MARIAANDERSSON ChemicallyInducedDNADamageinExtended-termCulturesofHumanLymphocytes 38 DigitalComprehensiveSummariesofUppsalaDissertationsfromtheFacultyofPharmacy

Full text

(1)Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Pharmacy 38. Chemically Induced DNA Damage in Extended-term Cultures of Human Lymphocytes MARIA ANDERSSON. ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2006. ISSN 1651-6192 ISBN 91-554-6674-5 urn:nbn:se:uu:diva-7179.

(2)

(3)

(4)

(5)

(6)

(7) .

(8) ! " # $$% $&'() * + * * ,+ + -! * ,+ ./ 0+

(9)

(10) 1

(11)

(12) 21+/ 3

(13)

(14) / $$%/ + 4

(15) 53

(16) 6

(17) 7 *

(18) 8 + / 3

(19)

(20) /

(21)

(22)

(23)

(24)

(25) 9/ $ / / 425 &(7))7%%#7)/ :

(26)

(27) * 53

(28)

(29)

(30)

(31)

(32) +

(33)

(34) /

(35) ;

(36) * +

(37) **

(38) *

(39) **

(40) +

(41)

(42)

(43)

(44) * +

(45)

(46) + +

(47) *

(48)

(49)

(50)

(51)

(52)

(53) 1 * / 0+ * + + 1 1+ + +

(54)

(55) * +

(56) +

(57) 7 * +

(58) + -60.

(59)

(60)

(61)

(62)

(63) *

(64)

(65)

(66) * +

(67) 1+

(68)

(69) +

(70)

(71) * + /

(72) +

(73) * *

(74)

(75) + 1 + + 60 1 *

(76) + 537

(77) ** 1 +

(78) < **

(79) 1+ +

(80)

(81) + /

(82)

(83) 1 + 17 +

(84)

(85) + + 8)(#9= + 60 + 1

(86) + 53

(87) ** * +

(88)

(89) +

(90)

(91) + / 3 + + + 1

(92)

(93)

(94)

(95)

(96)

(97) 53

(98) + ;

(99) 1+

(100)

(101) 60 * **

(102)

(103)

(104) *

(105) * +

(106) *

(107) * 537

(108)

(109) / 0+

(110) **

(111)

(112)

(113) + 7

(114) 53 1

(115) *+ + +

(116) 60 1 + **

(117) * + **

(118) + 1

(119) * + 537

(120)

(121) +

(122)

(123)

(124) /

(125) + + + 60

(126) + *

(127)

(128) 1+

(129)

(130) * +

(131) 53

(132) ** * + / >

(133)

(134)

(135) +

(136) +

(137) < 1+

(138)

(139) *

(140)

(141) *

(142)

(143) 7 *

(144) +

(145) *

(146)

(147)

(148) * +

(149) / 4

(150) * +

(151) 7 + 1 + 1 + 2&

(152) ?

(153) * 53

(154) @ + *;

(155) 1+

(156)

(157)

(158) * +

(159)

(160) * + / 537

(161)

(162) 53 6

(163) 7 *

(164) 8 + :

(165) 4

(166) 53 4

(167) 8 + . , > "

(168) 2 ! "

(169)

(170) # " # $ %&'" " ()*%'+, " A 3

(171)

(172) $$% 4225 (%)(7%(& 425 &(7))7%%#7)

(173) '

(174)

(175) ''' 7#(#& -+ '??

(176) /</? B

(177) C

(178) '

(179)

(180) ''' 7#(#&..

(181) LIST OF PAPERS. This thesis is based on the following original papers, which will be referred to in the text by their roman numerals. I. II. III. IV. Andersson, M., Agurell, E., Vaghef, H., Bolcsfoldi, G., Hellman, B. (2003) Extended-term cultures of human lymphocytes and the comet assay: A useful combination when testing for genotoxicity in vitro? Mutation Research 540, 43-55. Andersson, M., Hellman, B. (2005) Different roles of Fpg and Endo III on catechol-induced DNA damage in extended-term cultures of human lymphocytes and L5178Y mouse lymphoma cells. Toxicology In Vitro 19, 779-786. Andersson, M., Stenqvist, P., Hellman, B. Impact of interindividual differences in initial DNA repair capacity when evaluating DNA damage using extended-term cultures of human lymphocytes and the comet assay. Submitted. Andersson, M., Hellman, B. Evaluation of catechol-induced DNA damage in human lymphocytes: A comparison between freshly isolated lymphocytes and T-lymphocytes from extended-term cultures. Submitted.. Reprints were made with permission from the publishers..

(182)

(183) CONTENTS. INTRODUCTION .......................................................................................9 Cell cultures ...............................................................................................................9 Primary DNA damage..............................................................................................10 Measurement of DNA damage by the comet assay ...................................................11. AIMS OF THE PRESENT THESIS........................................................... 14 EXPERIMENTAL PROCEDURES ........................................................... 15 Isolation of human lymphocytes and the establishment of mass cultures and subsequent extended-term cultures .............................................................15 Mouse lymphoma cells .............................................................................................15 The comet assay ........................................................................................................16 Detection of reactive oxygen species ........................................................................16 DNA isolation, PCR and DNA sequencing ..............................................................17 Exposure conditions and substances .........................................................................17. RESULTS AND DISCUSSION .................................................................. 19 Validation of the extended-term cultures of human lymphocytes when screening for genotoxicity ..................................................................................19 Influence of interindividual differences between blood donors when using extended-term cultures of human lymphocytes for monitoring DNA damage and repair....................................................................................21 Comparison between extended-term cultures of human lymphocytes and mouse lymphoma L5178Y cells ...................................................................23 Comparison between extended-term cultures of human lymphocytes and freshly isolated peripheral blood lymphocytes .............................................27 Clouds – an unresolved issue ....................................................................................29. CONCLUSIONS AND FINAL REMARKS ..............................................32 ACKNOWLEDGEMENTS ........................................................................34 REFERENCES ............................................................................................36.

(184)

(185) ABBREVIATIONS. 4NQO 5-HU B(a)P CP CrPic Endo III ETC FA Fpg GR 20 H2O2 HIFCS LOEL MLC NOEL PBL PSGP R10P RD ROS RPMI 1640 S9 SR 20. 4-nitroquinoline-N-oxide 5-hydroxyurea benzo[a]pyrene cyclophosphamide chromium picolinate endonuclease III extended-term cultures of human lymphocytes formaldehyde formamidopyrimidine DNA glycosylase growth medium for human lymphocytes hydrogen peroxide heat-inactivated foetal calf serum lowest observed effect level mouse lymphoma cells no observed effect level freshly isolated peripheral blood lymphocytes penicillin/streptomycin/glutamine/sodium pyruvate culture medium for mouse lymphoma cells residual (DNA) damage reactive oxygen species Rosewell Park Memorial Institute 1640 medium post mitochondrial supernatant from a rat liver homogenate growth-stimulating medium for human lymphocytes.

