Mechanism of the ring contraction process in vitamin B 12 biosynthesis by the anaerobe Propionibacterium shermanii under aerobic conditions Katsumi Iida, Kuniaki Ohtaka and Masahiro Kajiwara Department of Medicinal Chemistry, Meiji Pharmaceutical University, Tokyo, Japan The anaerobic organism Propionibacterium shermanii was shown to produce vitamin B 12 from d-glucose as a carbon source under bubbling of N 2 gas in our previ- ous study [1]. This implies that d-aminolevulinic acid (ALA, a biosynthetic intermediate of tetrapyrrole) can be produced from d-glucose via the tricarboxylic acid cycle, an oxygen-dependent pathway, by P. shermanii. We were interested to know whether P. shermanii would be able to utilize oxygen from air instead, if it were cultured under aerobic conditions. Different mechanisms have been proposed for the ring contraction process in the biosynthetic pathway to vitamin B 12 [2–10]. In P. shermanii fed with [1- 13 C,1,1,4- 18 O 3 ]ALA under anaerobic conditions, 18 O-dilution of the 13 C 18 O-carbonyl oxygen of the ring A acetamide group of 13 C, 18 O-vitamin B 12 by oxygen from water in the medium led us to postulate that the ring contraction, which involves the migration of ring A, occurs via hydrolysis of our hypothetical d-lactone, which is proposed from Eschenmoser’s d-lac- tone [4], formed from the reaction of the ring A acetate group at C20 with methylation at C1 [5] as shown in the top row of Fig. 1. On the other hand, the isolation of factor IV, which was derived from Co-precorrin-4 as shown in Fig. 2, from P. shermanii cultured under anaerobic conditions led to the suggestion that ring contraction involves the migration of ring A after for- mation of the d-lactone from the ring A acetate group to C20, following the methylation of Co-precorrin-3 at C17 [6,7]. In the aerobe Pseudomonas denitrificans,an Keywords biosynthesis; 13 C-NMR; d-amino[1- 13 C,1,1,4- 18 O 3 ]levulinic acid; Propionibacterium shermanii; vitamin B 12 Correspondence K. Iida and M. Kajiwara, Department of Medicinal Chemistry, Meiji Pharmaceutical University, 2-522-1 Noshio, Kiyose-shi, Tokyo 204-8588, Japan Fax: +81 424 95 8612 Tel: +81 424 95 8611 E-mail: iida@my-pharm.ac.jp or kajiwara@my-pharm.ac.jp (Received 18 April 2007, revised 14 May 2007, accepted 14 May 2007) doi:10.1111/j.1742-4658.2007.05880.x The mechanism of the ring contraction process during vitamin B 12 biosyn- thesis by the anaerobe Propionibacterium shermanii was investigated under both aerobic and anaerobic conditions by means of feeding experiments with d-amino[1- 13 C]levulinic acid (a biosynthetic intermediate of tetra- pyrrole) and d-amino[1- 13 C,1,1,4- 18 O 3 ]levulinic acid in combination with 13 C-NMR spectroscopy. We showed that the characteristic mechanism of the ring contraction process (the generation of precorrin-3x from formation of the c-lactone from the ring A acetate group at C1 and hydroxylation at C20 by molecular oxygen catalyzed by CobG, and the migration of ring D by cleavage of the carbon–oxygen bond at C1 of precorrin-3x) in the aer- obe Pseudomonas denitrificans was not seen in P. shermanii under aerobic conditions, and the mechanism of the ring contraction process in P. sher- manii was the same irrespective of the presence or absence of oxygen. Abbreviation ALA, d-aminolevulinic acid. FEBS Journal 274 (2007) 3475–3481 ª 2007 The Authors Journal compilation ª 2007 FEBS 3475 aerobic feeding experiment with [4- 13 C,1,1,4- 18 O 3 ]ALA showed that precorrin-3x had the c-lactone formed from the ring A acetate group at C1 and a hydroxyl group generated from molecular oxygen, catalyzed by CobG, at C20, whereas an aerobic feeding experiment with [1- 13 C,1,1,4- 18 O 3 ]ALA revealed no 18 O-dilution of the 13 C 18 