Multiple roles for sialylated glycans in determining the cardiopulmonary tropism of adeno-associated virus 4.

Adeno-associated virus 4 (AAV4) is one of the most divergent serotypes among known AAV isolates. Mucins or O-linked sialoglycans have been identified as the primary attachment receptors for AAV4 in vitro. However, little is known about the role(s) played by sialic acid interactions in determining AAV4 tissue tropism in vivo. In the current study, we first characterized two loss-of-function mutants obtained by screening a randomly mutated AAV4 capsid library. Both mutants harbored several amino acid residue changes localized to the 3-fold icosahedral symmetry axes on the AAV4 capsid and displayed low transduction efficiency in vitro. This defect was attributed to decreased cell surface binding as well as uptake of mutant virions. These results were further corroborated by low transgene expression and recovery of mutant viral genomes in cardiac and lung tissue following intravenous administration in mice. Pharmacokinetic analysis revealed rapid clearance of AAV4 mutants from the blood circulation in conjunction with low hemagglutination potential ex vivo. These results were recapitulated with mice pretreated intravenously with sialidase, directly confirming the role of sialic acids in determining AAV4 tissue tropism. Taken together, our results support the notion that blood-borne AAV4 particles interact sequentially with O-linked sialoglycans expressed abundantly on erythrocytes followed by cardiopulmonary tissues and subsequently for viral cell entry.

Engraftment of a galactose receptor footprint onto adeno-associated viral capsids improves transduction efficiency.

New viral strains can be evolved to recognize different host glycans through mutagenesis and experimental adaptation. However, such mutants generally harbor amino acid changes that affect viral binding to a single class of carbohydrate receptors. We describe the rational design and synthesis of novel, chimeric adeno-associated virus (AAV) strains that exploit an orthogonal glycan receptor for transduction. A dual glycan-binding AAV strain was first engineered as proof of concept by grafting a galactose (Gal)-binding footprint from AAV serotype 9 onto the heparan sulfate-binding AAV serotype 2. The resulting chimera, AAV2G9, continues to bind heparin affinity columns but interchangeably exploits Gal and heparan sulfate receptors for infection, as evidenced by competitive inhibition assays with lectins, glycans, and parental AAV strains. Although remaining hepatotropic like AAV2, the AAV2G9 chimera mediates rapid onset and higher transgene expression in mice. Similarly, engraftment of the Gal footprint onto the laboratory-derived strain AAV2i8 yielded an enhanced AAV2i8G9 chimera. This new strain remains liver-detargeted like AAV2i8 while selectively transducing muscle tissues at high efficiency, comparable with AAV9. The AAV2i8G9 chimera is a promising vector candidate for targeted gene therapy of cardiac and musculoskeletal diseases. In addition to demonstrating the modularity of glycan receptor footprints on viral capsids, our approach provides design strategies to expand the AAV vector toolkit.

Glycan binding avidity determines the systemic fate of adeno-associated virus type 9.

Glycans are key determinants of host range and transmissibility in several pathogens. In the case of adeno-associated viruses (AAV), different carbohydrates serve as cellular receptors in vitro; however, their contributions in vivo are less clear. A particularly interesting example is adeno-associated virus serotype 9 (AAV9), which displays systemic tropism in mice despite low endogenous levels of its primary receptor (galactose) in murine tissues. To understand this further, we studied the effect of modulating glycan binding avidity on the systemic fate of AAV9 in mice. Intravenous administration of recombinant sialidase increased tissue levels of terminally galactosylated glycans in several murine tissues. These conditions altered the systemic tropism of AAV9 into a hepatotropic phenotype, characterized by markedly increased sequestration within the liver sinusoidal endothelium and Kupffer cells. In contrast, an AAV9 mutant with decreased glycan binding avidity displayed a liver-detargeted phenotype. Altering glycan binding avidity also profoundly affected AAV9 persistence in blood circulation. Our results support the notion that high glycan receptor binding avidity appears to impart increased liver tropism, while decreased avidity favors systemic spread of AAV vectors. These findings may not only help predict species-specific differences in tropism for AAV9 on the basis of tissue glycosylation profiles, but also provide a general approach to tailor AAV vectors for systemic or hepatic gene transfer by reengineering capsid-glycan interactions.

Intra- and inter-subunit disulfide bond formation is nonessential in adeno-associated viral capsids.

