Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

7.1 Introduction

A decade ago three groups simultaneously discovered the regulation of human antimicrobial peptide gene expression by vitamin D. In screening the human genome for vitamin D response elements (VDREs), White and colleagues identified potential VDREs in the cathelicidin antimicrobial peptide (CAMP) and β-defensin 4 (DEFB4) genes. They demonstrated induction of these two genes by treatment of isolated human keratinocytes, monocytes and neutrophils, and human cell lines with 1α,25-dihydroxyvitaminD3 [1,25(OH)2D3] concomitant with increased secretion of bactericidal activity from treated cells (Wang et al. 2004). Ståhle and colleagues demonstrated induction of CAMP in human keratinocytes in vitro by 1,25(OH)2D3 and in vivo after topical application of the 1,25(OH)2D3 analog calcipotriol to the skin of human volunteers (Weber et al. 2005). Our group discovered induction of CAMP by 1,25(OH)2D3 in acute myeloid leukemia (AML), immortalized keratinocytes, and colon cancer cell lines as well as normal human bone marrow (BM)-derived macrophages and fresh BM cells from two normal individuals and one acute myelogenous leukemia patient (Gombart et al. 2005). Each group demonstrated the requirement of a VDRE at about 700 base pairs upstream of the transcription start site in the CAMP gene by site-directed mutagenesis or deletion (Wang et al. 2004; Weber et al. 2005; Gombart et al. 2005). In the DEFB4 gene promoter, the VDRE is located about 1200 base pairs upstream of the transcription start site (Wang et al. 2004).

Interestingly, we observed a lack of this regulation in the mouse and discovered from a comparison of various mammalian genomes an evolutionarily conserved VDRE in a short interspersed nuclear element (SINE) in the CAMP promoter of primates that was absent in other mammalian genomes (Gombart et al. 2005). Further, we demonstrated that the VDRE in the CAMP gene originated from the exaptation of an AluSx SINE in the lineage leading to humans, apes, Old World monkeys, and New World monkeys and remained under purifying selection for the last 55–60 million years (Gombart et al. 2009a). Taken together, these findings revealed a novel activity of 1,25(OH)2D3 and the vitamin D receptor (VDR) in regulation of primate innate immunity due to an evolutionarily fixed, Alu-mediated divergence in steroid hormone nuclear receptor gene regulation between humans/primates and other mammals (Gombart et al. 2009a). Regulation of a murine homolog for DEFB4 by vitamin D has not been described.

In a series of subsequent in vitro experiments, Modlin and colleagues demonstrated that vitamin D was required for the induction of CAMP by Toll-like receptor (TLR) signaling, and findings indicated that insufficient serum 25-hydroxyvitamin D [25(OH)D] levels could lead to a lack of CAMP gene expression by macrophages in response to infection (Liu et al. 2006). In the current model, TLR signaling induces CYP27B1 and VDR expression. CYP27B1 activity hydroxylates 25(OH)D resulting in the production of the active 1,25(OH)2D that binds to the VDR and thus induces target genes including CAMP (Liu et al. 2006).

Activation of the vitamin D pathway by TLR signaling was identified in lung and skin epithelial cells, but was not observed in cell lines derived from the colon (Hansdottir et al. 2008; Schauber et al. 2007; Lagishetty et al. 2010). Wounding of the skin enhanced TLR2 function which enabled keratinocytes to respond to pathogen-associated molecular patterns, activate the vitamin D pathway, and increase CAMP levels in the skin to protect against infection (Schauber et al. 2007). Interestingly, the induction of CAMP gene expression by 1,25(OH)2D3 was observed in keratinocytes and monocytes, but not in epithelial cells of the colon (Schauber et al. 2006). In contrast butyrate induced CAMP in colonic cells, but not significantly in keratinocytes and monocytes suggesting that responses to these stimuli were cell type and microenvironment specific (Schauber et al. 2006). Several studies have demonstrated that butyrate and other histone deacetylase inhibitors together with 1,25(OH)2D3 cooperatively or synergistically induce CAMP gene expression in monocytes/macrophages, keratinocytes, and lung and colon cells (Gombart et al. 2007; Schauber et al. 2008; Mily et al. 2013; van der Does et al. 2014; Kulkarni et al. 2014).