O-carbonyl oxygen of the ring A acetate group of 13 C, 18 O-precorrin-5x methyl ester. These results led to the suggestion that ring contraction involved 13 C, 18 O-precorrin-4 formed from the migra- tion of ring D followed by cleavage of the carbon–oxy- gen bond of the c -lactone at C1 of 13 C, 18 O-precorrin- 3x [8–10] as shown in the lower row of Fig. 1. We were interested in the mechanism of cor- rin formation by the anaerobe P. shermanii in an oxygen-containing atmosphere. Therefore, we conduc- ted aerobic feeding experiments with [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA in P. shermanii.We also repeated our previous [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA feeding experiments [5] to con- firm the mechanism of the anaerobic ring contraction process in P. shermanii for comparison with the result of the aerobic feeding experiments. Results and Discussion Aerobic feeding experiments with [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA in P. shermanii 13 C-Vitamin B 12 (5.3 mg) and 13 C, 18 O-vitamin B 12 (4.3 mg) were isolated from cultures of P. shermanii aerobically cultivated in phosphate buffer containing [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA, respectively. Their purity was judged to be high, based on a compar- ison of the 1 H-NMR and UV spectra with those of authentic vitamin B 12 . The magnified 13 C-NMR spec- tra of 13 C-vitamin B 12 and 13 C, 18 O-vitamin B 12 are shown in Fig. 3A,B. In Fig. 3A, seven 13 C-enriched singlet signals (175.347, 175.699, 176.270, 176.395, 177.654, 178.409, and 178.724) can be observed. In Fig. 3B, seven pairs of 13 C-enriched singlet signals (175.318 and 175.347, 175.662 and 175.691, 176.233 and 176.263, 176.351 and 176.387, 177.618 and 177.647, 178.358 and 178.394, and 178.687 and 178.716), can be observed. By comparison of Fig. 3A with Fig. 3B, seven 13 C-enriched singlet signals (175.347, 175.699, 176.270, Fig. 1. The highlights of previously postulated mechanisms of the ring contraction process in the biosynthesis of vitamin B 12 in the anaerobe P. shermanii (top row) and the aerobe P. denitrificans (lower row) obtained by previous [1- 13 C,1,1,4- 18 O 3 ]ALA feeding experiments. Ring contraction mechanism of vitamin B 12 K. Iida et al. 3476 FEBS Journal 274 (2007) 3475–3481 ª 2007 The Authors Journal compilation ª 2007 FEBS 176.395, 177.654, 178.409, and 178.724) in Fig. 3A can be assigned to seven 13 C-amide carbons (C57, C38, C61, C27, C43, C32, and C50, respectively) derived from the 13 C-carbonyl carbon of [1- 13 C]ALA. Seven downfield 13 C-enriched singlet signals (175.347, 175.691, 176.263, 176.387, 177.647, 178.394, and 178.716) of each pair of signals in Fig. 3B can also be assigned to seven 13 C-amide carbons (C57, C38, C61, C27, C43, C32, and C50, respectively) derived from 13 CO-carbonyl carbons, {- 13 C(¼O)- 18 O-} and {- 13 C(¼O)-O-}, present to an extent of approximately 20% in [1- 13 C,1,1,4- 18 O 3 ]ALA. The other seven 13 C-enriched singlet signals (175.318, 175.662, 176.233, 176.351, 177.618, 178.358, and 178.687), which were shifted upfield from the above seven 13 C-enriched singlet signals by 2.21–2.76 Hz owing to the a-isotope effect of 18 O (Table 1), of each pair of signals in Fig. 3B can be assigned to seven 13 C- amide carbons (C57, C38, C61, C27, C43, C32, and C50, respectively, 13 C 18 O-amide carbons) bearing 18 O. Comparison of the signal intensity or half-width of the 13 C-enriched signal of the 13 CO-carbonyl carbon with that of the 13 C 18 O-carbonyl carbon of each 13 C 18 O-amide group of 13 C, 18 O-vitamin B 12 derived from [1- 13 C,1,1,4- 18 O 3 ]ALA in the 13 C-NMR spectrum (Fig. 3B) gave the 18 O-retention ratio for each 13 C 18 O- amide group in 13 C, 18 O-vitamin B 12 , as summarized in Table 1. The average 18 O-retention ratio of the 13 C 18 O-amide groups (C57, C38, C61, C43, C32, and C50) was 76%. Namely, the 13 C 18 O-carbonyl oxygens of [1- 13 C,1,1,4- 18 O 3 ]ALA used were completely trans- ferred to the 13 C 18 O-amide oxygens of 13 C, 18 O-vitamin B 12 , as they have a similar 18 O-ratio. However, the 18 O-retention ratio of the 13 C 18 O-amide group (C27) of the ring A was 34%, representing a decrease of 58% from the 18 O-retention ratio of the 13 C 18 O-carbo- nyl oxygen of [1- 13 C,1,1,4- 18 O 3 ]ALA. Our previous anaerobic [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA feeding experiments in P. shermanii [5] repeated Our previous [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA feeding experiments [5] were repeated to investigate the 18 O-retention ratio for each 13 C 18 O-amide group in 13 C, 18 O-vitamin B 12 anaerobically biosynthesized by P. shermanii. 13 C-Vitamin B 12 (3.3 mg) and 13 C, 18 O- vitamin B 12 (4.0 mg) were isolated from cultures of P. shermanii anaerobically cultivated in phosphate Fig. 2. The mechanisms of the ring contraction process in the biosynthesis of vitamin B 12 postulated on the basis of the isolation of factor IV from the anaerobe P. shermanii. K. Iida et al. Ring contraction mechanism of vitamin B 12 FEBS Journal 274 (2007) 3475–3481 ª 2007 The Authors Journal compilation ª 2007 FEBS 3477 buffer containing [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA, respectively. Their purity was judged to be high by comparison of the 1 H-NMR and UV spectra with those of authentic vitamin B 12 . As shown in Table 1, seven 13 C-enriched singlet signals (175.347, 175.691, 176.270, 176.395, 177.647, 178.401, and 178.724) can be assigned to seven 13 C-amide carbons (C57, C38, C61, C27, C43, C32, and C50, respectively) in the 13 C-NMR spectrum of 13 C-vitamin B 12 . Seven 13 C-enriched singlet signals (175.347, 175.699, 176.270, 176.395, 177.655, 178.409, and 178.724), which were assigned to 13 CO-amide carbons (C57, C38, C61, C27, C43, C32, and C50, respectively), as well as seven 13 C-enriched singlet signals (175.318, 175.669, 176.241, 176.365, 177.625, 178.380, and 178.694), which were assigned to 13 C 18 O-amide carbons (C57, C38, C61, C27, C43, C32, and C50, respectively) shifted upfield by 2.21 Hz (Table 1) owing to the 18 O a-isotope effect, can be observed in the 13 C-NMR spectrum of 13 C, 18 O-vitamin B 12 . These 13 C-NMR data for 13 C- vitamin B 12 and 13 C, 18 O-vitamin B 12 biosynthesized in anaerobic [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA feed- ing experiments were essentially identical to those that we reported previously [5]. The 18 O-retention ratio for each 13 C 18 O-amide group in 13 C, 18 O-vitamin B 12 is also summarized in Table 1. The average 18 O-retention ratio of the 13 C 18 O-amide groups (C57, C38, C61, C43, C32, and C50) was 81%, and so the 13 C 18 O-carbonyl oxygens of [1- 13 C,1,1,4- 18 O 3 ]ALA used were completely trans- ferred to the 13 C 18 O-amide oxygens of 13 C, 18 O-vitamin B 12 , which have the same 18 O-retention ratio. The 18 O- retention ratio of the 13 C 18 O-amide group (C27) of the ring A was 26%, representing a 68% decrement from Fig. 3. Magnified 13 C-NMR spectra of 13 C-amide carbons of (A) 13 C-vitamin B 12 biosynthesized from [1- 13 C]ALA and (B) 13 C, 18 O-vitamin B 12 aerobically biosynthe- sized from [1- 13 C,1,1,4- 18 O 3 ]ALA by the anaerobe P. shermanii. Ring contraction mechanism of vitamin B 12 K. Iida et al. 3478 FEBS Journal 274 (2007) 3475–3481 ª 2007 The Authors Journal compilation ª 2007 FEBS the 18 O-retention ratio of the 13 C 18 O-carbonyl oxygen of [1- 13 C,1,1,4- 18 O 3 ]ALA. Mechanism of the ring contraction process in the biosynthesis of vitamin B 12 by P. shermanii Two mechanisms of the ring contraction process in the anaerobe P. shermanii have been suggested [5–7]. One was based on the results of our anaerobic [1- 13 C,1,1,4- 18 O 3 ]ALA feeding experiment in P. sher- manii [5], shown in the top row of Fig. 1. The other was based on the isolation of factor IV from anaerobi- cally cultivated P. shermanii by Scott et al. [6] and Wang et al. [7], shown in Fig. 2. 