J Biol Chem, Vol. 275, Issue 14, 9937-9945, April 7, 2000

The Genomic Organization, Complete mRNA Sequence, Cloning, and Expression of a Novel Human Intracellular Membrane-associated Calcium-independent Phospholipase A₂^*

David J. Mancuso, Christopher M. Jenkins, and Richard W. Gross

From the Division of Bioorganic Chemistry and Molecular Pharmacology, Departments of Medicine, Chemistry and Molecular Biology, and Pharmacology, Washington University School of Medicine, St. Louis, Missouri 63110

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

During the sequencing of the long arm of chromosome 7 in the Human Genome Project, a predicted protein product of 40 kDa was identified, which contained two ~10-amino acid segments homologous to the ATP and lipase consensus sequences present in the founding members of a family of calcium-independent phospholipases A₂. Detailed inspection of the identified sequence (residues 79,671-109,912 GenBank^TM accession no. AC005058) demonstrated that it represented only a partial sequence of a larger undefined polypeptide product. Accordingly, we identified the complete genomic organization of this putative phospholipase A₂ through analyses of previously published expressed sequence tags, PCR of human heart cDNA, and 5'-rapid amplification of cDNA ends. Polymerase chain reaction and Northern blotting demonstrated a 3.4-kilobase message, which encoded a polypeptide with a maximum calculated molecular weight of 88476.9. This 3.4-kilobase message was present in multiple human parenchymal tissues including heart, skeletal muscle, placenta, brain, liver, and pancreas. Cloning and expression of the protein encoded by this message in Sf9 cells resulted in the production of two proteins of apparent molecular masses of 77 and 63 kDa as assessed by Western analyses utilizing immunoaffinity-purified antibody. Membranes from Sf9 cells expressing recombinant protein released fatty acid from sn-2-radiolabeled phosphatidylcholine and plasmenylcholine up to 10-fold more rapidly than controls. The initial rate of fatty acid release from the membrane fraction was 0.3 nmol/mg·min. The recombinant protein was entirely calcium-independent, had a pH optimum of 8.0, was inhibited by (E)-6-(bromomethylene)-3-(1-naphthalenyl)-2H-tetrahydropyran-2-one (IC₅₀ = 3 µM), and was predominantly present in the membrane-associated fraction. Collectively, these results describe the genomic organization, complete mRNA sequence, and sn-2-lipase activity of a novel intracellular calcium-independent membrane-associated phospholipase A₂.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Phospholipases A₂ catalyze the esterolytic cleavage of fatty acids from the sn-2-position of phospholipids, thereby regulating the release of lipid second messengers (e.g. eicosanoids and lysophospholipids), growth factors (lysophosphatidic acid), and membrane physical properties (1-5). In most cell types, the availability of nonesterified arachidonic acid is the rate-limiting step in the production of biologically active eicosanoid metabolites (1, 2, 6, 7). Thus, phospholipase A₂ activity modulates cellular growth programs (e.g. peroxisome proliferation (by prostaglandin J₂)), inflammation (e.g. prostaglandins and leukotrienes), vascular tone (e.g. 20-hydroxyeicosatetraenoic acid), and ion channel function (e.g. P450 products and arachidonic acid) (1, 2, 6, 7-11). Accordingly, substantial attention has focused on the molecular identification of the polypeptides that catalyze phospholipase A₂ activity, regulate its kinetic properties, and facilitate its topologic and topographic distribution in normal and stimulated cells.

Decades of painstaking research eventually illuminated several distinguishing kinetic and physical characteristics of the families of phospholipases A₂ that facilitated their categorization into several broad classes of enzymes based upon their requirement for calcium ion in in vitro activity assays (i.e. millimolar, nanomolar, or no calcium dependence) (e.g. see Refs. 5 and 12-16). For example, secretory phospholipases A₂ (secretory PLA₂)¹ were distinguished by their low molecular mass (14-18 kDa), heat stability, and obligatory dependence upon high (millimolar) concentrations of calcium ion for catalytic activity (5, 16, 17). A second group of calcium-facilitated phospholipases A₂ (i.e. the cPLA₂ family) did not absolutely require calcium ion for hydrolysis, although nanomolar amounts of calcium ion dramatically augmented their in vitro activity (13, 18) and facilitated their translocation to subcellular membrane targets (19). Finally, a third group of enzymes were identified that were entirely calcium-independent in in vitro activity assays (i.e. the iPLA₂ family) (15, 20, 21) and could be distinguished by their exquisite sensitivity to inhibition by (E)-6-(bromomethylene)-3-(1-naphthalenyl)-2H-tetrahydropyran-2-one (BEL) at 1-2 µM concentration (22, 23).

Initial application of molecular biologic approaches to the phospholipase A₂ field identified founding members and mechanistic insights into each of these three types of phospholipase A₂ catalytic activities (16, 24-27). For example, the secretory PLA₂s employ a calcium ion to polarize the sn-2-carbonyl for attack by a histidine-activated H₂O molecule, while the intracellular phospholipases employ a nucleophilic serine (17, 22, 24-27). Moreover, the cPLA₂ family is readily distinguished from the iPLA₂ family by the presence of a GXSGS consensus lipase motif, while the iPLA₂ family utilizes a GXSTG consensus motif. In addition, the iPLA₂ (but not the cPLA₂) gene family possesses a consensus sequence for nucleotide binding (26, 27). These insights have greatly accelerated our progress in the understanding of the molecular identities of the polypeptides responsible for phospholipase A₂ catalysis and their mechanisms of regulation in normal and disease states. More recently, global efforts aimed at identifying the complete human genome sequence have yielded a vast array of sequence information that has further delineated the role of individual phospholipases in biologic processes. For example, two recently described phospholipases A₂ (i.e. cPLA₂ and cPLA₂) were identified from initial insights gleaned from protein and nucleotide data bases (28, 29).

