Lysates of Methylococcus capsulatus Bath induce a lean-like microbiota, intestinal FoxP3+RORγt+IL-17+ Tregs and improve metabolism

Jensen, Benjamin A. H.; Holm, Jacob B.; Larsen, Ida S.; von Burg, Nicole; Derer, Stefanie; Sonne, Si B.; Pærregaard, Simone I.; Damgaard, Mads V.; Indrelid, Stine A.; Rivollier, Aymeric; Agrinier, Anne-Laure; Sulek, Karolina; Arnoldussen, Yke J.; Fjære, Even; Marette, André; Angell, Inga L.; Rudi, Knut; Treebak, Jonas T.; Madsen, Lise; Åkesson, Caroline Piercey; Agace, William; Sina, Christian; Kleiveland, Charlotte R.; Kristiansen, Karsten; Lea, Tor E.

doi:10.1038/s41467-021-21408-9

Download PDF

Article
Open access
Published: 17 February 2021

Lysates of Methylococcus capsulatus Bath induce a lean-like microbiota, intestinal FoxP3⁺RORγt⁺IL-17⁺ Tregs and improve metabolism

Nature Communications volume 12, Article number: 1093 (2021) Cite this article

6217 Accesses
22 Citations
11 Altmetric
Metrics details

Subjects

Abstract

Interactions between host and gut microbial communities are modulated by diets and play pivotal roles in immunological homeostasis and health. We show that exchanging the protein source in a high fat, high sugar, westernized diet from casein to whole-cell lysates of the non-commensal bacterium Methylococcus capsulatus Bath is sufficient to reverse western diet-induced changes in the gut microbiota to a state resembling that of lean, low fat diet-fed mice, both under mild thermal stress (T22 °C) and at thermoneutrality (T30 °C). Concomitant with microbiota changes, mice fed the Methylococcus-based western diet exhibit improved glucose regulation, reduced body and liver fat, and diminished hepatic immune infiltration. Intake of the Methylococcu-based diet markedly boosts Parabacteroides abundances in a manner depending on adaptive immunity, and upregulates triple positive (Foxp3⁺RORγt⁺IL-17⁺) regulatory T cells in the small and large intestine. Collectively, these data point to the potential for leveraging the use of McB lysates to improve immunometabolic homeostasis.

Microbiota in health and diseases

Article Open access 23 April 2022

A conserved complex lipid signature marks human muscle aging and responds to short-term exercise

Article 12 April 2024

Fasting-mimicking diet causes hepatic and blood markers changes indicating reduced biological age and disease risk

Article Open access 20 February 2024

Introduction

Gut microbes shape intestinal immunity¹ and increase the bioavailability of otherwise indigestible nutrients². A well-balanced community structure is therefore essential for immunometabolic homeostasis, whereas aberrant gut microbiota compositions associate with numerous diseases, both within and outside the gastrointestinal tract³.

While therapeutic implications of rebalancing a mistuned gut microbiota appear promising, inconsistent response rates in relation to both probiotics and fecal transfer studies, with occasional adverse events, emphasize the complexity of such approaches. One example relates to the otherwise promising probiotic candidate Akkermansia muciniphila⁴, where negative effects have been seen in immunocompromised recipients^5,6. Similarly, Prevotella copri aids in metabolizing fibers in healthy individuals⁷ and protects against bacterial invasion in high fiber, chow-fed mice⁸, yet associates with insulin resistance in prediabetic obese individuals and precipitates glucoregulatory impairments in diet-induced obese (DIO) mice⁹.

An alternative to administering viable microbes is to utilize whole-cell lysates, or selected cell components, of nonliving bacteria. Apart from alleviating global energy demands, if used as a nutrient source, such components may also potently affect host physiology as recently reported for A. muciniphila¹⁰ and Bifidobacterium bifidum¹¹. In the latter example, cell surface polysaccharides of B. bifidum were used to induce peripheral immune-tolerance via generation of regulatory T cells (T_regs). The authors reported a pronounced increase in Foxp3⁺RORγt⁺ T_regs (_pT_regs)_, specifically in lamina propria (LP) of the large intestine (LI)¹¹. This cell type is believed to be induced by commensal microbes and has emerged as a potent T_reg subset, exhibiting increased lineage stability and enhanced immunosuppressive capacity during intestinal inflammation compared to conventional Foxp3⁺RORγt^- T_regs (_nT_regs)¹².

RORγt is the canonical transcription factor controlling IL-17 expression; a pleiotropic cytokine with both proinflammatory and immune resolving actions depending on the eliciting cell type and physiological context¹³. Unfortunately, the studies describing _pT_reg function did not measure IL-17 secretion. It therefore remains unknown whether these cells exhibit normal, reduced or increased IL-17 levels, and how this translates to host physiology. The impact of this predominantly colonic cell subset on host metabolism also remains unknown. Still, mounting evidence points towards the importance of intestinal IL-17 for controlling metabolic homeostasis^14,15. IL-17⁺ _pT_regs may therefore be leveraged as a ‘dual hit’ strategy to curb immunometabolic dysfunction and gastrointestinal disturbances based on the immune-regulatory capacity of _pT_regs concomitant with the metabolic benefits of gut-delivered IL-17.

To this end, we hypothesized that environmental bacteria, who have not been under evolutionary scrutiny for host-microbe interactions, would provide an unexplored reservoir of immunomodulatory stimuli. In support of this hypothesis, the methanotropic noncommensal bacterium Methylococcus capsulatus Bath (McB) has previously been shown to interact with human dendritic cells (DC) modulating T cell responses in vitro¹⁶, and to reduce inflammation and disease activity in dextran sulfate sodium (DSS)-induced mouse colitis¹⁷. However, the impact on host metabolism and intestinal immune cells as well as mucus dynamics were not addressed.

We accordingly explored the effect of using whole-cell lysates from McB as protein source to reshape immunometabolism and the aberrant gut microbiota of DIO mice. We show that McB lysates augment Foxp3⁺RORγt⁺IL-17⁺ triple-positive _pT_regs in both SI- and LI-LP, and reset the obese microbiota concomitant with reversed key disease traits of diet-induced obesity.

Results

McB feeding reverses WD-induced gut microbiota changes and increases cecal SCFA levels

To induce obesity and immunometabolic dysfunctions, C57BL/6JRj mice were initially fed an obesogenic WD. After 12 weeks of WD feeding, the mice were stratified into new groups based on weight, fat mass and glucoregulatory capacity (Supplementary Fig. 1a), and fed experimental WDs for an additional 6 weeks. While dietary fat is known to elicit reproducible and lipid-dependent alterations in the murine gut microbiome across a variety of different diet compositions¹⁸, less is known about the microbiota-modulating impact of protein. Thus, to investigate if dietary protein (i.e., casein versus whole-cell bacterial lysates) would affect gut microbiota community structures, we analyzed freshly collected fecal samples before and throughout the dietary intervention. LFD and WD_REF fed mice showed distinct gut microbiota profiles after 12 weeks of feeding (intervention baseline, week 12 + 0; Fig. 1A, B), including ~10-fold lower abundance of the health-promoting genera, Parasutterella¹⁹ and Parabacteroides²⁰, countered by an equally increased abundance of the obesity-associated genus Desulfivobrio^21,22 (Fig. 1C) as well as a ~4-fold increase in the Firmicutes to Bacteroidetes (F/B) ratio (Fig. 1D). Interestingly, WD_CNTL fed mice showed negligible changes in the microbiome signature during the 6 weeks of intervention (Fig. 1A–E; Supplementary Fig. 2a, b), suggesting that the added lipid source had limited influence on the intestinal ecology. In contrast to this observation, we noted a pronounced shift in bacterial composition in mice fed WD_McB. Within the first 2 weeks of treatment, the general community structure in these mice shifted towards that of their LFD-fed counterparts (Fig. 1A–E, Supplementary Table 3). We next asked if the observed taxonomical differences between groups related to alterations in the functional potential. SCFAs are main end products of metabolized fibers, and to a lesser extent amino acids escaping digestion in the SI, with vast impact on host physiology^23,24. The highest levels of SCFAs are found in the cecum and proximal colon²⁵. We therefore investigated if cecal SCFA levels were different between groups. We found a consistent increase in the levels of the three major as well as three minor classes of SCFAs in the cecum of WD_McB fed mice compared to WD_CNTL fed counterparts pointing towards not just taxonomically, but also functionally, discrete microbiota profiles in the two groups of mice, supporting a beneficial health impact of dietary inclusion of McB lysates (Fig. 1F, G).

WD_McB feeding stimulates induction of gut-specific regulatory T cells

The intricate relationship between gut microbes and host immunity, combined with the immunoregulatory capacity of SCFAs²⁶, prompted us to investigate if the observed changes mediated by WD_McB feeding were associated with immune alterations.

We accordingly analyzed the immune cell profile of SI-LP and LI-LP in a subset of experimental mice (n = 6–10/group) using multicolor flow cytometry focusing on phenotypic characterization of group 3 innate lymphoid cells (ILC3), natural killer (NK) cells and T cells (consult Supplementary Fig. 3a, b for gating strategies). Numbers of ILC3s, NK cells and T cell receptor (TCR)-γδ⁺ T cells, were similar between groups (Supplementary Fig. 3d–f). The same was true for the numbers of TCRβ⁺ CD4⁺ T cells, as well as the proportion of T helper (T_H)1-, T_H17-, and _nT_regs cells (Fig. 2A, B; H, I). Interestingly, the proportion of _pT_regs was more than 2- and 3-fold increased in LI- and SI-LP, respectively of WD_McB fed mice compared to WD_CNTL fed counterparts (Fig. 2C, Jp < 0.001; S3G). Notably, this regulatory T cell subset has been shown to curb intestinal inflammation¹² and mediate immunological tolerance to the gut pathobiont Helicobactor hepaticus, thereby protecting against T_H17-mediated barrier dysfunction and subsequent colitis²⁷. Because RORγt is the hallmark transcription factor for T_H17 cell differentiation and essential for their IL-17 production, we next assessed if the _pT_regs induced by the different diets were also capable of expressing IL-17. Indeed, ex vivo stimulated _pT_regs produced substantial and diet-dependent amounts of IL-17 protein, where WD_McB feeding increased the proportion of IL-17⁺ cells within LI-LP _pT_regs (Fig. 2D–G, p < 0.001; 2k-N; S3H). We confirmed gut-specificity of the WD_McB-induced _pT_regs as only negligible amounts were observed in single-cell suspensions of liver homogenates obtained from ‘weight-matched’ mice fed the respective diets prophylactically for 7 weeks (Supplementary Figs. 1c, 4a). _pT_regs constituted <1% of all CD4⁺ T cells in the liver, hence contrasting the ~3 and 20% in SI- and LI-LP, respectively, of LFD- and WD_REF-fed mice and stunning ~12 and 30% in similar sites of WD_McB-fed mice (Supplementary Fig. 4b). Mechanistically, LI-, but not SI-, LP-derived McB-induced _pT_regs exhibited enhanced secretory capacity of the hallmark suppressive cytokine, IL-10, upon ex vivo stimulation (Fig. 2O, consult Supplementary Fig. 3c for gating strategy). Augmented IL-10 secretion was identified in both _nT_regs and _pT_regs populations in LI-LP of WD_McB-fed mice (Fig. 2O, P, p < 0.05 and <0.001, respectively), pointing towards enhanced immune regulation following WD_McB feeding. Although the absolute number of IL-10⁺_n/pT_regs was similarly increased in SI-LP of WD_McB-fed mice (Supplementary Fig. 4c, d), the relative proportion of IL-10⁺ cells within these T_regs remained similar between groups (Fig. 2O, P). Collectively, these data corroborate that the phenotypic shift of enhanced secretory capacity was restricted to the colon of WD_McB-fed mice. The amount of Ki67⁺ cells followed the patterns of _pT_reg abundances (Fig. 2Q, R; Supplementary Fig. 4e, f, consult Supplementary Fig. 3c for gating strategy).

