A molecular glue compound targeting the interface of phb1-phb2 heterodimer and applications thereof

Abstract

The application relates to the field of biological medicine, and discloses a molecular glue compound targeting a PHB1-PHB2 heterodimer interface and application thereof. The application first identifies a druggable shallow groove pocket located at the PHB1-PHB2 heterodimer interface, and screens out a molecular glue compound represented by N-(pyridin-4-ylmethyl)-3-(1-methyl-1H-indol-3-yl)propanamide. The compound can stabilize a PHB1-PHB2 heterodimer complex, significantly inhibit LPS-induced proinflammatory cytokine release and lung infiltration of neutrophils, reduce pathological damage of lung tissue, and significantly improve the survival rate of a mouse acute lung injury model, and thus is expected to be used for preventing and / or treating acute lung injury.

Description

Technical Field

[0001] This invention relates to the field of biomedicine, and more particularly to a molecular adhesive compound that targets the interface of PHB1-PHB2 heterodimers and its applications. Background Technology

[0002] Acute lung injury (ALI) and its severe form, acute respiratory distress syndrome (ARDS), are common critical illnesses in clinical intensive care medicine. They are often induced by various factors such as sepsis, severe trauma, severe pneumonia, aspiration, or major surgery. Their core pathological features include diffuse alveolar epithelial and vascular endothelial damage, increased alveolar-capillary barrier permeability, interstitial and alveolar edema, and uncontrolled inflammatory responses caused by massive neutrophil infiltration. Currently, clinical treatment for ALI / ARDS primarily focuses on protective mechanical ventilation, fluid management, and etiological control, lacking specific therapeutic drugs that can target the core molecular pathological aspects and significantly improve prognosis.

[0003] In the pathogenesis of ALI / ARDS, an uncontrolled "cytokine storm" is a key driver of widespread inflammatory damage to lung tissue. Lipopolysaccharide (LPS), pathogen-associated molecular patterns (PAMPs), or damage-associated molecular patterns (DAMPs) activate alveolar macrophages, monocytes, and epithelial cells, triggering a cascade release of pro-inflammatory cytokines such as IL-6, IL-8, IL-1β, and TNF-α, as well as neutrophil chemokines such as CXCL1 and CXCL2. This, in turn, recruits a large number of neutrophils to migrate into the lungs, aggregate, and release proteases and reactive oxygen species, ultimately causing diffuse damage to the alveolar epithelium and vascular endothelium. Therefore, precise intervention targeting key upstream molecular targets of the inflammatory mediator cascade release and neutrophil lung infiltration is an important direction for developing next-generation anti-ALI / ARDS drugs.

[0004] Mitochondrial functional status is considered a key factor in regulating lung tissue inflammatory responses and cell survival. The prohibitin (PHB) family of mitochondrial inner membrane scaffold proteins (including the PHB1 and PHB2 subunits) assembles in a 1:1 ratio on the inner mitochondrial membrane to form a cyclic supramolecular complex with a central channel. This complex functions as the mitochondrial inner membrane scaffold, maintains cristae structural stability, regulates the orderly assembly of the respiratory chain supercomplex, and supports the activity of homeostasis-related proteases (such as m-AAA proteases). Evidence suggests that in severe inflammatory pathological states such as sepsis, ALI / ARDS, and cytokine storms, the stability of the PHB1-PHB2 complex is significantly reduced, leading to mitochondrial dysfunction, increased reactive oxygen species release, and further amplification of inflammatory signals and exacerbation of lung tissue damage via the DAMPs pathway. However, currently, no small molecule drugs can specifically stabilize the PHB1-PHB2 heterodimer complex.

[0005] Molecular glue is a novel form of small molecule drug that has emerged in recent years. Its mechanism of action differs from traditional enzyme inhibitors or receptor antagonists. Instead, it binds to shallow groove pockets at the interface between two protein subunits, "glueing" and stabilizing loosely or transiently formed protein-protein interactions, thereby achieving precise regulation of their stable state without disrupting normal protein function. Based on the high-resolution cryo-electron microscopy structure of the PHB complex (e.g., PDBID: 9O6S), the development of "molecular glue" type small molecule stabilizers targeting the PHB1-PHB2 heterodimer interface holds promise as a novel therapeutic strategy to alleviate mitochondrial dysfunction and uncontrolled inflammation in ALI / ARDS. However, the structural basis of these shallow groove pockets, their druggability, and the small molecule compounds capable of effectively "glueing" this interface remain unclear. The synthetic routes, binding mechanisms, in vitro anti-inflammatory activity, and in vivo efficacy of related molecules are still under investigation. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a molecular glue compound targeting the PHB1-PHB2 heterodimer interface and its applications. This invention is the first to identify a druggable shallow groove pocket located at the PHB1-PHB2 heterodimer interface and screens for a molecular glue compound represented by N-(pyridin-4-ylmethyl)-3-(1-methyl-1H-indol-3-yl)propionamide. This compound can stabilize the PHB1-PHB2 heterodimer complex, significantly inhibit LPS-induced release of pro-inflammatory cytokines and neutrophil lung infiltration, alleviate pathological damage to lung tissue, and significantly improve the survival rate of a mouse model of acute lung injury. Therefore, it holds promise for the prevention and / or treatment of acute lung injury.

[0007] The specific technical solution of the present invention includes: First, the present invention provides a molecular glue compound that targets the PHB1-PHB2 heterodimer interface, which is a compound having the following molecular structure or a pharmaceutically acceptable salt, solvate, stereoisomer or prodrug thereof.

[0008] .

[0009] The compound shown in the above molecular structural formula is named N-(pyridin-4-ylmethyl)-3-(1-methyl-1H-indol-3-yl)propionamide, with the molecular formula C. 18 H 19N3O, with an exact molecular weight of 293.366 Da, has an 0 charge, contains one segment and five rotatable bonds, has a hydrogen bond donor (HBD) number of 1, a hydrogen bond acceptor (HBA) number of 4, a LogP of 2.8, and a topological polar surface area (TPSA) of 46.92 Å. 2 It conforms to the Lipinski five rules for class drugs. The expression for SMILES is O=C(CCc1cn(C)c2ccccc12)NCc1ccncc1; this compound has the function of binding to the shallow groove pocket at the interface of PHB1-PHB2 heterodimer and stabilizing the assembly of the PHB1-PHB2 complex.

[0010] The aforementioned molecular glue compound can bind to the shallow groove pocket at the PHB1-PHB2 heterodimer interface. The pocket is formed by the T141, I140, Q142, R143, E144, S147, R148, S151, and D167 residues on the PHB1 side and the V145, A146, V144, K142, S143, V140, N138, I136, E135, E167, and R171 residues on the PHB2 side. It simultaneously forms hydrogen bonds, salt bridges, and π-cation interactions with at least one of the residues R143, E144, L145, S147, R148, and S151 on the PHB1 side and K142, S143, V140, V144, V168, and K147 on the PHB2 side, thereby "gluing" and stabilizing the heterodimer interface.

