Structure Guided Design of Bacteriophage Qβ Mutants as Next Generation Carriers for Conjugate Vaccines

Vaccines are critical tools to treat and prevent diseases. For an effective conjugate vaccine, the carrier is crucial, but few carriers are available for clinical applications. In addition, a drawback of current protein carriers is that high levels of antibodies against the carrier are induced by the conjugate vaccine, which are known to interfere with the immune responses against the target antigen. To overcome these challenges, we obtained the near atomic resolution crystal structure of an emerging protein carrier, i.e., the bacteriophage Qβ virus like particle. On the basis of the detailed structural information, novel mutants of bacteriophage Qβ (mQβ) have been designed, which upon conjugation with tumor associated carbohydrate antigens (TACAs), a class of important tumor antigens, elicited powerful anti-TACA IgG responses and yet produced lower levels of anticarrier antibodies as compared to those from the wild type Qβ-TACA conjugates. In a therapeutic model against an aggressive breast cancer in mice, 100% unimmunized mice succumbed to tumors in just 12 days even with chemotherapy. In contrast, 80% of mice immunized with the mQβ-TACA conjugate were completely free from tumors. Besides TACAs, to aid in the development of vaccines to protect against COVID-19, the mQβ based conjugate vaccine has been shown to induce high levels of IgG antibodies against peptide antigens from the SARS-CoV-2 virus, demonstrating its generality. Thus, mQβ is a promising next-generation carrier platform for conjugate vaccines, and structure-based rational design is a powerful strategy to develop new vaccine carriers.

Synthetic and immunological studies of Salmonella Enteritidis O-antigen tetrasaccharides as potential anti-Salmonella vaccines

The first synthetic carbohydrate based potential anti-Salmonella Enteritidis vaccine has been developed by conjugating a synthetic tetrasaccharide antigen with bacteriophage Qβ. High levels of specific and long lasting anti-glycan IgG antibodies were induced by the conjugate, which completely protected mice from lethal bacterial challenges in a passive transfer model.

Biodegradable Viral Nanoparticle/Polymer Implants Prepared via Melt-Processing

Viral nanoparticles have been utilized as a platform for vaccine development and are a versatile system for the display of antigenic epitopes for a variety of disease states. However, the induction of a clinically relevant immune response often requires multiple injections over an extended period of time, limiting patient compliance. Polymeric systems to deliver proteinaceous materials have been extensively researched to provide sustained release, which would limit administration to a single dose. Melt-processing is an emerging manufacturing method that has been utilized to create polymeric materials laden with proteins as an alternative to typical solvent-based production methods. Melt-processing is advantageous because it is continuous, solvent-free, and 100% of the therapeutic protein is encapsulated. In this study, we utilized melt-encapsulation to fabricate viral nanoparticle laden polymeric materials that effectively deliver intact particles and generate carrier specific antibodies in vivo. The effects of initial processing and postprocessing on particle integrity and aggregation were studied to develop processing windows for scale-up and the creation of more complex materials. The dispersion of particles within the PLGA matrix was studied, and the effect of additives and loading level on the release profile was determined. Overall, melt-encapsulation was found to be an effective method to produce composite materials that can deliver viral nanoparticles over an extended period and elicit an immune response comparable to typical administration schedules.

Dual Functionalized Bacteriophage Qβ as a Photocaged Drug Carrier

Proteinatious nanoparticles are emerging as promising materials in biomedical research owing to their many unique properties and our interest focuses on integrating environmental responsivity into these systems. In this work, the use of a virus-like particle (VLP) derived from bacteriophage Qβ as a photocaged drug delivery system is investigated. Ideally, a photocaged nanoparticle platform should be harmless and inert without activation by light yet, upon photoirradiation, should cause cell death. Approximately 530 photocleavable doxorubicin complexes are installed initially onto the surface of Qβ by CuAAC reaction for photocaging therapy; however, aggregation and precipitation are found to cause cell death at higher concentrations. In order to improve solution stability, thiol-dibromomaleimide chemistry has been developed to orthogonally modify the VLP. This chemistry provides a robust method of incorporating additional functionality at the disulfides on Qβ, which was used to increase the stability and solubility of the drug-loaded VLPs. As a result, the dual functionalied VLPs with polyethylene glycol and photocaged doxorubicin show not only negligible cytotoxicity before photoactivation but also highly controllable photorelease and cell killing power.

Changes in protein domains outside the catalytic site of the bacteriophage Qβ replicase reduce the mutagenic effect of 5-azacytidine

The high genetic heterogeneity and great adaptability of RNA viruses are ultimately caused by the low replication fidelity of their polymerases. However, single amino acid substitutions that modify replication fidelity can evolve in response to mutagenic treatments with nucleoside analogues. Here, we investigated how two independent mutants of the bacteriophage Qβ replicase (Thr210Ala and Tyr410His) reduce sensitivity to the nucleoside analogue 5-azacytidine (AZC). Despite being located outside the catalytic site, both mutants reduced the mutation frequency in the presence of the drug. However, they did not modify the type of AZC-induced substitutions, which was mediated mainly by ambiguous base pairing of the analogue with purines. Furthermore, the Thr210Ala and Tyr410His substitutions had little or no effect on replication fidelity in untreated viruses. Also, both substitutions were costly in the absence of AZC or when the action of the drug was suppressed by adding an excess of natural pyrimidines (uridine or cytosine). Overall, the phenotypic properties of these two mutants were highly convergent, despite the mutations being located in different domains of the Qβ replicase. This suggests that treatment with a given nucleoside analogue tends to select for a unique functional response in the viral replicase.

IMPORTANCE

In the last years, artificial increase of the replication error rate has been proposed as an antiviral therapy. In this study, we investigated the mechanisms by which two substitutions in the Qβ replicase confer partial resistance to the mutagenic nucleoside analogue AZC. As opposed to previous work with animal viruses, where different mutations selected sequentially conferred nucleoside analogue resistance through different mechanisms, our results suggest that there are few or no alternative AZC resistance phenotypes in Qβ. Also, despite resistance mutations being highly costly in the absence of the drug, there was no sequential fixation of secondary mutations. Bacteriophage Qβ is the virus with the highest reported mutation rate, which should make it particularly sensitive to nucleoside analogue treatments, probably favoring resistance mutations even if they incur high costs. The results are also relevant for understanding the possible pathways by which fidelity of the replication machinery can be modified.

Engineered virus-like nanoparticle heparin antagonists

Virus nanoparticles provide a self-assembling, reproducible multivalent platform that can be chemically and genetically manipulated for the presentation of a wide array of epitopes. Presented herein are engineered bacteriophage Qβ nanoparticles that function as potent heparin antagonists. Three successful approaches have been used: 1) chemically appending poly-Arg peptides; 2) point mutations to Arg on the virus capsid; 3) incorporation of heparin-binding peptides displayed externally on the virus surface. Each approach generates particles with good heparin antagonist activity with none of the toxic side effects of protamine, the only drug currently FDA-approved for clinical use as a heparin antagonist.

Multivalent display and receptor-mediated endocytosis of transferrin on virus-like particles

The structurally regular and stable self-assembled capsids derived from viruses can be used as scaffolds for the display of multiple copies of cell- and tissue-targeting molecules and therapeutic agents in a convenient and well-defined manner. The human iron-transfer protein transferrin, a high affinity ligand for receptors upregulated in a variety of cancers, has been arrayed on the exterior surface of the protein capsid of bacteriophage Qbeta. Selective oxidation of the sialic acid residues on the glycan chains of transferrin was followed by introduction of a terminal alkyne functionality through an oxime linkage. Attachment of the protein to azide-functionalized Qbeta capsid particles in an orientation allowing access to the receptor binding site was accomplished by the Cu(I)-catalyzed azide-alkyne cycloaddition (CuAAC) click reaction. Transferrin conjugation to Qbeta particles allowed specific recognition by transferrin receptors and cellular internalization through clathrin-mediated endocytosis, as determined by fluorescence microscopy on cells expressing GFP-labeled clathrin light chains. By testing Qbeta particles bearing different numbers of transferrin molecules, it was demonstrated that cellular uptake was proportional to ligand density, but that internalization was inhibited by equivalent concentrations of free transferrin. These results suggest that cell targeting with transferrin can be improved by local concentration (avidity) effects.

Viral genomic single-stranded RNA directs the pathway toward a T=3 capsid

The molecular mechanisms controlling genome packaging by single-stranded RNA viruses are still largely unknown. It is necessary in most cases for the protein to adopt different conformations at different positions on the capsid lattice in order to form a viral capsid from multiple copies of a single protein. We showed previously that such quasi-equivalent conformers of RNA bacteriophage MS2 coat protein dimers (CP(2)) can be switched by sequence-specific interaction with a short RNA stem-loop (TR) that occurs only once in the wild-type phage genome. In principle, multiple switching events are required to generate the phage T=3 capsid. We have therefore investigated the sequence dependency of this event using two RNA aptamer sequences selected to bind the phage coat protein and an analogous packaging signal from phage Qbeta known to be discriminated against by MS2 coat protein both in vivo and in vitro. All three non-cognate stem-loops support T=3 shell formation, but none shows the kinetic-trapping effect seen when TR is mixed with equimolar CP(2). We show that this reflects the fact that they are poor ligands compared with TR, failing to saturate the coat protein under the assay conditions, ensuring that sufficient amounts of both types of dimer required for efficient assembly are present in these reactions. Increasing the non-cognate RNA concentration restores the kinetic trap, confirming this interpretation. We have also assessed the effects of extending the TR stem-loop at the 5' or 3' end with short genomic sequences. These longer RNAs all show evidence of the kinetic trap, reflecting the fact that they all contain the TR sequence and are more efficient at promoting capsid formation than TR. Mass spectrometry has shown that at least two pathways toward the T=3 shell occur in TR-induced assembly reactions: one via formation of a 3-fold axis and another that creates an extended 5-fold complex. The longer genomic RNAs suppress the 5-fold pathway, presumably as a consequence of steric clashes between multiply bound RNAs. Reversing the orientation of the extension sequences with respect to the TR stem-loop produces RNAs that are poor assembly initiators. The data support the idea that RNA-induced protein conformer switching occurs throughout assembly of the T=3 shell and show that both positional and sequence-specific effects outside the TR stem-loop can have significant impacts on the precise assembly pathway followed.

Heparin antagonism by polyvalent display of cationic motifs on virus-like particles

Particles to the rescue! The construction of cationic amino acid motifs on the surface of bacteriophage Qbeta by genetic engineering or chemical conjugation gives particles that are potent inhibitors of the anticoagulant action of heparin, which is a common anticlotting agent subject to clinical overdose.Polyvalent interactions allow biological structures to exploit low-affinity ligand-receptor binding events to affect physiological responses. We describe here the use of bacteriophage Qbeta as a multivalent platform for the display of polycationic motifs that act as heparin antagonists. Point mutations to the coat protein allowed us to generate capsids bearing the K16M, T18R, N10R, or D14R mutations; because 180 coat proteins form the capsid, the mutants provide a spectrum of particles differing in surface charge by as much as +540 units (K16M vs. D14R). Whereas larger poly-Arg insertions (for example, C-terminal Arg(8)) did not yield intact virions, it was possible to append chemically synthesized oligo-Arg peptides to stable wild-type (WT) and K16M platforms. Heparin antagonism by the particles was evaluated by using the activated partial thrombin time (aPTT) clotting assay; this revealed that T18R, D14R, and WT-(R(8)G(2))(95) were the most effective at disrupting heparin-mediated anticoagulation (>95 % inhibition). This activity agreed with measurements of zeta potential (ZP) and retention time on cation exchange chromatography for the genetic constructs, which distribute their added positive charge over the capsid surface (+180 and +360 for T18R and D14R relative to WT). The potent activity of WT-(R(8)G(2))(95), despite its relatively diminished overall surface charge is likely a consequence of the particle's presentation of locally concentrated regions with high positive charge density that interact with heparin's extensively sulfated domains. The engineered cationic capsids retained their ability to inhibit heparin at high concentrations and showed no anticlotting activity of the kind that limits the utility of antiheparin polycationic agents that are currently in clinical use.

