Dietary carbohydrates alter immune-modulatory functionalities and DNA inversions in Bacteroides thetaiotaomicron

Gal-Mandelbaum, Noa; Carasso, Shaqed; Kedem, Alon; Ziv, Tamar; Keshet-David, Roni; Abboud, Roberto; Zaatry, Rawan; Gefen, Tal; Geva-Zatorsky, Naama

doi:10.1038/s41467-025-60202-9

Download PDF

Article
Open access
Published: 28 May 2025

Dietary carbohydrates alter immune-modulatory functionalities and DNA inversions in Bacteroides thetaiotaomicron

Nature Communications volume 16, Article number: 4938 (2025) Cite this article

8371 Accesses
142 Altmetric
Metrics details

Subjects

Abstract

The gut bacteria environment is highly dynamic. Environmental conditions were shown to affect microbial composition. Yet, their influences on bacterial functionality (e.g., immune-modulation activity) are mostly overlooked. Distinct strains of the same species, and even the same bacterial strain, may have different effects on the immune system depending on their growth environment. Therefore, studying the functionality of strains under different conditions is crucial. We analyzed functional alterations in the gut symbiont Bacteroides thetaiotaomicron (B. theta) under different dietary components consumption in humans, upon white sugar consumption in mice, and in response to 190 different carbon sources in vitro. Dietary alterations affected the orientation of phase variable regions in B. theta in humans, in vivo, and in vitro, and altered B. theta’s proteome and immune-modulatory functionality. Studying the effects of dietary components on the immune-modulatory functionalities of key members of the gut microbiota will allow for personalized dietary recommendations.

Stratified dietary inflammatory potential identifies oral and gut microbiota differences associated with cognitive function in older adults

Article Open access 30 May 2025

Metaproteomic portrait of the healthy human gut microbiota

Article Open access 28 June 2024

Gut microbiota-derived metabolites in the regulation of host immune responses and immune-related inflammatory diseases

Article 11 March 2021

Introduction

Gut bacteria reside in an environment that changes according to nutrition, health status, diurnal cycles of the host, and reciprocally affect the host’s immune system through synergistic and redundant paths^1,2,3,4. Strains of the same bacterial species can regulate the immune system in distinct ways, sometimes with contradicting effects (e.g., anti- and pro-inflammatory effects)^5,6. Moreover, the same bacterial strain may pose different effects on the immune system depending on its growth environment, implying that neighboring microbes, bacteriophages, diet, host physiological state, among others, are key in determining the bacterial strains’ immune-modulatory functionality⁷. To date, the host diet has been shown to affect microbial composition, while its effects on bacterial immune-modulation activity on the strain level are largely overlooked. Dietary interventions aimed at altering gut bacterial composition have shown inconsistent effects on humans. These inconsistencies are largely attributed to the emerging complex diet-host-microbiota relationship, which leads to the need for further research on the specific strain-nutrient context-dependent effects on the host^8,9,10. A prominent member of the human gut microbiota is Bacteroides thetaiotaomicron (B. theta), a Gram-negative, obligate anaerobe¹¹, which was previously shown to have anti-inflammatory effects, increase mucosal barrier function, and protect the host from pathogenic invasion^{2,12,13,14,15}. Over 80 polysaccharide utilization loci (PULs) in the genome of B. theta enable the utilization of polysaccharides that reach the colon and are not digested by the mammalian host^16,17,18. The availability of these polysaccharides differs according to the hosts’ diet, and by using PULs, B. theta can quickly adapt to its ever-evolving environment¹⁸. One mechanism that allows B. theta to adjust its PULs is DNA inversion. The genome of B. theta contains multiple inverted repeats, which enable reversible DNA inversions, and result in molecular phase variation. The DNA inversions are controlled by invertases and consequently allow plasticity in gene transcription. One type of inversion occurs in promoter regions, where promoters switch from “ON” to “OFF” states¹⁹. A second type includes DNA rearrangements, in which previously inactive genes are brought to the active transcription site^7,20. In parallel to utilizing complex polysaccharides as a source for bacterial metabolism, B. theta also produces eight distinct capsular polysaccharides (CPSs). Many of the PULs and CPSs of B. theta are phase variable²⁰, and their expression alters under different environmental conditions^18,20,21,22. CPSs and PULs are expressed on the outer surface of B. theta, facilitating interactions with host epithelial and immune cells^{20,22,23,24,25,26,27,28,29,30}. Phase variable outer surface molecules are prevalent across the gut microbiota, some of which have prominent immune-modulatory effects^22,31,32,33. For example, the phase variable outer surface polysaccharide A (PSA) of Bacteroides fragilis NCTC 9343, was shown to upregulate colonic regulatory T cells and induce interleukin (IL)−10³⁴. Furthermore, the polysaccharides of B. fragilis are differentially expressed, in a phase variable manner, when grown in different conditions (in vitro vs. in vivo)⁷. Different growth conditions may also be achieved by altering the available nutrients for the bacteria. Since B. theta was shown to modulate the host immune system^{2,12,13,14,15} and can adapt in response to environmental conditions^18,19,20,21, we hypothesized that such alterations can be identified in humans depending on their diet, and further evaluated and analyzed in mice and experimental setups. This study combined analysis of human cohorts under different dietary conditions with an analysis in a reductionist experimental system, both focusing on the response of B. theta to changing dietary conditions and its effects on the host immune system. Our data demonstrates genomic variations in B. theta in correlation to diet in humans, alongside genomic alterations of B. theta supplemented with dietary components in mice and 190 different carbon sources in vitro. We further characterize the altered bacterial molecules and the effects of these alterations on the host immune system. Our results stress the importance of focusing on the functionality of bacterial strains under various conditions and the potential effects of dietary carbohydrates on bacterial immune-modulatory functionality.

Results

Diet correlates with DNA inversion frequencies in gut bacteria of human subjects, specifically B. theta

DNA inversion frequencies, in relation to food consumption, in the genomes of gut bacteria of IBD patients and healthy individuals from the IBDMDB human database³⁵ were identified and analyzed as described in Carasso et al.²². In brief, the PhaseFinder software³¹ was used to identify putative invertible regions in representative Bacteroidales genomes in the gut and to assess their frequencies of inversions. Invertible regions were detected by identifying inverted repeats and creating a database containing their forward and reverse orientations (compared with the submitted genome sequences). Metagenomic sequencing results were then aligned to the invertible regions, resulting in the ratio of reverse-oriented reads to all reads aligned for each region. Identified DNA inversion sites were associated with the reported dietary questionnaire of all individuals. Thirty-one different strains out of 27 species of bacteria had phase variable regions (PVRs) that were significantly affected by at least one food component (p-value < 0.05). Of these 27 species, 23 belonged to Bacteroides species (Fig. 1a, Supplementary Data 1). Since B. theta is known to adapt to its environment and was previously demonstrated to modulate the host immune system^{2,12,13,14,15} the DNA inversion frequencies in the genome of B. theta were evaluated in relation to food consumption and health or disease state. Next, the identified DNA inversion sites were correlated with the reported dietary questionnaire of all individuals who harbor B. theta. Lastly, the diet questionnaire items were filtered for dietary components in correlation with DNA inversion frequencies of the identified invertible regions (total of 35 regions). The inversion frequencies are analyzed by the ratio of reverse reads out of all forward and reverse reads of the specified inversion region. Principal component analysis (PCA) of the dietary components revealed three clusters of individuals based on different consumption frequencies (Fig. 1b).

Fig. 1: Consumption of different dietary components affects the orientation of invertible regions in gut bacteria, specifically *B. theta.*

Four PVRs were found to have significantly different inversion frequencies between the three clusters when comparing the 35 identified PVRs of B. theta (Fig. 1c–e, and Supplementary Fig. 1). PVR1 is located at the promoter of a gene encoding a two-component system sensor histidine kinase (Fig. 1c 1). A higher ratio of reverse-oriented reads was observed in cluster 3 in comparison to cluster 1 (Fig. 1c 2). Significant differences in the consumption of soft drinks, processed meat, starch, and whole grains distinguished cluster 3 from cluster 1, specifically for this region. The consumption of soft drinks and processed meat was also significantly different between clusters 3 and 2, however, the inversion frequencies were not significantly different (Fig. 1c 4,5). Therefore, it is more likely to assign this change to the consumption of starch and whole grains. While individuals in cluster 3 consumed high amounts of starch (score 3.89) and average amounts of whole grains (score 3.11), individuals in cluster 1 consumed lower amounts of starch (score 2.93) and almost no whole grains (score 1.9, Fig. 1c 4). Moreover, there was an increase in the reverse orientation in this region in CD patients, although not significant, which may indicate an additional correlation of PVR1 to disease state (Fig. 1c 3). PVR2 is found in proximity to a 2-isopropylmalate isomerase gene (Fig. 1d 1) and had a significantly higher reverse orientation ratio in cluster 3 compared to clusters 1 and 2 (Fig. 1d 2). In PVR2, as in PVR1, cluster 3 was distinguished from clusters 1 and 2 by the high consumption of soft drinks and processed meat (Fig. 1d 4,5). Since no significant differences were found in the inversion frequencies according to individuals’ health status (Fig. 1d 3), this region orientation is likely more affected by diet than disease state. PVR3 is located in proximity to a region of outer membrane protein (Fig. 1e 1) and had a significantly higher reverse orientation ratio in cluster 2 compared to clusters 1 and 3 (Fig. 1e 2). Individuals in cluster 2 consumed significantly more alcohol (score 2.6) and less fruit juice (score 1.71) compared to individuals in clusters 1 and 3 (Fig. 1e 4, 5). Moreover, healthy individuals exhibited a higher reverse orientation ratio (Fig. 1e 3) which indicates that this region is also affected by individuals’ health status. PVR4 is located in proximity to hypothetical genes, and had a significantly higher reverse orientation ratio in cluster 3 compared to clusters 1 and 2 (Supplementary Fig. 1). Moreover, CD patients displayed a higher reverse orientation ratio (Supplementary Fig. 1), indicating that this region is also affected by individuals’ health status. These results demonstrate DNA inversions in four distinct regions of the genome of B. theta in the human gut, in response to different dietary consumption. The genomic region which was most significantly altered in response to dietary components, specifically soft drinks, was PVR2. This region was also the most frequently occurring region across human subjects.

