Mutation Research, 269 (1992) 73-78 © 1992 Elsevier Science Publishers B.V. All rights reserved 0027-5107/92/$05.00
73
MUT 05135
Comparative toxicity and mutagenicity of N-hydroxy-2-acetylaminofluorene and 7-acetyl-N-hydroxy-2-acetylaminofluorene in human lymphoblasts * John R. Babson a, Nancy E. Gavitt a, Lakmai W. Boteju b and Patrick E. Hanna b,c a Department of Pharmacology and Toxicology, College of Pharmacy, Unicersity of Rhode Island, Kingston, RI 02881, USA and b Departments of Medicinal Chemistry and c Pharmacology, Unicersity of Minnesota, Minneapolis, MN 55455, USA
(Received 14 November 1991 ) (Accepted 24 February 1992)
Keywords: N-Hydroxy-2-acetylaminofluorene; 7.AcetyI-N-hydro~-2-acetylaminofluorene; N-Arylh~,droxamic acid N,O-acyltransferase
Lymphoblasts, human:
Summary Exponentially growing TK6 human lymphoblasts were exposed to either 0-50 /~M N-hydroxy-2acetylaminofluorene (N-OH-AAF) or 0-10 #M 7-acetyl-N.hydroxy-2-acetylaminofluorene (7-acetyI-NOH-AAF) in both the absence and presence of a partially purified preparation of hamster-liver N-arylhydroxamic acid N,O-acyltransferase (AHAT). Neither N-aryihydroxamic acid was toxic to the lymphoblasts, nor mutagenic at the thymidine kinase (tk) locus, in the absence of AHAT over the concentration range examined. In the presence of AHAT, an enzyme that activates N-arylhydroxamic acids to electrophilic N-acetoxyarylamine intermediates, both compounds caused toxicity and mutagenicity in TK6 cells. The 7.acetyl-N-OH-AAF was approximately 10.fold more toxic and mutagenic than the unsubstituted N-OH-AAF. These data demonstrate that metabolism of these N-arylhydroxamic acids, presumably to N-acetoxyarylamine intermediates by AHAT, is a key event in the biological activity of these agents. In addition, the presence of electron-withdrawing 7-acetyl substituent that is thought to stabilize N-acetoxy intermediates, appears to enhance the biological activity of the unsubstituted N-OH-AAF.
Correspondence: Dr. John R. Babson, Department of Pharmacology and Toxicology, College of Pharmacy, University of Rhode Island, Kingston, RI 02881, USA. Tel. 401-792-2363; Fax 401-792-2181. * This work was supported by grant BC-638 from the American Cancer Society to P.E.H. and J.R.B.
Abbreciations: N-OH-AAF, N-hydroxy-2-acetylaminofluorene; N-acetoxy-7-acetyI-AF. N-acetoxy-17-acetyl)-2-aminofluorene; 7-acetyI-N-OH-AAF, 7-acetyI-N-hydroxy-2acetylaminofluorene: 7-acetyI-N-OH-AF, N-hydroxy-(7acetyl)-2-aminofluorene" 2-AF, 2-aminofluorene: AHAT. Narylhydroxamic acid N,O-acyltransferase: CHAT. cytidine, hypoxanthine, aminopterim thymidine; DMSO, dimethylsulfoxide, Hepes, N.(2.hydroxyethyl)-l-piperazine-N'-(2-ethanesulfonic acid): THC, thymidine, hypoxanthine, cytidine; tk. thymidine kinase.
