Glycerol kinase 5 confers gefitinib resistance through SREBP1/SCD1 signaling pathway

This study was approved by the Research Ethics Committee of Zhongshan Hospital, Fudan University (Shanghai, China) and performed according to relevant guidelines and regulations. Written informed consent was obtained from all participating individuals. EDTA plasma samples from 17 individuals with lung adenocarcinoma, who were sensitive to EGFR TKIs, and 11 individuals with lung adenocarcinoma, who had acquired resistance to EGFR TKIs, admitted at the Department of Pulmonary Medicine, Zhongshan Hospital, Fudan University. The Invitrogen total exosome precipitation reagent (Thermo Fisher Scientific, MA, USA) was used to isolate the exosomes from plasma samples according to manufacturer’s instruction. The detection of exosomal GK5 mRNA, using tethered cationic lipoplex nanoparticles (TCLNs), was previously described [37, 38]. Cationic lipoplex nanoparticles, containing the GK5 molecular beacons (MBs, custom synthesized by Sigma-Aldrich, MO, USA), were tethered onto the glass slide surface by a biotin-avidin linkage. All the MBs were labeled with Fluorescein amidite (FAM). To detect the expression of GK5 mRNA in the exosomes, Total Internal Reflection Fluorescence (TIRF) Microscopy (Nikon, Japan) was used. A custom-developed MATLAB program was applied to analyze the images by generating a binary mask to remove the background and measure the sum intensity of each image. In this study, PBS controls were used to define the background. Any fluorescence signals from the samples that were equal or lower than the signals observed in the PBS controls were defined as background in the image analysis.

The human lung adenocarcinoma cell line PC9 and NCI-H1975 (H1975, intrinsically harbor EGFR L858R/T790 M) were purchased from the American Type Culture Collection (ATCC, Rockville, MD, USA). To induce gefitinib resistance, PC9 cells were exposed to increasing concentrations of gefitinib [39]. The gefitinib-sensitive PC9 and -resistance PC9R and H1975 cells were cultured in RPMI 1640 medium (Thermo Fisher Scientific, MA, USA) supplemented with 10% heat-inactivated fetal bovine serum and 100 U/ml penicillin/streptomycin. The gefitinib resistance in the PC9R and H1975 cells was maintained by adding 1 μΜ gefitinib (Selleckchem, TX, USA). The cells were grown as monolayers in a humidified atmosphere containing 5% CO₂ at 37 °C.

Short hairpin RNA (shRNA) vectors against the GK5 genes shGK5–1 and shGK5–2, and against the SCD1 genes shSCD1 were obtained from TRC (The RNAi Consortium). Lentiviral plasmids, containing GV112-shGK5–1, -shGK5–2, -shSCD1, and -negative, were obtained from GeneChem (Shanghai, China). Lentiviruses carrying overexpressing human EGFR (GenBank accession number NM_005228) lentiviral vectors (GV358) were from GeneChem. Lentiviruses carrying overexpressing human GK5 (GenBank accession number NM_001039547.2) and SCD1 (GenBank accession number AB032261.1) lentiviral vectors (Lv105) were from Genecopoeia (Guangzhou, China). The lentiviral particles were produced by transfecting the HEK 293 T cells with the lentiviral plasmids. For viral infection, PC9R or H1975 cells were plated in 6-well plates, grown to 50–70% confluence, incubated with medium containing virus and 4 μg/mL polybrene at an MOI (multiplicity of infection) of 10, and were then incubated with various concentrations of gefitinib after 24 h.

The Cell Counting Kit-8 (CCK-8; Dojindo Laboratories, Japan) was used to assess the rate of cell proliferation. Briefly, cells were plated in 96-well plates at approximately 1000 cells per well with 200 μL of culture medium. After 24 h, 10 μl of CCK8 solution was applied to each well, and the plates were incubated for 1 h at 37 °C. Finally, the absorbance values at 450 nm were determined using a microplate reader (Multiskan, Thermo Fisher Scientific, MA, USA) with a reference wavelength of 650 nm. All experiments were conducted at least in triplicate.

Cells were incubated with 10 μM EdU (5-ethynyl-2′-deoxyuridine, Thermo Fisher Scientific, MA, USA) for 4 h and were then fixed with 3.7% formaldehyde in PBS for 15 min at room temperature. The EdU was detected for EdU incorporation according to manufacturer’s recommendations. Confocal imaging was performed on a Nikon A1R confocal laser scanning microscope system (Nikon Corp., Tokyo, Japan). PC9R cells positive for EdU incorporation and positive for Hoechst 33342 staining were counted using Image J (v. 1.42, Wayne Rasband, NIH) and were used to calculate the percentage of EdU-positive cells.

