Induction of DNA-DNA Cross-Link Formation in Human Cells by Various Psoralen Derivatives1

[CANCER RESEARCH 45, 5394-5398, November 1985]

Dieter C. Gruenert,2 Michael Ashwood-Smith, Reginald H. Mitchell, and James E. Cleaver

Laboratory of Radiobiology and Environmental Health, University of California, San Francisco, California 94143 Â¡D.C. G., J. E. C.Â¡,and Department of Biology University of Victoria, British Columbia V8W2Y2, Canada [M.A-S..R.H.M.]

ABSTRACT from patients with two different hypersensitive diseases. We also measured the extent of interstrand DNA cross-linking from psor Furocoumarin-induced DNA damage, monoadducts, and alen derivatives using a method that was modified from that of cross-links were measured in normal human, xeroderma pigmentosum, and Panconi's anemia cells after exposure to near-UV Ball and Roberts (15). (356 nm). At similar concentrations and near-UV doses, photoad MATERIALS AND METHODS dition by 8-methoxypsoralen was twice that by angelicin and the Cells and Culture Conditions. SV40-transformed normal (GM637), substitution of bromodeoxyuridine for thymidine in one strand of excision-deficient xeroderma pigmentosum group A (XP12RO), and DNA did not alter the binding. The rate of cross-linking by 8- cross-link-sensitive Fanconi's anemia (FAH12) human fibroblasts (16) methoxypsoralen was twice that of 5-methoxypsoralen. Low were used in this study. The FAH12 cells were the gift of Dr. K. Sperling frequencies of cross-links were detected from angelicin and 3- (Institute fÃ¼rHuman Genetik der Freien UniversitÃ¤t, Berlin, Germany). carbethoxypsoralen but none were detected from 5-geranoxy- Cells were maintained in Eagle's minimal essential medium supplemented psoralen at concentrations up to 25 /Â¿g/mland near-UV doses with 15% fetal calf serum and antibiotics. up to 45,000 J/m2. Psoralen Derivatives. Angelicin was chemically synthesized. 8-MOP was purchased from Sigma Chemical Co. and further purified by recrys- tallization. 5-GYP and 5-MOP were extracted from lime oil and oil of INTRODUCTION bergamot, respectively. 3-CP was the generous gift of Dr. D. Averbeck Furocoumarins are photosensitive compounds that are used (Fondation Curieâ€”Institut du Radium, Section de Biologie, Paris, therapeutically (for review, see Ref. 1) although they have re France). All compounds were characterized by nuclear magnetic reso cently been associated with high levels of cancer among treated nance and mass spectroscopy. individuals (2-5). The biological activity of these compounds is The purity of psoralen derivatives was >99% as determined by high- related to the formation of mono- or diadducts in DNA following performance liquid chromatography. Stock solutions of 8-MOP, angelicin, 5-MOP, 5-GYP, and 3-CP were stored at -20Â°C at concentrations of absorption of one of two photons of near-UV (320-400 nm), 0.25 to 1 mg/ml in 100% ethanol. Stock solutions were diluted to the respectively (1,6). appropriate concentrations with phosphate-buffered saline. Untreated A comprehensive understanding of psoralen-induced carcino- and psoralen-treated cell cultures in 100-mm Petri dishes were exposed genesis requires both qualitative and quantitative analysis of the to near-UV filtered mercury arc lamp at a fluence of 25 J/m2/s (peak, 356 DNA lesions produced by the various psoralen derivatives. Pre nm; Magna Flux Corp., Chicago, IL). Dosimetry was performed with a vious studies have sought to correlate both the mutagenic and Yellow Springs Radiometer (Yellow Springs Instrument Co., Yellow carcinogenic effects with the degree of cross-linking (2, 7, 8). Springs, OH). However, because the lesions formed by DNA cross-linking Radiolabeled 8-Methoxypsoralen and Angelicin. 8-MOP and ange licin were labeled with 3H at the Amersham Corporation (Chicago, IL) by agents are a mixed population of monoadducts and interstrand cross-links, it is difficult to correlate a biological effect with a a catalytic exchange process. The crude product was purified by silica specific DNA lesion. gel column chromatography using CHCI3 as solvent (17). Radiochemical purity was determined by analytical thin-layer chromatography (silica gel) The psoralen derivative most commonly used as a monofunc- and absorption spectrum analysis using unlabeled 8-MOP or angelicin tional control has been the angular angelicin (9-12). Recent as a standard. Samples containing purified material were dried and studies, however, indicate that angelicin also forms a low fre redissolved in 100% ethanol. Concentration was determined on a Gary quency of DNA interstrand cross-links in PM2 and X-DNA (13, 219 spectrophotometer and specific activity was determined on a Pack 14). ard 3303 liquid scintillation spectrophotometer. The specific activity of Previous studies have measured psoralen photoaddition in [8-3H]MOP was 2.8 Ci/mmol and that of [3H]angelicin was 2.4 or 22 Ci/ native DNA, bacterial cells, and in eukaryotic cells such as yeast, mmol. hamster cells, mouse cells, and human cells (1). However, the The total amount of 8-MOP or angelicin bound to DNA was measured lack of information on the extent of cross-linking in human cells in confluent cell cultures prelabeled with [14C]dThd, 0.001 MCi/ml (57 mCi/mmol). The cells were rinsed and exposed to near-UV in the pres requires attention, especially in view of the use of psoralens in ence of various concentrations of [8-3H]MOP or [3H]angelicin in phos current dermatologica! practice. In this study we have compared the binding of 8-MOP ' and angelicin in normal cells and in cells phate-buffered saline on ice. The cells were harvested immediately after irradiation and the DNA was extracted and purified (18) to ensure removal of photoadducts in RNA and protein (1). Purified DNA was precipitated Received 5/17/84; revised 5/10/85; accepted 7/22/85. with ice-cold 100% ethanol and stored at -20Â°C for more than 1 h. The 1This work was supported by the United States Department of Energy (Contract DNA was centrifuged at 10,000 rpm for 30 min at 4Â°C and then No. DE-ACO3-76-SF01012) and National Cancer Institute Radiation Biophysics Training Grant 5A32 CA09272-04 (D. C. G.). resuspended in SSC. To ensure removal of any psoralen that was not 2 Present address: Hana Biologies, Inc., Therapeutics Division, 626 Bancroft covalently bound, the precipitation and resuspension procedure was Way, Berkeley, CA 94710. To whom requests for reprints should be addressed. 3 The abbreviations used are: 8-MOP, 8-methoxypsoraten; BrdUrd, bromo- phosphate-buffered saline; dThd, thymidine; SSC, standard saline citrate (0.15 M deoxyuridine; 3-CP, 3-carbethoxypsoralen; 5-GYP, 5-geranoxypsoralen; PBS, sodium chloride-0.015 M sodium citrate, pH 7.4).

