Vitamin D increases killing of intracellular Leishmania amazonensis in vitro independently of macrophage oxidative mechanisms

Abstract

Vitamin D has been reported to activate macrophage microbicidal mechanisms by inducing the production of antimicrobial peptides and nitric oxide (NO), but conversely has been shown to contribute to a greater susceptibility to Leishmania amazonensis infection in mice. Thus, this study aimed to evaluate the role of vitamin D during intracellular infection with L. amazonensis by examining its effect on macrophage oxidative mechanisms and parasite survival in vitro. Vitamins D₂ and D₃ significantly inhibited promastigote and amastigote growth in vitro. Vitamin D₃ was not able to induce NO and reactive oxygen species (ROS) production in uninfected macrophages or macrophages infected with L. amazonensis. In addition, vitamin D₃ in combination with interferon (IFN)-γ did not enhance amastigote killing and in fact, significantly reduced NO and ROS production when compared with the effect of IFN-γ alone. In this study, we demonstrated that vitamin D directly reduces parasite growth in infected macrophages (approximately 50–60% at 50 μm) but this effect is independent of the activation of macrophage oxidative mechanisms. These findings will contribute to a better understanding of the role of vitamin D in cutaneous leishmaniasis.

Introduction

Leishmaniasis is a protozoonosis affecting around 700 000 to 1 million individuals annually and causing 26 000–65 000 annual deaths (World Health Organization, 2019). During their life cycle, these parasites alternate between two main morphological forms: (1) amastigotes, which are intracellular and mainly found in macrophages, and (2) promastigotes, which are present in the digestive tract of the insect vector (Hoare and Wallace, 1966; Anversa et al., 2018). Pentavalent antimonials are used as the first-line drugs in the treatment of cutaneous leishmaniasis, but they are not recommended by the Food and Drug Administration (FDA), due to their toxicity and the clinical failure associated to parasite resistance (Aronson and Joya, 2019). For the treatment of visceral leishmaniasis, liposomal amphotericin B is recommended as the first-line antileishmanial drug, but this therapy is expensive and in East Africa and Brazil the efficacy of this drug is lower than expected, meaning high doses are required (van Griensven and Diro, 2019). However, in Brazil and in most parts of the world the treatment of choice for all clinical forms of leishmaniasis remains to be the pentavalent antimonials (Anversa et al., 2018; Maxfield and Crane, 2020).

Leishmania infection mainly occurs in poor communities, where diets are often inadequate. Furthermore, in Leishmania-infected individuals, malnutrition is more pronounced (Custodio et al., 2016). It has already been demonstrated that pre-existing malnutrition can interfere with cell-mediated immune responses against L. infantum, altering T cell migration and suppressing the control of parasite, which contributes to the pathophysiology of visceral leishmaniasis in malnourished individuals (Losada-Barragán et al., 2017).

Vitamin D is the precursor of the steroid hormone calcitriol (Rodriguez-Cortes et al., 2017). The major source of vitamin D for humans comes from exposure to sunlight (Holick, 2008), in which vitamin D is synthesized in the skin after exposure to UVB (Prietl et al., 2013). However, vitamin D can also be obtained orally through diet in the two main forms, vitamins D₂ (ergocalciferol) and D₃ (cholecalciferol). This vitamin reaches the liver, where it is converted into 25-hydroxyvitamin D [25(OH)D]. This pre-hormone then undergoes hydrolysis by the enzyme 1-α hydroxylasein in the kidneys and is converted to 1,25-dihydroxyvitamin D [1,25(OH)₂D], the hormone also known as calcitriol. Calcitriol is the most biologically active vitamin D metabolite and increases calcium absorption, which is essential for bone health. In addition, vitamin D or 25(OH)D can also reach other organs, such as the breast, colon, skin, brain, ovary and prostate where it is converted into 1,25(OH)₂D. In these organs, cells express the vitamin D receptor, through which 1,25(OH)₂D can regulate cell growth (Hollis and Wagner, 2013). Furthermore, the interaction between vitamin D and its receptor can regulate various cellular processes, such as intracellular signalling cascades, cytokine secretion, response to stress, cell communication and cell differentiation (Heikkinen et al., 2011).

