Targeting mitochondrial dysfunction to intervene in liver cancer

Mitochondrial structure

Mitochondria are important organelles within cells that are characterized by a complex structure. The outer mitochondrial membrane, which covers the surface, contains numerous porins that allow the passage of small molecules. The inner membrane is folded into cristae, effectively increasing the membrane surface area and providing space for the electron transport chain (ETC) for OXPHOS and ATP synthase. The mitochondrial matrix contains a variety of enzymes, proteins, mitochondrial DNA (mtDNA), and other components involved in metabolic processes.

Mitochondria serve as the cell power source and is primarily responsible for generating ATP through the process of OXPHOS, thereby supplying energy to the cell¹⁰. Additionally, mitochondria are involved in a multitude of metabolic processes, including lipid metabolism, Ca²⁺ regulation, and amino acid metabolism¹¹.

The key aspects of mitochondrial energy metabolism include the tricarboxylic acid cycle [TCA cycle (also known as the citric acid cycle)], OXPHOS, lipid β-oxidation, and pyruvate oxidative decarboxylation. The TCA cycle involves the conversion of acetyl-CoA into carbon dioxide and energy carriers, such as NADH and FADH₂, by enzymatic catalysis. Subsequently, these energy carriers enter the ETC¹². The ETC consists of a series of protein complexes (complexes I–IV). These complexes facilitate a series of redox reactions that transfer electrons from NADH and FADH₂ to oxygen during OXPHOS, ultimately reducing oxygen to water¹³. The formation of the proton-motive force (PMF) is a critical step in OXPHOS. Protons (H⁺) are pumped from the mitochondrial matrix into the intermembrane space, generating the PMF, which in turn provides the energy required for ATP synthase to catalyze ATP production. During this process, the establishment of the PMF is a pivotal step in OXPHOS. H⁺ are translocated from the mitochondrial matrix to the intermembrane space, resulting in generation of the PMF. This PMF drives the conversion of adenosine diphosphate (ADP) and inorganic phosphate (Pi) into ATP by ATP synthase¹⁴. The β-oxidation of lipids is a key process in fatty acid metabolism. Fatty acids first combine with ATP and CoA in the cytoplasm to form acyl-CoA, which then enters the mitochondria. On the inner mitochondrial membrane, the acyl group is transferred to carnitine (a key substance for fatty acids to enter the mitochondria) in a reaction catalyzed by carnitine-acyltransferase I, forming acylcarnitine. Acylcarnitine then enters the mitochondrial matrix through a transferase and the acyl group is re-transferred to CoA in a reaction catalyzed by carnitine-acyltransferase II to generate acyl-CoA. Energy is produced through the β-oxidation cycle of “dehydrogenation-hydration-dehydrogenation-thiolysis” in the mitochondrial matrix.

Mitochondrial bioenergetics also involve various regulatory mechanisms, such as fusion and fission, which influence cellular energy metabolism by altering mitochondrial mass¹⁵. Furthermore, multiple signaling molecules, such as AMP-activated protein kinase (AMPK) and mechanistic target of rapamycin (mTOR), regulate mitochondrial function to maintain normal cellular growth¹⁶.

Cell cycle and cell death

Mitochondria have a pivotal role in cell cycle progression and cell death. Regulation of the cell cycle is intimately connected to mitochondrial functionality, which influences cell cycle’ advancement by modulating energy metabolism and the redox balance¹⁷. Mitochondria are pivotal organelles in the processes of programmed cell death and necrosis. For example, changes in mitochondrial outer membrane permeabilization (MOMP) are crucial steps in initiating apoptosis. Activation of BAX and BAK proteins leads to the release of Cyt c from mitochondria, which subsequently triggers activation of the caspase cascade¹⁸. Furthermore, in the context of necroptosis, activation of RIP3 and MLKL proteins leads to mitochondrial dysfunction¹⁹. Mitochondrial dynamics influence the susceptibility of cells to apoptosis and necrosis²⁰. These studies have elucidated various mechanisms involving mitochondria within cells, thereby providing a foundation for understanding cellular processes in pathologic conditions.

Multidimensional regulation of mitochondrial function

As the core hub of cellular energy metabolism, mitochondrial function is precisely coordinated through multidimensional mechanisms to maintain cellular homeostasis and regulate pathologic processes. At the energy metabolism level mitochondria convert nutrients into ATP via OXPHOS. ETC complexes I–IV drive proton transmembrane pumping to form the membrane potential (ΔΨm). HCC cells often shift to glycolytic energy supply by activating the Warburg effect, accompanied by reduced activity of the mitochondrial oxidative respiratory chain²¹. ROS, as byproducts of energy metabolism (primarily from electron leakage in complexes I and III), also serve as key signaling molecules. Low-level ROS activate hypoxia-inducible factor 1-alpha (HIF-1α) and nuclear factor kappa B (NF-κB) to promote proliferation, while excessive ROS damage mtDNA and induce apoptosis or ferroptosis. The antioxidant system [e.g., manganese superoxide dismutase (SOD2) and glutathione (GSH)] maintains balance by scavenging ROS²². Mitophagy, a core quality control mechanism, selectively degrades damaged mitochondria via the PINK1/Parkin pathway. Autophagy inhibits carcinogenesis by reducing ROS damage in early-stage cancer. Autophagy is hijacked as a survival strategy in advanced HCC, degrading mitochondria for energy supply and promoting drug resistance²³.

The dynamic balance between fission and fusion regulates mitochondrial morphology and function. For example, the fission protein, dynamin-related protein 1 (DRP1), mediates mitochondrial fragmentation to adapt to metabolic demands (e.g., the glycolytic shift in tumor cells), while fusion proteins [mitofusin (MFN)1/2 and optic atrophy 1 (OPA1)] maintain mitochondrial network connectivity to ensure OXPHOS efficiency. Imbalance can lead to mitochondrial swelling or functional heterogeneity²⁴. Ca²⁺ signaling regulates metabolism and apoptosis via the mitochondrial calcium uniporter (MCU). Physiologic Ca²⁺ concentrations activate TCA cycle enzymatic reactions, whereas overload triggers the opening of mitochondrial permeability transition pores (mPTPs), inducing Cyt c release and apoptotic cascades²⁵. Genetic mutations further exacerbate mitochondrial dysfunction. The high mutation rate of mtDNA (lacking histone protection) leads to abnormal ETC complex function (e.g., decreased complex I activity). Nuclear gene mutations, such as peroxisome proliferator-activated receptor gamma coactivator (PGC-1α) (regulating mitochondrial biogenesis) or isocitrate dehydrogenase (IDH)1/2 [affecting accumulation of the metabolite, 2-hydroxyglutarate (HG)], drive tumor progression through epigenetic or metabolic reprogramming²⁶.

Overall, mitochondria form a complex regulatory network through the interplay of energy metabolism, ROS regulation, autophagy–dynamic balance, Ca²⁺ signaling, and genetic variations. Mitochondrial dysfunction serves as a stress-adaptive mechanism and a key therapeutic target in diseases, like HCC.

