Specific Hydrogen Bonds Research Articles

Design, Synthesis and Characterization of New Amide-Linked Chromone-Isoxazole Hybrids: In Vitro Anti-Bacterial and Antioxidant Evaluation, DFT Calculations, ADMET Profiling, Docking and Molecular dynamics simulation

In the present study, we report the synthesis, and biological evaluation of a new series of amide-linked chromone-isoxazole hybrids synthesized through a highly regioselective 1,3-dipolar cycloaddition (1,3-DC) reaction in a reproducible manner. The resulting products structures were identified by means of several spectroscopic techniques, including 13C NMR, 1H NMR, IR and MS spectrometry. In addition, the regio- and chemoselective synthesis of the newly synthesized chromone-isoxazole carboxamide hybrids were investigated via Density Functional Theory (DFT) calculations at the B3LYP/6-311G(d,p) level, which provided insights into molecular reactivity. Global and local reactivity indexes as well as the non-covalent interactions (NCI) analyses, and the findings explain the observed selectivity of the corresponding product. The antioxidant activity of all the new compounds was examined through the DPPH free radical scavenging assay, and the findings demonstrated a significant free radical scavenging power for the tested compounds. Notably, products (5c) and (5e) exhibited encouraging antioxidant effect. Antibacterial screening of the chromone-isoxazole hybrids against Escherichia coli, Porteus mirabilis, Bacillus subtilis and Staphylococcus aureus showed moderate activity, with compound 1 demonstrating the most promising antibacterial efficacy. To further explore the compound 1’s antibacterial potential, we employed molecular docking, dynamics simulations, ligand-protein contact analysis and ADMET profiling. Docking results revealed that compound 1 forms specific hydrogen bonds and hydrophobic interactions with bacterial targets 2W9S and 10FO, distinct from the interaction observed with standard antibacterial drug. Molecular dynamics simulation indicated stable protein-ligand conformations, despite slight ligand flexibility, and contact analysis identified a critical interaction stabilizing the complex. ADMET analysis confirmed that compound 1 possesses good drug-like properties, including oral bioavailability and low toxicity, indicating its potential as an effective antibacterial agent.

Assembly of 2′,3′-Cyclic guanosine Monophosphate-Adenosine monophosphate and their spontaneous intracellular disassembly for enhanced antitumor immunity via natural STING pathway activation

The stimulator of interferon genes (STING) pathway plays an important role in remodeling the immunosuppressive tumor microenvironment (TME). Cyclic dinucleotides (CDNs), the natural ligands of STING, activate innate immunity to enhance tumor immunogenicity. However, the anionic properties of CDNs result in poor membrane permeability, while their rapid metabolism by enzymes hinders the efficient targeting and activation of STING in vitro and in vivo. To improve the bioavailability and antitumor immunity of CDNs, we designed a hydrogen-bonding-based supramolecular approach for 2′,3′-cyclic guanosine monophosphate (GMP)-adenosine monophosphate (AMP) (cGAMP). Leveraging the synthetic molecules cytosine-uracil-hexane (CUH), we formulated supramolecular micelles of cGAMP via specific hydrogen bonds in aqueous solutions. Micelles comprising the CUH-cGAMP complex enhanced the biological efficacy of cGAMP by increasing its stability, thereby prolonging the activation of STING-mediated innate and adaptive immunity. We demonstrated that CUH-cGAMP stands as a promising candidate for immunotherapy, exhibiting significant inhibition of tumor growth in a CT26 syngeneic mouse model.

Visualization of nonsmall-cell lung cancer by near-infrared fluorescence imaging with tumor-targeting peptide ABT-510

