Mutation Research, 281 (1992) 137-141

137

© 1992 Elsevier Science Publishers B.V. All rights reserved 0165-7992/92/$05.00

MUTLET 00622

SOS system induction in Escherichia coli cells with distinct levels of ribonucleotide reductase activity Antonio Villaverde and Jordi Barb6 Department of Genetics and Microbiology, Autonomous University of Barcelona, Bellaterra, 08193-Barcelona (Spain) (Received 28 June 1991) (Revision received 8 October 1991) (Accepted 11 October 1991)

Keywords: SOS system; Ribonucleotide reductase genes; Escherichia coli

Summary The UV-mediated induction of recA and sfiA genes in Escherichia coli cells with distinct levels of dATP has been studied. Low levels of dATP were obtained by using either a temperature-sensitive ribonucleotide (RDP) reductase-deficient (nrdA) mutant or a wild-type strain treated with hydroxyurea. High pools of dATP were achieved by using a plasmid overproducing RDP reductase. The results obtained show that expression of the recA and sfiA genes was inhibited neither in the UV-irradiated nrdA mutant at 42°C nor in the wild-type strain in the presence of hydroxyurea. Likewise, the increase of the dATP pool did not enhance recA and sfiA gene expression after UV irradiation. All these data suggest that the basal level of dATP is not a limiting factor in the process of induction of the SOS system in Escherichia coli.

In Escherichia coli damage to DNA or arrest of the normal DNA replication induces the expression of a group of cellular activities involved in the process of cell survival which is known as the SOS system (Witkin, 1976). All of these SOS functions depend on the recA and lexA genes. The LexA protein is the common repressor of the SOS genes which include lexA and recA (Brent and Ptashne, 1981; Little et al., 1981). A model for regulation of the SOS network has been developed (Little and Mount, 1982; Walker, 1984): Correspondence: Dr. Jordi Barb6, Department of Genetics and Microbiology, Faculty of Sciences, Autonomous University of Barcelona, Bellaterra, 08193-Barcelona (Spain). Tel.: 343-5811837; Fax.: 343-5812003.

the RecA protein is activated by some signal and acquires some properties enabling it to catalyze the cleavage of the LexA protein repressing the recA gene. This derepression leads to the formation of large amounts of RecA protein, which, upon activation, causes the hydrolysis of more LexA proteins - which were repressing the target genes - and permits SOS expression. The characteristics of this inducing signal are still unknown, although it has been proposed that it would be single-stranded DNA regions which would originate after DNA damage was processed during DNA replication (Sassanfar and Roberts, 1990). Activated RecA protease is also able to cleave UmuD protein and many bacteriophage repressors (Walker, 1984; Burckhard et al., 1988; Nohmi

138

et al., 1988; Shinagawa et al., 1988). It has also been described that LexA protein may self-cleave in vitro under some conditions (Little, 1984). Thus, it is not established whether in vivoactivated RecA protein really hydrolyzes the LexA repressor or only facilitates self-cleavage of the LexA protein. The logic of the SOS circuit is not affected by any of these possibilities. In a DNAdamaged surviving cell, DNA repair decreases generation of the inducing signal so that the protease level drops. As a result, repressors accumulate and all induced genes become repressed again. It has been shown in E. coli that UV-irradiation induces about a 2-fold increase in the ATP concentration during the first 20 min (Barb6 et aI., 1983). In RecA + strains, the ATP concentration quickly recovers the same value as shown by non-irradiated ceils (Barb6 et al., 1983, 1986). This hydrolysis of ATP is related with the RecA protease-mediated LexA repressor cleavage, because neither recA (Def) nor lexA ( I n d ) mutants show any decrease in ATP level. In vitro, the RecA protease is more active using dATP than ATP as the cofactor (Phizicky and Roberts, 1981). Thus, it seems possible that dATP might be a limiting factor during the expression of the SOS response in vivo. To examine this possibility, we studied the induction of two SOS genes (recA

