The AAV receptor, known in the research world as AAVR and encoded by the gene KIAA0319L, is a membrane protein that most adeno-associated virus serotypes depend on to enter human cells. Its discovery in 2016 reshaped gene therapy research, because AAV vectors are the most widely used vehicles for delivering therapeutic genes. But AAVR is not just a doorway for viruses. The protein belongs to a gene family implicated in brain development and has its own biological functions that researchers are still working to understand, including emerging connections to neurodevelopmental conditions like dyslexia.
How Different AAV Serotypes Grab the Same Receptor
AAVR sits on the surface of cells with a string of five immunoglobulin-like domains, called PKD (polycystic kidney disease) repeats, reaching outward from the membrane. Different AAV serotypes latch onto different parts of this chain. AAV2, the most studied serotype, binds mainly to the second PKD domain (PKD2), which sits closer to the cell surface. Cryo-electron microscopy at near-atomic resolution shows that PKD2 docks onto the spike region of the AAV2 capsid shell, near its three-fold symmetry axis. Specific amino acids in strands B and E of PKD2, plus a loop between them, make direct contact with the virus, and when researchers mutate those residues, both binding and infectivity drop sharply.1Nature Microbiology. Adeno-associated virus 2 bound to its cellular receptor AAVR
AAV5 takes a different approach. Instead of PKD2, it engages predominantly with PKD1, the domain farthest from the cell membrane. Structural work reveals that PKD1 binds on the opposite side of the AAV5 capsid spike compared with where PKD2 contacts AAV1, using an entirely different set of contact residues.2Nature Communications. Divergent engagements between adeno-associated viruses with their cellular receptor AAVR Still other serotypes, including AAV1 and AAV8, require a combination of both PKD1 and PKD2 for efficient entry.3PubMed Central. Adeno-associated Virus Serotypes Have Distinctive Interactions with Domains of the Cellular AAV Receptor The upshot is that a single receptor serves as the gateway for a wide family of viruses, but each serotype has evolved its own handshake with that receptor. This has practical consequences for gene therapy design, because engineering a capsid to change its grip on AAVR can redirect which tissues a vector prefers.
Getting Inside Without Breaking Down the Door
Before AAV even contacts AAVR, it often needs a boost from sugar molecules on the cell surface. For AAV2, heparan sulfate proteoglycans serve as an attachment factor that concentrates virus particles near the membrane, improving the odds that the virus then encounters AAVR and binds productively.4Molecular Therapy. Adeno-associated virus (AAV) cell entry: structural insights Think of the sugar coating as a landing strip that slows the virus down long enough for the real docking to happen.
What happens after AAVR grabs the virus surprised researchers. The conventional assumption was that AAV particles get swallowed into endosomes, then punch through the endosomal membrane to reach the cell interior. Recent work tells a different story. AAVR’s cytoplasmic tail acts as cargo for the retromer-SNX3 complex, a piece of cellular machinery that shuttles proteins backward from endosomes to the trans-Golgi network (TGN). AAV2 essentially hitches a ride on this retrograde route. In cells that lack AAVR, virus particles still get internalized, but they pile up in early endosomes and never reach the nucleus. Critically, there is no sign of the membrane rupture that would accompany a brute-force endosomal escape, a mechanism used by some other delivery systems like lipid nanoparticles.5PubMed Central. AAV2 bypasses direct endosomal escape by using AAVR to access the trans-Golgi network en route to the nucleus
A trafficking adaptor protein called TBC1D23 has been identified as a key link in this chain. TBC1D23 physically binds the cytoplasmic tail of AAVR through an acidic cluster of residues, coupling the AAVR-virus complex to the vesicle transport machinery. Without TBC1D23, internalized AAV capsids fail to converge at the TGN, and nuclear import drops.6PubMed Central. TBC1D23-AAVR interaction drives endosome-to-TGN trafficking required for rAAV transduction This vesicle-guided, receptor-mediated route is a cleaner picture than the older escape model and could eventually inform how researchers design more efficient gene therapy vectors.
What AAVR Does When No Virus Is Around
AAVR was not built for the benefit of a virus. The gene that encodes it, KIAA0319L, is part of a family that includes KIAA0319, a gene that has been linked to dyslexia for over a decade. The protein’s normal job appears to involve cell migration during development. In chick embryos, KIAA0319L is expressed in the developing visual system, and when researchers knock it down in the optic tectum (the brain region responsible for visual processing in birds), neurons fail to migrate properly.7TSpace University of Toronto Repository. The Expression Profile of KIAA0319-like in Chick Embryos and its Involvement in Cell Migration in the Developing Optic Tectum
Mice engineered to completely lack AAVR are viable and show no obvious gross abnormalities, which is part of what made them useful for confirming that the receptor is essential for AAV infection in a living animal.8PubMed Central. An essential receptor for adeno-associated virus infection However, the fact that knockout mice survive does not mean the protein is unimportant. Many genes involved in fine-tuning brain development produce subtle phenotypes that standard cage assessments would miss. The growing evidence from its paralog, KIAA0319, suggests that these proteins play roles in cortical organization, and that their disruption could have consequences for cognitive development even when the animal appears superficially healthy.
