Molecular Epidemiology of Mitochondrial Cardiomyopathy: A Search Among Mitochondrial and Nuclear Genes

Mitochondrial Cardiomyopathy (MCM) is a common manifestation of multi-organ Mitochondrial Diseases (MDs), occasionally present in non-syndromic cases. Diagnosis of MCM is complex because of wide clinical and genetic heterogeneity and requires medical, laboratory, and neuroimaging investigations. Currently, the molecular screening for MCM is fundamental part of MDs management and allows achieving the definitive diagnosis. In this article, we review the current genetic knowledge associated with MDs, focusing on diagnosis of MCM and MDs showing cardiac involvement. We searched for publications on mitochondrial and nuclear genes involved in MCM, mainly focusing on genetic screening based on targeted gene panels for the molecular diagnosis of the MCM, by using Next Generation Sequencing. Here we report twelve case reports, four case-control studies, eleven retrospective studies, and two prospective studies, for a total of twenty-nine papers concerning the evaluation of cardiac manifestations in mitochondrial diseases. From the analysis of published causal mutations, we identified 130 genes to be associated with mitochondrial heart diseases. A large proportion of these genes (34.3%) encode for key proteins involved in the oxidative phosphorylation system (OXPHOS), either as directly OXPHOS subunits (22.8%), and as OXPHOS assembly factors (11.5%). Mutations in several mitochondrial tRNA genes have been also reported in multi-organ or isolated MCM (15.3%). This review highlights the main disease-genes, identified by extensive genetic analysis, which could be included as target genes in next generation panels for the molecular diagnosis of patients with clinical suspect of mitochondrial cardiomyopathies.


Introduction
Cardiomyopathies (CM) represent a wide group of human disorders, extending from isolated cardiac diseases to cardiac involvement in the context of complex diseases, such as neuromuscular disorders, mitochondrial pathology, and metabolic/infiltrative/storage disease [1,2]. An extensive clinical examination, comprising a systematic screening of several organs, should be, initially, conducted to define whether the cardiomyopathy is isolated or it is part of a multisystem disorder. Furthermore, the family history, specific symptoms, and the presence of typical clinical and instrumental findings may help the physicians in the differential diagnosis between mitochondrial cardiomyopathy (MCM) and sarcomere disease. Indeed, mitochondrial diseases (MDs) may manifest with structural and clinical features of hypertrophic (HCM), dilated (DCM), or restrictive (RCM) The most frequent cardiac disorders refereed to mitochondrial dysfunction are cardiomyopathies [3,16,25]. Although the true prevalence of mtDNA-related CM is unknown, based on the prevalence of mtDNA disease and frequency of cardiac manifestations, about 1/10,000-15,000 of the general population is affected [7]. Furthermore, several studies report different prevalence between pediatric and adult population, with cardiomyopathy estimated to occur in 20-40% of children with MDs [16,22,47,48].
Mitochondrial cardiomyopathy is a myocardial disorder, without concomitant coronary artery disease, valvular disease, congenital heart disease, and hypertension, characterized by abnormal myocardial structure and/or function, subsequent to genetic alterations and resulting in the impairment of the mitochondrial respiratory chain [4]. Hypertrophic cardiomyopathy, dilated cardiomyopathy, and left ventricular non compaction (LVNC) are the main manifestations of MCM. Histiocytoid cardiomyopathy (HICMP) (Purkinje fiber dysplasia) and restrictive cardiomyopathy, as well as Takotsubo syndrome (TTS) have been also observed [16,19].
Moreover, MCM, in addition to the myocardium, may affect other cardiac tissue, such as the cardiac conduction system, cardiac valves, aortic root, coronary arteries or the pericardium, and connective tissue, thus leading impulse generation/conduction insufficiency, valvular diseases, systolic dysfunction, or heart failure and pulmonary arterial hypertension (PAH). Conduction system disorders arise in 5-10% of patients with MD, especially in patients with associated neuromuscular disease [49].
The abnormal heart-muscle structure and/or function is due to genetic defects, impairing the functionality of the mitochondrial RC complexes, their assembly and stability factors, tRNAs, rRNAs, mtDNA maintenance, and coenzyme Q10 (CoQ10) synthesis [19,50].
The cardiac involvement in MDs may be already evident at onset, but in some cases, it may develop over time. More rarely, in patients without or with mild multisystem involvement, the heart may be the only or the main clinically affected organ.
The imbalance of mitochondrial function in MCM is supported by a large amount of scientific publications, investigating the cardiological aspect both as one of the manifestations of a syndromic mitochondrial disorder and, in a smaller number of cases, as an isolated cardiac disease. In particular, Whabi et al. showed that myocardial involvement in mtDNA diseases is progressive and, in association with other cardiovascular manifestations and risk factors (i.e., intraventricular conduction block, diabetes, premature ventricular contractions, and left ventricular hypertrophy-LVH), is an independent predictor of morbidity and early mortality [23].
Mutations in nuclear and mitochondrial genes encoding mitochondrial RC complex subunits and mitochondrial RC complex assembly factors have been associated with HCM [54][55][56][57][58]. Furthermore, mutations in nuclear TSFM, encoding a mitochondrial translation elongation factor, have been identified in HCM or DCM with multi-organ disease [59].
Point mutations in mtDNA, particularly in mt-tRNA genes, seem to be more frequently associated with hypertrophic phenotype.
Left ventricular hypertrophy has been identified in 38-56% of patients harboring the m.3243A>G mutation, in the tRNA for Leucine gene (MTTL), which is the most common mutation involved in several MDs, also reported in DCM patient [25].
Furthermore, heart failure with impaired systolic function and ventricular dilatation, leading sudden death has been reported in patients with LVH bearing the mutations m.3243A>G in MTTL gene or m.8344A>G in MTTK gene [4,60]. Sudden death may occur in patients bearing the m.3243A>G mutation also in absence of myocardial involvement [61]. Table 1 shows the main nuclear and mitochondrial gene mutations associated with different cardiological phenotype [62][63][64][65].

