Macrophages Infected by a Pathogen and a Non-pathogen Spotted Fever Group Rickettsia Reveal Differential Reprogramming Signatures Early in Infection

Abstract

Despite their high degree of genomic similarity, different spotted fever group (SFG) Rickettsia are often associated with very different clinical presentations. For example, Rickettsia conorii causes Mediterranean spotted fever, a life-threatening disease for humans, whereas Rickettsia montanensis is associated with limited or no pathogenicity to humans. However, the molecular basis responsible for the different pathogenicity attributes are still not understood. Although killing microbes is a critical function of macrophages, the ability to survive and/or proliferate within phagocytic cells seems to be a phenotypic feature of several intracellular pathogens. We have previously shown that R. conorii and R. montanensis exhibit different intracellular fates within macrophage-like cells. By evaluating early macrophage responses upon insult with each of these rickettsial species, herein we demonstrate that infection with R. conorii results in a profound reprogramming of host gene expression profiles. Transcriptional programs generated upon infection with this pathogenic bacteria point toward a sophisticated ability to evade innate immune signals, by modulating the expression of several anti-inflammatory molecules. Moreover, R. conorii induce the expression of several pro-survival genes, which may result in the ability to prolong host cell survival, thus protecting its replicative niche. Remarkably, R. conorii-infection promoted a robust modulation of different transcription factors, suggesting that an early manipulation of the host gene expression machinery may be key to R. conorii proliferation in THP-1 macrophages. This work provides new insights into the early molecular processes hijacked by a pathogenic SFG Rickettsia to establish a replicative niche in macrophages, opening several avenues of research in host-rickettsiae interactions.

Keywords: Rickettsia conorii; Rickettsia montanensis; host-pathogen interactions; macrophages; spotted fever group Rickettsia; transcriptional profiling.

Figure 1

SFG Rickettsia trigger reprogramming in THP-1 macrophages early in infection. (A,B) Volcano plots of log₂ fold change ratio of the expression levels in R. conorii- (A) and R. montanensis- (B) infected THP-1 macrophages over that in uninfected cells plotted against the -log₁₀ (q-value). Statistically differentially expressed genes found in increased abundance and in decreased abundance are represented in red and green, respectively (FDR < 0.05). See also Tables S1, S2. (C,D) Validation of RNA-seq data by comparing the transcriptional fold changes of 11 randomly selected genes determined by RNA-seq and an independent method (q-RT-PCR) for R. conorii- (C) and R. montanensis- (D) infected cells. Pearson analysis of correlation and respective significant test (two-tailed) were performed in GraphPad Prism. R. conorii-infected: r = 0.9208, N = 11, p < 0.0001 and R. montanensis-infected: r = 0.6437, N = 11, p < 0.05. Gene labeling: 1-B2M; 2-BTG2; 3-CD69; 4-EGR1; 5-EMC7; 6-G6PD; 7-IER3; 8-KLF10; 9-MTRNR2L6; 10-OTUD1; 11-PP1R15A. See also Table S3.

Figure 2

Gene expression patterns stimulated by infection of THP-1 macrophages with R. conorii or R. montanensis reveal a more robust modulation by the pathogenic species. (A) Venn diagram depicting the number and distribution of specific and common DE genes in each experimental condition. UP means increased abundance, DOWN means decreased abundance, RC is R. conorii-infected cells and RM is R. montanensis-infected cells. See also Table S4. (B) Prediction of the activation/inhibition state of the top 30 canonical pathways in R. conorii-infected cells (R.c.) ranked by -log(p-value) and corresponding prediction states in R. montanensis-infected cells (R.m.) according to Ingenuity Pathway Analysis (IPA). Pink colored heatmap shows the top 30 canonical pathways for R. conorii-infected cells (R.c.) (ranked by -log(p-value) and the respective -log(p-value) for the correspondent pathway in R. montanensis-infected cells (R.m.). Red-green heatmap shows the prediction of activation (red)/inhibition (green) state (Z-score) in R. conorii-infected cells (R.c.) and R. montanensis-infected cells (R.m.). Pathways are considered to be inhibited or activated for Z-scores values < −2.0 or > 2.0, respectively. See also Table S4.

