Cloning and recombinant expression of a novel human low molecular ...

32 downloads 0 Views 2MB Size Report
Roman numerals refer to coding exons; the only noncoding exon is unnumbered. of a Novel ..... Tam, S. W., Man, R. Y., and Choy, P. C. (1984) Can. J. Biochem.
T H E JOURNALOF Btotoctcu. CHEMISTRY Vol. 269, No.4, Issue of January 28, pp. 2365-2368, 1994 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Communication

Printed in U . S A .

fatty acid and lysophospholipids (1).Over the pasttwo decades, PLAz activities havebeen purifiedand characterizedfrom various tissues, cultured cells, and exudates from humans and different mammals (2-5). These enzymes varied in molecular weight and solubility, and their activities exhibited differing pH optima, Ca2+dependence, and substrate specificity. These observationssuggest that several different PLAzs existin mammals. Low molecular mass (14 kDa) PLA2s, which are character(Received forpublication, September 10, 1993, and in revised ized by a catalytic requirement for Ca2+ and a rigid threeform, November 10, 1993) dimensional structure maintained by disulfide bridges, rank Ju Chent, Sandra J. Englet, among the best characterized enzymes( 6 ) . Complete primary Jeffrey J. SeilhamerO, and Jay A. TischiieldSn sequences have been determined for more than 50 ( 6 , 7). Based From the $Department of Medical and Molecular on selected structural determinants, these PLAzs have been Genetics, Indiana University School of Medicine, classified into two groups (8).Group I enzymes have a disulfide Indianapolis, Indiana 46202 and 5Incyte bridge connecting cysteines a t positions 11 and 77 and a soPharmaceuticals Inc., Palo Alto, California94304 called "elapid loop." The only identified mammalian group I Extensive biochemical studies of phospholipase A2s PLA2(9,lO) is expressed most abundantly in the pancreas and (PLA,s) overthe lasttwo decades indicate that there are functions extracellularly in dietary lipid catabolism. Group I1 likely to be several distinct PLAz genes in mammals. PLA2s lack the 11-77 disulfide bridge, have a carboxyl-termiHere we report the cloning of a 1-kilobase pair cDNA nal extensionof 6 amino acids, and terminate in a half-cysteine P U S . joined to Cys-50 near the catalytic site His-48. Mammalian encoding a novel human low molecular weight The cDNA appears to encode a 118-amino acid mature group I1 PLA2s occur in trace amounts in tissues such liver, as peptide (M,= 13,692) preceded by a 20-residueprepep spleen, platelets, andsynovial fluid and are secretedfrom varitide. The deduced amino acid sequence encodes a proto appropriate ous cells of the immune system in response tein that lacks one of the seven disulfide bridges found stimuli (11-16). Group I1 PLAzs may be associated with the in similar P U a s and, therefore, represents a class of I and pathologies of several diseasesinvolving sepsis, tissue damage, enzymesdistinctfromthemammaliangroup and inflammation (17-20). Mammalian group I and I1 PLA2s group I1 enzymes. An RNA blothybridizedwiththe in share approximately 3040% amino acid homology; however, cDNAexhibited a putative 1.2-kilobase pair transcript heart and, less abundantly, in lung, as well as multiple 18 amino acids are invariantlyconserved in all functional low putative transcripts in placenta. When the cDNAwas molecular weight PLA2s (7). Johnson et al. (21) found a novel 120-bp putative PLA2 gene expressed using an Epstein-Barr virus-based vector in human 293s cells, PIAz activity accumulated in the cul-fragment (hlOa) homologous to the amino terminus encoding exon of low molecular weight PLA2s by screening a human turemedium.Conditionedmediumoptimallyhydrolyzed the phospholipids of [l-14C]oleate-labeledEsch- genomic DNA library with a 45-bp human PLAz group I1 olierichiu coli at neutral to alkaline pH with 10 m~ or gonucleotide probe. Hybridization of the 120-bp fragment to greater CaZ+. In assays done with individual substrates,DNA from other species suggested that the fragment hasbeen ~-a-l-palmitoyl-2-oleoyl phosphatidylcholine was more highly conserved during evolution. However, additional exons, efficiently hydrolyzed than ~-a-l-palmitoyl-2-arachidoa corresponding cDNA, or evidence of transcription in human nyl phosphatidylcholine, ~-a-l-palmitoyl-2-arachidonyl tissues was notfound. Here we report the cloning, expression, phosphatidylethanolamine, or ~-a-l-stearoyl-2-arachiand preliminary characterization of a novel Ca2+-dependent donyl phosphatidylinositol. mammalian PLA2 containing the 120-bp genomic fragment,

