Emerging Landscape of Cell-Penetrating Peptide-Mediated Organelle Restoration and Replacement

Abstract

Organelles are specialized subunits within a cell membrane that perform specific roles or functions, and their dysfunction can lead to a variety of pathophysiologies including developmental defects, aging, and diseases (cancer, cardiovascular and neurodegenerative diseases). Recent studies have shown that cell-penetrating peptide (CPP)-based pharmacological therapies delivered to organelles or even directly resulting in organelle replacement can restore cell function and improve or prevent disease. In this review, we summarized the current developments in the precise delivery of exogenous cargoes via CPPs at the organelle level, CPP-mediated organelle delivery, and discuss their feasibility as next-generation targeting strategies for the diagnosis and treatment of diseases at the organelle level.

1. Introduction

Organelles (e.g., mitochondria, lysosome, endoplasmic reticulum (ER), and Golgi apparatus) are membrane-enclosed/engulfed specialized compartments embedded within the cytoplasm and perform specific roles within a cell and their dysfunction can lead to a variety of pathophysiologies including developmental defects, aging, and diseases (cancer, cardiovascular and neurodegenerative diseases).^1,2 Therefore, developing pharmacological therapeutics for functional repair or replacing disordered organelles are urgently needed to restore cellular function, improve disease diagnosis, or even cure, stop, or prevent organelle-associated disease.

In the past few years, mitochondrial transfer via microinjection from donor to recipient (fertilized oocyte) has been successfully applied in human-assisted reproduction to save patients who have repeated pregnancy failures,^3,4 and this therapy has been used to promote precision medicine for the next generation of organelle level. However, as we noted, this microinjection-based approach is not feasible for nongermline cell-difunctionalized disease. Therefore, there is an urgent need to find an efficient strategy to conduct organelle targeting and replacement.

As a delivery tool, cell-penetrating peptides (CPPs) have been developed and extensively used in animal models for the treatment of various diseases. Some of CPP-coupled agents have entered phase I or II clinical trials, and some have also entered phase III.⁵ Research in CPP-mediated organelle targeting and organelle replacement is still relatively new, and two potential scenarios of this field can be considered depending on the degree of organelle damage.^6,7 One scenario is to achieve precise-targeting CPP-conjugated exogenous cargos (e.g., enzymes, DNA/RNA, antisense oligonucleotides (ASO), or peptide nucleic acids (PNA)) at the organelle level, including lysosomes,⁸ mitochondria,⁹⁻¹³ endoplasmic reticulum,¹⁴ extracellular vesicles,¹⁵ exosomes,¹⁶ and Golgi apparatus. Another scenario such as injury, damage, disease, or dysfunction leads to the majority of organelles within eukaryotic cells becoming irreparable; therefore, direct replacement of organelles (mitochondria,¹⁷⁻²⁰ lysosomes,²¹ and exosomes^22,23) may be the last way to restore cellular function. All of this progress has started to offer valuable insights and provides a therapeutic avenue of CPP-based organelle targeting and replacement precision medicine.

To generate widespread interest and discussion, the latest discoveries, recent advancements made by research groups, challenges, and future perspectives need to be summarized. Based on our interest and expertise of CPPs and mitochondria, our review article focuses on the current progress in CPP-mediated organelle targeting and organelle delivery. This will help readers to understand this new targeted therapeutic tool in a comprehensive manner and apply organelle-targeted therapies to more relevant disease areas.

2. Organelle-Associated Diseases

2.1. Mitochondria

Mitochondria are an essential organelle found in most eukaryotic cells, consisting of two separate and functionally distinct mitochondrial membranes. The outer mitochondrial membrane (OMM) and the inner mitochondrial membrane (IMM) encapsulate the intermembrane space (IMS) and matrix compartments which contain mitochondrial DNA (mtDNA) and proteins. mtDNA encodes only 13 proteins, which are essential components of the oxidative phosphorylation (OXPHOS) machinery producing aerobic ATP.²⁴ The nuclear genome (nDNA)-encoded mitochondrial proteins that are synthesized in the cytosol are actively imported and sorted into the mitochondrial compartment. Mitochondria play a crucial role in the energy supply for the cells and act as the central control points for the regulation of cellular differentiation, cell signaling, and apoptosis.²⁵ Mutations in the mtDNA, nDNA, as well as nongenetic causes such as environmental damage can cause a group of diseases called mitochondrial disorders, including metabolic disease, neurodegeneration, heart failure (HF), and cancer.^25,26 Most genetic mitochondrial disease might be transmitted multigenerationally via maternal mitochondrial DNA¹ (Figure 1).

Figure 1.

Organelle-related diseases and associated biological mechanisms. Studies have reported that infection, metabolic disease, neurodegenerative disorders, cancer, cardiovascular and autoimmune diseases are associated with organelles such as Golgi apparatus, lysosomes, mitochondria, and ER through different biological processes.

2.2. Golgi and Endoplasmic Reticulum

Endoplasmic reticulum and Golgi apparatus are important intracellular membrane-bound organelles in mammalian cells. The ER mediates the synthesis, modification, and processing of intracellular proteins, while Golgi receives proteins from the ER and enables further processing, packaging, and sorting of mature proteins and lipids into vesicles for transport to their target destinations.^27,28 Hence, the Golgi–ER network is an important network involved in a wide range of biological processes and multiple functions within the cytoplasm in eukaryotic cells. Several disorders such as cancer, autoinflammatory, and neurological diseases are closely associated with ER and Golgi dysfunction²⁹⁻³² (Figure 1).

