The Cross-Talk between miR-511-3p and C-Type Lectin Receptors on Dendritic Cells Affects Dendritic Cell Function

The innate immune system elicits first-line, nonspecific defense against foreign substances and pathogens and is responsible for initiating Ag-specific adaptive immune responses following Ag encounter. Dendritic cells (DCs) are professional APCs that play a crucial role in the induction of immunity and tolerance (1). Through a variety of pattern recognition receptors (PRRs), DCs are able to recognize and internalize Ags in peripheral tissues (2). Following PRR-mediated maturation and migration to lymph nodes, DCs activate Ag-specific T lymphocytes, culminating in the induction of appropriate effector responses (3). DCs express a plethora of PRRs, including the C-type lectin receptors (CLRs) and TLRs, among others (4, 5). Better understanding of the mechanisms that regulate these receptors could pave the way for the rational design of new intervention strategies for treating infectious and inflammatory diseases, including allergies.

Several studies have demonstrated that CLRs on DCs, such as the mannose receptor (MR) and DC-specific ICAM 3 nonintegrin (DC-SIGN), play a major role in uptake of glycosylated Ags and are vital in tailoring innate and adaptive immune responses to pathogens and allergens (6, 7). Like other CLRs, MR and DC-SIGN possess at least one carbohydrate recognition domain with conserved motifs that determine their specificity to carbohydrate ligands (8–10). Consequently, ligation of CLRs with Ag induces downstream pathways in DCs, enabling the acquisition of functional phenotypes that support Th polarization toward different subsets. Additionally, it is increasingly apparent that CLR activation is able to modulate TLR-induced signaling pathways, affecting NF-κB function and Th polarization. For example, we have previously demonstrated that MR plays a major role in the uptake of glycosylated allergens as well as allergen-driven Th2 adaptive immune responses both in vitro (11, 12) and in vivo (13). More recently, we investigated the molecular basis of these events and have shown partial involvement of IDO and the aryl hydrocarbon receptor, which converge on components of the noncanonical NF-κB signaling pathway, following MR engagement with glycosylated allergens and other MR ligands such as mannan (7). IDO is a rate-limiting enzyme that mediates tryptophan metabolism and is expressed in a variety of cell types, including DCs, where it has been shown to control several immune-regulatory processes (14, 15). Furthermore, the finding that a subset of DCs expressing IDO inhibit proliferation of T cells is indicative of IDO’s involvement in the development of immune tolerance (16–18). Thus, given the role of MR in host defense (19) and in immune modulation (e.g., through modulating IDO activity), it is important to gain further insight into regulatory mechanisms that control MR expression and downstream DC function.

MicroRNAs (miRNAs) represent a class of small, noncoding RNA molecules that function as posttranscriptional modulators of gene expression by promoting degradation of target mRNAs (20). They are crucial in health and disease, where they regulate biological processes such as cell development and differentiation, homeostasis, and immune activation. Several miRNAs have been implicated in shaping innate immune responses, including the development and function of DCs (miR-155, Let-7i, and miR-126), downstream control of PRR signaling (miR-146, miR-223, miR-9), and production of inflammatory cytokines (miR-106, miR-155, miR-21) (21, 22). Given the emerging roles of miRNAs in immune modulation, it is reasonable to assume that changes in miRNA expression may contribute to inflammatory processes. Recent studies have highlighted miR-511 as a multifunctional miRNA with roles in innate and adaptive immune responses and in regulation of human diseases. For instance, miR-511 was shown to suppress the proliferative capacity of human lung adenocarcinomas by targeting the TRIB2 oncogene involved in cell division (23). Furthermore, dysregulated expression of miR-511-3p, the active strand of miR-511, interfered with the genetic signature of tumor-associated macrophages, thereby affecting tumor growth in mouse models of cancer (24). Interestingly, miR-511 is embedded within the MRC1 gene that encodes MR (24, 25). Although studies regarding the role of miR-511-3p in human diseases are emerging, its role in the pathogenesis of inflammation is mostly uncharacterized. Therefore, in this study, we aimed to elucidate the link between the level of miR-511-3p expression in human DCs and CLRs’ expression and function and how it could influence DC function, including IDO activity and DC-mediated T cell polarization. These data provide new insights into the molecular basis of miR-511-3p–mediated immune regulation and its potential cross-talk with PRRs.

