Differing effects of breast cancer 1, early onset (BRCA1) and ataxia-telangiectasia mutated (ATM) mutations on cellular responses to ionizing radiation.

PURPOSE: The ataxia-telangiectasia mutated (ATM) gene encodes a protein kinase, the activation of which is an early event in the cellular response to ionizing radiation. One of the many substrates of ATM is BRCA1 (breast cancer 1, early onset gene), which has been associated with susceptibility to breast and ovarian cancer, and has been implicated in DNA repair processes. Various cellular responses to radiation were analysed in cells with mutations in ATM or BRCA1 in an attempt to clarify which effects of ATM can be mediated through BRCA1. MATERIALS AND METHODS: The response to radiation of cells with mutations in ATM or BRCA1 was examined, as were BRCA1-mutant tumour cells transfected with an exogenous wild-type BRCA1 allele. Assays included cell-survival curves, studies of potentially lethal damage repair, measurement of chromosomal aberrations and of G1 arrest, and Western blot analysis of lysates of irradiated cells to determine the phosphorylation of the product of the human Mdm2 gene (HDM2). RESULTS: Both ATM and BRCA1 mutations were associated with sensitivity to ionizing radiation, deficient repair of potentially lethal damage and markedly increased chromosomal aberrations. A BRCA1-mutated tumour cell line HCC1937, like ATM mutant cells, did not exhibit a normal G1 arrest but, unlike ATM mutant cells, did exhibit phosphorylation of HDM2. Expression of wild-type BRCA1 in HCC1937 cells partially restored radioresistance, restored repair of potentially lethal damage and markedly reduced radiation-induced chromosomal aberrations. G1 arrest, however, was not restored by expression of BRCA1. CONCLUSIONS: The results are consistent with a model in which ATM phosphorylation of BRCA1 regulates DNA repair functions, particularly those involved in potentially lethal damage repair and chromosomal integrity, but not other aspects of the cellular response to radiation such as G1 cell cycle arrest. To the authors' knowledge, this is the first demonstration of the ability of exogenously expressed BRCA1 to restore the ability to perform potentially lethal damage repair and maintain chromosomal integrity in irradiated cells.

ATM complexes with HDM2 and promotes its rapid phosphorylation in a p53-independent manner in normal and tumor human cells exposed to ionizing radiation.

To further understand the mechanism(s) by which DNA damage activates p53, we analysed the expression levels of p53 and HDM2 (the human homolog of murine MDM2) in various human diploid fibroblast and tumor cell strains during the period that precedes activation of known downstream effectors of p53. In X-irradiated human cells, HDM2 protein was rapidly phosphorylated in serine/threonine residues in a p53, p14ARF and p73-independent manner. In p53 wild-type cells, HDM2 phosphorylation precedes a detectable increase in the levels of p53 and is not observed in ataxia telangiectasia (AT) fibroblasts. The transfection of AT cells with a vector expressing ATM restored the ability to rapidly phosphorylate HDM2 following X-irradiation, confirming a role for ATM in its phosphorylation. We also show that ATM complexes with HDM2. The DNA lesions signaling the early rapid phosphorylation of HDM2 are a result of X-ray and not UV-type damage. The ATM-promoted early covalent modification of HDM2 in X-irradiated human cells may provide a mechanism to activate p53.

Response of human tumor cells of varying radiosensitivity and radiocurability to fractionated irradiation.

The cytotoxic effects of radiation delivered in daily fractions of 2.0 Gy were examined in plateau phase cultures of human tumor cells of varying in vitro radiosensitivity, derived from tumors of varying radiocurability. Among the eight cell lines examined, three types of responses to fractionated irradiation were observed. In the group composed of tumor cell lines that were radioresistant in culture (D0 > 2 Gy) and derived from known local radiation failures or from tumor histologies associated with radiation failure, a gradual linear reduction in surviving fraction versus total dose was observed. In a second group, composed of cell lines that were radiosensitive in culture (D0 approximately 1 Gy) but derived from known radiation failures, the surviving fraction initially declined and began to plateau after 6 Gy (three fractions of 2 Gy). In the third group, composed of radiosensitive cell lines derived from tumors associated with high radiocurability, a rapid decline in surviving fraction versus total dose was observed. The in vitro response of human tumor cells to fractionated irradiation delivered at clinically relevant doses appears to be independent of in vitro X-ray sensitivity and p53 status but related to clinical radiocurability, suggesting a possible role in predicting tumor response to radiotherapy.

