Chemoresistant cancers are difficult to treat using chemotherapy due to the stemness-causing factors include SOX2, Oct4, ERCC1, Pg-P, ALDH1, etc., to foster drug efflux and drug resistance (1). This stemness is leading to phenotypic cancer stem cells and forms a distinct subpopulation with substantial self-renewing capacity across tumor microenvironment (2–8) which enhance the tumor progression (9, 10). CRC is reported as 3^rd most commonly diagnosed cancer in men or women in United States, highlighting its profound public health impact (2). Approximately 35% of CRC risk is due to genetic inheritance (2). Genome-wide association studies (GWAS) have identified fifty genetic loci linked to an increased risk of CRC. These studies primarily highlight common variants located within the nuclear genome (3–7). But these identified loci account for only a small portion of the disease’s heritability, suggesting that additional genetic contributors remain undiscovered. More than seven decades ago, Otto Warburg described a hallmark metabolic alteration in cancer cells: despite the availability of sufficient oxygen for mitochondrial respiration, cancer cells acquire a higher glucose uptake and depends on the glycolysis for energy generation (8). This metabolic shift was indicative of a fundamental defect in mitochondrial respiration, which he hypothesized to be a primary cause of cancer (9). This metabolic reprogramming, now termed the Warburg effect, has been observed across numerous cancer types, including CRC, where enhanced glucose transport and glycolytic activity are frequently observed (10–12).

Human mtDNA is a circular, double-stranded and comprising 16,569 base pairs, with 10^3 to 10^4 copies present per cell. It encodes 37 genes, including two rRNA genes, 22 tRNA genes as well as thirteen protein-coding genes; these mitochondrial genes are crucial for oxidative phosphorylation (OXPHOS) (13, 14) and involved to foster the function of respiratory chain complexes: complex I (ND1, ND2, ND3, ND4, ND4L, ND5, ND6), complex III (cytochrome b), complex IV (COXI, COXII, COXIII), and complex V (ATPase6 and ATPase8) (15). Additionally, mtDNA contains a non-coding displacement loop (D-loop) region, crucial for regulating mtDNA replication and transcription (16, 17).

Variations within mtDNA, such as mitochondrial single nucleotide polymorphisms (mtSNPs), can profoundly affect mitochondrial function by altering the efficiency of OXPHOS and cause a higher ROS generation, which subsequently elevates the risk of cancer development (18–23). However, studies exploring mtDNA variants and CRC risk have yielded inconsistent results. For example, a Scottish study analyzing 132 mtSNPs in a cohort of 2,854 CRC patients and 2,822 controls described the absence of association between these variants and overall risk of acquisition of CRC (12, 24). At present, there is a lack of comprehensive research examining the relationship between mtDNA variations and CRC susceptibility across diverse racial and ethnic groups. A more focused, pathway-based methodology could potentially reveal novel connections between the mitochondrial genome and cancer risk. This approach would facilitate a more efficient analysis of variants that may have subtle effects on CRC susceptibility. Additionally, integrating multi-omics data and advanced bioinformatics tools could enhance our understanding of how mtDNA variations contribute to chemoresistant metastatic CRC pathogenesis, potentially leading to the discovery of new biomarkers and therapeutic targets. This review focuses on several key aspects of mtDNA in CRC. It explores how specific mtDNA mutations drive metastasis in CRC and their potential as prognostic markers or therapeutic targets. Additionally, it examines mitochondrial copy number variations (CNVs) and their implications for CRC. The review also highlights the connection between inherited and somatic mtDNA mutations with CRC susceptibility, particularly mutations in ND genes. Furthermore, it discusses how epigenetic dysregulation, referred to as mitoepigenetics, contributes to CRC progression and chemoresistance, emphasizing the role of epigenetic proteins in these processes.

We conducted a vivid literature review, drawing from a variety of reputable databases such as PubMed, Medline, Scopus, Google Scholar, National Library of Medicine (NLM), and ReleMed. Our analysis centered on evaluating published studies and reports that explore the role of mtDNA in colorectal cancer, with a particular emphasis on the D-loop region and its significance in metastasis. Additionally, we described the relationship between mutations in mtDNA and alterations in OXPHOS, which contribute to cancer growth and development of chemoresistance in colorectal cancer. The review also covered topics including the role of ROS, mitochondrial mutations in colorectal cancer progression, and the development of therapies targeting these mutations to address chemoresistance in metastatic colorectal cancers.

Recent integrative genomic frameworks such as DiffInvex have illuminated how selective pressures imposed by chemotherapy dynamically reshape the somatic mutational landscape across cancer types (32). By leveraging an empirical baseline mutation rate derived from non-coding DNA, DiffInvex identifies shifts in positive and negative selection acting on individual genes, providing a powerful lens for understanding adaptive resistance in metastatic colorectal cancer (mCRC). Application of this model to over 11,000 tumor genomes across ~30 cancer types revealed that chemotherapeutic exposure can induce treatment-associated selection in genes including PIK3CA, APC, MAP2K4, SMAD4, STK11, and MAP3K1, each of which contributes to critical signaling networks governing tumor survival, EMT, and immune evasion. These findings describe that mutational evolution under drug stress fosters clonal diversification and heterogeneity, key hallmarks of chemoresistant tumor phenotypes (32).

In mCRC, actionable genomic alterations such as HER2 amplification, BRAF V600E mutation, NTRK fusions, and MSI-H status have revolutionized therapeutic precision (33). However, the majority of CRC-associated mutations remain “undruggable,” and patients often develop adaptive resistance through secondary mutations or compensatory pathway activation. Recent evidence implicates GNAS mutations as potential molecular predictors of aggressive disease behavior and therapeutic refractoriness, highlighting their diagnostic and prognostic significance (34).

Moreover, the RNF43 gene exhibits pronounced mutational intratumoral heterogeneity (ITH) in both gastric and colorectal tumors, reinforcing the need for spatially resolved genomic profiling to accurately capture regional mutation spectra and therapeutic vulnerabilities (35). Complementary studies have identified RNF11 as another critical mediator of CRC progression, functioning through differential mRNA expression and ubiquitin ligase activity that promote tumor proliferation and immune escape, thereby representing an emerging therapeutic target (36).

Adding to this complexity, stromal-tumor crosstalk mediated by Wnt5a and hypoxia-induced fibroblasts (InfFib) establishes a pro-tumorigenic microenvironment in colorectal carcinoma. Wnt5a, expressed by inflammatory fibroblasts under hypoxic conditions, reinforces tumor angiogenesis suppression through VEGFR1 (Flt1)-dependent pathways and sustains a hypoxic niche that drives epiregulin production thereby potentiating tumor growth and metastasis (37). Concurrently, m6A RNA methylation regulators, including METTL3 and YTHDC1, orchestrate post-transcriptional control of metastasis-associated transcripts such as NRXN3, forming a METTL3–YTHDC1–NRXN3 axis that facilitates peritoneal dissemination of CRC (37, 38).

Collectively, these insights delineate a multifactorial interplay between mutation-driven selection, epigenetic remodeling, and microenvironmental adaptation, which together fuel the evolution of chemoresistance in metastatic colorectal cancer. Future research integrating spatiotemporal genomics, epigenetic mapping, and immune landscape profiling will be crucial for defining actionable vulnerabilities and developing multi-targeted immunoepigenetic therapies to overcome tumor heterogeneity and therapeutic resistance.

Mitochondria have prominent implications in the pathophysiology of diseases neurological ailments include dementia, other neurodegenerative conditions like Alzheimer’s disease, Parkinson’s disease and multiple sclerosis (39–50). However, the mitochondrial contributions to these diseases are often secondary and not yet fully understood. Neoplastic cells often exhibit metabolic imbalances, accumulating changes that manifest in advanced clinical phenotypes. Mitochondrial mutations are increasingly observed in cancers (44, 51–57), although whether these mutations are causative or consequential remains a question and warranted future studies. Severity of mitochondrial ailments is influenced by ‘biochemical threshold’, a point at which the proportion of mutant mtDNA exceeds a critical ratio inside the cells. This threshold can affect multiple tissues or be confined to specific ones, contributing to the diagnostic and mechanistic complexity of mitochondrial diseases. In another a few cells, normal mtDNA could mitigate the effects of mutated variants through rescue mechanisms include mitochondrial fission/fusion (39, 45, 58). Variations in mtDNA content can influence both metabolic processes and nuclear epigenetic modifications (43, 59–61). While it is hypothesized that the epigenetic modifications in mtDNA might be influenced by mtDNA copy number, this remains to be thoroughly investigated.

At birth, mtDNA is homoplasmic, meaning it is identical across all cells, though the copy number varies by tissue. With aging, mtDNA accumulates mutations leading to heteroplasmy, where cells contain a mix of different mtDNA sequences. This heteroplasmy influences cellular evolution and impacts disease severity and subtype (45, 58, 62). Heteroplasmy is altered depending on tissue type as well as energy requirements. This highlights the significance of considering both mtDNA sequence and copy number in cancer studies. Advanced genome sequencing techniques have increased the detection of heteroplasmy (63, 64), complicating the distinction between driver and ancillary mutations. Persistent heteroplasmy generally attributed to the phenotype instability in dividing cells (65), suggesting a selection pressure towards homoplasmy even in cancer cells (66, 67). Recent studies on iPSCs indicate that homoplasmy is crucial for maintaining pluripotency (68), though it is unclear if the same applies to cancer stem cells (69). The mtDNA could code quantitative trait loci (QTLs) which interact with nuclear genome for regulating the intricate disease process (70–72). It is reported that the phenotypic alterations in cells are induced due to the intricate interplay of SNPs, mutations, and environmental factors. While mitochondrial polymorphisms alone are unlikely to fully explain disease progression, they do influence disease progression by modulating gene-gene interactions subsequently alter the tumor microenvironment. This interaction is crucial since mtDNA is maternally inherited, yet signals from the tumor microenvironment can modulate metastasis efficiency. Not all mtDNA SNPs will act as QTLs for specific phenotypes. Identifying mtDNA mutations as drivers of cancer has been challenging due to experimental and technological limitations. In breast cancer studies, mtDNA mutational burden showed no correlation with survival (73), yet TCGA datasets have revealed intriguing correlations (66, 67). Determining definitive cause-and-effect relationships is challenging due to the presence of numerous copies of mtDNA per cell and the difficulty in manipulating all these copies simultaneously (45, 74, 75).

Mitochondrial CNVs in metastatic cancers: A few published reports indicates that mtDNA CNVs are present in various cancers (47, 51, 76–78). For instance, ovarian cancers often have more than 600 copies, while myeloid cancers have around 90 copies. Increased mtDNA is evident in the disease conditions such as chronic lymphocytic leukemia, lung squamous cell carcinoma, and pancreatic adenocarcinoma; the extent of mtDNA is lesser in the disease conditions include kidney clear cell carcinoma, hepatocellular carcinoma, and myeloproliferative neoplasms. Copy number variants is reported to have a positive correlation with age of the individuals diagnosed with prostate cancer and colorectal cancer (76, 79). These findings concluded the complexity of mitochondrial biogenesis regulation in oncogenesis and metastasis (69).

Approximately 35% of colorectal cancer cases are attributed to inherited susceptibility, with a small fraction due to known genetic mutations (2, 80). Genome-wide association studies reported several low-penetrance susceptibility loci correlated to CRC, demonstrating the role of common genetic variations (3, 5, 81). Mitochondria are crucial for energy metabolism, ROS generation, and apoptosis regulation, all of which are involved in cancer development (82–86). A higher generation of ROS in dividing tumor cells could cause oxidative stress subsequently fosters DNA damage, leading to genetic instability (87–89).

Somatic mtDNA mutations are found in several cancer types, including CRC (90). Although their causal role remains unclear, it is plausible that variant mitochondrial functions could cause cancer risk, as suggested by associations with breast cancer susceptibility (91). A comprehensive evaluation of mtDNA variants and CRC risk has involved genotyping 132 tagging variants, capturing about 80% of common mitochondrial variation, in a large case-control study (24). However, the potential role of low-frequency mtDNA variants or gene-environment interactions remains a possibility. Future research should focus on larger sample sizes and incorporate non-genetic covariates to effectively describe the implications of mitochondrial variations in CRC. The complex interaction between nuclear and mitochondrial genomes, along with the influence of environmental factors, describes the need for integrated studies to unravel the multifaceted roles of mtDNA in disease progression and metastasis.

Mitochondria contain extrachromosomal DNA. Mitochondrial haplogroups, which are defined by unique sets of mitochondrial single nucleotide polymorphisms (mtSNPs) describes specific ancestral populations; these are linked to incidence of various cancers, including breast cancer and nasopharyngeal cancers (92–95). However, research examining the association between mitochondrial haplogroups and CRC risk in European and Asian populations has produced inconsistent results (24, 94, 96). Another report described the functions of mtDNA in the risk of getting CRC in several ethnic groups by examining 185 mtSNPs (12). Germline mtDNA polymorphisms may contribute to cancer disparities. Implications of germline and somatic mutations and transcriptional activities of mitochondrial genes using whole-genome sequencing of 38 tumor types (76). According to this report (76), MT-ND5 is identified as the most recurrently mutated electron transport chain gene in diverse cancer types, while MT-ND4 and MT-COX1 were most commonly mutated in other cancer types include prostate cancer, lung cancer, breast cancer, and cervical cancer types respectively. Most mutations involved a C:G>T:A transition in over 50% of cases (76). Somatic mtDNA mutations arise early in neoplastic cell lineages and progressively shift towards homoplasmy over time. This progression towards homoplasmy could be due to asymmetric segregation at the time of cell proliferation or the selective advantage of specific mutations. In subsets of kidney or thyroid carcinomas lacking identifiable oncogenic drivers, non-silent mtDNA mutations imply a main role for these mutations. The selective pressure against truncating mutations in mtDNA-encoded proteins highlights the critical importance of maintaining an intact ETC for the survival of most cancer cells, with notable exceptions in kidney, colorectal, and thyroid cancers (69, 76).

Furthermore, oncogenes are involved in modulating the metabolism. For instance, the p53 P72R gain-of-function mutation could modulate the function of mitochondrial PGC1α, which could cause poorer cancer prognosis (97–100). Additionally, c-Myc is involved in the mtDNA fragmentation (101). This raises questions about whether oncogenes can modulate tumorigenesis with the involvement of mtDNA mutations or SNPs; it is crucial to explore whether combination effects of mtDNA as well as nuclear DNA QTLs involved in susceptibilities to cancer and metastasis. Mutations in mtDNA vary in frequency and location across different cancers; prostate cancer and colorectal cancers exhibiting the highest mutation rates, while heme malignancies generally exhibit minimal mutation rate (67). It is crucial to examine the influence of selective advantage of specific mtDNA mutations for CRC risk? and other tissue-specific alterations in mitochondrial DNA mutations for mediating oncogenesis or metastasis of chemoresistant CRC. Finally, the implications of germline mutations in mitochondrial DNA in specific to SNPs could explore the racial disparities in the oncogenesis and metastasis of CRC. For instance, a few previous reports described the relative risk associated with mitochondrial DNA haplotypes as described in Table 1 (69).

As discussed in the above, alterations in the mtDNA genome could cause CRC risk. Recent reports described the association of mtDNA variants with canonical haplotypes in CRC risk, for instance, the variants that capture 79% of all polymorphic variants with a minor allele frequency (MAF) >1% and 92% of variants with MAF >5% (24). Post hoc analyses suggested a stronger association between the A5657G variant and colon disease, instead of rectal disease, and a link between microsatellite instability (MSI) in CRC and the T4562C variant. Tumor hypoxia, which impairs the DNA mismatch repair system by downregulating MMR genes like MLH1, might explain these findings. Since A5657G is non-coding and T4562C is synonymous, their effects are likely indirect, possibly mediated by untyped SNPs (24, 102–104). The lack of consideration for mtDNA heteroplasmy in CRC yet requires future studies. A previous report examined whether common mtDNA variations influence CRC risk by genotyping 132 tagging mtDNA variants in 2854 CRC cases and 2822 controls, covering about 80% of common mtDNA variation (excluding the hypervariable D-loop). The strongest association in single marker tests was with A5657G individuals. Examining the cohorts by segregating into the nine common European haplogroups and comparing their distribution in cases and controls also showed no evidence of association between mtDNA genome variations and risk of CRC development but this association is yet to be proven in the chemoresistant metastatic CRC (24).

Future research directions should focus on exploring the functional impacts of low-frequency mtDNA variants and heteroplasmic mutations. Advanced sequencing technologies and larger cohorts will enhance the resolution of such studies. Investigating the interaction between mitochondrial and nuclear genomes, and how these interactions contribute to cancer progression under different environmental conditions, will be crucial. Additionally, longitudinal studies assessing mtDNA variation and heteroplasmy changes over time in cancer patients could provide insights into their roles in cancer progression and response to therapy (24).

Mitochondrial mutations, SNPs and colorectal cancer in specific populations: A previous report described the functional role of mitochondrial genome pertinent to CRC risk among 14,383 colorectal cancer cases and controls (12). This research systematically evaluated mitochondrial genome, and its pathways, gene sets, as well as implications of haplogroups across various maternal racial and ethnic groups in relation to CRC (12). This pathway analyses suggested a main role of mitochondrial genome as well as OXPHOS pathway in CRC risk in European Americans. Specifically, authors identified an association between the MT-ND2 gene with the risk of acquisition of CRC in European Americans, with a more pronounced correlation observed in colon cancers (12). Furthermore, haplogroup T is involved in the CRC risk among European Americans irrespective of global ancestry race. Thus, functional implications of the identified mitochondrial mutations related to CRC risk was described. For example, variations in MT-ND2 gene, which encodes a subunit of NADH dehydrogenase (Complex I), could potentially disrupt electron transport and increase ROS production, contributing to CRC pathogenesis. Another report described the overexpression of MT-ND2 in CRC tissues than normal tissues which has correlation with reduced mtDNA D-loop methylation, and correlated to stages of CRC pathogenesis (21, 105, 106). This report described the functional aspects of MT-ND2 in the development of CRC.

Additionally, the OXPHOS pathway’s involvement in CRC underscores the importance of mitochondrial bioenergetics in cancer development (12). Haplogroup T’s association with CRC risk suggests that inherited mitochondrial variations can influence cancer susceptibility. This finding aligns with previous research showing that certain mitochondrial haplogroups are linked to metabolic traits and disease risks. These insights into the mitochondrial genome’s contribution to CRC risk pave the way for future research to understand the complex interplay between mitochondrial genetics, cellular metabolism, and cancer. Further studies are needed to explore the mechanistic pathways through which mitochondrial variations influence metastatic CRC development and to explore potential therapeutic targets within the mitochondrial genome (12).

For instance, the distribution of mitochondrial haplogroups within the Multiethnic Cohort Study (MEC) aligns with previously reported data pertinent to U.S. population-based samples, corroborating existing population genetics research (107). Specifically, the prevalence of haplogroup T among European American controls (9.57%) corresponds with findings from the Mitomap database, which indicates a frequency range of 8% to 11% across Western to Eastern European populations, as well as with data from non-Hispanic Whites in the National Health and Nutrition Examination Surveys (NHANES), which reports a similar frequency of 9.6% (107, 108). Previous studies in Chinese and Scottish cohorts did not find associations between mtDNA haplogroups and incidence risk of CRC (24, 94). However, an association was noted between haplogroup B4 with the incidence risk of CRC risk in Korean patient cohort (12, 96). Another report described a correlation between haplogroup T with the incidence risk of CRC risk in European Americans, independent of overall genetic ancestry (12). Haplogroup T is distinguished by a set of 9 polymorphisms (109, 110), which include a total of 5 RNA variants (G709A, G1888A, T8697A, T10463C, G15928A), three synonymous mutations (G13368A, G14905A, A15607G), as well as one nonsynonymous mutation (A4917G). Mutation A4917G, which serves as defining marker for haplogroup T, is a conserved polymorphism within the MT-ND2 gene (95, 109, 110).

In the Scottish cohort, an analysis of 132 mtSNPs revealed no overall CRC risk association, though the A5657G variant in tRNA, with a minor allele frequency (MAF) of 0.01, was linked to colon tumors (24). The implications of SKAT common/rare approach, which enhances power by collectively testing multiple risk alleles with modest effects, addressing the limitations of single SNP tests, especially in the context of correlated SNPs and the need to balance the influence of rare variants (12, 111–114).

Most of the existing research focuses on mutations within the coding regions of mtDNA. A previous report found no significant overall correlation between mitochondrial haplogroups and CRC risk (24). However, they identified an association between the A5657G mutation in the non-coding region located between mt-tRNAAla and mt-tRNAAsp and the incidence of colorectal cancer, as opposed to rectal cancer. Furthermore, a synonymous mutation in the MT-ND2 gene (T4562C) was associated with microsatellite instability in CRCs, indicating a potential role in cancer pathogenesis (24).

Research involving mitochondrial-nuclear exchange (MNX) mice has provided crucial insights, building on Ishikawa’s pioneering work with cybrids, which demonstrated the influence of mitochondrial transfer on metastasis (115–117). Mutations in the mtDNA, particularly those disrupting complex I, such as the insertion mutation 13885insC in the MT-ND6 gene, have been shown to increase ROS production and enhance metastatic potential (115, 118). These mtDNA alterations were also found to upregulate the expression of genes involved in glycolysis and metastatic processes (118). Furthermore, specific mutations in the MT-ND6 (C12084T) and MT-ND5 (A13966G) genes were linked to increased metastatic activity, as exemplified by the MT-ND6 mutation, which heightened invasiveness in A549 lung cancer cells (117, 119). Additional mutations in NADH dehydrogenase genes, including “T3398C, T12338C, C3689G, G3709A, G3955A, T10363C, C11409T, G13103A, and T14138CC” in MT-ND1, as well as “G12813A, G13366A”, and a premature truncation 14504delA in MT-ND5 or MT-ND6, were involved in mediating distant metastasis (118, 120, 121). Two SNPs in MT-ND1 (C3497T and T3394C) were particularly noteworthy, suggesting that ancestral genetic differences might influence the cancer pathogenesis (118, 122). The association between various mtDNA haplotypes and the predicted risk of different cancers. Each row corresponds to a specific mtDNA haplotype (Tables 2, 3), defined by unique polymorphisms, and details the relevant mutations within mitochondrial genes.

Further evidence from studies on MNX mice indicated that mtDNA SNPs in the stroma could impact metastatic potential, paralleling findings of the T3394C mutation’s role in adjacent mucosal tissues in non-small cell lung cancer and colon tumors, pointing towards inherited susceptibilities to metastasis (118, 122). The metabolic shift from OXPHOS to glycolysis, accompanied by enhanced heteroplasmy, has been observed in invasive versus non-invasive breast cancer cells (122). However, this metabolic reprogramming is not universally described in different carcinoma types, suggesting variability in mitochondrial involvement (79, 123).

The role of mitochondrial antioxidants, particularly mitochondrial catalase (mtCAT), has also been emphasized in metastasis regulation. mtCAT has been shown to decrease macrophage infiltration and reduce the number of CD34+ endothelial cells, implying a suppression of angiogenesis, which is critical for tumor progression and metastasis (124, 125). This highlights the complex interplay between mitochondrial function, oxidative stress, and cancer progression, underscoring the need for further exploration into mitochondrial genetics and its impact on metastatic colorectal cancer biology (69).

The SNP A4917G in the T haplogroup was associated with risk of CRC incidence in a diverse population, but specific references for this SNP need to be verified separately. These Tables 2, 3 summarize the identified mtDNA mutations and polymorphisms associated with colorectal cancer based on the referenced studies. Each entry includes the cancer type, specific gene affected, mutation/polymorphism, and corresponding references (69). According to this report (69), In a study involving 2,453 cases of colorectal cancer and 11,930 control subjects, mtDNA-SNP of A4917G emerged as the most significant variant associated with cancer risk. This SNP, located within the T haplogroup, was identified across a diverse cohort, including American men and women of Asian, African, European, Latino, or Native Hawaiian descent (12). The presence of A4917G was correlated with a greater risk of acquiring CRC, suggesting a potential role in disease pathogenesis. Additionally, the G4655A SNP was reported with a heightened risk of CRC specifically in European-Americans. However, this association was not observed consistently across the broader population, indicating a possible interaction with nuclear genetic factors or environmental influences that modulate the impact of this mtDNA variant. Such differential risk profiles describe the complexity of mtDNA-nuclear DNA interactions and their influence on cancer susceptibility. Further research into these interactions could demonstrate the underlying mechanisms by which mtDNA variations contribute to colorectal cancer risk and offer insights into personalized risk assessment and targeted interventions (69).

