Binding Data Research Articles

Critical Evaluation of Childs Method for the NMR Spectroscopic Scaling of Effective Lewis Acidity: Limitations and Resolution of Earlier Discrepancies.

Quantifying Lewis acidity is essential for understanding and optimizing the performance of Lewis acids in diverse applications. Next to the widely accepted Gutmann-Beckett (GB) method, using triethyl phosphine oxide (TEPO) as a probe, the Childs method-employing trans-crotonaldehyde (TCA)-gained attention as an NMR-based technique for measuring effective Lewis acidity (eLA). Despite its steady use, the robustness of Childs method and its correlation with other measures remain underexplored. Previous comparisons between the GB and Childs scales revealed significant discrepancies, suggesting that hard and soft acid/base (HSAB) characteristics may be operative. In this study, we compare thermodynamic data for TCA binding to 117 Lewis acids (representing global Lewis acidity, gLA) with their corresponding NMR-induced chemical shifts in TCA. Our findings showcase notable deviations that reinforce key distinctions between eLA and gLA perspectives. Moreover, we identify significant limitations in the Childs method. First, the weak donor strength of TCA limits its applicability to only the strongest Lewis acids. Second, the exposed protons of TCA are prone to secondary interactions, obscuring the measurement of true Lewis acidity. Finally, our analysis reconciles discrepancies, refuting earlier assumptions that these arise from HSAB effects.

Encoding the lattice in the holography

One of the most wanted features of holography in its condensed matter physics application is to encode the structure of lattice, which is the most direct data of the material. In this paper, we propose a method to encode the lattice structure by embedding the tight binding data into the Dirac equation in the anti–de Sitter bulk. We explicitly worked out the idea for the graphene and Haldane models, and the result shows that some degrees of freedom escape the free-electron on-shell curve, and Green’s function loses the pole structure completely. It implies that the electronic structure is not described by the band structure only, which is consistent with what many angle-resolved photoemission spectroscopy data tell us, and it also implies that the system is in non-Fermi liquid even for the graphene, which is consistent with recent experiments for the clean graphene. Published by the American Physical Society 2024

CACHE Challenge #1: Targeting the WDR Domain of LRRK2, A Parkinson's Disease Associated Protein.

The CACHE challenges are a series of prospective benchmarking exercises to evaluate progress in the field of computational hit-finding. Here we report the results of the inaugural CACHE challenge in which 23 computational teams each selected up to 100 commercially available compounds that they predicted would bind to the WDR domain of the Parkinson's disease target LRRK2, a domain with no known ligand and only an apo structure in the PDB. The lack of known binding data and presumably low druggability of the target is a challenge to computational hit finding methods. Of the 1955 molecules predicted by participants in Round 1 of the challenge, 73 were found to bind to LRRK2 in an SPR assay with a KD lower than 150 μM. These 73 molecules were advanced to the Round 2 hit expansion phase, where computational teams each selected up to 50 analogs. Binding was observed in two orthogonal assays for seven chemically diverse series, with affinities ranging from 18 to 140 μM. The seven successful computational workflows varied in their screening strategies and techniques. Three used molecular dynamics to produce a conformational ensemble of the targeted site, three included a fragment docking step, three implemented a generative design strategy and five used one or more deep learning steps. CACHE #1 reflects a highly exploratory phase in computational drug design where participants adopted strikingly diverging screening strategies. Machine learning-accelerated methods achieved similar results to brute force (e.g., exhaustive) docking. First-in-class, experimentally confirmed compounds were rare and weakly potent, indicating that recent advances are not sufficient to effectively address challenging targets.

Benchmarking and building DNA binding affinity models using allele-specific and allele-agnostic transcription factor binding data

BackgroundTranscription factors (TFs) bind to DNA in a highly sequence-specific manner. This specificity manifests itself in vivo as differences in TF occupancy between the two alleles at heterozygous loci. Genome-scale assays such as ChIP-seq currently are limited in their power to detect allele-specific binding (ASB) both in terms of read coverage and representation of individual variants in the cell lines used. This makes prediction of allelic differences in TF binding from sequence alone desirable, provided that the reliability of such predictions can be quantitatively assessed.ResultsWe here propose methods for benchmarking sequence-to-affinity models for TF binding in terms of their ability to predict allelic imbalances in ChIP-seq counts. We use a likelihood function based on an over-dispersed binomial distribution to aggregate evidence for allelic preference across the genome without requiring statistical significance for individual variants. This allows us to systematically compare predictive performance when multiple binding models for the same TF are available. To facilitate the de novo inference of high-quality models from paired-end in vivo binding data such as ChIP-seq, ChIP-exo, and CUT&Tag without read mapping or peak calling, we introduce an extensible reimplementation of our biophysically interpretable machine learning framework named PyProBound. Explicitly accounting for assay-specific bias in DNA fragmentation rate when training on ChIP-seq yields improved TF binding models. Moreover, we show how PyProBound can leverage our threshold-free ASB likelihood function to perform de novo motif discovery using allele-specific ChIP-seq counts.ConclusionOur work provides new strategies for predicting the functional impact of non-coding variants.