Cloning and Recombinant Expression of a Novel Human Low Molecular Weight Ca2+-dependent Phospholipase A2*

EXPERIMENTALPROCEDURES RACE-RT-PCR--Total RNA was prepared according to the method of Phospholipase A2s (PLA2)' (phosphatide 2-acylhydrolase, EC 3.1.1.4) constitute a family of enzymes that hydrolyze the sn-2 Chomczynski and Sacchi (22).5' and 3' RACE-RT-PCR (rapid amplififatty acid acyl ester bond of phosphoglycerides, producing free cation of cDNA ends-reverse transcription-polymerase chain reaction) techniques were used to generate the full-lengthcDNA from total human brain RNA as described by Ishizaki (23).PCR amplifications were * This work was supported by the Children's Brain Diseases Foun- done using 30 cycles at 95 "C for 20 s,60 "C for 20 8 , and 72 "C for 75 s dation. The costs of publication of this article were defrayed in part by in 100 pl of buffer containing 1.5 m~ MgC12, 200 p~ dNTPs, 100 rn the payment of page charges. This article must therefore behereby Tris-HC1, pH8.3,and 3 units of Taq polymerase. Anchor(GGCCACGCmarked "aduertisement" in accordancewith 18 U.S.C.Section 1734 GTCGACTAGTAC(T)l,, 300 ng) and adaptor (GGCCACGCGTCGACsolely t o indicate this fact. ll To whom correspondence should beaddressed:Dept. of Medical and TAGTAC, 50 ng) primers were used in both 5' and 3' RACE-RT-PCR. Molecular Genetics,Indiana UniversitySchool of Medicine, IB130,975 Primers H10-C CGCAGAGGATG'R'GGGAAAGTAT, 300 ng) andH10-la W. Walnut St., Indianapolis, IN 46202-5251.Tel.: 317-274-5738; Fax: (ATCGGTGCCATCClTGGGGG'R',300 ng) wereused for 5' RACE-RTPCR. Primers H10-A (CTGGCTI'GG'R'CCTGGCTGTA, 300 ng) and 317-274-1069. H10-1 (GCAAGGAGGC'R'GCTGGACCTA, 300 ng)wereusedfor 3' Theabbreviationsusedare:PLA,,phospholipase A*; bp,base paids); kb, kilobase paids); RACE, rapid amplification of cDNA ends; RACE-RT-PCR. Library Screening-6 x lo6bacteriophage clonesfrom a human stomRT, reverse transcription; PCR, polymerase chain reaction; PC-oleoyl, 5 x lo5 clones froma ratgenomic ~-a-l-palmitoyl-2-[~~Cloleoyl phosphatidylcholine; PC-arachidonyl,~ - a - ach cDNA library (Clontech, Inc.) and l-palmitoyl-2-[14Clarachidonylphosphatidylcholine; PE-arachidonyl, DNA cosmid library (Clontech, Inc.) were screened according the sup~-a-l-palmitoyl-2-[~~Clarachidonyl phosphatidylethanolamine; PI-ara- plier's procedures. chidonyl, ~-a-l-stearoy1-2-['~C]arachidonyl phosphatidylinositol. Expression ofHPLA2-10cDNA in 293s Cells-An Epstein-Barr virus-

