THE L5178Y SUBLINES: a SUMMARY of 40 YEAR STUDIES in WARSAW Irena Szumiel

ISSN 1425-7351

PL0601824

RAPORTY IChTJ. SERIA B nr 2/2005

THE L5178Y SUBLINES: A SUMMARY OF 40 YEAR STUDIES IN WARSAW Irena Szumiel

INSTYTUT CHEMII I TECHNIKI JĄDROWEJ INSTITUTE OF NUCLEAR CHEMISTRY AND TECHNOLOGY RAPORTY IChTJ. SERIA B nr 2/2005

THE L5178Y SUBLINES: A SUMMARY OF 40 YEAR STUDIES IN WARSAW Irena Szumiel

Warszawa 2005 AUTHOR Irena Szumiel Department of Radiobiology and Health Protection, Institute of Nuclear Chemistry and Technology

This report is an extended version of the paper: " 'The L5178Y sublines: a look back from 40 years. 1.General characteristics. 2. Response to ionising radiation", published in International Journal of Radiation Biology, vol. 81 (2005) by kind permission of Taylor & Francis (http://www. tandf.co.uk)

ADDRESS OF THE EDITORIAL OFFICE Institute of Nuclear Chemistry and Technology Dorodna 16, 03-195 Warszawa, POLAND phone: (+4822) 811 06 56, fax: (+4822) 81115 32, e-mail: [email protected]

Papers are published in the form as received from the Authors The L5178Y sublines: a summary of 40 year studies in Warsaw The purpose of this report has been to review and summarise the results of 40 year studies concerning the general characteristics and response to UVC radiation, hydrogen peroxide and ionising radiation of the pair of L5178Y (LY) sublines, LY-R and LY-S, that differ in sensitivity to various DNA damaging agents. Comparison of karyotypes shows a number of differences in the banding patterns. Differences are found in ion transport and the ganglioside pattern of the plasma membranes, as well as in the content and turnover rate of poly(ADP-ribose) polymers. Nuclear matrix proteins show a differential affinity to these polymers. A unique property of the pair of LY sublines is inverse cross-sensitivity to X-rays and hydrogen peroxide, with cross-sensitivities to hydrogen peroxide and UVC, as well as to UVC and a platinum complex (cwplatin analogue). Initial DNA damage and repair and various aspects of the cellular response were determined in cells damaged with these agents. The higher sensitivity of LY-R cells to hydrogen peroxide, as compared to LY-S cells, is causally related to the higher content of iron ions in these cells and less efficient antioxidant defence system. Sensitivity of LY-R cells to UVC radiation and platinum complexes is explained by impaired excision repair (the incision step is missing). The reviewed data on the response of LY sublines to ionising radiation point to the key importance of DNA damage repair and fixation for the ultimate fate of the irradiated LY cell. The cause of slow double strand break (DSB) repair in LY-S cells is not identified but the defect (in non-homologous end-joining - NHEJ) explains most features of the cellular response to irradiation, as compared to the repair-competent LY-R cells. The most prominent are: very high radiosensitivity of Gl cells, extensive poly(ADP-ribose) dependent damage fixation, long G2 arrest, considerable chromosomal damage seen as premature chromatin condensation (PCC) fragments and aberrations in metaphase cells. The main cause of radiosensitivity difference between LY sublines lays in DNA repair/damage fixation ability. At the level of damage corresponding to a comparable lethal effect, the type of death differs between LY sublines; LY-S cells die by apoptosis, whereas LY-R cells - by necrosis. This observation is consistent with differential expression of proteins that are pro- or anti-apoptotic. The prominent role of poly(ADP-ribosylation) in the response of LY-S cells apparently is connected with damage fixation, but is in contrast with other hypersensitive to X/y-radiation cell lines with DSB repair defects. Comparison of LY sublines also indicates that their antioxidant defence system is of considerable importance in their susceptibility to hydrogen peroxide, but much less so in radiosensitivity.

Podlinie L5178Y: podsumowanie 40 lat badań w Warszawie Celem raportu jest przegląd i podsumowanie wyników 40 lat badań pary podlinii L5178Y (LY), LY-R i LY-S, różniących się wrażliwością na różne czynniki uszkadzające DNA. Bada- nia te dotyczyły ogólnej charakterystyki oraz odpowiedzi na nadtlenek wodoru, promieniowanie UVC i promieniowanie jonizujące. Porównanie kariotypów wykazało szereg różnic w układzie prążków. Znaleziono róż- nice w transporcie jonów, składzie gangliozydów błon plazmatycznych, zawartości i obrocie polimerów poli(ADP-rybozy). Białka zrębu jądrowego wykazują zróżnicowane powinowactwo do tych polimerów. Unikatową właściwością pary podlinii LY jest odwrotna krzyżowa wraż- liwość na promieniowanie X i nadtlenek wodoru z jednoczesną krzyżową wrażliwością na nadtlenek wodoru i promieniowanie UVC, a także na promieniowanie UVC i kompleks platyny (analog c/jplatyny). W komórkach uszkadzanych tymi czynnikami charakteryzowano uszkodzenia początkowe DNA, ich naprawę oraz różne aspekty odpowiedzi komórkowej. Wyższa wrażliwość na nadtlenek wodoru komórek LY-R w porównaniu z LY-S spowodowana jest wyższą zawartością jonów żelaza w komórkach tej podlinii i mniej skutecznym układem obrony antyoksydacyjnej. Wrażliwość komórek LY-R na promieniowanie UVC i kompleksy platyny wyjaśniono, charakteryzując defekt w naprawie z wycięciem (brak nacięcia nici DNA). Omówione w raporcie dane na temat odpowiedzi podlinii LY na promieniowanie joni- zujące wskazują na kluczową rolę naprawy i utrwalania uszkodzeń DNA dla dalszych losów napromienionych komórek LY. Przyczyna powolnego łączenia pęknięć podwójnoniciowych DNA w komórkach LY-S nie została do końca zidentyfikowana. Obserwowany defekt w ukła- dzie niehomologicznego łączenia pęknięć wyjaśnia jednak większość różnic w odpowiedzi na promieniowanie w porównaniu z komórkami LY-R, posiadającymi sprawny układ naprawy. Najważniejsze cechy odpowiedzi komórek LY-S są to: bardzo wysoka promieniowrażliwość w fazie Gl, bardzo znaczne utrwalanie uszkodzeń zależne od poli(ADP-rybozylacji), długie opóźnienie mitotyczne, liczne uszkodzenia chromosomów widoczne jako fragmenty PCC (metoda przedwczesnej kondensacji chromatyny) i aberracje w komórkach metafazalnych. Główną przyczyną różnicy w promieniowrażliwości podlinii LY jest relacja między zdolnością naprawy i zakresem utrwalania. Przy poziomie uszkodzeń dającym porównywalny skutek letalny, typ śmierci komórkowej różni się w podliniach: komórki LY-S umierają śmier- cią apoptotyczną, zaś LY-R - śmiercią nekrotyczną. Obserwacja ta odpowiada zróżnicowanej ekspresji białek pro- lub anty-apoptotycznych w obu podliniach. Znacząca rola poli(ADP-rybozylacji) w odpowiedzi komórek LY-S na promieniowanie najprawdopodobniej związana jest z utrwalaniem uszkodzeń; cechy tej jednak nie wykazują inne nadwrażliwe na promieniowanie X/y linie komórkowe z defektem naprawy pęknięć podwójnoniciowych. Z porówna- nia podlinii LY wynika też, że sprawność ich systemu obrony antyoksydacyjnej jest bardzo ważna we wrażliwości na nadtlenek wodoru, znacznie mniej natomiast - w promieniowrażli- wości. CONTENTS

1. INTRODUCTION 7

2. GENERAL CHARACTERISTICS 8 2.1. Cell cycle 8 2.2. Karyotype 8 2.3. Telomeres 9 2.4. DNA domains 9 2.5. Plasma membranes 9 2.5./ Lipid composition 10 2.5.2. /OH content and transport 10 2.5.3. Other transport systems 11 2.5.4. Osmotic fragility 12 2.5.5. Membrane potential 12 2.6. Arachidonic acid metabolism 12 2.7. Sensitivity profiles 12 2.8. PoIy(ADP-ribosylation) 13

3. RESPONSE TO HYDROGEN PEROXIDE 16 3.1. DNA damage related to iron content 16 3.2. Antioxidant defence system and NF-KB translocation 17

4. RESPONSE TO UVC RADIATION 19 4.1. Survival and DNA repair 20 4.2. Mutagenesis 21 4.3. Effects of combined UVC and benzamide or caffeine treatment 21 4.4. UVC-induced alterations in chromatin organisation 22

5. CELLULAR RESPONSE TO IONISING RADIATION 23 5.1. Cell cycle: progression and radiosensitivity fluctuations 24 5.2. Post-irradiation poly(ADP-ribosylation) 26 5.3. Post-irradiation metabolism of arachidonic acid 27 5.4. Stimulation of tyrosine kinase activity 28 5.5. Apoptosis 30 5.6. DNA damage induction and repair 32 5.7. Mutagenesis 34 5.8. NHEJ defect in LY-S cells 34 5.9. DNA damage fixation 36

6. CONCLUDING REMARKS 38

7. REFERENCES 39

1. INTRODUCTION

The original L5178Y cells were obtained from a methylcholantrene-induced lymphoma in DBA2 mouse on 8 July 1951 by Law at the Yale University; this fact is noted in the line's "family name": L for lymphoma, followed by the date of isolation and Y for Yale. This seemed somewhat lengthy and about 20 years ago was shortened to LY. The original acronym, however, is always given at the beginning of each paper. The L5178Y-S (LY-S) subline is the first mammalian radiation sensitive cell line to be reported. It was isolated from the parental L5178Y line (later called LY-R) and characterised by Alexander and Mikulski [1]. In 1983 the general characteristics of both LY sublines and their relation to other L5178Y lines was published [2] together with a description of sponta- neous conversion of LY-R phenotype into that of LY-S under in vitro conditions. Activation and silencing of numerous gene groups are more plausible as a mechanism of this conversion than a single mutation in a pleiotropic gene, since the comparison of karyotypes ([3], O. Ro- siek, unpublished, A. Wójcik, unpublished) shows a number of translocations and differences in the banding patterns of LY-R vs LY-S. Both sublines carry a heterozygous inactivating mutation in Tp53 [4, 5]. References are omitted to other, better known members of the L5178Y family, such as the LY 3.7.2C Tk+/- line applied in mutagenesis tests, the radiosensitive Japanese M10 mutant, the L5178Y-S/S cells grown by J.T. Lett in simplified culture medium and exhibiting the lowest mean lethal dose among mammalian cells, and the radioresistant variants obtained by V.D. Courtenay in the United Kingdom and H.H. Evans in the USA. In the passing years many new radiosensitive mutants of mammalian cells were obtained and characterised (reviewed in [6]); they have been widely used in research and have expanded our knowledge of DNA repair mechanisms that act in mammalian cells damaged by ionising radiation. The L5178Y model was left behind in this respect. Forty years ago, Prof. P. Alexander gave the LY cells to Dr J.Z. Beer who then worked with Prof. Alexander at the Chester Beatty Institute in London. Thanks to him, the cells were taken to Warsaw and became a favourite object of studies at the Department of Radiobiology and Health Protection of the Institute of Nuclear Research (now Institute of Nuclear Chemistry and Technology). Apart from the radiation sensitivity difference, the L5178Y model revealed another fascinating feature: the radiation resistant LY-R subline was more UVC-sensitive and hydrogen peroxide sensitive than the radiation sensitive LY-S subline. This unique sensitivity pattern is summarised in Table 1.

Table 1. The relative sensitivity pattern of two L5178Y murine lymphoma sublines, LY-R and LY-S. Relative sensitivity Subline Relative sensitivity to UVC to X/y-rays radiation Relative sensitivity to H2O2 LY-R Low High High LY-S High Low Low

The molecular and cellular mechanisms of the above mentioned sensitivity differences in LY sublines to these and other DNA damaging agents have been examined for more than 40 years in the Warsaw laboratory and elsewhere. This anniversary makes a good occasion to stand back from the current work to look at the whole picture, and to view biological features of this pair of sublines as reflecting the functions of interacting and interconnected subcellular systems. Although the experiments carried out with the old-fashioned techniques are now treated with some reservations, they brought solid facts which can still be useful for under- standing the cellular response in molecular terms. In spite of the lack of adequate genetic characteristics of the sublines, the time passed on examination of their phenotypes has not been lost. Knowledge of a gene expression profile is less valuable without information of its phenotypic reflection. Gene function often depends on the cellular context, a rather vague notion which should be better understood. One way to learn it, is to gather information on the cell's properties, even those that apparently are not connected with the main object of the study. The function of a single gene product may depend on the cellular context. Such an example are the murine cell mutants, M10 and LX830. They both are mutated at the xrcc4 locus, but the knowledge of this molecular defect does not explain why do they differ considerably in sensitivity to alkylating agents [7]. Here, I summarise the present knowledge gained in the Warsaw laboratory on the LY cellular model, with emphasis on radiation sensitivity in the last part of this report. Many observations are included that were made in other laboratories, usually with the contribution of Polish visiting scientists.

2. GENERAL CHARACTERISTICS

Cells and nuclei of both LY sublines are of similar diameter (about 15 and 9 um, respectively) and contain the same average amount of protein (1.25xlO"7 mg/cell) and DNA (0.9x10" mg/cell) in a population in the exponential growth phase [2, 8]. LY sublines differ in the ability to form tumours in non-immunosuppressed DBA2 mice: intraperitoneally injected LY-R cells easily form ascitic tumours, whereas LY-S cells form solid tumours with low frequency and after inoculation with a very high number of cells. LY-S cells are stable in vitro in contrast with LY-R cells. Therefore, the latter have been constantly passaged in DBA2 mice and trans- ferred into suspension culture at regular intervals [2]. Most experiments with LY sublines were carried out in Fischer's medium supplemented with bovine serum; only in the last few years RPMI 1640 medium and foetal calf serum were introduced. To form clones in soft agar, LY-R cells needed medium supplementation with sodium pyruvate and mercaptoethanol [8], consistently with the weak antioxidant defence (discovered much later [9]).

2.1. Cell cycle

For LY-R cells (generation time - 12.6 h) analysis of the kinetics of formation of labelled mitoses gave the following values: Gl - 3.0 h, S - 9.2 h, G2 - 1.4 h, M - 0.5 h. For LY-S cells (generation time - 10.3 h), Gl and S phases were shorter than those for LY-R cells (Gl - 1.6 h, S - 6.8 h). The duration of the G2 and M phases was comparable in both sublines (G2 - 1.5 h, M-0.5h)[8].

2.2. Karyotype

Karyotype analysis was carried at the Florida Institute of Technology 20 years ago [3]. A similar analysis was carried out at the same time in Warsaw, but the results were never published. Comparison of the banding patterns revealed differences between LY sublines and the DBA2 mouse cells. The modal chromosome number in LY-S cells is 39, including 32 identified and 7 marker chromosomes. There is only one copy of chromosomes 4, 5, 10, 18, 19 and X. Chromosome 15 is missing. There is a number of minor translocations and deletions, as well as a stable Robertsonian translocation involving chromosomes 1 and 6. In contrast with the rather stable LY-S karyotype, LY-R cells exhibit variability in the 10 marker chromosomes. The modal chromosome number in LY-R cells is 41. There is only one copy of chromosomes 1, 5, 9, 12, 13, 15 and X. Chromosome 19 is missing. Both sublines differ from DBA2 cells by numerous rearrangements and more subtle alterations in band width.

