Unlocking the potential of mitochondrial transplantation: overcoming challenges and paving the way for routine therapeutic application

Abstract

Mitochondrial medicine has shown great promise as a therapeutic approach for treating currently incurable diseases. Preclinical studies highlight its safety and efficacy, but significant challenges remain in translating these therapies from bench to bedside. Key unresolved issues include understanding the mechanisms behind the reparative potential of transplanted mitochondria, such as their viability and functionality in an extracellular environment, especially under elevated calcium ion concentrations. Additionally, challenges related to mitochondrial sourcing, delivery methods, and ethical considerations need to be addressed for broader clinical adoption. This review analyses these challenges and explores strategies to overcome them, including refining mitochondrial sourcing, delivery techniques, and storage solutions. We also emphasise the need for rigorous ethical guidelines and regulatory frameworks to ensure safe and global implementation, paving the way for mitochondrial medicine's broader clinical use.

Introduction

Mitochondria, the energy powerhouses of cells, evolved from ancient bacteria through endosymbiosis with eukaryotic cells. Over time, they became essential for various cellular functions, including metabolism, cell growth, survival, apoptosis, ferroptosis, calcium and iron homoeostasis, cell signalling, and redox balance. Emerging research is now exploring mitochondrial transplantation and transfer as a new therapeutic strategy. Recent efforts by Jonathan R. Brestoff and Keshav K. Singh have established a framework for mitochondrial transfer terminology, which may guide future research as this field rapidly advances (Brestoff et al. 2025).

Mitochondrial transfer occurs both under normal and pathological conditions, providing metabolic support to recipient cells and maintaining mitochondrial quality in donor cells, thus contributing to tissue remodelling and homoeostasis. Following the 2003 discovery of mitochondrial DNA (mtDNA) transfer in plants, numerous studies have evaluated the safety and efficacy of mitochondrial transplantation in vitro, in vivo, and in clinico (Neikirk et al. 2024).

However, significant challenges remain, including concerns about mitochondrial sources (autologous, heterologous, or xenografts), immunological reactions, mitochondrial uptake efficiency, delivery methods, and scaling. Ethical considerations also need to be addressed in future studies. Furthermore, the mechanisms by which transplanted mitochondria survive in the extracellular environment, especially in the presence of toxic calcium levels, are not well understood and warrant further research. Questions also remain about the optimal quantity of mitochondria required for best therapeutic effects.

This review highlights these key challenges and explores potential solutions to optimise mitochondrial transplantation, focusing on mitochondrial sources, delivery methods, uptake mechanisms, and ethical considerations, with the goal of making mitochondrial therapy a more feasible clinical approach.

Challenges in mitochondrial transplantation

Source of mitochondria

Identifying the ideal source of healthy, functional, and reparative mitochondria for clinical applications is a critical challenge in mitochondrial transplantation. The tissue selected should be easily accessible, minimally invasive to obtain, and rich in viable mitochondria, along with considering factors, such as the donor age, sex, and whether the source is autologous, heterologous, or xenogeneic (Roushandeh et al. 2019). Each of these factors significantly influences the quality and efficacy of the mitochondria used for therapeutic purposes. Several tissues and cell types have been studied as potential sources for mitochondrial isolation, including cell lines, stem cells, and various tissues, such as the liver, skeletal muscle, blood products like platelets, brain, placenta, heart, kidney, and adipose tissue. However, for clinical use, safety, and efficacy are paramount, and it is essential to avoid immunological rejection or other side effects. Figure 1 depicts the different sources of mitochondria and their advantages and disadvantages specially for clinical use.

Fig. 1.

Demonstrates the different sources for mitochondrial isolation. In addition, some benefits and limitations of these resources have been reviewed. In this picture, the resources vary from autologous, such as skeletal muscles or blood derived products, heterologous donors, such as cell lines and stem cell, and xenograft donor from pig tissues. It is noted that germ cells, such as sperms, oocytes and cumulus cells can be considered as valuable sources as well

Research supports the idea of using sources closely related to the damaged tissue, such as skeletal muscle-derived mitochondria for heart diseases or brain-derived mitochondria for ischaemic strokes. For instance, McCully et al. (2017) reported the successful use of autologous skeletal muscle-derived mitochondria for mitochondrial transplantation in heart disease, with positive outcomes like improved ventricular function and no significant adverse effects (Emani et al. 2017; Guariento et al. 2021). Although autologous mitochondria are generally preferred due to the low risk of immune rejection, there are limitations, the isolated mitochondria from patients with mitochondrial disorders or older individuals may be dysfunctional, rendering them unsuitable for transplantation. Additionally, repeated surgeries to harvest autologous mitochondria may not be feasible in certain clinical contexts, particularly for long-term treatments.

In such cases, heterologous mitochondria, derived from donors who are genetically similar (e.g., close family members) or from universal donor cell types, offer a potential alternative, especially when autologous sources are unavailable or unsuitable. Cell lines, such as mesenchymal stem cells (MSCs) and blood-derived products like platelets, are also promising sources for heterologous mitochondria due to their accessibility, ease of handling, and manageable quality control (Bamshad et al. 2023; Najafi-Ghalehlou et al. 2024; Pourmohammadi-Bejarpasi et al. 2020). However, this is costly, requires continuous supervision of the culture, as well as controlling for mitochondrial quality after increasing the number of cell passages (Bechet et al. 2025). Platelets have gained attention as a potential source due to their high mitochondrial content, ease of isolation, and relatively low risk of immunological issues. Our recent clinical trials have shown the safety and potential therapeutic benefits of platelet-derived mitochondria, such as improved cardiac function in STEMI patients (Baharvand et al. 2024).

In addition to autologous and heterologous sources, xenogeneic mitochondria, derived from animal sources, may also be considered as an alternative for certain conditions, such as cancer or mtDNA disorders. The advantages of xenogeneic transplantation stems from the ready availability and potential scalability of donor material. Although xenografts present the risk of innate immune response due to interspecies mtDNA incompatibility, and damage-associated molecular pattern (DAMPs)-induced inflammatory response, recent studies have demonstrated successful mitochondrial xenotransplantation without significant immune rejection or loss of mitochondrial viability in the short term. In detail, substantial evidence supports mtDNA as a functional ligand for TLR9, triggering immune activation, such that mitochondria contain various molecules that can function as DAMPs, which are recognised by Toll-like receptors (TLRs) on the immune cells and other pattern recognition receptors (PRRs). Subsequently, intracellular signalling cascades such as MyD88-dependent nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) activation (Yu et al. 2022), mitogen-activated protein kinase (MAPK) pathways, and interferon regulatory factors (IRFs) are initiated, resulting in the production of proinflammatory mediators like tumour necrosis factor alpha (TNF-α), interleukin-6 (IL-6), and interleukin-1β (IL-1β).

In regard to immune cells, following mtDNA release or delivery from donor cells to the endosomes of immune cells, TLR9 can be activated. TLR9 is characterised to be a PRR located within the endolysosomes of dendritic cells, macrophages, and other innate immune cells, which recognises unmethylated CpG motifs present in mtDNA and initiates downstream signalling pathways (Grazioli and Pugin 2018; Ma et al. 2024). For instance, engagement of TLR9 by mtDNA in neutrophils leads to the activation of NF-κB and MAPK pathways, promoting the release of proinflammatory cytokines, such as TNF-α and IL-6. Thus, mtDNA release or exposure as a result of mitochondria engulfment and degradation elicits immune responses similar to microbial DNA. In addition to TLR9, TLR4 is another surface receptor that is typically activated by bacterial lipopolysaccharide (LPS) as well as certain mitochondrial-derived molecules. One such ligand is cardiolipin, a unique phospholipid found in the inner mitochondrial membrane and can activate microglia via TLR4 when released extracellularly and stimulate the production of cytokines and chemokines, such as monocyte chemoattractant protein-1 (MCP-1) and interferon gamma (IFN-γ). Along with mtDNA, mitochondrial proteins like mitochondrial transcription factor A (TFAM) also act as DAMPs for TLR4. TFAM constitutes a high-mobility group box 1 (HMGB1)-like protein, which after being released from damaged mitochondria can engage TLR4 or receptor for advanced glycation end (RAGE), leading to TNF-α secretion by dendritic cells. Similarly, cytochrome c released from mitochondria has been shown to activate astrocytes through TLR4, initiating MAPK signalling and inducing the release of IL-1β and interleukin-8 (IL-8). Hence, these findings highlight the role of TLR4 as a key sentinel for mitochondrial DAMPs including lipids like cardiolipin and proteins like TFAM and cytochrome c, which converts mitochondrial stress signals into inflammatory responses.

