The Role of MicroRNA-155 in Chlamydia muridarum Infected Lungs

Abstract

Our laboratory has investigated the role of an evolutionarily conserved RNA species called microRNAs (miRs) in regulation of anti-chlamydial protective immunity. MiRs including miR-155 expressed in specific immune effector cells are critical for antigen specific protective immunity and IFN-γ production. Using miR-155 deficient mice, and a murine pulmonary model for chlamydial infection, we report here 1) the effect of host miR-155 on bacterial burden, and 2) identify probable immune genes regulated by miR-155.

1. Introduction

Evidence within the last decade has firmly established that small, non-coding regulatory RNA, such as microRNA (miR) influence a diverse variety of functions[1], ranging from immunity [2, 3] to physiology of reproductive [4] in vivo. MiRs modulate gene function post-transcriptionally by binding to target gene mRNA to effectively reduce expression by blocking translation or enhancing mRNA degradation [5]. Recent reports indicate Chlamydia trachomatis (Ct) infection induced miRs, may affect epithelial-mesenchymal transition[6], or serve as potential biomarkers of Ct infection [7]. Single nucleotide polymorphisms in miRs and inflammatory genes have been linked to Ct associated disease susceptibility in humans [8]. This role of miRs and miR-processing machinery [9], in immunity and immune effector cell function, has been recently reviewed [10].

Using a murine model, Gupta et al, characterized host miR expression in the female genital tract following Chlamydia muridarum (Cm; murine adapted strain of Ct) infection [11]. In that study, miRs were observed associated in a spatio-temporal regulating inflammation and immunopathology, thus, contributing to anti-Cm immunity [11]. Additionally, Cm infection induced miR-214 was shown to regulate ICAM-1 reducing upper genital pathology in IL-17A^−/− mice [12]. More recently, Gupta and coworkers, have demonstrated following Cm infection, dendritic cells (DC)- miR-155 acts synergistically with miR-182 in Ag-specific CD4⁺ T-cell to regulate IFN-γ, the cytokine critical for anti-Cm/Ct immunity [13]. The role of miRs in Ct infection at other mucosal sites (non-genital) including the eye and lung is also documented [14, 15].

While studies collectively underscore the previously unappreciated role for miRs in Ct biology including pathogenesis and immunity [6, 7, 11, 14, 15], the lack of miR-deficient in vivo murine studies during the course of Ct infection is apparent. Thus, pursuing our previous miR-155 findings describing its contribution to protective immunity and IFN-γ production in genital Cm infection [13], we report here an in vivo assessment of the 1), contribution of miR-155 to chlamydial infectivity, and 2), miR-155 dependent immune gene expression alteration.

2. Materials and Methods

2.1. Mice

Male, four-eight-week-old wild-type mice (WT, 000664 C57BL/6J), and miR-155^−/− (Mir155^tm1.1Rsky/J) mice were purchased from Jackson Laboratory (Bar Harbor, ME). All mice were allowed to acclimate to their new surroundings for 1 week. All animal experiments were conducted in compliance with an approved protocol (MU012) issued by The Institutional Animal Care and Use Committee (IACUC) of The University of Texas at San Antonio.

2.2. Chlamydia muridarum intranasal challenge and bacterial assessment

Cm seed stocks were propagated in HeLa 229 cells. At 24 h post infection, cells were mechanically disrupted and following high-speed centrifugation, bacterial pellets were purified on a Renografin gradient as previously described [12]. The same Cm seed stock was used throughout this study.Mice were intranasally infected with 2 × 10³ inclusion forming units (IFUs) of Cm in 40 μl Sucrose phosphate glutamate (SPG) buffer. Body weight was monitored daily. At day 3 and 9 post inoculation, mice were sacrificed following administration of light anesthesia (isoflurane) according to approved protocol, and lung tissue was aseptically removed. Lung tissue was homogenized in 500 ul cold Eagle’s Minimum Essential Medium (EMEM) buffer. A portion of the tissue lysate was subjected to assessment of bacterial burden using infected McCoy cells, and immunofluorescence [13].

