Directed Evolution of Adeno-Associated Virus for Retinal Gene Therapy

描述（由申请人提供）：基因疗法在治疗几种导致失明的单基因缺陷方面越来越成功。特别是，针对Leber's先天性黑朦2型（LCA2）的多项成功临床试验利用了一种已有25年历史的病毒传递载体，基于腺相关病毒（AAV）血清2型，将rpe65基因的功能拷贝传递到视网膜色素上皮（RPE）。这些试验在提高30多名患者的视觉功能方面取得了里程碑式的进展，成功证明了如果在一组患者中可以识别出致病基因，则可以将功能性替代基因包装并与AAV一起安全交付。然而，随着视网膜退行性疾病的大多数突变现在已经被确定，几乎所有的突变都编码光受体特异性转录物，这使得光受体成为视网膜基因治疗的主要靶点。此外，许多这些突变是常染色体显性的，因此基因替代策略不合适。在LCA2试验成功的基础上，至少有两个阻碍视网膜基因治疗更广泛应用的主要障碍必须克服。首先，基于天然AAV变体的载体需要载体的视网膜下介导基因传递到光感受器或RPE，伴随着视网膜脱离，在光感受器和潜在的RPE之间产生“气泡”。这种手术损害视网膜，可能加剧视网膜变性，并可诱发反应性神经胶质瘤。此外，视网膜下注射将治疗效果限制在水泡区域，超过该区域AAV不会扩散。从玻璃体中传递基因将大大减少创伤，并提供泛视网膜转导的潜力，这两者都代表着重大的进步。由于没有天然AAV血清型可以在小鼠或非人灵长类动物（NHP）模型中转导玻璃体中的光感受器，因此我们开发并实施了一种定向进化方法，正如我们最近发表的那样，已经产生了一种能够从小鼠玻璃体中转导光感受器的新型AAV，并在一定程度上在NHP视网膜中转导光感受器。我们现在建议在这一成功的基础上，设计AAV变体，以获得最佳的治疗基因递送到NHP视网膜。光感受器基因治疗的第二个问题是许多视网膜变性是常染色体显性的。虽然RNAi可以产生病理等位基因的部分敲除，但这些基因的完全消融是可取的。最近在位点特异性DNA核酸酶的开发方面取得了进展，这种核酸酶可以敲除目标基因，我们将在这些进展的基础上敲除导致视网膜变性的显性等位基因。因此，我们提出了一种独特的分子病毒学，蛋白质工程和翻译重要的动物模型的混合，以设计增强的遗传传递系统和货物，用于治疗人类视网膜疾病。

DESCRIPTION (provided by applicant): Gene therapy has been increasingly successful in treating several single-gene defects that cause blindness. In particular, multiple successful clinical trials for Leber's congenital amaurosis type 2 (LCA2) have utilized a 25 year-old viral delivery vehicle, based on adeno-associated virus (AAV) serotype 2, to deliver a functional copy of the rpe65 gene to the retinal pigment epithelium (RPE). These trials have taken landmark strides in enhancing visual function in over 30 patients, success that has established the proof of concept that if a causative gene can be identified in a group of patients, a functional replacement gene can be packaged and safely delivered with AAV. However, as the majority of mutations underlying retinal degenerative diseases have now been identified, it has become clear that almost all encode photoreceptor-specific transcripts, establishing photoreceptors as the primary target for retinal gene therapy. Furthermore, many of these mutations are autosomal dominant, such that gene replacement strategies are not suitable. To build upon the LCA2 trial successes, at least two major hurdles that impede broader application of retinal gene therapy must thus be overcome. First, vectors based on natural AAV variants require a subretinal of the vector to mediate gene delivery to photoreceptors or RPE, with accompanying retinal detachment with the creation of a "bleb" between the photoreceptors and underlying RPE. This procedure damages the retina, may exacerbate the retinal degeneration, and can induce reactive gliosis. In addition, subretinal injection limits the therapeutic effect to the area of th bleb, beyond which the AAV does not spread. Gene delivery from the vitreous would be considerably less traumatic and would offer the potential for pan-retinal transduction, both of which would represent significant advances. Since no natural AAV serotypes can transduce the photoreceptors from the vitreous in either murine or non-human primate (NHP) models, we developed and implemented a directed evolution approach that, as we have recently published, has yielded a novel AAV capable of photoreceptor transduction from the vitreous in the murine and to an extent in the NHP retina. We now propose to build upon this success and engineer AAV variants for optimal therapeutic gene delivery to the NHP retina. A second problem with photoreceptor gene therapy is that many retinal degenerations are autosomal dominant. While RNAi can yield a partial knockdown of pathological alleles, a full ablation of such genes would be desirable. There have been recent advances in the development of site-specific DNA nucleases that can knock out target genes, and we will build upon these advances to knock out dominant alleles that underlie retinal degeneration. We thus propose a unique blend of molecular virology, protein engineering, and a translationally important animal model to engineer enhanced genetic delivery systems and cargo for treating human retinal disease.