1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Despite many years of intense work investigating the function of nucleoid-associated proteins in prokaryotes, their role in bacterial physiology remains largely unknown. The two-dimensional protein patterns were compared and expression profiling was carried out on H-NS-deficient and wild-type strains of Escherichia coli K-12. The expression of approximately 5% of the genes and/or the accumulation of their protein was directly or indirectly altered in the hns mutant strain. About one-fifth of these genes encode proteins that are involved in transcription or translation and one-third are known to or were in silico predicted to encode cell envelope components or proteins that are usually involved in bacterial adaptation to changes in environmental conditions. The increased expression of several genes in the mutant resulted in a better ability of this strain to survive at low pH and high osmolarity than the wild-type strain. In particular, the putative regulator, YhiX, plays a central role in the H-NS control of genes required in the glutamate-dependent acid stress response. These results suggest that there is a strong relationship between the H-NS regulon and the maintenance of intracellular homeostasis.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Considerable progress has been made recently in genome analysis. The complete genome sequences of 35 microorganisms, including several pathogens, have been determined, and over 100 microorganisms are currently being sequenced ( These studies have emphasized our lack of physiological understanding of living organisms, including model microorganisms, such as Escherichia coli (Blattner et al., 1997) and Bacillus subtilis (Kunst et al., 1997), which have been studied extensively. Indeed, the function of approximately 40% of the predicted genes has not been assigned (Blattner et al., 1997; Kunst et al., 1997). Moreover, many of the annotations assigned to genomes based on reference models are often either irrelevant or even spurious (Kyrpides and Ouzounis, 1999).

In enterobacteria, nucleoid-associated proteins are required for the organization of chromosomal DNA. These include HU, IHF, FIS and H-NS, which are small, abundant proteins (Talukder et al., 1999), as are several eukaryotic DNA-binding proteins (Hayat and Mancarella, 1995). H-NS was initially described as a transcription factor (Jacquet et al., 1971) and was later shown to play a role in the structure and functioning of chromosomal DNA (Atlung and Ingmer, 1997; Williams and Rimsky, 1997). In E. coli, this ≈ 15 kDa protein affects DNA supercoiling (Tupper et al., 1994) and condenses DNA (Dame et al., 2000) by preferentially binding to curved DNA in vitro (Jordi et al., 1997). These properties seem to be dependent on the ability of the N-terminal domain of H-NS to form oligomers (Ueguchi et al., 1996; Williams et al., 1996; Spurio et al., 1997). This region is predicted to be highly α-helical (Dorman et al., 1999; Bertin et al., 1999; Ceschini et al., 2000; Smyth et al., 2000), whereas the C-terminal domain, which has been resolved by nuclear magnetic resonance (NMR; Shindo et al., 1995), is a mixed α–β structure.

In E. coli, hns expression is growth phase-dependent and is subjected to autorepression (Atlung and Ingmer, 1997). Cold shock increases H-NS synthesis in E. coli (La Teana et al., 1991), and the synthesis of the orthologous protein, VicH, in Vibrio cholerae (Tendeng et al., 2000). Mutations in hns result in various phenotypes, because its product is involved in the regulation of apparently unlinked genes (Atlung and Ingmer, 1997; Bertin et al., 1999). The control of the proU (Jordi and Higgins, 2000), rrnB (Schröder and Wagner, 2000), bglGFB (Caramel and Schnetz, 1998) and flhDC (Soutourina et al., 1999) operons by H-NS has been studied extensively. Despite this, the function of H-NS in bacterial metabolism remains unclear. To elucidate the physiological role of this nucleoid-associated protein, we compared both the proteome and the transcriptome of an H-NS-deficient strain and its wild-type parent. These large-scale methods provided a wider view of the global regulation of gene expression in bacteria in response to multiple stresses.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Comparative analysis of two-dimensional protein patterns of wild-type and H-NS-deficient strains

A representative pattern of silver nitrate-stained proteins extracted from hns mutant strain BE1410 and the quantitative differences in levels of accumulated proteins between BE1410 and the wild-type strain, FB8, are shown in Fig. 1. The cellular content of at least 60 proteins was altered in the H-NS-deficient strain: the amount of 49 polypeptides was increased and that of 11 polypeptides was decreased in the hns mutant strain compared with the wild type (Fig. 1A). These observations substantiate and extend our recent data (Laurent-Winter et al., 1997; Bertin et al., 1999; Soutourina et al., 1999). To ensure that these modifications resulted from the hns mutation itself and not from any alteration depending on the genetic context, identical experiments were carried out with the E. coli K-12 reference strain MG1655 and its hns derivative. The differences in protein profiles between the wild-type and hns mutant strains were similar to those observed in FB8 and BE1410 (data not shown). Twenty-three spots, with molecular masses ranging from 18 to 100 kDa and pI of 4–7, accumulated to different levels in the two strains and were characterized by comparison with the E. coli reference map ( or by mass spectrometry (Fig. 1B and C).


Figure 1. Two-dimensional silver-stained map of the H-NS-deficient strain, BE1410, and comparative analysis of the level of accumulated proteins in the wild-type and hns mutant strains. Two-dimensional gel electrophoresis (A) was carried out as described in Experimental procedures. The 49 proteins represented by a circle and the 11 proteins represented by a rectangle correspond to polypeptides whose accumulation level was increased and decreased, respectively, in all experiments by at least a factor of two in BE1410 compared with the FB8 wild-type strain. The 23 identified spots are named and indicated by an arrow; those characterized by mass spectrometry are indicated by *. Protein levels are expressed as percentage volume, which corresponds to the percentage ratio between the volume of a single spot and the total volume of all spots present in a gel. The accumulation level of proteins with known function (B) and of putative or unknown function (C) in the wild-type strain (□) and in the hns mutant (▪) was quantified using the melanie II software (Bio-Rad). The data are the mean values ± standard deviations of four independent experiments. The blastp program and the Prosite patterns database were used to compare the amino acid sequences of proteins with putative or unknown function (C) with the updated non-redundant databases. HdeA (P26604) has been proposed to have a chaperone-like function in extremely acidic conditions in pathogenic enteric bacteria. YeiN, YeeN and YtfE were similar to proteins of unknown function from Thermotoga maritima, B. subtilis and Haemophilus influenzae, respectively. YbcL belongs to the UPF0098 protein family, which comprises proteins of unknown function from Mycobacterium tuberculosis and Archeoglobus fulgidus. The YeiP (P33028) sequence was similar to elongation factors from various bacteria and contained the elongation factor P signature (PS01275); YfhO (P39171) was similar to pyridoxal phosphate enzymes and contained the corresponding signature (class V pyridoxal phosphate attachment site, PS00595). YhiU (P37636) was similar to the acriflavin resistance protein from E. coli and contained the prokaryotic membrane lipoprotein lipid attachment site (PS00013).

Download figure to PowerPoint

The cellular content of several proteins whose accumulation is known to be controlled by H-NS was altered in the hns mutant strain BE1410 (Fig. 1A). These included ProX, Gnd, GadA and HdeA. Mass spectrometry identified several additional proteins. For example, CbpA, a curved DNA-binding protein induced by starvation (Richmond et al., 1999), showed an increased accumulation in the hns mutant (Fig. 1B). Similarly, eight hypothetical proteins of unknown function were accumulated to increased levels in the hns mutant (Fig. 1C). The blastp program (Altschul et al., 1990) and the Prosite pattern database (Hofmann et al., 1999) were used to compare the amino acid sequences of these proteins with those in the updated non-redundant databases: YeiP, YfhO and YhiU were similar to proteins of known function (Fig. 1C).

Expression profiling of wild-type and H-NS-deficient strains

H-NS has been described as a transcriptional regulator (Atlung and Ingmer, 1997). Therefore, to study the effect of this protein on the whole E. coli genome, RNA was isolated from BE1410 and its wild-type counterpart and analysed using DNA arrays spotted with the whole gene set of E. coli. To account for unspecific variations, experiments were carried out using at least three independent RNA preparations, from which at least two hybridizations were performed and two different sets of DNA arrays. Furthermore, to study the effect of H-NS on genes expressed at a low level, a second set of hybridizations was carried out with 10 times more (10 µg) total RNA. Comparison of the signal intensity of arrays from duplicates or from independent hybridizations (see Experimental procedures) showed that the results were highly reproducible (Fig. 2). A representative overlay of hns mutant and wild-type patterns is shown in Fig. 3. The numerous differences between wild-type and hns mutant strains show that H-NS has a major effect on gene expression. In particular, strongly depressed genes identified by this method (Fig. 3) encoded proteins that were shown to be accumulated to high levels on two-dimensional gels (Fig. 1). Expression intensities were above the background level for 2986 genes, which is comparable with previous results obtained with specific oligonucleotide primers (Tao et al., 1999) or with random hexamers (Arfin et al., 2000). A non-parametric statistical test (see Experimental procedures) showed that the expression level of about 250 genes was significantly different in the hns mutant strain compared with the wild type (P-value ≤ 0.05). Of these H-NS-regulated genes, approximately 20% had a well-known or putative function in processes such as transcription and translation. Over 35% were regulated in response to multiple environmental conditions or were involved in cell envelope composition (Fig. 4). The genes whose expression level differed by a factor ≥ 2 between the wild-type and mutant strains are listed in Table 1.


