Nucleoid-associated protein H-NS has emerged as one of the most intriguing and versatile global regulators of enterobacterial gene expression acting primarily yet not exclusively at the transcriptional level where it generally acts as a repressor. H-NS is also believed to contribute to the architectural organization of the nucleoid by causing DNA compaction, although the evidence for such a role is not overwhelming. H-NS binds preferentially to DNA elements displaying intrinsic curvatures and can induce DNA bending. These functions are determined by its quaternary tetrameric structure. In turn, the existence of an intrinsic DNA curvature separating two or more H-NS binding sites seems to be characteristic of the H-NS-sensitive promoters and a prerequisite for the transcriptional repressor activity of this protein. In some cases, like that of the virF promoter, the temperature-sensitivity of the DNA curvature represents a key element in the thermo-regulation of pathogenicity gene expression.

Introduction

Histone-like nucleoid-structuring protein (H-NS), was discovered in Escherichia coli approximately two decades ago;^1,2 the claim of an earlier discovery of this protein is in fact devoid of any scientific foundation. Indeed, the H1 protein described in 1971-72^3,4 was a >10 kDa thermostable (at 100°C) transcriptional enhancer composed of 67-70 amino acids eluting as a monomer from size-exclusion chromatography (SEC).³ Instead, it is well established that H-NS, which is almost invariably a strong transcriptional repressor in vivo and under all in vitro conditions (for reviews see refs. 5-7), is inactivated by a brief exposure to 55°C⁸ and its monomeric mass is 15.6 kDa, being constituted by 136 amino acids.⁹ Furthermore, H-NS elutes from SEC as a mixture of dimers and tetramers even at μM concentrations^10,11 and its amino acid composition⁹ is distinctly different from that reported for H1.⁴

After elucidation of the H-NS primary sequence,⁹ the monocistronic gene (hns) encoding this protein was isolated and characterized.^12,13 However, due to a mistake in the orientation of the Kohara-Isono blot used in the physical mapping of the gene,¹² its chromosomal position was erroneously reported to be 6.1 min instead of 27 min where it is actually located.¹⁴ Further biochemical characterization demonstrated that H-NS binds double stranded (ds)DNA better than single-stranded DNA and RNA¹⁵ and displays a marked preference for bent DNA.^16-18 An architectural role in the organization of “bacterial chromatin”, was inferred from its abundance (20,000 copies/ cell),^1,19 from its biochemical properties and from its nucleoid localization,²⁰ but evidence also began to accumulate that H-NS controls, mainly at the transcriptional level, the expression of several genes since different hns mutations were shown to cause extremely pleiotropic phenotypes (for a review see ref. 14). More recently H-NS has emerged not only as one of the most important and intriguing global transcriptional regulators, but also as a participant in other processes such as translation and the control of RNA and protein stability.

Genes encoding proteins homologous to H-NS were in the meantime discovered in other enterobacteriaceae^13,21 but no H-NS-like protein was detected in Gram positive bacteria (Bacillus stearothermophilus and B. subtilis). Thus, unlike with the ubiquitous nucleoid-associated protein HU, H-NS was considered to be restricted to enteric bacteria. However, recent genomic data has shown that several Gram negatives from ecologically different habitats, contain H-NS-related proteins sharing a similar two-module structural organization.²² However, it is not clear whether the roles of these proteins and of enterobacterial H-NS are the same. Finally, a gene (stpA) orthologous to hns was also identified in E. coli²³ and its product (StpA) was shown to share at least some properties with H-NS.

In the following sections we describe in more detail some structural and functional facets of H-NS with emphasis on its regulation of transcription in combination with bent DNA.

Relationship between Function and Three-Dimensional and Quaternary Structure of H-NS

As mentioned above, the preference for curved DNA is a major characteristic of H-NS. This binding property is clearly demonstrated by biochemical evidence and by electron micrographs showing that H-NS clusters precisely in the position of an intrinsic bend within a linear segment of DNA (fig. 1). The functional significance of this preference for curved DNA^16-18 is suggested by the frequent occurrence of an intrinsic DNA bend flanked by two or more extended H-NS binding sites in the promoter regions susceptible to H-NS repression. (e.g., refs. 8, 24, 25) Among different types of curved DNA, H-NS seems to prefer AT-rich planar curvatures.^26,27 These findings, along with the fact that H-NS binds also to non-AT curvatures lacking a defined sequence-specificity,²⁸ suggest that this protein prefers a specific geometric structure of the duplex, probably corresponding to a narrow minor groove. However, little is known of the molecular nature of the H-NS-DNA interaction from both the protein and the nucleic acid side and nothing is known of the structural consequences locally induced by such an interaction. Thus, although H-NS binding results in the stabilization or induction of DNA bends from which transcriptional repression normally ensues, the microscopic consequences of the binding from which these macroscopic effects ultimately stem remain mysterious.

Figure 1

Specific recognition of curved DNA by H-NS. (A) Curvature prediction and (B) scanning force microscopy of a 1.1 kb linear fragment of naked DNA containing an in phase triple repeat of 5A and 6A tracts inserted at one third of its length; (C) specific (more...)

Structural and biochemical data show that H-NS consists of two domains whose three-dimensional (3D) structures have been elucidated^29-35 (fig. 2). However, considerable controversy surrounds the actual structure of the N-terminal domain for which three different analyses have yielded three completely different structures,33-35 two of which are shown in Figure 2.

Figure 2

3D structures determined in solution by NMR spectroscopy of (A) the C-domain (1HNR) and (B) N-domain (1NI8) of E. coli H-NS. The alternative 3D structure of the N-domain of Salmonella typhimurium H-NS (1LR1) is shown in (C). The structures were obtained (more...)

The C-terminal domain, which is separated from the N-domain by a linker, is considered to be the DNA binding domain. The interaction of this domain plus the linker (residues 60-137) with a 14 bp synthetic DNA (CAAAATATATTTTG) was investigated by nuclear magnetic resonance (NMR) spectroscopy; loops L1 and L2, which are spatially close to each other and display marked positive surface charges, were identified as the structural elements involved in this interaction. The N-terminal domain, which is = 45 Å away³⁶ from the C-domain, consists primarily of α-helices and its function is to promote protein dimerization. However, in addition to the controversy concerning the 3D structure of this domain, there is disagreement concerning the quaternary structure of H-NS which consists of a heterodisperse aggregate of trimers^32,33 or of monomers, dimers and tetramers in dynamic equilibrium.^9-11,37

In spite of the commonplace attribution of the DNA-binding and the protein dimerization functions to the C- and the N-domain, respectively, not all genetic and biochemical data can be easily fit within this schematic model. Therefore, the relationship between structure and function of H-NS may be much more complicated than it superficially appears.