The interaction of a self-assembled nanoparticle and a lipid membrane: Binding, disassembly and re-distribution

Here we report a detailed study of the interactions of nanoparticles, formed by the self-assembly of cholesterol-containing porphyrins, with lipid membranes. We show that the interaction is a two-step process: first, the docking and fusion, then, the redistribution of the building blocks of the self-assembled nanoparticles (SANs henceforth). Analysis of the binding and structural data is consistent with the docking step being driven by a multivalence cooperative effect and with the formation of SAN aggregates on the membrane, whilst the solubility of the cholesterol anchor in the membrane is key to both the fusion and redistribution of the SANs building blocks. The tendency of the SAN to aggregate in the membrane helps explain the photosensitizer properties of the SANs, essential to their anti-microbial activity. The solubility of the cholesteryl anchors drives fusion to the membrane and de-assembly of the SAN, explaining the capability of the SANs to deliver therapeutic cargos at the lipid interface. The subsequent redistribution of the SANs building blocks offer a plausible pathway to body clearance that is not immediately available to hard nanoparticles. These properties, and the modularity of the synthesis, point to the SANs being an excellent platform for the development of nanomedicines. An unexpected consequence of unraveling the mechanism of membrane interaction of these SANs is that it allows us to derive a value of the free energy of binding of cholesterol (the membrane anchor of the SAN building blocks) to a lipid membrane, that is consistent with the literature values. This is an additional property that can be exploited to determine the affinity of a variety of membrane anchors to membranes of various compositions.

Reconstructing the sequence specificities of RNA-binding proteins across eukaryotes.

RNA-binding proteins (RBPs) are key regulators of gene expression. Here, we introduce EuPRI (Eukaryotic Protein-RNA Interactions) - a freely available resource of RNA motifs for 34,736 RBPs from 690 eukaryotes. EuPRI includes in vitro binding data for 504 RBPs, including newly collected RNAcompete data for 174 RBPs, along with thousands of reconstructed motifs. We reconstruct these motifs with a new computational platform - Joint Protein-Ligand Embedding (JPLE) - which can detect distant homology relationships and map specificity-determining peptides. EuPRI quadruples the number of known RBP motifs, expanding the motif repertoire across all major eukaryotic clades, and assigning motifs to the majority of human RBPs. EuPRI drastically improves knowledge of RBP motifs in flowering plants. For example, it increases the number of Arabidopsis thaliana RBP motifs 7-fold, from 14 to 105. EuPRI also has broad utility for inferring post-transcriptional function and evolutionary relationships. We demonstrate this by predicting a role for 12 Arabidopsis thaliana RBPs in RNA stability and identifying rapid and recent evolution of post-transcriptional regulatory networks in worms and plants. In contrast, the vertebrate RNA motif set has remained relatively stable after its drastic expansion between the metazoan and vertebrate ancestors. EuPRI represents a powerful resource for the study of gene regulation across eukaryotes.

Cellular Phosphate Sensing and Anion Binding by an Azacrown-Calixpyrrole Hybrid.

A hybrid receptor-sensor for anions originating from the merging of positively charged ammonium moieties for electrostatic attraction/stronger binding of azacrowns with directionality of calixpyrrole hydrogen bond donors for selectivity is investigated. As demonstrated this hybrid receptor-sensor shows a remarkable selectivity for orthophosphate even in the presence of other phosphates and anions found in cellular materials (Kassoc H2PO4 ->H2P2O7 2->AMP-≫ADP2- or ATP3- over halides, nitrate, or hydrogen sulfate; all Na+ salts in water) but also cellular polyphosphate or phospholipids. This selectivity is harnessed in a real-time monitoring of cell lysis by lysozyme, which releases orthophosphate and other phosphates and anions from the cells. This sensitive (LOD 0.4 μM) fluorescence-based microscale method compares favorably with the state-of-the-art techniques but can easily be practiced in a high-throughput screening (HTS) manner. The anion binding and selectivity in aqueous solutions were investigated by NMR and put in context with phosphate binding of the parent calix[4]pyrrole. The microscopic understanding of anion binding by the hybrid receptor was then obtained from a combination of density functional theory (DFT), classical molecular dynamics (MD) with explicit water solvation, and ab initio MD (AIMD) simulations. Correlating the NMR and fluorescence binding data with studies of solvation of the receptor, phosphate anion, and the resulting complex confirms the binding is largely driven by entropic component (TΔS) associated with receptor and anion desolvation.

