PHP.B/Ly6A: Why only C57BL/6J works. Deverman et al. 2016 used the CREATES method to screen out AAV-PHP.B, IV to achieve widespread transduction in the whole brain in C57BL/6J [10-22]. Hordeaux et al. 2018 directly proved that its neurotropism is limited to C57BL/6J, and the advantage in the brain disappears in BALB/c, CD-1 and NHP; Batista et al. subsequently proved that FVB and DBA/2 are partially allowed, but BALB/c and CBA/J are not allowed [10-40][10-41]. Mechanistically, Ly6A/Sca-1 (lymphocyte activation antigen 6A, GPI-anchored protein) is a receptor for PHP.B that crosses the BBB. It is highly expressed on the luminal surface of brain microvascular endothelium in C57BL/6J and is not expressed in BALB/cJ due to promoter polymorphism [10-42]. Cynomolgus/rhesus monkeys do not have an ortholog of Ly6A, so PHP.B/PHP.eB is not superior to AAV9 in NHP brains [10-27]. Conclusions: The PHP.B family is a mouse strain-specific research tool and cannot be extrapolated to NHP/humans.
Humanized receptor designed capsid BI-hTFR1. Huang et al. 2024 screened BI-hTFR1 through directed evolution, which binds to the luminal epitope of human transferrin receptor 1 (TfR1/TFRC) (does not compete with endogenous transferrin ligand) [10-43]. It provides a 40-50-fold CNS reporter gene enhancement over AAV9 in human TFRC knock-in (KI) mice and is CNS-specific; this advantage is completely lost in wild-type (murine TfR1) mice. BI-hTFR1 does not recognize mouse TfR1 and must be evaluated in human TfR1 KI mice, human BBB models or NHP, and ordinary C57BL/6 is ineffective - this is a paradigm case of "receptor species determines capsid effectiveness" [10-43].
Conservative list. Relatively conservative and can be extrapolated across species: liver targeting of AAV8/AAVrh.10/AAVrh74 (both mice and cynomolgus monkeys have the most efficient capsid in the liver) [10-27][10-35]; AAV9 has a broad spectrum of cardiac/skeletal muscle (effective in mice, rats, dogs, and NHPs) [10-28]; AAV4 has liver off-targeting + lung/glomerular/islet bias (cross-targeting) Consistent with C57BL/6, BALB/c, and cynomolgus monkeys) [10-27]; MyoAAV muscle preference (both mice and NHP show 10-50 times the muscle mRNA of the native type) [10-44].
Not a conservative list. The differences are huge and cannot be directly extrapolated: BBB crossing of AAV-PHP.B/PHP.eB/CAP-B10 (only C57BL/6 line, NHP has no advantage) [10-27][10-40]; liver advantage of AAV-DJ (mouse only, NHP liver is not high) [10-27]; AAV-LK03 human hepatocyte tropism (only transduced humanized mouse/human hepatocytes, apparent transcriptional silencing of mouse hepatocytes) [10-20]; liver tropism of AAV6.2FF/LK03 (only NHP effective, not mice) [10-27]; brain endothelial tropism of AAV9-X1 (mouse CNS endothelium is strong, NHP cell type distribution is different) [10-9].
Examples of cross-species decay. Yang 2024 showed that although capsid DNA distribution was similar between mice and monkeys, RNA/protein expression was significantly lower in 20 of 28 tissues in monkeys than in mice (liver 66.5%→0.33%, heart 54%→7.8%, muscle 79%→34%, midbrain 30%→6%, tdTomato⁺ cell ratio) [10-27]. That is, high transduction in mice ≠ high transduction in humans/monkeys, and apparent transcriptional silencing is an important bottleneck in the low-dose monkey model.
Tissue distribution of commonly used NHP serotypes. AAV8 IV is highest in the whole body liver, and the liver is still the largest peripheral pool after IT [10-45]; AAV9 NHP is high in liver, heart, and skeletal muscle after IV, ICM/IT has extensive neuronal and glial transduction in the whole brain + spinal cord (Zolgensma clinical basis), neonatal/fetal NHP neuronal transduction is better than that in adults [10-45][10-46]; AAVrh.10 NHP Liver and muscle are high, and heart expression is lower than AAV9 in head-to-head barcode studies [10-36]; AAVrh74 NHP skeletal muscle is high (Elevidys clinical 2×10¹⁴ vg/kg grade), and liver is the main off-target [10-37][10-38]; MyoAAV-2A NHP muscle mRNA is 10-50 higher than the native type. times, liver off-target reduction [10-44].