A novel SNP in the 5’ regulatory region of organic anion transporter 1 is associated with chronic kidney disease

We aimed to analyze the associations of single nucleotide polymorphisms (SNP) in the 5′ regulatory region of the human organic anion transporter 1 (OAT1) gene with chronic kidney disease (CKD). A case-control study including age- and sex-matched groups of normal subjects and patients with CKD (n = 162 each) was designed. Direct sequencing of the 5′ regulatory region (+88 to −1196 region) showed that patients with CKD had a higher frequency of the −475 SNP (T > T/G) than normal subjects (14/162 vs. 2/162). The luciferase activity assay results indicated that the −475G SNP had a higher promoter efficiency than the −475T SNP. Chromatin immunoprecipitation (ChIP) and LC/MS/MS analyses showed that the −475G SNP up-regulated 26 proteins and down-regulated 74 proteins. The Southwestern blot assay results revealed that the −475G SNP decreased the binding of Hepatoma-derived growth factor (HDGF), a transcription repressor, compared to the −475T SNP. Overexpression of HDGF significantly down-regulated OAT1 in renal tubular cells. Moreover, a zebrafish animal model showed that HDGF-knockdown zebrafish embryos had higher rates of kidney malformation than wild-type controls [18/78 (23.1%) vs. 1/30 (3.3%)]. In conclusion, our results suggest that an OAT1 SNP might be clinically associated with CKD. Renal tubular cells with the −475 SNP had increased OAT1 expression, which resulted in increased transportation of organic anion toxins into cells. Cellular accumulation of organic anion toxins caused cytotoxicity and resulted in CKD.

The genes encoding OAT1 and other solute transporters are clustered on human chromosome 11, with OAT1 and OAT3 genes existing on chromosome 11 as a tandem pair. Previous studies have demonstrated that the selective pressure on the coding sequences of OAT1 is relatively small compared with that of other OAT family members 14 , suggesting that OAT1 is a conserved protein. Clinical studies have also indicated that OAT1 has low genetic and functional diversity in coding regions 15 . Therefore, this study aimed to analyze single nucleotide polymorphisms (SNPs) in the 5′ regulatory region of human OAT1 (SLC22A6) and their possible clinical associations with CKD. We screened for variants in the 5′ regulatory region of OAT1 in DNA samples from normal subjects and subjects with CKD (n = 162 for each group), and the associations between regulatory polymorphisms and CKD were analyzed. We also performed cellular studies to investigate the molecular mechanism of expression regulated by polymorphisms in the 5′ regulatory region of human OAT1.

Results
The reference sequence of the 5′ regulatory region of human OAT1 from +88 to −1196 bp and the results of promoter prediction are shown in Supplemental Fig. S1. There were 3 potential promoter regions predicted on the 5′ regulatory region of OAT1 from −1 to −1196 bp. To define possible clinical associations between the SNPs of OAT1 and CKD, DNA fragments containing the 5′ regulatory region of OAT1 (from +88 to −1196 bp) were analyzed with chromosomal DNA of peripheral leukocytes by direct sequencing (Fig. 1A). The study included 324 study subjects divided into age-and sex-matched normal (n = 162) and CKD (n = 162) groups. The characteristics of study subjects are listed in Table 1. Ten SNPs were found by direct sequencing, and the frequencies of these SNPs for each group are summarized in Table 2. Four of the SNPs found in this study (−118   of T > T/G was the most common SNP in the study population (16/324, 4.9%). In addition, the frequency of the −475 SNP in subjects with CKD was significantly higher than in normal subjects (14/162 vs. 2/162; P = 0.003). The frequencies of other SNPs did not differ significantly between normal (control) and CKD groups ( Table 2). These results suggest that the −475 SNP (T > T/G) of OAT1 is clinically associated with CKD. The odds ratio for subjects with the −475 SNP (T > T/G) having CKD was 7.57 (95% Confidence Interval: 1.69-33.86; P = 0.008).
