British Medical Bulletin Advance Access originally published online on February 2, 2007
British Medical Bulletin 2006 79-80(1):153-170; doi:10.1093/bmb/ldl020
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thiopurines, DNA damage, DNA repair and therapy-related cancer
Cancer Research UK London Research Institute, Clare Hall Laboratories, South Mimms, Herts EN6 3LD, UK
* Correspondence to: Peter Karran, Cancer Research UK London Research Institute, Clare Hall Laboratories, South Mimms, Herts EN6 3LD, UK. E-mail: peter.karran{at}cancer.org.uk
 |
Abstract |
|---|
Top
Abstract
IntroductionAims: The thiopurines, azathioprine, 6-mercaptopurine and 6-thioguanine are one of the success stories of chemotherapy. They are effective immunosuppressants and anti-cancer agents and are prescribed increasingly to treat inflammatory diseases. Although their metabolism has been studied in detail, the optimal use of thiopurines has been guided predominantly by clinical experience and the precise molecular events that underlie their therapeutic activity have remained unclear. The aim of this article is to review some of the properties of the thiopurines and relate them to possible therapeutic mechanisms. In particular, I consider the contribution that DNA substitution by 6-thioguanine makes to their effects as well as some of the possible harmful reactions that DNA 6-thioguanine might undergo.
Conclusions: The increased chemical reactivity of DNA 6-thioguanine underlies its cytotoxic effects and is an important contributor to the anti-leukaemic effects of the thiopurines. The same enhanced reactivity may contribute to the increased risk of acute myeloid leukaemia and skin cancer in thiopurine-treated organ transplant patients.
Keywords: Thiopurines • azathioprine • 6-thioguanine • 6-mercaptopurine • DNA mismatch repair • oxidative DNA damage • skin cancer • acute myeloid leukaemia
|
Introduction |
|---|
The thiopurines 6-mercaptopurine (6-MP), 6-thioguanine (6-TG), and azathioprine (Aza) (Fig. 1) are used as anti-cancer agents, in the treatment of inflammatory disorders, and as immunosuppressants. 6-MP and 6-TG were identified as growth inhibitory antimetabolites through the pioneering studies of Gertrude Elion and George Hitchins half a century ago (reviewed by Elion).1 They were soon introduced into clinical practice to treat acute leukaemia in children and their use has resulted in a dramatic increase in survival rates for this disease. Both drugs are also effective against colitis, psoriasis and rheumatoid arthritis; Aza, a modified 6-MP, is one of the standard immunosuppressants prescribed to prevent graft rejection in organ transplant patients.
|
Thiopurines are chemically more reactive than the normal DNA bases. This increased reactivity partly underlies their therapeutic effect but it also makes them susceptible to unwanted reactions and, as with many other powerful therapeutics, their long-term use is increasingly associated with malignancy. Here, I review some of the properties of thiopurines with particular emphasis on the mechanisms by which these agents might both cure and cause cancer.
|
Metabolism of thiopurines |
|---|
All three thiopurines are prodrugs. Their metabolism culminates in the formation of TG nucleotides that are precursors of nucleic acid synthesis. This metabolic activation has been reviewed in detail2 and is outlined here (Fig. 1). DeoxythioGTP (dthioGTP), the ultimate DNA precursor, is a good substrate for incorporation by DNA polymerases and estimates of steady-state levels of DNA 6-TG in patients suggest that during replication between 0.01 and 0.1% of DNA guanines (around 105 to 106 per cell) may be substituted by the thiobase.3,4 These low levels of DNA 6-TG fall below a threshold for toxicity (see below) and are tolerated. More extensive substitution is associated with chromosomal damage and cell death.
Thiopurines are also enzymatically inactivated. Xanthine oxidase—an important component of general purine catabolism—oxidizes 6-MP to 6-thiouric acid. A second, more important, pathway is initiated by thiopurine methyltransferase (TPMT) which converts 6-MP or 6-TG to inactive S-methyl derivatives. TPMT has received considerable attention as an example of the potential of pharmacogenomics to provide individualized therapy (for reviews, see Evans and co-workers).5,6 A number of polymorphic variants of the TPMT gene encode enzymes with significantly reduced activity. Approximately 0.3% of patients are homozygous for non-functional TPMT alleles and the consequent absence of TPMT activity is associated with severe, and potentially fatal, thiopurine-induced haematological toxicity. Pretreatment screening for inactive TPMT alleles can identify individuals at risk of severe bone marrow toxicity. Heterozygosity for a low activity allele identifies patients in whom lower doses of thiopurine can be used safely. Even normal TPMT activity does not abolish the risk of myelosuppression, however, and close monitoring of haematological parameters is still required. For this reason, whether knowledge of TPMT status provides significant practical benefits has been questioned.7
|
How do thiopurines kill cells? |
|---|
Despite five decades of clinical use, there is still uncertainty about the therapeutic mechanism of thiopurines. Their effectiveness as anti-leukaemic agents testifies to their ability to kill cells and an ability to inhibit the proliferation of activated T cells seems a likely contributor to immunosuppression. Early studies implicated thioguanine nucleotides (usually taken to mean the deoxy- and ribonucleoside triphosphates) and DNA substitution by 6-TG in cytotoxicity.8 Purine starvation—an alternative mechanism of toxicity that is independent of thioguanine nucleotide formation—is another possible contributor. This is because methylthioinosinemonophosphate (meTIMP) (Fig. 1) is a potent inhibitor of phosphoribosylpyrophosphate amidotransferase9,10—the first step of de novo purine biosynthesis. Since thioinosine monophosphate (TIMP) is not produced from 6-TG, purine starvation is a factor in the effects of 6-MP and its precursor Aza On the other hand, the effects of thioguanine nucleotide formation and DNA substitution are common to 6-MP, Aza and 6-TG. Although thioguanine nucleotide levels have been regarded as predictive for 6-MP cytotoxicity,11 there appears to be no current consensus as to the relative contributions of purine starvation and DNA substitution to cytotoxicity.
