Journal of Medicinal Chemistry

The classical estrogen receptor (ER) is a steroid hormone receptor known for its role in both health and disease. ER, like its counterpart ER{beta}, acts as a receptor for its endogenous ligand 17{beta}-estradiol (E2) and is capable of binding a diverse set of natural and synthetic small molecules. Several screening campaigns have identified various ER ligands, based on the oxabicyclo[3.3.1]-nonene scaffold, that act as ER agonists. One such campaign extensively characterized an oxabicyclic ER agonist, termed AB-1. Herein, we report the synthesis of a series of analogs based on the scaffold of AB-1 (herein termed 1), focusing on the enhancement of their binding to ER and functional potencies in ER mediated assays. We identify a new substituted compound 9 that achieves subnanomolar potencies for ER binding (IC50= 0.45 nM) and ER dependent transcription (EC50= 0.37 nM) and proliferation (EC50= 0.03 nM), with potencies approaching those of E2. The high potency of 9 provides a promising scaffold for the potential development of new ER-targeted therapeutics.

Fatty acid synthase (FASN) is the sole enzyme capable of lipogenesis in cells and is upregulated in many cancers. Herein we report the discovery and development of SBP-5917 (78S), a novel, reversible, small-molecule probe which targets the thioesterase (TE) domain of FASN. SBP-5917 and other compounds in the series possess IC50 < 1 M and < 5 M for inhibiting recombinantly purified FASN-TE and cellular FASN respectively. Co-crystal structures of our inhibitors bound to the FASN-TE domain show that they are positioned in the substrate tunnel of the enzyme. Biochemical and biophysical analyses quantify the strength of compound binding and demonstrate their selectivity for FASN-TE. Preliminary pharmacokinetic studies indicate promising in vivo potency for many of the compounds in the series. SBP-5917 is the first structurally validated, reversible inhibitor of FASN-TE with demonstrated cellular target engagement and can serve as a chemical probe to investigate FASN as a therapeutic target.

CC10137 is a first-in-class peripherally restricted 2A-adrenergic receptor (2AAR) agonist with broad-spectrum analgesic efficacy and a favorable safety profile. Systematic exploration of the chemical space surrounding first-in-class leads is important for defining series boundaries and guiding continued optimization, but conventional analogue-by-analogue medicinal chemistry samples only a small fraction of the accessible structural space. Here, we used a scaffold-seeded enumeration strategy to expand the chemical space surrounding CC10137 from four SAR-informed seed compounds comprising CC10137 and three closely related structural variants. Application of predefined medicinal chemistry transformation rules in StarDrop generated a virtual library of 16,601,163 unique structures. Morgan fingerprint-based principal component analysis indicated that the library occupied a highly multidimensional structural space involving variation in scaffold substitution, peripheral functional groups, and side-chain composition. A retrospective comparison set of 43 compounds independently designed and experimentally characterized in the earlier CC10137 program represented only approximately 0.00026% of the 16.6-million-member library, yet all 43 were recovered as exact structural matches. Three compounds selected directly from the virtual library retained 2AAR binding affinity and agonist potency below 25 nM. Five representative compounds further showed significant anti-allodynic effects in the in vivo spared nerve injury model, with inhibition rates ranging from 39.3% to 55.7%. These findings support scaffold-seeded computational enumeration as a practical strategy for systematic chemical-space mapping around a first-in-class lead and for identifying additional pharmacologically active structural regions for further optimization.

