The increase in total kidney volume (TKV) is a prognostic biomarker of decreased renal function in Autosomal Dominant Polycystic Kidney Disease (ADPKD) (1–3). Hypertension, which occurs prior to loss of kidney function in 60% of ADPKD patients, represents a significant independent risk factor for progression of the disease, contributing to cyst expansion and intrarenal ischemia hence activating intrarenal renin-angiotensin-system (RAS) (4–6). Besides genetic factors, predictors that may lead to cyst growth and increase in TKV in ADPKD also include male gender, high salt and protein intake, caffeine consumption, level of fluid intake, gross hematuria, nephrolithiasis, proteinuria among others (7–13). Inflammation, present even in initial stages of the disease may also influence the progression of ADPKD (14). Vitamin D₃ (cholecalciferol) is synthesized in human skin from the conversion of 7,8-dehydrocholesterol by UV radiation and then hydroxylated in the liver to give origin to the circulating form 25-hydroxyvitamin D₃ [25(OH)D or calcidiol]. The hormonal active metabolite of vitamin D, 1,25-dihydroxyvitamin D₃ [1,25(OH)₂D or calcitriol], is further produced in the kidneys though the enzyme CYP27B1 (1-α hydroxylase) and functions as the ligand for the nuclear vitamin D receptor (VDR) (15). Both forms of vitamin D are catabolized by CYP24A1. Vitamin D regulates the expression of pro-inflammatory genes and might act as an anti-inflammatory hormone (16). In an experimental non-orthologous model of PKD, cholecalciferol supplementation was capable to reduce proteinuria and interstitial inflammation (17). Nevertheless, few human studies focused on the relationship of vitamin D with BP and inflammation in ADPKD (18) and its impact on predictors of disease progression remains unknown. We aimed to investigate vitamin D status, serum levels of 1,25 (OH)₂D, the expression of regulatory enzymes [CYP24A1 and CYP27B1] and VDR and their association with BP, inflammatory markers and TKV in ADPKD patients.

Ninety eligible participants with ADPKD confirmed by positive family history and renal cysts according to ultrasonographic diagnostic criteria by Pei et al. (19), were recruited from the outpatient Polycystic Kidney Disease Unit of the Universidade Federal de São Paulo, since August 2011 through May 2016, to participate in this study. Exclusion criteria were age <18 years old, serum calcium >10.5 mg/dL and current/past use of calcium or vitamin D. The reason for excluding users of vitamin D supplements or hypercalcemic patients relied on the necessity of vitamin D supplementation foreseen for the patients exhibiting hypovitaminosis D, as a further planned intervention study, which is still ongoing. Hypertension was defined by either measurements upon enrollment, history of hypertension or use of antihypertensive medications. Body weight, height and waist circumference were obtained and Body mass index (BMI) calculated. Body fat composition was assessed by bioelectrical impedance analysis (BIA 101 Quantum, RJL Systems, Detroit, MI). Consumption of macronutrients, vitamin D, calcium, phosphorus and caffeine were assessed through a 24-h dietary recall and daily intakes were calculated as previously described (10). After this initial clinical evaluation, enrolled patients were scheduled for one blood sample drawn following an overnight fast and a 24-h urine collection, obtained during the preceding day. Subsequently, patients were scheduled to undertake a 24-h ambulatory blood pressure monitoring (ABPM) and a magnetic resonance imaging (MRI) scan to determine total kidney volume (TKV). Patients were divided in vitamin D-sufficient (>30 ng/mL) and vitamin D-insufficient (<30 ng/mL) groups for comparisons. The study was reviewed and approved by the Ethics Advisory Committee of the Universidade Federal de São Paulo, and each patient signed the informed consent form.

Vitamin D pleiotropic effects beyond mineral metabolism, such as potential reduction of arterial BP and inflammation (24), may be of relevance inADPKD (6, 14, 25). Increased inflammatory markers in serum, urine, and fluids of cysts have been detected in ADPKD patients (14, 26, 27). Preliminary data by Gitomer et al. (28) have suggested an inverse association between serum vitamin D and kidney volume in ADPKD but to the best of our knowledge, this is the first study to evaluate the association between vitamin D, its regulatory proteins and expression of VDR with predictors of ADPKD progression.

