Suppression of the pleiotropic effects of HisH and HisF overproduction identifies four novel loci on the Salmonella typhimurium chromosome: osmH, sfiW, sfiX, and sfiY

JOURNAL OF BACTERIOLOGY, Sept. 1995, p. 4841–4850 Vol. 177, No. 17 0021-9193/95/$04.0010 Copyright q1995, American Society for Microbiology Suppression of the Pleiotropic Effects of HisH and HisF Overproduction Identifies Four Novel Loci on the Salmonella typhimurium Chromosome: osmH,sfiW,sfiX, and sfiY AMANDO FLORES AND JOSEP CASADESU ´S* Departamento de Gene´tica, Universidad de Sevilla, Seville 41080, Spain Received 30 March 1995/Accepted 19 June 1995 Insertion mutations that suppress some or all the pleiotropic effects of HisH and HisF overproduction were obtained by using transposons Tn10dTet and Tn10dCam. All suppressor mutations proved to be recessive, indicating that their effects were caused by loss of function; thus, the suppressors identify genes that are necessary to trigger the pleiotropic response when HisH and HisF are overproduced. Genetic mapping of the suppressor mutations identifies four novel loci on the Salmonella typhimurium genetic map. Mutations in osmH (min 49) behave as general suppressors that abolish all manifestations of the pleiotropic response. Mutations in sfiY (min 83) suppress cell division inhibition and thermosensitivity but not osmosensitivity. Mutations that suppress only cell division inhibition define another locus, sfiX (min 44). A fourth novel locus, sfiW (min 19), is also involved in cell division inhibition. The phenotype of sfiW mutations is in turn pleiotropic: they suppress cell division inhibition, make S. typhimurium unable to grow in minimal media, and cause slow growth and abnormal colony and cell shape. The inability of sfiW mutants to grow in minimal medium cannot be relieved by any known nutritional requirement or by the use of carbon sources other than glucose. The hierarchy of suppressor phenotypes and the existence of epistatic effects among suppressor mutations suggest a pathwaylike model for the His c pleiotropic response. Constitutive expression of the Salmonella typhimurium histidine operon causes growth inhibition at 428C (24), wrinkled morphology of colonies grown in either 2% glucose or ‘‘green’’ plates (28, 42), and growth inhibition on plates containing three to five times the normal concentration of E salts (12, 37). The wrinkled-colony morphology is due to cell filamentation which is also observed in liquid cultures, especially if grown with a high concentration of glucose (25, 28, 42). This filamentation is sulA independent and unrelated to the SOS response (28). The His c pleiotropic response is caused by increased levels of proteins HisH and HisF (12, 25, 42). In the close relative Escherichia coli, a similar response is observed when HisH and HisF are overproduced (27). HisH and HisF catalyze the release of 5-amino-4-imidazole carboxamide riboside 59-monophosphate (AICAR), which is also an intermediate in purine biosynthesis (43, 44, 53). Since AICAR is a potential precursor of the alarmone 5-amino-4-imidazole carboxamide riboside 59triphosphate (9, 46), several authors had proposed that increased synthesis of AICAR might be responsible for the pleiotropic effects of his overexpression (11, 22). However, AICAR does not seem to be involved in the pleiotropic response (25, 26). We have addressed the identification of genes involved in the His c pleiotropic response by the classical approach of studying intergenic (external) suppression (30). Mutations that suppress some or all the manifestations of the pleiotropic response identify four novel loci on the S. typhimurium chromosome: osmH (min 49), sfiW (min 19), sfiX (min 44), and sfiY (min 83). The hierarchy of suppressor phenotypes and the existence of epistatic effects in certain suppressor combinations suggest that the pleiotropic response of His c mutants involves a complex network or pathway whose ultimate physiological significance remains unknown. MATERIALS AND METHODS Bacterial strains, bacteriophages, and strain construction. The S. typhimurium strains used in this study, all derived from strain LT2, are listed in Table 1. Strain TR6753, carrying the attenuator deletion hisO1242, was used as the standard His c strain (12). Strain hisD3050, obtained from J. R. Roth (Department of Biology, University of Utah, Salt Lake City), carries a complete deletion of the histidine operon. Transductional crosses using phage P22 HT 105/1 int201 (44a, 49) were used for strain construction; the transducing phage will be henceforth referred as P22 HT. Phage sensitivity was tested by cross-streaking with the clear-plaque mutant P22 H5. To obtain phage-free isolates, transductants were purified by streaking on green indicator plates. Plasmids and transposons. The episome F9128 pro 1 lac 1 zzf-1836::Tn10dCam was described by Elliott and Roth (23). pNK972 (Amp r ) is a pBR333 derivative carrying the IS10 transposase gene under the control of a tac promoter (41). pNK2880 contains the IS10 tnpA(ats1 ats2) gene expressed from a tac promoter; this plasmid contains a lacI q allele and confers ampicillin resistance (35). The ats1 and ats2 mutations confer relaxed specificity to Tn10 transposase (5). Tn10dTet, Tn10dCam, and Tn10dKan are deletion derivatives of transposon Tn10 that do not carry transposition functions (52). The E. coli episome F9128 pro 1 lac 1 zzf-1831::Tn10dTet and the S. typhimurium episome F9152 nad 1 zzf-1833::Tn10dKan were both obtained from J. R. Roth. Mud1-8(Amp Lac) (32) and MudI1734(Kan Lac) (13) are transposition-deficient Mu derivatives that generate operon fusions upon insertion. MudI1734(Kan Lac) (13) was renamed MudJ by Hughes and Roth (34). Media, chemicals, and culture conditions. The E medium of Vogel and Bonner (51) supplemented with 0.2% glucose was used as minimal medium. In certain experiments, glucose was replaced by 0.2% fumarate. High-salt (33E) plates contained three times the normal concentration of E salts (12). NCE medium is E medium without citrate. The rich medium was nutrient broth (NB; 8 g/liter; Difco) with added NaCl (5 g/liter). Solid media contained agar at a final concentration of 1.5%. Antibiotics were prepared and used as described by Maloy (38). For the selection of Tet s derivatives of Tet r strains, we used the medium of Bochner et al. (10) as modified by Maloy and Nunn (39). Green indicator plates were prepared as described by Chan et al. (14) except that methyl blue (Sigma) substituted for aniline blue. Nutritional supplements and auxanography pools were prepared as described by Davis et al. (19). Casamino Acids were used at concentrations of 0.2, 0.4, and 0.6%. Except when specifically indicated, cultures were incubated at 378C. Liquid cultures were grown in a Gallenkamp orbital incubator at 160 rpm. * Corresponding author. Mailing address: Departamento de Gene´tica, Facultad de Biologı´a, Universidad de Sevilla, Apartado 1095, Sevilla 41080, Spain. Phone: 345-455-7105. Fax: 345-455-7104. Electronic mail address: [email protected]. 