Bacillus subtilis implements several adaptive strategies to cope with nutrient limitation experienced at the end of exponential growth . The DegS-DegU two-component system is part of the network involved in the regulation of postexponential responses, suchas competence development, the production of exoenzymes, andmotility . The degU32[Hy] mutation extends the half-life of the phosphorylated form of DegU [DegU-P]; this in turn increasesthe production of alkaline protease, levan-sucrase, and otherexoenzymes and inhibits motility and the production of flagella.The expression of the flagellum-specific sigma factor SigD,of the flagellin gene hag, and of the fla-che operon is stronglyreduced in a degU32[Hy] genetic background . To investigate themechanism of action of DegU-P on motility, we isolated mutantsof degU32[Hy] that completely suppressed the motility deficiency.The mutations were genetically mapped and characterized by PCRand sequencing . Most of the mutations were found to delete atranscriptional termination signal upstream of the main flagellaroperon, fla-che, thus allowing transcriptional readthrough fromthe cod operon . Two additional mutations improved the ^A-dependent promoter sequence of the fla-che operon . Using an electrophoretic mobility shift assay, we have demonstrated that purified DegU binds specifically to the P_A promoter region of the fla-che operon . The data suggest that DegU represses transcription of the fla-che operon, and they indicate a central role of the operon in regulating the synthesis and assembly of flagella.

 
Swimming motility in bacteria depends upon the presence on thecell surface of the flagellar organelle, composed of the basalbody, the hook, and the filament . The production of flagellais of such adaptive value that most bacterial species are endowedwith flagella, despite the high energy requirement for the synthesisof the numerous flagellin monomers that are necessary to buildand maintain the flagellar filament.

In enterobacteria, the genes involved in flagellar formationare organized into regulons which are arranged into three hierarchical classes . The first class is constituted by the flhDC master operon, whose expression is necessary to turn on class II genes, coding for components of the export machinery and for the hookand basal body . The class II gene fliA encodes ²⁸, the transcriptionfactor for the class III genes, which include flagellar filamentstructural genes and the chemotaxis signal transduction system[7, 19] . In addition, many global regulators, such as CAP, H-NS,H-HU, Lrp, etc., have been reported to affect flagellar synthesisand assembly [5, 13, 24, 34].

In Bacillus subtilis, a bona fide master operon is missing and all genes corresponding to the enteric class II are clustered in a single fla-che operon . The expression of the operon depends upon a ^A-recognized promoter [fla-che P_A], with an additional ^D-dependent promoter [P_D-3] playing a minor role [1, 9] . Deletionof the fla-che P_A promoter renders the cells completely nonmotile,or Mot^– [40] . The sigD gene, encoding ^D, is at the endof the fla-che operon, and its expression is required for thetranscription of class III genes for flagellin, motor components,receptors, autolysins, and chemotaxis [27] . One of the lateflagellar gene products, FlgM, is an anti-sigma factor involvedin the control of expression of ^D-dependent promoters [3].

The regulation of motility in B . subtilis is integrated into a complex net of regulatory circuits controlling different adaptive responses, such as competence development and degradative enzyme synthesis and secretion [22] . The response regulator DegU controlsthe production of exoenzymes such as proteases, levan-sucrase,and -amylase and is involved in competence development and motility.DegU is the second element of a two-component signaling systemthat is phosphorylated by the first component, DegS . Missensemutations within the degS and degU genes, designated degS[Hy] and degU[Hy], that increase the half-life of the phosphorylated form of DegU [DegU-P] result in the overexpression of secreted enzymes and a lack of flagella [2, 8, 17, 23, 37] . One of the best-characterized such mutations is degU32[Hy] [15] . This mutationcauses a strong reduction in the expression of sigD, as seenwith a sigD-lacZ translational fusion, and prevents transcriptionfrom the cwlB ^D-dependent promoter [38] . The mechanism throughwhich the phosphorylated form of DegU negatively regulates theexpression of sigD is completely obscure.

To investigate the mechanism of action of DegU-P, we isolatedand characterized mutations that suppress the effect of degU32[Hy] on motility . Our results indicate that DegU-P represses the transcription of the fla-che operon, thereby preventing the expression of genes coding for the hook and basal body componentsof the flagellum and for the ^D transcription factor . Furthermore,we show that purified DegU binds to the regulatory region ofthe fla-che operon.

 
Bacterial strains and media. Table 1 lists the bacterial strains used for this study . B.subtilis strains were grown in Schaeffer sporulation medium[SM], PY broth [Pennassay antibiotic medium 3; Difco], or theminimal medium described by Kunst and Rapoport [18] . Motilitywas tested by spotting samples obtained from overnight coloniesby use of toothpicks . The motility medium had the followingcomposition:1% Bacto peptone, 8% Bacto gelatin, 1% Bacto agar,0.5% NaCl, and 25 µg of the appropriate growth requirement/ml.Escherichia coli cells were grown in Luria-Bertani broth . E.coli DH5 supE44 lacU169 [80 lacZM15] hsdR17 recA1 endA1 gyrA96thi-1 relA1 was used as a host for the construction of recombinantplasmids . E . coli C600 thi-1 thr-1 leuB6 lacY1 tonA21 supE44was used as an intermediate host for plasmids prior to the transformationof B . subtilis . For protein expression and purification, weused E . coli BL21 D3 [hsdS gal [ cIts857 ind1 Sam7 nin5 lacUV5-T7]gene1] . Antibiotics were used at the following concentrations:ampicillin, 100 µg/ml; erythromycin, 1 µg/ml; kanamycin,2.5 µg/ml; chloramphenicol, 5 µg/ml; phleomycin,5 µg/ml.

 
Genetic techniques. B . subtilis strains were transformed with chromosomal or plasmidDNA by the procedure of Kunst and Rapoport [18] . Transductionmapping with the PBS1 phage was performed according to Hochet al . [16] . For genetic landmarks, we used a set of mutantsconstructed by several laboratories [32] . E . coli transformation was performed according to standard protocols [31].