There are reports describing the antimicrobial (García-Barragán et al., 2018; Huang et al., 2019) and antiparasitic activities of vitamin D (Yamamoto et al., 2017, 2019). Moreover, studies have shown that vitamin D is able to modulate immune responses (Waters et al., 2001; García-Barragán et al., 2018). The nuclear vitamin D receptor and vitamin D₃-metabolizing enzymes are present in many cells of the immune system, such as dendritic cells, monocytes, macrophages, T cells and B cells, and this fact strongly suggests an important role of vitamin D as a modulator of immune responses (Vanherwegen et al., 2017). It has been reported that this vitamin is able to improve the phagocytic capacity of human monocytes (Xu et al., 1993). Calcitriol can also alter the function and morphology of dendritic cells to induce tolerogenic properties in these cells. This is characterized by a decrease in the expression of MHC class II and co-stimulatory molecules (CD40, CD80 and CD86) in dendritic cells, which leads to reduced antigen presentation (Prietl et al., 2013). Vitamin D can also control cells of the adaptive immune system. Direct effects of calcitriol on B cells have already been demonstrated, promoting the inhibition of memory cells and plasma cells generation and the induction of apoptosis in immunoglobulin-producing B cells (Chen et al., 2007; Prietl et al., 2013). In addition, calcitriol is able to suppress the proliferation and differentiation of T helper (Th) cells and modulate the cytokine production by these cells (Lemire et al., 1985), specifically by inhibiting the secretion of proinflammatory Th1 cytokines, such as interferon (IFN)-γ and tumour necrosis factor (TNF)-α, and stimulating the production of anti-inflammatory Th2 cytokines, including IL-4 and IL-5 (Prietl et al., 2013). Studies have already shown that vitamin D induces nitric oxide (NO) production, contributing to the effect of this vitamin against Mycobacterium (Waters et al., 2001; García-Barragán et al., 2018).

With regards to Leishmania parasites, the data reported on the effect of vitamin D are contradictory. It was verified that the treatment with vitamin D reduced lesion size of mice infected with L. mexicana, without affecting the parasite load (Ramos-Martínez et al., 2013). In accordance with this, Rodriguez-Cortes and colleagues have shown in a cohort study that dogs naturally infected with Leishmania and exhibiting symptoms of the disease presented lower vitamin D levels in the serum compared with asymptomatic or non-infected animals (Rodriguez-Cortes et al., 2017). In contrast, Leishmania amazonensis-infected mice receiving a vitamin D-deficient diet control lesion development better than those receiving a regular diet, which was shown to be due to the involvement of Th1 cells in both C57BL/6 and BALB/c mice (Bezerra et al., 2019). Furthermore, during L. major infection, C57BL/6 mice deficient for the vitamin D receptor (VDR knockout) developed significantly smaller lesions, with a reduced inflammatory response and healing time 3 weeks faster than wild-type (WT) C57BL/6 mice. However, these differences were not observed between VDR knockout in the BALB/c background and the BALB/c WT mice, which indicates the importance of the host background (Whitcomb et al., 2012). Due to these conflicting reports with in vivo models, it is necessary to investigate and understand the role of vitamin D at the cellular level using a Leishmania-macrophage in vitro model infection. For L. major, in vitro treatment with vitamin D on infected macrophages did not reduce parasite load and also inhibited the anti-microbial response promoted by IFN-γ; however, the effect on L. amazonensis-infected macrophages, that present a different type of parasitophorous vacuole, is still unknown.

The control of Leishmania infection depends on the mounting of a Th1 immune response, with production of IFN-γ and TNF-α, which activate microbicidal mechanisms in infected macrophages, such as NO and reactive oxygen species (ROS) production (Kaye and Scott, 2011; Santos and Brodskyn, 2014). These free radicals cause damage to lipids, proteins and DNA, causing parasite death (Kima, 2014).

In this study, we determined the in vitro antileishmanial effect of vitamin D (D₂, D₃ and calcitriol), and investigated the role of vitamin D during in vitro infection with the intracellular protozoan L. amazonensis, focusing on the microbicidal responses of the infected macrophages.

Materials and methods

Compounds

Vitamins D₂, D₃ and calcitriol were purchased from Sigma-Aldrich (St. Louis, Missouri, USA) and diluted in ethanol. Amphotericin B, obtained from Cristália (Itapira, São Paulo, Brazil), was used as a positive control in antileishmanial tests.

Parasite culture

Wild-type (WT) Leishmania amazonensis [IFLA/BR/67/PH8] (L. amazonensis–WT) and L. amazonensis [MHOM/BR/75/Josefa] transfected with green fluorescent protein (GFP) (L. amazonensis–GFP) were cultured in Warren's medium, brain-heart infusion (HiMedia, Mumbai, Maharashtra, India) plus haemin and folic acid – both obtained from Sigma-Aldrich, supplemented with 10% foetal bovine serum (FBS) (Cultilab, Campinas, São Paulo, Brazil) (v/v) and 0.1% penicillin/streptomycin solution (Sigma-Aldrich) (v/v), at 25 °C in a BOD incubator.