2.3. Lysosome

Lysosome is a single membranous, subcellular organelle that is highly acidic and stores approximately 60 different types of hydrolytic enzymes capable of degrading various biological macromolecules. Cargoes reach the lysosome via exogenous phagosomes or endosomes and endogenous autophagosomes. The involvement of lysosomes in various cellular processes, such as cholesterol homeostasis,³³ plasma membrane repair,³⁴ cell signaling,³⁵ immune response,³⁶ and cell death,³⁷ is a critical component of these cellular processes and is an attractive therapeutic target. Lysosomal dysfunction can lead to abnormal storage of undegraded substrates in endosomes and lysosomes, eventually triggering a group of lysosome storage disorders (LSDs), including autoimmune diseases such as lupus, rheumatoid arthritis (RA), and multiple sclerosis (MS), neurodegenerative disorders like Alzheimer disease (AD), Parkinson’s disease (PD), and Huntington diseases,^38,39 and cancer⁴⁰ (Figure 1).

2.4. Ribosome

Ribosomes are macromolecular complexes comprising small and large ribosomal subunits that are constructed, found in eukaryotic cells as well as bacterial cells, as the sites of biological protein synthesis.⁴¹ Given the fundamental role of mRNA translation, it is not surprising that disruption of this process represents one of the major strategies for the development of therapeutic agents, such as anticancer drugs used to inhibit the growth of tumor cells.^42,43

2.5. Cell-Penetrating Peptides

CPPs are a diverse group of long peptides of 4–30 amino acids capable of delivering bioactive cargos of proteins,⁴⁴⁻⁴⁶ peptides,^47,48 nucleic acids,⁴⁹⁻⁵³ nanoparticles,⁵⁴ and pharmacologic agents into the cellular cytoplasm through different internalization mechanisms mainly including energy-dependent endocytosis and energy-independent direct penetration.^5,54−56 Since the discovery of the first qualitative CPP-TAT,⁵⁷⁻⁵⁹ increasing research activities and preclinical evaluations of CPP-derived therapies have provided us with promising results in various disease models.^5,60 Thus, these breakthroughs bring new prospects for the development of CPP-derived therapeutics to treat human diseases related to intracellular organelles.

3. CPP-Based Organelle Functional Restoration

One of the effective treatments for various organelle-related diseases is the development of drugs that specifically act on specific intracellular organelles. Both invasive and noninvasive strategies have been used to facilitate the delivery of therapeutic agent to intracellular organelles. Invasive techniques such as electroporation, biolistic transfections, sonoporation, and microfluidic device-based electroporation may cause damage to the cell membranes,⁶¹⁻⁶⁴ while noninvasive strategies such as nanoparticles, liposomes and quantum dots suffer from poor bioavailability, low efficiency and poor specificity when administered for macromolecules delivery.⁶⁵⁻⁶⁸ Notably, CPP, as a noninvasive drug delivery system, meets stringent requirements to ensure efficient and targeted drug delivery to intracellular organelles with high biocompatibility and low side effects.^69,70

3.1. CPP-Mediated Mitochondrial Restoration

A series of delivery strategies based on CPPs have been recently reported for the delivery of therapeutic drugs to mitochondria for some specific diseases^{10,11,71−76} (Table 1). SS (Szeto–Schiller) peptides are a group of CPPs composed of alternating aromatic–cationic amino acids motif and are well-known for their antioxidant activity.^77,78 These peptide antioxidants targeting mitochondria can cross the cell membrane and localize in the IMM of mitochondria, exhibiting excellent mitochondrial protection in inhibiting mitochondrial swelling, decreasing mitochondrial ROS production, and preventing mitochondrial depolarization.⁷⁸ These therapeutic effects were further evaluated in numerous animal models of diabetes mellitus, ischemia–reperfusion injury, Alzheimer’s disease (AD), renal fibrosis, and cardiovascular diseases.⁷⁹⁻⁸³ To maximize their pharmacological benefits in mitochondrial disease, various SS peptide-based drug delivery systems have been extensively developed against various mitochondria-related diseases in vitro and in vivo.⁸⁴⁻⁸⁷ In a study by Kuang et al.,⁸⁴ SS-31 peptide-conjugated geranylgeranylacetone (GGA)-loaded poly(lactic-co-glycolic acid) (PLGA) nanoparticles were constructed as a new mitochondria-targeted drug delivery system to investigate its protective effect against gentamicin in a zebrafish model. Interestingly, the mitochondrial membrane potential (MMP, ΔΨm) was slightly enhanced and then returned to a steady-state, indicating the effect of SS-31-modified PLGA NPs on respiratory chain complexes in mitochondria. As expected, SS-31-conjugated PLGA NPs mainly accumulated in the mitochondria of hair cells and increased the survival of hair cells, whereas the unconjugated formulations showed no effects. In a similar manner, SS31 and PEG-modified PLGA nanoparticle (CsA@PLGA-PEG-SS31) were developed for the specific delivery of cyclosporin A to mitochondria of ischemic cardiomyocytes for the treatment of myocardial ischemia–reperfusion injury. After translocation of CsA@PLGA-PEG-SS31 in mitochondria, the viability of H/R-injured H9c2 cells were significantly increased, and further in vivo results showed that CsA delivered by PEG-SS31 could protect myocardium and improved cardiac function.⁸⁵

Table 1. Peptide-Based Mitochondria-Targeted Therapeutic Applications.

