J. Nutr. Sci. Vitaminol., 28, 225-236, 1982 the Mechanism of in Situ

J. Nutr. Sci. Vitaminol.,28, 225-236, 1982

The Mechanism of In Situ Reactivation of Glycerol Inactivated Coenzyme B12-Dependent Enzymes, Glycerol Dehydratase and Diol Dehydratasel

Kazutoshi USHIO, Susumu HONDA, Tetsuo TORAYA,2 and Saburo FUKUI3

Laboratory of Industrial Biochemistry, Department of Industrial Chemistry, Faculty of Engineering, Kyoto University, Yoshida, Sakyo-ku, Kyoto 606, Japan (Received September 28, 1981)

Summary In the previous paper (S. Honda, T. Toraya, and S. Fukui, J. Bacteriol., 143, 1458-1465 (1980)), we reported that the glycerol inactivated holoenzymes of adenosylcobalamin-dependent glycerol dehy dratase and diol dehydratase are rapidly and continually reactivated in toluene-treated cells (in situ) by adenosine 5•L-triphosphate (ATP) and divalent metal ions in the presence of free adenosylcobalamin. To elucidate the mechanism of this in situ reactivation, the nature of the binding of various irreversible cobalamin inhibitors to the dehydratases in situ was investigated. In the presence of ATP and Mn2+, enzyme-bound hydroxocobalamin, cyanocobalamin and methylcobalamin were rapidly displaced by added adenosylcobalamin. Without ATP and Mn2+, such displacement did not take place. In contrast, enzyme-bound adeninyl butylcobalamin and adenosylethylcobalamin were essentially not displace able by the free coenzyme even in the presence of ATP and Mn2+. Inosylcobalamin was a very weak inhibitor irrespective of the presence of ATP and Mn2+ . These results indicate that the relative affinity of the enzymes in situ for the cobalamins with simple Cof ligands was markedly lowered in the presence of ATP and Mn2+, whereas that for the cobalamins with adenine-containing ligands was not. When the glycerol inactivated holoenzymes in situ were dialyzed against a buffer containing ATP and Mg2+, the inactivated coenzyme moiety dissociated from the enzymes leaving apoproteins. Kinetic evidence was also obtained with the dehydratases in situ that continual displacement of the inactivated

1 This is Paper XVIII in the series concerning "Coenzyme B 12-Dependent Dehy dratases." 2 Address correspondence to this author at the present address: Department of

Chemistry, College of Liberal Arts and Sciences, Kyoto University, Yoshida, Sakyo-ku, Kyoto 606, Japan.

3 牛尾一利,本多進, 2 虎谷哲夫 ,福井三郎

225 226 K. USHIOet al.

coenzyme moiety by adenosylcobalamin takes place during the glycerol dehydration reaction in the presence of ATP and Mn2+. Since the adenosyl group of the bound coenzyme is irreversibly removed from the cobalamin moiety during inactivation by glycerol, all of these data constitute clear evidence that the inactivated holo-dehydratases are re activated in situ in the presence of ATP and Mn2+ by displacement of the modified coenzyme moiety by free intact adenosylcobalamin (i.e. selective B12-exchange mechanism). Key Words coenzyme B12, glycerol dehydratase, diol dehydratase, sui cide inactivation, mechanism of in situ reactivation, Toluene-treated cell

Coenzyme B12 (adenosylcobalamin)-dependent glycerol dehydratase (glycerol

hydro-lyase, EC 4.2.1.30) and diol dehydratase (DL-1, 2-propanediol hydro-lyase, EC 4.2.1.28) are isofunctional but immunologically different enzymes, both of

which catalyze the conversion of glycerol to ƒÀ-hydroxypropionaldehyde, that of 1, 2

- propanediol to propionaldehyde and that of 1, 2-ethanediol to acetaldehyde (1-6). As described in our previous paper (7), glycerol is an important physiological substrate for glycerol dehydratase and, in some cases, for diol dehydratase as well.

Recent studies in this and other laboratories, however, have indicated that both dehydratases undergo suicidal inactivation by glycerol during catalysis (2, 3, 8-10). This apparent inconsistency was eliminated by our recent finding that the glycerol

inactivated dehydratases are rapidly reactivated in toluene-treated cells (in situ) by ATP and divalent metal ions in the presence of free adenosylcobalamin (11).

