Anti-tumour and radiosensitising effects of PARP inhibitor on cervical cancer xenografts

Abstract This study evaluated the radiosensitising effect of niraparib; a poly(adenosine diphosphate-ribose) polymerase (PARP) inhibitor on HeLa cervical cancer cells in nude mice and explored its possible mechanism. Twenty-four 3–5-week-old female BALB/c nude mice, inoculated with HeLa cells into the right hind leg, were randomly assigned into eight groups with three mice per group and treated. The tumour volume was significantly reduced under niraparib + radiotherapy combination as compared to monotherapy and untreated mice. The tumour growth was significantly delayed by 23.33–39 days when treated with combination therapy (p<.05). Further, univariate analysis revealed prolonged time for tumour growth when radiotherapy was followed by niraparib (I.G.) rather than niraparib (I.P.) (p=.003). Combination therapy reduced levels of PARP-1 precursor, PARP-1 splicer, PAR and RAD51 protein with high expression of γ-H2AX/CC3 and low expression of Ki-67. Niraparib in combination with radiotherapy can enhance the formation of DNA double strand breaks in HeLa cells and up regulate the expression of γ-H2AX/CC3. IMPACT STATEMENT What is already known on this subject? Asia has the highest incidence of cervical cancer (58.2%). Poly(adenosine diphosphate-ribose) polymerases (PARPs) are family of enzymes involved in single-strand break (SSB) and double-strand break (DSB) repair pathways. Niraparib is an effective inhibitor of both PARP-1 and PARP-2 and has the ability to cross the blood–brain barrier. What the results of this study add? Our study demonstrated that the combination of niraparib and radiotherapy can significantly enhance the cytotoxicity induced by radiotherapy. The inhibition effect of radiotherapy combined with niraparib on the tumour growth of mice was prominent, thereby establishing the radio-sensitisation activity of niraparib. What are the implications of these findings for clinical practice and/or further research? Niraparib can improve the cytotoxic effect of radiotherapy by increasing the formation of DSBs and up regulating the expression of apoptotic protein in HeLa cells.


