Glutathione-dependent One-electron Transfer Reactions Catalyzed by a B12 Trafficking Protein*

Background: CblC is a B12-processing enzyme that converts cyanocobalamin to cob(II)alamin. Results: CblC catalyzes two previously unknown reactions: GSH-dependent reductive decyanation of CNCbl and reduction of OH2Cbl. Conclusion: Worm and human CblC activities show different susceptibilities to oxygen. Significance: The use of GSH for single-electron transfer is unexpected and expands the range of catalytic reactions supported by CblC. CblC is involved in an early step in cytoplasmic cobalamin processing following entry of the cofactor into the cytoplasm. CblC converts the cobalamin cargo arriving from the lysosome to a common cob(II)alamin intermediate, which can be subsequently converted to the biologically active forms. Human CblC exhibits glutathione (GSH)-dependent alkyltransferase activity and flavin-dependent reductive decyanation activity with cyanocobalamin (CNCbl). In this study, we discovered two new GSH-dependent activities associated with the Caenorhabditis elegans CblC for generating cob(II)alamin: decyanation of CNCbl and reduction of aquocobalamin (OH2Cbl). We subsequently found that human CblC also catalyzes GSH-dependent decyanation of CNCbl and reduction of OH2Cbl, albeit efficiently only under anaerobic conditions. The air sensitivity of the human enzyme suggests interception by oxygen during the single-electron transfer step from GSH to CNCbl. These newly discovered GSH-dependent single-electron transfer reactions expand the repertoire of catalytic activities supported by CblC, a versatile B12-processing enzyme.

Vitamin B 12 is a derivative of cobalamin, an essential cofactor involved in sulfur, branched-chain amino acid, odd-chain fatty acid, and cholesterol metabolism. Methylcobalamin (MeCbl) 2 and 5Ј-deoxyadenosylcobalamin (AdoCbl) are the two active forms of the cofactor that are required by the B 12 -dependent enzymes, methionine synthase, and methylmalonyl-CoA mutase, respectively (1). Because mammals are unable to synthesize B 12 , they possess an elaborate pathway for converting dietary cobalamins to the active cofactor forms and delivering them to target proteins (2)(3)(4). Defects in the trafficking proteins result in a functional B 12 deficiency and lead to cobalamin disorders that are classified into eight genetic complementation groups: cblA-F, cblJ, and cblX (5)(6)(7). Defects in the cobalamindependent enzymes, methionine synthase and methylmalonyl-CoA mutase, belong to the cblG (8 -10) and mut (11) genetic complementation groups, respectively.
Defects in the cblC locus affect both B 12 -dependent enzymes and result in combined methylmalonic aciduria and homocystinuria (12). This clinical phenotype indicates that the gene product of cblC (hereafter referred to as CblC) functions in the early and common part of the B 12 trafficking pathway before it bifurcates into the cytoplasmic (for methionine synthase) and mitochondrial (for methylmalonyl-CoA mutase) branches. The most common locus of mutations leading to cobalamin disorders is in the gene responsible for the cblC complementation group (13). Human CblC (also referred to as MMACHC or methylmalonic aciduria type C and homocystinuria) exhibits unusual and dual activities, catalyzing the flavin-dependent reductive decyanation of vitamin B 12 (or cyanocobalamin (CNCbl)) (Reaction 1) (14) and GSH-dependent dealkylation of alkylcobalamins (e.g. MeCbl or AdoCbl) (Reaction 2) (15). The products of reductive decyanation are cyanide and cob(II)alamin, which is paramagnetic, whereas a glutathione thioether and cob(I)alamin are formed in the dealkylation reaction. The superreactivity of cob(I)alamin results in its rapid oxidation to either cob(II)alamin or aquocobalamin (OH 2 Cbl) under aerobic conditions. pH. The crystal structures of human CblC (hCblC) revealed that the tail leading to dimethylbenzimidazole resides in a crevice on the surface of the protein (16,17).