Construction of mutants. For the construction of strains with a deletion of the promoterP_D-3, a sequence of 191 bp overlapping the promoter was replacedwith a kanamycin resistance determinant [Fig . 1A] . First a 485-bpDNA fragment corresponding to the intercistronic region downstreamof P_D-3 and extending into the coding sequence of flgB was amplified by a PCR with primers E8601 and B9070, which contained an EcoRIand a BamHI site, respectively . The template was chromosomalDNA from B . subtilis 168 . The amplified fragment was purifiedwith a Qiaquick PCR purification kit [Qiagen] and restrictedwith EcoRI and BamHI . After digestion, the DNA fragment waspurified with the same kit and ligated to EcoRI- and BamHI-digestedpJM114 [28] . The ligated plasmid was transformed into E . coliDH5 and called pGA10 . A second DNA fragment of 474 bp extendingfrom the middle of the codY coding sequence to four nucleotidesafter the codY stop codon was obtained by a PCR with primersK7936 and S8409, which contained a KpnI and a SalI recognitionsequence, respectively . The amplified DNA was treated as describedabove and cloned into pGA10 that had been digested with KpnIand SalI . The final plasmid [pGA11] contained two fragmentsof DNA derived from B . subtilis strain 168 flanking the kanamycinresistance determinant . The plasmid was verified by sequencing.The plasmid DNA was linearized by ScaI digestion and used totransform competent cells of B . subtilis 168, selecting forKan^r [2.5 µg/ml] . One of the transformants [PB5306] wascharacterized by PCR and sequencing and its DNA was used totransfer the deletion mutation into strain PB5213, generatingstrain PB5307.

 
To place gene codY under the control of the isopropyl-ß-D-thiogalactopyranoside [IPTG]-inducible Pspac promoter, we proceeded in the following way [Fig . 1B] . A 451-bp DNA fragment corresponding to the endof gene clpY [upstream of codY] and part of codY was obtainedby amplification of the chromosomal DNA with primers 7541 and7971 [Table 2] . The purified DNA was cloned into pUC18 by useof the SureClone ligation kit [Amersham Pharmacia Biotec] . Aftersequencing, one plasmid was chosen and digested with EcoRI andBamHI, and a DNA band of approximately 450 bp was purified andcloned into plasmid pMutin4 [39], which had previously beendigested with EcoRI and BamHI and dephosphorylated . The newplasmid, pMutin4/9, was verified by sequencing, transformedinto E . coli strain C600, extracted, and used to transform B.subtilis 168 competent cells . Selection was done on tryptoseblood agar base [Difco] plates with erythromycin at 1 µg/ml.One transformant was retained [PB5290], and after PCR verification,its DNA was used to transform strain PB5213 to obtain strainPB5291 . Strains PB5290 and PB5291 are thus derivatives of PB168and PB5213, respectively, with the codY gene under the controlof the IPTG-inducible Pspac promoter.

 
Plasmid pMutin4/9 was also used to transform strain PB5248 toobtain strain PB5368, in which the codY gene under the controlof the Pspac promoter was combined with the dhsA245 deletion [Fig . 1E] . For a tighter repression of transcription from Pspac in the absence of the IPTG inducer, plasmid pMAP65, which overproduces LacI, was introduced by transformation and selection for Phleo^r [30].

For the construction of the ylxF-lacZ transcriptional fusion, an internal fragment of the ylxF sequence was amplified by a PCR with the primers ylxF3 and ylxF2 [Table 2] . The amplifiedfragment [331 nucleotides] was restricted with EcoRI and BamHIand cloned into EcoRI- and BamHI-digested pMutin4 to generate pMutin331EB . The B . subtilis ylxF::pMutin4 [BFA2666] strain was obtained by Campbell-type integration of pMutin331EB into the chromosome of B . subtilis 168 . The mutant construction was verified by PCRs with combinations of the ylxF3 and ylxF2 primersand oligonucleotides corresponding to sequences of the vector.

To construct strain PB5323, in which the dhsA4 mutation was separated from the cod operon, we cloned the complete coding sequence of gene codY into the integrative plasmid pJM114 [28]. The codY gene was generated from the chromosomal DNA by a PCR with primers codY2 and codYB, which contained an EcoRI and a BamHI site, respectively [Table 2] . The amplified fragment was purified, digested, and ligated into pJM114, which had been digested with EcoRI and BamHI as reported above . The new plasmid [pPB01] was expanded in E . coli strain C600 and used to transform competent cells of strain PB5317 [dhsA4], selecting for Kan^r [Fig . 1D].

ß-Galactosidase activity assay. To assay ß-galactosidase activity, we diluted overnightcultures in SM in fresh medium and took samples at 30-min intervalsfor optical density readings at 525 nm and ß-galactosidaseactivity determinations . The calculation of ß-galactosidaseunits was done in modified Miller units [29].

Preparation of RNA. The following protocol, adapted from the work of Caldwell etal . [6], was used to prepare B . subtilis RNA samples for reversetranscription-PCR [RT-PCR] . Samples [5 ml] of cultures in SMwere collected and rapidly frozen by dripping into approximately40 ml of liquid nitrogen in a 50-ml Falcon tube . The tubes werestored at –80°C overnight . The frozen droplets weretransferred to a commercial hand-held coffee grinder that wasprechilled and contained crushed dry ice . The sample was groundfor 2 min in a cold room and the powder was transferred to a new Falcon tube and stored at –20°C overnight to allowthe dry ice to sublime . The frozen sample was treated with 5ml of phenol saturated with Tris-EDTA buffer [pH 7.8], mixed,and incubated at 65°C for 3 min . After cooling for 3 minon ice, the mixture was centrifuged at 12,000 x g for 10 min at 4°C . The upper aqueous phase was extracted with an equal volume of phenol, and the procedure was repeated four additional times . The last aqueous phase was transferred to a 1.5-ml Eppendorf tube, and 5 volumes of Trizol reagent [GIBCO BRL] was added.After mixing, the sample was incubated at room temperature for5 min . Chloroform was added [1 volume of the aqueous phase],mixed for 15 s, and incubated for 3 min at room temperature.After centrifugation at 12,000 x g for 15 min at 4°C, the aqueous phase was carefully removed and transferred to a new tube, and the RNA was precipitated with a one-half volume of isopropanol at room temperature for 10 min . After centrifugationat 12,000 x g for 10 min at 4°C, the supernatant was removedand the pellet was washed with 70% ethanol and centrifuged at7,500 x g for 5 min at room temperature . The pellet was airdried for 15 min and dissolved in diethyl pyrocarbonate-treatedwater . The yield was determined by UV spectroscopy.

RT-PCR. We used RT-PCR to evaluate the presence of readthrough transcriptionfrom the cod operon into the fla-che operon . Wild-type, degU₃₂, and degU₃₂ dhs mutant cells grown in SM were harvested at T_–1 [1 h before the transition point], T₀ [the transition pointbetween the exponential and postexponential growth phases], and T₂ [2 h after the transition point], and RNAs were isolatedas described above.