2.3. RNA extraction and quantitative real-time PCR

Total RNA was extracted from lung homogenate at day 9 post infection using either a miRNeasy RNA extraction Kit (Qiagen) or Aurum Total RNA Fatty and Fibrous Tissue Kit (Bio-Rad). RNA quantitation was assessed using a Nanodrop Spectrophotometer (ThermoScientific). RNA samples (1ug starting quantity) exhibiting A_260/280 values 1.8–2.0 were converted to cDNA using a miScript II RT Kit for miRNA (Qiagen), and iScript Advanced cDNA Synthesis Kit for gene expression (Bio-Rad) [12], or manufacturer’s instructions for IFN-γ signaling pathway array (Bio-Rad). A MiScript primer assay (for miR-155 expression levels) in lung tissue homogenates (Qiagen) was used as described[11]. PCR Amplification was carried out using the respective primers: miR-155 (Qiagen), single PCR gene primers (BioRad) or IFN-γ signaling pathway array (Pre-spotted primers in plate wells). Commercial control assays and synthetic templates were used for template and PCR quality-control. These included assays for DNA contamination, positive PCR reaction, RNA quality, reverse transcription control (BioRad) and performed as expected per manufacturer’s directions. PCR was carried out using a BioRad CFX96 Touch Real-Time PCR Detection System (Bio-Rad) and conditions in brief were- miRNA: Amplification conditions were miRNA: (95°C / 15 min and 40 cycles of 94 °C/ 15 s, 55 °C/ 30 s, 70°C/ 30 s). Gene: 95 °C/ 2 min and 40 cycles of 95 °C/ 5 sec, 60 °C/ 30 sec). Quantitation-normalization of gene expression was achieved using CFX Maestro software (Bio-Rad). MiR-155 expression was quantitated-normalized using RNU6 and Snord68 housekeeping targets; whereas, housekeeping genes (Gapdh & HSP90- for single PCR and Hprt &Tbp – for gene array) were used for determination of expression of all other gene targets. All miR and gene expression were reported as fold change differences normalized to control group which were WT mock infected mice and was determined using 2^{(−Average ΔΔCq)} 113 method [12].

2.4. IFN-γ ELISA

Assessment of IFN-γ production in lung tissue lysate supernatant was analyzed using BD OptELISA kits (BD Pharmingen, NJ, USA) as previously described [16]. Absorbance at 405nm and 560nm was measured using an ELISA microplate reader (Biotek Instruments, Winooski, VT).

2.5. Statistical analyses

Statistical analyses were performed using Prism 8 Software (GraphPad, La Jolla, CA, USA). Student’s t-test was used for comparison between two groups whereas for multiple group comparisons, 2-way ANOVA with Tukey’s post hoc test (parametric distribution) or Kruskal-Wallis post hoc test (nonparametric distribution) were used. Differences were considered statistically significant if P < 0.05.

3. Results

3.1. MiR-155 is regulated and alters disease progression

Prompted by our previous study using an intravaginal Cm infection model [13], we were interested in determining miR-155 involvement following intranasal Cm infection. Therefore, 4–6-week-old C57BL/6 (wild type-WT) mice were infected intranasally with 2 × 10³ Cm IFUs (Fig. 1a). A slight reduction in body weight was observed (Fig. 1b, supplementary table 1a). At day 9 post infection, mice were euthanized, and lung tissue evaluated for miR-155 expression using quantitative real-time PCR. As shown in Fig. 1a, miR-155 expression levels were significantly elevated in infected WT lung when compared to non-infected WT lung (15.8 ± 1.8), respectively (p<0.0001).

Figure 1. C. muridarum pulmonary infection results in increased expression of miR-155 and affects bacterial burden.

Four 8 week-old wild-type (WT, 000664 C57BL/6J), and miR-155^−/−(Mir155^tm1.1Rsky/J) mice were infected intranasally with C. muridarum (2 × 10³ IFU) or PBS (mock). At indicated time points, lung tissue was collected, homogenized, and miR-155/ bacterial burden quantitated as described under ‘Materials and Methods’. For miR-155 PCR, 1ug of RNA was converted to cDNA, and RNU6 and Snord68 were used as housekeeping miRs for miR-155 PCR assay. (a) Day 9 (D9) post Cm infection miR-155 expression in WT tissue (mean value for 2 different experiments) was observed to be significantly increased. * P<0.05 for miR-155 expression in WT Cm (red box) infected compared to WT mock (blue box) (b) Body weight change as a function of time demonstrated a significant difference between WT and miR-155^−/− mice following Cm infection (mean value for 6 different experiments). At indicated time points P<0.05, ** WT Cm infected (red line) compared to WT mock infected (blue line); Φ miR-155^−/− Cm infected (green line) compared to WT Cm infected (red line). (c) Lung bacterial burden in WT and miR-155^−/− mice at D9 and not D3 post infection (mean value for 2 different experiments). * P<0.05 bacterial burden in miR-155^−/− Cm infected (green box) compared to WT Cm infected (red box); For P value, student’s t-test was used for comparison between two groups whereas for multiple group comparisons, 2-way ANOVA with post hoc tests were used. P<0.05 was considered statistically significant.