CANCER RESEARCH VOL. 45 NOVEMBER 1985

5394

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1985 American Association for Cancer Research. BINDING OF PSORALEN DERIVATIVES TO CELLULAR DNA repeated until the specific activity of the DNA remained constant (usually tivity profiles. Equation B can then be written as two resuspensions). An aliquot of the DNA solution was precipitated on Whatman GF/A glass fiber filters and the radioactivity was determined (C) by liquid scintillation counting. The concentration of the DNA solution was determined by measuring the absorption at 260 nm. Psoralen binding to BrdUrd-substituted DNA was determined in the same fashion. The cross-link frequency was then expressed as m/Mâ€ž,andMâ€žisthe Some of the [3H]psoralen-labeled DNA was also centrifuged on alkaline number average molecular weight for double-stranded DNA which was CsCI gradients for further purification. determined on neutral or alkaline sucrose gradients. Measurement of Mâ€ž Isopyknic Gradient Analysis. Purified BrdUrd-substituted DNA was on neutral sucrose gradients avoided artifacts caused by anomalous banded in neutral or alkaline CsCi-Cs;.SO4 gradients (15,19). The DNA sedimentation of denatured DNA joined by cross-links (19). in SSC and 2 ITIMEDTA was sheared either by passing the solution 5 N and A/0were determined from the radioactivity profiles by measuring times through a 20-gauge needle or by agitation. The gradients were the radioactivity in equivalent fractions on the heavy side of the H peak. centrifuged in a Beckman 50 Ti rotor at 42,000 rpm and 20Â°Cfor up to This empirical approach facilitated the measurement of DNA remaining single-stranded and minimized the contribution of radioactivity from 48 h. After centrifugaron, the gradients were fractionated into 20 to 23 cross-linked DNA in the H peak. Corrections for small shifts in the location fractions/gradient while A?6nwas continuously monitored. The radioactiv of the peaks were made by measuring the distances between the peaks ity profile was obtained by precipitating the DNA in a 50-ul aliquot from of banded DNA. Gradients showing unusual banding were not consid each fraction on Whatman GF/A glass fiber filters or on Whatman 3-mm ered. paper filters and measuring the radioactivity by liquid scintillation count Three assumptions were required to determine the values of N and ing. No. (a) radioactivity in the heavy side of the H peak was representative Molecular Weight Measurements. A100- or 200-t

