HIF1 transcription factor regulates laminin-332 expression and ...

2992

Research Article

HIF1 transcription factor regulates laminin-332 expression and keratinocyte migration Giorgos Fitsialos1,2,*, Isabelle Bourget1,2,*, Séverine Augier1,2, Amandine Ginouvès3, Roger Rezzonico1,2, Teresa Odorisio4, Francesca Cianfarani4, Thierry Virolle1,2, Jacques Pouysségur3, Guerrino Meneguzzi1,2, Edurne Berra3,5, Gilles Ponzio1,2,‡ and Roser Buscà1,2 1

INSERM U634, 28 avenue de Valombrose, F-06107 Nice, France Université de Nice Sophia Antipolis, Faculté de Médecine, F-06000 Nice, France CNRS UMR 6543, Centre Antoine Lacassagne, 33, avenue de Valombrose, 06189 Nice, France 4 Laboratory of Molecular and Cellular Biology, IDI-IRCCS, via Monti di Creta, 104 00167 Roma, Italy 5 CICbioGUNE, Cell Biology and Stem Cells Unit, Technologic Park of Bizkaia, 48160 Derio, Spain 2 3

*These authors equally contributed to this work ‡ Author for correspondence (e-mail: [email protected])

Journal of Cell Science

Accepted 16 June 2008 Journal of Cell Science 121, 2992-3001 Published by The Company of Biologists 2008 doi:10.1242/jcs.029256

Summary Epidermal wound repair is a complex process involving the fine orchestrated regulation of crucial cell functions, such as proliferation, adhesion and migration. Using an in vitro model that recapitulates central aspects of epidermal wound healing, we demonstrate that the transcription factor HIF1 is strongly stimulated in keratinocyte cultures submitted to mechanical injury. Signals generated by scratch wounding stabilise the HIF1α protein, which requires activation of the PI3K pathway independently of oxygen availability. We further show that upregulation of HIF1α plays an essential role in keratinocyte migration during the in vitro healing process, because HIF1α

Introduction The epidermis is permanently submitted to stress and trauma, including UV radiation from sunlight, chemicals, infectious attacks and burn injuries. The keratinocytes provide a tight stratified barrier that maintains the integrity of the epidermis, which is crucial to ensure the protective capacity of this widely extended body organ. New differentiating keratinocytes continuously emerge from the proliferative basal layer of the epidermis to replenish upper layers, progressively differentiating into the exterior cornified and desquamating dead envelope. In order to understand the molecular nature of the mechanisms governing epidermal repair, we developed a ‘helicoidal scarificator’, which submits keratinocyte epithelial sheets, obtained in vitro, to massive mechanical injury. Such an injury mimics some aspects of epidermal wound healing and allows the screening of genes and proteins involved in this process (Turchi et al., 2002). Large-scale DNA array analysis at different times after wounding revealed specific transcriptomal signatures, including upregulation of numerous genes that had never been previously identified in other epidermal wounding systems (Fitsialos et al., 2007). A number of these genes are known to be targets of the HIF1 transcription factor. HIF1 is a master regulator of oxygen homeostasis and is composed of HIF1α , a finely regulated subunit, and HIF1β (also named ARNT), a ubiquitously expressed subunit (Semenza, 1998). HIF1 is a member of the basic helix-loop-helix (bHLH) family of transcription factors in which the HLH domains mediate α- and βsubunit heterodimerisation and the basic domain binds to a 5⬘RCGTG-3⬘ consensus sequence called the hypoxia-responsive

inhibition dramatically delays the wound closure. In this context, we demonstrate that HIF1 controls the expression of laminin-332, one of the major epithelial cell adhesion ligands involved in cell migration and invasion. Indeed, silencing of HIF1α abrogates injury-induced laminin-332 expression, and we provide evidence that HIF1 directly regulates the promoter activity of the laminin α3 chain. Our results suggest that HIF1 contributes to keratinocyte migration and thus to the reepithelialisation process by regulating laminin-332. Key words: Keratinocytes, HIF1, Migration, Laminin-332

element (HRE), which is found in HIF1 target gene promoters. HIF1 controls the expression of almost 70 genes involved in several aspects of cancer progression, including metabolic adaptation, apoptosis resistance, angiogenesis, invasion and metastasis (for reviews, see Pouysségur et al., 2006; Semenza, 2002). HIF1α is tightly regulated by multiple post-translational modifications that control protein stability in an oxygen-dependent manner through the ubiquitin-proteasome pathway (Maxwell et al., 1999). At normal oxygen levels (normoxia), the HIF1 prolyl hydroxylases (PHDs) hydroxylate HIF1α, allowing the binding of the E3 ligase von Hippel-Lindau (pVHL) and leading to HIF1α ubiquitylation, which targets the protein to the proteasome for degradation (Berra et al., 2006). Under hypoxia, the hydroxylase activity decreases, which reduces HIF1α hydroxylation and attenuates protein degradation. Although oxygen tension plays a major role in the control of HIF1 stability, this factor can be further modulated by either transcriptional (Buscà et al., 2005) or post-transcriptional mechanisms activated by hormones, or growth-factor-dependent pathways including the MAPK or PI3K cascades (for review, see Bardos and Ashcroft, 2005). However, HIF1 regulation in skin cells has been poorly investigated. Furthermore, very few studies dissect the role of this transcription factor and its target genes in cell migration, in particular in the context of the epidermis. The transcriptomal analysis of wounded keratinocytes in vitro demonstrated that a number of genes involved in adhesion and migration are upregulated after ‘injury’. Among these genes, we highlighted those encoding the three chains (α3, β3, γ2) of laminin332 (formerly laminin-5) (Aumailley et al., 2005). Laminin-332 is


HIF1 in cell migration and laminin expression a secreted extracellular matrix protein and a major adhesive component of the epidermal basement membrane (Carter et al., 1991; Rousselle et al., 1991), which tightly links the basal keratinocytes to the underlying dermis. Laminin-332 is a component of the anchoring filaments of the basement membrane, and its interaction with the integrin α6β4 induces nucleation of the hemidesmosomes (Borradori and Sonnenberg, 1999). In vitro, laminin-332 also promotes proliferation, spreading, scattering, survival and migration of normal keratinocytes through its interaction with integrins α3β1 and α6β4, and the subsequent activation of different signalling pathways (Kariya and Miyazaki, 2004; Murgia et al., 1998; Nguyen et al., 2000; Nikolopoulos et al., 2005; Rousselle and Aumailley, 1994). Laminin-332 also stimulates in vitro migration and invasion of human tumour cells (Fukushima et al., 1998; Kariya and Miyazaki, 2004). In the healing process, laminin-332 expression is induced in keratinocytes at the wound edge. The potent cell-migration-promoting activity of laminin-332 is thought to contribute to wound healing (Goldfinger et al., 1998; Nguyen et al., 2000) and tumour invasion (Katayama and Sekiguchi, 2004; Marinkovich, 2007; Pyke et al., 1995). Even though the role of laminin-332 in human skin wound healing has been largely confirmed, the characterisation of the signals that trigger its regulation in the particular context of keratinocyte reepithelialisation is currently not fully understood. In this paper, we demonstrate a strong upregulation of HIF1α after the disruption of the keratinocyte monolayer integrity in normoxia. Furthermore, we provide evidence for the role of HIF1 in keratinocyte migration. Finally, we demonstrate that laminin-332 is a new HIF1 target and might be largely responsible for the promigratory role of this transcription factor.

