Keratinocyte Collagenase-1 Expression Requires an EpidermalGrowth Factor Receptor Autocrine Mechanism*

(Received for publication, August 28, 1998, and in revised form, December 2, 1998)

Brian K. Pilcher‡§, JoAnn Dumin‡, Michael J. Schwartz‡, Bruce A. Mast¶, Gregory S. Schultzi,William C. Parks‡**, and Howard G. Welgus‡

From the ‡Division of Dermatology, Department of Medicine, and **Department of Cell Biology and Physiology,Washington University School of Medicine, St. Louis, Missouri 63110 and the Departments of iObstetrics/Gynecology and¶Surgery, University of Florida Health Sciences Center, Gainesville, Florida 32610

In response to cutaneous injury, expression of colla-genase-1 is induced in keratinocytes via a2b1 contactwith native type I collagen, and enzyme activity is es-sential for cell migration over this substratum. How-ever, the cellular mechanism(s) mediating integrin sig-naling remain poorly understood. We demonstrate herethat treatment of keratinocytes cultured on type I col-lagen with epidermal growth factor receptor (EGFR)blocking antibodies or a specific receptor antagonistinhibited cell migration across type I collagen and thematrix-directed stimulation of collagenase-1 produc-tion. Additionally, stimulation of collagenase-1 expres-sion by hepatocyte growth factor, transforming growthfactor-b1, and interferon-g was blocked by EGFR inhib-itors, suggesting a required EGFR autocrine signalingstep for enzyme expression. Collagenase-1 mRNA wasnot detectable in keratinocytes isolated immediatelyfrom normal skin, but increased progressively following2 h of contact with collagen. In contrast, EGFR mRNAwas expressed at high steady-state levels in keratino-cytes isolated immediately from intact skin but was ab-sent following 2 h cell contact with collagen, suggestingdown-regulation following receptor activation. Indeed,tyrosine phosphorylation of the EGFR was evident asearly as 10 min following cell contact with collagen.Treatment of keratinocytes cultured on collagen withEGFR antagonist or heparin-binding (HB)-EGF neutral-izing antibodies dramatically inhibited the sustainedexpression (6–24 h) of collagenase-1 mRNA, whereas ini-tial induction by collagen alone (2 h) was unaffected.Finally, expression of collagenase-1 in ex vivo woundedskin and re-epithelialization of partial thickness por-cine burn wounds was blocked following treatment withEGFR inhibitors. These results demonstrate that kera-tinocyte contact with type I collagen is sufficient toinduce collagenase-1 expression, whereas sustained en-zyme production requires autocrine EGFR activation byHB-EGF as an obligatory intermediate step, therebymaintaining collagenase-1-dependent migration duringthe re-epithelialization of epidermal wounds.

Efficient repair of a cutaneous wound requires a pro-grammed series of spatially and temporally regulated events.Among these, effective proteolytic degradation of extracellularmatrix (ECM)1 macromolecules is thought to be necessary toremodel the damaged tissue, promote neovascularization, andfacilitate efficient migration of cells during re-epithelialization(1). Matrix metalloproteinases (MMPs) constitute a family ofzinc-dependent enzymes with the collective capacity to degradevirtually all ECM components (2). Although most MMPs candegrade many ECM proteins with overlapping substrate spec-ificities, degradation of fibrillar type I collagen is initiated onlyby the catalytic activity of collagenases, a subgroup of the MMPgene family.

Previous studies from our group and others have shown that,in both normally healing wounds and chronic ulcers, basalkeratinocytes at the leading edge of re-epithelialization invari-antly express collagenase-1 (3–6). Collagenase-1 expression israpidly induced in wound-edge keratinocytes after injury, per-sists during the healing phase, and ceases following completere-epithelialization (7, 8). We demonstrated that induction ofcollagenase-1 by basal keratinocytes is mediated via a2b1 in-teraction with native type I collagen (9, 10), requires tyrosinekinase activity (11), and that the activity of this MMP is es-sential for cell migration over this matrix protein (10). Al-though much is known about the role of cell:matrix interactionsregulating collagenase-1 expression by keratinocytes, the sig-naling mechanism(s) following collagen binding and integrinreceptor occupancy remain poorly understood.

The epidermal growth factor receptor (EGFR; c-erbB1/HER1) is a transmembrane cell-surface tyrosine kinase that,upon ligand binding, phosphorylates downstream effector mol-ecules, leading to changes in cell function (12). EGFR-null micedemonstrate involvement of the receptor in a broad range ofdevelopmental processes, and these mice have pronounced de-fects in epithelial cell proliferation and differentiation (13–15).Activation of the EGFR by members of the EGF family ofgrowth factors (i.e. EGF, TGF-a, HB-EGF, and amphiregulin)is associated with multiple keratinocyte functions duringwound repair, including cell proliferation, migration, and stim-ulation of a2b1 integrin expression (16, 17). Keratinocyte mi-gration is essential for effective re-epithelialization, and ex-pression of the EGFR and its ligands is up-regulated following

* This work was supported by grants from the National Institutes ofHealth and the Dermatology Foundation. The costs of publication ofthis article were defrayed in part by the payment of page charges. Thisarticle must therefore be hereby marked “advertisement” in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

§ Recipient of a National Institutes of Health K-01 Mentored Re-search Fellowship and a Research Career Development Award from theDermatology Foundation. To whom correspondence should be ad-dressed: Dermatology, BJH North, Washington University School ofMedicine, 216 S. Kingshighway, St. Louis, MO 63110. Tel.: 314-454-8290; Fax: 314-454-8293; E-mail: bpilcher@im.wustl.edu.

1 The abbreviations used are: ECM, extracellular matrix; HB, hepa-rin-binding; EGF, epidermal growth factor; EGFR, epidermal growthfactor receptor; IFN, interferon; TGF, transforming growth factor; IL,interleukin; RA, receptor antagonist; MMP, matrix metalloproteinase;GAPDH, glyceraldehyde-3-phosphate dehydrogenase; ELISA, enzyme-linked immunosorbent assay; PBS, phosphate-buffered saline; DMEM,Dulbecco’s modified Eagle’s medium; Ab, antibody; mAb, monoclonalantibody; RT-PCR, reverse transcription-polymerase chain reaction;CAT, chloramphenicol acetyltransferase.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 15, Issue of April 9, pp. 10372–10381, 1999© 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org10372

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

injury in vivo (17, 18). Indeed, exogenously administered EGFRligands (EGF, TGF-a) stimulate keratinocyte motility in vitroand re-epithelialization in vivo (19–23), and wound healing-specific keratins (K6 and K16) contain upstream regulatorysequences responsive to EGFR activation (24). Furthermore,EGFR overexpression in cultured keratinocytes enhances li-gand-mediated motility (25).

Autocrine signaling mechanisms often regulate cell functionand behavior, and, in keratinocytes, autocrine activation of theEGFR can influence epithelial homeostasis and cutaneous re-pair. Indeed, recent evidence from Stoll et al. (51) demonstratesthat heparin-binding ligands, namely HB-EGF and amphi-regulin, mediate autocrine activation of the EGFR in a skinorgan culture model, suggesting that these ligands play animportant role in the amplification and transmission of thewound healing signal. Because collagenase-1 production is ty-rosine kinase-dependent (11) and because keratinocytes ex-press both the EGFR and its various ligands during woundrepair, we reasoned that an autocrine loop mechanism mayregulate keratinocyte collagenase-1 expression following a2b1

integrin-mediated collagen binding. Here, we report that theinitial induction of collagenase-1 by keratinocytes followingcontact with type I collagen requires only integrin receptoractivation. However, autocrine activation of the EGFR by HB-EGF is required for the sustained expression of collagenase-1by keratinocytes in vitro and during the full re-epithelializationprocess in vivo.

