Nanomaterial-Mediated Targeting of Mitochondrial Metabolism: Strategies and Applications in Cancer Therapy

Introduction

The unique metabolic reprogramming of cancer cells allows them to proliferate quickly and survive in hostile microenvironments.¹ Mitochondria, as the cell’s powerhouses, exert a pivotal influence in tumor cells’ metabolic regulation, redox balance, apoptosis and immune evasion. Due to metabolic reprogramming in cancer cells, mitochondrial metabolism is often altered to meet the demands of rapid proliferation.² These metabolic alterations not only support tumor cell proliferation, metastasis, and drug resistance but also lead to intracellular ROS accumulation, oxidative stress, and metabolic imbalances, posing significant challenges for cancer therapy.³ Significantly, modifications in mitochondrial metabolism can facilitate tumor cell proliferation while simultaneously inducing their demise through heightened oxidative stress and related mechanisms. By altering mitochondrial metabolic pathways, cancer cells can secure greater energy resources, improve antioxidant capabilities, and resist apoptosis, thus augmenting their flexibility and lifespan. This metabolic reprogramming also presents new targets for therapeutics focused on mitochondrial metabolism. In comparison to traditional treatments—such as surgery, chemotherapy, radiation, and immunotherapy—targeting mitochondrial metabolism is regarded as a highly promising approach. It can inhibit tumor proliferation and improve therapeutic effectiveness by precisely regulating mitochondrial function, while reducing toxicity to normal cells.⁴

Clinically, mitochondrial metabolic addiction underpins multiple high-mortality, treatment-refractory malignancies—including pancreatic ductal adenocarcinoma (PDAC), glioblastoma (GBM),⁵ and triple-negative breast cancer (TNBC)—that collectively account for over 60% of global cancer-related deaths, with dismal 5-year overall survival rates and minimal clinical progress over the past decades.⁶ Therapeutic failure in these tumors is directly driven by mitochondrial metabolic reprogramming, which mediates universal resistance to chemotherapy, radiotherapy, and immunotherapy.⁷ Even targeted metabolic agents have failed late-stage clinical trials due to poor tumor-specific delivery, highlighting the urgent clinical need for precise mitochondrial-targeted therapeutic strategies.

Although targeting mitochondrial metabolism represents a theoretically sound therapeutic paradigm, its clinical translation using conventional small molecules remains severely hindered by an array of physiological and pharmacokinetic barriers. Firstly, the lack of intrinsic mitochondrial selectivity often leads to off-target effects in healthy tissues, resulting in systemic toxicity.⁸ Furthermore, the highly negative mitochondrial membrane potential (ΔΨm) and the complex double-membrane structure pose formidable obstacles to the passive diffusion of pharmacological agents.⁹ Many small-molecule drugs fail to achieve therapeutic concentrations within the mitochondrial matrix (MM) or lack specificity for the tumor-specific ΔΨm gradient, resulting in sub-optimal intra-organelle accumulation within the tumor site.^10,11 Consequently, there is a critical need for delivery platforms that can circumvent these biological hurdles and ensure site-specific accumulation.

Nanomaterials present a transformative strategy to address these difficulties. Through the rational nanoengineering of precisely targeted nanomaterials, pharmacological agents can be effectively delivered to mitochondria, thereby improving their bioavailability and reducing toxicity to healthy cells. The dimensions, surface alterations, and targeting ligands of nanomaterials facilitate their selective infiltration into tumor cells and accumulation within mitochondria, so substantially impeding tumor cell proliferation and metastasis.¹²

Research on mitochondria-targeted nanomaterials in cancer therapy is currently progressing, especially in drug delivery, synergy with conventional therapies, and immunotherapy. Nevertheless, enhancing targeted specificity, optimizing drug release mechanisms, and minimizing side effects continue to be significant difficulties in this domain.

This review encapsulates the metabolic attributes of tumor mitochondria and offers an in-depth examination of the properties and targeted tactics of nanomaterials aimed at mitochondrial metabolism. This text comprehensively discusses recent advancements in mitochondria-targeted nanomaterials, emphasizing their potential in cancer therapy within the current research context. The paper additionally analyzes the benefits and obstacles of nanomaterial-mediated modulation of mitochondrial metabolism in malignancies. Ultimately, it emphasizes the existing challenges and future opportunities of mitochondria-targeted nanomaterial therapies, intending to offer direction for the development of highly effective and low-toxicity nanomedicines.

Metabolic Characteristics of Tumor Mitochondria

Mitochondria are organelles encased in a double membrane; they comprise the outer mitochondrial membrane (OMM), intermembrane space (IMS), inner mitochondrial membrane (IMM), and mitochondrial matrix (MM). Mitochondria chiefly produce adenosine triphosphate (ATP) via OXPHOS to provide energy for cellular functions. Moreover, they participate in the regulation of intrinsic apoptosis and signal transmission. In tumor biology, the Warburg effect asserts that cancer cells preferentially convert glucose to lactate via glycolysis, even in the presence of oxygen, instead of directing pyruvate into mitochondria for OXPHOS. This predilection was originally believed to stem from mitochondrial damage in cancer cells, resulting in compromised mitochondrial activity.¹³ However, recent data indicates that the Warburg effect is not attributable to mitochondrial malfunction. Conversely, the mitochondria of the majority of tumor cells are either intact or augmented, and mitochondria-mediated OXPHOS continues to be the principal mechanism for ATP synthesis in cancer cells.¹⁴ Beyond this cell-autonomous glycolysis, the metabolic landscape is further shaped by the “Reverse Warburg Effect” and intra-tumoral metabolic symbiosis. In this model, cancer-associated fibroblasts (CAFs) undergo aerobic glycolysis to secrete high-energy metabolites (eg, lactate and pyruvate), which are captured by adjacent oxidative cancer cells to fuel their mitochondrial OXPHOS. This metabolic compartmentalization underscores the heterogeneous efficacy of mitochondria-targeted therapies.^15,16 Mitochondria not only supply energy and metabolic precursors to promote rapid tumor cell proliferation, but also enable tumor escape of programmed cell death by modulating apoptotic pathways. Furthermore, mitochondrial metabolites and the reactive oxygen species (ROS) they produce can influence pro-tumorigenic signaling and epigenetic regulation, significantly contributing to tumor initiation and development.¹⁷ To meet the demands of rapid proliferation, tumor cells undergo metabolic reprogramming of their mitochondria, including alterations in energy metabolism, amino acid metabolism and lipid metabolism (Figure 1).

Figure 1 Differences in Mitochondrial Metabolism Between Normal and Cancer Cells (By BioRender). While normal mitochondria (left, blue panel) maintain a physiological balance of the TCA cycle, β-oxidation, and redox signaling, cancer mitochondria (right, red panel) undergo significant reprogramming to support rapid proliferation. This transformation is characterized by enhanced nutrient uptake, altered TCA cycle flux, and elevated ROS production leading to oxidative stress. Upward arrows (↑) explicitly mark upregulated processes in cancer cells, such as β-oxidation↑, Acetyl-CoA↑, and ROS↑, whereas downward arrows (↓) indicate decreased efficiency, such as the reduction of ATP↓ synthesis in the mitochondrial matrix. Yellow and red burst icons symbolize energy production and ROS-induced oxidative stress, respectively. Furthermore, all bold and underlined terms (eg, Anaplerotic Flux, Glutaminolysis, and Oxidative Stress) are defined as the hallmark metabolic phenotypes that distinguish tumor mitochondria from their normal physiological counterparts.

Energy Metabolism

Energy metabolism in tumor cell mitochondria relies on the stringent coupling between OXPHOS on the IMM and the tricarboxylic acid cycle (TCA cycle) in the MM. Under standard physiological settings, the TCA cycle functions as the universal oxidative route for carbohydrates, lipids, and proteins. It produces nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FADH₂). These reducing equivalents subsequently drive OXPHOS. This system consists of the electron transport chain (ETC) and ATP synthase (complex V). The electron transport chain has four protein complexes (I–IV) and two mobile carriers: coenzyme Q and cytochrome c (Cyt c). ATP is created efficiently through this method to sustain cellular function. In tumor cells, this metabolic pathway experiences substantial adaptive reprogramming.¹⁸ In OXPHOS, its activity does not merely diminish in line with the traditional Warburg effect. Rather, it is regulated with flexibility based on the type of malignancy.¹⁹ Most solid tumors, such as breast, gastric, and hepatocellular carcinomas, demonstrate downregulation of OXPHOS. Conversely, an increasing amount of evidence suggests that specific malignancies, including leukemia, lymphoma, pancreatic ductal adenocarcinoma, and melanoma, enhance OXPHOS to facilitate their proliferation. This metabolic profile is frequently linked to the emergence of medication resistance, therefore complicating treatment.^19,20 Meanwhile, in tumor cells, the TCA cycle is no longer primarily devoted to energy production but is instead rewired toward biosynthesis. Large amounts of TCA cycle intermediates are diverted into anabolic pathways. For example, citrate is transported to the cytosol through the citrate transporter solute carrier family 25 member 1 (SLC25A1) and subsequently cleaved by ATP-citrate lyase (ACLY) into acetyl-CoA and oxaloacetate. Acetyl-CoA functions as a crucial substrate for the biosynthesis of fatty acids and cholesterol necessary for cell membrane development, therefore facilitating rapid tumor cell growth.²¹ Nevertheless, this substantial utilization of intermediates results in their exhaustion. Tumor cells thus utilize two anaplerotic routes for compensation: glutamine (Gln) catabolism and pyruvate carboxylation, with Gln catabolism serving as the predominant anaplerotic route.⁴ Notably, certain genetic mutations in TCA cycle–related enzymes can directly drive tumorigenesis. For example, mutations in succinate dehydrogenase (SDH/complex II) occur in hereditary paraganglioma, fumarate hydratase (FH) mutations are found in hereditary leiomyoma, and isocitrate dehydrogenase (IDH1/2) mutations are observed in glioma. These alterations lead to the abnormal accumulation of oncometabolites, including succinate, fumarate, and 2-hydroxyglutarate (2-HG), respectively. These metabolites competitively inhibit α-ketoglutarate (α-KG)–dependent dioxygenases, resulting in widespread histone and DNA hypermethylation and thereby promoting malignant transformation.^22,23 Notably, IDH1/2 mutations in gliomas are also linked to suppressed OXPHOS activity, resulting in a diminished ΔΨm.²⁴ This metabolic nuance is critical for nanomedicine design, as a reduced ΔΨm may impair the accumulation of delocalized lipophilic cation (DLC)-based nanocarriers due to the weakened electrophoretic driving force. Therefore, understanding these metabolic shifts is essential for developing precise mitochondrial-targeted therapies.

