Fur J Riochem. 209,993-998 (1992) 'CI 1. kBS 1992

Unusual amino acid substitution in the anion-binding site of Lactobacillus plantarum non-allosteric L-lactate dehydrogenase Hayao TAGUCHT and Takahisa OIITA Dcpartmcnt of Agricultural Chemistry, The University of Tokyo, Japan (Received June 15/August 10. 1992)

-

EJR 92 0832

In Lactobacillus pluntarullz non-allosteric L-lactate dehydrogenase (L-LDH), the highly conserved His188 residue, which is involved in the binding of an allosteric effector, fructose 1,6-bisphosphate [Fru(1,6)P2], in allostcric L-LDH is uniquely substituted by an Asp. The mutant L . plantarum LLDH, in which Asp188 is replaced by a His, showed essentially the same Fru(1,6)P2-independent catalytic activity as the wild-type enzyme, except that the K, and V,,, values were slightly decreased. However, the addition of Fru( 1,6)P2 induced significant thcrmostabilization of the mutant enzyme, as in the case of many allosteric L-LDHs, while Fru(l,6)P2 showed no significant effect on the stability of the wild-type enzyme, indicating that only the single-point mutation, G+C, sufficiently induces the Fru(1 ,6)P2-bindingability of L . plunlarum L-LDH. The mutant enzyme showed higher thermostability than the wild-type enzyme in the presence of Fru(l,6)P2. In the absence of Fru(l,b)P,, on the other hand, the mutant enzyme was more labile below 65°C but more stable above 70°C.

The high variety of lactate dehydrogenases (LDH) has been demonstrated in Lactobacilli [l -31. Thcsc enzymes arc classified into three classes : allosteric and non-allosteric Llactate dehydrogenases (L-LDH) (EC 1.1.1.27) and n-lactate dehydrogenases (D-LDH) (EC 1.I .1.28). The distinct multiplicity of the LDHs makes them valuable taxonomic characteristics for species differentiation within this genus. However, little is known about the molecular bases of these characteristics. In a previous paper. we reported the sequences of the L- and D-LDH genes of L . plantarum, and demonstrated that D-LDH is a member of the D-isomer-specific 2-hydroxyacid dehydrogenase family, which is distinct from that of L-LDHs [4]. On the other hand, the amino acid sequence of L. planturunz L-LDH, which is a non-allosteric L-LDH, shows similarity to all known L-LDH sequences, and especially high identity with that of L. cusei allosteric L-LDH (66%), indicating that these two enzymes divcrged recently. In spitc of the especially high sequence identity, there are marked diffcrcnces in the catalytic properties between L . plcintarum and L. casei L-LDHs. L. casei L-LDH absolutely requircs Fru(1 ,6)Pz and certain divalent cations for its catalytic reaction at physiological pH. IJnder slightly acidic conditions, L . M . S ~ ~ L - L D shows H a sigmoidal saturation curvc for pyruvate in the absence of Fru(l,6)P2,but shows a hyperbolic saturation curve in the presence of Fru(l,6)P2 [5, 61. In con-

trast, L. plantarum L-LDH does not require Fru(l,6)P2 for its activity [l, 3, 71, and shows hyperbolic saturation curves for pyruvate, independently of Fru(l,6)P2, as in the case of L. casei enzyme in the presence of allosteric factors. L-LDHs are widely distributed in organisms and tissues; their structures and functions have been studied in detail [8]. Allosteric L-LDHs, which are usually activated by Fru(1 ,6)P2, have been found mostly in bactcrial cells [l]; mechanisms underlying the allostcric regulation have been extensively studied in the cases of the allosteric enzymes of L. casei [7,9 131, Thermus caldophilus [14- 191, Bacillus stearothermophilus [20-261 and Bifidobucterium longum [27-291. It has been indicated that the Fru(1,6)P2-binding site of allosteric enzymes correponds to the anion-binding site of non-allosteric vcrtcbrate enzymes, two Fru(1 ,6)P,-binding sites facing each other at the subunit interface. In particular, the Arg173 and His188 residues, which are highly conserved in both vertebrate non-allosteric and bacterial allosteric L-LDHs, have been demonstrated to be involved in the binding of Fru(l,6)P2 and the regulation of the enzyme activity. In this paper, we describe an unusual amino acid substitution in the anion-binding site of L. plantarum non-allosteric L-LDH, and analysis of the effect of this substitution on the enzyme propertics by means of sitc-directed mutagenesis.