4841 on July 25, 2017 by INSTITUTO DE PARASITOLOGFA Yhttp://jb.asm.org/Downloaded from Mutagenesis with Tn10dCam and isolation of salt-resistant derivatives of a His c strain. We used a variant of the nonhomologous transduction procedure described by Elliott and Roth (23). A lysate grown on strain SV2056 was used to transduce strain SV2057, selecting Cam r Amp r transductants. Transducing mixtures were made on NB plates and preincubated for 4 to6hat378C before replica plating to NB supplemented with chloramphenicol and ampicillin. Use of a recipient carrying the deletion hisD3050 prevented the insertion of the miniTn10 in the histidine operon. Cam r Amp r transductants were replica plated more than three times to NB-chloramphenicol plates containing EGTA (10 mM); the latter was added to prevent reinfection (thus allowing the isolation of phage-free TABLE 1. S. typhimurium strains used Strain Genotype Reference or source a LT2 Wild type Laboratory stock PP1002 cya-1091::Tn10 trpB223 SGSC b SA4247 aroA554::Tn10 SGSC SMS409 aspC409::Tn10 SGSC SV2056 hisD3050/F9128 lac 1 pro 1 zzf-1836::Tn10dCam SV2057 hisD3050/pNK972 SV2058 hisO1242/F9128 lac 1 pro 1 zzf-1831::Tn10dTet SV2059 hisO1242/pNK2880 SV2060 hisO1242 osmH1::Tn10dCam SV2061 hisO1242 osmH2::Tn10dCam SV2064 hisO1242 osmH3::Tn10dTet SV2068 hisO1242 sfiW1::Tn10dTet SV2069 hisO1242 sfiX1::Tn10dTet SV2070 hisO1242 sfiX2::Tn10dTet SV2071 hisO1242 sfiY1::Tn10dTet SV2072 hisO1242 sfiX3::Tn10dTet SV2073 hisO1242 sfiW2::Tn10dTet SV2074 hisO1242 sfiY2::Tn10dTet SV2082 hisO1242 osmH4::Tn10dTet SV2090 hisO 1 hisC9968::MudJ SV2091 hisO1242 hisC9968::MudJ SV2092 hisO 1 hisC9968::MudJ osmH1::Tn10dCam SV2093 hisO1242 hisC9968::MudJ osmH1::Tn10dCam SV2129 hisO1242 osmH3::Tn10dCam sfiX2::Tn10dTet SV2130 hisO1242 osmH3::Tn10dCam sfiY1::Tn10dTet SV2131 hisO1242 osmH3::Tn10dCam sfiW1::Tn10dTet SV2132 hisO1242 sfiW1::Tn10dTet sfiX2::Tn10dKan SV2133 hisO1242 sfiW1::Tn10dTet sfiY1::Tn10dKan SV2134 hisO1242 sfiX2::Tn10dKan sfiY1::Tn10dTet SV2135 hisO1242 osmH3::Tn10dCam sfiW1::Tn10dTet sfiY1::Tn10dKan SV2136 hisO1242 osmH3::Tn10dCam sfiW1::Tn10dTet sfiX2::Tn10dKan SV2137 hisO 1 hisIE9969::Mud1-8 SV2138 hisO1242 hisIE9969::Mud1-8 SV2139 hisO 1 hisIE9969::Mud1-8 osmH1::Tn10dTet SV2140 hisO1242 hisIE9969::Mud1-8 osmH1::Tn10dTet SV2150 hisO1242 DUP [hisH9962 pMudPpcysA1586] c SV2155 hisO1242 DUP [purE2154 pMudPppurB1879] c TR6753 hisO1242 12 TT2242 aroD5 hisW1824 purF145 rpsL metG319 zee-78::Tn10 SGSC TT7294 zea-1032::Tn10 SGSC TT10423 proAB47/F9128 pro 1 lac 1 zzf-1836::Tn10dTet J. Roth TT10425 nadA56/F9152 nad 1 zzf-1833::Tn10dKan J. Roth TT10604 proAB47/F9128 pro 1 lac 1 zzf-1836::Tn10dCam J. Roth TT13736 eut-153::MudJ SGSC TT15222 leuA414 r 2 m 1 fels2 thr-458::MudQ N. Benson TT15225 leuA414 r 2 m 1 fels2 pyrA2413::MudP N. Benson TT15228 leuA414 r 2 m 1 fels2 nadC220::MudP N. Benson TT15233 leuA414 r 2 m 1 fels2 purE2155::MudP N. Benson TT15234 leuA414 r 2 m 1 fels2 purE2155::MudQ N. Benson TT15251 leuA414 r 2 m 1 fels2 hisH9950::MudP N. Benson TT15252 leuA414 r 2 m 1 fels2 hisH9950::MudP N. Benson TT15252 leuA414 r 2 m 1 fels2 aroC566::MudP N. Benson TT15253 leuA414 r 2 m 1 fels2 cysA1586::MudQ N. Benson TT15259 leuA414 r 2 m 1 fels2 nadB226::MudP N. Benson TT15260 leuA414 r 2 m 1 fels2 nadB226::MudQ N. Benson TT15262 leuA414 r 2 m 1 fels2 cysHIJ1574::MudQ N. Benson TT15269 leuA414 r 2 m 1 fels2 ilvA2648::MudP N. Benson TT15274 leuA414 r 2 m 1 fels2 purD1874::MudP N. Benson TT17163 leuA414 r 2 m 1 fels2 hisH9962::MudP N. Benson a Omitted for strains first described in this study. b SGSC, Salmonella Genetic Stock Centre, University of Calgary, Calgary, Alberta, Canada. c Nomenclature for duplications (DUP) follows the rules described in references 15 and 33. 4842 FLORES AND CASADESU ´S J. BACTERIOL. on July 25, 2017 by INSTITUTO DE PARASITOLOGFA Yhttp://jb.asm.org/Downloaded from derivatives). Pools of 1,000 to 2,000 Cam r colonies were made and lysed with phage P22 HT. The pools were then used to transduce strain TR6753; Cam r transductants were selected on 33E plates. The total number of Cam r isolates pooled was around 60,000. To guarantee independence, only one Cam r , saltresistant isolate per pool was further studied. Mutagenesis with Tn10dTet. A lysate grown on strain SV2058 was used to transduce strain SV2059; the latter contains plasmid pNK2880 (35) to permit trans complementation of the defective Tn10dTet element by ATS transposase. Transducing mixtures were directly made on NB plates and preincubated for 4 h at 378C. The cultures were then replica plated to NB plates supplemented with tetracycline and to appropriate media to detect suppression of the mutation hisO1242:33E plates to select salt-resistant derivatives, E plates incubated at 428C to select thermoresistant isolates, and green plates to detect the presence of nonfilamenting (smooth) colonies. The total number of Tet r isolates scored (all of independent origin) was around 110,000. Candidates were made phage free on green plates and used to transduce TR6753. This second transduction allowed the isolation of plasmid-free Tet r derivatives. Transposon substitutions. Lysates grown on strains TT10604, TT10423, and TT10425 were used for transposon substitutions. These strains carry F-prime plasmids containing Tn10dCam, Tn10dTet, and Tn10dKan (Table 1). The lysates were irradiated with UV light, using a 15-W Sylvania lamp at a distance of 30 cm during 30 s; irradiation of the phage suspensions can be expected to increase recombination in the recipients (45). Transductions selecting the incoming marker (Tet r , Cam r ,orKan r ) were carried out. Transductants were then scored for loss of the resident marker (Tet r or Cam r ). These procedures allowed us to obtain allele variants tagged with different antibiotic resistance genes. Strains carrying allele variants are not included in Table 1; the allele variants available are listed in Table 2. Rapid mapping with Mud-P22 