Isocitrate dehydrogenase 1 (IDH1) catalyses the interconversion of isocitrate and -ketoglutarate and is overexpressed in several cancers, supporting tumour survival and treatment resistance. Inhibitors targeting the oncogenic mutant forms of IDH1 also inhibit the wild-type (WT) protein, but lose substantial potency due to greater competition with native substrates (isocitrate and Mg2+). We hypothesised that a covalent inhibitor may be more effective at overcoming the higher substrate affinity of wild-type IDH1. Here, a covalent fragment-based approach was used to develop IDH1-C269 selective inhibitors which prevent the key regulatory segment from forming an -helix required for catalytic competency. Following the identification of C269-selective fragments, we used X-ray crystallography to characterise IDH1-fragment bound complexes and guide structure-based design efforts. Inspired by the unexpected detection of bound isocitrate molecule in an X-ray crystal structure, fragment expansion yielded a series of compounds exploiting both an adjacent site and a unique water-mediated hydrogen bonding network. This series achieved strong potency (reaching IC50 = 47.5 nM, a 155-fold improvement in potency relative to the original fragment), retained activity in the presence of competing Mg2+, showed selectivity against IDH2 and reduced NADPH concentrations in a relevant PDAC cell model. Together, these findings present a mechanistic rationale for the covalent targeting of wild-type IDH1 and provide structurally validated inhibitors for further development.

The folate receptor (FR) is overexpressed in a range of human tumours including ovarian cancer cells. We propose that the overexpression of the FR on the surface of ovarian tumour cells could be exploited for the selective delivery of a DNA damage response inhibitor (DDRi) in the form of an intact folate drug conjugate (FDC). This approach would improve the therapeutic index of the parent DDRi facilitating combination studies of the DDRi-based FDC with DNA damaging chemotherapy. FR-mediated cellular uptake of the proposed folate drug conjugates is requisite for FDC selective delivery into tumours. In this study, we synthesised a series of olaparib-based folate conjugates that maintained the biochemical PARP1 inhibition associated with olaparib and showed binding affinity for the folate receptor. Significantly, we identified compounds 10b and 11 that selectively enter FR overexpressing tumour cells via folate receptor-mediated endocytosis in their intact form and engage with their target as demonstrated by the potent inhibition of PARylation (KB cells, PARylation IC50 = 5.7 and 3.9 nM; respectively).

Screening of the open-access CRESTdb small-molecule library identified the imidazothiazole carboxamide sCQG200 (1) as an initial hit, with activity against H37Rv and Erdman Mycobacterium tuberculosis strains in cholesterol-containing medium (MIC90= 6.88 uM and 4.48 uM, respectively) and against intracellular Mtb (EC50= 3.80 uM). sCQG200 retained activity against drug-resistant Mtb strains, including clinical MDR/XDR isolates. SAR optimization led to analogs 21 and 24 with substantially improved antimycobacterial potency. Mouse pharmacokinetic studies following oral and intravenous administration showed 62.1% and 29.5% oral bioavailability for 21 and 24, respectively. This structurally differentiated chemotype provides a promising starting point for further optimization as antitubercular agent.

Alkaline ceramidase 2 (ACER2) catalyzes the hydrolysis of ceramides into sphingosine and free fatty acids, playing crucial roles in regulating various biological processes, including but not limited to cell proliferation, survival, and the DNA damage response. However, a pharmacological tool that regulates its catalytic and signaling functions remains to be developed. In this study, N-[(1S,2R)-1-hydroxy-1-phenylpropan-2-yl]dodecane-1-sulfonamide (KPB82) was identified as a potent and selective ACER2 inhibitor. In vitro enzymatic assays showed that KPB82 inhibited ACER2 with an IC50 of 190 nM, while displaying minimal activity against acid and neutral ceramidases (IC50> 100 {micro}M) and no significant inhibition of alkaline ceramidase 1 or alkaline ceramidase 3. Enzymatic kinetic studies revealed that KPB82 is a competitive inhibitor of ACER2. The three-dimensional structure of ACER2 was modeled in the AlphaFold3 server, revealing key Zn2+-coordinating histidine residues and a substrate-binding pocket. Molecular docking studies demonstrated that KPB82 binds ACER2 within this pocket, forming hydrogen bonds with the catalytically important amino residues, Asp93 and His215, and interacting with several hydrophobic residues. Lactate dehydrogenase release assays revealed that KPB82 significantly reduced cytotoxicity in HeLa cells overexpressing ACER2, as well as in HCT116 cells, RT4 cells, and human umbilical vein endothelial cells treated with the DNA-damaging agent doxorubicin. Lipidomic analysis showed that KPB82 inhibited sphingosine production in response to ACER2 overexpression or doxorubicin treatment. Collectively, these findings demonstrate that KPB82 is a potent and selective ACER2 inhibitor capable of suppressing ACER2-dependent biological processes and highlight its potential as a tool for studying ceramide metabolism and ACER2-mediated signaling pathways.

