Exposure to UV Wavelengths in Sunlight Suppresses Immunity. To What Extent is UV-induced Vitamin D₃ the Mediator Responsible?

Abstract

Reduced immunity following exposure of skin to UV radiation (UVR) may explain the positive latitude gradient measured for a number of autoimmune diseases (greater incidence of disease with residence at higher latitudes), including multiple sclerosis, allergic asthma and diabetes. Humans obtain >80% of their vitamin D₃ by exposure of skin to UVR in sunlight. In experimental models, both vitamin D₃-dependent and vitamin D₃-independent pathways have been implicated in the mechanisms of UVR-induced systemic suppression of immunity. However, where does the balance of control lie? How important is vitamin D₃ other than providing a biomarker of sun exposure? Are other molecules/pathways activated by UVR more important? Murine and human studies suggest many molecules may play a role and their participation may vary with different diseases and the time of UVR exposure or vitamin D₃ sufficiency/deficiency. Although low vitamin D₃ levels have been associated with increased prevalence and progression of human autoimmune diseases, the benefits of supplementation with vitamin D₃ have not been definitive. Vitamin D₃ levels are a measure of past sun exposure but vitamin D₃-dependent and vitamin D₃-independent immunosuppressive effects of UVR may play a role in control of autoimmune diseases.

Introduction

It is now more than 30 years since the immunoregulatory properties of UVR (UVB 290–315 nm, UVA 315–400 nm) were first reported. It was recognised that most UVR-induced tumours are immunogenic but they are not rejected by a competent immune system because UVR also suppresses the immune processes that would otherwise have destroyed them.¹ Most importantly, the dampened immune responses can be detected not only at the skin site of UV-irradiation (i.e. UV-induced tumours) but also at non-irradiated skin sites and in remote tissues such as in the airways.^2,3 UVR suppresses immunity by similar mechanisms in humans and in mice with local and systemic immunosuppression reported in response to both erythemal and sub-erythemal amounts of UVR.^4–8 UV-induced immunosuppression may be important not only in determining the growth, or demise, of skin tumours but also may dampen responses to microbial agents, necrotic cells, and vaccine antigens.^9–11

Autoimmune diseases have complex aetiologies with both genetic and environmental contributions. Positive latitude gradients have been reported for most of the main autoimmune diseases with a higher incidence of disease reported for patients living furthest from the equator. In Australia, the highest incidence of multiple sclerosis and asthma has been reported in Tasmania, with the rate of multiple sclerosis in Tasmania being seven times higher than in Northern Queensland.^12,13 The incidence of type 1 diabetes is also influenced by the degree of latitude and UVB irradiance, both overseas and in Australia.^14,15 Not all studies have found the association of latitude with multiple sclerosis and asthma prevalence;¹⁶ place of birth rather than latitude of one’s current dwelling may contribute to disease incidence.¹⁷ In the most comprehensive review of multiple sclerosis prevalence to date, an exception to a positive association between multiple sclerosis prevalence and latitude globally was found in the Italian region and in northern Scandinavia.¹⁸ In that study, both genetic (HLA-DRB1) and behavioural-cultural variations contributed to disease incidence and pathogenesis. The seasonal effect of expression of some autoimmune diseases also supports UVR involvement with greater disease symptoms during winter.¹⁹ We may be more susceptible as neonates and children to the effects of UVR. The link between month of birth and latitude was recently systematically assessed in >100,000 patients and demonstrated a very significant effect of month of birth on subsequent multiple sclerosis risk.¹⁷ In another study, for 115,172 individuals born in the UK, the distributions of births for people with immune-mediated diseases peaked in April (i.e. spring, odds ratio 1.045) and troughed in October (i.e. autumn, odds ratio 0.945).²⁰ This seasonality of birth was measured in all the immune-mediated diseases investigated (multiple sclerosis, type 1 diabetes, rheumatoid arthritis, ulcerative colitis, systemic lupus erythematosis) except Crohn’s disease.²⁰ These correlations of the month/season of birth and risk of immune-mediated diseases highlight the important contribution of everyday environmental UV exposure.

