A Potential Role for Vitamin D on HIV Infection?
Nutrition Reviews, Vol. 64, No. 5 May 2006
Eduardo Villamor, MD, DrPH
Dr. Villamor is with the Department of Nutrition,
Harvard School of Public Health, Boston, Massachusetts,
Please address all correspondence to: Dr. Eduardo
Villamor, Department of Nutrition, Harvard
School of Public Health, 665 Huntington Ave., Boston,
MA, 02115; Phone: 617-432-1238; Fax: 617-432-
2435; E-mail: firstname.lastname@example.org.
Despite advances in the knowledge of vitamin D's potent immunomodulatory activity, its role on HIV disease progression is unknown. Decreased concentrations of 1a,25-hydroxyvitamin D3, or 1,25(OH)2D, the active form of vitamin D, have been reported among HIV-infected people and attributed to defects in renal hydroxylation and increased utilization. A few studies also described low levels of 25-hydroxyvitamin D3, 25(OH)D, the vitamin obtained from solar synthesis and diet. An inverse association between 1,25(OH)2D concentrations and mortality has been reported from a small cohort of HIV-infected adults, and some cross-sectional studies have indicated positive
correlations between 1,25(OH)2D and CD4+cell counts. Additional observational studies are needed to confirm the associations between vitamin D status and
HIV disease progression. These investigations would provide useful insights on the potential role of vitamin D supplementation to HIV-infected persons and the
planning of intervention trials.
"CONCLUSION: Several of the studies reviewed suggest that concentrations of 1,25(OH)2D are often decreased among HIV-infected persons. The causes of this decrease are not known but might include a defect in renal 1-a-hydroxylation of 25(OH)D mediated by pro-inflammatory cytokines or by the intake of antiretroviral drugs. Vitamin D deficiency is not quoted as a likely cause of decreased 1,25(OH)2D in the course of HIV infection, since low concentrations of 25(OH)D have been found in some but not all the studies. However, the limitations of these studies, mostly cross-sectional and observational in nature, preclude the exclusion of vitamin D deficiency as a potentially serious problem among HIV-infected individuals, particularly in populations with limited exposure to sunlight for cultural or geographical reasons. Studies on the prevalence of vitamin D deficiency in the settings most heavily hit by the HIV epidemic are urgently needed. These studies should include longitudinal comparisons of indicators of vitamin D status between groups of HIV-infected and uninfected persons. An association between poor vitamin D status and mortality among HIV-infected adults has been suggested by one small cohort study. This association needs to be confirmed, for example, by measuring metabolites of vitamin D in stored samples from patients who have participated in longitudinal HIV studies. Confirmation of this association would be crucial to advance the discussion of a potential role of vitamin D supplementation in HIV-infected individuals, and the planning of future intervention trials."
