Mitochondrial pharmaceutics

Mitochondrion 3 (2004) 229–244 www.elsevier.com/locate/mito

Mitochondrial pharmaceutics Volkmar Weissig*, Shing-Ming Cheng, Gerard G.M. D’Souza Department of Pharmaceutical Sciences, School of Pharmacy, Bouve College of Health Sciences, Northeastern University, 360 Huntington Avenue, Mugar 211, Boston, MA 02115, USA Received 2 November 2003; received in revised form 18 November 2003; accepted 20 November 2003

Abstract Since the end of the 1980s, key discoveries have been made which have significantly revived the scientific interest in a cell organelle, which has been studied continuously and with steady success for the last 100 years. It has become increasingly evident that mitochondrial dysfunction contributes to a variety of human disorders, ranging from neurodegenerative and neuromuscular diseases, obesity, and diabetes to ischemia-reperfusion injury and cancer. Moreover, since the middle of the ‘1990s, mitochondria, the ‘power house’ of the cell, have also become accepted as the cell’s ‘arsenals’ reflecting their increasingly acknowledged key role during apoptosis. Based on these recent developments in mitochondrial research, increased pharmacological and pharmaceutical efforts have lead to the emergence of ‘Mitochondrial Medicine’ as a whole new field of biomedical research. Targeting of biologically active molecules to mitochondria in living cells will open up avenues for manipulating mitochondrial functions, which may result in the selective protection, repair or eradication of cells. This review gives a brief synopsis over current strategies of mitochondrial targeting and their possible therapeutic applications. q 2004 Elsevier B.V. and Mitochondria Research Society. All rights reserved. Keywords: Mitochondria; Drug delivery; Mitochondrial gene therapy; Mitochondrial drug targeting; Intracellular drug delivery; Subcellular targeting

1. Introduction Mitochondrial research is presently one of the fastest growing disciplines in biomedicine (Singh, 2000). It has become progressively more evident that mitochondrial dysfunction contributes to a variety of human disorders, ranging from neurodegenerative and neuromuscular diseases, obesity and diabetes to ischemia-reperfusion injury and cancer. Most * Corresponding author. Tel.: þ1-617-373-3212; fax: þ 1-617373-8886. E-mail address: [email protected] (V. Weissig).

remarkably, during the last five years, mitochondria, the ‘power house’ of the cell, have also accepted as the ‘motor of cell death’(Brown et al., 1999) reflecting their recognized key role during apoptosis. In particular, the mitochondrial permeability transition pore complex (mPTPC) is widely accepted as being central to the process of cell death and presents therefore a privileged pharmacological target for cytoprotective and cytotoxic therapies (Kroemer, 1999). Based on these recent exciting developments in mitochondrial research, increasing pharmacological efforts have been made leading to the emergence of ‘Mitochondrial Medicine’ as a new

1567-7249/$20.00 q 2004 Elsevier B.V. and Mitochondria Research Society. All rights reserved. doi:10.1016/j.mito.2003.11.002

230

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

field of biomedical research (Larson and Luft, 1999; Murphy and Smith, 2000; Schon and DiMauro, 2003). However, the development of ‘Mitochondrial Pharmaceutics’, i.e. the design and evaluation of mitochondria-specific drug carrier or delivery systems is lagging behind. The targeted and carrier-based delivery of drugs and DNA to mitochondria hardly constitutes a field of research on its own yet and is still in its infancy. A PubMed query done at the writing of this review (October 2003) revealed that about 3% of all papers involving ‘drug carriers’ and only 0.2% of all papers involving ‘drug delivery systems’ contain the term ‘mitochondria’. To draw the attention of a broad audience in the drug delivery community to this fascinating cell organelle, Advanced Drug Delivery Reviews (Elsevier) dedicated in 2001 a Theme Issue to a multitude of aspects related to the targeting and delivery of drugs and DNA to mitochondria (Weissig and Torchilin, 2001a,b). Here, an attempt shall be made to briefly summarize all the efforts made to develop drug carrier systems able to selectively deliver biologically active molecules to mitochondria in living cells and also to discuss the therapeutic applications of such systems.

2. The rationale for the delivery of biological active molecules to mitochondria Mitochondria are the prime target for pharmacological intervention due to their main role in numerous fundamental metabolic pathways. Above all, mitochondria are vital for the cell’s energy metabolism and regulation of programmed cell death. In addition, mitochondria are critically involved in the modulation of intracellular calcium concentration and the mitochondrial respiratory chain is the major source of damaging reactive oxygen species. Consequently, mitochondrial dysfunction either causes or at least contributes to a large number of human diseases (reviewed in Modica-Napolitano and Singh (2002)). Malfunctioning mitochondria are found in several adult-onset diseases, including diabetes, cardiomyopathy, infertility, migraine, blindness, deafness, kidney, liver diseases and stroke. The accumulation of somatic mutations in the mitochondrial genome has been suggested to be involved in

aging, age-related neurodegenerative diseases as well as in cancer. Also, an increasing number of xenobiotics and pharmaceutics are being recognized to manifest their toxicity by interfering with mitochondrial functions (comprehensively reviewed in Wallace and Starkov (2000)). In conclusion, the delivery of both, the small drug molecules and large macromolecules to and into mitochondria may provide the foundation for a large variety of future cytoprotective and cytotoxic therapies: † The delivery of therapeutic DNA and RNA such as antisense oligonucleotides, ribozymes, plasmid DNA expressing mitochondrial encoded genes as well as wild-type mtDNA may provide the basis in the treatment of mitochondrial DNA diseases. † The delivery of anti-oxidants may protect mitochondria from oxidative stress caused by a variety of insults; perhaps even contribute to slowing down the natural aging process. † The delivery of mitochondria-specific naturally occurring toxins or synthetic drugs such as photosensitizers may open up avenues for new anti-cancer therapies. Moreover, delivering molecules known to trigger apoptosis by directly acting on mitochondria may overcome the apoptosisresistance of many cancer cells. † The delivery of drugs targeting mitochondrialuncoupling proteins may become a basis for treating obesity. † The delivery of peptides and proteins could become the basis for the treatment of a huge variety of mitochondrial disorders.

3. Are mitochondrially targeted drug delivery systems needed? It is undisputed that oral, intravenous, transdermal and other routes of systemic drug administration are generally associated with a series of problems all linked with each other. The drug’s lack of specific affinity toward the pathological site results in its more or less even biodistribution throughout the body only governed by the physico-chemical properties of the drug and the local blood flow. Therefore, in order to achieve a sufficiently high local drug concentration at the site in need, a large dose has to be administered.

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

However, large drug doses in turn are inevitably cause for non-specific drug toxicity and other adverse sideeffects (Torchilin, 2000). Therefore, the ability to target therapeutic agents selectively to the specific tissue in need has long been a goal in medicine and Paul Ehrlich’s dream of a ‘Magic Bullet’ has never ceased to fascinate the Pharmaceutical Scientist. Shouldn’t the concept of target site-specific drug delivery be extended onto the subcellular level? Unarguably, any effective nuclear- and mitochondria-based gene therapeutic approach requires that the therapeutic gene be transported into the nucleus and into mitochondria, respectively. Enzyme-replacement therapy aimed at the treatment of lysosomal storage diseases depends on the successful localization of the therapeutic enzyme inside the lysosomes. More examples could be listed. Yet, the need for the development of organelle-specific drug delivery systems, in particular mitochondria-specific drug carriers, has not fully been recognized yet. Due to the abundance of mitochondria in almost all cells, the notion that once inside the cell, the drug molecule will eventually either interact with components of the mitochondrial membrane and depending on its physico-chemical properties perhaps diffuse into the matrix, appears as plausible. Indeed, there is no doubt that the random, i.e. statistical collision of biologically active molecules with mitochondria will cause a biological effect. But so will the interaction of this molecule with other targets inside the cell. An example shall be given. The cessation of blood flow followed by reperfusion causes severe cellular damages by inducing a complex cascade of events, which involve among others an alteration of ionic homeostasis promoting Hþ and Ca2þ accumulation and the generation of free radicals. In this context, mitochondria appear to be highly vulnerable and seem to play a decisive role in the cell signaling leading to cell death. Therefore, recent efforts (comprehensively reviewed in Morin et al. (2001)) to find an effective therapy for ischemia-reperfusion injury have focused on mitochondrial drug targets, most prominent among them the mPTPC. The involvement of the mPTPC in the pathogenesis of necrotic cell death following ischemia-reperfusion was suggested for the first time by Crompton et al. (1987). It was noticed that all cellular conditions, which promote the PTP, like Calcium overload, high phosphate concentrations and