[0011] This invention is the first to identify an interface shallow groove pocket with high accessibility and drug development potential based on the cryo-electron microscopy structure of PHB1-PHB2 heterodimer. This pocket naturally forms a "dual anchoring" binding mode and exhibits a spatial distribution of "hydrophobic core – hydrophilic periphery", providing a clear structural basis for the development of stable molecular glue compounds of PHB1-PHB2 heterodimer. Based on this target structure, this invention further employs a de novo molecule generation method, combined with molecular docking scoring and ADMET-like drug efficacy evaluation, to obtain 10 lead candidate compounds (M1-M10). Through dose-response determination at the live cell level using the NanoBiT protein complementary fluorescence system, compound M4 (i.e., N-(pyridin-4-ylmethyl)-3-(1-methyl-1H-indole-3-yl)propionamide) was identified as the lead compound with the best activity (Log IC50=-10.87, Emax=179.2%). Furthermore, through Native PAGE and comparison with control molecules such as indole-3-propionic acid (IPA) and indomethacin, it was further confirmed that the stabilizing effect of M4 on the PHB1-PHB2 complex is structure-specific.

[0012] Second, the present invention provides a method for preparing a molecular adhesive compound having the above-mentioned molecular structural formula, which includes the following steps: Step (a): Starting with 1H-indole-3-propionic acid, the reaction is carried out in the presence of a base, a methylating agent, and a solvent to selectively complete the N1-methylation of indole, yielding the 1-methyl-1H-indole-3-propionic acid intermediate.

[0013] Step (b): The 1-methyl-1H-indole-3-propionic acid intermediate is subjected to an amide condensation reaction with 4-pyridinemethylamine in a solvent under the action of a condensing agent and an organic base. After the reaction is completed, the product is purified to obtain the target product.

[0014] The preparation method of this invention has a short synthetic route, readily available raw materials, mild reaction conditions, and good scale-up feasibility.

[0015] Preferably, in step (a), the base is NaH; the methylating agent is CH3I; and the solvent is tetrahydrofuran (THF).

[0016] Preferably, in step (a), the molar ratio of 1H-indole-3-propionic acid, NaH, and CH3I is 1:(4-6):(2-4).

[0017] Preferably, in step (a), the temperature of the reaction is from 0°C to room temperature.

[0018] Preferably, in step (b), the condensing agent is HATU (2-(7-benzotriazole oxide)-N,N,N′,N′-tetramethylurea hexafluorophosphate); the organic base is DIPEA (N,N-diisopropylethylamine); and the solvent is dichloromethane (DCM).

[0019] Preferably, in step (b), the reaction temperature is room temperature.

[0020] Preferably, in step (b), the molar ratio of 1-methyl-1H-indole-3-propionic acid, condensing agent, organic base, and 4-pyridinemethylamine is 1:(1.1-1.3):(1.8-2.2):(1.1-1.3). Preferably, in step (b), the purification is column chromatography purification, and the elution system is dichloromethane:methanol = (40-60):1 (volume ratio).

[0021] Third, the present invention provides the application of the above-mentioned molecular glue compound in the preparation of drugs that stabilize PHB1-PHB2 heterodimer complexes.

[0022] Fourth, the present invention provides the use of the above-mentioned molecular glue compound in the preparation of medicaments for the prevention and / or treatment of acute lung injury, acute respiratory distress syndrome or cytokine storm-related inflammatory lung diseases.

[0023] This invention has discovered that the above-mentioned molecular adhesive compounds exert their preventive and / or therapeutic effects on acute lung injury, acute respiratory distress syndrome, or cytokine storm-related inflammatory lung diseases through the following steps: 1) The compound binds to the shallow groove pocket at the PHB1-PHB2 heterodimer interface at a picomolar-nanomolar concentration, forming multiple specific interactions with key residues on both the PHB1 and PHB2 sides, including hydrogen bonds, salt bridges, and π-cations. 2) The assembly of the PHB1-PHB2 complex is "glued" and stabilized through the "molecular bridge" effect; 3) In an LPS-induced in vitro cellular inflammation model, it significantly inhibited the expression of IL-6 and IL-8 at the mRNA and secreted protein levels; 4) In an LPS-induced mouse model of acute lung injury, it significantly inhibited neutrophil infiltration in BALF and the release of pro-inflammatory cytokines and chemokines such as IL-6, IL-1β, TNF-α, CXCL1, and CXCL2; 5) Reduces pathological changes in lung tissue such as alveolar wall thickening, interstitial edema, and inflammatory cell infiltration; 6) Significantly improves the overall survival rate of ALI model mice.

[0024] Therefore, the drug has at least one of the following functions: a) Stabilize the assembly of the PHB1-PHB2 heterodimer complex; b) Inhibit the expression and secretion of lipopolysaccharide (LPS)-induced IL-6 and / or IL-8 pro-inflammatory cytokines at both the mRNA and protein levels; c) Inhibit the release of at least one pro-inflammatory cytokine and / or chemokine from LPS-induced IL-6, IL-1β, TNF-α, CXCL1 and CXCL2 in lung tissue; d) Inhibit neutrophil infiltration into lung tissue; e) Reduce the pathological changes in lung tissue caused by acute lung injury; f) Improve the survival rate of animals with acute lung injury.

[0025] Fifth, the present invention provides a medicament for the prevention and / or treatment of acute lung injury, acute respiratory distress syndrome or cytokine storm-related inflammatory lung disease, which uses the above-mentioned molecular glue compound as the active ingredient and contains a pharmaceutically acceptable carrier, excipient or solvent.

[0026] Preferably, the dosage form of the drug is an oral preparation or an injectable preparation.

[0027] More preferably, the oral preparation is a capsule, tablet, solution, or powder; the injectable preparation is a vesicle, injection, or injection powder.

[0028] Sixth, the present invention provides a method for screening molecular glue compounds that can prevent and / or treat acute lung injury, acute respiratory distress syndrome, or cytokine storm-related inflammatory lung diseases: using the shallow groove pocket at the PHB1-PHB2 heterodimer interface as the target site, the druggable shallow groove pocket located at the PHB1-PHB2 heterodimer interface is identified based on the high-resolution cryo-electron microscopy structure of the PHB complex (PDB ID: 9O6S), and the optimal molecular glue compound is identified by de novo molecule generation and virtual screening combined with NanoBiT cell activity assay.

[0029] The shallow groove pocket at the PHB1-PHB2 heterodimer interface is composed of T141, I140, Q142, R143, E144, S147, R148, S151, and D167 residues on the PHB1 side and V145, A146, V144, K142, S143, V140, N138, I136, E135, E167, and R171 residues on the PHB2 side. By screening or designing compounds that can bind to the pocket and stabilize the PHB1-PHB2 heterodimer complex, candidate drugs for the prevention and / or treatment of acute lung injury can be obtained.

[0030] Based on the novel discoveries of this invention regarding the structural basis and druggability of the shallow groove pocket at the interface of the PHB1-PHB2 heterodimer, the shallow groove pocket can be used as one of the targets in the process of screening or preparing drugs for the prevention and / or treatment of acute lung injury, acute respiratory distress syndrome, or cytokine storm-related inflammatory lung diseases. Compounds that can bind to the pocket and stabilize the PHB1-PHB2 heterodimer complex can be screened or designed using methods such as molecular docking scoring, de novo molecule generation, and live-cell NanoBiT protein complementation assays, thereby obtaining candidate drugs with good preventive and / or therapeutic effects against acute lung injury.

