Background
Na
+/H
+ exchanger regulatory factor1 (NHERF1) is an important molecule among the four protein family (NHERF1, NHERF2, NHERF3, NHERF4). NHERF1 is generally regarded as a multifunction cellular scaffold protein containing two structural domains of PDZ1 and PDZ2 upstream of an ezrin–radixin–moesin-binding element. NHERF1 is known to interact with a variety of proteins of great importance in human cancers, including platelet-derived growth factor receptor (PDGFR) and phosphatase and tensin homolog (PTEN) [
1]. By counterbalancing PI
3K/Akt oncogenic signals, NHERF1 enhanced cell responses to PDGFR inhibitor [
1]. Mutations at NHERF N-terminus within the conserved PDZ domains hampered its interaction with SYK, promoted the progression of breast cancer [
2]. NHERF1 binding to the C-terminal of EGFR inhibited its phosphorylation and attenuated downstream signaling [
3].
The differential and heterogeneous expression of NHERF1 is related to the progression of several tissues. In polarized colorectal epithelia, NHERF1 distributed at the apical luminal, but disrupted in cancer cases [
4]. In transformed cells, cytoplasmic ectopic NHERF1 expression increased cell proliferation and accelerated the development of malignant phenotype [
5]. During the early stages of carcinogenesis in colorectal cancers, nuclear NHERF1 was correlated with poor prognosis [
4]. In HCC models, NHERF1 and β-catenin colocalized in the nucleus and functioned as a positive regulator of Wnt signaling and contributed to the malignant phenotype [
6]. In breast cancer, increased cytoplasmic NHERF1 and decreased membranous localization during ductal carcinoma in situ (DCIS) transformed into invasive and metastatic types [
7]. These data implied that NHERF1 could contribute to the progression of cancers based on its expression and cellular distribution.
Lung cancers are the leading types of cancers over the world in both incidence and morbidity, approximately 85% of which are non-small cell lung cancers (NSCLC) [
8]. The five-year survival was merely about 15% following optimized chemotherapy procedures [
9]. The recent development using targeted therapy has greatly encouraged the screening of novel lung cancer markers for effective drugs or treatment assessments. The anaplastic lymphoma kinase gene (ALK) is a new driving gene of NSCLC following the discovery of EGFR mutations. ALK is a 1620 amino acid transmembrane protein consisting of extracellular ligand binding, transmembrane and intracellular tyrosine kinase domains [
10]. In 2011, the US FDA has approved crizotinib to be used for the treatment of ALK-translocated locally advanced or metastatic NSCLC patients [
11]. The diagnosis of ALK-translocated lung cancer took place in 3–5% in all NSCLC patients [
12]. Although resistance of crizotinib was reported in ALK positive patients subjected to the procedures of targeted therapy, the possible mechanism was poorly understood to date [
13]. Comparing to other drugs, genes that influenced the sensitivity of crizotinib were less reported due to the low frequency of ALK mutations in patients.
From previous studies in NSCLC, NHERF1 proteins in the plasma were increased in patients, suggesting NHERF1 might be useful for the evaluating tumor development and progression. A recent study involving 26 cases of NSCLC tumor puncture and 18 cases of matched tissue sections showed that the increase and mislocation of NHERF1 from immunohistochemistry could be used to indicate cancer invasion [
14]. It was also reported that NHERF1 elicited pharmacologic effect to enhance drug response of geftinib and imatinib in breast cancers [
15]. In cervical cancer cells, NHERF1, as a putative interaction partner, reduced the stability of MRP4 by facilitating its internalization from cell surface to sensitize cisplatin-refractory [
16]. These findings prompted us to evaluate the role of NHERF1 in ALK positive NSCLC in connection with the sensitivity to crizotinib treatments.
In the present study, both ALK-translocated and ALK-negative lung adenocarcinoma specimens in tissue sections were collected for immunohistochemistry. An ALK-positive and crizotinib-sensitive lung adenocarcinoma cell line H3122 was used as the laboratory model for experiments. Either a NHERF1 overexpression vector or agents for NHERF1 knockdown was used for crizotinib sensitivity measures, in association with cell viability and apoptosis assays. The results showed that NHERF1 ectopic expression was able to increase the drug sensitivity of crizotinib treatments. Our findings also suggested ALK activity and downstream signals were not only involved to regulated NHERF1 expression, but also affected the subcellular localization. NHERF1 could be a useful referencing marker to evaluate drug resistance, at least the sensitivity of crizotinib in NSCLC cells.
