Cellular and Molecular Life Sciences (2022) 79:524 https://doi.org/10.1007/s00018-022-04541-6 Cellular and Molecular Life Sciences ORIGINAL ARTICLE The ERK5/NF‑κB signaling pathway targets endometrial cancer proliferation and survival Nora Diéguez‑Martínez1,2 · Sergio Espinosa‑Gil1,2 · Guillermo Yoldi1 · Elisabet Megías‑Roda2,3 · Idoia Bolinaga‑Ayala1,2 · Maria Viñas‑Casas1,2 · Gokhan Gorgisen1,4 · Inés Domingo‑Ortí1 · Héctor Pérez‑Montoyo3 · Jose R. Bayascas1 · Eva Colas5 · Xavier Dolcet6 · Jose M. Lizcano1,2 Received: 23 May 2022 / Revised: 20 August 2022 / Accepted: 30 August 2022 / Published online: 19 September 2022 © The Author(s) 2022 Abstract Endometrial cancer (EC) is the most common type of gynecologic cancer in women of developed countries. Despite surgery combined with chemo-/radiotherapy regimens, overall survival of patients with high-risk EC tumors is poor, indicating a need for novel therapies. The MEK5-ERK5 pathway is activated in response to growth factors and to different stressors, including oxidative stress and cytokines. Previous evidence supports a role for the MEK5-ERK5 pathway in the pathology of several cancers. We investigated the role of ERK5 in EC. In silico analysis of the PanCancer Atlas dataset showed altera- tions in components of the MEK5-ERK5 pathway in 48% of EC patients. Here, we show that ERK5 inhibition or silencing decreased EGF-induced EC cell proliferation, and that genetic deletion of MEK5 resulted in EC impaired proliferation and reduced tumor growth capacity in nude mice. Pharmacologic inhibition or ERK5 silencing impaired NF-kB pathway in EC cells and xenografts. Furthermore, we found a positive correlation between ERK5 and p65/RELA protein levels in human EC tumor samples. Mechanistically, genetic or pharmacologic impairment of ERK5 resulted in downregulation of NEMO/ IKKγ expression, leading to impaired p65/RELA activity and to apoptosis in EC cells and xenografts, which was rescued by NEMO/IKKγ overexpression. Notably, ERK5 inhibition, MEK5 deletion or NF-kB inhibition sensitized EC cells to standard EC chemotherapy (paclitaxel/carboplatin) toxicity, whereas ERK5 inhibition synergized with paclitaxel to reduce tumor xenograft growth in mice. Together, our results suggest that the ERK5-NEMO-NF-κB pathway mediates EC cell prolifera- tion and survival. We propose the ERK5/NF-κB axis as new target for EC treatment. Keywords Map kinase · ERK5 · NF-kB · Apoptosis · Endometrial cancer · Anticancer drug * Jose M. Lizcano josemiguel.lizcano@uab.es; Jose.lizcano@vhir.org Introduction 1 Departament de Bioquímica i Biologia Molecular, Unitat Endometrial cancer (EC) is the most common type of de Medicina, and Institut de Neurociències, Universitat Autònoma de Barcelona (UAB), 08193 Barcelona, Spain gynecologic cancer and the fourth most frequent neopla- 2 sia in women of developed countries [1]. EC originates Protein Kinases in Cancer Research, Vall Hebron Institut de Recerca (VHIR), Edifici Collserola, Laboratory 143, Passeig from the uterine endometrial epithelium and is classified de la Vall d’Hebron, 119‑129, 08035 Barcelona, Spain into two different histological subtypes [2]. Endometrioid 3 Ability Pharmaceuticals, SL. Cerdanyola del Vallès, EC accounts for the majority of ECs. These tumors express 08290 Barcelona, Spain estrogen and/or progesterone receptors, which are used as 4 Department of Medical Genetics, Faculty of Medicine, Van targets for hormonal therapy. Non-endometrioid EC (NEEC) Yüzüncü Yıl University, Van, Turkey represents ~ 20% of cases and comprises serous and clear cell 5 Biomedical Research Group in Gynecology, Vall Hebron carcinomas, among others. Non-endometrioid EC tumors Institute of Research, Universitat Autònoma de Barcelona, tend to be estrogen-independent and show enhanced aggres- CIBERONC, 08035 Barcelona, Spain siveness and worse prognosis than endometrioid ECs [2]. 6 IRB Lleida, Oncologic Pathology Group, Departament de Advanced and hormone-independent ECs are currently Ciències Mèdiques Bàsiques, Institut de Recerca Biomèdica treated with systemic cytotoxic therapy with carboplatin/ de Lleida, IRBLleida, Universitat de Lleida, CIBERONC, paclitaxel [3]. However, its clinical effectiveness is poor 25198 Lleida, Spain 13 Vol.:(0123456789) 524 Page 2 of 20 N. Diéguez‑Martínez et al. in advanced, recurrent or metastatic EC tumors, indicat- 10% (rest of cells) fetal bovine serum (FBS, Gibco) and 1% ing a need for novel molecular therapies. Since EC tumors Penicillin/Streptomycin (Gibco). Cells were maintained at frequently concur with activating mutations of PI3KCA or 37 °C in a humidified atmosphere containing 5% ­CO2. Cell AKT, the PI3K-AKT-mTOR pathway has been proposed for lines were transfected using Lipofectamine 2000™ (Ther- EC targeted therapies. Unfortunately, EC tumors frequently moFisher Scientific, Waltham, MA, USA) as described develop resistance to PI3K, Akt, or mTOR inhibitors, mainly before [11]. due to hyper-activation of the Ras-MEK1 pathway [4]. Hence, there is an urgent need for new targeted therapies to Cell lysis and immunoblotting improve treatment of high-risk EC. The extracellular regulated kinase 5 (ERK5) is the most Cells were lysed in ice-cold RIPA buffer supplemented recent discovered member of the conventional MAP kinase with 1 mM sodium-orthovanadate, 50 mM NaF and 5 mM family. ERK5 is activated by its unique MAP2K MEK5, sodium-pyrophosphate, sonicated and stored at − 20 °C. in response to both mitogens and different forms of stress, Subcellular fractionation was performed following standard including oxidative stress and cytokines. These effectors lead procedures, as described before [12]. Proteins were resolved to the activation of the MEK5 upstream activators MAP3K2 in SDS-PAGE gels and electrotransferred onto a nitrocel- and MAP3K3 [5]. ERK5 is involved in the proliferation, lulose membrane (Sigma-Aldrich). After incubation with survival, apoptosis, and differentiation of different cell types the appropriated primary, detection was performed using [6]. Recently, ERK5 was shown to play an important role horseradish peroxidase-conjugated secondary antibodies and in the tumorigenesis, proliferation and survival of several enhanced chemiluminescence reagent (Bio-Rad). Primary cancer types [7, 8]. Because ERK5 inhibition or silencing antibodies used are given in Supplementary Table S1. compromises the viability of numerous cancer cell lines and tumor xenografts, ERK5 is emerging as a new target for DNA