Search

We have reviewed the literature on cellular membrane radiobiology over the last ten years and, in particular, report on the development of rapid techniques used to identify damage soon after irradiation. It is clear that damage can now be quantified after low doses, and further refinements can be expected. From the work summarised, it would appear that changes to membranes at low doses may occur soon after damage to other important macromolecules by intercommunicating processes. We believe that there now exists a variety of rapid methods of measuring deposition of damage in vital macromolecules, such as cellular membranes and DNA, which can give a fuller picture of the overall effects of radiation and lead to predictions of eventual cellular mortality.

We have re-examined critically the evidence for and against the involvement of membranes in determining the response of cells to acute and chronic heat stress. Although frequently dismissed by many in the past, we believe that the bulk of evidence presented supports the view that physical and compositional alterations of membrane lipid components, both during and subsequent to heat exposure may, at least in part, account for cell adaptation, malfunction and lethality. Our primary goal in this review is to generate renewed interest in testing the validity of this hypothesis.

In Escherichia coli K1060 grown at 37 degrees C we observed that the uptake of both L-[3H]leucine and L-[35S]methionine was inhibited by exposure of the cells to 48 degrees C. The calcium channel blockers diltiazem and verapamil, and the anti-arrhythmic agent quinidine, inhibited the uptake of L-[3H]leucine at both 37 degrees C and 48 degrees C. Verapamil also inhibited the uptake of L-[35S]methionine at 37 degrees C, but at 48 degrees C protected against some of the heat-induced decrease in the uptake of this amino acid. The local anaesthetic procaine markedly inhibited the uptake of both labelled amino acids at temperatures between 37 degrees C and 48 degrees C. Amino acid uptake and cell killing were not correlated.

Ionizing radiation inhibited the development of specific haemolysin-producing cells (PFC) and depressed the incorporation of (3H) thymidine by rabbit spleen explants responding to SRC in the culture medium. In contrast to these effects, the rates of incorporation of precursors for protein and RNA synthesis were much less affected. The depression of (3H) thymidine incorporation was found to result from a quantitative reduction of new DNA synthesis, without any change in the proportion of labelled cells, at any time after irradiation. The DNA synthesis occurring in these cells preparing to develop antibody-producing capacity was thus radio-sensitive, but the exact nature of the defect resulting from exposure to radiation requires further study.

The proliferation of human blood lymphocytes from ageing donors, responding to concanavalin A, showed greater sensitivity to inhibition by X-rays than similar cells from younger donors. This increased sensitivity was associated with deficiency in repair of X-ray-induced damage to nuclear material, as measured by density in sucrose gradients, and with increased incidence of chromosomal damage following exposure of freshly isolated lymphocytes. There was also an increased frequency of spontaneous chromosomal aberrations in ageing subjects whose lymphocytes were deficient in repair of DNA damage.

Apart from cancer and mutation induction, radiobiological effects on mammals are mostly attributable to cell 'death', defined as loss of proliferative capacity. Survival curves relate retention of that capacity to radiation dose, and often manifest a quasi-threshold ('shoulder'). The shoulder is attributable to an initial mechanism of repair ('Q-repair') which is gradually depleted as dose increases. Another form of repair, which is not depleted ('P-repair'), increases the dose required to deliver an average of one lethal event per cell (dose 'D0'). Neither form of repair can unambiguously be linked with repair of defects in isolated DNA. An important initial lesion may well be disruption of the complex structural relationship between the DNA, nuclear membrane and associated proteins. One form of P-repair may be restoration of that structural relationship.

