Disruption of Basal JNK Activity Differentially Affects Key Fibroblast Functions Important for Wound Healing

We used both a gene knockout approach and pharmacologic modulation to study the implication of the JNK pathway in regulating fibroblast motility, capacity to contract mechanically unloaded collagen gels, and type I collagen gene expression in vitro. These parameters, which are important for tissue repair, are positively regulated by transforming growth factor (TGF) -β, a cytokine viewed as playing a master role during wound healing. We demonstrate that basal JNK activity is critical for fibroblast motility because (a) mouse embryo jnk–/– fibroblasts exhibit significantly lower ability to close mechanically induced cell layer wounds than their wild-type (wt) counterparts, and (b) wound closure by human dermal fibroblasts is dramatically impaired by the specific JNK inhibitor SP600125. junAA fibroblasts, in which amino acids Ser63 and Ser73 of c-Jun are replaced by two Ala residues so that c-Jun cannot be phosphorylated by JNK, also exhibited impaired motility, suggesting that c-Jun phosphorylation by JNK is critical for fibroblast migration. In sharp contrast to their lesser motility on plastic, jnk–/– and junAA fibroblasts contracted free-floating, mechanically unloaded, collagen lattices markedly faster than wt fibroblasts. Furthermore, basal mRNA steady-state levels for types I and III collagen genes were similar in jnk–/– and wt fibroblasts. Likewise, overexpression of a dominant-negative mutant form of MKK4 in dermal fibroblasts did not affect collagen expression. We also demonstrate that basal JNK activity does not affect either TGF-β-induced collagen gene expression or lattice contraction, whereas on the other hand, the blockage of motility initiated by JNK inhibition cannot be overcome by TGF-β. Together these results demonstrate discrete, yet significant and highly specific, regulation of fibroblast functions important for wound healing by basal JNK activity. We used both a gene knockout approach and pharmacologic modulation to study the implication of the JNK pathway in regulating fibroblast motility, capacity to contract mechanically unloaded collagen gels, and type I collagen gene expression in vitro. These parameters, which are important for tissue repair, are positively regulated by transforming growth factor (TGF) -β, a cytokine viewed as playing a master role during wound healing. We demonstrate that basal JNK activity is critical for fibroblast motility because (a) mouse embryo jnk–/– fibroblasts exhibit significantly lower ability to close mechanically induced cell layer wounds than their wild-type (wt) counterparts, and (b) wound closure by human dermal fibroblasts is dramatically impaired by the specific JNK inhibitor SP600125. junAA fibroblasts, in which amino acids Ser63 and Ser73 of c-Jun are replaced by two Ala residues so that c-Jun cannot be phosphorylated by JNK, also exhibited impaired motility, suggesting that c-Jun phosphorylation by JNK is critical for fibroblast migration. In sharp contrast to their lesser motility on plastic, jnk–/– and junAA fibroblasts contracted free-floating, mechanically unloaded, collagen lattices markedly faster than wt fibroblasts. Furthermore, basal mRNA steady-state levels for types I and III collagen genes were similar in jnk–/– and wt fibroblasts. Likewise, overexpression of a dominant-negative mutant form of MKK4 in dermal fibroblasts did not affect collagen expression. We also demonstrate that basal JNK activity does not affect either TGF-β-induced collagen gene expression or lattice contraction, whereas on the other hand, the blockage of motility initiated by JNK inhibition cannot be overcome by TGF-β. Together these results demonstrate discrete, yet significant and highly specific, regulation of fibroblast functions important for wound healing by basal JNK activity. Efficient cutaneous wound repair implies carefully orchestrated molecular events allowing fibroblasts to migrate to the wound site (s), contract the wound, and synthesize extracellular matrix to restore skin integrity. Cellular signals underlying such events are not fully understood, but it is well accepted that transforming growth factor (TGF) 1The abbreviations used are: TGF, transforming growth factor; JNK, c-Jun NH2-terminal kinase; wt, wild-type; MAP, mitogen-activated protein; MKK, MAP kinase kinase; D/N, dominant-negative. 1The abbreviations used are: TGF, transforming growth factor; JNK, c-Jun NH2-terminal kinase; wt, wild-type; MAP, mitogen-activated protein; MKK, MAP kinase kinase; D/N, dominant-negative. -β, by means of its pleiotropic activities, plays a central role in orchestrating the various phases of wound healing (1Roberts A. B. Sporn M. B. Growth Factors. 