Apoptosis signal-regulating kinase 1 (ASK1) is an intracellular inducer of keratinocyte differentiation.

Cells differentiate in response to various extracellular stimuli. This cellular response requires intracellular signaling pathways. The mitogen-activated protein (MAP) kinase cascade is a core signal transduction pathway that determines the fate of many kinds of cell. MAP kinase kinase kinase activates MAP kinase kinase, which in turn activates MAP kinase. Apoptosis signal-regulating kinase (ASK1) was identified as a MAP kinase kinase kinase involved in the stress-induced apoptosis-signaling cascade that activates the SEK1-JNK and MKK3/MKK6-p38 MAP kinase cascades. Expression of the constitutively active form of ASK1 (ASK1-DeltaN) in keratinocytes induced significant morphological changes and differentiation markers, transglutaminase-1, loricrin, and involucrin. A transient increase in p21(Cip1/WAF1) reduced DNA synthesis, and cell cycle analysis verified the differentiation. p38 MAP kinase inhibitors, SB202190 and SB203580, abolished the induction of differentiation markers, transglutaminase-1, loricrin, and involucrin. In turn, the induction of differentiation with ceramide in keratinocytes caused an increase in ASK1 expression and activity. Furthermore, normal human skin expresses ASK1 protein in the upper epidermis, implicating ASK1 in in vivo keratinocyte differentiation. We propose that the ASK1-p38 MAP kinase cascade is a new intracellular regulator of keratinocyte differentiation.

Reverse Transcriptase-PCR Analysis-The epidermis was separated from normal human skin by incubation in phosphate-buffered saline at 60°C for 1 min, immediately followed by immersion in ice-cold phosphate-buffered saline. Total RNA was prepared with Isogen and treated with 50 units/ml DNase 1 (CLONTECH Laboratories, Inc., Palo Alto, CA) at 37°C for 30 min. Specific primers for ASK1 was produced by selecting specific nucleotide sequences from previously published sequences (5). The reverse transcriptase-PCR was performed using RT-PCR High Plus (Toyobo Co., Ltd., Osaka, Japan). The sequences of the primer pair, product size, annealing temperature, and number of cycles were as follows: 5Ј-TGACAGAGTCGTTTTAGGAA-3Ј and 3Ј-ACAAGC-AAGTCGTTAGCACA-5Ј, 759 base pairs, 57°C, and 25 cycles. The PCR products were sequenced to confirm the mRNA expression.
RNase Protection Assay-Analysis was performed using the multiprobe RNase protection assay system (PharMingen Co.) according to the manufacturer's instructions. Oligonucleotide probes were prepared by inserting the PCR-amplified human cDNA corresponding to oligonucleotides 2659 -2874 of ASK1 (Genbank TM /EBI accession number D84476), 866 -1133 of transglutaminase-1 (Genbank TM /EBI accession number D90287), 966 -1176 of loricrin (Genbank TM /EBI accession number M61120), and 74 -249 of involucrin (Genbank TM /EBI accession number M13903) into the EcoRI and HindIII sites of pPMG vector. 5 g of total RNA were hybridized with 32 P-labeled riboprobe and digested with RNase. The hybridization products were separated on a 5% polyacrylamide/8 M urea gel and exposed to film. GAPDH is shown as an internal standard. The intensity of each band was quantified using NIH Image, referring to the signal of the control as one unit.
Luciferase Assay-A reporter plasmid containing the involucrin promoter and firefly luciferase was constructed (pINV-Luc) as follows. The involucrin promoter cassette was a generous gift from Dr. Taichman (12). The coding region of firefly luciferase was digested from pGL3 basic (Promega Co., Madison, WI) and subcloned into the involucrin promoter cassette. The correct insertion and orientation were confirmed by sequencing. To normalize the transfection efficiency, a plasmid containing Renilla luciferase driven by herpes simplex virus thymidine kinase promoter (pRL-TK Promega Co.) was included in the assay. The reporter plasmids were introduced into the keratinocytes using Fu-GENE6 (Roche Molecular Biochemicals) according to the manufacturer's instructions. In each transfection, 1 g of pINV-Luc and 0.5 g of pRL-TK were introduced into 2 ϫ 10 5 keratinocytes in 6-well plates. After 24 h, the cells were infected with the indicated Ad at an MOI of 5 and were incubated for an additional 24 h. Then the cells were harvested with 250 l of lysis buffer (Promega Co.), and luciferase activity was measured using the Dual-Luciferase reporter assay system (Promega Co.) with a luminometer (Luminescencer JNR AB-2100; Atto Co., Osaka, Japan). Transfection was performed in triplicate. The relative luciferase activity was calculated by normalizing to the Renilla luciferase activity. Statistical analysis was performed using Student's t test.
