Somatic Cell and Molecular Genetics, Vol. 17, No. 3, 1991, pp. 229-238

Reactivation of Psoralen-Reacted Plasmid DNA in Fanconi Anemia, Xeroderma Pigmentosum, and Normal Human Fibroblast Cells Y. Sun and R.E. M o s e s 1 Department of Cell Biology, Baylor College of Medicine, ttouston, Texas 77030 Received 20 September 1990--Final 26 February 1991

A b s t r a c t - - W e have used a host cell reactivation system to study the effect of 8-methoxypsoralen

(8-MOP) reaction on CAT (chloramphenicol acetyltransferase) and NEO (aminoglycoside phosphotransferase) expression in normal human cells, as well as two cell lines with possibIe DNA repair-processing defects. Plasmid DNA was treated with psoralen plus near-ultraviolet (NUV) irradiation. The reacted plasmids, pSV2cat and pSV2neo, were transfected into Fanconi anemia (FA), xeroderma pigmentosum (XP), and normal human fibroblast cells for transient or stable assay. The cells were assayed for CAT activity at various times after transfection or selected for G418 resistance. The extent of adduct formation required to inhibit expression was much less (difference of D3ygreater than 2.5) in FA or XP cells' compared to normal. We conclude that in FA and XP cells, the reactivation of CA T was much less than in normal cells. The possibility of differential DNA uptake and~or degradation in transient assay was ruled out by analysis of plasmid DNA recovered from transfected cells. The data of the two independent assays indicate that FA and XP cells are deficient in cross-linked DNA repair.

INTRODUCTION

Monoadducts and interstrand crosslinks are produced when DNA is exposed to bifunctional alkylating agents, such as mitomycin C (MMC), nitrogen mustard, or psoralen plus near-ultraviolet (NUV) (1). Psoratens are tricyclic aromatic compounds and one of many carcinogenic and chemotherapeutic agents that intercalate into DNA molecules. They form covalent monoadducts specifically with the 5,6 double bond of pyrimidine bases, when exposed to NUV light (320-360 nm). Upon further irradiation, a fraction of the monoadducts can photoreact with a pyrimidine base on the opposite

strand to form an interstrand cross-link. Use of different wavelengths of NUV irradiation can enhance cross-link formation in a twostep treatment (2). The photochemical reaction of psoralens with DNA makes it possible to control the relative amount of the two qualitatively different ~pes of lesions, the bulky chemical monoadducts and covalent interstrand cross-links (3). It has been demonstrated that interstrand cross-links in DNA can block DNA replication and RNA transcription (4-6). It is not known if monoadducts inhibit transcription. We have chosen 8-MOP to study the effect of psoralen reaction on DNA expression. The mechanisms of cross-linked DNA

1Present address: Department of Medical Genetics, Oregon Healt]h Science University, Portland, Oregon 97201. 229

0740-7750/91/0500-0229506.50/0©1991 Plenum Publishing Corporation

230

repair have been studied in E. coli. Repair requires the uvr A B C nuclease and the recA gene product (4, 7-9). In human cells, it has been less well documented. A cell line with established deficiency in cross-link DNA repair would facilitate the study in human cells. Fanconi anemia (FA) is an autosomal recessive hematopoietic and skeletal disorder, characterized by growth retardation, several congenital malfunctions, and an increased incidence of leukemia and malignant solid tumors (10). Cells from FA patients are more sensitive to DNA crosslinking agents than normal cells (11). A deficient cellular DNA cross-link repair mechanism in FA cells has been suggested to explain the hypersensitivity to cross-linking agents. Conflicting data have been reported using different techniques (12-18). Xeroderma pigmentosum (XP) is a recessive disorder with skin fibroblasts sensitive to UV exposure due to a defect in removing pyrimidine dimers. There is evidence that XP cells show increased sensitivity to crosslinking agents, although the reports of cross-link DNA repair have also been conflicting (12, 17, 19, 20). We have used 8-MOP plus NUV irradiation to form adducts in pSV2cat and pSV2neo plasmid vectors, which can be expressed in human fibroblasts, to study the effect of DNA cross-links on plasmid transcription in human cells. This host cell reactivation assay has been used to examine the UV damage repair in mouse or human cells (16, 21-23). We found that CAT activity and the formation of G418-resistant colonies were sensitive to lower levels of psoralen treatment in both FA and XP(A) cells than in normal human fibroblast cells by either assay. The reacted DNA did not undergo differential degradation and/or replication in either cell line. Thus FA and XP(A) cells appear deficient in removal of DNA monoadducts or cross-links.

