In vitro and in vivo imaging of irradiation induced changes in PARP1 expression in oral cancer cell lines

Objectives Imaging of the expression dynamics of tumor specific markers during or after a therapeutic intervention would enable new ways to study tumor biology non-invasively and may eventually allow physicians to quickly adapt therapeutic strategies and regimens. Imaging of the DNA damage response of cancer cells is particularly interesting, since it is directly affected by common therapeutic interventions such as radiotherapy, and might therefore be a reliable biomarker for treatment success. Therefore, we investigated how the expression of the DNA damage repair enzyme Poly(ADP-ribose) Polymerase 1 (PARP1) is affected by radiotherapy in oral squamous cell carcinoma (OSCC), and if changes in its expression can be imaged with the PARP1 imaging agent PARPi-FL.

Methods Baseline PARP1 expression was determined in xenografts of OSCC cell lines (FaDu and Cal27) as well as normal mouse tissues (tongue, trachea and thigh muscle) using Immunofluorescence staining (IF). We determined the effect of 0 - 10 Gy irradiation on clonogenic survival. For Molecular Imaging of PARP1 in vitro and in vivo, we used PARPi-FL (MW: 640 g/mol), which is based on the PARP1 inhibitor Lynparza (Olaparib, Astra-Zeneca) and conjugated to the green fluorescent dye BODIPY-FL [1]. In vitro, changes in uptake of PARPi-FL at different time points post irradiation were measured via flow cytometry. For in vivo studies, FaDu tumors were subcutaneously inoculated in athymic nude mice and locally irradiated with 10 Gy using an image-guided microirradiator, while controls were not irradiated. At 24 and 48 h post irradiation, we a) determined uptake of PARPi-FL after i.v. injection using epifluorescence imaging and b) quantified PARP1 expression in tumors using IF staining.

Results We found a pronounced baseline expression of PARP1 in both cancer cell lines. Control tissues had a very low PARP1 expression. 10 Gy irradiation caused a strong decrease in the surviving fraction (>95% for FaDu, >99% for Cal27), however cell viability was > 80% 24 h post irradiation. With increasing irradiation doses (0, 2, 4 and 10 Gy) PARPi-FL uptake into surviving cells gradually increased at 24 and 48 h after irradiation, but not at 6 h. A significant uptake increase of up to 50% was observed after 10 Gy irradiation at 24 and 48 h post irradiation. In vivo irradiation led to strong inhibition of tumor growth up to 30 days after irradiation. This was reflected in an increase in the uptake of PARPi-FL in irradiated tumors compared to non-irradiated ones and a corresponding increase in PARP1 expression in the tumors at 24 and 48 h post irradiation. Upon irradiation, the PARP1 positive area increased stronger than the PARP1 intensity per nucleus, indicating changes in the tumor architecture. The correlation of PARPi-FL nuclear uptake in tumor cells and PARP1 IF staining was very strong (mean r2: 0.92), while stromal cells showed neither PARPi-FL uptake nor PARP1 staining.

Conclusions We could show that changes in PARP1 expression occur upon radiation treatment of oral cancer and that these changes can be measured using PARPi-FL. This indicates that PARPi-FL or other PARP1 targeted imaging agents could be used to delineate tissues exposed to radiation. Ultimately, implementation of other modalities, such as PET Imaging, using 18F-labeled PARP Inhibitors, will be critical for clinical translation of this approach. In future studies, we will elucidate the relationship between changes in PARP1 expression and long-term therapy outcome.