CANCER RESEARCH VOL 45 NOVEMBER 1985 5395

x C E Ã¬.

o 05 .1

< u ftÂ». 5 10 15 FRACTION NUMBER

n Ã–

> J/m2 x 10-3

> Chart 3. Fraction of heavy BrdUrd-substituted DNA remaining single stranded in alkaline isopyknic gradients after treatment of XP12RO cells with increasing concentrations of near-UV and 8-MOP or angelicin (ANG). 25 ng/m\. No, number of single-stranded molecules in uncross-linked (untreated) DNA; W. number of single- O stranded molecules remaining in treated DNA after cross-linking. O times4 and BrdUrd inhibits progression through the cell cycle U (24), lack of another round of DNA replication during a 40-h incubation is not unusual. The radioactivity profiles in alkaline isopyknic gradients of 4 8 12 16 20 control or 8-MOP-treated DNA, which was labeled with [14C]- FRACTION NUMBER dThd in one strand and [14C]BrdUrd in the opposite strand shows Chart 2. Alkaline CsjCI-CsSO4 isopyknic gradient profiles of DNA from XP12RO cells labeled with ("CJdThd and ("C]BrdUrd. A, cells treated with near-UV alone evidence of cross-linked molecules between H and L strand (A) or exposed to 7500 J/m2 plus 8-MOP, 25 jig/ml (A). Cross-linked DNA is positions. At higher concentrations or higher UV doses, the H reflected in the HL peak. B, , radioactivity profile of DNA from cells treated peak can be obscured by the banding of cross-linked DNA (Chart with 8-MOP, 10 /ig/ml, and 7500 J/m2 of near-UV; , normalized profile for 2). However, our analysis (see "Materials and Methods") elimi the H peak of DNA from control cells and represents the distribution of single- stranded BrdUrd molecules in the treated DNA. Cells were incubated as described nates any ambiguity. The increase in cross-link frequency with in "Materials and Methods." DNA content was normalized to the amount of DNA in cultures exposed to near-UV in the absence of 8-MOP. dose was quantified from changes in the proportion of labeled DNA remaining single stranded and therefore lacking cross-links (see "Materials and Methods") (Chart 3). dose of 7500 J/m2 was about 12 x 10~8 adducts/dalton (Chart IB), whereas the comparable value for angelicin was 7 x 10~8. Calculation of absolute cross-link frequency was dependent on molecular weights determined in neutral or alkaline sucrose The binding of these psoralens to DNA was essentially the same in xeroderma pigmentosum and Fanconi's anemia cells as in gradients. Preferential breakage in DNA due to BrdUrd photolysis by near-UV or long exposure to alkaline conditions was a concern normal cells. The ratio of angelicin and 8-MOP binding to DNA that was analyzed. The peak efficiency of BrdUrd photolysis is in which BrdUrd was substituted for dThd in one strand was closer to 310 nm than the 356 nm used in our experiments (25). 0.93 and 0.91, respectively. This is consistent with Ben-Hur and In addition, cells were irradiated through Petri dish covers and Riklis (20), although slightly greater binding to BrdUrd-substituted this filtered shorter wavelengths. Analysis on alkaline sucrose DNA was observed by Cassel and Latt (21). The radioactivity profiles of [8-3H]MOP- or [3H]angelicin- gradients of doubly labeled DNA which was unifilarly substituted with BrdUrd indicated no preferential breakage in the BrdUrd- treated DNA in alkaline isopyknic gradients indicated that all 3H substituted strand at near-UV fluences up to 60,000 J/m2 at 356 radioactivity was associated with the [14C]dThd-prelabeled DNA nm nor did exposure of purified DNA to alkaline conditions for rather than dispersed throughout the gradient (22, 23). This is up to 48 h at 20Â°C. further evidence that noncovalently bound psoralen has been The molecular weight of double-stranded DNA ranged from removed and that the psoralen associated radioactivity is cova- Mâ€ž=0.8 to 8.4 x 107 daltons with a mean of 3.54 x 107 Â±1.68 lently bound to the DNA. (SE) daltons. This range is not subject to significant centrifugation DNA Cross-Linking from Psoralens. Verification of labeling speed artifacts, which generally become important above 108 time indicated that cells incubated in the presence of BrdUrd did daltons (26). The variation in molecular weight occurred as a not enter another round of replication up to 40 h incubation. Since SV40-transformed cells appear to have longer cell cycle * L. N. Kapp, personal communication.

