Caroline A. Chadwick

Apoptosis, intrinsic radiosensitivity and prediction of radiotherapy response in cervical carcinoma

Apoptosis is an important mechanism of cell death in tumours and it is seen both prior to and following radiotherapy. In this study patients with proven carcinoma of the cervix had measurement made of the percentage of apoptotic cells (apoptotic index or AI) in pre-therapy biopsies. Measurements of intrinsic radiosensitivity (SF2), already shown to be a predictor of outcome, had previously been made on the same pre-therapy biopsies. Mitotic index (MI) and Ki-67 antigen staining were also recorded as markers for proliferation. Patients were divided into those with an AI above or below the median and in general increasing apoptosis was associated with poor prognosis. The 5-year survival rate for tumours with an AI below the median was 79% and was significantly greater than the rate of 47% for those with an AI above the median (p = 0.003). There was also a significantly increased 5-year local recurrence-free rate for patients with an AI below the median compared with those with an AI above the median (79 versus 61%, p = 0.012). In addition, AI and SF2 acted as independent prognostic indicators. Patients with both an SF2 and AI value above the median did badly (25% 5-year survival, 46% local control) compared with those with an SF2 and AI below the median (80% 5-year survival, 100% local control). Apoptosis showed correlation with MI (n = 66, r = 0.34, p = 0.002) and cell staining for the Ki-67 antigen (n = 57, r = 0.25, p = 0.03), but neither MI nor Ki-67 were related to patient outcome. This suggests that while apoptosis may be a reflection of tumour proliferation this cannot in itself explain the ability of apoptosis to predict clinical outcome for this series of patients. The study raises the possibility of AI and SF2 being used together as predictors of tumour response to radiotherapy.

Cell Death (Apoptosis) in the Mouse Small Intestine after Low Doses: Effects of Dose-rate, 14·7 MeV Neutrons, and 600 MeV (Maximum Energy) Neutrons

The production of dead (apoptotic) cells by low doses of gamma-rays was independent of dose-rate between 0.27 and 450 cGy per min. The r.b.e. for doses of 14.7 MeV neutrons between 1 and 15 cGy was about 4, and for neutrons generated by bombarding a beryllium target with 600 MeV protons the r.b.e. was about 2.7. The dose-incidence curves for all three radiation types reached a plateau at about 3-4 dead cells per crypt section, and this occurred at about 20-40 cGy of gamma-rays. These curves are compatible with exponential survival of the cell population at risk (D0 of 24 cGy for gamma-rays, 6 cGy for 14.7 MeV neutrons and 9 cGy for 600 MeV neutrons). Since the dose-response is exponential there is no indication of much higher r.b.e. values at very low doses, a point of concern in radiation protection. The spatial distribution of dead cells in the crypt was similar after doses of gamma-rays or neutrons, indicating that the same population of target cells was affected in both cases.

The detection of cyclobutane thymine dimers, (6-4) photolesions and the Dewar photoisomers in sections of UV-irradiated human skin using specific antibodies, and the demonstration of depth penetration effects

Ultraviolet irradiation of skin induces various DNA photolesions. Here we demonstrate that irradiation of human buttock skin with 300 nm UVR in situ induces thymine dimers and 6-4 photoproducts. Irradiation with 260 nm immediately followed by UVA (320 nm) induces the Dewar photoisomers of the 6-4 lesions. All three lesions can be detected in methanol-fixed paraffin sections using specific monoclonal antibodies. The sections have been analysed in an automated image analysis system (Discovery) and the level of immuno-DAB-peroxidase measured in individual epidermal cell nuclei as absorption at 460 nm (integrated optical density). The staining patterns with the antibodies showed no detectable change with epidermal depth by eye after 300 nm irradiation, however, the machine detected a fall off with depth of about 2.5% per cell layer. Following irradiation with a shorter wavelength (260 nm) there was a rapid fall off in staining with depth easily detectable by eye and machine (39% per cell layer).

