PAI-1 expression is required for epithelial cell migration in two distinct phases of in vitro wound repair

PAI-1 transcriptional regulation during the G0 --> G1 transition in human epidermal keratinocytes

Plasminogen activator inhibitor type-1 (PAI-1) is the major negative regulator of the plasmin-dependent pericellular proteolytic cascade. PAI-1 gene expression is normally growth state regulated but frequently elevated in chronic fibroproliferative and neoplastic diseases affecting both stromal restructuring and cellular migratory activities. Kinetic modeling of cell cycle transit in synchronized human keratinocytes (HaCaT cells) indicated that PAI-1 transcription occurred early after serum stimulation of quiescent (G0) cells and prior to entry into a cycling G1 condition. PAI-1 repression (in G0) was associated with upstream stimulatory factor-1 (USF-1) occupancy of two consensus E box motifs (5'-CACGTG-3') at the PE1 and PE2 domains in the PF1 region (nucleotides -794 to -532) of the PAI-1 promoter. Chromatin immunoprecipitation (ChIP) analysis established that the PE1 and PE2 site E boxes were occupied by USF-1 in quiescent cells and by USF-2 in serum-activated, PAI-1-expressing keratinocytes. This reciprocal and growth state-dependent residence of USF family members (USF-1 vs. USF-2) at PE1/PE2 region chromatin characterized the G0 --> G1 transition period and the transcriptional status of the PAI-1 gene. A consensus E box motif was required for USF/E box interactions, as a CG --> AT substitution at the two central nucleotides inhibited formation of USF/probe complexes. The 5' flanking sites (AAT or AGAC) in the PE2 segment were not necessary for USF binding. USF recognition of the PE1/PE2 region E box sites required phosphorylation with several potential involved residues, including T153, maping to the USF-specific region (USR). A T153A substitution in USF-1 did not repress serum-induced PAI-1 expression whereas the T153D mutant was an effective suppressor. As anticipated from the ChIP results, transfection of wild-type USF-2 failed to inhibit PAI-1 induction. Collectively, these data suggest that USF family members are important regulators of PAI-1 gene control during serum-stimulated recruitment of quiescent human epithelial cells into the growth cycle.

Games network and application to PAs system

In this article, we present a game theory based framework, named games network, for modeling biological interactions. After introducing the theory, we more precisely describe the methodology to model biological interactions. Then we apply it to the plasminogen activator system (PAs) which is a signal transduction pathway involved in cancer cell migration. The games network theory extends game theory by including the locality of interactions. Each game in a games network represents local interactions between biological agents. The PAs system is implicated in cytoskeleton modifications via regulation of actin and microtubules, which in turn favors cell migration. The games network model has enabled us a better understanding of the regulation involved in the PAs system.

β-Adrenergic Receptor Antagonists Accelerate Skin Wound Healing

The skin is our primary defense against noxious environmental agents. Upon injury, keratinocytes migrate directionally into the wound bed to initiate re-epithelialization, essential for wound repair and restoration of barrier integrity. Keratinocytes express a high level of beta2-adrenergic receptors (beta2-ARs) that appear to play a role in cutaneous homeostasis as aberrations in either keratinocyte beta2-AR function or density are associated with various skin diseases. Here we report the novel finding that beta-AR antagonists promote wound re-epithelialization in a "chronic" human skin wound-healing model. beta-AR antagonists increase ERK phosphorylation, the rate of keratinocyte migration, electric field-directed migration, and ultimately accelerate human skin wound re-epithelialization. We demonstrate that keratinocytes express two key enzymes required for catecholamine (beta-AR agonist) synthesis, tyrosine hydroxylase and phenylethanolamine-N-methyl transferase, both localized within keratinocyte cytoplasmic vesicles. Finally, we confirm the synthesis of epinephrine by measuring the endogenously synthesized catecholamine in keratinocyte extracts. Previously, we have demonstrated that beta-AR agonists delay wound re-epithelialization. Here we report that the mechanism for the beta-AR antagonist-mediated augmentation of wound repair is due to beta2-AR blockade, preventing the binding of endogenously synthesized epinephrine. Our work describes an endogenous beta-AR mediator network in the skin that can temporally regulate skin wound repair. Further investigation of this network will improve our understanding of both the skin repair process and the multiple modes of action of one of the most frequently prescribed class of drugs, hopefully resulting in a new treatment for chronic wounds.