HIV infection had affected an estimated 39.4 million people worldwide by the end of 2004, and is currently the single most infectious cause of mortality in developing countries, with over 3 million deaths in that same year.1 More than 90% of HIV-infected people live in areas of the world where nutritional deficiencies are also highly prevalent, and the interactions between HIV and
nutritional status have been widely documented. Deficiencies of several micronutrients have been associated with increased rates of progression to AIDS and HIV-related mortality,2-5 and selected nutritional interventions
including supplementation with vitamins B, C, and E have been found to decrease the risk of HIV disease progression and AIDS-related deaths.6 The beneficial effect of vitamin supplements on health and survival outcomes among HIV-infected persons is likely to be mediated by enhancements in specific aspects of immunity that include increases in CD4+ cell counts6,7 and
reduced viral loads.6,8
Vitamin D possesses strong immunomodulating properties. In the 1980s, the nuclear vitamin D receptor (VDR) was discovered in various immune cell lines that included T and B lymphocytes, monocytes, and dendritic cells.9-13 This discovery provided preliminary mechanistic evidence to link reported associations between vitamin D deficiency and the occurrence of respiratory infections, 14-21 diarrheal disease,15 and tuberculosis,22 with the regulatory actions of vitamin D on the immune system. Current research, reviewed in detail elsewhere, 23-25 suggests that such immunomodulating actions could be "paradoxical," as they depend on the specific immune response elicited by particular stimuli. For example, while reactions depending on T-cell helper responses of the pro-inflammatory Th1-type appear to be down-regulated by vitamin D, those that depend on Th2-type reactions may be up-regulated.23
It is uncertain whether vitamin D status could be associated with health and immunological outcomes related to HIV disease. This question becomes particularly intriguing as the knowledge of the immunological functions
of vitamin D continues to grow. The recent identification of high rates of subclinical vitamin D deficiency in various countries26-29 is an additional incentive to assess the functional consequences of deficiency under
a variety of pathophysiological conditions, including HIV infection. In Africa and southeast Asia, where the HIV epidemic is having the strongest impact, vitamin D
deficiency is likely to be a public health problem in subpopulation groups in which solar exposure is limited for cultural reasons.30 Updated reports on the prevalence of vitamin D deficiency in the general population of these regions are lacking.
This manuscript examines evidence available from experiments and epidemiological studies on the interactions between vitamin D status and HIV infection. Whether vitamin D deficiency is a prevalent condition among HIV-infected people and the potential mechanisms through which HIV infection could affect vitamin D status is explored first. Next, evidence on how vitamin D could affect HIV disease progression is reviewed.
VITAMIN D STATUS OF HIV-INFECTED
Vitamin D (D3) is obtained mostly by photolysis of dehydrocholesterol in the skin. It is transported to the liver by the vitamin D-binding protein, where a vitamin
D3-25-hydroxylase catalyzes its conversion into 25-hydroxyvitamin D3 (25[OH]D). Bound to the vitamin Dbinding protein, 25(OH)D is exported from the liver to the kidneys, where a 25(OH)D3 1a-hydroxylase converts it into the biologically active metabolite, 1a,25-hydroxyvitamin D3, also known as 1,25(OH)2D, the natural ligand of the VDR at a variety of target tissues. The
integrity of this pathway can be assessed using the concentrations of vitamin D metabolites. The circulating concentration of 25(OH)D is considered to be the best indicator of vitamin D status, and represents the vitamin D obtained from solar synthesis in the skin and from the diet; levels under 80 nmol/L (32 ug/L) are considered to be indicative of vitamin D deficiency.31 The concentrations
of 1,25(OH)2D provide an indication of renal 1a- hydroxylation and also of defects in disposal by the cytochrome P-450 enzyme CYP24.
Studies reporting on the concentration of vitamin D metabolites among HIV-infected subjects offer an opportunity to examine whether vitamin D deficiency is likely to be a problem in this subgroup and/or whether alterations in the vitamin's metabolic pathway are present in the course of HIV disease. In a study of HIV-infected adults from the United States, the serum concentration of 25(OH)D was below normal values in 2 out of 12 subjects (17%), whereas the concentration of
1,25(OH)2D was low in 3 of 27 individuals with the measurement (11%)32; the cutoff points for normality were the mean minus 2 SD in the same population, but
the absolute values were not presented. In a study from Germany, low concentrations of 25(OH)D (<50 nmol/L) were found in 10 out of 21 patients with AIDS who had hypocalcemia without any obvious cause.33 The authors
suggested that vitamin D deficiency could be an underlying cause of hypocalcemia among HIV-infected subjects. In a study of 22 HIV-infected subjects from Spain, half of whom had AIDS, average 25(OH)D and
1,25(OH)2D concentrations were 16.5 ug/L and 35.8 ng/L, respectively34; the subjects had not begun antiretroviral treatment. The mean serum levels of 25(OH)D and 1,25(OH)2D were 33.6 ug/L and 61.8 ng/L, respectively,
in 5 and 12 girls 6 to 15 years of age from the United States who had been infected perinatally with HIV.35 These girls had low calcium intake and all were
receiving antiretroviral therapy. Inferring whether HIV infection is associated with vitamin D deficiency from the studies above is limited by the lack of comparison
with an HIV-negative control group.