Eight specific SNPs (A16163G, C16186T, T16189C, C16223T, T16224C, C16295T, T16311C, T16519C) showed significant differences between CRC patients and controls, indicating that these SNPs might increase CRC risk or be in linkage disequilibrium with other functional SNPs contributing to cancer risk. Notably, a thymine-to-cytosine transition at position 16519 (T16519C) was found in 70% of CRC samples (126). This variant, located in the tRNAVal region, may lead to metabolic impairment and resistance to apoptosis, potentially worsening the prognosis for CRC patients. Previous studies have associated the T16519C SNP with increased risk for breast cancer and poorer outcomes in pancreatic cancer (91, 127). Despite its frequent occurrence in healthy controls (43%), the crucial functions of this mutation remains unclear and warrants further investigation (126).

Mitochondrial D-loop/ND genes mutations and colon cancers: Mutations in D-loop can influence mtDNA transcription, leading to mitochondrial dysfunction and potentially contributing to cancer initiation and progression (128). Elevated ROS levels can have deleterious effects, including apoptosis induction and genomic damage, and can alter cellular fates, shifting from apoptosis to necrosis, which in turn influences nuclear DNA mutations, cell division, and tumor growth. Whether mtDNA variations are causative factors or secondary results of the neoplastic process remains an open question. Given the multifactorial nature of cancer and the critical role of mitochondria in ROS production and apoptosis regulation, further exploration into mtDNA D-loop variations in cancer patients is essential (126). D-loop mutations result in a reduced mtDNA copy number or altered mitochondrial gene expression (Figure 1), thereby disrupting mitochondrial metabolism and the oxidative phosphorylation pathway. The exact role of these mutations in cancer progression is still under investigation; however, there is a consensus that mtDNA mutations are valuable cancer biomarkers (126, 129–134). Specifically, mutations within mtDNA displacement loop (D-loop) region have been identified in colorectal and gastric cancers (135) (Figure 1). This non-coding region has also been implicated in other malignancies, including lung, colon, ovarian, liver, and breast cancers (136–138). The investigation of mtDNA mutations offers a promising avenue for early cancer diagnostics, as these mutations can serve as potential biomarkers (139). Human mtDNA consists of a 16,569-base pair circular DNA encodes 13 polypeptides essential for the OXPHOS system, along with 12S and 16S rRNA and 22 tRNAs crucial for mitochondrial protein synthesis. D-loop, a critical noncoding region, regulates the replication as well as transcription of mtDNA and contains numerous common polymorphisms, especially within its highly variable segments (140). Spanning 1,124 base pairs (nucleotides 16,024 to 576), D-loop includes hypervariable regions HV1 (16,024-16,383) and HV2 (57-372), which serve as promoters for both the heavy (guanine-rich) and light (cytosine-rich) strands of mtDNA. These regions are particularly prone to mutations in various cancers (136).

Akouchekian et al. (126) analyzed mutation rate within the D-loop in CRC by sequencing mitochondrial control region in 40 CRC patient samples (20 females and 20 males) and comparing them with 150 normal control samples (79 females and 71 males). According to study, a high degree of polymorphism in mtDNA among individuals, with CRC patients exhibiting a higher frequency of variations in the D-loop region compared to controls was evident. Thirteen novel polymorphisms, not previously cataloged in the mitochondrial database (Mitomap), were identified, suggesting a potential link between these mtDNA variations and nuclear DNA mutations in CRC (126). Previous research has demonstrated a link between mtDNA mutations and several cancer incidences in human beings (135, 141, 142).

Mitochondria modulate OXPHOS pathway, which comprises complexes I to V. mtDNA possess seven subunits of complex I such as ND1, ND2, ND3, ND4L, ND4, ND5, and ND6; it also composed of a single subunit of complex III viz., cytochrome b (CYTB), as well as three subunits of complex IV viz., cytochrome c oxidase (CO) I, II, and III (118). Complex I is large and mediate proton-pumping NADH oxidoreductase activity, transferring electrons from NADH to ubiquinone (143, 144). The structure of Complex I includes a peripheral arm that extends perpendicularly to the membrane arm. This peripheral arm is divided into two main sections: distal PD module, composed of ND5 and ND4 subunits, and the proximal PP module, which includes the ND2, ND4L, ND6, ND3, and ND1 subunits to facilitate electron transfer pathway. ND1 subunit, confined to the proximal end, acts as a docking site for the Q module. Complex I is essential for the respiratory chain, playing a critical role in maintaining the NAD+/NADH ratio, regulating ROS levels, generating the mitochondrial membrane potential, as well as producing ATP. Malfunctions in Complex I are often linked to various mitochondrial diseases (118, 145–147).

Mutations in ND genes have profound effects on the malignancy of cancer cells, particularly in invasion and metastasis (146). Research has revealed that ROS-generating mtDNA mutations in ND6, such as G13997A and 13885insC cause higher metastasis in cancers (115, 117). This marked the initial discovery of pathogenic mutations associated with ND genes that driving metastasis (115, 117). Subsequent studies have validated these findings; for instance, the ND5 G13289A mutation has been shown to increase ROS production, invasion in human lung cancer cells (148). Similarly, studies in xenograft models, ND3 G10398A mutation has been observed to increase invasion as well as metastasis in human breast cancer cells, while ND6 missense as well as nonsense mutations exhibit similar effects in vitro (119, 149). Furthermore, ND6 gene mutations have been linked to lymph node metastasis in lung adenocarcinoma patients (119).

While some research suggests that increased complex I activity might reduce breast cancer metastasis (150), the connection between decreased complex I activity and metastasis remains ambiguous. The study of ND gene mutations is complicated by their random occurrence in cancer cells and the varied impacts on complex I activity. This study (118) aims to predict the pathogenicity of ND gene to explore their correlation to distant metastasis of CRC cancers (118).

A previous study (118), described that nonsynonymous single-nucleotide variants (SNVs) and SNPs in ND genes of NSCLC and colon cancer. Candidates likely to reduce complex I activity were selected based on Grantham value, evolutionary conservation, as well as protein structure and indicated a significant association between these SNVs and SNPs with distant metastasis (118).

ND6 13885insC mutation, in particular, has been shown to foster metastasis in low-metastatic cells. P29mtB82M cells with this mutation exhibit lower complex I activity, higher ROS production, and greater lung-colonizing ability compared to P29mtP29 cells with wild-type mtDNA (115, 118). P29mtB82M cells possess higher spontaneous metastatic potential, forming more metastatic foci. In this scenario, upregulation of metastasis-related genes such as Mmp11, Plaur, Ccl7, c-Myc, K-ras, Cd44, and VEGF-A (118).

The mutation induced a shift towards enhanced aerobic glycolysis (146), upregulating genes encoding for proteins include Glut1, hexokinase 1, phosphoglycerate kinase 1, and phosphofructokinase 1, while suppressing pyruvate dehydrogenase kinase 1. HIF-1a levels were upregulated which contributing to resistance against severe hypoxia. These changes suggest that the ND6 13885insC mutation enhances metastasis by stimulating metastasis-related genes, as well as metabolic reprogramming, tumor angiogenesis (118, 151, 152). Elevated expression of genes related to metastasis include Mmp11, Plaur, Ccl7, Kras, Myc, CD44, and VEGF-A. MMP11 and Plaur play roles in degrading the extracellular matrix (153, 154), while Ccl7 recruits tumor-associated macrophages, enhancing malignancy (155–157). Kras and Myc contribute to increased malignancy (158–160), whereas CD44 denotes a marker for cancer stem cells associated with metastasis (161). VEGF expression was higher in P29mtB82M cells, promoting tumor angiogenesis. Interestingly, upregulation in metastasis suppressor gene Mtss1 was evident whereas other metastasis enhancer genes Pnn, Lpar6, and Fxdy5 exhibited low expression, yet metastasis ability is more. Enhanced glycolysis and downregulation of PDK1 suggested increased acetyl-CoA generation. Increased HIF-1α expression in P29mtB82M cells likely led to upregulation of VEGF-A and glycolytic enzyme genes, contributing to hypoxia resistance and metastasis. These phenotypic changes collectively result in higher distant metastasis in P29mtB82M cells (118, 162).

