A general approach for label-free, ELISA-like detection of small molecules
$500,000 over two years
Liangcai Gu, Assistant Professor, Biochemistry
Frank DiMaio, Assistant Professor, Biochemistry
The analysis of small molecule metabolites and drugs in body fluids and tissue extracts plays a vital role in the fields of human health, food safety and environmental monitoring. Small molecule detection is traditionally associated with complex, time- and resource-consuming technologies, such as mass spectrometry. In contrast, antibody-based approaches, such as enzyme-linked immunosorbent assays (ELISAs), can be performed in almost any condition, with easily interpretable results available within a few hours. However, antibodies against small molecules are difficult to obtain by animal immunization because small molecules, by themselves, are non-immunogenic, and can only elicit antibodies upon conjugation to protein carriers. In addition, the detection assays require the labeling of small molecules—i.e., chemical linking of small molecule targets or their competitive inhibitors to solid support or reporter molecules—and thus, are not generally applicable because some modifications are difficult to chemically synthesize or can affect small molecule binding activity. Here, to overcome these limitations, we propose a general approach based on chemically-induced dimerization (CID) systems, in which two proteins dimerize only in the presence of a small molecule. Naturally occurring CID systems, such as rapamycin-inducible FKBP/FRB and gibberellin-inducible GAI/GID1 complexes, have been widely used as biosensors, transcriptional regulators, small molecule-gated therapeutics, etc. Nevertheless, a robust methodology to create CID systems with desired affinity and specificity for a specific small molecule remains an unsolved problem in the field of protein engineering. Indeed, so far there is no previous report of de novo engineered CID other than natural CID and derivatives. De novo engineering of new CID systems has been hampered by the barriers of computational protein design and high-throughput screening. First, it is highly challenging to design higher-order interactions involving a small molecule and two proteins, because protein-protein and protein-small molecule interacting surfaces, often undergoing significant conformational changes in CID systems, cannot be properly modeled by current computational methods. Secondly, the engineering of two proteins in a CID complex requires the screening of two large variant libraries to search a two-dimensional matrix of possible combinations (‘library-by-library’ screening), which is not affordable with conventional high-throughput approaches. Overcoming any of these barriers will fundamentally advance CID engineering.
We recently developed a single-molecular-interaction sequencing (SMI-seq) technology (Gu, et al., Nature, 2014) for large-scale ‘library-by-library’ protein-protein interaction (PPI) profiling in a single solution. SMI-seq can identify CID binder pairs by quantifying PPI changes between two binder libraries when titrated with small molecules. Two strategies will be applied to engineer CID systems: i) targeted screening of computationally designed binder libraries, and ii) random screening of vastly diverse binder libraries (>10^9), in particular, combinatorial synthetic single-domain antibody (or nanobody) libraries. We will assess the success rates, turnaround times, and cost-effectiveness of both strategies by testing important drugs and metabolites, all of which have unmet needs for label-free, in-solution detection. Finally, selected CID systems will be validated to be next-generation ELISA reagents to measure the drugs in blood samples. This two-year project will provide an affordable approach to creating CID systems for broad applications in research, diagnostics and theranostics.