8 Karyotype was also analysed in L5178Y-S/S cells at the Colorado State University [10]; the symbol S/S designates a variant of LY-S cells cultured in a simplified medium containing some bovine plasma protein fractions instead of the full serum. In contrast with the report quoted above [8], there, the Y chromosome was present, harbouring a large deletion.

2.3. Telomeres

Another difference between LY sublines, discovered more recently [11], consists in the average telomere length: about 7.1 kb (LY-S cells) and 48 kb (LY-R cells), in spite of equal telomerase activity. The difference may be related to some aspect of DNA repair defect in LY-S cells, as discussed by [11]. Figure 1 shows a view of metaphase LY cells stained with fluorescent probe to telomeric DNA. A similar impaired regulation of telomere length has recently been described in human HCT116 Ku86+/" cells [12] showing that inactivation of even a single allele of Ku86 can result in a considerably disturbed maintenance of telomeric DNA.

Fig.l. Metaphase LY-R and LY-S cells stained with fluorescent probe to telomeric DNA. Normally, each chromosome contains 4 discrete spots where the probe is bound. In LY-S cells the signal is much weaker than in LY-R cells and even absent in some chromosomes (courtesy of A. Wójcik, unpublished).

2.4. DNA domains

The sizes of DNA supercoiled domains were estimated using sedimentation of nucleoids from cells irradiated with doses from 1 to 7 Gy. These sizes were 2.44x109 and 5.13xlO8 Da for LY-R cells and 1.30xl09 and 4.07xl08 Da for LY-S cells. Higher radiosensitivity related to larger size of the DNA supercoiled domains was suggested by Filippovich et al. [13] for large and small murine thymocytes. However, no simple relation of this kind was found between the sizes of DNA supercoiled domains and the susceptibility of LY sublines to ionising radiation [14]. Nevertheless, a difference in DNA "packaging" between LY sublines was inferred from analysis of ionising radiation-induced DNA damage with the use of chromatin conformation-sensitive methods (described below).

2.5. Plasma membranes

Plasma membrane properties of LY sublines were characterised in an attempt to find differences that could be related to the differential response to X/y-radiation [15]. No such relationship was found, although some differences between sublines were apparent. Nevertheless, these properties may be pertinent to possibilities of modification of the cellular response by combination of radiation and drugs.

2.5.1. Lipid composition Fatty acid analysis showed similarities: the saturated to unsaturated fatty acid ratio in total lipids was 0.77 ± 0.01 and 0.79 ± 0.02 for LY-R and LY-S cells, respectively. Total lipid to protein ratio (uM/mg) was 0.213 ± 0.008 (LY-R) and 0.234 ± 0.009 (LY-S), cholesterol to protein ratio (uM/mg) - 0.049 ± 0.004 (LY-R) and 0.059 ± 0.003 (LY-S) (mean values, standard error indicated). Membrane fluidity, as measured with fluorescence polarisation with 1,6-di- phenyl-l,3,5-hexatriene did not differ between sublines either in untreated cells or 3 and 24 h after X-irradiation (10 Gy - aerobic conditions, 27 Gy - hypoxia). Lipid-bound sialic acid content was 30.4 ± 1.7 (LY-R) and 45.2 ± 2.0 nanomoles/108 cells. There was a difference in the composition of neutral glycosphingolipids (more slowly migrating bands in thin layer chromatography for LY-R than for LY-S cells, the latter containing more quickly migrating monohexosylceramides [16]). The difference in the ganglioside composition is of interest because of their ability to regulate activity of growth factor receptors (reviewed in [17]). LY-S cells contain more gangliosides co-migrating in thin layer chromatography with monosialosylganglio-tetraosylceramide GM1, and disialosyltetraosylceramides, GDI a and GDlb, than LY-R cells (ganglioside nomenclature follows Svennerholm). This difference in ganglioside composition may affect ligand binding to growth factor receptors or their dimerisation [18-20]. Hence, in LY-S cells relatively weaker (as compared to LY-R) signalling events can be expected that are generated in lipid rafts in response to growth factors. The slightly higher cholesterol content in LY-S as compared to LY-R cells may add to this effect, by decreasing growth factor binding [21]. The antigens that cause rejection of intraperitoneally injected LY-S cells in DBA2 mice were not identified.

2.5.2. Ion content and transport Interesting differences were found in the cation transport system (Fig.2): Na+/K+ ATP-ase activity is two times higher in LY-S cells than in LY-R cells. This is also reflected in differential sensitivity to ouabain (Na+/K+ ATP-ase inhibitor) and amilorid (Na+/H+ antiport inhibitor) (unpublished). For ouabain, the difference in survival was the largest (eightfold) at 0.1 mM and disappeared in the concentration range 0.5-3.0 mM. For amilorid, Djo differed less (Fig.2B). Since the Na+/K+ pump functions at the expense of a large part of the cellular reserve of energy in the form of ATP, this difference may be important for cellular properties, like osmotic fragility (see below).

Table 2. Total (mean value and standard error) or free (mean value and standard deviation) ion content of LY sublines.

Ion LY-R LY-S Method and reference Sodium 0.0821 ±0.0156 0.0563 ±0.0164 Potassium Total content, 0.6397 ± 0.0594 0.6711 ±0.0437 uM/100mg X-ray microanalysis [22] Magnesium dry weight 0.0288 ± 0.0026 0.0385 ± 0.0035 Chloride 0.2670 ±0.0185 0.2617 ±0.0256 Free calcium, nM 148.6 ±25.2 150.8± 1.8 FURA-2 fluorescence [23] Labile iron pool, uM 0.57 ±0.17 0.18 ±0.07 Calcein fluorescence [24] 6 Total iron (nucleus), ng/10 cells 7.7 ±1.8 3.1 ±0.9 Atomic absorption Total copper (nucleus), ng/106 cells 10.2 ±3.4 15.1 ±1.2 spectrometry [25]

10 Ion content of both sublines was examined by X-ray microanalysis [22]; there was a difference in sodium and potassium content, as shown in Table 2. A

Sensitivity to cation transport inhibitors

E1D10OUABAIN, mM DD10AMILORID, mg/ml

LY-R LY-S

Cation transport ATP-ases in LY sublines

LY-R LY-S

Fig.2. Features of cation transport in LY sublines. (A) sensitivity to cation transport inhibitors, expressed as Dio - dose reducing survival to 10% (unpublished data). Ouabain, concentration indicated, continuous treatment until clone scoring; amilorid, concentration indicated, 4 h treatment followed by cloning. (B) comparison of activities of cation transport ATP-ases in LY sublines (data from [15]).

The ratio of their sum to chloride content was about 2.7 in both sublines. Free calcium ions were at a comparable level. Marked differences in iron and copper (both in that localised in the nucleus and in the labile iron pool) turned to be the critical determinant of susceptibility to oxidants, as discussed below.

2.5.3. Other transport systems Na+-dependent amino acid transport system was shown to be similarly efficient in untreated cells and 3 or 24 h after irradiation (10 Gy - aerobic conditions, 27 Gy - hypoxia) for both sublines [15]. The drug transport system (responsible for multidrug resistance) measured by the rate of efflux of rhodamine 123 in the presence or absence of verapamil was identical in these LY sublines [26].

11 2.5.4. Osmotic fragility Osmotic fragility measured by exposure to hypotonic salt solution (0.75N KC1) at 37°C was considerably higher in LY-S than in LY-R cells: the cell numbers were reduced by 50% after 120 min (LY-S) and 260 min (LY-R) [27].

2.5.5. Membrane potential Membrane potential was measured with the use of Rb and tritiated methyltriphenyl phos- phonium method, according to Scott and Nicholls [28]. It was 47.7 mV in LY-S and 56.6 mV in LY-R cells; y-irradiation (20 Gy) did not alter it after 1 or 3 h post-irradiation incubation [29].

2.6. Arachidonic acid metabolism

Incorporation of exogenous (14C-labelled) arachidonic acid was higher in LY-R than in LY-S cells, whereas release was similar and not affected by inhibition of protein kinase C. Calcium ionophore, A23187, which activates phospholipase A2, increased the release to a greater extent in LY-R cells but in both sublines was not necessary for activation of synthesis of arachidonic acid metabolites, prostaglandins and 5-hydroxyeicosatetranoic acid (5-HETE). Synthesis of the latter metabolite was equally stimulated by hydrogen peroxide 1-10 uM, but there was a higher level of radioactive label in HETE in LY-R than in LY-S cells (11.31 ± 0.90 vs 8.37 ± 0.56 picomoles/108 cells, mean values with standard error). These data indicate that the basic features of arachidonic acid metabolism are similar in LY sublines [30]. Differences after X-irradiation are described below.

2.7. Sensitivity profiles

Sensitivity of LY sublines to various damaging agents was examined with the aim of finding common features in the cellular responses to these agents and ionising radiation. The relative (LY-R/LY-S) sensitivities are shown diagrammatically in Figs 3 (LY-R cells more sensitive than LY-S cells) and 4 (LY-R cells equally or more resistant than LY-S cells). The numerical values, experimental details and references were published in [26, 31-34]. As discussed below, in some cases the reason for high sensitivity to a particular agent was identified. This concerns, among others, the sensitivity of LY-R cells to hydrogen peroxide, platinum complex, cis-PAD (c«-dichlorobis(cyclopentylamine)platinum II) and UVC radiation. The peculiar hypersensitivity of LY-R cells to cooling in D2O-containing medium was not further studied.

camptothecin hydrogen peroxide 92% D2O medium, OoC, 290 min colchicine hyperthermia A23187 hydroxyurea cis-PAD (Pt complex) Ledakrin (acridine derivative) UV-C UV-A

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Relative sensitivity (LY-R/LY-S)

Fig.3. Relative (LY-R/LY-S) sensitivity/resistance to various damaging agents to which LY-R cells are more sensitive than LY-S cells [26, 31-34].

12 Cross-sensitivity of LY-S cells to X/y-rays, bleomycin, topoisomerase II poisons, etopo- side (VI6) and ellipticine as well as alkylating agents is similar to that reported for other radiosensitive cell lines (reviewed in [6]). The reason for the high sensitivity of LY-S cells has been expected to be impaired DSB (double strand break) repair. Therefore, the higher susceptibility of LY-R than LY-S cells to topoisomerase I poison, camptothecin [32, 35], came as surprise, since this drug exerts its lethal effect due to generation of DSB in replicating DNA [36]. This relative sensitivity pattern was explained by a higher phosphorylation of topoisomerase I molecules in LY-R cells (the reason for higher susceptibility of the enzyme's molecules to camptothecin) than in LY-S cells [37] and by the efficient repair of camptothecin- -generated DSB by the homologous recombination repair system in the LY-S subline [38]. Furthermore, the extent of topoisomerase I phosphorylation was shown to be dependent on poly(ADP-ribosylation). Activity of poly(ADP-ribose) polymerase (PARP) was higher in LY-S

mitomycin C bleomycin ellipticine VP16 N-methyl,N-nitrosourea ethylmetane sulfonate dimethylmyleran nitrogen mustard Triton X100 novobiocin ethanol 2% misonidasole adriamycin erythromycin KCN

12 3 4 Relative resistance (LY-R/LY-S)

Fig.4. Relative (LY-R/LY-S) sensitivity/resistance to various damaging agents to which LY-R cells are equally or more resistant than LY-S cells [26, 31-34]. than in LY-R nuclei (see below). When the activity of PARP was inhibited by treatment of LY-S cells with benzamide, casein kinase 2-catalysed phosphorylation of topoisomerase I increased. This was accompanied by an increase in sensitivity to camptothecin, as reflected in the diminished viability of LY-S cells [39].

2.8. PoIy(ADP-ribosylation)

There are remarkable differences in the content and metabolism of poly(ADP-ribose) polymers between LY sublines. The content in LY-S cells is exceptionally high as compared to other rodent and human cell lines and tissues and three times higher than that in LY-R cells [40]. The content and turnover are presented in Fig. 5. The decrease in polymer content was measured in the presence of 2 mM benzamide which inhibited polymer synthesis by 96%. hi LY-R cells, 20% of polymers decayed with a half-life of 96 min and the rest with a half-life of 5.5 h. In LY-S cells, 77% of polymers decayed with a half-life of 15 min and the remaining fraction was not further degraded.

13 200-

~ 160- z a ? 120-

o a 80- o •o cc 40-

180 Time of benzamide treatment, min Fig.5. Turnover of constitutive poly(ADP-ribose) in intact LY-R (o) and LY-S (•) cells. (A) absolute content, expressed as sRAdo (l,N6-ethenoribosyIadenosine). Difference between LY sublines in the first three points is statistically significant (p<0.01). (B) semilogarithmic plots of the time courses of polymer degradation. Benzamide present at 2 mM concentration. Data points represent mean values ± standard error from two experiments with 3-5 separate determinations each. Reproduced by permission from [40].

Fig.6. ADP-ribose polymer size distribution of constitutive polymers from LY-R and LY-S cells. Re- produced by permission from [42].

14 PARP activity in the nucleus is about five (5.38) times higher in LY-S than in LY-R cells [39], but equal in both cell sublines when measured in permeabilised cells, where DNA degradation is extensive [41] thus, inducing maximal PARP stimulation. The substrate for PARP, NAD+, is present in equal concentrations in both LY sublines (36.7 ± 3.2 nanomoles/mg DNA - LY-R, 36.9 ± 3.1 nanomoles/mg DNA - LY-S) [42]. Figure 6 shows the size distribution of constitutive polymers: computer-aided densitometric scanning reveals more polymers in the class from 5 to 13 ADP-ribose residues and fewer long polymers (37 residues and larger) in LY-S cells as compared to LY-R cells [42]. This is compatible with the preferential degradation of larger polymers by poly(ADP-ribose) glycohydrolase. After labelling of intact cells with tritiated adenosine, the distribution of label that re- sulted from post-translational protein ADP-ribosylation was examined in acid-soluble nuclear proteins [43]. It was found mainly in low mobility group proteins and histone H2A1 in equal amounts in both sublines. Additionally, more label was found in LY-S cells in histone H2B and in LY-R cells in histone H2A2.

-66 4 ®i;!Jpi^18B _ 45 S. B -3J.4 Ml -21-' -14-

LY-S

Fig.7. ADP-ribose polymer blot analyses of nuclear matrix proteins isolated from LY-R and LY-S cells; equal quantities (20 ug) of protein were loaded on 10% SDS-polyacryamide gels and electro- phoresed. (A) Coomassie blue stain, positions of Mr markers indicated, Hist - histones; (B) polymer blot; (C) densitometric analysis of the blot shown in (B). Reproduced by permission from [42].

It should be noted that up to 90% of the polymers co-extract with nuclear matrix proteins [44]. So, it seems important that polymer blot analyses of isolated nuclear matrix proteins point to an almost twofold reduction in polymer binding affinity in LY-S cells as compared to LY-R cells. This is shown in Fig.7. The reason for this may be an altered expression of several proteins found in the nucleoid fraction, as pointed out by Kapiszewska et al. [45] and Malyapa et al. [46]. According to the latter authors, lack of certain proteins in this fraction accompanies high radiation sensitivity. The difference in matrix proteins between LY sublines may be pertinent to chromatin function and the biological expression of DNA damage.

