The CYP24A1 Gene Variant rs2762943 Is Associated With Low Serum 1,25- Dihydroxyvitamin D Levels In Multiple Sclerosis Patients


 Background: Vitamin D is considered to play a role in multiple sclerosis (MS) etiopathogenesis. We recently identified a polymorphism located in the cytochrome P450 family 24 subfamily A member 1 (CYP24A1) gene, rs2762943, that was found to be associated with an increased risk for MS. CYP24A1 codes for a protein that is involved in the catabolism of the active form of vitamin D. Here, we investigated the immunological effects of carrying the risk allele for the rs2762943 polymorphism, as well as its role as genetic modifier in MS patients. Methods: Serum levels of 25‐hydroxyvitamin D (25OHD) and 1,25-dihydroxyvitamin D (1,25(OH)2D) were measured in a cohort of 167 MS patients. In a subgroup of these patients, expression levels of MHC class II and co-stimulatory molecules were determined by flow cytometry in blood cell populations, and the levels of proinflammatory (IFNG, GM-CSF, CXCL13) and anti-inflammatory (IL-10) cytokines and neurofilament light chain were measured by single-molecule array assays in serum samples. The effect of the rs2762943 polymorphism on disease activity and disability progression measures was evaluated in a cohort of 340 MS patients. Results: Compared to non-carriers, MS patients carrying the risk allele for rs2762943 were characterized by reduced levels of 1,25(OH)2D (p=0.0001), and elevated levels of IFNG (p=0.03) and GM-CSF (p=0.008), whereas no significant differences were observed between risk allele carriers and non-carriers groups for the other evaluated markers. The presence of the risk allele for rs2762943 had no significant impact on the annualized relapse rate, EDSS and MSSS measures during follow-up. However, risk allele carriers were younger at disease onset (p=0.04). Discussion: These findings suggest that the CYP24A1 rs2762943 gene variant plays a more important role on MS susceptibility than on disease prognosis, and is associated with lower 1,25(OH)2D levels and heightened pro-inflammatory environment in MS patients.


Introduction
Multiple sclerosis (MS) is a complex immune-mediated disorder of the central nervous system (CNS) wherein both a polygenic background and environmental factors contribute not only to MS risk but also to disease activity [1][2][3][4][5]. One of the environmental factors involved in MS etiopathogenesis is vitamin D status. In recent years, numerous studies have revealed that lower serum 25-hydroxyvitamin D (25OHD) levels are associated with an increased risk for MS [4,5]. Furthermore, vitamin D status is also associated with MS disease activity, and patients with lower serum 25OHD levels were shown to have higher disease activity, though ndings seem to be stronger for radiological rather than clinical outcomes [6]. Similarly, vitamin D supplementation in relapsing-remitting MS (RRMS) patients signi cantly improved the development of new magnetic resonance imaging lesions but had no signi cant effects on the annualized relapse rate or disability progression [7].
In a recent study, by means of targeted resequencing in 524 MS patients and 546 healthy controls and subsequent genotyping in an independent cohort of 3450 MS patients and 1688 healthy controls, we identi ed a variant located in the cytochrome P450 family 24 subfamily A member 1 (CYP24A1) gene, rs2762943, that was associated with an increased risk for MS [8]. The CYP24A1 gene encodes a protein that initiates the degradation of the physiologically active form of vitamin D [1,25-dihydroxyvitamin D -1,25(OH) 2 D]. Based on these ndings on CYP24A1, a gene involved in vitamin D metabolism, and considering the role of vitamin D in MS etiopathogenesis, as outlined above, in the present study we aimed to investigate the functional immunological effects and clinical consequences of carrying the rs2762943 risk allele in MS patients.