A cDNA were synthesized by RT, with purified RNA as a templateand with primer flgB1 [Table 2; also see Fig . 8], which wascomplementary to the flgB gene mRNA . For these experiments,3 µg of RNA template was mixed in a final volume of 12 µl with 20 pmol of the oligonucleotide flgB1 . The samplewas denatured for 5 min at 70°C and cooled for 1 min onice . The following reagents were added to this mixture: 4 µlof 5x first-strand buffer [GIBCO BRL], 2 µl of 0.1 M dithiothreitol,and 1 µl of 10 mM [each] deoxynucleoside triphosphates.After incubation at 42°C for 2 min, 1 µl [200 U] ofSuperScript II reverse transcriptase [GIBCO BRL] was added,and the incubation was prolonged for 1 h at 42°C . The samplewas treated with 1 µl of RNase H [2 U/µl; GIBCO BRL] for 10 min at 55°C and then allowed to cool . The resulting cDNA was used as a template to synthesize and amplify DNA fragments with Taq DNA polymerase [5 U/µl; GIBCO BRL] . The downstream primer was flgB2, which was complementary to the noncoding strand. Two different upstream primers, 7901 and PA, were used [Table 2; also see Fig . 8] . Primer 7901, which was complementary tothe codY transcript, was used to detect readthrough cod-flgBtranscripts, whereas primer PA was used to detect both readthroughtranscripts and transcripts initiating at the P_A promoter . Thesecond reaction contained 1.5 µl of the product of thefirst reaction and 15 pmol each of primer flgB2 and primer PA[or primer 7901] in a final volume of 20 µl . The conditionsfor PCR were as follows: 94°C for 1 min, 54°C for 1 min, and 72°C for 1 min for 40 cycles . The resulting RT-PCR products were analyzed by agarose gel electrophoresis [see Fig. 8] . RT-PCRs in the absence of reverse transcriptase were performedin parallel to check for DNA contamination . PCR products fromchromosomal DNAs from the parental and mutant strains obtained with the same primers [flgB2 and 7901] were run in parallel with the samples.

 
Purification of DegU. The degU gene was PCR amplified from wild-type [PB168] and degU32[Hy]mutant [PB5213] chromosomal DNAs with primers degU-Nt and degU-Ct[Table 2] . Both primers contained a restriction site for NdeI, which was used to clone the genes into the expression plasmids pET12a and pET16B [Novagen] . Four plasmids were thus obtained:two were derivatives of pET12a [pET12Uwt and pET12Uhy] codingfor the wild-type and mutant proteins, respectively, and twowere derivatives of pET16b [pET16Uwt and pET16Uhy] coding forthe two proteins with an N-terminal His₆-tag fusion [named H6-DegUand H6-DegU32, respectively] . For expression of the recombinantproteins, E . coli BL21 DE3[pLys] competent cells were transformedwith the plasmids, inoculated in Luria-Bertani medium containing100 µg of ampicillin/ml, grown at 37°C to an opticaldensity at 600 nm of 0.6, and induced with 1 mM IPTG, and growthwas then continued for 2.5 additional hours . Cells from 5-mlcultures were collected by centrifugation and washed twice inphosphate-buffered saline [140 mM NaCl, 27 mM KCl, 101 mM Na₂HPO₄, 18 mM KH₂PO₄, pH 7.3] . The pellets were resuspended in bufferA [20 mM Tris-HCl [pH 8.0], 200 mM NaCl] supplemented with 0.5mM phenylmethylsulfonyl fluoride and 1 mg of lysozyme/ml . After the cells were incubated on ice for 30 min, Triton X-100 [1% final concentration], 5 µg of DNase/ml, and 5 µgof RNase/ml were added . The samples were further incubated onice for 15 min, and the cells were then broken by sonication[five pulses of 15 s each with 15-s intervals, on ice] and centrifugedfor 10 min at 12,000 x g . The pellet was dissolved in 400 µlof buffer B [20 mM Tris-HCl [pH 8.0], 200 mM NaCl, 5 M urea] and gently stirred on ice for 1 h . The samples were centrifuged in a microcentrifuge for 10 min at 12,000 rpm, and the supernatants were recovered . For purification of the His₆-tagged proteins, the supernatants were each mixed with 20 µl of Ni-nitriloacetate agarose [Qiagen] and stirred on ice for 45 min . The resin was recovered by centrifugation and washed three times with bufferA . The fusion proteins were eluted with 100 mM imidazole [forH6-DegU] or 200 mM imidazole [for H6-DegU32] in buffer A . Priorto their utilization for gel retardation analysis, the proteinswere dialyzed against water.

Electrophoretic mobility shift assays. Electrophoretic mobility shift assays were performed as describedby Hamoen et al . [12] . DNA fragments from the codY-flgB intercistronicregion were obtained by PCR amplification with the primers describedin Fig. 10A and Table 2 . Chromosomal DNA from strain PB168 wasused as a template . After purification with a PCR purificationkit [Qiagen], the fragments were end labeled with T4 polynucleotidekinase and [-³²P]ATP [2,500 Ci/mmol, 10 mCi/ml; Amersham] . Theprobes were purified with a PCR purification kit [Qiagen], andapproximately 1 ng of probe was mixed with the protein in bindingbuffer [20 mM Tris-HCl [pH 8.0], 100 mM KCl, 5 mM MgCl₂, 0.5mM dithiothreitol, 10% [vol/vol] glycerol, 0.05% Nonidet P-40,0.05 mg of bovine serum albumin/ml] containing 0.05 µgof sonicated salmon sperm DNA/ml as a nonspecific competitor.After incubation at 37°C for 20 min, the samples were loadedinto a nondenaturing 6.5% polyacrylamide gel . Gels were runin TAE buffer [40 mM Tris-acetate [pH 8.0], 2 mM EDTA] at 10mA [65 V] for about 90 min and then were autoradiographed [HyperfilmMP] . B . subtilis specific competitor DNA was obtained by PCRand used at a 100-fold excess over the amount of labeled probe.

