Human macrophages differentiated in the presence of vitamin D3 restrict dengue virus infection and innate responses by downregulating mannose receptor expression

Background Severe dengue disease is associated with high viral loads and overproduction of pro-inflammatory cytokines, suggesting impairment in the control of dengue virus (DENV) and the mechanisms that regulate cytokine production. Vitamin D3 has been described as an important modulator of immune responses to several pathogens. Interestingly, increasing evidence has associated vitamin D with decreased DENV infection and early disease recovery, yet the molecular mechanisms whereby vitamin D reduces DENV infection are not well understood. Methods and principal findings Macrophages represent important cell targets for DENV replication and consequently, they are key drivers of dengue disease. In this study we evaluated the effect of vitamin D3 on the differentiation of monocyte-derived macrophages (MDM) and their susceptibility and cytokine response to DENV. Our data demonstrate that MDM differentiated in the presence of vitamin D3 (D3-MDM) restrict DENV infection and moderate the classical inflammatory cytokine response. Mechanistically, vitamin D3-driven differentiation led to reduced surface expression of C-type lectins including the mannose receptor (MR, CD206) that is known to act as primary receptor for DENV attachment on macrophages and to trigger of immune signaling. Consequently, DENV bound less efficiently to vitamin D3-differentiated macrophages, leading to lower infection. Interestingly, IL-4 enhanced infection was reduced in D3-MDM by restriction of MR expression. Moreover, we detected moderate secretion of TNF-α, IL-1β, and IL-10 in D3-MDM, likely due to less MR engagement during DENV infection. Conclusions/Significance Our findings reveal a molecular mechanism by which vitamin D counteracts DENV infection and progression of severe disease, and indicates its potential relevance as a preventive or therapeutic candidate.


Methods and principal findings
Macrophages represent important cell targets for DENV replication and consequently, they are key drivers of dengue disease. In this study we evaluated the effect of vitamin D 3 on the differentiation of monocyte-derived macrophages (MDM) and their susceptibility and cytokine response to DENV. Our data demonstrate that MDM differentiated in the presence of vitamin D 3 (D 3 -MDM) restrict DENV infection and moderate the classical inflammatory cytokine response. Mechanistically, vitamin D 3 -driven differentiation led to reduced surface expression of C-type lectins including the mannose receptor (MR, CD206) that is known to act as primary receptor for DENV attachment on macrophages and to trigger of immune signaling. Consequently, DENV bound less efficiently to vitamin D 3 -differentiated macrophages, leading to lower infection. Interestingly, IL-4 enhanced infection was reduced in D 3 -MDM by restriction of MR expression. Moreover, we detected moderate secretion of TNF-α, IL-1β, and IL-10 in D 3 -MDM, likely due to less MR engagement during DENV infection. PLOS