This was done as previously described (7, 26). Buffy coats were obtained from healthy donors following ethics committee approval (National Blood Service, Sheffield, U.K.), and PBMCs were separated by density gradient centrifugation on Histopaque (Sigma-Aldrich, Dorset, U.K.). Monocytes were purified by positive selection using the MACS CD14 Isolation kit (Miltenyi Biotec, Woking, U.K.) and cultured in 24-well plates, using RPMI medium supplemented with 10% heat-inactivated FBS, 100 U/ml penicillin, 100 U/ml streptomycin, and 2 mM l-glutamine (all from Sigma-Aldrich). Cells were incubated at 37°C with 5% CO₂ in a humidified incubator. The purity of CD14⁺ cells was always above 90% as measured by flow cytometry. DC differentiation was carried out over 6 d with 250 U/ml IL-4 and 50 ng/ml GM-CSF (Miltenyi Biotec). Fresh media were added on day 3.

Predesigned miR-511-3p inhibitors and mimics were purchased from Qiagen, and transfection was carried out using the HiPerFect Transfection Reagent according to manufacturer’s protocol (Qiagen). The miRNA-targeted sequence was 5′-AAUGUGUAGCAAAAGACAGA-3′. Briefly, CD14⁺ monocytes (1 × 10⁶ cells/ml) were suspended in Opti-MEM Reduced-Serum Media (Life Technologies) and seeded. Prior to transfection, miR-511-3p inhibitor or mimic was diluted in serum-free media with transfection reagent in separate tubes for 10 min at room temperature before adding dropwise into cells. The miScript Inhibitor Negative Control and the AllStars Negative Control siRNA (Qiagen) were used as controls for inhibitor and mimic, respectively. Transfection of small interfering RNAs (siRNA) was done as described before (7, 11). MR and DC-SIGN siRNA were the SMARTpool ON-TARGETplus siRNA from GE Healthcare, Gothenburg, Sweden. The ON-TARGETplus, nontargeting control siRNA was used as negative control. All transfections were carried out at a final concentration of 50 nM as determined during optimizations. Monocytes were differentiated into DCs after 6 h of incubation, with fresh Opti-MEM media supplemented with IL-4 and GM-CSF. Transfection efficiency was determined using the fluorescently labeled siGLO RISC-Free Control siRNA (GE Healthcare), and miRNA and/or mRNA expression was assessed on day 6.

Mouse mAbs against human CD14 (clone TÜK4), CD86 (clone FM95), CD11c (clone MJ4-27G12), HLA-DR (clone AC122), CD1a (clone HI149), and DC-SIGN (clone DCN47.5) were purchased from Miltenyi Biotec. Abs against human CD83 (clone HB15e) and CD80 (clone 2D10.4) were purchased from eBioscience. Anti-PDL1 (clone MIH1), PDL2 (clone MIH18), and ICOSL (clone 2D3/B7-H2) Abs were purchased from BD Biosciences (San Jose, CA). The TLR4 Ab, anti-CD284 (clone HTA125) was obtained from AbD Serotec, and the anti-CD206 Ab (clone 15-2) was purchased from BioLegend, London, U.K. Briefly, control and transfected cells were collected and washed twice in cold PBS buffer containing 0.5% BSA and 0.1% sodium azide (Sigma-Aldrich). Staining with labeled Abs was then carried out in the dark at 4°C for 20 min according to manufacturer’s instructions. Unless otherwise stated, specific Abs were conjugated to FITC, PE, or PE Cyanine 5.1 (PE/Cy5). Following staining, samples were washed twice with PBS buffer containing 0.5% BSA and 0.1% sodium azide and fixed in 0.5% formaldehyde solution before analysis. Nonreactive, isotype-matched Abs were used as controls. Flow cytometry was carried out on the FC500 Flow Cytometer (Beckman Coulter, London, U.K.). For intracellular staining, cells were fixed for 10 min at room temperature with 4% formaldehyde solution (Sigma-Aldrich) and washed twice with PBS buffer containing 0.5% BSA and 0.1% sodium azide with 0.5% saponin (Sigma-Aldrich). IFN-γ (Miltenyi Biotec) and IL-4 (BioLegend) staining was then carried out at 4°C for 30 min and washed twice before analysis. Intracellular staining for cytokines was analyzed in a MoFlo XDP Flow Cytometer (Beckman Coulter). Data analysis was carried out with Kaluza software (version 2.1) for Windows.