18 O-Dilution of the 13 C 18 O-carbonyl (C27) oxygen of the ring A acetamide group in 13 C, 18 O-vitamin B 12 biosynthesized via these two postulated anaerobic mechanisms should be detectable. Meanwhile, another mechanism for the ring contraction process by the aerobe P. denitrificans had been proposed from the results of aerobic [4- 13 C,1,1,4- 18 O 3 ]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA feed- ing experiments. According to this postulated mechanism in P. denitrificans,no 18 O-dilution of the 13 C 18 O-carbonyl (C27) oxygen of the ring A acetate group should be observed [8–10], as can be seen in the lower row of Fig. 1. Therefore, examination of the 13 C 18 O-carbonyl (C27) oxygen of the ring A acetate group in 13 C, 18 O-vitamin B 12 biosynthesized in [1- 13 C,1,1,4- 18 O 3 ]ALA feeding experiments should allow us to distinguish between the proposed aerobic and anaerobic mechanisms of the ring contraction process. When the 18 O-retention ratios of the 13 C 18 O-amide groups of 13 C, 18 O-vitamin B 12 biosynthesized aerobi- cally and anaerobically in P. shermanii are compared (Table 1), the 18 O-retention ratios are almost identical. These results show that vitamin B 12 was biosynthesized via the same biosynthetic pathways under both aerobic and anaerobic conditions in this organism. Further- more, 58% (aerobically) and 68% (anaerobically) 18 O- retention ratio decrements of the 13 C 18 O-amide (C27) oxygen in the ring A show that the mechanism of the ring contraction process in P. shermanii is the same irrespective of the presence or absence of oxygen. In other words, although we did not confirm the pres- ence of CobG in P. shermanii, we can conclude that the characteristic aerobic process (hydroxylation at C20 by molecular oxygen catalyzed by CobG) does not occur in P. shermanii, even when molecular oxygen is present. Table 1. 13 C-NMR data of 13 C-vitamin B 12 and 13 C, 18 O-vitamin B 12 biosynthesized from [1- 13 C]ALA or [1- 13 C,1,1,4- 18 O 3 ]ALA by the anaerobe P. shermanii under aerobic or anaerobic conditions. Carbon c Origin d Conditions of feeding experiments Aerobic Anaerobic 13 C-B 12 a13 C, 18 O-B 12 b13 C-B 12 a13 C, 18 O-B 12 b d C d C Value (Hz) e Ratio (%) f d C d C Value (Hz) e Ratio (%) f 50 13 CO 178.724 178.716 – – 178.724 178.724 – – 50 13 C 18 O – 178.687 2.21 78 – 178.694 2.21 82 32 13 CO 178.409 178.394 – – 178.401 178.409 – – 32 13 C 18 O – 178.358 2.76 78 – 178.380 2.21 83 43 13 CO 177.654 177.647 – – 177.647 177.655 – – 43 13 C 18 O – 177.618 2.21 75 – 177.625 2.21 81 27 13 CO 176.395 176.387 – – 176.395 176.395 – – 27 13 C 18 O – 176.351 2.76 34 – 176.365 2.21 26 61 13 CO 176.270 176.263 – – 176.270 176.270 – – 61 13 C 18 O – 176.233 2.21 76 – 176.241 2.21 83 38 13 CO 175.699 175.691 – – 175.691 175.699 – – 38 13 C 18 O – 175.662 2.21 69 – 175.669 2.21 74 57 13 CO 175.347 175.347 – – 175.347 175.347 – – 57 13 C 18 O – 175.318 2.21 77 – 175.318 2.21 82 a13 C-B 12 represents 13 C-vitamin B 12 biosynthesized from [1- 13 C]ALA. b13 C, 18 O-B 12 represents 13 C, 18 O-vitamin B 12 biosynthesized from [1- 13 C,1,1,4- 18 O 3 ]ALA. c Carbon represents carbon position of vitamin B 12 . d Origin indicates the origin of 13 C-enriched singlet signals assigned to 13 C-amide carbons, and 13 CO and 13 C 18 O represent 13 CO-amide and 