Materials and chemicals

PLGA-COOH (Shyuanye, Shanghai, China), dichloromethane (SCR, Shanghai, China), PVA (Sigma, USA), EDC (Ourchem, Shanghai, China), NHS (Shyuanye, Shanghai, China), PEI (1800 k) (Rhawn, Shanghai, China), RPMI-1640 medium and foetal bovine serum were purchased from GIBCO BRL (Grand Island, USA), Penicillin–Streptomycin Solution (C0222; Beyotime, Shanghai, China), Gefitinib (ZD1839; Selleck, USA), Lipo6000 Transfection Reagent(C0526FT; Beyotime, Shanghai, China), Cell Counting Kit-8(Yeasen, Shanghai, China), Hoechst 33342 (Solarbio, Beijing, China), Trizol Reagent Kit, Prime Script RT reagent Kit, 2× SYBR Green Pro Taq HS Premix reagent, ROX Reference Dye, SYBR Green Premix Pro Taq HS qPCR Kit and Evo M-MLV RT Kit with gDNA Clean for qPCR (Accurate Biotechnology, Hunan, China), RIPA Lysis Buffer(P0013B; Beyotime, Shanghai, China), Anti-PDLIM5/ENH antibody ab196559 (abcam, USA), GAPDH Rabbit Polyclonal antibody (Proteintech, USA), Peroxidase-Conjugated Goat anti-Rabbit IgG (H + L) (ORIGENE, Beijing, China), LC3B(D11)XP Rabbit mAb and SQSTM1/p62(D5E2) Rabbit mAb purchased from Cell Signaling (Massachusetts, USA).

Some siRNAs targeted gene PDLIM5 were designed from GenePharma (Shanghai, China), including PDLIM5-Homo-1782 (PDLIM5-1), PDLIM5-Home- 1375 (PDLIM5-2), PDLIM5-Homo-1048 (PDLIM5-3), and FAM labeled PDLIM5-Homo-1048, serial numbers shown in Table 1.

Table 1 Sequences of PDLIM5 siRNAsBioinformatic analysis of PDLIM5

The RNAseq data (level 3) and corresponding clinical information for NSCLC were obtained from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.com). Log rank was used to compare the Kaplan–Meier survival analysis differences between normal group and NSCLC group. For the Kaplan–Meier curve, P values and hazard ratio (HR) with a 95% confidence interval (CI) were obtained using the log rank test and univariate Cox regression. The above analyses were performed using the v 4.0.3 version of the R software (R Foundation for Statistical Computing, 2020). P < 0.05 was considered statistically significant.

Cells and culture conditions

Tyrosine kinase inhibitor (TKI)-sensitive LUAD cell line PC9 was purchased from Cellcook (Guangdong, China), and gefitinib-resistant LUAD cell line PC9GR was purchased from FUHENG BIOLOGY (Shanghai, China). PC9GR cells were routinely cultured in RPMI-1640 medium supplemented with 10% foetal bovine serum and 1% Penicillin–Streptomycin solution and incubated in 5% CO2 at 37 °C. Gefitinib was added to the culture medium at a concentration of 2 μmol/l to sustain the drug-resistance phenotype of PC9GR cells.

Selection of siRNA and the expression of PDLIM5 protein in PC9 and PC9GR cell lines

Specific ON-TARGET siRNAs were designed to silence PDLIM5. The sequences of PDLIM5 are listed in Table 1. The effects of PDLIM5 gene silencing were identified using western blot. ON-TARGET siRNAs were transfected into PC9GR cells using the Lipo6000 Transfection Reagent. The PC9GR cell line was divided into four groups: experimental groups were transfected with siRNAs targeted PDLIM5 (PDLIM5-1, PDLIM5-2 and PDLIM5-3) and the control group was without any interference.

Moreover, PDLIM5 protein expression levels in PC9 and PC9GR cells were compared to explore the correlation between PDLIM5 and gefitinib resistance in PC9 cells.

Formulation and synthesis of nanoparticles (NPs)

Briefly, 100 mg PLGA-COOH was directly dissolved in 4 ml dichloromethane. This mixture was then added into 20 ml 2% precooled PVA aqueous solution and homogenized. Under sonication, dichloromethane was removed. After high-speed centrifugation and DE precipitation, EDC, NHS, Milli-Q water and PEI were mixed and stirred. The NPs were collected via centrifugation and washed thrice with distilled water at the same parameters.

The determination of the encapsulation efficiency of nanobubbles (NBs)

The siRNA solution was fully mixed with PLGA-NBs solution (1 mg/ml) in a 1:4 ratio. After centrifugation, phosphate-buffered solution and NBs solution were added to the precipitate. Subsequently, NBs carrying siRNA were obtained.