On-virus construction of polyvalent glycan ligands for cell-surface receptors

Glycans arrayed on the exterior of virus particles were used as substrates for glycosyltransferase reactions to build di- and trisaccharides from the virus surface. The resulting particles exhibited tight and specific associations with cognate receptors on beads and cells, in one example defeating in cis cell-surface interactions in a manner characteristic of polyvalent binding. Combined with the ability of viruses to provide structurally well-defined attachment points, the methodology provides a convenient and powerful way to prepare complex carbohydrate ligands for clustered receptors.

Engineering thermal stability in RNA phage capsids via disulphide bonds

The RNA bacteriophages, a group that includes phages Qbeta and MS2, have a number of potential bionanotechnological applications, including cell specific drug delivery and as substrates for the formation of novel materials. Despite extensive sequence identity between their coat protein subunits, and an almost identical three-dimensional fold, Qbeta and MS2 capsids have dramatically different thermal stabilities. The increased stability of Qbeta has been correlated with the inter-subunit disulphide bonds present in that capsid and not present in MS2. We have tested this hypothesis directly using mass spectrometry. Analysis of the dissociated coat protein subunits suggests that inter-molecular disulphides are formed at the capsid five-fold but may not be at the three-fold axes. This conclusion has been tested by engineering disulphide cross-links into either the five-fold or three-fold positions of the recombinant MS2 capsid. Five-fold cross-linking results in a mutant with stability properties similar to those of Qbeta. Three-fold cross-linking results in a mutant unable to assemble T = 3 shells, implying that five-fold structures are on pathway to capsid assembly in these phages. The results demonstrate how it is possible to redesign the physical properties of phage shells and may be of general relevance to future applications of viruses and virus-like particles.

The influence of ionic strength on the interaction of viruses with charged surfaces under environmental conditions

The influence of ionic strength on the electrostatic interaction of viruses with environmentally relevant surfaces was determined for three viruses, MS2, Q beta, and Norwalk. The virus is modeled as a particle comprised of ionizable amino acid residues in a shell surrounding a spherical RNA core of negative charge, these charges being compensated for by a Coulomb screening due to intercalated ions. A second model of the virus involving surface charges only is included for comparison. Surface potential calculations for each of the viruses show excellent agreement with electrophoretic mobility and zeta potential measurements as a function of pH. The environmental surface is modeled as a homogeneous plane held at constant potential with and without a finite region (patch) of opposite potential. The results indicate that the electrostatic interaction between the virus and the oppositely charged patch is significantly influenced by the conditions of ionic strength, pH and size of the patch. Specifically, at pH 7, the Norwalk virus interacts more strongly with the patch than MS2 (approximately 51 vs approximately 9kT) but at pH 5, the Norwalk-surface interaction is negligible while that of MS2 is approximately 5.9kT. The resulting ramifications for the use of MS2 as a surrogate for Norwalk are discussed.

Replicable and recombinogenic RNAs

This paper summarizes results of the 40-year studies on replication and recombination of RNA molecules in the cell-free amplification system of bacteriophage Q. Special attention is paid to the molecular colony technique that has provided for the discovery of the nature of "spontaneous" RNA synthesis by Q replicase and of the ability of RNA molecules to spontaneously rearrange their sequences under physiological conditions. Also discussed is the impact of these data on the concept of RNA World and on the development of new in vitro cloning and diagnostic tools.

Mutilation of RNA phage Qbeta virus-like particles: from icosahedrons to rods

Icosahedral virus-like particles (VLPs) of RNA phage Qbeta are stabilized by four disulfide bonds of cysteine residues 74 and 80 within the loop between beta-strands F and G (FG loop) of the monomeric subunits, which determine the five-fold and quasi-six-fold symmetry contacts of the VLPs. In order to reduce the stability of Qbeta VLPs, we mutationally converted the amino acid stretch 76-ANGSCD-81 within the FG loop into the 76-VGGVEL-81 sequence. It led to production in Escherichia coli cells of aberrant rod-like Qbeta VLPs, along with normal icosahedral capsids. The length of the rod-like particles exceeded 4-30 times the diameter of icosahedral Qbeta VLPs.

