Decreased transcription-coupled nucleotide excision repair capacity is associated with increased p53- and MLH1-independent apoptosis in response to cisplatin
- 4.1k Downloads
One of the most commonly used classes of anti-cancer drugs presently in clinical practice is the platinum-based drugs, including cisplatin. The efficacy of cisplatin therapy is often limited by the emergence of resistant tumours following treatment. Cisplatin resistance is multi-factorial but can be associated with increased DNA repair capacity, mutations in p53 or loss of DNA mismatch repair capacity.
RNA interference (RNAi) was used to reduce the transcription-coupled nucleotide excision repair (TC-NER) capacity of several prostate and colorectal carcinoma cell lines with specific defects in p53 and/or DNA mismatch repair. The effect of small inhibitory RNAs designed to target the CSB (Cockayne syndrome group B) transcript on TC-NER and the sensitivity of cells to cisplatin-induced apoptosis was determined.
These prostate and colon cancer cell lines were initially TC-NER proficient and RNAi against CSB significantly reduced their DNA repair capacity. Decreased TC-NER capacity was associated with an increase in the sensitivity of tumour cells to cisplatin-induced apoptosis, even in p53 null and DNA mismatch repair-deficient cell lines.
The present work indicates that CSB and TC-NER play a prominent role in determining the sensitivity of tumour cells to cisplatin even in the absence of p53 and DNA mismatch repair. These results further suggest that CSB represents a potential target for cancer therapy that may be important to overcome resistance to cisplatin in the clinic.
KeywordsHCT116 Cell Nucleotide Excision Repair Prostate Cancer Cell Line DU145 Cell Xeroderma Pigmentosum
Cisplatin [cis-diammine-dichloroplatinum (II)] has been used in the treatment of neoplastic diseases for over 30 years . The effectiveness of cisplatin is dependent on its interaction with DNA. This drug forms a variety of DNA adducts but more than 99% of these are intrastrand DNA adducts, most often between adjacent purines, with only a very small number of interstrand-crosslinks and monoadducts . These intrastrand lesions are repaired by the nucleotide excision repair (NER) pathway so the response of tumours to cisplatin and other platinum-based drugs may be affected by nucleotide excision repair capacity of the tumour cells .
The vast majority of what is known about NER stems from studies using the model DNA damaging agent UV light but NER of cisplatin-induced DNA adducts is thought to occur through an identical mechanism [3, 4, 5, 6, 7, 8]. The rate-limiting step in NER is lesion recognition and this occurs through two distinct mechanisms yielding two interrelated yet genetically separable subpathways of NER . Global-genomic nucleotide excision repair (GG-NER) is responsible for the removal of the vast majority of UV and cisplatin-lesions throughout the genome whereas transcription-coupled nucleotide excision repair (TC-NER) is responsible for the selective removal of only those lesions that are present in the template strand of expressed genes [3, 5, 10, 11, 12].
Cockayne syndrome (CS) and xeroderma pigmentosum (XP) are heterogeneous disorders characterized by clinical photosensitivity . Based on cell fusion and complementation studies, patients with CS were classified into two groups (CS groups A and B) whereas XP patients were grouped into 8 groups (XP groups A through G and V). CS and XP cells (with the exception of the variant form, XP-V) have defects in NER. These defects can be specific to TC-NER, GG-NER or both sub-pathways of NER [14, 15]. Therefore, fibroblasts derived from patients with these UV sensitive syndromes have been instrumental in identifying proteins involved specifically in TC-NER and GG-NER and provided a model system to study the relative contribution of GG-NER and TC-NER to cisplatin response [6, 16, 17, 18].
All TC-NER-deficient (CS-B for example) and completely NER-deficient (XP-A for example) fibroblasts were found to be exquisitely sensitive to apoptosis induced by UV light and cisplatin [6, 16, 17, 18]. Many genetic alterations in cancer affect DNA damage responses. Notably, loss of the p53 tumour suppressor protein and DNA mismatch repair (MMR) proteins are among the most common genetic alterations in cancer and these alterations have been associated with resistance to cisplatin treatment [19, 20, 21, 22, 23, 24, 25, 26, 27]. Therefore, the effect of targeting TC-NER on the acute response of tumour cells to cisplatin could not be accurately predicted.