 
DegU-P represses motility. B . subtilis strains with the degU32[Hy] mutation are nonmotile[2, 17] [see Fig . 3] . Using the lacZ reporter gene transcriptionallyfused to the hag flagellin promoter [Phag], we confirmed thattranscription is completely abolished in the degU32[Hy] background[Table 3] . It has been reported that the degU32[Hy] mutationcauses a strong reduction in the expression of a sigD-lacZ translational fusion [38] . Using anti-SigD antibodies for an immunoblot analysis,we confirmed that in the mutant strain the SigD protein is barelydetectable [data not shown] . In addition, we measured the expressionof the fla-che operon, in which sigD is present, by using alacZ transcriptional fusion . For this purpose, we used the pMutin4vector [39], in which a fragment derived from the ylxF openreading frame was cloned . ylxF is the ninth open reading frameof the fla-che operon . As reported in Fig . 2, in the wild-type background ß-galactosidase activity reached the maximalvalue at approximately the end of exponential growth and eventually decreased . In the presence of the degU32[Hy] mutation, no ß-galactosidase activity could be detected [Fig . 2 and Table 3] . The expressionof hag-lacZ could be rescued by the production of SigD froma plasmid [Table 3] . This was obtained by the transformationof strain PB5267 [degU32] with plasmid pSigD, which containssigD under the control of the IPTG-inducible Pspac promoter[3] . Nevertheless, the transformants were nonmotile . In accordancewith this phenotype, the expression of the fla-che operon, measured as the ß-galactosidase activity of the ylxF-lacZ fusion,was not affected by the induction of SigD and remained at undetectable levels [Table 3] . This suggests that the lack of flagella and motility cannot be ascribed only to the absence of an active SigD factor.

 
We observed the same drastic reduction in Phag-dependent transcription in the presence of the degS200[Hy] mutation [Table 3] . Sinceboth mutations cause an increase in the half-life of the phosphorylatedform of DegU [DegU-P] [8, 37], we propose that the observedtranscriptional repression is dependent upon the presence ofDegU-P . This interpretation is indirectly supported by the lackof any effect on hag-lacZ expression of a deletion involvingdegS and degU [data not shown] . In addition, we transferredthe deletion of degSU by transformation into the degU32[Hy]strain . For this purpose, we used a mutant in which the degSUoperon is replaced with a kanamycin resistance determinant [23]. The Kan^r transformants recovered motility completely [data not shown] . The same deletion introduced into the Mot⁺ suppressed strain PB5248 [see below] did not alter the Mot⁺ phenotype. These results support an active role for DegU-P in repressing motility.

Isolation and mapping of a suppressor mutant. To understand the mechanism of action of DegU-P, we searchedfor a spontaneous suppressor mutant of the Mot^– phenotypein a degU32[Hy] background . Strain PB5213 [degU32 Mot^–]was spotted onto a motility plate . After incubation, cells fromthe peripheral boundaries of the growth zone were isolate andcharacterized . One motile strain, PB5248, which retained theoriginal degU32[Hy] mutation, as assessed by sequencing andthe protease hyperproduction phenotype, was chosen for characterization.The suppressing mutation was called dhsA245, for degU high suppressor.The Mot⁺ phenotype was stable, even after several passages througha single colony . On motility plates, the suppressed strain movedto an extent that was comparable to that of the parental strainPB168 [Fig. 3] . It also largely recovered expression from the hag promoter, as shown by the level of ß-galactosidase activity [Table 3].

The suppressor was genetically mapped by PBS1-mediated transduction, with lysates obtained from a collection of derivatives of strain PB168 with the erythromycin resistance marker present in defined genes around the chromosome used as donors [32] . Transductantsselected for Ery^r were screened for motility; the replacementof the dhsA245 mutant allele with the wild-type gene was expectedto confer a Mot^– phenotype . Of 18 markers tested, 5 consistentlyshowed significant linkage to dhsA245, indicating that it mappedat approximately 144° [Fig . 4] . It should be pointed outthat this position is far from the location of gene degU at311° . Transformation crosses allowed us to restrict theregion to the portion of the chromosome comprising the areabetween the ylqB and ylxF genes [Fig . 4] . Further refinement of the genetic map was not attempted . To pinpoint the dhsA245 mutation, we determined the nucleotide sequence of approximately20 kb in the region . The sequence derived from strain PB5248[degU32[Hy] dhsA245] differed from the sequence of the referencestrain 168 by only a 245-bp deletion [see Fig . 7] . The deletion encompassed the end of the codY gene and part of the intercistronic region between codY and flgB . As a result of the mutation, thededuced CodY protein is 251 residues long, instead of 259, and the last 8 residues differ from the wild-type protein . The span of the deletion should be sufficient to abolish the activityof the protein, as reported for an insertion mutation, cod-37,that causes a frame shift and results in a polypeptide of 251amino acids [35] . We monitored the expression of a hut-lacZ and a dpp-lacZ fusion, which were previously reported to be negatively controlled by CodY [10] . The expression of the twofusions was significantly increased in strains with the dhsA245mutation, in agreement with the inactivation of gene codY inthe suppressed mutant [data not shown] . Downstream of gene codY,the deletion eliminates the _D-dependent promoter P_D-3 of thefla-che operon, which contributes marginally to the expressionof the operon [9, 40].

 
To evaluate the role of the deletion as a suppressor of degU32[Hy], we transduced the recipient strain PB5248 [degU32[Hy] dhsA245] with a PBS1 lysate derived from the BFA 2626 strain, which has an Ery^r determinant linked to the pksL gene [Fig. 4] . As expectedfrom the distance between pksL and dhsA245, approximately 50%of the Ery^r transductants were Mot^– . We assayed for thepresence or absence of the deletion by PCR; all 11 Mot^– transductants tested had the donor [wild-type] configurationand all 10 Mot⁺ isolates retained the deletion of the recipient.We concluded that the deletion is responsible for the suppressionof the degU32[Hy] Mot^– phenotype.

Gene codY and the promoter P_D-3 are not responsible for suppression. Two explanations for the mechanism of suppression by the dhsA245deletion mutation are possible: it occurs either by the inactivationof gene codY or by the elimination of the P_D-3 promoter region.CodY has been implicated in the nutritional repression of flagellinexpression observed during early stages of exponential growthin complex media [4, 21] . Even though DegU phosphorylation is predominant late in exponential growth and at the beginningof stationary phase, a possible involvement of CodY in the repression exerted by DegU-P could not be ignored . Transcription from theP_D-3 promoter does not contribute significantly to the expressionof the fla-che operon [40], but the effects of a deletion mutationaffecting only P_D-3 have not been investigated . To address thesequestions, we analyzed the effects of single deletions of codYand P_D-3.

