Chem.-BioL Interactions, 48 (1984) 145--152 145 Elsevier Scientific Publishers Ireland Ltd.

GENETIC ACTIVITY OF BLEOMYCIN IN ESCHERICHIA COLI

KAZUO YAMAMOTO a, TAKEAKI HIRAMOTO a, HIDEO SHINAGAWA a and YOSHISADA FUJIWARA a

aDepartment of Radiation Biophysics, Kobe University School of Medicine, Kobe 650 and bDepartment of Experimental Chemotherapy, Rese~ch Institute for Microbial Diseases, Osaka University, Suita 565 (Japan)

(Received June 27th, 1983) (Revision received September 19th, 1983) (Accepted September 20th, 1983)

SUMMARY

The induction of umuC gene expression, cell lethality, induction of W-reactivation of UV-irradiated k-phage and the induction of mutagenesis caused by bleomycin (Blm) were studied in Escherichia coli K-12 strains with special references to the effects of SOS repair deficiencies. (1) The umuC gene is inducil~le by Blm and the induction is regulated by the lexA and recA genes. (2) The lexA and recA mutants are slightly more sensitive to Blm-killing than wild-type strain. (3) The plating efficiency of UV- irradiated k-phage increased by Blm treatment of the host cell. This increase was not observed in the urnuC mutant . The plating efficiency of UV-irradi- ated k-phage was drastically reduced in the lexA and recA strains treated with Blm. (4) No significant increase of the reversion of nonsense mutat ion (his-4 to His*) in ABl157 by the t reatment of Blm was observed. Possible implications of these results are discussed.

Key words: Bleomycin -- SOS functions -- umuC'-lac'Z fusion plasmid -- umuC gene expression -- W-reactivation -- Mutagenesis

INTRODUCTION

Blm is an antibiotic obtained from Streptornyces verticullus [1] : Initially it was investigated as both antimicrob~al and anti tumor agent. Blm has been shown to effect living cells which resemble those produced by ionizing radiation; loss of the ability of mammalian and bacterial cells to form

Abbreviations: Blm, bleomycin.

0009-2797/84/$03.00 © 1984 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

146

colonies [2] , product ion of chromosome aberration [3] , degradation of intracellular DNA [4], inhibition of DNA synthesis [5] and release of DNA from a complex with membrane [6]. Blm produces single-strand [7] and double-strand [8] breaks in DNA in solution, as does ionizing radiation. However, it has been repor ted that Blm induces no greater effect on colony forming ability of lexA and polA strains of E. coli than on wild-type strain [4,9] , even though the two mutants are more sensitive to X-rays. E. coli recA cells, which are quite sensitive to X-rays, have been reported to be equally sensitive to the drug as wild-type cells [9] . Mutagenic activity of Blm has not been demonstrated in bacteria, even in the supersensitive strains of Ames test [10] , although it has been shown in yeast [11,12] . Thus, biological effects of Blm appear to be different from those of X-rays, and Blm seems to be nonmutagenic or very weakly mutagenic in bacteria.

In E. coli, it was previously established that Blm induces the expression of the recA gene [13] , and prophage [14,15] . From these findings, it is reasonable to expect that Blm induces so-called SOS operons [16]. Included among the SOS operons is the umuC gene which is thought to be one of the genes directly involved in E. coil mutagenesis and W-reactivation [17] . However, as ment ioned above, no one has succeeded in demonstrating the mutagenicity of Blm in E. coli. To discriminate the alternatives, that Blm does no t induce mutagenic activity, or it does induce the activity bu t the mutagenic assay system used was no t proper, we examined the inducibility of the umu operon by Blm. A system for measuring the levels of umu expression became available recently [18] . Furthermore, inducibility of W-reactivation of UV-irradiated k-phage by Blm was studied. In this paper, we reported the results of these experiments and discuss the implications of our findings.

MATERIALS AND METHODS

Chemicals Clinical preparations of Blm, a generous gift from Nippon Kayaku Co.,

Tokyo, was dissolved in triple-distilled water at 1 mg/ml and kept frozen at - 2 0 ° C in small quantities.