The problem of understanding how thiopurines are cytotoxic and immunosuppressive is compounded by the fact that the presence of 6-TG in DNA is not, in itself, sufficient for its toxic effects. Further modification of DNA 6-TG as well as an active DNA mismatch repair (MMR) system is also required. DNA 6-TG undergoes in situ non-enzymatic methylation—probably by S-adenosylmethionine. This reaction, which is favoured by the increased chemical reactivity of the thiol group, affects an estimated one in 104 or 105 incorporated 6-TG bases.12 Because 6-TG substitution of DNA is so extensive, however, there are several thousand 6-meTG bases in a cell's DNA at toxic 6-TG doses. DNA 6-meTG preferentially base pairs with thymine during replication. The 6-meTG:T base pairs resemble replication errors and provoke processing by MMR. This results in cell death. An outline of the principal events in MMR is shown in Figure 2A. Briefly, its role is to correct errors committed by DNA polymerases during replication. Incorrect base pairs or supernumerary bases that escape proofreading during replication are recognized and engaged by the MutS
mismatch binding complex which comprises one subunit each of the MSH2 and MSH6 proteins. Further interaction with a second heterodimer, MutL comprising MLH1 and PMS2 leads to recruitment of exonucleases and DNA polymerases that remove the mismatched DNA section and replace it with an accurate copy. A defect in any of these four key proteins is sufficient to inactivate MMR and prevent the recognition of aberrant base pairs as well as true replication errors. MutS or MutL defective cells are therefore very resistant to thiopurines because they are unable to initiate lethal processing. They survive with substantial levels of DNA 6-TG.13 The resistance conferred by defective MMR is known as methylation tolerance because the same defects also confer substantial resistance to methylating agents that produce DNA O6-methylguanine. This methylation product miscodes in a similar fashion during replication. O6-methylguanine:T base pairs closely resemble 6-meTG:T base pair (Fig. 2B) and their processing by MMR also triggers cell death.
|
Does purine starvation contribute to the toxicity of thiopurines? An experimentally induced increase in meTIMP in cultured human cells is associated with increased sensitivity to 6-MP. TPMT overexpression which increases the levels of inhibitory methylthiopurine nucleotides, actually reduces the supply of thioguanine nucleotides for incorporation into DNA. This can increase 6-MP sensitivity but decreases sensitivity to 6-TG.14,15 At face value, these findings might suggest that 6-MP and 6-TG cause death by different mechanisms; purine starvation being more important for 6-MP toxicity and DNA substitution the major contributor to killing by 6-TG. These observations appear to be incompatible with a general requirement for active MMR and DNA 6-TG in the toxicity of the thiopurines. Furthermore, clinical findings are more consistent with thioguanine nucleotide formation and DNA 6-TG, rather than purine starvation, as a major determinant of thiopurine cytotoxicity.
Can these apparent contradictions be resolved? With hindsight, it seems that in vitro studies with cultured cells may have exaggerated the contribution of purine starvation. Many involve cell lines in which later work has revealed MMR defects.9,10,14,16,17 In the absence of MMR, purine depletion is clearly the most significant factor in toxicity. When MMR is operating, however, processing of 6-meTG base pairs is the major contributor to cell death. This conclusion, which is fully consistent with the clinical effects of different TPMT levels, is supported by observations of the effects of MMR inactivation in mice in which an Msh2 defect effectively abolishes the myelotoxicity associated with 6-MP treatment.18 Overall, the data suggest the formation of thioguanine nucleotides and subsequent DNA damage represent the major mechanism of thiopurine toxicity. Inhibition of de novo purine biosynthesis appears to play a less significant role. It is, of course, possible that the overt cytotoxic effects and the immunosuppressive effects are mechanistically different and that purine starvation is a significant contributor to the latter. It is noteworthy in this regard that mycophenolate mofetil, one of the later generation immunosuppressants, is an inhibitor of purine biosynthesis.
|
DNA 6-TG methylation and DNA damage signalling |
|---|
Although DNA 6-TG is a non-canonical DNA base, it is not strictly DNA damage. The damage is produced by post-replicative processing and the intervention of MMR is required to initiate DNA damage responses and to trigger DNA damage-related cell cycle checkpoints. Signalling events downstream of MMR recognition of 6-meTG base pairs are being clarified. Prolonged 6-TG treatment induces a sustained cell cycle arrest in G2 phase.19 Arrest is correlated with sensitivity. It requires functional MMR and occurs in thiopurine-sensitive repair proficient cells. Unlike most DNA damage checkpoint activation that occurs within hours of treatment, 6-TG-induced G2 arrest is a very late event and is not maximal until day two or three of continuous drug treatment.19 The onset of arrest coincides with phosphorylation of the CHK1 and CHK2 checkpoint kinases although the timing appears to correlate better with the former.20 CHK1 is the major transducer of signals generated by the ATR DNA damage sensor that is activated preferentially by treatments that interfere with replication fork progression, such as exposure to hydroxyurea or DNA polymerase inhibitors (reviewed in Sancar et al.).21 There are sound reasons to believe that ATR/CHK1 signalling is required to impose the G2 arrest after 6-TG treatment. First, reducing the level of ATR or CHK1 by the technique of siRNA silencing abrogates 6-TG-induced G2 arrest. In addition, 6-TG induces G2 arrest in ATM defective cells indicating that the alternative pathway in which CHK2 transduces signals from the ATM DNA damage sensor is less important.20 What triggers ATR activation? Analysis of DNA breakage after 6-TG treatment suggests a role for DNA single strand breaks.19 Although DNA double-strand breaks are detectable within hours of 6-TG exposure, their appearance is not MMR-dependent and they disappear well before detectable CHK1 activation. They do not seem to contribute significantly to cell death. At later times (from 24 h) an excess of single strand breaks is observed in MMR-proficient cells. Their appearance coincides temporally with CHK1 activation and G2 arrest. Although this might suggest that single-strand breaks trigger checkpoint activation, the presence of double-strand breaks derived from replication of DNA bearing incompletely processed 6-meTG base pairs, is not excluded by these findings. In view of the low number of 6-meTG lesions, any contingent double-strand breaks would probably escape detection by the rather insensitive analysis.
DNA 6-meTG and the DNA O6meG introduced by treatment with methylating drugs are both processed by MMR. Low doses of methylating agents also induce a delayed MMR-dependent G2 arrest that occurs in the second cell cycle after treatment. Arrest is dependent on ATR and CHK122 and requires MMR-dependent processing of O6-meG base pairs. Thus, DNA 6-meTG and DNA O6meG ultimately trigger the ATR/CHK1 DNA damage response, albeit indirectly. The timing of checkpoint activation suggests that an inducing signal is generated late and involves replication of regions of abnormal DNA structure produced when MMR processes DNA containing 6-meTG or O6meG bases.
It follows from this that MMR proteins themselves are not components, in any meaningful sense, of the ATR/CHK1 G2 checkpoint response. MMR and the G2 checkpoint are best thought of as separate systems. The signalling cascade that results in G2 checkpoint activation (ATR/CHK1 activation, CDC25A destruction) is intact in MMR defective cells and can be triggered by appropriate treatments. Indeed, MMR inactivation and abrogation of the G2 checkpoint have contrasting effects. Whereas the former increases resistance to 6-TG, siRNA-mediated inhibition of CHK1 or ATR expression sensitizes cells to 6-TG-induced apoptosis.20 MMR and the G2 checkpoint system simply overlap and MMR-mediated processing of 6-meTG (and O6meG) is required to generate the DNA strand discontinuities or other aberrant DNA structures that trigger activation of the ATR cascade.