White sugar causes alteration in the orientation of a phase variable region in B. theta and affects the immune-modulatory functionality of B. theta in monocolonized mice

To validate the effect of soft drinks on the orientation of PVR2 in B. theta in humans and its possible immune-modulatory effects, germ-free (GF) mice were either kept GF or monocolonized with B. theta. Ten days after monocolonization, GF and monocolonized mice were split into control groups that drank water and experimental groups that drank 9% white sugar supplemented water (“sugar water”). Twelve days following treatment, cell populations in the colon, spleen, and mesenteric lymph nodes (mLNs) were evaluated, as well as expression levels in the colon tissue of the cytokines tumor necrosis factor alpha (TNF-α), interleukin 6 (IL-6), and the tight junction protein zonula occludens (ZO-1, Fig. 2a, Supplementary Data 2). Analysis of the genomic DNA inversions of B. theta on the whole genome was performed by nanopore sequencing. Only PVR2 in the B. theta genome from mice stool was significantly more evident in its reverse orientation in monocolonized mice who drank sugar water compared to control mice, which was already apparent even seven days subsequent to the addition of sugar (Fig. 2b). An elevation in the reverse orientation ratio from day 10 to day 17 was observed in eight out of nine mice that drank sugar water compared to only three out of nine mice that drank normal water (Fig. 2b). Immune alterations were measured as an indication of functional changes in B. theta in response to white sugar consumption. Out of the analyzed immune parameters, we found that CD8+ CD62L+ cell percentages were significantly higher in the spleen and mLNs of GF mice drinking sugar water compared to water alone. A contrasting pattern of CD8+ CD62L+ cell percentages was observed in monocolonized mice that drank sugar water compared to water (Fig. 2c–e, Supplementary Fig. 6). Moreover, the expression of IL-6 and ZO-1 was downregulated in monocolonized mice that drank sugar water compared to water and compared to GF mice drinking sugar water (Fig. 2f–h). To evaluate the bacterial production of short-chain fatty acids (SCFA) in monocolonized mice that drank sugar water compared to water alone, SCFA concentrations were measured in mice feces on day 22. The concentrations of propanoic acid, acetic acid, and formic acid were significantly lower in monocolonized mice drinking sugar water compared to monocolonized mice drinking water (Supplementary Fig. 2 a, b, d), while the concentration of butanoic acid was not significantly altered (Supplementary Fig. 2 c). These results demonstrate the genomic and metabolic alterations in B. theta in response to white sugar and a correlation between the B. theta immune-modulatory functionality and its genomic DNA inversions.

Fig. 2: Consumption of white sugar affects the orientation of PVR2 in *B. theta* in monocolonized mice which in turn alters the immune-modulatory functionality of *B. theta.*

Reversible DNA inversions in B. theta in response to alternating white sugar consumption demonstrate immune-modulatory functional plasticity

To further demonstrate the plasticity of phase variation in PVR2 in B. theta in monocolonized mice and to reassure that this genomic change is correlated with the immune effects of B. theta, three groups of monocolonized mice with B. theta were used: monocolonized mice drinking water, monocolonized mice drinking sugar water, and monocolonized mice drinking sugar water that was exchanged to water after seven days. Twelve days following treatment (day 29), cell populations in the spleen and mLNs were evaluated, as well as expression levels in the colon tissue of the cytokine IL-6 and the tight junction protein ZO-1. Consequently, all were significantly altered in monocolonized mice that drank sugar water at day 22 (Fig. 2). PVR2 in the B. theta genome from mice stool was significantly more evident in its reverse orientation in monocolonized mice that drank sugar water compared to control mice, seven days following the addition of sugar. This significant elevation in the reverse orientation of PVR2 declined twelve days subsequent to exchanging sugar water to water (Fig. 3b). To note, the elevation in the reverse orientation of PVR2 seven days following sugar consumption was mostly driven by four mice, however, remained high in sugar water drinking mice but regressed in most mice that returned to drinking normal water (Fig. 3b).CD8+ CD62L+ cell percentages were significantly higher in the mLNs of monocolonized mice drinking sugar water compared to water alone. However, these cells were not significantly altered in monocolonized mice drinking sugar water that was replaced with water (Fig. 3c, d, Supplementary Fig. 6). Moreover, the expression of IL-6 and ZO-1 were altered in monocolonized mice drinking sugar water compared to water. However, their expression was not significantly altered in monocolonized mice drinking sugar water that was replaced with water (Fig. 3e, f). These results demonstrate the reversibility of the response of B. theta to white sugar pertaining to DNA inversions and altered immune-modulations.

Fig. 3: The effect of white sugar consumption on the orientation of PVR2 in *B. theta* in monocolonized mice is reversible, which in turn reverses the immune-modulatory effects of *B. theta.*

B. theta immune-modulatory effects are altered when grown on distinct carbon sources

Inversion frequencies of the identified regions in B. theta in humans were altered in correlation to the consumption of various sugars (soft drinks, whole grains, starch, and alcohol, Fig. 1). Furthermore, consumption of white sugar by moncolonized mice and the consumption of soft drink by humans showed a similar effect on the orientation of PVR2. Since white sugar is composed mainly of sucrose, a disaccharide of glucose and fructose, we further studied the effect of specific sugars on invertible DNA regions of B. theta and its immune-modulatory functionality. To examine the suspected effects, B. theta was grown in M9 minimal media with 190 distinct carbon sources using PM1 and PM2A Biolog Phenotype MicroArrays™ plates for 24 h. Bacterial growth was assessed by both OD₆₀₀ measurements and colony-forming unit (CFU) counts (Supplementary Fig. 3). B. theta utilized simple monosaccharides, disaccharides, oligosaccharides, and complex polysaccharides (a total of 22 carbon sources out of 190 carbon sources in the Biolog plates) within 24 h (Supplementary Fig. 3b–e). Previous research has shown that gut bacterial modulation of the host’s immune system can be mediated through secreted bacterial molecules directly to the environment or through secreted vesicles^36,37,38,39. Therefore, bacteria conditioned media (CM) was collected to assess whether growth on distinct carbon sources influences the immune-modulatory functionality of B. theta. CM was introduced to splenocytes isolated from specific pathogen-free (SPF) mice. The splenocytes were activated with anti-CD3 and supplemented with CM from each of the 22 different carbon sources. Mouse splenocytes enable the assessment of the impact of bacterial entities on multiple immune cell populations present in the spleen. The following immune markers were assessed after five days of incubation with bacteria CM: the percentage of active and proliferating CD8 + Ki67 + PD1 + T cells, interleukin 10 (IL-10) concentration, an anti-inflammatory marker, and IL-17 concentration, a pro-inflammatory marker. Log2 fold change from the control group (splenocytes exposed to minimal media containing each of the carbon sources without bacteria having grown in it) was calculated for each immune parameter. For all carbon sources, CD8 + Ki67 + PD1+ cell percentages were higher than the control group (Fig. 4b, c, Supplementary Fig. 7). In PM1, log2 fold change values of CD8 + Ki67 + PD1+ were significantly higher in CM of B. theta grown on N-Acetyl-D-Glucosamine, D-fructose, and sucrose compared to maltotriose (Fig. 4b). CM of B. theta grown on D-fructose, α-D-Lactose, and sucrose also had higher log2 fold change values of IL-10 compared to maltotriose and D-melibiose (Fig. 4d). IL-17 measurements showed no significant differences in B. theta CM grown on all PM1 carbon sources (Fig. 4f). In PM2A, CM of B. theta grown on N-Acetyl-D-Galactosamine, gentiobiose, and mannan showed higher log2 fold change values of CD8 + Ki67 + PD1+ compared to γ-Cyclodextrin and dextrin (Fig. 4c). CM of B. theta grown on N-Acetyl-D-Galactosamine, gentiobiose, mannan, and α-Cyclodextrin also had higher log2 fold change values of IL-10 compared to pectin and chondroitin sulfate C (Fig. 4e). The IL-17 log2 fold change value of CM from B. theta grown on α-Cyclodextrin significantly differed from all carbon sources except pectin (Fig. 4g). The Z-score was calculated for all immune parameters comparing log2 fold change values from both PM1 and PM2A. CM from B. theta grown on maltotriose and D-melibiose had a Z-score equal or higher than two in the IL-10 parameter, while α-Cyclodextrin had a Z-score equal or higher than two in the IL-17 parameter (Fig. 4h). These results demonstrate alterations in the immune-modulatory effects of B. theta, specifically its conditioned media, arising from the consumption of different carbohydrates.

**Fig. 4: Conditioned media of *B. theta* grown on distinct carbon sources has altered effects on the immune system.**

SCFA concentrations do not correlate with the conditioned media immune-modulatory effects

SCFA such as propionate, acetate, and butyrate are known to induce the secretion of anti-inflammatory cytokines such as IL-10 and reduce the secretion of pro-inflammatory cytokines such as IL-17^40,41. Since our study revealed altered immune-modulatory effects of the CM of B. theta, where SCFA can be present, we sought to analyze the SCFA concentrations with the intent of characterizing the bacterial molecules that mediate immune system regulation. We measured the concentrations of SCFAs in the conditioned media of B. theta focusing on carbon sources that showed significant alterations in immune parameters. While SCFA secretion by B. theta was affected by growth on distinct carbon sources, little to no correlation was found between SCFA concentrations and the immune-modulatory effects: in 11 out of 12 carbon sources, no correlation was found between the concentrations of propionate, butyrate, and acetate, measured in the CM, and the secretion of IL-10 or IL-17 by immune cells nor the CD8 + Ki67 + PD1+ cell percentage (Supplementary Fig. 4a–e, Fig. 4). A correlation was found between butyrate and acetate concentrations and IL-10 secretion, as well as CD8 + Ki67 + PD1+ cell percentage in the CM of B. theta grown on D-Fructose only (Supplementary Fig. 4a, f, Fig. 4). No correlations were found between formate concentrations in the CM of B. theta grown on all carbon sources and the CM immune-modulatory effects.

