Advertisement

Negative Feedback Mechanisms Surpass the Effect of Intrinsic EGFR Activation during Skin Chemical Carcinogenesis

  • Maik Dahlhoff
    Affiliations
    Institute of Molecular Animal Breeding and Biotechnology and the Laboratory for Functional Genome Analysis, Gene Center, LMU Munich, Munich, Germany
    Search for articles by this author
  • Christian Rose
    Affiliations
    Department of Dermatology, University Hospital Schleswig-Holstein, University of Lübeck, Lübeck, Germany
    Search for articles by this author
  • Martin Hrabé de Angelis
    Affiliations
    Institute of Experimental Genetics, Helmholtz Zentrum München, German Research Center for Environmental Health (GmbH), Neuherberg, Germany

    Institute of Experimental Genetics, Technical University of Munich, Munich, Germany
    Search for articles by this author
  • Eckhard Wolf
    Affiliations
    Institute of Molecular Animal Breeding and Biotechnology and the Laboratory for Functional Genome Analysis, Gene Center, LMU Munich, Munich, Germany
    Search for articles by this author
  • Marlon R. Schneider
    Correspondence
    Address reprint requests to Marlon R. Schneider, D.V.M., Gene Center, LMU Munich, Feodor-Lynen-Str. 25, 81377, Munich, Germany
    Affiliations
    Institute of Molecular Animal Breeding and Biotechnology and the Laboratory for Functional Genome Analysis, Gene Center, LMU Munich, Munich, Germany
    Search for articles by this author
Published:February 02, 2012DOI:https://doi.org/10.1016/j.ajpath.2011.12.017
      The negative feedback regulation of epidermal growth factor receptor (EGFR) and other tyrosine kinase receptors, including receptor dephosphorylation and endocytosis followed by degradation, is becoming recognized as a major determinant of receptor function. To evaluate the significance of the negative regulation of EGFR during carcinogenesis in vivo, we subjected the mutant mouse line Dsk5, in which the intrinsic activation of the receptor due to a point mutation is normally counterbalanced by increased posttranslational receptor down-regulation, to skin chemical carcinogenesis. Dsk5 mice showed reduced tumor numbers and tumor burden compared with control littermates, and Dsk5-derived tumors showed a reduction in the activation and total levels of EGFR. Furthermore, the transcript levels of several molecules known to act as negative regulators of EGFR were significantly increased in Dsk5-derived tumors. Another intriguing observation was the appearance of tumors with sebaceous differentiation in the ears of Dsk5 mice after chemical carcinogenesis. Further studies are necessary to reveal whether these tumors represent a cell type–specific evasion from EGFR negative feedback machinery. In conclusion, this study reveals that several negative feedback regulators contribute to suppression of the intrinsic activation of mutant EGFR during skin carcinogenesis, stressing the potential exploitation of negative regulators as either therapeutic targets or diagnostic tools in cancer and other diseases.
      The epidermal growth factor receptor (EGFR) system exerts multiple actions during development in adult tissue homeostasis and in diseases such as cancer.
      • Yarden Y.
      • Sliwkowski M.X.
      Untangling the ErbB signalling network.
      • Holbro T.
      • Hynes N.E.
      ErbB receptors: directing key signaling networks throughout life.
      • Citri A.
      • Yarden Y.
      EGF-ERBB signalling: towards the systems level.
      • Yarden Y.
      • Shilo B.Z.
      SnapShot: EGFR signaling pathway.
      These actions are predominantly the result of forward signaling by proteins composing numerous signaling cascades and immediate early genes encoding transcription factors. Recently, negative feedback (or inhibitory feedback) regulation of EGFR and other receptor tyrosine kinases, including receptor dephosphorylation and endocytosis followed by degradation, has become recognized as a major determinant of receptor signaling outcome.
      • Gotoh N.
      Feedback inhibitors of the epidermal growth factor receptor signaling pathways.
      • Avraham R.
      • Yarden Y.
      Feedback regulation of EGFR signalling: decision making by early and delayed loops.
      In accordance with their role in negative regulation of cell signaling, components of negative feedback pathways are often down-regulated in certain tumor types, with obvious clinical relevance. A particularly illustrative example is a phase I clinical trial of rapamycin, an inhibitor of the mammalian target of rapamycin, in patients with recurrent phosphatase and tensin homolog–deficient glioblastoma.
      • Cloughesy T.F.
      • Yoshimoto K.
      • Nghiemphu P.
      • Brown K.
      • Dang J.
      • Zhu S.
      • Hsueh T.
      • Chen Y.
      • Wang W.
      • Youngkin D.
      • Liau L.
      • Martin N.
      • Becker D.
      • Bergsneider M.
      • Lai A.
      • Green R.
      • Oglesby T.
      • Koleto M.
      • Trent J.
      • Horvath S.
      • Mischel P.S.
      • Mellinghoff I.K.
      • Sawyers C.L.