In relation to non-immune cells, exposure of endothelial cells to exogenous mitochondria has been shown to induce a pro-inflammatory state. These cells upregulate adhesion molecules like ICAM-1, VCAM-1, and E-selectin, which facilitate leukocyte tethering and extravasation, which is a key step in immune cell recruitment and inflammation (Sun et al. 2013).

However, comprehensive long-term follow-up studies are needed to monitor any potential immune sensitisation, potential chronic inflammation, or functional integration in the long term, especially in the case of multiple transplantations or large-scale transplant settings (Bhattacharya et al. 2023; Kim et al. 2024; Ramirez-Barbieri et al. 2019).

Among the autologous, heterogeneous, and xenogeneic sources, there exists no one-size-fits-all solution and requires a situational approach as a guiding principle. In detail, for acute conditions like myocardial infarction, autologous skeletal muscle or platelet-derived mitochondria may offer rapid and safe intervention. In regard to chronic degenerative diseases or unsuitability of autologous sources, heterologous MSCs or platelets may be preferred due to lower immunogenic risk and scalable production. Xenogeneic mitochondria due to the absence of widespread reports may be used for investigational purposes in research or oncology paradigms.

In conclusion, while autologous mitochondria remain the gold standard when available and viable, a growing body of evidence supports the use of heterologous platelet- or stem cell-derived mitochondria as practical, safe alternatives in specific clinical contexts. Ultimately, the choice of mitochondrial source should be tailored to the disease, patient characteristics, and therapeutic time frame, with continued research needed to refine these criteria.

Another often underappreciated and yet critical determinant in mitochondrial source selection relates to the donor age, as preclinical studies have provided compelling evidence supporting the fact that mitochondrial function naturally tends to decline with ageing marked by diminished ATP production, impaired oxidative phosphorylation, promoted reactive oxygen species (ROS) generation, and demoted capacity to buffer metabolic stress (Mietsch and Hinkel 2021). In contrast, mitochondria derived from younger donors generally show superior functionality, including higher bioenergetic capacity, better antioxidant defence systems, and greater structural resilience against stress. Therefore, isolating mitochondria from younger donors may enhance the therapeutic potential of mitochondrial transplantation. Studies in animal models have shown that transplanting mitochondria from young donors into aged recipients can improve systemic metabolism, reduce oxidative stress, and even improve cognitive and motor functions (Javani et al. 2022; Zhao et al. 2020).

The sex of the donor, though often left unnoticed, constitutes another considerable factor in mitochondrial transplantation as suggested by emerging studies proposing significant sex-based differences in mitochondrial oxidative capacity, calcium handling, and resistance to oxidative stress (Ventura-Clapier et al. 2017). In other words, mitochondria from female donors may have enhanced antitumour activity, more efficient mitochondrial-nuclear interference, and more significant resistance to bioenergetic stress compared to those from male donors. Such discrepancies are partly believed to stem from maternal inheritance of mtDNA and the effect of sex hormones like estrogens, which have been discovered to oversee mitochondrial activity, biogenesis, dynamics, and oxidative stress responses. In other words, males and females display distinct metabolic preferences, such that males tend to utilise protein-based substrates, while females favour lipid metabolism. Furthermore, female-derived mitochondria are found to be less susceptible to apoptosis because of lower calcium accumulation. These differences influence energy production as well as cellular resilience and survival under stress or injury conditions and thus could have considerable implications in the context of disease-targeted mitochondrial transplantation, such as cancer treatment (Arjmand et al. 2023; Demarest and McCarthy 2015; Kubat 2023; Ventura-Clapier et al. 2017; Yu et al. 2021).

Additionally, cryopreserved tissues, such as the placenta, have been explored as a reliable source for mitochondrial isolation because of operational ease, biobanking potential, and broader donor accessibility. From a clinical perspective, while this approach maintains viable mitochondria, offering an alternative option where tissue harvesting is not plausible, cryopreservation can instigate critical challenges to mitochondrial quality. In detail, freezing and thawing may result in ice crystal formation and cellular dehydration, adversely affecting mitochondrial membrane potential, disrupting the electron transport chain, and promoting oxidative stress. Although several studies have reported successful isolation of viable mitochondria, such outcomes may be accounted for by the cryopreservation protocol, necessitating the development of optimised freezing protocols to ensure functional recovery post-thaw (Nakamura et al. 2020).

Myocyte-derived mitochondria, particularly from physically active individuals or those with a higher proportion of type I (oxidative) fibres with high mitochondrial density, are also considered a strong candidate for mitochondrial transplantation. These mitochondria, compared to those derived from glycolytic (type II) fibres, often manifest enhanced oxidative phosphorylation capacity, improved membrane potential stability, and reduced ROS production. These attributes indicate better energy production efficiency post-transplantation in targeting age-related or metabolic conditions like sarcopenia (Turkel et al. 2023). In detail, recent studies have identified a strong connection among impaired iron homoeostasis in muscle tissue, mitochondrial dysfunction, and systemic inflammation in older adults (Picca et al. 2020). This aligns with the elevated labile iron pool observed in the skeletal muscles of the elderly with reduced physical performance, which is linked to dysregulation in the expression of mitochondrial quality control (MQC) markers and increased mtDNA damage, underscoring the critical role of inefficient MQC in age-related muscle decline. Specifically, ageing is marked by several structural and functional changes in mitochondria, including increased mtDNA mutations, elevated oxidative stress, impaired fission–fusion dynamics, reduced mitochondrial number and function, and diminished mitophagy efficiency. These alterations collectively compromise mitochondrial integrity and cellular health, such that regular exercise has been shown to significantly enhance mitochondrial biogenesis essential for maintaining cellular energy homoeostasis (Liu et al. 2020).

Ultimately, determining the optimal source of mitochondria for transplantation depends on several factors, including the individual patient attributes, disease pathology, and therapeutic objectives. Despite the promising viability of mitochondria derived from various sources, namely autologous, heterologous, xenogeneic, and tissue-specific options for clinical applications, further research is needed to evaluate their long-term safety, efficacy, and immunogenicity. As the field of mitochondrial transplantation evolves, understanding the nuances of donor-wise variables, including age, sex, tissue origin, and donor compatibility, will be crucial for optimising therapeutic outcomes (Roushandeh et al. 2019). Table 1 provides a summary of in vitro, in vivo and in clinico studies that explored different sources for mitochondria.

Table 1.