Figure 2. Comparison of signal intensity measured on DNA arrays from dot duplicates and from independent hybridizations. The reproducibility of the results obtained with 1 µg (A and C) and 10 µg of total RNA (B and D) extracted from the hns mutant was assessed before subtraction of the background.

A and B. Comparison of signal intensity for pairs of dots corresponding to each gene.

C and D. Comparison of signal intensity for each gene in two independent hybridizations. Similar results were obtained from wild-type strain cDNA hybridization (data not shown).

Download figure to PowerPoint


Figure 3. Overlay of DNA arrays hybridized with wild-type or hns mutant cDNA. Variations in gene expression were measured with DNA arrays containing the 4290 open reading frames (ORFs) of the E. coli genome hybridized with cDNA probes generated from RNA extracted from wild-type and hns BE1410 mutant strains. Macroarrays were scanned as described in Experimental procedures, and the adobe photoshop F1-4.0 software was used to produce the overlay of representative macroarrays. The macroarray hybridized with the hns mutant cDNA probes was coloured red and that with the wild-type cDNA probes was coloured green. Genes whose expression was not modified in H-NS-deficient and wild-type strains are coloured yellow, and those induced or repressed in the hns mutant are coloured red or green respectively. Genes whose product showed an altered accumulation level between wild-type and hns strains (Fig. 1) are indicated by an arrow and named.

Download figure to PowerPoint


Figure 4. Functional classification of H-NS-regulated genes identified by expression profiling on DNA arrays. Genes that showed a differential expression between wild-type and mutant strain (P-value ≤ 0.05) were classified according to their function (Moszer, 1998).

Download figure to PowerPoint

Table 1. Genes differentially expressed between E. coli K-12 strains FB8 (wild type) and BE1410 (hns mutant).
GeneabExpression ratio BE1410/FB8cd P-valueSWISSPROT and TrEMBL accession no.e Function Subcellular localizationf
  1. The results obtained are representative of at least four hybridizations from at least three independent RNA extractions. Only genes with a P-value ≤ 5E-2 in a Wilcoxon statistical test and showing a factor ≥ 2 differential expression between wild-type and hns strains were classified according to their function, i.e. cell envelope and cellular processes (A), intermediary metabolism (B), information pathways (C), other functions (D) or unknown functions (E). Genes that were previously shown to be regulated by H-NS (Atlung and Ingmer, 1997; Laurent-Winter et al., 1997) are indicated by a in the first column. The expression level of the genes indicated by b was also analysed with random hexamer primer-generated cDNAs synthesized from RNAs extracted by a rapid procedure and then hybridized on high-density filters designed in our laboratory. The differential expression measured between wild-type and mutant strains was similar to that obtained with Panorama arrays (F. Hommais, unpublished). An expression ratio measured from 10 µg of RNA is indicated by c in the second column, and genes whose expression is less than the background in one of the two strains are indicated by d in the third column. Columns four and five indicate protein accession number and protein function according to the SWISSPROT and TrEMBL databases ( or GenBank (e) respectively. The results from similarity searches carried out using the blastp program ( pL?) against non-redundant updated databases are shown in column five. Protein subcellular localization (column six) came from the Niceprot database ( or was predicted (f) using the psort software (