Sun and Moses

MATERIALS AND METHODS

8-MOP (Sigma) was dissolved in 100% ethanol and diluted to appropriate concentration prior to use. Cell Culture and Sensitivity Studies. The ceil lines used in these experiments were human fibroblast cells transformed by SV40. GM4312, GM0637, and GM6914 were purchased from the NIGMS Human Genetic Mutant Cell Repository. IMR90 and AG7217 were from the NIA Aging Cell Repository. IMR90XA is an immortalized cell line from strain IMR90, which was transformed with SV40 early DNA plasmid (in this laboratory). Cells were cultured in Dulbecco's modified Eagle's MEM containing 7.5% fetal calf serum. GM4312 is from complementation group A of XP cells, and GM6914 is a FA line, complementation group A. For kill curve studies, cells were plated in 10-cm plates at halt" a million per plate and incubated overnight. Medium was removed and the cells were rinsed with Hanks' balanced salt solution (HBSS). Cells then were incubated in HBSS with 10 ~xg/ml 8-MOP at room temperature for 1-2 h in darkness. The 8-MOP was rinsed away with HBSS. The cells were irradiated with NUV light through the bottom of the plate to filter away the UV fraction in the light source. The cells were covered with fresh medium and incubated for four to five days before counting. Plasmid Preparation and PsoraIen Treatment. pSV2cat (24) and pSV2neo (25) were prepared by lysozyme lysis and cesium chloride-ethidum bromide ultracentrifugation (26). DNA samples were checked on 1% agarose gel. For 8-MOP cross-linking, DNA was dissolved in 10 mM Tris HCt (pH 7.4) 1 mM EDTA (TE) at concentrations of i ixg/ixl and with 0.1 lxg/txl of 8-MOP. The mixture was incubated at 37°C for 10 min and then irradiated under NUV light (320-360 nm). Irradiation was through a plastic petri dish to

Cross-link Repair

231

absorb short-wavelength UV (4). The DNA was purified through a G-25 column and precipitated in EtOH and resuspended in TE. The treated DNA was denatured at 95°C for 5 min and quickly chilled in ice and run on a 1% agarose gel. The relaxed circle band and the single strand band were scanned on a densitometer. The degree of cross-linking (cross-links per molecule) was determined by Poisson distribution (27) based on DNA remaining at the relaxed circle position following denaturation-renaturation (Fig. 1). Transfection. We used the DEAE dextran method of transfection for CAT assay (28). Cells were plated at a density of half a miliion per 10-ram plate and cultured overnight. The medium was replaced with 2 ml of serum-free medium containing 10 p,g of plasmid DNA and 250 ixg/ml DEAE dextran. After a 4-h incubation, the cells were rinsed with Hanks' balanced salt solution and cultured in 10 ml of medium for 48 h. For the stable expression assay, we used the BES

buffer method (29). Cells were plated at half a million per plate and cultured for 16 h. Twenty micrograms of plasmid DNA was mixed with 0.5 ml of 0.25 M CaCt2 and 0.5 ml of 2 x BBS was added. The mixture was incubated for 10-20 min at room temperature. The calcium phosphate-DNA solution (1 rot) was added dropwise to the plate of cells, and the mixture was swirled gently and incubated for 15-24 h at 37°C under 3% CO2. The medium was removed and the cells were rinsed twice with growth medium, refed, and incubated for 24 h at 37°C under 7.5% CO 2. Transient CAT Assay. Cells were harvested 48 h after transfection by scraping with a rubber policeman, rinsed with Hanks', transferred to an Eppendorf tube, and resuspended in 100 ~I 0.25 M Tris, pH 7.5. The cells were lysed by freeze-thawing four times. The lysate was centrifuged and the supernatant was assayed for CAT activity (24). Fifty microliters of celt lysate, 8 p,1 [~4CJchloramphenicol, and 132 ~l of Tris