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5396

that angelicin had 0.15 times the efficiency of cross-linking DNA as did 8-MOP.

o DISCUSSION This study has examined the binding of various psoralen derivatives to cellular DNA in intact cells. We observed a maxi mum level of 8-MOP and angelicin binding to DNA of about 7 x 10~8 and 3.5 x 1(T8 adducts/dalton, respectively. Assuming approximately 109 thymines/genome, there is only about 1 adduct/104 thymines, which suggests that intact cells have a very low proportion of thymines which are potential binding sites. The level of angelicin binding is similar to previous measurements made in Ehrlich ascites tumor cells (27), although one study observed a 10-fold higher level (12). Saturation of covalent binding by psoralens may be due to limitations in binding sites imposed by chromatin structure (28, 29). Electron microscopic S 10 IS 20 studies of photobinding in Drosophila melanogaster (30) and CONCENTRATION (pg/ml) mouse chromatin (29) showed addition to every linker when isolated nuclei were treated with 4,5',8-trimethylpsoralen. Use of isolated nuclei and repeated applications of the psoralen may also have enhanced total binding above that possible in intact cells. The association constants for noncovalent binding to DNA by 8-MOP and angelicin are 10 x 103 and 13 x 103, respectively (31), which suggests that intercalation is not responsible for the 2-fold difference in photoaddition we observed. Rather an anal ysis of the absorption spectra of psoralens indicates that greater photobinding of 8-MOP is due to differences in absorbance between angelicin and 8-MOP at the emission maximum of the near-UV source (Xmax= 356 nm) (1, 32). The relative amount of cross-linking by 8-MOP, 5-MOP, and angelicin that we observed is consistent with the cytotoxicity (28, 33) and the photoaddition to purified DNA by each agent (34). 8-MOP has been shown to be more efficient in DNA interstrand cross-link formation than 5-MOP when studies were carried out using native DNA (35). The demonstration that DNA cross-linking in human cells by 5-MOP is one-half that of 8-MOP is consistent with this earlier study, although is has not been shown previously in human cells. 3-CP has a high photoaffinity for DNA (28) but forms very few cross-links. The low level of cytotoxicity detected in this previous study is consistent with 15 30 less toxicity from monoadducts than from interstrand cross-links. NUV DOSE (J/rt|2) Little is known about 5-GYP, although no cross-links were Chart 4. A cross-link frequency Â¡ncells exposed to various psoralens as a function of concentration. Cells were exposed to 7500 J/m2 near-UV plus increasing detected in human cells treated with this agent. concentrations of 8-MOP (A), angelicin (A), or 5-MOP (â€¢).B,cross-link frequency The studies showing that angelicin and 3-CP form only mon in cells exposed to various psoralens as a function of near-UV (NUV) dose. Cells oadducts used different methods of analysis or were based on were treated with 8-MOP, 5-MOP, angelicin (XING), 3-CP, or 5-GYP, 25 >ig/ml each, plus increasing doses of near-UV. Cross-link frequency was determined as de assumptions from work in nonmammalian systems (9,12,28). It scribed in 'Materials and Methods." Mâ€žofthe DNA was determined from analysis is possible for 3-CP to form cross-links by photochemical cleav on 5-20% neutral sucrose gradients or by averaging the molecular weights of DNA age of the carbethoxy group at wavelengths shorter than 340 from several gradients in one experiment when the DNA was subjected to uniform shear by passage through a 20-gauge needle. Bars, SE. nm (36). Because our near-UV source is not monochromatic, it is conceivable that there was enough near-UV of wavelength result of the DNA isolation procedure. The higher molecular less than 340 nm to cause such a conversion. The resulting weights were obtained by isolating the DNA with minimal me psoralen molecules would then be capable of cross-link forma chanical shear. tion. Cross-links/108 daltons were determined for various psoralen DNA cross-linking by angelicin has been observed at high drug derivatives as a function of drug concentration or near-UV dose concentrations and high doses (14) or in highly condensed DNA (Chart 4). 8-MOP and 5-MOP showed a high degree of cross- (13); however, ours is the first indication that angelicin also cross linking, whereas angelicin and 3-CP showed slight but reproduc links human DNA in intact cells. The highly compact structure of ible cross-linking and 5-GYP showed none. The graphs indicate mammalian chromatin may account for the cross-linking we

