General chemical information

All regents and solvents were purchased from commercially available sources and were used without further purification. All reactions were conducted under appropriate pressure and temperature in glassware that had been oven-dried prior to use. Thin-layer chromatography (TLC) was performed using precoated silica gel plates from Qingdao Haiyang Chemical Co. Flash column chromatography was performed with the silica gel (particle size of 40–63 μm). 1H and 13C nuclear magnetic resonance (NMR) spectra were recorded on Bruker AVANCE NEO 600 MHz (operating at 600 MHz for 1H and 151 MHz for 13C acquisitions) and JNM-ECZ400S spectrometers. High-resolution mass spectra Analysis were obtained on AB Sciex API 3000™ LC/MS.

Biological assaysReagents

Thalidomide (#T0213), pomalidomide (#T2384), lenalidomide (#T1642), CC-122 (#T3549), CC-220 (#T7791), CC-885 (#T14893), MLN4924 (#6332), dexamethasone (#T1076), tazemetostat (#T1788) and bortezomib (#T2399) were purchased from Targetmol, Boston, MA, USA. Cycloheximide (#S7418) was purchased from Selleck, Houston, TX, USA. The synthesis and characterization of molecular glue degraders in this study are described in Supplementary Information. All drugs were dissolved in dimethylsulfoxide (DMSO, #D8418; Sigma-Aldrich, MO, USA) at 10-200 mM, stored at -30 °C as stock solutions.

Plasmids

The pcDNA3.1(+) plasmids were purchased from Invitrogen, Carlsbad, CA, USA. The open reading frames (ORFs) of IKZF1, IKZF2, IKZF3, CK1αWT and CK1αG40N were amplified, and restriction enzyme sites were added by PCR and cloned into pcDNA3.1(+)-AGIA-MCS. The primers used in PCR experiments were available in Table S1

Cell culture and transfection

HEK293T (#CRL-3216) cells were cultured in high-glucose Dulbecco’s modified Eagle’s medium (DMEM, #SH30022.01; Hyclone, Logan, UT, USA). MM cell lines including RPMI-8226, NC-H929, OPM-2, and AML cell lines including U937, MOLM-13, and Ocl-Ly3 as well as DLBCL cells lines including SU-DHL-4, WSU-DLCL-2, TMD8 and U2932 were cultured in Roswell Park Memorial Institute (RPMI, #SH30809.01; Hyclone, Logan, UT, USA). KG-1 and MV-4-11 cells were cultured in Iscove’s Modified Dulbecco’s Medium (IMDM, #SH30228.01; Hyclone). All these cell lines were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA), cultured in medium containing 10% fetal bovine serum (FBS, #16000-044, Gibco), 100 U/mL penicillin, and 100 µg/mL streptomycin (#15140122, Thermo Fisher Scientific, MA, USA) at 37 °C under 5% CO2.

For the generation of HEK293T cell lines stably expressing IKZF1-HiBiT, IKZF2-HiBiT, IKZF3-HiBiT, CK1α-HiBiT, respectively, lentivirus was produced in HEK293T cells by transfection of pCSII-CMV-neosubstrate-HiBiT-IRES2-Bsd expression vector together with pCMV-VSV-G-RSV-Rev and pCAG-HIVgp vector as described in previous study [24]. HEK293T cells supplemented with 10 µg/mL polybrene (#TR-1003, Sigma-Aldrich, MA, USA) were infected with the lentivirus. After 48 h infection, puromycin (3 µg/mL) was utilized to select stable cell lines and pooled clones were screened by immunoblot analysis.

