22-((2-methyl-1-(1H-pyrrole-2-carboxamido)propan-2-yl)thio)-deoxy pleuromutilin (compound PDP) was prepared following the previously described methods (Li et al., 2024; Liu K. et al., 2024; Yi et al., 2025; zhang et al., 2018). With the assistance of dimethylcysteamine, the target pleuromutilin derivative PDP was synthesized by incorporating a pyrrole group into pleuromutilin. Pleuromutilin was first converted to intermediate 2, which was then reacted with dimethylcysteamine hydrochloride to afford intermediate 3 (zhang et al., 2018; Shang et al., 2014). In the current study, intermediate 3 was synthesized using a 20% (m/v) aqueous NaOH solution under rigorously controlled reaction conditions, including optimized temperature and time parameters (Scheme 1).

Pleuromutilin derivative PDP was synthesized by reacting intermediate 3 with prrole-2-carboxylic acid in the presence of either EDCI, HOBT, and DIPEA at room temperature. The yield of pleuromutilin derivative PDP was 53.4%. The purity of the target compounds was determined using high - performance liquid chromatography with an eluent consisting of acetonitrile in water. The results showed that the purity of the target compounds exceeded 95.0%. The synthesized derivative PDP was characterized using proton nuclear magnetic resonance spectroscopy (¹H NMR), carbon - 13 nuclear magnetic resonance spectroscopy (¹³C NMR), and high - resolution mass spectra (HR MS) analytical techniques (Supplementary Data Sheet 1).

The in vitro antibacterial activity of PDP was assessed against nine pathogenic bacterial strains. These strains included Staphylococcus aureus (ATCC 29213), Methicillin-resistant Staphylococcus epidermidis (MRSE, ATCC 51625), Methicillin-resistant S. aureus (MRSA, ATCC 43300), Escherichia coli (ATCC 25922), including clinical isolates of Streptococcus agalactiae and three distinct Staphylococcus dysgalactiae strains (designated as S. dysgalactiae-1, -2, and -3). Infections caused by these pathogens pose life-threatening risks and constitute a major public health concern. Of particular significance, MRSA has emerged as a predominant causative agent of hospital-acquired infections (Hammad et al., 2019). Streptococcus is capable of infecting a broad spectrum of both animals and humans (Cai et al., 2023; Torres et al., 2023). To further characterize its anti-MRSA activity, derivative PDP was evaluated against a panel of 48 clinical MRSA isolates in vitro. Forty-eight clinical MRSA isolates associated with dairy farms were collected from four Chinese provinces (Gansu, Shanghai, Sichuan, and Guizhou) (Zhou et al., 2023). These isolates were provided by the Lanzhou Institute of Husbandry and Pharmaceutical Sciences of the Chinese Academy of Agricultural Sciences.

Tiamulin and valnemulin were employed as reference compounds. Preliminary testing of the derivative PDP against nine bacterial strains demonstrated potent activity against Gram-positive bacteria, outperforming the control drugs tiamulin and valnemulin. The MIC of derivative PDP against tested Gram-positive bacteria is 0.008–0.016 μg/mL. However, derivative PDP exhibited limited activity against E. coli, demonstrating a minimum inhibitory concentration (MIC) of ≥16 μg/mL (Table 1).

The MIC ranges of PDP, tiamulin and valnemulin against MRSA clinical strains were 0.008–0.063, 0.125–1, and 0.031–0.25 μg/mL, respectively. The derivative PDP demonstrated significantly superior activity against MRSA clinical isolates compared to the control drugs tiamulin and valnemulin (Table 2).

To evaluate derivative PDP, we performed time-kill assays against MRSA at 1, 2, 4, 8 × MIC concentrations, using valnemulin as the reference drug. Both compounds showed concentration-dependent bactericidal activity (Figure 2). At concentrations of 1 × MIC and 2 × MIC, derivative PDP was observed to slow bacterial propagation. Specifically, in the 1×MIC treatment group, the bacterial concentration at 24 h remained higher than the initial inoculation concentration but was significantly lower than the bacterial concentration in the untreated control group. In the 2 × MIC treatment group, the bacterial concentration at 24 h was comparable to the initial inoculation concentration. At a concentration of 4 × MIC, derivative PDP exhibited a sterilizing effect, with the bacterial concentration at 24 h being lower than the initial inoculation concentration. When administered at 8×MIC, derivative PDP was able to almost completely eliminate MRSA within 24 h.