Introduction
During the last decades, there has been an expansion in the geographic range and incidence of dengue virus (DENV) due to spreading of its mosquito vectors, globalization and the lack of a protective tetravalent dengue vaccine [1][2][3]. It is estimated that nearly a third of the world population is at risk of infection with an annual incidence of 96 million symptomatic cases and high economic burden in countries where active DENV transmission has been identified [4,5]. Infection with DENV may result in a self-limiting febrile illness known as dengue fever with or without warning signs that can progress to severe dengue. Disease severity is hallmarked by hemodynamic compromises that can lead to organ failure, hypovolemic shock and ultimately death [6]. While only a small percentage of cases evolve to severe dengue, progression and severity of dengue disease can differ depending on eco-epidemiology, host genetic factors, age, and virus virulence [7,8]. Additionally, complex interactions between the host immune response and the virus have been proposed as critical factors contributing to the pathogenesis of the disease [9][10][11][12].
In general, upon biting by a dengue-infected mosquito, dermal dendritic cells (DC) and macrophages are the main targets of DENV. In the skin, these cells host viral replication and facilitate further dissemination to peripheral tissues [13,14], therefore becoming key drivers in the regulation of DENV-induced immune responses. To initiate infection, DENV has been shown to interact with C-type lectin receptors such as the mannose receptor (MR), Dendritic Cell-Specific Intercellular adhesion molecule-3-Grabbing Non-integrin (DC-SIGN) and Ctype lectin domain family 5 member A (CLEC5A) [15][16][17][18]. Although MR and DC-SIGN can bind DENV with high avidity facilitating attachment of the virus to the cell, in macrophages, only MR is thought to play a predominant role in virus binding and signaling [16,18]. Ligation of MR to DENV facilitates spatial interaction of the virus with a lower avidity receptor, CLEC5A [18], which in turn initiates signaling pathways that aim at the secretion of cytokines that potentiate immediate local response and priming of the immune system [17,18].
DENV-induced activation of target cells is generally believed to induce excessive production of pro-inflammatory cytokines such as TNF-α and IL-1β that affect endothelial integrity and consequently enhance capillary permeability [9,10,19,20]. Furthermore, early in the infection, components of mosquito saliva and local tissue damage at the site of infection cause neutrophils, basophils and mast cells to elicit a Th2 cytokine response with predominant production of IL-4 [21]. Notably, the presence of IL-4 induces up-regulation of MR expression on dermal macrophages and recruits monocyte-derived macrophages (MDM), thereby boosting further infection and pro-inflammatory events that may lead to disease progression [13,16,21].
Currently, no treatment for clinical improvement of dengue disease symptoms is available [3,22]; however, antiviral and immunomodulatory factors such as vitamin D may have the necessary potential. Apart from its classical role in maintaining calcium homeostasis, vitamin D 3 is a potent modulator of the immune system [23]. Indeed, binding of the biologically active form of vitamin D, 1,25-dihidroxyvitamin D 3 (vitamin D 3 ), to the vitamin D receptor (VDR) allows the VDR to act as a transcription factor that modulates the gene expression of proteins involved in calcium absorption, cell proliferation and differentiation [24]. Vitamin D 3 modulates the immune response to several pathogens including DENV [25][26][27][28]. In fact, epidemiological studies have associated genetic variants in the VDR with disease progression and vitamin D supplementation with early disease recovery [29][30][31]. In vitro, vitamin D 3 treatment of myelo-monocytic cell lines reduces DENV infection and to modulates the cytokine response [32][33][34]; yet the underlying mechanism remains elusive [35]. Therefore, we here investigated the phenotypic features, susceptibility and innate responses to DENV infection of monocytederived macrophages differentiated in the presence of D 3 (D 3 -MDM).

Ethics statement
Protocols for sample collection and written informed consent were approved by the Committee of bioethics Research of the Sede de Investigación Universitaria, Universidad de Antioquia (Medellín-Colombia).

DENV stocks and titration
The DENV-2 New Guinea C (NGC) strain was provided by the Center for Disease Control (CDC, CO, USA) and was propagated in C6/36 cells. Briefly, monolayers of C6/36 HT cells in 75-cm 2 tissue culture flasks were inoculated with DENV at a MOI of 0.05 in 1 mL of L-15 medium supplemented with 2% Fetal Bovine Serum (FBS). After 3 h, 10 mL of L15 medium containing 2% FBS were added and the cells were cultured for 5 days at 34˚C without CO 2 . The supernatants were obtained by centrifugation for 5 min at 1800 rpm to remove cellular debris and were stored at -70˚C. Virus titration was performed by flow cytometry as described [36]. Briefly, C6/36 HT cells were seeded in 12-well plates and cultured overnight at 34˚C without CO 2 . The cells were infected with 10-fold serial dilutions of the virus and harvested at 24 h post-infection (hpi). Indirect intracellular staining of DENV E protein with the monoclonal antibody 4G2 (Millipore, Darmstadt, Germany) and the secondary antibody goat anti-mouse IgG-FITC (Invitrogen, Life Technologies, CA, USA) was performed as described later below. The cells were analyzed by FACScanto flow cytometry using the FACSdiva software. The percentage of infected cells in each sample and the total number of cells seeded per well were used to calculate the final titer of the virus. Isolation of viral RNA from cell lysates and supernatants was performed according to manufacturer's instructions using the RNeasy mini kit and the QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany), respectively. The number of genome equivalent copies (GEc) was determined by RT-qPCR using DENV-2 specific primers (forward: 5'CAATATGCTGAAACGCGAGAGAAA 3', and reverse: 5' CCCCATCTATT CAGAATCCCTGCT 3'). The calculation of the GEc was performed based on a standard curve, as previously reported [37,38].

Cell lines
The mosquito C6/36 HT cell line was obtained from ATCC and cultured in Leibovitz L-15 medium (L-15) supplemented with 10% v/v heat-inactivated FBS, 4 mM L-glutamine, and 10 units/ml penicillin/0.1 mg/ml streptomycin (Sigma-Aldrich Chemical Co, MO, USA), at 34˚C in an atmosphere without CO 2 .

Blood donors
This study was conducted according to the principles expressed in the declaration of Helsinki. All samples of venous peripheral blood were obtained in Medellin-Colombia from an equal proportion of healthy women and men that were not vaccinated against yellow fever virus, were seronegative for the DENV NS1 antigen and DENV IgM/IgG, and were between 20 and 33 years old.

Monocyte-derived macrophage differentiation
Peripheral blood from donors was mixed with EDTA 4% v/v and human peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll-Histopaque (Sigma-Aldrich) gradient at 650 g during 30 min as described [39]. Platelet depletion was performed by washing with PBS (Sigma-Aldrich) three times at 250 g during 10 min. Monocytes were obtained from total PBMCs by plastic adherence as described [39]. Briefly, 5x10 5 CD14+ cells from total PBMCs were plated in 24 well plastic plates and were allowed for to adhere during 4 h in 1640 RPMI medium (Sigma-Aldrich) supplemented with 0.5% of heat inactivated human serum pool (HSP) at 37˚C and 5% of CO 2 . Non-adherent cells were removed by washing twice with PBS, and MDM differentiation was allowed for 144 h in RPMI medium 10% HSP at 37˚C and 5% of CO 2 . Additionally, MDMs were differentiated in presence of 1,25 di-hydroxyvitamin D 3 to obtain D 3 -MDMs. For this, 1,25 di-hydroxyvitamin D 3 (Sigma-Aldrich) was added to the culture media at a final concentration of 0.1 nM and replenished every 48 h until the final time point of differentiation was reached. Kinetics and concentration of the vitamin D dose was determined on the basis of cytotoxicity levels lower than 5% (measured by the MTT and trypan blue exclusion assays, of transcriptional induction of vitamin D signaling targets (VDR and CYP24A) and of modulation of immune responses, as previously described [40,41]. For each experiment, two equal fractions of PBMCs were used from the same donor for MDM and D 3 -MDM differentiation. At the indicated experiments, after MDM and D 3 -MDM differentiation, the cells were stimulated with IL-4 (100 ng/mL) (PeProtech) or mock-treated (culture medium) and incubated for an additional 48 h previous to DENV challenge.

Flow cytometry analysis
Expression of macrophage surface and intracellular molecules was evaluated by flow cytometry. Intracellular staining using anti-CD68-PE (BD Pharmingen) and surface staining using anti-CD14-FITC (BD Pharmingen), CD83-PeCy-7 (eBioscience) and anti-CD206-PE (BD Pharmingen) were performed on MDMs and D 3 -MDMs. For each experiment, unstained and isotype controls were included. DENV infection was measured by intracellular detection of the E protein using the murine monoclonal 4G2 antibody (Millipore, Darmstadt, Germany) and a secondary antibody, the goat anti-mouse IgG-FITC (Invitrogen). Unstained cells, mockinfected cells plus secondary antibody only, mock-infected plus detection pair and infected cells plus secondary antibody were included as controls for every experiment. The percentage of infected cells was expressed as number of 4G2 positive cells over the total number of cells analyzed. Surface and intracellular staining samples were read on a FACScanto flow cytometer (BD Biosciences) and data were analyzed using the FACSdiva (BD Biosciences) and Kaluza (Beckman Coulter) softwares.

