The NQR Complex Regulates the Immunomodulatory Function of Bacteroides thetaiotaomicron

Abstract

The gut microbiome and intestinal immune system are engaged in a dynamic interplay that provides myriad benefits to host health. However, the microbiome can also elicit damaging inflammatory responses, and thus establishing harmonious immune-microbiome interactions is essential to maintain homeostasis. Gut microbes actively coordinate the induction of anti-inflammatory responses that establish these mutualistic interactions. Despite this, the microbial pathways that govern this dialogue remain poorly understood. We investigated the mechanisms through which the gut symbiont Bacteroides thetaiotaomicron exerts its immunomodulatory functions on murine- and human-derived cells. Our data reveal that B. thetaiotaomicron stimulates production of the cytokine IL-10 via secreted factors that are packaged into outer membrane vesicles, in a TLR2- and MyD88-dependent manner. Using a transposon mutagenesis-based screen, we identified a key role for the B. thetaiotaomicron-encoded NADH:ubiquinone oxidoreductase (NQR) complex, which regenerates NAD+ during respiration, in this process. Finally, we found that disruption of NQR reduces the capacity of B. thetaiotaomicron to induce IL-10 by impairing biogenesis of outer membrane vesicles. These data identify a microbial pathway with a previously unappreciated role in gut microbe-mediated immunomodulation that may be targeted to manipulate the capacity of the microbiome to shape host immunity.

Figure 1.. B. theta secretes immunomodulatory factors that are sensed via a TLR2-MyD88 axis.

(A) Unfractionated wild-type splenocytes were stimulated with B. theta^VPI-5482 heat-inactivated pellet (OD =1.0) or B. theta^VPI-5482 conditioned media (1% v/v), or appropriate media control for 2 days and IL-10 was assessed in the supernatant by ELISA. (B) Wild-type splenocytes were stimulated with B. theta^VPI-5482 unfractionated conditioned media or TYG media (1% v/v), or isolated OMVs from the conditioned media or TYG control media that underwent the same process as that for OMV isolation (resuspended in an equivalent volume of PBS to that of the conditioned media from which they were isolated) (1% v/v). Samples were incubated for 2 days and IL-10 was assessed in the supernatant by ELISA. (C-E) Wild-type and MyD88^−/− (C) TLR2^−/− (D) or TLR4^−/− (E) splenocytes were stimulated with B. theta^VPI-5482 conditioned media or TYG media control (1% v/v) for 2 days and IL-10 was assessed in the supernatant by ELISA. Data points represent an independent technical replicate, and horizontal lines show the mean (n = 3). Note, the IL-10 values for B. theta conditioned media and TYG media from wild-type splenocytes in panel D represent the same data as the values in panel A, and the IL-10 values for B. theta conditioned media and TYG media from wild-type splenocytes in panel C represent the same data as the values in panel B; stimulations of the respective wild-type and mutant splenocytes were performed concurrently. Graphs are representative of at least 4 experiments (A-E). Statistical significance was determined using Student’s T test: (A, B), and two-way ANOVA with Sidak’s post hoc test, comparisons to wild-type (C-E). ns (not significant) P≥0.05; **P≤0.01, ***P≤0.001, ****P≤0.0001.

Figure 2.. Gene BT1160 is required for B. theta mediated induction of IL-10.