Boiling of B. theta conditioned media affects its immune-modulatory functionality

To further characterize the bacterial molecules mediating the effects on the immune system, CM was boiled, which consequently denatures proteins and mostly alters large molecules⁴² and was introduced to mouse splenocytes. Subsequently, CD8 + Ki67 + PD1+ cell percentage, IL-10, and IL-17 concentrations were monitored. A decrease in CD8 + Ki67 + PD1+ log2 fold change values was observed in boiled CM from B. theta grown on N-Acetyl-D-Glucosamine, D-fructose, sucrose, and maltotriose in PM1 (Fig. 5a, Supplementary Fig. 7) in addition to γ-Cyclodextrin, N-Acetyl-D-Galactosamine, and gentiobiose in PM2A (Fig. 5b, Supplementary Fig 7). A decrease in the log2 fold change values of IL-10 was observed in boiled CM from B. theta which was grown on N-Acetyl-D-Glucosamine, sucrose, maltotriose in PM1, chondroitin sulfate C, N-Acetyl-D-Galactosamine, and gentiobiose in PM2A (Fig. 5c, d). Most CM induced lower IL-17 secretion compared to the control (i.e., negative log2 fold change values). Boiled CM from B. theta which was grown on N-Acetyl-D-Glucosamine in PM1, dextrin, mannan, and N-Acetyl-D-Galactosamine in PM2A resulted in higher log2 fold change values of IL-17 (Fig. 5e, f). However, boiled CM from B. theta that was grown on γ-Cyclodextrin showed a decrease in the log2 fold change value (Fig. 5f). This decrease might be attributed to the exposure of different sites in the denatured proteins or the breakdown of molecule complexes that would expose new active molecules^42,43. CM of B. theta that was grown on α-Cyclodextrin and pectin did not exhibit alteration in its impact on the immune system after boiling, indicating that the active molecules in this CM are likely small molecules or metabolites unaffected by boiling (Fig. 5).

**Fig. 5: Boiling of *B. theta* conditioned media reduces some but not all conditioned media effects on the immune system.**

Proteomics analysis of B. theta cells and conditioned media reveals specific pattern for each carbon source and significant enrichment of starch utilization system (Sus) proteins

In order to characterize the protein content of the CM of B. theta introduced to immune cells, proteomics analysis was performed (Supplementary Data 3). Proteomics analysis was also done on B. theta cells to compare it to the CM protein content and to reveal intracellular processes that might affect the molecular content of the CM (Supplementary Data 5). These analyses included all carbon sources that showed significant differences in their effects on the immune system, except sucrose, which had low protein content and was excluded from the analysis (Data are available via ProteomeXchange with identifier PXD060215⁴⁴). CM of B. theta grown on carbon sources that showed significant alterations in the immune parameters demonstrated 136 significantly altered proteins (adjusted p-value < 0.01). Samples of CM were clustered using hierarchical clustering based on Pearson correlations. Each carbon source showed a distinct protein pattern except α-Cyclodextrin, γ-Cyclodextrin, and maltotriose, which were not separated into clusters (Fig. 6a). No unique proteins were found in the CM from B. theta grown on α-Cyclodextrin, which supports the assumption that the active molecule inducing IL-17 secretion is not a protein. Out of these 136 proteins, 70 proteins are outer membrane proteins, 35 are SusC proteins, 29 are SusD proteins and within those, 10 proteins are responsible for starch binding (Fig. 6b). Annotation enrichment analysis of all 136 proteins revealed enrichment of Sus proteins, ionic and organic substances transport, and macromolecule, carbohydrate and polysaccharide catabolic pathways (Fig. 6c, Supplementary Data 4). Over half of the proteins that were significantly altered in CM are outer membrane proteins, with most of them being Sus proteins, suggesting that these proteins are secreted into the media via outer membrane vesicles (OMVs) or cell debris. Samples of B. theta cells that grew on carbon sources that showed significant modifications in the immune parameters were clustered based on Pearson correlation. 1266 proteins were found to be significantly altered (adjusted p-value < 0.01) when B. theta cells grew in distinct carbon sources. These proteins were divided into eight different clusters using hierarchical clustering based on Pearson correlations, and annotation enrichment analysis was performed on each of the clusters (Supplementary Data 6). Proteins in cluster 1 were found mostly in B. theta cells that grew on mannan, N-Acetyl-D-Galactosamine, chondroitin sulfate C, α-Cyclodextrin, γ-Cyclodextrin, and maltotriose. Proteins enriched in this cluster belong to pathways of biosynthesis of secondary metabolites, ribosomes, biosynthesis of amino acids, microbial metabolism in diverse environments, and carbon metabolism (Fig. 6a, e). Proteins in cluster 2 were present in all samples except B. theta cells that grew on mannan. This cluster contained mainly proteins related to pathways of microbial metabolism in diverse environments, fructose and mannose metabolism, and pentose phosphate pathway (Fig. 6a, f). Cluster 3 consists of proteins found mostly in cells that grew on gentiobiose, N-Acetyl-D-Glucosamine, D-fructose, and pectin (Fig. 6a, g). Cluster 4 contains proteins that were mostly upregulated in cells grown on pectin (Fig. 6a, h). However, boiling of CM of B. theta that was grown on pectin did not alter the effect on the immune system, which indicated that these proteins are not immune-modulatory (Fig. 5d, f, h). Cluster 5 includes proteins that were significantly upregulated in B. theta cells grown on N-Acetyl-D-Galactosamine. Apart from pathways of glycan degradation and sugar metabolism, this cluster contains proteins related to quorum sensing, which were not found in any other cluster (Fig. 6a, i). Cluster 6 contains proteins that were found in cells grown on mannan, with only one pathway enriched- six-hairpin glycosidase superfamily (Fig. 6a, j). Cluster 7 consists of proteins found mostly in cells grown on chondroitin sulfate C, α-Cyclodextrin, γ-Cyclodextrin, and maltotriose. The pathways enriched in cluster 7 were catabolic processes of organic substances, cellular, starch, and polysaccharides (Fig. 6a, k).

**Fig. 6: Proteomics analysis of bacterial cells and conditioned media of *B. theta* grown on carbon sources with significant alterations in their immune-modulatory effects.**

Genomic orientations of specific invertible DNA regions (mostly PULs and CPSs) in B. theta alter in response to growth on distinct carbon sources

To identify the changes in B. theta that modulated its effects on the immune system in response to growth on different carbon sources, we analyzed the state of phase variable sites in its genome. DNA was extracted from stationary-phase B. theta grown on carbon sources, which showed significant altered effects (P < 0.01) on the immune system. Long-read Nanopore sequencing was used to compare the orientation of phase variable promoters to the reference genome. DNA inversion sites were detected, and reads were mapped using the PhaVa tool⁴⁵ (Data are available via the Sequence Read Archive (SRA) with identifier PRJNA1218017). The analysis was performed using only regions with ten or more reads supporting either forward or reverse orientations. Following the evaluation of all samples from all carbon sources, we found eleven regions with sufficient reads and Z-scores were calculated. A total of eight regions showed significantly differing inversion frequencies of higher or lower than two standard deviations from the mean (Fig. 7a), and six regions exhibited significant inversion frequencies between the different carbon sources (Fig. 7b–h). One of the eight regions with a Z-score higher than two standard deviations was PVR2 which was altered in humans and in vivo. However, no significant differences in the ratio of reverse orientation of PVR2 between B. theta grown on the distinct carbon sources, and specifically sucrose, were detected. Three of the regions with significant differences in inversion frequencies are in the promoter areas of loci responsible for the expression of CPSs previously shown to be altered in response to dietary alterations involving fiber²⁰ (Fig. 7b–d). CPS3 was significantly and predominantly in its “ON” orientation in B. theta cells grown on N-Acetyl-D-Galactosamine (Fig. 7b), and CM of B. theta grown on this carbon source induced high IL-10 secretion in splenocytes (Fig. 4e). Notably, CPS3 existed more frequently in its “OFF” orientation in IBD patients²². Furthermore, the PCA of both immune parameters and phase variable regions demonstrated that the IL-10 and CPS3 phase “ON” orientation loadings aligned (Fig. 7i, Supplementary Fig. 5). CPS5 was significantly more apparent in the “ON” orientation in B. theta cells that were grown on α-Cyclodextrin (Fig. 7c). CM from B. theta grown on α-Cyclodextrin caused high secretion of IL-17 by splenocytes (Fig. 4g), which might imply an immune-modulatory role of CPS5. The PCA analysis showed that IL-17 and CPS5 loadings aligned and were opposite to IL-10 and CPS3 (Supplementary Fig. 5). CPS8 was significantly more evident in the “ON” orientation in B. theta grown on N-Acetyl-D-Glucosamine (Fig. 6d). In line with CPS8 being highly immune-stimulative in a single-CPS expressing B. theta strain²³, CM from B. theta grown on N-Acetyl-D-Glucosamine induced high percentages of CD8 + Ki67 + PD1 + T cells in our earlier described experiment (Fig. 4b). Moreover, PCA analysis demonstrated that CPS8 “ON” orientation aligned with CD8 + Ki67 + PD1+ (Supplementary Fig. 5). PUL70/73 region was in a high “ON” orientation in B. theta grown on N-Acetyl-D-Galactosamine (Fig. 6f) and aligned with IL-10 and CPS3 loadings in the PCA (Supplementary Fig. 5). Assessing B. theta grown on D-fructose, a higher “ON” orientation was measured in the PUL78/80 region (Fig. 7g) and PUL12 region (Fig. 7h). To note, PUL70/73, PUL78/80, and PUL12 are known to be expressed in B. theta that was grown in the environment of mucin O-glycans¹⁷, which contain N-Acetyl-D-Galactosamine.

