Bntuh Mtdiail Bulltlm (1991) Vol. 47, No. 1, pp. 178-196 © The Brinih Council 1991
Drug resistance Molecular Oncology Laboratory, Imperial Cancer Research Fund Laboratories, Institute of Molecular Medicine, University of Oxford, John Radcliffe Hospital, Oxford, UK
Chemotherapy cures a minority of adult tumours (e.g. Hodgkin's and non-Hodgkin's lymphoma, acute leukaemia, teratoma) and the majority of childhood tumours. Prolongation of survival by chemotherapy has been shown for small cell lung cancer, ovarian cancer and breast carcinoma (when used as an adjuvant). However, in the majority of solid tumours there is a less than 20% response to chemotherapy and even curable tumours may relapse and become resistant. Resistance may be de novo, acquired as a stable change within the cell, or be rapidly inducible within the cell after drug administration. Several mechanisms have been described including multidrug resistance, glutathione transferases and DNA repair. Understanding these mechanisms may help to improve the therapeutic ratio and develop new approaches.
Mechanisms of drug resistance relating to pharmacokinetics can be considered to operate from the point of administration to achievement of cytocidal concentrations at the target site. Important factors may include absorption, distribution and metabolism of drug. Numerous effects mediate bioavailability of drug from poor absorption to poor rumour blood supply. These factors will not be discussed in diis review which will concentrate on some recent advances in understanding drug resistance at the molecular level. In the past few years correlation between in vitro cellular responses and the development of resistance in vivo has been most clearly shown in multidrug resistance.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
D Hochhauser A L Harris
DRUG RESISTANCE
179
MULTIDRUG RESISTANCE
Homologies of PGP Clear homologies exist between PGP and several bacterial transport proteins in particular with conservation of the ATP binding domains.3 These proteins function to transport individual peptide and carbohydrate species into bacterial cells. In bacteria the hydrophilic and hydrophobic domains are associated as independent integral membrane proteins rather than being linked in a single polypeptide chain. They are energized by hydrolysis of ATP. In eukaryotes the only denned normal role for an MDR homologue has been found in Saccharomyces cerevisiae. The product of the STE6 gene is actively involved in the transport of a factor mating (MAT) pheromone.4 It is not essential for cell viability and does not confer a drug resistant phenotype. Recent identification of the cystic fibrosis gene reveals a similarity of the cystic fibrosis transmembrane regulator (CFTR) to these transport systems with ATP binding domains and potential membrane spanning hydrophobic sequences. Cloning of the MDR gene Cloning of the MDR gene in higher eukaryotes has been achieved by a variety of approaches including in gel renaturation to detect amplified sequences, differential hybridization, or isolation of cDNA clones using antibodies to PGP. Definitive evidence for the involvement of these genes in the multidrug resistance phenotype has been obtained primarily by transfection studies as well as analysis of resistant cell lines, clinical material and the use of MDR antagonists.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
Multidrug resistance (MDR) is the phenomenon whereby exposure to one drug induces cross resistance to a variety of agents of different chemical classes to which the cell has never been exposed. These include primarily adriamycin, vinca alkaloids and mitomycin C but not platinum, bleomycin and alkylating agents. The hallmark of MDR is the expression of P glycoprotein (PGP), a 170 kD protein with six hydrophobic domains and a tandemly duplicated ATP binding domain.1'2 PGP acts as a transmembrane exporter of drugs but the normal substrates have yet to be identified.
180
THE CANCER CELL
Chromosomal location and genomic organisation of the MDR gene
Molecular genetic studies Major evidence for PGP involvement in drug resistance was through gene transfection studies. In man only the MDR1 gene has been demonstrated to mediate MDR through transfection experiments and MDR-3 has no effect. Expression of a full length cDNA conferred resistance to colchicine, doxorubicin and vinblastine in NIH3T3 cells.6 However, in rodents both PGP1 and PGP2 mediate resistance. Regulation of MDR gene P-glycoprotein is regulated as a stress protein.7 For example, the renal adenocarcinoma cell line HDB46 shows an 8-fold increase in MDR1 RNA levels in response to heat shock, ethanol, sodium arsenite and cadmium consistent with the role of MDR1 as a stress-inducible gene response to environmental insults. Induced transcripts were generated predominantly from the downstream promoter. Sequence analysis of the 5' region of MDR1 reveals several heat shock regulatory elements. The induction of MDR1 Table 1 Expression of PGP genes in human, mouse and hamster P-glycoprotein Gene Class Species
I
II
III
Human Mouse Hamster
mdr 1 mdr 3 PgP 1
mdr 1 Pgp2
mdr 3 mdr 2 Pgp3
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
In man the MDR1 gene is located on chromosome 7q 21-31 and consists of an open reading frame of 1280 amino acids. There are 2 human and 3 rodent MDR genes (Table 1). Control of expression of both human MDR genes occurs at the level of gene copy number, transcription, translation, and post-translationally. Isolation and sequencing of a one kilobase portion of genomic DNA encoding the promoter region of the human MDR-1 gene indicated a consensus CAAT box and two GC box like sequences but no TATA sequences.3 Although two potential promoter start sites have been found, only the downstream promoter was activated in the KB lines tested.
DRUG RESISTANCE
181
Protein modification Post-translational modification of PGP is also significant. In man a 140 kD precursor protein has been identified which is gradually converted to a 170 kD form over 2 to 4 hours. The amino acid sequence of PGP indicates 10 potential N-glycosylation sites, but only 3 appear on the external surface of the plasma membrane. However, glycosylation deficient mutants of MDR cells show no change in the resistance profiles compared with parental cells and glycosylation may therefore be of questionable significance. This was confirmed with studies in which treatment with tunicamycin (a glycosylation inhibitor) caused no change in resistance patterns.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
RNA occurs concomitantly with development of a 2- to 6-fold increase in drug resistance to vinblastine following heat shock. However, induction of glucose regulated proteins by glucose deprivation, anoxia, calcium ionophore A23187 and 2-deoxyglucose led to no significant change in the levels of PGP despite an increase in doxorubicin resistance. Furthermore, in several other cell lines (including human cervical carcinomas, liver carcinomas and fetal hepatocytes) no heat shock induction could be demonstrated, indicating that these effects may be tissue- and cell-type specific. Regulation of MDR1 by gene amplification was suggested by the presence of double minutes and homogeneously staining regions.8 Several genes may coamplify with P-glycoprotein. A major coamplified product is sorcin, an acidic 22 kD protein which has homologies to calpain and binds calcium. It may be that specific differential amplification of sorcin and other gene products is partially responsible for the differential drug sensitivities found in different MDR expressing resistant cell lines but no correlation has been shown with resistance pattern. Evidence of differential overexpression of MDR isotypes has come from studies on the murine macrophage line J7742 cells selected for resistance to colchicine, vinblastine or taxol.9 There was production of two distinctly sized precursors of 125 and 120 kD, with a switch from the 120 to the 125 kD precursor occurring during selection. In rodents a switch from one form to another may correlate with increased resistance to vinblastine, taxol and doxorubicin, with no detectable change in the total amount of P-glycoprotein produced. However, similar phenomena have not been detected in man as only one MDR gene confers resistance.
182
THE CANCER CELL
Mutations of MDR gene Other factors may alter patterns of cross resistance in multidrug resistance cells due to mutations within the MDR1 gene. Studies in KB cells revealed that mutations of codon 185 changing from glydne to valine, resulting in a changed preferential resistance pattern from vinblastine to colchicine.11 The nucleotide sequences were analysed using the polymerase chain reaction and it was found that the colchicine specific sequence exists de novo within KB-DNA prior to selection. Transfectants of wild type and colchicine resistant cDNAs conferred the appropriate preferential resistance. It is thus clear that amino acid 185 plays a key role in PGP drug interaction. This amino acid is located within the first hydrophobic region on the cytoplasmic side of the membrane and may be part of a drug binding site. RNA stabilization Recent evidence on the MDR-1 gene in regenerating rat liver reveals another possible level of control with mRNA degradation being controlled by sequences in the noncoding region.12 Summary of mechanisms of regulation Expression of the MDR gene is regulated by amplification, transcription and translation. The relative contributions of each may vary throughout the selection process. In a multidrug resistant ovarian carcinoma cell line overexpression of MDR1 mRNA was found without evidence of amplification at low levels of drug resistance. At intermediate drug concentrations, amplification became increasingly significant while at high drug level PGP increased without further changes in mRNA or gene copy number, consistent with translational modification or mutation. Apart from indicating that MDR may be mediated at different levels of protein in the same tumour under different selection pressures, this study
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
The MDR1 expressing mouse line J774-2 has been shown to be a substrate for cyclic AMP dependent protein kinase-A with phosphorylation of serine and threonine residues having been demonstrated in vitro. PGP seems to be phosphorylated in its basal state primarily by protein kinase C and this may play a role in the drug transport process.10
DRUG RESISTANCE
183
Tissue distribution of MDR expression Studies of tissue distribution have involved the use of immunohistochemistry with mouse and human monoclonal antibodies as well as ribonuclease protection assays and in situ mRNA hybridization. Specific monoclonals can distinguish between the various PGP isoforms.13 The highest levels of PGP were found in kidney, adrenal cortex (zona fasciculata and reticulata), stomach, duodenum, colon and placenta.14 Expression occurs primarily in specialized epithelial cells with secretory or excretory functions as well as in placental trophoblasts. Strong expression was also found on endothelial cells of capillary blood vessels at blood tissue barrier sites, primarily capillaries of the central nervous system and testes as well as the capillary dermis.' s The potential significance of these observations is that the brain and testis constitute sanctuary sites in which relapse following systemic chemotherapy in conditions such as acute lymphoblastic leukaemia occurs presumably because of failure of drug penetration. Location of PGP in these cells is primarily on the nominal surface consistent with the role of PGP in secretion of as yet unknown substrate. Clinical significance—expression in tumours Studies of the significance of MDR in human tumours have recently become available. The most extensive study reported on levels of MDR1 mRNA in over 400 human cancers.16 Quantification of mRNA was measured using slot blot analysis comparing tumour samples to known drug sensitive (KB-3-1) and resistant (KB-8-5) cells. Expression was denned as high if MDR1 mRNA was detectable in over 50% of cancers in each group (see Table 2). RNAse protection analysis revealed use of the downstream promoting site of each of these cancers. In general, there is an inverse correlation of MDR expression with relative chemosensitivity of the tumour type. Thus, breast and ovarian cancer have lower levels than colon and renal cell carcinoma. However, many tumours known to be resistant usually had low levels, e.g. sarcomas and non small cell lung cancers. This
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
also indicates that measurement of AiDRl mRNA alone may give a misleading impression of the amount of functional PGP present. Other studies in KB-Hela cells also showed overexpression of MDR mRNA preceding gene amplification during selection.
