Cytotoxicity of Carcinogenic Aromatic Amides in Normal and Xeroderma Pigmentosum Fibroblasts With Different DNA Repair Capabilities 1,2 Veronica M. Maher, Nancy Birch, James R. Otto, and J. Justin McCormick 3,4 SUMMARY-The effect of exposure to UV irradiation or to the N-acetoxy-ester derivatives of four carcinogenic aromatic amides, 4-acetylaminobiphenyl (AABP), 2-acetylaminofluorene (AAF), 2-acetylaminophenanthrene, and 4-acetylaminostilbene, on cell survival was compared in strains of cultured human fibroblasts possessing normal rates of excision repair of DNA and in thre,e strains of xeroderma pigmentosum (XP) cells, each differing in its rate of excision repair. The survival of each strain after exposure to UV reflected its capacity to repair DNA. Thus the slope of the survival curve for the XP strain with the poorest capacity for excision repair (XP12BE complementation group A) was 5.8-fold steeper than the exponential portion of the curve for the normally repairing strains; that of XP2BE (complementation group C) was 1.95-fold; and that of XP4BE (a variant capable of a normal rate of dimer excision) was only 1.3-fold steeper_ The slope of the survival curves after exposure to each N-acetoxy ester derivative for these same XP strains averaged 6.4, 2.0, and 1.4 times steeper, respectively, than that of the normal strains tested. The excision repair capacity of these lines after exposure to N-acetoxyAAF (50 I'M/ml) was tested with alkaline cesium chloride density gradient centrifugation to detect incorporation of tritiated thymidine into nonreplicated DNA. The normal strains and XP4BE exhibited DNA excision repair by this method, whereas XP patients 2 and 12 did not. The cytotoxic effect of the four parent aromatic amide carcinogens, their N-hydroxy derivatives, as well as the N-acetoxy ester of each of the four N-hydroxy compounds and the N-sulfate ester of N-hydroxyAAF and N-hydroxy-AABP in the XP2BE strain, was compared with their effect on the normal fibroblasts. The parent amides proved to be noncytotoxic at all doses tested. In contrast, the N-hydroxy derivatives of each aromatic amide were highly cytotoxic, as were the ester compounds. For each active derivative, the slope of the survival curve for XP2BE was 2-2.5 times steeper than that of the normally repairing strain.-J Natl Cancer Inst 1287-1294, 1975.
CELLULAR REP AIR ENZYMES can modulate the damage resulting from lesions induced in the DNA of cells by physical or chemical agents. For example, bacterial cells deficient in the ability to effect excision repair of pyrimidine dimers or postreplication repair of lesions resulting from dimers exhibit greatly increased susceptibility to killing by UV irradiation (1). Using an in vitro bacterial transforming DNA system, we have presented evidence that such cellular DNA repair capabilities operate not only on UV damage but also on le-
sions in DNA caused by covalent attachment of such carcinogenic chemicals as aromatic amide derivatives (2-4) or polycyclic hydrocarbons (5, 6). Evidence that human cells possess similar genetically determined repair capability has been obtained by comparison of the cloning capacity of cultured cells from xeroderma pigmentosum patients [(XP) inherited susceptibility to multiple carcinomas of the skin on exposed areas] with that of normal human fibroblasts (NF) after exposure to UV (7) and to derivatives of the carcinogen 4nitroquinoline-I-oxide (8, 9), as well as to benz[a]anthracene 5,6-oxide (10). Accompanying autoradiography studies provided physical evidence that normal human cells carryon more extensive unscheduled DNA synthesis than do XP cells after exposure to the above agents (8) and to N-acetoxy2-acetylaminoftuorene (N-AcO-AAF) and N-hydroxy2-acetylaminoftuorene (N-HO-AAF) (11). Further physical evidence of differential amounts of DNA repair by such cells after attack by N-AcO-AAF has been obtained by Setlow and Regan (12, 13) with their 313-nm radiation photolysis technique. The present studies were undertaken to compare 1) the cytotoxic effect on the cloning capacity of normal human cells in culture of four series of closely related active compounds from a major class of chemical carcinogens (the aromatic amides) , 2) the cytotoxicity of such carcinogens in strains of human fibroblasts differing in their sensitivity to UV irradiation and in their ability to repair UV-induced lesions in DNA, and 3) the ability of such strains to repair lesions induced in their DNA by these carcinogens by cesium chloride (CsCI) equilibrium-centrifugation techniques to detect incorporation of thymidine into parental DNA strands. 1 Received September 13, 1974; revised January 23, 1975; accepted February 21, 1975. 2 Supported by Public Health Service grants CAI3058 and CA14680 from the National Cancer Institute, grant CP33226-01 from the Division of Cancer Cause and Prevention of the National Cancer Institute, and by an institutional grant to the Michigan Cancer Foundation by the United Foundation of Greater Detroit. 3 Division of Biological Sciences, Michigan Cancer Foundation. 110 E. Warren Ave., Detroit, Mich. 48201. ! We thank Dr. Kerstin Philips, Chemistry Department, Michigan Cancer Foundation, for her valuable contribution in establishing the purity of the test compounds, and James Wessel, Susan Van Hoeck, and Marilyn Mittlestat for excellent technical assistance.