Antibodies

The following primary antibodies were used: CRBN (#71810, 1:1000), IKZF1 (#14859, 1:1000), IKZF2 (#42427, 1:1000), IKZF3 (#15103, 1:500) and were all from Cell Signaling Technology, Boston, MA, USA; CK1α (#ab108296, 1:1000) and ZNFX1 (#ab179452, 1:500) were purchased from Abcam, MA, USA; BRD9 (#24785-1-AP, 1:1000), GSPT1 (#10763-1-AP, 1:1000), STAT5 (#13179-1-AP, 1:500), c-Myc (#10828-1-AP, 1:1000), MDM2 (#27883-1-AP, 1:1000), P53 (#1044-1-AP, 1:1000), P21 (#10355-1-AP, 1:500), DTWD1 (#26810-1-AP, 1:1000), MBD3 (#14258-1-AP, 1:1000), MNT (#23742-1-AP, 1:1000) and MBD1 (#29998-1-AP, 1:1000) were purchased from Proteintech, Rosemont, IL, USA; ZMIZ2 (#GTX118779, 1:1000) were purchased from GeneTex, San Antonio, TX, USA; ZFP91 (#CSB-PA504466, 1:1000) was purchased from Cusibio, Houston, TX, USA; Anti-rabbit IgG (HRP-conjugated, Cell Signaling Technology, #7074, 1:5000), anti-mouse IgG (HRP-conjugated, Cell Signaling Technology, #7076, 1:5000) were used as secondary antibodies.

NCI-H929 CRBN knock-out

CRBN−/− NCI-H929 cells were created using CRISPR-Cas9 technology. NCI-H929 cells (1 × 106 cells/well) were cultured in 6-well dishes and were transiently transfected with Plasmid Transfection Medium (#sc-108062) precomplexed of CRBN CRISPR/Cas9 KO plasmid (#sc-412142), CRBN HDR plasmid (#sc-412142-HDR), and UltraCruz® Transfection Reagent (#sc-395739) following the instructions provided by the manufacturer (Santa Cruz Biotechnology, Inc, Dallas, TX, USA). After 72 h infection, puromycin (3 µg/mL) was utilized to select stable cell lines. Transfected cells (GFP+) were single-cell sorted by flow cytometry into 96-well tissue culture-treated plates 7 days, and isolated single cell clones were screened by DNA sequencing and western blot analysis.

Clinical specimens, cell culture and correlation analysis

The use of human experimental materials related to this study, including cDNA derived from cell lines and clinical specimens (the clinical specimens used in this study were mainly various specimens of multiple myeloma). The patients’ inclusion diagnosis and exclusion criteria; the information of clinical specimens; collection and preparation of clinical specimens were descripted in our previous publication [25] according to the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®), based on the Durie and Salmon, ISS and R-ISS prognostic staging criteria.

Briefly, the Diagnosis, inclusion and exclusion criteria of patients with multiple myeloma: (1) for patients with bone tumor or extra-bone tumor, biopsy-proven solitary lesion of bone or soft tissue results consisting clonal plasma cells; normal-random bone marrow biopsy results with no evidence of clonal plasma cells; normal skeletal survey, MRI or CT results except for the primary solitary lesion; absence of end-organ damage. (2) for patients with intraosseous / intra-marrow MM (e.g. the plasma cell myeloma), percentage ≥ 10% of clonal bone marrow plasma cell or biopsy-proven plasmacytoma ≥ 1 of the following myeloma-defining events. (3) patients with blood / intra-blood MM (e.g. the plasma cell leukemia), at circulating plasma cells for least 20% and in peripheral blood of at least 2 × 109/L a total plasma cell count.

In this study, clinical specimens were used for two purposes: (1) the cDNA samples reversely transcripted by total RNA from clinical specimens descripted in our previous publication [25]; (2) Patient-derived cell lines prepared from clinical specimens. The clinical specimens including (1) 35 samples of intra-marrow MM; (2) 42 samples of intra-blood MM; (3) 38 samples of MM with bone-tumor tissues in the form of masses or lumps; and (4) 21 MM samples with long-distance metastasis, or extra-bone masses or lumps forming a solid tumor tissue or mass in other organs.