Mice

BALB/c mice were obtained from the Central Animal Facility of UFRJ. The procedures using animals were performed according to protocols approved by the Ethical Committee for Animal Handling (CEUA 080/2018) from UFRJ.

Anti-parasite assays

Anti-promastigote test

Leishmania amazonensis–WT promastigotes, in the logarithmic phase of growth, were transferred to 96-well culture plates (3 × 10⁶ cells mL⁻¹) and exposed to different concentrations of vitamins D for 48 h at 25 °C. The cell viability was evaluated by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) method and readings were taken at 570 nm. The IC₅₀ values were calculated using GraFit software. Three independent experiments in triplicate were performed. Amphotericin B was used as the reference drug. MTT was obtained from Sigma-Aldrich.

Anti-amastigote test

Peritoneal macrophages of BALB/c mice (2 × 10⁶ cells mL⁻¹) were adhered in 24-well culture plates for 24 h at 37 °C under a 5% CO₂ atmosphere. Cells were infected with L. amazonensis–GFP promastigotes (20:1 ratio) or L. amazonensis–WT (10:1 ratio) for 4 h at 33 °C. After infection, the non-adhered promastigotes were washed and the cells were maintained in RPMI medium (Cultilab) supplemented with 10% FBS (v/v) and 0.5% penicillin/streptomycin (v/v) for 24 h. The vitamins were added to the cultures, at different concentrations for 48 h at 33 °C. For L. amazonensis–GFP, the parasite load was determined by measuring the fluorescence intensity of live amastigotes using a spectrofluorometer (485/528 nm). For L. amazonensis–WT, slides were stained with Giemsa and the number of amastigotes/200 macrophages was counted on an optical microscope, with subsequent determination of the percentage of infected macrophages and the number of amastigotes per macrophage. In each case (WT and GFP) three independent experiments in triplicate were performed. Infected macrophages that not received treatment were used as a control (incubated with culture medium containing FBS and antibiotic) and amphotericin B was used as the reference drug.

To test the effect of vitamin D₃ in combination with IFN-γ, vitamin D₃ (40 μm) and IFN-γ (1 ng mL⁻¹) were used together in the same infection protocol using L. amazonensis–GFP. In these experiments a treatment pre-infection and a treatment post-infection were performed. In the latter, the protocol used was the same as described above. In the former, the attached macrophages were treated with vitamin D, IFN-γ or both for 24 h, the cells were washed and infected with L. amazonensis–GFP promastigotes (20:1 ratio) for 4 h at 33  °C. The non-adhered promastigotes were washed and the cells were maintained in RPMI medium supplemented with 10% FBS (v/v) and 0.5% penicillin/streptomycin (v/v) for 48 h. The parasite load was determined by measuring the fluorescence intensity of live amastigotes using a spectrofluorometer (485/528 nm). Infected macrophages that not received treatment were used as a control (incubated with culture medium containing FBS and antibiotic). Three independent experiments in triplicate were performed. IFN-γ was obtained from the supernatant of L1210 cells cultivated in RPMI medium supplemented with 10% FBS (v/v), 1% non-essential amino acids (v/v) and 0.5% penicillin/streptomycin solution (v/v), at 37 °C with 5% CO₂.

Cytotoxicity assay

To evaluate the toxicity of the vitamins on mammalian cells, peritoneal macrophages from BALB/c mice (2 × 10⁶ macrophages mL⁻¹) were adhered in 96-well culture plates for 24 h at 37 °C in 5% CO₂. The vitamins were then added to the culture in different concentrations for 48 h at 37 °C and 5% CO₂. Cell viability was assessed by the MTT method. CC₅₀ values were calculated using GraFit software. Three independent experiments in triplicate were performed and 2% Triton X-100 (Dinâmica, Diadema, São Paulo, Brasil) was used as a positive control.

Reactive oxygen species assay

Peritoneal macrophages from BALB/c mice (1 × 10⁶ cells mL⁻¹) were plated in 96-well black microplates with a clear flat bottom. For the treatment post-infection, these cells were infected with L. amazonensis–WT promastigotes (10:1 ratio) for 4 h at 33 °C, non-adherent promastigotes were washed and after 24 h, the cells were treated with vitamin D₃ (40 μm), IFN-γ (1 ng mL⁻¹) or both for 48 h. For the treatment pre-infection, the macrophages were treated with vitamin D₃ (40 μm), IFN-γ (1 ng mL⁻¹) or both for 24 h, washed and infected with L. amazonensis–WT promastigotes (10:1 ratio) for 4 h at 33 °C, non-adherent promastigotes were washed and were incubated for 48 h. The ROS levels also were analysed in uninfected macrophages that had been treated for 48 h with vitamin D₃ (40 μm), IFN-γ (1 ng mL⁻¹) or both added at the same time as the post- and pre-treated infected macrophages. In all cases, the plates were washed with phosphate-buffered saline and 20 μm H₂DCFDA (Invitrogen, Carlsbad, California, USA), an oxidative probe, was added for 30 min. ROS production was evaluated using a spectrofluorometer (485/528 nm). Leishmania amazonensis-infected macrophages stimulated with H₂O₂ (4 mm) for 30 min were used as a positive control. Four independent experiments in triplicate were performed.