Indeed, probably because of the contribution of loop L1 to the H-NS-DNA interaction, the DNA affinity of H-NS depends substantially on the linker,³⁰ which is otherwise considered responsible for tetramerization of the protein.^38,39 Furthermore, also the N-terminal 46 residues, which should constitute the heart of the dimerization domain of H-NS^10,33-35 may contribute to the interaction with DNA.^10,35,38 This agrees with genetic data suggesting that deletion of the first 20 N-terminal residues abolishes the capacity of H-NS to repress transcription of at least some genes.¹⁰ On the other hand, both recognition of bent DNA and bending non-curved DNA by H-NS are severely impaired by mutations of P115, (within the DNA binding domain), which affect H-NS oligomerization without influencing its basal DNA binding capacity. Similar properties are also conferred by other mutations in both N-domain (residues 1-20) and C-domain (W108).^{10,11,35,38,39} Although these phenotypes could arise from defects of protein-protein and/or protein-DNA interaction, overall these results indicate the role played by H-NS oligomerization in the selective binding to bent DNA and induction of DNA curvature. The images of H-NS-plasmid DNA complexes obtained by atomic force microscopy (AFM)⁴⁰ (see also fig. 3) are fully compatible with this premise.

Figure 3

Opposite effects of HU and H-NS on DNA structure. A relaxed pUC19 plasmid (central panel) is opened by addition of 1 dimer HU / 9 bp (left panel) and compacted by 1 dimer H-NS / 12 bp (right panel). Reprinted by permission of Federation of the European (more...)

Indeed, at one H-NS dimer/12 bp of nicked circular plasmid, H-NS was found to cause two types of DNA condensation. In the first type of complex, large tracts of ds-DNA are held together by H-NS bridges while the rest of the DNA forms double stranded loops. The bridges were interpreted as the result of tetramerization of H-NS dimers bound to two separate double stranded helices. The second type of complex is characterized by globular foci of H-NS incorporating large amounts of DNA while the rest of the plasmid remains partly naked and partly subject to the lateral condensation characteristic of the first type of complex. The DNA contour length is reduced = 3% in the first type of complex, possibly due to interwinding of the two helices, and by = 25% in the second. When two-times more protein is added, unstable rod-like structures whose compaction may be as high as 50% may appear.⁴⁰

Although it is difficult to judge whether and how the complexes observed by EM and AFM⁴⁰ could be related to complexes surmised from other types of experimental evidence, the occurrence of at least two types of H-NS-DNA interactions leading to the formation of different types of nucleoprotein complexes is also indicated by fluorescence spectroscopy^15,41 and by H-NS footprinting analysis.^27,42 One type of complex could involve a small number of H-NS molecules nucleated around select positions of the chromosome and would depend on the H-NS capacity to form tetramers or small oligomers inducing duplex-duplex bridging. Another, less specific type of complex, would rely on the basic DNA binding capacity of H-NS and would engage a much larger number of H-NS molecules linearly polymerized along the duplex, eventually resulting in larger aggregates with strongly condensed DNA.^27,40,42,43 The moderate compaction of the nucleoids clearly observed in cells overproducing a mutant H-NS (ΔG112-P115) having wild-type (wt) affinity for non-curved DNA but with strongly reduced affinity for curved DNA^11,44 (fig. 4A) could be due to this second type of H-NS-DNA interaction. Likewise, the repression of the bgl operon by a transcriptional silencing mode could involve this type of H-NS-DNA interaction and could account for the suppression of the bgl phenotype apparently sustained by the formation of heterodimers between H-NS mutants lacking the DNA-binding domain and wt StpA.^45,46 Indeed, formation of these chimeric dimers containing only one DNA binding site is incapable of suppressing other, obviously more stringent hns phenotypes (e.g., see refs. 10, 47 and references therein).

Figure 4

Appearance of the nucleoids of E. coli K12ΔH1ΔTrp subjected to cryo-fixation before (A) and after (B) overexpression of H-NS (ΔGly112-Pro115), a mutant protein with intact basal DNA-binding capacity but unable to recognize bent (more...)

The observation that different H-NS-DNA complexes are formed as a function of the variation of environmental parameters⁴⁰ matches the finding that H-NS oligomerization equilibria are sensitive both in vitro³⁷ and in vivo (Stella et al., in preparation) to variations of physical parameters (protein concentration, temperature and ionic strength) corresponding to the environmental cues to which H-NS responds in vivo. These findings suggest that changes of the intracellular milieu may modulate H-NS function through the formation of different types of complexes.

Architectural Role of H-NS

DNA compaction is believed to involve DNA supercoiling and looping. In light of their capacity to affect these parameters, proteins HU and H-NS have long been considered responsible for condensing chromosomal DNA (= 1.5 mm in E. coli) inside the nucleoid. However, this viewpoint has been challenged by the observation that the HU/H-NS ratio varies considerably (2.5-fold) in the cell as a function of the growth phase,¹⁹ by data showing that none of the nucleoid-associated proteins contributes to chromosome looping⁴⁸ and by recent AFM observations suggesting that HU and H-NS have antagonistic effects on chromosomal architecture,⁴⁹ with HU stretching out and extending a circular duplex and H-NS causing instead its compaction (fig.3).⁴⁹

Thus, although H-NS is commonly credited with the dual roles of transcriptional repressor and of architectural protein of the nucleoid, the evidence for the latter role is somewhat weak and indirect. The nucleoid localization of H-NS in both cryo-fixed²⁰ and viable cells⁵⁰ (fig. 5) is compatible with but does not constitute proof in its favor. The early evidence that H-NS causes DNA compaction² has recently been challenged by Amit et al.⁵¹ who suggested that the increased sedimentation velocity acquired by a plasmid upon H-NS binding is due to the binding of a large number of protein molecules. Furthermore, the finding that H-NS polymerization (one dimer/15-20 bp) increased the bending rigidity of the double helix and increased the end-to-end distance of a DNA fragment clashes with the notion that H-NS causes DNA compaction.² Thus, the most convincing evidence for an architectural role comes from the H-NS-induced DNA condensation seen by electron microscopy (fig. 3)^24,40 and from the observation that nucleoids are clearly condensed upon overproduction of H-NS (ΔG112-P115), a non-lethal event which does not cause generalized transcriptional repression (fig. 4A,B).⁴⁴ The altered DNA-binding properties of this H-NS mutant (normal capacity to bind non-curved DNA but reduced capacity to bind curved DNA) support our previous interpretation that it is the non-specific coating of DNA by H-NS which ultimately causes clumping and compaction of the duplex.

