Mol Biotechnol (2010) 46:308–316 DOI 10.1007/s12033-010-9321-2

REVIEW

Multiple Drug Resistance Mechanisms in Cancer Bruce C. Baguley

Published online: 18 August 2010 Ó Springer Science+Business Media, LLC 2010

Abstract Multiple drug resistance (multidrug resistance; MDR), a phenomenon whereby human tumours that acquire resistance to one type of therapy are found to be resistant to several other drugs that are often quite different in both structure and mode of action, has been recognised clinically for several decades. An important advance in our understanding of MDR came with the identification of P-glycoprotein and other related transporters that were expressed in some cancer cells and could recognise and catalyse the efflux of diverse anticancer drugs from cells. A second advance came from an understanding of the mechanism of programmed cell death or apoptosis, leading to MDR mediated by increased to resistance to anticancer drug-induced apoptosis. A third advance came with the finding that the proliferation of human tumours was driven by a small population of self-renewing tumour cells, focussing attention on the MDR properties of these so-called tumour stem cells rather than on the cells that comprised the majority of the tumour population. A fourth advance was the delineation of features of the tumour microenvironment, including immunosuppression, which essentially provided tumour stem cells with an MDR phenotype. Most published work on the overcoming of MDR has concentrated on inhibition of drug transporters but the complexity of mechanisms contributing demands a broad strategy for the development of methods to overcome MDR in a clinical setting.

B. C. Baguley (&) Auckland Cancer Society Research Centre, The University of Auckland, Private Bag 92019, Auckland 1142, New Zealand e-mail: [email protected]

Keywords Cytokinetics  ABC transporters  Drug diffusion  Apoptosis  Tumour dormancy  Macrophages  Niche  Microenvironment

Introduction Multiple drug resistance (MDR) describes a phenomenon whereby resistance to one drug is accompanied by resistance to drugs whose structures and mechanisms of action may be completely different. The term was first applied to the area of antibiotic-resistant infection [1] but was soon applied to cancer chemotherapy. The identification of mechanisms of MDR in human cancer led to a corresponding development of therapeutic agents that might overcome MDR. There are many current reviews of MDR and of agents that overcome it [2–6]. The clinical problem of MDR can be divided into two main types, one acquired during treatment and the other pre-existing at the time of diagnosis. This can be illustrated by two theoretical examples; in the first, a woman is diagnosed with advanced ovarian cancer. Chemotherapy is commenced using combined carboplatin and paclitaxel and a complete remission is obtained. After an interval of 1 year, an abdominal mass is detected and combination therapy is reinstituted. However, in this case, there is no significant reduction in tumour mass and after four cycles, treatment with irinotecan is initiated. No response is obtained, and treatment is continued with doxorubicin, again with no response. In the second example, a patient diagnosed with metastatic pancreatic cancer is treated with the drug gemcitabine. Both the primary tumour and a lymph node metastasis continue to grow, and chemotherapy is changed to a combination of 5-fluorouracil and oxaliplatin, but again with no effect on tumour progression.

Mol Biotechnol (2010) 46:308–316

It is important to realise that multiple mechanisms are likely to contribute to clinical MDR. Historically, the first significant advance in our understanding of MDR came with the identification of the membrane transporter P-glycoprotein, followed by that of other transporters of the ATP-binding cassette (ABC) family, which could catalyse the efflux of a range of structurally unrelated anticancer drugs. However, clinical studies showed that clinical MDR applied to anticancer drugs that were not susceptible to the above transporters, suggesting that other MDR mechanisms were important. One such mechanism, termed ‘atypical MDR’ and caused by reduced cellular activity of topoisomerase II, was applied to a structurally varied class of topoisomerase poisons such as doxorubicin and etoposide [7] but did not explain the breadth of MDR observed in clinical studies. Resistance to apoptosis induction was identified as a much more ubiquitous resistance mechanism that could apply equally to targetted anticancer drugs and cytotoxic agents and was consistent with clinical studies. The identification of above two MDR mechanisms raised the question of whether either of these mechanisms could be identified in clinical tumour samples. A critical advance relevant to answering this question was the finding that the proliferation of human tumours was driven by a minor population of self-renewing tumour cells, often termed tumour stem cells. Evidence for the expression of MDR mechanisms by such stem cells stimulated further interest in strategies to overcome MDR. The concept that the resistance properties of tumour stem cells will dictate the resistance of the overall tumour also highlights two methodological problems in the investigation of resistance; cancer stem cells within a tumour population cannot easily be identified within a tumour population, and their properties may depend critically on their niche environment. The host stromal cells in the niche microenvironment, particularly those of fibroblast origin, provide soluble and matrix-linked factors to inhibit cell division and apoptosis while simultaneously preserving a primitive multipotent phenotype [8]. This microenvironment has been particularly well characterised for the bone marrow [9], where the low oxygen tension suggests diminished perfusion of oxygen and probably also diminished perfusion of potential toxins and mutagens [10]. Stem cells appear to express multiple transport proteins of ABC family that exclude toxins and mutagens [11]. In addition, stem cells express pathways such as NF-jB and bcl-2 that protect them from the induction of apoptosis [12]. The tumour microenvironment also includes immune cells and the finding that murine tumours could be sensitive to a given therapy when grown in immunoproficient mice but resistant when grown in immunodeficient mice [13] demonstrated a further factor in the tumour microenvironment that must be taken into account in studying MDR.

309

The aim of this review is to consider both the mechanisms that contribute to MDR and the possible approaches that can be used to overcome these mechanisms, with particular emphasis on tumour stem cell concepts. While most reviews on MDR have concentrated only tumour cells, this review also includes discussion of extrinsic contributions to MDR. The emerging picture is a complex one, implying that the development of strategies to combat clinical MDR needs considerable further research and mechanistic insight.