Methods
Tissue sample collection and database accession
Lung adenocarcinoma tissues were collected with the approval by the ethics committee from Beijing Xuanwu Hospital. A total of 10 ALK-positive patients were included with the confirmed diagnosis from the pathology department. ALK rearrangements were identified on formalin-fixed, paraffin-embedded (FFPE) tumors using Vysis ALK Break Apart FISH Probe Kit (Abbott Molecular, Abbott Park, IL, USA). Additional 10 ALK-negative lung adenocarcinoma tissues matched for similar clinical stage and grade were randomly selected for comparative studies. All tumor specimens were prepared at the time of surgical resection, and then fixed in 10% neutral buffered formalin for 24 h following an identical SOP. The processed samples were subjected to IHC analyses. For cross-reference, the NHERF1 gene expression data for normal lung or cancerous adenocarcinoma tissues were extracted from GEO dataset (GSE31210), and then used for statistical analyses.
Chemical reagents and plasmids
Crizotinib was purchased from HARVEY (Beijing, China). The siRNA targeting sequences for NHERF1 was designed as 5′-GCUAU GGCUU CAACC UGCAT T-3′ and 5′-GAAGG AGAAC AGUCG UGAAT T-3′. The synthetic interfering RNA in duplex was transfected into cells seeded in 6-well plates. A scrambled siRNA 5′-UUCUU CGAAC GUGUC ACGUT T-3′ (Gene-Pharma, Suzhou, China) was used in control samples. Heparin was purchased from Bellancom (Beijing, China). The NHERF1 overexpression plasmid fused with a 3 × FLAG tag was prepared as previously described [
17].
Cell culture and transfection
H3122, the human adenocarcinoma ALK-positive NSCLC cell line, harbouring EML4-ALK variant 1 fusion gene was purchased from Kebai Biotechnology Co. Ltd. (Nanjing, China) and maintained in RPMI 1640 (Gibco, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS) (ExCellBio, Shanghai, China) and 1% P/S antibiotics (Gibco BRC, Paisley, Scotland, UK). The transfection of plasmids or siRNAs were carried out using the Neofect™ DNA transfection reagent (Neofect Biotech, Beijing, China) or Lipofectamine® 2000 transfection reagent (Invitrogen, Paisley, Scotland, UK). The total proteins were extracted at 48 h post-transfection. The protein levels in treated cells following NHERF1 gene overexpression or its siRNA inhibition were determined by western blotting.
Cytoplasmic and nuclear extracts
Cells were seeded in 10-cm plates with or without treatments for 48 h, and then harvested in 1000 μl of cold phosphate-buffered saline (PBS). Collected cells were centrifuged (4 °C) at 500 g for 2 min, washed twice with cold PBS, and resuspended in 400 μl of cold lysis buffer (10 mM HEPES, 50 mM NaCl, 5 mM EDTA, 1 mM Benzamidine, 0.5% Triton X-100). The lysates were vortexed for 15 s and placed on ice for 20 min. The samples were centrifuged for 1 min (2000 g) at 4 °C and the supernatant were collected as the cytoplasmic fraction. The precipitate was resuspended in 40 μl nuclear lysis buffer. The cytoplasmic or nuclear lysate were placed on ice for 40 min and vortex for 10 min at 15 s with intervals. The final homogenate was centrifuged at 16,000 g for 10 min (4 °C) and the supernatant was used as the nuclear fraction. The prepared samples were stored at − 80 °C for subsequent analyses.
Western blotting
The treated cells were lysed following a standard protocol. The protein expression was assessed through sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and western blotting analysis. For ALK signaling experiments, confluent cell monolayer of the H3122 cells were serum-free starved 24 h and then exposed to 100 μg/ml heparin for 30 min. The anti-NHERF1 antibody was purchased from BD Transduction Laboratories (BD Biosciences, San Diego, CA, USA). The anti-GAPDH, anti-ERK, anti-p-ERK and anti-Coilin antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The anti-AKT, anti-p-AKT (Ser473), anti-ALK, anti-p-ALK (Tyr1604) antibodies were from Cell Signaling Technology (Danvers, MA, USA). The β-action antibody was obtained from ABclonal Biotechnology (Wuhan, Hubei, China). The protein expression levels were semi-quantified by densitometry of the western blot bands using the ImageJ2 software.
Cell proliferation assay
For each well, a number of 3000 cells were seeded in 96-well plates and continuously cultured for 0, 1, 2, 3 and 4 days. The cell growth was assayed by adding 100 μl of cell counting kit (CCK8) (Dojindo, Kamimashiki-gun, Kumamoto, Japan) solution at 1:10 dilution in culture medium for staining at 37 °C for 1 h. The plates were scanned at absorbance wavelength of 450 nm using a spectrophotometer (BioTek, Winooski, VT, USA) for detection. Each data point was measured in six duplicates. The experiments were repeated for 3 times independently.