Nonsmall-cell lung cancer (NSCLC) is the most frequent type of lung cancer, with early surgical treatment proving vital for prolonged patient survival. However, precise visualization of NSCLC remains a challenge due to limited molecular imaging probes, the unique anatomical structure of the lungs, and respiratory movement interference. In this study, we investigated the potential utility of CD36, which is highly expressed in NSCLC, as an imaging target. A selective and water-soluble fluorescent probe, MPA–ABT-510, was successfully constructed by coupling ABT-510 with MPA, a near-infrared (NIR) fluorescent dye. Molecular docking analysis shows that MPA–ABT-510 possesses strong binding affinity to the CD36 protein, with specific hydrogen bond interactions at defined amino acid residues. In vitro assays reveals that the fluorescein isothiocyanate-labeled peptide ABT-510 specifically binds to the CD36-high expressing NSCLC cell lines H1299 and A549. In vivo imaging verifies that the MPA–ABT-510 efficiently accumulates in the tumor site with a distinct fluorescent signal. Ex vivo imaging revealed that tumor-to-lung fluorescence ratios for subcutaneous and orthotopic H1299 mouse models were 7.19 ± 0.73 and 1.91 ± 0.42, respectively, while those for A549 mice were 5.53 ± 0.64 and 1.77 ± 0.41, respectively. Biodistribution analysis demonstrated efficient MPA–ABT-510 uptake in H1299 and A549 liver metastases models with tumor-to-liver fluorescence ratios of 2.47 ± 0.48 and 2.19 ± 0.22, respectively. High MPA–ABT-510 accumulation was evident in A549 intestinal metastases models, as evidenced by tumor-to-colorectal fluorescence ratios of 4.27 ± 0.11. MPA–ABT-510 exhibits superior imaging capabilities with minimal safety concerns, so it is a promising candidate for NSCLC surgical navigation.

Effect of DMSO and Triacetin Solvents on Polyvinylidene Fluoride Polymorphs: Molecular Dynamics Simulations

In this article, molecular dynamics simulations are performed to investigate the interaction between polyvinylidene fluoride (PVDF) polymorphs (β phase has more properties compared with α, such as polarity, higher mechanical strength, and piezoelectric, ferroelectric, and pyroelectric properties), and two solvents, dimethyl sulfoxide (DMSO) and glyceryl triacetate (GTA). Pure solvent boxes are built using two force fields (Compass and Dreiding) to study the stabilization of their density and solubility, which are used with the PVDF surfaces (for each solvent, computations are performed using two surfaces α and β). According to the radial distribution function results, DMSO shows a stronger interaction with two PVDF phases, allowing it to orient chains from the α phase to the β phase, while GTA exhibits a weaker interaction due to specific hydrogen bonds. These findings are confirmed by the experiment, where the fraction F(β) reaches 62% when using DMSO, while it is 36% when using GTA.

Comparative in-silico analysis of vitexin and orientin as potential antiphotoaging agents against MMP enzymes

Photoaging, a result of excessive UV exposure, increases ROS production and collagen degradation by MMPs, causing skin wrinkles and roughness. This study explores the potential of vitexin and orientin as natural antiphotoaging agents through in-silico molecular docking, comparing their efficacy against retinol in inhibiting MMP-1, MMP-3, and MMP-9 enzymes involved in photoaging. The research utilized Hyperchem 8 for compound optimization, Chimera 1.11 for target protein preparation, and AutodockTools 1.5.6 for docking analysis. Results demonstrated that vitexin and orientin exhibit stronger affinity towards MMP-1, MMP-3, and MMP-9, indicated by more negative binding energies than retinol. Their interaction with the MMP enzymes, characterized by specific hydrogen bonds with key amino acid residues, suggests a potent inhibitory effect. This affinity indicates vitexin and orientin’s potential as effective antiphotoaging agents, providing a basis for further exploration in skin care applications.

Entropy-Driven Design of Highly Impact-Stiffening Supramolecular Polymer Networks with Salt-Bridge Hydrogen Bonds.

Impact-stiffening materials that undergo a strain rate-induced soft-to-rigid transition hold great promise as soft armors in the protection of the human body and equipment. However, current impact-stiffening materials, such as polyborosiloxanes and shear-thickening fluids, often exhibit a limited impact-stiffening response. Herein, we propose a design strategy for fabricating highly impact-stiffening supramolecular polymer networks by leveraging high-entropy-penalty physical interactions. We synthesized a fully biobased supramolecular polymer comprising poly(α-thioctic acid) and arginine clusters, whose chain dynamics are governed by highly specific guanidinium-carboxylate salt-bridge hydrogen bonds. The resulting material exhibits an exceptional impact-stiffening response of ∼2100 times, transitioning from a soft dissipating state (21 kPa, 0.1 Hz) to a highly stiffened glassy state (45.3 MPa, 100 Hz) with increasing strain rates. Moreover, the material's high energy-dissipating and hot-melting properties bring excellent damping performance and easy hybridization with other scaffolds. This entropy-driven approach paves the way for the development of next-generation soft, sustainable, and impact-resistant materials.

Machine-learning accelerated structure search for ligand-protected clusters.

