Vitamin D₃ Affects Differentiation, Maturation, and Function of Human Monocyte-Derived Dendritic Cells¹

1α,25-Dihydroxyvitamin D₃ (1α,25-(OH)₂D₃)³ is a secosteroid hormone that binds to a nuclear receptor named vitamin D₃ receptor. During the past few years it has become apparent that 1α,25-(OH)₂D₃, in addition to its well-known role in mineral and skeletal homeostasis, regulates the differentiation, growth, and function of a broad range of cells, including cells of the immune system (1, 2, 3, 4). The immunological effects of pharmacological levels of 1α,25-(OH)₂D₃ or its analogues in vivo were demonstrated in studies of autoimmune disease (5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16) and studies of allograft rejection (17, 18, 19, 20). The effects of 1α,25-(OH)₂D₃ on the immune system were ascribed to its action on lymphocytes and monocytes/macrophages (21, 22, 23). When added to mitogen-stimulated human peripheral blood lymphocytes in vitro, 1α,25-(OH)₂D₃ inhibits their proliferation, Ig synthesis, and accumulation of transcripts for IL-1, IL-2, IL-6, TNF-α, and -β and IFN-γ (24, 25, 26). Of interest is that 1α,25-(OH)₂D₃ induces promyelocytes to differentiate into monocytes (27); in addition, 1α,25-(OH)₂D₃ differentiates myeloid leukemia cells to nonproliferating monocyte/macrophage-like cells in both humans and mice (28, 29) and promotes the differentiation of myeloid stem cells and normal peripheral blood monocytes toward a macrophage phenotype (30). 1α,25-(OH)₂D₃ also affects functional activities of monocytes and macrophages with contrasting results. Tumor cell cytotoxicity, phagocytosis, and mycobactericidal activity of monocytes/macrophages are enhanced by exposure to 1α,25-(OH)₂D₃ (31), but monocyte function as an APC appears be decreased (32, 33).

Over the past years in vitro methods have been described to differentiate dendritic cells (DC) from blood monocytes by in vitro culture with GM-CSF, IL-4 (34, 35, 36), or IL-13 (37). These cultured DC show functional and phenotypic characteristics typical of the immature stage of differentiation (i.e., high capacity of Ag uptake and processing, low capacity to stimulate T cell proliferation) and can be further differentiated in vitro into mature DC with TNF-α, LPS, IL-1, or CD40L (35, 38). As they are the most potent APC in vitro and in vivo, DC play a key role in the initiation of the immune response and are considered promising tools and targets for immunotherapy (39, 40, 41). It is therefore important to identify factors that might affect their process of differentiation and maturation (42, 43).

The aim of our work was to study the effects of 1α,25-(OH)₂D₃ on human monocyte-derived dendritic cell differentiation, maturation, and functional activities. Our results demonstrate that 1α,25-(OH)₂D₃ inhibited DC differentiation and maturation into potent APC. Moreover, D₃-DC promoted the onset of a nonspecific hyporesponsivity in T cells. These findings may have relevance in the development of new therapeutic treatments in the field of transplants and autoimmune diseases

Human recombinant GM-CSF (sp. act., 1.1 × 10⁴ U/mg) was obtained from Novartis (Basel, Switzerland). Human rIL-4 (sp. act., >2 × 10⁶ U/mg) and human rTNF-α (sp. act., >2 × 10⁷ U/mg) were obtained from PeproTech (London, U.K.). 1,25(OH)₂D₃ was purchased from Sigma (St. Louis, MO).