Allosteric modulation of G protein-coupled receptors (GPCR) is a rapidly advancing area of drug discovery with considerable therapeutic potential. In this study, we used various bioluminescence resonance energy transfer (BRET) assay configurations to investigate the differential activity of LY3154207, a dopamine D1 receptor (D1R)- positive allosteric modulator (PAM), in potentiating D1R-mediated signaling pathways. Our data showed that LY3154207 demonstrated clear bias toward {beta}-arrestin recruitment over G-protein engagement. Among the G-protein subtypes, LY3154207 preferentially potentiated Gs signaling induced by DA. Interestingly, when tested for its effect on the Gs-selective D1R agonist DHX and the Golf-selective agonist TAV, LY3154207 further enhanced the selectivity of these compounds, suggesting a probe-dependent effect. We also studied the structural mechanism underlying this subtype selectivity and found that W123 and R130 in the intracellular loop 2 (ICL2) of D1R were crucial for the activity of LY3154207. Furthermore, the N domain of the G-protein appeared to make minor contributions to this G-protein subtype selectivity of LY3154207. Lastly, our data demonstrated that LY3154207 induced cAMP production in the absence of the orthosteric ligand, indicating that this compound is an ago-PAM.

Albumin binding can substantially alter the unbound concentration and apparent cellular potency of small-molecule anticancer agents. BAY-1816032 is a selective BUB1 kinase inhibitor with activity in malignant pleural mesothelioma, but its interaction with serum albumin has not been defined. This study combined intrinsic bovine serum albumin (BSA) fluorescence measurements, molecular docking, and cell-viability assays to examine whether albumin association influences BAY-1816032 activity. BAY-1816032 produced pronounced quenching of BSA fluorescence at low drug-to-protein molar ratios, and the quenching response was greater at mildly acidic pH than at mildly basic pH. Docking to BSA (PDB 4F5S) placed the top-ranked pose in Sudlow site I (subdomain IIA), with a Vina score of -7.8 kcal/mol and predicted hydrogen-bond contacts with Tyr149, Arg256, and Asp450. In NCI-H28, NCI-H2052, NCI-H2452, and AB12 mesothelioma cells, supplementation of 10% fetal bovine serum medium with 2% (w/v; 20 mg/mL) BSA shifted BAY-1816032 concentration-response curves to the right. Relative to 10% serum alone, IC50 values increased from 2.4 to 11.1 uM in H28, 3.6 to 22.0 uM in H2052, 2.0 to 4.9 uM in H2452, and 2.4 to 11.1 uM in AB12. These data support an albumin-dependent reduction in the freely available drug fraction and identify site I as the leading computational binding-site hypothesis. Because fluorescence-intensity and docking data do not independently establish a dissociation constant or quenching mechanism, direct unbound-fraction measurements and competitive displacement experiments are required for quantitative confirmation.