[0031] Compared with the prior art, the beneficial effects of the present invention are: This invention, for the first time, identifies a novel druggable target—the shallow groove pocket at the PHB1-PHB2 heterodimer interface—based on the high-resolution cryo-electron microscopy structure of the PHB complex. Through de novo molecular generation, virtual screening, live-cell NanoBiT dose-response assays, and native PAGE specificity verification, the first molecular glue compound M4 (Log IC50 = -10.87, Emax = 179.2%) capable of specifically "gluing" this interface was obtained. This compound exhibits a concise synthetic route, readily available raw materials, mild reaction conditions, and excellent druggability. It demonstrates significant anti-inflammatory and protective effects in both LPS-induced in vitro cellular inflammation models and mouse acute lung injury models. It significantly inhibits the release of pro-inflammatory cytokine cascades, suppresses neutrophil lung infiltration, reduces pathological damage to lung tissue, and significantly improves the survival rate of ALI model mice. This provides a new target and strategy for the treatment of acute lung injury, acute respiratory distress syndrome, and cytokine storm-related inflammatory lung diseases. Attached Figure Description

[0032] Figure 1 Cryo-electron microscopy structure of the PHB complex and structural analysis of the drug-binding pocket at the interface of the PHB1-PHB2 heterodimer; based on PDB ID: 9O6S cryo-electron microscopy structure and molecular surface druggability analysis. A: Top view of the overall PHB complex structure, with the PHB1 subunit shown as a green ribbon and the PHB2 subunit as a cyan ribbon, arranged alternately in a 1:1 ratio to form a cyclic supramolecular assembly with a central channel. Candidate molecules docked within the interface pocket are indicated by purple rod-shaped models. B: Side view of the PHB complex, showing PHB1-PHB2 arranged in a columnar pattern and forming a transverse binding cavity across the two subunits at the interface. C: Magnified view of the surface model of the PHB1 (green)-PHB2 (cyan) heterodimer and a partial view of the shallow groove pocket at the interface, showing that the candidate molecules (purple rod-shaped) are highly complementary in shape to the interface pocket. D: A three-dimensional view of the interaction between key residues at the PHB1-PHB2 interface and candidate molecules. The PHB1 side interface residues (green sticks, including T141, I140, Q142, R143, E144, S147, R148, S151, D167, etc.) and the PHB2 side interface residues (cyan sticks, including V145, A146, V144, K142, S143, V140, N138, I136, E135, E167, R171, etc.) together constitute the molecular glue binding site. Figure 2A: NanoBiT dose-response evaluation of lead compounds M1–M10 on the interaction of PHB1–PHB2 and specificity comparison of M4 with the reference molecule. A: Dose-response curves of M1–M10 measured using the NanoBiT protein complementary fluorescence system. The vertical axis represents the NanoBiT signal (% basal value), and the horizontal axis represents the logarithmic concentration of the compound (Log[Molecule](M), approximately 10). -13 M-10 -6 M): Different colored curves correspond to compounds M1-M10 respectively; data are expressed as mean ± SEM (n≥3 independent experiments). B: Summary table of activity parameters for M1-M10, listing the molecular formula, Log IC50, and Emax of each compound; M4 is the molecule with the best activity (Log IC50 = -10.87, Emax = 179.2%). C: Native PAGE analysis of the effects of M4, indole-3-propionic acid (IPA), and indomethacin on the stability of the PHB1 / 2 complex band under LPS stimulation, with molecular weight gradient references of 240 / 440 / 670 kDa. D: Comparison of dose-response curves of M4, indole-3-propionic acid, indomethacin, and DMSO control on the same NanoBiT platform; Figure 3 : Molecular descriptor of compound M4. Structural information and key molecular descriptors of M4: SMILES, total number of atoms, exact molecular weight, charge, molecular formula, number of fragments, and two-dimensional structural formula; Figure 4 Drugability analysis of compound M4. The M4 drugability radar chart shows its distribution across multiple drugability indicators, including TPSA, HBD, LogP, MW, hepatotoxicity, hERG risk, CYP inhibition risk, BBB penetration, Caco-2 permeability, water solubility, and number of rotatable bonds. Figure 5 3D molecular docking structural analysis of the PHB1-PHB2 interface binding pocket. The results show the 3D molecular docking of M4 with the PHB1 (pink)-PHB2 (magenta) heterodimer interface binding pocket. M4 (green rod-shaped) is stably embedded in the shallow groove of the interface, forming hydrogen bonds (yellow dashed lines), salt bridges (purple dashed lines), and π-cation interactions (green dashed lines) with key residues on both the PHB1 and PHB2 sides. Figure 6 2D structural analysis of molecular docking in the PHB1-PHB2 interface binding pocket. Two-dimensional interaction fingerprint of M4 with residues in the PHB1-PHB2 interface pocket; Figure 7Chemical synthesis route and NMR structure confirmation of compound M4. A: Schematic diagram of the two-step chemical synthesis reaction route of M4. Reaction conditions: (a) CH3I, NaH, THF, 0℃ to room temperature; (b) HATU, DIPEA, DCM, room temperature. B: Synthetic product M4 1 H-NMR spectrum (solvent: DMSO-d6, residual solvent peak δ ≈ 2.50 ppm); Figure 8 The inhibitory effect of compound M4 on LPS-induced IL-6 and IL-8 expression at both mRNA and protein levels in vitro. AB: qRT-PCR was used to detect the relative expression levels of IL-6 (A) and IL-8 (B) mRNA induced by LPS (1 μg / mL) stimulation under DMSO or M4 pretreatment conditions, normalized to a baseline of LPS- / DMSO. CD: ELISA was used to detect the protein secretion concentrations (pg / mL) of IL-6 (C) and IL-8 (D) in the culture supernatant after LPS stimulation under DMSO or M4 pretreatment conditions. Data are expressed as mean ± SEM, n ≥ 6; statistical methods included two-way ANOVA with Tukey multiple comparisons, P < 0.0001. Figure 9 : In vivo efficacy evaluation of compound M4 in an LPS-induced mouse model of acute lung injury. A: Cell classification and count of bronchoalveolar lavage fluid (BALF) in each group of mice, including total cells (Total), macrophages (Mac), neutrophils (Neu), and lymphocytes (Lym). The vertical axis represents the number of BALF cells (× 10). 4 / mL); grouped as NS+DMSO, LPS+DMSO, NS+Molecular (M4), LPS+Molecular (M4). BF: ELISA results (pg / mL) of IL-6 (B), IL-1β (C), TNF-α (D), CXCL1 (E), and CXCL2 (F) in BALF / lung tissue homogenates; Figure 10 In vivo immunohistochemical and survival analysis of compound M4 in an LPS-induced mouse model of acute lung injury. A: Representative HE staining images of lung tissue from each group of mice (scale bar: 100 μm). B: Lung tissue pathological score (HE Score) based on HE-stained sections. C: 8-day survival curves of mice after LPS modeling; blue represents the DMSO group, and red represents the M4 treatment group; statistical method: Log-rank (Mantel-Cox) test, P < 0.001. Data in AC are expressed as mean ± SEM, n ≥ 6; statistical method: two-way ANOVA with Tukey multiple comparisons, P < 0.0001. Detailed Implementation

[0033] Example 1 I. Cryo-electron microscopy structural analysis of the PHB complex and identification of the druggable pocket at the interface of the PHB1-PHB2 heterodimer ( Figure 1 (AD) Based on the high-resolution cryo-electron microscopy structure of the PHB complex (PDB ID: 9O6S) from the RCSB PDB database, the 3D structure of the PHB1-PHB2 heterodimer was modeled and visualized using UCSF ChimeraX 1.6 and PyMOL 2.5. Druggability scanning of the PHB1-PHB2 heterodimer surface was performed using FPocket 4.0 and the SiteMap algorithm to identify shallow groove pockets located at the interface between the two subunits. The electrostatic surface potential of the pockets was analyzed using the "coulombic surface coloring" module of ChimeraX. The evolutionary conservation score (1-9) of the pocket residues was calculated using the ConSurf algorithm based on multiple sequence alignment (MUSCLE algorithm) of 100 PHB homologous sequences from vertebrate species and the maximum likelihood phylogenetic method.

