Background
Non-small-cell lung cancer (NSCLC) accounts for approximately 85% of the total lung cancer cases and represents a major reason for cancer-related deaths [
1]. It is known that the molecular epidemiology of NSCLC involves druggable genetic alterations such as the mutations of EGFR, KRAS, BRAF, MET, and HER2 [
2]. At present, a low radiation dose chest CT scan is recommended as a golden screening method of lung cancer for screen eligible population [
3]. However, unfortunately, NSCLC is usually diagnosed at the metastatic phase, and the median survival is only 1 year [
4]. To our acknowledge, pemetrexed (Pem), a type of folate analogue metabolic inhibitor used for mammalian cells, can be toxic to cancer cells through interference with their new biosynthesis of nucleotides to induce cell apoptosis and has been approved to be used for patients with NSCLC [
5].
Histone deacetylase (HDAC) inhibitors are representative of an extensive series of targeted anti-tumor agents, among which some have been approved for clinical treatment of cancers including givinostat (ITF2357) [
6]. Intriguingly, it has been reported that ITF2357 can aid in enhancement of the cytotoxicity of Pem via induction of apoptosis as well as autophagy in NSCLC [
7]. Of note, overexpressed HDAC2 could lead to reduced cisplatin sensitivity of NSCLC cells induced by valproic acid [
8]. Rad51, identified as the key protein in the homologous recombination pathway, shares great association with therapy resistance and genomic stability [
9]. Of note, decreased expression of Rad51 due to astaxanthin could result in promotion of mitomycin C-induced cytotoxicity in human NSCLC cells [
10].
Interestingly, our bioinformatic analyses suggested that MicroRNA (miR)-130a-3p, retrieved from a NSCLC-related microarray dataset, could target Rad51. MiRs, small single stranded non-coding RNA molecules with the ability of gene regulation at the posttranscriptional level, have been highlighted to play an important role in NSCLC [
11]. A previous study has found that miR-130a-3p, was notably downregulated in patients with NSCLC relative to that in healthy individuals [
12]. Moreover, a prior study has pointed out the suppression of HDAC3 on miR-130a-3p in breast cancer, which led to augmented escape of cancer cells from immune surveillance [
13]. Considering all the above findings, we hence hypothesized in the current study that HDAC inhibitor ITF2357 was likely to regulate the chemoresistance of mutant (mut)-KRAS NSCLC to Pem, with the involvement of the miR-130a-3p/HDAC2/Rad51 axis.
Methods
Ethics statement
The study protocol was in line with the ethical guidelines of the Helsinki Declaration and approved by Institutional Ethics Review Committee of The First Affiliated Hospital of Nanchang University. The animal experiments in this study were conducted under the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health.
Clinical sample collection
Sixty patients diagnosed with stage I, II or IIIa NSCLC who underwent resection surgery in The First Affiliated Hospital of Nanchang University from January 2016 to December 2017 were enrolled in this study. Patients with a recent history of other cancers, or with recurrent or primary NSCLC, or who had received chemotherapy or radiotherapy before surgery were excluded. Participants had complete clinical and pathological follow-up data. Paired NSCLC and adjacent normal tissues obtained from the 60 patients were immediately frozen in liquid nitrogen for RNA and protein extraction. Sixty collected NSCLC tissue samples were made into formalin-fixed and paraffin-embedded NSCLC specimens to detect the protein expression of HDAC2 by immunohistochemistry. Follow-up monitoring of the 60 patients were conducted to record their survival.
Cell culture
Human lung cancer cell lines, namely, mut-KARS A549 cell line and wild-type (wt) H1299 cell line, and human embryonic kidney cell line 293 T were purchased from Shanghai Institute of Biological Sciences, Chinese Academy of Sciences (Shanghai, China). The human lung cell line BEAS-2B was purchased from ATCC (Manassas, VA). The cells were cultured in DMEM medium (Gibco, Carlsbad, CA) supplemented with 10% heat-inactivated FBS, penicillin (100 U/mL) and streptomycin (100 μg/mL) in a CO2 incubator with 5% humidity at 37℃.
A549 cells were repeatedly exposed to a single high concentration of Pem (Pem inhibition concentration is IC50). Pem was purchased from Eli Lilly and Company (Indianapolis, IN; A762406C). Pem-treated cells were then washed with PBS and cultured in fresh medium without Pem. Following that, surviving cells were cultured. A549 cells exposed to Pem recovered and showed logarithmic growth after 2 weeks. Cells growing exponentially were trypsinized and exposed to Pem again for 48 h, the resistance of which was detected discontinuously until the experimental requirements were met. After 5 months, a Pem-resistant cell line was established and named A549R. The A549R cells were cultured (5% CO2, 37 °) for 1 month and passaged. The resistance of A549R cell line was confirmed again. Finally, A549R cells were cultured in Pem-free medium for another month.
