Biochimica et Biophysica Acta, 313 (1973) 150-155
~) Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands BBA 27172
C A R B O N M O N O X I D E P R O D U C T I O N F R O M H Y D R O X O C O B A L A M I N BY BACTERIA
ROLF R. ENGELa, SIV MODLERa, JOHN M. MATSEN b and ZBYSLAW J. PETRYKA~ a, bDepartment s of Pediatrics, bLaboratory Medicine and bMicrobiology, University of Minnesota Hospitals, Minneapolis, Minn. 55455 (U.S.A.) and cUniversity of Minnesota Medical Research Unit, Northwestern Hospital, Minneapolis, Minn. 55455 (U.S.A.)
(Received February 26th, 1973)
SUMMARY Microbial CO formation from vitamin B~2 analogues was investigated, since CO is a catabolic product of some other metal-containing cyclic tetrapyrroles. CO was evolved during aerobic incubation of hydroxocobalamin with either Bacillus cereus or ~-hemolytic Streptococcus mitis. The reaction resembles heme catabolism by these microbes in that CO was not formed under anaerobic conditions nor by a nonhemolytic strain of S. mitis. Cyanocobalamin partially inhibited CO formation from hydroxocobalamin.
INTRODUCTION Bacterial overgrowth of the small intestine can result in vitamin B 12 deficiency and in megaloblastic anemia. This association has prompted studies which show that enteric bacteria can compete in vivo t - a and in vitro 4 - 7 with intrinsic factor for the binding of vitamin B12. Synthetic and structural similarities between B12 and heme suggested the possibility that bacteria may also destroy B12 along catabolic pathways that resemble heme degradation s. In the case of heme catabolism by mammalian systems, the splitting of the porphyrin ring yields CO from the alpha methene bridge carbon 9-11 and molecular oxygen 12, as biliverdin and iron are released in equimolar amounts. The aerobic catabolism of various heme compounds by bacteria also resuits in CO formation s. Although the tetrapyrolle rings of cobalamin and heme are both synthesized by condensing porphobilinogen molecules, there are only three methene bridges in cobalamin as opposed to the four in heme ~3,14. The experiments reported here suggest that the microbial catabolism of B12 resembles heme degradation in generating CO as an aerobic byproduct. METHOD The experimental procedure is described in a previous report s on CO production from heme compounds by the same microbes. Vials containing 225 or 450
c o PRODUCTION FROM HYDROXOCOBALAMIN
nmoles o f crystalline h y d r o x o c o b a l a m i n or c y a n o c o b a l a m i n in 2 ml o f media were incubated in the dark at 37 °C, with either Bacillus cereus, ~-hemolytic Streptococcus mitis or a nonhemolytic S. mitis. After incubating individual vials for 18, 66 and 162 h in sealed glass syringes, evolved C O and methane were analyzed by introducing gas samples f r o m the syringe into a gas c h r o m a t o g r a p h . A l t h o u g h the analytic m e t h o d can detect less than 0.5 nmole o f C O in 100 ml o f air, no significance was ascribed to values o f less than 60 nmoles per week because sterile medium can evolve this m u c h CO, particularly if it has been freshly prepared and is exposed to light in an aerobic environment. RESULTS B. cereus and ~-hemolytic S. mitis always evolved more than 79 nmoles o f C O during a week o f aerobic incubation with h y d r o x o c o b a l a m i n (Table I). In contrast, control samples with substrate in sterile medium, or bacteria in substrate-free m e d i u m never accumulated more than 56 nmoles o f CO in 1 week. The nonhemolytic S. mitis did n o t raise C O levels significantly above sterile control samples. ~-Hemolytic S. mitis often formed more C O after an overnight incubation than B. cereus. However, by 7 days B. cereus f o r m e d the m o s t C O and still grew on subcultures, whereas the two strains o f S. mitis progressively lost viability between the 18-h and 7-day subcultures. U n d e r anaerobic conditions none o f these bacteria formed C O f r o m hyd r o x o c o b a l a m i n despite flocculant growth. C O accumulation remained at control levels when c y a n o c o b a l a m i n was substituted for h y d r o x o c o b a l a m i n and it was attenuated when the two substrates were incubated together with either o f the hemolytic microbes (Fig. 1).