**Fig. 2: WD_McB feeding stimulates induction of gut-specific regulatory T cells.**

WD_McB mitigates diet-induced obesity

The altered immune profile combined with a shift of the gut microbiota towards a state similar to that observed in lean LFD-fed mice, could potentially elicit crosstalk to glucoregulatory organs. To examine if McB lysates could reverse impaired glucose regulation, we performed OGTT and assessed GSIS concomitant with body mass composition in obese mice fed WD_REF for 11 weeks and after 5 weeks of dietary intervention allowing for temporal analyses (Supplementary Fig. 1a). All mice were stratified into experimental groups based on their pre-intervention glucoregulatory capacity (Supplementary Fig. 4g, h). While the response to glucose challenge remained largely unaffected from week 11 to week 12 + 5, regardless of experimental diets (Fig. 3A–C), both 5 h fasted insulin levels and glucose-stimulated insulin responses were significantly increased in mice fed WD_CNTL (Fig. 3E, p < 0.01 & p < 0.001, respectively) in accordance with our previous report on time-dependent alterations in glucose regulation²⁸. LFD- and WD_McB-fed mice were fully protected against this detrimental trajectory (Fig. 3D–F; Supplementary Fig. 4i, p = 0.24 and 0.68, respectively), and WD_McB-fed mice further exhibited modestly improved insulin sensitivity based on 5 h fasted glycemia (Fig. 3C, p < 0.05) and intraperitoneal insulin tolerance test (Fig. 3G, p < 0.05).

**Fig. 3: Dietary intervention with McB blunts progression of insulin resistance and fat mass accumulation.**

Overall, weight development mimicked the glucoregulatory capacity. As such, WD_McB-fed mice exhibited stability of weight, fat mass and lean mass when changed to experimental diets, contrasting the continuous weight and fat mass development of WD_CNTL-fed mice (Fig. 3H, I and Supplementary Fig. 4j). The absence of weight gain was not explained by decreased feed intake, but rather appeared to be associated with enhanced fecal energy secretion (Fig. 3J, K).

Since obesity and impaired glucose regulation are tightly associated with NAFLD²⁹, we next subjected paraffin-embedded liver sections to histological evaluation. These analyses revealed both diminished steatosis and hepatocellular ballooning in WD_McB-fed mice compared to WD_CNTL-fed counterparts, where especially hepatocellular ballooning was arrested in (or returned to) a state reminiscent that of lean LFD-fed mice (Fig. 3L, M, p < 0.05). Importantly, hepatocellular ballooning is instrumental in the development of the more severe liver disease, NASH²⁹.

WD_McB feeding resets the hepatic lipidome and decreases hepatic immune infiltration alleviating NAFLD

Based on the decreased NAFLD in WD_McB-fed mice housed at T_22°C we designed a new experiment (study outline, Supplementary Fig. 1b) using a recently described³⁰ method where thermoneutral housing (T_30°C) potentiates NAFLD in WT C57BL/6 J mice fed an obesogenic diet for 20–24 weeks. To more thoroughly investigate the effect of WD_McB-feeding, we also redesigned the diets and omitted macadamia oil in the WD_CNTL group, as this might lead to progression of obesity (Fig. 3) and related disorders. This new diet design entailed an increased fat/protein ratio in WD_McB compared to WD_REF, due to phospholipids inherently present in bacterial lysates³¹ (Supplementary Table 1). Despite the lower protein content in WD_McB compared to both other diets, the relative amounts of indispensable amino acids were similar between groups (Supplementary Fig. 1f) and protein availability was well beyond critical levels, corroborated by similar lean mass to WD_REF-fed mice post diet intervention (Supplementary Fig. 5a). Still, WD_McB-fed mice exhibited significantly improved 5 h fasting insulin levels and decreased fat mass (Fig. 4A, B), despite weight maintenance and significantly increased energy intake compared to both LFD and WD_REF-fed mice (Supplementary Fig. 5b–d). The decreased body fat mass was accompanied by a diminished NAS, supported by both pathological evaluation of H&E stained liver sections (Fig. 4C, D) and hepatocytic lipid content assessed by Oil-Red-O staining (Fig. 4E–G). We additionally observed augmented adiponectin secretion (Fig. 4H), pointing towards improved insulin sensitivity in the WD_McB group, further supported by the assessment of insulin tolerance and hepatic gene transcription activity of key metabolic enzymes (Fig. 4I, J). Of interest, we observed a >10-fold downregulation of Scd1 in the liver of WD_McB-fed mice, the hepatic expression of which is (a) regulated by the microbiota³² and (b) instrumental in de novo lipogenesis at the onset of metabolic syndrome³³.

**Fig. 4: Dietary McB intervention improves diet-induced metabolic and hepatic phenotype after prolonged WD feeding.**

We next assessed the hepatic lipidome by tandem mass spectrometry to elucidate if diminished NAFLD was associated with an altered lipid profile. Through comparison of WD_McB and WD_REF we identified 57 and 279 differentially regulated peaks in negative and positive ionization mode, respectively (Fig. 5A, B; Supplementary Fig. 5e, f, all FDR < 0.05). Of these, most classified lipids were changed with WD_McB in the direction of LFD-fed mice (Fig. 5C, D). Notably, 57% of upregulated species were odd-chain fatty acids, whereas 80% of downregulated species represented lipids with even carbon numbers (Fig. 5C, D and Supplementary Table 4), hence supporting previous reports where odd- rather than even-chain fatty acids are inversely associated with human insulin resistance⁹ and type 2 diabetes³⁴.

**Fig. 5: WD_McB feeding resets the hepatic lipidome and decreases hepatic immune infiltration alleviating NAFLD.**

To estimate the functional consequences of an altered lipid profile, we used the Lipidmaps database to identify affected pathways and plotted the observed changes on a log₂ scale comparing both LFD and WD_McB to WD_REF. The majority of affected pathways was similarly regulated in both direction and magnitude in LFD and WD_McB mice compared to WD_REF mice (Fig. 5E; Supplementary Fig. 5g). Notably, bile acids and ceramides, both of which were significantly downregulated in WD_McB-fed mice compared to WD_REF-fed counterparts, have been shown to mediate steatohepatitis by upregulation of IL-6 and TNF-α, respectively^35,36. We therefore measured these hepatic cytokines and observed similarly reduced levels in both LFD and WD_McB compared to WD_REF (Fig. 5F).

A key feature of diet-induced liver pathologies, including NAFLD, is recruitment of newly activated immune cells capable of eliciting a proinflammatory immune response. This process is generally hampered in mice housed at mild thermal stress, which therefore fail to phenocopy human pathophysiology. However, thermoneutral housing recapitulates some human disease traits³⁷, which combined with HFD-feeding accentuates intrahepatic infiltration of proinflammatory Ly6^high monocytes³⁸. These monocytes interact with tissue resident T cells and play a central role in the pathogenesis of liver injury, hence representing an attractive therapeutic target to mitigate NAFLD development and to curtail associated pathologies³⁶.

We therefore subjected liver tissues from representative mice to immunological evaluation by immunohistochemistry and observed a marked decrease in both CD3⁺ T cells and Ly6G⁺ neutrophils in WD_McB-fed mice compared to their WD_REF-fed counterparts (Fig. 5G). Diminished hepatic immune infiltration was mirrored by increased levels of circulating IL-22, IL-18, and IL-17 in WD_McB-fed mice compared to their WD_REF-fed counterparts (Fig. 5H, p < 0.01, <0.05, and <0.05, respectively).

We next initiated a short term experiment in ‘weight-matched’ mice to evaluate the hepatic immune profile before obesity onset (Supplementary Figs. 1C; 5h, i). Surprisingly, this experiment revealed increased numbers of tissue resident Tim4⁺ macrophages (i.e., Kupffer cells) in WD_McB-fed mice, suggesting that McB feeding either recruits or stimulate in situ proliferation of this key cell subset driving hepatic homeostasis (Fig. 5I). Kupffer cells are central to innate immunity and responsible for containment and clearance of foreign particles. Inflammatory activation of hepatic Kupffer cells potentiates obesity-associated insulin resistance, in part by recruiting neutrophils and T cells³⁹. Yet, Kupffer cells exhibit tremendous plasticity in their activation program, with anti-inflammatory properties in their alternative activation state ameliorating hepatic steatosis⁴⁰. While our staining panel did not allow us to identify the activation state of the enhanced Kupffer cell proportions, it is pertinent to note that none of the classically recruited cell types were altered in numbers (Supplementary Fig. 5j, k). Instead, we observed increased proportion of Ly6C⁺ monocytes. The mean fluorescence intensity (MFI) within these monocytes was marginally lower in WD_McB-fed mice than in their WD_REF-fed counterparts (Fig. 5J). Newly recruited monocytes express high levels of Ly6C in their inflammatory state; an expression that is gradually downregulated in immune resolving alternatively activated cells^36,41,42. Despite weight maintenance, WD_REF-fed mice exhibited ~50% increase in hepatic IL-17⁺ γδ T cells (Fig. 5K), an immunological precursor for subsequent NAFLD controlled by the gut microbiota⁴³. WD_McB-fed mice were fully protected from this trait, pointing towards extraintestinal regulation of innate immunity key to metabolic homeostasis.

WD_McB feeding reverses prolonged gut microbial dysbiosis and markedly improves colonic mucus production

The improved hepatic phenotype prompted us to further investigate potential traits in the gut-liver axis. We initially assessed if the observed cytokine responses associated with improved gut health in mice housed at T_30°C, where inflammation is expected to be increased³⁷. Indeed, WD_McB-fed mice were resistant to WD-induced colonic shortening closely associated with colonic inflammation (Fig. 6A).

