How a ketogenic diet could reduce autoimmune disease severity through host-microbiome interactions

Ketogenic diet shows promise in reducing autoimmune disease severity through microbiome and metabolic pathways.

Diet has a broad impact on autoimmune diseases, including rheumatoid arthritis, multiple sclerosis (MS), and inflammatory bowel disease. A low-carbohydrate, high-fat ketogenic diet (KD) ameliorates MS-related symptoms.

A KD is characterized by a metabolic shift to lipid oxidation, leading to higher circulatory levels of acetoacetate and β-hydroxybutyrate (βHB).KDs can shape the immune system by modulating the gut microbiota and microbiome. KD-related shifts in the gut microbiota reduce intestinal immune activation. However, their relevance to the disease is unknown.

The study and findings

In the present study, researchers evaluated the role of diet in host-microbiome interactions relevant to disease. First, conventionally raised (CONV-R) mice were fed a high-fat diet (HFD) or a KD. Ten days later, experimental autoimmune encephalitis (EAE) was induced in mice via myelin oligodendrocyte glycoprotein immunization.

Disease was evaluated based on four metrics: temporal changes in disease score, maximum disease developed, overall disease incidence rate, and mice distribution across maximum disease scores. KD-fed mice had significantly lower disease severity.

The KD group had significantly lower levels of cluster of differentiation 4 (CD4⁺) helper T (Th) cells co-producing interleukin (IL)-17a and interferon (IFN)-γ in the spleen and brain.

Next, they repeated these interventions in germ-free (GF) mice. While serum βHB levels were elevated in KD mice relative to HFD mice, there were no significant differences in disease metrics. In addition, unlike CONV-R mice, HFD-fed GF mice showed significantly reduced disease phenotypes.

Next, mice were fed HFD, KD, or HFD supplemented with a βHB-containing ketone ester (KE) before EAE induction.

HFD-fed mice had a higher disease score than KD-fed mice; however, KD and HFD-KE groups were indistinguishable by disease incidence rate or score. Next, the team generated transgenic mice (Hmgcs2^ΔIEC), with selective ablation of intestinal βHB production by the genetic deletion of 3-hydroxy-3-methylglutaryl-CoA synthase 2 (HMGCS2) in intestinal epithelial cells (IECs).

CONV-R Hmgcs2^ΔIEC mice and wildtype controls (Hmgcs2^WT) were treated with tamoxifen for five days and fed a KD for three days before EAE induction. Hmgcs2^ΔIEC mice had significantly elevated disease scores than Hmgcs2^WT mice.

However, circulating levels of βHB were not different between Hmgcs2^ΔIEC and Hmgcs2^WT mice. Next, CONV-R Hmgcs2^ΔIEC and Hmgcs2^WT mice were fed a KD or KD supplemented with KE for three days, followed by EAE induction.

KE supplementation significantly raised circulating βHB levels in mice of either genotype. While Hmgcs2^WT mice were unaffected by KE supplementation, it reduced disease scores and incidence rates in Hmgcs2^ΔIEC mice. Next, the researchers performed fecal microbiota transplantation (FMT) from CONV-R HFD-fed or HFD-KE-fed mice. Baseline microbiota in recipient mice were depleted before HFD or HFD-KE feeding using a common antibiotic cocktail.

FMT from HFD-KE donors reduced EAE phenotypes relative to HFD donors regardless of recipients’ diet. HFD-KE FMT reduced disease scores in both recipient groups. Similar FMT experiments were performed in KD-fed Hmgcs2^ΔIEC and Hmgcs2^WT mice; baseline microbiota in recipients were also depleted before EAE induction.

To this end, stable in vitro communities (SICs) established from gut microbiota were generated. Stool samples from mice fed a KD, HFD, or HFD-KE were passaged thrice in rich bacterial media. Subsequently, 16S rRNA sequencing was performed to evaluate microbial communities.

This revealed that differences between donors persisted despite three passages. After passaging, 16 unique amplicon sequence variants (ASVs) were detected in SICs.

Next, a representative Lactobacillus strain was isolated from SICs and designated Lactobacillus murinus KD6. The researchers investigated whether this strain possessed genes involved in indole-3-lactate (ILA) production, which inhibits IL-17a production by Th17 cells.

They identified four homologs to aminotransferases and four putative lactate dehydrogenase (LDH) genes in L. murinus KD6, which are involved in ILA production.

The team confirmed the inhibition of IL-17a production in Th17 cells by ILA in the skewing assay. Finally, they investigated the effects of ILA and L. murinus KD6 on EAE. CONV-R mice were fed HFD, followed by ILA or L. murinus KD6 administration and EAE induction.

Disease scores in ILA or L. murinus KD6-treated mice were significantly lower than untreated controls. Both treatment groups also had higher survival probability than controls.

Conclusions

The findings reveal a microbiome-dependent pathway through which diet protects from neurological disease. Notably, βHB-containing KE was sufficient to replicate the protective effects of the KD. Local βHB production within the gut was necessary for the protective effect of the KD.

In addition, L. murinus KD6 and ILA were sufficient to protect against EAE. Overall, the results suggest that diet alters the immunomodulatory potential of the gut microbiota by shifting host metabolism.

Journal reference: