Objective
Enzymes are important targets in diagnostics and drug discovery due to their involvement in etiology of most human diseases. Many diseases are diagnosed by quantification of disease-related enzymes in biological samples while others are treated by drugs that inhibit disease-related enzymes. Currently, the search of such inhibitory molecules is one of the major scopes of drug discovery programs. To find these inhibitors, small-molecule libraries are commonly screened by methods based on displacement of fluorescent active site probe. However, the limit of these methods is the requirement of large amounts of purified enzyme and high false positive rates that limits the ability of compound pooling.
Recently, we developed a novel assay called DIANA (DNA-linked Inhibitor ANtibody Assay) suitable for both enzyme quantification and screening of enzyme inhibitors. DIANA overcomes limitations of current state of the art methodologies in sensitivity, linear range, and the applicability to screen inhibitors with unpurified enzymes. In DIANA, the target enzyme is captured by an immobilized antibody or via affinity tag, probed with the detection probe consisting of a small-molecule ligand conjugated with a reporter DNA, and subsequently detected by qPCR. We showed on two clinically relevant targets CAIX (carbonic anhydrase IX) and PSMA (prostate specific membrane antigen) that DIANA has a linear range of up to six logs, allows selective detection of zeptomole amounts of targets, and is applicable to complex biological matrices. Furthermore, we showed that selectivity and sensitivity of DIANA enables quantitative determination of inhibition potency from single inhibitor concentration even when using microliters of blood sera containing picograms of target enzyme. [Navratil et al NAR 2017]
Here, we present an improved DIANA setup suitable for automation and high-throughput screening (HTS). We miniaturized the assay to 384-well plate format and automated each step of the protocol using standard lab equipment. In a pilot run, we screened a subset of proprietary IOCB library consisting of 2,816 compounds for CAIX inhibitors and obtained 13 hits. Even though we used endogenously expressed CAIX in a cell lysate, the assay showed signal-to-noise ratio of several logs leading to Z’ value >0.9. That enabled us to determine the inhibition potency of the tested compounds directly from single well measurements during the screen. At the same time, all hits and their potencies were confirmed in an orthogonal assay showing the accuracy of DIANA results. Additionally, we analyzed the ability of DIANA to screen pooled compounds. We pooled 11 compounds per well, performed the screen and deconvoluted the positive wells. All hits discovered by the non-pooled screening were recovered in the pooled screen with zero false positive rate. Finally, we showed that we are able to perform the pooled screen even when using human serum. These results confirm the robustness, sensitivity and selectivity of DIANA and its applicability to unpurified targets.
To conclude, DIANA is an accurate method suitable for efficient high-throughput screening, which reduces the time and cost of the screening. Using our ultra-fast qPCR protocol, it is possible to run up to 20 plates in 8 hours on a single qPCR instrument, enabling to analyze about 70,000 compounds when pooling 10 compounds per well. Moreover, throughput can be further scaled by implementation of automated robotic platforms. The development of the DIANA screening assay for new targets is straightforward and we have already developed the assay for more than ten relevant targets such as various influenza enzymes or insulin receptor. The assay will be further developed by the newly established company DIANA Biotechnologies, which will validate the pooled screening for large libraries with over 100,000 compounds and will offer custom assay development and screening.