Monocytes and DC samples were washed twice with cold PBS and stored at −80°C until RNA isolation. MiR-511-3p transfected cells as well as controls were stimulated on day 6 with or without mannan or LPS (Sigma-Aldrich) for 24 h before harvesting. Total RNA, including small RNAs, was isolated with TRIzol Reagent using the miRNeasy Mini Kit (Qiagen). The concentration of purified RNA was measured using a NanoDrop 1000 Spectrophotometer (Thermo Fisher Scientific, Pierce). First-strand cDNA was then generated with the miScript RT II Kit (Qiagen) according to manufacturer’s instructions. The reaction was incubated for 60 min at 37°C and at 95°C for 5 min to inactivate the reverse transcription enzyme. Reverse transcription was done with the T100 Thermal Cycler (Bio-Rad, Watford, U.K.), and the 5× HiFlex Buffer was used for parallel quantitative real-time PCR (qRT-PCR) quantification of mature miRNA and mRNA.

Comparative real-time PCR for miR-511-3p expression was carried out on the MxPro 3005P qRT-PCR system (Stratagene, San Diego, CA) using the miScript SYBR Green PCR kit (Qiagen) according to manufacturer’s instructions. Briefly, 12.5 μl of 2× QuantiTect SYBR Green Master mix, 2.5 μl of 10× universal primer, and 2.5 μl of primer assay (forward primer) was mixed with 3 ng of reverse transcription product. qRT-PCR cycling was initiated at 95°C for 15 min, followed by 40 cycles of 94°C for 15 s, 55°C for 30 s, and 70°C for 30 s. Mature miR-511-3p–specific primers were obtained from Qiagen, and relative expression was normalized to U6 (RNU6-2) small nuclear RNAs. qRT-PCR for mRNA expression was done with the Brilliant III Ultra-Fast SYBR Green qRT-PCR Master Mix (Agilent Technologies) as previously described (7). The two-step cycling reaction was initiated at 95°C for 3 min followed by 40 cycles of 95°C for 20 s and 60°C for 20 s. MR and DC-SIGN mRNA expression levels were normalized to GAPDH and calculated using the comparative Δ cycle threshold method, which determines the difference in cycle threshold values between the gene of interest and housekeeping gene. All experimental procedures were done in triplicate. Forward and reverse mRNA primers were selected using Roche Universal Probes Library and purchased from Eurofins Scientific, Wolverhampton, U.K. (Table I).

IDO activity was determined as described before (7, 27). Briefly, DCs (2.5 × 10⁵ cells/ml) were seeded in a 24-well plate with complete RPMI media supplemented with 100 μM l-tryptophan (Sigma-Aldrich). A colorimetric assay for IDO activity was determined after 24 h of stimulation by measuring the levels of l-kynurenine (KYN) produced in culture supernatants. The concentration of l-KYN was then calculated from a standard curve of defined concentrations from 0 to 200 μM.

Culture supernatants were collected and stored at −20°C before analysis. The levels of IL-10, IL-12p70, IL-6, TNF-α, and IL-1β were measured by sandwich ELISA using the Duo Set ELISA kit (R&D Systems, Abingdon, U.K.) according to manufacturer’s instructions.

This was done as previously described (7, 28). Briefly, DCs treated with miR-511-3p inhibitors or mimics were cocultured in the presence of 0.1 μg/ml LPS in 96-well round-bottom plates (Corning Life Sciences) with CD3⁺ autologous T cells at a ratio of 1:10. T cells were purified by negative selection using a mixture of immunomagnetic beads (Miltenyi Biotec). Following isolation, cells were resuspended in RPMI media supplemented with 5% human AB serum, 100 U/ml penicillin, 100 mg/ml streptomycin, and 2 mM l-glutamine (Sigma-Aldrich). The coculture media were refreshed every 3 d with media containing 5 ng/ml IL-2 (Miltenyi Biotec). After 6 d, T cells were harvested and restimulated with 2 μg/ml anti-CD3 (clone OKT3) and anti-CD28 (clone 28.2) Abs (Sigma-Aldrich) for 18 h, following which, culture supernatants were collected and stored.