Glutamine (Gln) Metabolism

Gln is a conditionally necessary amino acid and the most prevalent amino acid in the bloodstream. In tumor cells, Gln is carried into the cytoplasm by specialized transporters and degraded by glutaminase (GLS) to produce glutamate (Glu) and ammonia (NH₃). Based on metabolic reprogramming, tumor cells significantly enhance Gln catabolism by upregulating the expression of Gln transporters and GLS1, a process frequently termed “Gln addiction.”²⁵ Specifically, SLC1A5 (ASCT2) serves as the primary importer for the initial uptake of extracellular Gln. Subsequently, the intracellular Gln pool is utilized by the heterodimeric transporter SLC7A5/SLC3A2 (LAT1/CD98hc) as an exchange substrate. In this indirect mechanism, SLC7A5 facilitates the efflux of Gln to drive the inward transport of essential amino acids like leucine, thereby coupling metabolic supply with signaling activation.²⁶

Beyond its role in signaling, Gln serves as a critical carbon and nitrogen source for mitochondrial biosynthesis. Glu derived from Gln catabolism enters the mitochondria, where it undergoes oxidative deamination by glutamate dehydrogenase (GLUD), or is transformed via alanine or aspartate aminotransferases (ALT/AST), resulting in the formation of α-ketoglutarate (α-KG). As a crucial anaplerotic substrate, α-KG enters the TCA cycle to restore carbon supplies and maintain mitochondrial redox balance.²⁷ Furthermore, the amide nitrogen of Gln facilitates the synthesis of purines, pyrimidines, and amino sugars, while Glu serves as a indispensable precursor for the biosynthesis of the antioxidant glutathione (GSH).²⁸ In conjunction with NADPH, GSH assists tumor cells in removing surplus ROS and preserving cellular homeostasis.

To exploit this metabolic dependency, advanced nano-platforms have been designed to co-deliver GLS1 inhibitors alongside ROS-generating agents. Such integrated strategies not only deplete the cellular antioxidant pool by blocking glutamate production but also trigger mitochondria-mediated apoptosis through the cumulative stress of TCA cycle dysfunction and catastrophic oxidative damage. This dual-action approach represents a promising frontier in overcoming the metabolic flexibility and therapeutic resistance of refractory solid tumors.

Fatty Acid β-Oxidation

Fatty acid β-oxidation (FAO) is the mechanism via which long-chain fatty acids (LCFAs) are catabolized in the MM to produce acetyl-CoA, NADH, and FADH₂. Acetyl-CoA enters the TCA cycle as a carbon substrate, while NADH and FADH₂ contribute to OXPHOS to generate energy, rendering FAO a crucial energy source during fasting. Carnitine palmitoyltransferase 1 (CPT1) serves as the rate-limiting enzyme in FAO. At the molecular level, FAO is driven by the CPT1, which facilitates the transport of long-chain fatty acids across the mitochondrial membranes. Nanomaterials targeting this pathway typically focus on inhibiting the CPT1A isoform or downregulating the electron transfer flavoprotein (ETF) complex.²⁹ This disruption effectively cuts off the supply of NADH and FADH₂ to the mitochondrial respiratory chain, leading to a catastrophic ATP deficit. Tumor cells typically upregulate CPT1 expression to facilitate survival and proliferation in nutrient-deficient environments. In certain tumor types, this elevation of CPT1 has been associated with the induction of epithelial–mesenchymal transition (EMT) and the acquisition of stem-like characteristics, thereby enhancing metastatic potential.³⁰ However, this relationship appears to be highly context-dependent. While FAO provides the metabolic flexibility required for migration, emerging evidence suggests it may also support lipid remodeling for membrane dynamics during the metastatic cascade.^31,32 Significantly, when FAO is elevated beyond the tumor cell’s antioxidant systems’ buffering capacity, excessive ROS buildup can induce apoptosis, hence exerting an anticancer impact. Simultaneously, perturbing the oxidative equilibrium of neoplastic cells can efficiently surmount pharmacological resistance.³³

Calcium Ion Regulation

Calcium ions (Ca²⁺) are key intracellular signaling molecules, primarily stored in the ER, the main intracellular calcium reservoir. Mitochondria and the ER form close physical contacts through mitochondria-associated membranes (MAMs), allowing Ca²⁺ released from the ER to enter the MM via the mitochondrial calcium uniporter (MCU). In mitochondria, Ca²⁺ demonstrates concentration-dependent effects: moderate levels activate TCA cycle enzymes to enhance ATP synthesis for cellular proliferation, whereas excessive accumulation triggers mitochondrial permeability transition pore (mPTP) opening and initiates apoptosis.³⁴ The key regulators of intracellular Ca²⁺ levels are the MCU and mitochondrial calcium uptake protein 1 (MICU1). To maintain this delicate balance, MICU1 acts as a molecular “clamp,” establishing a gatekeeping threshold that inhibits MCU activity at low cytosolic Ca²⁺ levels to prevent chronic overload and premature mPTP opening. Tumor cells modulate MCU and/or MICU1 expression to maintain Ca²⁺ within a pro-survival window.³⁵ Specifically, in breast and colorectal cancers, tumor cells upregulate MCU to drive proliferation while relying on MICU1-mediated clamping to avoid crossing the lethal mPTP threshold. In contrast, ovarian cancer and melanoma further upregulate MICU1 to suppress Ca²⁺ influx, thereby enhancing apoptotic resistance.³⁶

ROS and Redox Homeostasis

ROS are extremely reactive molecules generated from oxygen, encompassing superoxide anion (O₂⁻), hydrogen peroxide (H₂O₂), hydroxyl radical (·OH), and singlet oxygen (¹O₂). Mitochondria serve as the principal generator of intracellular ROS. In the process of ATP generation through OXPHOS, roughly 1–2% of electrons escape from complexes I and III of the ETC and interact with adjacent oxygen molecules to generate O₂⁻.³⁷ Oxidative stress is initiated when ROS buildup surpasses the scavenging capability of cellular antioxidant mechanisms. Elevated amounts of ROS can cause peroxidation of mitochondrial membrane lipids and facilitate prolonged opening of the mPTP, thereby undermining membrane integrity. This results in the liberation of Cyt c into the cytosol, triggering the caspase cascade or precipitating cell death via irreversible oxidative damage to DNA, proteins, and lipids.³⁸ To counteract this, cells have developed a sophisticated antioxidant defense system, which primarily includes: (1) enzymatic antioxidants: superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GPX), glutathione reductase (GR), peroxiredoxin (PRX) and thioredoxin (TRX); (2) non-enzymatic antioxidants: GSH, flavonoids, vitamin A, vitamin C and vitamin E.³⁹ Mitochondrial ROS (mROS) levels in tumor cells are markedly increased—approximately 100–1000 times more than the ~20 nM found in normal cells (this range is highly context-dependent, varying significantly by tumor type, microenvironment, and detection methodologies such as MitoSOX-based flow cytometry)⁴⁰ —attributable to oncogene activation, loss of tumor suppressors, changes in TCA cycle enzymes, and hypoxia.⁴¹ To withstand apoptosis induced by high ROS levels, tumor cells maintain survival by upregulating their antioxidant defense systems and expressing higher levels of antioxidant proteins.⁴² In fact, ROS exhibit a “concentration dependent” double-edged sword effect in tumors. At moderate levels, they can promote tumor progression by driving metabolic reprogramming—such as activating hypoxia-inducible factor 1 (HIF-1) to enhance glycolytic gene expression and activating nuclear factor erythroid 2–related factor 2 (Nrf2) to boost antioxidant capacity. They also enhance tumor cell proliferation and survival through multiple oncogenic signaling pathways, including the epidermal growth factor receptor (EGFR) pathway and Akt/NF-κB-dependent mitochondrial transcription factor B2 (TFBM2), and inducing genomic instability. However, at high levels, ROS cause cellular damage and trigger apoptosis, necroptosis, or ferroptosis by increasing oxidative stress and activating ROS-dependent death signals.^43,44

Characteristics of Mitochondria-Targeting Nanomaterials

Nanomaterials are characterized as substances possessing at least one dimension within the range of 1 to 100 nm.⁴⁵ Nanomaterials exhibit unique advantages in drug delivery. They possess size-dependent effects, a high surface-area-to-volume ratio, and tunable functionality. The primary advantages encompass: (1) facilitating targeted drug delivery to significantly diminish systemic toxicity; (2) controlled drug release; (3) effective encapsulation and safeguarding of drugs from degradation, thereby extending their circulation time; and (4) promoting drug transport across biological barriers.^45–47 Building on this, mitochondria-targeted nanotherapies have emerged as a novel anticancer strategy. In this approach, nanomaterials primarily serve as carriers for drug delivery, with their targeting capability largely dependent on the incorporation of specific targeting moieties and the structural design of the nanoparticles. Because the IMM possesses a pronounced negative membrane potential, decorating nanomaterials with delocalized lipophilic cations (DLCs) has become a classical strategy for mitochondrial targeting. In addition, precise mitochondrial targeting can be achieved by incorporating mitochondrial targeting sequences (MTS) or exploiting the unique microenvironmental characteristics of mitochondria.⁴⁸ An overview of representative drug-loaded nanomedicines specifically engineered to target tumor mitochondria is provided in Table 1.

Table 1 Drug-Loaded Nanomedicines Targeting Mitochondria of Tumor Cells

Membrane Potential–Driven Targeting

The ΔΨm of normal cells is approximately –180 mV. In tumor cells, metabolic reprogramming—characterized by increased glycolysis and overexpression of hexokinase 2 (HK2)—results in significant hyperpolarization of ΔΨm, typically about 60 mV higher than in normal cells.¹⁴ DLCs are the most commonly used membrane potential–driven targeting moieties. They possess two essential characteristics: a positive charge and lipophilicity. These characteristics enable them to easily penetrate the cell membrane and be strongly attracted to the negative charge of the IMM, thereby passing through the double membrane and entering the MM to perform their function.⁶⁷ According to the Nernst equation, at 37°C, every 61.5 mV increase in ΔΨm leads to a tenfold increase in mitochondrial uptake of DLCs.⁶⁸ Exploiting the hyperpolarized ΔΨm characteristic of tumor cells, DLCs can selectively accumulate at high levels within tumor cell mitochondria. This can be achieved either by directly conjugating DLCs to drug molecules or by decorating them on the surface of other nanocarriers for cancer therapy.⁶⁹ Commonly used DLCs include triphenylphosphonium (TPP), rhodamine 123, dequalinium (DQA) and F16.

