The Biochemical Basis of Cobalamin Deficiency

The Biochemical Basis of Cobalamin DeficiencySubject Review The Biochemical Basis of Cobalamin Deficiency
AYALEW TEFFERI, M.D., AND RAJIV K. PRUTHI, M.D.
• Objective: In this report, our goal was to summa- rize the current knowledge of the biochemical basis for the impaired DNA synthesis and neuropathy asso- ciated with vitamin B12 deficiency.
• Material and Methods: We reviewed the pertinent literature and our clinical experience with cobalamin deficiency.
• Results: Studies have established that the megalo- blastic hematopoiesis associated with vitamin B]2 and folate deficiency is secondary to impaired DNA syn- thesis. Two mechanisms of impairment of DNA syn- thesis have been proposed: the "methylfolate trap hypothesis" and the "formate starvation hypothesis." One possibility is that both hypotheses may be con- tributory—that is, incoming dietary folate may be inaccessible for polyglutamation in accordance with
The megaloblastic anemias result from partially impaired DNA synthesis during hematopoiesis, which eventuates in morphologically evident larger-than-normal precursor cell's with delayed nuclear maturation. The impaired DNA syn- thesis is attributable to a perturbation of the enzymatic DNA repair or synthetic pathways. Ineffective enzymatic activity may be caused by either a deficiency or an inhibition of a cofactor activity. The latter mechanism is being exploited in cancer chemotherapy. Two of the cofactors necessary for DNA synthesis are vitamin B]2 and folate. A knowledge of their interdependent metabolic pathways is essential for un- derstanding the pathophysiologic aspects of megaloblastic anemia caused by vitamin B12 or folate deficiency.
VITAMIN B12
Vitamin B12 was first isolated in 19481 and synthesized in 1973.2 It is structurally classified as a corrinoid.3 The corrinoids are a family of compounds with a corrin ring (Fig. 1 ). The corrin ring consists of four reduced pyrrole subrings joined in a macrocyclic ring linked by the a positions of the pyrrole subrings in a planar configuration (corrin nucleus). The pyrrole subrings are linked to a central cobalt atom. The
From the Division of Hematology and Internal Medicine, Mayo Clinic Rochester, Rochester, Minnesota.
Address reprint requests to Dr. Ayalew Tefferi, Division of Hematology, Mayo Clinic Rochester, 200 First Street SW, Rochester, MN 55905.
the methylfolate trap hypothesis, whereas the formate starvation hypothesis may explain the failure to use already polyglutamated forms of folate.
• Conclusion: Although the pathophysiologic mechanisms of vitamin B12 and folate deficiency are not completely understood, nutritional anemias offer suitable models for the study of the biochemical basis of disease.
(Mayo Clin Proc 1994; 69:181-186)
Ado = adenosyl; C = carbon; Cbl = cobalamin; CN = cyanide; CoA = coenzyme A; IF = intrinsic factor; Me = methyl; MeMaCoA = methylmalonyl-coenzyme A; OH = hydroxyl; PGA = pteroylglutamic acid (folic acid); SAM = S-adeno- sylmethionine; SGMT = serine-glycine hydroxymethyltrans- ferase; TC = transcobalamin; THF = tetrahydrofolate
term "corrin" was initially proposed because this structure forms the core of the vitamin B]2 molecule, not because it contains cobalt.3
Subfamilies of the corrinoids are formed by the addition of various substituents on the cobalt atom, above and below the plane of the corrin nucleus. For example, the cobalamins (Cbls) are corrinoids in which 5,6-dimethylbenzimidazole (a nucleotide) is linked to the cobalt atom below the plane of the corrin nucleus (Fig. 1). Furthermore, the addition of various anionic group ligands, linked to the cobalt atom above the plane of the corrin nucleus, yields the various forms of the Cbls (Fig. 1). Accordingly, the addition of cyanide (CN) gives rise to cyano-Cbl (CN-Cbl), methyl (Me) to Me-Cbl, adenosyl (Ado) to Ado-Cbl, and hydroxyl (OH) to OH-Cbl.