Because the study of the cytoplasmic B 12 trafficking pathway has been hampered by the inability to express human methionine synthase, we have cloned the corresponding CblC-E-and CblG-encoding genes from Caenorhabditis elegans as a first step toward elucidating the interactions of the encoded proteins. Our characterization of the C. elegans CblC (ceCblC), which shares 36% identity and 53% similarity with hCblC, led to the unexpected finding that ceCblC catalyzes the reductive decyanation of CNCbl in the presence of GSH or other thiols, an activity that had been missed previously with hCblC. The decyanation product, cob(II)alamin, is stabilized by ceCblC. The discovery of this new catalytic activity of ceCblC led us to determine its relevance to hCblC. We report that hCblC also harbors a latent GSH-dependent decyanation activity, which is only readily observed under anaerobic conditions, indicating efficient interception of the electron transfer step by oxygen in the human versus the C. elegans CblC. Furthermore, both human and C. elegans CblC exhibit GSH-dependent OH 2 Cbl reductase activity, which might be advantageous for maintaining the cofactor in the cob(II)alamin oxidation state for subsequent transfer to an acceptor protein in the trafficking pathway.
The PCR products were subcloned using the ligation-independent cloning method into a ligation-independent cloning vector. A C-terminal His 6 tag was introduced into the ceCblC construct using primers 5Ј-GAGCACCACCACCACCACCA-CTAAGATCCCAACTCCATAAGGA-3Ј (forward) and 5Ј-GTGGTGGTGGTGGTGGTGCTCGAGGTCGATATTCTT-TGCTCCTCC-3Ј (reverse) to generate the expression construct, ceCblC-His 6 . MSR from C. elegans (ceMSR) was also subcloned into a ligation-independent cloning vector to generate the expression construct, His 6 -ceMSR.
Expression and Purification of ceCblC and ceMSR-Escherichia coli BL21 (DE3) was transformed with ceCblC-His 6 and grown overnight at 37°C in 100 ml of Luria Bertani medium containing ampicillin (100 g ml Ϫ1 ). Then, 6 ϫ 1 liter of the same medium containing ampicillin was inoculated with the starter culture and grown at 37°C. After 4 h, when the A 600 had reached 0.5-0.6, the temperature was reduced to 15°C. The cultures were induced with 100 M isopropyl ␤-D-1-thiogalactopyranoside, and the cells were harvested 16 h later. The cells pellets were stored at Ϫ80°C until use.
For purification of ceCblC, the cell pellets were suspended in 200 ml of buffer containing 100 mM Tris-HCl, pH 8.0, 150 mM KCl, 0.15 mg ml Ϫ1 lysozyme, 5% (v/v) glycerol, and two protease inhibitor cocktail tablets (Roche Applied Science). The cell suspension was stirred at 4°C for 20 min and then sonicated (power setting ϭ 6.5) on ice for 8 min at 15-s intervals separated by 45-s cooling periods. The sonicate was centrifuged at 38,000 ϫ g for 30 min, and the supernatant was loaded on to a Ni-NTA-agarose column (2.5 ϫ 5 cm, Qiagen) pre-equilibrated with Buffer A (100 mM Tris-HCl, pH 8.0, 150 mM KCl, and 5% (v/v) glycerol). The column was washed with 200 ml of a buffer containing 30 mM imidazole in Buffer A and eluted with 300 ml of a linear gradient ranging from 30 to 300 mM imidazole in Buffer A. The fractions containing ceCblC were identified by SDS-PAGE analysis, pooled and concentrated to 5 ml, and dialyzed overnight against 1 liter of dialysis buffer containing 100 mM HEPES, pH 7.0, 150 mM KCl, and 10% (v/v) glycerol. The dialyzed protein was loaded on to a Superdex 200 column (120 ml, GE Healthcare) pre-equilibrated with dialysis buffer. The fractions of interest were collected and stored at Ϫ80°C.
To purify His 6 -ceMSR, E. coli BL21 (DE3) was transformed with the His 6 -ceMSR expression vector and grown overnight at 37°C in 100 ml of Luria Bertani medium containing ampicillin (100 g ml Ϫ1 ). The following day, 6 ϫ 1 liter of Luria Bertani medium containing ampicillin was inoculated with the starter culture and grown at 37°C with shaking (220 rpm). After 4 h, when the A 600 was 0.5-0.6, the temperature was reduced to 18°C. The cultures were induced with 100 M isopropyl ␤-D-1thiogalactopyranoside, and the cells were harvested 16 h later. The cell pellets were stored at Ϫ80°C until use. His 6 -ceMSR was purified from cell lysate by Ni-NTA-agarose column using the same protocol as described above for ceCblC purification. The fractions containing ceMSR were identified using SDS-PAGE analysis, pooled, concentrated, dialyzed to remove imidazole, and stored at Ϫ80°C. Before use, ceMSR was thawed on ice and mixed with 1 eq of FMN. Excess FMN was removed by centrifugation using an Amicon Ultra-15 centrifugal filter (50-kDa cutoff).