To further assess the risk of microbial resistance development to derivative PDP during long-term use (Li et al., 2021), an induced drug resistance assay was conducted using MRSA 43300 (Figure 3). Tiamulin exhibited progressive resistance development, with MIC values increasing from baseline to 2 μg/mL (8-fold) by day 6, 4 μg/mL (16-fold) by day 12, and reaching 16 μg/mL (64-fold increase) by day 20. In contrast, derivative PDP showed only minimal resistance emergence, with a final MIC of 0.125 μg/mL. The observed resistance progression indicates a significantly slower development of resistance to PDP derivative compared with tiamulin.

Derivative PDP exhibited excellent in vitro antibacterial activity. To further investigate its mechanism of action, we explored the binding mode of PDP to the 50S ribosomal subunit (PDB ID: 1XBP). As a reference, tiamulin was docked into the same target using Smina, and the resulting root mean square deviation (RMSD) value was <2.0 Å (Figure 4; Trott and Olson, 2010; Quiroga and Villarreal, 2016).

The structural diagram of the ligand-receptor complex demonstrates that both tiamulin and PDP bind exclusively to ribosomal RNA. In the structural representation, the 50S ribosomal subunit core is shown in gray cartoon format, with ligands and their hydrogen-bonding residues rendered as stick models. Hydrogen bonds were formed between the hydroxyl groups located on the eight-membered rings of the docked compounds and the G-2484 residues, and these interactions were identified as key binding forces in the complex (Figures 4B,C). The ester carbonyl groups in the side chains of tiamulin and PDP form hydrogen bonds with the G-2044 residue (Figures 4B,C). Additionally, the acylamide carbonyl and pyrrole group on the PDP side chain form hydrogen bonds with residue U-2564 and C-2046, respectively (Figure 4C). When flexibly docked into the PTC, derivative PDP exhibited a higher binding affinity (ΔGb = −10.7 kcal/mol) compared to tiamulin (ΔGb = −9.0 kcal/mol). This difference in binding affinity is consistent with their respective antibacterial activities.

Given the strong antibacterial activity of PDP, we proceeded to assess its impact on protein synthesis using modified Staphylococcus aureus strains that ectopically express a green fluorescent protein (GFP) (Liu et al., 2023). The objective was to observe the reduction in green fluorescence intensity, which indicates hindered protein synthesis. Following a 4-h co-incubation with the GFP strains, it was evident that 0.25 μg/mL PDP can effectively suppress bacterial GFP expression compared with the control (Figure 5). The average fluorescence intensity results showed that compared to the control group, the fluorescence intensity decreased by 30.29% in the tiamulin group and by 79.99% in the PDP group. Similar to our earlier findings with pleuromutilin derivatives, this result was obtained using the BCA method (Qi et al., 2025).

Cytotoxicity assays were performed to evaluate the toxicity of derivative PDP against the mammalian cell lines HepG2 and RAW264.7 using the CCK-8 assay (Figure 6A). At a concentration of 128 μg/mL, derivative PDP induced viabilities of 65.88% in HepG2 cells and 74.66% in RAW264.7 cells, respectively. The half-maximal inhibitory concentration (IC₅₀) values of derivative PDP were ≥128 μg/mL for both cell types. These results indicated that derivative PDP exhibited low cytotoxicity toward mammalian cells, demonstrating satisfactory biocompatibility.