CPP	cargo	recipient	disease model	consequence	ref
Pen (penetratin)	proapoptotic peptide KLA	human cancer cell lines including A549, SK-N-SH, EpoB40, EpoB480 and SW480 cells	cancer	tumor cell growth inhibition	(75)
TAT	NADH dehydrogenase (ubiquinone) Fe–S protein 8 (NDUFS8)	T-REx-293 cells, human embryonic kidney cells	NDUFS8 defects	rescue complex I deficiency	(72)
MTPs	doxorubicin-based prodrug	HeLa cells	cervical cancer	mitochondrion-targeted cancer therapy	(71)
Chol-FRFK/D	antimycin A	A549 cell	lung cancer	anticancer	(73)
CAMP	hMT1A	MPP+-induced cellular PD model in SH-SY5Y human neuroblastoma cells, MPTP-induced PD model mice	Parkinson’s disease	mitochondrial damage alleviation	(74)
octa-arginine	oxaliplatin	HCT116, SW480, and SW620 colon cancer cells and colon cancer mouse model	colorectal carcinoma	antitumor activity	(11)
R8MTS	doxorubicin	breast cancer 4T1, MDA-MB-231 cells, and 4T1-bearing mice	breast cancer	breast cancer metastasis suppression	(10)
VDAC1-based MPPs
R-Tf-D-LP4	not applicable	human HepG2, HuH7, and mouse BNL1MEA.7R.1 liver cancer cell lines; hepatocellular carcinoma mouse models	liver cancer	apoptosis induction in cancer liver cell lines and tumor growth inhibition in liver cancer mouse models	(89)
D-ΔN-Ter-Antp and Tf-D-LP4	not applicable	U-87MG, U-251MG, U-118MG, LN-18, SH-SY5Y, GL-261MG Neuro-2a cells, primary astrocyte, and glia cells; intracranial-orthotopic xenograft mouse model	glioblastoma	apoptotic cell death induction and inhibition of tumor development	(88)
R-Tf-D-LP4	not applicable	steatosis-nonalcoholic-steatohepatitis (NASH), hepatocellular carcinoma (HCC) stelic animal model	nonalcoholic steatohepatitis hepatocellular carcinoma	suppression of steatosis and NASH-associated pathologies	(12)
R-Tf-D-LP4	not applicable	various cancer cell lines including U-251MG, U-118MG, U-87MG, LN-18 and etc.; lung, breast cancers and GBM xenograft mouse models	glioblastoma, lung cancer, and breast cancer	inhibition of tumor growth by inducing massive cancer cell lines death, including of cancer stem cells	(149)
Pal-N-Ter-TAT, pFL-N-Ter-TAT, and Pal-pFL-N-Ter-TAT	not applicable	A375 cells	melanoma cells	induction of mitochondria-mediated apoptosis	(150)
Antp-LP4	not applicable	peripheral blood mononuclear cells (PBMCs) isolated from CLL (chronic lymphocytic leukemia) patients	B-cell chronic lymphocytic leukemia	cell death induction	(151)
pHK-based MPPs
pHK-PAS	not applicable	HeLa cells	cervical cancer	led to mitochondrial dysfunction and apoptosis in cancer cells	(90)
PAS–pHK conjugate	miR-126	MCF-7 cells	breast cancer	miR-126 cargo synergistically enhances the anticancer effects of PAS-pHK	(13)
TAT-HK	not applicable	HeLa cells	cervical cancer	apoptosis induction	(152)
pHK-pKV	not applicable	A549 cells	lung cancer	lung cancer cell death	(153)
HK2pep	not applicable	HeLa cells, MDA-MB-231 cells, S462 cells, PN 04.4 cells, 4T1 cells, CT26 cells, RAW 264.7 cells, C2C12 cells, COLO 741 cells, and B-CLL MEC1 cells	various cancer cell model	Ca2+-dependent calpain activation and cell death	(154)

Voltage-dependent anion channel 1 (VDAC1) is an integral protein located in the OMM that can interact with more than 100 proteins to exchange ions and metabolites between the cytosol and the mitochondria. The amino acid residues of VDAC1 are thought to be mitochondrial-targeting sequences that can be incorporated into CPPs to form VDAC1-derived mitochondrial-targeting peptides (MPPs) to target mitochondria and dissociate the complexes of VDAC1 from its partners (Table 1). Based on this principle, Shoshan-Barmatz’s group designed a series of VDAC1-based MPPs and applied them in the mouse models of glioblastoma, nonalcoholic steatohepatitis, and hepatocellular carcinoma, successfully achieving the desired therapeutic effects.^12,88,89

Among the interaction partners of VDAC1, the N-terminal domain of hexokinase II (pHK) binds VDAC1 to form the VDAC–hexokinase II complexes, blocking a major cell death pathway and consequently inhibiting mitochondria-mediated apoptosis. Hence, various pHK-incorporated MPPs have been developed to reach mitochondria and disrupt the VDAC1–hexokinase II interaction for anticancer treatments (Table 1). Woldetsadik et al.⁹⁰ designed a novel CPP by covalently coupling pHK to a short, penetration-accelerating sequence (PAS), yielding pHK-conjugated MPP pHK-PAS. They demonstrated that pHK-PAS efficiently targeted mitochondria, disrupted the association between VDAC and HKII, and ultimately triggered apoptosis in cancer cells. Motivated by this result, they further extended this approach and application by investigating the ability of pHK-PAS to deliver several cell-impermeable anticancer cargoes in vitro and in vivo, thereby maximizing the therapeutic antitumor effects. pHK-PAS was assessed as a synthetic mimic for delivery of the tumor suppressor miR-126 in cancer cell lines. As expected, PAS-pHK-miR-126 conjugates were distributed throughout the cytosol and predominantly localized to mitochondria. PAS-pHK induced depolarization of mitochondrial membrane potential, inhibited metabolic activities, depleted intracellular ATP levels, and triggered cytochrome c release and apoptosis. Importantly, the anticancer effect of PAS-pHK is synergistically enhanced by miR-126 cargo.¹³