More Rare Cardiomyopathies
Dilated cardiomyopathy occurs more rarely in MD patients, in which DCM may represent the initial manifestation of cardiac involvement [66]. Furthermore, some studies found DCM more frequently associated with mitochondrial diabetes (MIDD, Maternally Inherited Diabetes and Deafness) (OMIM 520000) [67]. More commonly, the cardiac chamber dilation and systolic dysfunction represent the progression of pre-existing hypertrophy, reported in about 10% of HCM patients [68]. Mutations in tRNA or in genes encoding subunits of the mitochondrial RC complexes as well as single or large-scale mtDNA deletion and/or depletion have been observed (Table 1) [7,60,69]. Mitochondrial cardiomyopathy may also appear as RCM, LVNC, or HICMP [70,71].
Mutations in mtDNA are common in LVNC, whereas sarcomere or ion channel genes variants account for few LVNC cases [72]. LVNC is caused by defects during cardiac development with abnormal myofibrils compaction and may evolve towards progressive ventricular dilatation and systolic dysfunction. This disease is commonly reported as a cardiac manifestation in multisystem diseases, especially in pediatric populations [4,73,74]. Histiocytoid cardiomyopathy is another rare cardiomyopathy, showing severe cardiac arrhythmias or dilated cardiomyopathy and arising in infancy or childhood [75]. Autosomal recessive, X-linked, and maternal transmission have been described; nevertheless, several reports indicate that HICMP is a cardiac manifestation of MDs and should be considered as MCM [76][77][78]. Histiocytoid cardiomyopathy is characterized by infiltration of the myocardium by pathognomonic histiocyte-like cells with foamy or granular cytoplasm full of morphologically abnormal mitochondria, showing significantly decreased succinate-cytochrome C reductase (SCR) or NADH-cytochrome C reductase activity [78]. Known molecular defects associated with HICMP are in Table 1.