2993

Results HIF1 target genes and HIF1α protein are induced in keratinocytes upon in vitro wounding

In order to dissect the molecular events involved in epidermal repair, we performed a transcriptomal array that would identify the genes induced upon scratch wounding of confluent human keratinocyte cell cultures using a standardised and well-characterised experimental procedure (Fitsialos et al., 2007). Among the genes upregulated in wounded cells, several were bona fide HIF1 transcriptional targets, namely vascular endothelial growth factor A (VEGFA), plasminogen activator inhibitor 1 (PAI1), ETS1, insulin-like growth-factor-binding protein 3 (IGFBP3), N-myc downstream-regulated gene 1 protein (NDRG1) or cyclooxygenase 2b (COX2) (for review, see Denko et al., 2003) (Fig. 1A). Therefore, we focused our attention on the regulated α subunit of HIF1. However, when analysing the DNA array, we observed that HIF1α mRNA levels did not vary significantly after wounding. These results were further confirmed by real-time quantitative PCR (Fig. 1B). As clearly established in the literature, the main regulation of HIF1α takes place at a posttranslational level by stabilisation of the polypeptide. Thus, we analysed HIF1α protein levels by western blot. We found that HIF1α highly increased in a transient manner after scratch wounding of confluent NHKs or HaCaT keratinocyte monolayers in normal oxygen conditions (Fig. 2A). Immunofluorescence experiments also revealed an increase in the nuclear HIF1α staining within the keratinocytes migrating from the wound edge (Fig. 2B upper panel) as well as within the confluent monolayer further away from the scratch site (Fig. 2B lower panel). A mouse model was used to assess the induction of HIF1α protein

Fig. 1. Keratinocyte scratch wounding upregulates HIF1 target genes but not expression of HIF1α mRNA. (A) Wildtype human keratinocytes were scratch wounded using the ‘scarificator’ device. Total mRNA was extracted from the cells at different times after injury and submitted to biochip screening. Histograms represent the mRNA increase of six different HIF1 target genes found induced in the DNA array screening. (B) HIF1α mRNA levels in wounded cells measured by DNA array (upper panel) and by real-time quantitative PCR (lower panel). NW, non-wounded cells. Data are mean ± s.e.m. of three independent experiments performed in triplicate.


2994

Journal of Cell Science 121 (18)

Fig. 2. Wounding increases HIF1α protein concentration in keratinocyte cells. (A) Western blot detection of HIF1α from normal human keratinocytes (NHK) and HaCaT cell lysates at different times after scratch wounding. GAPDH was monitored as a gel-loading control. NW, not-wounded; W, wounded. (B) Immunofluorescence analysis of HIF1α in NHKs wounded using a micropipette tip (green labelling). Nuclei were stained with DAPI dye (blue). NW, not-wounded; W, 15 hours after wounding. In the upper panel, the dotted line corresponds to the wound incision and the arrows indicate the direction of migrating keratinocytes. The lower panel shows images depicting the activation of HIF1α in keratinocytes far from the wounded area of the cell monolayer. Scale bar: 20 μm. (C) HIF1α immunolabelling using peroxidase staining of paraffin slices obtained from mice epidermis harvested 1, 3, 5, 7 and 9 days (d) after injury. Arrows indicate HIF1α-positive keratinocytes. Scale bar: 60 μm.

upon injury in more physiological conditions. Mice were punched as described in the methods section, and skin histological slices were obtained from 1 to 9 days after punch injury. HIF1α protein was then detected using the anti-HIF1α antibody. As observed in Fig. 2C, HIF1α was not detected in the keratinocyte nuclei of nonwounded mouse epidermis. However, only some nuclei of the keratinocyte migrating tongue showed HIF1α-positive staining 1 day after injury. On day 3, the number of HIF1α-labelled cells increased greatly in the re-epithelialising keratinocytes all over the epidermis, and remained high at day 5. At day 7, the labelling decreased and HIF1α was no longer detected once the wound was completely closed at day 9.

HIF1 plays an important role in keratinocyte cell migration

In order to focus on the putative physiological role of HIF1, we analysed cell migration, one of the major biological processes involved in the closure of the wound bed. We knocked down HIF1α expression in HaCaT keratinocytes by specific siRNAs and evaluated cell migration using time-lapse videomicroscopy and Transwell Boyden chamber assays. Cells grown up to 80% confluence were either transfected with a control siRNA (siCtr) or

The PI3K pathway mediates the injury-dependent HIF1α increase

A compilation of reports has evoked a post-transcriptional regulation of HIF1α expression via the PI3K-AKT pathway in normoxia (for review, see Maynard and Ohh, 2007). Since our scratch experiments were performed in normoxia and because we have previously shown that keratinocyte wounding activates the PI3K pathway (Fitsialos et al., 2007), we hypothesised that this signalling cascade could account for the HIF1α post-transcriptional upregulation that we observed. Primary human keratinocytes and HaCaT cells were treated or not with the PI3K inhibitor LY294002 and HIF1α protein expression was analysed by western blotting at different times after injury. We observed that PI3K inhibition, determined by the decrease in AKT phosphorylation, strongly inhibited the injury-stimulated HIF1α expression (Fig. 3). These results demonstrate that, in wounded keratinocytes, the PI3K pathway activation is necessary to promote HIF1α expression. Furthermore, additional experiments demonstrated that activation of the PI3K pathway is mandatory to maintain high HIF1α protein levels (data not shown) meaning that stabilisation of HIF1α requires active PI3K signalling.

Fig. 3. PI3K inhibition abrogates the injury-dependent increased levels of HIF1α. HaCaT cells and normal human keratinocytes (NHK) were incubated for 1 hour in the presence of the PI3K inhibitor LY294002 (15 μM) and were then scratch wounded. Cells were collected 1 and 3 hours after injury and submitted to western blot analysis to detect HIF1α and phosphorylated AKT (as a reporter of the PI3K activity). NW, non-wounded cells. GAPDH was detected as a loading control.

HIF1 in cell migration and laminin expression a siRNA targeting HIF1α (siHIF1α) . For time-lapse microscopy, the confluent monolayer was manually scratch wounded and the wound closure was filmed for 24 hours. Images were extracted from the film immediately after scratching (t=0) as well as 12 and 24 hours after injury and the kinetics of wound closure were quantified (Fig. 4A and B). For Transwell in vitro migration assays, cells were trypsinised 48 hours after siRNA transfection and deposited into the Transwell insert. Migrating cells were analysed and quantified as described in the Materials and Methods. The results showed that, independently of the migration assay, HIF1α silencing inhibited the in vitro migration by almost 50% (Fig. 4A and B). It is worth mentioning that cell proliferation (assessed by cell counting experiments as well as BrdU incorporation or cyclin A labelling assays) was not affected by HIF1α silencing at the defined time points (not shown). This strongly suggests a specific role of HIF1 in cell migration. To confirm the abrogation of HIF1α expression by the transfected siRNAs, we performed real-time quantitative PCR and by western blot analysis (Fig. 4C).