15 3. RESPONSE TO HYDROGEN PEROXIDE

The LY sublines are inversely cross-sensitive to hydrogen peroxide and X-rays, as mentioned in the Introduction section and Table 1. Since with both damaging agents the main damaging species is the highly reactive hydroxy radical, this difference in sensitivity seemed worth closer examination.

3.1. DNA damage related to iron content

Hydrogen peroxide in the presence of transition metal ions (mostly iron in vivo) undergoes the Fenton reaction, that generates the highly reactive and damaging hydroxy radicals. So, the final effect of hydrogen peroxide treatment depends on the extent of this reaction, the ferrous ions available, and the supply of reducing equivalents from the ongoing metabolism that allow the recycling of iron ions. Figure 8 compares clonogenic ability and DNA strand breakage (the sum of single strand breaks, SSB, and DSB estimated by the alkaline comet assay) after hydrogen peroxide treatment (1 uM, 1 h) at 37 and 4°C [47]. It can be seen that the extent of metabolic activity (temperature-dependent) affected LY-R (sensitive) cells less than LY-S (resistant) cells.

Effects of hydrogen peroxide treatment

37 oC

0LY-S SLY-R 4oC

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Surviving fraction

B lillSt^i HHMHIfe warn 37 • HLY-S B i 0LY-R 4 mmi •iiiiiiiil ip mmmm ssi 0 5 10 15 a20 DNA damage, tail moment

Fig.8. Comparison of survival from clonogenic ability (A) and DNA strand breakage (the sum of breaks, SSB and DSB, estimated by the alkaline comet assay; (B)) after hydrogen peroxide treatment (1 uM, 1 h) at 37 and 4°C. Mean values ± standard error indicated. Data from Kraszewski et al. [47].

The survival data correspond with DNA damage, although it must be taken into consider- ation that repair is taking place during the treatment interval, especially at the higher tempera-

16 ture. Even with the limited metabolic supply of reducing equivalents at 4°C, the damage to DNA is very high in LY-R cells. A more detailed examination of DNA lesions (Fig.9) shows that both types of base damage (examined with the alkaline comet assay combined with specific endonucleases) are higher in LY-R cells than in LY-S cells [48]. Mutation frequency per 105 survivors at the hgprt (hypoxanthine: guanine phosphoribosyltransferase) locus after treatment (1 uM, 1 h) at 4°C is 2.4 ± 0.6 and 1.8 ± 0.1 in LY-R and LY-S cells, respectively. Treatment at 37°C induced 2.1 ± 0.3 (LY-R) and 3.4 ± 0.3 (LY-S) hgprt mutations per 105 survivors [47]. Here, the temperature effect is absent in the oxidant-sensitive LY-R cells, whereas present in the more resistant LY-S cells.

70 LY-R LY-S

60 •fl DNA breaks CZH ENDOIII sensitive sites § 50 FPG sensitive sites 40

0 CONTROL H2O2 CONTROL H2O2

Fig.9. DNA strand breakage and base damage estimated in LY cells treated with 25 uM hydrogen peroxide in phosphate buffered saline - PBS (1 h, 0°C) with the use of alkaline comet assay combined with endonuclease III (ENDOIII) or Fapy-DNA glycosylase (FPG). Mean values ± standard deviation indicated. Reproduced by permission from [48].

The data presented in Fig.8 were taken as indication of a differential availability of iron ions in the LY sublines. In LY-R cells, even the reduced metabolism-related recycling did not prevent generation of a considerable damage to DNA; therefore, it was concluded that the amount of iron must be very high as compared to LY-S cells. Indeed, both total iron content of the nuclei and the labile iron pool consisting of the chelatable iron ions that are available for Fenton reaction proved to differ as expected (Table 2). Consistent with this result was a higher protective effect of the iron chelator, desferroxamine, in hydrogen peroxide treated LY-R cells than in LY-S cells [49]. Further support came from the difference in the expression of the main proteins that control the cellular level of iron (transferrin receptor which is responsible for iron ion uptake and ferritin which sequestrates the ions in the cytoplasm). More details on iron content control in LY sublines can be found in [24].

3.2. Antioxidant defence system and NF-KB translocation

Further experimental data [9, 50, 51] point to a lower efficiency of the antioxidant defence system in the hydrogen peroxide sensitive LY-R cells than in the hydrogen peroxide resistant LY-S cells (see Table 3 for numerical data). In particular, activity of catalase is twofold higher and sensitivity to its inhibitor, AMT (3-amino-l,2,4-triazole) considerably lower in LY-S than in LY-R cells. The content of monobromobimane-reactive thiols is 54% higher in LY-S than in LY-R cells. On the contrary, activity of glutathione peroxidase (GPX) is about two times higher in LY-R than in LY-S cells; however, upon induction with selenium the activity increases 15.6 times in LY-R cells and as much as 50.3 times in LY-S cells [9].

17 Table 3. Activities of antioxidant defence enzymes in LY cells (mean values ± standard error indicated; data for total cell extracts from [9, 50,51]).

Enzyme LY-R LY-S Catalase (units/mg protein) 14.08 ±2.07 26.70 ± 2.68 Glutathione peroxidase (units x 103/mg protein) 5.16 ±2.28 2.49 ± 0.88 Glutathione peroxidase after induction with selenium (units x 103/mg protein) 79.78 ±15.36 126.64 ±16.37 Cu, Zn-superoxide dismutase (units/mg protein) 6.85 ±0.63 3.25 ±0.33

Pretreatment with selenium exerts a dramatic effect on DNA damage and growth of hydrogen peroxide treated cells. This is shown in Fig. 10. Alkaline comet assay applied to assess DNA damage shows the relative importance of catalase and GPX for cell protection. Inhibition of catalase with AMT has a negligible effect on the damage. In contrast, induction of GPX gives a pronounced protection, especially in LY-R cells. This is also reflected in the post-treatment growth (Fig. 10B).

B 1.0

0.5 ••

LY-S+Se, LY-S-Se,

LY-R+SeŁ LY-R-Se.A i 0.0 —t 0 12 H2O2 Fig. 10. Effects of GPX induction with selenium on the response to hydrogen peroxide. (A) DNA damage expressed as tail moment in LY cells following 1 h exposure at 37°C to hydrogen peroxide 50 uM, AMT (catalase inhibitor) and to combined simultaneous treatment with these agents (CTRL - control; AMT concentrations: LY-R cells - 5 mM, LY-S cells - 10 mM). Mean values ± standard error indicated. (B) Relative cell numbers (Nt/Nc, where Nt is the number of cells/ml in the treated cell culture, Nc is the number of cells/ml in the control, both grown from the same initial cell density during 48 h) in hydrogen peroxide treated LY cell cultures grown in Fischer's medium with or without selenium supplementation before the treatment. Mean values ± standard deviation indicated. Repro- duced by permission from [9].

18 One consequence of treatment with oxidants is activation of members of a Rel/NF-KB (nuclear factor KB) transcription factor family (reviewed in [52]). NF-KB is present in the cytoplasm as a heterodimer complexed to an inhibitory subunit, IKB. In consequence of specific stimuli the latter subunit becomes degraded. This is followed by translocation of the heterodimer to the nucleus which is related, among others, to the cellular response to oxidants. A differential induction is seen of p65-NF-KB nuclear translocation in LY-R and LY-S cells [53]: in LY-S cells there is a weaker nuclear translocation of p65-NF-KB subunit in response to hydrogen peroxide or to 1 Gy of X-rays than in LY-R cells. Other factors that act by generating reactive oxygen species (phorbol ester and lovastatin) also cause a weaker NF-KB translocation in LY-S than in LY-R cells. Altogether, induction of p65-NF-icB nuclear translocation corresponds to the antioxidant status, but is not related to the sensitivity of the LY subline to the inducing agent. So, the difference in sensitivity to oxidative stress between LY sublines depends primarily on differential iron homeostasis control and hence, iron content, especially that present in the form of the labile iron pool, potentially active in the Fenton reaction. Highly efficient generation of hydroxyl radicals in LY-R cells is the cause of higher DNA damage and in consequence, a more pronounced lethal effect of hydrogen peroxide treatment (at micromolar concentrations) in LY-R than in LY-S cells. Weaker antioxidant defence in LY-R cells further increases their susceptibility to oxidative stress as compared to LY-S cells.

4. RESPONSE TO UVC RADIATION

The LY sublines are cross-sensitive to hydrogen peroxide and UVC radiation while being inversely cross-sensitive to UVC radiation and X-rays, as mentioned in the Introduction section and in Table 1. The cross-sensitivity, however, does not stem from common mechanisms of sensitivity to these two damaging agents. As discussed below, the relative UVC sensitivity of LY-R cells as compared to LY-S cells is connected with DNA repair processes. The main features of the response of LY cells to UVC radiation are summarised in Table 4.

Table 4. Response of LY sublines to UVC radiation.

Characteristic feature LY-R LY-S Reference Sensitivity (cloning) High Low 2 2 - Do (exposure in Fischer's medium) 2.8 J/m 9.0 J/m [54] 2 2 - Do (exposure in PBS) 0.8 J/m 2.7 J/m Effect of divided exposure to UVC Sensitising Sparing [55] Excision repair (DNA strand incision) Absent Present [56-58] Percentage cells with chromatid aberrations 20 h 30 (10 J/m2) 26 (30 J/m2) [54] after equitoxic exposure 6 Mutation frequency at the D37 fluence per 10 surviving cells: - ouabain resistant 5.0± 1.1 8.4 ± 3.0 [59] - thioguanine resistant 129 ± 49 3.0 ±0.7 Sensitisation by 0.75 mM caffeine No Yes

Incorporation of tritiated thymidine: [60,61] - after exposure to 20 J/m2 UVC alone Equally inhibited - after exposure to 20 J/m2 UVC and 0.75 mM caffeine No effect of caffeine Sensitisation by 2 mM benzamide No Yes Incorporation of tritiated thymidine after exposure to 20 J/m2 No change as compared Inhibition [54] and 2 mM benzamide to UVC alone partly reversed

19 4.1. Survival and DNA repair

Table 4 and Fig.l 1A show that LY-R cells are considerably more UVC-sensitive than LY-S cells. Divided exposure of LY-R cells (Fig.l IB) results in a decrease in relative survival, with a gradual return to relative survival equal to unity for intervals between the first and second exposure from 5 to 28 h [55]. In contrast, the relative survival of LY-S cells after divided exposure remains stable (close to unity) for intervals 1-12 h and then increases to about 2 for intervals 12-24 h.

LY-R LY-S uv D n 1.o3 1.8 Do j/m* 2.8 9.O 10"' k- \ UV+Bz n 1.5 0,•9 Do J/nr"1 2.5 7. 1

10 :- L5178Y-R 32 erg. mm-2 or 16+16 erg. mrrr* Exp. A 529 1 . *. •/* J

7 _UV-Ughtj

L5178Y-S 214 erg mm"2 or

L-xp._A500_ i i.Ł^Łi.^ij.jujiiizrj 10 20 10 20 TIME INTERVAL HOURS

Fig.l 1. (A) Survival of UVC-irradiated (in Fischer's medium) LY cells determined by cloning with or without treatment with 2 mM benzamide (Bz) during the whole time of clone growth. For benzamide treated samples the results were normalised to their appropriated zero fluence controls. The data points are mean values of 2-4 experiments; standard error indicated when larger than the point plotted. Re- produced by permission from [54]. (B) Relative survival (i.e. survival related to that after single exposure, ordinate axis) after divided exposure to UVC fluences giving for undivided exposure 22% survival for LY-R cells and 6% survival for LY-S cells. Reproduced by permission from [55].

20 The high UVC sensitivity of LY-R cells found explanation in DNA repair studies carried out with three methods: nucleoid sedimentation [56], DNA unwinding combined with treatment with l-(3-D-arabinofuranosyl cytosine [57] and DNA sedimentation with the use of 5'-bromo- deoxy uridine photolysis [58]. All three methods gave consistent results that pointed to a low level of incisions in DNA from UVC-exposed LY-R cells as compared to LY-S cells. In other experiments, sedimentation in an alkaline sucrose gradient was carried out using LY cells exposed to 30 J/m2UVC and pulse labelled with tritiated thymidine 30 min after exposure [60]. In agreement with the apparent impaired excision of photoproducts in LY-R cells, it was found that the newly synthesised DNA in these cells had the same sedimentation profile as that from the unexposed (control) cells. It was concluded that no gaps were generated that might have been due to incomplete excision. In contrast, LY-S cells after exposure to UVC-synthesised DNA in fragments shorter than for the control, and this fragmentation was compatible with the ongoing excision. Treatment with 0.75 mM caffeine (applied for the whole interval between UVC exposure and clone scoring) markedly decreased survival of LY-S cells without sensitising LY-R cells [60].

4.2. Mutagenesis

Determination of UVC-induced mutations [59, 62, 63] gave results consistent with a repair defect in LY-R cells, since the mutation frequency in hgprt locus (measured as frequency of clones resistant to 6-thioguanine) was more than forty times higher in LY-R cells as compared to LY-S cells (Table 4); gene duplication in LY-S cells was excluded [62]. The resistance to Na+/K+ ATP-ase inhibitor, ouabain, was induced at a comparable frequency in both sublines [59]. Factors other than repair defect can contribute to differences in mutability in individual loci between LY sublines. A convincing analysis of mutations in LY sublines was done with the use of a polyomavirus-based shuttle vector; the target gene was supF [64]. This approach eliminated some of the confounding factors. Survival of transforming ability of plasmid in UVC-irradiated LY-R cells was lower and mutation frequency higher than in similarly treated LY-S cells, whereas the distribution of mutations was similar as in normal human and XPA (xeroderma pigmentosum, complementation group A) cells. In LY-S cells there were fewer tandem double mutations than in LY-R cells, consistent with more efficient excision of pyrimidine dimers, and there were fewer mutational hot spots in LY-S cells [64].