Probing the mechanism of nick searching by LIG1 at the single-molecule level.

DNA ligase 1 (LIG1) joins Okazaki fragments during the nuclear replication and completes DNA repair pathways by joining 3'-OH and 5'-PO4 ends of nick at the final step. Yet, the mechanism of how LIG1 searches for a nick at single-molecule level is unknown. Here, we combine single-molecule fluorescence microscopy approaches, C-Trap and total internal reflection fluorescence (TIRF), to investigate the dynamics of LIG1-nick DNA binding. Our C-Trap data reveal that DNA binding by LIG1 full-length is enriched near the nick sites and the protein exhibits diffusive behavior to form a long-lived ligase/nick complex after binding to a non-nick region. However, LIG1 C-terminal mutant, containing the catalytic core and DNA-binding domain, predominantly binds throughout DNA non-specifically to the regions lacking nick site for shorter time. These results are further supported by TIRF data for LIG1 binding to DNA with a single nick site and demonstrate that a fraction of LIG1 full-length binds significantly longer period compared to the C-terminal mutant. Overall comparison of DNA binding modes provides a mechanistic model where the N-terminal domain promotes 1D diffusion and the enrichment of LIG1 binding at nick sites with longer binding lifetime, thereby facilitating an efficient nick search process.

An in vitro and machine learning framework for quantifying serum albumin binding of per- and polyfluoroalkyl substances

Abstract Per- and polyfluoroalkyl substances (PFAS) are a diverse class of anthropogenic chemicals; many are persistent, bioaccumulative, and mobile in the environment. Worldwide, PFAS bioaccumulation causes serious adverse health impacts, yet the physiochemical determinants of bioaccumulation and toxicity for most PFAS are not well understood, largely due to experimental data deficiencies. As most PFAS are proteinophilic, protein binding is a critical parameter for predicting PFAS bioaccumulation and toxicity. Among these proteins, human serum albumin (HSA) is the predominant blood transport protein for many PFAS. We previously demonstrated the utility of an in vitro differential scanning fluorimetry assay for determining relative HSA binding affinities for 24 PFAS. Here, we report HSA affinities for 65 structurally diverse PFAS from 20 chemical classes. We leverage these experimental data, and chemical/molecular descriptors of PFAS, to build 7 machine learning classifier algorithms and 9 regression algorithms, and evaluate their performance to identify the best predictive binding models. Evaluation of model accuracy revealed that the top performing classifier model, logistic regression, had an AUROC statistic of 0.936. The top performing regression model, support vector regression, had an R2 of 0.854. These top performing models were then used to predict HSA-PFAS binding for chemicals in the EPAPFASINV list of 430 PFAS. These developed in vitro and in silico methodologies represent a high-throughput framework for predicting protein-PFAS binding based on empirical data, and generate directly comparable binding data of potential use in predictive modeling of PFAS bioaccumulation and other toxicokinetic endpoints.

Integrating computational and experimental approaches in discovery and validation of MmpL3 pore domain inhibitors for specific labelling of Mycobacterium tuberculosis

Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb), continues to pose a significant global health threat. Identifying new druggable targets is crucial for the advancement of drug development. Equally critical is the development of precise methods for monitoring Mtb to effectively combat this disease. Addressing these needs, our study pinpointed the pore domain (PD) of MtbMmpL3 as a new binding site for virtual screening, which led to the discovery of the small molecule ZY27. To confirm the binding site and action mode of ZY27, we employed cosolvent molecular dynamics (CMD), steered molecular dynamics (SMD), and long timescale molecular dynamics (MD) simulations of 5 μs. These in silico studies verified that ZY27 binds to the PD of MtbMmpL3. In antimicrobial activity tests, ZY27 exhibited potent anti-Mtb activity and high selectivity among mycobacterial species. Whole-genome sequencing of spontaneous ZY27-resistant Mtb variants, complemented by acid-fast staining experiments, confirmed that ZY27 specifically targets MtbMmpL3. Utilizing the ligand-protein binding data, we designed and synthesized two solvatochromic fluorescent probes, 27FP1 and 27FP2, based on ZY27. Further investigations through flow cytometry and confocal microscopy confirmed that these probes specifically label Mtb cells via the MtbMmpL3 binding mechanism.

