Age‐mediated gut microbiota dysbiosis promotes the loss of dendritic cells tolerance

Abstract The old age‐related loss of immune tolerance inflicts a person with a wide range of autoimmune and inflammatory diseases. Dendritic cells (DCs) are the sentinels of the immune system that maintain immune tolerance through cytokines and regulatory T‐cells generation. Aging disturbs the microbial composition of the gut, causing immune system dysregulation. However, the vis‐à‐vis role of gut dysbiosis on DCs tolerance remains highly elusive. Consequently, we studied the influence of aging on gut dysbiosis and its impact on the loss of DC tolerance. We show that DCs generated from either the aged (DCOld) or gut‐dysbiotic young (DCDysbiotic) but not young (DCYoung) mice exhibited loss of tolerance, as evidenced by their failure to optimally induce the generation of Tregs and control the overactivation of CD4+ T cells. The mechanism deciphered for the loss of DCOld and DCDysbiotic tolerance was chiefly through the overactivation of NF‐κB, impaired frequency of Tregs, upregulation in the level of pro‐inflammatory molecules (IL‐6, IL‐1β, TNF‐α, IL‐12, IFN‐γ), and decline in the anti‐inflammatory moieties (IL‐10, TGF‐β, IL‐4, IDO, arginase, NO, IRF‐4, IRF‐8, PDL1, BTLA4, ALDH2). Importantly, a significant decline in the frequency of the Lactobacillus genus was noticed in the gut. Replenishing the gut of old mice with the Lactobacillus plantarum reinvigorated the tolerogenic function of DCs through the rewiring of inflammatory and metabolic pathways. Thus, for the first time, we demonstrate the impact of age‐related gut dysbiosis on the loss of DC tolerance. This finding may open avenues for therapeutic intervention for treating age‐associated disorders with the Lactobacillus plantarum.


| INTRODUC TI ON
Age-inflicted inflamed microenvironment favors inflammatory and autoimmune responses with a concurrent decline in the protective immunity (Kim et al., 2017). Both Innate and adaptive arms of the immune system show age-related changes in their functional capacity, manifested by diminished antigen uptake and presentation capacity, phagocytic activity, thymic involution, antibody production, and reduced response to vaccination and infection. A gain of non-specific innate immunity with a loss of adaptive immunity is linked with the advancement of age (Lee et al., 2022).
Recognition of pathogen's danger signals, uptake of antigens, and their processing provides developmental cues to DCs for their maturation and activation. DCs are the only APCs with the capacity to activate and differentiate naive T cells (Trombetta & Mellman, 2005). Immunogenic DCs express a higher level of MHCII and co-stimulatory molecules CD40, CD80 and CD86, and release an elevated amount of pro-inflammatory cytokines like IL-12, IL-6, IL1β, and TNFα (Hackstein & Thomson, 2004). In contrast, tolerogenic DCs display comparatively lower levels of MHCII and costimulatory molecules and higher expression of inhibitory receptors such as Tim-3 and PDL-1 (Iberg & Hawiger, 2020). Tolerogenic DCs produce augmented quantities of anti-inflammatory cytokines like IL-10 and TGFβ (Vogel et al., 2022) and reduced production of IL-12 (Steinman et al., 2003). DCs maintain immunological tolerance by clonal deletion of T cells, induction of anergy, and generation of Tregs (Horton et al., 2017). In addition, they maintain peripheral tolerance against self-antigens and presentation to autoreactive T cells (Hawiger et al., 2001). This heterogeneity in the functionality of DCs establishes a fine balance between the activation and suppression of the immune system. The loss of DC tolerance is a gradual process connected with aging, as observed by the increased production of pro-inflammatory cytokines, reduced phagocytic capacity (Agrawal et al., 2017), deficient Tregs-inducing capacity, and failure to curb autoimmune and inflammatory responses. Although enhanced activation of NF-κB and pro-inflammatory responses are considered responsible for this phenomenon (Agrawal et al., 2007), the mechanism connected with age-associated DCs dysfunction remains extensively elusive.
Recently, the role of gut microbiota is increasingly being recognized in the development, maturation, and maintenance of homeostasis of the immune system through elegant experiments conducted on germ-free mice models (Honda & Littman, 2016). Gut microbiota induces peripheral tolerance through the induction of Tregs, IgAsecreting B cells, Th17 cells, and through the DC modulation (Zheng, Liwinski, & Elinav, 2020). An alteration in gut microbiota composition and function has been reported with age, and concomitantly, it is associated with various autoimmune diseases (Bosco & Noti, 2021).
Gut dysbiosis with age results in a loss of mainly Firmicutes and Bacteroides and the predominance of Proteobacteria and therefore predisposes to increased risk of immune dysregulation and persistence of chronic inflammation (Ragonnaud & Biragyn, 2021).
However, it is still largely unknown how aging provokes gut dysbiosis and loss of DC tolerance.
Taking into consideration the aforesaid facts, we studied the influence of aging on gut microbiota and its implication on the activation, differentiation, and function of DCs. Through comparison of DCs derived from young (DC Young ), old (DC Old ), and antibiotic-treated young animals (DC Dysbiotic ). We show the correlation between gut dysbiosis and loss of DC tolerance. Our study revealed that aging incited gut disruptions linked with the failure in the expansion of Tregs and downregulation of tolerance-associated gene network and signaling pathways that resulted in the loss of DC tolerance.
Importantly, the loss of DC tolerance was connected with the disappearance of the beneficial bacteria Lactobacillus. Interestingly, replenishing the gut of aged mice with Lactobacillus plantarum, restored the age-associated loss of tolerance in DCs (DC Old-LP ). The study suggests the therapeutic role of Lactobacillus plantarum in the maintenance of DC tolerance and its use as a remedial measure for alleviating age-associated immune system defects.