Laminin-332 expression increases upon scratch wounding and correlates with HIF1α induction

To investigate potential molecular mechanisms responsible for the HIF1-dependent regulation of keratinocyte migration, we searched, in our transcriptomal screening (Fitsialos et al., 2007), for genes encoding proteins involved in keratinocyte adhesion or motility. We

2995

highlighted the induction of the three genes encoding the different chains of laminin-332, namely LAMA3 for the α3 chain (Fig. 5A) that increased from 3 hours to 24 hours after scratch wounding, but also LAMB3 for the β3 chain and LAMC2 for the γ2 chain (not shown). This induction was confirmed by real-time quantitative PCR (Fig. 5A) and by immunofluorescence experiments (Fig. 5B). As a result of the upregulation of its α3, β3 and γ2 subunits, laminin332 protein clearly increased upon scratch wounding, and this increase correlated with the HIF1α protein upregulation occurring in the migrating primary keratinocytes (Fig. 5B upper panel), as well as in cells located further within the confluent monolayer (Fig. 5B lower panel). These increases were detected at early times after injury, but were visualised more easily after 24 hours (Fig. 5B). It is worth noting that two antibodies were used to detect laminin332 in immunofluorescence experiments: either GB3 monoclonal antibody recognising the native trimeric form of the protein, or the BM165 monoclonal antibody specific to the α3 chain (Rousselle et al., 1991). As expected, both antibodies revealed an identical pericellular and extracellular labelling pattern for laminin-332, indicating that the upregulation of the α3 chain was accompanied by an increase of the trimeric (α3β3γ2) protein expression and secretion (Fig. 5B). To confirm the correlation between HIF1α induction and laminin-332 upregulation, we induced the expression of HIF1α protein in keratinocytes by treatment for 15 hours with CoCl2

Fig. 4. Depletion of HIF1α inhibits keratinocyte cell migration. (A) HaCaT keratinocytes transfected with a nonrelevant siRNA (siCtrl) and a siRNA targeting HIF1α (siHIF1α) were scratch wounded with a micropipette tip and filmed for 24 hours using time-lapse videomicroscopy. The wound closure is illustrated by showing the wound size (arrow and a) immediately after injury and 12 and 24 hours after wounding. The right panel represents the quantification of the siHIF1α effect on the wound closure calculated by the measure of the diminution of the wound bed surface upon time by using Image J software. Results are represented as the percentage of the wound closure of monolayers obtained with cells transfected with the siHIF1α compared with that observed in the cultures of cells transfected with the nonrelevant siRNA (siCtrl) (taken as the 100% value). Each experiment was performed at least three times, and different areas of several scratches were randomly taken. (B) siRNA-transfected HaCaT keratinocytes were seeded in the upper chambers of Transwell inserts. After 24 hours, cells on the lower side of the filter were stained with DAPI and scored in five independent fields. Experiments were carried out three times and each condition was performed in triplicate. (C) Verification of HIF1α abrogation by the specific siRNA duplex. Transfected cells were scratch wounded, harvested 6 hours later then HIF1α mRNA and protein were analysed respectively by real time Q-PCR (left panel) and by western blotting (right panel).


2996


Fig. 5. Expression of laminin-332 is upregulated upon scratch wounding and correlates with HIF1α induction. (A) Histograms representing the quantification obtained by microarray analysis of the LAMA3 transcripts at different time points after cell wounding (left panel). Results were confirmed by real time Q-PCR (right panel). Values are mean ± s.e.m. of three independent experiments performed in triplicate. NW, not-wounded; W, wounded. (B) Immunofluorescence detection of HIF1α (green) and laminin-332 (BM165 antibody; red), 15 hours after wounding (W), in NHKs (upper panel) and HaCaT (lower panel), compared with nonwounded counterparts (NW). Nuclei were detected using DAPI (blue). The dotted line in the upper panel highlights the original position of the scratch wound edge, and the arrows indicate the direction of cell migration. The lower panel shows HaCaT cells in a nonwounded area of the cell culture far from the wound site. Scale bar: 20 μm. (C) Left panel, HaCaT cells untreated (NS) or treated with Cobalt (Co2+) for 15 hours to increase HIF1α protein. HIF1α protein that was measured by western blot analysis. GAPDH was detected as a loading control. Right panel, real-time quantitative PCR showing normalised LAMA3 mRNA expression in HaCaT cells nonstimulated (NS) or treated with Co2+ for 15 hours. (D) HaCaT (lower panel) and NHK cells (upper panel) were either left untreated (NS) or treated with Co2+. After 24 hours, cells were fixed and immunofluorescence analysis was carried out to detect HIF1α protein (green) and laminin-332 (red). Nuclei were stained using the DAPI. Scale bar: 20 μm.

(Co2+), which stabilises HIF1α protein (Yuan et al., 2003) (Fig. 5C). We next analysed laminin-332 expression by measuring, by real-time quantitative PCR, the LAMA3 mRNA (Fig. 5C). We clearly observed that Co2+ strongly induced the expression of the mRNA encoding the α3 subunit. Then we measured the laminin332 protein levels and distribution by immunofluorescence upon 15 to 24 hours of Co2+ exposure. We observed that laminin-332 labelling increased concomitantly with the upregulation of nuclear HIF1α. This happened in both HaCaT cells and normal human keratinocytes (NHKs) (Fig. 5D). We therefore conclude that an increase in HIF1α protein correlates highly with upregulation of laminin-332. HIF1 regulates laminin-332 expression

To definitively demonstrate the role of HIF1 in laminin-332 induction, we analysed the effect of HIF1α silencing on the expression of laminin-332 measured both by immunofluorescence and by real-time PCR quantification of LAMA3 mRNA expression. We transfected HaCaT keratinocytes, with a control siRNA or with

one of two different siRNAs targeting HIF1α. The confluent cell monolayer was then submitted to scratch wounding using the helicoidal scarificator. Total mRNAs were extracted from cells in non-wounded conditions (NW) as well as 6 hours after injury (W) and the LAMA3 transcripts were measured by real-time quantitative PCR. We found that HIF1α cell depletion inhibited the woundingdependent induction of the LAMA3 mRNA by more than 60% (Fig. 6A). The same experiment was performed to visualise the effect of HIF1α siRNA on the expression of the laminin-332 protein after scratch wounding or CoCl2 treatment by immunofluorescence in HaCaT cells. We observed that when the injury- or CoCl2-induced HIF1α expression was strongly diminished by the specific siRNA transfection, cells showed a reduced labelling for laminin-332 (Fig. 6B). The same result was obtained after transfection of a siRNA targeting PHD2 that stabilised and upregulated HIF1α expression (Berra et al., 2003). Indeed, we observed that, in confluent unwounded HaCaT cells, the PHD2-dependent HIF1α protein upregulation was accompanied by an increase in laminin-332 (Fig.


HIF1 in cell migration and laminin expression

Fig. 6. HIF1α depletion downregulates laminin-332. (A) Real-time quantitative PCR analysis of LAMA3 mRNA expression in HaCaT cells transfected with a nonrelevant siRNA (siCtrl) (grey) or a siRNA to HIF1α (siHIF1α) (black). Q-PCR analysis was performed using transfected cells either not wounded (NW) or harvested 6 hours after wounding (W). Results were obtained from three independent experiments performed in triplicate and are expressed as normalised LAMA3 mRNA ‘fold increase’ relative to the siCtrl-transfected non-wounded cells (siCtrl-NW). (B) Immunofluorescence detection of HIF1α (green) and laminin-332 (red) in HaCaT cells transfected with the nonrelevant siCtrl, a siRNA to HIF1α or a siRNA targeting PHD2 (that stabilised and upregulated HIF1α expression). Cells transfected with the siPHD2 siRNA were not wounded. Cells were fixed 15 hours after either scratch wounding (Wounded) or CoCl2 treatment (Co2+ treated). Nuclei were stained using DAPI. Scale bar: 40 μm.