To test the influence of changes in membrane fatty acid composition on the radiation response of mammalian cells, human LDV cells were cultured in medium containing delipidated serum supplemented with either oleic or linoleic acid. Analysis of lipid extracts of the cells by gas liquid chromatography showed that, after 3 or more days growth in oleic or linoleic acid supplemented media, there were substantial overall increases in the proportion of oleate and linoleate, respectively, in the cellular lipid. Smaller absolute changes were measured in nuclear phospholipid, which was found to be low in unsaturated phospholipid compared to the rest of the cell. Fluorescence polarization measurements using diphenylhexatriene indicated an increased membrane fluidity in cells grown in the presence of excess linoleic acid and to a lesser extent for oleic acid. The clonogenic capacity of the cells after irradiation in air or nitrogen was not altered by any of these membrane compositional changes. The lack of effect on radiation sensitivity, in contrast to that reported for bacteria (E. coli K1060), is consistent with the fact that little or no change was brought about in the nuclear membrane composition, since evidence from partial cell irradiation experiments indicates that the cell nucleus is the sensitive target for cell killing by ionizing radiation. The ability of the cell to maintain a low level of unsaturated phospholipids in its nuclear membrane may be an important general defence mechanism against free radical damage to chromatin mediated by lipid peroxidation.

Acholeplasma laidlawii, a mycoplasma, is unable to synthesize unsaturated fatty acids but it will incorporate them into its plasma membrane if they are supplied exogeneously. Thus the fatty acid composition of the cell membrane can be defined by growing the organism in media containing specific fatty acids. We obtained cells with predominantly one type of unsaturated fatty acid (either oleic, linoleic or linolenic acid) or cells with only saturated fatty acid in the cell membrane. The cells were irradiated with 7 MeV electrons and the effect of membrane fatty acid composition on cell survival was examined. At 200 Gy/min and 0.5 degrees C (melting ice) there was little difference in the radiation sensitivities of the cells grown in unsaturated fatty acids either in aerated or anoxic radiation conditions. However, the cells containing saturated fatty acids irradiated in anoxic conditions were markedly more sensitive than the cells containing unsaturated fatty acids. At 200 Gy/min and 37 degrees C the two types of cells were of similar sensitivity both in aerated and anoxic radiation conditions. At 5 Gy/min at 0.5 degrees C the cells containing linolenic acid (18:3) were less sensitive than those containing solely saturated fatty acids. However, at 5 Gy/min at 37 degrees C there was no difference in sensitivity between these two types of cell. Our results strongly argue against the involvement of lipid peroxidation as a molecular change leading to cell death.

Evidence has been obtained for a specific protein receptor for prostacyclin on cells of the NCB-20 somatic hybrid. A new stable prostacyclin analog, 5-[(E)-(1S,5S,6R,7R) - 7 - hydroxy-6-[(E) - (3S,4RS) -3-hydroxy-4-methyl-1 -octen-6-inyl]bicyclo[3.3.0]-octan-3-ylidene]pentanoic acid (Iloprost, ZK36374) activates adenylate cyclase of NCB-20 cell membranes to an extent similar to prostacyclin and with a comparable high affinity. The binding of [3H]Iloprost to NCB-20 membranes was rapid with an association rate constant (k+1) of 2.01 X 10(5) M-1 s-1 at 20 degrees C. The rate constant for the dissociation of the ligand-receptor complex (k-1) was 1.19 X 10(-3) s-1, giving a dissociation constant (k-1/k+1) of 5.9 nM. The equilibrium dissociation constant was 29.9 nM, and the membranes had a maximum binding capacity of 347 fmol mg-1 protein. Radiation inactivation has been employed to determine the molecular weights of the functional prostacyclin receptor and components of the adenylate cyclase system in the plasma membrane of the NCB-20 cells. Cell membranes were lyophilized prior to irradiation, which lead to the formation of high-molecular-weight aggregates. The aggregation was avoided, however, when membranes were prepared in an isotonic Tris-HCl buffer containing sucrose. Molecular weight values of 111,000 for the catalytic subunit of adenylate cyclase, 89,000 for the regulatory subunit, and 83,000 for the prostacyclin receptor were obtained. Loss of [3H]Iloprost binding capacity after irradiation of lyophilized membranes yielded a molecular weight value (mean +/- S.E.) for the prostacyclin receptor of 82,800 +/- 12,900 (n = 3).