|
Oxidation of thiopurines |
|---|
The increased reactivity of the thiol group means that 6-TG is susceptible to other chemical modification as well as to methylation. In particular, DNA 6-TG is very easily oxidized. As long ago as 1961, DNA 6-TG was identified as an oxygen-dependent radiation sensitizer in bacteria.23,24 Later, photochemical studies of thiopurines in solution indicated that the free bases are photodynamically active and undergo light-induced reactions that require molecular oxygen.25,26 They are much more susceptible to oxidation than any of the normal DNA bases. Thiopurines have spectral properties that distinguish them from normal DNA bases. The absorbance maxima of adenine and guanine lie in the UVC range (around 260 nm). Replacement of the 6-oxygen atom by sulphur causes a shift in the UV absorbance spectrum and the thiopurines absorb maximally in the UVA region at around 340 nm . In solution both 6-MP and 6-TG are rapidly destroyed by UVA in reactions that are dependent on oxygen.25,27 In these reactions, thiopurines (TPs) can act as Type I and Type II photosensitizers. In a Type I reaction, free 6-MP or 6-TG absorbs UVA photons to generate an excited triplet state, 3(TP)*. This highly unstable compound can interact with molecular oxygen leading to the formation of thiopurine radical species (TPs.) and superoxide: 3(TP)* + O2 –>O2.-+ TP. + H+. Superoxide can be converted into highly reactive hydroxyl radicals (OH.) by Fenton-like reactions. In a Type II reaction, the UVA energy absorbed by the thiopurines is transferred to molecular oxygen (3O2) to generate singlet oxygen (1O2). Both OH. and 1O2 are known to cause DNA damage and the latter is considered to be the predominant source of DNA damage in cells exposed to incident solar UV radiation (>90% of which is UVA). DNA is an important target of these highly unstable reactive oxygen species (ROS). Potentially harmful DNA oxidation products are generated at significant levels during normal oxidative metabolism. Importantly, DNA 6-TG is a significant source of ROS in cells exposed to low UVA doses. Since ROS react very close to their site of formation, the risk of oxidative DNA damage is likely to be particularly high when cells containing DNA 6-TG are exposed to UVA.
6-TG is itself very susceptible to oxidation and is rapidly converted to guanine sulphonate (GSO3). Photochemical GSO3 formation by UVA involves 1O2 and the reaction is inhibited in the presence of azide which quenches 1O2, and accelerated in deuterium oxide which prolongs the reactive lifetime of 1O2. These reactions happen in DNA and moderate UVA doses convert DNA 6-TG to GSO3.28 Photochemical activation of cellular DNA 6-TG has significant biological consequences and incorporated 6-TG sensitizes cells to the lethal and mutagenic effects of very low UVA doses.27 GSO3 has dramatic effects on DNA stability. When placed opposite any of the normal DNA bases in short oligonucleotide duplexes, it is unable to form any kind of stable base pair and destabilizes the double helix more effectively than even a purine:purine or pyrimidine:pyrimidine pair.28 Consistent with a significant deviation from a normal structure, GSO3 cannot be bypassed by the Klenow fragment of Escherichia coli DNA polymerase I in primer extension reactions with synthetic substrates. Human DNA polymerase 
and an archeal Y-family DNA polymerase, both of which have a less stringent fidelity, are able to bypass DNA GSO3, however.27 Whether recruitment of reduced fidelity bypass DNA polymerases contributes to the mutagenicity of 6-TG/UVA treatment is unknown. Consistent with the ability of DNA GSO3 to block DNA polymerases, 6-TG/UVA treatment induces a rapid inhibition of replication in human cells.28
|
Thiopurines as carcinogens |
|---|
One consequence of long-term treatment with powerful drugs, many of which are acknowledged carcinogens, is an increased incidence of iatrogenic malignancy. Thiopurines are no exception. Azathioprine is designated as a human carcinogen by the International Association for Research on Cancer.29 Long-term immunosuppression in organ transplant patients is associated with an increased risk of certain types of cancer.30,31 Although several factors including immunosuppression per se, are likely to contribute to this, there is good reason to believe that Aza treatment may play a significant role.
|
Leukaemia |
|---|
The development of drug resistance often limits the success of chemotherapy. 6-MP is widely used in the treatment of acute lymphoblastic leukaemia (ALL) and the emergence of drug-resistant disease contributes to treatment failure in perhaps 25% of cases. The development of resistance is mirrored by results from laboratory studies in which chronic exposure of cultured cells to toxic thiopurine concentrations allows the outgrowth of rare clones with stable thiopurine resistance (Fig. 3A, B). These 6-TG and 6-MP tolerant cells are MMR defective and have the microsatellite instability that is a defining feature of this repair deficiency.32 The emergence of tolerant clones may have clinical relevance. Studies of MMR protein expression in leukaemic blasts reveal large interindividual variations with frequent undetectable levels.18,16 Even a 2-fold reduction in MMR capacity in lymphoblastoid cell lines heterozygous for a mutated MSH2 gene confers a measurable increase in methylating agent resistance.33 Perhaps, this applies to thiopurines. Heterozygous cells do not exhibit microsatellite instability, retain sufficient capacity to repair replication errors, but have a phenotype that is only evident on drug exposure. In addition, although microsatellite instability is rare in ALL, many established ALL cell lines are MMR-defective.16,17,34 Strikingly, several lines appear to have been established from patients at relapse. It has been pointed out16 that this is consistent with the clonal expansion of rare MMR-defective 6-MP-resistant leukaemic cells during the development of refractory disease. It is noteworthy in this regard that haematopoietic cells are particularly sensitive to thiopurines. In mouse models this sensitivity is considerably attenuated in MMR-deficient Msh2–/– animals. Thiopurine treatment would provide conditions favouring the selective proliferation of rare MMR-defective variant cells. An example of selective clonal expansion during the development of therapy-related acute myeloid leukaemia (AML) has been described.35
Thiopurine treatment is implicated in the development of AML. One study36 describes five cases of AML among 493 children who received 6-MP during maintenance therapy for ALL. Each of the affected patients had low TPMT activity and/or abnormally elevated thioguanine nucleotide levels. Thoracic organ and kidney recipients have respectively a 5.1-fold and 2.5-fold increased risk of AML (Fig. 3C). The extent of the increased risk is correlated with Aza dosage. MMR defects are generally rare in leukaemia and less than 5% of AML cases display microsatellite instability. The incidence of microsatellite instability in organ transplant-related tAMLs appears to be significantly higher.32 This is again consistent with selection of drug-resistant, MMR-defective myeloid cell clones. Repair defective cells are not, in themselves, malignant but their high rate of spontaneous mutation increases the probability of malignancy development. A high frequency of microsatellite instability has also been reported in lymphomas that arose in immunocompromised organ transplant patients37 although the possible contribution of immunosuppressive drugs was not evaluated. Overall, it appears that the relationship between thiopurine resistance and MMR deficiency that has been well-established in laboratory models may have a clinical counterpart.
