Vector uncoating limits adeno-associated viral vector-mediated transduction of human dendritic cells and vector immunogenicity

Abstract

AAV vectors poorly transduce Dendritic cells (DC), a feature invoked to explain AAV’s low immunogenicity. However, the reason for this non-permissiveness remained elusive. Here, we performed an in-depth analysis using human monocyte-derived immature DC (iDC) as model. iDC internalized AAV vectors of various serotypes, but even the most efficient serotype failed to transduce iDC above background. Since AAV vectors reached the cell nucleus, we hypothesized that AAV’s intracellular processing occurs suboptimal. On this basis, we screened an AAV peptide display library for capsid variants more suitable for DC transduction and identified the I/VSS family which transduced DC with efficiencies of up to 38%. This property correlated with an improved vector uncoating. To determine the consequence of this novel feature for AAV’s in vivo performance, we engineered one of the lead candidates to express a cytoplasmic form of ovalbumin, a highly immunogenic model antigen, and assayed transduction efficiency as well as immunogenicity. The capsid variant clearly outperformed the parental serotype in muscle transduction and in inducing antigen-specific humoral and T cell responses as well as anti-capsid CD8 + T cells. Hence, vector uncoating represents a major barrier hampering AAV vector-mediated transduction of DC and impacts on its use as vaccine platform.

High-Throughput In Vitro, Ex Vivo, and In Vivo Screen of Adeno-Associated Virus Vectors Based on Physical and Functional Transduction

“…For example, AAV4 was the most abundant variant in T cells at the DNA level, but its relative expression at the RNA level in those cells was very low, implying that this serotype may be inefficient at one or multiple steps the vector must complete between cell entry and triggering transgene expression. While it has been previously shown that barriers such as inefficient nuclear translocation, 46 nuclear uncoating, 47,48 and second-strand synthesis/transcriptional regulation by the capsid 49 can affect function of AAV vectors, it was interesting and unexpected that a vector capable of such efficient cell entry was so inefficient at transgene expression. These results position AAV4 as a powerful genetic tool to study AAV biology and the interactions between AAV and human T cells.…”

The Interplay between Adeno-Associated Virus and Its Helper Viruses

“…Once located within the nucleus, AAV2 was found to accumulate in the nucleoli in an intact infectious state [ 55 ]. The mechanism of AAV uncoating is largely unknown and seems to be a limiting step during AAV transduction [ 56 , 57 ]. Both wildtype (wt) AAV as well as rAAV can integrate into the host genome [ 58 ].…”

Towards Clinical Implementation of Adeno-Associated Virus (AAV) Vectors for Cancer Gene Therapy: Current Status and Future Perspectives

“…Besides RGD4c, a portfolio of AAV targeting vectors are nowadays available for improved transduction of endothelial cells (reviewed in [ 10 , 62 , 63 ] and Table 2 ), some of which are also functional in the context of quiescent endothelial cells or endothelial progenitor cells which are discussed as tools for realizing “Trojan horse” concepts [ 81 ]. Moreover, for DC [ 83 ] and T cells [ 56 ] targeting AAV vectors are available.…”

“…These candidates are then produced as viral vectors and comprehensively characterized regarding production yield, efficiency, specificity and immune escape capabilities prior to application. In addition, these novel variants can be used as tools to identify the nature of barrier/s that impair transduction by the parental serotypes thereby adding to the understanding of the host-AAV interaction [ 81 , 83 ]. Finally, strategies are available to map the targeted receptors [ 85 ].…”