prophages. We followed the procedure described by Benson and Goldman (6), using 67 ‘‘locked-in’’ Mud-P22 prophages. In addition to the 54-lysogen collection described in reference 6, we used 13 additional Mud-P22 lysogens (strains TT15222, TT15225, TT15228, TT15233, TT15234, TT15251, TT15252, TT15253, TT15259, TT15260, TT15262, TT15274, and TT17163, all obtained from Nick Benson, Department of Biology, University of Utah). These additional lysogens (but not the original collection described in reference 6) are included in Table 1. Cotransductional mapping. Upon transduction with phage P22 HT, transductants were replica plated to suitable plates to score transduction of unselected markers. Compatible marker combinations were made possible by using different versions (Tet r , Kan r ,orCam r ) of the alleles to be mapped. Cotransduction frequencies are averages of more than four independent crosses, scoring at least 300 transductants from each cross. The relative order of markers was determined by three-factor crosses. Cotransduction frequencies below 1% were not taken into account (38). Table 1 contains the original strains carrying markers cotransducible with the loci studied (TT7294, TT15253, SA4247, SMS409, TT2242, PP1002, and TT15269) but not their derivatives constructed by adding one of the alleles to be mapped; such constructions were required to perform certain threefactor crosses. Strains used in unsuccessful cotransductional mapping attempts are also omitted in Table 1. Complementation analysis. Complementation was achieved by the construction of merodiploid strains (2, 15, 31). Duplications with predetermined endpoints were obtained by using a strategy adapted from reference 33. b-Galactosidase assays. Levels of b-galactosidase were assayed as described by Miller (40), using the CHCl 3 -sodium dodecyl sulfate permeabilization procedure. To measure his operon expression by using transcriptional lac fusions at hisC and hisIE, cultures were grown in E medium containing 0.2% glucose and 0.1 M histidine; overnight cultures were diluted into fresh medium to reach an optical density at 540 nm (OD 540 ) of 0.4 to 0.6. Osmosensitivity assays in liquid medium. All assays were carried out at 378C with shaking. Cultures were preadapted for .8 h in E medium containing 0.3 M NaCl; the cultures were then diluted to obtain an initial OD 540 of 0.05. Growth was monitored by nephelometric measurements with a Spectronic 200 spectrophotometer. Osmoprotection was achieved in the presence of 0.5 mM proline (17). Thermosensitivity assays. All assays were carried out in E medium. After overnight preincubation at 308C, the cultures were diluted to an OD 540 of 0.05 and transferred to a shaking water bath prewarmed at 428C. Growth was monitored by nephelometric measurements as described above. UV survival assays. Overnight cultures made in NB were diluted 1/10 in the same medium. When the cultures reached an OD of 0.5, the cells were harvested and resuspended in E buffer (E medium without glucose). Five-milliliter aliquots were transferred to sterile, empty petri dishes. Irradiation was achieved by opening the plates under a 15-W Sylvania UV lamp at a distance of 30 cm in the absence of daylight illumination. Cell suspensions were stirred during irradiation. After serial dilution in foil-covered tubes, irradiated cultures were plated on NB. Microscopic observation of cultures. Aliquots of exponentially growing cells in E medium containing 2% glucose were extracted and transferred to slides coated with a thin layer of agar (21). Cells were photographed with a Zeiss Ultraphot microscope under phase-contrast optics, using a Kodak Pan film. Photographs were developed with Kodak Technidol liquid developer. RESULTS Suppression of salt sensitivity by insertion of mini-Tn10 elements. After mutagenesis with either Tn10dTet or Tn10d Cam, salt-resistant derivatives of the His c strain TR6753 were isolated on 33E plates containing either tetracycline or chloramphenicol. Salt-resistant isolates appeared at a frequency of 1 per ;3,000 inserts tested. To be classified as suppressor mutations, the insertions were required to pass the following tests. (i) After more than two rounds of purification, the isolates carrying putative suppressors were lysed with phage P22 HT; the lysates were then used to transduce the parental strain TR6753, selecting either Tet r or Cam r . Antibiotic-resistant transductants were replica plated to 33E plates. Complete (100%) linkage between the transposon-encoded antibiotic resistance and the suppressor phenotype (salt resistance) was taken as evidence that suppression was caused by the insertion mutation. (ii) The lysates described above were also used to transduce strain hisD3050. His 1 transductants were replica plated to 33 E plates; salt sensitivity indicated inheritance of the mutation hisO1242. (iii) To confirm that the suppressor phenotype was caused by the transposon insertion (and not by another mutation located elsewhere in the genome), reconstruction experiments were carried out. First, the transposon insertion was transduced to strain hisD3050, selecting the transposon-encoded antibiotic resistance. The resulting transductants were then transduced to His 1 with a lysate grown on the His c strain TR6753. The presence of the suppressor was confirmed because all of the TABLE 2. Loci involved in the pleiotropic effects of HisH and HisF overexpression Locus Map location (min) Characteristics of mutants Original alleles fully characterized Other allele versions available a osmH 49 Suppression of osmosensitivity, thermosensitivity, and cell division inhibition osmH1::Tn10dCam Tet r osmH2::Tn10dCam Tet r osmH3::Tn10dTet Kan r , Cam r osmH4::Tn10dTet None sfiW 19 Suppression of cell division inhibition; inability to grow in minimal medium sfiW1::Tn10dTet Kan r sfiW2::Tn10dTet Kan r sfiX 44 Suppression of cell division inhibition only sfiX1::Tn10dTet Kan r sfiX2::Tn10dTet Kan r sfiX3::Tn10dTet Kan r sfiY 83 Suppression of cell division inhibition and thermosensitivity sfiY1::Tn10dTet Kan r sfiY2::Tn10dTet Kan r a Not included in the strain list. VOL. 177, 1995 PLEIOTROPIC EFFECTS OF HisH AND HisF OVERPRODUCTION 4843 on July 25, 2017 by INSTITUTO DE PARASITOLOGFA Yhttp://jb.asm.org/Downloaded from His 1 Tet r or His 1 Cam r transductants were salt resistant. As a control, when the same lysates were used to transduce strain hisD3050, all of the transductants were salt sensitive. Eleven (seven Cam r and