Decyanation of CNCbl by ceCblC-The reactions were carried out with 50 M ceCblC and 20 M CNCbl in 100 mM HEPES, pH 7.0, 150 mM KCl, and 10% (v/v) glycerol in a total volume of 200 l. The reactions were conducted at 20°C because ceCblC tended to precipitate at higher temperatures. Reactions were initiated by the addition of GSH (0.10 -2.0 mM) or other thiols (8 mM) or ceMSR (4 -10 M) in the presence of 0.2 mM NADPH and followed by the disappearance of CNCbl (decrease in absorbance at 530 nm, ⌬⑀ ϭ 3.7 mM Ϫ1 cm Ϫ1 ). To determine the kinetic parameters of GSH-dependent decyanation, the initial rate of the reaction was plotted against GSH concentration and fitted to the Hill equation below.
Decyanation of CNCbl by hCblC-hCblC (extending from residues 1-244) was purified as described previously (16). The reaction mixtures contained 50 M hCblC and 20 M CNCbl in 100 mM HEPES, pH 7.0 or 8.0, 150 mM KCl, and 10% (v/v) glycerol in a total volume of 200 l. Reactions were initiated by the addition of various thiols (8 mM) and carried out at 37°C under either anaerobic or aerobic conditions.
Reduction of OH 2 Cbl by ceCblC-OH 2 Cbl bound to ceCblC was generated by mixing 40 M ceCblC with 20 M OH 2 Cbl in 100 mM HEPES, pH 7.0, 150 mM KCl, and 10% (v/v) glycerol. Unbound OH 2 Cbl was removed using an Amicon Ultra-15 centrifugal filter (30-kDa cut off). The ceCblC⅐OH 2 Cbl complex was diluted in 100 mM HEPES, pH 7.0, 150 mM KCl, and 10% (v/v) glycerol to obtain a final concentration of 10 -20 M cobalamin. The reactions were initiated by the addition of various thiols (2-8 mM) or ceMSR (600 nM) with 0.2 mM NADPH in a total volume of 200 l and carried out at 20°C under aerobic conditions. The reaction was followed by the disappearance of OH 2 Cbl at 525 nm (⌬⑀ ϭ 5.5 mM Ϫ1 cm Ϫ1 ).
Reduction of OH 2 Cbl by hCblC-The reaction mixtures contained 50 M hCblC and 20 M OH 2 Cbl in 100 mM HEPES, pH 7.0, 150 mM KCl, and 10% (v/v) glycerol in a total volume of 200 l. Reactions were initiated by the addition of 8 mM GSH and carried out at 37°C under either anaerobic or aerobic conditions. MSR Activity Assay-Reduction of free OH 2 Cbl (35 M) by ceMSR (100 -300 nM) was carried out in 50 mM potassium phosphate buffer, pH 7.2, and 0.2 mM NADPH at 37°C, as described previously (18). Reduction of cytochrome c (60 M) was performed with 50 -150 nM ceMSR in 50 mM Tris-HCl (pH 7.5) and 0.2 mM NADPH at 37°C. Reactions were monitored by the increase in absorbance at 550 nm (⌬⑀ ϭ 21 mM Ϫ1 cm Ϫ1 ), as described previously (19).
HPLC Analysis of Cobalamins-Decyanation of CNCbl (50 M) by hCblC (20 M) in the presence of 8 mM GSH was carried out in 100 mM HEPES, pH 8.0, 150 mM KCl, and 10% (v/v) glycerol under aerobic conditions. After a 30-min reaction at 37°C, the protein was inactivated by heating (70°C, 10 min), and the cofactors were released into solution. The supernatant was analyzed by HPLC as described previously (20).
