Development of a Bispecific Antibody-Based Platform for Retargeting of Capsid Modified AAV Vectors

A major limiting factor for systemically delivered gene therapies is the lack of novel tissue specific AAV (Adeno-associated virus) derived vectors. Bispecific antibodies can be used to redirect AAVs to specific target receptors. Here, we demonstrate that the insertion of a short linear epitope “2E3” derived from human proprotein-convertase subtilisin/kexin type 9 (PCSK9) into different surface loops of the VP capsid proteins can be used for AAV de-targeting from its natural receptor(s), combined with a bispecific antibody-mediated retargeting. We chose to target a set of distinct disease relevant membrane proteins—fibroblast activation protein (FAP), which is upregulated on activated fibroblasts within the tumor stroma and in fibrotic tissues, as well as programmed death-ligand 1 (PD-L1), which is strongly upregulated in many cancers. Upon incubation with a bispecific antibody recognizing the 2E3 epitope and FAP or PD-L1, the bispecific antibody/rAAV complex was able to selectively transduce receptor positive cells. In summary, we developed a novel, rationally designed vector retargeting platform that can target AAVs to a new set of cellular receptors in a modular fashion. This versatile platform may serve as a valuable tool to investigate the role of disease relevant cell types and basis for novel gene therapy approaches.

Introduction of Surface Loops as a Tool for Encapsulin Functionalization.

Encapsulin based protein cages are nanoparticles with different biomedical applications, such as targeted drug delivery or imaging agents. These particles are biocompatible and can be produced in bacteria, allowing large scale production and protein engineering. In order to use these bacterial nanocages in different applications, it is important to further explore the potential of their surface modification and optimize their production. In this study we design and show new surface modifications of the Thermotoga maritima (Tm) and Brevibacterium linens (Bl) encapsulins. Two new loops on Tm encapsulin with a His-tag insertion after the residue 64 and the residue 127, and the modification of the C-terminal on Bl encapsulin, are reported. The multi-modification of the Tm encapsulin enables up to 240 different functionalities on the cage surface, resulting from 4 potential modifications per protein subunit. We furthermore report an improved protocol giving a better stability and providing a notable increase of the production yield of the cages. Finally, we tested the stability of different encapsulin variants over a year and the results show a difference in stability arising from the tag insertion position. These first insights in the structure-property relationship of encapsulins, with respect to the position of a function loop, allow for further study of the use of these protein nanocages in biomedical applications.

Structure of Blood Coagulation Factor VIII in Complex with an Anti-C1 Domain Pathogenic Antibody Inhibitor

Antibody inhibitor development in hemophilia A represents the most significant complication resulting from factor VIII (fVIII) replacement therapy. Recent studies have demonstrated that epitopes present in the C1 domain contribute to a pathogenic inhibitor response. In this study, we report the structure of a Group A anti-C1 domain inhibitor, termed 2A9, in complex with a B domain-deleted, bioengineered fVIII construct (ET3i). The 2A9 epitope forms direct contacts to the C1 domain at three different surface loops consisting of Lys2065-Trp2070, Arg2150-Tyr2156 and Lys2110-Trp2112. Additional contacts are observed between 2A9 and the A3 domain, including the Phe1743-Tyr1748 loop and the N-linked glycosylation at Asn1810. Most of the C1 domain loops in the 2A9 epitope also represent a putative interface between fVIII and von Willebrand factor (vWF). Lastly, the C2 domain in the ET3i:2A9 complex adopts a large, novel conformational change, translocating outward from the structure of fVIII by 20 Å. This study reports the first structure of an anti-C1 domain antibody inhibitor and the first fVIII:inhibitor complex with a therapeutically active fVIII construct. Further structural understanding of fVIII immunogenicity may result in the development of more effective and safe fVIII replacement therapies.