Explants of rabbit spleen, responding to sheep erythrocytes in vitro, were exposed to ionizing radiation at different times in culture. Radiation was found to depress the development of specific haemolysin-producing cells (PFC) at subsequent times, while cells already engaged in haemolysin production were relatively resistant to radiation. The depressing effects of radiation on PFC was most marked at times when the rate of increase of PFC was highest. Recovery from low doses (less than 2 Gy) occurred while higher doses of radiation had persistent depressing effects on the level of PFC. Assessment of synthesis of other macromolecules in these irradiated cultures showed that protein synthesis was highly resistant and DNA synthesis was sensitive like the immune response. Irradiation in hypoxic conditions showed that this protected the tissues from damage by radiation and there were also indications of enhancement of repair of this damage. Finally it was found that the results obtained with these cultures correlated well with previous studies of intact mice.

In recent years several lines of evidence have indicated that nuclear ADP-ribosyltransferase (ADPRT) activity is involved in DNA repair, although the requirement of ADPRT activity for cell survival has only been demonstrated in certain cases. We have investigated further the possible role of ADP-ribosylation in the response of cells to ionizing radiation, using 3-acetamidobenzamide (3-AAB), a highly effective inhibitor of ADPRT. With this compound we have demonstrated that marked enhancement of cell killing is observed if ADP-ribosylation is inhibited to a sufficient extent during the post-irradiation repair period, using concentrations of 3-AAB which are not toxic towards unirradiated cells. The critical period within which the inhibitor was effective was the first 90 min post-irradiation, and the half life of the recoverable radiation damage involved was estimated as approximately 20 min. Treatment with 3-AAB slowed the rate at which DNA strand breaks were repaired but did not prevent the ultimate repair of breaks, within the limits of resolution of the alkaline unwinding method used to determine DNA strand breakage. Although the majority of breaks were ultimately repaired, the frequency of radiation-induced sister chromatid exchanges and chromosome aberrations was increased, indicating that more genomic rearrangement had taken place, possibly as a consequence of the persistence of breaks when ADPRT activity was inhibited by 3-AAB during the post-irradiation repair period. It is suggested that the increased frequency of transposition and recombination which these observations reflect is likely to be associated with an increased risk of lethal mutations, some possibly involving chromosomal aberrations.

DNA-membrane complexes were isolated from lysed E. coli B/r and Bs-1, either by low g forces from a low salt solution, or by high g forces through a discontinuous sucrose gradient. The latter method was more gentle. Irradiation of the intact bacteria had no effect on the membrane macromolecules or on RNA components of these complexes. DNA loss was not significant after irradiation under anoxic conditions but complexes isolated from from Bs-1 irradiated in air showed an appreciable decrease in DNA content. In the presence of the appropriate nucleotide mixture, both 'free' DNA, found in the supernatant fractions, and rapidly sedimented membrane-associated DNA were able to synthesize DNA in the absence of added polymerase. DNA synthesis associated with 'free' DNA was more sensitive to radiation than that associated with DNA bound to the membrane, which appeared to moderate the effects of radiation on new DNA synthesis. It is concluded that the depression of DNA synthesis is primarily a result of irradiation-induced changes on genome-DNA. The interpretation of earlier work from our laboratories that DNA-membrane complexes contained the macromolecular structure which responded to radiation with a high o.e.r. is not supported by the evidence in this work.

On challenging with 2M NaCl, the nuclei of human lymphocytes yield an aggregate of DNA-protein material. The density of the material is less when isolated from irradiated cells than when isolated from unirradiated cells. The density of this material, designated histone-free-DNA (HF-DNA), from irradiated cells returns to that from unirradiated cells if the irradiated cells are allowed time at 37 degrees C in nutrient conditions. Lymphocyte HF-DNA from patients who have exhibited hypersensitivity to radiotherapy exhibit slower repair characteristics than lymphocyte HF-DNA from the average normal subjects. Neutrons are more effective than X-rays in producing density changes. Misonidazole and the ADPRT inhibitor 3-AAB substantially inhibit return to unirradiated densities. The oer for the initial effect is between 2 and 3. These properties of this DNA material from human lymphocytes suggest that alterations in the configuration of such material by the loss and re-establishment of relatively weak cellular bonds are closely related to the well-known changes observed in classical cell survival experiments. Where the proliferation of human lymphocytes has been observed by concanavalin A stimulation, oer, RBE and chemical modification has been of the same order as the measured density changes. Additionally, we have found that the density of HF-DNA is heavily dependent upon Cu content. This has led us to propose that cell killing or malfunction at the nuclear level caused by ionizing radiation is caused by the conversion CuII----CuI and also by specific .OH attack on DNA or proteins at a Cu site.(ABSTRACT TRUNCATED AT 250 WORDS)

