Biological and translational attributes of mitochondrial DNA copy number: Laboratory perspective to clinical relevance

Abstract

The mitochondrial DNA copy number (mtDNAcn) plays a vital role in cellular energy metabolism and mitochondrial health. As mitochondria are responsible for adenosine triphosphate production through oxidative phosphorylation, maintaining an appropriate mtDNAcn level is vital for the overall cellular function. Alterations in mtDNAcn have been linked to various diseases, including neurodegenerative disorders, metabolic conditions, and cancers, making it an important biomarker for understanding the disease pathogenesis. The accurate estimation of mtDNAcn is essential for clinical applications. Quantitative polymerase chain reaction and next-generation sequencing are commonly employed techniques with distinct advantages and limitations. Clinically, mtDNAcn serves as a valuable indicator for early diagnosis, disease progression, and treatment response. For instance, in oncology, elevated mtDNAcn levels in blood samples are associated with tumor aggressiveness and can aid in monitoring treatment efficacy. In neurodegenerative diseases such as Alzheimer’s and Parkinson’s, altered mtDNAcn patterns provide insights into disease mechanisms and progression. Understanding and estimating mtDNAcn are critical for advancing diagnostic and therapeutic strategies in various medical fields. As research continues to uncover the implications of mtDNAcn alterations, its potential as a clinical biomarker is likely to expand, thereby enhancing our ability to diagnose and manage complex diseases.

Key Words: Mitochondrial DNA copy number; Mitochondrial DNA; Quantitative polymerase chain reaction; Cancer; Neurodegenerative disease; Aging

Core Tip: This review highlights the critical biological and translational significance of mitochondrial DNA (mtDNA) copy number (mtDNAcn) variations and emphasizes their implications in both laboratory diagnostics and clinical settings. By exploring the mechanisms underlying mtDNA replication, the association of mtDNAcn with various diseases, and its potential as a biomarker for mitochondrial dysfunction, we underscore the need for standardized methodologies for measuring mtDNAcn. This perspective aims to bridge the gap between basic research and clinical applications by facilitating the integration of mtDNAcn assessments into routine diagnostic practices and therapeutic strategies for mitochondria-related diseases.

INTRODUCTION

Mitochondrial DNA and mitochondrial DNA copy number

Mitochondria are essential organelles found in almost all cells, and are pivotal energy generators in mammalian cells. Endowed with their own genomes, referred to as mitochondrial DNA (mtDNA), mammalian cells maintain hundreds to thousands of mtDNA molecules per cell[1]. Mitochondria are organelles that reside in the cytoplasm of eukaryotic cells and are involved in energy production via oxidative phosphorylation (OXPHOS), which utilizes oxygen and nutrients to produce adenosine triphosphate (ATP) accompanied by harmful byproducts of reactive oxygen species[2]. Mitochondria also play central roles in various cellular processes such as apoptosis, calcium homeostasis, and reactive oxygen species metabolism[1,2]. OXPHOS is orchestrated by four large complexes embedded in the mitochondrial inner membrane, and defects in OXPHOS are associated with mitochondrial diseases. mtDNA copy number (mtDNAcn) indicates the relative abundance of mtDNA relative to the total amount of nuclear DNA (nDNA) in the genome, is an essential indicator of mitochondrial health and integrity, and has been suggested as a potential disease biomarker as it affects mitochondrial biogenesis and cellular oxidative stress[1,3,4].

Mitochondrial biogenesis is orchestrated by the concerted expression of nuclear and mitochondrial genes, which is regulated by various transcription factors, cofactors, and coactivators. Key transcription factors involved in the regulation of mitochondrial biogenesis include peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α), nuclear respiratory factors (NRFs), and mitochondrial transcription factor A (TFAM)[3,5]. PGC-1α is a coactivator that controls the expression of mitochondrial and nuclear genes, thereby regulating mitochondrial biogenesis, energy substrate uptake, and OXPHOS activity in different tissues. PGC-1α activity is regulated by various signaling pathways that are activated in response to changes in cellular energy demand (e.g., exercise, cold exposure, and starvation). After activation, PGC-1α increases mitochondrial biogenesis by promoting the expression of NRFs and TFAM, mitochondrial dynamics and motility, and the mitochondrial quality control system[5,6]. Experimental overexpression of PGC-1α has been associated with increased mitochondrial biogenesis and proliferation, whereas PGC-1α downregulation is associated with mitochondrial dysfunction, decreased mitochondrial biogenesis, and several pathologies[5]. However, in different tissues, mtDNAcn may also be affected by factors such as cell size, oxygen consumption, energy expenditure, and age[2,5]. Understanding the biological attributes of mtDNAcn will provide insights into the pathogenesis of various diseases.

mtDNA is a circular genome containing 16569 base pairs (15-17 kb in length), is characterized by covalent closed-loop structures, and is commonly present in multiple copies per mitochondrion[7,8]. The number of mtDNA per cell varies according to tissue specificity, ranging from several hundred in non-energy-demanding tissues, such as fibroblasts, to thousands in high-energy-demanding tissues, such as the heart, skeletal muscle, and brain. mtDNA is thought to encode 37 genes necessary for the correct functioning of mitochondria, including 13 proteins of the OXPHOS system, two ribosomal RNAs, and 22 transfer RNAs[7,9]. Among the 13 polypeptides encoded by mtDNA, 12 are subunits of OXPHOS complexes I, III, IV, and V, and awareness of their importance stems from genetic mutations that lead to mitochondrial disorders, which can affect post-mitotic cells rich in mitochondria, such as neurons, cardiac muscle, and skeletal muscle[9,10]. Reversible lesions within mtDNA have been implicated in a variety of disorders, including oxidative stress, energetics, and signaling dysregulations[10-12]. Accordingly, mtDNA alterations can differentially influence cellular homeostasis and behavior, facilitating either adaptation or sensitivity to external perturbations. Indeed, numerous studies have indicated that mitochondrial biogenesis is associated with an increased mtDNAcn, and a decrease in mtDNAcn correlates with mitochondrial dysfunction[10-14]. Notably, several nuclear-encoded genes are involved in the regulation of mtDNAcn, suggesting that the mtDNAcn is regulated by a highly coordinated process. Since ATP synthesis is linked to mitochondrial respiration, which involves the electron and proton gradients of the respiratory chain, these two processes are also involved in mtDNA replication, which regulates the number of mtDNA copies (Table 1)[10,13].

Table 1 Key genes regulating mtDNA copy number.

Gene	Role in mtDNA regulation	Associated disorders	Ref.
PGC-1α	Master regulator of mitochondrial biogenesis	Neurodegeneration, cancer, metabolic syndrome	[3,5,6]
TFAM	Maintains mtDNA integrity and replication	Mitochondrial diseases, aging	[5,9]
POLγ	Essential for mtDNA replication	MELAS, Kearns-Sayre syndrome, cancer	[88-90]
NRFs (1 and 2)	Coordinate expression of mitochondrial genes	Neurodegeneration, metabolic disorders	[3,5]

mtDNA: Mitochondrial DNA; PGC-1α: Peroxisome proliferator-activated receptor gamma coactivator 1-alpha; TFAM: Mitochondrial transcription factor A; POLγ: Polymerase gamma; NRFs: Nuclear respiratory factors; MELAS: Mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes.

Alterations in mtDNAcn have been observed in aging and various disorders, including neurodegenerative, inflammatory, metabolic, and oncogenic diseases[10-14]. Aberrant changes in mtDNAcn are also believed to serve as potential biomarkers for early diagnosis of these diseases. It has been shown that mtDNAcn increases in a large number of human cancers, regardless of histological type, although the extent and distribution of mtDNAcn varies across tumors[15]. The first mtDNA sequence decoded in humans in 1981 led to the discovery of more than 150-point mutations in the genes encoded by mtDNA in patients with Leber’s hereditary optic neuropathy (LHON) and several large-scale deletions and duplications, mainly affecting mitochondrial coding genes and generating external frameshift mutations associated with mitochondrial myopathy and multiple mitochondrial dysfunction syndromes[15,16].

