Fibrotic response as a distinguishing feature of resistance and susceptibility to pulmonary infection with Mycobacterium tuberculosis in mice

Abstract

The differential susceptibility of inbred mouse strains DBA/2J (susceptible) and C57BL/6J (resistant) to pulmonary tuberculosis following aerosol infection is under complex genetic control. In this report, transcriptional profiling with RNAs from Mycobacterium tuberculosis-infected lungs was used to investigate the physiological response, cell type, and biochemical pathways underlying differential susceptibility to infection. Statistical analysis of cDNA-based microarrays revealed that 1,097 transcripts showed statistically significant changes in abundance (changes of > or = 1.5-fold) in at least one of four experimental group comparisons (C57BL/6J [day 0] versus DBA/2J [day 0] mice, C57BL/6J [day 90] versus DBA/2J [day 90] mice, C57BL/6J [day 90] versus C57BL/6J [day 0] mice, or DBA/2J [day 90] versus DBA/2J [day 0] mice). A group of genes showing very high degrees of significance (changes of > or = 2.0-fold) displayed enrichment for transcripts associated with tissue remodeling and the fibrotic response. The differential expression of fibrotic response genes (Sparc, Col1a1, Col1a2, Col4a1, and Col4a2) in the infected lungs of the two mouse strains was validated by another microarray platform (Affymetrix oligonucleotide chips) and by reverse transcription-PCR. Furthermore, the differential expression of additional genes known to be associated with fibrosis (Mmp2, Timp1, and Arg1) was also validated by these approaches. Overall, these results identify the differential fibrotic response as a pathological basis for the high susceptibility of DBA/2J mice to pulmonary tuberculosis.

FIG. 1.

Replication of M. tuberculosis in the lungs of C57BL/6J and DBA/2J mice. (A) Resistant C57BL/6J and susceptible DBA/2J mice were infected via the aerosol route with 10² M. tuberculosis H37Rv bacilli, and the number of M. tuberculosis bacilli were enumerated in the lungs (log₁₀ CFU) at 90 days postinfection. Horizontal bars represent means of CFU counts for each group. (B) Macroscopic examination of the lungs of C57BL/6J and DBA/2J mice 90 days after aerosol infection. (C) High-power micrographs of day 90 lung lesions of C57BL/6J and DBA/2J mice, showing the presence of a much larger number of acid-fast bacilli in the DBA/2J lesion. The C57BL/6J lesion is populated predominantly by macrophages, whereas the DBA/2J lesion is populated predominately by neutrophils undergoing degeneration.

FIG. 2.

Transcriptional response in the lungs following M. tuberculosis infection. (A) A closed-loop strategy was applied to monitor differences in transcript abundance between lungs (n = 6) from either C57BL/6J or DBA/2J strains either prior to (t = 0) or 90 days following infection with M. tuberculosis. Each arrow represents two microarray hybridizations, including dye swaps. The fluorescence ratios between lung pairs were averaged prior to analysis. (B) The transcriptional profiles from each lung comparison were used in a PCA to separate them in a two-dimensional space according to the similarities in their most discriminating profiles. (C) A total of 1,097 genes with statistically significant changes in transcript abundance of at least 1.5-fold were organized by two-dimensional hierarchical clustering and colored according to the difference in abundance between each lung pair (more abundant transcripts are shown in red, and less abundant transcripts are shown in green). Dendrograms are used to illustrate the profile similarities between groups of genes or lung pairs. Clusters of differentially expressed genes are labeled A to D (see Results).

FIG. 3.

Transcriptional response in the lungs following M. tuberculosis infection. (A) Scatter plot representing the average fluorescence ratios (day 90 over day 0) from pairwise hybridizations (n = 6) for DBA/2J (y axis) and C57BL/6J (x axis) mice (n = 12 microarrays). Transcripts significantly modulated by a factor of at least 2.0-fold (FDR, <0.05%) in C57BL/6J and DBA/2J mice are colored red and green, respectively. Transcripts modulated in both strains are colored yellow, and the genes that did not pass the statistical cutoff are colored gray. (B) The numbers of transcripts differentially expressed in the two mouse strains are depicted in a Venn diagram.

FIG. 4.

Semiquantitative RT-PCR analysis of differentially expressed genes involved in the fibrotic response. RNA samples (n = 3) from control and M. tuberculosis-infected lungs of C57BL/6J and DBA/2J mice were pooled and used for RT-PCR amplification, using the indicated PCR cycle numbers (16, 18, and 20). PCR products were separated by agarose gel electrophoresis, followed by Southern blotting and hybridization with the corresponding probes. (A) The intensities of the hybridization signals were quantitated using a phosphorimager. (B) Gapdh was used to standardize the mRNA levels of the target genes. (C) C57BL/6J and DBA/2J ratios of expression (day 90/day 0) for each target gene were established based on results from 18 PCR cycles. A ratio of expression with a relative unit of 1 indicates no gene modulation.

FIG. 5.

Semiquantitative RT-PCR analysis of Arg1, Mmp2, and Timp1. RNA samples were pooled (n = 3) and used for RT-PCR with various PCR cycle numbers (20, 22, 24, and 26). (A) Schematic representation of tissue remodeling regulation pathway. The expression of Arg1, Mmp2, and Timp1 was quantitated as described in the legend to Fig. 4 (B), and ratios of expression were calculated based on results obtained following 22 (Mmp2 and Timp1) and 24 (Arg1) PCR cycles (C). Gapdh was used to standardize the mRNA levels of the target genes (Fig. 4B). A ratio of expression with a relative unit of 1 indicates no gene modulation. (Panel A was adapted from reference with permission of the publisher.)