The concept of liver cancer

HCC is a highly malignant tumor of hepatocellular origin and is one of the most common cancers worldwide with a very high incidence, especially in Asia and Africa²⁷. Many chronic liver diseases, such as hepatitis and cirrhosis, are closely related to the onset of HCC. Due to the high recurrence and metastasis rates of tumors with a poor prognosis, the 5-year survival rate is low²⁸.

HCC cells have significant manifestations of metabolic reprogramming, in which mitochondria have a key role. Mitochondria are not only the cellular motility source but are also involved in regulation of apoptosis, oxidative stress, and calcium homeostasis²⁹. Mitochondrial dysfunction leads to abnormal metabolic pathways in HCC, such as the Warburg effect, which promotes tumor cell proliferation, invasion, and metastasis²⁷. Therefore, targeting mitochondrial dysfunction provides a novel strategy for the treatment of HCC (Figure 3).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

The pathogenesis of liver cancer. The pathologic mechanisms of liver cancer mainly include the following six parts: (A) Genetic and epigenetic abnormalities mainly include changes, such as TP53 mutations, β-catenin up-regulation, and methylation (e.g., regarding the molecular mechanism related to histone methylation modification and gene transcription regulation, the process by which Me3 mediated by PRC2 inhibits the transcription initiation of RNAPII). These changes drive the occurrence of liver cancer by influencing gene expression. (B) Abnormal activation of the signaling pathways, such as the PI3K/AKT/mTOR pathway, inhibits tumor cell autophagy and the abnormal Hippo signaling pathway, both of which promote the formation of liver cancer. (C) Chronic liver injury and inflammatory states, such as hepatitis virus infection and hepatic fibrosis, increase the oxidative stress of hepatocytes and ultimately lead to liver cancer. (D) Metabolic reprogramming, such as the Warburg effect, emphasize how glycolysis, mitochondrial oxidative stress, and electron transport chain abnormalities promote the development of liver cancer. (E) Synergy of the TME, including inflammatory factors, such as IL-6 and TNF-α, and growth factors, such as VEGF, regulate angiogenesis and the interaction of immune cells, thus aggravating liver cancer. (F) The immune escape mechanism, such as the immune checkpoint inhibitor, PD-L1, has a crucial role in this process. CAFs, cancer-associated fibroblasts; HSCs, hepatic stellate cells; Me3, histone H3 lysine 27 methylation; PRC2, polycomb repressive complex 2; RNAPII, RNA polymerase II; TME, tumor microenvironment.

Classification of liver cancer

There are many classification methods for liver cancer. HCC can be divided into HCC, cholangiocarcinoma (CC), and mixed cancer (cHCC-CC) based on histologic features. The most common type of primary liver cancer is HCC, which accounts for 75%–85% of all liver cancers³⁰. HCC usually develops in the context of chronic liver disease or cirrhosis, in which the hepatocyte of origin is HCC. CC originates from the bile duct epithelial cells, which accounts for 10%–15% of primary liver cancer. CHCC-CC is a rare liver cancer with dual features of HCC and CC²⁷.

HCC is classified into proliferative and non-proliferative types based on the gene expression profile and mutational characteristics. The proliferation class is characterized by high cell proliferation activity and is usually associated with TP53 mutations and activation of the Wnt/β-catenin signaling pathway. The non-proliferative class is usually associated with CTNNB1 mutations and altered expression of metabolism-related genes exhibiting lower cell proliferative activity³¹.

Mitochondrial function is also different in liver cancer subtypes. For example, the proliferative liver cancer subtype may be more dependent on glycolysis (Warburg effect), while the non-proliferative subtype may be more dependent on mitochondrial OXPHOS³². Therefore, personalized mitochondria-targeted therapeutic strategies are needed for different HCC subtypes.

Clinical staging of liver cancer

Two clinical staging systems [TNM staging system and Barcelona clinical liver cancer (BCLC)], are mainly used. The TNM staging system classifies liver cancer according to tumor size (T), degree of lymph node involvement (N), and distant metastasis (M), which is a common clinical staging method. In contrast, the BCLC staging system divides liver cancer from stage 0 (very early) to D stage (late) combined with the tumor characteristics, liver function status, and overall patient status. This system provides the basis for treatment decisions.

Etiology of liver cancer

Chronic viral hepatitis, liver cirrhosis, alcoholic liver disease, and non-alcoholic fatty liver disease (NAFLD) are the main causes of liver cancer. Hepatitis A virus (HAV), hepatitis B virus (HBV), excessive alcohol consumption, cigarette smoking, obesity, aflatoxin exposure, diabetes, and genetic factors can also lead to liver cancer.

In modern society increased stress and unhealthy lifestyle habits, such as cigarette smoking, alcohol consumption, and poor diet, have further elevated the risk of liver disease³³. A healthy lifestyle, such as a balanced diet and regular sleep patterns, is the common challenge that society needs to face. Mitochondrial dysfunction has an important role in the development of liver cancer. For example, infections with HBV and hepatitis C virus (HCV) can cause mtDNA damage and functional disorders³⁴. Alcohol metabolites can directly affect the mitochondrial membrane potential and respiratory chain function³⁵. NAFLD-related liver cancer is often accompanied by mitochondrial fatty acid oxidation (FAO) disorders³⁶.

Diagnosis and treatment of liver cancer

Currently, the diagnosis of liver cancer mainly involves imaging examinations (ultrasound, CT, and MRI), serum biomarkers [alpha-fetoprotein (AFP) and des-gamma-carboxy prothrombin (DCP)], and histologic features. Metabolites and gene expression profiles related to mitochondrial dysfunction may emerge as novel biomarkers for liver cancer diagnosis with increasing research on mitochondrial dysfunction. For example, some mitochondrial metabolic intermediates, such as succinate and fumarate, are significantly elevated in the serum of liver cancer patients and could serve as potential diagnostic biomarkers²⁷. In addition, imaging techniques based on mitochondrial functional abnormalities, such as metabolic imaging, also hold promise for improving the diagnostic accuracy of liver cancer³⁷.

The traditional treatment methods for liver cancer mainly include surgical therapy (liver transplantation and hepatectomy), local therapy [radiofrequency, microwave ablation, and transarterial chemoembolization (TACE)], and systemic therapy (sorafenib, lenvatinib, and PD-1/PD-L1 inhibitors). However existing treatment methods have not fully focused on the impact of mitochondrial function on liver cancer and have not received sufficient attention²⁷. For example, TACE may lead to increased mitochondrial oxidative stress in tumor cells, thereby inducing drug resistance. Targeted therapeutic agents may exert anti-tumor effects by influencing mitochondrial metabolic pathways³⁸. Therefore, combining mitochondrial targeting strategies with existing therapies is expected to improve therapeutic efficacy and overcome drug resistance.

In addition, the degree and type of mitochondrial dysfunction may impact the prognosis of liver cancer. For example, the extent of mitochondrial metabolic reprogramming may be associated with the aggressiveness and metastatic potential of the tumor³⁹. The activation status of the mitochondrial-dependent apoptotic pathway may influence tumor sensitivity to treatment⁴⁰. Therefore, assessing mitochondrial functional status is expected to be a new indicator to predict HCC prognosis and guide individualized treatment.