Fig. 7: Phase variable regions in *B. theta* genome alter their orientation in response to growth in distinct carbon sources. DNA was extracted from *B. theta* cells that grew on carbon sources with significant differences in their immune-modulatory effects.

Discussion

Studies have demonstrated the influence of diet on microbiome composition^45,46,47, and the effects on its global functionality (e.g. SCFA secretion)^48,49,50. However, the impact of diet on specific gut bacteria functionality, regardless of whether their abundance is altered, are relatively scarce^10,29,51,52. Recent reports have shown the ability of the gut microbiota to adjust to environmental stimuli in a reversible manner, termed functional plasticity^22,31,45,46. This functional plasticity enables the bacteria to adapt in response to various conditions such as host diet, physiological state, neighboring bacteria, among others, which can potentially modulate the host immune system^30,53,54. Considering gut bacterial functional plasticity in response to environmental stimuli such as diet, there is a considerable need to comprehensively examine gut bacteria-nutrient interactions and their effects on the host’s immune system and disease state⁹. Investigation of the IBDMDB human database, incorporating dietary surveys and microbiome metagenomic sequences, revealed 27 different species that underwent phase-variation in response to dietary components. Since B. theta can alter its functions via reversible DNA inversions and has immune-modulatory functions^2,29,55, we sought to delve into the effects of diet on the immune-modulatory functionality of B. theta. We studied the effects of dietary components on the functionality of B. theta in humans, gnotobiotic mice models, and in a reductionist in vitro system evaluating B. theta’s immune-modulatory plasticity upon exposure to specific carbon sources. In humans, our results showed correlations between the consumption of specific dietary components and the inversion frequencies of four invertible sites in the genome of B. theta (Fig. 1, Supplementary Fig. 1). The main dietary components correlating with DNA inversions contained different types of carbohydrates. The most notable one was soft drinks containing white sugar. Supplementing drinking water of B. theta monocolonized mice with white sugar resulted in a similar PVR orientation alteration found in humans consuming soft drinks, and altered the effects of B. theta on the mouse immune system (Fig. 2). At day 22, monocolonized mice drinking sugar water had lower IL-6 and ZO-1 expression in the colon tissue, which could be interpreted as compromised mucosal repair^56,57. Moreover, these mice had lower CD8+ CD62L+ effector-memory T-cells in the spleen and mLNs. Analysis of the SCFA produced by B. theta under these conditions did not correlate with the immune-modulatory effects^41,58. At day 29, mice drinking sugar water presented a higher percentage of CD8 + CD62L+ effector-memory T-cells in the mLNs and higher IL-6 and ZO-1 expression in the colon tissue compared to mice drinking water and mice drinking sugar water that were changed back to water (Fig. 3). These effects, although seem contrary to the effects observed in mice drinking sugar water, at day 22, demonstrate the altered immune-modulatory effects of B. theta. Furthermore, the bacterial immune-modulatory effect is plastic, therefore, the increase in these parameters compared to the decrease presented in Fig. 2 may be explained by the different timepoints in which the parameters were measured. These results support our initial hypothesis that dietary sugar intake alters B. theta functionality and strengthen the need to mechanistically understand the specific effects of different carbohydrates on B. theta’s functions.

Many DNA inversions occur in outer surface molecules, including polysaccharides, making it reasonable to hypothesize that these alterations can affect B. theta’s immune-modulatory effects on the host^23,53. Further investigation of the conditioned media (CM) of B. theta grown on an array of distinct carbon sources revealed differential immune-modulatory effects on immune cells, in vitro. Some significant differences in the concentrations of SCFA in the CM were measured (Supplementary Fig. 4), however, only the concentrations of acetate and butyrate correlated with the immune profile in only one out of twelve carbon sources that were assessed. On the other hand, a differential protein pattern was observed in CM of B. theta grown on each of the carbon sources, with most being membrane proteins belonging to PUL families (Fig. 6). These findings suggest that the altered proteins are secreted into the media, mediating the immune effects. In addition, the protein families enriched in the CM (predominantly PULs) indicated that the proteins secreted to the media could facilitate bacterial adaptation to altered environments⁵⁹. Lastly, we revealed genomic DNA inversions, mostly in PULs and CPSs, which are correlated with specific immune effects of B. theta grown on distinct carbon sources (Figs. 4 and 7). Notably, PVR2 which was more frequently observed in the reverse orientation in humans drinking soft drinks and mice drinking sugar water did not significantly change in response to sucrose in vitro. This might be explained by the inherent differences between the in vitro “clean” setting and the in vivo setting which includes many other host-related factors that may induce the phase variation in the bacteria. Some of the PUL-encoding regions in the genome of B. theta are phase variable and are regulated by the same recombinase⁶⁰. This mechanism allows the utilization of varied carbohydrate sources through the synthesis of different combinations of SusC/SusD outer membrane proteins, thus providing resilience to changing environments and nutrients. Notably, PUL 78/80 and PUL12 regions were more frequently in their “ON” orientation when B. theta that was grown on D-fructose. In parallel, acetate and butyrate concentrations were higher in the CM of B. theta grown on D-fructose.

These results may indicate that a change in the available carbohydrates may affect the orientation of phase variable PULs and, as a result, the production and secretion of SCFAs. PUL70/73 is expressed in B. theta when exposed to mucin O-glycans¹⁷ and was in its “ON” orientation when B. theta was grown on N-Acetyl-D-Galactosamine ̶ a component of mucin O-glycans in the gut known to be degraded by B. theta^17,61. Moreover, under this same growth condition, CPS3 was also more frequently in its “ON” orientation. Intriguingly, the CM of B. theta grown on N-Acetyl-D-Galactosamine induces secretion of IL-10 by immune cells. An opposite immune response occurs when B. theta is grown on α-Cyclodextrin, inducing the secretion of IL-17. Under these conditions, the CPS5 of B. theta was turned “ON”. PULs and outer surface polysaccharides were previously shown to have immune-modulatory effects on the host^30,34,62, and, therefore, might cause the observed immune modulations under these conditions. Taken together, supplementation of N-Acetyl-D-Galactosamine to diet could potentially affect B. theta’s carbon source utilization preferences⁶³, its DNA inversion pattern, and subsequently alter its immune-modulatory effects on the host. Our results, taking an experimental approach and based on human subjects’ analyses, emphasize the functional plasticity of B. theta when exposed to different carbohydrates (Fig. 8). By studying the effects of specific dietary components on the immune-modulatory functionality of key members of the gut microbiota, tailored dietary recommendations can be given to human subjects based on their microbiome composition.

**Fig. 8: An illustration portraying how dietary components such as white sugar (and various carbohydrates) affect gut bacterial immune-modulatory functionalities.**

Limitations

We based our human diet analyses on a questionnaire designed by Lloyd-Price et al.³⁵. Although this questionnaire was very helpful, its resolution was limited (e.g., “soft drinks” included tea/coffee with sugar, coke, juice without fruits, and others without specifying the amount of sugar in each). The study was based on the human subjects’ regular diet, which is very diverse, rather than a supervised diet-centered study. Sucrose is known to mostly degrade in the small intestine with <5% reaching the colon⁶⁴, therefore might reach B. theta in low percentages, which reside mostly in the colon. Although white sugar is composed of simple carbohydrates (i.e., sucrose, which is a disaccharide of glucose and fructose), the Biolog plates include sucrose, glucose and fructose – the source of the carbohydrates in the Biolog plates is not necessarily white sugar. Additionally, although the Biolog plates include 190 different carbohydrates, there are many additional carbohydrates relevant for gut bacteria which are not included in these plates (e.g. mucin and its derivatives). Some of the inversion sites detected using PhaseFinder in the IBDMDB cohort were not identified by PhaVa likely due to differences in detection criteria. PhaVa applies stringent filters, including MAPping Quality thresholds, read-span requirements, and mismatch rate cutoffs, which may exclude certain regions from the analysis. These regions had to be checked separately. While most of the proteins altered are outer surface proteins, and many are PULs, which are phase variable, it is hard to find the direct connection between the observed DNA inversion states and the protein array. The DNA inversions are calculated as percent of the population, and therefore, the protein array might include proteins from both orientations. Further research is required to unveil the dynamic process from the DNA inversion to protein presence. Lastly, the immune phenotyping was limited to the markers we have chosen to focus on. Consequently, we do not have the comprehensive immune profiling in both the in vitro and in vivo analyses.

Methods

Strain culture

B. theta ATCC29148 was thawed on Brain Heart Infusion agar plates (BHI, BD BBLTM) supplemented with 5 µg/ml hemin (Alfa Aesar) in 1 N NaOH and 2.5 µg/ml vitamin K (Thermo Fisher Scientific) in 100% EtOH (previously referred to as BHIS), at 37 °C in an anaerobic chamber, 85% N₂, 10% CO2, 5% H₂ (COY).

Biolog plates culture

B. theta ATCC29148 was transferred from BHIS agar plates into M9 minimal media (1% w/v NH₄Cl, 6% w/v Na₂HPO₄, 3% w/v KH₂PO₄, 0.5% w/v NaCl) supplemented with 0.1 mM CaCI₂.2H₂O, 1 mM MgSO₄.7H₂O, 0.5% w/v glucose, 0.05% w/v L-cysteine, 5 µg/ml hemin (Alfa Aesar) in 1 N NaOH and 2.5 µg/ml vitamin K (Thermo Fisher Scientific) in 100% EtOH, 0.0005% w/v vitamin B12 and was grown overnight at 37 °C under anaerobic conditions. B. theta was diluted 1:50 into M9 minimal media lacking any carbon source (i.e. glucose) and 100 µl was distributed to each well in PM1 and PM2A Biolog (Biolog Inc., Hayward, CA, USA) Phenotype MicroArrays™ plates. Plates were incubated for 24 h at 37 °C under anaerobic conditions. Bacterial conditioned media was collected and filtered with a 0.22-µm-pore-size filter.