[0034] II. Small Molecule Design and Virtual Screening Based on VD-Gen Based on the three-dimensional structural features of the shallow groove pocket at the PHB1-PHB2 heterodimer interface, a VD-Gen (Virtual Drug Generation) deep generation model was employed. Using the geometry, electrostatic and hydrophobic distribution of this binding pocket as input conditions, candidate small molecules with potential binding capacity were generated. During the generation process, the molecular weight range was set to 200-500 g / mol, the QED score ≥ 0.5, and compliance with Lipinski's five rules for drug-likeness was used as preliminary constraints. The generated candidate compounds were subsequently subjected to AutoDock Vina molecular docking scoring (binding free energy < -7.0 kcal / mol), and ADMET property prediction and drug-likeness screening were performed using SwissADME / pkCSM / ADMETlab2.0. Finally, the Top 10 lead candidate compounds M1-M10 were obtained (their structural formulas are shown below).

[0035]

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042]

[0043]

[0044]

[0045] III. Construction of NanoBiT protein complementary fluorescence system and determination of dose-response curves Expression plasmids fused with the SmBiT structure for PHB1 and the LgBiT structure for PHB2 were constructed into the pcDNA3.1 vector (Promega NanoBiT system). After verification by Sanger sequencing, HEK293T cells were transiently co-transfected using Lipofectamine 3000 (Invitrogen). Twenty-four hours after transfection, cells were seeded at a density of 5,000 cells / well in 384-well white transparent plates, and compounds M1-M10 were added to achieve final concentrations of 10-1. -13 10 -12 10 -11 10 -10 10 -9 10 -8 10 -7 10 -6 M, with three replicates for each concentration, and DMSO as a negative control. After 4 h of compound treatment, Nano-Glo LiveCell substrate (Promega) was added, and the NanoBiT luminescence signal was detected using a Pherastar FSX microplate reader (BMG Labtech). The percentage of NanoBiT signal relative to baseline signal (%basal value) was plotted on the ordinate, and the logarithmic concentration of the compound (Log[Molecule](M)) was plotted on the abscissa. Nonlinear regression fitting was performed using GraphPad Prism 9 to obtain LogIC50 and Emax values.

[0046] Figure 2 The AB results showed that M1-M10 all exhibited typical concentration-dependent S-type activation curves, among which M4 (LogIC50=-10.87, Emax=179.2%) showed significantly better activity than the other 9 candidate molecules; except for M4, the LogIC50 of the other compounds ranged from -9.78 (M2) to -6.88 (M5), and the Emax ranged from 164.8% (M6) to 182.0% (M9).

[0047] IV. Native PAGE analysis and specificity comparison with control compounds HEK293T cells stably expressing the PHB1-PHB2 complex were stimulated with DMSO (negative control) and LPS (1 μg / mL), and then pretreated for 1 h with M4 (100 nM), indole-3-propionic acid (IPA, 100 nM, with a simple analogue of the M4 backbone as a backbone control), or indomethacin (100 nM, a classic non-targeted indole anti-inflammatory drug). After treatment, the cells were lysed on ice for 15 min with lysis buffer (containing 1% digitalis saponin in 50 mM NaCl, 50 mM imidazole, 2 mM 6-aminocaproic acid, and 1 mM EDTA, pH 7.0), followed by centrifugation at 20,000 g for 30 min. After adding Coomassie Brilliant Blue G-250 to the supernatant, it was loaded onto a 3-12% gradient Native PAGE gel and electrophoresed at 4°C (100 V / 30 min, followed by 200 V until completion). The gel was then transferred to a PVDF membrane and analyzed by Western blot with PHB1 / PHB2 antibodies.

[0048] Figure 2 The CD results showed that the intensity of the PHB1 / 2 complex band was significantly decreased in the DMSO control group under LPS stimulation; M4 treatment significantly restored the intensity of the PHB1 / 2 complex band to near the unstimulated (LPS-) level, while IPA and indomethacin had no significant restorative effect on this band, confirming that the stabilizing effect of M4 is a structure-specific "molecular glue" mechanism, rather than non-specific binding to the indole skeleton (see details). Figure 2 (C). Dose-response curves of M4, IPA, indomethacin, and DMSO were compared on the same NanoBiT platform. The results showed that M4... -10 NanoBiT signaling can be significantly activated at around M (Emax≈180%), while IPA and indomethacin only activate at higher concentrations (≥10). -7 A slight increase in signal (Emax≈115-120%) was observed under M) and DMSO showed no significant activity.

[0049] V. Physicochemical Properties and ADMET Prediction Analysis of the Top 10 Lead Compounds Ten lead compounds, M1-M10, were systematically evaluated for their physicochemical properties (molecular weight, HBD, HBA, LogP, TPSA, number of rotatable bonds, compliance with the Lipinski five rules) and ADMET (water solubility, Caco-2 permeability, BBB permeability, CYP inhibition risk, hERG risk, predicted oral LD50 (rat), hepatotoxicity, estimated half-life, and GHS toxicity classification). The results showed ( Figure 3 , Figure 4 (See Tables 1, 2, and 3). The molecular weight of M4 is 293.366 g / mol, HBD is 1, HBA is 4, LogP is 2.8, and TPSA is 46.92 Å. 2 With 5 rotatable bonds, it conforms to the Lipinski five-rule drug system. ADMET evaluation results show that M4 has moderate water solubility, high Caco-2 permeability, BBB penetration capability, low CYP inhibition risk, low hERG risk, predicted oral LD50 of approximately 1500 mg / kg, low hepatotoxicity, estimated half-life of approximately 2-6 h, and GHS classification of Class IV, exhibiting balanced drug-like characteristics overall.

[0050] Table 1: Basic Structural Information: Summary of M1-M10 numbers, molecular names, molecular formulas, and molecular weights

[0051] Table 2: Physicochemical descriptors: Summary of compliance status of HBD, HBA, LogP, TPSA, number of rotatable bonds, and the five rules for Lipinski drugs for M1-M10.

[0052] Note: (1) Lipinski's five rules: MW≤500, LogP≤5, HBD≤5, HBA≤10; M7 (HBA=9, TPSA=137.21) is slightly not in compliance.

[0053] Table 3: Predictive Properties of ADMET: Key drug-likeness indicators are given, including water solubility, Caco-2 permeability, BBB penetration, CYP inhibition risk, hERG risk, estimated oral LD50 (rat), hepatotoxicity, estimated half-life, and GHS toxicity classification.

[0054]

[0055] Note: (1) The above ADMET data are estimated values ​​based on molecular descriptors. It is recommended to use SwissADME / pkCSM / ADMETlab 2.0 for verification; (2) Toxicity classification refers to GHS: Class III (50-300 mg / kg), Class IV (300-2000 mg / kg), Class V (>2000 mg / kg); (3) BBB penetration criterion: TPSA < 80 Å 2 Furthermore, LogP tends to be penetrable in the range of 1-3.