(186)

(187) DNA damage in human lymphocytes. INTRODUCTION. Humans are continuously exposed to genotoxic chemicals which can damage the DNA. Today it is well known that DNA damaging effects of chemical agents are associated with mutagenic and carcinogenic events, which could be the starting point for the development of cancer (Pitot and Dragan 1994). The basic concept describing the development of cancer is mutagenic events leading to either “achievement of functions” due to oncogene activation or “loss of functions” by inactivation of tumour suppressor genes (Ponder 2001). Consequently, a panel of tests have been developed to detect different types of genotoxic effects in order to predict the potential genotoxicity and mutagenicity of chemicals, including both pharmaceutical drugs and various types of environmental and occupational agents, as well as dietary factors. The information obtained in such studies are primarily used for hazard identification but can also be used in the subsequent risk assessment.. Cell cultures The in vitro screening for the potential genotoxicity of chemicals is usually performed in prokaryotic cells (e.g. in Salmonella typhimurium in Ames test) or in transformed cells from experimental animals (e.g. in Chinese hamster ovary cells in the HPRT assay or in mouse lymphoma cells in the TK assay). However, primary cells of human lymphocytes have also been used for this purpose. Established rodent cell lines, such as Chinese hamster ovary cells and mouse lymphoma L5178Y cells (MLC), have the practical advantages of being easy to handle and well characterized with regard to, i.e. growth pattern and DNA repair capacity. However, there are some, at least theoretical drawbacks associated with these cell lines. Their chromosomal material has undergone extensive rearrangement and their chromosome number may vary from cell to cell (Scott et al., 1990). Moreover, key genes involved in the viability of cells sustaining DNA damage such as p53, may be mutated. These abnormalities raise questions whether these cell lines respond to insults by mutagens and clastogenes in the same manner as human cells or the primary cells with a normal karyotype, from which they were originally derived. Primary cell cultures from human peripheral blood can easily be obtained from volunteers or blood donors, who regularly donates blood at a hospital. Primary cultures of human lymphocytes have the obvious advantage of being karyotypically normal, but they are poorly defined biologically and individual variability is observed due to day-to-day differences of human health and habits. Furthermore, only a restricted number of experiments can be performed on primary cultures of human 9.

(188) Maria Andersson. lymphocytes isolated from one particular blood sample, and these experiments should ideally be performed shortly after the blood sampling. Additionally, humans are known to be genetically polymorphic for various DNA repair enzymes (Harris 1989) and this interindividual variation can result in differences in responses to a particular chemical. To overcome the real and anticipated weaknesses with rodent cell lines and primary cultures of human lymphocytes, O’Donovan et al. (O’Donovan et al., 1995a) suggested that extended-term cultures of human T-lymphocytes (ETC) might be a convenient alternative when testing for the potential genotoxicity of chemicals in vitro. Following a simplified protocol using recombinant human interleukin-2 (rIL-2), heat-inactivated foetal calf serum (HIFCS) and freeze-killed feeder cells to develop mass cultures of T-lymphocytes from human peripheral blood, O’Donovan et al. (O’Donovan et al., 1995a) showed that cultures could be initiated and grown from a panel of normal human blood donors and that the growth of the T-cells from the different donors was very similar after the cryopreservation. O’Donovan et al. (1995a,b) have also shown that these cells are karyotypically normal for at least 15 days in culture, that cultures from different donors cryopreserved on day 8 and then grown for a further week on recovery contain very similar populations of cells as determined by surface antigen expression, and that T-cells from extended-term cultures can be cultured with little variability between experiments.. Primary DNA damage Chemical compounds can cause base alterations in the DNA sequence. Many electrophilic chemicals (e.g. metabolites of benzo[a]pyrene and 4-nitroquinolineN-oxide) can bind covalently to DNA, forming bulky adducts (Lutz 1979). Monofunctional alkylating agents, such as N-nitrosopiperidine, can add alkyl groups to DNA. Alkylated bases can also lead to secondary alterations in DNA, e.g. an alkyl adduct labilizes the bond that connects the base to deoxyribose, thereby stimulating base loss. Base loss from DNA leaves an apurinic or apyrimidinic site, commonly called an AP site. A subsequent insertion of an incorrect base into the AP site can cause a mutation (Laval et al., 1990). Polyfunctional alkylating agents, such as cyclophosphamide, can also alkylate DNA and induce various types of damage, including DNA/DNA crosslinks (Erickson et al., 1980, Crook et al., 1986, Little and Mirkes 1987, Anderson et al., 1995). The oxidation of DNA bases is produced by reactive oxygen species (ROS). ROS, such as hydrogen peroxide, superoxide and hydroxyl radicals, are generated both from endogenous sources and from the reactions of chemicals (Ames 1989, Frenkel 1992). There are a number of oxidized pyrimidines and purines that can lead to mutations if left unrepaired. The mutagenic potential of DNA oxidation is clearly shown by the mutagenicity of ionizing radiation which produces DNA oxidation (Hall et al., 1988). DNA strand breaks (single and double) are also a type of primary DNA damage, formed both endogenously and induced by different types of exogenous agents, including ionizing radiation. Replication,. 10.

(189) DNA damage in human lymphocytes. recombination, transcription and certain types of DNA repair are endogenous processes that generates DNA strand breaks (Gedik et al., 1992). Moreover, most of the DNA alterations mentioned above can potentially be transformed to single or double strand breaks, as the DNA repair machinery incise the DNA damage in order to remove and replace it with a new and undamaged DNA base (Preston and Hoffmann 2001). The induction and repair of DNA damages can be studied by different assays. One such method is the comet assay, which detects single strand breaks, double strand breaks and alkali labile sites.. Measurement of DNA damage by the comet assay The comet assay is a cheap, rapid and simple method that detects DNA damage in single cells. In 1984, Östling and Johanson developed a microgel electrophoresis technique, where they irradiated cells with Ƣ-rays and registered that nucleoids with damaged DNA were stretched toward the anode, while undamaged nucleoids had a round appearance (Ostling and Johanson 1984). In 1988, Singh et al. modified the assay by introducing an alkaline condition during the electrophoresis (Singh et al., 1988). Using the latter version of the assay, it is possible to detect DNA single strand breaks, alkali-labile sites, DNA/DNA–and DNA/protein crosslinks (Tice et al., 2000). The comet assay, which has become an increasingly popular method because of its high sensitivity, has been applied in a broad range of scientific fields, including genetic toxicology (Plappert et al., 1994, Rojas et al., 1999, Andersson et al., 2003, Andersson et al., 2006), ecotoxicology (Cotelle and Ferard 1999), DNA repair studies (Hu et al., 1995, Collins et al., 2001, Schmezer et al., 2001, Collins and Harrington 2002, Mayer et al., 2002, Marcon et al., 2003), and studies for human biomonitoring (Albertini et al., 2000, Moller 2005). By using the comet assay it is possible to detect low levels of DNA damage in individual cells. Moreover, only a small number of cells are needed in each experiment which means that, as long as a single cell suspension can be obtained, almost any tissue can be used. As single cells are visualized, it is possible to detect intercellular differences in response to DNA damaging agents. Recently, an expert panel reached a consensus that the alkaline (pH > 13) condition is currently the most optimal version of the comet assay when used for identifying agents with genotoxic actions (Tice et al., 2000). The fundamental principle of the comet assay is to detect DNA damage by monitoring movement of DNA in an agarose gel. Firstly, the cells are placed in a lysing solution with detergents and high salt concentration. The cells are permeabilised with the detergent, while the salt extracts the nuclear proteins (McKelvey-Martin et al., 1993). The remaining DNA is caught in a nucleus-like structure, the nucleoid (Cook and Brazell 1976). Secondly, the supercoiled DNA is relaxed when the nucleoids are put in an alkaline electrophoresis solution. Relaxation of the supercoiled DNA prepares the DNA for the electrophoresis. An electric field is applied on the agarose. 11.

