ERK5IsaCriticalMediatorofInﬂammation-Driven Cancer · Student t test or ANOVA (one- or two-way) depending on the typeofdata. , P

Tumor and Stem Cell Biology

ERK5 Is aCriticalMediator of Inflammation-DrivenCancerKatherine G. Finegan1, Diana Perez-Madrigal1, James R. Hitchin2, Clare C. Davies1,Allan M. Jordan2, and Cathy Tournier1

Abstract

Chronic inflammation is a hallmark of many cancers, yet thepathogenic mechanisms that distinguish cancer-associatedinflammation from benign persistent inflammation are stillmainly unclear. Here, we report that the protein kinase ERK5controls the expression of a specific subset of inflammatorymediators in the mouse epidermis, which triggers the recruit-ment of inflammatory cells needed to support skin carcino-genesis. Accordingly, inactivation of ERK5 in keratinocytesprevents inflammation-driven tumorigenesis in this model. In

addition, we found that anti-ERK5 therapy cooperates syner-gistically with existing antimitotic regimens, enabling efficacyof subtherapeutic doses. Collectively, our findings identifiedERK5 as a mediator of cancer-associated inflammation in thesetting of epidermal carcinogenesis. Considering that ERK5 isexpressed in almost all tumor types, our findings suggest thattargeting tumor-associated inflammation via anti-ERK5 therapymay have broad implications for the treatment of humantumors. Cancer Res; 75(4); 742–53. �2015 AACR.

IntroductionThe idea that the immune system influences tumorigenesis was

largely overlooked until epidemiologic studies identified chronicinfection and inflammation as major risk factors for developingvarious types of cancers (1). Furthermore, in cancerswhere there isno evidence for an underlying inflammatory reaction beforetumor formation, malignant cells, unlike their normal counter-parts, can attract leukocytes (1). As a result, immune cells create amicroenvironment that supports not only the de novo develop-ment of tumors by promoting the survival and sustained prolif-eration of cancer cells, but also many aspects of cancer progres-sion, including malignant transformation and angiogenesis (2).In addition, the resistance of tumor cells to chemotherapeuticagents appears to be enhanced bymacrophages (3). Consequent-ly, cancer-related inflammation (CRI) is now recognized as onehallmark of cancer (4).

Central to the inflammatory condition associated with tumor-igenesis is a complex network of chemokines and cytokinesproduced by both malignant and myeloid cells in the microen-vironment (2). Chemokines and cytokinesmediate their effect viathe activation of oncogenic transcription factors that have beenfound constitutively activated in many types of cancers (5). For

example, through the activation of nuclear factor kB (NFkB),interleukin (IL) 1a signaling via the IL1 receptor–MyD88 complexestablishes a positive feedback loop to reinforce the expression ofprotumorigenic factors in Ras-expressing keratinocytes (6). Col-lectively, these observations have advanced our molecular under-standing of the protumorigenic function of chemokines andcytokines in vivo. In contrast, very little is known about thepathogenic signaling mechanisms that initiate the inflammatoryreaction.

Mitogen-activated protein kinases (MAPK) are components ofcore signaling pathways essential for controlling a plethora ofcellular processes. Inparticular, the extracellular-regulated proteinkinase 5 (ERK5) is a nonredundant member of this family, whichhas been implicated in vascular integrity during development andin angiogenesis associatedwith the formationof tumor xenografts(7). Furthermore, a novel potent and specific ATP-competitiveinhibitor of ERK5, XMD8-92, was recently shown to suppresstumor growth through reduced cell proliferation in mice bearingxenografts (8). These findings are highly relevant to human cancerconsidering that elevated expression of MAPK/ERK kinase 5(MEK5) and ERK5 in human epithelial tumors correlates withunfavorable prognosis, that includes shorter disease-free inter-vals, increased risk of metastasis, and resistance to chemotherapy(9). However, the interpretation of results obtained from xeno-graft models can be limited due to immune deficiency of the host.Consequently, a causal relationship between ERK5 and tumori-genesis remained to be rigorously established.

Here, we have developed compoundmutantmice with a defectin erk5 gene expression in epidermal keratinocytes. Using the two-stage chemical carcinogenesis protocol to recapitulate in situ thestepwise formation of tumors in the presence of an intact immunesystem, we have discovered that ERK5 is an essential componentof the inflammatory response of the skin associated with tumorformation and maintenance. Accordingly, we demonstrate thatanti-ERK5 therapy combined with anti-mitotics more effectivelyregressed tumors than each treatment alone.

1Facultyof Life Sciences,UniversityofManchester,Manchester, UnitedKingdom. 2Drug Discovery Unit Cancer Research UK ManchesterInstitute, University of Manchester, Manchester, United Kingdom.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Authors: Cathy Tournier, Faculty of Life Sciences, University ofManchester, Michael Smith Building, Oxford Road, Manchester M13 9PT, UnitedKingdom. Phone: 44-161-275-5417; Fax: 44-161-275-5082; E-mail:[email protected]; and Katherine G. Finegan,[email protected]

doi: 10.1158/0008-5472.CAN-13-3043

�2015 American Association for Cancer Research.

CancerResearch

Cancer Res; 75(4) February 15, 2015742

on October 17, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst February 3, 2015; DOI: 10.1158/0008-5472.CAN-13-3043

http://cancerres.aacrjournals.org/

Materials and MethodsGenotype determination of mice and tissues

Mice carrying the erk5fl allele and the K14-CreERT2 transgenewere identified by PCR on genomic DNA isolated from tail andtissues, as previously described (10, 11). The mouse strains weremaintained in a pathogen-free facility at the University of Man-chester. All animal procedures were performed under license inaccordance with the UK Home Office Animals (Scientific Proce-dures) Act (1986) and institutional guidelines. In particular,careful clinical examination of the mice was carried out to allowdetection of deterioration of their physical condition associatedwith the progression of tumors. Animals showing signs of distresswere sacrificed before any further deterioration in conditionoccurred.