Vitamin D insufficiency was present in 62% and CYP24A1 was overexpressed in this group. Serum 25(OH)D and VDR expression were negatively correlated with htTKV, as was VDR with inflammatory markers.

Hypovitaminosis D was more prevalent for patients who had their blood samples drawn during winter, as already evidenced in our country (29, 30). Some studies have reported an association between hypovitaminosis D with BMI (29, 31) possibly due to either the low exposure of obese individuals to sunlight or sequestration and storage of vitamin D in adipose tissue (32). We did not observe a higher BMI, waist circumference or percentage of body fat in vitD-insufficient patients, differing from previous studies by our group and others in CKD or after renal transplant (29, 31), but corroborating with Gronborg et al. (33), who found no association. Among potentially modifiable factors affecting progression of ADPKD (8), daily intakes of protein, NaCl, caffeine, calcium, phosphorous, and vitamin D did not differ between groups with or without hypovitaminosis D but both presented an intake of vitamin D under the recommended allowance (600 IU/day) and of NaCl, three-fold higher than the recommended by the American Heart Association (4 g/day). Caffeine intake was low, as already observed by our group in a previous evaluation (10).

The percentage of hypertensive patients and 24-h ABPM parameters did not differ between vitD-insufficient vs. sufficient groups diverging from epidemiological data in general population using office BP measurements (34). The employment of 24-h ABPM in the current series reinforces the reliability of our findings, who agreed with other studies not showing an association (35, 36). On the other hand, it is possible that the anti-hypertensive treatment with ARB and/or ACEi by 72% of our patients could have contributed to the lack of association between vitamin D and BP. However, even under anti-hypertensive therapy, 31/70 (44%) of patients from the current series presented a non-dipping pattern, in accordance with several investigators (19, 37, 38), who have detected it even in otherwise normotensive ADPKD subjects (38), as an early manifestation of endothelial dysfunction (39). We observed a high percentage of albuminuria (48%), particularly among hypertensive patients, corroborating with data from Chapman et al. (40), but without association with hypovitaminosis D. The use of ARB and/or ACEi might have also accounted for by the absence of such association. As cAMP accumulation plays a central role in cystogenesis (41), we determined urine levels of cAMP in the present series, but no statistical difference has been detected between vitD-insufficient or sufficient groups. The median level of eGFR and the percentage of CKD patients (stages 3/4) distributed among vitD-sufficient (26%) and insufficient groups (41%) were not statistically different (p = 0.716), what rendered more adequate the comparison of all parameters between groups.