(190) Maria Andersson. gel and the DNA loops are able to extend towards the anode and they move in proportion to the size of the damaged DNA, creating the well-known comet trace profile. Finally, the agarose gel is gently neutralised to remove alkali and detergents to avoid interference with the ethidium bromide staining (Singh et al., 1988). In order to detect specific DNA damage, Collins et al. (Collins et al., 1993, Collins et al., 1996, Collins et al., 1997) introduced a modified version of the comet assay by introducing an enzymic DNA digestion step, using for example endonuclease III (Endo III) for detection of oxidized pyrimidines (Collins et al., 1993) and formamidopyrimidine DNA glycosylase (Fpg) for detection of oxidized purines (Collins et al., 1996). Endo III was historically characterized for having activity against oxidatively damaged thymine residues and urea residues (Katcher and Wallace 1983, Breimer and Lindahl 1984). In a first step, both enzymes function as DNA glycosylases, i.e. they remove damaged DNA bases. In a subsequent step, Endo III hydrolyses the DNA at the base-free site through its endonuclease activity, while Fpg instead nicks the DNA backbone at the base-free site by ơ-elimination through its lyase activity (Boiteux et al., 1987, Doetsch et al., 1987, Bailly et al., 1989). Today, Endo III is known to recognize a diverse array of oxidized pyrimidines, some of which are known to be highly premutagenic lesions (Basu et al., 1989, Feig et al., 1994, Kreutzer and Essigmann 1998). Fpg has been reported to excise a variety of modified bases from DNA, mainly oxidized purines. The main substrate of Fpg seems to be 8-oxodeoxyguanine (8-oxodG), which probably is the most abundant base oxidation product to be found in DNA. The 8-oxodG lesion is the most commonly used biomarker for oxidative DNA damage (Boiteux et al., 1992, Collins et al., 1996, Collins et al., 1997, Zharkov et al., 2003), and it is also known to be a premutagenic lesion (Shibutani et al., 1991, Cheng et al., 1992, Moriya 1993). The DNA migration can be monitored using various parameters such as the tail length, the tail moment and/or the percentage of DNA in the tail (Hellman et al., 1995). The approach to measure the length of the comet tail was one of the earliest. Singh et al. (1988) for example, used an eyepiece micrometer to measure the DNA migration on a photomicrograph. Another parameter is the tail moment, which was introduced by Olive et al. (Olive et al., 1990) and was defined as the tail length multiplied by the percent of the fluorescence in the tail. The tail moment has been frequently used ever since the introduction of image analysis systems, which made it easy to compute the tail moment. Moreover, the pioneers of the comet assay regarded the tail moment as one of the best indices of the break frequency determined by the comet assay (Collins 1992). However, there has been various ways to calculate the tail moment because there has been altered definitions of the starting point of the tail. As a result, different calculations of the tail moment have been based on the tail length from the centre of the head to the end of the tail (Olive et al., 1990), from the centre of the head to the centre of gravity of the tail (Hellman et al., 1995), and from the trailing edge of the head to the end of the tail (Ashby et al., 1995). Lately, there has been extensive discussions within the comet assay community about what comet parameter to use. Several authors recommend that the percent of DNA in the tail (%TDNA) should be the preferred parameter, since it gives a clear indication of 12.

(191) DNA damage in human lymphocytes. the appearance of comets and it is linearly related to the break frequency (Collins 2002, Hartmann et al., 2003). It has also been indicated that %TDNA facilitates a more accurate comparison and comparable measurement of DNA damage between different laboratories (Moller 2005).. 13.

(192) Maria Andersson. AIMS OF THE PRESENT THESIS. The present thesis focuses on whether the extended-term cultures of human lymphocytes can be used as an alternative in vitro system to the more commonly used transformed mammalian cell lines, and primary cell cultures from humans, when testing the potential genotoxicity of chemicals. The aims were: (i) (ii) (iii). (iv). 14. to validate the extended-term cultures with 7 different model substances, using the comet assay (Paper I) to compare the extended-term cultures with mouse lymphoma L5178Y cells, commonly used in genotoxicity testing (Paper II) to examine the importance of possible interindividual differences between different blood donors when evaluating induced DNA damage and its subsequent repair (Paper III) to evaluate differences between the extended-term cultures of human T-lymphocytes and its freshly isolated peripheral blood lymphocytes after exposure to DNA damaging agents (Paper IV)..

(193) DNA damage in human lymphocytes. EXPERIMENTAL PROCEDURES. Isolation of human lymphocytes and the establishment of mass cultures and subsequent extended-term cultures Isolation of human lymphocytes were made from fresh blood samples, collected from healthy male and female blood donors. Within 4 h after the blood collection, the initiation of mass cultures of lymphocytes was started, basically following the original protocol of O’Donovan et al. (1995). The blood was diluted with HBSS and the lymphocytes were separated by centrifugation over Percoll. The gained “buff y coat” was washed twice in HBSS before the final pellet was resuspended in a growth-stimulating medium, SR 20, containing RPMI 1640, 20 % HIFCS, 5 % PSGP, 500 U/ml rIL-2 and 0.5 g/ml PHA, to establish a mass culture. The mass culture was incubated together with feeder cells (freeze-killed GM 1899A Blymphoblastoids) at 37 °C in a humidified atmosphere of 5 % CO2 / 95 % air. On day 4, and daily thereafter, the cells were disaggregated by careful pipetting. On day 6, the cell suspension was diluted 1:1 with a growth medium, GR 20, consisting of all the components of SR 20 except the mitogen PHA. The cells were harvested on day 8 and directly suspended in a freeze medium, containing 10 % sterile DMSO in HIFCS (3 × 106 cells/ml), and subsequently frozen overnight at –70 °C (–1 °C/min). Thereafter, the cells were kept in –140 °C. In order to recover the cryopreserved lymphocytes and establish them as an extended-term culture, one ampoule of cells was rapidly thawed and washed accurately. The lymphocytes were finally resuspended in SR 20 and put into a cultivation flask before the culture was left in humidified air for 3 days. On day 4, and daily thereafter, the lymphocytes were disaggregated by pipetting. On day 7, GR 20 was added and the cells were left undisturbed for another 3 days before they were disaggregated again. Desired cell density, for the experiments, was normally reached on days 11 – 13.. Mouse lymphoma cells Heterozygous L5178Y TK+/- 3.7.2.C cells (originally obtained from Dr D. Clive, Burroughs Wellcome Co., Research Triangle Park, USA) were generously supplied to us by Dr G. Bolcsfoldi (AstraZeneca R&D Södertälje, Sweden). To initiate a culture of mouse lymphoma cells, one cryopreserved ampoule containing 3 × 106 cells was rapidly thawed and then instantly washed in an excessive volume of RPMI. 15.

(194) Maria Andersson. 1640 before they were resuspended in 10 ml growth medium (R10P), containing 9 % HIFCS, 4 % penicillin/streptomycin (4000 U/ml)/glutamine (40 mM)/ sodium pyruvate (40 mM) and 0.9 % Pluronic F68 solution. An aliquot of 5 ml of this suspension was transferred to a 75 cm2 cultivation flask with 95 ml R10P. Experiments with the cells were performed on cultivation day 4 or 5.. The comet assay Our standard procedure for the alkaline version of the comet assay (Papers I, III–V) is based on a slightly modified protocol of Singh et al. (Singh et al., 1988). Briefly, 60 l of a mixture of 30 l cell suspension in RPMI and 210 l 0.8 % (w/v) low-melting point agarose, was layered on top of an ordinary microscope slide precoated with 0.8 % (w/v) agarose. To avoid slippery gels, the slides were engraved about 5 mm from each of its four edges before the precoating, and they were also coded in order to make a blindfolded evaluation of DNA damage after the electrophoresis. The agarose was allowed to set on ice for 10 – 15 min and then the slides were immersed in cold lysis solution (2.5 M NaCl, 100 mM Na 2-EDTA, 10 mM Tris, pH adjusted to 10.0 with NaOH, and 1 % Triton X-100 and 10 % DMSO added before use) at 4 ºC for 1 h in order to remove cellular proteins. All slides were then placed in a horizontal gel electrophoresis chamber containing alkaline electrophoresis buffer (1 mM Na 2EDTA, 300 mM NaOH, pH > 13) for 40 min, where the supercoiled DNA can relax. Electrophoresis was in general performed for 10 min using a field strength of 0.7 V/cm (300 mA, 25 V). Subsequently, the slides were carefully rinsed with 0.4 M Tris, pH 7.5, for 15 min, dried at room temperature and stored in a sealed box (avoiding dust and other particles) until the day of image analysis (Vaghef et al., 1996). All steps from the lysis until the end of neutralization were performed under yellow light. By adding the lesion-specific repair enzymes Fpg and Endo III, from Escherichia coli, it was also possible to monitor oxidative base damage in the comet assay, since they convert oxidised purines or pyrimidines into single strand breaks (Collins et al., 1993, Collins et al., 1996, Collins et al., 1997). In the modified version of the alkaline comet assay (Paper II), the slides were washed 3 times in an enzyme buffer (40 mM HEPES, 0.1 M KCl, 0.5 mM EDTA, 0.2 mg/ml bovine serum albumin, pH 8.0) after the lysis, before being covered with 40 – 60 µl of either buffer alone, or enzyme (1 µg/ml of Fpg or Endo III) dissolved in buffer and sealed with a cover slip. The cells were incubated with the enzymes for 30 – 45 min at 37 ºC. After the enzyme incubation, the slides were placed in the electrophoresis buffer for 40 min, and the subsequent steps followed our standard protocol for the alkaline version of the comet assay.. Detection of reactive oxygen species A carboxy derivative of fluorescein, 5-(and-6)-carboxy-2’,7’-dichlorodihydro16.