Highly enriched monocytes (>80% CD14⁺) were obtained from buffy coats of 20 blood donors (through the courtesy of Centro Trasfusionale, Ospedale San Raffaele, Milan, Italy) by Ficoll and Percoll gradients and were purified by adherence. Monocytes were cultured for 7 days at 1 × 10⁶/ml in six-well tissue culture plates (Falcon, Becton Dickinson, Rutherford, NJ) in RPMI (Biochrom, Berlin, Germany) and 10% FCS (HyClone, Logan, UT) supplemented with 50 ng/ml GM-CSF and 10 ng/ml IL-4 and with (D₃-DC) or without (ctr-DC) various concentrations of 1α,25-(OH)₂D₃. In the control group (GM-CSF plus IL-4) the cell yield was about 80% of input cells. All cultures were tested for the presence of endotoxin (<0.03 U/ml; Lymulus test, BioWhittaker, Walkersville, MD).

LPS (10 ng/ml) was added to induce maturation of DC for at least 36 h of culture. Alternatively, J558L cells transfected with the ligand for CD40 (J558LmCD40L) were used to induce CD40 triggering on DC. Untransfected J558L cells were used as a control. After irradiation (10,000 rad) J558L cells were seeded together with DC at a 1:1 ratio in 24-well culture plates in culture medium (1 × 10⁶ cells/well). Cells were recovered after 48–72 h of culture.

Cell staining was performed using mouse mAbs followed by FITC-conjugated, affinity-purified, isotype-specific, goat anti-mouse Abs (Ancell, Bayport, MN). The following mAbs were used: L243 (IgG2a, anti-MHC class II), 32.2 (anti-CD32), and IV.3 (anti-CD64) from American Type Culture Collection (Manassas, VA); UCHM-1 (IgG2a, anti-CD14) and W6/32 (IgG2a, anti-MHC I) from Sigma; SK9 (IgG2b, anti-CD1a) from Becton Dickinson (San Jose, CA); B73.1 (IgG2a, anti-CD16) from Dr. G. Trinchieri (Philadelphia, PA); PAM-1 (IgG1 anti-mannose receptor) (44, 45); BB1 (IgM, anti-CD80), BU63 (IgG1, anti-CD86), and EA-5 (IgG1 anti-CD40) from Ancell; and HB15a (IgG2b, anti-CD83) from Immunotech (Marseilles, France). Results are expressed as the percentage of positive cells or as fluorescence intensity (FI), calculated according to the formula: FI = mean fluorescence (sample) − mean fluorescence (control).

Mannose receptor (MR)-mediated endocytosis was measured as the cellular uptake of FITC-dextran and was quantified by flow cytometry. Approximately 2 × 10⁵ cells/sample were incubated in medium containing FITC-dextran (1 mg/ml; m.w., 40,000; Sigma) for 0, 60, and 120 min. After incubation cells were washed twice with PBS to remove excess dextran and were fixed in cold 1% formalin. The quantitative uptake of FITC-dextran by the cells was determined by FACS. At least 8,000 cells/sample were analyzed. Fluid phase endocytosis through membrane ruffling was measured as the cellular uptake of 1 mg/ml of Lucifer Yellow (LY) dipotassium salt (Sigma) and was quantified by flow cytometry.

DC cultured in GM-CSF and IL-4 and with or without 1α,25-(OH)₂D₃ for 7 days were extensively washed, irradiated (3000 rad from a ¹³⁷Cs source), and added in graded doses to 1 × 10⁵ responder cells in 96-well flat-bottom Microtest plates (Costar, Cambridge, MA). Responder cells were purified allogeneic T cells depleted of autologous APC by passage with CD14- and CD19-coated Dynabeads (Unypath, Milan, Italy). Each group was performed in triplicate. Thymidine incorporation was measured on day 5 by a 16-h pulse with [³H]thymidine (1 μCi/well; spec. act., 5 Ci/mmol; Amersham, Aylesbury, U.K.).

