Regulation of Mycobacterium-Specific Mononuclear Cell Responses by 25-Hydroxyvitamin D3

The active vitamin D metabolite, 1,25-dihydroxyvitamin D3 (1,25(OH)2D3), has been shown to be an important regulator of innate and adaptive immune function. In addition, synthesis of 1,25(OH)2D3 from 25-hydroxyvitamin D3 (25(OH)D3) by the enzyme 1α-hydroxylase in monocytes upon activation by TLR signaling has been found to regulate innate immune responses of monocytes in an intracrine fashion. In this study we wanted to determine what cells expressed 1α-hydroxylase in stimulated peripheral blood mononuclear cell (PBMC) cultures and if conversion of 25(OH)D3 to 1,25(OH)2D3 in PBMC cultures regulated antigen-specific immune responses. Initially, we found that stimulation of PBMCs from animals vaccinated with Mycobacterium bovis (M. bovis) BCG with purified protein derivative of M. bovis (M. bovis PPD) induced 1α-hydroxylase gene expression and that treatment with a physiological concentration of 25(OH)D3 down-regulated IFN-γ and IL-17F gene expression. Next, we stimulated PBMCs from M. bovis BCG-vaccinated and non-vaccinated cattle with M. bovis PPD and sorted them by FACS according to surface markers for monocytes/macrophages (CD14), B cells (IgM), and T cells (CD3). Sorting the PBMCs revealed that 1α-hydroxylase expression was induced in the monocytes and B cells, but not in the T cells. Furthermore, treatment of stimulated PBMCs with 25(OH)D3 down-regulated antigen-specific IFN-γ and IL-17F responses in the T cells, even though 1α-hydroxylase expression was not induced in the T cells. Based on evidence of no T cell 1α-hydroxylase we hypothesize that activated monocytes and B cells synthesize 1,25(OH)2D3 and that 1,25(OH)2D3 down-regulates antigen-specific expression of IFN-γ and IL-17F in T cells in a paracrine fashion.