Figure 5

In vivo fluorescence localization of H-NS-GFP (green fluorescent protein) fusion protein (A) and DAPI-stained nucleoids (B) in exponential phase E. coli JM109 cells 30 min after IPTG induction of phns-gfp - ASV (Alanine, Serine, Valine). The bar represents (more...)

DNA Bending and H-NS Activity

Beyond their immediate informational significance, the nucleotide sequences characterizing the individual genomes determine intrinsic structural properties and spatial structures of the DNA which may represent an additional source of genetic information exploited by cells to regulate life processes. Sequence-directed DNA curvature represents one of the most important and widespread of these structural features. These curvatures, present in a substantial fraction of the promoters (Chapter 3), show conservative patterns of distribution in the genomes of mesophilic bacteria, being preferentially located 40-200 bp upstream from the nearest transcriptional start.⁵² Thus, DNA curvatures represent a well known example of the functional importance which the local architecture of the genome might acquire^53-55 since they can affect a large number of cellular phenomena, transcription being one of the most prominent.^56,57 Indeed, bent DNA has been shown to affect bacterial transcription sometimes facilitating the binding to promoters of RNA polymerase and/or of nucleoid-associated proteins (e.g., IHF, FIS, HU, H-NS, Lrp, Crp) functioning as activators and/or repressors, depending on the genetic system. Whereas in some cases bent DNA may play a direct role in transcriptional regulation, in other cases it plays an indirect role insofar as the binding of DNA-binding proteins stabilizes or enhances the curvature of a pre-existing DNA loop giving rise to nucleoprotein complexes which block downstream transcription. In fact, these nucleoid-associated proteins are not only able to recognize and bind curved DNA regions displaying more or less stringent consensus sequences, but are often able to induce DNA bending.^54-58 Concerning H-NS in particular the presence of intrinsically curved DNA regions is a common occurrence in H-NS-sensitive promoters and the molecular basis of its regulatory activity often resides in its preferential interaction with intrinsically curved DNA and in its ability to induce DNA bending.^11,59,60

Transcriptional Regulation by H-NS and Thermosensing through Curved DNA

H-NS affects the expression of a large number of enterobacterial genes, some coding for housekeeping functions, and many implicated in cellular responses to environmental changes, including the virulence factors whose expression is triggered by the passage from the external to the intestinal environment.^6,58 Thus, a variety of phenotypes is associated with hns mutations: increase in pH resistance;^61,62 loss of mobility;^63,64 serine-⁶⁵ and cold-susceptibility⁶⁶ the latter phenotype likely related to the fact that H-NS is a cold-shock protein.^67,68 However, H-NS does not regulate all its target genes through the same mechanism. In fact, although H-NS acts as a transcriptional repressor in the majority of the cases, it can also act as a translational repressor⁶⁹ or affect gene expression at different post-transcriptional levels.⁷⁰ Nevertheless, to focus on the specific scope of this review, here we shall describe only the cases in which H-NS acts as a transcriptional repressor, with emphasis on the cases in which the repression mechanism can be related to the presence and participation of a curved fragment of DNA. However, it should be borne in mind that even in those cases in which H-NS clearly acts as a transcriptional repressor on a promoter containing a curved segment of DNA, the repression may involve different mechanisms. In fact, depending on the genetic system, the repression has been attributed to the capacity of H-NS to increase DNA compaction, to alter its topology or to induce the formation of more or less complex nucleoprotein particles as a result of its preferential interaction with bent DNA.^71-73 Furthermore, while for some genes the actual molecular mechanism causing repression has been attributed to promoter occlusion (fig. 6A)^25,71,74 in other cases it has been demonstrated that it is due to RNA polymerase entrapment in the promoter caused by H-NS-mediated looping of DNA (Fig. 6B).^60,75

Figure 6

Schematic representation of two mechanisms by which H-NS may cause transcriptional repression. (A) Promoter occlusion as postulated in the case of hns autorepression. (B) RNA polymerase entrapping as shown for the P1 promoter of rrnB. Further details (more...)

As mentioned above, intrinsic DNA curvatures may be the primary actors in determining transcriptional regulation insofar as DNA curvatures are sensitive to changes in environmental parameters such as temperature, magnesium and polyamine concentrations.⁷⁶ Furthermore, the fact that these parameters differentially affect the curvature depending on the DNA sequence indicates that at least some DNA curvatures could be ideal and specific sensors implicated in gearing gene expression to environmental changes. It should not be ignored, however, that the same (or similar) physical parameters which can influence DNA curvature can also affect, both in vitro³⁷ and in vivo (Stella et al., manuscript in preparation), the oligomerization equilibria of H-NS which, in turn, can influence its activity on the DNA.