Here we report that silencing CSB by RNA interference (RNAi) reduced the TC-NER capacity of several prostate and colon cancer cell lines. This repair defect was associated with increased sensitivity of CSB-targeted cells to cisplatin-induced apoptosis. Importantly, the sensitivity of p53- and/or MMR-deficient tumour cells to cisplatin-induced apoptosis could be significantly increased by silencing CSB. These results suggest that TC-NER plays a major role in determining the sensitivity of these tumour cells to cisplatin and further suggests that CSB represents a potential therapeutic target for cancer therapy.
Cell Culture and UV-irradiation
HCT116, DU145 and PC-3 cells were obtained from the American Tissue Type Collection (Camden, NJ). The MLH1-corrected (HCT116 + chr3) and p53 nullizygous (HCT116p53-/-) cell lines were described previously [28, 29]. HCT116 derived cells were cultured in McCoy's media (Wisent, St. Bruno, QC) while DU145 and PC-3 cells were grown in DMEM (Hyclone, Logan, UT). McCoy's and DMEM were supplemented with 10% fetal bovine serum (Wisent, St. Bruno, QC). Where indicated, cisplatin (Mayne Pharma Canada Inc., Montreal, QC) was added to fresh, pre-warmed media at the indicated final concentration.
To UV-irradiate cells, medium was removed and cells were irradiated with the indicated dose using a germicidal bulb emitting predominantly at 254 nm at 1 J/m2/s as measured with a hand-held UV dosimeter (UVX Radiometer, UVP Inc., Uplands, CA). Fresh, pre-warmed media was replaced and dishes were returned to an incubator for the indicated period of time.
Sub-confluent cells were transfected with the indicated siRNA (Dharmacon, Lafayette, CO) using OptiMEM II and Oligofectamine (Invitrogen, Burlington, ON). The target sequences for CSB and XPA were GTGTGCATGTGTCTTACGA and AGAATTGCGGCGAGCAGTA, respectively. These RNA duplexes were used at a final concentration of 50 nM. A non-targeting control siRNA (TAGCGACTAAACACATCAA) was used as a negative control.
Preparation of nuclear lysates
Cells were rinsed with phosphate buffered saline (PBS) then trypsinized and collected by centrifugation. Cell pellets were resuspended in nuclear extraction buffer (320 mM sucrose, 10 mM HEPES, 5 mM MgCl2, 1% triton-X-100, pH 7.4), incubated on ice and then collected by centrifugation at 2500 × g. The resulting pellets were rinsed twice with nuclear wash buffer (320 mM sucrose, 10 mM HEPES, 5 mM MgCl2, pH 7.4) and collected by centrifugation at 2500 × g then resuspended in RIPA buffer (50 mM Tris-HCl pH6.8, 150 mM NaCl, 1 mM EDTA, 1% Triton-X-100, 1% sodium deoxycholate, 0.1% SDS). Pellets were disrupted using a sonicator equipped with a chilled microtip (Thermo Fisher Scientific, Ottawa, ON) and protein quantified using the Bradford assay (Bio-Rad, Mississauga, ON).
Two hundred micrograms of nuclear protein per well was subjected to gel electrophoresis using NuPAGE 3-8% gradient polyacrlyamide gels (Invitrogen), to visualize CSB protein whereas MLH1 and Ku86 were resolved using NuPAGE 4-12% gradient polyacrylamide gels (Invitrogen). Proteins were transferred to Hybond-C nitrocellulose (GE Healthcare, Baie d'Urfé, QC) and blots were stained with Ponceau S Red (5 mg/ml Ponceau S Red, 2% glacial acetic acid) to visualize total transferred proteins. Blots were then blocked in PBSMT-A (PBS, 5% nonfat milk powder, 0.05% Tween 20) proteins were detected using antibodies against XPA (FL-273), CSB (E-18) and Ku86 (M-20) (Santa Cruz Biotech, Santa Cruz, CA) and against MLH-1 (clone G168-15, BD Biosciences, Mississauga, ON) diluted in PBSMT-B (PBS, 0.5% nonfat milk powder) and were visualized using SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific) in combination with X-ray film (Kodak, Rochester, NY). Multiple proteins were detected using the same blots using Restore Western Blot Stripping Buffer (Thermo Fisher Scientific).