HPLC Analysis of GSH and GSSG-Reduction of CNCbl or OH 2 Cbl by ceCblC in the presence of 4 mM GSH was carried out in 100 mM HEPES, pH 7.0, 150 mM KCl, and 10% (v/v) glycerol. The amount of GSH and GSSG in the reaction mixtures was analyzed by HPLC using a modification of a previously described method (21). Briefly, 100 l of reaction mixture was mixed with an equal volume of metaphosphoric acid solution (16.8 mg ml Ϫ1 metaphosphoric acid, 2 mg ml Ϫ1 EDTA, and 9 mg ml Ϫ1 NaCl) to precipitate proteins. Thiols in the supernatant were alkylated with monoiodoacetic acid at a final concentration of 7 mg ml Ϫ1 , with the pH adjusted to 7-8 by saturated K 2 CO 3 . After incubation for 1 h in the dark at room temperature, an equal volume of a 2,4-dinitrofluorobenzene solution (1.5% v/v in ethanol) was added. The derivatization reaction was performed in the dark at room temperature for 4 h. The N-dinitrophenyl derivatives of GSH and GSSG were separated by HPLC on a Bondapak NH 2 column (Waters, 300 ϫ 3.9 mm, 10 m) at a flow rate of 1 ml min Ϫ1 . The mobile phase consists of solvent A (4:1 methanol/water mixture) and solvent B, which was prepared by mixing 154 g of ammonium acetate in 100 ml of water and 400 ml of acetic acid and adding 500 ml of the resulting solution to 1000 ml of solvent A. A gradient from 18 to 60% (v/v) solvent B was used to elute GSH and GSSG and the absorbance was monitored at 355 nm.

RESULTS
Expression and Purification of ceCblC-ceCblC was obtained in a yield of ϳ10 mg/liter of culture, and the protein was judged to be Ͼ90% pure by SDS-PAGE analysis following a single Ni-NTA purification step. The purified protein migrated as a single peak by gel filtration chromatography on a Superdex 200 column and eluted with an apparent molecular mass of ϳ30 kDa. This is consistent with ceCblC being a monomer (predicted mass ϳ32 kDa). AdoCbl and MeCbl bind to ceCblC in a base-off state as evidenced by the blue shift in the ␣/␤ bands from ϳ523 nm for the free cofactor to 453 nm for the ceCblC-bound forms ( Fig. 1).
Thiol-dependent Decyanation of CNCbl by ceCblC-CNCbl bound to ceCblC appears to exist as a mixture of the base-on ( max ϭ 548 nm) and base-off ( max ϭ 527 nm) species as suggested by the broad peak between 520 and 550 nm, which is As reported for bovine (22) and human CblC (16), the addition of GSH shifted the equilibrium to the base-off species, as revealed by the blue shift in the absorption maximum to 527 nm.
Unexpectedly, continued incubation of the ceCblC⅐CNCbl complex with GSH resulted in decyanation of CNCbl and formation of cob(II)alamin under aerobic conditions, as indicated by the increase in absorption at 473 nm (Fig. 3A). Isosbestic points for the conversion of CNCbl to cob(II)alamin were at 339, 370, and 491 nm. Cob(II)alamin production was also observed under anaerobic conditions (not shown). Formation of cob(II)alamin is consistent with a one-electron reductive elimination of the cyanide group, as previously observed with hCblC and reduced flavin or a flavoprotein (14,16). The initial rates of cob(II)alamin formation under aerobic and anaerobic conditions were virtually superimposable in the presence of 8 mM GSH (Fig. 3A, inset). The cob(II)alamin formed was stable for at least 2 h under aerobic conditions. The dependence of the k obs for cob(II)alamin formation on GSH concentration was determined under aerobic conditions by following the decrease in absorbance at 530 nm (Fig. 3B). From a Hill plot analysis of the data, the K act for GSH was estimated to be 291 Ϯ 24 M; k cat for decyanation was 0.046 Ϯ 0.003 min Ϫ1 , and the Hill coeffi-cient, n, was 2.1 Ϯ 0.3. The reaction rates increased with pH (Fig. 3B, inset). Other thiols, such as DTT, ␤-mercaptoethanol, and homocysteine, could substitute for GSH and convert ceCblC-bound CNCbl to cob(II)alamin (Fig. 3C), albeit with widely differing rates (Table 1) at a fixed 8 mM concentration of each thiol. The order of the reaction rates was ␤-mercaptoethanol Ͼ DTT Ͼ GSH Ͼ homocysteine. Neither cysteine nor tris(2-carboxyethyl)phosphine supported the reductive decyanation of ceCblC-bound CNCbl under the same conditions. Because the other biologically relevant thiols (cysteine and homocysteine) are present at 100 -1000-fold lower concentrations, respectively, in cells, the data obtained at 8 mM concentrations of each thiol yield a qualitative comparison, with the differences likely to be even greater at their physiologically relevant concentrations. Reductive decyanation of ceCblC-bound CNCbl was not observed in the absence of thiols (not shown). Unlike GSH, the other thiols did not shift the equilibrium of ceCblC-bound CNCbl to the base-off conformation.
