bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1 Title: Housing laboratory mice deficient for Nod2 and Atg16l1 in a natural environment

2 uncovers genetic and environmental contributions to immune variation.

3

4 Authors: Jian-Da Lin1,*, Joseph C. Devlin1,2,*, Frank Yeung2,3,*, Caroline McCauley1,

5 Jacqueline M. Leung4, Ying-Han Chen3, Alex Cronkite1,-3, Christina Hansen4, Charlotte

6 Drake-Dunn3, Kelly V. Ruggles5,6, Ken Cadwell1,3,7,§,†, Andrea L. Graham4,§ ,† and P’ng

7 Loke1,§, †

8

9 Affiliations:

10 1Department of Microbiology, New York University School of Medicine,

11 2Sackler Institute of Graduate Biomedical Sciences,

12 3Kimmel Center for Biology and Medicine at the Skirball Institute, New York University

13 School of Medicine, New York, NY 10016, USA.

14 4Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ

15 08544, USA.

16 5Division of Translational Medicine, Department of Medicine,

17 6Applied Bioinformatics Laboratories, New York School of Medicine, New York, NY 10016,

18 USA

19 7Division of Gastroenterology and Hepatology, Department of Medicine, New York

20 University Langone Health, New York, NY 10016, USA.

21

22 *These authors contributed equally

23 †These authors contributed equally

24

25 §Correspondence to:

26 [email protected]

27 [email protected]

28 [email protected] bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

29 SUMMARY

30 The relative contributions of genetic and environmental factors to variation in immune

31 responses are still poorly understood. Here, we performed a deep phenotypic analysis of

32 immunological parameters of laboratory mice released into an outdoor enclosure, carrying

33 susceptibility (Nod2 and Atg16l1) implicated in the development of inflammatory

34 bowel diseases. Variations of immune cell populations were largely driven by environment,

35 whereas production in response to stimulation was affected more by genetic

36 mutations. Multi-omic models identified transcriptional signatures associated with differences

37 in T cell populations. Subnetworks associated with responses against Clostridium perfringens,

38 Candida albicans and Bacteroides vulgatus were also coupled with rewilding. Hence,

39 exposing laboratory mice carrying different genetic mutations to a natural environment

40 uncovered important contributors to immune variation.

41

42 One sentence summary 43 Natural environment exposure in laboratory mice primarily promotes variation in population 44 frequencies of immune cells, whereas cytokine responses to stimulation are affected more by 45 genetic susceptibility to inflammatory bowel disease. 46 47 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

48 INTRODUCTION

49 Variation in the magnitude of the immune response is an important determinant of

50 susceptibility to pathogen infections, as well as a predisposition to autoimmunity and cancer

51 (Brodin and Davis, 2017; Schirmer et al., 2018). However, immunological studies with

52 laboratory mice typically aim to control variation in order to focus on genetic factors that alter

53 components of both the adaptive and innate immune system. Recent advances in technology

54 and throughput have facilitated a systems immunology approach towards deciphering the

55 relative genetic and environmental contributors to variation of the human immune system

56 (Bakker et al., 2018; Brodin et al., 2015; Li et al., 2016; Schirmer et al., 2016; Ter Horst et al.,

57 2016). By altering the environment in which laboratory mice are conventionally housed,

58 combined with deep phenotypic analysis of immune responses, we describe here a resource

59 dataset that we used to probe the role of genetic mutations versus environmental drivers of

60 heterogeneity in the immune system. Controlled experiments on specific groups with the

61 same genetic mutations exposed to a new environment would be difficult to conduct in human

62 studies.

63 Recently, there has been a growing effort to characterize wild mice and pet shop mice,

64 which appear to share greater similarity to the immunological state of humans than laboratory

65 mice (Abolins et al., 2017; Beura et al., 2016; Rosshart et al., 2019; Rosshart et al., 2017). We

66 have developed an outdoor enclosure facility that enables us to reintroduce laboratory mice

67 into a more natural environment, a process termed “rewilding” (Leung et al., 2018). The

68 enclosures include soil and vegetation as well as barnlike structures but no other mammals

69 (e.g., no wild conspecifics from whom lab-bred mice might acquire infections). Over the

70 course of several weeks, mice dig burrows and otherwise adapt to their new environment

71 (Supplemental Videos). Previously, we have observed C57BL/6 mice in this environment

72 exhibit a decreased type 2 and increased type 1 response to intestinal nematode (Trichuris

73 muris) infection and greater susceptibility to heavier worm burdens (Leung et al., 2018). This

74 facility provides us with a unique opportunity to rapidly alter the housing environment of bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

75 laboratory mice that carry genetic mutations and be able to recover them for subsequent

76 detailed immunological profiling.

77 -bacteria associations have been identified in both human and wild mice

78 indicating host genetic determinants of the gut microbial composition across mammals in

79 natural environments (Suzuki et al., 2019). To interrogate the impact of host genetic variants

80 on microbiota and inflammatory bowel disease (IBD) susceptibility in response to

81 environmental changes, we investigated Atg16l1 and Nod2 variants which are among the

82 highest risk factors for IBD associated with a dysregulated immune response to gut microbes

83 (Wlodarska et al., 2015; Wong and Cadwell, 2018). Atg16l1 is a key component of the

84 autophagy pathway, and Nod2 is involved in bacterial sensing. We and others previously

85 demonstrated that mice harboring mutations in Atg16l1 or Nod2 develop signs of

86 in a manner dependent on exposure to infectious entities (Biswas et al., 2010;

87 Cadwell et al., 2010; Lavoie et al., 2019; Matsuzawa-Ishimoto et al., 2017; Pott et al., 2018;

88 Ramanan et al., 2016; Ramanan et al., 2014). Mice with these particular mutations are

89 sensitive to the presence of murine norovirus (MNV), Helicobacter species, and other

90 commensal-like agents that are found in some animal facilities and excluded in others

91 (Biswas et al., 2010; Cadwell et al., 2010; Caruso et al., 2019; Kernbauer et al., 2014; Pott et

92 al., 2018; Ramanan et al., 2014). Since mice deficient in Atg16l1 or Nod2 can exhibit

93 dysregulated immune responses to colonization by organisms that are otherwise commensals

94 in wild type animals, we reasoned that they could be an interesting model system to

95 investigate gene-environment interactions in the outdoor enclosure.

96 We hypothesized that by introducing laboratory mice carrying IBD susceptibility

97 mutations into the outdoor enclosure, the drastic change in environment may trigger

98 alterations in the immune response and the microbiota, which would have greater adverse

99 effects on mutant mice than wildtype mice. Hence, to study the immunological consequences

100 of rewilding, we released Atg16l1T316A/T316A, Atg16l1T316A/+, Nod2-/- mice in addition to

101 C57BL/6J wild type (WT) mice to determine whether mutations in these genes alter the

102 immune response to microbial exposure in a natural environment. Here we present a detailed bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

103 systems immunology profile of both rewilded and laboratory mice to further investigate

104 variation in immune response and the relationship with environment and genetic mutations.

105

106 RESULTS

107 Study design for immune profiling of rewilded and laboratory mice

108 From 116 mice released into the outdoor enclosure for 6-7 weeks we recovered 104 mice

109 (25 WT, 28 Nod2-/-, 27 Atg16l1T316A/+, and 24 Atg16l1T316A/T316A) in time for analysis and

110 compared them to 80 matched controls (19 WT, 19 Nod2-/-, 20 Atg16l1T316A/+, 22

111 Atg16l1T316A/T316A) maintained under specific pathogen free (SPF) conditions (herein referred

112 to as lab mice) (Figure 1A). Blood samples were collected at the time of sacrifice and

113 analyzed by flow cytometry with a lymphocyte panel (Table S1). Additionally, cytokine

114 production in the plasma was assayed (Figure 1A). Fecal samples and cecal contents were

115 collected for microbial profiling, and for reconstitution experiments (Companion study,

116 Yeung et. al.). Mesenteric lymph node (MLN) cells were isolated for flow cytometry analysis

117 with the same lymphocyte panel as the blood and an additional myeloid cell panel (STAR

118 METHODS). Immune activity in the MLN was measured through transcriptional profiles by

119 RNA-seq and cytokine production assays in response to microbial stimulations (STAR

120 METHODS). All of the lab mice were sacrificed in one week, whereas the rewilded mice

121 were sacrificed over a 2-week period based on trapping frequency. This collection of multi-

122 parameter datasets from nearly 200 mice provides an opportunity to examine inter-individual

123 variation in innate and adaptive immune cell populations by flow cytometry analysis, evaluate

124 cytokine responses to microbial stimulation, and integrate highly dimensional transcriptomics

125 and microbiota data. In order to perform multi-omic analysis, 81 mice were completely

126 profiled with measurements from all four data types; flow cytometry, MLN RNA-Seq, fecal

127 microbiota and MLN restimulation and cytokine profiles. A systems level analysis also

128 enabled us to quantify effects of environmental influences and IBD susceptibility gene

129 mutations, Nod2 and Atg16l1, on the heterogeneity of immunological parameters among

130 individual mice. bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

131

132 Variation of immune cell populations in the blood is largely driven by the environment

133 In line with previous studies exposing lab mice to feral and pet shop mice (Beura et al.,

134 2016), we observed major differences in CD44hiCD62Llo or CD44hiCD62Lhi T cells

