Site-directed mutagenesis studies on the iron-binding domain and the determinant for the substrate oxygenation site of porcine leukocyte arachidonate 12-lipoxygenase

BBALIP

54322

a Department of Biochemistry, School of Medicine, Tokmhima Uniniuersity, Tokushima 770 (Japan), b Division of Enzyme Genetics, Institute for Enzyme Research, Tokushima University, Tokushimna770 (Japan), ’ Equipment Research Center, I(igawa Me&d School, Kagawa 741-07 (Japan) and d Department ofPharmacology, National Gardiouascular Center Research Institute, Osaka 565 (j”iayanl (Received 16 August 1993)

Key words: Site-directed mutagenesis; Arachidonic acid; 12-Lipoxygenase; Prokaryotic expression; ~~o~-~~~~ingdorr,ain; Positionat specificity; (Porcine) cDNA for arachidonate 12-lipoxygenase of porcine leukocytes was expressed in IC~he~ichia coli. The recombinant 12-iipcxygenase was purified by immunoaffinity chromatography to near homogeneity with a specific activity of about 1.5 pmoH/ min per mg protein. Each of eight histidine residues, which were well-conserved among various mammalian ~~~o~g~~as~~ an ’ esum,od as ligands for non-heme iron, was substituted with leucine by site-directed mutagenesis. Each mutant en_qme was i raoafk &y-purified to near homogeneity. Mutations of His-361, -366 and -541 caused a total loss of enzyme ac , and the irou content was much lower (0.10, 0.06 and 0.06 g atom/mol protein) than that of the wild-type enzyme (0.53). tations of His-128 and -356 gave 159% and 162% specific activity of the wild-type enzyme, and the iron contents were 0.55 and 052 g atom/moi protein. Substitution of His-426 decreased the activity to 5%, but the iron conten: was 0.4 g atom/mol protein. The expression level of mutants at His-384 and -393 was too low to precisely determine the iron content. Taken together, His-361, -366 and -546 may play important roles for iron-binding in catalytically active 12Aipoxygenase. Since a high homotogy of ammo acid sequence was known between porcine leukocyte 12-lipoxygenase and mammalian LNipoxygenases, we attempted to convert the 12-hpoxygenase to a 15-lipoxygenase. A double mutation of V&418 and -419 to Ile and Met increased the ratio af 15- and IZIipoxygenase activities from 0.1 to 5.7

Lipoxygenases are a group of enzymes w porate a molecular oxygen into cis, c&pen% of polyunsaturated fatty acids to yield t peroxide derivatives. Three types of hpoxygenase have been distinguished on the basis of their preferred position of oxygenation at the carbon-5, -12 or -15 of arachidonic acid [I-31. §- and 15-Eipoxygenases are mvoived in the biosyntheses of leukotrienes and hpoxins, and play important roles in allergy and i~fl~mmation [4]. In contrast, any enzymatic product with a biological activity like prostaglandins, tbromand leukotrienes has not been identified in the

* Corresponding author. Fax: + 81 88633 6409. Abbreviations: 12-HETE, 12-hydroxy-52,8Z,lOE,l4Z-eicosatetraenoic acid; I5-HETE, 15-hydroxy-52,8Z,llZ,13E -eicosatetraenoic acid; HPLC, high-performance liquid chromatography. SSDI OQOS-2760(93)E0197-W

ohc getaway initiate lar biological studi ave been carried oud these several years, and

cel%s (S), ~~-~~~~~~e~~~e~ fiw~-~~ II>, porcine leukocytes 112) 2nd Sovine trachea

4139, and

~~-~~~~~~e~a§~~

from

human

rabbit (15) ~~t~~~~~~~t~s~ T rnarnrna~~~~ hpoxygenases contain an aimos: equimolar amount of non-heme iron [I$-18!F and a linear relations e tent and the en I. In the central region of these

04)

