Cbl Analogues Research Articles

Spectroscopic and computational studies of cobalamin species with variable lower axial ligation: implications for the mechanism of Co-C bond activation by class I cobalamin-dependent isomerases.

5'-deoxyadenosylcobalamin (coenzyme B12, AdoCbl) serves as the cofactor for several enzymes that play important roles in fermentation and catabolism. All of these enzymes initiate catalysis by promoting homolytic cleavage of the cofactor's Co-C bond in response to substrate binding to their active sites. Despite considerable research efforts, the role of the lower axial ligand in facilitating Co-C bond homolysis remains incompletely understood. In the present study, we characterized several derivatives of AdoCbl and its one-electron reduced form, Co(II)Cbl, by using electronic absorption and magnetic circular dichroism spectroscopies. To complement our experimental data, we performed computations on these species, as well as additional Co(II)Cbl analogues. The geometries of all species investigated were optimized using a quantum mechanics/molecular mechanics method, and the optimized geometries were used to compute absorption spectra with time-dependent density functional theory. Collectively, our results indicate that a reduction in the basicity of the lower axial ligand causes changes to the cofactor's electronic structure in the Co(II) state that replicate the effects seen upon binding of Co(II)Cbl to Class I isomerases, which replace the lower axial dimethylbenzimidazole ligand of AdoCbl with a protein-derived histidine (His) residue. Such a reduction of the basicity of the His ligand in the enzyme active site may be achieved through proton uptake by the catalytic triad of conserved residues, DXHXGXK, during Co-C bond homolysis.

Oral Administration of Carbon-14 Labeled Cyanocobalamin (14C-Cbl) Reveals Variable Degradation of Vitamin B12 in the Gastrointestinal Tract That Impacts Vitamin B12 Absorption and Status.

Abstract 3018Poster Board II-994Recent evidence from our laboratory and others suggests that a variable portion of ingested cobalamin (Cbl), either crystalline or from food, is degraded in the gastrointestinal tract. We have developed a biosynthetic method to incorporate 14C into the lower axial ligand of cobalamin that has made it possible to study the fate of this vitamin during its passage through the gastrointestinal tract and to assess the presence of Cbl or its breakdown products in biological samples. Following oral administration of an aqueous physiological tracer dose of 14C-Cbl (1.3 μg, 50 nCi), blood, urine, and feces are analyzed for 14C by accelerator mass spectrometry. In 9 subjects, the plasma response was consistent with the expected behavior of peroral Cbl: 14C-Cbl first appeared in the plasma 3h post-dose reaching a peak level within 6-8h. Confirmation that this dose appears bound to the physiological transport protein transcobalamin (TC) was obtained in a subset of subjects by an immunoaffinity method using anti-human TC antibody-coated magnetic beads which selectively bound 95-98% of plasma 14C. Urinary excretion of 14C was maximal in the first 24h, with 14C first appearing in urine as early as 1.5h after dosing. Fecal excretion occurred variably over several days. The amount of 14C found in the urine (10-50% of the dose) was 100-fold greater than in previous reports using 57Co-labeled cyanocobalamin (0.1-0.5%), and fecal excretion was lower than expected (10-20% vs 30-70%). Urinary excretion of 14C was inversely correlated with the peak plasma level of 14C attained (r2=0.610; p<0.001). The bulk of urinary 14C was not associated with intact Cbl and first appeared in the urine before peak 14C levels were attained in the plasma. The peak plasma level of 14C attained also showed a strong positive correlation with plasma holotranscobalamin concentration measured before administration of the 14C-Cbl (r2=0.571; p<0.001). No such correlation was found with total plasma Cbl. In additional experiments on normal volunteers using eggs endogenously labeled with 14C Cbl following intramuscular injection of hens with 14C Cbl, comparably high urinary excretion of 14C was also observed. We conclude that a variable fraction of ingested Cbl is degraded in the gastrointestinal tract of normal individuals. This may be an important determinant of the amount of Cbl absorbed from food or supplement sources. Additionally, our findings suggest that the concentration of holoTC in the plasma reflects absorptive capacity and may therefore be a good surrogate for Cbl absorptive status. Our findings also have implications regarding the bioavailability of Cbl and may inform pending considerations to fortify food supplies with Cbl in order to mitigate the incidence of Cbl deficiency, particularly among the elderly. Intestinal degradation, either microbial or through the action of digestive enzymes, may also be a source of Cbl analogues that have previously been detected in the plasma and tissues. Cbl analogues may interfere with the physiological function of cobalamin, resulting in some of the manifestations of cobalamin deficiency. Disclosures:Green:Vitalea Science: Research Funding. Miller:Vitalea Science: Research Funding. Lee:Vitalea Science: Research Funding. Sutter:Vitalea Science: Research Funding. Allen:Vitalea Science: Research Funding. Dueker:Vitalea Science: Employment.

