Design of antibody fragments specific for amyloidogenic aggregates

描述（申请人提供）：生物分子研究中的一个重大挑战是合理设计以高亲和力和特异性结合靶抗原的抗体。本提案的目标是阐明设计抗体片段的互补决定区（CDR）以介导具有构象和序列特异性的聚集蛋白的识别的原理。我们的提议是基于我们最近的发现，即单结构域抗体可以被设计为使用介导与阿尔茨海默病相关的Aβ聚集的相同分子相互作用来识别Aβ42寡聚体和原纤维（Perchiacca et al.，PNAS，2012）。我们发现疏水性Aβ肽可以被接枝到单个CDR环中，并且所得的接枝淀粉样基序抗体（γ抗体）以纳摩尔亲和力结合Aβ寡聚体和原纤维。基于这些发现，我们认为可以根据它们的相对淀粉样蛋白原性预测能够介导γ抗体结合的其他Aβ肽。我们还假设，γ抗体的结合亲和力在小到中等CDR长度时最大，这些CDR长度足够长以展示Aβ自我识别肽，但不足以由于熵增加而不利于结合。此外，我们认为我们的设计方法不仅限于Aβ，还可以扩展到其他淀粉样蛋白多肽，包括IAPP（2型糖尿病）和α-突触核蛋白（帕金森病）。最后，我们假设甚至更高亲和力的γ抗体可以通过将多个淀粉样蛋白生成肽接枝到反向平行的CDR中来设计，所述反向平行的CDR以与原纤维的生长（模板化）末端处的相应肽相同的方式取向。因此，在目标1中，我们提出确定CDR 3的长度和序列如何影响两种Aβ γ抗体（Aβ15-24和Aβ33-42）的结合亲和力和特异性。然后，在目标2中，我们建议评估我们对其他Aβ肽的预测，这些肽在移植到CDR 3中时介导γ抗体与Aβ聚集体的结合。接下来，在目标3中，我们建议扩展目标1和2中进行的分析，以评估我们对来自其他两种淀粉样蛋白形成多肽（α-突触核蛋白和IAPP）的肽段的预测，这些肽段在移植到CDR 3中时介导γ抗体与其相应的聚集构象体结合。最后，在目标4中，我们提出评估是否可以通过将两种不同的淀粉样蛋白生成肽移植到反向平行CDR中以匹配原纤维生长末端处相应肽的方向来增加目标1-3中开发的Aβ和IAPP γ抗体的亲和力。我们的研究的一个重要成果将是阐明如何自我互补，淀粉样蛋白生成肽可用于介导抗体抗原识别。我们期望我们的发现将导致设计类似的单域和多域抗体的规则，这些抗体对不同的淀粉样蛋白具有特异性，包括与人类聚集性疾病如亨廷顿病和朊病毒病有关的蛋白。

DESCRIPTION (provided by applicant): A grand challenge in biomolecular research is to rationally design antibodies that bind to target antigens with high affinity and specificity. The goal of this proposal is to elucidate principles for designing the complementarity-determining regions (CDRs) of antibody fragments to mediate recognition of aggregated proteins with conformational and sequence specificity. Our proposal is based on our recent discovery that single-domain antibodies can be designed to recognize Aβ42 oligomers and fibrils using the same molecular interactions mediating Aβ aggregation associated with Alzheimer's disease (Perchiacca et al., PNAS, 2012). We find that hydrophobic Aβ peptides can be grafted into a single CDR loop, and the resulting Grafted AMyloid-Motif AntiBODIES (gammabodies) bind to Aβ oligomers and fibrils with nanomolar affinity. Based on these discoveries, we posit that additional Aβ peptides capable of mediating gammabody binding can be predicted based on their relative amyloidogenicity. We also hypothesize that the binding affinity of gammabodies will be maximal at small to intermediate CDR lengths that are sufficiently long to display Aβ self-recognition peptides but not long enough to disfavor binding due to increased entropy. In addition, we posit that our design approach is not limited to Aβ and can be extended to other amyloidogenic polypeptides, including IAPP (type 2 diabetes) and α-synuclein (Parkinson's disease). Finally, we hypothesize that even higher-affinity gammabodies can be designed by grafting multiple amyloidogenic peptides into anti-parallel CDRs that are oriented in the same manner as the corresponding peptides at the growing (templating) ends of fibrils. Therefore, in Aim 1, we propose to determine how the length and sequence of CDR3 impacts the binding affinity and specificity for two Aβ gammabodies (Aβ15-24 and Aβ33-42). Then, in Aim 2, we propose to evaluate our predictions of additional Aβ peptides that mediate gammabody binding to Aβ aggregates when grafted into CDR3. Next, in Aim 3, we propose to extend the analysis performed in Aims 1 and 2 to evaluate our predictions of peptide segments from two other amyloid-forming polypeptides (α-synuclein and IAPP) that mediate gammabody binding to their corresponding aggregated conformers when grafted into CDR3. Finally, in Aim 4, we propose to evaluate whether the affinity of the Aβ and IAPP gammabodies developed in Aims 1-3 can be increased by grafting two different amyloidogenic peptides into anti-parallel CDRs to match the orientation of the corresponding peptides at the growing ends of fibrils. A significant outcome of our studies will be the elucidation of how self-complementary, amyloidogenic peptides can be used to mediate antibody-antigen recognition. We expect that our findings will lead to rules for designing of similar single- and multidomain antibodies with specificity for diverse amyloidogenic proteins, including those linked to human aggregation disorders such as Huntington's and prion diseases.