**Fig. 6: WD_McB treatment improves colonic mucus production and reverses obesity-induced gut microbiota changes to resemble the composition found in lean LFD-fed mice.**

To gain additional insights to the immunomodulating properties of McB lysates, we focused on mucus production and -function. Because mucin production is a constitutive process where both secretion and adherence are constantly ongoing and rapidly adjust to environmental changes, immunohistochemical labelling for various MUC epitopes may not fully recapitulate the physical properties of the mucus. We therefore applied specialized mucin histochemistry staining allowing us to differentiate between neutral and acidic mucins according to the net charge of each molecule. Acidic mucins were further separated into sulfomucins and sialomucins. Sections revealed that neutral mucins in crypt-residing goblet cells were consistently downregulated in WD_REF-fed mice compared to LFD-fed counterparts in three well defined segments of the colon; i.e. proximal, middle and distal area (Fig. 6B, D). WD_McB-feeding not only reversed this pattern in all three segments but even also enhanced the production of neutral mucins exceeding the levels found in LFD- fed counterparts. This is a remarkable finding considering the continuous intake of a westernized diet high in fat and sucrose, known to hamper goblet cell function. We next evaluated crypt depth (CD) in stained sections. While the CD of the proximal and distal part of the colon was largely unaffected by diet, we observed increased CD in the middle segment of WD_McB-fed mice (Fig. 6C). WD_McB-feeding further enhanced the glycosylation pattern, particularly in the middle segment of colon, where this group exhibited a 3-fold increase in sulfomucins balanced by a similar (~2-fold) decrease in sialomucins compared to WD_REF-fed mice (Fig. 6E–G). No differences were observed between WD_REF and LFD-fed mice, indicating that the reciprocal regulation of mucin glycosylation status was specific to WD_McB-feeding.

With focus on the dynamic interactions between gut immunity, mucin glycosylation and commensal microbes⁴⁴, we next assessed the gut microbiota composition in temporally separated samples. This was done to determine if the changes towards a gut microbiota resembling that of lean LFD-fed mice was recapitulated in this intensified setup. In contrast to the first set of experiments, where we used cohoused mice shown to exhibit resilient microbiota profiles, we now employed single-housed mice to explore if the WD_McB-mediated community structures were persistent enough to induce consistent changes in the more dynamic communities of single-housed mice. Similar to our first experiments at T_22°C, we observed a normalization of the gut microbiota of WD_McB-fed mice, despite prolonged WD feeding prior to intervention (Fig. 6H–K). WD_McB-induced changes were surprisingly consistent with the first set of experiments, including a significantly lower F/B ratio (Fig. 6J), and a substantial reduction of Desulfovibrio abundance, countered by a similar bloom of the Parasutterella and Parabacteroides genera (Fig. 6K, Supplementary Table 5). Of note, the age-related increases in Desulfovibrio abundances recently reported²¹ was confirmed in this study where the general magnitude in both WD_REF- and LFD-fed mice increased 2-3-fold over the 5 week intervention (Fig. 6K). WD_McB-feeding fully prevented this trajectory and paired analyses even revealed a diminished relative abundance of Desulfovibrio in these mice.

McB lysates rely on adaptive immunity to favor Parabacteroides blooms

To assess if WD_McB-induced microbiota alterations were a result of diminished obesity or altered immunity we next evaluated the impact of WD_McB-feeding in ‘weight-matched’ C57BL6/N mice fed the respective diets for 7 weeks (Supplementary Fig. 1c). In agreement with previous reports, WD_REF feeding induced a ~3-fold decrease in Parabacteroides abundance independent of obesity, a feature that has been reported in several immune-competent mouse strains⁴⁵. WD_McB feeding not only prevented this trajectory but selectively enhanced the relative abundance of this genus > 10-fold (Fig. 7A, C; Supplementary Fig. 6a). The WD_McB-induced change in Parabacteroides abundance shifted the microbiota consortium away from their WD_REF-/LFD-fed counterparts (Fig. 7D, G). Considering the marked _pT_reg phenotype, we next evaluated microbial structures in RAG2^−/− mice deficient in adaptive immunity. Here, mice on LFD (Week 0) exhibited diminished Parabacteroides abundance as compared to their WT counterparts (Fig. 7A–C, E). Strikingly, WD_McB feeding was incapable of boosting the relative abundance in RAG2 deficient mice, indicating that the selective increase reported above was mediated by WD_McB through adaptive immunity, hence potentially by _pT_reg induction (Fig. 7B, E). The reciprocal actions on Desulfovibrio abundance was on the contrary independent of adaptive immunity (Fig. 7A, B; Supplementary Fig. 6a, b).

**Fig. 7: McB lysates rely on adaptive immunity to boost *Parabacteroides*.**

Despite some synergies between McB and T/B cells in affecting specific gut microbes (i.e., Parabacteroides), the global changes to microbial community structures induced by WD_REF-feeding in both WT and RAG2^-/- mice was generally mitigated by WD_McB-feeding. These traits include a diminished F/B ratio in WDMcB-fed RAG2^-/- mice (Supplementary Fig. 6c). Collectively, these findings suggest that the vast majority of WD_REF-induced changes and the McB-mediated protection against these, were independent of adaptive immunity (Fig. 7A–H; Supplementary Fig. 6a, b).

Altered gut microbiota by WD feeding affects glucose regulation

Despite a general protection against WD-induced global changes in the gut microbiota, the specific and substantial changes in key driver species (e.g., Parabacteroides) by WD_McB feeding solely in WT mice, could suggest that some of the metabolic effects reported above was mediated by the selective increase in _pT_regs. We thus evaluated glucoregulatory capacity in RAG2^−/− mice fed either diet for 6 weeks and observed a partial protection against WD-induced 5 h fasted hyperinsulinemia, insulin secretion during OGTT, body weight gain, and total fat mass (Fig. 8A–G). Both cecum weight and SCFAs were further enhanced to levels resembling those in the above-reported WT mice (Fig. 8H, I). Collectively, these data suggests that at least some of the metabolic effects observed in WD_McB-fed mice may occur independent of adaptive immunity.

We therefore designed CMT experiments in ABX treated mice and assessed the glucoregulatory capacity in cohorts fed either LFD or WD_REF (Supplementary Fig. 1e). While we failed to observe a donor-dependent effect of CMT in LFD-fed mice (Fig. 8J, K; Supplementary Fig. 6d–g), we did observe transient tendencies of metabolic improvements, notably on fat mass and GSIS, in WD-fed mice receiving cecal microbes from LFD- or WD_McB-fed donors (Fig. 8K–O).

Discussion

In this report, we explored the relationship between dietary nutrients and host-microbe interactions with a focus on immunometabolic response rates in the context of high fat, high sucrose, WD feeding. We found whole-cell lysates from the noncommensal methanotrophic bacterium, McB, capable of reversing hallmark signatures of WD feeding despite continued intake of an obesogenic diet. The signatures included diminished fat mass, improved intestinal immunity and glucoregulatory capacity accompanied by gut microbial community structures resembling that of lean LFD-fed counterparts.

The aberrant microbiota composition associated with obesity was recently shown to reflect HFD intake rather than obesity per se⁴⁵. It is therefore worth noting that we were able to normalize the dysregulated gut microbiota in mice remaining on a westernized, high fat, high sucrose diet; especially considering that it remains a clinical challenge to change dietary habits in individuals with lifestyle-related obesity. This remarkable change was reproduced in four substrains of mice originating from two different vendors, corroborating a robust phenotype not notably affected by baseline microbiota composition. To this end, a previous report on the resilience of the microbiota, argued for prolonged normalization⁴⁶. This study found that mice transferred to a low fat, fiber-rich chow diet after 12 weeks of HFD feeding, shifted their microbiota towards age-matched chow-fed control mice within 4 weeks, but only fully converged 10 weeks post diet change. This is particularly interesting as dietary fibers are known to be the most potent dietary regulator of the gut microbiota⁴⁷, by far exceeding that of dietary fat⁴⁸. Still, in our hands, WD_McB-feeding was able to reverse the obese microbiota at a higher pace than chow diet was in the previous report⁴⁶, despite similar fiber content in the two WDs. We further showed that the reversal of the microbiota traits was reproducible at different temperatures, altered reference diets, changed experimental duration, and in both cohoused and single-housed mice; all of which are prominent modulators of gut microbiota community structures. While many genera were similarly affected between experiments, others were either exclusively regulated at T_30°C (e.g., Bifidobacterium) or less pronounced affected at T_30°C (e.g., Barnesiella), suggesting dispensability for these specific microbes in the metabolic disease traits observed in this model. Similar observations were made for Akkermansia, which was ~3-fold upregulated at T_30°C, but discordantly affected (log2 FC bouncing from +9.3 to −10.4) between the two replication studies at T_22°C with pronounced intragroup variation.

Contrasting these discrepancies, there was a consistent downregulation of Desulfovibrio accompanied by >10-fold upregulation of Parabacteroides and Parasutterella in immunocompetent mice when changed from either WD_CNTL or WD_REF to WD_McB feeding. Our prophylactic experiment further showcased how Parabacteroides remained stable in LFD-, decreased in WD_REF- and dramatically increased in WD_McB-fed mice (~30-fold higher relative abundance than WD_REF-fed counterparts). Parabacteroides blooms were exclusively regulated in immunocompetent mice, suggesting that WD_McB feeding promotes selective microbial traits by McB-T/B cell interactions, and this likely through the pronounced induction of IL-10⁺/IL-17⁺ gut-specific _pT_regs, being the only investigated cell subset that was notably changed upon WD_McB feeding. It is worth noting, that not only was the prevalence of Parabacteroides not different between all groups and time points in RAG2^-/- fed mice, the relative abundance of this genus in these mice was also lowered to a level resembling that of WD_REF-fed WT mice, hence pointing towards a strong involvement of adaptive immunity to support a Parabacteroides favorable ecological niche. Other commensals, such as Barnesiella, Allobaculum, Clostridium IV, and Desulfovibrio remained as Parabacteroides stable over time in LFD-fed mice, but were substantially and inversely regulated between WD_REF- and WD_McB-fed mice (Supplementary Fig. 6a, b). The feed and time-dependent trajectories of these specific bacteria were independent of adaptive immunity and thus similarly regulated in both WT and RAG2^−/− mice. Interestingly, Desulfovibrio was most recently shown to flourish in aged, immunocompromised, obese mice with impaired glucose regulation²¹. Both T cell loss and Desulfovibrio administration per se, precipitated obesity-induced insulin resistance²¹. Still, in our hands, Desulfovibrio abundances neither were notably affected by the absence of T cells in RAG2^−/− mice, nor were the suppressive capacity of our McB lysate. This bacterium has moreover been shown to thrive in desulfonated colonic mucosa, associated with human⁴⁹ and mouse⁵⁰ colitis. Combined, these observations lend credence to the hypothesis of a mechanistic link between WD_McB-mediated reductions of Desulfovibrio and the observed improvements in mucin chemotype and metabolic response rates. In support of this notion, we observed a partial protection against WD-induced metabolic impairments in WD_McB-fed RAG2^−/− mice concomitant with Desulfovibrio suppression. Further pointing toward a role for the McB-induced microbiota changes to modulate host metabolism, WD_REF-fed, ABX treated recipient mice in CMT studies temporally exhibited a phenotype visually resembling that of their donors. Despite following a protective trajectory, the phenotype remained insignificant, likely due to a combination of the inherent variation in such experiments, a rather low n-number based on sample availability, and only a partial involvement of the gut microbial community structure in the precipitated phenotype.