135 (Companion study, Yeung et. al.). To visualize the overall landscape of flow cytometry data

136 in an unbiased manner, phenotypic heterogeneity was analyzed on total CD45+ cells per

137 mouse across the blood lymphocyte panel (Table S1) in lab and rewilded mice. We down

138 sampled to 1,000 CD45+ cells from each mouse for nearly 180,000 single CD45+ cell events

139 and single cell flow cytometry data was visualized by uniform manifold approximation and

140 projection (UMAP) (Becht et al., 2019). UMAP analysis was superior to t-distributed

141 stochastic neighborhood embedding (t-SNE) (Becht et al., 2019) (Figure S1A) for identifying

142 major immune cell subsets (Figures 1B and 1C; Figures S1B-E). Unexpected clusters of

143 CD44hiCD19+ cells were increased in the rewilded mice compared to lab mice (Figure 1B),

144 while lab mice had more CD62LhiCD4+ cells (Figure 1C). This unsupervised visualization

145 strategy was useful for identifying unexpected differences (e.g. in the CD19+ compartment

146 and other populations not investigated by (Beura et al., 2016; Rosshart et al., 2019; Rosshart

147 et al., 2017)) in cell clusters of the peripheral blood at a single cell level, which might have

148 been overlooked by traditional flow cytometry gating.

149 In order to identify known immune cell populations, we employed a traditional gating

150 strategy to systematically quantify the frequencies of immune cell subsets with a lymphocyte

151 panel (Figure S2; Table S2). The blood lymphocyte panel identified 13 lymphocyte

152 populations, a total myeloid cell population (CD11b/CD11c/DX5), and a total CD45+ cell

153 population. From the proportions of these immune cell populations, we examined the inter-

154 individual variation in lab and rewilded mice by principal component analysis (PCA) (Figure

155 1D). As shown by the separation of lab and rewilded mice along principle component (PC) 1

156 and 2, there is a strong effect of the environment on lymphocyte populations, and major

157 differences between these populations can be attributed to CD44hiCD62Lhi CD4 and

158 CD44hiCD62Lhi CD8 T cells as inferred by examining loading factors along PC1 and PC2 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

159 (Figure 1D). The reduced proportion of CD44loCD62Lhi T cells is consistent with some

160 previous reports on wild mice (Abolins et al., 2017; Beura et al., 2016). We also quantified

161 the extent of variability between lab and rewilded mice by measuring the Euclidean distances

162 between mice in our PCA. Notably, the within group distances in lab mice were significantly

163 less than rewilded mice (Figure 1E) suggesting that the natural environment increases

164 variability in immune cell populations. We then evaluated several population factors to

165 identify covariates with the largest effect sizes on immune cell populations (STAR

166 METHODS) (Figure 1F). As expected, the composition and activation state of immune cell

167 populations as determined by cell surface marker expression in the peripheral blood are

168 driven most strongly by environmental differences between the SPF facility and the outdoor

169 enclosure. These differences far outweighed any effects of genetic deficiency in Nod2 or

170 Atg16l1 in both environments (Figure 1F). These findings are consistent with the concept that

171 non-heritable influences explain the majority of the variation in the proportion of immune cell

172 subsets as reported in twin studies (Brodin et al., 2015).

173

174 Variation of cytokine production in response to stimulations is driven more by genetic

175 mutations than by the environment.

176 Cytokine production capacity in humans has been reported to be more strongly influenced

177 by genetic than environmental factors (Li et al., 2016). To investigate the relative roles of

178 Nod2/Atg16l1 deficiency and environment on induced responses, we performed ex vivo

179 stimulation experiments on total MLN cells from rewilded and lab mice. To examine cytokine

180 responses to microbial and T-cell specific antigens we stimulated with six different microbes

181 (C. perfringens, P. aeruginosa, B. vulgatus, B. subtilis, S. aureus and C. albicans),

182 αCD3/CD28 beads to activate T cells as well as PBS as a control (Figures 1A and 2A).

183 Despite no significant differences for PBS control samples between lab and rewilded mice

184 (data not shown), we chose to normalize cytokine production per mouse by expressing

185 cytokine data as a fold change levels over PBS (STAR METHODS). A broad array of 13 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

186 was measured: IFN-γ, IL-1α, IL-1ß, IL-13 IL-5, IL-10, IL-17, IL-6, CCL2, CXCL1,

187 CCL3, CCL4 and TNF-α. Average fold changes in cytokine production as compared to PBS

188 controls for both lab and rewilded mice indicate increases in IFN-γ and IL-17 after

189 αCD3/CD28 activation, consistent with the production of these cytokines by T cells in the

190 MLNs (Figure 2A). In general, bacterial antigens induced CCL3, IFN-γ, TNF-α, CCL4, and

191 IL-6 production, while the fungus C. albicans induced the regulatory cytokine IL-10 (Figure

192 2A). Bacterial antigens also induced CXCL1, which was not induced by C. albicans or

193 αCD3/CD28 (Figure 2A). As expected αCD3/CD28 was one of the most potent inducers of

194 cytokine production, especially for IFN-γ, CCL4, TNF-α and IL-5, and rewilded mice

195 produced higher levels of these cytokines than lab mice post stimulation (Figure 2B). IL-1α,

196 IL-1β, IL-13 and CCL2 were not appreciably altered by any stimulation (Figure 2A).

197 We next used PCA to assess the sources of variation in stimulated cytokine production for

198 both lab and rewilded mice (Figure 2C). Lab and rewilded mice do not clearly segregate

199 along PC1 or PC2 suggesting that environment is not a major source of variation. Effect size

200 measures based on the Euclidean distance indicated genotype as a much stronger source of

201 variation (Figure 2D). The differences in genotype are driven by IL-10 and IL-6 production as

202 indicated by the PCA loading factors (Figure 2C). More specifically, the rewilded Nod2-/-

203 mice appear to cluster distinctly from other genotypes, suggesting key differences in cytokine

204 production of rewilded Nod2-/- mice (Figure 2C). Genotype also had the greatest effect size on

205 variation in plasma cytokine levels of individual mice (Figure S3).

206

207 Rewilded Nod2-/- mice exhibit increased cytokine responses compared to wild type

208 rewilded mice.

209 To better understand how Nod2 and Atg16l1 deficiency was affecting cytokine production, we

210 performed two-way comparisons and calculated a p-value for each comparison to determine

211 which cytokine producing conditions were most significantly different between WT mice and

212 mice with Nod2 and Atg16l1 deficiencies from laboratory and rewilded environments bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

213 (Figures 3A and 3B). Compared to the lab mice, the rewilded mice had many more significant

214 changes in stimulated cytokine production across Nod2 and Atg16l1 deficient mice (Figure

215 3A and 3B). These changes were largely driven by significant differences in rewilded Nod2-/-

216 mice against rewilded WT mice (Figure 3A). Specifically, the Nod2-/- mice exhibited

217 increases in IL-17, IL-5 and IL-10 in response to C. perfringens suggesting an increased

218 responsiveness to this bacterial stimulant that was only observed in the rewilded condition

219 (Figures 3A and 3C). Production of IL-10 in response to C. albicans was also elevated in

220 Nod2 and Atg16l1 deficient mice, although Nod2-/- did not exhibit greater fungal colonization

221 than WT mice (Figure 3C) (see companion study Yeung. et.al.). Despite these differences in

222 cytokine responses, the rewilded Nod2-/- did not shown any signs of increased intestinal

223 inflammation by histology or changes in goblet cell numbers (data not shown), which we

224 have previously shown to be associated with susceptibility to B. vulgatus colonization

225 (Ramanan et al., 2016; Ramanan et al., 2014). Together these results indicate that exposure to

226 the outdoor environment per se that induces elevated cytokine responses is not a sufficient hit

227 by itself to trigger intestinal inflammation. Additional insults (i.e. a multi-hit model) may be

228 needed to trigger full blown pathology (Ramanan et al., 2016).

229

230 An integrated classification model identifies features predictive of environment and

231 genotype-specific effects.

232 In addition to stimulated cytokine profiles, MLN cells were assessed for immune cell

233 populations by flow cytometry analysis (Figure S4 and Table S2) and gene expression by

234 RNA-seq analysis (Figures S5A and S5B; Table S2). Consistent with the peripheral blood,

235 immune cell frequencies in MLNs showed a strong effect of environment (Figure S6).

236 Interestingly, a group of rewilded mice all sacrificed at week 7 of rewilding clustered

237 separately in the PCA analysis of lymphocyte populations (Figure S6B). Between week 6 and

238 7 of rewilding it is possible specific events in the enclosure (e.g., a severe thunderstorm, high

239 humidity and fungal blooms observed during that week), could have resulted in changes in T

240 cell activation. Similar to blood lymphocyte populations, RNA-seq analysis (Figures S5A and bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

241 S5B) and 16S rRNA profiling (Figures S5C-E; see companion study by Yeung et al.) showed

242 some environmental effects on variations. This large amount of data prompted us to use a

243 multi-omic classification model that could prioritize features predictive of environment and

244 genotype-specific effects, as well as identify relationships between these features.