Correspondence to H. Taguchi, Department of Agricultural Chcmistry, The University of Tokyo, Bunkyo-ku, Tokyo, Japan 113 Fax: +81 3 3812 0544. Abbreriiatiom. L-LDM. L-lactate dehydmgenase; D-LDH, Dlactate dehydrogenase; Fru(1 .6)P2, fructose 1.6-bisphosphatc. Enzyimes. L-Lactate dehydrogenase, (9-1actate:NAD' oxidoreductasc (EC 1.1.1.27); ~ - ] ~dehydrogenase, ~ t ~ t ~ (R)-lactatc:NA4D+ oxidoreductase (EC 1.1.1.28). A'ote. The novel amino acid sequence data published here have been dcpositcd with the EM BL sequence data bank and arc available under the accession number D90340.

MATERIALS AND METHODS Organisms and enzyme purification L. plantnuurnJCM 1558 (ATCC 8041) was obtained from the Japan Collection of Microorganisms, The Institute of and Research (Saitama, Japan), and tured as described previously [4]. The L-LDH enzyme was Purified essentially accordin&to a Previous report [71 as Preparations from both L. pluntarum cells and Escherichiu c d i strain MV1184 cells carry an expression plasmid for the L.

994 1

!0

20 2 2

40

30

50

60

LP LC BS

BL TC IIFM

4 T L i(

D K L I C A L A 80 81 8 3

70

100

90

103 105 1

110

120

*t

BL TC DFM 130

132

150

140

170

160

*

180

LP

Ic. BS

BL TC DFV

LP LC

BS

BL TC OEM

290

250

LP LC

BS BL TC

DFM 299 301

. .

310

320

LP LC

BS EL TC

DPM

330 a b

;ui i :

K

Re

F T R G - F

G - F K - F

Fig. 1. Comparison of the amino acid sequences of L-LDHs. The amino acid sequence of L. plantarurn L-LDH (LP) [4]is compared with those of 1.-LDITs from L. cusei ( L S ) [12], B. stearathermaphilus (BS) [34, 351, Bifidobacterium lungum ( B L ) [28], Therrnus cnldophilus (TC) [15] and dogfish muscle (DFM) [36]. Amino acids of the L. plantarurn L-LDH identical with ones of the other enzymes are boxed, and those identical in all L-LUHSare indicated by stars. Amino acids constituting the Fru(1 ,6)P,-binding site (anion-binding site) are indicated by closed circles. The residues are numbered according to Eventoff et al. [43]

plantarurn L-LDH gene; however, in the final step, the enzyme preparation was purified by fast protein liquid chromatography (FPLC) on Mono Q HR5/5 (Pharmacia LKB Biotechnology Inc.) with 0 -0.4 M NaCl gradient, instead of the second affinity chromatography. Protein concentrations were determined with the bicinchoninic acid protein reagent (Pierce Chemical Co.) [30], using bovine serum albumin as a standard protein. The purity of the enzyme preparation was examined by SDSIPAGE according to Laemmli [31]. Enzyme assay

The standard assay for the enzyme activity was performed at 30°C in 50 mM Na-Mops pH 7.0, containing 0.1 mM NADH and 20 mM pyruvate. One unit was defined as the amount of enzyme catalyzing the conversion of 1 pmol of su bstrate/min. Gene expression and site-directed mutagenesis

The enzymes used for DNA manipulation were purchased from Boeringer Mannheim, Takara Shuzo Co. (Kyoto, Japan) and Toyobo Biochemicals (Tokyo, Japan). Oligodeoxynucleotides 5‘-AAT TCA TGT CAA GCA TGC ATC GAT CTG CA-3’ and 5’-GAT CGA TGC ATG CTT GAG ATG-3’ were synthesized to construct an expression plasmid for the L . plantarum L-LDH gene in E. coli

cells; 5’-CCT CGC TCC GTG CAT GCT TAC ATC A-3’ was synthesized as a mutagenesis primer to replace Asp188 with a His in L. plantarum L-LDH, using an Applied Biosystems 380B DNA synthesizer. Site-directed inutagenesis was performed using a Muta-gene in vitro mutagenesis kit (BioRad), according to Kunkel [32]; the DNA fragment was sequenced to prove that only the mutation expected had occurred. DNA sequencing was performed by the dideoxy-chainterminator procedure [33] using an Applied Biosystems model 370A DNA sequencer, the chain elongation being carried out with Tth DNA polymerase (Toyobo Biochemica 1s) at 70°C.