309

tochrome c from Sigma, and iron standard solution from Wako. Restriction endonucleases and T7 DNA sequencing kit (Sequenase version 2.1) were purchased from Toyobo, and a restriction enzyme Eco311 was kindly donated by Mr. M. Fujinaka of Toyobo. A plasmid pKK223-3 was supplied by Pharmacia, and GeneAmp PCR Reagent kit from Takara Shuzo. Oligonucleotides were synthesized using an Applied Biosystems DNA synthesizer 380A. ~ri~c~t~o~ and usxzy of ~Z-l~~o~ge~~e. 12-Lipoxygenase of porcine leukocytes was purified as described previously [291. The purification steps included ammonium sulfate precipitation, DEB-cellulose chromatography, and immunoaffinity chromatography employing a monoclonal antibody (lox-2) against the enzyme [33]. 12-Lipoxygenase activity was determined using [l- 14C]arachidonic acid as previously described 1291. Specific activity of the purified enzymes was determined by 1-min incubations at 30°C and expressed as pmol/min per mg protein. Alternatively, the enzyme activity was determined by folIo~ng absorbance at 234 nm (E 27600 M-l cm-“> with arachidonic acid and other substrates. Protein concentration was determined by the method of Lowry et al. [34]. SDS-polyacrylamide gel electrophoresis was performed by the method of Laemmli 1351.Western blotting was carried out using poly~lonal antibody raised in rabbit against porcine leukocyte 1Zlipoxygenase [36].

assessed by X-ray absorption analysis [20,21]. Role of these histidines was investigated by site-directed mutagenesis in only human 5lipoxygenase [23-251. In this study, we extensively examined the roie of conserved histidines in porcine leukocyte 1Zlipoxygenase by sitedirected mutagenesis. A close resemblance of porcine leukocyte lZlipoxygenase to 15lipoxygenases has been well documented in terms of amino acid sequences and catalytic properties. Namely, the amino acid sequence of porcine leukocyte 1Zlipoxygenase showed 86% and 79% identities to human and rabbit reticulocyte 15lipoxygenases, respectively [2]. 12Lipoxygenases of several animal species exhibited a minor activity of 15 oxygenation with either arachidonic acid [26,27] or 5-hydro~-6,8,11,14-ei~osatetraenoic acid [28-301. On the other hand, 15”lipoxygenases of rabbit [31] and human [32] reticulocytes showed a small activity of 12-oxygenation. In connection with such a close relatedness of 12- and 15-lipoxygenases, we considered that the oxygenation site of porcine 1Zlipoxygenase could be changed from C-12 to C-15 if several amino acids of the enzyme were replaced by the corresponding amino acids of 15-lipoxygenase by site-directed mutagenesis. Materials and Methods Materials. [l- 14C]Arachidonic acid (1.85 GBq/ mmol) and [cw-32PldCTP (111 TBq/mmol) were purchased from Amersham, and arachidonic acid from Nu-Chek Prep. Vectastain ABC kit was supplied by Vector Laboratories, soybean Iipoxygenase and cy-

384

393

PSIHPIFKLLIPHFRYT

398

Construction of expression uector and site-directed ~~tuge~e~~~. An expression vector pKK223-3 carrying tat promoter was employed for the expression of the

porcine leukocyte 1Zlipoxygenase

419

128 __

356

361 .~

366

131

363

368

373

426 TGGGGH

533 --------------

cDNA (see Fig. 2).