During the sequencing of the long arm of chromosome 7 in the Human Genome Sequencing Project, a predicted protein product of 40 kDa was identified, which contained two ~10-amino acid segments homologous to the ATP and lipase consensus sequences present in the founding members of calcium-independent phospholipases A₂ (i.e. iPLA₂ (26) and iPLA₂ (27)). However, close inspection of the Human Genome Sequencing Project sequence demonstrated that it represented only the partial sequence of a larger undefined polypeptide product. Herein, we report the entire genomic organization, complete mRNA sequence, cloning, expression, and initial activity analyses of the protein encoded by this gene, which we term iPLA₂.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- [-³²P]dCTP (6000 Ci/mmol) and ECL detection reagents were purchased from Amersham Pharmacia Biotech. A human heart cDNA library was purchased from Stratagene, Inc. Human heart Marathon cDNA, QuickClone human skeletal cDNA, and human MTN multiple tissue Northern blots were purchased from CLONTECH. For PCR, a Perkin-Elmer Thermocycler was employed, and all PCR reagents were purchased from PE Biosystems. The pGEM-T vector and the TnT Quick Coupled Transcription/Translation System were obtained from Promega. Vector pcDNA1.1 was purchased from Invitrogen. Culture media, Cellfectin and LipofectAMINE reagents for transfection of baculovirus vectors, and competent DH110Bac Escherichia coli cells were purchased from Life Technologies, Inc. and used according to the manufacturer's protocol. QIAfilter plasmid kits and QIAquick gel extraction kits were obtained from Qiagen, Inc. Keyhole limpet hemocyanin was obtained from Pierce. L--Dipalmitoyl-2-[1-¹⁴C]palmitoyl phosphatidylcholine, L--1-palmitoyl-2-[1-¹⁴C]oleoyl phosphatidylcholine, L--1-palmitoyl-2-[1-¹⁴C]linoleoyl phosphatidylcholine, and L--1-palmitoyl-2-[1-¹⁴C]arachidonyl phosphatidylcholine were purchased from NEN Life Science Products. 1-O-(Z)-Hexadec-1'-enyl-2-[9,10-³H]oleoyl sn-glycerol-3-phosphocholine was synthesized and purified as described previously (30). BEL was obtained from Calbiochem. Most other reagents were obtained from Sigma. Searches of EMBL and NCBI data bases were performed using the Basic Local Alignment Search Tool (BLAST) (NCBI). Alignments of all sequences were performed with the MultAlign computer program (31).

PCR Amplification of iPLA₂-- For typical PCR analysis, a 30-cycle program was employed with steps at 53 °C for 30 s, 72 °C for 2 min, and 94 °C for 30 s per cycle. iPLA₂ was amplified utilizing oligonucleotides that flanked the predicted 5'- and 3'-coding region, M444 (5'-TTTTGTCGACATGTCTATTAATCTGACTGTAGATA-3') and M449 (5'-GCATACTCGAGTCACAATTTTGAAAAGAATGGAAGTCC-3'), respectively. PCR screening was performed utilizing human skeletal muscle cDNA (0.5 ng), human heart Marathon cDNA (0.5 ng), and a human heart cDNA library (~1 × 10⁹ plaque-forming units) as templates. To directly compare differences between our sequences and those previously reported, PCR amplification of the iPLA₂ sequence present in the original BAC genomic clone RG054 DO4 (Research Genetics) was used as template, and PCR was performed with primers M452 (5'-GTACATACGGTGGACAAGCCTA-3') and M446 (5'-CATTCCTCTCCCTTTCACTGGATCCACATAGCC-3'). All PCR products were resolved by 1% agarose gel electrophoresis. Candidate bands were extracted from the agarose gel using a QIAquick Gel extraction kit followed by blunt end ligation into the pGEM-T Vector (Promega) by standard procedures (32). Following bacterial transformation and growth of transformants, plasmids were purified using a QIAfilter plasmid kit (CLONTECH) and subjected to automated sequence analysis using either an ABI 373S or 377XL automated DNA sequencer (PE Biosystems).

Northern Blot Analysis-- Full-length iPLA₂ amplified by PCR was prepared for use as a probe by radiolabeling with [³²P]dCTP for 1 h at 37 °C in the presence of Ready-To-Go labeling beads (Amersham Pharmacia) according to instructions provided by the manufacturer. The radiolabeled probe was purified by gel filtration employing a 1-ml Sepharose G-25 spin column. For Northern analysis, an MTN blot (CLONTECH) containing 2 µg of poly(A)⁺ RNA/lane from human brain, heart, pancreas, liver, lung, and placenta tissue was prehybridized at 68 °C for 30 min in hybridization buffer, hybridized for 1 h at 68 °C with radiolabeled iPLA₂ (2 × 10⁶ cpm/ml), and washed in 2× SSC and 0.1% SDS twice for 30 min, followed by two washes with 0.1× SSC and 0.1% SDS for 40 min each at 50 °C as per the manufacturer's instructions. Hybridized sequences were identified by autoradiography for 16 h.

5'-RACE-- For 5'-RACE, a 45-cycle program with steps at 58 °C for 30 s, 72 °C for 2 min, and 94 °C for 30 s per cycle was employed. Human heart Marathon-Ready cDNA was used as template (0.5 ng), and primer AP1 (CLONTECH) was paired with M460 (5'-GAAAACCTCTTTGTAGACTGATGTGGCTTATCCTCCAG-3') to amplify products. Products were analyzed by electrophoresis utilizing a 1% agarose gel and visualized by ethidium bromide staining. PCR products were excised from the gel, purified with a QIAquick gel extraction kit, and subcloned into pGEM-T vector (Promega) for sequencing and alignment with the iPLA₂ sequence.

In Vitro Translation-- A full-length iPLA₂ construct in pcDNA1.1 (1 µg) was used in a coupled transcription/translation rabbit reticulocyte lysate system (Promega) with RNA synthesis from the T7 promoter of pcDNA1.1 using T7 RNA polymerase and translation in the presence of 20 µCi of [³⁵S]methionine for 90 min according to the manufacturer's instructions. Labeled protein products were resolved on a 10% SDS-polyacrylamide gel followed by autoradiographic visualization.