Introduction
Substantial evidence supports the notion that vitamin D insufficiency (serum 25(OH)D 3 concentrations ,32 ng/mL or 80 nM) results in inadequate immune function and thus increased risk for infectious and autoimmune diseases [1]. For instance, an inverse correlation exists between serum 25(OH)D 3 and the risk for upper respiratory tract infections [2], tuberculosis [3,4], and multiple sclerosis [5,6]. Vitamin D supplementation also decreases the risk influenza A infection [7], decreases the relapse rate in multiple sclerosis patients [8], and enhances ex vivo immunity to Mycobacteria tuberculosis [9]. The actions of 1,25-dihydroxyvitamin D 3 (1,25(OH) 2 D 3 ; the active hormone) on innate and adaptive immunity and the ability of immune cells to synthesize 1,25(OH) 2 D 3 [10] provides further evidence for a link between vitamin D status and immune function. Understanding the mechanisms of vitamin D signaling in the immune system, consequently, provides critical insight for the vitamin D requirements of the immune system.
The metabolism of 1,25(OH) 2 D 3 is critical for immune function because of the potent effects of 1,25(OH) 2 D 3 on innate and adaptive immunity. The enzyme that synthesizes 1,25(OH) 2 D 3 from 25-hydroxyvitamin D 3 (25(OH)D 3 ) is 1a-hydroxylase (1a-OHase) [36]. In the vitamin D endocrine system, 1a-OHase is expressed in the kidney and is tightly regulated in response to calcium homeostasis via the parathyroid hormone in order to control the circulating concentration of 1,25(OH) 2 D 3 [37]. However, the circulating concentration of 1,25(OH) 2 D 3 does not affect vitamin D-mediated immune responses [38,39] and circulating 1,25(OH) 2 D 3 does not increase when the immune system is activated [40]. Rather, monocytes and macrophages express 1a-OHase in response to toll-like receptor (TLR) signaling, and this has been shown for humans, cattle, and mice [20,41,42]. In addition, dendritic cells, B cells and T cells also have been found to express 1a-OHase to some degree upon activation [43,44]. However, 1a-OHase is predominantly upregulated in the CD14 + cells (monocytes/macrophages) from the inflamed mammary gland during mastitis in cattle [45]. Consequently, induction of 1a-OHase in immune cells enables regulation of 1,25(OH) 2 D 3 concentration at sites of inflammation and this localized regulation is evident from animal models of inflammation. In cattle, the gene for 24-hydroxylase, the vitamin D catabolic enzyme that is highly upregulated by 1,25(OH) 2 D 3 , is expressed much higher in inflamed mammary tissue than in healthy tissue or circulating immune cells during mastitis [45]. Also in cattle, 1,25(OH) 2 D 3 accumulated in granulomas during tuberculosis [46]. Finally, the concentration of 1,25(OH) 2 D 3 increased in the spinal cords of mice during EAE, but did not change in serum [47]. Therefore, the immune system has a mechanism to control 1,25(OH) 2 D 3 concentration locally independent of the endocrine system.
Subsequently, local control of 1,25(OH) 2 D 3 metabolism by the immune system has been shown to have a significant impact on innate immunity [48]. For example, synthesis of 1,25(OH) 2 D 3 by 1a-OHase in human monocytes induces their expression of cathelicidin [41]. Similarly, synthesis of 1,25(OH) 2 D 3 by 1a-OHase in bovine monocytes enhances their expression of iNOS and RANTES [20]. So, 1,25(OH) 2 D 3 produced in monocytes acts in an intracrine fashion to regulate vitamin-responsive genes.
As for adaptive immunity, monocyte production of 1,25(OH) 2 D 3 has been suggested to also regulate T cell responses in a paracrine fashion [48]. However, lymphocytes also may be a source of 1,25(OH) 2 D 3 and regulation of antigen-specific immune responses of T cells by conversion of 25(OH)D 3 to 1,25(OH) 2 D 3 in either monocytes or lymphocytes has yet to be shown. Therefore, the objectives of this study were to evaluate 1a-OHase gene expression in PBMC cultures in response to antigen stimulation and determine the effects of 25(OH)D 3 on innate and adaptive immune responses in PBMC cultures.
To accomplish the objectives of this study we use PBMCs from calves vaccinated with Mycobacterium bovis bacilli Calmette-Guerin (M. bovis BCG), which elicits strong IFN-c and IL-17 responses to purified protein derivative (PPD) of M. bovis [49]. The calf immune system has been found to serve as a good model of the human immune system for the study of tuberculosis and M. bovis BCG vaccination [50,51]. In addition, the concentration of 25(OH)D 3 circulating in blood is similar between cattle and humans with typical concentrations ranging from 20 to 100 ng/mL in both species [52,53,54]. In cattle and humans symptoms of vitamin D toxicity is rarely observed with circulating 25(OH)D 3 concentrations below 200 ng/mL [8,55,56]. Finally, as mentioned already, local control of 1,25(OH) 2 D 3 synthesis by the immune system and 1,25(OH) 2 D 3 -regulation of T cell responses is similar between cattle and humans. Therefore, the outcome of this study will provide insight on the mechanisms of vitamin D signaling in the human and bovine immune systems.

Animals
Twelve male Holstein calves that were approximately 5 months to 12 months of age were used for this study. At 14 d of age, 8 calves were vaccinated subcutaneously in the midcervical region with 10 7 cfu of M. bovis BCG (Pasteur strain). M. bovis BCG was prepared for vaccination as previously described [57]. The remaining 4 calves were not vaccinated. The NADC animal care and use committee approved the care and treatment of animals used in this study (Animal Protocol #ARS-3982).