EXPERIMENTAL PROCEDURES

Materials

Recombinant human EGF, HB-EGF, amphiregulin, and polyclonalneutralizing antiserum to HB-EGF were obtained from R & D Systems,Minneapolis, MN. Recombinant human TGF-a was purchased fromCollaborative Biomedical Products, Becton Dickinson, Inc., Bedford,MA. Human EGFR antagonist PD153035 was obtained as a generousgift from Dr. Robert Panek (Parke Davis, Inc., Ann Arbor, MI). Anotherspecific EGFR tyrosine kinase inhibitor, tyrphostin 1478, was pur-chased from Calbiochem, San Diego, CA. Recombinant IL-1a receptorantagonist (IL-1 RA) was a generous gift from Dr. David Carmichael(Synergen, Inc.). This compound is a soluble IL-1a receptor that com-petitively binds, blocking ligand binding and receptor activation (26).Polyclonal neutralizing antisera to TGF-a (clone 189-2130.1) (27), amonoclonal EGFR blocking antibody (clone 528) (28), and polyclonalEGFR Ab-4 (used for EGFR immunoprecipitation) (28) were purchasedfrom Oncogene Research Products, Cambridge, MA. Anti-phosphoty-rosine monoclonal antibody (PY-20) and anti-mouse IgG-horseradishperoxidase conjugate were purchased from Transduction Laboratories,Lexington, KY. Bovine type I collagen (Vitrogen-100) was obtained fromCelltrix Laboratories, Palo Alto, CA.

Isolation and Culture of Human Keratinocytes

Human keratinocytes were harvested from healthy adult skin fromreduction mammoplasties or abdominoplasties as described (11, 29).Briefly, the subcutaneous fat and deep dermis were removed, and theremaining tissue was incubated in 0.25% trypsin in PBS. After 16 h, theepidermis was separated from the dermis with forceps, and the kerati-nocytes were scraped into DMEM. The keratinocyte suspension wasadded to fresh DMEM supplemented with 5% fetal calf serum and 0.1%penicillin/streptomycin. Under these culture conditions, keratinocytesproliferate, migrate, differentiate, and cornify similarly to cells in vivo(29). A specified amount of keratinocyte suspension was then platedonto tissue culture dishes coated with 1 mg/ml type I collagen, which isnecessary for induction of collagenase-1 and keratinocyte adhesion (5,9, 11).

In Situ Hybridization

Collagenase-1 mRNA was detected in formalin-fixed tissue samplesby hybridization with 35S-labeled antisense RNA as described (30, 31).Punch biopsies (2 mm) of human skin were obtained and grown asexplant cultures in serum-containing DMEM for 4 days. Followingtreatment, the tissue was fixed in neutral buffered formalin for 24 hfollowed by washing in PBS and dehydration in graded ethanol. Sec-tions of tissue were hybridized with 2.5 3 104 cpm/ml 35S-labeled anti-

sense or sense RNA overnight at 57 °C. After hybridization the slideswere washed under stringent conditions, including RNase A treatment,and were processed for autoradiography. After development of thephotographic emulsion, slides were stained with hematoxylin-eosin.The specificity of the antisense RNA probe for collagenase-1 and thecomplete lack of reactivity by the sense probe have been demonstratedin previous studies (3, 7).

Migration Assays

Colony Dispersion—Primary human keratinocytes (1 3 104 cells)were plated within a siliconized cloning cylinder (6-mm inner diameter;BellCo Glass, Inc., Vineland, NJ) onto collagen-coated dishes. After a24-h incubation period to allow the cells to attach and become confluent,the cloning cylinder was removed, and cell colonies were allowed tomigrate for 96 h at 37 °C in a 5% CO2 humidified incubator. Keratino-cytes were fixed in neutral buffered formalin, stained with 1.5% Coo-massie Blue, and the area of the colony was determined by digitizedscanning analysis. Migration is expressed as the increase in colony arearelative to 0-h controls.

Colloidal Gold—Primary human keratinocytes were plated on cham-ber slides precoated with a mixture of 100 mg/ml type I collagen andcolloidal gold particles in serum-containing DMEM. Keratinocytes(;330 cells) were added to each chamber, and 20 min later, non-adherent cells were removed and the medium was replaced. Twentyhours after plating, cultures were fixed in 13 Histochoice tissue fixative(Amresco, Solon, OH), washed in PBS, and dehydrated through gradedethanol. Paths of cell migration (phagokinetic tracks) were identified asareas devoid of gold particles. A migration index was determined usingimage analysis software by measuring the area of the phagokinetictracks associated with cells in randomly chosen fields under dark fieldillumination at 1003 magnification. For each experiment, all conditionswere done in triplicate, and all experiments were repeated at least threetimes with keratinocytes from individual donors.

Enzyme-linked Immunosorbent Assay (ELISA)

The amount of collagenase-1 accumulated in keratinocyte condi-tioned medium was measured by indirect competitive ELISA (32). ThisELISA is completely specific for collagenase-1, has nanogram sensitiv-ity, and detects active and zymogen enzyme forms, as well as collagen-ase-1 bound to TIMP or bound to substrate. Results were obtained fromtriplicate determinations and were normalized to total cell protein asquantified by the BCA protein assay (Pierce) using bovine serum albu-min as a standard.

Metabolic Labeling

Post-confluent keratinocytes plated on type I collagen were culturedfor 24 h in the presence of serum-containing DMEM control or experi-mental solutions. The culture wells were then washed and replacedwith methionine-free DMEM containing 5% dialyzed fetal calf serum(to remove free amino acids), 1 mM sodium pyruvate, 2 mM L-glutamine,0.1 mM each of non-essential amino acids, 50 mCi/ml [35S]methionine(ICN Radiochemicals, Irvine CA), and the identical concentrations ofexperimental reagents. Conditioned medium was collected 24 h laterand analyzed by immunoprecipitation.

Immunoprecipitation and Western Immunoblotting

A specific polyclonal antiserum (33) was used to immunoprecipitatecollagenase-1 from keratinocyte conditioned medium as described (34).To immunoprecipitate the EGFR, cell layers were washed with PBS andtreated for 10 min at room temperature with cell lysis buffer (1.5%Triton X-100, 50 mM Tris, pH 7.5, 150 mM NaCl, 1 mM CaCl2, 1 mM

MnCl2, 1 mM MgCl2, 2 mM phenylmethylsulfonyl fluoride, 0.02 mg/mlaprotinin, and 0.01 mg/ml leupeptin). All samples were precleared withprotein A-Sepharose (Zymed Laboratories Inc., San Francisco, CA), andsupernatants were incubated with collagenase-1 or EGFR polyclonal(35) antibodies for 1 h at 37 °C then overnight at 4 °C. Immune com-plexes were precipitated with protein A-Sepharose and washed exten-sively. For visualization of collagenase-1, radiolabeled protein was re-solved by polyacrylamide gel electrophoresis and visualized byfluorography as described previously (36). Immunoprecipitated EGFRwas resolved by polyacrylamide gel electrophoresis and electrophoreti-cally transferred to Immobilon-P polyvinylidene difluoride membrane(Millipore Corp., Bedford, MA) using a semidry blotting apparatus(Bio-Rad). Tyrosine phosphorylation of the EGFR was then visualizedby incubating the membrane with anti-PY-20 (primary) and anti-mouseIgG-horseradish peroxidase (secondary) antibodies followed by detec-tion using the ECL system (Amersham Corp., Arlington Heights, IL)

Keratinocyte Collagenase-1 Expression Requires EGFR Activation 10373

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

according to the manufacturer’s instructions. Total incorporated radio-activity (new protein synthesis) was determined from keratinocyte con-ditioned medium by trichloroacetic acid precipitation as describedpreviously (37).

RNA Analysis

Collagenase-1, EGFR, EGF, TGF-a, amphiregulin, and GAPDHmRNAs were detected by modification of previously described reversetranscription-polymerase chain reaction (RT-PCR) assays (9, 38, 39).HB-EGF mRNA was amplified using primers designed with Gene-Works software (Oxford Molecular Group Inc., Campbell, CA). The59-sense primer was complementary to bases 791–810 in exon 4, andthe 39-antisense primer was complementary to bases 1017–1036 in exon6 of human HB-EGF (40). These primers amplify a fragment acrossthree exons; thus, the 246-base pair cDNA produced from HB-EGFmRNA would be easily distinguished from contaminating DNA or pre-processed mRNA. In addition, all resultant cDNAs from each of theprimer pairs contain restriction sites specific to the mRNA amplifiedand each product was subjected to digestion analysis to verify specific-ity. The primers used to amplify each mRNA, annealing temperatures,and resulting product sizes are listed in Table I.