Animal products are the primary dietary source of Cbls. Meat contains OH-Cbl and Ado-Cbl, and dairy products contain OH-Cbl and Me-Cbl.4 Crystalline vitamin B12, which is used for treating Cbl deficiency, is CN-Cbl. For the rest of our discussion, vitamin B12 will be referred to as CN- Cbl. The coenzymatically active forms are Me-Cbl and Ado-Cbl—the respective cofactors for methionine synthase (Fig. 2 A) and methylmalonyl-coenzyme A (MeMaCoA) mutase5 (Fig. 2 B). (These cofactors will be discussed in more detail subsequently in this report.) OH-Cbl and CN- Cbl are converted into the coenzymatically active forms (Me-Cbl and Ado-Cbl) in human tissue.
Mayo Clin Proc 1994; 69:181 -186 181 ©1994 Mayo Foundation for Medical Education and Research
182 BIOCHEMISTRY OF COBALAMIN DEFICIENCY Mayo Clin Proc, February 1994, Vol 69
Anionic ligand • CN · CH 3
• OH *Ado
Coffin ring
Anucteotide (dimethyl·
Fig. 1. Chemical structure of cobalamins. Chanarin I. The Megaloblastic Anaemias. Blackwell, 1979. By permission.)
(Modified from 2nd ed. Oxford:
Cbl in food is bound to protein and must undergo peptic digestion in the acidic environment of the stomach to be released. The freed Cbl initially binds to R-binder (a cobalophilin with a rapid electrophoretic mobility [hence the "R"], found in saliva and gastric juice), which is later de- graded by pancreatic enzymes in the duodenum; this process releases, once again, free Cbl. Thereafter, the Cbl binds tó parietal cell-derived intrinsic factor (IF). This relationship is a prerequisite for absorption of Cbl at the terminal ileum by means of IF receptors (Fig. 3).
Absorbed Cbl is transported from the enterocyte to Cbl- requiring tissues by transcobalamin (TC) II. Approximately 80% of plasma Cbl, however, is bound to other cobalophilins (TC I and the granulocyte-derived TC III)—which, unlike TC II, also bind Cbl analogues and transport them to the liver.6 The holotranscobalamin complex (TC II-Cbl com- plex) binds to specific cellular receptors and is internalized (Fig. 3).
FOLATE Unlike the Cbls, the dietary sources of folate are both animal products and leafy vegetables. Structurally, folic acid (pteroylglutamic acid [PGA]) consists of a pteridine moiety linked to a para-aminobenzoic acid residue attached to the glutamate side chain (Fig. 4). Although the synthetic vita-
min is a monoglutamate (PGA,), the naturally occurring folates, including the dietary forms, are polyglutamates (PGAn).7 The adult daily requirement is approximately 200 μg, but the recommended amount increases during preg- nancy and lactation.8
The monoglutamate form (PGA,) is necessary for trans- port across cell membranes (Fig. 5). PGA, is formed from PGAn by hydrolysis of the glutamate side chain at the jejunal brush-border surface by folate hydrolase (a deconjugase).9
Interaction with membrane-associated folate-binding pro- teins may result in internalization of the receptor-PGA, com- plex, as has been demonstrated in isolated nonintestinal cell systems.10
Within the enterocyte, PGA, is polyglutamated to PGAn
for two reasons: (1) to promote folate-dependent reactions (the active coenzyme forms are polyglutamates) and (2) to maintain a concentration gradient for the uptake of monoglutamates.I0 Subsequently, PGAn are reconverted into the monoglutamate form for release into the portal circula- tion after it undergoes methylation and reduction into 5- methyltetrahydrofolate (Me-THF,)" (Fig. 5). In the plasma, a third of the folate is free, and the rest is nonspecifically bound to albumin. Intracellular uptake may be mediated by folate-binding receptors.10
INTRACELLULAR COFACTOR ACTIVITIES OF Cbl AND FOLATE As previously discussed, the target cell (for example, an erythrocyte precursor) is provided with Me-THF, and a Cbl derivative. Coparticipation of these two compounds in inter-
(Homocysteine)
Cbl (enzyme bound)
0 = C - S C o A
Cofactor = Ado—Cbl
Fig. 2. A, Cofactor activity of methylcobalamin (CH3-Cbl) in methionine synthesis. THF = tetrahydrofolate. B, Cofactor activity of adenosylcobalamin (Ado-Cbt) in succinyl-coenzyme A {CoA) synthesis. MeMaCoA = methylmalonyl-coenzyme A.