(195) DNA damage in human lymphocytes. fluorescein diacetate (carboxy-H2DCFDA), was used as a cell-permeant indicator of reactive oxygen species (ROS). A chemically reduced and acetylated form of 2’,7’-dichlorofluorescein is not fluorescent until the acetate groups are removed by intracellular esterases and oxidation occurs within the cell (Brandt and Keston 1965, Keston and Brandt 1965, Cathcart et al., 1983, Jakubowski and Bartosz 2000). Before the probe was added to the cells they were washed in PBS. The probe was dissolved in HBSS to a final concentration of 20 M, and added to the cells for 30 minutes at 37 °C. Oxidation of the probe was detected by monitoring the increase in fluorescence with a 96-well plate in a FLUOstar multiwell fluorescence plate reader, using an excitation filter of 485 nm and an emission filter of 520 nm.. DNA isolation, PCR and DNA sequencing Genomic DNA from the extended-term cultures of human lymphocytes and mouse lymphoma cells were isolated using a Qiagen kit, DNeasy (Qiagen Inc., USA). The concentration and purity of DNA was determined by spectrophotometry and the 260/280 ratio, respectively. Human and mouse p53 exon 5 were individually amplified using designed primers to amplify fragments of 435 base pairs covering the region of interest. Primers for human exon 5 were: Fwd 5´-CCT CAA CAA GAT GTT TTG CCA ACT G-3´ and Rev 5´-TAG GGC ACC ACC ACA CTA TGT CG-3´. Primers for mouse exon 5 were: Fwd 5´-CCT CCC CTC AAT AAG CTA TTC TGC C-3´ and Rev 5ÁTA AGG TAC CAC CAC GCT GTG GC-3´. The DNA was amplified in a 20 l reaction [containing 116 ng DNA, 0.3 mM of each deoxynucleotide triphosphate, 1x Gold Buffer, 1.5 mM MgCl2, primers (0.3 mM each) and 7.5 U of AmpliTaq GoldTM DNA polymerase], in a PTC-200 thermal cycler. The PCR program was as follows: denaturation at 94 °C for 5 min followed by 5 cycles of 94 °C for 30 s, 64 °C for 30 s, 72 °C for 30 s, 30 cycles of 94 °C for 30 s, 59 °C for 30 s, 72 °C for 30 s, and a final extension of 72 °C for 10 min. PCR-amplified products were purified using QIAquick® Gel Purification Kit. The purified PCR products were sequenced using the Fwd and Rev PCR primers and the DYEnamicTM ET dye terminator kit and analyzed on a MegaBACE 1000 capillary instrument. Sequences were aligned using Sequencer 3.1.1 software (Gene Codes Corporation Inc., USA).. Exposure conditions and substances The exposure conditions and substances used in Papers I – IV are shown in Table 1. In this thesis, slightly different exposure conditions were used depending on what substance that was examined. The used substances were selected as model substances for DNA damage.. 17.

(196) 18 PBS. H2 O2 (exposure embedded H2 O2 (exposure embedded in gel for 10 min at 4 °C) in gel for 10 min at 4 °C). 3h 37 °C CAT. RPMI H2 O2 (exposure in medium for 10 min at 37 °C). No 3h 37 °C B(a)P CP EtOH FA 5-HU DMSO / ddH2 O 4NQO / NNP. Presence of DNA repair enzymes. Exposure time. Exposure temperature. Substance. Negative control. Positive control. CAT. H2 O2. RPMI. 37 °C. 3h. No. No. 4 °C. 10 min. Yes. No. The cells were exposed for in total 9 chemical compounds: 4-nitroquinoline-N-oxide (4NQO), benzo[a]pyrene (B(a)P), catechol (CAT), cyclophosphamide (CP), ethanol (EtOH), formaldehyde (FA), 5-hydroxyurea (5-HU), hydrogen peroxide (H2O2) and N-nitrosopiperidine (NNP). The negative control solutions were: double distilled water (ddH2O), dimethyl sulfoxide (DMSO), phosphate buffered saline (PBS) and Rosewell Park Memorial Institute medium (RPMI).. No. No. 0%. Yes. 0%. Presence of S9-mix. 5%. ETC / PBL. 20 %. ETC / MLC. Paper IV. Content of serum. ETC / MLC. Paper III. ETC. Paper II. Type of cells used. Paper I. TABLE 1. Exposure conditions used in Papers Ɛ ï ƐV.. Maria Andersson.

(197) DNA damage in human lymphocytes. RESULTS AND DISCUSSION. Validation of the extended-term cultures of human lymphocytes when screening for genotoxicity In order to evaluate the usefulness of the ETC in the comet assay, seven reference compounds were used. The bulky DNA adduct-forming compound 4-nitroquinoline-N-oxide (4NQO), is a direct-acting mutagen recommended as a positive control in mutagenicity assays performed without exogenous metabolic activation by mammalian microsomal enzymes (OECD 1981, Yen et al., 2001). 4NQO increased the primary DNA damage in the human lymphocytes in a dose-dependent manner and the lowest observed effect level (LOEL) was found to be 0.1 M. It has been shown that 4NQO induces DNA strand breaks, base pair change mutations, frameshift mutations, deletions, mitotic gene conversion, chromosomal aberrations and other types of genetic alterations without addition of an exogenous metabolic activation (S9) (Nagao and Sugimura 1976, Frenzilli et al., 2000), but it still needs to be metabolically activated before it can interact with nucleophilic sites in DNA and other cellular macromolecules (Endo 1971, Nagao and Sugimura 1976, Bailleul et al., 1989, Fronza et al., 1992, Inga et al., 1994). The enzyme that catalyses the reduction of 4NQO has been identified as NADP(H):quinone oxidoreductase 1 (QR 1) (Cavelier and Amzel 2001) and the results from Paper I clearly indicates that this enzyme is present in the ETC. In accordance with its direct-acting effect, the DNA-damaging effect of 4NQO was completely abolished when S9 was added during the exposure in the ETC. The DNA damaging effect of 4NQO has also been monitored in freshly isolated human leukocytes, using the alkaline version of the comet assay. Four hours of exposure without S9 was found to increase the DNA migration at the lowest concentration tested, i.e. 100 M (Frenzilli et al., 2000). Benzo[a]pyrene (B(a)P) is also a DNA adduct-forming agent but was shown to require exogenous metabolic activation (S9) to increase the tail moment after 3 h of exposure to 0.05 mM. The DNA damaging effect of B(a)P has also been monitored in freshly isolated leukocytes (Frenzilli et al., 2000) where an exposure of 4 h to a concentration as low as 0.004 mM, increased the DNA damage in the presence of S9. However, when the serum level was decreased during the exposure of the ETC, from approximately 20 % to 4 %, B(a)P was found to induce significant DNA damage in the extended-term cultures already at 0.01 mM in the presence of S9. The 19.