Materials And Methods
Quanti cation of serum levels of 25OHD, 1,25(OH) 2 D and calcium by chemiluminescence immunoassays.
A schematic ow chart summarizing the main steps performed in study design is represented in Fig. 1. Serum 25OHD, 1,25(OH) 2 D and calcium levels were measured in 167 patients with relapse-onset MS (RRMS and secondary progressive MS -SPMS). Of these, 44 (35.8%) patients were carriers of the risk minor allele for the rs2762943 polymorphism of the CYP24A1 gene (GT/TT) and 123 (64.2%) patients were non-carriers (GG). These patients were part of a full cohort of 340 MS patients with available serum samples to measure vitamin D and calcium levels. None of the 167 MS patients were receiving disease modifying therapies (DMT) at the time of blood collection and had never received vitamin D supplements. Table 1 shows a summary of demographic and main clinical characteristics of these patients.
Brie y, peripheral blood was collected by using standard venipuncture and allowed to clot spontaneously for 30 min. Serum was isolated by centrifugation and stored frozen at -80°C until used. Serum levels of 25OHD, 1,25(OH) 2 D and calcium were measured by commercially available chemiluminescence immunoassays (CLIA) according to the manufacturers' recommendations: Liaison 25 OH Vitamin D Total Assay (DiaSorin, USA) and Liason XL 1,25 Dihydroxyvitamin D (DiaSorin, USA), and a colorimetric method for Calcium Arseanzo (Beckman Coulter, USA). Assays were run on a fully automated Liason XL analyzer (DiaSorin) for both 25OHD and 1,25(OH) 2 D levels, and the AU5800 analyzer (Beckman Coulter) for calcium levels.
Immunophenotyping of peripheral blood cells and ow cytometric analysis.
Quanti cation of serum biomarker levels by single-molecule array assays.
Serum levels of C-X-C motif chemokine ligand 13 (CXCL13), interferon gamma (IFNG), interleukin 10 (IL-10) and granulocyte-macrophage colony-stimulating factor (GM-CSF) were measured in 20 MS patients, 10 risk allele carriers for the rs2762943 polymorphism and 10 non-carriers (Table S1). Serum neuro lament light chain (NfL) levels were measured in 49 MS patients, 26 risk allele carriers and 23 noncarriers (Table S1). Biomarker levels were quanti ed in the same serum samples used to determine the vitamin D status using commercially available immunoassay kits (CXCL13 cat#102635, IFNG cat#103337, IL-10 cat#101643, GM-CSF cat#102329, and NFL cat#103186, Quanterix) and run on the fully automated ultrasensitive Simoa HD-1 Analyzer (Quanterix). Samples were run in duplicate in accordance with manufacturers' instructions with appropriate standards and internal controls. Both the mean intra-assay coe cient of variation of duplicates and the mean inter-assay coe cient of variation were <11% for all assays. CYP24A1 rs2762943 polymorphism and disease activity and disability progression measures.
The role of rs2762943 as genetic modi er of MS was evaluated in the full cohort of 340 patients. Of these, 72 (21.2%) patients were risk allele carriers and 268 (78.8%) were non-carriers (Table 2).
In addition to demographic and main clinical characteristics such as sex, disease phenotype, disease duration, and percentage of MS patients receiving DMT, the following disability variables were recorded: Expanded Disability Status Scale (EDSS) and Multiple Sclerosis Severity Score (MSSS) at 10 years of disease duration and at year 2015. The annualized relapse rate (ARR) at 10 years from disease onset was recorded as in ammatory disease activity variable. We considered year 2015 as the maximum date of follow-up ("last follow-up time") to ensure that none of the MS patients had ever received vitamin D supplements. Based on the clinical evidence existing about vitamin D as predictor of disease activity and progression [9], from year 2016 serum vitamin D levels are routinely measured in our setting, and vitamin D supplementation is indicated for those MS patients with low vitamin D levels. Age at disease onset age was de ned as the age at which patients rst experienced neurologic symptoms suggestive of MS.

Statistical analysis.
Serum levels of 25OHD, 1,25(OH) 2 D, calcium, cytokines and NfL, and expression of HLA class II and costimulatory molecules by blood cell subsets were compared between risk allele carriers and non-carriers using the Mann-Whitney U nonparametric test. The effect of carrying the risk allele in age at disease onset, in ammatory disease activity and disability measures was evaluated using appropriate parametric and nonparametric tests depending on the normality of data. Analysis was performed in the whole MS population and in patients strati ed according to the MS phenotype. All statistical analyses were performed using the Statistical Package for Social Sciences (IBM Corp. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.). P-values <0.05 were considered statistically signi cant.

