Impairment of Thiamine Transport at the GUT-BBB-AXIS Contributes to Wernicke's Encephalopathy

Abstract

Wernicke's encephalopathy, a common neurological disease, is caused by thiamine (vitamin B1) deficiency. Neuropathy resulting from thiamine deficiency is a hallmark of Wernicke-Korsakoff syndrome in chronic alcohol users. The underlying mechanisms of this deficiency and progression of neuropathy remain to be understood. To uncover the unknown mechanisms of thiamine deficiency in alcohol abuse, we used chronic alcohol consumption or thiamine deficiency diet ingestion in animal models. Observations from animal models were validated in primary human neuronal culture for neurodegenerative process. We employed radio-labeled bio-distribution of thiamine, qualitative and quantitative analyses of the various biomarkers and neurodegenerative process. In the present studies, we established that disruption of thiamine transport across the intestinal gut blood-brain barrier axis as the cause of thiamine deficiency in the brain for neurodegeneration. We found that reduction in thiamine transport across these interfaces was the cause of reduction in the synthesis of thiamine pyrophosphate (TPP), an active cofactor for pyruvate dehydrogenase E1α (PDHE1α). Our findings revealed that decrease in the levels of PDHE1α cofactors switched on the activation of PD kinase (PDK) in the brain, thereby triggering the neuronal phosphorylation of PDHE1α (p-PDHE1α). Dysfunctional phosphorylated PDHE1α causes the reduction of mitochondrial aerobic respiration that led to neurodegeneration. We concluded that impairment of thiamine transport across the gut-BBB-axis that led to insufficient TPP synthesis was critical to Wernicke-neuropathy, which could be effectively prevented by stabilizing the thiamine transporters.

Keywords: Alcohol; Gut-BBB-axis; Neurodegeneration; Pyruvate dehydrogenase; Thiamine deficiency/transporters; Wernicke’s encephalopathy.

Fig. 1

Effects of alcohol or thiamine deficiency diet (TDD) on biodistribution of thiamine in different tissue organs: The ³H-thiamine hydrochloride (4 μCi in 500 μl saline water) was gavaged per animal following 10–12 weeks of ethanol/control liquid diet pair-feeding or TDD ingestion. Tissue organ homogenates in KRPH buffer mixed with scintillant fluid were measured for thiamine uptake in liquid scintillation counter, and data were quantified and expressed as nCi/g tissue. a Distribution of thiamine in different tissue organs of normal animal presented in pie diagram. Alterations of thiamine uptake in tissue organs under various experimental conditions, b small intestine, c liver, d blood, e brain, f lungs, g heart, and h kidney. Results are presented as mean values (± SEM, N = 5). Statistical significance indicates *p < 0.05, ***p < 0.001 compared with controls, ^#p < 0.05, compared with EtOH

Fig. 2

Alcohol/TD diets ingestion impairs thiamine transporters in the small intestine. a Representative immunofluorescent staining and microscopy analysis of THTR1 (red) merged with glucose transporter 1 marker, GLUT1 (green) in tissue cross-section of small intestine and b Western blot analysis of THTR1 protein levels in the small intestine tissue homogenates. c Representative immunofluorescent staining of THTR2 (red) merged with glucose transporter 1 marker, GLUT1 (green) in small intestine tissue cross-section and d quantitative validation of THTR2 protein levels in intestine tissue homogenates by Western blot. All quantitative data were normalized to beta-actin and results were expressed as ratio of THTR1/THTR2 to that of beta-actin bands. Results are presented as mean values (± SEM, N = 5). Statistical significance indicates *p < 0.05, ***p < 0.001 compared with controls, ^$p < 0.05 compared with TDD, and ^#p < 0.05 compared with EtOH. Scale bar = 50 μm

Fig. 3

Alcohol/TD diets ingestion impairs thiamine transporters at the BBB. a Representative immunofluorescent staining and microscopy analysis of THTR1 (red) merged with GLUT1 (green) in microvessel of brain tissue cross-section and b protein levels of THTR1 in isolated brain microvessels tissue homogenates. c Representative immunofluorescent staining of THTR2 (red) merged with endothelium marker von Willebrand Factor, vWF (green) in microvessel of brain tissue cross-section and d quantitative validation of THTR2 protein levels in isolated brain microvessels tissue homogenates. All quantitative data were normalized to beta-actin and results were expressed as ratio of THTR1/THTR2 to that of beta-actin bands. Results are presented as mean values (± SEM, N = 5). Statistical significance indicates *p < 0.05, ***p < 0.001 compared with controls, ^$p < 0.05 compared with TDD, and ^#p < 0.05 compared with EtOH. Scale bar = 50 μm

Fig. 4

Effects of EtOH on neuronal thiamine transporters. a Representative immunofluorescent staining and microscopy analysis of THTR1 (red) merged with neuronal cell marker, MAP2 (green) and DAPI (blue), and b protein levels of THTR1 in neurons. c Representative immunofluorescent staining of THTR2 (red) merged with MAP2 (green) and DAPI (blue), and d quantitative analysis of THTR2 protein levels in neuronal protein extracts. THTR1/THTR2 protein levels were normalized to beta-actin. Results were expressed as ratio of THTR1/THTR2 to that of beta-actin band and are presented as mean values (± SEM, N = 5). Statistical significance indicates *p < 0.05, ***p < 0.001 compared with controls, ^$p < 0.05 compared with TDD, and ^#p < 0.05 compared with EtOH. Scale bar = 50 μm