During inactivation by glycerol, the carbon-cobalt bond of the enzyme-bound

coenzyme is irreversibly broken forming 5•L-deoxyadenosine from the adenosyl

group (3, 12, and T. Toraya, T. Tobimatsu, T. Kawamura, and S. Fukui, manuscript in preparation). We have shown that the apoprotein itself is not

damaged during inactivation. Therefore, two possible mechanisms were considered for this reactivation in situ: the regeneration of adenosylcobalamin by cobalt

adenosylation of the bound inactivated coenzyme moiety (B12-adenosylation mechanism) and the displacement of the bound inactivated coenzyme moiety by free

adenosylcobalamin (B12-exchange mechanism). The experimental results reported in our previous paper indicated that the B12-adenosylation mechanism is very unlikely (11). In this paper, we report definite evidence that the glycerol-inactivated

glycerol and diol dehydratases are reactivated in situ through the B12-exchange mechanism.

MATERIALS AND METHODS

Chemicals. Crystalline adenosylcobalamin and methylcobalamin were gifts from Kyowa Hakko Kogyo Co., Ltd., Tokyo and Eisai Co., Ltd., Tokyo, respec tively. Adeninylbutylcobalamin, adenosylethylcobalamin, and inosylcobalamin were kindly supplied by Professor H. P. C. Hogenkamp, University of Minnesota,

J. Nutr. Sci. Vitaminol. REACTIVATION OF B12 ENZYMES 227

Minneapolis, Minnesota. Cyanocobalamin was obtained from Glaxo Laboratories

Ltd., Greenford, England. Hydroxocobalamin was prepared by aerobic photolysis

of methylcobalamin (13, 14). All other chemicals were reagent-grade commercial

products, and were used without further purification. Bacteria and growth. As described in the previous paper (11), glycerol-grown cells of Klebsiella pneumoniae ATCC 25955 (formerly Aerobacter aerogenes PZH

572, Warsaw) and glycerol-1, 2-propanediol-grown cells of K. pneumoniae ATCC 8724 (formerly A. aerogenes) were used as the sources of glycerol dehydratase and

diol dehydratase, respectively (8, 15). The bacteria were grown at 30•Ž or 37•Ž without aeration in a complex medium containing 5.4g of KH2PO4, 1.2g of

(NH4)2SO4, 0.4g of MgSO4 7H2O, 2.0g of yeast extract, 2.0g of tryptone, and 9.2g of glycerol (glycerol medium) or 9.2g of glycerol plus 5.7g of 1, 2-propanediol

(glycerol-1, 2-propanediol medium) in 1 liter of tap water (7, 8, 15). The medium was adjusted to pH 7.1 with KOH.

Permeabilized cells. Bacterial cells were permeabilized by toluene treatment, as described before (11). The concentration of the bacteria and toluene-treated cells was determined by comparison of the turbidity of the suspensions at 580nm with

standard curves (11).

Purification of diol dehydratase. The purified preparation of the diol dehy dratase of K. pneumoniae ATCC 8724 was obtained as described before (16) and used for the kinetic measurements as an in vitro system.

Enzyme assays. The activities of the glycerol dehydratase and diol dehydratase

in vitro and in situ were determined by the 3-methyl-2-benzothiazolinone hydrazone method of Toraya et al. (8, 17), as described in the previous paper (11).

RESULTS AND DISCUSSION

Effect of hydroxocobalamin on reactivation in situ As reported previously (11), hydroxocobalamin itself is not able to replace

adenosylcobalamin in the reactivation mixture for the glycerol-inactivated glycerol dehydratase in situ. When both adenosylcobalamin and hydroxocobalamin were

added together in equimolar quantities (15ƒÊM each), the reactivation was 46 inhibited (Table 1). It should be noted that this percentage is much smaller than that of inhibition of the dehydratase reaction obtained with hydroxocobalamin without

ATP and Mn2+ (>90%). This result suggested that the relative affinity of the in situ

enzyme for hydroxocobalamin was greatly lowered by the presence of ATP and Mn2+.

Displacement of the enzyme-bound cobalamin inhibitors in situ by adenosylcobalamin The result described above led us to examine reactivation of the inactive model complexes of the dehydratases with cyanocobalamin, hydroxocobalamin, methyl cobalamin, adeninylbutylcobalamin, and adenosylethylcobalamin. These model complexes formed in situ were incubated in the reactivation mixture, namely with Vol.28, No. 3, 1982 228 K. USHIO et al.