In the acute oral toxicity study conducted per OECD Guideline 423, no mortality or clinical abnormalities were observed in rats during the 24-h post-administration period or subsequent 7-day observation. All animals maintained normal physiological parameters including food intake, grooming behavior, neurological status, and excretory functions. Histopathological examination showed no significant tissue alterations in cardiac, splenic, pulmonary, or renal specimens from either control or treated groups. The 2000 mg/kg PDP-treated group exhibited only mild hepatic steatosis (Figure 6B), with no other drug-related pathology. Based on these findings, the LD₅₀ of derivative PDP was determined to exceed 2000 mg/kg body weight.

In the MRSA systemic infection model, mice received tail vein injections of PDP derivative, tiamulin, or valnemulin (20 mg/kg). Treatment with PDP (20 mg/kg) resulted in a 20% survival rate, demonstrating superior efficacy to tiamulin (10% survival) and equivalent activity to valnemulin (20% survival) at the same dose (Figure 7). Compared with the model, PDP significantly delayed death and improved survival rate.

We subsequently quantified bacterial burdens in the lungs, liver, and kidneys of mice following 7-day treatment (Figure 8). At equivalent doses, derivative PDP exhibited a significant reduction in the mean bacterial load in the lungs (2.109 and 1.522 Log₁₀ CFU), kidneys (1.959 and 1.996 Log₁₀ CFU), and liver (1.245 and 1.289 Log₁₀ CFU) when compared to tiamulin and valnemulin, respectively (p < 0.0001). These results demonstrate that the PDP derivative effectively attenuates MRSA-induced pathological damage across pulmonary, renal and hepatic tissues.

Histopathological examination revealed alveolar cell deformation and necrosis with pronounced inflammatory infiltration in pulmonary tissues from tiamulin- and valnemulin-treated groups. Renal sections similarly exhibited marked tubular degeneration and necrosis accompanied by inflammatory cell accumulation. Compared with the control drug treatment group, PDP had milder lung tissue lesions and more severe liver lesions. In addition, no significant pathological changes were observed in the tissue of PDP kidneys (Figure 9).

In the early stages of compound research and development, computer software can simulate and predict the ADMET properties of drugs. This capability allows researchers to forecast compound characteristics based on their chemical structures and assess the feasibility of potential drug candidates. For the present study, the pharmacokinetic properties of derivative PDP were predicted using the Swiss ADME platform. Table 3 summarizes the key pharmacokinetic profiles. The predicted lipophilicity values (log P (o/w)) for derivative PDP, tiamulin, and valnemulin were 4.52, 4.69, and 4.22, respectively, suggesting limited water solubility for all three compounds. Pharmacokinetic predictions indicated that tiamulin displayed high gastrointestinal absorption, whereas the absorption of derivative PDP and valnemulin was relatively lower. None of the three compounds can cross the blood-brain barrier (BBB); but valnemulin and PDP are substrates of P-glycoproteins. Derivative PDP may exert an inhibitory effect on CYP3A4 but shows no inhibition of CYP1A2, CYP2C19, CYP2C9, or CYP2D6. This finding is consistent with the characteristics of tiamulin and valnemulin. The skin permeability coefficients (Kp) for derivative PDP, tiamulin, and valnemulin were determined to be −5.74, −5.30, and −6.0 cm/s, respectively. Additionally, all three compounds achieved identical bioavailability scores.

All reagents and solvents were purchased from Macklin Biochemical Co., Ltd., (Shanghai, China) and Sinopharm Chemical Reagent Limited Corporation. Reaction progress was monitored by thin-layer chromatography (TLC) using silica gel-coated plates, with visualization achieved using either ultraviolet (UV) light or phosphomolybdic acid staining. All pleuromutilin derivatives were purified via column chromatography using 200–300 mesh silica gel. Proton nuclear magnetic resonance (¹H NMR) and carbon-13 nuclear magnetic resonance (¹³C NMR) spectra were recorded on Bruker spectrometers operating at 400 MHz and 101 MHz, respectively. High-resolution mass spectra (HRMS) were acquired using an Agilent Technologies 6530 Precision Mass Q-TOF LC/MS system.