[0056] VI. Molecular docking analysis of compound M4 with the PHB1-PHB2 interfacial pocket Rigid and flexible molecular docking was performed on the shallow trench pocket (based on PDB ID: 9O6S) at the interface of compound M4 and the PHB1-PHB2 heterodimer using the Schrödinger Glide SP / XP module and AutoDock Vina. Two-dimensional interaction fingerprints were then plotted using the Schrödinger Maestro 2D Interaction Diagram module. The results are as follows: Figure 5 and Figure 6 As shown: The optimal conformation of docking score (XP score) shows that M4 is stably embedded in the shallow groove of the interface, with its indole ring and pyridine ring anchored to the hydrophobic subpockets on the PHB1 and PHB2 sides, respectively. The propionamide linker arm extends along the interface direction, allowing M4 to form a "molecular bridge" across the two subunits. Docking analysis shows that M4 forms hydrogen bonds, salt bridges and π-cation interactions with residues such as R143, E144, L145, S147, R148, and S151 on the PHB1 side and K142, S143, V140, V144, V168, and K147 on the PHB2 side.

[0057] VII. Chemical Synthesis and Structural Characterization of Compound M4 Synthetic routes such as Figure 7 As shown in A, the specific steps are as follows: Step (a): Synthesis of 1-methyl-1H-indole-3-propionic acid: 1H-indole-3-propionic acid (1.5 g, 7.93 mmol, 1.0 equivalent) and anhydrous tetrahydrofuran (THF, 150 mL) were added to a 500 mL three-necked flask and cooled to 0 °C in an ice bath. NaH (956 mg, 39.64 mmol, 5.0 equivalent) was added in portions, and the mixture was stirred at 0 °C for 30 min. Iodomethane (CH3I, 3.38 g, 23.78 mmol, 3.0 equivalent) was then slowly added dropwise. The reaction was maintained at this temperature for 1 h, and the ice bath was removed. The reaction was continued at room temperature for 24 h. After the reaction was completed by TLC (ethyl acetate: n-hexane = 1:2 (v / v)), the pH of the reaction solution was quenched to 2.0 with 6 N HCl at 0 °C. The mixture was extracted three times with ethyl acetate, and the organic phases were combined, washed successively with saturated brine, dried over anhydrous Na2SO4, and concentrated under reduced pressure. The residue was recrystallized from ethyl acetate / n-hexane to give pure 1-methyl-1H-indole-3-propionic acid (pale yellow solid, 1.30 g, 81% yield). LC-MS (ESI): m / z = 204.1 [M+H] + .

[0058] Step (b): Synthesis of N-(pyridin-4-ylmethyl)-3-(1-methyl-1H-indole-3-yl)propionamide (M4): 1-methyl-1H-indole-3-propionic acid (250 mg, 1.23 mmol, 1.0 equivalent), HATU (562 mg, 1.48 mmol, 1.2 equivalent), and anhydrous dichloromethane (DCM, 30 mL) were added to a 100 mL round-bottom flask and stirred until completely dissolved. DIPEA (318 mg, 2.46 mmol, 2.0 equivalent) was added, and the mixture was activated at room temperature for 15 min. Then, 4-pyridinylmethylamine (160 mg, 1.48 mmol, 1.2 equivalent) was added, and the reaction was continued at room temperature for 2 h. After the reaction was monitored to be complete by TLC (dichloromethane:methanol = 50:1), the mixture was concentrated under reduced pressure. The residue was purified by silica gel column chromatography (dichloromethane:methanol = 50:1, v / v) to give the target compound M4 (pale yellow solid, 305 mg, yield 84%). LC-MS (ESI): m / z = 294.1 [M+H] + .

[0059] 1 Characterization results of H-NMR (DMSO-d6, residual solvent peak δ≈2.50 ppm) Figure 7(B): Characteristic multiplets of the indole ring and 4-pyridine ring aromatic hydrogens appear in the δ 7.57-6.99 ppm range (integrated for approximately 8 Hz); a characteristic peak of the methylene group (-NH-CH2-Py) linked by 4-pyridinemethylamine appears near δ 4.28 ppm (integrated for 2.03 Hz); a characteristic singlet of the indole N-methyl (N-CH3) appears near δ 3.71 ppm (integrated for 3.01 Hz); multiplets of the methylene groups (-CH2-CH2-C(=O)-NH-) on both sides of the propionamide linker arm appear in the δ 2.99-2.53 ppm range (integrated for approximately 4 Hz, divided into two segments of approximately 2.02 Hz and 2.06 Hz). The chemical shifts, peak shapes, and integration ratios of each signal are completely consistent with the target structure, confirming that the synthesized compound is N-(pyridin-4-ylmethyl)-3-(1-methyl-1H-indole-3-yl)propionamide.

[0060] VIII. In vitro cell inflammation model and IL-6 / IL-8 detection ( Figure 8 (AD) Human alveolar epithelial / monocyte-macrophage cells were seeded in 12-well plates (5 × 10⁻⁶ cells / wells). 5 (Number of samples per well) were divided into four groups: DMSO (-LPS), DMSO (+LPS), M4 (-LPS), and M4 (+LPS), with n ≥ 6 in each group. The pretreatment concentration of M4 was 100 nM, and the final concentration of DMSO was 0.1%. After 1 h of pretreatment, LPS (1 μg / mL) was added and cultured for another 6 h (for qRT-PCR) or 24 h (for ELISA).

[0061] qRT-PCR detection: Total RNA was extracted from cells using TRIzol reagent (Invitrogen), reverse transcribed into cDNA using the PrimeScript RT kit (TaKaRa), and qRT-PCR was performed using SYBR Premix Ex Taq II (TaKaRa). GAPDH was used as an internal control, and the relative expression levels of IL-6 and IL-8 mRNA were calculated using the 2^-ΔΔCt method. Results showed that under LPS stimulation, the relative expression levels of IL-6 and IL-8 mRNA in the DMSO group increased from approximately 1.6 and 1.3 to approximately 4.8 and 5.8-fold, respectively (P<0.0001); while in the M4 pretreatment group, under LPS stimulation, the IL-6 mRNA level decreased to approximately 2.3 and the IL-8 mRNA level decreased to approximately 3.0 (P<0.0001), a decrease of approximately 50%.

[0062] ELISA assay: Culture supernatants from each group were collected and analyzed according to the instructions of the Human IL-6 / IL-8 ELISA Kit (R&D Systems). Results showed that under LPS stimulation, the IL-6 concentration in the DMSO group increased from approximately 1,000 pg / mL at baseline to approximately 8,000 pg / mL, and the IL-8 concentration increased from approximately 150 pg / mL at baseline to approximately 1,800-2,000 pg / mL (both P < 0.0001). In the M4 pretreatment group, under LPS stimulation, the IL-6 concentration decreased to approximately 4,000 pg / mL, and the IL-8 concentration decreased to approximately 900 pg / mL, representing a decrease of approximately 50% compared to the LPS+DMSO group (P < 0.0001). Under LPS-negative conditions, M4 treatment alone had no significant effect on baseline IL-6 / IL-8 levels.

[0063] IX. LPS-induced mouse acute lung injury (ALI) model and in vivo efficacy evaluation ( Figure 9 AF and Figure 10 AC) All animal experimental procedures were performed in accordance with relevant regulations on laboratory animal ethics review. Male C57BL / 6J mice aged 6-8 weeks were randomly divided into four groups according to body weight: NS+DMSO, LPS+DMSO, NS+Molecular (M4), and LPS+Molecular (M4), with n ≥ 6 in each group. M4 mice were administered intraperitoneally at a dose of 10 mg / kg (solvent: PBS containing 5% DMSO), while the control group received an equal volume of the solvent. One hour after administration, the LPS group was infused intratracheally with LPS (Sigma-Aldrich, O111:B4) at a dose of 5 mg / kg to establish an ALI model, while the control group (NS group) received an equal volume of sterile saline.