Mitochondrial DNA in Conduction System and Coronary Heart Disease
MDs are responsible for defects of the contractile cardiac system, leading, indirectly, to dysfunction in cardiac electrical function. Arrhythmogenic cardiac diseases, including atrioventricular (AV) blocks, ventricular and supraventricular arrhythmias, associated with pre-excitation syndrome, represent the most common abnormalities of the cardiac electric system in MDs. Reduced ATP synthesis and increased ROS production, caused by the mitochondrial dysfunction, can directly damage the cardiomyocyte excitability.
In patients with mitochondrial dysfunction, the reduced ATP synthesis leads to decreased ATP/ADP which results in constitutive opening of the ATP-sensitive potassium (KATP) channels on the sarcoplasmic membrane; this reduces electrical transmission, by creating current sinks in the myocardium and decreases refractory periods, both increasing electrical instability and consequently, the arrhythmic risk [79].
In MDs patients, progression to high-grade AV block is often unexpected; hence, a fast identification of any conduction system disease is required for a timely intervention. In these patients, pacemaker is usually proposed on early PR-interval prolongation or bundle branch block [82].
Furthermore, coronary heart disease (CHD) may be a manifestation of MDs. CHD characterized by atherosclerosis of epicardial coronary arteries leading to vessel stenosis. Severe ischemia may lead to myocardial infarction and angina pectoris [83]. Genetic alterations and environmental factors increasing ROS production as well as insufficient antioxidant mechanisms, concur to the pathogenesis of atherosclerosis [84]. Numerous studies show that mtDNA mutations, altered mitochondrial DNA copy number (mtDNA-CN) and specific mtDNA haplogroups are involved in the CHD and myocardial infarction [85][86][87][88][89]. In particular, Kofler et al., found that mitochondrial haplogroup T was associated with CHD in Middle European Caucasian populations [86]. Mitochondrial mutations involved in atherosclerosis development and its clinical manifestations are showed in Table 1. These mutations cause defects in some mitochondrial respiratory chain proteins and tRNAs, lead-ing their reduced concentration with deficit of mitochondrial function, and contributing to the development of atherosclerosis [90,91].
Barth syndrome (OMIM 302060) is an X-linked disorder, whose most common cardiac manifestations are LVNC and DCM, while HCM is uncommon. Barth syndrome is caused by mutations in the gene coding for tafazzin (TAZ), which is an inner mitochondrial membrane phospholipid transacylase with key role in remodeling of cardiolipin [93].
Sengers syndrome (acylglycerol kinase deficiency) is characterized by 3-methylglutaconic aciduria. Patients show HCM, myopathy, congenital cataracts, lactic acidosis, and exercise intolerance. Heart failure causes death in half of patients within the first year of life. The syndrome is caused by mutations in AGK gene, coding an acylglycerol kinase involved in the assembly of the mitochondrial adenine nucleotide transporter ANT1 [94]. The transmembrane protein 70, encoded by TMEM70 gene, is a protein located in the mitochondrial inner membrane involved in assembling and stabilizing of the mitochondrial respiratory chain Complex I and V proteins.
Alterations in TMEM70 cause Complex V deficiency but are reported also in the less severe Complex I deficiency and in combined OXPHOS deficiency [95].
Mutations in TMEM70 gene (Table 1) are the most common cause of ATP synthase deficiency, resulting in a multi-organ mitochondrial disease, known as neonatal mitochondrial encephalo-cardiomyopathy, which is characterized by a large variety of symptoms, including 3-methylglutaconic aciduria, lactic acidosis, mitochondrial myopathy, and HCM [96]. The main known mutations in this gene are c.317-2 A>G, located in the splice site of TMEM70 intron 2, which leads to aberrant splicing and loss of TMEM70 transcript, with most patients not surviving the neonatal period and the c.366A>T, (p.Tyr112Ter) resulting in Nuclear Type 2 Mitochondrial Complex V deficiency, showing HCM. The study of Cameron J.M. et colleagues on mitochondrial morphology in patients with mutations in this gene, revealed abnormal mitochondria, with whorled cristae and disrupted nucleoid clusters of mtDNA [97].
Finally, HCM is the cardiac manifestation of Friedreich ataxia, an autosomal recessive neurodegenerative disease caused by mutations in frataxin (FXN) gene (Table 1) encoding a mitochondrial iron-binding protein involved in the synthesis of the iron-sulphur (Fe-S) clusters (ISCs). ISCs are inorganic redox-active protein cofactors required for the activity of the OXPHOS complexes [98,99]. The absence of FXN protein debars the ISC-assembly complex and increases the iron deposition and its oxidation. Cardiac dysfunction, mainly congestive heart failure or arrhythmia but also stroke, ischaemic heart disease, and pneumonia, are the most common causes of death in these patients [100].

