Secrete, Capture, Sort, Sequence: NGS Decoded Molecular Recognition Pairs

尽管基于单克隆抗体的技术占主导地位，但单克隆抗体的方法仍然存在。 抗体的产生和传播，以及IgG的大小和结构， 其用途有限，特别是用于研究目的。缺乏特异性和文件记录导致 实验再现性的危机，导致美国每年浪费约3.5亿美元的研究经费。 通过系统地产生测序的、经验证的单克隆抗体来纠正这一点， 每种抗体5万美元，对许多靶标都无效。 单克隆抗体的局限性将通过开发酵母“分泌和捕获”共展示来解决 用于重组非免疫纳米抗体（NB）的高通量分离和改进的系统， 衍生自骆驼科动物抗体VHH结构域的蛋白质亲和试剂。基于荧光的FACS技术， 定量展示的NB和捕获的靶蛋白结构域（TPD）将与下一代整合， 测序（NGS），其鉴定相关复合物。整合将极大地扩展 生物靶标，可以针对其分离同源NB，并促进创建聚焦NB工具包。 目前的非免疫支架筛选使用纯化的靶蛋白来分离候选结合物， 物理克隆和单独评估。这一资源密集型办法将由以下办法取代： (1)流式细胞仪报告分析，定量报告共表达的NB和TPD的相互作用， 根据特异性、亲和性和动力学，从而避免使用纯化的蛋白质。该分析基于融合 将表面展示的NB和分泌的TPD转化为光谱上不同的荧光团激活蛋白（FAP）， 当结合非荧光染料（荧光团）时发出荧光;荧光团可以通过PEG作为“扎染剂”灵活地连接。 可裂解的扎染剂用于稳定和报道分泌的TPD和同源物的细胞表面复合物， 在裂解后，从细胞表面释放TPD-FAP的动力学分析允许人们估计 NB/TPD亲和力。 (2)一种物理融合共表达的NB和TPD的编码基因型的方法， 分析以将复杂群体的FACS测定解析为单个克隆的结合表型，因此 消除了物理克隆的需要。在分别编码NB和NB的酵母文库的大量交配后， 为了在单独的质粒上产生分泌的TPD，CRISPR将用于迫使编码以下的序列的大量重组： NB和识别相关联的TPD的条形码到单个NGS可解码的读帧中。 (3)以生物信息学衍生的TPD查询集的形式进行多重筛选，以直接获得多组 相关试剂，从而最大限度地减少了连续评估单个克隆的需要。查询将用于隔离 NB探测以前很难接近的生物功能;我们的测试用例将是：（i） 神经连接蛋白剪接同种型;（ii）神经连接蛋白/神经连接蛋白复合物;和（iii）细菌表面蛋白胞外域。

Despite the predominance of monoclonal antibody based technologies, the means by which monoclonal antibodies are generated and disseminated, together with the size and structure of IgGs, impose fundamental limitations on their usefulness, especially for research purposes. Poor specificity and documentation has led to a crisis in experimental reproducibility, leading to an annual waste of ~$350 million in US research expenditures. Rectifying this by the systematic generation of sequenced, validated monoclonals would cost an estimated $50,000 per antibody and would fail for many targets. Limitations of monoclonals will be addressed by developing a yeast `secrete and capture' co-display system for the high throughput isolation and improvement of recombinant nonimmune Nanobodies (NBs), small protein affinity reagents derived from camelid antibody VHH domains. Fluorogen-based FACS technology that quantifies displayed NBs and captured target protein domains (TPDs) will be integrated with next generation sequencing (NGS) that identifies the associated complex. Integration will greatly expand the repertoire of biological targets for which cognate NBs may be isolated, and facilitate the creation of focused NB toolkits. Current nonimmune scaffold screens use purified target protein to isolate candidate binders that are physically cloned and individually evaluated. This resource-intensive approach will be replaced by the following: (1) A FACS reporter assay that quantitatively reports the interaction of co-expressed NB and TPDs in terms of specificity, affinity and kinetics, thus avoiding the use of purified protein. The assay is based on fusing surface-displayed NBs and secreted TPDs to spectrally distinct fluorogen activating proteins (FAPs) that fluoresce when binding non-fluorescent dyes (fluorogens); fluorogens may flexibly linked by PEG as a `tie-dye'. A cleavable tie-dye is used to stabilize and report on a cell surface complex of secreted TPD and cognate displayed NB; upon cleavage, kinetic analysis of TPD-FAP release from the cell surface allows one to estimate NB/TPD affinity. (2) A method that physically fuses the encoded genotypes of co-expressed NB and TPD, enabling NGS analysis to resolve FACS assays of complex populations into the binding phenotypes of individual clones, thus eliminating the need for physical cloning. After mass mating of yeast libraries respectively encoding NBs and secreted TPDs on separate plasmids, CRISPR will be used to force bulk recombination of sequences encoding the NB and a barcode that identifies the associated TPD into a single NGS decodable read frame. (3) Multiplexed screens in the form of bioinformatics derived TPD query sets to directly obtain groups of related reagents, thus minimizing the need to serially evaluate individual clones. Queries will be used to isolate NBs that probe biological functionalities that were previously very difficult to approach; our test cases will be: (i) neuroligin splice isoforms; (ii) neuroligin/neurexin complexes; and (iii) bacterial surface protein ectodomains.