Specificities of extracellular and ribosomal serine proteinases from Bacillus natto, a food microorganism

Biochimica et Biophvsica Acta 869 (1986) 178-184 Elsevier

178

BBA 32409

S p e c i f i c i t i e s o f e x t r a c e l l u l a r and r i b o s o m a l s e r i n e p r o t e i n a s e s f r o m Bacillus

natto, a f o o d m i c r o o r g a n i s m Eiji I c h i s h i m a , Y u k i h i r o T a k a d a *, K e i j i r o T a i r a ** a n d M i c h i o T a k e u c h i Laboratory of Enzyrnolog)' and Microbial Chemistry, Tokyo Noko Universi(v, Fuchu, Tol,To 183 (Japan) (Received July 30th, 1985)

Key words: Serine proteinase; Ribosomal proteinase; Proteinase specificity: ( Bacillus )

The specificities of extracellular and ribosomal serine proteinase from Bacillus natto, a food microorganism, were investigated. Both proteins have highly restricted and characteristic specificities. With the extraceilular serine proteinase, initial cleavage site was observed at Leu~5-Tyr t6, secondary site at Ser9-His I° and additional cleavage sites at GIn4-His 5 and HisS-Leu 6 in the oxidized insulin B-chain. Hydrolysis of proangiotensin with the extracellular serine proteinase was observed primarily at PheS-His 9 and secondary at Tyr4-Ile 5. The extracellular serine proteinase has a K., of 0.08 mM and kca t of 3 S-I for angiotensin hydrolysis. With the ribosomal proteinase, initial cleavage site of the oxidized insulin B-chain was observed at LeutS-Tyr t6 and additional cleavage site at Phez4-Phe zS. Hydrolysis of proangiotensin was observed at Tyr4-Ile 5 bond with the ribosomal proteinase.

Introduction Natto (Itohikinatto) is a typical, popular and economical fermented soybean food in Japanese diets [1]. For about a 1000 years Japanese people have been familiar with handling microorganisms through the production of such traditional food as natto. To make natto, soybean is cooked and on its surface Bacillus natto is grown, characteristic enzymes, tastes and flavours being produced. When fermented by B. natto, the surface of soybean becomes covered with a characteristic mucin [2] consisting of 58% ,/-polyglutamic acid and 40% polysaccharide. Natto contains a significant number of native enzymes, especially alkaline proteinase [3] and aamylase [4], which are important hydrolyzing en* Present address: Technical Research Laboratories, Snow Brand Milk Co. Ltd., Kawagoe-shi, Saitama-ken 350, Japan. ** Present address: Central Research Laboratories, Sanraku Co. Ltd., Fujisawa-shi, Kanagawa-ken 251, Japan.

zymes in fermented food processing. The alkaline proteinase is an important hydrolyzing enzyme in natto fermentation. This is a diisopropylfluorophosphate-sensitive enzyme with a similarity to subtilisins Carlsberg and Novo [5] (EC 3.4.21.14) in enzymatic activity and physicochemical properties. In the present study, we obtained purified preparations of extracellular and ribosomal serine proteinases from B. natto by successive ion-exchange chromatographies and ultracentrifugation, respectively. The specificities of extracellular and ribosomal serine proteinases were investigated, therefore, using the oxidized insulin B-chain and proangiotensin. These results are discussed in relation to the bond specificities of the other serine proteinases. Materials and Methods Materials. Crystalline bovine insulin (Lot 57950) was purchased from Fluka AG. Buchs SG,

0167-4838/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)