four Tet r ) salt-resistant isolates of independent origin were used for further study. All proved to be thermoresistant on E plates incubated at 428C, suggesting that osmosensitivity and thermosensitivity had been simultaneously suppressed by a single insertion mutation. Moreover, these salt-resistant isolates were smooth on green plates (although their colony morphology was slightly different from that of the parental strain: compare Fig. 1B and C). Microscopical observation of salt-resistant isolates grown in E liquid cultures containing 2% glucose confirmed that the insertions studied were able to suppress cell division inhibition, although short filaments were still formed (see Fig. 4C). Suppression of thermosensitivity by insertions of Tn10dTet. To ascertain whether suppression of thermosensitivity was always associated with high-salt resistance, thermoresistant derivatives of strain TR6753 were isolated by mutagenesis with Tn10dTet and plating on E medium at 428C (prewarmed). Thermoresistant isolates appeared at a frequency of 1 per 4,000 Tet r inserts. All (four of four, independent) proved to be salt resistant, thereby confirming that osmosensitivity and thermosensitivity had been simultaneously suppressed. These isolates were smooth on green plates, like the salt-resistant isolates described above. Insertions that suppress osmosensitivity, thermosensitivity, and cell division inhibition are alleles of a single locus. To ascertain whether the suppressor alleles belonged to one or more loci, recombination tests were carried out by transduction and replica printing. To assay every allele combination, transposon substitutions were performed. Two insertions were considered allelic whenever the resident allele was excluded from the recombinants (e.g., when .99% of the Tet r transductants were no longer Cam r ). These tests indicated that all general suppressors that abolish salt sensitivity, thermosensitivity, and cell division inhibition are alleles of a single locus. Henceforth, this locus will be called osmH (for osmosensitivity caused by HisH and HisF overproduction). Further work was carried out with the following alleles: osmH1::Tn10dCam, osmH2::Tn10dCam, and osmH3::Tn10dTet (isolated as salt resistant) and osmH4::Tn10dTet (isolated as thermoresistant). Suppression of cell division inhibition by Tn10dTet insertions. For the detection of insertions suppressing cell division inhibition, Tn10dTet inserts were obtained on strain TR6753. Tet r transductants were replica plated to green plates supplemented with tetracycline, to give 300 to 500 colonies per plate. The plates were scored for the presence of smooth colonies; these appeared at a frequency of 1 per ;2,000 Tet r isolates. The initial sample included 47 independent smooth isolates; 14 proved to be P22 resistant and were discarded. The remaining putative suppressor-containing isolates were subjected to tests similar to those described above, to correlate the presence of the insertion with the suppressor phenotype and to prove that the parental strain still contained the mutation hisO1242. None of the smooth isolates formed filaments in E liquid medium containing 2% glucose. However, these isolates belonged to three classes, on the basis of cell and colony morphology, nutritional requirements, and the ability to suppress other manifestations of the pleiotropic response. Class I isolates form small, smooth, dark colonies on green plates (Fig. 1F). They are unable to grow on minimal medium (either E or NCE). This inability to grow in minimal medium hampers the study of thermosensitivity and osmosensitivity, because HisO c strains grown in rich medium are osmoresistant and thermoresistant (data not shown). If insertions of this class are transferred to the wild-type strain LT2, the inability to grow in liquid medium and the abnormal colony shape are likewise observed. Thus, class I insertions cause these defects on their own, irrespective of the genetic background of the strain (His 1 or HisO c ). The growth defect of SfiW 2 mutants is not relieved (i) with any single component of an auxanography pool, (ii) with any pairwise combination of the nutrients of the auxanography test of Davis et al. (19), (iii) with any concentration of Casamino Acids (0.2, 0.4, or 0.6%), or (iv) with fumarate as a carbon source alternative to glucose. Class II isolates form large, smooth, light colonies on green plates (Fig. 1E). They are prototrophic and thermoresistant but osmosensitive. Class III isolates form large, smooth, light colonies on green plates (Fig. 1D). They are prototrophic, thermosensitive, and osmosensitive. Suppressors of cell division inhibition are alleles of three loci. Tn10dTet insertions suppressing cell division inhibition were replaced with Tn10dKan; allelic insertions were detected by transductional recombination as described above. These tests defined three groups of insertions, each corresponding to one of the classes described above (which, in turn, are not A F E DC B FIG. 1. Colonies of S. typhimurium HisO 1 , HisO c , and suppressorcontaining derivatives of the latter, all grown on green plates. (A) TR6753 (hisO1242); (B) LT2 (hisO 1 ); (C) SV2064 (hisO1242 osmH3::Tn10dTet); (D) SV2069 (hisO1242 sfiX1::Tn10dTet); (E) SV2071 (hisO1242 sfiY1::Tn10dTet); (F) SV2068 (hisO1242 sfiW1::Tn10dTet). 4844 FLORES AND CASADESU ´S J. BACTERIOL. on July 25, 2017 by INSTITUTO DE PARASITOLOGFA Yhttp://jb.asm.org/Downloaded from allelic with osmH insertions). Thus, each class of insertions defines a distinct locus. These loci will henceforth be called sfiW (class I), sfiY (class II), and sfiX (class III). The original collection contained 10 sfiW alleles, 16 sfiY alleles, and 7 sfiX alleles; representative alleles of each class are listed in Table 2. Genetic mapping of suppressors. For mapping, all Tn10 insertions characterized as suppressors of one or more manifestations of the pleiotropic response were transferred to strain LT2; the resulting strains are not included in Table 1. (i) Mapping of osmH alleles. Transductions with the locked-in lysate collection generated patches of Tet s transductants when strains TT15258 and TT15253 were used as donors. The Mud-P22 prophages of these strains map at min 50 and 49, respectively. Lysates from TT15258 are packaged clockwise; the packaging direction of TT15253 is unknown. Cotransductional mapping confirmed that osmH is 1% linked to cysA and 5% linked to zej-3271::Tn10. In turn, cysA and zej-3271::Tn10 are 