Enhancement of radiation killing of cultured mammalian cells by cordycepin

Effects of cordycepin on HepG2 and EA.hy926 cells: Potential antiproliferative, antimetastatic and anti-angiogenic effects on hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is a hypervascular tumor and accumulating evidence suggests that angiogenesis plays an important role in HCC development. Cordycepin, also known as 3′-deoxyadenosine, is a derivative of adenosine, and numerous cellular enzymes cannot differentiate the two. The aim of the present study was to determine whether cordycepin regulates proliferation, migration and angiogenesis in a human umbilical vein endothelial cell line (EA.hy926) and in a hepatocellular carcinoma cell line (HepG2). MTT was used to assess cell proliferation. Apoptosis was analyzed by flow cytometry (propidium iodide staining). Transwell and wound healing assays were used to analyze the migration and invasion of HepG2 and EA.hy926 cells. Angiogenesis in EA.hy926 cells was assessed using a tube formation assay. Cordycepin strongly suppressed HepG2 and EA.hy926 cell proliferation in a dose- and time-dependent manner. Cordycepin induced EA.hy926 cell apoptosis in a dose-dependent manner (2,000 μg/ml: 50.20±1.55% vs. 0 μg/ml: 2.62±0.19%; P<0.01). Cordycepin inhibited EA.hy926 cell migration (percentage of wound healing area, 2,000 μg/ml: 3.45±0.29% vs. 0 μg/ml: 85.48±0.84%; P<0.05), as well as tube formation (total length of tubular structure, 1,000 μg/ml: 107±39 μm vs. 0 μg/ml: 936±56 μm; P<0.05). Cordycepin also efficiently inhibited HepG2 cell invasion and migration. High-performance liquid chromatography analysis of the cytosol from EA.hy926 cells showed that cordycepin was stable for 3 h. In conclusion, cordycepin not only inhibited human HepG2 cell proliferation and invasion, but also induced apoptosis and inhibited migration and angiogenesis in vascular endothelial cells, suggesting that cordycepin may be used as a novel anti-angiogenic therapy in HCC.

An inhibitor of potentially lethal damage (PLD) repair reduces the frequency of gamma-ray-induced mutations in cultured Chinese hamster V79 cells

Cordycepin (3'-deoxyadenosine, 3'-dA) is an RNA antimetabolite and a radiosensitizer in cultured mammalian cells. In the present paper, the effects of 3'-dA on gamma-ray-induced lethality and 6-thioguanine (6TG)-resistant mutations in cultured Chinese hamster V79 cells were examined. 3'-dA had the effect of sensitizing the lethality induced by gamma-rays. The potentially lethal damage (PLD) repair produced by post-incubation of cells in Hanks' solution after gamma-irradiation was almost completely suppressed by 5 x 10(-5) M 3'-dA. When cells were irradiated with 10 Gy gamma-rays and incubated with 3'-dA for 5 h, the frequency of 6TG-resistant mutations induced by gamma-rays decreased to one-sixth of that of irradiated cells incubated without 3'-dA. The decrease in the frequency of gamma-ray-induced mutations was dependent on the length of incubation time with 3'-dA. It is suggested that the inhibition of PLD repair by 3'-dA may be that of error-prone repair.

Potentially lethal damage repair in human cells

Potentially lethal damage repair (PLDR) in human cells was investigated by delayed plating experiments with three malignant tumor cell lines (breast, colon and neuroblastoma) and two normal diploid lines (lung fibroblasts) in exponential and plateau phases. All cell lines demonstrated PLDR which was considerably higher for all the cell lines in plateau phase compared to log phase. At survival levels of 0.01, the two fibroblast lines had recovery ratios of 3.8 and 4.0 for 6 h delayed plating in plateau phase. The malignant lines showed recovery ratios of 5.7 (neuroblastoma), 3.2 (breast carcinoma) and 2.4 (colon carcinoma). The data were fitted to the linear quadratic equation and in addition, a repair factor was developed to compare survival for immediate and delayed plating: S(D) = exp[-gamma (alpha D + beta D2)] where gamma = 1 for immediate plating. Previously published data on human tumor and normal lines were also fitted to this equation. In all instances the three parameter equation (alpha, beta, gamma) allowed a good fit of the survival data for immediate and delayed plating. By using the same alpha, beta for the zero hour and 6 h survival curves, the gamma factor allows a much more useful method of comparing PLDR among different cell lines. The gamma factor for the two lung fibroblast lines were 0.70 and 0.71 while skin fibroblasts also had a 0.71 factor. The six malignant lines ranged from 0.62 to 0.93 with all but the neuroblastoma line having a higher factor (and therefore decreased PLDR capacity) than the normal lines. No correlation between PLDR and in vitro radiosensitivity or clinical radiosensitivity of the tumor type was found.

Response of cells of human origin, normal and malignant, to acute and low dose rate irradiation

Dose response curves were obtained for normal human fibroblasts and for several cell lines derived from human tumors, including melanomas and an osteosarcoma. Most of the tumor lines are similar in radiosensitivity to the normal fibroblasts, except for the melanoma lines, which are significantly more resistant. The two melanoma lines differ, one being much more radioresistant than the other. Potentially lethal damage repair (PLDR) has been studied in these cell lines as well. The extent of PLDR does not appear to correlate with radioresistance; for example, the most resistant melanoma line shows very little repair of PLD. In addition, the normal fibroblasts repair PLD at least as well as any of the tumor derived lines, which casts doubts on the wisdom of introducing into clinical practice inhibitors of PLD until a clear differential between normal tissues and tumors has been demonstrated in vivo. Low dose-rate studies with normal human fibroblasts indicate a smaller dose-rate effect than for most established cell lines of rodent origin. Indeed, in the human cells studied, the effect of sublethal damage repair is quantitatively similar to the repair of potentially lethal damage. Dose response curves for acute and protracted exposures have been obtained for cells derived from patients with cancer-prone syndromes including ataxia telangiectasia (AT) and Bloom's syndrome. Both cell lines are much more radiosensitive than normal human fibroblasts; the AT cells show a dose-rate effect, while Bloom's syndrome cells do not.