Crystal structure of human intrinsic factor: Cobalamin complex at 2.6-Å resolution

The structure of intrinsic factor (IF) in complex with cobalamin (Cbl) was determined at 2.6-A resolution. The overall fold of the molecule is that of an alpha(6)/alpha(6) barrel. It is a two-domain protein, and the Cbl is bound at the interface of the domains in a base-on conformation. Surprisingly, two full-length molecules, each comprising an alpha- and a beta-domain and one Cbl, and two truncated molecules with only an alpha- domain are present in the same asymmetric unit. The environment around Cbl is dominated by uncharged residues, and the sixth coordinate position of Co(2+) is empty. A detailed comparison between the IF-B12 complex and another Cbl transport protein complex, trans-Cbl-B12, has been made. The pH effect on the binding of Cbl analogues in transport proteins is analyzed. A possible basis for the lack of interchangeability of human and rat IF receptors is presented.

Mechanisms of Discrimination between Cobalamins and Their Natural Analogues during Their Binding to the Specific B12-Transporting Proteins

Three proteins, intrinsic factor (IF), transcobalamin (TC), and haptocorrin (HC), all have an extremely high affinity for the cobalamins (Cbls, Kd approximately 5 fM) but discriminate these physiological ligands from Cbl analogues with different efficiencies decreasing in the following order: IF > TC > HC. We investigated interactions of these proteins with a number of ligands: Cbl, fluorescent conjugate CBC, two base-off analogues [pseudo-coenzyme B12 (pB) and adenosyl factor A (fA)], and a baseless corrinoid cobinamide. Protein-ligand encounter and the following internal rearrangements in both molecules were registered as a change in the fluorescence of CBC (alone or mixed with other ligands), a transition in absorbance of pB and fA (base-off --> on-base conversion), and alterations in the molecular mass of two split IF domains. The greater complexity of the binding kinetics followed better Cbl specificity (HC < TC < IF). On the basis of the experimental results, we propose a general binding model with three major steps: (1) initial attachment of the ligand to the high-affinity C-domain, (2) primary assembly of N- and C-domains, and (3) slow adjustments and fixation of the ligand at the domain-domain interface. Since step 3 was characteristic of highly specific TC and especially IF, we suggest its particular importance for ligand recognition. The designed models revealed the absolute Kd values for a group of analogues. Calculations show that most of them could potentially bind to the specific transporters IF and TC under physiological conditions. Implications of this finding and the protective role of HC are discussed.

Mechanism of conversion of human apo- to holomethionine synthase by various forms of cobalamin.