Figure 3

Rickettsia conorii and R. montanensis differentially modulate innate immune responses during THP-1 macrophage infection. Combined list of the individual DE genes (and respective log₂ fold change values) categorized with the GO term inflammatory response (GO:0006954) and KEGG pathways: TLR signaling pathway (hsa04620), NF-κB signaling pathway (hsa04064), and TNF signaling pathway (hsa04668). DE genes in THP-1 macrophages infected with R. conorii are shown in black bars and in R. montanensis are shown in blue. The absence of bar means that the fold change of that gene for the respective experimental condition was not considered statistically significant. See also Table S6.

Figure 4

Rickettsia conorii switches macrophage immune responses into a hyporesponsive state. Quantification of TNFα concentration in the culture media of uninfected (white), R. conorii- (black) and R. montanensis-infected (blue) THP-1 macrophages upon stimulation with E. coli O26:B6 LPS. Results are shown as mean ± SD and differences were considered non-significant (ns) at P > 0.05 or significant at **P ≤ 0.01, ***P ≤ 0.001.

Figure 5

Rickettsia species differentially modulate the expression of several host apoptotic genes early in infection of THP-1 macrophages. Log₂ fold change values of DE genes categorized with GO term “negative regulation of apoptotic process” (GO:0043066) in R. conorii- (black) and R. montanensis-infected (blue) cells. Absence of bar means that the fold change of that gene for the respective experimental condition was not considered statistically significant. See also Table S9.

Figure 6

Rickettsia conorii modulates host cell apoptosis during infection. (A) Percentage of cleaved PARP-positive cells, a marker of intrinsic apoptosis, over the course of infection of THP-1 macrophages with R. conorii. Results are shown as the mean ± SD and differences were considered non-significant (ns) at P > 0.05 or significant at ^*P < 0.05. (B) Immunofluorescence microscopy of uninfected cells and THP-1 macrophages infected with R. conorii at 1, 3, and 5 days post-infection. Cells were stained with DAPI (blue) to identify host nuclei, mouse anti-Rickettsia antibody (5C7.31) followed by anti-mouse Alexa Fluor 594 (red) to identify R. conorii and rabbit anti-cleaved poly(ADP-ribose) polymerase (PARP) followed by anti-rabbit Alexa Fluor 488 (green) to identify cleaved PARP-positive cells. Scale bar = 10 μm.

Figure 7

Rickettsia conorii inhibits staurosporine-induced death of THP-1 macrophages at 5 days post-infection. (A) Uninfected and R. conorii-infected THP-1 macrophages at 1, 3, and 5 days post-infection were treated with staurosporine (750 nM) for 4 h. The percentage of cleaved PARP-positive cells in each experimental condition is shown as the mean ± SD and differences were considered ns (non-significant) at P > 0.05 or significant at ^*P < 0.05. (B) Immunofluorescence microscopy of uninfected cells and THP-1 macrophages infected with R. conorii at 1, 3, and 5 days post-infection upon challenge with staurosporine (750 nM) for 4 h. Cells were stained with DAPI (blue) to identify host nuclei, mouse anti-Rickettsia antibody (5C7.31) followed by anti-mouse Alexa Fluor 594 (red) to identify R. conorii and rabbit anti-cleaved poly(ADP-ribose) polymerase (PARP) followed by anti-rabbit Alexa Fluor 488 (green) to identify cleaved PARP-positive cells. Scale bar = 10 μm.

Figure 8

Rickettsia species differentially modulate the expression of several gene expression regulators early in infection of THP-1 macrophages. (A) Distribution of DE non-coding RNAs according to their category in R. conorii- (black) and R. montanensis- (blue) infected cells. scaRNAs (small Cajal body-specific RNAs), snoRNAs (small nucleolar RNAs), 5S-rRNAs (5S ribosomal RNAs), U-RNA (small nuclear RNAs), 7SL RNAs (signal recognition particle RNAs), 7SK RNAs (7SK small nuclear RNAs), lincRNAs (long intergenic noncoding RNAs), miRNAs (microRNAs). Number of genes for each orientation [increased abundance (UP) or decreased abundance (DOWN)] is represented in each bar. See Table S10 for details about individual transcripts. (B) STRING analysis of DE genes in R. conorii-infected cells categorized in the GO term “positive or negative regulators of transcription from RNA polymerase II promoter.” Nodes corresponding to DE genes categorized with transcriptional activator activity (GO:0001228) are in red and with transcriptional repressor activity (GO:0001227) are in blue. See also Table S11.