Unlike conventional nDNA quantification, mtDNA detection is based on polymerase chain reaction (PCR) amplification of a unique region of mtDNA, avoiding interference by nDNA[4-6,15]. Assessment of mtDNA and its structural alterations is emerging as a viable detection tool for various biological molecules, including cytosolic exosomes, body fluids, tissues, saliva, urine, and hair. The quantification of mtDNA content is underscored by the promise of a deeper understanding of cellular senescence, age-related cellular dysfunction, epigenetic regulation, tumorigenesis, and pharmacology. The emerging literature corroborates the correlation between mtDNAcn and its structural alterations (deletion and rearrangement) with the clinical progression of cancer, metabolic, and degenerative diseases. Studies have also suggested that alterations in mtDNAcn may be the earliest events in the progression of chronic diseases[4,12,15]. The modulation of mtDNAcn raises the possibility of a plausible contribution of mtDNAcn to cellular dysfunction, fate, and pathogenesis. Since mtDNAcn can be analyzed through non-invasive detection of peripheral blood, exploring the alterations of mtDNAcn in bodily cells and body fluids holds great promise for better understanding and improvement of the aforementioned diseases. Therefore, a comprehensive understanding of mtDNAcn regulation in cells is important for biomedical studies and has been substantiated. Recent studies have highlighted the clinical relevance of mtDNAcn in early diagnosis, prognosis of therapeutic intervention, and detection of disease recurrence.

Despite extensive research on mitochondrial function, the precise mechanisms regulating mtDNAcn and their quantification remain incompletely understood. Specifically, while various nuclear-encoded genes and transcription factors, such as PGC-1α, TFAM, and NRFs, have been identified as key regulators of mitochondrial biogenesis, significant gaps persist in the understanding of the intricate coordination of these factors in different tissues and the variability in mtDNAcn across tissues, its dynamic response to cellular energy demands, and its role in the pathogenesis of diverse diseases highlights a complex regulatory network that warrants further investigation. For example, the interplay between oxidative stress, mitochondrial dynamics, and mtDNA replication processes remains poorly characterized, particularly in the context of aging and chronic diseases[10,12]. These knowledge gaps underline the necessity of studying mtDNAcn as a potential biomarker of mitochondrial dysfunction. Addressing these unanswered questions could lead to improved diagnostic and therapeutic strategies, particularly for diseases for which mitochondrial health is critical.

METHODS FOR ASSESSING mtDNAcn

Methods for measuring mtDNAcn have proliferated alongside the growing application of mtDNA analysis. Because of the biased segregation of mtDNA mutations during cell division, some mutations in the mitochondrial genome become homoplasmic (existing in all copies) and persist at high copy numbers, while others remain heteroplasmic (existing only in a subset of copies) and diminish with time, indicative of cellular dysfunction[17,18]. All DNA in a diploid cell is referred to as its nuclear-equivalent DNA content, encompassing both nuclear and mitochondrial genomes[8,11,17]. Furthermore, all mtDNA in a diploid cell is referred to as its mtDNA content, which is directly measurable via quantitative PCR (qPCR) tests and other techniques[15,17,18].

Most methods for measuring mtDNAcn utilize qPCR in a real-time thermal cycler[19,20]. The assay can be performed using a two-target reference gene approach, normalized to one of the nuclear target genes, such as β-actin, or a single-target approach, where data are compared to a standard curve generated with known quantities of mtDNA input[19,20]. There is an implicit assumption that mtDNA and nDNA are under equal amplification conditions because of the use of the same PCR primer sets and reagents, with the difference being the number of target sequences[19,20]. Targeting and amplifying nDNA segments present in a chromosome with unequal pooling of homologs would confound any calculated ratio under equal amplification conditions. Although qPCR is the most common method employed, it has several caveats, particularly in the assessment of tissues with low-quality DNA, experiments with lower DNA concentrations, and heterogeneous samples. DNA preparation is also a rich source of systematic bias, and increased effort is needed to ensure that DNA samples are free of bias prior to mtDNAcn quantification[14,19,20].

Next-generation sequencing (NGS) platforms allow pyrosequencing of millions of individually sequenced DNA fragments in parallel[21]. NGS is a promising alternative mtDNAcn assessment tool that requires post-sequencing bioinformatic approaches that are currently either non-robust or non-widely adopted[8,14,21]. The design and optimization of the initial NGS experiments largely dictate read lengths and yields, amplicon sizes, primer dilution, PCR iterations, amplification conditions, and DNA concentrations in subsequent reactions using the same amplification primers for multiplexing, ProSEQ, resetting, and incorporating barcodes/indexes[20,21].

From a historical perspective, the southern blot (SB) hybridization has long been the gold standard for assessing the level and integrity of mtDNA in patient samples and model organisms[22]. Although SB techniques are highly reliable for determining mtDNA content, they are time-consuming, only semi-quantitative, and require a relatively large amount of DNA, which presents a significant drawback when studying human tissues[22]. Fluorescent in situ hybridization has also been used to spatially visualize mtDNA content with single-cell resolution; however, this method is only partially informative[23]. Its multi-step protocol is laborious and offers only a rough estimate of the changes in mtDNA levels.

Assessing mtDNAcn using qPCR

qPCR is the most widely used method for analyzing mtDNAcn[14,24]. It can be optimized to assess mtDNAcn in a high-throughput format, sensitive to low mitochondrial sample input, and conducive to sample pooling[24,25]. However, qPCR relies solely on specific DNA sequences. Thus, experimental design requires careful consideration and testing of qPCR primers to ensure robust, reproducible, and specific measurements, with low inter- and intra-assay variance being the most critical quality control step in this approach[24-26]. Traditional qPCR quantification methods typically analyze mtDNA and nDNA using a single primer set in two steps: The standard curve and the mtDNA/nDNA ratio[24,25]. However, for heterologous sample preparations containing nDNA genes homologous to mtDNA, this analysis will give rise to very large quantification errors in determining mtDNAcn[25-27]. This is attributed to the inability to monitor assay conditions and amplify efficiency with respect to specific sample preparation. As a result, such methods often lead to inconsistent data on mtDNAcns for the same sample preparation analyzed by different laboratories[25]. Quantification algorithms need more caution for qPCR assays to measure mtDNA-nDNA ratios and mtDNAcns, especially in samples containing nDNA genes with significant sequence similarity to mtDNA[26].

For sample preparation, all experiments must be performed while working on ice or at cold temperatures within a pre-cooled mini-cooled PCR cycler to prevent the unwarranted loss of fast-evolving wild-type genes[24-27]. The initial pre-preparation for the qPCR assay must avoid freezing and thawing of samples and resuspend all ultralow sample concentrations gently to produce a homogeneous distribution[24]. A low pipetting force should be used to prevent mtDNA shearing. As sample dilution and pooling may introduce batch effects and quenching of the qPCR signal, all samples to be pooled should be evenly split to avoid overconcentration of the same samples in specific wells[25,26]. An individual assay for each sample is advised for samples anticipated to fall below measurable concentrations.

mtDNAcn analysis using qPCR can be performed in either SYBR Green or TaqMan interrogation format, both of which target specific amplicon sequences[26-28]. The SYBR Green assay relies on the intercalation of a fluorescent dye into dsDNA, as it is amplified[27]. TaqMan assay is dependent on an oligonucleotide probe labeled with a fluorescent dye and a quencher dye. As amplification occurs, the 5’ nuclease activity of Taq polymerase cleaves the probe, releasing a fluorescent signal that is detected and proportional to the amount of PCR product generated[24,28]. When the experimental design requires the use of multiple primer pairs in the same plate layout and run, controls must be included on the same plate for each primer pair, including Q5 Hot Start High-Fidelity DNA Polymerase as a positive control, to verify the performance of the primer pairs used[26,27]. Compounds and probes should be included to test their effects on the protocol. This design allows for a validity check of the batch size. Preferably, qPCR analysis should be repeated for different batch sizes in a high-throughput qPCR setup[24-28].