Tissue Diffusion in Drug Resistance Before considering intrinsic cellular MDR mechanisms, it is important to remember that regardless of whether anticancer drugs are administered orally, intravenously, intraarterially or by other routes, they must diffuse from the bloodstream to individual tumour stem cells. The establishment of pharmacological sanctuaries of tumour stem cells that are protected from therapeutic drugs by a diffusion barrier might therefore appear clinically as MDR. The vascular density of the tumour, which determines the mean diffusion distance from blood supply to tumour stem cells, will have a major effect on diffusion time for some drugs. As with the case of normal bone marrow, there is evidence that tumour stem cells may exist in a state of low oxygen tension, suggesting the presence of a perfusion barrier which could limit the rate of penetration of anticancer drugs [10]. Tumour hypoxia may also be either intrinsic (related to vascular geometry) or intermittent (related to temporal changes in tumour blood flow) and both of these states may contribute to drug access and efficacy. Drug diffusion depends not only on tumour architecture and dynamics but also on drug properties such as the molecular weight and degree of protein binding. Albumin is a common binding protein that accounts for the low free drug fraction of many drugs and a-acid glycoprotein, an acute phase protein commonly elevated in cancer patients, can also have a major effect on free drug concentrations [14]. The free drug fraction can be quite low (\5%) for some anticancer drugs, particularly complex organic molecules, and this will affect the rate of extracellular diffusion [15]. Drugs move through cancer tissue not only by extracellular diffusion but also by cellular uptake and efflux, with rapid trans-cellular diffusion enhancing their ability to reach their target. The latter process is controlled by multiple processes such as the rate of passage between the extracellular fluid and the cytoplasm, the degree of binding to macromolecules within the cell, the rate of sequestration into intracellular vesicles and the rate of cellular efflux. Some anticancer drugs require transporters for uptake/ efflux, and the activity of this transporter may therefore

310

regulate the efficiency of drug delivery; an example is the copper transporter ATP7B in the case of cisplatin, carboplatin and oxaliplatin [16]. Drug efflux transporters of the ABC family, which are generally found both on the plasma membrane and on the membranes of cellular vesicles (see MDR mechanisms controlling drug efflux from cells section), can reduce the efficiency of trans-cellular diffusion of drugs by promoting their efflux from the cell or sequestration into vesicles. An example of the latter process is provided by the sequestration of cisplatin into melanosomes of melanoma cells [17]. Intracellular protein binding and DNA binding will also influence the rate of transcellular diffusion by reducing cytoplasmic free drug concentration. It is difficult to design treatment strategies to overcome diffusion resistance, i.e. the existence of pharmacological sanctuaries. One approach is to select drugs with high tissue diffusion rates; another is to use drugs that utilise cellular influx transporters to promote diffusion through tumour cells as well as through the extracellular matrix. A useful principle is to design simultaneously for high efficiency of tissue diffusion and long residence within the tumour tissue, thus allowing time for diffusion to all tumour compartments including those containing tumour stem cells. The use of slow drug infusion rates may facilitate good distribution within regions of tissue that have an inefficient and sometimes intermittent blood supply.

MDR Mechanisms Controlling Drug Efflux from Cells Studies with tumour cell lines have identified transporters that act either to promote drug efflux from the cell or to sequester it to cellular vesicles that are later eliminated by exocytosis. The first such transporter to be identified was P-glycoprotein, a membrane-associated protein typically found on the plasma membrane. Like albumin, P-glycoprotein was able to bind to a broad variety of small molecules, particularly those containing hydrophobic domains and positively charged areas [18], but in contrast to albumin, P-glycoprotein was able to carry out an ATP-dependent conformational change that moves the substrate to the exterior of the cell. At least, 48 structurally related transporters, known collectively as the ABC family, are known [19], and the three subfamilies concerned with drug transport are the ‘B’ subfamily that includes P-glycoprotein, the ‘C’ subfamily that includes MRP-related (multi-resistance protein related) transporters and the ‘G’ subfamily that includes the ABCG2, MXR and ABCP proteins. Some of these transporters act directly on the drug while others act on conjugates in concert with cellular conjugating enzymes that first link the drug to glutathione, glucuronide or sulphate. Expression of ABCG2 and MRP1

Mol Biotechnol (2010) 46:308–316

has been reported in side populations of tumour cells lines with some of the characteristics of stem cells [20], consistent with the hypothesis that tumour stem cells commonly express these drug efflux proteins [21, 22]. The expression of drug transporters that can affect the efficacy of a broad range of anticancer drugs in tumour stem cells is also well established [23, 24].

Overcoming of Transport MDR There are two main approaches to overcoming transport MDR [4]. The first is to co-administer a drug that inhibits the action of the transporter [3] while the second is to design or utilise anticancer drugs whose activity is not significantly altered by the presence of transporters [7]. A variant of this approach is to co-administer a second drug that increases the cellular uptake of the anticancer drug: this has been examined in the case of paclitaxel through the administration of releasable octaarginine transporters [5]. Drugs reversing transport-mediated MDR are often referred to as first, second and third generation inhibitors. First generation inhibitors include verapamil, cyclosporine A, tamoxifen and calmodulin antagonists but their use involved comparatively high doses and consequent unacceptable side effects because of interaction with other cellular targets. Second generation inhibitors are generally more dose-potent and include dexverapamil, valspodar and biricodar . Phase II evaluation of biricodar has been carried out but no clinical efficacy was observed [25]. One of the potential problems is that the drugs inhibit cytochrome P450 activity in altered pharmacokinetics of the anticancer drugs with which they were co-administered. Third generation inhibitors are designed to overcome these disadvantages and include the drugs tariquidar, zosuquidar and laniquidar. A recent Phase I trial of tariquidar revealed no untoward toxicity [26]. A carefully controlled Phase II study of laniquidar also showed no toxicity but failed to show clinical efficacy [27]. An alternative approach, that of designing drugs that are intrinsically insensitive to transport resistance mechanisms, generally requires that the drug is taken up efficiently into cells and is relatively resistant to conjugation with glutathione, glucuronide or sulphate. In this way, the rate of uptake rate greatly exceeds the potential rate of transportermediated efflux. Cellular uptake of drugs by diffusion is generally controlled by diffusion; lipophilic drugs can enter cells rapidly by diffusion, more hydrophilic drugs such as doxorubicin are taken up slowly and very hydrophilic drugs such as cytosine arabinoside require specific permeases [7]. Examples of lipophilic drugs that are thought to be substrates for drug transporters but whose uptake rates are rapid include the antileukaemia drug amsacrine [28]

