Receptor reprogramming. The essence of receptor reprogramming is to "replace receptors rather than cells." It usually retains AAVR dependence and adds a new target enrichment pathway. The PHP.B family inserts a 7-mer peptide into AAV9 VR-IV to obtain Ly6A binding, adding a "trans-brain endothelial Ly6A transcellular" pathway [10-22]; subsequent PHP.eB optimization efficiency [10-21], PHP.V1 partial to brain endothelium, PHP.N partial to neurons, PHP.A partial to glial and liver decreased by 152 times [10-9]. MyoAAV/AAVMYO inserts the RGD motif, actively binds to integrin αVβ6/αVβ8/α5β1, and reprograms AAV9 from galactose-dependent to integrin-dependent. Muscle mRNA is 10-50 times higher than AAV9 (NHP), liver off-target is reduced, and AAVR is still required [10-44][10-52]. BI-hTFR1 binds to the luminal epitope of human TfR1 to achieve trans-BBB receptor-mediated transcellular [10-43]. In terms of lung targeting, AAV-QPE/VNT and others achieve lung redirection by destroying AAV2 HSPG binding, and AAV2.5T (2/5 chimeric) is efficiently transferred to the airway epithelium through the airway lumen [10-9]; 9P31/9P36 binds to primate conserved carbonic anhydrase IV (CA-IV) [10-9].
Peptide insertion. Inserting targeting peptides into VR-IV/VR-VIII (GH loop) is the most commonly used method to change the affinity: RGD peptide → integrin binding (muscle/tumor blood vessels) [10-44]; TLAVPFK (PHP.B) → Ly6A [10-22]; AAV2.7m8 (AAP framework mutation + peptide display) → Intravitreal administration penetrates all layers of the retina, and phototransduction can reach about 60% (natural type only) 2-20%) [10-47]; DJ-NSS/DJ-MVN intranasal administration redirects to alveolar epithelial/alveolar macrophages respectively [10-9]. The inserted peptide must be exposed to the surface of the capsid tripod protrusions without disrupting AAVR binding and assembly.
Filter sources (brief description). CREATES/Cre relies on selection to enrich the correct capsids in specific cell types of Cre mice (PHP.B source) [10-22]; DELIVER in vivo directed evolution to retain muscle tropism across species (MyoAAV) [10-44]; barcode in vivo screening 21-100+ capsid pools in one injection, barcode sequencing to quantify tissue DNA/RNA (Yang 2024 paradigm) [10-27]; transcriptome-assisted screening (scRNA-seq/spatial transcriptome) analyzes single-cell level tropism [10-48]; humanized models (human liver chimeric mouse LK03 [10-20], human TfR1 KI mouse [10-43], human BBB chip) solve the problem of "mouse screening is effective, but human screening is invalid". See Chapter 12 for detailed screening procedures.