Serum 25(OH)D was negatively correlated with PTH, as expected (42), although median PTH did not differ between vitD-sufficient and insufficient groups. Surprisingly, the latter presented a significantly higher median 1,25(OH)₂D compared to the sufficient group, and a negative correlation between 25(OH)D with 1,25(OH)₂D in the whole sample was disclosed. These unforeseen findings might have been attributed, as suggested by Need et al. (43), to a biphasic relationship between calcidiol and calcitriol depending on the level of 25(OH)D: positive whenever it is in the normal range (attributed to substrate deficiency), but negative when it is low, due to secondary hyperparathyroidism. Although inflammatory parameters were similar among vitD-insufficient and sufficient patients, an inverse association of VDR expression with IL-6, IL-10, NFκB, and CRP was observed, supporting that VDR is directly involved in the regulation of inflammatory response (16). A chronic inflammatory milieu is observed in cystic PKD kidneys, as evidenced by the large numbers of interstitial macrophages ultimately promoting cyst epithelial cell proliferation, cyst expansion, and disease progression (44). Cultured Pkd1-deficient cells express the monocyte chemoattractant protein-1 (MCP-1) and CXCL16 (C-X-C Motif Chemokine Ligand 16) and large numbers of activated macrophages surrounding the cysts have been observed in orthologous models of PKD (45). Peda et al. (44) have demonstrated that cystic epithelial cells induce renal M2-like macrophage polarization which in turn enhances the ability to promote cyst cell proliferation. IL-10 was shown to be upregulated in human ADPKD tissue and present in cyst fluid, and although this regulatory cytokine has anti-inflammatory functions, M2-phenotype requires IL-10 secretion by the macrophages and IL-10-stimulated activation of STAT3 is required for this pathological macrophage differentiation (44). M2-like macrophages have been identified in interstitial areas juxtaposed to cysts in human ADPKD kidneys, potentially promoting cyst growth by stimulating nearby cyst lining epithelial cells (46). Recently, lysine methyltranferase SMYD2 was shown to increase cystic renal cell proliferation through methylation and activation of STAT3 and the p65 subunit of NFκB (47). VDR can form a complex with the p65 subunit of NFκB to produce anti-inflammatory actions (24) and vitamin D supplementation has already been reported to help reducing circulating levels of IL-6 in other populations such as end-stage renal disease patients (21). The inverse correlation between VDR with IL-10 and NFκB observed in the present series, suggest that targeting reduction of inflammation with vitamin D or other VDR-stimulating agents may represent an effective strategy for slowing PKD progression and further studies are needed to test this hypothesis. We cannot exclude the possibility of vitamin D deficiency being a consequence rather than the cause of inflammatory response, as excess of 1,25(OH)₂D is produced in an effort to up-regulate the VDR and 25(OH) is rapidly metabolized in this process (48). D. In the present series, median serum calcium, phosphorus, alkaline phosphatase did not differ between vitD-sufficient vs. insufficient patients. Median FGF-23 in the whole sample, composed mostly of ADPKD patients at CKD stages 1 and 2, was high if compared to CKD caused by other conditions (49, 50), being in agreement with Pavik et al. (51) who observed a four-fold increase in FGF-23 from ADPKD compared to CKD of other etiologies with similar eGFR, although no association with 25(OH) and 1,25 (OH)₂D was detected. Accordingly, in the present sample, FGF-23 was not statistically different between groups with or without hypovitaminosis D, nor correlated with 25(OH)D or 1,25(OH)₂D. Since both groups had been equally exposed to the use of ACEi (61 vs. 59%), the latter could have acted as a confounder, negatively regulating FGF-23 levels and disrupting the cross-talk between vitamin D and RAS (24). Moreover, as shown by Spichtig et al. (52), FGF-23 is detected in cells lining renal cysts of PKD animals but it fails to appreciably downregulate CYP27B1 due to resistance.

With regard to vitamin D regulatory enzymes, the vitD-insufficient D group presented a higher median expression of CYP24A1, which in turn correlated negatively with 25(OH)D, raising the hypothesis of vitamin D catabolism. As expected, we found a positive correlation between CYP27B1 with both 1,25(OH)₂D and VDR. In line with the data by Spichtig et al. (52) correlations between FGF-23 and CYP24A1 have not been disclosed in the current study as well (data not shown).

Finally, the negative correlation between both 25(OH)D levels and VDR expression with htTKV disclosed in the present study, suggested a potential contributionof hypovitaminosis D to kidney enlargement in ADPKD. However, the results of the multivariate regression analysis with htTKV as the dependent variable, including hypertension, CRP, eGFR, age, time since diagnosis, VDR and 25(OH)D adjusted for season of the year showed that only the first three parameters were independent predictors of the former.

Limitations of the present study included its cross-sectional design and the need of concomitant use of anti-hypertensive therapy, which could have contributed, at least in part, to the negative results obtained with respect to blood pressure. The high proportion of hypovitaminosis D obtained in samples collected during the winter might have influenced our results. Therefore, an adjustment of the serum 25(OH)D results according to the season of the year in the multiple regression analysis was performed to take into account this confounded limitation. On the other hand, our study also has strengths such as the employment of 24-h ABPM for BP measurements.

In conclusion, present findings suggested that low levels of serum 25(OH)D and VDR expression are associated with a higher kidney volume in ADPKD patients, but hypovitaminosis D does not represent an independent risk factor for increasing kidney volume.

The study was reviewed and approved by the Ethics Advisory Committee of the Universidade Federal de São Paulo, and each patient signed the informed consent form in accordance with the Declaration of Helsinki.

LV designed the study, performed the clinical study and the experiments, analyzed the data, and drafted the manuscript. IH designed the study, analyzed the data, provided intellectual content of critical importance to the work described, and revised the manuscript. MD, JdC, and JN participated in the analysis of data. MB helped in the conception and analysis of data.