Introduction

Cystinuria (OMIM 220100) is a congenital amino acid metabolic disorder that can lead to defective luminal transport of dibasic amino acids (cystine, ornithine, lysine, and arginine), and the defective reabsorption of cystine can lead to dramatically increased cysteine stone formation in the kidneys and other organs [8, 23, 27]. Cystine is a sulfur-containing homodimeric amino acid, which is composed of two cysteine molecules connected by disulfide bonds. Under physiological conditions, cystine is filtered freely by the glomerulus and reabsorbed by the proximal renal tubules, and its transport mechanism is not completely clear [5]. The transport proteins rBAT (encoded by SLC3A1) and b0, + AT (encoded by SLC7A9) participate in high-affinity transport of cystine and dibasic amino acids in kidney and intestine by forming a heterodimeric transport network connected by a covalent disulfide bond [3,  4, 10, 19, 21, 31]. Defects in transport protein cause impaired reabsorption of cysteine, which results in hyperexcretion of cystine in the urine [26], due to the low solubility of cystine in normal urine pH, patients tend to form cystine stones.

Cystinuria can be caused by mutations in the SLC3A1 and/or SLC7A9 gene [8, 14]. The SLC3A1 gene is located on chromosome 2p16.3–21, and the SLC7A9 gene is located on chromosome 19q13.11. In 2002, the International Cystinuria Consortium classified cystinuria based on mutations in different genes into type A (SLC3A1 mutations), type B (SLC7A9 mutations) and type AB (both the SLC3A1 and SLC7A9 mutations) [ 8, 13, 18, 27, 28]. A classification based on urinary amino acid excretion identified type 1 (normal cysteine excretion), type non-1 (moderate and high cysteine excretion), mixed, and unidentified cystinuria [14]. However, the correlation between the genotype and phenotype of cystinuria remains unclear. Cystinuria caused by SLC3A1 mutations is usually an autosomal recessive disease and SLC3A1 heterozygotes have no apparent phenotype, but cystinuria caused by SLC7A9 mutations is usually autosome dominant inheritance with 86% of SLC7A9 heterozygotes having abnormal urinary dibasic amino acid levels and some of them will develop into Cystine stones [28]. There is no difference between patients with the type A (biallelic SLC3A1 mutations) and type B (biallelic SLC7A9 mutations) in a variety of clinical parameters. However, carriers with a single heterozygous mutation also have different disease severity levels and cannot be distinguished from patients with biallelic SLC3A1/SLC7A9 mutations [7, 12, 23]. The complex genetic pattern and incidence of clinical forms of cystinuria differ across populations. The complex genetic pattern and incidence of clinical forms of cystinuria differ across populations. The frequency of cystinuria is approximately as 0.000143 in neonates, 0.0004 among Libyan Jews, 0.0000667 in Americans, and 0.000001 in Swedes [30]; in Chinese nephrolithiasis patients, the frequency of cystine stones is 0.00005 [27, 32]. However, the exact frequency of cystinuria carriers in different populations remains unclear because there are limited records of the prevalence of this disease in different populations [8].

In 1994, pathogenic variants of the SLC3A1 gene were identified as leading to cystinuria [3]. Currently, the Human Gene Mutation Database (HGMD® Professional 2022, http://www.hgmd.cf.ac.uk/ac/index.php, accessed December 4, 2022) contains nearly 260 pathogenic variants of SLC3A1 reported from different populations. Almost 95% of biallelic SLC3A1 mutations lead to cystine stone formation at some age, and the recurrence rate of stones and related metabolic diseases is as high as 60% [13]. Therefore, it is challenging to predict the phenotypes of individual cystinuria patients from different countries due to population-based variability [23, 25, 29, 30].

This study identified compound heterozygous variants of the SLC3A1 gene (c.898_905del (p.Arg301AlafsTer6) and c.1898_1899insAT (p.Asp634LeufsTer46)) in a 12-year-old Chinese boy with kidney stones. We also recorded the patient’s clinical characteristics and treatments to enhance our understanding of the link between the SLC3A1 genotype and the manifestations of cystinuria in the Chinese population.

