Inhibition of recA induction by the radioprotector 2-mercaptoethylamine

Mutation Research, 282 (1992) 203-207 © 1992 Elsevier Science Publishers B.V. All rights reserved 0165-7992/92/$05.00

203

MUTLET 0680

Inhibition of recA induction by the radioprotector 2-mercaptoethylamine Maria N islund, Ada Kolman and L. Ehrenberg Department of Radiobiology, Stockholm University, S-106 91 Stockholm, Sweden (Received 30 November 1991) (Revision received 24 March 1992) (Accepted 27 March 1992)

Keywords: Escherichia coli; RecA induction; Modification by thiols; UV; Gamma-radiation; Nalidixic acid

Summary Our earlier finding that the radioprotective action of 2-mercaptoethylamine (MEA) is counteracted by ascorbate suggests a biochemical mechanism of action, which is supported by observations that MEA is not radioprotective in R e c - E. coli strains. In this study we show that MEA inhibits the induction of the recA gene by UV- or y-irradiation or by nalidixic acid in Escherichia coli strain GE94, which contains a recA-lacZ fusion. This effect, which may be counteracted by cysteine, indicates that in general MEA inhibits the induction of SOS functions.

The radioprotective action of 2-mercaptoethylamine (MEA, cysteamine) is at present regarded as involving an influence on the balance of cellular repair systems. The mode of action of this radioprotective compound, discovered by Bacq et al. in 1951, is still not dear. Although it has long been recognized that MEA provokes a multitude of biochemical and physiological changes in living organisms (Bacq, 1965), radiobiological monographs are dominated by physico-

Correspondence: Dr. M. Niislund, Department of Radiobiology, Stockholm University, S-106 91 Stockholm, Sweden.

Abbreviations: MEA, 2-mercaptoethylamine (cysteamine); Cys, cysteine; MNU, N-methyl-N-nitrosourea; ONPG, Onitrophcnyl-/3-v-galactopyranoside; IPTG, iaopropyl-~O-Dthiogalactopyranoside; NAL, nalidixic acid; SEM, semi-enriched medium.

chemical interpretations of its action under aerobic conditions, such as proton donation (Alexander and Charlesby, 1955; Howard-Flanders, 1960) and radical scavenging (Braams, 1963; Nakken, 1965). Our earlier finding that ascorbate counteracted the radioprotective effect of MEA seemed to suggest a mechanism for the biochemical action of this agent (N~islund et al., 1976). Among the metabolic effects of MEA that might lead to an explanation of its mode of action, a marked inhibition of inducible protein synthesis was observed. This concerns induction of /3-galactosidase by isopropylthiogalactoside (IPTG), which, like the radioprotective action, could be reversed following y-irradiation in the presence of ascorbate (Niislund and Ehrenberg, 1978). The observed enhancement of the mutagenic action of N-methyl-N-nitrosourea by MEA in an adaptable E. coli strain (N~islund et al., 1986) is compatible with prevention of the induc-

204

tion of methyltransferase (Defais, 1985; Laval, 1985) (no such effect was observed in an a d a strain or in pre-induced bacteria - see Niislund et al., 1986). A possible target for MEA action might be one within the SOS repair system, which is a complex group of inducible functions of considerable importance in the survival of bacteria following treatment with various genotoxic agents (D'Ari, 1985; Walker, 1984; Little and Mount, 1982). A working hypothesis that some SOS functions are involved in the radioprotective action of MEA is supported by the observation that MEA exerts no radioprotective action in rec- E. coli WP10 (as compared to the wild-type WP2 where MEA is radioprotective - Hiilsewede and Schulte-Frohlinde, 1986). In these experiments, MEA did not increase survival following y-irradiation of the rec- bacteria; this suggests the basis for its biochemical mode of action. The present paper describes experiments carried out in order to clarify the action of MEA on recA induction, using the recA-lacZ gene fusion in E. coli GE94 (Weisemann et al., 1984) and various inducing agents. Our studies include effects of cysteine, shown in earlier studies (Niislund et al., 1986) to counteract the radioprotective action and other outcomes of MEA treatment.

