Elevated UMOD methylation level in peripheral blood is associated with gout risk

Uromodulin (UMOD) encodes an uromodulin glycoprotein, and its mutation results in uromodulin glycoprotein dysfunction and the occurrence of gout. The aim of our study was to assess whether UMOD methylation could predict the risk of gout. A total of 89 sporadic gout cases and 103 age and gender-matched healthy controls were recruited in this study. UMOD methylation level was determined by quantitative methylation-specific PCR (qMSP) in peripheral blood, and the percentage of methylated reference (PMR) was described to represent the methylation level. Our results showed that UMOD methylation was significantly higher in gout cases than controls (median: 1.45 versus 0.75, P < 0.001). The area under curve (AUC) of UMOD methylation in gout was 0.764 (P = 2.90E-10) with a sensitivity of 65.2% and a specificity of 88.3%. UMOD methylation level was shown to be significantly correlated with the serum level of uric acid (UA) (r = −0.208, P = 0.035). Besides, the luciferase reporter assay showed that UMOD CpG island region was able to upregulate gene expression (fold change = 2, P = 0.004). In conclusion, UMOD methylation assessment might be used to predict the occurrence of gout.

Plasma levels of glutamic pyruvic transaminase (ALT), glutamic oxalacetic transaminase (AST) were determined by the velocity method 17,18 . The concentrations of creatinine (CRE), uric acid (UA), blood glucose (Glu) and triglyceride (TG) in plasma were determined using the classic enzymatic methods [19][20][21][22] . Cholesterol level was measured using automated enzymatic methods 23 . High-density lipoprotein (HDL) cholesterol concentration was measured by enzymatic colorimetric methods with commercially available kits, and low-density lipoprotein (LDL) cholesterol concentration was measured by homogeneous assay 24 . The number of white blood cell (WBC) was measured by a standard blood test 25 .
DNA methylation analysis. The details of human genomic DNA extraction and concentration determination were as previously described 10 . DNA methylation was modified by EZ DNA Methylation-Gold TM kit (Zymo Research Corporation, Irvine, CA, USA). DNA methylation level was measured by quantitative methylation-specific PCR (qMSP) using the LightCycler ® 480 machine (Roche Diagnostics, Mannheim, Germany). To avoid errors that may occur from differences in the loading quantity of the samples, ACTB was taken as the internal reference. We used 100% SssI-treated sperm DNA as a positive control 26 , and nuclease-free water as a negative control for each panel. The qMSP was performed in a total volume of 10 µl and contained 5 µl of 2× SYBR Green Master Mix, 0.25 µl primers, 4 µl of ddH 2 O and 0.5 µl DNA. The primers were as follows: UMOD, forward 5′-GTTGTTGTTGGCGGAGTA-3′ and reverse 5′-CGACGATAACCTAACCTACG-3′; ACTB, forward 5′-TGGTGATGGAGGAGGTTTAGTAAGT-3′ and reverse 5′-AACCAATAAAACCTACTCCTCCCTTAA-3′. PCR amplification procedure included an initial denaturation at 95 °C for 10 min, 45 cycles of denaturation at 95 °C for 20 sec, annealing at 59 °C for 30 sec and extension at 72 °C for 30 sec. A melting curve procedure included 95 °C for 15 sec, 58 °C for 60 sec and 0.11 °C per second up to 95 °C. The amount of methylated DNA (PMR, percentage of methylated reference) at a specific locus was calculated by dividing the UMOD:ACTB ratio of a sample by the UMOD:ACTB ratio of SssI-treated human sperm DNA (presumably fully methylated) 26 .
Luciferase reporter gene assay. The human embryonic kidney 293 T (HEK293T) cell line was cultured as previously described 27 . The fragment of UMOD (+7151 bp to +7550 bp), GCKR (−173 bp to +227 bp), COMT (−386 bp to +14 bp) and CCL2 (−537 bp to −138 bp) were chemically synthesized and were digested with XhoI and KpnI (New England Biolabs, Ipswich, MA). The target DNA fragment, purified by Cycle Pure Kit (Omega, Norcross, GA, USA), was cloned to pGL3 Basic vector in the presence of DNA Ligation Kit (TaKaRa, Japan). The empty pGL3-Basic vector was used as negative control, and the pGL3-Control vector, (Promega, Madison city, WI, USA) containing an SV40 promoter upstream of the luciferase gene was used as positive control. Cells were prepared in 96-well plates and the details of plasmids transfection were as described previously 28 . After 18-72 h of HEK293T cells transfection, renilla and firefly luciferase activity was measured by SpectraMax 190 (Molecular Devices, Sunnyvale, USA). Luciferase activity was determined with the dual luciferase reporter assay system (Dual-Luciferase ® Reporter Assay Systems, Promega, Madison city, WI, USA).