Comparisons of vitamin D metabolites between HIV-infected and uninfected subjects have been presented in a few reports.36-38 Among 6 patients with
advanced AIDS in Switzerland, the concentrations of 25(OH)D and 1,25(OH)2D were not significantly different from those of 10 HIV-uninfected controls, although
their levels of parathyroid hormone and calcium were, on average, lower.37 In a study conducted in Germany, the mean concentration of 1,25(OH)2D was significantly lower among both men (n =65; mean = 30.3 ng/L) and women (n =35; mean = 22.4 ng/L) infected with HIV, compared with 20 men (mean = 67.5 ng/L) and 20 women (mean = 55.3 ng/L) who were uninfected.38 25(OH)D was nonsignificantly lower in HIV-infected men (mean =42.4 ug/L) and women (mean =37.3 ug/L) compared with HIV-negative controls (means = 69.5 ug/L in men and 62.7 ug/L in women). These patients were asymptomatic and were receiving antiretroviral treatment that did not include protease inhibitors. In a subsequent study in which the sample of HIV-infected women was expanded to 50 and compared against 50 uninfected controls, the same authors confirmed that 1,25(OH)2D and 25(OH)D average concentrations were significantly lower among HIV-positive individuals (19.4 ng/L and 37.3 ug/L, respectively) compared
with negative controls (47.3 ng/L and 61.5 ug/L, respectively).39 The authors suggested that low 1,25(OH)2D concentrations could be responsible for an
increased bone turnover rate that is observed as HIV disease progresses. In a study in Norway, the mean serum concentration of 1,25(OH)2D was significantly lower in 53 HIV-infected patients in a hospital (median = 42 ng/L) compared with that of 28 healthy, seronegative blood donors (median = 49 ng/L)36; 1,25(OH)2D was lowest in the 31 HIV-infected patients who were symptomatic, independent of the presence of opportunistic infections. In the same study, the concentrations of 25(OH)D or vitamin D-binding protein40 were
not associated with HIV status or disease stage. Since the 25(OH)D level, an indicator of vitamin D from diet or skin, was unaffected by HIV status, the low concentrations of 1,25(OH)2D in this study were not likely attributable to vitamin D deficiency; instead, they could have been due to a defect in the 1-a-hydroxylation of 25(OH)D into 1,25(OH)2D that should normally occur in
the kidney in response to low 1,25(OH)2D or parathyroid hormone. A mechanism to explain this potential defect is not known, but could be related to overactivation of tumor-necrosis factor-alpha (TNF-a) under certain circumstances
(such as infection with Mycobacterium avium complex41); excessive TNF-a might block the stimulatory effect of parathyroid hormone on the renal
1-a-hydroxylase.40,42 Alternatively, low 1,25(OH)2D could be due to increased utilization for maturation and proliferation of T lymphocytes during HIV infection.
Another potential reason the concentrations of 1,25(OH)2D seem to be lower in HIV-infected subjects is an effect of antiretroviral medications on the metabolic pathways of vitamin D. In a recent study comparing 1,25(OH)2D levels between 172 HIV-infected and 64 uninfected subjects, the average 1,25(OH)2D concentrations were lower among those infected with HIV compared with controls, irrespective of the type of antiretrovirals they were receiving or whether they were receiving any at all; however, the concentrations were lowest in the patients who were receiving proteinase inhibitors.43 The possibility that protease inhibitors suppress the activity of 1a-hydroxylase, the enzyme that converts 25(OH) into bioactive 1,25(OH)2D, was supported by an in vitro experiment in which human hepatocyte and monocyte cell lines were exposed to proteinase inhibitors. Exposed cells showed markedly reduced activities of both 25- and
1a-hydroxylase, particularly in response to ritonavir; there was also a mild inhibitory effect on 24-hydroxylase, an enzyme responsible for 1,25(OH)2D catabolism. 44 It is not known whether this effect could also occur in the kidney, where 1a-hydroxylase is most active in vivo. 1a-Hydroxylase can be also inhibited by ketoconazole,45 another medication frequently used
in HIV-infected subjects to protect against mycotic opportunistic infections.