and sfiA), as well as the recA-dependent changes in the ATP concentration in UV-irradiated cells of E. coli in 2 different physiological conditions: (i) inhibiting the ribonucleotide (RDP) reductase enzyme activity, and (ii) enhancing the intracellular concentration of RDP reductase. Two procedures were employed to decrease RDP reductase activity: a temperature-sensitive RDP reductase (nrdA) mutant (which immediately blocks the synthesis of deoxyribonucleotides) was shifted to 42°C (Fuchs et al., 1972), and wild-type cells were treated with hydroxyurea, which inhibits ribonucleotide reductase activity (Sinha and Snustad, 1972). RDP reductase activity was enhanced by using the pPS2 plasmid which overproduces this enzyme increasing the intracellular concentration of dATP about 4-fold (Platz et al., 1985). Our results show that these alterations in RDP reductase activity affected neither the normal pattern of ATP evolution nor the induction of the SOS system after UV irradiation. Material and methods

Bacterial and bacteriophage strains, media and growth conditions The bacterial strains used in this work are listed in Table 1. Derivatives of mutant nrdA containing the ArecA306 allele were constructed

TABLE 1 STRAINS OF Escherichia coil K-12 USED 1N THIS STUDY Strain

Relevant genotype

Source or reference

CR34 El01 EST1515 UA4202

wild type as CR34, but nrdAlO1 Arecv4306 srl:: TnlO as CR34, but ArecA306 srl:: TnlO

UA4203

as El01, but ArecA306 srl:: TnlO

Fuchs et al. (1972) Fuchs et al. (1972) Tessman and Peterson (1985) Tc r ArecA306 transductant of CR34, donor EST1515 Tc r ArecA306 transductant of El01, donor EST1515

GC2375

wild type but lysogenic for (A(recA :: lacZ)cl ind) wild type but lysogenic for ( A( sfiA :: lacZ )cl ind) as CR34, but lysogenic for (A(sf/A ::lacZ)cI ind) as El01, but lysogenic for (A(~f/A :: lacZ )cl ind)

GY4786 UA4273 UA4289

Casaregola et al. (1982) Huisman and D'Ari (1983) This work This work

139

by P l k c transduction using the EST1515 ArecA306 srl:: TnlO as a donor. P1 transductions were performed basically as described by Miller (1972). Presence of the ArecA306 allele in the Tc r transductant clones was detected by testing its sensitivity to 2 / x g / m l of nitrofurantoin. Afterwards, the response to the UV irradiation of those transductants which were nitrofurantoinsensitive was checked. Cultures were grown at 30°C or 42°C with shaking in liquid minimal medium AB (Clark and Maaloe, 1967) supplemented with thiamine (10/xg/ml), glucose (0.4% w/v), thymidine (25 /xg/ml) and casamino acids (0.4% w/v). When necessary, LB (Miller, 1972) was used as a rich medium. Hydroxyurea, nucleosides, nucleotides and firefly luciferin-luciferase assays were obtained from Sigma; bases, vitamins and mineral salts were purchased from Merck; casamino acids, tryptone, yeast extract and agar were provided by Oxoid.

Lysogen&ation of bacterial strains Lysogens for Ad(recA::lacZ)cI ind were obtained as described (Barb6 et al., 1985). Lysogens for Ad(sfiA::lacZ)cI ind were constructed by spotting the phage on a bacterial lawn, incubating overnight at 30°C, and then plating bacteria from the spot on LB-glucose X-Gal plates at 30°C. After one night, lysogens formed pale blue colonies which were subsequently isolated and tested for the presence of the sfiA :: lacZ fusion. UV irradiation of bacterial suspensions Cells were grown to exponential phase (2 × 108 cells/ml) in supplemented AB medium, and irradiated in a glass petri dish (10 cm diameter). After irradiation they were centrifuged and resuspended in the same volume of supplemented AB medium. Irradiation of cell suspensions was carried out with constant shaking in a layer less than 1 mm thick. All procedures were performed under yellow light or in the dark to prevent photoreactivation. Determination of the intracellular A TP concentration and 13-galactosidase assay The intracellular ATP concentration was measured by the firefly luciferin-luciferase assay as previously described (Barb6 et al., 1983). The

/?-galactosidase assay was performed as previously reported (Barb6 et al., 1985).

