Review
Approaches to mitochondrially mediated cancer therapy

https://doi.org/10.1016/j.semcancer.2007.12.006Get rights and content

Abstract

For some malignant cancers even combined surgical, radiotherapeutic and chemotherapeutic approaches are not curative, indeed, in certain tumour types even a modest survival benefit is difficult to achieve. There are various biological reasons which underlie this profound resistance but the propensity of cancer cells to repair breaks caused by DNA-damaging radiation and cytotoxic drugs is of major significance in this context. Such highly resistant tumours include the malignant gliomas which are intrinsic to and directly affect the brain and spinal cord. In evaluating approaches which do not elicit tumour cell death directly by DNA damage, it is intriguing to consider mitochondrially mediated apoptosis as a potentially effective alternative. Since the mitochondrial membrane potentials in cancer cells are frequently reduced in comparison with those of non-neoplastic cells this allows a window of opportunity for small molecule agents to enter the tumour cell mitochondria and reduce oxygen consumption with subsequent release of cytochrome c and activation of a caspase pathway to apoptosis which is cancer cell specific. In the quest for agents which can selectively destroy neoplastic cells in this manner, whilst leaving normal adjacent cells intact, various tricyclic drugs have come under scrutiny. In a range of laboratory assays we, and others, have established that certain cancers and, in particular, malignant glioma, are intrinsically sensitive to this approach. We have also established the cellular, molecular and biochemical mechanisms underlying this process. While such archival tricyclics as the antidepressants, clomipramine and amitriptyline, have been used in these experiments their commercial development in cancer therapy has not been forthcoming and their clinical use in glioma has been confined to anecdotal cases. In addition, the dose-dependant role of agents such as anticonvulsants and steroids commonly used in glioma patients in modulating efficacy of the tricyclics is a matter for continued investigation. Other ways of targeting the mitochondrion for cancer therapy include exploitation of the 18 kDa translocator protein (peripheral-type benzodiazepine receptor) within the mitochondrial permeability transition pore and enzymatic or molecular modification of a species of ganglioside (GD3/GD3A) expressed on the surface of neoplastic cells which are determinants of mitochondrially mediated apoptosis. It is hoped that such approaches may lead to clinical programmes which will improve the prognosis for patients suffering from highly resistant neoplasms.

Introduction

Therapeutic approaches for highly resistant malignancies have, in general, centred around various adjuvant regimens involving surgical resection and cytotoxic radiotherapy and chemotherapy. Cytotoxic drug therapy has, for the most part, focussed upon agents which damage the nuclear DNA of tumour cells. Sadly, not only are some normal cells as well as the neoplastic cells likely to be destroyed by this approach but, significantly, cells which comprise some of most aggressive and resistant tumours are highly efficient at repairing DNA damage breaks. Among these highly resistant neoplasms are the intrinsic, primary brain tumours; the vast majority of which are thought to be derived from the glial cells or their progenitors and are generically classified as glioma. These are graded from I to IV in increasing degree of histologically defined malignancy, with grade I and II being deemed “benign”. Biologically however, any expansive, space-occupying lesion within the cranium must be considered as potentially life-threatening, particularly if it encroaches on eloquent or vital areas of the brain. Moreover, in adults such “grade II, benign” tumours almost invariably undergo progressive genetic change and subsequently represent as high grade malignant tumours which are then among the most difficult of tumours to treat effectively and carry a dismal prognosis. To further complicate the issue, the presence of the blood–brain barrier (B–BB) precludes access for all but lipid-soluble or low molecular weight therapeutic agents to the brain. The damaged vasculature of the major tumour mass does, however, to some extent permit modest entry of certain cytotoxic agents which do not constitutively cross the B–BB. This too is counteracted by the ability of tumour cells at the periphery of the lesion to diffusely infiltrate deep (often several centimetres) into the contiguous normal brain where they are protected by virtue of an intact B–BB. In addition, these invading neoplastic cells transiently arrest from the cell cycle during their migratory phase thereby rendering themselves refractory to radiotherapy [1], [2]. Yet another biological feature of the gliomas contributes to the difficulty posed in achieving a successful therapeutic response; their marked cellular heterogeneity. Not only may gliomas demonstrate a mixed cell population (e.g. astrocytes and oligodendrocytes within the same tumour), but there are known, through both histological and in vitro studies, to be numerous sub-populations present within the same phenotype (e.g. astrocytes). Here, tumour cells identified by the presence of the astrocyte marker, glial fibrillary acidic protein (GFAP) (Fig. 1a), show distinct “clones” displaying various morphologies (e.g. polygonal, fusiform, stellate, etc.) with differential expression of cell surface molecules, enzyme and growth factor synthesis and secretion. Indeed, these sub-groups of tumour cells can interact in a paracrine fashion to drive behaviour and malignant potential of the tumour [3], [4], [5].