(A) Wild-type splenocytes were stimulated with conditioned media from a library of 2,009 individual B. theta^{ΔtdkΔcps1-8} transposon mutants (1% v/v) or Pam3CSK4 for 2 days and IL-10 levels were measured by ELISA. For each plate, IL-10 levels were normalized to the IL-10 level induced by the positive control Pam3CSK4. Data shows the Log2 transformed value normalized to the positive control Pam3CSK4 of the 1982 mutants whose growth was higher than background control media. (B) Wild-type splenocytes were stimulated with conditioned media from B. theta^{ΔtdkΔcps1-8} or transposon mutant 19_H4 (1% v/v) for 2 days and IL-10 was assessed by ELISA. (C) The transposon insertion in transposon mutant 19_H4 was mapped to gene BT1160 that is predicted to encode the first subunit of the NQR complex. (D) Wild-type splenocytes were stimulated with conditioned media (1% v/v) from B. theta harboring a truncated deletion of gene BT1160 (acapsular background, left; wild-type background, right), or appropriate parent strain, for 2 days, and IL-10 levels were measured by ELISA. (E) Human PBMCs from two healthy donors were stimulated with conditioned media (3% v/v) from B. theta harboring a truncated deletion of gene BT1160 in the wild-type background, parent strain (Δtdk), or TYG media for 2 days, and IL-10 levels were measured by ELISA. (F) Wild-type splenocytes were stimulated with conditioned media (1% v/v) from B. theta^Δtdk, B. theta^{ΔtdkΔBT1160_trunc}, or B. theta^{ΔtdkΔBT1160_trunc} complemented with BT1160 (B. theta^{ΔtdkΔBT1160_trunc::BT1160}) for 2 days, and IL-10 levels were measured by ELISA. (G) Wild-type splenocytes were stimulated with conditioned media (1% v/v) from B. theta harboring a truncated deletion of gene BT1160 (red), a full deletion of gene BT1160 (grey) (Δtdk background), or the parent strain B. theta^Δtdk (blue) for 2 days, and IL-10 levels were measured by ELISA. Individual data points show mean values, n=2 (A) or individual technical replicates (B, D-G), and horizontal lines represent the mean, n=3 (B, D-G). Graphs are representative of at least 4 individual experiments (B, D, F, and G) and 1 individual experiment (A, E). Statistical significance was determined by Student’s T-test: (B, D), and one-way ANOVA with Dunnett’s post-hoc test, comparisons to B. theta^Δtdk or between all (E-G). ns (not significant) P≥0.05, *P<0.05, **P≤0.01, ***P≤0.001, ****P≤0.0001.

Figure 3.. B. theta NQR complex promotes IL-10 induction.

(A) Schematic of the gene locus that encodes the NQR complex and flanking genes. (B) The expression (2^{Δ/Δ Ct values}) of nqr locus genes BT1155-BT1159 and control flanking genes (BT1154 and BT1161). B. theta^Δtdk, B. theta ^{ΔtdkΔBT1160_trunc}, and B. theta ^{ΔtdkΔBT1160_full} were grown in vitro in TYG media and cells were harvested at mid-log growth, and RNA expression profiles were assessed by RT-qPCR with the B. theta 16S rRNA gene as the housekeeping gene. (C-E) Wild-type splenocytes were stimulated with different doses of conditioned media (0.5%, 1%, 3% v/v) from B. theta harboring deletions of genes BT1155-BT1159 (C), or genes BT1155-BT1160 (D), and conditioned media (1% v/v) from B. theta harboring a deletion in gene BT1161 (E), on the B. theta^Δtdk background for 2 days, and IL-10 levels were measured by ELISA. Data points represent the mean of technical replicates of independent biological replicates, n=3 (B) or independent technical replicates, n=3 (E), or the mean of three technical replicates, n=3 (C, D) and horizontal bars represent the mean (B, E), and error bars shown the standard deviation (C, D). Graphs are representative of a single experiment (B) or 4 experiments (C-E). Note, the IL-10 values for B. theta^Δtdk in C and D represent the same data. Statistical significance was determined using one-way ANOVA with Dunnett’s post hoc test and comparisons between all groups: (B), and Student’s T test (E) on area under the curve: (C-D). ns (not significant) P≥0.05, *P<0.05, **P≤0.01, ***P≤0.001, ****P≤0.0001.

Figure 4.. NAD+ supplementation does not restore IL-10 induction by nqr deficient B. theta mutants.