100

90 8O

70 c~

6O

"0 d)

50 co

3

40

30 20

10

0 .......

l 1.5 NUV

] .,t 1 3.0 4 . 5.................. 6 . 0 Dose J/m 2 (xtO -4)

Fig. 1. Dose-related NUV-induced DNA crossqinking.

232

were preincubated for 5 rain. Twenty microliters of 4-raM acetyl coenzyme-A was added and the reaction was incubated for 3 h at 37°C. The reaction was stopped by adding 0.6 ml ethyl acetate. The mixture was vortexed, and the ethyl acetate layer was dried by spinning vacuum. The product was resuspended in 12 p.1 ethyl acetate, and paper chromatography was performed. The plate was exposed to X-ray film, and the spots were cut out and counted for radioactivity. Stable Expression Assay. After transfection, cells were split and plated at half a million per plate and cultured in medium containing 200 Ixg G418/ml. Two to three weeks after selection, the colonies were stained with methylene blue and counted.

Recovery and Analysis of Transfected Plasmid. Cell lines were transfected by the DEAE dextran method as described above. Cells were harvested by trypsinization 12 and 48 h after the transfection. Cells then were treated with DNase at a concentration of 50 l~g/ml at 37°C for 30 rain. This removed plasmid DNA from the cell surface (23). Plasmid DNA was extracted from the cells by the method of Hirt (30). The supernatant was extracted with phenol and chloroform, and the DNA was precipitated with ethanol. The pellet was resuspended in TE and digested with HindIII, which cuts once in the pSV2 plasmid. The DNA (3 ~g/lane) was electrophoresed on a t% agarose gel, transfered to a nitrocellulose membrane and hybridized to random primer-labeled 32p probe. The probe was prepared by the following procedure: pSV2 cat was digested with HindIII and ScaI, and the smaller band

Sun and Moses

was isolated. The band contained the CAT coding region. An aliquot of the DNA was cut with DpnI (Boehringer Mannheim Biochemicals) and analyzed in the same Southern procedure (25). RESULTS

Psoralen Treatment of Plasmid DNA and Cell Survival. The dose-dependence of 8-MOP damage for plasmid DNA is shown in Fig. 1. The treated DNA was analyzed on a denaturing gel (Fig. 2). An interstrand cross-link prevented separation of DNA strands under the denaturing conditions used. Therefore, the band of the relaxed circle increased as the adduct formation became greater. The density of the relaxed circle DNA band and single strand DNA (linear and circle) was scanned with a densitometer. The portion of the non-crosslinked DNA was plotted against the NUV energy (Fig. 1), and the number of cross-links per molecule was determined by Poisson distribution. At about five cross-links per molecule, essentially all DNA had been cross-linked. Under our conditions monoadducts would be much more numerous (31). Figure 3 shows the cytotoxicity of psoralen plus NUV treatment on different cell lines. The FA cells have a D37 for killing two to three times less that of the normal cells, and the XP(A) cells show similar sensitivity. These results agree with the reports cited above. Transient Reactivation Assay. FA and XP cells show increased sensitivity to DNA cross-linking agents (11, 12). An explanation

Fig. 2. Plasmid D N A cross-linked with 8 - M O P plus N U V on a d e n a t u r i n g gel. L a n e 1, no N U V irradiation; lane 2, 1.5 x 10 3 J/mZ; lane 3, 3.0 x 10 3 J/m2; lane 4, 7.5 x 103 j/m2; lane 5, 1.5 x 10 4 J/m2; lane 6, 2.3 x 10 4 J/mS; lane 7, 4.5 x 10 4 J/m2; lane 8, 9.0 x 10 4 J/m s.