The KIAA0319 Family and Dyslexia
Numerous genetic association studies have linked variants in KIAA0319 to susceptibility to dyslexia.9PLoS Genetics. A Common Variant Associated with Dyslexia Reduces Expression of the KIAA0319 Gene KIAA0319 is the paralog of KIAA0319L (AAVR), meaning the two genes descended from a common ancestor and share structural features. Research using human cortical organoids shows that knocking out KIAA0319 produces smaller organoids with morphological abnormalities, disrupted neurogenesis, and impaired neuronal maturation. Single-cell sequencing of those organoids revealed that neural progenitor cells favored a stunted radial glia fate instead of maturing into neurons. Strikingly, the disruption was not limited to KIAA0319 itself: a whole network of dyslexia-associated genes became coordinately dysregulated, affecting processes like primary cilia formation, cortical organization, and neural network connectivity.10PubMed. KIAA0319 Plays a Critical Role in Cortical Neuronal Maturation and Synaptic Development Through a Dyslexia-Associated Gene Network
This does not mean that AAVR mutations cause dyslexia. The two proteins are related but distinct, and KIAA0319L has not itself been linked to reading disabilities. Still, the family relationship raises the question of whether AAVR/KIAA0319L contributes to neurodevelopmental processes in ways that have not yet been tested. It is a thread worth watching, especially as more fine-grained phenotyping of KIAA0319L-deficient animals becomes available.
AAVs That Do Not Need AAVR
Although AAVR is essential for most AAV serotypes, a distinct lineage of capsids has found another way in. AAV4 and AAVrh32.33 can bind and transduce cells without AAVR entirely. These capsids do not interact with AAVR in binding assays, and AAVR knockout mice are just as susceptible to AAVrh32.33 as wild-type mice after systemic injection. Intriguingly, even some serotypes that normally depend on AAVR can manage a low level of transduction without it, hinting at a multimodal entry strategy for certain capsids.11PubMed Central. An Alternate Route for Adeno-associated Virus (AAV) Entry Independent of AAV Receptor
More recently, a second receptor called AAVR2 (also known as CPD) has been identified. AAVR2 specifically supports the transduction of clade E AAVs, including AAV8, AAVrh10, and AAVhu37. In cells where AAVR has been knocked out, overexpression of AAVR2 rescues transduction for those clade E serotypes but not for others.12Cell. AAVR2/CPD is an alternate receptor for adeno-associated virus This finding matters clinically because AAV8 is a leading candidate for liver-directed gene therapies. If some patients have reduced AAVR function but intact AAVR2, certain serotypes might still work where others would fail.
GPR108 and the Wider Entry Toolkit
AAVR is not the only host protein that matters for AAV entry. A protein called GPR108 has been identified as an entry factor required by more than 20 divergent AAV serotypes across all known clades, with one conspicuous exception: AAV5, which is unaffected by the loss of GPR108. In mice lacking GPR108, expression from AAV8 and AAVrh32.33 vectors drops by ten- to a hundred-fold. Mechanistically, cells without GPR108 show reduced nuclear import and virus particles accumulating in the cytoplasm rather than reaching the nucleus.13PubMed Central. GPR108 Is a Highly Conserved AAV Entry Factor The reason AAV5 escapes GPR108 dependence appears to trace to a region of the capsid’s VP1 domain that differs in AAV5 from other serotypes. When researchers swapped that region onto an AAV2 backbone, the chimeric capsid also became GPR108-independent.14Molecular Therapy. AAV Receptor Insights: Biological Role and Disease Links
The emerging picture is that AAV entry is not a simple lock-and-key event. It involves an initial sugar-mediated attachment, binding to AAVR (or in some cases AAVR2), retrograde trafficking through the Golgi, and GPR108-facilitated nuclear import. Each step represents a potential point of failure or a lever that researchers can pull to redirect a vector.
Human Genetic Variation in AAVR and What It Means for Patients
Because gene therapy depends on AAV getting into cells efficiently, natural variation in the AAVR gene could affect how well a treatment works from one patient to the next. A systematic survey of human KIAA0319L variants found hundreds of potentially deleterious missense mutations in public databases. Applying computational filters for likely functional impact narrowed this to around 119 mutations, of which 118 are rare and one is common.15Molecular Therapy: Methods & Clinical Development. Impact of natural human single nucleotide polymorphisms in AAVR on patient transduction variability
Laboratory experiments have started to pin down how specific variants affect transduction. The Lys3Thr variant cut AAV2-mediated gene expression by about half. Other variants affected some serotypes but not others in a strikingly selective pattern. For instance, the Gly6Arg variant reduced transduction by AAV8 and AAV9 by roughly 30% each but had no effect on AAVrh.10. The Pro302Gln variant reduced AAV8 transduction by about 30% but left AAV9 and AAVrh.10 untouched.16Molecular Therapy Advances. Consequences of human genetic variations in KIAA0319L, encoding adeno-associated virus receptor, on AAV-mediated gene transfer Some of these variants also lower AAVR protein levels. Seven out of eleven tested missense variants produced significantly less AAVR protein than wild-type, including Lys3Thr, Val396Leu, and several others affecting different domains of the protein.17Molecular Therapy. AAV Receptor Insights: Biological Role and Disease Links
This kind of serotype-specific sensitivity means that two patients receiving different AAV vectors could respond quite differently to the same AAVR mutation. It raises the possibility that genotyping AAVR could one day help guide the choice of which AAV serotype to use in a given patient, although that remains a research concept rather than a clinical reality.