Two murine lymphoma cell lines L5178YS and L5178YR were exposed to heat at 44 degrees C. In comparison with other cell lines there were found to be among the most sensitive to the lethal action of heat studied to date. The difference between these cell lines in response to heat was very small when compared to the differences observed with ionizing radiation. The pattern of leakage of radiolabelled leucine at 44 degrees C was measured in each cell line and small changes compared with the behavior at 37 degrees C were observed. No net breakdown or loss of preformed protein was observed during heating nor did any labelled protein leak from the cells.

Dispersions of a pure unsaturated phospholipid, dilinoleoylphosphatidyl choline, formed conjugated diene hydroperoxides when irradiated in air with 7 MeV electrons (150 Gy and 300 Gy). Peroxide formation was optimized when the dispersions were irradiated in air at 37 degrees C at a dose rate of 5 Gy/min. No significant loss of linoleic acid from the irradiated phospholipid dispersions was observed after doses of 150 or 300 Gy. Small amounts of thiobarbituric acid-reactive material were formed in irradiated unsaturated phospholipid dispersions. However, lipids or membranes isolated from 48 hour cultures of Acholeplasma laidlawii grown in media supplemented with either linoleic or linolenic acid did not appear to be peroxidized by irradiation under the same conditions.

The sensitizing effect of copper solutions has been studied on mammalian cells. A non-toxic concentration of cupric chloride was reduced under anoxia by irradiation to cuprous chloride, added to anoxic murine leukaemic L5178Y cells and irradiated either under aerobic or anoxic conditions. Sensitization (dose modification factor 1.54) was observed when the cells were irradiated in anoxia (oxygen-free nitrogen gassing). Sensitization was not observed when the cells were irradiated under aerobic conditions (air gassing). Electron spin resonance studies indicated that the conversion Cu II vector Cu I could be brought about by cells incubated under anoxic conditions at room temperature without irradiation.

The effect of rifampicin after irradiation with electrons or neutrons on cell-killing, DNA degradation and the ability of DNA-membrane complex extracted from irradiated bacteria to synthesize DNA was studied in strains E. coli B/r and BS minus 1. Post-irradiation treatment with rifampicin greatly enhanced the lethal effects of both electrons and neutrons on the repair-proficient strain E. coli B/r under anoxic conditions. However, the same post-irradiation treatment of the repair-deficient strain E. coli BS minus 1 with rifampicin gave no consequent increase in lethality. Rifampicin did not alter the extent of DNA degradation in either strain after irradiation with electrons or neutrons. The DNA-synthetic activity of DNA-membrane complex from electron-irradiated E. coli B/r and BS minus 1 was not altered by post-irradiation treatment with rifampicin, nor did this treatment affect synthesis by DNA-membrane complex isolated from neutron-irradiated BS minus 1; but the complex, isolated from E. coli B/r which had been irradiated under anoxic conditions with neutrons, showed a marked decrease in activity. We suggest that post-irradiation sensitization with rifampicin of the repair-proficient E. coli B/r, especially after irradiation under anoxic conditions, indicates that initiation of the synthesis of an RNA molecule is involved. In addition our results suggest that the structure of the membrane complex in E. coli B/r and BS minus 1 is different.

Little is known of the localisation of injected Corynebacterium parvum. We therefore developed a method of radiolabelling this vaccine and determined its distribution in the mouse. Live C. parvum (Wellcome, strain CN 6134) was grown in the presence of 3H-thymidine and subsequently killed by formalin. A high activity, 1-6 x 10(5) cpm/0.1 ml of a 7 mg dry weight/ml concentrated suspension, was obtained. Its biological properties (hepatosplenomegaly and antitumour effects) were similar to those of commercially available vaccine. After intravenous or intraperitoneal injection into normal mice, high activity was recovered in liver and moderate activity in spleen, lungs and small gut. In contrast, after subcutaneous injection, most activity was recorded at the injection site, and little in other tissues. A similar distribution of labelled C. parvum was found in tumour-bearing mice. Only moderate counts were detected in tumour.