231

oxidative stress, also prevail during ischemia-reperfusion. Consequently, a new pharmacological strategy, based on inhibiting the mPTPC, has been emerged (Morin et al., 2001). Cyclosporin A (CsA) was one of the first drug shown to inhibit PTP by binding to a mitochondrial cyclophilin with nanomolar affinity and to prevent the interaction of this protein with the adenine nucleotide translocator thus inhibiting pore opening (Crompton et al., 1999; Morin et al., 2001; Woodfield et al., 1998). Following the identification of CsA as a PTP inhibitor, it was shown that CsA is able to prevent or delay cell death caused by oxidative stress and to protect brain and heart from ischemiareperfusion injury. However, despite the apparent potential of CsA as an anti-ischemic drug, pharmacokinetic and pharmacodynamic parameters argue against the clinical use of CsA in ischemia. The drug targets at least eight other cyclophilins in the cell, whose roles are largely unknown and which are likely to bind a large portion of the administered drug. Therefore, the mitochondrial concentration of CsA is difficult to predict and a CsA treatment may require high, even toxic, concentrations to reach the mitochondrial target (Morin et al., 2001). Consequently, CsA as a potential anti-ischemic drug would almost certainly benefit from a mitochondria-specific drug carrier systems able to increase its therapeutic index.

4. Are mitochondrially targeted DNA delivery systems needed? For the treatment of mtDNA diseases via ‘direct mitochondrial gene therapy’ (Weissig and Torchilin, 2001a), the therapeutic DNA has to cross the plasma membrane as well as the outer and the inner mitochondrial membrane in order to reach the matrix space. While the transport of plasmid DNA into the mitochondrial matrix of living cells has proven to be elusive so far, the delivery of oligonucleotide and/or Peptide nucleic acids (PNA) conjugates with MLSpeptides to mitochondria of intact cells from outside the cell has been described (Chinnery et al., 1999; Flierl et al., 2003; Geromel et al., 2001). PNA – MLS conjugates were taken up by human myoblasts in their free form, i.e. without the help of either cationic liposomes or cationic polymers as vectors, which is not surprising considering the non-charged nature of

232

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

PNAs and their relative stability towards hydrolytic degradation. Once in the cytosol, free PNA molecules, i.e. PNAs not bearing a MLS peptide, localized to the nucleus, while conjugating the PNAs with a MLS peptide directed a subset of the PNA molecules to mitochondria. Although these experiments demonstrated that the addition of a MLS peptide facilitated mitochondrial targeting and import, they also showed that this process was not entirely efficient, as a fraction of the molecules were still localized to the nucleus (Chinnery et al., 1999). For the cellular uptake of negatively charged oligonucleotide-MLS peptide constructs and of PNA – MLS peptide – oligonucleotide complexes (PPO), cationic liposomes and cationic polyethylenimine (PEI) have been utilized (Flierl et al., 2003; Geromel et al., 2001). In the case of using cationic liposomes for the oligonucleotideMLS peptide delivery, the fluorescence of labeled oligonucleotide-MLS conjugates became associated with mitochondria (Geromel et al., 2001), yet it should be noted that oligonucleotide internalization into the mitochondrial matrix was not established. Most intriguingly, however, any attempts to utilize cationic liposomes for the delivery of PPO complexes into mitochondria of living myoblasts have failed despite the author’s efforts to try three different commercially available cationic liposome formulations (Flierl et al., 2003). Instead, it was found that the cytosolic delivery of PPO complexes using branched chain PEI was more successful. PEI has been developed in the 1990s as an efficient vector for nuclear-targeted gene and oligonucleotide transfer into cells in culture and in vivo (Boussif et al., 1995). Exposure of myoblasts to PEI – PPO complexes resulted in their efficient uptake into the cytosol followed by their initial peri-nuclear distribution and co-localization with lysosomes. Subsequently, the oligonucleotide fluorescence was described by the authors as forming bead-like structures thereby creating a pattern in which some fluorescence remained co-localized with lysosomes, but most of the fluorescence becoming co-localized with mitochondria (Flierl et al., 2003). Interestingly, it was found that the mitochondrial oligonucleotide fluorescence was in correlation to the proximity of the organelle to the PEI –PPO-complex aggregates. Only mitochondria in close proximity to

the PEI-PPO-complex aggregates rapidly developed high-level fluorescence (Flierl et al., 2003). Combining these three studies about the delivery of oligonucleotides and PNAs to mitochondria, following conclusions shall be drawn and questions raised: (i) While charge-neutral molecules (PNAs) can be taken up by cells in their free form (most likely via passive diffusion through the plasma membrane), negatively charged molecules need to complexed, i.e. charge-neutralized, with the help of either positively charged lipids or polymer. Once inside the cell, portions of oligonucleotide/PNA – MLS peptide complexes are being taken up by mitochondria. (ii) For the delivery of DNA into cells, a wide variety of systems have been developed, the most widely used ones are cationic lipid/liposomes and cationic polymers. All these DNA delivery systems have been designed and optimized for the delivery of DNA into the cell nucleus. Not much data are available about the intracellular fate of DNA/cationic liposome complexes (lipoplexes) and DNA/cationic polymer complexes (polyplexes) following their uptake by nonspecific endocytosis, which is widely believed to be their main mechanism of cell entry. However, two lines of discussions are prevailing in the literature. First, a large portion of the lipoplexes becomes entrapped in endosome-like vesicles and accumulates eventually in the peri-nuclear area (Zabner et al., 1899). Confirming these early data, we have found most recently using confocal fluorescence microscopy (Fig. 1) that Lipofectine, a commercially available transfection vector, transports the majority of fluorescence-labeled pDNA into close proximity of the nucleus seemingly bypassing mitochondria (Fig. 1b). Under identical experimental conditions, however, and in striking contrast to Lipofectin, CyclohexylDQAsomes (Weissig et al., 2001a), a newly developed mitochondria-specific DNA delivery vector derived from DQAsomes (Weissig et al., 1998a,b) deliver most of the pDNA directly to the site of mitochondria (Fig. 1a). Almost no fluorescencelabeled pDNA is detectable at or near the nucleus. It actually appears that the mitochondria-specific vector prevents pDNA from approaching the nucleus. In both cases, neither nuclear targeting sequence (NLS) nor MLS-peptides were present. Second, as evident from successful transgene expression, at least a small part of the DNA has to

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

233

Fig. 1. Confocal fluorescence micrographs of BT20 cells stained with Mitotracker Red CMXRos. Cells were exposed for 10 hours to fluorescein-labeled pDNA vectorized with the nuclear-targeted transfection vector Lipofectie (panel b) or with Cyclohexyl-DQAsomes (Weissig et al., 2001a–d), a newly developed mitochondrial-targeted transfection vector (panel a). Mitochondria are stained red, pDNA is indicated by green fluorescence. Confocal images of 0.2-(m sections were taken with a Zeiss Axioplan 2 microscope with auto Z focus shutter drives using Improvision’s Openlabe confocal imaging software (G.D’Souza, V. Weissig, 2003, manuscript in preparation).

