Skip to main content
NCBI home page
Biophysical Journal logoLink to Biophysical Journal
. 2015 Oct 6;109(7):1321–1329. doi: 10.1016/j.bpj.2015.08.016

H-NS Regulates Gene Expression and Compacts the Nucleoid: Insights from Single-Molecule Experiments

Ricksen S Winardhi 1,2, Jie Yan 1,2,∗, Linda J Kenney 1,3,4,5,∗∗
PMCID: PMC4601063  PMID: 26445432

Abstract

A set of abundant nucleoid-associated proteins (NAPs) play key functions in organizing the bacterial chromosome and regulating gene transcription globally. Histone-like nucleoid structuring protein (H-NS) is representative of a family of NAPs that are widespread across bacterial species. They have drawn extensive attention due to their crucial function in gene silencing in bacterial pathogens. Recent rapid progress in single-molecule manipulation and imaging technologies has made it possible to directly probe DNA binding by H-NS, its impact on DNA conformation and topology, and its competition with other DNA-binding proteins at the single-DNA-molecule level. Here, we review recent findings from such studies, and provide our views on how these findings yield new insights into the understanding of the roles of H-NS family members in DNA organization and gene silencing.

Main Text

Regulation of chromosome organization and gene expression by NAPs

Genetic information in bacteria is highly organized within the cell in a structure referred to as the nucleoid. Early architectural insights into bacterial chromosomes come from electron microscopy imaging, which reveals a higher-order structure that appears rosette-like, with supercoiled loop domains emanating from the central core (1–3). Multiple mechanisms contribute to the complex organization of bacterial chromosomes. One contributing factor is DNA supercoiling, produced by the activity of DNA gyrase and regulated by various topoisomerases (4,5). In addition, macromolecular crowding may also contribute to chromosomal packaging by promoting DNA condensation (6). However, these factors alone are insufficient to explain the complex organizational patterns of bacterial chromosomes. It is now clear that a set of abundant DNA-binding proteins play crucial roles in organizing the bacterial chromosome. They work in concert to shape the bacterial nucleoid in a dynamic manner and are referred to as nucleoid-associated proteins (NAPs).

The abundance of NAPs varies with the growth phase of bacteria (7). During exponential growth, the major NAPs include the factor for inversion stimulation, the host factor for phage Qβ, HU, the suppressor of td mutant phenotype A (StpA), the histone-like nucleoid-structuring protein (H-NS), and the integration host factor (IHF). During stationary phase or nutritional deprivation, the expression level of most of these NAPs decreases, whereas expression levels of DNA-binding proteins from starved cells (Dps) and IHF increase and these become the two most abundant proteins in the nucleoid. Such dynamic regulation of the differential abundance of NAPs allows bacterial cells to shape their nucleoid in response to various environmental changes.

NAPs typically have an equilibrium dissociation constant (Kd) on the order of 10–250 nM, depending on DNA sequence and solution conditions (8). Each NAP has an intracellular concentration over a wide range of a few to tens of micromolar, with a total concentration on the order of 100 μM (7), corresponding to roughly 100,000 total copies of NAPs. Escherichia coli has a 4.6 Mbp circular chromosome; therefore, the DNA in basepairs (bp) is estimated to be roughly two orders of magnitude higher than the intracellular NAPs. Simple chemical kinetics theory indicates that when the concentration of available protein-binding sites is greater than the protein concentration and the protein concentration is much greater than the Kd, the concentration of free/unbound proteins will be on the order of the Kd. This means that most of the NAPs are nucleoid associated, i.e., DNA-bound, which is supported by experimental evidence (9).

In addition to their role in chromosome organization, many NAPs have been shown to affect gene regulation on a global scale (10–13). In addition, the ability of bacterial cells to dynamically regulate the abundance of NAPs also enables global control of gene expression in response to environmental changes. The function of NAPs depends on how they bind to DNA locally at their binding sites and how they organize DNA over long length scales. Recent developments in single-DNA manipulation and imaging technologies make it possible to directly observe these processes and thus provide important insights into the function of these proteins.

NAPs typically cover 10–30 bp of DNA at their binding sites, deforming the DNA locally, depending on the particular NAP. Over a much longer length scale, NAPs can organize DNA into various conformations. This ability depends not only on local binding properties, but also on additional interactions between protein-bound DNA complexes with other protein-bound DNA complexes or with naked DNA segments. Therefore, it is necessary to distinguish between local DNA binding (at the binding site) and physical organization of DNA over long length scales. For example, as an NAP, IHF binds to nonspecific DNA and locally bends DNA at the binding site (<30°) (14,15). The DNA binding mode of IHF is hence defined as DNA bending. However, recent single-molecule manipulation and imaging experiments have shown that IHF can organize long DNA molecules into various complex conformations, including DNA condensation. This likely occurs (in the presence of magnesium) by bringing remote DNA sites together through DNA cross-linking, which is not obvious from its ability to locally bend DNA (15). Hereafter, we refer to this behavior as DNA juxtaposition. This review focuses on the DNA-binding properties of the H-NS family of NAPs and the implications of H-NS-dependent bacterial chromosome organization and gene regulation.

The H-NS family of proteins: universal gene regulators in bacteria

Proteins in the H-NS family, a focus of this review, share a common domain structure (Fig. 1). They consist of a C-terminal DNA-binding domain and a coiled-coil N-terminal domain that mediates oligomerization, forming higher-order homomeric or heteromeric complexes. At least two dimerization sites have been identified (see Fig. 1), which allows H-NS to form higher-order oligomers (16,17). The oligomerization and DNA-binding domains are joined via an unstructured flexible linker. Two NMR structures of the isolated N-terminus display parallel or antiparallel arrangement of the dimer interface (18,19). This obviously has important implications for how the DNA-binding domains are juxtaposed, if both conformations exist in vivo. H-NS generally binds nonspecifically to DNA, although it has a strong preference for AT-rich sequences, which are often found in DNA of external origin acquired by horizontal gene transfer (20).

