Molecular engineering of viral gene delivery vehicles

Abstract

Viruses can be engineered to efficiently deliver exogenous genes, but their natural gene delivery properties often fail to meet human therapeutic needs. Therefore, engineering viral vectors with new properties, including enhanced targeting abilities and resistance to immune responses, is a growing area of research. This review discusses protein engineering approaches to generate viral vectors with novel gene delivery capabilities. Rational design of viral vectors has yielded successful advances in vitro, and to an extent in vivo. However, there is often insufficient knowledge of viral structure-function relationships to reengineer existing functions or create new capabilities, such as virus-cell interactions, whose molecular basis is distributed throughout the primary sequence of the viral proteins. Therefore, high-throughput library and directed evolution methods offer alternative approaches to engineer viral vectors with desired properties. Parallel and integrated efforts in rational and library-based design promise to aid the translation of engineered viral vectors toward the clinic.

Figure 1

Schematic of the structure of viral particles and organization of the viral genomes. (a) Three-dimensional representation of the AAV capsid from VIPER database (http://viperdb.scripps.edu/) and schematic of AAV genome. (b) Representation of key components of the adenoviral capsid and genome organization. (c) Representation of key components of retroviral and lentiviral particles along with the genome organization of each virus.

Figure 2

Overview of rational protein engineering strategies for viral vectors. Rational design methods include (a) the use of a bispecific adaptor, (b) pseudotyping with an alternate capsid or VAP, (c,d) the generation of mosaic or chimeric particles, and (e) genetic engineering of VAP to insert peptide or point mutations.

Figure 3

Overview of library protein engineering strategies for viral vectors. Library generation and selection methods include (a) display of random peptide in defined location, (b) random insertional mutagenesis, (c) random point mutagenesis, and (d) in vitro recombination. Upon DNA library generation, a highly diverse viral library is produced (e). Finally, (f) high-throughput selection followed by (g) recovery of successful variants and iteration with steps (a)–(d) and (e) employs directed evolution to enhance desired gene delivery properties.