The protocol of preparation PDCs was descripted in our previous publication [25]. Briefly, for the MM cells in the bone, the bone marrow aspirate was directly sorting by CD38 (CD38+); (2) for the cells in peripheral blood, the peripheral blood lymphocytes are directly sorting by CD38+; (3) for cells in bone tumors, the tumor tissue was abraded by using a pre-sterilized 200-mesh steel sieve by the DMEM with 20% FBS to single-cell suspension and the single-cell suspension was washed; (4) for cells in peripheral organ tumors, the tumor tissue was abraded by using a pre-sterilized 200-mesh steel sieve by the DMEM with 20% FBS to single-cell suspension and the single-cell suspension was washed. The PDC lines were generated for one cell line by one patient only. PDC-1 (patients derived cell No. 1) and PDC-3 were directly separated, cultured, and stored (PDC-1 generated from intra-marrow MM patient; PDC-3 was generated from intra-blood MM patient). PDC-5 was generated from bone-tumor tissues, and PDC-7 was generated from extra-bone tumor tissues. As control, the primary B cells which not only derived from patients with multiple myeloma which were derived from PBMC from patients; but primary B cells from healthy individuals which were derived from the peripheral blood lymphocytes of the remaining part of the whole peripheral blood provided by the blood transfusion department except plasma and red blood cells. The primary B cells were purified based on the CD19+/CD45+/CD38-sorting. In the cell sorting process, the purity of primary B cells isolated from healthy human PBMCs could reach 99%; the purity of primary B cells isolated from patients could also reach 95–99%. When patient-derived cell lines were isolated from multiple myeloma patients, the purity of the cell lines isolated from peripheral blood was over 95% (up to 98-99%), and the purity of patient-derived cell lines isolated from tumor tissue was over 90%.

The correlation between the expression of BRD9 with IKZF1, IKZF2, IKZF3 and CK1α, respectively, were detected. At this time, taking the expression level of BRD9 as the abscissa and the expression of CRBN as the ordinate, each specimen can correspond to a data point. A group of specimens can correspond to a group of data points, which can be fitted and linearly regressed to obtain a regression equation and P value.

Cell viability and drug synergy assays

In order to screen the molecular glue library in MM, AML and DLBCL cell lines, RPMI-8226, NCI-H929, MV4-11, U937, and WSU-DLCL-2 cells were cultured in the complete medium recommended by the vendor. The cells were seeded in corning BC white 96-well assay plates at a density of 12,000 cells per well. During the experimental period of 4 days, the cells were treated with DMSO, thalidomide, lenalidomide, pomalidomide, CC-122, or the corresponding compounds at specific concentrations as indicated.

To assess the inhibitory effect of synthesized degraders, MM, AML and DLBCL cells were cultured in 96-well plates with DMSO, pomalidomide, or the corresponding compounds at specified concentrations for a duration of 4 days. The compounds’ ability to inhibit cell proliferation was evaluated using a cell counting kit-8 (CCK-8 Kit) following the instructions provided by the manufacturer (Dojindo Laboratories, Mamoto Ken, Japan).

Drug synergy assays were conducted by seeding cells in 96-well plates (100 µL per well) at a density of 12,000 cells per well. The cells were then treated with the specified drug doses for a duration of 96 h, with DMSO serving as the control. Cell viability was assessed using the CCK-8 Kit, following the manufacturer’s instructions. The synergy score plots for the indicated drug combinations were calculated using SynergyFinder.

Protein preparation and molecular docking

The crystal structure of CRBN with lenalidomide was downloaded from Protein Data Bank (PDB) entry 4TZ4 [26]. The proteins were prepared using Maestro by Protein Preparation Wizard, and the missing side chains and loops were filled by Prime [27]. The ionization state of the ligand suitable for pH 7.0 ± 2.0 was predicted by Epik [28] with the OPLS3 force field [29]. Grids for the binding site were defined using the crystal structure through Receptor Grid Generation. Ligands were docked into CRBN by Glide SP with default parameters.