First, we constructed a strain with one copy of the codY gene disrupted and with a second copy under the control of the IPTG-inducible Pspac promoter [Fig . 1B] . To this end, we used the integrativeplasmid pMutin4 [39], in which we cloned a fragment of 450 bp,extending from the middle part of clpY [the gene immediatelyupstream of codY] to the middle of codY . We transformed theparental PB168 strain with the pMutin4 derivative . The transformantswere Mot⁺ in both the presence and the absence of the inducerIPTG [data not shown] . We concluded that the inactivation ofcodY [absence of inducer] or [over]expression of CodY [in thepresence of 1 and 3 mM IPTG] did not interfere with motilityin an otherwise wild-type background . The chromosomal DNA ofone of the transformants [PB5290] was used to transfer the constructioninto PB5213 [degU32[Hy]] . The new transformants were still Mot^–,in both the presence and the absence of the IPTG inducer [datanot shown] . The same result, a lack of suppression of the degU32[Hy]Mot^– phenotype, was observed by the transfer into PB5213of the codY deletion mutation described by Serror and Sonenshein[34] . All transformants with the deletion were Mot^–, andthe level of expression of hag-lacZ fusions was very low, comparable to that obtained in the parental PB5213 strain [Table 3] . Weconcluded that the codY gene is not involved in the repressionexerted by DegU-P.

The possible involvement of the ^D-dependent promoter P_D-3 wasevaluated through the construction of a 191-bp deletion extendingfrom immediately downstream of the codY stop codon to approximatelyhalf of the codY-flgB intercistronic region . The P_D-3 sequencewas completely deleted, whereas the P_A and surrounding sequenceswere unaltered . The deleted sequence was replaced with the kanamycin resistance determinant, with the direction of transcriptionbeing opposite that of the fla-che operon [Fig . 1A] . In an otherwisewild-type background, the P_D-3 deletion did not interfere withmotility, with the expression of hag-lacZ fusions, or with transcriptionof the fla-che operon [Fig. 5 and data not shown] . When it wastransferred into the degU32[Hy] strain, the deletion did notrestore the Mot⁺ phenotype and did not allow expression of thehag-lacZ transcriptional fusion and the fla-che operon [Fig. 5] . We thus conclude that, under these experimental conditions,the P_D-3 promoter does not affect the expression of the fla-cheoperon and is not involved in the DegU-P-dependent repressionof motility.

 
Isolation of additional degU32[Hy] suppressor mutants. From the above results, it appears that suppression of the Mot^– phenotype of the degU32[Hy] mutants depends on the presenceof a deletion extending from the end of codY to upstream ofthe P_A promoter of the fla-che operon . It is also evident that deletion of the codY gene or of the sequence of the P_D-3 promoterdoes not restore the Mot⁺ phenotype . To obtain independent indicationsof the nature of the suppression and the mechanism of actionof DegU[-P], we isolated additional suppressor mutants . Nineindependent mutants were isolated and characterized at the geneticand molecular level, and they were designated dhsA1 to -6, -8,-10, and -11 . All of the mutants maintained the original highlevel of exoprotease production and the degU32 mutation [datanot shown] . All of the suppressor mutations were geneticallylinked to the erythromycin resistance determinant present inylq . The cotransduction frequencies were high, ranging from81% for the dhsA6 mutant to 98% for the dhsA11 mutant . The mutationswere characterized by PCR amplification [Fig . 6] and DNA sequencing. Seven mutations were deletions in the codY-flgB region and extending from bp 73 of dhsA11 to bp 1065 of dhsA5 . This last deletion extended into gene clpY, upstream of codY, and did not give a PCR product with the pair of oligonucleotides used [Fig . 6, lane 8] . The dhsA3 mutation was identical to the deletion in strain PB5248 . The only region missing from all of the deletions started at the nucleotide immediately following the codY stop codon and ended 37 nucleotides further downstream . The consequence of the deletion was the complete removal of the putative intrinsic termination site of transcription that is normally present immediately downstream of the codY gene [Fig . 7] . The simplest interpretationof this result is that in the absence of the termination signal,transcription from the cod operon can proceed through the intercistronicregion and bypass the control exerted by DegU[-P] . Two of thedhs mutations [dhsA6 and dhsA10] were identical point mutations,changing one A residue to a T residue [on the noncoding strand].The mutation altered the second position of the –35 hexamerof the P_A promoter in front of the fla-che operon . The change,from TAGACT to TTGACT, increases the match of the –35hexamer to the consensus sequence [TTGACA] and should increasethe affinity of RNA polymerase for the promoter [14].

 
Transcriptional readthrough. To evaluate the presence of readthrough transcripts in the suppressedstrain PB5248, we used RT-PCR analysis . RNAs were purified fromboth parental [PB168] and dhsA245 [PB5248] cells at three timepoints during the growth curve, i.e., at the transition fromexponential to stationary phase [T₀], 1 h earlier [T_–1],and 2 h into stationary phase [T₂] . The in vivo transcriptionproducts were analyzed by a two-step procedure [Fig . 8] . First,a single filament cDNA was obtained by RT primed with oligonucleotideflgB1, which corresponded to a sequence within the flgB openreading frame [Fig . 8A] . Subsequently, a small aliquot of the product of the first reaction was diluted 13 times, and twodifferent PCRs were performed . The primer flgB2 was used inboth PCRs; this primer was derived from within the flgB sequence,upstream of the flgB1 primer used for RT . In one set of reactions,amplification was carried out with primer PA, correspondingto the P_A promoter, to obtain a product of 115 bp derived fromthe transcript synthesized from the P_A promoter and from thehypothetical readthrough product . The second set of reactionsused oligonucleotide 7901, which hybridizes to the codY gene,in addition to the flgB2 primer [Fig . 8A] . The amplificationproduct derived from the codY-flgB readthrough transcript wasexpected to be 948 bp long . In the presence of the dhsA245 deletion, the amplification product was expected to be 703 bp long . Control reactions without reverse transcriptase were included in each experiment . As expected, a 115-bp product was obtained fromreactions derived from extracts of both the parental [PB168]and the suppressed PB5248 Mot⁺ strain; the degU32[Hy] RNA didnot give any product [data not shown] . For the parental strain,the 115-bp band was present at T_–1 and T₀ and absent fromthe T₂ sample . For the suppressed mutant, the band was presentat T₂ as well . The 703-bp PCR product corresponding to a readthroughfrom the cod operon was detected at all time points examinedfor the strain with the dhsA245 deletion [Fig . 8B] . The productwas absent from the RT-PCR performed with RNA from the parentalstrain PB168 [Fig . 8B] . These results confirm the interpretation that the Mot⁺ phenotype shown by the degU32[Hy] suppressed strainsis due to expression of the fla-che operon by readthrough transcriptiondriven from the cod operon promoter . For some experiments, afaint band of 948 bp was detected in the reaction mixture fromthe parental strain RNA, suggesting that some readthrough mayoccur in wild-type cells.