      Antitumor activity of rapamycin in a Phase I trial for patients with recurrent PTEN-deficient glioblastoma.
      Also, EGFR mutations characterized by basal, ligand-independent receptor activation that is sufficient to initiate downstream pathways but not strong enough to recruit signal transduction protein CBL-C (CBLC) to trigger receptor ubiquitinylation and subsequent degradation are frequently observed in lung cancer cells.
      • Shtiegman K.
      • Kochupurakkal B.S.
      • Zwang Y.
      • Pines G.
      • Starr A.
      • Vexler A.
      • Citri A.
      • Katz M.
      • Lavi S.
      • Ben Basat Y.
      • Benjamin S.
      • Corso S.
      • Gan J.
      • Yosef R.B.
      • Giordano S.
      • Yarden Y.
      Defective ubiquitinylation of EGFR mutants of lung cancer confers prolonged signaling.
      Thus, from a pathologic perspective, negative feedback regulation of receptor activity is likely to become an important factor for evaluating disease resistance to new drugs or for determining specific drug combinations.
      The skin is one of the tissues in which EGFR signaling is of prominent importance, and its deregulation is associated with several diseases, such as cancer, psoriasis, and defective wound healing.
      • Schneider M.R.
      • Werner S.
      • Paus R.
      • Wolf E.
      Beyond wavy hairs: the epidermal growth factor receptor and its ligands in skin biology and pathology.
      The N-ethyl-N-nitrosourea–induced Dsk5 mutation, recovered because affected mice displayed increased skin pigmentation, is characterized by a T→A transversion in the kinase domain of the Egfr gene leading to a Leu→Gln substitution.
      • Fitch K.R.
      • McGowan K.A.
      • van Raamsdonk C.D.
      • Fuchs H.
      • Lee D.
      • Puech A.
      • Herault Y.
      • Threadgill D.W.
      • Hrabe d.A.
      • Barsh G.S.
      Genetics of dark skin in mice.
      The mutation results in a considerable increase in basal EGFR signaling (estimated to be 5- to 10-fold higher than normal in in vitro studies) with a concomitant reduction in EGFR steady state protein levels.
      • Fitch K.R.
      • McGowan K.A.
      • van Raamsdonk C.D.
      • Fuchs H.
      • Lee D.
      • Puech A.
      • Herault Y.
      • Threadgill D.W.
      • Hrabe d.A.
      • Barsh G.S.
      Genetics of dark skin in mice.
      In fact, the posttranslational EGFR down-regulation probably explains the mild phenotype of the mice since such a high increase in EGFR activity would most likely immediately affect the health of mutant animals in different ways. For example, targeted overexpression of the EGFR ligand transforming growth factor-α in the skin of transgenic mice induces epidermal thickening and the development of papillomas,
      • Vassar R.
      • Fuchs E.
      Transgenic mice provide new insights into the role of TGF-α during epidermal development and differentiation.
      • Dominey A.M.
      • Wang X.J.
      • King Jr, L.E.
      • Nanney L.B.
      • Gagne T.A.
      • Sellheyer K.
      • Bundman D.S.
      • Longley M.A.
      • Rothnagel J.A.
      • Greenhalgh D.A.
      • Roop D.R.
      Targeted overexpression of transforming growth factor α in the epidermis of transgenic mice elicits hyperplasia, hyperkeratosis, and spontaneous, squamous papillomas.
      overexpression of the ligand amphiregulin results in psoriatic lesions,
      • Cook P.W.
      • Piepkorn M.
      • Clegg C.H.
      • Plowman G.D.
      • DeMay J.M.
      • Brown J.R.
      • Pittelkow M.R.
      Transgenic expression of the human amphiregulin gene induces a psoriasis-like phenotype.
      • Cook P.W.
      • Brown J.R.
      • Cornell K.A.
      • Pittelkow M.R.
      Suprabasal expression of human amphiregulin in the epidermis of transgenic mice induces a severe, early-onset, psoriasis-like skin pathology: expression of amphiregulin in the basal epidermis is also associated with synovitis.
      and ubiquitous overexpression of the EGFR ligand epigen causes epidermal thickening and highly enlarged sebaceous glands.
      • Dahlhoff M.
      • Muller A.K.
      • Wolf E.
      • Werner S.
      • Schneider M.R.
      Epigen transgenic mice develop enlarged sebaceous glands.
      Thus, the dynamic balance between EGFR forward signaling derived from the mutation and its extensive compensation by negative feedback mechanisms observed in Dsk5 mice seems to be a suitable model for studying EGFR negative feedback regulation.
      Herein, we asked whether a perturbation in this balance, brought about by tumorigenesis, would result in a hypermorphic or hypomorphic phenotype, reflecting enhanced or reduced EGFR activity. For this purpose, we used skin multistage chemical carcinogenesis, a widely used model involving initiation of epidermal keratinocytes by a genotoxic chemical followed by repeated promotion stimuli.
      • Yuspa S.H.
      • Poirier M.C.
      Chemical carcinogenesis: from animal models to molecular models in one decade.
      • Abel E.L.
      • Angel J.M.
      • Kiguchi K.
      • Digiovanni J.
      Multi-stage chemical carcinogenesis in mouse skin: fundamentals and applications.