MD simulations and binding free energy calculation

The CRBN underwent 500 ns MD simulation using Gromacs 2021.7 [30]. The lenalidomide and other ligands were parameterized with the general AMBER force field (GAFF) [31], and their topology and parameter files were prepared using the antechamber module [32], further converted into the GROMACS format using ACPYPE [33]. The simulation system was filled with TIP3P water to solvate the protein-ligand complexes. The AMBER14sb force field [34] was used throughout the calculation steps, followed by the addition of Na+ and Cl− in the water to render the system neutralized. And the sodium chloride molecules were added to reach the physiological concentration of 0.15 M. To eliminate unfavorable contacts between the protein and water molecules in the system, energy minimization and pre-equilibration simulations were conducted in three sections before the production simulation. The system was minimized by 5000 steps or Fmax < 10, and the 100 ps NVT and 1.0 ns NPT pre-equilibration was performed with time steps of 2 fs. The Particle-Mesh-Ewald (PME) method was used to calculate long-range electrostatic interactions, with the reference temperature of 310 K. H-bonds were constrained with the LINCS algorithm [35], and the Van der Waals cutoff was 1.0 nm. For pre-equilibration, the Nose-Hoover and Berendsen methods were employed for temperature and pressure coupling respectively. Finally, the production MD simulation was conducted at NPT ensemble with Nose-Hoover temperature coupling and Parrinello-Rahman pressure coupling after the system had been well-equilibrated at the desired temperature and pressure. The binding free energy of degraders were calculated using Molecular Mechanics/Poisson Boltzmann Surface Area (MM/PBSA) method.

Cell apoptosis assay

Cells (1 × 106 cells/well) were cultured in 6-well dishes. Subsequently, the cells were treated with a medium containing different concentrations of compounds for 3 days. To analyze cell apoptosis, the apoptosis rates were measured using the Annexin V-FITC apoptosis detection kit (Abcam, MA, USA). Sample analysis was performed using a FACS Calibur Flow Cytometer (Becton Dickinson, Franklin Lakes, NJ, USA).

Quantitative degradation assay using HiBiT system

To assess protein degradation of IKZF1, IKZF2, IKZF3, and CK1α, HEK293T cells that were stably expressing IKZF1-HiBiT, IKZF2-HiBiT, IKZF3-HiBiT, or CK1α-HiBiT were cultured in 96-well plates at a density of 12,000 cells per well. These cells were then treated with either DMSO or molecular glue degraders for a duration of 24 h. Subsequently, the cells were lysed using the Nano-Glo HiBiT Lytic Detection System (N3040, Promega) as per the manufacturer’s instructions. The luminescent signals of the HiBiT-tagged neosubstrates were detected using the SpectraMax iD3 (Molecular Devices) instrument.

Immunoblot analysis

Cells were seeded in 6-well plates and treated with various concentrations of compounds for the specified durations. Afterwards, the cells were washed with phosphate-buffered saline (PBS) and lysed using RIPA buffer. The protein concentration was measured, and the total protein lysates were then subjected to separation by 10% SDS-PAGE and transfer onto a nitrocellulose membrane. The membranes were subsequently probed with specific primary and secondary antibodies, and imaging was performed using the Bio-Rad Imaging system (Hercules, CA, USA).

RT-qPCR

RNA samples of MM patient cells or hematological cancer cells were isolated and extracted using TRIzol reagent according to the manufacturer’s instructions (Invitrogen). A minimum of 2 µg of total RNA was reverse transcribed into first strand cDNA with oligo (dT) primers using Moloney murine leukemia virus reverse transcriptase (Promega, Madison, WI USA). Quantitative polymerase chain reaction (qPCR) was performed in triplicates in a 20 µL reaction mixture containing 10 µL of SYBR Premix Ex Taq Master Mix (2×) (Takara Bio, Shiga, Japan), 0.5 mM of each of the primers and 10 ng cDNA. The relative expression level of the target was calculated using the comparative Ct method. β-actin was used as an internal control to normalize sample differences. The primers used in qPCR experiments were available in Table S1.