Generation and Purification of iPLA₂ Antibodies-- Anti-iPLA₂ polyclonal antibodies were made by immunizing rabbits with the synthetic peptide CENIPLDESRNEKLDQ. The peptide was conjugated to maleimide-activated keyhole limpet hemocyanin by incubation for 2 h at 22 °C followed by dialysis according to the manufacturer's instructions. After two booster injections of the peptide conjugate spaced 2 weeks apart, serum was collected, and antibodies against the peptide were affinity-purified using a thiopropyl-Sepharose column to which the peptide had been covalently coupled according to the instructions of the manufacturer.

iPLA₂ Expression and Sf9 Cell Fractionation-- PCR amplification with the primer pair m444 (5'-TTTTGTCGACATGCTATTAATCTGACTGTAGATA-3') and m458 (5'-GCATAGCATGCTCACAATTTTGAAAAGAATGGAAGTCC-3') was used to engineer appropriate restriction sites onto iPLA₂ for subsequent subcloning into SalI/SphI restriction sites of a pFASTBAC vector (Life Technologies, Inc.). The iPLA₂ and flanking sequences in pFASTBAC were sequenced in their entirety on both strands to verify the integrity of the sequence.

Sf9 cells were grown and infected as described previously in detail (33). In brief, Spodoptera frugiperda (Sf9) cells were cultured in 100-ml flasks equipped with a magnetic spinner containing supplemented Grace's medium (34). Sf9 cells at a concentration of 1 × 10⁶ cells/ml were prepared in 50 ml of growth medium and incubated at 27 °C for 1 h prior to infection with either wild-type virus or recombinant virus containing human iPLA₂ cDNA. After 48 h, cells were pelleted by centrifugation, resuspended in ice-cold phosphate-buffered saline, and repelleted. All subsequent operations were performed at 4 °C. The supernatant was decanted, and the cell pellet was resuspended in 5 ml of homogenization buffer (25 mM imidazole, pH 8.0, 1 mM EGTA, 1 mM dithiothreitol, 0.34 M sucrose, 20 µM transepoxysuccinyl-L-leucylamido-(4-guanidino) butane, and 2 µg/ml leupeptin). Cells were lysed at 0 °C by sonication (20 1-s bursts utilizing a Vibra-cell sonicator at a 30% output) and centrifuged at 100,000 × g for 1 h. The supernatant was saved (cytosol), and the membrane pellet was washed with homogenization buffer and resuspended using a Teflon homogenizer in 6 ml of homogenization buffer. After brief sonication (10 1-s bursts), the mixture was subjected to recentrifugation at 100,000 × g for 1 h. After removal of the supernatant, the membrane pellet was resuspended in 1 ml of homogenization buffer using a Teflon homogenizer and subsequently sonicated at 0 °C for 5 × 1-s bursts.

Immunoblot Analysis-- Sf9 cell cytosol and membrane proteins were separated by SDS-polyacrylamide gel electrophoresis (35) and transferred to Immobilon-P membranes by electroelution in 10 mM CAPS, pH 11, containing 10% methanol. Dry powdered milk (5% (w/v) in 20 mM Tris·HCl, pH 7.4, containing 137 mM NaCl and 0.1% Tween 20) was used to block nonspecific binding sites before incubation with the primary antibody (prepared as described above). Secondary antibody (anti-rabbit F(ab')₂ IgG-horseradish peroxidase conjugate) was incubated with the blot for 1 h, and immunoreactive bands were visualized utilizing an ECL detection system.

Phospholipase A₂ Enzymatic Assay-- Calcium-independent phospholipase A₂ activity was measured by quantitating the release of radiolabeled fatty acid from various radiolabeled phospholipid substrates in the presence of membrane fractions from Sf9 cells infected with wild-type or recombinant human iPLA₂ containing baculovirus. Reactions (200 µl) were incubated for up to 5 min at 37 °C in reaction buffer (100 mM Tris acetate, pH 8.0, containing 1 mM EGTA) prior to their termination by the addition of 100 µl of 1-butanol and vortexing. Phospholipids and fatty acids extracted into the butanol phase were separated by thin layer chromatography using Whatman Silica 60A prescored plates employing a mobile phase of 70:30:1 petroleum ether/ethyl ether/glacial acetic acid (v/v/v). Radiolabeled fatty acids were identified by staining of an overlaid fatty acid standard by exposure to iodine vapor. Regions corresponding to fatty acids were scraped into scintillation vials and subsequently quantitated by scintillation spectrometry after the addition of fluor. For experiments employing BEL, reactions were preincubated in reaction buffer for 3 min in the presence of selected concentrations of BEL or vehicle (EtOH) prior to the addition of radiolabeled substrate.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Characterization of the Full-length Message Encoding iPLA₂-- Inspection of the sequence encoding the putative 40-kDa phospholipase reported by the Human Genome Sequencing Project (BAC clone RG054D04; GenBank^TM accession no. AC005058) demonstrated that it did not begin with an initiator methionine codon. Accordingly, we performed a TBLASTN data base search (36) of GenBank^TM to find expressed sequence tags (ESTs) that could align with the 5'-end of the putative iPLA₂ sequence. EST clones vz36b01.ri Soares 2NbMT Mus musculus cDNA IMAGE:1328521 and Rattus norvegicus cDNA UI-R-C0-hp-c-06-0-UI (accession nos. AA915561 and AA998901, respectively) were found to overlap with the iPLA₂ sequence, thereby extending the known 5' sequence an additional 360 nt upstream (Fig. 1). Four other EST clones (Stratagene Homo sapiens colon cDNA clone IMAGE:588479, accession no. AA143503; Stratagene H. sapiens cDNA clone IMAGE:647744, accession no. AA205258; normalized rat ovary, Bento Soares Rattus sp. cDNA clone ROVAA46, accession no. AA801084; and NCI_CGAP_GCB1 H. sapiens cDNA clone IMAGE:825005, accession number AA504219) were also present in the data base and are in close spatial proximity with the putative PLA₂. However, when aligned with the BAC clone sequence, the 3'-end of the EST AA504219 sequence and the 5'-end of EST AA998901 are separated by a 150-nt gap (Fig. 1). Moreover, when the EST AA998901 sequence is back-translated through this gap and into EST AA504219, a continuous reading frame results. Thus, by overlapping known EST sequences with the 5'-end of iPLA₂ and back-translation through the 150-nt gap, the sequence could be extended approximately 1.2 kb upstream from the predicted GenBank^TM protein. The furthest upstream ATG codon that remained in frame with the coding sequence was located 1210 nt upstream from the originally reported sequence. Translation of the reported gene sequence further 5' from nt 122761 in BAC clone RG054D04 results in stop codons in all three reading frames. We performed PCR analysis using primers corresponding to the most 5' candidate initiator methionine and the known 3' stop codon (nt 79,673) in the gene sequence. PCR of human heart cDNA human and skeletal muscle libraries utilizing primers M444 and M949 gave rise to a single band, which was 2.4 kb in length (Fig. 2).

