Background
Melanoma is generally associated with poor outcome once metastatic disease stages have been reached. Compared to other solid cancers, this most aggressive form of skin cancer exhibits an extremely high prevalence of somatic mutations [
1,
2], which is almost entirely attributable to UV light exposure. Despite this high genetic heterogeneity, 40–60% of melanoma patients carry mutations in the Ser/Thr-kinase BRAF (most often V600E), which renders the BRAF kinase and the downstream MAPK signalling pathway constitutively active [
3]. The introduction of specific kinase inhibitors for melanoma patients carrying this BRAF mutation has revolutionised melanoma care. In 2011, BRAF inhibitors were FDA-approved showing convincing results at first [
4,
5] and since 2015 a combined inhibition of BRAF and MEK kinases is recommended [
6,
7], which has increased median survival from 18.7 to 25.1 months [
8,
9]. However, despite these unprecedented clinical responses, drug resistance arises rapidly within 3–12 months [
10,
11] leaving as only treatment options chemotherapy and in some cases immunotherapy. Most often, acquired resistance is driven by secondary mutations, which re-activate the MAPK signalling pathway resuming rapid proliferation.
Anaplastic lymphoma kinase (ALK) is a receptor tyrosine kinase that is normally involved in the development of the nervous system [
12]. In differentiated tissues, ALK can be activated by translocations or mutations making it an oncogene in a variety of malignancies, such as non-small cell lung cancer, anaplastic large cell lymphoma, neuroblastoma and many more [
13]. Additionally, in 2015, Wiesner and colleagues identified in 11% of melanoma tissues a truncated ALK transcript starting from intron 19 and resulting in a smaller protein, which was shown to be oncogenic [
14].
Here, we identified the overexpression of a novel truncated form of ALK, named ALKRES in the hereafter, as new mechanism driving acquired drug resistance in melanoma cells. In particular, we demonstrate that treatment of the ALKRES-expressing resistant melanoma cells with siRNA or ALK inhibitors in combination with either BRAF or MEK inhibitors, leads to efficient cell growth suppression and apoptosis, suggesting this combination to be an interesting clinical option for patients harbouring both BRAFV600E and expressing ALKRES, especially as more specific ALK inhibitors become available. Moreover, we show for the first time that the overexpressed ALKRES is secreted into extracellular vesicles (EVs) and is transferred to sensitive, ALK-negative melanoma cells. There, ALKRES is functional in activating the MAPK signalling pathway and thus is involved in transferring of drug resistance. Finally, the combination of BRAF and ALK inhibitor treatments of mice bearing ALK-positive melanoma tumours dramatically reduced tumour volumes, making ALK an exciting clinical target in melanoma patients.
Methods
Inhibitors
All inhibitors used in this study were purchased from Selleckchem and were dissolved in DMSO at a concentration of 10 mM and stored at − 20 °C.
Cell lines and cell culture
A375 melanoma cells were purchased from ATCC and cultured as previously described [
15]. Drug-resistant clones were generated by culturing parental A375 cells in presence of 1 μM PLX4032 for 6–8 weeks. 20 different clones were picked and grown independently under constant PLX4032 treatment. The clone A375X1 was selected for further experiments.
Microarray analysis
Total RNA was extracted with the miRNeasy mini kit (Qiagen) in triplicates following the manufacturer’s instructions. RNA quality was further assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies). Microarray analyses were performed at the Luxembourg Institute of Health (LIH) by using the Affymetrix HuGene 2.0 ST platform as described previously [
15]. The raw microarray data are accessible in the ArrayExpress database (
https://www.ebi.ac.uk/arrayexpress/) under the accession number E-MTAB-6596.
5’RACE, sequencing of amplified products and PCR
5’RACE was performed according to the manufacturer’s instructions using the GeneRacer™ kit (Invitrogen) and ALK specific primers binding to exon 21 and to the junction between exon 24 and 25 were designed. The final product was sequenced at GATC Biotech (Konstanz, Germany). In addition, ALK was fully sequenced.