3.2. Endoplasmic Reticulum–Golgi Network Targeting Therapy

Since the ER and Golgi apparatuses are key targets for certain diseases within eukaryotic cells, molecular delivery strategies developed to target the ER and Golgi may represent future diagnostic and therapeutic approaches. TAT peptide-bound QDs (TAT-QDs) were used as a model system in cell imaging and long-term tracking studies of live cells. As a result, an obvious and excellent fluorescence signal was observed in the Golgi apparatus, indicating the function of TAT-QDs as an effective tool for subcellular tracking.⁹¹ In 2019, Tan et al.⁹² constructed a PEG-modified CPP delivery system to facilitate mucous permeation and oral delivery of therapeutic proteins and peptides. As expected, CPP@NP showed a higher mucous permeation effect than the negative group without CPP involvement. In addition, a significant signal was observed in the ER–Golgi apparatus during the study of subcellular transport, showing a specific ability to target the ER–Golgi pathway. Recently, Jeong et al. proposed an intracellular delivery system with specific ER-targeting ability in which CPP- Penetratin was conjugated with lipid/polymer hybrid nanovehicles (LPNVs), paving the way for the development of intracellular ER-targeted drug delivery technology.¹⁴ KDEL (Lys-Asp-Glu-Leu) contains ER retention signals that can interact with the KDEL receptors localized in the Golgi apparatus,⁹³ and thus can be used as ER localization sequences for designing ER-targeted delivery tools. By linking TAT and KDEL with IL-24, a novel CPP- and KDEL-based delivery vector was developed, a delivery platform for delivering IL-24 to cancer cells for cancer therapy.⁹⁴ The results of In vitro experiments showed that TAT-IL-24-KDEL efficiently penetrated different tumor cell lines and was located mainly on ER, leading to inhibition of cell growth and induction of apoptosis via ER stress-mediated cell death pathway but without harm to normal NHLF cells. In contrast, treatment with TAT-IL-24-KDEL also significantly inhibited tumor growth in tumor xenograft animal models. Similarly, TAT-CaM, a CPP-adaptor system, can efficiently escape from endosomes and deliver cargoes to the ER by addition of KDEL to cargo proteins.⁹⁵ Although research papers are limited, these studies provided promising indications to design CPP-mediated drug delivery system to enable ER and Golgi targeting.

3.3. Lysosome-Targeted Therapy

An effective approach for lysosomal storage diseases (LSD), known as enzyme replacement therapy (ERT), is to break down toxic substances by trafficking desired enzymes to lysosomes within cells deficient in lysosomal enzymes. ERT has shown remarkable clinical benefits in the treatment of a series of LSDs, such as mucopolysaccharidosis,⁹⁶ Gaucher disease, Pompe disease,⁹⁷ and Fabry disease.⁹⁸ Although high efficacy was obtained in the liver and spleen after intravenous administration, the efficacy was poor in the heart, trachea and bronchi, joints, eyes, central nervous system, and ear, nose, and throat, due to limited penetration in various tissue barriers.⁹⁶ CPPs, as a tool capable of delivering cargos of various sizes across biological barriers, including the skin, epithelial tissue, intestinal mucosa, blood–brain barrier (BBB), cornea of the eye, and plasma barrier into various organs and some organelles like the nucleus, mitochondria, and ER–Golgi network, represent a promising candidate for the design of ERT-targeted delivery systems.⁵ In research of the cellular uptake and in vivo distribution of a series of cell-permeable polyhistidine peptides, Iwasaki’s group found that most polyhistidine-H16 peptides accumulate in lysosomes.⁹⁹ Based on this feature, they anticipated and further demonstrated that CPP-H16 could act as a valuable carrier for targeted delivery of drug to intracellular lysosomes.¹⁰⁰ H16 was first assessed for its ability to deliver large cargoes (e.g., liposomes), showing high potency. Then, H16-Lipo was investigated as a carrier to deliver a lysosomal enzyme, α-galactosidase A (GLA) to intracellular lysosomes. Further, they synthesized a polylysine–polyhistidine fusion peptide (K10H16 peptide) and developed a simple method to deliver GLA into the lysosomes via complexes formed by the electrostatic interaction of K10H16 peptide with enzymes, which efficiently restored the enzyme activity and improved the proliferation of GLA-deactivated cells.⁸ In another study conducted by Yang’s group, hybrid CPP-modified quantum dots (denoted as RhB-R9H6-QD) in which 9-mer polyarginine (R9) was applied as a vector for cell internalization, the histidine (His6) domain was used to mediate self-assembly onto the QD surface, and the RhB dye was used as pH indicator, thus generating a QD-based radiometric nanosensor.¹⁰¹ Finally, this platform is located both in the extracellular juxta membrane region and in the lysosomes, making it an ideal tool for simultaneous sensing and imaging of both extracellular and lysosome pH. To better ensure recognition by the lysosome during intracellular localization, a class of lysosome sorting peptides (LSPs),¹⁰² i.e., tyrosine-based peptide sequences containing degenerate motifs of 4–5 amino acids such as YXXØ or NPXY (X, any amino acid; Ø, bulky hydrophobic amino acid) were included in CPP-mediated drug delivery strategy. The CPPs (Pen, TAT)–LSP conjugates were shown to efficiently and selectively deliver 13 nm AuNP into the lysosomes with minimal cytotoxic effects.¹⁰³ However, despite the clear advantages shown by CPP–LSP, no further investigations have been performed on the application of this platform for lysosome imaging or therapy.

