Cloning and characterization of murine 1-acyl-sn-glycerol 3-phosphate acyltransferases and their regulation by PPARalpha in murine heart

Abstract

AGPAT (1-acyl-sn-glycerol 3-phosphate acyltransferase) exists in at least five isoforms in humans, termed as AGPAT1, AGPAT2, AGPAT3, AGPAT4 and AGPAT5. Although they catalyse the same biochemical reaction, their relative function, tissue expression and regulation are poorly understood. Linkage studies in humans have revealed that AGPAT2 contributes to glycerolipid synthesis and plays an important role in regulating lipid metabolism. We report the molecular cloning, tissue distribution, and enzyme characterization of mAGPATs (murine AGPATs) and regulation of cardiac mAGPATs by PPARalpha (peroxisome-proliferator-activated receptor alpha). mAGPATs demonstrated differential tissue expression profiles: mAGPAT1 and mAGPAT3 were ubiquitously expressed in most tissues, whereas mAGPAT2, mAGPAT4 and mAGPAT5 were expressed in a tissue-specific manner. mAGPAT2 expressed in in vitro transcription and translation reactions and in transfected COS-1 cells exhibited specificity for 1-acyl-sn-glycerol 3-phosphate. When amino acid sequences of five mAGPATs were compared, three highly conserved motifs were identified, including one novel motif/pattern KX2LX6GX12R. Cardiac mAGPAT activities were 25% lower (P<0.05) in PPARalpha null mice compared with wild-type. In addition, cardiac mAGPAT activities were 50% lower (P<0.05) in PPARalpha null mice fed clofibrate compared with clofibrate fed wild-type animals. This modulation of AGPAT activity was accompanied by significant enhancement/reduction of the mRNA levels of mAGPAT3/mAGPAT2 respectively. Finally, mRNA expression of cardiac mAGPAT3 appeared to be regulated by PPARalpha activation. We conclude that cardiac mAGPAT activity may be regulated by both the composition of mAGPAT isoforms and the levels of each isoform.

Figure 1. Amino acid sequences of mAGPAT2 and hAGPAT2, and alignment of predicted protein sequences of mAGPAT gene family members

(A) Alignment of the amino acid residues from the complete amino acid sequences of mAGPAT2 and hAGPAT2 (GenBank® accession no. NM_006412). The amino acid sequence is shown in single-letter code. * indicates amino-acid identities of the aligned polypeptides. Numbering of amino acids begins with the first amino acid residue of predicted mAGPAT2. (B) The predicted protein sequences from five mAGPAT gene families were aligned with the ClustalW program (http://www.ebi.ac.uk). The conserved motifs are underlined and identical amino acid residues are indicated with an asterisk. GenBank® accession numbers are summarized in Table 1.

Figure 2. Expression of mAGPAT in vitro and in vivo and enzyme activities of mAGPATs

(A) Expression of mAGPAT2, mAGPAT3, mAGPAT4 and mAGPAT5 protein using the in vitro protein translation system. Reticulocyte lysates were prepared and proteins were [³⁵S]methionine radiolabelled and visualized by exposure to X-ray film (upper panel). UPRL, unprocessed lysate. AGPAT enzyme activities were determined as described in the Materials and methods section (lower panel). (B) Expression of V5-His epitope tag fused to the C-terminus of mAGPAT3 and mAGPAT5 in COS-1 cells (upper panel). Appearance of the mAGPAT5 band is greater since the protein contains 13 methionine residues instead of 11 and 7 methionine residues in mAGPAT2 and mAGPAT3 respectively. Molecular mass markers are indicated on the left. An equal amount of protein from cell lysate was used to determine relative mAGPAT enzyme activitities as described in the Materials and methods section (lower panel).

Figure 3. Specificity of mAGPAT2 activity

AGPAT activities of mAGPAT2 from in vitro (A) and in vivo (B) sources. The film exposure time was 24 h for (A) and (B). (A) Lane 1, unprogrammed reticulocyte lysate; lane 2, reticulocyte lysate containing mAGPAT2; lane 3, [1-¹⁴C]oleoyl-CoA. (B) Lane 1, COS-1 cell lysate; lane 2, COS-1 cell lysate from mAGPAT2 transfected cells; lane 3, [1-¹⁴C]oleoyl-CoA. (C) GPAT activities in reticulocyte lysates. (D) LPCAT activities in reticulocyte lysates. (E) LPEAT activities in reticulocyte lysates. (F) LPGAT activities in reticulocyte lysates. Lane 1, unprogrammed reticulocyte lysate. Lane 2, reticulocyte lysate containing mAGPAT2. Lane 3, radiolabelled substrate [1-¹⁴C]oleoyl-CoA. The film was exposed for 72 h. Representative autoradiographs are depicted.

Figure 4. mRNA expression profiles of mAGPAT isoforms in murine tissues

Total RNAs were extracted and RT–PCR was performed as described in the Materials and methods section. Gel images obtained from RT–PCR showed the expression of AGPAT mRNAs in murine tissues. Results were from three independent RT–PCRs and representative gels are shown.

Figure 5. AGPAT activity in wild-type and PPARα null mice treated with or without clofibrate

Mice were fed 0.5% clofibrate or normal chow consecutively for 14 days and hearts were collected. Whole cell extracts were prepared and AGPAT activity was determined as described. Lane 1, wild-type control; lane 2, wild-type fed clofibrate; lane 3, PPARα null control; lane 4, PPARα null fed clofibrate. Values represent the means±S.D. (n=5), *P<0.05.

Figure 6. Expression of mAGPAT mRNAs in wild-type and PPARα null mice treated with or without clofibrate

The collected mouse heart samples were used to determine mRNA levels of mAGPATs by RT–PCR as described in the Materials and methods section. The representative blots are shown. The densitometry of each band was normalized to the signal generated from β-actin (n=4 in each group).