74
N-Hydroxy-2-acetylaminofluorene (N-OHAAF) is a proximate carcinogenic metabolite of 2-AF, a well studied member of an important class of carcinogenic arylamides. The bioactivation of arylamides is a complex, multistep process that initially involves the formation of proximate carcinogenic forms, the N-arylhydroxamic acids. These proximate carcinogens can be metabolized further to N-acetoxy esters, reactive electrophilic intermediates that are considered to be ultimate carcinogenic forms of the arylamides (Beland and Kadlubar, 1985; Hanna and Banks, 1985; Miller and Miller, 1985). Several enzyme activities can contribute to the conversion of N-arylhydroxamie acids to reactive N-acetoxy esters, including Ndeacylation followed by O-aeetylation of the resuiting N-hydroxyarylamine, and N,O-transaeetylation. The latter is an interesting process (Fig. 1) that is catalyzed by the en~me AHAT (Hanna and Banks, 1985). While the relative importance of O-aeetylation and N-deacylation activities to the mutagenic potency of N-OH-AAF has been studied in both bacterial (Stout et al., 1976; Schut et al., 1978; Beranek et al., 1982; McCoy et al., 1982, 1983; Saito et al., 1985) and mammalian (Heflich et al., 1988) assay systems, less is known about the significance of AHAT. The presence of endogenous activation and deactivation pathways complicates the use of bacterial assay systems to examine the significance of AHAT in the mutagenie activity of N-arylhydroxamic acids. How. ever, important information concerning the role of N-acetoxyfluorene intermediates has come from structure-activity studies of analogous compounds, the nitrofluorenes, in Salmonella systems
R
~
OH NICOCHs
Fig. 2. Structures of N-OH-AAF (R = H) and 7-acetyI-N-OHAAF (R = CH3OC-).
(Vance et al., 1985, 1987). These studies indicate that electron-withdrawing groups enhance the mutagenicity of compounds that may ultimately yield N-acetoxy intermediates similar to that derived from N-OH-AAF. In the current study, the role of AHAT activity in the toxicity and mutagenicity of two Narylhydroxamic acids, N-OH-AAF and 7-acetylN-OH-AAF (Fig. 2) was examined in a human cell line, TK6 lymphoblasts. A specific objective was to determine the effect of the electronwithdrawing 7-acetyl substituent on the biological activity of N-OH-AAF in the human cells. The TK6 line is a" thymidine kinase heterozygote derivative of human lymphoblast line and has proven effective for studying the toxicity and mutagenicity of a wide variety chemical agents (Skopek et al., 1978; Liber and Thilly, 1982). Results of this study indicate that N,O-transaeetylation, as catalyzed by AHAT, can contribute to the biological activity of N-arylhydroxamic acids in human lymphoblasts and that the presence of an electron-withdrawing, 7-acetyl moiety enhances both the toxicity and mutagenicity of N-OH-AAF. Materials and methods
OH I Ar-N-C-CH3 U o
•
OH AHAT I ~ Ar-N-H
2
O II + CH3-C-enzyme
_L..] reaction with biological nucleophiles
Fig. I. Bioactivation of N-arylhydroxamic acids via A H A T catalyzed N,O-transacetylation.
All chemicals and reagents used were the highest quality available. 4-Aminoazobenzene was obtained from Eastman Kodak Co. Culture media and all other reagents were purchased from Sigma Chemical Co. The horse serum was obtained from Hyclone Laboratories, the Primaria 96 well tissue culture plates and Falcon T-flasks from Becton/Diekinson Labware Co, and the male golden Syrian hamsters from Charles River Breeding Labs. The TK6 human lymphoblasts were the generous gift of the Gentest Corporation, Woburn, MA. N-OH-AAF and 7-acetyl-N-
75
O H - A A F were prepared as described previously (Marhevka et al., 1985). These compounds were characterized by NMR, MS, elemental analysis and TLC and found to contain no detectible impurities.
AlIA T preparation and assay AHAT from livers of 70-90 g golden Syrian hamsters was purified 2-3-fold from hamster-liver cytosoi by ammonium sulfate fractionation according to the method of King (1974). The final enzyme pellet was resuspended in 50% ammonium sulfate buffered with 50 mM pyrophosphate and containing 1 mM dithiothreitol, aliquots were distributed in 1.8-ml eppendorf centrifuge tubes and were centrifuged at 13 000 × g for 10 rain at 4°C. The supernatant was removed and the pellets were stored at -80°C. Just prior to use, frozen aliquots were thawed, resuspended in RPMI medium and filter-sterilized. It was determined that filter sterilizing neither reduced the amount of dissolved protein nor decreased the enzyme activity. The A H A T assay used N-OHAAF as a substrate and the transacetylation of 4-aminoazobenzene was measured spectrophotometrically by the procedure of Booth (1966). General cell culture methods TK6 cells were grown in RPMI 1640 medium supplemented with 10% horse serum (heattreated at 50°C for 2 h), 100 U / m l of penicillin, 100/zg/ml of streptomycin, and buffered with 24 mM sodium bicarbonate and 5% CO 2 at 37°C and 95% humidity. Cultures were maintained as suspension cultures with stock cultures grown in stirred culture and experimental cultures grown in unstirred culture. Prior to arylhydroxamic acid exposure, cells were treated with CHAT medium for two days to reduce background mutation fraction (Liber and Thilly, 1982). CHAT medium consisted of RPMI 1640 supplemented with 10 -5 M cytidine, 2 × 10 - 4 M hypoxanthine, 2 x 10 - 7 M aminopterin and 1.75 × 10 -5 M thymidine. Following CHAT exposure, cells were centrifuged, resuspended in THC medium, and diluted on subsequent days with regular medium. Cells were routinely exposed to mutagens within 4 - 5 days following CHAT treatment.
