Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
[CANCER RESEARCH 41, 5033-5038, December 1981] 0008-5472/81 /0041 -0000$02.00 Repair of Psoralen-treated DMA by Genetic Recombination in Human Cells Infected with Herpes Simplex Virus1 Jennifer D. Hall2 and Karen Scherer Department of Cellular and Developmental Biology, University of Arizona, Tucson, Arizona 85 721 ABSTRACT Herpes simplex virus type 1 was treated with 4,5',8-trimethylpsoralen (psoralen) plus near-ultraviolet light in order to produce lesions (monoadducts and DMA cross-links) in the viral DNA. Human fibroblasts were infected by damaged virus under conditions in which either a single virus particle or several particles entered a given cell, and the fraction of virusproducing cells was determined. This fraction was significantly greater for multiply infected cells than for singly infected cells, indicating that the psoralen lesions are repaired more efficiently in the presence of homologous, damaged DNA (multiplicity reactivation). Evidence is presented that herpes simplex virus may code for functions which participate in its own repair, both during multiplicity reactivation and during repair which occurs in singly infected cells: (a) host cells deficient in repair of lesions induced by psoralen (xeroderma pigmentosum) or the DNA cross-link ing agent mitomycin C (Fanconi's anemia) exhibited normal levels of multiplicity reactivation of psoralen-treated herpes virus; (b) while xeroderma pigmentosum cells have been pre viously shown to be deficient in repair of psoralen-treated adenovirus under conditions of single infection, herpes virus is repaired at near normal levels in these same cells. Recombination levels between genetically marked pairs of herpes viruses were found to increase after treatment of the parental viruses with psoralen, suggesting that psoralen dam age stimulates genetic recombination. This stimulation provides convincing evidence for a repair pathway in which genetic recombination between damaged viral genomes can lead to the production of viable virus. INTRODUCTION Psoralen3 reacts with DNA in the presence of near-UV3 light to produce both monoadducts and diadducts (cross-links) (4). Both types of lesions are repaired in prokaryotic and mamma lian systems (1, 5). While repair of monoadducts appears to involve excision repair pathways, there is considerable evi dence that cross-links are repaired by pathways involving both excision of adducts and genetic recombination (6). Genetic recombination is presumably required since both strands of a DNA duplex are damaged by the cross-link. In Escherichia coli, the mechanism of cross-link repair has been elucidated with genetic mutants deficient in excision repair of UV-damaged DNA and/or in genetic recombination. ' Support by Grant AG01689 from the NIH. 2 To whom requests for reprints should be addressed. 3 The abbreviations used are: psoralen, 4,5',8-trimethylpsoralen; XP, xero derma pigmentosum; FA, Fanconi's anemia; HSV-1, herpes simplex virus type 1. Received May 21, 1981 ; accepted July 30, 1981. These studies (6) suggest that excision enzymes cut one strand of the damaged DNA on both sides of a cross-link to remove the lesion from this strand (half-excision). The gap produced by excision is filled with homologous DNA from a sister chro mosome. Finally, the psoralen adduct is removed from the second strand by excision followed by DNA synthesis to fill this second gap. A role for genetic recombination in cross-linked repair has been further suggested by the ability of cross-links to stimulate genetic recombination (16). Damage produced by cross-linking treatments (such as by psoralen or mitomycin C) in bacterial systems is also repaired by a recombinational mechanism called multiplicity reactiva tion. In this mechanism, damaged bacteriophages are repaired more extensively in cells infected with several viral genomes than in singly infected cells (5, 15), providing that a functional recombination mechanism exists (15,17). Furthermore, cross links, as opposed to monoadducts, in bacteriophage \ DNA are not repaired unless 2 or more homologous viral genomes are present in the same infected cell (5), indicating the absolute requirement for multiplicity reactivation in repair of cross-links. Repair of psoralen-induced DNA cross-links in mammalian cells also appears to involve excision and recombination func tions. A role for genetic recombination has been suggested