A spectrophotometer was used to evaluate the efficiency of the combination of siRNA and NBs. To evaluate the binding of siRNA to NBs during the above ligation process, unbound PDLIM5 siRNA in the above suspension was determined using a spectrophotometer (Thermo ND2000, Thermo Science Company, USA). The experiment was performed in triplicates. Encapsulation efficiency of NBs (EE, %) = (total amount of siRNA in NB suspension-the total amount of unbound siRNA)/total amount of siRNA in NB suspension.

Material characterisation of NPsObservation of particle size, polymer dispersity index (PDI), zeta potentials and NP morphology

A Malvern Nano ZS detector (Malvern Instruments, Malvern, England) was used to measure the particle size, PDI and zeta potential of NPs according to the principle of DLS. The reported NP formulation values are presented as mean ± standard deviation of the lowest three individual measurements for each sample.

The NP morphology was investigated using TEM and SEM. For TEM, the NP suspension was deposited on a copper grid and dried in a desiccator. Subsequently, the morphology of each group of the NPs was observed under HT7800 TEM (Hitachi, Japan), and the images were acquired.

Stability and US sensitivity tests

To compare the stability of siRNA-NBs and NBs with traditional SonoVue Microbubbles (MBs), in vitro contrast enhanced US imaging experiments were performed using a custom made 3% (w/v) agarose mould. A total of 1 ml NBs or siRNA-NBs with a bubble concentration of 3 × 106 bubbles/ml was added to the sample wells. A high-frequency linear transducer of the clinical US scanner system (Philips EPIQ7, Philips, USA) was set to a frequency of 20 MHz, an ultrasound intensity of 4% and a gain of 30 dB. Three ultrasonic images were recorded for each sample in the initial and at 40 min for the off line gray-scale intensity examinations using the ImageJ software.

Cytotoxicity analysis of NPs

Cell viability was assessed using the Cell Counting Kit-8 (CCK8) assay (Yeasen, Shanghai, China), following the manufacturer’s suggestions. Briefly, the PC9GR cells transfected with NB-siRNA (untreated PC9GR cells were included as negative controls) were cultivated in five 96-well plates with six replicate wells, followed by incubation in a humidified incubator for 3 h. A total of 10 μl CCK-8 solution was added to each well, incubated for 2 h and assayed using a microplate reader at 450 nm.

Transfection efficiency measurements

To investigate the transfection efficiency of the NPs, confocal laser microscope observation (CLSM, Zeiss LSM880, Carl Zeiss, Germany) was performed. Briefly, PC9GR cells were seeded on a confocal dish. The experimental group cells were transfected with siRNA, siRNA-NBs, siRNA + US and siRNA-NBs + US. The following conditions were applied for UTMD: ultrasound intensity: 500 Mw/DM2; duty cycle: 20%; pulse rate1000 Hz, irradiation duration of 90 s. All siRNA were FAM-labelled PDLIM5 siRNA. Untreated PC9GR cells were included as negative controls. The nucleus was stained with Hoechst 33342 and cells were observed using a confocal laser scanning microscope.

Effect of autophagy in PC9GR cells

The effect of autophagy in PC9GR cells was analysed via autophagosome formation observed using TEM (HT7800, Japan), and autophagy-related proteins (p62 and LC3-II/I) were measured using western blot.

Molecular assaysReal-time polymerase chain reaction (PCR) assay for determining the mRNA level of PDLIM5

Total RNA was harvested from the cell using the Trizol Reagent Kit, according to the manufacturer’s protocol. RNA yield was determined using a NanoDrop spectrophotometer (Thermo Fisher, Waltham, USA). Real-time PCR was performed using TB Green qPCR Mix Plus (Accurate Biotechnology, Hunan, China) and the CFX96TM Real-time Detection System (Accurate Biotechnology, Hunan, China). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an endogenous reference. Data were analysed using the relative standard curve method, according to the manufacturer’s protocol. All data were normalised against GAPDH mRNA levels and expressed as fold increases relative to controls. The primer sequences of the tested genes are listed in Table 2.

Table 2 Primers used in this studyWestern blot analysis

Total protein was extracted and quantified using RIPA Lysis Buffer and bicine cholinic acid protein assay kit, respectively. Protein samples were separated on a 10% or 15% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to polyvinylidene difluoride (PVDF) membranes. After incubation with the blocking buffer, the membranes were incubated overnight at 4℃ with rabbit antibodies against PDLIM5 (1:1000 dilution), LC3 (1:1000 dilution) and P62 (1:1000 dilution). Simultaneously, the membranes were incubated with rabbit antibodies against GAPDH (1:25,000 dilution) as an internal standard for normalizing protein expressions. Peroxidase-Conjugated Goat anti-Rabbit IgG (H + L) (1:25,000 dilution) was used to amplify the signal. Protein signals were detected using a chemiluminescence system (Tanon 5200Uulti, Shanghai, China).

Statistical analyses

The data were statistically analysed using a one-factor analysis of variance (SPSS software, version 24.0, SPSS Inc. and GraphPad Prism 8). All data are expressed as the mean ± standard errors of the mean. P < 0.05 was considered statistically significant. All experiments were performed in triplicates.