Summary of in vitro, in vivo and clinical trials studies which applied mitochondrial transplantation

Type of study	Model/disease	Source of Mitochondria	Route of Delivery	Dosage	Grant/Clinical Trial Number	Trial Phase	References
In vitro	ρ0 cells	MSCs	Co-incubation	5 μg/mL	NA	NA	Kheirandish-Rostami et al. 2020
In vitro	PC3	MSCs	Co-incubation	5 μg/mL	NA	NA	Nikoo et al. 2023
In vitro	ρ0 cells	Fibroblast cells	Co-incubation	5 μg/mL	NA	NA	Takashi et al. 2020
In vitro	ρ0 cells	Fibroblast cells	Co-incubation	5 μg/mL	NA	NA	Roushandeh et al. 2020
In vitro	Melanoma model	BMMSCs	Co-incubation of mitochondria with CD8+ T	1 × 105–1.25 × 105 Mitochondria	NA	NA	Baldwin et al. 2024
In vivo	Acute renal failure	Skeletal muscle	Intra-arterial	7.5 × 106 mitochondria/mL	962,134	NA	Jabbari et al. 2020
In vivo	MCAO	Umbilical cord-MSCs	Intracerebroventricular	3 × 107 mitochondria	IR.GUMS.REC.1398.089	NA	Pourmohammadi-Bejarpasi et al. 2020
In vivo	Liver disease	Liver mitochondria	Intravenous	0.2–0.4 mg/kg daily for 7 days	cstc2018jcyjAX0612 cstc2019jcyj-zdxmX0035 cx2018086	NA	Zhao et al. 2021
In vivo	TBI	Umbilical cord-MSCs	Intracerebroventricular	3 × 107 mitochondria	400,032,407	NA	Bamshad et al. 2023
In vivo	TBI	Platelets	Intracerebroventricular	3 × 107 mitochondria	1,400,060,114	NA	Pourmohammadi-Bejarpasi et al. 2023
In vivo	Lung IRI model	Gastrocnemius muscle	left pulmonary artery	1 × 108 mitochondria	5 T32 HL 007734 5 R01 HL108107	NA	Moskowitzova et al. 2020
In vivo	Pulmonary hypertension	Skeletal muscle (soleus)	intravenous	100 μg mitochondria once weekly for 3 consecutive weeks	MOST 109-2314-b-650-007-MY2 MOST 109-2314-B-006-036 NCKUEDA 10710 EDPJ108033	NA	Hsu et al. 2022
In vivo	Spinal cord injury	Skeletal muscle (soleus) and PC12	Direct spinal cord injection	50, 100 and 150 μg	R21NS096670 2P30NS051220	NA	Gollihue et al. 2018a, b
In vivo	Bone defect	BMSCs	Direct injection/intraarticular	1 × 108 isolated mitochondria	82,201,031 82,201,062	NA	Wang et al. 2024b
In vivo	POI	MSCs	Intraperitoneal	200 μg twice a week for 3 weeks	ZK[2022]270 ZK[2021]459	NA	Liu et al. 2024
In vivo	Osteoarthritis	L6 rat cells	Direct injection/intraarticular	10 μg/50 μL volume twice weekly for three weeks	HI20C1496	NA	Lee et al. 2022a
In vivo	Hair loss model (aged animal with depilation)	Liver	Subcutaneous in dorsal skin	200 μg mitochondria/weekly/one month	106-CCH-IRP-101 08-CCH-IRP-003	NA	Wu et al. 2020
In vivo	Cerebral ischaemia–reperfusion	Cryopreserved mouse placenta	intravenous	100 μg protein	R01NS094756	NA	Nakamura et al. 2020
In vivo	Chronic ischaemic heart disease	hUMSCs	Oral administration via nitric oxide-releasing nanomotor	1 × 1010	IACUC-20200802	NA	Wu et al. 2024b
In vivo	Restraint stress	Brain	Intranasal	340 µg of mitochondrial	71,313	NA	Mafikandi et al. 2025
In vivo	Spinal cord injury	Skeletal muscle (soleus)	Direct spinal cord injection	100 μg	NCKUEDA10705 NCKUH-10904038	NA	Lin et al. 2022
In vivo	Dystrophin-deficient mdx	Skeletal muscle (quadriceps)	Intramuscular	1 μg mitochondrial protein/g body weight	20-75-10006 23-75-10006	NA	Dubinin et al. 2024
In vivo	LPS-induced inflammatory model	huMSCs (PN-101) and platelets	Intravenous	10 μg mitochondria	NRF-2016R1A2B4007640	NA	Yu et al. 2022
In vivo	MCAO		NA		AMUWEC20171288		Zeng et al. 2022
Clinical trial	Ischaemic heart	Skeletal muscle (rectus abdominis)	Direct epicardial injection	1 × 108 ± 105mitochondria	NA	NA	Emani et al. 2017
Clinical trial	Brain ischemia	Quadriceps femoris muscle	Femoral artery infusion	Not mentioned (from 0.1 g muscle)	NCT04998357	Phase I	Walker et al. 2024
Clinical trial	STEMI	Platelets	Intracoronary infusion	1.2 × 108 mitochondria	IRCT20210920052524N2	Phase I	Baharvand et al. 2024
Clinical trial	Ischaemic heart	Skeletal muscle (rectus abdominis)	Direct epicardial	1 × 108 to 1 × 109	NCT02851758	NA	Guariento et al. 2021
Clinical trial	Pressure wound	Vastus lateralis muscle	Direct into wound	7.5 × 108 mitochondria	E-56733164-202.99-217,154,006	NA	Taner et al. 2024
Clinical trial	SLSMDs	PBMCs	Co-incubation	0.88 mU	3428-16	NA	Jacoby et al. 2022
Clinical trial	Infertility/IVF	Ovarian cortex	Microinjection into oocyte	500 mt in 1–2 pL	NCT02586298	NA	Labarta et al. 2019

PBMCs peripheral blood mononuclear cells, SLSMDs single large-scale mitochondrial DNA (mtDNA) deletion syndromes, STEMI acute ST-elevation myocardial infarction, BMSCs bone marrow stem cells, POI premature ovarian insufficiency, LPS lipopolysaccharides, MFB medial forebrain bundle, IVF in vitro fertilisation, huMSCs human umbilical mesenchymal stem cells, MCAO middle cerebral artery occlusion, TBI traumatic brain injury, MSC mesenchymal stem cell, IR ischaemic reperfusion, NA not available/applicable

Delivery route of isolated mitochondria

The primary challenge in mitochondrial transplantation lies in the effective delivery of mitochondria to specific injured tissues or organs. The method of delivery is critical as it influences organelle uptake and determines how many of the transplanted mitochondria reach the target tissue. Various approaches for delivering isolated mitochondria into injured organs are available, depending on the study design. Preclinical and clinical studies have explored both direct and indirect routes of mitochondrial transplantation. Direct delivery methods include intramyocardial injections, injections into the kidney, brain, spinal cord, subcutaneous tissue in wound healing, nerve sheets, and others. Indirect delivery methods involve intra-arterial (IA) injections, intracoronary delivery through coronary arteries, intranasal (IN), intravenous (IV), intramuscular (IM), intraventricular cerebral (IVC) injections, and oral administration (Chernyak 2020; Cowan et al. 2016; Doulamis and McCully 2021; Emani and McCully 2018; Emani et al. 2017; Guariento et al. 2021; Hao et al. 2023; Hong et al. 2024; Javadpour et al. 2024; Jia et al. 2023; Kubat et al. 2021, 2024; Li and Mu 2025; Lightowlers et al. 2020; Luo et al. 2024; Masuzawa et al. 2013; McCully et al. 2016, 2017, 2022, 2023; Orfany et al. 2020; Ramirez-Barbieri et al. 2019; Ren et al. 2024; Riou et al. 2025a, b; Rossi et al. 2023; Roushandeh et al. 2019; Wu et al. 2024b; Yu et al. 2022; Zeng et al. 2022). Recently, our team proposed the use of the Mesogun for mitochondrial transplantation in burn patients or for wound healing (Pourmohammadi-Bejarpasi et al. 2023). Figure 2 illustrated the different delivery route of isolated mitochondria and highlighted some advantages and limitations of these methods.

Fig. 2.