1993; 8: 1-9Google Scholar, 2O'Kane S. Ferguson M. W. Int. J. Biochem. Cell Biol. 1997; 29: 63-78Google Scholar, 3Verrecchia F. Mauviel A. J. Invest. Dermatol. 2002; 118: 211-215Google Scholar). Extracellular stimuli elicit specific intracellular signals via activation of a family of so-called mitogen-activated protein (MAP) kinases, consisting of extracellular signal-regulated kinases, p38 MAP kinases, and the JNKs (4Whitmarsh A. J. Davis R. J. Science's STKE. 1999; (http: //stke. sciencemag. org/cgi/content/full/ocₛigtrans;1999/1/pe1) Google Scholar, 5Schramek H. News Physiol. Sci. 2002; 17: 62-67Google Scholar, 6Ichijo H. Oncogene. 1999; 18: 6087-6093Google Scholar). These MAP kinases phosphorylate transcription factors within the cell nucleus, thereby activating a number of cellular functions, including proliferation, apoptosis, differentiation, and regulation of gene expression. It has been reported that the extracellular signal-regulated kinases and p38 MAP kinases are activated in fibroblasts during collagen matrix contraction under isometric tension (7Lee D. J. Rosenfeldt H. Grinnell F. Exp. Cell Res. 2000; 257: 190-197Google Scholar) and that they cooperate in contraction-stimulated activation of the immediate early gene c-fos, although they are not required for lattice contraction per se (8Rosenfeldt H. Lee D. J. Grinnell F. Mol. Cell. Biol. 1998; 18: 2659-2667Google Scholar). In mechanically unloaded collagen lattices, on the other hand, extracellular signal-regulated kinase signaling is disrupted, a mechanism that may be responsible for the entry of fibroblast into a quiescent state after several days under these conditions (9Rosenfeldt H. Grinnell F. J. Biol. Chem. 2000; 275: 3088-3092Google Scholar). The JNK group of MAP kinases, also known as stress-activated kinases, are activated upon exposure of cells to cytokines, growth factors, and environmental stresses such as UV irradiation or heat shock (10Davis R. J. Cell. 2000; 103: 239-252Google Scholar). Three distinct genes, jnk1, jnk2, and jnk3, have been identified to encode JNKs. The former two genes are ubiquitously expressed, whereas jnk3 is selectively expressed in the heart, testis, and brain. Dual Thr and Tyr phosphorylation of JNK by the MAP kinase kinases (MKKs), MKK4 and MKK7, results in JNK activation and nuclear translocation. In the nucleus, JNKs phosphorylate transcription factors such as c-Jun (11Ip Y. T. Davis R. J. Curr. Opin. Cell Biol. 1998; 10: 205-219Google Scholar), a process that leads to maximal transcriptional activity of the latter (12Yang D. Tournier C. Wysk M. Lu H. T. Xu J. Davis R. J. Flavell R. A. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 3004-3009Google Scholar). Thus far, little is known about the role of JNK in the context of fibroblast ability to remodel collagen matrices, except that it was shown recently that JNK regulates the phenotypic modulation of lung fibroblasts into myofibroblasts induced by TGF-β (13Hashimoto S. Gon Y. Takeshita I. Matsumoto K. Maruoka S. Horie T. Am. J. Respir. Crit. Care Med. 2001; 163: 152-157Google Scholar), interleukin 4, and interleukin 13 (14Hashimoto S. Gon Y. Takeshita I. Maruoka S. Horie T. J. Allergy Clin. Immunol. 2001; 107: 1001-1008Google Scholar). Recent data from our laboratory have indicated that basal JNK activity in fibroblasts maintains a limited yet significant pool of phosphorylated c-Jun protein (15Verrecchia F. Tacheau C. Wagner E. F. Mauviel A. J. Biol. Chem. 2003; 278: 1585-1593Google Scholar). In this report, using pharmacologic and genetic approaches aimed at interfering with basal JNK activity, we demonstrate a critical role for the latter in allowing fibroblast motility, whereas it inhibits the ability of fibroblasts to contract free-floating collagen matrices and does not modify the expression of fibrillar collagen genes or their modulation by TGF-β. Details are provided herein. Cell Cultures—Human dermal fibroblasts were established by explanting neonatal foreskins. Immortalized fibroblast cell lines derived from wild-type (wt) and jnk1–/–-jnk2–/– (referred to henceforth as jnk–/–) mouse embryos (16Sabapathy K. Jochum W. Hochedlinger K. Chang L. Karin M. Wagner E. F. Mech. Dev. 1999; 89: 115-124Google Scholar), in which targeted disruption of jnk1 and jnk2 has been performed simultaneously, and junAA immortalized fibroblasts derived from mouse embryos carrying a mutant c-jun allele in which the JNK phospho-acceptor serines 63 and 73 are mutated to alanines (17Behrens A. Sibilia M. Wagner E. F. Nat. Genet. 