Future studies are warranted to elucidate host-microbe mutualism upon WD_McB feeding. On this note, mucin-producing goblet cells are capable of delivering luminal antigens to LP-residing DC⁵¹ instrumental for T cell polarization⁵². As McB potently induce DC maturation markers in vitro—even exceeding the effects of the well-described probiotic strain, Escherichia coli Nissle 1917⁵³—affecting subsequent T cell reponses¹⁶, we predict a direct link between McB intake and the corresponding immune profile, which might subsequently precision edit certain microbes, e.g., Parabacteroides.

In addition to the direct link proposed here, McB intake might also indirectly promote T_reg polarization through augmented SCFAs⁵⁴, a trait persisting in RAG2^−/− mice. These important metabolites also sustain mucus production and facilitate tissue crosstalk in the gut-liver axis². In keeping with this notion, we observed decreased hepatic bile acids and TNF-α levels combined with a pronounced reduction of intrahepatic CD3⁺ and Ly6G⁺ cells in obese WD_McB-fed mice. Importantly, while Ly6C^high cells represent proinflammatory, fibrogenic macrophages⁵⁵, their in situ differentiation to Ly6C^low cells⁴¹ facilitate tissue repair and improved NASH prognosis⁴². It is therefore encouraging to note, that the MFI of Ly6C was nominally decreased in ‘weight-matched’ mice fed WD_McB short term. These data indicate that, despite being increased in absolute numbers, Ly6C⁺ cells in WD_McB-fed mice were in the process of differentiating from Ly6C^high to Ly6C^low cells with immune resolving capabilities, although further studies are warranted to fully describe the kinetics in such responses. In agreement with an immune resolving phenotype, WD_McB-fed mice were fully protected from the ~50% increase in hepatic IL-17⁺ γδ T cells observed in WD_REF-fed counterparts before the onset of obesity. This cell subset has been shown to (a) be regulated by the gut microbiota and (b) both precede and causally induce NAFLD⁴³. The above affected targets are all proposed as relevant strategies to curb NAFLD and the more terminal liver disease, NASH³⁶.

The substantial, consistent, and global changes in the immunometabolic profile found in this study point towards clinical potential if further developed. Notably, only a single study has previously succeeded in inducing _pT_regs, and this was limited to LI-LP¹¹. Apart from inducing this cell subset throughout the gastrointestinal tract, we further show that these cells exhibited enhanced secretory capacity of both the hallmark immunosuppressive cytokine, IL-10, as well as IL-17 in LI-LP of WD_McB-fed mice compared to their WD_CNTL fed counterparts. To this end, emerging evidence indicates that the non-regulatory counterparts to _pT_regs, i.e. T_H17 cells, are purged from SI-LP in DIO mice¹⁴ and that reintroducing ex vivo differentiated gut tropic T_H17 cells curtail obesity development, hence improving insulin sensitivity¹⁵. Notably, despite intergroup similarities in the relative proportion of SI-LP IL-10⁺/IL-17⁺_pT_regs, the absolute numbers of these cells were markedly increased in WD_McB-fed mice, thus advocating enhanced immune regulation.

That IL-17 in this intervention is produced by T_regs and not T_H17 cells might further improve the safety profile of a putative medication, as the physiological impact and properties of T_H17 cells are context-dependent. As such, T_H17 cells may potentiate inflammatory bowel disease (IBD) in individuals with compromised barrier function⁵⁶, whereas _pT_regs have shown enhanced suppressive capacity against the same disease¹². This is a pertinent feature given the relatively high proportion of subjects with metabolic complications co-suffering from IBD⁵⁷. While our study was not designed to assess IBD susceptibility, it is well described that purified diets compromise barrier function^47,58,59 and thus represent clinical IBD-features (disease-cause) preceding inflammation (symptom)⁶⁰. In addition to the immunomodulating phenotype, WD_McB-feeding augmented neutral mucus production in all segments of the colon while enhancing IBD-protective sulfomucins specifically in the middle segment. In this segment, we also detected increased CD in WD_McB-fed mice, collectively pointing towards increased barrier function.

_pT_regs were most recently described to be transferred from mother to offspring via nongenetic inheritance through breastmilk IgA; a trait transmitted during a tight age window after birth but stable for life, and resistant to many microbial or cellular perturbations⁶¹. It is thus striking to note the pronounced and reproducible induction we here report in both SI- and LI-LP in three substrains of mice from different vendors. Unfortunately, we have not investigated the B cell component of the immune system in this study, and thus, we cannot conclude on the extent to which _pT_reg induction relies on B cells, McB lysate per se or the McB-induced SCFAs and/or gut microbial changes. It is, however, worth noting that apart from IgA-induced _pT_reg-control in early life, other reports have also convincingly demonstrated that this cell subset can be induced by bacterial metabolism of bile acids^62,63 as well as microbe-derived SCFAs in general⁵⁴ and butyrate in particular⁶⁴. The latter of which we found abundantly increased upon WD_McB feeding.

In summary, this study demonstrates a consistent activation of IL-10⁺/IL-17⁺ _pT_regs throughout the gastrointestinal tract and a reversion of gut dysbiosis in response to WD_McB feeding, even in the context of high dietary fat and sucrose intake. The prospects of using bacterial lysates as an alternative to traditional live probiotics merit further investigation, just as future studies are warranted to assess to what extent our results can be translated to humans. Future studies are also urgently needed to identify the biological effector molecule(s) in this bacterial lysate, with a potential for being developed into a medical product(s) rebalancing intestinal immunity while targeting gut-related dysbiosis and metabolic abnormalities.

Methods

Mice and ethical statements

All experiments were conducted in accordance with the EU directive 2010/63/EU as approved by the Danish Animal Experiments Inspectorate (#2014-15-2934-01,027). Six- to seven-week-old male C57BL/6JBomTac, C57BL/6JRj, C57BL/6 N, C57BL/6N-Rag2Tm1/CipheRj (RAG2^−/−) mice were acquired from Taconic Laboratories, Denmark, or Janvier Labs, France, respectively, as detailed below. All mice were allowed to acclimatize on regular chow diet for the first week upon arrival and then fed a low fat diet (LFD) for 2 weeks prior to study initiation. Mice were kept under specific pathogen free conditions at 22 °C (T_22°C) or 30 °C (T_30°C), as indicated, in 12 h light/dark cycle (6AM–6PM).

Housing, diets, and experimental setup

Standard pelleted diets as well as protein-free western diet (WD) powder were obtained from Ssniff Spezialdiäten GmBH, Germany, and stored at −20 °C throughout the duration of the experiment. Mice were fed a compositionally defined low fat diet (LFD, S8672-E050) until study initiation. A subgroup remained on LFD, whereas the remaining mice were transferred to a soy oil based reference WD (S8672-E025) (high fat, high sucrose, containing 0.15% cholesterol) for the run-in period (12 or 21 weeks dependent on the experiment). Experimental WDs were lot-matched and produced without protein (S9552-E021), which was subsequently added, and then pelleted, by the investigators as indicated; WD_REF = containing 19.5% (w/w) casein, WD_CNTL = containing 16.5% (w/w) casein and 3% Macadamia oil, WD_McB = 19.5% (w/w) whole-cell bacterial lysates (predominantly protein but up to ~15% phospholipids³¹). Diet composition is more extensively described in Supplementary Table 1 with amino acid composition presented in Supplemetary Fig. 1f. Detailed description of bacterial lysates is provided in the designated section below. All diets were freshly made for each experiment with independent batches of bacterial lysates to substantiate robustness and reproducibility of the observed findings. Mice were fed ad libitum, and feed intake was measured thrice a week. Water was changed weekly.

12 + 6 protocol T_22°C

Experimental setup is outlined in Supplementary Fig. 1a. C57BL/6JRj mice were purchased from Janvier Labs, France, and housed three mice per cage, reproduced in two independent experiments. For transparency, all data are depicted in respective graphs with mean ± SEM plus individual dots for the combined results of the two independent experiments. Experiment-dependent shapes enable visual discrimination of respective experiments as well as data distribution within and between experiments.

21 + 5 protocol T_30°C

Experimental setup is outlined in Supplementary Fig. 1b. C57BL/6JBomTac mice were purchased from Taconic Laboratories, Denmark, and initially housed 10 per cage to accelerate weight gain. Twelve weeks into the run-in period, mice were single caged to allow accurate assessment of food intake, and also monitor if the more sensitive microbiota of single-housed mice would uniformly change towards a composition typical for LFD-fed mice, or if such phenomenon relied on a few highly responding mice transferring their microbiota to cage mates under group housed conditions. The timing of single housing allowed affected mice to adapt to social isolation and stabilize their weight development before intervention (Supplementary Fig. 5b).

Prophylactic protocols

Experimental setup is outlined in Supplementary Fig. 1c, d. C57BL/6N-Rag2Tm1/CipheRj (RAG2^−/− mice) and matched C57BL6/N WT mice were purchased from Janvier Labs, France, and housed three mice per cage at thermoneutrality, acclimated on LFD and from 9 weeks of age fed the experimental diets for 6–7 weeks depending on genotype (see Supplementary Fig. 1c, d for details). This was reproduced in two independent experiments analogous to the 12 + 6 protocol described above.

Cecal microbiota transfer (CMT) protocol

Experimental setup is outlined in Supplementary Fig. 1e. C57BL/6JBomTac mice were purchased from Taconic Laboratories, Denmark, and housed three mice per cage at thermoneutrality, reproduced in two independent experiments. Experiment-dependent shapes enable visual discrimination of respective experiments as well as data distribution within and between experiments. We used mice from the 21 + 5 protocol as donors of their cecal microbiota, ensuring transfer between genetically identical mice hence limiting obvious confounders. Cecal microbiota transfer (CMT) was performed as after one week of antibiotic (ABX) treatment. Specifically, mice were housed in disposable IVC cages with free access to sterilized water and LFD during the 2 weeks of acclimatization. They were then treated with a two-leg ABX mixture (1 and 0.5 g ampicillin and neomycin, respectively, per liter water, changed thrice weekly) for seven days while remaining on LFD. After ABX treatment, experimental mice were fasted for 2 h and subsequently gavaged with Polyethylene Glycol (Gangiden, Sandoz) to clean the intestines from potential remnants. Six hours later, mice received the first of three boluses of cecal microbiota from either LFD-, WD_REF-, or WD_McB-fed donors. Cecal microbiota were pooled from all mice of the respective groups before transfer. Immediately after first CMT, mice were divided into either LFD- or WD_REF-fed recipients and fed their respective diets for additionally 5 weeks.