JOURNAL OF THE NATIONAL CANCER INSTITUTE, VOL. 54, NO.6, JUNE 1975
Downloaded from https://academic.oup.com/jnci/article-abstract/54/6/1287/920864 by University of California, Santa Barbara user on 04 March 2018
1287
1288
MAHER, BIRCH, OTTO, AND MCCORMICK
MATERIALS AND METHODS
Carcinogenic compounds.-The ester derivatives of the aromatic ami des tested were supplied by Dr. John Scribner, Fred Hutchinson Cancer Research Center, Seattle, Washington. The other compounds were prepared or obtained as described in (2, 4) . All were stored in a desiccator at - 20° C until use. The melting point of each chemical was identical to that in our previous experiments (2, 4). Except for the sulfate ester derivatives that were dissolved in sterile distilled water, stock solutions of each aromatic compound were prepared in 100% ethanol (EtOH) immediately before use; final concentration of EtOH in the medium during e:,"posure to the cells was less than O.S% The solutIOns were protected from light at all times. Cell cultures.-Stocks of NF derived from foreskins were established, cultured, and stored until needed in liquid nitrogen as described in (14-17). One normal strain from a IS-year old male was obtained from the American Type Culture Collection (Rockville, Md.). Fibroblasts, derived from skin biopsy specimens of XP patients (XP2BE, XP4BE, XPI2BE, and XPI3BE), also from the American Type Culture Collection, were similarly cultured. All cells were from stocks in passages 6-18 and were routiney shown to be free from mycoplasma according to the microbiologic culture techniques and polyacrylamide gel electrophoresis of labeled RNA (18). Culture medium.-Cells were cultured in a humid atmosphere of S% CO 2 and air at 37° C in Ham's F-lO medium (19) (Gibco, Grand Island, N.Y.), supplemented with IS% fetal calf serum (FCS) (BioQuest, Cockeysville, Md.) plus antibiotics (100 U penicillin/ml and 100 I'g streptomycin/ml). Assay for cytotoxicity of carcinogens.-We measured the cyto.toxicity of the compounds by plating an approprIate number of cells per dish (300-3,000), allowing 10-12 hours for cell attachment, replacing the culture medium with serumfree F-IO medium, and by introducing the test c?mp~unds .at speci~ed concentrations. All operatIOns mvolvmg carcmogen solutions were done in semidarkness. Microscopic examination demonstrated that, under the conditions used, less than 10% of the cells plated could be expected to divide before they were exposed to the compounds. Th~ ~xposure period (4 hr) was terminated by the addItIOn of fresh culture medium containing IS% FCS. The. I?edium should inactivate any remaining electrophilIc reactants (carcinogens), since proteins are strong nucleophiles. The cells were refed with fresh culture medium three times weekly until clones developed to macroscopic size. At this time they were fixed in methanol, stained with 2% methylene blue, and counted. The average number of clones per dish (10-12 dishes) divided by the number of cells inoculated yielded the cloning effic~ency for a particular dose. The cloning efficIency of the cells treated with a particular dose
