Methionine Biosynthesis in Brevibacterium ffavum: Properties and

Transcription

1 J. Biochem. 91, (1982) Methionine Biosynthesis in Brevibacterium ffavum: Properties and Essential Role of O-Acetylhomoserine Sulfhydrylase Hachiro OZAKI and Isamu SHIM Central Research Laboratories, Ajinomoto Co., Inc., Kawasaki-ku, Kawasaki, Kanagawa 210 Received for publication, August 20, 1981 Out of 27 strains of methionine auxotrophs of Brevibaclerium flavum, 14 strains did not grow on homoserine but grew on O-acetylhomoserine, and all were found to lack homoserine O-acetyltransferase [EC ] alone. Another 3 strains did not grow on O-acetylhomoserine but grew on homocysteine, and the two strains tested were found to lack O-acetylhomoserine sulfhydrylase (AHS) alone, without any changes in the activities of cystathionine ć-synthase [EC ] and Ĉ-cystathionase [EC ]. Prototrophic revertants of the AHS-lacking mutants showed concomitant reversion of AHS activity. None of the methionine auxotrophs grew on cystathionine. From these results it was concluded that the methionine biosyn thetic pathway of this bacterium involves formation of O-acetylhomoserine from homoserine by the action of homoserine O-acetyltransferase, and direct formation of homocysteine from O-acetylhomoserine by the AHS reaction. AHS synthesis was strongly repressed by methionine. AHS was purified to 70% purity. The purified preparation was activated by pyridoxal phosphate after treatment with hydroxylamine. The enzyme showed a molecular weight of 360,000, an optimum ph of 8.7 for activity, and specifically reacted with O-acetyl-L-homoserine and showed with O-acetyl-L-serine one hundredth as much activity as that with O-acetylhomoserine, but did not show activity with O-succinyl-L-homoserine, homoserine, or serine. The Km values for O-acetylhomoserine and H2S were 2.0 mm and 0.08 mm, respectively. The enzyme was inhibited 50, 23, and 29% by 10 mm L-methionine, L-homoserine, and O-acetyl-L-serine, respectively, but it was not inhibited by cystathionine or S-adenosyl-L-methionine. The first enzyme in the methionine biosynthetic pathway of Brevibacterium flavum has been found to be homoserine O-acetyltransferase which is inhibited by methionine and S-adenosylmethionine, Abbreviations: AHS, O-acetylhomoserine sulfhydry lase; HAT, homoserine O-acetyltransferase; ASS, 0- acetylserine sulfhydrylase. and repressed by methionine (1, 2). There are two possible methionine biosynthetic pathways from O-acylhomoserine; one is through cystathio nine formation and the other is ria direct formation of homocysteine from O-acylhomoserine. The cystathionine-pathway has been reported to be the main pathway in enteric bacteria and Neurospora (3). In yeast, there are two alternative pathways Vol. 91, No. 4,

2 1164 H. OZAKI and I. SHIM for homocysteine biosynthesis, but which of them is the major physiological pathway remains to be determined (3, 4). Therefore, in order to determine the methionine biosynthetic pathway of B. flavum, various methionine auxotrophs were derived from the bacterium, and growth responses on metabolic intermediates and enzyme activities of the auxotrophs were determined. The results showed the pathway via direct formation of homocysteine by O-acetylhomoserine sulfhydrylase (AHS) to be the sole methionine biosynthetic pathway in B. flavum. The present paper deals with the above results and some properties of a partially purified preparation of AHS. MATERIALS AND METHODS Bacterial Strains and Culture Media-Homoserine auxotroph H 1013 (5), methionine and threonine double auxotrophs, TM-12 (6) and TM- 15 (5), and methionine auxotrophs, M-78 (5) and M-116 (2) were all derived from Brevibacterium forum No (ATCC 14067). No (7) was a reisolated strain from citrate synthase rever tant No. 15 derived from No S-(2-Aminoethyl)-L-cysteine (AEC) resistant strain No (7) was derived from No ƒ -Amino-ƒÀ-hydroxyvaleric acid (AHV) resistant-proline-isoleucine-methionine-triple auxotroph ƒà-im-4 (8) was derived from Escherichia coli K-12. The composition of medium 5 (9) as well as those of medium 7, medium 10, and medium 13 (10) were described in the previous papers. Derivation of Methionine Auxotrophs-The method for derivation was essentially the same as described previously (7). After treatment of strains No and No with 750 ƒêg/ml of N-methyl-N Œ-nitro-N-nitrosoguanidine, colonies which did not grow on medium 10 (minimum glucose medium) but grew on medium 10 supple mented with 200 ƒêg/ml of Lmethionine were picked up. Preparation of Crude Enzyme-Enzyme extracts in 50 mm potassium phosphate buffer, ph 7.3, containing 0.05 mm pyridoxal phosphate and 0.1 mm ƒà-mercaptopropionic acid (buffer A), were prepared as described previously (1) by sonic treatment and then gel-filtered through a column of Sephadex G-50 using buffer A to remove substances of low molecular weight, before they were used as the crude enzyme solution. Purification of AHS-An enzyme extract was prepared by the method above from 14 g (wet weight) of H1013 cells grown on medium 5 sup plemented with 20 mg of Lmethionine plus 500 mg of Lthreonine per liter, and fractionated with ammonium sulfate. The precipitate obtained be tween 0.4 and 0.55 saturation was dissolved in a small volume of buffer A (8.9 ml, final volume), and gel-filtered through a column of Sephadex G-200 using buffer A. Fractions showing AHS activity were pooled (38 ml) and placed on a DEAE-cellulose (Midorijuji Co.) column (2.5 x 17 cm) previously equilibrated with 50 mm Tris-HCI buffer, ph 7.5, containing 0.1 mm ƒà-mercaptopropionic acid, 0.05 mm pyridoxal phosphate, and 0.2 M NaCl (buffer B). After the column was washed with 100 ml of buffer B, the enzyme was eluted with a concentration gradient of NaCl formed from 400 ml of buffer B and 400 ml of buffer B containing 0.5 M NaCI (final concentration). Fractions showing AHS activity were pooled, and dialyzed overnight against a 50-fold volume of 10 mm N-tris(hydroxymethyl)methyl-2- aminoethane sulfonic acid (TES)-NaOH buffer, ph 7.5, containing 0.1 mm ƒà-mercaptopropionic acid and 0.05 mm pyridoxal phosphate (buffer C). The dialyzed enzyme (40 ml) was placed on a hydroxylapatite column (BIO-GEL HT; Bio-Rad Laboratories) (2.5 x 17) previously equilibrated with buffer C. After the column was washed with 80 ml of buffer C, the enzyme was eluted with a concentration gradient of potassium phosphate formed from 400 ml of buffer C and 400 ml of buffer C containing 0.2 M potassium phosphate. Fractions showing AHS activity were pooled, and mixed with solid ammonium sulfate to give 90% saturation. The precipitate formed was dissolved in a small volume of buffer A and used as the partially purified enzyme. En yme Assav-Homoserine O-acetyltransferase (HAT) was assayed by the same method as reported previously (1). The decrease in absorb ance at 232 nm on consumption of actyl-coa and a molecular extinction coefficient of 4,500 (11) were used for calculation of the specific activity. Cystathionine ƒá-synthase was assayed by the twostep-assay method of Kaplan and Guggenheim (12) except that O-acetyl-L-homoserine was used instead of O-succinyl-DL-homoserine. ƒà-cysta- J. Biochem.