PCR amplification of both ALK and the fusion between MMLV and ALK were performed using specific primers. All primer sequences are listed in Additional file
1: Table S1.
Quantitative PCR
Total RNA was extracted using the Quick-RNA™ miniprep kit (Zymo Research) according to the manufacturer’s instructions and the concentration and quality was determined using a NanoDrop Spectrophotometer. Quantitative real time qPCR was performed as described previously [
15]. ALK primers listed in Additional file
1: Table S1.
ALK immunoprecipitation
ALK was precipitated from lysates of A375X1 cells. Cells were lysed in RIPA buffer and incubated with ALK antibody (1:100) overnight at 4 °C on an overhead shaker. The next day, lysates were incubated with protein G sepharose™ (GE Healthcare), which was previously washed with the lysis buffer, for 1 h at 4 °C on an overhead shaker. After three washing steps, the protein was released by heat treatment in 2× Laemmli buffer and separated by SDS-PAGE.
Small interfering RNAs and transfection
Three different ALK siRNAs were obtained from GE Dharmacon (ON-TARGETplus Human) (Additional file
1: Table S2). siRNA transfections were performed using 1.5 μl Lipofectamine RNAiMAX (Invitrogen) per reaction according to the manufacturer’s instructions. The final concentration of both ALK siRNA and scrambled control was 100 nM. siRNA transfections were performed 24 h prior to 48 or 72 h incubation with PLX4032 (1 μM), Trametinib (5 nM) or MK2206 (1 μM).
Western blot analyses and antibodies
Cell lysis and Western blot analysis were performed as described previously [
16,
17]. The following antibodies were used: phospho-ERK1/2, phospho-AKT, phospho-ALK and ALK (from Cell signaling), ERK1/2, tot-AKT and α-tubulin (from Santa Cruz), CD9 and CD81 (from System Biosciences) and TSG101 (from Abcam).
Real-time proliferation assays
25 X 103 cells/well of A375X1 melanoma cells were seeded in 24-well plates and 24 h later treated with both scrambled and ALK siRNA. Next, cells were incubated with PLX4032 (1 μM), Trametinib (5 nM) and MK2206 (1 μM). Cellular growth was monitored in the IncuCyte ZOOM live cell microscope (Essen BioScience) and images were taken in phase contrast every 3 h for a total of 90 h.
Dose-response analysis of kinase inhibitors
Black 96-well μclear plates (Greiner) were used. In case of ALK inhibitors, 5000 cells/well of resistant A375X1 cells were seeded in RPMI medium. In order to determine the dose-response, kinase inhibitors were serially diluted at a ratio of 1:3, starting at 10 μM for Crizotinib and ASP3026 and starting at 1 μM for Ceritinib, in a total reaction volume of 100 μl. A blank control (RPMI medium only), as well as an untreated control were included for each cell line. For dose-response to vemurafenib, 3500 cells/well of resistant A375X1 cells were seeded and pre-treated with 1 μM of Crizotinib and ASP3026 and 100 nM of Ceritinib. 24 h after the pre-treatment, vemurafenib was serially diluted at a ratio of 1:3, starting at 10 μM and added to the cells. For drug resistance transfer, 1000 cells/well of sensitive A375 were seeded in 100 μl of RPMI medium. The day after, EVs at a concentration of 10 μg/ml were added to the cells. 24 h later, does-response to vemurafenib was performed.
For all experiments, cell viability was measured 72 h later using the CyQuant proliferation assay. Fluorescence intensity was measured using the microplate reader CLARIOstarR (BMG-LABTECH). The blank corrected values were exported as Microsoft Excel files and analysed. Experiments were performed in technical and biological triplicates. Dose-response curves were generated using GraphPad Prism 5.
