Journal of Photochemistry and Photobiology, B.• Biology, 6 (1990) 283-289

283

PATTERNS OF PROTEIN SYNTHESIS IN A GROWTHDELAY MUTANT (nut,) OF Escherichia coli AFTERTREATMENT BY NEAR-UV RADIATION OR HYDROGENPEROXIDE

J. HOERTERDepartment of Natural Sciences, Stephens College, Columbia, MO 65215 (USA.)

A. EISENSTARKtDivision of Biological Sciences, University of Missouri, Columbia, MO 65211 (U.S.A .)

(Received July 7, 1989 ; accepted December 5, 1989)

Keywords . Near-UV radiation, hydrogen peroxide, protein induction .

Summary

When Escherichia coli cells are stressed by hydrogen peroxide (H202),synthesis of a large number of proteins is repressed, while several otherproteins are induced. Since there is evidence that some lethal effects of near-UV (NUV) radiation may be directly or indirectly due to hydrogen peroxidegenerated by NUV light, treatment of cells with NUV radiation or H 202 mightbe expected to repress and induce the same set of proteins . In this study,we compared the effects of H202 and NUV irradiation on patterns of proteininduction and/or repression which were separate from the 4-thiouridine-dependent response using growth delay mutants (nuv) . Concentrating initiallyon the proteins that ceased synthesis following NUV irradiation in an nuvmutant, we observed that these were not the same as those that ceasedsynthesis following H202 treatment. Inspection of two-dimensional poly-acrylamide gel electrophoresis proteins indicated that NUV irradiation re-pressed synthesis of a different set of proteins, although there was someoverlap between the two (45%) . It was also observed that the new proteinswhich appeared after each of the two treatments were different . This suggeststhat the induction and/or repression of new proteins following NUV irradiationis not triggered solely via oxidative stress, although there is some overlapbetween the proteins that are induced or repressed following the two treat-ments,

tAuthor to whom correspondence should be addressed .

1011-1344/90/$3 .50

C Elsevier Sequoia/Printed in The Netherlands

284

1. Introduction

Previous studies suggest that near-UV (NUV) light exerts toxic effectson Escherichia coli via oxidative mechanisms [1] . At high fluence rates,hydrogen peroxide may be a photoproduct of NUV irradiation and play arole in creating unique DNA lesions [2] . However, studies involving lowerfluence rates suggest that H 2O2 itself is not an important agent leading totoxicity [3] . Some studies suggest a common pathway for protection againstdamage by NUV light or H 2O2 [4, 5]. This study examines the patterns ofprotein induction and/or repression in E. coli immediately after treatmentby NUV light or H 202 to determine to what extent these two stresses inducethe same response involved in recovery and repair .

Wild-type E. coli cells, which contain 4-thiouridine in tRNA, undergogrowth delay following NUV irradiation [ 1 ] and thus do not readily incorporateradioactive amino acids during the first 20 min after treatment . This obstaclewas overcome by using a mutant (nuv) which lacks the 4-thiouridine in thetRNA [6-8] and, consequently, shows no growth delay following NUV stress .However, previous studies using nuv mutants in Salmonella typhimuriumsuggest that there are 4-thiouridine-dependent and 4-thiouridine-independentmechanisms which act in regulating synthesis of proteins necessary forresistance to NUV irradiation [9] . Therefore by using the nuv mutant in thisstudy, we can compare the effects of H2O2 and NUV irradiation on thesynthesis of proteins which are separate from those induced and/or repressedby 4-thiouridine, and can overcome the effects of growth delay on [35S]incorporation .

2 . Experimental procedure

2.1. StrainsE. coil nuv strain (4Srd- ) was kindly supplied by Dr . J. Jagger, and

was used in all the experiments .

2.2. Media and growth conditionsOvernight suspensions of cells grown in Luria broth were diluted 1 :100

in M9 buffer supplemented with 0 .4% glucose and 1% casamino acids, andincubated at 37 °C with continuous shaking to an optical density (OD) 500of 0.2 . Cells were also irradiated in this medium. After NUV treatment,survival rates were determined by diluting in M9 and plating cells on Luriaagar plates. Colonies were counted after overnight incubation at 37 °C .

2.3. NUV irradiation or H202 treatmentFor NUV irradiation mid-log cell suspensions (5 ml) were irradiated

(4X 10 3 J rn-2) at 37 °C in Pyrex test-tubes (Pyrex eliminates wavelengthsbelow 290 nm) by eight Sylvania F 15T8BL lamps as described previously[10] . Cells were continuously aerated during the exposure. For H202 exposure,

285

mid-log cell suspensions were treated with 60 pM H 202. Both treatmentsresulted in 85% survival .