179

Switzerland. Human proangiotensin (formerly designated angiotensin I) and angiotensin (formerly designated angiotensin II) were supplied by the Protein Research Foundation, Osaka, Japan. DEAE-Sephadex A-50 and CM-Sephadex C-50 were obtained from Pharmacia (Uppsala, Sweden). Acrylamide and bisacrylamide were from Wako Junyaku Co. Phenylmethanesulfonyl fluoride (PMSF) and diisopropyl fluorophosphate (DFP) were from Sigma, St. Louis, MO, U.S.A. Microorganisms. Wild-type strain of B. natto NC 2-1 was isolated from the commercial fermented soybean food, natto [1]. The temperaturesensitive mutant ts 25 strain of B. natto was used in this experiment. The ts 25 cells grew normally at 30°C, but did not grow at 43°C, The organism was maintained as stock cultures on a medium containing 1% beef extract, 1% polypeptone and 0.5% NaCi at pH 7.2, according to a previous paper [1]. The culture was performed by incubation for 12 h at 30°C. Cultivation. Shaken flasks (500 ml) containing 100 ml of the above described medium were inoculated after sterilization at 115°C for 15 min, and were then incubated on a reciprocal shaker (130 rev/min, 7 cm stroke) at 30°C. After 24 h, the culture broth described above was centrifuged at 13600 x g for 10 rain with a Hitachi model RP centrifuge. A crude proteinase solution was dialyzed against 0.05 M phosphate buffer (pH 7.5) and then stored in the cold. Purification of extracellular serine proteinase. A crude proteinase solution was prepared from the supernatant fluid by salting out (NH4)2SO 4 (0.85 saturation) and dialyzing against 0.01 M Tris-HC1 buffer (pH 7.5) containing 2 mM calcium acetate. Purification of extracellular serine proteinase was performed on successive chromatographic procedures with DEAE-Sephadex A-50 (2 × 50 cm) and CM-Sephadex C-50 (1 x 30 cm) by a method reported previously [6]. Preparation of diisopropylphosphoryl-proteinase. The diisopropylphosphoryl derivative of the extracellular serine proteinase was prepared according to the method of Matsubara et al. [7]. Ribosomes. Ribosomes from B. natto NC 2-1 ts 25 were prepared according to the method by Fortnagel et al. [8]. The ribosomal pellets were carefully rinsed with 10 mM imidazole buffer (pH

7.7) and resuspended. The suspension was adjusted to an absorbance at 260 nm of 1.00, and 0.3 ml of this ribosomal suspension was applied to a linear gradient of sucrose, from 10 to 30%. Each fraction obtained was centrifuged for 4 h at 150000 x g at 4°C with a Hitachi 65 P automatic preparative ultracentrifuge. The 70 S peak fraction thus obtained was quickly frozen and stored at - 7 0 ° C . Assay of extracellular serine proteinase activity. Proteinase activity at pH 7.5 was assayed by measuring the absorbance at 280 nm liberated from a casein solution as previously described [6]. 2 ml of 0.638% casein was digested with 0.5 ml of enzyme solution at 30°C for 30 min at pH 7.5. The absorbance at 280 nm was determined. One katal (kat) of the extracellular serine proteinase is defined as that amount of enzyme which yields the ultraviolet absorbance equivalent to 1 mol of tyrosine per s, using casein as a substrate at pH 7.5 and 30°C, according to the enzyme nomenclature of Florkin and Stotz [9] and the IUPAC and IUB recomendations [10]. Heat treatment of extracellular serine proteinase. After the sample from a wild-type or a temperature-sensitive mutant strain dissolved in 0.05 M phosphate buffer (pH 7.5) had been best for a specified time at 60, 65 or 70°C, it was transferred to a test-tube at 0°C to stop the reaction. The proteinase activity was then assayed by the method described above [6]. Assay of ribosomal serine proteinase. Proteinase activity was determined in Tris-HC1 buffer at pH 7.0-8.0 by a solid-phase assay according to the ninhydrin method reported previously [11], using 0.1 t~mol oxidized insulin B-chain as a substrate. One proteinase unit (katal) of the ribosomal serine proteinase was defined as that amount of enzyme which released peptides equivalent to 1 tool of tyrosine at 570 nm by the ninhydrin method per s at 30°C and pH 7.5. Hydrolysis of oxidized insulin B-chain. The oxidized insulin B-chain (0.02 gmmol, 0.7 mg) was dissolved in 1 ml distilled water and the pH was adjusted to 7,5 with 1 M NH4OH; 0.1 nkat of the extracellular serine proteinase from B. natto or 0.1 kat of the ribosomal proteinase was added to the solution. The mixture was incubated at 30°C for 5, 20 or 30 min. One drop of concentrated HCI was added to inactivate the enzyme. The samples of