Cross-link Repair

233 100,

10

v

> >

O3 03

O 1.0

--

oGM6914(FA) A GM4312(XPA) o|MRg0XA(Normal) ~GM0637(Norrna )

o.it. . . .

t

J

720

2000 NUV (J/m 2 )

Fig. 3. Survival curve of normal, FA, and XP cells after treatment with 8-MOP plus increasing NUV irradiation. The bars represent the range of triplicate survival tests.

would be failure of repair of cross-links in these cells relative to normal. A method to measure this indirectly would be to assess gene expression after cross-linking DNA. Remaining cross-links would inhibit gene expression due to a block of transcription (4, 5). We chose to monitor CAT activity following 8-MOP treatment of plasmids carrying the CA T gene followed by transfection into cells. We transfected different immortalized human cells with pSV2cat, pSV2cat has a CAT gene under the control of the SV40 early promoter, which permits its expression in immortalized human cells. Cells were transfected with plasmid DNA treated with

psoralen to form 5-10 cross-links per molecule (Figure 1). Forty-eight hours after the transfection, CAT assays were performed. The CAT activity was normalized against ceils transfected with nontreated plasmid. Figure 4 shows the reactivation by the different human cell lines. FA cells had four cross-links per molecule in transfecting DNA, XP cells had four per molecule, and the normal lines about 10 per molecule for 37% actiwity measured by this assay (Fig. 5). Monoadducts would have been present at higher levels. Stable Expression Assay. The above results showed that fewer adducts per plasmid molecule were sufficient to inhibit CAT

234

Sun and Moses

Fig. 4. Effect of 8-MOP treatment on CAT activity. Each cell line was transfected with untreated DNA (1), five cross-links per DNA molecule (2), and 10 cross-links per DNA molecule (3). CAT activity was assayed at 48 h. (A) AG 7217 (normal). (B) GM 4312 (XP-A). (C) IMR90XA1 (normal). (D) GM6914 (FA). CM- chloramphenicol, 1 CM and 3 CM- are 1-acetyl and 3-acetyl derivations, respectively.

loo

0--"

I

5 Cross-LinkslMolecule

Fig. 5. Effect of 8-MOP treatment on CAT activity at 48 h.

,,1

10

Cross-link Repair

235

expression in FA and XP cells. This suggested that these cell lines were less capable of repairing the cross-links, which remained and inhibited transcription. In order to investigate whether transcription became better in FA or XP cells at longer times, we performed a long-term or "stable" assay. We used the plasmid pSV2neo to transfect the human cell lines and selected for G418resistant colonies. Figure 6 shows the decrease of surviving colonies at three weeks as the psoralen reaction per molecule increased. At 37% G418-resistant colonies remaining, the FA and XP cells had nine cross-links per molecule and the normal lines had 19 cross-links per molecule in the transfecting DNA. Thus the results of the transient and stable assays suggest that removal of cross-links in FA or XP cells is not simply delayed. Analysis of RecoveredPlasmid. The above results suggested that FA and XP cells did not repair DNA normally after psoralen treatment. Alternative bases for the results

obtained might be that psoralen treatment affected the uptake of plasmid DNA and/or that such DNA was degraded or replicated less. We investigated these possibilities by recovering the plasmid from transfected cells at two different time points and analyzing the plasmid. Cells were transfected with treated and nontreated pSV2cat plasmid and, after 15 and 60 h, cells were harvested and plasmid was recovered (26). Plasmid DNA was dige,sted with HindIII, which cuts once in the plasmid; analyzed on agarose gel; transferred to nitrocellulose filter; and hybridized to a 32p probe from the CAT coding region (Fig. 7). A portion of the recovered plasmid was digested with DpnI, which cut the methylated transfected plasmid DNA but not any newly replicated nonmethylated plasmid DNA. This fraction then also was analyzed by Southern analysis (Fig. 8). The results showed that there was no difference in uptake and/or replication between crosslinked and non-cross-linked plasmid and that the plasmid did not replicate detectably in

100

03

i 5c ©

00

10 Cross-Links/Molecure

20

Fig. 6. Effect of 8-MOP treatment on N E O activity at three weeks.