As we discussed above, mtDNA related mutations occur randomly, leading to variability in each cancer cell. Despite this randomness, pathogenic missense as well as nonsense mutations in ND genes found to be crucial for distant metastasis (115, 117, 118, 149, 150). The study report by Nobuko Koshikawa et al. (118) sequenced genes such as ND1, ND2, ND3, ND4L, ND4, ND5, and ND6 in tissues from 45 primary NSCLC tumors and 37 brain metastases, as well as 22 primary colon cancer tumors and 11 distant metastases. They identified 51 somatic mutations which include a total of 22 nonsynonymous and 29 synonymous type, with a higher mutation frequency in ND6 compared to other ND genes. These mutations appeared as overlapping peaks (heteroplasmy) or single peaks (homoplasmy) on electropherograms, with homoplasmy being less prevalent in metastatic lesions (118).

Mitochondria, pivotal for cellular metabolites and energy, frequently exhibit varied dysfunctions in cancers, including CRC. Long-established Warburg effect characterizes cancer cells, emphasizing glycolysis and oxidative metabolism dysregulation, yet CRC uniquely relies on mitochondrial OXPHOS as its primary energy source. In addition, extent of mitochondria in CRC tissues surpasses that in normal colon mucosa, underscoring mitochondria’s critical, albeit unclear, role in CRC progression (168, 169).

Mitochondria possess their genome encoding 13 polypeptides crucial for electron transport chain and ATP synthase. Variations in mtDNA copy numbers correlate closely with various cancers: decreased in gastric, breast, hepatocellular, NSCLC, and renal cell carcinoma, yet increased in CRC (170–174). Recent studies implicate mtDNA depletion via TFAM mutation in fostering tumor progression as well as cisplatin resistance in microsatellite instability (MSI) CRC, with implications yet to be fully explored in microsatellite stable CRC (175) (Table 4). This study systematically explores how altered mtDNA copy numbers functionally affect MSS CRC progression, demonstrating that increased mtDNA promotes cell survival and metastasis via enhanced mitochondrial OXPHOS, suggesting novel therapeutic targets (30, 176–180).

OXPHOS predominantly generates cellular energy, with mtDNA encoding ETC components crucial for its function (187). CRC studies reveal increased mtDNA copy numbers, notably in early stages, implicating mtDNA in CRC initiation (181–183). A few other reports demonstrating that elevated mtDNA promotes MSS CRC cell survival, proliferation, and metastasis in vitro and in vivo (184–186). This contrasts with MSI CRC, where lower mtDNA levels correlate with increased glycolysis and chemoresistance (175). Differential mtDNA content may underpin these metabolic differences, necessitating further mechanistic studies (188) (30).

Mitochondria’s modest 13-gene genome contrasts with over 2000 proteins influencing diverse functions, including miRNAs originating from mitochondria (mitomiRs), regulating nuclear mRNA translation and cell phenotype (189–194). Xiacheng Sun et al. (30) described decreased mitochondrial COXIV-1 in CRC adenomas, crucial for mitochondrial-encoded complex IV and V regulation, affecting oxidative phosphorylation and ATP production (195, 196). Notably, miR-210 targets COX10, linking mitomiRs to CRC pathogenesis (197). Understanding these pathways aids in clarifying colorectal adenomatous polyps’ clinicopathological characteristics and early detection strategies (198). For instance, CRC pathogenesis involves mutations in tumor suppressor (e.g., P53, APC) and oncogenes (e.g., KRAS), regulated post-transcriptionally by miRNAs, influencing diverse cancer pathways (197, 199–204). MitomiRs, such as miR-21, miR-210, are implicated in ROS regulation, critical in CRC due to mitochondrial ROS production during oxidative phosphorylation (205–207). Mitochondrial gene expression changes during adenoma-carcinoma progression, with age-related accumulation of dysfunctional mitochondria contributing to CRC pathogenesis (206, 207). Studying mitomiRs (e.g., miR-24, miR-181, miR-210, miR-21, miR-378) across colorectal adenomatous polyps reveals varied expression patterns correlating with tumor architecture and progression, suggesting their regulatory roles in mitochondrial functional pathways (201, 204, 208–211).

In conclusion, mitoepigenetic studies demonstrate mitochondrial dynamics’ pivotal role in CRC evolution from adenomatous polyps to adenocarcinomas, urging further investigation into these intricate pathways for therapeutic and diagnostic advancements.

The impact of demethylation of D-loop region of mtDNA on mtDNA copy number, cell cycle progression, apoptosis, and cell proliferation in CRC remains uncertain (212). For instance, 5-AZA acts by irreversibly inhibiting DNA methyltransferases once incorporated into DNA, a mechanism predominantly utilized in treating hematologic malignancies and potentially applicable to other cancer types, including CRC (213). Numerous studies have indicated that 5-AZA can lead to reduced cell viability and a higher apoptotic rate in different CRC cell lines (214–219). Variations in results across studies might be attributed to differences in incubation periods and 5-AZA concentrations. For instance, Mossman et al. observed cell death in SW480 cells after a 72-hour incubation with 15 µM 5-AZA, whereas HCT116 cells did not exhibit cell death under the same conditions (218). Furthermore, zebularine, a similar DNA methyltransferase inhibitor, was found to stimulate Colo-205 cell growth at concentrations above 45 µM (218). Consequently, relatively minimal concentrations of 5-AZA (up to 10 µM) used for 24 hours in this study which has not induced a strong inhibition on CRC cells (219). Treatment of Colo-205 and Lovo colorectal cancer cells with 5 µM 5-AZA revealed notable alterations in mitochondrial and cell cycle dynamics. In Colo-205 cells, increased cell viability, delayed G0/G1 phase progression, minimal apoptosis, and elevated mitochondrial DNA (mtDNA) copy numbers were observed, while Lovo cells exhibited enhanced proliferation and mtDNA content following similar treatment. These findings suggest that elevated mtDNA levels may drive metabolic adaptation, providing the additional energy required for accelerated proliferation. Consistent with previous studies showing that mtDNA depletion impairs growth in breast and glioblastoma cells (220–223), the current observations imply that mtDNA abundance is closely linked to proliferative capacity. The extended G0/G1 phase in Colo-205 cells likely reflects an increased demand for mtDNA synthesis preceding genomic DNA replication, supporting the hypothesis that mitochondrial biogenesis and replication are tightly coordinated with cell cycle progression. Differences between Colo-205 and Lovo responses further highlight cell line-specific regulatory mechanisms governing mito-nuclear crosstalk during chemotherapeutic stress (220–223).

While mtDNA variation and demethylating agent studies in CRC offer intriguing mechanistic insights, it is critical to frame these findings accurately and avoid over-interpretation. For example, although a large Chinese cohort found that mtDNA haplogroup M7 was associated with worse prognosis in CRC in a northern Chinese population, this remains a population-specific finding and does not establish mtDNA haplogroups as reliable predictive or causative biomarkers across global CRC cohorts (212, 224). Likewise, studies of ND-gene mutations (such as ND6 13885insC) and OXPHOS upregulation in model systems demonstrate potential functional effects but lack evidence in large human metastatic CRC datasets to support the claim that these mutations drive metastasis rather than being passenger or adaptive events. In relation to epigenetic therapy, while 5-azacytidine (5-AZA) has been shown to alter mtDNA D-loop methylation and copy number in certain CRC cell lines (e.g., Colo-205) induced by treatment, this effect is highly cell line-specific and does not yet translate into robust clinical data addressing chemoresistance in CRC (212, 224).

D-loop encompassing 1122 base pairs, is critical for initiating mtDNA transcription and replication (225). While the role of D-loop methylation in mtDNA function is established, its relationship with mtDNA copy number is less understood. Various factors, including TFAM, that interacts with mtDNA and promotes transcription through the formation of initiation complex, and it can influence mtDNA copy number (226). Demethylation of these sites result in the enhanced number of mtDNA copy number, consistent with previous research linking demethylated D-loop regions to higher mtDNA copy numbers in CRC (227). Demethylated CpG sites might enhance mtDNA replication by facilitating TFAM binding and mtDNA transcription initiation, although further research is necessary to confirm this hypothesis (228) (212). Overall, demethylation across specific CpG sites in D-loop promoter may result in a higher mtDNA copy number in CRC, influencing biological behaviors such as enhanced cell proliferation and modulation of cell (212).