Assessing mtDNAcn using NGS

Complementary to qPCR methods, NGS has been developed to quantify mtDNAcn in a single run in a genome-wide manner without prior knowledge of mitochondrial sequences[28]. Using read-depth normalization strategies, NGS data can be used to measure the total and/or specific mtDNAcn changes[28]. Furthermore, their integration with existing whole-genome sequencing and whole-exome sequencing (WES) is enabled by the almost universal implementation of NGS in most laboratories that perform large-scale genome analysis[28-30].

The process begins with the extraction of total DNA, including both nDNA and mtDNA, from cells or tissues. The next step involves library preparation, wherein the extracted DNA is fragmented into smaller pieces to facilitate efficient sequencing[28-30]. Sequencing adapters, which are short pieces of synthetic DNA, are attached to the fragmented DNA to allow binding to the sequencing platform. This step is critical for ensuring that DNA fragments can be correctly amplified and sequenced. During sequencing, both nuclear and mitochondrial genomes were sequenced simultaneously, providing a comprehensive overview of the entire cellular DNA content. The sequencing platform generates millions of reads, which are short DNA sequences that correspond to fragments of both nDNA and mtDNA. Once the sequencing data are generated, bioinformatics tools are used to align the reads to their respective reference genomes, allowing for precise identification of whether the reads originate from the nuclear genome or the mitochondrial genome[28-32].

The number of reads mapped to the mitochondrial genome was compared to the number of reads mapped to the nuclear genome[29,30]. Because the nuclear genome is present in two copies per diploid cell, this comparison allows for the normalization and accurate calculation of the mtDNAcn per cell[29,31]. The formula used for this calculation typically normalizes the number of mtDNA reads to the size of the mitochondrial genome and the number of nDNA reads to the ploidy of the nuclear genome[32]. This allows for the estimation of the number of copies of the mitochondrial genome present in each cell. Given that the nuclear genome is usually constant in copy number across cells, any variation in the ratio of mtDNA to nDNA reads reflects changes in mitochondrial content, providing insights into mitochondrial biogenesis or degradation under different physiological or pathological conditions[30-33].

The diversity of experimental designs remains a potential source of bias, even in strictly controlled laboratory environments. Molecular experiments should be reported with the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines in mind, which would cover both laboratory knowledge and actively adopted best practices. Ideally, all design-specific factors should be reported so that all procedures leading to the normalization and quantification of results are transparent and open to independent validation.

For mtDNAcn assessment, technical considerations include qPCR-specific issues related to reaction volume, number of replicates or parallel reactions, reaction mixtures (acid/base ratio affecting amplification efficiency) leading to stricter biochemical requirements for samples, and reliance on qPCR properties, which can differ among thermocyclers. Systematic biases stemming from qPCR chemistry and thermocycling conditions are classically solved through plate-like replication of controls (standard/reference/sample) at a ratio of at least 1/8, flat-lined volumetric neutrality (classically 20 μL reaction volume), baseline filtering, and standard curve correction[34]. Controversies remain regarding the use of synthetic amplification additives (e.g., surfactants) that act as smoothing agents to flatten amplification results[34]. Environmental sources that directly impact laboratory design are unavoidable (e.g., geographical location, surrounding industry, and building architecture) (Table 2).

Table 2 Methods for assessing mitochondrial DNA copy number.

Method	Steps involved	Applications	Advantages	Limitations	Ref.
qPCR	DNA extraction → primer design → amplification → analysis	Widely used in clinical and research diagnostics	High sensitivity, high throughput, cost-effective, rapid	Susceptible to bias in low-quality DNA, issues with heterogeneous samples, requires careful primer design	[15,24-26]
NGS	DNA extraction → library preparation → sequencing → bioinformatics analysis	Genome-wide studies, detects mtDNA mutations alongside copy number analysis	Genome-wide analysis, accurate quantification, detects mtDNA heteroplasmy	Requires advanced bioinformatics, high cost, computational complexity	[21,28-32]
Southern blot hybridization	DNA extraction → gel electrophoresis → hybridization → quantification	Historically the gold standard, reliable for mtDNA integrity assessment	High reliability, detects large-scale deletions	Time-intensive, requires large DNA quantities, semi-quantitative	[22]
FISH	Sample preparation → probe hybridization → microscopy	Single-cell resolution studies, spatial visualization of mtDNA	Single-cell resolution, visualizes mtDNA distribution	Labor-intensive, provides only rough mtDNA estimates, technically demanding	[23]

qPCR: Quantitative polymerase chain reaction; NGS: Next-generation sequencing; FISH: Fluorescent in situ hybridization; mtDNA: Mitochondrial DNA.

mtDNAcn IN HEALTH AND DISEASE

Changes in mtDNAcn can lead to, or be a consequence of, mitochondrial dysfunction. mtDNA deletions and depletion (mtDNA deletions refer to the loss of a specific segment of the mtDNA molecule that results in a shortened mtDNA molecule that lacks certain genes or regulatory regions, and mtDNA depletion refers to a significant reduction in the overall copy number of mtDNA within cells or tissues, while the remaining mtDNA molecules remain intact and structurally normal), and mutations can occur as a result of damage from free radicals, ionizing radiation, chemotherapeutic agents, or toxins[4,8,9,15]. Accumulation of these mutations can lead to mitochondrial dysfunction and energy deficiency[2,4,15]. This can be compensated, in part, by an increase in mtDNAcn; however, it is hypothesized that there will be a deleterious cumulative effect on function and health.

Recent studies have linked alterations in mtDNAcn to a number of human disorders, such as neurodegenerative diseases [e.g., Alzheimer’s disease (AD), Parkinson’s disease (PD), Huntington’s disease], diabetes, Fanconi’s anemia, diseases of the skeletal muscle (e.g., mitochondrial myopathy, congenital defects like Pearson’s syndrome), cardiomyopathies, non-syndromic deafness, certain cancer types (e.g., breast, colorectal, endometrial, gastric, annular pancreas, gliomas, prostate, cervical, bladder carcinoma), and complex diseases (like prion diseases)[7,15]. Trimethylation of histone H3 lysine 9 has been implicated in the silencing of promoters of nuclear-encoded respiratory complex subunit genes, resulting in a gradual increase in mtDNAcn in response to mitochondrial dysfunction in animal models[35]. In contrast, hypomethylation of H3 lysine 4 chromatin marks at the Pol I transcription start sites was shown to be involved in the switch of the mtDNA replication mechanism from legacy to prominent mode in metazoan species[35]. The mtDNAcn varies widely among cell types[15], and a common medical interest is to determine the mtDNAcn in cells for diagnostic or prognostic purposes. With aging, somatic mutations accumulate in mtDNA, and a gradual decrease in the relative number of full-length mtDNA is observed in various organs of older mice, suggesting an age-dependent decline in mitochondrial function, substantiated by an altered mtDNAcn in the tissues or blood of aging animals or humans[4,15]. Cell-free mtDNA has recently attracted attention, especially because of its potential use as a non-invasive biomarker for both mitochondrial dysfunction and cancer.