T cell cytokine expression was measured in coculture medium using the RayBio Human T Cell Response Array C series (AAH-TCR-1) according to manufacturer’s instructions. Briefly, Ab-spotted membranes were blocked for 30 min and incubated with 1 ml of culture supernatants overnight at 4°C. Membranes were subsequently washed a total of five times with wash buffers I (3×) and II (2×) for 5 min and then incubated with the biotinylated Ab mixture for 2 h at room temperature. Membranes were washed as before, incubated with the HRP-conjugated streptavidin secondary Ab for 2 h, and developed by incubation with detection buffers for 2 min. Chemiluminescent signals were captured by means of the G:Box Chemi XRQ (Syngene), and exposure time was varied between 5 and 15 min to optimize signal-to-noise ratio of the images. Image analysis was done with the Odyssey Image Studio software (version 3.1.4) for Windows.

T cell proliferation was done using the CellTrace CFSE Cell Proliferation Kit (Thermo Fisher Scientific, U.K.), according to manufacturer’s instruction. Briefly, T cell cocultures were harvested on day 6 and counted. Prior to seeding, cells were incubated with 1 μl (5 μM) of CFSE solution in 1 ml of cell suspension for 20 min in the dark at 37°C before seeding into flat-bottom, 96-well plates (1 × 10⁵ cells per well). T cells were then stimulated with 2 μg/ml anti-human CD3 Ab (clone OKT3) for 96 h before harvesting. Cells were then washed, fixed in 4% formaldehyde solution, and analyzed. Viability analysis was done using the LIVE/DEAD Fixable Green Dead Cell Stain Kit following manufacturers protocol (Thermo Fisher Scientific). Cells were washed, stained with 1 μl of LIVE/DEAD stain, and incubated for 20 min in the dark at room temperature before fixing. Flow cytometry was carried out on the FC500 (Beckman Coulter), and analysis was done using Kaluza software (version 2.1) for Windows.

Data were analyzed using GraphPad Prism version 7.02 for Windows (GraphPad Software, San Diego, CA), and values expressed as mean ± SEM are from independent experiments unless otherwise stated. For nonparametric tests, the Wilcoxon signed rank test was used when comparing between two groups, and Kruskal–Wallis test was used for three or more groups. In some cases, data obtained were shown to follow a normal (Gaussian) distribution determined by the Shapiro–Wilk normality test. Therefore, paired t tests or ordinary one-way ANOVA was used when comparing between two groups or three or more groups, respectively. A p value <0.05 was considered statistically significant.

DCs were generated from monocytes by treating with GM-CSF and IL-4 for 6 d, and miR-511-3p expression was examined afterward. Immature DCs (iDCs) were confirmed by expression of CD11c and low expression of maturation marker CD83 (data not shown). Our data clearly show that miR-511-3p expression is significantly induced in iDCs compared with CD14⁺ monocytes as measured by qRT-PCR. Moreover, this increase in miRNA expression was also seen during different days of DC differentiation (Fig. 1A). To determine whether DC maturation influenced the level of miR-511-3p expression, we treated iDCs with increasing concentrations of LPS from Escherichia coli for 24 h. We observed a bell-shaped dose–response curve of miR-511-3p expression in mature DCs, all of which were significantly downregulated compared with iDCs (Fig. 1B). To examine the effect of CLR engagement on the pattern of miR-511-3p expression, we treated iDCs with mannan, a complex carbohydrate from Saccharomyces cerevisiae. In contrast to LPS treatment, we found that increasing concentrations of mannan upregulates miR-511-3p expression in human DCs, although there was a slight decrease from 1 to 10 μg/ml (Fig. 1C). Subsequent analysis of DC phenotype following treatment with mannan showed no changes to DC maturation and activation markers CD83, CD86 (B7-2), and HLA-DR (data not shown).

Next, we carried out DC differentiation from monocytes in the presence of miR-511-3p–specific inhibitors and mimics to examine whether down- or upregulation of miR-511-3p affects expression of key DC phenotypic markers. Transfection efficiency for introducing miR-511-3p inhibitors was above 80% (data not shown) and resulted in more than 50% downregulation of miR-511-3p expression (henceforth referred to as miR-511-3p^low) (Fig. 2A). Interestingly, we found for the first time, to our knowledge, that miR-511-3p^low DCs showed a significant increase in MR (CD206) and DC-SIGN (CD209) protein expression as measured by flow cytometry. This upregulation was also observed at mRNA level, where real-time PCR analysis of the miR-511-3p^low cells showed a 2-fold increase in MR and DC-SIGN mRNA levels (Table I) compared with controls (Fig. 2B, 2C). No changes were observed in expression of maturation or costimulatory markers, except PDL-1, which was significantly upregulated in the miR-511-3p^low cells. As shown in Fig. 2D, transfected cells were confirmed to be DCs but not monocytes, demonstrated by the increased expression of CD1a (an early DC development marker) and negligible expression of CD14 (CD1a^hiCD14^low DCs).

