The intestinal microbiota are necessary for stressor-induced enhancement of splenic macrophage microbicidal activity

Abstract

The indigenous microbiota impact mucosal, as well as systemic, immune responses, but whether the microbiota are involved in stressor-induced immunomodulation has not been thoroughly tested. A well characterized murine stressor, called social disruption (SDR), was used to study whether the microbiota are involved in stressor-induced enhancement of macrophage reactivity. Exposure to the SDR Stressor enhanced the ability of splenic macrophages to produce microbicidal mediators (e.g., inducible nitric oxide synthase (iNOS), superoxide anion, and peroxynitrite) and to kill target Escherichia coli. Exposure to the SDR Stressor also increased cytokine production by LPS-stimulated splenic macrophages. These effects, however, were impacted by the microbiota. Microbicidal activity and cytokine mRNA in splenic macrophages from Swiss Webster germfree mice that lack any commensal microbiota were not enhanced by exposure to the SDR Stressor. However, when germfree mice were conventionalized by colonizing them with microbiota from CD1 conventional donor mice, exposure to the SDR Stressor again increased microbicidal activity and cytokine mRNA. In follow-up experiments, immunocompetent conventional CD1 mice were treated with a cocktail of antibiotics to disrupt the intestinal microbiota. While exposure to the SDR Stressor-enhanced splenic macrophage microbicidal activity and cytokine production in vehicle-treated mice, treatment with antibiotics attenuated the SDR Stressor-induced increases in splenic macrophage reactivity. Treatment with antibiotics also prevented the stressor-induced increase in circulating levels of bacterial peptidoglycan, suggesting that translocation of microbiota-derived peptidoglycan into the body primes the innate immune system for enhanced activity. This study demonstrates that the microbiota play a crucial role in stressor-induced immunoenhancement.

Figure 1

Splenic macrophages from mice exposed to the SDR stressor kill more E. coli and produce more peroxynitrite than macrophages from non-stressed control mice. Spleen mass (Fig 1A), the number of splenic monocytes/macrophages (Fig. 1B), and the number of splenic neutrophils (Fig. 1C) were increased in conventional mice exposed to the SDR stressor (*p<0.05 vs. HCC Control). Exposure to the SDR stressor did not affect the number of splenic lymphocytes (Fig. 1D). The number of E. coli remaining alive within splenic macrophages from SDR stressor-exposed mice was significantly lower than the number remaining alive within macrophages from non-stressed HCC Control mice (*p<0.05 vs. HCC Control at 90 min) (Fig. 1E). Splenic macrophages from mice exposed to the SDR stressor produce significantly higher levels of peroxynitrite after stimulation in comparison to levels produced by cells from nonstressed HCC Control mice (*p<.05 vs. HCC Control at 60, 75, and 90 min) (Fig. 1F). Values are means ± SE [n=3 HCC and 3 SDR].

Figure 2

Intestinal microbiota are necessary for SDR stressor-induced enhancement in bacterial killing and peroxynitrite production. Fig 2A–C: Spleen mass, the number of splenic monocytes/macrophages, and number of splenic neutrophils were increased in both germfree and conventionalized mice exposed to the SDR stressor (*p<0.05 vs. HCC Control). Fig. 2D: Stressor exposure did not affect the number of splenic lymphocytes. Fig. 2E: Splenic macrophages from SDR stressor-exposed germfree mice that lack any endogenous microbiota failed to kill more E. coli than cells from nonstressed HCC Control germfree mice. Conventionalizing the Germfree mice with stool from conventionally-housed mice recapitulated the SDR stressor-enhanced bacterial killing. Splenic macrophages from conventionalized mice exposed to the SDR Stressor killed significantly more E. coli than did cells from nonstressed HCC Control conventionalized mice (*p<0.05). Fig 2F: Splenic macrophages from conventionalized mice exposed to SDR produced significantly more peroxynitrite at 45, 60, 75, 90 min compared to macrophages from mice in all other groups (*p<0.05). Values are means ± SE [n=9(HCC Control Germfree), n=8(SDR Stressor Germfree), n=9(HCC Control Conventionalized), n=8(SDR Stressor Conventionalized), *p<0.05)].