MS-based proteomic analysisSample preparation

NCI-H929 cells were treated with DMSO or MGD-4 (1, 10 µM), MGD-28 (1, 10 µM), pomalidomide (10 µM) and CC-220 (1 µM) for 4 h. Afterward, the cells were washed with PBS three times and lysed using 8 M UA (8 M urea, 100mM Tris-HCl, pH 8.0) containing complete protease inhibitor tablets. The supernatant was then collected by centrifuging the lysed cells at 14,000 rpm for 15 min at 4 °C. The collected supernatant was reduced with 10 mM Tris (2-carboxyethyl) phosphine hydrochloride solution (TCEP) at room temperature for 30 min, followed by cysteine alkylation with 50 mM 2-chloroacetamide (CAA) at room temperature for 30 min. Subsequently, the denatured proteins were digested overnight at 37 °C using trypsin (at a 1:50 ratio of enzyme to protein). The reactions were quenched by adding formic acid (FA) to a final concentration of 0.1%. To identify substrates targeted by MGD-4, MGD-28, or other IMiDs, samples were desalted and subjected to label-free proteomics analysis. For the investigation of substrates involved in IMiDs resistance mechanisms, samples were desalted, then labeled with TMTproTM-16plex reagents, and subsequently combined and desalted again for mass spectrometry analysis. TMT labeling steps were performed according to the manufacturer’s instructions, and the labeling channel design was described as followed: 126 QC; 127 N, 127 C, 128 N, 128 C, and 129 N represented five biological replicates of wild-type cells; 129 C, 130 N, 130 C, 131 N, and 131 C represented five biological replicates of cells resistant to lenalidomide; 132 N, 132 C, 133 N, 133 C, and 134 N represented five biological replicates of cells resistant to pomalidomide.

LC–MS/MS analysis

For samples to identify the substrates of MGD-4, MGD-28, and other IMiDs, the corresponding peptide mixtures were analyzed using an EASY-nLC 1200 liquid chromatography system (Thermo Fisher Scientific) with a home-made 15 cm C18 column (ID 150 μm, 1.9 μm, 100 Å). Peptide separation was carried out over a 78-minute gradient at a constant flow rate of 600nL/min: 7–12% B in 8 min, 12–32% B in 45 min, 32–45% B in 13 min, 45–95% B in 2 min, followed by a 10-minute hold at 95% B (Buffer A: 0.1% FA, Buffer B: 0.1% FA in 80% acetonitrile), with an additional 15-minute wash. The peptide mixture was analyzed using an Orbitrap Fusion mass spectrometer (Thermo Fisher Scientific). MS data were acquired using a data-dependent acquisition (DDA) method, with a dynamic exclusion duration of 18 s. For the MS1 scan, mass spectra were acquired in the positive-ion mode in the range of 300–1400 m/z, with a maximum ion injection time of 50ms and a resolution of 120,000 at m/z 200. Fragmentation of precursor ions was achieved by higher-energy collision dissociation (HCD) with a normalized collision energy of 33%. The MS2 spectra were acquired with an automatic gain control target value of 1.0e4 and a maximum injection time of 35 ms.

For samples to investigate the resistance mechanisms of lenalidomide and pomalidomide, the corresponding samples were analyzed using Orbitrap Exploris 480 mass spectrometer coupled with a high-field asymmetric waveform ion mobility spectrometer using − 45 V and − 65 V. The peptides were trapped in a home-made 2 cm solid-phase extraction column (ID 100 μm) and separated by a home-made 20 cm C18 column (ID 75 μm, 1.9 μm, 100 Å) with a constant flow rate of 300nL/min. The gradient was 7–12% B in 10 min, 12–30% B in 80 min, 30–45% B in 20 min, 45–95% B in 1 min, followed by a 9-minute hold at 95% B (Buffer A: 0.1% FA, Buffer B: 0.1% FA in 80% ACN), with an additional 15-minute wash. MS data were acquired using a data-dependent acquisition (DDA) method, with a dynamic exclusion duration of 30 s. MS1 scan was conducted at mass range of 375–1400 m/z, with a maximum ion injection time of 50 ms and a resolution of 120,000 at m/z 200. HCD energy was 34%. The MS2 spectra were acquired with resolution of 45,000, isolation window of 0.5 m/z, and maximum injection time of 120 ms.

MS data analysis

The label-free MS raw files were analyzed using MaxQuant (version 2.0.3.0), while the TMT-labeled MS files were analyzed using Proteome Discoverer (PD) (version 2.5.0.400). Both were matched against the UniProt Human database (downloaded in Sep 2022, containing 20398 entries). The protease was set as trypsin/P with a maximum of two missed cleavages. Carbamidomethyl (C) was designated as a fixed modification, while Oxidation (M) and Acetyl (Protein N-term) were defined as variable modifications. A false discovery rate (FDR) of ≤ 0.01 was applied at the spectra, protein, and modification levels. All other settings remained as default. Proteins identified from the contaminated and reversed database were excluded. The protein groups results generated by MaxQuant and PD were subsequently analyzed in R (version 4.2.1). Statistical analysis was performed using student’s t-test, and p values < 0.01 were considered statistically significant.