3 O-ACETYLHOMOSERINE SULFHYDRYLASE OF B. fin rum 1165 thionase was assayed by the DTNB-method of Kaplan and Guggenheim (12). ƒá-cystathionase activity was confirmed to be absent in the crude enzyme solution of B. flavum by the following procedure. L-Cystathionine was incubated with the crude enzyme solution under the same condi tions as for the ƒà-cystathionase assay for 30 min. Then the keto acid formed was converted to the 2,4-dinitrophenylhydrazone, and paper-chromatographed. The results showed the presence of pyruvate, a product of the ƒà-cystathionase reac tion, but not of ƒ -ketobutyrate, a product of the ƒá-cystathionase reaction. O-Acetylhomoserine sulfhydrylase (AHS) was assyed by the colorimetric method of Kredich and Becker (13). However, since the method gave a high absorbance at 540 nm for the blank, the method was slightly modified as follows. In order to decrease the blank value the reaction mixture was incubated for 10 min after addition of ammo nium sulfamate instead of 2-min incubation. The standard reaction mixture (0.2 ml) was composed of 150 mm Tris-HCl buffer, ph 7.4, 0.77 mm EDTA, 3 mm Na2S, 10 mm O-acetyl-L-homoserine, and enzyme, and incubated at room temperature (20-22 Ž) for 4 min. The reaction was started by adding Na2S dissolved in Tris-HCl buffer, ph 7.4, which were prepared just before use. For the blank experiment, O-acetylhomoserine was omitted from the standard reaction mixture. O-Acetylserine sulfhydrylase [EC ] (ASS) was assayed by the same method as for AHS except that O-acetyl-L-serine was used instead of O-acetyl-Lhomoserine. Chemicals-Na2S was a product of Junsei Chemical Co., Ltd. O-Acetyl-L-serine, O-succinyl- L-homoserine, pyridoxal phosphate, and L-cystathionine were purchased from Sigma. DL-Cystathionine was purchased from Koch-Light Laboratories Ltd. The other chemicals used were the same as reported in the previous paper (1). RESULTS Growth Responses on Intermediates of Meth ionine Biosynthesis and Defective Enzymes in Meth ionine Auxotrophs-Growth responses on interme diates of methionine biosynthesis were tested for 27 strains of methionine auxotrophs derived from B. flavum No Among them 14 strains (Group 1) did not grow on homoserine but did grow on O-acetylhomoserine. None of 9 strains (namely, H1013, 4 strains of Group 1, 2 strains of Group 2, and 2 strains of Group 3) tested grew on DL-cystathionine or L-cystathionine. On the other hand, methionine auxotroph (3-IM-4 of Escherichia coli, as a control, grew on DL-cystathionine. Three strains (Group 2) out of the 27 strains did not grow on or grew very slightly on O-acetylhomoserine but they did grow on DLhomocysteine. The other 10 strains (Group 3) did not grow on DL-homocysteine but did grow on Lmethionine. Then defective enzymes of methionine biosynthesis in these auxotrophs were determined by measuring each enzyme activity in the crude extract of the cells grown under limited methionine, because synthesis of the enzymes was probably repressed by methionine as observed for HAT (2). As shown in Table I, strains which belonged to Group 1 lacked HAT, and also showed an extreme decrease in AHS activity (approxi mately 1/10-1/100 of control strain H1013). On the other hand, cystathionine ƒá-synthase and ƒàcystathionase activities were the same levels as in H1013. M-63, which belonged to Group 2, did not show AHS activity at all, and M-35 of Group 2 which grew very slightly on homoserine and O-acetylhomoserine showed an extremely decreased level of AHS activity (approximately 1/100 of that of H1013). On the other hand, the other 3 en zyme activities of M-63 and M-35 were at the same levels as in H1013. The enzyme activities of two strains, M-78 and M-116, were determined out of 10 strains which belonged to Group 3. Both the strains showed all of the 4 enzymes at the same levels of activity as in H1013. Although homocysteine methyltransferase which catalyzes the formation of methionine from homocysteine was not determined, these strains appeared to lack the enzyme judging from the growth response. The reason why all the strains of Group I tested showed extremely low AHS activities, in contrast to the growth on O-acetylhomoserine, is not clear. It is possible that AHS of these mutants is labile or extremely sensitive to the repression control owing to the defect of HAT, the first enzyme of the methionine-biosynthetic pathway. The fact that all of the strains tested of Group 1 lacked HAT supports the previous conclusion (2) that this is the first step enzyme of methionine biosyn- Vol. 91, No. 4, 1982