13 C 18 O-amide groups, respectively. e Value indicates upfield shift owing to 18 O a-isotope effect in hertz. f Ratio is the 18 O-retention ratio of 13 C 18 O-amide oxygen in 13 C 18 O-vitamin B 12 (5). These ratios were calculated from a comparison of the intensities or half-widths of 13 C-enriched signals of 13 CO-amide and 13C18O -amide carbons in the 13 C-NMR spectrum of 13 C, 18 O-vitamin B 12 . K. Iida et al. Ring contraction mechanism of vitamin B 12 FEBS Journal 274 (2007) 3475–3481 ª 2007 The Authors Journal compilation ª 2007 FEBS 3479 Conclusions Aerobic and anaerobic feeding experiments with [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA in P. shermanii were consistent with the anaerobic ring contraction mechanisms of vitamin B 12 suggested by Kurumaya et al. [5], Scott et al. [6] and Wang et al. [7], irrespect- ive of the presence or absence of molecular oxygen. The characteristic aerobic processes of hydroxylation at C20 by molecular oxygen catalyzed by CobG, for- mation of the c-lactone from the ring A acetate group at C1, and migration of ring D by cleavage of the car- bon–oxygen bond at C1, observed in P. denitrificans, do not operate in P. shermanii even in the presence of oxygen. Experimental procedures Organism, chemicals, instruments, and equipment The bacterium used was Propionibacterium freudenreichii ssp. shermanii (ATCC 9614). [1- 13 C]ALA was synthesized according to our method [11], via the hydrolysis of ethyl 3-ethoxycarbonyl-5-phthalimido[1- 13 C]levulinate produced from the coupling of ethyl 4-phthalimidoacetoacetate and ethyl bromo[1- 13 C]acetate derived from sodium [1- 13 C]acetate (99 atom % 13 C), which was purchased from Cambridge Isotope Laboratories (Andover, MA, USA). [1- 13 C,1,1,4- 18 O 3 ]ALA was synthesized by heating [1- 13 C]ALA with [ 18 O]water (95–98 atom % 18 O; purchased from Cambridge Isotope Laboratories) in the presence of an acidic catalyst in a sealed tube, as we have described previously [5]. The 18 O-retention ratios at C1 of [1- 13 C,1,1,4- 18 O 3 ]ALA were shown to be {- 13 C(¼ 18 O)- 18 O-} ⁄ [{- 13 C(¼ 18 O)-O-} and {- 13 C(¼O)- 18 O-}] ⁄ {- 13 C(¼O)-O-} ¼ 68 : 26 : 6 by analysis of the 13 C-NMR signals. Authentic vitamin B 12 was purchased from Glaxo Operations UK Ltd (Greenford, UK). All other chemicals were of analytical grade. All 1 H-NMR (300 MHz) and 13 C-NMR (75 MHz) spectra were recorded on a Varian Gemini-300 spectrometer (Varian, Inc., Palo Alto, CA, USA). UV spectra were recorded on a Jasco UVIDEC-610C spectrometer (Jasco Corp., Tokyo, Japan). The air pump was a Nisso INNO- b4000 (Nisso Co., Saitama, Japan). Aerobic or anaerobic feeding experiments with [1- 13 C]ALA and [1- 13 C,1,1,4- 18 O 3 ]ALA in P. shermanii Wet cells (approximately 200 g) of P. shermanii were culti- vated with bubbling of air (aerobic conditions) or with bub- bling of N 2 gas (anaerobic conditions), and harvested as previously described [1,5]. These cells, together with a solution of [1- 13 C]ALA or [1- 13 C,1,1,4- 18 O 3 ]ALA (75 mg) and l-methionine (150 mg) in water (20 mL), which had been filtered through a membrane filter (0.2 lm), a solution of 5,6-dimethylbenzimidazole (75 mg) in 80% ethanol (2 mL), and 50% d-glucose solution (9.6 mL), were added to 0.07 m sodium phosphate buffer (pH 7.0, 300 mL) con- taining CoCl 2 .6H 2 O (3 mg) in two conical flasks. These were incubated at 27 °C with bubbling of air (aerobic con- ditions) or with bubbling of N 2 gas (anaerobic conditions), and adjusted to pH 7.0 with 20% Na 2 CO 3 , and 50% d-glu- cose solution (19.2 mL) was added during the fermentation. After 3 days, 13 C-vitamin B 12 and 13 C, 18 O-vitamin B 12 were isolated from the pellet obtained by centrifugation of the culture broths, using the methods described in our previous papers [1,5]. 