AZM and LZD are both purchased by Beijing Solarbio Science & Technology Co., Ltd.(Beijing, China), and MRX-I is provided by Shanghai Micurx Pharmaceutical Co., Ltd. (Shanghai, China). Both drugs were dissolved in dimethyl sulfoxide (DMSO), and the drug solution was prepared according to the suggestions provided by the Institute of Clinical and Laboratory Standards (CLSI) (Woods et al., 2011). The standard strain of M. abscessus ATCC 19977 was cultured on solid Lowenstein-Jensen medium at 37°C for 4-6 days. The MICs of AZM, LZD and MRX-I were determined by adding drugs to M. abscessus cultured in 96-well plates according to the recommended CLSI broth microdilution method. The broth was diluted twice, and both the concentrations of AZM, LZD and MRX-I ranged from 0.5-256μg/mL. A bacterial inoculum with turbidity equivalent to 0.5 McFarland standard dilution of 1∶200 was prepared for each strain. The MIC of M. abscessus was determined after 3 days of culture at 37°C together with antibiotics. Thereafter, 70 μl of Alamar Blue solution (Sirotec, 20μL Alamar Blue+50μL 5% Tween 80) was added to each well, and then the plates were incubated for another 24 hours. The color change from blue to pink indicates bacterial growth (Cowman et al., 2016). MIC is defined as the lowest drug concentration that no color change, that is, the lowest concentration that can inhibit the visible growth of the test isolate. Explain the results of drug sensitivity test (DST) according to the breakpoint recommended by CLSI.

This study was approved by the Ethics Committee of Beijing Chest Hospital, a hospital affiliated with Capital Medical University (2021-020). The wild-type zebrafish AB strain was maintained through natural paired mating to generate zebrafish progeny that were raised in water at 28 °C. Meanwhile, the smooth (S) morphology M. abscessus standard strain (ATCC19977) was incubated for 5 to 7 days at 30 °C in Middlebrook 7H9 broth (Becton Dickinson) containing 10% OADC (Becton Dickinson) and 0.05% Tween 80 (Sigma-Aldrich). Mid-log-phase M. abscessus cultures were centrifuged then the pellets were washed and cells were resuspended in phosphate buffered saline (PBS) containing 0.05% Tween 80. Next, the bacterial suspension was homogenised and sonicated then the tube containing dispersed cells was left upright for 5 to 10 min while bacterial cells settled to the bottom of the tube. The bacteria were then collected and resuspended in a smaller volume of PBS and labelled with the green fluorescent dye DIO (Thermo Fisher Scientific, USA or Hill Technology Co., Ltd., China). Labelled bacteria were next introduced into wild-type zebrafish at 2 days post-fertilisation (dpf) by micro-intravenous injection of about 3.6 × 10³ colony-forming units (CFUs) into the tail of each zebrafish in order to establish a zebrafish M. abscessus infection model. Zebrafish were anesthetized using 3-aminobenzoic acid ethyl ester methanesulfonate (C₉H₁₁NO₂·CH₄O₃S, MESAB). The MESAB was prepared by mixing MESAB and Na₂HPO₄·12H₂O in a total mass ratio of 1:5 to make a 4 mg/mL solution, which was stored at 4°C. For use, it was diluted with standard dilution water, with a final anesthetic concentration of 0.16 mg/mL. Then the zebrafish were injected bacteria by intravenous microinjection with each fish receiving approximately 3.6×10³colony forming units (CFUs)of the transplant to establish the zebrafish model of M. abscessus infection (Guo et al., 2020; Nie et al., 2020).