Chemical carcinogenesis protocolMice were backcrossed into an FVB background. Six-week old

littermates were subjected to intraperitoneal injection of a non-toxic amount of tamoxifen (200 mg) every day for 5 days. Twoweeks later, the shaved backs of the mice received a single

application of 7,12-dimethyl-benzanthracene (DMBA; 25 mg).One week after DMBA, the back skin was treated biweekly with12-O-tetradecanoylphorbol 13-acetate (TPA; 6 mg) for 20 weeks.In Fig. 1E, mice were exposed weekly to a single dose of DMBA(30 mg) for 29 weeks. Tumor burden was recorded (blind countsand photographs) every week for the duration of the experiment.For the tumor regression studies, mice were subjected to theDMBA/TPA protocol for 10 to 16 weeks before intraperitonealinjection of tamoxifen (200 mg; every day for 5 days) followed aweek later by three topical applications of 4-hydroxy-tamoxifen(4-OHT; 100 mg) on alternating days. Alternatively, XMD8-92(CRUK Manchester Institute, Manchester, United Kingdom; 25mmol/L) was administered topically every other day (three times,weekly) until the end of the experiments. Age-matched mock-treated (acetone) animals were used as controls. Doxorubicin(1.25 mg/kg per mouse) was administered intravenously weeklyfor 4 consecutiveweeks, 1week after the end of 4-OHTdosing or 1week after the beginning of XMD8-92 therapy. Biweekly TPAtreatment was continued for the duration of all regression experi-ments. When this was combined with XMD8-92, XMD8-92 wasapplied 30minutes before TPA. The number of tumors larger than

Figure 1.ERK5 is required for skin tumorformation. Adult erk5fl/flmice carrying(þ) or not (�) the K14-CreERT2

transgene were subjected tointraperitoneal tamoxifen injections.A–C, the specific inactivation of theerk5 gene in the epidermis of miceexpressing CreERT2 was confirmed byimmunoblot analyses (A and B) andby immunostaining of skin biopsieswith a specific antibody to ERK5(gray; C). Slides were counterstainedwith nuclear red. H&E stainingindicates that the absence of ERK5does not disrupt the architecture ofthe skin. D and E, representativepictures ofmice bearing papillomas 20weeks after DMBA/TPA treatment (D)and of mice bearing SCC 29 weeksafter repeated exposure to DMBA (E).Papillomas or SCC growing on theback skin of mice exposed toDMBA/TPA or DMBA alone werecounted. The data are represented asaverage number of tumors per mice� SE. Total number of animals usedwas 60 in D and 16 in E.

ERK5 Promotes Inflammation-Driven Tumorigenesis

www.aacrjournals.org Cancer Res; 75(4) February 15, 2015 743




3 mm in size was recorded (blind counts) every week for theduration of the experiment.

Preparation of epidermisesSkin biopsieswere removed andfloated dermis-down in 0.25%

trypsin-EDTA overnight at 4�C. Epidermises were then separatedfrom the dermis and analyzed as required.

Cell cultureHaCat cells were cultured in DMEM containing 10% FBS, 1%

penicillin/streptomycin, and 1% glutamine. Where indicated, thecells were pretreated for 30 minutes with 25 mmol/L XMD8-92and/or 10 mmol/L caspase-1 inhibitor (Ac-YVAD-CHO; Enzo LifeSciences).

Immunoblot analysisProteins were extracted from cells in RIPA buffer containing

inhibitors of proteases and protein phosphatases. Extracts (20 mg)were resolved by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and electrophoretically transferred to an Immobilon-Pmembrane (Millipore, Inc.). Themembraneswere incubatedwith3% nonfat dry milk or BSA for 60 minutes and probed overnightat 4�C with antibodies (1:1,000) to ERK5 (Upstate), human IL1b(R&D Systems), mouse IL1b (R&D Systems) or cleaved caspase 1(Cell Signaling Technology). Tubulin (Sigma) or actin (Sigma)were used to monitor protein loading. Immunecomplexes weredetected by enhanced chemiluminescence with immunoglobulinG coupled to horseradish peroxidase as the secondary antibody(GE Healthcare).

Protein kinase assayEndogenous ERK5 was immunoprecipitated from cellular or

epidermal extracts in the triton lysis buffer (TLB; 20 mmol/L TrispH 7.4, 137mmol/L NaCl, 2mmol/L EDTA, 1% Triton X-100, 25mmol/L b glycerophosphate, 10% glycerol, 1 mmol/L orthova-nadate, 1mmol/L phenylsulphonylfluoride, 10mg/mL leupeptin,and 10 mg/mL aprotinin) using an antibody to ERK5 (Upstate).Immunecomplexes were washed twice with TLB and twice withkinase buffer (25 mmol/L HEPES pH 7.4, 25 mmol/L b-glycer-ophosphate, 25 mmol/L MgCl2, 2 mmol/L DTT, and 0.1 mmol/Lorthovanadate) before being incubated at 30�C for 20minutes inkinase buffer containing 50 mmol/L [g-32P] ATP (10 Ci/mmol)and 1 mg of GST-MEF2C. The reactions were terminated byaddition of Laemmli sample buffer. Radioactivity incorporatedinto the recombinant protein was quantified after SDS-PAGE byPhosphoImager analysis.

ELISA assayRelease of mature IL1b into culture medium was measured

using a specific sandwich-type enzyme-linked immunosorbentassay (ELISA; Ready Set Go; Human Interleukin-1 b kit;eBioscience) according to the manufacturer's instructions.

Caspase-1 activity assayCellular or epidermal extracts were prepared in caspase buffer

(10 mmol/L HEPES pH 7.5, 150 mmol/L NaCl, 2 mmol/L EDTAcontaining 0.5%NP40). Extractswere incubatedwith 100mmol/LYVAD-AMC caspase-1–specific fluorogenic substrate (Enzo LifeSciences) for 1 hour. Cleavage of the substrate was measured byspectrofluorometer.

Histologic and immunohistochemical analysesFreshly isolated skin biopsies and tumors were fixed in 4%

paraformaldehyde, dehydrated, and embedded in paraffin. Five-micrometer thick sections were cut. For histologic analysis, sec-tions were stained with hematoxylin and eosin (H&E). Forimmunohistochemistry, sections were deparaffinized, rehydra-ted, and treated in boiling sodium citrate buffer (10 mmol/L pH6.0) for 10 minutes to unmask antigens. Endogenous peroxidaseactivity was quenched by treating the slides with 0.3% hydrogenperoxide for 10minutes. Sections were blocked in PBS containing3% to 10% goat serum and 0.1% Triton X-100 for 1 hour at roomtemperature before being incubated overnight at 4�C with pri-mary antibodies (1:100) to ERK5 (Eurogentec), IL1b (R&D Sys-tems) or MPO (Thermo Scientific). The following day, the slideswere rinsed in PBS and incubated at room temperature for 1 hourwith secondary biotinylated antibodies. The slides were processedusing the ABC Detection Kit (Vector Laboratories). The presenceof the antigens was revealed using the Vector SG (gray) ordiaminobenzidine (DAB; brown) peroxidase substrate kit (VectorLaboratories) and counterstained with nuclear red or with hema-toxylin (blue). For bromodeoxyuridine (BrdUrd) staining, themice were injected with 500 mg/kg BrdUrd and sacrificed 1 houror 2 hours later. Immunohistochemistry was performed using amouse monoclonal antibody to BrdUrd (Dako). For mast cellstaining, sections were incubated for 5 minutes in toluidine bluestain [0.1% toluidine blueO (Sigma), 1%NaCl, pH2.5]. All slideswere viewed using the Axioplan2 microscope.