| Bone marrow-derived hematopoietic cells exhibit declined differentiation potential with age and gut dysbiosis to DCs
Aged hematopoietic stem cells (HSCs) generate a dysfunctional immune system that contributes to immunosenescence (Geiger et al., 2013). Additionally, dysbiosis of the microbial population with antibiotic treatment adversely affects the bone marrow cells F I G U R E 1 Aging and Abx treatment of young animals disrupts gut microbiota and impairs the differentiation of BMCs to DCs. Young control, young dysbiotic (21 days of antibiotics treated) and old, groups of animals were compared; (a) quantification of facultative anaerobes and aerobes bacteria from fecal matter; (b) quantification of total BMCs; (c) comparative analysis of the total number of BMCs and DCs differentiated from BMCs; (d) phenotypic (CD11c + CD80 + population (top), CD11c + CD86 + (middle) and CD11c + MHCII + (bottom) analysis of differentiated (-LPS) DC Young , DC dysbiotic and DC Old ; (e) representative FACS plot and bar graph of syngeneic and allogeneic CD4 + T-cell proliferation induced by DC Young , DC Dysbiotic and DC Old ; (f) representative FACS plot and bar graph of CD4 + FoxP3 + Tregs differentiation by DC Young , DC Dysbiotic and DC Old . The data (mean ± SD) are from three independent experiments, with each point representing a pool of three animals for one independent experiment (n = 3 mice/group). Statistical analysis was done by one-way ANOVA followed by Tukey's multiple comparison tests except (c, d), where two-way ANOVA followed by Sidak's multiple comparisons test was done. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.    (Josefsdottir et al., 2017). These observations led us to evaluate the number and differentiation potential of HSCs in aged (20-22 months) and young dysbiotic mice (young mice treated with an antibiotic cocktail for 21 days) compared to young mice (2-4 months). Gut bacterial colony-forming units (CFUs) were enumerated and compared in all three groups under anaerobic and aerobic conditions from their fecal homogenate. We observed a significant decrease in the gut bacterial load in both facultative anaerobes and aerobes in old and young dysbiotic mice in comparison with young mice (Figure 1a).
Next, an absolute number of bone marrow cells (BMCs) and their differentiation potential towards dendritic cells were compared.
No significant alteration was observed in the absolute cell counts among the three groups ( Figure 1b) but a significant decrease in the differentiation of BMCs to DCs in both old and young dysbiotic groups ( Figure 1c) compared to the young group was noticed. We conclude that gut dysbiosis, either due to old age or antibiotic treatment, impairs the differentiation potential of bone marrow/myeloid progenitor cells.

| Age and gut dysbiosis affects the maturation and function of DCs
DCs play an important role in both, orchestrating antigen-specific T-cell responses and maintenance of the peripheral tolerance (Coquerelle & Moser, 2010). However, their functional capacity gets curtailed with chronological aging (Agrawal et al., 2007), and with gut microbiota modulation (Uribe-Herranz et al., 2020). The phenotype of the DCs can ascertain their activation or tolerization function. A significant upregulation in the expression of co-stimulatory molecules CD80, CD86, and MHCII molecules (Figure 1d), and CD40 ( Figure S1c) molecules was observed on the DC Dysbiotic and DC Old , as compared to DC Young upon differentiation.
One of the mechanisms for the induction of peripheral Tcell tolerance by dendritic cells is attributed to their capacity to phagocytose foreign pathogens, cancer cells, and self-apoptotic cells. A defect in this capacity leads to a breach of tolerance (Savill et al., 2002). Given that we observed the mature phenotype of DCs in our experimental setup, we checked their ability for phagocytosis.
We observed a significant decline in phagocytosis of the antigen (Dextran-FITC) by both DC Dysbiotic and DC Old compared to DC Young ( Figure S1d). Additionally, we noticed a notable decrease in the engulfment of apoptotic bodies ( Figure S1e). These results indicate reduced phagocytosis by DCs on aging and gut dysbiosis.
DCs induce immunogenic or tolerogenic T-cell stimulation based on their maturation state (de Heusch et al., 2004). We next examined the effect of gut dysbiosis on the ability of DCs to activate CD4 + T cells. DC Dysbiotic and DC Old induced rigorous proliferation of syngeneic CD4 + T cells in comparison with DC Young .This was further authenticated using allogeneic CD4 + T cells ( Figure 1e). These data suggest that DC Dysbiotic and DC Old exhibit hyperactivation of CD4 + T cells. Further to confirm this, we examined the Tregs-inducing potential of all three DCs, as the regulatory function of DCs is well recognized for the maintenance of central and peripheral tolerance by driving naive T cells to differentiate towards Tregs (Raker et al., 2015). A significant loss of the Tregs-inducing potential of DC Dysbiotic and DC Old was noted ( Figure 1f). These results indicate loss of tolerogenic and acquisition of immunogenic properties in old and young dysbiotic DCs, suggesting a substantial role of gut dysbiosis in the modulation of DCs function. Next, we focused our studies on understanding the possible mechanism of action for the loss of tolerogenic effect of BMDCs with Abx treatment and old age.