Curiously, a long wavelength band with an absorption maximum at 583 nm was observed under aerobic ( Fig. 3A) and anaerobic conditions (not shown) during the decyanation reaction with GSH but not with the other thiols. The origin of this band is not known. We speculate that it might represent a charge transfer band between the glutathionyl radical (GS ⅐ ) and cob(II)alamin.
Thiol-dependent Decyanation of CNCbl by hCblC-Under anaerobic conditions, incubation of the hCblC⅐CNCbl complex with 8 mM GSH at 37°C resulted in decyanation of CNCbl and generation of cob(II)alamin as indicated by the absorption maximum at 476 nm (Fig. 4A). The isosbestic points for the conversion of CNCbl to cob(II)alamin were at 336, 367, and 484 nm. A long wavelength band with an absorption maximum at 581 nm was observed with hCblC as with ceCblC, and in fact, was more pronounced with the human enzyme (Fig. 4A). Cysteine, homocysteine, ␤-mercaptoethanol, and DTT also supported the reductive decyanation under anaerobic conditions, generating cob(II)alamin ( Table 1).
As noted previously (14), and in contrast to the result with ceCblC, GSH-dependent decyanation of CNCbl bound to hCblC was not observed spectrally under aerobic conditions (Fig. 4B). Instead, the addition of GSH resulted in a blue shift of the CNCbl spectrum due to a shift in the equilibrium to the base-off state (Fig. 4B), as also seen with ceCblC (Fig. 2). Because the apparent lack of reaction with GSH under aerobic conditions with hCblC but not ceCblC was surprising, we sought to verify this result by HPLC analysis of the reaction mixture (Fig. 4B, inset). The data revealed that although the majority of CNCbl (retention time ϭ 19 min) remained intact after 30 min of incubation, a small amount (ϳ15%) was converted to glutathionylcobalamin (retention time ϭ 14.5 min) resulting from the reaction between OH 2 Cbl (retention time ϭ 12 min), the oxidation product of cob(II)alamin released from hCblC during sample preparation, and GSH present in the reaction mixture.
ceMSR Serves as an Electron Donor to ceCblC-We have previously reported that the flavoprotein, MSR, in the presence of NADPH can provide electrons to hCblC for the reductive decyanation of CNCbl (16). To assess whether the orthologous difla- vin oxidoreductase can serve an equivalent function for ceCblC, the gene encoding ceMSR was cloned, and the protein was purified. ceMSR was judged to be active using the cytochrome c or OH 2 Cbl reduction assay in the presence of 0.2 mM NADPH as described previously (18,19). ceMSR was more prone to losing activity following dilution than hMSR, presumably due to loss of the flavin cofactor. The k cat values for the ceMSR-catalyzed reduction of cytochrome c and free OH 2 Cbl at 37°C were estimated to be 1.15 Ϯ 0.07 s Ϫ1 and 40 Ϯ 4 min Ϫ1 , respectively. These activities are 3-5-fold lower than the values reported for hMSR (18,19).
In the presence of NADPH, ceMSR supported the decyanation of ceCblC⅐CNCbl, generating cob(II)alamin as evidenced by an increase in absorbance at 473 nm and a decrease in absorbance between 520 and 550 nm with isosbestic points of 497 and 572 nm (Fig. 5A). The reaction required both ceCblC and ceMSR and was not observed in the absence of either enzyme (Fig. 5A, inset). The k obs for the MSR-dependent decyanation was 0.085 Ϯ 0.008 min Ϫ1 for ceCblC and was comparable with that reported for hCblC (k obs ϭ 0.068 Ϯ 0.007 min Ϫ1 ) (14).
Reduction of CblC-bound OH 2 Cbl by Thiols-We tested whether OH 2 Cbl bound to ceCblC can be reduced by thiols (Fig. 6A). The addition of GSH to the mixture resulted in the conversion of OH 2 Cbl to cob(II)alamin (k obs ϭ 1.27 Ϯ 0.04   , Fig. 6A). Other thiols, such as DTT, ␤-mercaptoethanol, cysteine, and homocysteine, also reduced ceCblC-bound OH 2 Cbl to cob(II)alamin under aerobic conditions with rates that were either comparable with or slower than that obtained with GSH (GSH Ϸ ␤-mercaptoethanol Ϸ DTT Ͼ cysteine Ͼ homocysteine). In contrast, in the absence of CblC, glutathionylcobalamin was formed as evidenced by a red shift in the cobalamin spectrum to 535 and 553 nm (Fig. 6B).