Nitrite assay

Nitrite production was determined in the supernatant of the cultures (uninfected and infected macrophages with L. amazonensis) from treatment pre- and post-infection with vitamin D₃ (40 μm), IFN-γ (1 ng mL⁻¹) or both, by the Griess method (Green et al., 1982). Nitrite concentration was estimated using a sodium nitrite standard curve. Leishmania amazonensis-infected macrophages stimulated with 1 ng mL⁻¹ IFN-γ were used as a positive control. Four independent experiments in triplicate were performed.

Statistical analysis

All the graphs presented in this paper show a set of data taken from three or four independent experiments, performed in triplicate.

Comparison between groups was made using one-way analysis of variance (ANOVA) followed by Dunnett post-test and the significance was considered when P < 0.05.

Results

Anti-Leishmania amazonensis activity of vitamin D

It has been previously reported that vitamin D can regulate innate and adaptive immune mechanisms (Ramos-Martínez et al., 2015), such as through the modulation of macrophage effector functions (Gough et al., 2017). As macrophages are the main host cell for the Leishmania parasites (Carlsen et al., 2015), the role of vitamin D during infection of macrophages by L. amazonensis was investigated.

Initially, we evaluated separately the toxicity of vitamins D₂ and D₃ against host macrophages and L. amazonensis promastigotes. Vitamins D₂ and D₃ presented some toxicity against peritoneal macrophages, with CC₅₀ values of 61.95 and 59.97 μm, respectively (Table 1). The positive control, 2% Triton X-100, killed 93.36% of macrophages. Both of these forms of vitamin D were active against L. amazonensis promastigotes, with IC₅₀ values of 36.47 and 39.64 μm, for vitamins D₂ and D₃ respectively (Table 1). Amphotericin B was used as a positive control and showed an IC₅₀ of 0.12 μm (Table 1 and Fig. S1A). The selectivity index (SI = CC₅₀ of macrophages/IC₅₀ against L. amazonensis promastigotes) of vitamins D₂ and D₃ gave values of 1.70 and 1.51, respectively (Table 1), which indicates that these vitamins are more toxic to the parasite in the promastigote form than to the macrophage.

Table 1.

Effect of vitamins D against peritoneal macrophages, promastigotes of Leishmania amazonensis and SI

	Peritoneal macrophages – CC50 (μm)a	L. amazonensis promastigotes – IC50 (μm)b	SIc
Vitamin D2	61.95 ± 1.49	36.47 ± 2.88	1.70
Vitamin D3	59.97 ± 1.07	39.64 ± 1.88	1.51
Amphotericin Bd	–	0.12 ± 0.01	–

^aCytotoxic concentration of 50% of macrophages (CC₅₀).

^bInhibitory concentration of 50% of growth parasites (IC₅₀). Values are the mean ± s.d.

^cSI, calculated by: CC₅₀ against macrophages/IC₅₀ against parasite.

^dAmphotericin B was used as reference drug.

Next, the effect of vitamins D₂ and D₃ on intracellular L. amazonensis amastigotes was evaluated. To avoid toxicity to the macrophages, the highest concentration of the vitamins tested was 50 μm. First, the anti-amastigote effect was determined by fluorimetry using L. amazonensis–GFP, which showed a reduction of the amastigote viability of 51.16 and 54.23% at 50 μm for vitamins D₂ and D₃, respectively, after 48 h of incubation (Table 2). The anti-amastigote effect was also assessed using the L. amazonensis–WT by manually counting the intracellular parasites, and the reduction of the amastigote viability at 50 μm was 60.80 and 66.89% for vitamins D₂ and D₃, respectively (Table 2). Still regarding L. amazonensis–WT, vitamins D₂ and D₃ significantly reduced the percentage of infected macrophages at concentrations of 50, 25 and 12.5 μm – P < 0.05 (Fig. 1A and C and Fig. S2). Furthermore, vitamin D₂ significantly decreased the number of amastigotes per macrophage in all concentrations tested – 50 μm to 6.25 μm – P < 0.0001 (Fig. 1B and Fig. S2), while vitamin D₃ caused this reduction at 50 to 12.5 μm – P < 0.0001 (Fig. 1D and Fig. S2). Amphotericin B was used as a positive control and inhibited about 65% of the amastigote growth at a concentration of 0.5 μm, with an IC₅₀ of 0.28 μm (Fig. S1B).