dissociate from the cationic carrier, escape from the endosome and find its way into the nucleus. It is believed that anionic phospholipids located in the inner monolayer of the plasma membrane are able to form charge-neutral lipid pairs with the cationic lipid thereby displacing the DNA from its carrier upon interaction with the endosomal membrane (Xu and Szoka, 1996). As a result, the DNA would escape in its free form into the cytosol. Obviously, both of these proposed mechanisms, i.e. the peri-nuclear accumulation of the DNA/vector complex or the dissociation of the DNA from the vector upon contact with endosomal lipids would not facilitate any delivery of DNA to the site of mitochondria. Nevertheless, as shown (Flierl et al., 2003; Geromel et al., 2001), mitochondrial uptake of oligonucleotides, once inside the cytosol, does occur when conjugated to a MLS peptide. However, whether the mitochondrial uptake occurs just because the oligonucleotide is in extreme proximity of the organelle or whether either the MLS peptide alone or the carrier systems (liposome or polymer) transports the oligonucleotide to mitochondria is unclear. We believe that due to the abundance of mitochondria

and their widespread network relatively close to the nucleus, the oligonucleotide-MLS peptide conjugates are able to randomly interact with the cell organelle, once released into the cytosol. Note, that only mitochondria in close proximity to the PEI – PPOcomplex aggregates have been described to develop rapidly high-level fluorescence (Flierl et al., 2003). Likewise, when not using any delivery vector, i.e. in case of PNAs, only a subset of the PNA – MLS conjugates localized with mitochondria (Chinnery et al., 1999). (iii) Even if the presence of MLS-peptides at oligonucleotides ‘actively’ contributed to a targeting effect, the question would arise whether one MLS peptide conjugated to a large plasmid DNA molecule would be sufficient to ‘drag’ a large pDNA molecule all the way through the cytoplasm to the cell organelle. Otherwise, attaching more than one targeting peptide to the plasmid might actually counteract the uptake by the cell organelle as it has been shown for nuclear uptake of pDNA after the attachment of several nuclear targeting sequence (NLS) peptides to one single pDNA molecule (Behr, 2000; Zanta et al., 1999).

234

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

In summary, we believe that for efficient and selective targeted delivery of pDNA (and other biologically active molecules) a mitochondriaspecific delivery systems needs to be designed. Such a delivery system must bind and condense pDNA (or be able to encapsulate small molecules), chargeneutralize the DNA, protect it from nuclease digestion, mediate its cellular uptake, transport the DNA (or small molecules) selectively to mitochondria and finally become destabilized upon contact with mitochondrial membranes leading to the release of its cargo. To this end we have been exploring the use of self-assembling mitochondriotropic bolaamphiphiles (Weissig et al., 1998a,b; Weissig et al., 2001a – d). As result, we have developed DQAsomes (Weissig et al., 1998a,b), which meet all the above criteria, as a first model system for a mitochondrial drug and DNA delivery system (to be discussed in more details below) (D’Souza, 2003; Weissig and Torchilin, 2001b; Weissig et al., 2003).

5. Strategies for mitochondrially targeted drug and DNA delivery systems Most of the currently made efforts for transporting biologically active molecules to and into mitochondria within living mammalian cells are based on two distinct mitochondrial features: the high membrane potential across the inner mitochondrial membrane and the organelle’s protein import machinery (Muratovska et al., 2001; Murphy, 1997; Murphy and Smith, 2000). Contemporary drug delivery strategies, based on both features, either independent on each other or in combination, will be discussed below. In principle, however, also any other endogenous mitochondrial metabolite transporters (Dolce et al., 2001; Fiermonte et al., 2001, 2002; Palmieri et al., 1996) could potentially be utilized for transporting drug or DNA molecules to and/or into the matrix of mitochondria. In addition, any mitochondria-specific binding sites and unique protein receptor sites at the mitochondrial membranes could be used for drug targeting purposes. Most recently, in a very unique way, the transactivator of transcription (TAT) protein transduction domain (PTD) of the HIV virus has been utilized to mediate in vitro and in vivo the uptake of an exogenous fusion protein into

the mitochondrial matrix (Del Gaizo and Payne, 2003). A fusion protein was constructed composed of green fluorescence protein (GFP), a MLS peptide and TAT. It could be demonstrated that TAT mediates not only the uptake into the cell cytoplasm but also the crossing of both mitochondrial membranes. The presence of the MLS peptide (between GFP and TAT) allowed for processing by mitochondrial proteases, which lead to the removal of the TAT domain from the GFP, which in turn caused the accumulation of GFP inside the mitochondrial matrix. For control, fusion protein without MLS, i.e. only composed of GFP and TAT was found to diffuse in and out of mitochondria (and perhaps other cell organelles) displaying pseudofirst order kinetics (Del Gaizo and Payne, 2003).

6. Mitochondriotropics Among the first described mitochondriotropic cationic amphiphiles were phosphonium salts such as methyltriphenylphosphonium bromide, which were shown to be taken up rapidly by mitochondria in living cells (Liberman et al., 1969). The probably best known mitochondriotropic compound is Rhodamine 123, which has been used extensively as a relatively non-toxic stain for mitochondria in living cells since its introduction in 1982 (Chen et al., 1982). Other examples of mitochondriotropic cations are cyanine dyes such as N,N0 -bis (2-ethyl-1,3-dioxolane) kryptocyanine (Oseroff et al., 1986), Victoria Blue BO (Morgan et al., 1998) and dequalinium chloride (Weiss et al., 1987). Mitochondriotropic molecules have two structural features in common. First, they are all amphiphilic, i.e. they combine a hydrophilic charged center with a hydrophobic core. Second, in all structures the p-electron charge density extends over at least three atoms or more instead of being limited to the internuclear region between the heteroatom and the adjacent carbon atom. This causes a distribution of the positive charge density between two or more atoms, i.e. the positive charge is delocalized (hence a commonly used term is ‘delocalized cations’; ‘DLCs’). Both structural features have been recognized to be crucial for the accumulation of these organic cations inside the matrix of mitochondria. Sufficient lipophilicity, combined with delocalization of their positive charge to reduce

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

the free energy change when moving from an aqueous to a hydrophobic environment are prerequisites for their mitochondrial accumulation in response to the mitochondrial membrane potential (Weiss et al., 1987).

7. Mitochondriotropics as mitochondria-specific drug carriers 7.1. Stoichiometric carriers The first attempt to utilize mitochondriotropic molecules as drug carriers were undertaken in the middle of the 1980s. With the goal to develop a new drug which would combine both, the carcinoma cell specific uptake of rhodamine 123 (Bernal et al., 1982) and the anti-carcinoma activity of cis-diamminedichloroplatinum (II) (Kopf-Maier et al., 1981) platinum rhodamine 123, a tight ion pair composed of one platinum (II) tetrachlorodianion and two molecules of rhodamine-123, was synthesized (Teicher et al., 1986). During the same time period, i.e. during the 1980s, it has also been established that the driving forces for the preferred uptake of rhodamine-123 by carcinoma cells are their elevated plasma and mitochondrial membrane potentials, both of which are inside negative, which cause the intracellular and intramitochondrial accumulation of membrane-permeable, cationic molecules (Bernal et al., 1982; Lampidis et al., 1983; Summerhayes et al., 1982). In confirmation, the structural related but uncharged rhodamine 110 does not accumulate in mitochondria (Darzynkiewicz et al., 1982). Therefore, in hindsight, the design of platinum rhodamine 123 appears disputable, considering that the complex is chargeneutral, which makes it largely unresponsive to any electrical field. Nevertheless, the tight complex formation of the hydrophilic cis-diamminedichloroplatinum (II) with the amphiphilic rhodamine residue increased the accumulation of platinum, which entered the cell, by about 70-fold compared to non-complexed cis-diamminedichloroplatinum (II). Apparently, due to its amphiphilic character, rhodamine 123 renders the platinum complex membrane-permeable. But as it can be seen from its intracellular distribution, the charge-neutral platinum