541

CHFVTMCIFTCTGQHSSNH

PsIHPMFKLLIPHLRYT

TGGGGH

--------------

CHFVTMCIFTCTGQHSSTH

PGLHPIFKFLIPHIRYT

TGGGGH

--------------

CHFLTMCVFTCTAQHAAIN

PSIHPVFKLI~HLRYT

TGGGGH

--------------

CHFJTMCIFTCTGQHSSIH

PSIHPIFKLIIPHLRYT PAvHPLFKLLVAHVRFT

TGGGGti -------------TGGGGH --------------

CHFVTMCIFTCTGQHASVH

I- 1 PA~HPIFKLLVAHVRFT~~_G~INTK~RE~L:CECGLFDK~~TGGGGH 391 400

___________^__

SEYLTVVIFTASAQHAAVN SEY~TVV~FTASAQH~~

433 551 Fig. 1. Alignment of amino acid sequences of mammalian lipoxygenases deduced from their cDNAs. Amino acids are counted from the initiating methionine. The numbers above the sequences are those for porcine leukocyte 1Zlipoxygenase and those written in italic below the sequences are for human leukocyte 54ipoxygenase. Conserved histidines and cysteine are represented by bold letters. Boxed are amino acids that are specifically found in 12-, 15.. and Mipoxygenases irrespective of species differences. Porcine 12-LOX, porcine leukocyte 1Zlipoxygenase [12]; Bovine 12-LOX, bovine tracheal 12-lipoxygenase [13]; Human 12-LOX, human platelet IZlipoxygenase [9]; Rabbit Z-LOX, rabbit reticulocyte 15-lipoxygenase [El; Human E-LOX, human reticulocyte 15-lipoxygenase [14J; Rat SLOX, rat basophil 5-lipoxygenase [8]; Human 5-LOX, human leukocyte Slipoxygenase [6].

Synthetic adaptor

Fig. 2. Construction of an expression plasmid p 0x12 for 12lipoxygenase of porcine leukocytes. Three fragments of 12-lipoxygenase cDNA and the synthetic adaptor were ligated to an expression plasmid, pKK223-3 as described under Materials and Methods.

cDNA fragments were excised from three clones previously isolated [12]; pLOX134 (nucleotides

thesized to cover the synthetic oligonucle XT-3’ and 5’-GG phosphorylated at 5’ terminus kinase and annealed to create E at 5’ and 3’ ends, respectively.

lasmid, designated mto E. coEi strain W

merase was verified by eoxy chain-termination

using t

expression of the 124ipoxygeriase protein, and the incubation was contiriued for 5 h. Cells were harvested by centrifugation at 5 the pellet (approx. 10 g wet weig

311

A

123456

B

123456

Fig. 3. SDS-polyacrylamide gel electrophoresis of recombinant 12lipoxygenase at various purification steps. (A) Stained with Coomassie brilliant blue R250. (I31 Western blot. Lane 1, 105000X g supernatant of E. coli transformed with pKK223-3; Lane 2, 105000X g supernatant of E. coli transformed with pKKLOX12; Lane 3, ammonium sulfate fractionation; Lane 4, DEAE-cellulose chromatography; Lane 5, immunoaffinity chromatography; Lane 6, porcine leukocyte 1%lipoxygenase purified by immunoaffinity chromatography.

observed at each purification step. Specific activity of the 105 000 X g supernatant of the E. coli sonicate was 0.03-0.05 pmol/min per mg protein as determined with arachidonic acid as substrate. The enzyme was purified with a yield of about 4% to give a specific activity of 1-2 pmol/min per mg of protein, which was comparable with that of the purified enzyme from porcine leukocytes [29]. Approx. 2 mg of the purified enzyme was obtained from a 2-l culture. The amino terminal sequence of the recombinant enzyme, GLYRVRVSTGSSFYAGSQNQVQ, was identical with that of the porcine leukocyte enzyme [12]. The recombinant enzyme oxygenated a-linofenic, y-linolenic and linoleic acids all with 18 carbon atoms at 89%, 92% and 70% the rate of arachidonic acid oxygenation, respectively. The broad chain-Iength specificity of the substrates is characteristic of porcine leukocyte ZZlipoxygenase [29]. Hydroperoxy reaction products of the recombinant enzyme from arachidonic acid were reduced, and analyzed by reverse-phase HPLC. As shown in Fig. 4A, a major peak comigrated with 12-HETE, and its identity with 1ZHETE was confirmed by gas chromatography-mass spectrometry. The 12HETE was mlostly of S-configuration as examined by chiral-phase HPLC. Thus, the recombinant enzyme was indistinguishable from the native enzyme of porcine leukocytes in terms of physicochemical properties, immunogenicity and catalytic functions. A small peak co-migrating with 15-HETE was observed in about 10% the amount of 12-HETE (Fig. 4A). Such a minor 1:5-lipoxygenase activity had been