|
|
Brain tumours |
|---|
An important study by Relling et al.38 highlighted a possible interaction between ionizing radiation and thiopurine treatment in a clinical setting. During a series of trials aimed at optimizing treatment regimes for childhood ALL, they noted a remarkable increase in the incidence of brain tumours among a group of patients who had received 6-MP together with prophylactic cranial radiation. A careful analysis of patient groups indicated that the increased risk was particularly associated with long-term 6-MP treatment and high TGN levels. Genetically determined TPMT deficiency was markedly over-represented among the affected individuals. The effects were so striking that subsequent treatment protocols were modified to avoid concurrent antimetabolite and radiation treatment. Although this dramatic synergy still remains unexplained, the susceptibility of thiopurines to oxidation is a possible contributor. As noted above, radiosensitizing effects were among the first properties ascribed to DNA 6-TG. Since 6-MP has a limited distribution in brain, and cell division is largely restricted to glial cells, the contribution of DNA 6-TG may be limited. Is it possible that TGN or other non-DNA thiopurine metabolites might act as a source of ROS in brain cells? In some circumstances, unincorporated thiopurines do promote the formation of oxidative DNA damage in bacteria.39 It would certainly be interesting to investigate the chemical and biological consequences of interactions between ionizing radiation thiopurine metabolites, or DNA 6-TG in more detail.
|
Skin cancer |
|---|
Immunosuppressed organ transplant patients have a significantly increased cancer risk. Not all cancers are over-represented, however, and the frequencies of some of the more common tumours are not greatly different to those of the normal population.30 Among those that are, many are associated with viruses. There is, however, a dramatic increased risk of non-melanoma skin cancer and the incidence of cutaneous squamous cell carcinoma (SCC) is more than 100-fold higher among organ recipients than in the normal population.30 Sunshine exposure is an important factor in the increased risk. The possible contribution of viruses to the immunosuppression-related increase in SSC risk has been considered but although some sub-types of human papillomavirus are found more frequently in these tumours, there is no clear evidence for a causative association.40
Most transplant patients have been treated with Aza and UVA is the major component (>90 %) of incident ultraviolet radiation. The longer wavelengths of UVA penetrate deeper into skin than the more energetic UVB and up to 20% of incident UVA may reach the layers of skin containing the stem cells. As pointed out in a preceding section, DNA 6-TG is highly susceptible to oxidation by UVA in a reaction that generates highly damaging ROS (Fig. 4). The product of DNA 6-TG oxidation, GSO3 and collateral damage to normal DNA bases from the ROS are both potential contributors to the development of transplant-related skin cancer.
|
The systemic and long-term administration of 6-MP or Aza favours the accumulation of DNA 6-TG by dividing cells. Peripheral blood lymphocytes of children undergoing 6-MP treatment for ALL3 and skin cells of transplant patients immunosuppressed with Aza contain measurable DNA 6-TG.27 The skin of Aza-treated patients is particularly sensitive to UVA. Erythema, the localized reddening of exposed skin is an acute response to sunlight. Aza treatment significantly reduces the minimal erythema dose—the lowest dose of UVA required to provoke perceptible erythema. The same patients' response to UVB to which DNA 6-TG is not reactive, is unaltered.27 Erythema is linked to unrepaired DNA damage. In cells without DNA 6-TG, UVB induces DNA lesions directly and UVB sunscreens that protect against DNA damage also attenuate erythema.41,42 Studies with Xeroderma pigmentosum patients who are unable to excise UVB photoproducts and genetically modified animals with similar DNA repair defects link erythema to the persistence of cyclobutane pyrimidine dimers—one of the most abundant DNA photoproducts of UVB–in transcriptionally active DNA.43 The reduced MED for UVA in Aza-treated patients implies the photochemical formation (and possible persistence) of DNA lesions via 6-TG. Consistent with this possibility, in laboratory experiments, 6-TG and UVA are synergistically cytotoxic and mutagenic to cultured cells.27
As outlined in a previous section, DNA 6-TG and UVA interact to generate ROS. These cause damage to DNA and other cellular molecules. Oxidation of DNA 6-TG to GSO3 and the reactions between ROS and normal DNA bases are all potential sources of mutation and cancer. DNA GSO3 is a powerful block to replication in vitro and is bypassed by low fidelity Y-family DNA polymerases. DNA replication in living cells is also rapidly inhibited by the combination of DNA 6-TG and UVA—consistent with the formation of similar photoproducts and possible triggering of error-prone tolerance strategies. Immunosuppression in transplant patients is effectively life-long, and long-term chronic exposure to UVA/DNA 6-TG might result in a sustained mutator phenotype that would accelerate the accumulation of mutations. Fig. 5 outlines how these photochemical reactions of DNA 6-TG might contribute to mutation and the subsequent development of skin cancer.
