195
Mutation Research, 48 (1977) 195--204 © Elsevier/North-Holland Biomedical Press
EFFECT OF DNA REPAIR SYSTEMS ON ANTIBACTERIAL AND MUTAGENIC ACTIVITY OF AN ANTITUMOR PROTEIN, NEOCARZINOSTATIN
K. TATSUMI * and H. NISHIOKA
Department of Medicine, Kyoto University, Kyoto 606 and Laboratory of Biophysical Chemistry, Doshisha University, Kyoto 602 (Japan) (Received October 12th, 1976) (Accepted October 20th, 1976)
Summary Antibacterial and mutagenic effects of an antitumor protein, neocarzinostatin (NCS), on isogenic strains of Escherichia coli with normal or defective DNA repair systems were studied. Growth of the strains lacking recA gene was inhibited by NCS with much lower concentration than in the case of those possessing it, while this "differential inhibition for g r o w t h " {DIG) between the strains uvrA ÷ and uvrA- was not seen. NCS induces mutation in recA ÷ strains but not significantly in recA-, while no such difference of mutagenesis was noticed between the strains uvrA ÷ and uvrA-. These results suggest that NCS produces hardly excisable DNA damage which is repaired by an error-prone recombination process.
Introduction NCS, an antitumor antibiotic isolated from a culture filtrate of Streptomyces carzinostaticus Var F41 [11], is an acidic single-chain polypeptide of approx. 10,700 mol. wt. whose primary structure has been recently elucidated [16]. This antibiotic inhibits growth of several gram-positive organisms [ 11 ] and a number of transplantable solid tumors and leukemias in mice [5,11,15]. As to its clinical efficacy, a remission rate of 50% obtained in a group of patients with acute leukemia [10] and a regression observed in certain solid tumors [26] have been reported. Mode of action studies indicated that NCS inhibits DNA synthesis [20,21] and also causes degradation of existing DNA in Sarcina lutea [21]. It was also reported that NCS induces DNA breakages in L1210 cells [23,27] and HeLa-S3 * Present address: D e p a r t m e n t o f M i c r o b i o l o g y , the University o f Chicago, Chicago, Ill. 6 0 6 3 7 (U.S.A.).
196
cells [4,22] in vivo and in calf thymus DNA [27], E. coli DNA [12] and SV40 DNA [4] in vitro. The rejoining of the NCS-induced single-strand breaks observed in DNA of HeLa-S3 cells [22] and the unscheduled DNA synthesis detected in human lymphocytes treated with NCS [28] indicated the possibility that at least a part of DNA damage induced by this antibiotic is repaired. Thus the principal target for NCS-action appears to be cellular DNA, though its subcellular behavior is still unknown. Recently this protein antibiotic was found to produce repairable DNA damage and have mutagenic activity is bacterial systems [29]. In the present paper, we describe the antibacterial and mutagenic activities of NCS for isogenic strains of E. coli with normal or defective DNA repair systems, attempting to characterize further the type of DNA damage caused by NCS from the pattern of response of repair deficient strains. Materials and methods Chemicals NCS (Lot T-64) was the generous gift of Kayaku Antibiotics Research Co. Ltd., Tokyo, Japan. It was stored in the dark at --20°C and was dissolved in saline just before use. Mitomycin C (MMC) and kanamycin (KM) were provided from Kyowa Hakko Co. Ltd., Tokyo and Meiji Seika Co. Ltd., Tokyo, respectively. 4NQO, MNNG and EMS were purchased from Nakarai Chemicals Co. Ltd., Kyoto, Japan. Bacterial strains Four strains of E. coli possessing different DNA repairing capacities were used for DNA repair test and reversion test. They were (1) WP2 (trp-, uvrA ÷, recA ÷) derived from B/r by Witkin [30], (2) WP2uvrA (trp-, uvrA-, recA ÷) originally isolated by Hill [9], (3) CM571 (trp-, uvrA ÷, recA-)isolated from WP2 by Bridges [7] and (4) WP100 (trp-, uvrA-, recA-) constructed by Witkin [8,31]. These strains are nearly isogenic and have a tryptophan-deficiency that is suppressible by ochre suppressor mutations [6]. Culture media Nutrient broth: 8 g of Difco nutrient broth (dehydrated) and 0.5 g NaC1 were dissolved in 1000 ml of distilled water and sterilized; NB agar: 15 g (Difco) were added to 1000 ml of nutrient broth and solidified; M9 buffer: 1 g NH4C1, 15 g Na2HPO4 • 12H20, 3 g KH2PO4 and 0.5 g NaC1 were dissolved in 1000 ml distilled water, sterilized by autoclaving and then 3.3 ml of 0.5 M MgSO4 added; SEM agar [31] : 0.2% w/v glucose and 0.04% w/v nutrient broth as well as 15 g bacto-agar were added to M9 buffer. DNA repair test In order to detect DIG among the four strains of E. coli by NCS, the simple "streak test" and the super sensitive "test-tube test" [17] were adopted. "Streak test" The essential procedures were originally described by Kada et al. [13] as