Data availability

All the mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (https://proteomecentral.proteomexchange.org) via the iProX partner repository [36, 37] with the dataset identifier PXD053334.

Animal models for tumor growth

Animal research was conducted in accordance with the guidelines of the Animal Care and Use Committee of the Beijing Institute of Biotechnology. Six-week-old BALB/c nude mice and Non-obese diabetic Severe combined immunodeficiency (NOD/SCID) mice were obtained from SiPeiFu company, Beijing, China, and housed in a specific pathogen-free (SPF) animal facility. NCI-H929 cells (5 × 106) were injected subcutaneously into the dorsal flank of mice. For the evaluation of our compounds as single agent, once the NCI-H929 tumor volume reached approximately 80 mm3, BALB/c nude mice were divided into six groups and were treated with pomalidomide (10 mg/kg, p.o./qd), MGD-4 (3, 10 mg/kg, p.o./qd), MGD-28 (3, 10 mg/kg, p.o./qd), and a vehicle control (10% DMSO + 10% PEG300 + 5% Tween 80 + 75% H2O, p.o./qd) for 18 days. For the drug synergy assays, once the NCI-H929 tumor volume reached approximately 80 mm3, the BALB/c nude mice were divided into six groups and treated with single-agent dexamethasone (3 mg/kg, p.o./qd), MGD-4 (3 mg/kg, p.o.qd), MGD-28 (3 mg/kg, p.o./qd), and a vehicle control (10% DMSO + 10% PEG300 + 5% Tween 80 + 75% H2O, p.o./qd), as well as combination treatments of dexamethasone + MGD-4 and dexamethasone + MGD-28 for 14 days. To evaluate the synergistic effect of MGD-28 with bortezomib/tazemetostat, once the NCI-H929 tumor volume reached approximately 80 mm3, the NOD/SCID mice were divided into six groups and treated with single-agent MGD-28 (3 mg/kg, p.o./qd), bortezomib (1 mg/kg, p.o./qd), tazemetostat (100 mg/kg, p.o./qd), and a vehicle control (10% DMSO + 10% PEG300 + 5% Tween 80 + 75% H2O, p.o./qd), as well as combination treatments of MGD-28 + bortezomib and MGD-28 + tazemetostat for 12 days. To evaluate the synergistic effect of I-BRD9 with MGD-28/pomalidomide, once the NCI-H929 tumor volume reached approximately 80 mm3, the NOD/SCID mice were divided into six groups and treated with single-agent pomalidomide (10 mg/kg, p.o./qd), MGD-28 (3 mg/kg, p.o./qd), I-BRD9 (10 mg/kg, p.o./qd), and a vehicle control (10% DMSO + 10% PEG300 + 5% Tween 80 + 75% H2O, p.o./qd), as well as combination treatments of I-BRD9 + MGD-28 and I-BRD9 + pomalidomide for 24 days. The tumor volume was calculated using the formula V = (longest diameter × shortest diameter^2)/2. Tumor growth inhibition (TGI) was calculated to determine the inhibitory strength of the drugs on tumor growth. TGI (%) = (Vc - Vt) / (Vc - V0) × 100, where Vc is the median volume of the control group, Vt is the median volume of the treated groups at the end of the study, and V0 is the median volume of the control group at the start of the study. The body weight of the mice was measured every 2 days. The experiment was terminated when the maximum tumor size reached approximately 1.5 cm in diameter. Euthanasia was performed under deep anesthesia, and the tumors were then isolated from the animals, weighed, and photographed.

