CHCHD10S59L/+ mouse model: Behavioral and neuropathological features of frontotemporal dementia.

CHCHD10-related disease causes a spectrum of clinical presentations including mitochondrial myopathy, cardiomyopathy, amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). We generated a knock-in mouse model bearing the p.Ser59Leu (S59L) CHCHD10 variant. Chchd10S59L/+ mice have been shown to phenotypically replicate the disorders observed in patients: myopathy with mtDNA instability, cardiomyopathy and typical ALS features (protein aggregation, neuromuscular junction degeneration and spinal motor neuron loss). Here, we conducted a comprehensive behavioral, electrophysiological and neuropathological assessment of Chchd10S59L/+ mice. These animals show impaired learning and memory capacities with reduced long-term potentiation (LTP) measured at the Perforant Pathway-Dentate Gyrus (PP-DG) synapses. In the hippocampus of Chchd10S59L/+ mice, neuropathological studies show the involvement of protein aggregates, activation of the integrated stress response (ISR) and neuroinflammation in the degenerative process. These findings contribute to decipher mechanisms associated with CHCHD10 variants linking mitochondrial dysfunction and neuronal death. They also validate the Chchd10S59L/+ mice as a relevant model for FTD, which can be used for preclinical studies to test new therapeutic strategies for this devastating disease.

CHCHD10 and SLP2 control the stability of the PHB complex: a key factor for motor neuron viability.

CHCHD10 is an amyotrophic lateral sclerosis/frontotemporal dementia gene that encodes a mitochondrial protein whose precise function is unclear. Here we show that Coiled-Coil-Helix-Coiled-Coil-Helix Domain Containing protein 10 interacts with the Stomatin-Like Protein 2 and participates in the stability of the prohibitin complex in the inner mitochondrial membrane. By using patient fibroblasts and mouse models expressing the same CHCHD10 variant (p.Ser59Leu), we show that Stomatin-Like Protein 2 forms aggregates with prohibitins, found in vivo in the hippocampus and as aggresome-like inclusions in spinal motor neurons of Chchd10S59L/+ mice. Affected cells and tissues display instability of the prohibitin complex, which participates at least in part in the activation of the OMA1 cascade with OPA1 processing leading to mitochondrial fragmentation, abnormal mitochondrial cristae morphogenesis and neuronal death found in spinal cord and the hippocampus of Chchd10S59L/+ animals. Destabilization of the prohibitin complex leads to the instability of the mitochondrial contact site and cristae organizing the system complex, probably by the disruption of OPA1-mitofilin interaction. Thus, Stomatin-Like Protein 2/prohibitin aggregates and destabilization of the prohibitin complex are critical in the sequence of events leading to motor neuron death in CHCHD10S59L-related disease.

Mitochondrial defect in muscle precedes neuromuscular junction degeneration and motor neuron death in CHCHD10S59L/+ mouse.

Recently, we provided genetic basis showing that mitochondrial dysfunction can trigger motor neuron degeneration, through identification of CHCHD10 encoding a mitochondrial protein. We reported patients, carrying the p.Ser59Leu heterozygous mutation in CHCHD10, from a large family with a mitochondrial myopathy associated with motor neuron disease (MND). Rapidly, our group and others reported CHCHD10 mutations in amyotrophic lateral sclerosis (ALS), frontotemporal dementia-ALS and other neurodegenerative diseases. Here, we generated knock-in (KI) mice, carrying the p.Ser59Leu mutation, that mimic the mitochondrial myopathy with mtDNA instability displayed by the patients from our original family. Before 14 months of age, all KI mice developed a fatal mitochondrial cardiomyopathy associated with enhanced mitophagy. CHCHD10S59L/+ mice also displayed neuromuscular junction (NMJ) and motor neuron degeneration with hyper-fragmentation of the motor end plate and moderate but significant motor neuron loss in lumbar spinal cord at the end stage of the disease. At this stage, we observed TDP-43 cytoplasmic aggregates in spinal neurons. We also showed that motor neurons differentiated from human iPSC carrying the p.Ser59Leu mutation were much more sensitive to Staurosporine or glutamate-induced caspase activation than control cells. These data confirm that mitochondrial deficiency associated with CHCHD10 mutations can be at the origin of MND. CHCHD10 is highly expressed in the NMJ post-synaptic part. Importantly, the fragmentation of the motor end plate was associated with abnormal CHCHD10 expression that was also observed closed to NMJs which were morphologically normal. Furthermore, we found OXPHOS deficiency in muscle of CHCHD10S59L/+ mice at 3 months of age in the absence of neuron loss in spinal cord. Our data show that the pathological effects of the p.Ser59Leu mutation target muscle prior to NMJ and motor neurons. They likely lead to OXPHOS deficiency, loss of cristae junctions and destabilization of internal membrane structure within mitochondria at motor end plate of NMJ, impairing neurotransmission. These data are in favor with a key role for muscle in MND associated with CHCHD10 mutations.

