Animal experiments

Ten B. longum subsp. longum strains with divergent SNP statuses [five strains for each genotype (AGT or GTC in the genomic loci 891,726, 891,804, and 891,054)] in the arginine biosynthesis pathway were preliminarily selected, and their ability to alter the arginine level of the culture supernatant was determined. Briefly, each strain was inoculated into MRS broth with 0.1% l-cysteine HCl, and culture medium without inoculation served as a blank. After cultivation to early stationary phase (OD₆₀₀ = 5.0; the accuracy of the bacterial cell number was determined by plate counting), the culture was centrifuged to remove bacterial cells, and after precipitating protein by trichloroacetic acid (CAS number 76–03-9, Sinopharm Chemical Reagent Co., Ltd., China), the supernatant (including the medium blank) was directly used for determining the arginine level using an amino acid analyzer (L-8900, Hitachi, Japan). Alterations in the arginine level after bacterial cultivation were measured and expressed as ΔArg. Three pairs of B. longum subsp. longum strains (each pair comprised two phylogenetically close strains with different abilities to adjust arginine metabolism) were finally selected out of the 10 strains to conduct animal experiments.

Eight-week-old male C57black/6 J mice used in this study were purchased from the Shanghai Laboratory Animal Center (Shanghai, China). Animal care and study protocols were approved by the Ethics Committee of Jiangnan University, China (JN. No20181215b1000130[269]). All the applicable institutional and national guidelines for the care and use of animals were followed. The mice were kept in the mouse facility at the Laboratory Animal Center of the Department of Food Science and Technology, Jiangnan University, Wuxi, China, on a 12-h light/dark cycle in a temperature- (22 °C ± 1 °C) and humidity-controlled (55% ± 10%) room.

Mice were assigned to different experimental groups (n = 9 for each group). The aging model was generated by administering d-galactose (CAS number 59–23-4, Sinopharm Chemical Reagent Co., Ltd., China) via subcutaneous injection at a dose of 1000 mg/kg BW/d according to our preliminary results (data not shown) and a previous study [63]. Mice in the control group received sterile saline via subcutaneous injection, while the other groups were treated with saline-based d-galactose. For arginine supplementation, arginine was added to the normal mouse chow diet to ensure a dose of 0.4 mg/g BW/d according to a previous report [64]. B. longum strains were freshly cultured in MRS broth, then resuspended in sterile saline, each day, and plate counts were conducted to ensure a gavage dose of 10⁸–10⁹ CFU/d for each mouse. Related to the daily preparation of strains over 9 weeks, F2 cultures for each strain were used and grown from the same original cryo-stock to avoid possible genetic drift of strains. For the control group, an equal volume of sterile saline was administered. The body weight and feed intake of all of the mice were recorded daily, and the doses of arginine and d-galactose were adjusted correspondingly.

For the behavior tests, the testing room was fitted with an adjustable dimmer light within 280 lx, and mice were transferred into the room at least 30 min before testing. Testing equipment was cleaned regularly using 70% ethanol between events to avoid olfactory cuing. The open-field test, Morris water maze (MWM) test, step-through test, and Y-maze test were performed (see the Supplementary materials for detailed experimental settings and procedures).

One day after all of the behavior tests had been completed, the mice were euthanized. Tissues were collected immediately and stored at − 80 °C for measuring the antioxidative parameters within 1 week. The levels of MDA and the activities of GSH-Px, SOD, and CAT in the liver and brain were evaluated according to the instructions of the manufacturer using assay kits (Jiancheng Bioengineering Institute, Nanjing, China).