Construction of stably transduced cell line
After construction of lentiviral vectors carrying oe-NC and oe-Rad51, stable genetic A549 cell lines were obtained. The lentiviral vector was pLV-EGFP-N (overexpression vector, oe-), which was purchased from GenePharma (Shanghai, China). 293T cells were cultured in RPMI-1640 complete medium, and passaged every other day. The virus was collected, and then was added into A549R cells (1 × 108 TU/mL) for infection to obtain stably transduced cell lines.
miR-130a-3p mimic: 5'-CAGTGCAATGTTAAAAGGGCAT-3' (sense); mimic NC: 5'-GGTAACAATATCGGGTCAAGAT-3' (sense).
RNA quantification
Trizol reagent (Invitrogen, Carlsbad, CA) and PrimeScript RT Master Mix (Takara Bio Inc., Otsu, Shiga, Japan) were utilized to extract total RNA from cells and tissues. Then, mRNA was reversely transcribed into cDNA; A Poly(A) tailing reverse transcription kit (B532451-0010, Sangon, Shanghai, China) was utilized to reversely transcribed miRNA into cDNA. All primers were synthesized by GenePharma, as listed in Additional file
2: Table S1. β-actin was utilized as an internal reference for mRNA normalization and U6 was used for miR normalization. 2
−ΔΔCt method was employed to quantify the relative expression level of target genes.
Western blot assay
Tissue or cells were lysed with high-efficiency radioimmunoprecipitation assay (RIPA) lysis buffer (R0010, Solarbio, Beijing, China) to extract total protein, followed by measurement of protein concentration utilizing BCA kits (20201ES76, Yeasen, Shanghai, China). With the protein concentration adjusted to 1 μg/μL, the protein lysate was directly loaded to 10–12% SDS–polyacrylamide gel electrophoresis and transferred onto a PVDF membrane to carry out western blot (0.2 μM, Bio-Rad Laboratories, Hercules, CA). Cells were sealed with 5% skim milk for 1 h, and then incubated with the primary antibodies (Rad51; 1:500, sc-398587, Santa Cruz Biotechnology, Santa Cruz, CA), HDAC2 (1:1000; 5113, Cell Signaling Technology, Beverly, MA), and β-actin (1:2000, sc-8432, Santa Cruz Biotechnology) at 4 ℃ overnight. Afterwards, the membrane was incubated for 1 h with horseradish peroxidase (HRP)-labeled goat anti-rabbit against IgG (ab205718, 1:2000, Abcam, Cambridge, MA) or goat anti-mouse diluent (ab6789, 1:5000, Abcam) at room temperature. Immobilon Western chemiluminescence HRP substrate (Millipore, Billerica, MA) and Bio-Rad Chemi Doc MP were used to observe Western blots. Total protein and cytoplasmic protein were determined using β-actin as internal reference, and the ratio of gray value of target band and internal reference band indicated relative protein expression level, as analyzed by ImageJ 6.0 software (National Institutes of Health, Bethesda, MD).
MTT assay
NSCLC cells were seeded in a 96-well plate (3 × 104 cells/well), and cell viability was measured by MTT assay. Cells were treated with different concentrations of HDAC inhibitor ITF2357 or Pem. After incubation for 24 h, 10 μL MTT solution (5 mg/mL, GD-Y1317, Guduo Biotechnology, Shanghai, China) was added to each well of the 96-well plate, and then incubated in an incubator for 4 h, and the supernatant was subsequently discarded. Then, 100 μL dimethyl sulfoxide was added to each well to make methyl Zan crystal dissolve completely. The 96-well plate was placed in a microplate reader, and the absorbance at 570 nm and 630 nm were measured.
Transwell assay
Matrigel was used for invasion detection but not for migration assay. (Corelle, New York). Matrigel was added to the apical chamber of the 24-well Transwell plate (8 μm pore size), which was incubated in an incubator at 4℃ for 30 min to polymerize Matrigel into gel. Basement membrane hydration was performed before use. NSCLC cells were cultured in serum-free medium for 12 h. The cells were harvested and re-suspended in serum-free medium (105 cells) and seeded in the apical chamber. NSCLC cells (2 × 104 cells/well) were seeded in the basolateral chamber of the Transwell plate. After 24 h of incubation at 37 ℃, the invasive cells were stained with crystal violet for 15 min and photographed under an inverted light microscope (Carl Zeiss MicroImaging, Inc., Thornwood, NY). The invasive and migratory abilities of NSCLC cells were counted and analyzed by imagine J software.