7.2 The CAMP Gene

A majority of the interest in vitamin D-mediated regulation of antimicrobial peptide (AMP) gene expression has focused on induction of the CAMP gene because it is more robustly upregulated as compared to DEFB4 (Wang et al. 2004). The CAMP gene encodes an 18-kDa proprotein called hCAP18. It is processed to release a peptide called LL-37 that is expressed by neutrophils and macrophages for killing bacteria and by epithelial cells in barrier defense (Lehrer and Ganz 2002; Gombart 2009; White 2010). In addition, the LL-37 peptide can chemoattract T cells, dendritic cells, neutrophils, and monocytes (Chertov et al. 1996; Yang et al. 2001), which could allow LL-37 to influence cellular traffic at sites of infection or inflammation. Also, LL-37 affects dendritic cell activation and subsequent priming of T cells when added exogenously (Davidson et al. 2004), demonstrating that adaptive immune responses may be regulated by LL-37. The hCAP18 protein is present in γδ T cells, B cells, monocytes, and NK cells of the peripheral blood (Agerberth et al. 2000) following a general hierarchy of protein expression with neutrophils showing the highest levels, monocytes intermediate levels, and lymphocytes the lowest levels (Lowry et al. 2014). In the lymphocyte population, B cells, NK cells, CD4+ T cells, and CD8+ T cells all have similar levels of hCAP18 expression (Lowry et al. 2014). CAMP is secreted by tissues exposed to the environment and in saliva and seminal fluid (Frohm Nilsson et al. 1999; Murakami et al. 2002; Malm et al. 2000). It is critical for host barrier defense as mice lacking it are susceptible to infection (Gombart 2009; Nizet et al. 2001; Chromek et al. 2006).

7.3 The Vitamin D Pathway

The most effective way to acquire vitamin D is through synthesis in the skin or consumption of a purified supplement as diet, which is a poor source (Holick 2011). Ultraviolet B rays provided by natural or artificial sunlight cleave the B-ring of 7-dehydrocholesterol in the skin to produce cholecalciferol or vitamin D3. This is absorbed into the blood and hydroxylated in the liver by the cytochrome p450 enzyme CYP27A1 to calcidiol or 25(OH)D3. This form (together with 25(OH)D2; see below) is measured in the serum as an indicator of vitamin D status (Holick 2011). 25(OH)D3 is converted to its bioactive form, calcitriol or 1,25(OH)2D3, by the mitochondrial 1α-hydroxylase enzyme CYP27B1 in the kidney. A fungal-derived form of vitamin D is created by the UVB exposure of ergosterol to generate ergocalciferol. This form of vitamin D is hydroxylated in the liver to 25(OH)D2 and in the kidney to 1,25(OH)2D2 (Holick 2011). Both 1,25(OH)2D3 and 1,25(OH)2D2 bind to the VDR a steroid hormone nuclear receptor/transcription factor that binds to VDREs and recruits cofactors to activate and/or repress the expression of target genes (Mangelsdorf et al. 1995; Christakos et al. 1996).

Synthesis of 1,25(OH)2D in the kidney is essential for efficient uptake of dietary calcium in the gut and to maintain bone health. A drop in circulating Ca2+ levels stimulates the production of parathyroid hormone (PTH) which induces CYP27B1 expression by primary renal tubules. Increased 1,25(OH)2D production activates Ca2+ transporter expression via the VDR in the small intestine, thereby increasing circulating Ca2+ and suppressing PTH production (Holick 2011). In a negative feedback loop, activated VDR binds to the CYP27B1 promoter and represses its expression. Also, VDR induces fibroblast growth factor-23 in osteocytes which inhibits secretion of PTH and represses CYP27B1 expression and induces expression of CYP24A1, a mitochondrial enzyme that catabolizes both 1,25(OH)2D and 25(OH)D to limit 1,25(OH)2D levels and prevent hypercalcemia (Paz et al. 2007; Saito et al. 2003; Zierold et al. 1995).