Mol Biotechnol (2010) 46:308–316

and the experimental anticancer agent N-[2-(dimethylamino)ethyl]acridine-4-carboxamide [29]. Both drugs exhibit activity in cultured cells expressing MDR transporters [30].

311

itself can also provide a mechanism of apoptosis resistance [37].

Strategies to Overcome MDR Mechanisms Involving Resistance to Apoptosis MDR Mechanisms Involving Resistance to Apoptosis Suppression of pathways leading to apoptosis is thought to be an intrinsic feature of cancer cells [31]. Conversely, cytotoxic anticancer drugs commonly induce stress pathways such as p38 kinase [32, 33] or suppress signalling pathways such as those coordinated by phosphatidyl-3phosphate kinase (PI3K) and extracellular-regulated kinase-1 (ERK1) to reactivate pathways to apoptosis. One can therefore envisage pre-existing or acquired MDR mechanisms that lead to increased suppression of apoptosis pathways. There are many potential mechanisms whereby tumour stem cells acquire changes that increase resistance to apoptosis. Examples include inactivating mutations of the gene for p53 protein, activating mutations of the gene for PI3K, loss of expression of PTEN (a phosphatase controlling PI3K activity) and activating mutations of the genes for the RAS/RAF pathway. Modulation of these pathways affects the balance of activity of the bcl-2 family of proteins and their relatives [34], which in turn control transition to apoptosis by modulating the stability of the outer mitochondrial membrane. Thus, a change in balance of bcl-2 family members can lead to a so-called permeability transition [35] where the outer mitochondrial membrane is disrupted, and proteins that are normally stored in the space between the inner and outer mitochondrial membranes are released into the cytoplasm. These proteins include, depending on the cell, cytochrome C, endonuclease G and apoptosis-inducing factor and can act in various ways to induce cell death. Cytochrome C is well known for its ability to induce a cascade of caspase enzymes, which convert the cell to small fragments that can be recognised and ingested by nearby phagocytes. Loss of regulation of PI3K leads to increased activity of AKT (PKB), phosphorylation of bad, a member of the bcl-2 family, resultant protection of mitochondria from the permeability transition and increased resistance to cell death. Similarly, activating mutations in the RAS or RAF genes lead to activation of the ERK1 enzyme, inactivation by phosphorylation of bid, another member of the bcl-2 family and protection of mitochondria from the permeability transition. Phosphorylation of bcl-2, following activation of the NF-jB transcription factor and its downstream target twist-1 by cytokines and other cellular stress, can lead to apoptosis resistance [36]. Induced overexpression of bcl-2

There are now many examples of therapeutic drugs that act to counter resistance to apoptosis and some are currently in clinical trial [38–40]. Direct inhibition of bcl-2 family members, i.e. at the convergence of these pathways, has been demonstrated; for instance, obatoclax interferes with bcl-2 family-mediated resistance and restores sensitivity to several new anticancer drugs [41]. The transcription factor NF-jB specifies a number of pathways, including bcl-2 family members, that inhibit apoptosis [42] and the development of inhibitors of NF-jB signalling to overcome resistance is being investigated [43]. The development of inhibitors of PI3K, which induce apoptosis by reducing phosphorylation of AKT and thus of phosphorylation of one (bad) of the bcl2 family members is now a major aim of the pharmaceutical industry [44, 45]. Many other approaches are being investigated [46]. Two treatment strategies need to be distinguished here; the first uses a drug that promotes an apoptosis in combination with a cytotoxic agent to overcome an MDR mechanism that affects the cytotoxic agent. The second strategy uses a drug that promotes an apoptosis as a single agent to induce apoptosis without the need for combination therapy. Currently, most clinical trials are aimed at the second strategy. Combination clinical trials with of drugs that promote apoptosis with drugs that induce DNA damage or other cellular stress responses are still at an early stage.

Cytokinetic Factors in MDR Early experimental models for tumour growth were based particularly on transplantable murine leukaemias, where it was assumed that the majority of transplanted tumour cells were capable of forming tumours and that resistant cells could be identified. The development of stem cell theory for normal tissues, with its subsequent extension to tumour tissue, changed this concept by postulating that survival of normal or tumour tissue is controlled not by the whole population but only by cells that have the property of selfrenewal. The tumour stem cell model, which has had increasing general acceptance, implies that the resistance properties of the tumour stem cell population will dictate overall response to therapy. This in turn presents a challenge for the development of appropriate therapy [47].

312

Mol Biotechnol (2010) 46:308–316

TGF-β

IFN-γ

IFNGR1 TGFBR1/2

Smad2/3

p15CDKN2B

p21CDKN1A

Jak1/2

Stat-1

p21CDKN1A

Tumour stem cell quiescence

wnt, etc.