DNA Damage in UV-irradiated Human Skin in Vivo: Automated Direct Measurement by Image Analysis (Thymine Dimers) Compared with Indirect Measurement (Unscheduled DNA Synthesis) and Protection by 5-methoxypsoralen

The incidence of the various types of skin cancer in the general population has been increasing at an annual rate of 2-8% over the past 2 decades. In spite of considerable media coverage on the risk of skin cancer the acquisition of a suntan is still very popular. Thus the UV exposures required for tanning pose a serious carcinogenic risk, particularly to individuals who tan poorly. On the other hand, the presence of natural skin pigment, or the ability to tan easily can protect the skin against some of the harmful effects of subsequent UV exposures (Kollias et al. 1991). We have recently shown (Young et al. 1991) in human volunteers, using an indirect measurement of DNA damage (unscheduled DNA synthesis (UDS) detected by autoradiography), that a tan induced by UV in the presence of a UVB sunscreen (Parsol MCX) preparation containing 5-methoxypsoralen (5-MOP) is more effective at protecting the skin against a subsequent DNA-damaging challenge dose of UV than a tan induced by UV alone, particularly in individuals who tan poorly. No such protection was seen with the same sunscreen lacking 5-MOP. 5-MOP is an ingredient in natural citrus oils and in many other plants. Here we show the same pattern of protective action when measuring, for the first time, DNA damage directly using a monoclonal antibody to thymine dimers (a major category of DNA lesion induced by UV radiation) on fixed human skin sections and automated image analysis. There is a good correlation between UV exposure dose and the levels of thymine dimers in epidermal nuclei. The levels of thymine dimers (measured as absorption by the mean integrated optical density (IOD)) also correlated well with the levels of UDS (grains per nucleus). These findings are of importance in the comparative risk-benefit assessment of sunscreens with and without 5-MOP. The techniques described have applications for measuring other DNA lesions following UV and other exposures.

Epidermal p53 response and repair of thymine dimers in human skin after a single dose of ultraviolet radiation : Effects of photoprotection

A cellular p53 response, DNA repair enzymes and melanin pigmentation are important strategies utilized by skin keratinocytes against impairment caused by ultraviolet radiation (UVR). In this study a double-immunofluorescence technique was used to investigate UVR-induced thymine dimers and p53 protein simultaneously. Four healthy volunteers were irradiated on both sides of their buttock skin with a single dose of solar-simulating UVR. One side was pretreated with a topical sunscreen. Biopsies from different time-points were immunostained for visualization of thymine dimers, p53 and proliferation. One single physiological dose of UVR generated widespread formation of thymine dimers throughout the epidermis 4h after irradiation. The level of thymine dimers decreased over time and was followed by a p53 response in the same cells. A late proliferative response was also found. The formation of thymine dimers, the p53 response and the late proliferative response were partially blocked by topical sunscreen. Large inter-individual differences in the kinetics of thymine dimer formation and repair as well as in the p53 response were evident in both sunscreen-protected and unprotected skin.

Small Intestinal Growth Regulatory Factors Extracted by Simple Diffusion from Intact Irradiated Intestine and Tested In Vivo

Following a dose of 8 Gy of gamma-rays delivered to the entire body of BDF1 mice, the proliferative activity in the crypts of the small intestine changes. The labelling and mitotic activity both fall precipitously, but in the lower regions of the crypt recovery from this fall begins soon after irradiation with cyclic fluctuations. Forty-five to fifty hours after irradiation, control levels are reached after which there is an overshoot. The number of clonogenic cells in the crypt shows a somewhat similar pattern of regeneration and overshoot. It has been assumed that these changes reflect the production of endogenous signals for proliferation and inhibition and these might be extracted by diffusion through the gut wall. We report here that at appropriate times after irradiation stimulatory and inhibitory extracts could be prepared. Appropriate in vivo assay techniques have been developed for testing inhibitors or stimulators making similar use of the patterns of proliferative regeneration after irradiation. Extracts prepared at either 15 h or 39 h after irradiation, i.e. during the phase of active regeneration are quite potently stimulatory on recipient animals 96 h after irradiation (i.e., following the decline from a proliferative overshoot) when injected twice 3 h apart. Extract prepared 72 h after irradiation (shortly after the overshoot peak) is strongly inhibitory when tested on unirradiated animals, or animals 90 h after irradiation, when injected four times 2 h apart. An accompanying paper shows that the stimulatory extract is powerfully active on intestinal cell lines. The in vitro approach is currently being used to characterise the stimulatory factor.