Results
CYP24A1 rs2762943 polymorphism is associated with MS and vitamin D levels.
We perused the most recent large-scale genetic study of MS susceptibility to test the association of the CYP24A1 rs2762943 gene variant [3]. The variant had an odds ratio (OR) of 1.47 for the minor allele T that reached genome-wide level of statistical signi cance at the study's discovery phase (p=3.9x10 -8 ; Table S2). Rs2762943 was located underneath the association peak of a nearby variant also located in the CYP24A1 gene, rs2248137 (OR=1.12; p=7.8x10 -11 ; Fig. S1). After controlling for the top variant, association of rs2762943 with MS remained nominally signi cant (OR=1.28; p=1.2x10 -3 ; Table S2). The two variants have a large allele frequency difference, but the two risk alleles are part of the same haplotype (D'=0.9713; Table S3), suggesting a putative common contribution to MS susceptibility.
Next, we searched the GWAS catalog to identify any association of the CYP24A1 rs2762943 polymorphism with other phenotypes and traits [10]. We observed genome-wide associations with metabolites and metabolite biomarkers (p<5x10 -8 ), including vitamin D levels (Table S4), altogether suggesting that the variant is interrupting key metabolic mechanisms. Based on the abovementioned association between rs2762943 and vitamin D related traits, we rst investigated whether the presence of the risk allele for the CYP24A1 rs2762943 polymorphism was associated with changes in the serum levels of 25OHD and 1,25(OH) 2 D in our cohort of 167 MS patients .
As shown in Fig. 2a, the vast majority of MS patients (98.2%) had low serum 25OHD levels (<30 ng/ml) but no signi cant differences were observed between risk allele carriers (GT/TT) and non-carriers (GG) for rs2762943. Only 4 patients (2.4%) had serum 1,25(OH) 2 D levels below the normal limit of normality (19.9 pg/ml; Fig. 2a). However, in contrast to 25OHD, MS patients carrying the risk allele for rs2762943 had signi cantly lower serum 1,25(OH) 2 D levels compared to non-carriers (p=0.0001). These ndings in the vitamin D status of MS patients were not associated with changes in total calcium concentration and most of the patients had serum calcium levels within the intervals of reference (8.8 -10.6 mg/dl; Fig. 2b). As depicted in Fig. 2c, differences in serum 1,25(OH) 2 D levels between risk allele carriers and non-carriers were maintained irrespective of the season of blood collection. For serum 25OHD levels, seasonal strati cation did not result in statistically signi cant differences between risk allele carriers and non-carriers, although a trend towards lower 25OHD levels in risk allele carriers was observed (p=0.06) in blood samples collected in cold season.
Expression of the MHC class II and co-stimulatory markers in blood cell populations is similar between risk allele carriers and non-carriers.
We next investigated whether risk allele carriers for the rs2762943 polymorphism differed from noncarriers in terms of steady state myeloid and lymphoid cell activation pro les. For this, we assessed the expression levels of MHC class II and co-stimulatory molecules on classical monocytes and B cells. As shown in Fig. 3, the presence of the risk allele was not associated with changes in the expression levels for HLA-DR, CD40, CD80, and CD86. Similarly, in T cells, the risk allele was not associated with changes in the expression levels of HLA-DR.
Serum levels of the proin ammatory cytokines IFNG and GM-CSF are increased in risk allele carriers.
As a next step, we evaluated whether the presence of the risk allele for the rs2762943 polymorphism was associated with changes in the levels of pro-in ammatory (CXCL13, IFNG, GM-CSF) and antiin ammatory (IL-10) cytokines. As shown in Fig. 4a, serum levels of IFNG and GM-CSF were signi cantly higher in risk carriers compared to non-carriers (p=0.03 and p=0.008 respectively), whereas serum levels of CXCL13 and IL-10 were similarly distributed between both groups of patients.
Presence of the CYP24A1 rs2762943 risk allele has no impact on MS disease activity and disability progression measures but in uences age at disease onset.
Main demographic and clinical characteristics such as sex, MS phenotype, disease duration and percentage of patients who received DMT at any time during follow-up were comparable between risk allele carriers and non-carriers (Table 2).
When the rs2762943 polymorphism was evaluated as genetic modi er of MS, we observed that risk allele carriers were younger at disease onset compared to non-carriers (p=0.04; Table 3). Comparison of disability measures such as EDSS and MSSS both at 10 years from disease onset and at last follow-up did not reveal signi cant differences between risk allele carriers and non-carriers (Table 3). Similarly, ARR at 10 years from disease onset was comparable between both groups of patients (Table 3). Further strati cation of the whole MS group into disease phenotypes (RR, SP, and primary progressive) was not associated with signi cant differences for age at disease onset, in ammatory disease activity and disability measures between risk allele carriers and non-carriers (Table S5).
Based on the ndings of increased levels of pro-in ammatory cytokines in risk allele carriers, and in view of the strong associations reported in the literature between NfL and radiological measures of disease activity such as the number of contrast-enhancing lesions and number of T2 lesions [11], serum levels of NfL were also measured in a subgroup of MS patients as a proxy of CNS radiological in ammation. Fig.  4b shows the distribution of serum NfL levels in risk allele carriers and non-carriers, which were comparable among both groups of patients.