ACell envelope and cellular processes
appB b 2.672E-2P26458Cytochrome bd I oxidase subunit IIIntegral inner membrane
caiT −  2.279E-3P31553Probable PMF-driven uptake of carnitine and betainesIntegral inner membrane
cirA b −2.334E-2P17315Colicin I receptorOuter membrane
cpsB 2.705E-2P24174Mannose-1-phosphate guanylyltransferaseCytoplasmicf
csgA ab 2.834E-2P28307Major curlin subunitOuter membrane
cydA b 2.455E-2P11026Cytochrome d ubiquinol oxidase subunit IIntegral inner membrane
cydB 9.265E-2P11027Cytochrome d ubiquinol oxidase subunit IIIntegral inner membrane
cyoA b 10.205E-2P18400Cytochrome o subunit IIIntegral inner membrane
feoB 4.265E-2P33650Ferrous iron transport protein BIntegral inner membrane
fimB ab 12.395E-3P04742Fimbriae recombinaseCytoplasmic probable
fimI 6.612E-2P39264Fimbrin-like proteinOuter membrane
flgA −  2.444E-2P75933Basal body P-ring formation proteinPeriplasmic probable
flgE −  2.172E-2P75937Flagellar hook subunit proteinCytoplasmicf
flgH −  2.865E-2P75940Basal body l-ring proteinOuter membrane
flgI −  2.003E-2P75941Basal body P-ring proteinPeriplasmic
flhA −  2.009E-3P76298Flagellar synthesisIntegral inner membrane
fliC ab −12.509E-4P04949FlagellinExtracellular
fliR −  1.971E-3P33135Flagellar synthesisIntegral inner membrane
gadC b 13.675E-3P39183Putative amino acid antiporterIntegral inner membrane probable
glf 2.531E-2P37747UDP-galactopyranose mutaseInner membranef
gmd 8.693E-2P32054GDP-mannose 4,6-dehydrataseInner membranef
gmm 5.605E-2P32056HydrolaseCytoplasmicf
gspE 3.835E-2P45759Probable general secretion pathway protein ECytoplasmic probable
gspG 16.97c5E-2P41442Probable general secretion pathway protein GPeriplasmicf
gspM 3.055E-2P36678Probable general secretion pathway protein MInner membrane probable
gspO 4.205E-2P25960Type 4 prepilin-like protein leader peptide processing enzymeIntegral inner membrane probable
ompC a 3.241E-3P06996PorinIntegral outer membrane
ompF ab −3.853E-3P02931PorinIntegral outer membrane
ompN 2.502E-2P77747Similar to Q56111, outer membrane protein S2 (Salmonellatyphimurium); belongs to the OmpC/PhoE family of porinsIntegral outer membrane
ompX 3.585E-3P36546Similar to P16454, attachment invasion locus protein precursor (Yersinia enterocolitica); belongs to the Ail/OmpX/PagC/Lom family PS00694 and PS000695: Enterobacterial virulence outer membrane protein signaturesIntegral outer membrane
osmC ab 5.892E-3P23929Similar to osmotically inducible protein C (Deinococcus radiodurans)Inner membranef
rfaI 5.25c3E-2P27128Lipopolysaccharide 1,3-galactosyltransferasePeriplasmicf
rfaJ 3.97c2E-2P27129Lipopolysaccharide 1,2-glucosyltransferaseInner membranef
rfaK 4.52c5E-2P27242Lipopolysaccharide 1,2-N-acetylglucosaminetransferasePeripheral membrane
rfaL 2.002E-2P27243Lipopolysaccharide core biosynthesis, O-antigen ligaseIntegral membrane
rfaS 3.50c4E-2P27126Lipopolysaccharide core biosynthesis proteinInner membranef
rfaY 2.005E-2P27240Lipopolysaccharide core biosynthesisCytoplasmicf
slp b 22.964E-4P37194LipoproteinAttached to the outer membrane by a lipid anchor
tatA 2.002E-2P27856Translocation of periplasmic proteins, preservation of structures and ligandsIntegral membrane
tatE 2.305E-2P25895Translocation of periplasmic proteins, preservation of structures and ligandsInner membranef
wbbI 2.751E-2P37749Involved in lipopolysaccharide biosynthesisCytoplasmic probable
wbbJ b 4.20c5E-2P37750Putative lipopolysaccharide O-acetyltransferaseInner membranef
wbbK 2.152E-2P37751Lipopolysaccharide glycosyltransferase, transfer of UDP-galF and UDP-glucoseInner membrane-associated probable
wcaA 2.005E-2P77414Involved in lipopolysaccharide biosynthesisCytoplasmicf
wcaC 6.645E-2P71237Putative glycosyl transferaseInner membranef
wcaE 8.705E-2P71239Putative glycosyl transferaseCytoplasmicf
wcaF 4.695E-2P71240Putative acetyltransferaseCytoplasmicf
wcaI 6.305E-2P32057Putative glycosyl transferaseInner membranef
wcaJ 3.945E-2P71241UDP-glucose lipid carrier transferaseInner membranef
wzb 6.625E-2P77153Probable low-molecular-weight protein tyrosine phosphataseInner membranef
wzc b 7.125E-2P76387Putative membrane-associated ATP hydrolaseInner membranef
yadN 23.12cd5E-2P37050Similar to P12267, type 3 fimbrial protein MrkA precursor (Klebsiella pneumoniae)Outer membranef
yagX 2.525E-2P77802Similar to P25733, CFA/I fimbrial subunit C precursor (E. coli)Outer membranef
yaiP 22.70cd5E-2P77760Similar to Q54066, IcaA (Staphylococcus epidermidis)Inner membranef
ycbQ 8.861E-3P75855Similar to P12903, fimbrial subunit type 1 precursor (K. pneumoniae)Outer membranef
ycgH 2.502E-2AE000215eSimilar to AE000248 putative ATP-binding component of a transport system and adhesin  protein (E. coli)Outer membranef
yddA 10.065E-3P31826Similar to Q57335, Y036 hypothetical ABC transporter ATP-binding protein (Haemophilus influenzae) PS00211 and PS00017: ATP-GTP binding site motif A and ABC transporter familyIntegral membrane probable
ydeT 43.47c1E-2AE000247eSimilar to P30130 outer membrane protein; export and assembly of type 1 fimbriae (E. coli)Cytoplasmicf
yedV 6.171E-3P76339Similar to O44007, sensor protein CzcS (Ralstonia eutropha)Integral inner membrane probable
yhhZ b 2.565E-3P46855Similar to P80693 MajE, the major exported protein (Pseudomonas syringae)Extracellular probable
yjdE 6.133E-2P39269Similar to P23891, probable cadaverine/lysine antiporter (E. coli)Integral inner membrane probable
B Intermediary metabolism
gadA ab 56.004E-3P80063Glutamate decarboxylase-αCytoplasmic probable
gadB ab 56.324E-3P28302Glutamate decarboxylase-βCytoplasmic probable
gltD 2.765E-2P09832Glutamate synthase β-subunitCytoplasmic
gnd a 3.575E-3P003506-Phosphogluconate dehydrogenaseInner membranef
pflB 2.835E-2P09373Pyruvate formate lyase 1Cytoplasmicf
tdcE 2.754E-2P42632Pyruvate formate lyase or ketobutyrate formate lyaseCytoplasmic by similarity
ugd b 3.242E-2P76373UDP-glucose-6-dehydrogenaseInner membranef
ydeP b 7.476E-3P77561Similar to P06131, formate dehydrogenase α-chain (Methanobacterium formicicum)Cytoplasmicf
ydiO 32.70c5E-2P76200Similar to P31571, probable carnitine operon oxidoreductase (E. coli) PS00072 and PS00073: Acyl-coA dehydrogenases signatures 1 and 2Cytoplasmicf
ydjH b 4.604E-2P77493Similar to P44331, ribokinase RbsK homologue (H. influenzae) PS00583 and PS00584: carbohydrate kinase signatures 1 and 2Inner membranef
yfdW 8.081E-2P77407Similar to U82167 formyl-CoA transferase (Oxalobacter formigenes)Cytoplasmicf
C Information pathways
appY a 2.744E-3P05052Regulatory proteinCytoplasmic
aslB b 5.702E-2P25550Putative arylsulphatase regulatory proteinCytoplasmicf
evgA 7.35d2E-2P30854Regulator of the two-component regulatory system EvgA/EvgSCytoplasmic probable
gltF b 8.494E-3P28721Putative regulatory proteinCytoplasmic
hha 20.757E-3P23870Haemolysin expression modulating proteinCytoplasmicf
leuO a 8.692E-2P10151LysR regulatory proteinCytoplasmic
ligA 7.625E-2P15042DNA ligaseCytoplasmicf
lrp a 30.69c5E-2P19494Leucine-responsive regulatory proteinCytoplasmic
phoP b 2.721E-2P23836Regulator of the two-component system PhoP/PhoQCytoplasmic
pinQ b 2.697E-3P77170DNA invertaseCytoplasmicf
pinR b 8.723E-3P77574DNA invertaseCytoplasmicf
rcsA ab 2.601E-3P24210Regulatory proteinCytoplasmic
rplA b −3.702E-2P0238450S ribosomal subunit protein L1Cytoplasmic
rplB −  2.223E-2P0238750S ribosomal subunit protein L2Cytoplasmic
rplC −  2.001E-2P0238650S ribosomal subunit protein L3Cytoplasmic
rplF −  2.001E-2P0239050S ribosomal subunit protein L6Cytoplasmic
rplI b −3.335E-2P0241850S ribosomal subunit protein L9Cytoplasmic
rplK −  2.224E-3P0240950S ribosomal subunit protein L11Cytoplasmic
rplL −  2.082E-3P0239250S ribosomal subunit protein L7/L12Cytoplasmic
rplN −  4.005E-2P0241150S ribosomal subunit protein L14Cytoplasmic
rplO −  2.004E-4P0241350S ribosomal subunit protein L15Cytoplasmic
rplQ −  2.702E-2P0241650S ribosomal subunit protein L17Cytoplasmic
rplW −  2.772E-2P0242450S ribosomal subunit protein L23Cytoplasmic
rplX −  3.131E-2P0242550S ribosomal subunit protein L24Cytoplasmic
rpmC −  2.088E-3P0242950S ribosomal subunit protein L29Cytoplasmic
rpmF −  2.006E-3P0243550S ribosomal subunit protein L32Cytoplasmic
rpmG −  2.505E-2P0243650S ribosomal subunit protein L33Cytoplasmic
rpoS a 3.755E-2P13445RNA polymerase sigma factor 38Cytoplasmic
rpsH −  2.045E-2P0236130S ribosomal subunit protein S8Cytoplasmic
rpsI −  2.992E-2P0236330S ribosomal subunit protein S9Cytoplasmic
rpsM −  2.002E-2P0236930S ribosomal subunit protein S13Cytoplasmic
rpsN −  2.175E-2P0237030S ribosomal subunit protein S14Cytoplasmic
rpsQ −  2.504E-2P0237330S ribosomal subunit protein S17Cytoplasmic
rpsS b −2.862E-2P0237530S ribosomal subunit protein S19Cytoplasmic
stpA a 9.765E-2P30017DNA-binding proteinCytoplasmic
ycgE 14.004E-2P75989Similar to P33358, YehV putative transcriptional regulator (E. coli) PS00552: Bacterial regulatory protein MerR family signatureCytoplasmicf
ydeO b 3.872E-2P76135Similar to P05052, AppY regulatory protein (E. coli) PS00041: Bacterial regulatory proteins AraC family signatureCytoplasmic
yedW b 35.36c5E-2P76340Similar to O44006, regulatory protein CzcR (Ralstonia eutropha)Cytoplasmic probable
yfiA 3.724E-2P11285Similar to U32711, sigma54 modulation protein, putative (H. influenzae)Cytoplasmicf
yhiW b 32.41c2E-2P37638Similar to P05052, AppY regulatory protein (E. coli) PS00041: Bacterial regulatory proteins AraC family signatureCytoplasmicf
yhiX b 13.924E-4P37639Similar to P10805, regulatory protein EnvY (E. coli) PS00041: Bacterial regulatory proteins AraC family signatureCytoplasmicf
yjhI b 24.07c4E-2P39360Similar to P76268, KdgR transcriptional regulator (E. coli) PS01051: Bacterial regulatory proteins, IclR family signature Cytoplasmicf
D Other functions    
appA b 2.50d3E-2P07102Acid phosphatase, pH 2.5Periplasmic
cspB 4.105E-2P36995Similar to cold shock protein PS00352: Cold shock DNA-binding domain signatureCytoplasmic probable
cspC 4.731E-2P36996Similar to cold shock protein PS00352: Cold shock DNA-binding domain signatureCytoplasmic probable
cspE −  3.031E-3P36997Similar to cold shock protein PS00352: Cold shock DNA-binding domain signatureCytoplasmic probable
cspG b 9.105E-2Q47130Similar to cold shock protein PS00352: Cold shock DNA-binding domain signatureCytoplasmic probable
cspI b 8.718E-3P77605Similar to cold shock protein PS00352: Cold shock DNA-binding domain signatureCytoplasmic probable
dps 30.71c3E-2P27430DNA protection during starvation proteinCytoplasmic probable
ftn 3.548E-3P23887Iron storage proteinCytoplasmic
groL a 2.403E-2P06139Chaperone for assembly of enzyme complexesCytoplasmic
hdeA ab 48.404E-4P26604Putative chaperone-like protein involved in low pH resistancePeriplasmicf
hdeB ab 154.834E-4P26605Similar to hdeB (Shigella flexneri), involved in low pH resistancePeriplasmicf
proX ab 3.507E-3P14177Glycine-betaine binding proteinPeriplasmic
spy 8.835E-2P77754Spheroplast protein YPeriplasmic
yddT 2.924E-3P77790Similar to vancomycin resistance protein VanX (E. coli)Cytoplasmicf
E Unknown proteins
hdeD ab 4.804E-3P26603Unknown functionPeriplasmicf
yagK b 24.78c4E-2P77657Similar to P52125, YfjJ hypothetical protein (E. coli)Cytoplasmicf
yagY 8.345E-2P77188Similar to P77263, YagV hypothetical protein (E. coli)Periplasmicf
yaiS b 25.37c3E-2P71311Similar to P42981, YpjG hypothetical protein (B. subtilis)Cytoplasmicf
ybaJ 4.143E-2P37611Unknown functionCytoplasmicf
ybcL 3.415E-2P77368Similar to AE004189, hypothetical conserved protein (V. cholerae)Outer membranef
ybgE b 36.64c4E-2P37343PS00013: Membrane lipoprotein lipid attachment site signatureInner membranef
ycbW 3.002E-2P75862Unknown functionInner membranef
yccA 5.713E-2P06967Similar to O25578, hypothetical membrane protein (Helicobacter pylori) PS01243: uncharacterized protein family UPF005 signatureIntegral membrane probable
ycdT b 2.971E-3P75908Similar to P76236, YeaI hypothetical protein (E. coli)Integral membrane probable
ycgX b 3.02c3E-2P75988Similar to P76156, YdfO hypothetical protein (E. coli)Cytoplasmicf
yciF b 2.314E-4P21362Unknown functionCytoplasmicf
yciG b 6.263E-3P21361Similar to P56614, YmdF hypothetical protein (E. coli) PS00017: ATP-GTP binding site motif ACytoplasmicf
ydbA 17.628E-3P33666Unknown functionCytoplasmicf
ydbD b 2.935E-2P25907Unknown functionPeriplasmicf
ydcD b 5.031E-3P31991Unknown functionInner membranef
ydgK 4.025E-2P76180Unknown functionInner membrane probable
yeeN b 6.314E-4P76351Belongs to the UPF0082 familyCytoplasmicf
yegR b 13.335E-3P76406Unknown functionInner membranef
yegZ 4.082E-2AE000298eSimilar to P10312 D protein GPD bacteriophage P2Cytoplasmicf
yehA b 6.325E-2P33340Unknown functionCytoplasmic probable
ygaP b 6.225E-2P55734Similar to P73801 hypothetical protein (Synechocystis sp.)Inner membranef
ygbF 22.95c2E-2P45956Unknown functionCytoplasmicf
ygbT 2.293E-2Q46896Unknown functionCytoplasmicf
ygcF b 4.914E-2P55139Similar to D90908 hypothetical protein (Synechocystis sp.)Cytoplasmicf
ygcJ 4.045E-2Q46899Unknown functionCytoplasmicf
ygcK 4.352E-2P76632Unknown functionCytoplasmicf
ygcL b 8.172E-2Q46901Unknown functionInner or outer membranef
yhiE b 11.084E-4P29688Similar to P37195 YhiF hypothetical protein (E. coli)Cytoplasmicf
yhiF 12.463E-2P37195Similar to P29688 YhiE hypothetical protein (E. coli)Inner membranef
yhiM b 17.514E-4P37630Unknown functionInner membranef
ykgB 40.86c4E-2P75685Similar to gbAAB18029, hypothetical protein (H. influenzae)Integral inner membrane probable
ynaE 6.144E-4P76073Similar to P76154 YdfK hypothetical protein (E. coli)Cytoplasmicf
yncI b 15.309E-4AE000243eSimilar to P28912 H repeat-associated protein (E. coli)Inner membranef