Peripheral blood mononuclear cell cultures
Blood from the jugular vein was collected in 26 acid citrate dextrose. Blood was centrifuged and buffy coats were collected. Contaminating RBCs were removed by hypotonic lysis. PBMCs were isolated by density gradient centrifugation. PBMC were resuspended in RPMI 1640 (Sigma-Aldrich, St. Louis, MO) supplemented with 50 mg/ml gentamicin (Invitrogen, Carlsbad, CA). For gene expression assays, PBMCs were cultured at a concentration of 1.5610 7 cells/ml in 96-well (200 ml/well) or 6well (2 ml/well) tissue culture plates for 24 h at 37uC in 5% CO 2 . For determination of nitric oxide and IFN-c production, PBMCs were cultured at 1610 6

Relative gene expression
RNA was isolated from PBMC using the RNeasy Mini Kit (Qiagen, Valencia, CA). RNA samples were reverse transcribed to cDNA in 20 ml reactions using the High Capacity Reverse Transcription Kit with RNase inhibitor and random primers (Applied Biosystems, Foster City, CA). The reverse transcription reactions were incubated for 2 h at 37uC followed by 5 s at 85uC and finally cooled to 4uC. The cDNA samples were diluted 1:10 in water and stored at 220uC.
The amount of specific cDNA transcripts in each sample was determined using the 7300 Real-Time PCR System (Applied Biosystems). Each reaction contained 12.5 ml SYBR Green Master Mix (Applied Biosystems), 7.5 ml of cDNA sample, and 5 ml of 10 mM forward and 10 mM reverse primers. Reactions were incubated as follows: 95uC for 10 min followed by 40 cycles of 95uC for 15 s and 60uC for 60 s. Primer sets were designed with Primer3 (http://frodo.wi.mit.edu/primer3) [58] to span intronexon boundaries and are listed in Table 1. Primers were purchased from Integrated DNA Technologies (Coralville, IA). The efficiency of each primer set was determined as previously described [20] and fit the criteria required for quantification by real-time PCR [59]. The specificity of each primer set was verified by melting curve analysis and gel electrophoresis. The amounts of cDNA transcripts were normalized to ribosomal protein S9 (RPS9) cDNA. Expression of RPS9 also was compared to b-actin and GAPDH expression to verify its stability over treatment conditions. The relative expression of each gene was determined using the 2 2DDCt method [59]. The expression of each gene is relative to the normalized amount of each cDNA transcript in the nonstimulated controls for each experiment.

Measurement of nitric oxide production
Production of nitric oxide by PBMCs was determined by measurement of nitrite in the culture supernatant by using the Griess assay as previously described [20]. Supernatants (100 mL) from PBMC cultures were added to an equal volume of Griess reagent [0.5% sulfanilamide, 2.5% phosphoric acid, and 0.05% N-(1-naphthyl) ethylenediamine dihydrochloride; Sigma-Aldrich] in a 96-well clear bottom plate. Absorbance at 550 nm in each well was measured using a FlexStation 3 plate reader (Molecular Devices, Sunnyvale, CA). Absorbance values were converted to micromoles per liter using a standard curve that was generated by addition of 0 to 100 mM sodium nitrite to fresh culture media.

Measurement of IFN-c production
The concentration of IFN-c in PBMC culture supernatants was determined by an ELISA using the Endogen Bovine IFNc Screening Set (Pierce Biotechnology, Rockford IL) according to the manufacturers instructions. The absorbance at 450 nm minus the absorbance at 550 nm was measured with the FlexStation 3 plate reader and the values were converted to picograms per milliliter by using a standard curve.

Statistical Analysis
Analysis of variance was performed using PROC GLM of SAS (SAS Institute INC., Cary, NC). The model accounted for effects of treatment, cell type, and calf or vaccination status. DDCt values were used in the analyses of gene expression. The average DDCt values 6 SE were transformed using the equation 2 2DDCt . The expression of each gene is presented as the mean fold increase 6 SE relative to non-stimulated controls. Multiple comparison tests of the means were made using the Tukey adjustment.