In order to determine the contribution of miR-155 in the infection process, we next infected miR-155^−/− and WT mice intranasally with 2 × 10³ Cm IFU. Using weight as a measure of fitness, in Fig 1b, we observed no significant difference in weight reduction in WT and miR-155^−/− mice up to day 4 post infection or between WT and miR-155^−/− mock infected during the entire course of the experiment. In contrast, WT mice body weight appeared to stabilize by day 6 post infection while miR-155^−/− mice exhibited continued weight loss (p<0.0001). Due to continued significant body weight loss in miR-155^−/− mice, the experiment was terminated at day 9 post infection. Continued weight loss out to day 9 post infection in miR-155^−/− mice suggests systemic changes compared to the WT group since little difference in bacterial burdens of either group was observed at day 3 post infection in contrast to day 9 showing a 3.7-fold increase (WT mean value = 7.3 × 10⁶) vs miR-155^−/− (mean value = 2.7 × 10⁷, p<0.0039) (Fig. 1c).

3.2. Immune gene expression is significantly altered in C. muridarum infected miR-155 deficient lung tissue

Given that miR-155 is a master regulator of immunity [17], we next investigated a subset of genes implicated in anti-chlamydial immunity. SOCS-1 gene expression (fold change) was significantly increased in miR-155^−/− infected mice compared to WT infected mice (p<0.0001, Fig. 2a). Although miR-155^−/− infected mice exhibited increased SOCS-1 gene expression, a marked decrease in miR-155^−/− ICAM-1 gene expression from 4.7 to −0.1 (p<0.05) was observed (Fig. 2b). Similarly, a 4.6 and 4.9-fold reduction in CXCL10 and IL-33 gene expression, respectively, was observed post Cm infection (p<0.05, Fig 2c and d). Like CXCL10, CXCL-1 and −2 gene expression was reduced 1.5- (p<0.0001) and 3.9-fold (p<0.0001), respectively, in the absence of miR-155 (Fig 2e and f). Gene expression in WT and miR-155^−/− Cm infected groups were normalized to WT mock infected mice. Gene expression changes (Figs. 2 a–f) in WT and miR-155^−/− mock infected groups were statistically insignificant thereby indicting the changes were probably due to the effect of Cm infection in a miR-155^−/− environment and not just due to the miR alone.

Figure 2. Expression of immune gene(s) following C. muridarum infection in the lung.

Lung tissue was collected at day 9 post infection from WT and miR-155^−/− Cm infected mice. Tissues were homogenized and following RNA extraction, 1ug of RNA was converted to cDNA, and gene expression assessed by quantitative real-time PCR. Housekeeping genes (Gapdh & HSP90) were used for determination of expression of gene targets. Gene expression of (a) Socs-1 (Suppressor Of Cytokine Signaling 1); (b), ICAM-1 (Intercellular Adhesion Molecule 1); (c), CXCL-10 (C-X-C motif chemokine 10); (d), IL-33 (Interleukin- 33); (e), CXCL-1 and (f), CXCL-2 revealed significant regulation in miR-155^−/− Cm infected (black bars) mice compared to WT Cm infected mice (open bars). Respective graphs are representative of 2 different experiments (n = 3 mice/group). Gene expression of all groups was normalized to WT mock infected lungs. Gene expression changes were comparable and statistically insignificant in between WT and miR-155^−/− mock infected groups. ANOVA with post hoc tests were used and P<0.05 was considered statistically significant.