E. coli Bs-1 cells were exposed to acute doses of tritium-beta particles by suspension in tritiated water for known lengths of time. The resulting survival rate was compared with that obtained for external irradiation with 7 MeV electrons. The o.e.r. measured for tritium-beta s was not significantly different from the value of 2.15 measured for 7 MeV electrons. The r.b.e. of the tritium beta s relative to 7 MeV electrons was 1.21 in both air and nitrogen. These results were compared with existing data for low voltage electron irradiations and with track segment studies of the effect of varying LET on the radiosensitivity of E. coli Bs-1.

We have recently shown that 3-acetamidobenzamide (3-AAB), a highly effective inhibitor of ADP-ribosyl transferase (ADPRT), can act as a post-irradiation (electrons) sensitizer on the mouse lymphoma cell lines L5178Y R and S. We have now shown that this compound sensitizes human derived skin fibroblasts but to a lesser extent. Fibroblasts derived from normal, Friedreich's ataxia, and ataxia-telangiectasia individuals were equally sensitized by 3-AAB to electron radiation. 3-AAB was also effective in sensitizing the mouse lymphoma lines to fast neutron irradiation. In addition DNA strand break repair was retarded as had been found after electron irradiation. 3-Nitrobenzamide is structurally a potentially dual action radiation sensitizer with electron affinic and ADPRT inhibitory properties. It is a weaker inhibitor of ADPRT compared to 3-AAB, and results in a smaller sensitization of mouse lymphoma cells in air. However, a much greater sensitization is achieved in anoxia. This greater sensitization appears to be a synergistic rather than an additive combination of its electron affinic and ADPRT inhibitory properties.

The extent of strand breakage and repair in irradiated E. coli B/r and Bs-1 was studied using a DNA-unwinding technique in denaturing conditions of weak alkali. Although these two strains show widely different responses to the lethal effects of ionizing radiation, they both have an equal capacity to repair radiation-induced breaks in DNA. Oxygen enhancement ratios for the killing of B/r and Bs-1 were respectively 4 and 2; but after repair in non-nutrient or nutrient post-irradiation conditions, the oxygen enhancement values for the residual strand breaks were always the same for the two strains. The equal abilities of E. coli B/r and E. coli Bs-1 to remove the strand breaks measured by this weak-alkali technique leads us to suggest that some other type of damage to either DNA or another macromolecule may play a major role in determining whether or not the cells survive to proliferate.

E. coli K1060 a fatty acid auxotroph unable to synthesize unsaturated fatty acids was used to study the effect of membrane fluidity on survival and DNA damage after exposure to ionizing radiation. Oleic and elaidic acids were incorporated into the bacteria and the temperatures of irradiation chosen in order to give the maximum difference in survival between the "liquid crystal" and "gel" states. Maximum sensitization was achieved by cooling the bacteria for 5 min at ice temperature before irradiation. No sensitization was observed if cooling was after irradiation. Under non-nutrient conditions both oleic acid and elaidic acid grown bacteria were able to repair DNA strand breaks to the same extent. When irradiation was in nutrient conditions both bacteria at all temperatures of irradiation repaired DNA strand breaks equally well and to a greater extent than in non-nutrient suspension.

Post-irradiation changes in DNA synthesis and ADP-ribosyltransferase (ADPRT) activity in L5178YS and L5178YR, radiation sensitive and resistant murine lymphoma cells are described. DNA synthesis was inhibited to a greater extent in L5178YS than in L5178YR cells. The stimulation of ADPRT activity by irradiation was not significantly different between these two cell lines. These observations contribute to other evidence which has failed to confirm a general association of ADP-ribosylation with the DNA synthesis inhibition response. The contrast between the response of L5178Y cells and the corresponding behaviour of ataxia telangiectasia cells and normal human cells indicate that entirely different mechanisms are involved in determining the differences in radiosensitivity in these two systems.

