MiR-4792 regulates inflammatory responses in Cryptococcus neoformans infected microglia

Background: Investigating the factors that influence inflammatory response of microglial cells is important to understand the pathogenesis of cryptococcal meningitis (CM). MicroRNA (miRNA) have been shown to play an important role in inducing host defenses and activating immune response in the process of microbial infection; however, the regulatory mechanisms of miRNAs in cryptococcal meningitis are poorly defined. In our previous analysis, we assessed the miRNA profiles of BV2 cells following Cryptococcus neoformans (C. neoformans) infection. In this study, we characterized the expression of miR-4792 in CM patients to further our understanding of the host response to pathogen infections. Results: miR-4792 was downregulated in BV2 cells infected with C. neoformans while its target gene EGFR was upregulated. Infected cells with up-regulated miR-4792 exhibited decreased EGFR expression, reduced MAPK signaling and a decreased secretion of inflammatory cytokines. Following antifungal treatment in cryptococcal meningitis patients, the levels of miR-4792 in the CSF significantly increased, while the expression of EGFR significantly decreased. Conclusion: This study identified that miR-4792 and its target EGFR regulate the secretion of inflammatory cytokines in C. neoformans infected BV2 cells. This furthers our knowledge of the host immune responses to fungal infections in the CNS.

privileged and isolated from other peripheral organs due to separation by the blood-brain barrier (BBB). Microglial activation enhances innate immune response, enhancing the phagocytosis of invasive bacteria or fungi, thus protecting neuronal cells [3][4]. Assessment of the relationship between macrophage, innate immunity and C. neoformans can enhance our understanding of CM pathogenesis.
miRNAs are highly conserved small non-coding RNAs 18-22 nucleotides in length [5]. miRNAs regulate gene expression through binding to the 3′UTR of miRNAs that lead to a loss of gene expression through mRNA degradation or through preventing translation. Recently, a role for miRNAs in the immune response to fungal exposure has been documented. miR-21, miR-146, miR-132, miR-155, and the let-7 family regulate inflammatory responses following exposure to bacterial pathogens. The miRNA response to fungal exposure is comparable to that of inflammatory and allergic responses [6][7][8][9][10][11].
Approximately 50% of cryptococcal meningitis patients die within a year of infection due to a lack of successful therapy [12]. In the absence of effective anti-fungal agents, the immune response to fungal infection requires further understanding. The molecular mechanisms underlying these responses remain poorly defined for a range of fungal species. In this regard, knowledge of the miRNA response to pathogen infections can guide novel and effective approaches to antifungal therapeutics.

miR-4792 and EGFR expression in BV2 cells induced by C. neoformans (WM148)
It has been shown that miR-4792 is downregulated in nasopharyngeal carcinoma, uterine leiomyoma, during submucosal fibrosis and in glioblastoma-infiltrating CD14 + cells. In contrast, miR-4792 is upregulated in glioma microvesicles [13][14][15][16][17]. Our previous studies have shown that the expression of miR-4792 increases in THP-1 cells infected with C.neoformans (WM148) (Fig. 1a) [6]. To further study the expression of miR-4792 in microglia, BV2 cells were treated with WM148 for 0, 3, 6 and 9 and 12 h and qRT-PCR was performed to detect miR-4792 expression. We found that the expression of miR-4792 decreased over time, reaching its lowest after 6 h (Fig. 1b), while the expression of EGFR gradually increased, peaking at 6 h (Fig. 1c). ELISA assays showed that the inflammatory factors TNF-α and IL-6 secreted by microglia gradually increased over time, while IL-1β levels showed no obvious changes (Fig. 1d).
EGFR is targeted by miR-4792.
To further define the molecular mechanisms governing the regulatory effects of miR-4792, we performed targeted bioinformatics to investigate novel miR-4792 targets using MIRADA and target scan. These analyses revealed EGFR as miRNA target, the sequence of which was assumed to be in the 3'UTR (Fig. 2a). We additionally performed western blot analysis to assess the expression of the Phosphorylation of EGFR (pEGFR) in BV2 cells. We found that miR-4792 overexpression significantly reduced pEGFR levels, and dual luciferase assays confirmed EGFR as a direct target of miR-4792 ( Fig. 2b). The exogenous overexpression of miR-4792 significantly inhibited WT EGFR 3'-UTR activity, but had minimal effect on mutant EGFR 3'UTR sequence (Fig. 2c). These data strongly implicate EGFR as a direct target for miR-4792.