Methionine synthase catalyzes the conversion of N5-methyltetrahydrofolate and homocysteine to tetrahydrofolate and methionine. Methylcobalamin (Me-Cbl) is tightly bound to methionine synthase and is required for enzymatic activity. When added to crude tissue homogenates, Me-Cbl stimulates methionine synthase but similar stimulation is observed with hydroxocobalamin, cyanocobalamin (CN-Cbl), and adenosyl-Cbl, although the mechanisms involved are unknown. We prepared human apomethionine synthase and studied its activation in the presence of [14C]CN-Cbl and [14CH3]Me-Cbl with concentrations of 2-mercaptoethanol ranging from 0.15 to 100 mM. We observed that the removal of the labeled upper axial ligands from CN-Cbl and Me-Cbl both paralleled the activation of human apomethionine synthase. Spectral studies employing CN-Cbl and Me-Cbl showed that both forms of Cbl must be converted to Cob(II)alamin before they can bind to human apomethionine synthase and convert it to its activated holoenzyme form. Studies with 14 different Cbl analogues with alterations in various portions of the corrin ring and the nucleotide showed that all of the analogues were able to fully activate human methionine synthase when they were reduced with 2-mercaptoethanol. Full activation occurred at lower concentrations of many of the Cbl analogues than occurred with Cbl itself. We conclude that Me-Cbl and other forms of Cob(III)alamin do not bind to human apomethionine synthase and that all must first be reduced to Cob(II)alamin before such binding can occur. The fact that human methionine synthase shows little absolute specificity for alterations in various portions of the Cbl molecule suggests that the potent inhibition of mammalian methionine synthase activity observed in vivo with various Cbl analogues is due to inhibition of intracellular Cbl transport or to inhibition of the enzymatic formation of Cob(II)alamin rather than to direct inhibition of mammalian methionine synthase itself.

Inhibition of cobalamin-dependent enzymes by cobalamin analogues in rats.

To determine which parts of the cobalamin (cbl) molecule are required for enzyme activity and which parts, if altered, might inhibit cbl-dependent enzyme activity, we synthesized 16 cbl analogues and administered them to nutritionally normal rats. The cbl analogues, with either modifications of the propionamide side chains of the A-, B-, and C-rings, the acetamide side chain of the B-ring, or the nucleotide moiety, were administered to rats by continuous 14-d subcutaneous infusion. Infusion of cbl-stimulated, cbl-dependent activity. Changes in any part of the cbl molecule always abolished stimulation and, in some cases, caused potent inhibition of both cbl-dependent enzymes. The most inhibitory analogues, OH-cbl[c-lactam], a B-ring analogue, and OH-cbl[e-dimethylamide] and OH-cbl[e-methylamide], two C-ring analogues, decreased mean liver holo-L-methylmalonyl-coenzyme A mutase activity to 65% of control values and increased serum methylmalonic acid concentrations to as high as 3,200% of the control values. Liver methionine synthetase activity was decreased to approximately 20% of the control and mean serum total homocysteine concentrations were increased to 340% of control. A similar level of inhibition was demonstrated in rats who were exposed to 28 d of inhaled nitrous oxide or a prolonged period of dietary cbl deficiency. The inhibitory cbl analogues, nitrous oxide, and diet deficiency all depleted liver cbl. The naturally occurring cbl analogues with modifications of the nucleotide moiety had no effects. We conclude that all parts of the cbl molecule are necessary for in vivo cbl-dependent enzyme activity and that modifications of the side chains of the B and C rings are associated with potent in vivo inhibition of cbl-dependent enzyme activity.

The cDNA sequence and the deduced amino acid sequence of human transcobalamin II show homology with rat intrinsic factor and human transcobalamin I.

The cellular uptake of cobalamin (Cbl, vitamin B12) is mediated by transcobalamin II (TCII), a plasma protein that binds Cbl and is secreted by human umbilical vein endothelial (HUVE) cells. These cells synthesize and secrete TCII and, therefore, served as the source of the complementary DNA (cDNA) library from which the TCII cDNA was isolated. This full-length cDNA consists of 1866 nucleotides that code for a leader peptide of 18 amino acids, a secreted protein of 409 amino acids, a 5'-untranslated segment of 37 nucleotides, and a 3'-untranslated region of 548 nucleotides. A single 1.9-kilobase species of mRNA corresponding to the size of the cDNA was identified by Northern blot analysis of the RNA isolated from HUVE cells. TCII has 20% amino acid homology and greater than 50% nucleotide homology with human transcobalamin I (TCI) and with rat intrinsic factor (R-IF). TCII has no homology with the amino-terminal region of R-IF that has been reported to have significant primary as well as secondary structural homology with the nucleotide-binding domain of NAD-dependent oxidoreductases. The regions of homology that are common to all three proteins are located in seven domains of the amino acid sequence. One or more of these conserved domains is likely to be involved in Cbl binding, a function that is common to all three proteins. However, the difference in the affinity of TCII, TCI, and R-IF for Cbl and Cbl analogues indicates, a priori, that structural differences in the ligand-binding site of these proteins exist and these probably resulted from divergence of a common ancestral gene.