Etiology of mitochondrial mutations

Mutations in mtDNA can arise either from spontaneous errors during DNA replication or from incorrect repair of damaged DNA bases. Sequence analyses and experimental studies involving a large number of mtDNA genomes indicate that most mtDNA mutations result from spontaneous replication errors introduced by DNA polymerase gamma (POLγ) during mtDNA synthesis[36,37]. These pathogenic mtDNA mutations are commonly found in mitochondrial tRNA, rRNA, and protein-coding genes, leading to compromised mitochondrial gene expression and varying degrees of OXPHOS deficiency. Owing to the multicopy nature of the mitochondrial genome, mtDNA mutations may be homoplasmic or heteroplasmic. Many severe mtDNA mutations cannot be tolerated in the homoplasmic state, which makes their presence in the heteroplasmic state more common in patients. Typically, heteroplasmic pathogenic mtDNA mutations exhibit a functionally recessive nature, meaning that they only cause respiratory chain deficiencies when present above a certain threshold level[38]. Consequently, cells with mutated mtDNA levels exceeding this critical threshold develop respiratory chain dysfunction, whereas adjacent cells with lower mutation loads can maintain normal respiratory chain function. The threshold level is highly tissue-dependent and varies with the type of mtDNA mutation; for example, the threshold for single large deletions is approximately 50%-60%[39], while some tRNA gene point mutations may exceed 90% mutated mtDNA[39,40]. Throughout an individual’s life, levels of heteroplasmy can fluctuate owing to mitotic segregation, a result of relaxed replication, and random mitochondrial partitioning between daughter cells. Additionally, deleterious mtDNA alleles may be actively eliminated in proliferating tissues through purifying selection, a phenomenon observed in the blood for certain mtDNA mutations[40-42]. It is well established that point mutations in tRNA genes or single large deletions in mtDNA lead to functional impairments or deficiencies in one or more tRNAs. However, these pathogenic mtDNA mutations typically lack dominant effects, allowing tolerance to high levels of mutated mtDNA. The recessive nature of most human pathogenic mtDNA mutations results in a deficiency of wild-type gene products, impairing OXPHOS, and contributing to disease phenotypes. Beyond their role as direct causes of primary mitochondrial diseases, which are characterized by severe mitochondrial impairment in multiple tissues[42,43], mtDNA mutations have also been implicated in the pathophysiology of common age-associated diseases[15,44-46] and in the natural aging process[15].

Aging

The mtDNAcn during aging has been assessed in most tissues from humans to rodents. However, currently, there is no consensus regarding the relevance of mtDNAcn analysis. The impact of mtDNAcn changes on different tissues may not be equivalent. In a seminal paper, mtDNAcn was shown to decrease with age in the brain, heart, liver, skeletal muscle, and blood of humans[47,48] but increased in the skin, adipose tissue, peripheral blood mononuclear cells, and urine[15,49]. mtDNAcn levels were increased in the lymphocytes of TWAY mice, a model for human aging and age-related pathologies. Transient decreases in mtDNAcn were observed during early postnatal development in the brain, heart, liver, and skeletal muscle but not in the skin[50].

The amount of mtDNA in the tissues of people of various ages has been measured in numerous investigations[8,15]. During aging, mtDNA levels vary among different tissues. In lymphocytes and blood, there is a notable decrease in the mtDNAcn, as revealed by whole-genome sequencing and qPCR[51,52]. This decline in mtDNA has also been observed in skeletal muscle, where one study using NGS and digital droplet PCR reported a reduction[49]. However, other investigations in skeletal muscle using qPCR and southern blotting have shown that mtDNA levels remain unchanged[47,48]. Similarly, the heart showed no significant change in mtDNA levels, as confirmed by qPCR and SB analyses[47,48].

In contrast, certain tissues display increased mtDNA levels with age. For example, the liver exhibits elevated mtDNA content, based on NGS and digital droplet PCR data[49]. Likewise, mtDNA levels are higher in the substantia nigra (SN), as determined by qPCR[53]. Interestingly, some brain regions, such as the caudate nucleus, frontal lobe cortex, and cerebellar cortex, show no change in mtDNA levels during aging, as evidenced by SB analysis[47]. These findings suggest a complex and tissue-specific pattern of mtDNA regulation during the aging process, with some tissues experiencing a decline, others remaining stable, and some exhibiting increased mitochondrial content.

A significant reduction in mtDNAcn was observed in blood samples, with the decline becoming more noticeable in individuals in their 50 seconds. This decrease in mtDNA levels intensifies with age, particularly in older populations where a more dramatic reduction is evident[54,55]. The mtDNAcn has been estimated to decrease by a few percent per decade[56]. In individuals over 58 years of age, a low mtDNAcn in peripheral blood has been linked to higher mortality rates and poor health outcomes, including decline in both cognitive and physical performance[55]. However, studies conducted on long-lived families, including nonagenarians and centenarians, have produced puzzling and often contradictory results, complicating our understanding of mtDNA dynamics during extreme longevity[55,57].

The aging process is accompanied by a decline in mitochondrial function, along with noticeable changes in mitochondrial morphology, content (both in terms of number and protein levels), and OXPHOS[15,57,58]. Pathogenic mutations in mtDNA, including both large deletions and point mutations, have been identified in various tissues of aged individuals, affecting both post-mitotic and proliferating cells[59]. Experimental evidence has suggested that somatic mtDNA mutations play a role in driving certain age-related phenotypes. For example, in mice, a deficiency in the proofreading function of mitochondrial POLγ leads to progressive accumulation of mtDNA mutations, resulting in premature aging syndrome. This syndrome is characterized by reduced lifespan, decreased fertility, anemia, hair greying, hair loss, hearing loss, and stem cell dysfunction[15,47,60].

Diminished bioenergetic function and increased oxidative damage are hallmarks of aging in mammals. Mitochondria are susceptible to oxidative damage owing to their proximity to reactive oxygen species, resulting in mtDNA lesions. Autophagic clearance of damaged mitochondria decreases with age in various species. Overall, mtDNAcn dynamics should be explored more thoroughly across different tissues in animal models of aging.

mtDNAcn as a biomarker

mtDNAcn can be employed as a quantifiable biomarker, thereby providing a measurement that offers information related to cellular changes in disease diagnosis, severity prediction, and treatment efficacy evaluation in a multitude of human diseases, especially in cancers, neurodegenerative diseases, infections, maternal-fetal diseases, and inherited diseases. Advances in technology have enabled the assessment of mtDNAcn as part of genomic profiling in a high-throughput, rapid, and cost-effective manner because unlike nDNA, mtDNA is present in numerous copies, up to thousands per cell, and its copy number is subject to regulation[9,10,29]. Individuals vary in their mtDNAcn, and while they are relatively stable across tissues and over time, they may be modulated by diet, lifestyle, and exposure to various environmental factors[15,61]. mtDNAcn also appears to be regulated during mitochondrial biogenesis and in response to changes in cellular energy demand[11,13]. Preclinical studies involving cell lines and animal models, along with epidemiologic studies concerning human subjects with various diseases, have collectively revealed that the mtDNAcn is altered in a disease-dependent manner. As a result, there is a compelling interest in the clinical application of mtDNAcn as a diagnostic biomarker for potential use in clinical laboratories. To date, the diagnostic applications of mtDNAcn have been evaluated and assessed across multiple disciplines, including oncology, metabolic disorders, and infectious diseases, using different detection techniques in various tissues, including peripheral blood, buccal mucosa, saliva, lymphocytes, urine, breast tissues, tumor tissues, neurons in post-mortem brain samples, hair follicle tissues, and cultivated cells.