Cultivation of M. capsulatus Bath and preparation of bacterial lysate

Methylococcus capsulatus Bath (NCIMB 11132) was cultivated in nitrate mineral salts (NMS) medium to produce the single-strain bacterial lysate. NMS medium was composed from 5 stock solutions; 10× NMS Salts (98.9 mM KNO₃, 43.8 mM MgSO₄*6H₂O and 9.0 mM CaCl₂), 1000× NaMoO₄*2H₂0 (1.07 mM), 10,000× FeEDTA (103 mM), 1000× Trace Elements Solution (0.8 mM CuSO₄*5H₂0, 1.8 mM FeSO₄*7H₂0, 1.4 mM ZnSO₄*7H₂0, 0.24 mM H₃BO₃, 0.21 mM CoCl₂*6H₂0, 0.74 mM EDTA-Na₂, 0.1 mM MnCl₂*4H₂0, 42.07 uM NiCl₂*6H₂0) and 10× phosphate buffer pH 6.8 (198.25 mM Na₂HPO₄*12H₂0, 191.05 mM KH₂PO₄). When preparing 1× NMS medium from stock solutions 10x NMS Salts was add to 50% of the final volume with H₂O before the remaining stock solutions were added. Then pH was adjusted to 6.8 before the NMS medium was sterile filtered using a 0.45-uM filter. Medium was stored in the dark.

The 1000× Trace Element Solution was made from the following stock solutions and stored in the dark; 400.5 mM CuSO₄*5H₂0, 3.6 mM FeSO₄*7H₂0 (pH 3.0), 347.8 mM ZnSO₄*7H₂0, 161.7 mM H₃BO₃, 42.0 mM CoCl₂*6H₂0, 50.5 mM MnCl₂*4H₂0 and 42.0 mM NiCl₂*6H₂0. McB culture aliquots were frozen in liquid nitrogen and stored at −80 °C. Cultivations on agar plates and in shake flasks (orbital shaker incubator at 200 rpm) were performed at 45 °C in an atmosphere of 75% air, 23.25% CH₄, and 1.25% CO₂. Continuous cultivation was carried out in a 3-l bioreactor (Applikon, The Netherlands) with a working volume of 2 l. Cells were precultivated in shake flasks and used to inoculate the bioreactor to an optical density at 440 nm (OD₄₄₀) of 0.1. The temperature was maintained at 45 °C, stirring set to 650 rpm, and pH maintained at 6.8 by automatic addition of 2.5 M NaOH/2.5 M HCl. A gas mixture of 75% air and 25% methane was sparged into the bioreactor. The continuous culture was started after an initial batch phase, and the dilution rate was set to 0.01 h⁻¹. The OD₄₄₀ at steady state was generally sustained at ~10. Culture effluent was collected, and cells were harvested by centrifugation. Bacterial cell walls were disrupted by the use of a French press before freeze-drying of the material.

Glucose and insulin tolerance tests

Mice were subjected to magnetic resonance (MR)-scan using EchoMRI 4in1 (Texas, USA) to determine fat- and lean mass at indicated time points (Supplementary Fig. 1a–d). Mice were fasted 5 or 2 h prior to any oral glucose tolerance (OGTT) or intraperitoneal insulin tolerance test (ITT), respectively. Fasting blood glucose was measured by tail vein bleeding using the Bayer Contour glucometer (Bayer Health Care). Mice were subsequently gavaged with 3 µg glucose/g lean mass (OGTT) or intraperitoneally injected with 0.75 mU insulin/g lean mass (ITT). Blood were sampled at specified time points for blood glucose and insulin measurements to assess glucoregulatory capacity including glucose-stimulated insulin secretion (GSIS) as described in detail elsewhere²⁸.

Short chain fatty acid measurements

Short chain fatty acids (SCFAs) were determined by gas chromatographic analysis. Feces were suspended in MilliQ water (1:1 (w/v)) and subsequently homogenized in a FastPrep-96 (MP Biomedicals) sample preparation unit. The homogenates were diluted with 0.4% formic acid (1:1 (v/v)), transferred to Eppendorf tubes and centrifuged at 16,200 rcf for 10 minutes. 300 μl of the supernatants were applied to spin columns (VWR, 0.2 μm pore size) and centrifuged at 9560 rcf for 5 minutes. The eluates were transferred to 300-μl GC vials. Split injection mode was used with an injection volume of 0.2 μl. The gas chromatograph was a Trace 1310 (Thermo Scientific) equipped with an autosampler and a flame ionization detector. Helium was used as carrier gas, and the column was a 30 m long Stabilwax (Restek) with polyethylene glycol as stationary phase. Injector temperature: 250 °C, temperature intervals: 2 min at 90 °C before a 6 min increase to 150 °C, then a 2 min increase to 245 °C and a hold at this temperature for 4.9 min. Detector temperature: 275 °C.

RNA extraction and quantitative RT-PCR

Frozen liver tissue was cryo-grinded by mortar and pestle on liquid nitrogen and total RNA was extracted by TRIreagent (Sigma–Aldrich, USA) according to the manufacturer’s protocol using PRECELLYS® 24 for homogenization. One microgram of RNA was transcribed into cDNA by reverse transcriptase (Invitrogen, USA). Quantitative PCR analyses were performed using the SYBR Green qPCR Master mix (Thermo Scientific, USA) and the Stratagene Mx3000P qPCR System. Standard curve and no template control (NTC) were included on every 96-well plates. Validation of each target was based on Rsq of standard curve >0.995, amplification efficiency of 100 ± 10% and singular peak. Ct values for each target were related to 18 S reference Ct values of the same sample by 2^ΔCt and visualized as fold-change to the LFD reference group mean. Primer sequences are summarized in Supplementary Table 2.

Histology

Liver tissue was sampled promptly after euthanasia, fixed in 10% formalin and subsequently embedded in paraffin prior to sectioning according to standard procedures for light microscopy. Colon was emptied for content and subsequently rolled around a 27 G needle after which the generated Swizz Rolls were carefully preserved in liquid nitrogen (wrapped in aluminum foil) until paraffin embedding. Embedded tissues were cut 2-µm thick, and sections mounted onto glass slides. These sections were processed further and stained either with hematoxylin and eosin (H&E), Oil Red O, PAS, or HID-AB, or labelled with antibody of interest for immunohistochemical investigations, as detailed below. Samples were randomized and blinded to the pathologists performing the histological analyses.

Liver

Nonalcoholic fatty liver disease activity score (NAS). Liver sections were prepared by standard protocols and subsequently assessed by two independent observers in a blinded fashion, using the established NAFLD activity score (NAS) for evaluation of H&E stained liver sections⁶⁵. In short, score of 0–2 excludes nonalcoholic steatohepatitis (NASH), a score of 3–4 defines “borderline NASH”, and score ≥5 is considered as NASH.

Oil Red O staining. Liver biopsies were fixed in paraformaldehyde, cryoprotected in polyvinylpyrrolidone/saccharose, and frozen in liquid nitrogen. Cryosections were prepared and stained using the lipid-specific Oil red O (Sigma–Aldrich, Germany) and analyzed under an Axio Imager M2 microscope. Images were captured with the Axiocam 506 color camera (Zeiss, Jena, Germany). Areas of stained lipids were determined using ImageJ-1.50i software (https://imagej.nih.gov/ij/index.html).

Colon

The morphological and morphometrical evaluations of the colonic tissues were performed in a blinded fashion, and all measurements were made by the same trained pathologist. One H&E stained section of colon from each individual was evaluated histologically for any sign of pathology. To assess the quality of the mucins, colonic tissues were stained for acidic as well as neutral mucins. Neutral mucins were stained by Periodic Acid Schiffs (PAS). Acidic mucins were stained through a combination of High Iron Diamine (HID), for sulfated mucins (sulfomucins), and Alcian Blue (AB) for carboxylated mucins (sialomucins). Crypt depth (CD) measurements, as well as analyses of mucin type and amount, were performed on HID/AB- and PAS-stained sections in three histologically distinct areas of the colon, i.e., the proximal, middle and distal area. At each of these three locations, three longitudinally sectioned crypts were measured. Each data point is thus presented as an average of three (mucin type and amount) or six (CD) measures. Crypts were selected only when the entire crypt epithelium was visible from the lamina muscularis mucosae to the lumen. The histological sections of the colon were examined in an Axio Imager Z2 microscope (Zeiss, Jena, Germany), and digital images were obtained using an Axiocam 506 color camera (Zeiss, Jena, Germany). Micrographs were captured with the same ×20 objective magnification. Crypt depth measurements were performed using the software program Image J-Fiji version 1.52e Java 1.8.0_181⁶⁶. For quantitative evaluation and automated scoring of the different mucin types, a plugin for color deconvolution for the Image J-Fiji program was used⁶⁷.

Hepatic lipidomics

Samples were randomized and lipid extracted following standard protocols. In short, 50 mg of frozen tissue were homogenized in a prechilled 2 mL tube added 0.5 mL of ice-cold 50% MeOH with ~40 µM D5-tryptophan following standard protocols and stored at −80 °C until LCMS analysis. The peaklist obtained after preprocessing (features defined by their mass/charge value, retention time and peak area) was analyzed in MetaboAnalyst 4. Data were then auto-scaled (mean-centering and division by the square root of standard deviation of each variable) to enforce Gaussian distribution enabling relative comparison. Univariate and multivariate analysis were performed with t-tests, volcano plot, analyses of variance (ANOVA) followed by Tukey’s HSD test, and principal component analysis (PCA) to detect significant hits and to visually separate trends between groups. Features showing a similar pattern between LFD and WD_McB were identified by pattern matching function. The top 25 features obtained were then processed for identification using the online database Lipidmaps (http://lipidmaps.org) with a mass tolerance between the measured m/z value and the exact mass of 3 ppm.