Considerable exposure to radiation always causes posterior subcapsular cataract (PSC). This investigation was conducted to ascertain whether cellular hypersensitivity to radiation may be identified as a possible cause of cataract in persons exposed to low levels of radiation. Patients were studied in whom PSC had followed probable exposure to low levels of radiation or in whom PSC had developed before the age of 60 without known exposure. The patients with cataract were compared with age and sex matched controls without cataract. Radiation sensitivity was estimated by measuring clonal growth of skin fibroblasts and peripheral blood lymphocytes after exposure to graded doses of radiation and by measuring postirradiation reconstruction of separated nuclear material from lymphocytes. The results show variations in the level of radiation sensitivity between the patients, without significant differences from the controls. It is concluded that radiation hypersensitivity, as tested by the methods used in this study, is not normally associated with the development of posterior subcapsular cataract.

Corynebacterium parvum was labelled by growing live bacteria in the presence of [3H]thymidine. The bacteria were killed by formalin, washed thoroughly and resuspended at a concentration of 7 mg dry weight/ml. An activity of 1-6 X 10(5) ct/min/0-1 ml was obtained. The biological properties (inhibition of tumour growth and hepatosplenomegaly) of the labelled C. parvum were compared with those of commercially available vaccine, and were found to be similar. Labelled C. parvum was injected i.v., i.p., or s.c. into normal C57BL mice and the localization of activity determined at 4 h and 1,3,7 and 14 days after injection. After i.v. or i.p. injection, highest counts were recorded in the liver. Moderate activity was found in the spleen, lungs and small gut. After s.c. injection, the majority of radioactive label was detected at the site of injection and little found in other tissues. The distribution of injected C. parvum was also studied in mice bearing Lewis tumour, and was found to be similar to that in normal mice. Moderate amounts of labelled C. parvum were recovered from tumour. There appeared to be no relationship between the antitumour effect of C. parvum given by a particular route of injection and the concentration of C. parvum recovered from the tumour.

The proliferation of peripheral blood lymphocytes, cultured with Con A, can be inhibited by ionizing radiation. Lymphocytes from patients with conditions associated with autoimmunity, such as rheumatoid arthritis, systemic lupus erythematosus and polymyositis, are more radiosensitive than those from healthy volunteers or patients with conditions not associated with autoimmunity. The nuclear material isolated from the lymphocytes of patients with autoimmune diseases is, on average, lighter in density than the nuclear material from most healthy controls. This difference in density is not related to increased sensitivity to ionizing radiation but the degree of post-irradiation change in density (lightening) is proportional to the initial density, i.e. more dense nuclear material always shows a greater upward shift after radiation. The recovery of preirradiation density of nuclear material, 1 h after radiation exposure, taken as an indication of DNA repair, correlates with the radiosensitivity of lymphocyte proliferation (Con A response); failure to return to pre-irradiation density being associated with increased sensitivity of proliferative response. These results require extension but, taken with previously reported studies of the effects of DNA methylating agents, support the idea that DNA damage and its defective repair could be important in the aetio-pathogenesis of autoimmune disease.

There continues to be an oversimplification of the approach to correlate cellular lesions with radiation induced cell death. Both in the prokaryotic and eukaryotic cell the relationship between vital macromolecules such as DNA, RNA, membrane and proteins is not yet fully understood either in a structural or functional sense. These macromolecules are often closely associated and interdependent. In spite of these recognised relationships much work is still devoted to measuring relatively early changes induced only in the DNA molecule. However, at the present time the quaternary structure of DNA and its closely neighbouring macromolecules is becoming better defined, and disturbances in these vital interrelationships may prove to be the most important radiation lesions. In the attempts to relate identifiable radiation damage to cell malfunction several criteria must be applied. For instance, the measured lesions must exhibit sensitization, protection and shoulder changes in response to the variety of agents and conditions which produce these phenomena at cellular level. In addition the radiation doses employed to produce measurable change must be within the same dose range as those used to study cellular and tissue effects. In much of the published work these criteria have not been applied.