A role for the mitogen-activated protein kinase kinase kinase 1 in epithelial wound healing

The mitogen-activated protein kinase kinase (MEK) kinase 1 (MEKK1) mediates activin B signals required for eyelid epithelium morphogenesis during mouse fetal development. The present study investigates the role of MEKK1 in epithelial wound healing, another activin-regulated biological process. In a skin wound model, injury markedly stimulates MEKK1 expression and activity, which are in turn required for the expression of genes involved in extracellular matrix (ECM) homeostasis. MEKK1 ablation or down-regulation by interfering RNA significantly delays skin wound closure and impairs activation of Jun NH2-terminal kinases, induction of plasminogen activator inhibitor (PAI)-1, and restoration of cell-cell junctions of the wounded epidermis. Conversely, expression of wild-type MEKK1 accelerates reepithelialization of full-thickness skin and corneal debridement wounds by mechanisms involving epithelial cell migration, a cell function that is partially abolished by neutralizing antibodies for PAI-1 and metalloproteinase III. Our data suggest that MEKK1 transmits wound signals, leading to the transcriptional activation of genes involved in ECM homeostasis, epithelial cell migration, and wound reepithelialization.

TGF-beta 1-induced PAI-1 expression is E box/USF-dependent and requires EGFR signaling

Transforming growth factor-beta1 (TGF-beta1) transcriptionally regulates the expression of genes that encode specific proteins (e.g., plasminogen activator inhibitor-1; PAI-1) important in stromal remodeling and cellular invasion. Definition of molecular events underlying TGF-beta1-initiated PAI-1 transcription, therefore, may lead to the identification of new therapeutic targets for diseases associated with elevated PAI-1 synthesis (e.g., tissue fibrosis, vascular disorders, tumor progression). An intact upstream stimulatory factor (USF)-binding E box motif (5'-(-165)CACGTG(-160)-3') at the HRE-2 site in the rat PAI-1 gene was required for PAI-1 transcription in TGF-beta1-treated cells. Mutation of the CA dinucleotide to TC at position -165/-164 in a reporter construct driven by 764 bp of PAI-1 promoter sequence decreased TGF-beta1-dependent CAT activity by >80% indicating the necessity for a consensus hexanucleotide E box motif in induced expression. The same CA --> TC substitution eliminated USF binding to an 18-bp HRE-2 DNA target highlighting the importance of site occupancy to transcriptional activation. Transfection of a dominant-negative USF construct, moreover, completely inhibited formation of USF/HRE-2 probe complexes, attenuated PAI-1 promoter-driven luciferase activity and reduced the response of the endogenous PAI-1 gene to TGF-beta1 (to that approximating quiescent controls). Maximal immediate-early PAI-1 induction upon exposure to TGF-beta1 required EGFR, p21ras, MEK and pp60(c-src) signaling as pharmacologic or dominant-negative inhibition of any of the four intermediates (EGFR, p21ras, MEK, pp60(c-src)) virtually eliminated TGF-beta1-augmented PAI-1 levels. U0126 titering experiments, furthermore, revealed that the same MEK inhibitor concentration that blocked the TGF-beta1 increase in ERK1/2 phosphorylation (20 microM) also effectively attenuated the PAI-1 inductive response suggesting a requirement for stimulated ERK signaling in TGF-beta1-mediated PAI-1 expression. These data suggest a model whereby TGF-beta1 activates a complex signaling cascade to affect PAI-1 gene control and involves USF occupancy of a critical E box motif at the HRE-2 site in the PAI-1 gene.

The ternary complex factor Net regulates cell migration through inhibition of PAI-1 expression

Net, Elk-1, and Sap-1 are members of the ternary complex factor (TCF) subfamily of Ets transcription factors. They form ternary complexes with serum response factor (SRF) on serum response elements of immediate early genes such as c-fos and egr-1 and mediate responses to growth factors and mitogen-activated protein kinase signaling. Although the TCFs have been extensively studied as intermediates in signaling cascades, surprisingly little is known about their different target genes and physiological functions. We report that Net homozygous mutant mouse embryonic fibroblasts have a defect in cell migration. This defect results at least in part from increased expression of plasminogen activator inhibitor type 1 (PAI-1), a serine protease inhibitor (serpin) that controls extracellular proteolysis and cell matrix adhesion. The defect in cell migration can be reverted by the addition of a PAI-1 blocking antibody. Net represses PAI-1 promoter activity and binds to a specific region of the promoter containing Ets binding sites in the absence of SRF. We conclude that Net is a negative regulator of PAI-1 expression and is thereby involved in cell migration.