Apoptosis assay
Cells were seeded in 6-well plates at the density of 1 × 105 cells/well. The treatments with crizotinib at 10 μM were maintained for 24 h. In the control group, the cells were treated vehicle (DMSO) only for dissolving the drugs. The cell apoptosis was assessed using a MUSE™ Annexin V Dead Cell Kit (EMD Millipore Corporation, Hayward, CA, USA) following the manufacturer’s protocol.
Immunofluorescence
Cells were plated at a density of 1.0 × 105 per well into 6-well plates pre-placed with a glass coverslip. Following transfection, the cells on coverslips were treated with/without heparin and then washed three times with cold PBS. After fixed with 4% paraformaldehyde at room temperature for 30 min, the cells were permeabilized and blocked with a buffer containing 0.3% Triton X-100 and 5% bovine serum albumin for 30 min. To determine the subcellular localization of NHERF1, cells were probed with monoclonal mouse anti-human NHERF1 primary antibody (1:100) (BD Transduction Laboratories) and anti-Flag primary antibody (1:200) (Sigma-Aldrich, St. Louis, MO, USA). Following extensive wash for removing the primary antibodies, the cells were incubated with an Alexa Fluor® 488 donkey anti-mouse IgG (Life Technologies, MA, USA) at 1:200 dilution in the dark for 1 h. Prior to the examination under a fluorescent microscope, cell nuclei were stained with Hoechst 33258 (Sigma-Aldrich). Images were acquired with a confocal system (Leica Microsystems LAS AF-TCS SP5. Wetzlar, Germany). Control samples without adding primary or secondary antibodies were prepared for evaluating the non-specific staining in this study.
Immunohistochemistry (IHC)
The 5 μm sections of collected lung tissues were deparaffinized, rehydrated, rinsed with distilled water, and washed with Tris-buffered saline (TBS). Automated IHC staining was carried out using a Ventana BenchMark GX instrument (Ventana Medical Systems, Inc., Tucson, AZ, USA). NHERF1 was probed with an anti-NHERF1 antibody (1:100, Abcam, Cambridge, MA, USA). Image Pro Plus 6.0 (Media Cybernetics, Inc., Rockville, MD, USA) was used to calculate the intensity of NHERF1 staining. Randomly selected 10 microscopic fields were imaged for each sample subjected to computer-assisted imaging analyses. The results were presented as the mean ± standard deviation (SD). The determination of immunohistochemical staining intensity and positive areas for NHERF1 in IOD (integral optical density) measures were performed by two independent staffs who were mutually blinded to the sample grouping information.
Statistical analyses
The quantitative data were presented as mean ± SD and subjected to Student t-test. Statistical significance was called at p < 0.05. The charts were generated by Prism V5.0 (GraphPad Software, Inc., La Jolla, CA, USA).
Discussion
The upregulation of NHERF1, as a suggestive poor prognosis indication, was earlier discovered in breast cancers for its association with tumor malignancy and metastatic progression [
21]. Latest studies attempted to discriminate the diverged functions of NHERF1 with different subcellular distribution. Cytoplasmic NHERF1 was detected higher in primary cancer than in adjacent normal mucosa, and implicated nodal and distant metastases or lymphovascular invasion (LVI) [
7]. Nuclear NHERF1 was identified in colorectal cancer of early stages and was believed to participate the onset of the malignant phenotype during carcinogenesis leading to poor prognosis [
22]. In lung cancers, NHERF1 protein was increased in both plasma and tissues from NSCLC patients. NHERF1 interaction partners can be either cytoplasmic or nuclear factors, regulated by a wide range of signals in breast cancer or colorectal cancers, such as VEGFR and HIF-1ɑ, some of which closely associated with lymphatic metastasis [
17,
23]. Besides HIF-1ɑ, NF-κB was another important nuclear transcription factor demonstrated to interact with NHERF1 dependent to the activation of Ras/MEK/ERK or PI3K/Akt signaling [
24]. In primary macrophages and vascular smooth muscle cells, NF-κB in return was shown to promote the expression of NHERF1. From this study, a high level of NHERF1 expression was found in lung cancer tissues, especially in patients diagnosed with ALK-translocation (Fig.
1). ALK was a known activator of AKT and ERK signaling pathways upstream of NF-κB or HIF-1ɑ mediated transcription of the regulated genes, therefore it was worthwhile to exploit whether the activation of ALK signal was sufficient or responsible to modulate NHERF1 expression. As predicted, we found that ALK activation induced by heparin in coordination with ERK and AKT phosphorylation upregulated the transcription and expression of NHERF1 (Fig.