The link between reduced UVR exposure and increased autoimmune disease has been exemplified dramatically by the enhanced incidence of multiple sclerosis in Iran after the re-introduction of stricter Islamic traditions.²¹ With increased covering up of the skin of females in the last few decades, there has been a significant increase in the incidence of this autoimmune disease.²²

Control of Experimental Models of Immune Disease by UVR

UVR exposure is an important contributor to the environment-driven determinants of disease incidence. To support the epidemiological data, there are also studies in mice of UVR control of disease models. UVR can suppress the symptoms of the murine model of multiple sclerosis, experimental autoimmune encephalomyelitis.²³ We have shown that UV irradiation of the shaved skin of mice can reduce murine models of asthma, namely experimental allergic airways disease.^2,3 Further, UVR can be given either before allergen sensitisation or whilst the disease is developing in the pre-sensitised animal.²⁴

Several photoreceptors for UV photons, particularly the higher energy, shorter wavelength UVB photons, have been reported and the biochemical events initiated by their absorption studied. It is hypothesised that many of the pathways stimulated may be complementary, and at times redundant, depending on the health and circumstances of the irradiated skin. One important photoreceptor in keratinocytes is 7-dehydrocholesterol which in turn produces pre-vitamin D₃ (see next section). Other photoreceptors on skin cells include DNA and lipids,^4–11,25 as well as the histidine derivative, trans-urocanic acid in the stratum corneum. The membranes of skin cells may be oxidised as part of a ‘stress’ response to UV irradiation. As recently reviewed by both us and others, exposure to UVR can also affect the metabolic status of skin cells, and hence their capacity to repair UV-induced DNA damage and to secrete immunoregulatory molecules.^7,25

Multiple inflammatory and immunoregulatory molecules produced by the immune, neural and structural cells of UV-irradiated skin have been implicated in the migration of Langerhans cells, dermal dendritic cells and mast cells into the draining lymph nodes where presentation of antigens is such that reduced immune responses develop, fewer memory cells are created and in many models there is an induction of T regulatory cells.^{7–11, 25} The mechanisms behind immunosuppression at the ‘local’ UV-irradiated site are better understood. There is greater uncertainty about the mechanisms of UV-induced systemic immunosuppression; the extent of involvement of immune mediators and regulatory cells from UV-irradiated skin and the draining nodes in controlling presentation of antigens delivered to distant sites is not clear.

In summary, epidemiological studies in humans and experimental studies in mice suggest UVR exposure can suppress the initiation and development of immune-driven models of disease. Several signalling pathways, both locally in the skin and more systemically, may be stimulated by UVR exposure. Vitamin D synthesis is induced by UVR exposure; this review debates the importance of vitamin D as an immune mediator of the immunoregulatory effects of UVR.

Vitamin D

There have been many reviews detailing the biochemistry of vitamin D₃ production initially in the skin (7-dehydrocholesterol to vitamin D₃ via pre-vitamin D₃) and subsequently in the liver (to 25-hydroxy vitamin D₃ [25(OH)D₃], the storage form) and kidney (to the active molecule, 1,25-dihydroxy vitamin D3 [1,25(OH₂)D₃])(Figure 1).^26–28 1,25(OH)₂D₃ has a half-life of a few hours and circulates in the pmol/L range. The vitamin D receptor (VDR) that binds 1,25(OH)₂D₃ is a nuclear steroid receptor expressed by almost all cells of the body, and may potentially regulate 5–10% of the genome.^26–28 Circulating 25(OH)D₃, with a half-life of 30–40 days, is in the nmol/L range and the substrate for the enzyme CYP27B1. Blood levels of 1,25(OH₂)D₃ and parathyroid hormone are controlled by CYP27B1 expressed predominantly in the kidney; however the enzyme is also expressed in multiple cell types, including immune cells, to control levels of 1,25(OH₂) D₃ in local environments (Figure 1). The levels of CYP27B1, and the tissue need for 1,25(OH₂)D₃, may determine the level of circulating 25(OH)D₃ required for homeostasis and optimal health in that tissue, and diseases associated with that tissue.

Figure 1.

The production of 1,25(OH)₂D₃ from 7-dehydrocholesterol in skin. VDBP = vitamin D binding protein.