Tumour stem cell proliferation

Fig. 1 Hypothetical scheme for control of tumour stem cell proliferation by the microenvironment. Tumour stem cells can be maintained in a quiescent state by members of the TGF-b superfamily, thought to be an important mechanism in the stem cell niche of normal tissues. Tumour stem cells can also be maintained in a quiescent state by c-interferon, a suggested mechanism for tumour

cell dormancy. On the other hand, when tumour stem cells mover to a microenvironment where factors such as members of the wnt family are present, they become actively proliferating. The relative proportions of quiescent and proliferating cells may be important in dictating the resistance properties of the tumour as a whole

A salient feature of stem cells in normal tissue is that their microenvironment maintains their cytokinetic status. It is becoming apparent that the proliferation status of stem cells in normal tissue depends on their location [48]. As illustrated in Fig. 1, stromal cell production of factors in the stem cell niche, generally members of the TGFb/BMP superfamily, act on stem cells through corresponding surface receptors, smad family proteins and signalling pathways [49], increasing production of cyclin-dependent kinase inhibitors such as p15, p16, p21 and p27 and suppressing proliferation of stem cells. On the other hand, stem cells that have left the niche may enter a microenvironment where stromal cells produce factors of the wnt superfamily to induce proliferation. It is likely that tumour stem cells can also exist in these two microenvironments [50, 51]. Thus, slowly proliferating tumour stem cells are protected from a variety of anticancer drugs by MDR mechanisms related to their cytokinetic status. The clinical observation that a kidney transplant recipient developed melanoma with the genotype of the kidney donor, even though the donor had been free of melanoma for 16 years [52], points to mechanisms whereby tumour stem cells may be maintained for long periods in a nonproliferating state. There are many other examples of this phenomenon [53], raising the possibility that resistance to anticancer treatment may be related to loss of tumour stem cell dormancy following treatment. Surgery might in some cases also induce a loss of dormancy [54] and chemotherapy may have a similar effect, possibly through an

immunosuppressive effect. A preclinical model for tumour dormancy is provided by a study in which a group of mice were treated with a carcinogen. Under normal conditions, only a small proportion developed tumours, but when tumour-free mice were treated subsequently with antibodies to T cells or to c-interferon a high proportion developed tumours [55]. Examination of tumour-free mice before immunosuppressive treatment revealed the presence of microscopic groups of tumour cells associated with T cells. As shown in Fig. 1, one can hypothesise from these observations that IFNc acts in concert with other cytokines, probably through the induction of STAT1 [56] to constrain tumour cell proliferation [57]. Host cell-mediated mechanisms such as factors released by tumour fibroblasts also contribute to the maintenance of non-cycling tumour stem cells in a relatively drug-resistant state. Thus, tumour stem cell cytokinetics can contribute to MDR and our ability to combat this type of resistance will require a deeper understanding of host-tumour interactions..

The Role of Tumour Heterogeneity in Resistance Heterogeneity is an important feature which distinguishes tumour stem cell populations from stem cells in normal tissue. If this heterogeneity extends to expression of MDR mechanisms, the overcoming of one form of cell-mediated MDR may affect only a proportion of the total stem cell population, leaving the remainder to repopulate the niche.

Mol Biotechnol (2010) 46:308–316

There are a number of mechanisms for the generation of such heterogeneity; tumour cells contain a number of mutations or epigenetic changes that lead to defects both in their ability to differentiate and in their control of proliferation. Such changes affect not only cell cycle entry but also centrosome replication [58, 59], and centrosome defects lead during mitosis to chromosome instability, tetraploidy and aneuploidy. Other mechanisms, such as microsatellite instability and defective DNA repair can also contribute to a so-called mutator phenotype, whereby tumour cells may develop large numbers of genetic changes [60]. Thus, the observation that tumour stem cells are often aneuploid and heterogeneous may have a significant bearing on the development of resistance. It is of interest that when MCF-7 human breast cancer cells are cultured in the absence of oestrogen or in the presence of an antioestrogen, resistant cell lines develop with altered ploidy and altered proliferation rates, but these appear to be derived from existing minor populations of the original line [61].

MDR Involving Host Immune Responses The Role of Tumour Heterogeneity in Resistance section indicated that the body has a number of inbuilt mechanisms to prevent the proliferation of potential cancer stem cells. These mechanisms may induce senescence rather than quiescence; for instance, acquisition of genetic changes has the effect of inducing cellular stress responses involving the secretion of specific proteins, such as the interleukins IL-6 and IL-8, which induce senescence of tumour cells [62, 63]. The development of genetic changes in cancer stem cells can also lead to potentially cytotoxic T-lymphocyte responses [64]. Proteins such as HMGB1, when released from stressed or dying cells, can interact with TLR4 toll-like receptors macrophages and dendritic cells to induce potentially cytotoxic responses against the tumour [65]. Against these positive effects, tumours may also act to induce resistance to host immune mechanisms. As a consequence of cell turnover, tumour apoptotic is released continuously from the tumour and taken up by tumour macrophages and dendritic cells with a resultant immunosuppressive response [66]. Thus, tumours exist in a dynamic balance of progression or regression that is controlled largely by host-mediated effects. Tumours have a relatively high rate of cell turnover, as indicated by the fact that cell cycle times of individual tumour cells are measured in days while the volume doubling times of solid tumours are measured in months, and this turnover may be important for maintaining a state of relative