The addition of GSH to hCblC-bound OH 2 Cbl under aerobic conditions resulted in a blue shift in the OH 2 Cbl spectrum to 497 and 523 nm (from an initial broad peak between 500 and 530 nm) within the first minute, but no further change was observed over a 30-min period (Fig. 6C). The blue shift in the spectrum is likely to result from a conformational change perhaps reflecting a switch to the base-off state and does not correspond to the stable production of cob(II)alamin. Reduction of OH 2 Cbl bound to hCblC to cob(II)alamin could be observed under anaerobic conditions (k obs ϭ 0.36 Ϯ 0.02 min Ϫ1 , Fig. 6C, inset) consistent with the greater oxygen sensitivity of the electron transfer step from GSH to cobalamin in human versus the C. elegans CblC.

GSSG Is Produced during GSH-dependent Reduction of
CNCbl and OH 2 Cbl by ceCblC-We reasoned that GS ⅐ generated via one-electron oxidation of GSH during decyanation of CNCbl or reduction of OH 2 Cbl might result in the formation of GSSG. To test this hypothesis, we used HPLC to monitor the concentrations of GSH and GSSG. We observed conversion of 2 mM GSH to 1 mM GSSG in 15 min during aerobic decyanation of CNCbl in a reaction mixture originally containing 50 M ceCblC, 25 M CNCbl, and 4 mM GSH (Fig. 7A). In contrast, very little oxidation of GSH was observed during the same time period in the absence of either ceCblC or CNCbl. A similar concentration of GSSG formation was observed during reduction of ceCblC-bound OH 2 Cbl by GSH (Fig. 7B). The stoichiometry of GSSG formed versus cobalamin bound to the ceCblC active site suggests that cob(II)alamin generated during decyanation or reduction is prone to oxidation and forms OH 2 Cbl. The latter, once formed, is rapidly reduced to cob(II)alamin by GSH. The oxidation-reduction cycle is predicted to result in a futile cycle consuming GSH (Fig. 7C). This redox cycle could be limited by protecting cob(II)alamin from oxidation. This prediction was confirmed by HPLC analysis of the reactions con-  ducted under anaerobic conditions where GSSG production was significantly diminished (Fig. 7B).

DISCUSSION
CblC is proposed to be the first processing enzyme that B 12 encounters in the cytoplasm as it exits the lysosomal compartment (3,4). The role of CblC in the trafficking pathway is to convert dietary cobalamins to a common intermediate, cob(II)alamin, which can then be partitioned to AdoCbl and MeCbl synthesis to meet cellular needs for supporting methylmalonyl-CoA mutase and methionine synthase, respectively. hCblC exhibits remarkable chemical versatility, albeit at the cost of efficiency, and catalyzes a range of chemically distinct reactions. Thus, hCblC catalyzes the cleavage of the cobaltcarbon bond of alkylcobalamins via a heterolytic mechanism, transferring the alkyl group to the thiolate of GSH (15,23). With CNCbl, CblC exhibits an entirely different activity using an electron donor for the reductive elimination of cyanide and generating cob(II)alamin in the process (14). Both free and MSR-bound reduced flavin are able to serve as an electron donor to CblC during the decyanation reaction (16). GSH, which is a co-substrate in the dealkylation reaction, increases the binding affinity for CNCbl and shifts the equilibrium of CNCbl bound to hCblC to the base-off state (16,22). However, GSH was not observed to support decyanation under the aerobic conditions used to monitor this reaction (14). The physiological relevance of the slow decyanation and alkyltransferase reactions has been established by examining the fate of [ 57 Co]CNCbl or [ 57 Co]alkylcobalamins fed to fibroblasts from CblC patients versus normal individuals. Our study demonstrated that in the absence of functional CblC, the cobalamin precursors were not processed into the active cofactor forms (23).