[0064] Twenty-four hours after modeling, some mice were anesthetized by intraperitoneal injection of 1% sodium pentobarbital. Bronchoalveolar lavage fluid (BALF) was collected three times by lavaging the trachea with sterile PBS. The BALF was then combined and centrifuged at 500 g for 10 min at 4°C. The pellet was resuspended in PBS for cell classification and counting (Total, Mac, Neu, Lym, with the vertical axis unit being ×10⁻⁶). 4 / mL); BALF supernatant and lung tissue homogenate supernatant were used for ELISA detection of IL-6, IL-1β, TNF-α, CXCL1, and CXCL2 (R&D Systems). Results showed that the total BALF cell count in the LPS+DMSO group was approximately 15 × 10⁻⁶ cells compared to the NS+DMSO group. 4 / mL increased to approximately 80×10 4 / mL, Neu number from baseline approximately 5 × 10⁹ / mL 4 / mL increased to approximately 90×10 4 / mL (all P<0.0001); the total cell count in the M4 treatment group (LPS+Molecular) decreased to approximately 40×10⁹ / mL (all P<0.0001); 4 / mL, Neu number decreased to approximately 22 × 10 4 / mL (P<0.0001), there were no significant differences in Mac and Lym numbers among the groups.

[0065] ELISA results showed that the levels of IL-6 (approximately 210 pg / mL), IL-1β (approximately 690 pg / mL), TNF-α (approximately 185 pg / mL), CXCL1 (approximately 950 pg / mL), and CXCL2 (approximately 500 pg / mL) in the LPS+DMSO group were significantly higher than those in the NS+DMSO group (P<0.0001), exhibiting obvious characteristics of a cytokine storm. In the M4 treatment group, the levels of the above mediators decreased to approximately 70 pg / mL (IL-6), 200 pg / mL (IL-1β), 90 pg / mL (TNF-α), 500 pg / mL (CXCL1), and 220 pg / mL (CXCL2), respectively (P<0.0001), with a reduction of approximately 45-70%.

[0066] Pathological evaluation of lung tissue: The middle and lower lobes of the right lung were harvested 24 h after modeling, fixed in 4% paraformaldehyde for 24 h, and routinely embedded in paraffin (4 μm) for sectioning and hematoxylin-eosin (HE) staining. The tissues were scored in a blinded manner by two independent pathologists (the sum of scores from 0 to 4 for inflammatory cell infiltration, alveolar wall thickening, interstitial edema, and hemorrhage). Results showed that the lung tissue structure was normal in both the NS+DMSO and NS+M4 groups; the HE score in the LPS+DMSO group increased from approximately 0.3 in the NS group to approximately 3.8 (P<0.0001); the HE score in the LPS+M4 treatment group decreased to approximately 1.2 (P<0.0001), approaching the baseline level of the NS group.

[0067] Survival analysis: A separate group of 10 mice was established for each group, and the modeling and drug administration were performed according to the above protocol. Mice were observed for 8 consecutive days, and mortality was recorded. The Log-rank (Mantel-Cox) test was used to analyze survival differences. Results showed that after LPS modeling, mice in the DMSO group died concentratedly on days 4-6, with a survival rate of only about 5-10% on day 8; while the survival rate of mice in the M4 treatment group was significantly improved to about 65-70% (P<0.001).

[0068] 10. Statistical Analysis Unless otherwise stated, data are expressed as mean ± SEM of at least three independent experiments or n ≥ 6. Statistical analysis was performed using GraphPadPrism 9. Unpaired Student's t-test was used for comparisons between two groups; two-way ANOVA with Tukey's post-hoc test was used for comparisons among multiple groups; survival analysis was performed using the Log-rank (Mantel-Cox) test. P < 0.05 was considered statistically significant; P < 0.05, P < 0.01, P < 0.001, and P < 0.0001 are included in the figures.

[0069] Data Analysis (1) Discovery and structural characterization of shallow groove pockets at the interface of PHB1-PHB2 heterodimer Based on the high-resolution cryo-electron microscopy structure of the PHB complex (PDB ID: 9O6S) from the RCSB PDB database, this invention obtained a three-dimensional structural model of the PHB1-PHB2 heterodimer. The PHB complex consists of alternating PHB1 (green) and PHB2 (cyan) in a 1:1 ratio, forming a cyclic supramolecular assembly with a central channel. It exists on the inner mitochondrial membrane in a disk / barrel configuration and serves as the scaffold for the inner mitochondrial membrane and maintains homeostasis. Figure 1 A, top view; Figure 1 (B, side view). Using molecular surface analysis and structural druggability scanning, this invention successfully identified a shallow groove pocket with high accessibility and druggability potential at the PHB1-PHB2 heterodimer interface. Figure 1 The pocket extends along the PHB1-PHB2 interface. The PHB1 side (green) is composed of polar / basic residues such as R143, E144, S147, R148, and Q142, which provide positive charges and hydrogen bonds to the acceptor network, while also contributing polar / hydrophobic residues such as T141, I140, S151, and D167. The PHB2 side (cyan) is composed of polar / basic residues such as K142, A146, R171, and S143, and also contributes hydrophobic / polar residues such as V140, V144, V145, I136, N138, E135, and E167. Figure 1 (D). The pocket as a whole forms a spatial distribution of "hydrophobic core - hydrophilic periphery". Compared with the traditional pocket with a single subunit, the shallow groove of this interface spans two subunits and naturally forms a "double anchoring" binding mode, providing a unique and clear structural basis for the development of PHB1-PHB2 heterodimer stable molecular glue compounds.

[0070] (2) Evaluation of the structure and druggability of the top 10 lead compounds obtained by the VD-Gen method Based on the structural information of the shallow groove pockets at the interface of the PHB1-PHB2 heterodimers, this invention uses the VD-Gen de novo molecular generation method to score and rank these interface pockets, ultimately obtaining the Top 10 lead candidate compounds M1-M10 (their structural formulas are shown above). These 10 compounds cover a wide range of chemical skeletons, including bromoindole dioxopiperazines (M1), indole-oxazolospirones (M2), acenaphthene thioether cyclopentanes (M3), N-methylindolepropionamide-pyridinemethylamines (M4), naphthylmethylphosphonohydroxamic acids (M5), aminopyranosides (M6), triazole methoxyphenyl furanylbenzoic acid (M7), dibenzylamine succinate derivatives (M8), biotinylate thionyl sulfoxide bromophenyls (M9), and tetrahydroindolesulfonylpyrrolidone benzyl groups (M10), exhibiting good structural diversity. Ten compounds had molecular weights ranging from 293 to 492 g / mol, with LogP values ​​between 2.0 and 5.2. Nine compounds (M1, M2, and M4-M10) met the Lipinski five-rule criteria, while M3 was deemed non-compliant due to its LogP (5.2) exceeding the upper limit (LogP ≤ 5) (Table 2). ADMET property predictions (Table 3) showed that most candidate compounds had moderate water solubility, low to moderate Caco-2 permeability, BBB permeability, CYP metabolism, and hERG inhibition risk, and low hepatotoxicity and skin sensitization risk. The predicted oral LD50 (rat) was approximately 500-2,000 mg / kg. M4 had the highest molecular weight (293.37 g / mol), LogP (2.8), and TPSA (46.92 Å). 2 It exhibits balanced performance across indicators such as ), making it a key candidate for future monitoring.