Abundant epidemiological, clinical, and basic research has implicated vitamin D in preventing cancer, autoimmune disorders, cardiovascular disease, and infections (Grober et al. 2013). The synthesis of 1,25(OH)2D in nonrenal tissues and cells likely mediates these additional health benefits (Hewison et al. 2004). The extrarenal synthesis of 1,25(OH)2D occurs in lung, colon, parathyroid glands, bone, skin, and macrophages and is considered important for optimal immune response at sites of infection (Hewison et al. 2004).

7.4 Vitamin D and Immunity

The role of vitamin D in regulating the adaptive immune response is highlighted by numerous lines of evidence. The VDR is expressed in T and B cells, monocytes, macrophages, dendritic cells (DCs), and neutrophils (Provvedini et al. 1983; Bhalla et al. 1983; Deluca and Cantorna 2001; Adorini et al. 2004; Kreutz et al. 1993; Brennan et al. 1987; Takahashi et al. 2002; Mangelsdorf et al. 1984). 1,25(OH)2D3 inhibits Th17 development, increases the frequency of Th2 and regulatory T cells, decreases Th1 development, and modulates T-cell proliferation and cytokine expression (Lemire et al. 1995; Boonstra et al. 2001; Penna and Adorini 2000; Daniel et al. 2008). 1,25(OH)2D3 also promotes tolerance in dendritic cells and T cells and inhibits B-cell differentiation into plasma cells (Adorini et al. 2004; Mathieu and Adorini 2002; Chen et al. 2007). Overall vitamin D mediates an antiinflammatory response and promotes tolerance in the adaptive response.

In addition to responding to circulating 1,25(OH)2D3, dendritic cells, macrophages, and T cells can actively produce it (Hewison 2012). Initially, extrarenal production of 1,25(OH)2D3 by macrophages from some granulomatous disease patients was reported (Barbour et al. 1981; Adams et al. 1983). In vitro studies with normal macrophages indicated that CYP27B1 activity was induced as part of the immune response (Koeffler et al. 1985; Reichel et al. 1986). DCs confer specific homing properties upon T cells during the adaptive immune response, and DCs derived from the skin are able to synthesize 1,25(OH)2D3 from vitamin D3. This, in turn, induces expression of CC chemokine receptor 10 in T cells and suppresses expression of gut-homing receptors which enable T cells to migrate toward the chemokine CCL27 that is secreted by epidermal keratinocytes. These findings demonstrate that DCs produce locally high levels of 1,25(OH)2D3 to regulate T-cell epidermal tropism (Sigmundsdottir et al. 2007).

The production of potentially high local levels of 1,25(OH)2D3 is most likely important for intracrine and paracrine influences on the interactions between vitamin D, the immune system, and pathogens (Hewison 2012). During the mid-1980s, it was demonstrated that both 25(OH)D3 and 1,25(OH)2D3 increased the capacity of human monocytes to control Mycobacterium tuberculosis (Mtb) growth (Davies 1985; Rook et al. 1986). Nearly 20 years later, as described above, we and others discovered that vitamin D increased expression of the CAMP gene (Wang et al. 2004; Weber et al. 2005; Gombart et al. 2005). In addition, the human β-defensin 2 or DEFB4 gene was identified as a vitamin D inducible antimicrobial peptide gene, but its induction by vitamin D or TLR activation is much less robust than CAMP (Wang et al. 2004; Liu et al. 2006). These observations offered a mechanism by which vitamin D could directly enhance killing of Mtb.