2365

2366

Cloning and Expression of a Novel Human PLA,

based vector (pCEP)was used to express HPLA2-10 in human 293s cells (ATCC CRL1573). pCEPcontains two regionsof the Epstein-Barr virus genome required for episomalmaintenance (EBNA-1 and OriP), a drug resistance gene for selection in human cells (hyg), bacterial sequences for maintenance in Escherichia coli, a drug resistance gene for selection in E. coli (amp), and an expression cassette for the production of high levels of mRNA from an introduced sequence. The expression cassette utilizes an Rous sarcoma virus LTR promoter and an SV40 polyadenylation signaL2 PLA,ActiuityAssay-Autoclaved [l-14Cloleate-labeledE. coli JM109 was prepared according to the method of Elsbach (24) for use as a PLAz substrate. Assay times were adjusted to ensure that rates were within the linear range. Determination of p H Optimum and Ca2+Dependence-The pH optimum and calcium dependence weredetermined using 20 pl of medium (24). Sodium acetate buffer (25 m)was used for the pH range 4-6.5 and TridHCl buffer (25 m)for the pH range 7-9. Dependence of the PIAz activity on calcium was examined in the 0-400 IMI range. The buffer solution (TrieMCl, 25 m, pH 7.5) was prepared with highly purified water, and EDTA (300 w) was used to titrate residual calcium from the cell culture medium and E. coli extract. Sufficient EDTA was added such that no PLAzactivity was detected until exogenous calcium was added. Substrate Specificity Assay-Substrate specificity was determined using the method described by Seilhamer et al. (25) exceptthat 20 pl of medium wasused instead of purified protein. The dried thin-layer chromatography plate was visualized and substrateconversion quantitated phosphatidylusing an AMBIS 8-scanner. ~-(1-l-Palmitoyl-2-[~~Cl0le0yl phosphatidylcholine(PC-oleoyl), ~-a-l-palmitoyl-2-[~~Clarachidonyl phospha- ogy to the amino termini of mature low molecularweight choline (PC-arachidonyl), ~-cr-l-palmitoyl-2-[~~Clarachidonyl tidylethanolamine (PE-arachidonyl), and ~-a-l-stearoyl-2-[~~ClPLAZS. arachidonyl phosphatidylinositol (PI-arachidonyl) were obtained from H10-A, a primer located in putative exon I, and H10-1, a DuPont NEN.

primer located in putative exon 11, were then used for RT-PCR of total human brain and lymphoblast RNA. A unique 140-bp RESULTS AND DISCUSSION fragment common to both reactions was sequenced. As exCloning of HPLA2-1hIn order to clone a human cDNA con- pected, it contains only putative exons I and I1 and lacks any taining the hlOa sequence, two primers, H10-1 and H10-la, intervening sequences, indicating that RNA, not genomic DNA, were generated from the hlOa sequence (data not shown). PCR was amplified. This suggested that the putative exons were amplification with these primers was used to screen human part of an expressed gene and were not a cloning artifact or child brain, adult brain, liver, heart, and various white cell pseudogene. 5' and 3' RACE-RT-PCR techniques (23) were cDNA libraries. PCR amplification products were not obtained. used to generate the full-length cDNA sequence fromtotal high Since the screening of human cDNA libraries was unproduc- quality human brain RNA that was fortuitously available at tive and the hybridization of hlOa to multiple species DNA the time. The entire cDNA sequence, designated HPLA2-10, is blots had indicated that the sequence of hlOa was evolution- shown in Fig. 1.Exon-intronjunction sites were determined by arily conserved (211, a strategy based upon the identification of comparison to genomic DNA sequences.HPLA2-10 has an homologous PLAz-like sequences in human and rat was pur- exon-intron structure typical of group I1 PLAzs, with one 5'sued. A rat genomic DNA cosmid library (Clontech, Inc.) was untranslated exon followed by four protein coding exons. screened using a PCR-generated copy of hlOa as a probe. HyA human stomach cDNA library (Clontech, Inc.) and the bridization of three positive EcoRI-digested clones with hlOa human cDNA libraries described above were PCR-amplified identified a common 4.4-kb fragment. This fragment was sub- with primers H10-A and H10-C to detect sequences homologous cloned into EcoRI-digested pUC13and sequenced completely.A to HPLA2-10. A 400-bp and a 290-bp product were obtained region highly homologous to hlOa was identified in the 4.4-kb only from the stomach cDNA library. Each of the two PCR rat sequence. fragments werecloned into pUC19 and sequenced. The seTo search for additional regions of homology surrounding the quence of the 400-bp fragment is identical to the HPLA2-10 hlOa sequence, a 5-kb human genomic DNA clone containing RACE-RT-PCR-generated cDNA sequence from bases 148 to hlOa (data not shown) was sequenced completely. Computer 541. The 290-bp fragment is identical to the 400-bp fragment comparison of the human sequence to the above noted rat se- except that itis missing bases 316422, which encompass the quence identified two highly homologous regions. One homolo- 5' end of exon I11 (see Fig. 1).The same two PCR fragments gous regionappears to be exon113since it contains a consensus were also amplified from total human brain and lymphocyte splice acceptor site at its 5' end and a consensus splice donor RNA using primers H10-A and H10-C. The 290-bp PCRproduct site at its 3' end. The other region, located about 1.2-kb 5' of is much less abundant than the 400-bp product in amplificathis putative exon 11, contains a perfect consensus splice donor tions from human stomach and brain RNA. Since the 290-bp site at its 3'end and a potential in-frame ATG start codon at its product encodes onlythe signal peptide and the first 41 amino 5' end. It is likely to be exon I. Furthermore, when these two acids of the mature protein because of an in-frame stop codon putative exons are joined together using the assumed splice immediately following the 41st amino acid, any in vivo signifidonor and acceptor sites, the resulting sequence encodes a sig- cance of this transcript is unknown. nal peptide and 41 amino acids, which have significant homolIn order to obtain a full-length cDNA, the 400-bp PCR product was used to screen 6 x lo5 individual plaques from the human stomach cDNA library. Four positive cloneswith M e r J. J. Seilhamer, unpublished data. Roman numerals refer to coding exons;the only noncoding exonis ent size inserts were identified. The inserts of these clones were released by EcoRI digestion, subcloned into pUC19, and seunnumbered.