2.4. Radioactive labelingCells were pulse-labeled with L-[35S]-methionine (200 µCi ml- ') at 10

min intervals up to 50 min after NUV or H 2 02 exposure. Control cells werelabeled continuously for 20 min . Labeling was terminated by the additionof 10 µl of 0.1 M L-methionine .

2.5. Two-dimensional electrophoresisProteins were separated by two-dimensional (2-D) polyacrylamide gel

electrophoresis [11] . The isoelectric focusing gels were prepared using amixture of 2.2% pH 4-6 .5 ampholines and 2 .5% pH 5-8 ampholines . Eachexperiment was repeated three times and gels were prepared in duplicatefor each time period. Autoradiograms were carefully compared for consistentqualitative and quantitative differences in proteins using an LKB vetrascanXL (laser densitometer, coupled to an IBM computer) and were then assignedalphanumeric designations [12] .

3 . Results

A comparative analysis of the protein patterns in NUV- and H 202-treatedE. coli nuv cells revealed that there were major differences in proteinssynthesized during the first 10 min after each treatment . This analysisrepresents a composite summary of those proteins completely repressed ornewly induced on six separate gels from three independent experiments .The differences in protein patterns were greatest 10 min after each treatment .Each treatment continued to show different patterns of protein synthesis andrepression up to 30 min ; within 45 min after treatment protein patternswere almost identical for both treatments .

Repression of over 222 proteins (approximately 44% of the 501 proteinsclearly separated on 2-D gels in non-treated controls) occurred immediatelyafter NUV treatment (Fig . 1). In H2O2-treated E . coli nuv cells, repressionof over 396 proteins (approximately 78% of the 501 proteins clearly separatedon 2-D gels in non-treated controls) occurred immediately after H 202 treatment(Fig. 2). A comparative analysis of the 426 proteins repressed, revealed that192 (45%) of these proteins were common to both NUV and H2O2 treatments .

A comparative analysis of only those proteins that continued to besynthesized following NUV or H 202 treatment revealed that 205 proteins(64%) were unique to NUV, 40 proteins (13%) were unique to H2O,, and75 proteins (23%) were common to both treatments .

Ten new proteins were induced following NUV treatment (alphanumericdesignations : B34, C38, C42a, C42b, C45, D30, E22, G24, G102 and H98)and three new proteins were induced following H 202 treatment (alphanumericdesignations : A64, B30, H26) . These proteins were not synthesized in non-

286

Fig . 1 . Proteins separated by two-dimensional electrophoresis in E . coli nuv cells during 0-10min after NUV treatment and in non-treated controls . Exponentially growing cells were labeledwith 1 25S]-methionine for 10 min immediately following 15 min exposure to NUV light. Cellswere lysed and electrophoresed on an equilibrium isoelectric focusing gel followed by a 10%polyacrylamide-sodium dodecyl sulfate gel in the second dimension . Molecular sizes (kDa)for the standards are indicated in the center . Isoelectric zones (A-H ; acid to base) wereassigned as described by Phillips et at. [13]- New induced proteins are circled .

treated control cells at levels detectable on 2-D gels and were unique toeach treatment. None of the induced proteins correlated directly with thoseidentified in the gene protein index of E . coli (K-12 112], or with criticalenzymes known thus far to be involved in NUV and H202 stress and recovery .However, several known proteins in the index have coordinates that are verysimilar: DNA-directed DNA polymerase III-B-subunit (B36 .1), RecA (C39.3),phosphogluconate dehydrogenase (C42 .6), methionine adenosyltransferase(C44 .6), HTP protein (D33.4), acetate CoA-transferase (G24 .2) and DnaJ(H26.5) .

4. Discussion

Despite numerous observations which suggest that some NUV and H 202effects may be interchangeable, our analysis of protein synthesis in cells

287

Fig. 2 . Proteins separated by two-dimensional electrophoresis in E . coli nuv cells treated with60 µM H2O, and in non-treated controls. Exponentially growing cells were labeled with ['sS]-methionine for 10 min immediately following treatment with H2O,. Cells were lysed andelectrophoresed on an equilibrium isoelectric focusing gel followed by a 10% polyacryla-mide-sodium dodecyl sulfate gel in the second dimension . Molecular sizes (kDa) for thestandards are indicated in the center. Isoelectric zones (A-H ; acid to base) were assigned asdescribed by Phillips et al . [131. New induced proteins are circled .