| Loss of tolerogenic potential of DCs with gut dysbiosis is primarily mediated through secreted factors and modulation of regulatory and metabolic gene expression
In both proliferation and Tregs induction experiments, total CD4 + T cells were stimulated through plate-bound anti-CD3 and soluble anti-CD28 antibodies while co-culturing with DCs for optimal activation of T cells. This approach nullifies the effect of phenotypic difference among DCs for the observed effect; hence, we hypothesized that gut dysbiosis with old age and antibiotic treatment instigate cell-intrinsic properties of myeloid precursors and turn them towards immunogenic rather than tolerogenic, and this effect is mediated through secreted factors in the microenvironment of DC and T-cell co-culture.

F I G U R E 2
Mechanism of gut dysbiosis-mediated loss of tolerogenic potential in DCs. Co-culture supernatants of LPS-stimulated DC Young , DC Dysbiotic , and DC Old with syngeneic CD4 + T cells were analyzed with ELISA for cytokines secretion; (a) bar graph representation of pro-inflammatory cytokines IL-6, IL1β, TNFα, IL-12, and IFNγ; (b) bar graph representation of anti-inflammatory cytokines IL-2, IL-10, TGFβ, and IL-4. qRT-PCR based quantification of gene expression in DC Young , DC Dysbiotic , and DC Old . Bar graph representation of (c) relative gene expression of pro-inflammatory cytokine genes; (d) relative gene expression of anti-inflammatory cytokine genes; (e) relative gene expression of tolerogenic genes Irf4, Irf8, Pdl1, Btla, and Aldh2. LPS-stimulated DC SNs were estimated for NO secretion, and the same cells were examined for the expression of tolerogenic metabolic enzymes and phosphorylation status of the p65 subunit of NF-kB; (f) representative Western blots and protein level quantification of iNOS, IDO, and arginase; (g) relative expression of metabolic genes iNOS, Ido, and arginase-1 by qRT-PCR; (h) bar graph representation of NO secretion by Griess method; (i) representative Western blot and protein level quantification of phospho-p65 in DCs lysate (left panel), bar graph representation of the percentage of phospho-p65 + DCs through flow cytometry and relative expression of p65 encoding gene RelA through qRT-PCR. Data (mean ± SD) representing RT-PCR and flow cytometry are of three independent experiments, with each point representing a pool of three animals for one independent experiment, n = 3 mice/group, and Western blot data representing two independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Statistical analysis was done by one-way ANOVA followed by Tukey's multiple comparison test.   We analyzed the supernatant of DC:CD4 + T-cell co-culture for estimation of pro-inflammatory and anti-inflammatory cytokines concentration. A significant increase in pro-inflammatory cytokines IL-6, IL1β, TNFα, IL-12, and IFNγ was noticed when CD4 + T cells were co-cultured with DC Dysbiotic and DC Old compared to DC Young , which corroborates with a significant reduction in the anti-inflammatory cytokines IL-2, IL-10, TGFβ, and IL-4 release ( Figure 2a, b). These results were further validated at the transcript level. We found significantly increased expression of pro-inflammatory cytokines Il1b,Tnfa, while reduced transcripts were observed for anti-inflammatory cytokines and Arginase I (Arg 1), respectively, cooperate with DCs to confer their immunosuppressive effect (Mondanelli et al., 2017). These catabolizing enzymes deprive T cells of amino acids and lead to their suppression (Grohmann et al., 2003). IDO1 suppresses the allogeneic T-cell proliferation (Funeshima et al., 2005) and is implicated in the Tregs generation (Fallarino et al., 2006). On the contrary, DCderived nitric oxide (NO) determines either the regulatory or effector DC differentiation (Si et al., 2016). NO is synthesized from the metabolism of L-arginine by inducible NO synthase (iNOS) and induces tolerance in allograft models (Peche et al., 2005). Hence, it was of our interest to determine their role in gut dysbiosis-mediated DC modulation. We evaluated the protein, and transcript levels of IDO, iNOS, and Arginase1 in DC Young , DC Dysbiotic , and DC Old stimulated with LPS. A reduction in iNOS, arginase, and IDO at the protein level was observed in DC Dysbiotic and DC Old compared to DC Young through Western blot ( Figure 2f); concurrently, their transcript levels were also significantly reduced ( Figure 2g). Next, NO was measured in the culture supernatants, and a significant decrease in the production of NO was observed in both DC Dysbiotic and DC Old compared to DC Young (Figure 2h) complementing the decreased NO metabolizing enzyme, iNOS.