since cross-linking treatments induce recombination in lower eukaryotes (14, 21 ) and stimulate sister chromatid and nonsister homolog exchanges in mammalian cells (24, 27). The fact that psoralen-induced cross-links in adenovirus are not re paired during infection of human cells (8), under conditions of single infection, further implies that cross-links cannot be re paired in the absence of homologous DNA. Excision functions also appear to be required for repair of cross-links in human cells, since at least one cell line from a patient with XP3 is deficient in half-excision of psoralen-induced cross-links (18). These XP cells are also deficient in excision of UV-induced damage from DNA (12). FA3 cells are deficient in half-excision of cross-links induced by mitomycin C (9) but appear proficient in psoralen cross-link repair (18). These observations suggest that different types of cross-links may be repaired by different pathways in human cells. Monoadducts formed by psoralen treatment are presumably repaired by excision repair functions in bacteria (6). Excisiondeficient human XP cells show abnormally low survival of psoralen-treated adenovirus when infected by single viral par ticles (8). Since, under these conditions, viral DNA cross-links were not repaired, the excision function deficient in these XP cells appears to be required for repair of psoralen-induced monoadducts. Genetic recombination has been shown previously to be involved in the repair of UV-irradiated herpes simplex virus (10, 26). These experiments demonstrated multiplicity reactivation of irradiated virus and showed that UV irradiation of the virus DECEMBER 1981 Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1981 American Association for Cancer Research. 5033 J. D. Hall and K. Scherer stimulated genetic recombination between viral genomes. The role of genetic recombination in repair of psoralen-induced lesions in herpes-infected cells has not been studied previ suspended in a small volume of culture medium containing 1% calf serum, and sonicated. The virus suspensions were then clarified by centrifugation and stored at -80°. ously. The goal of the present experiments was to investigate whether psoralen damage in human cells can be repaired by genetic recombination. HSV-13 was used as a probe in these The procedure for plaque assays with intact virus on human or cells has been described (10). Psoralen Treatment. Psoralen was dissolved in ethanol at 1.5 ml and frozen at —20°.Before use, the psoralen suspension heated to 47° for 30 min to redissolve the psoralen precipitate studies because genetically marked viral mutants are available to study damage-induced recombination levels. In addition, the multiplicity of infections can be controlled so that repair of damaged virus can be compared in singly infected and multiply infected cells (i.e., under conditions in which genetic recom bination between viral genomes either cannot or can occur). We present evidence that repair of psoralen damage is sub stantially increased in multiply infected cells, suggesting that genetic recombination is involved in this repair. We have further shown that psoralen damage to viral DNA increases the yield of viral recombinants, providing further support for the idea that genetic recombination is involved in repair of psoraleninduced lesions. MATERIALS saline (NaH2PCvH2O, mg/ was and 0.34 g/ liter; Na2HPO4, 1.93 g/liter; NaCI, 8.5 g/liter) to give a final concen tration of 50 iig/ml. Virus was diluted 1:10 into the psoralen-buffer mixture and irradiated at ice temperature in a liquid layer approximately 1 mm thick. Samples were kept in the dark for 10 min prior to irradiation to allow the psoralen time to intercalate into viral DNA. This time was more than adequate to allow maximum photosensitization. Irradiation was carried out in plastic Retri dishes with the tops on or in plastic tubes to minimize exposure to shorter wavelengths. Controls (not shown) indicated that exposure of virus suspensions to irradiation in the absence of psoralen does not significantly alter the viability of the virus at the doses used. Irradiation was carried out with a Sylvania Blacklite blue lamp (F15T8). The dose rate was 8 to 10 J/sq m/sec as measured by a Blak-Ray UV meter (Ultraviolet Products, Inc., San Gabriel, Calif.). The suspension was swirled periodically during irradia tion to ensure a uniform exposure dose. UV Irradiation. UV irradiation of HSV-1 was performed