RESULTS Comparison of the amino acid sequence of the L . pluntuvurn L-LDHwith those of other LDHs

Fig. 1 compares the primary structure of L. plantarum L-LDH [4] with those of other representative L-LDHs. Its sequencc identities to the L. casei [12], Bacillus stearothermophilus [34,35], Bijdobacteriurn longuwi [28], Thcrmus caldophilus [IS] and dogfish musclc enzymes [36] are 66%, 51%, 39%, 37% and 36%, respectively. The especially high sequence identity between the L . plantarunz and L. rasei enzymes is consistent with the results of comparative analysis [7] and an immunochemical study [6] and indicates that these two

995 Luclohacillus enzymes diverged recently. The sequence comparison also showed obviously less conserved regions in these two enzymes: the NH2-and COOH-terminal and central (positions 201 -222) regions, which are also poorly conserved regions in all of the L-LDH sequences. The residues constituting the Fru(1 ,6)P,-binding site, which corresponds to the anion-binding site in non-allosteric vertebrate L-LDHs [13], are also especially highly conserved in L. pluntarum and L. casei L-LDHs (Fig. 1). As a striking difference, however, the highly conserved His188 residue, which is essential for the Fru(l,b)P,-binding of allosteric L-LDHs [I 8, 261 and the chemical modification of which induces the loss of the regulation by Fru(l,6)P2 of L . cuseiL-LDH [13],is uniquely substituted by an Asp in L. plantarm L-LDH.

Expression of the L-lactate dehydrogenase gene in E. coli To express the L. pluntarum L-lactate dehydrogenase gene in E. coli cells, we constructed an expression plasmid for the L-LDH gene using an expression vector, pEXP7 [37], which was constructed from pKK223-3 (tuc promoter, ShineDalgarno sequence, multicloning site and rm-BT, BT,) and pUCl19 (ori, p-lactamase gene and M13 intergenic region). The EmRI - PstI part of the multicloning site of pEXP7 was replaced with a synthetic DNA, which contained the N-terminal coding region of the gene from an initiating Met codon to the SphI site followed by the ClaI site. In this synthetic DNA, the TTG initiation codon of the original gene [4] was replaced with an ATG. The SphI - CluI 1.2-kb fragment of the cloned gene 141 was inserted between the SphI site and the ClaI site of this vector to construct pEXLP, an expression plasmid for the L-LDH gene. E. coli cells harboring pEXLP produced a high amount of the active L-LDH (about 8% of the soluble protein) and the production of the L-LDH was not affected by an inducer, isopropyl /I-n-thiogalactopyranoside, probably because the amount of the lucZq gene product produced from the gene of the host cell is not sufficient to repress the tuc promoter on the expression plasmid. Replacement of Asp188 with His by means of site-directed mutagenesis The EcoR1 0.6-kb fragment of pEXLP was inserted into the EcoRI site of phage M13mp19 RF DNA in a suitable orientation for site-directed mutagenesis, and then site-directed mutagenesis was performed as describcd under Materials and Methods. The EcoR1 0.6-kb region of pEXLP was replaced with the mutated EcoRI 0.6-kb fragment obtained to construct pEXLPD188H, an expression plasmid for the mutant enzyme, in which Asp188 was replaced with His. The wild-type and mutant enzymes were produced in E. coli MVll84 cells carrying pEXLP and pEXLPD188H, respectively, which were grown in a medium containing 1.6% tryptone, 1.0% yeast extract and 0.5% NaCl. 'The two types of enzyme werc purified as homogeneous proteins according to the purification procedure described under Materials and Methods.