(198) Maria Andersson. fact that the serum concentration during the exposure may be of great importance has also been reported by others (Anderson et al., 1994). Furthermore, we have shown that HIFCS can protect ETC from chromium picolinate (CrPic)-induced DNA damage (Andersson et al., 2006). It was quite obvious that the presence of serum during the exposure of cells completely abolished the DNA-damaging effect of this compound. The reason for this is not known, but it seems rather natural to assume that the serum contained components that inhibited the weak DNAdamaging effect of CrPic. It was not possible to elucidate the mechanisms behind the observed increase in DNA damage in the cells that had been exposed to CrPic in the absence of serum, but as suggested by other authors (Bagchi et al., 1997, Hepburn et al., 2003) the damage may very well have followed from a CrPic-induced oxidative stress. From this point of view it is interesting to note that HIFCS has been reported to reduce the level of oxidative DNA damage in human lymphocytes exposed to H2O2 in vitro (Anderson et al., 1994). Another, maybe more plausible explanation to the protective effect of serum, is that CrPic binds to serum proteins preventing its access to the cells. This suggests that the sensitivity of the system is increased if the cells are exposed in the absence of serum. Due to this increased sensitivity, we exposed the ETC to catechol in absence of serum in Paper IV, while we in a previous paper (Paper II) had exposed the cells to catechol dissolved in medium with 5 % serum. However, judging from our results in Paper II and Paper IV, the serum concentration does not seem to be of critical importance for the DNA-damaging effects, at least not when catechol is used as the DNA damaging agent. Still, the overall impression is that the serum concentration during the exposure is of great importance, although it should be taken into account that cells are exposed under non-physiological conditions when 0 % serum is used. The DNA-damaging effect of the polyfunctional alkylating agent cyclophosphamide was found to increase the tail moment in a dose-dependent manner, and the LOEL was found to be 0.05 mM after 3 h exposure in the presence of S9. Cyclophosphamide did not have any DNA-damaging effects when tested without S9. A similar LOEL (0.072 mM) was reported when the comet assay was used to study the DNA damaging effects of cyclophosphamide in freshly isolated human leukocytes (Frenzilli et al., 2000), but in the latter study 4 h of exposure was found to induce DNA damage even in the absence of S9 (LOEL: 0.36 mM). The monofunctional alkylating agent N-nitrosopiperidine was found to increase the DNA migration, both in the absence and presence of S9. There was a clear difference in LOELs (0.5 mM with S9 and 10 mM without S9). This would be expected since it is known that N-nitroso compounds require cytochrome P450dependent metabolic activation for the generation of DNA-reactive species (Kamataki et al., 2002). Whereas ethanol did not increase the tail moment at all, 5-hydroxyurea, a nonDNA-reactive mutagen, increased the tail moment at high concentrations (LOEL: 10 mM). The DNA/protein-cross linking agent formaldehyde was the only compound tested that significantly reduced the tail moment. This effect was observed when 20.

(199) DNA damage in human lymphocytes. the extended-term cultures were exposed in the absence of S9 (LOEL: 0.1 mM). Reduced DNA migration after formaldehyde exposure has also been reported by others. In freshly isolated human leukocytes, formaldehyde was found to reduce the tail moment after 4 h of exposure to 0.8 mM (Frenzilli et al., 2000). In a study on three different types of cross linking agents using the alkaline version of the comet assay, Merk and Speit (1999) noticed that the DNA/protein-cross linking agent formaldehyde reduced the Ƣ-ray induced DNA migration in V79 Chinese hamster cells. No such effect could be seen in cells that had been exposed to mitomycin C (an inducer of DNA/DNA-interstrand crosslinks) or cisplatin (an inducer of DNA/ DNA-intrastrand crosslinks) (Merk and Speit 1999).. Influence of interindividual differences between blood donors when using extended-term cultures of human lymphocytes for monitoring DNA damage and repair It is well known that there are interindividual differences in sensitivity to DNAdamaging agents and in DNA repair capacity between freshly isolated lymphocytes from different donors (Pero et al., 1978, Holz et al., 1995, Collins et al., 2001, Janssen et al., 2001). To investigate if different blood donors could influence the outcome of DNA damaging effects of chemicals and the subsequent DNA repair in established ETC, we used the DNA damaging agent hydrogen peroxide (H2O2) as a model substance when performing the alkaline version of the comet assay (Paper III). The exposure to 0.25 mM H2O2 was found to induce significant DNA damage in the extended-term cultures from all four blood donors but there were variations in sensitivity between the different cultures, especially after the initial period (30 min) of DNA repair (Fig. 1). The level of H2O2-induced DNA damage was lower in the cells derived from blood donor 2 (29-fold increase in DNA damage immediately after the H2O2-exposure) and 3 (25-fold increase), than in the cells derived from blood donor 1 (40-fold increase) and 4 (42-fold increase). The lymphocytes from donor 2 and 3 were also found to have a more efficient DNA-repair than the lymphocytes from donor 1 and 4. Whereas the residual DNA damage (RD) remaining 30 min after the H2O2exposure in the more H2O2 resistant lymphocytes from donor 2 and 3 varied between 10 and 20 % (suggesting a DNA repair capacity of 80 to 90 % during the first 30 min following the exposure), the RD’s varied between 40 and 45 % in the more H2O2-sensitive lymphocytes from donor 1 and 4 (suggesting an initial DNA repair capacity of 55-60 %) (Table 2). The observed differences in H2O2 sensitivity between the lymphocytes may be due to different levels of detoxifying enzymes, such as catalase, glutathione peroxidase or glutathione reductase, which normally can take care of at least some of the reactive oxygen species formed by H2O2. However, different DNA repair processes are also important in the cellular defence against DNA damaging agents. In fact, there is some evidence suggesting that the DNA 21.

(200) Maria Andersson. repair capacity actually is more important than detoxifying enzymes when it comes to H2O2 induced DNA damage (Duthie and Collins 1997). DNA damage after H2O2 exposure ***. *** ***. 500. ***. ***. 400 300 ***. 200. PBS H2O2 H2O2+30 min. ***. ***. DNA damage (TMom). 600. 100 0 Donor 1. Donor 2. Donor 3. Donor 4. FIGURE 1. DNA migration in extended-term cultures of human lymphocytes (ETC) from 4 different blood donors. The cells were subjected to 10 min exposure of vehicle (PBS) or 0.25 mM hydrogen peroxide, with (H 2O2+30 min) or without (H2O2) an incubation period of 30 min at 37 °C after the exposure. The DNA migration is presented as average median tail moments ± S.E.M. Differences between control cells and H2O2-exposed cells were evaluated using a two-tailed t-test for independent samples. *** P< 0.001. TABLE 2. Residual DNA damage in H2O2-exposed extended-term cultures of human lymphocytes (ETC) after 30 min of DNA repair. Cells. Blood donor. Residual damage (%). ETC. 1. 40. 2. 20. 3. 10. 4. 45. The cells were exposed to 0.25 mM H 2O2 for 10 min and then incubated for 30 min at 37 o C without H2O2 to allow for initial DNA repair. The following formulae was used when calculating the residual damage:  [(TMomH 2 O 2 + 30 min ) − (TMomPBS + 30 min )]   × 100 [(TMomH 2 O 2 )− (TMomPBS )]  . 22.