Future directions for liver cancer research

Future research should focus on precision medicine, novel therapeutic strategies, and translational research with the goal of achieving personalized treatment and long-term survival for patients with liver cancer. Precision medicine is a type of multi-omics research based on genomics, transcriptomics, and metabolomics and can provide a basis for precise classification and treatment of liver cancer. Precision medicine can combine the molecular characteristics of different liver cancer subtypes and the functional status of mitochondria and develop a personalized treatment plan for patients⁴¹. Second, new mitochondrial targeted drugs can be developed, such as mitochondrial reactive oxygen inhibitors, and agents targeting mitochondrial dynamics. Third, mitochondrial targeted therapy can be combined with the existing treatment methods to improve the efficacy of liver cancer treatment⁴².

In addition, translational studies should be conducted to establish preclinical liver cancer models closer to clinical practice, such as patient-derived xenograft (PDX) models and large-scale clinical trials should be conducted to evaluate the safety and efficacy of mitochondrial-targeted therapy⁴³.

Because mitochondria have an important role in hepatocytes, mitochondria are currently a hot topic of interest among scholars that are expected to occupy a more important position in future research involving liver cancer and make breakthrough progress in the treatment of liver cancer.

HCC is a highly heterogeneous malignancy, the occurrence and progression of which are closely related to mitochondrial dysfunction. By gaining a comprehensive understanding of the concepts, classification, etiology, diagnosis, treatment, and prognosis of liver cancer, we can better recognize the role of mitochondrial dysfunction in liver cancer. This knowledge also provides a theoretical basis for developing novel targeted therapeutic strategies.

Mitochondrial metabolic reprogramming and redox imbalance

Mitochondrial dysfunction drives tumor progression in HCC through a synergistic mechanism of metabolic network rewiring and redox homeostasis imbalance, specifically manifested as energy metabolic pattern switching, abnormal membrane potential, and disordered ROS regulatory networks.

Mitochondrial oxidative respiratory chain ETC dysfunction serves as the starting point of metabolic reprogramming in HCC. The activity of respiratory chain complexes I (NADH dehydrogenase) and III (cytochrome bc1 complex) is significantly reduced in HCC cells, leading to decreased electron transport efficiency and reduced ATP production⁴⁴. To compensate for energy crisis, tumor cells shift to glycolytic dependence (Warburg effect), which not only rapidly generates ATP through glucose metabolism but also provides precursors for the biosynthesis of nucleotides, lipids, amino acids, and biogenic amines to maintain the malignant proliferative phenotype⁴⁵. This metabolic switch is accompanied by abnormal ΔΨm. Indeed, ΔΨm is driven by proton gradients and participates in ATP synthesis and apoptosis regulation in normal cells, whereas ΔΨm is often abnormally elevated in HCC cells. This elevation blocks mPTP opening, inhibits Cyt c release, and suppresses mitochondria-dependent apoptosis⁴⁶. High ΔΨm further enhances the survival advantage of tumor cells by disrupting calcium homeostasis and metabolite transport (e.g., ADP/ATP exchange; Figure 4). Mitochondria serve as the main source of ROS (accounting for ~80% of total intracellular ROS) with ETC complexes I and III being the primary sites for superoxide anion (O₂⁻) generation. Specifically, ~1%–2% of electrons leaking from the ETC combine with oxygen to form O₂⁻, which is then converted to hydrogen peroxide (H₂O₂) by superoxide dismutase (SOD1/SOD2), and finally scavenged by catalase (CAT) and glutathione peroxidase (GPX)⁴⁷. This balance is disrupted in HCC. For example, in chronic liver diseases (e.g., hepatitis and fatty liver), pathogen infections, lipid accumulation, or metabolic disorders, the mitochondrial ETC is activated to overproduce ROS. In addition, mPTP opening causes calcium overload, further stimulating mitochondrial ROS generation⁴⁸. HCC cells exhibit downregulated SOD2 expression and reduced GPX4 activity, leading to impaired H₂O₂ scavenging and formation of an oxidative stress microenvironment (Figure 4). Excessive ROS drives HCC progression through multiple mechanisms, as follows: ① direct damage to mtDNA and respiratory chain proteins, exacerbating metabolic dysfunction; ② activation of NF-κB and HIF-1α signaling pathways, inducing proinflammatory cytokines (e.g., IL-6) and vascular endothelial growth factor (VEGF) to promote tumor invasion and angiogenesis⁴⁵; and ③ modification of cell membranes via lipid peroxidation, enhancing cell motility and inducing ferroptosis resistance. Notably, low-level ROS acts as a signaling molecule to promote tumor cell adaptation to metabolic stress, while high-intensity oxidative stress drives carcinogenesis by damaging genomic stability, reflecting the “dose-effect biphasicity” of ROS. Current therapeutic strategies focus on the cross-network of metabolic reprogramming and redox imbalance, as follows: In ETC-targeted intervention, complex I inhibitors (rotenone and 3-NPA) block the ETC, depleting ATP and inducing explosive ROS generation, demonstrating single-agent antitumor activity in HCC animal models⁴⁹. In membrane potential and mPTP regulation cyclosporine A (CsA) inhibits mPTP opening to alleviate oxidative stress damage, though balancing the immunosuppressive side effects is necessary⁵⁰. Dual strategies of anti-oxidation and pro-oxidation involve antioxidants, like vitamin E, which reduce oxidative damage in chronic liver diseases to delay carcinogenesis. Ferroptosis inducers targeting GPX4 (e.g., RSL3) deplete GSH to amplify mitochondrial ROS and induce tumor cell death⁵¹. Emerging technologies, such as mitochondria-targeted nanomedicines (e.g., doxorubicin-loaded triphenylphosphine-conjugated liposomes), enhance drug enrichment in tumor mitochondria, while reducing toxicity to normal tissues. Future research should further dissect key nodes in the metabolism-oxidative stress network (e.g., the SOD2/mPTP/HIF-1α axis) and screen personalized therapeutic markers via multi-omics technologies to accelerate the translation of mechanistic research to clinical applications.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