DENV infections
After differentiation, macrophage monolayers were washed with warm PBS and were infected with DENV at the indicated MOI in 300 uL of RPMI medium and 2% HSP per well. In some experiments, macrophage monolayers were pre-treated with α-Methyl D-Mannoside (MM) (Sigma-Aldrich) at 10 mM for 1 h before the virus was added to the cells. For these experiments, mock vehicle controls of cells pre-treated with the same volume of MM diluent alone were included. At 3 hpi, the cells were washed with warm PBS to remove unbound virus and were incubated at 37˚C with 5% CO 2 in RPMI medium and 10% PSH for another 21 h. At 24 hpi, macrophage monolayers were harvested and used for flow cytometry staining and analysis as mentioned above. Additionally, in some experiments cell lysates and supernatants were used to determine the GEc titer.

DENV binding measurement
Since low temperature can not only affect the dynamics of C-type lectin receptors and the fluidity of the cell membranes to facilitate endocytosis, but also the DENV conformational structure and attachment to target cells [42,43], we conducted our binding experiments at 37˚C, as previously described [44]. Macrophages were infected with DENV at a MOI of 10 for 1 h. Thereafter the cells were delicately washed twice with PBS to remove the unbound virus and lysed to determine the GEc titers as described above. Since rapid internalization may occur during this time, we report here percentage of bound and/or internalized GEc. The number of GEc titers detected in the inoculum served as 100% binding and/or internalization control.

Cytokine response
Levels of TNF-α, IL-6, IL-1β and IL-10 in cell supernatants were measured by ELISA (BD Biosciences, San Jose, CA, USA) according to the manufacturer's instructions.

Statistical analysis
Statistical comparisons were performed using the non-parametric Mann-Whitney test using the software GraphPad Prism 5 (GraphPad Prism, CA, USA). In addition, the Wilcoxon signed-rank test was used to compare paired MDMs and D 3 -MDM data. The critical value for statistical significance used for the analysis in the present study was p<0.05, denoted as Ã , p<0.01 denoted as ÃÃ , and p<0.001 denoted as ÃÃÃ .

D 3 -MDMs are less susceptible to DENV infection than MDMs
First, we assessed the effect of D 3 -MDM differentiation on the main phenotypic features of MDMs. To this end, CD14+ monocytes were obtained from human PBMCs by plastic adherence as described [39], and were cultured during 144 h in the absence (MDM) or presence of 0. Next, we compared the susceptibility of MDMs and D 3 -MDMs towards the DENV-2 NGC strain. Infection was evaluated 24 hpi by intracellular detection of the viral E glycoprotein using flow cytometry. In line with an earlier publication [34], we observed that DENV-2 infection in MDMs depends on the MOI. At a MOI of 10, 9.5% of the cells were infected (S1D Fig  and Fig 1A). Importantly, and in line with studies in immortal cells, infection of D 3 -MDMs at a MOI of 10 resulted in a significantly lower (p = 0.01) percentage of DENV-positive cells (~4%) (Fig 1A and 1B). To verify these results, we measured the number of DENV (GEc present in cell lysates and supernatants. In line with the infection data, we found that both, intracellular and secreted GEc titers were significantly lower (p = 0.03) in D 3 -MDMs when compared to MDMs (Fig 1C). Interestingly, whereas the percentage of infected cells was reduced only by 2 fold, the viral progeny release was reduced by 200 fold, suggesting that D 3 -MDM differentiation may contribute to viral restriction during early and late stages of the viral life cycle.