Conditions for exposure to arylhydroxamic acids The conditions for mutagen treatment and measurement of relative survival and mutagenicity at the tk locus were essentially those described by Liber and Thilly (1982) with the following modifications. Lymphoblasts were suspended to a density of 5.0 × 105 cells/ml and exposed to either various concentrations of arylhydroxamic acids dissolved in ethanol/DMSO (1:1) or just ethanol/DMSO (1% v/v). For mutagen exposure in the presence of AHAT, an equal amount of enzyme activity was added to each culture to a final concentration of 0.5 mg/ml. Cells were then exposed to each treatment in static culture in 25-cm 2 T-flasks at 37°C for 90 min. After mutagen exposure, cell suspensions were centrifuged at 100 × g for 5 min and resuspended in fresh medium. This step was repeated twice and the cells resuspended in medium, counted and a portion of the cells plated in 96-well tissue culture plates at 2-200 cells/well to determine relative survival. The remaining cells were maintained by daily dilution to 4 × 105 cells/ml in static culture until cultures returned to exponential growth. At that time cells from each culture were plated at 30000 cells/well in the presence of 2 /.tg/ml trifluorothymidine to determine mutation frequency at the tk locus and at 2 cells/well in the absence of trifluorothymidine to determine plating efficiency. Following incubation at 37°C, 95% humidity and 5% CO 2 plates were scored for colonies under 8 × magnification. The mutant fraction was calculated according to the procedure described (Liber and Thilly, 1982). At the end of the mutation assay, mutants were cloned, grown in the presence of trifluorothymidine for several days, then grown under nonselective conditions. The growth rates of mutant cells and nonmutant cells were then determined and found to be identical, thus indicating that mutants had no growth advantage over nonmutant TK6 cells under conditions used in this study. Results and discussion AHAT is the enzyme believed responsible for the metabolic conversion of N-arylhydroxamic acids to reactive N-acyloxyarylamines (Hanna and Banks, 1985). Results of the current study
76
demonstrate that AHAT catalyzed N,O-transacetylation can contribute to the biological activity of the arylhydroxamic acids, N-OH-AAF and 7-acetyl-N-OH-AAF in TK6 human lyrnphoblast cultures. The relative survival of TK6 cells was not diminished when cultures were exposed to either N-OH-AAF (0-50 p,M) or 7-acetyl-N-OH-AAF (0-10 t~M) in the absence of AHAT (Fig. 3). However, exposure to the same concentration of these arylhydroxamic acids in the presence of AHAT caused substantial toxicity. The mutagenic potency of these N-arylhydroxamic acids also was dependent on the presence of AHAT activity. The mutant fractions of cultures exposed to either of these N-arylhydroxamic acids in the absence of AHAT as shown in Fig. 4 were no different than those of controls exposed to solvent only. However, the presence of AHAT activity during N-arylhydroxamic acid exposure caused significant mutagenicity at the tk locus. Since added AHAT was required to induce both toxicity and mutagenicity, it was concluded that the lymphoblasts contain no enzyme activity capable of bioactivating N-arylhydroxamic acids. Indeed, no AHAT activity was detectable when lymphoblasts were lyscd and AHAT activity was measured.
10 20 C O N C E N T R A T I O N (p.M)
50
Fig. 3. Toxicity of N-OH.AAF and 7-acetyI-N-OH-AAF toward TK6 human lymphoblasts. Exponentially growing TK6 cells were'exposed for 90 rain to various concentrations of N-OH,AAF (o, o) or 7-acetyI-N-OH-AAF ( l l , n ) in the absence (open symbols) and the presence (closed symbols) of AHAT. Cells were then washed to remove enzyme and arylhydroxamic acid, plated in 96~well tissue-culture plates and the relative survival determined by colony-forming ability. Each point represents the mean + the standard deviation of at least three separate experiments.
z
6o
[.. .