3.4. Ribosomes Targeting Therapy

Within the ribosome, rRNA molecules, are responsible for the catalytic steps for peptide bond formation and act as mediators of mRNA/tRNA interactions designated to protein synthesis. Since the rapid proliferation of some cancer cells is associated with the overproduction of some oncoproteins, ribosome inactivation may represent an effective anticancer modality. Ribosome inactivation can be achieved by ribosome-inactivating proteins (RIPs), which are mainly of plant origin and can lead to the inhibition of protein synthesis by removing a specific adenine residue from the stem loop of rRNA through the rRNA N-glycosidase activity of RIP.¹⁰⁴ Nevertheless, the anticancer application of RIP has not yet been clinically realized due to its poor cellular uptake and inability to reach the ribosome.¹⁰⁵ To overcome this limitation, researchers attempted to apply CPP as a means for RIP delivery into the cytoplasm to target and enzymatically destroy ribosome, ultimately leading to protein translation inhibition and cell death.^106−109 In 2013, Yang et al.¹⁰⁶ modified RIP-gelonin with CPP-LMWP using both chemical conjugates and genetic recombination that produced a gelonin-LMWP chemical conjugate (cG-L) and recombinant gelonin-LMWP chimera (rG-L), respectively. Both cG-L and rG-L efficiently entered cells and showed significantly increased tumoricidal effects over rGel. And in vivo experiments further proved that rG-L significantly inhibited tumor growth with a very low dose, while the control group of rGel had no therapeutic effects. A year later, they proposed a novel strategy to link heparin-conjugated anti-CEA mAb (i.e., T84.66) with PTD-modified gelonin toxin for the treatment of CEA overexpressed cancer.¹⁰⁷ Most importantly, a significant inhibition of tumor growth (46%) was observed in the group treated with the TAT-gelonin/T84.66-Hep complex without systemic toxicity. Similarly, CPP-HBD was also used to deliver other RIPs, such as TCS and MAP30 to several types of tumor cells.^108,109 In vitro experiments have shown that TCS-HBD and rMAP30-HBD proteins not only exhibit efficient cellular uptake resulting in ribosomal incorporation and translation inhibition but also show more cytotoxic effect and apoptosis rates in tumor cell lines compared to those with HBD-free controls.

4. CPP-Mediated Organelle Replacement Therapy

Drug delivery systems based on CPPs to restore organelle dysfunction are certainly a promising strategy for the treatment of organelles-associated diseases, however, what if the damage to these organelles is irreversible? An alternative strategy for functional restoration in the treatment of organelles-associated disease is the direct transfer of these organelles from healthy cells to target cells with organelles deficiency, called organelle replacement therapy (ORT). The natural biological process of intercellular organelle exchange via tuned nanotube (TNT) and secreted cell bodies such as extracellular vesicles (EVs) exist in almost all mammalian cells.^110−112 TNTs are tubular structures containing skeletal and transport proteins, whereas EVs are spheroid structures surrounded by lipid bilayers, both of which mediate cell–cell communication by exchange of cytoplasmic materials including organelles.^113,114 Their small size and flexibility to change shape and length allow mitochondria and lysosomes to be surrounded and immobilized by membranes during transport, thus maintaining membrane integrity, protecting them from external damage, and ensuring the organelles function.¹¹⁵ One requirement that needs to be fulfilled for efficient and successful organelle transfer is to maintain the structural integrity within membrane structures throughout this process.^116,117 Even if TNT and EV-based organelle exchange satisfy this requirement, one of the biggest challenges is to establish a scalable manufacturing technology for reliable high-quality and high-volume production of products at a clinical scale.

4.1. Mitochondria Replacement

Several well-designed strategies were developed since the first formal mitochondrial transfer from one cell to another via coincubation completed by Clark and Shay in 1982^17,118−130 (Figure 2). Three methods are available for mitochondrial substitution, namely cytoplasmic transfer (CT), maternal spindle transfer (MST) and pronuclear transfer (PNT).¹³¹ Among these methods, only CT is a technique in which the nuclear genome was extracted from a mitochondria-deficient oocyte, subsequently implanted into a nucleus-removed donor cytoplast, and further fertilized with the father’s sperm has been successfully implemented in patients. In 2015, this technique was approved for use in the UK and the first individual baby has been born.^3,4 In this remarkable therapeutic strategy, mitochondrial transfer is based on microinjection method followed by passive uptake which limits uptake efficiency and cause side effects in vivo,¹¹⁷ evoking the importance of developing a more efficient mitochondria delivery strategy (Table 2).

Figure 2.

Schematic timeline of mitochondrial transplantation strategies. Timeline of research efforts in the development of mitochondrial replacement strategies over the past 40 years. Key findings in mitochondrial replacement are highlighted. A variety of mitochondria delivery methods have been discovered and continuous updated that mainly relies on the novel materials and devices.

Table 2. CPP-Mediated Transmission of Mitochondria In Vitro and In Vivo Models.

donor cells	recipient cells	disease model	consequence	ref
parent 143B cells	MERRF cybrid cells (MitoB2) and mtDNA-depleted Rho-zero cells (Mitoρ)	cell model of myoclonic epilepsy with ragged-red fibers (MERRF) syndrome	mitochondrial function recovery and cell survival	(123)
homogenous and normal skin fibroblast cells	MERRF fibroblasts	cellular models derived from patients with MERRF syndrome	better delivered efficiency and mitochondrial biogenesis, mitochondrial function recovered	(132)
143B osteosarcoma control cybrid	6-OHDA-lesioned rat model of PD	Parkinson’s disease	improved motor function, increased performance of tyrosine hydroxylase (TH) and survival of dopaminergic neuron, better mitochondrial complex (CI, III, and V) activities and higher copy number of mitochondrial DNA, up-regulation of OPA1, MFN2, and Drp1, and mitophagy induction shown by up-regulation of Parkin, PINK1, LC3, polyubiquitin	(133)
allogeneic PC12 cells and xenogeneic human osteosarcoma cybrids	PC12 cells	Parkinson’s disease	sustained mitochondrial function, improved cell viability and neurite outgrowth	(134)
allogeneic rat livers	Parkinson’s disease model rats	Parkinson’s disease	significant improvement of rotational and locomotor behaviors, dopaminergic (DA) neuron survival and recovery >60%	(19)
human 143B osteosarcoma cybrid cells	human cybrid cells with the mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes (MELAS) A3243G mutation	MELAS cybrid cell model with MELAS A3243G mutation in mtDNA	mitochondrial respiratory function recovery, evident mitochondria biogenesis; reversed fusion-to-fission ratio of mitochondrial morphology and improved cell survival	(135)
allogenic mice liver	aging mice	antiaging effect in hair aging of naturally aging mice	stimulated hair regrowth, increased thickness of subcutaneous fat, consist of increase of mitochondria in the subcutaneous muscle and mitochondrial DNA copies in the skin layer	(136)
exogenous human uterine endometrial gland-derived mesenchymal cells (EMCs) or H9c2 cells	neonatal rat cardiomyocytes (NRCMs)	acute myocardial infarction	reduced apoptotic rates and oxidative phosphorylation suppression	(18)
mesenchymal stem cells (MSCs)	RAW264.7 cells, macrophages (Mφ) and diabetic mice	diabetic nephropathy (DN)	inflammatory response restriction and kidney damage alleviation	(17)