MAP Kinase Activity-MAP kinase activity was measured with JNK and p38 kinase assay kits (BioLabs Inc., Beverly, MA). The lysate of 1 ϫ 10 6 keratinocytes was immunoprecipitated with a 1:100 dilution of rabbit antibody to p38 MAP kinase and protein G-Sepharose (Amersham Pharmacia Biotech). The resulting immunoprecipitate was then incubated with ATF-2 fusion protein at 30°C for 30 min in the presence of 200 M ATP. JNK was precipitated from the cell lysates with c-Jun fusion protein bound to glutathione-Sepharose beads and incubated with 100 M ATP at 30°C for 30 min. The phosphorylation of ATF-2 at Thr-71 and c-Jun at Ser-63 was detected by Western blotting using a 1:100 dilution of phosphospecific ATF-2 or c-Jun. To show that equal amounts of JNK and p38 MAP kinase were precipitated, the beads were incubated with SDS sample buffer at 97°C for 3 min and then subjected to Western blot analysis using a 1:1000 dilution of antibody to JNK and p38 MAP kinase (Santa Cruz Biotechnology, Inc., Santa Cruz, CA).
Immune Complex-coupled Kinase Assay for ASK1-The immune complex-coupled kinase assay has been described previously (7). In brief, cells were lysed in the lysis buffer containing 20 mM Tris-HCl, pH 7.5, 12 mM ␤-glycerophosphate, 150 mM NaCl, 5 mM EGTA, 10 mM NaF, 1% Triton X-100, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, and 1.5% aprotinin. The lysates of 1 ϫ 10 6 cells were immunoprecipitated with a 1:1000 dilution of anti-ASK1 (DAV, Ref. 10) using protein A-Sepharose. The beads were washed with washing buffer containing 150 mM NaCl, 20 mM Tris-HCl, pH 7.5, 5 mM EGTA, and 1 mM dithiothreitol and was subjected to kinase assay. GST⅐MKK6 (0.2 g) was first incubated with the immune complex for 10 min at 30°C in a final volume of 10 l in a solution containing 20 mM Tris-HCl, pH 7.5, 20 mM MgCl 2 , and 100 M ATP. Thereafter, the activated complex was incubated with 0.3 Ci of [␥-32 P]ATP and 1 g of GST⅐p38 ␥KN in the same solution (final volume 20 l) for 10 min at room temperature. Kinase reactions were stopped by adding SDS sample buffer and were analyzed by SDS-polyacrylamide gel electrophoresis under reducing conditions. Phosphorylation of GST⅐p38 ␥KN was analyzed using a Fuji BAS2000 image analyzer.
Cell Sorter Analysis-Involucrin-positive cells were analyzed with a cell sorter. Keratinocytes were harvested from the dishes with trypsin and fixed with 3.7% formaldehyde at room temperature for 8 min with methanol at Ϫ20°C for 4 min and then with acetone at Ϫ20°C for 2 min (13). After washing with Tris-buffered saline, pH 7.4, containing 0.2% Tween 20, the cells were reacted with a 1:100 dilution of rabbit antihuman involucrin antibody (Biomedical Technologies, Inc.) and a 1:100 dilution of fluorescein isothiocyanate-conjugated goat anti-rabbit antibody. The labeled cells were analyzed with a flow cytometer (Becton Dickinson Co.).
The cell cycle distribution was analyzed using a CycleTEST PLUS DNA reagent kit (Becton Dickinson Immunocytometry Systems) according to the manufacturer's instructions. Nuclei isolated by trypsinization were stained with propidium iodide and then run on a flow cytometer (Becton Dickinson Co.).
TdT-mediated dUTP Nick End Labeling (TUNEL)-Apoptotic cells were stained on chamber slides with an in situ cell death detection kit (Roche Molecular Biochemicals GmbH) according to the manufacturer's instructions. After treatment with 4% paraformaldehyde and 0.1% Triton X in sodium citrate, the cells were incubated with fluoresceinlabeled nucleotides and terminal deoxynucleotidyl transferase for 1 h at 37°C. Fluorescent specimens were observed by fluorescence microscopy.
BrdUrd Incorporation-BrdUrd incorporation was assessed with a cell proliferation kit (Amersham Pharmacia Biotech) according to the manufacturer's instructions. Cells were incubated with BrdUrd in thymidine-free culture medium for 60 min at 37°C. BrdUrd incorporated into cellular DNA was stained with monoclonal anti-BrdUrd diluted 1:100, peroxidase anti-mouse IgG diluted 1:70, and 3,3Ј-diaminobenzidine. Two hundred cells were counted in randomly selected fields to quantify the positive cells. Each experiment was repeated three times.