Mayo Clin Proc, February 1994, Vol 69 BIOCHEMISTRY OF COBALAMIN DEFICIENCY 183
Intestinal Lumen
-» TCn-Cbl complex
Fig. 3. Enteric processing and absorption of cobalamin (Cbl). IF = intrinsic factor; R-binder = a cobalophilin with a rapid (compared with IF) electrophoretic mobility; TCu = transcobalamin II.
mediary metabolic pathways leads to effective DNA synthe- sis. Although these pathways have not been precisely char- acterized, several prior investigations have provided a framework on which reasonable postulates can be con- structed and discussed.
The primary intracellular function of folates is to transfer one-carbon (1-C) units, at the oxidation levels of methyl (-CH3), méthylène (-CH2-), or formyl (HCO), to facilitate DNA synthesis. To achieve this coenzymatic activity, the folate analogues must be in both a reduced form (THF) and the polyglutamated form (THFn).
12
The primary function of Cbl is to provide coenzymatic activity for the synthesis of methionine and succinyl-coen- zyme A (CoA) (Fig. 2). For this function, the Cbl derivatives must be converted to Me- or Ado-Cbl, and the Cbl molecule must be in a reduced state for optimal enzyme binding.13
In DNA synthesis, the formation of methionine is the "central" reaction (Fig. 6). Both Cbl and folate are necessary for this reaction. Me-Cbl is the cofactor for methionine synthase in the conversion (by methylation) of homocysteine to methionine (Fig. 2 A and 6). The original methyl donor for the reaction is Me-THF in either the monoglutamate form (Me-THF,, obtained primarily from the diet) or the polyglutamate form (Me-THFn, obtained primarily from in- tracellular recycling of folate intermediates).13 The methyl group is first transferred from Me-THF to the enzyme-bound Cbl to form Me-Cbl, which then transfers the methyl group onto homocysteine to generate methionine (Fig. 6). This central reaction provides two products essential for DNA synthesis—THFn for 1-C unit transport and methionine,
which, although controversial, may facilitate the availability of the appropriate folate intermediary in the process.
FORMATION OF INTERMEDIATE FOLATE ANALOGUES Polyglutamation (by means of folylpolyglutamate synthase) of folate analogues is necessary for cellular retention12 and use in 1-C transfer pathways during DNA synthesis. Fur- thermore, Me-THF, must be demethylated before it can be polyglutamated.12 This sequence of events offers a partial explanation for the shortage of THFn during Cbl deficiency. Of note, the poor substrate activity for polyglutamation is restricted to Me-THF, and does not involve other monoglutamate forms, including methylene-THF, and formyl-THF,.14
/ V ° c
CH,
Fig. 4. Chemical structure of folic acid.
184 BIOCHEMISTRY OF COBALAMIN DEFICIENCY Mayo Clin Proc, February 1994, Vol 69
Jejunal Lumen
(Polyglutamated folate)
(No specific blood transport protein)
Fig. 5. Enteric processing and absorption of folates. Me = methyl; PGAt and PGAn = pteroylglutamic acid (folic acid)—monoglutamated and polyglutamated forms; THFI = monoglutamated form of tetrahydrofolate.
A méthylène group is added to THFn by serine-glycine hydroxymethyltransferase (SGMT) during the conversion of serine to glycine (Fig. 6). This is a reversible reaction in that methylene-THF can be converted back to THF .'2 Alterna-
J n n
tively, certain data suggest that, in vivo, methylene-THF may be produced by the oxidation of Me-THF,15 a reaction catalyzed by méthylène reductase (Fig. 6). Pertinent to a later discussion herein, the reduction of methylene-THF to Me-THF (by the same enzyme) is inhibited by methionine derivatives.'5
Methylene-THFn is a key intermediate and may be used in one of three ways: ( 1 ) it can provide the méthylène group to convert deoxyuridylate into thymidylate, (2) it can be oxi- dized to formyl-THFn, which provides 1-C fragments in purine synthesis, and (3) it can be reduced back to Me-THFn
to provide a methyl group during methionine synthesis (Fig. 6).
Formyl-THF is the other folate analogue that is directly involved with DNA synthesis (purine synthesis). Formyl- THF may be produced by oxidation of methylene-THF and vice versa (Fig. 6). Alternatively, formyl-THF may be pro- duced by direct formate transfer (through formyl-THF synthase)16 from THFn (Fig. 6). For the latter reaction, the sources of formate include methionine, which may indepen- dently facilitate the formate transfer without being the for- mate donor.16
The foregoing discussions support one point of view, which maintains that methionine may be important in pro- moting the availability of the appropriate folate analogues (methylene-THF and formyl-THF), which are essential for
DNA synthesis. Methionine inhibits the reduction of methylene-THF to Me-THF15 and facilitates formate transfer in the formation of formyl-THFn from THFn.
16 Therefore, when the production of methionine is directly curtailed by Cbl deficiency, DNA synthesis is indirectly affected.
MECHANISM OF IMPAIRED DNA SYNTHESIS IN Cbl DEFICIENCY The "Methylfolate Trap Hypothesis."—As an extension of the foregoing material, méthylène- and formyl-THF can be considered directly involved in thymidylate and purine syn- thesis, respectively. At least two sources of methylene-THFn
are available, one of which is the circulating Me-THF!. As discussed previously, Me-THF! cannot be demethylated in the absence of Cbl and thus is inaccessible for poly- glutamation.12 The second source involves a recycling of THFn as a by-product of the 1-C transfer reactions. Some of the methylene-THFn is reduced to Me-THFn (a reaction that is normally inhibited by a methionine derivative [S- adenosylmethionine or SAM]; see Figure 6), which could be "trapped" at the methionine synthase reaction stage—the basis for the methylfolate trap hypothesis.17
Observations that contradict the methylfolate trap hy- pothesis include the reversal of the defect by methionine,18
the possibility that Me-THF can be directly oxidized to methylene-THF,'5 and the inability of demethylated THF to correct the defect.19 These observations have been used as evidence to support an alternative hypothesis (see subse- quent section) that considers the availability and utilization of formate the main defect in Cbl deficiency.20
Mayo Clin Proc, February 1994, Vol 69 BIOCHEMISTRY OF COBALAMIN DEFICIENCY 185
Methionine synthase Adenosyl transferase
i | Homocysteine _^-—^_ »Methionine' » S-adenosylmethionine
- s (SAM) CrVCW CW \ /
ï Θ / I
Serlne^lyclne _ Serine T J F n ^ ™ F / è 5^jj5^ Oxidation .<«ss$$^
Méthylène y^reductase purine synthesis H Central reaction
S Key Intermediates Enzymes
Deoxyuridylate Thymidylate CH3-THFn
Fig. 6. Intracellular interdependent cofactor activity of cobalamin (Cbl) and folate. CH} = methyl group; THF and THF = monoglutamated and polyglutamated forms of tetrahydrofolate.