constructs molecular anticancer therapies [9]. In this work, we have investigated for the first time the The pEBG2T vector encoding GST-tagged human ERK5 role of the MEK5-ERK5 pathway in EC. We also report the was previously described [12]. The pcDNA3 vector encod- preclinical development of the small-molecule JWG-071, a ing for SR-IkBα was from J.X. Comella (VHIR Barcelona, selective ERK5 inhibitor with a good pharmacokinetic pro- Spain) [13]. pGL4-AP1 (containing six copies of an AP1- file. We present evidence supporting ERK5 inhibition as a response element fused to luciferase gene) and PGLA4.32- targeted therapy with great potential to treat EC. NF-κB (five copies of a NF-κB response element) vectors were from Promega (Madison, WI, USA). Materials and methods shERK5 lentiviral production Reagents Stable and efficient silencing of endogenous ERK5 protein was achieved using lentivirus. Two pLKO.1 lentiviral vec- JWG-071 was in-house synthesized as previously described tors encoding for specific shRNA for two human MAPK7 [10]. AX-15836, BAY-117082, paclitaxel and carboplatin mRNA sequences were used (TRCN0000010262/pLKO.1 were obtained from MedChem Express (Monmouth Junc- and TRCN0000010275/pLKO.1, Sigma). Lentiviral parti- tion, NJ, USA). Cisplatin, staurosporine and EGF were from cles were generated in HEK-293 cells by co-transfecting Sigma-Aldrich (Germany). Q-VD(OMe)-OPh was from the virion vectors (psPAX2 and pMD2G) and the pLKO.1 APExBio (Gentaur, Spain). ERK5-shRNA vectors. After 4 h, the medium was replaced with fresh medium. Forty-eight h post transfection, the Cells and cell culture medium containing the viral particles was collected, centri- fuged, filtered to remove cell debris, and stored at − 80 °C AN3CA and ARK1 EC cells and Hela cervical cancer cells until use. Lentiviral production is detailed at Supplementary were purchased from ATCC (Manassas, VA, USA). Ishikawa Materials and Methods. cells were from ECACC and were purchased from Sigma. T-large antigen-transformed ­BAX−/−/BAK−/− MEFs were Gene silencing generously provided by Dr Victor Yuste (UAB, Spain). HeLa, ARK1 and MEF cells were maintained in Dulbecco’s Validated siRNAs targeting NEMO were from Sigma (ref. modified Eagle’s medium (DMEM ThermoFisher). AN3CA EHU032271) and ThermoFisher (ref. AM51331). Vali- were maintained in DMEM-F12 medium (ThermoFisher). dated siRNAs targeting ERK5 were from ThermoFisher Medium was supplemented with 5% (Ishikawa cells) or (ref. 4390825 and 4392420). Scrambled (control) siRNA 13 The ERK5/NF‑κB signaling pathway targets endometrial cancer proliferation and survival Page 3 of 20 524 was from ThermoFisher. siRNAs were transfected into cells with 4% PFA, 0.1% NP40 in PBS, and nuclei were stained using Lipofectamine-2000, following the manufacturer’s with 1 μg/mL Hoechst 33258 (Invitrogen). recommended protocol. Quantitative real‑time PCR Human EC tumor samples Total RNA was isolated from cells using RNeasy kit (Qia- Patients with no previous therapy (50–80 years) underwent gen). cDNA was obtained using iScript™ cDNA Synthe- surgery for endometrial carcinoma at Hospital Vall d’Hebron sis Kit (Bio-Rad). Real-time quantitative PCR assays were (Barcelona, Spain). The protocol was approved by the Insti- performed using TaqMan Gene Expression Master Mix and tutional Review Boards, and informed consent was obtained the probes human NEMO/IKBKG (Hs00415849_m1) and from the patients involved. After surgery, endometrial nor- human GAPDH (Hs03929097_g1) (ThermoFisher Scien- mal tissue and tumor samples were collected from each tific). Amplifications were run in a Bio-Rad CFX96 real- patient and stored frozen (− 80 °C) until analysis. Samples time PCR system, using the following protocol: 50 °C for were lysed in RIPA buffer containing protease and phos- 2 min, 95 °C for 10 min, 39 cycles of 95 °C for 15 s, 55 °C phatase inhibitor cocktails (Sigma), and stored at -20˚C until for 1 min. Each value was normalized to GAPDH levels. use. Relative expression levels were determined using the ­2−ΔΔCt method. Real-time PCR was controlled by the Bio-Rad CFX Manager v 3.1 software. Gene reporter Luciferase assay Generation of MEK5 knockout cells by CRISPR/Cas9. Cells cultured in 12-well plates were transfected with 450 ng of NF-κB-driven (Stratagene) or AP 1-driven luciferase To generate CRISPR/Cas9 M ­ EK5−/− Hela or Ishikawa cells, (Promega) reporter plasmids, and 50 ng Renilla luciferase we used a kit from Santa Cruz technology, using vectors (Promega). Luciferase activity assay was monitored using containing sgRNAs (TAC​TTG​CTG​TTC​CAT​TAC​TG and the dual luciferase kit (Promega), following the manufac- CAG​ATT​CAC​TTC​CAA​GCA​AT) and the Cas9 endonucle- turer’s instructions. ase gene (sc-401688), in combination with a HDR-puromy- cin resistance vector (sc-401688). Cells were selected with 1 µg/ml puromycin (Sigma). Single colonies were picked EC xenografts up and monitored for MEK5 expression. Clones showing MEK5 null expression were selected. Athymic nude female mice (Hsd: Athymic Nude-Foxn1nu, Envigo, Spain) were injected subcutaneously with 4 × ­106 Cell viability, clonogenic and proliferation assays Ishikawa cells into the right flank. When tumors reached 80–100 ­mm3, mice were randomly distributed into groups Cell viability was determined by MTT (3-(4,5-dimethyl- and administered with the indicated treatments. JWG-071 2-thiazolyl)-2,5-diphenyl-2H-tetrazolium Blue, Sigma) was administered once a day intraperitoneally at dose of reduction assay. For clonogenic assays, cells were seeded in 50 mg/kg (monotherapy) or 30 mg/kg (combo treatment). 