Hfq Protein (PDB ID: 3QSU) from Staphylococcus aureus
Created by: Michael Huber
Hfq protein from the gram-positive Staphylococcus aureus (PDB ID: 3QSU) is a global post-transcriptional regulator that plays a key role in gene expression. Hfq is required for the fitness and virulence of both gram-positive and gram-negative pathogens (1). Protein Hfq is highly conserved and highly abundant, as there are approximately 30,000 to 60,000 copies per bacterial cell (2). Hfq protein forms stable sRNA-mRNA complexes by binding AU-rich and A-tract RNA sequences that are resistant to degradation. The stability of the sRNA-mRNA complex, facilitated by Hfq, is crucial for a cell’s ability to adapt to the environment and to stress signals. Deletion of the Hfq protein in bacterial cells leads to several pleiotropic effects such as growth defects, sensitivity to UV light, increased cell length, and impaired virulence in numerous pathogens. Despite the crucial role that Hfq protein plays in translational processes, the molecular details of how the protein functions remain unknown (3).
Protein Hfq is composed of 77 amino acids, has a molecular weight of 8,778.7 daltons, and exhibits an isoelectric point of 4.69 indicating an overall acidic charcter (4). Crystallization of an Hfq-A7 RNA complex was performed using the hanging-drop vapor diffusion method at room temperature. Protein Hfq is an asymmetric unit containing 14 protomers, which form two and one-third hexamers. Each hexamer is further divided into six subunits, capable of binding one RNA fragment to the distal face. Each of the 14 protomers share a similar secondary structure comprised of one alpha helix and five beta-sheets, and the remaining 11 or 12 residues on each strand are disordered. Because there are no meaningful structural changes upon RNA binding, the final 11 or 12 residues on each strand are believed to be relatively unimportant in regards to Hfq function. The solvent exposed amino acid side chains of residues His-53 and His-58, present on the proximal face, coordinate Zn2+ ions required for crystallization (3).
Hfq protein has two non-exclusive models of function in riboregulation: the first as a chaperone, and the second as a eukaryotic poly(A)-binding protein. As a chaperone, Hfq acts to alter the conformation of sRNA and mRNA by partially unfolding the ribonucleotides to allow hybridization. Hfq can also act as a poly(A)-binding protein, and functions by forming a stable sRNA-mRNA complex (3).
PSI-BLAST and the Dali server were used to reveal proteins that share similar primary and tertiary structure with S. aureus Hfq. The PSI-BLAST program is a tool for searching protein databases for sequence similarities (5). PSI-BLAST assigns a significant E-value to subjects that have sequence homology to the query protein, Hfq. The E-value is calculated by overlapping the sequence of Hfq to other proteins and assigning differences in amino acids, or gaps. Sequence homology reduces the magnitude of the E-value, while gaps increase the E-value. E-values less than 0.05 are significant for proteins. Secondly, the Dali server compares three-dimensional protein structures to reveal probable functional similarities (6). Specifically, Dali uses a sum-of-pairs method to compare intramolecular distances (7). Subjects are assigned Z-scores depending upon the degree of tertiary structure similarity; a Z-score greater than 2 is significant and suggests that the subject shares a similar fold with the query protein.
Investigations of Hfq function and mechanism have focused mainly on Hfq present in gram-negative bacteria, such as Escherichia coli (E. coli). Protein Hfq in Staphylococcus aureus (S. aureus) and Protein Hfq in E. coli (PDB ID: 1HK9), share a Z-score of 12.5 and an E-score of 1e-14 (7, 8). Furthermore, both S. aureus and E. coli Hfq are composed of repeating hexameric units. In E. coli, the pivotal role Hfq plays in riboregulation and stress adaptation is evidenced by several deleterious effects in response to an Hfq mutation (3). In gram-negative bacteria, Hfq functions by stabilizing sRNA regulators against degradation in response to certain stress or physiological conditions (9).
Hfq protein function in gram-positive bacteria differs markedly from Hfq function in gram-negative bacteria. Whereas Hfq function in gram-negative bacteria is crucial in a cell’s response to stress, Hfq function in gram-positive bacteria is strain dependent. For example, Hfq protein in S. aureus is typically not involved in resistance to antibiotics and is not implicated in metabolic pathways (10). However, Hfq protein in methicillin-resistant S. aureus (MRSA) contributes significantly to stress resistance and pathogenicity (3). Many of the functional differences between S. aureus Hfq and E. coli Hfq may be attributed to structural differences between the two proteins. Specifically, E. coli Hfq is only 68 amino acids in length and exhibits an isoelectric point of 9.99, as compared to the 77 amino acids found in S. aureus Hfq and the more acidic isoelectric point of 4.69 (4).