In 2011, the Institute of Medicine (USA) determined that circulating levels of 25(OH)D₃ of 50 nmol/L were sufficient for optimal bone health.²⁹ With respect to 25(OH)D₃ levels for optimal immune health, they concluded that there was insufficient evidence to recommend any active vitamin D supplementation. Not all have agreed with these recommendations and vitamin D insufficiency has been considered to occur between 50 and 75 nmol/L of 25(OH)D₃. The Endocrine Societies recommend circulating levels of 25(OH)D₃ exceed 75 nmol/L for optimal bone, immune and cardiovascular health and increased defence mechanisms against cancers and infectious agents.³⁰ As reviewed in this journal by Wootton³¹ and Grebe and Singh,³² measurements of 25(OH)D₃ are methodologically challenging and can be inconsistent from laboratory to laboratory. Thus, there may be some error in absolute levels of 25(OH)D₃ quoted. For general good health, most practitioners recommend serum levels of 25(OH)D₃ level of approximately 75 nmol/L.³³

The consequences of 1,25(OH)₂D₃ on the activity of immune cells are many but generally they have been defined as the stimulation of innate immunity and suppression of adaptive immunity. Dendritic cells and macrophages can synthesise 1,25(OH)₂D₃.^27,34 Activation of these cells through pathogen-recognition molecules (e.g. Toll-like receptors (TLR), NOD receptors) stimulates the upregulation of CYP27B1, and further downstream intracrine effects on immune cells including antimicrobial peptide synthesis³⁵ and increased cell activity.^36,37 A rate-limiting step in this local synthesis may occur during vitamin D₃ deficiency when there are low circulating levels of 25(OH)D₃. Indeed, immune cells may require a high level of 1,25(OH)₂D₃ for optimal functionality.³⁴ Less well recognised is how polymorphisms in the vitamin D binding protein can influence its binding affinity with 25(OH)D₃, the uptake of 25(OH)D₃ by immune cells and the subsequent 1,25(OH)₂D₃ production.³⁴ Vitamin D binding protein has its own immunoregulatory effects including activation of macrophages and stimulation of neutrophil and monocyte chemotaxis. Many immune cells respond to 1,25(OH)₂D₃ through the nuclear association of the VDR with the retinoid X receptor and numerous transcription factors. The expression of the VDR is up-regulated in T cells through activation of the T cell receptor, while for macrophages and other cells, cellular differentiation reduces VDR expression.²⁶ Further details of the effects of 1,25(OH)₂D₃ on innate and adaptive immune responses are described below.

Vitamin D and Innate Immunity

Innate immunity is the first line of defence by the body for combating infection, with the capacity to rapidly respond in a non-specific manner. Antimicrobial peptides (AMPs) like the cathelicidin peptide, LL-37, and the β-defensins are produced by monocytes, macrophages, neutrophils and epithelial cells in response to stimulation with 1,25(OH)₂D₃.³⁷ AMPs enhance microbe killing through disruption of microbial membranes and can also activate other anti-microbial pathways within infected cells.^37,38 AMPs can induce chemotaxis and have other effects on immunity.³⁹ In landmark studies with Mycobacterium tuberculosis, the infecting microbes initiated AMP induction through TLR-signaling, resulting in enhanced CYP27B1 expression, and synthesis of 1,25(OH)₂D₃.³⁵ Interestingly, as part of its immune evasion strategy, Mycobacterium leprae regulates the expression of microRNAs to modulate the expression of key immune genes, reducing CYP27B1 and IL-1β, and upregulating IL-10, to inhibit the expression of the cathelicidin and β-defensin-4 genes.⁴⁰ Molecules produced by T cells can have opposing effects on this pathway. IL-15 can act as a potential intermediary in promoting localised activity of CYP27B1,⁴¹ while IL-4 increases the expression of the 24-hydroxylase enzyme, which is responsible for inactivating 1,25(OH)₂D₃ and, in doing so, attenuates TLR-mediated induction of LL-37 in monocytes.⁴² Thus, innate and adaptive immune responses can interact to modulate vitamin D metabolism.