TPP

As a representative DLC, TPP features a positively charged phosphorus atom linked to three lipophilic phenyl groups. This unique structure facilitates its stable accumulation across mitochondrial membranes while offering the advantages of amphiphilicity and straightforward synthesis.^70,71 Studies have shown that, compared with other DLCs, TPP exhibits superior lipophilicity and, due to its relatively large ionic radius, requires lower activation energy to traverse biological membranes.^72,73 In addition, TPP-based compounds demonstrate excellent safety and minimal intrinsic cytotoxicity at therapeutic doses compared with other DLCs.⁷⁴

Currently, there are two approaches to using TPP for mitochondria-targeted cancer therapy. The first approach involves directly conjugating TPP to drug molecules, thereby enabling targeted delivery of the drugs to tumor cell mitochondria. Jiang et al designed a mitochondria-targeted TPP–resveratrol (RSV) conjugate and found that this conjugate enhanced the anticancer activity of RSV, increasing its antitumor effect by approximately threefold compared with RSV alone (Figure 2A).⁷⁵

Figure 2 The Mechanism of TPP Targeting Mitochondria (By BioRender); black arrows indicate directional transport processes; yellow arrows indicate the structural components and key chemical characteristics of the targeting moieties. The left panel represents Strategy 1: Direct Drug Conjugation, which focuses on the use of lipophilic cations to drive the direct accumulation of therapeutic agents within the mitochondria. The right panel represents Strategy 2: Nanocarrier Surface Decoration, illustrating how advanced nanoengineering (eg, liposomes, polymer NPs, and micelles) can be utilized to overcome the toxicity and release limitations associated with direct conjugates. (A) Direct conjugation of TPP for molecular targeting. Chemical structure and schematic of the TPP-resveratrol (RSV) conjugate. The integration of TPP enhances the mitochondrial accumulation and anticancer activity of RSV compared to the free drug. (B) TPP-functionalized liposomes for targeted drug delivery. Schematic illustration of DOX-loaded, TPP-modified liposomes (TPPLs). These nanocarriers exploit the hyperpolarized mitochondrial membrane potential (ΔΨm) of tumor cells for selective delivery, facilitating mitochondrial membrane fusion and programmed cell death. (C) Polymeric nanoparticles (NPs) for overcoming multidrug resistance. Structure of the TPH/PTX system (TPP-PF127-HA loaded with Paclitaxel). The TPP-guided polymeric NPs exhibit superior tumor-targeting capability and remarkable efficacy in reversing chemotherapy resistance. (D) ROS-responsive multifunctional micelles for synergistic therapy. Schematic of the triple-modality “CTC” micelles (CAT/CPT-TPP/PEG-Ce6). These micelles achieve mitochondrial targeting via TPP and integrate chemotherapy, PDT, and ROS amplification to induce immunogenic cell death (ICD) and activate the host immune system.

The second approach is employing TPP as a “guiding moiety” to functionalize the surface of nanomaterials for the purpose of targeting tumor mitochondria. Despite its prevalence as a mitochondria-targeting moiety, the clinical translation of TPP-conjugated agents is often constrained by a relatively narrow therapeutic window. Evidence suggests that TPP+ derivatives can exhibit significant dose- and structure-dependent mitochondrial toxicity.⁷⁶ Specifically, long-chain alkyl TPP+ conjugates (eg, decyl- and dodecyl-TPP+) have been shown to induce substantial proton leakage, inhibit respiratory chain complexes, and collapse ΔΨm at concentrations as low as 1 μM. Such bioenergetic dysfunction—often accompanied by compensatory increases in the extracellular acidification rate (ECAR)—limits the safe dosage for in vivo applications.⁷⁷ To mitigate these off-target effects, nanocarrier encapsulation has been strategically employed to shield the TPP+ moieties during circulation. This approach facilitates controlled release and site-specific accumulation within tumor mitochondria, thereby effectively widening the therapeutic window and enhancing the biocompatibility of TPP-functionalized platforms while maintaining high targeting efficiency.^78,79 Common mitochondria-targeted nanocarriers include liposomes, polymeric nanoparticles (NPs), micelles, vesicles, dendrimers, gold nanoparticles (AuNPs), carbon nanotubes (CNTs) and inorganic nanocarriers.⁷⁹

Liposomes are artificially synthesized, spherical vesicles whose membranes consist of one or more phospholipid bilayers surrounding an aqueous core. This structure allows liposomes to encapsulate both hydrophobic and hydrophilic drugs and deliver them into cells via membrane fusion or endocytosis.⁸⁰ Ekmekcioglu et al designed a polymer conjugate consisting of 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(polyethylene glycol)-2000] (DSPE-PEG2000-NH2)-(3-carboxypropyl) TPP (DSPE-PEG-TPP), which was utilized to create nontoxic, pH-sensitive, mitochondria-targeted doxorubicin-loaded liposomes (TPPLs). In comparison to DOX-loaded 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy (polyethylene glycol)-2000] (18:0 PEG2000 PE) liposomes (PPLs), empty TPPLs exhibited no intrinsic toxicity towards the evaluated cancer cells, hence confirming their biosafety as a delivery vehicle. Moreover, DOX-loaded TPPLs exhibited significant efficacy in killing cancer cells, suggesting that TPPL serves as a promising platform for targeted drug delivery to tumor mitochondria (Figure 2B).⁴⁹ Sivagnana et al functionalized 10,12-pentacosadiynoic acid (PCDA) with TPP and the fluorescent dye dansyl (DAN) group, and the TPP/DAN functionalized PCDA was incorporated into the 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) phospholipid to generate the functionalized PCDA/DMPC-based liposome (Lip-DT), and encapsulating DOX with Lip-DT yields a novel nanomedicine, Lip-DT-DOX.Lip-DT-DOX efficiently induces apoptosis in cancer cells while exhibiting no significant toxicity toward normal human embryonic kidney cells (HEK293) or cardiomyocytes (AC-16). In addition, the researchers innovatively incorporated DAN into Lip-DT-DOX, enabling real-time tracking of mitochondrial drug uptake and monitoring of drug delivery and release, thereby providing a more direct understanding of the underlying mechanisms of drug transport.⁵⁰

Polymeric NPs are solid colloidal particles composed of biodegradable or biocompatible synthetic or natural polymers, such as poly (lactic-co-glycolic acid) (PLGA), polycaprolactone (PCL) and chitosan. They typically range in size from 1 to 1000 nm. Their advantages include: (1) high stability, (2) controllable drug release, (3) high drug-loading capacity, (4) facile surface functionalization, and (5) excellent biocompatibility and biodegradability.⁸¹ He et al designed and synthesized a polymer NPs system, TPP-PF127-HA (TPH), where PF127 is a triblock copolymer (poly (ethylene oxide)-block-poly (propylene oxide)-block-poly (ethylene oxide), PEO-PPO-PEO). They discovered that the paclitaxel (PTX)-loaded TPH (TPH/PTX) not only exhibited excellent tumor-targeting capability and remarkable anti-tumor efficacy but also overcame the multidrug resistance associated with PTX (Figure 2C).⁵¹

Micelles are spherical nanostructures created through the self-assembly of amphiphilic block copolymers in solution. They possess a hydrophilic shell that safeguards the drug and facilitates evasion of immune detection, together with a hydrophobic core that solubilizes poorly water-soluble medicines. Consequently, micelles are extensively utilized as vehicles for the delivery of insoluble medicinal substances.^82,83 Hu et al designed and synthesized a ROS-responsive, triple-modality mitochondria-targeted polymeric micelle (CAT/CPT-TPP/PEG-Ce6, referred to as CTC). The CTC micelles are self-assembled from two amphiphilic polymer chains and encapsulate CAT. Chain I consists of camptothecin (CPT)–cinnamaldehyde (CA)–PEG–TPP, enabling chemotherapy and ROS amplification, while Chain II comprises PEG–photosensitizer Ce6 for photodynamic therapy (PDT), with CAT augmenting PDT efficacy. The researchers found that CTC could induce immunogenic cell death (ICD), not only directly killing tumor cells but also effectively activating the host immune system against cancer, demonstrating significant antitumor effects (Figure 2D).⁵³

DQA

DQA is an amphiphilic DLC consisting of two positively charged, hydrophilic quinolinium heads connected by a lipophilic decamethylene (10–CH₂) chain. In aqueous environments, DQA can self-assemble into vesicle-like nanocarriers called DQA micelles (DQAsomes). DQAsomes represent the first vesicular nanocarrier system designed for mitochondrial targeting and can serve as carriers for delivering DNA and drugs directly to mitochondria.⁸⁴ Wang et al designed a targeting molecule, dequalinium–polyethylene glycol–distearoylphosphatidylethanolamine (DQA–PEG2000–DSPE) and decorated it on the surface of liposomes to co-deliver RSV to mitochondria. They found that this nanomedicine exhibited significant antitumor activity and could reduce drug resistance. Moreover, when combined with vincristine-loaded liposomes, it markedly enhanced anticancer efficacy against resistant cells.⁵⁷ While foundational studies established the potential of DQAsomes for delivering DNA and drugs directly to mitochondria, recent research (2018–2025) indicates that its delivery advancement has remained relatively incremental and its application scope narrower compared to TPP.^85,86 Nonetheless, DQA remains a valuable alternative for specific cargo types that benefit from its unique self-assembly properties.

Rhodamine 123

Rhodamine 123 is a lipophilic, positively charged fluorescent compound that, like TPP and DQA, can selectively accumulate in mitochondria driven by the ΔΨm. However, current studies indicate that Rhodamine 123 is primarily used to assess mitochondrial function, activity, and ΔΨm, with relatively few applications exploiting its mitochondrial-targeting capability for drug delivery.⁸⁷ Currently, Rhodamine B is more commonly used for drug delivery applications.⁸⁸ Li et al designed a rhodamine-based mitochondria-targeted iridium complex, [((η⁵-Cp*) Ir (rhod)]²⁺ 2PF₆⁻ (Ir-rhod), and found that Ir-rhod could effectively overcome cisplatin resistance. It synergistically induces ferroptosis and apoptosis, efficiently eradicating resistant lung cancer tumors in animal models with low toxicity. Moreover, its intrinsic fluorescence enables the integration of therapy and real-time monitoring.⁸⁹ Notably, numerous studies indicate that rhodamine–metal complexes have emerged as a promising approach for combined anticancer therapy and diagnostic applications.⁹⁰

Microenvironment Responsiveness

Compared with normal tissues, the TME demonstrates notable modifications, such as decreased pH, hypoxia, elevated levels of GSH and ROS, and overexpression of certain specific enzymes. Nanocarriers can be designed to respond to these tumor-specific microenvironmental features. This allows targeted drug delivery, controlled release, and the ability to overcome multidrug resistance. Meanwhile, it minimizes side effects on normal cells.⁹¹