M. bovis PPD-activation of vitamin D signaling in PBMCs
By stimulating PBMCs from M. bovis-BCG-vaccinated calves with LPS, PWM, or M. bovis PPD, we found that 1a-OHase gene expression in the PBMCs was upregulated by LPS, PWM, and M. bovis PPD stimulation relative to non-stimulated PBMCs (P,0.001; Fig. 1A). In contrast, VDR gene expression in the PBMC cultures was not upregulated by any of the stimulants (Fig. 1B). We also measured iNOS, RANTES, IFN-c, IL-17A, and IL-17F gene expression. Neither iNOS nor RANTES was affected by M. bovis PPD or LPS stimulation, but RANTES was upregulated by PWM stimulation (Fig. 1C and D). IFN-c, IL-17A, and IL-17F were upregulated in PBMCs stimulated with PWM or M. bovis PPD, however, they were not affected by LPS stimulation (Fig. 1E-G).

Cell type-specific expression of 1a-OHase and VDR
Several cell types have been reported to express 1a-OHase, including activated monocytes, T cells, and B cells [41,43,44]. We sorted PBMCs that had been stimulated with M. bovis PPD from BCG-vaccinated animals according to surface expression of CD3, IgM, and CD14 by using FACS (Fig. 4) to determine what populations of cells in PBMCs were expressing 1a-OHase upon activation. By sorting the stimulated PBMCs, we found that 1a-OHase was predominantly expressed in the CD14 + population of cells (P,0.001; Fig. 4A). 1a-OHase expression was also induced in IgM + cells from vaccinated calves (P,0.001). Relative to 1a-OHase expression in non-stimulated, non-sorted PBMCs, the expression of 1a-OHase did not increase in the CD3 + cells isolated from the stimulated PBMC cultures (Fig. 4A). Unlike 1a-OHase, VDR gene expression did not differ significantly between cell types in PBMC cultures from vaccinated calves, but IgM + cells from 25(OH)D 3 treated cultures did have somewhat lower VDR expression (Fig. 4B).

Cell type-specific effects of 25(OH)D 3 on gene expression
We also compared gene expression in cells from M. bovis PPDstimulated PBMCs that were treated with 100 ng/ml 25(OH)D 3 with cells from M. bovis PPD-stimulated PBMCs that were not treated with 25(OH)D 3 . Treatment with 25(OH)D 3 increased 24-OHase, iNOS, and RANTES gene expression in both CD14 + cells and IgM + cells from the BCG-vaccinated calves (P,0.05; Fig. 4C-E). In contrast, 25(OH)D 3 treatment decreased expression of IFNc by over 60% and IL-17F by nearly 50% in the CD3 + cells from the BCG-vaccinated calves (P,0.05; Fig. 4F and H). IL-17A expression in the CD3 + cells was also down-regulated by 25(OH)D 3 treatment, but to a lesser extent (P.0.05; Fig. 4G).

Comparison of responses between BCG-vaccinated and non-vaccinated animals
Finally, we compared changes in gene expression caused by M. bovis PPD stimulation and 25(OH)D 3 in cells from non-vaccinated animals to the changes observed in cells from BCG-vaccinated animals. In PBMCs from non-vaccinated animals, 1a-OHase was induced in CD14 + cells by M. bovis PPD stimulation like in CD14 + cells from BCG-vaccinated animals (Fig. 4A). Unlike the PBMCs from BCG-vaccinated calves, VDR was not detected in IgM + cells from the non-vaccinated animals (Fig. 4B). VDR expression in CD3 + and CD14 + cells was similar between vaccinated and nonvaccinated calves. Neither 24-OHase, iNOS, or RANTES expression was affected by 25(OH)D 3 in CD14 + cells and IgM + cells from the non-vaccinated animals like it was in the BCGvaccinated animals (Fig. 4C-E). Finally, IFN-c, IL-17A, and IL-17F were not induced by M. bovis PPD stimulation in the CD3 + cells from non-vaccinated animals as they were in CD3 + cells from BCG-vaccinated animals (Fig. 4F-H).