as described AND METHODS Cell Lines and Virus Stocks. The characteristics of the cells and virus strains used are described in Tables 1 and 2. Cells were grown in Dulbecco's modified Eagle's medium supplemented with 20% fetal calf serum (human cells) or 10% calf serum (monkey cells) and incu bated in an atmosphere of 10% CO2 at 37°. Virus stocks were prepared on Vero cells at 37°(wild type) or 33° (temperature sensitive) by inoculating progeny from a single plaque onto cells at a multiplicity of infection of 0.01 plaque-forming unit/cell. When cytopathic then diluted into phosphate-buffered Vero effects were observed, infected cells were harvested, previously (10) using a General Electric germicida! lamp with an output maximum at 254 nm. Infective Center Assay. The infective center assay has been de scribed previously (10). Briefly, human cells were infected with a known number of virus particles. Cells infected with irradiated virus received the same total number of viral particles (damaged plus residual undam aged) as cells infected with unirradiated virus. Infected cells were then washed, trypsinized, and seeded with an excess of uninfected Vero cells in the presence of human y globulin. The ability of the initially infected cells to produce virus was then determined from the numbers Table 1 Cell lines Source Line CRL 1220 XP12BEorCRL1223 American Type Culture Collection (Rockville, Md.) American Type Culture Collection Genetic characteristics Human skin fibroblasts from a normal 15-yr-old male Human skin fibroblasts from a 7-yr-old female with XP (complementation Group A) skin fibroblasts from a 6-yr-old male with FA American Type Culture Collection Established cell line derived from African green monkey kidney cells Gift of P. Henson (Harvard University)Human Established proficientTable cell line derived from African green monkey kidney cellsPresumably Type Culture Collection CCL 122 Vero or CCL 81 TC-7American Ability to repair UV- or psoralen-induced age dam Presumably proficient Deficient in excision repair of UV damage (19), psoralen cross-links (18), psoralen monoadducts (8) proficient Presumably proficient Presumably 2 strainsComplementa HSV-1 tion group characteristicsTemperature (25)A16 Strain positionMaps map 1-1SourceGift F17 1-6 BU"-A PAA"-5 CI 101 5034 I-4 of P. Schaffer (Harvard Univer-Genetic sensitive; DMA-negativePhysical near mutants Cand D on genetic map (22) C and D map at 0.386-0.418 (3) sity) phenotype at nonpermissive tempera ture (2) Gift of P. Schaffer Temperature sensitive; DNA-positive Same complementation group as F18 which maps at 0.086-0.103 (23) phenotype at nonpermissive tempera ture (2) Spontaneous mutant of the KOS Resistant to bromodeoxyuridine; presum Presumably maps in the thymidine kinase strain of HSV-1 (J. Hall, unpub gene at 0.300-0.309 (11) ably deficient in viral thymidine kinase lished results) Gift of D. Coen (Harvard University) Maps in the DNA polymerase gene at 0.400Resistant to phosphonoacetic acid 0.418(3) Originally obtained from S. Kit (Bay Wild type lor College of Medicine) CANCER RESEARCH VOL. Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1981 American Association for Cancer Research. 41 Repair of Psoralen Damage in Herpes-infected Cells of plaques in this assay. The average multiplicities of infection were calculated as described previously (1 0) by comparing the virus titers in virus suspensions that were removed from cells infected with unirradiated virus and in suspensions prior to infection. The amount of adsorbed virus was then compared to the number of cells per dish to determine the average multiplicity of infection. Recombination Assay. The recombination assay has been de scribed previously (1 0). Briefly, human cells were multiply infected with pairs of genetically marked viruses. The viruses were treated with psoralen and irradiated prior to infection as indicated. Infected cells were incubated for 1 hr at 37° to allow viral adsorption. The viral suspensions were then removed, and 1 ml of fresh medium was added per plate. Incubation was continued at 33° (temperature-sensitive viruses) for 24 hr or at 37° (temperature-resistant viruses) for 18 hr, and the viral progeny were harvested. The progeny viruses were then titered to determine the recombination frequency. When temperaturesensitive parental viruses were used, this frequency was determined by measuring the relative yields of recombinant (temperature-resistant) viruses