A rising cohort of researchers is amalgamating stimuli-responsive nanocarriers with mitochondria-targeted entities, realizing a triadic hierarchical targeting technique that encompasses “tumor tissue → cancer cell → mitochondria.” This strategy ensures that drugs are released specifically at the tumor site. Such advanced drug delivery systems (DDS) enable more efficient and precise drug delivery while significantly reducing toxicity to normal tissues, representing a highly promising approach for cancer therapy.⁹²

pH Responsiveness

pH is the most widely studied endogenous stimulus for tumor drug delivery.⁹³ Rapid growth of solid tumors leads to hypoxia, which triggers the aberrant “Warburg effect,” producing large amounts of lactate. Additionally, tumors have abnormal vascular structures. These structures prevent the clearance of acidic metabolites. As a result, the pH of the tumor microenvironment is lower.⁹⁴ By exploiting the pH difference between the TME (pH ~6.5) and normal tissues (pH ~7.4), pH-responsive nanocarriers can be designed to achieve controlled, tumor-specific drug release. This strategy enhances therapeutic efficacy while minimizing side effects.⁹⁵ Currently, bond cleavage under acidic conditions and protonation of chemical groups are the two primary mechanisms underlying pH-responsive nanocarriers.⁹⁶ Zhang et al designed and synthesized a novel mitochondria-targeted conjugate, TPP–docetaxel (TD), which was incorporated into liposomes. The liposomes were composed of PEG–Schiff base cholesterol (a pH-sensitive material) and DSPE. Moreover, an EphA10 antibody was modified on the surface of the pH-responsive liposomes (EPSLP), forming the nanopreparation EPSLP/TD. EPSLP/TD exploits PEG–Schiff base cholesterol to target the acidic TME, the EphA10 antibody to target cancer cells, and TPP to target mitochondria. Through this triple-targeting strategy, precise drug delivery and controlled release were achieved. EPSLP/TD not only exhibited potent antitumor activity but also showed pronounced anti-angiogenic, anti-proliferative and pro-apoptotic effects.⁶⁰

In addition, the pH of the MM in cancer cells is approximately 8.0, which stands in sharp contrast to the neutral pH of the cytosol. It also differs greatly from the acidic pH of lysosomes and the TME.⁹⁷ Based on this property, alkaline-responsive nanocarriers can be rationally designed to achieve mitochondria-specific drug release. It is well established that acidic drugs are more soluble in alkaline than in neutral or acidic pH. Current studies focus on improving the solubility of encapsulated acidic drugs to enable mitochondrial pH-responsive release.⁴¹ Tan et al designed and synthesized a mitochondrial-targeted micellar system, CTPP-CSOSA/Cela, for controlled medication release in response to mitochondrial alkaline pH.In this system, CTPP, a DLC, confers mitochondrial targeting ability; CSOSA self-assembles into micelles as the nanocarrier; and Celastrol (Cela), a weakly acidic drug, serves as the therapeutic agent. At neutral cytoplasmic pH, Cela establishes modest electrostatic contacts with CSOSA, resulting in hydrophobic encapsulation within the micellar core and diminished premature leakage. Upon arriving at the MM, the alkaline pH environment increases the solubility of Cela and breaks its association with the micelles, hence initiating quick and significant drug release.⁶¹

GSH Responsiveness

In tumor tissues, GSH concentrations in the cytoplasm, extracellular microenvironment, and plasma are approximately 10, 1, and 0.1 mM, respectively, with cytosolic levels about fourfold higher than in normal cells.⁹⁵ Utilizing this characteristic, GSH-responsive disulfide-containing nanocarriers can be engineered for precise drug delivery and release.⁹⁸ Zhang et al designed and synthesized a pH/redox dual-responsive, cell membrane/mitochondria dual-targeting NPs. These NPs were fabricated through the self-assembly of TPP-grafted poly (ethylene glycol) (PEG)-poly (d,l-lactide) (PLA) copolymers (TPP-PEG-ss-PLA), utilizing a disulfide bond for redox responsiveness and TPP for mitochondrial targeting. In addition, the outer layer of these NPs was coated with chondroitin sulfate (CS). CS can mask the positive charge of TPP in the bloodstream, thereby prolonging the circulation time of the NPs. It also serves as a targeting ligand by binding to CD44 receptors overexpressed on cancer cells, enabling cell-specific targeting. Significantly, the protonation of carboxyl groups (-COO⁻ to -COOH) in the CS backbone in acidic environments induces a hydrophilic-to-hydrophobic transition, leading to shell dissociation and re-exposure of the underlying TPP.⁵⁹

ROS Responsiveness

Intracellular ROS levels in tumor cells are significantly higher than those in normal cells, and mitochondria are the primary organelles responsible for ROS production (~90%).⁴¹ Therefore, this characteristic can be exploited to design mitochondria-targeted, ROS-responsive nanocarriers. Such systems typically incorporate ROS-sensitive linkages, including thioether bonds, boronate ester bonds, and thioketal bonds.⁹⁹ Crucially, a distinction exists between cytoplasmic ROS (~0.1–1 μM) and the significantly higher concentrations within mitochondria, which dictates the choice of responsive chemistry. For instance, achieving cargo release specifically at mitochondrial ROS thresholds often requires high-affinity moieties like boronate esters, whereas thioketal bonds may be better suited for broader cytoplasmic responsiveness.¹⁰⁰ By targeting both cellular and mitochondrial pathways, these nanocarriers preferentially concentrate in tumor mitochondria, generate excessive ROS, and eliminate tumor cells. Zhang et al designed and synthesized a multi-prodrug nanoreactor (DT-PN) capable of simultaneously targeting cancer cells and mitochondria. A thioketal-linked prodrug monomer (CPTSM) was produced using CPT as a model drug, subsequently leading to the development of a cancer-targeting polymeric prodrug (cRGD-PDMA-b-PCPTSM) and a mitochondria-targeting polymeric prodrug (TPP-PDMA-b-PCPTSM). Their co-assembly in aqueous solution produced the dual-targeting nanoreactor DT-PN. Upon entering cancer cell mitochondria, low endogenous ROS triggers the release of CPT, which further enhances ROS production and propels an autocatalytic cycle of CPT → ROS → CPT. This self-amplifying loop results in a surge of medication release and ROS, effectively eliminating cancer cells.⁶³

Peptide-Mediated Mitochondrial Targeting

Peptide-mediated mitochondrial targeting is a strategy that employs specific short amino acid sequences (peptides) as targeting moieties. These peptides bind to receptors or characteristic molecules on the mitochondrial surface, enabling the selective delivery of loaded drugs to mitochondria within cells.¹⁰¹ Mitochondria-targeting peptides demonstrate affinity and membrane permeability, facilitating medication delivery across cellular and mitochondrial membranes into the MM. These peptides mainly include mitochondrial targeting signal (MTS) peptides, mitochondrial penetrating peptides (MPP), and Szeto–Schiller (SS) peptides.¹⁰²

Mitochondrial Targeting Signals (MTS) Peptides

Nearly 99% of mitochondrial proteins are encoded by nuclear genes and translated on cytosolic ribosomes. They are transported into mitochondria by translocase complexes, directed by MTS peptides.¹⁰³ MTS peptides (20–40 amino acids) exhibit an amphipathic α-helix structure, characterized by a hydrophobic side composed of uncharged residues and a hydrophilic side comprised of basic residues, which is crucial for mitochondrial import.¹⁰⁴ The cationic surface of MTS peptides interacts with TOM complex receptors on the OMM, facilitating protein translocation through the TOM channel into the intermembrane gap. The protein thereafter either integrates into the IMM or interacts with TIM23 on the IMM for translocation into the MM.¹⁰⁵ Thus, MTS peptides can selectively deliver proteins to the OMM, IMM or MM.¹⁰⁶ Although MTS peptides exhibits good biocompatibility as a natural targeting moiety, it has several limitations: (1) poor cell membrane permeability, preventing efficient entry into cells;¹⁰⁷ (2) the transported protein must remain partially unfolded or in a relaxed linear state, restricting the types of deliverable cargo;¹⁰⁸ (3) in cells with severely damaged mitochondria, the entire protein import machinery may be impaired, reducing targeting efficiency.¹⁰⁸ Therefore, its practical application is somewhat limited. Researchers have combined MTS with cell-penetrating peptides (CPPs) to enhance membrane permeability, facilitating effective plasma membrane traversal while preserving mitochondrial targeting.¹⁰⁹ Li et al designed and synthesized a mitochondria-targeted delivery system, P-D-R8MTS, utilizing an N-(2-hydroxypropyl) methacrylamide (HPMA) copolymer coupled with the hybrid peptide R8MTS (CPP R8 + MTS ALD5) connected to DOX. In tumor cells, acidic lysosomes activate P-D-R8MTS, facilitating the release of DOX-R8MTS from the HPMA backbone by pH-responsive cleavage. The R8MTS peptide subsequently guides DOX-R8MTS to mitochondria for targeted activity.⁶⁴

Mitochondrial Penetrating Peptides (MPPs)

MPPs (4–16 amino acids) are synthetic short peptides derived from CPPs. They combine cationic (eg, arginine, lysine) and lipophilic (eg, phenylalanine, cyclohexylalanine) residues, a configuration crucial for traversing both plasma and mitochondrial membranes.¹¹⁰ Due to the strong negative potentials of the plasma membrane and mitochondrial membranes, cationic MPPs can readily penetrate these barriers. However, since the IMM is more hydrophobic than the plasma membrane, ordinary CPPs lack sufficient lipophilicity to cross it. Thus, MPPs are engineered with strong lipophilicity for efficient IMM penetration and MM entrance, achieving high cellular uptake and targeted mitochondrial localization.^111–113 Compared with MTS peptides, MPP effectively overcomes poor cellular uptake. Yang et al designed and synthesized an MPP-modified DOX (MPP-DOX) conjugated to an HPMA copolymer (PM), and a nuclear-accumulating HPMA-DOX conjugate (PN) based on DOX’s inherent nuclear affinity. Co-delivery of these two polymers (PMN) enabled dual targeting of the nucleus and mitochondria. PMN exhibited significant anti-tumor and anti-metastatic efficacy, with PM mitigating metastasis and PN restraining tumor proliferation.Furthermore, the study demonstrated that simultaneous nuclear and mitochondrial damage holds promise for treating invasive tumors.⁶⁵