Vitamin D and Adaptive Immunity

Dendritic cells located in tissues like the skin and lung play a central role capturing, processing and presenting antigen to T and B cells in immune tissues like lymph nodes, allowing the generation of immune memory. The activation of dendritic cells is inhibited by 1,25(OH)₂D₃ through reduced costimulatory molecule expression, NFκB-induced pro-inflammatory cytokine production, and modified stimulation of T cells.³⁴ In particular, 1,25(OH)₂D₃ modulates T regulatory cell numbers and their suppressive abilities through dendritic cells.⁴³ Independently of dendritic cells, 1,25(OH)₂D₃ may have direct effects on T cells, promoting the development of T regulatory cells but not T helper type-1 and -17 cells. Reports that 1,25(OH)₂D₃ can stimulate the development of T helper type-2 cells are inconsistent,^26,43 where the effects of 1,25(OH)₂D₃ may be dependent on the activity and environment of the target cells at the time of exposure.³⁴

The immune cells and pathways that are regulated by 1,25(OH)₂D₃ have also recently been reviewed with more detail in a number of excellent publications.^27,34,36,44

Associations of Vitamin D with Human Immune Diseases

In response to the immunomodulatory properties for vitamin D₃ outlined above, many have proposed that a deficiency of vitamin D₃, or altered expression of the vitamin D binding protein or the VDR, may contribute to the pathogenesis of autoimmune diseases. For patients with multiple sclerosis, low serum 25(OH)D₃ levels have been associated with increased risk of disease, progression of disease, and increased relapse rate.^28,30 Several studies have quantified the benefits of higher serum 25(OH)D₃ levels and relapse rate for patients with relapsing-remitting multiple sclerosis. For patients in Tasmania, it was calculated that in a dose-dependent linear fashion, with each 10 nmol/L increase in serum 25(OH)D₃, there was a 12% reduction in risk of relapse.⁴⁵ In a study from the Netherlands, the association between serum 25(OH)D₃ concentrations and exacerbation rate was log linear without a threshold. With each doubling of the serum 25(OH)D₃ concentration, the exacerbation rate decreased by 27%.⁴⁶ In some studies, the risk of disease conferred by HLA-DRB1*1501, a strong risk factor in females, was moderately modulated by VDR variants.⁴⁷ However, polymorphisms in the VDR gene were not associated with multiple sclerosis risk in a recent meta-analysis of case-control studies.⁴⁸

For paediatric and adult patients with allergic asthma, low serum 25(OH)D₃ levels have correlated with increased allergen sensitivity (high IgE levels), bronchial hyperresponsiveness, poor lung function and reduced responses to steroids.^5,49 In a community-based cohort study in Perth, 25(OH)D₃ levels at ages 6 and 14 y were negatively associated with concurrent allergic phenotypes, particularly in boys.⁵⁰ Vitamin D levels at age 6 y were also significant predictors of subsequent atopy/asthma-associated phenotypes at age 14 y.

Lower serum 25(OH)D₃ levels have been associated with increased risk of diabetes in both children and adults.⁴⁴ In types 1 and 2 diabetes, vitamin D₃ may have effects not only on the immune cells (reducing autoimmune responses) but also on pancreatic beta cells rendering them more resistant to the cellular stress of the disease.⁵¹ In vitamin D₃ deficiency, insulin secretion by beta cells may be reduced, whilst vitamin sufficiency may protect them against cytokine-mediated insulin release, nitric oxide production and beta cell death.⁵¹ For both asthma and type 1 diabetes, VDR polymorphisms have been linked with risk of disease.

Vitamin D₃ Supplementation of Patients with Chronic Immune Diseases

In response to the associations of low vitamin D₃ levels in patients with increased incidence of autoimmune and other diseases, vitamin D₃ supplementation has been recommended by many medical specialists. In discussion of the outcome of many trials of vitamin D₃ supplementation, the dose of vitamin D₃ used, and whether supplemented with calcium must be considered. In November 2010, the Institute of Medicine (USA) recommended a daily vitamin D₃ allowance of 600 IU/day until 71 years of age, and 800 IU/day when older.²⁹ However, members of the Endocrine Societies have recommended vitamin D supplements of up to 4000 IU/day.³⁰