197 "rec-assay" for screening of chemical mutagens using wild and recombination deficient strains of Bacillus subtilis. The four strains of E. coli were grown overnight in nutrient broth. Each culture was diluted with M9 buffer to obtain a b o u t 0.3 of an optical density at 660 nm, giving approx. 2.5 • 108 cells per ml. The cultures were then streaked on NB agar plate from a central point in different directions using 0.1-ml pipette severally. A drug solution in 0.1 ml was spotted onto a filter disc (Whatman GFC, 24 mm in diameter) which had been placed on the plate so as to cover the starting points of the four streaks. All the plates were then incubated for 24 h at 37°C and the length of inhibition zone for each streak was measured.
"Test-tube test" This technique was recently developed by Nishioka et al. [17] for the purpose of obtaining further sensitive and quantitative results of DNA repair test. Using the technique, DIG among the bacterial strains different in DNA repairing capacities can be generally observed with even much lower concentration of the sample-solution. For instance, it has been noted [19] that K2Cr207 reported recently to be mutagenic by Nishioka [18] indicates a striking DIG with 5 • 10 -4 M for the "test-tube t e s t " while with 5 • 10 -2 M for the "streak test". Also a number of mutagens such as certain metal compounds and dyes which are difficult to be detected on solid agar plate have been newly found by the same author [19] from environment with the technique. An 0.1 ml aliquot of each culture diluted as described above was added to a small test tube (120 mm in length, 9 mm in diameter) containing 1.8 ml of M9 buffer supplemented with 0.4% w/v of glucose, 20 #g/ml of tryptophan and 0.1 ml of different concentrations of a drug to be tested. An optical density at 660 nm of the culture treated with the drug for 4 h incubation at 37°C with shaking in the small test tube was measured with Shimazu Spectronic 20 and then the concentrations of the chemical agent resulting in 50% growth inhibition in optical density for the four strains were compared. Reversion test Mutagenic activity was determined by the standard "spot t e s t " and the quantitative "treat and plate test". "Spot test" Bacteria harvested at their late log phase and washed by centrifugations were resuspended in M9 buffer as obtaining 0.3 oi~ an optical density at 660 nm. An 0.1 ml aliquot of the suspension was spread o n SEM agar plate and a filter disc (21 mm in diameter) containing 0.1 ml of each sample solution was placed on the center of each plate. All the plates were incubated at 37°C for 48 h and the number of revertant colonies grown around each paper disc was scored. "Treat and plate test" [8] The bacterial suspension described above was treated with different concentrations of a chemical for 30 min at room temperature with mild shaking. After treatment, the chemical was removed by washing with M9 buffer by centrifugations twice. For assaying low mutation yield, cell suspensions were concen-
198 trated about 5-fold or 50-fold. For measurement of the number of revertant cells, 0.1 ml portions of the suspension was spread on SEM agar plates in triplicate. The cell suspension treated was diluted appropriately and plated on the NB agar plates in triplicate for determining viability. All the plates were incubated at 37°C for 48 h, and then the numbers of colonies on mutation plates and on the viability plates were counted. Results