7.5 Cooperative Induction of Antimicrobial Peptide Gene Expression by Multiple Signaling Pathways

Robust induction of DEFB4 by vitamin D requires activation of additional signaling pathways. Co-treatment of monocytes with IL-1 and 1,25(OH)2D3 induced binding of both NF-κB and VDR to the DEFB4 promoter and was much more effective in inducing gene expression (Wang et al. 2004; Liu et al. 2009). Also, in the presence of muramyl dipeptide (MDP), the intracellular pattern recognition receptor nucleotide-binding oligomerization domain protein 2 (NOD2) activates NF-κB, and there is a modest induction of the DEFB4 gene (Voss et al. 2006; Wang et al. 2010); however, treatment with 1,25(OH)2D3 prior to addition of MDP strongly induces the DEFB4 gene (Wang et al. 2010). It was shown that 1,25(OH)2D3 strongly induced expression of NOD2 in primary human monocytic and epithelial cells which amplified the MDP signal (Wang et al. 2010). In total, studies have shown that the vitamin D pathway alone is insufficient to induce robust expression of DEFB4, and activation of additional signaling pathways is required (Liu et al. 2009; Wang et al. 2010).

Several published studies have demonstrated that cytokine expression also modulates vitamin D-mediated CAMP and DEFB4 expression. In human macrophages, TLR2/TLR1 signaling induces IL-15 expression which increases IL-32 which is essential for induction of CYP27B1 and the VDR (Krutzik et al. 2008; Montoya et al. 2014). The subsequent increased conversion of 25(OH)D3 to 1,25(OH)2D3 by CYP27B1 activates the VDR and induces CAMP expression and antimicrobial activity against Mtb (Krutzik et al. 2008; Montoya et al. 2014). In human monocytes, Th1 cytokine IFN-γ upregulates TLR2/TLR1 induction of CYP27B1 and the bioconversion of 25(OH)D3 to 1,25(OH)2D3 which enhances induction of CAMP (Edfeldt et al. 2010). Further, vitamin D is required for IFN-γ-mediated activity of human macrophages (Fabri et al. 2011). On the other hand, the Th2 cytokine IL-4 induces CYP24A1 expression which leads to the catabolism of 25(OH)D3 and downregulation of CAMP expression (Edfeldt et al. 2010). In contrast, the Th2 cytokine IL-13 enhances CAMP expression by 25(OH)D3 due to increased CYP27B1 expression and synthesis of 1,25(OH)2D3 (Schrumpf et al. 2012). No effect is observed with IL-17 in monocytes, but in the presence of 1,25(OH)2D3, IL-17 enhances CAMP expression in human keratinocytes via activation of the Act1 and MEK/ERK pathway (Peric et al. 2008). In addition to IL-4, other cytokines can inhibit antimicrobial peptide (AMP) expression. In macrophages, IFN-γ-induced vitamin D-dependent AMP expression was suppressed by IFN-β and IL-10 (Teles et al. 2013). Similarly, in placental cells, IL-10 inhibited β-defensin and CAMP expression, while 1,25(OH)2D3 treatment could override the suppression (Olmos-Ortiz et al. 2015). Further, TNF-α and 1,25(OH)2D3 enhanced β-defensin, and TNF-α reduced both basal and 1,25(OH)2D3-induced CAMP expression (Olmos-Ortiz et al. 2015). Taken together, the differential effect of T-cell cytokines on CAMP and DEFB4 expression represents mechanisms by which adaptive immune responses can regulate innate immune antimicrobial peptide defenses against pathogens. It remains to be determined how these various signaling pathways work together in vivo during infection.