Bacterial strains, phage and plasmid The bacterial strain used in this s tudy are; E. coli K-12 A B l 1 5 7 (argE3,

h/s-4, proA2, thr-1, thi-1, rpsL31, leu-6, gaIK2, lacY1, ara-14, xy/-5, mtl-1, supE44) and its derivatives KY1056 (recA56), Z-7 (lexA1) and TK712 (umuC36). TK712 was obtained from Dr. T. Kato. The other strains were described previously [19] . ),v~-phage (designated as phage k in this paper) was used for the W-reactivation experiments. The plasmid pSK1002 was described previously [18]. Briefly, a 5 kb EcoRI-BamHI chromosomal DNA fragment carrying the promoter of umu operon, the umuD gene and a part of the umuC gene, was ligated with EcoRI-BamHI digested pMC1403,

147

so-called Casadaban's vector [20] . The reading frame of umuC'-lac'Z was adjusted in vitro so that translation produces a hybrid protein with ~-galacto- sidase. Thus, resultant plasmid, pSK1002 which carried umuC'-lac'Z, enables us tc s tudy the levels of umuC gene expression by measuring ~-galactosidase activity in the cells carrying the plasmid, since the expression of iacZ is regulated b y the promoter of umu operon.

Media K medium, M-9 buffer, and L broth have been described [19] . For

mu ta t i on assay, we used SEM agar [21] . For the W-reactivation, ~ agar was used [22] .

Mutagenesis and W-reactivation Exponential ly growing cultures of the test strains in K medium (0.5--

1 × l 0 s cells/ml) were divided into several parts, to which Blm was added at various concentrations. The cultures were shaken at 37°C for 120 min. Then, a fraction of each cultures was diluted 100-fold in M-9 buffer and plated on L agar to count survivors after adequate dilutions. The remaining cultures were washed extensively in M-9 buffer, resuspended in M-9 buffer and plated on SEM agar to count His + revertants. The plates were incubated at 37°C for 48--72 h. For the W-reactivation experiments, the Blm treated cells were resuspended in 0.01 M MgSO4 and nine parts of the cell suspen- sion were mixed with one part o f k-phage suspensions (3 × 107 PFU/ml in 0.01 M MgSO4) which had been irradiated with UV or not. The suspensions of the cell-phage complexes were s tood at ice-temperature for 30 min, incubated for 30 min at 37°C and pour-plated on k agar using AB2480 (recA13 uvrA6) as an indicator strain. In view of the dependence of Blm- induced killing upon the growth phase of bacteria [9,23] , care was taken so that cultures of the strains to be compared were in early exponential growth phase. The W-reactivation experiments were done under yel low light condition.

Assay of •-galactosidase activity Exponentially growing cultures of the test strains with or wi thout

pSK1002 were t reated with Blm (5 ~g/ml) at 37°C, the samples were with- drawn from the cultures at intervals, pelleted by centrifugation at 4°C, resuspended in K medium, and ~-galactosidase activity was measured by the method described by- Miller [24].

RESULTS

Genetic regulation of umuC gene expression by Blm In order to examine the mutagenicity of Blm in E. coil, we first examined

whether Blm induces expression of umu operon, which is responsible for inducible mutagenesis. The strains carrying the plasmid, pSK1002, were very

148

much suited for the experiments, since the lacZ gene on pSK1002 is regu- lated by the umu promoter , in other words, the levels of umu ope~on expres- sion can be easily detectable by measuring the ~-galactosidase activity in the strains carrying the plasmid. The results {Fig. 1) show that Blm induces the expression of umu operon in the wild-type strain, bu t not in lexA and recA mutants. Blm does not induce lacZ gene expression from its authentic promoter as shown in the case of ABl157 which carried no pSK1002 plasmid. Thus, the increase of the ~-galactosidase activity in ABl157 / pSK1002 by Blm is due to the expression of the umu operon which is controlled by the lexA and recA gene products. We therefore concluded that Blm is one of the typical DNA damaging agents which induce SOS regulon in a manner dependent on IexA-recA genes.

Bacteriocidal effects of Blm We next examined ABl157 and its derivatives for their sensitivity to

Blm-killing. As shown in Fig. 2, KYlO56(recA56) and Z-7(lexA1) were about two-fold more sensitive to Blm than ABl157 (wild-type) and TK712- (umuC36). However, differences in the sensitivity be tween recA and wild- type cells to Blm were relatively small as compared to the differences seen when sensitivity to ~f-rays is measured, about 4-fold [9].