Regulation of genes of unknown function by H-NS

Most genes encoding hypothetical proteins (Table 1) were tentatively characterized using the blastp program to compare the amino acid sequence of their product with the updated non-redundant databases (Altschul et al., 1990). The pattern database Prosite from the ExPASy server (Hofmann et al., 1999) was used to search for conserved patterns. Twenty-two gene products were similar to proteins of well-known function from a variety of organisms (Table 1). Sixteen gene products were not similar to any of the proteins present in the databases, and 18 were predicted to be similar to proteins of unknown function from different microorganisms.

Cell envelope components controlled by H-NS

The cellular localization of H-NS-regulated gene products was determined using the Niceprot database ( or predicted using the psort program ( Approximately half the proteins encoded by genes controlled by H-NS were located in or associated with membranes or were present in the periplasmic space (Table 1). For example, there were higher mRNA levels of 10 genes belonging to an operon involved in the biosynthesis of capsular polysaccharide colanic acids, such as gmd and wzc (Whitfield and Roberts, 1999), and 10 genes involved in lipopolysaccharide (LPS) biosynthesis, such as rfaI and wbbJ, in an H-NS-deficient background. The synthesis of gspG transcripts, which encodes a general secretion pathway protein, increased over 10-fold, whereas the greatest decrease in mRNA levels in the mutant strain corresponded to seven of the flagellar cascade genes. These encode flagellum structural components, such as flagellin (encoded by fliC). This observation is consistent with the proteome analysis (Fig. 1) and also supports a key role for H-NS in the positive regulation of motility and flagellum synthesis in vivo (Bertin et al., 1994; Soutourina et al., 1999). Similarly, several genes encoding proteins involved in fimbriae (fimB and fimI) and curli biosynthesis (csgA) were induced in the hns mutant (Table 1). Finally, five genes encoding hypothetical proteins similar to fimbriae and ompX, which encodes a porin-like protein involved in virulence in different microorganisms, were upregulated in the hns mutant (Table 1).

Effect of H-NS on DNA-binding protein-encoding genes

Several genes that are upregulated in the hns mutant code for proteins with nucleic acid-binding properties (Table 1). These include dps whose product appears to protect DNA from oxidative damage in the stationary phase (Bridges and Timms, 1998), stpA, which encodes an H-NS paralogue in E. coli (Zhang et al., 1996), and the DNA-binding protein, Lrp, structural gene (Newman et al., 1992). We observed more than a threefold increase in the stationary phase sigma factor mRNA, suggesting that, in addition to its role in stabilizing RpoS (Barth et al., 1995; Yamashino et al., 1995), H-NS could also directly or indirectly regulate the expression of the rpoS gene (Table 1). The mRNA levels of three regulators from two-component systems were also increased in the hns mutant: phoP from the PhoPQ system, which is involved in the response to magnesium limitation (Kato et al., 1999); evgA from the EvgAS system, which is similar to the Bordetella pertussis bvgA gene; and yedW, the regulator from the two-component system YedVW (Table 1). Finally, five gene products of unknown function (ycgE, ydeO, yhiW, yhiX and yjhI) were predicted to encode regulatory proteins belonging to the AraC, IclR and MerR families (Table 1).

Changes in the levels of mRNAs of starvation-induced proteins in the hns mutant strain

Some genes induced by oxygen starvation, such as cydAB, which encodes the cytochrome d ubiquinol oxidase complex (Cotter et al., 1990), and appB, which encodes a cytochrome bd II oxidase (Sturr et al., 1996), showed increased expression in the hns mutant background (Table 1). The transcript levels of at least three genes involved in the carbon starvation response were also increased in the hns mutant; these included gnd, whose product is involved in the hexose monophosphate shunt by converting 6-phospho-d-gluconate into d-ribulose 5-phosphate (Sprenger, 1995), and slp, which encodes a lipoprotein that stabilizes the outer membrane (Alexander and St John, 1994). This was consistent with our two-dimensional data (Fig. 1). Finally, the mRNA levels of 21 genes encoding ribosomal proteins, which are repressed by growth in minimal medium (Gausing, 1977; Tao et al., 1999), were decreased in the mutant strain (Table 1).

Increased resistance to high osmolarity and low pH in an H-NS-deficient strain

DNA array experiments suggested that many H-NS-regulated genes encode proteins that are involved in responses to environmental stresses. Several genes are involved in the response to high osmolarity (Csonka and Epstein, 1996); for example, ompC and ompF, which both encode porins, and proX, which encodes a glycine-betaine binding protein, consistent with our two-dimensional data (Fig. 1) and previous studies (Atlung and Ingmer, 1997). The osmC gene, which is induced by high osmolarity, showed an increased mRNA level in the mutant strain. Finally, the evgA gene product (see above), which can induce ompC expression in an envZ mutant strain (Utsumi et al., 1992), was upregulated in the mutant strain. Owing to the number of H-NS-regulated genes involved in osmoregulation (Table 1), we analysed the role of this regulatory protein in the response to osmotic stress. A fivefold increase in resistance to high ionic strength was measured in the hns mutant strain compared with the wild-type strain (Table 2A).

Table 2. High osmolarity and low pH resistance in wild-type and hns strainsa.
Growth conditionsPercentage survivalb
FB8 (wild type)BE1410 (hns)
  • a

    . Resistance measurement to both stresses was carried out as described in Experimental procedures. Viable bacterial cells were counted on plates.

  • b . Percentage survival is calculated as 100 × the number of cfu ml −1 remaining after acid or osmotic treatment divided by the initial cfu ml−1 at time zero. Values presented are representative of three independent experiments performed with the wild-type FB8 or the hns mutant BE1410 strains.

  • c

    . Cells were incubated in K5 medium supplemented with 0.4% glucose and 3 M NaCl.

  • d

    . Cells were incubated in M9 medium supplemented with 0.2% glucose and 0.012% amino acid.