Multiorgan Clinical Expression of Mitochondrial Cardiomyopathy
The clinical expression of MCM is often accompanied by multisystem manifestations presenting neuromuscular, endocrine, and neuro-sensorial features [121,122]. Most patients with neuromuscular signs show creatine kinase enzyme in reference values or slightly elevated levels, while higher liver enzyme levels have been found in up to 10% of patients. Renal features may include nephritic syndrome, tubulopathy, tubulointerstitial nephritis, and nonspecific renal failure. Endocrinopathies include hypothyroidism, hypoparathyroidism, diabetes mellitus, adrenocorticotropic hormone deficiency, and hypogonadism. Gastrointestinal symptoms (diarrhea, constipation, abdominal pain, nausea, and chronic intestinal pseudo-obstruction) may also be present. The main ophthalmologic manifestation is retinitis pigmentosa. Sensorineural hearing loss occurs in 7% to 26% of patients, and its prevalence increases with age [4].

Diagnosis
Diagnosis of MCM is particularly complex because of wide clinical and genetic heterogeneity and requires a multidisciplinary approach, including extensive medical, laboratory and neuroimaging investigations. Physicians should suspect a possible MCM in patients with signs and symptoms of cardiomyopathy and multisystem involvement without a clear cause and refer the patients to a geneticist or other specialists with MDs expertise.
Genetic counselling is a fundamental part of the diagnostic workup and should be performed by specialized personnel. A detailed pedigree of the patient's family should be obtained, and all first-degree relatives should be referred to a specific cardiogenetic clinic. An integrated diagnostic approach, consisting of biochemical screening, histopathological studies, cardiological investigations, including magnetic resonance imaging, functional assays, and molecular genetic testing, should be conducted to reach a correct diagnosis. The identification of a causative mutation allows the presymptomatic identification and the follow up of family members. The advent of NGS techniques has revolutionized the diagnosis of MDs in terms of expanding the number of involved genes as well as the discovery of new genes potentially associated with the diseases.

Medical Examination
An initial careful clinical examination of several organs (including the evaluation of extra-cardiac features, i.e., diabetes and deafness) with different criteria for adults or children should initially be performed to determine whether the CM is an isolated pathology or part of a multisystem disease [4,16,121,123,124]. Cardiologists should suspect a mitochondrial dysfunction in patients presenting, in addition to cardiomyopathy, also hearing loss, renal dysfunction, diabetes mellitus, and peripheral myopathy.
A maternal inheritance pattern or family history of mitochondrial disease may support the suspicion of MCM [125].

Laboratory Investigation
As above reported, the laboratory investigations of suspected MD are very complex, and evaluation of selected biomarkers is necessary to guide genetic investigation. Recently, the Mitochondrial Medicine Society has provided consensus statements on the diagnosis and management of MDs [126,127].
Functional assays on tissue (typically skeletal muscle) based on measurement of RC complexes activity and mitochondrial oxygen consumption have been fundamental in the diagnosis of MDs, especially prior to the recent developments in genomics and remain important procedures to evaluate the mitochondrial function.
Although MDs present with a wide spectrum of clinical manifestations, the skeletal muscle is frequently affected and represents an excellent post-mitotic tissue with distinctive histological and histochemical hallmarks of mitochondrial pathology (in primary mtDNArelated disease). The typical presence of ragged-red fibers (RRF), by using the modified Gomori trichrome stains, represents the histologic hallmark of MDs, resulting from a compensatory response to a RC biochemical defect, although these fibers are typically absent in children and in many adults are present only in late-stage disease [4,8]. Furthermore, the histochemical activities of the partially mtDNA-encoded cytochrome C oxidase complex IV (COX) and the fully nuclear-encoded succinate dehydrogenase complex II (SDH) are measured in the diagnosis of adult-onset MDs [127].
The dual COX/SDH stain is used to determine whether mitochondrial anomalies, observed on Gomori trichrome stains, (and Hematoxylin and Eosin stain) are due to mtDNA or nDNA alterations. Mutations in mtDNA impair mainly synthesis of mtDNAencoded complex IV subunits, resulting in a decreased or absent complex IV activity and the affected fibers show either partial or complete lack of brown staining (the so-called "COX-negative fibers").
In contrast, SDH staining depends on the enzymatic activity of the Complex II, which is entirely encoded by nDNA. This complex maintains a normal activity even when mtDNA mutations are present [8,128].
The more recent quadruple immunofluorescence technology, using fluorescently labelled antibodies against subunits of complexes I-IV, provides accurate data on the relative amount of complex I and complex IV and could contribute to the diagnosis of MDs [129].
Furthermore, electron microscopy of the affected muscle biopsy can sometimes highlight an increased number of mitochondria with variable size and shape. In particular enlarged and swollen mitochondria, with irregular cristae and paracrystalline inclusions, are showed [130]. This mitochondrial morphology reflects the perturbation of the rates of fission and fusion, essential processes for the healthy maintenance of mitochondria, leading to either fragmented, punctiform mitochondria, or excessively long or interconnected mitochondria [131].