Effective Inclusion of Electronic Polarization Improves the Description of Electrostatic Interactions: The prosECCo75 Biomolecular Force Field.

prosECCo75 is an optimized force field effectively incorporating electronic polarization via charge scaling. It aims to enhance the accuracy of nominally nonpolarizable molecular dynamics simulations for interactions in biologically relevant systems involving water, ions, proteins, lipids, and saccharides. Recognizing the inherent limitations of nonpolarizable force fields in precisely modeling electrostatic interactions essential for various biological processes, we mitigate these shortcomings by accounting for electronic polarizability in a physically rigorous mean-field way that does not add to computational costs. With this scaling of (both integer and partial) charges within the CHARMM36 framework, prosECCo75 addresses overbinding artifacts. This improves agreement with experimental ion binding data across a broad spectrum of systems─lipid membranes, proteins (including peptides and amino acids), and saccharides─without compromising their biomolecular structures. prosECCo75 thus emerges as a computationally efficient tool providing enhanced accuracy and broader applicability in simulating the complex interplay of interactions between ions and biomolecules, pivotal for improving our understanding of many biological processes.

Improved Prediction of Stabilizing Mutations in Proteins by Incorporation of Mutational Effects on Ligand Binding.

While many computational methods accurately predict destabilizing mutations, identifying stabilizing mutations has remained a challenge, because of their relative rarity. We tested ΔΔG0 predictions from computational predictors such as Rosetta, ThermoMPNN, RaSP, and DeepDDG, using 82 mutants of the bacterial toxin CcdB as a test case. On this dataset, the best computational predictor is ThermoMPNN, which identifies stabilizing mutations with a precision of 68%. However, the average increase in Tm for these predicted mutations was only 1°C for CcdB, and predictions were poorer for a more challenging target, influenza neuraminidase. Using data from multiple previously described yeast surface display libraries and in vitro thermal stability measurements, we trained logistic regression models to identify stabilizing mutations with a precision of 90% and an average increase in Tm of 3°C for CcdB. When such libraries contain a population of mutants with significantly enhanced binding relative to the corresponding wild type, there is no benefit in using computational predictors. It is then possible to predict stabilizing mutations without any training, simply by examining the distribution of mutational binding scores. This avoids laborious steps of in vitro expression, purification, and stability characterization. When this is not the case, combining data from computational predictors with high-throughput experimental binding data enhances the prediction of stabilizing mutations. However, this requires training on stability data measured in vitro with known stabilized mutants. It is thus feasible to predict stabilizing mutations rapidly and accurately for any system of interest that can be subjected to a binding selection or screen.

High-affinity agonism at the P2X7 receptor is mediated by three residues outside the orthosteric pocket

P2X receptors are trimeric ATP-gated ion channels that activate diverse signaling cascades. Due to its role in apoptotic pathways, selective activation of P2X7 is a potential experimental tool and therapeutic approach in cancer biology. However, mechanisms of high-affinity P2X7 activation have not been defined. We report high-resolution cryo-EM structures of wild-type rat P2X7 bound to the high-affinity agonist BzATP as well as significantly improved apo receptor structures in the presence and absence of sodium. Apo structures define molecular details of pore architecture and reveal how a partially hydrated Na+ ion interacts with the conductance pathway in the closed state. Structural, electrophysiological, and direct binding data of BzATP reveal that three residues just outside the orthosteric ATP-binding site are responsible for its high-affinity agonism. This work provides insights into high-affinity agonism for any P2X receptor and lays the groundwork for development of subtype-specific agonists applicable to cancer therapeutics.

Oligomerization of protein arginine methyltransferase 1 and its effect on methyltransferase activity and substrate specificity.