      Materials and Methods

      Animals

      Dsk5 mice were donated by the Helmholtz Zentrum München, German Research Center for Environmental Health (GmbH) (Neuherberg, Germany). The animals were maintained in the C57BL/6N background under specific pathogen-free conditions in a closed barrier facility with free access to a standard rodent diet and water. Tail tip DNA was used for genotyping by PCR as previously described.
      • Fitch K.R.
      • McGowan K.A.
      • van Raamsdonk C.D.
      • Fuchs H.
      • Lee D.
      • Puech A.
      • Herault Y.
      • Threadgill D.W.
      • Hrabe d.A.
      • Barsh G.S.
      Genetics of dark skin in mice.
      All the experiments were approved by the local veterinary authority (Committee on Animal Health and Care of the local governmental body of the State of Upper Bavaria) and were performed according to the German Animal Welfare Act.

      Chemical Carcinogenesis and Tissue Processing

      Chemical carcinogenesis was performed according to internationally accepted standards.
      • Abel E.L.
      • Angel J.M.
      • Kiguchi K.
      • Digiovanni J.
      Multi-stage chemical carcinogenesis in mouse skin: fundamentals and applications.
      In experiment I, 7-week-old Dsk5 females and control littermates were shaved on their backs and received a single application (400 nmol) of the initiating agent 7,12-dimethylbenz(a)anthracene (DMBA; Sigma-Aldrich Chemie GmbH, Munich, Germany). This was followed by multiple applications (twice per week, 10 nmol each) of the promoting agent 12-O-tetradecanoylphorbol-13-acetate (TPA; Sigma-Aldrich Chemie GmbH) for 24 weeks; the mice were then euthanized (Figure 1A). The design of experiment II was the same as that of experiment I (Figure 1A) but without applications of TPA. All the mice were examined every week for the development of papillomas. After euthanasia, tumors and skin samples were fixed in 4% paraformaldehyde (Sigma-Aldrich Chemie GmbH), dehydrated, and embedded in paraffin or snap frozen and stored at −80°C.
      Figure thumbnail gr1
      Figure 1Tumor growth is significantly reduced in Dsk5 mice after multistage chemical carcinogenesis. A: Schematic representation of the protocols for chemical carcinogenesis used in Experiments I and II. B: Evaluation of the tumor incidence revealed no differences between groups. C: Tumor numbers are significantly reduced in Dsk5 mice compared with control littermates (group × time: P < 0.0001). D: The tumor burden is significantly reduced in Dsk5 mice (group × time: P < 0.0001). Histologic analysis of H&E-stained sections revealed no differences in tumor morphologic features between control (E) and Dsk5 (F) mice. Data are mean ± SEM; 12 control and 19 Dsk5 mice per group. *P < 0.05, **P < 0.01 by Student's t-test. Scale bars: 200 μm.