Tetanus toxin (TT)-responsive T cell lines were generated in our laboratory by culturing mononuclear cells with TT (36 μg/ml; Cannaught, Willowdale, Canada) for 1 mo in the presence of IL-2. TT-responsive T cells were tested at least 2 wk after the last PBMC stimulation and 5 days after the last addition of IL-2. DC were obtained from the same donor by culturing monocytes. After 7 days DC were preincubated with TT (6 μg/ml) for 12 h and with LPS for 48 h. Then DC were extensively washed, irradiated (3000 rad), and cocultured with autologous TT-responsive T cell lines for 72 h in 96-well microtiter plates, and [³H]thymidine uptake was measured during the last 12 h of culture (1 μCi/well; sp. act., 5 Ci/mmol; Amersham).

After 7 days of culture with GM-CSF (50 ng/ml) and IL-4 (10 ng/ml) in the presence or the absence of 1α,25-(OH)₂D₃, DC were washed twice and cultured for 3 days at 0.5 × 10⁶/ml in a 24-well flat-bottom plate (Costar). DC were either nonstimulated or stimulated with TNF-α (10 ng/ml), LPS (50 ng/ml), J558L cells transfected with CD40L (J558LmCD40L), or untransfected J558m cells. After 3 days medium was collected, and IL-12 p70 was quantified by ELISA (commercial kits from Endogen, Boston, MA).

Allogeneic T cells were prepared from human blood using Ficoll and Percoll gradients and subsequent depletion of B cells and monocytes by plastic adherence and by Ab-coated immunomagnetic beads (Unipath, Milan, Italy) according to a standard protocol. T cells were cocultured during the first incubation at a density of 1 × 10⁶/ml with 1 × 10⁴/ml ctr-DC or D₃-DC (DC were prepared as described above and matured with LPS). Three days later T cells were separated by Ficoll gradient (Sigma), extensively washed, depleted of DC using Ab-coated immunomagnetic beads (Ab anti CD32 and MR), and rested for 5 days in medium alone. Subsequently, T cells were restimulated with mature ctr-DC generated from the same donor used in the first coculture or from another unrelated donor.

DC were processed for electron microscopy. DC were fixed for 2 h in 2.5% glutaraldehyde in 0.1 M cacodylate buffer. Then they were postfixed in 1% OsO₄ in cacodylate buffer at 4°C for 1 h, dehydrated in graded ethanol up to propylene oxide, and finally embedded in an Epon-Araldite mixture. Well-preserved areas were identified by light microscopy of semithin sections (0.5 mm). Subsequently, serial ultrathin sections (80 nm) were mounted on 200-mesh copper grids, stained with uranyl acetate and lead citrate, and finally examined with a Zeiss CEM 902 electron microscope (New York, NY).

Data were expressed as the mean ± SD. Comparisons were performed using Student’s t test. A p value of <0.05 was considered statistically significant.

To investigate the effect of 1α,25-(OH)₂D₃ on DC differentiation from monocytes, we cultured monocytes in the presence of GM-CSF, IL-4 (control group, ctr-DC), and various concentrations (0.5–100 nM) of 1α,25-(OH)₂D₃ (D₃-DC). 1,25(OH)₂D₃ did not affect cell recovery at any concentration tested. The standard concentration of 1α,25-(OH)₂D₃ chosen for the study was 10 nM, the highest concentration considered physiological (4); this was also used in previous studies for leukocyte differentiation (30). Upon culture with GM-CSF and IL-4 for 7 days cells became nonadherent and clustered, with abundant cytoplasm and protruding veils typical of DC. Despite a similar morphology (Fig. 1), the presence of 1α,25-(OH)₂D₃ in culture interfered with the differentiation of monocytes into DC. Fig. 2 shows a representative experiment of surface phenotype. Control cells expressed high levels of CD1a and were negative or low positive for CD14 and CD16, while with 1α,25-(OH)₂D₃ the cells were negative or low positive for CD1a but expressed higher levels of CD14. Analysis of MHC class I showed an up-regulation of D₃-DC, whereas expression of MHC II, CD40, and CD86 molecules was decreased. Ag uptake molecules, such as CD32 and MR, were increased. DC obtained after 7-day culture with GM-CSF and IL-4 could be further differentiated in vitro into fully mature DC by exposure to LPS or CD40L. D₃-DC were not sensitive to maturation stimuli. In fact, after exposure to LPS (Fig. 3) or CD40L, D₃-DC were unable to up-regulate CD83 as well as the molecules involved in Ag presentation (MHC I, MHC II, CD80, CD86, and CD40) and to down-regulate the Ag uptake molecules (CD2 and MR). A summary of six different experiments is shown in Table I.