4.3. Effects of combined UVC and benzamide or caffeine treatment

The DNA repair experiments in UVC-exposed LY cells were not continued, and the molecular defect was not identified. Some indirect information on the repair defect in LY-R cells can be obtained from examination of the effects of benzamide and caffeine on UVC-exposed cells. Interestingly, the sensitivity pattern and the effects of benzamide and caffeine are similar in the case of UVC irradiation and treatment with an antitumour platinum complex, cis-?AD [8, 65]. Inhibition of poly(ADP-ribosylation) with 2 mM benzamide (applied for the whole interval between UVC exposure and clone scoring) sensitised LY-S cells without changing the survival of LY-R cells. The ratio of poly(ADP-ribose) polymers (quantified as 1 ,N6-etheno- ribosyl adenosine) to NAD+ in the presence of 2 mM benzamide decreased in UVC-exposed (8.5 or 32 J/m2) LY-S cells almost fourfold during 4 min, indicating a rapid catabolism. In LY-R cells this decrease was only about twofold [66]. It is plausible to assume that this difference reflects a diminished activation of poly(ADP-ribosylation) by DNA strand incisions generated in the course of excision repair to a lesser extent in LY-R cells. Apparently, this difference is of biological importance: benzamide treatment not only decreased survival of LY-S cells (Fig. 11 A), but also increased the frequency of chromatid aberrations in this cell line, without affecting LY-R cells. Interestingly, histone H4 was ADP-ribosylated in UVC-irradiated,

21 but not in control LY-S cells or LY-R cells; this modification accompanied the cellular resistance to the respective damaging agent, since an increase in histone H4 ADP-ribosylation was found in y-irradiated LY-R cells [43]. Inhibition of poly(ADP-ribosylation) does not affect UVC-irradiated human cells (e.g. [67, 68], but in rodent cells, PARP and Cockayne syndrome protein B seem to be important for pyrimidine dimer repair [69]. Conceivably, both proteins are involved in chromatin remodel- ling, improving accessibility of the damaged sites to the repair machinery [69]. If this is the case, then the repair deficient LY-R cells should be insensitive to inhibition of poly(ADP-ribosylation) after UVC exposure, and this is, in fact, observed (Fig.l 1A). Translesion replication probably is equal in both sublines, as judged from the percentage of G:C —> A:T transitions [64]. Why then is the LY-R subline, in contrast with LY-S cells, not sensitised by caffeine? Recently, Kaufmann et al. [70] postulated the existence of a S phase specific, polymerase r\ independent and caffeine sensitive repair. In a much older paper [71] the authors concluded that "...the replicative bypass repair process is either caffeine resistant or sensitive, depending on the cell type used, but not necessarily on the excision repair capability". Notably, established rodent and human cell lines, among them SV40 transformed XP cells, were sensitised to UVC by caffeine, whereas other human and rodent cells (among them excision deficient XP) were not. Post-UVC recovery of replication also differed in susceptibility to inhibition by caffeine: mouse cells and human XPV (xeroderma pigmentosum variant) cells were affected, whereas human normal and XPA cells were insensitive [71]. Since there are numerous translesion replication polymerases [72] with as yet undefined sensitivity to caffeine, explanation for the differential response of LY sublines to combined UVC + caffeine treatment must await the results of further studies.

4.4. UVC-induced alterations in chromatin organisation

The spectral index method [73] was applied to examine the response of chromatin to increase- ing NaCl and MgCk concentrations in UVC-irradiated LY cells. No alteration in chromatin properties was observed 1 h after UVC irradiation of LY-R cells. Chromatin from UVC-irradiated LY-S cells exhibited a lowered spectral index, indicating that at given Na+ and Mg2+ concentrations (1 or 200 mM NaCl, 0 or 0.5 raM MgCb) chromatin was more compact than that from the unirradiated cells. Benzamide treatment reversed the effect of UVC irradiation

i 1 1 -140 • LY-R o LY-S T §120 O

en 100

I 80 Q. O 60 • I

40 1 20 40 60 Time after UVC irradiation, min Fig. 12. Binding of 8-MOP to DNA in LY cells exposed o 85 J/m2 UVC (in Fischer's medium, un- corrected for absorption of 254 nm radiation in the medium; survival 2x10 and 7x10'3 for LY-R and LY-S, respectively). 8-MOP binding levels in sham-irradiated, identically treated control samples of LY-R and LY-S cells were identical within experimantal error. Mean results are shown from 3-5 experiments, expressed as percentage of the control samples. Standard error indicated. Reproduced by permission from [75].

22 in LY-S cells and did not change the response pattern of chromatin from LY-R cells or unir- radiatedLY-S cells [74]. Chromatin organisation in UVC-exposed LY cells was further probed with the UVA-as- sisted photobinding of tritiated 8-methoxypsoralen (MOP). Following UVC exposure, MOP binding was reduced by 40% at 15 min post-exposure in LY-S cells and gradually rose to the control level during the subsequent 45 min (Fig. 12). MOP preferentially binds to DNA in the chromatin sections that have "open" conformation, to the linker or transcribed DNA. So, this result indicated transient chromatin condensation in UVC-exposed LY-S cells and was con- firmed by increased sedimentation distance of nucleoids. In contrast, MOP binding increased by about 20% at 45 and 60 min post-exposure in LY-R cells and was consistent with slower nucleoid sedimentation [75]. These alterations of chromatin organisation in LY-S cells apparently accompany excision repair and are connected with poly(ADP-ribosylation). Since in LY-R cells the response of chromatin is strikingly different from that in LY-S cells, it is tempting to assume that there is a causal relationship between repair capacity and the observed conformational reorganisation of chromatin. Nevertheless, this work was not continued and the applied methods do not supply information sufficiently specific to be translated into molecular terms.

5. CELLULAR RESPONSE TO IONISING RADIATION

The preceding sections provided information on the origin and general features of L5I78Y sublines, LY-R (radioresistant) and LY-S (radiosensitive) and described their response to UVC radiation and hydrogen peroxide. One peculiar feature of this cellular model is inverse cross-sensitivity to X/y-rays and hydrogen peroxide, unusual in other pairs of radiation hypersensitive and parental, normally radioresistant cell lines (e.g. [76]). With both damaging agents, the main lesion-inducing species is the highly reactive hydroxyl radical, but the mechanisms of sensitivity to these two agents are different in LY sublines. LY-R cells are relatively sensitive to hydrogen peroxide beacuse of higher iron content and lower antioxidant defence than those in LY-S cells [9, 24, 47-51]. In LY-R cells, the high activity of the iron-catalysed Fenton reaction generating hydroxyl radicals is the cause of higher DNA damage upon hydrogen peroxide treatment than in the LY-S subline. In contrast, LY-S cells are more radiosensitive than LY-R cells beacuse of impaired repair of DSB, as described below. Table 5 shows the general characteristics of the response of LY-R and LY-S cells to ionising radiation. One striking feature of the response is lack of difference between LY sublines in the response to high LET (linear energy transfer) radiation. Radford [90] has postulated that high LET radiation inflicts a high proportion of complex DSB, which are difficult to repair. LY sublines have the same ability to deal with these lesions, whereas the sensitivity of LY-S cells to low LET X/y-rays is due to a defect in repair of a lesion type that is differentially induced by high- and low-LET radiation and is innocuous for cells with normal repair capability. Chronologically, the first indication that the high intrinsic radiosensitivity of LY-S cells is connected with DNA damage came from examination of tritium labelling effects [85]. Treat- ment with tritiated water, lysine and thymidine gave increasing lethal effects, depending on whether the tritium decays took place in the whole cell, in histones (which are lysine rich and placed in the near vicinity to DNA) or in the DNA itself (cf. Table 5). The difference between sensitivity of LY sublines to these tritium locations increased as the decays became located closer to the target, DNA. LY-S cells exhibit a unique feature: not only the sparing effect of X-ray dose fractionation is missing, but the first radiation dose seems to sensitise to the second one, as survival after divided radiation dose is lower than that after single dose [55]. This phenomenon was redis- covered by Yau et al. [91] in 1979 and named "inverse split-dose effect".

23 Table 5. Response of LY sublines to ionising radiation.

Characteristic feature LY-R subline LY-S subline Reference Response to X/y-irradiation Low High [2] Radiation sensitivity (cloning) (D0=l.lGy) (D0=0.5Gy) Oxygen enhancement ratio 2.7 2.7 [77] Effect of dose fractionation at 37°C Sparing Sensitising [55] Effect of low dose rate, as measured by the ratio: Do at 0.88 Gy/min Marked, 26.5 Small, 5.1 [78] Do at 0.00037 Gy/min At the same rate and to the same Repair of SSB (alkaline elution) [79] residual level Repair of DSB (neutral elution) Normal Impaired [79-82] Repair of chromatid breaks Normal (premature chromatin condensation) Impaired [80,81] Initial base damage (gas chromatography/mass spectrometry, GC/MS) Low High [83] At approximately the same rate Repair of base damage (GC/MS) [84] and to the same residual level Response to tritium P-radiation Relative susceptibility to tritiated compounds of different intracellular localisation from cell survival: 3 [85] - HOH 1 2.6 - 3H-lysine 1 7.5 - 3H-thymidine 1 40.0 Response to 9 GeV protons

Sensitivity (cloning) D0=1.04Gy D0=0.63Gy [86] Response to 1.6 MeV deuterons

Sensitivity (cloning) D0=0.67Gy D0=0.55Gy [87] Response to 2.5 MeV helium ions

Sensitivity (cloning) D0=0.34Gy D0 = 0.31Gy [88] Response to 6.2 MeV neutrons:

- sensitivity (cloning) Equal (Do = 0.42 Gy) [89] - repair of SSB (alkaline sedimentation) At the same rate and to the same residual level

5.1. Cell cycle: progression and radiosensitivity fluctuations

Alterations in progression of LY cells through the cell cycle after X-irradiation are illustrated in Fig. 13 [92]. The main features are lack of arrest at the Gl/S phase boundary and differential G2 phase arrest. The latter is much more pronounced in LY-S cells than in LY-R cells (11 and 4 h/Gy, respectively) [8]. G2 phase arrest can be prolonged by A23187 (calcium ionophore) treatment with a sparing effect on survival [22] or shortened by caffeine treatment with a sensitising effect in LY-S but not LY-R cells [41]. Sensitisation, however, is caused by caffeine's effect on DNA repair, not on the duration of arrest [93].

24 o 30U

GO _J LJ O o 10-

0 5 10 15 20 25 5 10 15 20 25 TIME AFTER IRRADIATION, h B TIME AFTER IRRADIATION, h

+0-

O 30 LY-R o o LY-S • •

o 10 15 20 25 TIME AFTER IRRADIATION, h

Fig. 13. Distribution of X-irradiated (5 Gy) LY-R and LY-S cells in Gl (A), S (B) and G2/M (C) phases of the cell cycle. Reproduced by permission from [92].

100 -

Hi ccO UJ a.

13 5 7 FRACTION NUMBER

Fig. 14. Cell age dependence of radiosensitivity of LY-S (closed circles) and LY-R (open circles). Cells were separated by centrifugal elutriation according to size and position in the cell cycle and X-irradiated with 2 Gy. Reproduced by permission from [80].

25 Cell age dependence of radiosensitivity of LY sublines is strikingly different [80], as shown in Fig. 14. In LY-R cells, survival after X-irradiation with 2 Gy remains almost unaltered (variation less than twofold), whereas in LY-S cells it varies about hundredfold from the very low level in Gl phase to a peak of resistance in late S phase.

5.2. Post-irradiation poly(ADP-ribosylation)

General features of poly(ADP-ribose) polymer metabolism in LY sublines have been discussed in section 2.8. (Figs 5 and 6); also, some ADP-ribose polymer-binding proteins missing in the A B

70 S 10 15 20 25 30 5 10 15 20 25 30 Time after irradiation (min) Time after irradladfon (rain) Fig. 15. NAD+ (A) and poly(ADP-ribose) (B) content in LY cells after X-irradiation (2 Gy). Mean results from three experiments and standard error are shown. Reproduced by permission from [42]. nuclear matrix of LY-S cells as compared to those in LY-R cells have been presented in Fig.7. Remarkable differences in poly(ADP-ribosylation) are noted in LY cells after X-irradiation [42]. LY-R cells respond by increasing about fourfold the amount of the polymer during the B 100 37 c -»- LY-R 90 \ ° -•< LY-R + AB <** an -O- LY-S jH 70 •O- LY-S + AB rc 60 •-•- • LY-R, X SO —•--LY-R, X+BZ < 40 • LY-S, X —•--LY-S, X+BZ Q 30 ł 20 >—-— 10 • — 0 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 Time (min) TIME AFTER X-IRRADIATION, h

Fig. 16. Differential effects of poly(ADP-ribose) synthesis inhibition by 2 mM benzamide (BZ) or 3-aminobenzamide (AB). (A) Comparison by neutral comet assay of the course of DSB rejoining in the presence or absence of 2 mM 3-aminobenzamide in LY-R and LY-S cells. Irradiation at time zero with 10 Gy X-rays. Mean results ± standard deviation from 4-5 experiments. (B) Frequency of chromatid exchanges in LY cells X-irradiated with 1 Gy with or without continuous 2 mM benzamide treatment (starting 30 min before irradiation). Reproduced by permission from [95].

26 first 15 min after X-irradiation and maintaining the NAD+ level unchanged or slightly higher than under control conditions. In contrast, in LY-S cells there is a 20% decrease in NAD+ between 5 and 15 min after irradiation and the polymer content remains almost unaltered (Fig. 15). These changes and the polymer size distribution are compatible with the high turnover rate of the polymer in LY-S cells and the preferential attack of large polymers by poly(ADP-ribose) glycohydrolase. Interestingly, an increase in histone H4 ADP-ribosylation takes place in y-irradiated LY-R but not LY-S cells; this modification seems to parallel the cellular resistance to the respective damaging agent, since it is seen in UVC-irradiated LY-S but not LY-R cells (cf. section 4.3., [43]). The described differences in polymer synthesis in X-irradiated cells are accompanied by a differential effect of poly(ADP-ribose) polymerase inhibitors: there is a decrease in DNA strand break rejoining [94, 95], and an increase in chromosomal damage in LY-S cells but not in LY-R cells [96], as shown in Fig. 16. Accordingly, the lethal effect of X-rays is enhanced by continuous 2 raM benzamide post-irradiation treatment only in LY-S cells (enhancement ratio from clonogenic survival is 1.53) [41].

5.3. Post-irradiation metabolism of arachidonic acid

As mentioned in section 2.6., the basic features of arachidonic acid metabolism are similar in LY sublines. However, differences between sublines are noted after X-irradiation [30]. In both sublines, irradiation (5 Gy) does not affect the uptake of exogenous arachidonic acid and does not potentiate its release. Stimulation of prostaglandin (PG) synthesis is seen in irradiated LY-R cells (from 3.93 ± 0.32 picomoles/108 cells in control to 5.10 ± 0.41 picomoles/108 cells 30 min after irradiation with 5 Gy X-rays, p<0.05).

Fig. 17. The effect of hydrogen peroxide on PG synthesis in control (open symbols) and X-irradiated (5 Gy - closed symbols) LY-R and LY-S cells. Standard error indicated. Reproduced by permission from [30].

Figure 17 shows the differential change in susceptibility of PG synthesis to hydrogen peroxide in unirradiated and X-irradiated (5 Gy) LY-R and LY-S cells. The higher catalase activity in LY-S cells (described above, cf. Table 3) is a plausible cause of the difference. Both PG and 5-hydroxyeicosatetranoic acid syntheses also are stimulated by calcium ionophore, A23187, in irradiated, but not in control cells. LY-S cells are totally unresponsive in these respects [30]. Eicosanoid biosynthesis is under control of tyrosine kinases [97] that are activated by ionising radiation. LY sublines may differ in this respect (see below). Additional- ly, the radioprotective effect of PG (e.g. [98]) may add to the relatively radioresistant phenotype of LY-R cells.

27 5.4. Stimulation of tyrosine kinase activity

Total tyrosine kinase activity in cells extracts was measured with the use of y-32P-ATP and two different peptide substrates after X-irradiation (1 Gy) and 1 or 3 h incubation [99] (Fig. 18). The control activity level was similar in both sublines, but about twofold stimulation was seen earlier in LY-R cells (after 1 h); in LY-S cells it took place after 3 h, while at that time the kinase activity in LY-R cells returned to the control value [99]. Part of this activity may stem from receptor kinases and the observed difference — from the effect of ganglioside composition on their activation, as mentioned in section 2.5.1.

o Control 1 3 Control 1 Q. Time after irradiation (Hours)

I Control 1 3 Control' 1 3 & Time after irradiation (Hours)

Fig. 18. Stimulation of tyrosine kinase activity by X-irradiation in LY cells. (A) PKS1 (protein kinase substrate 1) corresponds to amino acids 6-20 of the cell division kinase p34 cdc2. (B) PKS2 (protein kinase substrate 2) corresponds to the amino acid sequence 1-17 of gastrin. Cells were X-irradiated (1 Gy); after 1 or 3 h incubation, whole cell extracts were prepared and kinase activity determined. Reproduced by permission from [99].