| Gut dysbiosis due to aging and antibiotic treatment of young mice induces the NFκ B pathway in DCs to regulate their immunogenic/ inflammatory phenotype
To decipher the molecular mechanism that plays a role in influencing the tolerogenic properties of DCs upon gut dysbiosis, we investigated NF-κB signaling pathway. This pathway has a quintessential role in regulating DC tolerance as it is the master regulator of various genes involved in the DC maturation (Ade et al., 2007). Inhibiting NF-κB signaling has been reported to differentiate and maintain tolerogenic DCs in the context of cancer, autoimmune disorders, and aging (Carreno et al., 2011). First, to quantify NF-κB activation, we

| Gut dysbiosis influenced by aging alters the abundance of the genus Lactobacillus in the gut, and its replenishment with Lactobacillus plantarum restores dysbiosis and DC phenotype
Next, we were curious to decode the involvement of the gut microbiota in age-predisposed dysbiosis and loss of DC tolerance. We observed a significant difference in the overall microbiota composition of young dysbiotic and old mice compared to young mice. The beta diversity plot which measures the phylogenetic relationship of bacterial communities between groups through both weighted and unweighted principal coordinate analysis indicated separate clustering among all three groups, suggesting a difference in microbial communities ( Figure 3a). Alpha diversity analysis which measures the species richness and evenness within an ecological community through Chao1 (community richness) and Shannon index (evenness) (Hughes et al., 2001) showed distinct measures among the group.
Both indices indicated a gradual decrease in diversity and evenness of microbiota in young dysbiotic and old mice groups. The Chao1 diversity index revealed a decreased species richness in young dysbiotic and old groups of mice compared to the young group, but the difference was not significant ( Figure S3a). However, a significant decrease was noted in the Shannon diversity index (Figure 3b), indicating a biased microbial community structure in old and young dysbiotic mice. Further, phylum and genus level comparisons for relative abundance among all three groups were done. Phylum, like Firmicutes showed a substantial decrease in young dysbiotic and old mice with a concurrent increase in phylum Bacteroidetes with respect to the young group (Figure 3c). At the genus level within F I G U R E 3 Gut microbiota alteration with aging and antibiotics exposure and evaluation of the immunomodulatory role of Lactobacillus plantarum. The fecal DNA of young control, young dysbiotic (21 days of antibiotics treated), and old mice were analyzed with 16S rRNA sequencing and qRT-PCR; (a) adapted weighted and unweighted unifrac PCoA plots indicating differential beta diversity; (b) Shannon diversity index depicting alpha diversity; (c) bar graphs represent relative abundance at the phylum level; (d) bar graphs represent relative abundance at the genus level; (e) bar graph represents qRT-PCR data for the indigenous relative abundance of species L. plantarum; (f) bar graph represents qRT-PCR based quantification of relative accumulation of L. plantarumon the day 0, 15 and 21 of administration through oral gavage with (10 8 CFU/mL) in old mice; (g) dot plot and bar graph depicting percentage population of MHCII + CD80 + DCs; (h) MHCII + CD86 + DCs gated on CD11c + cells and (i) Representative histogram and frequency of dextran-FITC uptake, as assessed by flow cytometry in DC Young , DC Old and DC old-LP . All the data (mean ± SD) are from three independent experiments, with each point representing a pool of three animals for one independent experiment, n = 3 mice/group except (f, i), with each point in the bar graph indicating one animal, n = 5. The statistical analysis done for RT-PCR (f) and flow cytometry data (g, h) is by two-way ANOVA followed by Sidak's multiple comparisons test, and the rest is by one-way ANOVA followed by Tukey's multiple comparison test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.    (Figure 3i and Figure S3h).

| Lactobacillus plantarum modulates the tolerogenic regulatory program of old DCs by metabolic rewiring and downregulation of NFκ B pathway
Lactobacillus plantarum has been shown to modulate NF-κB mediated inflammatory pathway in the model of colitis (Yu et al., 2020) and bacterial pathogenesis (K. Li et al., 2022). Hence, we revisited (f) relative gene expression of iNOS, arginase1, and Ido1 evaluated by qRT-PCR; (g) representative Western blot and quantification of iNOS, IDO, and arginase1 at the protein level. Data (mean ± SD) are of 5 animals, with each point in the bar graph representing one animal, n = 5 except (a, g) data representing one independent experiment. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. One-way ANOVA followed by Tukey's multiple comparison test was performed for statistical analysis.  ( Figure S4). Further, we evaluated the expression of the Slc2a1 gene, encoding GLUT1, a known glucose transporter, and observed lesser expression in DC Old-LP ( Figure S4c). These results suggest that LP fixes the metabolic network to promote tolerance in DC Old-LP . These results suggest that administration of LP in the gut not only restores the phenotype and the features of DC Old but also reinvigorates their functional properties to the level of DC Young .