Zebrafish collected at 3 dpf under the microscope were randomly allocated to wells of 6-well plates (30 zebrafish per well). Meanwhile, azithromycin (AZM), LZD and MRX-I were prepared in dimethylsulfoxide (DMSO, Shanghai Aladdin Biochemical Technology Co., Ltd.) as initial 20.0 mg/mL stock solutions that were stored at -20 °C. For MTC determinations, drugs were diluted in water then were added to wells containing zebrafish to generate dilutions in wells of 6-well plates at final concentrations for AZM (62.5 μg/mL, 125 μg/mL, 250 μg/mL, 500 μg/mL, 1000 μg/mL), for LZD (15.6 μg/mL, 31.2 μg/mL, 62.5 μg/mL, 125 μg/mL, 250 μg/mL) and for MRX-I (15.6 μg/mL, 31.2 μg/mL, 62.5 μg/mL, 125 μg/mL, 250 μg/mL). The blank control group (zebrafish without injected M. abscessus and without drug dosing) and the model group (the negative control group of zebrafish injected with M. abscessus without drug dosing) were prepared at the same time using the same methods used to prepare the abovementioned drug-treated samples. After plates were incubated for 48 h at 35 °C, MTCs were determined as based on the highest concentration of each drug that did not cause zebrafish death.

According to previously reported experimental procedures, 3 dpf zebrafish were selected under the microscope then were randomly assigned to 6-well plates (30 zebrafish/well). Initial AZM, LZD and MRX-I stock solutions were diluted in water to generate working solutions of AZM (62.5 μg/mL), of LZD (15.6 μg/mL) and of MRX-I (0.488 μg/mL, 0.977 μg/mL, 1.95 μg/mL, 3.91 μg/mL, 7.81 μg/mL, 15.6 μg/mL). The control group and the model group were concurrently set up. All control and experimental samples were prepared in the same final volume of 3 mL/well. After treatment of zebrafish at 35 °C for 48 h, 10 zebrafish were randomly selected from each experimental group to photograph. The zebrafish were anesthetized and transferred onto methyl cellulose using a Pasteur pipette. An electrically controlled, continuously variable magnification fluorescence microscope (AZ100, Nikon, Japan) equipped with a green fluorescence channel was used at a magnification of 30x. Image processing software (NIS-Elements D 3.20) was used to analyse and collect fluorescence-based data. The efficacy of each drug, as based on inhibition of M. abscessus growth in zebrafish, was evaluated as based on fluorescence intensity of zebrafish whole-body, head and heart tissues.

3 dpf zebrafish were selected under the microscope and randomly assigned to 50-mL beakers (cups) with 60 zebrafish (experimental group) in a 20-mL volume per cup. Initial AZM, LZD and MRX-I stock solutions were diluted to generate a 62.5-μg/mL AZM MTC solution, a 15.6-μg/mL LZD MTC solution and MRX-I solutions at concentrations of 0.488 μg/mL, 0.977 μg/mL, 1.95 μg/mL, 3.91 μg/mL, 7.81 μg/mL and 15.6 μg/mL (MTC). Concurrently, control group and model group samples were set up in 20-mL volumes in cups then all control and experimental group cups were incubated at 35 °C. Numbers of zebrafish deaths for all groups were recorded every day and dead zebrafish were removed daily. At experiment completion, data were statistically analysed to calculate survival rates of zebrafish in each experimental and control group.

SPSS 26.0 was used to statistically analyse all data obtained in this study. One-way ANOVA was used for intergroup comparisons of CFU counts and fluorescence intensity expressed as mean ± SE values. For comparisons between two groups, t or t’ tests were used to compare normally distributed data from independent samples and non-parametric tests were used to compare data with non-normal distributions. One-way ANOVA analysis was chosen to compare normally distributed data with equal variance among multiple groups; otherwise non-parametric tests were used. Linearity between MRX-I concentration and fluorescence intensity was assessed using a linear trend analysis method. Kaplan-Meier survival analysis was performed using the log-rank test to visualise survival rates of zebrafish treated with different MRX-I concentrations. Intergroup differences with p values of <0.05 were considered statistically significant.

MICs of AZM, LZD and MRX-I were determined according to their effects on growth of the M. abscessus standard strain in zebrafish. The MIC of MRX-I for inhibition of growth of the M. abscessus standard strain was 16 μg/mL, and the MICs of AZM and LZD were 0.5µg/mL and 8μg/mL, respectively.