A mixture of pleuromutilin (5.0 g, 13.2 mmol) and p-toluenesulfonyl chloride (2.8 g, 14.5 mmol) in methyl tert-butyl ether (13.2 mL) was treated with the dropwise addition of 10 M sodium hydroxide solution (2.2 mL). The reaction mixture was stirred vigorously at 55 °C for 1 h. After the reaction, compound 2 was isolated by filtration, washed sequentially with water and methyl tert-butyl ether, and then dried in a desiccator. The product was obtained as a white solid (6.5 g, 93% yield).

A mixture of compound 2 (11.1 g, 20.8 mmol) and dimethylcystamine hydrochloride (9.0 g, 41.6 mmol) in tetrahydrofuran (80 mL) was treated with the dropwise addition of benzyltributylammonium chloride (0.65 g, 2.1 mmol) and 20% sodium hydroxide solution (15 mL). The reaction mixture was stirred at 50 °C for 4 h. After completion of the reaction, the products were extracted with dichloromethane. The organic extract was dried over anhydrous sodium sulfate, concentrated under reduced pressure, and subsequently purified by column chromatography using a solvent system consisting of petroleum ether, ethyl acetate, and diethylamine. Yellow powder, yield: 88.0%. ¹H NMR (400 MHz, CDCl₃) δ 6.47 (dd, J = 17.4, 11.0 Hz, 1H), 5.74 (d, J = 8.5 Hz, 1H), 5.33 (dd, J = 11.0, 1.5 Hz, 1H), 5.19 (dd, J = 17.4, 1.6 Hz, 1H), 3.34 (d, J = 6.5 Hz, 1H), 3.12 (d, J = 1.6 Hz, 2H), 2.59 (s, 2H), 2.39–2.26 (m, 1H), 2.27–2.15 (m, 2H), 2.13–2.01 (m, 2H), 1.76 (dd, J = 14.4, 3.1 Hz, 1H), 1.68–1.62 (m, 2H), 1.57–1.48 (m, 4H), 1.45 (s, 3H), 1.43–1.37 (m, 1H), 1.31 (d, J = 16.1 Hz, 1H), 1.23 (s, 6H), 1.16 (s, 3H), 1.10 (dd, J = 13.9, 4.4 Hz, 1H), 0.86 (d, J = 7.0 Hz, 3H), 0.72 (d, J = 6.9 Hz, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 217.30, 169.68, 139.29, 117.49, 74.86, 69.58, 58.47, 51.93, 48.80, 45.72, 44.99, 44.18, 42.05, 37.04, 36.27, 34.73, 31.51, 30.71, 29.95, 27.11, 26.59, 26.49, 25.11, 17.12, 15.19, 11.78. HRMS (ESI+): calcd for C₂₆H₄₃NO₄S [M + H]⁺, 466.2986; found, 466.2970.

Pyrrole-2-carboxylic acid (1.06 mmol) and compound 3 (0.49 g, 1.06 mmol) were dissolved in anhydrous dichloromethane (10 mL), followed by sequential addition of EDCI (0.24 g, 1.27 mmol), HOBt (0.17 g, 1.27 mmol), and N, N-diisopropylethylamine (DIPEA, 0.37 mL, 2.12 mmol). The reaction mixture was stirred at room temperature for 12–18 h (monitored by TLC). Sodium bicarbonate solution was added to the reaction mixture after complete reaction. Then, the products were extracted with dichloromethane, dried over anhydrous sodium sulfate, concentrated, and purified by column chromatography (Petroleum ether, ethyl acetate and diethylamine). White solid, yield: 53.4%. ¹H NMR (400 MHz, CDCl₃) δ 9.79 (s, 1H), 6.97–6.90 (m, 1H), 6.81–6.73 (m, 1H), 6.46 (dd, J = 17.4, 11.0 Hz, 1H), 6.28–6.22 (m, 1H), 5.75 (d, J = 8.5 Hz, 1H), 5.25–5.07 (m, 1H), 3.46–3.33 (m, 2H), 3.29–3.11 (m, 3H), 2.36–2.29 (m, 1H), 2.28–2.16 (m, 2H), 2.14–2.06 (m, 2H), 1.77 (dd, J = 14.5, 3.1 Hz, 1H), 1.71–1.50 (m, 4H), 1.45 (s, 3H), 1.42–1.31 (m, 2H), 1.29 (d, J = 7.2 Hz, 6H), 1.19–1.07 (m, 4H), 0.87 (d, J = 7.0 Hz, 3H), 0.72 (d, J = 7.0 Hz, 3H). ¹³C NMR (101 MHz, CDCl₃) δ 217.19, 170.79, 161.37, 138.97, 126.34, 121.66, 117.62, 109.94, 109.57, 74.81, 70.43, 58.34, 47.98, 47.57, 45.68, 45.15, 44.24, 42.04, 36.93, 36.21, 34.70, 31.75, 30.65, 27.13, 26.63, 26.59, 26.53, 25.08, 17.21, 15.13, 11.81. HRMS (ESI+): calcd for C₃₁H₄₆N₂O₅S [M + H]⁺, 559.3200; found, 559.3199.