235

rhodamine complex lacks any mitochondria-targeting capability. When measured by fluorescence or by 19 5mPtg-counting, 40– 54% of platinum rhodamine 123 was found in the nuclei and 27 – 35% was detected in the cytosol of SCC-25 or SCC-25/CP cells (Teicher et al., 1986). Since the middle of the 1990s, a variety of stoichiometric conjugates composed of biologically active molecules and the mitochondriotropic triphenylphosphonium cation have been synthesized (Muratovska et al., 2001; Murphy, 2001) to either probe, prevent or alleviate mitochondrial dysfunctions (Fig. 2). In Fig. 2, the conjugates (A), (B) and (E) represent mitochondrially targeted anti-oxidants, conjugates (C) and (D) are mitochondrially targeted thiol reagents and the conjugates (F) and (G) represent peptide nucleic acid oligomere specific for and targeted to the mitochondrial genome. 7.1.1. Mitochondrially targeted anti-oxidants During the mitochondrial generation of ATP by oxidative phosphorylation (OXPHOS), electrons are incessantly being diverted from the respiratory chain to give rise to the formation of free radicals and radical-derived reactive molecules involving nitric oxide, superoxide anions, hydrogen peroxide and related reactive oxygen species (ROS). At moderate concentrations, such reactive species play a role as regulatory mediators in cell signaling processes and many of the ROS-mediated responses actually protect the cell against oxidative stress (Droge, 2002). However, genetic defects of either the nuclear or the mitochondrial genome leading to the inhibition of OXPHOS cause the redirection of electrons from the respiratory chain into a subsequently enhanced ROS production, thus increasing oxidative stress (Wallace, 2001). Decline in mitochondrial energy production and increased oxidative stress in turn may alter the mPTPC thereby triggering apoptosis (Wallace, 2001). Although the exact molecular mechanism by which an increased production of ROS triggers programmed cell death still has to be elucidated, attempts to achieve cell protection using anti-oxidants have already successfully been undertaken, many of them utilizing the avid reactivity of fullerene compounds with free radicals. For example, C60 carboxyfullerene was found to exert a protective activity against oxidative stress-induced apoptosis in human peripheral

236

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

Fig. 2. The mitochondriotropic triphenylphosphonium cation as a stoichiometric carrier for biologically active molecules. All compounds (A)–(G) on the left half of the figure are covalently linked, in a 1:1 stoichiometric ratio, to the triphenylphosphonium residue shown on the right. (A) Active antioxidant moiety of vitamin E. Conjugate: ‘Mito Vit E’, TPPB, 2-[2-(triphenylphosphonio)ethyl]-3,4-dihydro-2,5,7,8-tetramethyl2H-1-benzopyran-6-ol bromide (Smith et al., 1999). (B) Redox-active ubiquinol. Conjugate ðn ¼ 8Þ : ‘Mitoquinol”, 10-(6(-ubiquinolyl)decyltriphenylphosphomium (Kelso et al., 2001). (C) Conjugate: Thiobutyltriphenylphosphonium bromide (TBTP) (Burns et al., 1995). (D) Conjugate: Iodobutyltriphenylphosphonium bromide (IBTP)(Coulter et al., 2000). (E) Fullerene derivative. Conjugate: ‘Mito C60’(Coulter et al., 2000). (F) Peptide nucleic acid (PNA) oligomere with sequence complementary to the human mitochondrial DNA L-chain (np 83398349) containing the A8344G point mutation (Muratovska et al., 2001). (G) Biotinylated PNA oligomere, same as in (F) (Muratovska et al., 2001). From (Weissig et al., 2003), with permission.

blood mononuclear cells (Monti et al., 2000), fullerene C60 in combination with ascorbic acid was shown to protect cultured chromaffin cells against levodopa toxicity (Corona-Morales et al., 2003), C3-fullero-tris-methanodicarboxylic acid was found to protect cerebellar granule cells from apoptosis (Bisaglia et al., 2000) and it was demonstrated that carboxyfullerene is able to attenuate oxidative injuries by transient ischemia-reperfusion in rat brain (Abdelhaleem, 2002).

The increase of mitochondrial concentrations of anti-oxidant drugs by selective targeting anti-oxidants to mitochondria in living cells should therefore be an effective therapy for a wide range of human diseases (Smith et al., 1999). To this end, mitochondriatargeted derivatives of naturally occurring antioxidants (Fig. 2, conjugates (A) and (B)), of the potent buckminsterfullerene anti-oxidant (Fig. 2, conjugate (E)) (Coulter et al., 2000; Kelso et al., 2001; Smith et al., 1999) and of two free radical

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

scavengers, 4-hydroxy-2,2,6,6-tetramethylpiperidinN-oxide (TEMPOL) and Salen-Mn(III) complex of o-vanillin (EUK-134) (Dessolin et al., 2002) have been successfully synthesized and partially tested in term of their anti-oxidant and anti-apoptotic properties. When cells were incubated with micromolar concentrations of the mitochondrially targeted version of vitamin E (TPPB, Fig. 2, conjugate A), they accumulated millimolar concentrations within their mitochondria, which corresponds to a severalhundred fold accumulation of the vitamin E derivative within the mitochondrial matrix (Smith et al., 1999). The amount of TPPB taken up by mitochondria was about 80-fold greater than endogenous levels of vitamin E. As a result, it was found that the mitochondrially targeted version of vitamin E protected mitochondria from oxidative damage induced by iron/ascorbate far more effectively than vitamin E itself, as measured by the level of both, lipid peroxidation (thiobarbituric acid reactive species) and protein damage (protein carbonyls) (Smith et al., 1999). TPPB also decreased the mitochondrial lipid peroxidation induced by tertbutylhydroperoxide. Two aspects of these tertbutylhydroperoxide experiments remarkably confirm the soundness of the triphenylphosphonium cationbased mitochondrial targeting approach. First, the mitochondrial accumulation of the vitamin E-conjugate is driven by the mitochondrial membrane potential. Accordingly, protection against peroxideinduced lipid oxidation was completely abolished by the presence of an uncoupler, which drastically decreases the mitochondrial membrane potential and thereby effectively blocks the uptake of TPPB into the mitochondria (Smith et al., 1999). Second, TPPB accumulates inside the mitochondrial matrix and is therefore unable to prevent peroxidation of lipids in the mitochondrial outer membrane or the outer leaflet of the mitochondrial inner membrane. Consequently, TPPB was less effective at preventing lipid peroxidation in experiments involving these outer membranes than vitamin E itself (Smith et al., 1999). 7.1.2. Mitochondria-specific thiol reagents To analyze putative redox changes of mitochondrial thiol proteins in response to physiological or