Szeto-Schiller (SS) Peptides

The SS peptide is a synthetic short peptide, typically fewer than 10 amino acids, with inherent antioxidant properties. It consists of alternating aromatic and basic amino acids, with a D-amino acid at position 1 or 2 carrying a +3 charge.¹¹⁴ The cellular uptake of SS peptides is ATP-independent, allowing entry into cells at physiological pH.¹¹⁵ Notably, SS peptides do not rely on the conventional membrane potential-driven mechanism to access the MM. Instead, they target mitochondria through electrostatic and hydrophobic interactions with cardiolipin on the IMM. This binding enables high local enrichment of SS peptides on the IMM surface rather than dispersing into the MM, thereby enhancing their antioxidant activity.¹¹⁶ SS peptides can act directly near the ETC. Their tyrosine or dimethyltyrosine residues serve as efficient electron donors, scavenging peroxynitrite and H₂O₂, and inhibiting lipid peroxidation. This reduces ROS levels, hence alleviating mitochondrial and cellular damage induced by oxidative stress.¹¹⁷ Due to their potent antioxidant properties, SS peptides are not utilized directly in cancer treatment. Instead, they serve to safeguard normal cells, alleviate cancer cachexia, and diminish chemotherapy toxicity.^118,119 Cancer cachexia is a multifaceted metabolic condition characterized by the increased breakdown of muscle and adipose tissue, resulting in irreversible weight loss. It impacts around 50% of individuals with advanced disease and constitutes up to 30% of cancer-related fatalities.^120,121 Ballarò et al shown that mitochondria-targeted SS-31 mitigates cancer- and chemotherapy-induced cachexia by reinstating SDH activity (a cardiolipin-dependent enzyme), conserving muscle ATP, and improving hepatic glucose/glycogen retention. SS-31 demonstrated optimal efficacy in intermediate cachexia, but offered minimal advantage in advanced or severe phases. The study preliminarily excluded negative impacts on tumor development or treatment, hence endorsing its safety profile. Overall, this work confirms that mitochondrial targeting with SS-31 is a viable and significant approach for cancer cachexia, emphasizing its dual function in antioxidant activity and metabolic enhancement, extending beyond simple antioxidant therapy.¹²²

Strategy of Nanomaterials Targeting Mitochondrial Metabolism

Inhibition of Energy Metabolism

Blockade of OXPHOS

OXPHOS is the primary mechanism via which mitochondria produce ATP. Considering that OXPHOS is significantly expressed in specific tumor cells, it can be therapeutically targeted with nanomaterials to obstruct this process. Nevertheless, the essential methods via which OXPHOS inhibition targets tumor cells vary.

(1) Targeted inhibition of OXPHOS to reduce the energy supply in tumor cells. Shen et al designed and synthesized a nanomedicine, PEG-Dendron-EPI@TPP-LND, which simultaneously inhibits OXPHOS and enhances chemotherapy. They first conjugated lonidamine (LND), an inhibitor of complexes I/II, with TPP to generate TPP-LND, enabling selective mitochondrial targeting to suppress OXPHOS and decrease ATP production in tumor cells. Subsequently, TPP-LND and the chemotherapeutic agent epirubicin (EPI) were co-loaded into a dendrimer-based nanocarrier. This combination chemosensitized TNBC tumors, markedly improving therapy efficacy (Figure 3A).¹²³

Figure 3 Strategies for Targeting Mitochondrial Energy Metabolism in Cancer Therapy Using Nanomaterials (By BioRender). Red upward arrows (↑) denote the induction of oxidative stress or the enhancement of therapeutic effects (eg, PDT↑, O2↑), whereas red downward arrows (↓) indicate the depletion of mitochondrial energy or the reduction of membrane potential (eg, ATP↓). T-shaped bars (⊣) represent the direct inhibition of mitochondrial respiratory chain complexes (I, II, or V) or key metabolic enzymes such as HK2 and PKM2. Other arrows point toward the ultimate biological outcomes, including Apoptosis, Cuproptosis, and ICD. Furthermore, all bolded terms (eg, Complete Energy Deprivation, Proteotoxic Stress) are used to emphasize the core therapeutic mechanisms and critical biological endpoints induced by the targeted mitochondrial treatment. The left panel represents Strategy 1: OXPHOS Inhibition, covering ATP depletion (A), apoptosis induction (B), and hypoxia alleviation for enhanced photodynamic therapy (C). The right panel represents Strategy 2: TCA Cycle Disruption, highlighting energy deprivation through cristae disruption (D) and the cuproptosis pathway (E). (A) Targeted inhibition of OXPHOS for chemosensitization. TPP-LND and epirubicin (EPI) co-loaded into dendrimer-based nanocarriers (PEG-Dendron-EPI@TPP-LND). TPP-LND selectively inhibits mitochondrial complexes I/II, decreasing ATP production and sensitizing TNBC cells to chemotherapy. (B) OXPHOS blockade for inducing mitochondria-mediated apoptosis. Hybrid membrane-camouflaged, ROS-responsive nanoparticles (HM-NPs@G) achieve the burst release of Gboxin in mitochondria. High-concentration Gboxin inhibits complex V, disrupting energy homeostasis and triggering programmed cell death in glioblastoma. (C) OXPHOS inhibition for tumor microenvironment (TME) remodeling. Metformin (Met) inhibits complex I to mitigate tumor hypoxia, thereby enhancing the generation of reactive oxygen species (ROS) during photodynamic therapy (PDT) and activating systemic anti-tumor immunity via immunogenic cell death (ICD). (D) Interference with the TCA cycle via multi-pathway energy blockade. The ZIAMH nanoplatform co-delivers Arenobufagin and ICG to inhibit HK2, PKM2, and LDHA, concurrently depleting substrates for the TCA cycle and inducing mitochondrial impairment through combined chemo-photothermal therapy. (E) Induction of cuproptosis through mitochondrial copper overload. The ROS-responsive CuET@PHF platform delivers excess copper ions to mitochondria, targeting lipoylated proteins (eg, DLAT) in the TCA cycle and causing proteotoxic stress. This process synergizes with photothermal therapy (PTT) to stimulate robust anti-tumor immune responses.

(2) Targeted inhibition of OXPHOS to induce mitochondria-mediated apoptosis. Yan et al designed and synthesized a cancer cell–mitochondria hybrid membrane–camouflaged, ROS-responsive NP loaded with Gboxin (HM-NPs@G). The hybrid membrane and ROS-responsive properties enabled burst release of Gboxin within mitochondria. High-concentration Gboxin inhibited complex V, disrupting energy metabolism and triggering mitochondrial death, thereby exhibiting significant efficacy against GBM (Figure 3B).¹²⁴

(3) Targeted inhibition of OXPHOS to relieve hypoxia in the TME. Zhao et al developed a novel nanotherapeutic platform, CyI&Met-Liposome (LCM), co-loading the photosensitizer CyI and metformin (Met). Met inhibits complex I, hence inhibiting oxidative phosphorylation and mitigating tumor hypoxia. The enhanced oxygen conditions facilitate CyI in effectively producing cytotoxic ROS, which directly eliminate tumor cells and augment PDT efficacy. Moreover, LCM remodeled the hypoxic TME, induced ICD, and activated immune cells, thereby strengthening anti-tumor immunity. Ultimately, LCM treatment significantly enhanced systemic immune responses, eradicated primary tumors, and effectively controlled tumor metastasis (Figure 3C).¹²⁵

Interference with the TCA Cycle

Tumor cells shift their TCA cycle function from primarily energy production to predominantly biosynthesis. Targeting the TCA cycle with nanomaterials can effectively inhibit tumor cell proliferation. Chen et al developed an intelligent nanoplatform, ZIF-8@ArBu&ICG@MPN@HA (ZIAMH), by sequentially incorporating Arenobufagin (ArBu, a chemotherapy/metabolic inhibitor) and Indocyanine Green (ICG, a photothermal agent) into Zeolitic Imidazolate Framework-8 (ZIF-8), encapsulating it with a metal–polyphenol network (MPN) self-assembled from iron and nitrocellulose, and applying a surface coating of HA.In ZIAMH, the combination of chemotherapy, CDT, and PTT impedes glycolysis and OXPHOS by inhibiting HK2, PKM2, and LDHA. The deficiency of substrates for the TCA cycle, coupled with mitochondrial impairment induced by CDT/PTT, further inhibits the cycle. This extensive energy blockade, incorporating many medicines, significantly improves antitumor effectiveness (Figure 3D).¹²⁶

In addition to disrupting energy metabolism by TCA cycle interference, the induction of cuproptosis via this pathway is becoming a prominent area of investigation in cancer research.

Cuproptosis is a type of programmed cell death induced by an excess of intracellular copper ions. The distinctive mechanism targets lipoylated proteins in the TCA cycle, particularly dihydrolipoamide S-acetyltransferase (DLAT) within the pyruvate dehydrogenase complex, as well as Fe-S cluster-containing proteins.This leads to proteotoxic stress and ultimately cell death. Ferredoxin 1 (FDX1) and Lipoic Acid Synthase (LIAS) are considered key genes regulating cuproptosis.^127,128 Xiao et al developed a ROS-responsive nanoplatform, CuET@PHF, in which copper–diethyldithiocarbamate (CuET) serves as a copper ion carrier. CuET is selectively delivered to tumor cell mitochondria, releasing high levels of copper ions and inducing cuproptosis. Additionally, the photothermal effect of CuET@PHF alleviates tumor hypoxia, creating a more favorable environment for cuproptosis. Moreover, the combination of cuproptosis and PTT synergistically induces ICD, enhances antitumor immunity, and markedly improves therapeutic efficacy against TNBC (Figure 3E).¹²⁹

Enhancing Oxidative Stress

Enhancement of ROS Generation

The mechanisms by which nanomaterials enhance mitochondrial ROS generation include: (1) serving as carriers to deliver ROS inducers (eg, DOX, cisplatin, 5-fluorouracil [5-FU], sorafenib, RSV,¹² etc.) directly to mitochondria; (2) photodynamic or sonodynamic therapy; and (3) Fenton or Fenton-like reactions.¹⁰⁰ Zhang et al designed a novel nanodrug delivery system, carbon-based nanomaterials (CBNs)-Pluronic F127-DOX (CPD). CPD generates ROS via DOX, which reduces ΔΨm, modulates apoptosis-related genes, and activates mitochondrial apoptosis, effectively killing tumor cells while mitigating DOX-induced systemic toxicity.¹³⁰

Additionally, ROS can be generated via PDT. Yang et al developed ISDN NPs through the carrier-free self-assembly of the photosensitizer ICG and the HSP90 inhibitor 17-DMAG.