Tissue-Level Differences in AAVR Expression
AAVR is not expressed equally across all cell types, and that unevenness helps explain why certain tissues are easy or difficult to transduce with AAV vectors. Muscle stem cells, for example, are among the cell types with the lowest endogenous AAVR expression in broad surveys of mouse tissues, which may contribute to the well-known difficulty of targeting these cells with standard AAV vectors.18Nature Methods. Hardwiring tissue-specific AAV transduction in mice through engineered receptor expression Researchers have shown that artificially boosting AAVR in those resistant cell types enhances transduction, essentially opening a door that was only slightly ajar.
On the flip side, cells with high AAVR expression tend to be the easy targets. The liver, which robustly expresses AAVR, is famously the primary destination for systemically administered AAV vectors. For therapies aimed at the brain or other organs, the liver’s eagerness to soak up AAV is a problem rather than a benefit, because most of the therapeutic dose gets trapped there instead of reaching the intended tissue.
Engineering Capsids to Change the AAVR Interaction
One strategy for steering vectors away from the liver and toward the brain exploits what researchers know about the AAVR binding interface. CAP-B10, an engineered capsid derived from AAV9, carries a seven-amino-acid substitution in a variable region that weakens its affinity for AAVR-PKD2. Where wild-type AAV9 binds AAVR-PKD2 with a dissociation constant of about 11 micromolar, CAP-B10’s affinity drops to around 61 micromolar. This weakened liver binding, combined with retained interactions with brain-expressed receptors, translates into reduced liver transduction and enhanced neurotropism.19Molecular Therapy. Structural insights into AAVR-mediated liver de-targeting and neurotropism of engineered AAV capsids The same substitution can be transferred to other AAV9-derived capsids to achieve similar liver de-targeting, suggesting this is a modular design principle rather than a one-off trick.
This approach essentially uses the structural map of the AAVR-capsid interface as a blueprint for rational engineering. By selectively loosening the grip on AAVR in the liver while preserving other entry pathways in the brain, gene therapists can potentially reduce off-target effects and lower the dose required to achieve a therapeutic effect in the central nervous system.
Evolutionary Variation Across Species
AAVR is not identical across mammals, and those differences can determine which AAV serotypes can infect which species. A phylogenetic comparison of AAVR across animals revealed that cats and their relatives carry a glutamate at position 353 in the PKD1 domain where humans have an arginine. This single amino acid difference is enough to disrupt binding of clade H AAVs, a group of serotypes that interacts primarily with PKD1.20PubMed Central. Evolutionary Analysis Reveals a Single Amino Acid in the AAV Entry Receptor (AAVR) of Cats That Disrupts Binding of a Major Phylogenetic Group of AAVs The finding underscores how sensitive the virus-receptor interface is to point mutations and complicates the use of AAV-based gene therapy in veterinary settings. A serotype that works well in mice or humans cannot be assumed to work in cats or other felids without checking receptor compatibility first.
From the virus’s perspective, this evolutionary variation may represent an arms race: host species accumulating receptor changes that block certain serotypes, while new AAV variants evolve to use different parts of the receptor or different receptors entirely. The existence of alternative pathways through AAVR2 and the AAVR-independent serotypes like AAV4 fits neatly into that picture.
An Unexpected Link to Systemic Sclerosis
Separate from gene therapy, AAV infection through particular receptors has been implicated in at least one autoimmune condition. Researchers investigating systemic sclerosis, a disease characterized by skin hardening and interstitial lung disease, found that AAV5 can infect cells through the receptor PDGFRα. The virus generates composite peptides that are recognized by agonistic immunoglobulins found in patients with systemic sclerosis but not in healthy controls.21PubMed. Adeno-Associated Virus Type 5 Infection via PDGFRα Is Associated With Interstitial Lung Disease in Systemic Sclerosis and Generates Composite Peptides and Epitopes Recognized by the Agonistic Immunoglobulins Present in Patients With Systemic Sclerosis This does not establish that AAV5 causes systemic sclerosis, but it suggests that prior AAV infection, acting through receptors other than AAVR, could contribute to autoimmune disease through molecular mimicry or altered self-antigen presentation. It is a reminder that understanding AAV receptor biology reaches beyond gene therapy and into natural infection and disease pathogenesis.