Proper protein arginine methylation by protein arginine methyltransferase 1 (PRMT1) is critical for maintaining cellular health, while dysregulation is often associated with disease. How the activity of PRMT1 is regulated is therefore paramount, but is not clearly understood. Several studies have observed higher order oligomeric species of PRMT1, but it is unclear if these exist at physiological concentrations and there is confusion in the literature about how oligomerization affects activity. We therefore sought to determine which oligomeric species of PRMT1 are physiologically relevant, and quantitatively correlate activity with specific oligomer forms. Through quantitative western blotting, we determined that concentrations of PRMT1 available in a variety of human cell lines are in the sub-micromolar to low micromolar range. Isothermal spectral shift binding data were modeled to a monomer/dimer/tetramer equilibrium with an EC50 for tetramer dissociation of ~20 nM. A combination of sedimentation velocity and Native polyacrylamide gel electrophoresis experiments directly confirmed that the major oligomeric species of PRMT1 at physiological concentrations would be dimers and tetramers. Surprisingly, the methyltransferase activity of a dimeric PRMT1 variant is similar to wild type, tetrameric PRMT1 with some purified substrates, but dimer and tetramer forms of PRMT1 show differences in catalytic efficiencies and substrate specificity for other substrates. Our results define an oligomerization paradigm for PRMT1, show that the biophysical characteristics of PRMT1 are poised to support a monomer/dimer/tetramer equilibrium in vivo, and suggest that the oligomeric state of PRMT1 could be used to regulate substrate specificity.

Neural network extrapolation to distant regions of the protein fitness landscape

Machine learning (ML) has transformed protein engineering by constructing models of the underlying sequence-function landscape to accelerate the discovery of new biomolecules. ML-guided protein design requires models, trained on local sequence-function information, to accurately predict distant fitness peaks. In this work, we evaluate neural networks’ capacity to extrapolate beyond their training data. We perform model-guided design using a panel of neural network architectures trained on protein G (GB1)-Immunoglobulin G (IgG) binding data and experimentally test thousands of GB1 designs to systematically evaluate the models’ extrapolation. We find each model architecture infers markedly different landscapes from the same data, which give rise to unique design preferences. We find simpler models excel in local extrapolation to design high fitness proteins, while more sophisticated convolutional models can venture deep into sequence space to design proteins that fold but are no longer functional. We also find that implementing a simple ensemble of convolutional neural networks enables robust design of high-performing variants in the local landscape. Our findings highlight how each architecture’s inductive biases prime them to learn different aspects of the protein fitness landscape and how a simple ensembling approach makes protein engineering more robust.

Highly quantitative measurement of differential protein-genome binding with PerCell chromatin sequencing.

We report a universal strategy for 2D chromatin sequencing, to increase uniform data analyses and sharing across labs, and to facilitate highly quantitative comparisons across experimental conditions. Within our system, we provide wetlab and drylab tools for researchers to establish and analyze protein-genome binding data with PerCell ChIP-seq. Our methodology is virtually no cost and flexible, enabling rapid, quantitative, internally normalized chromatin sequencing to catalyze project development in a variety of systems, including in vivo zebrafish epigenomics and cancer cell epigenomics. While we highlight utility in these key areas, our methodology is flexible enough such that rapid comparisons of cellular spike-in versus non spike-in are possible, and generalizability to nuclease-based 2D chromatin sequencing would also be possible within the framework of our pipeline. Through the use of well-defined cellular ratios containing orthologous species' chromatin, we enable cross-species comparative epigenomics and highly quantitative low-cost chromatin sequencing with utility across a range of disciplines.

Immunoreactivity of eastern small eyed snake (Cryptophis nigrescens) venom towards species-specific antibodies of five medically important venomous Australian elapids.

The eastern small eyed snake (Cryptophis nigrescens; CN) is an uncommon cause of snakebite in Australia despite the widespread distribution of the snake along the east coast of Australia. Diagnosis of envenomation relies on identification of the snake which is often not possible with animal snakebite cases. This study examined the immunoreactivity profile of CN venom towards specific rabbit IgG made against the medically relevant snake venom immunotypes found in Australia (tiger, brown, black, death adder and taipan). A simultaneous sandwich ELISA format was used to quantify CN venom binding to venom specific Protein A purified rabbit IgG. The binding profiles demonstrated weak binding of CN venom to rabbit IgG made against both tiger (N. scutatus) and black snake (P. australis) venoms with approximately 0.19% and 0.069% cross reactivity, respectively. However, the concentration of venom likely to be present in the urine of CN envenomed patients and the low cross reactivity suggest that envenomed veterinary patients are unlikely to be detected in the commercial snake venom detection kit. It is possible that CN envenomation is more common but may be underdiagnosed where snake venom antigen detection is relied upon solely. Serum biochemical abnormalities also overlap with other snake species found in the same geographical area. In respect of antivenom therapy, administration of tiger snake antivenom is supported by the binding data, but due to the low cross reactivity multiple vials may be required. Limited clinical evidence also supports the efficacy of tiger snake antivenom for envenomation by CN.

