From Phenotypic Screening to Target and Compound Prioritization for Autosomal Dominant Polycystic Kidney Disease

Autosomal Dominant Polycystic Kidney Disease (ADPKD) is a progressive kidney disorder with limited therapeutic options, highlighting an urgent need for safer, long-term treatments. This study presents a multi-step computational and experimental approach to identify and assess target-based hypotheses for mechanisms underlying the cAMP-dependent cyst swelling. We analyzed phenotypic screening data from a 3D mIMCD3-Pkd1^-/- forskolin-induced, cAMP-driven cyst swelling (CS) assay, classifying compounds based on their CS activity. These compounds were cross-referenced with the Papyrus database to identify known biological targets, which were then prioritized by considering their expression in the screened cell model and by deprioritizing targets associated with antineoplastic activity. Four prioritized targets, P2X purinoceptor 7 (P2RX7), Glucose Transporter 1 (GLUT1), Mineralocorticoid Receptor (MR), and Adenosine A1 Receptor (A₁AR), were experimentally evaluated by testing known ligands in the phenotypic assay. While P2RX7 modulation showed no effect, GLUT1 inhibitors significantly reduced cyst swelling. For A₁AR, known agonists reduced CS while the inverse agonist DPCPX enhanced it. In contrast, MR modulation showed compound-specific effects, with the antagonist Esaxerenone uniquely reducing CS among tested MR antagonists. Quantitative structure-activity relationship (QSAR) models for A₁AR and MR were used to select structural analogues of known active compounds within chemical vendor catalogues to further confirm identified target-ADPKD activity space. Radioligand displacement assays were conducted to confirm the QSAR-predicted A₁AR binding affinities and selectivity among adenosine receptor subtypes. A₁AR affinity and subtype selectivity was a common denominator among ligands with CS-reducing activity, with the A₁AR positive allosteric modulator MIPS521 showing the strongest effect. The workflow therefore generated a coherent A₁AR-linked hypothesis and identified Esaxerenone as a compound-specific phenotypic hit; both require mechanistic confirmation and evaluation in additional disease models.