Determination of c-Abl tyrosine kinase activity in immunoprecipitates showed a comparable activity in the nuclei of both sublines but a two times lower activity in the cytoplasm of LY-S cells in comparison with LY-R cells (Fig. 19). The 32P labelling and the co-preci- pitated proteins were practically identical in both sublines and were in a rather low molecular mass range, as shown in Fig.20. So, there was no indication of c-Abl interaction with large molecules like Atm or DNA-dependent protein kinase (DNA-PK) reported in other cell lines. Irradiation (1 or 5 Gy y-rays) did not appreciably alter the kinase activity measured in nuclear extracts 1 or 3 h later. Also, expression of the kinase protein (determined by Western blotting) was similar in both sublines and not altered by irradiation [100]. Cytoplasmic activated c-Abl exerts an anti-apoptotic effect, probably through phospha- tidyl inositol 3-kinase and Bad [101]. The pathways leading to this anti-apoptotic effect are

28 NUCLEAR EXTRACT WHOLE CELL EXTRACT •a c s

u 25001 i- <° > Y h- 3 O O •t O)

15001 1

Q. O O 500C D.

CONTROL CONTROL+CRK CONTROL CONTROL+CRK CONTROL+CRK CONTROL CONTROL+CRK CONTROL

Fig. 19. Activity of tyrosine kinase, c-Abl, in LY-R and LY-S cells, as determined with y-32P-ATP in immunoprecipitates from nuclear and cytoplasmic extracts. Phosphorylation of co-precipitating proteins (control) or of the added specific substrate, CRK; standard deviation indicated [100]. shown in Fig.21. Activation depends on ligand (growth factor) binding at the receptor tyrosine kinases sites. In LY sublines (in the cell culture conditions) the activity is constitutive. Hence,

NUCLEAR EXTRACT WHOLE CELL EXTRACT LY-R

70.22

LY-S

•

Fig.20. Electrophoretic determination of distribution of 32P-label in proteins present in c-Abl immunoprecipitates according to molecular mass [100]. The CRK protein was added as a marker. the relatively low cytoplasmic activity of the kinase in LY-S cells is consistent with the higher apoptotic proneness of these cells as compared to LY-R cells.

29 Receptor tyrosine kinase

plasma membrane P110 GRB2 P85 Ras mSOS Akt kinase

Active Abl 1 kinase Bad Mitochondrial membrane

Decrease in membrane potential MAPK/ERK Cytochrome c release signaling pathway Activation of caspases

APOPTOSIS

Fig.21. Diagram showing pathways leading to the anti-apoptotic effect of c-Abl. Arrows indicate activation. MAPK - mitogen-activated protein kinase, ERK - extracellular-signal-regulated kinase.

5.5. Apoptosis

The most thorough study of apoptosis in LY-S cells was carried out by video time-lapse microscopy [102]. Within 16-28 h after X-irradiation with 4 Gy about 97% of cells attempted cell division before undergoing apoptosis. During G2 arrest the cells become large (first noted by Okada in 1970 [103], as well as later, in camptothecin-treated cells [38]) and undergo a complicated sequence of unequal division and fusion events, followed by apoptosis with formation of apoptotic bodies. Also Tauchi and Sawada [104] described apoptosis in L5178Y cells as death following mitotic failure. LY-R LY-S

2GyX 48 h

5GyX 48 h

Tooo "o w TO w 'awT ici» Fig.22. Post-irradiation apoptosis measured by the TUNEL method in LY-R and LY-S cells, unirradiated and irradiated with 2 or 5 Gy y-rays. Reproduced by permission from [92].

30 The comparison of post-irradiation apoptosis in LY sublines indicated that the more radiosensitive cell line, LY-S, exhibited enhanced apoptosis in comparison with the radioresistant LY-R cells [92, 105] (Figs 22 and 23).

L5178Y-R L5178Y-S

OtB 1,0 0,5 1,0 2,0 Dose(Gy) Dose(Gy)

Fig.23. Dose-dependence and time course of post-irradiation apoptosis in LY sublines from flow-cyto- metric determination. Reproduced by permission from [105].

Both sublines are heterozygous for a p53 mutation [4, 5] in codon 170 that precludes the transactivation function. Typically for p53 mutants, the p53 protein is present in undamaged cells. The ratio LY-R/LY-S of p53 protein is 1.80 ± 0.23, as determined by Western blotting with the Mab 240 antibody that recognises mutant protein. Figure 24 shows that the increase in p53 in X-irradiated LY cells is negligible, if any. The presence of mutant p53 prevents cor- rect tetramerisation which is necessary for the transactivation function. Due to this function, activation of transcription of numerous genes takes place in cells under stress induced by various DNA damaging agents. The respective proteins carry out functions connected with control of cell cycle progression, DNA repair and apoptosis. Accordingly, in X-irradiated LY cells there is no Gl arrest and no change in expression in apoptosis-related proteins after irradiation, such as E2F-1, bax, bclxL, bcl2 [92]. Both sublines show high constitutive bax and c-Jun kinase content. LY-R cells lack bcl2 and LY-S - cdknla (formerly designated p21/WAFl) [92, 93]. Time after irradiation: 2h 24 h (dose, Gy) 5 contr 2 5 contr LY-R LY-S std p53

Fig.24. Expression of p53 in sham-irradiated (contr) and X-irradiated (2 or 5 Gy) LY cells 2 or 24 h after irradiation Western blots (60 ug protein per lane) are shown. Arrows indicate standard (breast cancer p53, cell lysate from a biopsy, 60 ug protein per lane). Reproduced by permission from [92].

Absence of cdknla may be another pro-apoptotic factor in LY-S cells, since this protein inhibits pro-caspase 3 activation [106]. Accelerated apoptosis takes place in caffeine-treated X-irradiated LY-S cells; such treatment increases the residual DNA damage (estimated by comet assay) by 25%. In contrast, in LY-R cells, the treatment affects neither residual DNA damage nor the timing of apoptosis [93]. So, apart from the expression levels of apoptosis-related proteins, the most probable reason for the difference in type of post-irradiation death between LY sublines is the persistence of residual unrepaired/misrepaired DNA damage. The DNA repair difference is discussed below.

31 5.6. DNA damage induction and repair

The molecular mechanism of radiosensitivity of LY-S cells is still not entirely resolved. The first indication that the high intrinsic radiosensitivity of LY-S cells is connected with DNA damage came from examination of tritium labelling effects (Table 5) [85]. Further studies, with the use of increasingly specific and sensitive methods indicated that DSB repair is impaired in the radiosensitive subline (Table 5). Base damage induction is higher in LY-S than in LY-R cells (in spite of a more efficient antioxidative defence), but base damage excision repair was shown to be equal. Thus, the possibility was eliminated that DSB rejoining difference in LY sublines is caused by repair of complex lesions involving a differentially efficient base excision component [84]. At alkaline pH, a difference in DNA strand break rejoining was found with the alkaline sedimentation technique after a high X-ray dose, 100 Gy [89, 107]. The repair curves were biphasic, corresponding to fast and slow components. In LY-R cells, about 80% of breaks were rejoined by the fast repair during 20 min and in LY-S cells about 60%. The slow repair also was more efficient in LY-R cells. After 90 min the residual damage was 8% of the initial in LY-R cells and 36% in LY-S cells. In contrast, the rejoining of breaks induced by irradiation with fast neutrons (40 Gy) was identical in both sublines: during the first 10 min, 45% breaks were rejoined by fast repair and the residual damage after 90 min was about 30%. Moreover, for LY-S cells the repair curve was identical for both types of radiation. So, the repair machinery in LY-R cells seemed to discern between neutron and X-ray-inflicted damage, whereas in LY-S cells, a class of breaks that was rejoined in repair-competent cells remained unrejoined during 1.5 h post-exposure incubation. Nevertheless, the repair rates in LY sublines were equal when examined with more sensitive methods at denaturing pH using the DNA unwinding technique [108], the alkaline filter elution [79] and the alkaline comet assay [49, 109] for cells X-irradiated with moderately high doses, 0.5-10.0 Gy. Difference in DNA strand break rejoining between LY sublines was observed at non-denaturing pH with methods that measure DSB, such as filter elution and fluorescent halo, with the surprising exception of neutral comet assay (Fig.l6A; the small difference between LY-R and LY-S cells is not statistically significant) [95]. The latter method was applied in a modification [95] which clearly revealed the DSB repair difference between CHO-K1 and xrs6 cells. The xrs6 cells have impaired non-homologous end-joining (NHEJ) due to loss of functional Ku80. Furthermore, a comparison was carried out of initial damage and repair estimated from neutral filter elution and premature chromosome condensation (PCC) in LY sublines [80, 81]. The results obtained with both methods were comparable and pointed to differences in cell age dependence both in the initial damage and its repair, reflected in survival fluctuations during cell cycle progression, shown in Fig. 14. Table 6. Initial DNA damage in X-irradiated LY cells as determined by PCC and cell age dependent survival (data from Wlodek and Hittelman [80]).

Feature examined LY-R cells LY-S cells Initial chromosome breaks (PCC fragments): -Gl cells 4.5 breaks/Gy 4.5 breaks/Gy - G2 cells 2.3 breaks/Gy 4.7 breaks/Gy Approximate surviving fraction, 2 Gy X-rays: -Gl cells 0.37 0.002 - Late S cells 0.37 0.37 - G2 cells 0.30 0.09

Table 6 shows that Gl cells of both sublines do not differ in the initial DNA damage but that the survival difference indicates a repair failure in LY-S cells. In contrast, LY-S cells in

32 G2 phase initially are more damaged than LY-R cells, but survival difference is much smaller than that in Gl phase cells (see also Fig. 14). Figure 25 shows that the relation between the initial chromosome damage and survival of LY-S cells dramatically differs from that in LY-R cells. This difference, however, concerns only LY-S cells in Gl phase, whereas chromosome damage estimated in G2 cells by PCC has identical effect on survival in both LY sublines.

0.01 -

0 50 '20 SO CHROMOSOME SflgAKS PER CEU. Fig.25. The relationship between initial chromosome damage (from PCC) and survival of LY-S cells (closed symbols) and LY-R cells (open symbols), X-irradiated in Gl (circles) or G2 (triangles) phase. Reproduced by permission from[80] .

Figure 26 shows repair in LY sublines estimated by PCC. In LY-S cells, survival after X-irradiation is the lowest in Gl phase and consistently, a large fraction of damage remains unrepaired (Fig.26A). In G2 phase, the difference in survival is smaller, and accordingly, the repair at the chromosome level differs less (Fig.26B). It should be mentioned, however, that into this consistent picture the DSB repair in G2 phase fits less well, as the difference in DSB repair between LY sublines differs in G2 phase almost as much as in Gl phase. Also, the high initial damage in G2 phase LY-S cells as compared to that in LY-R cells remains unexplained. Neutral elution is known to be sensitive to chromatin organisation (e.g. [110]). So, it is possible that the neutral elution data reflect a specific feature of chromatin architecture in LY-S cells that differs from that in LY-R cells. A B

0 30 60 90 TIME AFTER IRRADIATION (mitt) TJMS AFT5R IRRADIATION W Fig.26. (A) Chromosome break repair in LY-S (closed circles) and LY-R cells (open circles), X-irradiated (4 Gy) in Gl phase. Mean values from 3-4 experiments ± standard deviation. (B) Fraction of unrepaired chromatid breaks and gaps in LY-S cells (closed circles) and LY-R cells (open circles), X-irradiated (1 or 2 Gy, respectively) in G2 phase. Mean values from three experiments ± standard deviation. Reproduced by permission from [81].

33 With regard to changes in chromatin, X-rays induce a much smaller proportion of DNA-protein cross-links in LY-S cells than in its radioresistant counterpart (LY-R - 157 ± 29.8% of the control value, LY-S - 98.0 ± 22.9% of the control value) after X-irradiation (5 Gy) [111]. There are other chromatin architecture-related features of LY sublines. Pronounced differences were found in survival and DNA integrity (from the comet assay) of LY cells treated with neocupreine, a copper chelator, and desferroxamine (desferal), an iron chelator [112], corresponding to copper and iron content of the nuclei (cf. Table 2). The comet assay was also used to reveal differences in the percentage of DNA in the comet tail in nucleoids stepwise depleted of protein with the use of increasing (0.14-2.50 M) concentration of sodium chloride and X-irradiated (1.5 Gy) [113]. The initial DNA damage increased in nucleoids from LY-R cells with the increase in sodium chloride concentration, as could be expected from the protective role of chromatin proteins (e.g. [114-117]. In contrast, in LY-S cells the damage remained at a stable level, notwithstanding the amount of protein extracted.

5.7. Mutagenesis

In the inversely cross-sensitive to UVC and X/y-rays LY sublines, mutagenesis does not follow the pattern of survival and DNA repair. Both in UVC- and X/y-ray-induced mutagenesis, LY-R cells show higher mutant frequency. After UVC exposure, the difference in frequency between sublines is dependent on the locus examined (see also section 4.2.). X/y-ray-induced mutations were studied mostly in American laboratories [118-120]. LY-S cells were found to be less mutable than LY-R cells at the hgprt and Na+/K+ ATP-ase loci, not only following X-irradiation but also following treatment with alkylating agents to which these cells are more sensitive than LY-R cells [118, 120]. The hypomutability of LY-S cells was explained by formation of multilocus DNA lesions due to repair deficiency following treatment with ionising radiation or alkylating agents [120]. According to this explanation, multilocus lesions are the cause of poor recovery of viable mutants when the target locus, e.g. hgprt locus, is located next to essential, "house-keeping" genes on the hemizygous X chromosome. Loss of function of a neighbouring essential gene renders the mutant inviable, thus leading to an apparent decrease in mutant frequency. In the case of heterozygous loci, when active copies of neighbouring essential genes are present on the homologous (non-target) chromosome, the recovery of mutants is higher and comparable to that in LY-R cells. Kruszewski et al. [49] in a study of hydrogen peroxide-induced mutagenesis also compared X-ray-induced (0.5-3.0 Gy) mutations at the hgprt locus in LY sublines. In LY-R cells there was a dose-dependent increase in mutant frequency, and treatment with iron chelators of various affinity decreased the frequency. This observation suggested that hydrogen peroxide generated in X-irradiated LY-R cells entered the Fenton reaction and inflicted additional mutagenic lesions. In agreement with the observations of other authors mentioned above, the mutation level in X-irradiated LY-S cells was considerably lower than in LY-R cells. It was not modified by pretreatment with iron chelators.