2b). The result immediately suggested that NHERF1 might influence the sensitivity to crizotinib in ALK targeted therapies, as implied from Fig.
2a.
NHERF-interacting proteins, comprising transporters, membrane receptors, junction proteins or signaling molecules, have been identified with a function to influence the cell survival following treatments of chemo-drugs [
19]. It is no surprise for NHERF1 to alter the sensitivity or resistance to cancer drugs. In fact, NHERF1 has also been suggested to connect with tumor sensitivity to drugs targeting EGFR, MRP4 or MRP2 [
18]. The knockdown of NHERF1 expression reduced crizotinib sensitivity in ALK-positive H3122 cells via its inhibition on cell apoptosis (Fig.
3a,d) and promotion of cell growth (Fig.
4a). Further results from NHERF1-overexpressed H3122 cells (Figs.
5 and
6) were able to verify the role of NHERF1 on crizotinib sensitivity, except for the effects on cell growth was weak. These results were consistent with the ectopic overexpression of NHERF1 significantly reduced the heparin-induced cell growth in H3122 cells (suppl fig.
1). NHERF1 is a scaffold protein, which can form complex network through its two PDZ domains to regulate many cellular signal pathways [
16]. A majority of the molecular interactions involving the binding of NHERF1 is dependent to protein phosphorylation [
1]. The dentification for major NHERF1 interaction protein with its chemical modification status will be particularly helpful to understand the exact mechanism explaining the effect of NHERF protein on drug sensitivities. To the best of our knowledge, there is no evidence indicating a direct interaction between NHERF and ALK, the connection of NHERF1 to ALK activation was not previously reported and needed to be further studied.
NHERF1 was suggested to play an essential role in cancer initiation, development, progression and metastasis [
6]. An important finding from this study was the differential effect of NHERF1 functions linked to its localization. Ectopic expression of NHERF1 preferred a cytoplasm distribution and led to strong inhibition of cell growth (Figs.
5 and
7). Thus, it remained to be considered as a potential candidate for manipulation to treat cancers as previously recognized. The nuclear NHERF1 played a role on promoting tumor progression through the interaction with YAP and increased YAP translocation into the nucleus for transcription activation [
25]. In a few reports, it was also implied that NHERF1 lost its tumor suppressive function once mislocated from the cytoplasm into the nucleus, or even worse acted as a pro-oncogene product. We found that the translocation of NHERF1 happened to heparin-induced endogenous NHERF1 expression in ALK positive cancer cells (Fig.
7). The results partially explained why malignant cancer cells not only sustained high NHERF1 expression but also might benefit from the outcome for proliferation preference. Continuous explore about the complex mechanisms of NHERF1 regulating functions in cancers will supply more evidence to address the remaining questions.
Crizotinib is a first-generation ALK-TKI applied to NSCLC subtypes. Resistance to crizotinib can occur via secondary mutations or amplification in the ALK gene. Cancer cells may adopt various strategies, such as amplified KIT signaling, increased EGFR phosphorylation, KRAS mutations oncogene, or increased NRG1 for activating HER2/3 kinases to survive crizotinib treatments [
26]. Epithelial-mesenchymal transition (EMT) was suggested to be a critical process to confer resistance to crizotinib in NSCLC. NHERF1 suppressed EGFR phosphorylation and inhibited EGF-induced proliferation/migration in breast cancer cells [
27]. Previous studies showed that NHERF1 suppressed lung cancer cell migration by attenuating EMT process. Loss of the interaction between NHERF1 and EGFR induced EMT phenotypic features in biliary cancer cells [
28]. We found crizotinib reduced the expression of NHERF1 (Fig.
2a), implying that the inhibition of NHERF1 on EGFR phosphorylation was compromised. NHERF1 was discovered as putative interaction partners of MRP4 [
16] in cervical cancers [
19]. It implied that monitoring changes in NHERF1 levels could be used to predict loss of crizotinib sensitivity during drug treatments, as the inhibitory function of NHERF1 to EMT contributed to crizotinib drug resistance.
The role of NHERF1 in drug resistance in lung cancers deserves more investigation for clues to understand the underlining mechanism. By characterization the effects of NHERF1 with its regulation on tumor cell responses to crizotinib, we have demonstrated a dynamic interplay between NHERF1 and ALK signaling in lung adenocarcinoma cells. Upregulation of NHERF1 in lung adenocarcinoma needs to be scrutinized during the administration of ALK-targeting drugs. The reduction in NHERF1 expression in cancer cells treated with crizotinib is likely to suggest acceleration of tumor progression, as well as the resistance to drugs. The expression and subcellular distribution of NHERF1 might be a potential prognostic factor to predict the benefit in patients subjected to ALK targeted therapies.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.