Methylation of the Vitamin D Receptor (VDR) Gene, Together with Genetic Variation, Race, and Environment Influence the Signaling Efficacy of the Toll-Like Receptor 2/1-VDR Pathway

Background The disparity in prevalence of infectious diseases across the globe is common knowledge. Vitamin D receptor (VDR)-mediated toll-like receptor (TLR) 2/1 signaling produces antimicrobial peptides, which is critical as a first line of defense in innate immunity. Numerous studies disclosed the independent role of genetic polymorphisms in this pathway, vitamin D status or season and more recently epigenetics, as factors contributing to infectious disease predisposition. Few studies address the interaction between environment, genetics, and epigenetics. Here, we hypothesized that VDR-mediated TLR2/1 signaling is influenced by a combination of environment, epigenetics and genetics, collectively influencing differential innate immunity. Methods Healthy Black and White South Africans (n = 100) donated blood, while ultraviolet index (UVI) was recorded for the duration of the study. LC-MS/MS supported 25(OH)D3 quantification. Monocyte/macrophage cultures, supplemented with/without 1,25(OH)2D3, were activated with the TLR2/1 elicitor, Pam3CSK4. VDR, cathelicidin antimicrobial peptide, hCAP-18, and 25-hydroxyvitamin D3-24-hydroxylase expression were quantified by RT-qPCR or flow cytometry. Pyrosequencing facilitated VDR methylation analysis and single-nucleotide polymorphism (SNP) genotyping in regions pinpointed through a bioinformatics workflow. Results Season interacted with race showing 25(OH)D3 deficiency in Blacks. UVI correlated with 25(OH)D3 and VDR methylation, likely influencing race differences in the latter. Regarding the TLR2/1 pathway, race differences in SNP genotype distribution were confirmed and functional analysis of VDR-mediated signaling showed interaction between race, season, and 25(OH)D3 status. Multivariate OPLS-DA mirrored several interactions between UVI, 25(OH)D3 status, DNA sequence, and methylation variants. Methylation of the third cytosine-phosphate-guanine dinucleotide (CpG) in the promoter CpG island (CGI) 1062, CGI 1062 CpG 3, significantly discriminated a 5.7-fold above average mean in VDR protein level upon TLR2/1 elicitation, the variation of which was further influenced by 25(OH)D3 status and the VDR SNP TaqI. Conclusion Regulation of VDR-mediated TLR2/1 signaling is multifactorial, involving interaction between environment [UVI and consequent 25(OH)D3 status], epigenetics (VDR methylation at key regulatory sites), and genetics (TLR1, TIRAP, and VDR SNPs).

In addition to its role in maintaining calcium-phosphorus homeostasis, vitamin D is a potent modulator of both innate and adaptive immunity, is involved in the regulation of cell growth and differentiation, detoxification of xenobiotics, and activation of monocytes/macrophages (1,2). These actions of vitamin D are almost entirely dependent on the interaction between the most biologically active form of vitamin D, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3], and the vitamin D receptor (VDR) transcription factor. Bound to 1,25(OH)2D3, the VDR regulates the expression of a myriad of genes (3,4). Cathelicidin antimicrobial peptide (CAMP) and 25-hydroxyvitamin D3-24-hydroxylase (CYP24A1) are examples of two well-characterized vitamin D target genes, respectively, encoding the cathelicidin antimicrobial peptide (hCAP-18) and multifunctional vitamin D catabolizing enzyme. Toll-like receptor 2/1 (TLR2/1) triggering activates a signaling cascade inducing both VDR (5) and CYP27B1 in monocytes/macrophages. CYP27B1 catalyzes de novo production of 1,25(OH)2D3 from accumulated 25(OH)D3; delivered to the cells via the vitamin D binding protein (DBP), encoded by GC. The liganded VDR-transcription factor complex binds to vitamin D response elements (VDREs) in CAMP, activating CAMP expression and the production of hCAP-18. hCAP-18 is synthesized as a proprotein consisting of an N-terminal cathelin domain and a C-terminal LL-37 domain (6). While the cathelin domain is a cysteine protease inhibitor with broad spectrum antibacterial activity (7), LL-37 directly inhibits mycobacterial replication (5,8), has antifungal activity against Candida albicans (9), and antiviral activity against HIV (10).