Finding low-energy structures of ligand-protected clusters is challenging due to the enormous conformational space and the high computational cost of accurate quantum chemical methods for determining the structures and energies of conformers. Here, we adopted and utilized a kernel rigid regression based machine learning method to accelerate the search for low-energy structures of ligand-protected clusters. We chose the Au25(Cys)18 (Cys: cysteine) cluster as a model system to test and demonstrate our method. We found that the low-energy structures of the cluster are characterized by a specific hydrogen bond type in the cysteine. The different configurations of the ligand layer influence the structural and electronic properties of clusters.

Synthesis, molecular docking and pharmacological evaluations of novel naphthalene-pyrazoline hybrids as new orally active anti-inflammatory agents

The pyrazoline nucleus is widely present in diverse compounds as it serves as a valuable template for the discovery of new class of bioactive compounds. Various substituted 2-pyrazolines exhibit a spectrum of pharmacological activities, including antimicrobial, anti-inflammatory, analgesic, and antitumor properties. In view of this, we have synthesized a novel class of naphthalene-pyrazoline hybrids (PY1-PY20). All the synthesized hybrids were thoroughly characterized by means of standard spectroscopic techniques and tested for their in-vivo anti-inflammatory activity. Most of the compounds exhibited significant anti-inflammatory activity and in particular the compounds bearing electron withdrawing groups had potential in vivo activity. Further, in-silico molecular docking studies were carried out using the ‘Molegro Virtual Docker’ as a docking tool against the target proteins such as COX-1 (PDB ID: 1EQH) and COX-2 (PDB ID: 1PXX). The results revealed that the majority of pyrazolines form hydrogen interactions with key residues (Ser 530, Tyr 385, Tyr 355, Arg 120, Ser 353, Arg 83) in the active binding sites of COX-1 (1EQH) and COX-2 (1PXX). Notably, interactions with residues Tyr 355, Arg 120, Ser 530 for COX-1, and Tyr 355, Arg 120, Gln 192, Leu 352, Ser 353, His 90, Tyr 38, Ser 530, and Tyr 385 for COX-2 are crucial for the reported anti-inflammatory activity of diverse compounds. Ligand-protein inverse molecular docking studies highlight specific hydrogen bond interactions, with PY-8, PY-11, PY-15, and PY-17 showing significant interactions. The study was able to identify potential anti-inflammatory naphthalene-pyrazoline derivatives.

Innovative Piperidine-Catalyzation in Protecting Carbonyl Compounds with Implications for Angiogenesis and Inflammation

Introduction: This study explores the application of piperidine-catalyzed protection, an innovative technique, for safeguarding various carbonyl compounds functioning as acetals in a generic reaction. Specifically, we investigate the catalysis of 2-amino-1,3-propanediol-2-methyl with various aromatic aldehydes, leading to the production of cis and trans-5-methyl-2-(mono or di-substitutions-phenyl)-1,3-dioxane-5-amine. Methods: Identification of all newly formed products is achieved through the utilization of spectroscopic techniques, including IR, 1H-NMR, and 13C-NMR spectroscopy. Molecular interactions and potential therapeutic applications of compounds E1 and E2 are demonstrated with cAMP-specific phosphodiesterase (1zkl) and oxidized purine nucleoside triphosphate hydrolase (5ws7). Detailed structural analyses highlight specific hydrogen bonds, pi-pi stacked interactions, and alkyl contacts formed by these compounds with target proteins. A comprehensive bioinformatics approach involves GO enrichment, STRING protein-protein association networks, KEGG pathway analysis, and Reactome pathways to elucidate biological processes, molecular functions, and cellular components associated with the compounds. Results: Compounds E1 and E2 exhibit diverse enrichment profiles, suggesting their involvement in various signaling cascades, neurotransmission, immune responses, and cancer pathways. Comparative analysis of five compound pairs (A1/A2, B1/B2, C1/C2, D1/D2, E1/E2) reveals subtle distinctions in enrichment patterns, implying unique pharmacological advantages for each pair. Conclusion: This study introduces a separate investigation on piperidine-catalyzed protection of carbonyl groups in aldehydes, presenting a practical approach. Emphasizing the significance of understanding distinct biological signatures, our findings guide therapeutic applications and compound optimization, particularly in the context of anti-cancer therapeutics. The compounds show potential in modulating neuronal function, neurotransmission, cancer mechanisms, and immune responses, suggesting promising avenues for future research and development.

Bioinspired Herbicides-Ionic Liquids or Liquid Cocrystals?