Leveraging Machine Learning-Guided Molecular Simulations Coupled with Experimental Data to Decipher Membrane Binding Mechanisms of Aminosterols.

Understanding the molecular mechanisms of the interactions between specific compounds and cellular membranes is essential for numerous biotechnological applications, including targeted drug delivery, elucidation of the drug mechanism of action, pathogen identification, and novel antibiotic development. However, estimation of the free energy landscape associated with solute binding to realistic biological systems is still a challenging task. In this work, we leverage the Time-lagged Independent Component Analysis (TICA) in combination with neural networks (NN) through the Deep-TICA approach for determining the free energy associated with the membrane insertion processes of two natural aminosterol compounds, trodusquemine (TRO), and squalamine (SQ). These compounds are particularly noteworthy because they interact with the outer layer of neuron membranes, protecting them from the toxic action of misfolded proteins involved in neurodegenerative disorders, in both their monomeric and oligomeric forms. We demonstrate how this strategy could be used to generate an effective collective variable for describing solute absorption in the membrane and for estimating free energy landscape of translocation via on-the-fly probability enhanced sampling (OPES) method. In this context, the computational protocol allowed an exhaustive characterization of the aminosterol entry pathway into a neuron-like lipid bilayer. Furthermore, it provided accurate prediction of membrane binding affinities, in close agreement with the experimental binding data obtained by using fluorescently labeled aminosterols and large unilamellar vesicles (LUVs). The findings contribute significantly to our understanding of aminosterol entry pathways and aminosterol-lipid membrane interactions. Finally, the computational methods deployed in this study further demonstrate considerable potential for investigating membrane binding processes.

Discovery and Characterization of a Chemical Probe for Cyclin-Dependent Kinase-Like 2.

Acylaminoindazole-based inhibitors of CDKL2 were identified via analyses of cell-free binding and selectivity data. Compound 9 was selected as a CDKL2 chemical probe based on its potent inhibition of CDKL2 enzymatic activity, engagement of CDKL2 in cells, and excellent kinome-wide selectivity, especially when used in cells. Compound 16 was designed as a negative control to be used alongside compound 9 in experiments to interrogate CDKL2-mediated biology. A solved cocrystal structure of compound 9 bound to CDKL2 highlighted key interactions it makes within its ATP-binding site. Inhibition of downstream phosphorylation of EB2, a CDKL2 substrate, in rat primary neurons provided evidence that engagement of CDKL2 by compound 9 in cells resulted in inhibition of its activity. When used at relevant concentrations, compound 9 does not impact the viability of rat primary neurons or certain breast cancer cells nor elicit consistent changes in the expression of proteins involved in epithelial-mesenchymal transition.

Preclinical mitigation of 5-HT2B agonism-related cardiac valvulopathy revisited

Cardiac valvulopathy (Cardiac Valve Disease; CVD) associated with off-target activation of the 5-hydroxytryptamine (5-HT) 2B receptor has been well recognized, but is still poorly predicted during drug development. The regulatory guidance proposes the use of 5-HT2B binding data (i.e., Ki values) and free maximum therapeutic exposure (Cmax) to calculate safety margins as a threshold of detection (>10) for eliminating the risk of drug-induced cardiac valvulopathy. In this paper, we provide additional recommendations for preclinical prediction of CVD risk based on clinical pharmacodynamic and pharmacokinetic data obtained from drugs with or without 5-HT2B receptor activation. Our investigations showed that 5-HT2B agonist affinity of molecules tested in an in vitro 5-HT2B cell-based functional assay, placed in perspective to their sustained plasma exposure (AUCs) and not to their peak plasma exposure, Cmax (i.e., maximum therapeutic exposure) provide a solid basis for interpreting 5-HT2B data, for calculating safety margins and then, accurately differentiate drugs associated with a clinical risk of CVD from those which are not (despite having some agonist 5-HT2B activity). In addition, we discuss the risk of multi-organ fibrosis linked to 5-HT2B receptor activation, often underestimated, however well reported in FAERS for 5-HT2B agonists. We believe that our recommendations have the potential to mitigate the risk for the clinical development of CVD and fibrosis.