Triple blockade of EGFR, MEK and PD-L1 has antitumor activity in colorectal cancer models with constitutive activation of MAPK signaling and PD-L1 overexpression

BAY86–9766 (a selective MEK 1/2 inhibitor) was kindly provided by Bayer Italy (Milan, Italy). For in vitro applications, BAY86–9766 was dissolved in sterile dimethylsulfoxide (DMSO) and the 10 mM stock solution was stored in aliquots at − 20 °C. Working concentrations were diluted in culture medium just before each experiment. For in vivo applications, BAY86–9766 was solubilized in 0.5% Tween-80 in sterile Phosphate Buffered Saline (PBS) before use. Erlotinib (EGFR inhibitor) was provided by Chemietek (Indianopolis, IN, USA) and for in vivo experiments was dissolved in 0.5% Methylcellulose before use. Mouse PD-L1 inhibitor (clone 10F.9G2) was provided by BioXcell (Lebanon, NH, USA) and for in vivo applications was solubilized in PBS before use.

The human SW48 (catalogue number: HTL99020) and HCT116 (catalogue number: HTL99017) colon cancer cell lines were obtained from IRCCS “Azienda Ospedaliera Universitaria San Martino-IST Istituto Nazionale per la Ricerca sul Cancro” (Genova, Italy). The human LIM 1215 colon cancer cell line was obtained from Dr. Di Nicolantonio at Candiolo National Cancer Institute (Candiolo, Italy). The MC38 cell line derived from methylcholanthrene-induced C57BL6 murine colon adenocarcinoma and CT26 cell line derived from N-nitroso-N-methylurethane-induced BALB/c (H-2d) undifferentiated colon carcinoma were kindly provided from Dr. Kopetz at MD Anderson Cancer center (Houston, TX, USA).

SW48, LIM1215 and CT26 cells were grown in RPMI-1640 (Lonza) supplemented with 10% fetal bovine serum (FBS), 1 mM L-Glutamine, and Penicillin (100 μ/ml)–Streptomycin (0.1 mg/ml). HCT116 cell lines were grown in McCoy culture medium (Lonza, Cologne, Germany), supplemented with 10% FBS (Lonza), 1 mM L-Glutamine, and Penicillin (100 μ/ml)–Streptomycin (0.1 mg/ml). MC 38 cells were grown in DMEM (Lonza) supplemented with 10% FBS, 1 mM L-Glutamine, and Penicillin (100 μ/ml)–Streptomycin (0.1 mg/ml). All cell lines were grown in a humidified incubator with 5% of carbon dioxide (CO₂) and 95% air at 37 °C. All cell lines were routinely screened for the presence of mycoplasma (Mycoplasma Detection Kit, Roche Diagnostics, Monza, Italy).

Four- to six-week old female balb/c athymic (nu+/nu+) mice were purchased from Charles River Laboratories (Milan, Italy). The research protocol was approved and mice were maintained in accordance with the institutional guidelines of the University of Campania “Luigi Vanvitelli” Animal Care and Use Committee. Mice were injected subcutaneously (s.c.) with 3.5 × 10⁶ of SW48 cells that had been resuspended in 200 μl of matrigel (BD Biosciences, Milan, Italy). When tumors of approximately 200–300 mm³ in diameter were detected, mice were treated continuously by oral gavage with MEKi (BAY86–9766, 5 mg/kg, every day). Tumor size was evaluated twice per week by caliper measurements using the following formula: π/6 × larger diameter × (smaller diameter)². When tumors were resuming growth despite MEKi treatment, mice were sacrificed and tumors were removed, were cut in several pieces and recovered in vitro by enzymatic treatment [18] to generate MEKi resistant cell lines (SW48-MR). Cells were passaged in culture at least three times before any further testing.

HCT116 and LIM1215 cells were treated continuously in vitro to increasing concentrations of MEKi (BAY86–9766) for a period of 6 months. The starting dose was the dose causing the inhibition of 50% of cancer cell growth (IC₅₀). The drug dose was progressively increased to 1 μg/ml in approximately 2 months, to 5 μg/ml after other 2 months and, finally, to 10 μg/ml after additional 2 months. The established MEKi resistant HCT116 and LIM1215 cancer cell lines (HCT116-MR and LIM1215-MR) were then maintained in continuous culture with this maximally achieved dose of MEKi that allowed cellular proliferation.