Figure 1.

Figure 1

Open in a new tab

Domain organization of H-NS. H-NS is a 137 amino acid protein, functionally divided into an N-terminal oligomerization domain and a C-terminal DNA binding domain. H and β represent α-helices and β-sheets, respectively. At least two dimerization sites have been identified within the oligomerization domain, as indicated.

H-NS binding to DNA is important for two distinct functions in the cell, to shape the nucleoid (21–23) and to regulate gene expression. In general, H-NS functions to repress or silence genes, although it can indirectly act as a transcriptional activator by downregulating a negative regulator. For example, the flagellar master operon flhDC is negatively regulated by HdfR, which in turn is negatively regulated by H-NS (24). In addition, H-NS can also interact with some mRNAs to reposition the ribosome for more effective translation (25).

Many of the genes regulated by H-NS are involved in virulence in bacterial pathogens. The extent of H-NS binding to DNA and its silencing function are dependent on its oligomerization properties. In solution, H-NS exists as a complex mixture of differing oligomerization states (19,26,27), and it is believed that an H-NS dimer is the minimal functional binding unit (28,29). Amino acid substitutions in the N-terminus that render H-NS incapable of oligomerization abrogate gene silencing, suggesting that protein oligomerization plays a crucial role in silencing (27,30,31). Environmental stimuli such as temperature and osmolality can also alter the oligomerization states of H-NS in vivo and hence affect its gene silencing properties (32,33). The N-terminus of H-NS can mediate interaction with other proteins, such as StpA and gp5.5 protein, modulating other regulatory factors that affect gene repression (34,35).

The DNA-binding mode and physical organization of DNA by H-NS

To understand H-NS function, it is necessary to understand how it interacts with DNA, i.e., its DNA-binding modes as well as its physical organization of DNA. Recent imaging studies have provided direct visualization of the physical organization of DNA by H-NS, revealing that H-NS-family proteins were able to organize large DNA molecules into various conformations, including large extended filaments, hairpin-like large DNA bridges, and higher-order DNA condensations (22,23).

Atomic force microscopy (AFM) imaging experiments do not provide direct information as to the mode of binding at H-NS binding sites. However, the binding mode can be indirectly inferred from the force responses of single DNA molecules studied by single-molecule manipulation experiments (33,36,37). Analysis of the force-extension curves of DNA in the presence of H-NS has shown that H-NS stiffens DNA upon binding. Furthermore, analyses of force-extension curves of DNA complexed with H-NS at various H-NS concentrations revealed that H-NS bound to linear DNA tracks with positive cooperativity (38). Therefore, the DNA binding modes of H-NS can be described as cooperative DNA stiffening. Such cooperative binding predicts formation of patches of rigid H-NS nucleoprotein filaments, which was recently confirmed by AFM imaging (23).

Do the DNA-binding modes of H-NS account for the diversity of H-NS-mediated complex organization of large DNA molecules? The long, extended H-NS nucleoprotein filaments observed at low salt (50 mM KCl and <2 mM MgCl2) by Liu et al. (23) can be explained by cooperative DNA binding of H-NS over long lengths of DNA. Formation of large DNA hairpins (22,23), often referred to as DNA bridging, can also be formed when an H-NS filament interacts with naked DNA segments. This type of interaction would be expected to result in hairpin-like complexes or more complex structures (see Fig. 2, in particular Fig. 2 C). Such filament-mediated DNA bridging and higher-order complexes have been reported in several H-NS-family proteins, such as E. coli StpA, Pseudomonas aeruginosa MvaT and MvaU, and Mycobacterium tuberculosis Lsr2 (39–42).

Figure 2.

Figure 2

Open in a new tab

H-NS-filament-mediated DNA organization. (A) In 200 mM KCl in the absence of MgCl2, H-NS forms small patches of filaments (magenta arrows). These filaments can associate with naked DNA segments to form DNA bridges (green arrows). (B and C) Schematics of H-NS-filament-mediated nucleoprotein complexes. (B) H-NS binds cooperatively on DNA to form patches of stiff filaments. (C) Under certain solution conditions (e.g., in 200 mM KCl), additional DNA binding sites on the outer surface of H-NS filaments can associate with naked DNA, leading to formation of DNA bridges. (D) Schematic of H-NS-dimer-mediated DNA bridging. Individual H-NS dimers with two DNA binding sites bring two remote DNA sites together to form a bridge. Formation of neighboring bridges is facilitated, leading to clusters of H-NS bridges.

To demonstrate that H-NS-stiffened filaments could also form DNA bridges, we imaged H-NS nucleoprotein complexes formed by Salmonella H-NS under nonsaturating binding conditions (Fig. 2). Salmonella H-NS is 95% identical to E. coli H-NS. We used 200 mM KCl in the absence of MgCl2, a condition where H-NS binding affinity is reduced due to electrostatic screening effects, enabling a more specific protein-DNA interaction. Salmonella H-NS bound to a 755 bp csgD-csgB intergenic regulatory region at a stoichiometry of 2:1 (protein monomer/DNA bp). The binding pattern showed dispersed, small patches of H-NS nucleoprotein filaments (Fig. 2 A, schematized in Fig. 2 B). In addition, a major portion of DNA segments remained uncoated with H-NS. Some of these filaments were associated with another DNA segment, resulting in DNA bridges (Fig. 2 A, green arrows). This experiment provides direct evidence that H-NS filaments can mediate formation of DNA bridges (see Fig. 2 C). This is in contrast to previous AFM imaging experiments performed in 50 mM KCl in the absence of MgCl2, where large patches of H-NS filaments were observed in the absence of DNA bridging (23). Together, these results suggest that DNA bridges can be mediated by stiffened H-NS filaments, depending on solution conditions, as schematized in Fig. 2 C. DNA bridging by H-NS at 200 mM KCl was previously undetected in magnetic tweezers experiments, likely because the large DNA molecule (48,502 bp, ∼16.3 μm) used in the single-DNA stretching assay was insensitive to a few weak bridges formed by small, dispersed H-NS nucleoprotein filaments (23). The free intracellular Mg2+ concentration in bacterial cells is in the low, 1- to 2-mM range (43–45), a condition where H-NS nucleoprotein filaments and bridged DNA coexist (23). DNA bridging observed under these conditions likely occurs by the filament-mediated mechanism described above.