313

immunosuppression [67]. The application of cancer therapeutic agents will affect this balance and in some cases can increase immunosuppression, for instance by depleting immune cells or by increasing the production of potentially immunosuppressive apoptotic tumour cells. T-cell mediated release of interferon-c can potentially prevent the proliferation of tumour stem cells, and therapy that reduces interferon release could allow continuation and even stimulation of tumour growth. Thus, MDR may be a function of changes in host immune responses. Tumour tissue can be distinguished from normal tissue by an increased proportion of macrophages and dendritic cells, and there is increasing evidence that host cells play an important role in cancer chemotherapeutic response [68]. Following exposure to various sorts of stress, including those caused by anticancer drugs, tumour cells release certain proteins to the cytoplasm and from there to the external environment. Here, they interact with toll-like receptors, particularly with TLR4, on macrophages and dendritic cells, triggering responses that include cytokines and reactive oxygen species. The presence of a functioning immune system has been shown to be important for therapeutic outcome in a number of experimental models [69] and key proteins include HMGB1, a high mobility group protein associated with chromatin, and calreticulin [70], a protein associated with the endoplasmic reticulum. Administration of agents that optimise the production or effects of such proteins might therefore be used augment responses of tumour cells to chemotherapeutic agents. Existing cytotoxic agents appear to vary in their ability to induce this response, with one of the most effective being the anthracycline derivative doxorubicin. This drug induces DNA damage, which in turn activates poly ADP ribose polymerase (PARP). HMGB1 is ADP-ribosylated by PARP, allowing it to move to the cytoplasm, to be released from the cell and to interact with TLR4 receptors of host macrophages and host cells. Other stress pathways, probably involving calcium ions, appear to trigger co-translocation of calreticulin in association with the chaperone ERp57 (a disulphide isomerase protein) [70] to the plasma membrane where it can interact with TLR4 receptors. It has been found that tumour cells with low ERp57 expression of have a normal apoptotic response to doxorubicin in vitro but fail to respond to it in vivo [70], underlining the importance of this response in the overall antitumour effect of the drug. Tumour macrophages and dendritic cells must in themselves be competent in carrying out these responses to cancer chemotherapy. Release of chemokines within the tumour leads to the recruitment of immature macrophages to tumour tissue but until they are appropriately activated these may lack the ability to induce tumour cell responses.

314

Perspective The development of adequate methods to study MDR in human cancer remains a formidable challenge. Although initial studies in this area were dominated by the elucidation of outward drug transport mechanisms using immunohistochemistry and drug uptake/efflux measurements, subsequent studies have uncovered many different mechanisms that may confer simultaneous resistance to anticancer drugs with unrelated chemical structures. The material covered in this review describes several mechanisms of by which tumour stem cells exhibit MDR. A corollary of the tumour stem cell hypothesis is that the resistance properties of the tumour population as a whole may be different to the resistance properties of the tumour stem cells, meaning that the assessment of resistance by drug responses of primary cultures [71] or by gene expression arrays [72] will not necessarily predict the responses of the small proportion of the corresponding tumour stem cells. This may explain the limited success of such assays in predicting clinical outcome [73]. The multitude of mechanisms involved in the resistance of tumour stem cells to apoptosis, together with the known multiple resistance to normal tissue stem cells to cytotoxic insult, raises the question as to whether complete elimination of a tumour stem cell population is indeed possible. The results of many current clinical trials of antitumour agents that target the survival pathways have, with a few exceptions, provided extensions of tumour-free or overall survival measured in months rather than years, consistent with the hypothesis that these therapies are not having a major impact on the survival of tumour stem cells. As discussed earlier, both preclinical and clinical data are consistent with the concept that tumours can exist in a dormant state for long periods of time and studies in mice have suggested that the mechanism for the maintenance of this dormancy involves host-mediated suppression of cell proliferation through factors such as interferon-c. Therapies that lead to long-term arrest of tumour stem cells, either directly or by activation of host cell mechanisms therefore deserve serious consideration. As judged from studies of tumour cell lines, many cancers have defects such as loss p16 or p53 function, which compromise control of long-term cycle arrest. Strategies that reactivate p53 function [38] or that reverse epigenetic suppression of p16 [74] are examples of how cycle arrest might be approached. The contribution of the tumour microenvironment to treatment response can be illustrated by the following example. Murine methylcholanthrene-induced (MCA/129) tumours grow in BALB/c mice and respond well to radiotherapy (15 Gy). BALB/c mice lacking a single enzyme, acid sphingomyelinase, are phenotypically normal

Mol Biotechnol (2010) 46:308–316

and allow growth of MCA/129 tumours. However, such tumours are resistant to radiotherapy [75]. Radiation would be expected to cause the same amount of DNA damage to tumour cells regardless of the host, implying that tumour response is a function of the host rather than the tumour cells. The products of sphingomyelinase activity, ceramides are known to have multiple signalling roles in endothelial cells, macrophages and fibroblasts [76], supporting the concept that responses of host cells are essential, in combination with radiation-induced effects on the tumour population, for the observed tumour response. While further work is required to extend these observations to other tumour models, it is important to realise that the use of xenografts of human tumour cell lines growing in immunodeficient mice may be inappropriate for such studies. T-lymphocyte-induced interferon-c is important factor in macrophage maturation [77] and immunocompetent mice may for detailed studies. It is clear that further progress in understanding MDR in experimental cancer will have to employ a multidisciplinary approach with a combination of technologies to distinguish effects on tumour cells from those on host cells and to assess the relative importance of these two determinants in overall response. We are still at a very early stage with regard to the translation of these studies into clinical practice. The development of appropriate markers for host responses will be a critical factor in such research.

References 1. Bolhuis, H., Van Veen, H. W., Poolman, B., Driessen, A. J., & Konings, W. N. (1997). Mechanisms of multidrug transporters. FEMS Microbiology Reviews, 21, 55–84. 2. Kawase, M., & Motohashi, N. (2003). New multidrug resistance reversal agents. Current Drug Targets, 4, 31–43. 3. Szakacs, G., Paterson, J. K., Ludwig, J. A., Booth-Genthe, C., & Gottesman, M. M. (2006). Targeting multidrug resistance in cancer. Nature Reviews Drug Discovery, 5, 219–234. 4. Ozben, T. (2006). Mechanisms and strategies to overcome multiple drug resistance in cancer. FEBS Letters, 580, 2903–2909. 5. Dubikovskaya, E. A., Thorne, S. H., Pillow, T. H., Contag, C. H., & Wender, P. A. (2008). Overcoming multidrug resistance of small-molecule therapeutics through conjugation with releasable octaarginine transporters. Proceedings of the National Academy of Sciences of the United States of America, 105, 12128–12133. 6. Baguley, B. C. (2010). Multidrug resistance in cancer. Methods in Molecular Biology, 596, 1–14. 7. Baguley, B. C. (2002). Novel strategies for overcoming multidrug resistance in cancer. BioDrugs, 16, 97–103. 8. Baguley, B. C., & Marshall, E. S. (2008). The use of human tumour cell lines in the discovery of new cancer chemotherapeutic drugs. Expert Opinion on Drug Discovery, 3, 153–161. 9. Meads, M. B., Hazlehurst, L. A., & Dalton, W. S. (2008). The bone marrow microenvironment as a tumor sanctuary and contributor to drug resistance. Clinical Cancer Research, 14, 2519–2526.