| Lactobacillus plantarum modulates aged DCs by enhancing their migratory and regulatory function
The regulatory role of DCs depends on their ability to migrate to the draining lymph nodes to activate T cells (Hadeiba et al., 2008). CCR7 and CCR9 are important migratory molecules present on the DC surface, driving their trafficking in the skin and gut, respectively, to induce tolerance (Ohl et al., 2004;Pathak et al., 2020). Therefore, we studied the expression of these molecules and DC migration in vivo.

| Lactobacillus plantarum modulates the ageassociated gene expression profile of DC Old and directs them towards a tolerogenic phenotype
Lactobacillus influence the physiology of their hosts by diverse mechanisms, specifically by interacting with the immune system and play an imperative role in the development and maintenance of the same (Kemgang et al., 2014;van Baarlen et al., 2009). To unravel the molecular mechanism of LP-mediated immunomodulation, we performed the global transcriptomic profiling of DC Young , DC Old , DC Dysbiotic , and DC Old-LP groups. The differential gene expression analysis was filtered by logFC cutoff of 0.5 and a significant p value cutoff threshold of <0.05. Our data revealed a contrasting similarity between DC Dysbiotic with DC Old and DC Young with DC Old-LP as depicted in volcano plots ( Figure S5a) (1478) and DC Dysbiotic (839) Figure S5). All these findings indicate the substantial role of aging and antibiotic-mediated gut dysbiosis towards the generation of inflammatory phenotype, which interestingly resorted towards tolerogenic mode by replenishment of gut of old mice with LP.

| DISCUSS ION
Aging is an incessant and irrevocable physiological process accompanied by the phenotypic and functional changes of senescence.
It results in chronic inflammation, the decline in functional immunity, and altered gut microbial composition. The gut microbiome profoundly influences human health and disease, and recently, gut dysbiosis has been connected with several chronic diseases viz.    (Belkaid & Hand, 2014;Zheng, Fang, et al., 2020). However, the mechanism of age-associated immune dysfunction through gut dysbiosis remains largely unexplored. Consequently, in the current study, we have tried to uncover the influence of aging on gut microbiota and its impact on DCs. Additionally, we explored the prospect of a key probiotic strain for restoring age-mediated DC dysfunction.

F I G U R E 5 Lactobacillus
Dendritic cells play an essential role in regulating immunity and tolerance by connecting innate and adaptive immune systems (Raker et al., 2015). Tolerogenic DCs maintain central and peripheral tolerance by regulation of optimum effector T cell and regulatory T-cell responses (Hasegawa & Matsumoto, 2018). However, with age, the regulatory function of DCs is impaired. The current study explores the gut-age axis by monitoring dysbiosis-associated functional de- We observed that NF-κB signaling was required for the homeostatic maturation and function of DC Young while overactivation was observed in DC Dysbiotic and DC Old . NF-κB is a master regulator of inflammation, and its dysregulated activation is associated with several autoimmune, inflammatory pathogenesis (Barnabei et al., 2021) and aging (Adler et al., 2007). Over-expression of NF-κB subunit RelA/p65 induces a senescent phenotype in cultured cells, while the low level in the progeroid mouse model has been interrelated with the delay in age-associated pathologies (Flores et al., 2017;Seitz et al., 2000).
In agreement, we demonstrate that overt activation of NF-κB in DC Dysbiotic and DC Old is due to increased phosphorylation of p65, interestingly, p65 phosphorylation decline in DC Old-LP on feeding the old animals with LP. We assume that the decline in phosphorylation of suppress TGFβ/SMAD signaling (Bitzer et al., 2000). In our experimental settings, we have shown downregulation of Tgfb expression, and the corresponding decrease in TGFβ in the supernatant of DC Old and DC Dysbiotic CD4 + T-cell co-culture. Our transcriptomic profiling re-confirmed our results; we observed downregulation of Tgfb, Tgfb receptors and associated receptors (Tgfb1, Tgfr3, and Tgfbrap1) in F I G U R E 6 Inoculation of Lactobacillus plantarum into the gut of old mice modulates the transcriptomic gene profile. RNA was isolated from LPS-stimulated DC Old-LP , DC Old , DC Dysbiotic , and DC Young for gene array profiling by Agilent GeneChips. Data analysis was done, and differential gene expression between the groups was compared. (a) UpSet plot representing the intersection between the sets of differentially expressed genes from various comparisons in the limma setup with a vertical bar plot reporting the intersection size, the dot plot was reporting the set participation in the intersection, and the horizontal bar plot reporting the set sizes. The heatmap (b) represents the differentially expressed features associated with the immune response (left panel), differentially expressed metabolism-related features (right panel). Rows represent the features, and columns signify the samples. The 'Color Key' denotes the log-transformed expression values. Data are from three independent sets, with each data set representing a pool of three animals (n = 3 mice/group). The tolerogenic phenotype of DCs to induce T-cell hyporesponsiveness is orchestrated by a complex network of immunoregulatory molecules (Domogalla et al., 2017) and metabolic rewiring (Adamik et al., 2022). Transcription factors IRF4, IRF8, and iNOS, arginase 1 and IDO in DC Dysbiotic and DC Old that contributed to their immunogenic phenotype was also restored through LP reconstitution. The NO, iNOS, arginase 1, and IDO have been implicated in imparting tolerogenic behavior of DC (Panfili et al., 2019;Verinaud et al., 2015). NO responsible for the tolerogenic behavior of DCs is a product of L-arginine metabolism and its level is controlled in cells by iNOS (Moncada, 1999) while immunoregulatory enzyme IDO elicits peripheral tolerance in DCs. Another important observation in this study included the rewiring of metabolic processes in DC Old by switching from mitochondrial OXPHOS to a glycolytic pathway with greater consumption of glucose compared to DC Young and DC Old-LP .
Notably, DCs at a steady-state condition showing tolerogenic behavior normally engage in mitochondrial OXPHOS for energy requirements. However, after TLR-based activation they become immunogenic with the remarkable switch towards glycolysis, similar to Warburg metabolism (Krawczyk et al., 2010).
Overall, the parallel observations in DC Dysbiotic and DC Old suggest that disruption in gut microbiota due to aging or antibiotic treatment of young mice can contribute to the loss of tolerance in DCs. We have uncovered a notable contribution of gut microbiota contributing to the loss of immune tolerance in the dendritic cells on aging. Our data show a significant role of NF-κB signaling in immune tolerance at old age, through the regulation of cytokine signaling.
Furthermore, disruption of the gut on aging or antibiotic treatment of young group also results in the disappearance of beneficial commensals like Lactobacillus plantarum, resulting in the loss of tolerance in DCs. Interestingly, LP replenishment restores tolerogenic phenotype and properties of old DCs through modulation of metabolism and immunity-associated gene network profiling. The study for the first time demonstrates the role of aging on gut dysbiosis, which ultimately resulted in the loss of functionality in DCs in maintaining tolerance. Finally, this study suggests the immunotherapeutic role of LP to treat age-related disruption in the gut for the tolerogenic function of dendritic cells.