Under 808 nm near-infrared (NIR) light, ICG produces ROS for direct tumor destruction and ICD, while 17-DMAG mitigates hypoxia-induced immunosuppression.¹³¹ Likewise, sonodynamic therapy (SDT) employs ultrasound-activated sonosensitizers to produce ROS, facilitating the destruction of tumor cells, particularly in deep-seated malignancies.¹³² Chen et al designed and synthesized a novel HER2-targeted piezoelectric NP, PGd@tNBs. Under ultrasound stimulation, the P(VDF-TrFE) piezoelectric material generates a piezoelectric potential that modulates the cell membrane potential, promoting Ca²⁺ influx and inducing mitochondria-mediated apoptosis. Simultaneously, SDT triggers abundant ROS generation, causing oxidative damage and synergistically enhancing apoptosis. Notably, PGd@tNBs incorporate Gd/Fe ions, enabling magnetic resonance imaging (MRI), which, combined with ultrasound imaging, provides the advantages of dual-modality imaging. This “theranostic” nanoparticle platform demonstrates significant potential for clinical applications.¹³³

Finally, ROS can also be generated via Fenton or Fenton-like reactions. Bao et al developed FMUP, a core–shell nanopreparation with an upconversion NP (UCNP) core and an Fe³⁺-1,3,5-benzenetricarboxylic acid (BTC)-FA metal–organic framework (MOF) shell. Upon 808 nm NIR irradiation, Fe²⁺ and H⁺ are released, acidifying the TME. The combination of endogenous mitochondrial H₂O₂, released Fe²⁺, and the acidic TME triggers a Fenton reaction, generating abundant ·OH and causing severe oxidative damage. Simultaneously, TME acidification induces Ca²⁺ overload. The synergistic effect of ROS overproduction and Ca²⁺ dysregulation ultimately leads to tumor cell death.¹³⁴ Beyond traditional oxidative damage, the elevation of mitochondrial ROS is intrinsically linked to the induction of ferroptosis. Notably, this process is not merely a consequence of ROS accumulation but is actively regulated by mitochondrial metabolic pathways. Recent evidence underscores that ferroptosis is intrinsically linked to mitochondrial OXPHOS and the TCA cycle. Studies have demonstrated that cystine deprivation-induced ferroptosis is not a passive process but actively requires mitochondrial TCA activity and the ETC to drive the accumulation of lipid hydroperoxides.¹³⁵ This mechanistic interplay suggests that targeting mitochondrial metabolism—specifically the OXPHOS pathway—represents a synergistic strategy to enhance the efficacy of ferroptosis-inducing nanotherapies.

Disruption of the Antioxidant Defense System

As the principal endogenous antioxidant, GSH directly neutralizes ROS and preserves the redox equilibrium of the mitochondrial membrane. Nanomaterials can diminish GSH levels, hence compromising this defense and hindering ROS elimination. Wan et al developed a disulfide-bond-driven nanosystem (CM-HSADSP@[PS-Sor]) that reduces GSH by dual mechanisms: sorafenib-induced synthesis inhibition and disulfide-triggered oxidation to Glutathione Disulfide (GSSG), consequently hindering ROS clearance.¹³⁶ Zhang et al developed a DOX-loaded iron-coordination polymer NP (PCFD) that includes a ROS-sensitive cinnamaldehyde (CA) derivative (TCA). In this system, CA diminishes GSH and enhances ROS, DOX triggers ROS-mediated cytotoxicity, and Fe³⁺ facilitates ·OH production while oxidizing GSH. Their combination establishes a self-amplifying reactive oxygen species loop, markedly enhancing anticancer activity.¹³⁷ Various types of nanomaterials designed for modulating oxidative stress in cancer therapy are summarized in Table 2.

Table 2 Nanomaterials Targeting Oxidative Stress Modulation for Enhanced Cancer Therapy

Targeting FAO

FAO is upregulated in many tumors. Nanomaterial-mediated targeting of FAO-related enzymes reduces FAO, obstructs energy supply, and lowers tumor proliferation.CPT1 is the rate-limiting enzyme of FAO, and amiodarone is an inhibitor of CPT1A.¹⁴² Saorin et al synthesized high-loading amiodarone liposomes using microfluidics, utilizing the FDA-approved Doxil^® lipid formulation. They demonstrated markedly improved efficacy compared to the free medication against various ovarian cancer cell lines.¹⁴³ Zhang et al designed a cascade-responsive 2-DG nanocapsule delivery system. The surface was functionalized with the glycolysis inhibitor 2-DG, while the core encapsulated CPT1C siRNA (siCPT1C) and an anti-VEGFR2 monoclonal antibody (av). These components respectively inhibited glycolysis, FAO, and angiogenesis. This combinational strategy showed significant therapeutic efficacy against GBM.¹⁴⁴ However, an increasing number of studies suggest that appropriately enhancing FAO to generate more ROS and induce apoptosis is a highly promising therapeutic approach. Wang et al developed a liposomal NP system (Ato/CQ@L), which co-encapsulated atorvastatin (Ato) and chloroquine (CQ). Ato, an adenosine monophosphate‑activated protein kinase (AMPK) activator, enhances ROS generation by upregulating CPT1 levels, while CQ, an autophagy inhibitor, reduces protective autophagy in tumor cells. This combined treatment significantly improves the therapeutic efficacy against drug-resistant TNBC.¹⁴⁵

The clinical orientation between these two strategies is largely dictated by the tumor’s baseline metabolic profile. For instance, FAO inhibition is highly effective in “FAO-dependent” malignancies such as GBM, certain subtypes of TNBC, and acute myeloid leukemia (AML), where fatty acid catabolism is the primary driver of survival and drug resistance. In these contexts, blocking FAO leads to rapid ATP depletion and metabolic collapse.¹⁴⁶ Conversely, enhancing FAO to trigger a lethal ROS burst is more suitable for tumors with compromised antioxidant defenses, such as hepatocellular carcinoma (HCC) and melanoma.^147,148 In these cases, the cells’ limited capacity to buffer mitochondrial superoxide makes them vulnerable to the metabolic “overload” caused by FAO over-activation. Thus, identifying the tumor’s “metabolic flexibility” and “redox threshold” is essential for selecting the appropriate FAO-targeted intervention.

Targeting Gln Metabolism

Considering the significant “Gln addiction” of tumor cells, the targeted delivery of nanomaterials to glutaminolysis enzymes (particularly GLS) is a viable therapeutic approach. Li et al developed dual-starvation metal-phenolic nanocapsules (CG@Cap) to enhance the therapeutic window of metabolic inhibitors.While the free glutaminase inhibitor telaglenastat (CB-839) faced challenges in Phase II clinical trials (eg, CANTATA and ENTRATA)^149,150 due to suboptimal pharmacokinetics, its encapsulation within ZIF-8 (CG@Cap) provides a promising strategy to overcome these limitations through targeted delivery. In this system, CB-839 effectively inhibits Gln metabolism, while surface-adsorbed glucose oxidase (GOD) slows glycolysis. Combined with a tannic acid (TA)-Cu coating that promotes cuproptosis, this multifunctional nanoplatform initiates ICD and stimulates antitumor immunity. The synergy between precise metabolic deprivation and cupric-induced cell death effectively overcomes the clinical hurdles of free drugs to suppress tumor proliferation.¹⁵¹ Chen et al developed a novel folate-targeted nanoplatform (FA-DCNPs) based on poly (d,l-lactide-co-glycolic acid)–polyethylene glycol–folate (PLGA–PEG–FA), co-loaded with 6-diazo-5-oxo-L-norleucine (DON) and CaCO₃. In this system, DON inhibits Gln metabolism, while Ca²⁺ overload disrupts intracellular calcium homeostasis. Moreover, suppression of Gln metabolism promotes the repolarization of pro-tumorigenic M2 macrophages toward the antitumor M1 phenotype, thereby remodeling the tumor immune microenvironment.¹⁵²

Targeting Ca²⁺ Homeostasis

Current research primarily focuses on using nanomaterials to target calcium overload, inducing mitochondrial collapse. This strategy disrupts ATP synthesis and activates mitochondrial-mediated apoptosis, thereby achieving therapeutic effects in cancer treatment.

Chang et al designed and synthesized a hyaluronic acid (HA)-stabilized, hollow mesoporous calcium peroxide NPs (HMCPN-CE@HA) loaded with calcium ion efflux inhibitor 5(6)-carboxyeosin (CE). These NPs can directly release Ca²⁺ to induce calcium overload, while CE inhibits calcium efflux, leading to irreversible accumulation of Ca²⁺ in the cell. This disrupts mitochondrial function and triggers calcium-mediated cell death. Additionally, the NPs can release H₂O₂ to generate ROS, and the mitochondrial damage caused by calcium death further releases iron ions, which catalyze ROS production. The combined effects of calcium and iron death lead to lipid peroxidation, resulting in ferroptosis.¹⁵³ Moreover, to maximize therapeutic efficacy, Yang et al designed and synthesized a multifunctional synergistic nanotherapeutic platform integrating chemotherapy, CDT, PTT, and calcium overload therapy, termed DOX-CuS@CaCO₃@PL-PEG (DCCP). This platform was constructed by loading DOX and chemokinetic/photothermal agent CuS NPs into CaCO₃ NPs, followed by coating with a phospholipid–polyethylene glycol (PL-PEG) shell. DOX exerts potent chemotherapeutic effects. Under near-infrared irradiation, CuS NPs markedly enhance PTT and further improve CDT efficiency. In addition, CaCO₃ decomposition releases large amounts of Ca²⁺, while excessive ROS generated during CDT disrupt mitochondrial calcium buffering capacity, leading to calcium overload and calcium-dependent cell death.¹⁵⁴

Applications of Nanomaterial-Mediated Mitochondrial Metabolic Regulation in Cancer Therapy

Chemosensitization

Chemotherapy is a common systemic treatment for cancer. Many chemotherapeutic agents induce DNA damage that ultimately triggers apoptosis through the mitochondrial pathway, thereby killing tumor cells. However, chemotherapy has certain limitations. It lacks specificity and is unable to differentiate between cancer cells and normal fast-proliferating cells, leading to severe side effects. (eg, bone marrow suppression, gastrointestinal reactions, and hair loss).¹⁵⁵ Moreover, chemotherapy readily induces multidrug resistance (MDR) in tumor cells. MDR, characterized by simultaneous resistance to three or more classes of distinct chemotherapeutic agents, accounts for over 90% of treatment failures.¹⁵⁶ Nanomaterial-based co-delivery methods improve tumor targeting and counteract drug resistance, facilitating efficient apoptosis at diminished doses for increased efficacy and reduced toxicity.¹⁵⁷