H-NS is involved in many regulatory circuits controlling the expression of virulence genes^{6,58,71,77-85} and different mechanisms may account for this function. In fact, while in some cases transcription is prevented by direct binding of H-NS to target genes,⁷¹ in others, the effect is indirectly caused by the influence of H-NS on the expression of specific transcriptional regulators or by its inhibition of DNA modifications involved in gene activation.⁸⁴

A particular type of H-NS-dependent transcriptional regulation, which has recently emerged as a widespread strategy for controlling the expression of pathogenicity, is that mediated by environmental-dependent changes of DNA curvature. Some detailed information concerning this type of regulation is available in the case of Shigella virF, the gene whose product triggers the pathogenicity cascade in the etiological agent of human bacillary dysentery.⁸⁶ Like in other pathogenic enterobacteriaceae the transcription of the virulence genes of Shigella is strictly dependent on temperature, being repressed in non-intestinal environments characterized by lower temperature and osmolarity than in the intestine. Thus, the entry of the bacterium into the warm (37°C) host milieu represents a central cue triggering the expression of virulence factors and, ultimately, of the virulence phenotype.^87,88 The primary event following the temperature upshift is the synthesis of VirF, a transcriptional activator encoded by virF, a gene transcriptionally repressed below the threshold temperature of = 32°C.^81,89 In turn, the increased intracellular concentration of VirF triggers the activation of virB, a second regulator gene^89,90 whose product activates several invasion operons.^58,85 The transcriptional repression of virF below 32°C is mediated by H-NS and depends on the presence of an intrinsically bent region whose center at 4°C is located at 137 bp upstream from the transcriptional start. This bend is flanked by two rather extended H-NS sites, one of which overlaps the core elements of the promoter. Both extent and localization of this curvature were found to be temperature-sensitive. In fact, gel electrophoretic analyses between 20°C and 60°C revealed that with increasing temperature there is a reduction of DNA curvature (which is maximum at 4°C) so that the bend seems to collapse when the temperature approaches the transcription-permissive conditions (32°C) thereby allowing the transition of virF promoter from the repressed to the derepressed state.^81,83 Furthermore, mutations affecting the curvature's amplitude or the relative orientation of the two H-NS sites severely affected in vivo and in vitro H-NS binding to virF and the thermoregulation of its expression.⁸³ In addition, it was shown that when the temperature increases from 4° to 60°C, the center of the DNA bend within the virF promoter slides downstream by almost eight helical turns; it is also noteworthy that the sliding rate is not linear with temperature but undergoes the largest increase within the narrow range (28°C-32°C ) which corresponds to the transition from transcriptional repression to derepression. Within this range, the center of bending moves downstream by one helical turn going from -116 to -106 (fig. 7).⁸³ Sliding of a bending center has also been reported in other cases such as in the upstream region of the light-responsive promoters of the cyanobacterial psbA gene.^91,92 Although the biological significance of these slidings remains largely unclear, the observation that the bending center within the virF promoter region “moves” towards the boundary of a FIS binding site (site III) and that the movement of the bending center is slow up to 28°C while becoming rapid when the transition temperature is approached suggests that this shift might be related to the ability of FIS to partially relieve H-NS-induced virF repression at the transition temperature.⁸² If it could be demonstrated that the sliding of the virF bending center renders this promoter region more accessible to FIS binding, thereby facilitating this protein in antagonizing H-NS-mediated repression, it would be tempting to interpret the bend-movement in virF as an example of a (general?) strategy to optimize gene control in multifactorial regulatory systems by the gradual environmentally-guided unmasking of specific DNA targets.

Figure 7

Schematic representation of the S. flexneri virF promoter in the (A) repressed configuration and (B) derepressed configuration. As described in the text, the H-NS- mediated repression is switched on and off by the presence of a thermosensor represented (more...)

Taken together, the available data indicate that an intrinsic DNA bend within the promoter region of virF behaves as a thermal sensor and the geometric characteristics of this DNA curvature play an essential role in the thermoregulation of Shigella virulence expression (fig.7).^81,83

Other hypotheses were also presented in the past to explain the H-NS-dependent thermoregulation of virF expression. They were formulated in terms of changes in H-NS structure and modifications of DNA supercoiling (reviewed in refs. 31, 93). However, they have been disproven eventually. In fact, transcriptional repression by H-NS requires a supercoiled DNA target but the role of DNA supercoiling is only allowing H-NS to build a repressing complex and not acting as the temperature sensor.⁸¹ Furthermore, recent in vitro³⁷ and in vivo (Stella et al, manuscript in preparation) data indicate that H-NS oligomerization does change with temperature but not within the temperature interval in which the virF promoter undergoes the inactive/active transition.

In addition to the just described case of the virF thermoregulation, other cases are known in which a nucleic acid acts as a thermosensor in the control of virulence expression. For instance, a more stable binding of the YmoA repressor to the virF promoter of Yersinia enterocolitica, determined by a stronger DNA curvature induced by low temperature, results in silencing the virulence genes encoded by the Yops plasmid.^94,95 Furthermore, several genetic systems are known in which an increased DNA curvature induced by low temperature favors transcription instead of repressing it. For instance, like in the case of Shigella virF, the promoter of Clostridium perfringens plcC gene displays an increased bent structure at lower temperatures but in this case the expression is more efficient at low temperature because of the acquisition of a suitable template conformation.⁹⁶ Similarly, an increased DNA curvature at low temperature favors the binding of IHF and stimulates transcription from the PL promoter of bacteriophage λ.⁹⁷

A temperature-mediated change in DNA flexibility has also been implicated in the repression at low temperature of E.coli hly, an operon which is either plasmid- or chromosome-encoded, depending on the bacterial strain.⁷⁹ This operon, whose thermoregulated expression is under the control of H-NS and DNA curvature, encodes the α-hemolysin Hly, a toxin produced by several uropathogenic strains of E. coli. In the case of the hly operon carried by pHly152, a regulatory sequence (the 650 bp-long hlyR) is located > 1.5 kb upstream of the three promoters of hlyC, the first gene of the operon, being separated from the latter by an IS2 insertion element. Deletion of hlyR results in constitutive repression of hemolysin expression, a phenotype presumably connected with the loss of ops, a transcriptional antitermination element located in hlyR. Several lines of evidence indicate that the thermoregulated expression of hly operon is regulated by a nucleoprotein complex formed by H-NS and Hha, a protein homologous to YmoA, the above-mentioned temperature-dependent modulator of Yersinia virulence factors.⁹⁸ In fact, while hha mutants display only a partial derepression of temperature- and osmolarity-mediated expression of hemolysin, hha-hns double mutants were defective in both thermo- and osmo-regulation. Furthermore, while Hha displays a strong affinity for H-NS, this protein alone does not show any DNA binding preference, unlike its partner H-NS which preferentially binds to two sites in the regulatory region of the hly operon. The two sites, the first partially overlapping the core elements of the two most upstream promoters of hlyC and the second located = 2 kb upstream, are separated by an intrinsic DNA curvature predicted in silico and observed by AFM. Deletion analysis demonstrated that the upstream site is important for thermoregulation of the operon, and temperature was found to influence the affinity of H-NS for a DNA fragment containing both sites. The higher affinity of H-NS for its sites at low temperature correlates well with its more efficient transcriptional repression observed in vitro and with the constitutive hemolytic phenotype of the hns mutants observed in vivo. A current model explaining the H-NS/Hha-mediated temperature-dependent repression of hemolysin expression is that Hha generates hetero-oligomeric complexes with H-NS which are better suited than H-NS homo-oligomers in the temperature dependent repression of the hly operon. The H-NS binding specificity ensures that these complexes interact specifically with the DNA targets at the two sides of the bend, simultaneously occluding the transcription antiterminator sequence ops and at least two hlyC promoters.

