Identity between the Ca2+-independent phospholipase A2 enzymes from P388D1 macrophages and Chinese hamster ovary cells.

A novel Ca2+-independent phospholipase A2 (iPLA2) has recently been purified and characterized from P388D1 macrophages (Ackermann, E. J., Kempner, E. S., and Dennis, E. A. (1994) J. Biol. Chem. 269, 9227-9233). This enzyme appears to play a key role in regulating basal phospholipid remodeling reactions. Also an iPLA2 from Chinese hamster ovary (CHO) cells has been purified, molecularly cloned, and expressed (Tang, J., Kriz, R., Wolfman, N., Shaffer, M., Seehra, J., and Jones, S. S. (1997) J. Biol. Chem. 272, 8567-8575). We report herein that the cloned CHO iPLA2 is equivalent to the mouse enzyme purified from P388D1 cells. Polymerase chain reaction amplification of cDNA fragments from P388D1 cells using primers based on the CHO iPLA2 sequence, revealed a high degree of homology between the mouse and hamster enzymes at both the nucleotide and amino acid levels (92 and 95%, respectively). Identity between the two proteins was further demonstrated by using immunochemical, pharmacological, and biochemical approaches. Thus, an antiserum generated against the CHO enzyme recognized the P388D1 cell enzyme and gave similar molecular masses (about 83 kDa) for the two enzymes under the same experimental conditions. Further, the CHO enzyme has exactly the same sensitivity to inhibition by a variety of compounds previously shown to inhibit the P388D1 enzyme, including bromoenol lactone, palmitoyl trifluoromethyl ketone, and methyl arachidonyl fluorophosphonate. Additionally, covalent modification of the CHO enzyme by [3H]bromoenol lactone is dependent on active enzyme as is the P388D1 iPLA2. Finally, both enzymes have the same specific activities under identical experimental conditions.

Phospholipase A 2 (PLA 2 ) 1 comprises a superfamily of enzymes that regulate phospholipid metabolism and generate bioactive lipid mediators such as the eicosanoids and plateletactivating factor (for review see Ref. 1). Since the PLA 2 s have been implicated in a number of tissue dysfunctions ranging from inflammation to ischemia, much attention has been devoted to the study of the mechanism of action and biochemical characteristics of these enzymes. Depending on their site of action, the PLA 2 s can be subdivided into two types: the extracellular, secreted enzymes and the intracellular, cytoplasmic enzymes (1). Among the latter, two groups of enzymes can be considered, namely the group IV, Ca 2ϩ -dependent cytosolic PLA 2 , and the Ca 2ϩ -independent PLA 2 (iPLA 2 ).
In contrast to the group I-IV PLA 2 s, the Ca 2ϩ -independent PLA 2 s (iPLA 2 ) have been poorly studied, since most of them are labile, constitute only a minor fraction of the total cellular protein, and have lower specific activities. The iPLA 2 s have been grouped into three main categories based on their biochemical and localization characteristics: lysosomal, brush-border membranes, and intracellular (for review, see Ref. 2). The lysosomal and brush-border membrane iPLA 2 s appear to be conserved among the species where they have been identified. However, the intracellular iPLA 2 enzymes appear to represent a much more diverse and broad group of enzymes, whose relationship is not so evident. Only four intracellular iPLA 2 s have been purified to homogeneity, namely a 40-kDa enzyme from myocardium (3), a 39-kDa enzyme from bovine brain (4), an 80-kDa enzyme from P388D 1 macrophages (5), and very recently a 28-kDa enzyme from rabbit kidney (6). The enzymes from myocardium and P388D 1 cells are both modulated by ATP and form oligomeric complexes of about 400 kDa. However, besides their very distinct molecular mass, they also differ significantly in substrate preference and detergent sensitivity (3,5).