The focus of oncology is on cell-free mtDNAcn found in the plasma or serum of patients with various cancers. The rationale for investigating cell-free mtDNA is that it is released from dead or damaged cells, thereby creating an abundance of mtDNA in the circulatory system. The same is substantiated by the studies corroborating a link between reduced cell-free mtDNA in plasma or serum and the onset or progression of cancers, particularly with respect to lung, breast, ovarian, endometrial, colorectal, prostate, and hepatitis virus-induced liver cancer (Table 3)[15,62,63].

Table 3 Clinical relevance of mitochondrial DNA copy number in diseases.

Disease category	Role of mtDNAcn	Key observations	Diagnostic/prognostic value
Neurodegenerative disorders[64-70]	Biomarker for disease progression and severity	Reduced mtDNAcn in AD brains; increased mtDNAcn in peripheral blood of AD patients	Correlates with tau pathology in CSF; potential for non-invasive diagnosis using blood mtDNA levels
Cancer[15,62,100-102]	Indicator of tumor aggressiveness and treatment response	Elevated mtDNAcn associated with tumor proliferation; decreased mtDNAcn linked to poor prognosis	Distinguishes between cancerous and non-cancerous tissues; early-stage cancers show higher mtDNAcns, while advanced stages may show depletion
Metabolic disorders[4,15,80-83]	Reflects mitochondrial dysfunction	mtDNAcn dysregulated in diabetes and other metabolic syndromes, indicating stress or compensation mechanisms	Biomarker for mitochondrial stress in diabetes; changes in mtDNAcn can indicate early disease onset or progression
Aging[47-49,55,56]	Associated with age-related cellular dysfunction	Decline in mtDNAcn in various tissues (e.g., blood, muscle) with age; some tissues exhibit increased mtDNA	Low mtDNAcn linked with poor health outcomes in aging populations, including cognitive and physical decline
Inherited mitochondrial disorders[82-93]	Indicates heteroplasmy levels and disease severity	Variations in mtDNAcn linked to phenotypes like MELAS, Pearson’s syndrome, and Leber’s hereditary optic neuropathy	High mtDNAcn linked to milder phenotypes; can guide prognosis and therapy for conditions like Kearns-Sayre syndrome and mitochondrial encephalopathy

mtDNAcn: Mitochondrial DNA copy number; AD: Alzheimer’s disease; MELAS: Mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes; CSF: Cerebrospinal fluid.

mtDNAcn in neurodegenerative disorders

Mitochondrial dysfunction has emerged as a potential contributor to neurodegenerative disorders such as AD and PD[64,65]. Initial studies on mtDNAcn in AD focused solely on brain tissue. Several studies have used a more systematic approach to evaluate mtDNAcn levels in the peripheral tissues and biofluids. Follow-up studies have assessed mtDNAcn levels in a diverse array of tissues. Most studies have found that mtDNAcn is reduced in the brains of patients diagnosed with AD[66,67]. In support of these findings, decreased mtDNAcn is associated with increased phosphorylated tau levels in the cerebral spinal fluid of several AD cohorts[68,69]. Interestingly, contradictory findings have been reported for other tissues[70]. In peripheral blood and arteries, mtDNAcn has been primarily observed to be increased in patients with AD. As an important consideration, most of these studies used 12S rRNA to quantify mtDNAcn. Previous studies have shown that nDNA-located mitochondrial ribosomal RNA genes can be lost with age, potentially confounding the nature of the differences between groups.

In Parkinson’s, gene expression in various brain regions and biological fluids shows distinct patterns. In the geniculate nucleus region, expression is upregulated, whereas in the SN and blood, it is downregulated, as detected by qPCR techniques[69,71,72]. The cerebellum and cerebellar cortex remained unchanged based on WES analysis[73]. The frontal cortex also showed no change, but the qPCR results from the SN show downregulation[74]. In AD, the frontal cortex and hippocampus are downregulated according to qPCR results[75-77], whereas the cerebellum remains unchanged[76]. However, in some hippocampal and cerebellar cortex regions, downregulation was observed using both WES and qPCR[75]. Blood samples showed no changes in gene expression, whereas cerebral spinal fluid was downregulated[75,76]. These findings highlight the varied gene expression patterns across different regions and biological samples in both patients with PD and AD.

As an important modulator of mtDNAcn, PGC-1α has been implicated in the pathology of several neurodegenerative diseases. In a Drosophila model of synuclein-induced degeneration, PGC-1α exhibited protective effects by promoting mtDNA replication and mtDNAcn[78]. In human studies, PGC-1α was upregulated in the midbrain of sporadic PD patients[79]. However, it appears that some mutations in PGC-1α confer an increased vulnerability to nigral dopaminergic neuron death. In sporadic cases of AD, cerebrospinal fluid levels of the PGC-1α-targeted mitochondrial biogenesis driver sirtuin 1 negatively correlate with phosphorylated tau levels[78]. These studies indicate that a mechanism might promote an increase in mtDNAcn, whereas late disease mechanisms might promote its reduction. Importantly, these effectors interact with one another in several ways. Analysis of neurogenesis-enriched miRNAs revealed that the differential expression between AD subjects and controls could regulate many events.

mtDNAcn in metabolic disorders

In response to stresses or damage that compromise respiratory chain function, the cell can greatly amplify its mtDNAcn, which is intended to restore a normal ratio of mitochondrial and nuclear genomes, and is dependent on the transcriptional coactivator PGC-1α and its action on mitochondrial RNA POLγ. It is an attractive candidate for a biomarker because of the potential for it to point to early disease onset, the relative ease of obtaining tissue samples, and the feasibility of measuring mtDNAcn changes in archival collections, making it an attractive candidate.

Applications of mtDNAcn in the field of metabolism are largely directed toward mitochondrial disorders. Evidence suggests that disrupted mtDNA topology is disease dependent[80,81]. Studies that determined mtDNA deletion size and copy number in various tissues of subjects with different diseases indicated that mtDNA topology has the potential to serve as a disease signature at the genomic level[80]. Evaluation of mtDNA deletion size, which is readily detected by qPCR, has been employed as a clinical diagnostic tool for multiple mitochondrial disorders, including A3243G-induced mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS)[15]. The feasibility of mtDNA deletion size evaluation in maternal plasma has been investigated as a noninvasive prenatal test for the prenatal diagnosis of A3243G-induced MELAS. mtDNA depletion syndromes, a group of autosomal recessive disorders characterized by a significant tissue-specific reduction in mtDNA levels, arise from mutations in genes responsible for various aspects of mtDNA maintenance, including mitochondrial nucleotide metabolism, mtDNA replication, mitochondrial dynamics, and quality control mechanisms[15]. A few other examples of mtDNA mutations and their effects include Pearson’s syndrome, in which the mtDNA mutation involves deletions. Blood samples were used for analysis. mtDNA levels were elevated (“up”)[82], as quantified by qPCR. Similarly, in Kearns-Sayre syndrome (KSS), another deletion was identified in mtDNA, with blood and muscle samples used for analysis. The mtDNA levels were also elevated[82] using qPCR for quantification by qPCR. In MELAS, the mtDNA mutation, m.3243A>G, was detected in leukocytes. Studies have demonstrated that mtDNA levels can be elevated, unchanged, or decreased depending on the patient’s age, as reported by[83], using qPCR for quantification. Similarly, in myoclonic epilepsy with ragged red fibers (MERRF), the m.8344A>G mutation in leukocytes was studied by Liu et al[84], and mtDNA levels were found to vary (up, unchanged, or down) depending on age.

For LHON, in the case of the m.11778G>A mutation, peripheral blood cells showed elevated mtDNA levels in a study using qPCR[85]. Other studies have reported the m.11778G>A mutation in blood samples, where mtDNA levels were found to be either elevated or unchanged, with qPCR used for quantification[84,86]. Additionally, in LHON with the m.3460G>A mutation, blood samples were analyzed using qPCR, and mtDNA levels were either elevated or unchanged, as shown in a few studies[84,86]. Interestingly, it was found that copy number tended to increase in younger patients, whereas it remained unchanged or even decreased in older patients. Other point mutations linked to mitochondrial disorders that can be detected at the nucleic acid level have also been evaluated for potential application in maternal plasma tests, including A1555G-induced non-syndromic hearing loss and mtDNA4977 deletion-induced Pearson syndrome[87].