Cardiological and Neuroimaging Investigations
As above reported, all the major cardiomyopathy phenotypes may be shown in MDs, but hypertrophic remodeling is the dominant pattern of cardiomyopathy in MDs.
The early stages of MCM are characterized by progressive diastolic dysfunction and heart failure with preserved ejection fraction [125]. The association of LV hypertrophy (with or without apical trabeculation) with systolic dysfunction has been reported as a typical evolution of MCM. T-wave abnormalities are an ECG sign of MD, even in absence of LV hypertrophy. In a cohort of 32 MDs subjects, Limongelli et al. found an abnormal ECG in up to 68% patients [53]. Furthermore, the presence of a short PR interval, various degrees of AV block or Wolff-Parkinson-White syndrome, which is common in patients with mtDNA mutations, also supports the diagnosis.
Regarding cardiac magnetic resonance findings, characteristic patterns of cardiac association might be shown in some MDs, as reported by Florian A. et colleagues [132]. For example, concentric LV hypertrophy, with intramural late gadolinium enhancement (LGE) in the inferolateral wall may be considered a distinctive feature in patients with KSS [25].

Molecular Analysis of Mitochondrial Cardiomyopathies
Traditionally, for different pathologies as well as MDs, including MCM, molecular diagnosis is achieved through "candidate gene" studies. This approach is based on recognizable clinical manifestations and family history, followed by molecular testing performed by Sanger sequencing of known genes [133]. Indeed some MD phenotypes are linked to molecular defects in one or few mitochondrial and nuclear genes [12].
To date, for patients suspected of having maternally inherited mtDNA disorders, the molecular diagnosis is still initially based on the mtDNA Sanger sequencing. In adult onset cases, mtDNA aetiology is recurrent and mtDNA sequencing for point mutations or the whole mitochondrial genome, when common point mutations are not detected, remains the first approach [134].
Moreover, Sanger sequencing does not detect either heteroplasmic mutations below 15-20% or allow measuring the degree of heteroplasmy. It also does not show large deletions, which are, conversely, detectable by other different technologies (i.e., array comparative genome hybridization) [135][136][137].
The genetic diagnosis of many MDs, including MCM, is further impaired by numerous nuclear genes whose defects may cause mitochondrial diseases [28].
Nowadays, for the above mentioned reasons, the diagnosis of MCM based on clinical/biochemical and instrumental examinations and sequencing of candidate genes one by one, is not a practical approach because it leaves many undiagnosed cases.
Advances in genomics and high-throughput sequencing technologies, such as NGS, have revolutionized the diagnostic setting of MDs, allowing identification of a large number of nuclear genes simultaneously, but also providing the identification and an accurate measure of heteroplasmy, in case of mtDNA mutations [29,30,[138][139][140][141][142][143]. For making diagnosis of many complex diseases, characterized by a wide clinical spectrum and heterogeneous genetic involvement, next generation sequencing based approach is available and sometimes necessary [144,145].
NGS screening encompasses different configurations, ranging from a small number of genes (targeted gene panels: i.e., Complex I genes) to hundred or more genes (targeted exome sequencing: i.e., Mito Exome, that includes about 1500 genes related to mitochondrial structure and function), and even up to Whole Exome Sequencing (WES) [122,146,147]. The choice depends by clinical, laboratory, and instrumental data, based on the clinical indication, family history, biochemical, and histopathological findings and imaging studies. The costs, time, acceptable depth of coverage and the management of off-target findings have to be also evaluated.