EGFR blockade inhibits WM148 induced microglia activation
Microglia regulate the innate immune responses of the CNS [18,19]. Given that microglia are responsible for pro-inflammatory cytokine production [20], we investigated the effects of WM148 on microglia induction. Western blot analysis was performed to assess the expression of the cell surface molecule CD11b, considered a phenotypic marker of microglial activation [21]. The data showed that, after 6 h of WM148 treatment, higher levels of CD11b expression and EGFR activation (evident through the enhanced levels of pEGFR) occurred in treated BV2 cells (Fig. 3a). In contrast, 24 h treatment with EGFR inhibitor (AG1478) led to a loss of CD11b and pEGFR expression (Fig. 3b) confirmed by western blot analysis. Interestingly miR-4792 expression inversely correlated with EGFR levels, further highlighting EGFR as a miR-4792 target (Fig. 3c).
Inhibiting EGFR prevents MAPK kinase mediated cytokine production in BV2 cells Activated microglia stimulate neuronal inflammation and enhance the levels of proinflammatory cytokines including IL-1β and TNF-α in the CNS [22]. Extracellular signal-regulated kinases (Erk1/2), c-Jun terminal kinase (JNK) and p38 MAPK are key cellular signaling cascades MAPKs are known to enhance the production and secretion of cytokines [23][24][25]. For example, MAPK signaling promotes the LPS-induced synthesis of IL-1β and TNF-α [26]. MAPK is also stimulated following EGFR activation [27,28]. Consistent with previous findings, we found that AG1478 treatment inhibited MAPK phosphorylation leading to a loss of ERK1/2, p38MAPK and JNK activation in WM148 treated BV2 cells, while no significant differences in p-JNK levels were observed (Fig. 4a). Similarly, AG1478 treatment inhibited both IL-1β and TNF-α production in WM148 infected BV2 cells (Fig. 4b). We simultaneously assessed the secretion of other inflammatory factors in WM148 infected BV2 cells. The results showed that IL-1α, IL-12, Eotaxin, GM-CSF, MCP-1/CCL2 levels decreased after AG1478 treatment compared to the control group (Fig. 4c), but the differences were not significant.

miR-4792 regulates the EGFR/MAPK axis in BV2 cells
We next investigated the effects of the miR-4792-mediated regulation of EGFR on ERK1/2, JNK and P38MAPK signaling in BV2 cells infected with WM148. As shown in Fig. 2b and Fig. 5a, the levels of active pEGFR and p-ERK1/2, p-p38 and p-JNK decreased in the presence of miR-4792 mimics compared to the WM148 group, and the levels of the inflammatory cytokines IL-1β and TNF-α also decreased (Fig. 5b). While there is no significant different when BV2 cells were treated with miR-4792 inhibitors. All findings were confirmed by qRT-PCR analysis in which the expression of miR-4792 significantly increased, while EGFR expression decreased following the transfection of miR-4792 mimics for 24 h. The opposite phenotype was observed for miR-4792 inhibitors (Fig. 5c). We additionally found that active pEGFR levels were highly expressed when miR-4792 expressed lowly following WM148 stimulation ( Fig. 1). Furthermore, we assessed the secretion of other inflammatory factors in WM148 infected BV2 cells, and found that the levels of pro-inflammatory factors including IL-1α, IL-12, Eotaxin, GM-CSF, MCP-1/CCL2 decreased after treatment with miR-4792 mimics for 24 h (Fig. 5d). Taken together, these data highlight that in BV2 cells, miR-4792 regulates EGFR/MAPK signaling following WM148 infection and partially alleviates MAPK-mediated inflammatory responses therefore affect the production of pro-inflammatory factors following WM148 infection.

Expression of miR-4792 in patients with CM before and after treatment
We collected CSF from patients with CM before and after regular antifungal treatment. The results showed that the expression of miR-4792 in CSF significantly increased after treatment (Fig. 6a).
Receiver operating characteristics (ROC) curve analysis showed the areas under curves (AUCs) of 0.75 (95% CI, 0.54-0.96) (Fig. 6b). We also found the expression of EGFR in CSF significantly decreased after treatment (Fig. 6c). Receiver operating characteristics (ROC) curve analysis showed the areas under curves (AUCs) of 0.79 (95% CI, 0.6-0.98) (Fig. 6d). These data confirmed the reciprocal relationship between miR-4792 and EGFR in vivo setting of fungal infection and showed that the CSF miR-4792 and EGFR might be served as an useful biomarker for judging the effort of the treatment of CM patients.