      Immunohistochemical and Histologic Analyses

      H&E-stained sections were used for histopathologic analysis. For the detection of proliferating cells, immunohistochemical staining of Ki-67 was performed. Briefly, sections were boiled for antigen retrieval in 10 mmol/L sodium citrate buffer (pH 6.0), and the endogenous peroxidase was blocked with 3% H2O2 for 15 minutes. Slides were blocked with 5% rabbit serum and were incubated overnight at 4°C with a rat anti–Ki-67 antibody (Dako Denmark A/S, Glostrup, Denmark) diluted 1:200. After washing in Tris-buffered saline solution, the slides were incubated for 1 hour with a secondary biotin-conjugated rabbit anti-rat antibody (Dako Denmark A/S) diluted 1:200 and then for 30 minutes with streptavidin-biotin complex. 3,3′-diaminobenzidine (Kem-En-Tec Diagnostics A/S, Taastrup, Denmark) was used as chromogen. The procedure for immunohistochemical analysis against perilipin 2 was similar to that for Ki-67 staining using guinea pig anti-perilipin 2 antibody (Progen Biotechnik GmbH, Heidelberg, Germany) diluted 1:250 and rabbit anti-guinea pig (Dako Denmark A/S) diluted 1:250 as secondary antibody.
      For the quantitative evaluation of sebaceous gland area and proliferation index, three different sections from each ear of 6-month-old control (n = 4) and Dsk5 (n = 5) females that never received DMBA or TPA treatment were H&E stained. Pictures covering a length of 4.33 mm of epidermis were taken using a ×200 magnification lens and a Leica DFC425C digital camera (Leica Microsystems GmbH, Wetzlar, Germany), resulting in a total length of 26 mm of ear skin per animal to be measured. The area of all visible sebaceous glands was recorded using LAS Software version 3.8.0 (Leica Microsystems GmbH) and was used to calculate the mean gland area. For determining the proliferation index, the total number of cells and the number of proliferating cells per sebaceous gland were determined on the same samples after staining for Ki-67.

      Analysis of the Hras1 Mutation

      Genomic DNA was isolated from tail skin, back tumors, and ear tumors. Detection of the Hras1 mutation was performed as described previously.
      • Nagase H.
      • Mao J.H.
      • Balmain A.
      Allele-specific Hras mutations and genetic alterations at tumor susceptibility loci in skin carcinomas from interspecific hybrid mice.
      The PCR product flanking the mutations was digested with the restriction enzyme XbaI at 37°C for 3 hours. The digest was electrophoresed on a 3% high-resolution agarose gel (Sigma-Aldrich Chemie GmbH).

      Western Blot Analysis

      The Western blot experiments have been described in detail elsewhere.
      • Dahlhoff M.
      • Algul H.
      • Siveke J.T.
      • Lesina M.
      • Wanke R.
      • Wartmann T.
      • Halangk W.
      • Schmid R.M.
      • Wolf E.
      • Schneider M.R.
      Betacellulin protects from pancreatitis by activating stress-activated protein kinase.
      Briefly, samples were homogenized in extraction buffer, and equal amounts of protein were electrophoresed on 12% polyacrylamide-SDS gels and then blotted to polyvinylidene difluoride membranes (GE Healthcare, Munich, Germany). Incubation with primary antibodies was performed overnight at 4°C. The following antibodies, diluted in 5% dry milk, were used: goat anti–phospho-EGFR Tyr 1173 (1:250; #sc-12351; Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-EGFR (1:500; #sc-03; Santa Cruz Biotechnology), rabbit anti-HGS (1:500; #sc-30221; Santa Cruz Biotechnology), rabbit anti-LRIG1 (1:500; #sc-134435; Santa Cruz Biotechnology), and rabbit anti–α-tubulin (1:5000; #2125; Cell Signaling Technology Inc., Beverly, MA). Appropriate horseradish peroxidase–conjugated secondary antibodies were used. Immunoreactive bands were visualized by chemiluminescence using an ECL kit (GE Healthcare). Band intensities were quantified using the ImageQuant software package version 5.0 (Molecular Dynamics, Krefeld, Germany).