For treatment of patients with multiple sclerosis, there are many ongoing trials with vitamin D₃. The low incidence rate of the disease necessitates multi-centre participation and many years to complete. In one trial, some benefit was reported when vitamin D₃ was given as an add-on therapy with interferon β. In a report from Tasmania, an interaction between interferon β and 25(OH)D₃ was noted such that for patients on interferon β therapy, their serum 25(OH)D₃ levels increased approximately three times higher per hour of sun exposure than was measured for patients not on interferon β therapy.⁵² This interaction was only measured for patients with higher serum 25(OH)D₃ levels. Also, for reasons that are not understood, interferon β therapy was protective against relapse rate in patients with higher 25(OH)D₃ levels, whilst administration of interferon β increased relapse rate in vitamin D₃ insufficient patients. ⁵² In another study in Norway, and in direct contrast, no association was found between 25(OH)D₃ levels and disease activity after initiation of interferon therapy.⁵³ The benefits of vitamin D therapy for multiple sclerosis have been modest in many studies.⁵⁴ The lack of consistency of outcomes in studies of vitamin D₃ and multiple sclerosis is confusing and suggests that larger, more robust clinical trials are required.

For asthma, some encouraging but small supplementation studies have been reported for both newly-diagnosed patients⁵⁵ and for patients already on steroids and other medications.^56,57 Several investigators have supplemented pregnant women with vitamin D₃; however, reduced allergic asthma in the children has not been consistently measured.^58,59 It has been hypothesised that low vitamin D₃ levels in pregnant women may alter lung development as well as the development of the immune system.⁶⁰ For control of type 1 diabetes, variable responses have been measured. A meta-analysis of case-control and cohort studies suggested that vitamin D₃ supplementation in childhood offers some protection against the development of the disease but was less effective as a treatment for developed disease.⁶¹ Overall, there has not been strong evidence that maternal vitamin D₃ intake can reduce the risk of type 1 diabetes in offspring.⁵¹ With respect to the incidence of type 2 obesity-related diabetes, the outcomes of vitamin D₃ supplementation (generally with calcium) have again been variable with both reduced and unaltered incidence reported.⁵¹ In post-hoc analyses from eight trials among participants with normal glucose tolerance at baseline and in three small underpowered trials of patients with established type 2 diabetes, there was no effect of vitamin D₃ supplementation on glycaemic outcomes.⁶²

To summarise, many associations have been reported of low vitamin D₃ with increased incidence/severity of autoimmune diseases. However, there has been a lack of consistent improvement of disease incidence/severity with vitamin D₃ supplementation. Admittedly, there may be many shortcomings in the trials reported. They may not have been suitably powered, they may not have involved supplementation with adequate levels of vitamin D₃, or for sufficient lengths of time. There is always the question of the ‘optimal’ vitamin D₃ level necessary for the requirements of particular tissues and diseases associated with those tissues. In consideration of the ‘optimal’ vitamin D₃ level, it has been debated whether patients can have too much vitamin D and negative outcomes of supplementation occur. For example, a ‘J’ shaped curve has been described for associations of serum IgE levels and vitamin D₃ levels.⁶³ In that study, higher IgE levels were measured in the serum of patients with both the highest and lowest levels of 25(OH)D₃. ‘Optimal’ levels of vitamin D₃ are also an issue for studies investigating the benefits of daily, monthly or annual doses of supplemented vitamin D. Generally, sufficient vitamin D₃ levels in the elderly have been associated with fewer falls and fractures. However, in an Australian study of 2256 older women given a single annual dose of 500,000 IU of vitamin D₃ for 3–5 years, the patients given the high dose of vitamin D₃ had more falls and fractures (15.2% women) than those given a placebo treatment (12.1%).⁶⁴ In a subgroup of 75 women in the vitamin D₃-treatment group, serum 25(OH)D₃ levels were measured after dosing. After one month, 24% of these women recorded serum 25(OH)D₃ levels at 150 nmol/L or higher. By three months, the after-dose median levels were approximately 90 nmol/L, and remained higher than the placebo group 12 months after initial dosing.⁶⁴ This study suggests that 500,000 IU of vitamin D₃ may achieve toxicity in a small number of elderly women.