Since CAMP expression is dependent on vitamin D, and vitamin D deficiency has been linked to several infectious diseases including tuberculosis (11), sepsis (12), bacterial infections after kidney transplants (13), and HIV (14), it is not surprising that more than 100 clinical trials have assessed the efficacy of vitamin D supplementation as adjunct therapy in the treatment of various infectious diseases. However, the outcome of clinical trials has been conflicting and this is often attributed to differences in study design, baseline vitamin D status of participants, and outcome measurements. In fact, it appears that individuals can be classified into three groups: (i) those with a low response, (ii) those with a medium response, and (iii) those with a high response to vitamin D supplementation (15). These interindividual differences may result from variation in the regulation of VDR expression at both a genetic and epigenetic level (16). For example, VDR function to transactivate CAMP is influenced by the VDR single nucleotide polymorphism (SNP) FokI (rs2228570) and ethnicity (17), while vitamin D insensitivity in breast cancer cells has been attributed to CpG methylation of the VDR primary promoter (18). Thus, both genetics and epigenetics have the potential to influence the response to vitamin D. Indeed, a double-blind randomized controlled trial assessing the impact of high-dose vitamin D3 during intensive-phase antimicrobial treatment of pulmonary tuberculosis showed that vitamin D only increased the time for sputum culture conversion in participants carrying the CC genotype of the VDR SNP TaqI (rs731236) (19). Additionally, VDR expression is influenced by the environment. For example, narrow-band UVB induces miRNA-125b (20), which directly regulates VDR mRNA translation, decreasing VDR protein level (21,22). Seasonal variation in ultraviolet index (UVI) further correlates with circulating vitamin D (23). Thus, seasonal variation in UVI directly influences VDR function by altering the availability of the free 1,25(OH)2D3. Indirectly, changes in 1,25(OH)2D3 concentration may itself regulate VDR expression through multifunctional, 1,25(OH)2D3-responsive, enhancers located within the VDR itself (24). The complex regulation of the VDR through genetics, epigenetics, and environment (16) may therefore provide insight into inter-individual variation in response to vitamin D and the efficacy of vitamin D to enhance immune function.
Here, we evaluate (1) the effect of VDR methylation on the TLR2/1-VDR signaling pathway and (2) the impact of genetic and environmental factors on differential immune signaling. It was hypothesized that VDR-mediated TLR2/1 signaling is influenced by a combination of environment, epigenetics and genetics, collectively influencing differential innate immunity in healthy South Africans. Using an in vitro model, stimulating monocytes from healthy individuals with a TLR2/1 elicitor, we avoided pathogen-mediated changes in DNA methylation (25,26).
Results presented here provide support for multifactorial regulation of VDR-mediated, TLR2/1 signaling, involving interaction between environment, epigenetics, and genetics. UVI TLR2/

Bioinformatics
To identify putative functional loci that could influence VDR expression and function through genetic and/or epigenetic mechanisms, a bioinformatics workflow was developed (Methods S1.1, Figures S1 and S2, and Table S1 in Supplementary Material).