5.8. NHEJ defect in LY-S cells

The hypersensitivity of LY-S cells to X-rays in Gl phase, described above (Figs 14 and 25, Table 6), indicated that NHEJ was the repair system impaired in LY-S cells, by analogy to NHEJ mutants (reviewed in [6]). This hypothesis was supported by the observation that DNA repair (measured by comet assay) was refractory to the DNA-PK specific inhibitor, OK-1035, in contrast with that in the DSB repair-competent LY-R cells [109]. Nevertheless, in both LY sublines the DNA-PK activity was found in total cell extracts at almost equal levels [95]. In these experiments, the DNA-PK activity-deficient cells served as negative controls, including V3, a Chinese hamster cell line (mutant lacking the catalytic subunit) and CHO mutant xrs6

34 cells, harbouring a mutation in XRCC5/Ku80. CH0-K1 cells with functional DNA-PK served as positive control. This result indicates that a NHEJ defect in LY-S cells is not due to a DNA-PK mutation that would affect its activity in the in vitro test. Complementation studies carried out in Japan [121] have shown that the defect is not in xrcc4 or ligase IV. This leaves the possibility that other components of NHEJ may be defective in LY-S cells, such as AHNAK [122], Artemis or sir proteins (reviews in [123, 124]). Nevertheless, recent studies by Block et al. [125] indicate that the possible defect may rest in DNA-PK. It may concern autophosphorylation in the so-called ABCDE cluster of serine and threonine residues in the DNA-PK activities after irradiation

350

5 300 0. '•£. SI 1_ g. o ^ 250 0 o « ? & c S? § 200 - • LY-R SLY-S III •3 ioo • a: o

Fig.27. Activity of DNA-PK in LY sublines (determined by the pull-down method) after y-irradiation with 1 Gy. When the NHEJ process is completed, DNA-PK becomes phosphorylated and loses the ability to bind DNA and, in consequence, loses the catalytic activity. Since DSB repair in LY-S cells is slow, inactivation of DNA-PK after irradiation is delayed. Reproduced by permission from [95]. central part of the catalytic subunit. Such an autophosphorylation defect does not affect kinase activity nor autophosphorylation-induced loss of kinase activity but is a cause of failure in DNA end-joining. This corresponds exactly with the observations carried out on LY-S cells. Experi- ments to confirm this assumption are under way.

ISO 140 8 120 H 10W > 3 5

K Z

| 140 B 8 120 ioo< |. eo i BO -o 40 20 0-

0.0 0.5 1.0 13 2,0 TIME AFTER CAMPTOTHECIN REMOVAL, h Fig.28. Time-course of DSB repair in LY cells after 1 h treatment with 2 uM camptothecin. Remaining damage is shown, expressed as percentage of the initial damage taken as 100% determined by field inversion gel electrophoresis (A) and by comet assay in neutral pH (B). Closed circles - LY-S cells, open circles - LY-R cells. Standard deviation indicated. Reproduced by permission from [38].

35 Alterations in DNA-PK activity were noted in y-irradiated (1 Gy) LY-R and LY-S cells after various time intervals [95]. As shown in Fig.27, there was a decrease in the activity in LY-R cells with a minimum at 2 h. A comparable decrease in LY-S cells took place at 4 h after irradiation. This result was consistent with the slow functioning of NHEJ [126]. That LY-S cells are capable of DSB repair was also concluded from the high recovery of cells kept for 12-24 h at 34 or 25°C after X-irradiation [127-129]. From the increase in survival seen in LY-S cells progressing through the S phase (Fig. 14) it could be concluded that DSB in the replicated DNA can be repaired by homologous recombination. This type of repair operates in the case of DSB generated by collision of the camptothecin-DNA cleavable complex with replication forks and is functional in LY-S cells [38], as shown in Fig.28. The repair of DSB generated in freshly replicated (3H-thymidine-labelled) DNA in camptothecin-treated LY cells was determined by field inversion gel electrophoresis and by comet assay in neutral pH. As shown in Fig.28, LY-S cells proved to be more efficient than LY-R cells in the early (up to 2 h) repair of the DSB induced by camptothecin.

5.9. DNA damage fixation

As mentioned above, in contrast with other cell lines of various origin and radiation sensitivity, LY-S cells are sensitised by the first radiation dose to the second dose, so that the relative survival after fractionated irradiation is below unity. This is an unusual effect, since cell lines with DNA repair defects, e.g. scid (severe combined immunodeficiency) cells, at worst show the relative survival of about 1 after irradiation with divided doses. The relative survival of both LY sublines after fractionated X-irradiation is presented in Fig.29A. After X-irradiation (2+2 Gy), the relative (vs 4 Gy) survival decreases with the time interval between the first and A B

I 2 3 4 5 SPLIT DOSE INTERVAL* h)

12 3 4 TIME AFTER IRRADIATION [h]

Fig.29. (A) Relative survival curve for LY sublines (dashed lines - LY-R cells; solid lines - LY-S cells) after irradiation with divided doses of X-rays with incubation at 25°C (closed circles) or 37°C (open circles) between the first and second dose. Reproduced by permission from [129]. (B) Damage fixation in LY-S cells illustrated by survival data after 5 h interval split doses of 2+2 Gy relative to a single dose of 4 Gy of X-rays, as functions of post-irradiation incubation at 25°C (dashed lines) or 37°C (solid lines). Standard error of the mean indicated. Reproduced by permission from [130].

36 second irradiation until 8 h and then starts to rise [55]. The decrease can be prevented by incubation at 23 or 25°C between doses (Fig.29) [91, 128-130]. These results were discussed as indicating a temperature-dependent shift of balance between repair and fixation of radiation damage [130]. The nature of the fixation process was not understood. Differences in radiation response usually are ascribed to variation in DNA repair efficiency. On the other hand, it is generally acknowledged that the outcome of post-irradiation processes, as seen at the cellular level, depends on the competition between repair and fixation of damage. Better post-irradiation recovery of LY-S cells at 25°C makes the pair of LY sublines an eminently suitable model to study damage fixation. Recently, an attempt was made [95] to interpret the response of LY cells to X-irradiation in terms of Radford's model of damage fixation [131]. In this model, it is assumed that DSB fixation takes place in transcriptional factories. Two molecules of topoisomerase I act in concert on transcribed DNA strands localised close to one another. One enzyme molecule is located on an undamaged DNA strand, the other forms a complex with a DSB. An exchange event takes place, bringing about a chromosomal aberration which may be lethal. As diagrammatically shown in Fig.30, PARP-ef- fected inhibition of topoisomerase I (observed in X-irradiated cells, as shown by Boothman et al. [132]) prevents damage fixation. At 37°C, 2 mM 3-aminobenzamide decreased the repair of DSB in LY-S but not LY-R cells, in agreement with the previously observed radiosensitisation of LY-S cells by poly(ADP-ribosylation) inhibition (Fig. 16). This is understandable, considering that at 37°C LY-R cells are able to rejoin DSB at a rate that is sufficient to outdistance fixation, whereas in LY-S cells the rejoining is too slow. However, DSB rejoining in the repair competent cell line LY-R, also was affected by 3-aminobenzamide when the post-irradiation incubation was carried out at 25°C. Apparently, lowered temperature slowed down the rejoining enough to reveal the effect of poly(ADP-ribosylation) inhibition, in contrast with the absence of 3-aminobenzamide effect on repair at 37°C [95].

DNA DOUBLE STRAND DSB REPAIR BREAKS WITHIN TRANSCRIPTION FACTORIES

11 FIXATION (TOPOISOMERASE I) T DNA DOUBLE i PARP

DNA DOUBLE STRAND DSB REPAIR BREAKS WITHIN TRANSCRIPTION FACTORIES REJOINED

FIXATION (TOPOISOMERASE I) PARP

STRAND BREAKS FIXED 3-AMINOBENZAMIDE

Fig.30. Explanation of DSB fixation according to the model of Radford [131]. (A) In a repair-competent mammalian cell, the equilibrium between DSB repair and fixation is shifted towards repair. Usually, fixation is prevented, because PARP (activated by DNA damage) inhibits topoisomerase I. (B) PARP inhibitor, 3-aminobenzamide, enhances fixation and the number of correctly repaired DSB decreases.

37 Thus, the model explains the previously reported differential effect of poly(ADP-ribosylation) inhibition in X-irradiated LY-R and LY-S cells [41, 94, 96] (Fig. 16) as caused by its role in damage fixation.

6. CONCLUDING REMARKS

All experimental results reviewed above point to the key importance of DNA repair and fixation for the ultimate fate of the X-ray-damaged LY cell. The cause of slow DSB repair in LY-S cells is not identified, but the assumed DNA-PKcs autophosphorylation defect explains most features of the cellular response to irradiation, as compared to the repair-competent LY-R cells: high radiosensitivity of Gl cells, extensive damage fixation, long G2 arrest, considerable chromosomal damage seen as PCC fragments and aberrations in metaphase cells. On the other hand, at the level of damage corresponding to a comparable lethal effect, the type of death differs between LY sublines. This difference is consistent with altered expression of proteins that are pro- or anti-apoptotic and possibly, with differences in cellular signalling and protein translocations between cellular compartments which have not been sufficiently resolved. Another insufficiently characterised feature is chromatin architecture, which seems to be connected with the post-irradiation response of LY-S cells. The prominent role of poly(ADP-ribosylation) in the response of LY-S cells apparently is connected with damage fixation, but is in contrast with other hypersensitive to X/y-radiation cell lines with DSB repair defects. In LY-S cells, the role of poly (ADP-ribose)-binding nuclear matrix proteins remains to be established. Comparison of LY sublines also indicates that their antioxidant defence system is of considerable importance in susceptibility to hydrogen peroxide, but much less so in radiosensitivity. The unique inverse cross-sensitivity of LY sublines to X-rays and agents that inflict lesions to DNA repaired by excision has been an advantage in studies on effects of combined (X-rays + drug) treatment. The description of these combined treatments would increase the volume of this review beyond the acceptable limit. It may be mentioned, however, that especially in the case of combination of platinum anticancer complexes and X-irradiation, the use of LY sublines helped to formulate requirements for obtaining a more than additive effect of combined treatment [133, 134]. In a recent analysis of genes that are causally related to radiosensitivity in Saccharo- myces cerevisiae [135], it was found that many of the newly identified genes are not coding for proteins involved in DNA repair or cell cycle control machinery. They are involved in such cellular functions as transcription, transport and degradation of macromolecules, including nuclear-cytoplasmic traffic or local alterations in chromatin conformation. It can be expected that the role of genes not directly related to DNA repair will also be discovered in UVC- or hydrogen peroxide-damaged mammalian cells, in analogy to yeast. A much more detailed knowledge of gene expression control and protein functions in mammalian cells will be needed than is available now, for a full description of the cellular response to DNA damage in molecular terms. From this point of view, even the apparently unimportant observations may turn to be enlightening. I hope that time will come when it will be possible to confront gene expression profiles in LY cells with their phenotypic features. Technically, it already is feasible, but this kind of project without a direct practical application presently has small chances to get financial support.

Acknowledgements Permission for reproduction of figures is gratefully acknowledged from the following journals: International Journal of Radiation Biology (published by Taylor and Francis; www.tandf.co.uk/journals), Radiation Research (published by Radiation Research Society),

38 Photochemistry and Photobiology B, Prostaglandins, Leukotrienes and Essential Fatty Acids, Free Radicals in Biology and Medicine and Mutation Research (published by Elsevier), Cell Biochemistry and Function (published by John Wiley and Sons Ltd.), Radiation Environmental Biophysics (published by Springer Verlag), Nukleonika (published by the Instutite of Nuclear Chemistry and Technology). The greatest part of the reviewed data comes from the experimental work of my colleagues in various laboratories in Poland and abroad and I very much appreciate their cooperation. Financial support from the State Committee for Scientific Research (KBN) statutory grant for the Institute of Nuclear Chemistry and Technology is acknowledged.

7. REFERENCES

[1]. Alexander P., Mikulski Z.B.: Mouse lymphoma cells with different radiosensitivities. Nature, 192.572-573(1961). [2]. Beer J. Z., Budzicka E., Niepokójczycka E., Rosiek O., Szumiel I., Walicka M.: Loss of tumorigenicity with simultaneous changes in radiosensitivity and photosensitivity during in vitro growth of L5178Y murine lymphoma cells. Cancer Res., 43, 4736-4742 (1983). [3]. Sanchez C: Cytogenetic characteristics of the murine lymphoma cell lines L5178Y-R and L5178Y-S. M.Sc. Thesis, Florida Institute of Technology, Melbourne, Fla, USA 1983. [4]. Storer R.D., Kranyak A.R., McKelvey T.W., Elia H.C., Goodrow T.L., Deluca J.G.: The mouse lymphoma L5178Y TK+/-cell line is heterozygous for a codon 170 mutation in the p53 tumor suppressor gene. Mutat. Res., 373, 157-165 (1997). [5]. Radford I.R.: p53 status, DNA double-strand break repair proficiency, and radiation response of mouse lymphoid and myeloid cell lines. Int. J. Radiat. Biol., 66, 557-560 (1994). [6]. Thacker J., Zdzienicka M.Z.: The mammalian XRCC genes: their roles in DNA repair and genetic stability. DNA Repair, 2, 655-672 (2003). [7]. Itsukaichi H., Mori M., Nakamura A., Sato K.: Identification of a new G-to-A transition mutation at nucleotide position 129 of the Xrcc4 gene in ionizing radiation-hypersensitive mutant LX830 cells. J. Radiat. Res., 44, 353-358 (2003). [8]. Szumiel I.: Response of two strains of L5178Y cells to cis-dichlorobis(cyclopentylamine) platinum (II). I. Cross-sensitivity to cis-PAD and UV light. Chem. Biol. Interact., 24, 51-72 (1979). [9]. Boużyk E., Iwaneńko T., Jarocewicz N., Kruszewski M., Sochanowicz B., Szumiel I.: Anti- oxidant defense system in differentially hydrogen peroxide sensitive L5178Y sublines. Free Radic. Biol. Med., 22, 697-704 (1997). [10]. Rutledge M.H., Lett J.T.: The L5178Y S/S murine leukemie lymphoblast: a radiosensitive, malignant cell of stable karyotype. Radiat. Res., 81, 282-291 (1980). [11]. Mcllrath J., Bouffler S.D., Samper E., Cuthbert A., Wojcik A., Szumiel I., Bryant P.E., Riches A.C., Thompson A., Blasco M.A., Newbold R.F., Slijepcevic P.: Telomere length abnormalities in mammalian radiosensitive cells. Cancer Res., 61, 912-915 (2001). [12]. Myung K., Ghosh G., Fattah F.J., Li G., Kim H., Dutia A., Pak E., Smith S., Hendrickson E.A.: Regulation of telomere length and suppression of genomic instability in human somatic cells by Ku86. Mol. Cell Biol., 24, 5050-5059 (2004). [13]. Filippovich I.V., Sorokina N.I., Soldatenkov V.A., Romantzev E.F.: Supercoiled DNA repair in thymocyte fractions differing in radiosensitivity. Int. J. Radiat. Biol., 42, 31-44 (1982). [14], Walicka M., Godlewska E.: Radiation sensitivities are not related to the sizes of DNA supercoiled domains in L5178Y-R and L5178Y-S cells. Int. J. Radiat. Biol., 55, 953-961 (1989). [15]. Gonta-Grabiec K., Sarzała M.G., Szumiel I., Szymanska G., Włodek D.: No radiosensitivity-related change in plasma membrane properties in X- or gamma-irradiated L5178Y-R and L5178Y-S cells. Neoplasma, 32, 561-569 (1985). [16]. Bartoszewicz Z., Szumiel I., Dancewicz A.M.: Glycosphingolipids in L5178Y murine lymphoma sublines differing in radiosensitivity. Nukleonika, 33_, 233-245 (1988). [17]. Miljan E.A., Bremer E.G.: Regulation of growth factor receptors by gangliosides. Sci. STKE, 2002(160), RE 15 (2002).