Illustrates the different delivery routes for mitochondrial transplantation. All these methods have their pros and cons. In this figure, some advantages and disadvantages of these methods have been summarised. IVC Intraventricular cerebral, IT Intratracheal, IM Intramuscular, IV Intravenous, IA Intra-arterial, IP Intraperitoneal, SC Subcutaneous, IN Intranasal, CNS Central nervous system

Efficient mitochondrial delivery is crucial for successful transplantation, but current techniques are confronted by significant risks. Direct delivery into tissues like joints is feasible and minimally invasive but delivering mitochondria to deeper or more delicate organs like the heart, kidney, liver, brain, or spinal cord typically often requires surgical intervention or multiple injections. These procedures potentially present significant risks associated with local tissue damage, such that the needle or syringe injections can result in oedema, microvascular ruptures, or bleeding, especially in highly vascular or sensitive organs. Furthermore, infusing large volumes of mitochondria into confined anatomical spaces like the brain may cause oedema, intracranial pressure elevation and inflammatory responses. To overcome these challenges, future research should focus on improving delivery methods. In this context, image-guided injections like ultrasound and MRI can promote precision and demote off-target trauma. Furthermore, hydrogel-based mitochondrial encapsulation or biocompatible scaffolds allow gradual mitochondria release, mechanical disruption minimisation, and mitochondria viability preservation. Moreover, intravascular mitochondria delivery using catheter-based tools can empower targeted access to organs like the heart and kidney, saving the need for open surgery. Nanoparticle conjugation and mitochondrial surface modification may also introduce alternative options for enhancing the targeting specificity and tissue integrity conservation. Building upon these advancements, genetic engineering or mitochondrial reprogramming could address mtDNA incompatibility issues, improving the long-term functionality of transplanted mitochondria. McCully et al. (2017) introduced the concept of "end-organ homing", where transplanting mitochondria via direct or IA methods resulted in mitochondria localising to the target tissue (i.e., heart) without migrating to off-target sites. Using imaging techniques such as positron emission tomography (PET) and computed tomography (CT), the team found that mitochondria injected into the myocardium, coronary arteries, and even lung tissue via the pulmonary artery were retained in the heart and lungs, suggesting a phenomenon of end-organ homing. While the exact mechanism remains unclear, this finding is promising for clinical applications of mitochondrial transplantation (McCully et al. 2017).

Systemic injection through the veins is a common method for drug delivery, offering distinct advantages, such as ease of access to veins, feasibility for multiple injections, and minimal localised tissue disturbance. However, one of the main limitations of IV transplantation relates to the non-specific distribution of mitochondria throughout the body, which results in diluted mitochondrial concentration in the damaged tissue. Furthermore, high levels of circulating calcium may reduce the viability of healthy mitochondria administered intravenously, inducing calcium overload, and subsequently leading to the formation of permeability transition pores and disruption of mitochondrial integrity, as well as mitochondrial swelling, fragmentation, and apoptosis (Huang et al. 2023). In addition to efficacy concerns, potential safety risks of IV mitochondrial infusion need to be thoroughly considered. Infection is known to be a common risk with any IV procedure, which may arise as a result of inefficient mitochondria sterelisation (Hu et al. 2024). Additionally, the IV route can potentially lead to mitochondrial suspension aggregation or pathways activation and subsequently cause thromboembolism, especially in the case of glioblastom (Gonzalez-Delgado et al. 2023). Moreover, immune responses and inflammatory reactions can be elicited due to foreign mitochondrial proteins or DNA introduction, further exacerbating the risk of vascular complications. Despite these challenges, a recent study by Lee et al. (2022b) showed that exogenous mitochondria exhibit a remarkable capacity to preferentially migrate toward the damaged tissues. In a lung fibrosis model, mitochondria demonstrated a higher tendency to migrate to fibrotic lung tissue compared to the healthy control animals following systemic injection, offering hope for systemic administration of mitochondria in treating certain diseases. Although the mechanisms behind the preferential accumulation of transplanted mitochondria in injured tissues are not fully understood, evidence suggests this significantly greater accumulation in the damaged tissue can be accounted for by injury-induced signalling mechanisms, such as DAMPs, proinflammatory cytokines, ROS, and ATP depletion. In detail, injured tissues are known to release various chemotactic factors, such as ROS, cytokines, and DAMPs, which can create a biochemical gradient that guides mitochondrial migration. Moreover, injured cells may demonstrate elevated endocytic activity and upregulated surface receptors facilitating mitochondrial uptake. These factors collectively contribute to the selective recruitment and retention of exogenous mitochondria at the injury site. Therefore, albeit systemic delivery poses certain risks, IV constitutes a safe and reliable method for transplantation of mitochondria with injury-targeted homing capacity (Aoki et al. 2024; Hsu et al. 2022; Lee et al. 2022b; McCully et al. 2016, 2017, 2022, 2023; Mudgal et al. 2024; Neikirk et al. 2024; Phua et al. 2024; Sun et al. 2023; Xu et al. 2024a).

Delivery to the brain remains a specific challenge due to the blood–brain barrier (BBB). The BBB prevents many therapeutic molecules, including mitochondria, from entering the brain. Neurons have high energy demands and rely on functional mitochondria for their survival. In neurodegenerative diseases such as Alzheimer's, Parkinson's, and stroke, mitochondrial dysfunction is common. Mitochondrial transplantation offers a promising approach to restore neuronal function. Several studies have demonstrated that exogenously delivered mitochondria can successfully pass through the BBB, enter neurons, and restore damaged tissues, improving memory, learning, and sensorimotor functions, while reducing apoptosis (Bamshad et al. 2023; Chen et al. 2023; Dashkova et al. 2024; Ebadpour et al. 2024; Javadpour et al. 2024; Pourmohammadi-Bejarpasi et al. 2020; Riou et al. 2025a, b; Sun et al. 2024; Walker et al. 2024; Xu et al. 2024b; Zhang et al. 2022).

IN administration has gained popularity as a non-invasive method for delivering therapeutics, including mitochondria to the central nervous system (CNS), which is because this route bypasses the BBB via the olfactory and trigeminal nerve pathways and first-pass metabolism and consequently enables rapid delivery to brain tissue. Recent studies have shown that brain-derived mitochondria can exert therapeutic effects in animal models of stress, improving behaviour and reducing inflammation (Mafikandi et al. 2025). Despite the advantages of the IN route, several challenges include limited dosing volume and mucociliary escalator that may narrow delivery efficiency, which can critically affect the consistency and scalability of therapeutic outcomes.

IM injection is another widely used route for drug delivery, particularly in the context of musculoskeletal diseases. This method is easy to administer, minimally invasive, and supports targeted delivery to injured muscle tissues. In a recent study by Dubinin et al. (2024), the IM injection of healthy mitochondria into mdx mice models of Duchenne muscular dystrophy was observed to improve muscular strength, reduced skeletal muscle injury, and decreased serum creatine kinase levels, highlighting the potential of this method for treating myopathies. Despite the advantages of IM injection, the limitations in treating diseases that affect internal organs, such as heart, kidney or brain ischaemia. In detail, one major challenge relates to the limited therapeutic reach, as in restricted distribution of transplanted mitochondria from muscle tissues to distal organs. This is exemplified by the BBB in the case of brain ischaemia, which constitutes a formidable obstacle preventing transplanted mitochondria from passively crossing into the CNS. With regard to the heart and kidney, factors like their extracellular matrices, enzymatic activity, and immune surveillance can potentially interfere with the viability and uptake of mitochondria delivered from muscles located at distant sites.

In 2024, Wu et al. introduced a novel delivery method for mitochondrial transplantation using oral administration via a nitric oxide-releasing nanomotor. The mitochondria, protected by enteric capsules, were absorbed through the intestines and subsequently distributed to damaged cardiac tissue. This non-invasive method offers several advantages, including ease of administration, patient adherence, and exciting potential for repeated dosing in treating chronic conditions, such as ischaemic heart disease (IHD) by rebuilding cardiac energy metabolism and reducing myocardial fibrosis at the transcriptional level (Wu et al. 2024b). However, the challenges of this method relate to the potential mitochondrial membrane degradation in the gastrointestinal (GI) tract and limiting their bioavailability through interference with the gut microbiota and digestive enzymes.