The capsid proteins of adeno-associated viruses (AAV) have five conserved cysteine residues. Structural analysis of AAV serotype 2 reveals that Cys289 and Cys361 are located adjacent to each other within each monomer, while Cys230 and Cys394 are located on opposite edges of each subunit and juxtaposed at the pentamer interface. The Cys482 residue is located at the base of a surface loop within the trimer region. Although plausible based on molecular dynamics simulations, intra- or inter-subunit disulfides have not been observed in structural studies. In the current study, we generated a panel of Cys-to-Ser mutants to interrogate the potential for disulfide bond formation in AAV capsids. The C289S, C361S and C482S mutants were similar to wild type AAV with regard to titer and transduction efficiency. However, AAV capsid protein subunits with C230S or C394S mutations were prone to proteasomal degradation within the host cells. Proteasomal inhibition partially blocked degradation of mutant capsid proteins, but failed to rescue infectious virions. While these results suggest that the Cys230/394 pair is critical, a C394V mutant was found viable, but not the corresponding C230V mutant. Although the exact nature of the structural contribution(s) of Cys230 and Cys394 residues to AAV capsid formation remains to be determined, these results support the notion that disulfide bond formation within the Cys289/361 or Cys230/394 pair appears to be nonessential. These studies represent an important step towards understanding the role of inter-subunit interactions that drive AAV capsid assembly.

Engineering liver-detargeted AAV9 vectors for cardiac and musculoskeletal gene transfer.

We report the generation of a new class of adeno-associated virus serotype 9 (AAV9)-derived vectors displaying selective loss of liver tropism and demonstrating potential for cardiac and musculoskeletal gene transfer applications. Random mutagenesis of residues within a surface-exposed region of the major AAV9 capsid protein yielded a capsid library with mutations clustered at the icosahedral threefold symmetry axis. Using a combination of sequence analysis, structural models, and in vivo screening, we identified several functionally diverse AAV9 variants. The latter were classified into three functional subgroups, with respect to parental AAV9 displaying: (i) decreased transduction efficiency across multiple tissues; (ii) a selective decrease in liver transduction, or (iii) a similar transduction profile. Notably, variants 9.45 and 9.61 (subgroup II) displayed 10- to 25-fold lower gene transfer efficiency in liver, while transducing cardiac and skeletal muscle as efficiently as AAV9. These results were further corroborated by quantitation of vector genome copies and histological analysis of reporter (tdTomato) gene expression. The study highlights the feasibility of generating AAV vectors with selectively ablated tissue tropism, which when combined with other targeting strategies could allow sharply segregated gene expression. Liver-detargeted AAV9 variants described herein are excellent candidates for preclinical evaluation in animal models of cardiac and musculoskeletal disease.

Structural and functional divergence within the Dim1/KsgA family of rRNA methyltransferases.

The enzymes of the KsgA/Dim1 family are universally distributed throughout all phylogeny; however, structural and functional differences are known to exist. The well-characterized function of these enzymes is to dimethylate two adjacent adenosines of the small ribosomal subunit in the normal course of ribosome maturation, and the structures of KsgA from Escherichia coli and Dim1 from Homo sapiens and Plasmodium falciparum have been determined. To this point, no examples of archaeal structures have been reported. Here, we report the structure of Dim1 from the thermophilic archaeon Methanocaldococcus jannaschii. While it shares obvious similarities with the bacterial and eukaryotic orthologs, notable structural differences exist among the three members, particularly in the C-terminal domain. Previous work showed that eukaryotic and archaeal Dim1 were able to robustly complement for KsgA in E. coli. Here, we repeated similar experiments to test for complementarity of archaeal Dim1 and bacterial KsgA in Saccharomyces cerevisiae. However, neither the bacterial nor the archaeal ortholog could complement for the eukaryotic Dim1. This might be related to the secondary, non-methyltransferase function that Dim1 is known to play in eukaryotic ribosomal maturation. To further delineate regions of the eukaryotic Dim1 critical to its function, we created and tested KsgA/Dim1 chimeras. Of the chimeras, only one constructed with the N-terminal domain from eukaryotic Dim1 and the C-terminal domain from archaeal Dim1 was able to complement, suggesting that eukaryotic-specific Dim1 function resides in the N-terminal domain also, where few structural differences are observed between members of the KsgA/Dim1 family. Future work is required to identify those determinants directly responsible for Dim1 function in ribosome biogenesis. Finally, we have conclusively established that none of the methyl groups are critically important to growth in yeast under standard conditions at a variety of temperatures.

Recognition of a complex substrate by the KsgA/Dim1 family of enzymes has been conserved throughout evolution.

Ribosome biogenesis is a complicated process, involving numerous cleavage, base modification and assembly steps. All ribosomes share the same general architecture, with small and large subunits made up of roughly similar rRNA species and a variety of ribosomal proteins. However, the fundamental assembly process differs significantly between eukaryotes and eubacteria, not only in distribution and mechanism of modifications but also in organization of assembly steps. Despite these differences, members of the KsgA/Dim1 methyltransferase family and their resultant modification of small-subunit rRNA are found throughout evolution and therefore were present in the last common ancestor. In this paper we report that KsgA orthologs from archaeabacteria and eukaryotes are able to complement for KsgA function in bacteria, both in vivo and in vitro. This indicates that all of these enzymes can recognize a common ribosomal substrate, and that the recognition elements must be largely unchanged since the evolutionary split between the three domains of life.