In further debate of the value of supplementation, patients may have irreversible disease by the time they are supplemented with vitamin D₃ in adulthood. Co-factors may be required to ensure that the supplemented vitamin D is sufficiently effective. Further, reverse causality must be considered in that the immune diseases studied may be responsible for vitamin D deficiency in patients. Further, unwell patients may seek less sun exposure. Many large vitamin D₃ trials are ongoing and we eagerly await their outcome.

Vitamin D₃-Dependent and Vitamin D₃-Independent Pathways in Disease

Human Studies

It is frequently impossible to determine whether vitamin D₃ is responsible for the immune consequences ascribed to UV exposure. In the large UK study that found a season of birth effect in the immune-mediated diseases of multiple sclerosis, rheumatoid arthritis, ulcerative colitis and systemic lupus erythematosis, the risk of all diseases correlated with predicted second trimester UVB exposure and third trimester vitamin D status.²⁰ Both vitamin D₃-dependent and vitamin D₃-independent pathways may have been contributory to the outcome.

Patients from Scotland with inflammatory skin disease being treated during winter with UV light have provided an interesting analysis of the immune outcomes that may be vitamin D₃-dependent and those that may be vitamin D₃-independent.⁶⁵ Due to the season, the patients were starting from a relatively low vitamin D₃ level (34 nmol/L). They were given four weeks of narrowband UV light therapy, which both increased their vitamin D₃ levels and proportionally increased the levels of their circulating regulatory T cells. However, independently of changes in their vitamin D₃ levels, the UV exposure caused a reduced proliferative response and reduced IL-10 production by their blood cells stimulated in vitro with anti-CD3/CD28. This study suggests that immune functions may be independently controlled by UVR by vitamin D₃-dependent and vitamin D₃-independent mechanisms.⁶⁵ More studies in humans attempting to dissect these mechanisms are required. Of note, in an epidemiological study of demyelination of the central nervous system in patients with multiple sclerosis, both sun exposure and vitamin D₃ levels were considered independent risk factors.⁶⁶

Experimental Murine Studies

In the light of the uncertainty in human studies, many researchers including ourselves have turned to murine experimental models to clarify a role for UV-induced vitamin D₃ for the modulation of immune disorders.⁶⁷ We have recently demonstrated that UVR-induced 25(OH)D₃ is not essential for mediating the immunosuppressive effects of erythemal UVR in male mice.⁶⁸ In this model, vitamin D₃-deficient mice(<20 nmol/L 25(OH)D₃ in serum) were acutely irradiated with erythemal UVR. Serum 25(OH)D₃ levels significantly increased in vitamin D₃-deficient female but not male mice. The inability to alter serum 25(OH)D₃ levels after the irradiation of male mice with UVR may have been due to reduced skin levels of the vitamin D precursor, 7-dehydrocholesterol, and enhanced 1,25(OH)₂D₃ metabolism.⁶⁸ Increased expression of mRNA for the enzyme responsible for the synthesis of 1,25(OH)₂D₃ (CYP27B1) and the enzyme inactivating 1,25(OH)₂D₃ (CYP24A1) were detected in the kidneys of male mice.⁶⁸ Erythemal UVR suppressed contact hypersensitivity responses and allergic airway disease (murine asthma) in male and female vitamin D₃-deficient mice to a similar degree,indicating that UVR-induced vitamin D₃ is not absolutely required for immunosuppression.⁶⁸

Results from other studies have also indicated that vitamin D₃ is not a mediator of UVR-induced suppression of immunity. In studies similar to our own, others have shown that UVR suppressed the severity of experimental autoimmune encephalomyelitis (a murine model of multiple sclerosis) when there were minimal changes in serum 25(OH)D₃.²³ Schwarz and colleagues also demonstrated that UV-induced immunosuppression was not affected by the loss of the VDR in mice i.e. the extent of immunosuppression was similar in mice with and without the product of the VDR gene expressed.⁶⁹ However, in other studies, expression of the VDR was required for control of the anti-inflammatory capacity of mast cells in UV-irradiated skin after skin reconstitution of mast cell-deficient mice (WBB6F₁-Kit^W/Wv) with bone marrow-derived mast cells from wild type or VDR−/− mice.⁷⁰ These studies suggested that the VDR may be important for the ability of mast cells to suppress skin inflammation induced by UVR. In addition, dietary vitamin D₃ has been shown to suppress immunopathological outcomes in murine models of asthma,⁷¹ arthritis,⁷² type 1 diabetes⁷³ and inflammatory bowel disease. ⁶⁷ However, to add to the inconsistencies in the literature, if vitamin D₃ is immunoregulatory, one would expect enhanced severity of a model of multiple sclerosis in mice made vitamin D₃-deficient by dietary means. Instead, and as shown in theTable, the symptoms and intensity of experimental autoimmune encephalomyelitis in mice is significantly reduced when the mice are vitamin D₃ deficient. ^74,75