Differential p53, p21, mdm2 and Rb regulation in glioma cell lines that overexpress wild-type p53.

Gliomas are part of a subset of tumors in which overexpression of p53 protein in the absence of p53 gene mutation has been described. We have utilized a series of glioma cell lines to study the effects of ionizing radiation on the regulation of p53, p21, mdm2 and Rb proteins. The induction of p53 protein in glioma cell lines that overexpress wild-type p53 differs from normal control cells and glioma cell lines containing mutant p53. Alterations in the accumulation of p53 and p21 proteins are associated with diminished Rb hypophosphorylation. Gliomas that overexpress wild-type p53 also express high levels of mdm2 protein and exhibit a radiosensitivity that is intermediate between normal cells and cells with mutant p53. These findings suggest that, at least in certain glioma cell lines that over-express p53 which is wild-type in sequence, the function of p53 protein is abnormal.

Radiosensitivity of human biliary tract cancer cell lines in vitro.

The prognosis of biliary tract cancer is still poor. Although a number of clinical studies have suggested a role for radiation therapy in advanced biliary tract cancer, its value remains controversial. Moreover, the intrinsic radiosensitivity of bile duct cancer cell lines has not been described, and the molecular basis for the response of these tumors to ionizing radiation is poorly understood. The present study was designed to examine the intrinsic radiation sensitivity of human biliary tract cancer cells and its relationship to p53 status. Radiation response expressed by the parameters n, D-0, D-10, alpha, beta, (D) over bar (mean inactivation dose), and SF, of seven cell lines derived from gallbladder and bile duct cancers was determined. The results suggest that biliary tract cancer cell lines as a group are relatively radioresistant. The mean X-ray survival parameters for these seven cancer cell lines were D-0=2.13+/-0.29 Gy, D-10=5.73+/-0.59 Gy, (D) over bar=2.76+/-0.25 Gy, alpha=0.25+/-0.03, and SF2=0.54+/-0.05. One of the seven lines was more radiosensitive than the others (D-0=0.77+/-0.02 Gy, D-10=2.95+/-0.06 Gy, (D) over bar=1.57 Gy, alpha=0.35, SF2=0.35+/-0.03). Five of six lines examined expressed mutant p53 including the radiosensitive line; one radioresistant line expressed wild-type p53. Thus, although loss of wild-type p53 expression occurred frequently in these biliary cancer cell lines, radiosensitivity did not correlate with p53 status. In view of the intrinsic radioresistance of this type of tumor cell coupled with the poor tolerance of surrounding normal tissues, maximal surgical debulking and intraoperative radiation therapy may contribute to increased local control over resection and/or conventional fractionated radiotherapy.

Diminished capacity for p53 in mediating a radiation-induced G1 arrest in established human tumor cell lines.