Cloning and Expression of a Novel Human PLA, 1.8 T

2367

-

~~

-293s

PCHlOlB

-

PCHlOlD

I

i

PCHlOlF

A

4.4 kb 2.4 kb 9.5 kb 7.5 kb

1.35 kb-#

--t

PCHlO2G

--b

RASFZB

--C

0.4 0.2

RASFlC

RASFlD

-x-

RASFZD

0

FIG.3. PLA, activity in growth medium of pCH10- and pCRASF-transfected cell lines. 293s (ATCC CRL 1573) cells were grown in Dulbecco’s modified Eagle’s medium (Life Technologies, Inc.) supplemented with 10% fetal bovine serum. Approximately 7.5 x lo5 cells were transfected with 10 pgof purified supercoiled plasmid DNA accordingto the method of Kingston(32). Twenty-four hours after transfection, 200 unitdml hygromycin B (Calbiochem) was addedto the medium. Stably transfected, hygromycin B-resistant colonies were selected 10 days after transfection and maintainedDulbecco’s in modified Eagle’s medium containing200 unitdml hygromycin B. To test for PIA, activity, 2.0 x lo6 cells were transferred to a 25-cm2 culture flask and medium was collected a t 24-, 48-, and 72-h intervals. PIA, activity was 1 2 3 4 5 6 7 0 assayed as describedunder“ExperimentalProcedures.”pCHlO-IB, FIG.2. RNAblot hybridization. A, a human multiple Northern blot (Clontech, Inc.) was hybridized accordingto the manufacturer’s direc- pCH10-lD, pCH10-lF, and pCH10-2G are hygromycin-resistant cell tions with “P-labeled, a 400-bp PCR fragment generated using primerslines stably transfected with pCH10. RASF-2B, RASF-IC, RASF-lD, H10-A and HlO-C. Final wash conditions were 0.1 x SSC at 62“C for 1 and RASF-2D are hygromycin-resistant cell linestransfectedwith h, and the filter wasexposed for 72 h.B , the same blot was hybridized pCRASF. Substrate conversion is calculated according to the method of with a 450-bp fragment of human typeI1 PIA2 cDNA corresponding to Elsbach (25). bases 150-581 (11). Final wash and exposure conditions were a s described above. C,hybridization of the same blot with a manufacturer ern blot (Clontech, Inc.) was probed with a 400-bp fragment supplied human p-actin probe produced bands of expected size and generated by RT-PCR ofhuman brainRNA with primers H10-A intensity.