Caspase-3 activity assay
To measure apoptosis in A375 and A375X1 cells, 20000 cells/well were seeded in black 96-well μclear plates and treated with 1 μM or 100 nM of single or combined inhibitors (PLX4032 or ALK inhibitors). Cells treated with etoposide (200 μM) were included as an internal positive control for apoptosis. 24 h later, cells were lysed with a lysis buffer containing dithiothreitol (6 mM) and DEVD-AFC substrate (AFC: 7-amino-4-trifluoromethyl coumarin) (Alfa Aesar) for 30 min at 37 °C. Upon cleavage of the substrate by caspases, free AFC emits fluorescence, which can be quantified using a microplate reader (400 nm excitation and 505 nm emission). Additionally, we included a blank control (RPMI medium only), an untreated control as well as a negative control represented by cells treated with DEVD-CHO (Alfa Aesar), a synthetic tetrapeptide inhibitor for Caspase-3. Fluorescence intensity was measured using the microplate reader CLARIOstarR (BMG-LABTECH). The DEVD-CHO corrected values were exported as Microsoft Excel files and analysed.
In vivo assays
NOD/SCID gamma (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ) (NSG) mice were bred in-house. Approval by the University’s animal care and ethics committee was obtained (18-MDM-01) and in vivo experiments were performed according to applicable laws and regulations. Single A375X1 resistant cells (2*106 cells) were resuspended in 100 μL of 1:1 mixed serum-free medium and matrigel (BD Biosciences) and injected subcutaneously (right and left flank) of 6–8 week-old mice. Mice were randomized at day 10 (n = 5, tumour volume around 100mm3), and daily oral treatment was started for 7 consecutive days with vehicle, 45 mg/kg vemurafenib, 50 mg/kg ceritinib, or the combination of ceritinib and verumafenib. Drugs were formulated in 4% DMSO, 30% PEG 300, 5% Tween 80, ddH2O. Tumour growth was followed and tumour volume was calculated by the formula LxW2/2.
Patient samples and immunohistochemistry
Tumour samples were collected from melanoma patients at the Klinikum Dortmund (in Germany). Samples were obtained with patient consent and approval of the ethic committee (Ethikkommission der Ärztekammer Westfalen-Lippe und der Westfälischen Wilhemls-Universität, reference number 2015–178-f-S). Patient studies were conducted according to the Declaration of Helsinki and the Belmont Report.
Immunohistochemistry on formalin-fixed paraffin-embedded (FFPE) slides from melanoma samples was performed at the Integrated Biobank of Luxembourg (IBBL). Additional information is included in Additional file
2: Supplementary Methods.
Donor cells (both A375 and A375X1) were slowly adapted to serum-free medium (UltraCulture, Lonza BioWhittaker). Culture supernatants (100 ml) were harvested, centrifuged 2 × 10 minutes at 400 g, followed by 20 min at 2000 g to remove cells and cell debris. Extracellular vesicles were isolated by ultracentrifugation (70 min at 110000 g, 4 °C) by using a MLA-55 fixed rotor followed by flotation on an Optiprep cushion (Axis-Shield, 17%) for 75 min at 100000 g at 4 °C using a swinging MLS-50 rotor. After a PBS wash (110000 g, 70 min), extracellular vesicles were resuspended in PBS and frozen at − 80 °C. Protein quantification was performed using Pierce™ BCA Protein Assay Kit (Termo Fisher) according to the manufacturer’s instructions.
To label extracellular vesicles, culture supernatants were processed as mentioned above. After ultracentrifugation at 110000 g, the pellet was resuspended in 250 μl of PBS and stained with 5 μl of PKH67 (Sigma) for 30 min at 37 °C. To remove excess dye, this suspension was loaded on the Optiprep cushion, followed by a PBS washing step. 10 μg of labelled EVs were added to the cells; after 24 h cells were fixed and stained with SiR-actin kit (Spirochrome).