237

non-physiological stimuli, two triphenylphosphonium-based mitochondrially targeted thiol reagents have been developed (Burns and Murphy, 1997; Burns et al., 1995). Thiobutyltriphenylphosphonium bromide, TBTP (Fig. 2C) equilibrates with thiol proteins forming mixed disulfides, while iodobutyltriphenylphosphonium bromide, IBTP (Fig. 2D) reacts rapidly with free thiols forming a thioether link that irreversibly attaches the triphenylphosphonium moiety to exposed thiols (Coulter et al., 2000). The expected accumulation of TBTP in isolated mitochondria was confirmed using [14C] labeling. It was found that TBTP equilibrates in the mitochondrial matrix with endogenous thiols and becomes disulfide-bound to protein and non-protein thiols in response to oxidative stress (Burns et al., 1995). Most recently, the corresponding iodo derivative, IBTP, was used to specifically and above all, irreversibly label mitochondrial protein thiols (Lin et al., 2002). Due to the stability of the thioether bond between protein and triphenylphosphonium label, individual mitochondrial proteins that changed their redox state following oxidative insult could be identified by their decreased reaction with IBTP upon their isolation by two-dimensional electrophoresis. It could be shown, for example that exposure to peroxynitrate causes extensive redox changes to thiol proteins inside the mitochondrial matrix. In conjunction with blue native gel electrophoresis, IBTP was used to demonstrate that thiol groups are exposed on the matrix faces of respiratory complexes I, II and IV (Lin et al., 2002). The further use of these mitochondria-specific thiol reagents will open up avenues for exploring the role of redox-linked mitochondrial processes under physiological and pathological conditions. 7.2. Vesicular carriers 7.2.1. Self-assembling mitochondriotropics Dequalinium (DQA, Fig. 3) is a dicationic mitochondriotropic compound resembling bolaform electrolytes, i.e. it is a symmetrical molecule with two charge centers separated at a relatively large distance. Such symmetric bola-like structures are well known from archaeal lipids, which usually consist of two glycerol backbones connected by two hydrophobic chains (De Rosa et al., 1986). The self-assembly

238

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

Fig. 3. Structure of mitochondriotropic cationic bola amphiphiles (dequalinium derivatives) studied for their self-assembly behavior, i.e. for their ability to form liposome-like vesicles. From (Weissig, et al., 2001), with permission.

behavior of bipolar lipids from Archaea has been extensively studied (reviewed in Gambacorta et al. (1995)). It has been shown that these symmetric bipolar archaeal lipids can self-associate into mechanically very stable monolayer membranes. The most striking structural difference between dequalinium and archaeal lipids lies in the number of bridging hydrophobic chains between the polar head groups. Contrary to the common arachaeal lipids, in dequalinium there is only one alkyl chain

that connects the two cationic hydrophilic head groups. Therefore, we have named this type of bola lipids ‘single-chain bolaamphiphile’(Weissig and Torchilin, 1999). The self-assembly behavior of single-chain cationic bola amphiphiles has been thoroughly characterized by us using Monte Carlo computer simulations (Weissig et al., 1998a,b), Transmission and Freeze Fracture Electron Microscopy as well as dynamic laser light scattering (Weissig et al., 1998a,b 2001a – d). We found that

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

dequalinium forms upon sonication in aqueous medium vesicle-like aggregates (named ‘DQAsomes’) with diameters between about 70 and 700 nm. In a SAR study, the relationships between the structure of nine dequalinium-like single-chain bolaamphiphiles (Fig. 3.) and their ability to form vesicles (‘bolasomes’) were examined (Weissig et al., 2001a –d). We found that vesicle stability is significantly increased by substituting the methyl group in dequalinium by an aliphatic ring system (Fig. 3, compound 7). Vesicles made from this cyclohexyl derivative (‘Cyclohexyl-DQAsomes’) have in contrast to vesicles made from dequalinium with 169 ^ 50 nm a very narrow size distribution, which hardly changes at all, even after storage at room temperature for over 5 months. Delivery of pDNA to mitochondria in living mammalian cells using DQAsomes Plasmid DNA commonly used for nuclear-cytosolic gene therapy has a high negative surface charge with a j-potential ranging from 2 30 to 2 50 mV and a hydrodynamic diameter of larger than 100 nm in aqueous colloidal suspension (Ledley, 1996). Therefore, due to its size and charge, plasmid DNA cannot effectively cross phospholipid membranes. Viruses accommodate their large genome within the viral capside, which has usually a volume about 105 times less than that the free viral DNA would occupy. The packaging of DNA into such a small space is achieved by physical condensing with the help of viral proteins able to overcome the resistance of DNA toward condensation, which arises mainly from electrostatic repulsion between the negatively charged phosphates on the polyanion (Ledley, 1996). In an artificial system, the electrostatic barrier to DNA condensation is defeated by the use of multivalent inorganic or organic cations, a phenomenon that has been the basis for the successful development of non-viral transfection vectors utilizing cationic liposomes. The process of DNA packaging (DNA condensing or collapsing) employing cationic liposomes and polymers has extensively been studied and as a result, a variety of models have been proposed to describe the assembly of cationic lipid – DNA complexes, all of which have been most comprehensively reviewed (Duguid and Durlund, 1999). In short, the self-assembly of the cationic lipid into larger structures thus providing a positively charged surface area onto which the DNA

239

can bind and condense is the ultimate prerequisite for any artificial DNA delivery system. In view of that and due to its unique self-assembly into DQAsomes, dequalinium appears to be the molecule of choice for the development of a model for a mitochondriaspecific DNA delivery vector. Accordingly, we have developed a strategy for mitochondrial transfection (Weissig and Torchilin, 2000, 2001a,b), which involves the transport of a DNA-mitochondrial leader sequence (MLS) peptide conjugate to mitochondria using DQAsomes, the liberation of this conjugate from the cationic vector upon contact with the mitochondrial outer membrane followed by DNA uptake via the mitochondrial protein import machinery. In a series of papers, we have demonstrated that DQAsomes fulfill all essential prerequisites for a mitochondria-specific DNA delivery system: they bind and condense pDNA (Weissig et al., 1998), protect it from DNAse digestion, and mediate its cellular uptake (Lasch et al., 1999). DQAsome/DNA complexes (‘DQAplexes’) do not release the DNA upon contact with anionic liposomes mimicking cytoplasm membranes, but do release DNA when in contact with liposomes mimicking mitochondrial membranes (Weissig et al., 2000). The DNA release from DQAplexes at natural mitochondrial membranes was confirmed by incubating DQAplexes with isolated rat liver mitochondria (Weissig et al., 2001a – d) and it was also shown that MLS-peptides linked to DNA do not interfere with DQAsomal binding and release (Weissig et al., 2001a – d). Utilizing a newly developed protocol for selectively staining free pDNA inside the cytosol, we have recently demonstrated that DQAsomes, upon their endosomal escape, selectively deliver pDNA to and release the pDNA exclusively at the site of mitochondria in living mammalian cells (D’Souza, 2003). Free pDNA could not be detected anywhere else in the cytoplasm of cells treated with DQAplexes. Summarizing, our data demonstrate that DQAsomes not only mediate the cellular uptake of pDNA, but also transport the pDNA into the cytosol to the site of mitochondria and release pDNA exclusively upon contact with mitochondrial membranes. This is the ultimate prerequisite for the next step to follow, the uptake of DNA conjugated to a MLS peptide by the mitochondrial protein import machinery.