Following the observed effect of miR-511-3p knockdown on MR and DC-SIGN, we investigated the effect of increased expression of miR-511-3p on DC phenotype. To this end, we transfected DCs with 50 nM miRNA mimic, which resulted in a significant increase in miR-511-3p expression in DCs (henceforth referred to as miR-511-3p^hi) (Fig. 3A), which in turn led to a significant decrease in MR and DC-SIGN expression at both protein and mRNA levels (Fig. 3B, 3C). Interestingly, the expression of CD86 (B7-2) and HLA-DR showed significant increase in these cells (Fig. 3D); however, we did not observe changes in any other surface markers tested. Moreover, observed changes in mRNA and protein expression levels were supported by data from preliminary RNA-sequencing screen (Supplemental Fig. 1). It is noteworthy that changes in DC phenotype resulting from transfection with mimics and/or inhibitors were not associated with cell death as confirmed by annexin-V and propodium iodide staining (data not shown).

Considering the marked changes in MR expression following miR-511-3p modulation, we examined whether downregulation of MRC1, the gene controlling MR expression, would influence miR-511-3p expression in DCs. Therefore, we knocked down MR expression in DCs, using MRC1 siRNA, followed by quantifying miR-511-3p expression by qRT-PCR. MRC1 knockdown (≥60%) was confirmed by measuring MR mRNA levels (Fig. 4A). Interestingly, we found a significant increase in miR-511-3p expression in human DCs when MRC1 was inhibited compared with controls (Fig. 4B).

We and others have previously shown that CLRs such as MR and DC-SIGN play key roles in allergen recognition and downstream events leading to Th2 development (11, 28). In particular, we have shown that mannan and selected glycosylated allergens, through MR engagement, are able to downregulate LPS-induced IDO activity in DCs, favoring Th2 polarization (7). IDO mediates tryptophan metabolism and contributes to several immune-regulatory processes, including the induction of regulatory T cells (14, 29, 30). We therefore investigated whether changes in MR and DC-SIGN expression levels following inhibition or enhanced expression of miR-511-3p would influence IDO activity in human DCs. Accordingly, miR-511-3p^low and miR-511-3p^hi cells were stimulated with 10 μg/ml mannan in the presence and absence of LPS (0.1 μg/ml). As expected, we observed a significant downregulation in IDO activity when miR-511-3p^low cells were stimulated with mannan alone; however, in the presence of LPS, IDO activity was significantly upregulated (Fig. 5), suggesting that miR-511-3p^low DCs could be potent suppressors of T cell responses, depending on the nature of the stimulus. Conversely, miR-511-3p^hi cells showed no change in IDO activity when treated with mannan alone but showed a significant decrease in the presence of LPS (Fig. 5B). To further determine whether the observed changes in IDO activity in this study were influenced by the level of MR and/or DC-SIGN expression, we performed knockdown of both receptors (MR^low and DC-SIGN^low) in DCs and treated cells with mannan and LPS (Fig. 5C, 5D). Interestingly, the data demonstrate a significant increase in IDO activity following knockdown of MR, whereas there was a decrease after knockdown of DC-SIGN. These observations highlight an antagonistic relationship between MR and DC-SIGN in regulating IDO activity following stimulation with mannan and also the potential contribution of miR-511-3p knockdown particularly to an increase in DC-SIGN expression and subsequent increase in IDO activity (Fig. 5A).

Depending on the nature of signals received, DCs secrete a variety of cytokines that drive downstream immune responses. Having examined the impact of miR-511-3p on expression of DC surface markers, we investigated whether miR-511-3p modulation affects the production of IL-10, IL-6, TNF-α, IL-12, and IL-1β in DCs after treatment with LPS. We observed a significant increase in IL-10 and a decrease in IL-6 and IL-12p70 production in the miR-511-3p^low cells after treatment with LPS (0.1 μg/ml) (Fig. 6A). Although TNF-α and IL-1β production was downregulated, this did not reach statistical significance. In contrast to miR-511-3p^low cells, IL-10 production in miR-511-3p^hi cells was decreased following stimulation with LPS. Also, TNF-α and IL-1β showed a slight increase; however, this was not statistically significant (Fig. 6B). Interestingly, IL-12p70 showed a significant decrease in miR-511-3p^hi cells similar to the pattern observed for miR-511-3p^low cells, whereas no change was seen in IL-6 production. Furthermore, we observed a significant increase in expression of PDL-1 on miR-511-3p^low cells (Fig. 6A), an observation which was absent in miR-511-3p^hi cells (Fig. 6B).