In conclusion, the demethylation of specific CpG sites in D-loop promoter may increase the overall copy number of mtDNA in CRC, leading to increased cell proliferation, reduced apoptosis, and a delay in the G0/G1 phase. Thus, DNA methylation influence at D-loop region of mtDNA on the expression of rate-limiting enzymes, but their impact on OXPHOS in CRC remains unclear. Thus, the quantitative changes in ND2 expression as well as methylation across D-loop at the time of CRC progression, along with potential correlations to clinicopathological stages (106). In a study by Shi Feng et al, tumor and noncancerous tissues were subjected to surgical resection from 44 patients diagnosed with CRC. Authors evaluated Cox IV and ND2 expressions in all the specimens obtained from these patients. Correlating these findings with clinicopathological data revealed an association between changes in ND2 expression and clinicopathological stage of CRC (106). The increase in ND2 expression was evident as early as stage I and continued to rise through stages I to IV. Additionally, the proportion of unmethylated D-loop enhanced in tumor as well as non-cancerous tissues, paralleling the rise in ND2 expression (106) (Figure 1). Results indicated a higher ND2 expression in tumor tissues than non-cancerous tissues. D-loop region was methylated in 79.5% of non-cancerous tissues, while this percentage dropped to 11.4% in tumor tissues (106). This demethylation likely enhances mitochondrial function, contributing to the metabolic reprogramming observed in cancer cells (106).

Changes in mtDNA copy numbers are recognized as a crucial hallmark of cancers but the quantitative changes in mtDNA should be explored to vividly examine the initiation or progression of CRC (182). Shi Feng et al., 2011 (182) investigated quantitative alterations in mtDNA copy number during CRC initiation and progression and explores potential correlations with clinicopathological stages. Authors in this study surgically resected both tumor tissues as well as noncancerous tissues from 24 colon cancer patients and 20 rectal cancer patients respectively. mtDNA copy numbers were ascertained and the results of this study described the significant raise in mtDNA copy numbers in the CRC tissues when compared to noncancerous tissues (182). Furthermore, correlation with clinicopathological data revealed that changes in mtDNA copy number were associated with clinicopathological stage of CRC, with a marked increase in stages I and II (182). No significant association with gender was observed. Increased mtDNA content could enhance cellular energy production and biosynthetic capacity, supporting rapid cell proliferation and tumor growth. These findings suggest the significance of mtDNA copy number for the initiation as well as progression of CRC especially in the early stages (182).

The role of D-loop hypomethylation in regulating mtDNA copy number as well as ND2 expression in CRC remains unclear (227). This study investigates the association between D-loop methylation status, mtDNA copy number, and ND2 expression in 65 CRC tissue samples and the surrounding non-cancerous tissues. Additionally, a demethylation experiment conducted on Caco-2 CRC cell line using 5-Aza (227). Results of this study (227) described that typical decline in methylation across D loop in CRC tissues when compared to non-cancerous tissues was evident which characterized by the decline in D-loop methylation in clinicopathological stages III and IV than the stages I and II (227). According to this report, the demethylation of D-loop has resulted in the higher mtDNA copy number and ND2 levels. In addition, 5-AZA treatment increased mtDNA copy number as well as ND2 expression in Caco-2 cells (227). Future reports should describe the function of D-loop demethylation in CRC by exploring how D-loop demethylation influences mtDNA replication and ND2 gene expression (227). It is crucial to assess the efficacy of targeting D-loop methylation with demethylating agents or other epigenetic modulators in chemoresistant CRC treatment. By advancing the exploration of the epigenetic regulation of mtDNA in colorectal cancer, these studies could explore the novel epigenetic-based therapeutic strategies to ameliorate CRC or chemoresistant CRC (227).

It is crucial to ascertain the impact of demethylation in D-loop of mtDNA on mtDNA copy number and subsequent effect on the CRC cell proliferation, cell cycle. A previous study by Huan Tong et al. (212) described the higher mtDNA copy number in Colo-205 and Lovo cells upon 5-AZA treatment and the rate of cell cycle and apoptosis is higher upon this treatment. In these cell lines, enhanced methylation was evident at 4^th and 6^th/7^th CpG regions of D-loop which was mitigated upon 5-AZA treatment. However, the cell cycle and mtDNA copy number has not changed upon 5-AZA treatment in the CRC cell lines include HCT116, SW480, LS-174T, and HT-29 cells (212). Furthermore, authors have not observed any alterations in demethylation at 4^th and 6^th/7^th CpG regions of D-loop in these CRC cell lines upon 5-AZA treatment (212). Further research should explore the molecular mechanisms by which D-loop demethylation influences mtDNA replication and gene expression, including the role of transcription factors like TFAM in this process. It is crucial to explore the interplay between nuclear and mitochondrial epigenetic regulation and its impact on cancer metabolism, progression, and resistance to therapy in CRC (212). For instance, TFAM can regulate both mtDNA transcription and replication. Elevated TFAM expression has been observed in drug-resistant hepatocellular carcinoma cells, and its inhibition has been shown to restore the chemosensitivity of these resistant cells, suggesting a potential therapeutic target for overcoming chemoresistance (229, 230). In vivo model of Kras-driven lung cancer, deletion of the TFAM gene results in compromised mitochondrial function, leading to a reduced incidence of tumors (231). In CRC cells, there is an increase in mitochondrial divalent uptake that triggers the activation of phosphodiesterase 2, which in turn inhibits mitochondrial protein kinase A. This inhibition stabilizes the accumulation of TFAM within the mitochondria, fostering cell proliferation (232). Similarly, enhanced mitochondrial calcium uptake has been linked to upregulated TFAM expression, which fostering mitochondrial biogenesis and increases the production of mitochondrial ROS. This cascade activates the NF-κB signaling, thereby accelerating the progression of CRC (233).

Mutations in TFAM have been implicated in promoting increased cell proliferation and enhanced tumorigenic potential in xenograft models. Notably, silencing TFAM in CRC cells induces metabolic reprogramming (229, 230). This silencing disrupts the Wnt/β-catenin signaling pathway via an increase in α-ketoglutarate levels, ultimately inhibiting tumor initiation and progression (234, 235). MTERFs (mitochondrial transcription termination factors) comprise a protein family that, despite sharing structural homology, perform diverse functions crucial to mitochondrial homeostasis. Disruption of MTERF activity has been shown to impair mitochondrial function, leading to mitochondrial damage and contributing to the pathogenesis of various mitochondrial-related diseases (236–238). The precise relationship between MTERF proteins and processes such as OXPHOS, cell proliferation, and tumorigenesis remains incompletely demonstrated in proliferating CRC cells. However, research has indicated that MTERF1 can modulate mitochondrial gene expression and OXPHOS levels. In HeLa cells, overexpression of MTERF1 has been shown to enhance mitochondrial gene transcription, increase OXPHOS activity, and elevate cyclin D1 levels, which promotes cell proliferation. In contrast, MTERF1 knockdown leads to diminished ATP generation, lower cyclin D1 expression, as well as cell cycle arrest (239). Upregulated expression of MTERF1 in CRC cells could cause higher cell division and enhance migration and invasion of cancer cells to form tumors. Mechanistically, MTERF1 could control AMPK/mTOR pathway, which influences mtDNA replication, transcription, subsequently modulate protein synthesis (229). Additionally, MTERF1 overexpression reduces ROS production, further enhancing mitochondrial activity for OXPHOS and contributing to cancer progression (229, 240) (Figure 2). Moreover, the inhibition of TFAM expression facilitates the release of mtDNA into cytoplasm, and activates signaling pathways to modulate oncogenesis and it is crucial to develop novel small-molecule compounds to target mitochondrial RNA polymerase (POLRMT) to block mitochondrial transcription, thereby selectively blocking OXPHOS and curbing tumor cell proliferation in chemoresistant metastatic CRCs. These findings prompt questions regarding the role of oncogenes in tumorigenesis in terms of association with mtDNA mutations or SNPs and whether the combined effects of mtDNA as well as nuclear DNA quantitative trait loci might explain the variability in cancer susceptibility and metastatic potential across different individuals. mtDNA mutations exhibit variability in both frequency and location among various cancers, with prostate, stomach, and colorectal cancers showing the greater mutation rates. Given this variability, it is crucial to investigate the potential of targeting D-loop methylation through the use of demethylating agents or other epigenetic modulators as a therapeutic strategy, particularly in chemoresistant colorectal cancer. By deepening our understanding of the epigenetic regulation of mtDNA in CRC, these studies pave the way for the development of novel epigenetic-based therapies to combat CRC, including its chemoresistant forms.