Microbiome analysis and bioinformatic processing

Fresh feces samples were collected 3–4 h into the light cycle, snap frozen and stored at −80 °C until downstream processing. Bacterial DNA from fecal samples was extracted using a NucleoSpin soil kit (Macherey-Nagel) according to manufacturer’s instructions. 16 S rRNA gene amplification was performed using 515 F and 806 R primers and sequenced with 250PE on Ilumina MiSeq. USEARCH⁶⁸ and mothur⁶⁹ were used for the processing of the sequence data. Sequences were strictly dereplicated, discarding clusters smaller than 5. Sequences were clustered at 97% sequence similarity. Additional suspected chimeric operational taxonomic units (OTUs) were discarded on the basis of comparison with Ribosomal Database Project classifier training set v9¹⁷ using UCHIME⁷⁰. Taxonomic assignment of OTUs was done using the database from Ribosomal Database Project¹⁷. Initial data processing was performed as described elsewhere⁷¹. Data was rarefied to 23,107 reads per sample. The principal coordinate analysis (PCoA) was conducted based on the weighted UniFrac distance. To select OTUs and species enriched in different subgroups, DESeq2⁷² with default settings was used. The relative abundances of OTUs were aggregated to genus level. Genera present in >1/3 of the samples were used in the differential abundance testing. Genera with p.adj. < 0.05 comparing the last sampling point for WD_REF/WD_CNTL vs WD_McB with DESeq2 differential expression analysis based on the negative binomial distribution were reported. Permutational multivariate analysis of variance (permanova) using weighted UniFrac distance matrices (Adonis, Vegan R package) was used to evaluate overall microbial compositions.

Isolation of small intestine (SI) and large intestine (LI) lamina propria (LP) cells

Feces and mucus from LI were scraped off while SI was flushed with 1× HBSS (Gibco) containing 15 mM HEPES (Thermo Fisher Scientific). Peyer´s patches were carefully excised from SI. SI and LI were opened longitudinally, cut into approx. 1 cm pieces and washed three times with 1× HBSS (Gibco) containing 15 mM HEPES (Thermo Fisher Scientific), 5% FCS (Viralex^TM, PAA Laboratories), 50 µg/mL gentamycin (Gibco), and 2 mM EDTA (Invitrogen, Life Technologies) for 15 min at 37 °C. During the first incubation step, 0.15 mg/mL DL-dithiothreitol (Sigma–Aldrich) was added to LI samples. LI, but not SI samples, were shaken on an orbital shaker at 350 rpm during incubation. After each incubation step, SI, but not LI samples, were shaken by hand for 10 s. Media containing epithelial cells and debris were discarded by filtration through a 250 µm mesh (Tekniska Precisionsfilter, JR AB). The remaining tissue was digested for 20–28 min at 37 °C under magnetic stirring (350-500 rpm) in R-10 media (RPMI 1640 (Gibco) with 10% FCS (Viralex, PAA Laboratories), 10 mM HEPES (Thermo Fisher Scientific), 100 U/mL penicillin (Gibco), 100 µg/mL streptomycin (Gibco), 50 µg/mL gentamycin (Gibco), 50 µM 2-mercaptoethanol (Gibco), and 1 mM sodium pyruvate (Gibco)) containing 1 mg/ml collagenase P (Roche) and 0.03 mg/mL DNase I (Roche). After digestion, SI-LP and LI-LP cells were purified by density gradient centrifugation (600 rcf for 20 min at 22 °C, acceleration 5 and brake 0) with 40/70 % Percoll (GE Healthcare). Cell suspensions were subsequently filtered through 100 µm cell strainers (BD Biosciences) and restimulated in vitro prior to staining for flow cytometric analysis.

Isolation of liver cells

To isolate liver cells, livers were collected from 1× PBS (Gibco)-perfused mice, cut into small pieces and digested for 40 min at 37 °C in R-10 media (RPMI 1640 (Gibco) with 10% FCS (Viralex, PAA Laboratories), 10 mM HEPES (Thermo Fisher Scientific), 100 U/mL penicillin (Gibco), 100 µg/mL streptomycin (Gibco), 50 µg/mL gentamycin (Gibco), 50 µM 2-mercaptoethanol (Gibco), and 1 mM sodium pyruvate (Gibco)) containing 0.3 mg/ml collagenase IV (Roche), 0.21 mg/mL collagenase D (Roche) and 0.05 mg/mL DNase I (Roche) on an orbital shaker (370 rpm). After digestion, liver cells were filtered through 100 µm cell strainers (BD Biosciences) and subsequently purified by density gradient centrifugation (800 rcf for 20 min at 22 °C, acceleration 5 and brake 0) with 40/70% Percoll (GE Healthcare). Cell suspensions were either restimulated in vitro prior to or directly used for flow cytometric analysis.

Ex vivo stimulation of SI-LP, LI-LP, and liver cells

SI-LP, LI-LP, and liver cells were restimulated in vitro in R-10 media in the presence of either 20 ng/mL IL-23 (R&D Systems) or 250 ng/mL PMA (Sigma–Aldrich) in combination with 0.5 µg/mL ionomycin (Sigma–Aldrich) for 4 h at 37 °C. After 1 h incubation, 10 µg/mL brefeldin A (BioLegend) was added.

Flow cytometry

Flow cytometry was performed according to standard procedures. Cell aggregates (identified in FSC-H or FSC-W vs. FSC-A scatter plots) and dead cells identified by using Zombie Aqua or Zombie UV Fixable Viability Kit (BioLegend) were excluded from analyses. Intracellular staining was performed using the eBioscience FoxP3/Transcription Factor Staining Buffer Set (eBioscience) according to the manufacturer’s instructions. Data was acquired on a LSRII (BD Biosciences) and analyzed using FlowJo software (Tree Star).

Antibodies (Abs) and reagents

The following mAbs and reagents were used in the study: anti-CD3ε (17A2), anti-CD4 (GK1.5), anti-CD8α (53-6.7), anti- CD11b (M1/70), anti-CD11c (N418), anti-CD19 (6D5 and ID3), anti-CD45 (30F11), anti-CD45R/B220 (RA3-6B2), anti-CD90.2 (30-H12), anti-CD127 (A7R34), anti-Ly-6G/Ly6C (Gr-1) (RB6-8C5), anti-Ly6C (AL-21), anti-Ly-6G (1A8), anti-TER-119 (TER-119), anti-F4/80 (BM8), anti-I-A/I-E (M5/114.15.2), anti-Siglec-F (E50-2440), anti-TIM-4 (54(RMT4-54)), anti-TCRβ (H57-597), anti-TCRγδ (GL3), anti-NK1.1 (PK136), anti-NKp46 (29A1.4), anti-FoxP3 (FJK-16s), anti-RORγt (B2D), anti-T-bet (4B10), anti-Ki67 (B56), anti-IL-10 (JES5-16E3), anti-IL-17A (TC11-18H10.1), anti-IL-22 (IL22JOP), and anti-IFNγ (XMG1.2). All Abs were purchased from eBioscience, BioLegend or BD Biosciences. PECF594-conjugated and BV421-conjugated streptavidin were purchased from BD Biosciences and BioLegend, respectively.

Multiplex cytokine quantification

Liver tissues were crushed with a mortar and pestle on liquid nitrogen. 30–40 mg of tissue were transferred to a clean tube and 400-µL T-PER tissue protein extraction reagent (Thermo Scientific) buffer including proteinase inhibitors (Sigma–Aldrich) were added. Protein concentration was assessed by BCA (Thermo Scientific) according to manufacturer’s instruction. Cytokines were quantified using xMAP technology with combinations of Bio-Plex Pro (BioRad) premixed mouse cytokine panels, and the analyses were carried out on a Bio-Plex 200 system and the Bio-Plex Manager Software package.

Statistical analysis

Omics data were acquired as described in their appropriate sections. Normality residuals of remaining data were assessed by D’Augostino-Pearson omnibus (k2), Graphpad Prism 8, and subsequently evaluated by parametric (gaussian distributed) or non-parametric (non-gaussian distributed) tests as appropriate. Details of each test including post-hoc assessment are specified in figure legends. Data are expressed as indicated in individual figure legends, and all groups are compared by two-sided tests to the relevant WD_REF or WD_CNTL group as indicated.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

All data are in this manuscript are available on reasonable request to the corresponding authors. 16S sequencing data are uploaded to a public repository with study accession number: PRJEB41983 and available at the following link: http://www.ebi.ac.uk/ena/data/view/PRJEB41983. All non-omics datasets included in the current study are available in the figshare repository Jensen, B. A. H. et al. Data underlying the manuscript: Lysates of Methylococcus capsulatus Bath induce a lean-like microbiota, intestinal FoxP3+RORγt + IL-17+ Tregs and improve metabolism. figshare https://doi.org/10.6084/m9.figshare.13522283 (2021).

Code availability

Standard codes from the cited R-packages were used in this manuscript. Specific scripts are available upon reasonable request.