39 [18]. Mirkin B.L., Clark S.H., Zhang C: Inhibition of human neuroblastoma cell proliferation and EGF receptor phosphorylation by gangliosides GM1, GM3, GDI A and GT1B. Cell Prolif., 35_, 105-115(2002). [19]. Zurita A.R., Maccioni H.J., Daniotti J.L.: Modulation of epidermal growth factor receptor phosphorylation by endogenously expressed gangliosides. Biochem. J., 3_55_(Pt 2), 465-472 (2001). [20]. Miljan E.A., Meuillet E.J., Mania-Farnell B., George D., Yamamoto H., Simon H.G., Bremer E.G.: Interaction of the extracellular domain of the epidermal growth factor receptor with gangliosides. J. Biol. Chem., 277, 10108-10113 (2002). [21]. Roepstorff K., Thomsen P., Sandvig K., van Deurs B.: Sequestration of epidermal growth factor receptors in non-caveolar lipid rafts inhibits ligand binding. J. Biol. Chem., 277, 18954-18960 (2002). [22]. Szumiel I., Wlodek D., Lustyik G.: Ion content and the response of L5178Y-R and L5178Y-S cells to X rays and ionophore A23187. Radiat. Environ. Biophys., 23, 255-267 (1984). [23]. Szumiel I., Sochanowicz B., Buraczewska I.: Ca2+ mobilization is related to the lethal effect of X-irradiation in L5178Y cells. Int. J. Radiat. Biol., 58, 125-131 (1990). [24]. Lipinski P., Drapier J.C., Oliveira L., Retmańska H., Sochanowicz B., Kraszewski M.: Intra- cellular iron status as a hallmark of mammalian cell susceptibility to oxidative stress: a study of

L5178Y mouse lymphoma cell lines differentially sensitive to H2O2. Blood, 95, 2960-2966 (2000). [25]. Szumiel I., Kapiszewska M., Kraszewski M., Iwanenko T., Lange C.S.: Content of iron and copper in the nuclei and induction of pH 9-labile lesions in L5178Y sublines inversely cross-

-sensitive to H2O2 and X-rays. Radiat. Environ. Biophys., 34, 113-119 (1995). [26]. Kraszewski M, Zaim J., Grądzka I., Szumiel I.: Comparison of the effects of bleomycin and ionizing radiation in two sublines of murine lymphoma L5178Y. Nukleonika, 46, 81-86 (2001). [27]. Beer J.Z., Budzicka E., Niepokój czycka E., Szumiel I., Walicka M.: Institute of Nuclear Research internal report No. 53/X/79. [28]. Scott I.D., Nicholls D.G.: Energy transduction in intact synaptosomes. Influence of plasma- -membrane depolarization on the respiration and membrane potential of internal mitochondria determined in situ. Biochem. J., 186, 21-33 (1980). [29]. Wlodek D., Niepokojczycka E.: Institute of Nuclear Research internal report No. 21/TV/85. [30]. Sochanowicz B., Szumiel I.: Arachidonic acid metabolism in murine lymphoma cell sublines differing in radiation sensitivity. Prostaglandins Leukot. Essent. Fatty Acids, £5, 241-247 (1996). [31]. Szumiel I.: Odwrotna krzyżowa wrażliwość dwóch podlinii białaczki mysiej L5178Y na pro- mieniowania X i UVC. Post. Biol. Kom., 15,451-466 (1988). [32], Evans H.H., Ricanati M., Horng M.F., Mencl J.: Relationship between topoisomerase II and radiosensitivity in mouse L5178Y lymphoma strains. Mutat. Res., 211, 53-63 (1989). [33], Staroń K., Kowalska-Loth B., Szumiel I., Buraczewska I.: Sensitivity to inhibitors of type II to- poisomerases from mouse L5178Y lymphoma strains that are resistant or sensitive to X-radiation. Mutat. Res., 285, 175-179 (1993). [34]. Walicka M., Szmigiero L., Ciesielska E., Grądzka I.: Are lethal effects of nitracrine on L5178Y cell sublines associated with DNA-protein crosslinks? Biochem. Pharmacol., 46, 615-620 (1993). [35]. Kowalska-Loth B., Staroń K., Buraczewska I., Szumiel I., Kapiszewska M., Lange C.S.: Reduced sensitivity to camptothecin of topoisomerase I from a L5178Y mouse lymphoma subline sensitive to X-radiation. Biochim. Biophys. Acta, 1172. 117-123 (1993). [36]. Tsao Y.P., Russo A., Nyamuswa G., Silber R., Liu L.F.: Interaction between replication forks and topoisomerase I-DNA cleavable complexes: studies in a cell-free SV40 DNA replication system. Cancer Res., 53, 5908-5914 (1993). [37]. Staroń K., Kowalska-Loth B., Ząbek J., Czerwiński R.M., Nieznański K., Szumiel I:. Topo- isomerase I is differently phosphorylated in two sublines of L5178Y mouse lymphoma cells. Biochim. Biophys. Acta, 1260, 35-42 (1995). [38]. Grądzka I., Skierski J., Szumiel I.: DNA repair, cell cycle progression and cell death following camptothecin treatment in two murine lymphoma L5178Y sublines. Cell Biochem. Funct., 16± 239-252 (1998). [39]. Staroń K., Kowalska-Loth B., Nieznański K., Szumiel I.: Phosphorylation of topoisomerase I in L5178Y-S cells is associated with poly(ADP-ribose) metabolism. Carcinogenesis, 17, 383-387 (1996). [40]. Kleczkowska H.E., Szumiel I., Althaus F.R.: Differential poly(ADP-ribose) metabolism in repair-proficient and repair-deficient murine lymphoma cells. Mutat. Res., 235. 93-99 (1990).

40 [41]. Szumiel I., Wlodek D., Johnson K.J., Sundell-Bergman S.: ADP-ribosylation and post-irradiation cellular recovery in two strains of L5178Y cells. Br. J. Cancer Suppl., 6, 33-38 (1984). [42]. Kleczkowska H.E., Malanga M., Szumiel I., Althaus F.R.: Poly ADP-ribosylation in two L5178Y murine lymphoma sublines differentially sensitive to DNA-damaging agents. Int. J. Radiat. Biol., 78, 527-534 (2002). [43]. Kleczkowska H.E., Szumiel I: Histone acceptors for ADP-ribose in UVC or gamma-irradiated L5178Y sublines differing in DNA repair abilities. Nukleonika, 35_, 213-226 (1990). [44]. Cardenas-Corona M.E., Jacobson E.L., Jacobson M.K.: Endogenous polymers of ADP-ribose are associated with the nuclear matrix. J. Biol. Chem., 262, 14863-14866 (1987). [45]. Kapiszewska M., Wright W. D., Lange C. S., Roti Roti J. L.: DNA supercoiling changes in nucleoids from irradiated L5178Y-S and L5178Y-R cells. Radiat. Res., 119, 569-575 (1989). [46]. Malyapa R.S., Wright W.D., Roti Roti J. L.: DNA supercoiling changes and nucleoid protein composition in a group of L5178Y cells of varying radiosensitivity. Radiat. Res., 145, 239-242 (1996). [47]. Kruszewski M., Green M.H., Lowe J.E., Szumiel I.: DNA strand breakage, cytotoxicity and mutagenicity of hydrogen peroxide treatment at 4°C and 37°C in L5178Y sublines. Mutat. Res., 308,233-241 (1994). [48]. Kruszewski M., Iwaneńko T.: Labile iron pool correlates with iron content in the nucleus and the formation of oxidative DNA damage in mouse lymphoma L5178Y cell lines. Acta Biochim. Pol., 50, 211-215 (2003). [49]. Kruszewski M, Green M.H.L., Lowe J.E., Szumiel I.: Comparison of effect of iron and calcium

chelators on the response of L5178Y sublines to X-rays and H2O2. Mutat. Res., 326. 55-163 (1995). [50]. Jaworska A., Rosiek O., Witkowska K.: Activities of superoxide dismutase and catalase in two L5178Y murine lymphoma cell strains with different radiosensitivities. Nukleonika, 32, 193-201 (1987). [51]. Boużyk E., Grądzka I., Iwaneńko T., Kruszewski M., Sochanowicz B., Szumiel 1.: The response of L5178Y lymphoma sublines to oxidative stress: antioxidant defence, iron content and nuclear translocation of the p65 subunit of NF-kappaB. Acta Biochim. Pol., 47, 881-888 (2000). [52]. Aggarwal B.B., Takada Y., Shishodia S., Gutierrez A.M., Oommen O.V., Ichikawa K, Baba Y., Kumar A.: Nuclear transcription factor NF-kappa B: role in biology and medicine. Indian J. Exp. Biol., 42, 341-353 (2004). [53]. Sochanowicz B., Szumiel I., Grądzka I.: Nuclear translocation of the p65 subunit of NF-KB in L5178Y sublines differing in antioxidant defense. Radiat. Environ. Biophys., 38, 125-131 (1999).

[54]. Szumiel I., Wlodek D.5 Niepokójczycka E., Johanson K.J.: Differential response of excision proficient and deficient L5178Y cells to UVC irradiation and benzamide. J. Photochem. Photobiol. B., 3, 483-496 (1989). [55]. Beer J.Z., Szumiel I., Walicka M.: Cross-sensitivities to X-rays and UV-light of two strains of murine lymphoma L5178Y cells in vitro. Bull. Acad. Pol. Sci. (ser. Sci. Biol.), 21, 837-840 (1973). [56]. Walicka M., Godlewska E., Kleczkowska H.: Effect of UVC and araC on L5178Y-R and L5178Y-S cells. Nucleoid sedimentation. Acta Biochim. Pol., 34, 345-355 (1987). [57]. Szumiel I., Wlodek D., Johanson K.J.: Impaired repair of UVC-induced DNA damage in L5178Y-R cells: DNA unwinding studies with the use of 1-beta-D-arabinofuranosyl cytosine. Photochem. Photobiol., 48,201-204 (1988). [58]. Hagan M.P., Dodgen D.P., Beer J.Z.: Impaired repair of UVC-induced DNA damage in L5178Y-R cells: sedimentation studies with the use of 5'-bromodeoxyuridine photolysis. Photochem. Photobiol., 47, 815-821 (1988). [59]. Evans H.H., Horng M.F., Beer J.Z.: Lethal and mutagenic effects of radiation and alkylating agents on two strains of mouse L5178Y cells. Mutat. Res., 161., 91-97 (1986). [60]. Walicka M., Korner I., Malz W., Beer J.Z.: The effect of caffeine on post-replication repair and survival in two L5178Y cell lines with different sensitivities to UV irradiation. Mutat. Res., 52, 265-272(1978). [61]. Szumiel I., Wlodek D.: The effect of caffeine on DNA synthesis in L5178Y cells treated with an anti-tumour platinum complex or UV light. Nukleonika, 23, 865-871 (1978). [62]. Jacobson E.D., Krell K., Olempska-Beer Z., Beer J.Z.: UV-induced mutagenesis at the hypoxanthine-guanine phosphoribosyl transferase locus in two L5178Y mouse lymphoma cell strains with different UV sensitivities. Mutat. Res., 129, 259-267 (1984).

41 [63]. Beer J.Z., Jacobson E.D., Evans H.H., Szumiel I.: X-ray and UV mutagenesis in two L5178Y cell strains differing in tumorigenicity, radiosensitivity, and DNA repair. Br. J. Cancer Suppl., 6,107-111 (1984). [64]. Zernik-Kobak M., Szumiel I., Levine A.S., Dixon K.: Analysis of mutagenesis in UV-sensitive mouse lymphoma L5178Y-R cells with a polyomavirus-based shuttle vector. Mutat. Res., 344. 31-39(1995). [65]. Szumiel I.: Response of two strains of L5178Y cells to cis-dichlorobis(cyclopentylamine)platinum(II). II. Differential effects of caffeine. Chem.-Biol. Interact., 24, 73-82 (1979). [66]. Kleczkowska H.E., Szumiel I., Althaus F.R.: Adaptive changes in NAD+ metabolism in ultraviolet light-irradiated murine lymphoma cells. Cell Biol. Toxicol., 6, 259-268 (1990). [67]. James M.R., Lehmann A.R.: Role of poly(adenosine diphosphate ribose) in deoxyribonucleic acid repair in human fibroblasts. Biochemistry, 21., 4007-4013 (1982). [68]. Fujiwara Y., Goto K., Yamamoto K., Ichihashi M.: Roles of poly(ADP-ribose) synthesis in repair and replication in normal human, Cockayne syndrome, and xeroderma pigmentosum fibroblasts after UV irradiation. Princess Takamatsu Symp., H, 209-218 (1983). [69]. Flohr C, Burkle A., Radicella J.P., Epe B.: Poly(ADP-ribosyl)ation accelerates DNA repair in a pathway dependent on Cockayne syndrome B protein. Nucleic Acids Res., 3_i, 5332-5337 (2003). [70]. Kauftnann W.K., Heffernan T.P., Beaulieu L.M., Doherty S., Frank A.R., Zhou Y., Bryant M.F., Zhou T., Luche D.D., Nikolaishvili-Feinberg N., Simpson D.A., Cordeiro-Stone M.: Caffeine and human DNA metabolism: the magic and the mystery. Mutat. Res., 532, 85-102 (2003). [71]. Fujiwara Y., Tatsumi M.: Replicative bypass repair of ultraviolet damage to DNA of mammalian cells: caffeine sensitive and caffeine resistant mechanisms. Mutat. Res., 37, 91-110 (1976). [72]. Yagi T., Takebe H.: Similarity in the effect of caffeine on DNA synthesis after UV irradiation between xeroderma pigmentosum variant cells and mouse cells. Jpn. J. Cancer Res., 80, 754-758 (1989). [73]. Dixon D.K, Burkholder G.D.: The effects of sodium and magnesium ion interactions on chromatin structure and solubility. Eur. J. Cell Biol., 36, 315-322 (1985). [74]. Kleczkowska H.E., Buraczewska I., Szumiel I.: Benzamide-inhibitable alterations in spectral properties of chromatin accompany DNA repair in UVC-exposed L5178Y cells. Radiat. Environ. Biophys., 27, 213-218 (1988). [75]. Szumiel I., Walicka M., Buraczewska I., Beer J.Z.: Ultraviolet radiation-induced alterations in chromatin organization of L5178Y cell strains with different capacity for DNA repair. Nukleonika, 38,47-54 (1993). [76]. Sato K., Ito A., Shiomi T., Hama-Inaba H., Ishikawa H., Yoshizumi T., Nakazawa T.: X-ray-sensitive mutants of mouse mammary carcinoma cells are hypersensitive to bleomycin and hydrogen peroxide. Jpn. J. Cancer Res., 77, 456-461 (1986). [77]. Szumiel I., Budzicka E., Niepokójczycka E., Włodek D., Beer J.Z.: The response of L5178Y-R and L5178Y-S cells to X rays under oxic and hypoxic conditions. Radiat. Res., 87, 592-601 (1981). [78]. Beer J.Z., Mencl J., Horng M.F., Gregg E.C., Evans H.H.: Effects of low dose rate (0.003-0.025 Gy/h) chronic X-irradiation on radioresistant and radiosensitive L5178Y mouse lymphoma cells. Int. J. Radiat. Biol., 48, 609-619 (1985). [79]. Wlodek D., Hittelman W.N.: The repair of double strand breaks correlates with radiosensitivity of L5178Y-S and L5178Y-R cells. Radiat. Res., JJ2, 146-155 (1987). [80]. Wlodek D., Hittelman W.N.: The relationship of DNA and chromosome damage to survival of synchronized X-irradiated L5178Y cells. I. Initial damage. Radiat. Res., JJ5, 550-565 (1988). [81]. Wlodek D., Hittelman W.N.: The relationship of DNA and chromosome damage to survival of synchronized X-irradiated L5178Y cells. II. Repair. Radiat. Res., 115, 566-575 (1988). [82]. Evans H.H., Ricanati M., Horng M.F.: Deficiency in DNA repair in mouse lymphoma strain L5178Y-S. Proc. Natl. Acad. Sci. USA, 84, 7562-7566 (1987). [83]. Zastawny T.H.; Kraszewski M.; Olinski R.: Ionizing radiation and hydrogen peroxide induced oxidative DNA base damage in two L5178Y cell lines. Free Radic. Biol. Med., 24, 1250-1255 (1998). [84]. Kruszewski M., Zastawny T.H., Szumiel I.: Repair of gamma-ray-induced base damage in L5178Y sublines is damage type-dependent and unrelated to radiation sensitivity. Acta Biochim. Pol., 48, 525-533(2001).