      Quantitative RT-PCR

      Tumors were homogenized in TRIzol (Invitrogen, Karlsruhe, Germany) for RNA isolation. RNA samples, 5 μg, were reverse transcribed in a final volume of 35 μL using SuperScript II Reverse Transcriptase (Invitrogen) according to the manufacturer's instructions. Quantitative mRNA expression analysis was performed by real-time quantitative RT-PCR (RT-qPCR) using the LightCycler 480 System and the LightCycler 480 Probes Master (Roche Diagnostics GmbH, Penzberg, Germany). The final primer concentration was 0.5 μmol/L, and the probe concentration was 0.2 μmol/L. The final reaction volume was 10 μL, and cycle conditions were 95°C for 5 minutes for the first cycle, followed by 45 cycles of 95°C for 10 seconds, 60°C for 15 seconds, and 72°C for 1 second. Quantitative values were obtained from the CT number at which the increase in the signal associated with the exponential growth of PCR products begins to be detected. Transcript copy numbers were normalized to Actb mRNA copies. Results are expressed as fold differences in target gene expression relative to Actb transcripts. The ΔCT value of the sample was determined by subtracting the average CT value of the target gene from the average CT value of the Actb gene. Probes were labeled with the reporter dye FAM at the 5′ end and with the quencher dye TAMRA at the 3′ end. For each primer pair, we performed no-template control and no-RT control assays, which produced negligible signals that were usually >40 in CT value. Experiments were performed in duplicate for each sample. Sequences of the primers (Thermo Fisher Scientific, Schwerte, Germany) and probes (Roche Diagnostics GmbH) for the target transcripts are given in Table 1.
      Table 1Sequences of the Primers and Probes Used for RT-qPCR Analysis
      ACTB
       Forward primer5′-CGTGAAAAGATGACCCAGATCA-3′
       Reverse primer5′-CACAGCCTGGATGGCTACGT-3′
       Fluorogenic probe5′-TTTGAGACCTTCAACACCCCAGCCA-3′
      PTPRK
       Forward primer5′-TCCTTGCGATAAATCTCCTCA-3′
       Reverse primer5′-TCTGCCCAGCATTGACCT-3′
       Fluorogenic probe5′-TGGTGATG-3′
      PTPN1
       Forward primer5′-GGGCCCTTTACCAAACACA-3′
       Reverse primer5′-CTGCTCTTCTGCTCCCACA-3′
       Fluorogenic probe5′-CTGGGAGA-3′
      PTPRJ
       Forward primer5′-TGAACATCAGTGCCTCAAGC-3′
       Reverse primer5′-TACGGGACCCATCGTAATTG-3′
       Fluorogenic probe5′-GACCTGGA-3′
      ERRFI1
       Forward primer5′-CTGTTGTAGCAAGAGGCTTTCA-3′
       Reverse primer5′-TCAGACCCCAAACACTGATG-3′
       Fluorogenic probe5′-GCAGTGGA-3′
      CBLC
       Forward primer5′-CCTGTACCAGCTCACCAAGG-3′
       Reverse primer5′-GAAGCACGCACCGAACTC-3′
       Fluorogenic probe5′-GGAGGCAG-3′
      NEDD8
       Forward primer5′-TGCTAATTAAAGTGAAGACGCTGA-3′
       Reverse primer5′-TTGTCTGTGGGTTCGATGTC-3′
       Fluorogenic probe5′-GGAAGGAG-3′
      LRIG1
       Forward primer5′-GCCTTCTCCTACGTTAGCTTCA-3′
       Reverse primer5′-TCAGTAGATATGGCGTCTTCCA-3′
       Fluorogenic probe5′-CAGCCAG-3′
      HGS
       Forward primer5′-CTTCTGTGGCAAGTGCTCCT-3′
       Reverse primer5′-TCCTTCTCAATGCCGAACTT-3′
       Fluorogenic probe5′-CTCCACCA-3′
      STAMBP
       Forward primer5′-GCCTTTGAGGAGATGATCCA-3′
       Reverse primer5′-TTGAACAATTTTCAGCCGTTC-3′
       Fluorogenic probe5′-AGGAGCTG-3′
      USP8
       Forward primer5′-CTGAAATGGCTCCTTCGTCT-3′
       Reverse primer5′-CTGGGGCTTGACTTTGTGAG-3′
       Fluorogenic probe5′-GGAAGGAG-3′

      Statistical Analysis

      Data on the number of tumors per mouse and tumor burden were statistically evaluated by analysis of variance (PROC MIXED; SAS, release 8.2; SAS Institute, Inc., Cary, NC), taking the fixed effects of group (Dsk5 or control), time (relative to tumor initiation), and the group × time interaction as well as the random effect of animal into account. Differences in the numbers of tumors per mouse and in the tumor burden between Dsk5 and control mice for each time point were additionally tested for significance using a Student's t-test (GraphPad Prism version 5; GraphPad Software Inc., San Diego, CA). Differences in tumor-free lifetime were evaluated using Kaplan-Meyer survival analysis (SPSS, version 18.0; SPSS Inc., Chicago, IL). χ2 tests were used to test differences in tumor incidence between groups at specific time points. RT-qPCR values were related to the mean value of the control group and were compared by Mann-Whitney U-tests (GraphPad Prism). Data are presented as mean ± SEM (tumors per mouse and tumor burden) or as box plots with medians (RT-qPCR). Densitometric data from Western blot analysis using four back skin or three tumor samples per group and the proliferation index (n = 4 mice per group) were analyzed using Student's t-test. P < 0.05 was considered significant.

      Results

      Dsk5 Mice Show Impaired Tumor Development despite Intrinsic EGFR Activation

      In experiment I, including a single DMBA application followed by repeated administration of TPA (Figure 1A), tumors became visible on the backs of Dsk5 (n = 19) and control (n = 12) mice from weeks 6 to 7 after DMBA treatment, and the proportion of tumor-bearing mice did not differ between groups over time (Figure 1B; P = 0.405). The number of tumors per mouse (Figure 1C) increased significantly with time in both groups (effect of time: P < 0.0001), but the number of tumors was significantly lower in Dsk5 mice than in controls over time (effect of group × time: P < 0.0001). This difference was reflected in a significantly lower tumor burden (Figure 1D) in Dsk5 mice compared with controls (effect of group: P = 0.0169; effect of group × time: P < 0.0001). Histologic evaluation of H&E-stained sections of Dsk5 and control back skin tumors revealed typical papillomas with well-bordered basement membranes and surrounded by hyperplastic epidermis with orthologous keratosis (Figure 1, E and F).
      Since this rather surprising result was not predictable at the beginning of the investigation, a separate experiment (II) was initiated to evaluate the ability of the Dsk5 mutation to facilitate tumor progression without TPA treatment (Figure 1A). For this purpose, Dsk5 (n = 11) and control (n = 8) females received a single application (400 nmol) of the initiating agent DMBA without further applications of TPA. Back skin tumor development was never observed in any mouse independent of the genotype.

