One-Carbon Metabolism in Alzheimer's Disease and Parkinson's Disease Brain Tissue

Abstract

Disruptions in one-carbon metabolism and elevated homocysteine have been previously implicated in the development of dementia associated with Alzheimer's disease (AD) and Parkinson's disease (PD). Moreover, a PD diagnosis itself carries substantial risk for the development of dementia. This is the first study that explores alterations in one-carbon metabolism in AD and PD directly in the human brain frontal cortex, the primary center of cognition. Applying targeted liquid chromatography-tandem mass spectrometry (LC-MS/MS), we analyzed post-mortem samples obtained from 136 subjects (35 AD, 65 PD, 36 controls). We found changes in one-carbon metabolites that indicate inefficient activation of cystathionine β-synthase (CBS) in AD and PD subjects with dementia, the latter seemingly accompanied by a restricted re-methylation flow. Levodopa-carbidopa is known to reduce available vitamin B6, which would explain the hindered CBS activity. We present evidence of temporary non-protein-bound homocysteine accumulation upon levodopa intake in the brain of PD subjects with dementia but not in non-demented PD subjects. Importantly, this homocysteine elevation is not related to levodopa dosage, disease progression, or histopathological markers but exclusively to the dementia status. We hypothesize that this levodopa-induced effect is a direct cause of dementia in PD in susceptible subjects with reduced re-methylation capacity. Furthermore, we show that betaine best correlates with cognitive score even among PD subjects alone and discuss nutritional recommendations to improve one-carbon metabolism function.

Keywords: Alzheimer’s disease; Parkinson’s disease; betaine; brain frontal cortex; dementia; homocysteine; levodopa; metabolomics; one-carbon metabolism.

Figure 1

One-carbon metabolism. Red—measured metabolites; yellow highlighted—detected and quantified metabolites; light green—enzymes; purple—vitamin B cofactors. Abbreviations: BHMT, betaine-homocysteine methyltransferase; CBS, cystathionine β-synthase; CSE, cystathionine γ-lyase; DHF, dihydrofolate; DMG, dimethylglycine; FA, folic acid; MTHF, methyltetrahydrofolate; MTHFD, methylenetetrahydrofolate dehydrogenase; MTHFR, methylenetetrahydrofolate reductase; MTR, methionine synthase; SAH, S-adenosylhomocysteine; SAM, S-adenosylmethionine; SHMT, serine hydroxymethyltransferase; THF, tetrahydrofolate; TMG, trimethylglycine; X, methylation acceptor residue.

Figure 2

Group heatmap for differential analysis by diagnosis groups. Regardless of current levodopa status (a) and with levodopa interaction (b) as identified from DOPA concentrations. Red—upregulation; blue—downregulation. Font color: yellow—FDR ≤ 0.05; white—p-value ≤ 0.05; black—p-value > 0.05.

Figure 3

Boxplot of non-protein-bound homocysteine levels by diagnostic group with levodopa interaction. Points represent individual subjects. + and − signs attached as group suffixes denote DOPA+ and DOPA− subgroups. The measured DOPA levels (transformed and standardized) are further visualized as the point color.

Figure 4

Independence of non-protein-bound homocysteine elevation upon levodopa intake in PD subjects with dementia on medication dosage and measures of disease progression. (Top): measured DOPA, levodopa medication amount per dose and per day and a ratio of measured DOPA and levodopa dosage; (middle row): age, length of the diagnosis, UPDRS-M motor score, and MMSE cognitive score; (bottom): histopathological scores of neurofibrillary tangle density, senile plaque density, and Unified Lewy body stage. The DOPA+ PD subjects with dementia (red) have increased homocysteine levels than non-demented DOPA+ PD subjects (blue) irrespective of these variables. Levodopa dosage, UPDRS-M and MMSE scores are not available for all subjects. Abbreviations: norm, normalized (transformed and standardized).

Figure 5

Association of MMSE with betaine. The blue line represents a locally estimated scatterplot smoothing (LOESS) trend with 95% confidence interval shown as the dark gray area.

Figure 6

Differentially altered metabolic correlations by diagnosis group. Only DOPA− subjects are included to avoid levodopa-related pathway disturbances. PD-dem DOPA+ subjects are included for comparison with the effects presented for DOPA− PD subjects with dementia to demonstrate their similarity (apart from elevated Hcy) even upon levodopa intake.

Figure 7

Overview of differentially expressed metabolites and detected pathway abnormalities by group and levodopa status. Upregulated (red) and downregulated (blue) metabolites with p-value ≤ 0.05; additionally highlighted with yellow for FDR ≤ 0.05. The lightning icons denote potentially impacted pathways (regardless of primary causality) based on abnormalities found in differential analysis and correlation analysis. Notice that the alterations revealed in DOPA− PD subjects with dementia are well reflected in metabolite concentrations after levodopa-induced Hcy challenge. Light green—enzymes; purple—vitamin B cofactors. Abbreviations: BHMT, betaine-homocysteine methyltransferase; CBS, cystathionine β-synthase; CSE, cystathionine γ-lyase; MTHF, methyltetrahydrofolate; MTHFR, methylenetetrahydrofolate reductase; MTR, methionine synthase; SAH, S-adenosylhomocysteine; SAM, S-adenosylmethionine; SHMT, serine hydroxymethyltransferase; THF, tetrahydrofolate.