Many therapeutic treatments cause DNA damage and their overall utility is a balance between short-term therapeutic benefits, side effects, and longer-term hazards. Treatment-induced cancer can be a significant problem after some types of therapy. The DNA damage responses of human cells - including DNA repair - are a major determinant in the effectiveness of therapy and we are currently investigating the influence of DNA repair on the potential carcinogenicity of some types of therapy.
Skin cancer is the most frequently diagnosed malignancy. The UVB (wavelength 280-320nm) component of incident sunlight is an acknowledged contributor to skin carcinogenesis in the general population. The energy from UVB is directly absorbed by DNA bases and causes potentially mutagenic and cytotoxic photochemical DNA damage. Most (>95%) of the ultraviolet light that reaches the earth¿s surface is the longer wavelength UVA (320-400nm), however. This is poorly absorbed by DNA and consequently is less directly harmful. Because of the abundance of UVA in incident sunlight, any therapeutic treatment that renders DNA vulnerable to damage by UVA might be expected to increase the risk of skin cancer. Some drugs introduce UVA chromophores into DNA. These permit the absorption of energy from these longer wavelengths. The thiopurines (6-thioguanine (6-TG), 6-mercaptopurine (6-MP), and azathioprine) which have been used for several decades as anticancer and anti-inflammatory agents, provide an example of this. Azathioprine is a widely-prescribed immunosuppressant in organ transplant patients. This patient group, which requires life-long treatment with immunosuppressant drugs to ensure the survival of the engrafted organ, suffers a hugely elevated incidence of non-melanoma skin cancer in which sunlight exposure is a major contributory factor.
Thiopurines cause 6-TG, a purine analog, to accumulate in patients' DNA. We previously reported that DNA 6-TG absorbs UVA radiation and produces singlet oxygen (1O2), a form of reactive oxygen that is highly damaging to both DNA and proteins. Using cultured cells containing DNA 6-TG, we demonstrated that DNA 6-TG is an important target for damage by 1O2 and identified guanine sulfonate (GSO3) as a significant DNA photoproduct. We also showed that the PCNA DNA replication and repair factor is vulnerable to damage by 1O2 generated by the interaction of UVA with DNA 6-TG. These DNA photoproducts are refractory to excision by DNA repair enzymes and have profound inhibitory effects on replication and transcription.
We have continued to investigate the photochemical reactions of DNA 6-TG. In particular, analysis by Xiaolin Ren and Feng Li, (in collaboration with Yao-Zhong Xu, Open University) of the stoichiometry of the UVA-mediated destruction of 6-TG has identified a second photoproduct, guanine sulfinate (GSO2), as an intermediate in the oxidation of 6-TG to GSO3. The stepwise oxidation of DNA 6-TG to GSO3 is prevented by some free radical scavengers such as ascorbate although other antioxidant compounds form addition products with intermediates of 6-TG oxidation.
The photochemical vulnerability of DNA 6-TG reflects two things: its ability to absorb energy from UVA to generate 1O2 and its relatively low oxidation potential. The latter property makes 6-TG extremely vulnerable to chemical oxidation and Ilse Daehn has shown that cells containing DNA 6-TG are particularly sensitive to killing by oxidising chemicals and by reactive oxygen released by immune effector cells. 6-TG is incorporated freely into DNA of mitochondria. Mitochondria are potent sources of endogenous reactive oxygen species and Ilse has also shown that mitochondrial DNA of 6-TG treated cells rapidly accumulates high levels of GSO3. This is associated with mitochondrial malfunction and cell death and defines a novel mechanism by which the thiopurines may be cytotoxic.
The 6-TG/UVA-mediated damage to proteins has been further investigated. Azadeh Kia together with David Frith (2-D Gel Electrophoresis Laboratory, LRI Lincoln's Inn Fields), has examined changes in proteins induced by 6-TG/UVA treatment. In these experiments, altered proteins were identified by 2-D gel analysis of colour-derivatised, chromatin enriched fractions of control and 6-TG/UVA treated cells. Around 100 proteins were altered by a low UVA dose to 6- TG containing cells. So far, 60 of these have been identified by Maldi-Tof MS. The analysis confirmed the modification of PCNA and identified changes in the MSH2 DNA mismatch repair factor, the Ku80 DNA binding/repair protein, and the replication-associated Mcm proteins 3,7 & 6. The precise manner in which these proteins are modified is currently under investigation. In a parallel study, Quentin Gueranger has demonstrated extensive formation of covalent DNA protein crosslinks in cells treated with 6-TG and UVA.
Reactive oxygen species, including 1O2, cause DNA strand breaks which, in turn, trigger DNA damage responses. Reto Brem and Feng Li have been investigating the induction of DNA strand breaks by 6-TG/UVA treatment and the subsequent DNA damage responses (Figure 1). We previously showed that 6-TG/UVA induces DNA single strand breaks that are detectable by Comet assays. Curiously, these breaks are largely confined to the S phase of the cell cycle and little of no breakage is observed in G1 or G2 phase. Both the Chk2 and Chk1 proteins are phosphorylated indicating that the ATM and ATR-mediated DNA damage responses are triggered by the presence of photochemically-induced DNA lesions (Figure 1b). In agreement with the activation of ATM, DNA double strand breaks are detected in some cells. The distribution of breaks suggests that these breaks are also confined to cells in S phase. Feng has also shown that the 6-TG/UVAinduced DNA photoproducts are not excised. At high levels of photochemical DNA lesions, there is a reduction in the amount of the Chk1 protein that is recovered from cell extracts. This is consistent with continued turnover of Chk1 in the presence of irreparable DNA lesions although the possibility that DNA associated Chk1 may have become covalently crosslinked to DNA has not been excluded.
The vulnerability of DNA 6-TG to oxidation by reactive oxygen has clinical implications. The incidence of skin cancer is higher in patients who have been immunosuppressed with azathioprine combined with ciclosporin A. The accepted mode of action of ciclosporin is via its effect on calcineurin. Ciclosporin is also a source of reactive oxygen, however. Natalie Attard has shown that cells with defective double strand break rejoining capacity are particularly sensitive to killing by cyclosporine indicating that this drug causes significant DNA breakage. In view of the susceptibility of DNA 6-TG to conversion into irreparable DNA damage and DNA strand breaks by reactive oxygen, Natalie is investigating possible synergistic DNA damaging interactions between the two drugs.
Figure 1. 6-TG/UVA causes DNA double strand breaks and triggers DNA damage responses. a) MRC5VA cells grown in the presence of 6-TG were irradiated with 10kJ/m2 UVA or 5Gy ionising radiation. γH2AX foci indicating the presence of DNA double strand breaks were visualised by immunocytochemisty (Left Panels). DNA was stained with DAPI (Right Panels). 6-TG/UVA treatment induces γH2AX foci, but only in ≤50% of the cells. This contrasts to ionising radiation which causes breaks uniformly in all cells. b) MRC5VA cells grown in the presence of 6-TG were irradiated with 10kJ/m2 UVA and whole cell extracts were analysed by western blotting with the antibodies indicated. Both Chk1 and Chk2 are phosphorylated within 1h of UVA irradiation indicating activation of the ATR and ATM response pathways. At higher levels of photochemical DNA damage, the recovery of Chk1 protein is reduced - particularly at the later time.
For a list of refereed research papers, see Publications (in navigation on left).