There are a few case reports of hypercalcemia in HIV-infected patients46-51 that have been attributed to increased 1,25(OH)2D; however, in these cases the elevated 1,25(OH)2D was likely to be the result of extrarenal 1a-hydroxylation in tumors or monocytic cells. One in vitro experiment suggested a mechanism by which HIV infection could result in a paradoxical increase in the peripheral production of 1,25(OH)2D through an induction of the 1a-hydroxylase activity in peripheral blood monocytes by the p17 matrix protein in the presence
of 25(OH)D.52 Peripheral production of 1,25(OH)2D through this pathway, however, is unlikely to reflect in circulating concentrations of the metabolite.
It is not possible to ascertain whether low vitamin D concentrations preceded HIV infection in the cross-sectional studies reviewed above. To examine more accurately whether HIV infection has an impact on vitamin D status, the incidence of vitamin D deficiency should be compared between HIV-infected and uninfected subjects who are vitamin D sufficient at the outset. In the Norway
study,36 among 14 subjects who had normal concentrations of 1,25(OH)2D at baseline and had a second measurement within a year, the levels of 1,25(OH)2D decreased in two patients. However, a comparison with HIV-uninfected subjects was not offered. In a study in Italy among 27 HIV-positive patients who were receiving antiretrovirals and had low 1,25(OH)2D concentrations at baseline, a follow-up evaluation after 14 months revealed a further decrease in 1,25(OH)2D.43 One longitudinal animal study has suggested mechanisms regarding the potential impact of HIV infection on vitamin D status. This was a model of 8 Rhesus macaques that were experimentally infected with simian immunodeficiency virus (SIV). In this study, Kewenig et al.53 observed a
decrease in the serum concentration of 25(OH)D from baseline to 4 weeks post-infection in 6 of the monkeys. The values at 12 weeks were more than 25% lower than baseline in 4 of 6 monkeys for which the measurement was available at that time point. In contrast, no changes were observed during the study period in either of 2 monkeys that were kept uninfected as controls. The authors suggested that the decrease in vitamin D concentrations was due to intestinal malabsorption caused by an enteropathogenic effect of SIV in the distal small intestine mucosa. This suggestion was based on the observations that the decrease in vitamin D concentrations was correlated with villous atrophy and was more profound in 4 monkeys infected intrarectally compared with 4 infected intravenously; also, opportunistic intestinal infections that could explain malabsorption were not found. One potential implication of these findings is that profound dietary vitamin D deficiency could develop soon after HIV infection, when it is still asymptomatic. In humans, however, cutaneous synthesis could compensate for deficiency from malabsorption.
Low dietary intake could contribute to vitamin D deficiency in populations that may not receive adequate sunlight exposure for geographical or cultural reasons.
Some cross-sectional studies have indicated low dietary intake of vitamin D among HIV-infected subjects. In a survey of 81 HIV-positive subjects in South Africa, 64% had dietary intakes below the 67% of the recommended
dietary allowance (RDA), irrespective of their stage of disease, as measured with CD4+ cell counts.54 Among 119 HIV-infected adult men and women from Spain, the mean dietary intake of vitamin D represented 78% of the RDA.55
HOW VITAMIN D STATUS COULD AFFECT
HIV DISEASE PROGRESSION
The evidence on whether and how vitamin D could influence HIV susceptibility and disease progression will be reviewed from three groups of studies: 1) in vitro
experiments, 2) observational studies in humans, and 3) population studies of polymorphisms in the vitamin D receptor.