184
THE CANCER CELL
Table 2 Expression of MDR mRNA levels in tumours Low MDR
Colon carcinoma Renal cell carcinoma Hepatoma Adrenocortical cancer Phaeochromocytoma Pancreas islet-cell cancer CML (blast crisis) Carcinoid
Breast cancer Non small cell lung cancer Bladder cancer CML (Chronic phase) Small cell lung cancer Lymphomas Ovary Acute leukaemias
may reflect assay sensitivity and difficulty of knowing what level is important clinically. Clinical significance has also been assessed in a retrospective immunohistochemical study of rhabdomyosarcoma and undifferentiated sarcomas which showed that all 9 patients with MDR relapsed.17 Of 21 PGP-negative patients, 20 showed durable complete response. A study of various lung cancers and nontumourous lung tissue showed no significant difference in MDR and minimal change following therapy, though correlation has been found in other studies. The expression of MDR clearly varies with different cell and tumour types.18 There have been several other studies of MDR in tumours following treatment. It is often problematic to delineate the exact role of MDR in causing failure of therapy because of induction of a variety of different resistance pathways. The earliest report of PGP elevation in a human malignancy was in 5 cases of relapsed ovarian cancer where cells showed raised MDR compared with normal ovarian tissue which did not express MDR. 19 In one of these patients, increased MDR expression was shown following failure of therapy. In a study of 9 newly diagnosed patients with acute lymphoblastic leukaemia, 8 had low levels of MDR mRNA at time of diagnosis. One patient with elevated MDR mRNA failed to achieve remission. However, a significant number of relapsing patients (5 out of 19) in the same study also had low MDR mRNA levels.20 In another small study of 15 patients with acute non lymphoblastic leukaemia 3 out of 10 with significant MDR1 levels relapsed after a short interval, in contrast to 4 out of 5 with minimal or absent MDR expression who achieved remission. Immunohistochemistry of a series of MDR human myeloma cell lines showed correlation of increased P-glycoprotein with resistance and higher levels in relapsed patients, though no measurements were taken prior to treatment.21 The majority of studies so
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
High MDR
DRUG RESISTANCE
185
Reversal of MDR A variety of pharmacological agents reverse MDR and this has been extensively reviewed.22 These range from the calcium channel blockers, particularly verapamil and nifedipine to tamoxifen, phenothiazines and cyclosporin, among many others. The concentrations required to reverse MDR in vitro would produce toxicity in vivo and this has been a major limitation to effective clinical use of, for example, verapamil. Some of these agents probably act by binding to PGP and thereby cause intracellular accumulation of drug. Odiers may act by binding calmodulin or do not affect efflux but cause redistribution of subcellular localization. Normal basal to apical flux of vinblastine MDR in epithelial cells is reversed by verapamil with enhanced phosphorylation of PGP, but verapamil has also been noted to potentiate bleomycin, cis platdn and 5-fluorouracil toxicity which are not associated with the MDR phenotype. Preliminary clinical studies of verapamil have so far proved to be disappointing in most cases with effective plasma concentrations causing severe hypotension and heart block. However, recent studies on the use of verapamil in multiple myeloma have shown a significant improvement in remission rate and are encouraging.23 Possibly using different MDR antagonists together at lower levels such as tamoxifen with verapamil, may overcome MDR without undue toxicity seen with higher levels of each drug. Alternatively, chemotherapeutic drug analogues not binding to PGP, e.g. the adriamycin derivative aclacinomycin, could be developed. Although expression of P-glycoprotein thus far appears to be the most significant cause of the multidrug resistant phenotype, increasing attention has recently been paid to the so called atypical multidrug resistance (atMDR). This has been applied to circumstances where the multidrug resistance phenotype has occurred without overexpression of the MDR gene. The commonest pattern appears to be of drugs which act on cellular topoisomerase II as a target.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
far have constituted too small a number of patients to allow significance to be proven of the clinical relevance of MDR expression. Ongoing longitudinal studies will help to clarify the situation as to whether subgroups of tumour patients expressing MDR predict response to chemotherapy and will thus be appropriate for use of MDR modulators.
186
THE CANCER CELL
TOPOISOMERASE II
Topoisomerase II as a drug target Several classes of drugs are termed topoisomerase inhibitors though they do not inhibit the initial stages of topoisomerase action. These drugs include intercalating agents—such as acridines (mAMSA), anthracyclines (adriamycin, daunomycin), anthracene-diones (mitoxantrone) and ellipticines. There is also an important group of non-intercalating agents which appear to bind directly to topo II, most notably teniposide (VP26) and etoposide (VP16). These drugs bind to topo II forming a cleavable complex; so-called because denaturation of the trapped protein with SDS or alkali reveals the DNA strand breaks with protein covalently linked to the 5-phosphoryl end of each broken strand. Since each complex requires two proteins, an increase in the amount of topoisomerase will correlate with increase strand breakage and potential toxicity. It has been shown, for example, that CHO cell lines sensitive to etoposide overexpressed topoisomerase II. 2 6 Although the exact mode of cell death is unclear, presumably the cleavable complex prevents chromosome segregation as well as DNA synthesis. Almost all topoisomerase inhibiting drugs increase the number of topo II associated DNA strand breaks although this does not always correlate with cytotoxicity because the drugs may have other modes of action.
Regulation of topo II Topo II is expressed primarily during the proliferative phases of growth and is down regulated at plateau phase. The amount of
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
Topoisomerase II (topo II), the eukaryotic homologue of bacterial DNA gyrase, is a 170 kD homodimeric protein which plays a role in DNA replication, chromosome scaffold formation, chromosomal segregation, and possibly recombination and gene transcription. 24 ' 25 In Saccharomyces cerevisiae, topo II null mutations are lethal because of the inability to separate chromosomes at mitosis. The enzyme acts by producing DNA single and double strand breaks and it attaches covalently to the 5' ends of the break. Double strand breaks are staggered by 4 base pairs on opposite stands and the protein attaches as a dimer. Subsequently strand passage occurs with changes in both supercoiling and DNA relaxation.
DRUG RESISTANCE
187
Topoisomerase II and drug resistance There has been demonstration of so-called atypical multidrug resistant phenotype in numerous cell lines. Characteristically there is cross resistance to the full range of anti-topo II drugs though not to the vinca alkaloids. These mechanisms may include: Point mutations and abnormally functioning topoisomerases
An HL60 human leukaemia cell line showed 100-fold resistance to mAMSA in contrast to 2- to 3-fold increase in etoposide resistance. 28 Cleavage of DNA produced by mAMSA or etoposide had an absolute requirement for ATP in contrast to die situation in
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
topo II increases at the onset of DNA replication continues to increase through S and G2 phases, peaks in die late G2 to M phase and then drops after mitosis. At confluence topo II is switched off in untransformed but not in transformed cells. The regulation of topo II during the cell cycle may be relevant as to which phase is most sensitive to topo II inhibitors. Following serum stimulation of BALB-C3T3 cells, actively proliferating cells showed much greater sensitivity to etoposide than quiescent cells. The increase in drug sensitivity began during S phase and reached a peak just before mitosis with a maximal 2.5-fold increase in drug sensitivity. However, although the maximal number of topo II associated strand breaks occurred during G 2 , cytotoxicity was maximal during S phase, suggesting interactions with other mechanisms leads to cell death. Modulation of topo II activity during the cell cycle involves phosphorylation. This has been demonstrated for casein kinase and protein kinase C, and these changes may be crucial in switching on the activity of topoisomerase. Phosphorylation has been shown to lead to a 3-fold increase in enzyme activity and dephosphorylation inactivates die enzyme. Studies of 32 P incorporation into DNA topo II in vivo in chicken lymphoblastoid cells showed phosphorylation highest at G2 and M, suggesting phosphorylation may be involved in enzyme activation for sister chromatid disjunction.27 Glucose deprivation, hypoxia and the stress response to heavy metals have been shown to downregulate the amount of topo II while epidermal growth factor (EGF) is an upregulator. The latter effect may be simply due to causing increased cell proliferation.