Materials and methods Inorganic chemicals were obtained from Kebo (Sweden), and all other chemicals and biochemicals were from Sigma Chemical Company (St. Louis, MO, USA). For determination of fl-galactosidase, according to Miller (1972), two media were used. Medium I contained: 6 g NaEHPO4, 3 g KH2PO 4, 0.5 g NaCl, 1 g NH4CI, 0.2 g MgSO4, 0.0174 g CaCI 2 • 2H20, 2 g glucose and 100 ml of Luria broth per liter. Medium II contained: 16 g N a 2 H P O 4 . 7 H 2 0 , 5.5 g NaH2PO 4, 0.75 g KCI, 0.246 g MgSOg-7H20, and 2.7 ml 2-mercaptoethanol per liter. The GE94 strain of E. coli was obtained from Dr. G.M. Weinstock (University of Texas Medical School, Houston, TX, USA). Bacteria from an overnight culture in medium I were diluted 50 times with medium I, and the culture was aerated for about 3 h by shaking at 37°C until a cell

density of 5 x 108/ml, equal to OD 0.5 at 520 nm, was reached. The bacteria were diluted 3 times with fresh medium. Portions of the suspension were distributed into glass tubes containing the compounds to be tested. The rec-lac functions were induced by UV- or y-radiation or nalidixic acid (0.22 mM), and the influence of MEA (3 mM) or Cys (0.1 mM) or MEA + Cys, added 20 min before induction or at different times after induction, was studied. Gamma-irradiation. Cells suspended in medium I in glass tubes were irradiated with 137Cs radiation at a dose rate of 10 Gy/min, the dose rate being determined by Fricke dosimetry, UV-irradiation. Bacterial cells were irradiated at room temperature with monochromatic (254 nm) UV light, 6 J m2/min, under stirring in open petri dishes with suspension layers no more than 1 mm thick. After different times of treatment with inducer (UV- or y-radiation, nalidixic acid), 0.1-ml samples were withdrawn and rapidly mixed with 1 ml of medium II to which two drops of a 9:1 mixture of chloroform and toluene had been added. The samples were maintained at 37°C in a water bath for 20 min, then cooled down to 0°C and maintained at this temperature for 10 min. oNitrophenyl-/3-galactopyranoside (ONPG) was added and the mixture incubated at 28°C. The reaction was terminated by addition of Na2CO 3 to a final concentration of 0.5 M. Absorbances were determined at 420 nm against a reagent blank in a Perkin-Elmer Model 124 spectrophotometer. The intracellular specific activity of the enzyme was calculated according to the following equation, as described by Miller (1972): specific activity = 1.000 X [OD420 - 1.75 X OD550] - (t x v x OD600), where t is the duration of the assay in min, and v is the volume of the culture used per ml of assay mixture. All glassware was washed with EDTA and distilled water in order to remove metals that could catalyze sulfhydryl oxidation (cf. Ehrenberg et al., 1989).

Results Induction of the recA gene by UV, nalidixic acid (NAL) and 7-radiation, as indicated by ob-

205 TABLE 1 INFLUENCE OF MEA (3 mM) OR MEA (3 mM)+CYS (0.1 mM) ON THE INDUCTION BY UV OF/3-GALACTOSIDASE IN E. coli GE94 (WITH recA-lacZ FUSION) MEA and Cys were added 20 min before UV-radiation. Pretreatment

UV dose ( J / m 2)

none none none none MEA MEA MEA MEA+Cys MEA+Cys MEA+Cys

0 2.5 5.0 20 2.5 5.0 20 2.5 5.0 20

/3-Galactosidase activity, units at time (min) after UV irradiation 0

20

40

60

120

220

220 806 993 1155 361 360 317 966 1100 1206

238 638 1553 1472 361 361 317 776 1500 1516

225 472 1054 1138 309 309 290 760 1138 1320

226 352 904 1027 287 287 205 688 842 1170

TABLE 2 INFLUENCE OF MEA (3 mM) AND Cys (0.1 mM) ON THE INDUCTION BY NALIDIXIC ACID (NAL; 0.22 mM) OF /3-GALACTOSIDASE IN E. coli GE94 (WITH recA-lacZ FUSION) MEA and Cys were added 20 min before NAL, or, at indicated times after NAL. Experiments h and i were carried out on a different occasion. Expt.