|
|
Concluding discussion |
|---|
The biological effects of thiopurines—both desirable and undesirable—reflect the increased reactivity conferred by substitution of the single oxygen atom of guanine by sulphur. The lethal, immunosuppressive and anti-inflammatory properties derive, at least in part, from incorporation into DNA and they favoured in situ methylation of DNA 6-TG. This step is essential to trigger the DNA MMR system into generating the ultimately cytotoxic DNA damage. Loss of MMR provides sufficient selective growth advantage and this may account for the development of resistant disease as well as the association of MMR defects with thiopurine treatment-related AML. The biological implications of the increased susceptibility of 6-TG to oxidation remain to be determined. In particular, photochemical formation within DNA itself, of the highly damaging ROS that are implicated in mutagenesis and cancer appears to be a likely hazard. Although ROS are an unavoidable consequence of an oxygen-based metabolism, cells have numerous protective systems that counteract their genotoxicity. Treatments that tip the delicate balance towards an increased burden of oxidative DNA lesions are likely to have adverse long-term consequences. Aza in combination with UVA in transplant patients is likely to produce conditions of chronic oxidative stress. The idea that Aza might contribute to carcinogenesis in immunosuppressed patients is not new. It was suggested many years ago based on the observation that transplant patients with skin cancer had high levels of thioguanine nucleotides.44 More recently, immunosuppression per se, rather than any particular drug treatment has generally been regarded as the major contributor.45 It is perhaps worth revisiting the epidemiological findings now that there have been several more years of follow-up. There is no doubt that skin cancer remains a huge problem for transplant patients. It seems prudent to at least reconsider a possible role for Aza and the photochemical properties of DNA-6-TG provide a theoretical framework.
Ironically, one obstacle to resolving the separate contributions that overall immunosuppresion and Aza make to transplant-associated cancer may soon disappear. Aza is being phased out in favour of newer immunosuppressants like mycophenolate mofetil or sirolimus. These are generally considered to be more effective although this has been questioned.46 In time, this will provide a cohort of transplant patients who have undergone many years of immunosuppression without Aza. It will be interesting to review the effect of these changes on transplant-related SCC. Although it is generally considered that more powerful immunosuppression will increase the frequency of transplant-related cancers, this may not prove to be the case if Aza plays a significant role. At the same time, it might be worth keeping a close watch on the increasing number of patients who now receive Aza treatment for chronic inflammatory disorders.
|
Acknowledgements |
|---|
I am indebted to the past and current members of my laboratory, and in particular Drs Judith Offman, Peter O'Donovan and Conal Perrett, on whose work many of these opinions are based. I am also extremely grateful to Professor Peter Swann, Dr Yao-Zhong Xu, Professor Gerhard Opelz, Drs Catherine Harwood, Jane McGregor, Natalie Attard and Margherita Bignami all of those who played important roles in shaping these ideas.
Accepted for publication December 21, 2006.
|
References |
|---|
- Elion GB. The purine path to chemotherapy. Science (1989) 244:41–47.
[Abstract/Free Full Text] - Aarbakke J, Janka-Schaub G, Elion GB. Thiopurine biology and pharmacology. Trends Pharmacol Sci (1997) 18:3–8.[Medline]
- Warren DJ, Andersen A, Slordal L. Quantitation of 6-thioguanine residues in peripheral blood leukocyte DNA obtained from patients receiving 6-mercaptopurine-based maintenance therapy. Cancer Res (1995) 55:1670–1674.
[Abstract/Free Full Text] - Cuffari C, Li DY, Mahoney J, Barnes JY, Bayless TM. Peripheral blood mononuclear cell DNA 6-thioguanine metabolite levels correlate with decreased interferon-g production in patients with Crohn's disease on AZA therapy. Dig Dis Sci (2004) 49:133–137.[Medline]
- Evans WE, Relling MV. Pharmacogenomics: translating functional genomics into rational therapeutics. Science (1999) 286:487–491.