Variant Interpretation
The main difficulty in MDs molecular diagnosis, including MCM, is to distinguish rare or novel causative mutations from non-pathogenic polymorphisms in known/unknown genes [29].
Interestingly, rare variants may be also common in isolated genetic groups such as pathogenic mutations, which may show a higher carrier frequency within isolated populations due to the founder effect.
In the last decade, large-scale sequencing projects, integrating data from large population, have improved enormously the variants interpretation. In this context, exome data from different sequenced patients and healthy controls of the same population may contribute to assess the pathogenic role of the variant detected in the patient.
The databases of known nuclear/mitochondrial pathogenic mutations publicly accessible to define the MDs/MCM genetic diagnosis are listed in Supplementary data: online database.
Moreover, the correct classification of novel variants needs functional evidence, conferred by studies often complicated to perform and requiring specific laboratory competences [148,149]. Conversely, in silico evaluations, using several bioinformatics tools, are used to predict the likelihood of pathogenicity of missense and splicing variants.  Guidelines from the American College of Medical Genetics aim to standardize variants interpretation classifying them as "pathogenic", "likely pathogenic", "uncertain significance variant" (VUS), "likely benign", and "benign" considering simultaneously the weight of different criteria, also including in silico analysis [120,150].
However, the significance and the possible consequences of VUS, often detected by NGS technologies, are difficult to establish. Indeed, VUS interpretation and their consequent clinical management is a major challenge following NGS-based molecular testing. In this framework, an important step in the molecular MDs diagnostic approach is to investigate the family segregation of the detected variant in comparison with the clinical phenotype within the family. The detection of the same variant in affected family members, as well as in unrelated individuals with similar phenotype, may provide a strong additional data to support its possible pathogenic role.
Furthermore, interpretation of mtDNA variants for MDs, including MCM, should consider heteroplasmy level, maternal inheritance, and typical clinical manifestations as neuropathy and hearing loss, which provide accurate information for genetic counselling.

Literature Review
We reviewed and collected the literature data from July 1990 to October 2020, (one papers published in January 2021) by searching publications including genetic data from patient's cohort showing mitochondrial cardiomyopathy or mitochondrial disorder with cardiac involvement. Database searching and Inclusion/Exclusion Criteria are reported in Supplementary data: online database. The objective was to investigate the genotype/phenotype correlation in several studies, including patients with syndromic and non-syndromic features of mitochondrial cardiomyopathy. For this aim we investigated (accessed on 2 October 2020) the most known databases, available online.
Furthermore, in recent years, especially since the introduction of NGS technology, several studies, using extensive genetic analysis, highlighted the main disease-genes associated with MCM and MDs showing cardiomyopathy; this has contributed to increase our knowledge on the molecular background of these diseases.
The second aim of this review was to identify disease-genes, by searching publications of the last ten years, in order to prove the utility of genetic tests in MCM diagnosis in patients with syndromic or non-syndromic MDs and to identify a targeted gene panel for the molecular diagnosis.

Summary of Evidence
We found twelve case reports, four case-control studies, eleven retrospective studies, and two prospective studies for a total of twenty-nine papers ( Table 2).
Due to the number of studies identified, it was not possible to describe each of them individually. However, in the Table 2, we show the type study (case-control, case report, retrospective study, prospective study) and briefly the aim of study for each paper. Although with different inclusion criteria, ranging from genotype data (i.e., presence of specific mutation) or phenotype manifestations (i.e., mitochondrial disease, cardiomyopathy, respiratory chain disease) all the selected studies report the number of patients with cardiac phenotype and phenotype/genotype correlation.
Before the advent of NGS technology, few of the genes and mutations were known to be associated with mitochondrial diseases. For example, (as also showed in Table  2) the MT-TL1 m.3243A>G was one of the main mutations known to cause MDs (i.e., MELAS and MIDD phenotype). The use of next generation sequencing technology has revolutionized the MDs diagnosis, leading to the discovery of about half of the overall three hundred currently known genes. Thus, by searching publications regarding the most recent advances in understanding the molecular genetic basis of mitochondrial cardiomyopathy during the last ten years [7,16,19,48,151,152,154,155,[174][175][176] and investigating the above mentioned databases, we identified the main nuclear and mitochondrial genes reported in association with mitochondrial cardiomyopathy and MDs showing cardiac manifestations or conduction defects.   Table 3 shows the 130 genes (mtDNA and nDNA genes) collected according to their function showed by colored squares, as reported in Figure 1. This Table can thus represent a gene panel, from literature and database search, performed for targeted genes selection and could be used for the molecular screening of patients with MCM suspect. Table 3. List of major mitochondrial and nuclear (mtDNA and nDNA) genes collected through literature and database research over the past ten years in association with mitochondrial heart disease.