Molecular docking simulations were performed using Smina software (Trott and Olson, 2010; Quiroga and Villarreal, 2016). Ligand energy minimization was carried out using Chem3D. For the X-ray crystal structure (PDB ID: 1XBP), ligands, all water molecules, and metal ions were first removed. The binding pocket was defined with coordinates (x: 52.8, y: 122.3, z: 113.8) and a dimension of 40 × 50 × 40 Å. To validate the docking protocol, the co-crystal ligand tiamulin was redocked, and its conformation was compared with the original structure in the crystal. Finally, all docking conformations were visualized using PyMOL software.

Plasmids harboring erythromycin resistance genes and labeled with green fluorescent protein (GFP) were introduced into Staphylococcus aureus via electro-transformation, as described previously (Liu et al., 2023). Following transformation, erythromycin-based screening was performed to isolate successfully transformed bacterial colonies. The selected S. aureus strain was cultured until it reached the mid-logarithmic growth phase. Bacterial cells were then harvested by centrifugation, resuspended in phosphate-buffered saline (PBS), and co-incubated with compound PDP and tiamulin at final concentrations of 0.25 μg/mL and 2 μg/mL, respectively. This co-incubation was conducted at 37 °C for 4 h. Afterward, the cells were recollected, resuspended in PBS once more, and their fluorescence was observed using fluorescence microscopy.

Six-week-old SPF Kunming mice (weighing 25–30 g) were obtained from the Lanzhou Institute of Veterinary Medicine, Chinese Academy of Agricultural Sciences.

The mice were randomly divided into five groups (n = 10 per group) with balanced gender distribution. Control drugs (tiamulin and valnemulin) and compound PDP were dissolved in a vehicle solution containing 5% DMSO, 10% Tween 80, and 85% saline, followed by sterile filtration. Four days prior to inoculation, all mice received an intraperitoneal injection of cyclophosphamide at a dosage of 150 mg/kg. Subsequently, 1 day before inoculation, they were administered another intraperitoneal injection of cyclophosphamide at a reduced dosage of 100 mg/kg. Next, neutropenic mice were given an intraperitoneal injection of 0.56 mL MRSA inoculum with a concentration of 10⁸ CFU/mL. Half an hour later, those mice that had received the MRSA inoculum were injected with the compound at a dose of 20 mg/mL via the tail vein. The daily mortality of the mice was monitored over a 7-day period. After this period, the mice were anesthetized with an excess of carbon dioxide, and their lungs, livers, and kidneys were harvested.

Swiss ADME was employed to analyze the pharmacokinetic parameters of tiamulin, valnemulin, and PDP.

All animal experiments strictly adhered to the Animal Ethics Procedures and Guidelines of the People’s Republic of China. This research procedure received approval from the Animal Administration and Ethics Committee of Lanzhou Institute of Husbandry and Pharmaceutical Sciences of CAAS (Permit NO: SYXK-2023-006).

Statistical analyses were performed using IBM SPSS 24.0. One-way ANOVA with Dunnett’s post hoc testing was employed, with significance levels defined as p < 0.05 (significant) and p < 0.01 (highly significant).