Table 1 presents results related to efficacies of AZM, LZD and MRX-I against M. abscessus infection. The AZM MTC was 62.5 µg/mL and MTCs of LZD and MRX-I were both 15.6 µg/mL. Treatment of zebrafish with AZM at 2 times the MTC led to deterioration of zebrafish health; treatment with 8 times the MTC led to signs of cardiac congestion accompanied by body bending; treatment with 16 times the MTC led to zebrafish mortality approaching the rate of 33.33%. At the LZD MTC of 15.6 μg/mL, no notable zebrafish health differences were observed as compared to that of the model group, while LZD treatment of zebrafish at 2 times the MTC resulted in a dramatic increase in mortality rate to 23.33%. Similarly, treatment with MRX-I led to similar trends as those observed for LZD; at the MRX-I MTC of 15.6 μg/mL, no significant changes were observed in zebrafish phenotype as compared with that of the model group, while treatment with 2 times the MRX-I MTC led to a marked increase in mortality rate.

Fluorescence intensities of zebrafish whole-body, head and heart tissues of AZM MTC, LZD MTC and MRX-I MTC groups were lower than that of the model group ( Table 2 ). The LZD MTC group whole-body fluorescence intensity exceeded that of the MRX-I MTC group (509184 ± 23064 vs. 439040 ± 3647, p < 0.01). Comparisons of effects of different MRX-I concentrations revealed that the whole-body fluorescence intensity of the MRX-I MTC group was only lower than fluorescence intensities of the 1/32, 1/16 and 1/4 MTC MRX-I groups (439040 ± 3647 vs. 524203 ± 31487, P < 0.01; 439040 ± 3647 vs. 505230 ± 16923, p < 0.01; 439040 ± 3647 vs. 487036 ± 11374, P < 0.01, respectively), with zebrafish whole-body fluorescence intensity decreasing with increasing MRX-I concentration (p = 0.004). Zebrafish head fluorescence intensities were lower in the MRX-I MTC group than in the MRX-I 1/32 MTC group (74147 ± 2175 vs. 105710 ± 10573, P < 0.05), the MRX-I 1/8 MTC group (74147 ± 2175 vs. 93451 ± 5928, P < 0.01) and the MRX-I 1/4 MTC group (74147 ± 2175 vs. 83397 ± 3571, P < 0.05), while the head fluorescence intensity of the LZD MTC group was higher than that of the MRX-I MTC group (95996 ± 8054 vs. 74147 ± 2175, P < 0.05). As for whole-body fluorescence intensity, head fluorescence intensity decreased with increasing MRX-I concentration (p = 0.001). By contrast, the MRX-I MTC group zebrafish heart fluorescence intensity was lower than corresponding intensities of the MRX-I 1/16 MTC and 1/8 MTC groups (10011 ± 852 vs. 20258 ± 4171, P < 0.05; 10011 ± 852 vs. 14733 ± 1663, P < 0.05, respectively), with zebrafish cardiac fluorescence intensity increasing linearly with increasing MRX-I concentration (p = 0.005). Figure 1 shows zebrafish fluorescence intensity distributions for different tissues of zebrafish treated with different concentrations of the three drugs.Increased WX-081 concentration was associated with lower bacterial burdens in both whole-body, head and heart of zebrafish as determined by bacterial CFU ( Figure 2 ).

The survival rate of zebrafish in the MRX-I MTC group was 78.33%, a rate that was not significantly different from survival rates of MRX-I 1/16, 1/8, 1/4 and 1/2 MTC groups ( Table 3 ). In addition, the survival rate of the MRX-I MTC group was not significantly different from that of the AZM MTC group that was treated with 62.5 μg/mL (78.33% vs. 88.33%, p > 0.05) and that of the LZD MTC group treated with 15.6 μg/mL (78.88% vs. 76.67%, p > 0.05). As compared with the model group survival rate, no significant difference in LZD and MRX-I MTC (15.6 μg/mL) group survival rates were observed (66.67% vs. 76.67%, p > 0.05; 66.67% vs. 78.33%, p > 0.05, respectively), although MRX-I 1/32, 1/16 and 1/8 MTC group survival rates exceeded that of the model group (95.00% vs. 66.67%, p < 0.001; 90.00% vs. 66.67%, p < 0.01; 85.00% vs. 66.67%, p < 0.05, respectively). Figure 3 shows zebrafish survival curves for different concentrations of AZM, LZD and MRX-I.