CPP-based delivery platforms, which are easy to synthesize and optimize, have been explored as delivery tools for mitochondria. In 2013, Liu’s group designed the first delivery system for CPP-mediated mitochondria transplantation in vitro and in vivo^123,132,133 (Figure 3). They generated the peptide-mediated mitochondrial delivery (PMD) system in which Pep-1 was used as the vector for the delivery of wild-type mitochondria isolated from parent cybrid cells into MERRF (myoclonic epilepsy with ragged red fibers) cybrid cells (MitoB2) and mtDNA-depleted Rho-zero cells for recovery of human cells harboring the mitochondrial DNA mutation MERRF A8344G. This PMD system enabled efficient mitochondrial internalization, restoration of mitochondrial function, cell survival and increased mitochondrial biogenesis in both cell types. Moreover, they further demonstrated the feasibility of this PMD system in the treatment of MERRF syndrome caused by point mutations in the tRNA genes of mtDNA.¹³² More importantly, PMD system was confirmed as a useful approach to restore mitochondria functions, cell viability, neurite outgrowth and locomotive activity in treating 6-hydroxydopamine-induced model of Parkinson’s disease via local injection and intranasal infusion bypassing the blood–brain barrier, respectively.^19,133,134 In the same manner, PMD-mediated mitochondrial transplantation was validated to be effective in treatment of breast cancer.²⁰ The uptake efficiency of P-Mito was slightly higher compared to passive uptake. However, both transplants induced apoptosis; inhibited cell growth, decreased cellular oxidative stress; and enhanced cellular susceptibility of MCF-7 to doxorubicin and paclitaxel. They further wanted to determine whether Pep1-mediated mitochondrial transplantation affects the oxidative stress response and has a protective effect on human cybrid cells in MELAS disease.¹³⁵ Consequently, significant improvements in mitochondrial function, cell viability and normalized cytokine expression were observed compared to the control group. Interestingly, the antiaging effects of P-Mito on hair was further explored by using a stamped electric multineedle injector on naturally aging mice.¹³⁶ The P-Mito stimulated hair regrowth and increased the thickness of subcutaneous fat, validating the efficiency of PMD in restoring hair loss associated with aging. Pep1-mediated mitochondria transfer from MSC to Mφ in vitro and in vivo was first assessed by Yuan’s team.¹⁷ MSC-Mito was mixed with peptide Pep1 to form Pep1-Mito, followed by incubation with Mφ for 48 h to educate Mφ, which was further used to treat diabetic nephropathy (DN) mice via tail vein injection. Remarkably, this coculture system of Pep1-Mito successfully alleviated inflammation and kidney injury in DN mice. Recently, Satoshi, who focuses on mitochondria disease research, isolated mitochondria from human endometrial gland-derived mesenchymal cells (EMCs) and H9c2 cardiomyoblasts stably expressing DsRed-Mito or GFP-Mito and compared the use of TAT-dextran complex-modified mitochondria transport (TAT-Mito) with TNT-mediated mitochondrial transfer for mitochondrial replenishment in neonatal rat cardiomyocytes (NRCMs) in response to oxidative stress. Notably, significant higher levels of cellular uptake and increased inhibition of antiapoptosis and oxidative phosphorylation were observed in the TAT-Mito group compared to the simple TNT-Mito approach, suggesting the potential of CPP-mediated mitochondrial supplementation therapy in the treatment of mitochondria-related diseases.¹⁸

Figure 3.

Schematic illustration of CPP-mediated organelle delivery. Left: CPP-Pep-1 mediated the delivery of healthy mitochondria into the cell. Healthy mitochondria extracted from donor cells are assembled with CPP-Pep-1 and then further delivered to receipt cells mediated by peptide Pep-1. Right: His16-mediated delivery of healthy lysosome to lysosomal storage disorder (LSD) cells. After extraction and labeling with CPP-His16, lysosomes from donor cells can be efficiently delivered into LSD cells.

4.2. Lysosome Replacement

To address the need for therapeutic strategies in LSDs, organelle replacement therapy (ORT) is expected to be a new ERT to supply intact healthy lysosomes extracted from normal cells to lysosomes within LSD cells. To date, the only reported method of ORT is peptide-based lysosome replacement, which is still in the early stages of research. Cell-permeable H16-modified liposomes were applied as carriers of intercellular lysosome transfer in a model experiment by using human fibrosarcoma HT1080 cells as non-LSD cells²¹ (Figure 3). Healthy lysosomes were first extracted from human fibrosarcoma HT1080 cells, which have been used to produce recombinant lysosomes for ERT.^137−139 Stearyl-His16 was inserted to the hydrophobic region of the lysosomal membrane, designated as His16-Lyso. Cellular uptake and intracellular localization of His16-Lyso to endogenous lysosomes was demonstrated in model experiments. In the confirmation experiment of ORT-targeting LSD cells, His16-Lyso was found to internalize into fibroblasts of FD patients and restore cell proliferation, suggesting that His16-Lyso showed efficacy of ERT by supplementing lysosomes in endogenous lysosomes of LSD cells. This finding provides a promising drug candidate for the treatment of LSDs and points to a new direction for ORT.