      Multiple Negative Feedback Mechanisms Blunt the Intrinsic Activation of EGFR in Tumors

      The intrinsic activation of EGFR due to the Dsk5 mutation has been reported to cause posttranslational down-regulation of the receptor in liver cells.
      • Fitch K.R.
      • McGowan K.A.
      • van Raamsdonk C.D.
      • Fuchs H.
      • Lee D.
      • Puech A.
      • Herault Y.
      • Threadgill D.W.
      • Hrabe d.A.
      • Barsh G.S.
      Genetics of dark skin in mice.
      Western blot analysis confirmed increased phosphorylation of EGFR in the back skin of untreated Dsk5 mice compared with control littermates, whereas a reduction in total EGFR, as reported for liver cells, was not evident (Figure 2A; see also Supplemental Figure S1A at http://ajp.amjpathol.org). Analysis of the tumors, in contrast, did not show an increase in EGFR phosphorylation in samples from Dsk5 mice (Figure 2A). Although densitometric analysis revealed increased EGFR phosphorylation in Dsk5 tumors when normalized to total receptor levels (see Supplemental Figure S1A at http://ajp.amjpathol.org), this finding results from the remarkable reduction in the levels of total EGFR compared with control tumors (Figure 2A; see also Supplemental Figure S1A at http://ajp.amjpathol.org). Accordingly, EGFR phosphorylation is significantly increased in Dsk5 back skin but is unchanged in Dsk5 tumors compared with in control samples when expressed as a ratio of tubulin levels (see Supplemental Figure S1A at http://ajp.amjpathol.org).
      Figure thumbnail gr2
      Figure 2A: Detection of phospho-EGFR (p-EGFR), total EGFR, and tubulin in normal back skin and tumors from Dsk5 mice and control littermates by Western blot analysis. BD: Quantification of the transcript levels of negative feedback molecules for receptor activation (B), early endocytosis (C), and late endocytosis (D) in tumors by RT-qPCR (expressed as changes relative to controls). *P < 0.05 by Mann-Whitney U-test. Co, control. E: Detection of LRIG1, HGS, and tubulin in normal back skin and tumors from Dsk5 mice and control littermates by Western blot analysis. PTPRK, protein tyrosine phosphatase, receptor type, K; PTPN1, protein tyrosine phosphatase, nonreceptor type 1; PTPRJ, protein tyrosine phosphatase, receptor type, J; ERRFI1, ERBB receptor feedback inhibitor 1; NEDD8, neural precursor cell expressed, developmentally down-regulated 8; STAMBP, STAM binding protein; USP8, ubiquitin-specific peptidase 8.
      In an attempt to identify the specific mechanisms and molecules involved in the suppression of EGFR activity in tumors, we evaluated the levels of several known negative regulators of tyrosine kinase receptor signaling by RT-qPCR. We adopted the nomenclature put forward by Avraham and Yarden
      • Avraham R.
      • Yarden Y.
      Feedback regulation of EGFR signalling: decision making by early and delayed loops.
      and evaluated molecules in the following categories: receptor activation (four transcripts), early receptor endocytosis (two transcripts), and late receptor endocytosis (four transcripts). As shown in Figure 2B, the levels of all examined transcripts tended to be higher in Dsk5 tumors, with statistical significance detected for Ptprk. The same tendency was observed in the categories of early (Figure 2C) and late (Figure 2D) receptor endocytosis, with significantly higher transcript levels of Cblc, Lrig1, and Hgs in Dsk5 tumors than in controls. Next, using as a criterion the RT-qPCR results and the availability of reliable antibodies, we evaluated the protein levels of a subset of these molecules by Western blot analysis. Although no difference between groups was observed for CBLC and NEDD8 (data not shown), we detected increased leucine-rich repeats and Ig-like domains protein 1 (LRIG1) and hepatocyte growth factor–regulated tyrosine kinase substrate (HGS) protein levels in Dsk5 tumors compared with in tumors from control mice (Figure 2E; see also Supplemental Figure S1B at http://ajp.amjpathol.org).

      Growth of Sebaceous Adenomas in the Ears of Dsk5 Mice

      In addition to the tumors observed on the backs of the mice in experiment I, small papillomas were observed on the ears of Dsk5 mice starting 17 weeks after DMBA application (Figure 3A). At the end of the experiment, ∼60% of the females (11 of 19) showed at least one ear tumor, whereas tumors at this site were never observed in control females. Histologic analysis of H&E-stained sections revealed an array of well differentiated sebaceous glands in every ear tumor analyzed (Figure 3B), and the presence of lipid-laden sebocytes was confirmed by the detection of the lipid droplet–associated protein perilipin 2 (Figure 3C). Ki-67 staining revealed that cells at the basal layer of the sebaceous gland acini were highly proliferative (Figure 3D). Since basal (undifferentiated) sebocytes were restricted to a single peripheral cell layer, these tumors were classified as cases of sebaceous hyperplasia.
      • Eisen D.B.
      • Michael D.J.
      Sebaceous lesions and their associated syndromes: part I.
      A similarly high proportion of ear sebaceous hyperplasia was observed in Dsk5 mice (∼80%, 9 of 11 females) treated with only DMBA (experiment II), but, again, no ear tumors were detected in control females.
      Figure thumbnail gr3
      Figure 3Development of sebaceous hyperplasia in the ears of Dsk5 mice. A: Time course of ear tumor growth in Dsk5 females and control littermates from experiment I. B: H&E-stained sections of ear tumors showed well differentiated acinolobular glandular formations consisting of cells with a typical foamy cytoplasm and small round nuclei. C: The tumors are perilipin 2 positive, confirming their sebaceous nature. D: Ki-67 staining demonstrated high proliferative activity in the cells at the periphery of the acini. E: Representative H&E-stained ear skin sections from 6-month-old control (Co) and Dsk5 mice that never received DMBA or TPA. F: Quantitative evaluation of the mean area of sebaceous glands (SGs) in the ear skin of control and Dsk5 mice. G: Quantitative evaluation of the proliferation index (Ki-67–positive nuclei/105 nuclei) of SGs in the ear skin of control and Dsk5 mice. Data are mean ± SD; four control and five Dsk5 mice per group. *P < 0.05, **P < 0.01 by Student's t-test. H: The underlying principle of the strategy used to detect the DMBA-induced Hras codon 61 mutation. A 208-bp fragment flanking the Hras codon 61 was amplified by PCR and subjected to an XbaI digest to detect the A→T transversion. I: The Hras mutation is present in ear sebaceous hyperplasia derived from experiments (Exp.) I and II. A back skin papilloma and tail skin are included as positive and negative controls, respectively. Scale bars: 200 μm (BD); 50 μm (inset of D); and 100 μmol/L (E).
      To evaluate whether a change in sebaceous gland morphologic features may be the underlying reason for the DMBA-induced sebaceous hyperplasia in Dsk5 mice, we assessed H&E-stained ear skin sections. Although no noticeable differences in sebaceous gland morphologic features were detected, the glands of Dsk5 mice tended to be larger than those of control mice (Figure 3E). Quantitative analysis confirmed this impression and revealed a statistically significant increase in the mean sebaceous gland area on the ears of Dsk5 females (Figure 3F). Also, quantitative evaluation of Ki-67–positive cells revealed a significant increase in the proliferation index of Dsk5 ear sebaceous glands compared with the ears of control littermates (Figure 3G; see also Supplemental Figure S1C at http://ajp.amjpathol.org).
      To test whether the ear sebaceous hyperplasia is derived from cells carrying the DMBA-induced Hras mutation, we analyzed the typical A→T transversion at the second position of Hras1 codon 61 (Figure 3H) by PCR and restriction fragment length polymorphism analysis. The Hras mutation was detected in every Dsk5 ear sebaceous hyperplasia entity analyzed in experiments I and II (Figure 3I).