It has been reported that the p53 gene mediates an ionizing radiation-induced G1 arrest in mammalian cells. To further characterize this important phenomenon, a panel of seven human diploid fibroblast cell strains and 14 human tumor cell lines from a variety of sources with both wild-type and mutant p53 status were assayed for their susceptibility to G1 arrest after gamma-ray irradiation by a continuous labeling [3H]thymidine incorporation technique. An irreversible G1-block involving 20-70% of the cell population was observed in diploid fibroblasts irradiated with 4 Gy. The block was abolished by transfection with the Human Papilloma Virus E6 gene and in an ataxia telangiectasia (AT) cell line, indicating a role for the AT and p53 genes respectively in this process. In contrast to wild-type normal fibroblast cell strains, the G1-block in all tumor cell lines was significantly reduced, irrespective of their p53 status. None of the nine human tumor cell lines with mutant p53 genes showed a significant G1-block following irradiation with 4 Gy. Among the five tumor cell lines expressing wild-type p53, two showed no apparent G1-block. The remaining three showed a G1-block involving only 8-15% of the cell population, a block much smaller in magnitude than that seen in diploid fibroblasts. Finally, a diploid fibroblast cell strain and a tumor cell line, both showing a normal p53 and p21/WAF1 expression pattern, were examined for pRb phosphorylation before and after irradiation. The diploid fibroblast cell strain showed a significant G1-arrest and a clear inhibition of pRb phosphorylation by irradiation whereas the tumor cells showed no G1-arrest and no inhibition of pRb phosphorylation. These results suggest that (1) multiple genetic factors may modulate the occurrence and magnitude of the G1-arrest induced by exposure to ionizing radiation, (2) the capacity for p53 to mediate a radiation-induced G1 arrest is significantly reduced in tumor cells, (3) the disruption of G1-block modulating factor(s) other than p53 may be an important step in carcinogenesis.

Response of dermal fibroblast cultures from patients with unusually severe responses to radiotherapy and from ataxia telangiectasia heterozygotes to fractionated radiation.

We examined the cytotoxic effects of radiation delivered in daily fractions at clinically relevant doses in plateau phase cultures of skin fibroblast cell strains derived from ataxia telangiectasia (AT) heterozygotes, patients with unusually sensitive responses to radiotherapy, apparently normal patients, and cell bank controls. A gradual linear reduction in surviving fraction versus total dose was observed in the control group, comprised of apparently normal individuals and one patient with a normal clinical response to radiotherapy, after exposure to daily fractions of 2.0 Gy. There was a much steeper decline in surviving fraction among the AT heterozygotes and the group with sensitive responses to radiotherapy, such that after six daily fractions of 2.0 Gy (12 Gy total dose), the mean surviving fraction of the control group was significantly different from that of the AT heterozygotes (P = 0.0009) and that of the patients with unusually sensitive responses to radiotherapy (P = 0.0002). We propose that this assay may be a useful means of identifying cell strains from AT heterozygotes. Based on these results, the hypothesis is discussed that patients who suffer unusually sensitive clinical reactions to radiotherapy may be AT heterozygotes.

Relationship between gamma-ray-induced G2/M delay and cellular radiosensitivity.

Seven human and rodent cell lines with markedly differing cellular radiosensitivities were examined by the anti-Br-dUrd antibody and flow cytometric method in order to measure the progression of S phase cells and their accumulation in the G2 phase of the cell cycle after gamma-irradiation. Exponentially growing cells were labelled with 10 microM BrdUrd for 2 h, gamma-irradiated, then washed and cultured at 37 degrees C. At 2-h intervals postirradiation, the cells were harvested and fixed for flow cytometric analysis. Two parameter distributions of BrdUrd content and DNA content were analysed. The time intervals for unirradiated labelled cells to progress from S to G2/M phase were about 450 min for the human squamous cell carcinoma cell lines SCC-12B.2 (D0 = 2.66 Gy), SQ-20B (D0 = 2.39 Gy) and SCC-61 (D0 = 1.07 Gy) as well as for wild-type CHO cells (D0 = 2.62 Gy). After irradiation with 2 Gy, SCC-12B.2, SQ-20B, CHO and human diploid AG1521 cells showed similar small G2/M delays (about 1 h), whereas, a G2/M delay of about 2.2 h occurred in radiosensitive SCC-61 cells and delays of 5.0-7.7 h were observed in two extremely radiosensitive mutant cell strains (human AT homozygote and CHO xrs-5 respectively). When the cells were irradiated with doses yielding similar levels of survival (about 10%), however, the duration of the G2/M delay was generally similar (2-4 h) in all seven cell lines indicating a parallel relationship between radiation-induced G2/M delay and cellular radiosensitivity. These results suggest that the delay time may be related to the level of unrepaired damage present in the cell as it approaches mitosis.