and H10-C. As seen in Fig. 2 A , an abundant 1.2-kb transcript quenced completely. The sequences of the clones are identical to was detected in heart and, less abundantly, in lung RNA. In the correspondingregions of the RACE-RT-PCR generated addition, multiple transcripts were detected in placental RNA. This suggests thathigh levels of expression of H P W - 1 0 are HPLA2-10 sequence (data not shown). Predicted Protein Structure-The HPLA2-10 cDNA predicts relatively tissue specific and that placental transcription of a mature protein of 118 amino acids with a calculated molecu- HPLA2-10 may initiate or terminate at a distinct site(s). The lar mass of 13,592 daltons. The amino acid sequence has sig- apparent overabundance of HPLA2-10 mRNA in heart, given nificant homology to known PLA2s, showing 30.5% homology the relatively low levels of calcium-dependent PLA, activity with human group I PLA, and 40.6% with human group I1 previously observed in heart tissue(261, may result from activPLA,. All of the 18 invariantly conserved amino acids in low ity-limiting post-transcriptional regulation,rapid protein turnmolecular weightPLAzs ( 7 )are conserved in thisnovel protein. over, or inherent low activity of the enzyme. The lack of any RNA, despite its origiHomology alignment of the 18 conserved amino acids indicates detectable HPLA2-10 transcript in brain that the first 20 amino acids of HPLA2-10 are likely to be a nal isolation from that tissue, is not unusual since Northern blots are incapable of detecting rare transcripts that can fresignalpeptide similar to the ones observed in mammalian group I and group I1 PLA,s. HPLA2-10, therefore, is clearly quently be amplified by PCR. The same blot was also hybridized with a 450-bp fragment of human group I1 PLA2 cDNA related to the two known groups oflow molecularweight PLA2s.It is likely that HPLA2-10 and group I1 PLA,s diverged corresponding to bases 150-581 (11).A 1-kb transcript was more recently from each other than group I PLA2s given the detected in placenta RNA and, less abundantly, in liver and higher amino acid homology and the similar arrangement of skeletal muscle RNA (Fig. 2B 1. This indicates that the pattern of transcription of H P W - 1 0 is distinct from that of group I1 exons and introns. PLA2, further underscoring the novel nature of the gene. HuHPLA2-10, however, contains neither the disulfide bridge between cysteines 11and 77 or the elapidloop characteristic of man p-actin controls (Fig. 2C) demonstrate thateach lane congroup I PLA,s, nor does i t contain the 6-amino acid COOH- tains ample RNA. PLA, activities have been documented from heart, lung, and terminal extension characteristic of group I1 PLA2s. Unlike mammalian groupI and I1 PLA2s, which have 14 cysteine placenta of mammals such as dog, guinea pig, hamster, rat, residues, this putative PLA, only has 12 and, therefore, lacks rabbit, and humans (4, 5 , 27-30). These activities vary with one of the seven disulfidebonds. The predicted PIof this protein regard tophysical properties such as cellular location and prois 6.2, while that of human group I1 PLA2is 10.5. This suggests tein size and also with regard to enzymatic properties such as that HPLA2-10 may represent a new type of PIAz with func- pH optimum, Ca2+dependence, and inhibitor sensitivity, even species. Since no amino tions distinct from those of groups I and 11. Furthermore, the when isolatedfrom the same tissue and rat homolog of HPLA2-10 is functional and retains the struc- acid or nucleotide sequence is known in the majority of cases, one cannotdetermine if the reportedactivities represent tural features of H P ~ 4 2 - 1 0 . ~ Danscription PatternofHPLA2-10-To determine the tissue unique PLA,s or if the observed differences resulted from varyandor assay conditions. Consetranscription pattern of HPLA2-10, a human multiple North- ing purificationmethods quently, we cannot determine if the predicted HPLA2-10 proJ. Chen, S. J. Engle, J. J. Seilhamer, and J. A. Tischfield, manuscript tein is responsible for any PLA2 activity previously described. Expression of HPLA2-lO-To test whether or not this gene in preparation.