Loss of MICOS complex integrity and mitochondrial damage, but not TDP-43 mitochondrial localisation, are likely associated with severity of CHCHD10-related diseases.

Following the involvement of CHCHD10 in FrontoTemporal-Dementia-Amyotrophic Lateral Sclerosis (FTD-ALS) clinical spectrum, a founder mutation (p.Gly66Val) in the same gene was identified in Finnish families with late-onset spinal motor neuronopathy (SMAJ). SMAJ is a slowly progressive form of spinal muscular atrophy with a life expectancy within normal range. In order to understand why the p.Ser59Leu mutation, responsible for severe FTD-ALS, and the p.Gly66Val mutation could lead to different levels of severity, we compared their effects in patient cells. Unlike affected individuals bearing the p.Ser59Leu mutation, patients presenting with SMAJ phenotype have neither mitochondrial myopathy nor mtDNA instability. The expression of CHCHD10S59L mutant allele leads to disassembly of mitochondrial contact site and cristae organizing system (MICOS) with mitochondrial dysfunction and loss of cristae in patient fibroblasts. We also show that G66V fibroblasts do not display the loss of MICOS complex integrity and mitochondrial damage found in S59L cells. However, S59L and G66V fibroblasts show comparable accumulation of phosphorylated mitochondrial TDP-43 suggesting that the severity of phenotype and mitochondrial damage do not depend on mitochondrial TDP-43 localization. The expression of the CHCHD10G66V allele is responsible for mitochondrial network fragmentation and decreased sensitivity towards apoptotic stimuli, but with a less severe effect than that found in cells expressing the CHCHD10S59L allele. Taken together, our data show that cellular phenotypes associated with p.Ser59Leu and p.Gly66Val mutations in CHCHD10 are different; loss of MICOS complex integrity and mitochondrial dysfunction, but not TDP-43 mitochondrial localization, being likely essential to develop a severe motor neuron disease.

Mutations in MDH2, Encoding a Krebs Cycle Enzyme, Cause Early-Onset Severe Encephalopathy.

MDH2 encodes mitochondrial malate dehydrogenase (MDH), which is essential for the conversion of malate to oxaloacetate as part of the proper functioning of the Krebs cycle. We report bi-allelic pathogenic mutations in MDH2 in three unrelated subjects presenting with early-onset generalized hypotonia, psychomotor delay, refractory epilepsy, and elevated lactate in the blood and cerebrospinal fluid. Functional studies in fibroblasts from affected subjects showed both an apparently complete loss of MDH2 levels and MDH2 enzymatic activity close to null. Metabolomics analyses demonstrated a significant concomitant accumulation of the MDH substrate, malate, and fumarate, its immediate precursor in the Krebs cycle, in affected subjects' fibroblasts. Lentiviral complementation with wild-type MDH2 cDNA restored MDH2 levels and mitochondrial MDH activity. Additionally, introduction of the three missense mutations from the affected subjects into Saccharomyces cerevisiae provided functional evidence to support their pathogenicity. Disruption of the Krebs cycle is a hallmark of cancer, and MDH2 has been recently identified as a novel pheochromocytoma and paraganglioma susceptibility gene. We show that loss-of-function mutations in MDH2 are also associated with severe neurological clinical presentations in children.

Inactivation of Pif1 helicase causes a mitochondrial myopathy in mice.

Mutations in genes coding for mitochondrial helicases such as TWINKLE and DNA2 are involved in mitochondrial myopathies with mtDNA instability in both human and mouse. We show that inactivation of Pif1, a third member of the mitochondrial helicase family, causes a similar phenotype in mouse. pif1-/- animals develop a mitochondrial myopathy with respiratory chain deficiency. Pif1 inactivation is responsible for a deficiency to repair oxidative stress-induced mtDNA damage in mouse embryonic fibroblasts that is improved by complementation with mitochondrial isoform mPif1(67). These results open new perspectives for the exploration of patients with mtDNA instability disorders.

CHCHD10 mutations promote loss of mitochondrial cristae junctions with impaired mitochondrial genome maintenance and inhibition of apoptosis.