Introduction
Cervical cancer is the fourth most common malignancy (WHO 2021) globally, affecting 604,127 women.Asia has the highest incidence of cervical cancer (58.2%) (Singh et al. 2022) with East Asia alone reported with an estimated 129,567 new cases and 66,436 mortality cases annually by year 2020 (Singh et al. 2023).
Treatment strategy usually consists of surgery or radiotherapy with chemotherapy as a valuable adjunct (Bhatla et al. 2018).The combination of radiotherapy and standard chemotherapy drugs (such as cisplatin) has demonstrated an improved survival in patients with cervical cancer (Rose et al. 1999, Pearcey et al. 2002).In recent years, technical advances including intensity modulated radiation therapy (IMRT), proton radiotherapy and stereotactic body radiotherapy (SBRT) and other precise radiotherapy technologies have opened a new avenue in the area of radiation oncology (Albuquerque et al. 2016, Chundury et al. 2016, Mesko et al. 2017, Lazzari et al. 2018).Radiotherapy offers prolonged duration of disease-free survival and improves the overall survival (Vordermark 2016, Yang et al. 2019).However, despite the continuous use of radiotherapy in clinical practice, the curative effect of radical radiotherapy for cervical cancer remains unsatisfactory.There are still obstacles such as cancer stem cells and tumour heterogeneity making it difficult to use radiotherapy alone to cure tumours (Gong et al. 2021).Radiosensitisers in combination with radiotherapy have the ability to increase the radiosensitivity of tumour tissue and pharmacologically decrease the normal tissue toxicity in an efficient way to improve radiotherapy (Farhood et al. 2019).
Poly(adenosine diphosphate-ribose) polymerases (PARPs) are family of enzymes involved in single-strand break (SSB) and double-strand break (DSB) repair pathways, and also plays a vital role in homologous recombination (HR), nonhomologous end joining (NHEJ) and alternative microhomology mediated end joining repair (Scott et al. 2015, Konecny and Kristeleit 2016, Lord and Ashworth 2017, Ray Chaudhuri and Nussenzweig 2017, Ashworth and Lord 2018).Proficient DNA-repair in cancer cells provides a primary factor responsible for tumour resistance to radiation (Wang et al. 2012).PARP-1 and PARP-2 can be activated by DNA damage and promote DNA repair.On the contrary, their deletion will lead to hypersensitivity to ionising radiation.Therefore, inhibiting PARP mediated DNA damage repair can enhance the sensitivity of tumour cells to radiation.
Since 1990, Begg (1990) proposed that the PARP inhibitors can replace cisplatin as radiosensitisers by avoiding the nephrotoxicity caused by cisplatin.In addition, most of the PARP inhibitors are oral drugs, which are convenient to use and non-invasive.Many of these inhibitors have shown antitumor activity and ameliorate the antitumor efficacy of cytotoxic DNA-damaging agents including both chemotherapy (such as alkylating agents, topoisomerase 1 and platinum) and ionising radiation.However, due to limited sample size and clinical data, most of the trials are in the exploratory stage.Besides, several in vivo models have confirmed the radiosensitisation of PARP inhibitors (Brock et al. 2004, Calabrese et al. 2004, Albert et al. 2007, Senra et al. 2011, Tuli et al. 2014).
The current generation of PARP inhibitors targets both PARP-1 and PARP-2 enzymes to improve efficacy and selectivity.Niraparib is an effective inhibitor of both PARP-1 and PARP-2 and has the ability to cross the blood-brain barrier.Niraparib has demonstrated radiosensitisation in both preclinical and clinical trials, thus improving anti-tumour activity (Jones et al. 2009, Sandhu et al. 2010, Wang et al. 2012, Bridges et al. 2014).Herein, we evaluated the radiosensitisation effect of niraparib using HeLa cervical cancer xenograft in nude mice model.

Experimental cells
The human cervical carcinoma cell lines, HeLa cells were obtained from the cell bank of Shengjing Hospital Affiliated to China Medical University, Shenyang, China.The cells were maintained in Roswell Park Memorial Institute (RPMI)-1640 medium containing 10% foetal bovine serum, and 100 units of penicillin and streptomycin with HEPES buffer.The cells were cultured in a humidified incubator (5% CO 2 ) at 37 � C. Cells in logarithmic growth phase were collected for this study.

Animals
Three-to five-week-old female BALB/c nude mice were provided by the laboratory animal centre of Wanlei Biotechnology Co., Ltd.(Shenyang, China).A total of 24 white female BALB/C nude mice (weighing 12.92 ± 0.29 kg) were purchased from Wanlei Biotechnology Co., Ltd.(Shenyang, China).The mice were exposed to 12-hour light dark cycles, and allowed to acclimatise for 1 week before use.The room temperature was maintained at 22 � C ± 1 � C and the relative humidity was maintained at 45-55%.The mice were allowed to have food and water, ad libitum.For tumour generation, 2 � 10 6 cervical cancer HeLa cells were injected subcutaneously into the right hind leg of each BALB/C nude mice.All the animals were maintained as per the protocol and the experiment was approved by the Ethics committee of China Medical University (2020PS660K).Animal experiments were performed in accordance with the ARRIVE Guidelines 2.0.