92% linked. These results indicate that osmH lies on min 49; the marker order is osmH-zej-3271::Tn10-cysA. (ii) Mapping of sfiW alleles. Dense patches of Tet s transductants were obtained with locked-in lysates of strain TT15240, whose Mud-P22 prophage maps at min 21.5 and packages counterclockwise. Cotransductional mapping indicated that sfiW is 95% cotransducible with aroA and 7% cotransducible with aspC. The linkage between aroA and aspC is 3%. Thus, sfiW maps on min 19, and the marker order is aroA-sfiW-aspC. (iii) Mapping of sfiX alleles. Tet s transductants were obtained with lysates from TT15249 and TT17163, which package clockwise from min 40.5 and 42, respectively. Cotransductional mapping indicated that sfiX is 11% linked to zee-78::Tn10 and 3% linked to metG. In turn, zee-78::Tn10 and metG are 77% cotransducible. Thus, sfiX maps on min 44, and the marker order is sfiX-zee-78::Tn10-metG. (iv) Mapping of sfiY alleles. Lysates from TT15268 and TT15269, mapping at min 83 and packaging clockwise and counterclockwise, respectively, gave thick patches of Tet r transductants. A lysate from TT15271 (min 84, counterclockwise) gave a faint transduction patch. These data indicated that the sfiY locus maps around min 83 on the S. typhimurium chromosome. Cotransductional mapping confirmed that sfiY is 25% linked to ilvA and 72% linked to cya. The linkage between ilvA and cya is 6%; thus, sfiX lies between ilvA and cya on min 83. Effect of osmolytes on the growth of a His c strain and its mutant derivatives carrying suppressor mutations. The saltresistant phenotype of suppressor-containing derivatives of TR6753 is not observed only on 33E plates. With other osmolytes tested (NaCl, D-xylose, and D-sorbitol), clear differences were observed between the parental His c strain and its osmH derivatives. (i) When osmosensitivity was assayed on NCE plates, the following MICs were found for the His c strain: 0.3 M NaCl, 0.6 MD-xylose, and 1 M D-sorbitol. To prevent the use of D-xylose and D-sorbitol as carbon sources (29), the plates contained 0.2% glucose. The HisO 1 strain and the His c strains carrying an osmH suppressor still grew (slowly) on 1 M NaCl, .1M D-xylose, and .1MD-sorbitol. (ii) Growth curves obtained in the presence of 0.3 M NaCl are shown in Fig. 2A. Growth of the HisO c strain was severely inhibited, while the growth curves of hisO1242 osmH derivatives were nearly identical to that of the HisO 1 strain. Thus, mini-Tn10 insertions at osmH completely suppress the osmosensitivity of His-constitutive strains. In Fig. 2A, one of the strains shown carried an osmH allele (osmH4::Tn10dCam) isolated as a thermoresistant suppressor, while another strain carried the allele osmH3::Tn10dTet, isolated as osmoresistant; both curves are identical. Osmosensitivity was not suppressed by sfiX and sfiY alleles. Because HisO c strains are not osmosensitive in rich media (data not shown), sfiW alleles could not be tested for suppression of osmosensitivity. As an internal control, we tested the ability of the HisO c strain TR6753 and its derivatives carrying osmH,sfiX,orsfiY alleles to grow under low-osmolarity conditions (E liquid medium). All growth curves were similar to that of the wild-type (data not shown). These experiments confirmed that osmH alleles specifically suppress the inability of HisO c strains to grow in high-osmolarity media. (iii) Osmosensitivity of the HisO c strain and its derivatives containing sfiX and sfiY alleles was relieved by the presence of proline 1 mM (Fig. 2B). The alleviating effect of proline on the osmosensitivity of HisO c strains has been previously reported by Csonka (18). Growth of the His c strain and its suppressor-containing derivatives at 42&C. The HisO c strain and its derivatives carrying osmH,sfiX,orsfiY alleles grew normally in E medium at 308C, and their growth curves were similar to that of an isogenic HisO 1 strain (data not shown). In contrast, growth of the HisO c strain was severely inhibited at 428C, while its derivatives containing osmH or sfiY alleles were thermoresistant (Fig. 3). HisO c SfiX 2 strains grew poorly, confirming that sfiX mutations do not suppress thermosensitivity (Fig. 3). Addition of proline to E medium did not relieve the thermosensitive pheFIG. 2. Growth curves of S. typhimurium HisO 1 , HisO c , and suppressorcontaining derivatives of the latter, all grown in 33E liquid medium (A) and in 33E liquid medium supplemented with 1 mM proline (B). E, LT2 (hisO 1 ); F, TR6753 (hisO1242); Ç, SV2082 (hisO1242 osmH4::Tn10dTet); h, SV2064 (hisO1242 osmH3::Tn10dTet); å, SV2069 (hisO1242 sfiX1::Tn10dTet); ■, SV2071 (hisO1242 sfiY1::Tn10dTet). VOL. 177, 1995 PLEIOTROPIC EFFECTS OF HisH AND HisF OVERPRODUCTION 4845 on July 25, 2017 by INSTITUTO DE PARASITOLOGFA Yhttp://jb.asm.org/Downloaded from notype of the HisO c strain and its HisO c SfiX 2 derivatives (data not shown). Thus, thermosensitivity and osmosensitivity are two independent manifestations of the pleiotropic phenotype (and not two features of the same phenomenon). Suppression by osmH alleles is unrelated to changes in expression of the histidine operon. All phenotypes of the pleiotropic response are relieved or abolished by osmH mutations. Thus, the effect of osmH mutations is formally similar to that of cis-acting mutations (e.g., p 1 promoter mutations) that reduce his transcription (12). OsmH 2 mutations cannot be active in cis because they are unlinked to the his operon and recessive (see below). Although the possibility that a recessive mutation could lower his transcription seemed a priori unlikely because trans-acting positive regulators of the p 1 promoter are not known to exist (53), we used transcriptional b-galactosidase fusions to measure his transcription levels in osmH mutants. Suppressor osmH alleles were introduced in isogenic hisO 1 and hisO1242 strains that contained a MudJ insertion in hisC. While the presence of the attenuator deletion hisO1242 caused a 10to 15-fold difference in the b-galactosidase activity of the lac fusion hisC9968::MudJ, no significant difference was observed when the strains contained an osmH suppressor (Table 3). These results indicate that osmH mutations do not lower transcription from the p 1 promoter. However, HisH and HisF are also produced from transcripts originating at the internal p 2 promoter (48). The weakness of this promoter allows suppression of the pleiotropic effects of HisH and HisF overexpression by polar mutations upstream p 2 (12), indicating that