Together, these findings suggest that the cooperative DNA-stiffening mode of H-NS mediates the organization of large DNA molecules into various conformations depending on the binding saturation level and various environmental conditions (23,33). Many other H-NS family members share this cooperative DNA-stiffening mode of binding across several bacterial species (39–42).

As discussed above, DNA bridging can be mediated by an H-NS filament (Fig. 2). In previous studies performed in 10 mM MgCl2, DNA bridges were formed upon H-NS binding, whereas DNA stiffening was not observed (22,23,37). Two possible scenarios may explain this observation. 1) In 10 mM MgCl2, H-NS forms dispersed short filaments, similar to those observed in 200 mM KCl (Fig. 2 A), which are not detected by magnetic tweezers. These dispersed short filaments would then mediate the formation of DNA bridges as schematized in Fig. 2 C. 2) Alternatively, binding of individual H-NS dimers, each providing two DNA binding sites, could bring two remote DNA sites together to form a bridge. Binding of a subsequent H-NS dimer adjacent to the bridge is energetically favored, because two DNA duplexes are brought closer by prior bridge formation, allowing the arriving H-NS dimer to easily engage with two DNA binding sites (see Fig. 2 D). A theoretical framework describing such a scenario has been developed (46), and it is supported by Monte Carlo simulations (47). Obviously, this mechanism also leads to cooperative binding of H-NS, as formation of neighboring bridges is facilitated. This cooperativity causes clustering of H-NS bridges and is DNA-induced. This is distinct from the cooperativity of H-NS filament formation that arises from direct protein-protein interaction. The latter mechanism seems more likely to be physiologically relevant, as the free magnesium ion concentration in E. coli is only 1–2 mM (43–45). At this concentration, nucleoprotein filament formation by H-NS predominates (23).

Sequence selectivity of H-NS binding

In vitro DNA binding assays have shown that H-NS binds to DNA in a largely nonspecific manner, although DNase I footprinting experiments identified certain high-affinity sequence motifs (20). A major function of H-NS is to silence laterally acquired genes in bacterial pathogens, which are typically AT-rich (48–50). This implies a certain sequence selectivity of H-NS binding in vivo. Further, it has been reported that short inserts of high-affinity sequences in a large low-affinity DNA segment could switch binding from low affinity to high affinity (48). How does the cooperative, DNA-stiffening binding mode of H-NS inform these observations?

Formation of nucleoprotein filaments through cooperative binding implies the need for a nucleation site to initiate a nucleation-and-growth process. High-affinity consensus DNA sequence motifs have been reported at some promoter regions of genes silenced by H-NS (20,49), and these motifs may serve as nucleation sites that dictate at which locations H-NS filaments can form. In vivo, such high-affinity nucleation sites may play a crucial role in selective gene silencing, localizing the filament around specific sequences in the area of the bacterial genome to be silenced. It may also enable H-NS to distinguish between native genes and genes acquired from external sources, which are generally AT-rich (48). In the absence of nucleation sites and cooperative binding, indiscriminate binding of H-NS would lead to unregulated silencing that might reduce bacterial fitness. Therefore, the nucleation-and-growth process that initiates filament formation requires a certain level of specificity. Formation of an H-NS filament via cooperative binding also suggests that short insertions of high-affinity sequences can increase the overall binding affinity of a long low-affinity DNA segment due to nucleation of H-NS at these sites, which can spread to nearby sequences through cooperative binding. In such a way, the cooperative, DNA-stiffening binding mode of H-NS provides a basis for understanding the sequence selectivity and the switching of low-affinity to high-affinity binding by H-NS. Identifying the sequences where DNA stiffening initiates will provide a greater understanding of the sequence requirements of nucleation sites.

Oligomerization of H-NS and its implications for H-NS cooperative binding

The ability of H-NS to form oligomers also influences its ability to form stiffened filaments. In single-molecule experiments, functionally defective substitutions in dimerization site 1 resulted in the loss of nucleoprotein filaments (51). Similarly, substitutions in dimerization site 2 also caused a loss of nucleoprotein filaments (our unpublished results). As the H-NS nucleoprotein filament is crucial for both H-NS-mediated gene silencing and DNA organization functions, these results highlight the importance of both dimerization domains in H-NS oligomerization. Oligomerization is also important in other H-NS family proteins. For example, in P. aeruginosa MvaT, substitutions in the N-terminal oligomerization domain cause loss of gene silencing and are unable to form nucleoprotein filaments (41).

A role for two H-NS oligomerization domains in the formation of a continuous H-NS nucleoprotein filament can be understood as follows. In a continuous H-NS nucleoprotein filament, an H-NS molecule on DNA is linked to two adjacent H-NS molecules, which means that at least two dimerization sites are required to form a chain of linked H-NS molecules on DNA. Therefore, higher-order oligomerization mediated by at least two dimerization sites seems crucial for the integrity of H-NS nucleoprotein filaments.