Intraperitoneal (IP) injection is another potential delivery route that was used in a 2024 study to deliver MSC-derived mitochondria to a premature ovarian insufficiency (POI) animal model. The mitochondrial function was observed to be restored, inhibited oxidative stress, and promoted mitochondrial biogenesis (Liu et al. 2024). This method enables targeting organs near the peritoneal cavity as exemplified by the anatomical proximity of the ovaries to the peritoneal space. Thus, this delivery route may not be efficient for distant or less accessible organs inclusive of the brain, heart, lung.

Clinical studies have also explored various methods of mitochondrial delivery, such as epicardial injection and IA administration. McCully et al. demonstrated the direct delivery of mitochondria into ischaemic heart regions through epicardial injection, using a 1-mL syringe for 10–20 separate injections. This method allows the benefit of highly localised and precise mitochondrial delivery to the affected regions to enhance therapeutic efficacy but it is invasive and requires surgical exposure of the heart (Emani et al. 2017; Guariento et al. 2021). Less invasively, IA administration of mitochondria, such as in a recent clinical trial by our team, pertained to platelets-derived mitochondria injection into the coronary artery during angiography. This method provides targeted and efficient delivery to the target area without requiring open surgery to access the organ, although the necessity for arterial catheterisation and the use of contrast and radiation may limit its broader applicability (Baharvand et al. 2024). In a similar context, Walker et al. delivered mitochondria via IA infusion through the femoral artery to ischaemic brain stroke patients for the first time in a clinical setting. This method showed no adverse effects and provided easy access to organs like the heart, brain, kidney, and liver (Walker et al. 2024). The primary advantage of this method relates to ease of access to deep and vital organs through vasculature in a minimally invasive manner compared to the direct surgical approaches. However, the need for arterial catheterisation marks the necessity for contrast agents and radiation exposure and subsequently vascular injury, thrombosis, or embolism.

For superficial injuries, such as wounds and burns, subcutaneous administration of mitochondria constitutes an effective delivery method in a minimally invasive manner. In a chronic wound study, mitochondria were injected into the wound edges and base using insulin syringes, which was observed to promote healing (Taner et al. 2024). We have also demonstrated that administration of mitochondria to burn patients was associated with enhanced therapeutic outcomes. Mesogun is a clinical device that allows for the efficient delivery of large quantities of mitochondria to the skin (Pourmohammadi-Bejarpasi et al. 2023). The primary advantages of the subcutaneous route include ease of administration and localised delivery of mitochondria to the skin and peripheral tissue with minimal procedural risk. Therefore, this method would not be suitable for systemic or deep organ delivery due to restricted tissue penetration and distribution.

In vitro methods for mitochondrial delivery include co-incubation, centrifugation, microinjection, and mitoception, which are commonly used in cell culture environments (Kheirandish-Rostami et al. 2020; Nikoo et al. 2023; Roushandeh et al. 2020). In clinical trials, patients’ autologous CD34⁺ hematopoietic cells have been augmented by healthy mitochondria through co-incubation, leading to improvements in aerobic function and quality of life of patients (Jacoby et al. 2022). Moreover, in the context of cancer, recent research on CAR T-cell therapy has shown that co-incubating CD8⁺ T cells with exogenous mitochondria boosts mitochondrial function and enhances antitumor responses, even in the hostile tumour microenvironment (Baldwin et al. 2024). Microinjection has also been explored for oocyte mitochondrial transfer, though results have been inconclusive (Labarta et al. 2019). Table 1 is a summary of in vitro, in vivo, and clinical trial studies that used different routes in mitochondrial transplantation.

In closing, mitochondrial delivery routes differ in their invasiveness and translational potentials. Direct methods, such as intramyocardial or IM injection ensure site-specific delivery and high accumulation of mitochondria but are limited by being invasive and the risks of tissue damage. In contrast, although the indirect routes, including IV and IA are less invasive and more clinically scalable for a larger population, challenges relate to non-specific distribution and reduced targeting efficiency as well as vulnerability to systemic factors like calcium overload. However, injury-guided homing and end-organ retention phenomena suggest that such systemic routes can partially enable tissue specificity under pathological conditions. Alternatively, emerging techniques inclusive of hydrogel encapsulation, nanoparticle conjugation, and image-guided drug delivery offer promising avenues to enhance mitochondrial survival and delivery precision to protected sites, like the brain. Therefore, mitochondrial transplantation translation demands a holistic consideration of delivery efficiency, invasiveness, organ specificity, and patient safety across different diseases.

Ethical issues in mitochondrial transplantation

Exogenous mtDNA transfer

A key ethical issue in mitochondrial transplantation is the transfer of exogenous mtDNA. While using autologous mitochondria from the same individual raises fewer ethical concerns, the use of donor mitochondria from sources such as cell lines, stem cells, xenogeneic tissues, or blood introduces complications. Transferring foreign mtDNA may impact mitochondrial function and could be inherited by future generations if it integrates into the germline (Dimond 2015). Although mitochondrial transplantation does not involve nuclear DNA, which dictates personal traits, the transfer of mtDNA, even in small amounts, can still raise ethical dilemmas.

To address these concerns, using mitochondria from genetically related individuals, like mothers or siblings, could be a solution. If this is not feasible, haplotype matching might reduce differences between donor and recipient mtDNA. The issue of mtDNA haplotypes adds complexity, as these regions are linked to maternal lineages and influence various phenotypes and disease susceptibilities. mtDNA haplotypes affect mtDNA replication, gene expression, and cellular metabolism, potentially impacting transplantation outcomes. However, there is limited research on how different mtDNA haplotypes influence mitochondrial transplantation success, highlighting the need for future studies to explore their effects on mitochondrial function and patient health (Ali Pour et al. 2021; Dimond 2015; Fogleman et al. 2016; Zhang and Miao 2023).

Germline genetic interventions

A key ethical concern in mitochondrial transplantation is whether it constitutes germline genetic modification. Unlike somatic cell modifications, which affect only the individual, germline interventions can alter genetic material passed onto future generations. The introduction of foreign mtDNA into the germline raises concerns about potential unintended long-term effects, which are largely unknown. For example, mitochondrial replacement therapy (MRT), which relates to replacing defective mitochondria in a patient’s egg or embryo, has been legalised in the UK to prevent mitochondrial diseases, but such interventions are not yet approved in countries like the United States (Ali Pour et al. 2021). These regulatory differences highlight the need for global discussions and the development of ethical frameworks to guide mitochondrial transplantation (Dimond 2015).

Informed consent and donor issues

Informed consent is one of the most significant ethical challenges in mitochondrial transplantation. Unlike organ donation, which involves transferring whole organs from one individual to another, mitochondrial donation involves transferring mtDNA, an integral part of cellular function. It is critical that donors fully understand the implications of their donation, including the potential impact on the recipient's cellular integrity and any potential long-term genetic effects. Specifically, the introduction of foreign mtDNA into a recipient’s cells may have lasting effects on mitochondrial function, which could be passed down through the germline if incorporated into reproductive cells (Dimond 2015). Ethical concerns also arise when mitochondria are sourced from embryos, stem cells, or other genetic material, as this could lead to questions regarding the ethics of manipulating or creating embryos and the potential for exploitation. Additionally, it is essential to address the potential for unintended genetic alterations, such as the risk of “designer babies” or unforeseen long-term genetic consequences, particularly in cases where foreign mtDNA is integrated into the recipient's germline.

Equity and access to treatment

From a social equity perspective, there are several considerations about MRT. The exorbitant cost and limited availability of such an advanced technology may restrict ease of access for low-income populations and thus worsen health disparities. In other words, the availability of MRT currently being limited to a few countries with the necessary infrastructure and regulatory frameworks creates global partiality. Furthermore, ethical questions also arise regarding the priority-weighted allocation of resources and whether or not such technologies should be made barrier-free for everyone or remain limited to particular groups.