DNA techniques Plasmid DNA was isolated by the alkaline extraction procedure of Birnboim and Doly (1979). This procedure was followed by CsCI/ ethidium bromide isopycnic centrifugation. Plasmid DNA transformation was carried out as described (Maniatis et al., 1982). Results and discussion

Fig. 1A shows that cells of the nrdA mutant growing at 42°C presented, as the wild-type strain, a 2-fold increase in the cellular ATP concentration during the first 20 min following UV irradiation, and regardless of whether the UV treatment was applied at the same moment or 1 h after the shift to 42°C. Afterwards, the ATP level dropped to about the basal level. This decrease in ATP concentration in the UV-irradiated nrdA cells cultured at 42°C is recA-dependent because a

®

0_ I--

g 0/

15 o t.D

0a nr

0

c0

80

120 Time

0

z,0

80

120

(•in)

Fig. 1. (A) Response of ATP concentration in the NrdA RecA ÷ (©, z~) and N r d A - RecA (13, v ) strains growing at 42°C and following UV irradiation at 20 J / m 2. UV irradiation was performed at the moment of temperature shift ( o , [] ) or 60 min after ( A, • ) . Changes of cellular ATP concentration of the parental strains NrdA + RecA + (o) and NrdA + R e c A - ( • ) at 42°C and after UV irradiation at 20 J / m 2 are shown as a control. (B) Response of ATP concentration in the RecA ÷ (©,o, A ) a n d RecA (13, • , v ) s t r a i n s at 37°Cand following UV irradiation at 20 J / m 2 in the absence (open symbols) or the presence (closed symbols) of hydroxyurea at 0.01 M. UV irradiation was performed at the moment of the addition of hydroxyurea (o, • ) or 60 min after (A, v). Values refer to the level of specific concentration of ATP which each strain presents without any treatment.

140

600

®

"~- 500 400 300

200 o

100

i i

'

~o

'

go

'

~ko

i

40

i

i

80

i

i

120

Time (rain) Fig. 2. (A) Induction of sfiA gene expression, indicated by increased /3-ga[actosidase, in cultures of the NrdA mutant growing at 42°C and following UV irradiation at 20 J/m 2. UV irradiation was performed at the moment of temperature shift (©) or 60 min after (o). Expression of the sfiA gene in the NrdA strain growing at either 30°C ( v ) or 42°C ( • ) without any treatment is shown as a control. (B) Induction of sfiA gene expression, indicated by increased /3-galactosidase, in cultures of wild-type strain GY4786 following UV irradiation in the presence (11, • ) or the absence ( [ ] ) of hydroxyurea at 0.01 M. UV irradiation was performed at the moment of the addition of hydroxyurea (11) or 60 rain after ( • ). Expression of the sfiA gene in strain GY4786 treated with hydroxyurea but not UV-irradiated is also shown as control ( zx).

double niutant nrdA zlrecA also showed an increase in the cellular ATP, which did not decrease after UV irradiation, even at 42°C. Likewise, the presence of hydroxyurea at 0.01 M did not modify the A T P changes of the UV-irradiated RecA + and R e c A - strains even when the U V treatment was performed 1 h after the addition of the hydroxyurea (Fig. 1B). Furthermore, U V irradiation of both the R e c A + strain treated with hydroxyurea and the nrdA mutant growing at 42°C triggers the expression of sfiA (Fig. 2) and recA (data not shown) genes, showing that the inhibition of the R D P reductase activity does not block the UV-mediated induction of the SOS system even when this is displayed 1 h after the enzymatic activity has been abolished. It is worth noting that non-treated cells of the nrdA mutant incubated at 42°C showed a weak expression of the sfiA gene, in agreement with previous results indicating that this mutant at the restrictive temperature presents a recA-dependent inhibition of cell division

