Polymorphisms of nucleotide factor of activated T cells cytoplasmic 2 and 4 and the risk of acute rejection following kidney transplantation

The study protocol was in accordance with the ethical standards of the Declarations of Helsinki and Istanbul. The procedures were limited to the living-related transplantation of kidney tissues to lineal or collateral relatives not beyond the third degree of kinship or transplantation of kidney tissues from cadaveric allograft donors after cardiac death. The protocol of this study was approved by the local ethics committee of the First Affiliated Hospital with Nanjing Medical University. Written informed consent was obtained from all the transplant recipients. Moreover, none of the transplant donors were from a vulnerable population, and their written informed consent was obtained.

This was a retrospective, single-center, case–control study. A total of 200 renal transplant recipients who underwent kidney transplantation between 1st February 2010 and 1st December 2015, at our center of the First Affiliated Hospital with Nanjing Medical University, were enrolled in this study. According to the study criteria, we included (1) patients aged more than 18 years or less than 60 years; (2) patients who had experienced at least one AR episode after kidney transplantation that was confirmed by histological examination with hematoxylin–eosin staining and immunohistological staining according to the Banff07 criteria [25] (these patients were assigned to the AR group); (3) patients with stable serum creatinine levels (< 120 μmol/L; fluctuation, < 20%) for at least 3 months who had not experienced any AR episodes, delayed graft dysfunction or opportunistic infection after kidney transplantation (these patients were assigned to the stable group); and (4) patients who had undergone post-transplant follow-up for more than 6 months. Further, we excluded (1) patients who did not fulfill the inclusion criteria, (2) patients with chronic viral infections such as HIV and chronic viral hepatitis B and C, and (3) pregnant women. The medical records of the enrolled patients were critically reviewed, and relevant data were extracted independently by two authors (Z.J. Wang and K. Wang).

All recipients received immunosuppressive protocols that included three or four drugs: cyclosporin A (CsA, n = 98) or tacrolimus (TAC, n = 102) combined with mycophenolate mofetil (MMF) and prednisone, with or without sirolimus (n = 15) (Table 1). Methylprednisolone was intravenously administered as a dosage of 500 mg/day on the day of surgery and until 2 days after the kidney transplantation. Following this, the dosage was reduced to 400, 300, 200 mg and then 80 mg on each subsequent day. This was followed by oral administration of prednisone at 30 mg/day as maintenance therapy. Moreover, 20 mg basiliximab was intravenously administered at 30 min before the surgery and on the fourth post-transplant day. MMF was administered at 24 to 48 h after transplantation, and the starting dosage was 0.75 g/d to 1.0 g/d (BID). The dose was adjusted according to the serum creatinine (Scr) and drug concentrations. The starting dosage of CsA and TAC was 8 mg/kg/day and 0.2 mg/kg/day, respectively; these dosages were later adjusted according to the Scr levels and drug concentrations. In the case of patients who experienced AR episodes, methylprednisolone was intravenously administered at a dosage of 200 mg/day for 3–5 days.

Peripheral blood samples (2 ml) from each recipient were collected. In particular, blood samples from recipients in the AR group were taken before the intravenous administration of methylprednisolone and the allograft biopsy. The blood samples were transferred to the laboratory and stored in the refrigerator at − 80°C.

DNA of the blood samples was extracted using the QIAmp DNA mini kit according to the manufacturer’s instructions (Qiagen, Hilden, Germany). The concentration and purity of genomic DNA (gDNA) were quantitatively analyzed using NanoDrop ND2000 (Thermo, MA, USA), while gene integrity was assessed using agarose gel electrophoresis. The gDNA samples extracted were considered to be acceptable if the total mass was ≥ 1 μg, and the A260/A280 absorbance ratio was ≥ 1.80 and ≤ 2.0. The gDNA was added to a mixture of upstream and downstream oligonucleotides specific to the target regions of interest for hybridization (see supplementary files). Then, the gDNA was fragmented using a Bioruptor Interrupt (Diagenode, Belgium), and quantitative detection was performed to ensure that the average fragment size was 150–250 bp. Fragmentation was followed by end repair, dA tailing, and sequencing adaptor ligation with the ABI 9700 PCR instrument (ABI, USA). The adapter-ligated DNA was amplified by selective, limited-cycle PCR for five cycles and then quantitatively analyzed using the Qubit dsDNA HS assay kit (Invitrogen, USA). The prepared library (750 ng) was hybridized by overnight incubation (for 8–16 h) at 65°C with 11 µl of hybridization blocking buffer (Allwegene, China), 20 µl of hybridization buffer (Allwegene, China), and a mixture of 5 µl RNase block (Invitrogen, USA) and 2 µl probes (Allwegene, China). The hybridized products were mixed with 200 µl Dynabeads MyOne Streptavidin T1 magnetic beads (Invitrogen, USA) for 30 min at room temperature. The products were then washed two times with a wash buffer (Allwegene, China), and the mixture was amplified for 16 PCR cycles and quantitatively assessed using the Qubit dsDNA HS assay kit (Invitrogen, USA). The captured libraries were denatured and loaded onto an Illumina cBot instrument at a concentration of 12–16 pmol/l for cluster generation according to the manufacturer’s instructions. Up to 20 WUCaMP libraries were sequenced per HiSeq lane. A PhiX control (Illumina) was added to lane 8 of each flow cell.