4 1166 H. OZAKI and I. SHIIO TABLE I. Growth responses on intermediates and specific activities of enzymes related to methionine biosynthesis in methionine auxotrophs. Growth responses were observed after cultivation 30 Ž on medium 10 (basal 48 h at medium) supplemented 0.1 ml 1% solution of L-homoserine (HS), O-acetyl-L-homoserine (AH), DL-cystathionine with of DL-homocysteine (HC), or L-methionine (Met) in 'penicillin-cup.' For auxotroph, (CT), a a homoserine and methionine and threonine double auxotrophs, supplemented mg/liter of used. medium 10 with 500 L-threonine was 0.1 ml suspension (108/ml) was spread on the agar plate. determine the enzyme specific activities the of cell To auxotrophs were cultured at 30 Ž for 24 h medium 5 medium for supplemented with mg/liter L in (or 13 *) 30 of methionine. For the homoserine auxotroph, and methionine and threonine double auxotrophs, 30 mg/liter of L- methionine and 500 mg/liter Lthreonine were added. -30, -49, -35, -63, and were derived of M-27, -70, TM-66-1 from strain No. 15-8, and M-248 from No strain a Cystathionine ƒá-synthase. b ƒà-cystathionase. C N indicates that no experiment was carried out. thesis in this bacterium. The fact that strains of Group 2, especially M-63, lacked only AHS indicates that the enzyme(s) catalyzing the conversion of O-acetylhomoserine to homocysteine is AHS and not both cystathionine ƒá-synthase and ƒà-cystathionase. Moreover, in order to confirm the participation of AHS in methionine biosyn thesis, prototrophic revertants were derived from methionine auxotroph M-63 which lacks AHS. These revertants showed similar AHS activity to the wild-type original strain, No. 15-8, as shown in Table II. These low AHS activities in the revertants and wild-type strain seemed due to repression of the enzyme by methionine synthe sized de novo. From these results, it was reasonably concluded that the methionine biosynthetic pathway of B. flavum is as follows: homoserine O-acetylhomoserine homocysteine methionine. Repression of AHS Synthesis by Methionine- Examination was carried out of the effect of methi onine on syntheses of HAT and AHS, which were found to participate in methionine biosynthesis, as described above, ƒà-cystathionase which did not participate in methionine synthesis, and O-acetyl serine sulfhydrylase (ASS), which seemed to be one of enzymes for cysteine biosynthesis. As shown in Table III, AHS formation was strongly repressed by excess methionine, similar to the case of HAT (2). On the other hand, ƒà-cystathionase synthesis was only slightly (about 1/2) repressed, and ASS synthesis was not repressed at all by excess methionine. The strong repression of AHS J. Biochem.

5 O-ACETYLHOMOSERINE SULFHYDRYLASE OF B. flavum 1167 TABLE U. AHS activities of prototrophic revertant mutants from strain M-63. Revertants were obtained by picking up colonies which appeared spontaneously on medium 10 (glucose minimum medium) containing / plate of M-63 at 30 Ž. To determine AHS activity, No and revertants were cultured in medium 13, and M-63 in medium 13 supplemented 30 mg/liter of L-methionine, at 30 Ž for 24 h. a Absorbancy at 562 nm after 26-fold dilution. o Growth on medium 13 supplemented with 30 mg/liter of L- methionine. TABLE V. Effect of methionine on HAT and AHS syntheses. Strain H1013 was cultured at 30 Ž for 24 h in medium 5 supplemented with 500mg/liter of Lthreonine and 20 or 500 mg/liter Of Lmethionine. TABLE W. Purification of AHS from B. flavum. AHS was purified from an extract of B. flarum H1013 cells (14 g as wet weight) by the method described in "MATERIALS AND METHODS." suggests its participation in methionine biosynthesis. The results also indicate that AHS is a different protein from ASS in contrast to the enzyme in yeast (14). Purification of AHS-AHS was partially puri fied from strain H1013 grown in a methionine limited medium by the method described in "MATERIALS AND METHODS." As shown in Table IV, the specific activity of the purified preparation was 24-fold over the crude extract obtained from derepressed cells and 1,100-fold over that from repressed cells of the wild-type strain (Table II). The purity of AHS was determined by disc-gel electrophoresis which showed two minor protein bands and a main band showing AHS activity; it was calculated to be about 70% Vol. 91, No. 4, 1982