The recovery of RNA synthesis
Sub-confluent cells were transfected with siRNA and subsequently grown in media supplemented with 5 μCi/mL of [14C] thymidine (GE Healthcare) to uniformly label DNA. Seventy-two hours later, cells were treated with 10 J/m2 UV. One hour before indicated collection time, the media was replaced with fresh media containing 50 μCi/mL [3H] uridine (GE Healthcare) to label nascent RNA. Samples were rinsed in PBS containing 0.2% sodium azide (PBS-Z), collected by trypsin in PBS-Z, rinsed with PBS-Z and cell pellets were stored at -80°C. Samples were lysed in 1% SDS and nucleic acids were precipitated in 10% trichloroacetic acid (TCA)/0.1 M sodium pyrophosphate (NaPPi) and precipitated nucleic acids were collected on glass fiber filters (Thermo Fisher Scientific). Incorporation of [3H] and [14C] was determined using a scintillation counter and [3H] counts were normalized to [14C] counts to control for cell number. RNA synthesis is expressed as the proportion of [3H] uridine incorporated in UV-treated samples compared to unirradiated controls.
Host cell reactivation
Recombinant adenovirus expressing the lacZ gene under control of the murine cytomegalovirus promoter , was suspended in a minimal volume of PBS and was subsequently irradiated with the indicated dose of UV light on ice, as previously described . Cells were infected with UV- or mock-treated AdlacZ at a multiplicity of infection of 5 plaque forming units per cell. Forty-eight hours following infection, media was removed and monolayers were incubated with 1 mM chlorophenolred-β-D-galactopyranoside (Fluka Biochemika, Buchs, Switzerland) in 0.01% Triton X-100, 1 mM MgCl2, and 100 mM phosphate buffer (pH 8.3) . Absorbance at 570 nm was determined using a Thermo Multiskan Ascent microplate photometer (Thermo Fischer Scientific). β-galatosidase activity from the indicated dose is expressed relative to the activity obtained by infection with un-irradiated virus.
Cells were treated with the indicated dose of UV light or cisplatin, 72 hours following transfection of the indicated siRNA. Detached and adherent cells were collected 48 hours following treatment, fixed in 70% ethanol for a minimum of 2 hours at -20°C and stained in 30 μM propidium iodide (Sigma-Aldrich, Oakville, ON) in PBS with 40 μg/mL of RNAse A (Sigma-Aldrich) . Samples were analyzed by fluorescence activated cell sorting using a Becton Dickenson LSR II Facstation and CellQuest software (Becton Dickinson, Franklin Lakes, NJ) and data files were analyzed using FCS Express (De Novo Software, Los Angeles, CA). Apoptosis was quantified as the proportion of cells with subdiploid DNA content.
Caspase activity assays
Cells were transfected with the indicated siRNA and treated with either UV light or cisplatin. Twenty four hours following treatment, cells were collected with trypsin, rinsed thoroughly with PBS and cell number was determined using an automated cell counter (Vi-Cell XR, Beckman Coulter). Caspase 3, 8 and 9 activities were determined from 1 × 106 cells using ApoAlert caspase-3, caspase-8 or caspase-9/6 Fluorescent Assay Kits as specified by the manufacturer (Clontech, Mountain View, CA). Fluorescence was measured using a Thermo Fluoroskan Ascent microplate fluorometer (Thermo Fisher Scientific).
RNA interference against CSB increased the sensitivity of prostate cancer cells to cisplatin-induced apoptosis
The expression of CSB was similarly inhibited by RNAi and the sensitivity of PC-3 and DU145 cells to cisplatin-induced apoptosis was assessed. Silencing CSB led to a large increase in the sensitivity of DU145 cells to cisplatin-induced apoptosis (Figure 1D). The increased sensitivity of PC-3 cells was evident albeit less pronounced (Figure 1D). These results indicate that the ability of these prostate cancer cell lines to repair transcription-blocking DNA lesions by TC-NER was significantly reduced when CSB levels were decreased by RNAi. Importantly, these results suggest that disruption of CSB may be beneficial in combination with platinum-based drugs in the management of prostate cancer.