Visualization of EVs
For electron microscopy, a drop of extracellular vesicles suspended in PBS was deposited on Formvar-carbon-coated electron microscopy grids. The samples were fixed with 2% PFA, labelled with anti-CD63 (Abcam) and immunogold-labelled using protein A coupled to 10 nM gold (PAG10) as previously described [
18].
EV mass spectrometry
A liquid chromatography-tandem mass spectrometry (LCMS/MS) system was used to study the protein composition of EVs. The detailed protocol is shown in the Additional file
2: Supplementary Methods.
EV transfer experiments
For the transfer assays, 25000 cells in 24 well plates were seeded in RPMI medium. The day after, following 1 h pre-treatment with 1 μM of PLX4032, increasing concentrations of resistant EVs were added to the cells. After 7 h, cells were collected for western blot analysis.
Immunofluorescence
For ALK immunofluorescence, A375 or A375X1 cells grown on glass coverslips were treated with 10 μg of EVs for 24 h. Cells were washed with PBS and fixed with 4% paraformaldehyde in PBS for 10 min at room temperature. The coverslips were washed three times in PBS-Tween (0.05% Tween 20). Then, cells were permeabilised with PBS 0.5% Triton X-100 for 10 min at room temperature, and blocked in PBS plus 2% bovine serum albumin (BSA) for 15 min. Cells were incubated with ALK antibody, diluted in PBS plus 2% BSA, for 1 h at room temperature. Coverslips were washed 3 times with PBS and treated with Alexa Fluor 488 donkey anti-rabbit IgG (Invitrogen) for 1 h at room temperature. Coverslips were washed and mounted with Gold antifade reagent with DAPI (Invitrogen). The cells were visualised by Andor Revolution Spinning Disk confocal microscopy, mounted on a Nikon Ti microscope (60× oil objective) and the images were analysed with ImageJ software.
Statistical analysis
Statistical analysis was performed with the GraphPad Prism software (version 5). All data are presented as mean of three biological replicates ± s.d. and were analysed either with paired Student’s t-test or one-way ANOVA coupled with Tukey’s multiple comparison tests. Differences in tumour volumes among groups of treated mice were tested using a two-way ANOVA (treatment factor p = 0.0004) followed by multiple comparison t-tests corrected with the Holm-Sidak method; data are presented as mean tumour volumes (mm3) ± SEM. Tumour weights were analysed by unpaired student’s t-tests with Welch’s correction at end-point conditions and represented as mean tumour weights (mg) ± SEM.
Supplemental information includes Additional file
2: Supplementary methods, Additional file
1: Tables S1 and S2 and Additional file
3: Figures S1–S7.
Discussion
Over the past few years, the implementation of accurate screening programs together with major advances in treatment choices have vastly improved the life expectancy for advanced stage melanoma patients [
22]. The availability of specific inhibitors targeting mutated BRAF and the downstream MAPK signalling pathway or other kinases activated in melanoma, together with immunotherapies that de-block inhibition of T cell responses against the tumour, offer potent ways to fight this cancer [
23]. However, immunotherapies are only successful in less than 30% of cancer patients, often have severe side effects, lead to resistance and are still very costly [
22,
24,
25]. On the other hand, treatment of BRAF-mutant melanoma patients with BRAF inhibitors in monotherapy or in combination with MEK inhibitors is limited by both acquired and intrinsic drug resistance [
11]. The re-activation of the MAPK signalling pathway due to secondary mutations is one of the key mechanisms driving acquired resistance to BRAF inhibitors. Promising new drugs such as compounds inducing ER stress, targeting mitochondria biogenesis or metabolic pathways (PDKi) that are effective in both intrinsically and acquired resistant cells and/or xenografts have recently been postulated as potential candidates for second line treatments [
17,
26‐
28]. A deeper understanding of the re-activation mechanisms of the MAPK pathway will aid the selection of appropriate novel therapies to improve survival of melanoma patients.
