The interaction of hydroxyurea and ionizing radiation in human cervical carcinoma cells

The results from prior in vitro and in vivo studies and recent phase 3 clinical trials suggest a significant potential role for hydroxyurea (HU) as a clinical radiosensitizer for cervix cancer. However, a detailed study of possible cellular mechanisms of radiosensitization in human cervix cancer cells as a consequence of dose and timing of HU and ionizing radiation (IR) has not been performed. This in vitro study analyses the interactions of HU and IR in a human cervical carcinoma cell line, Caski.

Exponentially growing Caski cells were continuously exposed to clinically achievable but minimally cytotoxic concentrations of HU (0.3-3.0 mM) for various time intervals (6, 12, 18, 24, and 30 hours) up to one population doubling time either prior to or immediately following IR (2-8 Gy). The radiation survival data were analyzed using our modification of the linear-quadratic model to test for an interaction (greater than additive). The effects of HU alone, IR alone, and the combination on cell cycle progression and on apoptotic cell death in exponentially growing Caski cells were measured.

We report a significant HU-IR interaction (radiosensitization) based on the sequence of HU exposure (post- > pre-IR) and with increasing concentrations of HU (0.3-3.0 mM), but no effect on radiosensitization with the duration of exposure to HU for up to one cell population doubling (6-30 hours). HU concentration has a significant effect on both alpha and beta linear-quadratic values in the post-IR sequences. Exposures of exponentially growing Caski cells to 1 mM and 3 mM HU alone result in a complete block in early S phase throughout the 30-hour exposure, while 0.3 mM HU causes a transient early S-phase block over the initial 12 to 18 hours of exposure. HU alone has no effect on cell cycle progression in G1 or G2/M populations but results in a large apoptotic population (31% following 1 mM HU x 30 hours), which appears to be the principal mechanism of drug cytotoxicity in these cells. IR alone (4 or 6 Gy) results in a significant G2 delay for 6 to 18 hours following IR but no G1 delay and a small apoptotic population at 30 hours post-IR (5.4% vs 2.1% in non-IR controls). The use of HU (0.3 or 1.0 mM) following IR (4 or 6 Gy) results in a significantly larger G2 delay compared with IR alone, but with only an additive effect on the apoptotic population.

These in vitro data demonstrate that radiosensitization of Caski cells is more significant with post-IR exposures to clinically achievable concentrations of HU. This HU-IR interaction is associated with an increased G2 delay, suggesting a reduction in IR damage repair. However, this interaction appears to be independent of the cytotoxicity (principally by apoptosis) from HU alone.