7.6 Fighting Infection Through Increased AMP Expression

Historically, sources of vitamin D were used as treatments for tuberculosis (Martineau et al. 2007). In the 1940s, physicians effectively treated cutaneous Mtb infection with high-dose vitamin D2, but this fell out of favor with the advent of effective antibiotics (Martineau et al. 2007; Dowling 1946; Gaumond 1948). In the 1980s, epidemiological studies pointed to a correlation between higher rates of tuberculosis and vitamin D deficiency (Davies 1985). Further, 1,25(OH)2D3 was shown to enhance intracellular killing by human monocytic cells (Rook 1986). Knockdown of either DEFB4 or CAMP expression in monocytes/macrophages decreased killing of Mtb indicating their importance for fighting infection (Liu et al. 2009). The induction of CAMP by vitamin D is required for promoting autophagy to kill Mtb (Hoyer-Hansen et al. 2005; Wang et al. 2008; Yuk et al. 2009). Additional findings support a paracrine macrophage-lung epithelial cell signaling pathway that is driven by IL-1β and 1,25(OH)2D3 (Verway et al. 2013). In this model, 1,25(OH)2D3 increased IL-1β secretion in Mtb-infected macrophages. The secreted IL-1β induced DEFB4 expression from airway epithelial cells which enhanced control of Mtb growth in co-cultured macrophages in vitro (Verway et al. 2013). Taken together these studies support an important role for vitamin D in modulating the immune response to Mtb infection.

These findings have renewed interest in potentially using vitamin D to treat tuberculosis. A review of clinical trials and case series indicates that numerous studies are methodologically flawed (Martineau et al. 2007) or an insufficient vitamin D dose was used (Wejse et al. 2009). Two small randomized studies indicate some benefit from vitamin supplementation of TB patients (Nursyam et al. 2006; Morcos et al. 1998). More recently, pulmonary tuberculosis patients receiving standard therapy and a 100,000 IU dose of vitamin D3 every 2 weeks showed accelerated sputum conversion if they possessed the tt genotype of the vitamin D receptor as compared with placebo (Martineau et al. 2011). Further, vitamin D supplementation accelerated resolution of inflammation during tuberculosis treatment (Coussens et al. 2012). A recent randomized, double-blinded, multicenter, placebo-controlled clinical study involving 258 patients showed that 600,000 IU vitamin D3 once per month for 2 months led to a significant increase in average weight gain and lower residual disease by chest x-ray as compared to placebo (Salahuddin et al. 2013).

Deficiencies in vitamin D are associated with poor outcomes in HIV-infected individual, bacterial vaginosis in the first trimester of pregnancy, increased influenza A infections, and increased respiratory tract infections (Bodnar et al. 2009; Villamor 2006; Aloia and Li-Ng 2007; Sabetta et al. 2010). Supplementation with vitamin D lowered the incidence of seasonal flu in school children, the elderly, and African-American women and lowered the severity of respiratory tract infections (Urashima et al. 2010; Avenell et al. 2007; Aloia et al. 2005; Kenny et al. 2012). In contrast, vitamin D supplementation did not reduce the incidence and duration of severity of upper respiratory tract infection (Li-Ng et al. 2009). In a meta-analysis of 11 placebo-controlled studies involving 5660 patients, vitamin D showed a protective effect against respiratory tract infections with once-daily dosing being better than bolus doses (Bergman et al. 2013). The authors noted that there was significant heterogeneity and evidence of publication bias in the field and warned that results should be carefully interpreted (Bergman et al. 2013). It should be noted that in all of these studies including those with tuberculosis, the role of CAMP induction in these outcomes is unknown. Future studies must optimize dose, dosing frequency, and target populations that are deficient in vitamin D to detect modest effects.

