This study provides evidence that AAV-mediated gene transfer of the human anti-gp120 b12 minibody to primary cervical and vaginal epithelial cells can protect against virus challenge in vitro
. To date, there is no highly effective microbicide against HIV-1 infection. Although a recent trial showed that a tenofovir containing vaginal gel could reduce HIV-1 infection rates by 39% 
, daily application of the gel or before and after sexual acts was impractical for many users and adherence to the use of the gel dropped over time. Such behavior-related issues may be avoided by delivering anti-HIV agents with more durable activity. This present study provides proof-of-principle for a novel microbicide strategy against HIV-1 utilizing an AAV vector, that potentially offers long-term stable transduction of cervico-vaginal stem cells with secretion of a potent and broadly neutralizing anti-HIV gp120 minibody.
In preclinical animal models of human disease, AAV vectors have emerged as a favored gene transfer system due to their safety profile and the potential to transduce non-dividing cells. To date there have been 12 different serotypes of AAV isolated from human and non-human primates 
. Different AAV serotypes are capable of transducing a wide variety of tissue types, including muscle, lung, brain and eye 
. However, specific AAV transduction of mucosal tissues in the female genital tract has not yet, to our knowledge, been published. In the present study, AAV-GFP gene transfer studies were performed on immortalized human endocervical, ectocervical and vaginal epithelial cell lines as well as huPGEC. Among the 8 serotypes tested, AAV-2 and 6 were the most efficient for transduction of these cells (). In this study, AAV-6 was chosen over AAV-2 serotype because AAV-6 (a hybrid of AAV-1 and AAV-2) has shown lower immunogenicity than AAV-2 
, and so may provide significant advantages over AAV-2 for gene transfer to female genital epithelial stem cells.
The choice of utilizing the b12 mAb in this study was based on evidence from both in vitro
and in vivo
studies. The in vitro
studies demonstrated that both b12 IgG1and IgA2 are able to inhibit transfer of cell-free HIV-1 to ME-180 cells and are able to block viral attachment to and uptake by epithelial cells 
. Additional evidence from macaque models indicate that the broadly neutralizing antibody b12 IgG1 is capable of conferring protection against SHIV infection when administered by intravenous or intravaginal (topical) routes 
. In the present study b12 antibody was produced in the scFvFc (minibody) format as opposed to full-length IgG for AAV gene transfer for several reasons. First, as AAV vectors do not package more than ~5 kb of foreign DNA efficiently, the AAV vector cannot accommodate conventional antibody expression cassettes to drive the mAb heavy chains and light chains from two individual promoters 
although, 2A sequences have been used successfully to express full-length IgG 
. Second, the minibodies are generally expressed at high levels in mammalian cells, and third, the smaller size of the minibodies may allow greater tissue penetration and targeting of neutralizing epitopes that may be inaccessible or sterically hindered as a whole IgG 
The generated b12 minibodies were confirmed to be as potent as the full-length b12 IgG both in their capacity to bind to HIV-1 gp120 and to neutralize HIV-1bal
virus (). Furthermore, b12 minibodies were able to block HIV-1 transfer and infectivity of cell-free virus transferred through VEC tissues ( and ). These data are in agreement with previously published studies 
, which showed that b12 IgG inhibited the transfer of cell free HIV-1 through the human cervical epithelial cell line ME-180 and that inhibition was due to the ability of IgG b12 to block both viral attachment to and uptake by epithelial cells. In addition, studies on rhesus macaques 
demonstrated that b12 IgG applied vaginally can afford protection against SHIV challenge. Thus, the data clearly demonstrate that the activity of b12 minibodies is comparable to the full-length b12 antibodies in their capacity to inhibit in vitro
transfer of cell free HIV-1 through human VEC tissues. The findings also strongly suggest that b12 minibodies should protect the vaginal mucosa of rhesus macaque against SHIV challenge in planned future studies.
The present study also demonstrates that primary genital epithelial stem cells could be transduced in vitro
. This same strategy should work effectively in vivo
if the AAV vector has ready access and sufficient time to transduce the epithelial stem cell populations. There are two distinct epithelial stem cell populations that differ in morphology and location. The endocervix is naturally thin, and there is only a one cell thick layer above the epithelial stem cells (). We propose that this population of columnar epithelial stem cells may be the most straightforward and efficient to transduce in vivo
. In contrast, the in vivo
transduction of squamous epithelial stem cells of the vaginal and ectocervix could be less efficient due to their thick layers (), numerous tight junctions 
and their continuous shedding. It could be challenging for the AAV vectors to reach the ectocervical and vaginal epithelial stem cells located in the basal layer. To overcome this, AAV vectors could be applied during the secretory phase of the menstrual cycle when mucosa is dominated by progesterone and the cervico-vaginal layer is at its thinnest 
or alternatively the vector could be applied with a gene gun 
, or on gently abraided tissue. These approaches could facilitate AAV vector penetration and transduction of epithelial stem cells in the basal layer. To achieve stable and durable gene transfer in vivo
and to create a milieu rich in BnAbs, transduction of the genital epithelial stem cells in both endocervical and ectocervical/vaginal locations may be essential for replenishing a clinically relevant number of b12 BnAb secreting cells.
In summary, this study represents a novel HIV-1 microbicide strategy that is both feasible and attractive in light of the absence of a highly effective prophylactic HIV-1 vaccine. The results provide proof of concept that AAV-BnAb gene transfer to primary cervical and vaginal epithelial cells can protect against HIV-1 challenge in vitro.
Importantly, demonstration of AAV mediated transduction of cervico-vaginal epithelial stem cells highlights the potential of this work as a means to convey durable and long-lasting expression of BnAbs against HIV-1 in the female genital tract. Further studies are required to examine the effectiveness of this strategy in primate models, including testing additional BnAbs and syndecan decoys such as those recently reported 
before moving this approach into human clinical trials. If clinically relevant levels of anti-HIV biological agents can be achieved in vivo
, this approach may not only provide protection against HIV-1 infection but a means by which local neutralization escape, a hallmark of HIV-1 pathogenesis, can be inhibited.