4.3. CPP-Mediated EV Replacement

So far, we have summarized a series of CPP-based organelle-targeted delivery to restore organelle function and CPP-mediated organelle delivery for organelles-related disease treatment. In contrast to the organelles mentioned above, EVs, including microvesicles (MVs), exosomes, and apoptotic bodies are organelles with the intrinsic ability to transport biomolecules between cells, mainly through endocytosis. As natural vehicles, EVs exhibit high delivery efficiency, low immunogenicity, and good compatibility. Due to these desirable natural advantages, EVs are a new generation of drug delivery of external drugs or molecules to targeted cells. However, limitation in low drug loading capacity and clinical-grade production limitations hinder the application of EV-mediated drug delivery.¹⁴⁰

Compared to EVs, CPPs are natural or synthetic carriers that are generated in a faster and easier way, and therefore can be obtained by various techniques (Figure 4). Hence, combining these two carriers for drug development may achieve higher delivery efficiency and correspondingly better treatment effect. Based on these potential advantages, many researchers explored to combine these two tools on a single platform to achieve efficient internalization and site-specific drug release in different disease models^{16,22,141−145} (Table 3).

Figure 4.

EVs modified with CPPs for drug delivery. After cargo molecules encapsulated into EVs, CPP-modified EVs can be sufficiently taken up and trafficked and finally released from endosomes to the cytosol for further functions.

Table 3. Delivery of CPP-Modified EVs in Different Disease Models.

donor	CPP	size (in diameter)	recipient	cargo	disease model	consequence	ref
CD63-GFP-HeLa cells	GALA	∼78 nm	HeLa	dextran and saporin	not applicable	enhanced cytosolic delivery of exosomes	(145)
HeLa	hexadeca-arginine (R16)	30–200 nm	CHO-K1 cells and CHO-A745 cells	SAP	not applicable	potent anticancer activity	(22), (141)
HeLa	stearyl-r8	196.9 ± 46.7 nm	A431 cells	SAP	squamous cell carcinoma	induction of cytotoxicity in targeted cells	(142)
HeLa	CAP18-derived (sC18)2 peptides	∼200 nm	MCF-7	SAP	breast cancer	cell death induction	(23)
HepG2	R9	∼200.9 nm	HepG2, MCF-7, MDA-MB231, HeLa cells, and L-02 hepatocytes	Bcl-2 targeted ASOs	various cancer cell models	down-regulation of antiapoptotic Bcl-2	(16)
Raw 264.7	TAT	∼79 nm	HeLa cells, tumor-bearing mice	carbon dots (CDs:Gd,Dy)	cervical cancer	imaging-guided cancer therapy via burning of tumor cell nuclei and the ablation of tumors in vivo under NIR laser irradiation	(148)
WPMY-1	TAT-DRBD fusion protein	144.9 ± 15.6 nm	LNCaP-AI cells	cocktail siRNAs targeting FLOH1, NKX3, DHRS7 genes	castration-resistant prostate cancer (CRPC)	selective gene down-regulation for CRPC treatment	(15)
HeLa	R16	95 ± 13 nm	C6-glioma cells	disodium mercaptoundecahydro-closo-dodecaborate (BSH)	glioblastoma	boron neutron capture therapy	(147)
MSCs	CPP (KETWWETWWTEWSQPKKKRKV)	not determined	MCF-7, A549, Colo201, B16F10 melanoma tumor model	TNF-α	various cancer	anticancer targeting therapy	(144), (145)

The most employed cotransfer technique is to encapsulate drugs into EVs and further modify the surface of EV membranes with CPPs for effective transport. In 2017, one study conducted by Nakase’s group encapsulated RIP-saporin (SAP) into EVs and modified it with R16 peptide, which effectively attained increased anticancer activity.¹⁴¹ However, in the development EVs for functional and therapeutic applications, EVs are usually frozen for long-term storage. To avoid the destruction of EVs caused by repeated freeze–thawing cycles, a new preservation method, lyophilization, has been developed for storage of functional EVs.¹⁴⁶ Therefore, Nakase’s group further applied the lyophilization method to arginine-rich CPP-modified EVs and investigated the effects of lyophilization on the properties of EVs, including vesicular structure, internalization efficacy, and cytosolic release of SAP encapsulated within EVs. The results showed that lyophilization and subsequent recovery with water did not affect the particle size, structure, zeta-potential, and cellular uptake efficacy of EVs. However, the anticancer activity of the lyophilized encapsulated SAP was inhibited.²² Based on this progress, they further introduced a dimer peptide (sC18)₂ to EV membranes, which also greatly enhanced the cellular uptake of EVs, as well as the cellular membrane delivery of the bioactive protein SAP, with the help of hydrophobic molecules.²³ Recently, they successfully applied this technique to intracellular delivery of therapeutic boron compounds for boron neutron capture therapy (BNCT) by encapsulating fluorescently labeled boron compounds (FITC-BSH) into EVs with additional R16 incorporated in EVs, achieving efficient cellular uptake of FITC-BSH-EV with the help of R16 and attaining remarkable cancer-killing BNCT activity under irradiation with thermal neutrons.¹⁴⁷