Table 1. Effect of hydroxocobalamin on the reactivation in situ of glycerol-inactivated

glycerol dehydratase. Toluene-treated cells of K. pneumoniae ATCC 25955 were incubated at 37•Ž for 30min in a mixture containing 15ƒÊM adenosylcobalamin, 0.2M

glycerol, 0.05M KCl and 0.04M potassium phosphate buffer (pH 8.0). Free adenosyl cobalamin and glycerol were removed by washing and dialysis of the cells against 0.05M

potassium phosphate buffer (pH 8.0). The experiment for reactivation in situ was carried out at 37•Ž for 10min with 0.009mg (dry weight) of the inactivated cells in the reactivation mixture. The complete system contained 0.2M glycerol, 0.05M KCl, 0.03M

potassium phosphate buffer (pH 8.0), 15ƒÊM adenosylcobalamin, 3mM ATP, 3mM MnCl2 and inactivated cells, in a total volume of 1.0ml.

Fig. 1. Time course of displacement in situ of the glycerol dehydratase-bound coba lamin inhibitors by adenosylcobalamin. Toluene-treated cells (0.022-0.025mg of dry cells) of K. pneumoniae ATCC 25955 were incubated at 37•Ž in a mixture containing 0.2 mmol of 1, 2-propanediol, 0.05 mmol of KCl, 0.035 mmol of

potassium phosphate buffer (pH 8.0) and 15 nmol of hydroxocobalamin (A), 15 nmol of cyanocobalamin (B), or 5 nmol of adeninylbutylcobalamin (C), with (•¤) or without (_??_) 3ƒÊmol of ATP and 3ƒÊmol of MnCl2, in a total volume of 0.9ml. Adenosylcobalamin (15 nmol) was added to the mixture at 10min of incubation (•¤, _??_) to a final volume of 1.0ml, and the amount of propionaldehyde formed by the incubation at 37•Ž for the indicated time was determined. Controls 1 and 2; minus cobalamin inhibitor with (•¢) or without (•›) ATP and MnCl2. Controls 3 and 4; A cobalamin inhibitor and adenosylcobalamin were added at the same time (zero time) with (•£) or without (•œ) ATP and MnCl2.

J. Nutr. Sci. Vitaminol. REACTIVATIONOF B12ENZYMES 229

ATP, Mn2+ and adenosylcobalamin. The extent of displacement of the enzyme bound cobalamin inhibitors by the added coenzyme was estimated from the appearance of the 1,2-propanediol-dehydrating activity. As illustrated in Figs. 1A and B, inactive complexes of glycerol dehydratase with hydroxocobalamin and cyanocobalamin in situ were rapidly reactivated in the presence of ATP, Mn2+ and adenosylcobalamin. The enzyme-methylcobalamin complex was also reactivated in a similar manner (data not shown). These results indicate that these enzyme-bound inhibitors were rapidly displaced by the added coenzyme in the presence of ATP and divalent cations. The same results were obtained with the complexes of diol dehydratase in situ with these cobalamin inhibitors (data not shown). Without ATP and Mn2+, such displacement did not take place at all. When adenosylcobalamin and these cobalamin inhibitors were added together at the same time (zero time) with ATP and Mn2+, only very slight inhibition was seen. Under the same conditions without ATP and Mn2+, the degrees of inhibition of the enzyme reaction by hydroxocobalamin and cyanocobalamin observed were about 90% and 60%, respectively. Thus, it is evident that the relative affinity of the dehydratases in situ for hydroxocobalamin, cyanocobalamin, and methylcobalamin was greatly weak ened by ATP and Mn2+. In contrast, the enzyme-bound adeninylbutylcobalamin was not displaced by the added coenzyme in the presence of ATP and Mn2+ (Fig. 1C). The enzyme bound adenosylethylcobalamin was also essentially not displaceable by the added coenzyme (data not shown). When adenosylcobalamin and adeninylbutylcobalamin were added at the same time (zero time), marked inhibition by this analog was observed both in the presence and absence of ATP and Mn2+. Inosylcobalamin, which bears a nucleoside ligand but not an adenine moiety, was a very weak competitive inhibitor irrespective of the presence of ATP and Mn2+ . Similar results were obtained with diol dehydratase in situ as well. Therefore, it was strongly suggested that the relative affinity of the dehydratases for the adenine-containing cobalamins was not lowered even in the presence of ATP and divalent cations.