SW48-MR and LIM1215-MR cells were grown in RPMI-1640 (Lonza) supplemented with 10% FBS, 1% penicillin/streptomycin. HCT116-MR cell line was grown in McCoy culture medium (Lonza, Cologne, Germany), supplemented with 10% FBS (Lonza), 1% penicillin/streptomycin (Lonza).

C57BL/6 and BALB/c immune-competent mice were respectively injected with MC38 and CT26 colon cancer syngeneic cell lines. When established tumors of approximately 200–300 mm³ in diameter were detected, mice were treated continuously by oral gavage injection of MEKi daily (BAY86–9766, 5 mg/kg). When tumors were resuming growth despite MEKi treatment, MEKi resistant tumors were surgically removed and homogenized into single-cell suspensions used to generate in vitro MC38 and CT26 MEKi resistant cell lines as described above. MC38-MR cell line was grown in DMEM (Lonza) supplemented with 10% FBS, 1% penicillin/streptomycin (Lonza). CT26-MR cell line was grown in RPMI-1640 (Lonza) supplemented with 10% FBS, 1% penicillin/streptomycin (Lonza).

Agilent (Agilent Technologies, Waldbronn, Germany) microarray analyses were performed to assess baseline gene expression profile for SW48 and SW48-MR colon cancer cells using a one color labeling microarray system. The absolute amount and purity (A260/280 nm ratio) of total RNA samples were determined by spectrophotometry (Nanodrop, Thermofisher) and the size distribution was assessed by Agilent Bioanalyzer. Eight hundred nanograms of total RNA were converted into labelled cRNA with nucleotides coupled to a fluorescent dye (either Cy3 or Cy5) following the manufacturer’s protocol (Quick Amp Kit, Agilent). Yield and purity (A260/280 nm ratio) of cRNAs were determined by spectrophotometry (Nanodrop, Thermofisher). 825 ng of cRNA-labeled from SW48 and SW48-MR colon cancer cell lines were hybridized to Agilent Human Whole Genome 4 × 44 k Microarrays. Data were extracted from slide image using Agilent Feature Extraction software (v.10.5). The raw data and associated sample information were loaded and processed by Gene Spring® 11.5X (Agilent Technologies). For identification of genes significantly altered in resistant cells, total detected entities were filtered by signal intensity value (upper cut-off 100th and lower cut-off 20th percentile) and flag to remove very low signal entities. Experiments were performed in triplicate and data were analyzed using Student’s t test (p < 0.05) with a Benjamani-Hochberg multiple test correction to minimize selection of false positives. Of the significantly differentially expressed RNA, only those with greater than 2-fold increase or 2-fold decrease as compared to the controls were used for further analysis. Functional and network analyses of statistically significant gene expression changes were performed using Ingenuity Pathways Analysis (IPA) 8.0 (Ingenuity® Systems, http://​www.​ingenuity.​com). Analysis considered all genes from the data set that met the 2-fold (p-value < 0.05) change cut-off and that were associated with biological functions in the Ingenuity Pathways Knowledge Base. The significance of the association between the data set and the canonical pathway was measured in 2 ways: 1- Ratio of the number of genes from the dataset that map to the pathway divided by the total number of genes that map to the canonical pathway is displayed; 2- Fisher’s exact test was used to calculate a p value determining the probability that the association between the genes in the dataset and the canonical pathway is explained by chance alone.

HCT116, HCT116-MR, LIM1215 and LIM1215-MR cells were seeded into 24-well plates (1 × 10⁴ cells per well) and were treated with different doses of drugs for 96 h. Cell proliferation was measured with the 3-(4, 5-dimethylthiazol-2-yl)-2, 5- diphenyltetrazolium bromide (MTT) (Sigma) assay (final concentration, 5 mg/mL-Sigma-Aldrich). The MTT solution was removed and remained formazan crystals were extracted with Isopropanol supplemented 1% HCl (200 μl/well). The 24-well were shaker for 10 min then 100 μl was subsequently transferred to 96-well. Absorbance of the formazan’s solution in Isopropanol-HCl was measured spectrophotometrically at a wavelength of 550 nm. The IC50 value was determined by interpolation from the dose-response curves. Results represent the median of three separate experiments, each performed in triplicate.