β2‐Adrenergic receptor activation delays wound healing

Keratinocytes migrate directionally into the wound bed to initiate re-epithelialization, necessary for wound closure and restoration of barrier function. They solely express the beta2-adrenergic receptor (beta2-AR) subtype of beta-ARs and can also synthesize beta-AR agonists generating a hormonal mediator network in the skin. Emerging studies from our laboratory demonstrate that beta-AR agonists decrease keratinocyte migration via a protein phosphatase (PP) 2A-dependent mechanism. Here we have extended our investigations to observe the effects of beta2-AR activation on keratinocyte polarization, migration, and ERK phosphorylation at the wound edge, cytoskeletal organization, phospho-ERK intracellular localization, proliferation, human skin wound re-epithelialization, wound-induced ERK phosphorylation, and murine skin wound healing. We demonstrate that in keratinocytes, beta2-AR activation is anti-motogenic and anti-mitogenic with both mechanisms being PP2A dependent. beta2-AR activation dramatically alters the organization of the actin cytoskeleton and prevents localization of phospho-ERK to the lamellipodial edge and its colocalization with vinculin. Finally, we demonstrate a beta2-AR-mediated delay in re-epithelialization and decrease in wound-induced epidermal ERK phosphorylation in human skin wounds and a delay in re-epithelialization in murine tail-clip wounds. Our work uncovers novel keratinocyte biology and a previously unrecognized role for the adrenergic hormonal mediator network in the wound repair process.

Periodic direct current does not promote wound closure in an in vitro dynamic model of cell migration

BACKGROUND AND PURPOSE

A prevailing paradigm is that electrical fields can promote cell migration and tissue healing. To further validate this paradigm, we tested the hypothesis that periodic direct current (DC) can enhance wound closure using an in vitro dynamic model of cell migration.

METHODS AND RESULTS

Layers of primary fibroblasts were wounded and treated with DC under various voltages. Repair area, cell velocity, and directionality as well as lamellipodium area were evaluated at different times. Direct current had no beneficial effect on cell migration. Moreover, prolonged stimulation under the highest voltage led to significant reduction in wound closure and cell velocity. The reduction of membrane protusions in stimulated cells may be associated with the deleterious effect of DC.

DISCUSSION AND CONCLUSION

Contrary to the authors' expectations, they found that periodic DC did not promote wound closure, a finding that emphasizes the need to clarify the complex effects of electrical fields on migrating cells.

Smad4 dependency defines two classes of transforming growth factor {beta} (TGF-{beta}) target genes and distinguishes TGF-{beta}-induced epithelial-mesenchymal transition from its antiproliferative and migratory responses

In response to transforming growth factor beta (TGF-beta), Smad4 forms complexes with activated Smad2 and Smad3, which accumulate in the nucleus, where they both positively and negatively regulate TGF-beta target genes. Mutation or deletion of Smad4 is found in about 50% of pancreatic tumors and in about 15% of colorectal tumors. As Smad4 is a central component of the TGF-beta/Smad pathway, we have determined whether Smad4 is absolutely required for all TGF-beta responses, to evaluate the effect of its loss during human tumor development. We have generated cell lines from the immortalized human keratinocyte cell line HaCaT or the pancreatic tumor cell line Colo-357, which stably express a tetracyline-inducible small interfering RNA targeted against Smad4. In response to tetracycline, Smad4 expression is effectively silenced. Large-scale microarray analysis identifies two populations of TGF-beta target genes that are distinguished by their dependency on Smad4. Some genes absolutely require Smad4 for their regulation, while others do not. Functional analysis also indicates a differential Smad4 requirement for TGF-beta-induced functions; TGF-beta-induced cell cycle arrest and migration, but not epithelial-mesenchymal transition, are abolished after silencing of Smad4. Altogether our results suggest that loss of Smad4 might promote TGF-beta-mediated tumorigenesis by abolishing tumor-suppressive functions of TGF-beta while maintaining some tumor-promoting TGF-beta responses.