Exploring the ability of metabolic enzymes to interact with diverse small molecules can provide insights into their ligand recognition and activity regulation. Glucose-6-phosphate isomerase (GPI) is an enzyme that catalyzes the second step of glycolysis, a critical node that links glucose metabolism to other cytosolic pathways, such as pentose phosphate and hexosamine biosynthetic pathways. Despite its importance in cellular metabolism, little is known about the broad ligandability of GPI. Here, we performed a structure-based virtual screening using molecular docking to identify new modulators predicted to engage with the human GPI active site. Enzyme kinetic and binding studies confirmed eleven compounds as active-site binders of GPI, exhibiting inhibitory activities and binding affinities in upper micromolar to sub-millimolar range. The binding of several ligands to GPI active site was further supported by X-ray crystallographic data. In addition, in vitro assays were performed to evaluate the modulatory effects of the GPI hits on cellular glycolysis and viability. The identified inhibitors included the mono-phosphorylated metabolite of ribavirin, which inhibited glycolytic flux and cell viability; two compounds with phosphate bioisosteres; and endogenous phosphate-containing metabolites from the pentose phosphate, hexosamine, nucleotide, and mevalonate pathways. Collectively, these findings expanded the repertoire of human GPI ligands and revealed previously unrecognized GPI-metabolite interactions that may contribute to modulation of glycolytic flux.

The full efficacy range of ligands at CB1 and CB2 (i.e., inverse agonists, neutral agonists, partial agonists, full agonists, and high-efficacy agonists) has been reported collectively across numerous laboratories and assay platforms. Canonical Gi/o and {beta}-arrestin signaling have each been suggested to shape physiological effects differently via different classes of cannabinoid ligands. These prior efficacy classifications, however, were obtained in multiple assay systems, often using indirect, amplified readouts and different reference agonists, making them difficult to compare across ligands. Here we profiled a structurally diverse panel of 22 cannabinoid ligands: synthetic cannabinoid receptor agonists (SCRAs), {Delta}-tetrahydrocannabinol ({Delta}-THC) analogues, endocannabinoid analogues, and antagonists/inverse agonists at CB1 and CB2 receptors using bioluminescence resonance energy transfer (BRET) assays, measuring Gi1 engagement and {beta}-arrestin 2 recruitment and normalizing every response to the reference full agonist CP55,940. At CB1, SCRAs and {Delta}-THC analogues acted as high-efficacy agonists. At CB1, these high-efficacy agonists were also highly efficacious in {beta}-arrestin 2 recruitment. Across the panel, the most consistent trend was a loss of agonist efficacy from CB1 to CB2. Among antagonists, inverse agonists suppressed constitutive CB1 activity more than neutral antagonists. Together, these CP55,940-normalized measurements provide a common efficacy scale and a framework for interpreting pharmacodynamics of CB1 and CB2 cannabinoid receptors.

Dysregulation of insulin-like growth factor (IGF) signaling contributes to diverse human diseases. Therapeutic strategies have largely targeted the IGF-1 receptor, but the complexity of the IGF system has limited their clinical efficacy. Pregnancy-associated plasma protein-A (PAPP-A) is a secreted metalloproteinase that regulates local IGF signaling through proteolytic cleavage of IGF-binding proteins (IGFBPs) and represents an attractive therapeutic target. Here, we report the first synthetic PAPP-A inhibitors designed to target the enzyme active site using an IGFBP-5-derived anchor sequence. Replacing the scissile bond with phosphinate or phosphonate moieties yielded potent and selective PAPP-A inhibitors. Cryo-electron microscopy revealed the molecular basis of inhibition and key active-site interactions. The compounds inhibited cleavage of IGFBP-4 and IGFBP-5, two physiological PAPP-A substrates. They suppressed IGF-1 signaling in A549 lung cancer cells, with efficacy comparable to the endogenous PAPP-A inhibitor stanniocalcin-1. They also inhibited Papp-aa-dependent IGF signaling and cell proliferation in wild-type and stc1a-/- mutant zebrafish in vivo in a dose-dependent manner. Together, these findings establish the first class of synthetic, selective PAPP-A inhibitors capable of modulating IGF signaling. They provide new chemical tools for studying and modulating pathological IGF signaling and offer new directions for developing PAPP-A-targeted therapeutic strategies.