180 hydrolysate were stored at - 2 0 ° C , and then the frosen digest was lyophilized. Hydrolysis of proangiotensin. Proangiotensin (0.2 btmol, 0.25 mg) was dissolved in 1 ml of distilled water adjusted to pH 7.5 with 1 M N H 4 O H and incubated at 30°C with 0.1 nkat of the extracellular serine proteinase for 5 and 20 min, or 0.1 kat of the ribosomal proteinase for 12 h. Separation and identification of the peptide. Separation and identification of the peptide in the freeze-dried digest were performed by high-performance liquid chromatography (HPLC) as described in a previous paper [12]. HPLC was performed on a column (4.0 × 250 ram) of Zorbax BP-ODS (Dupon) equiped with a Shimadzu model LC-3A delivery system. Chromatographic recording was performed at 220 nm with a Shimadzu SPD-2A spectrophotometer. Peptides isolated were dissolved in 2 ml of 6 M HC1 and hydrolyzed at l l 0 ° C for 24 h. Amino acids in the hydrolyzates were determined with a Hitachi amino acid analyzer, model 834-30. Determination of kinetic parameter. The values of K m and kcat were graphically determined according to the Lineweaver-Burk and EadieHofstee plots. Initial rates towards angiotensin were determined at the concentrations from 0.05 mM to 1 mM at pH 7.5. Results

Extracellular serine proteinase The residual activities of extracellular serine proteinase from the ts 25 strain at 60°C for 20 rain, 65°C for 10 min and 70°C for 10 rain in 0.05 M phosphate buffer (pH 7.5) were 20%, 24% and 3%, respectively. In comparing the extracellular serine proteinase from the wild-type strain, B. natto NC 2-1, the experiments showed that the heat stability of the enzyme from the ts 25 strain was similar to that of the enzyme from the wildtype strain. The specific activities of highly purified preparation of extracellular serine proteinase from the ts 25 strain were 0.151 k a t / k g of enzyme at pH 7.5, and 0.267 k a t / k g of enzyme at pH 10.0, respectively. The specific activity of the wild-type strain NC 2-1 was 0.160 k a t / k g of enzyme at pH 7.5. The proteinase activity of 0.1 nkat extracellular

proteinase was completely inhibited by 1 mM phenylmethanesulfonyl fluoride (PMSF), 0.8 mM diisoopropyl fluorophosphate (DFP) or 1 mM benzyloxycarbonyl-L-phenylalanyl chloromethylketone (ZPCK), respectively. The diisopropylphosphoryl proteinase migrates as a single band on disc gel electrophoresis at pH 9.4. However, the highly purified preparation could not migrate on disc gel electrophoreses at pH 9.4 and pH 2.3. The hydrolysis of oxidized insulin B-chain by the extracellular serine proteinase was examined. The two peptides, Phe~-Leu ~5 and Tyrl~-Ala~, obtained after 5 rain incubation were isolated and purified as described. The five peptides obtained after 20 min incubation were isolated and purified. Seven peptides were obtained after 30 min incubation, All of the peptides were readily identifiable from the known sequence of the oxidized insulin B-chain. The results in Fig. 1 show that the extracellular serine proteinase was hydrolyzed primarily at the peptide bond in the oxidized insulin B-chain, the LeulS-Tyr ~6, secondarily hydrolyzed at the other bond Ser~-His ~°, and additionally at two others (Gln4-His 5, and HisS-Leu6). The extracellular serine proteinase could not hydrolyze the C-terminal peptide, Tyr~6-Ala3", Similar specificity was observed for the extracellular proteinase from the wild-type strain, B. natto NC 2-1. When proangiotensin was digested with the extracellular serine proteinase for 5 min, the products could be resolved into two peaks by HPLC. After 20 rain incubation, the peak of proangiotensin was completely disappeared and four peptides were isolated and purified. The peptide bond primarily hydrolyzed by the extracellular serine proteinase was the Phe~-His 9 bond and the peptide bond secondarily hydrolyzed was the Tyr4-lle 5 bond, as shown in Fig. 2. When angiotensin was digested with the extracellular serine proteinase, the enzyme was hydrolyzed at one peptide bond at Tyr4-11e 5. A similar specificity was observed by the extracellular proteinase from the wild-type strain, B. natto NC 2-1. Kinetic parameters of the extracellular serine proteinases are summarized in Table I. The parameters obtained from Lineweaver-Burk plots were similar to those obtained from Eadie-Hofstee plots.