3.3. IFN-γ is significantly altered in C. muridarum infected miR-155 deficient lung tissue

Similar to our previous findings on overexpression of miR-155 positively correlating with production of IFN-γ [13], we observed Cm infection of WT lung significantly elevated IFN-γ production (Fig 3a–b) with concomitant overexpression of miR-155 (Fig. 1a). Importantly, in line with our previous report [13], lack of miR-155 results in significant elevation of IFN-γ production in Cm infected miR-155^−/− compared to WT Cm infected lungs (Fig 3b). We additionally investigated probable immune genes contributing to IFN-γ signaling pathway(s), and performed transcriptomics analyses in lung tissue, using an IFN-γ PrimePCR signaling pathway commercial array in WT and miR-155 ^−/− lungs (Fig 3c). Clusterogram analyses revealed close alignment of similarly expressed genes such as Cbt and IRF-1 or Pik3cd and Jak2 amongst others in WT and miR-155^−/− Cm infected lungs indicating probable similar functional roles in vivo (Fig 3c). Biological variations due to below-detection level expression of certain genes including brca1, ifngr2, cbt, plcg2, akt2, akt3 and makp3 was observed in Cm infected mice (Fig 3c). These variations were summarized in graphical (Fig 3c) and tabular formats (supplementary table 1b). Following Cm infection in WT lungs, genes such as brca1, cbt, icam1, ifngr1, ifngr2, irf1, jak2, myc or calm 2, calm3 and map2k6 were observed to be either upregulated or downregulated respectively compared to uninfected WT mice (Fig 3d). Interestingly, a greater proportion of genes regulated in miR-155^−/− lungs compared to WT lungs following Cm infection were down-regulated in the absence of miR-155(Fig 3d, supplementary table 1b). These genes included akt1, akt2, clam1, clam3, camk2g, crebbp, jak1, jak2, mapk1 and mapk14. Overall, in conclusion, our gene-wise clusterogram heat-map (Fig 3c) and graphs (Fig 3d, supplementary table 1b) displayed differential profiling of IFN-γ signaling related immune genes in miR-155 deficient lungs compared to WT (155 sufficient) lungs.

Figure 3. C. muridarum pulmonary infection results in altered IFN-γ expression in miR-155^−/− lung tissue.

Four-8 week-old wild-type (WT, 000664 C57BL/6J), and miR-155^−/−(Mir155^tm1.1Rsky/J) mice were infected intranasally with C. muridarum (2 × 10³ IFU) or PBS (mock). At day 9 lung tissue was collected, homogenized, and IFN-γ quantitated as described under ‘Materials and Methods’. Housekeeping genes (Gapdh & HSP90- for IFN-γ PCR and Hprt &Tbp – for gene array) were used for determination of expression of all gene targets.

(a) Lung INF-γgene PCR; (b), Lung INF-γ cytokine ELISA (mean value for 2 different experiments) revealed a significant increase in expression/ secretion respectively of IFN-γ in WT Cm infected mice (yellow bar) compared to WT mock infection (Maroon (* P<0.05). However Cm infected mir-155^−/− mice (green bar) produced significantly higher lung INF-γ in comparison to miR-155^−/− mock infected (light blue bar) (** P<0.05) and upon comparison to WT Cm infected mice (yellow bar) (§ P<0.05); (c),Clusterogram analyses of IFN-γ signaling pathway related genes identified using a commercial array as described under Materials and Methods’. Results of biological replicates (M1-M3- mouse 1–3 for each group revealed differential regulation of genes upon comparison of WT and mir-155^−/− mice following Cm Denotes lack of heat indices for respective gene(s) in biological sample(s) due to non-detection of melt curve above the threshold of detection in PCR run analyses. (d), Graphical representation of relative normalized expression of clusterogram of IFN-γ signaling pathway genes in WT and mir-155^−/− Cm and mock infected mice in comparison to WT mock infected mice revealed significantly altered profiles in experimental groups. For P value calculations, student’s t-test was used for comparison between two groups whereas for multiple group comparisons, ANOVA with post hoc tests were used. P<0.05 was considered statistically significant.

4. Discussion

Chlamydia sp is a leading cause of sexually transmitted infections worldwide, and its ability to infect multiple mucosal surfaces makes it an ideal pathogen to study organ specific immune response. The host’s ability to differentially regulate multiple targets through processes including phosphorylation, histone modification, and protein expression enables the immune system to orchestrate a rapid, defensive cascade in response to an invading pathogen such as Ct [18].

The immune-modulatory potential of miR-155 in inflammation and immunity is well-documented [19]. Reports on miR-155 following Ct infection via genital [7, 13] and ocular routes[14], highlights the importance of miR-155 at multiple mucosal surface(s) naturally infected by Ct. To date, our previous study remains the only report describing the role of miR-155 in antigen (Ag)-specific cross-talk between effector populations and IFN-γ production in a gram negative, intracellular pathogen namely, Cm [13]. The current study provides additional insight into the in vivo role of miR-155, and its disruption of the inflammatory immune milieu following Cm infection.

