Gut microbiota dysbiosis as a novel pathogenic factor of diabetic peripheral neuropathy

To date, a considerable number of studies have shown the involvement of human gut microbiota in the development and progression of various kinds of diseases, including metabolic and neurological disorders. Fecal microbiota transplantation (FMT), the process of transplanting gut microbiota of healthy donors into intestines of patients, is considered a comprehensive therapy to improve enteric environments. FMT has been shown to suppress the progression of type 1 diabetes mellitus1 and restore insulin sensitivity in type 2 diabetes mellitus2 patients in randomized control trials (RCT). These findings indicate critical roles of gut microbiota dysbiosis in insulin deficiency and resistance. In addition, animal and preliminary human studies imply that FMT can be a potential therapy for neurodegenerative disorders, such as Alzheimer's disease, Parkinson's disease and multiple sclerosis. Diabetic peripheral neuropathy (DPN) is the most frequent and early-onset chronic complication of diabetes mellitus. Although DPN shows a variety of nerve injury patterns, its major form is distal symmetric polyneuropathy (DSPN). Ultrastructural studies have shown that the main pathology of DSPN is axonal degeneration of myelinated and unmyelinated fibers, which might characterize the disease as both metabolic and neurodegenerative disorders. As the etiology and onset mechanisms of DSPN remain largely unclear, no disease-modifying therapies against it are currently available. A recent study by Yang et al.3 shed light on gut microbiota dysbiosis as a novel pathogenic factor of DSPN; the details are described in the following paragraphs. In the first study, fecal microbiota from individuals with normal glucose levels, patients with diabetes mellitus but not DSPN and patients with DSPN (M-DSPN) were transplanted into genetically diabetic db/db mice. M-DSPN induced more severe neurological manifestations (reduced nociception and nerve conduction velocities) and histopathology (reduced intraepidermal nerve fiber density and downregulated expression of 200-kD neurofilament, myelin basic protein, and brain-derived neurotrophic factor in spinal sensory ganglia and sciatic nerves) than microbiota from individuals with normal glucose levels or patients with diabetes mellitus but not DSPN. In addition, the M-DSPN-treated group of mice showed more severe gut barrier dysfunction (reduced expression of tight junction proteins in colon tissue, and increased fluorescent dextran permeability) and systemic inflammation (increased plasma levels of lipopolysaccharide-binding protein, tumor necrosis factor-α and interleukin-6) than the other groups. Lipopolysaccharide-binding protein can bind to antigens produced by bacteria, thereby being considered a marker of bacterial antigen load in the blood. Consistent with the mouse phenotypes, genomic and metabolomic analyses with the gut microbiota resulted in significant increases in potentially pathogenic bacteria that might produce endotoxins in M-DSPN compared with microbiota from individuals with normal glucose levels and patients with diabetes mellitus but not DSPN. These findings suggest that M-DSPN exacerbates peripheral neuropathy in diabetic mice, possibly through the disruption of gut barrier integrity and the provocation of systemic inflammation through bacterial antigen load. Conversely, DSPN can be alleviated by improving enteric environments, as shown in the next clinical study. Fecal microbiota from healthy donors (FMD) or saline with plant powder (Placebo) were transplanted to patients with DSPN. The RCT for 12 weeks showed that FMD significantly alleviated DPSN, whereas Placebo showed no beneficial effects. There were no significant differences in glucose and lipid metabolism between the FMD- and Placebo-transplanted groups, whereas the severity of neuropathy evaluated by the Toronto Clinical Scoring System and visual analog scale was significantly reduced in the former, but not the latter, group. The FMD-treated group also showed diminished grades of anxiety, depression and sleep disorders, and increased grade of quality of life, as a result of respective specialized analyses. These outcomes imply that FMD-induced alleviation of neuropathic symptoms provides favorable influences on the mental states of DSPN patients. In addition to the improvement of subjective symptoms, FMD restored objective test scores for DSPN, such as sensory and motor nerve conduction velocities, and perception threshold. Furthermore, the post-RCT for 12 weeks, where the Placebo-treated patients in the RCT received FMD, resulted in a significant improvement in all the subjective and objective scores described above. These findings corroborate the efficacy of FMT toward DSPN; however, its precise action mechanisms remain to be solved. In the RCT study, biopsy specimens of colon tissue were obtained from patients before and 12 weeks after FMD or Placebo transplantation. Immunohistochemical analyses with the specimens showed the tendency of more abundant expression of tight junction proteins in FMD than Placebo. In addition, the serum levels of lipopolysaccharide-binding protein, tumor necrosis factor-α and interleukin-6 were significantly reduced in FMD, but not Placebo. These findings are in contrast to the M-DSPN effects on diabetic mice, and suggest that the beneficial effects of FMT on DSPN patients can be, at least partly, attributed to the restoration of gut barrier integrity and suppression of systemic inflammation. Metagenomic analyses on FMD and the fecal samples from the patients before and after transplantation showed that FMD treatment induced rapid and significant changes in the gut mycobiome in the patients. Among the gut microbial genomes, 21 and 33 genomes were positively and negatively correlated with the Toronto Clinical Scoring System score (the severity of neuropathic symptoms), respectively. These genomes are organized into two competing groups ("guilds"); the genomes negatively correlated with the Toronto Clinical Scoring System score were formed as a potentially beneficial guild 1, whereas those positively correlated with the score were formed as a potentially harmful guild 2. Guild 1 has a higher capacity to synthesize butylate, which might protect the mucosal barrier and maintain the host immune response. Guild 2 is involved in the antigen biosynthesis and load (Figure 1). The findings that FMD treatment increased guild 1 and decreased guild 2 in the gut mycobiome of the patients accord well with the possible action mechanisms of FMT described above. Despite these problems, this study has provided us with a novel etiology of DSPN and the therapeutic efficacy of FMT against it. Currently, no effective remedies for DPN have been established, and the only way to suppress its progression is glycemic control and improvement of lifestyle habits. Therefore, it is hoped that FMT will be available for patients as a promising treatment in the near future.