Radiosensitivity in vitro of human soft tissue sarcoma cell lines and skin fibroblasts derived from the same patients.

Skin fibroblast cell strains and tumour cell lines were established from 12 patients with various types of soft tissue neoplasms, and radiation survival curve parameters were measured in vitro. Soft tissue sarcoma cells were consistently more sensitive to X-irradiation than fibroblasts isolated from the same patient, and were also more sensitive as a group than cell lines derived from 34 other human tumours. There was a general correlation in radiosensitivity between fibroblasts and tumour cells derived from the same patient, indicating that some component of tumour cell sensitivity may relate to genetic factors in the host. Such genetic factors, however, do not explain all of the heterogeneity in tumour cell response. The response of soft tissue sarcoma in vivo may be dependent on complex radiomodifying factors other than inherent radiation sensitivity, thus making it difficult to predict clinical outcome by use of assays which use survival of irradiated tumour cell lines in vitro as an endpoint.

Differential sensitization of human tumor cells by ara-A to X irradiation and its relationship to inherent radioresponse.

We have previously shown that pretreatment of plateau-phase cultures of human tumor cells with ara-A can markedly sensitize them to the cytotoxic effects of X irradiation; the degree of sensitization varied in two different cell lines. The present study was undertaken to determine whether variability in radiosensitization by ara-A occurs at random in human tumor cell lines or if it is related to their intrinsic radiosensitivity (human tumor radioresponse). The interaction between ara-A and X irradiation was examined in plateau-phase cultures of early-passage tumor cell lines of varying radioresponse (D0 range 0.85-3.15 Gy) subcultured immediately after irradiation to measure survival. In six of the eight cell lines studied, pretreatment with ara-A greatly enhanced the lethal effects of X irradiation in a concentration-dependent fashion. Little or no effect was observed in the two radiosensitive cell lines. When ara-A sensitization was plotted as a function of D10 or D, a linear relationship was observed. These data suggest that pretreatment with ara-A is effective in sensitizing radiation-resistant human tumor cells to the lethal effects of X rays, and that this phenomenon may be dependent upon inherent tumor cell radiosensitivity.

In vitro radiosensitivity of human diploid fibroblasts derived from women with unusually sensitive clinical responses to definitive radiation therapy for breast cancer.

Between January 1985 and December 1986, 811 patients were treated for carcinoma of the breast at the Joint Center for Radiation Therapy by an identical protocol. Of these 811 patients, five patients (0.6%) were identified as having an unusually sensitive clinical response to routine external beam irradiation. This unusual clinical response was characterized by severe skin erythema and edema during the first few weeks of treatment, requiring treatment breaks. Skin fibroblast cell strains were established from these five women as well as from six women with a normal clinical response to breast irradiation (chosen at random from the population of 811 patients). Radiation survival parameters were determined by a colony formation assay from complete survival curves in coded and blinded samples. Cells from the sensitive patients were significantly more sensitive to the cytotoxic effects of radiation in vitro as determined by the parameters D0, D, D10, and n, than were the strains derived from patients with a normal response. We conclude that an unusually severe response to standard fractionated radiotherapy may be associated with greater intrinsic radiation sensitivity of the individual's somatic cells.

Sensitization of noncycling human diploid cells to X-irradiation by pretreatment with 9-beta-D-arabinofuranosyladenine.