Objective To systematically evaluate the predictability of drug chemical structures against the complete set of MedDRA Preferred Terms (PTs), construct a comprehensive structure adverse reaction landscape, validate the utility of molecular fingerprint representations in pan-screening tasks, and establish quantitative tools for drug safety pre-screening. Methods Drug adverse event association data were compiled from SIDER 4.1 and its derivative compilation Nsider, encompassing 2,005 approved drugs, of which 1,120 small-molecule drugs with successfully retrieved chemical structures were retained for analysis. For each Preferred Term (PT) meeting the minimum sample-size threshold ([&ge;]25 positive and [&ge;]30 negative drugs; n = 768), XGBoost classifiers were trained using MACCS 166-bit fingerprints, with model performance evaluated via 5-fold stratified cross-validation. A PT was designated as high-confidence if it achieved a validation AUC [&ge;] 0.75, a cross-validation AUC [&ge;] 0.70, and passed Benjamini Hochberg false discovery rate (FDR) correction (q < 0.05). The entire analytical pipeline was replicated using Morgan ECFP4 2048-bit fingerprints to assess the robustness of fingerprint selection. The robustness of significance screening was independently validated by permutation testing (500 random label permutations), and model generalizability was evaluated through Bemis-Murcko scaffold-based stratified split validation. SHAP (SHapley Additive exPlanations) analysis was employed to identify critical structural features, and mechanism-based case studies further demonstrated concordance between model-identified features and established pharmacophores. Results Among the 768 modelable PTs, 71 (9.2%) met the high-confidence significance criteria (FDR q < 0.05), predominantly comprising nervous system disorders (10 PTs) and psychiatric disorders (9 PTs). The toxicities exhibiting the strongest structural predictability included hyperprolactinaemia (AUC = 0.968), dermatitis acneiform (AUC = 0.966), tardive dyskinesia (AUC = 0.952), and breast tenderness (AUC = 0.979). Under a unified XGBoost hyperparameter configuration, the low-dimensional MACCS fingerprint (166-bit) demonstrated near-equivalent performance to the high-dimensional Morgan fingerprint (2048-bit) in pan-screening tasks ({Delta}AUC = -0.012, Cohen's d = 0.16), and achieved comparable performance at a substantially lower dimensionality across 18 of 22 organ systems. SHAP analysis identified key structural fragments corresponding to established pharmacophores (six-membered rings, tertiary amine cores, heteroatomic bonds, and bridged heteroatomic chains), and mechanism-based case studies further demonstrated that the model captured structural patterns concordant with known pharmacophores of D2 antagonists, EGFR inhibitors, and related agents. Bemis-Murcko scaffold-based stratified split validation revealed a median AUC decrease of merely 0.042, with 66% of PTs maintaining an AUC [&ge;] 0.70 under scaffold-split conditions. Immune-mediated and multifactorial endpoints remained structurally opaque under the current data conditions. Conclusion This study constructed a comprehensive structure-adverse reaction predictability landscape encompassing 768 MedDRA PTs; seventy-one PTs achieved high-confidence criteria. The MACCS 166-bit fingerprint achieved near-equivalent performance to the Morgan 2048-bit fingerprint at a markedly lower dimensionality, and its chemically explicit features enable direct translation of model outputs into structural alerts, providing a computational tool for prospective screening of potential adverse reaction profiles for any novel compound. A tiered assessment strategy is proposed: machine learning ensemble models for high-confidence screening, substructure alert rules for high-specificity exclusion, and pharmacovigilance signal detection for complementary coverage of structurally unpredictable endpoints. Bemis-Murcko scaffold-based stratified split validation further demonstrated that the median generalization performance decrease was merely 0.042 AUC, with 66% of PTs maintaining an AUC [&ge;] 0.70 under scaffold-split conditions, corroborating the existence of a genuine chemical basis underlying structure-adverse reaction mappings.