at 39°and total viruses (temperature resistant plus temperature sensitive) at 33° on Vero cells. When drug-resistant parental viruses were used, this frequency was determined by measuring the relative yields of recombinant viruses [those able to form plaques in the presence of phosphonoacetic acid (100 /ig/ml) and bromodeoxyuridine (100 fig/ml)] and the total viral yield in the absence of these drugs. The assays with drug-resistant viruses were performed on TC7 cells. Reversion of temperature-sensitive viruses or production of spontaneous drug-resistant mutants occurred in these assays at very low levels and did not affect the calculation of recombination frequen cies (data not shown). Cellular thymidine kinase (present at low levels in monolayer cells) did not affect the results of these measurements since similar results were obtained with iododeoxycytidine, a com pound which is phosphorylated by the viral, but not the cellular, thymidine kinase. The percentage frequency of recombination (RF) was calculated as: Recombinant — virus titer Total virus titer X ¿.DU RESULTS In order to determine whether genetic recombination might be involved in repair of psoralen damage in human cells, HSV1 was treated with psoralen plus light and allowed to infect cells at various multiplicities of infection. This approach facili tated comparison of repair in multiply infected cells, in which genetic recombination between viral genomes could occur, and in singly infected cells, in which genetic recombination is absent. In this assay, human fibroblasts were first infected at the appropriate multiplicities and then seeded with an excess of uninfected monkey kidney cells. The survival of plaque forma tion of the infected cells was then determined (Chart 1). Multi plicities of infection were determined from control infections with undamaged virus. Cells infected with psoralen-treated virus received the same number of viral particles (damaged plus residual plaque forming) as cells in these control infec tions. Two multiplicities of infection were used. Multiplicities less than 2 x 10~2 plaque-forming unit/cell were chosen to ensure that the cells would be singly infected. Since herpes virus preparations typically contain 10 to 100 defective parti cles for every infectious virus, this multiplicity should also minimize multiple infections with defective particles. A multi plicity of infection of approximately 3 plaque-forming units/cell was also used. DECEMBER 1981 300 600 300 600 IRRADIATION DOSE ( J/tq m ) 600 900 Chart 1. Survival of plaque formation by human fibroblasts infected at various multiplicities with herpes simplex virus treated with psoralen plus light. The infective center assay was performed as described in "Materials and Methods." Human fibroblasts used in these experiments were: A, CRL 1220 (normal); B, CRL 1223 (XP, Group A); and C, CCL 122 (FA). The average multiplicities of infection (plaque-forming units/cell) for cells infected with unirradiated virus are indicated in parentheses in the figures. Cells infected with irradiated virus were infected with the same number of viral particles as cells infected with unirradiated virus. Plating efficiencies of infective centers were greater than 34%. , theoretical survival curves for plaque formation by cells infected at the higher multiplicity of each graph, assuming that plaque formation occurred only by infection of cells with undamaged virus. The equation from which these curves were generated has been derived previously (10), assuming a Poisson distribution for the multiplicity of infection by both damaged and undamage virus particles. Theoretical values were calculated to an accuracy of 0.3%. A comparison of the survivals of plaque formation by singly or multiply infected normal human fibroblasts is shown in Chart ÃŒA. Clearly, the survival for multiply infected cells is substan tially greater than for singly infected cells. Theoretical curves based on the expected survival at the higher multiplicity due to infection by residual undamaged viral particles are presented in Chart 1 as dashed lines. These theoretical values are lower than the observed survivals for multiply infected cells at irradia tion doses of 300 J/sq m or greater. This result suggests that survival in multiply infected normal human cells occurs by pathways not available in singly infected cells. This increased repair will be referred to as multiplicity reactivation. Two other