To elucidate possible effects of the −475 SNP on expression of OAT1, a promoter activity assay with wild-type and −475 mutant promoters (−1 to −1196 nt) was performed ( Fig. 2A). The −475 mutant promoter increased luciferase activity relative to the wild-type promoter (2.5 vs. 1.0; P = 0.019) (Fig. 2B). Real-time PCR results also showed that the −475 mutant promoter increased luciferase mRNA expression relative to the mutant promoter (P = 0.003) (Fig. 2C). These results indicate that the −475 SNP of OAT1 might up-regulate OAT1 expression by affecting transcription factor binding. A chromatin immunoprecipitation/LC/MS/MS analysis using wild-type and −475 mutant oligonucleotides (−463 to −487; 25 bp) was conducted to identify potential binding transcription factors. The chromatin immunoprecipitation analysis flow is summarized in Fig. 3. Compared with the wild-type oligonucleotide, 26 proteins were up-regulated and 74 proteins were down-regulated (2-fold changes,    (Table 3). Hepatoma-derived growth factor (HDGF) is known as a transcription repressor 16,17 . LC/MS/MS results showed that the −475 mutant oligonucleotide significantly down-regulated HDGF binding (Table 3). Alignment analysis of the HDGF binding site of the SMYD1 promoter (30 nt) with the 5′ regulatory region of OAT1 (−463 to −487, 25 nt) revealed 43.3% sequence identity over a 30 nt overlap. In addition, the nucleotide at the −475 position of OAT1 was conserved in the HDGF binding site on the SMYD1 promoter ( Fig. 4A) 17 . Southwestern blot analysis with HDGF produced by in vitro translation and synthetic wild-type and −475 mutant oligonucleotides (−463 to −487, 25 bp) showed that the −475 mutant oligonucleotide significantly down-regulated HDGF binding (Fig. 4B). To define the regulatory effects of HDGF on OAT1 expression, expression of OAT1 by HK2 cells over-expressing HDGF was analyzed by Western blot. Compared with control cells, cells over-expressing HDGF had significantly decreased OAT1 expression (Fig. 4C). These results suggest that the −475 SNP of OAT1 might down-regulate HDGF binding, resulting in over-expression of OAT1.
To define the association of HDGF with kidney malformation, HDGF-deficient zebrafish embryos were obtained by injection of antisense morpholino oligonucleotides. The zebrafish embryos were produced from the green fluorescent kidney transgenic zebrafish line Tg(wt1b:egfp), which enables easier observation of kidney malformations. The results showed that embryos derived from HDGF3-MO injection displayed more malformed kidney phenotypes at 48 hpf than did embryos of the uninjected control group (defect rate 23.1% vs. 3.3%, n = 30; Fig. 5C,D). Differences in defects in the glomerulus, pronephric tube, and pronephric duct were observed between the uninjected control and HDGF3-MO-injected groups, particularly in fish with severe defects (Fig. 5A,B). These results indicate that HDGF3 expression is essential for kidney development.

Discussion
The central role of the kidney in the elimination of potential internal or external toxins from the blood into the urine is well documented. Substrates for OAT1 are varied and range from the classic small organic anion para-aminohippurate to several clinically important drugs, herbicides, and endogenous substances. For these reasons, there has been much interest in the possibility that polymorphisms in SLC22A6 may be partially responsible for variation in the handling and efficacy of many commonly used drugs and toxins that are transported by OAT1 14,18,19 .
Substantial evidence indicates that OAT1 activity is critical in renal function and injury. A key role for OAT1 in the handling of uremic toxins derived from the gut microbiome was identified using Oat1-knockout mice 20 . The results of studies in Oat1-knockout mice suggest that uremic toxins and solutes are significantly retained in Oat1-knockout mice. On the other hand, OAT1 activity also has important roles in the pathogenesis of drug-related kidney injury. In vivo studies with Oat1-knockout mice verified that disruption of OAT1 activity can prevent renal toxicity of drugs or chemicals 21,22 . In Oat1-knockout mice, the loss of function of OAT1 was associated with decreased renal accumulation of arachidonic acid and lessened the severity of renal injury compared to wild-type animals 8 .