Smad4 dependency defines two classes of transforming growth factor {beta} (TGF-{beta}) target genes and distinguishes TGF-{beta}-induced epithelial-mesenchymal transition from its antiproliferative and migratory responses

In response to transforming growth factor beta (TGF-beta), Smad4 forms complexes with activated Smad2 and Smad3, which accumulate in the nucleus, where they both positively and negatively regulate TGF-beta target genes. Mutation or deletion of Smad4 is found in about 50% of pancreatic tumors and in about 15% of colorectal tumors. As Smad4 is a central component of the TGF-beta/Smad pathway, we have determined whether Smad4 is absolutely required for all TGF-beta responses, to evaluate the effect of its loss during human tumor development. We have generated cell lines from the immortalized human keratinocyte cell line HaCaT or the pancreatic tumor cell line Colo-357, which stably express a tetracyline-inducible small interfering RNA targeted against Smad4. In response to tetracycline, Smad4 expression is effectively silenced. Large-scale microarray analysis identifies two populations of TGF-beta target genes that are distinguished by their dependency on Smad4. Some genes absolutely require Smad4 for their regulation, while others do not. Functional analysis also indicates a differential Smad4 requirement for TGF-beta-induced functions; TGF-beta-induced cell cycle arrest and migration, but not epithelial-mesenchymal transition, are abolished after silencing of Smad4. Altogether our results suggest that loss of Smad4 might promote TGF-beta-mediated tumorigenesis by abolishing tumor-suppressive functions of TGF-beta while maintaining some tumor-promoting TGF-beta responses.

Nucleofection: a new, highly efficient transfection method for primary human keratinocytes*

Transfection is an essential tool for numerous in vitro applications including studies of gene expression, promoter analysis, and intracellular signaling pathways and also for therapeutic strategies such as tissue engineering and gene therapy. However, transfection of primary cells including keratinocytes with common methods such as calcium phosphate, DEAE-dextran, liposome-mediated transfer, electroporation or viral vectors is problematic because of low transfection efficiency and the induction of terminal differentiation. Here we analyzed the use of nucleofection, a new, electroporation-based transfection method that enables the DNA to enter directly the nucleus, for the transfection of keratinocytes. Several different conditions were tested and optimized, resulting in a final transfection efficiency of 56% in primary human epidermal keratinocytes. This efficiency is superior to all non-viral transfection methods reported so far. The number of non-viable keratinocytes after nucleofection was low, varying between 14 and 16%. In contrast to other transfection protocols, nucleofection did not induce terminal differentiation in the transfected keratinocytes. In addition, nucleofection is a fast method, because the results can be analyzed within 7 h. In summary, nucleofection is a fast, easy and highly effective alternative for the transfection of primary human keratinocytes, which offers new opportunities for various research applications.

Upstream stimulatory factor regulates E box-dependent PAI-1 transcription in human epidermal keratinocytes

Certain growth factors (e.g., TGF-beta1) initiate a "plastic" response in human keratinocytes (HaCaT cells) characterized by changes in gene expression and increased cell motility. While microarray analyses identified a number of involved genes, plasminogen activator inhibitor type 1 (PAI-1) is among the subset most highly responsive to TGF-beta1. Previous antisense attenuation of PAI-1 synthesis confirmed an essential role for this protease inhibitor in cell motility (Providence et al., 2002, J Cell Sci 115:3767-3777; Providence and Higgins, 2004, J Cell Physiol 200:297-308). It was important, therefore, to clarify molecular mechanisms underlying PAI-1 expression control in human keratinocytes. A consensus E box motif (5'-CACGTG-3') at nucleotides -566 to -561 in the PE2 region of the PAI-1 gene was required for TGF-beta1-induced transcription of a PAI-1 promoter-driven luceriferase reporter. Truncation of the PE2 E box or mutation of the CACGTG hexanucleotide to CAATTG inhibited growth factor-stimulated promoter function confirming the importance of this site in inducible expression. A similar mutation at the PE1 E box (nucleotides -682 to -677), in contrast, did not result in reduced luciferase activity. Competing CACGTG-containing DNAs, regardless of the presence or absence of PAI-1-specific flanking sequences or lacking accessory sequences (i.e., Smad-binding sites, AAT trinucleotide spacer), inhibited complex formation between HaCaT cell nuclear factors and a 45-mer PE2 region probe. A deoxyoligonucleotide that differed from the consensus E box by a CG --> AT substitution (the same base change incorporated into the PAI-1p806-lucerifase reporter by site-directed mutagenesis) but with random (i.e., non-PAI-1) flanking sequences also failed to compete with the PE2 region probe for protein binding whereas the same construct with an intact CACGTG motif was an effective competitor. The major protein/DNA interactions in the PE2 segment, therefore, are E box-dependent. USF-1, a member of the upstream stimulatory factor family, bound the PE2 construct suggesting a role for USF proteins in E box residence and PAI-1 gene expression. Chromatin immunoprecipitation, using primers designed to amplify a 300-bp PE2-associated promoter fragment and containing no other E box motifs except the target CACGTG at nucleotides -566 to -561, confirmed that this site was occupied by USF-1 or a USF-1-containing complex in both quiescent and TGF-beta1-stimulated cells. Transfection of a dominant-negative USF construct effectively attenuated serum- and TGF-beta1-induced PAI-1 synthesis as well as TGF-beta1-stimulated Matrigel barrier invasion. Dominant-negative USF-expressing keratinocytes, moreover, specifically had a reduced capacity for Matrigel barrier invasion. USF elements, therefore, are important regulators of growth factor-initiated PAI-1 transcription (as predicted from the identification of PAI-1 as a direct USF target gene) and the associated epithelial migratory response.