Monocyte/Macrophage culture and Treatment
To estimate TLR2/1 pathway efficacy, VDR mRNA, VDR protein, CAMP mRNA, hCAP-18 peptide, and CYP24A1 mRNA, hereafter referred to as functional variables, were quantified following different treatments of monocyte/macrophage cultures that were established as previously described (17). Some monocytes were retained for functional analysis at time zero (baseline

mrna and Protein Quantification
The relative level of VDR, CAMP, and CYP24A1 mRNA was quantified by RT-qPCR and VDR protein and intracellular hCAP-18 peptide by flow cytometry as previously described (17). Gene normalization was performed against two stably expressed reference genes: ubiquitin C (UBC) and tyrosine-3-monooxygenase/tryptophan-5-monooxygenase activation protein, zeta polypeptide (YWHAZ). Gene expression was quantified using the comparative CT method according to the MIQE guidelines, using inter-run calibrators and qBASE PLUS software. To compensate for variability in fluorescence readings between experiments on the flow cytometer, the median fluorescence intensity (MFI) of broad-spectrum calibration beads was used to normalize data and thereby provide a calibrator for instrument-related variation in the flow cytometry readings over time.   Pearson or Spearman's rho. A general linear model was used for multivariate analysis of variance to assess main effects and factor interaction. Mann-Whitney U tested methylation differences.

Tlr-VDr Pathway genetics Differs between races
To assess genetic variation between individuals in the TLR2/1-VDR signaling pathway, SNPs in several genes of the pathway  Table 2). Treatment had a significant main effect on functional variables (Figure 3), while season, race, and 25(OH)D3 status showed several complex interactions regarding VDR mRNA and VDR protein (Figure 4), CAMP mRNA and CYP24A1 mRNA ( Figure 5) and hCAP-18 ( Figure 6). TLR2/1 elicitation induced VDR protein ( Figure 3B, Considering interactions, 25(OH)D3-deficient Blacks had significantly lower VDR mRNA in summer than deficient Whites or Whites and Blacks with a normal 25(OH)D3 status (P < 0.050, Figure 4A). In contrast, 25(OH)D3-deficient Whites had significantly lower VDR mRNA in autumn than deficient Blacks or Whites and Blacks with a normal 25(OH)D3 status (P < 0.050). VDR protein dropped significantly in summer and autumn for Whites and Blacks, respectively, showing a significant race difference in summer (P < 0.050, Figure 4B). CAMP mRNA increased significantly from summer to autumn in Whites, being significantly higher than Blacks (P < 0.050, Figure 5A). Whites with a normal 25(OH)D3 status had significantly higher CYP24A1 mRNA than 25(OH)D3-deficient Whites or Blacks and Blacks with a normal 25(OH)D3 status (P < 0.050, Figure 5B). A notable decrease in hCAP-18 was observed in Whites from spring through summer to autumn being significantly higher in spring and significantly lower in autumn, compared to Blacks (P < 0.050, Figure 6A). All individuals with a normal 25(OH)D3 status showed a similar, significant decrease in hCAP-18 from spring to autumn (P < 0.050, Figure 6B). Blacks with a normal 25(OH)D3 status had significantly more hCAP-18, than normal Whites and deficient Blacks or Whites (P < 0.050, Figure 6C).
To confirm that the intracellular decrease in hCAP-18 in response to seasons with higher 1,25(OH)2D3 reflects hCAP-18 processing and LL-37 secretion, we performed Western blotting on 20 additional randomly selected healthy Black (n = 10) and White (n = 10) South Africans. These individuals, collected in winter and for whom no other variables were quantified, were also included in 25(OH)D3 quantification (shown in Figure 1). Western blotting showed individual-specific hCAP-18 processing and LL-37 secretion, which depended on 25(OH)D3 status, extent of 1,25(OH)2D3 supplementation, and incubation time (Figure 7). For example, an individual with a sufficient 25(OH)D3 status (>50 nmol/L, Figure 7A) had the highest level of intracellular LL-37 under control condition and already  Figure 7D). It should be noted that secreted LL-37 was undetectably low in 13 of the 20 randomly selected individuals subjected to Western blotting.