The first molecular cloning of an iPLA 2 from CHO cells is reported in the accompanying manuscript (7). Due to the role of iPLA 2 in P388D 1 cell metabolism (8), it was essential to determine whether or not the enzyme present in these cells was the same as that cloned by Tang et al. (7). Using a variety of techniques, we demonstrate herein that the enzyme from P388D 1 cells and the cloned enzyme from CHO cells is the same molecular entity expressed in different species. Thus, the availability of the macrophage iPLA 2 sequence should ensure rapid progress in understanding its physiological function. iPLA 2 Assay-The iPLA 2 activity has been previously described (5). Briefly, 10 -50 ng of purified iPLA 2 was assayed in a buffer consisting of 100 mM Hepes, 400 M Triton X-100, 100 M 1,2-dipalmitoyl-sn-glycero-3-phosphocholine, 1-palmitoyl-2-[ 14 C]palmitoyl-sn-glycero-3-phosphocholine (200,000 cpm), 5 mM EDTA, and 0.1 mM ATP (pH 7.5). The * This work was supported by National Institutes of Health Grants HD 26,171 and GM 20,501. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) U88624.
[ 3   lated using the guanidinium isothiocyanate/phenol chloroform extraction method (Stratagene). Polyadenylated mRNA was purified by affinity chromatography using an oligo(dT)-cellulose column (Stratagene). The mRNA was converted to cDNA using the Stratagene reverse transcriptase-PCR kit. Amplification of specific cDNA fragments was achieved by using recombinant Pfu polymerase (Stratagene). Primers were based on the CHO iPLA 2 sequence (Ref. 7; GenBank TM accession number I15470). The 5Ј-primer used (AGG ATG CAG TTC TTC GGA CGC C) included the translation initiation codon ATG. The 3Ј-primer was a reverse primer (CAG TTG ATG GAG CCA GTT GTC C) situated 223 bases after the termination codon TGA. To determine the sequence of the 5Ј-end (primer region), we performed a standard rapid amplification of cDNA ends-PCR (9). Briefly, cDNA was amplified by PCR using the mouse reverse primer TTC CTA GGA GCT GTA GCA CCT G, which is equivalent to nucleotides 598 -619 in the hamster sequence. After attaching a poly(A) tail to the 5Ј-end of the PCR product obtained, amplification of this product was accomplished by PCR using a dT(18) primer and a mouse reverse primer, ACT CCA GTT GGA AGA GCC GGA A, which is equivalent to nucleotides 205-226 in the hamster sequence. A final round of amplification was performed using the mouse reverse primer TTC CTC CCG GAC ACG TTC ACT T (equivalent to nucleotides 101-123 in the hamster sequence). PCR conditions were as follows: denaturing, 95°C for 30 s; annealing, 55°C for 45 s; extension, 72°C for 120 s; 40 cycles. A last extension step lasted 10 min. PCR products were purified by agarose gel electrophoresis and sequenced by automatic DNA cycle sequencing (Applied Biosystems 373 automated DNA sequencer, Perkin-Elmer).

RESULTS AND DISCUSSION
The intracellular iPLA 2 from P388D 1 cells was purified and characterized in our laboratory as an approximately 80-kDa protein (5). The iPLA 2 purified by Tang et al. (7) from CHO cells and the protein expressed by the cDNA have a molecular size of 84.5 kDa. Besides this similarity, the purification strategy used by Tang et al. (7) contains some analogous steps to those used for the macrophage iPLA 2 (5), and the enzymatic profile of the CHO enzyme is comparable. Therefore, we hypothesized that both proteins would share a high degree of similarity if not identity. We began to address this hypothesis by taking advantage of the polyclonal antibody generated by Tang et al. (7) against the C-terminal portion of the CHO protein. The antiserum recognized pure iPLA 2 from P388D 1 cells, as shown by immunoblot (Fig. 1, lane 2). Purified CHO iPLA 2 was run in parallel as a control and showed identical mobility, focusing both proteins in the 80 -85-kDa molecular mass range (Fig. 1,  lane 1).
To further establish a similarity between the two proteins, a series of inhibition studies were conducted. We have previously shown that the iPLA 2 from P388D 1 cells is potently and irreversibly inhibited by bromoenol lactone (BEL). After a preincubation time of 5 min at 40°C, BEL inhibits pure iPLA 2 with an IC 50 of about 60 nM (10). Likewise, iPLA 2 from CHO cells was inhibited in a dose-dependent manner by BEL with an IC 50 of 120 nM ( Fig. 2A). This inhibition was irreversible even after exhaustive dialysis of treated enzyme compared with control (Fig. 2B). Also, P388D 1 iPLA 2 is reversibly inhibited by PACOCF 3 (IC 50 ϭ 4 M), and again CHO iPLA 2 was inhibited by the same reagent with an IC 50 of 3 M (Fig. 3).