Materials and methods

Study subjects

A 12-year-old Chinese boy was admitted for paroxysmal pain in the left lumbar region with no apparent cause. He was born in Pingdingshan, Henan, China, to a non-consanguineous family with no family history of genetic disease. He underwent a physical examination, liver and kidney function tests, and plain abdominal X-ray imaging of the kidney, ureter, and bladder (KUB). A physical examination showed a boy with normal development and neurological behaviors; his weight was 55 kg, height 155 cm, body temperature 36.4℃, and blood pressure 99/54 mmHg. A plain KUB x-ray showed stones and hydronephrosis in his left kidney (Fig. 1); most stones were relatively small, with two bigger ones measuring 0.3 × 0.2 cm and 1 × 0.9 cm (Fig. 1). The patient underwent extracorporeal shock wave lithotripsy and the smaller stones were passed successfully. Infrared spectroscopy indicated that the stones were made of cystine (Supplementary Fig. s1). We prescribed potassium citrate and tried to maintain the patient’s urinary pH > 7.5, but failed. Six months later, residual fragments were found in his left kidney and the stone had grown (Fig. 1E). Therefore, retrograde intrarenal surgery was used to remove the stones. In addition to potassium citrate, the patient was given tiopronin and sodium bicarbonate, which maintained a urinary pH > 7.5. The patient recovered and no additional kidney stones were detected. In this family, the proband was the only one with cystinuria-induced nephrolithiasis. His parents and brother were healthy, without any kidney abnormalities. After the patients’ parents signed the informed consent form, peripheral blood was drawn from the proband and his parents for genetic testing. His brother refused to be involved in the study.

Fig. 1
figure 1

Radiographic findings in the patient. R: The right side of the body, P: The front of the body. A KUB film. B The red arrow indicates the stones in the left kidney (arrow 1); the average CT value of the arrow is 199.63 HU. C The red arrow indicates the stones in the left ureter (arrow 2). D Hydronephrosis was seen in the left kidney. E Six months later, residual fragments were found in his left kidney. The average CT value of the arrow is 752.25 HU. (The reference CT value for cystine stones is between 200 ~ 1100HU)

Full size image

Whole-exome Sequencing (WES)

Genomic DNA was extracted from peripheral blood with a GenMagBio Genomic DNA Purification Kit (Changzhou, China) and fragmented using a Bioruptor® Pico (Diagenode, Ougrée, Belgium). We checked its quality with an Agilent 2100 Bioanalyzer and 4200 TapeStation System (Agilent Technologies, Inc., Santa Clara, CA, USA) and recorded the total DNA concentration with a NanoDrop™ One Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). A VAHTS™ Universal DNA Library Prep Kit for Illumina V3 (Vazyme Biotech, Nanjing, China) was used to build a DNA library. Capture and amplification were conducted with SureSelect Human All Exon V7 (Agilent Technologies, Inc.). Finally, the HiSeq 4000 system (Illumina, Inc., San Diego, CA, USA) was used to sequence the whole exome.

Bioinformatics analysis

Trimmomatic-0.39 [2] was used to remove the sequencing adaptors and low-quality reads. The Burrow Wheeler Aligner (ver. 0.7.17-r1188) was used to align the cleaned reads to the human reference genome GRCh37 [16]. Genome Analysis Toolkit ver. 4 (GATK4) HaplotypeCaller was used to call single-nucleotide variants and small indels [6]. The variants were annotated by Vcfanno [20] based on the 1000 Genomes Project [11], Exome Aggregation Consortium (ExAC) [15], and gnomAD (v2.1) [17] annotation databases. We prioritized variants that occurred in known cystinuria-related genes SLC3A1 and SLC7A9 depending on the clinical diagnosis and phenotype information.