| Animals and ethical statement
Animal experiments were performed with C57BL/6 female mice.
2-3 months old mice were considered young and 22-24 months were considered old. All the animals were procured from IMTECH Center for

| Gut dysbiosis mice model
Young mice were given drinking water with a broad-spectrum antibiotic (Abx) cocktail (Ampicillin, 1 g/L; neomycin sulfate, 1 g/L; metronidazole, 1 g/L and vancomycin, 0.5 g/L) Himedia (Mumbai, India) ad libitum in drinking water for 21 days to disrupt the gut microbiota, with change in Abx containing water after every 3 days. To assess the effect of Abx-mediated dysbiosis on the gut bacterial burden, serially diluted fecal samples from all groups were plated on a BHI medium under aerobic and anaerobic conditions. Facultative anaerobic conditions were maintained by keeping the plates in airtight sealed glass chambers in presence of Anaerogas packs-LE002A-5NO, Himedia (Mumbai, India).

| Bacterial strains used in the in vivo experiments
Lactobacillus plantarum MTCC 2621 obtained from Microbial Type Culture Collection (MTCC), IMTECH (Chandigarh, India) was cultured in DeMan, Rogosa Sharpe broth (Merck, Darmstadt, Germany) at 37°C, 5% CO 2 . The bacterial suspension was pellet down by centrifugation at 3000 g for 10 min and washed with 1xPBS twice, and CFU was adjusted to 1 × 10 8 . Thereafter, the same CFU in 200 μL sterile PBS was orally gavaged to mice every alternate day for 21 days, until mice were sacrificed. Fecal samples were obtained on Days 0, 7, and 21 to assess the accumulation of bacteria inside the gut.

| Flow cytometry
Single-cell suspension was incubated for 30 min at 4°C with Fc block (anti-mouse CD16/32 Ab) to prevent non-specific binding of Abs.
Next, cell-surface staining was done with fluorochrome-tagged monoclonal antibodies for another 30 min at 4°C as per the experiment.
For intracellular staining, cell surface stained cells were fixed and permeabilized using True-nuclear™ transcription factor kit-Biolegend (San Diego, CA) and stained for intracellular targets in accordance with the manufacturer's protocol. The data was obtained with BD FACSVerse and analyzed through BD FlowJo software (San Jose, CA).

| Cytokine estimation
The cytokines were estimated in the culture SNs by sandwich ELISA. Briefly, 96-well ELISA plates were coated with purified rat anti-mouse IL-6, IFNγ, IL-17, IL-12, TGFβ, IL-4, TNFα, IL-2 (all at 2 μg/mL), and IL-10 (4 μg/mL) antibodies in phosphate buffer (0.01 M-pH 9.2 or pH 6) overnight (O/N) at 4°C as per the manufacturer's protocol BD Biosciences (San Diego, CA). The next day, blocking was done with BSA (1%) in PBS for 2 h at room temperature (RT). Subsequently, 50 μL samples (culture SNs) along with the respective standards of the recombinant cytokines were added to each well followed by O/N incubation at 4°C. After incubation and subsequent washing, 50 μL biotinylated anti-mouse antibodies (2 μg/ mL) for respective cytokines were added to plates and incubated for 2 h at RT. Further, plates were incubated with streptavidin-HRP at 1:10,000 dilution at 37°C for 40 min. After subsequent washing, 1× TMB solution was used for color development and the reaction was stopped by H 2 SO 4 (7%). Absorbance was read at 450 nm on a spectrophotometer, Synergy H1, BioTek (Santa Clara, CA). The quantification of cytokines (pg/mL) was done using a standard curve of recombinant cytokines log2 serial dilutions.