The pathologic mechanism underlying mitochondrial dysfunction in hepatocellular carcinoma cells. This figure illustrates four parallel mechanisms (A–D), where each sub-step is numbered (1–15) to correspond to the signaling events and processes in the diagram. (A) Abnormal mitophagy and mitochondrial outer membrane permeability involves steps 1–3, as follows: 1. Ubiquitin-dependent mitophagy (PINK1/Parkin pathway). When mitochondrial function is impaired (e.g., loss of membrane potential), PINK1 accumulates on the OMM and undergoes autophosphorylation, thereby activating Parkin. Activated Parkin mediates ubiquitination of OMM proteins, such as MFN2 and NIX. The ubiquitinated mitochondria are then recognized by autophagic adaptor proteins (P62, NDP52, and OPTN), which bind to ubiquitin chains via ubiquitin-binding domains and interact with LC3 to promote autophagosome formation and encapsulation. Finally, the encapsulated mitochondria fuse with lysosomes to complete degradation. 2. Ubiquitin-independent mitophagy (BNIP3/NIX and FUNDC1 pathways). Under hypoxia or ROS stress, the OMM proteins (BNIP3 and NIX) directly interact with LC3 to induce mitophagy, cooperating with the PINK1/Parkin pathway to maintain mitochondrial quality. Loss of FUNDC1 leads to the accumulation of damaged mitochondria, activates inflammasomes, and triggers the release of pro-inflammatory cytokines (e.g., IL-1β and IL-18), exacerbating the hepatic inflammatory microenvironment. 3. Abnormal MOMP. In hepatocellular carcinoma cells, there is an imbalance between pro- and anti-apoptotic protein systems. Specifically, anti-apoptotic proteins, such as Bcl-2 and Bcl-xL, are highly expressed due to gene amplification and transcriptional upregulation, while pro-apoptotic proteins, such as Bax and Bak, lose function due to promoter methylation or accelerated degradation. This process leads to a “closing tendency” of MOMP, making MOMP difficult to open normally to initiate apoptosis. Cytochrome c in the mitochondrial intermembrane space can still be released through non-classical pores mediated by Bax/Bak, activating the caspase cascade. However, hepatocellular carcinoma cells inhibit this process by upregulating anti-apoptotic mechanisms. Meanwhile, impaired MOMP results in the loss of ΔΨm, which further reduces ATP synthesis and promotes ROS accumulation, forming a vicious cycle. (B) Cascade amplification of mitochondrial oxidative stress involves steps 4–8, as follows: 4. Activation of NADPH oxidase. The NADPH oxidase complex is activated by oncogenic signals in hepatocellular carcinoma cells. The NADPH oxidase complex uses cytoplasmic NADPH as an electron donor to catalyze the generation of O₂⁻ from O₂. In addition, NADPH is converted into NADP⁺ and H⁺. O₂⁻ is further converted into H₂O₂, which diffuses into the cytoplasm and initiates oxidative stress. 5. ROS generation mediated by the respiratory chain. O₂⁻ is one of the major ROS in mitochondrial oxidative stress. Mutations in mitochondrial DNA (mtDNA) or functional defects in respiratory chain complexes (I and III) lead to electron leakage, which combines with O₂ to form O₂⁻. A decrease in ΔΨm exacerbates electron transport blockage, further promoting ROS production, accompanied by a reduction in ATP synthesis. 6. Inactivation of antioxidant systems. The enzymatic antioxidant system is the main way to scavenge ROS in cells. Cytoplasmic SOD1 and mitochondrial matrix SOD2 convert O₂⁻ into H₂O₂ in normal hepatocytes, which is then degraded into non-toxic water by CAT, GPX, and PRDX. However, the above antioxidant enzymes are downregulated due to epigenetic silencing or accelerated degradation in hepatocellular carcinoma cells, leading to ROS accumulation. 7. Vicious cycle of ONOO⁻. Excessive O₂⁻ reacts rapidly with NO to generate ONOO⁻, driving the excessive accumulation of ONOO⁻. As a highly oxidizing RNS, excessive ONOO⁻ exacerbates mitochondrial dysfunction through oxidative and nitrosative modifications of key mitochondrial molecules, such as irreversible damage to respiratory chain complexes, destruction of mitochondrial membrane structure, and damage to mtDNA. 8. Fe-S damage. O₂⁻ and other ROS directly oxidize iron-sulfur clusters, leading to the release of Fe²⁺. Free Fe²⁺ reacts with H₂O₂ via the Fenton reaction (Fe²⁺ + H₂O₂ → •OH + OH⁻) to generate highly reactive •OH, which further damage Fe-S clusters and mitochondrial structures. This process establishes a self-perpetuating cycle (“ROS accumulation → Fe-S cluster damage → Fe²⁺ release → more ROS generation”). (C) Adaptive changes in mitochondrial metabolic reprogramming involve steps 9–11, as follows: 9. Enhanced glucose transport. Glucose transport relies on the GLUTs family, with GLUT1 and GLUT2 being the main transporters. In the context of mitochondrial dysfunction, the abnormal enhancement of glucose transport primarily depends on the high expression and increased membrane localization of GLUT1. This compensatory mechanism not only represents adaptation of the cell to mitochondrial energy production defects but also serves as the metabolic basis for maintaining malignant proliferation, invasion, and metastasis. 10. Hyperactive FAO. LCFAs are activated in the cytoplasm by ACS to form acyl-CoA. Through the carnitine shuttle system (CPT1 catalyzes the combination of acyl-CoA and carnitine to generate acylcarnitine → CACT mediates trans-inner membrane transport → CPT2 regenerates acyl-CoA), acyl-CoA enters the mitochondrial matrix. Acyl-CoQ is then degraded via FAO to acetyl-CoA, which supplements the TCA cycle for energy supply. Hepatocellular carcinoma cells exhibit high expression of CPT1/2 and CACT, enhancing FAO to adapt to hypoxic environments. 11. Glutamine metabolic compensation. Cytoplasmic glutamine is converted to glutamic acid by GLS (which is highly expressed in hepatocellular carcinoma). Glutamic acid then enters mitochondria and is converted to α-KG via GDH or ALT/AST, replenishing the TCA cycle and further promoting glutamine metabolism. In addition, NH₄⁺ generated from glutamine metabolism participate in nucleotide synthesis. (D) Calcium homeostasis imbalance and abnormal apoptosis regulation involve steps 12–15, as follows: 12. Enhanced cytoplasmic calcium influx. Abnormal activation of exogenous VOCs and endogenous ROCs leads to increased calcium ions entering the cytoplasm, which further enter the mitochondrial matrix through VADC channels. 13. Endoplasmic reticulum-mitochondria calcium transfer. MAMs form high-calcium microdomains. IP3R on the endoplasmic reticulum directly binds to VDAC1 on the outer mitochondrial membrane, efficiently transferring Ca²⁺ to the mitochondrial intermembrane space, which is then transported to the matrix via MCU on the inner membrane. This process is enhanced in hepatocellular carcinoma cells due to abnormal MAMs structure. 14. Dysfunction of calcium exchangers. NCLX (sodium-calcium exchanger) on the inner mitochondrial membrane relies on the Na⁺ gradient to efflux Ca²⁺ from the matrix, while LETM1 (calcium-hydrogen exchanger) assists in Ca²⁺ efflux depending on the proton gradient. However, the Na⁺ gradient (due to Na⁺-K⁺-ATPase inhibition) and proton gradient (due to respiratory chain defects) are disrupted in hepatocellular carcinoma cells, resulting in impaired Ca²⁺ efflux and matrix calcium overload. 15. MPTP opening and apoptotic escape. Matrix calcium overload and ROS accumulation jointly activate the mPTP, leading to loss of ΔΨm, mitochondrial swelling, and release of apoptotic factors, such as cytochrome c, AIF, and SMAC/DIABLO. These factors activate the caspase cascade in normal cells, but hepatocellular carcinoma cells inhibit this process by overexpressing Bcl-2/Bcl-xL and simultaneously activate the NF-κB pathway to promote survival, achieving apoptotic escape. In conclusion, mitochondrial dysfunction in hepatocellular carcinoma cells induces ROS accumulation, ATP deficiency, and loss of ΔΨm through four major mechanisms (abnormal autophagy, oxidative stress, metabolic reprogramming, and calcium homeostasis imbalance). These changes further activate pathways, such as HIF-1α (which promotes angiogenesis) and NF-κB (which triggers the release of inflammatory factors). In addition, by inhibiting caspase activity and enhancing anti-apoptotic signals, the pathways ultimately facilitate tumor proliferation, invasion, and drug resistance. It can be concluded that the oxidative stress of the patient with liver cancer increases, the electron transport chain decreases, and some membrane potential changes will increase Cytochrome c and calcium ions, which will make the mitochondria lose normal function and aggravate the course of liver cancer. α-KG, α-ketoglutarate; ACS, acyl-CoA synthetase; AIF, apoptosis-inducing factor; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ATP, adenosine triphosphate; Bak, Bcl-2 antagonist/killer; Bax, Bcl-2-associated X protein; Bcl-2, B-cell lymphoma 2; Bcl-xL, B-cell lymphoma-extra large; BNIP3, Bcl-2/adenovirus E1B 19 kDa protein-interacting protein 3; CACT, carnitine-acylcarnitine translocase; CAT, catalase; CPT1, carnitine palmitoyltransferase 1; CPT2, carnitine palmitoyltransferase 2; ΔΨm, mitochondrial membrane potential; FAO, fatty acid oxidation; Fe²⁺, ferrous ion; Fe-S, iron-sulfur cluster; FUNDC1, FUN14 domain-containing protein 1; GDH, glutamate dehydrogenase; GLUT1, glucose transporter 1; GLUT2, glucose transporter 2; GLUTs, glucose transporters; GPX, glutathione peroxidase; GLS1, glutaminase 1; H₂O₂, hydrogen peroxide; HIF-1α, hypoxia-inducible factor 1α; •OH, hydroxyl radical; IL-1β, interleukin-1β; IL-18, interleukin-18; IP3R, inositol 1,4,5-trisphosphate receptor; LETM1, leucine zipper-EF-hand containing transmembrane protein 1; LCFAs, long-chain fatty acids; LC3, microtubule-associated protein 1 light chain 3; MAMs, mitochondria-associated endoplasmic reticulum membranes; MCU, mitochondrial calcium uniporter; MOMP, mitochondrial outer membrane permeabilization; mPTP, mitochondrial permeability transition pore; mtDNA, mitochondrial DNA; MFN2, mitofusin 2; Na+-K+-ATPase, sodium-potassium ATPase; NADPH, nicotinamide adenine dinucleotide phosphate; NF-κB, nuclear factor-kappa B; NH₄⁺, ammonium ion; NCLX, sodium-calcium exchanger; NDP52, nuclear dot protein 52 kDa; NO, nitric oxide; X, BNIP3-like protein; OMM, outer mitochondrial membrane; ONOO⁻, peroxynitrite; OPTN, optineurin; O₂⁻, superoxide anion; PINK1, PTEN-induced kinase 1; PRDX, peroxiredoxin; RNS, reactive nitrogen species; ROCs, receptor-operated calcium channels; ROS, reactive oxygen species; SOD1, superoxide dismutase 1; SOD2, superoxide dismutase 2; SMAC/DIABLO, second mitochondria-derived activator of caspases/direct IAP-binding protein with low pI; TCA, tricarboxylic acid cycle; VADC, voltage-dependent anion channel; VDAC1, voltage-dependent anion channel 1; VOCs, voltage-operated calcium channels.