Evidence for changes in relative affinity of the diol dehydratase in situ for adenosylco balamin and cobalamin inhibitors in the presence and absence of ATP and Mn2+ To prove the change in relative affinity of the diol dehydratase in situ for cobalamin inhibitors in the presence and absence of ATP and Mn2+, the ratio of Ki for cobalamin inhibitor to Km for adenosylcobalamin (Ki/Km)was determined with toluene-treated K. pneumoniae ATCC 8724 (containing diol dehydratase) by conventional kinetic experiments (Table 2). K. pneumoniae ATCC 25955 cells (containing glycerol dehydratase) were not used in these experiments, since double reciprocal plots obtained with them did not give straight lines. This may be due to the presence of a small amount of diol dehydratase in the cell (7, 18), because glycerol and diol dehydratases are quite different in relative affinity for adenosyl cobalamin (18). The Ki/Kmratio obtained for hydroxocobalamin with the enzyme in situ in the presence of ATP and Mn2+ was 2.2, which is considerably higher than the Vol.28, No. 3, 1982 230 K. USHIO et al.

Table 2. Effects of the presence of ATP and Mn2+ on ratios of Ki for cobalamin inhibitors to Km for adenosylcobalamin obtained with diol dehydratase in situ and in vitro. Ki/Km values were obtained by the conventional kinetic method using toluene treated K. pneumoniae ATCC 8724 cells (in situ) or purified diol dehydratase (in vitro).

value obtained in the absence of ATP and Mn2+. In contrast, the ratio obtained in situ for adeninylbutylcobalamin was still small even in the presence of ATP and Mn2+. Ki/Km values determined in vitro for hydroxocobalamin and adeninylbutyl cobalamin with purified diol dehydratase were not affected at all by the presence of ATP and Mn2+. These data provide additional evidence that the relative affinity of the enzyme for hydroxocobalamin was lowered in situ by ATP and Mn2+, whereas that for the adenine-containing cobalamin was not.

Inhibition of the continual reactivation by adeninylbutylcobalamin in situ The conclusion reached using model complexes is very important. In order to gain an insight into the mechanism of reactivation of the glycerol-inactivated dehydratases in situ, adeninylbutylcobalamin, a strong inhibitor even in the presence of ATP and Mn2+, was added to the reaction mixture. As shown in Fig. 2B, the dehydration of glycerol by the glycerol dehydratase in situ ceased very rapidly after the addition of adeninylbutylcobalamin. The same results obtained with diol dehydratase in situ (data not shown). In clear contrast, dehydration of 1,2 - propanediol was inhibited only slightly by this analog (Fig. 2A). It is evident from the results that adeninylbutylcobalamin competes with adenosylcobalamin for the coenzyme binding site of the dehydratases even in the presence of ATP and Mn2+. Thus, the data in Fig. 2 indicate rapid and continual displacement of the inactivated coenzyme moiety in the glycerol-inactivated holoenzyme by the added adenosyl cobalamin during the glycerol dehydration reaction and much less frequent displacement during the 1,2-propanediol dehydration reaction. This conclusion is quite reasonable, since 1,2-propanediol is a substrate which does not cause significant inactivation of the holoenzyme in its dehydration reaction (8). In the absence of ATP and Mn2+, the binding of adenosylcobalamin to the enzyme in situ was virtually irreversible (Fig. 2A), since the adeninylbutylcobalamin added at 10 min of incubation did not inhibit the reaction. These results constitute strong J. Nutr. Sci. Vitaminol. REACTIVATION OF B12 ENZYMES 231

Fig. 2. Inhibition of the continual reactivation in situ of glycerol dehydratase by adeninylbutylcobalamin during 1, 2-propanediol (A) and glycerol (B) dehydration reactions. Toluene-treated cells (0.020mg of dry cells) were incubated at 37•Ž in a mixture containing 0.2mmol of 1, 2-propanediol (A) or glycerol (B), 0.05mmol of KCl, 0.035mmol of potassium phosphate buffer (pH 8.0) and 15nmol of adenosylcobalamin, with (•¢) or without (•›) 3ƒÊmol of ATP and 3ƒÊmol of MnCl2, in a total volume of 0.9ml. Adeninylbutylcobalamin (5nmol) was added to the mixture at 10min of incubation (•£, •œ), to a final volume of 1.0ml, and the products formed by the incubation at 37•Ž for the indicated time were determined. evidence for the B12-exchange mechanism that is, the inactivated holoenzymes of glycerol dehydratase and diol dehydratase are reactivated in situ by displacement of the inactivated coenzyme moiety by the added intact coenzyme.