Statistical analysis.
All the statistical analyses were performed by SPSS software version 18.0 (SPSS, Inc., Chicago, IL, USA). Comparisons of the PMR differences between the gout cases and controls were performed by non-parametric test. The correlations between UMOD methylation and clinical features were assessed by Spearman test. Receiver operating characteristic (ROC) curves were generated to confirm the diagnostic accuracy of UMOD. P value less than 0.05 was considered to indicate a statistically significant difference.

Results
In the current study, only the male samples were selected since gout was predominant in males (a male/female ratio of 4:1) 29, 30 . As shown in Table 1, a total of 11 clinical characteristics were collected from all the individuals. Significantly lower level of HDL was found in the gout cases than controls (mean ± sd: 1.23 ± 0.31 versus 1.50 ± 0.35, P < 0.001). Meanwhile, significantly higher levels of ALT, UA, Glu, cholesterol, TG and WBC were found in the gout cases than controls (all P ≤ 0.001). A fragment located in CpG (cytosine-phosphate-guanine) island of UMOD (Chr16: 20,344,373-20,364,037), hg19) was selected for the methylation assay (Fig. 1A). DNA sequence analysis showed that the bisulphite conversion of the template DNA was successful (Fig. 1B). Capillary electrophoresis confirmed that the amplified fragment length was 73 bp (Fig. 1C). As shown in Fig. 2, UMOD hypermethylation was significantly associated with the risk of gout. UMOD methylation was elevated in the gout cases compared with the controls [median (interquartile range): 1.45 (0.87, 3.54) versus 0.75 (0.59, 0.92), P < 0.001]. Subsequently, we analyzed the diagnostic role of UMOD hypermethylation in peripheral blood, obtaining an AUC of 0.763 (P = 2.90E-10, Fig. 3). The ROC curve showed that UMOD methylation was a promising biomarker for gout (sensitivity = 65.2%, specificity of 88.3%).
In order to investigate the relationship between UMOD methylation and the pathogenesis of gout, the correlation tests were performed between UMOD methylation levels and clinical features in control samples. Significant inverse correlation was found with UMOD methylation level and UA (r = −0.208; P = 0.035, Table 2). However, there was no significant association between clinical features (age, ALT, AST, CRE, Glu, cholesterol, HDL, LDL, TG, WBC) and UMOD methylation (all P > 0.05, Table 2).
We performed a dual-luciferase reporter assay to check whether the UMOD CpG island region (+7151 bp to +7550 bp) was able to regulate gene expression. Our results showed that the transcriptional activity of recombinant pGL3-UMOD plasmid was higher compared with that of empty vector pGL3-basic (mean ± sd: 36.22 ± 2.15 versus 17.11 ± 0.16, fold change = 2, P = 0.004, Fig. 4).