Total RNA was isolated from cultured keratinocytes by phenol-chlo-roform extraction (41) and treated with RQ1 RNase-free DNase (Pro-mega, Madison, WI) to remove any contaminating DNA as described(42). DNase-treated RNA was reverse transcribed with random hexam-ers using kit reagents and under the manufacturer’s recommendedconditions (GeneAmp RNA PCR kit, Perkin Elmer Cetus, Norwalk, CT).Signal strength for each RT-PCR cDNA product increased exponen-tially between 21 and 29 cycles using 10 ng of DNase-treated RNA andincreased linearly between 1 and 25 ng of total RNA at 25 cycles. Toamplify each mRNA, we used 10 ng of total RNA and 25 cycles. PCRproducts were separated through a 2% agarose gel and visualized byethidium bromide staining. Specificity was determined by overnighttransfer to Hybond N1 membrane (Amersham Corp.) followed bySouthern hybridization with a radiolabeled product-specific oligonu-cleotide probe (EGF family members and EGFR) or radiolabeled cDNAprobes (collagenase-1 and GAPDH). The probes used to detect EGFfamily members and EGFR transcripts following RT-PCR were devel-oped to recognize only specific sequences that were amplified in thePCR reaction. For each oligonucleotide, a BLAST search was per-formed, and no sequence similarity to other known cDNAs was found.In addition, a parallel reaction was run without reverse transcriptase toassure that products were not generated by contaminating DNA. Oli-gonucleotide probes were labeled by terminal transferase (BoehringerMannheim), and cDNA probes were labeled by random priming (Am-ersham Corp.) with [a–32P]dCTP (ICN Radiochemicals). Following hy-bridization, the membranes were washed and exposed to x-ray film foran appropriate duration.

Transient Transfection

Expression constructs contained the chloramphenicol acetyltrans-ferase (CAT) reporter gene with a 2.2-kilobase pair fragment of thehuman collagenase-1 promoter, pCLCAT (provided by Dr. StevenFrisch, Burnham Institute, La Jolla, CA), and pSV-b-galactosidase(Promega, Madison, WI). Primary keratinocytes were transfected at50% confluence with 2 mg/well of each construct using 10 ml/well Lipo-fectAMINE (LifeTechnologies, Inc.). After 16 h, the medium was re-placed, and the cells were incubated with control or experimental solu-tions for an additional 24 h and harvested. Detection of b-galactosidaseand CAT activity was performed as described previously (37). Under theculture conditions used, we routinely achieve 85% transfection effi-ciency of keratinocytes (9, 11, 37).

In Vivo Studies

Partial thickness thermal burns were created on the dorsal skin ofpigs as described (43). Briefly, two domestic male pigs, each weighing 35pounds, were anesthetized with ketamine and xylazine, and anesthesiawas maintained with halothane inhalation. The dorsal skin was chem-ically depilitated, and six partial thickness burns measuring 3 3 3 cmwere created by contact for 10 s with a solid brass block weighing 714 gheated previously in a water bath maintained at 70 °C. The six burnswere arranged in two rows of three burns on each side of the spine.Blister roofs were removed and two burns were treated topically with 3ml of Silvadene® cream (Marion Laboratories, Kansas City, MO) con-taining 20 mg/ml tyrphostin 1478 (Calbiochem) (44). Two burns weretreated with Silvadene® cream alone, and two burns were left un-treated. Each burn was individually covered with a 6 3 6-cm2 adhesiveocclusive dressing (Steri-Drape™ 2, 3M, St. Paul, MN). Dressings wereremoved daily, the burns were retreated, and fresh dressings wereapplied. Five days after injury, the pigs were sacrificed and a full-thickness biopsy was taken diagonally across each burn, fixed in 10%buffered formalin, and embedded in paraffin. Sections were stainedwith hematoxylin and eosin. The extent of epithelial healing for eachburn was calculated by measuring the distance between intact epithe-lial edges divided by the total length of the wound and was expressed aspercentage of re-epithelialization. Epithelial healing for each of thethree treatment groups was averaged and compared for statisticalsignificance by analysis of variance and Tukey’s HSD post-test.

RESULTS

Blockade of EGFR Signaling Inhibits Migration on Type ICollagen—We used specific inhibitors of EGFR occupancy andtyrosine kinase activity to determine if signaling through thisreceptor was related to collagenase-1 induction and keratino-cyte migration across type I collagen. Migration was assessedusing colloidal gold and colony-dispersion motility assays asdescribed (10). In the colloidal gold assay, keratinocytes wereplated on chamber slides coated with a colloidal gold type Icollagen mixture, and migration was quantified 20 h later. Inthe colony dispersion assay, cells were cultured within cloningcylinders for 24 h. After cell attachment to the matrix, thecloning cylinders were removed and migration was assessed at96 h. As we have shown (10), keratinocytes migrated efficientlyover type I collagen in both migration assays (Fig. 1, A and D).

As we reported (10), treatment of keratinocytes with anaffinity-purified collagenase-1 antibody (1:4 antiserum dilu-tion), which blocks enzymatic activity, markedly inhibited cellmigration across type I collagen, reconfirming that collagen-ase-1 activity is required for motility over this matrix (Fig. 1D).Similar to collagenase-1 antiserum, the addition of PD153035(500 nM), a highly specific EGFR tyrosine kinase antagonist(45), inhibited keratinocyte migration by about 80% (Fig. 1, Aversus B and D). EGFR phosphorylation and EGFR-dependentcellular functions in vitro are inhibited by this compound atconcentrations of 40–300 nM, whereas other tyrosine kinasesare unaffected below 10 mM (45). In addition, anti-EGFR mAb528 (1.0 mg/ml), which blocks ligand binding and subsequentreceptor activation, was equally effective at inhibiting kerati-nocyte motility on collagen (Fig. 1, C and D). Addition of anIL-1a receptor antagonist (IL-1 RA) (500 ng/ml) did not affectkeratinocyte migration (Fig. 1D). Results for all conditions

TABLE ISpecifications for collagenase-1 (C’ase-1), EGF, TGF-a, amphiregulin (AR), HB-EGF, EGFR, and GAPDH RT-PCR

Transcript Forward primer Reverse primer Annealing temperature Product length

°C base pairs

C’ase-1 59-GGACTCACACCATGTGTTTTCC-39 59-GGGCTGTTCAGGGACAGAATGT-39 60 400EGF 59-AAGATGACCACCACTATTCCG-39 59-AGCGTGGCGCAGTTCCCACCA-39 45 253TGF-a 59-ATGGTCCCCTCGGCTGGACA-39 59-CTGCAGGTTCCATGGAAGCA-39 58 182AR 59-CAGTCAGAGTTGAACAGG-39 59-GCTGTGAGTTTTCATGGA-39 50 268HB-EGF 59-TGTCTGTCTGCTGGTCATCG-39 59-CAATCCTAGACGGCAACTGG-39 55 246EGFR 59-CGCTGCTGGCTGCGCTCTG-39 59-AGCCACCTCCTGGATGGTC-39 45 221GAPDH 59-CTGAGAACGGGAAGCTTGTCATCAA-39 59-GCCTGCTTCACCACCTTCTTGATGTC-39 55 610

Keratinocyte Collagenase-1 Expression Requires EGFR Activation10374

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

paralleled one another in each migration assay (Fig. 1D). Thus,inhibition of EGFR function blocks keratinocyte motility acrossa type I collagen matrix.

Blockade of EGFR Function Inhibits Keratinocyte Collagen-ase-1 Production—Because keratinocyte migration across typeI collagen is dependent on both collagenase-1 activity andEGFR function, we determined if EGFR signaling was requiredfor collagen-mediated collagenase-1 production. Human kera-tinocytes were plated on native type I collagen (all conditionsshown except for “gelatin”) or heat-denatured collagen (gelatin)and treated with inhibitors of EGFR signaling, and accumula-tion of collagenase-1 protein in the conditioned medium wasquantified by ELISA. We used the IL-1 RA as a control in theseexperiments because signaling through this receptor is re-quired for induction of collagenase-1 expression in fibroblasts(46, 47). Treatment with PD153035 or EGFR Ab 528 inhibitedcollagen-mediated collagenase-1 production in a dose-depend-ent manner, reducing enzyme levels to those detected in cellsplated on gelatin (negative control) (Fig. 2A). In contrast, theIL-1 RA did not reduce collagenase-1 expression but had aslight stimulatory effect. Keratinocyte viability was not af-

fected by treatment with either EGFR inhibitor as assessed bytotal protein synthesis (data not shown).