The interactions between X-irradiation and 9-beta-D-arabinofuranosyladenine (ara-A) were studied in confluent, density-inhibited cultures of normal human diploid fibroblast cell strain AG1522. ara-A (0.1 to 3.0 x 10(-3) M) was added to the cultures either 2 or 24 h prior to irradiation. The cells were subcultured at low density to measure the surviving fraction either immediately after irradiation, or they were returned to the incubator in the continued presence of ara-A for an additional 4 or 24 h. In cells subcultured immediately after irradiation, pretreatment with ara-A greatly enhanced cell killing by X-rays in a concentration-dependent fashion. The D0 of the survival curve decreased from 143 cGy to 74 or 49 cGy following pretreatment with 1.0 or 3.0 x 10(-3) M ara-A, respectively. ara-A by itself had little effect on the cloning efficiency of nonirradiated cells at concentrations up to 10(-3) M, but higher levels were cytotoxic. When subculture was delayed for 4 or 24 h, the enhancement in survival reflecting potentially lethal damage recovery was not reduced by continuing incubation with ara-A at concentrations up to 10(-3) M. Higher levels significantly inhibited potentially lethal damage repair, as well as producing marked sensitization. With 24-h delayed subcultivation, the D0 of the survival curve decreased from 324 cGy to 63 cGy following incubation with 3.0 x 10(-3) M ara-A. These results indicate that pretreatment with nontoxic concentrations of ara-A can markedly sensitize human cells to the lethal effects of X-rays. This effect appears to be independent of its ability to inhibit postirradiation cellular recovery processes.

Differential response of human and rodent cell lines to chemical inhibition of the repair of potentially lethal damage.

We have examined the effects of several classes of metabolic inhibitors on the repair of potentially lethal damage in density-inhibited cultures of two rodent and two human cell systems which differ in their growth characteristics. Aphidicolin, 1-beta-D-arabinofuranosylcytosine (ara-C) and hydroxyurea showed no effect on PLD repair, whereas the effects of 9-beta-D-arabinofuranosyladenine (ara-A) and 3-aminobenzamide (3-AB) were cell line dependent. For example, 3-AB suppressed PLD repair almost completely in CHO cells, but showed no inhibitory effects in human diploid fibroblasts. These results indicate that inhibitors of DNA replication and poly(ADP-ribose) synthesis are not efficient inhibitors of cellular recovery in irradiated cells and, moreover, that such effects may be cell line dependent.

Heterogeneity of radiation response of a parent human epidermoid carcinoma cell line and four clones.

We examined the radiobiological parameters of a parent tumor cell line and four tumor clones of a human skin squamous cell carcinoma. The parent line and all clones have a tumor morphology, aneuploid karyotype, and the ability to passage infinitely in vitro. The parent cell line and three of four clones formed tumors in nude mice. The parent line, SCC-12, has a D0 of 1.59 Gy and an n of 7.5. In the four clones, D0 ranges from 1.31 Gy to 2.66 Gy and n ranges from 2.1 to 22.8. Potentially lethal damage repair (PLDR) following 11 Gy ranges from 1.7 to 13.1, at 24 hours, however PLDR following equitoxic doses of radiation is similar. The mean inactivation dose value (D) in the parent tumor cell line is 3.48 and ranges from 3.31 to 4.17 in the tumor clones. Radiobiological interpretation of heterogeneity may complicate predictive assays for clinical radiotherapy.

Normal cytotoxic response of skin fibroblasts from patients with Li-Fraumeni familial cancer syndrome to DNA-damaging agents in vitro.

Skin fibroblasts from patients with the Li-Fraumeni familial cancer syndrome have been reported to show abnormalities in their response to X-irradiation. We have examined the response of fibroblasts from affected and nonaffected individuals in three families to treatment with four DNA-damaging agents: X-rays, UV light, N-methyl-N'-nitro-N-nitrosoguanidine, and mitomycin C. Test cells along with additional cell bank control strains were received coded and blinded. The same strains were studied on 2 or 3 separate occasions; each of these groups was coded differently. The cytotoxic effects of the four agents were examined by a colony formation assay. Sensitivity to the induction of chromosomal aberrations by X-rays was also measured. In all cases, the response of cells from affected individuals did not differ significantly from that of cells from unaffected (not a risk) family members nor of cell bank controls. The response of somatic cells from members of Li-Fraumeni cancer families to DNA-damaging agents does not appear to be a fruitful approach to the detection of at-risk individuals.