divided by the cloning efficiency of the control cells, which received solvent only, determined the cytotoxicity of the compound and was expressed as percent. No cytotoxic effect was caused by the concentration of the solvent used. Exposure to UV irradiation.-After time was allowed for cell attachment, the culture medium was removed, and the cells were rinsed with 0.9% saline and exposed in semidarkness to the specified doses of irradiation with a shortwave UVS-12 lamp (Ultra-Violet Products, San Gabriel, Calif.) havmg an output of 2-4 ergs/mm 2/second. UV irradiation was measured with a J22S metering unit, also from U~tra-Violet Products. Cells were protected from lIght for 48-72 hours from the time of irradiation to prevent possible photoreactivation. Detection of repair of DNA.-Cell cultures of normal or XP fibroblasts in 2S0-ml flasks were prel.abeled with HC-thymidine (O.S ,1.tCi/ml; S6 mCI/mmole; Nuclear Dynamics, El Monte, Calif.) in culture medium for 20 hours. Unlabeled culture medium containing S-bromodeoxyuridine (BUDR, 10 I'g/ml; P. L. Biochemicals Inc., Milwaukee Wis) was supplied 1 hour before exposure to th~ carcmogens to label the density of the most recently replicated DNA fragments. .Cells "":ere exposed to N-AcO-AAF (SO I'M) 60 mmutes m the dark in serum-free F-IO medium plus B~DR (10 ~g/ml). Hydroxyurea (HU, 20 mM; SIgm.a ChemIcal Co., St. Louis, Mo.) was ~dd~d dUrIng the last 20 minutes to stop DNA replIcatIOn. After 60 minutes of carcinogen treatment, the medium was removed and the cells were given freshly prepared culture medium containing BUDR (10 t.tg/ml), HU (20 mM), and tritiated thymidine (3~-TDR, 10 t.tCi/ml, 40 Ci/mmole; Nuclear DynamICS) for 4 hours to label DNA during repair synthesis before the cells were harvested. If residual DNA synthesis were to occur despite the HU block, 3~-TDR incorporated during normal DNA syntheSIS rather than by repair synthesis would be d~tected in the region of higher density on the gradIent because of the simultaneous incorporation of BUDR. DNA extraction.-Cells were rinsed with saline, scraped from flasks, suspended in O.IS M NaCl, O'0.1S M sodium citrate (SSC) , pH 7.2, washed t~lCe, resuspended in SSC, and lysed with 0.2% sodIUm lauryl sulfate. PaI?-creatic (SO t.tg/ml) and Tl RNase (SO U /ml; CalbIOchem, Los Angeles, Calif.) was added for 30 minutes at 37° C, followed by nuclease-free Pronase (100 I'g/ml; Calbiochem) for 1 hour and deproteinization with chloroform: butanol (4: 1) was effected, after which the solution underwent dialysis against 2 liters of SSC for 2 hours. AI~aline CsCI centrifugation.-Alkaline CsCI gradIents w.ere prepared and centrifuged at 34,000 rpm accordmg to the methods of McCormick et al. (20). Fractions were collected, and the doublelabel radioactivity was determined in a spectrometer
Downloaded from https://academic.oup.com/jnci/article-abstract/54/6/1287/920864 by University of California, Santa Barbara user on 04 March 2018
EFFECT OF REPAIR ON CYTOTOXICITY OF CARCINOGENS
(Packard Tricarb, Packard Instrument Co., LaGrange, Ill.) as described in (20). The region of normal density (unreplicated parental) DNA was located from the 14C-TDR peak and the amount of 3H-TDR incorporated under this peak determined. These fractions were pooled and rebanded if necessary to insure that the tritium counts did not represent physically trapped material. RESULTS Differential Sensitivity to UV Irradiation
We compared the sensitivity of several strains of human cells in culture to UV light by measuring the survival of their cloning capacity. The cells thus compared represented several strains of normal diploid fibroblasts started from foreskin cultures, one normal strain started from a biopsy specimen of a 15-year-old, and strains of cells cultured from biopsy material from XP patients selected on the basis of the different ability of their lymphocytes and fibroblasts to incorporate thymidine in repair of DNA after UV irradiation (21). Thus XPl2BE from genetic complementation group