Immature DC, such as cells derived by culturing monocytes with GM-CSF and IL-4 for 7 days, express a potent ability to uptake external molecules, essentially via two main mechanisms: a receptor-mediated endocytosis and a fluid phase endocytosis (macropinocytosis). To study the endocytic capacity of D₃-DC, we used two fluorescent markers: LY, a nonspecific fluid phase marker, and FITC-DX, which is mainly taken up via the MR. DC cultured with GM-CSF and IL-4 in the presence of 1α,25-(OH)₂D₃ showed a vigorous endocytosis of FITC-dextran, higher than control DC (Fig. 4,A). The same behavior was seen when we used LY as marker of fluid phase pinocytosis (Fig. 4,B). DC are potent stimulators of allogeneic T cells. We tested whether D₃-DC were able to stimulate allogeneic T lymphocytes in MLR. D₃-DC showed very little ability to induce allogeneic T lymphocyte proliferation (Fig. 5,A). Moreover, the immunostimulatory capacity of D₃-DC in MLR was not increased by LPS or was increased to a much lower extent by CD40L, compared with that of ctr-DC (Fig. 5). In view of the fact that Ag capture was increased in D₃-DC but the stimulatory capacity was impaired in MLR, we evaluated the ability to present soluble Ag that need to be taken up and processed. Cells differentiated in the presence of 1α,25-(OH)₂D₃ showed much lower efficiency in presenting TT to specific autologous T cell lines (Fig. 6).

To investigate the capacity of 1α,25-(OH)₂D₃ to interfere with IL-12 production, after 7 days of culture with GM-CSF and IL-4 with or without 1α,25-(OH)₂D₃, DC were washed, seeded in the presence of maturation-inducing stimuli, and cultured for 3 days. Supernatants were quantified for IL-12 p70. IL-12 p70 production was significantly decreased when D₃-DC were exposed to TNF-α (53.5 vs 72.2 pg/0.5 × 10⁶ cells/ml; p = 0.05) or LPS (45.7 vs 79.8 pg/0.5 × 10⁶ cells/ml; p = 0.02), or CD40L (38.6 vs 481.7 pg/0.5 × 10⁶ cells/ml; p = 0.002; Fig. 6).

As T cell stimulation via TCR in the absence of a second signal by costimulatory molecules and/or secreted cytokines may induce a state of hyporesponsivity or anergy, we tested whether D₃-treated DC induced an alloantigen-specific tolerance. In these experiments allogeneic T cells were first cocultured with ctr-DC or D₃-DC and exposed to LPS for 3 days. Then cells were extensively washed, depleted of remaining DC by MR or CD2 using Immuno-Dynabeads (Unipath, Milan, Italy), and rested for 5 days. T cells (viability, >90%) were then restimulated in a second coculture with mature ctr-DC. T cells first cocultured with ctr-DC responded vigorously to restimulation with mature ctr-DC. In contrast, T cells first cocultured with D₃-DC were hyporesponsive to further stimulation with ctr-DC (Fig. 7,A). To determine whether this hyporesponsivity was alloantigen specific, the rescued T cells were restimulated with DC generated from an unrelated donor. In this case also, T cells showed a profound inhibition of proliferative capacity compared with T cells cocultured with untreated DC (Fig. 7 B). These results indicate that D₃-Dc exposed to LPS induce a state of hyporesponsivity in T cells, which was not alloantigen restricted. Of interest, when immature D₃-DC were used as stimulator in the first coculture, T cells were not inhibited (data not shown).