(201) DNA damage in human lymphocytes. Even if DNA repair by rejoining of DNA strand breaks is considered to be a very rapid process in most cells (Frankenberg-Schwager 1989, Collins 2004), it has been shown that the repair of H2O2-induced DNA damage can be rather slow, at least in freshly isolated lymphocytes (Torbergsen and Collins 2000). There was a correlation between the sensitivity to the DNA damaging effect of H2O2 and the DNA repair capacity (the lymphocytes from the two donors with the lowest levels of H2O2-induced DNA damage were also found to have the highest DNA-repair capacity). However, the residual level of DNA damage was only monitored 30 min after the H2O2 exposure, thus the reported differences in residual DNA damage in the lymphocytes from the different donors may actually reflect differences in DNA repair kinetics rather than actual differences in DNA repair capacity. It should also be emphasized that the way of monitoring the DNA repair in Paper III does not give any information about the fidelity of the repair. It is known that there is a close relationship between DNA repair and DNA replication and that many proteins, including various checkpoint kinases and polymerases, are involved in both processes (Shechter et al., 2004, Petermann et al., 2005, Andreassen et al., 2006). It is also well known that both DNA repair and DNA replication are associated with an increased number of DNA strand breaks. It has previously been reported that incubation of control lymphocytes at 37 oC for 1 h in PBS agarose can result in comet formation (McKelvey-Martin et al., 1993) and the same phenomenon was also observed in the present study where the background level of DNA strand breaks were found to be increased in some of the control cells after the incubation for 30 min at 37 oC. Another possible explanation for the rather dramatic increase in the background level of strand breaks could be that the repair of oxidative DNA damage was initiated during the 30 min incubation, which consequently generates more strand breaks and thereby higher background levels of DNA damage.. Comparison between extended-term cultures of human lymphocytes and mouse lymphoma L5178Y cells In order to evaluate the sensitivity of ETC towards different DNA damaging agents, MLC were used for comparison. The latter cells are known to be susceptible to several classes of DNA damaging agents and they are also frequently used in various types of in vitro assays when screening for the potential genotoxicity of chemicals. In order to evaluate the ability of catechol to induce DNA damage in ETC and MLC, we used the repair enzymes Fpg and Endo III and the comet assay (Paper II). In Paper II we used the mean of logarithmic transformed tail moments (ln TMom) to analyze our comet data. It has been suggested that the mean of the logarithmic transformed data is the most sensitive measure when using the tail moment as the indicator of DNA damage (Wiklund and Agurell 2003). However, in Paper I, Paper III and Paper IV, we used the average median tail moments. The DNA migration can be monitored using many different types of measures such as the tail length, the. 23.

(202) Maria Andersson. tail moment and/or the percentage of DNA in the tail (Hellman et al., 1995). The major disadvantage with the tail moment is that it has no commonly accepted units and different image analysis programs can therefore report quite different values for the same level of DNA damage. This makes it almost impossible to directly compare tail moments between different laboratories (if they use different software’s for the comet analysis). It is also difficult for a reader who is unfamiliar with the image analysis program used in a specific study to know whether a given tail moment is “high” or “low”. The percentage of the DNA in the tail (%TDNA) is probably a better parameter for interlaboratory comparisons and the %TDNA may also give the reader a better feeling for the magnitude of damage. Nevertheless, in e.g. Paper II similar results were obtained when using %TDNA, TMom or ln TMom (Table 3). All concentrations of catechol tested induced DNA damage in the MLC when the assay was performed without adding lesion specific repair enzymes after the lysis. In the human lymphocytes, the lowest concentrations of catechol was found to decrease the migration of DNA, but at the highest concentration tested (3 mM), this compound was found to induce significant DNA damage also in these cells. The same pattern was seen when the repair enzyme Fpg was added after the lysis (Table 4). However, when Endo III was added, a different pattern of DNA damage emerged in the ETC. With this lesion specific repair enzyme, there was no decrease in DNA migration and only an increased DNA damage was observed in the extended-term cultures of human lymphocytes, the LOEL being 0.5 mM (Table 4). It is worth noting that reduced DNA migration only occurred in lymphocytes that had been exposed to the lowest concentrations of catechol. There was no such decrease in DNA migration in the MLC. Whether or not a lower concentration than 0.5 mM would have decreased the ln tail moment in the MLC is not known, since this was the lowest concentration tested in these cells. Hydrogen peroxide (H2O2), had a rather limited DNA damaging effect in Paper II, both in the ETC and the MLC. However, we believe that the limited effect of 0.25 mM H2O2 was a result of the exposure condition used for H2O2. The cells were suspended in medium in the presence of serum and exposed for 10 minutes. On the other hand, in Paper III and Paper IV we exposed the cells for 0.25 mM H2O2 for 10 min but they were instead directly exposed embedded in agarose on slides. The exposure condition used for H2O2 in Paper III, has also been used by others (Singh et al., 1988), and it is rather tough for the cells. Paper III clearly showed that the MLC were noticeably more resistant to the DNA damaging effect of H2O2 than any of the human lymphocyte cultures (5 – 8 times). The MLC were also found to have the least efficient DNA repair after 30 min (the RD being 60 %, suggesting a DNA repair capacity of only 40 %). However, it should be emphasized that it is difficult to directly compare the DNA repair capacity in these two different cell types since a direct comparison ideally should be done on cells having the same background level of damage. The fact that the ETC are human cells with a normal karyotype, and that MLC are transformed and immortalized cells with an abnormal karyotype could explain 24.

(203) 0.25. 0 0.5 3.0. 0.25. H2O2 (10 min). Catechol (3h). H2O2 (10 min). (22) (16) (22) (14) (12). (n). (22) (16) (22) (14) (12). (n). 84 ± 10***. 2.98 ± 0.38*** (10). (10). 29 ± 4 (9) 77 ± 6*** (10) 147 ± 18*** (10). 141 ± 12*** (12). 28 ± 4 23 ± 3 49 ± 13 56 ± 24 62 ± 6. TMom. Without enzyme. 1.11 ± 0.16 (9) 2.74 ± 0.21*** (10) 4.61 ± 0.49*** (10). 4.84 ± 0.37*** (12). 1.22 ± 0.15 1.00 ± 0.11 1.30 ± 0.28 1.34 ± 0.41 2.47 ± 0.24***. % TDNA. Without enzyme. 4.07 ± 0.07***. 2.56 ± 0.14 3.50 ± 0.11*** 4.63 ± 0.07***. 4.76 ± 0.03***. 2.82 ± 0.06 2.59 ± 0.07* 2.78 ± 0.07 2.47 ± 0.10** 3.91 ± 0.05***. ln TMom. (548). (447) (498) (491). (589). (1142) (799) (1096) (698) (597). (n). Without enzyme. The cells were subjected to alkaline single cell gel electrophoresis after 3 h of exposure to different concentrations of catechol. Hydrogen peroxide (H2O2) was used as positive control. The DNA migration is presented as average median tail DNA ± S.E.M. (%TDNA), average median tail moment ± S.E.M. (TMom) and mean values of the ln transformed tail moment ± S.E.M. (ln TMom). The data is pooled from 2 ï 9 individual experiments, where 3 ï 4 slides/treatment and 50 cells/slide were analyzed in each experiment. Differences between control cells and exposed cells were evaluated using a two-tailed t-test for independent samples. * P < 0.05 ** P < 0.01 *** P < 0.001. MLC. 0 0.1 0.5 1.0 3.0. Catechol (3h). ETC. (mM). Concentration. Treatment. Cells. TABLE 3. DNA migration in extended-term cultures of human lymphocytes (ETC) and in mouse lymphoma cells (MLC).. DNA damage in human lymphocytes. 25.