181 1

5

FVNQ

HLC+GS

10 HLV

15 E AL

1

Vlll(3&5)

t

2

I, ,~

t

IV ' ( 2 L ~ )

V111(38.5)

11

~ _ !n_(1_7.6_) ,. < . . . . . . . iv (2~.8)

25 30 FY TPKA

IX ( 3 9 . 4 )

" I1,11

IX ( 3 9 . 4 )

t

t

3

ERGF

t

v (3o.o)

w~ ~ 36 o)_

"

20 YLVC*G

_v L (_3_5.j) _ _ - > v(3o.o)

IX ( 3 9 . 4 )

Fig. 1. Summary of the specificity of the extracellular serine proteinase from B. natto NC 2-1 ts 25 towards the oxidized insulin B-chain at pH 7.5 and 30°C. The enzymatic digest of the chains was directly separated as described in the text. Abbreviations of amino acids follow the alphabetical system. C* indicates cysteine sulfonic acid. The sign.(, ~ ) shows the peptides recovered. Perpendicular arrows ( t , T, t ) indicate the bond split, the degree of hydrolysis being follows: f > I" > t . The values in parentheses denote the retention times of the peptides obtained on HPLC: (1) 5 min; (2) 20 min; (3) 30 min digestion.

Ribosomal serine proteinase

The proteinase activity associated with the 70 S ribosomes from B. natto NC 2-1 ts 25 was characterized, and its optimal pH was 7.5 with oxidized insulin B-chain. The proteinase activity of 0.1 kat was completely inhibited by 1 mM PMSF. The enzyme activity was also inhibited by 0.8 mM diisopropyl fluorophosphate and 1 mM ZPCK. p-Chloromercuribenzoate, monoiodoacetic acid, leupeptin, chymostatin and pepstatin had no effect on its activity. The hydrolysis of oxidized insulin B-chain by 0.1 kat of ribosomal serine proteinase in 6 h was

D

•9

R

V

Y

I

IV(186)

H

P

F

H

L

l I(58) _--- = z-

= [~(64).I=mc~62) .I IV (186)

i (.~8)

Fig. 2. Summary of the specificity of the extracellular serine proteinase from B. natto NC 2-1 ts 25 towards the proangiotensin at pH 7.5 and 30°C. The sign ( ~, ) shows the peptide recovered. Perpendicular arrows (T) indicate the bond split. The values in parentheses denote the retention times of peptides obtained on HPLC: (A) 5 min; (B) 20 min digestion.

examined. Four peptides were obtained on HPLC. The retention times of Phe2LAla 3°, Tyr~6-Phe z4` Phel-Leu 15 and Tyr~6-Ala 3° in the enzymatic hydrolysate were 26.0, 31.1, 38.0 and 38.6 min, respectively. The recoveries of Phe2LAla 3° and Tyr16-Phe24 were lower than those of Phel-Leu 15 and Tyr16-Ala3°. The initial cleavage site of the ribosomal serine proteinase on the oxidized insulin B-chain was observed at the Leu~LTyr 16 bond, and an additional cleavage site was the Phe24-Phe 25 bond. The present studies indicate that the secondary cleavage site of oxidized insulin B-chain at Phe 24Phe 25 by the ribosomal proteinase was different from the specificities of the extracellular serine proteinase from B. natto and subtilisins Carisberg and Novo. Further HPLC experiments after 24 h digestion of oxidized insulin B-chain by ribosomal serine proteinase treated previously with 1 mM EDTA showed that four similar peptides were produced. A similar specificity was observed with the ribosomal proteinase from the wild-type strain of B. natto. When proangiotensin was incubated at pH 7.5 for 12 h at 30°C, with the 70 S ribosomal fraction, small peaks of three peptides, A s p l - T y r 4, Aspl-Phe 8 and IleS-Leu 1°, were obtained on HPLC. The clear cleavage site of proangiotensin

182

was observed at Tyr4-Ile 5. Similar specificity was observed with the ribosomal proteinase from the wild-type strain of B. natto NC 2-1. Discussion

A summary of the specificities of various microbial serine proteinases (EC 3.4.21.14) is shown in Fig. 3. The highly restricted specificity in the oxidized insulin B-chain with the extracellular serine proteinase from B. natto is much narrower than those previously reported from subtilisins Carlsberg and Novo [13], the alkaline proteinase from thermophilic Streptomyces rectus [14], '~ F V N Q