Dissociation of the inactivated coenzyme moiety from the glycerol-inactivated dehy dratases in situ by dialysis against an ATP and Mg2+ -containing buffer To test the possibility of the intermediary formation of the apoenzyme during reactivation, the glycerol-inactivated dehydratases in situ were dialyzed against an ATP and Mg2+ -containing buffer. Instead of Mn2+, Mg2+ was used in these experiments because of its high solubility in the dialysis buffer. As shown in Table 3, 26% of the original activity was recovered as apoenzyme when the inactivated glycerol dehydratase was dialyzed in the presence of ATP and Mg2+ . No holoenzyme activity was detected. In the absence of ATP and Mg2+ in the dialysis buffer, essentially no activity was recovered. Thus, it can be concluded that dissociation of the inactivated coenzyme moiety from the enzyme takes place even in the absence of the added coenzyme. Since hydroxocobalamin, a possible product from the inactivated coenzyme, sticks to other proteins nonspecifically (19), K2SO3 was added to the dialysis buffer to effect the dissociation of the modified coenzyme. SO32- is known to react with hydroxocobalamin forming sulfitocobalamin (20). As

Vol. 28, No. 3, 1982 232 K. USHIO et al .

Table 3. Dissociation of the inactivated coenzyme moiety from the glycerol inactivated dehydratases in situ by dialysis. The samples were dialyzed at 4•Ž for 12hr

against 1,000 volumes of 0.01M potassium phosphate buffer (pH 8.0) plus 2% 1 , 2 - propanediol with the indicated additions. The concentration of ATP, MgCl2 and K2SO3 added were 0.25mM, 3mM and 50mM, respectively. Excess reagents were removed by the

second dialysis against 1,000 volumes of 0.01M potassium phosphate buffer (pH 8 .0) plus 1, 2-propanediol with two buffer changes. The enzyme activity recovered was determined by the standard assay method (11) after dilution. The holoenzyme activity was also measured without addition of adenosylcobalamin, but was less than 1% of the total original activity.

a In situ glycerol-inactivated glycerol dehydratase and diol dehydratase were obtained as described in the footnote to Table 1. Free adenosylcobalamin was removed by dialysis against 0.01M potassium phosphate buffer (pH 8.0) containing 2% 1,2-propanediol.

given in Table 3, the recovery of the active apoenzyme was markedly increased by this addition (recovery of activity, 68% for glycerol dehydratase and 97% for diol dehydratase). The difference between the two dehydratases in the recovery of apoenzyme may reflect their difference in affinity for cobalamins (18). At any rate, these results offer additional strong evidence that the glycerol-inactivated glycerol dehydratase and diol dehydratase in situ are reactivated through the B12-exchange mechanism.

The B12-exchange mechanism for reactivation in situ Based on the data obtained so far, we propose a plausible mechanism for the reactivation of the glycerol-inactivated dehydratases in situ, which is shown in Fig . 3. The apoenzyme binds adenosylcobalamin to form catalytically active holoen J. Nutr. Sci. Vitaminol. REACTIVATION OF B12 ENZYMES 233

Fig. 3. Schematic representation of the B12-exchange mechanism for reactivation in

situ of glycerol and diol dehydratases. ATP and divalent cation-dependent dis

sociation of the bound cobalamins from the apoproteins is shown by bold arrows.

(C o), cobalamin; Ad-5•L-d-ribose, 5•L-deoxyadenosine; Ad, adenine; AdoCbl, adeno

sylcobalamin. zyme. The resulting holoenzyme catalyzes dehydration of glycols to the correspond ing aldehydes. When glycerol is used as substrate, suicide inactivation occurs at a high probability. The inactivated enzyme contains the modified coenzyme moiety whose adenosyl group is irreversibly dissociated from the cobalt atom (3, 12, and T. Toraya, T. Tobimatsu, T. Kawamura, and S. Fukui, manuscript in preparation). In the presence of ATP and divalent cations, the inactivated coenzyme moiety is bound only loosely and dissociates readily from the enzyme to leave apoenzyme, as shown by bold arrow. Similar dissociation in the presence of ATP and Mn2+ takes place Vol.28, No. 3, 1982 234 K. USHIO et al.