2368

Cloning and Expression of a Novel Human PLAD

HPLA2-10 activity toward PC-oleoyl, PC-arachidonyl, PEarachidonyl, and PI-arachidonyl was assayed (see “Experimental Procedures”). As seen in Fig. 4, PC-oleoyl is hydrolyzed slightly more efficiently than the other substrates. Using the same assay, the control group I1 PLAz exhibited a 4-fold preference for PE-arachidonyl over either phosphatidylcholine-containing substrate or PI-arachidonyl (data not shown, Ref. 25). Supernatant from untransfected 293s cells demonstrated no hydrolytic activity toward any of the substrates. In summary, a cDNA clone encoding a novel, human, low molecular weight, Ca2+-dependentPLAz and its genomic coun0 2 6 7 10 16 terpart were obtained. It is likely to be a representative of a YM FIG.4. Substrate specificity of HPLA2-10. The amount (nmol) of distinct group of PLA,s. Possible roles for HPLA2-10, given its substrate hydrolyzed per milliliter of pCH10-1D supernatant was cal- apparent relative tissue specificity and substrate specificity, culated by multiplying the amount of substrate at time zero (0.22 nmol) might include the production of lung surfactant, theremodelby the fraction of hydrolyzed substrate at a given time and the dilution ing of cardiac muscle membrane, or the regulation of cardiac factor. muscle through eicosanoid production. encodes a n active, secreted PIA,, a n Epstein-Barr virus-based vector (pCEP) was used to express the HPLA2-10 cDNA in human 293s cells. HPLA2-10-5’, a 5’ primer beginning at base 126 of HPLA2-10 and introducing an NheI linker, and HPLA210-3’, a 3’ primer endingat base 555 and introducing an XhoI linker, were used for RT-PCRof total human brain RNA to generate the appropriate cDNA insert. The PCR product was blunt-end-ligated to HincII-digested pUC 19 and sequenced. The insertwas then released by digestion with NheI andXhoI and cloned into the NheI-XhoI sites of pCEP. The resulting plasmid was designated pCH10. Human groupI1 PIAz cDNA was cloned into pCEPfor use as a positive control. PCR primers RASF-5’ and RASF-3’ were used to amplify pRASF, a plasmid containing the entire human PLA2 group I1 cDNA, from base 130 to581 (11).The resulting plasmid was designated pCRASF. Medium samples from multiple cell lines transfected with either pCHlO or pCRASF were assayed for PLA, activity (Fig. 3). PIAz activity from both types of cell lines accumulated in the medium. Neither 293s cells alonenor cell lines transfected with an unrelated PLA,cDNA inactivated by a 1-base pair deletion at the 5’ end of the matureprotein produced detectable PLAz activity in the mediumeven after 72 h. Cell lysates from the cell cultures used to collect the 72-h conditioned medium samples were prepared by sonication in 0.9% NaCl and showed approximately 50% of the activity of 72-h medium samples (data not shown). Characterization of Enzymatic Activity-TWO cell lines, pCH10-1D expressing pCHlO and C W F - 2 B expressing pCRASF, were chosen for further study. MaximalPLAz activity of HPLA2-10 is observed at pH 6.5 and remainsrelatively high up to pH 9.0, whereas the control group I1 PLA, reaches peak activity at pH 7.0 and then steadilydeclines (data notshown). HPLA2-10 appears to be a calcium-dependent PLAz having at about 10 IILM Ca2+. The activity of maximalactivity HPLA2-10 decreases in a linear manner to about half-maximal 400 II~M(data not activity as the Ca2+ concentration approaches shown). This activity profile differs from that of the group I1 control, which shows maximal activity at 1 mM Ca2+ and becomes inactive at Ca2+concentrations over 100 mM. Since HPLA2-10 expresses at least half of its maximal activity at Ca2+concentrationsbetween 1 and 100 ~llhl,similar to the previously described mammalian group I1 PIAz (311, HPLA2-10 may be capable of functioning at both intracellular calcium concentrations. (0.1-2 m) and extracellular (1111~)

Acknowledgments-We thank Drs. L. Carr and H. Thomasson for providing libraries and Dr. A. Siakotos for providing~-cr-l-palmitoyl-2oleoyl phosphatidylcholine.In addition, we are grateful to Drs. R. Harris and H. Edenburg for criticism of the manuscript. REFERENCES 1. Dennis, E. A. (1983)in The Enzvmes (Bover,

P..

