Targeting AAV Vectors to the Central Nervous System by Engineering Capsid–receptor Interactions That Enable Crossing of the Blood–brain Barrier
Qin Huang, Albert T. Chen, Ken Y. Chan, Hikari Sorensen, Andrew J. Barry, Bahar Azari, Qingxia Zheng, Thomas Beddow, Binhui Zhao, Isabelle G. Tobey, Cynthia Moncada-Reid, Fatma-Elzahraa Eid, Christopher J. Walkey, M. Cecilia Ljungberg, William R. Lagor, Jason D. Heaney, Yujia A. Chan, Benjamin E. Deverman
Abstract
Viruses have evolved the ability to bind and enter cells through interactions with a wide variety of cell macromolecules. We engineered peptide-modified adeno-associated virus (AAV) capsids that transduce the brain through the introduction of de novo interactions with 2 proteins expressed on the mouse blood–brain barrier (BBB), LY6A or LY6C1. The in vivo tropisms of these capsids are predictable as they are dependent on the cell- and strain-specific expression of their target protein. This approach generated hundreds of capsids with dramatically enhanced central nervous system (CNS) tropisms within a single round of screening in vitro and secondary validation in vivo thereby reducing the use of animals in comparison to conventional multi-round in vivo selections. The reproducible and quantitative data derived via this method enabled both saturation mutagenesis and machine learning (ML)-guided exploration of the capsid sequence space. Notably, during our validation process, we determined that nearly all published AAV capsids that were selected for their ability to cross the BBB in mice leverage either the LY6A or LY6C1 protein, which are not present in primates. This work demonstrates that AAV capsids can be directly targeted to specific proteins to generate potent gene delivery vectors with known mechanisms of action and predictable tropisms.
Introduction
Gene therapy with recombinant adeno-associated viruses (AAVs) shows promise for treating diseases at their root genetic cause, but remains constrained by the inefficiency of delivery to disease-relevant organs and cell types. Natural AAV capsids can be modified to produce vectors with dramatically improved in vivo tropisms. An effective engineering strategy has been to generate diverse libraries of capsid variants via peptide insertions and to subject these libraries to multiple rounds of in vivo selection to identify capsids with the desired properties such as central nervous system (CNS)-wide transduction [1–3], brain vascular endothelium targeting [4,5], retrograde transduction in the CNS [6], transduction of human hepatocytes in a xenograft system [7], photoreceptor transduction [8], and muscle transduction [9,10]. However, these screening efforts have been limited to function-focused approaches, where capsids are selected for a particular biodistribution or cell type tropism without discriminating for mechanism of action.
Materials and method
Capsid library cloning
The RNA expression system for the selection of functional AAV capsids was used as previously described [4] with a modification to include a Woodchuck Hepatitis Virus (WHV) Posttranscriptional Regulatory Element (WPRE) between the restriction enzyme site SalI and HindIII. The wild-type AAV9 capsid gene sequence was synthesized (GenScript) with nucleotide changes at S448 (TCA to TCT, silent mutation), K449R (AAG to AGA), and G594 (GGC to GGT, silent mutation) to introduce XbaI and AgeI restriction enzyme recognition sites for library fragment cloning.
Results
A high-throughput purified protein assay identifies capsids selective for LY6A or LY6C1
To assess the potential of a mechanism-focused approach to develop capsids with enhanced CNS tropisms, we targeted 2 surface proteins present on brain vascular endothelial cells: LY6A, the known receptor for the AAV-PHP.B family of capsids, as a positive control, and a related protein, LY6C1, a novel target likewise highly expressed on CNS endothelial cells [11,32]. LY6C1 was selected based on the hypothesis that it may share LY6A’s ability to mediate AAV transport into the CNS, given that the LY6 family possesses a conserved protein structure and subcellular localization [33]. We generated LY6A and LY6C1 proteins as Fc fusions and used a magnetic bead-based pull-down assay to perform initial (Round 1) screens of 2 independently generated 7-mer-modified AAV9 libraries (random 7-mer amino acid sequences were inserted between residues 588–589 in VP1)—named Library 1 and Library 2, respectively—for variants that bind to LY6A-Fc, LY6C1-Fc, or an Fc-only control (Library 1 data is shown in Fig 1B, 1D and 1E; Library 2 data is shown in S1 Fig).