CHCHD10-related diseases include mitochondrial DNA instability disorder, frontotemporal dementia-amyotrophic lateral sclerosis (FTD-ALS) clinical spectrum, late-onset spinal motor neuropathy (SMAJ), and Charcot-Marie-Tooth disease type 2 (CMT2). Here, we show that CHCHD10 resides with mitofilin, CHCHD3 and CHCHD6 within the "mitochondrial contact site and cristae organizing system" (MICOS) complex. CHCHD10 mutations lead to MICOS complex disassembly and loss of mitochondrial cristae with a decrease in nucleoid number and nucleoid disorganization. Repair of the mitochondrial genome after oxidative stress is impaired in CHCHD10 mutant fibroblasts and this likely explains the accumulation of deleted mtDNA molecules in patient muscle. CHCHD10 mutant fibroblasts are not defective in the delivery of mitochondria to lysosomes suggesting that impaired mitophagy does not contribute to mtDNA instability. Interestingly, the expression of CHCHD10 mutant alleles inhibits apoptosis by preventing cytochrome c release.

A mitochondrial origin for frontotemporal dementia and amyotrophic lateral sclerosis through CHCHD10 involvement.

Mitochondrial DNA instability disorders are responsible for a large clinical spectrum, among which amyotrophic lateral sclerosis-like symptoms and frontotemporal dementia are extremely rare. We report a large family with a late-onset phenotype including motor neuron disease, cognitive decline resembling frontotemporal dementia, cerebellar ataxia and myopathy. In all patients, muscle biopsy showed ragged-red and cytochrome c oxidase-negative fibres with combined respiratory chain deficiency and abnormal assembly of complex V. The multiple mitochondrial DNA deletions found in skeletal muscle revealed a mitochondrial DNA instability disorder. Patient fibroblasts present with respiratory chain deficiency, mitochondrial ultrastructural alterations and fragmentation of the mitochondrial network. Interestingly, expression of matrix-targeted photoactivatable GFP showed that mitochondrial fusion was not inhibited in patient fibroblasts. Using whole-exome sequencing we identified a missense mutation (c.176C>T; p.Ser59Leu) in the CHCHD10 gene that encodes a coiled-coil helix coiled-coil helix protein, whose function is unknown. We show that CHCHD10 is a mitochondrial protein located in the intermembrane space and enriched at cristae junctions. Overexpression of a CHCHD10 mutant allele in HeLa cells led to fragmentation of the mitochondrial network and ultrastructural major abnormalities including loss, disorganization and dilatation of cristae. The observation of a frontotemporal dementia-amyotrophic lateral sclerosis phenotype in a mitochondrial disease led us to analyse CHCHD10 in a cohort of 21 families with pathologically proven frontotemporal dementia-amyotrophic lateral sclerosis. We identified the same missense p.Ser59Leu mutation in one of these families. This work opens a novel field to explore the pathogenesis of the frontotemporal dementia-amyotrophic lateral sclerosis clinical spectrum by showing that mitochondrial disease may be at the origin of some of these phenotypes.

The human MSH5 (MutSHomolog 5) protein localizes to mitochondria and protects the mitochondrial genome from oxidative damage.

MutS homologs play a central role in maintaining genetic stability. We show that MSH5 (MutSHomolog 5) is localized into the mitochondria of germ and somatic cells. This protein binds to mtDNA and interacts with the Twinkle helicase and the DNA polymerase gamma. hMSH5 stimulates mtDNA repair in response to DNA damage induced by oxidative stress. Furthermore, we observed a subsarcolemmal accumulation of hMSH5 in COX negative muscle fibers of patients presenting a mitochondrial myopathy. We report a novel localization for hMSH5 suggesting that this protein may have functions other than those known in meiotic recombination.

hMSH5 is a nucleocytoplasmic shuttling protein whose stability depends on its subcellular localization.

MSH5 is a MutS-homologous protein required for meiotic DNA recombination. In addition, recent studies suggest that the human MSH5 protein (hMSH5) participates to mitotic recombination and to the cellular response to DNA damage and thus raise the possibility that a tight control of hMSH5 function(s) may be important for genomic stability. With the aim to characterize mechanisms potentially involved in the regulation of hMSH5 activity, we investigated its intracellular trafficking properties. We demonstrate that hMSH5 possesses a CRM1-dependent nuclear export signal (NES) and a nuclear localization signal that participates to its nuclear targeting. Localization analysis of various mutated forms of hMSH5 by confocal microscopy indicates that hMSH5 shuttles between the nucleus and the cytoplasm. We also provide evidence suggesting that hMSH5 stability depends on its subcellular compartmentalization, hMSH5 being much less stable in the nucleus than in the cytoplasm. Together, these data suggest that hMSH5 activity may be regulated by nucleocytoplasmic shuttling and nuclear proteasomal degradation, both of these mechanisms contributing to the control of nuclear hMSH5 content. Moreover, data herein also support that in tissues where both hMSH5 and hMSH4 proteins are expressed, hMSH5 might be retained in the nucleus through masking of its NES by binding of hMSH4.