188
THE CANCER CELL
Post translational
In a study of teniposide resistant human CEM lines the amount of topo II bound to the nuclear matrix decreased due to a mutant enzyme which differs in salt stability thereby affecting attachment to the nuclear matrix.30 It is probable that the functional topo II is that which is associated with the matrix, and with newly replicated DNA. Methylation Several studies have demonstrated hypermethylated sites. This could be of significance in regulating the transcription of nonmutant alleles. Glucose deprivation and hypoxia
These factors may be relevant within the tumour mass and contribute towards decreased topoisomerase expression. Upregulation of topoisomerase I
This has been shown in several cases. Topoisomerase I, an enzyme which produces single strand breaks in DNA and is involved in transcription, may increase as a compensatory effect widi topo II downregulation (as occurs in S. cerevisiae). Switch to a novel form of topo II ($) Purification of a topoisomerase II from mAMSA resistant P388 leukaemia cells indicated another form of topo II of 180 kD differing in several respects from the previously known 170 kD form.31 Partial cloning of the gene shows that it may lack the leucine zipper motif which is probably important in dimerisation of the 170 kD form and that it differs in expression throughout the cell cycle (topo II P levels peak during G! and fall during the rest of the cell cycle), and in drug sensitivity. The 170 kD topo II (topo I la) is expressed in rapidly proliferat-
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
wild-type cells. A new restriction enzyme polymorphism was found in the resistant forms. This correlated with a mutation in the ATP binding site which has been found to induce mAMSA resistance in the bacteriophage T4 DNA topoisomerase. The extent to which downregulation of the total amount of topo II is a significant resistance mechanism remains unclear, although reduced levels of topo II shown in some studies could be due to decreased half-life of the mutant protein. Several studies have demonstrated abnormal functional forms of topo II in resistant cells with different drug cleavage patterns.29
DRUG RESISTANCE
189
Tissue distribution of topoisomerase II Studies of topo II activity in normal tissues showed the highest levels in spleen and thymus.33 Expression within tumours was highest in breast cancer and leiomyosarcoma. However, levels within these tumours were in the same range as normal tissues; possibly normal tissue topo II is relatively teniposide resistant. These results were unexpected since one would expect highest topo II levels in proliferating cells. Further studies on tissue levels are needed using antibodies and activity assays. Clinical implications Further studies are required to elucidate the clinical significance of altered topo II as well as possible mediods of circumvention. It may be possible to reduce topo II resistance by molecular design. Mechanisms of resistance to the topo II inhibitor mAMSA include altered transport and MDR. By altering substituents on the anilino acridine nucleus resistance to both mechanisms was overcome.34 Upregulation of topo I could be significant as inhibitors, notably camptothecin, could be used and cells exhibiting this mechanism of resistance may be hypersensitive to topo I inhibitors. Another route could be through the development of competitive topoisomerase inhibitors rather than the current stoichiometric agents i.e. competition with topo II for binding to DNA would be more effective in low topo II expressers. There appears to be a link between MDR and topo II regulation with possible inverse correlation of these.
GLUTATHIONE TRANSFERASES The involvement of glutathione and glutathione transferases (GST) in the multi-drug resistant phenotype has been difficult to
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
ing cells and its ratio increases compared to the 180 kD form (topo IIP) in ras transfected NIH3T3 cells.32 As the 170 kD protein is preferentially sensitive to teniposide and merbarone this could account for drug effectiveness against tumours which possess a higher percentage of topo I la—i.e. transformed cells as opposed to normal cells. Similarly, switch of expression from topo Ila to topo Up could lead to an 8-fold degree of resistance with equivalent protein expression.
190
THE CANCER CELL
GST expression in drug resistance Several anticancer drugs may be inactivated by GST catalysis— e.g. melphalan, chlorambucil and cyclophosphamide. There have been numerous reports correlating increases in GST isozymes, particularly the TC family, with the onset of drug resistance. There may be increased amounts of enzymes involved in glutathione synthesis and also glutathione peroxidase. GSTTI has also been thought to be significant as it constitutes the predominant isozyme found in human cancers with 2- to 4-fold increases in RNA levels in tumours of the colon, bladder, ovary and stomach relative to normal tissues. Also it has been found to be overexpressed in several multi-drug resistant lines, particularly in adriamycin resistant lines. Studies in ovarian cancer patients show expression of a and n isozymes following treatment. Transfection studies Transfection of GSTTI C D N A into ras transformed NIH3T3 cells led to elevated GSTTI enzyme levels but although there was a moderate 1.8- to 3-fold increase in resistance to adriamycin and ethacrynic acid (a substrate of GSTTT) there was no change in resistance to cis platinum, chlorambucil and melphalan.37 In another study a doxorubicin resistant line of human breast cancer MCF7 with both GSTir and MDR expression was analysed. The MDR and GSTTI CDNA'S from these cells were cloned and transfected but although MDR conferred a resistance phenotype GSTTI gave only moderate resistance to ethacrynic acid. The transfection of GSTTC, together with MDR showed no significant additive effect.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
clarify. Glutathione transferases are enzymes involved in detoxification, distributed in all organs but primarily in the liver and kidney. Their mode of action is by conjugation of glutathione (a tripeptide present in all organs in mM amounts) via the sulphur atom of its cystine residue to various electrophiles. GST's may also act as intracellular binding proteins (e.g. for bilirubin and steroids) and may constitute up to 10% of the total cellular protein in the liver. In man GST exists in cytosolic and microsomal forms. There are 3 major classes of cytosolic GST: n, a, \i. Several comprehensive recent reviews deal with classification of the GST's. 3 5 3 6
DRUG RESISTANCE
191
Glutathione depletion
DNA REPAIR The importance of basal DNA repair mechanisms in carcinogenesis and drug resistance is suggested by human repair deficiency diseases in which cancer susceptibility is high.38 For example, cells from patients with ataxia -telangiectasia are hypersensitive to topoisomerase II inhibitors and x-ray irradiation while xeroderma pigmentosa cells are hypersensitive to UV irradiation and cis platinum. There are multiple mechanisms of DNA repair {see Table 3) and probably at least 100-200 genes involved in DNA repair in mammalian cells. In view of the interaction of anticancer drugs with DNA by intercalation, alkylation and interference with DNA synthesis, it is likely that DNA repair is of key significance in the development of resistance. This has been clearly demonstrated with regard to platinum toxicity in which lack of repair is well correlated with sensitivity in cell lines. Several studies show DNA repair deficient cells as more sensitive to cis platinum than the
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
In view of the large amount of intracellular glutathione, it may be that non enzymatic reaction of glutathione with electrophiles could be relevant. Several studies have shown that glutathione depletion with buthionine sulfoximine (BSO), a blocker of glutathione synthesis, potentiates melphalan toxicity in numerous cell lines. It has not been shown so far that GST activity changes amount to more than associated stress responses following treatment with alkylating agents rather than definitive resistance mechanisms. However, the increased protection against adriamycin, although small, has been consistently found to be correlated with GSTrc. It is of interest that transformation with vH-ras and v-raf oncogenes induces resistance to a number of cytotoxic agents and induces both PGP and GSTTT in NIH3T3 cells. Definitive evidence with regard to gene transfection experiments awaits cloning the GST isoenzymes into a variety of cell types and the testing of different drug classes. Evidence so far suggests that resistance induced would be less than 3-fold. However, coordinate changes reducing ability to regenerate or transport glutathione may interact synergistically with other resistance pathways. Studies are in progress involving the clinical use of BSO and ethacrynic acid which deplete glutathione.
192
THE CANCER CELL
Table 3 Mechanisms of DNA repair This involves incision on both sides of a DNA adduct, filling in of the gap with polymerase(s), and rejoining with ligase(s). The types of lesions recognized by excision repair enzymes are caused by UV light, numerous carcinogens, and DNA cross-linking alkylating agents such as mitomycin C and cij-platinum
Glycosylases
These recognize specific abnormal bases (e.g. alkylated adenine and giianine residues). They excise the base to leave an apurinic site which is further processed by endonucleases. The lesions repaired by glycosylases are caused by methylating agents such as methyl nitrosourea (MNTJ) and methyl methane sulphonate (MMS)
Error-free repair mechanisms
A specific suicide protein, which dealkylates of O6 position of guanine residues in DNA, exists in both bacteria and eukaryotes. Photolyases reverse pyrimidine dimer formation induced by UV light
Recombination repair
This involves strand exchange between chromosomes and is implicated in doublestrand break repair. This type of damage is considered to be main cytotoxic lesion induced by x-rays and bleomycin. Recombination is also important in DNA cross-link repair in E. colt
Mismatch repair
This removes mispaired bases and preferentially repairs GT in favour of GC. This may protect against miscoding of deaminated 5methyl-cytosine residues. Defects in this system are associated with sensitivity to cis-platinum and methylating agents in E. coh
corresponding wild-type. Repair capacity is measured by unscheduled DNA synthesis (reflecting DNA synthesis in absence of DNA replication), reduction in the levels of cis platinum bound to DNA, DNA polymerase activity and the capacity of cell extracts in vitro to repair cis platinum damaged plasmids. Conversely cells made more resistant may have enhanced repair and other mechanisms are also involved. These include changes in transport, DNA binding proteins and metallothioneins.39 Resistance to nitrosoureas (e.g. chloroethylnitrosureas used in therapy of brain tumours, and melanomas) correlates with expression of the enzyme O 6 alkylguanine alkyltransferase (AT).40 AT dealkylates the oxygen atom and covalently transfers the alkyl group to a cysteine residue within the protein. Possibly AT inhibitors could potentiate anticancer effects and this has been demon-
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
Excision repair
DRUG RESISTANCE
193
SUMMARY Apart from recent advances in understanding different types of multidrug resistance, there is increasing evidence of the importance of membrane transport systems which are well reviewed.41 Almost every chemotherapeutic agent has been demonstrated to have altered transport after induction of resistance and in some cases this is the major mode of resistance, e.g. methotrexate. Drug resistance is usually mediated through more than one pathway and this is well exemplified by cis platinum in which DNA repair is of significance as well as altered transport, induction of metallothioneins and changes in GSH. Another example is adriamycin which, apart from resistance mediated by both topoisomerase and MDR pathways, may have other mechanisms of resistance, e.g. GSTs. Another cause of resistance may be the loss of regulatory sequences in key genes, a potential target of DNA damaging drugs such that they function in the presence of the drug. For example, in an SV40 viral test system isolates resisting platinum treatment were found to have acquired deletions within the promoter sequences known as GC boxes and drug resistance hence rested on loss of regulatory DNA sequences. DNA damage due to platinum may be more selective than previously thought.42 Resistance may be due to genetic variations in drug metabolism. A study of 6-mercaptopurine response in childhood acute lymphoblastic leukaemia correlated poor response with higher levels of the enzyme thiopurine methyltransferase which inactivates the drug.43 Gene amplification has been shown to be of major significance in methotrexate resistance with amplification of the dihydrofolate reductase gene (DFHR) leading to overproduction of the drug target.44
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
strated in pretreatment of murine leukaemia cells with O 6 methylguanine. This substrate of the protein competes with the alkyl lesion in DNA and hence prevents removal of the latter. Most DNA repair inhibitors are too toxic to be used clinically though hydroxyurea, a ribonucleotide reductase inhibitor inhibits both DNA repair and replication and has been used clinically. Other antimetabolites that inhibit DNA polymerases have also been studied (e.g. ara-C) but in vivo inhibition of repair in human tumours has not been demonstrated.