      Discussion

      Herein, we used the N-ethyl-N-nitrosourea-induced–mutant mouse line Dsk5 to investigate whether the dynamic balance between increased intrinsic EGFR forward signaling derived from a point mutation and its extensive compensation by negative feedback mechanisms would become disrupted during multistage skin chemical carcinogenesis. To our surprise, the tumor number and the tumor burden in Dsk5 mice were significantly reduced compared with those in control littermates, suggesting that the negative feedback mechanisms surpass the intrinsic activation of the receptor. In addition, Western blot analysis revealed that Dsk5 tumors show unchanged activation of EGFR concomitantly with a clear reduction in total EGFR protein levels. This is in contrast to the situation in normal Dsk5 skin samples, in which strong intrinsic activation of EGFR in the absence of down-regulation of total EGFR was observed. This tissue- or cell type–specific difference may explain why several phenotypical skin alterations (including hyperkeratosis, epidermal thickening, and increased skin pigmentation) but no alterations in the liver of Dsk5 mice have been reported.
      • Fitch K.R.
      • McGowan K.A.
      • van Raamsdonk C.D.
      • Fuchs H.
      • Lee D.
      • Puech A.
      • Herault Y.
      • Threadgill D.W.
      • Hrabe d.A.
      • Barsh G.S.
      Genetics of dark skin in mice.
      RT-qPCR revealed a significant increase in the transcript levels of several molecules known to act as negative regulators of EGFR, including receptor-type tyrosine-protein phosphatase kappa (PTPRK), a membrane protein previously reported to dephosphorylate EGFR in human keratinocytes,
      • Xu Y.
      • Tan L.J.
      • Grachtchouk V.
      • Voorhees J.J.
      • Fisher G.J.
      Receptor-type protein-tyrosine phosphatase-kappa regulates epidermal growth factor receptor function.
      CBLC, an early mediator of EGFR ubiquitin-dependent sorting and subsequent lysosomal and proteosomal degradation,
      • Levkowitz G.
      • Waterman H.
      • Zamir E.
      • Kam Z.
      • Oved S.
      • Langdon W.Y.
      • Beguinot L.
      • Geiger B.
      • Yarden Y.
      c-Cbl/Sli-1 regulates endocytic sorting and ubiquitination of the epidermal growth factor receptor.
      and two late stimulators of EGFR endocytosis: HGS
      • Chin L.S.
      • Raynor M.C.
      • Wei X.
      • Chen H.Q.
      • Li L.
      Hrs interacts with sorting nexin 1 and regulates degradation of epidermal growth factor receptor.
      and LRIG1.
      • Gur G.
      • Rubin C.
      • Katz M.
      • Amit I.
      • Citri A.
      • Nilsson J.
      • Amariglio N.
      • Henriksson R.
      • Rechavi G.
      • Hedman H.
      • Wides R.
      • Yarden Y.
      LRIG1 restricts growth factor signaling by enhancing receptor ubiquitylation and degradation.
      • Laederich M.B.
      • Funes-Duran M.
      • Yen L.
      • Ingalla E.
      • Wu X.
      • Carraway III, K.L.
      • Sweeney C.
      The leucine-rich repeat protein LRIG1 is a negative regulator of ErbB family receptor tyrosine kinases.
      Some additional molecules, such as tyrosine-protein phosphatase non-receptor type 1 (P = 0.056) and STAM-binding protein (P = 0.093), showed an increase of expression in Dsk5 tumors close to statistical significance, indicating that further molecules are likely to be involved in the suppression of EGFR function in Dsk5 mice. Western blot analysis confirmed up-regulation of LRIG1 and HGS at the protein level.
      An additional observation during these studies was the appearance of tumors with sebaceous differentiation in the ears of Dsk5 mice following both protocols of chemical carcinogenesis (spontaneous tumors of any kind were never observed). These tumors carried the typical DMBA-induced A→T transversion at the second position of Hras codon 61. Sebaceous neoplasias include a broad spectrum of lesions ranging from benign to highly malignant.
      • Eisen D.B.
      • Michael D.J.
      Sebaceous lesions and their associated syndromes: part I.
      The tumors observed in the ears of Dsk5 mice were classified as sebaceous hyperplasia because they appeared as an array of well-differentiated sebaceous lobules with no more than two layers of peripheral undifferentiated cells. The proportion of undifferentiated basal cells and mature sebocytes in the acini is often used as a criterion to distinguish among sebaceous hyperplasia, sebaceous adenoma, and sebaceoma.
      • Eisen D.B.
      • Michael D.J.
      Sebaceous lesions and their associated syndromes: part I.
      In humans, sebaceous hyperplasia exists in several forms and localizations, but its cause remains enigmatic.
      • Eisen D.B.
      • Michael D.J.
      Sebaceous lesions and their associated syndromes: part I.
      Whereas papillomas with a sebaceous component are commonly detected after chemical carcinogenesis of the skin,
      • Sundberg J.P.
      • Sundberg B.A.
      • Beamer W.G.
      Comparison of chemical carcinogen skin tumor induction efficacy in inbred, mutant, and hybrid strains of mice: morphologic variations of induced tumors and absence of a papillomavirus cocarcinogen.
      • Zinser G.M.
      • Sundberg J.P.
      • Welsh J.
      Vitamin D(3) receptor ablation sensitizes skin to chemically induced tumorigenesis.
      the massive and dominant presence of sebaceous hyperplasia as observed in Dsk5 mice is rather unusual. However, there are reports of similar lesions after application of the strong irritant tetradecane in rabbits
      • Ito M.
      • Motoyoshi K.
      • Suzuki M.
      • Sato Y.
      Sebaceous gland hyperplasia on rabbit pinna induced by tetradecane.
      or after a high dose of DMBA in mice.
      • Rice J.M.
      • Anderson L.M.
      Sebaceous adenomas with associated epidermal hyperplasia and papilloma formation as a major type of tumor induced in mouse skin by high doses of carcinogens.
      Although the researchers classified the tumors as sebaceous adenoma, the latter study included features resembling the present study, such as the appearance of tumors at sites distant from the application spot of the initiator and no requirement of promotion by TPA.
      • Rice J.M.
      • Anderson L.M.
      Sebaceous adenomas with associated epidermal hyperplasia and papilloma formation as a major type of tumor induced in mouse skin by high doses of carcinogens.
      Further studies are necessary to clarify the specific pathways responsible for the development of sebaceous hyperplasia in DMBA-treated Dsk5 mice and whether this effect represents a cell type–specific evasion from the EGFR inhibitory feedback machinery. Quantitative evaluation revealed a significant increase in the area and in the proliferation index of sebaceous glands of Dsk5 females never treated with DMBA or TPA, suggesting that the availability of a larger number of putatively responsive cells may be a factor contributing to this intriguing phenotype.
      In conclusion, this in vivo study reveals that several negative feedback regulators contribute to suppress the intrinsic activation of mutant EGFR during skin carcinogenesis. These results highlight the potential exploitation of receptor tyrosine kinase–negative regulators as therapeutic targets or as diagnostic tools in cancer and other diseases.

