NMR assignments for the insertion domain of bacteriophage Sf6 coat protein

The P22 bacteriophage group is a subgroup of the λ phage supercluster, comprised of the three major sequence types Sf6, P22, and CUS-3, based on their capsid proteins. Our goal is to investigate the extent to which structure–function relationships are conserved for the viral coat proteins and I-domains in this subgroup. Sf6 is a phage that infects the human pathogen Shigella flexneri. The coat protein of Sf6 assembles into a procapsid, which further undergoes maturation during DNA packaging into an infectious virion. The Sf6 coat protein contains a genetically inserted domain, termed the I-domain, similar to the ones present in the P22 and CUS-3 coat proteins. Based on the P22 example, I-domains play important functional roles in capsid assembly, stability, viability, and size-determination. Here we report the ¹H, ¹⁵N, and ¹³C chemical shift assignments for the I-domain of the Sf6 phage coat protein. Chemical shift-based secondary structure prediction and hydrogen-bond patterns from a long-range HNCO experiment indicate that the Sf6 I-domain adopts a 6-stranded β-barrel fold like those of P22 and CUS-3 but with important differences, including the absence of the D-loop that is critical for capsid assembly and the addition of a novel disordered loop region.     

NMR assignments for the cis and trans forms of the hemolysin II C-terminal domain

Pathogenic bacteria secrete pore-forming toxins (PFTs) to selectively defend against immune cells and to break through cellular barriers in the host. Understanding how PFTs attack cell membranes is not only essential for therapeutic intervention but for designing agents to deliver drugs to specific human cell subtypes, for example in anti-cancer or anti-viral therapies. Many toxins contain accessory domains that help recognize specific molecular epitopes on the membranes of target cells, including proteins, carbohydrates, and lipids. Here we report NMR assignments for the 94-residue 10 kDa C-terminal accessory domain of Bacillus cereus hemolysin II, HlyIIC, that has no known structural or functional homologues. The HlyIIC domain exists in a dynamic equilibrium due to cis/trans isomerization of its Gly86–Pro87 peptide bond. The cis and trans forms are about equally populated and are in slow exchange on the NMR timescale, giving rise to separate signals for approximately half of the residues in the domain. Assignments for the cis and trans forms were achieved with the aid of a P87M mutant that stabilizes the trans form, and separate sequential walks for the two forms in 3D NMR spectra of the wild-type HlyIIC. Based on backbone chemical shifts, the domain has a α1–α2–β1–β2–β3–β4–α3–β5 order of secondary structure elements. The last strand in the trans form and in the P87M mutant is shortened near Pro87 compared to the cis form. Both cis/trans isomerization and the P87M mutation cause large chemical shift changes throughout HlyIIC, suggesting that the proline is important in stabilizing the structure of the domain. The NMR assignments pave the way for solving the structures of the multiple conformational forms of HlyIIC and establishing their mechanism of interconversion.     

The role of pili in the interactions of pathogenic Neisseria with cultured human endothelial cells

The influence of the two surface structures of Neisseria meningitidis, capsule and pili, in bacterial interactions with human endothelial cells was investigated. Increased association correlated with the presence of pili on bacteria while capsule type had no apparent effect. Strains expressing both Class I and Class II pili associated with endothelial cells in significantly larger numbers compared with the non-piliated variants of the same strains (greater than 10x). Variants of Neisseria gonorrhoeae strain P9 expressing antigenically distinct pili also associated with endothelial cells in larger numbers (greater than 30x) compared with the non-piliated variant. Electron microscopic studies confirmed these data and showed that gonococci were internalized more frequently compared with meningococci. One consequence of increased association was an increase in the cytopathic effect of bacteria on the target cells.     

Mechanism of Protein Denaturation: Partial Unfolding of the P22 Coat Protein I-Domain by Urea Binding

The I-domain is an insertion domain of the bacteriophage P22 coat protein that drives rapid folding and accounts for over half of the stability of the full-length protein. We sought to determine the role of hydrogen bonds (H-bonds) in the unfolding of the I-domain by examining ³J_NC’ couplings transmitted through H-bonds, the temperature and urea-concentration dependence of ¹HN and ¹⁵N chemical shifts, and native-state hydrogen exchange at urea concentrations where the domain is predominantly folded. The native-state hydrogen-exchange data suggest that the six-stranded β-barrel core of the I-domain is more stable against unfolding than a smaller subdomain comprised of a short α-helix and three-stranded β-sheet. H-bonds, separately determined from solvent protection and ³J_NC’ H-bond couplings, are identified with an accuracy of 90% by ¹HN temperature coefficients. The accuracy is improved to 95% when ¹⁵N temperature coefficients are also included. In contrast, the urea dependence of ¹HN and ¹⁵N chemical shifts is unrelated to H-bonding. The protein segments with the largest chemical-shift changes in the presence of urea show curved or sigmoidal titration curves suggestive of direct urea binding. Nuclear Overhauser effects to urea for these segments are also consistent with specific urea-binding sites in the I-domain. Taken together, the results support a mechanism of urea unfolding in which denaturant binds to distinct sites in the I-domain. Disordered segments bind urea more readily than regions in stable secondary structure. The locations of the putative urea-binding sites correlate with the lower stability of the structure against solvent exchange, suggesting that partial unfolding of the structure is related to urea accessibility.