Pretreatment

Treatment

Posttreatment

a b c d e f g h i

none none Cys MEA MEA+Cys MEA MEA none none

none NAL NAL NAL NAL NAL NAL NAL NAL

Cys(25min) Cys(35min) none MEA(15min)

/3-Galactosidase activity, units at time (min) after NAL treatment 0

15

30

113

130 214 260 160 228 183 160 376 376

172 422 672 160 761 472 160 697 697

45

60

120

761 870

170 779 1188 183 1472 713 783 1072 944

151 898 1804 297 2183 1317 1240 2486 1027

TABLE 3 INFLUENCE OF MEA (3 mM) OR MEA (3 mM)+ Cys (0.1 mM) ON THE INDUCTION BY 3,-RADIATION OF/3-GALACTOSIDASE IN E. coli GE94 (WITH recA-lacZ FUSION) MEA and Cys were added 20 min before irradiation. Pretreatment

none none none MEA MEA MEA+Cys MEA+Cys

Gamma-ray dose (Gy)

fl-Galactosidase activity, units at time (min) after y-irradiation 0

15

30

45

60

120

0 20 40 20 40 20 40

92

92 315 553 194 220 363 601

92 672 1029 194 220 862 1101

90 804 1004 204 229 971 1504

95 959 1040 209 208 1103 1603

91 896 982 178 150 1083 1383

206

served increases in specific activity of /3-galactosidase, is recorded in Tables 1, 2 and 3, respectively. When MEA was added prior to any of the three inducing treatments, induction was strongly suppressed. This suppression was counteracted by a low concentration of Cys added at the same time as the MEA. Later addition of Cys was also shown to counteract the suppression produced by MEA (Table 2, Expts. f, g), whereas if MEA was added after the inducing treatment, induction was not suppressed (Table 2, Expts. h, i).

Discussion Mechanism o f action o f M E A

Irrespective of the inducing agent applied (3'radiation, UV light, nalidixic acid), addition of 3 mM MEA appeared to suppress the induction of the recA gene. Addition of MEA at an early stage following treatment with the inducer (e.g., 15 min after administration of nalidixic acid - see Table 2) gave results which indicate that the induction phase rather than the ensuing protein synthesis is inhibited by MEA. This is compatible with our earlier finding that RNA synthesis (as measured by the rate of incorporation of [14C]uridine) is rapidly suppressed by MEA (Ehrenberg et al., 1974). The ability of Cys, at a concentration (0.1 mM) much lower than that of MEA, to counteract the suppression caused by MEA was further studied with respect to the mechanisms involved. The enhancement by Cys of the response to the inducing agent (see Tables 1, 2 and 3) indicates that Cys plays an important role in induction. The fact that the response to MEA + Cys is often even stronger than the response to Cys alone may have the simple explanation that the presence of excess MEA assists in maintaining Cys in the reduced state. Even if measures were taken to eliminate trace metals which catalyze thiol 'autoxidation' (Ehrenberg et al., 1989), amounts of catalyst sufficient to render it difficult to prevent oxidation of Cys at a concentration as low as 0.1 mM are often associated with bacteria (unpublished data). Kinetic studies, based on intracellular concentrations of thiols, have been initiated for the purpose of clarifying whether competition

between Cys and its decarboxylated product, MEA, occurs. The suppression of gene induction by MEA and its reversion by Cys is not specifically limited to the rec gene but may also affect other genes, as may be concluded from the effects of MEA + Cys on IPTG induction of /3-galactosidase (N~islund et al., 1986) and on adaptation to MNU treatment in E. coli (N~islund et al., 1986). Radiobiological considerations

From the absence of a marked radioprotective effect of MEA in rec- strains, it might be anticipated that inducibility of SOS functions is in some way essential to the exertion of the radioprotective action of MEA. Since inactivation of the recA gene, as for example in rec- mutant strains (Hiilsewede and Schulte-Frohlinde, 1986), leads to increased radiosensitivity, there would appear to be no simple relationship between recA induction and radiosensitivity. An important consequence of the possible demonstration of a role of the SOS system in the radioprotective action of MEA is the possibility that a similar situation may be found to exist in eukaryotic cells or organisms, in which MEA has been shown to exert radioprotective effects that can also be counteracted by ascorbate (Forsberg et al., 1978). The failure of explanations based on physico-chemical mechanisms to account for the action of MEA in either prokaryotic or eukaryotic cells follows from reaction-kinetic data for proton donation and free-radical scavenging (Hi~lsewede and Schulte-Frohlinde, 1986). The radioprotective action of MEA thus lends support to other suggestions that there may be an analogue of the bacterial rec system in higher organisms (Rossman and Klein, 1985; cf. Maher et al., 1987); MEA may offer a useful tool for studies of such systems.