A recent structural study suggests that the head-to-head and tail-to-tail dimerization of H-NS by the two dimerization sites could lead to formation of a right-handed helical filament with a helical pitch of 28 nm and a diameter of 19 nm (16). A model was proposed based on this helical structure, where DNA follows the H-NS helical conformation (see Fig. 3 A). This model predicts that the axial length of the complex is significantly shortened to around 50% of the contour length of the naked DNA. However, recent single-DNA stretching experiments revealed that an H-NS-DNA complex has a similar apparent contour length to that of naked DNA (23,38). Therefore, one possible mechanism is that the H-NS filament is flexible, which allows deformation into a thinner and longer structure with increased helical pitch upon DNA binding, hence forming a cohelical structure with DNA (see Fig. 3 B). Alternatively, the H-NS filament is relatively rigid, but it wraps around DNA such that DNA is not deformed significantly (see Fig. 3 C). In these scenarios, the exposed C-terminal DNA binding sites on the outer surface of the helical filament would be able to interact with other DNA segments, allowing filament-mediated DNA bridging. These scenarios are depicted in Fig. 3.

Figure 3.

Figure 3

Open in a new tab

H-NS forms a right-handed helical nucleoprotein filament. An H-NS molecule has two dimerization domains (Fig. 1), which enable H-NS to form a chain of linked H-NS molecules on DNA. Three H-NS nucleoprotein filament structures are proposed. (A) DNA wraps around an H-NS helix, reducing the contour length to ∼50%. (B) A flexible H-NS nucleoprotein filament allows deformation into a thinner and longer structure. (C) H-NS wraps around the DNA in a helix, forming thick nucleoprotein filaments. The DNA is drawn to scale to highlight the change in contour length. Yellow circles (free DNA-binding domains) that are directed toward the outside of these helical filaments may allow interaction with naked DNA, hence forming filament-mediated DNA bridges under certain conditions.

H-NS filament formation leads to gene silencing

Previous experiments have revealed the vital role that H-NS plays in gene silencing. Questions remain, however, regarding how the binding mode of H-NS influences its gene-silencing ability. An initial mechanism of H-NS-mediated gene silencing was proposed based on the formation of DNA bridges. Bridges formed on promoter regions were reported, which were proposed to trap and immobilize RNA polymerase (52–54). In addition to this mechanism, DNA bridging might also stabilize plectonemes formed downstream of a transcribing RNAP (55,56), serving as a roadblock for further translocation.

An interesting question is whether cooperative filament formation, without formation of DNA bridges, could lead to gene silencing. Fig. 4 depicts two of several scenarios by which an H-NS filament could promote gene silencing. 1) A continuous H-NS filament formed on a promoter sequence may effectively block the accessibility of the promoter by RNA polymerase. Indeed, recent DNase I digestion experiments demonstrated that the accessibility of DNA was effectively blocked when filaments were formed by H-NS family proteins (StpA, MvaT, MvaU, and Lsr2) in the absence of bridging (39,40,42,57). 2) Formation of an H-NS filament downstream in the gene could act as a roadblock for RNAP translocation along DNA. RNAP can exert up to 25 pN of force during translocation (58), and the stiffened H-NS filament might prevent translocation. This possibility has not been tested and warrants further investigation.

Figure 4.

Figure 4

Open in a new tab

Mechanisms of H-NS-filament-dependent gene silencing. H-NS-mediated gene silencing can be achieved by inhibiting RNAP binding to the promoter region or by blocking RNAP translocation. This is possible by formation of an H-NS nucleoprotein filament in the promoter region or downstream of the promoter. The filament can associate with naked DNA to form DNA bridges, which may further restrict accessibility of RNAP to DNA.

Can H-NS filament-mediated DNA bridging play a role in gene silencing? If the bridging occurs at the promoter region, it may further restrict DNA accessibility (scenario 1 above). If it occurs downstream in the gene, bridged H-NS filaments may become a stronger roadblock for RNAP translocation compared to a stretch of H-NS filament alone (scenario 2 above). In addition, bridged filaments may also constrain DNA supercoiling (59), creating a closed topological domain and accumulating torsional stress during transcription, which promotes pausing for Rho-dependent termination (60). It is worth emphasizing that all of these strategies employ an H-NS filament.

The importance of filament-mediated silencing has been highlighted in recent studies reporting that functionally defective mutants often lead to a loss of filament formation (41,51). In addition, it was shown that SsrB, an H-NS antisilencing protein, effectively competes with H-NS binding under conditions where H-NS filament formation was predominant (61).

Importantly, the H-NS filament was found to be sensitive to various solution factors; lowering pH, temperature, KCl, or MgCl2 favors filament formation (23,33). The sensitivity to temperature is particularly interesting, as the formation of an H-NS filament was drastically decreased at 37°C (23,33). H-NS filament formation would be less favorable after bacterial cells invade humans, which may partly contribute to relief of H-NS-silenced virulence genes. This provides a mechanism by which H-NS might play a direct role in thermoregulation of gene expression.

Implications of H-NS filament formation for relief of gene silencing

H-NS-mediated gene silencing has to be tightly regulated, because genes that are repressed by H-NS need to be activated at some point for bacteria to survive and thrive and for pathogens to initiate their virulence pathways. Indeed, bacteria have evolved multiple mechanisms to counteract H-NS silencing (54). Some DNA-binding proteins, referred to as antisilencing proteins in this review, are able to antagonize H-NS silencing functions. The proposed mechanisms of antisilencing proteins mainly include competing for DNA binding sites and heteromeric protein interaction (54). Among the antisilencing proteins that antagonize H-NS by competing for DNA binding sites, only SsrB and Ler have well-characterized DNA binding properties (38,61). H-NS silencing can also be relieved by disrupting the oligomeric state of H-NS by heteromeric protein interaction, as reported for gp5.5 from phage T7 and a family of naturally occurring truncated H-NS derivatives (H-NST) (34,62,63). Here, we review these well-characterized antisilencing proteins, and discuss how H-NS filament formation influences their antisilencing activity (see Fig. 5).

Figure 5.