Pharmacokinetic experiments in ratsPharmacokinetic experiments in rats

Animal experiments were conducted at the Beijing Center for Drug Safety Evaluation, which is approved by the Institutional Animal Care and Use Committee of the Center. These experiments were conducted in accordance with the guidelines set forth by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC). Male Sprague-Dawley (SD) rats weighing between 180 and 210 g were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. and were kept in a controlled environment with regulated temperature, humidity, and a 12-hour light/dark cycle. It was determined that live animals were necessary to achieve the study objectives, as no alternative methods were available. The compounds MGD-4 and MGD-28 were dissolved in a 5% glucose solution, with a final concentration of less than 2% DMSO. A total of 24 rats were randomly divided into four groups and administered MGD-4 and MGD-28 via intravenous bolus (1 mg/kg) and gavage (3 mg/kg), respectively. Rats in the oral administration group underwent a 12-hour fasting period before drug administration, with unrestricted access to water. Blood samples (0.15 mL) were continuously collected prior to administration and at designed time points up to 24 h post intravenous dosing, and oral administration. These blood samples were collected in heparin anticoagulation collection tubes, centrifuged at 4℃ for 10 min (2500 × g) to separate the plasma, and then stored at -20℃. Prior to determining the plasma drug concentration, the samples were thawed at room temperature and a 20 µL plasma sample was taken for quantitative extraction. Acetonitrile (20 µL) and IS (propranolol 10 ng/mL, 100 µL) were added to precipitate the protein through vortexing. After centrifugation at 15,000 g for 10 min, the supernatant was collected and a 5 µL aliquot of the sample was injected into the LC-MS/MS system for drug concentration determination.

Bioanalysis method

The concentrations of MGD-4 and MGD-28 were determined using an LC-MS/MS system. The system consisted of an LC instrument (LC-20AD, Shimadzu) and an 8060 triple quadrupole mass spectrometer detector (Shimadzu, Japan). The compounds were separated on a Phenomenex C18 column (2.1 × 50 mm, USA). An LC gradient was employed, which consisted of a 0.1% formic acid aqueous solution (v/v, mobile phase A) and a 0.1% formic acid in acetonitrile (v/v, mobile phase B). The flow rate was set at 0.6 mL/min, and the run duration was 4 min. The LC separation program was set as follows: 0–0.3 min, 5% B; 0.3–2.0 min, a gradient from 5 to 95% B; 2.0–2.5 min, held at 95% B; 2.6 min, returned to 5% B. The analytes and internal standard were detected using positive ion spray in the multiple-reaction-monitoring modes (MRM). The injection volume was 5 µL. The monitored precursor/product ion mass transitions were as follows: m/z 289.05/178.1 for MGD-4, m/z 596.2/307.95 for MGD-28, and m/z 260.1/116.1 for propranolol (IS). The calibration linear ranges for both MGD-4 and MGD-28 were 1-1000 ng/mL.

Data analysis

The pharmacokinetic parameters were calculated using WinNonlin 9.0 (Pharsight, CA) by noncompartmental method. The maximal plasma concentration (Cmax) and time to reach the peak (Tmax) were obtained directly from the observed data. The area under the plasma concentration-time curve (AUC) from time 0 to the last time point with a measurable concentration was calculated using the trapezoidal method. The area from the last datum point to time infinity was estimated by dividing the last measured plasma concentration by the terminal rate constant. The terminal elimination rate constant (λz) was calculated by log-linear regression of the terminal phase of the plasma concentration-time curves using at least three time points, and the half-life t1/2 was calculated from ln2/λz. Absolute bioavailability was evaluated calculated as: bioavailability (F, %) = [(AUCextra−venous route /AUCiv) × dose iv/doseextra−venous route)] ×100.

Statistical analysis was conducted by Student’s t test between different groups for major pharmacokinetic parameters (Cmax, Tmax, AUC, volume of distribution (V), and clearance (CL). A P-value < 0.05 was considered statistically significant.

Statistical analysis and reproducibility

All in vitro experiments were conducted in triplicate. Differences between variables were evaluated using the Bonferroni correction, two-tailed Student’s t-test or one-way analysis of variance (ANOVA). Statistical analyses were performed using SPSS 13.0 or GraphPad Prism 8.0. The statistical data were expressed as the mean ± standard deviation (SD). A P value < 0.05 was considered statistically significant for all assays.