Quantitative real-time PCRTotal RNA was isolated from cells or epidermises using the

TRIzol reagent and cDNA synthesis carried out, as previouslydescribed (12). Quantitative real-time PCRs were performedusing the SYBR Green I Core Kit (Eurogentec). Sequences of theforward and reverse primers are indicated in Supplementary TableS1. PCR products were detected in the ABI-PRISM 7700 sequencedetection systems (Applied Biosystems). Results were analyzedusing the 2�DDG method. The level of expression of mRNA wasnormalized to actinmRNA in mouse tissues or GAPDH and Pgk1mRNA in HaCat cells.

Statistical analysisStatistical significance was determined using the two-tailed

Student t test or ANOVA (one- or two-way) depending on thetype of data. �, P < 0.05 was taken to be statistically significant. nsindicates no statistical difference. For all mouse tissue experi-ments, images are representative of cohorts of at least three mice.

ResultsInactivation of ERK5 in the skin prevents carcinogen-inducedtumorigenesis

Mice lacking erk5 die before birth (13). To circumvent this earlylethality and permit the study of the pathologic function of ERK5,we developed a mouse model in which the erk5 gene was flankedwith LoxP sites [referred to as the flox (fl) allele; 10]. Homozygouserk5fl mice expressing Cre fused to a mutated form of the ligand-binding domain of the estrogen receptor (CreERT2) under thecontrol of the keratin 14 (K14) promoter (14), were generated.Immunoblot analysis demonstrates that this system allows thespecific ablation of ERK5 in the epidermis following injection ofthemice with tamoxifen (Fig. 1A and B). Immunostaining of skin

Finegan et al.

Cancer Res; 75(4) February 15, 2015 Cancer Research744




sectionswith an antibody to ERK5 confirmed the absence of ERK5in the epidermis of K14-CreERT2;erk5fl/fl mice treated with tamox-ifen (Fig. 1C). We verified that a control IgG antibody producedno staining under the same experimental conditions (data notshown). The loss of ERK5 in adult mice did not cause any obviousmorphologic defect, suggesting that ERK5 was not required forskin homeostasis (Fig. 1C). In subsequent experiments, erk5fl/fl

and K14-CreERT2;erk5fl/fl animals injected with tamoxifen will bereferred to as erk5wt and erk5Depidermis animals, respectively.

To examine the role of ERK5 in tumorigenesis, mice weresubjected to the classic two-stage chemical carcinogenesis proto-col, which gives rise to skin papillomas (15). This simple andhighly reproducible system is particularly suited to explore themechanism of epithelial cancer-associated inflammation becausethe repeated treatment of themouse skin with TPA recapitulates apersistent and nonresolving inflammation that promotes theclonal expansion of keratinocytes harboring oncogenic Ha-rasmutation (caused by a single, cutaneous application of thegenotoxic carcinogen DMBA; ref. 16). Importantly, resolutionof inflammation via cessation of TPA treatment causes signif-icant inhibition of tumor growth, as well as frequent sponta-neous tumor regression (16). We discovered that tumor burden(number of tumors per mouse) was markedly decreased in theabsence of ERK5 (Fig. 1D). In fact, papillomas that developedon the back skin of erk5Depidermis animals were genotypicallywild-type. This was predicted because deletion of floxed genesfollowing tamoxifen-induced Cre expression does not occurwith 100% efficiency.

In addition, we observed that the malignant conversion ofpapillomas, which occurred at a frequency �5% to 10%, wasrestricted to erk5wt mice. However, it is possible that malignanttumors in the erk5Depidermis group may have remained undetected,simply because of the low overall total tumor number. Therefore,to assess the role of ERK5 inmalignancy,we compared the effect ofthe repeated treatment of erk5wt and erk5Depidermis animals withDMBA alone (Fig. 1E). This complete carcinogenesis protocolleads directly to the formation of predominantly invasive squa-mous cell carcinomas (SCC; ref. 17). As with the DMBA/TPAmodel (Fig. 1D), the absence of ERK5 caused amarked decrease intumor burden upon exposure to DMBA (Fig. 1E). SCC formationwas significantly delayed in erk5-deleted skin, with themajority oftumors detected in erk5Depidermis mice significantly smaller than inwild-type littermates (Fig. 1E). Together, these results clearlydemonstrate that ERK5 is required for the development of malig-nant skin cancer.

To test whether the resistance of ERK5-deficient mice to tumor-igenesis could be attributed to a cell proliferation defect, weperformed a short time–course analysis of TPA exposure. Epider-mal cell proliferation was maximally induced 24 hours after asingle application of TPA on the dorsal skin of the wild-typemice,as demonstrated by a 10-fold increase in the number of BrdUrd-positive cells (Fig. 2AandB). Thiswas significantly impaired in theabsence of ERK5 (Fig. 2A and B). However, 6 hours after TPAexposure, we did not observe any marked difference in thenumber of BrdUrd-positive cells between wild-type and ERK5-deficient epidermises (Fig. 2B). Instead, by this time, the mutantskin displayed a noticeably lower number of infiltrating cells inthe dermis compared with wild-type controls (Fig. 2C). Thisindicated that the proliferative defect displayed by erk5-deletedkeratinocytes was preceded by a failure of themutant epidermis torespond to inflammatory stimulation.

Cutaneous inflammation is defective in erk5Depidermis miceTo confirm the requirement of ERK5 in mediating the inflam-

matory response of the skin to TPA, we examined the effect ofepidermal erk5 gene deletion on neutrophil infiltration into thedermis. Neutrophils rapidly migrate to the skin exposed to TPAand impaired neutrophil recruitment is sufficient to prevent TPA-induced cutaneous inflammation, hyperproliferation, and papil-loma formation (18, 19). Accordingly, the number of neutrophils(MPO-positive cells) in skin sections of erk5wt mice increased 10-fold 6 hours after TPA treatment and 16-fold after 24 hours (Fig.2D). This effect was significantly reduced in the absence of ERK5,with only a 4- to 5-fold increase in neutrophil number detected inerk5Depidermis animals exposed to TPA for 6 or 24 hours (Fig. 2D). Asimilar defect was observed in skin sections of mice treated withDMBA/TPA for 20 weeks (25-fold increase in neutrophil numberin wild-type skin, 5- to 6-fold increase in mutant skin).

Two major neutrophil chemoattractants produced by kerati-nocytes, namely chemokine (C-X-Cmotif) ligand 1 (CXCL1) andCXCL2, can trigger the recruitment of neutrophils in the mouseskin (20). We confirmed that TPA induced a transient upregula-tion of CXCL1 and CXCL2 transcripts in the epidermis, with amaximum after 6 hours (Fig. 2E). This effect was abolished in theabsence of ERK5, consistent with decreased number of neutro-phils trafficking through the dermis of the mutant skin (Fig. 2Dand E). Furthermore, erk5Depidermis mice were resistant to cutane-ous chronic inflammation caused by repeated long-term TPAexposure (Fig. 2F). The establishment of chronic inflammationin the wild-type skin subjected to the DMBA/TPA protocol for 10and 20 weeks is demonstrated by the accumulation of residentmast cells in the dermis before papilloma formation, and in thetumor stroma (Fig. 2F). Together, our results are consistent withevidence that neutrophil recruitment is a prerequisite to theestablishment of a chronic inflammatory microenvironment andthatmice deficient inmast cells are resistant to skin carcinogenesis(21).