Figure 5

Open in a new tab

Antisilencing mechanisms based on H-NS filament formation. Shown at center is the H-NS bound filament, which is transcriptionally silent. Protein-dependent antisilencing can be achieved by 1) progressive displacement of terminal H-NS subunits, hence alleviating H-NS-mediated gene silencing (left), or 2) disrupting H-NS oligomerization through interaction with the N-terminal region of H-NS, thereby affecting its DNA-binding capability (right). The latter mechanism has been shown for naturally occurring truncated H-NS derivatives (H-NST), which lack the DNA-binding domain.

SsrB is a DNA-binding protein that regulates expression of virulence genes when Salmonella is in the macrophage vacuole. Normally, virulence genes are repressed by H-NS until the appropriate signals dictate the start of pathogenesis. Previous studies showed that SsrB was able to outcompete H-NS under conditions that favored H-NS filament formation, i.e., once SsrB bound to DNA, H-NS was unable to bind (61). In addition, SsrB was able to dissociate H-NS from preformed H-NS filaments (our unpublished data). This behavior can be understood based on the distinct DNA-binding properties of SsrB compared to H-NS. SsrB bends DNA (64) and can condense DNA via DNA juxtaposition (61), resulting in highly curved DNA conformations. Thus, we propose that one mechanism of antisilencing is via production of bent and juxtaposed DNA conformations that are energetically unfavorable for formation of a straight rigid H-NS nucleoprotein filament (61). DNA bending by SsrB can also promote slow, progressive disassembly of a preformed H-NS filament, most likely via displacing H-NS subunits from the ends. Therefore, the opposing DNA binding modes between SsrB (bending) and H-NS (filament formation) provide a molecular explanation of SsrB antisilencing activity. This may be a general mechanism shared by other antisilencing proteins and warrants additional studies of other DNA binding antisilencing proteins.

Another example of antisilencing by displacing H-NS from DNA comes from a recent study on Ler from enterohemorrhagic E. coli and enteropathogenic E. coli (38,65). Although Ler is described as an H-NS-family protein because of its sequence similarity to H-NS (primarily in the C-terminal domain), Ler cannot complement H-NS functions in an hns-deficient host. Therefore, this protein is better described as an H-NS-antagonizing protein than as an H-NS-family protein. The similarities in the C-terminal DNA-binding domains of Ler and H-NS may explain their overlapping binding sites at the locus-of-enterocyte-effacement promoter regions, and it also implies that antisilencing of H-NS-repressed genes by Ler likely occurs via competition for DNA-binding sites (66,67).

In terms of DNA binding, Ler binds to DNA noncooperatively, in contrast to the cooperative binding of H-NS (38). Thus, Ler, unlike H-NS, cannot form a continuous patch of nucleoprotein filament on DNA. On the other hand, Ler can robustly displace a preformed H-NS filament (38,65). How do the differences in DNA binding between Ler and H-NS explain the antisilencing function of Ler? A continuous H-NS filament formed by cooperative binding is expected to have a very low exchange rate between proteins inside the filament and proteins in solution. This is because such exchange requires breaking attractive interactions between adjacent N-terminal dimerization sites, in addition to forming protein-DNA interactions. Therefore, dissociation should primarily occur at the ends of filaments. In the case of Ler, because it binds noncooperatively, dissociation of Ler can occur throughout the DNA with similar probability, i.e., not just at the ends. By outcompeting H-NS in binding to DNA, Ler provides RNAP with easier access to the promoter region and is less of a barrier to translocation, both of which result in alleviation of H-NS-mediated gene silencing.

Some antisilencing proteins employ protein-protein interactions that alleviate silencing by directly interacting with the N-terminus of H-NS. Such interactions would be likely to prevent formation of H-NS filaments, which require oligomerization. An example of this mechanism was demonstrated by gp5.5 protein of phage T7, which tightly associated with H-NS and antagonized H-NS-mediated silencing of the proU operon and T7 promoters (62). The antisilencing function of gp5.5 was via a direct interaction of gp5.5 with dimerization site 2 within H-NS, and it resulted in the disruption of higher-order nucleoprotein complexes (34). This DNA-independent mechanism of antisilencing was also employed by H-NST, which acts as an H-NS antagonist by forming heteromeric complexes with H-NS (63). Thus, filament-mediated gene silencing by H-NS is overcome in antisilencing, either by competition in binding, direct protein-protein interaction, or bending DNA to drive off H-NS.

Conclusions

This article has focused on the physical basis underlying the function of H-NS proteins, which play a major role in regulating gene expression in bacteria. Such knowledge is made possible by recent advances in single-molecule experiments that unraveled the binding mode of H-NS and various H-NS family proteins. From these studies, the nucleoprotein filament has emerged as the general feature among gene-silencing proteins across several gram-negative bacterial species. Increasing evidence suggests that this nucleoprotein filament is crucial for H-NS-mediated gene silencing, and the filament also serves as the basis for organizing DNA into various conformations. Understanding the nucleoprotein filament also provides insights into mechanisms of antisilencing by various antisilencing proteins. In summary, the filament-based model elegantly describes the dual roles of H-NS in gene regulation and chromosomal DNA packaging.

Author Contributions

R.S.W., J.Y., and L.J.K. wrote the manuscript. R.S.W. performed the AFM experiment depicted in Fig. 2.

Acknowledgments

We are grateful to Cindy Zhang Bo and Wong Chun Xi of the Science Communications core in MBI for the artwork in Fig. 3.

This research was supported by VA 5IO1BX000372 (to L.J.K. at JBVAMC), the Ministry of Education of Singapore, National Research Foundation of Singapore Grant MOE 2013-T2-1-154 (to J.Y.), and the Research Centre of Excellence Grant in Mechanobiology from the Ministry of Education, Singapore (to L.J.K. and J.Y.).

Editor: Lois Pollack.

