PDF
Citation
Print
INTRODUCTION
Citrin deficiency is an autosomal recessive disorder resulting from pathogenic variants in SLC25A13, located on chromosome 7q21.3 [1]. Citrin deficiency presents with different age-dependent clinical phenotypes, including neonatal intrahepatic cholestasis caused by citrin deficiency (NICCD, OMIM #605814) and adult-onset type II citrullinemia (CTLN2, OMIM #603471) [2]. NICCD typically manifests in the first few months of life with symptoms such as intrahepatic cholestasis, multiple aminoacidemia, and fatty liver. Although NICCD often resolves within the first year, some patients may develop CTLN2, characterized by recurring neuropsychiatric symptoms associated with hyperammonemia, later in life [2]. The clinical phenotype varies widely, ranging from nearly asymptomatic to severe cases involving coma, brain edema, and liver failure. Prompt diagnosis and appropriate management are important for a favorable long-term prognosis [3].
The prevalence of heterozygous carriers of pathogenic SLC25A13 variants is high in East Asian populations, particularly in Japan, with a frequency of 1/42 to 1/69 [4–6]. Data from a recent large-scale newborn screening study in China revealed a disease-associated allele frequency of approximately 1/45 [7], similar to that in Japan and higher than those of Thai and Korean populations (1/90 [8] and 1/112 [4], respectively). In East Asian populations, a few recurrent pathogenic SLC25A13 variants exist. Among Chinese patients with NICCD, variants such as c.852_855del, c.1638_1660dup, c.615+5G>A, and IVS16ins3kb account for >80% of all cases [3]. In Japan, 11 pathogenic variants, including c.852_855del, c.1177+1G>A, c.1311+1G>A, and c.674C>A, constitute approximately 95% of all cases [5]. Studies on the Korean population are limited. Regarding the only previous study primarily targeting Japanese pathogenic variants, Lu, et al. [4] reported a carrier frequency of 1/112 based on 2,455 Korean control samples. Notably, some common high-frequency variants in Korean patients, such as a 3-kilobase insertion in intron 16 of SLC25A13 (IVS16ins3kb) [9], were excluded, and the overall sample size was small.
The IVS16ins3kb variant is a large insertion of 2,677 nucleotides processed from NUS1 (formerly C6orf68) into intron 16 in SLC25A13. IVS16ins3kb was identified using long-range PCR amplification and gel electrophoresis or sequencing, both of which are time- and labor-intensive [6]. The availability of genomic databases, such as the Genome Aggregation Database (gnomAD) [10], provides an opportunity to study the carrier frequency in large, healthy populations [11]. However, because the corresponding databases rely on general callers for detecting copy number or structural variants, such as Manta [12] in gnomAD, accurately determining the frequency of the large transposon insertion variant may be challenging [13].
Estimating mutation carrier frequency and disease prevalence is essential for genetic counseling and public health planning, which enables the identification of individuals at genetic risks for recessive monogenic genetic disorders and helps reduce the disease prevalence in the neonatal population [14–16]. We analyzed genomic population databases to evaluate the burden of known pathogenic SLC25A13 variants and the prevalence of citrin deficiency in Korean and other East Asian populations. In addition, the IVS16ins3kb variant in SLC25A13 was identified using an in silico grep method to search the mutant chimeric sequence in in-house whole-genome sequencing (WGS) data from 681 Korean individuals.
MATERIALS AND METHODS
Literature search and compilation of SLC25A13 pathogenic variants
This cross-sectional study was conducted at Samsung Medical Center, Seoul, Korea, to estimate the carrier frequency and prevalence of citrin deficiency in East Asians and Koreans. First, we conducted a comprehensive literature search on PubMed using the keyword “citrin deficiency” on August 19, 2024. Studies were included based on the following criteria: 1) included ≥20 East Asian patients; 2) clearly described SLC25A13 pathogenic variants in the patients; 3) provided individual patient allele frequencies for the entire patient cohort; and 4) included only unrelated probands. From the six studies selected, we compiled a list of all pathogenic variants and cross-referenced these with genomic population and variant databases.
