Hassan Al-Ali

Though 100 kinase inhibitors are approved for clinical use, most kinases in the genome remain poorly characterized. Moreover, though kinase inhibitors have revolutionized the care of selected cancers, most others respond transiently or not at all to these drugs, typically directed against signal-transduction pathways. Given cellular regulation at every step by phosphorylation events, we hypothesized that phenotypic screening and machine-learning analysis can reveal novel kinase targets. Diffuse large B-cell lymphoma (DLBCL), the most common lymphoma, responds poorly to signal transduction-directed kinase inhibitors. We applied a library of probes to DLBCL systems broadly representing its pathogenic heterogeneity and subjected results to iterative target-deconvolution analysis aimed at tumor selectivity (idTRAX). These efforts identified the cyclin G-associated kinase (GAK) as a tumor-selective target. Mechanistic studies showed GAK kinase activity is required for mitotic progression, with inhibition leading to G2/M arrest and cell death and increased efficacy in the absence of the retinoblastoma (RB) tumor suppressor. GAK-directed kinase inhibitors, including repurposed clinical compounds, showed promising single-agent anti-DLBCL activity in vivo. GAK kinase function is therefore a targetable dependency, opening novel treatment strategies suitable for rapid clinical translation. Moreover, we establish phenotypic screening with target deconvolution as a generalizable strategy to reveal tumor-selective kinase dependencies.

The p21-activated kinases (PAKs) are a group of serine-threonine kinases central to multiple signaling pathways that govern cell survival and proliferation. Aberrant activity of PAK1, the most well characterized member of the PAK family, drives progression of several malignancies and brain disorders, including Alzheimer’s disease and neurodevelopmental disorders. Despite growing interest in PAK1 as a drug target for these diseases, there is no assay to evaluate the intracellular target engagement of PAK1 inhibitors. To address this need, we developed first-in-class NanoBRET assays for wild-type PAK1 and a neurodevelopmental disorder-causing gain-of-function PAK1 mutant. Furthermore, we executed our novel PAK1 NanoBRET assay to evaluate target engagement of PAK1 inhibitors in primary hippocampal neurons. To the best of our knowledge, this is the first demonstration of a NanoBRET cellular target engagement assay in primary neurons, thereby increasing the relevance of our work by confirming PAK1 inhibitor binding to the aberrant form of the protein in cerebral cells.

Axon regrowth in the central nervous system (CNS) is constrained by robust regulatory networks. Here we show that optimal neurite outgrowth in rodent and human CNS neurons is achieved by co-inhibition of kinases across four closely related clades within the protein kinase A, G, and C (AGC) family. The kinases derive from ancestral regulators of cytoskeletal dynamics, resource allocation, and polarized cell growth. Their shared domain architecture makes polypharmacology (co-engagement by a single small molecule) feasible. Phenotype-guided optimization of a tool compound with established efficacy in mouse spinal cord injury models yielded TMP-316, a drug candidate engaging these AGC kinases with selectivity against the broader kinome. A single intrathecal dose of TMP-316 produced sustained motor recovery in a rat cervical hemicontusion model. These findings reveal conditions under which polypharmacology is simultaneously required and enabled by shared evolutionary origins, illuminating a therapeutic discovery principle for pathologies governed by functionally overlapping targets.

Intrathecal (IT) drug delivery is a critical technique for bypassing the blood-brain and blood-spinal cord barriers in preclinical CNS research. However, conventional rat catheterization methods suffer from high rates of neurologic complications, poor reliability, and unverified dosing due to epidural reflux and inconsistent supraspinal distribution. Our objective was to develop and validate an improved method for lumbar IT catheterization in rats that ensures distribution to the brain and to confirm supraspinal pharmacodynamic target engagement. We describe a refined microsurgical technique using dural puncture under direct visual control at the L6-S1 interlaminar space, a site chosen for its anatomical safety margin. The method uses a small-bore (0.33 mm OD) polyurethane (PU) catheter to minimize durotomy size, air-bubble tracking to compensate for catheter dead volume, and epidural sealing with Surgifoam ^® to minimize reflux. Visualization using Evans Blue dye confirmed complete neuraxial distribution from a single 30 µL lumbar bolus injection, with dye reaching the ventral/dorsal brain cisterns. Pharmacodynamic validation of cortical exposure was achieved using an S6 kinase 1 (S6K1) inhibitor. Lumbar IT administration over a period of 6 hours via a pump resulted in significant supraspinal S6K1 engagement, demonstrated by a reproducible reduction in S6 phosphorylation in the cerebral cortex. Highlights: Method for lumbar intrathecal catheterization in rats under direct visual control, using basic surgical tools CNS distribution validated by Evans Blue dye reaching ventral and dorsal brain Pharmacodynamic confirmation of supraspinal target engagement following lumbar intrathecal delivery of a small molecule kinase inhibitor Serves as a faithful preclinical model for therapeutics intended for clinical intrathecal administration Provides a screening route for early-stage compounds not yet optimized for CNS penetrance, supporting efficacy testing prior to medicinal chemistry investment

The p21-activated kinases (PAKs) are a group of serine-threonine kinases central to multiple signaling pathways that govern cell survival and proliferation. Aberrant activity of PAK1, the most well characterized member of the PAK family, drives progression of several malignancies and brain disorders, including Alzheimer’s disease and neurodevelopmental disorders. Despite growing interest in PAK1 as a drug target for these diseases, there is no assay to evaluate the intracellular target engagement of PAK1 inhibitors. To address this need, we developed first-in-class NanoBRET assays for wild-type PAK1 and a neurodevelopmental disorder-causing gain-of-function PAK1 mutant. Furthermore, we executed our novel PAK1 NanoBRET assay to evaluate target engagement of PAK1 inhibitors in primary hippocampal neurons. To the best of our knowledge, this is the first demonstration of a NanoBRET cellular target engagement assay in primary neurons, thereby increasing the relevance of our work by confirming PAK1 inhibitor binding to the aberrant form of the protein in primary neurons.