In a previous study with an ethnically diverse sample of 96 individuals, only one polymorphism was found in the 5′ regulatory region of OAT1 19 . In our study population (n = 324), there were 10 SNPs. Furthermore, our study also indicated that polymorphisms in the 5′ regulatory region of human OAT1 had significant clinical associations with CKD. Our results suggest that subjects with the −475 SNP (T > T/G) of OAT1 have increased risk of CKD. Our study also found that the −475 SNP with T to G transversion could increase OAT1 promoter activity that might result in increased OAT1 expression. OAT1 plays a major role in the renal uptake of uremic toxins on the basolateral membrane of renal tubules. Previous studies have indicated that OAT1 expression is associated with intracellular accumulation of organic anion toxins in the renal tubular cells of patients with CKD 7 . It has also been shown that transporter molecules such as OAT1 transport anionic uremic toxins into cells, where they accumulate and can cause oxidative stress and ultimately kidney injury [23][24][25] . Our results suggest that increased OAT1 expression due to the −475 SNP with T to G transversion might increase intracellular organic anion uremic toxins accumulation such that it exceeds the excretion rate, resulting in nephrotoxicity (Fig. 6).
Another important finding of our study was that the −475 SNP of OAT1 with T to G transversion could decrease HDGF binding. HDGF is considered a multi-functional protein and is suggested to have important roles in organ development 26 . HDGF shows proliferative activity, and expression of HDGF has been reported in many different tumor types and correlated with prognosis 27,28 . HDGF is also known as a transcription repressor. A microarray study with mouse primary aortic vascular smooth muscle cells demonstrated that expression of HDGF significantly down-regulated a large group of genes and increased expression of a relatively small number of genes 17 . Similarly, our study demonstrated that over-expression of HDGF could down-regulate OAT1 expression in cultured kidney cells, indicating that HDGF might be a transcription repressor for OAT1. The findings above suggest that the −475 SNP of OAT1 with T to G transversion might increase OAT1 expression by down-regulating HDGF binding to the OAT1 promoter thus lessening the transcriptional repression of OAT1.
In the present study, we identified a clinical association between SNPs of OAT1 and CKD. Our study demonstrated that SNPs of OAT1 can alter the transcriptional regulation of OAT1, which might affect CKD outcomes. Despite the low PCR error rate, this study may have been confounded by PCR errors that may have caused false SNP signals. Collectively, our data provide the first evidence of the clinical significance of SNPs of OAT1 on CKD and suggest that testing SNPs of OAT1 might serve as a valuable tool for CKD prevention and therapy.

Methods
Study subjects. A case-control study was conducted with sex-and age-matched groups. The inclusion criterion was adults aged >18 but <80 years. Patients were excluded from the study if they had diabetes mellitus, autoimmune disease, malignant disease, polycystic kidney disease, organ transplantation, infections requiring admission to the hospital in the past 3 months or an unwillingness to participate in the study. In total, 162 normal subjects and 162 subjects with CKD (eGFR < 30 ml/min/1.73 m 2 ) were recruited into the study. This study adhered to the Declaration of Helsinki and was approved by the Ethics Committee of the Institutional Review Board at Chang Gung Memorial Hospital (Approval No. 103-0344C). Informed consent for all participants (162 normal subjects and 162 subjects with CKD) was obtained and kept at the Chang Gung Memorial Hospital.
Leukocyte chromosomal DNA preparation. In brief, leukocytes were separated from a specimen of whole human blood by mixing the specimen with a hypotonic EDTA solution (1 mM). White blood cells were separated by centrifugation. The chromosomal DNA was extracted with an automatic nucleic acid extraction system according to the product instructions (LabTurbo 96 Standard System, Taigen Bioscience Corporation, Taipei, Taiwan).