1

5 kt

I:D 15 20 L C" G S H L V £ A L Y L V C G E

t

~

t t t t t ttt t t t t

t ~

~

~

t t ~

30 P K A

t

t t tttt

~o

2~ R G F F Y T

t

2

7

Streptomyces fradie proteinase type III [15], Streptomyces griseus alkaline proteinase C [16], Streptomyces apkalophilic keratinase [17], Saccharomyces cerevisiae proteinase B [18] and Aspergillus sojae alkaline proteinase I [19]. The initial cleavage site at Leu~5-Tyr ~6 in the oxidized insulin B-chain was observed with the extracellular serine proteinase. The extracellular serine proteinase could not hydrolyze the C-terminal peptide, Tyr16-Ala 3°. We could not explain the reason for this from the amino acid sequence of primary structure of the substrate. All of the microbial serine proteinases in Fig. 3 have a similar specificity of the initial or major cleavage site at

f t

'

t

t t

tt

ft ~t~

t ttt ~

t

tff tftf tf

tt

ff ttt tf

Fig. 3. Summary of the specificity of various microbial alkaline proteinases from Bacillus, Strepton~vces. Saccharomw'es and Aspergillus towards oxidized insulin B-chain. (1) The extracellular serine proteinase from Bacill~ nato, 30 min incubation at pH 7;5 and 30°C. (2) The ribosomal serine proteinase from Bacillis natto, 6 h incubation at pH 7.5 and 30°C. (3) Subtilisin Carlsberg from Bacillus subtilis 2 h incubation at p H 8.0 and 30°C [13]. (4) Subtilisin Novo from Bacillus amyloliquefaciens, 2 h incubation at pH 8.0 and 30°C [13]. (5) Streptomyces rectus vat. proteolyticus alkaline proteinase B, 4-h incubation at pH 8.0 and 30°C [14]. (6) Streptornycesfradie proteinase type III, at pH 9 and room temperature [15].

(7) Streptomyces griseus alkaline proteinase C, 2 h incubation at pH 9.2 and 30°C [16]. reptomyces alkalopholic keratinase, 24 h incubation at pH 11.6 and 37°C [17] (9) Saccharomyces cerevisiae proteinase B, 6 h incubation at pH 7.2 and 25°C [18]. (10) Aspergillus sojae alkaline proteinase, 90 min incubation at pH 7.5 and 30°C [19]. (11) Aspergillopeptidase B from Aspergillus oryzae, 30 min incubation at pH 10.2 and 0°C [20].

183 LeulS-Tyr 16 in the oxidized insulin B-chain. The cleavage sites at Glna-His 5, Serg-His H~ and LeulS-Tyr ~6 in the oxidized insulin B-chain by the extracellular serine proteinase were identical to those observed in the studies on subtilisins Carlsberg and Novo [13]. However, the extracellular serine proteinase could not split the Leu~-Va112, Tyr16-Leu17, LeulV-Val 1~ and Phe25-Thr 26 bonds in the oxidized insulin B-chain, which are known cleavage sites with subtilisins Carlsberg and Novo [13]. The single serine and single histidine residues of subtilisins are essential for catalysis. Inhibition of the extracellular serine proteinase by idiisopropyl fluorophosphate and ZPCK indicates that the active sites of the extracellular serine proteinase are similar to those of the subtilisins. It may be concluded that the specificity of the extracellular serine proteinase from B. natto is different from those of subtilisins Carlsberg and Novo. All of the proteinase in Fig. 3, except for the extracellular and ribosomal proteinases from B. natto and Streptomyces alkalophilic keratinase, hydrolyzed the Leu It-Val ~2 bond. The sites of cleavage at Gln4-His 5, HisS-Leu6, LeulS-Tyr16 in the oxidized insulin B-chain by the extracellular serine proteinase were identical with those in the work with the alkaline proteinase from thermophilic S. rectus var proteolyticus [14]. However, the extracellular serine proteinase could not split the Glut3-Ala14, LeulT-Va118 and Phe24-Phe 25 bonds in the oxidized insulin B-chain, which are known cleavage sites with S. rectus alkaline proteinase [14]. S. rectus alkaline proteinase possesses a single active serine residue and a active thiol group [14]. Similar observations are reported with S. cerevisiae proteinase B [18]. When proangiotensin was digested for 5 or 20 min with the extracellular serine proteinase, the primary hydrolysis of the peptide bond at Phe sHis 9 and the secondary hydrolysis of the bond at Tyr4-Ile 5 were observed. When proangiotensin was digested with S. rectus alkaline proteinase for 1 h, the products could be resolved into equal amounts of three peptides, Aspl-Tyr 4, IleS-Phe 8 and Hisg-Leu 1°. These results suggest that the specificity of the extracellular serine proteinase is different from that of S. rectus alkaline proteinase [14]. Determination of proteolytic activity in the