1999; 21: 326-329Google Scholar) were a kind gift from Dr. Erwin F. Wagner (Institute for Molecular Pathology, Vienna, Austria). Cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal calf serum, 2 mm glutamine, and antibiotics (100 units/ml penicillin, 50 μg/ml streptomycin G, and 0. 25 μg/ml Fungizone™). Human recombinant TGF-β1, purchased from R 108: 73-81Google Scholar, 19Bennett B. L. Sasaki D. T. Murray B. W. O'Leary E. C. Sakata S. T. Xu W. Leisten J. C. Motiwala A. Pierce S. Satoh Y. Bhagwat S. S. Manning A. M. Anderson D. W. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 13681-13686Google Scholar), was purchased from Calbiochem-Novabiochem. Collagen Matrix Contraction—Fibroblasts were harvested from monolayer culture with 0. 25% trypsin and 1 mm EDTA, and then trypsin was neutralized with 10% fetal calf serum-containing medium. Collagen lattices in 60-mm dishes were prepared with 7 ml of a mixture containing 106 fibroblasts and 1 mg/ml native type I collagen (Biocoat™; BD Biosciences) in medium supplemented with 10% fetal calf serum. When needed, TGF-β was added to the mixture before polymerization of the collagen matrix. Polymerization of collagen matrices required ∼60 min at 37 °C. To initiate lattice contraction, freshly polymerized matrices were released from the underlying culture dish with a few gentle taps on the dish. Wound Closure and Transwell™ Motility Assays—For wound closure assays, confluent cell monolayers were wounded by manually scraping the cells with a pipette tip. After wounding, wound size was verified to ensure that all wounds were the same width (see corresponding figures). The cell culture medium was then replaced with fresh medium, and wound closure was monitored by microscopy at various times. Transwell™ migration assays were performed utilizing 8-μm pore, 6. 5-mm polycarbonate Transwell™ filters (Falcon, Franklin Lakes, NJ). In some experiments, a type I collagen solution (10 μg in 100 μl) was allowed to polymerize in the upper well for 1 h at 37 °C. Single cell suspensions were seeded onto the upper surface of the filters in medium containing 10% fetal calf serum (without or with prior collagen coating) and allowed to migrate through the membrane. After a 16-h incubation period, cells on the upper surface of the filter were wiped off with a cotton swab, and the cells that had migrated to the underside of the filter were fixed, stained with DiffQuick™ (Dade Behring, Düdingen, Switzerland), and counted by bright-field microscopy at ×200 in six random fields. Northern Blotting—Total RNA was obtained using an RNeasy kit (Qiagen GmbH, Hilden Germany) and analyzed by Northern hybridization (20 μg/lane) with 32P-labeled cDNA probes for COL1A1 (20Chu M. L. de Wet W. Bernard M. Ramirez F. J. Biol. Chem. 1985; 260: 2315-2320Google Scholar), COL1A2 (21Chu M. L. Myers J. C. Bernard M. P. Ding J. F. Ramirez F. Nucleic Acids Res. 1982; 10: 5925-5934Google Scholar), COL3A1 (22Chu M. L. Weil D. de Wet W. Bernard M. Sippola M. Ramirez F. J. Biol. Chem. 1985; 260: 4357-4363Google Scholar), and GAPDH (23Fort P. Marty L. Piechaczyk M. el Sabrouty S. Dani C. Jeanteur P. Blanchard J. M. Nucleic Acids Res. 1985; 13: 1431-1442Google Scholar). Hybridization signal was revealed with a PhosphorImager (Storm 840; Amersham Biosciences). Western Blotting—Whole cell extracts were prepared in 10 mm Tris, pH 7. 4, 1% SDS, and 1 mm sodium vanadate; treated with Benzon nuclease (Sigma) for 5 min at room temperature; and denatured by heating at 95 °C for 3 min. Protein concentration in each lysate was assayed with a one-step colorimetric method (Bio-Rad protein reagent; Bio-Rad), and 25 μg of protein was resolved by SDS-PAGE. After electrophoresis, proteins were transferred to Hybond ECL nitrocellulose filters (Amersham Biosciences). Filters were placed in blocking solution (1× Tris-buffered saline and 5% nonfat milk) for 1 h and immunoblotted with either goat anti-type I collagen (Southern Biotech, Birmingham, AL), rabbit anti-phospho-c-Jun (Upstate Biotechnologies, Lake Placid, NY), or at a in Tris-buffered and 5% nonfat for 1 mouse (Sigma) at a in 5% nonfat was used as a of phosphorylated nonfat was replaced by serum After filters were and with for 1 Filters were then to Amersham and revealed with a PhosphorImager (Storm 840; Amersham Biosciences). Cell and basal JNK activity, we used a derived from the consisting of a and a a protein consisting of the of and the of c-Jun that phosphorylation by JNK to fully were performed using the with a kit was in to activity was with a kit of a dominant-negative MKK4 