View larger version (6K): [in this window] [in a new window]   Fig. 1.   Mapping of the ESTs overlapping the iPLA2 Sequence. Amino acid numbering starting from the most 5' potential translation initiation site is indicated at the top, with a schematic representation of the previously identified 40-kDa region of iPLA2 shaded below. The EST sequences (solid bars) were utilized in mapping the full-length sequence of iPLA2 and identification of a potential N-terminal initiator methionine. ESTs used to map the full-length coding sequence of iPLA2 correspond to clones vz36b01.ri Soares 2NbMT M. musculus cDNA IMAGE:1328521 (accession no. AA915561), R. norvegicus cDNA UI-R-C0-hp-c-06-0-UI (AA998901, Stratagene H. sapiens colon cDNA clone IMAGE:588479 (accession no. AA143503), Stratagene H. sapiens cDNA clone IMAGE:647744 (accession no. AA205258), normalized rat ovary, Bento Soares Rattus sp. cDNA clone ROVAA46 (accession no. AA801084), and NCI_CGAP_GCB1 H. sapiens cDNA clone IMAGE:825005 (accession no. AA504219). Accession numbers for each EST are indicated above each solid bar. The arrow at the bottom indicates the position and orientation of PCR primers M444 (5'-end) and M449 (3'-end) used to amplify full-length 2.4-kb iPLA2 coding sequence from human heart cDNA.

View larger version (45K): [in this window] [in a new window]   Fig. 2.   PCR amplification of human iPLA2. PCR was performed using human heart cDNA (0.5 ng) (lane 1), a cDNA library prepared from human heart (lane 2), human skeletal muscle cDNA (lane 3), and a blank control (lane 4) as templates as described under "Experimental Procedures." PCR primers M444 and were utilized with M449 positioned at the 5'- and 3'-ends of iPLA2 coding sequence (respectively) in 30 cycles of amplification (53 °C for 30 s, 72 °C for 2 min, and 94 °C for 30 s). PCR products were analyzed on a 1% agarose gel and visualized by ethidium bromide staining. Molecular size markers are shown on the left in kb. The arrow indicates the size of the major PCR band (2.4 kb).

Sequencing of iPLA₂-- The PCR product was subcloned into pGEM-T and sequenced in both directions (Fig. 3). Based on amino acid residue 1 being the initiator methionine, the message encoded a 782-amino acid polypeptide with a calculated molecular weight of 88,476.9. Contained within this sequence were an ATP binding motif (amino acid residues 449-454) and a lipase consensus sequence (amino acid residues 481-485) as well as multiple potential cAMP phosphorylation sites, PKC phosphorylation sites, CK2 phosphorylation sites, and a microbody C-terminal targeting sequence as determined by a Prosite pattern search (37, 38). Kyte-Doolittle hydrophobicity analysis (39) demonstrated that this putative iPLA₂ had two major hydrophobic domains, one at the extreme putative N terminus and a second centered near the lipase consensus sequence (Ser⁴⁸³) (Fig. 4).

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Schematic representation of the strategy utilized for sequencing the PCR product encoding iPLA2. The size in nucleotides is indicated at the top, below which is a representation of the full-length iPLA2 coding sequence with the locations of the ATP binding (nt 1345-1362) and lipase (nt 1441-1445) consensus sequences indicated. The arrows represent the direction and length of the sequences obtained from individual sequencing reactions. At the bottom is a representation of the region of the differences from our sequencing and that previously reported in the BAC clone RG054D04.

View larger version (11K): [in this window] [in a new window]   Fig. 4.   Hydropathy plot of the deduced amino acid sequence of human iPLA2. The deduced amino acid sequence of iPLA2 was analyzed using the method of Kyte and Doolittle (39) with a window size of 17 amino acids. Negative values on the y axis represent increasing hydrophobicity.

This sequence analysis demonstrated 10 putative differences located between amino acid residues 351 and 358 in comparison with the original GenBank^TM report. Three pieces of information substantiate the integrity of the sequence shown in Fig. 3. First, rat clone UI-R-C0-hp-c-06-0-UI (accession number AA998901) EST sequence agreed in its entirety with the sequence we identified. Second, our sequencing of this region in both directions from multiple different libraries gave identical results. Third, we sequenced the original BAC clone in both directions and obtained identical results to those shown in Fig. 3. Accordingly, we conclude that multiple errors are present in GenBank^TM sequence RG054D04 in residues 351-358 and that the sequence shown in Fig. 3 is correct.