To further characterize the requirement of ERK5 in CRI, weanalyzed the expression pattern of a panel of inflammatorymediators implicated in tumorigenesis. We found that ERK5 wasrequired for TPA-induced IL1a and IL1bmRNAexpression and forthe upregulation of cyclooxygenase-2 (COX-2), a downstreameffector of IL1 (22) (Fig. 3A). Prostaglandin E2 (PGE2) is thepredominant COX-2–derived product in the skin. It acts bybinding to its cognate G-protein–coupled receptors, namelyEP1-4. Although all EP receptors, except EP3, participate in skintumor development, genetic analyses have suggested an essentialrole of EP2 in tumorigenesis (23). Here, we confirmed that TPAincreased the expression of EP1 and EP2, but not EP3, mRNA inthe epidermis (Fig. 3A and Supplementary Fig. S1). Consistentwith its very low level of expression in the skin (24), the EP4transcript was undetectable (data not shown). More importantly,TPA-induced EP2 expression was abolished in the absence ofERK5 (Fig. 3A), whereas nomarked difference was detected in thelevel of EP1 between wild-type and erk5-deleted epidermises,under basal or stimulated conditions (Supplementary Fig. S1).Likewise, TPA-mediated increased TNFa and IL6 mRNA expres-sion was independent of ERK5 (Fig. 3A), thereby demonstratingthe selective requirement of ERK5 signaling in controlling aspecific subset of inflammatory mediators.

These results establish ERK5 as a novel regulator of cutaneousinflammation via the upregulation of neutrophil chemoattrac-tants and of the IL1–COX-2 pathway. CXCL-mediated dermal






recruitment of neutrophils and COX-2–generated PGE2 signalingthrough the EP2 receptor constitute potentially importantmechanisms by which ERK5 contributes to skin tumorigenesis.

Regulation of IL1b by ERK5IL1a and IL1b are synthesized as pro-forms.While IL1a is active

as a membrane precursor, caspase-1 mediated cleavage of pro-IL1b is required for IL1b to be active as a secreted form (25). Oncecleaved, IL1b is quickly released into themicroenvironment (26).We found that the intracellular level of cleaved IL1b was signif-icantly lower in the mutant epidermis compared with wild-typesamples (Fig. 3B). Consistent with the requirement of ERK5 formediating the proteolytic cleavage of pro-IL1b, erk5-deleted epi-dermises were resistant to TPA-induced caspase-1 activation (Fig.

3C). Elevated caspase-1 activity in the wild-type skin exposed toTPA is indicative of rapid turn-over of the pro-form and subse-quently, secretion of the cleaved fragment. This dynamic processexplains why no difference was detected in the level of IL1bprotein expressed in the wild-type epidermis between basal con-ditions and following exposure to TPA (Fig. 3B), despite ourprevious evidence that TPA increased the expression of the tran-script (Fig. 3A).

The secreted form of IL1, mostly IL1b, contributes to therecruitment of leukocytes (mainly neutrophils andmacrophages)but also, acts on stromal and infiltrating cells to induce theproduction of inflammatory cytokines, including IL1b itself, tosustain and propagate the inflammatory reaction (27). Accord-ingly, a transient increase in the number of IL1b-positive cells was

Figure 2.erk5 gene deletion reduces TPA-induced inflammation. Dorsal skins oferk5wt and erk5Depidermis mice wereexposed to TPA or DMBA/TPA for theindicated times. A and B, miceexposed to a single dose of TPA wereinjected intraperitoneally with BrdUrd1 hour before the end of treatment.Skin biopsies were processed forBrdUrd immunoreactivity. Positivecells are stained in gray (A). Slideswere counterstained with nuclear red.The number of BrdUrd-positive cellsper 100 cells in 10 fields per sectionwas blind counted by two users andthe averagenumberwas recorded (B).The experimental mean � SE wasgenerated from 16 slides (foursections from four mice). C, H&Estaining of skin biopsies shows thatthe dermis of the mutant skin displaysa marked defect in inflammatory cellinfiltration following TPA treatment.D, the number of neutrophils in theskin after TPA treatment wasestimated by quantitative analysis ofMPO-positive cells. The experimentalmean � SE was generated from eightslides (two sections from four mice).E, total RNA was extracted fromepidermises and the amount of CXCL1and CXCL2 transcripts was measuredby quantitative real-time PCR. Thedata correspond to the mean � SE ofthree independent experiments inwhich one mouse was used perexperimental condition. Total numberof animals used in E was 42. F, chronicinflammation was assessed in skinbiopsies by toluidine blue staining ofmast cells. Scale bars, 50 mm.

Finegan et al.





detected in the dermis of the skin exposed to TPA (Fig. 3D).Chronic cutaneous inflammation caused by long-term DMBA/TPA treatment correlated with the presence of a population ofIL1b-positive cells permanently residing in the dermis (Fig. 3D).In contrast, and consistent with impaired IL1b secretion, noinfiltrating IL1b-positive cells were detected in the dermis of themutant skin exposed to short-term TPA treatment or subjected tothe two-stage chemical carcinogenesis protocol for 20 weeks (Fig.3D). On the basis of these results, we concluded that ERK5-dependent release of IL1b in the epidermis was required forpromoting a secondary inflammatory reaction in the dermis.

To confirm that increased IL1b expression in the epidermis wasa primary response to ERK5 activation in keratinocytes exposed toTPA, we undertook an analysis of ERK5 function in HaCat cells, a

human immortalized keratinocyte cell line, to reproduce in vitrothe immune response defect associated with the epidermal loss ofERK5 in vivo. We demonstrated that TPA stimulation inducedphosphorylation of ERK5, as indicated by an electrophoreticmobility shift of the protein analyzed by immunoblot, andincreased ERK5 activity (Fig. 4A). This was prevented by preincu-bating the cells with the pharmacologic inhibitor of ERK5, XMD8-92 (Fig. 4A). Using this model system, we confirmed the require-ment of ERK5 inmediating TPA-induced upregulation of IL8 (thefunctional homolog of murine CXCL1/2), and IL1b, but notTNFa, mRNA (Fig. 4B). Next, HaCat cells were incubated witha caspase-1 inhibitor (YVAD-CHO) to prevent pro-IL1b frombeing cleaved and secreted (Fig. 4C). Under these conditions, wecould detect TPA-mediated increased expression of pro-IL1b