expression kind gift from Dr. A. A. S. Davis C. J. Biol. Chem. 1997; Scholar), human dermal fibroblasts were with a GmbH, Germany) to the was to be not by of a protein expression We the JNK under the conditions used in the To this we the levels of and in jnk–/– and wt mouse fibroblasts, the latter of which were treated or not treated with SP600125. in the of expression and the of phosphorylation of c-Jun in jnk–/– also demonstrate that at a concentration of the basal phosphorylation levels of JNK and c-Jun in wt fibroblasts. These results were in the under and the transcriptional of under the various conditions We the Western data because (a) activity is dramatically lower in jnk–/– fibroblasts as with their wt counterparts, of an of JNK activity in the former cell and (b) inhibits activity in wt fibroblasts, with its role as a JNK inhibitor (18Han Z. Boyle D. L. Chang L. Bennett B. Karin M. Yang L. Manning A. M. Firestein G. S. J. Clin. Invest. 2001; 108: 73-81Google Scholar, 19Bennett B. L. Sasaki D. T. Murray B. W. O'Leary E. C. Sakata S. T. Xu W. Leisten J. C. Motiwala A. Pierce S. Satoh Y. Bhagwat S. S. Manning A. M. Anderson D. W. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 13681-13686Google Scholar) on basal kinase activity. These results are in with our that overexpression of a dominant-negative form of the kinase of JNK, transcription in human dermal fibroblasts (15Verrecchia F. Tacheau C. Wagner E. F. Mauviel A. J. Biol. Chem. 2003; 278: 1585-1593Google Scholar). the JNK was in our experiments, we the role of its basal activity on fibroblast we the motility of wt, and junAA immortalized mouse embryo fibroblasts in a wound closure shown in wt fibroblasts migrated into the wound and the wound within h the other hand, in jnk–/– fibroblast the wounds in cell did not for the wt and jnk–/– fibroblasts because wt fibroblasts the wound in the or of a that cell not c-Jun is a for JNK, junAA fibroblasts, in which c-Jun Ser63 and Ser73 have been mutated into alanines so that c-Jun be phosphorylated by JNK, were used to c-Jun phosphorylation by JNK was important for the of basal JNK activity on fibroblast shown in 2 junAA fibroblasts were identified as in the same wound closure from jnk–/– fibroblasts These data that the basal state of c-Jun phosphorylation by JNK is critical for cell migration. because TGF-β is a known of fibroblast motility J. J. Exp. Med. Scholar), we to TGF-β be to fibroblast migration in the of basal JNK activity. When either jnk–/– or junAA fibroblasts were treated with TGF-β significant in migration within the wound be as with that TGF-β cannot overcome the in motility induced by the of basal JNK activity or by the of basal levels of The in cell motility wt and jnk–/– immortalized fibroblasts the of such a may be specific for either the knockout or the immortalized to pharmacologic inhibition of JNK activity in either wt immortalized fibroblasts or human dermal fibroblasts their motility in the same wound closure this we used SP600125, a specific JNK inhibitor (18Han Z. Boyle D. L. Chang L. Bennett B. Karin M. Yang L. Manning A. M. Firestein G. S. J. Clin. Invest. 2001; 108: 73-81Google Scholar, 19Bennett B. L. Sasaki D. T. Murray B. W. O'Leary E. C. Sakata S. T. Xu W. Leisten J. C. Motiwala A. Pierce S. Satoh Y. Bhagwat S. S. Manning A. M. Anderson D. W. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 13681-13686Google Scholar), at a concentration of which is to basal JNK activity, as by either Western for or in a modified specific for (see these wt fibroblasts treated with were to close a mechanically wound in a confluent cell layer whereas cells treated with the did with the results obtained using jnk–/– fibroblasts, TGF-β was to the of in wt fibroblasts. results were obtained with human skin fibroblasts dermal fibroblasts migrated into the wound, and the latter was within h the other hand, were in their ability to close the wound, treated with TGF-β these basal JNK activity as a critical of the cell allowing fibroblast migration. these results our of a role for basal JNK activity in a pool of phosphorylated c-Jun protein in fibroblasts, to gene under conditions (15Verrecchia F. Tacheau C. Wagner E. F. Mauviel A. J. Biol. Chem. 2003; 278: 1585-1593Google Scholar). We to the JNK pathway is during collagen contraction by fibroblasts, a that both extracellular matrix and cell motility in F. J. Cell Biol. Scholar). To this we the of wt, and junAA fibroblasts placed in free-floating, mechanically unloaded, collagen The of collagen contraction was then a shown in 4, jnk–/– and junAA fibroblasts were significantly that their wt in collagen