Conclusion

H-NS has emerged as a major actor in the transcriptional regulation of gene expression, especially for genes involved in the cellular response to environmental changes and in virulence. While a general feature of the genes subject to H-NS control is the presence of an intrinsic DNA curvature, more work is necessary to establish whether a unitary mechanism underlies the H-NS functions and to clarify the molecular basis of the specificity governing H-NS-DNA interaction. Also to be clarified are the mechanism and the structural consequences of the binding.

Acknowledgements

This work was supported by MIUR grants PRIN 2001 (to CLP and COG) and PRIN 2002 (to COG).

References

1.

Lammi M, Paci M, Pon C. et al. Proteins from the prokaryotic nucleoid. biochemical and 1H-NMR studies on three bacterial histone-like proteins. In: Hubscher H, Spadari S, eds. Proteins Involved in DNA Replication. New York: Plenum Publishing Co. 1984:467–477. [PubMed: 6395665]

2.

Spassky A, Rimsky S, Garreau H. et al. H1a, an E. coli DNA-binding protein which accumulates in stationary phase, strongly compacts DNA in vitro. Nucleic Acids Res. 1984;12:5321–5340. [PMC free article: PMC318922] [PubMed: 6379600]

3.

Jacquet M, Cukier-Kahn R, Pla J. et al. A thermostable protein factor acting on in vitro DNA transcription. Biochem Biophys Res Commun. 1971;45:1597–1607. [PubMed: 4942735]

4.

Cukier-Kahn R, Jacquet M, Gros F. Two heat-resistant, low molecular weight proteins from Escherichia coli that stimulate DNA-directed RNA synthesis. Proc Natl Acad Sci USA. 1972;69:3643–3647. [PMC free article: PMC389839] [PubMed: 4566454]

5.

Ussery DW, Hinton JC, Jordi BJ. et al. The chromatin-associated protein H-NS. Biochimie. 1994;76:968–980. [PubMed: 7748941]

6.

Atlung T, Ingmer H. H-NS: a modulator of environmentally regulated gene expression. Mol Microbiol. 1997;24:7–17. [PubMed: 9140961]

7.

Schroeder O, Wagner R. The bacterial regulatory protein H-NS - a versatile modulator of nucleic acid structures. Biol Chem. 2002;383:945–960. [PubMed: 12222684]

8.

Tippner D, Afflerbach H, Bradaczek C. et al. Evidence for a regulatory function of the histone-like Escherichia coli protein H-NS in ribosomal RNA synthesis. Mol Microbiol. 1994;11:589–604. [PubMed: 7512187]

9.

Falconi M, Gualtieri MT, La Teana A. et al. Proteins from the prokaryotic nucleoid: primary and quaternary structure of the 15-kD Escherichia coli DNA binding protein H-NS. Mol Microbiol. 1988;2:323–329. [PubMed: 3135462]

10.

Ueguchi C, Suzuki T, Yoshida T. et al. Systematic mutational analysis revealing the functional domain organization of Escherichia coli nucleoid protein H-NS. J Mol Biol. 1996;263:149–162. [PubMed: 8913298]

11.

Spurio R, Falconi M, Brandi A. et al. The oligomeric structure of nucleoid protein H-NS is necessary for recognition of intrinsically curved DNA and for DNA bending. EMBO J. 1997;16:1795–1805. [PMC free article: PMC1169782] [PubMed: 9130723]

12.

Pon CL, Calogero R, Gualerzi CO. Identification, cloning, nucleotide sequence and chromosomal map location of hns, the structural gene for Escherichia coli DNA-binding protein H-NS. Mol Gen Genet. 1988;212:199–202. [PubMed: 2841565]

13.

La TeanaA, Falconi M, Scarlato V. et al. Characterization of the structural genes for the DNA-binding protein H-NS in Enterobacteriaceae. FEBS Lett. 1989;244:34–38. [PubMed: 2494066]

14.

Higgins CF, Hinton JC, Hulton CS. et al. Protein H1: a role for chromatin structure in the regulation of bacterial gene expression and virulence? Mol Microbiol. 1990;4:2007–2012. [PubMed: 2128526]

15.

Friedrich K, Gualerzi CO, Lammi M. et al. Proteins from the prokaryotic nucleoid. Interaction of nucleic acids with the 15 kDa Escherichia coli histone-like protein H-NS. FEBS Lett. 1989;229:197–202. [PubMed: 2450042]

16.

Bracco L, Kotlarz D, Kolb A. et al. Synthetic curved DNA sequences can act as transcriptional activators in Escherichia coli. EMBO J. 1989;8:4289–4296. [PMC free article: PMC401636] [PubMed: 2512122]

17.

Yamada H, Muramatsu S, Mizuno T. An Escherichia coli protein that preferentially binds to sharply curved DNA. J Biochem (Tokyo). 1990;108:420–425. [PubMed: 2126011]

18.

Yamada H, Yoshida T, Tanaka K. et al. Molecular analysis of the Escherichia coli hns gene encoding a DNA binding protein, which preferentially recognizes curved DNA sequences. Mol Gen Genet. 1991;230:332–336. [PubMed: 1745240]

19.

Azam TA, Iwata A, Nishimura A. et al. Growth phase-dependent variation in protein composition of the Escherichia coli nucleoid. J Bacteriol. 1999;181:6361–6370. [PMC free article: PMC103771] [PubMed: 10515926]

20.

