br Discussion br We investigated
We investigated the response of cervical cancer SP13786 to low dose UV exposure. Particularly, our eﬀorts illuminate the increased likelihood of UV-induced apoptosis in these cells. HPV oncogenes seem to counter the inclination towards programed cell death as their shRNA mediated knockdown caused increased p53- and BAX-associated apoptosis, sug-gesting dying cervical cancer cells sustain a greater abundance of DNA damage. This is somewhat suspected given the ability of HPV onco-genes to impair cellular DNA repair. We also generated clonal popu-lations of cervical cancer cells that were resistant to chemical- or ra-diation-induced DNA crosslinks and measured their sensitivity to other crosslinking agents and a PARP1-inhibitor.
4.1. HPV oncogenes have both protective and detrimental eﬀects on a cell's response to UV
One realization from this study is that HPV oncogenes sensitize cells to UV while also protecting them from UV-induced apoptosis. HPV oncogenes cause constitutive activation of both ATM and ATR, two keystone repair kinases (Gillespie et al., 2012; Johnson et al., 2017; Moody and Laimins, 2009). While this indicates repair initiation, suc-cessful repair also requires resolution of these signaling events as well. HPV E6 also mislocalizes repair proteins from sites of damage allowing the existence of active repair complexes without resolution of the da-mage they were activated in response to (Mehta and Laimins, 2018; Wallace et al., 2017). Such a disruption of repair signaling is consistent with the increased sensitivity to crosslinked DNA. HPV oncogenes protect the cell from BAK-mediated apoptosis, consistent with the vir-us's extensive eﬀorts to avoid immune detection by minimizing its cy-totoxic eﬀects (Jackson et al., 2000).
While the benefits of evading the immune system are obvious, the evolutionary pressures that drove HPV to sensitize cells to UV are more cryptic. Unlike cutaneous members of the papillomavirus family, in-fections in the genital tract would not encounter sunlight often, making Gene 688 (2019) 44–53
Fig. 6. Sensitivity to cross-linking agents in UV and cisplatin resistant cells. A. This graph depicts the sensitivity of HeLa cells as measured by MTT to ultra-violet radiation before and after acquiring resistance to cisplatin. The square points and solid line represent parental HeLa. The circle points and dotted line represent the Clone A of HeLa cells that acquired resistance to cisplatin. B. This graph depicts the sensitivity of HeLa cells to cisplatin measured by MTT before and after acquiring UV resistance. The square points and solid line represent the parental HeLa. The circle points and dotted line depict data from Clone A of UV resistant HeLa cells. The triangle points and the dashed line represent data from Clone B of UV resistant HeLa cells. C. This graph depicts the PARP1 inhibitor (Olaparib) sensitivity of HeLa cells measured by MTT before and after cisplatin resistance was acquired. Black squares represent the parental HeLa line while open circles depict data from Clone A of cisplatin resistant HeLa cells. The untreated controls are set at 100. For all, n = 3, *p < 0.05 by unpaired t-test and error bars represent mean ± SD.
it unlikely they evolved a mechanism to change the cellular response to UV. The increased toxicity is more likely an unintended consequence of the role repair factors play in replicating the HPV genome. Perhaps being recruited to sites of viral replication prevents ATM and ATR from coordinating the robust response required to avoid the deleterious ef-fect of UV-damage.
4.2. Resistance to crosslinking agents does not guarantee resistance to other genotoxic agents
We generated cervical cancer cell lines that were resistant to two diﬀerent sources of DNA crosslinks. The clonal populations of resistant cells insured that the acquisition of resistance occurred separately in each cell line. Although we did not determine the mechanism of re-sistance in these cells, the clonality suggests that cells gained the ability to survived either UV or cisplatin through diverse means. A significant amount is known about the ways that cells become tolerant of geno-toxic drugs. Potential resistance strategies include reducing the func-tional concentration of the drug in the cell by pumping it out of the cell or obtaining a mutation that restores expression of a repair factor (Michels et al., 2013; Rosell et al., 2003; Srivastava et al., 2015; Zhu et al., 2016). Less is known about the ways cells come to be more tol-erant of UV, but our observations are in line with previous reports in-dicating UV-resistance does not confer resistance to other DNA cross-linking agents (Petersen et al., 1995). Instead, UV-resistance appears to have fitness costs as cells grow slower and can be more sensitive to cisplatin (Data not shown, Fig. 6 and (Petersen et al., 1995)). Similarly, sensitivity to small molecule PARP1 inhibitor may or may not accom-pany resistance to cisplatin (Fig. 6 and Table 1). Moreover, cisplatin resistance was not accompanied with proportionate resistance to UV (Fig. 6 and Table 1). Together, this adds to the evidence that under-standing underlying mechanisms of resistance is critical for predicting the response of tumors to other genotoxic drugs.