Radioimmunoprecipitation (RIP)
RIP Kits (17-701, Millipore) were used to detect the binding of Rad51 to Ago2 protein according to the instructions of the kits. When the cell confluence reached 80–90%, the medium was discarded. Cells were lysed with RIPA lysis buffer (P0013B, Beyotime, Shanghai, China) for 5 min, and centrifuged at 4 ℃ for 10 min at 14,000 rpm, with the supernatant subsequently obtained. One part of the cell extract was taken out as input, and the other part was incubated with antibody for coprecipitation. The specific steps were as follows: 50 μl magnetic beads were washed and then resuspended in 100 μL RIP wash buffer, and 5 μg antibody was added according to the experimental grouping for binding purpose. The magnetic bead-antibody complex was resuspended in 900 μL RIP wave buffer and incubated with 100 μL cell extract at 4 ℃ overnight.. The complex was then harvested, and samples and inputs were detached with proteinase K to extract RNA for subsequent qPCR detection of Rad51. Rabbit anti-mouse against Ago2 (1:100, ab32381, Abcam) and goat anti-mouse against IgG (1:100, ab205719, Abcam) were as negative control (NC).
RNA pull-down assay
The cells were transfected with biotin-labeled Bio-NC, Bio-Rad51-wt and Bio-Rad51-mut RNA (50 nM each). After 48 h of transfection, the cells were collected and washed with PBS. Then, the cells were incubated with specific cell lysate (20164, Thermo Fisher Scientific Inc., Waltham, MA) for 10 min. After that, 50 mL sample cell lysate was separately packed. The residual lysate was incubated at 4℃ for 3 h with M-280 streptavidin magnetic beads pre-coated with RNase-free and yeast tRNA (R8759, Sigma-Aldrich, St Louis MO), followed by two rinses with cold lysate, three rinses with low-salt buffer, and one rinse with high-salt buffer. The binding RNA was purified by Trizol, and then the enrichment of miR-130a-3p was detected by RT-qPCR.
Dual luciferase reporter gene assay
The targeting relationship between miR-130a-3p and Rad51 was verified using biological prediction website and luciferase reporter gene assay. The binding sites of miR-145 and Rad51 were analyzed, and the fragment sequence containing the action site was obtained. The 3’UTR region of Rad51 and the sequences after site-directed mutagenesis of miR-130a-3p binding site were cloned into the target sequences of psiCheck2 plasmid downstream of luciferase reporter gene (Hanbio Biotechnology, Shanghai, China). They were named Rad51-mut and Rad51-wt respectively. The luciferase activity was measured using luciferase assay kits (Promega, Madison, WI). After incubation for 48 h, the cells were lysed and the luciferase activity of firefly was measured using the dual luciferase system (Promega).
Chromatin immunoprecipitation (ChIP)
ChIP assay kits (Millipore) were used as follows: A total of 1 × 107 cells were fixed with 1% formalin for 10 min to make DNA and protein cross-linked. After that, the cells were subjected to SDS lysis, and the DNA was randomly sonicated into 500–1000 bp fragments. After 12,000 g centrifugation at 4 ℃, the supernatant was collected and placed into two tubes, which were respectively mixed with the specific antibodies of the target proteins. Pol II (05–623, 10 mL/mg total protein, Millipore), AcH4 (06-866, 10 mL/mg total protein, Millipore), HDAC2 (17-10237, 5 mL/mg total protein, Millipore) and IgG in the control group. For AcH4 ChIP, sodium butyrate (20 mM, 19-137, Millipore) was added to all solutions to maintain histone acetylation. ChIP DNA was purified and eluted with 100 μL H2O, and 2.5 μL ChIP DNA was extracted for qPCR detection. Primers: RNA Pol II: 5ʹ-CAGGGACTGGGAGAAGGA-3ʹ (forward), 5ʹ-CACTGCTAGTGACAGGTGCA-3ʹ (reverse); AcH4 and HDAC2: 5ʹ-GCCCCATCCCCTGCTGCT-3ʹ (forward), 5ʹ-CAGGCCCAGCGACTCACC-3ʹ (reverse).