DENV-2 binds less efficiently to D 3 -MDMs
To assess whether DENV infectivity in D 3 -MDMs is restricted at early stages of infection, we evaluated the number of bound and/or internalized DENV-2 particles in MDMs and D 3 -MDMs at 1 hpi. The number of bound and/or internalized DENV-2 GEc was measured by RT-qPCR as described in the Methods section. Fig 2 shows the percentage of bound and/or internalized DENV-2 GEc in relation to the number of added GEc titers. For MDMs, the percentage of bound and/or internalized DENV-2 GEc ranged from 15% to 80% depending on the blood donors. Notably, only for D 3 -MDMs, was the percentage of bound and/or internalized DENV-2 GEc comparable between the donors and significantly (p = 0.03) lower in comparison to their MDM counterparts. This led us to propose that macrophage differentiation in the presence of vitamin D 3 reduces expression of receptors required for DENV to gain entry into macrophages.

D 3 -MDM modulates the expression of the mannose receptor
Among several C-type lectins expressed on macrophages that can facilitate virus attachment, MR is thought to represent a key binding receptor [15,16,18,46]. Thus, we next tested whether  (Fig 2B). In addition, the mean fluorescence intensity (MFI) values of CD206 were lower in D 3 -MDMs (~750) as compared with those observed in MDMs (~1400) (Fig 3A and 3B). Interestingly, we observed a positive correlation between the percentage of CD206 positive cells and the percentage of infection in MDMs and D 3 -MDMs (p = 0.001, r = 0.761) (Fig 3C). Moreover, and in line with our flow cytometry data, we also found that the transcriptional activity of MR was significantly This is in line with previous reports where transcriptional and surface expression of DC-SIGN has not been detected in MDMs obtained under the same conditions as reported here [47]. Since MR is critical for the MR/CLEC5A complex function, the reduced susceptibility of D 3 -MDMs to DENV could be attributed to the lower expression of MR on the cell surface.
To confirm this result, MR and other C-type lectins were blocked, reasoning that this treatment would have no effect on DENV infection in D 3 -MDMs. Accordingly, cells were pretreated with α-Methyl-D-mannoside (MM), a mannose binding site competitor and inhibitory sugar of C-type lectins [48]. As shown in Fig 3D, pre-treatment of MDMs with 10 mM MM reduced the intracellular DENV-2 GEc titers by approximately~1 Log as compared with those observed in mock pre-treated control cells. Of note, the intracellular GEc titers in MDMs treated with MM were virtually identical to those found in D 3 -MDMs. Importantly, MM treatment had no effect on the GEc titers observed in D 3 -MDMs thereby confirming that low accessibility of C-type lectins indeed restricts infection of D 3 -MDMs.

Vitamin D and dengue virus infection D 3 -MDMs attenuate IL-4-induced susceptibility to DENV
To substantiate the contribution of limited MR accessibility to reduced infection in D 3 -MDMs, we sought to up-regulate MR expression by treatment with IL-4, a well-known inducer of MR expression in macrophages [16,46]. To this end, MDMs and D 3 -MDMs were treated with IL-4 (100 ng/mL) for 48 h, as described [16]. Analysis of MR expression by flow cytometry showed that IL-4 treatment induced a significant (p = 0.007) increase in the percentage of MR positive cells in MDMs but not in D 3 -MDMs (Fig 4A and 4B). In addition, no differences were observed in the MFI values of CD206 positive cells. This indicates that D 3 -MDMs are refractory to the canonical effects of IL-4 on MR expression. In line with this, we observed that in MDMs, IL-4 treatment induced an 8-fold increase in the transcriptional activity of the MR gene when compared to mock-treated, whereas in IL-4-treated D 3 -MDMs, the fold-change induction was only 2-fold (S2D Fig). Likewise, induction of CLEC5A transcriptional activity was lower in D 3 -MDMs when compared with MDMs (S2E Fig), indicating that other IL-4-induced c-type lectins may also be less responsive to the effect of vitamin D.
We then tested whether this limited IL-4-induced increase of MR expression in D 3 -MDMs was also linked to reduced susceptibility to DENV-2. Thus, IL-4-treated MDMs and D 3 -MDMs were infected with DENV-2 at a MOI of 10 and 24 hpi the number of DENV GEc was determined by RT-qPCR in cell lysates (Fig 4C). IL-4 treatment increased the susceptibility of both MDMs and D 3 -MDMs as compared with their control, the mock-treated cells. As expected, for the mock-treated control D 3 -MDMs, the number of DENV GEc titers were~1.5 log lower as compared with mock-treated control MDMs. However, the number of intracellular DENV GEc observed in IL-4-treated D 3 -MDMs was significantly lower than that found in IL-4-treated MDMs (p = 0.007). In spite of this, there was no detectable increase in MR expression in D 3 -MDMs, IL-4 did increase the susceptibility to DENV infection. This rather surprising finding prompted us to investigate the role of MM in controlling DENV infectivity in IL-4-treated cells.
To this end, the number of DENV GEc was determined in IL-4-treated cells at 24 hpi (Fig 4D). MM treatment lowered GEc production by~2 Log (p = 0.01) and~1 Log (p = 0.01) in MDMs and D 3 -MDMs, respectively, as compared with mock-treated control cells. All these data suggest the importance of MR during DENV infection of MDMs and D 3 -MDMs and provide insights of in the mechanism by which vitamin D limits DENV infection in human macrophages.