A incorporated less than 2% of the amount of 3HTDR that normal cells did during the first 3 hours after exposure to UV irradiation (21); XP2BE, from group C, incoq:orated 15-25% of normal (22), and XP4BE 100% of normal (22). The latter XP patient, whose excision repair rate appeared normal, was designated an XP variant (7). Text-figure I compares the percent survival of the cloning capacity (as a function of UV irradiation) of the three XP strains with that of NF. The repairing NF exhibited a shoulder on the UV survival curve, which suggested DNA repair. A smaller shoulder was seen for the XP4BE cells, but the exponential portion of its survival curve had a slope only approximately 1.3-fold steeper than that of the normal cells. The slope of the XP2BE cells was 1.9 times steeper than that of normal cells, but because it lacked a shoulder, it required only 8
1289
ergsjmm 2 to reduce the survival level to 37% compared with approximately 48 ergsjmm 2 for the NF. Strain XPl2BE was more sensitive to the killing effect of UV irradiation. It had no shoulder and the slope of its survival curve was 5.8 times steeper than that of the linear portion of the curve for normal cells. A dose of only approximately 3 ergsjmm 2 brought these cells to a survival level of 37% of untreated controls. The survival of the cloning ability, as a function of UV, of NF taken from a 15-year-old boy was also determined to see if the difference between XP4BE and the NF could be attributed to age of the donor (XP4BE came from a 27-year-old male XP patient). There was no difference in sensitivity to UV between the normal strains taken from newborns and that of the 15-year-old. A second XP variant, XPI3BE, was then tested for survival of its cloning ability as a function of UV; this strain exhibited survival levels identical with those of XP4BE. Differential Cytotoxicity of Active Derivatives of Carcinogens
The survival of the cloning ability of the three XP strains, compared with that of normal fibroblasts, was then determined as a function of the concentration of the active N·acetoxy ester derivatives of four aromatic amide carcinogens, 4-acetylaminobiphenyl, 2-acetylaminoftuorene, 2-acetylaminophenanthrene, and 4-acetylaminostilbene (AABP, AAF, AAP, and AAS, respectively) to which they were exposed. Text-figure 2 compares these strains for the percent survival of their cloning capacity as a function of the dose of the four esters. No shoulder was detected in the survival curves after exposure to these carcinogens, for each of which the ratio between the slope of the curve for the particular XP strains and that measured for the NF was similar. The slope of the curve for the XP4BE variant was 1.2- to 1.7-fold steeper than that of the normal cells. The curve for XP2BE was 1.8- to 2.3-
UV DOSE (IrVl/mml ) 120
100
140
I.-Percent survival of cloning ability of strains of human skin fibroblasts with different DNA repair capabilities. Survival of cloning ability of irradiated cells divided by cloning efficiency of untreated control cultures is expressed in percent. Cloning efficiencies for controls ranged from 15 to 35% for normal celis, 10 to 25% for the XP2BE, XP13BE, and XP4BE cells, and 5 to 15% for XPI2BE. Exponential portion of these survival curves was drawn by the least squares method. Each symbol represents survival of cloning efficiency averaged from a series of 8-12 replicate dishes of cells receiving the specified dose of UV. (See "Materials and Methods.") Variation in number of colonies found per point was smaller than size of symbol used.