after NUV or H 202 treatment did not entirely support this view . Repressionof protein synthesis occurred to a greater degree in H 202 (78%) than inNUV light (44%) . If NUV and H202 effects were interchangeable and directlyrelated to each other, we would expect that H 202 and NUV light wouldrepress and induce the same set of proteins . Although we observed someoverlap in the proteins repressed, major differences existed . In addition, thenumber and kinds of new proteins induced after each of the treatments weredifferent. This suggests that NUV light may not produce its effects on thecell solely through the generation of H202 and subsequent conversion toOz- . The lack of similarity between proteins induced by NUV light and thoseinduced by oxidative stress (methyl viologen) in superoxide dismutase (sodAsodB) double mutants [14] also suggests that NUV light does not inflictinsults on cells solely through the generation of high levels of 02 and doesnot induce the superoxide inducible (soi) response .

The observation that of all the proteins repressed, 45% were commonto both H2O, and NUV light, suggests that there may be a common intracellular

288

signal generated by both NUV and H202 stress, manifesting itself as anoverlap of protein repression for a short time until the cell initiates thenecessary metabolic activities required for recovery from the stress . Themetabolic pathways associated with recovery from NUV light may be quitedifferent from those associated with recovery from H202, as indicated bythe continued synthesis of 280 proteins in cells treated with NUV lightcompared with 115 proteins in cells stressed with H202, and by the lack ofsimilarity of the new proteins induced. However, some interrelationshipsexist in the recovery pathways as shown by the fact that 23% of the proteinsthat continued synthesis were common to both NUV and H202 stress .

Studies using nuv mutants (lacking 4-thiouridine) in S. typhimuriumalso show similarities and differences in patterns of protein induction afteroxidative stress or NUV irradiation compared with wild-type cells, suggestingthat 4-thiouridine in the tRNA may play a key role in triggering (via adenylateddinucleotides) the synthesis of proteins necessary for resistance to NUVirradiation [9 ] . Therefore our study using the nuv mutant in E. coli suggeststhat there are proteins involved in protection (separate from those triggeredby 4-thiouridine) which are different for NUV and H 202 stress .

This study was restricted to a comparison of proteins repressed and/orsynthesized early after treatment with NUV light of H 202 in order to distinguishthe `oxidative stress" component of NUV light . It is recognized that theremay be unique NUV (non-oxidation) effects or photoreceptors in membranesand other cellular components, and these may explain some of the differencesin protein patterns .

Acknowledgments

This investigation was supported in part by the National Science Foun-dation (DMS-85027-08), the National Institute of Environmental Health (ESO4889h) and the University of Missouri Institutional Biomedical ResearchSupport Grant RFR 07053 (N .I .H.) . We thank John Jagger for the nuv(4Srd -) strain of E . coli .

References

1 J. Jagger, Solar-UV Actions On Living Cells, Praeger, New York, 1985 .2 H. Ananthaswamy and A . Eisenstark, J. Bacteriol., 130 (1977) 187-191 .3 G. R Kramer and B. N. Ames, J. Bacteriol ., 169 (1987) 2259-2266 .4 R. M. Tyrrell, Mutat. Res., 145 (1985) 129-136 .5 L. J. Sammartano and R. W. Tuveson, Photochem. Photobiol ., 41 (1985) 367-370 .6 G. Thomas and A . Favre, Eur. J. Biochem., 113 (1980) 67-74.7 S. Tsai and J . Jagger, Photochem . PhotobioL, 33 (1981) 825-834 .8 E. Hajnsdorf and A . Favre, Photochem. PhotobioL, 43 (1986) 157-164 .9 G. F. Kramer, J. C. Baker and B . N. Ames, J. Bacteriol., 170 (1988) 2344-2354 .10 H. Ananthaswamy and A. Eisenstark, Photochem. Photobiol ., 24 (1976) 390-442 .

289

11 P . O'Farrell, Biol. Chem., 259 (1975) 4007-4021 .12 F . C . Neidhardt, V . Vaughn, T. A. Phillips and P. L. Bloch, Microbiol. Rev., 47 (1983)

231-284 .13 T . A. Phillips, V . Vaughn, P . L. Bloch and F. C. Neidhardt, in F. C. Neidhardt (ed .),

Escherichia coli and Salmonella typhimurium Cellular and MolecularBiology, AmericanSociety for Microbiology, Washington DC, 1987, pp. 919-979 .

14 L. K . Walkup and T. Kogoma, J. Bacteriol., 171 (1989) 1476-1484.