Table 2.

Effect of vitamins D against intracellular amastigotes of L. amazonensis (WT and GFP), promastigotes of L. amazonensis and peritoneal macrophages at 50 μm

	% inhibition at 50 μma
	L. amazonensis amastigotes		L. amazonensis promastigotes – WTb	Peritoneal macrophages
	WTb	GFPc
Vitamin D2	60.80 ± 4.67***	51.16 ± 2.02***	89.35 ± 9.81***	13.68 ± 7.72ns
Vitamin D3	66.89 ± 2.3***	54.23 ± 6.44***	64.19 ± 4.31***	18.64 ± 11.97ns

^aPercentage of inhibition compared to untreated control ± standard error of the mean (s.e.m.).

^bL. amazonensis – WT = not transfected.

^cL. amazonensis – GFP = transfected with green fluorescence protein (GFP).

***P < 0.0001 and ^nsnot significant (P = 0.0803), when compared with the percentage of inhibition of control group, that is 0%.

Fig. 1.

Effect of vitamins D₂ and D₃ against intracellular amastigotes of Leishmania amazonensis. Peritoneal macrophages infected with L. amazonensis amastigotes were treated with vitamin D₂ or D₃ for 48 h and the parasitic load was determined by counting the intracellular parasites after staining with Giemsa. Three independent experiments in triplicate were performed. (A and C) Percentage of infected macrophages with L. amazonensis amastigotes after treatment with vitamin D₂ (A) and D₃ (C). (B and D) The number of amastigotes per macrophage after treatment with vitamin D₂ (B) and D₃ (D). Graphs were prepared using the GraphPad Prism software, in which one-way ANOVA was used to realize the statistical analysis and Dunnett post-test was applied to compare the groups (50, 25, 12.5 and 6.25) with 0 μm (untreated control): P < 0.001 (***); P < 0.01 (**); P < 0.05 (*) and ns (not significant, P > 0.05).

Overall, Table 2 shows that at 50 μm the vitamins had low toxicity on peritoneal macrophages: 13.68 and 18.64% of cell death after treatment with for vitamins D₂ and D₃, respectively, when compared to untreated control, and these values showed no statistical difference in relation to the untreated control (P = 0.0803). However, these vitamins significantly inhibited (P < 0.0001) the growth of L. amazonensis promastigotes (89.35 and 64.19% for vitamins D₂ and D₃, respectively, compared to untreated control) and intracellular amastigotes (both vitamins inhibited about 50% of the GFP parasites and 60% of the WT parasites, when compared to untreated control).

The calcitriol, the most biologically active vitamin D metabolite, was not toxic to macrophages up to the highest concentration tested (5 μm) (Fig. S3A) and did not show an antileishmanial effect up to 5 μm (Fig. S3B and S3C).

Combination of vitamin D₃ and IFN-γ did not alter parasite load

It has previously been demonstrated that vitamin D₃ and IFN-γ in combination can increase the activity of human monocytes against Mycobacterium tuberculosis reducing the proliferation of this microorganism (Rook et al., 1986). Based on this, the combination of vitamin D₃ and IFN-γ during L. amazonensis infection in vitro was examined. Infected macrophages were treated during 48 h with 40 μm vitamin D₃ together with 1 ng mL⁻¹ IFN-γ but this did not enhance the amastigote killing compared to the vitamin D₃ alone (Fig. 2A). IFN-γ (1 ng mL⁻¹) inhibited about 15% of amastigote growth (Fig. 2A), when compared to untreated control. Pre-infection treatment with vitamin D₃ did not cause a significant inhibition of amastigote growth (only about 5%) and IFN-γ (1 ng mL⁻¹) inhibited about 10% of amastigote growth (Fig. 2B). Vitamin D₃ in combination with IFN-γ (pre-infection treatment) did not alter the effect of vitamin D₃ used separately (Fig. 2B). This combination nullified the antileishmanial effect of IFN-γ (Fig. 2B), when used separately.

Fig. 2.

