Low-dose oral cholecalciferol is associated with higher numbers of Helios + and total Tregs than oral calcitriol in renal allograft recipients: an observational study

In this retrospective study of renal allograft recipients, we tested a possible association between vitamin D supplementation with either cholecalciferol or calcitriol and the numbers of Tregs. We found that supplementation with cholecalciferol was associated with higher absolute numbers of Helios⁺ Tregs (CD4⁺CD25⁺CD127^?Foxp3⁺Helios⁺) than calcitriol or no vitamin D supplementation, and with higher absolute numbers of total Tregs (CD4⁺CD25⁺CD127^?Foxp3⁺) than with calcitriol supplementation. The absolute numbers of Treg subsets were similar in the cholecalciferol group to healthy controls. We did not find a significant difference in the numbers of Helios^? (CD4⁺CD25⁺CD127^?Foxp3⁺Helios^?) and IFNg⁺ (CD4⁺CD25⁺CD127^?Foxp3⁺IFNg⁺) Tregs among the 3 renal allograft patient groups.

We assessed absolute number of cells/?l in addition to the ratio of Tregs to CD4⁺ and CD8⁺ cells. Liu et al. demonstrated that the absolute numbers of Tregs, rather than the ratios of Tregs to peripheral lymphocytes, was associated with long-term survival of renal allografts [26].

Interestingly, as shown in Table 1, in our study cholecalciferol at an average dose of 7000 IU weekly was associated with higher numbers of Helios⁺ and total Tregs than calcitriol at an average dose of 1.75 ?g weekly. In vivo trials reported cholecalciferol supplementation at doses of 140,000 IU monthly or 20,000 IU daily, and 0.5 ?g of daily calcitriol in another study [17, 20, 21, 27]. The average dose of cholecalciferol supplemented in our study was thus about 5–20 times lower than the dosages prescribed in these trials, whereas the average dose of calcitriol was about 3 times lower than that administered in a trial conducted by Ardalan et al. [17]. Our results show that cholecalciferol has an effect on Tregs in transplant patients even at relatively low doses.

Although calcitriol can directly affect T cells in vitro, the required doses are much higher than the physiological doses as demonstrated in many studies [16, 28–31]. In addition, it was demonstrated in healthy persons as well as uremic patients that the bioavailability of 1, 25(OH)₂ D₃ was 70 % of the supplemented dose of oral calcitriol [32]. De Sévaux et al. showed that calcitriol supplementation in a daily dose of 0.25 ?g did not improve the serum 25(OH) D₃ level in renal transplant patients [33]. Consistent with this finding, Marcen et al. demonstrated that calcitriol in a daily dose of 0.25–0.5 ?g failed to improve vitamin D deficiency in a cohort of renal transplant patients [34]. In contrast, low dose oral cholecalciferol in a weekly dose of 5000 IU for 15 weeks increased serum 25(OH) D₃ from 18.4?±?8.2 to 68.6?±?17.7 nmol/l in a cohort of 34 hemodialysis patients without causing any episode of hypercalcemia [35]. Moreover, oral cholecalciferol showed good long-term 25(OH) D₃ and 1, 25 (OH)₂ D₃ bioavailability, even after 3 months of supplementation with a single dose of 600,000 IU, with a maximal effect at one month [36]. Based on these findings, we speculate that the calcitriol doses supplemented in our study were much lower than the doses required for induction of Tregs in patients, in contrast to cholecalciferol, which can improve vitamin D deficiency even at low doses.

Our study challenges the findings of Ardalan et al., who reported that calcitriol administered to donors 5 days before transplantation at a dose of 0.5 ?g and continued at the same dose in the recipients for 1 month after transplantation and thereafter at a dose of 0.25 ?g for another 5 months, was associated with a significant increase in the numbers of CD4⁺CD25⁺ cells [17]. This finding is surprising since the doses administered seem too low for the induction of a significant increase of Tregs in the light of other studies. The most likely explanation of the discrepancy is that the Treg cell population of our study was defined as CD4⁺ CD25⁺ Foxp3⁺ CD127 ^low/-, whereas in Ardalan’s trial Tregs were defined only as CD3⁺CD4⁺CD25⁺ T cells. It was shown that a subpopulation of Tregs, termed IL-10-Tregs, expressed CD4 and CD25, whereas these cells lacked the expression of Foxp3 [37]. The marker combination CD4⁺ CD25⁺ IL10⁺ characterizes a Treg subset termed Tr1 cells. Because only the CD4⁺ T cell subset, which expresses the highest level of CD25 (CD25 ^high), has a suppressive effect, it is likely that the cell population studied by Ardalan et al. was a mixture of Tr1 cells in addition to both CD4⁺ CD25⁺ effector cells and conventional Tregs [38, 39]. Moreover, serum 1, 25 (OH)₂ D₃ was not measured before and six months after transplantation. This would have provided an idea about the effects of the supplemented doses on serum 1, 25 (OH)₂ D₃. To test whether the higher doses of calcitriol administered in Ardalan’s study were responsible for the increase in the numbers of Tregs, we compared the patients who were supplemented with comparable or higher doses of calcitriol as the patients in Ardalan’s trial with the other two groups of transplanted patients in our study. The higher- dose calcitriol group received an average dose of 3.5 ?g weekly. We found that the Tregs were still about two-folds higher in the cholecalciferol arm. This finding suggests that cholecalciferol is superior to calcitriol even when the latter is prescribed at higher doses.