13 C-NMR measurements of 13 C-vitamin B 12 and 13 C, 18 O-vitamin B 12 13 C-NMR (75 MHz) spectra were recorded for solutions of 13 C-vitamin B 12 and 13 C, 18 O-vitamin B 12 in 2 H 2 O with 1,4-dioxane (67.4 p.p.m.) as an internal standard. The spec- tral width was 18 102.9 Hz with 65 536 data points, which corresponds to a resolution of 0.28 Hz per point. The deter- mined 10° pulse width was 2.3 ls, the acquisition time was 1.504 s, the pulse delay time was 0.496 s, and the number of scans was approximately 15 000. The assignments of 13 C-NMR signals of vitamin B 12 were carried out on the basis of reported data [1,5]. References 1 Iida K & Kajiwara M (2000) Evaluation of biosynthetic pathways to d-aminolevulinic acid in Propionibacterium shermanii based on biosynthesis of vitamin B 12 from d-[1- 13 C]glucose. Biochemistry 39, 3666–3670. 2 Blanche F, Cameron B, Crouzet J, Debussche L, Thibaut D, Vuilhorgne M, Leeper FJ & Battersby AR (1995) Vitamin B 12 : how the problem of its bio- synthesis was solved. Angew Chem Int Ed Engl 34, 383–411. 3 Scott AI, Roessner CA & Santander PJ (2003) Genetic and mechanistic exploration of the two pathways of vitamin B 12 biosynthesis. In The Porphyrin Handbook, Vol. XII (Kadish KM & Guilard R, eds), pp. 211–228. Academic Press, San Diego, CA. 4 Eschenmoser A (1988) Vitamin B 12 : experiments con- cerning the origin of its molecular structure. Angew Chem Int Ed Engl 27, 5–39. 5 Kurumaya K, Okazaki T & Kajiwara M (1989) Studies on the biosynthesis of corrinoids and porphyrinoids. I. The labeling of oxygen of vitamin B 12 . Chem Pharm Bull 37, 1151–1154. 6 Scott AI, Stolowich NJ, Wang J, Gawatz O, Fridrich E &Mu ¨ ller G (1996) Biosynthesis of vitamin B 12 : factor Ring contraction mechanism of vitamin B 12 K. Iida et al. 3480 FEBS Journal 274 (2007) 3475–3481 ª 2007 The Authors Journal compilation ª 2007 FEBS IV, a new intermediate in the anaerobic pathway. Proc Natl Acad Sci USA 93, 14316–14319. 7 Wang J, Stolowich NJ, Santander PJ, Park JH & Scott AI (1996) Biosynthesis of vitamin B 12 : concerning the identity of the two-carbon fragment eliminated during anaerobic formation of cobyrinic acid. Proc Natl Acad Sci USA 93, 14320–14322. 8 Scott AI, Roessner CA, Stolowich NJ, Spencer JB, Min C & Ozaki S (1993) Biosynthesis of vitamin B 12 : discov- ery of the enzymes for oxidative ring contraction and insertion of the fourth methyl group. FEBS Lett 331, 105–108. 9 Spencer JB, Stolowich NJ, Roessner CA, Min C & Scott AI (1993) Biosynthesis of vitamin B 12 : ring contraction is preceded by incorporation of molecular oxygen into precorrin-3. J Am Chem Soc 115, 11610–11611. 10 Spencer JB, Stolowich NJ, Santander PJ, Pichon C, Kajiwara M, Tokiwa S, Takatori K & Scott AI (1994) Mechanism of the ring contraction step in vitamin B 12 biosynthesis: the origin and subsequent fate of the oxy- gen functionalities in precorrin-3x. J Am Chem Soc 116, 4991–4992. 11 Kurumaya K, Okazaki T, Seido N, Akasaka Y, Kawaj- iri Y, Kajiwara M & Kondo M (1989) A facile synthesis of d-aminolevulinic acid (ALA) regioselectively labeled with 13 C and direct observation of enzymatic transfor- mation from ALA to porphobilinogen (PBG). J Label Compd Radiopharm 27, 217–235. K. Iida et al. Ring contraction mechanism of vitamin B 12 FEBS Journal 274 (2007) 3475–3481 ª 2007 The Authors Journal compilation ª 2007 FEBS 3481 . Mechanism of the ring contraction process in vitamin B 12 biosynthesis by the anaerobe Propionibacterium shermanii under aerobic conditions Katsumi. 2007) doi:10.1111/j.1742-4658.2007.05880.x The mechanism of the ring contraction process during vitamin B 12 biosyn- thesis by the anaerobe Propionibacterium shermanii was investigated under both