human fibroblast cells lines were tested for repair of psoralen-damaged herpes virus as described above. The XP cell line, CRL 1223, has been previously reported to be defi cient in repair of psoralen-induced cross-links (18). Certain FA cells have been found to be deficient in repair of mitomycin Cinduced cross-links (9), while another cell line was found to repair psoralen cross-links normally (18). Both the FA (CCL 122) and XP (CRL 1223) cell lines showed increased viral repair in multiply infected cells as compared to singly infected cells (Chart 1, ßand C), suggesting that they are proficient in multiplicity reactivation. Apparently, the host functions deficient in these mutant cells are not required for multiplicity reactiva tion of psoralen-damaged herpes virus. Repair of psoralen-induced monoadducts appears to involve excision repair functions in bacteria (6) and is substantially reduced in human XP cells singly infected with psoralen-dam aged adenovirus (8). To determine whether cellular excision functions might also be utilized during repair of monoadducts in herpes simplex virus DMA,we compared the relative abilities of various human host cells to repair psoralen-damaged herpes virus under conditions of single infection. The results in Table 3 show that for FA cells (CCL 122) the survival of plaque 5035 Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1981 American Association for Cancer Research. J. D. Hall and K. Scherer Table 3 Survival of HSV-/ plaques on human fibroblasts I.O (A) Human fibroblasts were infected with HSV-1, as described in the legend to Chart 2. D37values for the curves shown in Chart 2 and for the CCL 122 cell line (curve not shown) were determined as the doses giving 1/e survival. lineCRL Cell , • for psoralen-treated virus192(1.00)" 1220 (normal) CRL1223(XP) CCL 122 (FA)O Numbers in parentheses, 1220) cells. 152(0.79) 282 (1.47)D37 for virus39.5(1.00) UV-irradiated O.I 13.0(0.33) Not done relative D,,- values compared to normal (CRL Õ formation, as indicated by the D37 value, was similar to that seen for normal cells. Survival on XP (CRL1223) fibroblasts showed only a slight reduction compared to normal cells. This result suggests either that repair of psoralen damage in singly infected cells does not require the repair functions deficient in FA or XP cells or that viral functions can complement these deficiencies. Previous studies have shown that repair of HSV-1 damaged by UV is abnormally low in singly infected XP cells (10, 20, 26), suggesting that host cell excision functions are required for repair of UV-induced lesions in herpes DMA. In contrast, as shown above, host excision functions are not required for repair of psoralen damage in herpes DNA. We wished, there fore, to confirm this difference by comparing the role of host excision functions in repair of psoralen-treated and UV-irradi ated HSV-1 under the same experimental conditions and with the same cell lines. For these studies, both the normal human fibroblast line (CRL 1220) and the XP line (CRL 1223) were used. As shown in Chart 2 and Table 3, psoralen-treated virus was repaired at near normal levels in XP (CRL 1223) cells. In contrast, UV-irradiated virus was repaired at significantly lower levels in XP (CRL 1223) cells than in normal cells (CRL 1220). The D37 value for survival of UV-irradiated HSV-1 on XP (CRL 1223) cells was 3.0-fold less than that seen on normal cells (CRL 1220). This result suggests that, under conditions of single infection, host cell excision repair functions are utilized during repair of UV-irradiated HSV-1 in normal human cells but that psoralen-treated virus does not require these same func tions for repair. In bacterial systems, genetic recombination is induced by DNA cross-links (16). In addition, genetic recombination of herpes virus is stimulated by UV irradiation (10). If genetic recombination is also involved in multiplicity reactivation of psoralen-treated herpes virus, then psoralen damage may also stimulate genetic exchanges between herpes genomes. To test this possibility, cells were infected with pairs of psoralen-treated herpes mutants and the formation of recom binant progeny was measured (Chart 3). Cells were multiply infected with each mutant to allow recombination between damaged viral genomes to take place. Infection was conducted with complementing mutants to avoid any effect of defective viral functions on the recombination levels measured. The