Bacteroides uniformis CECT 7771 alleviates inflammation within the gut-adipose 1 tissue axis , involving TLR 5 signaling , in diet-induced obese mice 2

6 Microbial Ecology, Nutrition & Health Research Unit, Institute of Agrochemistry and 7 Food Technology, National Research Council (IATA-CSIC). Valencia, Spain 8 9 *Corresponding Author: 10 Yolanda Sanz; IATA-CSIC. C/ Catedrático Agustín Escardino Benlloch 7, 46980, 11 Paterna-Valencia, Spain 12 Tel: +34 963900022; Fax: +34 963636301 13 E-mail: yolsanz@iata.csic.es 14 15 Running title: Bacteroides uniformis CECT 7771 reduces obesity-associated 16 inflammation 17


52
Obesity has become a major global health challenge due to its increasing prevalence. In 53 2016, more than 1.9 billion adults (39%) 18 years and older were overweight and of these 54 over 650 million (13%) were obese, according to the WHO 1 . Obesity frequently results 55 in a state of chronic low-grade inflammation that is considered a precipitating factor of 56 metabolic complications, such as type 2 diabetes, cardiovascular disease and non-57 alcoholic fatty liver disease 2 . Inflammation of the white adipose tissue (WAT) is 58 considered a major driver of metabolic alterations and, therefore, has been investigated 59 in depth. WAT inflammation is mediated by an overall increase in macrophages largely  significant increases in the potentially pathogenic genus Helicobacter were observed in 212 obese mice under the HFHFD, whereas abundance of this genus was reduced in the 213 HFHFD+B group, which were similar to the control group (SD). Further analysis via 214 BLAST of the DNA sequence associated with the OTU classified to this genus revealed 215 a single species identified as Helicobacter ganmani (100% identity). Ruminococcaceae 216 UCG-014 was reduced by the HFHFD but the administration of the bacteroides strain did 217 not restore this alteration.

218
Correlations between metabolic, immune and gut microbiota features 219 Increased weight gain was positively correlated (q < 0.05) with EAT weight (WAT), 220 blood glucose, leptin, B cells (from EAT), and total macrophages and ratios of M1/M2 221 (from EAT) (Fig. 5). Negative correlations (q < 0.05) were observed between expression 222 of TLR5 in either PP or EAT with obesity markers such as increased body weight gain, 223 cholesterol, triglycerides, EAT weight, blood glucose and leptin as well as the blood pro-224 inflammatory markers IFNγ and IL-1α, while TLR5 from EAT positively correlated with 225 anti-inflammatory makers IL-10 and TSLP from EAT (Fig. 5). Furthermore, negative 226 correlations (q < 0.05) between TLR2 or TLR4 from PP were observed with weight gain,

227
EAT weight, blood glucose and leptin, as well as with B cells and total macrophages 228 (from EAT) and (Fig. 5).

237
This study shows that a Western-style diet locally impacts the intestine, altering the  This study showed that the Western diet-induced fecal microbiota changes and/or their tissues. Consistent with this hypothesis, study models suggest that metabolic 329 inflammation associated with Western diets originates in the intestine before affecting the 330 WAT. The intestine is the first tissue exposed to the diet and also the first to respond by 331 recruiting pro-inflammatory macrophages that, in turn, activate cytokine production and 332 alter gut permeability, ultimately resulting in inflammation and insulin resistance in 333 WAT, while inhibition of intestinal macrophage recruitment prevents insulin resistance 334 9 . Specifically, using a model of adipose tissue inflammation independent of the diet, it 335 was proven that the microbiota drives metabolic inflammation, affecting ultimately the 336 WAT 48 . Further studies using knock-outs for the monocyte chemoattractant protein 337 CCL2 indicated that gut microbiota is responsible for induction of CCL2, which in turn 338 enhances macrophage accumulation in WAT. The study established gut microbiota as a factor aggravating inflammation during diet-induced obesity and, therefore, as a suitable 340 target for therapies against associated metabolic perturbations 13 , as shown in our study.

341
All in all, this study reinforces the idea that diet-induced microbiota changes cooperate 342 with obesogenic diets, aggravating the immune-metabolic deregulation in obesity. The   Glucose tolerance test (GTT) 407 The GTT was performed in vivo after 10 weeks of dietary intervention as described in

518
Bioinformatic processing of data was carried out using the software QIIME 51 , Mothur 52 , 519 and UPARSE 53 . Briefly, using QIIME, paired-end forward and reverse Illumina reads 520 were joined into contigs, barcodes were extracted and reads were demultiplexed. Primers 521 were then removed using the software program Mothur. Using UPARSE, chimeras were 522 removed and reads were clustered at 97 % identity into OTUs using default settings. An

534
Data from animal experiments were analyzed using Graph Pad Prism software (LaJolla, 535 CA). Data distribution was assessed by the Kolmogorov-Smirnov normality test. For 536 normally distributed data, differences were determined with one or two-way ANOVAs 537 (as appropriate) and post hoc Bonferroni's tests. Non-normally distributed data were 538 analyzed with the non-parametric Mann-Whitney U test. In every case, p values < 0.05 539 were considered statistically significant. Gut microbiota statistics and data visualization 540 of sequencing data were carried out using the R statistical software and related R packages 541 or QIIME 51,57 . Comparison between dietary groups of relative abundances of taxonomic 542 groups was carried out using a Kruskal-Wallis test followed by a Wilcoxon rank-sum test 543 to identify significant differences. All p values were corrected for multiple comparisons 544 using false discovery rate where (q < 0.05) was a cutoff for significance. Comparisons of 545 beta diversity between dietary groups using generalized UniFrac distances were