4-ethylphenyl-cobalamin impairs tissue uptake of vitamin B12 and causes vitamin B12 deficiency in mice

Abstract

Coβ-4-ethylphenyl-cob(III) alamin (EtPhCbl) is an organometallic analogue of vitamin B12 (CNCbl) which binds to transcobalamin (TC), a plasma protein that facilitates the cellular uptake of cobalamin (Cbl). In vitro assays with key enzymes do not convert EtPhCbl to the active coenzyme forms of Cbl suggesting that administration of EtPhCbl may cause cellular Cbl deficiency. Here, we investigate the in vivo effect of EtPhCbl in mice and its ability, if any, to induce Cbl deficiency. We show that EtPhCbl binds to mouse TC and we examined mice that received 3.5 nmol/24h EtPhCbl (n=6), 3.5 nmol/24h CNCbl (n=7) or NaCl (control group) (n=5) through osmotic mini-pumps for four weeks. We analyzed plasma, urine, liver, spleen, submaxillary glands and spinal cord for Cbl and markers of Cbl deficiency including methylmalonic acid (MMA) and homocysteine (tHcy). Plasma MMA (mean±SEM) was elevated in animals treated with EtPhCbl (1.01±0.12 µmol/L) compared to controls (0.30±0.02 µmol/L) and CNCbl (0.29±0.01 µmol/L) treated animals. The same pattern was observed for tHcy. Plasma total Cbl concentration was higher in animals treated with EtPhCbl (128.82±1.87 nmol/L) than in CNCbl treated animals (87.64±0.93 nmol/L). However, the organ levels of total Cbl were significantly lower in animals treated with EtPhCbl compared to CNCbl treated animals or controls, notably in the liver (157.07±8.56 pmol/g vs. 603.85±20.02 pmol/g, and 443.09±12.32 pmol/g, respectively). Differences between the three groups was analysed using one-way ANOVA and, Bonferroni post-hoc test. EtPhCbl was present in all tissues, except the spinal cord, accounting for 35-90% of total Cbl. In conclusion, treatment with EtPhCbl induces biochemical evidence of Cbl deficiency. This may in part be caused by a compromised tissue accumulation of Cbl.

Figure 1. Structural formulas of some relevant cobalamins.

Left: Active forms of cobalamins. R denotes any of the following ligands: 5’-deoxyadenosyl (5′-deoxyadenosyl-cobalamin (AdoCbl)); methyl (methylcobalamin (MeCbl)); CN (vitamin B₁₂ (CNCbl)) or H₂O (aquocobalamin (H₂OCbl)). Right: Structural formula of Co_β-4-ethylphenyl-cob(III) alamin (EtPhCbl) [3].

Figure 2. Binding of Co_β-4-ethylphenyl-cob(III) alamin (EtPhCbl) to mouse TC.

Increasing concentration of EtPhCbl (with and without exposure to light) or vitamin B₁₂ (CNCbl) was incubated with⁵⁷[Co]-Cbl and mouse TC. The quantity of bound⁵⁷[Co]-Cbl was expressed relative to the amount bound when only⁵⁷[Co]-Cbl was present. A logarithmic scale was used on the X axis.

Figure 3. Cbl and MMA levels in mouse urine.

Cbl (upper panel) and MMA (lower panel) levels in urine collected before and one to four weeks after insertion of minipumps delivering NaCl (controls, n=5) or 3.5 nmol/day vitamin B₁₂ (CNCbl, n=7) or Co_β-4-ethylphenyl-cob(III) alamin (EtPhCbl, n=6). Mean±SEM of Cbl/creatinine (Cbl) levels are shown for all three groups, and mean±SEM of MMA/creatinine levels are shown for mice treated with EtPhCbl. For both Cbl and MMA level, unpaired t-test, Welch correction, was employed to compare the difference between each experimental group at each time point and the baseline values (time 0). ^* p<0,05; ^**p<0.0001 vs. baseline values of each experimental group. Note that the Y-scale for Cbl is presented in two segments. crea: creatinine; u: urinary.

Figure 4. Cbl content in plasma and tissue extracts from mice.

Scattergram of Cbl levels in plasma and tissues of mice treated for four weeks with NaCl (controls, n=5), 3.5 nmol/day vitamin B₁₂ (CNCbl, n=7) or Co_β-4-ethylphenyl-cob(III) alamin (EtPhCbl, n=6). The horizontal lines represent the mean level ± SEM. The figure shows the content of Cbl expressed per gram wet weight of tissue or per ml (plasma). One way ANOVA and Bonferroni post-hoc test were employed to compare Cbl levels between the groups. For all the groups ANOVA gave an F value which is statistically significant with three degrees of freedom. The levels of statistical significance of Bonferroni test are: *** indicates increased values, p< 0.001 compared to controls, zzz indicates decreased values, p<0.001 compared to controls, xxx indicates decreased values, p<0.001 compared to CNCbl, VVV indicates increased values, p<0.001 compared to CNCbl.

Figure 5. Forms of Cbl in pools of plasma or tissues of mice treated for four weeks with 3.5 nmol/day Co_β-4-ethylphenyl-cob(III) alamin (EtPhCbl).

Extracted Cbl was subjected to separation by HPLC. X-axis indicates the fraction number. Y-axis indicates the amount of Cbl present in each fraction expressed in percent of the total amount of Cbl recovered. The elution of reference preparations of each form of Cbl is indicated by the arrows. H₂OCbl: aquocobalamin; CNCbl: vitamin B₁₂; AdoCbl: 5′-deoxyadenosyl-cobalamin; CH ₃Cbl: methylcobalamin; EtPhCbl: Co_β-4-ethylphenyl-cob(III) alamin.