Figure 3

Exposure to SDR increases iNOS and proinflammatory cytokine gene expression in LPS-stimulated splenic macrophages from mice with intact intestinal microbiota. Fig 3A: iNOS mRNA expression was significantly increased in LPS-stimulated splenic macrophages from stressor-exposed conventionally-housed and conventionalized mice in comparison to cells from nonstressed controls (*p<0.05 vs. all other groups; †p<0.05 vs. unstimulated HCC Control). Fig 3B: Stressor exposure increases TNF-α mRNA expression in splenic macrophages from conventional and conventionalized mice regardless of stimulation (*p<0.05 main effect for stress). Stressor exposure failed to enhance TNF-α mRNA in splenic macrophages from germfree mice. Fig 3C: IL-1β mRNA expression is higher in LPS-stimulated macrophages (†p<0.05 vs. unstimulated HCC Control). Exposing mice with microbiota to the SDR stressor further increased IL-1β mRNA expression (*p<0.05 vs. all other groups). Stressor exposure failed to increase IL-1β mRNA expression in macrophages from germfree mice even with LPS-stimulation. Values are means ± SE [n=5(HCC Control Conventional), n=5(SDR Stressor Conventional), n=9(HCC Control Germfree), n=8(SDR Stressor Germfree), n=9(HCC Control Conventionalized), n=6(SDR Stressor Conventionalized)].

Figure 4

Antibiotic reduction of the intestinal microbiota abrogated the SDR stressor-induced increase in bacterial killing and peroxynitrite production. Fig. 4A–D: Exposure to the SDR stressor increased spleen mass, the number of splenic monocytes/macrophages, the number of splenic neutrophils, and the number of splenic lymphocytes in both vehicle- and antibiotic-treated mice (*p<0.05 vs. HCC Control). Fig 4E: Exposing vehicle-treated mice to the SDR stressor significantly increased phagocytosis (*p<0.05 vs. HCC Control at 20 min) and bacterial killing (*p<0.05 vs. HCC Control and 90 min). However, SDR stressor exposure failed to enhance bacterial killing by macrophages from antibiotic-treated mice. Fig 4F: Splenic macrophages from vehicle-treated mice exposed to the SDR stressor produced significantly higher levels of peroxynitrite (*p<0.05 vs. all other groups at 60, 75, and 90 min of stimulation). Values are means ± SE [n=8(HCC Control Vehicle), n=8(SDR Stressor Vehicle), n=8(HCC Control Antibiotics), n=8(SDR Stressor Antibiotics)].

Figure 5

Antibiotic treatment attenuates the SDR stressor-induced increase in splenic macrophage reactivity to stimulation. Fig 5A. Splenic macrophages from vehicle-treated mice exposed to the SDR stressor produce significantly higher levels of superoxide anion (*p<0.05 vs. all other groups at 15, 30, 45, 60, 75, and 90 min). Fig. 5B. iNOS gene expression is significantly higher in LPS-stimulated macrophages from vehicle-treated mice exposed to the SDR stressor (*p<0.05 vs. all other groups). SDR exposure results in significantly lower iNOS gene expression in unstimulated macrophages from vehicle-treated mice (†p<0.05 vs. unstimulated HCC Control). Antibiotic administration prevented these effects. Fig 5C: TNF-α mRNA expression is increased in LPS-stimulated macrophages from vehicle-treated stressor-exposed mice in comparison to macrophages from nonstressed controls. Antibiotic treatment prevented this increase. Fig 5D: IL-1β mRNA expression was significantly increased in LPS-stimulated macrophages from vehicle-treated stressor-exposed mice compared to macrophages from nonstressed HCC Controls (*p<0.05 vs. all other groups). Vehicle treatment significantly reduced IL-1β gene expression in unstimulated splenic macrophages from mice exposed to the SDR stressor (†p<0.05 vs. unstimulated HCC Controls). Values are means ± SE [n=5 for all groups].

Figure 6

Stressor-induced increases in TNF-α and IL-1β production by splenic CD11b+ cells were attenuated in mice treated with antibiotics. Splenic CD11b+ macrophages were stimulated with 1 μg of E. coli-derived LPS. After 18 hr in culture, TNF-α (Fig. 6A) and IL-1β (Fig. 6B) levels were measured via ELISA. **p< 0.05 vs. all other groups. *p<0.05 vs. HCC Antibiotic+LPS and vs. SDR Vehicle+LPS. Values are means± SE [n=5 for all groups].