A High osmolarity stressc
 3 M NaCl2.612.5
B Acid stressd
 pH 2.5 ≤  0.010.9
 pH 2.5 + 0.012% lysine0.12.9
 pH 2.5 + 0.012% arginine0.213.0
 pH 2.5 + 0.012% glutamate4.138.0

The cadBA operon and the adiA gene, which encode a lysine decarboxylase, an antiporter and an arginine decarboxylase, respectively, have been shown to be regulated by H-NS in rich medium under anaerobic conditions (Atlung and Ingmer, 1997). These genes are involved in the response to acid stress. Comparison of expression profiling (Table 1) showed that at least seven genes encoding proteins induced at low pH were increased in the hns mutant. For example, gadC (formerly called xasA) encodes a probable GABA/glutamate antiporter (De Biase et al., 1999), and gadA and gadB encode two glutamate decarboxylases (Table 1). The hdeA-encoded protein has been proposed to have a chaperone-like function under extremely acidic conditions (Gajiwala and Burley, 2000). The corresponding increase in its mRNA level (Table 1) is consistent with the increased accumulation of its gene product observed on two-dimensional gels (Fig. 1). This prompted us to study the effect of the hns mutation on resistance to acidic stress via lysine, arginine or glutamate. The wild-type strain had a low level of resistance to acid stress in the presence of lysine or arginine, compared with that in the absence of any amino acid (Table 2A). This can be explained by the minimal growth medium and aerobic growth conditions used in this study (see above). In contrast, the presence of glutamate resulted in a moderate increase in acid resistance. Similarly, in the H-NS-deficient strain BE1410, a moderate increase in resistance was measured in the presence of lysine or arginine. Furthermore, a large increase in resistance (nearly 40% viable cells) was measured in the hns mutant strain in the presence of glutamate compared with that in the wild-type strain.

To investigate the regulation underlying this acid resistance in the hns mutant strain, we overexpressed E. coli genes from a genomic library in the wild-type strain FB8 subjected to acid stress, as described in Experimental procedures, and screened for an acid resistance phenotype in the presence of glutamate. Analysis of clones resistant to low pH allowed us to select pDIA567, which increased the resistance of the wild-type strain in the presence of glutamate to similar levels to that of an hns mutant (44.5% viable cells). Sequence determination of the DNA insert showed that this plasmid carries a fragment encompassing both yhiX and gadA. Castanie-Cornet et al. (1999) recently suggested that the overexpression of gadA could result in the acid resistance measured in the wild-type strain. The yhiX gene, whose expression was increased over 13-fold in the H-NS-defective strain (Table 1), encodes a protein similar to the AraC family of regulatory proteins. To determine whether this YhiX putative regulator could confer acid resistance, its structural gene was amplified by polymerase chain reaction (PCR) and then cloned and overexpressed from pDIA570 in FB8 wild-type strain. Remarkably, 55.3% of these bacteria survived exposure to acid stress in the presence of glutamate. This strongly suggests that this regulatory protein is involved in the control of bacterial adaptation to low pH in the presence of glutamate. Thus, the corresponding gene was renamed gadX. To test the hypothesis that gadX could play a role in the control of H-NS-regulated genes induced by low pH, we analysed the expression levels of some pH-regulated genes in a wild-type strain overexpressing gadX. DNA fragments corresponding to the whole protein coding sequence (CDS) of various pH-regulated genes were amplified by PCR and spotted onto nylon membranes in duplicate, and these membranes were hybridized with cDNA probes generated from 1 µg of RNA. Quantitative measurements showed that the level of hdeD transcripts increased by up to threefold (Fig. 5). More importantly, the expression of gadA/gadB and gadC genes, which are specifically involved in the response to low pH, increased over eightfold. This further supports a role for GadX in the positive regulation of genes involved in low pH response, especially in the presence of glutamate.


Figure 5. Effect of gadX overexpression on pH-regulated genes. DNA fragments were amplified by PCR using primers specific for the 5′ and 3′ ends of the corresponding nucleotide sequence (Sigma–GenoSys Biotechnologies) and spotted in duplicate using a Q Pix arrayer (Genetix). Wild-type FB8 strains in the presence or not of pDIA570 overexpressing gadX were grown to A600 = 0.5 in minimal medium, pH 5.5, supplemented with 0.4% glucose and 0.012% glutamate. RNAs were extracted by a rapid procedure (I. Guillouard, unpublished), and cDNA probes were prepared with random hexamer primers (F. Hommais, unpublished). After hybridization, the results were quantified as described in Experimental procedures.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Proteome and transcriptome analyses demonstrated that the nucleoid-associated protein, H-NS, plays a major role in bacterial physiology by directly or indirectly controlling the expression of many genes and/or the synthesis of their product (these data are accessable at the web site HTM). In our growth conditions, we could visualize up to 1200 spots, of which the accumulation level of at least 5% was altered in the mutant strain (Fig. 1). Similarly, the expression level of nearly 5% of the E. coli 4290 CDSs was altered on DNA macroarrays (Fig. 3). The expression of over 80% of these genes was induced in the hns mutant strain, which further supports the role of H-NS as a repressor of gene expression. Of all the E. coli genome CDSs, many H-NS-regulated genes are involved in bacterial responses to multiple environmental conditions and/or in cell envelope composition. In addition, a number of genes that were up- or downregulated in the hns mutant have putative functions or have been described as hypothetical in databases ( Our results are the first demonstration that these genes can be expressed, at least under some conditions. Similarity searches, pattern identification and cellular localization suggested that many of these genes encode proteins that are localized at the cell surface (Table 1). It is therefore tempting to speculate that most of these gene products of unknown function (Table 1) also play a role in bacterial adaptation to different environmental conditions. In particular, YadN, YaiP, YagX and YcbQ, which are similar to fimbriae or adhesins, could be involved in bacterial colonization and/or virulence.

Among the targets of H-NS, many genes (for example, ompC, gnd, proX, fliC, yciF and yeeN) encoding proteins that were differentially accumulated in the hns strain (Fig. 1) showed, as expected, altered transcript levels on DNA arrays (Table 1). In contrast, some proteins, such as CbpA, were preferentially accumulated in the hns strain, but no significant difference in their transcript level was observed between wild-type and mutant strains. Moreover, several genes and proteins known to present a differential rate of expression or synthesis, respectively, in an hns mutant (Atlung and Ingmer, 1997; Laurent-Winter et al., 1997) did not show any difference in transcript level in wild-type or mutant strains on DNA arrays (Table 1). Several hypotheses may explain these discrepancies. First, some H-NS-regulated genes (for example, cadBA and adiA) are preferentially expressed in growth conditions that differ greatly from those used in the present study (aerobiosis and minimal medium supplemented with glucose). Similarly, these growth conditions could explain the lack of induction of the β-glucoside metabolism operon, bglGFB, in the hns mutant strain. Moreover, the transcript level of some genes, such as yeiP, was largely below the background level (data not shown), suggesting a low expression level and/or a rapid degradation of their corresponding mRNA. Finally, H-NS may affect the preferential accumulation of some proteins by a post-transcriptional mechanism, as observed recently for YceI: the lack of H-NS resulted in a minor effect on yceI expression (F. Hommais, unpublished), which is inconsistent with the great level of accumulation of YceI observed in the mutant strain on two-dimensional gel (Laurent-Winter et al., 1997).

H-NS post-transcriptionally regulates the synthesis and stability of the stationary sigma factor RpoS (Barth et al., 1995; Yamashino et al., 1995). An increase in the amount of rpoS mRNA was observed in the hns strain on DNA arrays (Table 1), suggesting that H-NS may also directly or indirectly regulate this gene at the transcriptional level and/or mRNA stability. We observed recently that the accumulation of a subset of H-NS-regulated gene products (GadA, Slp and HdeB) on two-dimensional gels is also controlled by RpoS and/or entry into stationary phase (C. Laurent-Winter, unpublished results). Strikingly, their genes showed the greatest increase in transcript level, between the hns mutant BE1410 and the wild-type FB8, ranging from a 23-fold to a 155-fold induction (Table 1). This effect on gene expression could be, at least partly, mediated by an increase in rpoS mRNA (Table 1) and/or RpoS concentration in the hns mutant and further supports the existence of numerous interactions between both regulatory systems (Hengge-Aronis, 1999). Furthermore, these observations suggest that several other genes that are regulated by H-NS (Table 1) are also controlled by RpoS (Fig. 6).


Figure 6. H-NS regulon genes and their regulation by environmental parameters. Genes of known function that were found to be either directly or indirectly controlled by H-NS on two-dimensional gels (Fig. 1) and/or macroarrays (Table 1) were clustered according to the conditions known to modulate their expression. Rectangles represent genes belonging to oxygen, temperature, osmolarity and pH regulon respectively. Genes written in shaded are those identified previously as RpoS-regulated genes.

Download figure to PowerPoint

Several H-NS-regulated genes encode DNA-binding proteins (Table 1). An increase in the transcript level of several regulators from two-component systems (PhoP, EvgA and YedW) was observed in the hns strain. In such systems, the control of gene expression depends on the phosphorylation state of the regulatory protein (Stock et al., 1989). Our results suggest that H-NS could initiate a new kind of regulation by specifically modulating the expression level of those regulators. Moreover, the effect of H-NS on genes of the colanic acid operon could be mediated by the increased expression of their activator-encoding gene, rcsA (Ebel and Trempy, 1999), in the hns mutant (Table 1). This suggests that H-NS plays an indirect role in the regulation of many genes identified in this study and further supports a high position of H-NS in the regulatory hierarchy of bacterial physiology. In particular, we demonstrated here that the overexpression of gadX (formerly called yhiX), which encodes a putative AraC family regulator (Table 1), in a wild-type strain increased the expression of H-NS-regulated genes involved in the response to low pH. Moreover, this strain showed a similar level of survival to the hns mutant at low pH in the presence of glutamate. These results strongly support an important role for GadX in the control of the genes involved in acid stress resistance with glutamate.