245 We integrated multi-omic data from flow cytometry populations, gene expression,

246 cytokine, and microbial profiles (Table S2) in a subset of mice (n-=81) which had measures

247 for all four data types. A concatenated matrix (Table S3) of these features was used to build a

248 random forest classification model (STAR METHODS) to distinguish samples by

249 environment or genotype (Figure 4). The contribution of each feature to the model was also

250 assessed to identify features predictive of environment or genotype-specific effects (Figures

251 4A and 4C). When classifying the environment, a hub of transcription factors (including

252 Zfp36, Atf4, Fos, several Jun and Klf family members) and Cd69 are strong predictive

253 features of environment (Figure 4A). Pairwise correlation analysis between these top features

254 also reveals associations to CD44hiCD62Lhi populations for CD4+ and CD8+ T cells in the

255 blood and MLN, as well as IL-5 and TNF-α production in response to anti CD3/CD28

256 stimulation (Figure 4B). For the genotype model, our classification was not as accurate as the

257 environment, as measured by area under the receiver operator curve (AUC) of 0.78 versus 1

258 on the test set (STAR METHODS). Pairwise correlation analysis revealed fewer associations

259 between these features and overall less connected nodes (Figure 4D). Instead of cytokines

260 associated with T cell activation by αCD3/CD28, IL-17 responses to microbial antigen (B.

261 vulgatus and P. aeruginosa) stimulation were correlated (Figures 4B and 4D). Interestingly,

262 expressions of a number of ribosomal (Rps and Rpl proteins) were also predictive of

263 genotype (Figures 4C and 4D). In summary, we identified a network of transcription factor

264 associated with T cells in the MLNs based on transcriptional profiling data that is strongly

265 predictive of the effects of the different environments.

266

267 Integrative network analysis identifies co-regulated modules in rewilded mice bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

268 To evaluate the interactions between data types, we next utilized an unsupervised sparse

269 partial least squares (sPLS) regression model to integrate the different data types into a multi-

270 component network (Figures 5A and S7). To condense the size of our network we collapsed

271 our gene expression profiles according to known Gene Ontologies (STAR METHODS) (2015;

272 Ashburner et al., 2000) for immune system processes in Mus musculus, which also focuses

273 our attention on the most relevant genes. In short, we compared our gene expression profiles

274 to all child terms of the term “immune system process” (GO:0002376) and

275 collapsed our genes into modules based on these pre-defined ontologies. 954 genes were used

276 to generate 91 specific gene ontologies from our gene expression data all related to the parent

277 gene ontology “immune system process”. These modules were used as inputs alongside the

278 cytokine profiles, microbial profiles and flow cytometry populations to generate an

279 unsupervised multi-omic network. sPLS-regression models were built pairwise between each

280 data type to generate a 188-node covariance network with 577 total connections between the

281 four different types of features (Figures 5A and S7). In the resulting model, all microbial taxa

282 fell below our 0.6 covariance threshold and were therefore removed (Table S4). This

283 indicated that bacterial composition, determined by 16S sequencing alone, was not an

284 important component of the network. Perhaps other strategies to assess microbial function

285 (e.g. transcriptomics, metagenomics or metabolomics) would have yielded more connections

286 to immune function.

287 While total CD4+ and CD8+ T cell populations in the MLN were among the most

288 connected nodes of the network specific gene ontologies, cytokine responses and other

289 immune populations were also strongly interconnected (Figure 5A). As expected from earlier

290 analysis (Figure 1), the blood CD44hiCD62Lhi CD4 T cell population was strongly

291 interconnected and enriched in rewilded mice (Figures 5A and 5B). Additionally, by

292 collapsing our gene expression profiles into annotated gene ontologies, we identified a

293 module (GO:0061844: antimicrobial humoral immune response mediated by antimicrobial

294 peptides) as the most connected gene module that was enriched in rewilded mice (Figures 5A

295 and 5B). Interestingly, IL-5 production in response to C. perfrigens stimulation was the most bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

296 connected cytokine response and was found to be higher in rewilded mice (Figures 5A and 5B;

297 Figure S7). This node is part of a 15-node subnetwork connected to total CD4 and CD8 T cell

298 populations in the MLN, several other cytokine responses and the GO:0030593:

299 module (Figure 5C). This module is of particular interest because of the increased

300 neutrophilia observed in the rewilded mice (Yeung et.al. companion paper), which is driven

301 by increased fungal colonization. Indeed, this module is tightly linked with cytokine

302 responses to C. albicans antigen stimulation (Figure 5C). While expressions of genes in this

303 module are increased by the rewilding environment, there does not appear to be genotype

304 specific differences regulating expression of the genes in this module (Figure 5D). This multi-

305 omics network analysis therefore substantiated the omics-by-omics inferences (Figures 1 and

306 2) but also provided new insight by identifying an important role for C. albicans responses

307 during rewilding, which is more fully addressed in the companion study (Yeung et.al.

308 companion paper).

309

310

311 DISCUSSION

312 In summary, we found that changing the environment profoundly contributes to the

313 inter-individual variations on immune cell frequencies, whereas cytokine responses to

314 pathogen stimulation were more affected by genetic deficiencies in the IBD susceptibility

315 genes Nod2 and Atg16l1. These observations are consistent with human twin studies whereby

316 the majority of cell population influences (measured by flow and mass cytometry) are

317 determined by non-heritable factors (Brodin et al., 2015), whereas cytokine production

318 capacity in response to stimulation may be more strongly influenced by genetic factors (Li et

319 al., 2016). Another large human study indicated that features of innate immunity were more

320 strongly controlled by genetic variation than lymphocytes, which were driven by

321 environmental effects (Patin et al., 2018). Thus, although studies of lab mice, by design,

322 often emphasize genetic effects on immune phenotype (including cell population

323 distributions), here we show that even brief exposure to a natural environment renders bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

324 predictors of immune phenotype in mice a better match for predictors of human immune

325 phenotypes. Additionally, our dataset described here will be a useful resource for other

326 investigators to delve into the immunological consequences of rewilding.

327 An attractive hypothesis arising from these findings is that perhaps environment is a

328 primary driver of the composition of the immune system, but genetics is a stronger driver of

329 per-cell responsiveness. However, in this study we only investigated how Nod2 and Atg16l1

330 influenced the immunological parameters we measured. To test this hypothesis further would

331 require us to release mice from diverse genetic backgrounds into the outdoor enclosure in a

332 similar experiment. This experiment only examined the effects of Nod2 and Atg16l1

333 deficiency in the C57BL/6 background. For example, analysis of macrophage activation from

334 five different strains of mice that provided genetic variation on the order of the human

335 population yielded considerable insights into how genetic variation affects transcriptional

336 regulation mechanisms (Link et al., 2018). Alternatively, mice from the collaborative cross

337 may enable high-resolution genome mapping for such complex traits as cytokine production

338 in response to stimulation (Noll et al., 2019). Studies of wild animal populations may also

339 determine if per cell cytokine responses show a higher genetic variance than do immune cell

340 population distributions. Hence, the combination of new environmental challenge strategies

341 such as rewilding of mouse models, combined with modern multi-omic approaches in fully

342 wild systems, should enable us to better define determinants of immune variation at a

343 molecular level.

344 This study was initially designed to test the hypothesis that Nod2 and Atg16l1

345 deficient mice may respond to the re-wilding environment in an adverse way. Our previous

346 studies had found that B. vulgatus colonization of Nod2 and norovirus colonization of Atg16l1

347 deficient mice predisposed them to intestinal inflammation. Hence, we wanted to examine if

348 these mice with mutations in genes associated with the development of IBD would be

349 associated with environmental triggering of intestinal inflammation. However, we did not

350 observe significant differences in intestinal inflammation based on histology for either the

351 Nod2 or Atg16l1 deficient mice, although we find an elevated production of cytokines in the bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

352 Nod2 deficient mice in response to the rewilding environment. While there were not

353 significant differences in intestinal inflammation, it is possible that additional insults (i.e. a

354 multi-hit model) are needed to trigger intestinal inflammation that is sufficiently severe to be

355 discernable by histologic examination. In our previous studies, an additional piroxicam (for

356 the Nod2 deficient mice (Ramanan et al., 2016) or dextran sodium sulfate (DSS) (for the

357 Atg16l1 deficient mice (Matsuzawa-Ishimoto et al., 2017)) insult was required to drive

358 pathogenesis of the B. vulgatus/norovirus colonized mice.

359 A systems level analysis identified interconnected networks of transcriptional

360 signatures, immune cell populations and cytokine profiles to microbial stimulation,

361 highlighting a potentially important role for fungal stimulation from rewilding (Companion

362 study, Yeung et. al.). This unanticipated result may have been missed through conventional

363 comparisons. Increased colonization by commensal fungi is perhaps the most striking

364 environmental effect of the rewilding experiment that we performed. While Nod2 deficiency

365 did not affect fungal colonization, Atg16l1 deficient mice are more susceptible (Companion

366 study, Yeung et. al.). Future studies with mice from more genetically diverse backgrounds

367 will better define the genetic basis of host susceptibility to commensal fungi colonization

368 through rewilding and determine how this may affect the innate and adaptive immune

369 responses.