73
MUT 05135
Comparative toxicity and mutagenicity of N-hydroxy-2-acetylaminofluorene and 7-acetyl-N-hydroxy-2-acetylaminofluorene in human lymphoblasts * John R. Babson a, Nancy E. Gavitt a, Lakmai W. Boteju b and Patrick E. Hanna b,c a Department of Pharmacology and Toxicology, College of Pharmacy, Unicersity of Rhode Island, Kingston, RI 02881, USA and b Departments of Medicinal Chemistry and c Pharmacology, Unicersity of Minnesota, Minneapolis, MN 55455, USA
(Received 14 November 1991 ) (Accepted 24 February 1992)
Keywords: N-Hydroxy-2-acetylaminofluorene; 7.AcetyI-N-hydro~-2-acetylaminofluorene; N-Arylh~,droxamic acid N,O-acyltransferase
Lymphoblasts, human:
Summary Exponentially growing TK6 human lymphoblasts were exposed to either 0-50 /~M N-hydroxy-2acetylaminofluorene (N-OH-AAF) or 0-10 #M 7-acetyl-N.hydroxy-2-acetylaminofluorene (7-acetyI-NOH-AAF) in both the absence and presence of a partially purified preparation of hamster-liver N-arylhydroxamic acid N,O-acyltransferase (AHAT). Neither N-aryihydroxamic acid was toxic to the lymphoblasts, nor mutagenic at the thymidine kinase (tk) locus, in the absence of AHAT over the concentration range examined. In the presence of AHAT, an enzyme that activates N-arylhydroxamic acids to electrophilic N-acetoxyarylamine intermediates, both compounds caused toxicity and mutagenicity in TK6 cells. The 7.acetyl-N-OH-AAF was approximately 10.fold more toxic and mutagenic than the unsubstituted N-OH-AAF. These data demonstrate that metabolism of these N-arylhydroxamic acids, presumably to N-acetoxyarylamine intermediates by AHAT, is a key event in the biological activity of these agents. In addition, the presence of electron-withdrawing 7-acetyl substituent that is thought to stabilize N-acetoxy intermediates, appears to enhance the biological activity of the unsubstituted N-OH-AAF.
Correspondence: Dr. John R. Babson, Department of Pharmacology and Toxicology, College of Pharmacy, University of Rhode Island, Kingston, RI 02881, USA. Tel. 401-792-2363; Fax 401-792-2181. * This work was supported by grant BC-638 from the American Cancer Society to P.E.H. and J.R.B.
Abbreciations: N-OH-AAF, N-hydroxy-2-acetylaminofluorene; N-acetoxy-7-acetyI-AF. N-acetoxy-17-acetyl)-2-aminofluorene; 7-acetyI-N-OH-AAF, 7-acetyI-N-hydroxy-2acetylaminofluorene: 7-acetyI-N-OH-AF, N-hydroxy-(7acetyl)-2-aminofluorene" 2-AF, 2-aminofluorene: AHAT. Narylhydroxamic acid N,O-acyltransferase: CHAT. cytidine, hypoxanthine, aminopterim thymidine; DMSO, dimethylsulfoxide, Hepes, N.(2.hydroxyethyl)-l-piperazine-N'-(2-ethanesulfonic acid): THC, thymidine, hypoxanthine, cytidine; tk. thymidine kinase.
74
N-Hydroxy-2-acetylaminofluorene (N-OHAAF) is a proximate carcinogenic metabolite of 2-AF, a well studied member of an important class of carcinogenic arylamides. The bioactivation of arylamides is a complex, multistep process that initially involves the formation of proximate carcinogenic forms, the N-arylhydroxamic acids. These proximate carcinogens can be metabolized further to N-acetoxy esters, reactive electrophilic intermediates that are considered to be ultimate carcinogenic forms of the arylamides (Beland and Kadlubar, 1985; Hanna and Banks, 1985; Miller and Miller, 1985). Several enzyme activities can contribute to the conversion of N-arylhydroxamie acids to reactive N-acetoxy esters, including Ndeacylation followed by O-aeetylation of the resuiting N-hydroxyarylamine, and N,O-transaeetylation. The latter is an interesting process (Fig. 1) that is catalyzed by the en~me AHAT (Hanna and Banks, 1985). While the relative importance of O-aeetylation and N-deacylation activities to the mutagenic potency of N-OH-AAF has been studied in both bacterial (Stout et al., 1976; Schut et al., 1978; Beranek et al., 1982; McCoy et al., 1982, 1983; Saito et al., 1985) and mammalian (Heflich et al., 1988) assay systems, less is known about the significance of AHAT. The presence of endogenous activation and deactivation pathways complicates the use of bacterial assay systems to examine the significance of AHAT in the mutagenie activity of N-arylhydroxamic acids. How. ever, important information concerning the role of N-acetoxyfluorene intermediates has come from structure-activity studies of analogous compounds, the nitrofluorenes, in Salmonella systems
R
~
OH NICOCHs
Fig. 2. Structures of N-OH-AAF (R = H) and 7-acetyI-N-OHAAF (R = CH3OC-).