194
THE CANCER CELL
REFERENCES 1 Endicott JA, Ling V. The biochemistry of P-glycoprotein mediated multidrug resistance. Annu Rev Biochem 1989; 58: 136-171 2 Deuchars KL, Ling V. P-glycoprotein and multidrug resistance in cancer chemotherapy. Semin Oncol 1989; 16 (2): 156-165 3 Juranka PE, Zastawny RL, Ling V. P-glycoprotein: multidrug resistance and a superfamily of membrane associated transport proteins. FASEB J 1989; 3: 2583-2592 4 McGrath JP, Varshavsky A. The yeast STE6 gene encodes a homoguc of the mammalian multidrug resistant P-glycoprotein. Nature 1989; 340: 400-404 5 Ueda K, Pastan I, Gottesman MM. Isolation and sequence of the promoter region of the human multidrug resistance (P-glycoprotein) gene. J Biol Chem 1987; 262 (36): 17432-17436 6 Gros P, Ben-Nariah Y, Croop JM, Housman DE. Isolation and expression of a complementary DNA that confers multidrug resistance. Nature 1986; 323: 728-3 7 Chin KV, Tanaka S, Darlington G, Pastan I, Gottesman MM. Heat shock and arsenite increase expression of the multidrug resistance (MDR1) gene in human renal cell carcinoma cells. J Biol Chem 1990; 275 (1): 221-226 8 Scotto KW, Biedler JL, Melera PW. Amplification and expression of genes associated with multidrug resistance in mammalian cells. Science 1986; 232: 751-755 9 Hsu SI, Lothstein L, Horwitz SB. Differential overexpression of 3 MDR gene family members in multidrug resistance J774.2 mouse cells. J Biol Chem 1989; 264 (20): 12053-12062 10 Chambers TC, McAvoy EM, Jacobs JW, Eilon G. Protein Kinase C phosphorylates P-Glycoprotein in Multidrug Resistant Human KB Carcinoma Cells. J Biol Chem 1990 265 (13): 7679-7686 11 Choi K, Chen C-J, Kriegler M, Roninson IB. An altered pattern of cross resistance in multidrug resistant cells results from spontaneous mutations in the MDR1 (P-glycoprotein) gene. Cell 1988; 53: 519-529 12 Marino PA, Gottesman MM, Pastan I. MDR gene regulation in rat liver. Cell Growth Differ 1990; 1: 57-62 13 Weinstein RS, Kuszak JR, Kluskensl F, Coon JS. P-glycoprotein in pathology: the multidrug resistance gene family in humans. Hum Pathol 1990; 21 (1): 34-^8 14 Fojo AJ, Ueda K, Slamon DJ, Poplack DG, Gottcsman, MM, Pastan I. Expression of a multidrug resistance gene in human rumours and tissues. Proc Natl Acad Sri USA 1987; 84: 265-269 15 Gordon-Cardo C, O'Brien JP, Casals D, et al. Multidrug resistance gene (P-glycoprotein) is expressed by endothelial cells at blood-brain barrier sites. Proc Natl Acad Sri USA 1989; 86: 65-68 16 Goldstein LJ, Galski H, Fojo A, et al. Expression of a multidrug resistance gene in human cancers. J Natl Cancer Inst 1989; 8 (2): 116-124
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
Understanding these mechanisms has been helpful in developing strategies to overcome resistance by blocking relevant pathways through drug design. It has also contributed to major understanding of the normal tissue protection against carcinogens and environmental toxins and may be relevant to individual cancer risk and familial cancers. In future profiles of these resistance mechanisms may help select therapy for patients.
DRUG RESISTANCE
195
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
17 Chan HSL, Thorner PS, Haddad G, Ling V. Immunohistochemical detection of P-glycoprotein: prognostic correlation in soft tissue sarcoma of childhood. J Clin Oncol 1990; 81 (4): 689-704 18 Lai S-L, Goldstein LJ, Gottesman MM, et al. MDR1 gene expression in lung cancer. J Nad Cancer Inst 1989; 81 (15): 1144-1150 19 Bell DR, Gerlach JH, Kartner N, Buick, RN, Ling V. Detection of P-glycoprotcin in ovarian cancer: a molecular marker associated with multidrug resistance. J Clin Oncol 1985; 3 (3): 311-315 20 Rothenberg ML, Mickley LA, Cole, DE, et al. Expression of the MDR1/P170 gene in patients with acute lymphoblastic leukaemia. Blood 1989; 74 (4): 13881395 21 Dalton WS, Grogan TM, Rybski JA, et al. Immunohistochemical detection and quantitation of P glycoprotein in multiple drug resistant human myeloma cells. Blood 1989; 73 (3): 747-752 22 Stewart DJ, Evans WK. Non-chemotherapeutic agents that potentiate chemotherapy efficiency. Cancer Treat Rev 1989; 16: 1-40 23 Dalton WS, Grogan JM, Meltzer PS, et al. Drug resistance in multiple myeloma and non-Hodgkin's lymphoma: detection of P-glycoprotein and potential circumvention by addition of verapamil in chemotherapy. J Clin Oncol 1989; 7: 415-424 24 Wang JC. DNA topoisomerases. Annu Rev Biochem 1985; 54: 664-697 25 Liu LF. DNA topoisomcrase poisons as antitumour drugs. Annu Rev Biochem 1989; 58: 351-375 26 Davies SM, Robson CN, Davies SL, Hickson ID. Nuclear topoisomerase II levels correlate with the sensitivity of mammalian cells to intercalating agents and cpipodophyllotoxins. J Biol Chem 1988; 263 (33): 17724-17729 27 Heck, MMS, Hittelman WN, Earnshaw WL. In vivo phosphorylation of the 170-kDa form of eukaryotic DNA topoisomerase II. J Biol Chem 1989; 264 (26): 15161-15164 28 Zwelling LA, Hinds M, Chan D, et al. Characterisation of an amsacrine resistant line of human leukaemia cells. J Biol Chem 1989; 264 (28): 16411-16420 29 Pommier Y, Kerrigan D, Schwartz R, et al. Altered DNA topoisomerase II activity in Chinese hamster cells resistant to topoisomerase II inhibitors. Cancer Res 1976; 46: 3075-3081 30 Fernandes DJ, Danks MK and Beck WT. Decreased nuclear matrix DNA topoisomerase II in human leukaemia cells resistant to VM-26 and mAMSA. Biochemistry 1990; 29: 4235-4241 31 Chung TDY, Drake FH, Tan KB, Per SR, Crooke ST, Mirabelli EK. Characterisation and immunological identification of cDNA clones encoding 2 human DNA topoisomerase II isozymes. Proc Nat] Acad Sci USA 1989; 86: 94319435 32 Woessner RD, Chung TDY, Hofmann GA et al. Differences between normal and ros-transformed NIH-3T3 cells in expression of The 170kD and 180kD forms of topoisomerase II. Cancer Res 1990; 50: 2901-2908 33 Holden JA, Rolfson DH, Winner CT. Human DNA topoisomerase II: evaluation of enzyme activity in normal and neoplastic tissues. Biochemistry 1990; 29: 2127-2134 34 Findlay GJ, Baguley BC, Snow K, Judd W. Multiple patterns of resistance of human leukemia cell sublines to amsacrine analogues. J Nad Cancer Inst 1990: 82 (8): 662-667 35 Pickett CB, Lu AYH. Glutathione S-transferases: gene structure, regulation and biological function. Annu Rev Biochem 1989; 58743-58764 36 Morrow CS, Cowan KH. Glutathione S-transferases and drug resistance. Cancer Cells 1990; (2) 1: 15-22 37 Nakagawa K, Saijo N, Tsuchida S, et al. Glutathione S-transferase pi as a
196
38
40 41 42 43 44
determinant of drug resistance in transfectant cell lines. J Biol Chem 1990; 265 (8): 4296-4301 Harris AL. DNA repair and resistance to chemotherapy. Cancer Surv, 1985; 4 (3): 601-624 Andrews PA, Howell SB. Cellular pharmacology of cisplatin: perspectives on mechanisms of acquired resistance. Cancer Cells 1990; (2) 2: 35-43 D'lncalci M, Sifti L, Tavema P, Catapano, CV. Importance of the DNA repair enzyme O 6 alkyl guanine alkyltransfcrase in cancer chemotherapy. Cancer Treat Rev 1988; 15: 279-292 Fry DW, Jackson RL. Membrane transport alterations of a mechanism of resistance to anticancer agents. Cancer Surv 1986; 5 (2): 48-79 Buchanan RL, Gralla JD. Cis-platin resistance and mechanisms in a viral test system. Biochemistry 1990; 29: 3436-3442 Lennard L, Lilleyman JS, Van Loon J, Weinshilboum RM. Genetic variation in response to 6-mercaptopurine for childhood Acute Lymphoblastic Leukemia. Lancet 1990 (336); 8709: 225-229 Stark GR, Deban'ssc M, Giulotto E, Wahl GM. Recent progress in understanding mechanisms of mammalian DNA amplification. Cell 1989; (57): 901-908
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
39
THE CANCER CELL
Drug resistance Molecular Oncology Laboratory, Imperial Cancer Research Fund Laboratories, Institute of Molecular Medicine, University of Oxford, John Radcliffe Hospital, Oxford, UK
Chemotherapy cures a minority of adult tumours (e.g. Hodgkin's and non-Hodgkin's lymphoma, acute leukaemia, teratoma) and the majority of childhood tumours. Prolongation of survival by chemotherapy has been shown for small cell lung cancer, ovarian cancer and breast carcinoma (when used as an adjuvant). However, in the majority of solid tumours there is a less than 20% response to chemotherapy and even curable tumours may relapse and become resistant. Resistance may be de novo, acquired as a stable change within the cell, or be rapidly inducible within the cell after drug administration. Several mechanisms have been described including multidrug resistance, glutathione transferases and DNA repair. Understanding these mechanisms may help to improve the therapeutic ratio and develop new approaches.