370 Hence, rewilding laboratory mice of different genetic backgrounds and careful

371 monitoring of other non-heritable influences (e.g. differential behavior via which divergent

372 immunological experience of individuals may accrue, by analogy with causes of divergence

373 in neurological phenotype (Freund et al., 2013) observed during rewilding (Cope et al., 2019)

374 could be a powerful new approach towards dissecting drivers of immune response

375 heterogeneity under homeostatic as well as during disease settings or infectious challenges.

376

377

378

379 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

380

381 References:

382 (2015). Gene Ontology Consortium: going forward. Nucleic Acids Res. 43(Database 383 issue), D1049-1056. Published online 2014/11/28 DOI: gku1179 [pii] 384 10.1093/nar/gku1179. 385 386 Abolins, S., King, E.C., Lazarou, L., Weldon, L., Hughes, L., Drescher, P., Raynes, J.G., 387 Hafalla, J.C.R., Viney, M.E., and Riley, E.M. (2017). The comparative immunology of 388 wild and laboratory mice, Mus musculus domesticus. Nat Commun. 8, 14811. Published 389 online 2017/05/04 DOI: ncomms14811 [pii] 390 10.1038/ncomms14811. 391 392 Ashburner, M., Ball, C.A., Blake, J.A., Botstein, D., Butler, H., Cherry, J.M., Davis, A.P., 393 Dolinski, K., Dwight, S.S., Eppig, J.T., et al. (2000). Gene ontology: tool for the 394 unification of biology. The Gene Ontology Consortium. Nat Genet. 25(1), 25-29. 395 Published online 2000/05/10 DOI: 10.1038/75556. 396 397 Bakker, O.B., Aguirre-Gamboa, R., Sanna, S., Oosting, M., Smeekens, S.P., Jaeger, M., 398 Zorro, M., Vosa, U., Withoff, S., Netea-Maier, R.T., et al. (2018). Integration of multi- 399 omics data and deep phenotyping enables prediction of cytokine responses. Nat 400 Immunol. 19(7), 776-786. Published online 2018/05/23 DOI: 10.1038/s41590-018-0121-3 401 10.1038/s41590-018-0121-3 [pii]. 402 403 Becht, E., McInnes, L., Healy, J., Dutertre, C.A., Kwok, I.W.H., Ng, L.G., Ginhoux, F., 404 and Newell, E.W. (2019). Dimensionality reduction for visualizing single-cell data using 405 UMAP. Nat Biotechnol. Published online 2018/12/12 DOI: nbt.4314 [pii] 406 10.1038/nbt.4314. 407 408 Beura, L.K., Hamilton, S.E., Bi, K., Schenkel, J.M., Odumade, O.A., Casey, K.A., 409 Thompson, E.A., Fraser, K.A., Rosato, P.C., Filali-Mouhim, A., et al. (2016). 410 Normalizing the environment recapitulates adult human immune traits in laboratory 411 mice. Nature. 532(7600), 512-516. Published online 2016/04/21 DOI: nature17655 [pii] 412 10.1038/nature17655. 413 414 Biswas, A., Liu, Y.J., Hao, L., Mizoguchi, A., Salzman, N.H., Bevins, C.L., and 415 Kobayashi, K.S. (2010). Induction and rescue of Nod2-dependent Th1-driven 416 granulomatous inflammation of the ileum. Proc Natl Acad Sci U S A. 107(33), 14739- 417 14744. Published online 2010/08/04 DOI: 1003363107 [pii] 418 10.1073/pnas.1003363107. 419 420 Brodin, P., and Davis, M.M. (2017). Human immune system variation. Nat Rev 421 Immunol. 17(1), 21-29. Published online 2016/12/06 DOI: nri.2016.125 [pii] 422 10.1038/nri.2016.125. 423 424 Brodin, P., Jojic, V., Gao, T., Bhattacharya, S., Angel, C.J., Furman, D., Shen-Orr, S., 425 Dekker, C.L., Swan, G.E., Butte, A.J., et al. (2015). Variation in the human immune 426 system is largely driven by non-heritable influences. Cell. 160(1-2), 37-47. DOI: 427 10.1016/j.cell.2014.12.020. 428 429 Cadwell, K., Patel, K.K., Maloney, N.S., Liu, T.C., Ng, A.C., Storer, C.E., Head, R.D., 430 Xavier, R., Stappenbeck, T.S., and Virgin, H.W. (2010). Virus-plus-susceptibility gene 431 interaction determines Crohn's disease gene Atg16L1 phenotypes in intestine. Cell. 432 141(7), 1135-1145. Published online 2010/07/07 DOI: S0092-8674(10)00545-3 [pii] bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

433 10.1016/j.cell.2010.05.009. 434 435 Caruso, R., Mathes, T., Martens, E.C., Kamada, N., Nusrat, A., Inohara, N., and Nunez, 436 G. (2019). A specific gene-microbe interaction drives the development of Crohn's 437 disease-like colitis in mice. Sci Immunol. 4(34). Published online 2019/04/21 DOI: 438 4/34/eaaw4341 [pii]10.1126/sciimmunol.aaw4341. 439 440 Cope, E.C., Opendak, M., LaMarca, E.A., Murthy, S., Park, C.Y., Olson, L.B., Martinez, 441 S., Leung, J.M., Graham, A.L., and Gould, E. (2019). The effects of living in an outdoor 442 enclosure on hippocampal plasticity and anxiety-like behavior in response to nematode 443 infection. Hippocampus. 29(4), 366-377. Published online 2018/09/27 DOI: 444 10.1002/hipo.23033. 445 446 Freund, J., Brandmaier, A.M., Lewejohann, L., Kirste, I., Kritzler, M., Kruger, A., 447 Sachser, N., Lindenberger, U., and Kempermann, G. (2013). Emergence of individuality 448 in genetically identical mice. Science. 340(6133), 756-759. Published online 2013/05/11 449 DOI: 340/6133/756 [pii]10.1126/science.1235294. 450 451 Kernbauer, E., Ding, Y., and Cadwell, K. (2014). An enteric virus can replace the 452 beneficial function of commensal bacteria. Nature. 516(7529), 94-98. DOI: 453 10.1038/nature13960. 454 455 Lavoie, S., Conway, K.L., Lassen, K.G., Jijon, H.B., Pan, H., Chun, E., Michaud, M., 456 Lang, J.K., Gallini Comeau, C.A., Dreyfuss, J.M., et al. (2019). The Crohn's disease 457 polymorphism, ATG16L1 T300A, alters the gut microbiota and enhances the local 458 Th1/Th17 response. Elife. 8. Published online 2019/01/23 DOI: 10.7554/eLife.39982 459 39982 [pii]. 460 461 Leung, J.M., Budischak, S.A., Chung The, H., Hansen, C., Bowcutt, R., Neill, R., 462 Shellman, M., Loke, P., and Graham, A.L. (2018). Rapid environmental effects on gut 463 nematode susceptibility in rewilded mice. PLoS Biol. 16(3), e2004108. Published online 464 2018/03/09 DOI: 10.1371/journal.pbio.2004108 465 pbio.2004108 [pii]. 466 467 Li, Y., Oosting, M., Smeekens, S.P., Jaeger, M., Aguirre-Gamboa, R., Le, K.T.T., Deelen, 468 P., Ricano-Ponce, I., Schoffelen, T., Jansen, A.F.M., et al. (2016). A Functional Genomics 469 Approach to Understand Variation in Cytokine Production in Humans. Cell. 167(4), 470 1099-1110 e1014. Published online 2016/11/05 DOI: S0092-8674(16)31400-3 [pii] 471 10.1016/j.cell.2016.10.017. 472 473 Link, V.M., Duttke, S.H., Chun, H.B., Holtman, I.R., Westin, E., Hoeksema, M.A., Abe, 474 Y., Skola, D., Romanoski, C.E., Tao, J., et al. (2018). Analysis of Genetically Diverse 475 Macrophages Reveals Local and Domain-wide Mechanisms that Control Transcription 476 Factor Binding and Function. Cell. 173(7), 1796-1809 e1717. Published online 2018/05/22 477 DOI: S0092-8674(18)30511-7 [pii]10.1016/j.cell.2018.04.018. 478 479 Matsuzawa-Ishimoto, Y., Shono, Y., Gomez, L.E., Hubbard-Lucey, V.M., Cammer, M., 480 Neil, J., Dewan, M.Z., Lieberman, S.R., Lazrak, A., Marinis, J.M., et al. (2017). 481 Autophagy ATG16L1 prevents necroptosis in the intestinal epithelium. J Exp 482 Med. 214(12), 3687-3705. Published online 2017/11/02 DOI: jem.20170558 [pii] 483 10.1084/jem.20170558. 484 485 Noll, K.E., Ferris, M.T., and Heise, M.T. (2019). The Collaborative Cross: A Systems 486 Genetics Resource for Studying Host-Pathogen Interactions. Cell Host Microbe. 25(4), 487 484-498. Published online 2019/04/12 DOI: S1931-3128(19)30161-1 [pii] bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