Figure 3.In vivo regulation of proinflammatorycytokines by ERK5. Dorsal skins oferk5wt and erk5Depidermis mice wereexposed to TPA or DMBA/TPA for theindicated times. A, total RNA wasextracted from epidermises and theamount of transcripts was measuredby quantitative real-time PCR. Thedata correspond to the mean � SE ofthree independent experiments inwhich one mouse was used perexperimental condition. Total numberof animals used in A was 42. B and D,IL1b expression in skin biopsies wasdetermined by immunoblot analysis(B) and by immunohistochemistry(D), with a specific antibody againstIL1b. Scale bar, 50 mm. C, caspase-1activity in epidermal extracts wasmeasured by caspase assay. The datacorrespond to the mean� SE of threeindependent experiments in whichonemousewas used per experimentalcondition. Total number of animalsused in C was 12.






protein (Fig. 4C). This was prevented by preincubating the cellswith XMD8-92 (Fig. 4C). Likewise, inhibition of ERK5 signifi-cantly reduced the amountof IL1b released in themedia fromcellstreated with TPA (Fig. 4D). Accordingly, ERK5 activation wasrequired for both TPA-induced increased caspase-1 activity (Fig.4E) and the proteolytic cleavage of caspase-1 to its active form(p20; Fig. 4F). Together, these results firmly established that ERK5was essential for IL1b production by keratinocytes and revealed anovel role for ERK5 in the regulation of IL1b signaling and IL8expression in human cells.

ERK5 constitutes a novel therapeutic target to treatinflammation-driven cancer

To investigate whether our findings could be translated into auseful approach for cancer treatment, we examined the pheno-typic consequence of the epidermal loss of ERK5 expression in

skin tumors. We found that, under low tumor burden, the loss ofERK5 in neoplastic keratinocytes halted tumor growth induced bybiweekly exposure to TPA (Fig. 5A). This was associated with areduction in the number of proliferating cells labeled withBrdUrd, together with a collapse of tumor architecture (Fig.5B). To determine whether an anti-ERK5 therapy could also blockthe growth of more advanced tumors, a similar experiment wasperformed in cohorts of animals exhibiting a high tumor burdenbefore disrupting ERK5 function by either erk5 gene deletion orXMD8-92 treatment (Fig. 5C and D). Firstly, we demonstratedthat XMD8-92 effectively blocked TPA-induced ERK5 activationin the skin (Supplementary Fig. S2A). Moreover, like geneticdeletion, pharmacologic inhibition of ERK5 activity caused byexposing the skin to XMD8-92 prevented increased epidermal cellproliferation (Supplementary Fig. S2B), cells from infiltrating thedermis (Supplementary Fig. S2C), and increased caspase-1–

Figure 4.Pharmacologic inhibition of ERK5 inHaCat cells blocks IL1b secretionfollowing TPA stimulation. HaCat cellswere mock treated (DMSO) or treatedwith XMD8-92 before being incubatedwith TPA for the indicated times. A,the effect of TPA treatment on ERK5expression and posttranslationalmodification was examined byimmunoblot analysis. Similar resultswere obtained in two independentexperiments. Endogenous ERK5activity was measured by proteinkinase assay (KA). The radioactivityincorporated in the GST-MEF2substrate was quantified byphosphoImager. The data correspondto the mean � SE of threeindependent experiments. B, totalRNA was extracted and the level ofIL8, IL1b, and TNFa transcripts wasquantitated by real-time PCR. Thedata correspond to the mean � SE ofthree independent experimentsperformed in triplicate. C, caspase-1activity was inhibited by incubatingthe cells with YVAD-CHO. Proteinlysates were analyzed by immunoblotwith an antibody against IL1b.Signals corresponding to the pro-IL1bprotein were quantified with theImageQuantifier software (BioImage).Thedata correspond to themean�SEof three independent experiments. D,the amount of IL1b released in culturesupernatants wasmeasured by ELISA.Thedata correspond to themean�SEgenerated from three independentexperiments performed in triplicate. E,caspase-1 activity was measured bycaspase assay. Thedata correspond tothe mean � SE generated from threeindependent experiments performedin triplicate. F, protein lysates wereanalyzed by immunoblot with anantibody against caspase 1. Similarresults were obtained in twoindependent experiments.

Finegan et al.





mediated IL1b cleavage (Supplementary Fig. S2D and S2E) inresponse to TPA stimulation.

As expected, the number of papillomas larger than 3 mm insize increased over the time course of TPA treatment. Althoughthis was not prevented by the genetic loss of erk5, represen-tative pictures show that tumors were noticeably smaller inerk5Depidermis than in erk5wt mice (Fig. 5C). Similarly, exposureof the mouse skin to XMD8-92 affected tumor developmentwith quantitative evidence of halted tumor growth (Fig. 5D).In contrast with the genetic strategy that permits the specificloss of ERK5 in keratinocytes, XMD8-92 will inhibit ERK5activity in all cell types present in the epidermal and dermalcompartments of the skin, including immune cells. This couldexplain the stronger effect of the inhibitor on tumor growth.However, we cannot exclude the possibility that the com-pound may exert some of its effect by partially inhibitingadditional protein kinases involved in the inflammatoryreaction.

Many effective anticancer regimens use more than one ther-apeutic agent. Therefore, we sought to explore the possibilitythat targeting ERK5 signaling could enhance the effectiveness ofexisting chemotherapeutic drugs. We used a subtherapeuticdosing regimen of chemotherapy (doxorubicin; 1.25 mg/kgweekly intravenous injection for 4 consecutive weeks) thatexerted a limited effect on tumor growth (Fig. 6A and B),comparable with that caused by erk5 gene deletion in miceharboring a high tumor burden (Fig. 5C). Importantly, micesubjected to this low-dose doxorubicin did not display tran-sient gastrointestinal side effects as exhibited by littermatesreceiving high-dose doxorubicin (2.5 mg/kg weekly intrave-nous injection for 4 consecutive weeks). We found that epi-dermal erk5 gene deletion or XMD8-92 therapy combined withdoxorubicin had an additive effect in preventing tumor growth,with evidence of tumor regression indicated by a decreasednumber of large tumors (Fig. 6A and B). This is substantiated bythe demonstration that the functional disruption of ERK5

Figure 5.The functional disruption of ERK5blocks the growth of papillomas.erk5fl/fl and K14-CreERT2;erk5fl/fl

littermates were subjected to theDMBA/TPA protocol until theappearance of papillomas (Start). Thistook 10 weeks or 14 to 16 weeks forreaching low (A and B) or high (C andD) tumor burden, respectively. Micewere subsequently subjected totamoxifen (A, B, and C) to induce erk5gene deletion in neoplastickeratinocytes expressing Cre ortreated with XMD8-92 (D) to inhibitERK5 activity in the skin. BiweeklyTPA administration was continued fora further 11 (tamoxifen) or 8 (XMD8-92) weeks (Finish). A, C, and D,representative images of the tumorburden exhibited by animals at thestart and finish of the experiment. Thenumber of tumors �3 mm in size wasblind-recorded during theexperiment. The data correspond tothe mean � SE of three independentexperiments in which at least six wild-type and six mutant animals wereanalyzed. Total number of animalsused per panel was�36. B, mice wereinjected intraperitoneally with BrdUrd2 hours before being sacrificed at theend of the experiment (Finish).Representative sections of tumorsstained with eosin or processed forBrdUrd immunoreactivity are shown.The slides were counterstained withhematoxylin (blue). Arrows, BrdUrd-positive cells. Scale bars, 50 mm.






signaling in mice concurrently treated with doxorubicin severe-ly affected tumor architecture and markedly reduced the num-ber of proliferating cells in the tumors (Fig. 6C).