A higher level of mRNA induction and a greater amount of protein accumulation was observed in the H-NS-deficient strain than in the wild type (Table 1 and Fig. 1) for GadA and GadB, two glutamate decarboxylases involved in the response to low pH with glutamate (Lin et al., 1995). Moreover, the transcript levels of several genes and operons, such as ompC, ompF and proU, which are involved in the bacterial response to high osmolarity (Kempf and Bremer, 1998), and the accumulation of their products were also affected by the hns mutation (Table 1 and Fig. 1). At high osmolarity, an increase in the intracellular glutamate concentration and a K+ accumulation is required to produce potassium glutamate, the primary cell protector against high osmolarity (Kempf and Bremer, 1998). Potassium glutamate may be an internal signal for the onset of the second phase of osmoadaptation, the uptake of compatible solutes, such as glycine betaine, which requires the ompC and proU gene products. The response to both low pH and high osmolarity results in the consumption of glutamate, which could explain the increase in gltD mRNA levels (Table 1). These observations suggest that the hns mutation severely perturbs glutamate metabolism.

The lack of H-NS has usually been considered as a loss of function, such as loss of motility (Soutourina et al., 1999) or reduction in growth rate (Barth et al., 1995). Paradoxically, survival assays showed that an hns mutation can constitute an advantage for bacterial cells at high osmolarity or low pH. In addition to the genes directly involved in low pH and high osmolarity resistance, numerous H-NS-regulated genes are controlled by both stresses (Fig. 6). In particular, some of them, such as those encoding porins, flagellum components and colanic acid, are also controlled by oxygen starvation or high temperature (Shi et al., 1993; Whitfield and Roberts, 1999), which reflects the growth conditions frequently encountered by enterobacteria inside their host (Mahan et al., 1996). All these responses imply cation exchange which is required for the establishment of homeostasis (Booth, 1999). These results and the observation that the modification of membrane protein composition (Table 1) is usually associated with the adaptation of bacteria to stressful environmental conditions (Kadner, 1996) suggest a strong relationship between the control of gene expression by H-NS and the maintenance of intracellular homeostasis. Thus, an hns strain could constitute a unique model for identifying the genes that enable pathogens to survive better inside their hosts.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains and growth conditions

Escherichia coli FB8 (Bruni et al., 1977) and MG1655 (Bachmann, 1983) strains and their hns mutant derivatives, BE1410 (Laurent-Winter et al., 1997) and BE1816 (Soutourina et al., 1999), respectively, containing a Tn5seq1 transposon in the 20th codon of the hns coding sequence (E. Krin, unpublished), were used. Bacteria were grown at 37°C in M63 minimal medium (Miller, 1992) supplemented with 0.4% glucose.

Construction of an E. coli genomic DNA library

pDIA561 was constructed by replacing the fragment between the XbaI site and the SacI site in pBCKS+ (Stratagene) with the pcDNA 2.1 polylinker that contained two BstXI sites (Invitrogen). Genomic DNA was isolated from E. coli mutated in the hns and stpA genes and nebulized for 30 s at 1 bar. Fragments ranging from 1.5 to 4.5 kb were ethanol precipitated, filled in by T4 polymerase and linked to BstXI adapters (Invitrogen). Restriction fragments and pDIA561 digested by BstXI were ligated by T4 DNA ligase at 16°C for 15 h. The ligation mixture was introduced into XL1-Blue strain (Stratagene) by electrotransformation. About 60 000 independent clones were selected on LB plates containing 20 µg ml−1 chloramphenicol and pooled. Large-scale plasmid DNA isolation was carried out using the JETstar kit (Genomed). This library was used to transform FB8.

Construction of plasmids

To overexpress yhiX, its coding sequence was amplified by PCR from genomic DNA using primers 5′-ATGGAATTCCAATCACTACATGGGAATTG-3′ and 5′-CGGGATCCTATAATCTTATTCCTTC-3′. These primers introduced an EcoRI cloning site into its 5′ end and a BamHI site in its 3′ end. The PCR product was inserted into the EcoRI and BamHI sites of the pTRC99A (Pharmacia), thus resulting in pDIA570.

High osmotic resistance

Bacteria were grown overnight in K5 medium (Epstein and Kim, 1971) supplemented with 0.4% glucose, 0.1% casamino acids. Cells were diluted 1:1000 in fresh K5 medium supplemented with 0.4% glucose and 3 M NaCl for 1 h at 37°C and then plated on LB. Viable cells were counted after 16 h at 37°C.

Low pH resistance

Bacteria were grown overnight in M9 medium (Miller, 1992), pH 5.5, supplemented with 0.4% glucose, 0.1% casamino acids. Cells were diluted 1:1000 in M9 medium, pH 2.5, supplemented with 0.2% glucose, 0.012% amino acids (glutamate, arginine or lysine) (Lin et al., 1995) for 2 h at 37°C and then plated on LB. Viable cells were counted after 16 h at 37°C.

Two-dimensional gel electrophoresis

Total protein extracts and two-dimensional gel electrophoresis were carried out as described previously (Laurent-Winter et al., 1997) with some modifications. Total protein was extracted from E. coli cells (A600 = 0.6), and 50 µg was loaded onto isoelectrofocusing (IEF) gels containing ampholines with pH ranging from 3.5 to 10 (Millipore). This optical density of the culture corresponds to a late-exponential phase in minimal medium for both strains, despite a slightly slower growth rate of the hns mutant. The second dimension was carried out in 12.5% acrylamide slab gels. Resolved proteins were detected by silver nitrate staining (Morrissey, 1981). The gels were analysed, and the spots were quantified on the melanie II software (Bio-Rad). Values were expressed as percentage volume, which is the percentage ratio between the volume of a single spot and the total volume of all the spots on a gel. Volume is a function of optical density and spot area, which takes into account the variability resulting from silver staining. Spots of interest were excised, digested in trypsin (Shevchenko et al., 1996), desalted using Zip Tips C18 (Millipore) and eluted with 50% acetonitrile. MALDI-TOF spectra of peptides were obtained with an STR-DE mass spectrometer (Perspective Biosystems) using 2,5-dihydroxybensoic acid as a matrix. The sample was loaded on the target by the dried droplet method. Spectra obtained for the whole protein were calibrated externally using the [M+ H+] ion from Des Arg Braddykin peptide (904.47) and ACTH (2465.13). Peptides from the autodigestion of tryptin were used for the internal calibration. Samples were analysed in the reflectron mode, with an accelerating voltage of 25 000 V, an extraction delay of 200 ns and an average of 250 scans. Proteins were identified by comparing the spectra with non-redundant databases (SWISSPROT, Trembl and GenBank) using the ProFound, PeptIdent and MS-Fit web servers (, http://www.,

Handling of RNA

Bacterial cells were grown to A600 = 0.6 (see above), and 5 ml of bacterial culture was centrifuged at 4°C for 10 min at 6000 g and stored at −80°C to prevent RNA degradation. Cells were lysed, and RNA was extracted according to the manufacturer's recommendations (Sigma-GenoSys Biotechnologies) using phenol, pH 4.5, at 65°C. RNA was ethanol precipitated and redissolved in 420 µl of 10 mM Tris, 1 mM EDTA, pH 7.6 (TE). RNA was incubated in 3 mM MgCl2 with 20 µl of DNase I RNase free (Roche) for 30 min at 37°C to remove genomic DNA. DNase I was removed by a phenol–chloroform extraction. Purified RNA was ethanol precipitated, redissolved in 100 µl of TE and quantified by measuring A260 and A280. RNA purity and integrity were controlled by separating a sample on an agarose gel and ensuring that mRNA, tRNA and rRNAs could be seen.

cDNA probe synthesis

Hybridization probes were generated from 1 µg of RNA following standard cDNA synthesis using [α-33P]-dCTP (2000–3000 Ci mmol−1 from New England Nuclear) as recommended by Sigma–GenoSys Biotechnologies. To prevent any subsequent amplification, the AMV reverse transcriptase (25 U µl−1; Roche) with RNase H activity was used for cDNA synthesis. Alternatively, hybridization probes were generated from 10 µg of RNA in a final volume of 120 µl by incubating 4 µl of E. coli labelling primers (Sigma–Genosys), 0.25 mM each dATP, dGTP and dTTP with 60 µCi of [α-33P]-dCTP and 6 µl of AMV reverse transcriptase for 3 h at 42°C. Unincorporated nucleotides were removed from labelled cDNA by gel filtration through a G-25 Sephadex column (Roche).