Many UV-induced mediators may contribute to vitamin D₃-independent suppression of immunity (Figure 2). Many of the pathways were discussed in our review in 2011 in Nature Reviews Immunology.²⁵ Very recently, it was reported that UVR exposure of keratinocytes could alter double-stranded domains of some noncoding RNAs and provide a damage-associated molecular pattern that, via Toll-like receptor 3, may contribute to the mechanisms of UV-mediated systemic immunosuppression.⁷⁶ Many studies also support the involvement of UVR-induced prostaglandin E₂ in UV-mediated systemic immunosuppression.^77,78 By binding to the prostaglandin E receptor subtype 4, signalling from the receptor stimulates increased receptor activator of NF-κB ligand (RANKL) expression in the epidermis and an elevation of T regulatory cells in the draining lymph nodes.⁷⁷ We also propose that prostaglandins may mediate UV-induced immunosuppression by reducing the ability of progenitor dendritic cells from the bone marrow to develop into effective antigen-presenting cells.⁷⁸ A potential mechanism for this effect could be through epigenetic modification of committed progenitor cells in the bone marrow.^67,68 Another UVR-induced molecule is cis-urocanic acid.⁷⁹Trans-urocanic acid is a histidine derivative in the stratum corneum that isomerises to the more soluble cis form upon UV-irradiation.⁷⁹ Multiple mechanisms by which cis-urocanic is immunomodulatory have been proposed; these include an ability to stimulate the production of reactive oxygen species in keratinocytes causing oxidative DNA damage and immunosuppression.⁸⁰ Other studies have suggested that cis-urocanic acid can modulate the production of immunosuppressive molecules from keratinocytes, nerves and mast cells.^81–83 The ability of cis-urocanic acid to be immunosuppressive systemically has also been demonstrated in a murine model of chemically-induced colitis,⁸⁴ and in human skin during a contact hypersensitivity response.⁸⁵ Further research on the immunoregulatory properties of cis-urocanic acid, as well as other UVR-induced molecules in the skin, are required.than We longer exposure of propose that with evolution, multiple pathways have developed to reduce overzealous immune responses to antigens associated with skin damage and may contribute to UVR-induced systemic immunosuppression (Figure 2).

Figure 2.

Proposed mechanisms of systemic immunosuppression following UV irradiation of skin.

Sun Exposure

In response to the uncertainty of the role of vitamin D₃-dependent and vitamin D₃-independent pathways in UVR-induced immunoregulation, moderate sun exposure must be recommended for optimal benefits of UVR-induced immunoregulation. However, advocacy of moderate sun exposure may enhance the risk of development of skin carcinomas. In a recent position statement for Australians published in the Medical Journal of Australia, it was recommended that ‘for moderately fair-skinned people, a walk with arms exposed for 6–7 minutes mid-morning or mid-afternoon in summer, and with as much bare skin exposed as feasible for 7–40 minutes (depending on latitude) at noon in winter, on most days, is likely to be helpful in maintaining adequate vitamin D levels in the body’.⁸⁶ Whether this is sufficient sun exposure is a matter of debate and not for discussion here. However, enhanced vigilance and education in identifying skin lesions is recommended. Repeated short sun exposures to a larger body surface area may increase vitamin D₃ levels more efficiently than longer exposure of smaller body areas.^9–11 Further, one does not need to obtain reddening of the skin for either vitamin D₃ production or for the vitamin D₃-independent immunomodulatory effects of UV exposure; UVR-induced systemic immunosuppression has been measured in humans after exposure to non-inflammatory, suberythemal amounts of UVR.^8,9