Flow cytometry
After 48 h of transfection, cells were detached with 0.25% trypsin (without EDTA) and collected, followed by centrifugation and removal of the supernatant. Based on the instructions of Annexin-V-fluorescein isothiocyanate (FITC) cell apoptosis detection kit (556547, Shanghai Shuojia Biotechnology Co., Ltd., Shanghai, China), Annexin-V-FITC, PI and HEPES buffer solution were prepared into Annexin-V-FITC/PI staining solution at the ratio of 1:2:50. Next, per 100 μL of dye solution was used to resuspend 1 × 106 cells, followed by 15-min incubation with 1 mL HEPES buffer solution. The 515 nm and 620 nm band-pass filters were stimulated at 488 nm wavelength to detect FITC and PI fluorescence.
Xenografts in nude mice
Female athymic nude mice (Envigo, Shanghai, China; 5–6 weeks old) were raised in a SPF animal room (25 ℃, 70% humidity) under 12-h light/dark alternation. About 5 × 106 H1299, A549 or A549R cells were suspended in 100 μL PBS containing 20% Matrigel and injected subcutaneously into the right flank of each mouse. The tumor size was measured weekly with a digital caliper and calculated by the following formula: tumor size = a × b2/2, where a represents the larger size and b represents the smaller size. When the average tumor volume of mice was about 100 mm3, A549R tumor-bearing mice were randomly into four groups (5 in per group) and accordingly injected with (1) PBS, (2) ITF2357 (HDAC inhibitor, 50 mg/kg), (3) Pem (100 mg/kg) and (4) ITF2357 + Pem (sequential injection of Pem and then of ITF2357) every other day via tail vein. The injection volume of each was 200 μL. The tumor volume was closely monitored with time. After 20 days, the mice were euthanized, and tumor xenografts were collected, weighed and analyzed.
TUNEL staining
The experiment was carried out following the protocols of apoptosis detection kit (40306ES60/40308ES60, Yeasen Company). The slices were treated with 20% normal bovine serum for 30 min at room temperature, and then incubated with 50 μL of TUNEL reaction mixture at 37 ℃ for 90 min (negative pairs were not added with TUNEL reaction mixture). Next, the slices were treated with 3% H2O2 methanol solution for 10 min at room temperature, and then incubated with horseradish peroxidase (POD) solution (50 μL/tablet) at 37 ℃ for 30 min. Following diaminobenzidine/hydrogen peroxide (DAB/H2O2) color development, the slices were stained with hematoxylin and then photographed under a positive fluorescence microscope (BX63, Olympus, Tokyo, Japan).
Immunohistochemistry and in situ hybridization
The formaldehyde fixed (4%) tumor tissues were embedded in paraffin. After being baked, the samples were successively placed in xylene solution, and then soaked for 15 min after xylene was replaced. Then the samples were dehydrated with gradient ethanol. Finally, the samples were washed with double distilled water. Each slice was dipped with 3% H2O2 to block endogenous peroxidase. After addition of citric acid buffer, the samples were boiled in a microwave oven for 3 min, treated with antigen repair solution, and allowed to rest at room temperature for 10 min. Next, normal goat serum blocking solution (Sangon, Shanghai, China) was used to incubate the samples at room temperature for 30 min. Diluted primary antibodies against HDAC2 (1:200, ab32117, Abcam) and Rad51 (1:500, ab133534, Abcam) were dropped to the samples for incubation at 4 ℃ overnight. The following day, the secondary antibody goat anti-rabbit against IgG (ab6721, 1:5000, Abcam) was used to incubate the samples for 30 min. Following incubation with SABC (Vector Company) in a 37 ℃ incubator for 30 min, DAB reagent kit (P0203, Beyotime). The samples were dyed in hematoxylin and sealed with neutral resin, and observed under an upright microscope (BX63, Olympus).
According to the instruction of in situ hybridization detection kit (Bio-high technology, Shijiang Huang, China), digoxigenin-labeled oligonucleotide probes (5ʹ-GCCCTTTTACATGCACTG-3ʹ) were used to detect the presence of miR-130a-3p in tumor tissues.
Hematoxylin and eosin (HE) staining
Complete tumor tissue sections were fixed at room temperature for 30 s, and stained with hematoxylin (60 ℃) for 60 s. Following a rinse with 1% hydrochloric acid alcohol differentiation solution, sections were counterstained with eosin for 3 min, dehydrated, and cleared with xylene. After that, sections were cleared, sealed, and observed under a microscope (BX63, Olympus).