Although the methylfolate trap hypothesis can explain both plasma and tissue accumulations of Me-THF in Cbl deficiency, alternative explanations also exist. Plasma Me- THF accumulations in Cbl deficiency can result from re- duced cellular uptake21 rather than leakage of trapped cellu- lar Me-THF. Similarly, the lack of methionine and its de- rivative (SAM) in Cbl deficiency results in excess Me-THF because of lack of inhibition, by SAM, of methylene-THF reductase, which converts methylene-THF to Me-THF15
(Fig. 6). The "Formate Starvation Hypothesis."—An alternative
hypothesis, the formate starvation hypothesis, considers the lack of methionine (as a result of Cbl deficiency) the most important causative factor in impaired DNA synthesis.20
This theory is based on several observations, including the substantial but incomplete reversal of Cbl deficiency states with methionine18 and the lack of formyl-THF generation and increased formate accumulation during Cbl deficiency, emphasizing the importance of methionine in formate trans- fer.16 The incomplete reversal of clinical features with me- thionine may reflect the failure of methionine to improve cellular uptake of folates rather than suboptimal coenzy- matic activity in formate delivery and utilization.19
In addition, methionine may be adenosylated (SAM) and may be used as a source of methyl and formate groups20·22
(Fig. 6). With methionine deficiency, concentrations of
SAM decline, a situation that promotes methylene-THF re- duction to Me-THF15 (SAM inhibits the forward conversion of methylene-THF to Me-THF; see Figure 6). Furthermore, an effort to conserve SAM for essential methylation might result in decreased SGMT activity and lead to reduced gen- eration of methylene-THFn23 (SGMT adds a méthylène group to THFn during conversion of serine to glycine; see Figure 6).
Moreover, the inability of THF, but not formyl-THF, to correct folate-dependent reactions in Cbl deficiency has been cited as further evidence in support of the formate starvation hypothesis.20 A recent study, however, has dem- onstrated that thymidylate synthesis in Cbl deficiency can be corrected by THF, although with less efficiency than formyl- THF.24
Could Both Hypotheses Be Contributory?—One possi- bility is that circulating methylfolate (Me-THF^ may be handled differently than recycled cellular polyglutamated methylfolate (Me-THFn). Thus, incoming dietary folate (Me-THFj) may be inaccessible for polyglutamation in ac- cordance with the methylfolate trap hypothesis, whereas the formate starvation hypothesis may explain the failure to use already polyglutamated forms of folate. Nevertheless, nei- ther hypothesis can satisfactorily explain the reversal of megaloblastic hematopoiesis in some patients treated with pharmacologie doses of folic acid.
186 BIOCHEMISTRY OF COBALAMIN DEFICIENCY Mayo Clin Proc, February 1994, Vol 69
BIOCHEMICAL BASIS OF NEUROPATHY IN Cbl DEFICIENCY The absence of neuropathy in patients with folate deficiency suggested that methionine synthesis may not be causally related to Cbl-associated neuropathy, and attention initially focused on the other Cbl-dependent reaction, which is the conversion of MeMaCoA to succinyl-CoA catalyzed by MeMaCoA mutase5 (Fig. 2 ß). For this reaction, Ado-Cbl is the cofactor, and the oxidative state and intracellular site of the Cbl molecule differ13 from those of Me-Cbl, factors that may explain the initial absence of methylmalonic aciduria in N20 poisoning.20
Earlier studies showed that excess MeMaCoA inhibited long-chain fatty acid synthesis by being incorporated into terminal positions and resulting in the formation of abnormal branched-chain fatty acids.25 Similarly, an excess of propionyl-CoA, the immediate precursor of MeMaCoA, re- sults in odd-chained fatty acid synthesis.26 Therefore, inves- tigators have presumed that these abnormal fatty acids par- ticipated in abnormal myelin formation.