We also found that other known stimulators of collagenase-1expression by keratinocytes required intermediate signalingthrough the EGFR. Collagen-mediated induction of collagen-ase-1 expression was augmented by transforming growth fac-tor-b1 (TGF-b1), hepatocyte growth factor/scatter factor, phor-bol ester, and interferon-g (IFN-g). However, in the presence ofPD153035 (Fig. 2, B and C), all such soluble factor-stimulatedenzyme production was inhibited. In contrast, IL-1 RA had noeffect on collagenase-1 stimulation by each of these factors(data not shown). Therefore, agents that stimulate collagen-ase-1 production by keratinocytes, whether matrix or soluble,appear to require an obligatory intermediate EGFR signalingstep.

EGFR Blockade Inhibits Collagen-induced Collagenase-1mRNA Levels and Promoter Activity in Keratinocytes—To gainan understanding of the molecular level at which EGFR block-ade inhibits collagen-directed collagenase-1 production, totalRNA was harvested from keratinocytes grown on collagenalone, or on collagen and treated with PD153035 or EGFR mAb

FIG. 1. Blockade of EGFR signaling inhibits keratinocyte migration on type I collagen. A–C, primary human keratinocytes were platedon culture slides coated with a mixture of type I collagen (1.0 mg/ml) and colloidal gold particles, treated with vehicle control (collagen alone) (A),PD153035 (500 nM) (B), or EGFR blocking antibody 528 (1.0 mg/ml) (C), and fixed after 20 h. D, colloidal gold. Primary human keratinocytes wereplated on culture slides coated with colloidal gold and type I collagen (1.0 mg/ml) in the presence of vehicle control (collagen alone), affinity-purifiedcollagenase-1 antiserum (1:4 dilution), PD153035 (500 nM), EGFR blocking Ab 528 (1.0 mg/ml), or IL-1 RA (500 ng/ml) and were fixed 20 h later.Keratinocyte migration was quantified as described under “Experimental Procedures,” and the data are shown as means 6 S.D. of triplicatesamples from three experiments. For colony dispersion, primary human keratinocytes were cultured for 24 h within cloning cylinders on type Icollagen (1.0 mg/ml). After 24 h, the cloning cylinders were removed and the cells were allowed to migrate for an additional 96 h in the presenceof vehicle control (collagen alone), collagenase-1 antiserum (1:4 dilution), PD153035 (500 nM), EGFR blocking Ab 528 (1.0 mg/ml), or IL-1 RA (500ng/ml). The data presented are means 6 S.D. of values from three separate wells per treatment group, and migration is expressed as arbitrary unitsrelative to 0-h controls.

Keratinocyte Collagenase-1 Expression Requires EGFR Activation 10375

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

528 for 24 h. To detect collagenase-1 mRNA present in eachsample, we used a semiquantitative RT-PCR assay (9). Onlycollagenase-1-specific products are detected by Southern hy-bridization or by ethidium bromide staining. Collagenase-1mRNA was readily observed in keratinocytes cultured on col-lagen alone (Fig. 3A). In contrast, and paralleling the proteindata, treatment of keratinocytes with PD153035 (500 nM) orEGFR mAb 528 (1.0 mg/ml) inhibited collagenase-1 mRNA ex-pression by 96% and 92%, respectively, indicating pretransla-tional regulation.

To determine if collagenase-1 transcription was affected fol-lowing inhibition of EGFR signaling, keratinocytes on collagenwere transfected with a CAT expression construct containing2.2 kilobase pairs of the human collagenase-1 promoter. Asshown in previous studies, changes in the activity of this pro-moter construct parallel changes in the activity of the endoge-nous collagenase-1 gene (48), and the activity of the promoterconstruct is about 5-fold greater in keratinocytes plated onnative collagen than in cells on heat-denatured collagen (gela-tin) (9). The normalized level of CAT activity detected in kera-tinocytes plated on collagen was reduced in cells treated withPD153035 or EGFR Ab 528 by 75% and 87.5%, respectively(Fig. 3B), indicating that inhibition of EGFR function blockscollagenase-1 production by a transcriptional mechanism.

Keratinocyte Contact with Type I Collagen Induces EGFRPhosphorylation—Becuase keratinocyte contact with collagenis the primary and requisite inductive event for collagenase-1

expression (9, 10) and because EGFR blockade inhibits enzymeproduction, we reasoned that EGFR activation should followinitial matrix binding. To assess this temporal relationship, weanalyzed changes in keratinocyte steady-state mRNA levels ofcollagenase-1, EGF family members, and the EGFR at differenttime points after plating on collagen (0–48 h) (Fig. 4A). As wereported previously (9), no collagenase-1 mRNA was detectedin keratinocytes prior to matrix contact (Fig. 4A, C’ase-1, 0-h).Low levels of collagenase-1 mRNA were observed as early as2 h after contact with collagen, and expression increased mark-edly and progressively over the next 8 h. Collagenase-1 mRNAdropped to lower levels between 24 and 48 h after plating,coincident with the keratinocytes reaching confluence. Expres-sion of the mRNAs for EGF, TGF-a, and amphiregulin paral-leled that of matrix-stimulated collagenase-1 expression (Fig.4A).

In contrast, mRNAs for HB-EGF and EGFR were constitu-tively expressed at high levels in keratinocytes prior to collagenbinding (Fig. 4A). EGFR mRNA became undetectable following2 h of contact with collagen (Fig. 4A, EGFR), but then increasedprogressively from 4 to 24 h, diminishing slightly at 48 h.HB-EGF mRNA was expressed at near-constant levelsthroughout the entire time course, except for a single drop at4 h.

We next determined whether collagen binding inducedEGFR activation. Primary keratinocytes were plated on colla-gen, and total cell lysates were harvested over the next 120 min(Fig. 4B). Phosphorylated species were immunoprecipitatedwith an anti-phosphotyrosine monoclonal antibody, and phos-phorylated EGFR was detected by Western immunoblotting.

FIG. 2. Blockade of EGFR signaling inhibits matrix- and solu-ble factor-induced collagenase-1 production by keratinocytes.Primary human keratinocytes were cultured on type I collagen (colla-gen alone) or heat-denatured collagen (gelatin) until confluent. A, cellson collagen were treated with PD153035 (100 or 500 nM), EGFR block-ing antibody 528 (0.1 or 1.0 mg/ml), or IL-1 RA (250 or 500 ng/ml). B andC, keratinocytes on collagen were treated with TGF-b (25 ng/ml), hep-atocyte growth factor (25 ng/ml), phorbol ester (20 ng/ml), or IFN-g(1000 units/ml) in the presence or absence of PD153035 (500 nM).Collagenase-1 protein in the conditioned medium after 72 h was quan-tified by ELISA, and values were normalized to total protein content.Data shown are the means 6 S.D. of triplicate observations from thesame cell preparation and are representative of up to four separateexperiments.

FIG. 3. Blockade of EGFR signaling inhibits collagen-inducedcollagenase-1 mRNA levels and promoter activity in keratino-cytes. Primary human keratinocytes were cultured on type I collagen.A, keratinocytes were treated with vehicle control (collagen alone),PD153035 (100, 500 nM), or EGFR blocking Ab 528 (0.1 or 1.0 mg/ml).Total RNA was harvested 24 h later and processed for collagenase-1 orGAPDH RT-PCR as described under “Experimental Procedures.” Acontrol sample was processed without reverse transcriptase (2RT) toassure RNA purity. B, keratinocytes cultured on type I collagen weretransfected at ;60% confluence with a human collagenase-1 promoterconstruct containing the CAT reporter gene and a construct containingthe b-galactosidase gene for normalization of transfection efficiency.After 16 h, keratinocytes were washed and treated with vehicle control(collagen alone), or with media containing PD153035 or EGFR blockingAb 528 at the indicated concentrations. CAT activity was quantified byscintillation counting of non-acetylated and acetylated chloramphenicoland shown as a percent conversion to the acetylated forms normalizedto b-galactosidase activity.