As a biomarker, the mtDNAcn reflects the maintenance and modification of the mitochondrial genome. In surface cells, larger copy numbers are maintained by the replicative machinery involving mitochondrial POLγ. POLγ subunit A (POLGA), which encodes the catalytic subunit of mitochondrial POLγ, has been identified as one of several nuclear genes associated with mtDNA depletion syndrome. Over 300 mutations in the POLGA gene have been identified, contributing to a wide range of mtDNA defects that lead to varying disease severities, symptoms, and ages of onset. Interestingly, a mutation that severely impairs the proofreading activity of POLGA in mice resulted in numerous mtDNA mutations but had no effect on mtDNAcn[88].

Patients with POLGA-related disorders can present with either a significant reduction in mtDNAcn or an accumulation of mtDNA mutations and deletions without affecting overall mtDNA levels[89]. The crystal structure of POLγ has provided insights into how different mutations affect enzyme function and replication fidelity[90]. However, the relationship between genotype and phenotype remains complex and unpredictable, with significant phenotypic variability occurring within members of the same family[91-93]. Pathogenic mtDNA mutations, whether homoplasmic or heteroplasmic, are responsible for various mitochondrial syndromes, such as mitochondrial encephalopathy, MELAS, LHON, MERRF, Pearson’s syndrome, and KSS. Studies have shown that mtDNAcn is elevated only in small cohorts of patients with either Pearson’s or KSS, which is caused by large heteroplasmic mtDNA deletions, and that there is no correlation between mtDNAcn and the size or position of the deletion[15,46].

More extensive investigations have been conducted on mtDNAcn variations in patients with mtDNA point mutations. A study involving patients with MELAS carrying the heteroplasmic m.3243A>G mutation found that high mtDNAcn and low heteroplasmy levels were associated with less severe disease[93,94]. Other studies involving smaller cohorts reported that MELAS and MERRF patients with moderately elevated mtDNAcn s exhibited milder phenotypes[95]. In patients with LHON, who mostly carry homoplasmic mutations, mtDNAcn has been implicated in disease penetrance, with asymptomatic carriers of the m.11778G>A or m.3460G>A mutations showing higher mtDNAcns than visually impaired patients[96,97].

mtDNAcn in cancer

The mtDNA polyploidy biomarker can classify different cancer types; changes in mtDNAcn have been reported in solid tumors and hematological malignancies, with higher levels reported in some cancer types and lower levels reported in others[15,62,80]. Cancer tissues normally display significant alterations in mtDNAcn (one hundredfold lower than normal tissue; mtDNA amplification up to twentyfold)[98]. An increase in mtDNA has been associated with mutations in the oncogenes Ras and Myc, and downregulation of Tfam expression (at mRNA and protein levels)[98,99].

Increased mtDNAcn has been proposed and observed as a potential mechanism to eliminate the effects of mitochondrial mutations in proliferative cancer cells. This association is explained by the selective growth advantage of tumor cells that have a higher mtDNA content than their non-cancerous neighbors[99,100]. Over time, this leads to gradual accumulation and propagation of mutant mitochondrial genomes in cellular and tissue mixtures[101]. Accordingly, early stage cancers with higher mtDNAcns are often accompanied by a lower mutation load.

Other cancer-related biomechanisms related to increased mtDNAcn include accelerated mitochondrial biogenesis, the Warburg effect, and increased bioenergetic demand, which are often proposed in parallel to maintain mitochondrial function and ATP production in cancer cells[101]. mtDNAcn levels can significantly distinguish cancerous tissues from non-cancerous tissues in matched samples[62,100,101]. This observation suggests the potential use of mtDNAcn level as a sensitive biomarker for early cancer screening. Decreased mtDNAcn has also been reported in breast cancer and is associated with poor prognosis[102]. Based on the relative detection thresholds of somatic mutations across different variant allele frequencies, some early-stage cancers with mtDNA deletions are difficult to detect.

mtDNAcn variation plays a significant role in cancer progression and varies across different cancer types, with both increases and decreases in mtDNA levels correlating with cancer risk and severity. In peripheral blood lymphocytes, studies have shown that an increase in mtDNA levels is associated with elevated cancer risk, as observed through a meta-analysis of the literature using qPCR[103]. In contrast, decreased mtDNA levels in bone tissue and peripheral blood lymphocytes are associated with a decreased risk of cancer[103].

In brain gliomas, elevated mtDNA levels in blood samples are associated with an increased risk of cancer[104], whereas in breast cancer, similar elevations in blood mtDNA levels have been reported to correlate with an increased cancer risk[105,106]. Blood samples from colon and rectal cancers exhibit increased mtDNA levels, further supporting the link between elevated mtDNA levels and cancer risk[107]. However, in the kidney, reduced mtDNA levels have been observed in peripheral blood lymphocytes from cancer patients[108], suggesting a complex role of mtDNA in different tissues. For lung cancer patients, increased mtDNA levels in blood samples are indicative of a higher cancer risk[109], and in pancreatic cancer cases, upregulation of mtDNA is similarly associated with cancer progression[110]. Patients with skin cancer also show increased mtDNA levels[106].

In matched tissue samples, changes in mtDNA levels correlated with disease severity. In bladder cancer, mtDNA levels are downregulated, contributing to an increase in disease severity[99]. Similar reductions have been observed in Ewing sarcoma[111], gliomas[112], and primary breast tumors[99,113], indicating a potential link between reduced mtDNA levels and more severe cancer stages. Conversely, increased mtDNA levels have been reported in colorectal carcinoma[114,115], colorectal adenoma[99,100], and lung cancer, where both small- and non-small-cell lung cancers show upregulated mtDNA, contributing to disease progression[116-118].

In kidney cancers, specifically chromophobe renal cell carcinoma, mtDNA levels are increased[119], whereas renal oncocytomas are decreased[120,121]. Hepatocellular carcinoma also shows a decline in mtDNA levels, which aligns with an increase in disease severity[99,122]. A notable decrease in mtDNA is observed in head and neck cancers, such as squamous cell carcinoma[99], as well as in certain prostate cancer types, particularly adenocarcinoma, where reduced mtDNA levels are linked to worse outcomes[122,123]. In the oral/digestive tract, increased mtDNA levels have been observed in esophageal cancer[99], whereas stomach cancer shows elevated levels[124]. Pancreatic endocrine tumors also display upregulation of mtDNA[123], and thyroid adenocarcinoma also exhibits an increase[123]. These variations in mtDNA levels across different cancers suggest that mtDNA could be a critical biomarker for both cancer risk assessment and monitoring disease progression.

These data reveal that mutations in mtDNA accumulate in virtually all types of cancer and have been associated with diagnostic or prognostic purposes. The pathophysiological relevance of these changes is still unknown; however, it is still unclear whether mtDNA mutations directly cause oncogenesis or whether they are merely the result of faster mtDNA replication in rapidly growing cancer cells. The proliferation of pre-existing heteroplasmic mutations or polymorphisms, which then undergo passive clonal expansion after several rounds of cell division during tumor formation, may be the cause of these mutations[125].

The notion that mtDNA mutations directly cause cancer has been refuted by strong evidence. Most tumor types, with the notable exception of kidney, colorectal, and thyroid cancers, counter-select for mutations that result in the truncation of mitochondrial proteins[123,126,127]. mtDNA levels have been found to be upregulated in these specific cancer types, indicating a compensatory mechanism meant to maintain mitochondrial function despite harmful mutations[123]. This observation underscores the complex relationship between mitochondrial dysfunction and tumor biology.