Several differences exist between Hfq-RNA binding in E. coli and S. aureus. Notably, gram-positive Hfq proteins have a key mutation within the A-site binding pocket that replaces Ile-30 with Phe-30. The bulky aromatic side-chain inhibits the binding of a ribonucleotide due to sterics. A second key mutation found in gram-positive Hfq is the insertion of Gly-50, an extra residue in the loop connecting β3 to β4. This extra residue causes the loop to protrude 2.4 Å further into the A-site binding pocket, thereby further precluding nucleotide binding. As a consequence of these structural constraints, the adenosine nucleotide in gram-positive bacteria is unable to interact with the same binding site found in gram-negative bacteria; therefore, a new binding site is created on the distal face of S. aureus Hfq (3).
The new binding site of S. aureus Hfq employs a bipartite-binding (R-L) motif. The R-site is a purine nucleotide-binding site and the L-site is an R-site linker. The six (R-L) motifs of S. aureus Hfq provide the capacity to optimally bind 12 ribonucleotides to the distal face. The (R-L) binding motif does not allow the ribonucleotide to have any exposed nucleotides directed towards the solvent, thereby increasing stability. The conserved Phe-30/Tyr-30 among gram-positive bacteria suggests that the bipartite (R-L) bindng motif functions throughout the gram-positive bacteria domain (3).
Compared to the binding site of Hfq in E. coli, the rotated binding site of Hfq in S. aureus strongly prefers adenosine, is more aromatic, and permits deeper insertion of the adenine ring (3). The binding pocket of the S. aureus Hfq-RNA complex is formed between β2 and β2’strands of neighboring subunits, where the prime indicates location on a neighboring subunit. The adenine base inserts into this highly aromatic pocket and stacks against the side chains of residues Phe-25, Phe-30', and Phe-26'. The distinctive aromatic character of the binding site of S. aureus Hfq favors a downward rotation of the adenosine so that the base sticks deeper into the cleft. In addition to these stacking interactions, a hydrogen bond network exists between Asn-28, Gly-29, the sugar O4’ and the adenine N3 that act to further strengthen ribonucleotide binding. The added hydrogen bonds increase S. aureus Hfq protein's preference for A-rich RNA over DNA. Additional stabilizing hydrogen bonds are formed between the N1 and N6 atoms of the adenine and the hydroxyl groups of Ser-61 and Thr-62. Overall, the R-site of S. aureus Hfq makes a large number of contacts to bound adenosine, which contributes to the overall stability of the Hfq-RNA complex (3).
Most of the sRNAs present in S. aureus include a C-rich motif that is critical for association with their target mRNAs. Interestingly, one of these sRNAs, coined RsaE, was found to be conserved in another species of gram-positive bacterium, Bacillus subtilis (B. subtilis). RsaE present in B. subtilis is predicted to regulate expression of metabolic genes through the C-rich motif (8). Protein Hfq in S. aureus and Protein Hfq in B. subtilis (PDB ID: 3AHU), share a Z-score of 13.1, thereby suggesting a conserved tertiary structure (7). Binding studies on S. aureus Hfq strongly point to a conserved distal-face-binding mechanism among gram-positive bacteria. The crystal structure of B. subtilis reveals a highly similar mode of A-tract RNA binding to that used by S. aureus Hfq (11). Additionally, B. subtilis Hfq and S. aureus Hfq share a similar primary structure of 78 amino acids. However, because S. aureus Hfq does not play a role in the stress response, or in RNA stability, the functional similarities among gram-positive bacteria may not be as related as predicted by structural similarity (9).
The R-site-mechanisms of S. aureus Hfq and B. subtilis Hfq do differ in several important details. First, S. aureus Hfq is capable of forming more hydrophobic interactions with the bound adenosine at residues Phe-26’, Leu-27’ and Met-32’. Furthermore, S. aureus Hfq is able to form a crucial hydrogen bond between Asn-28’ and the bound adenosine that is absent in B. subtilis Hfq. By contrast, B. subtilis is capable of forming a hydrogen bond between Gln-30 and the 3’-hydroxyl group of the ribose. S. aureus Hfq is incapable of forming this additional hydrogen bond because its equivalent residue, Gln-31is 3.5 Å from the 3’-hydroxyl group of the ribose (3).
The functional consequence of the different mechanisms on the distal face of Hfq homologues from gram-positive bacteria have yet to be elucidated, but is likely to continue to be an area of active investigation. Although sRNA-mRNA interactions in S. aureus are decisive for gene expression regulation, they do not require the RNA-chaperone protein Hfq. These interactions may possibly require an RNA-chaperone protein other than Hfq, which remains to be found (10).