Comparison of bactericidal effect on NCS among E. coli B/r strains with different DNA repairing capacities A typical example of the simple "streak t e s t " is shown in Fig. 1. The distance (mm) between the edge of a filter disc and that of each streak indicates clearly the level of inhibition by the drug for bacterial growth of each strain. The results of the "streak t e s t " for NCS with differeni doses and for other chemicals are shown in Table I. While KM, which has no mutagenic activity, exhibited inhibitory effect on the growth of four strains to similar extents, NCS as well as the standard mutagenic agents such as MMC, 4-nitroquinolineN-oxide (4NQO) or N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) was more inhibitory to the recA- strains (CM571 and WP100) than to the recA ÷ counterparts (WP2 and WP2uvrA). The uvrA- strains (WP2uvrA and WP100) do n o t differ significantly from the respective uvrA* counterparts (WP2 and CM571)
Fig. 1. P h o t o g r a p h of a p l a t e in the " s t r e a k t e s t " of NCS. Strains WP2 ( u v r A +, recA+), W P 2 u v r A ( u v r A - , recA+), C M 5 7 1 ( u v r A +, r e e A - ) a n d W P 1 0 0 ( u v r A - , r e e A - ) were g r o w n o v e r n i g h t in n u t r i e n t b r o t h . E a c h c u l t u r e was d i l u t e d 10 t i m e s w i t h fresh n u t r i e n t b r o t h and s t r e a k e d o n n u t r i e n t b r o t h agar plate. T e n m g o f N C S w e r e s o a k e d in a filter disc ( W h a t m a n G F C , 24 r a m in d i a m e t e r ) a n d p l a c e d o n t h e starting part o f 4 s t r e a k s so as t o c o v e r t h e i r b e g i n n i n g . T h e l e n g t h o f i n h i b i t i o n z o n e s for e a c h s t r e a k w a s m e a s u r e d a f t e r t h e i n c u b a t i o n for 24 h a t 3 7 ° C . U p p e r , WP2, 1 m m i n h i b i t i o n ; left, W P 2 u v r A . 1 r a m ; right, CM5710 9 ram; lower. WP100, 9 ram.
199 Table 1
RESULTS OF THE ~'STREAK TEST" OF CHEMICALS Chemicals
Control NCS
MMC 4NQ0 MNNG KM
Dose /~g/dlsc
0 5 1 5 1 1 2 2 8
X 102 X 103 X 103 )
Mutation Research, 48 (1977) 195--204 © Elsevier/North-Holland Biomedical Press
EFFECT OF DNA REPAIR SYSTEMS ON ANTIBACTERIAL AND MUTAGENIC ACTIVITY OF AN ANTITUMOR PROTEIN, NEOCARZINOSTATIN
K. TATSUMI * and H. NISHIOKA
Department of Medicine, Kyoto University, Kyoto 606 and Laboratory of Biophysical Chemistry, Doshisha University, Kyoto 602 (Japan) (Received October 12th, 1976) (Accepted October 20th, 1976)
Summary Antibacterial and mutagenic effects of an antitumor protein, neocarzinostatin (NCS), on isogenic strains of Escherichia coli with normal or defective DNA repair systems were studied. Growth of the strains lacking recA gene was inhibited by NCS with much lower concentration than in the case of those possessing it, while this "differential inhibition for g r o w t h " {DIG) between the strains uvrA ÷ and uvrA- was not seen. NCS induces mutation in recA ÷ strains but not significantly in recA-, while no such difference of mutagenesis was noticed between the strains uvrA ÷ and uvrA-. These results suggest that NCS produces hardly excisable DNA damage which is repaired by an error-prone recombination process.
Introduction NCS, an antitumor antibiotic isolated from a culture filtrate of Streptomyces carzinostaticus Var F41 [11], is an acidic single-chain polypeptide of approx. 10,700 mol. wt. whose primary structure has been recently elucidated [16]. This antibiotic inhibits growth of several gram-positive organisms [ 11 ] and a number of transplantable solid tumors and leukemias in mice [5,11,15]. As to its clinical efficacy, a remission rate of 50% obtained in a group of patients with acute leukemia [10] and a regression observed in certain solid tumors [26] have been reported. Mode of action studies indicated that NCS inhibits DNA synthesis [20,21] and also causes degradation of existing DNA in Sarcina lutea [21]. It was also reported that NCS induces DNA breakages in L1210 cells [23,27] and HeLa-S3 * Present address: D e p a r t m e n t o f M i c r o b i o l o g y , the University o f Chicago, Chicago, Ill. 6 0 6 3 7 (U.S.A.).