results in Chart 3A show that, with 2 different temperaturesensitive viral mutants (A16 and F17), psoralen treatment stim ulated the yields of temperature-resistant recombinants. In Chart 36, 2 mutants resistant either to bromodeoxyuridine [BUR-A, deficient in the viral thymidine kinase (28)] or to phosphonoacetic acid [PAAR-5, carrying a mutation in the viral DNA polymerase (13)] were tested for the production of recombi- 5036 O.I O 4OO 800 IRRADIATION DOSE( J/«qm) Chart 2. Survival of plaque formation by human fibroblasts infected with HSV1 treated with psoralen plus light or with UV light. Virus was treated with UV (A) as described (10) or with psoralen plus light (B) as described in "Materials and Methods." Human fibroblasts were then infected with virus in a plaque assay, and the survival of plaque formation was determined. Curves in B and upper curve in A represent composites of 2 experiments. Lmes indicate the least squares fit of an exponential curve to the points. •,CRL 1220 (normal); O, CRL 1223(XP). 1.0 (B) (A) 0.5 OX) 600 0 300 600 IRRADIATION DOSE ( J/sq m ) Chart 3. Yield of temperature-resistant herpes virus recombinants in human fibroblasts infected with virus treated with psoralen plus light. Crosses between pairs of viral mutants and calculation of the recombination frequencies were performed as described in "Materials and Methods." In each experiment, only 300 one parental virus was exposed to psoralen and/or light. Spontaneous recom bination frequencies in each experiment are shown by error bars. A, A16 x F17; A16 was treated with psoralen and/or light; B, BUB-A x PAAn-5, PAAR-5 was treated with psoralen and/or light. •,virus treated with psoralen plus light; O, untreated virus; G, virus treated with light in the absence of psoralen. Plaqueforming units (titers determined prior to irradiation) added per cells were: 8.2 (A16); 1.2(F17); 3.2 (BUR-A); 3.2 (PAAB-5). nants resistant to both drugs. Again, the recombination fre quency increased following psoralen damage. These results clearly demonstrate that psoralen damage stimulates viral re combination and may be involved in repair of these lesions. CANCER RESEARCH VOL. 41 Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1981 American Association for Cancer Research. SÉF- Repair of Psoralen Damage in Herpes-infected DISCUSSION The present results demonstrate that repair of psoralendamaged herpes simplex virus occurs more readily in multiply infected human cells than in singly infected cells, a phenome non called multiplicity reactivation. In addition, under condi tions of multiple infection, psoralen damage stimulates the production of viral genetic recombinants. Since these pro cesses were both observed under the same experimental con ditions, it seems probable that they are functionally related. Taken together, these results strongly suggest that genetic recombination is involved in the repair of psoralen-induced lesions, if multiple copies of genomes are available. A role for genetic recombination in multiplicity reactivation is made more plausible by the requirement previously established for recom bination in multiplicity reactivation in bacterial systems (15,17) and in human cells infected with UV-irradiated herpes virus (10). While psoralen-induced cross-links are removed from cellu lar DNA in diploid human cells (18), they are not removed from adenovirus DNA under conditions in which human cells are infected with single-viral particles (8). These results suggest that repair of DNA cross-links in mammalian cells may require genetic recombination between homologous DNA duplexes, as found previously for bacterial systems (5, 6). Repair of monoadducts, on the other hand, occurs extensively in the absence of genetic recombination, as indicated by the substantial re duction in repair of psoralen-treated adenovirus in singly in fected XP cells compared to normal cells (8). Consequently, our results with psoralen-treated HSV-1 suggest that the in creased survival of virus in multiply infected cells is due pri marily to an ability of these cells to repair DNA cross-links in viral DNA. However, we cannot rule out the possibility that psoralen-induced monoadducts also are repaired by multiplic ity reactivation and stimulate recombination in herpes-infected cells. We are presently measuring the yield of psoralen-induced cross-links produced under our conditions of treatment to distinguish between repair of monoadducts and