Comparison of catalytic properties of the wild-type and mutant enzymes The wild-type enzyme showed a hyperbolic saturation curve for pyruvate (Fig. 2 ) , the K , values being 1.3 mM and 2.7 mM, and the maximum reaction rates 1650 and 1400units/ mg at pH 5 and 7, respectively. These values were consistent

0

0

0

1200

0

0

0

0

0

8

0 0

0

0

= I rl

6oo

i%

Pyruvate (mM) Fig. 2. Pyruvate saturation curves of the wild-type and mutant L. plum m u m L-LDHs.Thc reaction velocity was measured at pH 5.0 (50 mM sodium acetate buffer) (0, M) or pH 7.0 (50 mM Na Mops buffer) (0, 0 ) with the wild-type (Cl,0) or mutant (m, 0 ) L. plantarurn L-LDH.

with those for the enzyme purified from the original L. pluntarum cells (data not shown). The mutant enzyme also showed a hyperbolic saturation curve for pyruvate, independently of Fru(1.6)P2, as in the case of the wild-type enzyme, and showed slightly decreased K, values of 0.7 mM and 1.1 mM, and V,,, values of 1300 and 1050 units/mg at pH 5 and 7, respectively. In the case of both the wild-type and mutant enzymes, the enzyme reaction was inhibited by a high concentration of pyruvate at pH 5 (Fig. 2). On the other hand, the K , values for NADH, which were measured at pyruvate saturation (30 mM), were essentially the same for the two types of enzymes (30 pM). The mutant enzyme, together with the wildtype enzyme, showed hyperbolic saturation curves in the pH range 5.0 - 8.4, and slightly decreased K, and V,,, values, compared to the wild-type enzyme (Fig. 3). For both the wildtype and mutant enzymes, the p K value of the catalytic His (His195) obtained from the pH profile is about 7.0, and there was no significant difference in the pH profile between the two types of enzymes. Fru(1 ,6)P2 at 5 mM induced no significant effect on the saturation curves of the two enzymes at pH 5 or 7 (data not shown). Besides Fru(l,6)P,, the activity of L. casei allosteric L-LDH is regulated by certain divalent cations such as Mn2+ [6, 71. However, 10 mM MnS04 showed no significant effect on the activity of the mutant enzyme in the assay mixture containing or not containing 5 mM Fru(l,6)PZ,as in the case of the wild-type enzyme (data not shown). On the basis of the above results, we concluded that the mutant enzyme i s a Fru(l,6)P2-independent enzyme, like the wild-type enzyme.

Thermostability of the mutant enzyme Both the wild-type and mutant enzymes were treated at various temperatures for 60 min in the absence and presence

996

100

:

h

5

u

; Av

B

10

x m

10

5

1

1:

Fructose 1.6-bisphosphate tmM)

F i g . 5 Effect of Fru(l,6)P2 on the thermostahility of the mutant L. plantarum I-LDH. Thc mutant L-LDH (0.1 mgjml) was incubated at 62‘C for 60 min in 100 mM sodium acetate pH 5.5 containing the various concentralrions of Fru(l,6)P2 indicated, and then the remaining activity was mcasured.

0

“

0.1

120

‘

‘

‘

~

‘

m

’

0.1

~

the wild-type enzyme occurred morc abruptly around 6 5 ’C than that of the mutant enzymc. In this temperature range, Fru(l,6)P2 showed a significant protective effect on the mutant enzyme but not on the wild-type enzyme, indicating that Fru(l,6)P2induces thermostabilization of the mutant enzyme. To analyze the dose effect of Fru(l,6)P2 on the thermostability of the mutant enzyme, the latter was treated at 62°C for 1 h in the presence of various concentrations of Fru(1,6)P2 (Fig. 5). The dose effect on the thermostability showed a saturation profile, a maximum being reached below 1 mM Fru(1,6)P2. It has been shown that Mn2+ also stabilizes L. casei and L. curvatus L-LDHs[6], but 10 mM MnS04 induced no significant stabilization of the Rild-type or mutant L. ylantarum L-LDH (data not shown).