References

Honda, K. & Littman, D. R. The microbiota in adaptive immune homeostasis and disease. Nature 535, 75–84 (2016).
Article ADS CAS PubMed Google Scholar
Koh, A., De Vadder, F., Kovatcheva-Datchary, P. & Bäckhed, F. From dietary fiber to host physiology: short-chain fatty acids as key bacterial metabolites. Cell 165, 1332–1345 (2016).
Article CAS PubMed Google Scholar
Lynch, S. V. & Pedersen, O. The human intestinal microbiome in health and disease. N. Engl. J. Med. 375, 2369–2379 (2016).
Article CAS PubMed Google Scholar
Cani, P. D. Human gut microbiome: Hopes, threats and promises. Gut https://doi.org/10.1136/gutjnl-2018-316723 (2018).
Seregin, S. S. et al. NLRP6 protects Il10−/−mice from colitis by limiting colonization of Akkermansia muciniphila. Cell Rep. 19, 733–745 (2017).
Article CAS PubMed PubMed Central Google Scholar
Shono, Y. et al. Increased GVHD-related mortality with broad-spectrum antibiotic use after allogeneic hematopoietic stem cell transplantation in human patients and mice. Sci. Transl. Med. 8, 339ra71 (2016).
Article ADS PubMed PubMed Central CAS Google Scholar
Kovatcheva-Datchary, P. et al. Dietary fiber-induced improvement in glucose metabolism is associated with increased abundance of prevotella. Cell Metab. https://doi.org/10.1016/j.cmet.2015.10.001 (2015).
Rolhion, N. et al. Specific targeting of intestinal Prevotella copri by a Listeria monocytogenes bacteriocin. bioRxiv https://doi.org/10.1101/680801 (2019).
Pedersen, H. K. et al. Human gut microbes impact host serum metabolome and insulin sensitivity. Nature 535, 376–381 (2016).
Article ADS CAS PubMed Google Scholar
Plovier, H. et al. A purified membrane protein from Akkermansia muciniphila or the pasteurized bacterium improves metabolism in obese and diabetic mice. Nat. Med. 23, 107–113 (2017).
Article CAS PubMed Google Scholar
Verma, R. et al. Cell surface polysaccharides of Bifidobacterium bifidum induce the generation of Foxp3+ regulatory T cells. Sci. Immunol. 3, eaat6975 (2018).
Article PubMed Google Scholar
Yang, B.-H. et al. Foxp3+ T cells expressing RORγt represent a stable regulatory T-cell effector lineage with enhanced suppressive capacity during intestinal inflammation. Mucosal Immunol. 205, 1381–1393 (2015).
Google Scholar
Cavallari, J. F., Denou, E., Foley, K. P., Khan, W. I. & Schertzer, J. D. Different Th17 immunity in gut, liver, and adipose tissues during obesity: the role of diet, genetics, and microbes. Gut Microbes 7, 82–89 (2016).
Article CAS PubMed PubMed Central Google Scholar
Garidou, L. et al. The gut microbiota regulates intestinal CD4 T cells expressing RORγt and controls metabolic disease. Cell Metab. 22, 100–112 (2015).
Article CAS PubMed Google Scholar
Hong, C. et al. Gut-specific delivery of T-helper 17 cells reduces obesity and insulin resistance in mice. Gastroenterology 152, 1998–2010 (2017).
Article CAS PubMed Google Scholar
Indrelid, S., Kleiveland, C., Holst, R., Jacobsen, M. & Lea, T. The soil bacterium Methylococcus capsulatus bath interacts with human dendritic cells to modulate immune function. Front. Microbiol. 8, 1–13 (2017).
Article Google Scholar
Cole, J. R. et al. Ribosomal Database Project: data and tools for high throughput rRNA analysis. Nucleic Acids Res. 42, 633–642 (2014).
Article CAS Google Scholar
Bisanz, J. E., Upadhyay, V., Turnbaugh, J. A., Ly, K. & Turnbaugh, P. J. Meta-analysis reveals reproducible Gut microbiome alterations in response to a high-fat diet. Cell Host Microbe 26, 265–272 (2019).
Article CAS PubMed PubMed Central Google Scholar
Ju, T., Kong, J. Y., Stothard, P. & Willing, B. P. Defining the role of Parasutterella, a previously uncharacterized member of the core gut microbiota. ISME J. 13, 1520–1534 (2019).
Article CAS PubMed PubMed Central Google Scholar
Wang, K. et al. Parabacteroides distasonis alleviates obesity and metabolic dysfunctions via production of succinate and secondary bile acids. Cell Rep. 26, 222–235 (2019).
Article CAS PubMed Google Scholar
Petersen, C. et al. T cell-mediated regulation of the microbiota protects against obesity. Science 365, eaat9351 (2019).
Article CAS PubMed PubMed Central Google Scholar
Luck, H. et al. Gut-associated IgA+ immune cells regulate obesity-related insulin resistance. Nat. Commun. 10, 3650 (2019).
Article ADS PubMed PubMed Central CAS Google Scholar
Macfarlane, G. T. & Macfarlane, S. Bacteria, colonic fermentation, and gastrointestinal health. J. AOAC Int. 95, 50–60 (2012).
Article CAS PubMed Google Scholar
Tremaroli, V. & Bäckhed, F. Functional interactions between the gut microbiota and host metabolism. Nature 489, 242–249 (2012).
Article ADS CAS PubMed Google Scholar
Cummings, J. H., Pomare, E. W., Branch, W. J., Naylor, C. P. & Macfarlane, G. T. Short chain fatty acids in human large intestine, portal, hepatic and venous blood. Gut 28, 1221–1227 (1987).
Article CAS PubMed PubMed Central Google Scholar
Mowat, A. M. & Agace, W. W. Regional specialization within the intestinal immune system. Nat. Rev. Immunol. https://doi.org/10.1038/nri3738 (2014).
Xu, M. et al. c-MAF-dependent regulatory T cells mediate immunological tolerance to a gut pathobiont. Nature https://doi.org/10.1038/nature25500 (2018).
Damgaard, M. T. F. et al. Age-dependent alterations of glucose clearance and homeostasis are temporally separated and modulated by dietary fat. J. Nutr. Biochem. 54, 66–76 (2018).
Article CAS PubMed Google Scholar
Friedman, S. L., Neuschwander-Tetri, B. A., Rinella, M. & Sanyal, A. J. Mechanisms of NAFLD development and therapeutic strategies. Nat. Med. 24, 1 (2018).
Article CAS Google Scholar
Giles, D. A. et al. Thermoneutral housing exacerbates nonalcoholic fatty liver disease in mice and allows for sex-independent disease modeling. Nat. Med. 23, 829–838 (2017).
Article CAS PubMed PubMed Central Google Scholar
Makula, R. A. Phospholipid composition of methane-utilizing bacteria. J. Bacteriol. 134, 771–777 (1978).
Article CAS PubMed PubMed Central Google Scholar
Kindt, A. et al. The gut microbiota promotes hepatic fatty acid desaturation and elongation in mice. Nat. Commun. 9, 3760 (2018).
Article ADS PubMed PubMed Central CAS Google Scholar
Singh, V. et al. Microbiota-dependent hepatic lipogenesis mediated by stearoyl CoA desaturase 1 (SCD1) promotes metabolic syndrome in TLR5-deficient mice. Cell Metab. 22, 983–996 (2015).
Article CAS PubMed PubMed Central Google Scholar
Forouhi, N. G. et al. Differences in the prospective association between individual plasma phospholipid saturated fatty acids and incident type 2 diabetes: the EPIC-InterAct case-cohort study. Lancet Diabetes Endocrinol. 2, 810–818 (2014).
Article CAS PubMed PubMed Central Google Scholar
Pagadala, M., Kasumov, T., McCullough, A. J., Zein, N. N. & Kirwan, J. P. Role of ceramides in nonalcoholic fatty liver disease. Trends Endocrinol. Metab. 23, 365–371 (2012).
Article CAS PubMed PubMed Central Google Scholar
Ju, C. & Tacke, F. Hepatic macrophages in homeostasis and liver diseases: From pathogenesis to novel therapeutic strategies. Cell. Mol. Immunol. 13, 316–327 (2016).
Article CAS PubMed PubMed Central Google Scholar
Ganeshan, K. & Chawla, A. Warming the mouse to model human diseases. Nat. Publ. Gr. 13, 458–465 (2017).
Google Scholar
Tian, X. Y. et al. Thermoneutral housing accelerates metabolic inflammation to potentiate atherosclerosis but not insulin resistance. Cell Metab. 23, 165–178 (2016).
Article CAS PubMed Google Scholar
Baffy, G. Kupffer cells in non-alcoholic fatty liver disease: the emerging view. J. Hepatol. 51, 212–223 (2009).
Article CAS PubMed PubMed Central Google Scholar
Odegaard, J. I. et al. Alternative M2 activation of Kupffer cells by PPARδ ameliorates obesity-induced insulin resistance. Cell Metab. 7, 496–507 (2008).
Article CAS PubMed PubMed Central Google Scholar
Dal-Secco, D. et al. A dynamic spectrum of monocytes arising from the in situ reprogramming of CCR2+ monocytes at a site of sterile injury. J. Exp. Med. 212, 447–456 (2015).
Article CAS PubMed PubMed Central Google Scholar
Ramachandran, P. et al. Differential Ly-6C expression identifies the recruited macrophage phenotype, which orchestrates the regression of murine liver fibrosis. Proc. Natl Acad. Sci. USA 109, E3186–E3195 (2012).
Article CAS PubMed PubMed Central Google Scholar
Li, F. et al. The microbiota maintain homeostasis of liver-resident γδ T-17 cells in a lipid antigen/CD1d-dependent manner. Nat. Commun. 7, 13839 (2017).
Article ADS PubMed CAS Google Scholar
Lathrop, S. K. et al. Peripheral education of the immune system by colonic commensal microbiota. Nature 478, 250–254 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Xiao, L. et al. High-fat feeding rather than obesity drives taxonomical and functional changes in the gut microbiota in mice. Microbiome 5, 1–12 (2017).
Article Google Scholar
Zhang, C. et al. Structural resilience of the gut microbiota in adult mice under high-fat dietary perturbations. ISME J. 6, 1848–1857 (2012).
Article CAS PubMed PubMed Central Google Scholar
Desai, M. S. et al. A dietary fiber-deprived gut microbiota degrades the colonic mucus barrier and enhances pathogen susceptibility HHS Public Access Graphical abstract. Cell 167, 1339–1353 (2016).
Article CAS PubMed PubMed Central Google Scholar
Dalby, M. J., Ross, A. W., Walker, A. W. & Morgan, P. J. Dietary uncoupling of gut microbiota and energy harvesting from obesity and glucose tolerance in mice. Cell Rep. 21, 1521–1533 (2017).
Article CAS PubMed PubMed Central Google Scholar
Lennon, G. et al. Correlations between colonic crypt mucin chemotype, inflammatory grade and Desulfovibrio species in ulcerative colitis. Color. Dis. 16, 161–169 (2014).
Article Google Scholar
Dawson, P. A. et al. Reduced mucin sulfonation and impaired intestinal barrier function in the hyposulfataemic NaS1 null mouse. Gut 58, 910–919 (2009).
Article CAS PubMed Google Scholar
Shan, M. et al. Mucus enhances gut homeostasis and oral tolerance by delivering immunoregulatory signals. Science 342, 447–453 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Persson, E. K., Scott, C. L., Mowat, A. M. & Agace, W. W. Dendritic cell subsets in the intestinal lamina propria: Ontogeny and function. Eur. J. Immunol. 43, 3098–3107 (2013).
Article CAS PubMed PubMed Central Google Scholar
Christoffersen, T. E. et al. Effects of the non-commensal Methylococcus capsulatus Bath on mammalian immune cells. Mol. Immunol. 66, 107–116 (2015).
Article CAS PubMed Google Scholar
Smith, P. M. et al. The microbial metabolites, short-chain fatty acids, regulate colonic Treg cell homeostasis. Science 341, 569–573 (2013).
Article ADS CAS PubMed Google Scholar
Karlmark, K. R. et al. Hepatic recruitment of the inflammatory Gr1+ monocyte subset upon liver injury promotes hepatic fibrosis. Hepatology 50, 261–274 (2009).
Article CAS PubMed Google Scholar
Moschen, A. R., Tilg, H. & Raine, T. IL-12, IL-23 and IL-17 in IBD: immunobiology and therapeutic targeting. Nat. Rev. Gastroenterol. Hepatol. 16, 185–196 (2019).
Article CAS PubMed Google Scholar
Bregenzer, N. et al. Increased insulin resistance and beta cell activity in patients with Crohn’s disease. Inflamm. Bowel Dis. 12, 53–56 (2006).
Article PubMed Google Scholar
Chassaing, B. et al. Lack of soluble fiber drives diet-induced adiposity in mice. Am. J. Physiol. Gastrointest. Liver Physiol. 309, G528–G541 (2015).
Article CAS PubMed PubMed Central Google Scholar
Zou, J. et al. Fiber-mediated nourishment of gut microbiota protects against diet-induced obesity by restoring IL-22-mediated colonic health. Cell Host Microbe 23, 41–53 (2018).
Article CAS PubMed Google Scholar
Stange, E. F. & Schroeder, B. O. Microbiota and mucosal defense in IBD: an update. Expert Rev. Gastroenterol. Hepatol. 00, 1–14 (2019).
Google Scholar
Ramanan, D. et al. An immunologic mode of multigenerational transmission governs a gut Treg setpoint. Cell 181, 1276–1290 (2020).
Article CAS PubMed PubMed Central Google Scholar
Song, X. et al. Microbial bile acid metabolites modulate gut RORγ+ regulatory T cell homeostasis. Nature 577, 410–415 (2020).
Article CAS PubMed Google Scholar
Campbell, C. et al. Bacterial metabolism of bile acids promotes generation of peripheral regulatory T cells. Nature 581, 475–479 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Furusawa, Y. et al. Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells. Nature 504, 446–450 (2013).
Article ADS CAS PubMed Google Scholar
Kleiner, D. E. et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 41, 1313–1321 (2005).
Article PubMed Google Scholar
Schneider, C. A., Rasband, W. S. & Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675 (2012).
Article CAS PubMed PubMed Central Google Scholar
Ruifrok, A. C. & Johnston, D. A. Quantification of histochemical staining by color deconvolution. Anal. Quant. Cytol. Histol. 23, 291–299 (2001).
CAS PubMed Google Scholar
Edgar, R. C. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10, 996–998 (2013).
Article CAS PubMed Google Scholar
Schloss, P. D. et al. Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 75, 7537–7541 (2009).
Article CAS PubMed PubMed Central Google Scholar
Edgar, R. C., Haas, B. J., Clemente, J. C., Quince, C. & Knight, R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27, 2194–2200 (2011).
Article CAS PubMed PubMed Central Google Scholar
Larsen, I. S. et al. Human Paneth cell α-defensin-5 treatment reverses dyslipidemia and improves glucoregulatory capacity in diet-induced obese mice. Am. J. Physiol. Metab. 317, E42–E52 (2019).
CAS Google Scholar
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Article PubMed PubMed Central CAS Google Scholar