Cyclic peptides can engage targets beyond the reach of small molecules, but their discovery typically relies on libraries reused across targets, diversification around a known interaction motif, or computational methods limited to a handful of non-canonical building blocks. Here, we present nCycle-Forge, a simulated annealing framework that designs cyclic peptide binders by iteratively substituting building blocks and scoring each proposal by cofolding with the target. We designed, synthesized, and screened 1,472 cyclic peptides against the active-site pocket of Thrombin and the shallow p53-binding surface of MDM2 from the target sequence alone and without fixing binding mo-tifs. The library yielded chemically diverse hits against both targets in a single round without experimental optimization, with hit rates comparable to or up to 24-fold higher than in larger experimental libraries. Purified designs reached sub- to low-micromolar potency, with the best Thrombin inhibitor at Ki 0.14 M and the best MDM2 binder at EC50 2.4 M. Surface plasmon resonance confirmed direct binding, competition with or-thosteric inhibitors at both sites, and weaker binding under conditions expected to linearize the designs. nCycle-Forge requires no building block-specific parameterization, conformer generation, or retraining, and provides a target-focused route to functional cyclic peptides beyond the canonical amino acid alphabet.

The dopamine D2 receptor is a principal target of antipsychotic and antiparkinsonian drugs, yet well-characterized fluorescent ligands for the D2 receptor remain scarce compared with its relative, the D3 receptor. Here we show that fluorescent ligands developed for the D3 subtype can be repurposed as tools for studying the D2 receptor pharmacology. Using a machine-learning-assisted, live-cell fluorescence microscopy assay, we screened twelve fluorescent probes and identified eight ligands that bound specifically to the D2 receptor at 10 nM. Four ligands exhibited classical saturable binding, yielding Kd in the nanomolar range, whereas the remaining probes displayed unsaturated, time-dependent binding profiles. Kinetic analyses revealed substantial differences in association and dissociation behavior, highlighting the importance of time-resolved measurements. Competition binding with dopamine and haloperidol yielded affinity estimates consistent with literature values. These results expand the toolkit of validated D2 receptor fluorescent probes and support cross-receptor repurposing as a strategy for fluorescent ligand development.

Reactivation of telomerase confers immortality to approximately 90% of human tumors by enabling continuous elongation of the DNA at chromosome ends, or telomeres. The telomerase catalytic subunit TERT adds TTAGGG repeats using a portion of the telomerase RNA component hTR as a template. Because telomerase is inactive in most normal somatic cells, it remains an attractive therapeutic target; however, no telomerase inhibitor has yet demonstrated robust clinical efficacy with acceptable safety. Here we evaluate thiomorpholino oligonucleotides (TMOs) as a new class of antisense oligonucleotides targeting the template region of hTR. TMOs incorporate morpholino rings and phosphorothioate linkages, which enhance nuclease resistance, RNA binding and nuclear uptake. Two anti-hTR TMOs inhibited telomerase activity in vitro with an IC50 below 1 nM, whereas two control TMOs were at least 100-fold less active. HeLa cells treated with anti-hTR TMOs showed progressive telomere shortening, detectable after one week of treatment. Growth inhibition was observed after substantial telomere erosion, and both telomere length and proliferation recovered upon withdrawal of TMOs. These findings establish TMOs as a promising new chemistry for telomerase-targeted therapeutics.

COVID-19 highlighted the need for antivirals that can be rapidly deployed against both current and future coronavirus threats. CMX990 is a clinical stage coronavirus main protease (Mpro) inhibitor bearing a unique trifluoromethoxymethyl ketone warhead. Here, we broadly evaluate its pan-coronavirus target engagement, antiviral breadth, and pharmacologic differentiation. Across a reporter panel of 24 Mpro enzymes spanning alpha-, beta-, delta-, and gamma-coronaviruses, CMX990 retained robust potent activity, including against nirmatrelvir-resistant SARS-CoV-2 variants. In direct comparison, nirmatrelvir and ensitrelvir showed diminished potency against non-SARS-CoV-2 coronaviruses. Live-virus assays corroborated this breadth, with potent, selective activity against SARS-CoV-2, SARS-CoV-1, MERS-CoV, and additional human coronaviruses. Structural studies revealed a conserved covalent binding mode centered on the catalytic Cys of coronavirus Mpro. In serial-passage studies, resistance emerged only after approximately twice as many serial passages under CMX990 selection as under nirmatrelvir selection, without selecting for the canonical nirmatrelvir-resistance substitution E166V. Extracellular washout experiments revealed superior persistence of antiviral activity relative to nirmatrelvir, consistent with durable target engagement. In a murine SARS-CoV-2 infection model, once-daily CMX990 protected against weight loss and reduced lung viral titers to near baseline levels. Collectively, these findings identify CMX990 as a structurally and functionally validated pan-coronavirus protease inhibitor supporting its potential as a pandemic-ready antiviral candidate.