20

b

815

o

i,° 5

~ ~ O ~ o

4 ~ • , , '~ ,

®

£ 1 0

o ~ O

[ ]

0 1 2 5 I0

T i m e . h r B l m (~JcJ/m [)

Fig. 1. Kinetics of induction of ~-galactosidase in the strains carrying a umuC'-lac'Z plasmid, pSK1002. Exponential cultures growing in K medium were added with Blm (5 ~g/ml) at the t ime indicated 0 h. o, ABl157/pSK1002; 4 KY1056(recA56)/pSK1002; =, Z-7(lexA1)/pSKlO02; ×, ABl157; o, ABl157/pSK1002 without Blm treatment.

Fig. 2. Killing of E. coil strains by Blm. o, ABl157; ~, KYlO56(recA56); o, Z-7(lexA 1); o, TK712(umuC36).

149

W-reactivation of UV-irmdiated k-phage in Blm-treated bacteria Stimulated by the results shown in Figs. 1 and 2, which suggest the

induction of SOS repair capacity by Blm, we measured the inducibility of W-reactivation. kv~-phage was irradiated with 240 J /m 2 UV (254 nm UV) in 0.01 M MgSO4 and Blm-treated lexA, recA and umuC host strains were infected with the irradiated phages. Plaque forming ability of X-phage in these strains was shown in Fig. 3. First, the capacities of these E. coli strain~, treated with various doses of Blm, to propagate unirradiated X-phage differed only slightly. The plating efficiencies of 240 J /m 2 UV irradiated k-phage increased in ABl157 (wild-type) with increasing dose of Blm by a factor of approx. 6. On the other hand, Blm decreased the ability of the lexA and the recA cells to repair the UV-irradiated phage, and the drug, at the concentrat ion of 10 pg/ml, almost completely eliminated the ability of these two mutants to repair UV-irradiated phage. The sensitivity of the umuC strain to Blm was identical tc the umuC ÷ counterpart (Fig. 2). How- ever, as shown in Fig. 3, no W-reactivation in the Blm-treated umuC cells was observed.

Blm mutagenesis in E. coli We measured the ability of Blm to induce reversion from his-4 to His + in

ABl157 , and the results were summarized in Table I. Using our t reatment

~ o . ~ ~ ~ - o

10"

t~

5,

01 25 BIm to Bacteria(pg/m[)

Fig. 3. W-reactivation of UV-irradiated ~,-phage (lower data) and capacity for k-phage infection (upper data) plotted against Blm dose to host bacteria before absorption. The W-reactivation factor is the ratio (plaque forming unit of UV 240 J /m 2 phage on t h e bacteria treated with Blm dose indicated o n t h e abscissae)/(plaque forming unit of UV240 J /m 2 phage on untreated bacteria). Experiments were carried out 3 times and were highly reproducible. Each result represents the mean value of 2 plates of 1 represen- tative experiment, o, e, ABl157 ; ~, A, KY1056(recA56); D, m, Z-7(lexA1); 0, , , TK712- (umuC36).

150

TABLE I

BIm-INDUCED REVERSION OF h/~4 MUTATIONS IN ESCHERICHIA COLI ABl157

Numerals in the brackets show the averages with S.D.

Blm (~ g/ml) No. of total colonies/L plate

No. of His+ revertants/S.E.M, plate (x ± %- 1)

0 281 X 10 ~ 6,6,3,3,6,4, (4.67 +- 1.51) 5 208 X 10 ~ 8,12,7,13,6, (9.2 -+ 3.11)

10 100 × 106 4,5,10, (6.33 +- 3.21)

condition, no mutagenic activity of Blm was observed for h/s-4 reversion. Similarly, Blm did not exhibit mutagenesis of h/s-4 to His ÷ in an ABl157 strain carrying the mul t icopy umuC+-plasmid (data not shown).

DISCUSSION

In the present work, it was evidenced that Blm induces the expression of the umu operon (Fig. 1), p roduct of which is known to be required for SOS mutagenesis and at least partly responsible for W-reactivation [17]. We have also shown that the induction of the umu operon by Blm is regu- lated by the lexA-recA functions (Fig. 1). We were, therefore, able to demonstrate that Blm induces W-reactivation in a wild-type strain but not in the umuC, lexA and recA strains (Fig. 3). Furthermore, it was shown that the recA and lexA mutants are more sensitive to this drug than the wild- type strain (Fig. 2). Taken together, the results that RecA protein and prophage are inducible by Blm treatment [13--15] , our present results all suggest that Blm induces a set of SOS functions including those related to DNA repair and mutagenesis.