In line with these studies, Xu et al.¹⁶ designed a CPP-equipped exosome platform in which polyarginine peptide (R9) and HepG2 cell-derived exosomes were easily assembled under mild reaction condition. The platform not only enhances the penetration ability of exosomes but also facilitates the loading efficiency of antisense oligonucleotides (ASOs). To facilitate tumor-targeted therapy, Yang and colleagues¹⁴⁸ designed and constructed exosome- and peptide-based delivery nanoplatforms for imaging-guided and targeted anticancer therapy. The surface of an exosomes was modified with the tumor-targeting peptide RGD to form RGD peptide-engineered exosomes (Exo-RGD) that bind with high affinity to tumor cells with the integrin αvβ3 overexpressed. TAT peptide-modified carbon dots (CDs:Gd,Dy-TAT) were encapsulated into Exo-RGD to facilitate accurate delivery into the nucleus for cancer imaging diagnosis and photothermal therapy. In vitro and in vivo experiments showed that CDs:Gd,Dy-TAT@Exo-RGD effectively accumulated at cancer sites. Under NIR irradiation, the temperature of tumors in mice treated with CDs:Gd,Dy-TAT@Exo-RGD rose above 50 °C, leading to local hyperpyrexia and significant ablation of tumors. The mice treated with CDs:Gd,Dy-TAT@Exo-RGD showed 100% survival and higher contrast-enhanced MRI/CT imaging of the tumor site after 60 days, both of which were higher than that in the control groups.

In contrast to the traditional encapsulation method of loading drugs into CPP-modified exosomes, Diao et al. reported a different transfer strategy using EVs to encapsulate CPP–cargo conjugates for in vitro delivery.¹⁵ They first applied a TAT-DRBD fusion protein to bind and mediate siRNAs delivery into EVs, and engineered EVs were further internalized by cancer cell lines, successfully achieving the desired therapeutic effects on castration-resistant prostate cancer (CRPC) cells. In a similar way, CTNF-α-exosome-SPIONs platform was designed by conjugation of superparamagnetic iron oxide nanoparticles (SPIONs) to the surface of CTNF-α (CPP and TNF-α)-anchored exosomes for antitumor targeting therapy.¹⁴⁴ In this system, recombinant plasmids encoding both CPP and TNF-α were constructed and stably expressed in MSC cell line, enabling TNF-α to anchor in the cell membrane and thus producing exosomes anchored with CTNF-α. In vitro and in vivo data showed that tumor cell growth was significantly inhibited by CTNF-α-exosome-SPIONs with tumor-targeting capacity. Unfortunately, due to the limited number of reports on this strategy, the efficiency of targeted delivery and potential for clinical application remain unclear.

5. Concluding Remarks

An organelle is a specialized subunit that plays a specific role or function in the cytoplasm, and if dysfunction occurs in the cellular homeostasis, it leads to various pathophysiology in development, aging and diseases (cancer, cardiovascular and neurodegenerative diseases), reflecting that organelle targeting with pharmacological therapeutics can restore cellular function, improve disease diagnosis, or even cure, stop or prevent disease.

Over the past few decades, CPPs have been developed and extensively used in animal models for the treatment of various diseases, some of which have entered phase I or II clinical trials, and some have also entered phase III. Since protein–protein detection platforms (phage or yeast display, protein fragment complementation, protein arrays, affinity chromatography, tandem affinity purification, coimmunoprecipitation, NMR spectroscopy and X-ray crystallography) became feasible, a variety of organelle-, cell-, or tissue-specific ligands have been identified. Therefore, conjugation of these ligands or experimentally modified ligands to CPPs has become an increasingly popular delivery for targeting tissues, cells, or organelles.

In this review, we summarized CPP-based delivery approaches for targeting different organelles, as well as CPP-based organelle replacement therapies. The precise targeting of exogenous cargo such as enzymes, DNA/RNA, antisense oligonucleotides (ASO) or peptide nucleic acids (PNA) at the level of organelles including lysosomes, mitochondria, endoplasmic reticulum, and Golgi apparatus has proven to be valuable and offer a therapeutic avenue for the next generation of precision medicine. However, in some scenario, such as injury, damage, disease, or dysfunction accumulates and most organelles within eukaryotic cells become irreparably lost and cannot remodeled, and direct organelle replacement might be the last way to restore cellular function. As a representative example of organelle replacement, mitochondria transfer from a donor into a recipient has been successfully conducted in assisted human reproduction to rescue those patients that had repeatedly pregnancy failure.^3,4 Nevertheless, heterologous organelle replacement therapy is clinically complicated and has limited sources of healthy donors. Therefore, how to generate enough organelles with good quality for replacement will become an urgent issue.

Since the advent of novel biotechnology such as induced pluripotent stem cell (iPSCs) and clustered regularly interspaced short palindromic repeats (CRISPR) gene editing, access to patients or their specific source cells or organelles⁵⁶ is no longer a major obstacle for cell or organelle replacement therapies (illustrated in Figure 5). By implementing advanced measurements at the genomic, transcriptomic, proteomic, and metabolomic levels, rigorous quantitative and qualitative control of cells and/or organs can be achieved. In combination with the nonviral delivery tool CPPs, CPP-based organelle therapy strategies may be feasible in the future. In this view, the delivery of CPP-based organelle-targeted therapeutics and the next generation of CPP-based delivery at the organelle level offer a new frontier for diagnostic and therapeutic applications.

Figure 5.

Schematic diagram of potential applications of CPP-mediated organelle replacement therapy (ORT), novel induced pluripotent stem cell (iPSC)-based technology procedures for cell replacement therapy and organelle replacement therapy. CPP-based reprogramming factors (RFs) induce reprogramming of iPSCs from patient biopsies, and then with or without CPP-based mutation correction, we can eventually generate healthy cells for cell replacement and healthy organelles for organelle replacement therapy, as well. Quality control through DNA, RNA, protein, and metabolite levels during the preparation of healthy cells or organelles should be carefully evaluated prior to cell or organelle transplantation.

The authors declare no competing financial interest.