Role of p53 and MLH1 in determining cisplatin response of CSB-targeted colon cancer cells
Tumour cells frequently express mutant forms of DNA damage response proteins that may influence their sensitivity to therapeutic agents like cisplatin. For example, loss of p53 and MMR have been associated with resistance to cisplatin [19, 20, 21, 22, 23, 25]. Notably, DU145 cells carry point mutations in both alleles of p53 and do not express MLH1 so they are MMR- and p53-deficient [39, 40]. Similarly, PC-3 cells do not express detectable p53 and they exhibit microsatellite instability indicative of a defect in MMR [39, 40]. Given the effect of targeting CSB in these p53 and MMR-deficient prostate cancer cell lines, an isogenic series of cell lines derived from HCT116 colorectal cancer cells were used to better assess the impact of p53- and MMR-deficiency under conditions in which CSB is silenced by RNAi.
HCT116 were transfected with control and anti-CSB siRNAs and their sensitivity to cisplatin-induced apoptosis was assessed. Consistent with the results in prostate cancer cell lines, targeting CSB in HCT116 cells increased their sensitivity to cisplatin-induced apoptosis (Figure 2D) and this was associated with increased caspase activity (Figures 2E-G). Taken together, silencing CSB by RNAi reduced the capacity of these MMR-deficient cells to repair transcription-blocking DNA lesions and greatly increased their sensitivity to cisplatin-induced apoptosis.
There was no significant increase in the proportion of control siRNA transfected HCT116p53-/- cells undergoing apoptosis following exposure to up to 15 μM of cisplatin indicating that these cells were relatively resistant to cisplatin-induced apoptosis compared to isogenic controls (Figure 4D). This is likely due to the disruption of the well-described pro-apoptotic activity of p53 in these nullizygous cells . However, decreased expression of CSB was again associated with a significant increase in the sensitivity of tumour cells to cisplatin-induced apoptosis (Figure 4D). Apoptosis was associated with a significant increase in the activity of caspases 3, 8 and 9 (Figure 4E-G). So, while p53 nullizygous cells were less sensitive to cisplatin-induced apoptosis, targeting TC-NER was similarly effective at sensitizing HCT116p53-/- cells to cisplatin-induced cell death. Clearly, p53 was not absolutely required to sensitize tumour cells to cisplatin when CSB expression was decreased by RNAi.
RNA interference against XPA sensitizes cancer cells to cisplatin
Link between nucleotide excision repair and sensitivity to cisplatin
The contribution of TC-NER to cell survival following UV-irradiation or cisplatin treatment is thought to be cell type specific. For example, CS-B fibroblasts have a specific defect in TC-NER and are more sensitive to UV- and cisplatin-induced apoptosis than GG-NER deficient XP-C fibroblasts [6, 16, 17, 18]. A similar relationship was reported in CSB nullizygous mouse embryonic fibroblasts and murine keratinocytes exposed to UV light [7, 16, 17, 47, 48, 49, 50, 51, 52, 53]. By contrast, CSB null murine embryonic stem cells were not hypersensitive to this DNA damaging agent . The cell type-specific roles for TC-NER in determining the fate of UV-irradiated cells raised the question as to whether the response of tumour cells to UV light and cisplatin would more closely resemble that of primary cells [7, 16, 17, 18, 33, 47, 48, 49, 54, 55] or embryonic stem cells .
The CSB protein is required for a rate limiting step of TC-NER [56, 57, 58]. Here we show that a series of prostate and colon cancer cell lines retained the ability to repair transcription-blocking DNA lesions and that silencing CSB functionally impaired this repair process. Therefore, we were able to test the role of TC-NER in cisplatin response using this in vitro system. Decreased expression of CSB was associated with increased sensitivity of cells to cisplatin-induced apoptosis. Similarly, silencing CSB led to increased apoptosis following exposure to UV light (data not shown), consistent with a role for CSB in determining the sensitivity of cells to apoptosis induced by both of these agents [6, 16, 17, 18, 50, 59, 60]. Lu and coworkers reported that antisense oligonucleotides designed to inhibit the expression of CSB increased the responsiveness of wildtype p53 and DNA mismatch repair proficient ovarian tumour xenografts to cisplatin . Targeting TC-NER may represent a means of increasing the responsiveness of a variety of tumours to cisplatin.