(204) 26 0.25. 0 0.5 3.0. 0.25. H2O2 (10 min). Catechol (3h). H2O2 (10 min). (697) (747) (741). (735). 5.19 ± 0.02***. (293). 5.05 ± 0.05***. 3.54 ± 0.06 4.08 ± 0.06*** 4.56 ± 0.05***. (594) (398) (598) (347) (294). (n). 4.20 ± 0.05 3.94 ± 0.06*** 3.87 ± 0.06*** 3.79 ± 0.11*** 4.57 ± 0.04***. ln TMom. Fpg. 4.70 ± 0.03***. 3.90 ± 0.05 4.21 ± 0.04*** 4.75 ± 0.04***. 5.07 ± 0.05***. 3.87 ± 0.05 3.99 ± 0.05 4.12 ± 0.10* 4.10 ± 0.10** 4.70 ± 0.04***. ln TMom. EndoIII. (690). (692) (743) (792). (296). (599) (519) (399) (398) (297). (n). The cells were exposed for 3 h to different concentrations of catechol before they were digested with enzymes in the comet assay. Cells treated with formamidopyrimidine glycosylase (Fpg) were digested for 30 min and those treated with endonuclease III (Endo III) were digested for 45 min. Hydrogen peroxide (H 2O2) was used as positive control. After adding a small constant (0.001) to the calculated tail moment (TMom), these were ln transformed and presented as mean values ± S.E.M. Differences between control cells and exposed cells were evaluated using a two-tailed t-test for independent samples. * P < 0.05 ** P < 0.01 *** P < 0.001. MLC. 0 0.1 0.5 1.0 3.0. Catechol (3h). ETC. (mM). Concentration. Treatment. Cells. TABLE 4. DNA migration in extended-term cultures of human lymphocytes (ETC) and mouse lymphoma cells (MLC) after enzyme treatment in the comet assay.. Maria Andersson.

(205) DNA damage in human lymphocytes. the observed difference in sensitivity to the H2O2-induced DNA damage between the two different cell types, but exactly why the human lymphocytes were more sensitive remains an unresolved issue. The relatively low extent of DNA repair in the MLC could possibly be explained by the fact that they had a heterozygous mutation in exon 5 in the p53 gene (Paper III). The p53 gene is an important gene mediating various cellular responses to an induced DNA damage. When DNA damage occurs, the p53 gene activates several other genes that are involved, not only in nucleotide and base excision DNA repair processes (Zhou et al., 2001, Smith and Seo 2002, Achanta and Huang 2004), but also in cell cycle arrest and apoptosis (Geske et al., 2000).. Comparison between extended-term cultures of human lymphocytes and freshly isolated peripheral blood lymphocytes To investigate whether ETC and freshly isolated peripheral blood lymphocytes (PBL) (established from the same blood samples) differed to catechol-and H2O2-induced DNA damage, we used three different blood donors (Paper IV). In the experiments with catechol, the cells were exposed to two different concentrations (0.5 or 3 mM) for 3 hours in RPMI. Under these exposure conditions, the lower concentration was found to be the no observed effect level (NOEL) for the DNA-damaging effect of catechol both in the ETC and the PBL. In contrast, 3 mM catechol was found to induce significant DNA damage in both cell types, and at this concentration the freshly isolated lymphocytes were clearly more sensitive to the DNA damaging effect of catechol than the ETC. H2O2 (0.25 mM) was used as a positive control, but in this case the cells were exposed for 10 min embedded in agarose on the microscope slides using a similar protocol as in Paper III. H2O2 was found to induce significant DNA damage in both types of cells, but in this case there was no clear and consistent difference in sensitivity between the PBL and the ETC. The data from Paper II and Paper IV, clearly shows that an exposure to 3 mM catechol for 3 h induces substantial DNA damage in the extended-term cultures of human lymphocytes. However, one of the more interesting findings in Paper II was that an exposure to lower concentrations of catechol (0.1 – 1 mM) actually reduced the background level of damage in ETC (but not in MLC), especially when the repair-specific enzyme formamido pyrimidine glycosylase was included in the assay (Table 3 and 4). This reduction might have been due to an antioxidative effect of low concentrations of catechol, but it could not be repeated in Paper IV. The reason for this is not known, but one explanation could be that the background level of DNA damage in the extended-term cultures in Paper IV was “as low as it could be” already from the start, making it almost impossible to detect a further reduction in the exposed cells. Another explanation could be that the study in Paper IV was performed without repair specific enzymes. The slightly different levels of DNA damage before and after the exposure to 27.

(206) Maria Andersson. catechol in the ETC in Paper II and Paper IV may also be due to the fact that different blood donors were used in the two studies. We have shown that there are interindividual differences both in induced DNA damage and its subsequent repair between extended-term cultures from different blood donors (Paper III), and also for freshly isolated lymphocytes it is well known that inter- and intraindividual variations between the blood donors exist (Pero et al., 1978, Holz et al., 1995, Collins et al., 2001, Janssen et al., 2001). It has also been shown that the basal level of DNA damage in human lymphocytes can be subjected to seasonal variations (Giovannelli et al., 2006). Using the alkaline version of the comet assay, it was quite clear that the PBL from all three donors were much more sensitive towards the catechol-induced DNA damage than the corresponding ETC, established from the same blood samples. The reason for this is not known, but their different sensitivity is most likely due to their different proliferative status. Even if their DNA can be both broken and rejoined, PBL are non-proliferating cells arrested in G0-phase (Carson et al., 1988). In contrast, the ETC are proliferating cells, and actively dividing cells are known to have a higher DNA repair activity than non-dividing cells (Bohr et al., 1989, Bartosova et al., 1996). The higher level of catechol-induced DNA damage in the PBL may therefore be the result of a less efficient rejoining of strand breaks in these cells during the 3 h of exposure to catechol. We did not have enough PBL to monitor ROS in these cells (all PBL’s isolated from a blood sample were either used directly for the comet assay or to establish an ETC), but we could measure ROS in the extended-term cultures at various time points during a 3-hour exposure to catechol. As expected, catechol was found to increase the level of ROS, but there was no obvious relationship between the production of ROS and the catechol-induced DNA damage (Paper IV). At 0.5 mM catechol, the level of ROS was significantly increased but not the level of DNA damage, and the ETC from the blood donor producing the highest level of ROS did not have the highest level of DNA damage. It may very well be so that the level of ROS produced by 0.5 mM catechol could be handled by various radical scavengers like catalase, glutathione peroxidase or superoxid dismutase, and that a threshold concentration of ROS was passed between 0.5 and 3 mM catechol. However, it may also be so that the catechol-induced DNA damage observed at 3 mM followed another, nonoxidative, pathway possibly involving an adduct. There was an interindividual variation in sensitivity towards the catechol- and H2O2-induced DNA damage between the lymphocytes from the different blood donors (especially for the PBL’s), and this is in accordance with our findings in experiments with ETC’s in Paper III. However, it should be emphasized that the number of slides from the experiments with the freshly isolated lymphocytes examined was rather restricted (and so was the number of blood donors included in the study). The experiments with PBL were based on blood samples from three to us anonymous blood donors, and they could therefore only be performed one time for each blood donor (i.e. on the same day as the PBL’s were isolated). The maximum number of slides per treatment was therefore limited to 5 slides in the 28.

(207) DNA damage in human lymphocytes. PBL experiments. Only 2 slides per treatment were analysed in the experiments with the PBL from blood donor I, due to technical reasons. All experiments with the extended-term cultures (which were established from the PBL on the same days as the blood samples were processed) were performed in duplicate, and this explains why the ETC experiments with the comet assay were based on 10 slides per treatment.. Clouds – an unresolved issue Dead or dying cells can undergo rapid DNA fragmentation which should be expected to increase DNA migration in the comet assay (Tice et al., 2000). The presence of extremely high levels of DNA migration in the comet assay may be associated with cytotoxicity in the form of necrosis or apoptosis. Due to an excessive cytotoxicity there is a potential problem of false positive calls in the comet assay (Hartmann et al., 2001). It has been suggested that apoptotic nuclear fragmentation produces a small comet head and a large fan-like tail, also known as a ‘cloud’ (also referred to as ‘hedgehogs‘, Fig. 2) (Olive et al., 1993, Fairbairn et al., 1995). Cells undergoing necrosis exhibit larger comet heads and thin long tails, which are indistinguishable from comets formed from DNA damage induced by a genotoxin (Tice et al., 2000).. FIGURE 2. Comet image of extended-term lymphocytes from human (ETC). The picture illustrates a ‘cloud’ (the upper cell) next to two almost undamaged cells (the lower cells). The image was kindly provided by Oskar Karlsson.. We have always excluded comets without clearly identifiable heads (i.e. ‘clouds’) from our image analysis (Hellman et al., 1995). The approach of excluding comets with practically all DNA in the tail after the electrophoresis in the comet assay has also been used by others (Hartmann and Speit 1997). These authors showed that cytotoxicity was associated with an increased number of cells with extremely fragmented DNA in V79 Chinese hamster cells and human white blood cells. Moreover, non-mutagenic and non-carcinogenic agents known to induce DNA double strand breaks, secondary to induced cytotoxicity, revealed negative results in. 29.