TABLE I CO PRODUCTION FROM HYDROXOCOBALAMIN BY HEMOLYTIC BACTERIA Bacteria
Nonhemolytic S. mitis
~-Hemolytic S. mit~ B. cereus
0 225 450 0 225 450 0 225 450 0 225 450
CO accumulation* (nmoles) After 18 h incubation
After 7 days incubation
8 7 10 8 6 16 8 42 49 7 30 47
2 12 15 7 14 19 4 24 41 6 14 15
3 4 8 3 8 14 3 16 10 1 1 7
43 34 39 55 52 56 48 113 120 38 428 637
28 41 51 30 50 48 23 86 136 24 80 172
17 30 28 8 28 33 12 52 55 3 3 39
* Each value is the mean for three or more experiments. All control samples evolved less than 60 nmoles of CO during incubation for 7 days at 37 °C.
R. R. ENGEL et al.
Mean + I
.c _~ 4OO
Hydroxocobolomin Gyanocobalomin Hydtoxocobolomin Hydroxocobalomin 225 n moles
225 n moles
plus Cyonocobolomin 225 n moles of eoch
450 n moles
Fig. 1. B. cereus and or-hemolyticS. mitis formed CO from hydroxocobalamin but not from cyanocobalamin, during 1 week of incubation at 37 °C in 02. The addition of 225 nmoles of cyanocobalamin to an equal amount of hydroxocobalamin decreased CO formation by these microbes below levels observed with either 225 or 450 nmoles of hydroxocobalamin alone. The color of the post-incubation mixture suggests that CO production was always accompanied by hydroxocobalamin destruction. When 225 or 450 nmoles of hydroxocobalamin were added to the golden yellow culture medium, the solution became red. Nonhemolytic S. m i t i s did not alter this color after 1 week of aerobic incubation, but a-hemolytic S. m i t i s returned the less concentrated samples to a golden yellow color and the more concentrated samples became orange. After aerobic incubation with B. cereus, samples with the two concentrations of hydroxocobalamin became light and dark orange. Under anaerobic conditions each of the three bacterial strains developed an orange color from the red hydroxocobalamin solutions. To confirm that CO formation was coupled with substrate utilization, absorption spectra were obtained on the supernatant from the post-incubation mixture (Fig. 2). Samples containing 250 nmoles of hydroxocobalamin displayed absorption maxima around 530 and 504 nm before and after 1 week of sterile aerobic or anaerobic incubation at 37 °C. In 0 2 B. cereus and or-hemolytic S. m i t i s eliminated these maxima, whereas the nonhemolytic S. m i t i s did not alter them. In N 2 the two strains of S. m i t i s eliminated the above absorption maxima more rapidly than the B. cereus. All combinations that reduced the 530 and 504 nm peaks also formed a transient peak around 470 nm, which was retained to a variable degree depending on the gas phase, sub-
CO PRODUCTION FROM HYDROXOCOBALAMIN 0-~
I t I I
Non-hemolytic Sffep. mi/is
[' ' ' ] i ' ' ' Alpho hemo/yl/c ~ / i Sffepm#/s / B cereus
f \ l~ / I//
co N 6©
80 i I0©
Fig. 2. Absorption spectra of hydroxocobalamin in Todd-Hewitt medium after aerobic and an aerobic incubation with and without bacteria for 7 days. Initially each sample contained 225 nmoles of hydroxocobalamin in 2 ml of Todd-Hewitt broth. This medium has progressively increasing absorption below wavelength 650 rim, corresponding to the two curves after incubation with ~-hemolytie S. mitis. Variable dilution of the supernatants, with water, has introduced shifts in the vertical position of the above absorption spectra. strate concentration, incubation period and strain of bacteria. Helgeland e t aL 15 have also observed microbial derivatives of Blz with absorption maxima at 474 nm. Possibly because these investigators used cyanocobalamin, B . c e r e u s did not form this pigment and none of their organisms converted more than 3 % of the B12 to this pigment. The present finding of a transitory absorption maximum around 474 nm in the absence of CO formation could correspond to their derivative which contained CO and also accumulated primarily anaerobically. Doubling the hydroxocobalamin content to 450 nmoles or reducing the incubation period to 3 days produced transitional changes in the absorption spectra. Incubations in air resulted in absorption spectra that were intermediate between those obtained in 02 and N 2 with each microbe. The more rapid initial CO production by ~-hemolytic S. m i t i s than by B. c e r e u s is matched by greater changes in the absorption spectra produced by ~-hemolytic S. m i t i s than by B. c e r e u s during the first day. N o changes were observed in the absorption spectrum and color of cyanocobalamin solutions with each of these bacteria, which is in accord with the absence of CO production from this substrate. DISCUSSION The data provide presumptive evidence that CO is a byproduct in the bacterial catabolism of hydroxocobalamin. This finding invites more definitive studies to determine whether the reaction is analogous to the expulsion of a methene carbon of heine by mammalian tissues. The analogy between heme and cobalamin catabolism
R.R. ENGEL et al.