Total RNA was prepared using TRIzol reagent (Life Technologies) and reverse-transcribed into cDNA by SensiFast reverse transcriptase (Bioline) according to the manufacturer instruction. Expression levels of genes encoding for STAT3, PD-L1 and EGFR were analyzed using Real Time quantitative PCR. Amplification was conducted using the SYBER Green PCR Master Mix (Applied Biosystems). All samples were run in duplicate using a Quant studio 7 Flex (Applied Biosystem) and the expression levels of target genes were standardized by housekeeping gene 18S using the 2^-ΔΔCt method.

The small inhibitor duplex RNAs (siRNA) (ON-target plus SMARTpool) siSTAT3 (human: # L-003544-00-000) and siCD274 (human: #L-015836-01-000) were from Dharmacon (Lafayette, CO). The siCONTROL Non-Targeting Pool (#D-001206-13-05) was used as a negative (scrambled) control. Cells were transfected with 100 nM siRNAs using Dharmafect reagent following manufacturer’s instructions. The day before transfection, the cells were plated in 35 mm dishes at 40% of confluence in medium supplemented with 5% FBS without antibiotics. Cells were harvested 48 h after transfection. PCR for STAT3 and PD-L1 expression was done. RNA extraction was performed by the RNeasy Kit (Qiagen, Crawley, West Sussex, UK) following manufacturer’s instructions. The RNA was quantified by Nanodrop (Thermo Scientific, Wilmington, DE) and RNA integrity was analyzed by the 2100 Bioanalyzer (Agilent Technologies).

Western blot analysis was performed as previously described [10, 11]. The protein concentration was determined using a Bradford assay (Bio-Rad) and equal amounts of proteins were separated by SDS-PAGE gel and transferred to nitrocellulose membrane (Bio-Rad). The membranes were probed with primary antibodies followed by incubation with HRP-conjugated secondary antibodies. The following antibodies: EGFR monoclonal antibody (#4267), pEGFR monoclonal antibody (#3777), E-cadherin, monoclonal antibody (#3195), STAT3 monoclonal antibody (#4904), pSTAT3 monoclonal antibody(#9145), AKT policlonal antibody (#9272), pAKT monoclonal antibody (#4060), Vimentin monoclonal antibody (#5741), PD-L1 monoclonal antibody (#13684), p44/42 MAPK polyclonal antibody (#9102), phospho-p44/42MAPK monoclonal antibody (#9106), MEK 1/2 monoclonal antibody(#4694), pMEK1/2 monoclonal antibody (#9154), Cleaved Caspase 9 antibody (#9505), Caspase 3 antibody (#9662), SNAIL monoclonal antibody (#3879), SLUG monoclonal antibody (#9585) and Bcl-2 monoclonal antibody (#4223) were from Cell Signaling (Beverly, MA, USA). Mouse PD-L1 antibody (#ABM4E54) was for Abcam (Cambridge, UK). Monoclonal anti-α-tubulin antibody (T8203) was from Sigma Chemical Co. (St. Louis, MO, USA). Secondary antibodies goat anti-rabbit IgG and rabbit anti-mouse IgG were from Bio-rad (Hercules, CA, USA). Immunoreactive proteins were visualized by enhanced chemiluminescence (ECL plus, Thermo Fisher Scientific, Rockford, IL, USA). Each experiment was done in triplicate.

The cells were fixed with 4% paraformaldehyde for 10 min, permeabilized in 0.5% Triton X-100 for 10 min and blocked in PBS supplemented with 3% BSA for 30 min. After each step, the cells were rinsed in PBS, incubated for 1 h at room temperature with primary antibodies anti-EGF Receptor (Abcam), anti-PD-L1(Cell signaling) followed by incubation with secondary antibodies Alexa Fluor 488 goat anti-mouse IgG (ThermoFisher) or Alexa Fluor 532 goat anti-rabbit IgG (ThermoFisher) for 30 min at RT. Finally, the coverslips were incubated for 15 min at 37 °C with 4′,6-Diamidino-2-phenylindole dihydrochloride (DAPI) and then examined under the fluorescence confocal microscope Zeiss LSM 700 (Zeiss, Oberkochen, Germany).