An annexin-V-allophycocyanin (APC) and 4′,6-diamidino-2-phenylindole (DAPI) double staining kit (Thermo Fisher Scientific) was used to evaluate apoptosis. The transfected PC9R cells were seeded in 6-well plates (5 × 10⁵ cells/well) and were treated with 1 μM gefitinib. Cells were then digested with trypsin (Gibco® Trypsin-EDTA, Thermo Fisher Scientific, MA, US), washed with PBS three times, suspended in 500 μl of binding buffer, and were then incubated with 5 μl APC-conjugated annexin-V and 3 μl DAPI for 15 min at room temperature in the dark. The stained cells were detected using the BD FACS Aria II flow cytometer (BD biosciences, CA, USA).

The transfected PC9R cells were seeded in 6-well plates (5 × 10⁵ cells/well) and were treated with 1 μM gefitinib. The treated cells were then collected, washed with PBS, and fixed in 70% ethanol for 24 h at 4 °C. The fixed cells were stained with PI in the dark for 30 min at room temperature. Finally, cell cycle distribution was analyzed by flow cytometry.

Total RNA was extracted using the TRIzol reagent (Thermo Fisher Scientific, MA, USA), and cDNA was synthesized using reverse transcriptase (TOYOBO, Japan). The RNA (1%) was reverse transcribed into complementary deoxyribonucleic acid, and 20 ng of complementary DNA was used as the template for RT-PCR. The amplification cycling parameters (40 cycles) were set as follows: 15 s at 95 °C; 15 s at 60 °C and 45 s at 72 °C. The primer sequences included the following:

GK5 sense 5’-TGAGGGACACAAGCCACAAT-3′,

GK5 anti-sense 5’-GGAAGCAGCACTCTCCAAAC-3′.

SCD1 sense 5’-GTACCGCTGGCACATCAACTT-3′,

SCD1 anti-sense 5’-TTGGAGACTTTCTTCCGGTCAT-3′.

β-actin sense 5’-CTGGCACCCAGCACAATG-3′,

β-actin anti-sense 5’-CCGATCCACACGGAGTACTTG-3′.

The gene expressions were normalized to β-actin and were measured by 2^-ΔΔCT. The RT-PCR assay was performed at least 3 separate times in triplicate.

Total protein was extracted using a RIPA kit (Beyotime Biotechnology, China), separated on polyacrylamide gels, and transferred to PVDF membranes. The membranes were incubated with anti-GK5 (Proteintech Group, IL, USA), anti-EGFR (Cell Signaling Technology (CST), MA, USA), anti-survivin (CST), anti-PLK1 (CST), anti-Bcl2 (CST), anti- cleaved caspase-3 (Asp175, CST), anti-caspase-3 (CST), anti- cleaved PARP (Asp214, CST), anti-PARP (CST), anti-SCD1 (Abcam, MA, USA), anti-SREBP1 (Abcam), and anti-actin (CST) antibodies at 4 °C overnight and were then incubated with horseradish peroxidase-conjugated goat anti-rabbit or anti-mouse immunoglobulin G at room temperature for 1 h. Proteins were visualized using Pierce ECL Western blotting substrate and autoradiography. The blots were analyzed using Quantity One 4.6.

Total RNA was extracted from the negative control cells and the PC9R cells transfected with shGK5–1 using the RNeasy Plus Mini Kit (Qiagen, MD, USA). The extracted RNA was converted to double-stranded cDNA and was amplified using a OneArray plus RNA amplification kit (Phalanx Biotech Group, Taiwan). The Cy5-labeled RNA targets were hybridized to the Human Whole Genome OneArray (Phalanx Biotech Group, Taiwan). The intensity of signals was measured by the Agilent Microarray Scanner (G2505C, Agilent, USA) and was analyzed by the Rosetta Resolver System (Rosetta Biosoftware). Testing was performed in triplicate, and the statistical analyses were performed using the proprietary modeling techniques from the Rosetta Resolver.

The animal experiments were approved by the Institutional Animal Care and Use Committee at Zhongshan Hospital from Fudan University, China. Twenty four- to 6-week-old male BALB/c nude mice were obtained from the Shanghai Experimental Animal Center of the Chinese Academy of Sciences (Shanghai, China). 1 × 10⁷ PC9R cells transfected with shNEG or shGK5–1 were subcutaneously injected into the right flanks of ten nude mice. The tumor volume was monitored and was calculated according to the following formula: Volume = length x width²/2. After one month, tumors were dissected out and weighed for quantification. Ki67 immunohistochemistry staining was applied to identify the proliferating cells in the paraffin sections of the xenograft tumors. The Ki67-positive cells were quantified in randomly selected fields from each tissue section using Image J.