A comprehensive study that assessed the mutational landscape of over two-thousands cancer patients highlighted a strong strand bias in mtDNA mutations, indicating a replication-driven mechanism underlying their generation[123]. One likely scenario is that pre-existing heteroplasmic mutations, or mutations introduced by replication errors of DNA POLγ, may clonally expand to high frequencies in fast-dividing cancer cells, becoming fixed within certain tumor subpopulations over time. Oncocytomas, a type of tumor characterized by the accumulation of mtDNA mutations that lead to severe OXPHOS dysfunction, provide a compelling example of this mechanism. In these tumors, mitochondrial dysfunction is accompanied by a compensatory increase in mitochondrial mass, which is likely to sustain bioenergetic capacity despite the OXPHOS defects[128,129]. Although oncocytomas are driven by high levels of mtDNA mutations, they typically present as benign lesions with low invasiveness and nonaggressive clinical phenotypes.

However, the extent to which mitochondrial dysfunction acts as a barrier for cancer progression remains unclear. Although some forms of mitochondrial impairment may limit the malignant potential of certain tumors, there is evidence to suggest that in some cancer types, the accumulation of mtDNA mutations can provide a selective advantage. This facilitates the survival and proliferation of transformed cells, directly contributing to cancer development. Supporting this hypothesis, a study demonstrated that metabolic alterations induced by mtDNA mutations promote tumorigenesis in colon cancer[130].

In addition to the mutational burden, the abundance of mtDNA within a cell has emerged as a critical biomarker of a cell’s reliance on OXPHOS for energy production. Therefore, the regulation of mtDNAcn may be indicative of the bioenergetic adaptations that cancer cells undergo in response to mitochondrial dysfunction. The relationship between mtDNA mutations, their functional consequences, and their contribution to cancer progression continue to be the subject of intense investigation in the field of cancer biology.

Mounting evidence suggests that mtDNAcn may be dysregulated during carcinogenesis, aging, and neurodegenerative disease progression, among other disorders. Owing to its unique characteristics, such as maternal inheritance and absence of histones[2,6,15,48,63], mtDNA can be isolated not only from tissues but also from body fluids, such as plasma or saliva, among others. Moreover, the qPCR quantification of mtDNA is simple, inexpensive, and widely available. Hence, diagnostic applications of mtDNAcn in body fluids have gained significant attention. Nevertheless, there are important challenges that must be overcome to fully realize this potential. For example, a pre-analytical step must be conducted prior to any qPCR quantification of plasma to prevent potential contamination of qPCR reagents, calculation bias, and amplification of unreliably detected mtDNA. There is also the question of whether mtDNAcn response in tissues can be replicated in bodily fluids. Integration across the ranges of fluid and tissue types studied is necessary to understand the potential cross-application of tools and techniques and to identify species that cannot be integrated and require tailored approaches.

Further research is needed to elucidate these matters while ensuring that the use of mtDNAcn in fluid biopsies never compromises its application to tissues. There is also a substantial demand for computational tools that enable comparisons between different cohorts to harmonize nonlinear data. Moreover, sample pooling to reduce processing and qPCR costs can seriously confound the diagnostic strategies. Population-level cutoff norms for mtDNAcn must be established to ensure better comparability of the results. Finally, factors affected by sampling strategy and ethnic diversity must be explored further, as they can largely contribute to the inter-study variability already seen in the mtDNAcn literature. Addressing all these known hurdles will be fundamental to ensuring proper diagnostics using mtDNAcn in fluid biopsies in the future (Table 4).

Table 4 Role of mtDNA copy number in various stages of cancer development and progression.

Aspect	Role of mtDNAcn	Mechanism/impact	Examples
Cancer risk	Altered mtDNAcn (increase or decrease) may predispose individuals to cancer development	Imbalance in ROS production	Decreased mtDNAcn linked to breast cancer risk
		Compromised cellular energy metabolism	Increased mtDNAcn linked to lung cancer risk
Tumor initiation	Changes in mtDNAcn can affect mitochondrial biogenesis and metabolic reprogramming	Promotes a shift to aerobic glycolysis (Warburg effect)	Low mtDNAcn observed in colorectal cancer tissues
		Increases ROS, leading to genomic instability
Tumor progression	Dynamic changes in mtDNAcn support adaptation to tumor microenvironment	High mtDNAcn enables oxidative metabolism in hypoxic conditions	Elevated mtDNAcn associated with metastatic breast cancer
		Supports invasive and metastatic properties
Therapeutic resistance	Altered mtDNAcn contributes to drug resistance	High mtDNAcn enhances oxidative phosphorylation, reducing sensitivity to certain chemotherapies	Increased mtDNAcn linked to resistance in lung cancer treatments
Prognostic biomarker	mtDNAcn alterations can predict cancer outcomes	Low mtDNAcn correlates with poor prognosis in many cancers	Reduced mtDNAcn in gastric cancer linked to poor survival
		High mtDNAcn may predict aggressive tumor behavior
Immune evasion	Changes in mtDNAcn influence immune responses within the tumor microenvironment	mtDNA release into the cytoplasm activates inflammatory pathways	mtDNA-derived DAMPs in melanoma
		Alters immune surveillance mechanisms
Angiogenesis	mtDNAcn modulates energy demand and oxidative stress, indirectly affecting vascular growth	High mtDNAcn supports angiogenic signaling	Increased angiogenesis in glioblastoma with altered mtDNAcn
Metastasis	Altered mtDNAcn facilitates energy supply for metastatic spread	Provides metabolic flexibility for survival in secondary sites	Elevated mtDNAcn in metastatic colorectal cancer

mtDNAcn: Mitochondrial DNA copy number; ROS: Reactive oxygen species; DAMPs: Damage-associated molecular patterns.

THERAPEUTIC IMPLICATIONS OF MODULATING mtDNAcn

Abnormal mtDNAcn is involved in the pathogenesis of many human diseases, and numerous clinical studies have shown that alterations in mtDNAcn are correlated with disease conditions. Therefore, modulation of mtDNAcn has therapeutic implications for human health and disease. The development of a versatile genome editing tool is required for the examination of mtDNAcn and for a fundamental understanding of the pathogenic mechanisms of altered mtDNAcns. Platform technologies for the modulation of mtDNAcn and their application in human diseases are also desired.

The metabolic stress sensor 5’ adenosine monophosphate-activated protein kinase (AMPK) is activated by a decrease in the ATP/adenosine monophosphate ratio at the onset of various stress conditions, including energy deprivation, hypoxia, and nutrient starvation[131]. Once activated, AMPK maintains cellular energy homeostasis by enhancing ATP-generating pathways, such as fatty acid oxidation in the mitochondria, and the inhibition of energy-consuming biosynthetic pathways, such as lipid and protein synthesis[131,132]. Reports also have demonstrated that AMPK also upregulates mitochondrial biogenesis through the activation of the transcriptional coactivators PGC-1α and PGC-1β[133]. Furthermore, the activation of AMPK increases mtDNAcn in several human somatic cells, including endothelial and cancer cells. Depletion of mtDNA by chemical inhibition of mtDNA replication decreases the levels of mtDNA-encoded genes and OXPHOS complexes, resulting in the activation of stress response pathways, including the mitochondrial unfolded protein response[131-133]. Activation of unfolded protein response is known to enhance mitochondrial biogenesis and the clearance of damaged mitochondria, thus improving mitochondrial homeostasis[134].