This complex biology leaves currently used traditional chemotherapy approaches offering little long-term benefit for most malignant glioma patients and treatment is frequently associated with adverse side-effects that diminish the quality of life. Hence, new approaches are needed that can provide long-term management for malignant brain tumours, whilst permitting a decent quality of life [6]. In order to overcome the difficulties arising through the biological behaviour of primary brain tumours which severely hampers current therapeutic strategies it may be useful to consider the pre-requisites of an ideal agent to treat glioma:

  • (a)

    Passage across the B–BB.

  • (b)

    Low toxicity/side effects.

  • (c)

    Selective cytotoxicity—ability to destroy tumour but not normal brain cells.

  • (d)

    Trigger apoptosis through a mechanism not involving the nuclear DNA.

In consideration of the above features it has long been known that an alternative cellular pathway for chemotherapy exists—via the tumour cell mitochondria.

Section snippets

Mitochondria as cancer therapeutic targets

The rationale for targeting the mitochondria for therapeutic gain lies in the knowledge that this organelle plays a critical role in the regulation of energy metabolism, reactive oxygen species (ROS) production and apoptosis. It would appear difficult to target the energy metabolism of tumours as the host cells also rely upon the essential pathways for ATP supply. By identifying the most significant differences in the energy metabolism between tumour cells and normal brain cells, termed

Tricyclic drugs

In our laboratories we have investigated, in vitro, the efficacy of several alternative approaches to cancer therapy using the mitochondrial route. In particular, we and others, have shown that certain tricyclic antidepressants (Table 1) which have been used to competitively inhibit the serotonin and norepinephrine transporters for over 40 years, possess the potential to induce apoptosis in malignant glioma and neuroblastoma cells in vitro [18], [19], [20]. These lipophilic compounds, which

Cancer stem cells

The 120 kDa five-transmembrane cell surface protein, CD133 originally shown to be a haematopoietic stem cell marker [24], [25] has also been localised on normal human neural stem cells [26]. Subsequently, Singh et al. [27] demonstrated CD133 on “cancer stem cells” in brain tumour. Such CD133-positive, brain tumour stem cells can be magnetic immunobead and fluorescence activated cell sorted by use of dissociated cell suspensions and are also seen to express nestin but fail to express markers

Dexamethasone and valproic acid

In the UK most malignant glioma patients receive the steroid dexamethasone to reduce raised intracranial pressure. Current literature on the effects of dexamethasone on glioma cells is conflicting. Glucocorticoids have been reported to fulfil a functional role at the level of the mitochondrion [31] and have been shown to be neurotoxic, playing a role in neuronal cell death [32]. Dexamethasone also enhances necrotic cell death in glioma cells following serum deprivation [32]. In addition, it

Gangliosides

While drug-based approaches to induction of mitochondrially mediated apoptosis in cancer treatment carry potential there exist further possibilities to the same ends by modulating endogenous cellular processes using gene-based strategies. In this context, simple gangliosides may warrant further attention. Gangliosides are acidic glyco-sphingolipids that are present in high concentration, and are involved in, the early developmental stages of the brain when simple gangliosides such as GD3 are

Blood samples/clomipramine distribution

Blood plasma samples are taken at regular intervals, from clomipramine-treated anecdotal brain tumour patients and analysed using standard high-performance liquid chromatography (HPLC), to detect both clomipramine and its metabolite norclomipramine. A methodology is currently being developed for the measurement of dexamethasone via HPLC, and amitriptyline/nortriptyline can potentially be added to the range of tricyclics tested. The data taken from the analysis of blood plasma will be used to

Conclusions

The mitochondrial route to apoptosis in cancer treatment is, to a large extent, unexploited but there is, with some justification, an increased research activity in this area. In our own laboratories we have considered the potential of tricyclic drugs, hitherto used in control of obsessive compulsive disorder and clinical depression, in the treatment of malignant glioma. These drugs have only limited, and generally mild, side effects, cross the B–BB and are sequestered in the lung and brain in

Acknowledgements

We are indebted to the Samantha Dickson Brain Tumour Trust for supporting the tricyclic drug research presented and to Ali's Dream and Charlie's Challenge brain tumour charities for provision of ongoing funding for our ganglioside research programme also reported herein.

References (61)

  • K. Morita et al.

    Dexamethasone enhances serum deprivation-induced necrotic death of rat C6 glioma cells through activation of glucocorticoid receptors

    Brain Res

    (1999)
  • L. Goya et al.

    Effect of growth factors on the in vitro growth and differentiation of early and late passage C6 glioma cells

    Int J Dev Neurosci

    (1996)
  • K. Renner et al.

    Changes of mitochondrial respiration, mitochondrial content and cell size after induction of apoptosis in leukaemia cells

    Biochim Biophys Acta

    (2003)
  • A.M. Gorman et al.

    Dexamethasone pre-treatment interferes with apoptotic death in glioma cells

    Neuroscience

    (2000)
  • W.D. Beecken et al.