Download references

Acknowledgements

This work was supported by the Norwegian Research Council (project 267655). B.A.H.J. was supported by Lundbeck Foundation (grant number: R232-2016-2425) and Novo Nordisk Foundation (grant number: NNF17OC0026698). A.-L.G. and A.M. were supported by Canadian Institutes for Heart Research and Sentinel North from the Canada First Research Excellence Fund.

Author information

Jacob B. Holm
Present address: Clinical Microbiomics, Copenhagen, Denmark
These authors contributed equally: Jacob B. Holm, Ida S. Larsen, Nicole von Burg
These authors jointly supervised this work: Benjamin A. H. Jensen, Karsten Kristiansen, Tor E. Lea.

Authors and Affiliations

Laboratory of Genomics and Molecular Biomedicine, Department of Biology, Faculty of Science, University of Copenhagen, Copenhagen, Denmark
Benjamin A. H. Jensen, Jacob B. Holm, Ida S. Larsen, Si B. Sonne, Simone I. Pærregaard, Mads V. Damgaard, Lise Madsen & Karsten Kristiansen
Department of Medicine, Faculty of Medicine, Cardiology Axis of the Québec Heart and Lung Institute, Laval University, Laval, QC, Canada
Benjamin A. H. Jensen, Ida S. Larsen, Anne-Laure Agrinier & André Marette
Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
Benjamin A. H. Jensen
Mucosal Immunology, Department of Health Technology, Technical University of Denmark, Copenhagen, Denmark
Nicole von Burg, Simone I. Pærregaard, Aymeric Rivollier & William Agace
Institute of Nutritional Medicine, University Hospital Schleswig-Holstein, Campus Lübeck, Lübeck, Germany
Stefanie Derer & Christian Sina
Novo Nordisk Foundation Center for Basic Metabolic Research, Integrative Metabolism and Environmental Influences, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
Mads V. Damgaard, Karolina Sulek & Jonas T. Treebak
Faculty of Chemistry, Biotechnology and Food Science, Norwegian University of Life Sciences, Oslo, Norway
Stine A. Indrelid, Yke J. Arnoldussen, Inga L. Angell, Knut Rudi, Charlotte R. Kleiveland & Tor E. Lea
Institute of Marine Research, Bergen, Norway
Even Fjære & Lise Madsen
Department of Anatomy and Pathology, Faculty of Veterinary Medicine, Norwegian University of Life Sciences, Oslo, Norway
Caroline Piercey Åkesson
Immunology Section, Department of Experimental Medical Science, Lund University, Lund, Sweden
William Agace
Institute of Metagenomics, BGI-Shenzhen, Shenzhen, P.R. China
Karsten Kristiansen

Authors

Benjamin A. H. Jensen
View author publications
You can also search for this author in PubMed Google Scholar
Jacob B. Holm
View author publications
You can also search for this author in PubMed Google Scholar
Ida S. Larsen
View author publications
You can also search for this author in PubMed Google Scholar
Nicole von Burg
View author publications
You can also search for this author in PubMed Google Scholar
Stefanie Derer
View author publications
You can also search for this author in PubMed Google Scholar
Si B. Sonne
View author publications
You can also search for this author in PubMed Google Scholar
Simone I. Pærregaard
View author publications
You can also search for this author in PubMed Google Scholar
Mads V. Damgaard
View author publications
You can also search for this author in PubMed Google Scholar
Stine A. Indrelid
View author publications
You can also search for this author in PubMed Google Scholar
Aymeric Rivollier
View author publications
You can also search for this author in PubMed Google Scholar
Anne-Laure Agrinier
View author publications
You can also search for this author in PubMed Google Scholar
Karolina Sulek
View author publications
You can also search for this author in PubMed Google Scholar
Yke J. Arnoldussen
View author publications
You can also search for this author in PubMed Google Scholar
Even Fjære
View author publications
You can also search for this author in PubMed Google Scholar
André Marette
View author publications
You can also search for this author in PubMed Google Scholar
Inga L. Angell
View author publications
You can also search for this author in PubMed Google Scholar
Knut Rudi
View author publications
You can also search for this author in PubMed Google Scholar
Jonas T. Treebak
View author publications
You can also search for this author in PubMed Google Scholar
Lise Madsen
View author publications
You can also search for this author in PubMed Google Scholar
Caroline Piercey Åkesson
View author publications
You can also search for this author in PubMed Google Scholar
William Agace
View author publications
You can also search for this author in PubMed Google Scholar
Christian Sina
View author publications
You can also search for this author in PubMed Google Scholar
Charlotte R. Kleiveland
View author publications
You can also search for this author in PubMed Google Scholar
Karsten Kristiansen
View author publications
You can also search for this author in PubMed Google Scholar
Tor E. Lea
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

B.A.H.J., J.B.H., C.R.K., K.K., and T.E.L. conceived and designed the study. B.A.H.J., J.B.H., I.S.L., S.B.S., and M.T.F.D. conducted the in vivo experiments. B.A.H.J., K.K., and T.E.L. supervised all parts of the study. J.T.T., A.M., L.M., W.A., and C.S. supervised parts of the study. N.v.B., S.I.P., and A.R. performed flow cytometry analyses. B.A.H.J., J.B.H., I.S.L., S.D., C.P.Å., N.v.B., A.L.A., A.R., S.A.I., Y.J.A., K.S., E.F., M.T.F.D., and SIP performed key experiments and analyzed the data. B.A.H.J. integrated the data and wrote the manuscript with inputs from J.B.H., I.S.L., K.K., and T.E.L. All authors edited, revised, and approved the final version of the manuscript.

Corresponding authors

Correspondence to Benjamin A. H. Jensen, Karsten Kristiansen or Tor E. Lea.

Ethics declarations

Competing interests

B.A.H.J., J.B.H., I.S.L., K.K., C.R.K., and T.E.L. are co-inventors of International (PCT) Patent Application No. PCT/EP2018/071076 based on the enclosed data. The remaining authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Reporting summary

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Jensen, B.A.H., Holm, J.B., Larsen, I.S. et al. Lysates of Methylococcus capsulatus Bath induce a lean-like microbiota, intestinal FoxP3⁺RORγt⁺IL-17⁺ Tregs and improve metabolism. Nat Commun 12, 1093 (2021). https://doi.org/10.1038/s41467-021-21408-9

Download citation

Received: 07 November 2019
Accepted: 27 January 2021
Published: 17 February 2021
DOI: https://doi.org/10.1038/s41467-021-21408-9

This article is cited by

The Current and Future Perspectives of Postbiotics
- Bing Liang
- Dongming Xing
Probiotics and Antimicrobial Proteins (2023)
Regulation of intestinal immunity by dietary fatty acids
- Jinxin Qiu
- Yanhui Ma
- Ju Qiu
Mucosal Immunology (2022)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Subjects

Abstract

Similar content being viewed by others

Introduction

Results

McB feeding reverses WD-induced gut microbiota changes and increases cecal SCFA levels

WDMcB feeding stimulates induction of gut-specific regulatory T cells

WDMcB mitigates diet-induced obesity

WDMcB feeding resets the hepatic lipidome and decreases hepatic immune infiltration alleviating NAFLD

WDMcB feeding reverses prolonged gut microbial dysbiosis and markedly improves colonic mucus production

McB lysates rely on adaptive immunity to favor Parabacteroides blooms

Altered gut microbiota by WD feeding affects glucose regulation

Discussion

Methods

Mice and ethical statements

Housing, diets, and experimental setup

12 + 6 protocol T22°C

21 + 5 protocol T30°C

Prophylactic protocols

Cecal microbiota transfer (CMT) protocol

Cultivation of M. capsulatus Bath and preparation of bacterial lysate

Glucose and insulin tolerance tests

Short chain fatty acid measurements

RNA extraction and quantitative RT-PCR

Histology

Liver

Colon

Hepatic lipidomics

Microbiome analysis and bioinformatic processing

Isolation of small intestine (SI) and large intestine (LI) lamina propria (LP) cells

Isolation of liver cells

Ex vivo stimulation of SI-LP, LI-LP, and liver cells

Flow cytometry

Antibodies (Abs) and reagents

Multiplex cytokine quantification

Statistical analysis

Reporting summary

Data availability

Code availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding authors

Ethics declarations

Competing interests

Additional information

Supplementary information

Rights and permissions

About this article

Cite this article

Share this article

This article is cited by

Comments

Search

Quick links

WD_McB feeding stimulates induction of gut-specific regulatory T cells

WD_McB mitigates diet-induced obesity

WD_McB feeding resets the hepatic lipidome and decreases hepatic immune infiltration alleviating NAFLD

WD_McB feeding reverses prolonged gut microbial dysbiosis and markedly improves colonic mucus production

12 + 6 protocol T_22°C

21 + 5 protocol T_30°C