Modulation of mtDNAcn has therapeutic implications in human diseases such as age-associated neurodegenerative diseases. Modification of the mtDNAcn is an attractive option for the development of therapeutic modalities against age-associated neurodegenerative diseases. Thus far, classes of small molecules, vitamins, and ionophores that increase mtDNAcn have been identified. Chemical screening using human cell models is a promising approach for the identification of additional small molecules that alter mtDNAcns. Two primary therapeutic strategies have been explored to enhance OXPHOS capacity and restore mitochondrial function. The first involves increasing the overall mitochondrial mass, while the second focuses on selectively altering mtDNA to modulate its copy number and/or heteroplasmy levels. These approaches predominantly target treatment of primary mitochondrial disorders. However, if human clinical trials yield positive outcomes, they may have broader applications in the management of age-related diseases.

One promising strategy to improve the balance in favor of functional mitochondria is to focus on selective elimination of damaged mitochondria[135]. This is accomplished by promoting autophagy, a cellular mechanism that breaks down and recycles a variety of components, including mitochondria, to increase mitochondrial turnover. Inhibition of mammalian target of rapamycin complex 1 (mTORC1), a crucial regulator of food sensing and metabolism, is a method of inducing autophagy[136,137]. The intricacy of the mTORC1 pathway, which controls a variety of cellular processes, is probably the reason why rapamycin-mediated mTORC1 suppression has been inconsistently beneficial in improving mitochondrial function in certain animal models[137]. However, in one study, rather than autophagy activation, the effect of rapamycin in a mouse model of leigh syndrome appeared to be due to a metabolic shift toward amino acid catabolism[138]. As a result, off-target effects may limit the therapeutic use of mTORC1 inhibitors such as rapamycin[139,140].

By increasing longevity in model organisms such as Caenorhabditis elegans and enhancing muscle function in elderly mice, more specialized autophagy enhancers, including urolithin A, have shown promise[140]. However, the regulatory processes controlling autophagy and mitochondrial quality are extremely intricate and poorly understood. Excessive mitochondrial clearance caused by the dysregulation of these mechanisms may result in negative long-term repercussions. Before embarking on therapeutic development based on these principles, caveats and concerns associated with manipulating mtDNAcns should be considered (Table 5).

Table 5 Comparison of techniques for modulating mtDNA copy number therapeutic strategy.

Therapeutic strategy	Mechanism	Applications	Advantages	Challenges
AMPK activation	Enhances mitochondrial biogenesis via PGC-1α activation	Neurodegenerative diseases, aging	Promotes energy balance	Off-target effects, limited clinical trials
Genome editing tools	Targets mtDNA mutations or modulates copy number	Mitochondrial diseases, cancer therapy	Precision targeting	Ethical concerns, technical challenges
Autophagy induction	Removes damaged mitochondria	Improves mitochondrial quality	Enhances cellular health	Excessive clearance may have long-term side effects
Small molecules and vitamins	Increases mtDNAcn	Metabolic and neurodegenerative disorders	Cost-effective	Limited understanding of long-term effects

AMPK; Adenosine monophosphate-activated protein kinase; PGC-1α: Peroxisome proliferator-activated receptor gamma coactivator 1-alpha; mtDNA: Mitochondrial DNA; mtDNAcn: Mitochondrial DNA copy number.

Despite years of effort aimed at elucidating the regulation of mtDNAcn, our understanding remains rudimentary. There are several obstacles in mtDNAcn research. First, the rate-limiting steps of mtDNA replication remain elusive, forcing scientists to imagine rather than experiment with the mtDNA replication mechanism. Second, the expression of genes involved in the regulation of mtDNAcn is tissue-selective, and the regulatory mechanisms responsible for mtDNAcn alterations remain poorly defined. Additionally, discrepancies between the effects of mitochondrial respiration on mtDNAcn observed in different laboratory settings remain unclear. Unexpectedly, an increased mtDNAcn was observed under high glucose conditions in a cell type and in an experimental background-dependent manner. Considered as a simple internal control in qPCR, chloroplast DNA copy number was also reported to be altered in proud low mtDNAcn cells (ND7 and 964) but not in other cells. Third, it has been difficult to evaluate changes in mtDNAcn in human tissues owing to the complexity of the tissue architecture and biological variability. Controversies regarding the increase or decrease in mtDNAcn have been observed among comparable disorders across different laboratories. These confounding factors collectively hinder the understanding and comparison of mtDNAcn alterations in various diseases (Table 6).

Table 6 Broad spectrum of therapeutic applications of mtDNA copy number.

Application	Role of mtDNAcn assessment	Therapeutic implications	Examples
Cancer prognostication	mtDNAcn serves as a biomarker for predicting cancer outcomes	Guides risk stratification and treatment intensity	Low mtDNAcn linked to poor prognosis in gastric and colorectal cancers
		Identifies patients with aggressive disease
Therapeutic targeting	Abnormal mtDNAcn highlights mitochondrial vulnerabilities	Enables development of drugs targeting mitochondrial pathways (e.g., OXPHOS inhibitors)	mtDNAcn modulation as a target in ovarian and pancreatic cancer therapies
Monitoring treatment response	Changes in mtDNAcn reflect tumor response to therapy	Serves as a real-time marker to monitor chemotherapy, radiotherapy, or immunotherapy efficacy	mtDNAcn alterations used to monitor cisplatin therapy in ovarian cancer
Personalized medicine	mtDNAcn variations help tailor therapies based on mitochondrial function	Facilitates selection of specific treatment modalities (e.g., glycolysis inhibitors vs OXPHOS inhibitors)	mtDNAcn guiding metabolic therapy choices in lung and breast cancer
Radiotherapy sensitization	Altered mtDNAcn may increase sensitivity or resistance to radiotherapy	Identifies patients who might benefit from combined mitochondrial and radiotherapy interventions	Elevated mtDNAcn linked to radio-resistance in glioblastoma
Metabolic modulation	mtDNAcn assessment reveals metabolic dependencies of tumors	Guides therapies targeting cancer metabolism (e.g., ketogenic diets, mitochondrial uncouplers)	Low mtDNAcn tumors treated with glycolysis inhibitors
Early disease detection	mtDNAcn alterations in blood or tissue serve as a non-invasive biomarker for early cancer detection	Allows early initiation of treatment, potentially improving outcomes	Reduced mtDNAcn detected in circulating cell-free DNA in lung and breast cancers
Combination therapies	mtDNAcn dynamics predict synergy between mitochondrial-targeted drugs and conventional therapies	Combines metabolic modulators with standard chemotherapy or immunotherapy for enhanced efficacy	mtDNAcn-directed combination strategies in melanoma treatment
Toxicity management	mtDNAcn levels predict susceptibility to mitochondrial toxicity from certain drugs	Assists in preemptive dose adjustments or alternative drug selection to avoid adverse effects	Monitoring mtDNAcn to prevent cardiotoxicity from anthracyclines
Rare mitochondrial disorders	mtDNAcn assessment aids in the diagnosis and management of mitochondrial diseases with cancer overlap	Develops therapies that normalize mtDNAcn or enhance mitochondrial biogenesis	mtDNAcn restoration therapies in mitochondrial depletion syndromes

mtDNAcn: Mitochondrial DNA copy number; OXPHOS: Oxidative phosphorylation.

CONCLUSION

mtDNAcn is a crucial biomarker for understanding mitochondrial health, cellular energy metabolism, and disease pathogenesis. Its quantification using methods such as qPCR and NGS plays a vital role in the early diagnosis, prognosis, and monitoring of treatment responses in a wide range of diseases, including cancer, neurodegenerative disorders, and metabolic conditions. However, challenges, such as standardizing methodologies, addressing inter-study variability, and ensuring accurate and reproducible measurements, remain. Overcoming these obstacles is essential to effectively integrate mtDNAcn assessments into clinical diagnostics and therapeutic strategies.