240

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

Mitochondriotropic vesicles (DQAsomes) as carriers for apoptosis-inducing anti-cancer agents We have recently proposed the use of DQAsomes as a tumor cell- and mitochondria-specific delivery system for drugs known to trigger apoptosis by directly acting on mitochondria (Weissig et al., 2003). As a first step, we studied the encapsulation of paclitaxel into DQAsomes. Paclitaxel is a potent anti-tubulin agent used in the treatment of malignancies (Eisenhauer and Vermorken, 1998). Its therapeutic potential, however, is limited (Seligson et al., 2001) due to a narrow span between the maximal tolerated dose and intolerable toxic levels. In addition, paclitaxel’s poor aqueous solubility requires the formulation of emulsions containing Cremophor ELw, an oil of considerable toxicity by itself (Seligson et al., 2001). It has recently been demonstrated that clinically relevant concentrations of paclitaxel target mitochondria directly and trigger apoptosis by inducing cytochrome-c (cyt c) release in a permeability transition pore (PTP)-dependent manner (Andre et al., 2002). This mechanism of action is known from other pro-apoptotic, directly on mitochondria acting agents (Fulda et al., 1998). A 24-hour delay between the treatment with paclitaxel (Andre et al., 2002) or with other PTP inducers and the release of cyt c in cell-free systems compared to intact cells has been explained by the existence of several drug targets inside the cell making only a subset of the drug available for mitochondria (Andre et al., 2002). Considering that many cancer cells possess in comparison to normal cells both, an elevated mitochondrial and a higher plasma membrane potential (Chen, 1988; Modica-Napolitano and Aprille, 1987, 2001), which cause the selective accumulation of mitochondriotropics in tumor cell mitochondria, the encapsulation of paclitaxel (and other drugs) into DQAsomes would potentially have two advantages. First, with DQAsomes being a colloidal drug delivery system, solubility problems would be overcome. Second, with DQAsomes responding to negativeinside membrane potentials, a ‘double-targeting’ of the drug could potentially be achieved, that is to say, on the cellular level (i.e. carcinoma cells vs. normal cells) and on the subcellular level (i.e. mitochondria vs. rest of the cell). Such ‘double-targeting’ could be the basis for new anti-cancer chemotherapies.

Fig. 4. Transmission electron microscopic image (uranyl acetate staining) of DQAsomal incorporated paclitaxel (0.67 mol paclitaxel/mol dequalinium). Image taken by Bill Fowle, Northeastern University, Boston, MA (from (Weissig et al., 2003) with permission).

Moreover, one could even speculate that hiding the drug inside a vesicle while entering the cell might potentially overcome P-glycoprotein mediated drug resistance. Transmission EM images (Fig. 4.) and cryo-EM images (not shown) of an identical sample show with a remarkable conformity worm- or rod-like structures roughly around 400 nm in length, as confirmed by size distribution analysis (not shown) (Weissig et al., 2003). Interestingly, such structures of aggregates, i.e. tubular clusters, have been predicted by us with Monte Carlo computer simulation of coarse-grained dequalinium (Weissig et al., 1998a,b). The detailed structure of DQAsomal incorporated paclitaxel, however, remains elusive at this time. The formation of worm-like micelles, as recently discussed for self-assembling amphiphilic block copolymers (Discher and Eisenberg, 2002) appears possible. In a most recent study we were able to demonstrate that paclitaxel encapsulated in DQAsomes was able to inhibit human colon carcinoma growth in nude mice by 50% over controls—despite concentrations of paclitaxel and dequalinium where free paclitaxel or empty DQAsomes did not show any effect on tumor growth (S.M. Cheng, S. Pabba, V.P. Torchilin, V. Weissig, manuscript in preparation).

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

8. Concluding remarks Based on significant advancements in our knowledge about the molecular pathology of mitochondrial diseases, increasing pharmacological and pharmaceutical efforts have been undertaken to find effective therapies for disorders associated with malfunctioning mitochondria thus leading to the emergence of Mitochondrial Medicine. In traditional medicine it has been hypothesized almost a century ago that the site-specific, i.e. selective and targeted delivery of drugs will potentially overcome many problems associated with the random distribution of the drug throughout the body. Most important, the dose required to achieve a therapeutically effective concentration at the site of the disease could be reduced when the drug is being selectively delivered to the target site, which in turn would contribute to reducing non-specific toxic side-effects of that drug. Subsequently, the development and marketing of drug delivery technology has become a mainstay for the pharmaceutical industry during the second half of the last century. However, just like any pharmaceutical agent distributes throughout the human body governed only by its own physico-chemical properties and blood flow, any biologically active molecule taken up by a cell will distribute in the cytoplasm depending on its physical and chemical properties and possibly influenced by intracellular trafficking events. Accordingly, the concept of sub-cellular (or intracellular) drug targeting has gained increasingly acceptance in recent years. The authors are aware that a review about ‘Mitochondrial Pharmaceutics’ is not complete without discussing pharmacodynamic and pharmacokinetic aspects relating to drug action at or inside mitochondria, i.e. without addressing what actually will happen to the drug once inside the organelle. Discussing the biochemical and physiological effects of drugs delivered into mitochondria would be beyond the scope of this review, since each drug would have to be dealt with separately, i.e. each delivered drug will have a different therapeutic effect. This review tried only to indicate future therapeutic areas in mitochondrial medicine. The main purpose was to summarize and discuss current strategies, which apply the concept of site-specific drug delivery to

241

a sub-cellular level, i.e. to the selective delivery of drugs and DNA to mitochondria in living mammalian cells. Considering that the area of sub-cellular, i.e. mitochondria-specific delivery of drugs is still in its infancy, it should also be no surprise that hardly any data are available yet about the metabolism and elimination of drugs once transported into mitochondria. An appropriate review about the metabolism and elimination of mitochondria-specific drugs will have to be written at a later time.

Acknowledgements We would like to acknowledge Prof. Vladimir P. Torchilin (Boston, MA) for his numerous helpful discussions. The author’s (V.W.) work was supported in part by a Research Development Grant from the Muscular Dystrophy Association (Tucson, AZ).

References Abdelhaleem, M., 2002. A role for the subcellular localization of Bax in differentiation-induced resistance to apoptosis in HL-60 cells. Anticancer Res. 22, 177–181. Andre, N., Carre, M., Brasseur, G., Pourroy, B., Kovacic, H., Briand, C., Braguer, D., 2002. Paclitaxel targets mitochondria upstream of caspase activation in intact human neuroblastoma cells. FEBS Lett. 532, 256– 260. Behr, J.-P., 2000. Overcoming barriers to nonviral gene delivery. In: Gregoriadis, G., McCormack, B. (Eds.), NATO Science Series: Targeting of Drugs, Strategies for Gene Constracts and Delivery, IOS Press, Washington, DC, pp. 76–80. Bernal, S.D., Lampidis, T.J., Summerhayes, I.C., Chen, L.B., 1982. Rhodamine-123 selectively reduces clonal growth of carcinoma cells in vitro. Science 218, 1117–1119. Bisaglia, M., Natalini, B., Pellicciari, R., Straface, E., Malorni, W., Monti, D., Franceschi, C., Schettini, G., 2000. C3-fullero-trismethanodicarboxylic acid protects cerebellar granule cells from apoptosis. J. Neurochem. 74, 1197–1204. Boussif, O., Lezoualc’h, F., Zanta, M.A., Mergny, M.D., Scherman, D., Demeneix, B., Behr, J.P., 1995. A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc. Natl Acad. Sci. USA 92, 7297–7301. Brown, G.C., Nicholls, D.G., Cooper, C.E., 1999. Mitochondria and Cell Death. Princeton University Press, Princeton, NJ, pp. vii–viii. Burns, R.J., Murphy, M.P., 1997. Labeling of mitochondrial proteins in living cells by the thiol probe thiobutyltriphenylphosphonium bromide. Arch. Biochem. Biophys. 339, 33–39.