As part of our long term goal of studying the interactions between the C. elegans cytoplasmic B 12 trafficking proteins and methionine synthase, we made the unexpected observation that ceCblC displays a heretofore unobserved activity, i.e. decyanation of CNCbl in the presence of GSH (Fig. 8). This activity was surprising for two reasons. First, as noted above, it had not been observed with hCblC under aerobic conditions (14), and second, the use of GSH as a single-electron donor was unexpected. Although the decyanation activity was also supported by nonphysiological reductants, e.g. DTT and ␤-mercaptoethanol (Table 1), it was considerably slower with homocysteine, which unlike GSH is present at low micromolar versus 1-10 mM concentrations in the cell. Hence, GSH, which binds with a relatively high affinity to ceCblC (K act ϭ 291 Ϯ 24 M), is likely to be the only physiologically relevant thiol that supports its decyanation activity.
The GSH-dependent decyanation activity of ceCblC under aerobic and anaerobic conditions (Fig. 3A, inset) led us to reexamine the activity of hCblC under similar conditions. Under anaerobic conditions, the decyanation activity of hCblC could be monitored spectrally in the presence of GSH (Fig. 4A) and other natural and unnatural thiols (Table 1). In contrast, the major spectral change observed upon the addition of GSH to hCblC under aerobic conditions was a shift in the equilibrium of bound CNCbl to the base-off state (Fig. 4B) as seen previ-ously (16), without detectable formation of cob(II)alamin. Because cob(II)alamin bound to hCblC is prone to oxidation and the spectrum of the oxidation product, base-off OH 2 Cbl, is very similar to that of base-off CNCbl, we used HPLC analysis to determine whether decyanation was occurring, albeit at a low level. Indeed, hCblC does catalyze inefficient GSH-dependent decyanation even under aerobic conditions as evidenced by the presence of glutathionylcobalamin in the HPLC trace (Fig. 4B, inset). This result indicates that the decyanation product, cob(II)alamin, is formed inefficiently under aerobic conditions and is rapidly oxidized to OH 2 Cbl. The latter is released from the protein during sample preparation and complexes with GSH, present in excess in the sample. The resulting product, glutathionylcobalamin, is detected by HPLC. Hence, oxygen appears to readily intercept the electron transfer step between GSH and CNCbl in the active site of hCblC but not ceCblC. Furthermore, ceCblC rapidly reduces OH 2 Cbl to cob(II)alamin in the presence of GSH (Fig. 6A) and does not form glutathionylcobalamin. Human CblC does not convert bound OH 2 Cbl to glutathionylcobalamin as reported previously for bovine CblC (24). In contrast to ceCblC (Fig. 4A), stable reduction of OH 2 Cbl to cob(II)alamin by hCblC is seen only under anaerobic but not aerobic conditions (Fig. 6C).
It is interesting that GSH can support the chemically distinct decyanation and dealkylation activities catalyzed by CblC. Although the dealkylation reaction has clear precedence both  within the family of glutathione transferases as well as in MeCbl-dependent methionine synthase, which transfers a methyl group from MeCbl to homocysteine to generate methionine (25), the use of GSH as a single-electron donor is not common. Clearly, the choice between using GSH as a nucleophile versus an electron source is determined by the nature of the cofactor bound to CblC allowing dichotomous reaction mechanisms to be supported: heterolytic cleavage of the cobalt carbon bond resulting in alkyl group transfer versus reductive homolytic cleavage resulting in cyanide elimination (Fig. 8). A shorter and physiologically relevant thiol, such as homocysteine, supports the decyanation reaction (Table 1) but not the dealkylation reaction (15).
GSH is a major cellular antioxidant and is utilized to reduce H 2 O 2 and other oxidized metabolites or protein residues and for conjugation to electrophilic xenobiotics. Formally, oxidation of GSH can occur via a one-or two-electron process. Single-electron oxidation of GSH, although not common, has been observed in reactions catalyzed by lactoperoxidase (26) and horseradish peroxidase (27). The redox potential and mechanism of reduction of CNCbl is very sensitive to the presence of a lower axial ligand (28). In the base-on state, reductive decyanation is a two-electron process yielding cob(I)alamin with a redox potential of ϳϪ0.86 V versus the standard hydrogen electrode (29). In the base-off state, reduction involves a oneelectron process with a redox potential of Ϫ0.11 V (28). Oneelectron oxidation of GSH by CNCbl would generate the glutathionyl radical (GS ⅐ , Reaction 3), which can react with a second mole of GSH to give the radical anion, GSSG ⅐ Ϫ (Reaction 4).