488 10.1016/j.chom.2019.03.009. 489 490 Patin, E., Hasan, M., Bergstedt, J., Rouilly, V., Libri, V., Urrutia, A., Alanio, C., 491 Scepanovic, P., Hammer, C., Jonsson, F., et al. (2018). Natural variation in the 492 parameters of innate immune cells is preferentially driven by genetic factors. Nat 493 Immunol. 19(3), 302-314. Published online 2018/02/25 DOI: 10.1038/s41590-018-0049-7 494 10.1038/s41590-018-0049-7 [pii]. 495 496 Pott, J., Kabat, A.M., and Maloy, K.J. (2018). Intestinal Epithelial Cell Autophagy Is 497 Required to Protect against TNF-Induced Apoptosis during Chronic Colitis in Mice. 498 Cell Host Microbe. 23(2), 191-202 e194. DOI: 10.1016/j.chom.2017.12.017. 499 500 Ramanan, D., Bowcutt, R., Lee, S.C., Tang, M.S., Kurtz, Z.D., Ding, Y., Honda, K., 501 Gause, W.C., Blaser, M.J., Bonneau, R.A., et al. (2016). Helminth infection promotes 502 colonization resistance via type 2 immunity. Science. 352(6285), 608-612. Published 503 online 2016/04/16 DOI: science.aaf3229 [pii]10.1126/science.aaf3229. 504 505 Ramanan, D., Tang, M.S., Bowcutt, R., Loke, P., and Cadwell, K. (2014). Bacterial 506 sensor Nod2 prevents inflammation of the small intestine by restricting the expansion of 507 the commensal Bacteroides vulgatus. Immunity. 41(2), 311-324. DOI: 508 10.1016/j.immuni.2014.06.015. 509 510 Rosshart, S.P., Herz, J., Vassallo, B.G., Hunter, A., Wall, M.K., Badger, J.H., 511 McCulloch, J.A., Anastasakis, D.G., Sarshad, A.A., Leonardi, I., et al. (2019). 512 Laboratory mice born to wild mice have natural microbiota and model human immune 513 responses. Science. 365(6452). DOI: 10.1126/science.aaw4361. 514 515 Rosshart, S.P., Vassallo, B.G., Angeletti, D., Hutchinson, D.S., Morgan, A.P., Takeda, K., 516 Hickman, H.D., McCulloch, J.A., Badger, J.H., Ajami, N.J., et al. (2017). Wild Mouse 517 Gut Microbiota Promotes Host Fitness and Improves Disease Resistance. Cell. 171(5), 518 1015-1028 e1013. DOI: 10.1016/j.cell.2017.09.016. 519 520 Schirmer, M., Kumar, V., Netea, M.G., and Xavier, R.J. (2018). The causes and 521 consequences of variation in human cytokine production in health. Curr Opin Immunol. 522 54, 50-58. Published online 2018/06/19 DOI: S0952-7915(17)30173-5 [pii] 523 10.1016/j.coi.2018.05.012. 524 525 Schirmer, M., Smeekens, S.P., Vlamakis, H., Jaeger, M., Oosting, M., Franzosa, E.A., 526 Horst, R.T., Jansen, T., Jacobs, L., Bonder, M.J., et al. (2016). Linking the Human Gut 527 Microbiome to Inflammatory Cytokine Production Capacity. Cell. 167(7), 1897. 528 Published online 2016/12/17 DOI: S0092-8674(16)31658-0 [pii] 529 10.1016/j.cell.2016.11.046. 530 531 Suzuki, T.A., Phifer-Rixey, M., Mack, K.L., Sheehan, M.J., Lin, D., Bi, K., and 532 Nachman, M.W. (2019). Host genetic determinants of the gut microbiota of wild mice. 533 Mol Ecol. 28(13), 3197-3207. Published online 2019/05/30 DOI: 10.1111/mec.15139. 534 535 Ter Horst, R., Jaeger, M., Smeekens, S.P., Oosting, M., Swertz, M.A., Li, Y., Kumar, V., 536 Diavatopoulos, D.A., Jansen, A.F.M., Lemmers, H., et al. (2016). Host and 537 Environmental Factors Influencing Individual Human Cytokine Responses. Cell. 167(4), 538 1111-1124 e1113. Published online 2016/11/05 DOI: S0092-8674(16)31401-5 [pii] 539 10.1016/j.cell.2016.10.018. 540 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

541 Wlodarska, M., Kostic, A.D., and Xavier, R.J. (2015). An integrative view of 542 microbiome-host interactions in inflammatory bowel diseases. Cell Host Microbe. 17(5), 543 577-591. DOI: 10.1016/j.chom.2015.04.008. 544 545 Wong, S.Y., and Cadwell, K. (2018). There was collusion: Microbes in inflammatory 546 bowel disease. PLoS Pathog. 14(9), e1007215. DOI: 10.1371/journal.ppat.1007215. 547 548 549 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

550 Acknowledgements 551 We wish to thank William Craigens, Daniel Navarrete Prado, Allison Lee, and Veena 552 Chittamuri for assistance with trapping and husbandry in the field, the PU Lab Animal 553 Resources staff for logistical support. We wish to thank the NYU School of Medicine Flow 554 Cytometry and Cell Sorting, Microscopy, Genome Technology, and Histology Cores for use 555 of their instruments and technical assistance (supported in part by National Institute of Health 556 (NIH) grant P31CA016087, S10OD01058, and and S10OD018338). Funding: this research 557 was supported by US National Institute of Health (NIH) grants DK103788 (K.C. and P.L.), 558 AI121244 (K.C.), HL123340 (K.C.), DK093668 (K.C.), AI130945 (P.L.), R01 HL125816 559 (K.C.), HL084312, AI133977 (P.L.), research station and research rebate awards from PU 560 EEB (A.L.G.), pilot award from the NYU CTSA grant UL1TR001445 from the National 561 Center for Advancing Translational Sciences (NCATS) (K.C., P.L.), pilot award from the 562 NYU Cancer Center grant P30CA016087 (K.C.), AI100853 (Y.C.), and DK122698 (F.Y.). 563 This work was also supported by the Department of Defense grant W81XWH-16-1-0256 564 (P.L.), Faculty Scholar grant from the Howard Hughes Medical Institute (K.C.), Crohn’s & 565 Colitis Foundation (K.C.), Merieux Institute (K.C.), Kenneth Rainin Foundation (K.C.), 566 Stony-Wold Herbert Fund (K.C.), and Bernard Levine Postdoctoral Research Fellowship in 567 Immunology (Y.C.). K.C. is a Burroughs Wellcome Fund Investigator in the Pathogenesis of 568 Infectious Diseases. Author contributions: Design of experiments, data analysis, data 569 discussion, and interpretation: J.D.L., J.C.D., F.Y., K.C., A.L.G, and P.L.; primary 570 responsibility for execution of experiments: J.D.L., F.Y., C.M., J.M.L., Y.H.C., A.C., C.H., 571 and C.D.D.; MLN cell RNA and 16S analysis: J.C.D., K.V.R., J.D.L., and F.Y. Supervised 572 and Unsupervised machine learning model analysis: J.C.D., K.V.R., and J.D.L. All authors 573 discussed data and commented on the manuscript. Competing interests: K.C. and P.L. 574 receive research funding from Pfizer. K.C. has consulted for or received an honorarium from 575 Puretech Health, Genentech, and Abbvie. P.L. consults for and has equity in Toilabs. Data 576 and materials availability: Raw sequence data from 16S, ITS, and RNA sequencing 577 experiments are deposited in the NCBI Sequence Read Archive under BioProject accession 578 number PRJNA559026 and gene expression omnibus (GEO) accession number GSE135472. bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

579 Figure Legends: 580 581 Figure 1. Environmental change drives inter-individual variation in immune cell 582 populations. (A) 25 C57BL/6+/+, 28 Nod2-/-, 27 Atg16l1T316A/+, 24 Atg16l1T316A/T316A mice 583 (total=104) were housed in the wild enclosure (Rewilded) for 6-7 weeks and successfully 584 trapped for flow cytometry analysis of lymphoid and myeloid cell populations in the blood 585 and mesenteric lymph nodes (MLNs). Plasma was also collected for cytokine profiles. Age 586 matched 19 C57BL/6+/+, 19 Nod2-/-, 20 Atg16l1T316A/+, 22 Atg16l1T316A/T316A mice (total=80) 587 housed under SPF conditions (Lab) were analyzed as controls. Feces were collected for 16S 588 rRNA sequencing. MLN cells were cultured with indicated stimulates for 48 hours and 589 supernatants were collected for cytokine profiling. Samples that fail quality control are not 590 included in downstream analyses. (B and C) UMAP projections of ~180,000 CD45+, 591 ~110,000 CD19+, and ~36,000 CD4+ cells on flow cytometry data of the blood from lab and 592 rewilded mice. Events are color-coded according to CD19, CD44 (B), CD4, and CD62L (C) 593 fluorescence level. Box plots show the abundance of CD44hi CD19+ cells (B) and CD62Lhi 594 CD4+ cells (C) in the blood of individual lab and rewilded mice. (D) Principal component 595 analysis (PCA) of gated immune cell populations in the blood and the density of each 596 population along the principal component (PC). Right panel indicate biplots of the gated 597 immune cell populations that are projected onto PC1 and PC2. (E) Box plot of pairwise 598 Euclidean distance measures based on blood immune cell population in the blood of lab 599 versus lab, rewilded versus rewilded and lab versus rewilded mice. (F) Bar plot showing the 600 pseudo R2 measure of effect size on the entire distance matrix used to calculate the PCA of 601 immune cell populations in the blood (D). ***P < 0.001 by Kruskal-Wallis test between 602 groups (B, C) and ***P < 0.001 by Mann-Whitney-Wilcoxon test between groups (E). 603 604 Figure 2. The genetic contribution towards variation in cytokine production in response 605 to microbial stimulation is masked when mice are rewilded. (A to D) MLN cells from lab 606 and rewilded mice were ex vivo cultured with B. subtilis, S. aureus, C. perfringens, B. 607 vulgatus, P. aeruginosa, C. albicans, and αCD3/CD28 beads for 48 hours and the 608 supernatants were assayed for 13 cytokines. (A) Heat map of average fold changes compared 609 to cytokine levels of MLN cells cultured with PBS controls from all mice. (B) Box plot