Inhibition of Energy Metabolism

The primary cause of tumor drug resistance is enhanced efflux of chemotherapeutic agents.¹⁵⁸ The ATP-binding cassette (ABC) transporter superfamily plays a central role in this process, among which P-glycoprotein (P-gp/MDR-1/ABCB-1) is the most important transporter.¹⁵⁹ Cancer cells enhance the expression of these transporters, utilizing ATP hydrolysis to actively expel medicines and sustain sub-toxic intracellular concentrations, thus promoting resistance.¹⁶⁰ However, previous studies targeting P-gp as a therapeutic approach have not been very successful (P-gp inhibitors exhibit high toxicity and low bioavailability).¹⁶¹ Consequently, energy metabolism may be identified as a strategic focal point. Targeting mitochondrial respiration to diminish ATP levels inactivates the ATP-dependent efflux pump P-gp, hence obstructing the evacuation of chemotherapeutics.¹⁶² Met is a commonly used anti-diabetic drug, primarily for type 2 diabetes (T2DM). However, an increasing body of research suggests that Met can also act as an anti-cancer agent and has significant potential in cancer treatment.¹⁶³ Met primarily acts by inhibiting complex I, leading to reduced ATP production and an increased AMP/ATP ratio.¹⁶⁴ It not only hinders P-gp-mediated efflux to mitigate chemoresistance but also stimulates AMPK, which subsequently inhibits mammalian target of rapamycin (mTOR) and curtails tumor development.¹⁶⁵ Notably, Met can also alleviate chemoresistance by directly suppressing P-gp expression.¹⁶² Huang et al designed and synthesized a novel self-assembled nanomedicine platform, PMDDH, for the co-delivery of DOX and Met. Met was conjugated with linear polyethyleneimine (PEI) to create bioactive PMet, which co-assembled with DOX-intercalated dsDNA into the PMDDH core and was surface-coated with HA to improve targeting and stability. The study demonstrated that PMDDH not only enhanced the antitumor efficacy of DOX but also mitigated its cardiotoxicity. Meanwhile, Met sensitized chemotherapy by activating the AMPK pathway, downregulating PD-L1 expression, and promoting ICD, making the combined treatment safer and more effective than DOX alone.¹⁶⁶

Lowering the Apoptotic Threshold

Numerous chemotherapeutics induce apoptosis in cancer cells through mitochondrial pathways. Anti-apoptotic proteins (eg, Bcl-2, Bcl-xL and Mcl-1) are located on the OMM. Cancer cells enhance the expression of these proteins to inhibit Cyt c release, hence reducing apoptosis, fostering survival, and facilitating treatment resistance.¹⁶⁷ Notably, the overexpression of Bcl-2 not only induces drug resistance in tumor cells but also promotes the progression, metastasis, and recurrence of various cancers.¹⁶⁸ Consequently, to surmount tumor treatment resistance, Bcl-2 inhibitors (sometimes referred to as BH3 mimetics, like Venetoclax) may be employed. These agents directly suppress Bcl-2 activity, relieve its inhibition on apoptosis, and thereby trigger the suicide program of cancer cells, enhancing chemosensitivity.^169,170 Yang et al engineered and fabricated a T22-peptide-modified, disulfide-crosslinked polymer micelle (TM). The micelle facilitated the C-X-C motif chemokine receptor 4 (CXCR4)-mediated co-delivery of the Bcl-2 inhibitor (venetoclax, VEN) and the FLT3 inhibitor (sorafenib, SOR). The resultant dual-inhibitor nanomedicine was designated TM-VS. VEN binds to Bcl-2 to obstruct pro-apoptotic connections and induce apoptosis, whereas SOR directly inhibits FLT3 to curtail proliferation. They simultaneously inhibit Bcl‑2/Bcl‑xL/Mcl‑1 and promote mitochondrial apoptosis. Additionally, the T22 peptide facilitates CXCR4-targeted transport to AML cells, while disulfide links provide GSH-responsive drug release. TM-VS demonstrated significant anticancer efficacy with negligible hematological and organ damage. This dual-inhibitor micelle attained reduced dose, minimized side effects, and improved efficacy, thereby ensuring safety and chemosensitization.¹⁷¹

Radiosensitization

Radiotherapy (RT) is a primary treatment for various cancers, and its efficacy largely depends on tumor radiosensitivity. RT directly inflicts damage on DNA by ionizing radiation (IR) and indirectly through ROS produced from water radiolysis. Clinical data indicate that over two-thirds of IR-induced DNA damage results from indirect effects.¹⁷² Moreover, tumor hypoxia and proficient DNA repair mechanisms contribute to radioresistance, diminishing the efficiency of RT.¹⁷³ Given that mitochondria are the primary site of endogenous ROS generation and mediate intrinsic apoptosis pathways, combined with the targeting capabilities of nanomedicines, radioresistance of tumor cells can be overcome by (1) inhibiting mitochondrial oxygen consumption to alleviate tumor hypoxia, and (2) increasing ROS production, thereby achieving radiosensitization.¹⁷⁴

Amelioration of Hypoxia

Hypoxia is a primary cause of radioresistance.¹⁷² After radiation energy deposition, O₂ must react with DNA free radicals to form irreparable peroxides, “fixing” and amplifying the damage. Under hypoxic conditions, this process is inefficient, allowing the damage to be repaired and reducing RT efficacy by 2–3 fold.¹⁷⁵ The administration of mitochondrial respiratory inhibitors using nanomaterials decreases tumor oxygen consumption, mitigates local hypoxia, and improves radiosensitivity. Zhou et al designed and synthesized IR-TAM@Alb NPs. Tamoxifen (TAM) was coupled with mitochondria-targeted heptamethine cyanine dye IR-68 to create IR-TAM, which then self-assembled with albumin to generate IR-TAM@Alb. TAM suppressed OXPHOS by suppressing complex I, thereby reducing tumor hypoxia, activating AMPK, and downregulating PD-L1/TGF-β, which combined enhanced DNA damage and T-cell infiltration. This approach reversed RT-induced immune tolerance and radiation-induced pulmonary fibrosis (RIPF), achieving radiosensitization.¹⁷⁶

ROS Amplification

The ROS-mediated indirect effect is the primary mechanism by which RT kills cancer cells.¹⁷² Therefore, artificially elevating ROS levels in tumor cells can overload their antioxidant defenses, exacerbate oxidative DNA damage, and achieve radiosensitization. Liu et al designed and synthesized a novel nanozyme, MDP, exhibiting both peroxidase (POD) and CAT-mimetic activities. MDP was synthesized by surface-conjugating MnCO and Ru enzymes onto TPP-targeted dendritic mesoporous silica nanoparticles (DMSN). Mn²⁺ liberated from MnCO facilitates both ROS production through Fenton-like processes and cGAS-STING-mediated antitumor immunity. Ru, as a high atomic number element, enhances local energy deposition, produces abundant secondary electrons and ROS, and thus provides physical RT sensitization. MDP employs Mn²⁺/Ru to replicate POD (transforming H₂O₂ into ·OH) and CAT (breaking down H₂O₂ into O₂), thereby producing oxidative stress while alleviating hypoxia. TPP-mediated mitochondrial targeting facilitates in situ ·OH production, enhancing mitochondrial death.¹⁷⁷ Zhao et al designed and synthesized a novel mitochondrial-targeted and protein sulfonation-reactive gold NPs (dAuNP-TPP). These NPs were formed by incorporating TPP and 1,3-cyclohexanedione (CHD) into AuNPs. AuNPs facilitate physical RT sensitization by the formation of secondary electrons and ROS, whereas CHD forms covalent crosslinks with high levels of protein sulfonation (PSA) in tumor mitochondria.TPP-mediated mitochondrial targeting and CHD anchoring facilitate the retention of AuNP in tumor mitochondria, directing AuNP-generated ROS to induce mitochondrial death. CHD fixation exacerbates mitochondrial function by depleting ATP, hence increasing cell death.¹⁷⁸

Synergistic Immunotherapy

Cancer immunotherapy is one of the most promising approaches in current oncology, working by mobilizing and activating the body’s own immune system to recognize, attack, and eliminate cancer cells. It functions either by stimulating ICD to activate the latent immune system,¹⁷⁹ or by directly augmenting the cytotoxicity of immune cells, as demonstrated by immune checkpoint inhibitors (ICIs) and chimeric antigen receptor T-cell (CAR-T) therapies.¹⁸⁰ Current studies indicate that mitochondria are essential for CD8⁺ T-cell functionality and the effectiveness of immunotherapy. Modulating mitochondrial metabolism can improve the outcome of cancer immunotherapy.⁴

Awakening the Immune System: Inducing ICD

ICD is a specialized form of cell death. It is characterized by the release of damage-associated molecular patterns (DAMPs), including calreticulin (CRT) exposure, ATP secretion, and high-mobility group box 1 protein (HMGB1) release.¹⁸¹ These signals stimulate adaptive immunity, enabling the particular identification and eradication of tumor antigens, hence fostering memory T cell development to ensure enduring protection against cancer recurrence.¹⁸² The ability of cancer therapies to induce ICD largely depends on their capacity to trigger ER stress and generate ROS.¹⁸³ As the main intracellular source of ROS, mitochondria can produce high levels of ROS to induce ER stress and initiate ICD, while also generating ATP to provide the energetic basis for this process. During ICD, ATP is released into the extracellular space, acting as a key DAMP molecule to activate immune responses.¹⁸⁴ Guo et al developed a pH-responsive nanocarrier, cRGD-mix Man-pRNC_{DEA+Thioether}/R848. It was based on an immunoactive triblock copolymer (P_{Thioether+DEA}). The copolymer self-assembled into a nanocarrier to encapsulate the immune adjuvant Resiquimod (R848). The NP surface was modified with cRGD and mannose (Man) as targeting ligands. P_{Thioether+DEA} facilitates direct mitochondrial targeting, stimulates OXPHOS, increases ROS levels, and upregulates gasdermin D (GSDMD) to initiate ICD. Concurrently, cRGD-conjugated R848 specifically targets melanoma (B16F10) cells for in situ tumor vaccination, whereas mannose (Man) targets tumor-associated macrophages (TAMs) and encourages M2-to-M1 polarization. cRGD-mix Man-pRNC_{DEA+Thioether}/R848 proficiently elicit ICD, dendritic cell (DC) maturation, TAM reprogramming, and cytotoxic T lymphocyte (CTL) infiltration, demonstrating considerable anticancer activity.¹⁸⁵ Recent studies have further revealed that targeting mitochondria-specific cell death pathways can directly induce ICD.¹⁸⁶ Chen et al designed and synthesized a novel OPDEA-PCL/ICT NP, composed of the copolymer poly(2-(N-oxide-N,N-diethylamino) ethyl methacrylate)-b-poly(ε-caprolactone) (OPDEA-PCL) encapsulating Icaritin (ICT). ICT can induce ICD via mitophagy, and the mitochondrial-targeting property of OPDEA-PCL enables ICT to act specifically on mitochondria, greatly enhancing its ICD-inducing capability.¹⁸⁷

Strengthening the Immune System: Activating and Empowering T Cells

Recently, cancer immunotherapy has progressed significantly. Strategies such as ICIs, adoptive cell treatment, and cancer vaccines have significantly enhanced survival rates in patients with refractory malignancies.¹⁸⁰ However, researchers have found that current immunotherapies benefit only a small subset of patients, with the majority showing no response.¹⁸⁸ A primary reason for restricted efficacy is the functional depletion of tumor-infiltrating lymphocytes (TILs). Mitochondria are crucial for T-cell activation, differentiation, and functionality as cellular energy metabolism centers.¹⁸⁹ Recent studies indicate that mitochondrial depletion and malfunction are significant contributors to T cell exhaustion. Targeting mitochondria may consequently rejuvenate T cell functionality and enhance the efficacy of immunotherapy.¹⁹⁰