TEXT' FIGURE
> ::i iii
• • • •
NF NF XP XP o XP 4 XP
I-
CELLULAR REP AIR ENZYMES can modulate the damage resulting from lesions induced in the DNA of cells by physical or chemical agents. For example, bacterial cells deficient in the ability to effect excision repair of pyrimidine dimers or postreplication repair of lesions resulting from dimers exhibit greatly increased susceptibility to killing by UV irradiation (1). Using an in vitro bacterial transforming DNA system, we have presented evidence that such cellular DNA repair capabilities operate not only on UV damage but also on le-
sions in DNA caused by covalent attachment of such carcinogenic chemicals as aromatic amide derivatives (2-4) or polycyclic hydrocarbons (5, 6). Evidence that human cells possess similar genetically determined repair capability has been obtained by comparison of the cloning capacity of cultured cells from xeroderma pigmentosum patients [(XP) inherited susceptibility to multiple carcinomas of the skin on exposed areas] with that of normal human fibroblasts (NF) after exposure to UV (7) and to derivatives of the carcinogen 4nitroquinoline-I-oxide (8, 9), as well as to benz[a]anthracene 5,6-oxide (10). Accompanying autoradiography studies provided physical evidence that normal human cells carryon more extensive unscheduled DNA synthesis than do XP cells after exposure to the above agents (8) and to N-acetoxy2-acetylaminoftuorene (N-AcO-AAF) and N-hydroxy2-acetylaminoftuorene (N-HO-AAF) (11). Further physical evidence of differential amounts of DNA repair by such cells after attack by N-AcO-AAF has been obtained by Setlow and Regan (12, 13) with their 313-nm radiation photolysis technique. The present studies were undertaken to compare 1) the cytotoxic effect on the cloning capacity of normal human cells in culture of four series of closely related active compounds from a major class of chemical carcinogens (the aromatic amides) , 2) the cytotoxicity of such carcinogens in strains of human fibroblasts differing in their sensitivity to UV irradiation and in their ability to repair UV-induced lesions in DNA, and 3) the ability of such strains to repair lesions induced in their DNA by these carcinogens by cesium chloride (CsCI) equilibrium-centrifugation techniques to detect incorporation of thymidine into parental DNA strands. 1 Received September 13, 1974; revised January 23, 1975; accepted February 21, 1975. 2 Supported by Public Health Service grants CAI3058 and CA14680 from the National Cancer Institute, grant CP33226-01 from the Division of Cancer Cause and Prevention of the National Cancer Institute, and by an institutional grant to the Michigan Cancer Foundation by the United Foundation of Greater Detroit. 3 Division of Biological Sciences, Michigan Cancer Foundation. 110 E. Warren Ave., Detroit, Mich. 48201. ! We thank Dr. Kerstin Philips, Chemistry Department, Michigan Cancer Foundation, for her valuable contribution in establishing the purity of the test compounds, and James Wessel, Susan Van Hoeck, and Marilyn Mittlestat for excellent technical assistance.
JOURNAL OF THE NATIONAL CANCER INSTITUTE, VOL. 54, NO.6, JUNE 1975
Downloaded from https://academic.oup.com/jnci/article-abstract/54/6/1287/920864 by University of California, Santa Barbara user on 04 March 2018
1287
1288
MAHER, BIRCH, OTTO, AND MCCORMICK
MATERIALS AND METHODS
Carcinogenic compounds.-The ester derivatives of the aromatic ami des tested were supplied by Dr. John Scribner, Fred Hutchinson Cancer Research Center, Seattle, Washington. The other compounds were prepared or obtained as described in (2, 4) . All were stored in a desiccator at - 20° C until use. The melting point of each chemical was identical to that in our previous experiments (2, 4). Except for the sulfate ester derivatives that were dissolved in sterile distilled water, stock solutions of each aromatic compound were prepared in 100% ethanol (EtOH) immediately before use; final concentration of EtOH in the medium during e:,"posure to the cells was less than O.S% The solutIOns were protected from light at all times. Cell cultures.-Stocks of NF derived from foreskins were established, cultured, and stored until needed in liquid nitrogen as described in (14-17). One normal strain from a IS-year old male was obtained from the American Type Culture Collection (Rockville, Md.). Fibroblasts, derived from skin biopsy specimens of XP patients (XP2BE, XP4BE, XPI2BE, and XPI3BE), also from the American Type Culture Collection, were similarly cultured. All cells were from stocks in passages 6-18 and were routiney shown to be free from mycoplasma according to the microbiologic culture techniques and polyacrylamide gel electrophoresis of labeled RNA (18). Culture medium.