Percentage of growth inhibition of L. amazonensis amastigotes after 24 h for treatment pre-infection and 48 h for treatment post-infection with vitamin D₃ (40 μm) – (1), IFN-γ (1 ng mL⁻¹) – (2) and vitamin D₃ (40 μm) + IFN-γ (1 ng mL⁻¹) – (1) + (2). Each value was calculated based on the control, which caused 0% of growth inhibition. The parasite load was determined by measuring the fluorescence intensity of live amastigotes (L. amazonensis–GFP) using a spectrofluorometer (485/528 nm). Three independent experiments in triplicate were performed. Graphs were prepared using GraphPad Prism software, in which one-way ANOVA was used to realize the statistical analysis and Dunnett post-test was applied to compare the groups indicated in the figure: P < 0.001 (***); P < 0.01 (**); P < 0.05 (*) and ns (not significant, P > 0.05). Control = macrophages that were infected with L. amazonensis, but did not receive the treatment.

Vitamin D reduced microbicidal mechanisms in macrophages

Vitamin D₃ has been reported to increase NO production from peripheral blood mononuclear cells (PBMCs) in Mycobacterium bovis-infected cattle (Waters et al., 2001). Therefore, we investigated the effect of vitamin D₃ in the production of NO and ROS by murine macrophages uninfected and infected with L. amazonensis.

The NO levels produced by L. amazonensis-infected macrophages and uninfected macrophages after treatment with vitamin D₃ (40 μm) were comparable to the untreated control (Fig. 3A). IFN-γ (1 ng mL⁻¹) significantly stimulated the NO production by both infected and uninfected macrophages (Fig. 3A). However, the combination of vitamin D₃ and IFN-γ deactivated the NO production induced by the IFN-γ (Fig. 3A). Similar results were observed for the pre-infection treatment with vitamin D₃ and vitamin D₃ with IFN-γ (Fig. 3B). Pre-infection treatment with vitamin D₃ (40 μm) maintained NO levels comparable to the untreated control; IFN-γ (1 ng mL⁻¹) significantly stimulated the NO production by both infected and uninfected macrophages; while the combination of 40 μm vitamin D₃ and 1 ng mL⁻¹ IFN-γ maintained NO levels compared to vitamin D₃ alone or to untreated control. Therefore, this pre-infection combination also was able to deactivate the NO production induced by IFN-γ used separately (Fig. 3B).

Fig. 3.

Effect of post-infection treatment (48 h) and pre-infection treatment (24 h) with vitamin D₃ (40 μm) – (1), IFN-γ (1 ng mL⁻¹) – (2) and vitamin D₃ (40 μm) + IFN-γ (1 ng mL⁻¹) – (1) + (2) on NO (A and B) and ROS (C and D) production by uninfected macrophages and macrophages infected with L. amazonensis amastigotes. (A and B) The NO levels were evaluated by the Griess method and the NO concentration is represented in μm. The treatment with IFN-γ 1 ng mL⁻¹ was used as a positive control. (C and D) The ROS levels were evaluated by fluorimetry after labelling with H₂DCFDA and the ROS amount is expressed in relative fluorescence units (RFU). Uninfected and infected macrophages with L. amazonensis treated with H₂O₂ (4 mm) were used as a positive control. Four independent experiments in triplicate were performed. Graphs were prepared using GraphPad Prism software, in which one-way ANOVA was used to realize the statistical analysis and Dunnett post-test was applied to compare all the groups with control and these analyses are represented above each column: P < 0.001 (***); P < 0.01 (**); P < 0.05 (*) and ns (not significant, P > 0.05). The uninfected groups only are compared with the uninfected control and the infected groups only are compared with the infected control. The IFN-γ 1 ng mL⁻¹ treatment – (2) was also compared with (vitamin D₃ + IFN-γ) treatment – (1) + (2) and these analyses are indicated above the line between these groups: P < 0.001 (***); P < 0.01 (**) and ns (not significant, P > 0.05). In the experiments using only uninfected macrophages, the control is macrophages untreated with vitamin D₃ or IFN-γ or both. In experiments with infected macrophages the control is macrophages that were infected with L. amazonensis, but did not receive the treatment.

Vitamin D₃ at 40 μm reduced ROS production from uninfected and L. amazonensis-infected macrophages (Fig. 3C). IFN-γ (1 ng mL⁻¹) and H₂O₂ (4 mm) stimulated significantly the ROS production by both infected and uninfected macrophages (Fig. 3C). The combination of vitamin D₃ and IFN-γ deactivated the ROS production induced by IFN-γ used alone (Fig. 3C), which was observed for both uninfected and infected macrophages, similarly to that which occurred for the NO production. Pre-infection treatment with vitamin D₃ maintained the ROS levels comparable to the untreated control, and the combination of vitamin D₃ and IFN-γ deactivated the ROS production induced by IFN-γ used separately in macrophages infected with L. amazonensis (Fig. 3D).