On the basis of our previous results, we examined whetherthe effectiveness of anti-ERK5 therapy in regressing existingtumors was attributable to the requirement of ERK5 for sus-tained inflammation in the skin. This hypothesis was con-firmed by evidence that the loss of ERK5 expression (Fig. 7A)or activity (Fig. 7B) was associated with a significant reductionin the number of neutrophils (MPO-positive cells) infiltratingthe neoplastic epidermis (E) and the tumor stroma (S), inde-pendently of doxorubicin treatment. This correlated with adecreased number of mast cells (toluidine blue–stained cells)present in the tumor stroma (Fig. 7A and B). These resultsare consistent with evidence that persistent neutrophil andmast cells recruitment is required for tumor maintenance (16).

Furthermore, tumors displayed strong and diffuse IL1b stain-ing in the papilloma epithelium (E), indicative of IL1b secre-tion in the tumor (Fig. 7A and B). This was accompanied byan influx of IL1b-positive cells in the tumor stroma (S; Fig. 7Aand B). A reduction in both, IL1b secretion and the numberof IL1b-positive cells was observed following the functionaldisruption of ERK5 signaling, thereby demonstrating therequirement of ERK5 for continued IL1b production in estab-lished tumors (Fig. 7A and B). Importantly, none of theseinflammatory markers were affected by doxorubicin treatmentalone.

Together, these results indicate that neutralizing inflamma-tion by blocking ERK5 in existing tumors represents a novelstrategy by which to treat CRI. In addition, we demonstrate thatcombining anti-ERK5 therapy with chemotherapeutic agentsaugments the antitumorigenic effect of anti-ERK5 treatment

Figure 6.The functional disruption of ERK5 signalingsynergizes with subtherapeutic doxorubicintreatment todecrease tumor burden.erk5fl/fl

and K14-CreERT2;erk5fl/fl littermates weresubjected to the DMBA/TPA protocol untilthe appearance of papillomas (14–16 weeks,Start). Mice were subsequently subjected totamoxifen (A and C) or XMD8-92(B and C), together with doxorubicin.Biweekly TPA administration was continuedfor a further 11 (tamoxifen) or 8 (XMD8-92)weeks (Finish). A and B, representativeimages of the tumor burden exhibited byanimals at the start and finish of theexperiment. The number of tumors�3mm in sizewas blind-recordedduring theexperiment. The data correspond to themean � SE of three independentexperiments in which at least six wild-typeand six mutant animals were analyzed. Totalnumber of animals used per panel was �36.C, mice were injected intraperitoneallywith BrdUrd 2 hours before being sacrificedat the end of the experiment (Finish).Representative sections of tumors stainedwith eosin or processed for BrdUrdimmunoreactivity are shown. The slideswere counterstained with hematoxylin(blue). Arrows, BrdUrd-positive cells(brown). Scale bars, 50 mm.

Finegan et al.





while concurrently lowering the therapeutic threshold of theseagents, consequently reducing the associated side effects.

DiscussionWe have discovered that epidermal expression of ERK5 is

required for mediating inflammation in the skin. Furthermore,we have demonstrated that inactivation of ERK5 in epidermalkeratinocytes prevented inflammation-driven tumorigenesis,with evidence that ERK5 is critical for the development of skinSCC. SCC is one of the most common types of human nonme-lanoma skin cancer (28). The lesions arise from keratinocytes insun-exposed areas of the epidermis and are characterized by anincreased risk of metastasis compared with basal cell carcinomas.Therefore, these findings are highly relevant because elevated

MEK5/ERK5 expression in human epithelial tumors correlateswith unfavorable prognosis (9). In addition, inflammation is ahallmark of most, if not all, cancer (4). In particular, besidesultraviolet light, inflammatory skin diseases also predisposepatients to developing cutaneous SCC (28). Accordingly, target-ing CRI has become one of the best options by which to advancecancer treatment. This therapeutic strategy is already supported byevidence that tumor burden can be reduced in mice in which keymediators of the immune response are blocked by genetic orpharmacologic means (2). Importantly, several cytokine antago-nists are being tested in phase II clinical trials with some encour-aging results (2).

The idea that ERK5 provides a mechanistic link betweeninflammation and cancer originated from our observation thatthemigration of neutrophils andmast cells to the skin, after shortand long-term exposure to TPA, was impaired in the absence ofepidermal ERK5. Likewise, the functional inhibition of ERK5 inthe neoplastic epidermis correlated with a significant reduction inthe number of infiltrating neutrophils and mast cells in all tumorcompartments. The recruitment of neutrophils in inflamed skin isregulated by chemokines, notably CXCL1 and CXCL2, whichsignal through the same receptor, CXCR2. Like erk5 gene deletion,CXCR2 deficiency was shown to suppress inflammation-driventumorigenesis (18, 19) and it was postulated that CXCR2-drivenneutrophil recruitment stimulates epithelial cell proliferation inthe skin (19). Furthermore, depletion of Ly6Gþ cells, whichinclude neutrophils, was associated with increased tumor celldeath (19). Therefore, it is reasonable to propose that the upre-gulation of CXCL1 and CXCL2 constitutes an important mech-anism by which ERK5 triggers the neutrophilic inflammationrequired for tumor cell proliferation and viability.

The causal relationship between ERK5 and inflammation isfurther supported by the demonstration that epidermal ERK5controls the expression of a specific subset of proinflammatorymediators currently under preclinical therapeutic investigationfor the treatment of human cancer (2). In particular, the func-tional loss of ERK5 in keratinocytes prevented TPA-inducedupregulation of IL1a, IL1b, and COX-2 transcripts. However, incontrast with malignant mesothelioma cell lines (29), we foundno evidence that epidermal ERK5 was required for IL6 geneexpression. This may reflect a cell-specific function of ERK5 atvarious stages of malignant transformation. Likewise, no differ-ence was observed in the level of TNFamRNA between wild-typeand ERK5-deficient epidermis under basal or stimulated condi-tions. Mice lacking TNFa have been reported to be resistant to thedevelopment of skin tumors induced by DMBA/TPA (30, 31).However, TNF receptor deficiency does not prevent neutrophilinfiltration in the epidermis exhibiting increased PKCa activity(32). Therefore, our findings are consistent with the idea thatTNFa does not contribute to ERK5-mediated recruitment ofneutrophils in the skin exposed to inflammatory stimuli.