242

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

Burns, R.J., Smith, R.A.J., Murphy, M.P., 1995. Synthesis and characterization of thiobutyltriphenylphosphonium bromide, a novel thiol reagent targeted to the mitochondrial matrix. Arch. Biochem. Biophys. 322, 60– 68. Chen, L.B., 1988. Mitochondrial membrane potential in living cells. Annu. Rev. Cell Biol. 4, 155–181. Chen, L.B., Summerhayes, I.C., Johnson, L.V., Walsh, M.L., Bernal, S.D., Lampidis, T.J., 1982. Probing mitochondria in living cells with rhodamine 123. Cold Spring Harb. Symp. Quant. Biol. 46, 141 –155. Chinnery, P.F., Taylor, R.W., Diekert, K., Lill, R., Turnbull, D.M., Lightowlers, R.N., 1999. Peptide nucleic acid delivery to human mitochondria. Gene Ther. 6, 1919–1928. Corona-Morales, A.A., Castell, A., Escobar, A., Drucker-Colin, R., Zhang, L., 2003. Fullerene C60 and ascorbic acid protect cultured chromaffin cells against levodopa toxicity. J. Neurosci. Res. 71, 121–126. Coulter, C.V., Kelso, G.F., Lin, T.K., Smith, R.A., Murphy, M.P., 2000. Mitochondrially targeted antioxidants and thiol reagents. Free Radic. Biol. Med. 28, 1547–1554. Crompton, M., Costi, A., Hayat, L., 1987. Evidence for the presence of a reversible Ca2 þ -dependent pore activated by oxidative stress in heart mitochondria. Biochem. J. 245, 915–918. Crompton, M., Virji, S., Doyle, V., Johnson, N., Ward, J.M., 1999. The mitochondrial permeability transition pore. Biochem. Soc. Symp. 66, 167–179. Darzynkiewicz, Z., Traganos, F., Staiano-Coico, L., Kapuscinski, J., Melamed, M.R., 1982. Interaction of rhodamine 123 with living cells studied by flow cytometry. Cancer Res. 42, 799 –806. De Rosa, M., Gambacorta, A., Gliozi, A., 1986. Structure, biosynthesis, and physicochemical properties of archaebacterial lipds. Microbiologic. Rev. 50, 70–80. Del Gaizo, V., Mark Payne, R., 2003. A novel TAT-Mitochondrial signal sequence fusion protein is processed, stays in mitochondria, and crosses the placenta. Mol. Ther. 7, 720–730. Dessolin, J., Schuler, M., Quinart, A., De Giorgi, F., Ghosez, L., Ichas, F., 2002. Selective targeting of synthetic antioxidants to mitochondria: towards a mitochondrial medicine for neurodegenerative diseases? Eur. J. Pharmacol. 447, 155–161. Discher, E.D., Eisenberg, A., 2002. Polymer vesicles. Science 297, 967–973. Dolce, V., Fiermonte, G., Runswick, M.J., Palmieri, F., Walker, J.E., 2001. The human mitochondrial deoxynucleotide carrier and its role in the toxicity of nucleoside antivirals. Proc. Natl Acad. Sci. USA 98, 2284–2288. Droge, W., 2002. Free radicals in the physiological control of cell function. Physiol. Rev. 82, 47–95. D’Souza, G.G., Rammohan, R., Cheng, S.M., Torchilin, V.P., Weissig, V., 2003. DQAsome-mediated delivery of plasmid DNA toward mitochondria in living cells. J. Control. Release 92, 189 –197. Duguid, J.G., Durlund, R.H., 1999. Advanced Gene Delivery. In: Roland, A., (Ed.), Harwood Academic Publishers, Amsterdam, pp. 45 –63. Eisenhauer, E.A., Vermorken, J.B., 1998. The taxoids. Comparative clinical pharmacology and therapeutic potential. Drugs 55, 5–30.

Fiermonte, G., Palmieri, L., Todisco, S., Agrimi, G., Palmieri, F., Walker, J.E., 2002. Identification of the mitochondrial glutamate transporter. Bacterial expression, reconstitution, functional characterization, and tissue distribution of two human isoforms. J. Biol. Chem. 277, 19289–19294. Fiermonte, G., Dolce, V., Palmieri, L., Ventura, M., Runswick, M.J., Palmieri, F., Walker, J.E., 2001. Identification of the human mitochondrial oxodicarboxylate carrier. Bacterial expression, reconstitution, functional characterization, tissue distribution, and chromosomal location. J. Biol. Chem. 276, 8225–8230. Flierl, A., Jackson, C., Cottrell, B., Murdock, D., Seibel, P., Wallace, D.C., 2003. Targeted delivery of DNA to the mitochondrial compartment via import sequence-conjugated peptide nucleic acid. Mol. Ther. 7, 550 –556. Fulda, S., Susin, S.A., Kroemer, G., Debatin, K.M., 1998. Molecular ordering of apoptosis induced by anticancer drugs in neuroblastoma cells. Cancer Res. 58, 4453–4460. Gambacorta, A., Gliozi, A., De Rosa, M., 1995. Archaeal lipids and their biotechnological applications. World J. Microbiol. Biotechnol. 11, 115–131. Geromel, V., Cao, A., Briane, D., Vassy, J., Rotig, A., Rustin, P., Coudert, R., Rigaut, J.P., Munnich, A., Taillandier, E., 2001. Mitochondria transfection by oligonucleotides containing a signal peptide and vectorized by cationic liposomes. Antisense Nucleic Acid Drug Dev. 11, 175–180. Kelso, G.F., Porteous, C.M., Coulter, C.V., Hughes, G., Porteous, W.K., Ledgerwood, E.C., Smith, R.A., Murphy, M.P., 2001. Selective targeting of a redox-active ubiquinone to mitochondria within cells: antioxidant and antiapoptotic properties. J. Biol. Chem. 276, 4588–4596. Kopf-Maier, P., Wagner, W., Liss, E., 1981. Cytokinetic behavior of Ehrlich ascites tumor after in vivo treatment with cisdiamminedichloroplatinum(II) and metallocene dichlorides. J. Cancer Res. Clin. Oncol. 102, 21 –30. Kroemer, G., 1999. Mitochondrial control of apoptosis: an overview. In: Brown, G.C., Nicholls, D.G., Cooper, C.E. (Eds.), Mitochondria and Cell Death, Princeton University Press, Princeton, NJ, pp. 1–15. Lampidis, T.J., Bernal, S.D., Summerhayes, I.C., Chen, L.B., 1983. Selective toxicity of rhodamine 123 in carcinoma cells in vitro. Cancer Res. 43, 716– 720. Larson, N.-G., Luft, R., 1999. Revolution in mitochondrial medicine. FEBS Lett. 455, 199–202. Lasch, J., Meye, A., Taubert, H., Koelsch, R., Mansa-ard, J., Weissig, V., 1999. Dequalinium vesicles form stable complexes with plasmid DNA which are protected from DNase attack. Biol. Chem. 380, 647–652. Ledley, F.D., 1996. Pharmaceutical approach to somatic gene therapy. Pharm. Res. 13, 1595–1614. Liberman, E.A., Topaly, V.P., Tsofina, L.M., Jasaitis, A.A., Skulachev, V.P., 1969. Mechanism of coupling of oxidative phosphorylation and the membrane potential of mitochondria. Nature 222, 1076–1078. Lin, T.K., Hughes, G., Muratovska, A., Blaikie, F.H., Brookes, P.S., Darley-Usmar, V., Smith, R.A., Murphy, M.P., 2002. Specific modification of mitochondrial protein thiols in response to