Innovative therapies targeting these links are expected to break through the current therapeutic bottlenecks. In the future it will be necessary to deepen mechanistic research and accelerate clinical translation.

Apoptosis and autophagy

Mitochondria-mediated apoptosis is a “double-edged sword” for HCC cell fate. Mitochondria have a significant role in the apoptotic pathway. Mitochondria dysfunction, such as increased MOMP, can lead to the release of Cyt c into the cytosol. Cyt c binds to apoptotic protease-activating factor 1 (Apaf-1) to form the apoptosome, which activates caspase-9 and downstream effector caspases-3/7, ultimately triggering programmed cell death⁴⁰. In addition to the classic apoptotic pathway, mitochondria also release pro-apoptotic factors, such as SMAC/DIABLO, to antagonize apoptosis inhibitors, thereby reinforcing apoptotic signals. This is a canonical mechanism of mitochondrial apoptosis.

HCC cells often thwart MOMP by upregulating the expression of anti-apoptotic Bcl-2 family members, such as Bcl-2, Bcl-xL, and MCL-1, or evade apoptosis by downregulating the expression of pro-apoptotic factors, including Bax, Bak, and Bim⁵². ΔΨm disruption and the exacerbation of reactive ROS accumulation further enhance resistance to apoptosis, which has become a key mechanism for chemoresistance in HCC⁵³. BH3 mimetics, such as ABT-737 and navitoclax, induce MOMP by antagonizing anti-apoptotic members of the Bcl-2 family. Mitochondrial-targeted drugs, such as Mito-CP, selectively accumulate in mitochondria, directly triggering the release of Cyt c⁴⁰. In addition, combining radiotherapy or chemotherapeutic agents (e.g., sorafenib) synergistically enhances mitochondrial apoptosis sensitivity⁵⁴.

Under physiologic conditions, mitophagy mediated by the PINK1/Parkin pathway maintains metabolic homeostasis by clearing damaged mitochondria and prevents genomic instability caused by excessive accumulation of ROS⁵⁵. However, autophagy can be hijacked as a pro-survival mechanism in HCC. Specifically, the hypoxic microenvironment activates HIF-1α, which induces BNIP3/NIX-dependent mitophagy. This process clears dysfunctional mitochondria to maintain energy supply and supports the survival of tumor cells under stress conditions⁵⁶. When mitochondrial damage exceeds the repair threshold, persistent activation of autophagy may promote the death of HCC cells through “autosis” or crosstalk with apoptotic signaling, such as the interaction between Beclin-1 and Bcl-2)⁵⁷. For example, rapamycin analogs induce protective autophagy by inhibiting mTORC1, while combination with autophagy inhibitors, such as chloroquine, can block lysosomal degradation, leading to the accumulation of autophagosomes and triggering apoptosis⁵⁸.

Selective inhibition of pro-survival autophagy (e.g., targeting ULK1 or ATG5) can enhance chemosensitivity. Conversely, inducing excessive autophagy [e.g., using the mitochondrial uncoupler, carbonyl cyanide n-chlorophenylhydrazone (CCCP)] may overcome the tolerance threshold of HCC cells. The level of autophagy needs to be dynamically regulated in combination with tumor stage and microenvironmental characteristics⁵⁹. Apoptosis-related proteins, such as caspases cleaving ATG proteins, can inhibit autophagy, while autophagy-related proteins, such as p62/SQSTM1, indirectly regulate apoptotic signaling by degrading ubiquitinated substrates⁶⁰. Factors released by mitochondria, such as apoptosis-inducing factor, can participate in both apoptotic and non-apoptotic cell death programs. Combining apoptosis induction (e.g., using BCL-2 inhibitors) with autophagy modulation (e.g., using HDAC inhibitors) can overcome the limitations of single-pathway targeting⁶¹. For example, inhibition of protective autophagy by MCL-1 while blocking protective autophagy significantly enhanced cell death in sorafenib-resistant HCC⁵⁷.