high levels of HisH and HisF products are not obtained in the absence of constitutive expression from p 1 . However, processed mRNA species derived from the main his transcript are known to have increased stability (1). Thus mutations affecting the production and/or the stability of processed mRNAs containing the distal part of the operon can be expected to affect HisH and HisF levels. As a consequence, these (hypothetical) mutations might suppress the pleiotropic response. Since all processed mRNA species known include hisIE (1), we investigated the effect of osmH mutations on the activity of hisIE9969::Mud1-8 lac fusions. The results (Table 3) indicate that osmH mutations do not affect hisIE expression levels. Thus, the possibility that osmH mutations mimic suppression by lowering HisH and HisF expression seems unlikely. Effect of suppressor mutations on cell division. Strains LT2 (HisO 1 ), TR6753 [HisO c ], SV2064 [HisO c OsmH 2 ], SV2069 [HisO c SfiX 2 ], and SV2071 [HisO c SfiY 2 ] were grown in E medium containing 2% glucose; strain SV2068 [HisO c SfiW 2 ] was grown in NB containing 2% glucose. Photographs obtained from late-exponential-phase cultures are shown in Fig. 4. Division of the parental HisO c strain is severely inhibited. Filamentation is not observed in its suppressor-containing derivatives, although short filaments seem to coexist with normal cells in strain SV2064 [HisO c OsmH 2 ]. None of the sfi mutants forms filaments, but all show abnormal cell morphology; lysis might be occurring in the strains containing sfiX and sfiY suppressors, while HisO c SfiW 2 cells are small and dark. Suppression of cell division inhibition in HisO c strains does not cause UV sensitivity. To explore the possibility that mutations that suppress HisO c division inhibition can affect genes related to the SOS pathway, we determined the UV sensitivity of strains carrying osmH,sfiX,sfiW, and sfiY insertion mutations. No differences were observed between the suppressorcontaining isolates and the parental His c strain TR6753; in turn, the UV sensitivity of TR6753 was similar to that of the parental strain LT2 (data not shown). These results agree with earlier observations indicating that the division inhibition pathway triggered by HisH and HisF overexpression is unrelated to the SOS response (28). Complementation of suppressor alleles and dominance studies. Although most insertion mutations can be expected to be recessive (7, 36), Tn10 outward promoters can also activate adjacent genes (16, 50). Thus, we carried out complementation tests to ascertain whether the suppressor mutations studied are dominant or recessive. (i) For complementation analysis of sfiW mutants, we constructed a duplication with endpoints at min 12 and 25, using a strategy adapted from that of Hughes and Roth (33). Two strains, each carrying a Mud-P22 prophage in the same orientation in purE (TT15235) and purB (TT15241), were lysed with P22 HT. The lysates were mixed to transduce strain TR6753, selecting Cam r on minimal plates. Prototrophic Cam r transductants can be expected to contain a duplication of the purEpurB region, with an hybrid locked-in Mud-P22 prophage in the middle. This assumption was confirmed because segregation in the absence of antibiotic selection yielded only Cam s prototrophs (see reference 33). One isolate carrying one such Mud-held duplications between min 12 and 25 was propagated as strain SV2155. This strain was rough on green plates and showed all the additional manifestations of the pleiotropic response. To examine whether sfiW mutations are dominant or recessive, a sfiW::Tn10dTet allele was introduced into strain SV2155 by P22 HT transduction, selecting Tet r . To maintain the duplication, chloramphenicol was also added to the selective plates. The resulting Cam r Tet r transductants contain one copy of the sfiW 1 allele and one sfiW::Tn10dTet insertion. All Tet r FIG. 3. Growth curves of S. typhimurium HisO 1 , HisO c , and suppressorcontaining derivatives of the latter, all grown in E liquid medium at 428C. E, LT2 (hisO 1 ); F, TR6753 (hisO1242); Ç, SV2082 (hisO1242 osmH4::Tn10dTet); h, SV2064 (hisO1242 osmH3::Tn10dTet); å, SV2069 (hisO1242 sfiX1::Tn10dTet); ■, SV2071 (hisO1242 sfiY1::Tn10dTet). TABLE 3. Activities of transcriptional hisC::lac and hisIE::lac fusions in OsmH 1 and OsmH 2 backgrounds Strain Genotype b-Galactosidase activity (Miller units) SV2090 hisO 1 hisC9968::MudJ 493 SV2091 hisO1242 hisC9968::MudJ 5,953 SV2092 hisO 1 hisC9968::MudJ osmH1::Tn10dTet 445 SV2093 hisO1242 hisC9968:MudJ osmH1::Tn10dTet 6,028 SV2137 hisO 1 hisIE9969:Mud1-8 52 SV2138 hisO1242 hisIE9969:Mud1-8 494 SV2139 hisO 1 hisIE9969::Mud1-8 osmH1::Tn10dTet 55 SV2140 hisO1242 hisIE9969::Mud1-8 osmH1::Tn10dTet 471 4846 FLORES AND CASADESU ´S J. BACTERIOL. on July 25, 2017 by INSTITUTO DE PARASITOLOGFA Yhttp://jb.asm.org/Downloaded from FIG. 4. Phase-contrast micrographs of late-exponential-phase cultures of S. typhimurium HisO 1 , HisO c , and suppressor-containing derivatives of the latter, all grown in liquid E medium containing 2% glucose. (A) TR6753 (hisO1242); (B) LT2 (hisO 1 ); (C) SV2064 (hisO1242 osmH3::Tn10dTet); (D) SV2069 (hisO1242 sfiX1::Tn10dTet); (E) SV2071 (hisO1242 sfiY1::Tn10dTet); (F) SV2068 (hisO1242 sfiW1::Tn10dTet). VOL. 177, 1995 PLEIOTROPIC EFFECTS OF HisH AND HisF OVERPRODUCTION 4847 on July 25, 2017 by INSTITUTO DE PARASITOLOGFA Yhttp://jb.asm.org/Downloaded from Cam r transductants were rough and light green, indicating that sfiW mutations are recessive. The presence of both sfiW 1 and sfiW alleles in the strain was confirmed by segregation: upon nonselective streaking on green plates, Cam s isolates were obtained. These belonged to two types: rough, light green, Tet s and smooth, dark, Tet r . (ii) For complementation analysis of sfiX mutants, we constructed a Mud-held duplication with endpoints in hisH (min 42) and cysA (min 50). The resulting strain (SV2150) was then used as the recipient of a sfiX::Tn10dTet allele. All the Tet r transductants were rough, indicating that sfiX mutations are recessive. The merodiploid nature of these transductants was confirmed by segregation as described above. (iii) Complementation analysis of osmH and sfiY mutants was analyzed with a slightly different strategy. Since osmH is linked to cysA, both the suppressor and a linked cysA::MudP insertion can be cotransduced to a His c recipient (TR6753). Simultaneous selection of prototrophy and the two incoming markers allows