(Vance et al., 1985, 1987). These studies indicate that electron-withdrawing groups enhance the mutagenicity of compounds that may ultimately yield N-acetoxy intermediates similar to that derived from N-OH-AAF. In the current study, the role of AHAT activity in the toxicity and mutagenicity of two Narylhydroxamic acids, N-OH-AAF and 7-acetylN-OH-AAF (Fig. 2) was examined in a human cell line, TK6 lymphoblasts. A specific objective was to determine the effect of the electronwithdrawing 7-acetyl substituent on the biological activity of N-OH-AAF in the human cells. The TK6 line is a" thymidine kinase heterozygote derivative of human lymphoblast line and has proven effective for studying the toxicity and mutagenicity of a wide variety chemical agents (Skopek et al., 1978; Liber and Thilly, 1982). Results of this study indicate that N,O-transaeetylation, as catalyzed by AHAT, can contribute to the biological activity of N-arylhydroxamic acids in human lymphoblasts and that the presence of an electron-withdrawing, 7-acetyl moiety enhances both the toxicity and mutagenicity of N-OH-AAF. Materials and methods
OH I Ar-N-C-CH3 U o
•
OH AHAT I ~ Ar-N-H
2
O II + CH3-C-enzyme
_L..] reaction with biological nucleophiles
Fig. I. Bioactivation of N-arylhydroxamic acids via A H A T catalyzed N,O-transacetylation.
All chemicals and reagents used were the highest quality available. 4-Aminoazobenzene was obtained from Eastman Kodak Co. Culture media and all other reagents were purchased from Sigma Chemical Co. The horse serum was obtained from Hyclone Laboratories, the Primaria 96 well tissue culture plates and Falcon T-flasks from Becton/Diekinson Labware Co, and the male golden Syrian hamsters from Charles River Breeding Labs. The TK6 human lymphoblasts were the generous gift of the Gentest Corporation, Woburn, MA. N-OH-AAF and 7-acetyl-N-
75
O H - A A F were prepared as described previously (Marhevka et al., 1985). These compounds were characterized by NMR, MS, elemental analysis and TLC and found to contain no detectible impurities.
AlIA T preparation and assay AHAT from livers of 70-90 g golden Syrian hamsters was purified 2-3-fold from hamster-liver cytosoi by ammonium sulfate fractionation according to the method of King (1974). The final enzyme pellet was resuspended in 50% ammonium sulfate buffered with 50 mM pyrophosphate and containing 1 mM dithiothreitol, aliquots were distributed in 1.8-ml eppendorf centrifuge tubes and were centrifuged at 13 000 × g for 10 rain at 4°C. The supernatant was removed and the pellets were stored at -80°C. Just prior to use, frozen aliquots were thawed, resuspended in RPMI medium and filter-sterilized. It was determined that filter sterilizing neither reduced the amount of dissolved protein nor decreased the enzyme activity. The A H A T assay used N-OHAAF as a substrate and the transacetylation of 4-aminoazobenzene was measured spectrophotometrically by the procedure of Booth (1966). General cell culture methods TK6 cells were grown in RPMI 1640 medium supplemented with 10% horse serum (heattreated at 50°C for 2 h), 100 U / m l of penicillin, 100/zg/ml of streptomycin, and buffered with 24 mM sodium bicarbonate and 5% CO 2 at 37°C and 95% humidity. Cultures were maintained as suspension cultures with stock cultures grown in stirred culture and experimental cultures grown in unstirred culture. Prior to arylhydroxamic acid exposure, cells were treated with CHAT medium for two days to reduce background mutation fraction (Liber and Thilly, 1982). CHAT medium consisted of RPMI 1640 supplemented with 10 -5 M cytidine, 2 × 10 - 4 M hypoxanthine, 2 x 10 - 7 M aminopterin and 1.75 × 10 -5 M thymidine. Following CHAT exposure, cells were centrifuged, resuspended in THC medium, and diluted on subsequent days with regular medium. Cells were routinely exposed to mutagens within 4 - 5 days following CHAT treatment.