-Cells were cultured in a humid atmosphere of S% CO 2 and air at 37° C in Ham's F-lO medium (19) (Gibco, Grand Island, N.Y.), supplemented with IS% fetal calf serum (FCS) (BioQuest, Cockeysville, Md.) plus antibiotics (100 U penicillin/ml and 100 I'g streptomycin/ml). Assay for cytotoxicity of carcinogens.-We measured the cyto.toxicity of the compounds by plating an approprIate number of cells per dish (300-3,000), allowing 10-12 hours for cell attachment, replacing the culture medium with serumfree F-IO medium, and by introducing the test c?mp~unds .at speci~ed concentrations. All operatIOns mvolvmg carcmogen solutions were done in semidarkness. Microscopic examination demonstrated that, under the conditions used, less than 10% of the cells plated could be expected to divide before they were exposed to the compounds. Th~ ~xposure period (4 hr) was terminated by the addItIOn of fresh culture medium containing IS% FCS. The. I?edium should inactivate any remaining electrophilIc reactants (carcinogens), since proteins are strong nucleophiles. The cells were refed with fresh culture medium three times weekly until clones developed to macroscopic size. At this time they were fixed in methanol, stained with 2% methylene blue, and counted. The average number of clones per dish (10-12 dishes) divided by the number of cells inoculated yielded the cloning effic~ency for a particular dose. The cloning efficIency of the cells treated with a particular dose
divided by the cloning efficiency of the control cells, which received solvent only, determined the cytotoxicity of the compound and was expressed as percent. No cytotoxic effect was caused by the concentration of the solvent used. Exposure to UV irradiation.-After time was allowed for cell attachment, the culture medium was removed, and the cells were rinsed with 0.9% saline and exposed in semidarkness to the specified doses of irradiation with a shortwave UVS-12 lamp (Ultra-Violet Products, San Gabriel, Calif.) havmg an output of 2-4 ergs/mm 2/second. UV irradiation was measured with a J22S metering unit, also from U~tra-Violet Products. Cells were protected from lIght for 48-72 hours from the time of irradiation to prevent possible photoreactivation. Detection of repair of DNA.-Cell cultures of normal or XP fibroblasts in 2S0-ml flasks were prel.abeled with HC-thymidine (O.S ,1.tCi/ml; S6 mCI/mmole; Nuclear Dynamics, El Monte, Calif.) in culture medium for 20 hours. Unlabeled culture medium containing S-bromodeoxyuridine (BUDR, 10 I'g/ml; P. L. Biochemicals Inc., Milwaukee Wis) was supplied 1 hour before exposure to th~ carcmogens to label the density of the most recently replicated DNA fragments. .Cells "":ere exposed to N-AcO-AAF (SO I'M) 60 mmutes m the dark in serum-free F-IO medium plus B~DR (10 ~g/ml). Hydroxyurea (HU, 20 mM; SIgm.a ChemIcal Co., St. Louis, Mo.) was ~dd~d dUrIng the last 20 minutes to stop DNA replIcatIOn. After 60 minutes of carcinogen treatment, the medium was removed and the cells were given freshly prepared culture medium containing BUDR (10 t.tg/ml), HU (20 mM), and tritiated thymidine (3~-TDR, 10 t.tCi/ml, 40 Ci/mmole; Nuclear DynamICS) for 4 hours to label DNA during repair synthesis before the cells were harvested. If residual DNA synthesis were to occur despite the HU block, 3~-TDR incorporated during normal DNA syntheSIS rather than by repair synthesis would be d~tected in the region of higher density on the gradIent because of the simultaneous incorporation of BUDR. DNA extraction.-Cells were rinsed with saline, scraped from flasks, suspended in O.IS M NaCl, O'0.1S M sodium citrate (SSC) , pH 7.2, washed t~lCe, resuspended in SSC, and lysed with 0.2% sodIUm lauryl sulfate. PaI?-creatic (SO t.tg/ml) and Tl RNase (SO U /ml; CalbIOchem, Los Angeles, Calif.) was added for 30 minutes at 37° C, followed by nuclease-free Pronase (100 I'g/ml; Calbiochem) for 1 hour and deproteinization with chloroform: butanol (4: 1) was effected, after which the solution underwent dialysis against 2 liters of SSC for 2 hours. AI~aline CsCI centrifugation.-Alkaline CsCI gradIents w.ere prepared and centrifuged at 34,000 rpm accordmg to the methods of McCormick et al. (20). Fractions were collected, and the doublelabel radioactivity was determined in a spectrometer
Downloaded from https://academic.oup.com/jnci/article-abstract/54/6/1287/920864 by University of California, Santa Barbara user on 04 March 2018
EFFECT OF REPAIR ON CYTOTOXICITY OF CARCINOGENS