Discussion
In the present study, we investigated the functional immunological consequences of carrying the risk allele for the rs2762943 polymorphism in the CYP24A1 gene, as well as its role as genetic modi er in MS patients. This gene variant, which is located 672 base pairs upstream from the transcriptional start site of the CYP24A1 gene [10], was found to be associated with an increased risk for MS in a recent targeted resequencing study carried out by our group [8] and also in the largest GWAS up to now conducted in MS [2]. Interestingly, rs2762943 remains as an independent signal for association with MS and is linked to vitamin D related traits [10].
CYP24A1 encodes a mitochondrial P450 enzyme that can hydroxylate both 25OHD and 1,25(OH) 2 D producing the inactive metabolites 24,25(OH) 2 D and 1,24,25(OH) 3 D, respectively [12,13]. In consequence, CYP24A1 limits the amount of the active form of vitamin D in target tissues by speeding up its catabolism and also by reducing the pool of 25OHD available for 1 α-hydroxylation in the kidney [12,13]. Although it can hydroxylate both, 1,25(OH) 2 D is the preferred substrate for CYP24A1 [14]. In our study, MS patients carrying the risk allele for rs2762943 (T) had signi cantly reduced serum levels of 1,25(OH) 2 D compared to non-allele carriers. Of note, the lower 1,25(OH) 2 D levels in risk allele carriers were not associated with signi cantly reduced serum 25OHD levels in these patients, despite that 25OHD levels were overall low in most MS patients. Although not proved, these ndings may be consistent with an increased enzymatic activity for CYP24A1 that results in an accelerated catabolism of the active form of vitamin D in those MS patients carrying the risk allele for rs2762943.
Although the lower 1,25(OH) 2 D levels observed in risk allele carriers were not of su cient magnitude to alter the calcium concentration, insomuch as serum calcium levels were within the normal range and similar to non-carriers, we further explored the potential immunological effects associated with reduced serum 1,25(OH) 2 D levels in these patients. Investigation of the expression levels for HLA class II and costimulatory molecules such as CD40, CD80, and CD86 in the major PBMC populations did not reveal signi cant differences between risk allele carriers and non-carriers. In contrast, levels of the proin ammatory cytokines IFNG and GM-CSF were found to be signi cantly elevated in carriers of the risk allele for rs2762943. Considering that vitamin D has immunomodulatory properties and has shown to suppress Th1 and Th17 responses [15,16], the ndings of increased IFNG and GM-CSF levels in MS patients carrying the risk allele may be associated with heightened Th1 and Th17 immune responses as a result of the lower blood 1,25(OH) 2 D levels observed in these patients.
In a second part of the study, we evaluated whether the presence of the risk allele for the rs2762943 polymorphism in the CYP24A1 gene could act as a genetic modi er in MS. Low vitamin D is considered a moderate risk factor for MS susceptibility based on a number of observational studies demonstrating an association between low serum 25OHD levels and increased MS risk [17][18][19]. The role of vitamin D in uencing disease prognosis in MS is more controversial. Higher 25OHD levels have been associated with lower relapse risk [20,21], lower risk of subsequent development of new T2 lesions and contrast-enhancing lesions on brain MRI [21], and lower change in EDSS scores [22]. However, other studies have not reported an association of vitamin D levels with MS disease activity [23]. Furthermore, a number of randomized clinical trials indicate that vitamin D supplementation does not seem to have bene cial effects on annualized relapse rate or EDSS scores in MS patients [24]. In our study, the presence of the risk allele for rs2762943 in MS patients had no effect on disease course, since it was not associated with increased in ammatory disease activity evaluated by the ARR or higher risk of disability progression evaluated by the EDSS or MSSS during follow-up. A limitation of the study was the evaluation of abovementioned in ammatory disease activity and disability progression measures at de ned moments in time such as 10 years from disease onset and at year 2015 to avoid the potential confounding effect of vitamin D supplements administered to MS patients in these clinical parameters. Another limitation of the study was the lack of MRI parameters to evaluate radiological disease activity. In order to circumvent this limitation, we measured serum NfL levels in a subgroup of MS patients as a proxy of CNS radiological in ammation. Although it cannot totally ruled out, the nding of similar NfL levels between risk allele carriers and non-carriers does not support the presence of signi cant differences in radiological measures of disease activity such as the number of T2 lesions or the number of contrast-enhancing lesions between both groups of patients. Noteworthy, risk allele carriers were younger at disease onset compared to non-carriers, indicating that the CYP24A1 rs2762943 gene variant seems to have more impact on MS susceptibility rather than on disease prognosis.

Conclusion
Altogether, these ndings indicate that the risk allele for the rs2762943 polymorphism in the CYP24A1 gene is associated with lower serum 1,25(OH) 2 D levels and heightened pro-in ammatory environment in MS. Additional studies are needed to further investigate the mechanisms by which the presence of the risk allele for the CYP24A1 rs2762943 gene variant is associated with an increased risk for MS. The authors would like to thank the MS patients and nurses for the participation and collaboration in the study.
-Consent for publication Not applicable.

-Availability of data and materials
The data that support the ndings of this study are available from corresponding author on reasonable request.

-Competing interests
The authors declare that they have no competing interests. Drafting/revision of the manuscript for content, including medical writing for content. M.C.: Drafting/revision of the manuscript for content, including medical writing for content; Study concept or design; Analysis or interpretation of data.
All authors reviewed the manuscript.
24. Zheng C, He L, Liu L, Zhu J, Jin T. The e cacy of vitamin D in multiple sclerosis: a meta-analysis.