6B, lower panel). Taken together, these results demonstrate that HIF1α contributes to the regulation of laminin-332 expression in keratinocytes. To further evaluate the direct impact of HIF1 on laminin-332 transcriptional upregulation, we performed luciferase reporter assays to measure the activity of the human LAMA3 promoter in basal conditions and when HIF1α was overexpressed. To this aim, HaCaT cells were cotransfected with a plasmid encoding a promoter fragment of the human laminin-332 α3 chain cloned upstream of the luciferase reporter gene (Virolle et al., 2002) and either an empty vector (pcDNA3) or a construct encoding the HIF1α gene (HIF1A). Luciferase values revealed that the activity of the LAMA3 promoter significantly increased upon HIF1α

2997

Fig. 7. HIF1α directly binds to the human LAMA3 promoter and stimulates its activity. (A) Luciferase reporter assays of the human LAMA3 promoter activity performed in HaCaT cells transfected with either an empty vector (pcDNA3) or constructs encoding either HIF1α or USF1. Results are expressed as the fold stimulation of the luciferase activity measured in cells transfected with the pcDNA3 empty vector. (B) Functional activity of the HIF1α construct was verified by cotransfecting HIF1α together with a reporter plasmid containing three HRE consensus sites cloned upstream a minimal thymidine kinase (TK) promoter and the luciferase reporter gene. (C) Partial DNA nucleotide sequence of the LAMA3 promoter. Nucleotide numbering is relative to the first nucleotide of the transcription initiation site (+1). AP1a, AP1b and AP1c indicate the position of the three reported AP1 sites. The HRE consensus sequence is represented by a box at position –108. Arrows indicate the transcription start site and the beginning of the DNA coding sequence. (D) Effect of HIF1α on the transcriptional activity of the human LAMA3 promoter mutated at the –108 HRE consensus site. The Mut hLAMA3 promoter driving the expression of the luciferase reporter gene was cotransfected in HaCaT cells with either an empty vector (pcDNA3) or a construct encoding HIF1α. A control condition where the Mut hLAMA3-Luc promoter was cotransfected with an USF1-encoding plasmid was also performed. In A, B and D, the histograms show the mean ± s.e.m. of the results obtained in three independent experiments resulting from transfections performed using different plasmid preparations. (E) A ChIP assay was performed on extracts from 6 hour CoCl2-stimulated HaCaT keratinocytes. HIF1α immunoprecipitation was performed using a specific anti-HIF1α antibody. Primers spanning the LAMA3 (upper panel) or the VEGF (lower panel) promoter regions were used for PCR amplification. In (input) represents the control of the PCR amplification performed using HaCaT genomic DNA, which showed a 142 bp and a 104 bp band corresponding to the amplification of the LAMA3 and VEGF promoter regions, respectively. The IgG (M,R) and Pol2 lanes indicate the PCR amplification obtained when the extracts were immunoprecipitated with a non immune mouse immunoglobulin (M), a non relevant rabbit antibody (R) or an antibody raised against the RNA polymerase II (Pol2). The Pol2 antibody was included in the kit and represents a supplemental positive control, according to the manufacturer. PG, amplification after immunoprecipitation with protein-G-coupled beads only.


2998


cotransfection, albeit to a lesser extent than upon cotransfection with a plasmid encoding the USF1 transcription factor that was used as a positive control (Virolle et al., 2002) (Fig. 7A). To verify the efficiency of the HIF1A construct we transfected a reporter plasmid containing three HRE consensus sites upstream of a minimal thymidine kinase promoter and the luciferase reporter gene. As expected, HIF1A cotransfection exerted a strong stimulatory effect on this HRE-promoter activity (Fig. 7B). Taking into account that the human transfected LAMA3 promoter contains a putative HRE consensus site (GCGTG) at position –108 from the transcription initiation site (Fig. 7C), we investigated whether this sequence could be required for the HIF1-dependent regulation of the promoter activity. We therefore mutated this site and created a novel mutant construct designated Mut hLAMA3 promoter. When we cotransfected this reporter plasmid together with the HIF1A construct, we observed a loss of the HIF1-dependent promoter induction (Fig. 7D). By contrast, USF1 stimulated the mutant promoter activity as efficiently as the non-mutant one. This led us to hypothesise that this HRE site is functional for HIF1 binding and it is involved in the HIF1-dependent LAMA3 stimulation. Moreover, these experiments suggest that HIF1 could regulate LAMA3 via a direct interaction with its promoter sequence. To verify this assertion, we performed chromatin immunoprecipitation assays (Fig. 7E). HaCaT cells were stimulated with 200 μM CoCl2 for 6 hours to maximally increase HIF1α protein expression. Chromatin complexes were immunoprecipitated with the antiserum 2087 to human HIF1α and a PCR amplification was performed using specific primers to the human LAMA3 promoter region. As shown in Fig. 7E (upper panel), we detected a specific amplification of the LAMA3 promoter region in cells treated with CoCl2 after chromatin immunoprecipitation using anti-HIF1α antibody. As a positive control, we used specific primers spanning the VEGF promoter region, which are well known targets of HIF1 (Fig. 7E, lower panel). Specific bands corresponding to LAMA3 and VEGF were detected when using human genomic DNA (Fig. 7E, upper and lower panels, respectively). In the presence of a mouse immunoglobulin or a nonrelevant rabbit antiserum, no specific amplification was observed. We therefore conclude that HIF1 binds to the LAMA3 promoter in the chromatin context of HaCaT cells where HIF1α has been induced. Taken together, these results show that the role of HIF1 in laminin-332 expression in keratinocytes is mediated by direct control of the transcriptional activity of LAMA3 promoter.

Discussion In the present work we describe the post-transcriptional and oxygenindependent induction of HIF1α protein upon the scratch disruption of a human keratinocyte sheet using a specific experimental approach that recapitulates some aspects of epidermal reepithelialisation, a major event in skin wound healing. We also report that HIF1α protein increases in vivo upon injury in mouse epidermis. Furthermore, we identify HIF1 as a key player in keratinocyte migration and demonstrate that laminin-332 expression is regulated by HIF1 transcription factor. Considering the promigratory role of laminin-332, it is likely that, by regulating its expression, injury-stimulated HIF1 promotes keratinocyte cell migration in normal oxygen conditions and contributes to wound closure (see model proposed in Fig. 8). HIF1 transcription factor has been shown to play a key role in tumour progression by controlling the metabolism, the proliferation and migration of tumoral cells (Semenza, 1998). However, despite the multiple analogies existing between tumorigenesis and wound healing (Dvorak, 1986), the role of HIF1 in skin repair has not been clearly established, although some studies have started to address this question. A transcriptional upregulation of HIF1α in mouse skin wound healing has been previously evoked by Elson and coworkers (Elson et al., 2000). These authors showed, by using in situ hybridisation, that HIF1Α mRNA was induced after injury, but they did not analyse the regulation of HIF1α protein nor its physiological impact in skin repair. Hypoxia is considered to be the main signal that upregulates HIF1α protein levels (Zagorska and Dulak, 2004), most notably in tumours, and it is thus tempting to link hypoxia and wound healing. Authors have thus proposed that acute in vivo skin injury creates a local hypoxic microenvironment (Hunt et al., 1972; Niinikoski et al., 1972; Varghese et al., 1986). Nevertheless, HIF1α protein induction in this specific context and its putative functional role in the healing process, have never been investigated. Others have documented the hypoxia and thereby HIF1α expression in specific subcompartments of the epidermis, such as hair follicles and sebaceous glands. In other areas of normal human skin, HIF1α is present at only low levels. Our work clearly shows that HIF1α protein increases after in vitro wounding, which strongly suggests an active role in wound closure. Interestingly, this upregulation occurs in conditions of normoxia, indicating that a local hypoxia is not absolutely required to induce HIF1 in normal epidermal cells. However, if hypoxia is not at the origin of HIF1α upregulation, what are the molecular

Fig. 8. Illustration of a hypothetical mechanism for the role of HIF1 in the control of keratinocyte migration during wound closure. The disruption of the integrity and cohesion of the keratinocyte monolayer consequent to scratch wounding activates PI3K signalling, which leads to an increased expression of HIF1α and results in the upregulation of the HIF1 target genes among which is LAMA3 that encodes the α3 subunit of laminin-332 and reflects the upregulated synthesis of laminin-332. Enhanced deposition of laminin-322 to the ECM, strongly influences the keratinocyte adhesion and their migration capacity.