ribosomes was complicated by the fact that [1] the activity was very low; and milk casein was not split by the ribosomes. Therefore, we used oxidized insulin B-chain as substrate for the proteinase associated with ribosomes. The proteinase associated with the 70 S ribosomes from B. natto was characterized. The inhibition experiments suggested the involvement of active serine and histidine groups in the enzyme activity. The major cleavage site at LeulS-Tyr t6 in the oxidized insulin B-chain with the ribosomal serine proteinase was similar to that of the extraceUular serine proteinase. With the ribosomal serine proteinase, an additional cleavage site, Phe24-Phe 25, was observed, which could not split by the extracellular serine proteinase. However, various alkaline proteinases from Streptomyces [14-17[ could split the Phe24-Phe 25 bond in the oxidized insulin B-chain. In comparison with the extracellular serine proteinase and subtilisins Carlsberg and Novo [13], the ribosomal proteinase showed a quite unique specificity towards the oxidized insulin B-chain and proangiotensin. In the work by Orekhovich et al. [21] the broad specificity of the neutral proteinase (EC 3.4.99.33, cathepsin R) from rat liver ribosomes towards insulin B-chain was investigated. However, the presence of contaminating proteinase activity in the ribosomal preparation from rat liver still remained a possibility.

Acknowledgments This research was supported in part by a grant from the Research Council, Ministry of Agriculture, Forestry and Fisheries of Japan, for research projects on biotechnology and a grant from Ajinomoto Co. Ltd.

References 1 Ichishima, E., Kato, M., Wada, Y., Takeuchi, M., Takahashi, T., Takinarni, K. and Hirose, Y. (1982) Food Chem. 8, 1-9 2 Saito, T., Iso, N., Mizuno, H., Kaneda, H., Suyama, H., Kawamura, S. and Osawa, S. (1974) Agric. Biol. Chem. 38, 1941-1946 3 Yoshimoto,T., Fukumoto, J. and Tsuru, D. (1971) Int. J. Protein Res. 3, 285-295 4 Yamaguchi, K., Matsuzaki, H. and Maruo, B. (1969) J, Gen. Appl. Microbiol. 15, 97-107

184 50ttesen, M., and Svendsen, I. (1970) Methods Enzymol. 19, 199-215 6 Uehara, H,, Yoneda, Y., Yamane, K. and Maruo, B. (1974) ,I. Bacteriol. 119, 82 91 7 Matsubara, H., Kasper, C.B., Brown, D.M. and Smith, E.L. (1965) J. Biol. Chem. 240, 1125-1130 8 Fortnagel, P., Bergmann, R.. Hafemann. B, and Legelsen, C. (1975) in Spores (Gerhardt, P., Costilow, R.N. and Sadoff, H.L., eds.), Vol. 6, pp. 301 306. Am. Soc. Microbiol., Washington, DC 9 Florkin, M. and Stotz, E.H. (1973) in Comprehensive Biochemistry (Florkin, M. and Stotz, E.H., eds.), Vol. 13, pp. 26-27, Elsevier, Amsterdam 10 Enzyme Commission, IUB (1978) Enzyme Nomenclature, Academic Press, New York 11 lchishima. E. (1972) Biochim. Biophys. Acta 258, 274-288 12 Ichishima, E., Maeba, H., Amikura, T. and Sakata, H. (1984) Biochim. Biophys. Acta 786, 25 31

13 Johansen, J.T., Ottesen, M., Svendsen, I. and Wybrandt, G. (1969) C.R, Trav. Lay. Carlsberg 36, 365-384 14 Matsue, M., Majima, E. and Ichishima, E, (1982) Agric. Biol. Chem. 46, 2485 2490 15 Morihara, K. and Tsuchiya, H. (1969) Arch. Biochem. Biophys. 129, 620 634 16 Narahashi, Y. and Yoda, K. (1973) J. Biochem. 73, 831-841 17 Nakanishi, T. and Yamamoto, T. (1974) Agric. Biol. Chem. 38, 239l-2397 18 Kominami, E., Hoffschute, H.. Leuschel, L., Maier, K. and Holzer, H. (1981) Biochim. Biophys. Acta 661, 136-141 19 Ichishima, E., Hamamatsu, M. and Yamamoto, N. (1983) Food Chem. 11. 187-200 20 Spadari, S., Subramanian, A.R, and Kalnitsky, G. (1974) Biochim. Biophys. Acta 359, 267 272 21 Le,~jant, M.I., Bylinkina, V,S., Spivak, V.A. and Orekhovich, V.N. (1978) Biokhimiya43, 1423 1428177 178 179