6 1168 H. OZAKI and I. SHIIO by measuring optical density after protein staining of the disc-gel after electrophoresis. The enzyme preparation did not show any HAT or ƒà-cystathionase activity, but showed very slight ASS activity, although most of the ASS activity was removed during purification of AHS. In order to clarify whether ASS activity of the partially purified AHS Fig. 1. DEAE-cellulose column chromatography of the partially purified AHS preparation. The pooled fraction (17.5 ml) obtained by hydroxylapatite column chromatography was placed on a DEAE-cellulose column (1.5 ~ 10 cm) which had been equilibrated with buffer B. The column was then washed with 15 ml of buffer B and 15 ml of buffer B containing 0.2 M NaCl, and eluted with a concentration gradient of NaCI (---) formed from 50 ml of buffer B containing 0.2 M NaCl and 50 ml of buffer B containing 0.5 xi NaCl. ASS( ) was assayed in the standard reaction mixture except that Tris-HCl buffer, ph 8.7, was used instead of Tris-HCl buffer, ph 7.4. AHS ( ) and cystathionine ƒá-synthase ( ) were assayed by the methods described in "MA TERIALS AND METHODS." TABLE V. Treatment of AHS with hydroxylamine. The DEAE-cellulose fraction (0.46 mg protein) de scribed in Table IV was gel-filtered through a Sephadex G-50 column using 0.05 M potassium phosphate buffer, ph 7.3. Hydroxylamine hydrochloride was added to the gel-filtered enzyme to a final concentration of 0.1 M. After incubation at 22 Ž for 30 min, the mixture was gel-filtered as above to remove hydroxylamine hydrochloride. AHS activity was determined in the standard reaction mixture except that Tris-HCl buffer, ph 8.7, was used instead of Tris-HCl buffer, ph 7.4. TABLE Y. Substrate specificity and effects of metabolic intermediates on AHS activity. Reactions were carried out in the standard reaction mixture except that Tris-HCl buffer, ph 8.7, was used instead of Tris-HCl buffer, ph 7.4. The enzyme added (1 ƒêg as protein) was the ammonium sulfate (0-0.9 saturation) precipitate of the AHS frac tion obtained in the 2nd DEAE-cellulose column chromatography as shown in Fig. 1. a 2.25 mm S-adenosyl-L-meth ionine was added. J. Biochem.

7 O-ACETYLHOMOSERINE SULFHYDRYLASE OF B. forum 1169 preparation is due to AHS itself or to contamina tion by a different enzyme, DEAE-cellulose chro matography of the AHS preparation was carried out again. As shown in Fig. 1, the ASS activity was separated into two peaks; one was eluted in the same fraction as AHS activity was and the other in fractions where most ASS activity was eluted when the crude preparation was chromatographed under the same conditions. Therefore, AHS seems to have slight ASS activity. On the other hand, the behavior of the cystathionine ć-synthase activity was similar to that of AHS both during the course of purification and in the rechromatography on DEAF-cellulose, as shown in Fig. 1. There is the possibility that these two activities are attributable to the same enzyme. Properties of AHS-The molecular weight of AHS was determined to be 360,000 by gel-filtra tion of the partially purified enzyme on Sephadex G-200. The enzyme activity showed an optimum ph of 8.7 and higher activities were shown in phosphate buffer than in Tris-HCI buffer especially on the acidic ph side. As shown in Table V, the enzyme activity was neither decreased after gel- Fig. 2. Double reciprocal plots of reaction velocity against O-acetylhomoserine (A) or H2S (B) concentration at various fixed concentrations of H2S or O-acetylhomoserine, respectively. The reactions were carried out in a reaction mixture containing 150 mm potassium phosphate buffer, ph 8.5, 0.77 mm EDTA, the indicated concentrations of O-acetylhomoserine, 0.05 mm ( ), mm ( ) or 0.1 mm H2S ( ), and 1 Đg (as protein) of the partially purified enzyme in (A). Concentrations of O-acetylhomoserine in (B) were 1.5 mm ( ), 2 mm ( ), 2.5 mm ( ), 5 mm ( ), and 10 mm ( ). V is A540/4 min. R is the inter cept on the vertical axis of each plot. Vol. 91, No. 4, 1982