HIF1 in cell migration and laminin expression signals responsible for such induction? It is tempting to speculate that a large set of soluble factors released by keratinocyte wounding such as proheparin-binding EGF-like growth factor (HBEGF), proepiregulin (EREG), amphiregulin (AREG), protransforming growth factor α (TGFA) and platelet-derived growth factor (PDGF) or parathyroid hormone-related protein (PTHLH) (see Wenger, 2002), and also the pro-inflammatory cytokines, leukaemia inhibitory factor (LIF), interleukin-1 (IL-1) and IL-18 (Chandrasekar et al., 2005; Ohbayashi et al., 2007), could account for this high HIF1α upregulation. The majority of these factors are known to activate PI3K, which is in agreement with the observation that, in our model, this kinase plays a central role in HIF1α protein synthesis. However, these soluble factors probably only constitute a part of the many molecular mechanisms responsible for HIF1α induction. Indeed, we observed that the exposure of keratinocytes to conditioned medium collected from wounded cells only modestly upregulated HIF1α protein expression (not shown), thereby suggesting that additional signals, and particularly those generated by the alteration of the mechanical strength, namely shear stress, could initiate mechanotransduction, thereby driving the strong HIF1α increase that we detect. In our cell system, this rupture induces matrix metalloproteinase 9 (MMP9) expression (Turchi et al., 2003). Mechanical stress, which is at the origin of many cellular processes (reviewed in Li et al., 2005), has also been described to induce HIF1α in the myocardium, a very different cell context (Kim et al., 2002). Strong cell-cell cohesion and cell-ECM adhesion is essential for epithelium homeostasis, and the rupture of the cell monolayer upon scratch wounding launches signals that stimulate cell migration and proliferation to restore the lost ‘tissue’ integrity. It is worth noting that increased HIF1α was detected in migrating cells as well as in keratinocytes inside the cell monolayer. We can thus infer that the monolayer rupture induces a global reorganisation of the keratinocyte sheet. This can be easily observed by videomicroscopy, which clearly detects a global cell movement that correlates with the observed HIF1α induction in cells far from the wound edge (data not shown). This capacity of the keratinocyte sheet to globally reorganise is probably essential for proper wound bed closure. Indeed, time-lapse analysis showed that, in contrast to fibroblasts that migrate individually into the scratch ‘bed’ after injury, keratinocyte cells do not dissociate from each other but rather slide together into the wound because of high epithelial cellcell cohesion. We found that HIF1 induction is a key element of the molecular cascade generated by injury and controlling keratinocyte motility, because its depletion by specific siRNAs strongly delayed the closure of a scratch performed on confluent keratinocytes. Interestingly, HIF1α silencing did not affect keratinocyte cell proliferation at the analysed time points (not shown), suggesting a prevailing role of HIF1 in the control of cell migration in our experimental situation. Although several studies performed in other cell systems have shown that HIF1 can modulate integrin expression (Cowden Dahl et al., 2005), FAK activity (Corley et al., 2005), MMP expression (Miyoshi et al., 2006), the expression of ephrins (Vihanto et al., 2005) as well as the classical c-met receptor (Pennacchietti et al., 2003) arguing for an involvement of the factor in cell adhesion, motility and invasion, the direct action of HIF1 in keratinocyte migration has never been demonstrated. Here we identify a novel HIF1 target, laminin-332 (laminin 5), which is one of the main keratinocyte-secreted extracellular matrix proteins. We show that HIF1 upregulates the expression of the heterotrimeric laminin-332

2999

by stimulating the transcription of LAMA3 (the α3 subunit of laminin-332, which fully comprises α3, β3, γ2). The regulation of laminin-332 by HIF1 is in total agreement with the promigratory action of HIF1 that we observe. Indeed, laminin-332 is largely known to be induced during skin wound healing and to play a major part in keratinocyte adhesion and migration in physiological and pathological contexts (Amano et al., 2004; Kariya and Miyazaki, 2004; Nguyen et al., 2000; Ryan et al., 1994; for reviews, see Hartwig et al., 2007 and Nguyen et al., 2000). Considering the major role of laminin-332 in promoting keratinocyte migration, its control by HIF1 could explain the antimigratory effect of HIF1α siRNA in our in vitro wounding model, and supports the notion that HIF1 is indeed a major player in keratinocyte motility. Our results seem to contradict another report (O’Toole et al., 1997), which observed a decreased secretion of laminin-332 upon hypoxia. However, our experimental conditions were not hypoxic and these authors did not look further into the role of HIF1 in this process. We confirmed the involvement of HIF1 in laminin-332 regulation by showing that HIF1 is involved in the control of the transcription of the α3 subunit (LAMA3), which is mandatory for the proper biological function of this laminin (Gagnoux-Palacios et al., 2001; Schneider et al., 2007). Moreover, we provide precise evidence demonstrating a new mechanism of laminin-332 regulation in the wound-healing context. Upon overexpression of HIF1α and by performing chromatin immunoprecipitation assays, we further demonstrate that HIF1 directly binds the HRE consensus sequence located at the position –108 of the LAMA3 promoter and regulates the transcriptional activity of this laminin-332 subunit. To our knowledge, this is the first time that direct transcriptional HIF1-dependent regulation of an extracellular matrix protein is reported. We further suggest that laminin-332 is an HIF1 target protein that might mediate the HIF1controlled keratinocyte migration. Taking into account the reported involvement of laminin-332 in keratinocyte motility, in tumour growth and invasion (for a review, see Miyazaki, 2006), we can envisage that, upon repeated injuries or in chronic wounds, a deregulation of HIF1α levels could trigger the enhancement of the invasion and metastatic potential of keratinocytes in aggressive epidermal carcinomas. Thus, the expression of HIF1α and laminin332 in these cancers might help to evaluate their tumoral potential. Finding ways to target HIF1 in specific skin pathological situations might provide novel therapeutic tools. Materials and Methods Cell culture Human keratinocytes isolated from healthy neonatal foreskin as described (Rheinwald and Green, 1975) were cultured on a feeder of lethally irradiated 3T3-fibroblasts (Turchi et al., 2002). The HaCaT keratinocytes (Boukamp et al., 1988) were grown in DMEM (Gibco-BRL/Invitrogen, Carlsbad, CA) containing 10% fetal calf serum (Perbio Science AB, Helsingborg, Sweden). To inhibit PI3K, cells were treated with LY294002 (Calbiochem, San Diego, CA) 1 hour before scratch wounding.