7

DISCUSSION

U

60

20[ 0 30

,

, 40

,

, 50

,

,

,

60

,\k 1 70

80

Temperature (“C)

Fig. 4. Thennostability of the wikl-type and mutant L. plantarum L-LDHs. The wild-type (0, 0 ) and mutanl (n, D ) L. plunturuni T LDHs (0.1 mg/ml) wcrc incubatcd at vaiious tcmperatures for 60 min in 100 mM sodium acetate pH 5.5 (25°C) in the absence (0, 0)or presence ( 0 ,W ) of 2 mM Fru(1 ,6)Pz, and then the remaining activity was mcasured. Each point indicatcs the inean value of three separate experiments

of 2 mM Fru(l,6)P2 (Fig. 4). In the absence of Fru(l,6)P2, the activity of the mutant enzyrnc was slightly decreased, even below 50“C, while the wild-type enzyme was stable up to 60 “C. At temperatures higher than 70°C. however, the mutant enzyme was more heat-resistant than the wild-type enzyme, even in the absence of Fru(1,6)P2, since the inactivation of

His1 88 is a highly conscrved residue in all known L-LDH sequences. In the case of vertebrate non-allosteric L-LDHs, X-ray crystallography [&, 381 revealed that the His188 residue constitutes the anion-binding site and that it is involved in the binding of citratc molecules in this site which lies at the Paxis subunit interface. This structural observation strongly suggests that the occupation of‘this site with an anion may influence the stability of the subunit interaction since the vertebrate enzymes are denatured at low ionic strengths. In the bacterial allosteric L-LDHs, His188, together with the conserved Argl73 residue, constitutes thc binding site for an allosteric factor, Fru( 1,6)P2, which corresponds to the anionbinding site ofthe vertebrate enzymes [13], and plays a particularly important roles in the Fru(l,6)P2 binding and regulation of the catalytic activity of allosteric L-LDHs [13, 17, 18, 23, 241. X-ray crystallography of B. steurothcrmuphilus allosteric L-LDH [25, 261 revealed that the two His188 residues of the two Fru(l,6)P2-binding sites, which face each other across the P-axis subunit interface, undergo ionic interactions with phosphate groups of one Fru(1,6)P2molecule. It is, therefore, surprising that the conserved His1 88 residue is uniquely substituted by an Asp in L . plantarum L-LDH, though L. pluntarum L-LDH shows especially high identity to L. casei allosteric L-LDH, not only in the overall sequencc but also in the amino acid residues in the Fru(l,h)P,-binding site (Fig. 1). In the case ofallostcric L-LDHs, it has been suggested that thc occupation of the binding site by Fru(l,6)P2 induces a change in the relative position of the residues involved in