Plasminogen activator inhibitor type-1 gene expression and induced migration in TGF-β1-stimulated smooth muscle cells is pp60c-src/MEK-dependent

Transforming growth factor-beta1 (TGF-beta1) stimulates expression of plasminogen activator inhibitor type-1 (PAI-1), a serine protease inhibitor (SERPIN) important in the control of stromal barrier proteolysis and cell-to-matrix adhesion. Pharmacologic agents that target MEK (PD98059, U0126) or src family (PP1) kinases attenuated TGF-beta1-dependent PAI-1 transcription in R22 aortic smooth muscle cells. Pretreatment with PP1 at concentrations that inhibited TGF-beta1-dependent PAI-1 expression also blocked ERK1/2 activation/nuclear accumulation suggesting that the required src kinase activity is upstream of ERK1/2 in the TGF-beta1-initiated signaling cascade. The IC(50) of the PP1-sensitive kinase, furthermore, specifically implied involvement of pp60(c-src) in PAI-1 induction. Indeed, addition of TGF-beta1 to quiescent R22 cells resulted in a 3-fold increase in pp60(c-src) autophosphorylation and kinase activity. Transfection of a dominant-negative pp60(c-src) construct, moreover, reduced TGF-beta1-induced PAI-1 expression levels to that of unstimulated controls or PP1-pretreated cells. A >/=170 kDa protein that co-immunoprecipitated with TGF-beta1-activated pp60(c-src) was also phosphorylated transiently in response to TGF-beta1. TGF-beta1 is known to transactivate the 170 kDa EGF receptor (EGFR) by autocrine HB-EGF or TGF-alpha mechanisms suggesting involvement of EGFR activation in certain TGF-beta1-initiated responses. Incubation of quiescent R22 cells with the EGFR-specific inhibitor AG1478 prior to growth factor (EGF or TGF-beta1) addition effectively blocked EGFR activation as determined by direct visualization of receptor internalization. AG1478 suppressed (in a dose-dependent fashion) EGF-induced PAI-1 protein levels and, at a final concentration of 2.5 muM, virtually eliminated EGF-dependent PAI-1 synthesis. More importantly, AG1478 similarly repressed inducible PAI-1 levels in TGF-beta1-stimulated R22 cultures. PP1, PD98059, and U0126 also inhibited TGF-beta1-dependent cell motility at concentrations that significantly attenuated PAI-1 expression. Consistent with the AG1478-associated reductions in EGF- and TGF-beta1-stimulated PAI-1 expression, pretreatment of R22 cell cultures with AG1478 effectively suppressed growth factor-stimulated cell motility. These data indicate that two major phenotypic characteristics of TGF-beta1-exposure (i.e., transcription of specific target genes [e.g., PAI-1], increased cell motility) are linked in the R22 vascular smooth muscle cell system, require pp60(c-src) kinase activity and MEK signaling and involve activation of an AG1478-sensitive (likely EGFR-dependent) pathway.