Mol Biotechnol (2010) 46:308–316 10. Parmar, K., Mauch, P., Vergilio, J., Sackstein, R., & Down, J. D. (2007). Distribution of hematopoietic stem cells in the bone marrow according to regional hypoxia. Proceedings of the National Academy of Sciences of the United States of America, 104, 5431–5436. 11. Huls, M., Russel, F. G., & Masereeuw, R. (2009). The role of ABC transporters in tissue defense and organ regeneration. Journal of Pharmacology and Experimental Therapeutics, 328, 3–9. 12. Turco, M. C., Romano, M. F., Petrella, A., Bisogni, R., Tassone, P., & Venuta, S. (2004). NF-kappaB/Rel-mediated regulation of apoptosis in hematologic malignancies and normal hematopoietic progenitors. Leukemia, 18, 11–17. 13. Suzuki, E., Kapoor, V., Jassar, A. S., Kaiser, L. R., & Albelda, S. M. (2005). Gemcitabine selectively eliminates splenic Gr-1?/ CD11b? myeloid suppressor cells in tumor-bearing animals and enhances antitumor immune activity. Clinical Cancer Research, 11, 6713–6721. 14. Finlay, G. J., & Baguley, B. C. (2000). Effects of protein binding on the in vitro activity of antitumour acridine derivatives and related anticancer drugs. Cancer Chemotherapy and Pharmacology, 45, 417–422. 15. Hicks, K. O., Pruijn, F. B., Baguley, B. C., & Wilson, W. R. (2001). Extravascular transport of the DNA intercalator and topoisomerase poison N-[2-(dimethylamino)ethyl]acridine-4-carboxamide (DACA): Diffusion and metabolism in multicellular layers of tumor cells. Journal of Pharmacology and Experimental Therapeutics, 297, 1088–1098. 16. Nakagawa, T., Inoue, Y., Kodama, H., Yamazaki, H., Kawai, K., Suemizu, H., et al. (2008). Expression of copper-transporting Ptype adenosine triphosphatase (ATP7B) correlates with cisplatin resistance in human non-small cell lung cancer xenografts. Oncology Reports, 20, 265–270. 17. Chen, K. G., Valencia, J. C., Lai, B., Zhang, G., Paterson, J. K., Rouzaud, F., et al. (2006). Melanosomal sequestration of cytotoxic drugs contributes to the intractability of malignant melanomas. Proceedings of the National Academy of Sciences of the United States of America, 103, 9903–9907. 18. Ling, V. (1997). Multidrug resistance: Molecular mechanisms and clinical relevance. Cancer Chemotherapy and Pharmacology, 40, S3–S8. 19. Gillet, J. P., Efferth, T., & Remacle, J. (2007). Chemotherapyinduced resistance by ATP-binding cassette transporter genes. Biochimica et Biophysica Acta, 1775, 237–262. 20. Ejendal, K. F., & Hrycyna, C. A. (2002). Multidrug resistance and cancer: The role of the human ABC transporter ABCG2. Current Protein & Peptide Science, 3, 503–511. 21. Sarkadi, B., Ozvegy-Laczka, C., Nemet, K., & Varadi, A. (2004). ABCG2—a transporter for all seasons. FEBS Letters, 567, 116–120. 22. Loebinger, M. R., Giangreco, A., Groot, K. R., Prichard, L., Allen, K., Simpson, C., et al. (2008). Squamous cell cancers contain a side population of stem-like cells that are made chemosensitive by ABC transporter blockade. British Journal of Cancer, 98, 380–387. 23. Hadnagy, A., Gaboury, L., Beaulieu, R., & Balicki, D. (2006). SP analysis may be used to identify cancer stem cell populations. Experimental Cell Research, 312, 3701–3710. 24. Keshet, G. I., Goldstein, I., Itzhaki, O., Cesarkas, K., Shenhav, L., Yakirevitch, A., et al. (2008). MDR1 expression identifies human melanoma stem cells. Biochemical and Biophysical Research Communications, 368, 930–936. 25. Gandhi, L., Harding, M. W., Neubauer, M., Langer, C. J., Moore, M., Ross, H. J., et al. (2007). A phase II study of the safety and efficacy of the multidrug resistance inhibitor VX-710 combined with doxorubicin and vincristine in patients with recurrent small cell lung cancer. Cancer, 109, 924–932.