42 [85]. Szumiel I., Walicka M., Budzicka E., Beer J.Z.: Susceptibility of L5178Y-R and L5178Y-S cells to HTO, 3H-lysine and 3H-thymidine exposure. Curr. Top. Radiat. Res. Q., 12,157-167 (1978). [86]. Kraszewski M., Ovodkov J.V., Gerasimenko V.N., Karpovski A.L., Rosiek O., Sawicki Z.: The influence of radiation quality on survival of two mouse lymphoma L5178Y cell strains of different radiation sensitivity. 2. Effects of 9 GeV protons. Nukleonika, 33, 339-343 (1988). [87]. Kraszewski M., Ovodkov J.V., Krasavin E.A., Lobachevsky P.N., Rosiek O., Sawicki Z., Tcherevatenko A.P.: The influence of radiation quality on survival of two mouse lymphoma L5178Y cell strains of different radiation sensitivity. 5. Effects of 1.6 MeV deuterons. Nukleo- nika, 34, 315-321 (1989). [88]. Kraszewski M., Ovodkov J.V., Krasavin E.A., Lobachevsky P.N., Rosiek O., Sawicki Z., Tcherevatenko A.P.: The influence of radiation quality on survival of two mouse lymphoma L5178Y cell strains of different radiation sensitivity. 4. Effects of 2.5 MeV helium ions. Nukleonika, 34, 109-114 (1989). [89]. Walicka M., Koerner I.J., Malz W.: Survival and DNA single strand break rejoining in L5178Y-Rand L5178Y-S cells after neutron irradiation. Studia Biophys., 76, 49-50 (1979). [90]. Radford I.R.: DNA lesion complexity and induction of apoptosis by ionizing radiation. Int. J. Radiat. Biol., 78, 457-466 (2002). [91]. Yau T.M., Kim S.C., Cregg E.C., Nygaard O.F.: Inverse X-irradiation split-dose effect in a murine lymphoma cell line. Int. J. Radiat. Biol., 35, 577-581 (1979). [92]. Szumiel I., Jaworska A., Kapiszewska M., John A., Grądzka I., Sochanowicz B.: Differentia] induction of apoptosis in X-irradiated L5178Y sublines bearing p53 mutation. Radiat. Environ. Biophys., 39, 33-40 (2000). [93]. Szumiel I., Kapiszewska M., John A., Grądzka I., Kowalczyk D., Janik P.: Caffeine-inhibitable control of the radiation-induced G2 arrest in L5178Y-S cells deficient in non-homologous end-joining. Radiat. Environ. Biophys., 40, 137-143 (2001). [94]. Johanson K.J., Sundell-Bergman S., Szumiel I., Wlodek D.: Effects of irradiation and inhibition of adenosine diphosphate ribosyl transferase in radiation sensitive and resistant mouse lymphoma (L5178Y) cells. ActaRadiol. Oncol., 24, 451-457 (1985). [95]. Wojewódzka M., Kraszewski M., Sochanowicz B., Szumiel I.: Differential DNA double strand break fixation dependence on poIy(ADP-ribosylation) in L5178Y and CHO cells. Int. J. Radiat. Biol., 35, 577-581 (2004). [96]. Szumiel I., Wlodek D., Johanson K. J.: Differential effect of benzamide on NAD+ content and the frequency of chromatid aberrations in X-irradiated L5178Y-R and L5178Y-S cells. Acta Oncol., 27, 851-855(1988). [97]. Glaser K.B., Sung A., Bauer J., Weichman B.M.: Regulation of eicosanoid biosynthesis in the macrophage. Involvement of protein tyrosine phosphorylation and modulation by selective protein tyrosine kinase inhibitors. Biochem. Pharmacol., 45, 711-721 (1993). [98]. Van Buul P.P., Van Duyn-Goedhart A., Sankaranarayanan K.: In vivo and in vitro radioprotective effects of the prostaglandin El analogue misoprostol in DNA repair-proficient and -deficient rodent cell systems. Radiat. Res., 152, 398-403 (1999). [99]. Buraczewska I., Gasińska A., Grądzka I., Jarocewicz N., Sochanowicz B., Szumiel I.: Erbsta- tin-induced increase in apoptosis does not radiosensitize L5178Y cells. Nukleonika, 44, 561-578 (1999). [100]. Sochanowicz B., Szumiel I.: Cytoplasmic activity of tyrosine kinase protein, c-Abl, in L5178Y murina lymphoma sublines - a link to differential apoptosis proneness? In: INCT Annual Report 2000. Institute of Nuclear Chemistry and Technology, Warszawa 2001, pp.104-105. [101]. Neshat M.S., Raitano A.B., Wang H.G., Reed J.C., Sawyers C.L.: The survival function of the Bcr-Abl oncogene is mediated by Bad-dependent and -independent pathways: roles for phospha- tidylinositol 3-kinase and Raf. Mol. Cell Biol., 20, 1179-1186 (2000). [102]. Endlich B., Radford I.R., Forrester H.B., Dewey W.C.: Computerized video time-lapse microscopy studies of ionizing radiation-induced rapid-interphase and mitosis-related apoptosis in lymphoid cells. Radiat. Res., 153, 36-48 (2000). [103]. Okada S.: Formation of giant cells. In: Radiation Biochemistry. Vol. 1. Cells. Eds. K.I. Altman, G.B. Gerber, S. Okada. Academic Press, New York and London 1970, pp.239-246. [104]. Tauchi H., Sawada S.: Analysis of mitotic cell death caused by radiation in mouse leukaemia L5178Y cells: apoptosis is the ultimate form of cell death following mitotic failure. Int. J. Radiat. Biol., 65, 449-455 (1994).

43 [105]. Jaworska A., Szumiel I., De Angelis P., Olsen G., Reitan J.: Evaluation of ionizing radiation sensitivity markers in a panel of lymphoid cell lines. Int. J. Radiat. Biol., 77, 269-280 (2001). [106]. Suzuki A., Ito T., Kawano H., Hayashida M, Hayasaki Y., Tsutomi Y., Akahane K., Nakano T., Miura M., Shiraki K.: Survivin initiates procaspase 3/p21 complex formation as a result of interaction with Cdk4 to resist Fas-mediated cell death. Oncogene, 19, 1346-1353 (2000). [107]. Walicka M., Koerner I.J.: Lethal effects and DNA break rejoining in two L5178Y cell lines irradiated with fast neutrons and X-rays. Studia Biophys., 87, 47-56 (1982). [108]. Johanson K.J., Wlodek D., Szumiel I.: DNA repair and replication in radiation-sensitive and -resistant mouse lymphoma cells gamma-irradiated under aerobic and hypoxic conditions. Int. J. Radiat. Biol., 41, 261-270 (1982). [109]. Kraszewski M., Wojewódzka M., Iwanenko T., Szumiel I., Okuyama A.: Differential inhibitory effect of OK-103 5 on DNA repair in L5178Y murine lymphoma sublines with functional or defective repair of double strand breaks. Mutat. Res., 409, 31-36 (1998). [110]. Wlodek D., Olive P.L.: Neutral filter elution detects differences in chromatin organization which can influence cellular radiosensitivity. Radiat. Res., 132, 242-247 (1992). [111]. Szumiel I., Buraczewska I., Grądzka I., Gasińska A.: Effects of topoisomerase I-targeted drugs on radiation response of L5178Y sublines differentially radiation and drug sensitive. Int. J. Radiat. Biol., 67, 441-448 (1995). [112]. Kraszewski M., Iwaneńko T., Boużyk E., Szumiel I.: Chelating of iron and copper alters properties of DNA in L5178Y cells, as revealed by the comet assay. Mutat. Res., 434, 53-60 (1999). [113]. Kraszewski M., Iwaneńko T.: Induction of DNA breakage in X-irradiated nucleoids selectively stripped of nuclear proteins in two mouse lymphoma cell lines differing in radiosensitivity. Acta Biochim. Pol., 45, 701-704 (1998). [114]. Ljungman M.: The influence of chromatin structure on the frequency of radiation-induced DNA strand breaks: a study using nuclear and nucleoid monolayers. Radiat. Res., 126. 58-64 (1991). [115]. Ljungman M., Nyberg S., Nygren J., Eriksson M., Ahnstrom G.: DNA-bound proteins contribute much more than soluble intracellular compounds to the intrinsic protection against radiation-induced DNA strand breaks in human cells. Radiat. Res., 127, 171-176 (1991). [116]. Olive P.L., Banath J.P.: Radiation-induced DNA double-strand breaks produced in histone-depleted tumor cell nuclei measured using the neutral comet assay. Radiat. Res., 142, 144-152 (1995). [117]. Elia M.C., Bradley M.O.: Influence of chromatin structure on the induction of DNA double strand breaks by ionizing radiation. Cancer Res., 52, 1580-1586 (1992). [118]. Beer J.Z., Jacobson E.D., Evans H.H., Szumiel I.: X-ray and UV mutagenesis in two L5178Y cell strains differing in tumorigenicity, radiosensitivity, and DNA repair. Br. J. Cancer Suppl. 6, 107-111(1984). [119]. Evans H.H., Horng M.F.: The repair of potentially lethal damage and sublethal damage in strains of mouse L5178Y lymphoma cells differing in radiation sensitivity. Radiat. Res., 113, 183-190(1988). [120]. Evans H.H., Mencl J., Horng M.F., Ricanati M., Sanchez C, Hozier J.: Locus specificity in the mutability of L5178Y mouse lymphoma cells: the role of multilocus lesions. Proc. Natl. Acad Sci. USA, 83, 4379-4383 (1986). [121]. Sato K., Chen D.J., Eguchi-Kasai K., Itsukaichi H., Odaka T., Strniste G.F.: Interspecific complementation between mouse and Chinese hamster cell mutants hypersensitive to ionizing radiation. J. Radiat. Res., 36, 38-45 (1995). [122]. Stiff T., Shtivelman E., Jeggo P., Kysela B.: AHNAK interacts with the DNA ligase IV-XRCC4 complex and stimulates DNA ligase IV-mediated double-stranded ligation. DNA Repair, 3, 245-256 (2004). [123]. Lees-Miller S.P., Meek K.: Repair of DNA double strand breaks by non-homologous end joining. Biochimie, 85, 1161-1173 (2003). [124]. Pfeiffer P., Goedecke W., Kuhfittig-Kulle S., Obe G.: Pathways of DNA double-strand break repair and their impact on the prevention and formation of chromosomal aberrations. Cytogenet. Genome Res., 104, 7-13 (2004). [125]. Block W.D., Yu Y., Merkle D., Gifford J.L., Ding Q., Meek K., Lees-Miller S.P.: Auto- phosphorylation-dependent remodeling of the DNA-dependent protein kinase catalytic subunit regulates ligation of DNA ends. Nucleic Acids Res., 32,4351-4357 (2004). [126]. Kharbanda S., Pandey P., Jin S., Inoue S., Bharti A., Yuan Z., Weichselbaum R.R., Weaver D., Kufe D.: Functional interaction between DNA-PK and c-Abl in response to DNA damage Nature, 386, 732-735 (1997).

44 [127]. Beer J.Z., Lett J.T., Alexander P.: Influence of temperature and medium on the X-ray sensitivities of leukemia cells in vitro. Nature, 199, 193-194 (1963). [128]. Evans H.H., Horng M.F.: The repair of potentially lethal damage and sublethal damage in strains of mouse L5178Y lymphoma cells differing in radiation sensitivity. Radiat. Res., 113, 183-190(1988). [129]. Kapiszewska M., Lange C.S.: The effects of reduced temperature and/or starvation conditions on the radiosensitivity and repair of potentially lethal damage and sublethal damage in L5178Y-R and L5178Y-S cells. Radiat. Res., U3, 458-472 (1988). [130]. Kapiszewska M., Lange C.S.: The effect of hypothermic treatment on the repair or expression of X-ray damage measured in split dose experiments on L5178Y-S cells. Radiat. Environ. Biophys.,28, 303-317 (1989). [131]. Radford I.R.: Transcription-based model for the induction of interchromosomal exchange events by ionizing irradiation in mammalian cell lines that undergo necrosis. Int. J. Radiat. Biol., 78, 1081-1093(2002). [132]. Boothman D.A., Fukunaga N., Wang M.: Down-regulation of topoisomerase I in mammalian cells following ionizing radiation. Cancer Res., 54, 4618-4626 (1994). [133]. Szumiel I.: Requirements for potentiation of radiation effect by a platinum complex. Int. J. Radiat. Biol., 33, 605-608 (1978). [134]. Grądzka I., Buraczewska I., Kuduk-Jaworska J., Romaniewska A., Szumiel I.: Radiosensitizing properties of novel hydroxydicarboxylatoplatinum(II) complexes with high or low reactivity with thiols: two modes of action. Chem.-Biol. Interact., 146, 165-77 (2003). [135]. Bennett C.B., Lewis L.K., Karthikeyan G., Lobachev K.S., Jin Y.H., Sterling J.F., Snipe J.R., Resnick M.A.: Genes required for ionizing radiation resistance in yeast. Nat. Genet., 29, 426-434 (2001).

45 UKD: 576.32/36 INIS: S63 KEY WORDS: L5178Y MURINE LYMPHOMA, LY SUBLINES, UVC RADIATION, IONISING RADIATION, PLATINUM COMPLEXES, HYDROGEN PEROXIDE, SUR- VIVAL, APOPTOSIS, DNA REPAIR, DOUBLE STRAND BREAK REPAIR, BASE DAMAGE REPAIR, DNA-PK, DNA DAMAGE FIXATION, ION CONTENT, LIPID COM- POSITION OF PLASMA MEMBRANES, CHROMATIN CONFORMATION, KARYO- TYPE, DNA DAMAGE RESPONSE, RADIOSENSITIVITY, PHOTO SENSITIVITY, ANTI- OXIDANT DEFENCE, CAFFEINE, IRON, LABILE IRON POOL, TELOMERES, PO- LY(ADP-RIBOSYLATION).