Cysteines are known to play critical roles in the assembly of human papillomavirus 
, portal ring formation in herpes simplex virus 
, hexameric capsomer stabilization in HIV 
and infectivity of SV40 virions 
. The current study is focused on investigating the role of cysteines in disulfide bond formation within AAV2 capsids. Of the five cysteines tested, VP subunits containing C230S and C394S mutations are rapidly degraded by the proteasomal machinery within the host cell. Whether these mutant VP subunits exist in monomeric or oligomeric form remains to be determined.
Interpretation of the individual role(s) played by each cysteine at the 230 or 394 position proved exceedingly difficult. The observation that a Cys-to-Val substitution can be tolerated at the 394 position, but not the 230 position suggests that disulfide bond formation by the Cys230/394 pair might be nonessential. Further, examining the secondary structure of the VP subunit reveals that Cys230 is located at the beginning of the βA strand in the VP subunit, while Cys394 is located between βE and βF strands 
. Thus, it is plausible that mutagenesis of the Cys230 residue is more likely to disrupt secondary structure than mutations at the 394 position. It is also noteworthy to mention that mutagenesis of neighboring residues, namely, Trp228, His229 and His231 has been proposed to yield an assembly-deficient phenotype 
(Pulicherla and Asokan, unpublished). However, this interpretation is confounded by the fact that mutations in open reading frame ORF1, which encodes capsid protein subunits also results in mutations within ORF2, which encodes the assembly-activating protein (AAP) 
. Thus, it is likely that mutations within this region of AAP could also adversely affect AAV capsid assembly. Interestingly, complementation assays with wild type AAP are unable to rescue capsid assembly suggesting that the C230S mutant might indeed affect capsid subunit stability (Supporting Information S1
). However, the possibility that mutant AAP might exert a transdominant negative phenotype cannot be ignored. Additional mutagenesis studies that might help understand interactions of Cys394 with residues in the neighboring VP subunit are difficult to carry out due to overlap between AAP and this capsid domain. Nevertheless, the latter results highlight the importance of understanding AAP-capsid protein interactions.
Why is the Cys230/394 pair unlikely to form a disulfide bond? A closer examination of the structural environment at the five-fold interface reveals that the Cys230/394 pair is nestled in a hydrophobic environment created by Phe398, Tyr397 and Trp228 residues (Supporting Information S1
). The buried environment in proteins has been shown to significantly affect the pKa values and hydrogen bonding potential of cysteine thiol side chains 
. Therefore, it is conceivable that the buried cysteines at the AAV capsid pentamer interface have high pKa values and remain protonated without forming disulfide bonds. This is further corroborated by the observation that the C394V mutation is well tolerated and does not adversely affect AAV capsid formation. Nevertheless, it is plausible that transient disulfide formation within the Cys289/361 or the Cys230/394 pair might occur in the event that conformational changes result in surface exposure of one or both cysteine thiol side chains. In summary, our results support a model wherein inter- as well as intra-subunit disulfide bond formation between neighboring cysteine residues appears nonessential for AAV capsid formation. These studies are a step towards understanding the nature of subunit interactions that drive AAV capsid assembly.