To evaluate the effects of 1α,25-(OH)₂D₃ on differentiated immature DC, monocytes were cultured for 7 days in the presence of GM-CSF and IL-4. Cells were then washed and incubated again with IL-4, GM-CSF, and 1α,25-(OH)₂D₃ for 3 or 7 additional days. Control cells were incubated with GM-CSF and IL-4 for the same period of time. Treatment with 1α,25-(OH)₂D₃ partially reversed DC differentiation, as demonstrated by a down-regulation of CD1a and an up-regulation of CD14; expression of CD80, CD86, and MHC I was not affected. In contrast, MHC II and CD40 were significantly down-regulated (Fig. 8). After exposure to LPS or CD40L, D₃-DC showed a lower expression of MHC I, MHC II, CD80, CD86, CD83, and CD40. A summary of four experiments is shown in Table II. Finally, we evaluated the influence of 1α,25-(OH)₂D₃ on the endocytic activity of immature DC and on the capacity to stimulate T lymphocytes in MLR. 1α,25-(OH)₂D₃ significantly increased the uptake of FITC-dextran and Ag uptake receptor expression, while the capacity to stimulate allogeneic T lymphocytes was decreased compared with that of untreated DC (data not shown). Overall, these results demonstrate that 1α,25-(OH)₂D₃ impaired the maturation of DC even when added tn already differentiated cells.

It is well known that the active form of vitamin D, 1α,25-(OH)₂D₃, modulates lymphocyte and macrophage functions (46). We demonstrated a new target of 1α,25-(OH)₂D₃ action on the immune system: DC. Because DC have the unique property to activate naive T cells and are required for the induction of a primary response (47), the suppression of DC function may very efficiently control the specific immune response (48). 1α,25-(OH)₂D₃ showed complex effects on DC. 1α,25-(OH)₂D₃ partially blocked the GM-CSF- and IL-4-driven differentiation of monocytes to DC. In fact, in the presence of 1α,25-(OH)₂D₃, despite a quite similar morphology, the expression of CD1a was inhibited and CD14 expression was increased, a marker of monocytes/macrophages normally not present on DC (49). In previous studies high expression of CD14 was found in monocyte-derived DC cultured in the presence of immunosuppressive factors such as glucocorticoid and IL-10 (50, 51). The intensity of CD14 expression was lower in vitamin D₃-derived DC than in glucocorticoid-derived or IL-10-derived DC, and CD14 was not seen with low concentration (5%) of FCS (data not shown) Despite the persistence of CD14, other markers of monocytes/macrophages, such as CD16 and CD68, were not present in vitamin D₃-treated DC. Therefore, we conclude that vitamin D₃ inhibits a full differentiation of monocytes into DC, but, unlike IL-10, does not promotes differentiation toward macrophages. In the past, a clear effect of D₃ has been shown on monocytes, bone marrow precursors, and monocytic leukemic cell lines on the differentiation toward macrophages, but under different culture conditions (28, 30, 52, 53). 1α,25-(OH)₂D₃ is also important in osteoclast generation, defining the commitment of monocytes differentiating into osteoclasts as a cooperative associative mechanism involving osteoblastic cells (54, 55). Monocytes may be considered relatively immature precursors with multiple differentiation potentials that depend upon the microenvironment (56). Our data showed that similar to cytokines such as M-CSF, GM-CSF, TGF-β, and IL-4, the hormone 1α,25-(OH)₂D₃ may play an important role in the final decision determining whether monocytes will acquire DC, macrophage, or osteoclastic characteristics and functions, in particular inhibiting DC differentiation.