8 1170 H. OZAKI and I. SHIM filtration on a Sephadex G-50 column in an attempt to remove pyridoxal phosphate in the prepa ration, nor increased by the addition of 0.1 mm pyridoxal phosphate. However, when the gelfiltered preparation was treated with hydroxylamine, according to the method of Kaplan and Flavin (15), it lost the activity almost totally, but was markedly activated by addition of 0.1 mm pyridoxal phosphate to 50% of the original activ ity. The results indicate that the enzyme has pyridoxal phosphate as a coenzyme. Table VI shows substrate specificity: the enzyme reacted spe cifically with O-acetyl-L-homoserine, and showed with O-acetyl-L-serine only one hundredth as much activity as that with O-acetyl-L-homoserine. Dou ble reciprocal plots of reaction velocity against O-acetyl-L-homoserine concentration are shown in Fig. 2A. The Km for H.S was determined to be mm from secondary plots of each intercept on the vertical axis against 1/[H2S], as shown in the inset of Fig. 2A. Similarly, double reciprocal plots of Fig. 2B showed the Km for O-acetylhomoserine to be 2.O mm. The enzyme was in hibited 50, 23, and 29% by (each 10 mm) L-methionine, L-homoserine, and O-acetyl-L-serine, respectively. No inhibition was observed with Lserine, glycine, cystathionine, or S-adenosyl-L-methionine, as shown in Table Y. DISCUSSION From growth responses on the biosynthetic intermediates and defective enzymes of methionine auxotrophs, it can be concluded that the methio nine biosynthetic pathway in B. flavum is as follows; homoserine O-acetylhomoserine homocysteine methionine. Enteric bacteria, Neurospora crassa and yeast have enzymes which participate in both the pathway described above (sulfhydrylase pathway) and the pathway via cysta thionine (3), but it has been reported that the cystathionine pathway in enteric bacteria and Neu rospora provides a major alternative one for methi onine biosynthesis and the sulfhydrylase pathway provides an ineffective alternative because Ĉ-cysta thionase lacking mutants of these organisms grow only slightly in the absence of methionine (3). In yeast, however, which of them is the major physiological pathway is not clear (3, 4). Therefore, this is the first example of the sulfhydrylase pathway being the sole biosynthetic pathway for methionine. Despite that extracts of M-35 and M-63 showed low cystathionine ć-synthase and Ĉ-cysta thionase activities as did that of the parent, these mutants, especially M-63, are auxotrophic with respect to methionine. Therefore these two en zymes are probably inoperative under in vivo con ditions. The inability of cystathionine to support the growth of methionine auxotrophs which grew on O-acetylhomoserine appears to be due to the impermeability of the cells for cystathionine, or the inoperativeness of Ĉ-cystathionase in vivo. At present there is no evidence which supports opera tion of the cystathionine pathway in B. flavum. It has been reported that cystathionine ćsynthase of enteric bacteria also catalyzes the O- succinylhomoserine sulfhydrylase reaction (16). It is possible that AHS and cystathionine ć-synthase of B. flavum are the same protein because both the enzymes behaved in the same way during purification. In yeast, however, it has been reported that AHS does not show cystathionine y-synthase (4), and that AHS and ASS are the same protein (14). AHS of B. flavum was sepa rated from most ASS, and the ASS activity of the partially purified AHS was only 1/100 of the AHS activity, corresponding to 1/6,000 to 1/500 of the total ASS activity shown by the crude enzyme solution (Tables V, Y). On the contrary, the Neurospora enzyme does not show either ASS activity or cystathionine ć-synthase activity at all (17). The Km value of the enzyme for 0-acetylhomoserine was 2.0 mm, which is smaller than those of Neurospora (7 mm) (17), yeast (4.55 mm) (18), and Salmonella (5 mm O-succinylhomoserine) (16). The Km value of the enzyme for H2S was mm, which is also much smaller than those of Neurospora (0.8 mm) (17) and Salmonella (3 mm) (16). The Km value of B. flavum AHS for H,S is small enough to be comparable to that (less than 0.1 mm) (19) of ASS of Salmonella which appears to be physiologically active in vivo. Thus the Km value of B. flavum AHS for H2S seems to be of a physiologically significant level. AHS of Corynebacterium acetophilum which is suggested to be inoperative for homocysteine and methio nine synthesis in vivo was reported to have Km values of 13 mm and 2.4 mm for 0-acetylhomo- J. Biochem.