This research provides information about combinations of several amino acids, including l-proline (Pro), l-arginine (Arg), and l-histidine (His), with phenoxyacetic acid herbicides (MCPA and 2,4-D). Five amino acid ionic liquids (AAILs), one amino acid higher-melting salt (AAHMS), and two amino acid liquid cocrystals (AALCs) were obtained in high yields (>90%). The ionization of the six new structures was confirmed by NMR, IR, and molecular modeling. X-ray crystallography was used to definitively confirm the binding location of the mobile hydrogen. Furthermore, we propose a computational method for estimating the energy of specific hydrogen bond(s) in AAIL crystals based on the NBO and QTAIM hydrogen bond parameters obtained by model calculations. An in-depth analysis of the structures allowed to answer the question posed in the title, ionic liquids or liquid cocrystals? AAILs based on arginine and histidine were obtained. In contrast, combining proline with MCPA and 2,4-D led to AALCs. Finally, the compounds were analyzed to measure their herbicidal activity. These studies proved that the novel form of MCPA or 2,4-D improved its ability to control weeds compared to commercial formulations containing the same active ingredients.

High-strength and ultra-tough supramolecular polyamide spider silk fibers assembled via specific covalent and reversible hydrogen bonds

Achieving ultra-high tensile strength and exceptional toughness is a longstanding goal for structural materials. However, previous attempts using covalent and non-covalent bonds have failed, leading to the belief that these two properties are mutually exclusive. Consequently, commercial fibers have been forced to compromise between tensile strength and toughness, as seen in the differences between nylon and Kevlar. To address this challenge, we drew inspiration from the disparate tensile strength and toughness of nylon and Kevlar, both of which are polyamide fibers, and developed an innovative approach that combines specific intermolecular disulfide bonds and reversible hydrogen bonds to create ultra-strong and ultra-tough polyamide spider silk fibers. Our resulting Supramolecular polyamide spider silk, which has a maximum molecular weight of 1084 kDa, exhibits high tensile strength (1180 MPa) and extraordinary toughness (433 MJ/m3), surpassing Kevlar's toughness 8-fold. This breakthrough presents a new opportunity for the sustainable development of spider silk as an environmentally friendly alternative to synthetic commercial fibers, as spider silk is composed of amino acids. Future research could explore the use of these techniques and fundamental knowledge to develop other super materials in various mechanical fields, with the potential to improve people's lives in many ways. Statement of significance• By emulating synthetic commercial fibers such as nylon and polyethylene, we have successfully produced supramolecular-weight polyamide spider silk fibers with a molecular weight of 1084 kDa through a unique covalent bond-mediated linear polymerization reaction of spider silk protein molecules. This greatly surpasses the previous record of a maximum molecular weight of 556 kDa. • We obtained supramolecular polyamide spider silk fibers with both high-tensile strength and toughness. The stress at break is 1180 MPa, and the toughness is 8 times that of kevlar, reaching 433 MJ/m3. • Our results challenge the notion that it is impossible to manufacture fibers with both ultra-high tensile strength and ultra-toughness, and provide theoretical guidance for developing environmentally friendly and sustainable structural materials that meet industrial needs.

Tautomeric contributions to the absorption spectrum of [2,2′-bipyridyl]-3,3′-diol in water unveiled by molecular dynamics with accurate quantum mechanically derived force-fields

In this work, we provide a realistic description of the tautomers of [2,2′-bipyridyl]-3,3′-diol (BPOH2) and their relative populations in water solution, eventually validating the computational predictions through the comparison of the simulated absorption spectrum with the one experimentally measured at ambient conditions. To this end, we present an original computational protocol, which exploits accurate and specific quantum mechanically derived force-fields in a MD/QM sequential procedure, allowing for a first principle characterization of all BPOH2 tautomers, including their thermalized conformational dynamics and relative Boltzmann weights. To our knowledge, this represents an unprecedented result demonstrating that the lower energy bands, appearing in water, are due to the absorption of the diketo tautomer of BPOH2, which is stabilized by a strong and specific hydrogen bond network. Considering its general applicability, the proposed computational procedure paves the way to future application on more complex systems of biological or technological relevance, as for instance protic solvents micro-confined in large organic aggregates.

Excited-state symmetry breaking is an ultrasensitive tool for probing microscopic electric fields.