To understand the association of GK5 expression and EGFR TKI resistance, we collected EDTA plasma samples from 28 lung adenocarcinoma patients, of which 17 and 11 were sensitive and resistant to EGFR-TKIs, respectively (Table 1). There were no differences in the clinical characteristics of these patients. Analysis of exosomal GK5 mRNA by a tethered cationic lipoplex nanoparticle based assay showed that the level of GK5 mRNA in plasma exosomes was downregulated in EGFR TKI-sensitive patients compared to patients who are TKI-resistant (Fig. 1a, b). To further understand the mechanism of gefitinib resistance, we generated gefitinib-resistant PC9R cells by exposing gefitinib-sensitive PC9 cells to increasing concentrations of gefitinib [39]. We found that in the gefitinib-resistant PC9R cells, GK5 was significant upregulated compared to that in gefitinib-sensitive PC9 by quantitative reverse transcription qRT-PCR and Western blotting analysis (Fig. 1c, d).

To evaluate the function of GK5 in gefitinib resistance, we constructed a lentivirus vector carrying the complete open reading frame of GK5 that allows GK5 overexpression in gefitinib-sensitive PC9 cells. After confirming GK5 overexpression in lentivirus-infected gefitinib-sensitive PC9 cells by qRT-PCR (Fig. 1e), we evaluated the viability and apoptosis of GK5-overexpressing and control cells by CCK8 assay and annexin V-APC/DAPI double staining followed by flow cytometry analysis, after treatment with gefitinib at concentrations ranging from 0 to 0.1 μM. GK5 induced drug resistance in gefitinib-sensitive PC9 cells (Fig. 1f, g), confirming that GK5 expression is sufficient to induce gefitinib resistance in cancer cells.

To further confirm the relationship between gefitinib-resistance and GK5 expression, gefitinib-resistant PC9R cells were transfected with either GK5 short hairpin (sh) RNA or negative control shRNA to silence GK5 expression. The qRT-PCR and Western blot analyses showed that shGK5–1 and − 2 significantly inhibited GK5 expressions (Fig. 1h, i). Morphological examination showed that GK5 knockdown reduced the number of PC9R cells transfected with shGK5 compared to those transfected with negative control shRNA for 24 h and then treated with 1 μM gefitinib for 72 h (Fig. 1j). The CCK8 assays revealed that transfection of PC9R cells with either shGK5–1 or − 2 enhanced gefitinib-induced apoptosis, especially at 10.0 μM gefitinib, increasing the percentage of apoptotic cells from 45 to 68 or 64%, respectively (Fig. 1k).

To verify the relationship between gefitinib-resistance and GK5 expression, gefitinib-resistant H1975 cells, which contain the exon 20 T790 M mutation, were used. To validate gefitinib resistance in H1975 cells, the viability of PC9 and H1975 cells upon gefitinib treatment was measured by CCK-8 assay. It was found that the viability of the PC9 cells was dramatically reduced in the presence of 0.1 μM gefitinib, whereas the viability of the H1975 cells was not significantly decreased by up to 1 μM gefitinib treatment (Fig. 2a). Western blotting demonstrated that the protein level of GK5 was upregulated in gefitinib-resistant H1975 cells compared to PC9 cells (Fig. 2b). The qRT-PCR analyses showed that shGK5–1 and − 2 significantly inhibited GK5 expressions in H1975 cells (Fig. 2c). Morphological examination showed that GK5 knockdown reduced the number of H1975 cells transfected with shGK5 compared to those transfected with negative control shRNA for 24 h and then treated with 1 μM gefitinib for 72 h (Fig. 2d). The CCK8 assays revealed that transfection of H1975 cells with either shGK5–1 or − 2 enhanced gefitinib-induced apoptosis (Fig. 2e).

To confirm that GK5 knockdown suppresses the proliferation of gefitinib-resistant cells in the presence of gefitinib, we used 5-ethynyl-2′-deoxyuridine (EdU) staining to analyze the effect of GK5 silence on proliferation of gefitinib-resistant cells following gefitinib treatment. The fraction of EdU-positive cells was significantly lower in GK5-silenced cells compared with control cells, indicating a lower rate of cell proliferation in the absence of GK5 (Fig. 3a, b). We also analyzed the apoptosis by annexin-V-APC and DAPI double staining followed by flow cytometry. The results showed that cell apoptosis increased ~ 35% in PC9R cells transfected with shGK5 relative to negative control cells in the presence of 1 μM gefitinib (Fig. 3c, d). Analysis of cell cycle distribution by flow cytometry showed that GK5 silencing resulted in cell number increased in the G0/G1 phase and decreased in S phase in the presence of 1 μM gefitinib, indicating that cell cycle arrest was induced by GK5 knockdown (Fig. 3e, f).