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5397

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1985 American Association for Cancer Research. BINDING OF PSORALEN DERIVATIVES TO CELLULAR DNA detected. The condensation and DNA supercoiling could cause Biochim. Biophys. Acta, 607: 215-220,1980. 14. Lown, J. W., and Sim, S.-K. Photoreactionof psoralenand other furocoumarins different sections of the double helix to be adjacent making the with nucteicacids. Bioorg. Chem., 7: 85-95,1978. formation of intermolecular cross-links possible. Intermolecular 15. Ball, C. R., and Roberts, J. J. Estimation of interstrand DNA cross-linking cross-links involving opposite strands would be indistinguishable resulting from mustard gas alkylation of HeLa cells. Chem.-Biol. Interact., 4. from interstrand cross-links when measured by our density shift 297-303,1971. 16. Zakrzewski, S., and Sperling, K. Genetic heterogeneity of Panconis anemia technique. demonstrated by somatic cell hybrids. Hum. Genet., 56: 81-84,1980. Although the possibility that outlying counts in the isopyknic 17. Isaacs, S. T., Shen, C. J., Hearst, J. E., and Rapoport, H. Synthesis and distribution may be due to small diffusion-driven molecules can characterization of new psoralen derivatives with superior photoreactivity with DNA and RNA. Biochemistry, 76: 1058-1064,1977. not be ignored, we feel that the linearity demonstrated at high 18. Cleaver, J. E. Methods for studying excision repair of DNA damaged by levels of cross-linking supports our empirical approach for inter physicaland chemicalmutagens.In: B. Kilbey(ed.), Handbookof Mutagenicity Test Procedures, pp. 19-48. Amsterdam: Elsevier/North-HollandBiomÃ©dical pretation of the data. In addition, we have made individual Press, 1977. molecular weight determinations of the DNA in >75% of the 19. Roberts, J. J., and Friedlos, F. Quantitative aspects of the formation and loss isopyknic gradients used in this study. Variations in the value of of DNA interstrand crosslinks in Chinese hamster cells following treatment m were offset by differences in Mâ€žsuchthat the values of m/Mn with c/s-diammniedichloroplatinumfll)(cisplatin) I: Proportion of DNA-platinum reactions involved in DNA crosslinking. Biochim. Biophys. Acta, 655: 146- were consistent for the same experimental conditions. This 151,1981. consistency is further demonstrated by the close agreement our 20. Ben-Hur, E., and Riklis, E. Photochemical interaction of furocoumarins with measurements have with specific findings in previous investiga bromodeoxyuridine and polydeoxynucleotidescontaining bromodeoxyuridine: its biological implications. Photochem. Photobiol., 27: 551-557,1978. tions and by the linearity of the curve at high levels of cross- 21. Cassel, D. M., and Latt, S. A. Relationship between DNA adduct formation linking. and sister chromatic)exchangeinduction by [3H]8-methoxypsoralenin Chinese The use of 8-MOP therapeutically and of 5-MOP as a com hamster ovary cells. Exp. Cell Res.. 128:15-22,1980. 