Dürrenberger M, La TeanaA, Citro G. et al. Escherichia coli DNA-binding protein H-NS is localized in the nucleoid. Res Microbiol. 1991;142:373–380. [PubMed: 1871423]

21.

Marsh M, Hillyard DR. Nucleotide sequence of hns encoding the DNA-binding protein H-NS of Salmonella typhimurium. Nucleic Acids Res. 1990;18:3397. [PMC free article: PMC330951] [PubMed: 2113274]

22.

Tendeng C, Krin E, Soutourina OA. et al. A novel H-NS-like protein from an antarctic psychrophilic bacterium reveals a crucial role for the N-terminal domain in thermal stability. J Biol Chem. 2003;278:18754–18760. [PubMed: 12637536]

23.

Zhang A, Belfort M. Nucleotide sequence of a newly-identified Escherichia coli gene, stpA, encoding an H-NS-like protein. Nucleic Acids Res. 1992;20:6735. [PMC free article: PMC334595] [PubMed: 1480493]

24.

Dame RT, Wyman C, Goosen N. Structural basis for preferential binding of H-NS to curved DNA. Biochimie. 2001;83:231–234. [PubMed: 11278073]

25.

Falconi M, Higgins NP, Spurio R. et al. Expression of the gene encoding the major bacterial nucleoid protein H-NS is subject to transcriptional auto-repression. Mol Microbiol. 1993;10:273–282. [PubMed: 28776853]

26.

Zuber F, Kotlarz D, Rimsky S. et al. Modulated expression of promoters containing upstream curved DNA sequences by the Escherichia coli nucleoid protein H-NS. Mol Microbiol. 1994;12:231–240. [PubMed: 8057848]

27.

Rimsky S, Zuber F, Buckle M. et al. A molecular mechanism for the repression of transcription by the H-NS protein. Mol Microbiol. 2001;42:1311–1323. [PubMed: 11886561]

28.

Jordi BJ, Fielder AE, Burns CM. DNA binding is not sufficient for H-NS-mediated repression of proU expression. J Biol Chem. 1997;272:12083–12090. [PubMed: 9115277]

29.

Shindo H, Iwaki T, Ieda R. et al. Solution structure of the DNA binding domain of a nucleoid-associated protein, H-NS, from Escherichia coli. FEBS Lett. 1995;360:125–131. [PubMed: 7875316]

30.

Shindo H, Ohnuki A, Ginba H. et al. Identification of the DNA binding surface of H-NS protein from Escherichia coli by heteronuclear NMR spectroscopy. FEBS Lett. 1999;455:63–69. [PubMed: 10428473]

31.

Dorman CJ, Hinton JC, Free A. Domain organization and oligomerization among H-NS-like nucleoid-associated proteins in bacteria. Trends Microbiol. 1999;7:124–128. [PubMed: 10203842]

32.

Smyth CP, Lundback T, Renzoni D. et al. Oligomerization of the chromatin-structuring protein H-NS. Mol Microbiol. 2000;36:962–972. [PubMed: 10844682]

33.

Renzoni D, Esposito D, Pfuhl M. et al. Structural characterization of the N-terminal oligomerization domain of the bacterial chromatin-structuring protein, H-NS. J Mol Biol. 2001;306:1127–1137. [PubMed: 11237622]

34.

Esposito D, Petrovic A, Harris R. et al. H-NS oligomerization domain structure reveals the mechanism for high order self-association of the intact protein. J Mol Biol. 2002;324:841–850. [PubMed: 12460581]

35.

Bloch V, Yang Y, Margeat E. et al. The H-NS dimerization domain defines a new fold contributing to DNA recognition. Nat Struct Biol. 2003;10:212–218. [PubMed: 12592399]

36.

Schroeder O, Tippner D, Wagner R. Toward the three-dimensional structure of the Escherichia coli DNA-binding protein H-NS: a CD and fluorescence study. Biochem Biophys Res Commun. 2001;282:219–227. [PubMed: 11263995]

37.

Ceschini S, Lupidi G, Pon CL. et al. Multimeric self-assembly equilibria involving the histone-like protein H-NS. J Biol Chem. 2000;275:729–734. [PubMed: 10625601]

38.

Williams RM, Rimsky S, Buc H. Probing the structure, function, and interactions of the Escherichia coli H-NS and StpA proteins by using dominant negative derivatives. J Bacteriol. 1996;178:4335–4343. [PMC free article: PMC178199] [PubMed: 8755860]

39.

Ueguchi C, Seto C, Suzuki T. et al. Clarification of the dimerization domain and its functional significance for the Escherichia coli nucleoid protein H-NS. J Mol Biol. 1997;274:145–151. [PubMed: 9398522]

40.

Dame RT, Wyman C, Goosen N. H-NS mediated compaction of DNA visualised by atomic force microscopy. Nucleic Acid Res. 2000;28:3304–3510. [PMC free article: PMC110753] [PubMed: 10982869]

41.

Tippner D, Wagner R. Fluorescence analysis of the Escherichia coli transcription regulator H-NS reveals two distinguishable complexes dependent on binding to specific or nonspecific DNA sites. J Biol Chem. 1995;270:22243–22247. [PubMed: 7673203]

42.

Badaut C, Williams R, Arluison V. et al. The degree of oligomerization of the H-NS nucleoid structuring protein is related to specific binding to DNA. J Biol Chem. 2002;277:41657–41666. [PubMed: 12200432]

43.

Rimsky S, Spassky A. Sequence determinants for H1 binding on Escherichia coli lac and gal promoters. Biochemistry. 1990;29:3765–3771. [PubMed: 2160266]

44.

Spurio R, Duerrenberger M, Falconi M. et al. Lethal overproduction of the Escherichia coli nucleoid protein H-NS: ultramicroscopic and molecular autopsy. Mol Gen Genet. 1992;231:201–211. [PubMed: 1310520]

45.

Johansson J, Eriksson S, Sonden B. et al. Heteromeric interactions among nucleoid-associated bacterial proteins: localization of StpA-stabilizing regions in H-NS of Escherichia coli. J Bacteriol. 2001;183:2343–2347. [PMC free article: PMC95143] [PubMed: 11244076]

46.