    Valproic acid modulates NCAM polysialylation and polysialyltransferase mRNA expression in human tumor cells

    Int Immunopharmacol

    (2005)
  • I. Kracun et al.

    Gangliosides in the human brain development and aging

    Neurochem Int

    (1992)
  • S. Hakomori et al.

    Glycosphingolipid-dependent cross-talk between glycosynapses interfacing tumour cells with their host cells: essential basis to define tumour malignancy

    FEBS Lett

    (2002)
  • S. Ladisch et al.

    Detection of medulloblastoma and astrocytoma-associated ganglioside GD3 in cerebrospinal fluid

    Cancer Lett

    (1997)
  • S. Koochekpour et al.

    Growth factors and gangliosides stimulate laminin production by glioma cells in vitro

    Neurosci Lett

    (1995)
  • S.L. Maidment et al.

    The effect of exogenous gangliosides on matrix metalloproteinase secretion by human glioma cells in vitro

    Eur J Cancer

    (1996)
  • S. Koochekpour et al.

    Vascular endothelial growth factor production is stimulated by gangliosides and TGF-ß isoforms in human glioma cells in vitro

    Cancer Lett

    (1996)
  • H. Satake et al.

    Genes modulated by expression of GD3 synthase in Chinese hamster ovary cells. Evidence that the Tis21 gene is involved in the induction of GD3 9-O-acetylation

    J Biol Chem

    (2003)
  • H. Takematsu et al.

    Lysosomal and cytosolic sialic acid 9-O-acetylesterase activities can be encoded by one gene via differential usage of a signal peptide-encoding exon at the N terminus

    J Biol Chem

    (1999)
  • G.J. Pilkington

    Tumour cell migration in the CNS

    Brain Pathol

    (1994)
  • A.J. Bolteus et al.

    Migration and invasion in brain neoplasms

    Curr Neurol Neurosci Rep

    (2001)
  • G.J. Pilkington

    Glioma heterogeneity in vitro: the significance of growth factors and gangliosides

    Neuropathol Appl Neurobiol

    (1992)
  • J.C.A. Knott et al.

    A human glioma cell line retains expression of GFAP and gangliosides recognised by A2B5 and LB1 antibodies, after prolonged passage

    Neuropathol Appl Neurobiol

    (1990)
  • P.G. Fisher et al.

    Malignant gliomas in 2005: where to GO from here?

    JAMA

    (2005)
  • R. Moreno-Sanchez et al.

    Energy metabolism in tumor cells

    FEBS J

    (2007)
  • J.S. Armstrong

    Mitochondrial medicine: pharmacological targeting of mitochondria in disease

    Br J Pharmacol

    (2007)
  • Cited by (59)

    • Metal complexes targeting the Translocator Protein 18 kDa (TSPO)

      2017, Coordination Chemistry Reviews
      Citation Excerpt :

      A prolonged pore opening induces permeabilization of IMM and consequent osmotic swelling of the matrix with release of Cyt c and of the apoptosis inducing factor (AIF). These two pro-apoptotic proteins cause the activation of intracellular pathways responsible for the typical nuclear alterations of apoptotic cells [4,92,93]. Myxoma Poxvirus M11L is an anti-apoptotic protein, localized in the mitochondria, which can regulate the MPT pore complex by direct modulation of TSPO, thus preventing the loss of the mitochondrial inner membrane potential preluding to induction of apoptosis [94].

    • Biocatalytic conversion and detoxification of imipramine by the laccase-mediated system

      2016, International Biodeterioration and Biodegradation
      Citation Excerpt :

      They can induce apoptosis in the cells of malignant glioma and neuroblastoma in vitro and zebra fish embryos. These effects result from restriction of the electron transport chain in mitochondria (Pilkington et al., 2008; Cordero et al., 2009; Parker et al., 2012). Also, TCAs are metabolized and conjugated by the liver, where some of them become pharmacologically active, and excreted by the kidneys (Kerr et al., 2001).

    • Mitochondrial dysfunction in gliomas

      2013, Seminars in Pediatric Neurology
      Citation Excerpt :

      In a small clinical trial of patients with glioblastoma, DCA was shown to decrease tumor growth and angiogenesis.110 Various tricyclic antidepressant drugs, such as clomipramine and amitriptyline, can enter the mt of tumor cells causing reduction of oxygen consumption, release of cytochrome c, and activation of the caspase cascade hence exerting their tumoricidal action through apoptosis in vitro; as such, these drugs can be used as potential anticancer agents aimed at targeting mt in gliomas.113,114 Clomipramine and its metabolite norclomipramine have been found to be potent inhibitors of cellular respiration in anaplastic astrocytoma cells in vitro thereby enhancing apoptotic cell death, an effect that is synergistically accentuated by dexamethasone.114

    View all citing articles on Scopus
    View full text