correlation analysis
To further explore the relation between independent variables (UVI, 25(OH)D3 concentration, age and regional methylation) and dependent functional variables, following different treatments, correlation analysis was performed ( The combined impact of genetics, epigenetics, and environment on Tlr2/1-VDr signaling To identify the main variables underlying differential levels of VDR and downstream targets (CAMP, hCAP-18, and CYP24A1), the multivariate OPLS-DA statistical method was performed and validated as described (Methods S1.3 in Supplementary Material). Evaluation of the loadings S-plots (Methods S1.3 and Figure S4 in Supplementary Material) and descriptive assessment of the scores space for each model, identified combined effects of genetics, VDR methylation, vitamin D status, and UVI on the efficacy of TLR2/1-VDR signaling, as assessed by the level of functional variables produced in response to various treatments. Functional variables were categorized as above/ below average and X (independent) variables that significantly (P ≤ 0.050) and/or measurably (≥1.5-fold or ≤0.667) discriminate mean values for above/below average response were recorded with their correlation ( Table 4). Methylation at CGI 1062, CpG 3 significantly and most notably discriminate mean values for above and below average VDR protein level, particularly with TLR2/1 elicitation. Other sizeable methylation-function interactions observed that were significant and occurred in at least two treatments of a functional variable included 1060 CpG 6 [positive impact on VDR mRNA with 1,25(OH)2D3

DiscUssiOn
Using a healthy South African cohort, we studied the combined effect of VDR methylation, TLR2/1-VDR pathway SNPs, and environment on TLR2/1 signaling and inter-individual variation in the response to vitamin D supplementation.  Results support race-related seasonal variation in 25(OH)D3 status (29), though, on average, Blacks were 25(OH)D3 deficient irrespective of season (Figure 1). Since TLR2/1-VDR signaling to induce CAMP greatly depends on the availability of 1,25(OH)2D3, Black South Africans may have an overall weaker immune response to bacterial pathogens and may benefit from vitamin D supplementation all year round. However, 25(OH)D3, season, and race showed complex interactions that influence TLR2/1-VDR signaling, rendering blanket supplementation presumptuous. For example, not all Blacks in the current study were 25(OH)D3 deficient (14% were sufficient). Western blotting of secreted LL-37 showed a decreased LL-37 secretion upon supplementation of 25(OH)D3 sufficient individuals (Figure 7), supporting the proposed U-shaped relationship between serum 25(OH)D3 and health (30).
As expected for regional methylation of expressed genes (31), VDR enhancer (1066) and promoter (1062) CGIs were hypomethylated, while the gene-body CGIs (1061 and 1060) were hypermethylated (Figure 2). Blacks had significantly higher methylation at CGI 1062 and CGI 1060a, but lower methylation at 1060b than Whites. The race-specific variation in methylation observed here agrees with Heyn et al. (32) and Adkins et al. (33) who independently showed genome-wide methylation differences between populations, contributing to natural variation. Similarly, Andraos et al. (34) showed significantly higher methylation levels in the Nigerian Yoruba population compared to European Caucasians at several CpG sites within VDR CGI 1060. CGI 1060 spans key features; the 5′ splice site for exon 9 (partly encoding the ligand binding domain for VDR), TaqI/CpG 6 embedded in a putative VDRE (35) and the promoter of an untranslated transcript (AK024830) for which the transcription start site is a few bp downstream of a miRNA-125b target site. Thus, differential methylation within this region of the VDR may have profound effects on the expression of VDR and subsequent efficacy of the TLR2/1-VDR signaling pathway. Indeed, we have previously shown ethnicity-dependent methylation of VDR CGI 1060 to distinguish tuberculosis cases from controls (34).
The significant correlation between VDR methylation and plasma 25(OH)D3 supports the proposed relationship between vitamin D and the epigenome (36)(37)(38)(39). The inverse relationship between vitamin D and VDR methylation, especially at the primary promoter-spanning CGI 1062, suggests that in addition to the decrease in ligand, increased promoter methylation may be present in vitamin D-deficient individuals, further dampening the TLR2/1-VDR signal. Thus, vitamin D may interact with the epigenome to influence immune function. Indeed, the higher the methylation at CGI 1062 and 1060a, the less VDR protein is present in response to 1,25(OH)2D3 supplementation and/or TLR2/1 elicitation ( Table 3).