6-well plates, treated for 14 days and stained with 0.5% crys- Paclitaxel was administered twice a week intraperitoneally tal violet solution. Colonies with > 40 cells were counted. at dose of 15 mg/Kg. Tumors volumes were measured as Cell proliferation was measured by cell counting assay, using (length × w­ idth2)/2, thrice a week. When tumors reached a TC10TM automated cell counter (Bio-Rad, Hercules, CA, more than 1000 m ­ m3 or body weight decreased 20% of the USA) using Trypan Blue. Cell viability was also determined initial measure, mice were euthanized by cervical disloca- using a LIVE/DEAD viability/cytotoxicity kit assay (Ther- tion. Tumors were excised from mice, and a fraction of each moFisher), as reported before [11]. tumor was homogenized in RIPA buffer containing protease inhibitor cocktail (Sigma), sonicated, centrifuged to remove Flow cytometry analysis of apoptosis cell debris, and stored at − 20 °C until use. All procedures involving animals were performed with approval of the UAB Cells were seeded and treated for the indicated times. Cell Animal Experimentation Committee, according to Spanish were then trypsinized, washed sequentially with FBS and official regulations. Individual tumor growth curves are binding buffer, and incubated in binding buffer with Annexin given in Supplementary Fig. 16. V and/or propidium iodide (Invitrogen, Waltham, MA, USA) for 15 min. Samples were analyzed in a Beckman Coulter FC500 flow cytometer. For nuclear staining, cells were fixed 13 524 Page 4 of 20 N. Diéguez‑Martínez et al. A B Uterine Corpus Endometrial Carcinoma (TCGA, PanCancer Atlas) MAPK7 12% 100 100 Progression Free (%) MAP2K5 8% 90 90 Overall Survival (%) MAP3K2 10% 80 Unaltered ERK5 80 Unaltered ERK5 (n = 489) (n = 472) MAP3K3 11% 70 70 MEF2A 10% 60 60 MEF2B 10% 50 50 MEF2C 10% 40 High ERK5 40 High ERK5 30 (n = 25) 30 (n = 20) MEF2D 16% 20 20 Mutations (unknown significance) No alterations 10 Log Rank P: 0.000087 10 Log Rank P: 0.000940 Genetic Amplification mRNA High 0 0 Alteration Deep Deletion mRNA Low 0 40 80 120 160 200 240 0 40 80 120 160 200 240 Months Months Ishikawa cells AN3CA cells HeLa cells C D 50 nM EGF 50 nM EGF 50 nM EGF JWG - - 0.01 0.03 0.1 0.3 1 3 10 µM - - 0.01 0.03 0.1 0.3 1 3 10 µM - - 0.01 0.03 0.1 0.3 1 3 10 µM ERK5 p-ERK1/2 ERK1/2 JWG-071 Hsp90 JWG-071 (µM) Ishikawa cells AN3CA cells E F Veh. 1 2.5 5 7.5 10 G 100 100 100 Ishikawa *** *** *** *** % Proliferation 80 % Proliferation cells 80 80 60 AN3CA 60 60 40 Ishikawa cells 40 40 AN3CA HeLa 20 20 20 HeLa cells 0 0 0 0.1 0.3 1 3 10 20 Scr. 1 2 Scr. 1 2 JWG-071 (µM) siERK5 siERK5 Ishikawa cells HeLa cells H I $$$ 100 1.5 $ 2.0 100 Cell number (x105) % Proliferation 80 80 % Proliferation 1.0 1.5 ** *** (day 4) 60 60 ** 1.0 Ishikawa cells 40 40 Ishikawa cells 0.5 *** AN3CA cells AN3CA cells 0.5 20 20 0 0 0 10 25 50 100 250 500 0 0.01 0.02 0.1 0.3 1 3 1 µM JWG-071 - - + - - - + - LRRK2 inhibitor (nM) DCLK1 inhibitor (μM) 1 µM AX15836 - - - + - - - + EGF EGF J Veh. 50 nM EGF K JWG (3 µM) L ** act. (arbitrary units) AP1 transcriptional 15 - JWG EGF - 20 40 60 - 20 40 60 min C N C N C N ERK5 10 ERK5 c-Jun 5 CREB GAPDH Hsp90 0 MEK5 - + - + + ERK5 - - + + + JWG - - - - + Immunohistochemistry protocols. Briefly, deparaffinized samples were subjected to heat-induced antigen retrieval in citrate buffer. After Samples from tumor xenografts were dissected, formalin- blocking endogenous peroxidase with 3% ­H2O2, and reduc- fixed and paraffin-embedded as described previously [14]. ing unspecific binding with and 5% goat serum and 0,1% Three µm-thickness sections were stained with haematoxy- Triton X-100, slices were then incubated with anti-cleaved lin and analyzed by immunohistochemistry using standard caspase-3 antibody or anti-Ki67 antibody (Cell Signaling) 13 The ERK5/NF‑κB signaling pathway targets endometrial cancer proliferation and survival Page 5 of 20 524 ◂Fig. 1  ERK5 inhibition impairs EC cell proliferation. A Genomic mining from the cBioPortal for Cancer Genomics (Uterine profiles of components of the MEK5-ERK5 pathway in EC patients corpus endometrial carcinoma TCGA, PanCancer Atlas) obtained from Uterine Corpus Endometrial Carcinoma data set (TCGA, PanCancer Atlas). Data set from cBiportal (https://​www.​ showed alterations (gene copy number, mutations or mRNA cbiop​ortal.​org/). MAPK7, ERK5 gene; MAP2K5, MEK5. B Kaplan– alterations) in all components of ERK5 signaling, includ- Meier plots overall survival (left) and progression-free (right) in uter- ing the upstream kinase activators MAP3K2 and MAP3K3, ine corpus endometrial carcinoma patients with high (red), quartile the MEK5 and ERK5 genes (MAP2K5 and MAPK7, respec- 1) or normal (blue, quartile 2–3) ERK5 mRNA levels (data set from cBioportal). P values were obtained using log-rank test. C Chemical tively), and the ERK5 substrates MEF2A, MEF2B, MEF2C, structure of JWG-071. D Immunoblot analysis of the effect of JWG- and MEF2D transcription factors [16]. Of note, 48% of 071 on EGF stimulation. E MTT cytotoxicity (48 h) assay in a panel human EC patients (250 out of 517) showed alterations of of human EC cells. F 14-day clonogenic assay. G Effect of ERK5 components of ERK5 signaling, mostly associated to gene silencing on the proliferation of EC cell. ERK5 silencing was car- ried out with two different specific siRNA sequences. Scr., scramble amplification and high mRNA levels (33%, 171 out of 517) siRNA. Cell proliferation was measured at three days of culture using (Fig. 1A). Interestingly, EC is the human cancer that showed crystal violet assays. ***, P < 0.001. H LRRK2 or DCLK1 inhibition the higher percentage of cases wherein ERK5, MEK5, does not affect human EC cell proliferation. Ishikawa or AN3CA MAP3K2 and MAP3K3 are mutated (Supplementary cells were incubated for 48 h with either the specific DCLK1 inhibi- tor DCLK1-IN-1 or the LRRK2 specific inhibitor MLi-2, and cell Fig. 1). Finally, patients with high levels of ERK5 mRNA proliferation was assessed by crystal violet assay. I Effect of ERK5 expression showed lower overall survival (p = 0.000087) and inhibitors JWG-071 and AX15836 in EGF-induced cell prolifera- reduced progression-free survival (p = 0.00094) (Fig. 1B). tion. Histograms show number of cells at day 4 referred to cells at Taken together, these data suggest a role of ERK5 signaling day 0. $, P < 0.05; $$$, P < 0.001 EGF vs. vehicle. **, P < 0.01; ***, P < 0.001 ERK5i vs. EGF. J, JWG-071 blocks EGF-induced ERK5 in EC. nuclear localization. HeLa cells were pre-incubated with 3 µM JWG- 071 prior stimulation with 50 nM EGF (30 min), fractionated into ERK5 inhibition impairs basal and EGF‑induced EC the cytosolic and nuclear fractions, and protein expression analyzed cell proliferation by immunoblot. Hsp90, cytosolic marker; CREB-1, nuclear marker. K ERK5 inhibition impairs EGF-induced c-Jun expression in Ishi- kawa cells. Immunoblot analysis. L JWG-071 impairs ERK5-medi- Early work reported that ERK5 mediates serum- and EGF- ated AP-1 transcriptional activity. Hela cells transfected with vectors induced proliferation of cervical cancer HeLa cells [5], by encoding pAP-1–luciferase reporter and pRL-CMV-Renilla were sub- phosphorylating and promoting transcriptional activation of jected to a dual-luciferase reporter assay. **, P < 0.005 MEF2C and transcription of c-Jun [16]. However, recent development of ERK5 