610 showing the top 10 log2 fold changes of cytokine responses to stimulation for MLN cells 611 from lab and rewilded mice. (C) PCA plot of of fold change cytokine stimulation profiles 612 with histograms of the total variance explained by each PC indicated on the axis. Right panel 613 indicates the loading factor of each stimulated cytokine profile that contributes to each PC. (D) 614 Bar plot showing the pseudo R2 measure of effect size on the entire distance matrix used to bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

615 calculate the PCA of cytokine profiles (C). * P <0.05, ** P <0.01, *** P < 0.001 by Mann- 616 Whitney-Wilcoxon test between groups (B). 617 618 Figure 3. Rewilded Nod2-/- mice display greater changes in cytokine production 619 compared to rewilded WT mice. (A to C) MLN cells from lab and rewilded mice were ex 620 vivo cultured with B. subtilis, S. aureus, C. perfringens, B. vulgatus, P. aeruginosa, C. 621 albicans, and αCD3/CD28 beads for 48 hours and the supernatants were assayed for 13 622 cytokines. (A and B) Heatmap of significant p-values from Mann-Whitney-Wilcoxon test 623 between group results comparing fold changes of stimulated cytokine production in rewilded 624 (A) and lab (B) WT mice against Nod2-/-, Atg16l1T316A/+ and Atg16l1T316A/T316A mice 625 respectively. Within each cytokine condition the order of stimulation remains constant as 626 follows; B. subtilis, B. vulgatus. C. albicans, αCD3/CD28, C. perfringens, P. aeruginosa, S. 627 aureus. (C) Representative box plots of the most significant changes between WT and Nod2-/- 628 mice. * P <0.05, ** P <0.01, *** P < 0.001 by Mann-Whitney-Wilcoxon test between groups 629 (C). 630 631 Figure 4. Multi-omic classification models identify predictive features of environment 632 and genotype. (A) Bar plot of the top 40 features predictive of the environment and (B) 633 interaction network analysis of correlations (R2 > 0.6 or R2 < -0.6) between features (genes, 634 immune cell populations, OTUs or cytokines) that are predictive of environmental changes. 635 (C) Bar plot of the top 40 features predictive of the genotype and (D) network analysis of the 636 associations between features that are associated with genotype differences. Data types are 637 colored according to cytokine profiles (green), genes (red), immune populations (blue), and 638 OTUs (purple). The size of the nodes in B and D are scaled proportionally to the number of 639 connections with other nodes in each network. 640 641 Figure 5. An unsupervised multi-component model identifies a C. albicans and C. 642 perfringens response network associated with environmental changes. (A) Data matrices 643 from flow cytometry data, transcriptional profiles, microbial profiles and stimulated cytokine 644 profiles were used as inputs for a sPLS-regression model to generate a covariance network. (B) 645 The most connected nodes, Blood CD4 T-cells CD44hi CD62Lhi, IL-5 in response to C. 646 perfringens, and GO:0061844 (antimicrobial humoral immune response mediated by 647 antimicrobial peptides) were evaluated for a difference between lab and rewilded mice 648 populations. (C) The IL-5 C. perfringens subnetwork was derived from the network and re- 649 calculated to highlight the connections most related to IL-5 production. (D) GO: 0030593 650 (neutrophil chemotaxis) was identified as a significant connection in the subnetwork and the bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

651 genes found in that GO term were plotted as a heatmap. Color bars indicate environment and 652 genotype. The size of the nodes in (A) and (C) are scaled proportionally to the number of 653 connections with other nodes in each network. Data types are colored according to cytokine 654 profiles (green), GO term (red), immune populations (blue). * P <0.05, ** P <0.01, *** P < 655 0.001 by Mann-Whitney-Wilcoxon test between groups (B). 656 657 Figure S1: Identification of immune cell populations that differ between lab and 658 rewilded mice with UMAP. (A) Projection of ~180,000 CD45+ cells visualized by tSNE vs 659 UMAP. (B) Identification of the major lymphocyte populations on UMAP based on CD19, 660 CD3, CD4 and CD8 expression. (C to E) Expression of cell surface molecules on CD19+ 661 cells (C), CD4+ cells (D) and CD8+ cells (E). In (C), (D) and (E), expression of 1B11, PD1, 662 CD25, CD44, CD127, CXCR3, KLRG1, CD27, CD69, and CD62L are shown as high (purple) 663 or low (grey) from cut-off values based on overall distribution of fluorescence. 664 665 Figure S2: Flow cytometry gating strategy for the quantification of lymphoid 666 populations in the blood of lab and rewilded mice. (A to H) Representative flow cytometry 667 plots illustrating gating strategy (A) of lymphoid populations in the blood including the total 668 CD45+ cells (B), total myeloid cells (C), total B and T cells (D), total CD4 and CD8 T cells 669 (E), CD44loCD62Lhi, CD44hiCD62Lhi, CD44hiCD62Llo, CD44loCD62Llo CD8 T cells (F), 670 CD44loCD62Lhi, CD44hiCD62Lhi, CD44hiCD62Llo, CD44loCD62Llo CD4 T cells (G), and 671 CD25+CD4+ T cells (H). 672 673 Figure S3: A genetic contribution towards variation in plasma cytokine levels of lab and 674 rewilded mice. (A and B) Levels of RANTES, MIP-3a, KC, TNF-α, MCP-1, IP-10, MIP-1a, 675 MIP-1b, IL-1α, IL-6 cytokines were measured from the plasma of lab and rewilded mice at 676 the time of sacrifice with bead-based immunoassays (LEGENDplex). (A) PCA of plasma 677 cytokine levels in the blood of individual mice and the density of each population along the 678 principal components (PC). Right panel indicate biplots of the plasma cytokines are projected 679 onto PC1 and PC2. (B) Bar plot showing the pseudo R2 measure of the effect size of different 680 variables on the variance of plasma cytokine levels. 681 682 Figure S4: Memory CD4+ and CD8+ T cells are increased in the mesenteric lymph 683 nodes (MLNs) of rewilded mice. (A to F) Representative flow cytometry plots for immune 684 cell populations after gating for CD45+ cells in the MLNs of lab or rewilded C57BL/6 mice; 685 for the total abundance of myeloid cells (A), B and T cells (B), CD4 and CD8 T cells (C), 686 CD44loCD62Lhi, CD44hiCD62Lhi, CD44hiCD62Llo, CD44loCD62Llo CD4 T cells (D), bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

687 CD44loCD62Lhi, CD44hiCD62Lhi, CD44hiCD62Llo, CD44loCD62Llo CD8 T cells (E), and 688 CD25+CD4+ T cells (F). 689 690 Figure S5: Environment has the greatest effect on variation of transcriptional profiles 691 and microbial composition. (A to B) Transcriptional profiles of MLNs were determined by 692 Celseq. (A) PCA based on gene profiles in the MLN and the density of each population along 693 the principal components (PC). Right panel indicates biplots of the genes projected onto PC1 694 and PC2. (B) Bar plot showing the pseudo R2 measure of effect size on the entire distance 695 matrix used to calculate the PCA of gene expression levels. (C) PCoA based on Bray Curtis 696 distances of 16s rRNA sequencing profiles in all lab and rewilded mice across all genotypes, 697 (D) lab wild type (WT) and Atg16l+/+ only, (E) lab wild type (WT), Atg16l+/+ and Nod2-/- and 698 (F) lab wild type (WT), Atg16l+/+ and Nod2-/- by respective cages. P-values were determined 699 by an Adonis test from Bray Curtis distances. 700 701 Figure S6: Unmeasured environmental effects drive additional variation of MLN 702 immune cell populations for the rewilded mice. (A and B) PCA based on lymphoid 703 populations in the MLN and the density of each population along the principal components 704 (PC). In plot (A), dot color coded by rewilded (red) and lab (purple) and in plot (B), dot color 705 coded by lab (green), rewilded week 6 (yellow), rewilded week 7 (blue) (C) Biplots of the 706 MLN lymphocyte populations are projected onto PC1 and PC2. (D) Bar plot showing the 707 pseudo R2 measure of effect size on the entire distance matrix used to calculate the PCA of 708 immune populations. 709 710 Figure S7: An unsupervised PLS-regression multi-component network model for multi- 711 omic integration. A fully annotated multi-component network as shown in Figure 5A. 712 713 714 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 2 ● Rewilded A B ● Lab IFN− CD3/CD28