with the model complexes of the enzymes with cobalamins bearing simple CoƒÀ

ligands, such as hydroxo-, cyano-, and methylcobalamin. The apoenzyme formed can be reconstituted to the active holoenzyme with added adenosylcobalamin,

which leads to reactivation of the inactivated enzyme. In the presence of ATP and Mn2+, apparently no inactivation was seen (11). This suggests continual displace

ment of the inactivated coenzyme moiety by the added intact coenzyme during the

glycerol dehydration reaction. If adeninylbutylcobalamin is added to the re activation mixture, this analog is bound to the enzyme very tightly even in the presence of ATP and Mn2+. Thus, this analog competes with adenosylcobalamin for the active site of the enzyme, and therefore, the addition thereof brings about

the rapid cessation of the glycerol dehydration reaction. When 1, 2-propanediol is used as substrate, it does not cause significant inactivation of the enzyme.

Therefore, such displacement does not occur, or occurs at a much lower probability. As a result, even when adeninylbutylcobalamin is added to the reaction system, it inhibits the enzyme only slowly during the 1, 2-propanediol dehydration reaction.

Reactivation in situ of the other types of inactivated glycerol and diol dehydratases Incubation of the holoenzymes of diol dehydratase and glycerol dehydratase in the absence of substrate but in the presence of oxygen leads total inacti

Fig. 4. Reactivation in situ of the oxygen-inactivated glycerol dehydratase. The oxygen-inactivated glycerol dehydratase was obtained by incubation of toluene treated cells (0.015mg of dry cells) of K. pneumoniae ATCC 25955 at 37•Ž for 1hr in a mixture containing 0.05mmol of KCl, 0.035mmol of potassium phosphate buffer (pH 8.0) and 15nmol of adenosylcobalamin, in a total volume of 0.8ml. 1, 2 - Propanediol (0.2mmol) was then added, and the time course of the propional dehyde formation at 37•Ž was measured with (•£) or without (•œ) 3ƒÊmol of ATP and 3ƒÊmol of MnCl2 (final volume, 1.0ml). Apoenzyme controls in situ; with (•¢) or without (•›) ATP and MnCl2.

J. Nutr. Sci. Vitaminol. REACTIVATION OF B12 ENZYMES 235 vation (10, 21). This inactivation is believed to result from the irreversible cleavage of the activated carbon-cobalt bond of the coenzyme by reaction with oxygen. Figure 4 shows that the oxygen-inactivated glycerol dehydratase was also re activated in situ in the presence of ATP, Mn2+ and adenosylcobalamin, although the rate of reactivation was rather slow. This slow rate may be attributed at least in part to damage to the apoprotein itself, to extraordinarily tight binding of the compound derived from the coenzyme, or to denaturation of the reactivating system during the oxygen-inactivation procedure. It is known that diol dehydratase is gradually inactivated during dehydration of 1,2-ethanediol (1, 3). Diol dehydratase also undergoes gradual inactivation during catalysis when arabino-adenosylcobalamin is used as a coenzyme instead of adenosylcobalamin (17). The inactivated enzymes thus obtained were also re activated in situ by ATP, Mn2+ and adenosylcobalamin, although the data are not shown here.

This work was supported in part by a Grant-in-Aid for Encouragement of Young Scientists (1979), No. 41934 (to T. T.), from the Ministry of Education, Science and Culture of Japan.

REFERENCES

1) Lee, H. A., Jr., and Abeles, R. H. (1963): Purification and properties of dioldehydrase,

an enzyme requiring a cobamide coenzyme. J. Biol. Chem., 238, 2367-2373.

2) Toraya, T., Shirakashi, T., Kosuga, T., and Fukui, S. (1976): Substrate specificity of

coenzyme B12-dependent diol dehydrase: glycerol as both a good substrate and a potent

inactivator. Biochem. Biophys. Res. Commun., 69, 475-480.

3) Bachovchin, W. W., Eagar, R. G., Jr., Moore, K. W., and Richards, J. H. (1977):

Mechanism of action of adenosylcobalamin: glycerol and other substrate analog as

substrate and inactivators for propanediol dehydratase kinetics, stereospecificity,

and mechanism. Biochemistry, 16, 1082-1092.

4) Smiley, K. L., and Sobolov, M. (1962): A cobamide-requiring glycerol dehydrase from

an acrolein-forming Lactobacillus. Arch. Biochem. Biophys., 97, 538-543.