Tumour implantation and treatment regimens
The experimental treatment was started in the mice when the tumour grew to 10 mm in length.The usage and dosage of the drug are based on the existing research data: niraparib 50 mg/kg/d (obtained from China Tianjin Kailaiying Pharmaceutical Co., Ltd., Tianjin, China), intragastric (I.G.) or intraperitoneal (I.P.) injection, before or two hours after radiotherapy.X-ray (3 Gy/time, once every other day) was used for local radiotherapy.Before radiotherapy, vaseline with a thickness of 1.5 cm was evenly applied on the skin surface of the tumour-bearing leg of the mice to compensate the dose, ensuring that the other body parts were covered.The mice were randomly assigned into eight groups with three mice per group and allocated treatment as follows: (A) control (no treatment); (B) niraparib (I.P.) (50 mg/kg/d, d1-7); (C) niraparib (I.G.) (50 mg/kg/d, d1-7); (D) radiotherapy (3 Gy, every alternative day, d1/3/5); (E) niraparib (I.P.) followed by radiotherapy (two hours after I.P. injection of niraparib 50 mg/kg/d, radiotherapy was given 3 Gy/time for once every alternative day, d1/3/5); (F) niraparib (I.G.) followed by radiotherapy (two hours after I.G. administration of niraparib 50 mg/kg/d, radiotherapy was given 3 Gy/time once every alternative day, d1/3/5); (G) radiotherapy before niraparib (I.P.) (after 3 Gy/time of radiotherapy, 50 mg/kg/d of niraparib was injected I.P. two hours later, once every alternative day, d1/3/5); (H) radiotherapy before niraparib (I.G.) (after 3 Gy/time of radiotherapy, 50 mg/kg/d of niraparib was gavaged two hours later, once every alternative day, d1/3/5).

Tumour growth delay assay
Tumour growth delay (TGD) was the endpoint used to determine the antitumor efficacy of niraparib alone and in combination with radiotherapy.The tumour growth of mice was measured with vernier calliper every other day after the treatment.Tumour volume was calculated with the following formula: V ¼(long diameter �short diameter 2 )/2.The average TGD was the time (days) difference between the treatment group and the control group when the tumour grew to four times the volume before the treatment.The weight change of mice in each group was recorded.The body weight changes in mice were monitored to assess the systemic toxicity of niraparib, radiotherapy and combination therapy.

Histopathological analysis
For pathological examination of heterotopic lesions, tumour tissues were collected and cell lysate along with other reagents was added to obtain total RNA.The concentration of RNA in each sample was determined by NanoDrop TM 2000 (Thermo Fisher, Waltham, MA).Reverse transcription was performed to obtain the corresponding cDNA, and fluorescence quantitative analysis was performed by Exicycler TM 96 Real-Time Quantitative Thermal Block (Bioneer Corporation, Daejeon, South Korea).

Western blot analysis
To determine whether niraparib can enhance the DNA damage function in radiotherapy tissues, the levels of PARP-1 precursor (Wanleibio, Shenyang, China), PARP-1 splicer (Wanleibio, Shenyang, China), PAR (Santa, Northern California, USA) and RAD51 (Wanleibio, Shenyang, China) protein were examined by Western blot analysis.The lysate was used to lyse the tumour tissue and adjust the concentration of sample protein.Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) was used to isolate the proteins at 80 V voltage.After electrophoresis for 2.5 hours, the proteins were transferred to polyvinylidene fluoride (PVDF) membrane, sealed and incubated with a primary antibody.After the first antibody was incubated, the PVDF membrane was placed in goat antirabbit IgGHRP (Thermo Fisher, Waltham, MA) and rabbit anti-mouse IgG-HRP (Wanleibio, Shenyang, China) which was used as second antibody.After the second antibody was incubated, enhanced chemiluminescence reagent was used to develop the colour.With b-actin for internal reference, the optical density of the target strip was analysed by the gel image processing system (Gel-Pro-Analyzer software, Bethesda, MD).