Purification and partial characterization of a cobalamin-binding protein from chicken egg yolk.

Cobalamin-binding protein has been purified from chicken egg yolk by using DEAE-cellulose with a NaCl gradient. The resultant protein fraction was subjected to bioaffinity chromatography. The Mr was 38,000 by SDS-PAGE and 39,000 by gel filtration, and indicated that it was a glycoprotein. The Stokes radius was 4.3 nm and the pI 4.1. The protein bound 57CO.B12 with a molar ratio of 1:1 and a Kd of 0.41 microM. The CBP composed 296 amino acids residues. The protein-ligand interaction was inhibited by Cbl analogues.

Cobalamin-Dependent Replication of L1210 Leukemia Cells and Effects of Cobalamin Analogues

L1210 cell line cells were made deficient in Cbl by propagation in a medium devoid of CNCbl in which fetal calf serum had been replaced by bovine serum albumin. These Cbl-deficient cells gradually ceased to multiply when the medium contained 5-CH3THF, although cell growth was resumed following the addition of CNCbl, OHCbl or folic acid. The results of this study provide experimental proof for the 'methyl trap' hypothesis. In contrast to the above effect of CNCbl, cobinamide and Cbl analogues which were produced by a reaction of OHCbl with ascorbic acid did not have any growth-inducing effect on the cells which had been cultured in a 5-CH3THF-supplemented medium and had ceased to multiply. Nor did these analogues have an inhibitory effect on CNCbl-dependent growth.

A specific radioimmunoassay for 5'-deoxyadenosyl cobalamin in serum.

A specific radioimmunoassay for 5'-deoxyadenosyl cobalamin (ado-Cbl) has been developed using an antiserum raised to this cobalamin (Cbl). At a 1:100 or greater dilution the antiserum did not bind radiolabelled methyl-, hydroxo-, sulphito- or cyano-Cbl and these Cbl analogues did not compete in the radioimmunoassay even at 100-fold higher concentration. The serum concentration (range and mean +/- SD) of ado-Cbl in 30 normal subjects was 47-134 and 81 +/- 16 pg/ml. The corresponding values for total Cbl in these sera were 189-610 and 355 +/- 144 pg/ml, and the computed values for methyl-Cbl were 142-476 and 274 +/- 127 pg/ml. The coefficient of variation was substantially greater for methyl-Cbl than for ado-Cbl (46% v. 21%, respectively). The ado-Cbl concentration was in the normal range (75-95 pg/ml) in five healthy subjects with a low normal concentration (189-217 pg/ml) of total Cbl. In two subjects with low total Cbl (118 and 170 pg/ml) and without any clinical evidence of Cbl deficiency, ado-Cbl was normal (63 and 95 pg/ml, respectively). Thus, in this group, low methyl-Cbl accounted for the lower total Cbl. The concentration (mean +/- SD) of serum ado-Cbl and methyl-Cbl in six patients with low total Cbl (57 +/- 25 pg/ml) and clinical evidence for Cbl deficiency was 35 +/- 12 pg/ml and 22 +/- 22 pg/ml, respectively. This difference from the normal mean for each cofactor was highly significant (P less than 0.001). The decrease in the concentration of methyl-Cbl in Cbl deficiency was relatively greater than the decrease in ado-Cbl (92% v. 57%, respectively) which raised the relative concentration of ado-Cbl in Cbl deficiency to 61% of the total Cbl. Although a low serum methyl-Cbl is a sensitive indicator of Cbl deficiency, it may not be as specific as a decrease in serum ado-Cbl. Factor(s) other than tissue stores of Cbl may lower serum methyl-Cbl below the 95% confidence limit in the absence of clinical Cbl deficiency.