cross-links. XP cells have been shown to be deficient in repair of psora len-induced cross-links (18). However, in the present study, these cells exhibit extensive multiplicity reactivation of psora len-treated herpes virus. Since multiplicity reactivation of herpes virus probably repairs DNA cross-links, the proficiency with which XP cells carry out multiplicity reactivation of herpes virus suggests that cross-links in viral DNA are repaired nor mally in these cells. This result further suggests that cross links in viral DNA are repaired by a different mechanism than are cross-links present in cellular chromosomes, possibly due to the presence of viral-induced repair functions. Repair of psoralen-damaged HSV-1 is only slightly reduced in singly infected XP cells compared to wild-type cells. In contrast, repair of psoralen-treated adenovirus is substantially reduced in XP cells (8). These results suggest that herpes virus but not adenovirus might provide one or more functions which can complement the deficiency in XP cells during repair of psoralen damage. Since it is probable that monoadducts, not cross-links, are repaired under conditions of single infection, these postulated herpes repair functions appear to act on monoadducts. Alternatively, structural differences in herpes genomes (such as a paucity of histones) might allow the repair of psoralen damage in XP cells, while similar lesions in either DECEMBER 1981 Cells adenovirus or cellular chromosomes might be inhibited. Unlike repair of psoralen damage, repair of UV-irradiated herpes virus is significantly reduced in singly infected XP cells relative to wild type. Repair of UV-irradiated adenovirus, like wise, is inhibited in XP cells (7). Apparently, herpes functions cannot complement the excision repair deficiency in XP cells for repair of UV-induced lesions. XP cells are thought to be deficient in the UV endonuclease which initiates excision repair oadducts (8). To explain the results with herpes virus, we hypothesize that herpes repair functions are more specific than the corresponding host function and act on psoralen adducts but fail to recognize UV damage. In summary, multiplicity reactivation of herpes simplex virus can best be explained as involving a genetic recombination mechanism. Multiplicity reactivation of HSV-1 has previously been observed for repair of UV damage (10, 26) and alkylation damage." Therefore, this mechanism may constitute a major repair pathway in multiply infected cells, capable of acting on many different types of lesions. The possibility that herpes viruses are repaired by genetic recombination also suggests that a similar mechanism may exist in human cells, made possible by the diploid nature of these cells. ACKNOWLEDGMENTS The authors thank Dr. D. W. Mount for many helpful discussions. R. E. Almy for excellent technical assistance. Drs. P. Shaffer and D. Coen for the generous gift of the herpes mutants, and Dr. P. Henson for kindly providing the TC-7 cells. REFERENCES 1. Ben-Hur, E., and Elkind, M. M. DNA cross-linking in Chinese hamster cells exposed to near ultraviolet light in the presence of 4,5',8-trimethylpsoralen. Biochim. Biophys. Acta. 33Õ. 181-193, 1973. 2. Benyesh-Melnick, M., Schaffer, P. A., Courtney, R. J., Esparza, J., and Kimura, S. Viral gene functions expressed and detected by temperaturesensitive mutants of herpes simplex virus. Cold Spring Harbor Symp. Quant. Biol.,34. 731-746, 1974. 3. Chartrand. P., Crumpacker, C. S., Schaffer, P. A., and Wilkie, N. M. Physical and genetic analysis of the herpes simplex virus DNA polymerase locus. Virology, 103: 311-326, 1980. 4. Cole, R. S. Psoralen monoadducts and interstrand cross-links in DNA. Biochim. Biophys. Acta, 254. 30-39, 1971. 5. Cole, R. S. Inactivation of Escherichia coli. F' episomes at transfer, and bacteriophage lambda by psoralen plus 360-nm light: significance of deoxyribonucleic acid cross-links. J. Bacteriol., »07:846-852. 1971. 6. Cole, R. S. Repair of DNA containing ¡nterstrand crosslinks in Escherichia coli: sequential excision and recombination. Proc. Nati. Acad. Sei. U. S. A., 70. 1064-1068. 1973. 7. Day, R. S. Cellular reactivation of ultraviolet-irradiated human adenovirus 2 in normal and xeroderma pigmentosum fibroblasts. Photochem. Photobiol., Õ9:9-13, 1974. 8. Day, R. S., Giuffrida, A. S., and Dingman, C. W. Repair by human cells of adenovirus-2 damaged by psoralen plus near ultraviolet light treatment. Mutât.Res., 33: 311-320, 1975. 