Microscopic electric fields are increasingly found to play a pivotal role in catalysis of enzymatic and chemical reactions. Currently, the vibrational Stark effect is the main experimental method used to measure them. Here, we demonstrate how excited-state symmetry breaking can serve as a much more sensitive tool to assess these fields. Using transient infrared spectroscopy on a quadrupolar probe equipped with nitrile groups we demonstrate both its superior sensitivity and that it does not suffer from the notorious hydrogen-bond induced upshift of the C[triple bond, length as m-dash]N stretch frequency. In combination with conventional ground-state infrared absorption, excited-state symmetry breaking can be used to disentangle even weak specific hydrogen bond interactions from general field effects. We showcase this capability with the example of weak C-H hydrogen bonds in polar aprotic solvents. Additionally, we reveal for the first time symmetry breaking driven not by solvent but by the entropy of the pendant side chains of the chromophore. Our findings not only enhance our understanding of symmetry-breaking charge-transfer phenomena but pave the way toward using them in electric field sensing modality.

A novel FA1-targeting fluorescent probe for specific discrimination and identification of human serum albumin from bovine serum albumin.

A novel probe C1 combining benzothiazole with a spiropyran section was developed for the specific detection of human serum albumin (HSA). The molecular docking suggested that the sulphonic acid group modification allowed C1 to form specific hydrogen bonds with lysine (Lys137) at fatty acid site 1 (FA1) of HSA, thus enabling fluorescence differentiation between HSA and BSA.

Dynamics of the splicing factor SRSF6 protein show metastable conformations with potentially druggable sites revealed by Markov state model

The splicing factor SRSF6 has emerged as a promising target for cancer treatment. In order to gain a comprehensive understanding of the structure and dynamics of SRSF6 and identify the potential binding sites of inhibitors, a combined approach of microseconds-time scale molecular dynamics simulations and Markov state models, followed by molecular docking, was employed here. We predicted that SRSF6 protein exhibits a variety of characteristic metastable conformations that are stabilized by specific hydrogen bonds and/or salt bridges formed between the RRM domains and the loop linker. The transitions between the metastable states are in the time scales ranging from nanoseconds to microseconds. For the most populated states, we identified their potential binding pockets. Then we performed molecular docking of the FDA-approved drugs library, and 15 drugs including reproterol were selected as potential hits inhibiting SRSF6. This work provides valuable insights into the structure and dynamics of SRSF6, which will aid in the structure-based drug design targeting SRSF6 to inhibit tumorigenesis.

Alleviation of carbendazim toxicity effect by Moringa oleifera oil and Linum usitatissimum L. oil on testes of male rats: Physiological, histological and in silico study

Carbendazim (CBZ) is a widely used fungicide that is used to control the unwanted growth of fungi on fruits and vegetables. Sixty male rats were divided into six groups, each having ten. Group one served as control, animals belonging to group two were exposed to CBZ in the measure of 200 mg/kg body weight (BW). In the third and fourth groups, rats were administered 800 mg/kg BW of Moringa oleifera (moringa oil) and Linum usitatissimum L. (flaxseed oil), plus CBZ with the same dose given to group two. Groups five and six were administered with moringa and flaxseed oils respectively for six weeks. A marked decline was seen in oxidative stress markers, reduced glutathione (GSH), superoxide dismutase (SOD), catalase (CAT), and a rise in malondialdehyde (MDA) level in group two with severe histological disruptions. Moringa oil and flaxseed oil were used to alleviate these changes. In addition, a biocomputational molecular docking analysis of three proteins found in male rats was performed. In relation to CBZ (CID:10584007) the screened proteins namely testis-expressed protein (TX101_RAT), EPPI_RAT, and glutathione peroxidase 5 (GPX5_RAT) were docked, and their docking score were obtained (−5.9 kcal/mol), (−5.8 kcal/mol) and (−5.6 kcal/mol) respectively. By examining these interactions in 2D and 3D structures, a detailed understanding of the unique and specific binding affinity, hydrogen bonds, hydrophobic interactions, ionic bonds, and water bonds were obtained. Structure-based virtual screening (SBVS) molecular docking analysis showed that protein interaction with CBZ causes reproductive complications in protein expression and functions by hampering their normal function and blocking active sites.

Lambda CI Binding to Related Phage Operator Sequences Validates Alignment Algorithm and Highlights the Importance of Overlooked Bonds.