Purification, Some Properties and Possible Physiological Role of an Extracellular Cobalamin Binding Protein from Euglena gracilis

The extracellular cobalamin (Cbl) binding protein from Euglena gracilis was purified and some properties of the protein were studied for the elucidation of its physiological role. The protein was purified about 20-fold with a yield of 15% and was homogeneous on PAGE. SDS-PAGE indicated that the protein had a single type of polypeptide of M r 56000. The protein could bind some Cbl analogues with different β-coordination moieties, over a wide range of pH values from 4·0 to 9·0, and the K s value for cyanocobalamin was 1·1 nm. The extracellular Cbl binding protein was located on the cell surface of E. gracilis, probably bound in the muciferous layer.

Biliary excretion of cobalamin and cobalamin analogues in man.

Using dialysis, gel filtration, isoelectrofocusing and radioaffinity assay, we studied the unsaturated and saturated binders of bile and the biliary concentration of cobalamin (Cbl) and Cbl analogues compared to the corresponding serum concentrations in 7 choledochodomized patients. Bile contained a single saturated or unsaturated R binder with a molecular mass of about 120,000. Differences in the isoelectrofocusing pattern were observed between unsaturated and saturated R binders and could correspond to two secretion origins, mucosal secretion and hepatocyte clearance, respectively. The concentration of total corrinoids is about 4 times higher in bile than in serum, and this could be explained by a hepatic clearance of serum Cbl analogue-R binder complexes, as previously described in the rabbit. Moreover, the enterohepatic circulation of Cbl seems likely in healthy individuals since the saturated biliary R binder is degraded by pancreatic juice.

Nitrous oxide has multiple deleterious effects on cobalamin metabolism and causes decreases in activities of both mammalian cobalamin-dependent enzymes in rats.

In man, use of the general anesthetic nitrous oxide, N2O, is associated with hematologic and neurologic abnormalities that mimic those seen in cobalamin (Cbl, vitamin B12) deficiency. We have measured a number of aspects of Cbl metabolism in rts exposed to various concentrations of N2O for various periods of time. As little as 2% N2O given for 15 h resulted in 30% inhibition of methionine synthetase (MS) in rat liver. With 50% N2O, inhibition of 70% occurred with 1 h and did not change during the next 48 h. Under these conditions, no inhibition of methylmalonyl-CoA mutase (MMCoAM) was observed. The recovery of MS activity was slow and was only 80% of control values 72 h after N2O was stopped. Studies employing rats previously injected with [57Co]Cbl showed that N2O displaced [57Co]Cbl from MS in a manner that temporally and quantitatively paralleled the loss of MS activity. Recovery of MS activity paralleled the reappearance of [57Co]Cbl on MS. N2O also caused the hepatic content of CH3-[57Co]Cbl to decrease by 20-60%. When [57Co]-Cbl was extracted from liver and analyzed by paper chromatography, [57Co]Cbl analogues were present (10-40% of total [57Co]Cbl) in rats exposed to N2O, but not in control animals. When rats were exposed to 50% N2O for 33 d, the total of endogenous Cbl and Cbl analogues in liver decreased to 35% of control values and endogenous Cbl decreased to 10% of control values. At this time, MS activity was 15% of control values and MMCoAM was only 26% of control values. We conclude that N2O causes multiple defects in Cbl metabolism that include the following: (a) rapid inhibition of MS activity with a slow recovery when N2O is stopped; (b) displacement of Cbl from MS; (c) decreased CH3-Cbl; (d) conversion of Cbl to Cbl analogues; (e) the gradual development of Cbl deficiency and (f) an eventual decrease in MMCoAM activity with a further decrease in MS activity.