Discussion
For over two decades now, 1,25(OH) 2 D 3 has been known as an important regulator of adaptive immunity, suppressing lymphocyte proliferation and IFN-c production [28,60]. The implications of 1,25(OH) 2 D 3 on adaptive immunity are further realized in animal models of T cell-mediated autoimmunity as 1,25(OH) 2 D 3 inhibits disease progression [32,34,61,62]. Recently, 1,25(OH) 2 D 3 also was found to be an important regulator of innate immunity by enhancing antimicrobial properties of macrophages [16,18]. Vitamin D-mediated immune responses, however, do not correlate with the circulating concentration of 1,25(OH) 2 D 3 [38,39]. Therefore, local synthesis of 1,25(OH) 2 D 3 is a critical factor in regulating both innate and adaptive immunity. Previously, induction of 1a-OHase expression in macrophages was shown to occur upon activation by TLR 2/1 or TLR 4 signaling and enable them to convert 25(OH)D 3 to 1,25(OH) 2 D 3 [20,41,63]. Synthesis of 1,25(OH) 2 D 3 in macrophages, in turn, enhanced their innate antimicrobial properties in an intracrine fashion [48]. In this study, we give evidence that endogenous synthesis of 1,25(OH) 2 D 3 also occurs in antigen-stimulated PBMC cultures and regulates key aspects of adaptive immunity.
In this study we found that 1a-OHase gene expression was induced in CD14 + cells (monocytes/macrophages) and IgM + (B cells), but not in CD3 + (T cells) cells in M. bovis PPD-stimulated PBMC cultures. Furthermore, treatment of M. bovis PPDstimulated PBMC cultures with 25(OH)D 3 enhanced iNOS and RANTES expression in monocytes and B cells and suppressed antigen-specific IFN-c and IL-17F responses in T cells. Based on this evidence, we hypothesize that 1,25(OH) 2 D 3 was produced in monocytes and B cells acted on monocytes and B cells in an intracrine fashion to upregulate iNOS and RANTES expression and on T cells in a paracrine fashion to suppress M. bovis PPDspecific IFN-c and IL-17F responses (Fig. 5).
M. bovis PPD is a crude extract and as such likely contains antigens that activate both the innate and adaptive immune systems. Therefore, in PBMC cultures innate antigen presenting cells (e.g., monocytes) recognize TLR ligands, such as lipoproteins, become activated and then express 1a-OHase. The APCs also internalize protein from the M. bovis PPD, process it and then present it on their surface as peptide associated with MHC. Activation of T cells specific for M. bovis PPD is then caused by the interaction of the specific T cell receptor (TCR) with its cognate MHC/antigen. Likewise, B cells recognize antigen through IgM on their surfaces, and along with co-stimulation from T cells, become activated and express 1a-OHase. We suggest that production of 1,25(OH) 2 D 3 by 1a-OHase in activated monocytes and B cells can alter the IFN-c and IL-17F responses that are the result of the TCR/MHC/antigen interaction between T cells and APCs.
There are multiple possibilities as to how 1,25(OH) 2 D 3 suppressed IFN-c and IL-17F gene expression in T cells. VDR expression in the T cells was similar to that in monocytes in this study and purified T cells do respond to 1,25(OH) 2 D 3 [25,64]. Also, T cell VDR expression is required for 1,25(OH) 2 D 3 -mediated inhibition of experimental autoimmune encephalomyelitis in mice [35]. Therefore, the T cells in the PBMC cultures likely had the ability to respond to 1,25(OH) 2 D 3 secreted from the monocytes and B cells. Consequently, activation of the VDR in T cells could have directly suppressed IFN-c and IL-17F expression. However, 1,25(OH) 2 D 3 failed to suppress IFN-c production in fully differentiated Th1 cells [61]; so, 1,25(OH) 2 D 3 may have regulated genes in T cells that influenced T cell differentiation or sensitized them to apoptosis. Alternatively, up-regulation of nitric oxide production by 1,25(OH) 2 D 3 in monocytes and B cells could have induced apoptosis