Sanger sequencing

Sanger sequencing of SLC3A1 (NM_000341.4) was performed on participating family members to determine whether the candidate variant co-segregated with the phenotype in the family. Two pairs of primers for c.898_905del and c.1898_1899insAT (Supplementary Table s3) were designed using NCBI Primer-BLAST and synthesized by ShangYa Bio Technology (Shanghai, China). The PCR products were purified using PCR purification kits (LifeSciences, Hangzhou, China) and sequenced on a SeqStudio Genetic Analyzer System sequencer (Applied Biosystems, Waltham, MA, USA).

Results

Laboratory findings

The patient’s initial renal and liver functions were normal, except for an elevated blood urea nitrogen (BUN) of 7.81 (normal 3.9–7.1) mmol/L and uric acid (UA) concentration of 512 (normal 150–440) μmol/L. Due to nephrolithiasis, the urine red blood cell count was 201 per μL and a bacterial infection was detected (Supplementary Table s1). After the lithotripsy and second operation, his BUN normalized and the UA concentration decreased gradually from 471 to 413 μmol/L (Table 1).

Table 1 Biochemical features of the patient before and after two treatments
Full size table

WES revealed two pathogenic variants of the SLC3A1 gene

WES generated 14.66G of original data with a Phred quality score of 30 (Q30) of 93.5%, and a mapping rate of reads to the human reference genome > 99.9%. The average sequencing depth was 100 × , with > 98% of the target sequence reaching 20 × .

We identified a novel compound heterozygous variant (c.[898_905del];[1898_1899insAT]) in the SLC3A1 gene of this Chinese patient (Supplementary Table s2). Both mutations were frameshift variants: one was a deletion variant and the other an insertion variant. The deletion variant (NM_000341.4, c.898_905del, p.Arg301AlafsTer6) came from the mother and results in the formation of alanine instead of arginine, generating a stop codon after six amino acids. According to the standards and guidelines for interpreting sequence variants by the American College of Medical Genetics and Genomics (ACMG) [24], the variant is a frameshift variant leading to a null function. The null variant of the SLC3A1 gene with an established loss of function is a known disease mechanism (PVS1 evidence); this variant has not been detected in the healthy population in the 1000 Genomes Project, ExAC, and gnomAD databases (PM2_supporting evidence). The patient’s phenotype was highly specific for the disease caused by SLC3A1 (PP4 evidence). Therefore, we classified this variant as pathogenic.

The other mutation (NM_000341.4, c.1898_1899insAT, p.Asp634LeufsTer46) was inherited from the father and is a two-nucleotide insertion (AT) that generates leucine instead of aspartic acid and a stop codon after 46 amino acids. This variant is also a frameshift mutation leading to a loss of function (PVS1 evidence). The genotype frequency in the 1000 Genomes Project, ExAC, and gnomAD databases was 0.000 (PM2_supporting evidence). The patient’s phenotype is highly specific for the disease caused by SLC3A1 (PP4 evidence). According to the ACMG guidelines, this variant (c.1898_1899insAT) was classified as pathogenic. Both variants were validated by Sanger sequencing of the proband and his parents (Fig. 2).

Fig. 2
figure 2

Pedigree and pathogenic variants in a Chinese family with cystinuria. A The family pedigree. The proband was the oldest boy in this family; his younger brother was not involved in the study. B Sanger sequencing revealed the SLC3A1 gene mutations c.898_905del, and c.1898_1899insAT

Full size image

Genotype and phenotype association

Some studies have found no obvious pattern between phenotype and genotype in cystinuria, which involves at least two genes [9, 18, 23]. Cystinuria caused by mutations in SLC3A1 is classified as type A cystinuria, while mutations in SLC7A9 cause type B. Most patients with cystinuria carry two mutations (AA, BB, or AB) [8]. Mutational analyses revealed biallelic SLC3A1 mutations (AA genotype) in our proband. We observed that an older age of onset age in patients with the AA genotype. The proband in this study was first diagnosed with cystine stones at the age of 12 years, which is significantly older than the mean reported age of 5.59 years for cystinuria urolithiasis onset [1, 29]. The two variants carried by the proband were truncating mutations, which usually lead to a more severe phenotype. In our study, the BUN (mmol/L), creatinine (Cre; μmol/L), and UA (μmol/L) concentrations of the proband were all higher than the average values found in other study (Table 2) [27].

Table 2 Clinical comparison between our patient (WCZ) and 13 Chinese cystinuria patients (13C)
Full size table

Discussion

Cystinuria is a genetic disorder with high population heterogeneity globally. In a study of 13 Chinese cystinuria patients, Shen et al. [27] reported an average age at the first confirmed stone of 6.8 (range 0.8–16) years. Our study updates the clinical and genetic spectrum of cystinuria, and we have comprehensively studied the genetic mutations in SLC3A1 that occurred in a Chinese family in which only a 12-year-old boy with thpe A cystinuria. We identified two heterozygous mutations (c.898_905del and c.1898_1899insAT) in SLC3A1; one of which has not been described in other cystinuria patients.

As of 2010, 133 mutations have been identified in SLC3A1 [4] and the number is continuing to increase, but most identified to date are missense mutations [13, 27]. The deletion we found was in exon 5 and the AT insertion was in exon 10. Both variants alter the amino acids, leading to premature termination of protein translation(c.898_905del generating a stop codon after 6 amino acids and c.1898_1899insAT generating a stop codon after 46 amino acids) and disruption of gene function, and could be important in the rBAT coding protein functions, which is thought to be crucial in dibasic amino acid transepithelial transporters. The insertion variant c.1898_1899insAT identified in our case was reported in a Japanese patient in 2000 [9], the same study described four other mutations: three missense mutations (L346P, I445T, and C673R) and one deletion (1820delT). In a cohort of Japanese cystinuria patients, mutations in SLC3A1 were found in around 10%, most of which were missense mutations [29]. In a study of 13 pediatric patients with cystine stones, Shen et al. [27] found 17 missense mutations in SLC3A1 and SLC7A9; more than 90% (17/18) of these were missense mutations. It is uncommon to find two frameshift mutations in SLC3A1 in Chinese cystinuria patients, and the deletion and insertion mutations in our study both caused frameshifts, but not a stop codon in the open reading frame. Our findings suggest that the genetic pattern of cystinuria is heterogeneous not only among different groups of people, but even within a local population.

Several researchers have examined the correlation between the clinical and genetic characteristics of cystinuria [13, 23,  27]. In our cystinuria patient, the BUN (mmol/L), creatinine (Cre; μmol/L), and UA (μmol/L) levels were all higher than the average values found in other studies (Table 2), we suspect that this may be related to the variant type of SLC3A1 gene in the patient, as frameshift variants lead to more severe phenotypes than missense ones; after traditional treatments, including extracorporeal shock wave lithotripsy, alkalize urine and Oral medication treatment, these serum values normalized. Besides kidney stones and hydronephrosis, no other metabolic disorders were found in our patient, while in a study of 13 Chinese cystinuria patients, Shen et al. [27] found hyperoxaluria in 61.5%, hypercalciuria in 46.2%, hypocitraturia in 30.8%, and hyperuricosuria in 7.7%. Due to the limited number of cystinuria cases reported and analyzed in China, the correlation between SLC3A1 mutation pattern and clinical presentation remains unclear. Rhodes et al. [23] found no association between the genotype and clinical course in cystinuria patients in the United Kingdom. Since cystinuria is caused by mutations in multiple genes, this makes it difficult to define the genetic pattern and etiology of cystinuria. Besides SLC3A1 and SLC7A9, other genes can lead to similar presentations with cystine stones, such as SLC7A13 and PBX1 [18, 22]. Due to the complex genetic pattern of cystinuria, it is difficult to define a universal correlation between the phenotype and genotype. This study updates the clinical and mutational spectrum of cystinuria, especially in Chinese cystinuria patients.Therefore, broadening the spectrum of clinical and genetic variants could be crucial for the diagnosis and therapy of cystinuria in the long-term.