P′2, and P′3 are occupied while P2 and P′1 are not; all three IHF‐binding sites are occupied but the XIS and FIS sites are vacant. During excision, Int sites P1, P2, P′1, and P′2 are occupied, as are IHF sites H2 and H′ and all of the Xis and FIS sites; IHF site H1 and Int site P3 are vacant.
Figure 1.18 The interactions of the nucleoid‐associated proteins FIS, HU, and IHF with DNA. The FIS protein is homodimeric, while HU and IHF are heterodimeric. HU and IHF are closely related at the level of amino acid sequence and the alpha and beta subunits of each protein are similar in sequence and secondary structure. The tertiary structures of HU and IHF are also quite similar, as their modes of binding to DNA. Each inserts a beta sheet from each subunit into the minor groove of its DNA target. IHF differs from HU in having a strict nucleotide sequence requirement for binding and in making more contacts with the DNA at its binding site. IHF also induces a much greater bend angle in the DNA. The bends on either side of the HU‐DNA complex are non‐coplanar. FIS binds in the major groove of DNA using an alpha helix, one of four alpha helices found in each FIS monomer. The protein uses an induced fit binding mechanism that compresses the minor groove lying between the two sites of insertion of the alpha helices in the major groove. FIS binds to a variety of sites with differing binding affinities; a Logo has been assembled that summarises the chief sequence characteristics of the highest affinity sites (see Stella et al. 2010).
1.36 HU, a NAP with General DNA‐binding Activity
The HU protein from E. coli is the founder member of a superfamily of related NAPs found throughout the prokaryotic world and beyond (Burroughs et al. 2017). HU interacts with DNA in the minor groove and this encourages the bound DNA to follow a looped path (Figure 1.18). This property helps HU to overcome the resistance of the DNA to loop formation by overcoming DNA's intrinsic stiffness (Johnson et al. 1986). HU‐assisted loop formation contributes to the formation of nucleoprotein complexes involved in the control of transcription and site‐specific recombination (Haykinson and Johnson 1993; Semsey et al. 2004). It also has RNA‐binding activity, enabling it to influence translation (Balandina et al. 2001).
Each HU subunit inserts a beta sheet with an apical proline amino acid into the minor groove of the DNA at the binding site, inducing the DNA to bend (Figure 1.18). The bend angle is typically in the range of 105°–140° and bends are not coplanar, having a dihedral angle that is consistent with the path taken in negatively supercoiled DNA (Swinger et al. 2003). The flexibility in the bend angle, coupled with the absence of a strict nucleotide sequence for DNA binding, may allow HU to participate as an architectural component in a wide variety of DNA‐based transactions.
The α and β subunits of HU are encoded, respectively, by the hupA and hupB genes, located at distinct positions on the chromosome: hupA hupB double mutants that fail to express the HU protein display a filamentous cell phenotype because of disruption of the cell cycle due to the arrest of DNA replication (Dri et al. 1991). HU interacts with the DnaA protein at oriC where it stimulates formation of the initiation complex in chromosome replication. Like IHF, HU is usually heterodimeric and is composed of an alpha and a beta subunit. The alpha subunit seems to have the primary responsibility for interacting with DnaA. This preference for the alpha subunit may facilitate enhanced HU–DnaA interaction at early stages of growth when an HU α2 homodimer predominates rather than the αβ heterodimer (Chodavarapu et al. 2008). HU can also influence chromosome replication initiation indirectly by repressing the expression of the gene that encodes SeqA (Lee, H., et al. 2001), the protein that sequesters oriC and excludes DnaA (Han et al. 2003; Slater et al. 1995; von Freiesleben et al. 1994).
The HU protein can form nucleosome‐like structures in E. coli that are dependent on the local HU‐to‐DNA ratio (Sagi et al. 2004). It has been described as insulating transcription units on the chromosome by preventing changes in DNA supercoiling caused by transcription in one unit from influencing an adjacent one (Berger et al. 2016). HU may be particularly important for the maintenance of DNA supercoiling levels in the Ter macrodomain as the bacterium enters the stationary phase of the growth cycle (Lal et al. 2016). It has been reported to induce, together with FIS, weak and transient domain boundaries around the E. coli chromosome (Wu et al. 2019).
In laboratory‐grown cultures, the subunit composition of the HU protein changes as a function of growth phase: In lag phase, as the bacterium adapts to its new environment, the α2 form of HU occurs; in exponential growth the αβ form predominates and the β2 form is detected as the culture enters stationary phase (Claret and Rouvière ‐Yaniv 1997). The changing subunit composition of HU and the different DNA interaction properties of the distinct HU forms may contribute to processes that differentially compact the chromosome in the nucleoid and affect gene expression patterns (Hammel et al. 2016). Transcriptomic studies in Salmonella have shown that each form of the HU protein seems to govern a distinct group of genes, with overlaps between the three sub‐regulons (Mangan et al. 2011).
1.37 The Very Versatile FIS Protein
FIS is the Factor for Inversion Stimulation, so called because it was discovered originally as an important architectural element in the DNA inversion mechanisms responsible for the phase‐variable expression of flagella in Salmonella (Johnson et al. 1986) and of tail fibre proteins in bacteriophage Mu (Koch and Kahmann 1986). FIS is now known to contribute to a wide range of molecular events in bacteria, including DNA replication (Cassler et al. 1995; Filutowisz et al. 1992; Gille et al. 1991), site‐specific recombination (Dhar et al. 2009; McLean et al. 2013), transposition (Weinreich and Reznikoff 1992), transcription regulation (Grainger et al. 2008; Hirvonen et al. 2001; Kelly et al. 2004; Pemberton et al. 2002), bacteriophage life cycles (Betermier et al. 1993; van Drunen et al. 1993; Papagiannis et al. 2007; Seah et al. 2014), illegitimate recombination (Shanado et al. 1997), and chromosome domain boundary formation (Hardy and Cozzarelli 2005; Wu et al. 2019).
1.38 FIS and the Early Exponential Phase of Growth
FIS is a homodimeric NAP that is encoded by the second gene in the dusB‐fis operon and shows strong homology to the DNA‐binding domain of the NtrC transcription factor (Bishop et al. 2002; Morett and Bork 1998). Transcription of the fis gene is maximal in the early stages of exponential growth and FIS plays an important role in boosting the expression of genes that encode components of the translational machinery of the cell (Appleman et al. 1998; Ball and Johnson 1991; Hirvonen et al. 2001; Osuna et al. 1995). FIS binds to the major groove of the DNA using a helix‐turn‐helix (HTH) motif that interacts with A+T‐rich sites that match a weak consensus sequence (Hancock et al. 2016). The protein uses an induced fit binding mechanism that compresses the minor groove between those parts of the major groove that accommodate the HTH motifs of the two subunits (Figure 1.18) (Hancock et al. 2016; Stella et al. 2010). This creates a bend in the DNA of 65° according to FIS‐DNA co‐crystal structure data (Stella et al. 2010) with bends of up to 90° also being reported (Kostrewa et al. 1992; Pan et al. 1996).
Transcription factors that introduce bends into DNA can facilitate additional contacts between DNA (including proteins bound to that DNA) located upstream of the promoter and bound RNA polymerase, increasing the efficiency of transcription initiation (Huo et al. 2006; Rivetti et al. 1999; Verbeek et al. 1991). FIS acts as a ‘conventional’ transcription factor at some promoters, making protein–protein contacts with RNA polymerase (Bokal et al. 1997) and its DNA‐bending activity has the potential to enhance the efficiency