196
cells [4,22] in vivo and in calf thymus DNA [27], E. coli DNA [12] and SV40 DNA [4] in vitro. The rejoining of the NCS-induced single-strand breaks observed in DNA of HeLa-S3 cells [22] and the unscheduled DNA synthesis detected in human lymphocytes treated with NCS [28] indicated the possibility that at least a part of DNA damage induced by this antibiotic is repaired. Thus the principal target for NCS-action appears to be cellular DNA, though its subcellular behavior is still unknown. Recently this protein antibiotic was found to produce repairable DNA damage and have mutagenic activity is bacterial systems [29]. In the present paper, we describe the antibacterial and mutagenic activities of NCS for isogenic strains of E. coli with normal or defective DNA repair systems, attempting to characterize further the type of DNA damage caused by NCS from the pattern of response of repair deficient strains. Materials and methods Chemicals NCS (Lot T-64) was the generous gift of Kayaku Antibiotics Research Co. Ltd., Tokyo, Japan. It was stored in the dark at --20°C and was dissolved in saline just before use. Mitomycin C (MMC) and kanamycin (KM) were provided from Kyowa Hakko Co. Ltd., Tokyo and Meiji Seika Co. Ltd., Tokyo, respectively. 4NQO, MNNG and EMS were purchased from Nakarai Chemicals Co. Ltd., Kyoto, Japan. Bacterial strains Four strains of E. coli possessing different DNA repairing capacities were used for DNA repair test and reversion test. They were (1) WP2 (trp-, uvrA ÷, recA ÷) derived from B/r by Witkin [30], (2) WP2uvrA (trp-, uvrA-, recA ÷) originally isolated by Hill [9], (3) CM571 (trp-, uvrA ÷, recA-)isolated from WP2 by Bridges [7] and (4) WP100 (trp-, uvrA-, recA-) constructed by Witkin [8,31]. These strains are nearly isogenic and have a tryptophan-deficiency that is suppressible by ochre suppressor mutations [6]. Culture media Nutrient broth: 8 g of Difco nutrient broth (dehydrated) and 0.5 g NaC1 were dissolved in 1000 ml of distilled water and sterilized; NB agar: 15 g (Difco) were added to 1000 ml of nutrient broth and solidified; M9 buffer: 1 g NH4C1, 15 g Na2HPO4 • 12H20, 3 g KH2PO4 and 0.5 g NaC1 were dissolved in 1000 ml distilled water, sterilized by autoclaving and then 3.3 ml of 0.5 M MgSO4 added; SEM agar [31] : 0.2% w/v glucose and 0.04% w/v nutrient broth as well as 15 g bacto-agar were added to M9 buffer. DNA repair test In order to detect DIG among the four strains of E. coli by NCS, the simple "streak test" and the super sensitive "test-tube test" [17] were adopted. "Streak test" The essential procedures were originally described by Kada et al. [13] as
197 "rec-assay" for screening of chemical mutagens using wild and recombination deficient strains of Bacillus subtilis. The four strains of E. coli were grown overnight in nutrient broth. Each culture was diluted with M9 buffer to obtain a b o u t 0.3 of an optical density at 660 nm, giving approx. 2.5 • 108 cells per ml. The cultures were then streaked on NB agar plate from a central point in different directions using 0.1-ml pipette severally. A drug solution in 0.1 ml was spotted onto a filter disc (Whatman GFC, 24 mm in diameter) which had been placed on the plate so as to cover the starting points of the four streaks. All the plates were then incubated for 24 h at 37°C and the length of inhibition zone for each streak was measured.