In the future it will be necessary to focus on an analysis of spatial and temporal dynamic regulatory network, develop personalized mitochondrial targeted therapy based on patient stratification, and provide new dimensions for the treatment of liver cancer. For example, a highly specific mitochondrial-targeted delivery system was developed to reduce off-target toxicity⁶². Exploring biomarkers of mitochondrial pathway dynamics in HCC heterogeneity [e.g., mtDNA mutation and transcription factor A mitochondrial (TFAM) expression]⁶³. Deciphering the paracrine regulation of mitochondrial function by the TME (e.g., cancer-associated fibroblasts) expands potential targets for combination therapies⁶⁴.

Genic mutation

mtDNA lacks histone protection, has a weak repair mechanism, and is exposed to a high ROS environment. As a result, the mutation rate is 10–100 times higher than nuclear DNA⁶⁵. Common mutations in liver cancer include mutations in genes encoding subunits of the OXPHOS complexes, such as ND1-6, CYTB, and COX1-3, leading to dysfunction of the ETC^66,67. Mutations in the D-loop region (a key area controlling mtDNA replication and transcription) can affect the copy number of mtDNA and mitochondrial biogenesis⁶⁸. Mutations in genes within mtDNA that regulate mitochondrial function, such as SDHB, IDH1/2, fumarate hydratase (FH), and TFAM, can indirectly impair mitochondrial function through epigenetic modifications or metabolic reprogramming⁶⁹. For example, mutations in IDH1/2 lead to an increase in 2-HG, which competitively inhibits α-ketoglutarate-dependent enzymes, such as ten-eleven translocation (TET) enzymes and histone demethylases. This inhibition disrupts epigenetic regulation and mitochondrial metabolism, contributing to the pathogenesis of cancers associated with IDH1/2 mutations⁷⁰. Mutations in the subunits of the succinate dehydrogenase (SDH) complex, such as SDHB, lead to the accumulation of succinate, which inhibits prolyl hydroxylase. This inhibition stabilizes HIF-1α, thereby promoting glycolysis and angiogenesis⁷¹.

Mutations in mtDNA lead to dysfunction of ETC complexes, such as complexes I and III, reducing ATP generation and forcing HCC cells to rely on the Warburg effect (aerobic glycolysis) for energy supply. At the same time, the NADH/NAD + ratio imbalance activates the SIRT 1/AMPK pathway, promotes FAO and glutamine hydrolysis, and maintains metabolic plasticity. Dysfunction of the ETC increases electron leakage, resulting in excessive production of ROS, such as O₂⁻, which attack nuclear DNA and induce oncogenic mutations (e.g., TP53 and CTNNB1 mutations)⁷². In addition to the direct effects of mtDNA mutations on mitochondrial function, these mutations can also indirectly impact tumor immunity by activating the TLR9/STING pathway. Specifically, mtDNA mutations lead to the release of damage-associated molecular patterns (DAMPs), such as mtDNA fragments, which activate the TLR9/STING pathway. This activation creates a pro-inflammatory microenvironment that facilitates tumor immune evasion^73–75 Mitochondrial dysfunction triggers calcium signaling (e.g., via the MCU channel), metabolic reprogramming mediated by metabolites [e.g., acetyl-CoA and α-ketoglutarate (α-KG)], and ROS, leading to nuclear transcriptional reprogramming. Activation of the NF-κB/STAT3 pathway promotes cell proliferation and anti-apoptosis. Stabilization of HIF-1α enhances glycolysis and angiogenesis. Upregulation of the MYC/mTORC1 axis drives anabolic metabolism and protein translation⁷⁶.

Mutations in mtDNA lead to altered levels of TCA cycle intermediates, such as α-KG, succinate, and fumarate), which affect the activity of histone/DNA-modifying enzymes, such as histone deacetylases (HDACs) and TETs. This results in global DNA hypomethylation or local hypermethylation (e.g., silencing of tumor suppressor gene promoters)⁷⁴. The specific mutational spectrum of liver cancer includes high-frequency mutation genes, as follows: the D-loop region of mtDNA (approximately 30%–50% of liver cancer patients); NADH dehydrogenase 5 (ND5); Cyt b; and nuclear genes, such as IDH2 and SDHB⁷⁷. The mutation burden and prognosis are as follows: mutations in the D-loop region of mtDNA are positively correlated with poor prognosis in HBV-related HCC. HCC with IDH1 mutations show a poor response to immune checkpoint blockade (ICB) therapy. The content and mutations of cell-free mtDNA (cf-mtDNA) in the plasma of HCC patients (e.g., ND1 G3460A) can assist in early diagnosis and therapeutic monitoring, outperforming traditional AFP⁷⁸.

Intervention strategies targeting mitochondrial gene mutations mainly include three approaches: targeting metabolic vulnerabilities; gene-editing technologies targeting mutated mtDNA; and antioxidant and anti-inflammatory therapies.

Inhibitors targeting metabolic vulnerabilities primarily include ETC complex and IDH mutation inhibitors. Complex I inhibitors, such as metformin, selectively target HCC subpopulations dependent on OXPHOS. IDH mutation inhibitors, such as enasidenib (AG-221), target IDH2 mutations to reduce 2-HG levels and restore epigenetic stability⁷⁹. Gene-editing technologies targeting mutated mtDNA include two main approaches. The first approach is the mtZFN/TALEN technology, which specifically cleaves mutated mtDNA (e.g., ND4 G11778A) to promote preferential replication of wild-type mtDNA^80,81. The second approach is the mito-CRISPR technology, which involves delivering CRISPR-Cas9 to mitochondria using lipid nanoparticles. However, this technique is still in the experimental stage⁸². Antioxidant and anti-inflammatory therapies mainly include two types. The first type is mitochondrial-targeted antioxidants, such as MitoQ and SkQ1, which scavenge mitochondrial ROS and block mutation-driven genomic instability⁸³. The second type is STING pathway inhibitors, such as H-151), which inhibit chronic inflammation induced by the release of mtDNA⁸⁴.

In the future there will be an urgent need to address the challenge of tumor heterogeneity. HCC cells exhibit intratumor heterogeneity in mitochondrial mutations, which requires the combined use of single-cell sequencing and spatial multiomics for detailed analysis. It is also critical to develop dynamic monitoring technologies, such as highly sensitive mtDNA mutation detection platforms that combine droplet digital PCR with next-generation sequencing. To achieve precision treatment, personalized combination therapies should be designed based on mutation types. For example, combining IDH inhibitors with ICB can be a promising strategy⁴⁴.

Calcium homeostasis imbalance

The imbalance of Ca²⁺ homeostasis is closely related to the development and progression of malignant tumors. Within the cell, the endoplasmic reticulum (ER), mitochondria, lysosomes, and other organelles, as well as the plasma membrane, work together to form a Ca²⁺ regulatory network that maintains intracellular Ca²⁺ balance⁸⁵. The mitochondrial calcium uniporter (MCU), which is situated on the mitochondrial inner membrane, functions as the principal pathway for calcium ion influx into the mitochondria. The MCU is essential for maintaining mitochondrial calcium homeostasis and has a critical role in energy metabolism, reactive ROS production, and apoptosis⁸⁶. Ca²⁺ can enter mitochondria through various pathways, such as the voltage-dependent anion channel (VDAC) or the MCU, and can also be transported to mitochondria from the ER via inositol trisphosphate receptors (IP3Rs). These pathways collectively contribute to the maintenance of intracellular Ca²⁺ homeostasis⁸⁷. The mitochondrial sodium/calcium exchanger (NCLX), a protein located on the inner mitochondrial membrane, regulates mitochondrial calcium homeostasis by facilitating sodium-dependent calcium efflux from the mitochondria. This process has a crucial role in the maintenance of intracellular Ca²⁺ homeostasis and cellular viability⁸⁸.