In Vitro Experiments
The effects of vitamin D on HIV infection in vitro have been examined in terms of its potential role on monocyte/macrophage function and on HIV expression
and replication in monocytes/macrophages. With respect to monocyte/macrophage function, one study of monocytes from 10 AIDS patients showed that incubation with 1,25(OH)2D resulted in significant increases in chemotaxis. 56 In another experiment, 1,25(OH)2D tended to decrease the number of M. avium bacteria in macrophages from HIV-positive patients, whereas the opposite effect was apparent in macrophages of HIV-uninfected
controls.57 These findings suggest that 1,25(OH)2D could enhance some macrophage functions such as the respiratory burst in cells from HIV-positive persons, or it may have a direct effect against M. avium replication in these subjects. A study on the effect of 1,25(OH)2D on monocyte function showed that vitamin D improved growth and maturation parameters of monocytes from
both HIV-infected and uninfected patients, but among the former, the improvements in vitro were greater in patients with low CD4+ cell counts or symptomatic disease compared with those at less-advanced stages.58
No effects were observed in monocytes from the sickest subjects, suggesting that very severe monocytic dysfunction may not be responsive to 1,25(OH)2D.
Regarding the effect of vitamin D on HIV replication, the results are conflicting. Early studies showed that 1,25(OH)2D could decrease the expression of CD4+
surface receptors in a promyelocytic leukemia cell line59 and in human monocytes.60 If confirmed, these results could constitute a mechanism to control viral entry and cytopathogenic effects. In another study, pretreatment of
human peripheral blood monocytes with 1,25(OH)2D, but not with 25(OH)D, decreased HIV infection by 95%.61 There was also a reduction in the expression of CD4+, but this happened after the decrease in p24 antigen production, which suggested that a 1,25(OH)2Dmediated inhibition of CD4+-dependent binding of HIV was not the only potential factor involved. In another study, pretreatment of monocytes with 1,25(OH)2D for 5 days was related to resistance to HIV infection, apparently as a consequence of inhibited monocyte proliferation. 62 In an experiment with monocyte-derived macrophages, which can differentiate spontaneously in culture, pretreatment with 1,25(OH)2D for 3 to 5 days before
infection resulted in strong suppression of viral replication. 63 In contrast, promonocytic U937 cells that were exposed to 1,25(OH)2D before infection showed an increase in HIV replication. A similar adverse effect on the
U937 line was shown in an experiment in which vitamin D incubation promoted both macrophage-like differentiation and increased HIV production.64 1,25(OH)2D also increased HIV replication in studies of promyelocytic/
myeloblastic HL-60 cells65 and in A3.5 cells and peripheral blood monocytes that had been acutely infected with HIV.66 Vitamin D analogs enhanced HIV-1 replication in proportion to their affinity for the VDR. Marked increases in viral replication were also reported from an experiment in which human macrophages were either pretreated for 3 days with 1,25(OH)2D before HIV infection
or incubated with 1,25(OH)2D only after infection. 67
Since 1,25(OH)2D stimulates maturation of monocytes to macrophages, its effect on HIV expression could be partly related to its role on cellular differentiation, and
the fact that investigators have used cells at different stages of differentiation could be a partial explanation for the inconsistency of results. It has been shown that 1,25(OH)2D-induced maturation of chronically infected promonocytic cells is not always accompanied by increased HIV replication, but this depends on the synergistic effect of other factors such as TNF-a.68 It has also been reported that in one single cell line, such as the U937, not all the clones are equally susceptible to HIV infection,69 and that 1,25(OH)2D-dependent increased
HIV-1 replication may occur in some, but not all, clones, depending on their ability to sustain HIV infection.70 It appears that vitamin D could enhance viral replication by upregulating the expression of CXCR4 in promonocytic
cells that are normally poorly efficient to sustain HIV infection.70
Observational Studies in Humans