Free A, Porter ME, Deighan P. et al. Requirement for the molecular adapter function of StpA at the Escherichia coli bgl promoter depends upon the level of truncated H-NS protein. Mol Microbiol. 2001;42:903–917. [PubMed: 11737635]

47.

Ohta T, Ueguchi C, Mizuno T. rpoS function is essential for bgl silencing caused by C-terminally truncated H-NS in Escherichia coli. J Bacteriol. 1999;181:6278–6283. [PMC free article: PMC103760] [PubMed: 10515915]

48.

Brunetti R, Prosseda G, Beghetto E. et al. The looped domain organization of the nucleoid in histone-like protein defective Escherichia coli strains. Biochimie. 2001;83:873–882. [PubMed: 11698109]

49.

Dame RT, Goosen N. HU: promoting or counteracting DNA compaction? FEBS Lett. 2002;529:151–156. [PubMed: 12372591]

50.

Giangrossi M, Exley RM, Le Hegarat F. et al. Different in vivo localization of the Escherichia coli proteins CspD and CspA. FEMS Microbiol Lett. 2001;202:171–176. [PubMed: 11520610]

51.

Amit R, Oppenheim AB, Stavans J. Increased bending rigidity of single DNA molecules by H-NS, a temperature and osmolarity sensor. Biophys J. 2003;84:2467–2473. [PMC free article: PMC1302812] [PubMed: 12668454]

52.

Bolshoy A, Nevo E. Ecologic genomics of DNA: upstream bending in prokaryotic promoters. Genome Res. 2000;10:1185–1193. [PMC free article: PMC310903] [PubMed: 10958636]

53.

Ohyama T. Intrinsic DNA bends: an organizer of local chromatin structure for transcription. Bioessays. 2001;23:708–715. [PubMed: 11494319]

54.

Travers AA. DNA conformation and protein binding. Annu Rev Biochem. 1989;58:427–452. [PubMed: 2673015]

55.

Travers AA. Reading the minor groove. Nat Struct Biol. 1995;2:615–618. [PubMed: 7552720]

56.

Perez-Martin J, de LorenzoV. Clues and consequences of DNA bending in transcription. Annu Rev Microbiol. 1997;51: 593–628. [PubMed: 9343361]

57.

McLeod SM, Johnson RC. Control of transcription by nucleoid proteins. Curr Opin Microbiol. 2001;4:152–159. [PubMed: 11282470]

58.

Prosseda G, Falconi M, Nicoletti M. et al. Histone-like proteins and the Shigella invasivity regulon. Res Microbiol. 2002;153:461–468. [PubMed: 12405354]

59.

Afflerbach H, Schroeder O, Wagner R. Conformational changes of the upstream DNA mediated by H-NS and FIS regulate E. coli rrnB P1 promoter activity. J Mol Biol. 1999;286:339–353. [PubMed: 9973555]

60.

Schroeder O, Wagner R. The bacterial DNA-binding protein H-NS represses ribosomal RNA transcription by trapping RNA polymerase in the initiation complex. J Mol Biol. 2000;298:737–748. [PubMed: 10801345]

61.

Hommais F, Krin E, Laurent-Winter C. et al. Large-scale monitoring of pleiotropic regulation of gene expression by the prokaryotic nucleoid-associated protein, H-NS. Mol Microbiol. 2001;40:20–36. [PubMed: 11298273]

62.

De BiaseD, Tramonti A, Bossa F. et al. The response to stationary-phase stress conditions in Escherichia coli: role and regulation of the glutamic acid decarboxylase system. Mol Microbiol. 1999;32:1198–1211. [PubMed: 10383761]

63.

Soutourina O, Kolb A, Krin E. et al. Multiple control of flagellum 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. 1999;181:7500–7508. [PMC free article: PMC94207] [PubMed: 10601207]

64.

Soutourina OA, Krin E, Laurent-Winter C. et al. Regulation of bacterial motility in response to low pH in Escherichia coli: the role of H-NS protein. Microbiology. 2002;148:1543–1551. [PubMed: 11988529]

65.

Lejeune P, Bertin P, Walon C. et al. A locus involved in kanamycin, chloramphenicol and L-serine resistance is located in the bglY-galU region of the Escherichia coli K12 chromosome. Mol Gen Genet. 1989;218:361–363. [PubMed: 2506428]

66.

Dersch P, Kneip S, Bremer E. The nucleoid-associated DNA-binding protein H-NS is required for the efficient adaptation of Escherichia coli K-12 to a cold environment. Mol Gen Genet. 1994;245:255–259. [PubMed: 7816034]

67.

La TeanaA, Brandi A, Falconi M. et al. Identification of a cold shock transcriptional enhancer of the Escherichia coli gene encoding nucleoid protein H-NS. Proc Natl Acad Sci U S A. 1991;88:10907–10911. [PMC free article: PMC53041] [PubMed: 1961761]

68.

Tendeng C, Badaut C, Drin E. et al. Isolation and characterization of vicH, encoding a new pleiotropic regulator in Vibrio cholerae. J Bacteriol. 2000;182:2026–2032. [PMC free article: PMC101921] [PubMed: 10715012]

69.

Yamashino T, Ueguchi C, Mizuno T. Quantitative control of the stationary phase-specific sigma factor, __in Escherichia coli: involvement of the nucleoid protein H-NS. EMBO J. 1995;14:594–602. [PMC free article: PMC398118] [PubMed: 7859747]

70.

Deighan P, Free A, Dorman CJ. A role for the Escherichia coli H-NS-like protein StpA in OmpF porin expression through modulation of micF RNA stability. Mol Microbiol. 2000;38:126–139. [PubMed: 11029695]

71.

Goransson M, Sonden B, Nilsson P. et al. Transcriptional silencing and thermoregulation of gene expression in Escherichia coli. Nature. 1990;344:682–685. [PubMed: 1691451]

72.

Ueguchi C, Mizuno T. The Escherichia coli nucleoid protein H-NS functions directly as a transcriptional repressor. EMBO J. 1993;12:1039–1046. [PMC free article: PMC413305] [PubMed: 8458322]

73.

Tupper AE, Owen-Hughes TA, Ussery DW. et al. The chromatin-associated protein H-NS alters DNA topology in vitro. EMBO J. 1994;13:258–268. [PMC free article: PMC394800] [PubMed: 8306968]

74.