Autonomy and reproductive rights

MRT raises complex ethical questions about autonomy and reproductive rights. Although MRT can prevent the transmission of mitochondrial diseases from pregnant women, it also brings up concerns about the permissibility of germline modification and the autonomy of future generations inheriting altered mtDNA without consent. The creation of "three-parent" babies, where genetic material is derived from both parents and a mitochondrial donor, complicates these issues further. It raises questions about the traditional notions of parenthood and reproductive identity, highlighting the demand for comprehensive ethical regulations that acknowledge individual rights along with taking the broader societal impacts into account (Dimond 2015).

Regulatory oversight and safety concerns

Mitochondrial transplantation and replacement therapies involve advanced biotechnologies that raise significant safety concerns. Regulatory agencies like the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) play a vital role in ensuring that these therapies are safe for patients. However, the novelty and complexity of mitochondrial transplantation present challenges for these agencies in establishing appropriate safety standards and procedures. Issues such as immune rejection of transplanted mitochondria and the long-term effects of integrating foreign mtDNA into recipient cells must be carefully considered. Rigorous preclinical and clinical trials are necessary to evaluate the safety and efficacy of these therapies (Dimond 2015).

International regulatory harmonisation

A critical challenge in mitochondrial transplantation regulation is the lack of harmonisation across countries, as different nations have varying ethical standards and regulatory frameworks for genetic technologies. To ensure global patient safety and consistent ethical application, international collaboration among regulatory bodies is essential. A unified approach would help prevent legal loopholes and ensure equitable access to these therapies worldwide (Dimond 2015).

Despite the UK's leadership in legislating mitochondrial donation, ethical concerns remain, particularly regarding mitochondrial-nuclear incompatibility and the transmission of harmful mutations. Researchers have suggested using mitochondria from the same species or closely related individuals to reduce incompatibility risks. Additionally, the higher mutation rate of mtDNA, due to its lack of histone protection and inefficient repair mechanisms, raises concerns about potential risks. These issues highlight the need for clear ethical guidelines to ensure the safe and effective use of mitochondrial transplantation as a therapeutic strategy.

Dosage

Currently, no universally standardised dosage exists for mitochondrial transplantation and thus determining an effective and reliable dosage for mitochondrial transplantation is essential for treating diseases and requires further research. In detail, dosage involves pharmacokinetics and pharmacodynamics, considering factors like body weight, metabolic rate, and species differences. Mitochondria’s molecular size, variability, and potential immunogenicity add to the complexity of this challenge. Therefore, scalability, reproducibility, and cost-effectiveness are key to the success of mitochondrial transplantation as a therapeutic approach. In this sense, preclinical and clinical studies condition the dosage of transplanted mitochondria based on either the number of mitochondria or the protein content. However, one concern in this field is the vitality and the fate of transplanted mitochondria after the procedure. Furthermore, it remains unclear whether the exogenous mitochondria undergo fission and fusion with endogenous mitochondria, which may isolate the beneficial mitochondrial population. Hence, given that the half-life of mitochondria ranges from several days to weeks, a single injection is unlikely to offer long term protective benefits, especially in chronic or neurodegenerative diseases that span decades (Gollihue et al. 2018a, b; Vodičková et al. 2022). Based on current studies, particularly clinical trials, most investigations have applied a single dose of mitochondria. However, future studies should consider multiple doses to assess their potential efficacy. In this section, we will review in vitro, in vivo, and clinical trials that focus on mitochondrial dosage.

In vitro studies

The mitochondrial dosage in cell culture systems is often based on the protein content and is typically expressed in μg/mL of mitochondrial protein. For instance, we have co-incubated cancer cells with 5 μg/mL of mitochondria from hUC-MSCs, HUVECs, and HEK cells. These treatments made cancer cells more sensitive to ferroptosis and increased their susceptibility to cisplatin (Kheirandish-Rostami et al. 2020; Nikoo et al. 2023; Roushandeh et al. 2020). Similarly, for investigating the involvement of mitochondria in H2O2-induced ferroptosis and examining the molecules regulating ferroptosis, the exogenous mitochondria have been introduced to different cancer ρ0 cells with a concentration of 5 μg/mL (Takashi et al. 2020).

In vivo studies

The mitochondrial dosage in animal models is variable and is expressed in terms of either total protein (μg or mg) or absolute mitochondrial number. In this sense, in a recent study, Wang et al. demonstrated that transplantation of isolated mitochondria from 2 × 10⁵ cells of BMSCs improved angiogenesis in both in vitro and in vivo bone defect models. They also tracked the transplanted mitochondria and found them in the cytoplasm of recipient cells 5-, 10-, and 15-min post-transplantation (Wang et al. 2024b). For spinal cord injury, Jenna et al. isolated mitochondria from PC12 cells and rat soleus muscle using differential centrifugation. Exogenous mitochondria were injected at doses of 50 µg, 100 µg, or 150 µg into the injured spinal cord's medial and lateral grey matter. The study showed that exogenous mitochondria helped maintain the energy supply to the injured spinal cord in a dose-dependent manner, enhancing oxidative phosphorylation and antioxidative capacities. The study also revealed the presence of exogenous mitochondria in cells, such as microglia, endothelial cells, astrocytes, pericytes, and oligodendrocytes, all of which contributed to improved spinal cord function (Gollihue et al. 2018a, b; Wang et al. 2024a). Another study involving a 100-μg bolus of soleus-derived mitochondria directly transplanted into the injured spinal cord showed enhanced recovery in locomotor and sensory functions after 28 days. The transplanted mitochondria alleviated demyelination, cellular apoptosis, mitochondrial fragmentation, oxidative stress, and inflammation, leading to better spinal cord function (Lin et al. 2022). McCully and his team as pioneers in mitochondrial transplantation, especially for heart ischaemia–reperfusion, conducted an experiment showing that a small number of mitochondria can significantly alter organ function. They found that concentrations less than 2 × 10⁶ mitochondria per gram of heart tissue were associated with decreased cardioprotective efficacy, while concentrations greater than 2 × 10⁶ per gram did not yield additional benefits. In their studies, they used mitochondrial concentration ranges of 2 × 10⁵ to 2 × 10⁶ per gram wet weight, which equates to approximately 1 × 10⁹ mitochondria in the adult 400 g heart. Similar these concentrations were used for kidney and lung models (Arroum et al. 2024; Bechet et al. 2025; Celik et al. 2023; Doulamis et al. 2020, 2022; Doulamis and McCully 2021; Emani and McCully 2018; Emani et al. 2017; Guariento et al. 2021; McCully et al. 2016, 2017, 2022, 2023; Moskowitzova et al. 2020; Ramirez-Barbieri et al. 2019).