To determine if this PCR product represented the true 5'-end of the coding sequence and to locate additional message sequence 5' of nt 122,761, we performed 5'-RACE utilizing a reverse primer near the 5'-end of the 2.4-kb PLA₂ PCR product (M460) and a sense primer (AP1) to the adapter sequence flanking the cDNA template. This maneuver extended the 5' sequence from the putative initiator methionine residue an additional 225 base pairs upstream. Within the 225-base pair sequence, it was obvious that no additional coding regions were present, since this sequence contained stop codons in all three reading frames. The most 5' ATG site that is in frame with the iPLA₂ sequence is located at nt 226. Additional in frame ATG sites are located at nt 526, 589, and 886. Initiation of polypeptide synthesis at these potential methionine start sites would result in polypeptides of 77, 74, and 63 kDa, respectively. At the 3'-end of the gene, a signal site for 3' poly(A) processing (AATAAA) (40) was identified 757 nt 3' of the TGA stop signal. The actual cleavage site for poly(A) addition usually occurs 10-30 nt 3' of the poly(A) signal, and at this location, a CA is the preferred sequence immediately 5' to the cleavage site (41). Additionally, GU-rich or U-rich elements are also typically found downstream of the poly(A) site (42). Since a CA dinucleotide occurs 35 nt 3' of the iPLA₂ poly(A) signal sequence and highly U-rich sequences occur at nucleotides 3339-3343 and 3373-3392, the likely point of poly(A) addition occurs following nucleotide 3372. Accordingly, these results identify a putative transcription initiation site, 5'-untranslated region, coding sequence, and 3'-untranslated region that together result in a 3.4-kb mature message (Fig. 5).

View larger version (69K): [in this window] [in a new window]   Fig. 5.   Full-length human iPLA2 message sequence. The deduced amino acid sequence is shown below the nucleotide sequence. The 225-nt 5'-noncoding region determined by 5'-RACE is shown with numbering beginning at the putative transcription initiation site. Potential alternative initiator methionines at amino acid residues 1, 101, 122, and 221 are in boldface type. Within the coding region, the ATP binding (GXGXXG) and lipase (GXSTG) motifs are underlined, while the C-terminal peroxisome localization signal is double underlined. The location of the polyadenylation motif (AATAAA) within the 3'-noncoding region is underlined, and the presumed site for poly(A) addition 792 nt after the TGA stop codon occurs after a CA dinucleotide (after nt 3372), which is indicated by a triangle. The 3.4-kb message sequence is predicted to encode a 782-amino acid protein with a predicted molecular weight of 88,476.9.

To determine if this 3.4-kb message was the full-length (or nearly full-length) message or if additional, as yet unidentified, regions of the gene were transcribed to potentially serve a promoter function, Northern blot analysis was performed. Northern blotting of human heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas mRNA demonstrated that each of these tissues possessed a 3.4-kb message that tightly bound to the full-length probe (Fig. 6). The largest amount of message was present in the heart, followed by the placenta and skeletal muscle, with smaller amounts of message in the brain, kidney, pancreas, lung, and liver. Within the limits of resolution of this technique, only a single band was identified. Collectively, these results demonstrate that the identified message is either full-length or nearly full-length. We cannot exclude the possibility that the in vivo message extends slightly beyond the residue that we identified, but multiple additional 5'-RACE reactions from multiple libraries did not identify additional sequence.

View larger version (52K): [in this window] [in a new window]   Fig. 6.   Northern blot of human mRNA obtained from multiple human tissues. The tissue distribution of iPLA2 was examined by hybridization of 32P-labeled full-length iPLA2 with ~2 µg of human poly(A)+ mRNA and autoradiography as described under "Experimental Procedures." Lane 1, heart; lane 2, brain; lane 3, placenta; lane 4, lung; lane 5, liver; lane 6, skeletal muscle; lane 7, kidney; lane 8, pancreas. The positions of RNA size markers are shown. Relative size in kb is indicated on the left based on the mobility of the RNA standard ladder.

iPLA₂ Genomic Organization-- Examination of the sequence upstream from the putative transcription start site did not reveal the presence of either TATA box or CAAT box consensus sequences. The full-length message of iPLA₂ was aligned with BAC genomic clone RG054D04 sequence to determine the location of intron/exon boundaries. Two intron/exon boundaries utilizing conventional AG/GT splicing rules (43) were identified within the iPLA₂ 5'-untranslated region by alignment with the genomic clone sequence. Thus, exon 1 of the iPLA₂ gene is 96 nt in length (nucleotides 133,299-133,394 of BAC genomic clone RG054D04) and is followed by a 4.5-kb intron. Exon 2 is 46 nt in length (nucleotides 128,746-128,791 of BAC genomic clone RG054D04) and is followed by a 5.9-kb intron. The first candidate ATG start site occurs 84 nt downstream from the start of exon 3 (nucleotides 121,692-122,844 of BAC genomic clone RG054D04), which is 1151 nt in length and followed by a 139-nt intron 3. Exon 4 (nucleotides 121,411-121,522 of BAC genomic clone RG054D04) is 139 nt in length and followed by an 11.5-kb intron. Exon 5 begins the coding sequence previously reported in the GenBank^TM BAC clone report (Fig. 7). Based on these findings, the iPLA₂ gene on chromosome 7 is 54 kb in size and contains 11 exons.

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Genomic organization of the human iPLA2 gene. The intron-exon boundaries of the iPLA2 gene are shown in scale (in kilobases). The 11 exons of the iPLA2 gene are indicated as open boxes. Spaces between the exons represent the relative sizes of the 10 introns contained within the iPLA2 gene. Regions of the gene that correspond to the ATP binding, lipase, and peroxisomal localization consensus sequences are indicated in exons 5, 6, and 11, respectively. The dashed box enclosing exons 5-11 corresponds to previously identified exons in the BAC genomic clone (GenBankTM accession no. AC005058). The boxes at the bottom indicate the nucleotide numbers (corresponding to the original BAC genomic clone report) with the sizes of each exon in nucleotides and in amino acids shown within.