Retention Time (min) Fig. 4. Reverse-phase HPLC of products from arachidonic acid by the wild-type and mutant 12lipoxygenases. lmmunoaffinity-purified enzymes were incubated with 25 PM arachidonic acid at 30°C for 1 min. The borohydride-reduced products were injected into a reverse-phase column at a flow rate of 1 ml/min with a solvent system of methanol/water/acetic acid (80:20:0.01). A234 due to conjugated diene was monitored. (A) wild-type enzyme, (B) Ile-106 was replaced by Val, (C) Met-398 by Leu, (D) Val-419 by Met, (El Val-418 by Be, (F) both Val-418 and -419 were replaced by Ile and Met, respectively. Retention times of authentic 12-HETE and 15HETE are as indicated.

observed with 1Zlipoxygenases of several animal tissues [26-301, but the possibility of 15-lipoxygenase contamination could not be ruled out. Since E. coli lacks its own lipoxygenase, the presence of 15-lipoxygenase activity in the recombinant 1Zlipoxygenase shows that the porcine leukocyte 1Zlipoxygenase can oxygenate C-15 of arachidonic acid to a lesser extent in addition to C-12 as the major oxygenation site. Site-directed mutagenesis of conserved histidines

Fig. 1 shows an alignment of amino acid sequences of various mammalian lipoxygenases deduced from their cDNAs [6,8,9,12-151 *. Eight histidine residues at 128, 356, 361, 366, 384, 393, 426 and 541 of porcine 12-lipoxygenase were conserved in all the other lipoxygenases. In addition to these His residues, Cys-533 was conserved in 12- and 15-lipoxygenases, but not in 5lipoxygenases [12]. These nine amino acids were selected as targets for site-directed mutagenesis of

* Numbering of the amino acids of human leukocyte 5-lipoxygenase is inconsistent among researchers 16-81. In this paper the position of the initiation methionine is referred to as No. 1 as shown in Fig. 1.

enzymes

in

cob a

positivespsts,but

t

mutants at 366, 384, in E. coki at YX, and were more s temperature [41]. When we reduced

servation, when the E. co&i was grown at 37” 5lipoxygenase was expressed as an inso presumably encapsulated ian the inclusion by reducing the expression te

temperature

dependency of the expression was also tant 5lipoxygenases [24j and rice ch a temperature

Fig. 5. Immunoblot ar;mlysis of the mutaat 124ipozqgenases ifi E. c3Z grown at 37°C and 20°C. Aliqvots of the 105 OQOX g superaatan:of each clone 419 ;i~g) were b1ot:ed on nitrocelhkse membrane, and were immucostained with anti-12-lipozqgenase polyclnnai antibody. Sham, E. coli transformed with pKM223-3 (lacking the cDNA for 12-lipoxygenase).