MTERF3 has been recognized as an oncogene across multiple cancer types, with gene amplification and elevated expression of MTERF3 levels strongly correlating with poor overall survival rates in cancer patients (241). Its overexpression has been shown to promote tumorigenesis both in vitro and in vivo, and enhancing proportion of cells in S phase of the cell cycle, thereby accelerating proliferation (242). In CRC specifically, MTERF3 has been implicated in the upregulation of pro-inflammatory cytokines such as IL-6 as well as IL-11, that not only modulate tumor growth but also enhance resistance to radiotherapy (243). These findings described the oncogenic role of MTERF3 in promoting both tumor progression and therapeutic resistance (229). However, in-depth exploration is crucial to uncover the intricate molecular signaling by which TFAM and MTERF family members regulate mitochondrial function and influence metastasis and chemoresistance. Specifically, studies should focus on understanding how these proteins modulate mitochondrial biogenesis, energy production, and ROS signaling in different chemoresistant cancers including CRC (229). Given their pivotal role in mitochondrial regulation and cancer progression, TFAM and MTERFs hold promise as therapeutic targets. Preclinical studies exploring inhibitors or modulators of these proteins may yield novel treatment options for chemoresistant CRC (229). Furthermore, combining mitochondrial-targeted therapies with conventional cancer treatments like chemotherapy or radiotherapy could improve efficacy and overcome resistance. For example, TFAM or MTERF inhibitors could be used alongside traditional treatments to synergistically halt tumor growth and sensitize cells to therapeutic interventions. Future studies must focus on deciphering how mtDNA single nucleotide polymorphisms and interactions with nuclear DNA quantitative trait loci influence cancer susceptibility and metastatic behavior. Furthermore, targeting epigenetic modifications within the mitochondrial genome especially through D-loop demethylation or the application of epigenetic modulators represents a promising therapeutic strategy to reprogram mitochondrial function and overcome chemoresistance. Collectively, these insights reinforce the emerging paradigm that mitochondrial epigenetic regulation serves as a pivotal determinant of CRC progression and therapeutic response, providing a conceptual framework for developing next-generation mitochondrial-targeted and epigenetic-based therapies.

Immunotherapy has emerged as a transformative approach in mCRC, yet its clinical efficacy remains largely confined to a limited subset of microsatellite instability-high (MSI-H) tumors. The majority of microsatellite-stable (MSS) mCRC cases continue to rely on combinations of chemotherapy with targeted pharmacotherapies such as anti-VEGF (e.g., bevacizumab) and anti-EGFR (e.g., cetuximab, panitumumab) drugs (33). Despite their established therapeutic benefit, both modalities are hampered by intrinsic and acquired resistance mechanisms that arise from tumor heterogeneity, dynamic clonal evolution, and metabolic adaptation.

Anti-VEGF therapy, a standard-of-care since 2004, effectively suppresses angiogenesis but lacks well-defined molecular predictors of responsiveness. Conversely, anti-EGFR therapy provides benefit only in RAS and BRAF wild-type patients, excluding approximately 60% of CRC cases due to mutation-driven intrinsic resistance (33, 244). Even in responsive subgroups, selective pressure under anti-EGFR therapy rapidly fosters secondary mutations within the EGFR extracellular domain and downstream MAPK pathway components, conferring adaptive resistance (33, 244). Longitudinal circulating tumor DNA (ctDNA) profiling now enables noninvasive detection of these emergent mutations, uncovering a dynamic interplay between drug exposure, clonal selection, and treatment relapse. Moreover, 8q chromosomal gains, frequently encompassing MYC amplification, have been correlated with resistance to EGFR blockade, supporting the rationale for combined EGFR and pan-KRAS inhibition as a next-generation therapeutic strategy (33, 244) (Table 5).

Expanding beyond receptor-targeted resistance, novel studies have implicated TP53RK as a critical regulator of replication stress tolerance. Its overexpression sensitizes CRC cells to CDC7 inhibition, suggesting a synthetic-lethal vulnerability exploitable through replication checkpoint modulation (244). Similarly, mitochondrial dynamics have been recognized as central to therapeutic resistance, wherein the PINK1–Parkin–Drp1 axis preserves mitochondrial integrity under chemotherapeutic stress. Hyperactivation of Drp1-mediated fission promotes mitophagy and survival, while its inhibition re-sensitizes CRC cells to apoptosis, positioning mitochondrial quality control as a viable metabolic checkpoint for combination therapy (244).

Recent preclinical and translational advances have further illuminated the therapeutic promise of integrating targeted inhibitors, metabolic modulators, and immunotherapies. The MEK inhibitor trametinib, when combined with 5-fluorouracil, demonstrates synergistic suppression of MAPK signaling and enhanced neoadjuvant chemoradiotherapy response. Dual blockade strategies such as cabozantinib (anti-c-MET) plus durvalumab (anti-PD-L1) have shown activity in MSS CRC by concurrently suppressing angiogenesis and reprogramming the tumor immune microenvironment. Moreover, nivolumab–regorafenib co-administration augments T-cell infiltration, particularly benefiting patients without hepatic metastases.

Emerging ferroptosis-based interventions are also redefining metabolic-immune synergy. The SLC7A11 inhibitor erastin enhances oxaliplatin efficacy through iron-dependent oxidative stress, reversing chemoresistance in preclinical CRC models (27). Complementary innovations in nanotechnology have enabled mitochondrial-targeted nanoinducers (mitoNIDs) that disrupt redox balance and potentiate the activity of CAR-T and CD8⁺ T cells, offering new hope for immunotherapy-refractory MSS CRC (27).

At the epigenetic interface, SALL1 promoter methylation and cg13755795 site hypermethylation have been identified as prognostic biomarkers distinguishing CRC from its metastatic counterpart, highlighting the intertwined influence of mitochondrial metabolism and epigenetic regulation in immune modulation (245). Furthermore, RNF4-mediated ubiquitination of PDHA1, a key enzyme bridging glycolysis and the TCA cycle, drives metabolic reprogramming, proliferation, and metastasis linking proteostasis to energy metabolism and chemoresistance (246) (Table 5).

Additional molecular mediators, including GNG2, a tumor suppressor limiting brain metastasis and FSTL3, which promotes tumor progression via HIF1α-dependent pathways, have emerged as actionable biomarkers for metastatic control (247, 248) (Table 5). Pharmacological inhibition of HIF1α using KC7F2 effectively suppresses FSTL3-driven metastasis, restoring tissue integrity in xenograft and tail-vein models. Collectively, these discoveries delineate an evolving paradigm in mCRC therapy as one that integrates immunomodulation, metabolic disruption, mitochondrial targeting, and epigenetic reprogramming. Such multimodal strategies transcend traditional cytotoxic approaches by simultaneously dismantling tumor survival networks, curbing adaptive resistance, and reactivating antitumor immunity. The convergence of multi-omic profiling, ctDNA-based precision monitoring, and nanotechnology-driven delivery systems is thus redefining the next frontier of precision immunoepigenetic therapy for chemoresistant metastatic colorectal cancer.

Epigenetic modifications mediated by histone writer enzymes orchestrate dynamic histone modifications crucial for establishing and maintaining epigenetic landscape (263). These modifications are controlled by a complex interplay among writer enzymes, reader proteins, and erasers. Dysregulation of these processes can drive pathogenesis, including in chemoresistant CRC (263) (Tables 6, 7).

Dynamic histone modifications and transcriptional regulation: Histone acetylation, initially linked with transcriptional activation, neutralizes the positive charge on histone tails, thereby modulating DNA-histone interactions and facilitating transcription factor binding and RNA polymerase activity (264). Notably, histone acetylation at H3K27 (H3K27ac) could be considered as the main process at active promoter and enhancer regions, regulating specific gene expression (265). Targeting histone acetyltransferases (HATs) like EP300 and CREBBP, which catalyze H3K27ac, has emerged as a therapeutic strategy to modulate transcriptional programs implicated in CRC progression (266, 267).

Therapeutic targeting of EP300/CBP: EP300 and CBP not only catalyze H3K27ac but also acetylate other histone and non-histone substrates, influencing proliferation and differentiation (268). Small-molecule inhibitors like C646 and A485 selectively target the catalytic activity or bromodomain function of EP300/CBP, respectively, offering potential therapeutic approaches (267, 269). Strategies that simultaneously target multiple domains of EP300/CBP have shown promise in inhibiting CRC cell proliferation synergistically (270, 271).