22. Cleaver.J. E., and Gruenert, D. C. Repair of psoralen adducts in human DNA: ponent of certain suntan preparations is cause for concern. differences among xeroderma pigmentosum complementation groups. J. In Photoaddition and cross-linking will occur as long as the individ vest. Dermatol., 82: 311-315,1984. ual is exposed to a source of near-UV because of the presence 23. Gruenert, D. C. Psoralen damage in human cells: DNA repair and effects on replication. Ph.D. Dissertation. Universityof California, Berkeley, CA, 1982. of residual drug. It is therefore important that we define the repair 24. Cleaver, J. E. Thymidine metabolism and cell kinetics. Front. Biol., 6: 108, of psoralen monoadducts and cross-links and the extent of the 1967. interaction between cross-links and other lesions in human cells 25. Hutchinson, F. The testonsproduced by UV in DNA containing 5-bromouracil. Q. Rev. Biophys.. 6: 201-246, 1973. (23, 37, 38). We have elucidated some of these phenomena in 26. Hutchinson,F.The dependenceof DNA sedimentationon centrifugation speed. the accompanying study (39) to further our understanding of the In: P. C. Hanawalt and R. B. Settow (eds.), Molecular Mechanisms for DNA role DNA cross-links play in mutagenesis and carcinogenesis. Repair, Part B, pp. 703-707. New York: Plenum Press, 1975. 27. Bordin, F., Cariassare, F., Baccichetti, F., Guiotto, A., Rodighiero.P., Vedaldi, D., and Dall'Acqua, F. 4,5'-DÂ¡methylangelicin:anew DNA-photobinding mon- REFERENCES ofuncttonal agent. Photochem. Photobiol., 29:1063-1070,1979. 1. Ben-Hur, E., and Song, P.-S. The photochemistry and photobtology of furo- 28. Averbeck, D., Moustacchi, E., and Bisagni, E. Biological effects and repair of coumarins (psoratens).Adv. RadiÃ¢t.Bio).,11:131-177,1984. damage photoinduced by a derivative of psoralen substituted at the 3,4 2. Grekm. D. A., and Epstein, J. H. Psoratens,UVA (PUVA)and photocarcmoge- reaction site: photoreactivity of this compound and lethal effect in yeast. nesis. Photochem. Photobiol., 33. 957-960,1981. Biochim. Biophys. Acta, 578: 464-481,1978. 3. Stem, R. S., Thibodeau,L. A., Kiemerman.R. A., Parrish,J. A., and Fitzpatrick, 29. Cech, T., and Pardue, M. L. Cross-linking of DNA with trimethylpsoralen is a T. B. Risk of cutaneous carcinoma in patients treated with oral methoxsalen probe for chromatin structure. Cell, 71: 631-640,1977. photochemotherapy for psoriasis. N. Engl. J. Med., 300: 809-813,1979. 30. Wiesehahn, G. P., Hyde, J. E., and Hearst, J. E. The photoaddition of 4. Verdich, J. Squamous cell carcinoma:occurrence in mycosis fungoidestreated trimethylpsoralen to Drosophila melanogaster nuclei: a probe for chromatin with psoratens plus long-wave ultraviolet radiation. Arch. Dermatol., 115: substructure. Biochemistry, 76: 925-932, 977. 1338-1339,1979. 31. Dall'Acqua, F., Terbojevich, M., Marciani, Sâ€žVedaldi,D., and Recher, H. 5. Roberts, L. K., Schmitt, M., and Daynes, R. A. Tumor-susceptibilitygenerated Investigationon the dark interaction between furocoumarins and DNA. Chem.- in mice treated with subcarcinogenic doses of 8-methoxypsoralen and long Biol. Interact., 27: 103-115,1978. wave ultraviolet light. J. Invest. Dermatol., 72: 306-309,1979. 32. Musajo, L., and Rodighiero. G. Studies on the photo-C4-cyclo-additionreac 6. Johnston, B. H., Johnson, M. A., Moore, C. B., and Hearst, J. E. Psoralen- tions between skin-photosensitizing furocoumarins and nucleic acids. Photo DNA photoreaction: controlled production of mono- and diadducts with nano chem. Photobiol., Ã•7:27-35,1970. second ultraviolet laser pulses. Science(Wash. DC), 797: 906-908,1977. 