Growth analysis

Bacteria were diluted 1:50 into M9 minimal media lacking any carbon source (i.e. glucose) and 100 µl was distributed to each well in PM1 and PM2 Biolog (Biolog Inc., Hayward, CA, USA) Phenotype MicroArrays™ plates. Twenty microliters of mineral oil were added on top of the minimal media and plates were placed in a plate reader for 24 h at 37 °C under anaerobic conditions. OD600 was measured every 20 min. Colony-forming units (CFU) were determined after 24 h for each well.

Media boiling

Filtered bacterial conditioned media and controls were boiled for 5 min at 95 °C.

Splenocytes culture

Spleens were harvested from C57BL/6 SPF mice and were mechanically disrupted on a 40 µm cell strainer with 1 ml of ACK lysing buffer (Thermo Fisher Scientific) for 1–2 min. Cells were then transferred to 20 ml ice-cold RPMI (Sartorius) medium and centrifuged for 10 min at 4 °C and 300 × g. Cells were washed twice with 10 ml of ice-cold RPMI, counted and diluted with 37 °C RPMI to a final concentration of 2 million cells/ml. Cells were then supplemented with 0.05 mM β-Mercaptoethanol (Merck) and anti-CD3 (0.5 µg/ml, 145-2C11, BioLegend) for suboptimal activation of the T cells. 100 µl of cells were plated in each well of a 96-well plate and were added with 100 µl of diluted bacterial conditioned media (1:50 dilution, with similar CFU/ml of each carbon source). Cells and their media were collected 5 days later for flow cytometry and ELISA analysis.

Flow cytometry

A constant panel of antibodies was used for consistency. For splenocytes in vitro assay the panel included antibodies against CD4 (RMA-5, #100559), CD8 (53-6.7, #100730), TCRβ (H57-597, #100320), CD19 (6D5, #115530), Ki67 (16A8, #652418), PD-1 (29 F.1A12, #135206, all from BioLegend) and zombie UV (Biolegend) for dead cells staining. For intracellular staining, cells were stained for surface markers and fixed in Fix/Perm buffer (eBioscience) for 30–60 min at RT, and permeabilized in permeabilization buffer (eBioscience) at RT for 30 min in the presence of antibodies. Cells were acquired with a BD BioSciences® LSRFortessa, and analysis was performed with Kaluza® Analysis Software. The concentration, clone, and source of antibodies were kept constant to ensure consistency in staining.

Mice monocolonization and water supplementation

All mouse work was in accordance with protocols approved by the local Technion IACUC committee under approval numbers: IL-0620421 and IL-1350921. Mice were fed Ad libitum (Autoclaved, Teklad 2018S, ENVIGO) and grown at 23 °C with 60% humidity and 12:12 dark-light cycles. GF C57BL/6 wild-type mice (males and females, Taconic) were transferred from isolators to isocages at 4 weeks of age. Mice were either kept GF or were orally inoculated by gavage with 200 µl containing 10^9 CFU/ml of B. theta ATCC29148 that was grown overnight in BHIS. Monocolonized and GF mice were kept in the same conditions for 10 days, then 9 g per 100 ml white sugar was added to the drinking water. For immunological assays mice were kept under the specified conditions for 12 days. Each group of mice was housed in gnobiotic isocages under sterile conditions for the course of the experiment. Fecal material was collected over the course of the experiment, plated and incubated both aerobically and anaerobically to ensure monocolonization by a single bacterial strain. The identity of all colonizing microbial species was confirmed by 16S sequencing using the 27F and 1492R primers (Supplementary Data 2) and Sanger sequencing at the Technion Genome Center Facility.

Preparation of lymphocytes and flow cytometry

Single-cell suspensions of colon tissues lymphocytes were generated using Lamina Propria Dissociation Kit, mouse (Miltenyi Biotec) following the manufacturer’s instructions. Mesenteric lymph nodes (mLNs) were mechanically disrupted on a 40 µm cell strainer with 1 ml Hanks’ Balanced Salt Solution (HBSS, Sigma) medium containing 10 mM HEPES (Thermo Fisher Scientific). Spleens were mechanically disrupted on a 40 µm cell strainer with 1 ml of ACK lysing buffer (Thermo Fisher Scientific) for 1–2 min. Cells were then transferred to 10 ml ice-cold HBSS medium containing 10 mM HEPES and centrifuged for 10 min at 4 °C and 300 × g. Cells were washed twice with 10 ml of ice-cold HBSS medium containing 10 mM HEPES. Single-cell suspensions from all tissues were washed twice with PBS and stained for surface and intracellular markers.

A constant panel of antibodies was used for consistency. The panel included antibodies against CD45 (30F11, #103146) CD4 (RMA-5, #100559), CD8 (53-6.7, #100730), TCRβ (H57-597, #100320), CD19 (6D5, #115530), CD62L (MEL-14, #104432), CD44 (IM7, #103057), Ki67 (16A8, #652418), PD-1 (29 F.1A12, #135220), mhHelios (22F6, #137220, all from BioLegend), mhRORgt (AFKJS-9, #12-6988-82), Foxp3 (FJK-16s, #17-5773-82, both from eBioscience) and zombie UV (Biolegend) for dead cells staining. For intracellular staining, cells were stained for surface markers and fixed in Fix/Perm buffer (eBioscience) for 30–60 min at RT, and permeabilized in permeabilization buffer (eBioscience) at RT for 30 min in the presence of antibodies. Cells were acquired with a BD BioSciences® LSRFortessa, and analysis was performed with Kaluza® Analysis Software. The concentration, clone, and source of antibodies were kept constant to ensure consistency in staining.

RNA extraction and RT-qPCR

RNA was extracted from 0.5 cm pieces of colon tissue. Tissues were flash frozen in liquid nitrogen and then transferred into 1 ml Trizol and cut into small pieces using surgical scissors. Trizol containing tissues was transferred to 2 mL tubes containing two 3.2 mm stainless steel beads (Next Advance) and homogenized for three rounds of 30 s at 4m/s using the Bioprep-24 Homogenizer (Allsheng). Samples were centrifuged at 12,000 × g for 5 min, Trizol was transferred to clean tubes and 300 μl of chloroform was added. Sample tubes were mixed thoroughly by inversion, incubated for 3 min at room temperature, and centrifuged for 15 min at 4 °C. The aqueous phase was transferred to clean tubes and 0.5 ml of isopropanol was added. Samples were incubated at RT for 10 min, subsequently, samples were centrifuged for 10 min at 12,000 × g at 4 °C. The supernatant was then discarded and the RNA pellet was washed twice with 1 ml of 75% ethanol following a 7500 × g centrifuge of 5 min. Ethanol was vacuum dried for 5 min and RNA was resuspended in 20 μl of molecular-grade water. DNA leftovers were digested using DNase I (RNase-free, NEB) and cleaned using Monarch RNA Cleanup Kit (NEB). RNA quality was measured using RNA ScreenTape (Agilent). cDNA reaction was done using qScript Flex cDNA Synthesis Kit (Quantabio) according to the manufacturers’ instructions. RT-qPCR was performed on both RNA and cDNA to evaluate DNA contamination of the samples. Then, RT-qPCR was performed on cDNA samples (10 ng DNA per reaction) using primers for TNF-α, IL-6 and ZO-1. The reference gene used was Rplp0. All primers used are listed in Supplementary Data 2.

ELISA assay

IL-10 and IL-17 concentrations in splenocytes media were measured using Mouse IL-10 and IL-17 ELISA MAX^TM Standard Kit (BioLegend) following the manufacturer’s instructions. ELISA limits of detection: IL-10—31.5–2000 pg/ml, IL-17—15.6–1000 pg/ml.

Analysis of B. theta DNA inversion regions in IBDMDB

Data processing

PhaseFinder³¹ (v1.0) was used to identify invertible regions in metagenomics samples of the IBDMDB cohort, of 132 subjects, as previously described²². In brief, putative inversion regions were detected by identifying inverted repeats in representative Bacteriodales genomes as described in Supplementary Data 1 including B. theta VPI-5482 (NC_004663.1) using the ‘locate’ function. A database containing the inversion regions forward and reverse orientations was created using the tool’s ‘create’ function. Using the tool’s ‘ratio’ function, metagenomic samples from the publicly available IBDMDB cohort were then aligned to the database, resulting in the ratio of reverse-oriented reads out of all reads assigned to each region. Results were filtered for identified regions with >20 reads supporting either the forward or reverse orientations combined from the paired-end method, mean inversion frequency >0.005 across all samples. Dietary questionnaires available in the IBDMDB metadata file were used to assess food consumption (1-5, where 5 is the highest consumption in the previous week).

Mixed-effect linear regression analysis

To examine the impact of food components on B. theta DNA inversion ratios, we performed mixed-effect linear regression analysis. Food components were used as independent variables, with random intercepts accounting for variation within each patient (ID) nested in the health status (Status) group. The analysis was conducted using the lme4 package (version 1.1.35.4) in R. Visualization was performed using the packages ggplot2 (version 3.4.3) and ggpubr (version 0.6.0).

Clustering and statistical testing

Significant dietary components identified from the regression analysis were used to cluster samples via Partitioning Around Medoids (PAM) with k = 3 and the “manhattan” metric, defining distinct dietary patterns (‘Clusters’). Differences in DNA inversion ratios across clusters were assessed using the Kruskal-Wallis test, with pairwise comparisons conducted via the Conover-Iman test. Significant food components were compared between clusters using the Wilcoxon rank-sum test. After subsetting data by food components and PVRs, FDR correction (Benjamini–Hochberg method, p < 0.05) was applied. Clustering results were visualized using factoextra (version 1.0.7) and ggplot2 (version 3.4.3).

Linear regression and heatmap

Linear regression analysis was performed to evaluate independent effects of each dietary component on PVR orientation. Significant associations (Benjamini-Hochberg-based FDR-adjusted p < 0.05) were evaluated using t-tests provided by the lmerTest package (version 3.1.3), where the null hypothesis for each diet was that its estimated coefficient equals zero (no effect). All associations were visualized in a heatmap created using pheatmap (version 1.0.12) with significant associations marked in red. PVRs genomic context was visualized using gggenes (version 0.5.1).