Recently, autocrine activation of NFkB through an IL1 recep-tor–MyD88 axis has emerged as an essential positive feedbackloop that reinforces the expression of protumorigenic factors inRas-transformed keratinocytes (6). Importantly, targeted deletionof the MyD88 gene in keratinocytes conferred a significant resis-tance of the skin to DMBA/TPA-induced tumorigenesis, therebyconfirming the important contribution of epidermal IL1 signalingin cancer (6). However, although transformation by oncogenicRas induced both IL1a and IL1bmRNA, IL1bwas undetectable inthe culture supernatants (6). Therefore, we can speculate that

Figure 7.Anti-ERK5 therapy neutralizes tumor-associated inflammation. erk5fl/fl andK14-CreERT2;erk5fl/fl littermates were subjected to the DMBA/TPA protocoluntil the appearance of papillomas (14–16 weeks). Mice were subsequentlysubjected to tamoxifen (A) or XMD8-92 (B), together without or withdoxorubicin. Biweekly TPA administration was continued for a further 10weeks. A and B, immunohistochemical analysis of sections of tumorscollected at the end of the experimentwas performedwith antibodies toMPOand IL1b. Arrows, positively stained neutrophils. Mast cells were detected bytoluidine blue staining. E, neoplastic epidermis; S, tumor stroma. Thedemarcation between these two compartments is indicated by a dashed lineon IL1b-immunostained tumor sections. Scale bars, 50 mm.






increased caspase-1 activity, caused by elevated ERK5 signaling inkeratinocytes, independently of oncogenic mutation, may besufficient to trigger a nonresolving inflammatory condition thatincreases cancer risk. This raises the interesting possibility ofpotential prophylactic applications for anti-ERK5 therapy incancer development. Furthermore, consistent with the require-ment of ERK5 for caspase-1–mediated cleavage of pro-IL1b,tumor regression associated with the epidermal loss of ERK5 wasaccompaniedwith the depletion of secreted IL1b in the papillomaepithelium. It is well established that IL1b can act on stromal cells(fibroblasts and inflammatory cells) to induce the production of acascade of inflammatory cytokines but also, adhesion moleculesand angiogenic factors to increase tumor invasiveness and metas-tasis (27, 33). Therefore, impairing tumoral IL1b signaling viaanti-ERK5 therapy may also yield therapeutic results in the pre-vention of metastatic progression.

Collectively, genetic evidence that the functional loss of ERK5in keratinocytes causes, simultaneously, the downregulation ofseveral inflammatory mediators overexpressed in various humancancers, suggest that an anti-ERK5 therapymay have the potentialto be clinically more efficacious than therapies directed againstindividual inflammatory mediators. Furthermore, the pathwayorganized in a typical three-tiered kinase (MAPKKK, MAPKK, andMAPK) cascade offers multiple potential targets, highly suitablefor drug design. In particular, an alternative therapeutic strategytargeting MEK5, instead of ERK5, may allow the specific inhibi-tion of pathologic ERK5 activation with reduced undesirable sideeffects. Indeed, while the systemic administration of XMD8-92, aselective and potent ATP-competitive inhibitor of ERK5, had noapparent adverse effects inmice, the specific loss of ERK5 signalingin endothelial cells during embryogenesis or in adult miceresulted in lethality as a consequence of vascular instability(8, 34). A similar approach has been used for the ERK1/2 MAPKsignaling pathway, where most current efforts target theirupstream activators, namely MEK1/2, rather than inhibitingERK1/2 directly. A clinical study involving patients with cancerhas confirmed decreased ERK1/2 phosphorylation in tumors andseveral partial remissions when MEK1/2 were inhibited (35).Hence, the rationale for MAPK kinase blockade in humans isquite clear.

To conclude, we propose that ERK5 is part of a core signaltransduction pathway that drives the expression of a specificsubset of inflammatory mediators in epithelial cells that canact in an autocrine and paracrine manner to initiate animmune response. As such, anti-ERK5 therapy provides a newopportunity to neutralize inflammation associated withcancer.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: K.G. Finegan, C. TournierDevelopment of methodology: K.G. Finegan, A.M. JordanAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): K.G. Finegan, C.C. DaviesAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): K.G. Finegan, D. Perez-Madrigal, C. TournierWriting, review, and/or revision of themanuscript: K.G. Finegan, A.M. Jordan,C. TournierAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): K.G. Finegan, D. Perez-Madrigal, C. TournierStudy supervision: K.G. Finegan, C. TournierOther (synthesis of compounds used in the study): J.R. HitchinOther (provision of chemical matter used to obtain the key results): A.M.Jordan

AcknowledgmentsThe authors thank Peter March (Bioimaging Facility, University of Man-

chester) for very helpful advice with the microscopy and the staff at theUniversity of Manchester Biological Services Facility for looking after themice.

Grant SupportThis work was supported by grants mainly fromWorldwide Cancer Research

(10-0134) and partly from Cancer Research UK (C18267/A11727) and theWellcome Trust (097820/Z/11/B) .

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received October 24, 2013; revised November 10, 2014; accepted November13, 2014; published OnlineFirst February 3, 2015.

References1. Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related inflammation.

Nature 2008;454:436–44.2. Balkwill FR, Mantovani A. Cancer-related inflammation: common themes

and therapeutic opportunities. Semin Cancer Biol 2012;22:33–40.3. DeNardoDG, BrennanDJ, Rexhepaj E, Ruffell B, Shiao SL,Madden SF, et al.

Leukocyte complexity predicts breast cancer survival and functionallyregulates response to chemotherapy. Cancer Discov 2011;1:54–67.

4. Mantovani A. Cancer: inflaming metastasis. Nature 2009;457:36–7.5. Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and cancer.

Cell 2010;140:883–99.6. Cataisson C, Salcedo R, Hakim S, Moffitt BA, Wright L, Yi M, et al. IL-1R-

MyD88 signaling in keratinocyte transformation and carcinogenesis. J ExpMed 2012;209:1689–702.

7. Hayashi M, Fearns C, Eliceiri B, Yang Y, Lee JD. Big mitogen-activatedprotein kinase 1/extracellular signal-regulated kinase 5 signaling path-way is essential for tumor-associated angiogenesis. Cancer Res 2005;65:7699–706.

8. Yang Q, Deng X, Lu B, Cameron M, Fearns C, Patricelli MP, et al. Phar-macological inhibition of BMK1 suppresses tumor growth through pro-myelocytic leukemia protein. Cancer Cell 2010;18:258–67.