Chromosomal DNA sequencing and polymorphism identification. The 5′ regulatory region of OAT1 (−1196 to +88 relative to the transcription start site) was amplified by polymerase chain reaction (PCR) with chromosomal DNA. PCR was performed in 25 µL SYBR Green PCR Master Mix (Applied Biosystems, Waltham, MA) containing 0.6 mol/L primers (Table 4) and 1 µg DNA using an iQ5 PCR detection system (Bio-Rad, Berkeley, CA). Then, the PCR products (500 ng) were identified and purified by gel electrophoresis and   29 .    Table 4.
Cell culture and transfection. HK2 cells were obtained from ATCC and cultured as suggested by ATCC. After electrophoresis, the gels were stained with VisPRO 5 minutes Protein Stain kit (Visual Protein, Taiwan). After staining, the gels were washed in Milli-Q water and stored at 4 °C until processing for in-gel digestion. The gel lanes corresponding to the sample were cut into 5 slices, and each slice was processed for in-gel digestion according to the Shevchenko method. Briefly, each slice was washed/dehydrated three times in 50 mM ammonium bicarbonate (ABC, pH 7.9)/50 mM ABC +50% acetonitrile (ACN). Subsequently, cysteine bonds were reduced by incubating slices in 10 mM dithiothreitol for 1 h at 56 °C and alkylated by incubating slices in 50 mM iodoacetamide for 45 min at room temperature (RT) in the dark. After two subsequent wash/dehydration cycles, the slices were dried for 10 min in a vacuum centrifuge (ThermoFisher, Breda, The Netherlands) and incubated overnight with 6.25 ng/μL trypsin in 50 mM ABC at 25 °C. Peptides were extracted into 100 μL of 1% formic acid and then extracted twice into 100 μL of 50% ACN in 5% formic acid. The volume was reduced to 50 μL in a vacuum centrifuge before LC-MS/MS analysis. Peptides were separated using an Ultimate  Western and Southwestern blotting. Total protein was extracted using a commercial kit according to the manufacturer's instructions (Protein Extraction Kit, Millipore, Billerica, Massachusetts). Then, 30 μg of protein from each sample was mixed with sample-loading buffer and loaded onto separate lanes of a 12% sodium dodecyl sulfate-polyacrylamide gel. The proteins were electrotransferred onto polyvinylidene fluoride membranes (0.2 μm: Immun-Blot, Bio-Rad) and then immunoblotted with antibodies against HDGF (Abcam), OAT1 (Abcam), and β-actin (Abcam). The intensity of each band was quantified using NIH Image software (Bethesda, Maryland), and the densitometric intensity corresponding to each band was normalized against β-actin expression. HDGF protein for Southwestern blotting was synthesized using purified HDGF mRNA with a eukaryotic Fish embryo staging and morpholino injection. Mature Tg(wt1b:EGFP) 30 zebrafish were maintained at 28 °C with a photoperiod of 14-h light and 10-h dark in an aquarium supplied with freshwater and aeration. Embryos were produced using standard procedures 31 and were staged according to standard criteria (hours postfertilization, hpf) 32 or by days postfertilization (dpf). Antisense morpholino oligonucleotides targeting the 5′ untranslated region and the translation initiation site of HDGF (HDGF3ATG-MO: 5′-GGCGAGCCATGCCGACACAC-3′) were designed and obtained from Gene Tools (Philomath, OR). MOs were dissolved in 1× Danieau solution containing 0.5% Phenol red, and 2.3 nl of MO solution of the indicated concentration was injected into 1-cell-stage Tg(wt1b:egfp) embryos. All of the embryos were observed under a microscope (DM 2500, Leica, Wetzlar, Germany) equipped with a GFP fluorescent module. Pictures of the embryos were captured at particular stages with a digital camera (SONY, Tokyo, Japan).