For this study, we used a previously well-established, characterized model of Cm infection [15, 20, 21]. Respiratory (intranasal) infection with Cm in the lung (originally called MoPn or the causative agent of mouse pneumonitis) [22], presented us with an additional tissue/ site [23], to study miR-mediated mucosal immunity against Cm [11, 12]. It offered us an opportunity to extend our miR-155 results on Ag-presenting dendritic cells, and induction of local IFN-γ secretion by Ag-specific CD4⁺ T cells following genital mucosae infection[13]. Importantly, similar to genital infection, our previous neonatal-pulmonary Cm studies reiterated the role of IFN-γ and/or IFN-γ signaling in abrogating infection and affecting tissue pathology [16]. This further encouraged us to extend our genital tract-miR findings to the lung, especially in light of limited reports on Cm associated miRs at this tissue site [15].

Our results indicate that miR-155 regulates SOCS-1 expression, and the absence of miR-155 leads to increased SOCS-1 gene expression [24]. Similar to previous reports, we observed increased expression following Cm pulmonary challenge in miR-155^−/− mice (Fig. 2a). SOCS-1 inhibits the IFN-γ pathway through Jak1 and Jak2 [25]. Differential expression of these IFN-γ pathway genes was noted following pulmonary challenge with Cm in miR-155 deficient mice (Fig 1d). As previously shown in endothelial hCMEC/D3 cells, inhibition of miR-155 leads to decreased ICAM-1 under normal culture conditions [26]. In the absence of miR-155, Cm pulmonary challenged mice exhibited decreased ICAM-1 gene expression (Fig. 2b) [12]. Additionally, we observed regulation of CXCL-10/ interferon gamma-induced protein 10 (IP-10) gene expression in miR-155^−/− mice (Fig. 2c) [27]. MiR-155’s over-expression in Cm infected WT mice (Fig. 1a) was concomitant with increase in IFN-γ (Fig. 3a, b), and SOCS-1 (Fig. 2a), CXCL-10 expression in these tissues (Fig. 2b), similar to previous reports on signaling interaction(s) of these molecules in vivo [28, 29], or in primary cultures of immune cells [30, 31], respectively. Interestingly, in contrast, Cm infected miR-155^−/− lungs demonstrated upregulation of IFN-γ (Fig. 3a, b), with an upregulated SOCS-1 (Fig. 2a) and reduced CXCL-10 expression (Fig. 2b). These miR-155^−/− findings indicated that additional gene targets (Fig. 3c, d, supplementary table 1b) may be involved in a SOCS-1-CXCL-10-independent pathway to regulate IFN-γ production in a miR-155 null background. Additionally, it is known that miR-155^−/− mice lack functional regulatory T cells (T_reg) [32], and that miR-155 regulates T_reg fitness via direct regulation of SOCS-1[33]. Given that, upon Cm infection, Ag-specific T_reg cells promote Th₁₇ differentiation[34], and Th₁₇ in turn promotes Th ₁and IFN-γ production [35], miR-155’s role in immune effector cell homeostasis requires investigation. Moreover, to this end, the role of additional miR(s) from immune cells involved in cognate interaction resulting in Ag-specific anti-Cm immunity and in turn affecting IFN-γ signaling [13], needs consideration. Taken together, these mechanisms may constitute a potential ‘work- around established gene targets’ by the host in pulmonary IFN-γ regulation in miR-155^−/−cells (Fig 3c). Along with IFN-γ regulatory genes, we observed inhibition of IL-33 expression with concomitant decreased levels of CXCL-1 and −2 (Figs. 2d–f) [36, 37]. Removal of miR-155 following Cm pulmonary infection ultimately leads to a compromised ability for the host to mount an effective immune response in the lung and result in increased bacterial burdens (Fig. 1 c).

While our report provides evidence for a promising in vivo role for miR-155 previously implicated by us and others to be induced in humans and laboratory animals, the experimental design reported here presents with a few noteworthy limitations. Our observations on immune gene regulation in miR-155 sufficient and deficient tissues is based upon a selected thus limited, or rather an incomplete list of targets (Fig. 2). With bioinformatic evidence for only SOCS-1 (Fig. 2a) being directly regulated by miR-155, regulation of other genes (Figs 2b–f) may be due simply to an unknown ‘bystander’ effect, i.e., happenstance in a miR-155 deficient/ mutants infected with Cm, and not the direct target of miR-155. Further investigation into the direct or indirect role of miR-155 regulation of these immune genes would require miR-gene ‘gain or loss’ of target function studies (Fig. 2, Fig 3c). Additionally, our findings are in total lung tissue homogenates which contain a conglomerate of Cm infection induced immune effector cells post infection. Comparative analyses of miR-155 mediated target genes (downselected from Figs. 2, 3c and supplementary table 1b) within such cells is warranted. However, despite the limitations of the current study’s design, our findings strongly imply an important role for miR-155 in anti-Cm/ Ct immunity.