315 26. Abraham, J., Edgerly, M., Wilson, R., Chen, C., Rutt, A., Bakke, S., et al. (2009). A phase I study of the P-glycoprotein antagonist tariquidar in combination with vinorelbine. Clinical Cancer Research, 15, 3574–3582. 27. Ruff, P., Vorobiof, D. A., Jordaan, J. P., Demetriou, G. S., Moodley, S. D., Nosworthy, A. L., et al. (2009). A randomized, placebo-controlled, double-blind phase 2 study of docetaxel compared to docetaxel plus zosuquidar (LY335979) in women with metastatic or locally recurrent breast cancer who have received one prior chemotherapy regimen. Cancer Chemotherapy and Pharmacology, 64, 763–768. 28. Robbie, M. A., Baguley, B. C., Denny, W. A., Gavin, J. B., & Wilson, W. R. (1988). Mechanism of resistance of noncycling mammalian cells to 40 -(9- acridinylamino)methanesulfon-m-anisidide: Comparison of uptake, metabolism, and DNA breakage in log- and plateau-phase Chinese hamster fibroblast cell cultures. Cancer Research, 48, 310–319. 29. Haldane, A., Finlay, G. J., Hay, M. P., Denny, W. A., & Baguley, B. C. (1999). Cellular uptake of N-[2-(dimethylamino)ethyl] acridine-4-carboxamide (DACA). Anti-Cancer Drug Design, 14, 275–280. 30. Davey, R. A., Su, G. M., Hargrave, R. M., Harvie, R. M., Baguley, B. C., & Davey, M. W. (1997). The potential of N-[2(dimethylamino)ethyl]acridine-4-carboxamide to circumvent three multidrug-resistance phenotypes in vitro. Cancer Chemotherapy and Pharmacology, 39, 424–430. 31. Hanahan, D., & Weinberg, R. A. (2000). The hallmarks of cancer. Cell, 100, 57–70. 32. Sharma, S. V., Gajowniczek, P., Way, I. P., Lee, D. Y., Jiang, J., Yuza, Y., et al. (2006). A common signaling cascade may underlie ‘‘addiction’’ to the Src, BCR-ABL, and EGF receptor oncogenes. Cancer Cell, 10, 425–435. 33. Olson, J. M., & Hallahan, A. R. (2004). p38 MAP kinase: A convergence point in cancer therapy. Trends in Molecular Medicine, 10, 125–129. 34. Danial, N. N. (2007). BCL-2 family proteins: Critical checkpoints of apoptotic cell death. Clinical Cancer Research, 13, 7254– 7263. 35. Forte, M., & Bernardi, P. (2006). The permeability transition and BCL-2 family proteins in apoptosis: Co-conspirators or independent agents? Cell Death and Differentiation, 13, 1287–1290. 36. Pham, C. G., Bubici, C., Zazzeroni, F., Knabb, J. R., Papa, S., Kuntzen, C., et al. (2007). Upregulation of twist-1 by NF-kappaB blocks cytotoxicity induced by chemotherapeutic drugs. Molecular and Cellular Biology, 27, 3920–3935. 37. Osford, S. M., Dallman, C. L., Johnson, P. W., Ganesan, A., & Packham, G. (2004). Current strategies to target the anti-apoptotic Bcl-2 protein in cancer cells. Current Medicinal Chemistry, 11, 1031–1039. 38. Foster, B. A., Coffey, H. A., Morin, M. J., & Rastinejad, F. (1999). Pharmacological rescue of mutant p53 conformation and function. Science, 286, 2507–2510. 39. Sebolt-Leopold, J. S. (2004). MEK inhibitors: A therapeutic approach to targeting the Ras-MAP kinase pathway in tumors. Current Pharmaceutical Design, 10, 1907–1914. 40. Serra, V., Markman, B., Scaltriti, M., Eichhorn, P. J., Valero, V., Guzman, M., et al. (2008). NVP-BEZ235, a dual PI3K/mTOR inhibitor, prevents PI3K signaling and inhibits the growth of cancer cells with activating PI3K mutations. Cancer Research, 68, 8022–8030. 41. Nguyen, M., Marcellus, R. C., Roulston, A., Watson, M., Serfass, L., Murthy Madiraju, S. R., et al. (2007). Small molecule obatoclax (GX15-070) antagonizes MCL-1 and overcomes MCL-1mediated resistance to apoptosis. Proceedings of the National Academy of Sciences of the United States of America, 104, 19512–19517.

316 42. Nakanishi, C., & Toi, M. (2005). Nuclear factor-kappaB inhibitors as sensitizers to anticancer drugs. Nature Reviews Cancer, 5, 297–309. 43. Wang, W., McLeod, H. L., & Cassidy, J. (2003). Disulfirammediated inhibition of NF-kappaB activity enhances cytotoxicity of 5-fluorouracil in human colorectal cancer cell lines. International Journal of Cancer, 104, 504–511. 44. Bunney, T. D., & Katan, M. (2010). Phosphoinositide signalling in cancer: Beyond PI3K and PTEN. Nature Reviews Cancer, 10, 342–352. 45. Cleary, J. M., & Shapiro, G. I. (2010). Development of phosphoinositide-3 kinase pathway inhibitors for advanced cancer. Current Oncology Reports, 12, 87–94. 46. Hersey, P., Zhuang, L., & Zhang, X. D. (2006). Current strategies in overcoming resistance of cancer cells to apoptosis melanoma as a model. International Review of Cytology, 251, 131–158. 47. Saini, V., & Shoemaker, R. H. (2010). Potential for therapeutic targeting of tumor stem cells. Cancer Science, 101, 16–21. 48. Li, L., & Clevers, H. (2010). Coexistence of quiescent and active adult stem cells in mammals. Science, 327, 542–545. 49. Massague, J. (2008). TGFbeta in cancer. Cell, 134, 215–230. 50. Vermeulen, L., De Sousa, E. M., van der Heijden, M., Cameron, K., de Jong, J. H., Borovski, T., et al. (2010). Wnt activity defines colon cancer stem cells and is regulated by the microenvironment. Nature Cell Biology, 12, 468–476. 51. Roesch, A., Fukunaga-Kalabis, M., Schmidt, E. C., Zabierowski, S. E., Brafford, P. A., Vultur, A., et al. (2010). A temporarily distinct subpopulation of slow-cycling melanoma cells is required for continuous tumor growth. Cell, 141, 583–594. 52. MacKie, R. M., Reid, R., & Junor, B. (2003). Fatal melanoma transferred in a donated kidney 16 years after melanoma surgery. New England Journal of Medicine, 348, 567–568. 53. Aguirre-Ghiso, J. A. (2007). Models, mechanisms and clinical evidence for cancer dormancy. Nature Reviews Cancer, 7, 834–846. 54. Demicheli, R., Retsky, M. W., Hrushesky, W. J., & Baum, M. (2007). Tumor dormancy and surgery-driven interruption of dormancy in breast cancer: Learning from failures. Nature Clinical Practice Oncology, 4, 699–710. 55. Koebel, C. M., Vermi, W., Swann, J. B., Zerafa, N., Rodig, S. J., Old, L. J., et al. (2007). Adaptive immunity maintains occult cancer in an equilibrium state. Nature, 450, 903–907. 56. Kortylewski, M., Komyod, W., Kauffmann, M. E., Bosserhoff, A., Heinrich, P. C., & Behrmann, I. (2004). Interferon-gammamediated growth regulation of melanoma cells: Involvement of STAT1-dependent and STAT1-independent signals. Journal of Investigative Dermatology, 122, 414–422. 57. Muller-Hermelink, N., Braumuller, H., Pichler, B., Wieder, T., Mailhammer, R., Schaak, K., et al. (2008). TNFR1 signaling and IFN-gamma signaling determine whether T cells induce tumor dormancy or promote multistage carcinogenesis. Cancer Cell, 13, 507–518. 58. Kramer, A., Lukas, J., & Bartek, J. (2004). Checking out the centrosome. Cell Cycle, 3, 1390–1393. 59. McDermott, K. M., Zhang, J., Holst, C. R., Kozakiewicz, B. K., Singla, V., & Tlsty, T. D. (2006). p16(INK4a) prevents centrosome dysfunction and genomic instability in primary cells. PLoS Biology, 4, e51. 60. Loeb, L. A., Bielas, J. H., & Beckman, R. A. (2008). Cancers exhibit a mutator phenotype: Clinical implications. Cancer Research, 68, 3551–3557.