Falconi M, Brandi A, La Teana. et al. Antagonistic involvement of FIS and H-NS proteins in the transcriptional control of hns expression. Mol Microbiol. 1996;19:965–975. [PubMed: 8830277]

75.

Dame RT, Wyman C, Wurm R. et al. Structural basis for H-NS mediated trapping of RNA polymerase in the open initiation complex at the rrnB P1. J Biol Chem. 2002;277:2146–2150. [PubMed: 11714691]

76.

Ussery DW, Higgins CF, Bolshoy A. Environmental influences on DNA curvature. J Biomol Struct Dyn. 1999;16:811–823. [PubMed: 10217451]

77.

Maurelli AT, Sansonetti PJ. Identification of a chromosomal gene controlling temperature regulated expression of Shigella virulence. Proc Natl Acad Sci USA. 1988;85:2820–2824. [PMC free article: PMC280091] [PubMed: 3282241]

78.

Beloin C, Dorman CJ. An extended role of the nucleoid structuring protein H-NS in the virulence gene regulatory cascade of Shigella flexneri. Mol Microbiol. 2003;47:825–838. [PubMed: 12535079]

79.

Madrid C, Nieto JM, Paytubi S. et al. Temperature-and H-NS-dependent regulation of a plasmid-encoded virulence operon expressing Escherichia coli haemolysin. J Bacteriol. 2002;184:5058–5066. [PMC free article: PMC135304] [PubMed: 12193622]

80.

Tobe T, Yoshikawa M, Mizuno T. et al. Transcriptional control of the invasion regulatory gene virB of Shigella flexneri: activation by VirF and repression by H-NS. J Bacteriol. 1993;175:6142–6149. [PMC free article: PMC206708] [PubMed: 7691791]

81.

Falconi M, Colonna B, Prosseda G. et al. Thermoregulation of Shigella and Escherichia coli EIEC pathogenicity. A temperature-dependent structural transition of DNA modulates accessibility of virF promoter to transcriptional repressor H-NS. EMBO J. 1998;17:7033–7043. [PMC free article: PMC1171051] [PubMed: 9843508]

82.

Falconi M, Prosseda G, Giangrossi M. et al. Involvement of FIS on the H-NS mediated regulation of the virF gene of Shigella and Escherichia coli EIEC. Mol Microbiol. 2001;42:439–452. [PubMed: 11703666]

83.

Prosseda G, Falconi M, Giangrossi M. et al. The virF promoter in Shigella: more than just a curved DNA stretch. Mol Microbiol. 2004;51:523–537. [PubMed: 14756791]

84.

White-Ziegler CA, Angus HillML, Braaten BA. et al. Thermoregulation of Escherichia coli pap transcription: H-NS is a temperature-dependent DNA methylation blocking factor. Mol Microbiol. 1998;28:1121–1137. [PubMed: 9680203]

85.

Dorman CJ, Porter ME. The Shigella virulence gene regulatory cascade: a paradigm of bacterial gene control mechanisms. Mol Microbiol. 1998;29:677–684. [PubMed: 9723908]

86.

Sansonetti PJ. Rupture, invasion and inflammatory destruction of the intestinal barrier by Shigella, making sense of prokaryote-eukaryote cross-talks. FEMS Microbiol Rev. 2001;25:3–14. [PubMed: 11152938]

87.

Hurme R, Rhen M. Temperature sensing in bacterial gene regulation-what it all boils down to. Mol Microbiol. 1998;30:1–6. [PubMed: 9786180]

88.

Konkel ME, Tilly K. Temperature-regulated expression of bacterial virulence genes. Microbes Infect. 2000;2:157–166. [PubMed: 10742688]

89.

Colonna B, Casalino M, Fradiani PA. et al. H-NS regulation of virulence gene expression in enteroinvasive Escherichia coli harboring the virulence plasmid integrated into the host chromosome. J Bacteriol. 1995;177:4703–4712. [PMC free article: PMC177236] [PubMed: 7642498]

90.

Durand JMB, Dagberg B, Uhlin BE. et al. Transfer RNA modification, temperature and DNA superhelicity have a common target in the regulatory network of the virulence of Shigella flexneri: the expression of the virF gene. Mol Microbiol. 2000;35:924–935. [PubMed: 10692168]

91.

Agrawal GK, Asayama M, Shirai M. A novel bend of DNA CIT: changeable bending-center sites of an intrinsic curvature under temperature conditions. FEMS Microbiol Lett. 1997;147:139–145. [PubMed: 9037772]

92.

Asayama M, Kato H, Shibato J. et al. The curved DNA structure in the 5'- upstream region of the light-responsive genes : its universality, binding factor and function for cyanobacterial psbA transcription. Nucleic Acid Res. 2002;30:4658–4666. [PMC free article: PMC140650] [PubMed: 12409456]

93.

Dorman CJ, Deighan P. Regulation of gene expression by histone-like proteins in bacteria. Curr Opin Genet Dev. 2003;13:179–184. [PubMed: 12672495]

94.

Rohde JR, Luan X, Rohde H. et al. The Yersinia enterocolitica pYV virulence plasmid contains multiple intrinsic DNA bends which melt at 37°C. J Bacteriol. 1999;181:4198–4204. [PMC free article: PMC93920] [PubMed: 10400576]

95.

Drlica K, Perl-Rosenthal NR. DNA switches for thermal control of gene expression. Trends Microbiol. 1999;7:425–426. [PubMed: 10542416]

96.

Katayama S, Matsushita O, Jung C. et al. Promoter upstream bent DNA activates the transcription of the Clostridium perfringens phospholipase C gene in a low temperature-dependent manner. EMBO J. 1999;18:3442–3450. [PMC free article: PMC1171423] [PubMed: 10369683]

97.

Giladi H, Goldenberg D, Koby S. et al. Enhanced activity of the bacteriophage λPL promoter at low temperature. Proc Natl Acad Sci USA. 1995;92:2184–2188. [PMC free article: PMC42448] [PubMed: 7892244]

98.

Cornelis GR, Sluiters C, Delor I. et al. ymoA, a Yersinia enterocolitica chromosomal gene modulating the expression of virulence functions. Mol Microbiol. 1991;5:1023–1034. [PubMed: 1956283]