is reinforced by similar results that were obtained for both of these substrates with three bacteria strains a. Thus nonhemolytic S. m i t i s did not make appreciable amounts of CO f r o m red cells, hemoglobin, myoglobin, cytochrome c, heme or hydroxocobalamin. In contrast, a-hemolytic S. m i t i s and B. c e r e u s generated CO from each of these substrates aerobically. The absence of CO formation in N 2 is consistent with reports 12 on mammalian heme oxygenase where molecular oxygen and not the oxygen in water is incorporated into CO during heme catabolism. The absorption spectra indicate that in N 2 and in air, substrate disappearance can exceed CO accumulation. Evidence for heme catabolism in excess of CO production has also been obtained with these bacteria s and with mammalian systems 9-11. Mammalian heme oxygenase can be induced by heme 4 and inhibited by mesohemin IX, deuterohemin IX, coprohemin IX, K C N and CO 12. Possibly the initial delay in CO production by B. cereus and the inhibiting effect of cyanocobalamin also represent substrate induction and analog inhibition of a bacterial enzyme system. The only combination yielding more than 1 mole of CO per mole of hydroxocobalamin was B . c e r e u s in 02. It is uncertain whether this extra CO is derived from hydroxocobalamin catabolism, since enhanced turnover of other compounds has not been excluded. I f the alpha or g a m m a methene bridge carbons of hydroxocobalamin are converted to CO, then the fate of the attached methyl groups 14 also remains to be determined. Methane did not accumulate in these experiments. ~-Hemolytic S. m i t i s may have a different mechanism for cleaving cyclic tetrapyrroles than B. cereus. Thus the time of maximum CO production from hydroxocobalamin (Table I) and heine compounds 8 is earlier for a-hemolytic S. m i t i s than for B. cereus. Also, B. c e r e u s has strong catalase activity whereas a-hemolytic S. m i t i s (but not the nonhemolytic S. m i t i s ) produces H 2 0 z. The addition of H202 to sterile solutions of hydroxocobalamin or heme also resulted in CO formation. ACKNOWLEDGEMENTS Supported by U.S. Public Health Service Grant RO1-HDO4487 from the National Institute of Child Health and H u m a n Development. Crystalline hydroxocobalamin and cyanocobalamin were supplied through the courtesy of Mr Robert W. Wilson of Eli Lilly and Company, Indianapolis, Indiana.
REFERENCES 1 2 3 4 5 6 7 8 9
Strauss, E. W., Donaldson, Jr, R. M. and Gardner, F. H. (1961) Lancet 2,736-738 Booth, C. C. and Heath, J. (1962) Gut 3, 70-73 Donaldson, Jr, R. M. (1962) Gastroenterology 43,271-281 Schjonsby, H., Peters, T. J., Hoffbrand, A. V. and Tabaqchali, S. (1970) Gut 11,371 Hoff-Jorgensen, E. (1952) Arch. Biochem. Biophys. 36, 235-236 Donaldson, Jr, R. M., Corrigan, H. and Natsios, G. (1962) Gastroenterology 43,282-290 Dellipiani, A. W., Samson, R. R. and Girdwood, R. H. (1968) Am. J. Dig. Dis. 13, 718-726 Engel, R. R., Matsen, J. M., Chapman, S. S. and Schwartz, S. (1972) J. Bacteriol. 112, 1310-1315 Pimstone, N. R., Tenhunen, R., Seitz, P. T., Marver, H. S. and Schmid, R. (1971) J. Exp. Med. 133, 1264-1281 10 Landaw, S. A., Callahan, Jr, E. W. and Schmid, R. (1970) J. Clin. Invest. 49, 914-925 11 Coburn, R. F., Williams, W. J., White, P. and Kahn, S. B. (1967) J. Clin. Invest. 46, 346-356
CO P R O D U C T I O N F R O M H Y D R O X O C O B A L A M I N 12 13 14 15
Tenhunen, R., Marver, H. S. and Schmid, R. (1969) J. Biol. Chem. 244, 6388-6394 Schwartz, S., Ikeda, K., Miller, I. M. and Watson, C. J. (1959) Science 129, 40-41 Burnham, B. F. (1969) in Metabolic Pathways (Greenberg, D. M., ed.), Vol. III, p. 404, New York Helgeland, K., Jonsen, J. and Laland, S. (1961) Biochem. J. 81,260-265