Dysregulation of lipid homeostasis is associated with a wide range of pathologies encompassing neurological, metabolic, cardiovascular, oncological, and renal disorders. We previously showed that lipid droplet (LD) accumulation in podocytes contributes to the progression of diabetic kidney disease (DKD) and reducing LDs preserves podocyte function and prevents albuminuria. Here, we sought to identify compounds that treat pathological LD accumulation. We developed a phenotypic assay using human podocytes and deployed it to screen a combinatorial library comprising over 45 million unique small molecules. This led to the identification of a compound series that effectively reduces LD accumulation in stressed podocytes. Mechanistic studies revealed that these compounds activate lipophagy, reduce LD accumulation, and rescue podocytes from cell death. In contrast, compounds known to induce general autophagy failed to mimic these effects, indicating a novel lipophagy-specific mechanism of action (MoA), which was confirmed by unbiased phenotypic profiling. An advantage of this therapeutic strategy is its potential to not only halt the progression of pathological lipid accumulation but also reverse it. These compounds will serve as tools for uncovering novel drug targets and therapeutic MoAs for treating DKD and other diseases with similar etiologies.

Precision medicine approaches using ex-vivo drug sensitivity testing (DST) have received attention in the cancer research community as a means to improve treatment stratification in populations where multiple treatment attempts are not feasible, or no standard-of-care treatment exists, such as ultra-rare cancers with a significant clinical need for effective treatment options, like osteosarcoma. DST has the potential to supplement existing patient stratification approaches by providing tumor-specific response data to aid in treatment selection at the time of treatment decision. We present the case of a pediatric osteosarcoma patient who was evaluated using DST at the time of standard-of-care treatment to evaluate treatment sensitivity. The DST screen indicated significant treatment sensitivity to anthracyclines and methotrexate, consistent with the first-line standard-of-care therapy (MAP). Clinical follow-up showed treatment sensitivity to standard-of-care MAP treatment and pathology results of 90% necrosis. The present case shows that DST screening is feasible from a technical standpoint, can be performed in a clinically relevant time frame that does not delay treatment start, and provides personalized drug sensitivity information on clinically available agents, and the DST results align with the clinical treatment response.

Dysregulation of lipid homeostasis is associated with a wide range of pathologies encompassing neurological, metabolic, cardiovascular, oncological, and renal disorders. We previously showed that lipid droplet (LD) accumulation in podocytes contributes to the progression of diabetic kidney disease (DKD) and reducing LDs preserves podocyte function and prevents albuminuria. Here, we sought to identify compounds that treat pathological LD accumulation. We developed a phenotypic assay using human podocytes and deployed it to screen a combinatorial library comprising over 45 million unique small molecules. This led to the identification of a compound series that effectively reduce LD accumulation in stressed podocytes. Mechanistic studies revealed that these compounds activate lipophagy, reduce LD accumulation, and rescue podocytes from cell death. In contrast, compounds known to induce general autophagy failed to mimic these effects, indicating a novel lipophagy-specific mechanism of action (MoA), which was confirmed with unbiased phenotypic profiling. An advantage of this therapeutic strategy is its potential to not only halt the progression of pathological lipid accumulation but also reverse it. These compounds will serve as tools for uncovering novel drug targets and therapeutic MoAs for treating DKD and other diseases with similar etiologies. GRAPHICAL

The term chondrosarcoma refers to a rare and heterogeneous group of malignant cartilaginous tumors that are typically resistant to chemotherapy and radiotherapy. Metastatic chondrosarcoma has a poor prognosis, and effective systemic therapies are lacking. Isocitrate dehydrogenase (IDH) mutations represent a potential therapeutic target, but IDH inhibitors alone have shown limited clinical efficacy to date. Although the role of conventional chemotherapy is still subject to debate, some evidence suggests it may provide therapeutic benefits in advanced cases. In this study, we aimed to identify effective compounds for combination therapy in chondrosarcoma. Using high-throughput screening, we evaluated a panel of anticancer agents in IDH1-mutant chondrosarcoma cell lines and their mutant IDH1 knockout derivatives. The top 20 most potent compounds were identified across all cell lines, irrespective of IDH mutation status. Representative drugs selected for further investigation included docetaxel, methotrexate, panobinostat, idarubicin, camptothecin, and pevonedistat. These drugs inhibited colony formation, induced apoptosis and cell cycle arrest, and exhibited synergistic antitumor activity in two-drug combinations. In conclusion, we identified several highly effective agents with potent anti-tumor activity in chondrosarcoma cells, independent of IDH mutation status. These agents represent promising candidates for chondrosarcoma therapy and warrant further preclinical investigation and potential inclusion in clinical trials.