Cells were lysed in Buffer B2 (Hepes pH 7.9 10 mM, EDTA 1 mM, KCl 60 mM, DTT 1 mM) supplemented with protease inhibitor, phosphatase inhibitor and 0.2% NP-40. Centrifugation at 3000×g for 15 min and the supernatant was centrifuged further at 13.000×g for 15 min to obtain the cytosolic fraction. The nuclear pellet was resuspended in buffer containing 60% sucrose and separated by centrifugation at 6000×g for 10 min. Nuclei were dissolved in Buffer B4 (Tris-HCl pH 7.8250 mM, KCl 60 mM, DTT 1 mM) supplemented with protease inhibitor and phosphatase inhibitor. Nuclei were dissolved using alternated temperatures (− 80 °C and 37 °C) and clarified by centrifugation at 9500×g for 10 min.

C57BL/6 N mice and BALB/c mice from Charles River were maintained in the animal facilities of MDACC following standard animal regulation and strict health controls. 1 × 10⁵ MC38-MR and 3 × 10⁵ CT26-MR cells were suspended in 200 μl of Matrigel (BD Biosciences, Milan, IT): PBS (1:1) and were subcutaneously injected to the right flank of C57BL/6 and BALB/c mice respectively. Two weeks after injections when tumors were between 200 and 300 mm³, mice were randomly assigned to one of the following groups (ten mice per group): group 1: vehicle, administrated intraperitoneally (i.p.); group 2: MEKi (BAY86–9766 25 mg/kg every day for 5 days a week, by oral gavage); Group 3: EGFRi (Erltonib 10 mg/Kg very day for 5 days a week, by oral gavage); Group 4: PD-L1i (clone 10F.9G2 injected twice a week i.p. at total dose of 200 μg/mouse); Group 5: MEKi plus PD-L1i at same doses mentioned above; Group 6: MEKi plus PD-L1i plus EGFRi at same doses mentioned above. The treatment was continued for 3 weeks and mice were followed for additional 5 weeks after end of treatment. When tumors reached approximately 2.000 mm³ mice were euthanized. In all experiments, mice body weights were monitored daily. Tumor size was evaluated twice a week by calliper measurements using the following formula: π/6 × larger diameter × (smaller diameter)². For assessment of tumor response to treatment, we used volume measurements and adopted a classification methodology loosely inspired by clinical criteria: (i) tumor regression (or shrinkage) was defined as a decrease of at least 50% in the volume of target lesions, taking as reference the baseline tumor volume; (ii) at least a 35% increase in tumor volume identified disease progression; and (iii) responses that were neither sufficient reduction to qualify for shrinkage or sufficient increase to qualify for progression were considered as disease stabilization.

Results were expressed as mean ± SD, and n refers to the number of sample replicates. The statistical differences between the means were determined by one-way ANOVA followed by Tukey’s multiple comparison tests with Prism software (version 6.01; GraphPad, USA). p < 0.05 was considered to be statistically significant. The false discovery rate (FDR) was applied as a multiple test correction method. Average gene expression values between experimental groups were compared (on log scale) by means of a modified ANOVA (p < 0.05). Genes in the resistant group were identified as being differentially expressed compared to parental cells if they had a fold-change (FC) in expression of at least 3 and a p-value < 0.05.