Despite these findings, some questions remain unanswered regarding the scaling and determination of mitochondrial dosage. For example, it is still unclear whether single or multiple mitochondrial transplants should be prioritised for initial translational studies. In the case of serial applications, the choice of disease model, animal size, and the metabolic demand of tissues must be carefully considered (Bodenstein et al. 2024). Some studies have explored multiple mitochondrial injections. For instance, Dubinin et al. transplanted two doses of mitochondria into the muscles of dystrophin-deficient mdx mice, which resulted in improvements in skeletal muscle injury, grip strength, and motor activity (Dubinin et al. 2024). In another study on POI, 200 µg of MSC-derived mitochondria combined with Pyrroloquinoline quinone (PQQ) were transplanted IP twice a week for three weeks. The treatment significantly improved ovarian function, antioxidant capacity, and mitochondrial biogenesis, and reduced follicle loss and apoptosis (Liu et al. 2024). Hsu et al. applied 100 μg of soleus muscle-derived mitochondria to treat pulmonary hypertension, administering the doses once weekly for three consecutive weeks. The results showed increased ATP production, improved right ventricle function, and restoration of right ventricular mass and wall thickness (Hsu et al. 2022). In osteoarthritis (OA) models, Lee et al. administered 10 μg of mitochondria per 50 μL volume into the knee joint of monosodium iodoacetate-induced OA rats twice weekly for three weeks. The treatment reduced cartilage destruction, bone loss, and inflammation, while enhancing mitochondrial function (Lee et al. 2022a). In addition, in liver disease, Zhao et al. transplanted mitochondria intravenously at two doses (0.2 mg/kg and 0.4 mg/kg) for seven days. They found that the mitochondria improved hepatocyte function, ATP production, and reduced free radical damage, with higher doses showing enhanced benefits (Zhao et al. 2021). Hair loss, often linked to mitochondrial dysfunction, was addressed in a study where liver-derived mitochondria were transplanted to aged mice to promote hair regrowth. Weekly transplantation of combination of 200 μg liver mitochondria and Pep-1 (P-Mito) for 1 month significantly improved hair regrowth compared to PRP, with the P-Mito group showing enhanced mtDNA and anti-aging marker expression (Wu et al. 2020). A study by Guariento et al. explored the outcomes of single and serial doses of mitochondrial transplantation in a donation after circulatory death (DCD) model. While both approaches resulted in improved myocardial function, oxygen consumption, and decreased infarct size, no significant difference was observed between the single and serial injections of 5 × 10⁹ mitochondria between groups (Guariento et al. 2020). Our team has also investigated various mitochondrial doses in both preclinical and clinical settings. We have applied both mitochondrial number and protein content to scale the dosage of mitochondria. For example, in acute renal ischaemia, we injected skeletal muscle-derived mitochondria at a concentration of 7.5 × 10⁶ mitochondria/mL into the arterial system. Our findings were promising and demonstrated the regenerative effects specially on renal tubules (Jabbari et al. 2020). In traumatic brain injury models, we transplanted 3 × 10⁷ hU-MSC-derived mitochondria into the lateral ventricle of rats (Bamshad et al. 2023). In this study, we increased the number of mitochondria because the route of delivery was through ICV and was different from our previous study on kidney where the isolated mitochondria have been transplanted through renal artery (Bamshad et al. 2023). In another preclinical study that conducted by our team, 3 × 10⁷ MSCs-derived mitochondria transplanted into the lateral ventricle of middle cerebral artery occlusion (MCAO) animal model (Pourmohammadi-Bejarpasi et al. 2020). In clinical trials, we transplanted 1.2 × 10⁸ platelets-derived mitochondria intracoronary in STEMI patients (Baharvand et al. 2024).

Clinical studies

In the context of clinical studies, mitochondrial dosage is predominantly reported in terms of mitochondrial count. McCully et al. used 1 × 10⁸ ± 10⁵ mitochondria for heart ischaemia in paediatric patients, while Guariento et al. applied 1 × 10⁸ to 1 × 10⁹ mitochondria for similar treatments (Emani et al. 2017; Guariento et al. 2021). In a clinical trial conducted by our team, we transplanted 1.2 × 10⁸ platelets-derived mitochondria into STEMI patients, finding that the mitochondria were well-tolerated, with no major adverse cardiac events (MACE) reported (Baharvand et al. 2024).

Taking the above studies into account, most in vitro studies administer 5–10 μg/mL of mitochondrial protein, whereas in vivo studies, the dosage ranges from 50 μg to 3 × 10⁷ mitochondria per injection in acute conditions and repeated dosing in chronic conditions. As for clinical studies, effective mitochondrial dosage clusters around 1 × 10⁸ mitochondria per patient. Thus, we recommend 1–2 doses of 50–150 μg or 10⁶–10⁷ mitochondria for preclinical acute models. For chronic models, we suggest repeated dosing over multiple weeks. As for clinical applications, ~ 1 × 10⁸ mitochondria appears to be both safe and effective for initial intervention. However, future studies need to explore dose–response relationships to determine the effective treatment range and long-term efficacy. Table 1 summarises the doses in in vitro and studies in vivo as well as clinical trials used in mitochondrial transplantation.

Mitochondria uptake after transplantation

One of the most significant challenges in the field of mitochondrial transplantation is the rate at which recipient cells take up the transplanted mitochondria, as well as the viability of these mitochondria once inside the cells. The first documented instance of naked mitochondrial uptake by cells was reported by Clark and Shay in Nature in 1982, marking a landmark moment in mitochondrial transplantation (Clark and Shay 1982; McCully et al. 2023).

Despite the promising potential, there have been varying and even controversial results regarding the mitochondrial uptake efficiency post-transplantation. Mitochondria are known to have a short survival span outside the cells and must remain viable and functional even in calcium-rich extracellular microenvironments. Additionally, the dynamic nature of mitochondria and their integration with intracellular organelles complicates their fate in recipient cells. Once inside the cell, the fate of these exogenous mitochondria is contingent on several factors, including the recipient cell type, mitochondrial viability, and the delivery method (Nakamura et al. 2024).

Following mitochondrial internalisation into target cells, various intracellular fates may occur. Some are degraded by lysosomes through mitophagy if they are recognised as damaged or foreign pathogens. Others successfully evade degradation with at least 80% of the transplanted mitochondria being linked to early and late endosomes and at least 70% of the endocytosed mitochondria fusing with the endogenous mitochondrial network and exchange mtDNA and membrane components (McCully et al. 2023). The literature also indicates that some mitochondria may undergo intercellular transfer, which relates to the mitochondria persisting independently within the cytoplasm and supporting cellular function without immediate integration (Borcherding and Brestoff 2023).

While some researchers have concluded that mitochondrial transplantation shows low efficiency, with only 3% to 7% of mitochondria successfully entering target cells, other preclinical studies paint a more optimistic picture. In certain models, up to 44% of donor mitochondria have been found to colocalise in recipient cells, such as cardiomyocytes, with the remainder residing in other cell types or the interstitial space (Bodenstein et al. 2024; Cowan et al. 2016; Kaza et al. 2017; McCully et al. 2023). Almost all studies highlight the efficacy of transplanted mitochondria for treating various diseases, underlining their ability to integrate into recipient cells. However, it is universally agreed that improving the efficiency of mitochondrial uptake is crucial for enhancing their therapeutic potential.

In order to achieve this, several strategies to enhance mitochondrial uptake include (I) conjugating mitochondria to cell-penetrating peptides (CPPs), such as P-Mito or transactivator of transcription (TAT) peptide to enhance membrane translocation and uptake (Koren and Torchilin 2012), (II) encapsulating mitochondria in liposomes or extracellular vesicles (EVs) to protect them from extracellular degradation and enhance endocytosis (Ekmekcioglu et al. 2024; Wu et al. 2024a), (III) and conjugating mitochondria with magnetic nanoparticles to guide them to specific tissues using an external magnetic field and increase their local delivery and uptake (Lin et al. 2022). In addition, ongoing efforts focus on developing artificial mitochondrial nanocarriers that mimic the biophysical properties of mitochondria and offer enhanced targeting and fusion.

Recent advances in artificial mitochondrial delivery methods

In order to address the restraints of conventional mitochondria delivery methods, recent advancements have concentrated on artificial mitochondria delivery modalities. One encouraging scheme pertains to mitochondria encapsulation within EVs like exosomes and microvesicles. EVs facilitate efficient cellular uptake by shielding mitochondria from immune detection and subsequent enzymatic degradation. Studies have demonstrated that EV-encapsulated mitochondria benefit from enhanced stability in circulation and improved bioenergetic rescue in models of neuronal and cardiac injury, compared to free mitochondria (Wang et al. 2025; Wu et al. 2024a).

Another novel and cutting-edge approach leverages nanomotor complexes for oral or systemic mitochondrial delivery, which is exemplified by a 2024 study by Wu et al. (2024b) introducing a nitric oxide-releasing nanomotor system that package mitochondria in enteric capsules, catapult mitochondria through the GI tract acid erosion, and facilitate their absorption into circulation following oral administration. Therefore, these nanomotors enhance mitochondrial penetration and retention in target tissues and emerge as a promising modality for chronic disease treatment due to their minimally invasive nature.