      Acknowledgments

      We thank Dr. Ingrid Renner-Müller and Petra Renner for excellent animal care and Stefanie Riesemann and Josef Millauer for assistance with mouse genotyping and Western blot analysis.

      Supplementary data

      • Supplemental Figure S1

        Statistical analysis of densitometric data from the detection of phospho-EGFR (p-EGFR), total EGFR, and tubulin (A) as well as LRIG1, HGS, and tubulin (B) in normal back skin and tumors from Dsk5 mice and control littermates by Western blot analysis. C: Representative Ki-67–stained ear skin sebaceous glands from 6-month-old control (left) and Dsk5 (right) mice that never received DMBA or TPA. Data are mean ± SD; four back skin and three tumor samples per group. *P < 0.05, **P < 0.01 by Student's t-test. Scale bars: 25 μm.

      References

        • Yarden Y.
        • Sliwkowski M.X.
        Untangling the ErbB signalling network.
        Nat Rev Mol Cell Biol. 2001; 2: 127-137
        • Holbro T.
        • Hynes N.E.
        ErbB receptors: directing key signaling networks throughout life.
        Annu Rev Pharmacol Toxicol. 2004; 44: 195-217
        • Citri A.
        • Yarden Y.
        EGF-ERBB signalling: towards the systems level.
        Nat Rev Mol Cell Biol. 2006; 7: 505-516
        • Yarden Y.
        • Shilo B.Z.
        SnapShot: EGFR signaling pathway.
        Cell. 2007; 131: 1018
        • Gotoh N.
        Feedback inhibitors of the epidermal growth factor receptor signaling pathways.
        Int J Biochem Cell Biol. 2009; 41: 511-515
        • Avraham R.
        • Yarden Y.
        Feedback regulation of EGFR signalling: decision making by early and delayed loops.
        Nat Rev Mol Cell Biol. 2011; 12: 104-117
        • Cloughesy T.F.
        • Yoshimoto K.
        • Nghiemphu P.
        • Brown K.
        • Dang J.
        • Zhu S.
        • Hsueh T.
        • Chen Y.
        • Wang W.
        • Youngkin D.
        • Liau L.
        • Martin N.
        • Becker D.
        • Bergsneider M.
        • Lai A.
        • Green R.
        • Oglesby T.
        • Koleto M.
        • Trent J.
        • Horvath S.
        • Mischel P.S.
        • Mellinghoff I.K.
        • Sawyers C.L.
        Antitumor activity of rapamycin in a Phase I trial for patients with recurrent PTEN-deficient glioblastoma.
        PLoS Med. 2008; 5: e8
        • Shtiegman K.
        • Kochupurakkal B.S.
        • Zwang Y.
        • Pines G.
        • Starr A.
        • Vexler A.
        • Citri A.
        • Katz M.
        • Lavi S.
        • Ben Basat Y.
        • Benjamin S.
        • Corso S.
        • Gan J.
        • Yosef R.B.
        • Giordano S.
        • Yarden Y.
        Defective ubiquitinylation of EGFR mutants of lung cancer confers prolonged signaling.
        Oncogene. 2007; 26: 6968-6978
        • Schneider M.R.
        • Werner S.
        • Paus R.
        • Wolf E.
        Beyond wavy hairs: the epidermal growth factor receptor and its ligands in skin biology and pathology.
        Am J Pathol. 2008; 173: 14-24
        • Fitch K.R.
        • McGowan K.A.
        • van Raamsdonk C.D.
        • Fuchs H.
        • Lee D.
        • Puech A.
        • Herault Y.
        • Threadgill D.W.
        • Hrabe d.A.
        • Barsh G.S.
        Genetics of dark skin in mice.
        Genes Dev. 2003; 17: 214-228
        • Vassar R.
        • Fuchs E.
        Transgenic mice provide new insights into the role of TGF-α during epidermal development and differentiation.
        Genes Dev. 1991; 5: 714-727
        • Dominey A.M.
        • Wang X.J.
        • King Jr, L.E.
        • Nanney L.B.
        • Gagne T.A.
        • Sellheyer K.
        • Bundman D.S.
        • Longley M.A.
        • Rothnagel J.A.
        • Greenhalgh D.A.
        • Roop D.R.
        Targeted overexpression of transforming growth factor α in the epidermis of transgenic mice elicits hyperplasia, hyperkeratosis, and spontaneous, squamous papillomas.
        Cell Growth Differ. 1993; 4: 1071-1082
        • Cook P.W.
        • Piepkorn M.
        • Clegg C.H.
        • Plowman G.D.
        • DeMay J.M.
        • Brown J.R.
        • Pittelkow M.R.
        Transgenic expression of the human amphiregulin gene induces a psoriasis-like phenotype.
        J Clin Invest. 1997; 100: 2286-2294
        • Cook P.W.
        • Brown J.R.
        • Cornell K.A.
        • Pittelkow M.R.
        Suprabasal expression of human amphiregulin in the epidermis of transgenic mice induces a severe, early-onset, psoriasis-like skin pathology: expression of amphiregulin in the basal epidermis is also associated with synovitis.
        Exp Dermatol. 2004; 13: 347-356
        • Dahlhoff M.
        • Muller A.K.
        • Wolf E.
        • Werner S.
        • Schneider M.R.
        Epigen transgenic mice develop enlarged sebaceous glands.
        J Invest Dermatol. 2010; 130: 623-626
        • Yuspa S.H.
        • Poirier M.C.
        Chemical carcinogenesis: from animal models to molecular models in one decade.
        Adv Cancer Res. 1988; 50: 25-70
        • Abel E.L.
        • Angel J.M.
        • Kiguchi K.
        • Digiovanni J.
        Multi-stage chemical carcinogenesis in mouse skin: fundamentals and applications.
        Nat Protoc. 2009; 4: 1350-1362
        • Nagase H.
        • Mao J.H.
        • Balmain A.
        Allele-specific Hras mutations and genetic alterations at tumor susceptibility loci in skin carcinomas from interspecific hybrid mice.
        Cancer Res. 2003; 63: 4849-4853
        • Dahlhoff M.
        • Algul H.
        • Siveke J.T.
        • Lesina M.
        • Wanke R.
        • Wartmann T.
        • Halangk W.
        • Schmid R.M.
        • Wolf E.
        • Schneider M.R.
        Betacellulin protects from pancreatitis by activating stress-activated protein kinase.
        Gastroenterology. 2010; 138: 1585-1594
        • Eisen D.B.
        • Michael D.J.
        Sebaceous lesions and their associated syndromes: part I.
        J Am Acad Dermatol. 2009; 61: 549-560
        • Xu Y.
        • Tan L.J.
        • Grachtchouk V.
        • Voorhees J.J.
        • Fisher G.J.
        Receptor-type protein-tyrosine phosphatase-kappa regulates epidermal growth factor receptor function.
        J Biol Chem. 2005; 280: 42694-42700
        • Levkowitz G.
        • Waterman H.
        • Zamir E.
        • Kam Z.
        • Oved S.
        • Langdon W.Y.
        • Beguinot L.
        • Geiger B.
        • Yarden Y.
        c-Cbl/Sli-1 regulates endocytic sorting and ubiquitination of the epidermal growth factor receptor.
        Genes Dev. 1998; 12: 3663-3674
        • Chin L.S.
        • Raynor M.C.
        • Wei X.
        • Chen H.Q.
        • Li L.
        Hrs interacts with sorting nexin 1 and regulates degradation of epidermal growth factor receptor.
        J Biol Chem. 2001; 276: 7069-7078
        • Gur G.
        • Rubin C.
        • Katz M.
        • Amit I.
        • Citri A.
        • Nilsson J.
        • Amariglio N.
        • Henriksson R.
        • Rechavi G.
        • Hedman H.
        • Wides R.
        • Yarden Y.
        LRIG1 restricts growth factor signaling by enhancing receptor ubiquitylation and degradation.
        EMBO J. 2004; 23: 3270-3281
        • Laederich M.B.
        • Funes-Duran M.
        • Yen L.
        • Ingalla E.
        • Wu X.
        • Carraway III, K.L.
        • Sweeney C.
        The leucine-rich repeat protein LRIG1 is a negative regulator of ErbB family receptor tyrosine kinases.
        J Biol Chem. 2004; 279: 47050-47056
        • Sundberg J.P.
        • Sundberg B.A.
        • Beamer W.G.
        Comparison of chemical carcinogen skin tumor induction efficacy in inbred, mutant, and hybrid strains of mice: morphologic variations of induced tumors and absence of a papillomavirus cocarcinogen.
        Mol Carcinog. 1997; 20: 19-32
        • Zinser G.M.
        • Sundberg J.P.
        • Welsh J.
        Vitamin D(3) receptor ablation sensitizes skin to chemically induced tumorigenesis.
        Carcinogenesis. 2002; 23: 2103-2109
        • Ito M.
        • Motoyoshi K.
        • Suzuki M.
        • Sato Y.
        Sebaceous gland hyperplasia on rabbit pinna induced by tetradecane.
        J Invest Dermatol. 1985; 85: 249-254
        • Rice J.M.
        • Anderson L.M.
        Sebaceous adenomas with associated epidermal hyperplasia and papilloma formation as a major type of tumor induced in mouse skin by high doses of carcinogens.
        Cancer Lett. 1986; 33: 295-306