Besides epigenetic differences, SNP frequency distribution for all, except VDR BsmI and TaqI, differed significantly between races with large effect sizes observed for GC rs7041, TLR1 A7202G, N248S and I602S, and VDR Cdx-2 ( Table 1). Similar results were obtained for the 1000 Genomes Project's YRI and CEU populations, except for TaqI and BsmI being significant, but not ApaI (Table S2 in Supplementary Material). These striking differences in frequency distribution of disease-associated or functionally relevant SNPs support the likelihood of interindividual variation in TLR2/1-VDR signaling, response to vitamin D supplementation and immune function. For example, the two GC SNPs rs7041 and rs4588 create the three common Gc/DBP isoforms (Gc1F, Gc1S, and Gc2) showing significant geographical-and race-specific distribution patterns. The Gc1F alleles (rs7041: T allele, rs4588: C allele) are more common among African-Americans and Africans, while the Gc1S alleles (rs7041: G allele, rs4588: C allele) are more common among Europeans (40,41). Gc1F and Gc1S have a stronger affinity for 25(OH)D compared to Gc2 (42), proposed to deliver 25(OH)D more efficiently to target tissues (43). Gc1F/Gc1F homozygotes have the lowest DBP level and Gc1S/Gc1S the highest, yet the bioavailable (unbound or free) 25(OH)D is similar between the isoforms (44). Thus, the efficacy of 25(OH)D3 delivery to target cells may be influenced by genetics and may contribute to the differential response to vitamin D supplementation. The in vitro model used confirmed the induction of VDR by TLR2/1 elicitation (5) and ligand dependance of VDR transactivation of CAMP and CYP24A1 (Figure 3). Observed interactions, regarding race, season and 25(OH)D3 status (Figures 4-6) supported the observed correlation between UVI and 25(OH)D3 status (23), influencing gene expression and hCAP-18 processing; both processes seemingly hampered/delayed in Blacks or 25(OH)D3 deficient individuals. This was also observed for LL-37 secretion in Western blot findings (Figure 7). Significant down regulation of VDR protein in Whites in summer likely reflected negative auto-regulation (24) or UVI-mediated miR125b regulation (20)(21)(22).
The relation between VDR methylation and functional variables ( Table 4) was best observed for VDR protein levels, with CpG sites across the enhancer (CGI 1066), promoter (CGI 1062), exon 3 (CGI 1061) and exon 9 (CGI 1060) showing power to discriminate individuals with above average VDR protein levels from those with below average levels. This supports, in part, the univariate correlation of regional methylation observed most commonly with VDR protein ( Table 3). The significant, large discriminatory power of CGI 1062 CpG 3 in the primary promoter of VDR to distinguish a 5.7-fold above average mean VDR protein level upon elicitation, may relate to the colocation of a binding site for the E2F transcription factor 7, a member of the V$E2FF matrix family (Matrix Library 10.0, Genomatix 2016 (35)), implicated in negative regulation of DNA binding and transcription (45). Comparing the V$E2FF matrix to the sequence around CGI 1062 CpG 3, showed a matrix and core similarity of 1 and 0.875, respectively, with high conservation across the CpG 3 cytosine-guanine dinucleotide that forms part of the matrix core. Notably, a CpG-ruinous SNPs (C/G) in the second position of the dinucleotide, unique to Africans (5% "C, " 1000 Genomes Browser), have been reported. The positive correlation between VDR protein level and methylation at CGI 1062 CpG 3 may support methylation-sensitive suppressor activity, alleviated by DNA methylation. The positive impact of CGI 1060 CpG 6 methylation, possible only when TaqI is "C, " seen for VDR mRNA with supplementation or elicitation, support a TB case control finding from our laboratory showing concomitant decreased methylation