Discussion
We present a rapid method for enhancing the tropism of AAV vectors by introducing de novo interactions with proteins expressed on target cells. Our approach generated BBB-crossing capsids by first screening for direct in vitro interactions with specific proteins rather than by selecting immediately for in vivo success. This mechanism-focused strategy identified thousands of capsids that specifically bind to the mouse brain endothelial cell surface proteins, LY6A or LY6C1, and many of these capsids exhibited enhanced CNS tropisms when validated in vivo, both in a pooled library and when tested individually. Importantly, the tropisms observed with LY6A- and LY6C1-binding capsids in different mouse strains matched expectations based on the strain-specific expression of these proteins. These results demonstrate how new virus capsid–receptor interactions can be introduced through the addition of short linear insertions into AAV capsid proteins.
Acknowledgments
We thank the members of the Deverman laboratory for continuous discussions of the project; and Alexa E. Martinez, Denise G. Lanza, and John R. Seavitt for providing logistical and technical support for the validation studies at Baylor College of Medicine.
Citation: Huang Q, Chen AT, Chan KY, Sorensen H, Barry AJ, Azari B, et al. (2023) Targeting AAV vectors to the central nervous system by engineering capsid–receptor interactions that enable crossing of the blood–brain barrier. PLoS Biol 21(7): e3002112. https://doi.org/10.1371/journal.pbio.3002112
Academic Editor: Chaitan Khosla, Stanford University, UNITED STATES
Received: November 29, 2022; Accepted: April 6, 2023; Published: July 19, 2023
Copyright: © 2023 Huang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All code used in this study is available on GitHub: https://github.com/vector-engineering/AAV_capsid_receptor/. All data required to reproduce each plot are available from the Zenodo open repository with DOI: 10.5281/zenodo.7689795. These data include RPMs and enrichment values from the libraries used in this study. Supplementary Data contain analyzed and processed values derived from the data available on Zenodo. Some of the data presented in S9 Fig is available as part of the NIH SCGE Toolkit at https://scge.mcw.edu/toolkit/data/experiments/group/1441 All materials were acquired from commercial vendors as described. Packaging plasmids carrying the individually characterized LY6A (AAV-BI28: 203532; AAV-BI48: 203533; AAV-BI49:203534) or LY6C1 binding capsids (AAV-BI28: 203532; AAV-BI62:203535; AAV-BI65:203536), AAV-CAG-NLS-mScarlet-2A-Luc-WPRE-pA (203539), AAV-GfABC1D-SaCas9-WPRE-pA (203540), and AAV-GfABC1D-GFP-U6-L1-U6-R2 (203541) will be made available through Addgene under the indicted Addgene IDs.
Funding: Work in this study was supported by the National Institute of Neurological Disorders and Stroke and NIH Common Fund through the Somatic Cell Genome Engineering (SCGE) Program (UG3NS111689 to B.E.D); the Stanley Center for Psychiatric Research, Broad Institute (B.E.D); Apertura Gene Therapy (B.E.D); a Brain Initiative award funded through the National Institute of Mental Health (UG3MH120096 to B.E.D). F.E.E. was supported by a Broad Shark Tank Award and Y.A.C. was supported by a Broad Ignite Award. Work at the BCM-Rice was supported by NIH SCGE grant (U42OD026645 to J.D.H. and W.R.L). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: BED is a scientific founder at Apertura Gene Therapy and a scientific advisory board member at Tevard Biosciences. BED, QH, KYC, and FEE are named inventors on patent applications filed by the Broad Institute of MIT and Harvard related to this study. Remaining authors declare that they have no competing interests.
Abbreviations: AA, amino acid; AAV, adeno-associated virus; BBB, blood–brain barrier; CNS, central nervous system; ML, machine learning; NGS, next-generation sequencing; OCT, optimal cutting temperature; PBS, phosphate buffered saline; PWM, position weight matrix; RPM, reads per million; SVAE, supervised variational auto-encoder
https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002112#ack