In this study, we report ALK to be involved in driving resistance in a subclone of BRAF-resistant cells. Several translocations, mutations or amplifications render ALK oncogenic in different cancer types [
13]. So far, 22 different genes have been described to fuse with the C-terminal part of ALK making the ALK locus particularly prone to activating translocations [
13]. The various translocations normally produce constitutively activated ALK fusion proteins, which can signal through the MAPK signalling pathway, the PI3K/AKT pathway or the JAK/STAT pathway contributing to cell proliferation and survival [
12]. Therefore, ALK fusion proteins are already important clinical targets in non-small cell lung cancer (EMLA4-ALK) but have also been described in diffuse large cell lymphoma (NPM-ALK) and in inflammatory myofibroblastic tumour (TPM3-ALK). In addition, a new ALK transcript consisting of a fragment of intron 19 followed by exons 20–29 that resulted from an alternative transcription initiation was recently identified in 11% of melanoma patients [
14]. In our study, an activating translocation with a murine leukemia viral sequence was observed, which leads to a truncated protein lacking the N-terminal part (exons 1–17). We confirmed by whole genome sequencing that this MMLV was stably inserted in our A375 cells (data not shown). The identification of MMLV has been reported for many cancer cell lines, including melanoma, across several laboratories [
29,
30] suggesting MMLV as a regular resident in cancer cells. Nevertheless, the activation of ALK by a murine retrovirus suggests that other sequences from human retroviruses or their closely related human retrotransposons or any other translocating sequence can activate this oncogene in humans.
Most of the ALK variants described so far (overexpressed wild-type ALK, EML4-ALK, NPM-ALK, ALK
ATI, ALK
R1275Q, ALK
F1174L) were shown to trigger proliferation and tumourigenesis and to be sensitive to ALK inhibitors [
14,
31‐
34]. In this context, a phase 2 clinical trial has been launched to test the effect of ALK inhibitor in melanoma patients harboring ALK alterations or aberrant ALK expression (
https://clinicaltrials.gov/ct2/show/NCT03420508#studydesc).
In our study, to determine therapeutic responses, we tested three different ALK inhibitors in combination with BRAF inhibitor. As expected, both knock down and inhibition of ALKRES did not have any effect per se on the growth of resistant cells as phosphorylation of ERK was not inhibited. Only with the combination of BRAF inhibition (and subsequently ERK), cell growth was suppressed and apoptosis induced. This demonstrates that ALKRES modulates sensitivity to BRAF inhibition. The combined inhibition of BRAF and ALK could therefore be of immediate clinical relevance to those patients who acquired secondary mutations within ALK or for those who carry BRAFV600E together with an oncogenic isoform of ALK and show intrinsic resistance to BRAF inhibitor monotherapy.
Importantly, the presence of ALK
RES in resistant cells was mirrored in the corresponding EVs, suggesting that circulating vesicles might be useful diagnostic tools to identify biomarkers of resistance. The detection of ALK
RES in EVs prompted us to examine whether this new oncogenic protein could also be transferred to other melanoma cells. The transfer of phenotypic traits through EVs is an emerging field of research [
35,
36]. Here, we describe for the first time a functional transfer of a truncated kinase (ALK
RES) by EVs likely involved in the propagation of a drug resistance phenotype in melanoma. Of note, the modest effect induced by resistant-EVs (Fig.
5a and
b, Fig.
6b and
c) is not surprising: EV preparations represent an heterogeneous mixture of vesicles [
37] and if only a subtype of EVs carries ALK, its efficacy will be diluted by the presence of other types of EVs, which also transport a spectrum of different proteins and small RNAs [
21,
37]. Furthermore, the isolation protocol might affect the real biological activities of EVs. In addition, it is important to note that ALK might not be the only mediator of drug resistance dissemination and that several players are likely working together to contribute to this phenotype.