CRM1-dependent nuclear export and dimerization with hMSH5 contribute to the regulation of hMSH4 subcellular localization.

MSH4 and MSH5 are members of the MutS homolog family, a conserved group of proteins involved in DNA mismatch correction and homologous recombination. Although several studies have provided compelling evidences suggesting that MSH4 and MSH5 could act together in early and late stages of meiotic recombination, their precise roles are poorly understood and recent findings suggest that the human MSH4 protein may also exert a cytoplasmic function. Here we show that MSH4 is present in the cytoplasm and the nucleus of both testicular cells and transfected somatic cells. Confocal studies on transfected cells provide the first evidence that the subcellular localization of MSH4 is regulated, at least in part, by an active nuclear export pathway dependent on the exportin CRM1. We used deletion mapping and mutagenesis to define two functional nuclear export sequences within the C-terminal part of hMSH4 that mediate nuclear export through the CRM1 pathway. Our results suggest that CRM1 is also involved in MSH5 nuclear export. In addition, we demonstrate that dimerization of MSH4 and MSH5 facilitates their nuclear localization suggesting that dimerization may regulate the intracellular trafficking of these proteins. Our findings suggest that nucleocytoplasmic traffic may constitute a regulatory mechanism for MSH4 and MSH5 functions.

Association between MSH4 (MutS homologue 4) and the DNA strand-exchange RAD51 and DMC1 proteins during mammalian meiosis.

During meiotic prophase, chromosomes must undergo highly regulated recombination events, some of which lead to reciprocal exchanges. In yeast, MSH4, a meiosis-specific homologue of the bacterial MutS protein, is required for meiotic recombination. In mice, disruption of the Msh4 gene results in male and female infertility due to meiotic failure. To date, the implication of MSH4 mutations has not been established in human sterility. However, it is noteworthy that mutant mice exhibit a defect in the chromosome synapsis, strikingly similar to the clinical observations found in human infertility. As a step towards understanding the molecular mechanisms underlying the role of MSH4 in human gametogenesis, we decided to determine whether this protein interacts with recombination machinery enzymes. Our results provide biochemical evidence indicating that the human MSH4 protein physically interacts with both RAD51 and DMC1, two RecA homologues known to initiate DNA strand-exchange between homologous chromosomes. Immunolocalization analyses show that some MSH4 foci, located on mouse meiotic chromosomes, colocalize with DMC1/RAD51 complexes. Our data support the view that MSH4 is associated with the early meiotic recombination machinery in mammals. We consider the possibility that MSH4 is involved in the regulation of recombination events by exerting a function closely after DNA strand-exchange has been initiated.

The DNA mismatch-repair MLH3 protein interacts with MSH4 in meiotic cells, supporting a role for this MutL homolog in mammalian meiotic recombination.

The mismatch-repair (MMR) system plays a central role in maintaining genetic stability and requires evolutionarily conserved protein factors, including MutS and MutL homologs. Since the discovery of a link between the malfunction of post-replicative MMR and human cancers, a number of works have focused on the function of MutS and MutL homologs in the correction of replication errors. However, several MutS-like and MutL-like proteins also participate in meiotic recombination. The MutL homolog MLH3 has been recently identified in mammals. Several pieces of evidence support a role for this protein in post-replicative MMR. To investigate whether MLH3 also acts during meiotic recombination, we analyzed its expression in mammalian germ cells. The MLH3 gene is expressed in mouse meiotic cells and in human testis, and, as revealed by immunoprecipitation assays, the MLH3 protein is found in mouse spermatocytes. We further demonstrate that the meiosis-specific MSH4 protein, known to participate to meiotic recombination, is co-immunoprecipitated with MLH3 from mouse meiotic cell extracts. In addition, the two MLH3 protein isoforms potentially expressed in human testis (hMLH3 and hMLH3 Delta 7) interact in vitro with the hMSH4 protein. These interaction data suggest that MLH3 is associated with MSH4 in mammalian meiotic cells, and strongly support the possibility that MLH3 plays a role in mammalian meiotic recombination.