These results suggest that vitamin D₃ is able to modulate effector functions of macrophages, leading to a decrease in ROS levels produced by uninfected macrophages and macrophages infected with L. amazonensis. In addition, combination of vitamin D₃ and IFN-γ deactivates macrophage microbicidal mechanisms induced by IFN-γ, such as the NO and ROS production, by both uninfected and infected macrophages.

Discussion

Vitamin D interferes with the regulation of the immune response (Ramos-Martínez et al., 2015), with several immune targets such as dendritic cells, monocytes, macrophages, T cells and B cells (Hart et al., 2011), which are important for the treatment of infectious diseases (Ramos-Martínez et al., 2015). Interestingly, vitamin D₃, also known as cholecalciferol, has been shown to be effective against different strains of Mycobacterium spp., like M. avium subspecies paratuberculosis, M. avium subspecies avium and M. tuberculosis complex, inhibiting bacterial growth (Greenstein et al., 2012). It was also shown by Kim and colleagues that vitamin D contributes to the antimicrobial activity of macrophages against M. leprae, with the bacterial burden decreasing significantly with an increasing concentration of vitamin D (Kim et al., 2018). Against Leishmania parasites, another intracellular pathogen like Mycobacterium, the role of vitamin D is still under debate, as some studies show a positive role while others a negative. For example, Rodriguez-Cortes and colleagues have shown that symptomatic dogs naturally infected with Leishmania present lower vitamin D levels compared with asymptomatic or non-infected dogs (Rodriguez-Cortes et al., 2017). In agreement with this study, L. mexicana-infected mice treated with vitamin D exhibit a reduction in lesion size (Ramos-Martínez et al., 2013). In contrast, L. amazonensis-infected mice that received a vitamin D-deficient diet controlled the lesion development better than mice that received a regular diet (Bezerra et al., 2019). In addition, C57BL/6 mice deficient for the VDR and infected with L. major developed significantly smaller lesions and had a reduced inflammatory process when compared to C57BL/6 WT mice (Whitcomb et al., 2012). Therefore, it is necessary to use a simpler model to investigate the role of vitamin D, for this, the direct effect on macrophages during Leishmania infection can be evaluated.

In this study, an in vitro model of cutaneous leishmaniasis (infection by L. amazonensis) was used to assess the role of vitamin D. Vitamin D (D₂ and D₃) was slightly effective against L. amazonensis. Vitamin D is produced in the skin (Prietl et al., 2013), therefore it is in close proximity to the L. amazonensis parasite, which resides within cells, primarily macrophages, in the host skin (Anversa et al., 2018). In addition, the chemical structure of vitamin D reveals that it has lipophilic properties, which could favour the penetration of this compound into the host macrophage. These two facts tend to favour the effect of vitamin D against this intracellular parasite, as we observed. However, it is important to highlight that the in vitro concentrations used in this study would not be achieved in plasma in an in vivo model, since the plasma concentrations of vitamin D are: 50 nmol L⁻¹ (=0.02 μg mL⁻¹ or 0.05 μm), according to the Institute of Medicine (IOM) and 75 nmol L⁻¹ (=0.03 μg mL⁻¹ or 0.08 μm), according to the Endocrine Society (ES) (Bendik et al., 2014). In addition, it is worth noting that the concentration of vitamin D₃ in an inflammatory site appears to be greater than in the blood because activated monocytes and T lymphocytes, cells found in inflammatory infiltrates, metabolize vitamin D₃ to the active form (calcitriol) (Xu et al., 1993).

IFN-γ, a potent macrophage-activating cytokine, can induce an antimicrobial response in human macrophages cultured in vitamin D-sufficient sera, which does not occur with cells cultured in sera containing low amounts of vitamin D (Fabri et al., 2011). Rook and colleagues also showed that the combination of vitamin D₃ and IFN-γ increases the activity of human monocytes against M. tuberculosis (Rook et al., 1986). Previous study by Qi and colleagues observed an increase in the replication of the L. amazonensis within infected macrophages after treatment with IFN-γ (post-infection treatment) (Qi et al., 2004). In this study, IFN-γ poorly inhibited the L. amazonensis amastigote growth and the combination of vitamin D₃ and IFN-γ was not favourable to the leishmanicidal effect, as it caused a similar effect in the growth inhibition percentage as the vitamin D₃ alone. Helming and colleagues also showed that vitamin D suppressed the listericidal activity of bone marrow-derived macrophages stimulated by IFN-γ, due to an inhibition of oxidative burst (Helming et al., 2005). It was shown by Henard and colleagues that the overexpression of a tryparedoxin peroxidase in L. amazonensis increases the parasite resistance to toxicity mediated by peroxynitrite produced by activated macrophages (Henard et al., 2014). In a previous study conducted by Ehrchen and colleagues, the treatment of L. major-infected macrophages with vitamin D led to the conclusion that although there was no reduction in the parasitic load within the macrophages, this vitamin caused a reduction in the microbicidal properties of macrophages induced by IFN-γ, which appeared to be dependent on signalling by VDR (Ehrchen et al., 2007).