Exploring selective inhibition and functional roles: Despite structural homology, EP300 and CBP exhibit distinct functional roles in gene regulation and cancer pathogenesis, concluding the need for the development of selective inhibitors (271). Ongoing efforts focus on dissecting their specific contributions to CRC biology using chemical and genetic tools (272).

Expanding Targets Beyond EP300/CBP: Beyond EP300 and CBP, PCAF and GCN5 have emerged as context-dependent regulators in cancer, with inhibitors such as GSK4037 and L-Moses revealing their potential as therapeutic targets (273, 274). Therefore, development of PROTACs like GSK983/GSK699 for PCAF underscores new strategies to modulate protein function and understand their roles in cancer progression and chemoresistance (263).

Histone methylation dynamics: Histone methylation, traditionally considered irreversible, modulates gene expression through diverse mechanisms, including recruitment of TFs and chromatin remodeling complexes (263). Inhibition of histone lysine methyltransferases (PKMTs) like DOT1L and EZH2 has shown therapeutic promise in leukemia and solid tumors, respectively (275, 276).

Targeting EZH2: EZH2, a core component of the PRC2 complex catalyzing H3K27me3, regulates gene silencing in cancer. Inhibitors such as GSK126 and EPZ6438 have demonstrated efficacy in various malignancies, including CRC and glioma (277, 278). Combination therapies involving EZH2 inhibitors with other epigenetic or signaling pathway inhibitors offer synergistic effects against chemoresistance (279) but their efficacy should be examined against chemoresistant/radioresistant CRC.

PRMTs as emerging targets: Protein arginine methyltransferases (PRMTs), implicated in cancer stemness and DNA damage response, are promising targets for therapeutic intervention with inhibitors like EPZ015666 and HLCL-61 showing efficacy in lymphomas and AML (263, 280). Future directions in targeting epigenetic modifiers in chemoresistant CRC include developing selective inhibitors against less explored histone modifications (e.g., serotonylation, crotonylation) and employing innovative strategies like PROTACs to degrade oncogenic proteins (281, 282). Integrating genomic and epigenomic data will identify critical vulnerabilities in CRC, paving the way for personalized therapeutic strategies targeting epigenetic dysregulation (Tables 6, 7).

Epigenetic co-activators CBP/p300 couple chromatin state to mitochondrial biogenesis and stress signaling by directly regulating transcriptional programs that control nuclear-encoded mitochondrial genes (283). CBP/p300 acetylates and coactivates PGC-1α and other transcription factors at enhancers/promoters of mitochondrial genes, increasing mitochondrial biogenesis and OXPHOS capacity; CBP/p300 is also required for the mitochondrial unfolded protein response, linking mitochondrial proteotoxic stress to adaptive nuclear transcriptional programs. Loss or inhibition of CBP/p300 therefore impairs mitobiogenesis, lowers OXPHOS capacity and weakens adaptive mito-nuclear signaling (123, 283–285). PRMTs modulate mitochondrial metabolism through dual routes: (1) direct substrate methylation of metabolic enzymes (for example, PRMT-mediated methylation of glycolytic or TCA cycle enzymes alters their activity or stability), and (2) chromatin/histone arginine methylation that reprograms transcriptional networks controlling mitochondrial function. By changing the balance of glycolysis versus OXPHOS, PRMT activity can shift cellular reliance on mitochondria and indirectly influence ROS generation and mitochondrial quality control. Several PRMTs have been linked to altered metabolic enzyme function and tumor metabolic phenotypes in cancer models (123, 283–285). Histone methyltransferase EZH2 (PRC2 catalytic subunit) influences mitochondrial biology largely via transcriptional repression and chromatin remodeling: EZH2-mediated H3K27me3 can silence nuclear genes encoding mitochondrial regulators (including factors that modulate lipid metabolism, mitochondrial dynamics, or antioxidant responses), thereby reprogramming cellular metabolism toward or away from oxidative phosphorylation (285). In addition, EZH2 perturbation alters ER-mitochondrial contacts and organelle homeostasis in tumor cells, changing Ca²⁺ flux, bioenergetic coupling, and ROS signaling that feedback on epigenetic enzyme activity. Thus, EZH2 inhibition may derepress mitochondrial programs (or destabilize mitochondrial ER crosstalk), altering mitochondrial respiration, redox balance and susceptibility to cell death (285). PGC-1α and mito-nuclear feedback: PGC-1α functions as a nodal integrator of many of these epigenetic inputs (acetylation by CBP/p300, deacetylation by SIRT1). Changes in CBP/p300, PRMTs or EZH2 activity converge on PGC-1α-driven transcriptional networks and on downstream factors such as NRF1/TFAM that control mtDNA transcription/replication; therefore, epigenetic modulation alters mitochondrial mass, OXPHOS capacity, and ROS output, key determinants of metastatic fitness and response to therapy (123, 284, 285).

In cancer (including CRC), these mechanistic axes predict that CBP/p300 inhibition will reduce mitobiogenesis and sensitize OXPHOS-addicted tumors, PRMT inhibition may shift metabolic flux and affect mitochondrial enzyme activity, and EZH2 inhibition can rewire mitochondrial gene expression and ER–mitochondrial homeostasis, each producing distinct changes in mtDNA copy number, oxygen consumption rate (OCR), mitochondrial membrane potential, and mtROS (123, 283–285). Key experiments to validate these links include: chromatin immunoprecipitation at nuclear-encoded mitochondrial gene loci (for CBP/p300 and EZH2), mass-spec methylome and proteome profiling (to identify PRMT substrates), functional respirometry (Seahorse OCR/ECAR), mtDNA quantification, and assays of ER–mitochondrial contact (e.g., split-GFP or electron microscopy) and future studies should explore these concepts using these significant techniques (123, 283–285).

Other targeted therapies for metastatic colorectal cancer in clinical trials: According to clinicaltrials.gov. (https://clinicaltrials.gov/), the following targeted therapies that focus on genetic mutations driving tumor growth represent a key area of research for metastatic colorectal cancer (Tables 6, 7). The goal is to develop agents that inhibit the activity of abnormal proteins produced by these mutations. Encorafenib (Braftovi) targets the BRAF protein and is approved for treating patients with metastatic CRC harboring specific BRAF mutations. It is used in combination with cetuximab (Erbitux) in adults who have undergone prior treatment. Similarly, vemurafenib (Zelboraf) targets mutant B-Raf proteins and has demonstrated effectiveness in NCI-supported trials for CRC with BRAF mutations. It is administered in combination with cetuximab and irinotecan (Camptosar). The SOLARIS trial is investigating the addition of vitamin D3 to chemotherapy and bevacizumab to enhance treatment efficacy in metastatic CRC. Additionally, tucatinib (Tukysa) and trastuzumab (Herceptin) target the HER2 protein and were approved in January 2023 for advanced CRC with HER2 overexpression. In the MOUNTAINEER clinical trial, over one-third of participants experienced tumor shrinkage or disappearance, while another third achieved stable disease (Table 7).

Exploring combination therapies: Future research should focus on identifying optimal combinations of targeted therapies with conventional treatments to enhance efficacy and reduce resistance. Long-term efficacy and safety: Longitudinal studies are needed to evaluate the long-term efficacy and safety of these targeted therapies in diverse patient populations. Expansion to other mutations: Investigating the potential of targeted therapies against other genetic mutations in metastatic chemoresistant colorectal cancer could broaden the scope of treatment options available. By continuing to advance targeted therapy research, we can develop more effective treatments for metastatic chemoresistant colorectal cancer, ultimately improving patient outcomes.

The intricate relationship between mtDNA mutations and CRC metastasis describes the pivotal role of mitochondria in tumor progression. Specific mtDNA mutations can act as key drivers of CRC metastasis, potentially serving as prognostic markers and therapeutic targets for metastatic chemoresistant colorectal cancer. Variations in mitochondrial DNA copy number are closely linked to CRC progression, highlighting their potential contribution to both cancer initiation and metastasis. The exploration of mitochondrial CNVs offers insights into tumor heterogeneity and resistance mechanisms. Involvement of mitoepigenetic mechanisms, including DNA methylation and histone modifications, plays a crucial role in modulating mitochondrial gene expression in CRC. Dysregulation of these pathways contributes to both tumor growth and chemoresistance, suggesting that targeting epigenetic machinery holds promise for reversing resistance and controlling tumor progression.