Retroviral Infection and Commensal Bacteria Dependently Alter the Metabolomic Profile in a Sterile Organ

Abstract

Both viruses and bacteria produce "pathogen associated molecular patterns" that may affect microbial pathogenesis and anti-microbial responses. Additionally, bacteria produce metabolites, while viruses could change the metabolic profiles of the infected cells. Here, we used an unbiased metabolomics approach to profile metabolites in spleens and blood of murine leukemia virus-infected mice monocolonized with Lactobacillus murinus to show that viral infection significantly changes the metabolite profile of monocolonized mice. We hypothesize that these changes could contribute to viral pathogenesis or to the host response against the virus and thus open a new avenue for future investigations.

Keywords: commensal bacteria; metabolites; microbiota; retroviruses.

Figure 1

Flow chart illustrating the processing of sera and spleen samples for metabolomics. Mass spectrometry for serum samples was conducted in two separate procedures (A,B). Therefore, unknown metabolites could not be matched between the experiments and were removed. Spleen (C) samples were subjected to mass spectrometry under positive and negative ionization modes. Metabolites within a cohort of mice with a relative standard deviation (RSD) > 30% between technical replicates were discarded, leaving the indicated number of known and unknown metabolites.

Figure 2

MuLV infection and bacteria colonization alter the metabolomic profile in the sera and spleen. (A, C) PCoA plot of all metabolites within the sera (A) or spleen (C) of the indicated mice. Each point is an individual mouse. (B,D) Heatmaps of metabolite abundances in the sera (B) or spleen (D) of the four groups of mice. Each row is a metabolite.

Figure 3

Abundances of certain metabolites is altered by colonization and MuLV infection. (A,C) Venn diagrams illustrating the numbers of metabolites significantly altered by colonization, infection, or both in the sera (A) or spleen (C). (B,D) Heatmaps of identified metabolites, clustered by chemical type, whose abundances are influence by colonization, MuLV infection, or both in the sera (B) or spleen (D). Normalized values for each group were averaged and constructed into a heat map where red color indicates high metabolite levels, and blue color indicates lower metabolite levels. p-values were adjusted using Bonferroni correction [27] followed by two-way ANOVA to identify metabolites influenced by two independent variables: the presence of L. murinus and the virus. Each row is a metabolite.

Figure 4

Intestinal permeability is unaffected by MuLV infection. FITC permeability assay was used to determine the permeability of the intestines in SPF BALB/cJ infected and uninfected mice. Each data point is an individual plasma sample. n, number of mice used. p-value calculated using unpaired t-test. ns, not significant.