We did not detect significant changes in cytokine production in either knockdown or overexpressed conditions after treatment with mannan only (data not shown). As shown in Supplemental Fig. 2, LPS treatment may account for changes in gene ontology analysis of biological processes and molecular function.

Activation of T cells by DCs promotes rapid induction of T cell proliferation and differentiation into distinct Th subsets such as Th1, Th2, Th17, and regulatory T cells, which facilitates various aspects of immunity. We therefore sought to investigate whether up- or downregulation of miR-511-3p expression in DCs affected T cell polarization in autologous DC–T cell coculture experiments. Immature miR-511-3p^low or miR-511-3p^hi DCs were stimulated with 0.1 μg/ml LPS for 24 h before coculture with CD3⁺ autologous pan T cells in complete medium containing human AB serum. Control samples did not receive LPS. T cell responses were examined after 7 d of coculture by measuring the level of cytokines produced in culture supernatants using the Human T Cell Response Array from RayBio as a screening platform (Fig. 7). Purity of isolated T cells was measured by flow cytometry. Interestingly, our data show a significant increase in IL-4 and a decrease in IL-17 cytokine production from T cells cocultured with miR-511-3p^low DCs, suggesting a bias toward type 2 T cell responses (Fig. 7A, 7B). Additionally, IL-10 and TGF-β were also increased, whereas IFN-γ was decreased, but these were not statistically significant. T cells cocultured with miR-511-3p^hi DCs, in contrast, showed a significant decrease in TGF-β and IL-9 production. Although IL-4 and IL-10 were decreased, these changes did not reach statistical significance (Fig. 7D). Furthermore, the percentage of IFN-γ– and IL-4–producing cells was determined by intracellular cytokine staining following coculture with miR-511-3p^low/hi DCs. Data from these experiments show a decrease in the percentage of IFN-γ/IL-4 double-positive T cells (from 18% in control cells to 8% in T cell cocultured with miR-511-3p^low DCs), which was in contrast to an increase in the percentage of IFN-γ/IL-4 double-positive T cells cocultured with miR-511-3p^hi DCs (9% in control cells versus 15% in cells cocultured with miR-511-3p^hi DCs) (Supplemental Fig. 3). This is in line with data from the cytokine arrays particularly for IFN-γ. Additionally, T cell proliferation seemed to be comparable between cocultures with either miR-511-3p^low or miR-511-3p^hi DCs in a CFSE-based proliferation assay. Also, no changes were seen in T cell viability between cultures with mimic or inhibitor treatment (Supplemental Fig. 4).

miRNAs represent a class of newly discovered gene regulators that bind the 3′UTR of target genes and cause translational inhibition or mRNA degradation (20, 31). Because of their increasing role in regulating various cell processes, miRNAs represent crucial regulators in both human health and disease. In this study, we examined the influence of miR-511-3p up- or downregulation with synthetic mimics or antisense inhibitors, respectively, on human DC phenotype and function, particularly within the context of surface phenotype (including CLR expression), cytokine secretion, IDO activity, and downstream T cell polarization.