MAPK-activated protein kinase 2 (MAPKAPK2, MK2) is a downstream substrate of p38 MAPK that regulates inflammatory signaling and cytokine synthesis, making it an attractive therapeutic target. Although several crystal structures of MAPKAPK2 with ATP-competitive inhibitors have supported drug discovery, structural elucidation of selective non-ATP-competitive inhibitors has remained challenging. MK-25 is a non-ATP-competitive furan-2-carboxamide MAPKAPK2 inhibitor developed at Merck to overcome the selectivity and cellular-efficacy limitations of ATP-competitive chemotypes. Here, we show through isothermal titration calorimetry that MK-25 binds the activated MAPKAPK2, consistent with state-dependent allosteric binding. To determine the binding mode of MK-25, we solved a 2.8 A resolution crystal structure of an activated MAPKAPK2 construct in ternary complex with the ATP-site inhibitor PF-3644022 and a bromo-analogue of MK-25. The ligand occupies a pocket in the C-lobe overlapping the protein-substrate recognition region and formed through rearrangements of D and G helices and the activation segment. These interactions are consistent with the structure-activity relationships reported for the furan-2-carboxamide chemotype and identify a ligandable allosteric pocket that may facilitate future structure-guided design of selective, non-ATP-competitive MAPKAPK2 inhibitors.

This study aims to identify potential inhibitors of the Ag85C enzyme from Mycobacterium tuberculosis through primary docking, pharmacophore modelling, and ADME/pharmacokinetics analysis. Delamanid and Pretomanid emerged as the top hits in primary docking, demonstrating strong binding affinities and interactions with key active site residues. A robust pharmacophore model was developed, identifying essential features for binding and used to screen 4254 antituberculotic compounds, yielding three promising candidates: F0314-0040, F0398-0505, and F0349-2716. These compounds exhibited high binding affinities in major docking studies. ADME and pharmacokinetics analysis revealed favorable Drug likeness and absorption profiles, supporting their potential as effective antituberculotic agents. The integration of computational methods provided a comprehensive approach for rational drug design.

The pharmacokinetics concentration-time profile encodes ADME dynamics. Its optimization in drug discovery curbs late-stage attrition, yet forecasting human pharmacokinetics from chemical structure remains difficult. Machine learning models ignore continuous ADME dynamics, whereas physiologically based pharmacokinetics (PBPK) models demand unavailable parameters and impose rigid, misspecified compartmental structures. We present PK-MUSE (PharmacoKinetic hybrid Mechanistic Universal System Equation), augmenting a two-compartment model with bounded, state- and time-dependent neural corrections to elimination and intercompartmental distribution. Molecular structure parameterizes compound-specific behavior, while learned corrections refine dynamic equations. Across scaffold and temporal out-of-distribution evaluations, PK-MUSE achieved the lowest mean exposure-endpoint MdFE, reducing MdFE by 7.2%, 45.2%, and 26.2% versus state-of-the-art machine learning (HML- RF), deep learning compartmental (DeepCt), and PBPK (PK-Sim) models. Interrogating the corrections localized misspecification to distribution rather than elimination. By extending mechanistic simulators with data-driven constrained neural dynamics that preserve interpretability and generalize to unseen chemotypes, PK-MUSE advances digital twins for drug discovery.