DENV-induced inflammatory response is diminished in D 3 -MDMs
Besides the role of macrophages as target cells for DENV infection, they also play an important role in the production of pro-inflammatory cytokines that can enhance the pathogenesis of dengue disease [12]. Since vitamin D 3 has been previously suggested to act as an important modulator of immune responses [41], we next evaluated changes in cytokine response of MDMs and D 3 -MDMs following DENV infection. To this end, we measured by ELISA in culture supernatants, the production of several cytokines related to DENV pathogenesis, including TNF-α, IL-1β and IL-10. As shown in Fig 5A, baseline levels (mock-infected cells) of the cytokines are similar between MDMs and D 3 -MDMs. However, upon DENV infection, the induced cytokine levels were significantly lower (p<0.01 and p<0.05) in D 3 -MDMs as compared with MDMs ( Fig 5A). This observation could be a direct consequence of the lower infection or poor engagement of C-type lectin receptors, such as MR, due to the overall downregulation of these molecules observed in D 3 -MDMs.
DENV attachment to MR allows spatial interaction and cooperation of the MR/DENV complex with CLEC5A, that interacts with the virus, and triggers the DENV-induced proinflammatory signaling pathways [18]. Since we observed down-regulation of MR and CLEC5A in D 3 -MDMs, we hypothesized reduced participation of these receptors during DENV-induced cytokine response in these cells. Accordingly, we evaluated the DENVinduced cytokine response after blockade of C-type lectin ligation to DENV using MM. As shown in Fig 5B, pre-treatment of the cells with 10 nM MM had no effect on the levels of TNF-α, as compared with mock-treated control cells. The MM pretreatment also did not affect TLR4/CD14-mediated signaling indicating that incubation with MM does not interfere with the secretion of pro-inflammatory mediators induced by C-type lectin receptor-independent pathways. Of note, and in line with previous reports [32,41], LPS did not induce TNF-α production in D 3 -MDMs, substantiating the reduction of inflammatory responses in D 3 -MDMs. Additionally, MM treatment itself did not alter the secretion of TNF-α and IL-1β in MDMs and D 3 -MDMs. Importantly, upon DENV infection, MM treatment reduced the secretion of TNF-α and IL-1β in MDMs but not in D 3 -MDMs. On the other hand, in D 3 -MDMs, these cytokine levels were significantly (p<0.05) lower than in MDMs and showed the same levels as those found in MM-treated cells. Taken together, these data suggest that in MDMs and D 3 -MDMs, C-type lectin ligation to DENV triggers cytokine release and supports the hypothesis that down-regulation of these molecules can contribute in modulating the DENV-induced cytokine response in D 3 -MDMs.

