Intestinal bacteria induce a different profile of gene expression and immune response in macrophages that could be related to their tolerogenic potential

J Rodríguez-Ruiz1, FB Velhote2, C Youssif1, O Reina3, D Rossell3, J Gálvez4, A Celada1, J Xaus4, M Comalada1,4. 1Institute for Research in Biomedicine (IRB), Macrophage Biology Group 08028, Spain, 2University of Sao Paulo (USP), Pharmacology Department, Brazil, 3Institute for Research in Biomedicine (IRB), Biostatistics and Bioinformatics Unit 08028, Spain, 4CIBER-ehd, University of Granada, Pharmacology Department 18071, Spain

Background: The intestinal microbiota plays a crucial role in the maturation of the neonatal immune system and thus in the development of the immune tolerance phenomena. The “hygiene hypothesis” proposes that there is a dysregulation in the immune system due to the lack of exposure to microorganisms at early stages of life. Probiotic administration has been considered as a suitable strategy to counteract these immunological alterations. Considerable evidence shows that the activation of macrophages and dendritic cells is crucially dependent on the detection of conserved microbial motifs, the so-called pathogen associated molecular patterns (PAMPs), by pattern recognition receptors (PRRs), of which the family of TLRs are the most studied. However, relatively little is known about the capacity of these cells to discriminate between beneficial/commensal and potential pathogen bacteria, and thus to induce oral tolerance or the initiation of immune responses.

Aims: The objective of this work is to evaluate the immune and genetic response in macrophages to different bacterial strains in order to characterize a potential tolerogenic response against particular bacteria.

Methods: Bone marrow-derived macrophages (BMDM) from male balb/c mice (8 wk-old) were differentiated in the presence of M-CSF (Macrophage-Colony Stimulating Factor) during 7 days. Then, starved macrophages were stimulated with LPS (lipopolysaccharide) from Escherichia coli (10 ng/ml) or by 5x107 cfu/ml of several heat-inactivated bacterial strains for 5 h (for RNA expression analysis) or 24 h (for cytokine production). Then, RNA was extracted to perform full mice genome expression by Affymetrix Mouse Genome 430 2.0 PM array analysis and confirmed by Real Time-PCR. Arrays were processed in Bioconductor, using RMA (Robust Multi-array Average) background correction, quantile normalization and RMA summarization. Fold changes between samples were computed after MA mean and variance normalization using the GAM (Generalized Additive Models) method. An empirical Bayes partial density model was used to identify differentially expressed genes within a FDR (False Discovery Rate) of 5% and a |log2(Fold-change)| > 2.5, (that is, an up/down fold in expression of 5.65 or more). Moreover, supernatants of the cultures were analyzed for cytokine expression by Cytometric Bead Assay (CBA).

Results: The number of probes that were significantly modified by any bacteria strain is quite reduced, representing less than 3% of the total gene diversity, Surprisingly, no phylogenic relation could be done based on the gene response induced in macrophages, and no clear differentiation exists even between Gram positive and Gram negative bacteria, or between different bacterial families of the phyla Firmicutes. When the existence of a potential bacterial signature common for all bacterial treatments and LPS was evaluated, 138 probesets were down-regulated by all bacteria while 293 were up-regulated. Among them, genes involved in cytoskeleton reorganization, cell differentiation and senescence are usually repressed by all the bacterial strains tested, while those genes related to wound healing and bacterial defense are commonly up-regulated. In addition, a differential response was characterized by each bacterial strain which could be related to their tolerogenic potential effect based on the relative expression and secretion of cytokines such as TNFα, IL-10 or IL-12 or the expression of different chemokines, surface receptors and/or other immunological markers. Qualitative cytokine expression profile is shown in table 1.

Table 1: Qualitative profile of cytokine expression induced by bacterial strains

Conclusion: Independently of its phylogeny, different intestinal bacterial strains can induce diverse gene responses in macrophages (tolerogenesis vs activation) that could be related with their differential recognition by these cells.