Statistical analysis
All data i were analyzed utilizing the SPSS 21.0 statistical software (IBM, Armonk, NY). Each experiment was repeated three times. The measurement data were expressed by mean ± standard deviation. Paired t test was used to compare data between NSCLC tissue and adjacent normal tissue. Data between two groups were compared by unpaired t test, and those among multiple groups by one-way analysis of variance (ANOVA), followed by Tukey’s post hoc tests. Data among multiple groups at different time points were compared by repeated measures ANOVA, followed by Bonferroni post hoc tests. Kaplan–Meier method was used to calculate the survival rate of patients. p < 0.05 was indicative of statistically significant difference.
Discussion
Lung cancer is one of the leading causes of cancer death worldwide [
18], and almost 80–85% of lung cancers are NSCLC [
19]. In spite of the improvement in chemotherapy or targeted therapies, there is still a low survival rate of NSCLC [
20]. In the current study, we explored the regulatory mechanism of HDAC inhibitor ITF2357 in the chemo-resistance of mut-KRAS NSCLC to Pem and found that HDAC inhibitor ITF2357 could reduce resistance of mut-KRAS NSCLC to Pem by regulating the miR-130a-3p/HDAC2/Rad51 axis.
It has been well-established that HDACs are enzymes functioning to remove the acetyl groups from hyperacetylated histones [
21] and participate in cancer progression via modulation of tumor cell proliferation, differentiation and cell cycle [
22]. Recently, an increasing number of studies have unveiled the participation of HDAC2 in the development and chemotherapeutic effect of NSCLC. For instance, HDAC2 could facilitate the migration and invasion of NSCLC cells by increasing fibronectin [
23]. Besides, upregulated HDAC2 by FKBP3 was found to contribute to enhanced viability of NSCLC [
24]. In addition, overexpressed HDAC2 could lead to reduced cisplatin sensitivity of NSCLC cells induced by valproic acid [
8]. Consistent with the previous studies, we found in the present study that HDAC2 expression was upregulated in NSCLC tissues and cells.
Intriguingly, accumulating evidence has further indicated that HDAC inhibitors, such as entinostat, SAHA, trichostatin A and KA2507, could suppress the growth of cancer cells and induce apoptotic cell death in a variety of malignancies including metastatic uveal melanoma, lung cancers, glioblastoma, etc. [
25‐
28]. ITF2357 (givinostat) is a safe and tolerable pan-HDACi with broad anti-inflammatory properties [
29], and the anti-tumoral activity of ITF2357 has been reported in various tumors, including NSCLC [
7,
30,
31]. In the current study, we illuminated that HDAC inhibitor ITF2357 inhibited the expression of HDAC2, thereby reducing the resistance of mut-KRAS NSCLC cells to Pem. In line with our results, it also has been highlighted that ITF2357 could promote the cytotoxicity of Pem in NSCLC and contribute to increased suppression of xenografted tumor growth in vivo as well as prolonged mice survival [
7]. Moreover, a previous study has revealed that ITF2357 could induce the death of ruxolitinib-resistant cells and augment the effect of chemotherapy in B-cell precursor acute lymphoblastic leukemia [
32].
Importantly, our molecular mechanism exploration in this study revealed that inhibition of HDAC2 promoted the expression of miR-130a-3p, thereby inhibiting Rad51, which ultimately reduced the resistance of mut-KRAS NSCLC to Pem. To our knowledge, the regulatory relationship between HDAC2 and miR-130a-3p has been rarely reported. There is also a paucity of reports unveiling the interaction between miR-130a-3p and Rad51. In the current study, we discovered a negative correlation between HDAC2 and miR-130a-3p and a positive correlation between Rad51 and HDAC2 in NSCLC based on the results from RT-qPCR and Western blot assay. In consistency with our finding, several previous studies have unveiled the role of Rad51 in NSCLC as an oncogene. For example, upregulation of Rad51c is predictive of poor prognosis of patients with NSCLC and promotes cell resistance to cisplatin as well as radiation in this malignancy [
33]. Intriguingly, knockdown of endogenous Rad51 could suppress the growth of the A549 lung cancer cells through accumulation of cells in G1 phase and induction of cell death, thereby serving as an independent prognostic gene in patients with NSCLC [
34].
Of note, previous research has also identified miR-130a-3p as an inhibitor of the development of NSCLC. It was demonstrated that miR-130a-3p displayed marked downregulation in patients with NSCLC in comparison to that in healthy individuals [
12]. In addition, miR-130a-3p was found to be underexpressed in A549 and H1299 lung cancer cell lines and could suppress the migration of lung cancer cells [
35]. Furthermore, Hu et. al. revealed that miR-130a-3p could target SOX4 to augment cisplatin resistance of NSCLC cells and thus enhance efficacy of chemotherapy for NSCLC patients [
17].
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.