Function
Gene           Table 3 associated with mitochondrial heart disease, according to their function. A large proportion of these genes (34.3%) encode for key proteins in oxidative phosphorylation system, either as directly OXPHOS subunits (22.8%) and as OXPHOS assembly factors (11.5%). Furthermore, mutations in several mitochondrial tRNA genes have been reported with multi-organ or isolated MCM (15.3%). Mainly HCM, but also DCM or histiocytoid cardiomyopathy, are the main cardiomyopathies associated with pathogenic variants in genes encoding mitochondrial tRNAs. Mutations in mt-tRNA may damage overall mitochondrial translation by alteration of functional mt-tRNAs.

Discussion
CMs are reported by the European Society of Cardiology (ESC) Working Group on Myocardial and Pericardial Diseases as structurally and functionally myocardial disorders, in the absence of coronary artery disease, hypertension, valvular heart disease, and congenital heart disease explaining the myocardial defect [177].
The contribution of clinical molecular biology in the evaluation and prevention of cardiological risk, in particular sudden cardiac death, and in the therapeutic approach [178], both in patients and in healthy subjects and athletes, is an important goal that involves many areas of the medical profession [179][180][181][182][183].
MCM, mainly HCM followed by DCM and LVNC or, more rarely, RCM and HICMP, are common cardiac symptoms of MDs and should be considered in patients with multiorgan involvement. Cardiac abnormalities, such as cardiomyopathy, arrhythmias, and conduction defects are, indeed, present in up to 60% of patients with MD and represent often a major determinant of morbidity and mortality [53]. The wide genetic heterogeneity associated with mitochondrial cardiomyopathies, supports the opinion that the "cardiological phenotype" should often be considered as a common feature of many mitochondrial disorders and more rarely may be the first, or even the only, clinical manifestation.
Moreover, because MCM may also occur in non-syndromic MDs or in patients with only mild multisystem involvement and the differential diagnosis with non-mitochondrial cardiomyopathy through cardiological and imaging investigations is often difficult a considerable number of MCM remain undiagnosed or delayed [125]. An early recognition of MCM is crucial to initiate therapeutic procedures avoiding heart failure and arrhythmias as well as mitochondrial crisis.
Currently, the molecular screening for MCM, is a fundamental part of the management of MDs that, through an integrated multidisciplinary diagnostic approach allows getting to the diagnosis.
Recently, the Mitochondrial Medicine Society has provided consensus statements on the diagnosis and management of MDs, including measurements of mitochondrial biomarkers in plasma, urine, and spinal fluid [126].
Moreover, in patients with suspected of MD, biochemical screening, as well as histopathological studies and detailed clinical/cardiological evaluation should be followed by genetic investigation to make a definitive diagnosis.
Molecular analysis of mtDNA may be, initially performed when a clear maternally inheritance is observed, while both the mtDNA and nDNA should be considered if the inheritance pattern is uncertain.
The first objective of this review was to identify the major genes frequently associated with mitochondrial cardiomyopathies, in order to propose, by using Next Generation Sequencing technology, a novel targeted gene panel for the molecular diagnosis of the diseases. For this purpose, we explored literature and public data concerning the cardiac manifestations in mitochondrial diseases/disorders, in order to investigate the genotype/phenotype correlation in patients with MCM. This data collection allowed us to extract twenty-nine papers showing patients with mitochondrial cardiomyopathy and mtDNA and/or nDNA mutations.
The most of these papers report mtDNA mutations in association with the cardiological phenotype, in agreement with the higher frequency, in adults, of mutations in the mitochondrial genome. Moreover, many of these studies are not carried out with highthroughput genetic technologies that are able to highlight several disease-genes associated with MCM and MDs showing cardiomyopathy.
In this context, the next-generation sequencing technology has allowed, for several diseases, a huge expansion of the diagnostic screening, increasing the spectrum of involved genes and suggesting new genotype/phenotype correlations [184]. This correlation is particularly important in the diagnosis of MDs, including MCM and for the identification of pre-symptomatic family members to take in follow up.
The actual NGS strategies allow screening patients with suspected MD through a wide range of approaches, including targeted gene panel sequencing, mitochondrial gene panel, clinical exome panel and Whole Exome Sequencing (WES). The choice of the technique depends on several factors, such as funding, bioinformatics, and laboratory expertise.
WES or clinical exome panel may be considered for group of genetic disorders, which may be examined following the same pipeline. Then, the use of virtually assembled gene panel (i.e., Mitochondrial Cardiomyopathy gene panel, metabolic gene panel, etc.) allows to select only data in genes of interest and filtering for further genes or exome analysis, in a subsequent bioinformatics re-evaluation (i.e., if, in the selected panels, no mutations are found and the clinical suspicion is further deepened and confirmed). This is very important because the MD phenotype overlapped with other metabolic and neuromuscular disorders.