236

Sun and Moses

Fig. 7. Southern analysis of recovered plasmid. (A) DNA recovered 15 h after transfection. (B) DNA recovered 60 h after transfection. Lane 1, pSV2cat. Lane 2, GM4312 non-cross-linked. Lane 3, GM4312, five cross-links per molecule. Lane 4, GM4312, 10 cross-links per molecule. Lane 5, GM6914, non-cross-linked. Lane 6, GM6914, five cross-links. Lane 7, GM6914, 10 cross-links. Lane 8, IMR90XA, non-cross-linked. Lane 9, IMR90XA, five cross-links. Lane 10, IMR90XA, 10 cross-links.

the human cell lines since the DpnI digestion abolished the 5.0-kb band completely. This indicated the deficient reactivation in FA or XP cells compared to normal cell lines was due to differences in restoration of transcription, rather than differential DNA uptake, degradation, or replication.

DISCUSSION

The purpose of our study was to investigate 8-MOP repair in FA, XP(A), and control cells, using a plasmid reactivation assay similar to the one described by Dean et al. (16). This assay system was developed to

Fig. 8. Southern analysis of recovered plasmid treated with DpnI. (A) DNA recovered 15 h after transfection. (B) DNA recovered 60 h after transfection. Lane arrangement is the same as in Fig. 7.

Cross-link

Repair

demonstrate the cellular ability for repair of in vitro-damaged plasmid with an assayable function. We have found that XP and FA cells cannot reactivate gene expression to the same level as normal cells. These results lead us to conclude that the increased sensitivity of these cells to 8-MOP is due to failure of D N A repair. Our results are in agreement with those of Chu and Berg (32) for repair of cisplatin cross-linking in XP(A), except these authors noted increased expression at over 20 days of culture. A report by Fujiwara and Tatsumi (t2) suggested that XP cells were able to remove cross-links induced by treatment with mitomycin C. Day et al. (14) have shown that neither FA nor normal cells can repair cross-linked (psoralen + NUV) using a viral survival assay. It was also reported (12, 13) that F A cells are deficient in removing cross-links from DNA, but Poll et al. (15) obtained different results using the same techniques. Kaye et aL (17) observed no abnormality in the incision step on 8-MOP-produced damage. Dean et al. (16) used an F A line, GM6914, and nitrogen mustard as the in vitro cross-linker with the CAT assay. They found no difference between F A and normal cell lines after cross-linking of CAT activity. We used the same FA line, GM6914, and failed to observe normal reactivation. The D37values for C A T g e n e inactivation differ by a factor of 2.5 or more in our work. One possible explanation for the results we obtained might be that there are different repair mechanisms for different cross-linking agents. Dean et al. used nitrogen mustard, and we used psoralen plus N U V irradiation; this might produce different obselvations. Alternate explanations for these results were sought. We demonstrated that psoralen treatment of the plasmid did not affect the level of transfection and that there was no differential degradation of the plasmid after transfection. The best explanation for the results obtained in XP and F A cells is a

237 defect in removal of psoralen adducts from D N A and, therefore, failure to allow' transcription.

ACKNOWLEDGMENTS This work was supported by USPHS grants CA 37860, GM24711, and a grant from the Fanconi Anemia Research Fund, Inc.