However, we were unable to detect any significant increase of mutagenic activity of Blm in E. coli by the present assay system which measured the reversion of ochre mutat ion in his-4 to His+ (Table I). The apparent contra- diction might be resolved by assuming that Blm induces mutagenic activity in the cell but not the premutagenic lesions required for mutagenesis. Accordingly, the resultant mutat ions are specific and were no t detectable by the present back mutat ion assay. This assumption may be supported by the finding that DNA strand scissions by Blm have nucleotide specificity [25]. Furthermore, Moore [11] has presented evidence in yeast that Blm induces back mutations at the highly different frequencies, depending on the genetic marker examined. While she demonstrated an increased reversion frequency from ilv 1-92 to pro to t rophy of about 300-fold, she obtained only a 2--3-fold increase in the case of lys2, hisl and trp2 markers. Similar results were also obtained in yeast by Hannan and Nasim [12] . Thus, at least in yeast, the mutagenic effect of Blm seems highly locus (sequence} specific. With regard to nucleotide specific DNA scissions, it is no tewor thy that Blm produces double-strand breaks in DNA [8 ]. Thus, another plausible

151

scenar io is t h a t doub le - s t r and breaks p r o d u c e d b y Blm in D N A m a y be a l m o s t all l e tha l even in the wi ld - type E. coli, and hence n o n o m u t a g e n i c in bacter ia . We are cu r r en t ly engaged in the e x p e r i m e n t s to e x a m i n e w h e t h e r these are t rue in E. coli or not .

As s h o w n in Fig. 3, Blm t r e a t m e n t r educed the abi l i ty o f lexA and recA cells t o repa i r UV- i r rad ia ted ) ,-phage a t t he doses which did n o t a f fec t s ignif icant ly t he abi l i ty of these m u t a n t s to p r o p a g a t e the un i r r ad i a t ed phage. Since Blm is n o t able to induce SOS repa i r func t ions in these m u t a n t s , the l imi ted a m o u n t s o f cons t i tu t ive ly expressed D N A repa i r factors , for e x a m p l e R e c A pro te in , m a y be mob i l i zed for repai r ing Blm- d a m a g e d hos t D N A and m a y n o t be avai lable for cont ro l l ing the levels o f nuclease(s) in vivo. Thus, s u b s e q u e n t l y in fec t ed UV- i r rad ia ted phage D N A m a y be easily degraded b y the ac t ion of nuclease(s) , such as recSC nuclease. Similar p h e n o m e n o n , t h a t UV i r rad ia t ion reduces t he abi l i ty of t he recA hos t t o r eac t iva te the UV-i r rad ia ted phage, has been r e p o r t e d [ 2 6 , 2 7 ] .

ACKNOWLEDGEMENT

We are gra teful to N i p p o n K a y a k u Co., Ltd. for generous gifts o f blec- m y c i n and to Dr. T. K a t o for strains. This w o r k was s u p p o r t e d by Grant~ in-Aid for Scient i f ic Resea rch f r o m the Minis t ry o f Educa t ion , Science and Culture , Japan .

REFERENCES

1 H. Umezawa, K. Maeda, T. Takeuchi and Y. Okami, New antibiotics, bleomycin A and B, J. Antibiot., 19 (1966) 200.

2 H. Umezawa, M. Ishizuka, K. Maeda and T. Takeuchi, Studies on bleomycin, Cancer, 20 (1967) 891.

3 K. Ohama and T. Kadotani, Cytological effect of bleomycin on cultured human leukocytes, Jap. J. Human Genet., 14 (1970) 293.

4 T. Ohnishi, K. Shimada and Y. Takagi, Effects of bleomycin on Escherichia coli strains with various sensitivities to radiations, Biochiw_ Biophys. Acta, 312 (1973) 248.

5 H. Suzuki, K. Nagai, H. Yamaki, N. Tanaka and H. Umezawa, Mechanism of action of bleomycin, Studies with the growing culture of bacterial and tumor cells, J. Anti- blot., 21 (1968) 379.