of CGI 1060a associating with protection from TB (34) and correlating with increased VDR levels in the current study (   to its location adjacent to an NF-κB1 and SP-1 binding site in the primary promoter of VDR. Overall, it appears that individuals who respond with an above average level of VDR protein upon TLR2/1 elicitation display hypermethylation at CGI 1066 CpG 1 and 3, as well as at CGI 1062 CpG 3, while displaying hypomethylation at CGI 1061 CpG 9 and CGI 1060a CpG 1-4 (Figure 8). While the VDR SNP TaqI and 25(OH)D3 influenced the variation within the study population with elicitation, without 1,25(OH)2D3 supplementation, 25(OH)D3 status was no longer identified as a contributing factor upon 1,25(OH)2D3 supplementation. Moreover, the effect of several SNPs became apparent only in the presence 1,25(OH)2D3 supplementation. This suggests that 25(OH)D3 status may have a larger effect on TLR2/1, VDRmediated signaling than these genetic variables. (Pam3CysSerLys4), a synthetic tripalmitoylated lipopeptide, mimics the acylated amino terminus of bacterial lipoproteins and is recognized by TLR2, cooperating with TLR1 to induce signaling that activates the pro-inflammatory transcription factor NF-κB, inducing CYP27B1 and VDR. Factors found to influence VDR expression are indicated in red. UVI correlated with 25(OH)D3, known to autoregulate VDR. Individuals who responded with an above average level of VDR protein upon TLR2/1 elicitation were hypermethylated at CGI 1066 (CpG 1 and 3) and CGI 1062 (CpG 3), while hypomethylated at CGI 1061 (CpG 9) and CGI 1060 (CpG 1-4). With in vitro 1,25(OH)2D3 supplementation CGI 1062 CpG 3 was slightly less discriminatory upon TLR2/1 elicitation, but remained a significant variable to discriminate above/below average VDR protein level. However, with supplementation the population clustered by race and all genetic variants indicated (TLR1 SNP I602S, TIRAP SNP S180L, and VDR SNPs BsmI, ApaI, and TaqI, see Table 4), compared to clustering by vitamin D status and TaqI alone, without supplementation. Methylation at CpG 6 of CGI 1060 (dictated by TaqI, with which it overlaps) together with UVI and circulating 25(OH)D3, significantly discriminated a 1.2-1.5-fold change in VDR mRNA with supplement or elicitation. Abbreviations: CAMP, cathelicidin antimicrobial peptide; CGI, CpG island; CpG, cytosine-phosphate-guanine dinucleotide; CYP, cytochrome P-450 enzyme; CYP24A1, 25-hydroxyvitamin D3-24-hydroxylase; D2, ergocalciferol; D3, cholecalciferol; DBP, vitamin D binding protein; GC, group component or gene encoding DBP; hCAP-18, human cathelicidin antimicrobial peptide 18; LL-37, cathelicidin antimicrobial peptide fragment; 1,25(OH)2D3 1,25-dihydroxycholecalciferol; 25(OH)D3, 25-hydroxycholecalciferol; NF-κB, nuclear factor kappa B; Pam3CSK4, synthetic tripalmitoylated lipohexapeptide, PTH parathyroid hormone; RXR retinoid X receptor; TIRAP, TIR, toll-interleukin 1 receptor domain-containing adaptor protein; TLR, toll-like receptor; UVB, ultraviolet B rays; UVI, ultraviolet index; VDR, vitamin D receptor gene; VDR, vitamin D receptor protein; VDRE, vitamin D receptor element.
Taken together, results presented here provide support for multifactorial regulation of VDR-mediated TLR2/1 signaling, involving interaction between environment, epigenetics, and genetics. UVI influences 25(OH)D3 status, which regulates VDR expression through VDR methylation, while enhancing the extent and rate of VDR transactivation of CAMP encoding the antimicrobial peptide hCAP-18. The complex interaction between these factors may shed further light on the disparity in infectious diseases across the globe.

eThics sTaTeMenT
In accordance with the Declaration of Helsinki, ethical clearance was obtained from the South African National Blood Service (SANBS) and the Faculty of Science, University of Johannesburg.