Sequencing data, such as the number of altered chromosomes, genomic alterations, and the depth of the sequencing coverage, were analyzed. All analyses were based on the human reference sequence UCSC build hg19 (NCBI build 37.2) using the Burrows–Wheeler Aligner. Local alignment and removal of duplicates were performed using the Genome Analysis Toolkit (Version 3.7-0, Broad Institute, Cambridge, MA, USA) and the Picard software (Broad Institute, Cambridge, MA, USA). Detection of SNPs was performed using dbSNP 132. Damaging or deleterious SNPs were predicted using the Gemini software, and prediction tools, including sorting intolerant from tolerant and polymorphism phenotyping, were used for analysis of all human non-synonymous SNPs. In addition, putative somatic variant calls were detected using two separate programs, MuTect 1.1.5 and VarScan 2.3.6, by pairing each sample with its matched blood sample.

Conformance to the Hardy–Weinberg equilibrium (HWE) was determined based on gene frequencies obtained from a single gene counting. The Chi-square test was used to compare the observed and expected values. To perform genotype association analysis, five models were used: the dominant model (minor allele homozygotes plus heterozygotes vs. major allele homozygotes), recessive model (minor allele homozygotes vs. heterozygotes plus major homozygotes), additive model (major homozygotes vs. heterozygotes vs. minor homozygotes), HET model (major homozygotes vs. heterozygotes) and HOM model (major homozygotes vs. minor homozygotes). Genotypic frequencies of the control and AR groups were compared by the Chi-square test. In addition, we explored linkage disequilibrium blocks using Haploview version 4.2. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated using the SPSS 13.0 software (SPSS Inc., Chicago, IL, USA). P < 0.05 was considered to indicate statistical significance. The OR provides an effect estimate: when its value is less than 1, a protective effect can be assumed, whereas when it is more than 1, the disease risk is considered to be increased. In addition, the genotypic distributions of the NFATC2/C4 SNPs in recipients with AR and stable recipients were analyzed using logistic regression models adjusted for age, gender and immunosuppressive protocol.

The baseline clinical characteristics of the renal transplant recipients in our study are presented in Table 1. A total of 69 recipients who experienced at least one AR episode (42 men and 27 women) were assigned to the AR group, and 131 recipients who had not experienced any AR episodes (82 men and 49 women) were assigned to the stable group. No significant differences (P > 0.05) were observed in baseline characteristics, such as age, gender and immunosuppressive protocol, between the stable and AR groups. No patient in two groups was tested with positive results of panel reactive antibodies before kidney transplantation.

A total of 71 SNPs of the NFATC2 and NFATC4 genes (NFATC2: 34 SNPs, NFATC4: 37 SNPs) were identified using the NGS technology and HWE analysis (Supplemental Tables 1 and 2). The genetic distribution of these SNPs in the AR and stable group are shown in Supplemental Tables 3 and 4. Among these SNPs, 9 NFATC2 SNPs and 11 NFATC4 SNPs are being reported here for the first time. We selected 27 SNPs with MAF (minor allele frequency) more than 0.05 to further test the relationship with the AR episodes.

After adjustment for age, gender and immunosuppressive protocols in all five models (additive, dominant, recessive, HOM and HET), we found that rs2426295 on the NFATC2 gene was significantly correlated with the risk of AR episodes following kidney transplantation in the HET model (AA vs. AC: OR = 0.43, 95% CI = 0.19–0.98, P = 0.045, Table 2). No significant association was observed for the other 26 SNPs in all the models (Supplemental Table 5).