High catalytic rate of the cold-active Vibrio alkaline phosphatase depends on a hydrogen bonding network involving a large interface loop

The role of surface loops in mediating communication through residue networks is still a relatively poorly understood part of cold-adaptation of enzymes, especially in terms of their quaternary interactions. Alkaline phosphatase (AP) from the psychrophilic marine bacterium Vibrio splendidus (VAP) is characterized by an analogous large surface loop in each monomer, referred to as the large-loop, that hovers over the active site of the other monomer. It presumably has a role in VAP high catalytic efficiency that accompanies extremely low thermal stability. We designed several different mutagenic variants of VAP with the aim of removing inter-subunit interactions at the dimer interface. Breaking the inter-subunit contacts from one residue in particular (Arg336) caused diminished temperature stability of the catalytically potent conformation and a drop in catalytic rate by a half. The relative B-factors of the R336L crystal structure, compared to the wild-type, confirmed increased surface flexibility in a loop on the opposite monomer, but not in the large-loop. Contrary to expectations, the observed reduction in stability with an expected increase in dynamic mobility resulted in reduced catalytic rate. This contradicts common theories explaining high catalytic rates of enzyme from cold-adapted organisms as being due to reduced internal cohesion bringing increased dynamic flexibility to catalytic groups. The large-loop increases the area of the interface between the subunits through its contacts and may facilitate an alternating structural cycle demanded by a half-of-sites reaction mechanism through stronger ties, as the dimer oscillates between high affinity (active) or low phosphoryl-group affinity (inactive).

Expression of influenza A virus-derived peptides on a rotavirus VP6-based delivery platform

Recombinant protein technology enables the engineering of modern vaccines composed of a carrier protein displaying poorly immunogenic heterologous antigens. One promising carrier is based on the rotavirus inner-capsid VP6 protein. We explored different VP6 insertion sites for the presentation of two peptides (23 and 140 amino acids) derived from the M2 and HA genes of influenza A virus. Both termini and three surface loops of VP6 were successfully exploited as genetic fusion sites, as demonstrated by the expression of the fusion proteins. However, further studies are needed to assess the morphology and immunogenicity of these constructs.

Conserved Sequence Features in the Spike Protein Provide Evidence Suggesting the Origin of SARS-CoV-2 (COVID-19) Virus by Recombination between SARS virus and Another Sarbecovirus

Both SARS-CoV-2 (COVID-19) and SARS coronaviruses (CoVs) are members of the subgenus Sarbecovirus. To understand the origin of SARS-CoV-2, protein sequences from sarbecoviruses were analyzed to identify highly-specific molecular markers consisting of conserved inserts or deletions (termed CSIs) in the spike (S) and nucleocapsid (N) proteins that are specific for either particular clusters/lineages of these viruses or are commonly shared by specific lineages. Three novel CSIs in the N-terminal domain of the spike protein S1-subunit (S1-NTD) are uniquely shared by the SARS-CoV-2, BatCoV-RaTG13 and most pangolin CoVs, distinguishing this cluster of viruses (SARS-CoV-2r) from all others. In the same positions, where these CSIs are found, related CSIs are also present in two other sarbecoviruses (viz. CoVZXC21 and CoVZC45 forming CoVZC cluster), which form an out group of the SARS-CoV-2r cluster. These three CSIs are not found in the SARS-CoVs. However, both SARS and SARS-CoV-2r CoVs contain two large CSIs in the C-terminal domain of S1 (S1-CTD), which binds the human ACE-2 receptor, that are absent in the CoVZC cluster of CoVs. These results indicate that while the S1-NTD of the SARS-CoV-2r viruses possesses the sequence characteristics of the CoVZC cluster of CoVs, their S1-CTD resembles the SARS viruses. Thus, the spike protein of SARS-CoV-2r viruses has likely originated from a recombination event between the S1-NTD of the CoVZC viruses and the S1-CTD of SARS viruses. This inference is also supported by the amino acid sequence similarity of the S1-NTD and S1-CTD from SARS-CoV-2 compared to the CoVZC and SARS CoVs. We also present evidence that one of the pangolin-CoV_MP789, whose receptor-binding domain is most similar to the SARS-CoV-2, is also derived by a recent recombination between the S1-NTD of the CoVZC CoVs and the S1-CTD of a SARS-CoV-2 related virus. Several other identified CSIs are specific for others clusters of sarbecoviruses including a clade consisting of bat SARS-CoVs (BM48-31/BGR/2008 and SARS_BtKY72). Structural mappings studies show that the identified CSIs are located within surface-exposed loops and form distinct patches on the surface of the spike protein. These surface loops/patches are predicted to interact with other host components and play important role in the biology/pathology of SARS-CoV-2 virus. Lastly, the CSIs specific for the SARS-CoV-2r clade provide novel means for development of new diagnostic and therapeutic targets for these viruses.