5) Pawelkiewicz, J., and Zagalak, B. (1965): Enzymic conversion of glycerol into ƒÀ - hydroxypropionaldehyde in a cell-free extract from Aerobacter aerogenes. Acta

Biochim. Pol., 12, 207-218.

6) Schneider, Z., Larsen, E. G., Jacobson, G., Johnson, B. C., and Pawelkiewicz, J. (1970): Purification and properties of glycerol dehydrase. J. Biol. Chem., 245, 3388-3396.

7) Toraya, T., Kuno, S., and Fukui, S. (1980): Distribution of coenzyme B12-dependent diol dehydratase and glycerol dehydratase in selected genera of Enterobacteriaceae and

Propionibacteriaceae. J. Bacteriol., 141, 1439-1442.

8) Toraya, T., and Fukui, S. (1977): Immunochemical evidence for the difference between coenzyme-B12-dependent diol dehydratase and glycerol dehydratase. Eur. J. Biochem.,

76, 285-289.

9) Poznanskaya, A. A., Yakusheva, M. I., and Yakovlev, V. A. (1977): Study on the mechanism of action of adenosylcobalamin-dependent glycerol dehydratase from

Aerobacter aerogenes. II. The inactivation kinetics of glycerol dehydratase complexes

with adenosylcobalamin and its analogs. Biochim. Biophys. Acta, 484, 236-243.

Vol. 28, No. 3, 1982 236 K. USHIO et al.

10) Schneider, Z., and Pawelkiewicz, J. (1966): The properties of glycerol dehydratase isolated from Aerobacter aerogenes, and the properties of the apoenzyme subunits . Acta Biochim. Pol.,13, 311-328. 11) Honda, S., Toraya, T., and Fukui, S. (1980): In situ reactivation of glycerol-inactivated coenzyme B12-dependent enzymes, glycerol dehydratase and diol dehydratase . J. Bacteriol., 143, 1458-1465. 12) Bachovchin, W. W., Moore, K. W., and Richards, J. H. (1978): Mechanism of action of adenosylcobalamin: hydrogen transfer in the inactivation of diol dehydratase by glycerol. Biochemistry, 17, 2218-2224. 13) Dolphin, D., Johnson, A. W., and Rodrigo , R. (1964): Reactions of alkylcobalamins. J. Chem. Soc., 3186-3193. 14) Pratt, J. M. (1964): The chemistry of vitamin B12. Part II. Photochemical reactions. J. Chem. Soc., 5154-5160. 15) Toraya, T., Honda, S., Kuno, S., and Fukui , S. (1978): Coenzyme B12-dependent diol dehydratase: regulation of apoenzyme synthesis in Klebsiella pneumoniae (Aerobacter aerogenes) ATCC 8724. J. Bacteriol.,135, 726-729. 16) Poznanskaja, A. A., Tanizawa, K., Soda, K., Toraya , T., and Fukui, S. (1979): Coenzyme B12-dependent diol dehydratase: purification , subunit heterogeneity, and reversible association. Arch. Biochem. Biophys., 194, 379-386. 17) Toraya, T., Ushio, K., Fukui, S., and Hogenkamp, H . P. C. (1977): Studies on the mechanism of the adenosylcobalamin-dependent diol dehydrase reaction by the use of analogs of the coenzyme. J. Biol. Chem., 252, 963-970 . 18) Forage, R. G., and Foster, M. A. (1979): Resolution of the coenzyme B12-dependent dehydratases of Klebsiella sp. and Citrobacter freundii . Biochim. Biophys. Acta, 569, 249-258. 19) Toraya, T., Kondo, M., Isemura, Y., and Fukui, S. (1972): Coenzyme B12-dependent propanediol dehydratase system. Nature of cobalamin binding and some properties of apoenzyme-coenzyme B12 analog complexes. Biochemistry, 11, 2599-2606. 20) Bernhauer, K., and Wagner, O. (1963): Synthesen auf dem Vitamin B12-Gebiet XVI. Uber die Cobinamid and Cobalamin-sulfonate sowie verwandte Verbindungen . Biochem. Z., 337, 366-382. 21) Wagner, O. W., Lee, H. A., Jr., Frey, P. A., and Abeles, R. H. (1966): Studies on the mechanism of action of cobamide coenzymes: chemical properties of the enzyme coenzyme complex. J. Biol. Chem., 241, 1751-1762.

J. Nutr. Sci. Vitaminol.