Moderate sun exposure will allow one to increase vitamin D₃ levels the ‘natural’ way. In the skin there are inbuilt control mechanisms that prevent excessive vitamin D₃ production. As shown in a publication by Norval and colleagues,⁸⁷ and in Figure 1, limited amounts of pre-vitamin D₃ convert thermochemically to vitamin D₃ in keratinocytes of the skin, and are subsequently released from the cell membrane to the extracellular space. Upon production of large amounts of pre-vitamin D₃, other photoproducts are synthesised including lumisterol, toxisterols and tachysterol.⁸⁷ For vitamin D₃ supplementation by ingestion, there are not the inbuilt controls to limit production of excessive 25(OH)D₃. If one is vitamin D₃-insufficient (50–75 nmol/L) rather than deficient (<50 nmol/L), a lifestyle change to obtain more sun exposure is frequently recommended by medical practitioners.

Conclusions

This review has addressed the effects of the UV component of sunlight on immune processes and chronic diseases associated with over-zealous immune responses to autoantigens or allergens. The main diseases addressed were multiple sclerosis, allergic asthma and types 1 and 2 diabetes. Similar questions have been asked about the role of UV and vitamin D₃ in control of other common immune system-driven diseases like psoriasis²⁵ and the inflammatory arthritic diseases. The literature supporting the necessity for sufficient vitamin D₃ for the development and maintenance of bone health is strong,²⁷ and there is considerable data emerging suggesting that vitamin D₃ may contribute to patterns of brain development.⁸⁸ This review covers exposure to UVR, vitamin D₃ and immune health, with many of the main points summarised in the Table.

Table.

Similarities and inconsistences of published studies that have addressed the vitamin D₃-dependent and vitamin D₃-independent effects of UV exposure on reduced immunity in chronic immune diseases.

	Multiple sclerosis	Allergic asthma	Diabetes
Latitude gradient for disease incidence	Yes18	Yes13	Yes14,15
Animal models: Effect of UVR	Reduced expression of experimental autoimmune encephalomyelitis23	Reduced expression of experimental allergic airways disease2,3,24
Association of low vitamin D3 levels with increased incidence/relapse rate/severity of human disease	Yes28,30,45,46	Yes 49,50	Yes44,51
Impact of vitamin D3 supplementation of patients with disease	Inconsistent outcomes52–54	Inconsistent outcomes55–59	Inconsistent outcomes51,61,62
Impact on offspring of vitamin D3 supplementation during pregnancy		No consistent reduction of disease58,59	No consistent reduction of disease51
Animal models: Effect of vitamin D3 deficiency (if vitamin D3 regulatory, increased disease expected)	Reduced expression of experimental autoimmune encephalomyelitis74,75	Increased expression of some aspects of experimental allergic airways disease71	Increased expression of experimental type 1 diabetes with early life vitamin D deficiency73

Both latitude of residence and serum vitamin D₃ levels have been negatively associated with increased incidence of immune diseases. However, intervention studies with vitamin D₃ of patients with immune diseases have not been definitive. The determination of the extent of involvement of vitamin D₃-dependent and vitamin D₃-independent pathways for immune homeostasis and regulation of immune system-driven diseases has been difficult. If UV-induced vitamin D₃ is responsible for the immunoregulatory effects of UVR, supplementation with vitamin D₃ would provide a cheap and easy therapeutic approach for these diseases. In murine models of disease, the balance of contribution of vitamin D₃-dependent and vitamin D₃-independent pathways to UVR control of disease suggests that the outcome may depend on the disease model/tissue under study. It is proposed that complementary vitamin D₃-dependent and vitamin D₃-independent pathways for immunoregulation are operative after exposure of human skin to UVR. There is a need for more research in this area, particularly as it is important for evidence-based recommendations to the public for skin protection when in the sun. Further, if not UVR-induced vitamin D₃, the identity of the active molecules involved in the UVR-induced, vitamin D₃-independent pathways for immune homeostasis are not known. Further research is required to identify the UVR-induced molecules, both vitamin D₃ and others, their modes of regulation and their interactions, and how they can contribute to immune homeostasis and regulation of immune-driven diseases. UVR is an important component of our everyday environment so it is important that we know more about its effects on the skin, how it influences our immune balance, and whether it can be manipulated for therapeutic use.