Recent observations, however, do not support the afore- mentioned hypothesis. First, hereditary MeMaCoA mutase deficiency and defects in Ado-Cbl synthesis do not cause Cbl neuropathy,13 whereas N20 poisoning, which inactivates methionine synthase, does not appreciably affect MeMaCoA mutase and yet causes neuropathy. Second, hereditary de- fects of the methionine synthesis reaction cause Cbl neuropathy,23 and the administration of methionine to Cbl- deficient animals ameliorates the neuropathy.27 Therefore, dysfunction of methionine synthesis—rather than succinyl- CoA synthesis—may be responsible for Cbl neuropathy.
Methionine is adenosylated into SAM, which along with its derivatives is needed in transmethylation reactions and polyamine synthesis, including those that occur in the central nervous system. Although defective amine turnover in the central nervous system is thus a possibility,28 the exact mechanism by which reduced methionine causes demyelina- tion remains to be elucidated. A recent communication suggested that cultured human glial cells may be suitable for study of the biochemical basis of Cbl neuropathy.29
REFERENCES 1. Rickes EL, Brink NG, Koniuszy FR, Wood TR, Folkers K. Crystal-
line vitamin B|2. Science 1948; 107:396-397 2. Maugh TH II: Vitamin B12: after 25 years, the first synthesis.
Science 1973; 179:266-267 3. IUPAC-IUB Commission on Biochemical Nomenclature. The no-
menclature of corrinoids (1973 recommendations). Biochemistry 1974; 13:1555-1560
4. Herbert V. Recommended dietary intakes (RDI) of vitamin B-12 in humans. AmJClinNutr 1987;45:671-678
5. Barker HA. Corrinoid-dependent enzymic reactions. Annu Rev Biochem 1972:41:55-90
6. Herzlich B, Herbert V. Depletion of serum holotranscobalamin II: an early sign of negative vitamin B12 balance. Lab Invest 1988; 58:332-337
7. Davis RE, Nicol DJ. Folie acid. Int J Biochem 1988;20:133-139 8. Herbert V. Recommended dietary intakes (RDI) of folate in humans.
AmJClinNutr 1987;45:661-670 9. Halsted CH. Jejunal brush-border folate hydrolase: a novel enzyme.
WestJMed 1991;155:605-609 10. Kamen BA, Smith AK, Anderson RGW. The folate receptor works
in tandem with a probenecid-sensitive carrier in MA 104 cells in vitro. J Clin Invest 1991; 87:1442-1449
11. Perry J, Chanarin I. Intestinal absorption of reduced folate com- pounds in man. BrJHaematol 1970;18:329-339
12. Schirch V, Strong WB. Interaction of folylpolyglutamates with enzymes in one-carbon metabolism. Arch Biochem Biophys 1989; 269:371-380
13. Cooper B A, Rosenblatt DS. Inherited defects of vitamin B|2 metabo- lism. Annu Rev Nutr 1987;7:291-320
14. Cichowicz DJ, Shane B. Mammalian folylpoly-Y-glutamate syn- thetase. 2. Substrate specificity and kinetic properties. Biochemistry 1987;26:513-521
15. Pearson AGM, Turner AJ. Folate-dependent 1-carbon transfer to biogenic amines mediated by methylenetetrahydrofolate reductase. Nature 1975;258:173-174
16. Deacon R, Perry J, Lumb M, Chanarin I. Formate metabolism in the cobalamin-inactivated rat. BrJHaematol 1990;74:354-359
17. Herbert V, Zalusky R. Interrelations of vitamin B|2 and folic acid metabolism: folic acid clearance studies. J Clin Invest 1962; 41:1263-1276
18. Shin YS, Buehring KU, Stokstad ELR. The relationships between vitamin B|2 and folic acid and the effect of methionine on folate metabolism. Mol Cell Biochem 1975;9:97-108
19. Perry J, Chanarin I, Deacon R, Lumb M. Chronic cobalamin inacti- vation impairs folate polyglutamate synthesis in the rat. J Clin Invest 1983;71:1183-1190
20. Chanarin I, Deacon R, Lumb M, Perry…