Acknowledgements The authors wish to thank Dr. G.M Weinstock (University of Texas Medical School, Houston, TX, USA) for generous gifts of bacterial strains. The work was supported financially by the Bank of Sweden Tercentenary Foundation.

207

References Alexander, P., and A. Charlesby (1955) Physico-chemical methods of protection against ionizing radiation, in: Z.M. Bacq and P. Alexander (Eds.), Radiobiology Symposium, Liege, Butterworth, London, pp. 49-59. Bacq, Z.M. (1965) Chemical Protection Against Ionizing Radiation, Charles C Thomas, Springfield, IL. Bacq, Z.M., A. Herv6, J. Lecomte, P. Fisher and J. Blavier (1951) Protection contre le rayonnement X par la /3mercapto6thylamine, Arch. Int. Physiol., 59, 442-447. Braams, R. (1963) A mechanism for the direct action of ionizing radiations, Nature (London), 200, 752-754. D'Ari, R. (1985) The SOS system, Biochimie, 67, 343-347. Defais, M. (1985) The adaptive response in E. coli, Biochimie, 67, 357-360. Ehrenberg, L., I. Fedorcs~k, M. Harms-Ringdahl and M. N~islund (1974) The role of H 2 0 2 in the reversible inhibition of RNA synthesis by thiols in E. coli, Acta Chem. Scand. B, 28, 960-962. Ehrenberg, L., M. Harms-Ringdahl, I. Fedorcsfik and F. Granath (1989) Kinetics of the copper- and iron-catalysed oxidation of cysteine by dioxygen, Acta Chem. Scand., 43, 177-187. Forsberg, J., M. Harms-Ringdahl and L. Ehrenberg (1978) Interaction of ascorbate with the radioprotective effect of mercaptoethylamine: an exploratory study in mice, whole animals and cell cultures, Int. J. Radiat. Biol., 34, 245-252. Howard-Flanders, P. (1960) Effect of oxygen on the radiosensitivity of bacteriophage in the presence of sulphydryl compounds, Nature (London), 186, 485-487. Hulsewede, J.W., and D. Schulte-Frohlinde (1986) Radiation protection of E. coli strains by cysteamine in the presence of oxygen, Int. J. Radiat. Biol., 50, 861-869. Laval, F. (1985) Repair of methylated bases in mammalian cells during adaptive response to alkylating agents, Biochimie, 67, 361-364.

Little, J.W., and D.W. Mount (1982) The SOS regulatory system of E. coli, Cell, 29, 11-22. Maher, V.M., K. Sato, S. Kateley-Kohler, H. Thomas, S. Michaud, J.J. McCormick, M. Kraemer, H.J. Rahmsdorf and P. Herrlich (1987) Evidence of inducible error-prone mechanisms in diploid human fibroblasts, in: R. Moses and W.C. Summers (Eds.), DNA Replication and Mutagenesis, ASM, Washington, DC, pp. 465-471. Miller, J.H. (1972) Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Nakken, K.F. (1965) Radical scavengers and radioprotection, in: M. Ebert and A. Howard (Eds.), Current Topics in Radiation Research, Vol. I, North Holland, Amsterdam, pp. 51-90. N~islund, M., and L. Ehrenberg (1978) Suppression of induced /3-galactosidase synthesis by cysteamine and its reversion by y-irradiation in the presence of ascorbate, Int. J. Radiat. Biol., 34, 411-415. N~islund, M., L. Ehrenberg and G. Djalali-Behzad (1976) Antagonism of ascorbate against the radioprotective action of cysteamine, Int. J. Radiat. Biol., 30, 95-99. N~islund, M., A. Kolman and L. Ehrenberg (1986) Enhancement by cysteamine of N-methyI-N-nitrosourea mutagenesis in E. coli, Mutation Res., 173, 163-168. Rossman, T.G., and C.B. Klein (1985) Mammalian SOS system: a case of misplaced analogies, Cancer Invest., 3, 175-187. Walker, G.C. (1984) Mutagenesis and inducible responses to deoxyribonucleic acid damage in E. coli, Microbiol. Rev., 48, 60-93. Weisemann, J.M., C. Funk and G.M. Weinstock (1984) Measurement of in vivo expression of the recA gene of Escherichia coli by using lacZ gene fusions, J. Bacteriol., 160, 112-121. Communicated by K. Sankaranarayanan