Discussion
In the present study, we reported for the first time that UMOD hypermethylation was significantly associated with the risk of gout in Chinese male patients. Moreover, the methylation levels of UMOD could be served as a predictive biomarker for the risk of gout.
DNA methylation has been studied in many metabolic diseases. Prdx2 and SCARA3 hypermethylation played an important role in the pathogenesis and progression of diabetes mellitus 31 . In diabetic ketoacidosis, POMC hypomethylation might make the patients' condition worse 32 . Moreover, AR methylation was shown to be  Scientific RepoRts | 7: 11196 | DOI:10.1038/s41598-017-11627-w associated with hyperuricemia 33 . However, there were few articles between DNA methylation and gout. Previous studies showed that uromodulin (UMOD) played an important role in gout 34 . UMOD encoded the uromodulin glycoprotein. The mutations of UMOD led to uromodulin glycoprotein dysfunction and gout 35 .
As shown in the genecards website, UMOD expression level is able to be detected in the whole blood according to both the microarray and the RNAseq technologies. UMOD expression level is the highest in kidney, and uromodulin is the most abundant urine protein 36 . Decreased serum uromodulin is often correlated with the increase of serum inflammatory cytokines and the aggravation of diseases including kidney disease, hypertension and diabetes 11,[36][37][38] . In addition, the increase of serum uromodulin was a promising prognostic biomarker for recovery from acute kidney injury 39 . Besides, another kidney-specific gene, Klotho (KL) was reported to be much less expressed in peripheral blood cells compared in kidney 40 . KL hypermethylation in peripheral blood mononuclear cells was detected to be associated with the aggravation of chronic kidney disease 41 .
In the current study, elevated UMOD methylation in peripheral blood was shown to be associated with the risk of Gout, which is characterized by urate crystal-induced inflammation 42 . Since UMOD expression was often inversely associated with the levels of inflammatory cytokines in peripheral blood 11 , we speculate that elevated UMOD methylation in Gout might reduce the expression of UMOD, which triggers an immune response and leads to the risk of gout. In addition, our study couldn't exclude the possibility that UMOD hypermethylation (and possibility of other genes) in peripheral blood cells could be secondary to increased circulating levels of uric acid (or of other molecules found to be increased in cases). Future study is warranted to investigate the correlation of UMOD methylation with UMOD expression in peripheral blood, kidney and other tissues.
In our study, a significantly higher serum UA level was found in gout patients than that in normal controls, and this finding might support that an elevated serum UA concentration was the main cause of gout 43 . But a significant inverse correlation was found between UMOD methylation level and serum UA level in controls. Due to the limited samples, we didn't measure uromodulin levels in serum or urine in cases and controls in time. Therefore, we couldn't test the correlation of UMOD expression and UMOD methylation in the samples. The relationship between UMOD methylation and the pathogenesis of gout needs further investigation.
Joint aspiration with synovial fluid analysis for monosodium urate crystals were the reference standard in early diagnosis of gout, however, rarely patients used this method in the early diagnosis of gout due to the risk of infection 44 . Our ROC curve analysis showed a moderate sensitivity of 65.2% and a high specificity of 88.3%. Moreover, increased levels of uric acid in blood is one of the clinical diagnostic criteria for gout 45,46 . However, the blood uric acid index does not seem sensitive enough, patients with early-onset gout do not have a significant increase in uric acid levels 47 . And the detection rate of gout by using serum uric acid had a relatively low AUC of 0.61 48 . These findings suggested that UMOD methylation could be a diagnostic biomarker for gout. Dual-luciferase  reporter system assay is a common tool to verify whether the cloned DNA fragment can play a regulation role in the expression of the luciferase reporter gene 49 . HK293T cell line was chosen for its easy culture and transfection.
In the current study, pGL3-UMOD recombinant plasmid was constructed, and it was co-transfected into cells along with an internal control vector (pRL-SV40). Our results showed that the specific fragment (+7151 bp to +7550 bp) in UMOD CpG island region could induce a significantly higher expression of reporter gene than the control. Besides, as shown in the Supplementary Figure 1, other 400-bp inserts didn't show obvious promoter activities, suggesting the UMOD fragment contained DNA elements with gene up-regulation. According to the TCGA dataset (https://genome-cancer.ucsc.edu/), there were five CpGs (cg03140788, cg06294373, cg21996068, cg09792189 and cg00376654) on the 400 bp fragment and three CpGs (cg06294373, cg21996068 and cg09792189) in the 73 bp fragment. Using the TCGA data, we found all the five CpGs were in positive correlation (r > 0.25, P < 0.001), suggesting that the selected CpGs might represent the neighbor CpG sites. In addition, UCSC Genome Browser website showed that the fragment was overlapped with several transcription factors binding sites, such as CTCF and ZNF143. We used P-Match method 50 to predict TFBS in the selected fragment, there were Nkx2-5, c-Rel, NF-kappaB(P65), NF-kappaB in this fragment. Further study should be performed to explore the regulatory roles of CpG region in UMOD expression.
In conclusion, our study found that UMOD DNA hypermethylation in peripheral blood might be used to predict the risk of gout.