Initial examination of miRNA expression showed that miR-511-3p is highly expressed in monocyte-derived DCs compared with their monocyte precursors. IL-4 and IL-13, cytokines that play key roles in Th2-mediated allergic responses, have been demonstrated to induce miR-511 expression in M2-type macrophages in mice, as opposed to M1 macrophages (32). Moreover, IL-4 in particular drives the Mrc1 gene expression in mice, which houses miR-511 in its fifth intron (24, 33). In the current study, DCs were differentiated in the presence of IL-4 and GM-CSF over 6 d, which could account for the observed upregulation in miR-511-3p expression in DCs. Our data also indicate that miR-511-3p expression in DCs is significantly downregulated following treatment with LPS. Similar results relating to suppression of miR-511 expression were shown in mouse macrophages treated with IFN-γ and heat-inactivated E. coli (32). Several miRNAs are differentially expressed in immune cells following LPS treatment most likely to protect against TLR4-mediated inflammation as demonstrated for miR-155 and miR-146a (34–36). Interestingly, overexpression of miR-511 has been shown to protect against LPS-induced endotoxemia in SPRET/Ei mouse models, which display dominant resistance against TNF-induced shock (37). Similar to these reports, we have shown that miR-511-3p is downregulated in response to TLR4 stimulation with LPS (Fig. 1B). Nonetheless, mechanisms leading to LPS-mediated downregulation of miR-511-3p in human DCs are not fully understood. It is possible that downregulation of MR expression triggered by DC maturation with LPS (38, 39) may drive suppression of miR-511-3p expression.

We have shown for the first time, to our knowledge, that transfection with miR-511-3p inhibitors in human DCs resulted in a significant increase in MR as well as DC-SIGN (CD209) protein levels, which also correlated with an increase in mRNA levels as measured by qRT-PCR. This is particularly interesting considering in mice, Mrc1 is coregulated with miR-511 as previously shown (24). Furthermore, using TargetScan and a combination of different algorithms, Tserel et al. (25) highlighted the ligand-activated transcription factor, peroxisome proliferator–activated receptor γ (PPARγ), as a putative target of miR-511 activity. Interestingly, a number of potential links between PPARγ and C-type lectins, particularly MR, have also been demonstrated. These include data showing that interaction between MR and Man-LAM from Mycobacterium tuberculosis can induce PPARγ activation, which promotes intracellular survival of the bacterium (40). Furthermore, they showed that knockdown of MR also leads to downregulation of PPARγ expression. Therefore, although miR-511-3p does not seem to directly target MR or DC-SIGN mRNA, it is likely that silencing miR-511-3p may promote MR expression via inducing changes in PPARγ activity. Generally, intronic miRNAs are generated using their host transcription machinery and should be coordinately expressed with their respective host mRNA (41). However, intronic miRNAs can also be transcribed from independent promoters, and there have been reports of discordant expression between host genes and their intronic miRNAs (31, 42–44), which could explain why host gene and miRNA expression are not always correlative. Also, the use of synthetic antisense inhibitors, which sequester the endogenous/physiological miRNA via complementary base pairing, making it unavailable for normal function (45, 46), could account for observed changes in MR and DC-SIGN expression. Moreover, we found that downregulation of MRC1 in human DCs resulted in an increase in miR-511-3p expression (Fig. 4). Our data also show a decrease in MR and DC-SIGN protein and mRNA levels following miRNA overexpression, further highlighting the likely existence of discordant expression between MRC1 and miR-511-3p. Considering that DCs were differentiated from transfected monocytes, it was possible that miR-511-3p transfection could affect the overall DC differentiation process. However, iDCs differentiated from miR-511-3p–transfected monocytes still showed high expression of CD1a (a characteristic marker expressed in early DC development) and negligible expression of CD14 (monocyte marker), which is in line with the expected phenotype of immature monocyte-derived DCs.

In a string of previous studies, we have shown that MR and DC-SIGN play a central role in the uptake of clinically relevant allergens and in modulating downstream allergic inflammation and Th2 cell polarization (7, 11–13, 28). It is therefore reasonable to assume that the cross-talk between miR-511-3p, MR, and DC-SIGN could have significant implications in response to allergens. This is in line with recent data showing MR modulates macrophage polarization and allergic inflammation through miR-511-3p in a mouse model of allergic sensitization (47). The lack of Mrc1 (Mrc1^−/− mice) in these studies significantly reduced allergen uptake as well as miR-511-3p expression in lung macrophages and exacerbated allergic inflammation, suggesting that allergen uptake via MR may promote miR-511-3p expression (47). However, in the current study, we found a reciprocal antagonistic relationship between MR and miR-511-3p in human DCs, which also brings to light the complex differences that may exist between cell types (i.e., macrophages and DCs) as well as species, with MR expression, for instance, present on chromosome 10p12 in humans, whereas present on chromosome 2.2 in mice (48).