Immunohistochemistry analysis
Immunohistochemistry (IHC) was used to detect CC3 (cleaved caspase-3), Ki-67 and c-H2AX levels.The specific steps were as follows: the tumour tissue was fixed in neutral formaldehyde for at least 24 hours, and later embedded in xyleneþparaffin.Paraffin embedded tumour tissues were cut into 5 lm sections and stained using immunofluorescence technique.For quality control of IHC assay, paraffin sections were compared with CC3, Ki-67 and c-H2AX (Abcam, Cambridge, UK) primary antibody and later with biotinylated goat anti-mouse (Beyotime, Nantong, China) antibody.After IHC staining, the sections were observed under a fluorescence microscope (BX53, OLYMPUS, Shibuya, Japan) with magnification of 400 times.The integrated optical density (IOD) was analysed by DP73 microscope photography system (OLYMPUS, Shibuya, Japan).

Statistical methods
The measurement data of different groups were described by the mean and the standard deviation.The comparison between the two groups was analysed using independent sample t-test, and one-way ANOVA.Statistical analysis was performed using SPSS 22.0 software (SPSS Inc., Chicago, IL).p < .05 was considered as statistically significant.

Tumour growth curve
Niraparib inhibited tumour growth to a certain extent (p < .05),and the tumour volume in the radiotherapy group was significantly inhibited compared with the untreated mice and niraparib monotherapy, indicating that the effect of radiotherapy on tumour growth of mice was significantly better than that of niraparib (p < .05)(Table 1).
The increase in tumour volume for the combination therapy was significantly smaller compared with tumour volume for the niraparib monotherapy and radiotherapy, indicating that the inhibition effect of the radiotherapy combined with niraparib on the tumour growth of mice was prominent, establishing the radio-sensitisation activity of niraparib.Furthermore, the tumour volume was the same when niraparib was given either by I.G. or I.P., indicating that niraparib exhibited the same anti-tumour effect irrespective of the route of administration (p > .05)(Figures 1 and 2).

Tumour growth delay time
Niraparib significantly delayed tumour growth by 8.33 days (p ¼ .026),whereas the treatment with radiotherapy delayed the tumour growth by 17.33 days (p ¼ .001).The combined effect of radiotherapy and niraparib was more profound than monotherapy as the tumour growth was visible after 23.33-39 days (p < .05) of treatment.Furthermore, univariate analysis revealed that the TGD was significantly longer in mice treated with radiotherapy and niraparib (I.G.) than those treated with radiotherapy and niraparib (I.P.) (p ¼ .003)(Table 2 and Figure 3).

Weight change
During the treatment, the mental state and the diet of mice in each group were normal, and there were no behavioural problems (such as sleepiness) or cachexia.No obvious skin lesions, contracture or erythema were found in the area of     irradiation.As observed in supplementary Figure S1, the weight of untreated nude mice steadily increased during the treatment (before vs. after treatment: 15.67 ± 0.74 vs. 17.10 ± 0.46), whereas there was only a slight weight gain in the mice treated with niraparib monotherapy.The weight of the mice treated with radiotherapy decreased by 12.26%, whereas the average weight of the combination therapy decreased by 6%.Under the same conditions, there was less weight loss in the mice treated with niraparib via I.P. route (before vs. after treatment: 15.47 ± 0.91 and 16.60 ± 1.11) than those treated via I.G. route (15.53 ± 0.23 vs. 17.03 ± 0.61).The results of the following experiment shows that niraparib followed with radiotherapy is highly effective and well tolerated with less weight gain in mice (mean change: 1.13 ± 1.59 (when niraparib is given I.P.) and 0.33 ± 0.81 (when niraparib is given via I.G.)) (Table 3).