in the surrounding T cells and resulted in suppressed IFN-c and IL-17F expression. A combination of several mechanisms also is possible and we have not ruled out the In any case, treatment of antigen-stimulated PBMCs with 25(OH)D 3 suppressed antigen-specific IFN-c and IL-17F expression in T cells, which indicates that synthesis of 1,25(OH) 2 D 3 by immune cells has significant implications in regulating adaptive immunity. IFN-c is a potent activator of macrophages and is mainly produced by Th1 cells [65]. Th1-mediated responses are critical in the defense against intracellular infections, such as tuberculosis [66,67]. IL-17A and IL-17F are produced by Th17 cells and play major roles in neutrophil recruitment and protection against intracellular and extracellular bacterial infections [68,69,70]. Self reactive Th1 and Th17 cells, however, are involved in the development of autoimmune disorders [71] and inhibition of animal models of autoimmunity by 1,25(OH) 2 D 3 is thought to occur, in part, by suppression of self-reactive Th1 and Th17 cells [72]. Although suppression of Th1 and Th17 responses to bacterial antigens by 1,25(OH) 2 D 3 would seem to attenuate the immune response against bacterial infections, keep in mind that 1,25(OH) 2 D 3 also enhances the antimicrobial activity of macrophages [41]. So overall, production of 1,25(OH) 2 D 3 by immune cells serves to limit inflammation caused by Th1 and Th17 effector cells, but ultimately improves defense against bacterial infections by boosting the innate antimicrobial response.
In addition to suppression of T cell responses by 1,25(OH) 2 D 3 synthesis in PBMC cultures, treatment of PBMCs with 25(OH)D 3 upregulated antigen-specific B cell iNOS and RANTES expression. We had previously shown that monocyte iNOS and RANTES expression depends on availability of 25(OH)D 3 [20], but not B cell iNOS and RANTES. Nitric oxide produced by iNOS in macrophages is considered to be an antimicrobial molecule. However, nitric oxide produced by the monocytes and B cells may suppress proliferation of T cells [73]. So, as mentioned above, 1,25(OH) 2 D 3 may suppress T cell responses in part by enhancing B cell and monocyte iNOS expression. RANTES is a chemokine originally found to be expressed by T cells [74], but also has been found to be expressed by alveolar macrophages in cattle [75]. We speculate that upregulation of RANTES in monocytes and B cells by 1,25(OH) 2 D 3 would enhance recruitment of immune cells to the site of inflammation, but the implications of 1,25(OH) 2 D 3 -upregulation of RANTES in monocytes and B cells will need to be investigated.
Finally, the ability of 1a-OHase in monocytes and B cells to synthesize 1,25(OH) 2 D 3 , and subsequently regulate 1,25(OH) 2 D 3mediated immune responses, depends on the availability of 25(OH)D 3 . The circulating concentration of 25(OH)D is primarily regulated by dietary intake of vitamin D 3 and sun exposure [76]. Current recommendations for vitamin D in humans and cattle target a circulating concentration of 25(OH)D of 20 to 50 ng/ml [55,77]. However, 25(OH)D concentrations above 30 ng/ml may be necessary for optimal immune function [1]. In addition, vitamin D insufficiency (serum 25(OH)D below 30 ng/ml) and even deficiency (serum 25(OH)D below 20 ng/ml) is widespread, indicating that current recommendations for vitamin D 3 intake may be inadequate [78,79]. Previously, and here, we have shown that 1,25(OH) 2 D 3 -regulated innate immune responses increase linearly from 0 to 125 ng/ml of 25(OH)D 3 [20]. This observation leads to the question of what concentration is necessary for optimal immune functionality if below 30 ng/ml is insufficient? Based on the requirement of 25(OH)D 3 by the immune system for signaling mechanisms and evidence from epidemiological studies, vitamin D requirements need to be re-evaluated to ensure proper immune function.