LITERATURE CITED l. Cimino, G.D., Gamper, H.B., Isaacs, S.T., and Hearst, J.E. (1985).Annu. Rev. Biochem. 54"11511194. 2. Averbeck, D., Papadopoulo, D., and Moustacchi, E. (1988). CancerRes. 48:2015-2020. 3. Hearst,J.E., Isaacs, S.T., Kanne, D., Rapoport, H., and Straub, K. (1984). Q. Rev. Biophys. 17:1-44. 4. Sladek, F.M., Munn, M.M., Rupp, W.D., and Howard-Flanders, P. (1989). Jr- Biol. Chem. 264: 6755-6765. 5. Piette, J.D., and Hearst, J.E. (1983). Proc. Natl. Acad. Sci. U.S.A. 50:5540-5544. 6. Gruenert, D.C., Ashwood-Smith, M., Mitchell, R.H., and Cleaver, LE. (1985). Cancer Res. 45:5394-5398. 7. Jones, B.K., and Yenng, A.T. (1988). Proc. Natl. Acad. Sci U.S.A. 85:8410-8414. 8. Cole, R.S. (1973). Proc, Natl. Acad. Sci. U.S.A. 70:1064-1068. 9. vanHouton,B., Gamper, H., Holbrook, S., Hearst, J., and Sancar, A. (1986). Proc. Natl. Acad. Sci U.S.A. 83:8077-8081. 10. Friedberg, E°C. (i985). In DNA Repair (W.H. Freeman, San Francisco), pp. 505-575. I 1. Weksberg,R, Buchwald, M., Sargent, P., Thompson, M.W., and Siminovitch, L. (1979). J. Cell. PhysioL 101:311-324. 12. Fujiwara,Y., and Tatsumi, M. (1977).J. Mol. Biol. 113:635-649. 13. Fujiwara,Y. (1982).Biochim. Biophys. Acta 699:217225. 14. Day,P.S., lII, Giuffrida,A.S., and Dingman, C.W. (1975). Murat. Res. 33:311-320. 15. Poll, E.H.A., Arwert, F., Kortbeek, H.T., and Eriksson, A.W. (1984).Hum. Genet. 68:228-234. 16. Dean, S.W., Sykes, H.R., and Lehmann, A.R. (1988). Murat. Res. 194:57-63. 17. Kaye,J., Smith, C.A., and Hanawa|t, P.C. (1979). Cancer Res. 40:696-702. 18. Fornace, A.J., Jr., Little, J.B., and Weichselbaum, R.R. (1979).Biochim. Biophys. Acta 561:9%109. 19. Cleaver,J.E., and Bootsma, D. (1975).Annu. Rev. Genet. 9:19-38.

238

20. Bredberg, A., Lambert, B., and Soderhall, S. (1982). Murat. Res. 93:221-234. 21. Lehmann, A.R., and Domen, A. (1985). Nucleic Acids Res. 13:2087-2095. 22. Protic-Sabljic, M., and Kraemer, K.H. (1985), Proc. Natl. Acad. Sci. U.S.A. 82:6622-6626. 23. Sheibani, N., Jennerwein, M.M., and Eastman, A. (1989). Biochemistry 28:3120-2124. 24. Gorman, C.M., Moffat, L.F., and Howard, B.H, (1982). Mol. Cell. Biol. 2:1044-1051. 25. Southern, P., and Berg, P. (1982). J, Mol. Appl. Genet. 1:327-341. 26. Maniatis, T., Fritsch, E.F., and Sambrook, J.

Sun and Moses

27. 28. 29. 30. 31. 32.

(1982). (Cold Spring Harbor Laboratory, Cold Spring Harbor, New York). Kuhnlein, U., Penhoet, E.E., and Linn, S. (1976). Proc. Natl, Acad. Sci. U.S.A. 73:1169-1173. McCutchan, J.H., and Pagano, J.S. (1968). J. Natl. Cancer Inst. 41:351-357. Chen, C., and Okayama, H. (1987). Mol. Cell. Biol. 7:2745-2752. Hirt, B. (1967).J. Mol. Biol. 26:365-369. Kanne, D., Straub, K., Rapoport, H., and Hearst, J.E. (1982). Biochemistry 21:861-871. Chu, G., and Berg, P. (1987). Mol. Biol. Med. 4:277-290.

Reactivation of psoralen-reacted plasmid DNA in Fanconi anemia, xeroderma pigmentosum, and normal human fibroblast cells.

We have used a host cell reactivation system to study the effect of 8-methoxypsoralen (8-MOP) reaction on CAT (chloramphenicol acetyltransferase) and ...
735KB Sizes 0 Downloads 0 Views