ed) PP. 307-353. Academic Press, New York 2. Rordorf, G., Uemura, Y., and Bonventre, J. V (1991)J. Neurosci. 11, 18291836 3. Jordan, L. M., and Russo-Marie, F. (1992)J. Chromatogr: 597,299-308 4. Tam, S . W., Man, R. Y., and Choy, P. C. (1984)Can. J. Biochem. Cell Biol. 62, 1269-1274 5. Lindahl, M., von Schenck, H., and Tagesson, C. (1989)Biochim. Biophys. Acta 1005,282-288 6. Heinribon, R. L. (1991) Methods Enzymol. 197, 201-214 7. Davidson, F. E, and Dennis, E.A. (1990)J. Mol. Euol. 31,228-238 8. Heinribon, R. L., Krueger, E. T., and Keim, P. S . (1977)J. Biol. Chem. 252, 4913-4921 9. Seilhamer, J. J., Randall, T. L., Yamanaka, M., and Johnson, L.K. (1986)DNA 5, 519-527 10. Ohara, O.,Tamaki, M., Nakamura, E., Tsuruta,Y., Fujii,Y., Shin, M., Teraoka, H., and Okamoto, M. (1986)J. Biochem. (Tokyo)99, 733-739 11. Seilhamer, J. J., Pruzanski, W., Vadas, P., Plant, S., Miller, J. A., Kloss,J., and Johnson, L. K. (1989)J. Biol. Chem. 264,5335-5338 12. Kramer, R. M., Hession, C., Johansen, E., Hayes, G., McGray, P., Chow, E. P., Tizard, R., and Pepinsky, R. 3.(1989)J. Biol. Chem. 264,5768-5775 13. Komada, M., Kudo, I., Mizushima, H., Kitamura, N., and Inoue, K. (1989)J. Biochem. (Tokyo) 106,545-547 Biophys. 14. Kusunoki, C., Satoh, S., Kobayashi, M., and Niwa, M.(1990)Biochim. Acta 1087,95-97 15. Aarsman, A. J., de Jong, J. G., Amoldussen, E.,Neys, F. W., van Wassenaar, P. D., and Van den Bosch, H. (1989)J. Biol. Chem. 264, 1000~10014 16. Ono, T., Tojo,H., Kuramitau, S., Kagamiyama, H., and Okamoto, M. (1988)J . Biol. Chem. 263, 57325738 17. Vadas, P., and Pruzanski, W. (1986)k b . Invest. 55,391404 18. Pruzanski, W., and Vadas, P. (1990)Adv. Exp. Med. Biol. 279,239-251 19. Uhl, W., Buchler, M., Nevalainen, T. J., Deller, A,, and Eeger, H. G . (1990)J . ?Fauma 30,12861290 20. Malfertheiner, P., Nevalainen, T., Uhl, W., Schadlich, H., and Buchler, M. (1989)Klin. Wochenschr. 67, 183-185 21. Johnson, L. K., Frank, S . , Vades, P.,Pruzanski, W., Lusis, A. J.,and Seilhamer, J. J. (1990)Adu. Exp. Med. Biol. 275, 17-34 22. Chomczynski, P., and Sacchi, N. (1987)AnaZ.Biochem. 162, 15&159 23. Ishizaki, J., Ohara, O.,Nakamura, E., Tamaki, M., Ono,T.,Kanda, A., Yoshida, N., Teraoka, H., Tujo, H., and Okamoto, M. (19891Biochem. Biophys. Res. Commun. 162, 1030-1036 24. Elsbach, P., and Weiss, J. (1991)Methods Enzymol. 197, 24-31 25. Seilhamer, J. J., Plant, S., Pruzanski, W., Schilling, J., Stefanski, E.,Vadas, P., and Johnson, L. K.(1989)J. Biol. Chem. 106,3842 26. Nevalainen, T. J., and Haapanen, T.J. (1993)Znflammation 17,453459 27. Nalbone, G., and Hostetler, K.Y. (1985)J. Lipid Res. 26, 104-114 28. Bennett. C. F.. McCarte. A,, and Crooke, S . T. (1990)Biochim. Biophys. Acta 1047,- 271-283 29. Franson, R. C., and Weir, D. L. (1982)Lung 160,275-284 30. Cao, Y.Tam. S . W., Arthur, G., Chen, H., and Choy, P. C. (1987)J. Biol. Chem. 262. 16927-16935 31. Marshall, L. A,, and McCarte-Roshak, A. (1992)Biochem.Pharmncol. 1849-1858 32. Kingston, R. E. (1989)in Current Protocols in Moleculnr Biology (Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A,, Struhl, K., eds) Wiley Interscience, Ng.w York ”

44.