[Abstract/Free Full Text] - Krynetski E, Evans WE. Drug methylation in cancer therapy: lessons from the TPMT polymorphism. Oncogene (2003) 22:7403–7413.[Medline]
- Marshall E. Preventing toxicity with a gene test. Science (2003) 302:588–590.
[Abstract/Free Full Text] - Tidd DM, Paterson AR. A biochemical mechanism for the delayed cytotoxic reaction of 6-mercaptopurine. Cancer Res (1974) 34:738–746.
[Abstract/Free Full Text] - Bokkerink JPM, Stet EH, dAbreu RA, et al. 6-mercaptopurine: cytotoxicity and biochemical pharmacology in human malignant T-lymphoblasts. Biochem Pharmacol (1993) 45:1455–1463.[Medline]
- Stet EH, Abreu RAd, Bokkerink JPM, et al. Reversal of 6-mercaptopurine and 6-mercaptopurine ribonucleoside cytotoxicity by amidoimidazole carboxamide ribonucleoside in Molt4 human malignant T-lymphoblasts. Biochem Pharmacol (1993) 46:547–550.[Medline]
- Lennard L, Rees CA, Lilleyman JS, Maddocks JL. Childhood leukemia: a relationship between intracellular 6-mercaptopurine metabolites and neutropenia. Br J Clin Pharmacol (1983) 16:359–363.[Medline]
- Swann PF, Waters TR, Moulton DC, Xu Y-Z, Edwards M, Mace R. Role of postreplicative DNA mismatch repair in the cytotoxic action of thioguanine. Science (1996) 273:1109–1111.[Abstract]
- Karran P, Offman J, Bignami M. Human mismatch repair, drug-induced DNA damage, and secondary cancer. Biochimie (2003) 85:1149–1160.[Medline]
- Dervieux T, Blanco JG, Krynetski EJ, Vanin EF, Roussel MF, Relling MV. Differing contribution of thiopurine methyltransferase to mercaptopurine versus thioguanine effects in human leukemic cells. Cancer Res (2001) 61:5810–5816.
[Abstract/Free Full Text] - Coulthard SA, Hogarth LA, Little M, et al. The effect of thiopurine methyltransferase expression on sensitivity to thiopurine drugs. Mol Pharmacol (2002) 62:102–109.
[Abstract/Free Full Text] - Matheson EC, Hall AG. Assessment of mismatch repair function in leukaemic cell lines and blasts from children with acute lymphoblastic leukaemia. Carcinogenesis (2003) 24:31–38.
[Abstract/Free Full Text] - Taverna P, Liu L, Hanson AJ, Monks A, Gerson SL. Characterization of MLH1 and MSH2 DNA mismatch repair proteins in cell lines of the NCI anticancer drug screen. Cancer Chemother Pharmacol (2000) 46:507–516.[Medline]
- Krynetskaia NF, Brenner TL, Krynetski EY, et al. Msh2 deficiency attenuates but does not abolish thiopurine hematopoietic toxicity in msh2-/- mice. Mol Pharmacol (2003) 64:456–465.
[Abstract/Free Full Text] - Yan T, Berry SE, Desai AB, Kinsella TJ. DNA mismatch repair (MMR) mediates 6-thioguanine genotoxicity by introducing single-strand breaks to signal a G2-M arrest in mismatch repair-proficient RKO cells. Clin Cancer Res (2003) 9:2327–2334.
[Abstract/Free Full Text] - Yamane K, Taylor K, Kinsella TJ. mismatch repair-mediated G2/M arrest by 6-thioguanine involves the ATR-Chk1 pathway. Biochem Biophys Res Commun (2004) 318:297–302.[Medline]
- Sancar A, Lindsey-Boltz LA, Unsal-Kacmaz K, Linn S. Molecular mechanisms of DNA repair and the DNA damage checkpoints. Ann Rev Biochem (2004) 73:39–85.[Medline]
- Stojic L, Mojas N, Cejka P, et al. Mismatch repair-dependent G2 checkpoint induced by low doses of SN1 type methylating agents requires the ATR kinase. Genes Dev (2004) 18:1331–1344.
[Abstract/Free Full Text] - Kaplan HS, Smith KC, Tomlin P. Radiosensitization of E. coli by purine and pyrimidine analogues incorporated into deoxyribonucleic acid. Nature (1961) 190:794–796.[Medline]
- Kaplan HS, Zavarine R, Earle J. Interaction of the oxygen effect and radiosensitization produced by base analogues incorporated into deoxyribonucleic acid. Nature (1962) 194:662–664.[Medline]
- Hemmens VJ, Moore DE. Photo-oxidation of 6-mercaptopurine in aqueous solution. Chem Soc Perkin Trans (1984) II:209–211.