Our data are further supported by results from IDO activity analysis. As previously indicated, IDO catalyzes tryptophan metabolism and participates in several immune-regulatory processes via induction of regulatory T cells (49–51). Recently, we showed that different airborne allergens through MR engagement are able to downregulate LPS-induced IDO activity in DCs, thereby affecting Th polarization (7). Similarly, the current study shows that mannan, a complex carbohydrate, is able to significantly downregulate IDO activity in miR-511-3p^low DCs but has no effect on LPS-induced IDO activity. This may be due to the increased expression of MR in the miR-511-3p^low cells. However, in the presence of LPS, miR-511-3p^low cells showed increased IDO activity, which supports the notion that miR-511-3p^low DCs may exert an anti-inflammatory role. It is important to note that TargetScan prediction software did not show complementary base pairing between miR-511-3p and IDO mRNA and hence, miR-511-3p does not seem to putatively target IDO. Gene ontology analysis of biological processes in our RNA-sequencing data, however, indicates regulation of cellular catabolic processes by miR-511-3p^low DCs, which was absent in miR-511-3p^hi conditions (Supplemental Fig. 2). We have also demonstrated that MR and DC-SIGN can have opposing effects on IDO activity, whereby DC-SIGN knockdown in DCs leads to a decrease in IDO activity, which is opposite to the increase in IDO activity following MR suppression. However, it is worth confirming the roles of MR or DC-SIGN by inclusion of double knockdown populations (miR-511-3p and either MR or DC-SIGN) to determine the overall contribution of either receptors to miR-511-3p–induced modulation of IDO activity in human DCs.

Nevertheless, our data indicate significant changes in DC phenotype and function following transfection with miR-511-3p inhibitors and mimics. As shown, it is clear that knockdown of miR-511-3p promotes the establishment of an anti-inflammatory DC phenotype characterized by increased IL-10 production and decrease in IL-12p70 and IL-6. Immune cells such as macrophages and DCs rapidly respond to LPS-induced inflammation through proinflammatory cytokine production, which are then downregulated following induction of anti-inflammatory signals and resolution of inflammation. Moreover, we showed that miR-511-3p^low cells treated with LPS significantly increase PDL-1 expression. Several studies have demonstrated a key role for the transmembrane protein PDL-1 (52, 53) and IL-10 secretion (54, 55) in promoting immune suppression. These include our own studies showing that PDL-1 upregulation and IL-10 secretion following LPS conditioning promotes development of human DCs with anti-inflammatory properties (56). PDL-1 belongs to the B7 family of costimulatory molecules that also includes B7-1 (CD80), B7-2 (CD86), as well as PDL-2, and T cell engagement (via PD-1) with each of these costimulatory receptors can induce inhibitory and, in some cases, stimulatory signals. Our data therefore strongly suggest that the induction of PDL-1 and IL-10 following knockdown of miR-511-3p may promote immune suppression during inflammation. In line with this, our T cell coculture data indicate that miR-511-3p^low DCs may promote Th2-type responses, evidenced by the increase in the levels of IL-4 in coculture supernatant. Posttranscriptional regulation by miRNAs is vital for controlling numerous immunological events, including the development and function of effector T cells. For instance, the miR-17 ∼ 92 cluster was shown to be important for Th1 polarization evidenced by the lack of IFN-γ and T-bet in miR-17 ∼ 92–deficient mice (57). Recently, the miR-21 ∼ 27 ∼ 24 family cluster has been shown to limit Th2 responses by directly and indirectly targeting IL-4 and GATA3 (58). In the current study, a significant increase in IL-4 and decrease in IL-17 production from T cells was observed in coculture with miR-511-3p^low DCs, which may suggest a bias toward Th2 phenotype. Although IL-10 was increased, this was not statistically significant (Fig. 7A, 7B). The bias toward Th2 phenotype, driven by miR-511-3p^low DCs, is also backed by the finding that TGF-β is significantly reduced in T cells cocultured with miR-511-3p^hi DCs, which may indicate suppression of regulatory T cells.

In conclusion, we have shown that manipulation of miR-511-3p expression in human DCs plays a key role in regulating DC phenotype and function partly through regulating MR and DC-SIGN expression. Additionally, miR-511-3p knockdown may contribute towards induction of anti-inflammatory DC phenotype and promote immune suppression at least within the context of LPS-induced inflammation. Our data provide new insight into miR-511-3p–mediated immune regulation and could pave the way for rational design of new therapies against inflammatory diseases.

Authors would like to acknowledge support from the Flow Cytometry facility (School of Life Sciences, University of Nottingham).

The authors have no financial conflicts of interest.