Macroarrays were Panorama E. coli gene arrays obtained from Sigma–GenoSys Biotechnologies. Prehybridization and hybridization were carried out according to the manufacturer's recommendations with some modifications. Hybridization and washing steps were carried out as described by the manufacturer using SSPE solution (0.18 M NaCl, 10 mM NaH2PO4 and 1 mM EDTA, pH 7.7). The arrays were prehybridized in 15 ml of hybridization solution (5× SSPE, 2% SDS, 1× Denhardt's reagent, 100 µg of sheared salmon sperm DNA ml−1) in roller bottles and hybridized for 16 h in a hybridization oven with 10 ml of hybridization solution containing the entire cDNA probe. Blots were washed in 0.5× SSPE, 0.2% SDS, slightly dried on a Whatman paper, wrapped and sealed in a damp Saran 25 µm film (Dow). Blots were exposed to a PhosphorImager screen (Molecular Dynamics) for 22 h. The arrays were stripped with boiled buffer (10 mM Tris, pH 7.6, 1 mM EDTA, 1% SDS) four times for 15 min and stored at room temperature in a sealed plastic bag with 20 ml of 0.5% SSPE, 0.2% SDS. A single array could be used up to 10 times.

Data analysis

Exposed PhosphorImager screens were scanned on a 445SI PhosphorImager (Molecular Dynamics) with a pixel size of 177 µm. The intensity of each dot on the resulting TIFF image files was measured with the xdotsreader software (Cose) and analysed using an excel spreadsheet. The accuracy and reproducibility of the results were analysed by comparing the distribution of intensity in each spot between duplicate dots and between independent hybridizations. In all experiments performed with 1 µg and 10 µg of mRNA, the results (Fig. 2) showed a high degree of correlation (> 0.80), in agreement with recent data (Richmond et al., 1999; Hoheisel and Vingron, 2000). The background noise was calculated from the expression of dots without DNA and subtracted from the intensity of each dot. Dot intensity was normalized according to the mean value of the total intensities of all spots on each DNA array, which allowed direct comparison of the two strains. The hns mutant expression profiles were compared with those of the wild-type strain by calculating the consistency of differential expression across replicate hybridizations using the Wilcoxon signed rank test, a non-parametric statistical method contained in the statview 5.0.1 package. This tested the hypothesis that one of the paired variables is greater or less than the other variable regardless of the magnitude of the difference and is appropriate for the analysis of small samples.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We wish to thank I. Martin, B. Dujon, J. Rossier and O. Barzu for their kind interest and B. Llorente, C. Fabry and A. Maitournam for helpful technical assistance in membrane hybridization and analysis. Financial support came from the Pasteur Institute and the Centre Nationale de la Recherche Scientifique (URA 2171).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Alexander, D.M. & St John, A.C. (1994) Characterization of the carbon starvation-inducible and stationary phase-inducible gene slp encoding an outer membrane lipoprotein in Escherichia coli. Mol Microbiol 11: 10591071.
  • Altschul, W., Gish, W., Miller, W., Myers, E.W., Lipman, D.J. (1990) Basic local alignment search tool. J Mol Biol 215: 403410.
  • Arfin, S.M., Long, A.D., Ito, E.T., Tolleri, L., Riehle, M.M., Paegle, E.S., Hatfield, G.W. (2000) Global gene expression profiling in Escherichia coli K12. J Biol Chem 275: 2967229684.DOI: 10.1074/jbc.m002247200
  • Atlung, T. & Ingmer, H. (1997) H-NS: a modulator of environmentally regulated gene expression. Mol Microbiol 24: 717.
  • Bachmann, B.J. (1983) Linkage map of Escherichia coli K-12. Microbiol Rev 47: 180230.
  • Barth, M., Marschall, C., Muffler, A., Fischer, D., Hengge-Aronis, R. (1995) Role for the histone-like protein H-NS in growth phase-dependent and osmotic regulation of σ-dependent genes in Escherichia coli. J Bacteriol 177: 34553464.
  • Bertin, P., Terao, E., Lee, E.H., Lejeune, P., Colson, C., Danchin, A., Collatz, E. (1994) The H-NS protein is involved in the biogenesis of flagella in Escherichia coli. J Bacteriol 176: 55375540.
  • Bertin, P., Benhabiles, N., Krin, E., Laurent-Winter, C., Tendeng, C., Turlin, E., et al. (1999) The structural and functional organization of H-NS-like proteins is evolutionarily conserved in Gram-negative bacteria. Mol Microbiol 31: 319330.
  • Blattner, F.R., Plunkett, G., Bloch, C.A., Perna, N.T., Burland, V., Riley, M., et al. (1997) The complete genome sequence of Escherichia coli K-12. Science 277: 14531474.DOI: 10.1126/science.277.5331.1453
  • Booth, I.R. (1999) The regulation of intracellular pH in bacteria. In Bacterial Responses to pH. Wiley, J. (ed.). Chichester: Novartis Foundation Symposium, pp. 1937.
  • Bridges, B.A. & Timms, A. (1998) Effect of endogenous carotenoids and defective RpoS sigma factor on spontaneous mutation under starvation conditions in Escherichia coli: evidence for the possible involvement of singlet oxygen. Mutat Res 403: 2128.DOI: 10.1016/s0027-5107(98)00013-x
  • Bruni, C.B., Colantuoni, V., Sbordone, L., Cortese, R., Blasi, F. (1977) Biochemical and regulatory properties of Escherichia coli K-12 his mutants. J Bacteriol 130: 410.
  • Caramel, A. & Schnetz, K. (1998) Lac and lambda repressors relieve silencing of the Escherichia coli bgl promoter. Activation by alteration of a repressing nucleoprotein complex. J Mol Biol 284: 875883.DOI: 10.1006/jmbi.1998.2191
  • Castanie-Cornet, M., Penfound, T., Smith, D., Elliott, J., Foster, J. (1999) Control of acid resistance in Escherichia coli. J Bacteriol 181: 35253535.
  • Ceschini, S., Lupidi, G., Coletta, M., Pon, C.L., Fioretti, E., Angeletti, M. (2000) Multimeric self-assembly equilibria involving the histone-like protein H-NS – a thermodynamic study. J Biol Chem 275: 729734.
  • Cotter, P.A., Chepuri, V., Gennis, R.B., Gunsalus, R.P. (1990) Cytochrome o (cyoABCDE) and d (cydAB) oxidase gene expression in Escherichia coli is regulated by oxygen, pH, and the fnr gene product. J Bacteriol 172: 63336338.
  • Csonka, L.N. & Epstein, W. (1996) Osmoregulation. In Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. Neidhardt, F.C. (ed.). Washington, DC: American Society for Microbiology Press, pp. 12101224.
  • Dame, R.T., Wyman, C., Goosen, N. (2000) H-NS mediated compaction of DNA visualised by atomic force microscopy. Nucleic Acids Res 28: 35043510.DOI: 10.1093/nar/28.18.3504
  • De Biase, D., Tramonti, A., Bossa, F., Visca, P. (1999) The response to stationary-phase stress conditions in Escherichia coli: role and regulation of the glutamic acid decarboxylase system. Mol Microbiol 32: 11981211.
  • Dorman, C.J., Hinton, J.C.D., Free, A. (1999) Domain organization and oligomerization among H-NS-like nucleoid-associated proteins in bacteria. Trends Microbiol 7: 124128.DOI: 10.1016/s0966-842x(99)01455-9
  • Ebel, W. & Trempy, J.E. (1999) Escherichia coli RcsA, a positive activator of colanic acid capsular polysaccharide synthesis, functions to activate its own expression. J Bacteriol 181: 577584.
  • Epstein, W. & Kim, B.S. (1971) Potassium loci in Escherichia coli K-12. J Bacteriol 108: 639644.
  • Gajiwala, K.S. & Burley, S.K. (2000) HDEA, a periplasmic protein that supports acid resistance in pathogenic enteric bacteria. J Mol Biol 295: 605612.DOI: 10.1006/jmbi.1999.3347
  • Gausing, K. (1977) Regulation of ribosome production in Escherichia coli: synthesis and stability of ribosomal RNA and of ribosomal protein messenger RNA at different growth rates. J Mol Biol 115: 335354.
  • Hayat, M.A. & Mancarella, D.A. (1995) Nucleoid proteins. Micron 26: 461480.DOI: 10.1016/0968-4328(95)00022-4
  • Hengge-Aronis, R. (1999) Interplay of global regulators and cell physiology in the general stress response of Escherichia coli. Curr Opin Microbiol 2: 148152.DOI: 10.1016/s1369-5274(99)80026-5