In the Norway study among HIV-infected adults,36 the concentration of 1,25(OH)2D was positively correlated with CD4+ (r =0.35; P <0.05) and CD8+ (r =0.33; P <0.01) cell counts, whereas no relation was found with 25(OH)D. Similarly, a positive correlation between CD4+ and 1,25(OH)2D was found in a study in Germany among men and women (r =0.23, P <0.05)38 and among women alone (r = 0.45, P < 0.05),39 although no correlation existed with 25(OH)D. In a smaller study of 22 Spanish patients, no correlations were found
between either of these metabolites and CD4+ cell counts,34 but in a different population of Spanish subjects, there was a significant, positive correlation be tween vitamin D intake from the diet and CD4+ cell counts (r =0.51; P < 0.001).55 After adjustment for age, sex, antiretroviral treatment, and dietary protein intake, every 1 ug of vitamin D in the diet was associated with
a significant increase of 34 CD4+cells (95% CI = 6-167; P <0.001). In this study, the assessment of diet with a 24-hour recall represents current intake, and it is
possible that advanced disease may have induced changes in the patients' diet, opening the possibility for reverse causality. No correlations between 1,25(OH)2D and CD4+ or viral load were found in a study of 172 Italian HIV patients, 152 of whom were receiving HAART.43 Among 12 peri-adolescent girls who had been infected perinatally, O'Brien et al.35 found no correlation between the Centers for Disease Control (CDC) classification of disease stage and the concentration of 1,25(OH)2D in a cross-sectional study. The positive correlations between vitamin D status and CD4+ cell counts could indicate a link between 1,25(OH)2D and HIV disease progression, but the cross-sectional nature of the comparisons and the possibility of confounding by other factors, including antiretroviral treatment, do not allow causal inferences at this point.
One longitudinal study examined the impact of vitamin D concentrations on survival among HIV-infected individuals.36 In this small study, Norwegian patients who had 1,25(OH)2D concentrations under 25 pg/mL at baseline (N =9) had a significantly shorter survival time than those with normal concentrations (N=44), after adjustment for CD4+ cell counts. The association appeared
to be stronger in patients with CD4+ cell counts under 50/mm3. Baseline concentrations of 1,25(OH)2D in this study had been particularly low among AIDS patients with M. avium complex infection, and correlated
inversely with serum concentrations of immunoreactive TNF-a (r = -0.78; P <0.05).41 It is possible that low 1,25(OH)2D could contribute to impaired antimycobacterial activity or that excessive TNF-a activation may inhibit the antimycobacterial effect of 1,25(OH)2D, either of which could lead to accelerated disease progression and increased mortality.
There exist case reports on the successful use of vitamin D analogs in the treatment of specific HIV-related conditions, such as calcipotriol as a topical therapy for psoriasis,71 but evidence from large epidemiological studies is lacking.
Vitamin D Receptor and HIV Infection Susceptibility
The risk of HIV disease infection and progression has been examined in relation to the presence of a number of polymorphisms in the VDR gene. In a study of Caucasians from Spain, Barber et al.72 examined the prevalence of mutations in a region of the VDR gene corresponding to the BsmI restriction enzyme in relation to HIV status. The BsmI polymorphism represents the variability of the 3" untranslated region (UTR) of the VDR gene, in which another two restriction fragment length polymorphisms (ApaI and TaqI) and a poly (A) microsatellite polymorphism have also been identified. These four polymorphisms exhibit almost full linkage disequilibrium in white populations.73 There were no significant differences in the distribution of genotypes for the BsmI polymorphism between 120 controls and 185 HIV-positive intravenous drug users, suggesting that this polymorphism does not affect susceptibility to HIV-1 infection.Among the 185 HIV-positive subjects, those who were homozygous for the B allele (corresponding to the absence of the BsmI restriction site) were at significantly higher risk of either progression to AIDS or a decline in CD4+ cells to under 200/mm3 after 7 years of follow-up, compared with those homozygous forthe b allele and heterozygous subjects. The results were adjusted for age at time of first HIV-positive test, age, and antiretroviral status.
