Probiotic-derived ecto-5'-nucleotidase produces anti-inflammatory adenosine metabolites in Treg-deficient scurfy mice

Abstract

Probiotic Limosilactobacillus reuteri DSM 17938 (DSM 17938) prolonges the survival of Treg-deficient scurfy (SF) mice and reduces multiorgan inflammation by a process requiring adenosine receptor 2A (A _2A ) on T cells. We hypothesized that L. reuteri -derived ecto-5'-nucleotidase (ecto-5'NT) activity acts to generate adenosine, which may be a central mediator for L. reuteri protection in SF mice. We evaluated DSM 17938-5'NT activity and the associated adenosine and inosine levels in plasma, gut and liver of SF mice. We examined orally fed DSM 17938, DSM 17938Δ5NT (with a deleted 5'NT gene), and DSM 32846 (BG-R46) (a naturally selected strain derived from DSM 17938). Results showed that DSM 17938 and BG-R46 produced adenosine while "exhausting" AMP, whereas DSM 17938∆5NT did not generate adenosine in culture. Plasma 5'NT activity was increased by DSM 17938 or BG-R46, but not by DSM 17938Δ5NT in SF mice. BG-R46 increased both adenosine and inosine levels in the cecum of SF mice. DSM 17938 increased adenosine levels, whereas BG-R46 increased inosine levels in the liver. DSM 17938Δ5NT did not significantly change the levels of adenosine or inosine in the GI tract or the liver of SF mice. Although regulatory CD73 ⁺ CD8 ⁺ T cells were decreased in spleen and blood of SF mice, these regulatory T cells could be increased by orally feeding DSM 17938 or BG-R46, but not DSM 17938Δ5NT. In conclusion, probiotic-5'NT may be a central mediator of DSM 17938 protection against autoimmunity. Optimal 5'NT activity from various probiotic strains could be beneficial in treating Treg-associated immune disorders in humans.

Figure 1

Probiotic-derived 5’NT activity in bacterial culture. a: Proteomics analysis detected 5’NT encoded peptides in bacterial lysates (n=6). b: Probiotic-derived 5’NT converting AMP to adenosine in culture medium with incubation time (n=4 indepednet experiments). c: AMP levels in the culture medium (n=4 independent experiments). ***p<0.001, ****p<0.0001.

Figure 2

Plasma levels of adenosine, inosine and 5’NT activity in SF mice after feeding DSM 17938, BG-R46, or DSM 17938D5NT. a: plasma adenosine level in SF mice. b: plasma inosine level in SF mice. c: plasma 5’NT activity in SF mice. The levels were measured by HPLC. n=6–9 mice per group. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001.

Figure 3

Adenosine and inosine levels in gut and liver of SF mice after probiotic treatment. a: Adenosine levels in duodenum (a), jejunum (b), ileum (c), cecum (d), colon (e), and liver (f). b: Inosine levels in duodenum (a), jejunum (b), ileum (c), cecum (d), colon (e), and liver (f). n=5–8 mice per group. Note that only in panels (d)-(f) were there any significant differences in regional expression. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001.

Figure 4

Regulatory CD73⁺CD8⁺T cells affected by orally feeding DSM 17938, BG-R46, or DSM 17938D5NT to SF mice. a: The proportion of CD73⁺CD8⁺T cells among CD8⁺T cells in spleen and blood analyzed by flow cytometry. b: The proportion of CD8⁺T cells in lymphocyte population in spleen and blood. c: The proportion of activated CD4⁺T cells among CD4⁺T cells in spleen and blood. d: The correlation between adenosine levels and CD73⁺CD8⁺T cells in the blood. n=6–9 mice per group. *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001.