Mol Biotechnol (2010) 46:308–316 61. Leung, E., Kannan, N., Krissansen, G. W., Findlay, M. P., & Baguley, B. C. (2010). MCF-7 breast cancer cells selected for tamoxifen resistance acquire new phenotypes differing in DNA content, phospho-HER2 and PAX2 expression, and rapamycin sensitivity. Cancer Biology & Therapy, 9(9), 717–724. 62. Kuilman, T., Michaloglou, C., Vredeveld, L. C., Douma, S., van Doorn, R., Desmet, C. J., et al. (2008). Oncogene-induced senescence relayed by an interleukin-dependent inflammatory network. Cell, 133, 1019–1031. 63. Acosta, J. C., O’Loghlen, A., Banito, A., Raguz, S., & Gil, J. (2008). Control of senescence by CXCR2 and its ligands. Cell Cycle, 7, 2956–2959. 64. Smyth, M. J., Godfrey, D. I., & Trapani, J. A. (2001). A fresh look at tumor immunosurveillance and immunotherapy. Nature Immunology, 2, 293–299. 65. Zitvogel, L., Apetoh, L., Ghiringhelli, F., & Kroemer, G. (2008). Immunological aspects of cancer chemotherapy. Nature Reviews Immunology, 8, 59–73. 66. Fonseca, C., & Dranoff, G. (2008). Capitalizing on the immunogenicity of dying tumor cells. Clinical Cancer Research, 14, 1603–1608. 67. Baguley, B. C. (2006). Tumor stem cell niches: A new functional framework for the action of anticancer drugs. Recent Patents on Anti-Cancer Drug Discovery, 1, 121–127. 68. Chen, R., Alvero, A. B., Silasi, D. A., Steffensen, K. D., & Mor, G. (2008). Cancers take their Toll—the function and regulation of Toll-like receptors in cancer cells. Oncogene, 27, 225–233. 69. Hicks, A. M., Riedlinger, G., Willingham, M. C., AlexanderMiller, M. A., Kap-Herr, C., Pettenati, M. J., et al. (2006). Transferable anticancer innate immunity in spontaneous regression/complete resistance mice. Proceedings of the National Academy of Sciences of the United States of America, 103, 7753–7758. 70. Panaretakis, T., Joza, N., Modjtahedi, N., Tesniere, A., Vitale, I., Durchschlag, M., et al. (2008). The co-translocation of ERp57 and calreticulin determines the immunogenicity of cell death. Cell Death and Differentiation, 15, 1499–1509. 71. Baguley, B. C., & Marshall, E. S. (2004). In vitro modelling of human tumour behaviour in drug discovery programmes. European Journal of Cancer, 40, 794–801. 72. Sotiriou, C., & Piccart, M. J. (2007). Taking gene-expression profiling to the clinic: When will molecular signatures become relevant to patient care? Nature Reviews Cancer, 7, 545–553. 73. Samson, D. J., Seidenfeld, J., Ziegler, K., & Aronson, N. (2004). Chemotherapy sensitivity and resistance assays: A systematic review. Journal of Clinical Oncology, 22, 3618–3630. 74. Claus, R., & Lubbert, M. (2003). Epigenetic targets in hematopoietic malignancies. Oncogene, 22, 6489–6496. 75. Garcia-Barros, M., Paris, F., Cordon-Cardo, C., Lyden, D., Rafii, S., Haimovitz-Friedman, A., et al. (2003). Tumor response to radiotherapy regulated by endothelial cell apoptosis. Science, 300, 1155–1159. 76. Vit, J. P., & Rosselli, F. (2003). Role of the ceramide-signaling pathways in ionizing radiation-induced apoptosis. Oncogene, 22, 8645–8652. 77. Duff, M. D., Mestre, J., Maddali, S., Yan, Z. P., Stapleton, P., & Daly, J. M. (2007). Analysis of gene expression in the tumorassociated macrophage. Journal of Surgical Research, 142, 119–128.