The time of onset and duration of 5-methoxypsoralen photochemoprotection from UVR-induced DNA damage in human skin.

Sites of previously unexposed buttock skin of eight human volunteers (skin type II) were treated daily for 3, 5, 8, or 10 days with suberythemogenic doses of solar‐simulated radiation (SSR) in the presence of a UVB sunscreen containing 5‐methoxypsoralen (5‐MOP) at 30p.p.m., or daily for 10 days with SSR+the same sunscreen without 5‐MOP. One week after cessation of treatment, these sites, together with a control unexposed site, were challenged with 2 minimal erythema doses (2 MED) of SSR. Biopsy samples were taken within 15 min of the challenge dose, and were incubated for 1 h in tritiated thymidine. UV‐induced DNA damage was measured indirectly by unscheduled DNA synthesis (UDS), and directly using a monoclonal antibody to thymine dimers, and automated image analysis. The level of pigmentation was assessed in sections in a semiquantitative fashion with Masson‐Fontana staining, and the number of layers in the stratum corneum was used to assess changes in epidermal thickness. Using the UDS and dimer measurements, the level of photochemoprotection afforded by 5‐MOP was determined from the reduction in the level of DNA damage observed. The photochemoprotection was expressed as a ratio of the 5‐MOP‐treated sites compared with the sites that did not receive 5‐MOP treatment.

A Potent Stimulator of Small Intestinal Cell Proliferation Extracted by Simple Diffusion from Intact Irradiated Intestine: In Vitro Studies

The epithelium lining the small intestine is one of the most rapidly proliferating body tissues yet it rarely develops cancers. The proliferation, migration and differentiation of the stem cell progeny appears to be under very strict control. After 8 Gy gamma irradiation the murine epithelium contains surviving stem cells from which the epithelium rapidly and effectively regenerates, presumably in response to stimulatory signals, and then returns to steady state conditions after overshoots in proliferation. Here we describe the isolation and preliminary characterisation in vitro of a potent stimulatory extract obtained by diffusion from intact murine small intestine, post-irradiation. In addition to in vivo responses the extract stimulates intestinal epithelial lines very effectively, most notably the rat IEC 18 line where it can replace the serum requirement. The extent of the induced increase in proliferation could not be reproduced by any other single growth factor tested. Preliminary evidence suggests the extract contains either a potent stimulatory cocktail of factors or a novel intestinal growth factor(s).

The quantitation and kinetics of unscheduled (repair) DNA synthesis in ultraviolet-irradiated human skin by automated image analysis.

Summary Grain counting by eye is a tedious and time‐consuming technique but one with great potential in cell kinetics and for the study of DNA excision repair activity (unscheduled DNA synthesis or UDS). We have been investigating the levels of UDS in human skin sections exposed in situ to ultraviolet radiation using a short‐term incubation in tritiated thymidine and autoradiography and the decline in UDS levels with time (repair Kinetics). We have adapted an automated image analysis system automatically to assess the number of grains over epidermal cell nuclei in autoradiographs of sections of epidermis. An excellent correlation was observed between visual counting and machine measurement of the area (in pixels) occupied by silver grains. The levels of UDS declined with time as lesions are progressively repaired. The half time (± standard deviation) for the reduction in UDS is 7.25± 0.18 h. The grain counts can be significantly increased by increasing the autoradio‐graphic exposure, by increasing the concentration of tritiated thymidine and by increasing the incubation time.

5-MOP Induced Protection Against Epidermal DNA Damage by Ultraviolet Radiation in Human Skin

It is now well recognized by the dermatological community that solar exposure results in an increased risk of skin cancer, especially in individuals of skin types I and II who sunburn easily and have a limited ability to tan (MacKie et al., 1987). In the last few years this information has been widely publicized in the press and other media. Despite this, a suntan is still a prized social asset and epidemiological data indicate a continued rise in the incidence of skin cancers. It also seems reasonable to suppose that the possible depletion of the stratospheric ozone layer, over populated areas of the earth’s surface, would result in a further long-term increase of skin cancer (van der Leun, 1988).