| Quantitative real-time PCR (qRT-PCR)
Trizol reagent was used to isolate total RNA from LPS-stimulated  Table S1.

| DC-T cell co-cultures
CD4 + T cells were purified from single-cell suspension of the mouse splenocytes with BD IMag™ mouse CD4 T lymphocyte enrichment set-DM, BD Biosciences (San Diego, CA) by MACS negative selection following the manufacturer's protocol. Later, sorted T cells were co-cultured with LPS-stimulated DCs in a 1:5 ratio (DC: T cell) with a total seeding population of 2 × 10 5 cells per well of 96-well plates.
The proliferation of labeled T cells was assessed, using flow cytometry. The culture SNs were collected to later estimate the cytokines through ELISA. For syngeneic and allogeneic setup, CD4 + T cells were obtained from C57BL/6 and BALB/c mice, respectively.

| In vitro Tregs induction assay
The same procedure was followed as described in DC-T cell co-

| Nitric oxide (NO) production
After 24 hours, the culture SNs of LPS-stimulated BMDCs were collected and NO was measured according to Griess method (Pahari et al., 2016). In brief, a 1:1 ratio of SNs was added to the Griess reagent (50 μL)-Sigma Aldrich (St. Louis, MO) and incubated for 5 min at room temperature. Later, the absorbance was measured at 550 nm.
NO was quantified in comparison with sodium nitrite (NaNO 2 ) as a standard (μM).
After proper washing, uptake was then assessed by flow cytometry.
For phagocytic uptake of apoptotic bodies, Jurkat T cells labelled with CFSE were treated with actinomycin D for 14-15 h. to induce apoptosis in the same. Later, apoptotic cells were co-cultured with LPS-stimulated DCs for 3-4 h, and uptake was monitored by flow cytometry.

| In vivo DC migration
LPS-stimulated DCs from different groups of mice were labelled with CFSE (2 μM), and 5 × 10 6 cells were adoptively transferred through an intravenous (i.v.) route into Old mice in 200 μL of sterile PBS. After 5 days, animals were sacrificed and cells from the spleen, mesenteric lymph node and Peyer's patches were stained for fluorochrome-tagged anti-mouse CD11c antibody. Later, the CD11c + CFSE + population was monitored to assess migrated DC population. Furthermore, cells from the same tissues as mentioned were stained by fluorochrome-tagged anti-mouse CD4, CD127, CD25 and FOXP3 to assess the induction of Tregs using flow cytometry.

| Inhibition of the NFκ B pathway
The DCs harvested on the seventh day of culture were stimulated with LPS (1 μg/mL) in presence of a selective inhibitor for nuclear translocation of NF-κB p65, JSH-23 (7 μM) #481408, Sigma Aldrich

| Dextran-FITC intestinal permeability assay
Evaluation of intestinal epithelial barrier permeability was done utilizing the dextran-FITC permeability assay (Furuta et al., 2001;Meisel et al., 2018). In brief, mice were fasted for 4 h and then 60 mg dextran-FITC (MW 4000), #46944, Sigma Aldrich (St. Louis, MO) per 100 g body weight was given through oral gavage. After 3 h of gavage, blood samples (400-500 μL) were obtained from the tail vein and plasma was extracted by centrifugation (2000 g for 10 min at 4°C). Thereafter, 50 μL blood plasma was transferred in duplicates into a flat-bottom 96-well plate (Corning, NY), and fluorescence reading was measured in fluorescence spectrophotometer setup in Synergy H1, BioTek (Santa Clara, CA) with emission and excitation wavelengths of 520 nm and 490 nm, respectively. Plasma dextran-FITC concentration was assessed using a standard curve established by serial dilution of the same.

| Western blotting
The whole cell lysate was prepared by lysing LPS-stimulated DCs in RIPA lysis buffer (Tris HCl25 mM, NaCl150 mM, EDTA5 mM, Triton X-1000.1%, sodium deoxycholate 1%, SDS 0.1%) supplemented with PMSF, phosphatase and protease inhibitor cocktail after 40 min incubation on ice. Lysates were centrifuged at 8000 g for 10 min, and the protein concentration of the lysate was measured by the BCA method. 40 μg of protein was added to Laemmli buffer and boiled (10 min, 95°C). Proteins were separated in 10-12% SDS-PAGE gels and transferred to the PVDF membrane. Later, 5% BSA was used to block the membranes followed by incubation with the primary antibody on a shaker overnight at 4°C. All primary antibodies were diluted at a 1:500-1000 ratio for immunoblotting.
HRP-conjugated secondary anti-mouse and anti-rabbit antibodies were used (1:10000) for the detection of primary antibody binding.
Each step included regular washings and incubations. Finally, blots were developed using Novex™ ECL Chemiluminescent Substrate Reagent Kit, Invitrogen-Thermo Fisher Scientific, (Waltham, MA) and visualization was done on iBright™ FL1500 Imaging System, Invitrogen-Thermo Fisher Scientific (Waltham, MA). Blot analysis and quantification were done using ImageJ analysis software. Abs to phospho p65, p65, anti-IDO, iNOS, arginase1, and β-actin (loading control) were used for Western blotting. Sequencing libraries were prepared from amplified V3-V4 region of 16S rRNA by Index PCR using Nextera XT index kit as per manufacturer instructions (Illumina, #15044223 Rev. B). The quality of the final libraries was checked using high-sensitivity D1000 screen tape in Tape-Station 2200 Agilent Technologies (Santa Clara, CA), and final library quantification was performed in Qubit Fluorometer.