9 1 O-ACETYLHOMOSERINE SULFHYDRYLASE OF B. flavum 1171 serine and HAS, respectively (20), which are much larger than those of B. flavum AHS. With regard to metabolic regulation, it has been reported that HAT, the first enzyme of the methionine biosynthetic pathway of B. flavum, is subject to feedback inhibition by S-adenosyl-Lmethionine (1). The methionine concentration giving 50% inhibition of AHS was 10 mm. This concentration is much higher than that (0.26 mm) (1) of S-adenosyl-L-methionine which gives 50% inhibition of HAT. Therefore, the methionine inhibition of AHS seems not to be physiological significant. From the results of the previous and present studies, it can be concluded that methio nine biosynthesis in B. flavum is regulated through feedback inhibition of HAT by S-adenosylmethio nine (1) and repression of HAT (2) and AHS syntheses by methionine or its metabolites. The authors are indebted to Drs. T. Akashi and Y. Hirose of the Central Research Laboratories for encouragement during this work. REFERENCES. Shiio, I. & Ozaki, H. (1981) J. Biochem. 89, Miyajima, R. & Shiio, 1. (1973) J. Biochem. 73, Flavin, M. (1975) in Metabolic Pathways (Green berg, D.M., ed.) Vol. 7, pp , Academic Press, New York 4. Savin, M.A. & Flavin, M. (1972) J. Bacteriol. 112, Sano, K. & Shiio, 1. (1967) J. Gen. Appl. Microbiol. 13, Shiio, I. & Sano, K. (1973),4gric. Biol. Chem. 37, Shiio, I., Ozaki, H., & Ujigawa-Takeda, K. (1982) Agric. Biol. Chem. 46, Shiio, I. & Nakamori, S. (1969) Agric. Biol. Chem. 33, Sugimoto, S. & Shiio, I. (1977) J. Biochem. 81, Shiio, I. & Ujigawa, K. (1978) J. Biochem. 84, Data for Biochemical Research (Dawson, R.M.C., Elliott, D.C., Elliot, W.H., & Jones, K.M., eds.) (1969) 2nd Ed., pp , Oxford University Press, Oxford 12. Kaplan, M. & Guggenheim, S. (1971) in Methods in Enzymology (Tabor, H. & Tabor, C.W., eds.) Vol. 17B, pp , Academic Press, New York 13. Kredich, N.M. & Becker, M.A. (1971) in Methods in Enzymology (Tabor, H. & Tabor, C.W., eds.) Vol. 17B, pp , Academic Press, New York 14. Yamagata, S., Takeshima, K., & Naiki, N. (1974) J. Biochein. 75, Kaplan, M.L. & Flavin, M. (1966) J. Biol. Client. 241, Flavin, M. & Saughter, C. (1967) Biochim. Biophys. Acta 132, Kerr, D.S. (1971) J. Biol. Chem. 246, Yamagata, S. (1971) J. Biochem. 70, Becker, M., Kredich, N.M., & Tomkins, G.M. (1969) J. Biol. Client. 244, Murooka, Y., Kakihara, K., Miwa, T., Seto, K., & Harada, T. (1977) J. Bacteriol. 130, Vol. 91, No. 4, 1982