997 substrate binding and catalysis, such as Argl71 and His195 [23, 24, 261. However, the mutant L. plantarum enzyme, in which Asp188 is replaced with a His, still showed Fru(l,6)P2insensitive catalytic activity and the catalytic properties of the mutant enzyme were essentially the same as those of the wildtype enzyme except that the K, and V,,, values slightly decreased. This is not an unexpected result because His188 is a highly conserved residue even in most of the non-allosteric I,LDHs. It is likely that the catalytic site of L. plantarum LLDH constitutionally has an 'active' structure independent of the Fru( 1,6)P2-binding site and insufficient flexibility to be changed by the allosteric signal of Fru(l,6)Pz. Recently, Zuelli et al. reported that, in Bacilli L-LDHs, minor amino acid substitutions at positions 207,209B and 209C [39] or positions 29, 37, 39 and 40 [40], which are located on or near the P-axis or Q-axis subunit interface, respectively, can induce a change not only in the thcrinostability, but also in the catalytic activity and even in the allosteric properties; they indicated the existence of elements involved in the allosteric communication bctween the catalytic site and the regulator site. In Lactobacilli L-LDH, we have not identified such elements as yet, but it seems reasonable that somc are also located at the subunit interface and involved in subunit interaction, since the subunit interaction may be closely associated with the allosteric regulation of Lactobacilli L-LDHs [9]. In the case of allosteric L-LDHs, Fru(l ,6)P2usually affccts not only the catalytic activity but also the thermostability [ l , 41,421. In spitc of the Fru(l,6)P2 insensitivity of the catalytic activity, however, the mutant enzyme was significantly protected against heat-inactivation by Fru(l,6)P2; this contrasts with the wild-type enzyme, in which Fru(l,6)P, showed no significant effect on the thermostability (Fig. 4). This protective effect of Fru(1 ,6)Pz showed a saturation profile, an apparent maximum being reached at less than 1 mM Fru(1,6)Pz (Fig. 5), suggesting the high affinity and specificity of Fru(l,6)Pz to the cnzyme. These results indicate that only the single amino acid substitution, Asp+His, which results from only the single-point mutation, G+C, sufficiently induces the Fru( l,6)Pz-binding ability of L. ylmtarum L-LDH. In other words, only the His+Asp (C+G) substitution is responsible for the loss of the Fru(l,6)Pz-binding ability of L . plantarum L-LDH. The mutant enzyme showed higher thermostability than the wild-type enzyme in the presence of Fru(1,6)Pz (Fig. 4). In the absence of Fru(l,6)Pz, on the other hand, the mutant enzyme was more labile below 65°C but more stable above 70°C than the wild-type enzyme, since inactivation of the wild-type enzyme occurred abruptly at about 65 "C (Fig. 4). This abruptness indicates more cooperative heat inactivation of the wild-type enzyme than that of the mutant enzyme and suggests some change in the interactions within the protein molecule. Our results indicate that the unusual His+Asp substitution in L. planturum enzyme is not essential for Fru(l,6)Pzindependent expression of the catalytic activity of the enzyme but it contributcs to stabilization of the enzyme at a physiological temperature, independently of Fru(l,6)P2. In addition, the significant stabilization of the mutant enzyme in the presence of Fru(l,6)P2 also indicates that Fru(l,6)Pz, or certain anions as in the case of the vertebrate enzymes, can play an important role through an interaction with His188, such as stabilization of the enzyme, even in the case of non-allosteric L-LDHs. This work was supportcd by a research grant from the Ministry of Education, Sciencc and Culture of Japan.