Fig. 5

EtOH downregulates thiamine pyrophokinase 1 levels (TPK1). a Representative immunofluorescent staining and microscopy analysis of TPK1 (red) and merged DAPI (blue) in brain tissue from EtOH and EtOH + ALC administered mice compared with controls. Western blot analyses of TPK1 protein levels in brain tissue homogenates (b), and human neurons (c). Bar graphs show the results expressed as ratio of TPK1 to that of beta-actin bands, and data are presented as mean SEM (n = 5). Statistically significant, *p < 0.05 compared with controls, and ^#p < 0.05 compared with EtOH (second bar). Scale bar: 40 μm in all panels

Fig. 6

Alcohol-induced deficiency of thiamine decreased the levels of pyruvate dehydrogenase E1α (PDHE1α) in intestine and brain cortex. a Immunofluorescent staining and microscopy analysis of PDHE1α (red) merged with GLUT1 (green) in intestinal tissue cross-sections. b Western blot analysis of PDHE1α in intestinal tissue homogenates. c Representative immunofluorescent staining and microscopy analysis of PDHE1α in microvessel of brain tissue cross-sections merged with endothelial marker, vWF (green), and d quantitative validation of PDHE1α protein levels in isolated brain microvessel tissue homogenates. All quantitative data were normalized to beta-actin and results were expressed as ratio of PDHE1α to that of beta-actin bands. Results are presented as mean values (± SEM, N = 5). Statistical significance indicates *p < 0.05, **p < 0.001 compared with controls, and ^#p < 0.05 compared with EtOH. Scale bar = 20 μm

Fig. 7

Alcohol/TDD downregulates PDHE1α protein levels by upregulating the levels of PDHE1α phosphorylation (p-PDHE1α) in the brain. a Representative immunofluorescent staining of PDHE1α (red) in cortical brain tissue merged with neuronal nuclear antigen marker, NeuN (green) and DAPI (blue), and b Western blot analysis of PDHE1α protein levels in mice cortical brain tissue homogenates. c Immunofluorescent staining and microscopy analysis of PDHE1α phosphorylation at amino acid residue Ser293 (p-PDHE1αSer293, red) in cortical brain tissue sections merged with NeuN (green) and DAPI (blue), and d Western blot analysis of p-PDHE1αSer293 levels in cortical brain tissue homogenates. All quantitative data were normalized to beta-actin and results were expressed as ratio of PDHE1α or p-PDHE1αSer293 to those of beta-actin bands. Results are presented as mean values (± SEM, N = 5). Statistically significant, indicates *p < 0.05, **p < 0.001 compared with controls, and ^#p < 0.05 compared with EtOH. Scale bar = 40 μm

Fig. 8

Alcohol/TDD down-regulates PDHE1α levels by upregulating p-PDHE1α levels in neurons. a Representative immunofluorescent staining and microscopy analysis of PDHE1α (red) in primary human neuronal culture merged with neuronal marker neurofilaments, NF (green) and b changes in PDHE1α protein levels in neuronal lysate proteins in different treatment conditions. c Immunofluorescent staining and microscopy analysis of phosphorylated PDHE1α (p-PDHE1αSer293) (red) in neurons merged with neuronal microtubule marker, MAP-2 (green), and d Western blot analysis of p-PDHE1αSer293 levels in neuronal lysate proteins. All quantitative data were normalized to beta-actin and results were expressed as ratio of PDHE1α or p-PDHE1αSer293 to those of beta-actin bands. Results are presented as mean values (± SEM, N = 5). Statistically significant, indicates *p < 0.05, **p < 0.001 compared with controls, and ^#p < 0.05 compared with EtOH. Scale bar = 40 μm

Fig. 9

Increase phosphorylation of PDHE1α is mediated by upregulation of PD kinase-1 (PDK1) and down-regulation of PD phosphatase (PDP) in response to alcohol-induced deficiency of thiamine. a Immunofluorescent staining and microscopy analysis of PDK1 (red) in cortical brain tissue sections from EtOH, EtOH + ALC and controls. Nuclei counter stained with DAPI (blue). Western blot analysis of pyruvate dehydrogenase kinase-1 (PDK1) protein levels in b human neurons and c brain cortical tissue with respective treatment conditions. Alterations of pyruvate dehydrogenase phosphatase (PDP) protein levels in d human neurons and e brain cortical tissue lysates in different experimental conditions. All quantitative data were normalized to beta-actin and results were expressed as ratio of PDK1/PDP to those of beta-actin bands. Results are presented as mean values (± SEM, N = 5). Statistically significant, indicates *p < 0.05, **p < 0.001 compared with controls, and ^#p < 0.05 compared with EtOH. Scale bar = 40 μm

Fig. 10

Proposed mechanisms of alcohol-induced thiamine deficiency triggering the development and progression of Wernicke-Korsakoff Syndrome. Inhibition of THTR1/THTR2 at the gut-BBB-axis decreases the synthesis of PDHE1α active co-factor thiamine pyrophosphate (TPP) in the brain. Deficiency of TPP at PDHE1α catalytic site (see X) triggers PDK1 activation and PDP deactivation, which promotes PDK1-elicited neuronal PDHE1α phosphorylation (inactive PDHE1α). Long-term reduction in brain neuronal oxidative phosphorylation (aerobic respiration) leads to neurodegeneration as observed in Wernicke-neuropathy