Analysis of B. theta DNA inversion regions from Biolog plates culture

DNA was extracted from B. theta ATCC29148 using ZymoBIOMICS DNA Miniprep kit (Zymo, #D4300). In samples were only PVR2 region was of interest, PCR reaction was conducted to amplify this specific region (NEBNext LongAmp Hot Start Taq 2X Master Mix #M0533L, all primers used are listed in Supplementary Data 2.). DNA concentrations were measured using Quant-iT™ dsDNA Assay Kit, high sensitivity (Thermo Fisher Scientific, #Q33120). One hundred nanograms of each sample were FFPE repaired (NEBNext FFPE DNA repair v2 module, #E7360) and end-prepped (NEBNext Ultra II End repair/dA-Tailing module, #E7546) according to the manufacturer protocol. Each end-prepped sample was barcoded and adapters were added, using the Native barcoding kit V14 (Nanoporetech, #SQK-NBD114.96) according to the manufacturer protocol. The pooled library was loaded on R10.4.1 flow cell (Nanoporetech, #FLO-MIN114) and ran for 72 h.

DNA inversion regions were identified in the Nanopore whole genome sequencing results using PhaVa⁴⁵ (v0.2.3). Putative inversion regions were detected by identifying inverted repeats in B. theta VPI-5482 (NC_004663.1) genome using the ‘locate’ function. A database containing the inversion regions forward and reverse orientations was created using the tool’s ‘create’ function. Using the tool’s ‘ratio’ function, sequencing reads were aligned to the database, resulting in the ratio of reverse oriented reads out of all reads assigned to each region in each sample. We filtered the results by removing identified regions with <10 reads supporting either the forward or reverse orientations. Genomic regions were visualized using the R package gggenes (version 0.5.1). “ON”/”OFF” assignment was decided according to the Bacteroides conserved promoter motif (TTG-AT-rich region (19-21 bp)-TANNTTTG)³¹.

Proteolysis

Cells were lyzed in 8.5 M Urea, 400 mM ammonium bicarbonate and 10 mM DTT, sonicated twice (90%, 10-10, 5 min), and centrifuged (10,000 × g, 10 min). Protein amount was estimated using Bradford readings. The samples from the conditioned media were denatured by 8.5 M Urea, 400 mM ammonium bicarbonate and 10 mM DTT. The samples were reduced (60 °C for 30 min), modified with 35.2 mM iodoacetamide in 100 mM ammonium bicarbonate (room temperature for 30 min in the dark) and digested in 1.5 M Urea, 66 mM ammonium bicarbonate with modified trypsin (Promega), overnight at 37 °C in a 1:50 (M/M) enzyme-to-substrate ratio. An additional second digestion with Trypsin was done for 4 h at 37 °C in a 1:100 (M/M) enzyme-to-substrate ratio. The tryptic peptides were desalted using homemade C18 stage tip, dried and re-suspended in 0.1% Formic acid.

Mass spectrometry analysis

Two micrograms of the resulted peptides from the cell pellets were analyzed by LC-MS/MS using an Exploris 480 mass spectrometer (Thermo) fitted with a capillary HPLC (Easy nLC 1200, Thermo-Fisher) and 50% of all peptides from the conditioned media were analyzed by Q Exactive plus mass spectrometer (Thermo). The peptides were loaded in solvent A (0.1% formic acid in water) on a homemade capillary column (30 cm, 75-micron ID) packed with Reprosil C18-Aqua (Dr. Maisch GmbH, Germany). The peptides mixture was resolved with a 6–30% linear gradient of solvent B (80% acetonitrile with 0.1% formic acid) for 120 min (for the cell pellets samples) or 60 min (for the conditioned media samples) followed by gradient of 15 min of 30–80% and 15 min at 80% acetonitrile with 0.1% formic acid in water at flow rates of 0.15 μl/min. Mass spectrometry was performed in a positive mode (m/z 300–1500). For the cells’ samples, a repetitively full MS scan was used followed by high collision dissociation (HCD, at 27 normalized collision energy) of the 30 most dominant ions (>1 charges) selected from the first MS scan. The resolution was 120,000 for MS1 and 15,000 for MS2 and the intensity threshold for triggering MS/MS analysis was 1 × 104. A dynamic exclusion list was enabled with an exclusion duration of 30 s. For the CM samples, a repetitively full MS scan was used followed by collision induced dissociation (HCD, at 25 normalized collision energy) of the 10 most dominant ions (>1 charges) selected from the first MS scan. The resolution was 70,000 for MS1 and 17,5000 for MS2. A dynamic exclusion list was enabled with an exclusion duration of 20 s. The mass spectrometry data was analyzed using the MaxQuant software version 2.1.1⁶⁵ for peak picking and identification using the Andromeda search engine, searching against the Bacteroides thetaiotaomicron proteome from the Uniprot database with mass tolerance of 6 ppm for the precursor masses and 20 ppm for the fragment ions. Oxidation on methionine and protein N-terminus acetylation were accepted as variable modifications and carbamidomethyl on cysteine was accepted as static modifications. Minimal peptide length was set to seven amino acids and a maximum of two miscleavages was allowed. The data was quantified by label free analysis using the same software. Peptide- and protein-level false discovery rates (FDRs) were filtered to 1% using the target-decoy strategy. Protein table was filtered to eliminate the identifications from the reverse database and common contaminants.

Extraction of SCFAs

Conditioned media samples: B. theta was grown in each of the carbon sources in three independent biological repeats. SCFAs were extracted from the media by diluting 50 µl of media with 50 µl of DW acidified to pH 2–3 using phosphoric acid (50% v/v) containing 4-methyl valeric acid as internal standard. The samples were vortexed for 10 min. 160 µl Ethyl acetate were added and the samples were vortexed again for 10 min, followed by centrifugation at max speed for 10 min. The caps of the tubes were opened and the phases were left to fully separate for 5 min. The upper organic layer was transferred to GC/HPLC glass vial and kept at −80 °C until GCMS analysis.

Stool samples: Each wet stool sample was weighed and its weight was recorded for normalization. The stool was transferred into homogenization tube prefilled with 500ul of phosphoric acid (50%) and CK14 ceramic beads (Bertin Corp, P000926-LYSK0-A). Samples were homogenized using Precellys 24 tissue homogenizer (Bertin Corp, P002391-P24T0-A.0). Homogenization conditions were set to three cycles, 30 s each, 6500 rpm with a 30 s gap between each of the cycles. 800 µl Ethyl acetate were added and the samples were vortexed for 10 min, followed by centrifugation at max speed for 10 min. The caps of the tubes were opened and the phases were left to fully separate for 5 min. The upper organic layer was transferred to GC/HPLC glass vial and kept at −80 °C until GCMS analysis.

GC–MS analysis of SCFAs

The organic phase was injected into a gas chromatograph instrument (Agilent 7890A, Agilent Technologies, CA, USA), equipped with TG-WaxMS A column (30 m length, 0.25 mm diameter, 0.25 μm film thickness, Thermo Fisher Scientific) coupled to a mass spectrometer (Agilent 7010B, Agilent Technologies, CA, USA). The injection was made in a split mode (5:1) with an injection volume of 1 μl. The injector temperature was set to 250 °C. Helium was used as a carrier gas at a constant flow rate of 1 ml/min. The initial oven temperature was held at 70 °C for 0.75 min, then increased to 160 °C at a rate of 5 °C/min, to 230 °C at a rate of 20 °C/min, and finally held at this temperature and flow rate for 5 min (total time of 27.25 min). The solvent delay was set to 6 min. The SCFAs were detected using a scan mode, in the range of 30–150 m/z. The MS source temperature was set to 230 °C. SCFAs were detected based on retention time and mass spectra generated in house from commercial standards. SCFA were quantified using a standard calibration curve and expressed as mM in the media samples and µmol/g in the stool samples. Heptanoic acid was used as an internal standard, to correct SCFAs concentration. Data integration was performed using the Mass Hunter Quantitative Analysis for QQQ V12.1. Following every ten samples injected, a pool sample from the biological batch was injected for quality control.

Statistical analysis

Principal components analysis of immune parameters and phase variable regions was performed in R using the ‘prcomp’ function (package ‘stats’ version 4.3.1). Prior to analysis, the data was scaled and centered. Loadings were extracted and plotted to understand the relationship between variables and principal components (Supplementary Fig. 5).

Heatmaps were generated for proteomics data using the ‘pheatmap’ package in R (version 1.0.12). The data was filtered to include only values with an adjusted p-value below 0.01 (ANOVA test with permutation-based FDR and 250 randomizations). Hierarchical clustering of both rows and columns was performed using the average linkage method based on Pearson correlations. Rows were split into clusters based on this hierarchical clustering. All other statistical analyses were done using Prism-GraphPad.

Reporting summary

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

Data availability

The Sequences data that supported the findings of this study have been deposited in the Sequence Read Archive (SRA) with the dataset identifier PRJNA1218017. (https://www.ncbi.nlm.nih.gov/bioproject/1218017). Mass spectrometry proteomics data that support the findings of this study have been deposited to the ProteomeXchange Consortium via the PRIDE⁶⁶ partner repository with the dataset identifier PXD060215. (https://protect.checkpoint.com/v2/r02/___https://www.ebi.ac.uk/pride/archive/projects/PXD060215___.YzJlOnRlY2huaW9uOmM6bzo0YWM2ZjFiZjc4NzI3OTVjMzk2NzE2Yjc0MDBjZTdkNDo3OjZhNWE6ZTNhN2Y4Y2M4NjcyNzFlYmYyOTA4NDFiM2ZiYWRiNjUyN2I3MGM4ZDljMTZkNTdlZjJmMGEwMTA2MDE5MGU1Yzp0OlQ6Tg). Mass spectrometry metabolomics data of the study have been deposited to Mass spectrometry Interactive Virtual Enviroment (MassIVE) repository with the dataset identifier MSV000097823. (https://doi.org/10.25345/C5BZ61M8W). Source data are provided with this paper.