Cancer cells can actively commandeer mitochondria from T cells by tunneling nanotubes (TNTs), enhancing their own metabolism and survival, while undermining the antitumor efficacy of T cells and facilitating immune evasion.¹⁹⁰ One technique to combat cancer immune evasion, based on nanotube-mediated mitochondrial hijacking, is to reduce mitochondrial content in cancer cells or augment it in T cells. A more direct strategy involves reversing the direction of mitochondrial translocation, relocating mitochondria from cancer cells to T cells. This may augment antitumor immunity and potentially boost the efficacy of immunotherapy.^191,192 Leng et al designed and synthesized a nanoplatform, NBP@PLGA-HA. It is composed of PLGA encapsulating nebivolol hydrochloride (N) and 3-bromopyruvate (BP), with HA surface modification. NBP@PLGA-HA not only obstructs tumor OXPHOS and glycolysis through N and BP, but also impedes TNTs synthesis in tumor cells with intraperitoneal L-778123 hydrochloride, therefore averting mitochondrial appropriation by CTLs.¹⁹³ Pan et al designed and synthesized a nanoinitiator, MitoNIDs. It consists of AuNPs modified with a mitochondrial membrane-binding peptide and an LC3-targeting peptide (targeting autophagosomes). MitoNIDs can selectively degrade mitochondria via autophagosomes within cancer cells. This reduces mitochondrial content and enhances CD8⁺ T-cell-mediated tumor killing. It also overcomes immune tolerance by increasing tumor immunogenicity.¹⁹⁴ Zhang et al designed and synthesized a hydrogel-based nanodrug delivery system, O-TMV@ABP. This system employs oxidized sodium alginate-modified tumor cell membrane vesicles (O-TMV) as carriers, encapsulating axitinib (AXT) within the lipid bilayer, while concurrently loading 4–1BB antibodies and PF-06446846 (a PCSK9 inhibitor) NPs in the hydrogel cavity. The 4–1BB antibody stimulates the T-cell PGC-1α pathway to enhance mitochondrial biogenesis, whereas AXT, a VEGFR inhibitor, alleviates hypoxia in the tumor microenvironment. They collaboratively mitigate T-cell depletion.¹⁹⁵

Metabolic reprogramming of tumor cells depends on aerobic glycolysis rather than OXPHOS, resulting in an immunosuppressive TME characterized by nutritional deprivation, acidosis, and hypoxia. This directly results in T-cell fatigue and diminished cytotoxicity. The activation, proliferation, and effector function of T-cells are significantly reliant on mitochondrial OXPHOS. Thus, tailored modulation of this system can ameliorate the TME, counteract T-cell depletion, and bolster antitumor immunity.¹⁹⁶ Xiong et al developed a mitochondria-targeted atovaquone prodrug (Mito-ATO), whereby TPP is conjugated to ATO through a decyl linker. ATO functions as an inhibitor of mitochondrial respiration. Mito-ATO specifically modulates energy metabolism in specific immune cells: it augments OXPHOS and glycolysis in CD8⁺ T cells, while inhibiting metabolism in granulocytic myeloid-derived suppressor cells (G-MDSCs). This dual impact amplifies anticancer immune cells while suppressing protumor immune cells, hence fostering antitumor immunity.¹⁹⁷ Yang et al synthesized ATO/CABO@PEG-TK-PLGA NPs utilizing PEG-thioketal-(poly (lactic-co-glycolic acid)) (PEG-TK-PLGA) as the carrier for the co-encapsulation of ATO and cabozantinib (CABO). ATO reduces oxygen consumption in tumor cells and alleviates TME hypoxia, whereas CABO, functioning as an MDSC scavenger, directly mitigates immunosuppression. This intervention decreased MDSC prevalence and augmented CTL efficacy, demonstrating significant anticancer and antimetastatic effects.¹⁹⁸ Mitochondrial metabolism regulation has emerged as a potent strategy to overcome drug resistance, with various nanomedicines designed for tumor sensitization and combination therapy summarized in Table 3.

Table 3 Applications of Nanomedicines Based on Mitochondrial Metabolism Regulation in Tumor Sensitization and Combination Therapy

Challenges and Future Perspectives

Optimization of Targeting Efficiency: Overcoming the Lysosomal Barrier

As discussed in Characteristics of Mitochondria-Targeting Nanomaterials, while DLCs like TPP leverage the hyperpolarized ΔΨm for mitochondrial entry (Figure 4C), the preceding intracellular journey remains a formidable challenge. Most nanocarriers are internalized via endocytosis, which frequently leads to their sequestration within lysosomes. If nanocarriers fail to achieve timely lysosomal escape, the acidic environment and proteolytic enzymes can degrade the payload or the targeting ligands, rendering the mitochondrial-targeting strategy ineffective (Figure 4A).⁹² Four mainstream mechanisms address this barrier: the proton sponge effect, pH-responsive membrane disruption, ROS-responsive escape, and CPP-assisted escape (Figure 4B).²⁰⁶ Future work should rationally integrate these modules to ensure efficient endolysosomal escape, while preserving the integrity of TPP-based targeting moieties.

Figure 4 Intracellular Journey and Endolysosomal Escape of Mitochondria-targeting Nanocarriers (By BioRender); all arrows indicate directional transport processes. The cross symbol (×) indicates payload degradation. (A) Conventional Pathway of Endocytosis and Lysosomal Degradation: This panel illustrates the standard internalisation process where TPP-functionalized nanoparticles enter the cell via endocytosis. Following internalization, the nanoparticles reside within early endosomes, which mature into late endosomes. These vesicles eventually fuse with lysosomes, where hydrolytic enzymes lead to the sequestration and degradation of the therapeutic payload.(B) Endolysosomal Escape Mechanisms: This panel highlights four primary strategies engineered to bypass lysosomal entrapment: (1) Proton Sponge Effect: Cationic polymers induce H+ and Cl− influx, leading to osmotic swelling and subsequent rupture of the endolysosome. (2) pH-Responsive Membrane Disruption: pH-sensitive polymers undergo conformational changes (from coiled to extended) in the acidic endolysosomal environment, causing membrane destabilization. (3) ROS-Responsive Escape: The presence of ROS triggers the cleavage of sensitive linkers, resulting in stimulated cargo release. (4) CPP-Assisted Escape: Cell-penetrating peptides (CPPs) facilitate direct membrane translocation or endolysosomal exit. (C) Mitochondria Targeting: Following successful escape from the endolysosomal system, the nanomaterials achieve targeted accumulation within the mitochondrial matrix, a process primarily driven by the high negative mitochondrial membrane potential (ΔΨm).

Comparative Analysis of Targeting Platforms: Efficacy vs Safety

A critical evaluation of the strategies in Characteristics of Mitochondria-Targeting Nanomaterials reveals a clear trade-off between targeting precision and biological safety. DLCs (eg, TPP), discussed in Tpp, remain the gold standard for mitochondrial entry due to their robust electrophoretic drive; however, their potential to depolarize membranes often leads to off-target toxicity in mitochondria-rich healthy organs. In contrast, peptide-based ligands offer superior biocompatibility but suffer from poor metabolic stability.²⁰⁷ From a structural standpoint, organic carriers (liposomes/micelles) feature excellent biocompatibility but limited drug loading and poor stability, while inorganic frameworks (eg, ZIF-8 in Targeting Gln Metabolism) offer high structural rigidity and high drug loading yet carry inherent toxicity risks and poor biodegradability.²⁰⁸ Ultimately, the transition from in vitro potency to in vivo success requires hybrid systems that balance the high affinity of DLCs with the controlled release and safety profiles of advanced nanocarriers.

Bridging the Gap: From Lab to Clinical Translation

Despite the innovative designs of ROS-responsive micelles (eg, CTC micelles) and synergistic platforms (eg, TPH/PTX), the clinical translation of mitochondrial-targeted nanomedicine is still in its infancy. A major hurdle identified in our review is the complexity of the TME, which can alter ΔΨm and drug release kinetics, leading to inconsistent therapeutic outcomes. To move beyond fundamental research, future efforts must focus on simplifying nanoplatform architectures while enhancing their “theranostic” capabilities—enabling real-time monitoring of mitochondrial accumulation to adjust dosages and mitigate side effects.

Summary and Outlook

In recent years, with the deepening of research on tumor metabolic reprogramming, the metabolic changes of mitochondria in tumor cells have garnered increasing attention. The indispensable roles of mitochondria in energy production, redox balance, and cell death regulation provide a fundamental theoretical framework for targeting mitochondrial metabolism. In this context, mitochondria-targeting nanomaterials with adjustable physicochemical characteristics and multifunctionality have emerged as essential instruments for accurate mitochondrial intervention. The rational design of NP size, surface charge, and modification with mitochondrial-targeting ligands (DLCs, MTS) or microenvironment-responsive structures can substantially improve mitochondrial targeting and therapeutic efficacy of nanomaterials. Based on this, various therapeutic strategies targeting mitochondrial metabolism have been developed. They focus on key pathways such as energy metabolism inhibition, oxidative stress regulation, FAO, and amino acid metabolism. These approaches exhibit superior antitumor efficacy compared with traditional delivery systems.

So far, nanomaterial–mitochondrial metabolism regulation systems have been widely studied and applied in the synergistic enhancement of chemotherapy, RT, and immunotherapy. These systems not only augment tumor treatment sensitivity but also assist in surmounting tumor heterogeneity and therapeutic resistance. Nonetheless, numerous significant obstacles persist in this domain. A significant number of nanomedicines remain in fundamental research and have yet to be utilized in clinical settings. Furthermore, challenges like as optimizing delivery efficiency, ensuring biosafety, and overcoming tumor therapeutic resistance require immediate attention.

Looking ahead, nanotherapy strategies targeting mitochondrial metabolism are expected to evolve towards higher precision, stronger controllability, and greater clinical relevance. On one hand, intelligent TME-responsive nanomaterials will be developed, and their synergistic mechanisms with various therapies explored, enabling more effective tumor suppression. On the other hand, further research will focus on theranostic nanoplatforms. These platforms enable drug delivery and real-time imaging feedback on mitochondrial drug accumulation and early therapeutic responses, achieving precise tumor treatment. Overall, nanomaterials targeting mitochondrial metabolism provide a precise and efficient treatment strategy for cancer, and are expected to play an important role in the diagnosis and treatment of tumors in the future.