Data integration with population and variant databases
We utilized gnomAD Exomes v2.1.1 [10], Korean Variant Archive (KOVA) [17], and Tohoku Medical Megabank Organization (ToMMo)-8.3KJPN as population databases [18]. gnomAD is a global genomic database that contained 125,748 exomes at the time of the study, including 9,197 from East Asian populations. As a reference for genetic variants within the Korean population, we used KOVA, which included data from 5,305 healthy Korean individuals, comprising WGS data for 1,896 individuals, and whole-exome sequencing data from 3,409 individuals. For the Japanese population, we used ToMMo-8.3KJPN [18], which contained allele and genotype frequency WGS data for 8,380 Japanese individuals. All variants were described using the MANE Select reference transcript of SLC25A13 (NM_014251.3) and the nomenclature of the Human Genome Variation Society [19]. The SLC25A13 pathogenic variants found through the literature search were verified using the Human Gene Mutation Database (HGMD Professional 2024.2) [20].
Citrin deficiency carrier frequency and prevalence estimations
We calculated the frequencies of individuals carrying citrin deficiency in the Korean and other East Asian subpopulations utilizing genomic population databases. To estimate the prevalence of citrin deficiency, we used the Hardy–Weinberg equilibrium principle (1=p2+2pq+q2), where p represents the major allele (non-disease), and q is the minor allele (disease). Given the approximation of the major allele p being close to 1, carriers are indicated by 2pq, and the disease state is indicated by q2. By deriving the q value from carrier frequency data obtained from genomic population databases, we estimated the disease prevalence (q2).
Detecting IVS16ins3kb in SLC25A13 via WGS
To investigate the IVS16ins3kb variant frequency, we used the Linux grep command to search for the junction point in the reference sequence of SLC25A13 between intron 16 and the inserted sequence of NUS1 in compressed FASTQ files of previously generated WGS data. The WGS data were from 681 consecutive, unrelated Korean individuals suspected of Mendelian disorders other than inborn errors of metabolism who underwent WGS as part of a Korean National Project (Bio Big Data) in 2021. These data were of sufficient quality for in silico analysis, and no individuals were excluded. We designed a 26-nucleotide-long chimeric sequence containing an insert boundary, 5′-AAA TGGAGAAATCGGGGGGCGGGGCT-3′ (SLC25A13-NUS1) (Fig. 1), in both the forward and reverse directions. The forward and reverse sequences were specific for the mutant allele and did not match any other region of the genome. The grep commands returned the number of reads containing the matching sequences in each file. After this screening, the variant allele frequency (VAF) was calculated as the percent of mutant alleles to the total number of alleles. In addition to the commands used for the chimeric sequence, other commands were those employed by Won, et al. [21] to search for Alu insertions in RP1. When IVS16ins3kb was suspected in the grep screening results, soft-clipped sequences from the Binary Alignment Map (BAM) file were visually inspected using the Integrative Genomics Viewer (IGV) to determine whether the observed sequence was identical to the known IVS16ins3kb retrotransposon junction.
The Institutional Review Board (IRB) of Samsung Medical Center approved this WGS study (IRB No.: SMC 2020-10-042) and granted an exemption to review the public database analysis study (IRB No.: SMC 2024-10-117). The study protocol adhered to the tenets of the Declaration of Helsinki. Informed consent for genetic testing and using biological and related clinical data for research was obtained from all individuals investigated. Images, video, or details that could identify the individuals were not used.
Statistical analysis
Comparisons between gnomAD exomes and KOVA data for Korean individuals were calculated using the chi-square test. The calculations were performed with VassarStats (http://vassarstats.net/), and 95% confidence intervals [CIs] were calculated for each value. P was based on two-sided comparisons, and P<0.05 was considered to reflect a statistically significant difference.