REACTION 6
A rate constant of 6.2 ϫ 10 8 M Ϫ1 s Ϫ1 has been reported for formation of the GSH radical anion (Reaction 4) (30). Under aerobic conditions, GS ⅐ and GSSG ⅐ Ϫ can react with oxygen to form GSH peroxide (Reaction 5) or glutathione disulfide (GSSG) and superoxide (Reaction 6), respectively. In the ceCblC reactions, the 2:1 stoichiometry of GSH consumption to GSSG formation (Fig. 7A) is consistent with the reaction of GSSG ⅐ Ϫ rather than GS ⅐ with oxygen.
A complete understanding of the reaction mechanism of GSH-dependent decyanation awaits detailed kinetic studies on the formation of the GSH oxidation product(s) under aerobic and anaerobic conditions, and these studies are in progress. The oxygen-dependent reactions, subsequent to GSH-dependent dealkylation or decyanation (Fig. 7C), demonstrate that CblC can be a potential source of reactive oxygen species (ROS). It is likely that strategies exist to minimize futile redox cycling by CblC that would result in wasteful consumption of GSH and generation of noxious O 2 . , and subsequently, H 2 O 2 .
Elevated ROS is a hallmark of many inborn errors of metabolism, and in fact, CblC patients show high levels of urinary oxidative damage markers (31). Higher ROS and increased susceptibility to apoptosis have also been reported in fibroblasts from CblC patients versus controls suggesting that ROS production could be a phenotypic modifier in CblC disorders (32). OH 2 Cbl had a protective effect in CblC patient fibroblasts, decreasing ROS production (32). It is possible that some pathogenic mutations promote futile cycling of cob(II)alamin bound to hCblC. The oxygen sensitivity of the human versus the C. elegans CblC impedes GSH-dependent decyanation by hCblC in air. The stability of cob(II)alamin bound to ceCblC in the presence of 8 mM GSH, a physiologically relevant concentration (33), for Ͼ2 h is remarkable. Stabilization by ceCblC is aided by the ability of GSH to reduce the oxidation product OH 2 Cbl back to cob(II)alamin (k obs ϭ 1.27 Ϯ 0.04 min Ϫ1 , Fig. 6A), which comes at the price of ROS production. The oxygen sensitivity of hCblC may have evolved to minimize ROS generation and GSH consumption. In contrast to ceCblC, GSH-dependent reduction of OH 2 Cbl by human CblC is barely detectable under aerobic con-ditions (Fig. 6B). It appears likely that ceCblC occludes oxygen from the GSH binding site more effectively than hCblC.
Cob(II)alamin is presumably the cofactor form that is further processed in the B 12 trafficking pathway, and its stabilization would be advantageous for its subsequent transfer from CblC to downstream acceptors. The mechanisms by which ceCblC stabilizes cob(II)alamin and protects against oxygen interception during the decyanation reaction are not known. Sequence alignment reveals that ceCblC lacks the last ϳ40 C-terminal residues present in mammalian CblCs (Fig. 9), which are predicted to be disordered (16). The C terminus of hCblC does not play an essential role in its catalytic function because deletion of the C-terminal 38 residues does not impair activity (16). A truncated form of CblC is observed in human fibroblasts and in murine tissue (16,34), and the role of the C-terminal extension in mammalian CblCs is open to question. Although the residues lining the B 12 binding pocket are largely conserved between the human and C. elegans active sites, there are a few differences (e.g. Ala-117, Ser-146, Cys-149, and Ile-160 in hCblC are substituted by methionine, isoleucine, serine, and phenylalanine, respectively, in ceCblC) (Fig. 9). The potential role of these residues in protecting against interception of the electron transfer step by oxygen will be interesting to elucidate.
Human CblC has been shown to interact with CblD, a protein involved in B 12 trafficking, albeit its function is not understood. Stable interaction between CblC and CblD required the presence of alkylcobalamins and GSH (35). However, the chemical and oxidation state of the cobalamins in these experiments was not established and was likely to have been mixed due to the propensity of hCblC-bound B 12 to undergo oxidation. In light of the newly discovered role of GSH in reduction of OH 2 Cbl to cob(II)alamin, reexamination of the cofactor form needed to promote complex formation between CblC and CblD is warranted. In summary, we have described two new activities associated with the human and C. elegans CblC, GSH-dependent decyanation and GSH-dependent reduction of OH 2 Cbl. Both activities could be physiologically relevant to B 12 trafficking in the intracellular milieu where GSH concentrations are high and oxygen is limiting milieu.