Interestingly, it has also been shown that vitamin D supplementation increases NO production from PBMCs of M. bovis-infected cattle, which contributes to bacterial death (Waters et al., 2001). Whitcomb and colleagues determined that, when pre-incubated with vitamin D prior to infection with L. major, NO production from dendritic cells of VDR knockout C57BL/6 mice was increased compared to dendritic cells from WT C57BL/6 mice. Nevertheless, no differences were observed in NO production by the macrophages from VDR knockout or WT C57BL/6 mice after pre-stimulation with vitamin D followed by infection with L. major (Whitcomb et al., 2012). In this study, we show that vitamin D (D₃) was able to modulate effector mechanisms of uninfected and L. amazonensis-infected macrophages, decreasing the ROS levels produced by these cells. However, the combination of vitamin D₃ and IFN-γ inhibited the microbicidal mechanisms, the NO and ROS production, in both uninfected and L. amazonensis-infected macrophages induced by IFN-γ, whereby the levels were lower than the treatment with IFN-γ alone.

Bacchetta and colleagues reported that vitamin D supplementation increases the production of antimicrobial peptides, such as cathelicidins, from peritoneal macrophages of patients on peritoneal dialysis (Bacchetta et al., 2014). Vanherwegen and colleagues also reported that vitamin D₃ can induce the production of cathelicidins (Vanherwegen et al., 2017), and this could be the mechanism involved in the antileishmanial effect of vitamin D against L. amazonensis amastigotes within macrophages, since it appears that NO and ROS production are not related to this process. It has already been reported that vitamin D may induce the production of antimicrobial peptides, such as LL-37, a member of the cathelicidin family (Kościuczuk et al., 2012). It has also been reported that vitamin D induces LL-37 expression, contributing to the host immune response against M. tuberculosis (Jo, 2010). Rivas-Santiago and colleagues (2008) stated that vitamin D stimulated the production of LL-37 in macrophages by sunlight through the skin. Because of this, sunbathing is recommended for patients with tuberculosis, in order to favour the production of LL-37 and consequently the kill of M. tuberculosis (Rivas-Santiago et al., 2008). In addition, the LL-37 peptide has been shown to have antileishmanial activity against promastigotes and amastigotes of L. donovani and L. major (Marr et al., 2016). It also has been shown that cathelicidins interact with membranes of pathogens causing permeation and disintegration of these cell membranes (Kościuczuk et al., 2012).

It has already been reported that vitamin D decreases the production of IL-12 and TNF-α from monocytes and macrophages, and IFN-γ from CD4⁺ T cells. In addition, vitamin D is able to decrease the differentiation of Th1 cells and increase the differentiation of regulatory T cells (Hart et al., 2011). All these effects promoted by vitamin D in the immune system are unfavourable to the development of an effective immune response against Leishmania spp., which thus favours the parasite growth and the consequent disease progression in the host. Our previous data have demonstrated that in the absence of vitamin D in the diet, there is a decrease in the lesion size of infected mice due to an increase of Th1 cells (Bezerra et al., 2019). However, in this paper, we have investigated the vitamin D role at the host cell level, and we suggest that vitamin D (D₂ and D₃) can directly cause a slight increase in the control of the parasite load. During in vivo experiments, a combination of effects may happen, it is important to point out that despite the direct effect on the macrophage, during a Th1 response with production of IFN-γ, vitamin D could inhibit the NO and ROS response, thus increasing susceptibility. These results provide important information to better understand the complexity of the role of vitamin D directly on macrophages during in vitro infection by the protozoan parasite L. amazonensis.

Financial support

This study was funded by Programa Jovem Cientista do Nosso Estado (FAPERJ – E-26/203.215/2015); Productivity Fellowships from CNPq (304712/2016-7), FAPERJ (E-26/202-422/2017) and the Agency for Support and Evaluation of Graduate Education (CAPES) Finance code 001.

Ethical standards

The procedures using animals were performed according to protocols approved by the Ethical Committee for Animal Handling (CEUA 080/2018) from UFRJ.

Conflict of interest

The authors declare that they do not have any conflict of interests.