Keratinocyte Collagenase-1 Expression Requires EGFR Activation10376

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

EGFR was not phosphorylated in freshly isolated keratinocytesfrom normal skin (Fig. 4B, 0 min). In contrast, marked phos-phorylation of the EGFR was observed as early as 10 minfollowing contact with collagen and persisted up to 120 min(Fig. 4B). Therefore, keratinocyte contact with native type Icollagen rapidly induces phosphorylation of the EGFR. Fur-thermore, despite EGFR mRNA down-regulation at 2 h follow-ing matrix contact (Fig. 4A), the receptor remains present andfunctional on the cell surface as evidenced by continued phos-phorylation (Fig. 4B).

EGFR Activity Mediates Sustained, but Not Early Collagen-directed Keratinocyte Collagenase-1 Expression—In the exper-iments shown in Figs. 1–3, inhibitors of EGFR function blockedcollagen-mediated collagenase-1 expression as measured byenzyme protein and mRNA levels at 24 h. We next performed atime course of collagenase-1 expression in the presence ofEGFR inhibitors to determine if the initial induction of colla-genase-1 by collagen was mediated by EGFR signaling. Freshlyisolated keratinocytes from normal skin were incubated withvehicle control or PD153035 (500 nM) for 2 h prior to plating oncollagen. After plating, total RNA was isolated over a timecourse (0–24 h), and collagenase-1 mRNA was assessed byRT-PCR. Surprisingly, the initial matrix-induced expression ofcollagenase-1 at 4 h was unaffected by inhibition of EGFRsignaling (Fig. 5, A and B, 1PD153035). In contrast, collagen-ase-1 mRNA was slightly diminished at 6 h and almost com-pletely absent at 24 h in cells treated with PD153035 whencompared with vehicle controls (Fig. 5, A and B). To ensure thatPD153035 inhibited EGFR activation, keratinocytes wereplated on collagen and treated with vehicle, 30 ng/ml EGF, orEGF 1 PD153035 (500 nM). Stimulation of collagenase-1 ex-pression by EGF was inhibited by PD153035 (data not shown).Therefore, the sustained expression of collagenase-1 mRNArequires EGFR signaling, whereas matrix contact alone is suf-ficient to induce enzyme expression.

Heparin-binding Epidermal Growth Factor Is the LigandThat Mediates EGFR-dependent Sustained Collagenase-1 Ex-

FIG. 5. Sustained keratinocyte collagenase-1 expression is me-diated via EGFR signaling. Keratinocytes were isolated from normalhuman skin as described under “Experimental Procedures” and pre-treated for 2 h with vehicle control or PD153035 (500 nM). After pre-treatment some keratinocytes were collected for RNA isolation (0 h).The remaining keratinocytes were plated on collagen-coated dishes inthe continued presence of vehicle control or PD153035 and total RNAwas isolated over a time course (0–24 h). A, collagenase-1 mRNA wasamplified by RT-PCR analysis of DNase-treated RNA and visualized bySouthern hybridization with a radiolabeled cDNA probe. Duplicatesamples were processed without reverse transcriptase (2RT) to assureRNA purity. B, collagenase-1 mRNA hybridization bands were quanti-fied by scanning digitized densitometry and values were plotted overthe time course.

FIG. 4. Collagenase-1, EGF, TGF-a, amphiregulin, HB-EGF, EGFR expression, and EGFR activation in keratinocytes cultured on type I collagen.Normal human skin was processed for keratinocyte isolation as described under “Experimental Procedures.” As the trypsin-dispersed cellsuspension was prepared, some cells were collected for RNA isolation (0 h). The remaining keratinocytes were plated on collagen-coated dishes. A,total RNA was isolated over a time course (0–48 h) and the mRNAs for collagenase-1, EGF, TGF-a, amphiregulin, HB-EGF, EGFR, and GAPDHwere amplified by RT-PCR and analyzed by Southern hybridization with product-specific cDNA (c’ase-1 and GAPDH) or oligonucleotide probes (allothers). Duplicate samples were processed without reverse transcriptase (2RT) to assure RNA purity. B, total proteins were isolated fromkeratinocytes grown on collagen (1 mg/ml) at the indicated time points and processed for immunoprecipitation of tyrosine-phosphorylated speciesfollowed by Western immunodetection of the EGFR as described under “Experimental Procedures.” Results presented in A and B are represent-ative of data obtained from at least three individual skin donors.

Keratinocyte Collagenase-1 Expression Requires EGFR Activation 10377

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

pression—Members of the EGF family, namely EGF andTGF-a, up-regulate matrix metalloproteinase expression inseveral cell types, including keratinocytes (49, 50). Further-more, collagen-directed collagenase-1 expression in keratino-cytes is enhanced by treatment with EGF (9, 10). Recent evi-dence has shown that autocrine production of HB-EGF is likelyresponsible for potentiation and transmission of the healingresponse during epidermal wound repair (51), yet effects of thiscytokine on collagenase-1 expression have not been studied.Because HB-EGF was the only EGFR ligand constitutivelyexpressed in keratinocytes from intact skin (Fig. 4A), we hy-pothesized that sustained collagen-mediated production of ke-ratinocyte collagenase-1 is regulated via a HB-EGF/EGFR au-tocrine loop mechanism. We treated primary keratinocyteswith several EGF family members to determine their effect oncollagen-directed collagenase-1 expression (Fig. 6A). Collagen-

ase-1 production, as assessed by immunoprecipitation of met-abolically labeled proteins, was induced in keratinocytes cul-tured on collagen alone, and this expression was slightlyenhanced in cells treated with EGF (Fig. 6A). Collagenase-1production was substantially increased in keratinocytestreated with HB-EGF or TGF-a, whereas cells treated withamphiregulin showed no stimulation above that induced bycollagen. Thus, of the EGF family members tested, HB-EGFand TGF-a were the most potent modulators of matrix-medi-ated collagenase-1 production by keratinocytes, making themlikely candidates for autocrine activation of the EGFR follow-ing contact with collagen.

To address this issue, keratinocytes were isolated from nor-mal human skin and pre-incubated for 2 h with vehicle controlor neutralizing antibodies to HB-EGF or TGF-a prior to platingon collagen. After plating, total RNA was harvested over thenext 0–24 h, and collagenase-1 mRNA was assessed by RT-PCR. The expression of collagenase-1 by keratinocytes wasinhibited by treatment with the HB-EGF neutralizing antibodyat both time points tested (Fig. 6B, 8 and 24 h). In contrast,collagenase-1 expression at 24 h was unaltered by TGF-a neu-tralizing antiserum (Fig. 6C). Therefore, the sustained produc-tion of keratinocyte collagenase-1 in the presence of collagenrequires an intermediate HB-EGF/EGFR autocrine signalingstep.

Blockade of EGFR Signaling Inhibits ex Vivo EpidermalCollagenase-1 Expression and Prevents Re-epithelialization ofPorcine Burn Wounds—We treated ex vivo wounded humanskin explants with inhibitors of EGFR tyrosine kinase activityto determine the requirement, at the tissue level, of EGFRactivation for keratinocyte collagenase-1 expression. Punch bi-opsies of normal human skin were cultured for 4 days withvehicle control or PD153035 (500 nM). As demonstrated by insitu hybridization of the control specimens, collagenase-1mRNA was expressed by migrating keratinocytes at the lead-ing edge of re-epithelialization (Fig. 7, A and A’). Collagenase-1was also expressed by some dermal fibroblasts. In markedcontrast to control injured skin, no expression of collagenase-1mRNA was evident in wound-edge keratinocytes in specimenstreated with PD153035, whereas fibroblast production was un-affected (Fig. 7, B and B’).