(Taschner et al., 1987). In a similar way, cultures of the wild-type strain treated with 0.01 M of hydroxyurea also presented a slight induction of the sfiA gene, as expected since this compound is able to induce the SOS response at higher concentrations (Barb6 et al., 1987). However, the kinetics of the UV-mediated induction of the sfiA gene was the same regardless of whether the cells were or were not previously incubated at 42°C, or treated with hydroxyurea (Fig. 2). Finally, the effect on SOS expression was studied of a high cellular concentration of dATP, because of the presence of the pPS2 plasmid overproducing R D P reductase (Platz et al., 1985). Data obtained show that cells harboring pPS2 present the same evolution of cellular A T P and induction of both recA and sfiA genes as those strains without this plasmid (data not shown). On the other hand, it has been shown that the increase in the intracellular A T P level by the addition of adenine gives rise to a stronger stimulation in both recA441 and D N A damage-mediated induction of the SOS system (Llagostera et al., 1985). Likewise, depletion of cellular ATP pools by either cytidine or guanosine dramatically decreases the expression of the SOS response in vivo (Llagostera et al., 1985). So, whereas cellular A T P variations affect induction of the SOS network, cellular dATP changes do not. All data reported here lead us to suggest that the cellular pool of dATP is not a limiting factor in the recA-mediated cleavage of the LexA repressor. This may be due to the fact that A T P may be able to substitute dATP in vivo, despite this last nucleotide being more effective in vitro. In this respect, it has been shown that various nucleoside triphosphate (NTP) species show different efficiencies in promoting R e c A protease activity in vitro (Phizicky and Roberts, 1981; Weinstock and McEntee, 1981). For instance, for lambda repressor cleavage by the RecA441 mutant protein, dATP and A T P are by far the most effective; dUTP, UTP, dCTP and CTP are about 10, 20, 100 and 1000 times less effective than dATP, respectively; d G T P and G T P promote only trace amounts of R e c A protease activity; and T T P promotes non-detectable R e c A protease activity. Furthermore, it has also been reported that activation of protease-constitutive R e c A proteins

141

may be carried out by several nucleoside triphosphates (Wang et al., 1988). Acknowledgements

We thank Drs. M. Blanco, A. Jimenez and B.M. Sjoberg for their generous gifts of several strains and plasmids, and J.M. Cuartero for drawing the figures. This work was supported by Grant No. PB88-0246 of the Comision Interministerial de Ciencia y Tecnologia (CICYT), Spain. References Barb& J., R. Guerrero and A. Villaverde (1983) Evolution of cellular ATP concentration after UV-mediated induction of SOS system in Escherichia coli, Biochem. Biophys. Res. Commun., 117, 556-561. Barb& J., J.A. Vericat, J. Cairo and R. Guerrero (1985) Further characterization of SOS system induction in recBC mutants of Escherichiu co/i, Mutation Res., 146, 23-32. Barb& J., A. Villaverde, J. Cairo and R. Guerrero (1986) ATP hydrolysis during SOS induction in Escherichia coli, J. Bacterial., 167, 1055-1057. Barb& J., A. Villaverde and R. Guerrero (1987) Induction of the SOS response by hydroxyurea in Escherichiu coli K12, Mutation Res., 192, 105-108. Birnboim, H.C., and J. Doly (1979) A rapid alkaline extraction procedure for screening recombinant plasmid DNA, Nucleic Acids Res., 7, 1513-1524. Burckhardt, S.E., R. Woodgate, R.H. Scheuerman and H. Echols (1988) UmuD mutagenesis protein of Escherichia coli: overproduction, purification and cleavage by RecA, Proc. Nat]. Acad. Sci. (U.S.A.), 85, 1811-1815. Brent, R., and M. Ptashne (1981) Mechanism of action of the IexA gene product, Proc. Natl. Acad. Sci. (U.S.A.), 76, 4202-4208. Casaregola, S., R. D’Ari and 0. Huisman (1982) Quantitative evaluation of recA gene expression in Escherichiu coli, Mol. Gen. Genet., 185, 430-439. Clark, D.J., and 0. Maalae (1967) DNA replication and the division cycle of Escherichiu coli, J. Mol. Biol., 23, 99-112. Fuchs, J.A., H.O. Karlstrom, H.R. Warner and P. Reichard (1972) Defective gene product in dnaF mutant of Escherichia coli, Nature New Biol., 238, 69-71. Huisman, O., and R. D’Ari (1983) Effect of suppressors of SOS-mediated filamentation on $4 operon expression in Escherichia coli, J. Bacterial., 153, 169-175. Little, J.W. (1984) Autodigestion of 1exA and phage A repressor, Proc. Natl. Acad. Sci. (U.S.A.), 58, 1903-1910. Little, J.W., and D.W. Mount (1982) The SOS regulatory system of Escherichiu co/i, Cell, 29, 11-22. Little, J.W., D.W. Mount and C.R. Yanisch-Perron (1981) Purified IexA protein is a repressor of the recA and lexA genes, Proc. Natl. Acad. Sci. (U.S.A.), 78, 4199-4203.