The significance of transcriptional readthrough in the suppression of the degU32 Mot^– phenotype was also supported by the behavior of strain PB5323, in which the dhsA4 mutation was moved away from the cod operon . The strain was derived from the suppressedMot⁺ dhsA4 mutant by the insertion of plasmid pPB01, a derivativeof pJM114 carrying the complete codY coding sequence, into thechromosome . As a result of a Campbell-type insertion [Fig . 1D],an intact codY gene was present at the original position inthe cod operon, whereas the fla-che operon and its regulatoryregion were separated from the cod operon by plasmid sequences.As a result of this manipulation, the strain derived from thesuppressed Mot⁺ dhsA4 mutant became Mot^– [Fig . 3] . This result can be interpreted as due to a lack of readthrough and thus a lack of expression of the fla-che operon, whose transcription is repressed by DegU[-P].

Further support for the interpretation that transcriptional readthrough can account for the suppression of the degU32 Mot^– phenotype was obtained by combining the dhsA245 deletion with a codY gene under the control of the Pspac promoter [Fig. 1E].In this genetic background, the suppression [Mot⁺] was strictlydependent on IPTG induction [Fig. 1E and 9] . In the controlstrain PB5291, with the putative transcription terminator downstreamof codY, IPTG induction did not affect the Mot^– phenotype [Fig . 1B and 9].

 
DegU specifically binds the P_A promoter of the fla-che operon. The results reported above strongly suggest that DegU[-P] acts as a repressor of transcription of the fla-che operon . DegU has been shown to bind to the comK and aprE promoter regions [12, 26] . Electrophoretic mobility shift assays were used tomonitor the ability of DegU to bind to the fla-che promoterregion . Probes of various lengths derived from the intercistronicregion between codY and flgB [Fig . 10A] were radiolabeled andincubated with the DegU protein . We used purified DegU and DegU32as N-terminal His₆-tag fusion proteins [Fig . 10B] . The sameresults were obtained with both proteins . In addition, we alsotested partially purified DegU and DegU32 proteins without theHis tag, again with the same type of results [data not shown].

As shown in Fig . 10C, DegU binds specifically to the DNA regioncontaining the P_A promoter of the fla-che operon . The DegU proteinfailed to bind to probes from upstream or downstream of P_A [Fig.10A]; moreover, these DNA fragments failed to compete with thespecific probe for DegU binding [Fig . 10C] . The binding specificity was further demonstrated by the complete competition of thereaction obtained with an excess of unlabeled DNA probe [Fig. 10C] . Thus, we conclude that the P_A promoter region of the fla-cheoperon is a target for DegU binding.

 
The isolation and characterization of suppressor mutations ofthe Mot^– phenotype due to the presence of the degU32[Hy]allele have shed some light on the mechanism of action of DegUon motility . All of the independent mutants isolated were alteredin the codY-flgB region of the chromosome.

No additional locus was found to be involved in suppression, strongly suggesting that DegU[-P] is the only trans-acting factor and that it has only one target, the codY-flgB intercistronic region.

The observed mutations fell into two groups, namely, deletionsand point mutations . The deletions had different lengths, butall had in common the removal of the transcription terminationsignal of the cod operon . Our results indicate that as an effectof the deletion, transcription from the cod promoter can continue into the fla-che operon . Since this readthrough is sufficient to allow the production of flagella and the acquisition of theMot⁺ phenotype, it means that DegU[-P] acts by preventing transcriptionof the fla-che operon . This is a simple, straightforward model: DegU[-P] binds to the sequence in the proximity of P_A, thus preventing binding of the RNA polymerase and transcription initiation. In fact, by performing a gel retardation analysis, we demonstrated that DegU binds to the P_A promoter region of the fla-che operon.DegU has been shown to bind to the comK promoter [12], and morerecently, to the aprE promoter [26] . In our model, deletionof the transcription terminator allows RNA polymerase to proceedbeyond the cod operon and to overcome the block representedby the DegU[-P] repressor bound to the fla-che operator site.

The repressor-operator model was also supported by the effectof two point mutations that changed the –35 hexamer ofthe P_A promoter to a nearly perfect consensus sequence [14]. Such a modification should allow a more efficient interaction between the RNA polymerase and the P_A promoter and should increase the stability of the closed complex . If the operator sequence were to interdigitate with the promoter sequence, which is possible, a point mutation increasing the affinity of RNA polymerase forthe promoter would reduce the effect of steric hindrance ofa negative regulator, favoring transcription against repression.

A repressive role of DegU[-P] on the fla-che operon is supported by the results of a whole-genome transcriptional analysis, in which the overexpression of DegU mimicked the phosphorylationstate of the response regulator [25] . The fla-che operon geneswere among those that were down-regulated by the overproduction of DegU [20, 25] . Furthermore, the expression of degSU is inducedby growth in a high-salinity medium, and under these conditions,transcription of the fla-che operon is strongly repressed [36].

Upon entering stationary phase, several different options are available to the starving bacterial cell . Among these for B. subtilis is the production of extracellular enzymes that scavenge the environment and free new resources . We may assume that the synthesis and secretion of degradative enzymes require a fairamount of energy, which is especially valuable in cells confrontedwith the depletion of nutrients . A different option consistsof motility and chemotactic behavior, enabling the cells tomove to new territories . This reaction also requires a largeamount of energy, especially to assemble and maintain the flagellarfilaments . Hence, it is not unexpected that the two responsesare alternative . In fact, motility and enzyme secretion appearto be sequentially activated, with motility reaching its maximalactivity at approximately T_–1 and the secretion of exoenzymespeaking only after the beginning of the stationary phase . TheDegS-DegU two-component system is involved in the regulationof the two alternative adaptive pathways . We suggest that thelevel of phosphorylated DegU negatively controls motility andactivates exoenzyme production . An increase in DegU phosphorylationat the end of the exponential growth phase could be responsiblefor the observed decrease in transcription of the flagellum-relatedgenes [3] . The role of DegU-P on degradative enzyme synthesisis well documented [22] . The temporal sequence of events occurringaround the transition phase may be dictated by the state ofthe DegU response regulator, which in turn affects the stabilityof the protein.

