enzymatic activity for AGPAT3 was determined as the conversion of various lysophospholipids to their corresponding [¹⁴C]phospholipids in the presence of [¹⁴C] oleoyl-CoA or [¹⁴C]arachidonoyl-CoA and expressed as [¹⁴C]phospholipids product formed in nmol per min per mg protein. LacZ was used as a control. Shown are the mean values from two independent assays carried out in triplicate. Each bar represents mean ± SD. The P values are shown above the bars. B: The enzymatic activity of AGPAT5 for various lysophospholipids (determined as described for AGPAT3). Shown are the mean values from two independent assays carried out in triplicate. Each bar represents mean ± SD. The P values are shown above the bars. AGPAT, 1-acylglycerol-3-phosphate-O-acyltransferase; LacZ, recombinant adenovirus β-galactosidase; LPA, lysophosphatidic acid; LPE, lysophosphatidylethanolamine; LPI, lysophosphatidylinositol; LPC, lysophosphatidylcholine; LPS, lysophosphatidylserine; LPG, lysophosphatidylglycerol.

Fig. 6.

Kinetic studies of recombinant human AGPAT3 and AGPAT5 expressed in AD-293 cells. A: The saturation curve of AGPAT3 enzymatic activity as determined by varying the concentration of oleoyl-LPA spiked with a known amount of [¹⁴C]oleoyl-CoA. B: The saturation curve of AGPAT3 enzymatic activity as determined by incubating increasing concentrations of LPA with a fixed amount of [¹⁴C]oleoyl-CoA. In A and B, the data were transformed to fit the Lineweaver-Burk plot to determine the K_m and V_max (shown as an insert). The enzymatic activities were determined in triplicate with each experiment carried out in triplicate. C: The saturation curve of AGPAT5 enzymatic activity as determined by varying the concentration of oleoyl-LPA spiked with a known amount of [¹⁴C]oleoyl-CoA. D: The saturation curve of AGPAT3 enzymatic activity as determined by incubating increasing concentrations of LPA with a fixed amount of [¹⁴C]oleoyl-CoA. In C and D, the enzymatic data were transformed as above to obtain the K_m and V_max (shown as an insert). Only representative graphs are shown. All experiments were performed in triplicate. The values were determined using GraFit software (Erithacus Software Ltd., Surrey, RH6 9YJ, UK). AGPAT, 1-acylglycerol-3-phosphate-O-acyltransferase; LPA, lysophosphatidic acid.