Differential expression of DNA damage related proteins
The administration of niraparib either as monotherapy or combination with radiotherapy significantly reduced the PARP-1 levels.Although the PAR levels in the combination therapy group were higher than that of niraparib monotherapy, it was still significantly lower compared with radiotherapy group (supplementary Table S1).The RAD51 levels in HeLa cells were significantly down-regulated following treatment with combination therapy compared to those with only niraparib and radiotherapy treatment (supplementary Figure S2).c-H2AX was used as a marker of DSB, and its expression in HeLa tumour cells was detected by fluorescence imaging.As shown in supplementary Figure S3, the levels of c-H2AX were significantly increased in case of radiotherapy followed by the niraparib (I.G.) group (group H).These results suggest that combination of niraparib with radiotherapy not only retains the function of inhibiting DNA repair, but also improves the radiosensitivity of the xenograft tumour cells (supplementary Table S2 and supplementary Figure S3).
The expressions of CC3 and Ki-67 levels were analysed by the IHC.The results of IOD analysis showed that the CC3 was significantly increased in tumours from the combination therapy compared to their monotherapy treatment arms (p < .05).In addition, Ki-67 was significantly reduced in tumours following combination therapy compared to the treatment with niraparib or radiotherapy.These results indicate that the combination of niraparib and radiotherapy can significantly enhance the cytotoxicity induced by radiotherapy (supplementary Figure S4).

Discussion
Radiation therapy plays a central role in cancer treatment strategy, as a first-line treatment and either alone or in combination with radiosensitiser, for low operable tumours such as glioblastoma or advanced lung cancers.Nevertheless, the cure rates remain discouraging, whereas the concomitant chemotherapy generates systemic toxicity, thus leading to search of new radiosensitisers (Lesueur et al. 2017).High energy radiation can induce many types of DNA damage, including base modification, crosslinking, SSB and DSB.Its radiosensitivity is mainly affected by many factors, such as tumour intrinsic radiosensitivity, reoxygenation process, cell cycle redistribution, tumour tissue regeneration and tumour repair ability.At present, many radiosensitisers are available, such as chemotherapy drugs, targeted drugs, etc. (Pollom et al. 2017), but the improvement of their efficacy in treatment of cervical cancer is limited.An ideal radiosensitiser should have two main qualities: one is to provide better protection to normal tissues, the other is to improve the efficiency of tumour treatment (Lesueur et al. 2017).A large number of data show that DNA damage in cells can enhance the activity of PARP (Pommier et al. 2016, Mao et al. 2018).The application of PARP inhibitors or PARP deficient cells can enhance the sensitivity of tumour cells to various kinds of damage.Simultaneously, PARP inhibitors can expand tumour blood vessels, increase tumour perfusion, effectively reduce hypoxia and anti-radiation parts of tumour, making the tumour more sensitive to the radiation (Curtin and Szabo 2013).Therefore, we selected a highly selective and potent PARP inhibitor, niraparib, and tested its combination with radiotherapy to determine whether it increased the radiosensitivity in cervical cancer HeLa cells.In a study by Murai et al., niraparib demonstrated the most potent anti-tumour activity among all PARP inhibitors for trapping the PARP-DNA complexes that contribute to the PARP cytotoxicity (Murai et al. 2012).
In the experiment, we found that the tumour volume of the treated mice with combination of radiotherapy and niraparib was significantly inhibited compared with the other treatment groups, indicating that niraparib can increase the curative effect of radiotherapy and inhibits the tumour growth.At the same time, we found that the delayed time of tumour growth in mice treated with radiotherapy followed by I.G. administration of niraparib was up to 39 days, which was significantly longer compared with the other treatment groups.We hypothesise that oral administration of niraparib can not only increase the sensitivity of radiotherapy, but also act on tumour cells for a long time.
Recent studies have shown that the absence of PARP-1 can make the cells hypersensitive to ionising radiation.Therefore, we speculate that inhibition of PARP-1 could make HeLa cells more sensitive to radiotherapy.Using histones c-H2AX as a marker of DNA strand breaks, we found that niraparib combined with radiotherapy can better inhibit the DNA repair of tumour cells.High Ki-67 indicates a high level of metabolism in malignant cells with poorly differentiated tumour cells, as compared with the normal tissue (Li et al. 2015).In the Western blot and immunofluorescence experiments, the CC3 levels increased and Ki-67 levels decreased in the tumour tissue on treatment with niraparib radiotherapy combination therapy compared with only radiotherapy, while the immunofluorescence showed that the DNA damage repair decreased, indicating that the sensitivity of cells to radiation can be enhanced by increasing cell apoptosis and reducing a series of activities such as cell damaged DNA repair, thus inhibiting the tumour growth.Previous study reports that PARP-1 inhibition down-regulates RAD51, a key factor in the HR repair pathway (Hegan et al. 2010) which is consistent with our study findings.
Radiosensitisation of niraparib could be improved by using the prognostic biomarkers by informing the severity of cervical cancer.Biomarkers such as peritoneal human papillomavirus (HPV) positivity to detect early cervical lesion, p16 ink 4a, p16, E-cadherin, Ki67, pRb and p53 to detect the intraepithelial lesions that have more chances to evolve into invasive forms, could augment radiosensitiser therapy strategies and to monitor the therapy outcomes.Exploitation of these biomarkers at the translational level could help in achieving maximum treatment results of niraparib radiosensitisation (Valenti et al. 2017, Bizzarri et al. 2021).
The critical role of radiation in the treatment of cervical cancer is well established.In this study, we provide the preclinical evidence to support the introduction of the PARP-1/2 inhibitor niraparib in combination with radiation for patients with cervical cancer, for improving long-term clinical outcomes.Lack of data from an orthotopic model can be considered as one of the limitations of this study.Therefore, further investigations are recommended to establish and characterise such a model for the HeLa cell line.