We examined mitochondrial membrane potential of PC9R cells transfected with shGK5 or negative control shRNA. Using JC-1 staining, which emits red fluorescence as J-aggregates in intact mitochondria and green fluorescence in the monomeric form in damaged mitochondria. The JC-1 monomer ratio of PC9R cells transfected with shGK5 was higher than that of cells transfected with the negative control in the presence of 1 μM gefitinib, indicating that the loss of GK5 expression was associated with mitochondrial damage (Fig. 4a, b). Mitochondria dysfunction can accelerate caspase-3 activation, as evidenced by Western blotting showing increased amounts of cleaved caspase-3 and poly-ADP-ribose polymerase (PARP) in GK5-depleted cells compared to control cells (Fig. 4c).

To determine whether GK5 knockdown suppresses lung tumor growth in vivo, PC9R cells transfected with shGK5–1 or negative control shRNA were injected into nude mice. PC9R cells transfected with shGK5–1 or negative control shRNA grew into visible tumors one week after injection, although the size of tumors derived from GK5-silenced cells was significantly smaller than that of tumors from control cells (Fig. 5a). A similar trend in tumor mass was seen one month after injection (Fig. 5b, c). To assess the effects of GK5 knockdown on cell proliferation, Ki67 expression in sections of tumors was detected by immunohistochemistry. The percentage of Ki67-positive cells decreased in tumors derived from GK5-silenced cells compared with tumors from control cells (Fig. 5d, e).

To evaluate the molecular mechanism underlying GK5-mediated gefitinib resistance, we used a whole human cDNA array to identify alterations in GK5 signaling pathways. This assay showed that GK5 depletion altered the expression of metabolic, apoptosis and cell cycle-associated molecules (Fig. 6a). Western blotting confirmed that GK5 knockdown in gefitinib-resistant PC9R cells significantly repressed protein expression of SREBP1 and SCD1 (Fig. 6b). The expression of EGFR, Survivin, PLK1, and Bcl-2 were also repressed.

To confirm the relationship between GK5 and EGFR, we performed a rescue assay by overexpressing EGFR in PC9R cells. CCK8 assays showed that EGFR overexpression in PC9R cells partly rescued the apoptotic effects of GK5 depletion in PC9R cells after treatment with gefitinib (Fig. 6c). YM155, a potent survivin inhibitor, also decreased the proliferation of PC9R cells relative to the negative control, especially in cells transfected with shGK5, indicating that GK5 induces drug resistance by interfering survivin signaling (Fig. 6d).

Furthermore, we found that SCD1 was upregulated in GK5 overexpressed PC9 cells (Fig. 7a). We also found that SCD1 was significantly increased in the PC9R cells compared with PC9 cells (Fig. 7b). We constructed a lentivirus vector carrying the complete open reading frame of SCD1 that allows SCD1 overexpression in PC9 cells. After confirming SCD1 overexpression in lentivirus-infected gefitinib-sensitive PC9 cells by qRT-PCR and Western blot (Fig. 7c, d), we evaluated the viability of SCD1-overexpressing and control cells by CCK8 assay after treatment with gefitinib. We found that SCD1 induced drug resistance in gefitinib-sensitive PC9 cells (Fig. 7e), confirming that SCD1 expression is sufficient to induce gefitinib resistance in cancer cells.

To further confirm the relationship between gefitinib-resistance and SCD1, PC9R cells were transfected with either SCD1 short hairpin (sh) RNA or negative control shRNA to silence SCD1 expression. The qRT-PCR and Western blotting showed that shSCD1 significantly inhibited SCD1 expressions (Fig. 7f, g). Morphological examination showed that SCD1 knockdown reduced the number of PC9R cells transfected with shSCD1 compared to those transfected with negative control shRNA for 24 h and then treated with 1 μM gefitinib for 72 h (Fig. 7h). To verify this, the CCK8 assay was performed and revealed that transfection of PC9R cells with shSCD1 enhanced gefitinib-induced apoptosis (Fig. 7i). Furthermore, A939572, a specific SCD1 inhibitor, decreased the proliferation of PC9R cells in a dose-dependent manner (Fig. 7j).

Then we constructed a lentivirus vector carrying the complete open reading frame of SCD1 that allows SCD1 overexpression in PC9R cells. After confirming SCD1 overexpression in lentivirus-infected gefitinib-sensitive PC9R cells by qRT-PCR and Western blot (Fig. 7k, l), a rescue assay was performed in the SCD1-overexpressed PC9R cells. CCK8 assays showed that SCD1 overexpression in PC9R cells transfected with shGK5 cells rescued PC9R cell viability in the presence of gefitinib (Fig. 7m). These results suggest that SCD1 overexpression decreased cell apoptosis induced by GK5 depletion.