1α,25-(OH)₂D₃-treated DC showed other important modifications in phenotype. Normally we can identify two major phases in the life of DC (40, 41, 57, 58): an immature stage characterized by a high efficiency in taking up and processing Ags associated with high expression of molecules involved in Ag uptake as MR, such as CD32; and a mature stage in which the Ag uptake capacity is lost, the cell migrates toward regional lymph nodes, and the function shifts to become a potent APC (59, 60) associated with a high expression of molecules involved in Ag presentation and T cell stimulation, such as MHC I, MHC II, CD80, CD86, CD40, and CD83. Exposure of differentiating monocytes to 1α,25-(OH)₂D₃ increased the expression of molecules involved in Ag capture (CD32, MR), while some important costimulatory molecules (CD86, CD40) were inhibited. This phenotype correlates with impaired Ag-presenting function for T lymphocytes and higher endocytic activity. Moreover, 1α,25-(OH)₂D₃ strongly inhibited DC maturation, as demonstrated by a low or absent increase in the expression of MHC I, MHC II, CD80, CD86, CD40, and CD83 and by the impaired stimulatory capacity for T lymphocytes after exposure to LPS or CD40L. Finally, as recently reported with already differentiated DC (61), D₃-DC showed impaired IL-12 production after CD40L, LPS, or TNF-α exposure. These results extend the immunosuppressive effects of this hormone and confirm a role for 1α,25-(OH)₂D₃ as a regulator of immune cell differentiation and function.

The effects of 1α,25-(OH)₂D₃ on immature DC that have been differentiated for 7 days in the presence of GM-CSF and IL-4 appear to be similar to, but not identical with, the effects of 1α,25-(OH)₂D₃ included at the beginning of the culture. Overall, the effects can be summarized as follows: a partial conversion to a monocyte/macrophage phenotype, an impaired capacity to reach maturation, and a decreased ability to stimulate T cells (the latter not shown). These results confirm the in vitro instability of immature DC generated with GM-CSF and IL-4 (56, 62). Palucka et al. (56) showed that upon removal of both GM-CSF and IL-4 and/or reculture with M-CSF, immature CD1a⁺/CD14⁻ DC easily converted to a macrophage phenotype expressing CD14 with a decreased ability to stimulate allogeneic T cells. Thus 1α,25-(OH)₂D₃ acts at two different steps of DC life: 1) inhibiting the differentiation from monocytic precursors and thus impairing the normal turnover of DC in tissues, and 2) inhibiting the terminal maturation of DC into a potent APC.

The inhibitory effect of 1α,25-(OH)₂D₃ on DC maturation and differentiation is very similar to that of IL-10, an anti-inflammatory cytokine, and to that of glucocorticoids. In fact, both IL-10 and glucocorticoids were shown to prevent monocyte differentiation and maturation to DC, to impair IL-12 production, and to increase Ag uptake (43, 50, 51).

Of interest is the fact that 1α,25-(OH)₂D₃-differentiated DC matured with LPS or CD40L induced hyporesponsivity of allogeneic T cells. T cells cocultured with D₃-DC showed impaired proliferation to a second stimulation with control DC from the same as well as from an unrelated donor. A direct effect of D₃ on T cells is excluded, as D₃-DC were extensively washed before coculture. Moreover, T cell viability before the second stimulation was >90%, and T cells were able to proliferate upon addition of exogenous IL-2 (data not shown). The induction of T cell anergy has been reported by Steinbrink et al. (63) with DC differentiated in the presence of IL-10, but several differences can be outlined, showing different mechanisms of action between IL-10 and D₃. In IL-10-treated DC the induction of anergy is associated only with immature DC and is alloantigen specific. In our work it was not alloantigen specific and was not observed when we used immature D₃-DC. It appeared at least in part linked to a soluble factor(s) secreted by D₃-DC exposed to LPS. In fact, supernatants from these cells, when added in a first coculture to control mature DC and allogeneic T cells, inhibited T cell proliferation, while supernatants from immature D₃-DC or mature ctr-DC did not. The generation of DC able to induce T cell hyporesponsivity might be a first step in the development of treatments for patients at risk of transplant rejection or with autoimmune or allergic diseases. The therapeutic use of these cells, however, requires further studies, as the hyporesponsivity was not specific, and putative factors involved in the induction of T cell hyporesponsivity remain to be defined.