  • Hofmann, K., Bucher, P., Falquet, L., Bairoch, A. (1999) The PROSITE database, its status in 1999. Nucleic Acids Res 27: 215219.DOI: 10.1093/nar/27.1.215
  • Hoheisel, J.D. & Vingron, M. (2000) Transcriptional profiling: is it worth the money? Res Microbiol 151: 113119.DOI: 10.1016/s0923-2508(00)00120-0
  • Jacquet, M., Cukier-Kahn, R., Pla, J., Gros, F. (1971) A thermostable protein factor acting on in vitro DNA transcription. Biochem Biophys Res Commun 45: 15971607.
  • Jordi, B.J. & Higgins, C.F. (2000) The downstream regulatory element of the proU operon of Salmonella typhimurium inhibits open complex formation by RNA polymerase at a distance. J Biol Chem 275: 1212312128.
  • Jordi, B.J., Fielder, A., Burns, C.M., Hinton, J.C.D., Dover, N., Ussery, D.W., Higgins, C.F. (1997) DNA binding is not sufficient for H-NS-mediated repression of proU expression. J Biol Chem 272: 1208312090.
  • Kadner, R.J. (1996) Cytoplasmic membrane. In Escherichia coli and Salmonella: Cellular and Molecular Biology. Neidhardt, F.C. (ed.). Washington, DC: American Society for Microbiology Press, pp. 5887.
  • Kato, A., Tanabe, H., Utsumi, R. (1999) Molecular characterization of the PhoP-PhoQ two-component system in Escherichia coli K-12: identification of extracellular Mg2+-responsive promoters. J Bacteriol 181: 55165520.
  • Kempf, B. & Bremer, E. (1998) Uptake and synthesis of compatible solutes as microbial stress responses to high-osmolarity environments. Arch Microbiol 170: 319330.DOI: 10.1007/s002030050649
  • Kunst, F., Ogasawara, N., Moszer, I., Albertini, A.M., Alloni, G., Azevedo, V., et al. (1997) The complete genome sequence of the gram-positive bacterium Bacillus subtilis. Nature 390: 249256.DOI: 10.1038/36786
  • Kyrpides, N.C. & Ouzounis, C.A. (1999) Whole-genome sequence annotation: going wrong with confidence. Mol Microbiol 32: 886887.
  • La Teana, A., Brandi, A., Falconi, M., Spurio, R., Pon, C.L., Gualerzi, C.O. (1991) Identification of a cold shock transcriptional enhancer of the Escherichia coli gene encoding nucleoid protein H-NS. Proc Natl Acad Sci USA 88: 1090710911.
  • Laurent-Winter, C., Ngo, S., Danchin, A., Bertin, P. (1997) Role of Escherichia coli histone-like nucleoid-structuring protein in bacterial metabolism and stress response. Eur J Biochem 244: 767773.
  • Lin, J., Lee, I.S., Frey, J., Slonczewki, J.L., Foster, J.W. (1995) Comparative analysis of extreme acid survival in Salmonella typhimurium, Shigella flexneri, and Escherichia coli. J Bacteriol 177: 40974104.
  • Mahan, M.J., Slauch, J.M., Mekalanos, J.J. (1996) Environmental regulation of virulence gene expression in Escherichia, Salmonella and Shigella sp. In Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. Neidhardt, F.C. (ed.). Washington, DC: American Society for Microbiology Press, pp. 28032813.
  • Miller, J.H. (1992) A Short Course in Bacterial Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • Morrissey, J.H. (1981) Silver stain for proteins in polyacrylamide gels: a modified procedure with enhanced uniform sensitivity. Anal Biochem 117: 307310.
  • Moszer, I. (1998) The complete genome of Bacillus subtilis: from sequence annotation to data management and analysis. FEBS Lett 430: 2836.DOI: 10.1016/s0014-5793(98)00620-6
  • Newman, E.B., Lin, R.T., D'Ari, R. (1992) The leucine-Lrp regulon in E. coli: a global response in search of a raison d'être. Cell 68: 617619.
  • Richmond, C.S., Glasner, J.D., Mau, R., Jin, H., Blattner, F.R. (1999) Genome-wide expression profiling in Escherichia coli K-12. Nucleic Acids Res 27: 38213835.DOI: 10.1093/nar/27.19.3821
  • Schröder, O. & Wagner, R. (2000) The bacterial DNA-binding protein H-NS represses ribosomal RNA transcription by trapping RNA polymerase in the initiation complex. J Mol Biol 298: 737748.
  • Shevchenko, A., Wilm, M., Vorm, O., Mann, M. (1996) Mass spectrometric sequencing of proteins from silver-stained polyacrylamide gels. Anal Chem 68: 850858.DOI: 10.1021/ac950914h
  • Shi, X., Waasdorp, C., Bennett, G. (1993) Modulation of acid-induced amino acid decarboxylase gene expression by hns in Escherichia coli. J Bacteriol 175: 11821186.
  • Shindo, H., Iwaki, T., Ieda, R., Kurumizaka, H., Ueguchi, C., Mizuno, T., et al. (1995) Solution structure of the DNA binding domain of a nucleoid-associated protein, H-NS, from Escherichia coli. FEBS Lett 360: 125131.DOI: 10.1016/0014-5793(95)00079-o
  • Smyth, C.P., Lundback, T., Renzoni, D., Siligardi, G., Beavil, R., Layton, M., et al. (2000) Oligomerization of the chromatin-structuring protein H-NS. Mol Microbiol 36: 962972.DOI: 10.1046/j.1365-2958.2000.01917.x
  • Soutourina, O., Kolb, A., Krin, E., Laurent-Winter, C., Rimsky, S., Danchin, A., Bertin, P. (1999) Multiple control of flagella biosynthesis in Escherichia coli: Role of H-NS protein and the cyclic AMP/catabolite activator protein complex in transcription of the flhDC master operon. J Bacteriol 181: 75007508.
  • Sprenger, G.A. (1995) Genetics of pentose-phosphate pathway enzymes of Escherichia coli K-12. Arch Microbiol 164: 324330.DOI: 10.1007/s002030050270
  • Spurio, R., Falconi, M., Brandi, A., Pon, C.L., Gualerzi, C.O. (1997) The oligomeric structure of nucleoid protein H-NS is necessary for recognition of intrinsically curved DNA and for DNA bending. EMBO J 16: 17951805.DOI: 10.1093/emboj/16.7.1795
  • Stock, A.M., Mottonen, J.M., Stock, J.B., Schutt, C.E. (1989) Three-dimensional structure of CheY, the response regulator of bacterial chemotaxis. Nature 337: 745749.
  • Sturr, M.G., Krulwich, T.A., Hicks, D.B. (1996) Purification of a cytochrome bd terminal oxidase encoded by the Escherichia coli app locus from a delta cyo delta cyd strain complemented by genes from Bacillus firmus OF4. J Bacteriol 178: 17421749.
  • Talukder, A.A., Akira, I., Akiko, N., Susumu, U., Akira, I. (1999) Growth phase-dependent variation in protein composition of the Escherichia coli nucleoid. J Bacteriol 181: 63616370.
  • Tao, H., Bausch, C., Richmond, C., Blattner, F.R., Conway, T. (1999) Functional genomics: expression analysis of Escherichia coli growing on minimal and rich media. J Bacteriol 181: 64256440.
  • Tendeng, C., Badaut, C., Krin, E., Gounon, P., Ngo, S., Danchin, A., et al. (2000) Isolation and characterization of vicH, encoding a new pleiotropic regulator in Vibrio cholerae. J Bacteriol 182: 20262032.
  • Tupper, A.E., Owen-Hughes, T.A., Ussery, D.W., Santos, D.S., Ferguson, D.J.P., Sidebotham, J.M., et al. (1994) The chromatin-associated protein H-NS alters DNA topology in vitro. EMBO J 13: 258268.
  • Ueguchi, C., Suzuki, T., Yoshida, T., Tanaka, K., Mizuno, T. (1996) Systematic mutational analysis revealing the functional domain organization of Escherichia coli nucleoid protein H-NS. J Mol Biol 263: 149162.DOI: 10.1006/jmbi.1996.0566
  • Utsumi, R., Katayama, S., Ikeda, M., Igaki, S., Nakagawa, H., Miwa, A., et al. (1992) Cloning and sequence analysis of the evgAS genes involved in signal transduction of Escherichia coli K-12. Nucleic Acids Symp Series 27: 149150.
  • Whitfield, C. & Roberts, I.S. (1999) Structure, assembly and regulation of expression of capsules in Escherichia coli. Mol Microbiol 31: 13071319.
  • Williams, R.M. & Rimsky, S. (1997) Molecular aspects of the E. coli nucleoid protein, H-NS: a central controller of gene regulatory networks. FEMS Microbiol Lett 156: 175185.DOI: 10.1016/s0378-1097(97)00397-2
  • Williams, R.M., Rimsky, S., Buc, H. (1996) Probing the structure, function and interactions of the Escherichia coli H-NS and StpA proteins using dominant negative derivatives. J Bacteriol 178: 43354343.
  • Yamashino, T., Uegushi, C., Mizuno, T. (1995) Quantitative control of the stationary phase-specific sigma, σS, in Escherichia coli: involvement of the nucleoid protein H-NS. EMBO J 14: 594602.
  • Zhang, A., Rimsky, S., Reaban, M.E., Buc, H., Belfort, M. (1996) Escherichia coli protein analogs StpA and H-NS: regulatory loops, similar and disparate effects on nucleic acid dynamics. EMBO J 15: 13401349.