Although cholecalciferol was associated with higher numbers of Helios⁺ Tregs (CD4⁺CD25⁺CD127^?Foxp3⁺Helios⁺), we could not detect a significant difference in the numbers of Helios^? Tregs (CD4⁺CD25⁺CD127^?Foxp3⁺Helios^–) among cholecalciferol, calcitriol, no vitamin D supplementation, and control groups. Since Helios^? Tregs represent a mixture of tTregs (Bona fide Tregs) and a majority of Tregs activated when exposed to antigens (pTregs) [4], we speculate that cholecalciferol might have affected only the bona fide tTregs rather than peripherally activated Tregs. We think that even if cholecalciferol caused an increase in the numbers of bona fide Helios^? Tregs, the increase might have been too small to be statistically significant, considering that the majority of Helios^? Tregs were pTregs and that the sample size was small. If this hypothesis proves to be true, the important question is how such relatively low doses of cholecalciferol can affect the bona fide Tregs (either Helios⁺ or Helios^?) while not affecting the Helios^? pTregs. This may be attributed to the vitamin D-binding protein (DBP). A likely explanation is that the Helios^? bona fide tTregs are induced in the thymus where DBP concentrations are much lower than in the serum [40, 41]. The lower concentration of DBP in the thymus renders relatively lower concentrations of cholecalciferol capable of inducing Helios^? bona fide tTregs. As Helios^? pTregs are induced in the periphery, where the concentrations of DBP are much higher, higher concentrations of 25 (OH) D₃ may be required to induce pTregs. T cells express CYP27B1, which is a 1?-hydroxylase, and have the capacity to convert 25(OH) D₃ to 1, 25(OH)₂ D₃ in sufficient concentrations to affect the vitamin D-responsive genes [22].

To find out whether the increase in Treg subset numbers associated with cholecalciferol supplementation was real, we compared cholecalciferol patients with healthy controls. Interestingly, both cholecalciferol groups showed similar Treg subset numbers, whereas calcitriol and no vitamin D supplementation groups showed lower Treg numbers. Accordingly, it appears that cholecalciferol has a stabilizing effect on Treg (particularly Helios⁺ subset) rather than a proliferation-inducing one.

CTLA-4 is a marker associated with the suppressive capacity of Tregs. Cholecalciferol was associated with significantly higher numbers of these cells in comparison to calcitriol and there was a trend of higher numbers also in the no vitamin D supplementation group, whereas the numbers were comparable to healthy controls. We tested the correlation between CTLA-4⁺ Tregs (CD4⁺CD25⁺CD127^?Foxp3⁺CD152⁺) and Helios⁺ Tregs (CD4⁺CD25⁺CD127^?Foxp3⁺Helios⁺) in the four groups to estimate whether Helios⁺ Tregs co-express CTLA-4 and use CTLA-4 for cell-cell contact suppression. Helios⁺ Tregs were associated with CTLA-4 in all 4 groups suggesting that they use CTLA-4 for suppression.

Immunosuppressive drugs have variable effects on Tregs. Cyclosporine has long been known to inhibit the activation of T cells through suppression of calcium-dependent phosphatase calcineurin leading to suppression of IL-2 synthesis. Melony et al. showed that calcineurin inhibitors (CNI) led to expansion of the Treg population in lung allograft recipients [42]. Intriguingly, Ruppert et al. showed that Tregs resisted apoptosis caused by cyclosporine through expression of CD44 [43]. Kogina et al. reported that tacrolimus suppressed T cell receptor-mediated cell division of conventional T cells (CD4⁺ T cells), whereas it enhanced division of Tregs in vitro [44]. In contrast, other studies showed a harmful effect of CNI on Tregs [45, 46]. In our study, it is unlikely that calcineurin inhibitors were the cause of the increased Treg numbers in the cholecalciferol group as we found no significant difference in CNI prescription among the three patient groups. Mycophenolic acid is one of the most widely used drugs in solid organ transplantation. It exerts its function through inhibition of inosine monophosphate dehydrogenase leading ultimately to B and T cell suppression. Recently, Scotta et al. showed that administration of methylprednisolone, tacrolimus and mycophenolic acid suppressed the viability and proliferation of Tregs. They have also showed that in vivo administration of sirolimus; an inhibitor of the mechanistic target of rapamycin, maintained proliferation and survival of adoptively transferred Tregs [47]. A recent in vitro study published by our group showed variable effects of the immunosuppressive agents on IFNg⁺ and total Tregs [48].