During the progression of cancer, the concentration of Ca²⁺ and the spatiotemporal signaling patterns in the extracellular space, cytoplasm, ER, and mitochondria are frequently disrupted due to genetic mutations, dysregulated expression, imbalanced regulation, and alterations in the subcellular localization of Ca²⁺ transport proteins. Disruption of Ca²⁺ homeostasis leads to improper regulation of signaling effectors that depend on Ca²⁺. This imbalance further exacerbates the pathophysiologic features of cancer, such as promoting cell proliferation, enhancing cell survival, and increasing invasiveness⁸⁹. Thus, maintaining Ca²⁺ homeostasis is essential for suppressing the aberrant behavior of cancer cells.

Dynamics

Mitochondrial dynamics refer to the fission and fusion processes of mitochondria within the cell, which are crucial for maintaining cellular energy balance and overall cell homeostasis. Mitochondrial fission is the process by which mitochondria divide into smaller, ring-like structures within the cell, typically mediated by mitochondrial dynamin-related proteins, such as DRP1. In contrast, the process of mitochondrial fusion involves the merging of two or more mitochondria into a larger structure, a process that is facilitated by mitochondrial fusion proteins, such as MFN2 and OPA1. Imbalances in these processes can lead to mitochondrial morphologic abnormalities, dysfunction, and apoptosis, thereby impacting liver health⁹⁰.

Changes in mitochondrial dynamics are closely associated with the development of metabolic diseases in the liver, such as NAFLD. For example, mitochondrial dynamics become imbalanced due to various factors in NAFLD. The levels of MFN1, MFN2, and OPA1 protein expression in the hepatocytes of patients with NAFLD, which are involved in mitochondrial fusion, are decreased compared to normal hepatocytes, while the levels of fission mitochondrial 1 (FIS1) and DRP1 protein expression, which are associated with mitochondrial fission, exhibit an upward trend. This change results in impairment of mitochondrial fusion and augmentation of mitochondrial fission. Mitochondrial dynamics imbalance can alter the normal morphologic structure of mitochondria, disrupt the tubular mitochondrial network, leading to mitochondrial fragmentation, swelling, and transformation into short rod-like or spherical shapes. This imbalance is accompanied by a decrease in mitochondrial membrane potential and disruption or loss of cristae structures (Figure 5). These changes inhibit cellular metabolism, impair mitochondrial function, and accelerate cell death⁹¹.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5

Targeting mitochondrial dynamics for liver cancer intervention. This figure illustrates the mechanisms underlying mitochondrial dynamic disorders during the progression from normal liver-to-liver cancer. The left part presents the pathologic evolution of liver tissue undergoing malignant transformation due to mitochondrial dynamic imbalance, while the right part focuses on the dynamic cycle of mitochondrial fusion and fission. During fission, upregulation of DRP1 and FIS1 proteins drives mitochondria to divide into multiple smaller mitochondria. During fusion, downregulation of MFN1, MFN2, and OPA1 proteins inhibits mitochondrial fusion, disrupting the normal dynamic balance. DRP1, dynamin-related protein 1; FIS1, mitochondrial fission 1 protein; MFN1, mitofusin 1; MFN2, mitofusin 2; OPA1, optic atrophy 1. https://app.biorender.com/illustrations/67aaf0588086827cc67759d6

Drug intervention

Drug intervention represents the most common therapeutic approach targeting mitochondria at present. Drugs can act on key mitochondrial proteins or pathways, such as improving mitochondrial function, inhibiting mitochondrial fission, and promoting mitophagy. Additionally, some antioxidants can alleviate oxidative stress, thereby suppressing the growth of HCC cells.

Mitochondrial transplantation

Termed “mitochondrial transplantation”, this emerging research field shows promise in in vitro, in vivo, and various clinical applications, offering benefits, such as preventing cell death, reducing inflammation, and restoring cellular metabolism and oxidative balance¹¹⁶. Mitochondrial transplantation involves multiple techniques, including co-culture, electroporation, microinjection, magnet-mediated transfer, photothermal nanoknife, and exosome-encapsulated delivery, to achieve directional transplantation of mitochondria into HCC cells. Enrichment efficiency in tumor tissues is enhanced via TOM20-targeted peptide modification¹¹⁷. For donor mitochondrial screening, high-activity mitochondria derived from M1 macrophages or normal hepatocytes are preferred to avoid genetic mutations in tumor cell mitochondria (e.g., ND1 gene deletion). Researchers first proposed targeting the transplantation of mitochondria from M1 macrophages into M2 tumor-associated macrophages (TAMs) to induce M2 TAMs to shift toward glycolytic metabolism, thereby reprogramming the immunophenotype of M2 TAMs, reversing the tumor immunosuppressive microenvironment, and enhancing cancer ICB therapy¹¹⁸. Nanotube connections established between bone marrow stromal and T cells serve as channels for transplanting stromal cell mitochondria into CD8⁺ T cells. CD8⁺ T cells receiving mitochondrial transplantation exhibit enhanced adaptive metabolism, improved tumor-killing ability, and reduced exhaustion, overcoming the barrier of mitochondrial loss and dysfunction-driven T cell exhaustion after CAR-T therapy¹¹⁹. In vitro and animal models have confirmed that mitochondrial transplantation significantly inhibits tumor growth by arresting the cell cycle and inducing apoptosis¹²⁰. Despite its promise, mitochondrial transplantation faces challenges, as follows: 1) Immunological rejection of mitochondria may trigger NLRP3 inflammasome activation by allogeneic mtDNA, which can be mitigated by HLA typing for donor screening or using autologous fibroblast mitochondria. 2) Degradation of exosome carriers by matrix metalloproteinase (MMP)-9 in the tumor microenvironment can be addressed by modifying exosome membranes with polyethyleneimine to enhance stability. 3) Insufficient targeting efficiency and low mitochondrial fusion efficiency in HCC cells can be improved via translocase of outer mitochondrial membrane (TOM)20 peptide modification combined with an ultrasound-targeted microbubble delivery system. Future strategies may also adopt combination therapy, such as concurrent use with mitochondria-targeted drugs (e.g., MitoQ), to reduce tumor recurrence through dual inhibitory effects.

Gene therapy

In the field of gene therapy for HCC, multiple strategies have been synergistically developed to address disease challenges. At the level of tumor suppressor gene restoration, p53 gene replacement therapy delivers wild-type p53 gene via an AAV9 vector, activating the Bax/caspase-9 pathway and significantly increasing the HepG2 cell apoptosis rate. PTEN gene repair uses CRISPR-Cas9 technology to correct PTEN mutations, inhibiting the PI3K/Akt/mTOR pathway and effectively reducing tumor angiogenesis¹²¹. Novel lipid nanoparticle-encapsulated CTNNB1 siRNA (LNP-CTNNB1) has exhibited significant antitumor effects in multiple immunocompetent mouse models of β-catenin-mutated HCC¹²². To further enhance efficacy, combination therapy strategies have become a research hotspot.