(208) 30 0.25. 0 0.5 3.0. 0.25. H2O2 (10 min). Catechol (3 h). H2O2 (10 min) 5.6 ± 0.7. 5.3 ± 0.8 6.6 ± 1.1 10.0 ± 2.6. 16.0 ± 2.2. 10.3 ± 1.9 4.5 ± 1.1* 11.3 ± 1.9 6.0 ± 1.1 20.9 ± 1.2***. (11). (9) (10) (10). (25). (23) (16) (24) (15) (12). 4.1 ± 0.8. 3.6 ± 0.9 4.7 ± 0.6 5.6 ± 1.1. 3.5 ± 0.6. 4.2 ± 1.1 6.5 ± 1.7 4.9 ± 1.0 4.3 ± 1.8 5.3 ± 1.4. (15). (14) (15) (15). (13). (12) (8) (12) (8) (6). 8.9 ± 1.0. 6.4 ± 2.7 7.3 ± 1.2 7.9 ± 1.9. 6.7 ± 1.0. 4.7 ± 0.7 7.5 ± 1.3* 5.4 ± 1.0 7.1 ± 1.2 3.7 ± 0.8. Clouds (%, mean ± S.E.M.). EndoIII. (14). (14) (15) (16). (12). (12) (8) (12) (8) (6). (n). The cells were subjected to alkaline single cell gel electrophoresis for 10 min after exposure to different concentrations of catechol or 0.25 mM H2O2 (positive control). The number of ‘clouds’ on each slide is presented as average percentages ± S.E.M. after pooling data from 2 ï 9 individual experiments (n = 6 ï 25 slides/treatment). Differences between exposed cells and vehicle controls were evaluated using a two-tailed t-test for independent samples. * P < 0.05 *** P < 0.001. MLC. 0 0.1 0.5 1.0 3.0. Catechol (3 h). ETC. Clouds (n) (%, mean ± S.E.M.). Clouds (%, mean ± S.E.M.). (mM). (n). Fpg. Without enzyme. Treatment. Cells. Concentration. TABLE 5. Percentage of ‘clouds’ in extended-term cultures of human lymphocytes (ETC) and mouse lymphoma cells (MLC).. Maria Andersson.

(209) DNA damage in human lymphocytes. the comet assay when the ‘clouds’ were excluded from the analysis. The percentage of ‘clouds’ in ETC and MLC after exposure to catechol and H2O2 are shown in Table 5 (unpublished data from Paper II). Even if the number of ‘clouds’ increased in the ETC with the lowest concentration of catechol and Endo III and with the highest concentration of catechol without enzyme, the overall impression is that catechol did not affect this type of comets much. In addition, the 10 min exposure to the positive control H2O2 did not affect the number of ‘clouds’ neither in the ETC nor in the MLC (Table 5). There has been an extensive debate within the comet assay community whether ‘clouds’ are dying cells, representing cytotoxicity, or not. When using visual scoring in the analysis of images from a comet assay, the assigned value of 4 is very similar to a ‘cloud’ and thereby it is recommended that these cells are included in the analysis (Collins 2004). It has also been shown that there is a very close agreement between visual and computer scoring (Collins et al., 1997). The approach of not excluding comets with practically all DNA in the tail is also recommended by others (Hellman et al., 1995, Kiffe et al., 2003, Rundell et al., 2003). Using CHO cells, Rundell et al. (2003) showed that it is possible to discriminate apoptotic cells from DNA-damaged cells on the basis of their ability to repair their genomic DNA. For this purpose liquid-holding recovery was used, in which DNA lesions are repaired while cells are prevented from dividing (Roberts and Aldous 1949). Cells that were extremely damaged by 0.25 mM H2O2 (with the appearance of many ‘cloud’ figures) showed complete recovery after 4 h of liquid holding. Taken together, my personal opinion, based on my present experience, is now shifting towards that ‘clouds’ should be included in the analysis of comets, since they may represent a genotoxic effect. The phenomenon of excluding the ‘clouds’ (due to the assumption that they represent cytotoxicity) has lead to a neglect of genotoxic effects from compounds with steep or unusual dose response curves (Marie Vasquez, contribution to the Comet Assay Forum, comet-assay@list.nih.gov). To examine whether the ‘clouds’ originate from a cytotoxic effect, a cytotoxicity assay should be performed on the same population of cells. Ideally, this should be performed by having twice the number of slides in a comet assay, where half of them undergo the normal computerized image analysis (e.g. stained with ethidium bromide) while the other half could be stained with, e.g. propidium iodide, to examine the cytotoxicity.. 31.

(210) Maria Andersson. CONCLUSIONS AND FINAL REMARKS. This thesis demonstrates the value of using extended-term cultures of human lymphocytes in the comet assay when evaluating the potential DNA damaging effects of chemicals. Out of the seven reference compounds used in Paper I, the bulky DNA adduct forming compounds (4-nitroquinoline-N-oxide and benzo[a]pyrene) were the most potent agents in the ETC. The second most potent substances tested were the alkylating agents (cyclophosphamide and N-nitrosopiperidine). Only one of the non-DNA-reactive agents (5-hydroxyurea and ethanol) slightly increased the DNA migration in the extended-term cultures of human lymphocytes. The cross linking agent formaldehyde was also tested, and as expected it was the only compound that significantly reduced the DNA migration. To summarize, the ETC were found to detect the DNA-damaging effects of the seven reference compounds with similar results to what has been reported in other cell types. A comparison between the ETC and the MLC indicated that the latter cells are somewhat more sensitive to catechol-induced DNA damage than the extendedterm cultures (Paper II). However, data obtained from H2O2-induced DNA damage showed that ETC were clearly more sensitive to the DNA damaging effect of H2O2 than MLC (Paper III). The two types of in vitro systems may indeed show different sensitivity to the DNA damaging effects of different types of genotoxic agents, but the overall conclusion is that they give the same qualitative answer when evaluating primary DNA damage with the comet assay. Using different blood donors for the establishment of ETC confirmed that there is an interindividual variation in induced DNA damage and the subsequent repair (Paper III). These variations may be of importance from a safety evaluation and risk assessment perspective, but they do not seem to be of crucial importance when using the ETC to identify DNA damaging agents. The evaluation of PBL and the corresponding ETC, established from the same blood samples, clearly indicated that the PBL were more sensitive to catechol-induced DNA damage (Paper IV). This was most likely due to their different proliferative status but despite this difference, both in vitro systems were able to identify catechol as a DNA damaging agent at the same concentration. In conclusion, the ETC and the comet assay are a useful combination when testing for the potential DNA damaging effects of chemicals. Representing easily cultivated cells possessing the normal human karyotype, where one blood sample can be used for numerous experiments performed over a long time, extended-term cultures appear to be a practical alternative, both to transformed mammalian cell. 32.

(211) DNA damage in human lymphocytes. lines, and primary cell cultures from humans. In fact, I suggest that the extendedterm lymphocytes, with or without S9 and/or lesion specific DNA repair enzymes, should be used more frequently when screening for the potential genotoxicity of chemicals.. 33.

No results found