Decreased Brain Levels of Vitamin B12 in Aging, Autism and Schizophrenia

Abstract

Many studies indicate a crucial role for the vitamin B12 and folate-dependent enzyme methionine synthase (MS) in brain development and function, but vitamin B12 status in the brain across the lifespan has not been previously investigated. Vitamin B12 (cobalamin, Cbl) exists in multiple forms, including methylcobalamin (MeCbl) and adenosylcobalamin (AdoCbl), serving as cofactors for MS and methylmalonylCoA mutase, respectively. We measured levels of five Cbl species in postmortem human frontal cortex of 43 control subjects, from 19 weeks of fetal development through 80 years of age, and 12 autistic and 9 schizophrenic subjects. Total Cbl was significantly lower in older control subjects (> 60 yrs of age), primarily reflecting a >10-fold age-dependent decline in the level of MeCbl. Levels of inactive cyanocobalamin (CNCbl) were remarkably higher in fetal brain samples. In both autistic and schizophrenic subjects MeCbl and AdoCbl levels were more than 3-fold lower than age-matched controls. In autistic subjects lower MeCbl was associated with decreased MS activity and elevated levels of its substrate homocysteine (HCY). Low levels of the antioxidant glutathione (GSH) have been linked to both autism and schizophrenia, and both total Cbl and MeCbl levels were decreased in glutamate-cysteine ligase modulatory subunit knockout (GCLM-KO) mice, which exhibit low GSH levels. Thus our findings reveal a previously unrecognized decrease in brain vitamin B12 status across the lifespan that may reflect an adaptation to increasing antioxidant demand, while accelerated deficits due to GSH deficiency may contribute to neurodevelopmental and neuropsychiatric disorders.

Fig 1. Cobalamin-related redox metabolic pathways in neuronal cells.

Endocytosis brings TC-bound Cbl species to lysosomes where axial ligands are removed by MMACHC and MeCbl or AdoCbl are subsequently formed by SAM and ATP-dependent pathways, respectively. MeCbl is a required cofactor for methionine synthase, whose activity supports a large number of methylation reactions, including DNA methylation, as well as dopamine-stimulated phospholipid methylation, carried out by the D4 dopamine receptor (D4R). AdoCbl supports MMACoA mutase in mitochondria. Cysteine, which is rate-limiting for GSH synthesis, can be provided either by cellular uptake via the cysteine/glutamate transporter EAAT3 (excitatory amino acid transporter 3) or by transsulfuration of HCY via cystathionine. The latter pathway is restricted in human brain, increasing the importance of growth factor-dependent cysteine uptake by EAAT3.

Fig 2. Cobalamin status in human frontal cortex.

(a) The general structure of Cbl species in which “X” represents various ligands linked to the cobalt atom, giving rise to the five different Cbl species measured in postmortem frontal cortex. (b) Total Cbl levels in frontal cortex of control subjects divided into four age groups: 0–20 yrs (n = 12), 21–40 yrs (n = 5), 41–60 yrs (n = 10) and 61–80 yrs (n = 12). (c) Levels of five individual Cbl species of control subjects in four age groups. (d) Age-dependent decrease of MeCbl in human frontal cortex (n = 43). Inset: Age trends of serum Cbl, frontal cortex total Cbl and MeCbl. Serum Cbl data is from Ref. 30. (e) Total Cbl levels in placenta (n = 6), frontal cortex of fetal (n = 4) and control (0–20 yrs) subjects (n = 12). (f) Levels of five individual Cbl species in placenta (n = 6), frontal cortex of fetal (n = 4) and control (0–20 yrs) subjects (n = 12). * Indicates a significant difference from 0–20 yrs group (* p < 0.05, ** p < 0.01, *** p < 0.001).

Fig 3. Cobalamin status in autism and schizophrenia.

(a) Total Cbl levels in frontal cortex of autistic subjects (n = 12) and aged-matched controls (n = 9). (b) Levels of five individual Cbl species in frontal cortex of autistic subjects (n = 12) and aged-matched controls (n = 9). (c) Total Cbl levels in frontal cortex of schizophrenic subjects (n = 9) and aged-matched controls (n = 9). (d) Levels of five individual Cbl species in frontal cortex of schizophrenic subjects (n = 9) and aged-matched controls (n = 9). * Indicates a significant difference from control group (* p < 0.05, ** p < 0.01, *** p < 0.001).

Fig 4. Redox and methylation metabolites in aging and autism.

(a) Redox and methylation pathway metabolites in control subjects of 0–20 yrs (n = 12) compared to subjects of 61–80 yrs (n = 10). (b) GSH/GSSG ratio (left) and SAM/SAH ratio (right) in aging. (c) Redox and methylation pathway metabolites in frontal cortex of autistic subjects (n = 9) compared to age-matched controls (n = 9). (d) GSH/GSSG ratio (left) and SAM/SAH ratio (right) in autism. * Indicates a significant difference from 0–20 yrs group (panels a and b) or control group (panels c and d) (* p < 0.05, ** p < 0.01, *** p < 0.001).

Fig 5. Methionine synthase activity in autism.

Methionine synthase activity in frontal cortex of autistic and age-matched control subjects measured either with only endogenous Cbl or with the addition of OHCbl. * Indicates a significant difference from control group (* p < 0.05, ** p < 0.01, *** p < 0.001).

Fig 6. Redox and methylation metabolite and cobalamin status in GCLM KO mice.

(a) Redox and methylation metabolite levels in frontal cortex of GCLM KO mice at P40 and P90 (n = 7). Results are expressed as a percentage of the WT level of each metabolite. (b) Levels of five individual Cbl species in frontal cortex of GLCM KO and WT mice at P40 and P90. Inset indicates total Cbl levels. * Indicates a significant difference from control group (* p < 0.05, ** p < 0.01, *** p < 0.001).