Contributor Information

Jie Yan, Email: phyyj@nus.edu.sg.

Linda J. Kenney, Email: kenneyl@uic.edu.

References

  • 1.Kavenoff R., Bowen B.C. Electron microscopy of membrane-free folded chromosomes from Escherichia coli. Chromosoma. 1976;59:89–101. doi: 10.1007/BF00328479. [DOI] [PubMed] [Google Scholar]
  • 2.Kavenoff R., Ryder O.A. Electron microscopy of membrane-associated folded chromosomes of Escherichia coli. Chromosoma. 1976;55:13–25. doi: 10.1007/BF00288323. [DOI] [PubMed] [Google Scholar]
  • 3.Delius H., Worcel A. Electron microscopic studies on the folded chromosome of Escherichia coli. Cold Spring Harb. Symp. Quant. Biol. 1974;38:53–58. doi: 10.1101/sqb.1974.038.01.008. [DOI] [PubMed] [Google Scholar]
  • 4.Vos S.M., Tretter E.M., Berger J.M. All tangled up: how cells direct, manage and exploit topoisomerase function. Nat. Rev. Mol. Cell Biol. 2011;12:827–841. doi: 10.1038/nrm3228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Stuger R., Woldringh C.L., Westerhoff H.V. DNA supercoiling by gyrase is linked to nucleoid compaction. Mol. Biol. Rep. 2002;29:79–82. doi: 10.1023/a:1020318705894. [DOI] [PubMed] [Google Scholar]
  • 6.Murphy L.D., Zimmerman S.B. Macromolecular crowding effects on the interaction of DNA with Escherichia coli DNA-binding proteins: a model for bacterial nucleoid stabilization. Biochim. Biophys. Acta. 1994;1219:277–284. doi: 10.1016/0167-4781(94)90049-3. [DOI] [PubMed] [Google Scholar]
  • 7.Ali Azam T., Iwata A., Ishihama A. Growth phase-dependent variation in protein composition of the Escherichia coli nucleoid. J. Bacteriol. 1999;181:6361–6370. doi: 10.1128/jb.181.20.6361-6370.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Azam T.A., Ishihama A. Twelve species of the nucleoid-associated protein from Escherichia coli. Sequence recognition specificity and DNA binding affinity. J. Biol. Chem. 1999;274:33105–33113. doi: 10.1074/jbc.274.46.33105. [DOI] [PubMed] [Google Scholar]
  • 9.Azam T.A., Hiraga S., Ishihama A. Two types of localization of the DNA-binding proteins within the Escherichia coli nucleoid. Genes Cells. 2000;5:613–626. doi: 10.1046/j.1365-2443.2000.00350.x. [DOI] [PubMed] [Google Scholar]
  • 10.Dorman C.J. H-NS: a universal regulator for a dynamic genome. Nat. Rev. Microbiol. 2004;2:391–400. doi: 10.1038/nrmicro883. [DOI] [PubMed] [Google Scholar]
  • 11.Hommais F., Krin E., Bertin P. Large-scale monitoring of pleiotropic regulation of gene expression by the prokaryotic nucleoid-associated protein, H-NS. Mol. Microbiol. 2001;40:20–36. doi: 10.1046/j.1365-2958.2001.02358.x. [DOI] [PubMed] [Google Scholar]
  • 12.Salmon K.A., Hung S.P., Gunsalus R.P. Global gene expression profiling in Escherichia coli K12: effects of oxygen availability and ArcA. J. Biol. Chem. 2005;280:15084–15096. doi: 10.1074/jbc.M414030200. [DOI] [PubMed] [Google Scholar]
  • 13.Prieto A.I., Kahramanoglou C., Luscombe N.M. Genomic analysis of DNA binding and gene regulation by homologous nucleoid-associated proteins IHF and HU in Escherichia coli K12. Nucleic Acids Res. 2012;40:3524–3537. doi: 10.1093/nar/gkr1236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Ali B.M., Amit R., Stavans J. Compaction of single DNA molecules induced by binding of integration host factor (IHF) Proc. Natl. Acad. Sci. USA. 2001;98:10658–10663. doi: 10.1073/pnas.181029198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Lin J., Chen H., Yan J. Physical organization of DNA by multiple non-specific DNA-binding modes of integration host factor (IHF) PLoS One. 2012;7:e49885. doi: 10.1371/journal.pone.0049885. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Arold S.T., Leonard P.G., Ladbury J.E. H-NS forms a superhelical protein scaffold for DNA condensation. Proc. Natl. Acad. Sci. USA. 2010;107:15728–15732. doi: 10.1073/pnas.1006966107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Renault M., García J., Pons M. Protein oligomers studied by solid-state NMR: the case of the full-length nucleoid-associated protein histone-like nucleoid structuring protein. FEBS J. 2013;280:2916–2928. doi: 10.1111/febs.12297. [DOI] [PubMed] [Google Scholar]
  • 18.Bloch V., Yang Y., Kochoyan M. The H-NS dimerization domain defines a new fold contributing to DNA recognition. Nat. Struct. Biol. 2003;10:212–218. doi: 10.1038/nsb904. [DOI] [PubMed] [Google Scholar]
  • 19.Esposito D., Petrovic A., Ladbury J.E. H-NS oligomerization domain structure reveals the mechanism for high order self-association of the intact protein. J. Mol. Biol. 2002;324:841–850. doi: 10.1016/s0022-2836(02)01141-5. [DOI] [PubMed] [Google Scholar]
  • 20.Bouffartigues E., Buckle M., Rimsky S. H-NS cooperative binding to high-affinity sites in a regulatory element results in transcriptional silencing. Nat. Struct. Mol. Biol. 2007;14:441–448. doi: 10.1038/nsmb1233. [DOI] [PubMed] [Google Scholar]
  • 21.Spurio R., Dürrenberger M., Gualerzi C.O. Lethal overproduction of the Escherichia coli nucleoid protein H-NS: ultramicroscopic and molecular autopsy. Mol. Gen. Genet. 1992;231:201–211. doi: 10.1007/BF00279792. [DOI] [PubMed] [Google Scholar]