| Glucose uptake assay
Paired-end (2 × 250 bp) sequencing of these libraries was performed in NovaSeq 6000 (Illumina, San Diego, CA). Sample preparation and sequencing were done at the National Institute of Biomedical Genomics (Kalyani, India). The raw data were first checked for quality, and then, DADA2 was used to remove chimeric information. The demultiplexed sequences were then trimmed to lengths of 250 and 230 for forward and reverse, respectively. After that, the frequency table and data were obtained. Model classifier, which was created using the reference Green gene database, was used to classify taxa.
The taxonomic plots were then created using the model classifier parameters such as forward primer CCTAC GGG NGG CWG CAG and reverse primer GACTA CHV GGG TAT CTA ATCC. The OTU table has been created from taxonomic classification. MAFFT was used to align the sequences, which were then masked to obtain masked aligned sequences. Subsequently, for diversity analysis, phylogenetic trees were created. With the parameter sampling depth 582,778, PCoA and diversity analysis plots were obtained using rooted and unrooted trees.

| Microarray-based gene expression analysis
RNA was isolated from DCs after LPS stimulation (1 μg/mL) using TRIzol reagent according to the manufacturer's protocol, Sigma Aldrich (St. Louis, MO). Later, the DNase1 enzyme was added for the digestion and removal of genomic DNA. Thereafter, RNA was quantified with the NanoDrop ND-100 Spectrophotometer, NanoDrop Technologies (Wilmington, DE), and quality for the same was examined with the Tapestation 4200™, Agilent Technologies (Santa Clara, CA). Samples with RIN (RNA integrity number) score of more than 6 and a 28S:18S rRNA ratio of around 2∶1 were included for the further process. Thereafter, labelling and hybridization were done using Agilent one-color (Cy3 fluorochrome), a microarray-related gene expression platform in accordance with the manufacturer's instructions for evaluating mRNA expression. Briefly, 500 ng of RNA was labelled with a one-color Low-Input Quick Amp labelling kit, 5190-2305 (Agilent Technologies, Palo Alto, CA, USA). Cy3-labelled mRNA samples were hybridized onto a mouse 8 × 60 K Gene Expression V2 Array kit, G4858A, Agilent Technologies, (Palo Alto, CA) for 16 h at 55°C in a rotator oven with subsequent washing. DNA microarray scanner, Agilent Technologies, (Palo Alto, CA) was used to scan the array slides, and Agilent Feature Extraction software 10.5 (Palo Alto, CA) to extract hybridization signals. Thereafter, with the help of limma R package, the quantile normalization method was implemented to normalize the quantitative microarray data. Using the same package, differential feature estimation was performed by lmFit () function to fit a linear model to expression data of each feature and eBayes (empirical Bayes) function used to obtain adj.p.value and logFC. Threshold adj.p. value ≤ 0.05 and abs (logFC) ≥ 0.5 were used to obtain significant differential features. R package gprofiler2 was used for the enrichment of functional annotations. Summarization of gene ontology and treemap plot were done with R package rrvgo.

| Statistical analysis
One-way ANOVA followed by Tukey's multiple comparison test and two-way ANOVA followed by Sidak's multiple comparisons test were done using GraphPad Prism 6 software (San Diego, CA). A p value <0.05 was considered significant.

AUTH O R CO NTR I B UTI O N S
HB performed the experiments and contributed to experimental design, data analysis, discussion and writing. RPS helped in flow cytometry data acquisition with HB and analyzed the 16S rRNA sequencing data. SS conducted experiments with HB. JNA conceptualized, designed and supervised the study, and contributed to the discussion and writing of the manuscript. RK conceptualized, designed and supervised the study, analyzed the data and wrote the manuscript. All authors revised and commented on the manuscript.
All authors approved the final manuscript.

FU N D I N G I N FO R M ATI O N
This study is supported by the grants from CSIR, India grants OLP134, HUM (No. BSC0119) and CSIR-FIRST (No. MLP062).

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare that they have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The 16 s rRNA sequencing data from this study have been deposited at the GenBank Sequence Read Archive with the accession number PRJNA879291. The microarray raw data are submitted to the GEO repository with GEO accession number GSE 213155.