A variant of RAG1 gene identified in severe combined immunodeficiency: a case report

WES was performed to identify potential pathogenic variants, which contributes to diagnosis confirmation. A homozygous variant c.2147G>A [NM_000448.2: exonme2: c.2147G>A (p.Arg716Gln)] in the RAG1 gene was detected. The evidence for pathogenicity based on the ACMG guidelines was “PM2_supporting” and “PP3” and the variation category was defined as “uncertain significance.” PP4 evidence was also arguable met. The variant was inherited from the parents who were heterozygous carriers (Fig. 2a). We further confirmed the variant using Sanger sequencing (Fig. 2b). This variant was listed in the ClinVar database (https://​www.​ncbi.​nlm.​nih.​gov/​clinvar/​mutation/​418448/​) as “likely pathogenic” and was described in a multi-institutional study in India [8]; however, there were no reports on its experimental verification. The allele frequency of c.2147G>A is 6.57e^-6 according to the public database gnomAD (http://​gnomad.​broadinstitute.​org), suggesting that the homozygous variant of p.Arg716Gln is extremely rare. The variant was predicted to be disease-causing according to multiple software (Table 3). We analyzed the domain arrangement of the RAG1 protein and found that the mutation is located in the V(D) J recombination-activating region, which is important for protein function (Fig. 2d). Species conservation analysis showed that this region is highly conserved in multiple species (Fig. 2c).

The Swiss-Pdb Viewer was used to predict and compare the structures of wild-type (WT) and mutant RAG1 [9]. The 3D protein structure of RAG1 used as a template was 89.9 % similar to that of the target protein. The arginine-to-glutamine substitution at position 716 resulted in the disruption of a hydrogen bond with lysine at position 722 (Fig. 3). The distance between residues 716 and 722 increased from 2.67 Å to 4.12 Å in the variant protein, which may impact its stability and function.

To further investigate the impact of the mutation on protein function, recombinant expression plasmids (pEGFP-N1-RAG1-wt/mut and phage-RAG1-wt/mut) were constructed and transfected into 293T cells. Samples were collected 48 h after transfection. The c.2147G>A (p.Arg716Gln) variation was successfully introduced into recombinant expression plasmids and validated by sequencing (Fig. 4a). Quantitative polymerase chain reaction (qPCR) showed that the expression of the variant gene was significantly lower than that of the WT, using the two sets of plasmids. The levels of variant mRNA decreased to 72.63 % and 70.9 % compared with those of WT mRNA using the phage and pEGFP-N1 plasmids, respectively (Fig. 4b). Western blotting showed that the protein levels of mutant RAG1 were lower compared with those of WT RAG1 using both phage (120 kd) and pEGFP-N1 (141.8 kd) plasmids (Fig. 4c). Overall, these results indicate a lower expression of variant RAG1 at both mRNA and protein levels.

The patient’s parents gave informed consent and agreed to participate in this study. To further explore the etiology of the disease, blood samples were drawn from the proband and his parents after obtaining consent. Sample sequencing, database filtering, and variant interpretation proceeded subsequently. Variant pathogenicity was predicted using SIFT [10], Polyphen-2 [11], Provean [12], Mutation Assessor [13], Mutation Taster [14], and FATHMM [15] software. The pathogenic classification was determined according to the ACMG guidelines.

Molecular modeling analysis was performed to estimate the impact of p.Arg716Gln on the protein structure. The structures of WT and variant RAG1 proteins were predicted using the Swiss-Model program. The Swiss-Pdb Viewer software was used for analyzing the structure of WT and mutant RAG1 proteins.

The 293T cells were cultured in DMEM medium with 10 % fetal bovine serum. We cloned the WT and variant fragments into two eukaryotic expression vectors (pEGFP-N1-RAG1-wt/mut and phage-RAG1-wt/mut). The primer sequences of phage-RAG1 (PHAGE-RAG1-SalI-F, PHAGE-RAG1-SalI-R) are shown in Supplemental Table 2. The plasmids were transfected into 293T cells using the Lipo2000 Transfection Reagent (Lipofectamine^TM 2000,ThermoFisher Scientific, Waltham,MA,USA) according to the manufacturer’s protocol.

The 293T cells were collected 48 hours after transfection. Total RNA was extracted using Trizol. cDNA was synthesized after digestion of DNA using the Prime Script RT Reagent kit with gDNA Eraser (RR047A, TaKaRa). The difference in mRNA expression between WT and mutant genes was detected by qPCR analysis, using the SYBR Green Realtime PCR Master Mix (QPK-201, Toyobo) in a CFX Connect Real-Time PCR Systems (BioRad). The two-pair primers used were RAG1-QPCR-F and RAG1-QPCR-R and the detail of primer sequences was shown in Supplemental Table 2. The statistical method used for analysis of qPCR data was the t-test.

We collected cells to detect the protein expression levels of the WT and mutant samples. RIPA Lysis Buffer (C1053, PPLYGEN) was used to extract the total proteins. A BSA kit (20201ES76, YEASEN) was used to determine the protein concentration and then perform protein denaturation treatment. The total protein was loaded per lane on an SDS-PAGE gel and transferred to a polyvinylidene difluoride membrane. The membrane was incubated with primary antibodies overnight at 4 °C after blocking with 5 % skim milk for 1 h. The details of the primary antibodies used in this study are described in Supplemental Table 3.