(Packard Tricarb, Packard Instrument Co., LaGrange, Ill.) as described in (20). The region of normal density (unreplicated parental) DNA was located from the 14C-TDR peak and the amount of 3H-TDR incorporated under this peak determined. These fractions were pooled and rebanded if necessary to insure that the tritium counts did not represent physically trapped material. RESULTS Differential Sensitivity to UV Irradiation
We compared the sensitivity of several strains of human cells in culture to UV light by measuring the survival of their cloning capacity. The cells thus compared represented several strains of normal diploid fibroblasts started from foreskin cultures, one normal strain started from a biopsy specimen of a 15-year-old, and strains of cells cultured from biopsy material from XP patients selected on the basis of the different ability of their lymphocytes and fibroblasts to incorporate thymidine in repair of DNA after UV irradiation (21). Thus XPl2BE from genetic complementation group A incorporated less than 2% of the amount of 3HTDR that normal cells did during the first 3 hours after exposure to UV irradiation (21); XP2BE, from group C, incoq:orated 15-25% of normal (22), and XP4BE 100% of normal (22). The latter XP patient, whose excision repair rate appeared normal, was designated an XP variant (7). Text-figure I compares the percent survival of the cloning capacity (as a function of UV irradiation) of the three XP strains with that of NF. The repairing NF exhibited a shoulder on the UV survival curve, which suggested DNA repair. A smaller shoulder was seen for the XP4BE cells, but the exponential portion of its survival curve had a slope only approximately 1.3-fold steeper than that of the normal cells. The slope of the XP2BE cells was 1.9 times steeper than that of normal cells, but because it lacked a shoulder, it required only 8
1289
ergsjmm 2 to reduce the survival level to 37% compared with approximately 48 ergsjmm 2 for the NF. Strain XPl2BE was more sensitive to the killing effect of UV irradiation. It had no shoulder and the slope of its survival curve was 5.8 times steeper than that of the linear portion of the curve for normal cells. A dose of only approximately 3 ergsjmm 2 brought these cells to a survival level of 37% of untreated controls. The survival of the cloning ability, as a function of UV, of NF taken from a 15-year-old boy was also determined to see if the difference between XP4BE and the NF could be attributed to age of the donor (XP4BE came from a 27-year-old male XP patient). There was no difference in sensitivity to UV between the normal strains taken from newborns and that of the 15-year-old. A second XP variant, XPI3BE, was then tested for survival of its cloning ability as a function of UV; this strain exhibited survival levels identical with those of XP4BE. Differential Cytotoxicity of Active Derivatives of Carcinogens
The survival of the cloning ability of the three XP strains, compared with that of normal fibroblasts, was then determined as a function of the concentration of the active N·acetoxy ester derivatives of four aromatic amide carcinogens, 4-acetylaminobiphenyl, 2-acetylaminoftuorene, 2-acetylaminophenanthrene, and 4-acetylaminostilbene (AABP, AAF, AAP, and AAS, respectively) to which they were exposed. Text-figure 2 compares these strains for the percent survival of their cloning capacity as a function of the dose of the four esters. No shoulder was detected in the survival curves after exposure to these carcinogens, for each of which the ratio between the slope of the curve for the particular XP strains and that measured for the NF was similar. The slope of the curve for the XP4BE variant was 1.2- to 1.7-fold steeper than that of the normal cells. The curve for XP2BE was 1.8- to 2.3-
UV DOSE (IrVl/mml ) 120
100
140
I.-Percent survival of cloning ability of strains of human skin fibroblasts with different DNA repair capabilities. Survival of cloning ability of irradiated cells divided by cloning efficiency of untreated control cultures is expressed in percent. Cloning efficiencies for controls ranged from 15 to 35% for normal celis, 10 to 25% for the XP2BE, XP13BE, and XP4BE cells, and 5 to 15% for XPI2BE. Exponential portion of these survival curves was drawn by the least squares method. Each symbol represents survival of cloning efficiency averaged from a series of 8-12 replicate dishes of cells receiving the specified dose of UV. (See "Materials and Methods.") Variation in number of colonies found per point was smaller than size of symbol used.
TEXT' FIGURE
> ::i iii
• • • •
NF NF XP XP o XP 4 XP
I-
