Mutation analysis of the ATP7B gene and genotype-phenotype correlation in Chinese patients with Wilson disease

doi:10.21203/rs.3.rs-135373/v1

Download PDF

Research Article

Mutation analysis of the ATP7B gene and genotype-phenotype correlation in Chinese patients with Wilson disease

https://doi.org/10.21203/rs.3.rs-135373/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

AIM To discover the novel ATP7B mutations in 103 southern Chinese patients with Wilson disease (WD), and to determine the spectrum and frequency of mutations in the ATP7B gene and genotype-phenotype correlation in a large-scale sample of Chinese WD patients.

Methods One hundred three WD patients from 101 unrelated families in southern China were enrolled in this study. Genomic DNA was extracted from the peripheral blood. Direct sequencing of all 21 exons within ATP7B was performed. Subsequently, an extensive study of the overall spectrum and frequency of ATP7B mutations and genotype-phenotype correlation was performed in all Chinese patients eligible from the literature, combined with the current southern group.

Results In 103 patients with WD, we identified 48 different mutations (42 missense mutations, 4 nonsense mutations and 2 frameshifts). Of these, 7 mutations had not been previously reported: 1510_1511insA, 2075T>C (Leu692Pro), 2233C>A (Leu745Met), 3209C>G (Pro1070Arg),3677C>T (Thr1226Ile), 3793G>T (Val1265Leu) and 3824T>C (Leu1275Ser). The 2333G>T (Arg778 Leu) at exon 8, was the most common mutation with an allelic frequency of 18.8%, followed by 2975C>T (Pro992Leu) at exon 13, with an allelic frequency of 13.4%. In the comprehensive study, 233 distinct mutations were identified, including 154 missense mutations, 23 nonsense mutations and 56 frameshifts. Eighty-five variants were identified as novel mutations. The 2333G>T (Arg778 Leu) and 2975C>T (Pro992Leu) were the most common mutations, with allelic frequencies of 28.6% and 13.0%, respectively. Exons 8, 12, 13, 16 and 18 were recognised as hot spot exons. Phenotype-genotype correlation analysis suggested that 2333G>T (Arg778 Leu) was significantly associated with lower levels of serum ceruloplasmin (P=0.034). 2975C>T (Pro992Leu) was correlated with earlier age of disease onset (P=0.002). Additionally, we found that the 3809A>G (Asn1270Ser) mutation significantly indicated younger onset age (P=0.012), and the 3884C>T (Ala1295Val) mutation at exon 18 was significantly associated with hepatic presentation (P=0.048). Moreover, the patients with mixed presentation displayed the initial WD features at an older onset age than the groups with either liver disease or neurological presentation (P=0.039, P=0.015, respectively). No significant difference was observed in the presence of KF rings among the three groups with different clinical manifestations.

Conclusion In this study, we identified seven novel mutations in 103 WD patients from the southern part of China, which could enrich the previously established mutational spectrum of the ATP7B gene. Moreover, we tapped into a large-scale study of a Chinese WD cohort to characterise the overall phenotypic and genotypic spectra and assess the association between genotype and phenotype, which enhances the current knowledge about the population genetics of WD in China.

Gastroenterology & Hepatology

Wilson disease

ATP7B gene

mutations

correlation

Chinese

Wilson disease (WD), also known as hepatolenticular degeneration (HLD), is an autosomal recessive inherited disorder of copper metabolism, resulting from pathogenic mutations in the ATP7B gene. It is characterised by deficient incorporation of copper into ceruloplasmin and decreased biliary copper excretion, leading to excessive copper accumulation, primarily in the liver, brain and eyes. The toxic deposition of copper in the body results in highly heterogeneous clinical presentations, such as liver impairment, neurological disturbance and/or other derangements.

The age of onset ranged from 1 to 72 years ^[1],[2]. Most of the existing literature regards the worldwide prevalence of WD as approximately 1 in 30,000 to 1 in 100,000, with an estimated carrier rate of 1 in 90 ^[3]. However, in Korea, where WD is one of the most common inherited metabolic disorders, the carrier frequency and incidence of WD are estimated to be 1 in 88.2 and 1 in 30,778, respectively ^[4]. In Latvia, the estimated prevalence of WD is 1 in 24,000 cases ^[5]. Screening of WD in the UK population suggests that the frequency of individuals predicted to carry two mutant pathogenic ATP alleles is 1 in 7026, which is considerably higher than the typically reported prevalence ^[6]. Ceruloplasmin-based screening for WD in the population of Japan suggested a frequency as high as 1 in 1,500 ^[7].

WD is caused by mutations in the ATP7B gene, discovered in 1993, that encodes a copper transporting P-type ATPase containing 1,465 amino acids ^[8–10]. It is located on chromosome 13q14.3 and consists of 21 exons and 20 introns. Genetic disorders of the ATP7B gene disrupt the synthesis and function of the ATP7B protein, and further impair the copper excretion pathway, leading to the abnormal deposition of copper in the body. Currently, there are records of at least 800 distinct disease-causing mutations in the ATP7B gene, characterised by a few hotspot mutations and a wide spectrum of rare mutations, with obvious ethnic and regional differences.

Traditionally, the diagnosis of WD mainly depends upon clinical manifestations and conventional biochemical indicators, including elevated 24-h urinary copper, low serum ceruloplasmin and increased hepatic copper content. However, biochemical tests can be misleading, making WD diagnosis difficult ^{[11, 12]}. Hence, molecular detection is warranted for establishing a precise and decisive diagnosis of WD, particularly in asymptomatic patients and siblings of the proband in a WD-affected family. Previous studies on mutations in Chinese WD patients have been based on diverse genetic detection methods with different detection rates. Therefore, a consensus has not yet been reached regarding the spectrum and frequency of mutations in the ATP7B gene in the Chinese WD population. On the other hand, previous studies have failed to identify WD mutations in a significant number of clinically diagnosed cases, resulting in incomplete understanding of the patterns and frequencies of hot spots in the ATP7B gene and controversial correlations between genotypes and phenotypes in the Chinese population with WD.

Here, to obtain the best identification rate and accuracy, we used direct sequencing to detect the WD mutations. This method is considered as the gold standard to identify mutations in molecular genetics ^[13] and has been documented as having a high detection rate and accuracy ^[14]. In our study, we first analysed the genotypic profile and determined the novel mutations of 103 cases with WD from the south of China by means of direct sequencing. Subsequently, we conducted a comprehensive literature search for available studies on WD mutations to identify the overall spectrum of ATP7B mutations and the mutation hotspots observed in the Chinese WD population, and to explore the potential correlation between genotype and phenotype.

To the best of our knowledge, this is the first study to undertake a comprehensive literature study to identify the molecular genetic features and correlations with clinical phenotypes in a large-scale sample of Chinese WD patients. Understanding the genotypic pattern of WD in China could pave the way for offering diagnostic mutational analysis of WD in the future. Our genetic investigation of WD patients from the southern part of China could extend the previously established spectrum of ATP7B mutations, and the comprehensive mutation analysis would enhance the current knowledge about the genotypic and phenotypic profiles of WD in China and provides insights into the association between genotype and phenotype in the Chinese population with WD.

Patients

A total of 103 WD patients (66 males and 37 females) from 101 independent families were enrolled in this study, with a mean age at presentation of 18.2 ± 12.9 years. They came from different parts of China (40.9% subjects from Hunan Province, 11.3% from Jiangxi Province, 6.8% from Hubei Province and 4.5% from the northern China). All the patients were identified and diagnosed at the Second Xiangya Hospital in Hunan Province between January 2014 and January 2020. Diagnosis of WD was based on a combination of characteristic clinical symptoms, Kayser-Fleischer (KF) rings, abnormal brain magnetic resonance imaging and biochemical parameters, including low serum ceruloplasmin (< 0.2 g/L), increased urinary copper excretion (> 100 µg/24 h) and high hepatic copper content (> 250 µg/g dry weight). The control group consisted of 37 subjects with neither family history nor clinical features of WD. This study was approved by the local ethics committee, and informed consent was obtained from all recruited subjects or from their parents.

DNA Extraction And Amplification

The peripheral venous blood was obtained from the WD patients and controls. Genomic DNA was extracted with a Genomic DNA Purification Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. Genomic DNA was stored at -20℃. ALL 21 exons of the WD gene were amplified by polymerase chain reaction (PCR). The amplification was conducted as follows: pre-denaturation at 94℃ for 5 min; 35 cycles of denaturation at 94℃for 45 s, annealing at 55℃ for 45 s (except exon 2.2 and exon 17), or annealing at 51℃ for 45 s (to exon 2.2), or annealing at 61℃for 45 s (to exon17) and extension at 72℃ for 1 min, with a final elongation at 72℃ for 10 min.

DNA Sequencing

Direct sequencing of the amplified PCR products was performed to detect the disease-causing mutations and single nucleotide polymorphisms at Sangon Biotech (Shanghai, China), and the sequenced products were compared with published normal sequences deposited in GenBank using a blast search program. Samples showing abnormal results were subjected to bidirectional sequencing.

Systematic Review

A comprehensive literature search of PubMed was carried out for articles published from inception until May 2020. Index terms used were Wilson Disease (Title/Abstract) or hepatolenticular degeneration (Title/Abstract). Papers that described the mutations in the WD gene were handpicked by examining online abstracts. Subsequently, further selection was conducted using the full-length publications. Eligibility of the articles was based on the inclusion criteria: (1) observational studies published as original articles that focused on the genetic analysis of WD patients in China; (2) patient cohorts including both paediatric (< 18 years old) and adult (≥ 18 years old) patients; (3) the 21 exons of the ATP7B gene were amplified by PCR; and (4) direct sequencing of the PCR products was performed. We excluded the papers in which the subjects failed to satisfy the foregoing criteria, and which were not written in English. To avoid reporting bias, we manually collected additional relevant studies listed as references of these retrieved articles. This systematic review was conducted in accordance with the PRISMA (preferred reporting items for systematic reviews and meta-analysis) guideline.

A standardised data collection form was utilised to derive the following information: first author, country of origin, year of publication, study design and methods, studied population, gender, age at onset of presentations, WD features (clinical manifestations and biochemical markers), and mutations in ATP7B gene. A systematic analysis for the spectrum of the ATP7B mutations in China was performed on all Chinese WD patients available from qualifying literature, combined with the current southern group. An analysis for the genotype and phenotype correlation in the large-scale Chinese WD cohorts was also performed.

Statistical analysis

Allele and genotype frequencies were calculated by the direct count. Statistical analyses were performed using SPSS for Windows(Version 20.0, SPSS, Inc., Chicago, IL, USA). The distributed analysis of numeric variables was carried out at first. Normally distributed variables were expressed as mean and standard deviation and were compared between groups using the Scheffe test. Variables that were not normally distributed have been presented as median and interquartile range (IQR) and were compared between groups using the Mann-Whitney U test. Frequencies of qualitative variables were compared between groups by the chi-squared (X) test. A P value less than 0.05 was considered statistically significant. Bonferroni correction was applied in case of pairwise comparisons out of larger groups.

Mutation Analysis In 103 WD Patients

Among the 103 WD patients derived from 101 unrelated families, we have identified 48 distinct mutations, including 42 missense mutations, 4 nonsense mutations and 2 frameshifts, as presented in Table 1. None of these 48 mutations was detected among 68 alleles in healthy individuals. According to the records in the WD mutation database of the University of Alberta (http://www.wilsondisease.med.ualberta.ca/database.asp) and the Human Gene Mutation Database (HGMD) professional (http://www.hgmd.cf.ac.uk), this is the first time that the following mutations have been reported: 1510_1511insA, 2075T > C (Leu692Pro), 2233C > A (Leu745Met), 3209C > G (Pro1070Arg), 3677C > T (Thr1226Ile), 3793G > T (Val1265Leu) and 3824T > C (Leu1275Ser). All novel missense variants were tested for the possibility of being pathogenic in nature using PolyPhen-2 software (http://genetics.bwh.harvard.edu/pph2/), with the results indicating that they were significantly more likely to alter protein function, as shown in Table 1. Moreover, 15 known polymorphisms that do not disrupt ATP7B gene function were detected (data is available in Table 2). Mutation analysis of the ATP7B gene by direct sequencing of 21 exons yielded a mutation detection rate of 80.7% (163/202). There were eight patients having no detectable mutation, indicating that the remaining mutations were possibly located in the intron or the regulator. In the present study, the 2333G > T (Arg778Leu) at exon 8 was the most frequent mutation, with an allelic frequency of 18.8% (38/202), followed by the2975C > T (Pro992Leu) at exon 13, with an allelic frequency of 13.4% (27/202).

Table 1

Spectrum of mutations in the ATP7B gene of a large-scale sample of Chinese patients with Wilson’s disease
Exon	Nucleotide mutation	Amino acid change	Mutation type	Domain	Allelic frequency in southern cohort	Allelic frequency in large-scale cohort	PolyPhen-2
Exon	Nucleotide mutation	Amino acid change	Mutation type	Domain	Allelic frequency in southern cohort	Allelic frequency in large-scale cohort	Score	Prediction
2	121A > G	Asn41Asp	Missense	before Cu1	0	0.04%(1/2,604)	0.029	Benign
2	254G > T	Gly85Val	Missense	Cu1	0	0.04%(1/2,604)
2	268_271DEL	Lys90PhefsX10	Deletion	Cu1	0	0.04%(1/2,604)	NA	NA
2	287A > G	Asp96Gly	Missense	Cu1	0	0.31%(8/2,604)
2	314C > A	Ser105X	Nonsense	Cu1	0	0.15%(4/2,604)
2	367DELG	Ala123ProfsX30	Deletion	Cu1	0	0.04%(1/2604)	NA	NA
2	433G > T	Val145Phe	Missense	Cu2	0	0.08%(2/2,604)
2	523INSA	NA	Insertion	Cu2	0	0.08%(2/2,604)	NA	NA
2	525DupA	Val176SerfsX28	Insertion	Cu2	0.50%(1/202)	1.11%(29/2,604)
2	588C > A	Asp196Glu	Missense	Cu2	0.99%(2/202)	0.35%(9/2,604)
2	592A > G	Arg198Gly	Missense	Cu2	0	0.04%(1/2,604)	0.735	Possibly damaging
2	685insA	NA	Insertion	Cu3	0	0.04%(1/2,604)
2	695DELC	Pro232GlnfsX30	Deletion	Cu3	0	0.04%(1/2,604)	NA	NA
2	748G > A	Gly250Arg	Missense	Cu3	0	0.04%(1/2,604)	0.004	Benign
2	813DELC	Cys271TrpfsX3	Deletion	Cu3	0	0.04%(1/2,604)
2	898_902DEL AAGTA	NA	Deletion	Cu3	0	0.04%(1/2,604)	NA	NA
2	994G > T	Glu332X	Nonsense	bet Cu3/Cu4	0	0.42%(11/2,604)
2	1057DELC	Gln353ArgfsX10	Deletion	bet Cu3/Cu4	0	0.04%(1/2,604)	NA	NA
2	1162C > T	GIn388X	Nonsense	Cu4	0	0.04%(1/2,604)
2	1168A > G	Ile390Val	Missense	Cu4	0	0.58%(15/2,604)	0.019	Benign
2	EX2 DEL	NA	Deletion	Cu4	0	0.08%(2/2,604)
3	1366G > C	Val456Leu	Missense	bet Cu4/Cu5	0	0.08%(2/2,604)
3	1403_1416DEL	Ala468GlyfsX33	Deletion	bet Cu4/Cu5	0	0.04%(1/2,604)	NA	NA
3	1426G > A	Ala476Thr	Missense	bet Cu4/Cu5	0	0.04%(1/2,604)	0.002	Benign
3	1448_1455DEL GAGCAGTG	Arg483SerfsX20	Deletion	Cu5	0	0.04%(1/2,604)
3	1449_1456DEL	Arg483SerfsX20	Deletion	Cu5	0	0.08%(2/2,604)	NA	NA
3	1470C > A	Cys490X	Nonsense	Cu5	0.50%(1/202)	0.38%(10/2,604)
3	1492A > T	Thr498Ser	Missense	Cu5	0	0.04%(1/2,604)
3	1510_1511INSA	NA	Insertion	Cu5	0.50%(1/202)	0.04%(1/2,604)	NA	NA
3	1516_1517DELAT	NA	Deletion	Cu5	0	0.04%(1/2,604)
3	1531C > T	Gln511X	Nonsense	Cu5	0	1.38%(36/2,604)
4	1544G > T	Gly515Val	Missense	Cu5	0	0.04%(1/2,604)
4	1545DELT	Gly515GlyfsX9	Deletion	Cu5	0	0.04%(1/2,604)	NA	NA
4	1552_1553DELTC	Ser518ArgfsX15	Deletion	Cu5	0	0.04%(1/2,604)	NA	NA
4	1639C > T	Gln547X	Nonsense	Cu5	0	0.04%(1/2,604)
5	1745_1746DEL TA	NA	Deletion	Cu6	0	0.04%(1/2,604)	NA	NA
5	1760C > T	Thr587Met	Missense	Cu6	0	0.04%(1/2,604)	0.005	Benign
5	1782T > A	Tyr594X	Nonsense	Cu6	0	0.04%(1/2,604)	NA	NA
5	1802DELC	NA	Deletion	Cu6	0	0.04%(1/2,604)	NA	NA
5	1803DELC	Ser602AlafsX46	Deletion	Cu6	0	0.04%(1/2,604)
5	1817T > G	Val606Gly	Missense	Cu6	0	0.08%(2/2,604)
5	1820DUPA	Phe608ValfsX2	Insertion	Cu6	0	0.08%(2/2,604)
5	1831G > A	Glu611Lys	Missense	Cu6	0	0.04%(1/2,604)
5	1846C > T	Arg616Trp	Missense	Cu6	0	0.04%(1/2,604)
6	1875_1876INS AATT	NA	Insertion	Cu6	0	0.04%(1/2,604)
6	1925A > G	Asp642Gly	Missense	bet Cu6/TM1	0	0.04%(1/2604)	0.899	Possibly damaging
7	1950G > A	Trp650X	Nonsense	bet Cu6/TM1	0	0.04%(1/2,604)
7	1994T > G	Met665Arg	Missense	TM1	0	0.04%(1/2,604)	0.836	Possibly damaging
7	2012_2013INS ATAT	NA	Insertion	TM1	0	0.04%(1/2,604)	NA	NA
7	2038C > T	Gln680X	Nonsense	bet TM1/TM2	0.50%(1/202)	0.12%(3/2,604)
7	2043DELC	Ser681SerfsX15	Deletion	bet TM1/TM2	0	0.04%(1/2,604)	NA	NA
7	2075T > C	Leu692Pro	Missense	TM2	0.50%(1/202)	0.08%(2/2,604)	0.996	Probably damaging
7	2078C > G	Ser693Cys	Missense	TM2	0.50%(1/202)	0.15%(4/2,604)
7	2097_2099DELCTT	Phe700del	Deletion	TM2	0	0.08%(2/2,604)
7	2120A > G	Gln707Arg	Missense	TM2	0	0.04%(1/2,604)
8	2128G > A	Gly710Ser	Missense	TM2	0	0.12%(3/2,604)
8	2145C > A	Tyr715X	Nonsense	TM2	1.49%(3/202)	0.12%(3/2,604)
8	2156A > G	Tyr719Cys	Missense	bet TM2/TM3	0	0.04%(1/2,604)	0.990	Probably damaging
8	2157C > A	Tyr719X	Nonsense	bet TM2/TM3	0	0.12%(3/2,604)	NA	NA
8	2185A > G	Met729Val	Missense	bet TM2/TM3	0	0.04%(1/2,604)
8	2192T > A	Val731Glu	Missense	TM3	0	0.08%(2/2,604)
8	2195T > C	Leu732Pro	Missense	TM3	0	0.04%(1/2,604)
8	2223T > A	Tyr741X	Nonsense	TM3	0	0.04%(1/2,604)
8	2231C > T	Ser744Phe	Missense	TM3	0	0.04%(1/2,604)	1.000	Probably damaging
8	2233C > A	Leu745Met	Missense	TM3	0.50%(1/202)	0.04%(1/2,604)	0.786	Possibly damaging
8	2251G > T	Ala751Ser	Missense	TM3	0	0.04%(1/2,604)	0.831	Possibly damaging
8	2261A > G	Glu754Gly	Missense	bet TM3/TM4	0	0.04%(1/2,604)	0.960	Probably damaging
8	2267C > G	Ala756Gly	Missense	bet TM3/TM4	0	0.04%(1/2,604)
8	2293G > A	Asp765Asn	Missense	TM4	0	0.04%(1/2,604)
8	2294A > G	Asp765Gly	Missense	TM4	1.49%(3/202)	0.35%(9/2,604)
8	2297C > T	Thr766Met	Missense	TM4	0	0.08%(2/2,604)
8	2298INS C	NA	Insertion	TM4	0	0.08%(2/2,604)	NA	NA
8	2299INSC	Pro767ArgfsX28	Insertion	TM4	0	0.04%(1/2,604)	NA	NA
8	2302DUPC	NA	Insertion	TM4	0	0.08%(2/2,604)
8	2304DUPC	Met769HisfsX26	Insertion	TM4	0	1.08%(28/2,604)
8	2304DELC	Met769CysfsX38	Deletion	TM4	0	0.04%(1/2,604)
8	2305A > G	Met769Val	Missense	TM4	0	0.12%(3/2,604)
8	2308C > T	Leu770Phe	Missense	TM4	0	0.08%(2/2,604)	1.000	Probably damaging
8	2316_2317INS CTCTTTGTG	Val772insLeuPheVal	Insertion	TM4	0	0.04%(1/2,604)	NA	NA
8	2332C > T	Arg778Trp	Missense	TM4	0.99%(2/202)	0.19%(5/2,604)
8	2333G > T	Arg778Leu	Missense	TM4	18.81%(38/202)	28.57%(744/2,604)
8	2333G > A	Arg778Gln	Missense	TM4	0	1.42%(37/2,604)
8	2336G > A	Trp779X	Nonsense	TM4	0	0.04%(1/2,604)
8	2341G > A	Glu781Lys	Missense	TM4	0	0.04%(1/2,604)	0.998	Probably damaging
9	2383C > T	Leu795Phe	Missense	bet TM4/Td	0.50%(1/202)	0.08%(2/2,604)
9	2390C > T	Ser797Phe	Missense	bet TM4/Td	0	0.04%(1/2,604)	0.999	Probably damaging
10	2455C > T	Gln819X	Nonsense	bet TM4/Td	0	0.04%(1/2,604)
10	2464DUPA	Met822AsnfsX32	Insertion	bet TM4/Td	0	0.19%(5/2,604)
10	2506G > A	Gly836Arg	Missense	Td	0	0.04%(1/2,604)	0.998	Probably damaging
10	2509G > T	Gly837X	Nonsense	Td	0	0.04%(1/2,604)
10	2510DELG	Gly837GlufsX35	Deletion	Td	0	0.04%(1/2,604)
10	2519C > T	Pro840Leu	Missense	Td	0	0.04%(1/2,604)
10	2525A > G	Asp842Gly	Missense	Td	0	0.04%(1/2,604)	0.999	Probably damaging
10	2549C > T	Thr850Ile	Missense	Td	1.49%(3/202)	0.23%(7/2,604)
10	2561A > G	Glu854Gly	Missense	Td	0	0.04%(1/2,604)	0.998	Probably damaging
10	2564C > A	Ser855Tyr	Missense	Td	0	0.04%(1/2,604)
11	2587C > T	Pro863Ser	Missense	Td	0	0.04%(1/2,604)	0.950	Probably damaging
11	2593_2594INS GTCA	NA	Insertion	Td	0	0.04%(1/2,604)	NA	NA
11	2605G > A	Gly869Arg	Missense	bet Td/TM5	0	0.15%(4/2,604)
11	2620G > C	Ala874Pro	Missense	bet Td/TM5	0.50%(1/202)	0.27%(7/2,604)
11	2621C > T	Ala874Val	Missense	bet Td/TM5	0.50%(1/202)	2.42%(63/2,604)
11	2648_2649DEL	Val883AlafsX3	Deletion	bet Td/TM5	0	0.04%(1/2,604)
11	2659del G	Ala887LeufsX14	Deletion	bet Td/TM5	0	0.04%(1/2,604)
11	2662A > C	Thr888Pro	Missense	bet Td/TM5	1.49%(3/202)	0.61%(16/2,604)
11	2668G > A	Val890Met	Missense	bet Td/TM5	0	0.12%(3/2,604)
12	2740C > T	Gln914X	Nonsense	bet Td/TM5	0	0.04%(1/2,604)
12	2755C > G	Arg919Gly	Missense	bet Td/TM5	2.97%(6/202)	1.76%(46/2,604)
12	2755C > T	Arg919Trp	Missense	bet Td/TM5	0	0.08%(2/2,604)
12	2761A > C	Ser921Arg	Missense	bet Td/TM5	0	0.04%(1/2,604)
12	2785A > G	Ile929Val	Missense	TM5	0	0.04%(1/2,604)
12	2790_2792DEL	Ile930DEL	Deletion	TM5	0	0.23%(6/2,604)	NA	NA
12	2794_2795INSGT	Ser932CysfsX4	Insertion	TM5	0	0.04%(1/2,604)	NA	NA
12	2804C > T	Thr935Met	Missense	TM5	2.97%(6/202)	4.45%(116/2,604)
12	2810DELT	Val937GlyfsX5	Deletion	TM5	0	0.46%(12/2,604)
12	2827G > A	Gly943Ser	Missense	TM5	00	0.27%(7/2,604)
12	2828G > A	Gly943Asp	Missense	TM5	0.50%(1/202)	2.04%(53/2,604)
12	2848G > T	Val950Phe	Missense	bet TM5/TM6	0	0.04%(1/2,604)	0.978	Probably damaging
12	2853_2856DEL	Gln951HisfsX15	Deletion	bet TM5/TM6	0	0.04%(1/2,604)	NA	NA
13	2885DELC	NA	Deletion	bet TM5/TM6	0	0.04%(1/2,604)	NA	NA
13	2887C > T	Gln963X	Nonsense	bet TM5/TM6	0	0.04%(1/2,604)
13	2905C > T	Arg969Trp	Missense	TM6	0	0.04%(1/2,604)
13	2906G > A	Arg969gGln	Missense	TM6	0.99%(2/202)	0.12%(3/2,604)
13	2924C > A	Ser975Tyr	Missense	TM6	0.50%(1/202)	0.77%(20/2,604)
13	2930C > T	Thr977Met	Missense	TM6	0	0.08%(2/2,604)
13	2939G > A	Cys980Tyr	Missense	TM6	0.50%(1/202)	0.12%(3/2,604)
13	2944G > A	Ala982Thr	Missense	TM6	0	0.04%(1/2,604)
13	2957C > T	Ser986Phe	Missense	TM6	0	0.08%(2/2,604)
13	2975C > T	Pro992Leu	Missense	bet TM6/Ph	13.37%(27/202)	13.02%(339/2,604)
13	3007G > A	Ala1003Thr	Missense	bet TM6/Ph	0.50%(1/202)	0.19%(5/2,604)
13	3008C > T	Ala1003Val	Missense	bet TM6/Ph	0	0.04%(1/2,604)
13	3010C > T	Gln1004X	Nonsense	bet TM6/Ph	0	0.04%(1/2,604)	NA	NA
13	3028A > G	Lys1010Glu	Missense	bet TM6/Ph	0	0.04%(1/2,604)	0.997	Probably damaging
13	3029INST	Lys1010AsnfsX18	Insertion	bet TM6/Ph	0	0.08%(2/2,604)
13	3029A > C	Lys1010Thr	Missense	bet TM6/Ph	0	0.12%(3/2,604)
13	3041C > T	Pro1014Leu	Missense	bet TM6/Ph	0	0.04%(1/2,604)
13	3044T > C	Leu1015Pro	Missense	bet TM6/Ph	0	0.04%(1/2,604)	0.999	Probably damaging
13	3053C > T	Ala1018Val	Missense	bet TM6/Ph	0	0.12%(3/2,604)
13	3056A > C	His1019Pro	Missense	bet TM6/Ph	0	0.04%(1/2,604)	0.993	Probably damaging
14	3083A > G	Lys1028Arg	Missense	Ph	0	0.04%(1/2,604)	0.914	Probably damaging
14	3085A > G	Thr1029Ala	Missense	Ph	0.50%(1/202)	0.04%(1/2,604)
14	3087DELT	Gly1030AlafsX91	Deletion	Ph	0	0.04%(1/2,604)
14	3089G > A	Gly1030Asp	Missense	Ph	0	0.19%(5/2,604)
14	3095T > C	Ile1032Thr	Missense	Ph	0	0.04%(1/2,604)	0.997	Probably damaging
14	3098C > T	Thr1033Ile	Missense	Ph	0	0.04%(1/2,604)	0.999	Probably damaging
14	3104G > T	Gly1035Val	Missense	Ph	0	0.04%(1/2,604)
14	3121C > T	Arg1041Trp	Missense	ATP loop	0	0.08%(2/2,604)
14	3122G > C	Arg1041Pro	Missense	ATP loop	0	0.08%(2/2,604)
14	3140A > T	Asp1047Val	Missense	ATP loop	0	0.27%(7/2,604)
14	3155C > T	Pro1052Leu	Missense	ATP loop	0	0.12%(3/2,604)
14	3157DUPC	Leu1053ProfsX16	Insertion	ATP loop	0	0.04%(1/2,604)
14	3209C > G	Pro1070Arg	Missense	ATP loop	0.50%(1/202)	0.23%(6/2,604)	1.000	Probably damaging
14	3221C > T	Ala1074Val	Missense	ATP loop	0	0.04%(1/2,604)
14	3236G > T	Cys1079Phe	Missense	ATP loop	0	0.04%(1/2,604)
15	3263T > C	Leu1088Ser	Missense	ATP loop	0	0.15%(4/2,604)	1.000	Probably damaging
15	3271T > C	Cys1091Arg	Missense	ATP loop	0	0.04%(1/2,604)	0.960	Probably damaging
15	3274A > C	Thr1092Pro	Missense	ATP loop	0	0.08%(2/2,604)	0.832	Possibly damaging
15	3284A > C	Gln1095Pro	Missense	ATP loop	0	0.04%(1/2,604)
15	3293C > G	Pro1098Arg	Missense	ATP loop	0	0.04%(1/2,604)
15	3307DELG	NA	Deletion	ATP loop	0	0.04%(1/2,604)	NA	NA
15	3310T > C	Cys1104Arg	Missense	ATP loop	0	0.04%(1/2,604)
15	3311G > A	Cys1104Tyr	Missense	ATP loop	0	0.04%(1/2,604)
15	3316G > A	Val1106Ile	Missense	ATP loop	2.97%(6/202)	1.08%(28/2,604)
15	3368C > T	Pro1123Leu	Missense	ATP loop	0	0.04%(1/2,604)	0.001	Benign
15	3376DELC	His1126ThrfsX2	Deletion	ATP loop	0	0.04%(1/2,604)
15	3377_3378DELAC	His1126ProfsX3	Deletion	ATP loop	0	0.12%(3/2,604)	NA	NA
16	3426G > C	Gln1142His	Missense	ATP loop	0.50%(1/202)	1.04%(27/2,604)
16	3443T > C	Ile1148Thr	Missense	ATP loop	3.47%(7/202)	3.19%(84/2,604)
16	3445G > A	Gly1149Arg	Missense	ATP loop	0	0.04%(1/2,604)	1.000	Probably damaging
16	3446G > C	Gly1149Ala	Missense	ATP loop	0.50%(1/202)	0.04%(1/2,604)
16	3446G > A	Gly1149Glu	Missense	ATP loop	0.50%(1/202)	0.27%(7/2,604)
16	3451C > G	Arg1151Gly	Missense	ATP loop	0	0.04%(1/2,604)	1.000	Probably damaging
16	3451C > T	Arg1151Cys	Missense	ATP loop	0	0.08%(2/2,604)
16	3452G > A	Arg1151His	Missense	ATP loop	0	0.19%(5/2,604)
16	3459G > T	Trp1153Cys	Missense	ATP loop	0.99%(2/202)	0.19%(5/2,604)
16	3502G > C	Ala1168Pro	Missense	ATP loop	0	0.04%(1/2,604)
16	3517G > A	Glu1173Lys	Missense	ATP loop	0.50%(1/202)	0.54%(14/2,604)
16	3532A > G	Thr1178Ala	Missense	ATP loop	4.95%(10/202)	0.77%(20/2,604)
17	3563T > G	Leu1188Arg	Missense	ATP loop	0	0.04%(1/2,604)	0.998	Probably damaging
17	3577G > C	Ala1193Pro	Missense	ATP loop	0	0.04%(1/2,604)
17	3584C > T	Ala1195Val	Missense	ATP loop	0	0.04%(1/2,604)	0.997	Probably damaging
17	3587A > G	Asp1196Gly	Missense	ATP loop	0	0.04%(1/2,604)	1.000	Probably damaging
17	3605C > G	Ala1202Gly	Missense	ATP loop	0	0.08%(2/2,604)
17	3646G > A	Val1216Met	Missense	ATP bind	1.49%(3/202)	1.34%(35/2,604)
17	3653T > C	Leu1218Pro	Missense	ATP bind	0	0.04%(1/2,604)	0.999	Probably damaging
17	3659C > T	Thr1220Met	Missense	ATP bind	0	0.04%(1/2,604)
17	3677C > T	Thr1226Ile	Missense	ATP bind	0.50%(1/202)	0.08%(2/2,604)	0.990	Probably damaging
17	3679G > C	Ala1227Pro	Missense	ATP bind	0	0.04%(1/2,604)	0.999	Probably damaging
17	3682A > T	Arg1228X	Nonsense	ATP bind	0	0.04%(1/2,604)
18	3700DELG	Val1234LeufsX96	Deletion	ATP bind	0	0.23%(6/2,604)	NA	NA
18	3715G > T	Val1239Phe	Missense	ATP bind	0	0.12%(3/2,604)	0.997	Probably damaging
18	3716T > G	Val1239Gly	Missense	ATP bind	0.50%(1/202)	0.04%(1/2,604)
18	3733C > G	Pro1245Ala	Missense	ATP hinge	0	0.04%(1/2,604)	1.000	Probably damaging
18	3741C > G	His1247Gln	Missense	ATP hinge	0	0.04%(1/2,604)
18	3744G > C	Lys1248Asn	Missense	ATP hinge	0	0.08%(2/2,604)
18	3766_3767DUPCA	Gln1256ProfsX75	Insertion	ATP hinge	0	0.04%(1/2,604)	NA	NA
18	3767INSCA	NA	Insertion	ATP hinge	0	0.08%(2/2,604)	NA	NA
18	3776G > T	Gly1259Val	Missense	ATP hinge	0	0.12%(3/2,604)	0.988	Probably damaging
18	3791T > C	Met1264Thr	Missense	ATP hinge	0	0.04%(1/2,604)	0.990	Probably damaging
18	3793G > T	Val1265Leu	Missense	ATP hinge	0.50%(1/202)	0.04%(1/2,604)	0.980	Probably damaging
18	3796G > C	Gly1266Arg	Missense	ATP hinge	0	0.04%(1/2,604)	0.998	Probably damaging
18	3799G > A	Asp1267Asn	Missense	ATP hinge	0	0.04%(1/2,604)
18	3802G > A	Gly1268Arg	Missense	ATP hinge	0	0.04%(1/2,604)
18	3809A > G	Asn1270Ser	Missense	ATP hinge	1.98%(4/202)	1.88%(49/2,604)
18	3818C > T	Pro1273Leu	Missense	ATP hinge	0	0.08%(2/2,604)
18	3818C > A	Pro1273Gln	Missense	ATP hinge	0	0.15%(4/2,604)
18	3824T > C	Leu1275Ser	Missense	ATP hinge	0.50%(1/202)	0.08%(2/2,604)	1.000	Probably damaging
18	3836A > G	Asp1279Gly	Missense	ATP hinge	0	0.19%(5/2,604)
18	3843DUPT	Val1282CysfsX22	Insertion	ATP hinge	0	0.08%(2/2,604)
18	3848C > T	Ala1283Val	Missense	ATP hinge	0	0.04%(1/2,604)	1.000	Probably damaging
18	3851_3876DEL	NA	Deletion	ATP hinge	0	0.08%(2/2,604)	NA	NA
18	3859G > A	Gly1287Ser	Missense	ATP hinge	0.99%(2/202)	0.19%(5/2,604)
18	3877G > A	Glu1293Lys	Missense	ATP hinge	0	0.04%(1/2,604)
18	3884C > T	Ala1295Val	Missense	bet ATP hinge/TM7	1.98%(4/202)	0.61%(16/2,604)
18	3889G > A	Val1297Ile	Missense	bet ATP hinge/TM7	0	0.04%(1/2,604)
18	3896T > G	Leu1299Arg	Missense	bet ATP hinge/TM7	0	0.04%(1/2,604)	0.996	Probably damaging
18	3901_3902INSA	Arg1301PhefsX3	Insertion	bet ATP hinge/TM7	0	0.08%(2/2,604)	NA	NA
19	3955C > T	Arg1319X	Nonsense	bet ATP hinge/TM7	0	0.15%(4/2,604)
19	3960G > C	Arg1320Ser	Missense	bet ATP hinge/TM7	0.50%(1/202)	0.12%(3/2,604)
19	3965G > C	Arg1322Pro	Missense	bet ATP hinge/TM7	0.50%(1/202)	0.04%(1/2,604)
19	3982G > A	Ala1328Thr	Missense	TM7	0.99%(2/202)	0.23%(6/2,604)
19	4003G > C	Gly1335Arg	Missense	TM7	0	0.23%(6/2,604)
19	4005_4006INS	Gly1335INS LXWVA	Insertion	TM7	0	0.08%(2/2,604)	NA	NA
20	4043T > A	Ile1348Asn	Missense	TM7	0	0.04%(1/2,604)
20	4057T > C	Trp1353Arg	Missense	TM8	0	0.08%(2/2,604)
20	4059G > A	Trp1353X	Nonsense	TM8	0	0.04%(1/2,604)
20	4064G > A	Gly1355Asp	Missense	TM8	0	0.15%(4/2,604)
20	4094_4097DELCTGT	Ser1365TrpfsX27	Deletion	TM8	0	0.04%(1/2,604)
20	4112T > C	Leu1371Pro	Missense	TM8	0.50%(1/202)	0.27%(7/2,604)
20	4114C > T	Gln1372X	Nonsense	TM8	0.50%(1/202)	0.42%(11/2,604)
21	4162DELG	Ala1388ArgfsX5	Deletion	after TM8	0	0.04%(1/2,604)
21	4175T > A	Met1392Lys	Missense	after TM8	0	0.04%(1/2,604)
21	4272T > G	Tyr1424X	Nonsense	after TM8	0	0.04%(1/2,604)	NA	NA
21	4333G > C	Ala1445Pro	Missense	3COOH	0	0.04%(1/2,604)
Notes: Novel mutations are highlighted in bold; TMS: transmembrane domain; TDS: transduction domain.

Table 2

Polymorphisms in *ATP7B* identified in 103 WD patients
Exon	Nucleotide change	Polymorphism	Nucleotide sequence	Area of protein	Type	Frequency(%) [Patients(n = 103)]
1 2 2 2 3 8 10 12 12 13 13 14 16 18 18	9A > G 870G > C 1216T > G 1168A > G 1366G > C 2310C > G 2495G > A 2855G > A 2785A > G 3009G > A 2913T > A 3188C > T 3419T > C 3889G > A 3798G > T	Glu3Glu Val1290Val Ser406Ala Ile390Val Leu456Val Leu770Leu Arg832Lys Arg952Lys Ile929Val Ala1003Ala Ala971Ala Ala1063Val Val1140Ala Val1297Ile Gly1266Gly	GAG > GGA GTG > GTC TCT > GCT ATA > GTA GTG > CTG CTC > CTG AGG > AAG AGA > AAA ATC > GTC GCG > GCA GCT > GCA GCG > GTG GTC > GCC GTC > ATC GGG > GGT	Before Cu1 Cu3 Cu4 Cu4 Cu4/Cu5 TM4 TM4/Td TM5 TM5 Bet TM6/Ph TM6 ATP loop ATP loop ATP hinge ATP hinge	Missense Silent Missense Missense Missense Silent Missense Missense Missense Silent Silent Missense Missense Missense Silent	89.47% 95.14% 87.37% 0.97% 87.37% 35.92% 66.02% 74.76% 0.97% 3.88% 0.97% 0.97% 75.73% 1.94% 1.94%

The exons harbouring the highest percentage of mutations were exons 8, 13, 16, 12 and 18. The total mutation detection rate on these five exons was 63.4% (128/202), suggesting that these exons could be important regions for detecting mutations in the southern Chinese WD cohort. The mutations on exons 8 and 13 accounted for 28.8% (n = 47) and 19.6% (n = 32) of the total mutant alleles (n = 163), respectively. The detection rate of other mutations on exons 16, 12 and 18 spread from 14–8.0% (Fig. 1).

Mutation spectrum of ATP7B in a large-scale sample of Chinese patients

A search of the literature for studies on the overall spectrum of mutations in large-scale sample of Chinese WD patients was conducted. Of the 5,868 studies initially, 5,848 publications were removed due to their irrelevant titles and abstracts. After full-length review of 20 included studies, we further excluded 6 articles that failed to meet the inclusion criteria. Finally, 14 eligible articles with 1,201 WD probands were included, and an additional 101 probands from the southern part of China in the present study were also included, for a grand total of 1,302 index patients with WD in the systemic analysis. In total, 233 different mutations in the coding region of the ATP7B gene were detected in our patient pool, including 154 missenses, 23 nonsenses and 56 insertions or deletions. Eighty-five variants were identified as novel mutations in the WD databases mentioned above. The computational predictive analysis of the missense variants by PolyPhen-2 was also shown in Table 1. Most of the missense substitutions showed a significant effect on the protein. All mutations accounted for 87.0% (n = 2265) of the alleles studied (n = 2604), with 2333G > T (Arg778Leu) and 2975C > T (Pro992Leu) being the two most common mutations at a frequency of 28.6% and 13.0%, respectively. Mutations were distributed in all exons except exon 1. We found that exons 8, 13, 12 and 16 were the hot spot exons in this large WD pool, accounting for 64.1% (1668/2604) of studied alleles. The mutations accumulated on exons 8, 13, 12 and 16 showed higher prevalence, accounting for 38.0%, 17.3%, 10.9% and 7.4% of the mutant chromosomes, respectively, as depicted in Fig. 2.

It was interesting to note that the prevalent mutation 2333G > T (Arg778Leu) was almost exclusively linked with 2310C > G (Leu770Leu) polymorphism. 2310C > G (Leu770Leu) polymorphism was rare in the normal population but appeared frequently in the WD cohort, suggesting that, to some extent, this linkage may impact the expression of the ATP7B protein.

Characterisation Of Clinical Phenotype

Screening patients from the 14 eligible reports ^[14–27] identified 108 patients with clear WD features. We enrolled 196 patients with WD in the final analysis, including the current 88 patients with detailed data from southern China. Of these, 50.5% (99/196) patients had a primary hepatic manifestation, 18.9% (37/196) showed a primary neurological manifestation and 13.8% (27/196) had combined hepatic and neurological manifestations. Thirty-three of the 196 patients (16.8%) presented with no symptoms. The mean age at symptom onset was 19.3 years (from 1 year to 62 years). The median ceruloplasmin level was 82.2 mg/L (range 20–962 mg/L).

We observed that the patients with mixed manifestation were significantly older at symptom onset than patients with hepatic symptoms (24.4 V 19.4 years of age, P = 0.039) and patients with neurological symptoms (24.4 V 17.9 years of age, P = 0.015). There is no significant difference in terms of onset age between patients with hepatic phenotype and patients with neurological presentation (P > 0.05). In addition, as shown in Fig. 3A, the patients with typically clinical manifestations displayed later age of onset than the clinically asymptomatic patients, which was a statistically significant difference (19.8 V 12.9 years of age, P = 0.000).No significant difference was observed in the presence of KF rings among the three clinical subtypes with different manifestations(Figure 3B)

Correlation Between Genotype And Phenotype

To describe the picture of correlation between genotype and phenotype based on the cohort investigation, we initially studied the rarely reported association between the exons and clinical subtypes. First, we examined the hot spot exons in the available 196-patient WD cohort. The results showed that exons 8, 13 and 16 harboured the highest percentage of mutations, consistent with the results described in the large WD patient pool in the comprehensive analysis. Secondly, in different clinical subtypes, we sorted 21 exons in order of mutation frequencies to gain more insights into the most frequent exons in different types of clinical presentations. Notably, in the mixed presentation group, the second most prevalent exon was exon 11, found in 11.5% of mutant alleles, which was much higher than that in the other two groups. Hence, we assumed that the mutations in exon 11 might play an important role in the development of combined presentation. In the primary hepatic group, exon 18 was the third most mutant exon, contributing 10.4% of mutant chromosomes. The possible association between the mutations in exon 18 and the hepatic manifestation is discussed below. In this course, we observed that 2621C > T (Ala874Val), the most prevalent mutation in exon 11, frequently occurred with 2333 G > T (Arg778Leu) except in single heterozygotes. Another observation was that 3884C > T (Ala1295Val), the most prevalent mutation in exon 18, only mutated in patients with hepatic symptoms. The difference between the 3884C > T patients with non-3884C > T patients in three clinical groups was statistically significant (P = 0.048), as shown in Fig. 4A. This indicated that the 3884C > T (Ala1295Val) mutation in exon 18 was significantly associated with hepatic symptoms.

Next, we studied the correlation between specific mutations and the phenotypes. We examined the most prevalent mutations in the available 196-patient WD cohort. The results showed that the two most common mutations were 2333G > T (Arg778Leu) and 2975C > T (Pro992Leu), accounting for 21.4%(84/392)and 12.5%༈49/392༉of the alleles, in good agreement with the results demonstrated in the comprehensive analysis. With regard to the most frequent mutation, 2333 G > T (Arg778Leu), we observed that patients carrying the 2333 G > T (Arg778Leu) mutation had a lower serum ceruloplasmin levels than patients with other mutations at both alleles. When comparing 2333 G > T homozygous (39.3 ± 23.5 mg/L) or heterozygous patients (67.7 ± 48.1 mg/L) with non-2333 G > T patients (79.7 ± 47.1 mg/L), we found significant differences (P = 0.018, P = 0.049, respectively). However, the difference between 2333G > T homozygous and heterozygous patients was not significant, as shown in Fig. 4B. With regard to the second most common mutation, 2975 C > T (Pro992Leu), we found that the patients with 2975 C > T mutation often exhibited symptoms earlier than the patients without 2975 C > T mutation at both chromosomes (Fig. 4C). The difference in the age of onset between 2975 C > T homozygous (9.7 ± 4.2 years of age) and non-2975 C > T patients (20.5 ± 12.2 years of age) was significant (P = 0.01), and the difference between 2975 C > T heterozygous (15.7 ± 11.6 years of age) and non-2975 C > T patients was also significantly different (P = 0.017). No significant difference was observed in the age of onset between 2975 C > T homozygous and heterozygous patients (P > 0.05). We also found a dramatic association between the 3809 A > G (Asn1270Ser) mutation and the disease onset age. Statistics showed that the patients with 3809 A > G mutation had an earlier age of onset (10.8 ± 7.4 years of age) than the non-3809 A > G patients (19.3 ± 12.0 years of age) (P = 0.012, Fig. 4D).

It should be noted that 82.2% of patients carried at least two mutations at both alleles, while 17.8% patients only carried one mutation at two chromosomes. We did not find a significant correlation between the different forms of mutation (homozygous V heterozygous mutations, combined mutations V single heterozygous mutations) and the several clinical indices in terms of the age of onset, clinical manifestations, ceruloplasmin level and the presence of KF rings.

In the present study, we explored the mutations in the ATP7B gene in 101 WD probands from southern China. Forty-eight mutations were found, including 7 novel variants. These novel variants were not found in the control chromosomes. Substituted amino acids with a PolyPhen-2 score close to 1.000 could be predicted to be potentially damaging. However, insertion, deletion and premature stop mutations failed to yield acceptable results from PolyPhen-2 analysis. We regarded the insertion mutation (1510–1511 ins A) as a clearly pathogenic mutation since it caused a frameshift leading to a premature stop codon. According to the PolyPhen-2 score, the other six variants (2075 T > C (Leu692Pro), 2233 C > A (Leu745Met), 3209C > G (Pro1070Arg), 3677C > T (Thr1226Ile), 3793 G > T (Val1265Leu), 3824 T > C (Leu1275Ser)) were all predicted to affect protein function. A mutation detection rate of 80.7% was achieved in the southern cohort, but 19.3% of alleles remained unidentified. Failure to detect the remaining mutations may be explained by some objective factors, such as the primers, the PCR procedure and/or the sequence alignment. In our recent studies, the detection rate of direct sequencing could reach as high as 95% with the newly designed PCR primers and the improved amplification requirements. Another reason may be due to the presence of mutations outside the open reading frame of the gene, i.e., in the promoter, introns, the presence of gene rearrangements or possible mutations in other copper-transport chaperone gene, such as ATOX1 and COMMD1^[28].

2333G > T (Arg778Leu) was the most frequent mutation in our study and was also described as the most common mutation in China ^{[16–21, 23, 24]}, accounting for 18.8% of alleles studied here. The second most common mutation was 2975 C > T (Pro992Leu) among the WD patients, with an allelic frequency of 13.4%, consistent with the frequency previously reported in China ^{[17, 19, 20, 23]}. However, Hong et al ^[24] suggested that 3443 T > C (Ile1148Thr) was the second most common mutation instead of 2975 C > T in their cohort study of 103 Chinese WD patients. An earlier study of 114 WD patients from northern China demonstrated that 2621 C > T (Ala874Val) was the second hot-spot mutation, followed by 2975 C > T, at an allelic frequency of 6.1% ^[29]. We speculate that different gene-level tests or a limited number of patients is largely responsible for the differential conclusions. A large-scale or prospective study, based on the same detection standard, is imperative.

In our current study, exon 8 was the most frequent mutational site, found in 28.8% of mutant alleles, followed by exon 13 in 19.6% and exon 16 in 14.1%, indicating that these three exons could be important regions for detecting mutations.

We conducted a comprehensive analysis of the spectrum and frequency of ATP7B mutations in a large-scale sample of Chinese WD patients from more than 30 provinces, autonomous regions and municipalities of China. A total of 233 distinct mutations were detected, of which 85 were novel. The computational predictive analysis software PolyPhen-2 interpreted most of the novel missense variants as disease-causing mutations, with the exception of several benign variants (121A > G (Asn41Asp), 748G > A (Gly250Arg), 1168A > G (Ile390Val), 1426G > A (Ala476Thr), 1760C > T (Thr587Met), 3368C > T (Pro1123Leu)). It cannot be ruled out that the silent mutations interpreted as benign could affect protein function.

The most prevalent mutation in the 1302 WD patients pool was 2333G > T (Arg778Leu), in exon 8, with an allelic frequency of 28.6%. The 2333G > T (Arg778Leu) mutation is frequently found in reports of Asian patients, with an allele frequency of 12 to 50% ^{[30], [14, 16–20]}. In contrast, the 3207 C > A (His1069Gln) mutation, the most common mutation in European and North American populations, accounting for 30 to 70% of the alleles studied ^[30], was not detected in any Chinese patients. The next most frequent mutation in this large cohort of Chinese patients was 2975 C > T (Pro992Leu), with an allelic frequency of 13.0%. To our knowledge, the highest frequency of 2975 C > T described so far was 27% ^[14].

All exons except exon 1 were affected. Notably, exons 8, 13, 12 and 16 were the hot-spot exons identified in the large WD population, accounting for 73.6% of mutant alleles, consistent with previous results that 60.5–74% of mutations were located on the above hot-spot exons ^[31].

The spectrum of WD mutations in the large cohort of Chinese patients consisted of a small number of relatively frequent mutations and a greater number of rare mutations. This further indicated a high degree of mutational heterogeneity, in agreement with previously published findings ^{[32], [33]}. Moreover, we found that many mutations were located a short distance away, in line with the preliminary results ^[34]. Additionally, 64.3% patients were found to stay in a compound heterozygotic state, compared with 13.8% patients in a homozygotic state and 17.4% patients in a single heterozygotic state, which can be explained by the low percentage of consanguinity in our investigated population. No significant difference in phenotypic profiles were found when comparing homozygous or combined heterozygous patients with the patients who had only one mutation at two alleles. We suspected that, to the patients with a single mutation, the remaining unidentified mutations would probably be located in non-coding regions of the ATP7B gene. Other mutational mechanisms should also be taken into consideration.

One polymorphism with substitution of leucine with leucine at codon 770 in the transmembrane region of ATP7B has been found to be linked with the 2333 G > T (Arg778Leu) mutation. Perhaps, the coexistence of the 2333G > T mutation and the 2310C > G polymorphism would have a special effect on the ATP7B protein. Further investigation of the functional implications of both is needed.

A well-defined landscape of the genotype-phenotype correlation will promote the development of clinical studies. However, most of previous studies devoted to genotype-phenotype association have addressed rare or conflicting conclusions. Verification of this requires a cohort study. Our study significantly described a systemic and quantitative analysis of the genotype-phenotype correlation in a large cohort of Chinese patients with WD.

In the demonstration of hot-spot exons in different types of clinical presentations, we identified that exon 11 was ranked as the second most mutational exon in the mixed presentation. The difference in the proportion of patients with mutations in exon 11 between the hepatic and the mixed group was significant (P = 0.046), while the difference between the neurological and the mixed group was not significant. We could not reach the correlation between the mutations in exon 11 and the mixed manifestation. 2621 C > T (Ala874Val), the predominant mutation in exon 11, frequently presented with 2333 G > T (Arg778Leu) substitution. Krishna et al considered that the hydrophobicity and conformational stability of the hydrophobic domains, such as transmembrane domains, may be altered due to the valine amino acid ^[35]. We speculated that the transmembrane domain region of ATP7B with valine at the 874 domain region and with leucine at 778 could probably destabilise the formation or influence the expression of protein. Functional studies of mutations are required for the validation of our speculation. Another finding in the analysis of exon hot spots in different clinical presentations was that exon 18 ranked as the third exon with the most mutations in the hepatic presentation group, with a higher mutation frequency (10.38%) than that in the other two groups (6.25% and 3.85% in the neurological and mixed presentation, respectively). This is probably attributable to the potential association between the mutations in exon 18 and the hepatic involvement. Fortunately, we identified that 3884 C > T (Ala1295Val), one kind of the mutations in exon 18, only mutated in the patients with hepatic manifestation. Statistical analysis revealed that there was a significant association between the 3884 C > T (Ala1295Val) mutation and the hepatic phenotype, which was consistent with previous observations that mutations in the conserved ATP hinge region were associated with liver disease without neurological presentation ^[36], and when the mutation affected the ATP hinge, it resulted in hepatic failure ^[37].

Furthermore, we found a statistically significant correlation between the 2333 G > T (Arg778Leu) mutation and lower serum ceruloplasmin levels. The difference in the serum ceruloplasmin level between 2333 G > T homozygous or heterozygous patients and non-2333 G > T patients was significant. A recent study in a large cohort of Chinese WD patients ^[1] showed that 2333 G > T was related to lower levels of ceruloplasmin as well. That study also suggested that 2333 G > T was related to younger onset age. However, in our study, we did not find the significant difference between 2333 G > T and the onset age, and we did not find a considerable difference between 2333 G > T and the hepatic manifestation either, as previously reported by Liu et al ^[38]. Significant difference in the age of onset was observed between 2975 C > T homozygous or combined heterozygous patients and non-2975 C > T patients. Collectively, our finding revealed that the patients with 2975 C > T mutation often presented with WD profiles at an earlier age, usually before 13.7 years old, than the patients with other mutations, while Hua et al ^[19] described that the patients with 2975C > T often presented WD features before 12 years old. We also found a remarkable association between 3809 A > G (Asn1270Ser) and the disease onset age. Statistical findings showed that the patients with 3809 A > G mutation usually manifested the WD features before 10.8 years old, much earlier than the patients with other mutations at two chromosomes.

We also observed a visible correlation between the onset age and the characteristic clinical manifestations. Our findings showed that the patients with mixed manifestation had a later age of onset than the groups with either liver disease or neurological phenotype. However, the two latter onset ages were not significantly different from one another, inconsistent with the understanding that patients having predominantly neuropsychiatric symptoms usually manifest symptoms later than patients with hepatic presentation ^[39][40][41]. Our results showed that the liver and brain could be affected by WD simultaneously.

In addition, we discovered interesting clinical differences between the symptomatic group and the asymptomatic group in terms of the presence of KF rings. KF rings of symptomatic patients were found to be significantly higher than KF rings of asymptomatic cases, consistent with the earlier findings ^[42]. We also found that the patients who had KF rings were significantly older at symptom onset than the cohort without KF rings, which is in line with the finding identified in our study that the patients with typical clinical manifestations significantly displayed later age of onset than the patients who were clinically asymptomatic. In summary, the cohort with clinical symptoms presented with a later age of onset and higher prevalence of KF rings than the asymptomatic cohort. One report claimed that patients with purely neurological symptoms were susceptible to KF rings ^[43]. Our current study did not show any clear difference in the presence of KF rings among the three groups with different clinical manifestations.

In conclusion, we characterised a complete genotypic and phenotypic profile of Chinese patients with WD. The seven novel mutations identified in the southern Chinese WD patients could considerably extend the previously established spectrum of the ATP7B mutations. Comprehensive mutation analysis will enhance the current knowledge of WD genetics in China. The findings of correlation between specific mutations and clinical features, as well as the age of onset and several clinical profiles provides new insights into the relationships between genotype and phenotype. Additional large studies are required for validation of our conclusions.

Acknowlegements

We thank all the patients who agreed to participate in this study.

Funding

This study was supported by the Hunan Province Development and Reform Commission Foundation (KZ, Luo N0.2016-65) and the National Natural Science Foundation of China (Nos. 81500455). The funders had no influence on the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Contributions

Xu Yang and Kaizhong Luo designed the experiments. Mingming Li developed the method and executed the experiments. Ji Ma and Wenlong Wang performed Statistical analysis. Mingming Li wrote the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Central South University Medical College (Reference number: S073(2013)). All patients had signed informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Availability of data and material

All data generated or analysed during this study are included in this published article，except the sequencing data.

The sequencing data generated in this study are available from the corresponding author on reasonable request.

Corresponding author

Correspondence to Kaizhong Luo ( [email protected])

Guidelines and regulations

All methods were performed in accordance with the Declaration of Helsinki

Cheng N, Wang H，Wu W，Yang R，Liu L，Han Y，Guo L，Hu J，Xu L，Zhao J，Han Y，Liu Q，Li K，Wang X，Chen W. Spectrum of ATP7B mutations and genotype-phenotype correlation in large-scale Chinese patients with Wilson Disease. Clin Genet 2017: 92(1): p. 69-79 [PMID:27982432 DOI:1111/cge.12951]
Aggarwal A and M Bhatt. Update on Wilson disease. Int Rev Neuropil 2013: 110: p. 313-48 [PMID:24209444 DOI:10.1016/B978-0-12-410502-7.00014-4]
Pfeiffer RF. Wilson's Disease. Semin Neurol 2007: 27(2): p. 123-32 [PMID:17390257 DOI:10.1055/s-2007-971173]
Kim GH, Yang JY, Park JY, Lee JJ, Kim JH, Yoo HW. Estimation of Wilson's disease incidence and carrier frequency in the Korean population by screening ATP7B major mutations in newborn filter papers using the SYBR green intercalator method based on the amplification refractory mutation system. Genet Test 2008: 12(3): p. 395-9 [ PMID:18652531 DOI:10.1089/gte.2008.0016]
Zarina A, Tolmane I, Kreile M, Chernushenko A, Cernevska G, Pukite I, Micule I, Krumina Z, Krumina A, Rozentale B, Piekuse L. Genetic variation spectrum in ATP7B gene identified in Latvian patients with Wilson disease. Mol Genet Genomic Med 2017: 5(4): p. 405-409 [PMID:28717664 DOI:10.1002/mgg3.297]
Coffey A.J, Durkie M, Hague S, McLay K, Emmerson J, Lo C, Klaffke S, Joyce C J, Dhawan A, Hadzic N, Mieli-Vergani G, Kirk R, Elizabeth Allen K, Nicholl D, Wong S, Griffiths W, Smithson S, Giffin N, Taha A, Connolly S, Gillett G T, Tanner S, Bonham J, Sharrack B, Palotie A, Rattray M, Dalton A, Bandmann O. A genetic study of Wilson's disease in the United Kingdom. Brain 2013. 136(Pt 5): p. 1476-87 [PMID:23518715 DOI:10.1093/brain/awt035]
Ohura T, Abukawa D, Shiraishi H, Yamaguchi A, Arashima S, Hiyamuta S, Tada K, Iinuma K. Pilot study of screening for Wilson disease using dried blood spots obtained from children seen at outpatient clinics. J Inherit Metab Dis 1999. 22(1): p. 74-80 [PMID:10070620]
Petrukhin K, Fischer SG, Pirastu M, Tanzi RE, Chernov I, Devoto M, Brzustowicz LM, Cayanis E, Vitale E, Russo JJ. Mapping, cloning and genetic characterization of the region containing the Wilson disease gene. Nat Genet 1993: 5(4): p. 338-43 [PMID:8298640 DOI:10.1038/ng1293-338]
Bull PC, Thomas GR, Rommens JM, Forbes JR, Cox DW. The Wilson disease gene is a putative copper transporting P-type ATPase similar to the Menkes gene. Nat Genet 1993: 5(4): p. 327-37 [PMID:8298639DOI:10.1038/ng1293-327]
Tanzi RE, Petrukhin K, Chernov I, Pellequer JL, Wasco W, Ross B, Romano DM, Parano E, Pavone L, Brzustowicz LM. The Wilson disease gene is a copper transporting ATPase with homology to the Menkes disease gene. Nat Genet 1993: 5(4): p. 344-50 [PMID:8298641 DOI:10.1038/ng1293-344]
Roberts EA and ML Schilsky. Diagnosis and treatment of Wilson disease: an update. Hepatology 2008: 47(6): p. 2089-111 [PMID:18506894 DOI:10.1002/hep.22261]
Medici V, L Rossaro and GC Sturniolo. Wilson disease--a practical approach to diagnosis, treatment and follow-up. Dig Liver Dis, 2007: 39(7): p. 601-9 [PMID:17382611 DOI:10.1016/j.dld.2006.12.095]
Poon KS, KM Tan and ES Koay. Targeted next-generation sequencing of the ATP7B gene for molecular diagnosis of Wilson disease. Clin Biochem 2016: 49(1-2): p. 166-71. [PMID:26483271 DOI:10.1016/j.clinbiochem.2015.10.003]
Ye S, Gong L, Shui QX, Zhou LF. Wilson disease: identification of two novel mutations and clinical correlation in Eastern Chinese patients. World J Gastroenterol, 2007: 13(38): p. 5147-50 [PMID:17876883]
Dong Y, Ni W, Chen WJ, Wan B, Zhao GX, Shi ZQ, Zhang Y, Wang N, Yu L, Xu JF, Wu ZY. Spectrum and Classification of ATP7B Variants in a Large Cohort of Chinese Patients with Wilson's Disease Guides Genetic Diagnosis. Theranostics 2016: 6(5): p. 638-49 [PMID: 27022412 DOI: 10.7150/thno.14596]
Geng J, Wang J, Yao RE, Liu XQ, Fu QH. Identification of one novel and nine recurrent mutations of the ATP7B gene in 11 children with Wilson disease. World J Pediatr, 2013: 9(2): p. 158-62 [PMID:23275100 DOI:10.1007/s12519-012-0388-7]
Gu S, Yang H, Qi Y, Deng X, Zhang L, Guo Y, Huang Q, Li J, Shi X, Song Z, Deng H. Novel ATPase Cu (2+) transporting beta polypeptide mutations in Chinese families with Wilson's disease. PLoS One 2013: 8(7): p. e66526 [PMID:23843956 DOI: 10.1371/journal.pone.0066526]
Gu YH, Kodama H. Du S L, Gu Q J, Sun H J, Ushijima H. Mutation spectrum and polymorphisms in ATP7B identified on direct sequencing of all exons in Chinese Han and Hui ethnic patients with Wilson's disease. Clin Genet 2003: 64(6): p. 479-84 [PMID:14986826]
Hua R, Hua F, Jiao Y, Pan Y, Yang X, Peng S, Niu J. Mutational analysis of ATP7B in Chinese Wilson disease patients. Am J Transl Res 2016: 8(6): p. 2851-61 [PMID:27398169]
Li XH, Lu Y, Ling Y, Fu QC, Xu J, Zang G Q, Zhou F, De-Min Y, Han Y, Zhang DH, Gong Q M, Lu ZM, Kong XF, Wang JS. Zhang XX. Clinical and molecular characterization of Wilson's disease in China: identification of 14 novel mutations. BMC Med Genet 2011: 12: p. 6 [PMID:21219664 DOI:10.1186/1471-2350-12-6]
Liu Y, Zhou H, Guo H, Bai Y. Genetic and Clinical Analysis in a Cohort of Patients with Wilson's Disease in Southwestern China. Arch Med Res 2015: 46(2): p. 164-9 [PMID: 25704634 DOI: 10.1016/j.arcmed.2015.02.001]
Lu CX, Qing Lin, Huang W Q, Tzeng CM. New mutations and polymorphisms of the ATP7B gene in sporadic Wilson disease. Eur J Med Genet 2014: 57(9): p. 498-502 [PMID:24878384 DOI:10.1016/j.ejmg.2014.04.016]
Wan L, Tsai C H, Hsu C M, Huang CC, Yang CC, Liao CC, Wu CC, Hsu YA, Lee CC, Liu SC, Lin WD, Tsai FJ. Mutation analysis and characterization of alternative splice variants of the Wilson disease gene ATP7B. Hepatology, 2010: 52(5): p. 1662-70 [PMID: 20931554 DOI: 10.1002/hep.23865]
Wei Z, Huang Y, Liu A, Diao S, Yu Q, Peng Z, Hong M. Mutational characterization of ATP7B gene in 103 Wilson's disease patients from Southern China: identification of three novel mutations. Neuroreport, 2014: 25(14): p. 1075-80 [PMID: 25089800 DOI: 10.1097/WNR.0000000000000216]
Zhang DF and JF Teng. Direct sequencing of mutations in the copper-transporting P-type adenosine triphosphate (ATP7B) gene for diagnosis and pathogenesis of Wilson's disease. Genet Mol Res, 2016: 15(3). [PMID:27706781 DOI: 10.4238/gmr.15038746]
Zhu HW, Tao ZB, Su G, Jin QY, Zhao LT, Zhu JR, Yan J, Yu TY, Ding JX, Li YM. Identification of two novel mutations in the ATP7B gene that cause Wilson's disease. World J Pediatr, 2017: 13(4): p. 387-391 [PMID:28856630 DOI:10.1007/s12519-017-0055-0]
Chen L, Li X, Zheng Z, Lu X, Lin M, Pan C Liu J. A novel ATP7B gene mutation in a liver failure patient with normal ceruloplasmin and low serum alkaline phosphatase. Gene 2014: 538(1): p. 204-6 [PMID:24476933 DOI:10.1016/j.gene.2013.10.044]
Park S, Park JY, Kim GH, Choi JH, Kim KM, Kim JB. Yoo HW. Identification of novel ATP7B gene mutations and their functional roles in Korean patients with Wilson disease. Hum Mutat, 2007: 28(11): p. 1108-13 [PMID:17587212 DOI:10.1002/humu.20574]
Li K, Zhang WM, Lin S, Wen L, Wang ZF, Xie D, Wei M, Qiu ZQ, Dai Y, Lin MC, Kung HF, Yao FX. Mutational analysis of ATP7B in north Chinese patients with Wilson disease. J Hum Genet, 2013: 58(2): p. 67-72 [PMID:23235335 DOI:10.1038/jhg.2012.134]
Ferenci P. Regional distribution of mutations of the ATP7B gene in patients with Wilson disease: impact on genetic testing. Hum Genet 2006: 120(2): p. 151-9 [PMID:16791614 DOI: 10.1007/s00439-006-0202-5]
Wu ZY, Wang N, Lin MT, Fang L, Murong SX, Yu L. Mutation analysis and the correlation between genotype and phenotype of Arg778Leu mutation in Chinese patients with Wilson disease. Arch Neurol 2001: 58(6): p. 971-6 [PMID: 11405812]
Al Jumah M, Majumdar R, Al Rajeh S, Awada A, Al Zaben A, Al Traif I, Al Jumah AR, Rehana ZA Clinical and genetic study of 56 Saudi Wilson disease patients: identification of Saudi-specific mutations. Eur J Neurol 2004: 11(2): p. 121-4 [PMID: 14748773]
Vrabelova S, Letocha O, Borsky M, Kozak L. Mutation analysis of the ATP7B gene and genotype/phenotype correlation in 227 patients with Wilson disease. Mol Genet Metab 2005: 86(1-2): p. 277-85 [PMID:15967699DOI: 10.1016/j.ymgme.2005.05.004]
Haas R, Gutierrez-Rivero B, Knoche J, Boker K, Manns MP, Schmidt HH. Mutation analysis in patients with Wilson disease: identification of 4 novel mutations. Mutation in brief no. 250. Online. Hum Mutat 1999: 14(1): p. 88. [PMID: 10447265 DOI: 10.1002/(SICI)1098-1004(1999)14:1<88::AID-HUMU15>3.0.CO;2-H]
Krishna BM, Chaudhary S, Panda AK, Mishra DR, Mishra SK. Her2 (Ile)655(Val) polymorphism and its association with breast cancer risk: an updated meta-analysis of case-control studies. Sci Rep 2018: 8(1): p. 7427 [PMID: 29743533 DOI: 10.1038/s41598-018-25769-y]
Barada K, El-Atrache M, El Hajj II, Rida K, El-Hajjar J, Mahfoud Z, Usta J. Homozygous mutations in the conserved ATP hinge region of the Wilson disease gene: association with liver disease. J Clin Gastroenterol 2010: 44(6): p. 432-9. [PMID: 20485189 DOI: 10.1097/MCG.0b013e3181ce5138]
Abdelghaffar TY, Elsayed SM, Elsobky E, Bochow B, Buttner J, Schmidt H. Mutational analysis of ATP7B gene in Egyptian children with Wilson disease: 12 novel mutations. J Hum Genet 2008: 53(8): p. 681-7. [PMID: 18483695 DOI: 10.1007/s10038-008-0298-7]
Liu XQ, Zhang YF, Liu TT, Hsiao KJ, Zhang JM, Gu XF, Bao KR, Yu LH, Wang MX. Correlation of ATP7B genotype with phenotype in Chinese patients with Wilson disease. World J Gastroenterol 2004: 10(4): p. 590-3. [PMID: 14966923]
Aggarwal A, Chandhok G, Todorov T, Parekh S, Tilve S, Zibert A. Bhatt M, Schmidt HH. Wilson disease mutation pattern with genotype-phenotype correlations from Western India: confirmation of p.C271* as a common Indian mutation and identification of 14 novel mutations. Ann Hum Genet 2013: 77(4): p. 299-307. [PMID: 23551039 DOI:10.1111/ahg.12024]
Merle U, Schaefer M, Ferenci P, Stremmel W. Clinical presentation, diagnosis and long-term outcome of Wilson's disease: a cohort study. Gut 2007: 56(1): p. 115-20 [PMID:16709660 DOI:10.1136/gut.2005.087262]
Kumar S, Thapa B, Kaur G, Prasad R. Analysis of most common mutations R778G, R778L, R778W, I1102T and H1069Q in Indian Wilson disease patients: correlation between genotype/phenotype/copper ATPase activity. Mol Cell Biochem 2007: 294(1-2): p. 1-10. [PMID: 17160357 DOI: 10.1007/s11010-005-9028-z]
Rodman R, Burnstine M. Esmaeli B, Sugar A, Martonyi C, Johnson V, Brewer G. Wilson's disease: presymptomatic patients and Kayser-Fleischer rings. Ophthalmic Genet 1997: 18(2): p. 79-85 [PMID:9228244]
Seo GH, Kim YM, Oh SH, Chung SJ, Choi IH, Kim GH, Yum MS, Choi JH, Kim KM, Ko TS, Lee BH. Biochemical and molecular characterisation of neurological Wilson disease. J Med Genet 2018 [PMID:29618506 DOI:10.1136/jmedgenet-2017-105214]

Download PDF

Editorial decision: Major revision
11 Feb, 2021
Reviews received at journal
09 Feb, 2021
Reviewers agreed at journal
28 Jan, 2021
Reviews received at journal
23 Jan, 2021
Reviewers agreed at journal
21 Jan, 2021
Reviewers invited by journal
21 Jan, 2021
Editor assigned by journal
21 Jan, 2021
Editor invited by journal
18 Jan, 2021
Submission checks completed at journal
18 Jan, 2021
First submitted to journal
24 Dec, 2020

You are reading this latest preprint version

Mutation analysis of the ATP7B gene and genotype-phenotype correlation in Chinese patients with Wilson disease

Status:

Version 1

Abstract

Figures

Introduction

Material And Methods

Results

Discussion

Declarations

References

Status:

Version 1