The effects described in our work were seen in a range of 1α,25-(OH)₂D₃ concentrations from 5 × 10⁻¹¹ to 10⁻⁷. Importantly, calcitriol is effective at concentrations that are considered physiological (10⁻¹⁰–10⁻⁸ mol/l) (64) and that correspond to the accepted affinity value for its receptor (65). The physiological role of vitamin D in immune responses is not clearly defined. In vivo, both an excess and a deficiency of vitamin D suppress the delayed hypersensitivity response (66) or Ig production (67). Vitamin D-deficient animals and humans have a higher risk of infection, probably related to impaired macrophage function (68). The monocyte function as APC seem to be decreased (32). The NK cell activity is enhanced by 1α,25-(OH)₂D₃. This enhancing effect of the nonspecific immune defense contrasts with an inhibition of the Ag-specific immune response, as demonstrated by decreased T cell proliferation and activity (decreased IL-2, IFN-γ, GM-CSF synthesis and secretion). It is very difficult to clarify the endocrine activity of 1α,25-(OH)₂D₃ in the immune system. A possible paracrine or autocrine activity may be postulated. The secretory role of macrophages may be central to the production of localized concentration of 1α,25-(OH)₂D₃ within immune microenvironments. Normal macrophages have been shown to synthesize 1α,25-(OH)₂D₃ when activated by agents such as IFN-γ and LPS (1, 4). In granulomatous disorders such as sarcoidosis and tuberculosis, macrophages are able to produce 1α,25-(OH)₂D₃ and appear to be insensitive to feedback control by 1α,25-(OH)₂D₃ itself as well as other regulators, such as calcium and parathyroid hormone. It is tempting to speculate that 1α,25-(OH)₂D₃ produced by macrophages, by inhibiting differentiation and function of DC, may contribute to the peripheral anergy in sarcoidosis and to the persistence of granulomatous lesion in tuberculosis.

Another site of interest for 1α,25-(OH)₂D₃ action is the skin. It is likely that the major source of vitamin D for human is not dietary, but results from its manufacture by a chemical photolysis reaction in skin. Vitamin D₃ itself is a biologically inert molecule. It must be activated by 25-hydroxylation in the liver to produce the major circulating form of vitamin D, 25-hydroxyvitamin D₃. However, 25-hydroxyvitamin D₃ is also biologically inactive at physiological concentrations, and it is finally activated in the proximal convoluted tubule cells of the kidney to produce 1α,25-(OH)₂D₃ (8). Keratinocytes, the most important cells of the skin, possess both the 24- and 1α-hydroxylase enzymes and thus can produce small amount of 1α,25-(OH)₂D₃. Therefore, UV exposure induces a systemic and a local increase in 1α,25-(OH)₂D₃ in the skin. It is known that after UV exposure, Langerhans cells (epidermal CD1a⁺ cells) disappear from the healthy skin, and CD11b⁺ macrophage-like cells appear in few days (69). Moreover, UV radiation induces apoptosis and suppresses the immune function of epidermal Langerhans cells (70). Although other cytokines, such as IL-10, are known to play an important role in UV-induced immunosuppression (71), it is tempting to speculate that 1α,25-(OH)₂D₃ could also contribute to some of the modifications of Langerhans cells and could be responsible for the decrease in DC in skin after UV exposition.

In conclusion, our data suggest that 1α,25-(OH)₂D₃ may modulate the immune system, acting at the very first step of the immune response through the inhibition of DC differentiation and maturation into potent APC.

We thank Dr. Luciano Adorini (Roche, Milan, Italy) for stimulating discussion.