Although coverage for WES is constantly increasing and may be generally comparable to that obtained by targeted gene panel analysis, challenges in obtaining sufficient coverage are due to the need, for economic reasons, to collect a greater number of samples in the same analytical session.
Furthermore, the most challenging aspect of this approach is given by the large amount of unknown identified variants or novel disease genes, which need further investigation both by in silico analysis and through functional studies [185]. In this regards, the role of molecular biology research is fundamental to expand the list of pathogenic variants and increase our knowledge about MCM.
Conversely, smaller or targeted gene panels, such as the one proposed in this study, achieve a high diagnostic yield, at lower operating costs, in particular when they are designed to target the high yield genetic content, or most relevant and clinically associated genes. This is an ideal NGS approach for the diagnosis of heterogeneous disorders, like MDs, involving dual genome.
In the post-genomic era, the benefits of NGS in molecular diagnosis of rare diseases are well established. However, the use of NGS-based approach in diagnostics is increasing the yield of VUS. The significance and possible consequences of these variants are difficult to establish. The possibility to study a specific VUS, in the context of a large pedigree by analyzing the variant's segregation, as compared with the clinical phenotype may be useful; however, it is often unfeasible. Additionally, when clinical features suggest an underlying disease, the presence of a VUS may deserve reinterpretation regarding its pathogenicity during follow-up according to accumulating literature knowledge. We believe that the interpretation of variants should be verified approximately every 6 months. Indeed, it is not uncommon that the interpretation of variants changes after further studies are published over time. Moreover, we are fully aware that in silico predictions are not enough to establish the variants' pathogenicity and in vitro functional studies would be needed to have more conclusive information on each identified variant.
Mitochondrial cardiomyopathies are very rare disorders; in this setting, despite laboratory capacity to generate sequence data has greatly increased, with the advent of high-throughput next generation sequencing, the capacity to interpret genomic data and to identify disease genes can be difficult. The literature review highlighted about thirty high-confidence genes known to cause mitochondrial diseases in large cohorts of MCM patients. At the same time, we selected many other genes rarely reported associated with MCM (<1% of cases, each), which are reported as potential disease-genes based on known gene function, segregation with disease in families, computational analysis, etc., although more information is needed so that they can be classified as clinically actionable genes. Consequently, the probability of finding pathogenic variants is higher in high confidence genes than in low confidence ones, using the updated guidelines for the clinical interpretation of sequence variants developed by the American College of Medical Genetics and Genomics (ACMG).
Employment, in a cohort of patients showing clinical features of MCM, of the genes panel designed by carefully selecting genes previously associated with MCM, will yield rapidly actionable results (variants in high-confidence genes) and also identify poten-tially causal variants (in low-confidence genes) that will require further information to be clinically actionable.
Although a lot of work is still needed to easily and quickly diagnose a Mitochondrial Cardiomyopathy, a significant number of undiagnosed patients may be identified by NGS technology.
In this review, we identified 130 disease-genes associated with MCM, by searching molecular data from papers published in the last ten years, reporting new sequencing technologies. These genes could be included as target genes to build a novel gene panel for the molecular diagnosis of patients with clinical suspect of MCM.

Conclusions
Mitochondrial cardiomyopathy is a challenging diagnosis, requiring wide medical, neuroimaging, and laboratory examinations. The cardiologists have to keep the degree of suspicion high, looking also for signs from other tissues involved and the family history. Echocardiography and neuroimaging findings are variable, but the hypertrophic cardiomyopathy is the most common phenotype. Moreover, the increasing use of Next Generation Sequencing technologies allows collecting more genetic data, continually improving the diagnostic success. This review collects the major genes associated with mitochondrial cardiomyopathy in recent decades, underlining how the link between research and diagnosis is crucial to expand the list of pathogenic variants and concomitantly enhance our knowledge of mitochondrial cardiomyopathy.

Limitation
This review has some limitations concerning both the possibility that the filters utilized for the literature and database research, the keywords used, and their association may not have been enough to avoid losing papers. Furthermore, we did not differentiate pediatric and adult studies. Finally, the panel of genes selected from such a large literature will have to be validated to know the diagnostic yield in mitochondrial cardiomyopathies.

Conflicts of Interest:
The authors declare no conflict of interest.