Llagostera, M., R. Guerrero, A. Villaverde and J. BarbC (1985) Effect of adenine, cytidine and guanosine on the expression of the SOS system in Escherichia co/i, J. Gen. Microbial., 131, 113-118. Maniatis, T., E.F. Fritsch and S. Sambrook (1982) Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Miller, J.H. (1972) Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Nohmi, T., J.R. Battista, L.A. Dodson and G.C. Walker (1988) RecA-mediated cleavage activates UmuD for mutagenesis: mechanistic relationship between transcriptional derepression and posttranslational activation, Proc. Natl. Acad. Sci. (U.S.A.), 85, 1816-1820. Phizicky, E.M., and J.W. Roberts (1981) Induction of SOS functions: regulation of proteolytic activity of E. coli RecA protein by interaction with DNA and nucleoside triphosphate, Cell, 25, 259-267. Platz, A., M. Karlsoon, S. Hahne, S. Eriksson and B.M. Sjoberg (1985) Alterations in intracellular deoxyribonucleotide levels of mutationally altered ribonucleotide reductases in Escherichia coli, J. Bacterial., 164, 1194-1199. Sassanfar, M., and J.W. Roberts (1990) Nature of the SOS-inducing signal in Escherichiu coli. The involvement of DNA replication, J. Mol. Biol., 212, 79-96. Shinagawa, H., H. Iwasaki, T. Kato and A. Nakata (1988) RecA protein-dependent cleavage of UmuD protein and SOS mutagenesis, Proc. Nat]. Acad. Sci. (U.S.A.), 85, 1806-1810. Sinha, N., and P. Snustad (1972) Mechanism of inhibition of deoxyribonucleic acid synthesis in Escherichiu co/i by hydroxyurea, J. Bacterial., 112, 1321-1334. Taschner, P.E., J.G.J. Verest and C.L. Woldringh (1987) Genetic and morphological characterization of ftsB and nrdB mutants of Escherichiu coli, J. Bacterial., 169, 19-25. Tessman, E.S., and P. Peterson (1985) Plaque color method for rapid isolation of novel recA mutants of Escherichiu coli K-12: new classes of protease-constitutive recA mutants, J. Bacterial., 163, 677-687. Walker, G.C. (1984) Mutagenesis and inducible responses to deoxyribonucleic acid damage in Escherichiu co/i, Microbiol. Rev., 48, 60-93. Wang, W.B., M. Sassanfar, 1. Tessman, J.W. Roberts and E.S. Tessman (1988) Activation of protease-constitutive recA proteins of Escherichiu cob by all of the common nucleoside triphosphates, J. Bacterial., 170, 4816-4822. Weinstock, G.M., and K. McEntee (1981) RecA protein-dependent proteolysis of bacteriophage lambda repressor. Characterization of the reaction and stimulation by DNAbinding proteins, J. Biol. Chem., 256, 10883-10888. Witkin, E.M. (1976) Ultraviolet mutagenesis and inducible DNA repair in Escherichiu cob, Bacterial. Rev., 40, 869907.

Communicated

by F.H. Sobels