References

Atarashi, K. et al. Treg induction by a rationally selected mixture of Clostridia strains from the human microbiota. Nature 500, 232–236 (2013).
Article ADS CAS PubMed Google Scholar
Geva-Zatorsky, N. et al. Mining the human gut microbiota for immunomodulatory organisms. Cell 168, 928–943 e911 (2017).
Article CAS PubMed PubMed Central Google Scholar
Hooper, L. V., Littman, D. R. & Macpherson, A. J. Interactions between the microbiota and the immune system. Science 336, 1268–1273 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Surana, N. K. & Kasper, D. L. Erratum: Moving beyond microbiome-wide associations to causal microbe identification. Nature 554, 392 (2018).
Article ADS CAS PubMed Google Scholar
Sela, U., Euler, C. W., Correa da Rosa, J. & Fischetti, V. A. Strains of bacterial species induce a greatly varied acute adaptive immune response: The contribution of the accessory genome. PLoS Pathog. 14, e1006726 (2018).
Article PubMed PubMed Central Google Scholar
Hajjo, H. & Geva-Zatorsky, N. Strain-level immunomodulatory variation of gut bacteria. FEBS Lett. 595, 1322–1327 (2021).
Article CAS PubMed Google Scholar
Ben-Assa, N. et al. Analysis of a phase-variable restriction modification system of the human gut symbiont Bacteroides fragilis. Nucleic Acids Res. 48, 11040–11053 (2020).
Article CAS PubMed PubMed Central Google Scholar
Zmora, N., Suez, J. & Elinav, E. You are what you eat: diet, health and the gut microbiota. Nat. Rev. Gastroenterol. Hepatol. 16, 35–56 (2019).
Article CAS PubMed Google Scholar
Ross, F. C. et al. The interplay between diet and the gut microbiome: implications for health and disease. Nat. Rev. Microbiol. 22, 671–686 (2024).
Article CAS PubMed Google Scholar
Pudlo, N. A. et al. Phenotypic and genomic diversification in complex carbohydrate-degrading human gut bacteria. mSystems 7, e0094721 (2022).
Article PubMed Google Scholar
Bjursell, M. K., Martens, E. C. & Gordon, J. I. Functional genomic and metabolic studies of the adaptations of a prominent adult human gut symbiont, Bacteroides thetaiotaomicron, to the suckling period. J. Biol. Chem. 281, 36269–36279 (2006).
Article CAS PubMed Google Scholar
Hooper, L. V., Wong, M. H., Thelin, A., Hansson, L., Falk, P. G. & Gordon, J. I. Molecular analysis of commensal host-microbial relationships in the intestine. Science 291, 881–884 (2001).
Article ADS CAS PubMed Google Scholar
Wrzosek, L. et al. Bacteroides thetaiotaomicron and Faecalibacterium prausnitzii influence the production of mucus glycans and the development of goblet cells in the colonic epithelium of a gnotobiotic model rodent. BMC Biol. 11, 61 (2013).
Article PubMed PubMed Central Google Scholar
Kelly, D. et al. Commensal anaerobic gut bacteria attenuate inflammation by regulating nuclear-cytoplasmic shuttling of PPAR-gamma and RelA. Nat. Immunol. 5, 104–112 (2004).
Article CAS PubMed Google Scholar
Charlet, R., Bortolus, C., Sendid, B. & Jawhara, S. Bacteroides thetaiotaomicron and Lactobacillus johnsonii modulate intestinal inflammation and eliminate fungi via enzymatic hydrolysis of the fungal cell wall. Sci. Rep. 10, 11510 (2020).
Article CAS PubMed PubMed Central Google Scholar
Xu, J. et al. A genomic view of the human-Bacteroides thetaiotaomicron symbiosis. Science 299, 2074–2076 (2003).
Article ADS CAS PubMed Google Scholar
Martens, E. C., Chiang, H. C. & Gordon, J. I. Mucosal glycan foraging enhances fitness and transmission of a saccharolytic human gut bacterial symbiont. Cell Host Microbe 4, 447–457 (2008).
Article CAS PubMed PubMed Central Google Scholar
Liu, H. et al. Functional genetics of human gut commensal Bacteroides thetaiotaomicron reveals metabolic requirements for growth across environments. Cell Rep. 34, 108789 (2021).
Article CAS PubMed PubMed Central Google Scholar
Krinos, C. M., Coyne, M. J., Weinacht, K. G., Tzianabos, A. O., Kasper, D. L. & Comstock, L. E. Extensive surface diversity of a commensal microorganism by multiple DNA inversions. Nature 414, 555–558 (2001).
Article ADS CAS PubMed Google Scholar
Porter, N. T., Canales, P., Peterson, D. A. & Martens, E. C. A subset of polysaccharide capsules in the human symbiont bacteroides thetaiotaomicron promote increased competitive fitness in the mouse gut. Cell Host Microbe 22, 494–506 e498 (2017).
Article CAS PubMed PubMed Central Google Scholar
Dapa, T., Ramiro, R. S., Pedro, M. F., Gordo, I. & Xavier, K. B. Diet leaves a genetic signature in a keystone member of the gut microbiota. Cell Host Microbe 30, 183–199 e110 (2022).
Article CAS PubMed Google Scholar
Carasso, S. et al. Inflammation and bacteriophages affect DNA inversion states and functionality of the gut microbiota. Cell Host Microbe 32, 322–334 e329 (2024).
Article CAS PubMed PubMed Central Google Scholar
Hsieh, S. et al. Polysaccharide capsules equip the human symbiont bacteroides thetaiotaomicron to modulate immune responses to a dominant antigen in the intestine. J. Immunol. 204, 1035–1046 (2020).
Article CAS PubMed Google Scholar
Cecil JD, Sirisaengtaksin N, O’Brien-Simpson NM, Krachler AM. Outer membrane vesicle-host cell interactions. Microbiol. Spectr. 7 https://doi.org/10.1128/microbiolspec.PSIB-0001-2018 (2019).
Stentz, R. et al. The proteome of extracellular vesicles produced by the human gut bacteria bacteroides thetaiotaomicron in vivo is influenced by environmental and host-derived factors. Appl. Environ. Microbiol. 88, e0053322 (2022).
Article PubMed Google Scholar
Shen, Y., Giardino Torchia, M. L., Lawson, G. W., Karp, C. L., Ashwell, J. D. & Mazmanian, S. K. Outer membrane vesicles of a human commensal mediate immune regulation and disease protection. Cell Host Microbe 12, 509–520 (2012).
Article CAS PubMed PubMed Central Google Scholar
Kaparakis-Liaskos, M. & Ferrero, R. L. Immune modulation by bacterial outer membrane vesicles. Nat. Rev. Immunol. 15, 375–387 (2015).
Article CAS PubMed Google Scholar
Durant, L. et al. Bacteroides thetaiotaomicron-derived outer membrane vesicles promote regulatory dendritic cell responses in health but not in inflammatory bowel disease. Microbiome 8, 88 (2020).
Article CAS PubMed PubMed Central Google Scholar
Wegorzewska MM, et al. Diet modulates colonic T cell responses by regulating the expression of a Bacteroides thetaiotaomicron antigen. Sci. Immunol 4, eaau9079 (2019).
Pedersen, T. K. et al. The CD4(+) T cell response to a commensal-derived epitope transitions from a tolerant to an inflammatory state in Crohn’s disease. Immunity 55, 1909–1923 e1906 (2022).
Article CAS PubMed PubMed Central Google Scholar
Jiang, X. et al. Invertible promoters mediate bacterial phase variation, antibiotic resistance, and host adaptation in the gut. Science 363, 181–187 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Ni, M. et al. Epigenetic phase variation in the gut microbiome enhances bacterial adaptation. Preprint at bioRxiv https://doi.org/10.1101/2025.01.11.632565 (2025).
Bergman, M., Del Prete, G., van Kooyk, Y. & Appelmelk, B. Helicobacter pylori phase variation, immune modulation and gastric autoimmunity. Nat. Rev. Microbiol 4, 151–159 (2006).
Article CAS PubMed Google Scholar
Mazmanian, S. K., Round, J. L. & Kasper, D. L. A microbial symbiosis factor prevents intestinal inflammatory disease. Nature 453, 620–625 (2008).
Article ADS CAS PubMed Google Scholar
Lloyd-Price, J. et al. Multi-omics of the gut microbial ecosystem in inflammatory bowel diseases. Nature 569, 655–662 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Atarashi, K. et al. ATP drives lamina propria T(H)17 cell differentiation. Nature 455, 808–812 (2008).
Article ADS CAS PubMed 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
Nahui Palomino, R. A., Vanpouille, C., Costantini, P. E. & Margolis, L. Microbiota-host communications: Bacterial extracellular vesicles as a common language. PLoS Pathog. 17, e1009508 (2021).
Article CAS PubMed PubMed Central Google Scholar
Okumura, R. & Takeda, K. Roles of intestinal epithelial cells in the maintenance of gut homeostasis. Exp. Mol. Med. 49, e338 (2017).
Article CAS PubMed PubMed Central Google Scholar
Sun, M. et al. Microbiota-derived short-chain fatty acids promote Th1 cell IL-10 production to maintain intestinal homeostasis. Nat. Commun. 9, 3555 (2018).
Article ADS PubMed PubMed Central Google Scholar
Li, M., van Esch, B., Wagenaar, G. T. M., Garssen, J., Folkerts, G. & Henricks, P. A. J. Pro- and anti-inflammatory effects of short chain fatty acids on immune and endothelial cells. Eur. J. Pharmacol. 831, 52–59 (2018).
Article CAS PubMed Google Scholar
Boob, M., Wang, Y. & Gruebele, M. Proteins: “Boil ‘Em, Mash ‘Em, Stick ‘Em in a Stew. J. Phys. Chem. B 123, 8341–8350 (2019).
Article CAS PubMed PubMed Central Google Scholar
Jin, B., Li, L., Liu, G. Q., Li, B., Zhu, Y. K. & Liao, L. N. Structural changes of malt proteins during boiling. Molecules 14, 1081–1097 (2009).
Article CAS PubMed PubMed Central Google Scholar
Deutsch, E. W. et al. The ProteomeXchange consortium at 10 years: 2023 update. Nucleic Acids Res 51, D1539–D1548 (2023).
Article PubMed Google Scholar
Chanin, R. B. et al. Intragenic DNA inversions expand bacterial coding capacity. Nature 634, 234–242 (2024).
Duncan, S. H., Belenguer, A., Holtrop, G., Johnstone, A. M., Flint, H. J. & Lobley, G. E. Reduced dietary intake of carbohydrates by obese subjects results in decreased concentrations of butyrate and butyrate-producing bacteria in feces. Appl Environ. Microbiol. 73, 1073–1078 (2007).
Article ADS CAS PubMed Google Scholar
Wastyk, H. C. et al. Gut-microbiota-targeted diets modulate human immune status. Cell 184, 4137–4153 e4114 (2021).
Article CAS PubMed PubMed Central Google Scholar
Beam A, Clinger E, Hao L. Effect of diet and dietary components on the composition of the gut microbiota. Nutrients 13, 2795 (2021).
Mohr, A. E. et al. Gut microbiome remodeling and metabolomic profile improves in response to protein pacing with intermittent fasting versus continuous caloric restriction. Nat. Commun. 15, 4155 (2024).
Article ADS CAS PubMed PubMed Central Google Scholar
Asnicar, F. et al. Microbiome connections with host metabolism and habitual diet from 1,098 deeply phenotyped individuals. Nat. Med. 27, 321–332 (2021).
Article CAS PubMed PubMed Central Google Scholar
Sinha, A. K. et al. Dietary fibre directs microbial tryptophan metabolism via metabolic interactions in the gut microbiota. Nat. Microbiol. 9, 1964–1978 (2024).
Article CAS PubMed PubMed Central Google Scholar
Lobel, L., Cao, Y. G., Fenn, K., Glickman, J. N. & Garrett, W. S. Diet posttranslationally modifies the mouse gut microbial proteome to modulate renal function. Science 369, 1518–1524 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Troy, E. B., & Kasper, D. L. Beneficial effects of Bacteroides fragilis polysaccharides on the immune system. Front. Biosci. 15, 25–34 (2010).
Article CAS PubMed Central Google Scholar
Mazmanian, S. K., Liu, C. H., Tzianabos, A. O. & Kasper, D. L. An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system. Cell 122, 107–118 (2005).
Article CAS PubMed Google Scholar
Delday, M., Mulder, I., Logan, E. T. & Grant, G. Bacteroides thetaiotaomicron ameliorates colon inflammation in preclinical models of Crohn’s disease. Inflamm. Bowel Dis. 25, 85–96 (2019).
Article PubMed Google Scholar
Tripathi, A. et al. Identification of human zonulin, a physiological modulator of tight junctions, as prehaptoglobin-2. Proc. Natl. Acad. Sci. USA 106, 16799–16804 (2009).
Article ADS CAS PubMed PubMed Central Google Scholar
Kuo, W. T. et al. The tight junction protein ZO-1 is dispensable for barrier function but critical for effective mucosal repair. Gastroenterology 161, 1924–1939 (2021).
Article CAS PubMed Google Scholar
Trompette, A. et al. Dietary fiber confers protection against Flu by shaping Ly6c(-) patrolling monocyte hematopoiesis and CD8(+) T cell metabolism. Immunity 48, 992–1005 e1008 (2018).
Article CAS PubMed Google Scholar
Hickey, C. A. et al. Colitogenic Bacteroides thetaiotaomicron antigens access host immune cells in a sulfatase-dependent manner via outer membrane vesicles. Cell Host Microbe 17, 672–680 (2015).
Article CAS PubMed PubMed Central Google Scholar
Trzilova, D. & Tamayo, R. Site-specific recombination - how simple DNA inversions produce complex phenotypic heterogeneity in bacterial populations. Trends Genet. 37, 59–72 (2021).
Article CAS PubMed Google Scholar
Bergstrom, K. S. & Xia, L. Mucin-type O-glycans and their roles in intestinal homeostasis. Glycobiology 23, 1026–1037 (2013).
Article CAS PubMed PubMed Central Google Scholar
Henke, M. T., Kenny, D. J., Cassilly, C. D., Vlamakis, H., Xavier, R. J. & Clardy, J. Ruminococcus gnavus, a member of the human gut microbiome associated with Crohn’s disease, produces an inflammatory polysaccharide. Proc. Natl. Acad. Sci. USA 116, 12672–12677 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Comstock, L. E. & Coyne, M. J. Bacteroides thetaiotaomicron: a dynamic, niche-adapted human symbiont. Bioessays 25, 926–929 (2003).
Article CAS PubMed Google Scholar
Di Rienzi, S. C. & Britton, R. A. Adaptation of the gut microbiota to modern dietary sugars and sweeteners. Adv. Nutr. 11, 616–629 (2020).
Article PubMed Google Scholar
Cox, J., Hein, M. Y., Luber, C. A., Paron, I., Nagaraj, N. & Mann, M. Accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ. Mol. Cell Proteom. 13, 2513–2526 (2014).
Article CAS Google Scholar
Perez-Riverol, Y. et al. The PRIDE database at 20 years: 2025 update. Nucleic Acids Res. 53, D543–D553 (2025).
Article PubMed Google Scholar