Discussion
Containment of DENV infection and controlled secretion of inflammatory cytokines by macrophages are crucial events needed to avoid progression of dengue disease [12,49]. Our findings show that differentiation in the presence of vitamin D 3 restricts DENV infection in human MDMs by affecting DENV binding to cells. We found that MR is reduced in D 3 -MDMs and given the importance of this receptor for DENV attachment, we argue its accessibility as a limiting factor for less virus binding. Although we cannot rule out the participation of other C-type lectins such as CLEC5A, engagement of the MR receptor during DENV infection depends on the formation of the MR-DENV complex [18], attributing to MR an essential role for binding and signaling. Indeed, we show that reduced expression of MR and likely of CLEC5A contributes to a lower secretion of DENV-induced TNF-α, IL-1β and IL-10.

Vitamin D and dengue virus infection
The expression of MR is under the control of the pro-adipogenic peroxisome proliferatoractivated receptor γ (PPARγ) [50], a transcription factor that is responsible for the induction of MR expression via IL-4/IL-13 signaling and polarization of inflammatory responses in macrophages [51]. Interestingly, vitamin D activity was recently found to downregulate PPARγ expression and co-regulate PPARγ-induced transcriptional activity in macrophages [52,53]. This could explain the reduced expression of MR in D 3 -MDMs and a limited effect of IL-4 on these macrophages. Importantly, this may also have in vivo relevance in the context of dengue pathogenesis, as since inflammatory skin conditions caused by mosquito bites are accompanied by secretion of IL-4 that leads to enhancement of infection and cytokine response [16,46]. It is important to note, that the effects of vitamin D on MR expression reported here may contrast previous reports [54,55], where vitamin D 3 concentrations and target cells were different to from those used in the present study.
Furthermore, we showed here that secretion of TNF-α, IL-1β and IL-10 was significantly lower in DENV-infected D 3 -MDMs than in MDMs. This observation is in line with a previous report in the human cell line U937 [56] and can be a direct consequence of either lower cell activation due to the decreased viral load and less availability of viral antigens, or to the immunomodulatory features of vitamin D 3 . In the latter scenario, it is recognized that vitamin D can control cytokine responses by indirect modulation of NF-κB activity or by direct regulation of VDR-dependent genes [32,[57][58][59]. Both mechanisms may contribute to variations in the expression of several PRRs that are important to trigger cytokine responses. Indeed, our data shows that MDM differentiation in the presence of vitamin D modulates several DENV-relevant PRRs, such as MR and CLEC5A. MR orchestrates the MR/DENV/CLEC5A complex functions to provide an essential connection between binding and triggering of downstream signaling pathways that aim at cytokine secretion [18]. Since we found diminished MR expression in D 3 -MDMs, it is likely that a lower participation of the MR/CLEC5A complex during DENV infection can occur. Certainly, blockade of MR ligation to DENV in MDMs decreased secretion of TNF-α and IL-1β to the same levels as observed in D 3 -MDMs, showing the effect of MR availability in the induction of cytokine responses during DENV infection.
This study provides a mechanism underlying the resistant phenotype of human MDMs differentiated in the presence of vitamin D 3 to DENV infection in vitro. Interestingly, we also demonstrated that the anti-DENV effect provided by vitamin D 3 was retained after IL-4 stimulation. Since this cytokine can enhance C-type lectin receptor-mediated-DENV infection and the cytokine response [46,60], our observations indicate a potential in vivo role of vitamin D 3 in down-tuning the immune response during infection by DENV. Interestingly, susceptibility to severe dengue disease has been associated with variations on in VDR and Fc receptors [30], that are crucial for antibody-dependent enhancement of infection during secondary encounters with the virus. However, future and critical studies are required to assess the clinical importance of our findings and the potential role of vitamin D as a preventive or therapeutic target to treat disease severity. Interestingly, several reports have already anticipated VDR genetic variants with clinical outcomes of dengue disease and oral vitamin D supplementation with disease recovery and moderate inflammation [29][30][31].