REFERENCES Garvic. E. I. (1980) Mirrobiol. Rev. 44, 106-139. Gasser, F. & Gasscr, C. (1970) J . Bacterinl. 106, 113-125. Gasser, r. (1970) J . Gen. Microbiol. 62, 223-239. Taguchi,H.&Ohla,T.(1991)J. B i d . Chem.266.12588-12594. Gordon, C . L & Doelle, H. W. (1976) Eur. J. Biochrm. 67,543 555. 6. Hensel, R., Mayr, U., Stetter, K. 0. & Kandler, 0. (1977) Arch. Microbiol. 112, 81 -93. 7 . Hensel, R., Mayr, U., Fujiki, H. & Kandler, 0. (1977) Eur. J. Biochem. 80, 83 -92. 8. Holbrook, J. J., Liljas, A., Steindel, S. J. & Rossmann, M. G. (1975) in The enzymes, 3rd edn (Boycr, P. D., ed.) vol. 1 1 , pp. 191 -292, Academic Press, New York. 9. Mayr, U., Hensel, R. & Kandler, 0. (1980) Eur. 1.Biochem. 110, 527 - 538. 10. Mayr, U., Hensel, R., Deparade, M., Pauly, H. E., Pfleiderer, G. & Trommer, W. E. (1982) Eur. J . Biochem. 126, 549-558. 1 1 . Buehner, M. & Hecht, H.-J. (1982) J . Mol. Biol. 162, 819 -838. 12. Hensel, R., Mayr, U. & Yang, C.-Y. (1983) Eur. J . Biochem. 134; 503-511. 13. Hensel, R., Mayr, U . & Woenckhaus, C. (1983) Eur. J . Biochem. 135,359-365. 14. Taguchi, H., Matsuzawa, H. & Ohta, T. (1984) Eur. J . Riochenz. 145, 283 -290. 15. Kunai, K., Machida, M., Matsuzawa, H. & Ohta, T. (1986) Eur. J . Biochem. 134,433-440. 16. Machida, M., Yokoyama, S., Matsuzawa, H., Miyazawa. T. & Ohta, T. (1985) J . Biol. Chem. 260, 16143-16147. 3 7. Matsuzawa, H., Machida, M., Kunai, K., Ito, Y. & Ohta, T. (1988) FEBS Lett. 233, 375-378. 1 8 . Schroeder, G., Matsuzawa, H. & Ohta, T. (1988) Biochem. Biophys. Res. Commun. 152. 1236- 1241. 19. Koide, S., Yokoyama, S.: Matsuzawa, H., Miyazawa, T. & Ohta, T. (1989)J . Bid. Chem. 264, 8676-8679. 20. Wirz, B., Suter, F. & Zuber, H. (1983) Z.Physiol. Chem. 364, 893 - 909. 21. Clarke, A. R., Waldman, A. D. B., Munro, I. & Holbrook, J. J. (1985) Biochim. Biophys. Acta 828, 375-379. 22. Clarke. A. R., Atkinson, T., Campbell, J. W. & Holbrook, J. J. (1985) Biochim. Biophys. Acta 829, 387- 396. 23. Clarke, A. R., Evington, J. R. N., Dunn; C . R., Atkinson, T. & Holbrook, J. J. (1986) Biochinz. Rioiphys. Arta 870, 112- 126. 24. Clarke, A. R., Wigley, D. B., Barstow, D. A , , Chia. W. N., Atkinson, T. & Holbrook. J. J. (1987) Biochim. Biophys. Acta 913, 72-80. 25. Piontek, P., Chakrabarti, P., Schaer, H.-P., Rossmann, M. G. & Zuber, H. (1990) Proteins: Structure, Function, and Genetics 7; 74-92. 26. Wigley, D. B., Gabblin, S. J., Turkenhurg, J. P., Dobson, E. J., Piontek, K., Muirhead, H. & Holbrook, J. J. (1992) J. Mol. B i d . 223, 3 1I - 335. 27. Iwata, S., Minowa, T., Sakai, H. & Ohta, T. (1989) Agric. Bid. Chem. 53, 3365 - 3366. 28. Minowa, T., Iwata. S., Sakai, H., Masaki, H. & Ohta, T. (1989) Gene XlY5,161 - 168. 29. Iwata, S., Minowa, T., Mikami, B.: Morita, Y. &Ohta, T. (1989) J . Biochem. (Tokyo) 106, 558-559. 30. Smith, P. K., Krohn, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. TI., Provenzano, M. D., Eujimoto, E. K., Goeke, N. M., Olson, B. J . & Klenk, D. C. (1985) Anal. Biochem. 150. 76 - 85. 31. Laemmli, U. K . (1970) Nature 227, 680-685. 32. Kunkcl, T. A. (1985) Proc. Natl Acad. Sci. U S A 82, 488-492. 33. Sanger, F., Nicklen, S. & Coulson. A. R. (1977)Proc. Natl Acad. Sci. U S A 74, 5463 - 5467. 34. Barslow, D. A., Clarke, A. R., Chia, W. N., Wigley, D., Sharman, A. F., Holbrook, J. J., Atkinson, T. & Minton, N. P. (1986) Gene 46,47-55. 35. Zuelli, F.. Weber. H. 8~ Zuher, H. (1987) B i d . C h m . HoppcSeyler 368, 1 1 67 - 1177. 1. 2. 3. 4. 5.

36. Taylor, S. S. (1Y77) J . Biol. Chein. 252, 1799-1806. 37. Ono, M.; Matsuzawa, H. & Ohta, T. (1990) J. Biochem. (Tokyo) 107,21-26. 38. Adams, M. J., Liljias, A & Rossmann, M. G. (1973) J. Mol. Biol. 76, 519-531. 39. Zuelli, F., Weber, H. & Zuber, H. (1990) Biol. Chern. HoppeSeyler 371,655-662. 40. Zuelli, F., Schneiter, R., Urfer, R . &Tuber, H. (1991) Biol. Chem. Hoppe-Seyler 371,655-662.

41. Goets, F. & Schleifer, K. H. (1975) Arch. Microbiol. 10.5, 303312. 42. Taguchi, H., Yamashita, M., Matsuzawa, H. & Ohta, T. (1982) J . Biochem. (Tokyo) 91, 1343-1348. 43. Eventoff, W., Rossmann, M. G., Taylor, S. S., Torff, H.-J., Meyrer, H., Kcil. W. & Kiltz. H.-H. (1977) Proc. Natl Acad. Sci. USA 74,2677 - 2681.