Expression of iPLA₂ in an in Vitro Reticulocyte Lysate Translation System and in Sf9 Cells Infected with Baculovirus Encoding iPLA₂-- To determine the molecular weight of the protein(s) translated by this message, the 2.4-kb PCR product in pcDNA1.1 vector was incubated with an in vitro reticulocyte lysate translation system in the presence of [³⁵S]methionine. Translated radiolabeled proteins were resolved by SDS-polyacrylamide gel electrophoresis and visualized by autoradiography. Two radiolabeled protein products were detected corresponding to molecular masses of ~77 and 63 kDa (Fig. 8). These products correspond to translation initiation at methionine residues 101 and 221 (Fig. 5). The possibility that translation occurred at methionine residue 1 and that the observed bands are proteolytic products of a larger polypeptide precursor cannot be definitively excluded. To compare the results of in vitro translation with translation in an intact eukaryotic cell, the 2.4-kb message was inserted into the Sf9 cell vector pFASTBAC. Spinner cultures of Sf9 cells were infected with either wild type pFASTBAC or pFASTBAC containing the 2.4-kb message encoding iPLA₂. Western analysis of membrane fractions from Sf9 cells demonstrated the presence of two major bands corresponding to molecular masses of 77 and 63 kDa, which were present in the membrane fraction of cells infected with vector harboring iPLA₂ but not in Sf9 cell cultures infected with wild type virus (Fig. 9). In contrast, cytosolic fractions from control or iPLA₂-transfected cells did not contain any detectable immunoreactive protein. Collectively, these results suggest the presence of two translation initiation start sites at methionine residues 101 and 221, but the possibility that proteolytic processing generated some of the immunoreactive bands cannot be ruled out.

View larger version (61K): [in this window] [in a new window]   Fig. 8.   In vitro translation of human iPLA2. Sizes of molecular weight standards in kDa are shown to the left. The full-length iPLA2 PCR product was cloned into pcDNA1.1 vector so that the T7 promoter region of pcDNA1.1 was upstream from the iPLA2 sequence. For coupled in vitro transcription/translation RNA was synthesized from 1 µg of iPLA2-pcDNA1.1 in the presence of T7 RNA polymerase. Translation of RNA in the rabbit reticulocyte lysate system was performed in the presence of [35S]methionine as described under "Experimental Procedures." Following translation, 5 µl of the labeled product was boiled for 2 min in SDS loading buffer, and protein products were electrophoresed and visualized by autoradiography. In vitro translated products corresponding to iPLA2 () and a negative control (ctl) are indicated.

View larger version (61K): [in this window] [in a new window]   Fig. 9.   Western analysis of iPLA2 expression in Sf9 cells. Sf9 cells were infected with either wild-type baculovirus (Ctl) or recombinant baculovirus encoding human iPLA2 (). At 48 h postinfection, the cells were collected, and cytosolic (Cyto) and membrane (Memb) fractions were prepared as described under "Experimental Procedures." Proteins (100 µg/lane) from each fraction were loaded on a 10% polyacrylamide gel, resolved by SDS-polyacrylamide gel electrophoresis, transferred to a polyvinylidene difluoride membrane, and incubated with immunoaffinity-purified antibody directed against iPLA2. Following incubation with an anti-rabbit IgG horseradish peroxidase conjugate, immunoreactive bands were visualized by ECL. The results are typical of three independent experiments.

To determine if the Sf9 cells infected with recombinant baculovirus encoding iPLA₂ catalyze the hydrolysis of phospholipids, cytosolic and membrane fractions were prepared as described under "Experimental Procedures." Phospholipase A₂ activity was assessed by quantifying the release of radiolabeled sn-2-fatty acid from 1-hexadecanoyl-2-[1-¹⁴C]octadec-9'-enoyl-sn-glycero-3-phosphocholine as a function of time. Membrane fractions from cells infected with vector harboring the iPLA₂ insert released fatty acid from the sn-2-position of phosphatidylcholine ~10-fold faster than from membrane fractions prepared from wild type vector controls (Fig. 10). Liberation of sn-2-radiolabeled fatty acid was nearly linear for 2 min with a velocity of ~0.3 nmol/mg·min. Phospholipase A₂ activity was entirely calcium-independent (0-10 mM) and possessed a pH optimum of 8.0 (data not shown). In contrast, phospholipase A₂ activities in the cytosolic fractions from cells infected with wild type vector and in vector harboring the iPLA₂ insert were found to be similar (data not shown). Incubation of membrane fractions containing iPLA₂ with phospholipids containing three distinct types of unsaturated fatty acids at the sn-2-position (oleic, linoleic, and arachidonic acids) gave similar specific activities that were each 6-10 times greater than activities manifest in membrane preparations derived from cells infected with wild type vector (Fig. 11A). Interestingly, incubation of membranes from Sf9 cells infected with vector harboring the iPLA₂ insert with dipalmitoyl phosphatidylcholine did not show any increase in phospholipase activity in comparison with cells infected with wild type vector.

View larger version (20K): [in this window] [in a new window]   Fig. 10.   Initial rate analysis of iPLA2 phospholipase A2 activity. Sf9 cells were infected with either wild-type baculovirus () or recombinant with baculovirus encoding human iPLA2 (). Membrane fractions from control or cells expressing iPLA2 were incubated with 40 µM L--1-palmitoyl-2-[1-14C]oleoyl phosphatidylcholine in 100 mM Tris-HAc, pH 8.0, containing 1 mM EGTA at 37 °C. Aliquots of the reaction were removed at the indicated times, and the amount of [1-14C]oleic acid released was quantitated as described under "Experimental Procedures." Results are representative of three independent experiments.

View larger version (27K): [in this window] [in a new window]   Fig. 11.   Substrate selectivity profile of iPLA2 phospholipase A2 activity. A, membrane fractions from Sf9 cells infected with either wild-type baculovirus (stippled bars) or recombinant baculovirus encoding human iPLA2 (solid bars) were incubated in 100 mM Tris·HAc, pH 8.0, containing 1 mM EGTA for 2 min at 37 °C in the presence of either 40 µM L--dipalmitoyl-2-[1-14C]phosphatidylcholine (PPPC), L--1-palmitoyl-2-[1-14C]oleoyl phosphatidylcholine (POPC), L--1-palmitoyl-2-[1-14C]linoleoyl phosphatidylcholine (PLPC), or L--1-palmitoyl-2- [1-14C]arachidonyl phosphatidylcholine (PAPC). Reactions were terminated by butanol extraction, and radiolabeled fatty acids were resolved by TLC and quantitated by scintillation spectrometry. Results are representative of three independent experiments. B, membrane fractions from control or iPLA2-transfected Sf9 cells (as described above) were incubated with either 40 µM L--1-palmitoyl-2-[1-14C]oleoyl phosphatidylcholine or 40 µM 1-O-(Z)-hexadec-1'-enyl-2-[9,10-3H]oleoyl-sn-glycerol-3-phosphocholine (Plasmenyl-PC) for 2 min at 37 °C. Reaction products were extracted into butanol, separated by TLC, and quantitated by scintillation spectrometry.