Plasmids, siRNAs and PCR oligonucleotide sequences The plasmid encoding the human LAMA3 promoter fragment upstream of the luciferase reporter gene is described (Virolle et al., 2002). The 3-HRE-LUC reporter vector contains three copies of the HRE (hypoxia responsive element) from the erythropoietin promoter cloned upstream of a minimal thymidine kinase promoter and the luciferase reporter gene (Berra et al., 2003). The plasmids encoding HIF1α and USF1 have been previously described (Richard et al., 1999) and (Pognonec et al., 1992), respectively. For HIF1α silencing, two siRNA DNA sequences were used: 5⬘-GCCACTTCGAAGTAGTGCT-3⬘ and 5⬘-CCTATATCCCAATGGATGATG-3⬘ (Buscà et al., 2005; Warnecke et al., 2003). To silence PHD2, the DNA sequence was: 5⬘CTTCAGATTCGGTCGGTAAGG-3⬘ (Berra et al., 2003). As non-relevant siRNAs, we used the siLUC: 5⬘-CGTACGCGGAATACTTCGA-3⬘ (Buscà et al., 2005) and the Drosophila melanogaster HIF1α gene (SIMA): 5⬘-CCTCATCCCGATCGATGATG-3⬘ as described elsewhere (Berra et al., 2003; Busca et al., 2005).

3000


Antibodies

Chromatin immunoprecipitation assay

The rabbit polyclonal anti-HIF1α antibody (antiserum 2087) was described (Richard et al., 1999). The mouse monoclonal antibodies GB3 to native laminin332 (Verrando et al., 1987) and BM165 to laminin α3 (Rousselle et al., 1991) were used at dilutions of 1:200 and 1:100 respectively. The anti-phospho-AKT antibody (1:6000) was from Cell Signalling Technology (Invitrogen, Carlsbad, CA). The polyclonal anti-GAPDH antibody (Santa Cruz Biotechnology, Santa Cruz, CA) was used at a 1:8000 dilution. The anti-mouse and anti-rabbit peroxidase-conjugated antibodies (Dakopats Hamburg, Germany) were used at 1:20,000 dilution. The Alexa-Fluor-488 and -594 anti-mouse and anti-rabbit (Molecular Probes, Eugene, OR) were used at 1:400 and 1:500 dilutions, respectively. An anti-Hsp60 (from Santa Cruz Biotechnology) was also used to perform gel-loading controls.

HaCaT cells were stimulated for 6 hours with CoCl2 and were then treated with 1% formaldehyde for 5 minutes at 37°C and for 30 minutes at 4°C. Cells were harvested and the chromatin immunoprecipitation was performed using the Chromatin Immunoprecipitation (ChIP) Assay Kit from Upstate Biotechnology (Upstate Biotechnology, Lake Placid, NY) according to the manufacturer’s instructions. After immunoprecipitation, the genomic captured fragments were then recovered by phenolchloroform extraction. Identification of the LAMA3 or VEGF promoter fragments was performed by PCR analysis using specific promoter primers. 35 cycles of PCR were performed and the amplified products were analysed on a 2% agarose gel.The antiserum used for immunoprecipitation was directed against the C-terminus of HIF1α and has been described (Richard et al., 1999). The oligonucleotide sequences used to amplify the LAMA3, and VEGF promoters by PCR were respectively LAMA3: F, 5⬘-CTTCCTCCATCTCTTTTTAGCCTG-3⬘, R, 5⬘-CATCACCTTCCACCCACCTCAC-3⬘; VEGF: F, 5⬘-GAGAGAGACGGGGTCAGAGGAGAG-3⬘, R, 5⬘-CAGCGCGACTGGTCAGCTG-3⬘.

RNA extraction real-time quantitative PCR and DNA array screening Total RNAs were extracted using TRIzol reagent (Invitrogen) and RNeasy protect mini kit (Qiagen, Courtaboeuf, France). mRNA levels were quantified by one-step real-time RT-PCR using SYBR-Green PCR Master Mix (Applied Biosystems, Foster City, CA). The oligonucleotide sequences corresponding to the different genes analysed are available upon request. The DNA array screening used in this study has been described (Fitsialos et al., 2007).

Scratching protocol and western blotting


The standardised scratch wounding conditions using the ‘helicoidal scarificator’ were described previously (Turchi et al., 2002). After wounding, cells were washed with PBS and frozen at –80°C. Proteins were lysed in Laemmli buffer and analysed by western blot (Turchi et al., 2002). For time-lapse and immunofluorescence experiments, scratch wounding was performed using a micropipette tip.

Transfections and luciferase assays, site-directed mutagenesis HaCaT cells were transiently transfected using lipofectamine 2000 (GIBCO-BRL), according to the manufacturer’s protocol, lysed 48 hours later in cell lysis buffer (Promega France, Charbonnières, France) and assayed for luciferase activity (Promega). Directed mutagenesis of the HRE site was performed using the Quickchange directed mutagenesis kit (Stratagene/Agilent Technologies, Wilmington, DE) according to the manufacturer’s protocol. The oligonucleotide DNA sequence was 5⬘-CACCCTCCTACCCAGCTCGAGTTCCCTGCCGTGACTC-3⬘. The siRNA duplex transfection (50 nM) was carried out using either the calcium phosphate technique (Berra et al., 2003), or the RNAiMax lipofectamine reagent as devised by the manufacturer (Invitrogen). HaCaT cells were scratch wounded 36 hours after transfection when the culture reached confluence.

Immunofluorescence microscopy Cells grown on coverslips were fixed in paraformaldehyde (PFA) and permeabilised in 0.1% Triton X-100. The cells were incubated for 1 hour with the primary antibody diluted in PBS with 0.1% bovine serum albumin, then with the appropriate conjugated secondary antibody. Nuclei were stained with DAPI, coverslips were mounted on glass slides and viewed with an Axiophot fluorescent microscope (Carl Zeiss S.A.S., Le Pecq, France). To acquire the images, a COHU (High performance CCD camera) and the Leica Q-Fish (Leica Microsystems, Tokyo, Japan) informatics software were employed.

Mice wound-healing experiments and immunohistochemistry A 6-mm-diameter full-thickness wound was punched on the dorsal midline of 8week-old C57Bl/6 male mice using a biopsy punch. Wounded areas surrounded by unwounded skin were dissected at day 1, 3, 5, 7 and 9 after injury, fixed in PFA and embedded in paraffin. Wound specimen sections (4 μm) were deparaffinised, rehydrated, and antigen retrieval was performed by microwave treatment at 650 W for 6 minutes in citrate buffer pH 6.0. After blocking of nonspecific binding, wound specimens were incubated for 2 hours with the anti-HIF1α antibody (1:5000) then with peroxidase-conjugated anti-rabbit antibody. Immunoreactivity was visualised with peroxidase reaction using aminoethylcarbazole, and specimens were counterstained with haematoxylin.

In vitro migration analysis and time-lapse videomicroscopy For Transwell migration, siRNA-transfected HaCaT cells were seeded in the upper chambers of Falcon Transwell inserts (8 μm size pores) in serum-free medium. Modified Eagle’s medium containing 10% fetal bovine serum was added to the lower compartment as chemoattractant. After 24 hours, cells on the lower side of the filter were fixed with 3% PFA and nuclei were stained with DAPI. Four fields of defined size were photographed and counted by using Image J software (http://rsb.info.nih.gov/ij/). For time-lapse experiments siRNA-transfected cells were wounded with a micropipette tip and different fields were filmed using a Zeiss videomicroscope then analysed using Metamorph software (Molecular Devices, Sunnyvale, CA).