The fla-che operon comprises the majority of the genes whose products are structural components of the flagellum, in addition to several genes involved in chemotaxis . The sigD cistron, coding for ^D, is at the end of the 26-kbp operon, and its expressionis essential for transcription of the late genes . In additionto being a target of DegU, the fla-che operon promoter regionis the target of the CodY nutritional repressor that controlsthe expression of the operon in response to nutrient availability[4] . The high expression level of the fla-che operon observedin strains with the dhsA245 deletion [Fig . 2] is probably a consequence of a lack of control exerted by both CodY and DegU upon the operon . The deletion inactivates the codY gene and removes the transcription terminator, allowing readthrough.The operon is released from the combined effects of CodY andDegU and is transcribed at an unusual level.

The on-off switch of the fla-che operon appears to be the first step in the cascade of events leading to the synthesis of the flagellar components and to the assembly and functioning ofthe organelle . Thus, the fla-che operon is in a pivotal position in the regulation of motility and chemotaxis in B . subtilis.

 
We are grateful to Eugenio Ferrari for suggestions during theearly stages of this work, to Cinzia Calvio for critical readingsof the manuscript, and to Elisabetta Andreoli for technicalhelp . We thank the anonymous reviewer who suggested the experimentreported in Fig. 9.

This work was supported by Università degli Studi diPavia [Fondo d'Ateneo per la Ricerca], the "C.N.R . Target Projecton Biotechnology," and grants from MIUR [Ministero dell'Istruzione, Università e Ricerca, Italy].

 
* Corresponding author . Mailing address: Dipartimento di Genetica e Microbiologia, Via Abbiategrasso 207, 27100 Pavia, Italy . Phone: 39-0382-505548 . Fax: 39-0382-528496 . E-mail: galizzi@ipvgen.unipv.it.

G.A . and P.B . contributed equally to this work.

Present address: Smurfit Institute, Department of Genetics,Trinity College, Dublin, Ireland.