  • 22.Dame R.T., Wyman C., Goosen N. H-NS mediated compaction of DNA visualised by atomic force microscopy. Nucleic Acids Res. 2000;28:3504–3510. doi: 10.1093/nar/28.18.3504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Liu Y., Chen H., Yan J. A divalent switch drives H-NS/DNA-binding conformations between stiffening and bridging modes. Genes Dev. 2010;24:339–344. doi: 10.1101/gad.1883510. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Ko M., Park C. H-NS-Dependent regulation of flagellar synthesis is mediated by a LysR family protein. J. Bacteriol. 2000;182:4670–4672. doi: 10.1128/jb.182.16.4670-4672.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Park H.S., Ostberg Y., Uhlin B.E. Novel role for a bacterial nucleoid protein in translation of mRNAs with suboptimal ribosome-binding sites. Genes Dev. 2010;24:1345–1350. doi: 10.1101/gad.576310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Smyth C.P., Lundbäck T., Ladbury J.E. Oligomerization of the chromatin-structuring protein H-NS. Mol. Microbiol. 2000;36:962–972. doi: 10.1046/j.1365-2958.2000.01917.x. [DOI] [PubMed] [Google Scholar]
  • 27.Stella S., Spurio R., Gualerzi C.O. Nature and mechanism of the in vivo oligomerization of nucleoid protein H-NS. EMBO J. 2005;24:2896–2905. doi: 10.1038/sj.emboj.7600754. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Badaut C., Williams R., Rimsky S. 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. doi: 10.1074/jbc.M206037200. [DOI] [PubMed] [Google Scholar]
  • 29.Johansson J., Eriksson S., Uhlin B.E. Heteromeric interactions among nucleoid-associated bacterial proteins: localization of StpA-stabilizing regions in H-NS of Escherichia coli. J. Bacteriol. 2001;183:2343–2347. doi: 10.1128/JB.183.7.2343-2347.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Spurio R., Falconi M., Gualerzi C.O. 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. doi: 10.1093/emboj/16.7.1795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Ueguchi C., Suzuki T., Mizuno T. Systematic mutational analysis revealing the functional domain organization of Escherichia coli nucleoid protein H-NS. J. Mol. Biol. 1996;263:149–162. doi: 10.1006/jmbi.1996.0566. [DOI] [PubMed] [Google Scholar]
  • 32.Stella S., Falconi M., Pon C.L. Environmental control of the in vivo oligomerization of nucleoid protein H-NS. J. Mol. Biol. 2006;355:169–174. doi: 10.1016/j.jmb.2005.10.034. [DOI] [PubMed] [Google Scholar]
  • 33.Amit R., Oppenheim A.B., Stavans J. Increased bending rigidity of single DNA molecules by H-NS, a temperature and osmolarity sensor. Biophys. J. 2003;84:2467–2473. doi: 10.1016/S0006-3495(03)75051-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Ali S.S., Beckett E., Navarre W.W. The 5.5 protein of phage T7 inhibits H-NS through interactions with the central oligomerization domain. J. Bacteriol. 2011;193:4881–4892. doi: 10.1128/JB.05198-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Williams R.M., 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. doi: 10.1128/jb.178.15.4335-4343.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Yan J., Marko J.F. Effects of DNA-distorting proteins on DNA elastic response. Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 2003;68:011905. doi: 10.1103/PhysRevE.68.011905. [DOI] [PubMed] [Google Scholar]
  • 37.Dame R.T., Noom M.C., Wuite G.J. Bacterial chromatin organization by H-NS protein unravelled using dual DNA manipulation. Nature. 2006;444:387–390. doi: 10.1038/nature05283. [DOI] [PubMed] [Google Scholar]
  • 38.Winardhi R.S., Gulvady R., Yan J. Locus of enterocyte effacement-encoded regulator (Ler) of pathogenic Escherichia coli competes off histone-like nucleoid-structuring protein (H-NS) through noncooperative DNA binding. J. Biol. Chem. 2014;289:13739–13750. doi: 10.1074/jbc.M113.545954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Lim C.J., Whang Y.R., Yan J. Gene silencing H-NS paralogue StpA forms a rigid protein filament along DNA that blocks DNA accessibility. Nucleic Acids Res. 2012;40:3316–3328. doi: 10.1093/nar/gkr1247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Winardhi R.S., Castang S., Yan J. Single-molecule study on histone-like nucleoid-structuring protein (H-NS) paralogue in Pseudomonas aeruginosa: MvaU bears DNA organization mode similarities to MvaT. PLoS One. 2014;9:e112246. doi: 10.1371/journal.pone.0112246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Winardhi R.S., Fu W., Yan J. Higher order oligomerization is required for H-NS family member MvaT to form gene-silencing nucleoprotein filament. Nucleic Acids Res. 2012;40:8942–8952. doi: 10.1093/nar/gks669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Qu Y., Lim C.J., Yan J. Mechanism of DNA organization by Mycobacterium tuberculosis protein Lsr2. Nucleic Acids Res. 2013;41:5263–5272. doi: 10.1093/nar/gkt249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Lusk J.E., Williams R.J., Kennedy E.P. Magnesium and the growth of Escherichia coli. J. Biol. Chem. 1968;243:2618–2624. [PubMed] [Google Scholar]