the isolation of rare transductants in which the transduced fragment has recombined with one copy of a preexisting duplication. The OsmH 1 /OsmH 2 merodiploids obtained proved to be salt sensitive, thermosensitive, and rough on green plates, indicating that osmH mutations are recessive. Segregation of the merodiploids provided an internal control for the experiment. Since sfiY is linked to ilvA, an analogous procedure was used to obtain SfiY 1 /SfiY 2 merodiploids: the HisO c strain TR6753 was transduced with a P22 HT lysate grown on a donor carrying an sfiY suppressor and a linked ilvA::MudQ insertion, selecting Ilv 1 Cam r Tet r transductants. The SfiY 1 /SfiY 2 merodiploids obtained formed wrinkled colonies on green plates (and filaments in liquid medium) and were thermosensitive, indicating that sfiY mutations are recessive. Combinations of suppressor mutations. Taking advantage of the allele variants harboring different mini-Tn10 insertions (Tn10dTet, Tn10dCam, and Tn10dKan), we carried out transductional crosses to construct His c strains containing more than one suppressor mutation. The relevant conclusions from these experiments are as follows. (i) All allele combinations are viable. (ii) The hierarchy of phenotypes defined by single suppressors remains unaltered in doubleand triple-suppressor mutants; for instance, osmH sfiX,osmH sfiY, and osmH sfiX sfiY strains are osmoresistant and thermoresistant and do not form filaments. In turn, sfiX sfiY mutants are thermoresistant and do not form filaments. (iii) The cell shape of osmH sfiX and osmH sfiY mutants resembles that of osmH strains (compare Fig. 5 with Fig. 4C). Thus, the cell morphology conferred by osmH alleles is epistatic over those of sfiY and sfiX. The same conclusion applies to colonial morphology (data not shown). (iv) When sfiY and sfiX alleles are combined, both the colony shape and the cell morphology of the double-suppressor mutants remain like those of the parental strains, which are in turn hard to distinguish (Fig. 1 and 4D and E). These results do not clarify whether the phenotypes conferred by sfiX and sfiY mutations are epistatic or additive. However, the fact that sfiX sfiY mutants are thermoresistant indicates that sfiX mutations are not epistatic over sfiY. Altogether, these observations suggest a model in which the pleiotropic response involves the hierarchical expression of osmH,sfiY, and sfiX (Fig. 6). The observation that sfiX sfiY mutants are thermoresistant places sfiX below sfiY (and not in a parallel position). The existence of sfiW suppressors complicates the situation. When combined with osmH,sfiX,orsfiY alleles, the resulting doubleand triple-suppressor mutants do not grow in minimal medium and show the colony and cell shape typical of SfiW 2 mutants. Thus, the sfiW effects on growth pattern, cell shape, and colony morphology are epistatic over those of osmH,sfiX, and sfiY. The growth defect caused by sfiW alleles does not allow testing HisO c OsmH 2 SfiW 2 and HisO c SfiX 2 SfiW 2 strains for suppression of thermosensitivity and osmosensitivity; thus, the consequences (if any) of combining sfiW alleles with osmH and sfiY cannot be analyzed. DISCUSSION This report describes insertion mutations that suppress one or more manifestations of the His c pleiotropic response. All of the suppressor mutations are recessive; therefore, they identify genes that trigger the pleiotropic response when HisH and HisF are overproduced. With the exception of sfiW alleles, FIG. 5. Phase-contrast micrograph of a late-exponential-phase culture of strain SV2129 (hisO1242 osmH3::Tn10dCam sfiX2::Tn10dTet) grown in liquid E medium containing 2% glucose. FIG. 6. A pathway-like model for the His c pleiotropic response, based on the hierarchy of suppressor mutations and on the epistatic effects observed among certain suppressor phenotypes. 4848 FLORES AND CASADESU ´S J. BACTERIOL. on July 25, 2017 by INSTITUTO DE PARASITOLOGFA Yhttp://jb.asm.org/Downloaded from none of the suppressors studied confers a recognizable phenotype to the wild type. Insertions in osmH (min 49) suppress osmosensitivity, thermosensitivity, and cell division inhibition. Insertions in sfiY (min 83) suppress thermosensitivity and cell division inhibition (but not osmosensitivity). Insertions in sfiX (min 44) suppress cell division inhibition only. The general suppression exerted by osmH insertions might indicate that the pleiotropic response triggered by HisH and HisF overexpression has a single cellular target, the product of the osmH gene. However, the existence of partial suppressors, the hierarchy of their phenotypes, and the existence of epistatic effects among the suppressors suggest that the pleiotropic response is a pathway which can be interrupted at different stages (Fig. 6). The existence of a fourth type of suppressor, sfiW, adds further complexity to the picture. Certain phenotypes of sfiW mutations are epistatic over those of osmH,sfiY, and sfiX, suggesting that sfiW might be placed above osmH in the tentative pathway of Fig. 6. Alternatively, sfiW might suppress filamentation by a mechanism unrelated to the pathway defined by osmH,sfiY, and sfiX. The latter explanation seems more reasonable because the distinct phenotypes conferred by sfiW mutations (bizarre morphology of cells and colonies, inability to grow in minimal medium) are likewise observed in a HisO 1 background. On this rationale, sfiW has been placed outside the osmH-sfiY-sfiX pathway drawn in Fig. 6. None of the suppressors studied causes UV sensitivity; thus, the corresponding wild-type genes must be involved in an SOSindependent process of filament formation. This assumption agrees with the previous observation that the pleiotropic effects of HisH and HisF overproduction are SOS unrelated and sulA independent (28). In E. coli, SOS-independent division inhibitors are encoded by the minB operon (8, 20). The map positions of the osmH,sfiW,sfiX, and sfiY loci on the S. typhimurium chromosome are all different from that of the E. coli minB operon (4). Thus, the cellular functions that cause filamentation upon HisH and HisF overproduction seem to identify novel loci involved in an SOS-independent process of cell division inhibition. The uniqueness of the His c pleiotropic response (and thus of the scores used for suppressor isolation) may explain why novel genes putatively related to cell division have been detected in our hunts. The list of suppressors of the His c pleiotropic response described