6 M. Miyaki, T. Katayama and T. Ono, Breakage of DNA-membrane complex by bleomycin, J. Antbiot., 27 (1974) 647.

7 I. Sbirakawa, M. Azegami, S. Ishii and H. Umezawa, Reaction of bleomycin with DNA strand scission of DNA in the absence of sulfhydryl or peroxide compounds. J. Antibiot., 24 (1971) 761.

8 L.F. Povirk, W. Wiibker, W. KShnlein and F. Hutchinson, DNA double-strand breaks and alkali-labile bonds produced by bleomycin, Nucleic Acid Res., 4 (1977) 3573.

9 I~ Yamamoto and F. Hutchinson, Response to bleomycin of Escherichia coli mutants deficient in DNA repair, J. Antibiot., 32 (1979) 1181.

10 W.F. Benedict, M.S. Baker, L. Haroun, E. Choi and B.N. Ames, Mutagenicity of cancer chemotherapeutic agents in the Salmonella/microsome test, Cancer Res., 37 (1977) 2209.

11 C.W. Moore, Bleomycin-induced mutation and recombination in Saccharomyces cerevisiae, Mutat. Res., 58 (1978) 41.

152

12 M.A. Hannan and A. Nasim, Genetic activity of bleomycin: differential effects on mitotic recombination and mutations in yeast, Mutat. Res., 53 (1978) 309.

13 L.T. Gudas and A.B. Pardee, DNA synthesis inhibition and induction of protein X in Escherichia colt J. Mol. Biol., 101 (1976) 459.

14 C.W. Haidle, K . ~ Weiss and M.L. Mace Jr., Induction of bacteriophage by bleomy- cin, Biocherm Biophys. Res. Commun., 48 (1972) 1179.

15 M.-E. Armengod and M. Blanco, Influence of the recF143 mutation of Escherichia coli K12 on prophage k induction, Mutat. Res., 52 (1978) 37.

16 J.W. Little and D.W. Mount, The SOS regulation system of Escherichia coli, Cell, 29 (1982) 11.

17 T. Kato and Y. Shinoura, Isolation and characterization of mutants of Escherichia coli deficient in induction of mutations by ultraviolet light, Mol. Gen. Genet., 156 (1977) 121.

18 H. Shinagawa, T. Kato, T. Ise, K. Makino and A. Nakata, Cloning and characteriza- tion of the umu operon responsible for inducible mutagenesis in Escherichia colt Gene, 23 (1983) 167.

19 K. Yamamoto, M. Satake and H. Shinagawa, A multicopy phr-plasmid increases ultraviolet resistance of a recA strain of Escherichia coli, Mutat. Res. (1983) in press.

20 M.J. Casadaban, J. Chou and S.N. Cohen, In vitro gene fusions that join the enzymati- cally active ~-galactosidase segment to amino-terminal fragments of exogeneons proteins: Escherichia coli plasmid vectors for the detection and cloning of transla- tional initiation signals, J. Bacteriol., 143 (1980) 971.

21 T. Kato, R.H. Rothman and A.J. Clark, Analysis of the role of recombination and repair in mutagenesis of Escherichia coli by UV radiation, Genetics, 87 (1977) 1.

22 T. Horiuchi and H. Inokuchi, Temperature-sensitive regulation system of prophage lambda induction, J. MoL BioL, 23 (1976) 217.

23 I~L Suzuki, Killing action of bleomycin on radiation-sensitive mutants of Escherichia col~ Jap. J. Genet., 46 (1971) 277.

24 J.H. Miller, Experiments in molecular genetics, Cold Spring Habor Laboratory, Cold Spring Habor, New York, 1972.

25 A.D. D'Andrea and W.A. Haseltine, Sequence specific cleavage of DNA by the anti- tumor antibiotics neocartinostatin and bleomycin, Proc. NatL Acad. Sci. U.S.A., 75 (1978) 3608.

26 M. Defais, P. Fauquet, M. Radman and M. Errera, Ultraviolet reactivation and ultra- violet mutagenesis of k in different genetic systems, Virology, 43 (1971) 495.

27 M. Radman and R. Devoret, UV-reactivation of bacteriophage k in excision repair- deficient hosts: independence of red functions and attachment regions, Virology, 43 (1971) 504.