MMR proteins and p53 in cisplatin resistance
Cisplatin is among the most widely used anti-neoplastic agents. Although tumours often respond favorably to cisplatin treatment, there is eventual disease progression in many cases. This recurrent disease can be refractory to subsequent treatment with platinum-based drugs [1, 62, 63, 64, 65, 66, 67, 68, 69]. Drug resistance can in some cases be attributed to increased DNA repair response but may also result from a variety of other alterations including decreased apoptotic signaling in response to this form of DNA damage [64, 67, 69, 70, 71, 72, 73, 74]. Two key genetic changes that have been associated with resistance to cisplatin are p53- and MMR-deficiency [19, 21, 22, 23, 25, 64, 75, 76, 77]. Despite the association of p53 and MMR defects with cisplatin resistance, RNAi against CSB and XPA in this panel of tumour cell lines deficient in p53 (PC-3, DU145 and HCT116p53-/-) and MMR (PC-3, DU145, HCT116 and HCT116p53-/-) resulted in a significant increase in the sensitivity of cells to cisplatin-induced cell death. The present results indicate that targeting TC-NER sensitizes tumour cells to cisplatin-induced apoptosis and that this was largely independent of p53 and MMR. This makes TC-NER an attractive target for combined cancer therapy.
We found that both p53 and MMR-deficient colorectal and prostate cancer cells retain the ability to perform TC-NER and targeting TC-NER in these cells increased their sensitivity to UV light and cisplatin. CSB and XPA may represent rational targets to augment cisplatin responsiveness of tumours, independent of MMR capacity and p53. This work further suggests that targeting other proteins involved in TC-NER may similarly represent a promising approach for cancer therapy. Notably, recent clinical evidence suggests that ERCC1 levels predict response to platinum-based therapies in non-small cells lung cancer [1, 78, 79, 80], small cell lung cancer , esophageal cancer , head and neck cancer , bladder cancer  and testicular cancer. ERCC1, like XPA, is required for both TC-NER and GG-NER. It is possible that the predictive value of ERCC1 is related to its role in TC-NER and that the expression of other proteins involved in TC-NER, like CSB, may also be predictors of therapeutic response in diverse tumours.
This work was supported by the Prostate Cancer Research Initiative, the National Canadian Institute of Cancer with funds from the Terry Fox Run and the Ottawa Regional Cancer Foundation with funds from the Motorcycle Ride for Dad. L.J.S. was supported by an Ontario Premier's Research Excellence Award to B.C.M. and an Ontario Graduate Scholarship in Science and Technology. J.M.S. was supported with an Ontario Graduate Scholarship in Science and Technology. B.C.M. was supported with a Research Scientist of the National Cancer Institute of Canada supported with funds provided by the Canadian Cancer Society.
- 19.Fajac A, Da Silva J, Ahomadegbe JC, Rateau JG, Bernaudin JF, Riou G, Benard J: Cisplatin-induced apoptosis and p53 gene status in a cisplatin-resistant human ovarian carcinoma cell line. Int J Cancer. 1996, 68 (1): 67-74. 10.1002/(SICI)1097-0215(19960927)68:1<67::AID-IJC13>3.0.CO;2-3.CrossRefPubMedGoogle Scholar
- 28.Koi M, Umar A, Chauhan DP, Cherian SP, Carethers JM, Kunkel TA, Boland CR: Human chromosome 3 corrects mismatch repair deficiency and microsatellite instability and reduces N-methyl-N'-nitro-N-nitrosoguanidine tolerance in colon tumor cells with homozygous hMLH1 mutation. Cancer Res. 1994, 54 (16): 4308-4312.PubMedGoogle Scholar