Discussion
Cryptococcal meningitis (CM) is 100% fatal if not immediately treated and predominates in immunocompromised individuals [29]. CM diagnostics have rapidly improved over the last 10 years due to advancements in point-of-care testing that is highly specific, sensitive, accurate and capable of data production within 10 min of test initiation [30,31]. Despite diagnostic advances, the discovery of effective anti-fungal agents has stalled. In this study, we present data that has relevance for the discovery of immunomodulatory therapies that provides a perspective to improve the immune modulation to fungal infection, providing hope for future CM treatment regimens.
MicroRNAs protect the host from infection by regulating key genes involved in host immune defenses amongst different phenotypes of macrophages and immune response [32,33]. In vitro, macrophage phagocytosis directly correlates with the clinical outcome [34, 35]. Following C. neoformans exposure in human monocytic THP-1 cells, miR-146a is upregulated and inhibits NF-κB activation and inflammatory cytokine release [6]. MiRNAs act in concert to induce host defenses and are frequently dysregulated in an array of disease models, including the exposure to fungal pathogens. As such, miRNAs serve as novel biomarkers for fungal infections. By screening differentially expressed microRNAs in C. neoformans infection and studying their interactions with microglia, we found that miR-4792 represents a potential immunomodulatory therapy of cryptococcosis that can improve the design of antifungal drugs and may hold value as a biomarker for C. neoformans exposure.
In BV2 cells, exposure to WM148 led to a significant upregulation of IL-1β, TNF-α, and IL-6. Proinflammatory cytokines are key mediators of neuroinflammation in many CNS pathologies, highlighting the ability of C. neoformans to induce a neuroinflammatory phenotype. BV2 cells are RAF/Myc immortalized murine neonatal microglia models and act as surrogates for primary microglia [36]. The suitability of BV2 microglia as an alternative model system has been established previously [36,37]. These in vitro data further highlight the inflammatory responses induced by C. neoformans.
Microglia have been heavily implicated in neuroinflammation in response to a broad range of intrinsic or external stimuli [38]. Microglia are key effector cells in the host defenses to microbial infections and act as antigen presenting cells that can produce active substances that promote inflammatory cell death [39,40]. Following C. neoformans exposure, microglia produce TNF-α, IL-1β and IL-6 to upregulate MHC Class II and CD11c [41,42]. Moreover, microglia can phagocytose C. neoformans and upregulate iNOS with anti-fungal effects which is dependent on GPR43 expression [43][44][45][46]. Despite the phagocytic capabilities of microglia, they are unable to destroy yeast cells and remain susceptible to latent intracellular infections [46,47]. Microglia are also activated in response to injury in which cell-surface CD11b is a typical phenotypic marker [48]. Microglia continually survey the microenvironment for noxious agents and injurious processes, and respond to extracellular signals, clear debris and toxic substances, and secret trophic factors, providing neuroprotection following central nervous system (CNS) injury. Increasing evidence implicates microglia activation as a major cause of CNS inflammation, the suppression of which reduces tissue damage and morphological alterations [49][50][51]. This highlights microglia as critical for analysis in C. neoformans infection.
In this study, we confirm a role for EGFR/MAPK signaling in WM148 mediated inflammation in the CNS.
EGFR is expressed in astrocytes, neurons oligodendrocytes and microglia [52,53]. The inhibition of EGFR/MAPK signaling prevents microglial inflammatory responses, through dampening down Ras-Raf-MAPK signaling [54]. MAPK activation is essential for the production of inflammatory cytokines, including IL-1β, TNF-α and IL-6 and regulates cell survival, differentiation and proliferation through its effects on gene expression [23,55 ]. Oral epithelial cells infected with Candida albicans activate three MAPK subfamilies and enhance the production of inflammatory mediators. Sporothrix schenckii yeast induces robust activation of JNK, ERK1/2 and p38 MAPK in dendritic cells, which is related to IL-6 and TNF-α secretion. Recent study demonstrated that miR-4792 participates in the apoptotic induction of A549 cells by RTHF via MAPK pathway and determines the potential apoptotic mechanism [56]. In this study, we found that miR-4792 mimics/inhibitors did not significantly influence EGFR expression at the transcription level, suggestive of translational regulation. This study found that the inhibition of EGFR leads to the inhibition of downstream MAPK signaling, preventing microglial activation and inflammatory cytokine production. Taken together, these data highlight the importance of EGFR and its effect on MAPK signaling and pro-inflammatory factors during the microglial inflammatory response to C. neoformans.
The exposure to dangerous pathogenic fungal infections poses a risk to human health. Enhancing our understanding of the mechanisms governing the host innate immune response to pathogenic infections can advance the discovery of anti-fungal agents. Previous studies have solely focused on the end-points of the host response to fungal exposure but the specific molecular mechanisms underlying these responses differ for fungal species. Although an array of studies have investigated the pulmonary immunological response following both chronic and acute exposure to lethal fungal spores, the miRNAs that mediate these responses/deficiencies are poorly characterized. In this study, using an up-to-date models, and the assessment of miR-4792 and EGFR in clinical samples, we specifically analyzed the role of miR-4792 and EGFR in the host immune response to fungal exposure, Our data provide a platform for the future development of anti-fungal immune therapeutic regimens and permit the identification of at-risk populations, enabling targeted treatments to those deemed most at risk and can provide a new way to judge the effort of the treatment of CM patients. However, some limitations exist, for example, human primary microglia are difficult to culture in vitro, and so BV2 cells were used as an alternative. As miRNAs are highly conserved, the results can guide future human assessments. Further studies on the role of miR-4792 in cryptococcal meningitis animal models are now required through the intramedullary injection of miR-4792 liposomes or exosomes.