  • 44.Martin-Orozco N., Touret N., Grinstein S. Visualization of vacuolar acidification-induced transcription of genes of pathogens inside macrophages. Mol. Biol. Cell. 2006;17:498–510. doi: 10.1091/mbc.E04-12-1096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Alatossava T., Jütte H., Kellenberger E. Manipulation of intracellular magnesium content in polymyxin B nonapeptide-sensitized Escherichia coli by ionophore A23187. J. Bacteriol. 1985;162:413–419. doi: 10.1128/jb.162.1.413-419.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Wiggins P.A., Dame R.T., Wuite G.J. Protein-mediated molecular bridging: a key mechanism in biopolymer organization. Biophys. J. 2009;97:1997–2003. doi: 10.1016/j.bpj.2009.06.051. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.de Vries R. Influence of mobile DNA-protein-DNA bridges on DNA configurations: coarse-grained Monte Carlo simulations. J. Chem. Phys. 2011;135:125104. doi: 10.1063/1.3636383. [DOI] [PubMed] [Google Scholar]
  • 48.Navarre W.W., Porwollik S., Fang F.C. Selective silencing of foreign DNA with low GC content by the H-NS protein in Salmonella. Science. 2006;313:236–238. doi: 10.1126/science.1128794. [DOI] [PubMed] [Google Scholar]
  • 49.Lang B., Blot N., Travers A. High-affinity DNA binding sites for H-NS provide a molecular basis for selective silencing within proteobacterial genomes. Nucleic Acids Res. 2007;35:6330–6337. doi: 10.1093/nar/gkm712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Gordon B.R., Li Y., Liu J. Structural basis for recognition of AT-rich DNA by unrelated xenogeneic silencing proteins. Proc. Natl. Acad. Sci. USA. 2011;108:10690–10695. doi: 10.1073/pnas.1102544108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Lim C.J., Lee S.Y., Yan J. Nucleoprotein filament formation is the structural basis for bacterial protein H-NS gene silencing. Sci. Rep. 2012;2:509. doi: 10.1038/srep00509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Dame R.T., Wyman C., Goosen N. 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. doi: 10.1074/jbc.C100603200. [DOI] [PubMed] [Google Scholar]
  • 53.Dorman C.J., Kane K.A. DNA bridging and antibridging: a role for bacterial nucleoid-associated proteins in regulating the expression of laterally acquired genes. FEMS Microbiol. Rev. 2009;33:587–592. doi: 10.1111/j.1574-6976.2008.00155.x. [DOI] [PubMed] [Google Scholar]
  • 54.Stoebel D.M., Free A., Dorman C.J. Anti-silencing: overcoming H-NS-mediated repression of transcription in Gram-negative enteric bacteria. Microbiology. 2008;154:2533–2545. doi: 10.1099/mic.0.2008/020693-0. [DOI] [PubMed] [Google Scholar]
  • 55.Liu L.F., Wang J.C. Supercoiling of the DNA template during transcription. Proc. Natl. Acad. Sci. USA. 1987;84:7024–7027. doi: 10.1073/pnas.84.20.7024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Ma J., Bai L., Wang M.D. Transcription under torsion. Science. 2013;340:1580–1583. doi: 10.1126/science.1235441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Winardhi R.S. National University of Singapore; Singapore: 2014. Single-molecule studies on the role of nucleoid-associated proteins in bacterial chromatin. PhD thesis. [Google Scholar]
  • 58.Wang M.D., Schnitzer M.J., Block S.M. Force and velocity measured for single molecules of RNA polymerase. Science. 1998;282:902–907. doi: 10.1126/science.282.5390.902. [DOI] [PubMed] [Google Scholar]
  • 59.Lim C.J., Kenney L.J., Yan J. Single-molecule studies on the mechanical interplay between DNA supercoiling and H-NS DNA architectural properties. Nucleic Acids Res. 2014;42:8369–8378. doi: 10.1093/nar/gku566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Kotlajich M.V., Hron D.R., Landick R. Bridged filaments of histone-like nucleoid structuring protein pause RNA polymerase and aid termination in bacteria. eLife. 2015;4:e04970. doi: 10.7554/eLife.04970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Walthers D., Li Y., Kenney L.J. Salmonella enterica response regulator SsrB relieves H-NS silencing by displacing H-NS bound in polymerization mode and directly activates transcription. J. Biol. Chem. 2011;286:1895–1902. doi: 10.1074/jbc.M110.164962. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Liu Q., Richardson C.C. Gene 5.5 protein of bacteriophage T7 inhibits the nucleoid protein H-NS of Escherichia coli. Proc. Natl. Acad. Sci. USA. 1993;90:1761–1765. doi: 10.1073/pnas.90.5.1761. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Williamson H.S., Free A. A truncated H-NS-like protein from enteropathogenic Escherichia coli acts as an H-NS antagonist. Mol. Microbiol. 2005;55:808–827. doi: 10.1111/j.1365-2958.2004.04421.x. [DOI] [PubMed] [Google Scholar]
  • 64.Carroll R.K., Liao X., Kenney L.J. Structural and functional analysis of the C-terminal DNA binding domain of the Salmonella typhimurium SPI-2 response regulator SsrB. J. Biol. Chem. 2009;284:12008–12019. doi: 10.1074/jbc.M806261200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Mellies J.L., Benison G., Larabee F.J. Ler of pathogenic Escherichia coli forms toroidal protein-DNA complexes. Microbiology. 2011;157:1123–1133. doi: 10.1099/mic.0.046094-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Sperandio V., Mellies J.L., Kaper J.B. Activation of enteropathogenic Escherichia coli (EPEC) LEE2 and LEE3 operons by Ler. Mol. Microbiol. 2000;38:781–793. doi: 10.1046/j.1365-2958.2000.02168.x. [DOI] [PubMed] [Google Scholar]
  • 67.Mellies J.L., Barron A.M., Carmona A.M. Enteropathogenic and enterohemorrhagic Escherichia coli virulence gene regulation. Infect. Immun. 2007;75:4199–4210. doi: 10.1128/IAI.01927-06. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Biophysical Journal are provided here courtesy of The Biophysical Society

RESOURCES