Because EGFR signaling is required for sustained keratino-cyte collagenase-1 expression both in vitro and ex vivo, andbecause the activity of this enzyme is essential for cell migra-tion across type I collagen (10), we next determined if EGFRinhibition affected re-epithelialization of in vivo wounds. Par-tial thickness burn wounds were created in pigs and weretreated with Silvadene® cream containing tyrphostin 1478 (20mg/ml), a highly potent and specific EGFR tyrosine kinaseinhibitor (52), vehicle control (Silvadene® cream), or occlusivedressing only. All wounds were covered with occlusive dressing.In tyrphostin 1478-treated wounds, epithelial healing was sig-nificantly impaired when compared with burns treated withvehicle control or covered with occlusive dressing only (Fig. 8,A–D). Specifically, wounds treated with tyrphostin 1478 hadre-epithelialized only 22 6 16% (mean 6 standard error) of thesurface of the burn, whereas wounds treated with vehicle con-trol had closed an average of 90 6 5% of the burn area (p 50.008). Burns covered with occlusive dressing only had re-epithelialized an average of 82% 6 12% of the wound surface(p 5 0.016 versus 1478-treated wounds). There was no signifi-cant difference between epithelial healing of burns treated withvehicle control or covered with occlusive dressing (p 5 0.88).

To verify that tyrphostin 1478 blocked collagenase-1 expres-sion, human keratinocytes plated on type I collagen weretreated with PD153035 (100–500 nM) or tyrphostin 1478 (0.1–

FIG. 6. Sustained collagen-directed keratinocyte collagenase-1expression is dependent on HB-EGF activity. Primary humankeratinocytes were cultured on type I collagen until confluent. A, kera-tinocytes were treated for 24 h with vehicle control (collagen alone),HB-EGF (25 ng/ml), amphiregulin (25 ng/ml), EGF (30 ng/ml), or TGF-a(30 ng/ml). Proteins were metabolically labeled in the presence of eachsolution for an additional 24 h, and conditioned medium was analyzedby immunoprecipitation for collagenase-1. Results from one cell popu-lation are shown and represent data obtained from three separate skindonors. B, keratinocytes were cultured on collagen over a time course(0–24 h) and treated with vehicle control or with a HB-EGF neutraliz-ing antibody (10 mg/ml). Total RNA was harvested at the indicated timepoints and processed for collagenase-1 or GAPDH RT-PCR as describedunder “Experimental Procedures.” A control sample was processedwithout reverse transcriptase (2RT) to assure RNA sample purity. C,keratinocytes were cultured on collagen and treated with vehicle con-trol or with a TGF-a neutralizing antibody at the indicated concentra-tions. Total RNA was harvested 24 h later and processed for collagen-ase-1 RT-PCR as described under “Experimental Procedures.” A controlsample was processed without reverse transcriptase (2RT) to assureRNA sample purity. Results presented in B and C are representative ofdata obtained from at least two separate skin donors.

Keratinocyte Collagenase-1 Expression Requires EGFR Activation10378

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

1.0 mM) for 72 h and collagenase-1 secreted into conditionedmedium was quantified by ELISA (Fig. 8E). Both PD153035and tyrphostin 1478 blocked collagen-induced collagenase-1expression in a dose-dependent manner (Fig. 8E). Taken to-gether, these findings support the hypothesis that inhibition ofre-epithelialization in porcine burn wounds was due, at least inpart, to blocking collagenase-1 production.

DISCUSSION

Activation of the EGFR in basal keratinocytes following in-jury provides several critical signals required for proper heal-ing of the tissue defect. For example, the wound-associatedkeratins 6 and 16 are markedly up-regulated (53), and theirpromoters contain regulatory elements that are responsive toEGFR activation (24). Continued signaling through the EGFRpromotes Bcl-X-L-mediated prevention of keratinocyte apopto-sis, thereby potentiating re-epithelialization by maintainingsurvival of the migrating cells (54). We show here that thesustained expression of collagenase-1 in two systems thatmimic epidermal healing (keratinocyte migration on type Icollagen and ex vivo skin explants) requires an EGFR autocrineloop mechanism. In previous studies, we showed that collagen-ase-1 expression is induced at the onset of re-epithelializationas keratinocytes migrate onto dermis, and that collagenase-1activity is required for cell migration over a type I collagensubstratum (10, 55). Thus, EGFR activation is central to yetanother important phase of the wound repair process.

Our findings that both matrix- and soluble factor-induced

collagenase-1 expression in keratinocytes require autocrinesignaling through the EGFR (Fig. 2) draw interesting parallelsto reports in human fibroblasts showing that all up-regulatorsof collagenase-1 production operate via an IL-1a/IL-1a receptorautocrine loop. Fini and colleagues (46, 47) have demonstratedthat only IL-1a directly stimulates collagenase-1 gene tran-scription in fibroblasts. All other inducers, ranging from cyto-kines such as TNF-a to phorbol esters to agents causing cy-toskeletal rearrangement, must first stimulate IL-1a genetranscription and subsequent release of IL-1a protein. Thiscytokine then binds to the cell-surface IL-1a receptor, resultingin triggering of collagenase-1 expression. Consequently, theaddition of an IL-1a receptor antagonist to fibroblast culturesblocks collagenase-1 gene expression by any stimulating agent.Our findings suggest an analogous role for HB-EGF/EGFR inkeratinocytes and offer the possibility that autocrine signalingpathways may represent a general biologic mechanism for in-ducing the expression of collagenase-1, and perhaps even otherMMPs, in different cell types.

Our results implicate HB-EGF as the key ligand that medi-ates autocrine signaling through the EGFR, resulting in sus-tained keratinocyte production of collagenase-1. Evidence dem-onstrating multiple effects of HB-EGF on keratinocytes hasaccumulated through studies of its role in epidermal woundrepair. This newer member of the EGF family is synthesized bymultiple cell types, including vascular endothelium, smoothmuscle, and keratinocytes (56). Exogenously added HB-EGF,

FIG. 7. Blockade of EGFR signaling inhibits collagenase-1 expression in ex vivo wounded epidermis. Punch biopsies (2 mm) of humanskin were treated in culture for 4 days in the presence of vehicle control (A and A’) or PD153035 (500 nM) (B and B’). The tissue was fixed, andsections were hybridized with a 35S-labeled antisense RNA probe specific for collagenase-1 mRNA. In the control section (A and A’), prominentautoradiographic signal for collagenase-1 was detected in basal keratinocytes (arrows) at the migrating front of the epidermis (E). Collagenase-1mRNA expression was also observed in several fibroblasts (arrowheads) within the dermis (D). In contrast, explants treated with PD153035 (B andB’) showed no detectable collagenase-1 mRNA signal in keratinocytes at the wound margin (arrows), whereas expression was unchanged in dermalfibroblasts (arrowheads) when compared with controls (A and A’).

Keratinocyte Collagenase-1 Expression Requires EGFR Activation 10379

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

TGF-a, or EGF promotes autoinduction of HB-EGF mRNA,suggesting that this protein is an autocrine growth factor forthese cells (56). The requirement for activation of the EGFR tosustain matrix-induced collagenase-1 production is an exampleof wound keratinocyte phenotype modulation by HB-EGF. Al-though HB-EGF mRNA is expressed in freshly isolated kera-tinocytes from normal skin, the membrane-bound form of thisprotein may not be processed and released in a soluble formuntil later in the wound healing response. In fact, other groupshave shown that soluble HB-EGF does not appear in condi-tioned medium from excisional wounds or human skin explantsuntil 24 h after injury (51, 57). These findings correlate withour observations that complete inhibition of collagenase-1 ex-pression by blocking EGFR or HB-EGF activity does not occuruntil 24 h following keratinocyte contact with collagen.