γ α ***

CCL4 αCD3/CD28 ** B. subtilis

TNF−α αCD3/CD28 *** S. aureus

IL−5 αCD3/CD28 *** C. perfringens Log2

CXCL1 B. subtilis *** FoldChange B. vulgatus 6 IL−10 B. vulgatus *** 4

P. aeruginosa 2 CCL4 B. vulgatus *** 0 C. albicans CCL4 S. aureus ** Cytokine Stimulation

IL−10 C. albicans * αCD3/CD28 IL−10 P. aeruginosa γ α *** IL−5 IL−6 CCL3 IFN− IL−17 IL−1α IL−1β IL−13 CCL2 CCL4 IL−10 TNF− CXCL1 0 5 10 Log2 Fold Change C Lab Rewilded D ● Atg16l1T316A/+ IL−10 P. aeruginosa Atg16l1T316A/T316A Genotype IL-10 B. vulgatus IL-10 αCD3/CD28 ● WT Nod2-/- IL-10 C. perfringens IL-10 B. subtilis IL-10 C. albicans Environment ● Rewilded ● Lab ● ● ● ● Diarrhea ● ●● ● ● ● ● ● ● ● ● IL-6 C. perfringens Sex ● PC2 4.69% expl. variation ● ● IL-6 B. subtilis IL-5 S. aureus WeightGain IL-6 αCD3/CD28 IL-6 B. vulgatus IL−6 P. aeruginosa

PC1 5.86% expl. variation 0.00 0.04 0.08 0.12 Effect Size bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

A B Figure 3 C IL-17 C. perfringens IL-5 C. perfringens Rewilded Lab * *** CCL2 CCL2 *** 6 3 CCL3 CCL3

CCL4 CCL4 4 2 CXCL1 CXCL1 2 1 IFN-γ IFN-γ Log2 FoldChange

IL-10 IL-10 0 −log10 0 IL-13 IL-13 p−value 5 IL-17 IL-17 4 IL-10 C. perfringens IL-10 C. albicans 3 ** 10.0 ** IL-1α IL-1α ** ** 2 ** 1 ** IL-1β IL-1β 8 7.5 IL-5 IL-5 5.0

IL-6 IL-6 4

2.5 TNF-α TNF-α Log2 FoldChange

-/- -/-

T316A/+ T316A/+ 0 0.0 Nod2 Nod2 T316A/T316A T316A/T316A

Atg16l1 Atg16l1 -/- -/- WT vs. WT vs. WT WT Atg16l1 Atg16l1 T316A/+ T316A/+ T316A/T316ANod2 T316A/T316ANod2

WT vs. WT vs. Atg16l1 Atg16l1 WT vs. WT vs. Atg16l1 Atg16l1 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Figure 4 A B Environment Model Features Zfp36 Atf4 Ubc DataType Cd69 MLN CD8 Jund Blood CD8 T-cells TNF-α αCD3/CD28 Cytokine CD44hi CD62Lhi MLN CD4 T-cells Eif1 Gene MLN CD8 T-cells T-cells Klf2 ImmunePop Blood CD4 T-cells Klf6 Blood CD8 T-cells Fos OTU CD44hi CD62Lhi● CD44lo CD62Lhi Nfkbia ● MLN CD4 T-cells Btg2 Cd79b MLN CD4 T-cells CD25+ Hspa1a Cd69 IL-5 αCD3/CD28 Hsp90aa1 ● ●Serinc3 Dnajb1 ● Atp5e Rn4.5s TNF-α αCD3/CD28 Blood CD4 T-cells CD44hi CD62Lhi ● Hspe1 Junb Eif1 Cd79b Jun H2−Eb1 ●Tmsb10 IL-5 αCD3/CD28 Dnajb1 ● Hspe1 Gm1821 Chchd2 Rps21 Ppp1r15a Nfkbia Hsp90aa1 Btg2 Rn4.5s Hspa5 ● Egr1 Jund Gm1821 Klf2 Dusp1 Serinc3 Atp5e Gimap6 Dusp1 Blood CD8 T-cells CD44lo CD62Lhi Zfp36 Egr1 Jun Klf6 ● Chchd2 Gimap6 Ppp1r15a IL-1β S. aureus Blood CD8 T-cells CD44hi CD62Lhi Hspa5 Junb Hspa1a Atf4 Rps21 Fos Ubc H2−Eb1 Tmsb10 MeanDecreaseGini

C D

Genotype Model Features MLN CD8 T-cells CD44hi CD62Lhi Hmha1 Erdr1 MLN CD8 T-cells H2−K1 CD44hi CD62Lhi MLN CD8 T-cells IL-17 P. aeruginosa CD44lo CD62Lhi MLN Eosinphils Blood CD8 T-cells Rps8 CD44lo CD62Llo Sell DataType MLN CD8 T-cells Ucp2 ● IL-10 C. perfringens Cytokine CD44lo CD62Llo MLN CD4 T-cells Rps9 Blood CD4 T-cells Rps5 Gene CD44hi CD62Lhi ● MLN CD4 T-cells CD44hi CD62Lhi ImmunePop CD44lo CD62Lhi Serinc3 MLN Ly6C OTU Rpl4 Rac2 Blood CD8 T-cells Cfl1 MLN Ly6C Jun ● Blood CD8 T-cells CD44lo CD62Llo CD44hi CD62Lhi MLN Total myeloid CD11b CD11c ●Monocyte MLN IL-6 P. aeruginosa Ucp2 Sell ●Cd69 MLN CD8 T-cells CD44lo CD62Llo ● ● Rac2 Rpl23 Rpl4 MLN CD8 T-cells CD44lo CD62Lhi Cfl1 Rps24 Rps26 IL-17 P. aeruginosa IFN-γ αCD3/CD28 ● Blood CD8 T-cells CD44hi CD62Lhi ● Rps7 Rpl23 Rps5 Jun ● BacteroidesV IL.17 Tpt1 ● Rpl32 Rps8 IL-17 B. vulgatus Rpsa Tpt1 Rpl32 Cox6a2 Cd69 Rps9 ● Rps24 Rps7 Rps26 CCL3 αCD3/CD28 IFN-γ C. perfringens Blood CD4 T-cells CD44lo CD62Lhi MeanDecreaseGini bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Blood CD4 T-cells CD44hi CD62Lhi 15 *** A B ● Blood CD4 T-cells MLN CD4 T-cells Data Type CD44hi CD62Lhi 10 ● ● Cytokine ● GO Term ● MLN CD8 ● ● Immune ● ● ● ● ● ● ● 5 ● Population T-cells ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ●● ●● ● ● ● ● ● ● ● ● 0 ● ● Lab Rewilded

GO:0061844 antimicrobial humoral immune response mediated by antimicrobial peptides 7 *** GO:0061844 antimicrobial ● humoral immune response 6 ● mediated by antimicrobial peptides

● 5 ● GO:0030593 neutrophil ●● chemotaxis ● 4 ● ● ● ● ● ● ● ● ●● ● Log2 Value ● ● ● ●● ● 3 ● ●● ● ● ●● ● ● ●● ● ● ●● ● ●● ● ● ● ● ●● ● ●● ● ● ●● ● ● ● ●● ● ●● ● ●● ● ● ● ● ●● ● ● ● 2 ●

● 1 Lab Rewilded IL−5 C. perfringens *** ●

4 ● ● ●

● 3 ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● 2 ● ● ● ●● ● ● ● ●● ●● ●●● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● 1 ● ● ● ● IL-5 ●● ● ● ● ● ● ● ●● ● ●● C. perfringens ● ● 0 ● Lab Rewilded

C D GO:0030593 neutrophil chemotaxis

MLN Total T-cells Rewilded Environment MLN CD4 T-cells MLN CD8 T-cells Lab Genotype Prex1 Ccl25 T316A/+ IL-5 Atg16L1 Itgam GO:0030593 IL-17 Atg16L1T316A/t316A Cxcl2 ● B. vulgatus Ccl5 neutrophil C. perfringens WT Ccl22 chemotaxis ● IL-5 Vav3 Nod2-/- Cxcl13 ● C. albicans Ccl19 ● ● Ccl2 IL-13 ● Ifng IL-10 IL-17 2 Spp1 C. albicans C. albicans 1 S100a9 C. albicans 0 S100a8 ●IL-5 Il1b −1 IL-5 B. subtilis Fcgr3 IFN-γ ● −2 Fcer1g C. perfringens Ccl9 C. albicans Lgals3 Cxcl9 Prkca IL-1α ● Ccl17 IL-1α Cxcl1 ● C. albicans Pf4 IL-1α B. subtilis Cklf Xcl1 B. vulgatus Ccl3 Ccl4 Cxcl10 Pde4b Ccl6 Cxcr2 Itga9 Slc37a4 Pde4d Gbf1 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Figure 1 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Figure 2 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Figure 3 A B

Serum Cytokines TNF-α

GenotypeGenotype ● ● ● Environment ● ● ● Atg16l1T316A/+ ● ●● KC Environment ●● Atg16l1T316A/T316A ●● ●● ● WT ● ● Nod2-/- RANTES WeightGain ● ●● ● MIP-1a MIP-3a WeightGain ●● ● ● Rewilded ● ● IP-10 ● ● Lab Gender MIP-1b PC2 20.32% expl. variation ● MCP-1 ●● ● ● Sex ● IL-1 ● α Diarrhea