inhibitors has led to question about and further processed with secondary antibody (LSAB2 the role of ERK5 kinase activity in cancer cell proliferation horseradish peroxidase kit; Dako, Copenhagen, Denmark). [17–20]. This is due to the discovery that widely used ERK5 ATP-competitive classical inhibitors (such as XMD8-92 and Statistical analysis ERK5-IN-1) are unspecific, since they bind the acetyl-lysine domain of Bromodomain-Containing Protein-4 BRD4, a All in vitro data were assessed using one-way ANOVA fol- ubiquitous regulator of transcription and mediator of can- lowed by Bonferroni multiple comparison test or two-tailed cer cell proliferation. To address this question, we used the Student’s t test. Tumor volumes of mice were compared benzo[e]pyrimido-[5,4-b]diazepine-6(11H)-one core to using two-way ANOVA followed by Bonferroni, except for develop a new ATP-competitive ERK5 inhibitor (JWG-071, combinatorial experiment (two-way ANOVA Tukey´s test). Fig. 1C), with no BRD4 inhibitory activity [10]. JWG-071 Statistical significance between the groups was assessed with shows selectivity for ERK5 ­(IC50 90 nM, in vitro kinase the log-rank test (GraphPad software, San Diego, CA, USA). assay, Supplementary Fig. 2) over BRD4 ­(IC50 6 µM, [10]). Data in figures are presented as mean ± SD, unless otherwise In this work, we present the preclinical development of indicated. Statistical significance was set at p < 0.05. Syner- JWG-071. gism analysis was performed using the Compusyn software Endometrial carcinoma cells express EGFR, and they rep- [15]. resent a good model to investigate the role of EGF in cancer cell proliferation. Therefore, to study the role of ERK5 in basal and EGF-induced proliferation, we tested the JWG-071 Results compound in two different human endometrioid cancer cell lines that show EGFR expression: Ishikawa and AN3CA ERK5 signaling pathway is altered in EC cells. We also used HeLa cells as a control, which also express EGFR. JWG-071 potently inhibited EGF-induced To investigate the impact of the MEK5-ERK5 signaling ERK5 phosphorylation in the three cancer cell lines tested pathway in EC, we performed an in silico data analysis (active ERK5 auto-phosphorylates, resulting in a slower of the components of this pathway. Gene expression data migrating band), without affecting EGF-induced ERK1/2 13 524 Page 6 of 20 N. Diéguez‑Martínez et al. phosphorylation (Fig. 1D). Proliferation assays showed that impact of ERK5 activity on cell physiology. As expected, JWG-071 impaired cell proliferation (Fig. 1E) and colony- and in contrast to wild type cells, ­MEK5−/− Ishikawa or formation (Fig. 1F) in the cell lines tested. Furthermore, ­MEK5−/− HeLa cells did not show ERK5 activation/phos- ERK5 silencing with two different specific siRNAs also phorylation in response to EGF (Fig. 2A). Importantly, resulted in a significant impairment of basal proliferation of genetic deletion of MEK5 did not affect phosphorylation of Ishikawa and AN3CA cells, but not in cells transfected with ERK1/2, Akt or S6 proteins in response to EGF (Fig. 2A) scramble siRNA (Fig. 1G). or insulin (Supplementary Fig. 3A), and neither affected Kinase selectivity of JWG-071 was previously deter- phosphorylation of stress-associated MAP kinases p38 and mined by profiling the inhibitor against a panel of 468 JNKs in response to osmotic stress (Supplementary Fig. 3B). human kinases, using an in vitro ATP-site competition These results confirm that genetic deletion of MEK5 does binding assay. Apart of ERK5, JWG-071 also inhibited the not affect the activity of other MAP kinases or the activity activity of Doublecortin Like kinases DCLK1 and DCLK2, of the Akt-mTORC1 axis. as well as the activity of the leucine-rich repeat kinase 2 As it happened for ERK5 inhibitor JWG-071, LRRK2 [10]. To investigate whether the antiproliferative ­MEK5−/− Ishikawa and M ­ EK5−/− HeLa cells showed sig- effect of JWG-071 was due to DCLK1/2 and/or LRRK2 nificantly impaired proliferation in response to EGF in cell off-target inhibition, we used the specific inhibitor DLCK1- counting experiments (Fig. 2B). More importantly, concen- IN-1 (which targets both DCLK kinases [21]) and the trations of specific ERK5 inhibitors JWG-071 or AX15386 MLi-2 compound (which specifically targets LRRK2 [22]). that impaired EGF-induced proliferation in wild type cells, DLCK1-IN-1 or MLi-2 compounds had no effect on the pro- had no effect in the absence of MEK5 (Fig. 2C). These liferation of Ishikawa or AN3CA cells (Fig. 1H), indicating results confirm that activation of the MEK5-ERK5 path- that DCLK1/2 or LRRK2 do not mediate the cytotoxicity way is needed for EGF-induced proliferation. Regarding induced by JWG-071 in EC cells. basal proliferation, ­MEK5−/− Ishikawa cells phenocopied Next, we investigated the role of ERK5 activity in EGF- the effect of JWG-071, since they showed less proliferation induced cell proliferation. Cell counting experiments showed (Fig. 2D) and colony-formation (Fig. 2E) capacity. Finally, increased number of Ishikawa or Hela cells in response to to investigate the effect of MEK5 deletion in vivo, wild type EGF, which was significantly impaired by 1 µM JWG-071. ­MEK5+/+ and ­MEK5−/− Ishikawa cells were subcutaneously Similar results were obtained using 1 µM AX15836 (Fig. 1I), injected in athymic nude mice, and tumor growth was moni- another ERK5 specific inhibitor recently reported [17]. tored. The deletion of MEK5 caused a drastic reduction on Active ERK5 trans-locates to the nucleus, where it phospho- tumor growth in vivo (Fig. 2F), which was also evidenced rylates and activates MEF2 transcription factors, which in by a reduction on the levels of the proliferation marker Ki67 turn promote c-Jun expression and cell proliferation [16]. In (Fig. 2G). our cell models, EGF treatment resulted in a rapid (30 min) In summary, our results support the use of JWG-071 as a accumulation of active/phosphorylated nuclear ERK5, specific ERK5 inhibitor for in vitro and in vivo studies. Fur- whereas incubation with JWG-071 abolished nuclear active/ thermore, we provide evidence showing for the first time that phosphorylated ERK5 (Fig. 1J). Accordingly, the ERK5i ERK5 kinase activity mediates both, basal and EGF-induced impaired EGF-induced expression of c-Jun (Fig. 1K) and proliferation in EC. AP-1 transcriptional activity (Fig. 1L). These results suggest that ERK5 mediates EGF-induced proliferation by activating ERK5 inhibition induces apoptosis in human EC