Download references

Acknowledgements

We would like to thank the Geva-Zatorsky lab members for fruitful discussions and contributions. Special thanks to Rachel Herren and Melody Kasher for proofreading and to Tomm Blum for the illustration of Fig. 8, all from the Geva-Zatorsky lab. We also thank Tomer Mamman and Aviv Sasson from the Geva-Zatorsky lab for assisting with the experiments. We appreciate and thank Profs. Laurie Comstock, Emma Slack and the HMB CIFAR group for scientific discussions and insights. We thank Drs. Amir Grau, Ofer Shenker, and the biomedical core facility at the Technion Rappaport Faculty of Medicine for their help with flow cytometry; the Smoler Proteomics Center at the Technion for help with proteomics; Ifat Gavish-Abramovich for the metabolomics; Noy Ifrah and The Pre-Clinical Research Authority for the animal care. This work was supported by the Technion Institute of Technology, “Keren haNasi,” Cathedra, the Rappaport Technion Integrated Cancer Center, the Alon Fellowship for Outstanding Young Researchers, the Israeli Science Foundation (3165/20, N.G.-Z.), the D. Dan and Betty Kahn Foundation’s gift to the University of Michigan, the Weizmann Institute, the Technion–Israel Institute of Technology Collaboration for Research, the Seerave Foundation, CIFAR (grant FL-000969/FL-001245/FL-001381, N.G.-Z.), and the European Union (ERC, ExtractABact, 101078712, N.G.-Z.). Views and opinions expressed are, however, those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency. Neither the European Union nor the granting authority can be held responsible for them. N.G.-Z. is a CIFAR fellow in the Humans & the Microbiome program, a Kavli fellow, and a Horev Fellow (Taub Foundation). N.G.M is supported by the RTICC- Rubinstein fellowship and the Jacob’s excellence scholarships, S.C. is supported by the Gutwirth excellence scholarship.

Author information

Authors and Affiliations

Department of Cell Biology and Cancer Science, Rappaport Faculty of Medicine and Research Institute, Rappaport Technion Integrated Cancer Center (RTICC), Technion-Israel Institute of Technology, Haifa, Israel
Noa Gal-Mandelbaum, Shaqed Carasso, Alon Kedem, Roni Keshet-David, Roberto Abboud, Rawan Zaatry, Tal Gefen & Naama Geva-Zatorsky
Smoler Proteomics Center, Lokey Interdisciplinary Center for Life Sciences & Engineering, Technion-Israel Institute of Technology, Haifa, Israel
Tamar Ziv
CIFAR Humans & the Microbiome program, CIFAR, Toronto, Canada
Naama Geva-Zatorsky

Authors

Noa Gal-Mandelbaum
View author publications
Search author on:PubMed Google Scholar
Shaqed Carasso
View author publications
Search author on:PubMed Google Scholar
Alon Kedem
View author publications
Search author on:PubMed Google Scholar
Tamar Ziv
View author publications
Search author on:PubMed Google Scholar
Roni Keshet-David
View author publications
Search author on:PubMed Google Scholar
Roberto Abboud
View author publications
Search author on:PubMed Google Scholar
Rawan Zaatry
View author publications
Search author on:PubMed Google Scholar
Tal Gefen
View author publications
Search author on:PubMed Google Scholar
Naama Geva-Zatorsky
View author publications
Search author on:PubMed Google Scholar

Contributions

Study conceptualization: N.G.-M., T.G., and N.G.-Z., all experiments and analyses were performed by N.G.-M. and T.G. T.Z. performed all the proteomics analyses. S.C., A.K. and R.K.-D. performed the computational analyses. R.A. and R.Z. helped with the experiments. N.G.-M., T.G., and N.G.-Z. wrote the manuscript. All authors contributed to the article and approved the submitted version. N.G.-Z. conceived and planned the study, supervised it, interpreted the experiments, and wrote the manuscript.

Corresponding author

Correspondence to Naama Geva-Zatorsky.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

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

Supplementary information

Supplementary Information

Description of Additional Supplementary Files

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

Supplementary Data 4

Supplementary Data 5

Supplementary Data 6

Reporting Summary

Transparent Peer Review file

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, 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 licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence 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 licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Gal-Mandelbaum, N., Carasso, S., Kedem, A. et al. Dietary carbohydrates alter immune-modulatory functionalities and DNA inversions in Bacteroides thetaiotaomicron. Nat Commun 16, 4938 (2025). https://doi.org/10.1038/s41467-025-60202-9

Download citation

Received: 15 August 2024
Accepted: 15 May 2025
Published: 28 May 2025
DOI: https://doi.org/10.1038/s41467-025-60202-9