Combination gene therapy and immunotherapy involve using AAV to deliver the OX40L gene for CD8⁺ T cell activation, which when combined with ICB therapy and in situ tumor vaccine therapy, elicits synergistic antitumor immune responses. Stereotactic body radiation therapy combined with sorafenib improves clinical outcomes in HCC patients, particularly those with major vascular invasion, extending the median overall survival (OS) by 3.5 months and median progression-free survival (PFS) by 3.7 months, while enhancing quality of life¹²³. Additionally, emerging research explores the integration of the ultrasound-activated artificial enzyme, P-Por-Os, with precision gene therapy (PGT), in which external ultrasound stimulation regulates artificial enzyme activity to generate ROS for precise HCC ablation¹²⁴. In animal model validation, a gene therapy inducing endogenous microRNA-22 (miR-22) production successfully treated mice with HCC, demonstrating good safety and efficacy. However, clinical translation of gene therapy faces multiple bottlenecks. Non-specific expression in normal hepatocytes causes hepatotoxicity, which can be mitigated by introducing an AFP promoter for HCC-specific expression. Immunologic rejection induced by viral vectors can be alleviated by alternating multiple serotype AAVs (e.g., AAV8→AAV9). Exosome carriers are easily cleared by Kupffer cells, resulting in insufficient delivery efficiency, which can be improved via macrophage depletion pretreatment (clodronate disodium liposomes). These challenges and countermeasures provide directions for subsequent research.

Intercellular mitochondrial transfer and therapeutic intervention

Mitochondrial transfer, defined as the intercellular transmission of mitochondria or their components, was first discovered in 2006 when mesenchymal stem cells (MSCs) were shown to transfer functional mitochondria into mtDNA-deficient (ρ0) cells, restoring the OXPHOS capacity. A 2023 consensus statement standardized terminology, defining the natural process of intercellular mitochondrial transmission as “mitochondrial transfer,” while “mitochondrial transplantation” refers to the therapeutic application of exogenous mitochondria.

Mitochondrial transfer relies on diverse structural bases, primarily achieved via tunnel nanotubes (TNTs), cell fusion, gap junctions, and extracellular vesicles¹²⁵. The mitochondrial Rho GTPase (Miro) family has a central role in regulatory mechanisms. Miro1 localizes to the outer mitochondrial membrane, mediating connections between mitochondria and microtubule motor proteins. Endoplasmic reticulum-mitochondria contact sites (ERMES) and the mitochondrial inner and outer membrane organizing system (MICOS) synergistically mediate non-vesicular transport of phosphatidylserine, with lipid transfer proteins, like ORP5/8 and VPS13D, also participating. Mitochondrial transfer in the TME exhibits bidirectional effects. Stromal-to-cancer cell transfer promotes tumor progression, such as glioblastoma acquiring astrocyte mitochondria via growth-associated protein 43 (GAP43)-dependent junctions, which enhances oxidative metabolism and tumorigenicity¹²⁶. Conversely, mitochondrial transfer can exert antitumor effects, as occurs in osteocyte-to-metastatic cancer cell transfer triggering cGAS/STING-mediated antitumor immune responses¹²⁷.

Selectively enhancing mitochondrial transfer in MSCs may optimize stem cell therapy for tissue repair in therapeutic applications, while inhibiting cancer cells from hijacking mitochondria via cancer-associated fibroblasts (CAFs) or immune cells in the TME emerges as a novel antitumor strategy. Studies have shown cancer cells transfer mitochondria carrying mtDNA mutations to tumor-infiltrating lymphocytes (TILs) via TNTs and small extracellular vesicles (sEVs), inducing metabolic dysfunction, senescence, and loss of antitumor efficacy in T cells. This mechanism not only enables tumor immune evasion but also induces resistance to ICB therapy¹²⁸. Molecules, like USP30, have key roles by mediating mitochondrial resistance to autophagy and homotypic replacement⁷³. Overall, mitochondrial transfer occurs in physiologic/pathologic contexts, especially under oxidative stress from excessive ROS, though the exact triggers and regulatory mechanisms require further study¹²⁹. Current technical limitations (e.g., difficulty in precisely tracking and visualizing mitochondrial transfer in living tissues) have limited direct evidence of in vivo mitochondrial transfer in humans. Nevertheless, targeting mitochondrial transfer represents a promising therapeutic strategy, poised to drive innovative changes in organ treatment models for regenerative medicine, oncology, and other fields.

Other treatments

In addition to pharmacologic and gene therapies, other treatment modalities include photodynamic therapy and nanotechnology. Photodynamic therapy uses light of specific wavelengths to activate photosensitizers within mitochondria, generating ROS that kill cancer cells. Nanotechnology, in contrast, uses nanoparticles as carriers to deliver drugs directly to mitochondria, enhancing the targeting and efficacy of the drugs. Nanoparticle drug delivery systems can precisely deliver drugs to mitochondria, enhancing therapeutic efficacy while reducing toxicity. For example, researchers have designed a mitochondria-targeted system based on mesoporous silica nanoparticles. By modifying the nanoparticle surface with triphenyl phosphonium and folic acid, stepwise targeting of mitochondria in tumor cells was achieved¹³⁰.

Despite significant advances in laboratory studies of mitochondrial targeting for HCC, many challenges face clinical application. For example, the specific mechanism underlying mitochondria in HCC should be further explored and more targeted drugs should be developed.

Challenges

In the research on targeting mitochondrial dysfunction to intervene in HCC, specificity, drug resistance, and personalized treatment are the main challenges. First, most of the drugs targeting mitochondria lack tumor specificity and are prone to damage normal tissues. Therefore, there is an urgent need to develop an efficient, low toxic, and mitochondrial targeted delivery system with HCC specificity. Second, HCC cells are prone to developing resistance to mitochondria-targeted drugs. The mechanisms are complex and involve multiple aspects, such as metabolic reprogramming and enhanced antioxidant defenses. A deep investigation into these resistance mechanisms and the development of effective strategies to overcome the resistance mechanisms are of vital importance.

Finally, because HCC is highly heterogeneous, there is a significant variation in patient responses to mitochondria-targeted therapies. Therefore, establishing a reliable biomarker system to drive the development of personalized precision treatment is an important direction for the future.

Future prospects

With the in-depth research on mitochondrial biology and the metabolic mechanisms underlying HCC, more potential mitochondrial targets will be revealed, laying a solid theoretical foundation for the development of novel targeted drugs. In addition, several mitochondrial treatments are combined to form a new and more efficient therapeutic strategy, which is expected to overcome the limitations of monotherapy and significantly improve the therapeutic effect. In addition to the unique advantages of nanotechnology in drug delivery, which provide new tools for the treatment of HCC, nanotechnology can enhance the specificity, stability, and bioavailability of mitochondria-targeted drugs, thereby optimizing therapeutic outcomes. In summary, targeting mitochondrial dysfunction to intervene in HCC shows great potential. With the continuous progress of research and technology this area may bring new hope for liver cancer patients despite the need to overcome many challenges.