Presence of cobalamin analogues in animal tissues.

Cobalamin (Cbl, vitamin B-12) has been extracted and isolated from a number of animal tissues by using (i) reverse-affinity chromatography on R protein-Sepharose followed by adsorption to and elution from charcoal-coated agarose and (ii) paper chromatography. Radioisotope dilution assays showed that only 75-97% of the Cbl chromatographed in the position of crystalline Cbl. The remaining 3-25% was present in a number of slower and faster moving fractions. This suggested that Cbl analogues are present in animal tissues because appropriate controls ruled out the possibility that this material was artifactually derived from Cbl during the extraction and purification procedures. With a large-scale isolation from rabbit kidney, the material in five such fractions contained cobalt and had absorption spectra that were similar to but different from the spectrum of Cbl, indicating that they were Cbl analogues. Compared to Cbl, these Cbl analogues had decreased but definite affinities for Cbl-binding proteins with the following order of strength of binding: R protein > transcobalamin II > intrinsic factor. Compared to Cbl, they also had decreased but definite growth-promoting activity for two microorganisms, Euglena gracilis and Lactobacillus leichmannii, which require Cbl for growth. These Cbl analogues differed from each other and from 18 synthetic Cbl analogues, including the most common Cbl analogues synthesized by microorganisms, in at least one of the above features. These studies indicate that animal tissues contain a number of Cbl analogues whose origins, structures, and biologic activities remain to be determined.

Absorption, plasma transport, and cellular retention of cobalamin analogues in the rabbit. Evidence for the existence of multiple mechanisms that prevent the absorption and tissue dissemination of naturally occurring cobalamin analogues.

Analogues of cobalamin (Cbl; vitamin B(12)) are prevalent in nature as a result of bacterial synthesis, and are of additional interest because of their potential use as antimetabolites and chemotherapeutic agents. We have synthesized 14 Cbl analogues containing (57)Co and have compared their gastrointestinal absorption, plasma transport, and cellular retention to that of [(58)Co]Cbl in rabbits. Many of the Cbl analogues were bound with low affinity by intrinsic factor, and none of these [(57)Co]Cbl analogues were taken up by the ileum or absorbed into the body in amounts comparable to that of [(58)Co]Cbl. The Cbl analogues that were bound by intrinsic factor with high affinity were taken up by the ileum but, in many cases, they were retained there in significant amounts. Most of the Cbl analogues were bound by plasma transcobalamin II with high affinity and all of these transcobalamin II-[(57)Co]Cbl analogue complexes were taken up by a variety of tissues in a manner that was indistinguishable from that of transcobalamin II-[(58)Co]Cbl. The few analogues that were bound by transcobalamin II with low affinity were taken up by tissues in lesser amounts, and 20-70% of these analogues was rapidly excreted in the urine as occurs with native Cbl when it is present in plasma in unbound form. All of the Cbl analogues were bound by the granulocyte R-type Cbl-binding protein with high affinity and all of the R-type protein-[(57)Co]Cbl analogue complexes were cleared rapidly from plasma exclusively by hepatocytes as occurs with R-type protein-[(58)Co]Cbl. Some Cbl analogues were released back into the plasma and were disseminated among a variety of tissues via transcobalamin II as occurs with native Cbl. Other Cbl analogues were retained in the liver and eventually excreted in the feces and urine without accumulating in other tissues. These studies indicate that intrinsic factor and the ileum prevent certain Cbl analogues from entering the body and that the granulocyte R-type protein and hepatocytes prevent the dissemination of certain Cbl analogues that may gain entry such as during infections with Cbl analogue-producing bacteria. The fact that transcobalamin II binds and transports a large number of Cbl analogues indicates that these protective mechanisms can be circumvented and supports the feasibility of using Cbl analogues as antimetabolites in vivo.