In the same population, Nieto et al.74 compared the frequency of an SNP in the endonuclease restriction site of FokI between HIV-infected patients and healthy controls. The Fok-I region is located at the initiation codon of the VDR gene. This polymorphism is a T3C transition that eliminates the first translation initiation codon of the VDR gene; in homozygous individuals for the mutation (F allele), VDR has a length difference of three amino acids, which may affect the receptor function. There were no significant differences in the Fok-I genotype frequencies between HIV-infected individuals and uninfected controls, suggesting that the polymorphism was not associated with susceptibility to HIV infection. However, among the HIV-infected subjects, those who were heterozygous (Ff) were overrepresented among those who progressed to AIDS during an 8-year follow-
up period (50%) compared with non-progressors (36%; P =0.06) and among patients who had a decline in CD4" to under 200/mm3 (52%) compared with those with no decline (36%; P =0.04). In time-to-event analyses, the heterozygous genotype was also significantly associated with the progression to AIDS and CD4+ count drop compared with the FF and ff genotypes together, after adjusting for sex and age at first HIV-positive test. There appeared to be a significant interaction between the Fok-I and the Bsm-I polymorphisms,
because carriers of the lowest-risk genotypes, non-Ff/ non-BB, had the lowest risk of progression to AIDS or experiencing a decrease in CD4_ cell counts.
Although the mechanisms through which the associations between VDR polymorphisms and HIV disease progression are not known, one possible explanation is that mutations in the VDR gene affect the immunomodulatory
actions of 1,25(OH)2D through VDR. The VDR-BB and VDR-Ff genotypes could be associated with reduced response to the immunosuppressive actions
of vitamin D, which would normally induce a switch toward activation of Th2-type cells and inhibition of Th1 responses. Lack of this inhibition would result in increased proliferation of Th1 cells that are susceptible to HIV infection.
The heterozygous genotype (Tt) for a polymorphism identified by the TaqI enzyme in 3' non-coding sequences of exon 9, was reported to be related to reduced risk of HIV-1 infection in a large case-controlled study in Uganda75; unfortunately, no additional details are available from this study, as those particular results were unpublished. This polymorphism has been previously
associated with susceptibility to different forms of leprosy. Some possible candidate mutations in the VDR gene that could be studied in relation to HIV infection susceptibility have been identified in an ongoing project in South Africa, including SNPs in codons 1 and 19 and intron 3.76
Several of the studies reviewed suggest that concentrations of 1,25(OH)2D are often decreased among HIV-infected persons. The causes of this decrease are not known but might include a defect in renal 1-a-hydroxylation of 25(OH)D mediated by pro-inflammatory cytokines or by the intake of antiretroviral drugs. Vitamin D deficiency is not quoted as a likely cause of decreased 1,25(OH)2D in the course of HIV infection, since low concentrations of 25(OH)D have been found in some but not all the studies. However, the limitations of these studies, mostly cross-sectional and observational in nature, preclude the exclusion of vitamin D deficiency as a potentially serious problem among HIV-infected individuals, particularly in populations with limited exposure to sunlight for cultural or geographical reasons. Studies on the prevalence of vitamin D deficiency in the settings most heavily hit by the HIV epidemic are urgently needed. These studies should include longitudinal comparisons of indicators of vitamin D status between
groups of HIV-infected and uninfected persons. An association between poor vitamin D status and mortality among HIV-infected adults has been suggested by one small cohort study. This association needs to be confirmed, for example, by measuring metabolites of vitamin D in stored samples from patients who have
participated in longitudinal HIV studies. Confirmation of this association would be crucial to advance the discussion of a potential role of vitamin D supplementation in HIV-infected individuals, and the planning of future intervention trials.