Conclusions
In summary, this study confirmed the radiosensitisation effect of niraparib using cervical cancer HeLa cell xenograft model.Niraparib can improve the cytotoxic effect of radiotherapy by increasing the formation of DSBs and up regulating the expression of apoptotic protein in HeLa cells.The mechanism of combination therapy with niraparib and radiotherapy thus include augmented inhibition of DNA repair.The mode of cell death involves apoptotic cell death based on the increased CC3 levels and decreased Ki-67 levels and significant down-regulation of PAR in tumours from the mice treated with niraparib with radiotherapy.This study provides an experimental base for further investigation of PARP inhibitors in cervical cancer combined with radiotherapy.

Figure 1 .
Figure 1.Comparison of tumour growth in each treatment groups.

Figure 2 .
Figure 2. Growth curve of tumours after inoculation of HeLa cells.

Figure 3 .
Tumour growth delay time represented by univariate analysis.

Table 1 .
Effect of niraparib based on average tumour volume of HeLa tumour xenografts in nude mice.
HeLa: Henrietta Lacks.For treatment groups: (A) control (no treatment); (B) niraparib (I.P.);(C)niraparib (I.G.); (D) radiotherapy; (E) niraparib (I.P.) followed by radiotherapy; (F) niraparib (I.G.) followed by radiotherapy; (G) radiotherapy before niraparib (I.P.); (H) radiotherapy before niraparib (I.G.).(1) P0 is the comparison between each group and the control group; (2) P BC is the comparison between groups B and C ¼ 0.662; P BD is the comparison between groups B and D ¼ 0.009; P CD is the comparison between groups C and D ¼ 0.018; (3) P DE is the comparison between groups D and E ¼ 0.018; P DF is the comparison between groups D and F ¼ 0.031; P DG is the comparison between groups D and G ¼ 0.025; P DH is the comparison between groups D and H ¼ 0.012; (4) P BE is the comparison between groups B and E ¼ 0.001; P BF is the comparison between groups B and F ¼ 0.002; P BG is the comparison between groups B and G ¼ 0.001; P BH is the comparison between groups B and H ¼ 0.001; (5) P CE is the comparison between groups C and E ¼ 0.003; P CF is the comparison between groups C and F ¼ 0.005; P CG is the comparison between groups C and G ¼ 0.003; P CH is the comparison between groups C and H ¼ 0.003.

Table 2 .
Tumour growth delay time in various treatment groups of HeLa tumour xenografts in nude mice.

Table 3 .
Weight changes in various treatment groups of HeLa tumour xenografts in nude mice.