105 000 x g s~pe~~a~~~~ in yields as the protein amount of the wild-type en-

se mutaM

eazy

Enzyme actiuities and iron contents of wild-type and mutant I24poxygexases

Wild-type and mutant 12-lipoxygenases carrying the indicated mutations were expressed at 20°C in 2 I of E. coii culture, and were was assayed with arachidonic acid immunoaffinity-purified. The amount of the purified enzyme protein was determined, and the enzym,p &vity i as substrate [29]. The purified enzymes were subjected to atomic absorption spectrometry, and tbe iron content was cahlated as reported of several separate experiments, previously [17]. n.d., assayed but not detectable. -, not determined. Data ar-- vepresentative I Mutation

Wild-type His-128 + His-356 + His-361 + His-361 + His-366 + His-384 + His-393 + His-426 + His-541 + Cys-533 +

Codon change

Leu Leu Leu Cln Len Lea Leu Leu Leu Ser

CAC + CTC CAT + CTT CAC + CTC CAC --) CAA CAC + CTC CAC + CTC CAC --j CTC CA@ -+ CTC CAC --) CTC TGT + TCT

Total Protein mg (%I

Specific activity pmol/min

2.21 (100) 2.33 (105) 2.17 (96) 2.46 (Ill) 2.35 (106) 1.81 (82) 0.20 (9) 0.53 (24) 2.58 (113) 2.26 (102) 2.07 (94)

1.52 (100) 2.41 (159) 2.46 (162) n.d. ¶.d. nd. (3.66(43) Rd. 0.08 (5) n.d. 2.20 (145)

Iron content ig atom/mol) per mg (%I 0.53 0.55 0.52 0.10 0.01 0.06 _ 0.40 0.06 0.66

-_

313

;Da

Wild H128,L H356bL H361 ,L H366bL H384bL H393bL H426bL H541bL c533e Fig. 6. SDS-polyacrylamide gel electrophoresis of purified recombinant 12-lipoxygenases. E. cd was transformed by each recombinant plasmid carrying a mutation as denoted, and the expressed enzymes were purified by immunoaffinity chromatography with a monoclonal anti-12-lipoxygenase antibody as a ligand 1291.The purified enzyme (2 pg of protein) was subjected to SDS-polyacrylamide gel electrophoresis (10%) followed by staining with Coomassie brilliant blue R250.

specific enzyme activities were 159% and 162% that of the wild-type enzyme. These mutant enzymes contained 0.55 and 0.52 g atom of iron per mol of enzyme, which were almost the same as the iron content of the wild-type enzyme (0.53 g atom/mol). The results suggested that these His residues were not crucial for the iron-binding and the enzyme activity. These values of iron content were not very close to 1 g atom/m01 of enzyme. However, these enzyme preparations were not activated by the addition of 20 PM ferrous ammonium sulfate to the assay mixture. According to site-directed mutagenesis studies on 5-lipoxygenase [23-251, His-363 was not considered to be essential for the 5-lipoxygenase activity. This His residue is equivalent to His-356 of porcine 12-lipoxygenase as illustrated in Fig. 1. Secondly, the mutations of His-361, -366 and -541 gave totally inactive 12Jipoxygenases and very low iron contents (0.10, 0.06 and 0.06 g atom/mol). Moreover, the mutation of His-361 to Gln also brought about a totally inactive enzyme and a very low iron content (0.01 g atom/mol). The concomitant loss of the enzyme activity and the iron suggested an important role