Conditions for exposure to arylhydroxamic acids The conditions for mutagen treatment and measurement of relative survival and mutagenicity at the tk locus were essentially those described by Liber and Thilly (1982) with the following modifications. Lymphoblasts were suspended to a density of 5.0 × 105 cells/ml and exposed to either various concentrations of arylhydroxamic acids dissolved in ethanol/DMSO (1:1) or just ethanol/DMSO (1% v/v). For mutagen exposure in the presence of AHAT, an equal amount of enzyme activity was added to each culture to a final concentration of 0.5 mg/ml. Cells were then exposed to each treatment in static culture in 25-cm 2 T-flasks at 37°C for 90 min. After mutagen exposure, cell suspensions were centrifuged at 100 × g for 5 min and resuspended in fresh medium. This step was repeated twice and the cells resuspended in medium, counted and a portion of the cells plated in 96-well tissue culture plates at 2-200 cells/well to determine relative survival. The remaining cells were maintained by daily dilution to 4 × 105 cells/ml in static culture until cultures returned to exponential growth. At that time cells from each culture were plated at 30000 cells/well in the presence of 2 /.tg/ml trifluorothymidine to determine mutation frequency at the tk locus and at 2 cells/well in the absence of trifluorothymidine to determine plating efficiency. Following incubation at 37°C, 95% humidity and 5% CO 2 plates were scored for colonies under 8 × magnification. The mutant fraction was calculated according to the procedure described (Liber and Thilly, 1982). At the end of the mutation assay, mutants were cloned, grown in the presence of trifluorothymidine for several days, then grown under nonselective conditions. The growth rates of mutant cells and nonmutant cells were then determined and found to be identical, thus indicating that mutants had no growth advantage over nonmutant TK6 cells under conditions used in this study. Results and discussion AHAT is the enzyme believed responsible for the metabolic conversion of N-arylhydroxamic acids to reactive N-acyloxyarylamines (Hanna and Banks, 1985). Results of the current study
76
demonstrate that AHAT catalyzed N,O-transacetylation can contribute to the biological activity of the arylhydroxamic acids, N-OH-AAF and 7-acetyl-N-OH-AAF in TK6 human lyrnphoblast cultures. The relative survival of TK6 cells was not diminished when cultures were exposed to either N-OH-AAF (0-50 p,M) or 7-acetyl-N-OH-AAF (0-10 t~M) in the absence of AHAT (Fig. 3). However, exposure to the same concentration of these arylhydroxamic acids in the presence of AHAT caused substantial toxicity. The mutagenic potency of these N-arylhydroxamic acids also was dependent on the presence of AHAT activity. The mutant fractions of cultures exposed to either of these N-arylhydroxamic acids in the absence of AHAT as shown in Fig. 4 were no different than those of controls exposed to solvent only. However, the presence of AHAT activity during N-arylhydroxamic acid exposure caused significant mutagenicity at the tk locus. Since added AHAT was required to induce both toxicity and mutagenicity, it was concluded that the lymphoblasts contain no enzyme activity capable of bioactivating N-arylhydroxamic acids. Indeed, no AHAT activity was detectable when lymphoblasts were lyscd and AHAT activity was measured.
10 20 C O N C E N T R A T I O N (p.M)
50
Fig. 3. Toxicity of N-OH.AAF and 7-acetyI-N-OH-AAF toward TK6 human lymphoblasts. Exponentially growing TK6 cells were'exposed for 90 rain to various concentrations of N-OH,AAF (o, o) or 7-acetyI-N-OH-AAF ( l l , n ) in the absence (open symbols) and the presence (closed symbols) of AHAT. Cells were then washed to remove enzyme and arylhydroxamic acid, plated in 96~well tissue-culture plates and the relative survival determined by colony-forming ability. Each point represents the mean + the standard deviation of at least three separate experiments.
z
6o
[.. .