[0071] (3) The dose-effect relationship of lead compounds M1–M10 on the interaction of PHB1–PHB2 and the specificity comparison of M4 with the reference molecule. Based on the NanoBiT protein complementary fluorescence system, this invention investigated the M1-M10 protein expression in HEK293T-SmBiT-PHB1 / LgBiT-PHB2 co-expressing cells for approximately 10... -13 M to 10 -6 The dose-response relationship was determined within the M range, and nonlinear regression fitting was performed on each curve to obtain the corresponding Log IC50 and Emax values. All 10 compounds exhibited typical concentration-dependent S-shaped activation curves. Figure 2 The A and Emax values ​​were significantly higher than the baseline signal, consistent with the characteristics of molecular glue-like interactions. Among them, M4 showed the best activity, with Log IC50 = -10.87 (IC50 is approximately 10). -11The M-order value (on the order of M) and Emax = 179.2% significantly outperformed the other nine candidate molecules (except for M4, whose Log IC50 values ​​ranged from -9.78 to -6.88 and whose Emax values ​​ranged from 164.8% to 182.0%). Figure 2 B).

[0072] To further confirm the structure-specific stabilizing effect of M4 on the PHB1-PHB2 complex, this invention uses indole-3-propionic acid (IPA, a simple analog of the M4 skeleton) and the classic non-targeted indole anti-inflammatory drug indomethacin as control molecules for native PAGE analysis. Figure 2 The results showed that the intensity of the PHB1 / 2 complex band in the DMSO control group was significantly reduced under LPS stimulation; M4 treatment significantly restored the band intensity to near the unstimulated level, while IPA and indomethacin had no significant restorative effect. Comparison of dose-response curves of control molecules on the same NanoBiT platform ( Figure 2 The D) shows that M4 is in 10 -10 NanoBiT signaling can be significantly activated at around M (Emax ≈ 180%), while IPA and indomethacin only activate at ≥10 M. -7 A slight increase in signal intensity (Emax ≈ 115-120%) was observed at the M concentration, while DMSO showed no significant activity, further supporting the specific molecular glue effect of M4 from a functional perspective. Based on the above data, M4 was identified as the top-1 hit compound screened in this invention.

[0073] (4) Analysis of the molecular descriptor of compound M4 and its binding mode to the PHB1-PHB2 interface pocket After confirming M4 as the most active leading compound, this invention performed detailed molecular descriptor calculations and molecular docking analysis. The SMILES of M4 are O=C(CCc1cn(C)c2ccccc12)NCc1ccncc1, with a total of 41 atoms (22 heavy atoms), an exact molecular weight of 293.366 Da, a charge of 0, and the molecular formula C1. 18 H 19 N3O, consisting of one segment and five rotatable bonds, has a simple overall structure and is charge-neutral, belonging to a typical "leader-like" segmental molecule; its structure is composed of three parts: an N-methylindole ring, a propionamide linker arm, and 4-pyridinemethylamine. Figure 3 Drug-grade radar chart of M4 ( Figure 4 The results show that it falls within the recommended drug-like space in multiple dimensions of drug-likeness, including TPSA, HBD, LogP, MW, hepatotoxicity, hERG risk, CYP inhibition risk, BBB penetration, Caco-2 penetration, and water solubility.

[0074] Molecular docking results show ( Figure 5 M4 (green rod-shaped) can be stably embedded in the shallow groove of the PHB1 (pink)-PHB2 (magenta) heterodimer interface. Its indole and pyridine rings are anchored to the hydrophobic subpockets on the PHB1 and PHB2 sides, respectively. The propionamide linker arm extends along the interface direction, allowing M4 to form a "molecular bridge" across the two subunits. M4 simultaneously forms hydrogen bonds, salt bridges, and π-cation interactions with residues such as R143, E144, L145, S147, R148, and S151 on the PHB1 side and K142, S143, V140, V144, V168, and K147 on the PHB2 side. Two-dimensional interaction fingerprint ( Figure 6 This further confirms that M4 utilizes both polar / charged residues and hydrophobic residues to participate in complex binding, which is highly consistent with the "polar-hydrophobic mixed" property of the PHB1-PHB2 interface.

[0075] (5) Chemical synthesis route and NMR structure confirmation of compound M4 This invention completes the chemical synthesis and structural characterization of M4. The synthetic route is as follows: Figure 7 As shown in A: Starting from 1H-indole-3-propionic acid, a 1-methyl-1H-indole-3-propionic acid intermediate was obtained by mild N1-position methylation with CH3I / NaH / THF in step (a); this intermediate was then subjected to amide condensation with 4-pyridinemethylamine in step (b) at room temperature using HATU / DIPEA / DCM to obtain the target compound M4. Specific feed ratios and experimental procedures are detailed in section VII, "Chemical Synthesis and Structural Characterization of Compound M4," above. The ¹H-NMR (DMSO-d6) spectrum of the synthesized compound M4 is shown below. Figure 7 B) shows: multiplets of aromatic hydrogens of indole / 4-pyridine ring at δ 7.57-6.99 ppm (integrated for about 8H), characteristic peak of pyridine methylamine methylene at δ 4.28 ppm (2H), singlet of indole N-methyl at δ 3.71 ppm (3H), and multiplets of propionamide chain methylene at δ 2.99-2.53 ppm (total about 4H); the chemical shifts, peak shapes, and integration ratios of each signal are in perfect agreement with the expected structure of M4, confirming that the synthesized compound is the target structure M4 and has good purity.

[0076] (6) Inhibitory effect of compound M4 on LPS-induced IL-6 and IL-8 expression in vitro To evaluate the in vitro inhibitory effect of M4 on acute lung injury-related inflammatory response, this invention used an LPS-induced human alveolar epithelial / monocyte-macrophage inflammation model, and detected changes in IL-6 and IL-8 at the mRNA level (qRT-PCR) and the secreted protein level (ELISA), respectively. qRT-PCR results ( Figure 8 A and Figure 8 (B) showed that, compared with the LPS− / DMSO group, the relative expression levels of IL-6 and IL-8 mRNA in the DMSO group stimulated by LPS increased from approximately 1.6 and 1.3 to approximately 4.8 and 5.8 times, respectively (P<0.0001); in the M4 pretreatment group, the IL-6 mRNA level decreased to approximately 2.3 and the IL-8 mRNA level decreased to approximately 3.0 under LPS stimulation (P<0.0001), a decrease of approximately 50%. ELISA results ( Figure 8 C and Figure 8 (D) The results showed that, under LPS stimulation, the concentration of IL-6 in the culture supernatant of the DMSO group increased from approximately 1,000 pg / mL at baseline to approximately 8,000 pg / mL, and the concentration of IL-8 increased from approximately 150 pg / mL at baseline to approximately 1,800-2,000 pg / mL. In the M4 pretreatment group, the concentration of IL-6 decreased to approximately 4,000 pg / mL, and the concentration of IL-8 decreased to approximately 900 pg / mL, representing a decrease of approximately 50% compared with the LPS+DMSO group (both P < 0.0001). Under LPS-negative conditions, M4 had no significant effect on baseline levels, suggesting that M4 has good specificity. The above data indicate that M4 can significantly inhibit the expression of LPS-induced IL-6 / IL-8 pro-inflammatory cytokines at both the transcriptional and secretory protein levels, and has a clear in vitro anti-inflammatory activity.

[0077] (7) In vivo efficacy evaluation of compound M4 in an LPS-induced mouse model of acute lung injury This invention further evaluated the in vivo efficacy of M4 in an LPS-induced mouse acute lung injury (ALI) model. BALF cell differential count ( Figure 9 A) showed that the total number of BALF cells and neutrophils in the LPS+DMSO group were significantly higher than those in the NS+DMSO group (P<0.0001), indicating severe alveolar inflammatory cell infiltration. In the M4 treatment group (LPS+Molecular), the total number of BALF cells and Neu count were significantly lower (P<0.0001), while the number of macrophages and lymphocytes did not differ significantly, suggesting that M4 mainly inhibits neutrophil-mediated acute pulmonary inflammatory infiltration.