of these His residues for the enzyme activity and the iron-binding. In relation to these observations, mutations of the equivalent His residues of human 5-lipoxygenase (His-368, -373 and -551) gave totally inactive enzymes [23-251. Thirdly, a mutation of His-426 to Leu decreased the specific enzyme activity to only 5%, but the enzyme had a comparable amount of iron (0.40 g atom/mol) with the wild-type enzyme. The results suggest that His-426 is not essential for the iron-binding of the 12-lipoxygenase, but is important for the enzyme activity. With regard to this mutation, Ishii et al. reported that the mutant 5-lipoxygenases with replacement of His-433 (equivalent to His-426 of porcine leukocyte 1Zlipoxygenase) by Asn and Ala showed 56% and 83% activity of the wild-type enzyme [25]. On the basis of the finding that the molar volume of these amino acids are in the order of Leu, Asn and Ala (107.5, 73.8 and 60.4 cm3/mol, Ref. 441, the decrease of the enzyme activities of the mutant 5- and 12-lipoxygenases seemed to correlate to the mass of the amino acids. The bulky side chain of the amino acids might decrease the enzyme activities by interferring with the binding of substrate to the active site of the enzymes or causing a significant change of the protein structure. The purified His-384 mutant enzyme showed a significantly high specific activity (43% of that of the wild-type enzyme), but the total enzyme protein was only 9% of that of the wild-type enzyme, which was too low to determine an accurate iron content. In accordance to our findings, a 5-lipoxygenase carrying a mutation at the equivalent His (His-391) was expressed with 67% the activity of the wild-type enzyme in an insect cell system [23]. However, in E. coli system, mutations of the His residue to various amino acids markedly decreased the production of the mutant enzymes [23-251. These findings suggested that the expression of these mutant 1Zlipoxygenase and 5-lipoxygenases were significantly influenced by experimental conditions, and further investigations are required to conclude the importance of the amino acid (His-384 of porcine 1Zlipoxygenase and His-391 of human 5lipoxygenase). A mutant enzyme at His-393 was totally inactive, and the total protein was only about one fourth of the wild-type enzyme. On electrophoretic analysis this mutant enzyme gave a major band corresponding to the wild-type enzyme and another minor band with a smaller molecular weight (Fig. 6). This result suggested that the mutation at His-393 made the 1Zlipoxygenase susceptible to proteolysis. Limited availability of this mutant enzyme did not encourage the determination of its iron content. Previously, mutations of the equivalent His residue at 400 of human 5-lipoxygenase were carried out using E. coli (23-25) and insect cell [23] expression systems. In the E. coli system a mutation of

5lipoxygenase

wi

er-400 was expressed with

75%

human Wpmygenase

of the role of these Cys-533 of porcine cyte ~2-~~~~~ge~a~e was ~~~~~~~~~ypresumed to co tit&e structure with several hist shown in Table I, the neither the enzyme activi gesting no essential role this amino acid for t ~~~~-~~~d~~g. e of His-361, -366

is residues for the catalytic activity and

-541 of porcine 12-21 th the enzyme act concluded that these

-361 to Leu 0.1

the iron content,

and.

or

g

ato

. The other two mutants were obtained in too small amounts In a very recent report [46], the three ucture of soybean ~~~Q~ge~a~e-~ was -ray c~stallog~a~~y. The issn in the

enzyme

was

315

rat pineal gland by our group [50], and the three amino acids corresponding to Gln-417, Val-418 and Val-419’ of porcine 12-lipoxygenase were Lys-Ala-Met at position 417 to 419 in the rat 1Zlipoxygenase. A mutation of Lys-Ala-Met to Gln-Be-Met [%I and that of Ala to Ile [50] did not increase 15lipoxygenase activity of the rat 12~lipoxygenase. Thus, the finding with the rat 1Zlipoxygenase is not consistent with our conclusion obtained from the findings with human reticulocyte 15lipoxygenase and porcine leukocyte 12-lipoxygenase that the two amino acids (Be-417 and Met-418 for the former and Val-418 and Val-419 for the latter) play a crucial role for the determination of the positional specificity of arachidonate oxygenation. The solution of this subject awaits further investigations. Acknowledgements This work was supported by grants-in-aid from the Ministry of Education, Science and Culture of Japan, and grants from the Japanese Foundation of Metabolism and Diseases, the Japan Foundation for Applied Enzymology, the Ono Medical Research Foundation, and the CIBA-GEIGY Foundation for the Promotion of Science. References 1 2 3 4 5 6