lattice contraction by jnk–/– and junAA fibroblasts was 2 days after to a of the lattice whereas contraction by wt fibroblasts a not significant in cell within the collagen were with either cell type the of the experiments, as by cells after of the lattices not When added after the cell in the collagen TGF-β significantly contraction by wt, and junAA fibroblasts. is shown in such lattice contraction the same with either wt fibroblasts or jnk–/– and junAA fibroblasts, although the was in the of wt fibroblasts these results demonstrate that basal JNK activity the capacity of fibroblasts to contract collagen but does not the of this by TGF-β. It is that we recently identified JNK as a for c-Jun to of gene transcription in junAA fibroblasts (15Verrecchia F. Tacheau C. Wagner E. F. Mauviel A. J. Biol. Chem. 2003; 278: 1585-1593Google Scholar). we that TGF-β signaling in junAA fibroblasts, whereas it does not so in jnk–/– fibroblasts. dominant-negative mutant form of MKK4 that JNK activation the of TGF-β in junAA that this mechanism is on JNK and a other than means of we that for to the activity of signaling in a with the latter in the which was in the of cytokine basal JNK activity may be junAA and jnk–/– fibroblasts both exhibit impaired motility and activity as with their wt that other JNK is to the mutant in the context of either cell motility or capacity of fibroblasts to contract collagen be that basal expression not induced by in fibroblasts is and is not to transcriptional whereas c-Jun is expressed at levels to basal JNK activity. may from the that is a transcriptional than c-Jun P. Karin M. Cell. Scholar, J. J. M. S. J. J. Cell. Scholar, A. Y. W. J. Curr. Biol. 1993; Scholar). In the context of the latter at in some of the JNK signals c-Jun expression or activation by JNK is (15Verrecchia F. Tacheau C. Wagner E. F. Mauviel A. J. Biol. Chem. 2003; 278: 1585-1593Google Scholar). the protein has been in both cell motility and matrix contraction F. Cell Biol. 2000; 10: Scholar). Furthermore, a in both fibroblast motility and fibroblast capacity to contract free-floating has been to the process Mech. Dev. 2001; Scholar). fibroblasts derived from tissue lower motility than fibroblasts from the latter of which also exhibit a capacity to contract free-floating collagen It that the collagen contraction and cell motility is and has been established to these two the results such we to of the of cell to collagen may modify fibroblast this we motility on the Transwell™ which is to the cell through in the or of a collagen (see and shown in jnk–/– fibroblasts exhibited a lower capacity to migrate through the Transwell™ than their wt counterparts, and this capacity was not either positively or by the collagen These in with obtained using the in wound closure in 2 and to that basal JNK activity is an important allowing fibroblast We basal JNK activity may basal fibrillar collagen gene expression and its by TGF-β. a to this confluent wt and jnk–/– fibroblast were for h in the or of after which expression of type I and type III collagen genes was by Northern of shown in similar basal steady-state mRNA levels for and COL3A1 were in wt and jnk–/– fibroblasts. incubation with TGF-β in of the expression of each of these genes, which was in its in both cell I collagen by Western basal levels in both wt and jnk–/– fibroblasts, as well as similar in to TGF-β. human dermal fibroblasts were with either or MKK4 expression by means of a (see and after which collagen gene expression was by Northern in demonstrate that MKK4 expression does not modify either basal 3 TGF-β-induced collagen mRNA steady-state Together with the results obtained using jnk–/– fibroblasts, these data demonstrate that basal JNK activity in fibroblasts is not a for either basal fibrillar collagen gene expression or the of its activation by with the critical role by JNK activation in the ability of to the of fibrillar collagen genes by TGF-β (15Verrecchia F. Tacheau C. Wagner E. F. Mauviel A. J. Biol. Chem. 2003; 278: 1585-1593Google Scholar). migration has been a responsible for collagen contraction by fibroblasts under mechanically unloaded The results in this that a the two is because cellular signaling by basal JNK activity these two cellular we for a critical role for basal JNK activity in allowing fibroblasts to at the same it their capacity to contract collagen and does not affect fibrillar collagen gene expression. data the basal activity of the JNK pathway as highly in several fibroblast functions for tissue repair, cell matrix contraction, and collagen In to wound healing We A. F. and Erwin F. Wagner for and