RESULTS
Frequencies of common pathogenic SLC25A13 variants in East Asian patients with citrin deficiency
On August 19, 2024, we retrieved 289 publications spanning February 2002 to July 2024 from PubMed using the search term “citrin deficiency.” Among these publications, six provided detailed descriptions of pathogenic SLC25A13 variants for all patients, enabling variant frequency determinations [4, 6, 9, 22–24]. The 23 pathogenic variants retrieved from the literature are summarized in Table 1. Because three of the six publications were from the same research group [4, 6, 23], only the most recent publication [6] is included in Table 1. The most frequent pathogenic SLC25A13 variant was c.852_855del, p.(Met285Profs*2), with an allele frequency of 30.3%–58.4% across the studies. Other common pathogenic variants included c.1177+1G>A, IVS16ins3kb, c.1638_1660dup, and c.1311+1G>A, with varying frequencies across East Asian populations.
Allele and carrier frequencies of common pathogenic SLC25A13 variants in Korean population databases
The allele count and carrier frequency for each pathogenic SLC25A13 variant are presented in Table 2. Among 7,214 Koreans (gnomAD Koreans, N=1,909 and KOVA, N=5,305), pathogenic alleles were identified in 84 individuals, corresponding to a carrier frequency of 1 in 86 (95% confidence interval [CI], 1/110–1/71). The most frequent pathogenic SLC25A13 variant in this population was c.1177+1G>A, p.?, with an allele frequency of 0.21% (95% CI, 0.14%–0.29%) and a carrier frequency of 1 in 233 (95% CI, 1/360–1/173). The next most common variant was c.852_855del, p.(Met285Profs*2), with an allele frequency of 0.12% (95% CI, 0.07%–0.18%) and a carrier frequency of 1 in 401 (95% CI, 1/745–1/275).
Carrier frequency of IVS16ins3kb in SLC25A13 in the Korean population based on WGS
Among the WGS samples from 681 consecutive, unrelated Korean individuals, the IVS16ins3kb variant of SLC25A13 was heterozygous in three individuals, with an allele frequency of 0.22% (3/1,362) and an estimated carrier frequency of 1 in 228 (Table 2). This frequency is similar to the estimated carrier frequency of 1 in 233 of the most common variant, c.1177+1G>A. The three individuals in whom the IVS16ins3kb variant was detected included one female patient with breast cancer, one male child with global developmental delay, and one male patient suspected of having Charcot–Marie–Tooth disease, none of whom were suspected of having a metabolic disorder. The IVS16ins3kb variant discovered with the in silico grep method was confirmed through visual inspection using IGV, and characteristic nucleotide sequences were identified (Fig. 1).
Comparisons of the citrin deficiency carrier frequency and prevalence in East Asian and Korean populations
The overall carrier frequency in East Asian populations in the gnomAD and ToMMo-8.3KJPN databases was 1 in 62 (95% CI, 1/70–1/55; Table 3). For the Korean gnomAD exomes, the carrier frequency was 1 in 107, whereas the Japanese and other East Asian populations exhibited higher carrier frequencies of 1 in 100 and 1 in 40, respectively. The prevalence of citrin deficiency varied across populations, being 1 in 15,083 in East Asian populations, 1 in 39,810 in the Japanese population, and 1 in 44,991 (based on gnomAD data) in the Korean population. However, it should be noted that the estimated prevalence was 1 in 25,843 based on KOVA data. When integrating gnomAD and KOVA data from 7,214 individuals, the carrier frequency of citrin deficiency in the Korean population was 1.16%, with an estimated prevalence of 1 in 29,502. To compare the results with the gnomAD and ToMMo-8.3KJPN genomic population databases, carrier frequency and disease prevalence estimates were calculated using the allele count of SLC25A13 IVS16ins3kb.
DISCUSSION
To our knowledge, this was the first study to investigate the carrier frequency and prevalence of citrin deficiency using genomic population databases. The results expand the limited data on the prevalence and distribution of pathogenic SLC25A13 variants among East Asian populations, including Koreans. The citrin deficiency carrier frequency in Koreans estimated in this study (1 in 86) was higher than the previously reported frequency of 1 in 112 [4].
The most common pathogenic SLC25A13 variant among Koreans was c.1177+1G>A, in contrast to c.852_855del, which was reported as the most prevalent pathogenic variant in previous East Asian studies [6, 9, 22, 24]. SLC25A13 c.1177+1G>A occurs in a canonical donor splice site and has been confirmed to disrupt this splice site by inducing exon 11 skipping (r.1019_1177del) based on RNA analysis [25]. c.1177+1G>A occurs as a homozygous or compound heterozygous variant in patients with citrin deficiency [25–28]. c.852_855del, identified as the second most common variant in this study and predicted to result in a frameshift (p.Met285Profs*2) and premature translational termination signal in SLC25A13, has been identified in multiple individuals affected by citrin deficiency [1, 26, 29–32]. In the gnomAD database, c.1177+1G>A and c.852_855del are found exclusively within East Asian populations, not in European, American, or South Asian populations.
IVS16ins3kb is a common variant in Chinese [22, 24, 33] and Japanese [6] individuals with citrin deficiency. The insertion results in 2,677 additional nucleotides in intron 16 of SLC25A13 and a premature termination codon in the inserted sequence (p.Ala584Valfs*2) [6]. This variant was identified at a high allele frequency of 33.3% in a study involving 33 unrelated Korean patients with citrin deficiency [9]. In our analysis of in-house WGS data from 681 unrelated individuals, the variant allele was detected in three cases (0.22%, Table 2). The reason for the frequency difference between the present and previous studies remains unclear. The previous study [9] was focused on patients diagnosed as having citrin deficiency, whose characteristics differ from those of the individuals in our study, and included only 33 patients, which is insufficient to draw definitive conclusions. Further research with a larger patient cohort is required to better clarify the discrepancy. We found an IVS16ins3kb variant carrier frequency of 1 in 228, which is similar to that of the most common pathogenic variant, c.1177+1G>A. Clinical laboratories should implement strategies to prevent overlooking this variant during genetic diagnostic testing for citrin deficiency.
This study had some limitations. First, we focused on 23 pathogenic SLC25A13 variants selected through a literature review, which may have led to an underestimation of the carrier frequency. In addition, the IVS16ins3kb variant frequency analysis results based on the grep command were not validated for the 681 patients included in the WGS dataset.
We utilized another dataset comprising 234 unrelated Korean individuals (parents of pediatric patients suspected of having Mendelian disorders other than inborn errors of metabolism) for whom nucleic acids were available. Using this dataset, we performed in silico analysis using the grep command, ultimately identifying the IVS16ins3kb variant in two unrelated individuals. The frequency of the variant was 0.43% (2/468, 95% CI, 0.12%–1.55%), with no significant difference in allele frequencies between the two WGS datasets (P=0.823). Two specimens identified as positive in silico were confirmed using long-range PCR, nested PCR, and Sanger sequencing, and the results indicated that the WGS analysis approach used in this study was appropriate. The PCR amplification and Sanger sequencing results obtained to validate the presence of IVS16ins3kb (suspected based on grep analysis and IGV inspection) are provided in Supplemental Data Table S1 and Figs. S1 and S2. Despite the above limitations, this was the most extensive study to analyze SLC25A13 in East Asians to date, particularly Koreans, providing a more accurate prediction of the carrier frequency and prevalence of citrin deficiency among East Asians and Koreans. Considering the recent progress in surveillance and treatments for citrin deficiency [34–36], identifying the carrier frequency and prevalence is crucial for effective management and intervention.
In conclusion, the carrier frequency of pathogenic SLC25A13 variants responsible for citrin deficiency is relatively high among Koreans, and the variant spectra differ by ethnicity. These data can serve as a valuable resource for clinical geneticists to inform population-adjusted genetic screening programs, particularly in understudied ethnic groups.
XML Download