In previous works, we have demonstrated that MEK is a key downstream effector of EGFR pathway that must be inhibit to prevent and/or delay the onset of acquired resistance to anti-EGFR treatment [3, 10, 11, 18‐21]. Nevertheless, we have found that some tumors after an initial benefit to MEKi treatment, started to regrowth limiting its use [11]. In order to understand the mechanism underlying MEKi resistance, we have generated MEKi resistant cell line, SW48-MR as described in Material and Methods. To investigate the potential molecular pathways involved in MEKi-resistance mRNAs from SW48 and SW48-MR cells were extracted and assessed for global gene expression changes by microarray analysis (Fig. 1). We conducted gene set enrichment analyses (GSEA) using previously described signatures of pathway activity and well-characterized cellular processes to characterize the molecular pathways specific for this MEKi resistant model. Among the genes that were up-regulated in SW48-MR versus SW48 cells we have identified several genes involved in the PD-L1 pathway. In particular PD-L1 gene was up-regulated in the resistant cells compare to parental cells (Fig. 1). Moreover, genes overexpressed in MEKi resistant cells were functionally related in pathways involving immune cell activation, inflammation, and antigen processing and presentation such as Programmed Cell Death 1 (PDCD1), CD86 molecule (CD86), CD8 molecule (CD8), Pore Forming protein 1 (Perforin 1), Interferon Regulatory Factor 1 (IRF1), Cytotoxic and regulatory T cell molecule (CRTAM) (Fig. 1). Moreover our analysis revealed an increase levels of genes known to be involved in mesenchymal progression, EMT transition and in matrix modeling such as Transforming Growth Factor beta 1 (TGFB1), Transforming Growth Factor beta activated kinase 2 (TAB2), Transforming Growth Factor beta kinase 3 (TAB3), Transforming Growth Factor alpha (TGFA), Signal Transducer and Activator of Trascription 1 (STAT1), Signal Transducer and Activator of Trascription 2 (STAT2), Signal Transducer and Activator of Trascription 3 (STAT3), Signal Transducer and Activator of Trascription 4 (STAT4), Janus kinase 3 (JAK3), SMAD specific E3 ubiquitin protein ligase (SMURF), SMAD family member (SMAD4), Twist homolog 1 (Drosophila) (TWIST1), Vimentin (VIM), SNAIL family zinc Finger 1 (SNAI1), SNAIL family zinc Finger 2 (SLUG), SNAIL family zinc Finger 3 (SNAI3), Matrix Metallo Peptidase 9, 28, 7, 1, 19 (MMP9, MMP28, MMP7, MMP1, MMP19), intercellular adhesion molecule 4 (ICAM4), elastin microbril interfacer 2 (EMILIN2). These results demonstrate an enhanced immune-reactive microenvironment in MEKi resistant cells compared with parental ones. In fact, high expression of genes specific to Treg cells, myeloid-derived suppressor cells (MDSCs), monocyte-derived cells and TH17 cells, present in our MEKi resistant tumor are typically seen in the microenvironment of immune-tolerant malignancies (Fig. 1). This ‘inflamed’ immune profile was similar to CMS4 subtype of CRC that is characterized by marked up-regulation of immune suppressive factors, such as Transforming Growth Factor Receptor beta (TGFβ) and Chemokine (C-X-C motif)12 (CXCL12), and high expression of genes encoding chemokines that attract myeloid cells, including C-C motif chemokine ligand 2 (CCL2) and the related cytokines IL-23 and IL-17 [14‐17]. Moreover, CMS4 tumors were characterized by activation of pathways related to EMT and steaminess, such as TGFβ pathway members and integrin, and showed marked overexpression of proteins implicated in extracellular matrix remodeling and complement signaling [14‐17]. All the genes correlated to these signatures were overexpressed in our MEKi resistant cells, confirming the correlation between our MEKi resistant cell lines and CMS4 tumors (Fig. 1).

Other less well-known genes involved in these processes were also up-regulated: Platelet-derived Growth Factor Receptor (PDGFRB), Fibroblast Growth Factor 3 (FGF3), Fibroblast Growth Factor 13 (FGF13), Insulin-like growth factor binding protein 3 (IGFBP3), Insulin-like growth factor binding protein 7 (IGFBP7). In addition, all EGFR ligands are up-regulated in the resistant cells compared to parental ones, underlying a role of EGFR activation in MEKi resistance landscape (Fig. 1).

To further expand these observations, we have selected other two CRC cell lines sensitive to MEK blockade, such as HCT116 (KRAS G13D) and LIM1215 (all RAS WT), to generate by an in vitro selection new models of acquired resistance to MEKi (Fig. 2, Additional file 1: Figure S1). After the establishment of MEKi resistant cell lines we have characterized the resistant phenotype by cell proliferation analysis using a 3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide (MTT) and by 5-bromo-2-deoxyuridine (BrdU) incorporation assay, in the presence of MEKi. To assess whether gene expression–based subtypes are recapitulated at the protein level, we performed a western blot analysis and we found that all the pathways that are up regulated in CMS4 subtype were up-regulated in these resistant cell lines (Fig. 2, Additional file 1: Figure S1). As depicted in Fig. 2, moving forward the MEKi phenotype we assisted to the loss of epithelial features and acquisition of mesenchymal markers and morphology. In particular, we showed an up-regulation of SLUG, SNAIL and Vimentin and down regulation of E-cadherin in the resistant cells compared to parental ones. Moreover, this change in the morphology is accompanied by up-regulation of PD-L1 expression and activation of EGFR and its downstream pathway in the resistant models (Fig. 2, Additional file 1: Figure S1).