Bacteriophage λ's CI repressor protein controls a genetic switch between the virus's lysogenic and lytic lifecycles, in part, by selectively binding to six different DNA sequences within the phage genome-collectively referred to as operator sites. However, the minimal level of information needed for CI to recognize and specifically bind these six unique-but-related sequences is unclear. In a previous study, we introduced an algorithm that extracts the minimal direct readout information needed for λ-CI to recognize and bind its six binding sites. We further revealed direct readout information shared among three evolutionarily related lambdoid phages: λ-phage, Enterobacteria phage VT2-Sakai, and Stx2 converting phage I, suggesting that the λ-CI protein could bind to the operator sites of these other phages. In this study, we show that λ-CI can indeed bind the other two phages' cognate binding sites as predicted using our algorithm, validating the hypotheses from that paper. We go on to demonstrate the importance of specific hydrogen bond donors and acceptors that are maintained despite changes to the nucleobase itself, and another that has an important role in recognition and binding. This in vitro validation of our algorithm supports its use as a tool to predict alternative binding sites for DNA-binding proteins.

Experimental and Computational Approach to Studying Supramolecular Structures in Propanol and Its Halogen Derivatives.

A series of four alcohols, n-propanol and its halogen (Cl, Br, and I) derivatives, were selected to study the effects of variation in polarity and halogen-driven interactions on the hydrogen bonding pattern and supramolecular structure by means of experimental and theoretical methods. It was demonstrated on both grounds that the average strength of H-bonds remains the same but dissociation enthalpy, the size of molecular nanoassemblies, as well as long-range correlations between dipoles vary with the molecular weight of halogen atom. Further molecular dynamics simulations indicated that it is connected to the variation in the molecular order introduced by specific halogen-based hydrogen bonds and halogen-halogen interactions. Our results also provided important experimental evidence supporting the assumption of the transient chain model on the molecular origin of the structural process in self-assembling alcohols.

Channel Gating in Kalium Channelrhodopsin Slow Mutants

Kalium channelrhodopsin 1 from Hyphochytrium catenoides (HcKCR1) is the first discovered natural light-gated ion channel that shows higher selectivity to K+ than to Na+ and therefore is used to silence neurons with light (optogenetics). Replacement of the conserved cysteine residue in the transmembrane helix 3 (Cys110) with alanine or threonine results in a >1,000-fold decrease in the channel closing rate. The phenotype of the corresponding mutants in channelrhodopsin 2 is attributed to breaking of a specific interhelical hydrogen bond (the “DC gate”). Unlike CrChR2 and other ChRs with long distance “DC gates”, the HcKCR1 structure does not reveal any hydrogen bonding partners to Cys110, indicating that the mutant phenotype is likely caused by disruption of direct interaction between this residue and the chromophore. In HcKCR1_C110A, fast photochemical conversions corresponding to channel gating were followed by dramatically slower absorption changes. Full recovery of the unphotolyzed state in HcKCR1_C110A was extremely slow with two time constants 5.2 and 70 min. Analysis of the light-minus-dark difference spectra during these slow processes revealed accumulation of at least four spectrally distinct blue light-absorbing photocycle intermediates, L, M1 and M2, and a UV light-absorbing form, typical of bacteriorhodopsin-like channelrhodopsins from cryptophytes. Our results contribute to better understanding of the mechanistic links between the chromophore photochemistry and channel conductance, and provide the basis for using HcKCR1_C110A as an optogenetic tool.

Machine Learning Assisted Discovery of Novel p38α Inhibitors from Natural Products.

P38α, emerging as a hot spot for drug discovery, is a member of the mitogen- activated protein kinase (MAPK) family and plays a crucial role in regulating the production of inflammatory mediators. However, despite a massive number of highly potent molecules being reported and several under clinical trials, no p38α inhibitor has been approved yet. There is still demand to discover novel p38α to deal with the safety issue induced by off-target effects. In this study, we performed a machine learning-based virtual screening to identify p38α inhibitors from a natural products library, expecting to find novel drug lead scaffolds. Firstly, the training dataset was processed with similarity screening to fit the chemical space of the natural products library. Then, six classifiers were constructed by combing two sets of molecular features with three different machine learning algorithms. After model evaluation, the three best classifiers were used for virtual screening. Among the 15 compounds selected for experimental validation, picrasidine S was identified as a p38α inhibitor with the IC50 as 34.14 μM. Molecular docking was performed to predict the interaction mode of picrasidine S and p38α, indicating a specific hydrogen bond with Met109. This work provides a protocol and example for machine learning-assisted discovery of p38α inhibitor from natural products, as well as a novel lead scaffold represented by picrasidine S for further optimization and investigation.