A branched stem-loop structure in the M-site of bacteriophage Qbeta RNA is important for template recognition by Qbeta replicase holoenzyme

An internal site on bacteriophage Qbeta RNA, the M-site (map position 2545 to 2867), was recently shown by us to be required for the efficient initiation of minus strand synthesis by Qbeta replicase. In a more detailed mutational analysis, we show here that the essential elements within the M-site consist of two successive stem-loop structures followed by a bulge loop of unpaired purines, located at nucleotides 2696 to 2754 on the tip of a long, imperfectly base-paired stalk. Mutational changes affecting the sequences of paired or unpaired nucleotides in this segment reduced the template efficiency only mildly. The only severe effects were observed when one of the helical stems or the unpaired bulge was completely deleted or substantially shortened. We conclude that the three-dimensional backbone arrangement of these three elements constitutes the feature recognized by replicase. The role of the long stalk remains undetermined, because mutations that either stabilized or disrupted its base-pairing barely affected template activity, and even deletion of a major portion of one of its strands did not cause complete inactivation. Earlier evidence had implicated protein S1 (the alpha subunit of replicase) as the mediator of the M-site interaction. The lack of an active M-site on the Qbeta RNA template has the same quantitative and qualitative effects on template recognition as the absence of the S1 protein from replicase in the presence of wild-type RNA. We therefore believe that the M-site interaction explains most of the role of S1 protein in the replication of Qbeta RNA by replicase.

Structural plasticity in RNA and its role in the regulation of protein translation in coliphage Q beta

We have analyzed both conformational and functional changes caused by two large cis-acting deletions (delta 159 and delta 549) located within the read-through domain, a 850 nucleotide hairpin, in coliphage Q beta genomic RNA. Studies in vivo show that co-translational regulation of the viral coat and replicase genes has been uncoupled in viral genomes carrying deletion delta 159. Translational regulation is restored in deletion delta 549, a naturally evolved pseudorevertant. Structural analysis by computer modeling shows that structural features within the read-through domain of delta 159 RNA are less well determined than they are in the read-through domain of wild-type RNA, whereas predicted structure in the read-through domain of evolved pseudorevertant delta 549 is unusually well determined. Structural analysis by electron microscopy of the genomic RNAs shows that several long range helices at the base of the read-through domain, that suppress translational initiation of the viral replicase gene in the wild-type genome, have been destabilized in delta 159 RNA. In addition, the structure of local hairpins within the read-through region is more variable in delta 159 RNA than in wild-type RNA. Stable RNA secondary structure is restored in the read-through domain of delta 549 RNA. Our analyses suggest that structure throughout the read-through domain affects the regulation of viral replicase expression by altering the likelihood that long-range interactions at the base of the domain will form. We discuss possible kinetic and equilibrium models that can explain this effect, and argue that observed changes in structural plasticity within the read-through domain of the mutant genomes are key in understanding the process. During the course of these studies, we became aware of the importance of the information contained in the energy dot plot produced by the RNA secondary structure prediction program mfold. As a result, we have improved the graphical representation of this information through the use of color annotation in the predicted optimal folding. The method is presented here for the first time.

The crystal structure of bacteriophage Q beta at 3.5 A resolution

BACKGROUND

The capsid protein subunits of small RNA bacteriophages form a T = 3 particle upon assembly and RNA encapsidation. Dimers of the capsid protein repress translation of the replicase gene product by binding to the ribosome binding site and this interaction is believed to initiate RNA encapsidation. We have determined the crystal structure of phage Q beta with the aim of clarifying which factors are the most important for particle assembly and RNA interaction in the small phages.

RESULTS

The crystal structure of bacteriophage Q beta determined at 3.5 A resolution shows that the capsid is stabilized by disulfide bonds on each side of the flexible loops that are situated around the fivefold and quasi-sixfold axes. As in other small RNA phages, the protein capsid is constructed from subunits which associate into dimers. A contiguous ten-stranded antiparallel beta sheet facing the RNA is formed in the dimer. The disulfide bonds lock the constituent dimers of the capsid covalently in the T = 3 lattice.