Radiation-resistant and repair-proficient human tumor cells may be associated with radiotherapy failure in head- and neck-cancer patients.

Inherent cellular radioresistance and repair of x-ray damage was studied in 19 early-passage squamous cell carcinoma lines derived from head- and neck-cancer patients with known clinical results following radiotherapy. Human tumor cells that were radioresistant and/or proficient in accumulation/repair of x-ray damage were cultured from patients unsuccessfully treated with radiotherapy. Thus, the presence of radiation-resistant and repair-proficient tumor cells was associated with clinical radiation failure, suggesting the possibility of a predictive assay based on in vitro radiobiological parameters.

Inherently radioresistant cells exist in some human tumors.

We investigated the survival of 29 human tumor cell lines after treatment with ionizing radiation during exponential growth. The radiosensitivity of 20 tumor cell lines was within the range of normal fibroblasts. However, 9 cell lines derived from radio-incurable tumors appeared inherently radioresistant. When the ability to repair potentially lethal x-ray damage was studied in plateau phase cultures, some tumor cell lines derived from radio-incurable tumors were more efficient in this process. These results indicate that the presence of radioresistant and/or repair-proficient cells, which exist in some human tumors, may be responsible for failure to obtain local control following therapeutic x-ray treatment.

Influence of nutrient and energy deprivation on cellular response to single and fractionated heat treatments.

The effect of hypoglycemia and hypoxia on cell survival following single and fractionated 44 degrees C heat treatments was investigated. These studies show that short-term glucose deprivation or oxygen deprivation does not increase sensitivity to single or fractionated heat treatments compared to normal nutrient conditions. However, simultaneous hypoglycemia and hypoxia leads to an increased sensitivity to single heat treatments, and to an even greater increase in sensitivity to fractionated heat treatments. Sensitivity to heat treatment under nutrient-deprived conditions correlates with intracellular ATP levels at the time of treatment. These data indicate that it is not nutrient concentration per se, but rather the effect of nutrient level on intracellular ATP concentration that affects thermal sensitivity.

Cellular X-ray repair parameters of early passage squamous cell carcinoma lines derived from patients with known responses to radiotherapy.

We have investigated X-ray survival parameters and repair of potentially lethal damage ( PLDR ) in ten early passage squamous cell carcinoma cell lines derived from patients who were biopsied before initiation of radiotherapy or after radiation therapy failure. Radiosensitivity (D0) ranged from 1.07 to 1.93 (Gy), extrapolation numbers (-n) from 1.17 to 2.14 and PLD recovery at 24 h from 1.4 to 20.3. Despite significant differences in these parameters amongst the cell lines, a firm correlation between radiocurability and any individual radiobiological parameter could not be established. Our data suggest that the mechanisms associated with radioresistance are complex and that any single radiobiological parameter may not predict clinical success or failure.

Lethal effects of 1-beta-D-arabinofuranosylcytosine incorporation into deoxyribonucleic acid during ultraviolet repair.

We have previously demonstrated that 1-beta-D-arabinofuranosylcytosine (ara-C) incorporates into leukemic cell DNA and that the extent of this incorporation correlates significantly with loss of clonogenic survival. These studies have now been extended by monitoring the incorporation of ara-C in DNA undergoing repair of UV-induced damage. The present studies have been performed using human foreskin diploid fibroblasts that undergo density-dependent growth arrest. The results demonstrate that ara-C incorporates specifically into DNA undergoing repair of damage by UV light. The extent of ara-C incorporation in AG1522 DNA during UV repair correlates significantly (p less than 0.0006) with cell lethality. These findings confirm that incorporation of ara-C into DNA undergoing either semiconservative or unscheduled DNA synthesis results in lethal cellular events. The present findings may provide the experimental basis for the development of new therapeutic approaches using ara-C in combination with agents that damage DNA.