9. Fujiwara, Y , Tatsumi, M., and Sasaki, M. S. Cross-link repair in human cells and its possible defect in Fanconi's anemia cells. J. Mol. Biol.. / 13 635649, 1977. 10. Hall, J. D., Featherston, J. D., and Almy, R. E. Evidence for repair of ultraviolet light-damaged herpes virus in human fibroblasts by a recombi nation mechanism. Virology, »05.490-500, 1980. 11. Halliburton, I. W., Morse. L. S., Roizman, B., and Quinn, K. E. Mapping of the thymidine kinase genes of type 1 and type 2 herpes simplex viruses using intertypic recombinants. J. Gen. Virol., 49. 235-253, 1980. 12. Hanawalt, P. C., Cooper, P. K., Ganasan, A. K., and Smith, C. A. DNA repair in bacteria and mammalian cells. Annu. Rev. Biochem., 48: 783-836, 1979. 13. Hay, J., and Subak-Sharpe, J. H. Mutants of herpes simplex virus types 1 and 2 that are resistant to phosphonoacetic acid induce altered DNA polymerase activities in infected cells. J. Gen. Virol., 3Õ. 145-148, 1976. ' J. Hall, unpublished results. 5037 Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1981 American Association for Cancer Research. J. D. Hall and K. Scherer 14. Holliday, R. The induction of mitotic recombination by mitomycm C in Ustilago and Saccharomyces. Genetics, 50: 323-335, 1964. 15. Holmes, G. E., Schneider, S., Bernstein, C., and Bernstein, H. Recombinational repair of mitomycin C lesions in phage T4. Virology, »03:299-310, 1980. 16. Howard-Flanders, P., and Lin, P.-F. Genetic recombination induced by DNA cross-links in repressed phage lambda. Genetics, 73: (Suppl.): 85-90, 1973. 17. Huskey, R. J. Multiplicity reactivation as a test for recombination function. Science (Wash. D.C.), 164: 319-320, 1969. 18. Kaye, J., Smith. C. A., and Hanawalt, P. C. DNA repair in human cells containing photoadducts of 8-methoxypsoralen or angelicin. Cancer Res., 40:696-702, 1980. 19. Kleijer, W. J.. de Weerd-Kastelein. E. A., Sluyter, M. L., Keijzer, W., de Wit, J.. and Bootsma, D. UV-induced DNA repair synthesis in cells of patients with different forms of xeroderma pigmentosum and of hétérozygotes. Mutât. Res., 20. 417-428, 1973. 20. Lytle. C. D.. Aaronson, S. A., and Harvey, E. Host-cell reactivation in mammalian cells. II. Survival of herpes simplex and vaccinia virus in normal human and xeroderma pigmentosum cells. Int. J. Radiât.Biol.. Relat. Stud. Phys. Chem. Med.. 22. 159-165, 1972. 21. Morpurgo, G. Induction of mitotic crossing-over in Aspergillus nidulans by 5038 bifunctional alkylating agents. Genetics, 48: 1259-1263, 1963. 22. Parris, D. S., Courtney, R. J., and Schaffer, P. A. Temperature-sensitive mutants of herpes simplex virus type 1 defective in transcriptional and posttranscriptional functions required for viral DNA synthesis. Virology, 90:177186, 1978. 23. Parris, D. S., Dixon, R. A. F., and Schaffer, P. A. Physical mapping of herpes simplex virus type 1 ts mutants by marker rescue: correlation of physical and genetic maps. Virology, 100: 275-287, 1980. 24. Perry, P., and Evans, H. J. Cytological detection of mutagen-carcinogen exposure by sister chromatid exchange. Nature (Lond.), 256. 121-125, 1975. 25. Schaffer, P. A.. Carter, V. C., and Timbury, M. C. Collaborative complemen tation study of temperature-sensitive mutants of herpes simplex virus types 1 and 2. J. Virology, 27. 490-504, 1978. 26. Selsky, C. A., Henson, P., Weichselbaum, R. R.. and Little, J. B. Defective reactivation of ultraviolet light-irradiated herpesvirus by a Bloom's syndrome fibroblast strain. Cancer Res., 39: 3392-3396, 1979. 27. Shaw, M. W., and Cohen, M. M. Chromosome exchanges in human leuko cytes induced by mitomycin C. Genetics, 51:181-190, 1975. 28. Summers, W. P., Wagner, M., and Summers, W. C. Possible peptide chain termination mutants in thymidine kinase gene of a mammalian virus, herpes simplex virus. Proc. Nat). Acad. Sei. U. S. A., 72: 4081-4084, 1975. CANCER RESEARCH VOL. Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1981 American Association for Cancer Research. 41 Repair of Psoralen-treated DNA by Genetic Recombination in Human Cells Infected with Herpes Simplex Virus Jennifer D. Hall and Karen Scherer Cancer Res 1981;41:5033-5038. Updated version E-mail alerts Reprints and Subscriptions Permissions Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/41/12_Part_1/5033 Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected]. Downloaded from cancerres.aacrjournals.org on June 16, 2017. © 1981 American Association for Cancer Research.