Our findings indicate that matrix-induced keratinocyte col-lagenase-1 expression involves two distinct pathways, an ini-tial response not requiring EGFR activation and a sustainedresponse obligatory to EGFR activation. Both of these re-sponses are integrin-mediated, but early collagenase-1 expres-sion is EGFR-independent. Our data demonstrate that a2b1

integrin activation alone is sufficient to induce the early (0–8h) expression of collagenase-1 in keratinocytes following con-tact with type I collagen. However, the sustained production($8 h) of this MMP requires signaling through the EGFR inaddition to a2b1 binding (Figs. 5 and 6B). Indeed, blocking a2b1

activity blocks all matrix-induced collagenase-1 production(10), whereas only the sustained expression is inhibited when

EGFR activity is blocked (this report) (Fig. 5). Stimulation byboth integrin adhesion and cell binding of a soluble factor arerequired for a variety of cellular responses during tissue mor-phogenesis. Unique to our system, however, is that both inte-grin and EGFR signaling are required only after prolongedexposure to type I collagen (the inducing stimulus).

Similar to our findings, EGFR autophosphorylation is in-duced by contact of glomerular epithelial cells with type Icollagen (58). Our data demonstrate that EGFR (and HB-EGF)mRNA is expressed in keratinocytes freshly isolated from in-tact skin and that the receptor is rapidly autophosphorylatedfollowing contact with type I collagen. These findings representfurther examples of how keratinocytes in unwounded skin are“primed” to respond to injury. Indeed, keratinocytes in intactskin express endogenous a2b1 integrin (59). Interestingly, al-though the EGFR is autophosphorylated rapidly following ke-ratinocyte binding to collagen, collagenase-1 expression doesnot require EGFR signaling until later in the healing response.We speculate, however, that other early functions of the woundkeratinocyte, such as non-enzymatic migration-related eventsor cell proliferation may require early EGFR signaling, al-though this remains to be determined.

In addition to our in vitro data, we also show a role for EGFRsignaling in the re-epithelialization of porcine burn wounds invivo. Blocking receptor activity with a specific inhibitor of theEGFR tyrosine kinase (tyrphostin 1478) inhibited keratinocytecollagenase-1 production and markedly delayed re-epithelial-ization when compared with normal controls. Similarly, treat-

FIG. 8. Re-epithelialization of partial-thickness porcine burn wounds is inhibited by blocking EGFR activity. Partial thicknessthermal burns were created on the dorsal skin of pigs and treated topically with Silvadene® cream (A, vehicle control), Silvadene® cream containing20 mg/ml tyrphostin 1478 (B) or occlusive dressing only (C). All wounds were covered with occlusive dressing. The two large arrowheads in eachpanel mark the original wounded edges of the epidermal burn. The extent of epithelial healing for each burn was assessed by measuring thedistance between intact epithelial edges divided by the total length of the wound and expressed as percent re-epithelialization 6 S.E. D, epithelialhealing for the three treatment groups was averaged and compared for statistical significance by analysis of variance and Tukey’s HonestStatistical Difference post-test. E, primary human keratinocytes were cultured on type I collagen until confluent. Cells were treated with vehiclecontrol (collagen alone), PD153035 (100 or 500 nM), or tyrphostin 1478 (0.1–1.0 mM). Collagenase-1 protein in 72-h conditioned medium wasquantified by ELISA. Data shown are the means 6 S.D. from triplicate observations from the same cell preparation and are representative of atleast two separate skin donors.

Keratinocyte Collagenase-1 Expression Requires EGFR Activation10380

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

ment of ex vivo human skin punch biopsies with PD153035inhibited collagenase-1 production in keratinocytes at thewound edge. Taken together, these data suggest that blockingEGFR activity inhibits keratinocyte migration across the der-mal matrix and that this is due, at least in part, to inhibition ofcollagenase-1 expression. Although blocking EGFR signalingmay inhibit other biological events necessary for the completewound repair (e.g. cell proliferation), we suggest that EGFR-dependent collagenase-1 expression is critical for sustainedkeratinocyte migration and normal re-epithelialization.

Our findings in this report add substantially to understand-ing the mechanisms that regulate collagenase-1 expressionduring wound repair. Previously, we found that collagenase-1gene transcription is induced following injury when keratino-cytes move off their underlying basement membrane and con-tact type I collagen of the dermal matrix (3, 7, 5). We identifiedthe cell-surface recognition integrin as a2b1 and showed thatcollagenase-1 activity was requisite for keratinocyte migrationover a type I collagen substratum (10). Our present data indi-cate that intact skin is primed and ready to respond to injurywith high endogenous levels of keratinocyte EGFR and HB-EGF mRNA. Upon keratinocyte a2b1 binding to type I collagen,EGFR phosphorylation occurs within minutes. Collagenase-1mRNA levels are induced within 2 h, however, by a mechanismthat is EGFR-independent. Nevertheless, as collagenase-1 ex-pression continues, the sustained high levels of enzyme pro-duction from $8 h following contact with type I collagen aredependent on an EGFR/HB-EGF autocrine signaling loop.Since re-epithelialization of even minor wounds takes days,this sustained collagenase-1 expression is likely essential formost keratinocyte migration and for complete re-epithelializa-tion. Important unanswered questions raised by our findingsinclude: 1) what signaling pathway does a2b1 use to transmitthe rapid initial, EGFR-independent induction of collagenase-1?; and 2) upon completion of re-epithelialization, does thecessation of collagenase-1 production involve the dismantlingof the EGFR/HB-EGF autocrine loop? Studies to address theseimportant questions are currently in progress.

Acknowledgments—We thank Dr. Steven Frisch, La Jolla CancerResearch Foundation, La Jolla, CA, for the human collagenase-1 pro-moter construct and Dr. Robert Panek, Parke-Davis Warner Lambert,Ann Arbor, MI for the PD153035 EGFR antagonist. We also thankJennifer Gaither-Ganim, Molly R. Cofman, and Theresa J. Burke fortechnical assistance, and Jill Roby (Barnes-Jewish Hospital) for assist-ance with in situ hybridization studies.

REFERENCES

1. Mignatti, P., Rifkin, D. B., Welgus, H. G., and Parks, W. C. (1996) in TheMolecular and Cellular Biology of Wound Repair (Clark, R. A. F., ed) 2ndEd., pp. 427–474, Plenum Press, New York

2. Parks, W., and Mecham, R. (eds) (1998) Matrix Metalloproteinases, 1st Ed.,Academic Press, New York

3. Saarialho-Kere, U. K., Chang, E. S., Welgus, H. G., and Parks, W. C. (1992)J. Clin. Invest. 90, 1952–1957

4. Stricklin, G., and Nanney, L. (1994) J. Invest. Dermatol. 103, 488–4925. Saarialho-Kere, U. K., Vaalamo, M., Airola, K., Niemi, K.-M., Oikarinen, A. I.,

and Parks, W. C. (1995) J. Invest. Dermatol. 104, 982–9886. Inoue, M., Kratz, G., Haegerstrand, A., and Ståhle-Backdahl, M. (1995) J. In-

vest. Dermatol. 104, 479–4837. Saarialho-Kere, U. K., Kovacs, S. O., Pentland, A. P., Olerud, J., Welgus, H. G.,

and Parks, W. C. (1993) J. Clin. Invest. 92, 2858–28668. Inoue, M., Kratz, G., Haegerstrand, A., and Ståhle-Backdahl, M. (1995) J. In-

vest. Dermatol. 104, 479–4839. Sudbeck, B. D., Pilcher, B. K., Welgus, H. G., and Parks, W. C. (1997) J. Biol.

Chem. 272, 22103–2211010. Pilcher, B. K., Sudbeck, B. D., Dumin, J., Krane, S. M., Welgus, H. G., and

Parks, W. C. (1997) J. Cell Biol. 137, 1445–145711. Sudbeck, B. D., Parks, W. C., Welgus, H. G., and Pentland, A. P. (1994) J. Biol.

