The Gut Microbiome Modulates Body Temperature Both in Sepsis and Health

Abstract

Rationale: Among patients with sepsis, variation in temperature trajectories predicts clinical outcomes. In healthy individuals, normal body temperature is variable and has decreased consistently since the 1860s. The biologic underpinnings of this temperature variation in disease and health are unknown. Objectives: To establish and interrogate the role of the gut microbiome in calibrating body temperature. Methods: We performed a series of translational analyses and experiments to determine whether and how variation in gut microbiota explains variation in body temperature in sepsis and in health. We studied patient temperature trajectories using electronic medical record data. We characterized gut microbiota in hospitalized patients using 16S ribosomal RNA gene sequencing. We modeled sepsis using intraperitoneal LPS in mice and modulated the microbiome using antibiotics, germ-free, and gnotobiotic animals. Measurements and Main Results: Consistent with prior work, we identified four temperature trajectories in patients hospitalized with sepsis that predicted clinical outcomes. In a separate cohort of 116 hospitalized patients, we found that the composition of patients' gut microbiota at admission predicted their temperature trajectories. Compared with conventional mice, germ-free mice had reduced temperature loss during experimental sepsis. Among conventional mice, heterogeneity of temperature response in sepsis was strongly explained by variation in gut microbiota. Healthy germ-free and antibiotic-treated mice both had lower basal body temperatures compared with control animals. The Lachnospiraceae family was consistently associated with temperature trajectories in hospitalized patients, experimental sepsis, and antibiotic-treated mice. Conclusions: The gut microbiome is a key modulator of body temperature variation in both health and critical illness and is thus a major, understudied target for modulating physiologic heterogeneity in sepsis.

Keywords: heterogeneity; microbiome; sepsis; subphenotypes; temperature.

Figure 1.

Subphenotypes of patients with sepsis on the basis of temperature trajectories. Using serial temperature measurements from 13,051 hospital admissions at the University of Michigan Hospital, we used group-based trajectory modeling to identify subphenotypes of patients on the basis of initial temperature trajectories. These subphenotypes were identical in number and trajectory to those previously identified by Bhavani and colleagues (4): “hyperthermic, slow resolvers” (n = 852 [6.4% of the cohort]); “hyperthermic, fast resolvers” (n = 3,690 [28.3%]); “normothermic” (n = 8,343 [63.9%]); and “hypothermic” (n = 166 [1.3%]).

Figure 2.

The gut microbiome is highly heterogeneous in hospitalized patients and is predictive of temperature trajectory. We characterized gut bacteria in rectal swabs obtained from 118 patients hospitalized at the University of Michigan and compared community composition across temperature trajectories. (A) Each row and column represents the gut community of a unique hospitalized patient, and the intersection represents the number of shared bacterial families identified. (B) Random forest analysis of the top 20 operational taxonomic units (OTUs) associated with temperature trajectory in our human cohort. All 20 OTUs are classified within the Firmicutes (Bacillota) phylum. Lachnospiraceae was the most commonly represented family among the top 20 OTUs (denoted in red). (C) Relative abundance of the top 20 OTUs from B within each temperature trajectory. Overall P = 1.3 × 10⁻⁸ (Kruskal-Wallis test). Between-group differences were analyzed using Tukey’s honestly significant difference test for multiple comparisons. *P < 0.05 and ****P < 0.0001. Bars represent mean and SEM.

Figure 3.

The microbiome is a major source of temperature variability in a murine model of sepsis. Germ-free and conventional mice were injected with 5 mg/kg intraperitoneal LPS or vehicle (negative control). Temperature was measured rectally at baseline and after exposure. (A) Day 1 and 3 temperatures were compared with baseline temperature (Day 0) to determine temperature change. The number of mice per cohort was 28–46 for conventional mice and 6 or 7 for germ-free mice. For the Day 1 comparison, P = 0.02 (conventional with LPS vs. germ-free with LPS). (B) Genetically identical (C57BL/6) mice were obtained from two vendors in multiple shipments to ensure microbial variation and were treated with 5 mg/kg intraperitoneal LPS. Temperatures were measured at baseline and 1 day after injection. Variation in temperature change was extreme, ranging from −14.3°C to 0°C. Mean change from Day 0 to Day 1 was −6.63°C ± 4.76°C (n = 37). (C) Germ-free mice from two cohorts were treated with 5 mg/kg LPS, and temperatures were measured at baseline and 1 day after injection. Among germ-free mice, variation in temperature change was small, ranging from −3°C to −0.3°C. Mean change from Day 0 to Day 1 was −1.31°C ± 0.64°C (n = 20). An F test for unequal variances revealed significant differences in temperature variability between conventional and germ-free mice treated with LPS (F = 54.83; P < 0.0001). Error bars denote SEM.

Figure 4.

Gut bacteria explain variation in temperature response during experimental sepsis in mice. Genetically identical (C57BL/6) mice (n = 37) were treated with intraperitoneal endotoxin (5 mg/kg), and temperature was measured rectally after 24 hours and normalized to the preendotoxin baseline for each animal. Variation in temperature change was significantly correlated with the community composition of cecal microbiota (P < 0.0001, permutational multivariate ANOVA). (A) Random forest analysis of temperature change identified Lachnospiraceae as the bacterial family most strongly associated with temperature change during sepsis. (B) Variation in relative abundance of Lachnospiraceae relative to body temperature change.

Figure 5.

The gut microbiome participates in the calibration of basal temperature in healthy mice. (A) Rectal temperatures of conventional (control) and germ-free mice were evaluated using the same thermometer at the same time of day (n = 27–30). (B) Conventional mice were treated with oral antibiotic regimens in drinking water or drinking water alone (control) for 7 days. Body temperatures, determined using a rectal thermometer, were compared at Day 7 with the Day 0 pretreatment baseline (n = 10–20 per group). (C) Conventional mice were treated with daily IP injections of 50 mg/kg ceftriaxone in sterile saline, sham injection (sterile saline alone), or control (no injections) for 4 days. Body temperatures were compared at Day 4 and Day 0 (pretreatment baseline) (n = 10 per treatment). *P < 0.05. Error bars denote SEM. IP = intraperitoneal.

Figure 6.

In germ-free mice, monocolonization influences body temperature in a species-specific manner. (A and B) Germ-free mice were monocolonized with the specified microorganisms. Rectal temperatures were measured at Day 0 (before monocolonization) and Day 14 after monocolonization. Error bars denote SEM. (A) Germ-free mice were colonized with Lachnospiraceae D4, a previously characterized murine isolate (n = 11), and compared with mice “sham colonized” with vehicle (n = 12). **P < 0.01 (two-tailed t test). (B) Germ-free mice were colonized with Lactobacillus johnsonii or Clostridium propionicum or were sham colonized with vehicle (n = 12–23). *P < 0.05 (two-tailed paired t test). N.S. = not significant (P > 0.05).