CONCLUSIONS

The unusual stability of the Q beta particle is due to the tight dimer interactions and the disulfide bonds linking each dimer covalently to the rest of the capsid. A comparison with the structure of the related phage MS2 shows that although the fold of the Q beta coat protein is very similar, the details of the protein-protein interactions are completely different. The most conserved region of the protein is at the surface, which, in MS2, is involved in RNA binding.

Secondary structure model for the first three domains of Q beta RNA. Control of A-protein synthesis

We present a secondary structure model for the first 860 nucleotides of Q beta RNA. The model is supported by phylogenetic comparison, nuclease S1 structure probing and computer prediction using energy minimization and a Monte Carlo approach. To provide the necessary data for the comparative analysis we have sequenced the single-stranded RNA coliphages MX1, M11 and NL95. Together with the known sequences of Q beta and SP, this yields five sequences with sufficient sequence diversity to be useful for the analysis. The part of the Q beta genome examined contains the 60 nucleotide 5' untranslated region and the first 800 nucleotide of the maturation protein gene. The RNA adopts a highly ordered structure in which all hairpins are held in place by a network of long-distance interactions, which form three-way and four-way junctions. Only the 5'-terminal hairpin is unrestrained, while connected by a few single-stranded nucleotides to the body of the RNA. The start region of the A-protein gene, which is part of the network of long-distance interactions, is base-paired to three non-contiguous downstream sequences. As a result, translation is expected to be progressively quenched when the length of the nascent chains increases. This feature explains the previous observation that A-protein synthesis on Q beta RNA can start only on short nascent strands. Translational control of the A protein in the distantly related phage MS2 was recently shown to be controlled by the kinetics of RNA folding. This basic difference and its possible biological purpose can be explained by the different RNA folding pathways in Q beta and MS2. Interestingly, due to the presence of G-U pairs, structure prediction for the minus strand differs in some aspects from that for the plus strand. More specifically, there is a minus-strand specific, long-distance interaction bordering the minus-strand equivalent of the 5'-terminal hairpin. This interaction extends at the expense of the lower part of the terminal helix, thereby exposing the terminal C residues at which replication starts. This long-distance interaction, which was recently shown to be required for minus-strand replication, is strongly supported by our comparative data.

Secondary structure model for the last two domains of single-stranded RNA phage Q beta

We have determined the nucleotide sequence of three positive single-stranded RNA coliphages and have used this information, together with the known sequences of the related phages Q beta and SP, to construct a secondary structure model for the two distal domains of Q beta RNA. The 3' terminal domain, which is about 100 nucleotides long, contains most of the 3' untranslated region and folds into four short, regular hairpins. The adjacent 3' replicase domain contains about 1100 nucleotides. Hairpins in this protein-coding domain are much longer and more irregular than in the 3' untranslated region. Both domains are defined by long-distance interactions. The secondary structure is not a collection of hairpin structures connected by single-stranded regions. Rather, the RNA stretches between the stem-loop structures are all involved in an extensive array of long-distance interactions that contract the molecule to a rigid structure in which all hairpins are predicted to have a fixed position with respect to each other. A general feature of the model is that helices tend to be organized in four-way junctions with little or no unpaired nucleotides between them. As a result, there is a potential for coaxial stacking of adjacent stems. The essential features of the model are supported by the S1 nuclease cleavage pattern. Viral RNA sequences are strongly constrained by their coding function. As a result, structural evolution by simple base-pair substitution is not always possible, as this usually requires the juxtaposition of the codon wobble positions in stems. Rather, we often observe co-ordinate base substitutions that maintain the stem, but tend to change the position at which bulges or internal loops are found. Structures that differ in this way are apparently equally fit. Also, the relative position of hairpin loops can shift several nucleotides through an alignment based on maximal sequence identity i.e. amino acid homology. The fact that these structural irregularities do not occur at the 3' untranslated region suggests indeed that the coding function of the RNA constrains the secondary structure. Hairpins with the stable tetraloop motif GNRA and UNCG or their complement are over-represented. This suggests their involvement in segregation of plus and minus strand. The genome of the coliphages contains a well-defined high affinity binding site for the coat protein, which serves to suppress replicase translation and also acts as a nucleation point in capsid formation. Close to the 3' end we find an additional conserved helix that meets the described consensus criteria for coat-protein binding.