- 38.Rainbow AJ, McKay BC, Francis MA: Recombinant adenoviruses as expression vectors and as probes for DNA repair in human cells. Gene Ther Mol Biol. 2000, 5: 87-100.Google Scholar
- 51.Brash DE, Wikonkal NM, Remenyik E, Horst van der GT, Friedberg EC, Cheo DL, van Steeg H, Westerman A, van Kranen HJ: The DNA damage signal for Mdm2 regulation, Trp53 induction, and sunburn cell formation in vivo originates from actively transcribed genes. J Invest Dermatol. 2001, 117 (5): 1234-1240. 10.1046/j.0022-202x.2001.01554.x.CrossRefPubMedGoogle Scholar
- 52.de Waard H, Sonneveld E, de Wit J, Esveldt-van Lange R, Hoeijmakers JH, Vrieling H, Horst van der GT: Cell-type-specific consequences of nucleotide excision repair deficiencies: Embryonic stem cells versus fibroblasts. DNA Repair (Amst). 2008, 7 (10): 1659-1669. 10.1016/j.dnarep.2008.06.009.CrossRefGoogle Scholar
- 53.Garssen J, van Steeg H, de Gruijl F, de Boer J, Horst van der GT, van Kranen H, van Loveren H, van Dijk M, Fluitman A, Weeda G, et al: Transcription-coupled and global genome repair differentially influence UV-B-induced acute skin effects and systemic immunosuppression. J Immunol. 2000, 164 (12): 6199-6205.CrossRefPubMedGoogle Scholar
- 59.Lu Y, Lian H, Sharma P, Schreiber-Agus N, Russell RG, Chin L, Horst van der GT, Bregman DB: Disruption of the Cockayne syndrome B gene impairs spontaneous tumorigenesis in cancer-predisposed Ink4a/ARF knockout mice. Mol Cell Biol. 2001, 21 (5): 1810-1818. 10.1128/MCB.21.5.1810-1818.2001.CrossRefPubMedPubMedCentralGoogle Scholar
- 60.Liu F, Yu ZJ, Sui JL, Bai B, Zhou PK: siRNA-mediated silencing of Cockayne Cyndrome group B gene potentiates radiation-induced apoptosis and antiproliferative effect in HeLa cells. Chin Med J (Engl). 2006, 119 (9): 731-739.Google Scholar
- 72.Petersen LN, Mamenta EL, Stevnsner T, Chaney SG, Bohr VA: Increased gene specific repair of cisplatin induced interstrand crosslinks in cisplatin resistant cell lines, and studies on carrier ligand specificity. Carcinogenesis. 1996, 17 (12): 2597-2602. 10.1093/carcin/17.12.2597.CrossRefPubMedGoogle Scholar
- 81.Ceppi P, Longo M, Volante M, Novello S, Cappia S, Bacillo E, Selvaggi G, Saviozzi S, Calogero R, Papotti M, et al: Excision repair cross complementing-1 and topoisomerase IIalpha gene expression in small-cell lung cancer patients treated with platinum and etoposide: a retrospective study. J Thorac Oncol. 2008, 3 (6): 583-589. 10.1097/JTO.0b013e3181734f24.CrossRefPubMedGoogle Scholar
- 82.Kim MK, Cho KJ, Kwon GY, Park SI, Kim YH, Kim JH, Song HY, Shin JH, Jung HY, Lee GH, et al: Patients with ERCC1-negative locally advanced esophageal cancers may benefit from preoperative chemoradiotherapy. Clin Cancer Res. 2008, 14 (13): 4225-4231. 10.1158/1078-0432.CCR-07-4848.CrossRefPubMedGoogle Scholar
- 83.Handra-Luca A, Hernandez J, Mountzios G, Taranchon E, Lacau-St-Guily J, Soria JC, Fouret P: Excision repair cross complementation group 1 immunohistochemical expression predicts objective response and cancer-specific survival in patients treated by Cisplatin-based induction chemotherapy for locally advanced head and neck squamous cell carcinoma. Clin Cancer Res. 2007, 13 (13): 3855-3859. 10.1158/1078-0432.CCR-07-0252.CrossRefPubMedGoogle Scholar
- 84.Bellmunt J, Paz-Ares L, Cuello M, Cecere FL, Albiol S, Guillem V, Gallardo E, Carles J, Mendez P, de la Cruz JJ, et al: Gene expression of ERCC1 as a novel prognostic marker in advanced bladder cancer patients receiving cisplatin-based chemotherapy. Ann Oncol. 2007, 18 (3): 522-528. 10.1093/annonc/mdl435.CrossRefPubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2407/10/207/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.