Mechanisms of drug resistance relating to pharmacokinetics can be considered to operate from the point of administration to achievement of cytocidal concentrations at the target site. Important factors may include absorption, distribution and metabolism of drug. Numerous effects mediate bioavailability of drug from poor absorption to poor rumour blood supply. These factors will not be discussed in diis review which will concentrate on some recent advances in understanding drug resistance at the molecular level. In the past few years correlation between in vitro cellular responses and the development of resistance in vivo has been most clearly shown in multidrug resistance.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
D Hochhauser A L Harris
DRUG RESISTANCE
179
MULTIDRUG RESISTANCE
Homologies of PGP Clear homologies exist between PGP and several bacterial transport proteins in particular with conservation of the ATP binding domains.3 These proteins function to transport individual peptide and carbohydrate species into bacterial cells. In bacteria the hydrophilic and hydrophobic domains are associated as independent integral membrane proteins rather than being linked in a single polypeptide chain. They are energized by hydrolysis of ATP. In eukaryotes the only denned normal role for an MDR homologue has been found in Saccharomyces cerevisiae. The product of the STE6 gene is actively involved in the transport of a factor mating (MAT) pheromone.4 It is not essential for cell viability and does not confer a drug resistant phenotype. Recent identification of the cystic fibrosis gene reveals a similarity of the cystic fibrosis transmembrane regulator (CFTR) to these transport systems with ATP binding domains and potential membrane spanning hydrophobic sequences. Cloning of the MDR gene Cloning of the MDR gene in higher eukaryotes has been achieved by a variety of approaches including in gel renaturation to detect amplified sequences, differential hybridization, or isolation of cDNA clones using antibodies to PGP. Definitive evidence for the involvement of these genes in the multidrug resistance phenotype has been obtained primarily by transfection studies as well as analysis of resistant cell lines, clinical material and the use of MDR antagonists.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
Multidrug resistance (MDR) is the phenomenon whereby exposure to one drug induces cross resistance to a variety of agents of different chemical classes to which the cell has never been exposed. These include primarily adriamycin, vinca alkaloids and mitomycin C but not platinum, bleomycin and alkylating agents. The hallmark of MDR is the expression of P glycoprotein (PGP), a 170 kD protein with six hydrophobic domains and a tandemly duplicated ATP binding domain.1'2 PGP acts as a transmembrane exporter of drugs but the normal substrates have yet to be identified.
180
THE CANCER CELL
Chromosomal location and genomic organisation of the MDR gene
Molecular genetic studies Major evidence for PGP involvement in drug resistance was through gene transfection studies. In man only the MDR1 gene has been demonstrated to mediate MDR through transfection experiments and MDR-3 has no effect. Expression of a full length cDNA conferred resistance to colchicine, doxorubicin and vinblastine in NIH3T3 cells.6 However, in rodents both PGP1 and PGP2 mediate resistance. Regulation of MDR gene P-glycoprotein is regulated as a stress protein.7 For example, the renal adenocarcinoma cell line HDB46 shows an 8-fold increase in MDR1 RNA levels in response to heat shock, ethanol, sodium arsenite and cadmium consistent with the role of MDR1 as a stress-inducible gene response to environmental insults. Induced transcripts were generated predominantly from the downstream promoter. Sequence analysis of the 5' region of MDR1 reveals several heat shock regulatory elements. The induction of MDR1 Table 1 Expression of PGP genes in human, mouse and hamster P-glycoprotein Gene Class Species
I
II
III
Human Mouse Hamster
mdr 1 mdr 3 PgP 1
mdr 1 Pgp2
mdr 3 mdr 2 Pgp3
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
In man the MDR1 gene is located on chromosome 7q 21-31 and consists of an open reading frame of 1280 amino acids. There are 2 human and 3 rodent MDR genes (Table 1). Control of expression of both human MDR genes occurs at the level of gene copy number, transcription, translation, and post-translationally. Isolation and sequencing of a one kilobase portion of genomic DNA encoding the promoter region of the human MDR-1 gene indicated a consensus CAAT box and two GC box like sequences but no TATA sequences.3 Although two potential promoter start sites have been found, only the downstream promoter was activated in the KB lines tested.
DRUG RESISTANCE
181
Protein modification Post-translational modification of PGP is also significant. In man a 140 kD precursor protein has been identified which is gradually converted to a 170 kD form over 2 to 4 hours. The amino acid sequence of PGP indicates 10 potential N-glycosylation sites, but only 3 appear on the external surface of the plasma membrane. However, glycosylation deficient mutants of MDR cells show no change in the resistance profiles compared with parental cells and glycosylation may therefore be of questionable significance. This was confirmed with studies in which treatment with tunicamycin (a glycosylation inhibitor) caused no change in resistance patterns.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
RNA occurs concomitantly with development of a 2- to 6-fold increase in drug resistance to vinblastine following heat shock. However, induction of glucose regulated proteins by glucose deprivation, anoxia, calcium ionophore A23187 and 2-deoxyglucose led to no significant change in the levels of PGP despite an increase in doxorubicin resistance. Furthermore, in several other cell lines (including human cervical carcinomas, liver carcinomas and fetal hepatocytes) no heat shock induction could be demonstrated, indicating that these effects may be tissue- and cell-type specific. Regulation of MDR1 by gene amplification was suggested by the presence of double minutes and homogeneously staining regions.8 Several genes may coamplify with P-glycoprotein. A major coamplified product is sorcin, an acidic 22 kD protein which has homologies to calpain and binds calcium. It may be that specific differential amplification of sorcin and other gene products is partially responsible for the differential drug sensitivities found in different MDR expressing resistant cell lines but no correlation has been shown with resistance pattern. Evidence of differential overexpression of MDR isotypes has come from studies on the murine macrophage line J7742 cells selected for resistance to colchicine, vinblastine or taxol.9 There was production of two distinctly sized precursors of 125 and 120 kD, with a switch from the 120 to the 125 kD precursor occurring during selection. In rodents a switch from one form to another may correlate with increased resistance to vinblastine, taxol and doxorubicin, with no detectable change in the total amount of P-glycoprotein produced. However, similar phenomena have not been detected in man as only one MDR gene confers resistance.
182
THE CANCER CELL
Mutations of MDR gene Other factors may alter patterns of cross resistance in multidrug resistance cells due to mutations within the MDR1 gene. Studies in KB cells revealed that mutations of codon 185 changing from glydne to valine, resulting in a changed preferential resistance pattern from vinblastine to colchicine.11 The nucleotide sequences were analysed using the polymerase chain reaction and it was found that the colchicine specific sequence exists de novo within KB-DNA prior to selection. Transfectants of wild type and colchicine resistant cDNAs conferred the appropriate preferential resistance. It is thus clear that amino acid 185 plays a key role in PGP drug interaction. This amino acid is located within the first hydrophobic region on the cytoplasmic side of the membrane and may be part of a drug binding site. RNA stabilization Recent evidence on the MDR-1 gene in regenerating rat liver reveals another possible level of control with mRNA degradation being controlled by sequences in the noncoding region.12 Summary of mechanisms of regulation Expression of the MDR gene is regulated by amplification, transcription and translation. The relative contributions of each may vary throughout the selection process. In a multidrug resistant ovarian carcinoma cell line overexpression of MDR1 mRNA was found without evidence of amplification at low levels of drug resistance. At intermediate drug concentrations, amplification became increasingly significant while at high drug level PGP increased without further changes in mRNA or gene copy number, consistent with translational modification or mutation. Apart from indicating that MDR may be mediated at different levels of protein in the same tumour under different selection pressures, this study
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
The MDR1 expressing mouse line J774-2 has been shown to be a substrate for cyclic AMP dependent protein kinase-A with phosphorylation of serine and threonine residues having been demonstrated in vitro. PGP seems to be phosphorylated in its basal state primarily by protein kinase C and this may play a role in the drug transport process.10
DRUG RESISTANCE
183
Tissue distribution of MDR expression Studies of tissue distribution have involved the use of immunohistochemistry with mouse and human monoclonal antibodies as well as ribonuclease protection assays and in situ mRNA hybridization. Specific monoclonals can distinguish between the various PGP isoforms.13 The highest levels of PGP were found in kidney, adrenal cortex (zona fasciculata and reticulata), stomach, duodenum, colon and placenta.14 Expression occurs primarily in specialized epithelial cells with secretory or excretory functions as well as in placental trophoblasts. Strong expression was also found on endothelial cells of capillary blood vessels at blood tissue barrier sites, primarily capillaries of the central nervous system and testes as well as the capillary dermis.' s The potential significance of these observations is that the brain and testis constitute sanctuary sites in which relapse following systemic chemotherapy in conditions such as acute lymphoblastic leukaemia occurs presumably because of failure of drug penetration. Location of PGP in these cells is primarily on the nominal surface consistent with the role of PGP in secretion of as yet unknown substrate. Clinical significance—expression in tumours Studies of the significance of MDR in human tumours have recently become available. The most extensive study reported on levels of MDR1 mRNA in over 400 human cancers.16 Quantification of mRNA was measured using slot blot analysis comparing tumour samples to known drug sensitive (KB-3-1) and resistant (KB-8-5) cells. Expression was denned as high if MDR1 mRNA was detectable in over 50% of cancers in each group (see Table 2). RNAse protection analysis revealed use of the downstream promoting site of each of these cancers. In general, there is an inverse correlation of MDR expression with relative chemosensitivity of the tumour type. Thus, breast and ovarian cancer have lower levels than colon and renal cell carcinoma. However, many tumours known to be resistant usually had low levels, e.g. sarcomas and non small cell lung cancers. This
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
also indicates that measurement of AiDRl mRNA alone may give a misleading impression of the amount of functional PGP present. Other studies in KB-Hela cells also showed overexpression of MDR mRNA preceding gene amplification during selection.