9. Lochhead PA, Gilley R, Cook SJ. ERK5 and its role in tumour development.Biochem Soc Trans 2012;40:251–6.

10. Finegan KG, Wang X, Lee EJ, Robinson AC, Tournier C. Regulation ofneuronal survival by the extracellular signal-regulated protein kinase 5.Cell Death Differ 2009;16:674–83.

11. Finegan KG, Tournier C. Themitogen-activated protein kinase kinase 4 hasa pro-oncogenic role in skin cancer. Cancer Res 2010;70:5797–806.

12. Kayahara M, Wang X, Tournier C. Selective regulation of c-jun geneexpression by mitogen-activated protein kinases via the 12-o-tetradeca-noylphorbol-13-acetate- responsive element and myocyte enhancer factor2 binding sites. Mol Cell Biol 2005;25:3784–92.

13. Hayashi M, Lee JD. Role of the BMK1/ERK5 signaling pathway: lessonsfrom knockout mice. J Mol Med 2004;82:800–8.

14. Indra AK, Li M, Brocard J, Warot X, Bornert JM, Gerard C, et al. Targetedsomatic mutagenesis in mouse epidermis. Horm Res 2000;54:296–300.

15. Abel EL, Angel JM, Kiguchi K, DiGiovanni J. Multi-stage chemical carci-nogenesis in mouse skin: fundamentals and applications. Nat Protoc2009;4:1350–62.

16. MuellerMM. Inflammation in epithelial skin tumours: old stories and newideas. Eur J Cancer 2006;42:735–44.


Finegan et al.




17. Hennings H, Shores R, Wenk ML, Spangler EF, Tarone R, Yuspa SH.Malignant conversion of mouse skin tumours is increased by tumourinitiators and unaffected by tumour promoters. Nature 1983;304:67–9.

18. Cataisson C, Ohman R, Patel G, Pearson A, Tsien M, Jay S, et al. Induciblecutaneous inflammation reveals a protumorigenic role for keratinocyteCXCR2 in skin carcinogenesis. Cancer Res 2009;69:319–28.

19. Jamieson T, Clarke M, Steele CW, Samuel MS, Neumann J, Jung A, et al.Inhibition of CXCR2 profoundly suppresses inflammation-driven andspontaneous tumorigenesis. J Clin Invest 2012;122:3127–44.

20. CataissonC, PearsonAJ, TsienMZ,Mascia F, Gao JL, Pastore S, et al. CXCR2ligands and G-CSF mediate PKCalpha-induced intraepidermal inflamma-tion. J Clin Invest 2006;116:2757–66.

21. Coussens LM, Raymond WW, Bergers G, Laig-Webster M, Behrendtsen O,Werb Z, et al. Inflammatory mast cells up-regulate angiogenesis duringsquamous epithelial carcinogenesis. Genes Dev 1999;13:1382–97.

22. Gabay C, Lamacchia C, Palmer G. IL-1 pathways in inflammation andhuman diseases. Nat Rev Rheumatol 2010;6:232–41.

23. Rundhaug JE, SimperMS, Surh I, Fischer SM. The role of the EP receptors forprostaglandin E2 in skin and skin cancer. Cancer Metastasis Rev 2011;30:465–80.

24. Lee JL, KimA, Kopelovich L, BickersDR, AtharM.Differential expression ofE prostanoid receptors in murine and human non-melanoma skin cancer.J Invest Dermatol 2005;125:818–25.

25. Dinarello CA. IL-1: discoveries, controversies and future directions. Eur JImmunol 2010;40:599–606.

26. Eder C. Mechanisms of interleukin-1beta release. Immunobiology 2009;214:543–53.

27. Apte RN, Dotan S, Elkabets M, White MR, Reich E, Carmi Y, et al. Theinvolvement of IL-1 in tumorigenesis, tumor invasiveness, metastasis andtumor-host interactions. Cancer Metastasis Rev 2006;25:387–408.

28. Ratushny V,GoberMD,Hick R, Ridky TW, Seykora JT. Fromkeratinocyte tocancer: the pathogenesis and modeling of cutaneous squamous cell car-cinoma. J Clin Invest 2012;122:464–72.

29. Shukla A,Miller JM, Cason C, SayanM,MacphersonMB, Beuschel SL, et al.Extracellular signal-regulated kinase 5: a potential therapeutic target formalignant mesotheliomas. Clin Cancer Res 2013;19:2071–83.

30. Moore RJ, Owens DM, Stamp G, Arnott C, Burke F, East N, et al. Micedeficient in tumor necrosis factor-alpha are resistant to skin carcinogenesis.Nat Med 1999;5:828–31.

31. SuganumaM, Okabe S, Marino MW, Sakai A, Sueoka E, Fujiki H. Essentialrole of tumor necrosis factor alpha (TNF-alpha) in tumor promotion asrevealed by TNF-alpha-deficient mice. Cancer Res 1999;59:4516–8.

32. Cataisson C, Pearson AJ, Torgerson S, Nedospasov SA, Yuspa SH. Proteinkinase C alpha-mediated chemotaxis of neutrophils requires NF-kappa Bactivity but is independent of TNF alpha signaling in mouse skin in vivo.J Immunol 2005;174:1686–92.

33. Voronov E, Shouval DS, Krelin Y, Cagnano E, Benharroch D, Iwakura Y,et al. IL-1 is required for tumor invasiveness and angiogenesis. Proc NatlAcad Sci U S A 2003;100:2645–50.

34. Hayashi M, Kim SW, Imanaka-Yoshida K, Yoshida T, Abel ED, Eliceiri B,et al. Targeted deletion of BMK1/ERK5 in adult mice perturbs vascularintegrity and leads to endothelial failure. J Clin Invest 2004;113:1138–48.

35. KohnoM, Tanimura S,Ozaki K. Targeting the extracellular signal-regulatedkinase pathway in cancer therapy. Biol Pharm Bull 2011;34:1781–4.






2015;75:742-753. Published OnlineFirst February 3, 2015.Cancer Res Katherine G. Finegan, Diana Perez-Madrigal, James R. Hitchin, et al. ERK5 Is a Critical Mediator of Inflammation-Driven Cancer

Updated version

10.1158/0008-5472.CAN-13-3043doi:

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/75/4/742.full#ref-list-1

This article cites 35 articles, 12 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/75/4/742.full#related-urls

This article has been cited by 10 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/75/4/742To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-3043

http://cancerres.aacrjournals.org/content/75/4/742.full#ref-list-1

http://cancerres.aacrjournals.org/content/75/4/742.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/75/4/742


Documents

ERK5IsaCriticalMediatorofInﬂammation-Driven Cancer · Student t test or ANOVA (one- or two-way) depending on the typeofdata. , P