(A) Wild-type splenocytes were stimulated with conditioned media (1%, 3%, or 5 % v/v) from wild-type B. theta (Δtdk), B. theta harboring a deletion of gene BT1160_trunc, or genes BT1155-BT1160, or control TYG media for 2 days, and IL-10 levels were measured by ELISA. The conditioned media was generated from bacterial cultures that were grown in TYG media supplemented with NAD+ at a final concentration of 0 μM, 1 μM, 10 μM, or 100 μM. (B) Wild-type splenocytes were plated in RPMI supplemented with NAD+ to a final concentration of 0 μM, 1 μM, 10 μM, or 100 μM, and stimulated with conditioned media from wild-type B. theta (Δtdk), B. theta harboring a deletion of gene BT1160_trunc, or genes BT1155-BT1160, or control TYG media (1, 3, or 5% v/v) for 2 d, and IL-10 levels were measured by ELISA. Data points represent the mean of two technical replicates and error bars shown the standard deviation, n=2 (A/B). Graphs are representative of two experiments (A/B). Statistical significance was determined using two-way ANOVA with Tukey’s post hoc test and comparisons between all conditioned media groups (relevant statistical comparisons shown): (A/B), ns (not significant) P≥0.05, ****P≤0.0001.

Figure 5.. OMV production is directly linked to the ability of B. theta to induce IL-10 via the NQR complex.

(A) Schematic of NQR gene locus present within each B. theta mutant. (B) Wild-type splenocytes were stimulated with conditioned media (1% v/v) or OMVs isolated from the conditioned media (1% v/v of OMVs resuspended in PBS back to original volume from which they were isolated) from B. theta^Δtdk or B. theta^{ΔtdkΔBT1160_trunc} for 2 days, and IL-10 levels were measured by ELISA. (C-E) Wild-type splenocytes were stimulated with OMVs isolated from the conditioned media (1% v/v of OMVs resuspended in PBS as in (B)) from B. theta^Δtdk or B. theta^{ΔtdkΔBT1155-BT1159} (C), B. theta^Δtdk or B. theta^{ΔtdkΔBT1155-BT1160} (D), or B. theta^Δtdk or B. theta^{ΔtdkΔBT1161} (E) for 2 days, and IL-10 levels were measured by ELISA. (F, G) Protein concentration of independent biological batches of OMVs isolated from the conditioned media of B. theta^Δtdk, B. theta^{ΔtdkΔBT1160_trunc}, or B. theta^{ΔtdkΔBT1160_trunc::BT1160} (F), or B. theta^Δtdk, B. theta^{ΔtdkΔBT1155-BT1159}, or B. theta^{ΔtdkΔBT1155-BT1160} (G), as measured by Qubit. Protein concentration was normalized to the value from a biological batch prepared concurrently from the parent strain (B. theta^Δtdk) which was set to 100. (H) Wild-type splenocytes were incubated with OMVs isolated from the conditioned media of B. theta^Δtdk or B. theta^{ΔtdkΔBT1160_trunc} (1% v/v). Non-normalized OMVs were resuspended in an equivalent volume of PBS to the original volume of conditioned media they were isolated from (blue and red) and normalized OMVs were concentrated to normalize the protein concentration between the mutant and the parent strain (pink). Samples were incubated for 2 days, and IL-10 was measured in culture supernatants by ELISA. Data points represent independent technical replicates (B-E, H) or independent biological replicates (F, G) and horizontal bars represent the mean, n=3 (B-E, H), or median (F, G), n=5 (F), n=4 (Δtdk) n=3 (mutants) (G). Note, the IL-10 values for B. theta^Δtdk isolated OMVs in panels B and D represent the same data; stimulations were performed concurrently. Graphs are representative of 4 experiments (B-E, H), 5 pooled experiments (note, graph shows pooled data from all experiments) (F), 3 pooled experiments (G). Statistical significance was determined using Student’s T test: (B-E), Kruskal-Wallis test with Dunn’s post-hoc test, comparisons between all: (F, G), or one-way ANOVA with Dunnett’s post-hoc test, comparisons between all: (H). ns (not significant) P≥0.05, *P<0.05, **P≤0.01, ***P≤0.001, ****P≤0.0001.