184
THE CANCER CELL
Table 2 Expression of MDR mRNA levels in tumours Low MDR
Colon carcinoma Renal cell carcinoma Hepatoma Adrenocortical cancer Phaeochromocytoma Pancreas islet-cell cancer CML (blast crisis) Carcinoid
Breast cancer Non small cell lung cancer Bladder cancer CML (Chronic phase) Small cell lung cancer Lymphomas Ovary Acute leukaemias
may reflect assay sensitivity and difficulty of knowing what level is important clinically. Clinical significance has also been assessed in a retrospective immunohistochemical study of rhabdomyosarcoma and undifferentiated sarcomas which showed that all 9 patients with MDR relapsed.17 Of 21 PGP-negative patients, 20 showed durable complete response. A study of various lung cancers and nontumourous lung tissue showed no significant difference in MDR and minimal change following therapy, though correlation has been found in other studies. The expression of MDR clearly varies with different cell and tumour types.18 There have been several other studies of MDR in tumours following treatment. It is often problematic to delineate the exact role of MDR in causing failure of therapy because of induction of a variety of different resistance pathways. The earliest report of PGP elevation in a human malignancy was in 5 cases of relapsed ovarian cancer where cells showed raised MDR compared with normal ovarian tissue which did not express MDR. 19 In one of these patients, increased MDR expression was shown following failure of therapy. In a study of 9 newly diagnosed patients with acute lymphoblastic leukaemia, 8 had low levels of MDR mRNA at time of diagnosis. One patient with elevated MDR mRNA failed to achieve remission. However, a significant number of relapsing patients (5 out of 19) in the same study also had low MDR mRNA levels.20 In another small study of 15 patients with acute non lymphoblastic leukaemia 3 out of 10 with significant MDR1 levels relapsed after a short interval, in contrast to 4 out of 5 with minimal or absent MDR expression who achieved remission. Immunohistochemistry of a series of MDR human myeloma cell lines showed correlation of increased P-glycoprotein with resistance and higher levels in relapsed patients, though no measurements were taken prior to treatment.21 The majority of studies so
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
High MDR
DRUG RESISTANCE
185
Reversal of MDR A variety of pharmacological agents reverse MDR and this has been extensively reviewed.22 These range from the calcium channel blockers, particularly verapamil and nifedipine to tamoxifen, phenothiazines and cyclosporin, among many others. The concentrations required to reverse MDR in vitro would produce toxicity in vivo and this has been a major limitation to effective clinical use of, for example, verapamil. Some of these agents probably act by binding to PGP and thereby cause intracellular accumulation of drug. Odiers may act by binding calmodulin or do not affect efflux but cause redistribution of subcellular localization. Normal basal to apical flux of vinblastine MDR in epithelial cells is reversed by verapamil with enhanced phosphorylation of PGP, but verapamil has also been noted to potentiate bleomycin, cis platdn and 5-fluorouracil toxicity which are not associated with the MDR phenotype. Preliminary clinical studies of verapamil have so far proved to be disappointing in most cases with effective plasma concentrations causing severe hypotension and heart block. However, recent studies on the use of verapamil in multiple myeloma have shown a significant improvement in remission rate and are encouraging.23 Possibly using different MDR antagonists together at lower levels such as tamoxifen with verapamil, may overcome MDR without undue toxicity seen with higher levels of each drug. Alternatively, chemotherapeutic drug analogues not binding to PGP, e.g. the adriamycin derivative aclacinomycin, could be developed. Although expression of P-glycoprotein thus far appears to be the most significant cause of the multidrug resistant phenotype, increasing attention has recently been paid to the so called atypical multidrug resistance (atMDR). This has been applied to circumstances where the multidrug resistance phenotype has occurred without overexpression of the MDR gene. The commonest pattern appears to be of drugs which act on cellular topoisomerase II as a target.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
far have constituted too small a number of patients to allow significance to be proven of the clinical relevance of MDR expression. Ongoing longitudinal studies will help to clarify the situation as to whether subgroups of tumour patients expressing MDR predict response to chemotherapy and will thus be appropriate for use of MDR modulators.
186
THE CANCER CELL
TOPOISOMERASE II
Topoisomerase II as a drug target Several classes of drugs are termed topoisomerase inhibitors though they do not inhibit the initial stages of topoisomerase action. These drugs include intercalating agents—such as acridines (mAMSA), anthracyclines (adriamycin, daunomycin), anthracene-diones (mitoxantrone) and ellipticines. There is also an important group of non-intercalating agents which appear to bind directly to topo II, most notably teniposide (VP26) and etoposide (VP16). These drugs bind to topo II forming a cleavable complex; so-called because denaturation of the trapped protein with SDS or alkali reveals the DNA strand breaks with protein covalently linked to the 5-phosphoryl end of each broken strand. Since each complex requires two proteins, an increase in the amount of topoisomerase will correlate with increase strand breakage and potential toxicity. It has been shown, for example, that CHO cell lines sensitive to etoposide overexpressed topoisomerase II. 2 6 Although the exact mode of cell death is unclear, presumably the cleavable complex prevents chromosome segregation as well as DNA synthesis. Almost all topoisomerase inhibiting drugs increase the number of topo II associated DNA strand breaks although this does not always correlate with cytotoxicity because the drugs may have other modes of action.
Regulation of topo II Topo II is expressed primarily during the proliferative phases of growth and is down regulated at plateau phase. The amount of
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
Topoisomerase II (topo II), the eukaryotic homologue of bacterial DNA gyrase, is a 170 kD homodimeric protein which plays a role in DNA replication, chromosome scaffold formation, chromosomal segregation, and possibly recombination and gene transcription. 24 ' 25 In Saccharomyces cerevisiae, topo II null mutations are lethal because of the inability to separate chromosomes at mitosis. The enzyme acts by producing DNA single and double strand breaks and it attaches covalently to the 5' ends of the break. Double strand breaks are staggered by 4 base pairs on opposite stands and the protein attaches as a dimer. Subsequently strand passage occurs with changes in both supercoiling and DNA relaxation.
DRUG RESISTANCE
187
Topoisomerase II and drug resistance There has been demonstration of so-called atypical multidrug resistant phenotype in numerous cell lines. Characteristically there is cross resistance to the full range of anti-topo II drugs though not to the vinca alkaloids. These mechanisms may include: Point mutations and abnormally functioning topoisomerases
An HL60 human leukaemia cell line showed 100-fold resistance to mAMSA in contrast to 2- to 3-fold increase in etoposide resistance. 28 Cleavage of DNA produced by mAMSA or etoposide had an absolute requirement for ATP in contrast to die situation in
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
topo II increases at the onset of DNA replication continues to increase through S and G2 phases, peaks in die late G2 to M phase and then drops after mitosis. At confluence topo II is switched off in untransformed but not in transformed cells. The regulation of topo II during the cell cycle may be relevant as to which phase is most sensitive to topo II inhibitors. Following serum stimulation of BALB-C3T3 cells, actively proliferating cells showed much greater sensitivity to etoposide than quiescent cells. The increase in drug sensitivity began during S phase and reached a peak just before mitosis with a maximal 2.5-fold increase in drug sensitivity. However, although the maximal number of topo II associated strand breaks occurred during G 2 , cytotoxicity was maximal during S phase, suggesting interactions with other mechanisms leads to cell death. Modulation of topo II activity during the cell cycle involves phosphorylation. This has been demonstrated for casein kinase and protein kinase C, and these changes may be crucial in switching on the activity of topoisomerase. Phosphorylation has been shown to lead to a 3-fold increase in enzyme activity and dephosphorylation inactivates die enzyme. Studies of 32 P incorporation into DNA topo II in vivo in chicken lymphoblastoid cells showed phosphorylation highest at G2 and M, suggesting phosphorylation may be involved in enzyme activation for sister chromatid disjunction.27 Glucose deprivation, hypoxia and the stress response to heavy metals have been shown to downregulate the amount of topo II while epidermal growth factor (EGF) is an upregulator. The latter effect may be simply due to causing increased cell proliferation.