Native Sf9 cell membranes (membranes derived from Sf9 cells in the absence of pFASTBAC infection) contain large amounts of lysophospholipase activity (>10 nmol/mg·min). Accordingly, definitive and precise regiospecific analysis of the initial site of hydrolysis in an unpurified preparation is difficult. To identify the major site of hydrolysis (i.e. sn-1 versus sn-2), radiolabeled 1-O-(Z)-hexadec-1'-enyl-2-[9,10-³H]octadec-9'-enoyl-sn-glycero-3-phosphocholine (plasmenylcholine) was synthesized and incubated with membranes containing iPLA₂, which resulted in a 7-fold increase in measurable phospholipase A₂ activity in comparison with controls (Fig. 11B). Since the rate of hydrolysis utilizing plasmenylcholine (which contains a relatively enzymatically resistant vinyl ether linkage at sn-1) was similar to that utilizing sn-2-[1-¹⁴C]phosphatidylcholines, these results support the concept that hydrolysis occurred predominantly at the sn-2-position. Of course, the rigorous and detailed kinetic characterization and regiospecific analysis of iPLA₂ activity awaits the solubilization and purification of each iPLA₂ isoform to homogeneity and detailed kinetic analysis.

Inhibition of iPLA₂ by BEL-- BEL has previously been shown to be a potent and irreversible mechanism-based inhibitor of myocardial cytosolic and membrane-associated calcium-independent phospholipase A₂ with nearly complete inhibition of myocardial cytosolic calcium-independent PLA₂ at concentrations of ~2-5 µM (22, 44-46). Preincubation of membranes harboring iPLA₂ with selected concentrations of BEL for 3 min prior to the addition of substrate resulted in the inhibition of ~70% of iPLA₂ phospholipase A₂ activity at a concentration of 5 µM (Fig. 12). The IC₅₀ for BEL inhibition of iPLA₂ was ~3 µM.

View larger version (20K): [in this window] [in a new window]   Fig. 12.   Inhibition of iPLA2 phospholipase A2 activity by BEL. Membrane fractions from Sf9 cells infected with either wild-type control baculovirus () or recombinant baculovirus encoding human iPLA2 () were incubated at room temperature with the indicated concentrations of BEL or ethanol vehicle for 3 min in 100 mM Tris·HAc, pH 8.0, containing 1 mM EGTA. L--1-palmitoyl-2-[1-14C]oleoyl phosphatidylcholine (40 µM final concentration) in ethanol was then added to each reaction, followed by incubation at 37 °C for 2 min. Released [1-14C]oleic acid was quantitated as described under "Experimental Procedures." Results are representative of three independent experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The present results identify the complete genomic organization (i.e. intron-exon boundaries), complete mRNA sequence (including the 5'- and 3'-untranslated regions), and the phospholipase A₂ catalytic activity of one or more polypeptide products of the iPLA₂ gene encoded on the long arm of chromosome 7. The iPLA₂ gene is 54 kb in length and contains 11 exons ranging in size from 46 to 1151 nucleotides. Interestingly, both the ATP binding motif (amino acid residues 449-454) and the lipase consensus sequence (amino acid residues 481-485) of iPLA₂ require splicing of adjacent exons to form complete functional sequences. This is in contrast to iPLA₂, which contains its functional ATP and lipase sites within a single exon (47).

The promoter region of the gene encoding PLA₂ contains neither a TATA box nor a CCAAT sequence within 400 nucleotides flanking the putative transcription start site. However, Tfsearch analysis (48) reveals other promoter elements (e.g. several Sp1/GC boxes) that are located in this region as well as several potential cap signals (one of which is located immediately upstream of the predicted transcription initiation site (i.e. residues 6 to 13). Prior work has established that TATA-less promoters typically respond to a variety of stimuli with a range of transcriptional regulatory responses including differential expression during embryogenesis, tissue-specific distribution, and regulation by either viral or pharmacologic stimuli (49). Many of these genes are growth-regulated with low levels in nongrowing cells, which are up-regulated as cells proliferate. Although the present results do not identify the specific role of any of these sites in the transcriptional regulation of the iPLA₂ gene, they do provide the foundation for future studies in which the identified promoter region can be fused to a reporter gene and subsequently dissected to determine important transcriptional regulatory elements by deletional mutagenesis.

Translation of the mature message encoding iPLA₂ in either an in vitro reticulocyte lysate translation system or in a baculovirus expression system in Sf9 cells gave rise to two major protein products at 77 and 63 kDa. Polypeptides of this size are most consistent with translation initiation at Met¹⁰¹ and Met²²¹. However, some proteins migrate anomalously on SDS-polyacrylamide gel electrophoresis, and the potential role of proteases in cleaving larger proteins to the observed polypeptides is unknown at present. None of the potential methionine initiator sites were strong matches for the Kozak consensus sequence for initiator methionines (50, 51). We point out that the use of alternative methionine start sites is frequently observed in genes encoding regulatory polypeptides such as cytokines, receptors, protein kinases, and growth factors (52). Whatever the case, membrane preparations from Sf9 cells expressing the iPLA₂ gene clearly possessed robust phospholipase A₂ catalytic activity, thereby unambiguously identifying this gene as one encoding bona fide phospholipase A₂.