33. Ashwood-Smith, M. J., Poulton, G. A., Barker, M., and Mildenberger, M. 5- 7. Pani, B., Babudri, N., VenturingS., Tamaro, M., Bordin, F., and Monti-Bragadin, Methoxypsoralen, an ingredient in several suntan preparations, has lethal, C. Mutation induction and killing of prokaryotic and eukaryotic cells by 8- mutagenic and clastogenic properties. Nature (Lond.),285: 407-409, 1980. methoxypsoraten, 4,5'-dimethylangelicin, 5-methylangelicin, 4'-hydroxy- 34. Rodighiero, G., Musajo, L., Dall'Acqua, F., Marciani, S., Caparale, G., and methyl-4,5'-dimethylangelicin. Teratog. Carcinog. Mutagen., 1: 407-415, Ciavatta, L. Mechanism of skin photosensitizatJonby furocoumarins: photo- 1981. reactivity of various furocoumarins with native DNA and with ribosomal RNA. 8. Seki, T., Nozu, K., and Kondo, S. Differentialcauses of mutation and killing in Biochim. Biophys. Acta, 277: 40-49,1970. Escherichia coli after psoralen plus light treatment: monoadducts and cross 35. Dall'Acqua, F., Marciani, S., Ciavatta. L., and Rodighiero, G. Formation of links. Photochem. Photobiol., 27: 19-24,1978. inter-strand cross-linkings in the photoreactions between furocoumarins and 9. Dal!' Acqua, F. Photochemical reactions of furocoumarins. In: C. HÃ©lÃ¨ne,M. DNA. Z. Naturforsch Teil B Anorg. Chem. Org. Chem., 26: 561-569, 1971. Chartter,T. Montenay-Garestter,and G. Laustriat (eds.), Trends in Photobiol 36. Grossweiner, L. I., and Smith, K. C. Sensitivity of DNA repair-deficient strains ogy, pp. 267-277. New York: Plenum Press, 1982. of Escherichiacoli K-12 to various furocoumarins and near-ultravioletradiation. 10. Kaye,J., Smith, C. A., and Hanawalt, P. C. DNA repairin humancellscontaining Photochem. Photobiol., 33: 317-323,1981. photoadducts of 8-methoxypsoraten or angelicin. Cancer Res., 40: 696-702, 37. Gruenert, D. C., and Cteaver,J. E. Repairof ultraviolet damage in human cells 1980. also exposed to agents that cause strand breaks, crosslinks, monoadducts 11. Zolan, M. E., Cotropassi, G. A., Smith, C. A., and Hanawalt, P. C. Deficient and alkylatkxis. Chem.-Btol.Interact., 33: 163-177, 1981. repair of chemical adducts in a DNA of monkey cells. Cell, 28:613-619,1982. 38. Gruenert, D. C., Kapp, L. N., and Cleaver,J. E. Inhibitionof DNA synthesis by 12. Zolan, M. E., Smith, C. A., Calvin, N. M., and Hanawalt, P. C. Rearrangement psoralen-induced testons in xeroderma pigmentosum and Panconi's anemia of mammalian chromatin structure following excision repair. Nature (Lond.), fibroblasts. Photochem. Photobiol., 47: 543-550,1985. 299: 462-464, 1982. 39. Gruenert, D. C., and Cteaver,J. E. Repair of psoraten-inducedcross-links and 13. Kittter, L., Hradecna, Z., and Suhnel, J. Cross-link formation of phage lambda monoadducts in normal and repair-deficient human fibroblasts. Cancer Res., DNA in situ photochemically induced by the furocoumarin derivative angelicin. 45:5399-5404,1985.

CANCER RESEARCH VOL. 45 NOVEMBER 1985 5398

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1985 American Association for Cancer Research. Induction of DNA-DNA Cross-Link Formation in Human Cells by Various Psoralen Derivatives

Dieter C. Gruenert, Michael Ashwood-Smith, Reginald H. Mitchell, et al.

Cancer Res 1985;45:5394-5398.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/45/11_Part_1/5394

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/45/11_Part_1/5394. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.