We are very grateful to P. Pognonec for the plasmid encoding USF1. We also thank M. Lemonier for his help in the site-directed mutagenesis experiments and M. P. Simon for her advice concerning the ChIP experiments. We kindly thank Philippe Lenormand and Marta Jansà for their grammatical assistance. This work was financed by INSERM (Institut National de la Santé et la Recherche Médicale) and was carried out within the framework of an associated laboratory between the INSERM U634 and the Laboratory of Molecular and Cell Biology at the IDI-IRCCS (INSIDI project). Funding was also obtained from ARC (Association pour la Recherche contre le Cancer) grant n° 3903 and by the SFD (Societé Française de Dermatologie). G.F. was recipient of a PhD fellowship from the ARC association.

References Amano, S., Akutsu, N., Ogura, Y. and Nishiyama, T. (2004). Increase of laminin 5 synthesis in human keratinocytes by acute wound fluid, inflammatory cytokines and growth factors, and lysophospholipids. Br. J. Dermatol. 151, 961-970. Aumailley, M., Bruckner-Tuderman, L., Carter, W. G., Deutzmann, R., Edgar, D., Ekblom, P., Engel, J., Engvall, E., Hohenester, E., Jones, J. C. et al. (2005). A simplified laminin nomenclature. Matrix Biol. 24, 326-332. Bardos, J. I. and Ashcroft, M. (2005). Negative and positive regulation of HIF-1: a complex network. Biochim. Biophys. Acta 1755, 107-120. Berra, E., Benizri, E., Ginouves, A., Volmat, V., Roux, D. and Pouyssegur, J. (2003). HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state levels of HIF-1alpha in normoxia. EMBO J. 22, 4082-4090. Berra, E., Ginouves, A. and Pouyssegur, J. (2006). The hypoxia-inducible-factor hydroxylases bring fresh air into hypoxia signalling. EMBO Rep. 7, 41-45. Borradori, L. and Sonnenberg, A. (1999). Structure and function of hemidesmosomes: more than simple adhesion complexes. J. Invest. Dermatol. 112, 411-418. Boukamp, P., Petrussevska, R. T., Breitkreutz, D., Hornung, J., Markham, A. and Fusenig, N. E. (1988). Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. J. Cell Biol. 106, 761-771. Busca, R., Berra, E., Gaggioli, C., Khaled, M., Bille, K., Marchetti, B., Thyss, R., Fitsialos, G., Larribere, L., Bertolotto, C. et al. (2005). Hypoxia-inducible factor 1{alpha} is a new target of microphthalmia-associated transcription factor (MITF) in melanoma cells. J. Cell Biol. 170, 49-59. Carter, W. G., Ryan, M. C. and Gahr, P. J. (1991). Epiligrin, a new cell adhesion ligand for integrin alpha 3 beta 1 in epithelial basement membranes. Cell 65, 599-610. Chandrasekar, B., Mummidi, S., Valente, A. J., Patel, D. N., Bailey, S. R., Freeman, G. L., Hatano, M., Tokuhisa, T. and Jensen, L. E. (2005). The pro-atherogenic cytokine interleukin-18 induces CXCL16 expression in rat aortic smooth muscle cells via MyD88, interleukin-1 receptor-associated kinase, tumor necrosis factor receptor-associated factor 6, c-Src, phosphatidylinositol 3-kinase, Akt, c-Jun N-terminal kinase, and activator protein-1 signaling. J. Biol. Chem. 280, 26263-26277. Corley, K. M., Taylor, C. J. and Lilly, B. (2005). Hypoxia-inducible factor 1alpha modulates adhesion, migration, and FAK phosphorylation in vascular smooth muscle cells. J. Cell Biochem. 96, 971-985. Cowden Dahl, K. D., Robertson, S. E., Weaver, V. M. and Simon, M. C. (2005). Hypoxiainducible factor regulates alphavbeta3 integrin cell surface expression. Mol. Biol. Cell 16, 1901-1912. Denko, N. C., Fontana, L. A., Hudson, K. M., Sutphin, P. D., Raychaudhuri, S., Altman, R. and Giaccia, A. J. (2003). Investigating hypoxic tumor physiology through gene expression patterns. Oncogene. 22, 5907-5914. Dvorak, H. F. (1986). Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. N. Engl. J. Med. 315, 1650-1659. Elson, D. A., Ryan, H. E., Snow, J. W., Johnson, R. and Arbeit, J. M. (2000). Coordinate up-regulation of hypoxia inducible factor (HIF)-1alpha and HIF-1 target genes during multi-stage epidermal carcinogenesis and wound healing. Cancer Res. 60, 6189-6195. Fitsialos, G., Chassot, A. A., Turchi, L., Dayem, M. A., LeBrigand, K., Moreilhon, C., Meneguzzi, G., Busca, R., Mari, B., Barbry, P. et al. (2007). Transcriptional signature of epidermal keratinocytes subjected to in vitro scratch wounding reveals selective roles


HIF1 in cell migration and laminin expression for ERK1/2, p38, and phosphatidylinositol 3-kinase signaling pathways. J. Biol. Chem. 282, 15090-15102. Fukushima, Y., Ohnishi, T., Arita, N., Hayakawa, T. and Sekiguchi, K. (1998). Integrin alpha3beta1-mediated interaction with laminin-5 stimulates adhesion, migration and invasion of malignant glioma cells. Int. J. Cancer 76, 63-72. Gagnoux-Palacios, L., Allegra, M., Spirito, F., Pommeret, O., Romero, C., Ortonne, J. P. and Meneguzzi, G. (2001). The short arm of the laminin gamma2 chain plays a pivotal role in the incorporation of laminin 5 into the extracellular matrix and in cell adhesion. J. Cell Biol. 153, 835-850. Goldfinger, L. E., Stack, M. S. and Jones, J. C. (1998). Processing of laminin-5 and its functional consequences: role of plasmin and tissue-type plasminogen activator. J. Cell Biol. 141, 255-265. Hartwig, B., Borm, B., Schneider, H., Arin, M. J., Kirfel, G. and Herzog, V. (2007). Laminin-5-deficient human keratinocytes: defective adhesion results in a saltatory and inefficient mode of migration. Exp. Cell Res. 313, 1575-1587. Hunt, T. K., Niinikoski, J. and Zederfeldt, B. (1972). Role of oxygen in repair processes. Acta Chir. Scand. 138, 109-110. Kariya, Y. and Miyazaki, K. (2004). The basement membrane protein laminin-5 acts as a soluble cell motility factor. Exp Cell Res 297, 508-520. Katayama, M. and Sekiguchi, K. (2004). Laminin-5 in epithelial tumour invasion. J. Mol. Histol. 35, 277-286. Kim, C. H., Cho, Y. S., Chun, Y. S., Park, J. W. and Kim, M. S. (2002). Early expression of myocardial HIF-1alpha in response to mechanical stresses: regulation by stretchactivated channels and the phosphatidylinositol 3-kinase signaling pathway. Circ. Res. 90, E25-E33. Li, S., Huang, N. F. and Hsu, S. (2005). Mechanotransduction in endothelial cell migration. J. Cell Biochem. 96, 1110-1126. Marinkovich, M. P. (2007). Tumour microenvironment: laminin 332 in squamous-cell carcinoma. Nat. Rev. Cancer 7, 370-380. Maxwell, P. H., Wiesener, M. S., Chang, G. W., Clifford, S. C., Vaux, E. C., Cockman, M. E., Wykoff, C. C., Pugh, C. W., Maher, E. R. and Ratcliffe, P. J. (1999). The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399, 271-275. Maynard, M. A. and Ohh, M. (2007). The role of hypoxia-inducible factors in cancer. Cell. Mol. Life Sci. 64, 2170-2180. Miyazaki, K. (2006). Laminin-5 (laminin-332): Unique biological activity and role in tumor growth and invasion. Cancer Sci. 97, 91-98. Miyoshi, A., Kitajima, Y., Ide, T., Ohtaka, K., Nagasawa, H., Uto, Y., Hori, H. and Miyazaki, K. (2006). Hypoxia accelerates cancer invasion of hepatoma cells by upregulating MMP expression in an HIF-1alpha-independent manner. Int. J. Oncol. 29, 1533-1539. Murgia, C., Blaikie, P., Kim, N., Dans, M., Petrie, H. T. and Giancotti, F. G. (1998). Cell cycle and adhesion defects in mice carrying a targeted deletion of the integrin beta4 cytoplasmic domain. EMBO J. 17, 3940-3951. Nguyen, B. P., Ryan, M. C., Gil, S. G. and Carter, W. G. (2000). Deposition of laminin 5 in epidermal wounds regulates integrin signaling and adhesion. Curr. Opin. Cell Biol. 12, 554-562. Niinikoski, J., Grislis, G., Hunt, T. K., Hunt, T. K., Niinikoski, J. and Zederfeldt, B. (1972). Respiratory gas tensions and collagen in infected wounds: role of oxygen in repair processes. Ann. Surg. 175, 588-593. Nikolopoulos, S. N., Blaikie, P., Yoshioka, T., Guo, W., Puri, C., Tacchetti, C. and Giancotti, F. G. (2005). Targeted deletion of the integrin beta4 signaling domain suppresses laminin-5-dependent nuclear entry of mitogen-activated protein kinases and NF-kappaB, causing defects in epidermal growth and migration. Mol. Cell. Biol. 25, 6090-6102. Ohbayashi, N., Ikeda, O., Taira, N., Yamamoto, Y., Muromoto, R., Sekine, Y., Sugiyama, K., Honjoh, T. and Matsuda, T. (2007). LIF- and IL-6-induced acetylation of STAT3 at Lys-685 through PI3K/Akt activation. Biol. Pharm. Bull. 30, 1860-1864. O’Toole, E. A., Marinkovich, M. P., Peavey, C. L., Amieva, M. R., Furthmayr, H., Mustoe, T. A. and Woodley, D. T. (1997). Hypoxia increases human keratinocyte motility on connective tissue. J. Clin. Invest. 100, 2881-2891.