7.7 The Impact of Vitamin D on CAMP Levels

To date, in vivo studies demonstrating that vitamin D status or supplementation affects the levels of CAMP/hCAP18 are inconclusive. High levels of hCAP18 are found in the blood; therefore, we hypothesized that vitamin D levels may correlate with hCAP18 levels (Sorensen et al. 1997; Gombart et al. 2009b). In an early study on dialysis patients, we found only a modest positive correlation between hCAP18 and 1,25(OH)2D, but not 25(OH)D levels, but high hCAP18 levels were associated with a significant decrease in 1-year mortality (Gombart et al. 2009b). For sepsis patients, a positive association between 25(OH)D and hCAP18 levels was observed in all patients (Jeng et al. 2009). In healthy individuals, a positive association between hCAP18 and 25(OH)D levels was observed at levels of 25(OH)D below 32 ng/ml, but not above (Bhan et al. 2011; Dixon et al. 2012). In the elderly and in atopic dermatitis patients and normal controls, a positive correlation was observed without applying a cutoff (Alvarez-Rodriguez et al. 2012; Kanda et al. 2012). On the other hand, in cord blood samples, patients with active TB and patients with pneumonia, a correlation between serum 25(OH)D and hCAP18 was not observed (Yamshchikov et al. 2010; Mandic Havelka et al. 2010; Leow et al. 2011). Supplementation of atopic dermatitis patients with 4000 IU/day vitamin D for 3 weeks increased CAMP in skin lesions and unaffected skin, but a second study with more patients was negative (Hata et al. 2008, 2014). Several studies using high-dose supplementation (50,000–60,000 IU/week) did not observe increased hCAP18 in the blood (Adams et al. 2009; Alvarez et al. 2013; Das et al. 2014). In a randomized controlled trial in patients with severe sepsis, 1,25(OH)2D3 did not increase plasma hCAP18 levels (Leaf et al. 2014). In a study of 15 hereditary vitamin D-resistant rickets patients (possess a nonfunctional VDR) and 17 normal controls, it was shown that VDR is required for induction of CAMP by vitamin D in adherent mononuclear cells cultured for 24 h, but basal expression of CAMP in various cell types, fluids, or tissue samples was not determined (Tiosano et al. 2013). Additional studies are required to determine the effect of vitamin D status or treatment on in vivo CAMP expression, particularly on the in vivo induction of CAMP in immune cells like macrophages during infection.

7.8 Vitamin D-Mediated Regulation of AMPs: An Animal Model

The difficulty in determining the role CAMP in mediating the effects of vitamin D on the immune response is the lack of a good animal model that replicates the pathway as it is found in humans. As described earlier, vitamin D does not regulate CAMP expression in mice or other mammals (Gombart et al. 2009a). In addition, work from our own group and others revealed a striking difference in the use of vitamin D by human versus murine macrophages. As described above, activation of human macrophages by TLR ligands induces expression of CYP27B1 and the bioconversion of 25(OH)D3 to 1,25(OH)2D3. This, in turn, leads to the induction of various VDR target genes including CAMP. In contrast, TLR activation of murine macrophages does not induce CYP27B1 expression; thus, bioactive 1,25(OH)2D3 is not synthesized by murine macrophages (Kapetanovic et al. 2012; Ooi et al. 2014), and vitamin D target genes are not induced (our unpublished findings). This major difference in the utilization of vitamin D by macrophages highlights the importance of caution when using the mouse model to elucidate the role of vitamin D on immune function in humans. Macrophages are very likely important for producing locally high levels of 1,25(OH)2D3 at sites of infection in humans, but not in mice.

7.9 Conclusion

An abundance of in vitro evidence exists to demonstrate the regulation of AMPs, particularly CAMP, by vitamin D. Also, historical, epidemiological, and clinical data is consistent with the vitamin D-CAMP pathway providing protection against infection. Nevertheless, there is a paucity of in vivo data that demonstrates that induction of CAMP mediates important antibacterial or viral activities that are attributed to vitamin D. Due to the current lack of a viable animal model, this evidence will need to come from carefully designed human and/or primate studies. Outstanding questions remain on how vitamin D status or supplementation affects CAMP and DEFB4 expression and can active forms of vitamin D increase levels of AMPs to improve immunity.