188
THE CANCER CELL
Post translational
In a study of teniposide resistant human CEM lines the amount of topo II bound to the nuclear matrix decreased due to a mutant enzyme which differs in salt stability thereby affecting attachment to the nuclear matrix.30 It is probable that the functional topo II is that which is associated with the matrix, and with newly replicated DNA. Methylation Several studies have demonstrated hypermethylated sites. This could be of significance in regulating the transcription of nonmutant alleles. Glucose deprivation and hypoxia
These factors may be relevant within the tumour mass and contribute towards decreased topoisomerase expression. Upregulation of topoisomerase I
This has been shown in several cases. Topoisomerase I, an enzyme which produces single strand breaks in DNA and is involved in transcription, may increase as a compensatory effect widi topo II downregulation (as occurs in S. cerevisiae). Switch to a novel form of topo II ($) Purification of a topoisomerase II from mAMSA resistant P388 leukaemia cells indicated another form of topo II of 180 kD differing in several respects from the previously known 170 kD form.31 Partial cloning of the gene shows that it may lack the leucine zipper motif which is probably important in dimerisation of the 170 kD form and that it differs in expression throughout the cell cycle (topo II P levels peak during G! and fall during the rest of the cell cycle), and in drug sensitivity. The 170 kD topo II (topo I la) is expressed in rapidly proliferat-
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
wild-type cells. A new restriction enzyme polymorphism was found in the resistant forms. This correlated with a mutation in the ATP binding site which has been found to induce mAMSA resistance in the bacteriophage T4 DNA topoisomerase. The extent to which downregulation of the total amount of topo II is a significant resistance mechanism remains unclear, although reduced levels of topo II shown in some studies could be due to decreased half-life of the mutant protein. Several studies have demonstrated abnormal functional forms of topo II in resistant cells with different drug cleavage patterns.29
DRUG RESISTANCE
189
Tissue distribution of topoisomerase II Studies of topo II activity in normal tissues showed the highest levels in spleen and thymus.33 Expression within tumours was highest in breast cancer and leiomyosarcoma. However, levels within these tumours were in the same range as normal tissues; possibly normal tissue topo II is relatively teniposide resistant. These results were unexpected since one would expect highest topo II levels in proliferating cells. Further studies on tissue levels are needed using antibodies and activity assays. Clinical implications Further studies are required to elucidate the clinical significance of altered topo II as well as possible mediods of circumvention. It may be possible to reduce topo II resistance by molecular design. Mechanisms of resistance to the topo II inhibitor mAMSA include altered transport and MDR. By altering substituents on the anilino acridine nucleus resistance to both mechanisms was overcome.34 Upregulation of topo I could be significant as inhibitors, notably camptothecin, could be used and cells exhibiting this mechanism of resistance may be hypersensitive to topo I inhibitors. Another route could be through the development of competitive topoisomerase inhibitors rather than the current stoichiometric agents i.e. competition with topo II for binding to DNA would be more effective in low topo II expressers. There appears to be a link between MDR and topo II regulation with possible inverse correlation of these.
GLUTATHIONE TRANSFERASES The involvement of glutathione and glutathione transferases (GST) in the multi-drug resistant phenotype has been difficult to
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
ing cells and its ratio increases compared to the 180 kD form (topo IIP) in ras transfected NIH3T3 cells.32 As the 170 kD protein is preferentially sensitive to teniposide and merbarone this could account for drug effectiveness against tumours which possess a higher percentage of topo I la—i.e. transformed cells as opposed to normal cells. Similarly, switch of expression from topo Ila to topo Up could lead to an 8-fold degree of resistance with equivalent protein expression.
190
THE CANCER CELL
GST expression in drug resistance Several anticancer drugs may be inactivated by GST catalysis— e.g. melphalan, chlorambucil and cyclophosphamide. There have been numerous reports correlating increases in GST isozymes, particularly the TC family, with the onset of drug resistance. There may be increased amounts of enzymes involved in glutathione synthesis and also glutathione peroxidase. GSTTI has also been thought to be significant as it constitutes the predominant isozyme found in human cancers with 2- to 4-fold increases in RNA levels in tumours of the colon, bladder, ovary and stomach relative to normal tissues. Also it has been found to be overexpressed in several multi-drug resistant lines, particularly in adriamycin resistant lines. Studies in ovarian cancer patients show expression of a and n isozymes following treatment. Transfection studies Transfection of GSTTI C D N A into ras transformed NIH3T3 cells led to elevated GSTTI enzyme levels but although there was a moderate 1.8- to 3-fold increase in resistance to adriamycin and ethacrynic acid (a substrate of GSTTT) there was no change in resistance to cis platinum, chlorambucil and melphalan.37 In another study a doxorubicin resistant line of human breast cancer MCF7 with both GSTir and MDR expression was analysed. The MDR and GSTTI CDNA'S from these cells were cloned and transfected but although MDR conferred a resistance phenotype GSTTI gave only moderate resistance to ethacrynic acid. The transfection of GSTTC, together with MDR showed no significant additive effect.
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
clarify. Glutathione transferases are enzymes involved in detoxification, distributed in all organs but primarily in the liver and kidney. Their mode of action is by conjugation of glutathione (a tripeptide present in all organs in mM amounts) via the sulphur atom of its cystine residue to various electrophiles. GST's may also act as intracellular binding proteins (e.g. for bilirubin and steroids) and may constitute up to 10% of the total cellular protein in the liver. In man GST exists in cytosolic and microsomal forms. There are 3 major classes of cytosolic GST: n, a, \i. Several comprehensive recent reviews deal with classification of the GST's. 3 5 3 6
DRUG RESISTANCE
191
Glutathione depletion
DNA REPAIR The importance of basal DNA repair mechanisms in carcinogenesis and drug resistance is suggested by human repair deficiency diseases in which cancer susceptibility is high.38 For example, cells from patients with ataxia -telangiectasia are hypersensitive to topoisomerase II inhibitors and x-ray irradiation while xeroderma pigmentosa cells are hypersensitive to UV irradiation and cis platinum. There are multiple mechanisms of DNA repair {see Table 3) and probably at least 100-200 genes involved in DNA repair in mammalian cells. In view of the interaction of anticancer drugs with DNA by intercalation, alkylation and interference with DNA synthesis, it is likely that DNA repair is of key significance in the development of resistance. This has been clearly demonstrated with regard to platinum toxicity in which lack of repair is well correlated with sensitivity in cell lines. Several studies show DNA repair deficient cells as more sensitive to cis platinum than the
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
In view of the large amount of intracellular glutathione, it may be that non enzymatic reaction of glutathione with electrophiles could be relevant. Several studies have shown that glutathione depletion with buthionine sulfoximine (BSO), a blocker of glutathione synthesis, potentiates melphalan toxicity in numerous cell lines. It has not been shown so far that GST activity changes amount to more than associated stress responses following treatment with alkylating agents rather than definitive resistance mechanisms. However, the increased protection against adriamycin, although small, has been consistently found to be correlated with GSTrc. It is of interest that transformation with vH-ras and v-raf oncogenes induces resistance to a number of cytotoxic agents and induces both PGP and GSTTT in NIH3T3 cells. Definitive evidence with regard to gene transfection experiments awaits cloning the GST isoenzymes into a variety of cell types and the testing of different drug classes. Evidence so far suggests that resistance induced would be less than 3-fold. However, coordinate changes reducing ability to regenerate or transport glutathione may interact synergistically with other resistance pathways. Studies are in progress involving the clinical use of BSO and ethacrynic acid which deplete glutathione.
192
THE CANCER CELL
Table 3 Mechanisms of DNA repair This involves incision on both sides of a DNA adduct, filling in of the gap with polymerase(s), and rejoining with ligase(s). The types of lesions recognized by excision repair enzymes are caused by UV light, numerous carcinogens, and DNA cross-linking alkylating agents such as mitomycin C and cij-platinum
Glycosylases
These recognize specific abnormal bases (e.g. alkylated adenine and giianine residues). They excise the base to leave an apurinic site which is further processed by endonucleases. The lesions repaired by glycosylases are caused by methylating agents such as methyl nitrosourea (MNTJ) and methyl methane sulphonate (MMS)
Error-free repair mechanisms
A specific suicide protein, which dealkylates of O6 position of guanine residues in DNA, exists in both bacteria and eukaryotes. Photolyases reverse pyrimidine dimer formation induced by UV light
Recombination repair
This involves strand exchange between chromosomes and is implicated in doublestrand break repair. This type of damage is considered to be main cytotoxic lesion induced by x-rays and bleomycin. Recombination is also important in DNA cross-link repair in E. colt
Mismatch repair
This removes mispaired bases and preferentially repairs GT in favour of GC. This may protect against miscoding of deaminated 5methyl-cytosine residues. Defects in this system are associated with sensitivity to cis-platinum and methylating agents in E. coh
corresponding wild-type. Repair capacity is measured by unscheduled DNA synthesis (reflecting DNA synthesis in absence of DNA replication), reduction in the levels of cis platinum bound to DNA, DNA polymerase activity and the capacity of cell extracts in vitro to repair cis platinum damaged plasmids. Conversely cells made more resistant may have enhanced repair and other mechanisms are also involved. These include changes in transport, DNA binding proteins and metallothioneins.39 Resistance to nitrosoureas (e.g. chloroethylnitrosureas used in therapy of brain tumours, and melanomas) correlates with expression of the enzyme O 6 alkylguanine alkyltransferase (AT).40 AT dealkylates the oxygen atom and covalently transfers the alkyl group to a cysteine residue within the protein. Possibly AT inhibitors could potentiate anticancer effects and this has been demon-
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
Excision repair
DRUG RESISTANCE
193
SUMMARY Apart from recent advances in understanding different types of multidrug resistance, there is increasing evidence of the importance of membrane transport systems which are well reviewed.41 Almost every chemotherapeutic agent has been demonstrated to have altered transport after induction of resistance and in some cases this is the major mode of resistance, e.g. methotrexate. Drug resistance is usually mediated through more than one pathway and this is well exemplified by cis platinum in which DNA repair is of significance as well as altered transport, induction of metallothioneins and changes in GSH. Another example is adriamycin which, apart from resistance mediated by both topoisomerase and MDR pathways, may have other mechanisms of resistance, e.g. GSTs. Another cause of resistance may be the loss of regulatory sequences in key genes, a potential target of DNA damaging drugs such that they function in the presence of the drug. For example, in an SV40 viral test system isolates resisting platinum treatment were found to have acquired deletions within the promoter sequences known as GC boxes and drug resistance hence rested on loss of regulatory DNA sequences. DNA damage due to platinum may be more selective than previously thought.42 Resistance may be due to genetic variations in drug metabolism. A study of 6-mercaptopurine response in childhood acute lymphoblastic leukaemia correlated poor response with higher levels of the enzyme thiopurine methyltransferase which inactivates the drug.43 Gene amplification has been shown to be of major significance in methotrexate resistance with amplification of the dihydrofolate reductase gene (DFHR) leading to overproduction of the drug target.44
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
strated in pretreatment of murine leukaemia cells with O 6 methylguanine. This substrate of the protein competes with the alkyl lesion in DNA and hence prevents removal of the latter. Most DNA repair inhibitors are too toxic to be used clinically though hydroxyurea, a ribonucleotide reductase inhibitor inhibits both DNA repair and replication and has been used clinically. Other antimetabolites that inhibit DNA polymerases have also been studied (e.g. ara-C) but in vivo inhibition of repair in human tumours has not been demonstrated.