Conclusions
In summary, both our in vivo and in vitro data highlight how miR-4792 regulates microglial activation and neuroinflammation in response to fungal infection. We further highlight the role of the EGFR/MAPK axis in mediating these responses.

Materials And Methods
Patients CSF samples were obtained from 11 patients with cryptococcal meningitis (7 males and 4 females; age range 16-62 years, mean 45 years old, all HIV(-). CSF were collected before and after effective antifungal therapy at our local hospital).All of the patients fulfilled the criteria of the Infectious Diseases Society of America (IDSA). All protocols were approved by the ethics committee of our institute and were performed according to the declaration of Helsinki guidelines.

Cell Culture
Given the limited availability of primary cultures, BV2 cells were used as representative immortalized microglia cell lines due to their known similarities to glial cells [57,58]. Cells were cultured under standard tissue culture conditions in complete DMEM (Gibco) containing 10 FBS and 1 % penicillin/streptomycin. The cell densities were never allowed to exceed 5×10 5 cells/cm 2 .

Induction of BV2 cells
Cells were seeded into 6-well plates and stimulated with C. neoformans (WM148) for 6 h to create an inflammatory environment. Cells were then treated with the AG1478, miR-4792 inhibitor, miR-4792 inhibitor NC, miR-4792 mimics and miR-4792 mimics NC for 24 h. Untreated BV2 cells served as negative controls.

Western blot analysis
For western blot (WB) analysis, BV2 cells were harvested in lysis buffer and total cell proteins were resolved by SDS-PAGE electrophoresis. Proteins were transferred to PVDF membranes which were blocked and labeled with the appropriate primary antibodies overnight at 4°C. Membranes were then washed and labeled with HRP-conjugated secondary antibodies for 1 h. ECL was used to visualize protein bands. Membranes were exposed on X-ray films, developed and imaged. Briefly, 50μL of antibody-conjugated beads were added to the assay plates to which 50 μL of tissue lysates, standards, and blank controls were added in the dark at room temperature (RT) with shaking at 850 rpm for 2h. After washing, 50 μL of biotinylated antibodies were added to the plate in the dark at RT with shaking at 850 rpm for 1h. Plates were then washed and 50 μL of streptavidinphycoerythrin (PE) was added in the dark at RT with shaking at 850 rpm for 30 min. Plates were then washed and read using a Bio-Plex MAGPIX Multiplex Reader (Bio-Rad, Hercules, CA, USA). Bio-Plex Manager 6.0 software was used for data acquisition and analysis.
qRT-PCR and ELISA analysis Cells (1×10 5 cells/cm 2 ) were seeded onto coverslips and treated with AG1478 for 24 h following WM148 induction. The cell culture media form each sample was assessed by ELISA and RNA was extracted at various timepoints post-harvesting.
For qRT-PCR analysis, total mRNA was extracted using MagExtractor and 1 µg of mRNA was subject to RT using ReverTra Ace (Toyobo). PCRs were performed using hot-star PCR mixes from 1 µl of cDNA template in a reaction volume of 25 μl. Successful PCRs were confirmed following gel electrophoresis visualized on a Gene Genius Bio-Imaging system (Syngene). Target gene expression was calculated relative to β-actin and values were normalized to untreated controls.