cell the MEF2C-c-Jun axis, at least in endometrial and cervical lines cancer cells that express EGFR. The above results prompted us to investigate the anticancer CRISPR/Cas9 MEK5 knockout cells show impaired activity of ERK5 inhibition in EC. To do so, we used the basal and EGF‑induced cell proliferation endometrioid cancer cell lines Ishikawa and AN3CA cells (that carry PTEN deficiency and P53 mutation), as well as To ensure that the antiproliferative effect of JWG-071 in the more clinically aggressive non-endometrioid serous ade- endometrial and cervical cancer cells expressing EGFR was nocarcinoma cell line ARK1 (PI3KCA mutation). In cell due to inhibition of ERK5, we analyzed the effect of MEK5 viability assays, JWG-071 induced cytotoxicity in the three deletion by generating stable CRISPR/Cas9 MEK5 knock- EC cell lines tested, with I­ C50 values ranging 2–3 µM. In out Ishikawa and HeLa cell lines. MEK5-ERK5 constitute a fact, JWG-071 induced cytotoxicity in a wide panel of can- unique signaling module, since MEK5 is the only upstream cer cells, including prostate, cervical, neuroblastoma, and activating kinase for ERK5, and MEK5 has ERK5 as its only glioblastoma cancers (Supplementary Fig. 4). known substrate [23]. Therefore, ­MEK5−/− cells lack ERK5 Several authors have reported that the unspecific ERK5 kinase activity and represent a useful tool to investigate the inhibitor XMD8-92 induces apoptosis in different models 13 The ERK5/NF‑κB signaling pathway targets endometrial cancer proliferation and survival Page 7 of 20 524 A B C D E F G Fig. 2  MEK5 genetic deletion impairs EC cell proliferation. A impaired proliferation, and impaired number of colonies. E 14-day CRISPR/Cas9 ­MEK5−/− cells show impaired ERK5 activation in clonogenic assay. F, In vivo tumor growth of subcutaneous injected response to EGF stimulation. Cells were treated with 50 nM EGF and ­MEK5+/+ (n = 8) and ­MEK5−/− (n = 7) Ishikawa cells. Data are pre- expression of proteins was monitored by immunoblot. B CRISPR/ sented as mean ± SEM. **, P < 0.01; ****, P < 0.0001, two-way Cas9 ­MEK5−/− cells show impaired EGF-induced proliferation. His- ANOVA Bonferroni. Pictures show each of the tumors at final day togram shows the number of cells at day 4 referred to cells at day of treatment (21 days). G, Immunohistochemical analysis of Ki67 0. C ERK5 inhibitors do not impair EGF-induced proliferation in ­ EK5+/+ and M expression in M ­ EK5−/− Ishikawa cell tumors. Right ­MEK5−/− HeLa cells. $$$, P < 0.001 EGF vs. vehicle. ***, P < 0.001 histogram shows the corresponding quantification. B, D, E, G, ***, ERK5i vs. EGF. D CRISPR/Cas9 M ­ EK5−/− Ishikawa cells show P < 0.001 13 524 Page 8 of 20 N. Diéguez‑Martínez et al. A Vehicle JWG-071 B *** 4.9% 48.9% Control JWG-071 Ishikawa cells % Apoptotic nuclei 40 Ishikawa cells << << 20 1.8% 6.9 % 0 Veh. JWG < < < < < < < < Vehicle JWG-071 ** Propidium Iodide (PI) % Apoptotic nuclei 1% 33% AN3CA cells AN3CA cells 40 << 20 0.6% 15.2% 0 Veh. JWG < < < < < < < < Vehicle JWG-071 ** ARK1 cells < % Apoptotic nuclei < Q1 3.9% < Q2 Q1 28.4% Q2 40 ARK1 cells 20 0 < Q3 11.1% < Q4 < Q3 7.5% 1, antagonism; CI = 1, summary effect; CI < 1, synergism). RELA activity and apoptosis, which was rescued when B 14-day clonogenic assay. Left, representative images of single-cell clone proliferation, stained with crystal violet. Right, quantification of NEMO/IKKγ was transiently overexpressed. Although number of colonies. C MEK5 deletion sensitizes EC cells to chemo- we found that ERK5 inhibition or silencing significantly therapy. Cell viability (MTT) assay of MEK5+/+ and MEK5−/− Ishi- reduced NEMO/IKKγ mRNA levels, we cannot discard kawa cells. A–C *, P < 0.05; **, P < 0.005; ***, P < 0.001. D ERK5i a role of ERK5 in regulating NEMO/IKKγ stability, as sensitizes EC tumor xenografts to chemotherapy. Athymic nude mice bearing EC tumor xenografts (Ishikawa cells) were treated with described for other ERK5 substrates such as KLF2 [48]. vehicle, 30 mg/Kg JWG-071 (i.p., daily), 15 mg/Kg paclitaxel (i.p., Further work will be necessary to establish if ERK5 phos- BIW) or combo. Values represent the mean ± SEM of 6 mice for each phorylates NEMO/IKKγ, and the implications on the sta- group. Low panel shows the corresponding scatter plot for tumors at bility of this protein. day 15 of treatment. **, P < 0.05; ****, 0.0001, two-way ANOVA Tukey. E Tumors were collected and cleaved caspase-3 was evaluated We also provide evidence showing, for the first time, by immunohistochemical analysis. Right histogram shows the corre- that the NEMO-IKKβ-p65 axis mediates EC cell survival. sponding quantification. $, P < 0.05 combo treatment vs. single treat- Hence, NEMO silencing, IKKβ inhibition, or p65 inhibition ments. * P < 0.05; ** P < 0.001 ERK5i or paclitaxel vs. vehicle (by the SR-IkB repressor) resulted in apoptotic death of EC cells. In all the cases, we observed activation of the JNK- cells, [40], and ERK5 inhibition sensitizes these cells to anti- Bim apoptotic pathway in response to canonical NF-kB inhi- HER2 treatments [41]. 20%-40% of patients with advanced bition, as it has been previously reporter [28, 29]. Interest- non-endometrioid EC show HER2 overexpression [42], and ingly, MEK5 genetic deletion also resulted in activation of Trastuzumab added to standard chemotherapy improved sur- the JNK-Bim apoptotic pathway, supporting a role for ERK5 vival of patients with ­HER+ aggressive serous EC, according in regulating NF-kB canonical activity. However, the spe- to a recent phase 2 study [43]. Thus, it will be relevant to cific JNK inhibitor JNK-IN-8 did not rescue the cytotoxic- investigate whether ERK5 also mediates HER2 signaling in ity induced by the ERK5 inhibitor (Supplementary Fig. 15), aggressive EC, and whether ERK5 inhibition could be of indicating that other pathways might contribute to the cyto- help to sensitize these cells to anti-HER2 therapies. toxicity induced by the ERK5i. In this regard, we have The NF-κB pathway is a key regulator of proliferation and recently described that ERK5 inhibition induces cytotoxic survival of many types of cancers, by controlling the expres- autophagy in human pancreatic cancer cells [11]. Hence, it sion of a plethora of genes involved in proliferation, angi- will be important to investigate whether impairment of the ogenesis, inflammation, and survival, among others [44]. ERK5 pathway also results in cytotoxic autophagy in EC Early work suggested a role for the NF-κB pathway in