The primary sequence of iPLA₂ contains two signature sequence motifs. The ATP binding motif and the serine lipase site are present in all known members of this family of lipases (i.e. iPLA₂, iPLA₂, and iPLA₂) (26, 27, 53). Additionally, a region of nine amino acids in iPLA₂ (residues 627-635) is also highly conserved in iPLA₂, but not in iPLA₂ (26, 27). The functional significance of the 627-635 sequence is unknown. Excluding these three short amino acid motifs, there is no known homology between iPLA₂ and any other known phospholipase. Additionally, there is an SKL sequence present at the C terminus of iPLA₂, which is a known microbody localization sequence (54). Through the elegant studies of Subramani and others (reviewed in Ref. 55), the biochemical mechanisms leading to incorporation of proteins synthesized on cytosolic ribosomes into the peroxisomal compartment has been elucidated. Since iPLA₂ is tightly bound to the membrane fraction in cell homogenates, it is extremely likely that the major subcellular localization of iPLA₂ is in the peroxisomal matrix tightly associated with the peroxisomal membrane. As far as we are aware, there are no known exceptions to proteins having a C-terminal SKL sequence residing predominantly in the peroxisomal compartment (except in genetic diseases in which peroxisome assembly is defective (e.g. Zellweger syndrome (56, 57)).

The present results clearly identify the iPLA₂ gene product(s) as catalysts of cleavage of the sn-2-fatty acid of choline glycerophospholipids. This is most convincingly demonstrated in the case of plasmenylcholine, where esterolytic hydrolysis of the sn-1-aliphatic chain is precluded by the presence of a vinyl ether. However, the precise clarification of the detailed kinetic characteristics and substrate specifications of the iPLA₂ gene products is complicated by multiple issues. First, iPLA₂ gene protein products are tightly membrane-bound (i.e. no immunoreactive material was present in the soluble fraction even after intense sonication). Thus, the precise mechanism through which exogenous radiolabeled phospholipids interact with the membrane-associated protein is unclear. Several possibilities exist. For example, if the radiolabeled substrates must first insert into the plane of the membrane prior to interaction with iPLA₂, then at least two (and likely more) kinetic steps are relevant. If insertion into the membrane is rate-limiting, then the observed substrate selectivities reflect the rate of insertion into the plane of the membrane bilayer and do not necessarily reflect the intrinsic interaction energies of substrate with enzyme. Accordingly, before detailed kinetic analysis is undertaken, it is necessary to remove iPLA₂ from its membrane-associated state. Thus far, all attempts at detergent solubilization and salt extraction with retention of activity have failed, thereby making definitive kinetic characterization and identification of substrate selectivities impossible at present. Nonetheless, several important characteristics of the enzyme can be identified. First, iPLA₂ is a calcium-independent enzyme, since calcium is not an obligatory cofactor in catalysis. Second, the pH optimum for hydrolysis appears to be at or near physiologic pH. Third, BEL is an effective inhibitor of iPLA₂. We point out that the different polypeptide products produced by this gene may have significantly different substrate selectivities, pH profiles, and sensitivities to BEL. Accordingly, it is necessary not only to solubilize the enzyme from its membrane environment but also to purify each of the different isoforms prior to definitive kinetic analysis.

In prior studies, we identified iPLA₂ activity in the cytosolic and membrane fractions of canine and human myocardium (15, 58). Moreover, we and others have demonstrated an increase in membrane-associated, BEL-inhibitable iPLA₂ activity during myocardial ischemia or hypoxia (59-62). We have previously proposed that increased myocardial iPLA₂ activity during ischemia contributes to ventricular arrhythmias and hemodynamic dysfunction through production of lysophospholipids and arachidonic acid, each of which has potent electrophysiologic effects (15, 63, 64). Recently, another cPLA₂ that shares homology with the catalytic domain of cPLA₂ has been cloned and expressed (28). Northern analysis demonstrated the presence of cPLA₂ message in skeletal and heart muscle (28). We have cloned and expressed cPLA₂ in Sf9 cells and have confirmed the membrane localization and calcium independence of cPLA₂ as illuminated by Underwood et al. (28). In recently completed studies, we have determined that cPLA₂ is not inhibited by BEL at <50 µM [BEL].² The large majority of membrane-associated iPLA₂ activity from ischemic hearts or hypoxic myocytes in culture is membrane-associated and exquisitely sensitive to inhibition by BEL. Accordingly, the results presented herein identify iPLA₂ as a candidate for the polypeptide catalyzing the increase in ischemia-induced calcium-independent phospholipase A₂ activity. Of course, the possibility that other, as yet unidentified calcium-independent phospholipase A₂ activities are involved cannot be ruled out. Based on our previous studies in ischemic myocardium, it is tempting to speculate that the peroxisomal compartment is a source of regulatory and potentially arrhythmogenic eicosanoid metabolites and lysophospholipids.

In summary, the protein products of the iPLA₂ gene identified herein are potential candidates for the iPLA₂ activity that increases during myocardial ischemia and may contribute to myocardial dysfunction during the ischemic process. Experiments in our laboratory are currently in progress pursuing this intriguing possibility.

	FOOTNOTES

^* This work was supported jointly by Juvenile Diabetes Foundation International Grant 996003 and National Institutes of Health Grants 1 PO1 HL 57278-02, 2 R02 HL 41250-08A1, and 5R01 AA11094.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Washington University School of Medicine, Division of Bioorganic Chemistry and Molecular Pharmacology, 660 S. Euclid Ave., Campus Box 8020, St. Louis, Missouri 63110. Tel.: 314-362-2690; Fax: 314-362-1402.

² C. M. Jenkins, D. J. Mancuso, and R. W. Gross, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: PLA₂, phospholipase(s) A₂; iPLA₂, calcium-independent PLA₂; cPLA₂, calcium-dependent PLA₂; plasmenylcholine, 1-O-(Z)-hexadec-1'-enyl-2-[9,10-³H]octadec-9'-enoyl-sn-glycero-3-phosphocholine; BEL, (E)-6-(bromomethylene)-3-(1-naphthalenyl)-2H-tetrahydropyran-2-one; PCR, polymerase chain reaction; RACE, rapid amplification of cDNA ends; CAPS, 3-(cyclohexylamino)propanesulfonic acid; EST, expressed sequence tag; nt, nucleotide.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES