Abstract
As an important multifunctional protein involved in regulation of mitochondrial metabolism, CHCHD2 was identified as a causative gene for Parkinson’s disease (PD), yet the relationship between CHCHD2 and neurodegenerative dementia is not well understood. We directly sequenced the entire coding region of CHCHD2 gene in 150 AD patients, 84 FTD patients, and 417 controls. Four rare putative pathogenic variants of CHCHD2, including rs142444896 (c.5C>T, p.P2L), rs752705344 (c.15C>G, p.S5R), rs145190179 (c.94G>A, p.A32T), and rs182992574 (c.255T>A, p.S85R) were identified from a cohort composed of 150 AD and 84 FTD patients. These results suggest the CHCHD2 gene may an important role in other neurodegenerative disorders from our dementia study in China.
INTRODUCTION
As one important member of CHCHD protein family, CHCHD2 is a multifunctional protein and takes part in regulation of mitochondrial metabolism and synthesis of respiratory chain components [1]. In 2015, CHCHD2 was identified as a novel causative gene for familial autosomal dominant Parkinson’s disease (PD) through next generation sequencing [2]. To date, CHCHD2 mutations have been found in familial and sporadic PD in both Caucasian and Asian populations [3–6]. Additionally, it was reported that rare variants in CHCHD2 might be a risk factor for Lewy body disorders (DLB), which expanded the disease spectrum caused by CHCHD2 variants to neurodegenerative disease. Meanwhile, CHCHD10, the very close homologue of CHCHD2, is believed to be causative of frontotemporal dementia (FTD) and amyotrophic lateral sclerosis according to recent study [7].
As well known, Alzheimer’s disease (AD) is the most common progressive neurodegenerative disease among the elderly, which accounts for an estimated 60–80% of all dementia cases [8]. It is characterized by memory and other cognitive decline, a variety of neuropsychiatric symptoms, and restriction in daily living activities, which places a considerable burden on the society [9]. AD is generally considered a polygenic disease resulting from complex interactions among multiple genes and environmental factors. Four AD causative genes have been reported, including the amyloid precursor protein gene (APP) [10], and the presenilin 1 (PSEN1) [11] and presenilin 2 genes (PSEN2) [12]. Additionally, at least 10 loci have been found contributing to the risk of AD, including PICALM, CLU, CR1, BIN1, CD2AP, EPHA1, MS4A4A, CD33 [14], and ABCA7 [15]. Meanwhile, as the second most common neurodegenerative dementia, FTD primarily affects individuals younger than 65 years and comprises about 10–20% of all dementias worldwide [16]. The clinical presentations of FTD are heterogeneous mainly with two clinical variants: behavioral variant [17] and language variants (semantic dementia and progressive nonfluent aphasia) [18]. Patients with FTD can develop motor neuron disease (FTD-MND) or atypical parkinsonism [19]. Clinical presentations of FTD associated motor dysfunction are variable depending on the symptoms, age of symptom onset, the symptoms occurrence into the disease course, and disease duration. FTD also has strong genetic basis and seven FTD causative genes have been reported, including C9orf72, CHMP2B, FUS, GRN, MAPT, TARDBP, and VCP from our and other previous studies [20, 21]. Furthermore, at least two loci have been found contributing to the risk of FTD, including the HLA locus at 6p21.3 and locus at 11q14, encompassing RAB38/CTSC [22]
Therefore, what happens to CHCHD2 dysfunction in AD and FTD? A series of investigations revealed that CHCHD2 dysfunction affects mitochondrial structure and results in mitochondrial dysfunction and even cell apoptosis [23, 24], a prevalent feature of many neurodegenerative diseases including PD and DLB and potentially AD and FTD. Therefore, we put forward a hypothesis that mutation of CHCHD2 may be also causative of AD and/or FTD. Generally, this study aims to investigate whether CHCHD2 mutation is associated with both AD and FTD.
MATERIALS AND METHODS
Study sample
150 AD patients (female:male = 79:71) and 84 FTD patients (female:male = 44:40) were recruited from the outpatient neurology clinics of the Ruijin Hospital affiliated to Shanghai Jiao Tong University School of Medicine, Huashan Hospital affiliated to Fudan University, and Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China. The mean age at recruitment for AD was 73.8 years old (SD = 9.2) and Mini-Mental Scale Examination (MMSE) was 15.8 (SD = 5.5). The mean age at recruitment of FTD was 61.4 years old (SD = 9.8) and MMSE was 13.5 (SD = 8.8) and 39 patients with behavioral variant FTD and 45 patients with primary progressive aphasia. All patients underwent a set of standardized neurological examinations by two or three neurologists who specialized in dementia. The diagnosis of AD and FTD was based on the NINCDS-ADRDA criteria for AD [25] and the Work Group on FTD and Pick’s disease [16], respectively. 417 healthy controls (female:male = 236:181) were recruited from the hospital at the same period of time, and mild cognitive impairment was excluded. The average of age at recruitment and MMSE scores is 67.99 years old (SD = 9.56) and 28.95 (SD = 1.13), respectively. This study was approved by the ethics committee of the Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University.
Genetic analysis
Genomic DNA was isolated using a QIAGEN kit from peripheral blood leukocytes from all subject. We performed mutation analysis on all the exons of CHCHD2 using Sanger sequencing. The quality and quantity of DNA were assessed with a fluorometer. All DNA samples were normalized to 50 ng/mL. Each PCR product was sequenced using identical forward and reverse primers by Sanger sequencing on an ABI 3100 automated sequencer (Applied Biosystems, Foster City, CA). Alignment and analysis were carried out with the biosoftware DNAStar (DNAStar, Inc. Madison, WI). The identified positive genetic variants were screened in healthy controls and checked with SNPs database (http://www.ncbi.nlm.nih.gov/snp) and conservative protein amino acid alignment across different species (http://www.uniprot.org) to confirm which plays a causal role. Furthermore, SIFT tool (http://sift.jcvi.org/), Polyphen-2 tool (http://genetics.bwh.harvard.edu/pph2/[tics.bwh.harvard.edu/pph2/]), Mutation Taster tool (http://www.mutationtaster.org) (Table 1) and InterVar (http://wintervar.wglab.org) were used to predict functional consequences of the identified novel missense variants.
Rare variants in CHCHD2 gene for AD and FTD
AAO, age at onset; AD, Alzheimer’s disease; FTD, frontotemporal dementia; SD, semantic dementia; bvFTD, behavioral variant FTD.
RESULTS
By directly sequencing the entire coding region of CHCHD2 gene in 150 AD and 84 FTD patients, four rare variants of CHCHD2 (Table 1, Fig. 1), including rs142444896 (c.5C>T, p.P2L, 2/150 in AD, 1/84 in FTD, probably damaging in-silico analysis), rs752705344(c.15C>G, p.S5R, 1/150 in AD, probably damaging in-silico analysis) in exon1, rs145190179 (c.94G>A, p.A32T, 1/150 in AD, probably damaging in-silico analysis) and rs182992574 (c.255T>A, p.S85R, 1/84 in FTD, probably damaging in-silico analysis) in exon2, were identified. Then we submitted these four variants for further analysis to determine their allele frequencies in 417 controls. These gene variants were not found in 417 control subjects. We further analyzed the APOE gene in four AD cases and common FTD causative genes in the two FTD cases carrying CHCHD2 gene mutations. We found one out of four AD cases carried the APOE ɛ4 allele and our previous study showed that none of the FTD cases carried the MAPT, GRN, CHCHD10, or C9orf72 hexanucleotide repeat expansions [21].

Sanger sequencing showed CHCHD2 mutations in patients with dementia. A) CHCHD2 mutation (c.5C>T, p.P2L) was identified in one sporadic FTD patient. The arrow in the sequence diagrams indicates the mutation site. B) CHCHD2 mutation (c.15C>G, p.S5R) was identified in one sporadic AD. The arrow in the sequence diagrams indicates the mutation site. C) CHCHD2 mutation (c.94G>A, p.A32T) was identified in one sporadic AD. The arrows in the sequence diagrams represent the mutation site. D) CHCHD2 mutation (c.255T>A, p.S85R) was identified in one sporadic FTD patient. The arrow in the sequence diagram represents the mutation site. E) Protein sequence alignment of CHCHD2 in primary species and the evolutionary conservation of the CHCHD2 mutations using Uniprot software (http://www.uniprot.org).
Among them, 1) interestingly, the rare variant rs142444896 was found in both AD and FTD patients. Clinically, the semantic dementia patient carrying the rs142444896 variant had disease onset at age 58 and his MMSE was 22 after 4 years into the disease course. The two AD patients carrying the rs142444896 variant had disease onset at 65 and 74 years old, respectively, and their MMSE scores were 18 and 12 after 3 and 4 years into the disease course. 2) The AD patient carrying the rs752705344 variant had disease onset at age 70 and his MMSE was 15 after 5 years into the disease course. 3) The AD patient carrying the rs145190179 variant had disease onset at age 67 and his MMSE was 17 after 3 years into the disease course. 4) The rare variant rs182992574 was found in one sporadic behavioral variant FTD patient, who had stereotypic behaviors at age 65 and subsequently developed tremor and rigidity of the right arm. By age 68, his memory and executive function had substantially declined.
DISCUSSION
In this study, we found a total of three rare exonic variants in four sporadic AD patients, including rs142444896, rs752705344, and rs145190179. All of them were located in the mitochondrial targeting sequence (MTS) of CHCHD2, their amino acids are conservative across different species (Fig. 1), and they are predicted to be “Disease causing” by in silico analysis including Mutation Taster, Polyphen-2, and SIFT software (Table 2). Variants in MTS of CHCHD2 are likely to change mitochondrial metabolism. According to the 2015 ACMG-AMP regulations, rs752705344 and rs145190179 fall into the class of “uncertain significance”, but rs142444896 falls into the class of “benign” by InterVar. The high prevalence of 2.7% (4/150) of CH CHD2 mutations in Chinese AD patients indicates its potential important contribution to AD in China.
In silico analyses of rare variants in CHCHD2 gene
In addition, we found two sporadic FTD patients carrying rs142444896 (c.5C>T, p.P2L) and rs182992574 (c.255T>A, p.S85R) respectively, suggesting that CH CHD2 might play a role in the pathogenesis of FTD. The rs142444896 is located in the MTS of CHCHD2, and is predicted to be “Disease causing” by in silico analysis including Mutation Taster, Polyphen-2, and SIFT software. However, rs182992574 is predicted to be “Disease causing” by Mutation Taster and SIFT software, but “Benign” by Polyphen-2 (Table 2). Their amino acids are conservative across different species (Fig. 1). According the 2015 ACMG-AMP rules, rs182992574 falls into the class of “uncertain significance” by InterVar. Maybe CH CHD2 variants are common risks for AD and FTD in China because this study shows that CHCHD2 variants account for up to 2.7% AD (4/150) and 2.4% FTD (2/84).
The present study has several limitations. Firstly, we did not include cases with familial AD or FTD, so the potential role of CHCHD2 variants in familial disease remains unclear. Secondly, the diagnosis of AD or FTD patients was confirmed by clinical phenotype rather than pathology. Therefore, our result should be replicated in a large cohort in future.
To our knowledge, this is the first study assessing the frequency of CHCHD2 mutations in FTD and AD over the world. Our results showed that CHCHD2 mutations might be a common genetic risk in Chinese AD and FTD patients. Multi-center clinical cooperative studies involving larger number of subjects are necessary to validate CHCHD2 that is an AD and FTD causative gene and to explore the variants associated clinical and pathological phenotypes. Furthermore, functional studies shall be conducted to elucidate the molecular functional alternation caused by those variants.
Footnotes
ACKNOWLEDGMENTS
We thank the patients and their families for their participation in this project. This study was supported by the National Natural Science Foundation of China (No. 81171019; 81671043), National 973 Project (2013CB530900, 2013CB530904, 2011ZX09307-001-03) and the “Shuguang Program” supported by Shanghai Education Development Foundation and Shanghai Municipal Education Commission.