endo- cells and tumors. metrial carcinogenesis [27]. This was later confirmed by an Our results point to NF-κB as an interesting target for IHC study that found increased nuclear NF-κB expression EC treatment, as it has been proposed for other cancer types in patients with EC, compared with benign and hyperplasic [44]. However, several IKKα or IKKβ inhibitors have failed endometrial tissues, and suggested a role for NF-κB path- in clinical trials, mainly due to toxicity concerns [49]. This way in the development of malign endometrial changes [45]. was probably due to the fact that these inhibitors block the These observations are in agreement with those reported activation of the IKK complex (IKKα /IKKβ, necessary for for other gynecologic cancers, such ovary [46] and cervical the activation of the canonical NF-κB pathway) as well as [47] cancers, where NF-κB is involved in tumor progression. the activity of individual IKKβ and IKKα (necessary for Here, we found a significant and positive correlation the activation of the non-canonical pathway) [49]. On the between ERK5 and p65/RELA protein levels in human contrary, NEMO/IKKγ inhibition affects the activity of the EC tumor samples, and that ERK5 inhibition or silenc- NF-κB pathway mediated by the IKK complex (canonical ing, or MEK5 genetic deletion, resulted in impairment pathway), but not the activity of individual IKKα or IKKβ of the NF-κB canonical pathway in EC cells and tumor or the non-canonical pathway. Concordantly, a small com- xenografts. These results are in agreement with those pound that blocks the binding of NEMO/IKKγ to IKKβ has reported for colorectal [26] and leukemic [25] cancers, shown anticancer activity in colorectal cancer tumor xeno- where MEK5/ERK5 overexpression results in enhanced grafts [50]. Considering that ERK5 inhibition or silencing p65/RELA nuclear localization and transcriptional activ- impaired NEMO/IKKγ expression in EC cells and xenograft ity. Furthermore, ERK5 is essential for NF-κB-induced tumors, we speculate that ERK5 inhibition could be an effec- survival in leukemic cells [25], and ERK5 protein lev- tive and safe strategy to impair the NF-κB pathway in those els positively correlate with p65/RELA levels in colon cancers where NF-κB acts as a driver, such as EC. 13 524 Page 18 of 20 N. Diéguez‑Martínez et al. Finally, here we report that ERK5 inhibition sensitizes (MINECO, grant SAF2015-64237-R), and the Spanish Ministry of Sci- EC cells and xenograft tumors to the standard chemo- ence and Innovation (grant PID2019-107561RB-I00), and co-funded by the European Regional Development Fund (ERDF). therapy agents paclitaxel and carboplatin (1), suggesting that ERK5i could be used as an effective targeted therapy Data availability The original contributions presented in the study are to improve chemotherapy of EC patients. Our results are included in the article/supplementary material, and further inquiries in line with those reported for other cancer types. Thus, can be directed to the corresponding author. among others, ERK5 inhibitor XMD8-92 improves the cytotoxicity of docetaxel and/or doxorubicin in lung [51], Declarations triple negative breast [52] and squamous skin [53] can- cer cells and tumors, as well as the antitumor action of Conflict of interest Elisabet Megias-Roda and Héctor Pérez-Montoyo are Ability Pharmaceuticals employees; Jose M Lizcano serves as an 5’-fluoroacil in p53-dependent colorectal cancer cells [54]. uncompensated member of Ability Pharmaceuticals advisory board, Furthermore, ERK5 silencing also potentiates doxorubicin and holds shares of the company. and cisplatin therapies in malignant mesothelioma cells [55], or doxorubicin, cyclophosphamide, and paclitaxel in Ethical approval The animal study was approved by the Ethics Com- mittee on Animal Experiments of the Universitat Autònoma de Barce- TNBC cells [56]. The mechanism by which ERK5 inhibi- lona (CEEAH) and the Ethics Commission in Animal Experimentation tion sensitizes several types of cancer cells remains to be of the Generalitat de Catalunya, Spain. described, although it has been suggested that it could be due to inhibition of some of the ABC transporters Consent to participate The use of samples from patients was approved by the Institutional Review Boards of Hospital Vall d’Hebron (Bar- [57]. Since we found that pharmacologic inhibition of the celona, Spain), and written informed consent was obtained from the NF-κB pathway (IKKβ inhibitor BAY-117082) also poten- patients involved. tiated the cytotoxic effect of paclitaxel and carboplatin in EC cells, it is plausible that NF-κB inhibition may medi- Consent to publish Not applicable. ate the sensitization of EC cells to chemotherapy exerted by ERK5 inhibitors. Moreover, because activation of the Open Access This article is licensed under a Creative Commons Attri- NF-κB pathway is implicated in resistance to taxanes and bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long platinum-based drugs in several preclinical models of as you give appropriate credit to the original author(s) and the source, cancer [58], it is tempting to speculate that ERK5 inhibi- provide a link to the Creative Commons licence, and indicate if changes tion might prevent NF-κB-mediated resistance to chemo- were made. The images or other third party material in this article are therapy in EC cells. Further work will be necessary to included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in investigate this possibility. the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will Supplementary Information The online version contains supplemen- need to obtain permission directly from the copyright holder. To view a tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 00018-0​ 22-0​ 4541-6. copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. Acknowledgements S. Espinosa-Gil is a recipient of a fellowship from FI-AGAUR (2020-FISDU-00575). N. Dieguez-Martinez and I. Bolinaga-Ayala are recipients of PIF fellowships from UAB. Gokhan References Gorgisen is a recipient of the 2219 International Postdoctoral Research Fellowship Program for Turkish Citizens from the Scientific and Tech- 1. Siegel RL, Miller KD, Jemal A (2018) Cancer statistics, 2018. nological Research Council of Turkey (TUBİTAK). We are grateful Cancer J Clin 68:7–30. https://​doi.​org/​10.​3322/​caac.