194
THE CANCER CELL
REFERENCES 1 Endicott JA, Ling V. The biochemistry of P-glycoprotein mediated multidrug resistance. Annu Rev Biochem 1989; 58: 136-171 2 Deuchars KL, Ling V. P-glycoprotein and multidrug resistance in cancer chemotherapy. Semin Oncol 1989; 16 (2): 156-165 3 Juranka PE, Zastawny RL, Ling V. P-glycoprotein: multidrug resistance and a superfamily of membrane associated transport proteins. FASEB J 1989; 3: 2583-2592 4 McGrath JP, Varshavsky A. The yeast STE6 gene encodes a homoguc of the mammalian multidrug resistant P-glycoprotein. Nature 1989; 340: 400-404 5 Ueda K, Pastan I, Gottesman MM. Isolation and sequence of the promoter region of the human multidrug resistance (P-glycoprotein) gene. J Biol Chem 1987; 262 (36): 17432-17436 6 Gros P, Ben-Nariah Y, Croop JM, Housman DE. Isolation and expression of a complementary DNA that confers multidrug resistance. Nature 1986; 323: 728-3 7 Chin KV, Tanaka S, Darlington G, Pastan I, Gottesman MM. Heat shock and arsenite increase expression of the multidrug resistance (MDR1) gene in human renal cell carcinoma cells. J Biol Chem 1990; 275 (1): 221-226 8 Scotto KW, Biedler JL, Melera PW. Amplification and expression of genes associated with multidrug resistance in mammalian cells. Science 1986; 232: 751-755 9 Hsu SI, Lothstein L, Horwitz SB. Differential overexpression of 3 MDR gene family members in multidrug resistance J774.2 mouse cells. J Biol Chem 1989; 264 (20): 12053-12062 10 Chambers TC, McAvoy EM, Jacobs JW, Eilon G. Protein Kinase C phosphorylates P-Glycoprotein in Multidrug Resistant Human KB Carcinoma Cells. J Biol Chem 1990 265 (13): 7679-7686 11 Choi K, Chen C-J, Kriegler M, Roninson IB. An altered pattern of cross resistance in multidrug resistant cells results from spontaneous mutations in the MDR1 (P-glycoprotein) gene. Cell 1988; 53: 519-529 12 Marino PA, Gottesman MM, Pastan I. MDR gene regulation in rat liver. Cell Growth Differ 1990; 1: 57-62 13 Weinstein RS, Kuszak JR, Kluskensl F, Coon JS. P-glycoprotein in pathology: the multidrug resistance gene family in humans. Hum Pathol 1990; 21 (1): 34-^8 14 Fojo AJ, Ueda K, Slamon DJ, Poplack DG, Gottcsman, MM, Pastan I. Expression of a multidrug resistance gene in human rumours and tissues. Proc Natl Acad Sri USA 1987; 84: 265-269 15 Gordon-Cardo C, O'Brien JP, Casals D, et al. Multidrug resistance gene (P-glycoprotein) is expressed by endothelial cells at blood-brain barrier sites. Proc Natl Acad Sri USA 1989; 86: 65-68 16 Goldstein LJ, Galski H, Fojo A, et al. Expression of a multidrug resistance gene in human cancers. J Natl Cancer Inst 1989; 8 (2): 116-124
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
Understanding these mechanisms has been helpful in developing strategies to overcome resistance by blocking relevant pathways through drug design. It has also contributed to major understanding of the normal tissue protection against carcinogens and environmental toxins and may be relevant to individual cancer risk and familial cancers. In future profiles of these resistance mechanisms may help select therapy for patients.
DRUG RESISTANCE
195
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
17 Chan HSL, Thorner PS, Haddad G, Ling V. Immunohistochemical detection of P-glycoprotein: prognostic correlation in soft tissue sarcoma of childhood. J Clin Oncol 1990; 81 (4): 689-704 18 Lai S-L, Goldstein LJ, Gottesman MM, et al. MDR1 gene expression in lung cancer. J Nad Cancer Inst 1989; 81 (15): 1144-1150 19 Bell DR, Gerlach JH, Kartner N, Buick, RN, Ling V. Detection of P-glycoprotcin in ovarian cancer: a molecular marker associated with multidrug resistance. J Clin Oncol 1985; 3 (3): 311-315 20 Rothenberg ML, Mickley LA, Cole, DE, et al. Expression of the MDR1/P170 gene in patients with acute lymphoblastic leukaemia. Blood 1989; 74 (4): 13881395 21 Dalton WS, Grogan TM, Rybski JA, et al. Immunohistochemical detection and quantitation of P glycoprotein in multiple drug resistant human myeloma cells. Blood 1989; 73 (3): 747-752 22 Stewart DJ, Evans WK. Non-chemotherapeutic agents that potentiate chemotherapy efficiency. Cancer Treat Rev 1989; 16: 1-40 23 Dalton WS, Grogan JM, Meltzer PS, et al. Drug resistance in multiple myeloma and non-Hodgkin's lymphoma: detection of P-glycoprotein and potential circumvention by addition of verapamil in chemotherapy. J Clin Oncol 1989; 7: 415-424 24 Wang JC. DNA topoisomerases. Annu Rev Biochem 1985; 54: 664-697 25 Liu LF. DNA topoisomcrase poisons as antitumour drugs. Annu Rev Biochem 1989; 58: 351-375 26 Davies SM, Robson CN, Davies SL, Hickson ID. Nuclear topoisomerase II levels correlate with the sensitivity of mammalian cells to intercalating agents and cpipodophyllotoxins. J Biol Chem 1988; 263 (33): 17724-17729 27 Heck, MMS, Hittelman WN, Earnshaw WL. In vivo phosphorylation of the 170-kDa form of eukaryotic DNA topoisomerase II. J Biol Chem 1989; 264 (26): 15161-15164 28 Zwelling LA, Hinds M, Chan D, et al. Characterisation of an amsacrine resistant line of human leukaemia cells. J Biol Chem 1989; 264 (28): 16411-16420 29 Pommier Y, Kerrigan D, Schwartz R, et al. Altered DNA topoisomerase II activity in Chinese hamster cells resistant to topoisomerase II inhibitors. Cancer Res 1976; 46: 3075-3081 30 Fernandes DJ, Danks MK and Beck WT. Decreased nuclear matrix DNA topoisomerase II in human leukaemia cells resistant to VM-26 and mAMSA. Biochemistry 1990; 29: 4235-4241 31 Chung TDY, Drake FH, Tan KB, Per SR, Crooke ST, Mirabelli EK. Characterisation and immunological identification of cDNA clones encoding 2 human DNA topoisomerase II isozymes. Proc Nat] Acad Sci USA 1989; 86: 94319435 32 Woessner RD, Chung TDY, Hofmann GA et al. Differences between normal and ros-transformed NIH-3T3 cells in expression of The 170kD and 180kD forms of topoisomerase II. Cancer Res 1990; 50: 2901-2908 33 Holden JA, Rolfson DH, Winner CT. Human DNA topoisomerase II: evaluation of enzyme activity in normal and neoplastic tissues. Biochemistry 1990; 29: 2127-2134 34 Findlay GJ, Baguley BC, Snow K, Judd W. Multiple patterns of resistance of human leukemia cell sublines to amsacrine analogues. J Nad Cancer Inst 1990: 82 (8): 662-667 35 Pickett CB, Lu AYH. Glutathione S-transferases: gene structure, regulation and biological function. Annu Rev Biochem 1989; 58743-58764 36 Morrow CS, Cowan KH. Glutathione S-transferases and drug resistance. Cancer Cells 1990; (2) 1: 15-22 37 Nakagawa K, Saijo N, Tsuchida S, et al. Glutathione S-transferase pi as a
196
38
40 41 42 43 44
determinant of drug resistance in transfectant cell lines. J Biol Chem 1990; 265 (8): 4296-4301 Harris AL. DNA repair and resistance to chemotherapy. Cancer Surv, 1985; 4 (3): 601-624 Andrews PA, Howell SB. Cellular pharmacology of cisplatin: perspectives on mechanisms of acquired resistance. Cancer Cells 1990; (2) 2: 35-43 D'lncalci M, Sifti L, Tavema P, Catapano, CV. Importance of the DNA repair enzyme O 6 alkyl guanine alkyltransfcrase in cancer chemotherapy. Cancer Treat Rev 1988; 15: 279-292 Fry DW, Jackson RL. Membrane transport alterations of a mechanism of resistance to anticancer agents. Cancer Surv 1986; 5 (2): 48-79 Buchanan RL, Gralla JD. Cis-platin resistance and mechanisms in a viral test system. Biochemistry 1990; 29: 3436-3442 Lennard L, Lilleyman JS, Van Loon J, Weinshilboum RM. Genetic variation in response to 6-mercaptopurine for childhood Acute Lymphoblastic Leukemia. Lancet 1990 (336); 8709: 225-229 Stark GR, Deban'ssc M, Giulotto E, Wahl GM. Recent progress in understanding mechanisms of mammalian DNA amplification. Cell 1989; (57): 901-908
Downloaded from https://academic.oup.com/bmb/article-abstract/47/1/178/392184 by INSEAD user on 05 November 2018
39
THE CANCER CELL