Understanding molecular mechanisms that modulate PD-L1 expression would likely lead to improve treatments for patients. The regulation of PD-L1 expression is likely complex and is regulated by multiple signaling pathways, some that have been investigated including the JAK/STAT (Janus kinase signal transducer and activation of transcription) [21, 22]. Since in our MEKi resistant model we have found an up-regulation of STAT3 gene, we investigated the mechanism of its activation (Fig. 1). STAT3 can be activated by several secreted factors, cytokines and different tyrosine kinases receptors (TKRs) including EGFR activation [23]. To evaluate the role of STAT3 in the PD-L1 modulation and subsequently in mediating MEKi resistance and to understand its correlation with EGFR we have first modulated the expression of STAT3 by using a siRNA technology in the MEKi resistant cell line (HCT116-MR), and then we have assessed the PD-L1 and EGFR expression by performing western blot analysis (Fig. 3 a-b). Transfection with a specific STAT3 siRNA significantly reduced PDL1 and EGFR mRNA and protein expression in HCT116-MR cells at 72 and 48 h respectively, as shown in Fig. 3 a-b. These data confirmed a direct correlation between STAT3, EGFR and PD-L1 expression. Moreover, STAT3 silencing also restored the sensitivity of HCT116-MR to the inhibitor effect of MEKi, as shown in Fig. 3 c-d. To further determine if PD-L1 activation could be involved in the acquisition of MEKi resistance in HCT116-MR cells, we investigated whether reduction of PD-L1 expression could modulate STAT3 and EGFR expression (Fig. 4). The inhibition of PD-L1 expression by siRNA determined a significant reduction of STAT3 and EGFR expression, underlying the strong correlations among these three factors and their potential role in the development of resistance to MEKi (Fig. 4).

PD-L1 is generally a membrane protein and its localization could change in different caner type [24]. Several groups have already demonstrated that mislocalization of it in the nucleus could promote drug resistance and is associated with short survival durations [24, 25].

With this aim, to determine the location of the PD-L1 and EGFR protein in our resistant models, we observed the staining pattern of anti-PD-L1 and anti-EGFR affinity-purified antibodies by indirect immunofluorescence. HCT116 and HCT116-MR cells were plated on slides, serum starved overnight and then treated with TGFα at different time point (5 min, 20 min and 1 h). As depicted in Fig. 5 a, in HCT116-MR cells treatment with TGFα increased the expression of PD-L1 in the nucleus, this effect is more evident at 5 min time point (Fig. 5a). To better characterize this result, we studied the distribution of PD-L1 and EGFR on the surface cells by flow cytometry in HCT116-MR cells. According to immunofluorescence staining after TGFα treatment and also after treatment with MEKi at dose of 0,5 μM the expression of PD-L1 on cell-surface decreased as well as expression of both PD-L1 and EGFR on cell-surface compared to control cells (Fig. 5b). The same observations were made using confocal microscopy after Z-stacks images; PD-L1 is more expressed in the HCT116-MR cells on the cell surface but above all the presence of the protein is evident in the nucleus (Fig. 5c).

To confirm the presence of PD-L1 in nuclear compartment, nuclear and cytoplasmic extracts were performed in HCT116 and HCT116-MR cell line. Western blot analysis revealed that PD-L1 was predominantly in the nuclear fraction, as also p-EGFR and p-STAT3. Histone H3 and tubulin were used as nuclear and cytoplasmic controls, respectively (Fig. 5d).

Based on these findings, PD-L1 expression in MEKi resistant cells is more evident in the cell nuclear compartment. This effect could explain also anti-apoptotic activity present in the MEKi resistant cell line compared to parental one. We tested expression of the anti-apoptotic components by Western blot analysis. As shown in Additional file 1: Figure S1B, we observed a strong increasing in expression levels of BcL2 in HCT116-MR cells, suggesting a probable activation of anti apoptotic mechanisms in this resistant cell line.