​21442 to Cristina Gutierrez, and Neus Ontiveros for tissue culture, to Mar 2. Colombo N, Creutzberg C, Amant F et al (2016) ESMO-ESGO- Hernández for IHC assistance. We also thank Miguel Segura and Victor ESTRO consensus conference on endometrial cancer: diagnosis, Yuste for helpful discussions. We thank the following UAB Services: treatment and follow-up. Ann Oncol 27:16–41. https://​doi.​org/​10.​ Servei de Cultius Cel·lulars INc, Laboratori de Luminiscència Espec- 1093/​annonc/​mdv484 troscopia and Servei Genòmica Informàtica. We also thank to the UAT 3. Morice P, Leary A, Creutzberg C et al (2016) Endometrial cancer. Service from VHIR. Lancet 387:1094–1108. https://​doi.​org/​10.​1016/​S0140-​6736(15)​ 00130-0 Author contributions Conceived and designed the experiments: 4. Bregar AJ, Growdon WB (2016) Emerging strategies for targeting ND-M, SE-G, XD and JML. Performed experiments: ND-M, SE-G, PI3K in gynecologic cancer. Gynecol Oncol 140:333–344. https://​ GY, EM-R, IB-A, MV-C, GG, ID-O and JML. Analysis of the data: doi.​org/​10.​1016/j.​ygyno.​2015.​09.​083 ND-M, SE-G, HP-M, EC, JRB, XD and JML. Writing–original draft: 5. Kato Y, Tapping RI, Huang S et al (1998) Bmk1/Erk5 is required ND-M, SE-G and JML. All the authors read, critically revised, and for cell proliferation induced by epidermal growth factor. Nature approved the final manuscript. 395:713–716. https://​doi.​org/​10.​1038/​27234 6. Drew BA, Burow ME, Beckman BS (2012) MEK5/ERK5 path- Funding Open Access Funding provided by Universitat Autonoma way: the first fifteen years. Biochim Biophys Acta 1825:37–48. de Barcelona. The JM Lizcano research group was supported by https://​doi.​org/​10.​1016/j.​bbcan.​2011.​10.​002 grants from the Spanish Ministry of Economy and Competitiveness 13 The ERK5/NF‑κB signaling pathway targets endometrial cancer proliferation and survival Page 19 of 20 524 7. Lochhead PA, Gilley R, Cook SJ (2012) ERK5 and its role in 25. Garaude J, Cherni S, Kaminski S et al (2006) ERK5 activates tumour development. Biochem Soc Trans 40:251–256. Biochem NF-kappaB in leukemic T cells and is essential for their growth Soc Trans 40:251–256 in vivo. J Immunol 177:7607–7617. https://​doi.​org/​10.​4049/​ 8. Stecca B, Rovida E (2019) Impact of ERK5 on the hallmarks of jimmu​nol.​177.​11.​7607 cancer. Int J Mol Sci 20:1426 26. Simões A, Pereira DM, Gomes SE et al (2015) Aberrant 9. Pereira DM, Rodrigues CMP (2020) Targeted avenues for cancer MEK5/ERK5 signalling contributes to human colon cancer treatment: the MEK5-ERK5 signaling pathway. Trends Mol Med progression via NF-κB activation. Cell Death Dis 6:e1718. 26:394–407. https://​doi.​org/​10.​1016/j.​molmed.​2020.​01.​006 https://​doi.​org/​10.​1038/​cddis.​2015.​83 10. Wang J, Erazo T, Ferguson FM et al (2018) Structural and atropi- 27. Pallares J, Martínez-Guitarte JL, Dolcet X et al (2004) Abnor- someric factors governing the selectivity of pyrimido-benzodiazi- malities in the NF-κB family and related proteins in endometrial pinones as inhibitors of kinases and bromodomains. ACS Chem carcinoma. J Pathol 204:569–577. https://​doi.​org/​10.​1002/​path.​ Biol 13:2438–2448. https://d​ oi.o​ rg/1​ 0.1​ 021/a​ csche​ mbio.7​ b006​38 1666 11. Gámez-García A, Bolinaga-Ayala I, Yoldi G et al (2021) ERK5 28. Tang G, Minemoto Y, Dibling B et al (2001) Inhibition of JNK inhibition induces autophagy-mediated cancer cell death by acti- activation through NF-κB target genes. Nature 414:313–317. vating ER stress. Front Cell Dev Biol 9:1–16. https://​doi.​org/​10.​ https://​doi.​org/​10.​1038/​35104​568 3389/​fcell.​2021.​742049 29. Papa S, Zazzeroni F, Pham CG et al (2004) Linking JNK signaling 12. Erazo T, Moreno A, Ruiz-Babot G et al (2013) Canonical and to NF-κB: a key to survival. J Cell Sci 117:5197–5208. https://d​ oi.​ kinase activity-independent mechanisms for extracellular signal- org/​10.​1242/​jcs.​01483 regulated kinase 5 (ERK5) nuclear translocation require disso- 30. Gilmore TD (2006) Introduction to NF-κB: players, pathways, ciation of Hsp90 from the ERK5-Cdc37 complex. Mol Cell Biol perspectives. Oncogene 25:6680–6684. https://​doi.​org/​10.​1038/​ 33:1671–1686. https://​doi.​org/​10.​1128/​MCB.​01246-​12 sj.​onc.​12099​54 13. Marques-Fernandez F, Planells-Ferrer L, Gozzelino R et al (2013) 31. Gribble FM, Loussouarn G, Tucker SJ et al (2000) A novel TNFα induces survival through the FLIP-L-dependent activation method for measurement of submembrane ATP concentration. J of the MAPK/ERK pathway. Cell Death Dis 4:e493. https://​doi.​ Biol Chem 275:30046–30049. https://d​ oi.o​ rg/1​ 0.1​ 074/j​ bc.M ​ 0010​ org/​10.​1038/​cddis.​2013.​25 10200 14. Erazo T, Lorente M, Lopez-Plana A et al (2016) The new antitu- 32. Knight ZA, Shokat KM (2005) Features of selective kinase inhibi- mor drug ABTL0812 inhibits the Akt/mTORC1 axis by upregu- tors. Chem Biol 12:621–637. https://​doi.​org/​10.​1016/j.​chemb​iol.​ lating tribbles-3 pseudokinase. Clin Cancer Res 22:2508–2519. 2005.​04.​011 https://​doi.​org/​10.​1158/​1078-​0432.​CCR-​15-​1808 33. Ching JK, Rajguru P, Marupudi N et al (2010) A role for AMPK 15. Chou TC (2010) Drug combination studies and their synergy in increased insulin action after serum starvation. Am J Physiol quantification using the chou-talalay method. Cancer Res 70:440– Physiol 299:C1171–C1179. https://d​ oi.o​ rg/1​ 0.1​ 152/a​ jpcel​ l.0​ 0514.​ 446. https://​doi.​org/​10.​1158/​0008-​5472.​CAN-​09-​1947 2009 16. Kato Y, Kravchenko VV, Tapping RI et al (1997) BMK1/ERK5 34. Monti M, Celli J, Missale F et al (2022) Clinical significance and regulates serum-induced early gene expression through transcrip- regulation of ERK5 expression and function in cancer. Cancers tion factor MEF2C. EMBO J 16:7054–7066. https://​doi.​org/​10.​ 14:348. https://​doi.​org/​10.​3390/​cance​rs140​20348 1093/​emboj/​16.​23.​7054 35. Konecny GE, Santos L, Winterhoff B et al (2009) HER2 gene 17. Lin ECK, Amantea CM, Nomanbhoy TK et al (2016) ERK5 amplification and EGFR expression in a large cohort of surgically kinase activity is dispensable for cellular immune response and staged patients with nonendometrioid (type II) endometrial can- proliferation. Proc Natl Acad Sci USA 113:11865–11870. https://​ cer. Br J Cancer 100:89–95. https://​doi.​org/​10.​1038/​sj.​bjc.​66048​ doi.​org/​10.​1073/​pnas.​16090​19113 14 18. Rovida E, Di Maira G, Tusa I et al (2015) The mitogen-activated 36. Rovida E, Di MG, Tusa I et al (2015) The mitogen-activated protein kinase ERK5 regulates the development and growth of protein kinase ERK5 regulates the development and growth of hepatocellular carcinoma. Gut 64:1454–1465. https://​doi.