7

8

9

10 11 12

13 14

15

Yamamoto, S. (1991) Free Radic. Biol. Med. 10, 149-159. Yamamoto, S. (1992) Biochim. Biophys. Acta 1128, 117-131. Sigal, E. (1991) Am. J. Physiol. 260, L13-28. Samuelsson, B., Dahltn, S.-E., Lindgren, J. A, Rouzer, C.A. and Serhan, C.N. (1987) Science 237, 1171-1176. Brash, A.R. (1985) Circulation 72, 702-707. Dixon, R.A.F., Jones, R.E., Diehl, R.E., Bennett, C.D., Kargman, S. and Rouzer, C.A. (1988) Proc. Natl. Acad. Sci. USA 85, 416-420. Matsumoto, T., Funk, C.D., Rbdmark, O., H&g, J.-O., Jdrnvall, H. and Samuelsson, B. (1988) Proc. Natl. Acad. Sci. USA 85, 26-30 and 3406 (correction). Balcarek, J.M., Theisen, T.W., Cook, M.N., Varrichio, A., Hwang, S.-M., Strohsacker, M.W. and Crooke, S.T. (1988) J. Biol. Chem. 263, 13937-13941. Yoshimoto, T., Yamamoto, Y., Arakawa, T., Suzuki, H., Yamamoto, S., Yokoyama, C., Tanabe, T. and Toh, H. (1990) Biochem. Biophys. Res. Commun. 172, 1230-1235. Izumi, T., Hoshiko, S., RHdmark, 0. and Samuelsson, B. (1990) Proc. Natl. Acad. Sci. USA 87, 7477-7481. Funk, C.D., Furci, L. and FitzGerald, G.A. (1990) Proc. Natl. Acad. Sci. USA 87, 5638-5642. Yoshimoto, T., Suzuki, H., Yamamoto, S., Takai, T., Yokoyama, C. and Tanabe, T. (1990) Proc. Natl. Acad. Sci. USA 87, 21422146. De Marzo, N., Sloane, D.L,., Dicharry, S., Highland, E. and Sigal, E., (1992) Am. J. Physiol. 262, L198-L207. Sigal, E., Craik, C.S., Highland, E., Grunberger, D., Costello, L.L., Dixon, R.A.F. and Nadel, J.A. (1988) Biochem. Biophys. Res. Commun. 1.57, 457-464. Fleming, J., Thiele, B.J., Chester, J., O’Prey, J., Janetzki, S.,