Chem. 269, 30022–3002912. van der Geer, P., Hunter, T., and Lindberg, R. (1994) Annu. Rev. Cell Biol. 10,

251–33713. Miettinen, P., Berger, J., Meneses, J., Phung, Y., Pederson, R., Werb, Z., and

Derynck, R. (1995) Nature 376, 337–34114. Theadgill, D., Dlugosz, A., Hansen, L., Tennenbaum, T., Lichti, U., Yee, D.,

LaManti, C., Mourton, T., Herrup, K., Harris, R., Barnard, J., Yuspa, S.,Coffee, R., and Magnuson, T. (1995) Science 269, 230–234

15. Sibilia, M., and Wagner, E. (1995) Science 269, 234–23816. Chen, J., Kim, J., Zhang, K., Sarret, Y., Wynn, K., Kramer, R., and Woodley,

D. (1993) Exp. Cell. Res. 209, 216–22317. Nanney, L., and King, J. L. E. (1996) in The Molecular and Cellular Biology of

Wound Repair (Clark, R., ed) 2nd Ed., pp. 171–194, Plenum Press, NewYork

18. Stoscheck, C., Nanney, L., and King, J. L.E. (1992) J. Invest. Dermatol. 99,645–649

19. Ju, W., Schiller, J., Kazempour, M., and Lowy, D. (1993) J. Invest. Dermatol.100, 628–632

20. Chen, Y. Q., Kahari, V.-M., Bashir, M. M., Rosenbloom, J., and Uitto, J. (1992)J. Invest. Dermatol. 98, 615 (Abstr.)

21. Cha, D., O’Brien, P., O’Toole, E., Woodley, D., and Hudson, L. (1996) J. Invest.Dermatol. 106, 590–597

22. Brown, G., Nanney, L., Griffin, J., Cramer, A., Yancey, J., Curtsinger, L.,Holtzin, L., Schultz, G., Jurkiewicz, M., and Lynch, J. (1987) N. Engl.J. Med. 321, 76–79

23. Schultz, G., White, M., Mitchell, R., Brown, G., Lynch, J., Twardzick, D., andTodaro, G. (1987) Science 235, 350–352

24. Jiang, C.-K., Magnaldo, T., Ohtsuki, M., Freedberg, I., Bernerd, F., and Blu-menberg, M. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 6786–6790

25. McCawley, L., O’Brien, P., and Hudson, L. (1997) Endocrinology 138, 121–12726. Wilcox, B., Dumin, J., and Jeffrey, J. (1994) J. Biol. Chem. 269, 29658–2966427. Sorvillo, J. M., McCormack, E. S., Yanez, L., Valenzuela, D., and Reynolds, J.,

F. H. (1990) Oncogene 5, 377–38628. Kawamoto, T., Sato, J., Le, A., Polikoff, J., Sato, G., and Mendelsohn, J. (1983)

Proc. Natl. Acad. Sci. U. S. A. 80, 1337–134129. Pentland, A. P., and Needleman, P. (1986) J. Clin. Invest. 77, 246–25130. Prosser, I. W., Stenmark, K. R., Suthar, M., Crouch, E. C., Mecham, R. P., and

Parks, W. C. (1989) Am. J. Pathol. 135, 1073–108831. Saarialho-Kere, U. K., Welgus, H. G., and Parks, W. C. (1993) J. Biol. Chem.

268, 17354–1736132. Cooper, T. W., Bauer, E. A., and Eisen, A. Z. (1982) Collagen Relat. Res. 3,

205–21133. Stricklin, G. P., Eisen, A. Z., Bauer, E. A., and Jeffrey, J. J. (1978) Biochemistry

17, 2331–233734. Welgus, H. G., Campbell, E. J., Bar-Shavit, Z., Senior, R. M., and Teitelbaum,

S. L. (1985) J. Clin. Invest. 76, 219–22435. Gullick, W. J., Downward, J., Parker, P. J., Whittle, N., Kris, R., Schlessinger,

J., Ullrich, A., and Waterfield, M. D. (1985) Proc. R. Soc. Lond. B 226,127–134

36. Pilcher, B., Gaither-Ganim, J., Parks, W., and Welgus, H. (1997) J. Biol. Chem.272, 18147–18154

37. Dunsmore, S. E., Rubin, J. S., Kovacs, S. O., Chedid, M., Parks, W. C., andWelgus, H. G. (1996) J. Biol. Chem. 271, 24576–24582

38. Imai, T., Kurachi, H., Adachi, K., Adachi, H., Yoshimoto, Y., Homma, H.,Tadokoro, C., Takeda, S., Yamaguchi, M., Sakata, M., Sakoyama, Y., andMiyake, A. (1995) Biol. Reprod. 52, 928–938

39. Forsyth, I. A., Taylor, J. A., Keable, S., Turvey, A., and Lennard, S. (1997) Mol.Cell. Endocrinol. 126, 41–48

40. Fen, Z., Dhadly, M. S., Yoshizumi, M., Hilkert, R. J., Quertermous, T., Eddy,R. L., Shows, T. B., and Lee, M. (1993) Biochemistry 32, 7932–7938

41. Chomczynski, P., and Sacchi, N. (1987) Anal. Biochem. 162, 156–15942. Swee, M. H., Parks, W. C., and Pierce, R. A. (1995) J. Biol. Chem. 270,

14899–1490643. Brown, G. L., Curtsinger, I., L., Brightwell, J. R., Ackerman, D. M., Tobin,

G. R., Polk, J., H. C., George-Nascimento, C., Valenzuela, P., and Schultz,G. S. (1986) J. Exp. Med. 163, 1319–1324

44. Levitzki, A., and Gazit, A. (1995) Science 267, 1782–178845. Fry, D., Kraker, A., McMichael, A., Ambroso, L., Nelson, J., Leopold, W.,

Connors, R., and Bridges, A. (1994) Science 265, 1093–109546. Fini, M., Strissel, K. J., Girard, M. T., Mays, J. W., and Rinehart, W. B. (1994)

J. Biol. Chem. 269, 11291–1129847. West-Mays, J., Strissel, K. J., Sadow, P. M., and Fini, M. E. (1995) Proc. Natl.

Acad. Sci. U. S. A. 92, 6768–677248. Doyle, G. A. R., Pierce, R. A., and Parks, W. C. (1997) J. Biol. Chem. 272,

11840–1184949. Saarialho-Kere, U. K., Chang, E. S., Welgus, H. G., and Parks, W. C. (1993)

J. Invest. Dermatol. 100, 335–34250. Woodley, D. T., Kalebec, T., Baines, A. J., Link, W., Prunieras, M., and Liotta,

L. (1986) J. Invest. Dermatol. 4, 418–42351. Stoll, S., Garner, W., and Elder, J. (1997) J. Clin. Invest. 100, 1271–128152. Levitski, A., and Gazit, A. (1995) Science 267, 1782–178853. Paladini, R., Tkakahashi, K., Bravo, N., and Coulombe, P. (1995) J. Cell Biol.

132, 381–39754. Stoll, S., Benedict, M., Mitra, R., Hiniker, A., Elder, J., and Nunez, G. (1998)

Oncogene 16, 1493–149955. Saarialho-Kere, U. K., Vaalamo, M., Airola, K., Niemi, K.-M., Oikarinen, A. I.,

and Parks, W. C. (1995) J. Invest. Dermatol. 105, 982–98856. Hashimoto, K., Higashimaya, S., Asada, H., Hashimura, E., Kobayashi, T.,

Sudo, K., Nakagawa, T., Damm, D., Yoshikawa, K., and Taniguchi, N.(1994) J. Biol. Chem. 269, 20060–20066

57. Marikovsky, M., Breuing, K., Liu, P. Y., Eriksson, E., Higashiyama, S., Farber,P., and Abraham, J. M. K. (1993) Proc. Natl. Acad. Sci. U. S. A. 90,3889–3893

58. Cybulsky, A., McTavish, A., and Cyr, M.-D. (1994) J. Clin. Invest. 94, 68–7859. Larjava, H., Salo, T., Haapasalmi, K., Kramer, R. H., and Heino, J. (1993)

J. Clin. Invest. 92, 1425–1435

Keratinocyte Collagenase-1 Expression Requires EGFR Activation 10381

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from

William C. Parks and Howard G. WelgusBrian K. Pilcher, JoAnn Dumin, Michael J. Schwartz, Bruce A. Mast, Gregory S. Schultz,

Receptor Autocrine MechanismKeratinocyte Collagenase-1 Expression Requires an Epidermal Growth Factor

doi: 10.1074/jbc.274.15.103721999, 274:10372-10381.J. Biol. Chem. 

  http://www.jbc.org/content/274/15/10372Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/274/15/10372.full.html#ref-list-1

This article cites 56 references, 23 of which can be accessed free at

by guest on October 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from