Collectively, such innovations portray a shift toward controlled and precision mitochondria delivery, which is capable of overcoming challenges like rapid degradation, poor targeting, and immune clearance. However, scalability, legal authorisation, and long-term safety remain areas for future research.

Storage and cryopreservation of mitochondria advances

The storage and cryopreservation of isolated mitochondria remain a significant challenge in mitochondrial transplantation. Mitochondria survive only about two hours on ice, limiting their use in time-sensitive procedures and thus the ability to effectively cryopreserve mitochondria could increase their therapeutic efficacy and enable batch preparation. Based on the sensitivity of the organelle to osmotic stress, ice crystal formation, and oxidative damage, cryopreservation techniques have been established for cells and tissues to optimise protocols and ascertain the effective cryoprotectants that preserve mitochondrial structural integrity and functionality. This is crucial for heterogeneous xenografts or when multiple doses are needed. (Nukala et al. 2006). The choice of cryoprotectant is crucial to prevent the formation of ice crystals during freezing, which can further harm the mitochondria. Common cryoprotectants, such as glycerol and dimethyl sulfoxide (DMSO), have been shown to protect mitochondrial structure and oxidative phosphorylation activity when used at a 10% concentration. Nukala et al. found that cryopreserved mitochondria retained intact inner and outer membranes, tight cristae, and higher levels of cytochrome c (Nukala et al. 2006). In a separate study, Gnaiger et al. used a 4-(2-Hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)–sucrose-based buffer containing antioxidants, ATP, histidine, colloidal agents, and cytochrome c at 0 °C to preserve mitochondria. This solution significantly improved mitochondrial function after short-term cold storage (up to 1 h). When rat heart mitochondria were isolated and stored for four days in a preservation solution, respiration rates remained above 50% of control levels, and mitochondria remained coupled (Gnaiger et al. 2000). Recently, researchers have explored trehalose as an effective component for mitochondrial storage. A buffer containing 300 mM trehalose, 10 mM HEPES, 10 mM KCl, 1 mM EGTA, and 0.1% BSA has been shown to exert cryoprotective effects. Yamaguchi et al. demonstrated that trehalose preserved mitochondrial morphology and cytochrome c levels after thawing, as well as biological functions such as ATP synthesis, calcium-induced swelling, transmembrane potential, and protein precursor import and processing (Yamaguchi et al. 2007). In a more recent study by Cloer et al., direct storage of isolated mitochondria at -80 °C in a trehalose buffer maintained both morphology and function even after one year (Cloer et al. 2023). This study suggests that freezing mitochondria in trehalose buffer at − 80 °C might be more effective than storing them on ice at 4 °C, especially in situations where long surgical procedures are required, as prolonged storage on ice can lead to damage to the mitochondrial membranes and a loss of ATP production capacity (Yamada et al. 2020).

Effective time management is crucial in mitochondrial transplantation, particularly for autologous, heterologous, and xenograft sources of mitochondria, as obtaining donor samples takes time. Freezing organs, such as the liver, could streamline the surgical process by allowing more time for mitochondrial sampling and quality assessment. A recent study by Heinze et al. introduced a new method for isolating an enriched mitochondrial fraction (EMF) from frozen tissue using a customised potter. This technique revealed minimal changes in trace elements (e.g., Mn and Cu) between frozen and fresh tissues, with only slight decreases in EMF iron. This method simplifies mitochondrial isolation, especially in tissues like the liver, which contain less connective tissue (Heinze et al. 2024).

Although cryopreservation and storage of mitochondria have demonstrated promising results, several key parameters still require optimisation. Gao et al. (2025) reported that mitochondria cryopreservation media optimisation resulted in cryopreserved mitochondria maintaining both structural and functional integrity post-thawing and restoring respiration and ATP production in recipient cells. Similarly, Giovarelli et al. (2023) optimised a cryopreservation protocol to safeguard mitochondrial morphology and membrane potential and respiratory capacity. Furthermore, the study by Riou et al. offer important insights into how pathological tau proteins impact mitochondrial function and intercellular transfer. Their findings, although not directly addressing cryopreservation, highlight the broader importance of maintaining mitochondrial integrity across various contexts, including during storage to preserve cellular viability and function (Aurélien Riou et al. 2025a, b). Hence, critical factors requiring further refinement include the choice of freezing protocols (slow vs. snap), optimal storage duration, thawing conditions, and post-thaw quality control. For clinical applications, additional considerations, such as the source of mitochondria are essential to ensure therapeutic efficacy.

Conclusion

Mitochondria, once known as the cellular powerhouse, are now being explored as a potential treatment for incurable diseases. While preclinical studies show promise, challenges remain in translating mitochondrial transplantation into clinical practice. Issues include ensuring mitochondrial viability in high calcium environments, accessibility of donor mitochondria, and scaling methods. Future studies should focus on large animals, organoids, and patient-derived cultures, alongside more clinical trials. Ethical guidelines will be crucial for safe, global application. Figure 3 provides an overview of the challenges and opportunities in mitochondrial transplantation.

Fig. 3.

Provides a comprehensive overview of the challenges and opportunities associated with mitochondrial transplantation. Key factors, such as ethical considerations, dose escalation, mitochondrial storage, and cryopreservation, as well as safety and efficacy, are recognised as significant hurdles in the advancement of this therapeutic approach. However, these challenges also present valuable opportunities for further research aimed at developing solutions to overcome these limitations. Finally, these investigations will facilitate the translation of current bench-side studies into clinical applications, paving the way for future therapeutic advancements

Abbreviations

HEPES

4-: 2-Hydroxyethyl-1-piperazineethanesulfonic acid

BBB

Blood–brain barrier

CPP

Cell-penetrating peptide

CNS

Central nervous system

CT

Computed tomography

DAMP

Damage-associated molecular pattern

DCD

Donation after circulatory death

EMF

Enriched mitochondrial fraction

EMA

European Medicines Agency

EV

Extracellular vesicle

FDA

Food and Drug Administration

GI

Gastrointestinal

DMSO

Glycerol and dimethyl sulfoxide

HMGB1

High-mobility group box 1:

IFN-γ

Interferon gamma:

IRF

Interferon regulatory factor:

IL

Interleukin

IA

Intra-arterial

IM

Intramuscular

IN

Intranasal

IP

Intraperitoneal

IV

Intravenous

IVC

Intraventricular cerebral

LPS

Lipopolysaccharide

IHD

Ischaemic heart disease

MACE

Major adverse cardiac events

MSC

Mesenchymal stem cell

MCAO

Middle cerebral artery occlusion

mtDNA

Mitochondrial DNA

MQC

Mitochondrial quality control

MRT

Mitochondrial replacement therapy

TFAM

Mitochondrial transcription factor A

MAPK

Mitogen-activated protein kinase

MCP-1

Monocyte chemoattractant protein-1

NF-κB

Nuclear factor kappa-light-chain-enhancer of activated B cells

OA

Osteoarthritis

PRR

Pattern recognition receptor

P-Mito

Pep-1

PET

Positron emission tomography

POI

Premature ovarian insufficiency

PQQ

Pyrroloquinoline quinone

ROS

Reactive oxygen species

RAGE

Receptor for advanced glycation end

TLR

Toll-like receptor

TAT

Transactivator of transcription

TNF-α

Tumour necrosis factor alpha

Author contributions

AMR and MHR: Wrote the main manuscript text. AMR and KT prepared figures, KT, YK, NNG, TS and MHR; Writing—review & editing and Investigation, MHR: supervision. All authors reviewed the manuscript.

Funding

This research was not supported by any specific grants from public, commercial, or non-profit funding agencies.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethical approval

Not applicable.

Research involving human and animal participants

This article does not involve any studies with human or animal subjects conducted by the authors.