[0078] BALF / lung tissue homogenate cytokine detection results ( Figure 9BF (Bronchoalveolar blotting) showed that the levels of IL-6 (approximately 210 pg / mL), IL-1β (approximately 690 pg / mL), TNF-α (approximately 185 pg / mL), CXCL1 (approximately 950 pg / mL), and CXCL2 (approximately 500 pg / mL) in the LPS+DMSO group were significantly higher than those in the NS+DMSO group (P<0.0001), exhibiting a clear "cytokine storm" characteristic; M4 treatment reduced the levels of the above mediators to approximately 70 pg / mL (IL-6), 200 pg / mL (IL-1β), 90 pg / mL (TNF-α), 500 pg / mL (CXCL1), and 220 pg / mL (CXCL2), respectively (P<0.0001), a reduction of approximately 45-70%. Lung tissue HE staining and pathological scoring results ( Figure 10 The results of the AB study showed that the lung tissue structure was normal in the NS+DMSO and NS+M4 groups; the LPS+DMSO group showed significant alveolar wall thickening, interstitial edema and dense inflammatory cell infiltration, and the HE score increased from about 0.3 in the NS group to about 3.8 (P<0.0001); the pathological changes in lung tissue were significantly reduced in the LPS+M4 treatment group, and the HE score decreased to about 1.2 (P<0.0001), close to the baseline level of NS.

[0079] Survival analysis ( Figure 10 The results (C) showed that mice in the DMSO group died rapidly on days 4-6 after LPS modeling, with a survival rate of only about 5-10% on day 8; while the survival rate of mice in the M4 treatment group was significantly improved to about 65-70% (P<0.001, Log-rank (Mantel-Cox) test). Based on the above in vitro and in vivo pharmacodynamic data, as a molecular glue compound targeting the PHB1-PHB2 heterodimer interface, M4 significantly inhibited LPS-induced neutrophil lung infiltration, release of pro-inflammatory cytokines and chemokines, and pathological damage to lung tissue by stabilizing the PHB1-PHB2 complex, and significantly improved the survival rate of ALI model mice at the overall level, demonstrating consistency between "mechanism-efficacy-overall survival," providing sufficient pharmacodynamic evidence for the use of PHB1-PHB2 interface stabilizers in the treatment of acute lung injury and related inflammatory lung diseases.

[0080] Example 2: Preparation of a pharmaceutical composition containing compound M4 As a preferred embodiment of the present invention, a representative pharmaceutical formulation is prepared using compound M4 as the active ingredient according to the following formulation: 1) Injectable solutions: Formulation: Compound M4 5 mg, phosphate buffer (PBS, pH 7.4) as needed, 5% (v / v) dimethyl sulfoxide (DMSO) as a cosolvent, polyethylene glycol 15-hydroxystearate (Solutol HS 15) as a surfactant, and water for injection to 1 mL.

[0081] Preparation process: Dissolve compound M4 in DMSO, then add Solutol HS 15 and stir until homogeneous. Slowly add PBS to bring the volume to 1 mL, filter through a 0.22 μm sterile filter membrane, and dispense into sterile glass vials.

[0082] 2) Oral capsules: Formula: Compound M4 10 mg, microcrystalline cellulose 100 mg, lactose 80 mg, croscarmellose sodium 5 mg, magnesium stearate 2 mg.

[0083] Preparation process: Mix compound M4 with the above excipients in equal increments until homogeneous, pass through a 60-mesh sieve, and fill into hard capsules.

[0084] The dosage forms and formulations of the above-mentioned pharmaceutical compositions are only preferred representative embodiments of the present invention. Those skilled in the art can use other pharmaceutically acceptable dosage forms (including tablets, solutions, powders, vesicles, injectable powders, etc.) and formulations to formulate the compositions according to actual needs. All such variations are within the protection scope of the present invention.

[0085] Unless otherwise specified, the raw materials and equipment used in this invention are all commonly used in the field; unless otherwise specified, the methods used in this invention are all conventional methods in the field.

Claims

1. A molecular adhesive compound targeting the PHB1-PHB2 heterodimer interface, characterized in that: The following are compounds having the following molecular structural formulas or their pharmaceutically acceptable salts, solvates, stereoisomers or prodrugs; 。 2. A method for preparing the molecular adhesive compound as described in claim 1, characterized in that: Includes the following steps: Step (a): Starting with 1H-indole-3-propionic acid, the reaction is carried out in the presence of a base, a methylating agent, and a solvent to selectively complete the N1-methylation of indole, yielding the 1-methyl-1H-indole-3-propionic acid intermediate. Step (b): The 1-methyl-1H-indole-3-propionic acid intermediate is subjected to an amide condensation reaction with 4-pyridinemethylamine in a solvent under the action of a condensing agent and an organic base. After the reaction is completed, the product is purified to obtain the target product.

3. The preparation method according to claim 2, characterized in that: In step (a), The alkali is NaH; The methylating agent is CH3I; The solvent is tetrahydrofuran; The molar ratio of 1H-indole-3-propionic acid, NaH, and CH3I is 1:(4-6):(2-4). The reaction is carried out at a temperature of 0°C to room temperature.

4. The preparation method according to claim 2, characterized in that: In step (b), The condensing agent is 2-(7-benzotriazole oxide)-N,N,N′,N′-tetramethylurea hexafluorophosphate; The organic base is N,N-diisopropylethylamine; The solvent is dichloromethane; The reaction was carried out at room temperature. The molar ratio of 1-methyl-1H-indole-3-propionic acid, condensing agent, organic base, and 4-pyridinemethylamine is 1:(1.1-1.3):(1.8-2.2):(1.1-1.3). The purification was performed by column chromatography.

5. The use of the molecular glue compound as described in claim 1 or the molecular glue compound obtained by the preparation method according to any one of claims 2-4 in the preparation of a drug for a stable PHB1-PHB2 heterodimer complex.

6. The use of the molecular glue compound of claim 1 or the molecular glue compound obtained by the preparation method of any one of claims 2-4 in the preparation of medicaments for the prevention and / or treatment of acute lung injury, acute respiratory distress syndrome or cytokine storm-related inflammatory lung diseases.

7. A drug for the prevention and / or treatment of acute lung injury, acute respiratory distress syndrome, or cytokine storm-related inflammatory lung disease, characterized in that: The active ingredient is a molecular adhesive compound as described in claim 1 or a molecular adhesive compound prepared by any one of claims 2-4, and includes a pharmaceutically acceptable carrier, excipient or solvent.

8. The drug as described in claim 7, characterized in that: The drug is available in oral or injectable form.

9. The medicament as described in claim 8, characterized in that: The oral preparation is a capsule, tablet, solution, or powder; The injectable preparation is a vesicle, injection, or injection powder.

10. A method for screening molecular glue compounds that can prevent and / or treat acute lung injury, acute respiratory distress syndrome, or cytokine storm-related inflammatory lung disease, characterized in that: Using the shallow groove pocket at the PHB1-PHB2 heterodimer interface as the target site, druggable shallow groove pockets located at the PHB1-PHB2 heterodimer interface were identified based on the cryo-electron microscopy structure of the PHB complex. The optimal molecular glue compound was identified by de novo molecule generation and virtual screening combined with NanoBiT cell activity assay.

Images