Diarrhea Pregnant

0.000.00 0.01 0.01 0.02 0.03 0.02

PC1 31.43% expl. variation Effect Size bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Figure 4 bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Figure 5 A B MLN RNA−Seq ● Atg16l1T316A/+ Environment ● Atg16l1T316A/T316A WT ● ● Nod2-/- Cfd ●● Ctrb1 Genotype ● ● Rewilded ● ● ● Lab Try4 Pnlip Cpa1 ● Amy2b Try5 Cela1 Cel Cpb1 Amy2a5 ● Ctrl Pnliprp1 Sex ● Prss2 Cela3b PC2 4.74% expl. variation ●● Reg1 Zg16 ● Cela2a ● ● ● Gm13011 Rnase1 ● WeightGain ●

Diarrhea PC1 5.91% expl. variation

0.00 0.03 0.06 0.09 Effect Size C D

0.4 0.4 ● ● ● ● p = 0.703

Genotype Genotype +/+ 0.2 ● Atg16l1T316A/+ 0.2 Atg16l1 ● Atg16l1T316A/T316A WT ● WT Environment ● ●● -/- ● ● Nod2 ● Lab ● ● ● ● ● ● 0.0 ● ● ● 0.0 ● ● ● Environment ● ●●● ● ● Lab ● ● ● ●● ● Rewilded ● PCoA_2 12.12% expl. variation PCoA_2 12.12% expl. PCoA_2 12.12% expl. variation PCoA_2 12.12% expl. ● ● ●● ● ● ● ● −0.2 ● −0.2 ● ●

−0.4 −0.2 0.0 0.2 −0.4 −0.2 0.0 0.2 E PCoA_1 19.08% expl. variation F PCoA_1 19.08% expl. variation 0.4 0.4

p = 0.842 p = 0.712 ● ● ●

0.2 Genotype 0.2 ● Cage Atg16l1+/+ ● ● ● 2517957 WT ● ● 2517959 Nod2-/- ● ● 2537331 ● 0.0 Environment 0.0 ● ● ● ● 2537332 ● Lab ● ● PCoA_2 12.12% expl. variation PCoA_2 12.12% expl. PCoA_2 12.12% expl. variation PCoA_2 12.12% expl. ● ● −0.2 −0.2 ● ●

−0.4 −0.2 0.0 0.2 −0.4 −0.2 0.0 0.2 PCoA_1 19.08% expl. variation PCoA_1 19.08% expl. variation bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Figure 6 A C MLN Lymphocyte Panel CD8 T-cells CD44hi CD62Llo ● ● ● ●● Total myeloid CD11b CD11c ●● ● T316A/+ CD4 T-cells CD44hi CD62Llo ● Atg16l1 CD4 T-cells Atg16l1T316A/T316A CD44hi CD62Lhi CD4 T-cells CD44lo CD62Llo ● ●● ●● WT CD8 T-cells CD4 T-cells CD25+ ● -/- ● ● Nod2 CD44hi CD62Lhi ● ● CD4 T-cells ● ●● ● ●● CD8 T-cells CD44lo CD62Llo ●●●● ● Rewilded ● ● ● CD8 T-cells ● ● ● ● ● Lab ● ● ●

PC2 15.45% expl. variation ● ● ● ● ● ● CD4 T-cells CD44lo CD62Lhi CD8 T-cells CD44lo CD62Lhi

PC1 22.12% expl. variation B D

MLN Lymphocyte Panel ● ● ● Environment

●● T316A/+ ●● ● ● Atg16l1 Atg16l1T316A/T316A WT Genotype ● ● ● Nod2-/- ● ●● ● ● ●● ● ● Lab ● ●● ● ●●● ● ● Rewilded (Week 6) Diarrhea ● ● ●● ● ● ● ● ● Rewilded (Week 7) ●● ●●

PC2 15.45% expl. variation ● ● ● Sex ● ●

WeightGain

0.0 0.1 0.2 0.3 Effect Size PC1 22.12% expl. variation bioRxiv preprint doi: https://doi.org/10.1101/741470; this version posted December 9, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Supplementary Figure 7

Blood CD4 T-cells CD44hi CD62Lhi MLN CD4 T-cells

MLN Total immune cells CD45+

MLN CD11c MHCII

MLN CD4 T-cells MLN CD8 T-cells CD44hi CD62Lhi MLN Total T-cells MLN CD103neg CD11c MHCII DC

MLN CD4 T-cells CD44lo CD62Llo MLN CD4 T-cells Blood CD8 T-cells Blood Total T-cells MLN CD8 T-cells CD44lo CD62Lhi CD44lo CD62Llo Blood CD4 MLN CD8 T-cells Blood CD4 T-cells T-cells CD25+ CD44lo CD62Lhi CD44hi CD62Llo MLN Total myeloid Blood CD8 T-cells Blood CD4 T-cells CD11b CD11c Blood Total immune cells CD45+ CD44hi CD62Lhi MLN CD8 T-cells MLN Total B cells CD44hi CD62Llo MLN CD4 T-cells Blood Total myeloid Blood CD8 T-cells MLN CD103 CD44hi CD62Llo CD11c MHCII DC CD11b CD11c CD44lo CD62Lhi MLN Neutrophils MLN CD8 T-cells Blood CD4 T-cells CD44hi CD62Lhi Blood Total B cells Blood CD8 T-cells CD44lo CD62Llo Blood CD4 T-cellsCD44lo CD62Llo CD44lo CD62Lhi MLN Ly6C IL-6 IL-17 GO:0042113 GO:0033077 Blood CD8 T-cells Monocyte CD44hi CD62Llo C. albicans B. vulgatus CXCL1 IL-5 GO:0042110 IL-1 C. albicans B. vulgatus IL-10 P. aeruginosa α GO:0006955 CCL3 P. aeruginosa GO:0061844 C. albicans GO:0048535 IL-1β B. vulgatus CCL4 CD3/CD28 GO:0030593 IL-6 IL-1α α GO:0048536 GO:0002548 IL-1 GO:0002376 B. subtilis B. vulgatus α IL-6 P. aeruginosa CCL3 B. subtilis C. perfringens GO:0002281 CCL4 GO:0071346 CXCL1 B. vulgatus IL-10 TNF-α B. vulgatus IL-10 GO:0019731 GO:0002377 GO:0001913 C. albicans C. albicans GO:0019732 CCL3 B. vulgatus C. perfringens GO:0035855 CCL2 S. aureus CCL2 CXCL1 B. subtilis TNF-α C. albicans IL-6 S. aureus GO:0001779 GO:0071425 GO:0002315GO:0048821 IL-13 C. albicans GO:0030225 IL-1β IL-17 C. albicans GO:0030098 B. subtilis S. aureus CCL3 S. aureus IL-5 S. aureus GO:0002218 GO:0030218 IL-5 C. albicans CCL2 IL-10 GO:0030220 P. aeruginosa IL-5 C. perfringens GO:0002318 B. vulgatus IFN.y IL-5 αCD3/CD28 GO:0030217 GO:0042088 CCL4 P. aeruginosa CD3/CD28 GO:0001774 GO:0035162 IL-5 B. subtilis α IFN-γ S. aureus GO:0019886 GO:0002250 IL-10 B. subtilis GO:0002437 GO:0042098 IFN-γ C. perfringens GO:0050901 IL-13 C. perfringens IL-6 TNF-α IL-1β C. albicans GO:0001771 GO:0050852 CCL2 IL-13 B. subtilisB. vulgatus B. subtilis IL-1β TNF-α C. perfringens GO:0030099 αCD3/CD28 IL-6 CCL2 B. vulgatus GO:0030097 GO:0048539 GO:0031295 S. aureus GO:0030316 GO:0002260 αCD3/CD28 CCL2 P. aeruginosa CCL2 B. subtilis GO:0072540 IL-13 B. vulgatus GO:0051607 IL-6 IFN-γ GO:0001780 IL-17 S. aureus GO:0010818 GO:0002262 C. perfringens αCD3/CD28 GO:0048245 GO:0002755 IL-1α B. subtilis IL-10 S. aureus GO:0002476 CXCL1 IL-17 GO:0006958 GO:0002523 αCD3/CD28 IFN-γ GO:0016446 GO:0097028 B. subtilis P. aeruginosa GO:0043011 IL-10 IL-13 CCL3 P. aeruginosa CXCL1 B. vulgatus IL-13 C. albicans C. perfringens GO:0016064 αCD3/CD28 GO:0034138 CCL4 C. albicans GO:0050853 GO:0045190 CCL3 GO:0006956 GO:0002224 IL-1α CXCL1 B. vulgatus IL-1α C. perfringens αCD3/CD28 P. aeruginosa GO:0045087 GO:0042267 CXCL1 S. aureus GO:0043303 IL-1β αCD3/CD28 CCL3 IL-17 C. perfringens GO:0030595 C. perfringens GO:0019882 GO:0016447 IFN-γ C. albicans GO:0050900 IL-5 C. perfringens