Since BEL is an irreversible inhibitor, it was possible to label the macrophage iPLA 2 by incubation with [ 3 H]BEL (10). However, labeling with BEL required the presence of active enzyme, since the inhibitory agent is generated in situ from BEL by the hydrolytic action of the enzyme on the lactone ring (10). Indeed, in previous studies with the P388D 1 iPLA 2 , DTNB prevented labeling of the enzyme with [ 3 H]BEL (10). Although DTNB also inhibits the CHO iPLA 2 (Fig. 4), we have employed herein the irreversible inhibitor MAFP (11), because it is a much more specific reagent than DTNB. MAFP (20 M) completely inactivated the CHO iPLA 2 (Fig. 4) and prevented labeling of the enzyme with [ 3 H]BEL (Fig. 5). Moreover, in the absence of MAFP treatment, autoradiographic analysis of [ 3 H]BEL-treated enzyme revealed a single spot at about 85 kDa. This is the same molecular size as that obtained for the P388D 1 iPLA 2 utilizing a similar strategy (10) as well as that found by immunoblot (Fig. 1).
The nucleotide sequence of the P388D 1 iPLA 2 was obtained by analyzing mRNA from these cells by PCR using primers based on the CHO iPLA 2 sequence (Fig. 6). Comparison of the nucleotide sequence of the two proteins revealed a 92% sequence homology. Most of the changes of the nucleotide sequence occurred at the third nucleotide of the codon, reflecting the 95% identity at the amino acid level between two closely related species. These results further support the notion that the hamster iPLA 2 cloned by Tang et al. (7) is the species equivalent to the murine enzyme purified and characterized by us.
Our previous studies on the P388D 1 cell iPLA 2 have suggested that this enzyme may play an important role in regulating fatty acid remodeling reactions in the cells (8,12), and a similar role has subsequently been attributed to the iPLA 2 activity present in rat pancreatic islets (13). This implies that the iPLA 2 enzyme might ultimately regulate several key as-FIG. 6-continued pects of cell physiology, such as new membrane synthesis that allows cell proliferation, or fatty acid exchange within phospholipids that allows adaptive homeostatic changes. Although the discovery by Gross and co-workers (14) that some iPLA 2 s are potently and selectively inhibited by BEL has accelerated research on this enzyme, study of the iPLA 2 is hampered by the fact that it is extremely difficult to obtain sufficient amounts of pure protein for biochemical and sequence analysis.
We have presented evidence herein that the enzyme purified and cloned by Tang et al. (7) is the equivalent in hamster to the enzyme we identified (15), purified (5), characterized (5,10,11), and studied at the cellular level (8,12) in mouse P388D 1 macrophages. By using a wide variety of approaches, we have found biochemical, immunological, pharmacological, and genetic similarities between the two proteins, strongly suggesting that the same molecular entity is expressed in different species. There is, however, an apparent difference between the two enzymes concerning ATP sensitivity. Tang et al. (7) failed to observe any ATP effect on a partially purified preparation of CHO iPLA 2 . We have confirmed this finding using the assay system employed for the P388D 1 iPLA 2 . We have previously reported on the apparent activation of P388D 1 iPLA 2 by ATP (5). The effect was shown to depend on the presence of Triton X-100 in the assay system. This, along with the fact that several other nucleotides manifested the same effect (i.e. ADP, UTP, GTP) raised the possibility that it might not be relevant as a regulatory mechanism. We have new data with P388D 1 iPLA 2 showing that, rather than stimulating the iPLA 2 , the ATP stabilizes the enzyme and protects it from denaturation; hence, higher activity is found in the presence than in the absence of ATP. 2 Considering the fact that the two proteins have slightly different amino acid sequences and purification schemes, it is possible that the CHO enzyme lacks the region with which ATP interacts. Alternatively, it is possible that the CHO enzyme is more resistant to denaturation than the P388D 1 enzyme.
The availability of the iPLA 2 cDNA and protein sequence of the 80-kDa iPLA 2 , which is now classified as a Group VI PLA 2 (16), will allow new experimental avenues to be explored in defining the role(s) of iPLA 2 in phospholipid metabolism and cellular function.