A&ken, A., Anton, I.A., Rapoport, S.M. and Harrison, P.R. (1989) Gene 79, 181-188. 16 Percival, M.D. (1991) J. Biol. Chem. 266, 10058-10061. 17 Matsuda, S., Suzuki, H., Yoshimoto, T., Yamamoto, S. and Miyatake, A. (1991) Biochim. Biophys. Acta 1084, 202-204. 18 Schewe, T., Rapoport, S.M. and Kiihn, H. (1986) Advances in Enzymology 58, 191-272. 19 Shibata, D., Steczko, J., Dixon, J.E., Andrews, P.C., Hermodson, M. and Axelrod, B. (1988) J. Biol. Chem. 263, 6816-6821. 20 Navaratnam, S., Feiters, MC., Al-Hakim, M., Allen, J.C., Veldink, G.A. and Vliegenthart, J.F.G. (1988) Biochim. Biophys. Acta 956, 70-76. 21 Dunham, W.R., Carroll, R.T., Thompson, J.F., Sands, R.H. and Funk, M.O. (1990) Eur. J. Biochem. 190, 611-617. 22 Funk, C.D., Gunne, H., Steiner, H., Izumi, T. and Samuelsson, B. (1989) Proc. Natl. Acad. Sci. USA 86, 2592-2596. 23 Nguyen, T., Falgueyret, J.-P., Abramovits, M. and Riendeau, D. (1991) J. Biol. Chem. 266, 22057-22062. 24 Zhang, Y.Y., Ridmark, 0. and Samuelsson, B. (1992) Proc. Natl. Acad. Sci. USA 89,485-489. 25 Ishii, S., Noguchi, M., Miyano, M., Matsumoto, T. and Noma, M. (1992) Biochem. Biophys. Res. Commun. 182, 1482-1490. 26 Takahashi, Y., Ueda, N. and Yamamoto, S. (1988) Arch. Biochem. Biophys. 266, 613-621. 27 van der Donk, E.M.M., Vervaat, J.M.A., Verhagen, J., Veldink, G.A. and Vliegenthart, J.F.G. (1992) Biochim. Biophys. Ada 1128, 14-25. 28 Yoshimoto, T., Miyamoto, Y., Ochi, K. and Yamamoto, S. (1982) Biochim. Biophys. Acta 713, 638-646. 29 Yokoyama, C., Shinjo, F., Yoshimoto, T., Yamamoto, S., Oates, J.A. and Brash, A.R. (1986) J. Biol. Chem. 261, 16714-16721. 30 Hada, T., Ueda, N., Takahashi, Y. and Yamamoto, S. (1991) Biochim. Biophys. Acta 1083, 89-93. 31 Bryant, R.W., Bailey, J.M., Schewe, T. and Rapoport, S.M. (1982) J. Biol. Chem. 257, 6050-6055. 32 Sigal, E., Grunberger, D., Highland, E., Gross, C., Dixon, R.A.F. and Craik, C.S. (1990) J. Biol. Chem. 265, 5113-5120. 33 Shinjo, F., Yoshimoto, T., Yokoyama, C., Yamamoto, S., Izumi, S., Komatsu, N. and Watanabe, K. (1986) J. Biol. Chem. 261, 3377-3381. 34 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 35 Laemmli, U.K. (1970) Nature 227, 680-685. 36 Maruyama, T., Ueda, N., Yoshimoto, T., Yamamoto, S., Komatsu, N. and Watanabe, K. (1989) J. Histochem. Cytochem. 37, 1125-1131. 37 Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) in Molecular Cloning: A Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 38 Higuchi, R., Krummel, B. and Saiki, R.K. (1988) Nucleic Acids Res. 16, 6545-6551. 39 Sanger, F., Nicklen, S. and Co&on, A.R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463-5467. 40 Van OS, C.P.A., Rijke-Schilder, G.P.M., Van Halbeek, H., Verhagen, J. and Vliegenthart, J.F.G. (1981) Biochim. Biophys. Acta 663, 177-193. 41 Schein, C.H. (1989) Bio/Technology 7, 1141-1149. 42 Noguchi, M., Matsumoto, T., Nakamura, M. and Noma, M. (1989) FEBS Lett. 249, 267-270. 43 Shirano, Y. and Shibata, D. (1990) FEBS Lett. 271, 128-130. 44 Zamyatnin A.A. (1984) Ann. Rev. Biophys. Bioeng. 13, 145-165. 45 Zhang, Y.Y., Lind, B., RBdmark, 0. and Samuelsson, B. (1993) J. Biol. Chem. 268, 2535-2541. 46 Boyington, J.C., Gaffney, B.J. and Amzel, L.M. (1993) Science 260, 1482-1486. 47 Arakawa, T., Oshima, T., Kishimoto, K., Yoshimoto, T. and Yamamoto, S. (1992) J. Biol. Chem. 267, 12188-12191.

48 Sloane, D.L., Leung, Hp.,Craik, C.S. and Sigd, E. (1991) Nature 354, 149-152. , J.F., eggstr6m, J.Z., FSMnark, 0. end 49 Watanabe, T., Eur. 9 ‘o&em. 212, 605-612. Samuelsson, B. ., Suzuki, W., Arakawa, T., Nakamwa, M., 50 Hada, T., Hagiya,

Matsuda, S., Yoshimoto, T., UaEamoeo, S., _Azh!#it, ‘r., Mor:t:i. u., ishimura, K. and :Kirn>H.-Y. :1993) Bitchim. Biophys. &?a, e, 7. and Haeggslriim, I.Z. 11993) Biochcnr. Biophys. es. commun. 192, 1023-1029,