​org/​10.​ hepatocellular carcinoma. Gut 64:1454–1465. https://​doi.​org/​10.​ 1136/​gutjnl-​2014-​306761 1136/​gutjnl-​2014-​306761 19. Tusa I, Gagliardi S, Tubita A et al (2021) The hedgehog-GLI 37. Mehta PB, Jenkins BL, McCarthy L et al (2003) MEK5 overex- pathway regulates MEK5-ERK5 expression and activation in pression is associated with metastatic prostate cancer, and stimu- melanoma cells. Int J Mol Sci 22:11259 lates proliferation, MMP-9 expression and invasion. Oncogene 20. Sánchez-Fdez A, Re-Louhau MF, Rodríguez-Núñez P et al (2021) 22:1381–1389. https://​doi.​org/​10.​1038/​sj.​onc.​12061​54 Clinical, genetic and pharmacological data support targeting the 38. Lochhead PA, Clark J, Wang LZ et al (2016) Tumor cells with MEK5/ERK5 module in lung cancer. npj Precis Oncol 5:78. KRAS or BRAF mutations or ERK5/MAPK7 amplification are https://​doi.​org/​10.​1038/​s41698-​021-​00218-8 not addicted to ERK5 activity for cell proliferation. Cell Cycle 21. Ferguson FM, Nabet B, Raghavan S et al (2020) Discovery of a 15:506–518. https://​doi.​org/​10.​1080/​15384​101.​2015.​11209​15 selective inhibitor of doublecortin like kinase 1. Nat Chem Biol 39. Mulloy R, Salinas S, Philips A, Hipskind RA (2003) Activation of 6:635–643. https://​doi.​org/​10.​1038/​s41589-​020-​0506-0 cyclin D1 expression by the ERK5 cascade. Oncogene 22:5387– 22. Fell MJ, Mirescu C, Basu K et al (2015) MLi-2, a potent, selec- 5398. https://​doi.​org/​10.​1038/​sj.​onc.​12068​39 tive, and centrally active compound for exploring the therapeutic 40. Esparis-Ogando A, Az-Rodriguez E, Montero JC et al (2002) Erk5 potential and safety of lrrk2 kinase inhibition. J Pharmacol Exp participates in neuregulin signal transduction and is constitutively Ther 355:397. https://​doi.​org/​10.​1124/​jpet.​115.​227587 active in breast cancer cells overexpressing ErbB2. Mol Cell Biol 23. Gomez N, Erazo T, Lizcano JM (2016) ERK5 and cell prolifera- 22:270–285. https://​doi.​org/​10.​1128/​MCB.​22.1.​270-​285.​2002 tion: nuclear localization is what matters. Front Cell Dev Biol. 41. Montero JC, Ocana A, Abad M et al (2009) Expression of Erk5 in https://​doi.​org/​10.​3389/​fcell.​2016.​00105 early stage breast cancer and association with disease free survival 24. Zong W-X, Lindsten T, Ross AJ et al (2001) BH3-only proteins identifies this kinase as a potential therapeutic target. PLoS ONE that bind pro-survival Bcl-2 family members fail to induce apop- 4:e5565. https://​doi.​org/​10.​1371/​journ​al.​pone.​00055​65 tosis in the absence of Bax and Bak. Genes Dev. https://​doi.​org/​ 42. Saito A, Yoshida H, Nishikawa T, Yonemori K (2021) Human 10.​1101/​gad.​897601 epidermal growth factor receptor 2 targeted therapy in endometrial 13 524 Page 20 of 20 N. Diéguez‑Martínez et al. cancer: clinical and pathological perspectives. World J Clin Oncol 52. Al-Ejeh F, Miranda M, Shi W et al (2014) Kinome profiling 12:868–881. https://​doi.​org/​10.​5306/​wjco.​v12.​i10.​868 reveals breast cancer heterogeneity and identifies targeted thera- 43. Fader AN, Roque DM, Siegel E et al (2020) Randomized phase II peutic opportunities for triple negative breast cancer. Oncotarget Trial of carboplatin-paclitaxel compared with carboplatin–pacli- 5:3145–3158. https://​doi.​org/​10.​18632/​oncot​arget.​1865 taxel–trastuzumab in advanced (stage III–IV) or recurrent uterine 53. Finegan KG, Perez-Madrigal D, Hitchin JR et al (2015) ERK5 serous carcinomas that overexpress Her2/Neu (NCT01367002): is a critical mediator of inflammation-driven cancer. Cancer Res updated overall survival analysis. Clin Cancer Res 26:3928–3935. 75:742–753. https://​doi.​org/​10.​1158/​0008-​5472.​CAN-​13-​3043 https://​doi.​org/​10.​1158/​1078-​0432.​CCR-​20-​0953 54. Pereira DM, Simoes AE, Gomes SE et al (2016) MEK5/ERK5 44. Xia Y, Shen S, Verma IM (2014) NF-κB, an active player in signaling inhibition increases colon cancer cell sensitivity to human cancers. Cancer Immunol Res 2:823–830. https://​doi.​org/​ 5-fluorouracil through a p53-dependent mechanism. Oncotarget 10.​1158/​2326-​6066.​CIR-​14-​0112 7:34322–34340. https://​doi.​org/​10.​18632/​oncot​arget.​9107 45. Yilmaz E, Coskun EI, Gul M et al (2017) Nuclear factor-kappa 55. Shukla A, Miller JM, Cason C et al (2013) Extracellular signal- β pathway and endometrial cancer: a pilot study. Eur J Gynaecol regulated kinase 5: a potential therapeutic target for malignant Oncol 38:536–540. https://​doi.​org/​10.​12892/​ejgo3​471.​2017 mesotheliomas. Clin Cancer Res 19:2071–2083. https://​doi.​org/​ 46. Harrington BS, Annunziata CM (2019) NF-κB signaling in ovar- 10.​1158/​1078-​0432.​CCR-​12-​3202 ian cancer. Cancers (Basel) 11:1182. https://​doi.​org/​10.​3390/​ 56. Pavan S, Meyer-Schaller N, Diepenbruck M et al (2018) A cance​rs110​81182 kinome-wide high-content siRNA screen identifies MEK5- 47. Tilborghs S, Corthouts J, Verhoeven Y et al (2017) The role of ERK5 signaling as critical for breast cancer cell EMT and nuclear factor-kappa b signaling in human cervical cancer. Crit metastasis. Oncogene 37:4197–4213. https://​doi.​org/​10.​1038/​ Rev Oncol Hematol 120:141–150. https://​doi.​org/​10.​1016/j.​critr​ s41388-​018-​0270-8 evonc.​2017.​11.​001 57. Wang F, Li D, Zheng Z et al (2020) Reversal of ABCB1-related 48. Brown HA, Williams CAC, Zhou H et al (2021) An ERK5-KLF2 multidrug resistance by ERK5-IN-1. J Exp Clin Cancer Res 39:50. signalling module regulates early embryonic gene expression and https://​doi.​org/​10.​1186/​s13046-​020-​1537-9 telomere rejuvenation in stem cells. Biochem J 478:4119–4136. 58. Godwin P, Baird A-M, Heavey S et al (2013) Targeting nuclear https://​doi.​org/​10.​1042/​BCJ20​210646 factor-kappa b to overcome resistance to chemotherapy. Front 49. Ramadass V, Vaiyapuri T, Tergaonkar V (2020) Small molecule Oncol. https://​doi.​org/​10.​3389/​fonc.​2013.​00120 NF-κB pathway inhibitors in clinic. Int J Mol Sci 2:5164. https://​ doi.​org/​10.​3390/​ijms2​11451​64 Publisher's Note Springer Nature remains neutral with regard to 50. Yu Z, Gao J, Zhang X et al (2022) Characterization of a small- jurisdictional claims in published maps and institutional affiliations. molecule inhibitor targeting NEMO/IKKβ to suppress colorectal cancer growth. Signal Transduct Target Ther 7:71. https://d​ oi.o​ rg/​ 10.​1038/​s41392-​022-​00888-1 51. Yang Q, Liao L, Deng X et al (2013) BMK1 is involved in the regulation of p53 through disrupting the PML–MDM2 interaction. Oncogene 32:3156–3164. https://​doi.​org/​10.​1038/​onc.​2012.​332 13