V. Weissig et al. / Mitochondrion 3 (2004) 229–244 oxidative stress: a proteomics approach. J. Biol. Chem. 277, 17048– 17056. Modica-Napolitano, J.S., Aprille, J.R., 1987. Basis for the selective cytotoxicity of rhodamine 123. Cancer Res. 47, 4361–4365. Modica-Napolitano, J.S., Aprille, J.R., 2001. Delocalized lipophilic cations selectively target the mitochondria of carcinoma cells. Adv. Drug Deliv. Rev. 49, 63–70. Modica-Napolitano, J.S., Singh, K.K., 2002. 11 April mitochondria as targets for detection and treatment of cancer. Expert Rev. Mol. Med., 1–19. Monti, D., Moretti, L., Salvioli, S., Straface, E., Malorni, W., Pellicciari, R., Schettini, G., Bisaglia, M., Pincelli, C., Fumelli, C., Bonafe, M., Franceschi, C., 2000. C60 carboxyfullerene exerts a protective activity against oxidative stress-induced apoptosis in human peripheral blood mononuclear cells. Biochem. Biophys. Res. Commun. 277, 711–717. Morgan, J., Whitaker, J.E., Oseroff, A.R., 1998. GRP78 induction by calcium ionophore potentiates photodynamic therapy using the mitochondrial targeting dye victoria blue BO. Photochem. Photobiol. 67, 155–164. Morin, D., Hauet, T., Spedding, M., Tillement, J., 2001. Mitochondria as target for antiischemic drugs. Adv. Drug Deliv. Rev. 49, 151–174. Muratovska, A., Lightowlers, R.N., Taylor, R.W., Wilce, J.A., Murphy, M.P., 2001. Targeting large molecules to mitochondria. Adv. Drug Deliv. Rev. 49, 189–198. Murphy, M.P., 1997. Selective targeting of bioactive compounds to mitochondria. Trends Biotechnol. 15, 326–330. Murphy, M.P., 2001. Development of lipophilic cations as therapies for disorders due to mitochondrial dysfunction. Expert Opin. Biol. Ther. 1, 753–764. Murphy, M.P., Smith, R.A., 2000. Drug delivery to mitochondria: the key to mitochondrial medicine. Adv. Drug Deliv. Rev. 41, 235–250. Oseroff, A.R., Ohuoha, D., Ara, G., McAuliffe, D., Foley, J., Cincotta, L., 1986. Intramitochondrial dyes allow selective in vitro photolysis of carcinoma cells. Proc. Natl Acad. Sci. USA 83, 9729–9733. Palmieri, F., Bisaccia, F., Capobianco, L., Dolce, V., Fiermonte, G., Iacobazzi, V., Indiveri, C., Palmieri, L., 1996. Mitochondrial metabolite transporters. Biochim. Biophys. Acta 1275, 127–132. Schon, E.A., DiMauro, S., 2003. Medicinal and genetic approaches to the treatment of mitochondrial disease. Curr. Med. Chem. 10, 1241–1253. Seligson, A.L., Terry, R.C., Bressi, J.C., Douglass, J.G. III, Sovak, M., 2001. A new prodrug of paclitaxel: synthesis of Protaxel. Anticancer Drugs 12, 305–313. Singh, K.K., 2000. Mitochondrial me and the mitochondrion journal. Mitochondrion 1, 1–2. Smith, R.A., Porteous, C.M., Coulter, C.V., Murphy, M.P., 1999. Selective targeting of an antioxidant to mitochondria. Eur. J. Biochem. 263, 709–716. Summerhayes, I.C., Lampidis, T.J., Bernal, S.D., Nadakavukaren, J.J., Nadakavukaren, K.K., Shepherd, E.L., Chen, L.B., 1982. Unusual retention of rhodamine 123 by mitochondria in muscle and carcinoma cells. Proc. Natl Acad. Sci. USA 79, 5292–5296.

243

Teicher, B.A., Holden, S.A., Jacobs, J.L., Abrams, M.J., Jones, A.G., 1986. Intracellular distribution of a platinum-rhodamine 123 complex in cis-platinum sensitive and resistant human squamous carcinoma cell lines. Biochem. Pharmacol. 35, 3365–3369. Torchilin, V.P., 2000. Drug targeting. Eur. J. Pharm. Sci. 2 (Suppl 2), S81–S91. Wallace, D.C., 2001. A mitochondrial paradigm for degenerative diseases and ageing. Novartis Found Symp. 235, 247 –263.discussion 263– 246. Wallace, K.B., Starkov, A.A., 2000. Mitochondrial targets of drug toxicity. Annu. Rev. Pharmacol. Toxicol. 40, 353–388. Weiss, M.J., Wong, J.R., Ha, C.S., Bleday, R., Salem, R.R., Steele, G.D. Jr, Chen, L.B., 1987. Dequalinium, a topical antimicrobial agent, displays anticarcinoma activity based on selective mitochondrial accumulation. Proc. Natl Acad. Sci. USA 84, 5444–5448. Weissig, V., Torchilin, V.P., 1999. Cationic single-chain bolaamphiphiles as new materials for application in medicine and biotechnology, Materials Research Society Meeting, Boston, MA (USA). Weissig, V., Torchilin, V.P., 2000. Mitochondriotropic cationic vesicles: a strategy towards mitochondrial gene therapy. Curr. Pharm. Biotechnol. 1, 325 –346. Weissig, V., Torchilin, V.P., 2001a. Towards mitochondrial gene therapy: DQAsomes as a strategy. J. Drug Target 9, 1–13. Weissig, V., Torchilin, V.P., 2001b. Cationic bolasomes with delocalized charge centers as mitochondria-specific DNA delivery systems. Adv. Drug Deliv. Rev. 49, 127–149. Weissig, V., Lizano, C., Torchilin, V.P., 2000. Selective DNA release from DQAsome/DNA complexes at mitochondria-like membranes. Drug Deliv. 7, 1–5. Weissig, V., D’Souza, G.G., Torchilin, V.P., 2001a. DQAsome/ DNA complexes release DNA upon contact with isolated mouse liver mitochondria. J. Control. Release 75, 401 –408. Weissig, V., Mogel, H.J., Wahab, M., Lasch, J., 1998aa. Computer simulations of DQAsomes, International Symposium on Controlled Relative Bioactive Materials. Controlled Release Society, Las Vegas, Nevada, USA, pp. 312. Weissig, V., Lasch, J., Erdos, G., Meyer, H.W., Rowe, T.C., Hughes, J., 1998b. DQAsomes: a novel potential drug and gene delivery system made from Dequalinium. Pharm. Res. 15, 334 –337. Weissig, V., Lizano, C., Ganellin, C.R., Torchilin, V.P., 2001b. DNA binding cationic bolasomes with delocalized charge center: a structure–activity relationship study. S.T.P. Pharma Sciences 11, 91–96. Weissig, V., Seibel, P., Seibel, M., Torchilin, V.P., 2001c. Binding and release of DNA-peptide conjugates by cationic mitochondriotropic vesicles (DQAsomes), International Symposium on Controlled Release Bioactive Materials. Controlled Release Society, San Diego, CA (USA), pp. 850–851. Weissig, V., Cheng, S.M., D’Souza, G.G., Torchilin, V.P., Schubert, R., Kimpfler, A., 2003. Towards mitochondriaspecific delivery of apoptosis-inducing agents: DQAsomal incorporated paclitaxel, International Symposium on Controlled

244

V. Weissig et al. / Mitochondrion 3 (2004) 229–244

Bioactive Materials. Controlled Release Society, Glasgow, Scotland. Weissig, V., Torchilin, V.P., Eds, ., 2001d. Drug and DNA delivery to mitochondria. Adv. Drug Deliv. Rev. 49, 1–215.Theme Issue. Woodfield, K., Ruck, A., Brdiczka, D., Halestrap, A.P., 1998. Direct demonstration of a specific interaction between cyclophilin-D and the adenine nucleotide translocase confirms their role in the mitochondrial permeability transition. Biochem. J. 336 (Pt 2), 287–290.

Xu, Y., Szoka, F.C. Jr, 1996. Mechanism of DNA release from cationic liposome/DNA complexes used in cell transfection. Biochemistry 35, 5616–5623. Zabner, J., Fasbender, A.J., Moninger, T., Poellinger, K.A., Welsh, M.J., 1899. Cellular and molecular barriers to gene transfer by a cationic lipid. J. Biol. Chem. 270, 7 –19007. Zanta, M.A., Belguise, P., Behr, J.-P., 1999. Gene delivery: a single nuclear localization signal (NLS) peptide is sufficient to carry DNA to the cell nucleus. Proc. Natl Acad. Sci. USA 96, 91 –96.