- Hemmens VJ, Moore DE. Photochemical sensitization by azathioprine and its metabolites-II. Azathioprine and nitroimidazole metabolites. Photochem Photobiol (1986) 43:257–262.[Medline]
- O'Donovan P, Perrett C, Zhang X, et al. Azathioprine and UVA light generate mutagenic oxidative DNA damage. Science (2005) 309:1871–1874.
[Abstract/Free Full Text] - Zhang X, Jeffs G, Ren X, et al. Novel DNA lesions generated by the interaction between therapeutic thiopurines and UVA light. DNA Repair. (2006) epublished ahead of print.
- Azathioprine. IARC Monographs. (1987) 119. Supplement 7.
- Penn I. Incidence and treatment of neoplasia after transplantation. J Heart Lung Transplant (1993) 12:S328–S336.[Medline]
- Penn I. Post-transplant malignancy: the role of immunosuppression. Drug Safety (2000) 23:101–113.[Medline]
- Offman J, Opelz G, Doehler B, et al. Defective DNA mismatch repair in acute myeloid leukemia/myelodysplastic syndrome after organ transplantation. Blood (2004) 104:822–828.
[Abstract/Free Full Text] - Marra G, D'Atri S, Corti C, et al. Tolerance of human MSH2+/– lymphoblastoid cells to the methylating agent temozolomide. Proc Natl Acad Sci USA (2001) 98:7164–7169.
[Abstract/Free Full Text] - Molenaar JJ, Gerard B, Chambon-Pautas C, et al. Microsatellite instability and frameshift mutations in BAX and transforming growth factor b RII genes are very uncommon in acute lymphoblastic leukemia in vivo but not in cell lines. Blood (1998) 92:230–233.
[Abstract/Free Full Text] - Blanco JG, Dervieux T, Edick MJ, et al. Molecular emergence of acute myeloid leukemia during treatment for acute lymphoblastoc leukemia. Proc Natl Acad Sci USA (2001) 98:10338–10343.
[Abstract/Free Full Text] - Thomsen JB, Schroder H, Kristinsson J, et al. Possible carcinogenic effect of 6-mercaptopurine on bone marrow stem cells. Cancer (1999) 86:1080–1086.[Medline]
- Larson RS. Microsatellite instability in natural killer cell-like T cell lymphomas in immunocompromised and immunocompetent individuals. Blood (1997) 89:114–115.
- Relling MV, Rubnitz JE, Rivera GK, et al. High incidence of secondary brain tumours after radiotherapy and antimetabolites. Lancet (1999) 354:34–39.[Medline]
- Komeda K, Iwamoto S, Kominami S, Ohnishi T. Induction of cell killing, mutation and umu gene expression by 6-mercaptopurine or 2-thiouracil with UVA irradiation. Photochem Photobiol (1997) 65:115–118.[Medline]
- Purdie KJ, Surentheran T, Sterling JC, et al. Human papillomavirus gene expression in cutaneous squamous cell carcinomas from immunosuppressed and immunocompetent individuals. J Invest Dermatol (2005) 125:98–107.[Medline]
- Young AR, Sheehan JM, Chadwick CA, Potten CS. Protection by ultraviolet A and B sunscreens against in situ dipyrimidine photolesions in human epidermis is comparable to protection against sunburn. J Invest Dermatol (2000) 115:37–41.[Medline]
- Liardet S, Scaletta C, Panizzon R, Hohlfeld P, Laurent-Applegate L. Protection against pyrimidine dimers, p53, and 8-hydroxy-2'-deoxyguanosine expression in ultraviolet-irradiated human skin by sunscreens: difference between UVB + UVA and UVB alone sunscreens. J Invest Dermatol (2001) 117:1437–1441.[Medline]
- Berg RJ, Ruven HJ, Sands AT, deGruijl FR, Mullenders LH. Defective global genomic repair in XPC mice is associated with skin cancer susceptibility but not with sensitivity to UVB induced erythema and edema. J Invest Dermatol (1998) 110:405–409.[Medline]
- Lennard L, Thomas S, Harrington CJ, Maddocks JL. Skin cancer in renal transplant recipients is associated with increased concentrations of 6-thioguanine nucleotide in red blood cells. Br J Dermatol (1985) 113:723–729.[Medline]
- Bavinck JNB, Hardie DR, Green A, et al. The risk of skin cancer in renal transplant recipients in Queensland, Australia. A follow-up study. Transplantation (1996) 61:715–721.[Medline]
- Remuzzi G, Lesti M, Gotti E, et al. Mycophenolate mofetil versus azathioprine for prevention of acute rejection in renal transplantation (MYSS): a randomised trial. Lancet (2004) 364:503–512.[Medline]
- Waters TR, Swann PF. Cytotoxic mechanism of 6-thioguanine: hMutSalpha, the human mismatch binding heterodimer, binds to DNA containing S6-methylthioguanine. Biochemistry (1997) 36:2501–2506.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||