in this report may be incomplete, first because all of the suppressors described are insertion mutations, and thus any essential genes that might be involved in the His c response would be overlooked in our hunts, and second because use of ATS transposase can be expected to broaden Tn10 target specificity, but target preferences must still exist (5). This may explain why one type of mutation which has been previously shown to suppress the pleiotropic response of His c mutants (envB [3]) has not been found among the suppressors described in this work. ACKNOWLEDGMENTS This study was supported by grants PB90-0898 and PB93-649 from the DGICYT of the Government of Spain. Additional funds were obtained from the Regional Government of Andalusia (Junta de Andalucı´a, Spain). We thank Miguel Vicente for advice and David A. Cano for helping in mapping experiments. Cell pictures were taken by Manuel Sa´nchez and Pilar Palacios. We acknowledge the efficient and generous help received from Andrew Hessel and Kenneth Sanderson, Salmonella Genetic Stock Center, University of Calgary, Calgary, Alberta, Canada. Strains were also provided by Nick Benson and John Roth. We are also grateful to Andre´s Garzo´n, Molly Schmid, Alan Campbell, Richard D’Ari, Diana Downs, John Roth, Lionello Bossi, Rafael Maldonado, and Juan Aguilar for helpful suggestions and to Eduardo Santero and Andre´s Aguilera for critical reading of the manuscript. The technical assistance of Asuncio´n Blasco, Ana Moreno, Jose´Co´rdoba, and Luis Romanco is also appreciated. REFERENCES 1. Alifano, P., C. Piscitelli, V. Blasi, F. Rivellini, A. G. Nappo, C. B. Bruni, and M. S. Carlomagno. 1992. Processing of a polycistronic mRNA requires a 59 cis element and active transcription. Mol. Microbiol. 6:787–798. 2. Anderson, R. P., and J. R. Roth. 1979. Gene duplication in bacteria: alteration of gene dosage by sister-chromosome exchanges. Cold Spring Harbor Symp. Quant. Biol. 43:1083–1087. 3. Anto´n, D. N. 1979. Positive selection of mutants with cell envelope defects in aSalmonella typhimurium strain hypersensitive to the products of genes hisH and hisF. J. Bacteriol. 137:1271–1281. 4. Bachmann, B. J. 1990. Linkage map of Escherichia coli K-12, edition 8. Microbiol. Rev. 54:130–197. 5. Bender, J., and N. Kleckner. 1992. IS10 transposase mutations that specifically alter target site recognition. EMBO J. 11:741–750. 6. Benson, N. R., and B. S. Goldman. 1992. Rapid mapping in Salmonella typhimurium with Mud-P22 prophages. J. Bacteriol. 174:1673–1681. 7. Berg, C. M., D. E. Berg, and E. A. Groisman. 1989. Transposable elements and the genetic engineering of bacteria, p. 879–925. In D. E. Berg and M. M. Howe (ed.), Mobile DNA. American Society for Microbiology, Washington, D.C. 8. Bi, E., and J. F. Lutkenhaus. 1990. Interaction between the minB locus and ftsZ. J. Bacteriol. 172:5610–5616. 9. Bochner, B. R., and B. N. Ames. 1982. ZTP (5-amino-4-imidazole carboxamide riboside 59-triphosphate): a proposed alarmone for 10-formyl-tetrahydrofolate deficiency. Cell 29:929–937. 10. Bochner, B. R., H. C. Huang, G. L. Schieven, and B. N. Ames. 1980. Positive selection for loss of tetracycline resistance. J. Bacteriol. 143:926–933. 11. Brown, D. A., R. D’Ari, and E. B. Newman. 1990. A relationship between L-serine degradation and methionine biosynthesis in Escherichia coli K12. J. Gen. Microbiol. 136:1017–1023. 12. Casadesu´s, J., and J. R. Roth. 1989. Absence of insertions among spontaneous mutants of Salmonella typhimurium. Mol. Gen. Genet. 216:210–216. 13. Castilho, B. A., P. Olfson, and M. J. Casadaban. 1984. Plasmid insertion mutagenesis and lac gene fusion with mini-Mu bacteriophage transposons. J. Bacteriol. 158:488–495. 14. Chan, R. K., D. Botstein, T. Watanabe, and Y. Ogata. 1972. Specialized transduction of tetracycline by phage P22 in Salmonella typhimurium. II. Properties of a high frequency transducing lysate. Virology 50:883–898. 15. Chumley, F. G., and J. R. Roth. 1980. Rearrangement of the bacterial chromosome using Tn10 as a region of homology. Genetics 94:1–14. 16. Ciampi, M. S., M. B. Schmid, and J. R. Roth. 1982. Transposon Tn10 provides a promoter for transcription of adjacent sequences. Proc. Natl. Acad. Sci. USA 79:5016–5020. 17. Csonka, L. N. 1981. Proline over-production results in enhanced osmotolerance in Salmonella typhimurium. Mol. Gen. Genet. 182:82–86. 18. Csonka, L. N. 1989. Physiological and genetic responses of bacteria to osmotic stress. Microbiol. Rev. 53:121–147. 19. Davis, R. W., D. Botstein, and J. R. Roth. 1980. Advanced bacterial genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 20. De Boer, P. A. J., R. E. Crossley, and L. I. Rothfield. 1988. Isolation and properties of minB, a complex genetic locus involved in correct placement of the division site in Escherichia coli. J. Bacteriol. 170:2106–2112. 21. Donachie, W. D., K. J. Begg, and M. Vicente. 1976. Cell length, cell growth and cell division. Nature (London) 264:328–333. 22. Downs, D. M. 1987. Purine metabolism and cryptic prophages in Salmonella typhimurium. Ph.D. Thesis. University of Utah, Salt Lake City. 23. Elliott, T., and J. R. Roth. 1988. Characterization of Tn10d-Cam: a transposition-defective Tn10 specifying chloramphenicol resistance. Mol. Gen. Genet. 213:332–338. 24. Fink, G. R., T. Klopotowski, and B. N. Ames. 1967. Histidine regulatory mutants in Salmonella typhimurium. IV. A positive selection for polar histidine-requiring mutants from histidine operator constitutive mutants. J. Mol. Biol. 30:81–95. 25. Flores, A., M. Fox, and J. Casadesu´s. 1993. The pleiotropic effects of Salmonella typhimurium his overexpression do not involve AICAR-induced mutagenesis. Mol. Gen. Genet. 240:360–364. 26. Fox, M., N. Frandsen, and R. D’Ari. 1993. AICAR is not an endogenous mutagen in Escherichia coli. Mol. Gen. Genet. 240:355–359. 27. Frandsen, N., and R. D’Ari. 1993. Excess histidine enzymes cause AICARindependent filamentation in Escherichia coli. Mol. Gen. Genet. 240:348– 354. 28. Gibert, I., and J. Casadesu´s. 1990. sulA-independent division inhibition in His-constitutive strains of Salmonella typhimurium. FEMS Microbiol. Lett. 69:205–210. 29. Gutnick, D., J. M. Calvo, T. Klopotowski, and B. N. Ames. 1969. Compounds which serve as the sole source of carbon or nitrogen for Salmonella typhimurium LT-2. J. Bacteriol. 100:215–219. VOL. 177, 1995 PLEIOTROPIC EFFECTS OF HisH AND HisF OVERPRODUCTION 4849 on July 25, 2017 by INSTITUTO DE PARASITOLOGFA Yhttp://jb.asm.org/Downloaded from