To extend these in vitro findings, we performed an in vivo study using MC38 and CT26 MEKi resistant syngeneic models. MC38 (all RAS WT) and CT26 (KRAS G12D) are the only two syngeneic colon cancer cells available. These two cell lines should recapitulate the whole scenario of acquired resistance mechanisms to MEKi in both WT and RAS mutated contest. We generated MEKi resistant cells starting from MEKi sensitive MC38 and CT26 syngeneic colon cancer cells, which were grown as tumor xenografts in in vivo C57BL/6 and BALB/c immune-competent mice that were exposed to continuous treatment with an optimal therapeutic dose of the drug. Continuous treatment caused tumor growth suppression within the first weeks of treatment (Additional file 2: Figure S2A). However, after approximately 12 weeks, in most mice tumors resumed growth despite MEKi treatment, reaching a growth rate comparable to untreated control MC38 and CT26 tumors within 18–20 weeks (Additional file 2: Figure S2A). MEKi resistant tumors were surgically removed and homogenized into single-cell suspensions used to generate in vitro MC38 and CT26 MEKi resistant cell lines, which displayed resistance to the growth inhibitory effects of MEKi treatment (Additional file 2: Figure S2B). These models are a valuable tool to provide additional information on the potential for clinical activity of immune therapies, and could also improve the understanding of immune response in mCRC. MC38 and CT26 MEKi resistant colon cancer cells have been injected subcutaneously in C57BL/6 and BALB/c mice respectively. Mice have been treated with MEKi, EGFR inhibitor (EGFRi), murine PD-L1 inhibitor (PD-L1i) and their combination (Fig. 6a). We used erlotinib a tyrosine kinase inhibitor as EGFRi that is able to recognize and inhibit the mouse EGFR. Treatments were started 2 weeks after the injection of tumor cells and lasted for 3 weeks. After the end of treatment mice were followed for additional 5 weeks. At the end of 3 weeks treatment, all mice treated with vehicle, MEKi or EGFRi alone reached the maximum allowed tumor size of 2000 mm³. Among the single agent treatments, the group treated with PD-L1i showed the greatest tumor growth inhibition in both xenografts. As shown in Figs. 6a, both combined treatment regimens markedly suppressed tumor growth compared with vehicle, MEKi alone, and EGFRi alone. Combined treatment of MEKi, EGFRi and PD-L1i resulted in a marked inhibition of tumor growth in both MC38–MR and CT26-MR xenograft models. Indeed, at the end of treatment period, the triplet combination treatment caused a more evident synergistic effect, with no palpable tumor present in 4 out of 10 mice engrafted with MC38-MR cells and 3 out of 10 mice engrafted with CT26-MR cells (Fig. 6a). The synergistic effect of triplet combination treatment was also more marked at the end of observation period elucidating the long-term effect of this combination. This project reveals a strategy to potentially improve the efficacy of MEK inhibition by co-treatment with EGFR and PD-L1 inhibitors via modulation of host immune responses.

To understand whether the synergistic anti-tumor activity obtained by the combined treatment with MEKi, EGFRi and PD-L1i was due to a more effective inhibition of key intracellular signals for cell survival and proliferation, tumors were collected at the end of the treatment from mice engrafted with both MC38-MR and CT26-MR cell lines. One mouse per arm was sacrificed from MEKi, EGFRi, PD-L1i, MEKi plus PD-L1i and MEKi plus PD-L1 plus EGFRi treatment groups. As control, we used one mouse engrafted with both resistant cell lines that has not undergone any type of treatment. First, we assessed the phosphorylation status of EGFR and its downstream effectors AKT and MAPK by western blot analysis. As shown in Fig. 6b, in tumor specimens treated with MEKi EGFRi and PD-L1i alone no change or partially inhibition in expression of pEGFR, pAKT and pMAPK was observed. The combined treatment with MEKi and PD-L1i substantially inhibited phosphorylation of EGFR, MAPK and AKT compared to single agent treatments (Fig. 6b). In particular the expression of pMAPK was almost completely ablated only in the triplet treatment arm in both MC38-MR and CT26-MR models underling the synergistic effect of this combination. Moreover, also PD-L1 expression was totally abolished only in the triplet treatment group while no change or partial inhibition of its expression was observed in the other treatment group.