1. Allmansberger, R. 1997 . Temporal regulation of sigD from Bacillus subtilis depends on a minor promoter in front of the gene . J . Bacteriol . 179:6531-6535.
2. Ayusawa, D., Y . Yoneda, K . Yamane, and B . Maruo. 1975 . Pleiotropic phenomena in autolytic enzyme content, flagellation, and simultaneous hyperproduction of extracellular -amylase and protease in a Bacillus subtilis mutant . J . Bacteriol . 124:459-469.
3. Barillà, D., T . Caramori, and A . Galizzi. 1994 . Coupling of flagellin gene transcription to flagellar assembly in Bacillus subtilis . J . Bacteriol . 176:4558-4564.
4. Bergara, F., C . Ibarra, J . Iwamasa, J . C . Patarroyo, R . Aguilera, and L . M . Márquez-Magaña. 2003 . CodY is a nutritional repressor of flagellar gene expression in Bacillus subtilis . J . Bacteriol . 185:3118-3126 .
5. Bertin, P., E . Terrao, E . H . Lee, P . Lejoune, C . Colson, A . Danchin, and E . Collatz. 1994 . The H-NS protein is involved in the biogenesis of flagella in Escherichia coli . J . Bacteriol . 176:5537-5540.
6. Caldwell, R., R . Sapolsky, W . Weyler, R . R . Maile, S . C . Causey, and E . Ferrari. 2001 . Correlation between Bacillus subtilis scoC phenotype and gene expression determined using microarrays for transcriptome analysis . J . Bacteriol . 183:7329-7340 .
7. Chilcott, G . S., and K . T . Hughes. 2000 . Coupling of flagellar gene expression to flagellar assembly in Salmonella enterica serovar Typhimurium and Escherichia coli . Microbiol . Mol . Biol . Rev . 64:694-708 .
8. Dahl, M . K., T . Msadek, F . Kunst, and G . Rapoport. 1992 . The phosphorylation state of the DegU response regulator acts as a molecular switch allowing either degradative enzyme synthesis or expression of genetic competence in Bacillus subtilis . J . Biol . Chem . 267:14509-14514 .
9. Estacio, W., S . Santa Anna-Arriola, M . Adedipe, and L . M . Márquez-Magaña. 1998 . Dual promoters are responsible for transcription initiation of the fla/che operon in Bacillus subtilis . J . Bacteriol . 180:3548-3555 .
10. Fisher, S . H., K . Rohrer, and A . E . Ferson. 1996 . Role of CodY in regulation of the Bacillus subtilis hut operon . J . Bacteriol . 178:3779-3784.
11. Fredrick, K., T . Caramori, Y.-F . Chen, A . Galizzi, and J . D . Helmann. 1995 . Promoter architecture in the flagellar regulon of Bacillus subtilis: high-level expression of flagellin by the ^D RNA polymerase requires an upstream promoter element . Proc . Natl . Acad . Sci . USA 92:2582-2586.
12. Hamoen, L . W., A . F . Van Werkhoven, G . Venema, and D . Dubnau. 2000 . The pleiotropic response regulator DegU functions as a priming protein in competence development in Bacillus subtilis . Proc . Natl . Acad . Sci . USA 97:9246-9251 .
13. Hay, N . A., D . J . Tipper, D . Gygi, and C . Hughes. 1997 . A nonswarming mutant of Proteus mirabilis lacks the Lrp global transcriptional regulator . J . Bacteriol . 179:4741-4746.
14. Helmann, J . D. 1995 . Compilation and analysis of Bacillus subtilis ^A-dependent promoter sequences: evidence for extended contact between RNA polymerase and upstream promoter DNA . Nucleic Acids Res . 23:2351-2360.
15. Henner, D . J., E . Ferrari, M . Perego, and J . A . Hoch. 1988 . Location of the targets of the hpr-97, sacU32 [Hy], and sacQ36 [Hy] mutations in upstream regions of the subtilisin promoter . J . Bacteriol . 170:296-300.
16. Hoch, J . A., M . Barat, and C . Anagnostopoulos. 1967 . Transformation and transduction in recombination-defective mutants of Bacillus subtilis . J . Bacteriol . 93:1925-1937.
17. Kunst, F., M . Pascal, J . Lepesant-Kejzlarová, J.-A . Lepesant, A . Billault, and R . Dedonder. 1974 . Pleiotropic mutations affecting sporulation conditions and the synthesis of extracellular enzymes in Bacillus subtilis 168 . Biochimie 56:1481-1489.
18. Kunst, F., and G . Rapoport. 1995 . Salt stress is an environmental signal affecting degradative enzyme synthesis in Bacillus subtilis . J . Bacteriol . 177:2403-2407.
19. Macnab, R . M. 1996 . Flagella and motility, p . 123-145. In F . C . Neidhart et al . [ed.], Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed . ASM Press, Washington, D.C.
20. Mäder, U., H . Antelmann, T . Buder, M . K . Dahl, M . Hecker, and G . Homuth. 2002 . Bacillus subtilis functional genomics: genome-wide analysis of the DegS-DegU regulon by transcriptomics and proteomics . Mol . Genet . Genomics 268:455-467.
21. Mirel, D . B., W . F . Estacio, M . Mathieu, E . Olmsted, J . Ramirez, and L . M . Márquez-Magaña. 2000 . Environmental regulation of Bacillus subtilis ^D-dependent gene expression . J . Bacteriol . 182:3055-3062 .
22. Msadek, T. 1999 . When the going gets tough: survival strategies and environmental signalling networks in Bacillus subtilis . Trends Microbiol . 7:201-207.
23. Msadek, T., F . Kunst, D . Henner, A . Klier, G . Rapoport, and R . Dedonder. 1990 . Signal transduction pathway controlling synthesis of a class of degradative enzymes in Bacillus subtilis: expression of the regulatory genes and analysis of mutations in degS and degU . J . Bacteriol . 172:824-834.
24. Nishida, S., T . Mitzushima, T . Miki, and K . Sekimizu. 1997 . Immotile phenotype of an Escherichia coli mutant lacking the histone-like protein HU . FEMS Microbiol . Lett . 150:297-301.
25. Ogura, M., H . Yamaguchi, K . Yoshida, Y . Fujita, and T . Tanaka. 2001 . DNA microarray analysis of Bacillus subtilis DegU, ComA and PhoP regulons: an approach to comprehensive analysis of B . subtilis two-component regulatory systems . Nucleic Acids Res . 29:3804-3813 .
26. Ogura, M., K . Shimane, K . Asai, N . Ogasawara, and T . Tanaka. 2003 . Binding of response regulator DegU to the aprE promoter is inhibited by RapG, which is counteracted by extracellular PhrG in Bacillus subtilis . Mol . Microbiol . 49:1685-1697.
27. Ordal, G . W., L . Márquez-Magaña, and M . J . Chamberlin. 1993 . Motility and chemotaxis, p . 765-784 . In A . L . Sonenshein, J . A . Hoch, and R . Losick [ed.], Bacillus subtilis and other gram-positive bacteria: biochemistry, physiology, and molecular genetics . American Society for Microbiology, Washington, D.C.
28. Perego, M. 1993 . Integrational vectors for genetic manipulation in Bacillus subtilis, p . 615-624 . In A . L . Sonenshein, J . A . Hoch, and R . Losick [ed.], Bacillus subtilis and other gram-positive bacteria: biochemistry, physiology, and molecular genetics . American Society for Microbiology, Washington, D.C.
29. Perego, M., and J . A . Hoch. 1988 . Sequence analysis and regulation of the hpr locus, a regulatory gene for protease production and sporulation in Bacillus subtilis . J . Bacteriol . 170:2560-2567.
30. Petit, M., E . Dervyn, M . Rose, K . Entian, S . McGovern, S . D . Ehrlich, and C . Bruand. 1998 . PcrA is an essential DNA helicase of Bacillus subtilis fulfilling functions both in repair and rolling-circle replication . Mol . Microbiol . 29:261-273.
31. Sambrook, J., E . F . Fritsch, and T . Maniatis. 1989 . Molecular cloning: a laboratory manual . Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
32. Schumann, W., S . D . Ehrlich, and N . Ogasawara [ed.]. 2001 . Functional analysis of bacterial genes . A practical manual . John Wiley and Sons, Ltd . Chichester, United Kingdom.
33. Serror, P., and A . L . Sonenshein. 1996 . CodY is required for nutritional repression of Bacillus subtilis genetic competence . J . Bacteriol . 178:5910-5915.
34. Silverman, M., and M . Simon. 1974 . Characterization of Escherichia coli flagellar mutants that are insensitive to catabolite repression . J . Bacteriol . 120:1196-1203.
35. Slack, F . J., P . Serror, E . Joyce, and A . L . Sonenshein. 1995 . A gene required for nutritional repression of the Bacillus subtilis dipeptide permease operon . Mol . Microbiol . 15:689-702.
36. Steil, L., T . Hoffmann, I . Budde, U . Völker, and E . Bremer. 2003 . Genome-wide transcriptional profiling analysis of adaptation of Bacillus subtilis to high salinity . J . Bacteriol . 185:6358-6370 .
37. Tanaka, T., M . Kawata, and K . Mukai. 1991 . Altered phosphorylation of Bacillus subtilis DegU caused by single amino acid changes in DegS . J . Bacteriol . 173:5507-5515.
38. Tokunaga, T., M . H . Rashid, A . Kuroda, and J . Sekiguchi. 1994 . Effect of degS-degU mutations on the expression of sigD, encoding an alternative sigma factor, and autolysin operon of Bacillus subtilis . J . Bacteriol . 176:5177-5180.
39. Vagner, V., E . Dervyn, and S . D . Ehrlich. 1998 . A vector for systematic gene inactivation in Bacillus subtilis . Microbiology 144:3097-3104.
40. West, J . T., W . Estacio, and L . M . Márquez-Magaña. 2000 . Relative roles of the fla/che P_A, P_D-3, and P_sigD promoters in regulating motility and sigD expression in Bacillus subtilis . J . Bacteriol . 182:4841-4848 .

Free Online Full-text Article

What Is Biofilter?, What Is Molecular Microbiology?, What Is Protein?, What Is Bioengineering?, What Is Biotechnology?, r, Microorganisms, o, Bacterium, e, Bacteriology, i, Microorganism, a, Microbes, s, Biological filters, n, Microbiological, s, Escherichia coli, e, Bacillus, o, Streptococci, e, Escherichia coli, r, Sepsis, c, Multidrug resistant, c, Bacillus subtilis, e, S. cerevisiae, r, Salmonella typhimurium, o, S. cerevisiae, e, Salmonella typhimurium, r, Microbiological, i, Cell suspensions, r, Escherichia coli, o, Antibiotics, n, Antibiotics, e, Biocontactor, n, O157, n, Saccharomyces yeast




 

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005