3001

Pennacchietti, S., Michieli, P., Galluzzo, M., Mazzone, M., Giordano, S. and Comoglio, P. M. (2003). Hypoxia promotes invasive growth by transcriptional activation of the met protooncogene. Cancer Cell 3, 347-361. Pognonec, P., Kato, H. and Roeder, R. G. (1992). The helix-loop-helix/leucine repeat transcription factor USF can be functionally regulated in a redox-dependent manner. J. Biol. Chem. 267, 24563-24567. Pouyssegur, J., Dayan, F. and Mazure, N. M. (2006). Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature 441, 437-443. Pyke, C., Salo, S., Ralfkiaer, E., Romer, J., Dano, K. and Tryggvason, K. (1995). Laminin-5 is a marker of invading cancer cells in some human carcinomas and is coexpressed with the receptor for urokinase plasminogen activator in budding cancer cells in colon adenocarcinomas. Cancer Res. 55, 4132-4139. Rheinwald, J. G. and Green, H. (1975). Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6, 331-343. Richard, D. E., Berra, E., Gothie, E., Roux, D. and Pouyssegur, J. (1999). p42/p44 mitogenactivated protein kinases phosphorylate hypoxia-inducible factor 1alpha (HIF-1alpha) and enhance the transcriptional activity of HIF-1. J. Biol. Chem. 274, 32631-32637. Rousselle, P. and Aumailley, M. (1994). Kalinin is more efficient than laminin in promoting adhesion of primary keratinocytes and some other epithelial cells and has a different requirement for integrin receptors. J. Cell Biol. 125, 205-214. Rousselle, P., Lunstrum, G. P., Keene, D. R. and Burgeson, R. E. (1991). Kalinin: an epithelium-specific basement membrane adhesion molecule that is a component of anchoring filaments. J. Cell Biol. 114, 567-576. Ryan, M. C., Tizard, R., VanDevanter, D. R. and Carter, W. G. (1994). Cloning of the LamA3 gene encoding the alpha 3 chain of the adhesive ligand epiligrin. Expression in wound repair. J. Biol. Chem. 269, 22779-22787. Schneider, H., Muhle, C. and Pacho, F. (2007). Biological function of laminin-5 and pathogenic impact of its deficiency. Eur. J. Cell Biol. 86, 701-717. Semenza, G. (2002). Signal transduction to hypoxia-inducible factor 1. Biochem. Pharmacol. 64, 993-998. Semenza, G. L. (1998). Hypoxia-inducible factor 1, master regulator of O2 homeostasis. Curr. Opin. Genet. Dev. 8, 588-594. Turchi, L., Chassot, A. A., Rezzonico, R., Yeow, K., Loubat, A., Ferrua, B., Lenegrate, G., Ortonne, J. P. and Ponzio, G. (2002). Dynamic characterization of the molecular events during in vitro epidermal wound healing. J. Invest. Dermatol. 119, 56-63. Turchi, L., Chassot, A. A., Bourget, I., Baldescchi, C., Ortonne, J. P., Meneguzzi, G., Lemichez, E. and Ponzio, G. (2003). Cross-talk between RhoGTPases and stress activated kinases for matrix metalloproteinase-9 induction in response to keratinocytes injury. J. Invest. Dermatol. 121, 1291-1300. Varghese, M. C., Balin, A. K., Carter, D. M. and Caldwell, D. (1986). Local environment of chronic wounds under synthetic dressings. Arch. Dermatol. 122, 52-57. Verrando, P., Hsi, B. L., Yeh, C. J., Pisani, A., Serieys, N. and Ortonne, J. P. (1987). Monoclonal antibody GB3, a new probe for the study of human basement membranes and hemidesmosomes. Exp. Cell Res. 170, 116-128. Vihanto, M. M., Plock, J., Erni, D., Frey, B. M., Frey, F. J. and Huynh-Do, U. (2005). Hypoxia up-regulates expression of Eph receptors and ephrins in mouse skin. FASEB J. 19, 1689-1691. Virolle, T., Coraux, C., Ferrigno, O., Cailleteau, L., Ortonne, J. P., Pognonec, P. and Aberdam, D. (2002). Binding of USF to a non-canonical E-box following stress results in a cell-specific derepression of the lama3 gene. Nucleic Acids Res. 30, 1789-1798. Warnecke, C., Griethe, W., Weidemann, A., Jurgensen, J. S., Willam, C., Bachmann, S., Ivashchenko, Y., Wagner, I., Frei, U., Wiesener, M. et al. (2003). Activation of the hypoxia-inducible factor-pathway and stimulation of angiogenesis by application of prolyl hydroxylase inhibitors. FASEB J. 17, 1186-1188. Wenger, R. H. (2002). Cellular adaptation to hypoxia: O2-sensing protein hydroxylases, hypoxia-inducible transcription factors, and O2-regulated gene expression. FASEB J. 16, 1151-1162. Yuan, Y., Hilliard, G., Ferguson, T. and Millhorn, D. E. (2003). Cobalt inhibits the interaction between hypoxia-inducible factor-alpha and von Hippel-Lindau protein by direct binding to hypoxia-inducible factor-alpha. J. Biol. Chem. 278, 15911-15916. Zagorska, A. and Dulak, J. (2004). HIF-1: the knowns and unknowns of hypoxia sensing. Acta Biochim. Pol. 51, 563-585.