"Test-tube test" This technique was recently developed by Nishioka et al. [17] for the purpose of obtaining further sensitive and quantitative results of DNA repair test. Using the technique, DIG among the bacterial strains different in DNA repairing capacities can be generally observed with even much lower concentration of the sample-solution. For instance, it has been noted [19] that K2Cr207 reported recently to be mutagenic by Nishioka [18] indicates a striking DIG with 5 • 10 -4 M for the "test-tube t e s t " while with 5 • 10 -2 M for the "streak test". Also a number of mutagens such as certain metal compounds and dyes which are difficult to be detected on solid agar plate have been newly found by the same author [19] from environment with the technique. An 0.1 ml aliquot of each culture diluted as described above was added to a small test tube (120 mm in length, 9 mm in diameter) containing 1.8 ml of M9 buffer supplemented with 0.4% w/v of glucose, 20 #g/ml of tryptophan and 0.1 ml of different concentrations of a drug to be tested. An optical density at 660 nm of the culture treated with the drug for 4 h incubation at 37°C with shaking in the small test tube was measured with Shimazu Spectronic 20 and then the concentrations of the chemical agent resulting in 50% growth inhibition in optical density for the four strains were compared. Reversion test Mutagenic activity was determined by the standard "spot t e s t " and the quantitative "treat and plate test". "Spot test" Bacteria harvested at their late log phase and washed by centrifugations were resuspended in M9 buffer as obtaining 0.3 oi~ an optical density at 660 nm. An 0.1 ml aliquot of the suspension was spread o n SEM agar plate and a filter disc (21 mm in diameter) containing 0.1 ml of each sample solution was placed on the center of each plate. All the plates were incubated at 37°C for 48 h and the number of revertant colonies grown around each paper disc was scored. "Treat and plate test" [8] The bacterial suspension described above was treated with different concentrations of a chemical for 30 min at room temperature with mild shaking. After treatment, the chemical was removed by washing with M9 buffer by centrifugations twice. For assaying low mutation yield, cell suspensions were concen-
198 trated about 5-fold or 50-fold. For measurement of the number of revertant cells, 0.1 ml portions of the suspension was spread on SEM agar plates in triplicate. The cell suspension treated was diluted appropriately and plated on the NB agar plates in triplicate for determining viability. All the plates were incubated at 37°C for 48 h, and then the numbers of colonies on mutation plates and on the viability plates were counted. Results
Comparison of bactericidal effect on NCS among E. coli B/r strains with different DNA repairing capacities A typical example of the simple "streak t e s t " is shown in Fig. 1. The distance (mm) between the edge of a filter disc and that of each streak indicates clearly the level of inhibition by the drug for bacterial growth of each strain. The results of the "streak t e s t " for NCS with differeni doses and for other chemicals are shown in Table I. While KM, which has no mutagenic activity, exhibited inhibitory effect on the growth of four strains to similar extents, NCS as well as the standard mutagenic agents such as MMC, 4-nitroquinolineN-oxide (4NQO) or N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) was more inhibitory to the recA- strains (CM571 and WP100) than to the recA ÷ counterparts (WP2 and WP2uvrA). The uvrA- strains (WP2uvrA and WP100) do n o t differ significantly from the respective uvrA* counterparts (WP2 and CM571)
Fig. 1. P h o t o g r a p h of a p l a t e in the " s t r e a k t e s t " of NCS. Strains WP2 ( u v r A +, recA+), W P 2 u v r A ( u v r A - , recA+), C M 5 7 1 ( u v r A +, r e e A - ) a n d W P 1 0 0 ( u v r A - , r e e A - ) were g r o w n o v e r n i g h t in n u t r i e n t b r o t h . E a c h c u l t u r e was d i l u t e d 10 t i m e s w i t h fresh n u t r i e n t b r o t h and s t r e a k e d o n n u t r i e n t b r o t h agar plate. T e n m g o f N C S w e r e s o a k e d in a filter disc ( W h a t m a n G F C , 24 r a m in d i a m e t e r ) a n d p l a c e d o n t h e starting part o f 4 s t r e a k s so as t o c o v e r t h e i r b e g i n n i n g . T h e l e n g t h o f i n h i b i t i o n z o n e s for e a c h s t r e a k w a s m e a s u r e d a f t e r t h e i n c u b a t i o n for 24 h a t 3 7 ° C . U p p e r , WP2, 1 m m i n h i b i t i o n ; left, W P 2 u v r A . 1 r a m ; right, CM5710 9 ram; lower. WP100, 9 ram.
199 Table 1
RESULTS OF THE ~'STREAK TEST" OF CHEMICALS Chemicals
Control NCS
MMC 4NQ0 MNNG KM
Dose /~g/dlsc
0 5 1 5 1 1 2 2 8
X 102 X 103 X 103 )