Induction of Keratinocyte Differentiation with ASK1-⌬N-Ad
carrying ASK1-⌬N (Ad-ASK1-⌬N) was constructed as described previously (14). In this study, Ad expressing a bacterial ␤-galactosidase gene (Ad-␤-gal) and no exogenous gene (Ad-1W) were used as controls to exclude the effect of Ad itself. Gene expression was found in almost all of the keratinocytes with Ad (data not shown). We infected normal human keratinocytes with Ad-ASK1-⌬N or control Ad at an MOI of 5 or 50. As expected from our previous reports, infection of Ad-ASK1-⌬N but not Ad-␤-gal at a higher MOI (50) strongly induced apoptosis as determined by morphology and TUNEL staining (Fig. 1). Surprisingly however, infection of Ad-ASK1-⌬N at a lower level (MOI of 5) induced dramatic morphological changes without any sign of apoptotic phenotypes. The cells became enlarged and flattened, showing a differentiated phenotype 48 h after infection with Ad-ASK1-⌬N. There were no morphological changes in keratinocytes infected with Ad-␤-gal and Ad-1W at an MOI of 5 compared with no-vector (data not shown). Because, the morphology is apparently different from apoptotic cells, and the TUNEL staining was negative, we conclude that the morphological change induced by the infection of Ad-ASK1-⌬N at an MOI of 5 is not apoptosis. In the following experiments, keratinocytes were infected with Ad at an MOI of 5.
In addition to the morphological changes, ASK1-⌬N expression induced differentiation markers including involucrin protein and transglutaminase-1 mRNA ( Fig. 2A) as seen with 10% FCS, a potent inducer of keratinocyte differentiation (15). Transglutaminase-1 enzymatically cross-links its substrate proteins including involucrin and loricrin, forming a cornified envelope in terminally differentiated keratinocytes (16). Similar results were obtained with three keratinocyte strains (data not shown).
To further confirm that Ad-ASK1-⌬N infection at an MOI of 5 induces keratinocyte differentiation, the time course for protein and mRNA expression, the percentage of involucrin-positive cells, 5-bromo-2Ј-deoxyuridine (BrdUrd) incorporation and the cell cycle were analyzed (Figs. 2 and 3). ASK1-⌬N protein appeared within 3 h of infection with Ad-ASK1-⌬N, reaching a maximum level at 48 h, which lasted until 72 h (Fig. 2B). The expression of mRNA of differentiation markers including transglutaminase-1, loricrin, and involucrin significantly increased at 24 h (Fig. 2C). The increase of p21 Cip1/WAF1 , the cyclin-dependent kinase inhibitor, occurred 6 h after infection, which was earlier than the transglutaminase-1, loricrin, and involucrin induction (Fig. 2, B and C), and reached a maximum at 24 h. The transient increase of p21 Cip1/WAF1 with differentiation is consistent with a previous report (17,18); p21 Cip1/WAF1 expression is up-regulated in the early stage of keratinocyte differentiation and then declines in the late stages of differentiation. ASK1-⌬N expression increased the percentage of involucrin-positive cells from 5.5 to 29.1% (Fig. 2D), which is quite similar to that seen with 10% FCS (13). Cell growth is suppressed in differentiated cells. ASK1-⌬N expression suppressed the percentage of BrdUrd-positive cells from 38.7 (control) to 13.8% 48 h after the infection (Fig. 3A), indicating that cell growth declined. The cell cycle distribution of the keratinocytes 48 h after Ad infection (Fig. 3B) indicates that ASK1-⌬N expression causes G 0 /G 1 and G 2 /M arrest. Differentiated epidermal keratinocytes in vivo are in G 0 /G 1 arrest (19). However, in cultured keratinocytes the induction of differentiation by suspension culture and FCS results in G 0 /G 1 and G 2 /M arrest (15), which is consistent with our results. G 0 /G 1 and G 2 /M arrest by differentiation stimuli suggests that in cultured keratinocytes differentiation signals are not restricted to the cell cycle stage.
To investigate the role of ASK1 in the regulation of involucrin expression, a luciferase assay was performed. Keratinocytes transfected with an involucrin promoter-luciferase reporter plasmid (pINV-Luc) were infected with Ad at an MOI of 5. The reporter activity increased 5.5-fold with Ad-ASK1-⌬N, whereas Ad-␤-gal had no effect (Fig. 4), suggesting a positive role of ASK1 in the regulation of involucrin expression. All of these data (Figs. 2-4) verify that ASK1-⌬N differentiates keratinocytes.
Involvement of p38 MAP Kinase in ASK1-induced Keratinocyte Differentiation-ASK1 is a MAPKKK that activates SEK1-JNK and MKK3/MKK6-p38 MAP kinase cascades. Therefore, we examined whether the expression of ASK1-⌬N enhanced the JNK and p38 MAP kinase activities in keratinocytes. The activities of both JNK and p38 MAP kinase started to increase 6 -12 h after Ad-ASK1-⌬N infection (Fig. 5A). This increase paralleled the level of ASK1-⌬N protein shown in Fig. 2A, suggesting that ASK1-⌬N activates the SEK1-JNK and MKK3/ MKK6-p38 MAP kinase cascades in keratinocytes.
The involvement of p38 MAP kinase in ASK1-⌬N-induced differentiation was shown using p38 MAP kinase inhibitors: SB202190 and SB203580 (Fig. 5B). They significantly reduced the induction levels of transglutaminase-1, loricrin, and involucrin mRNA mediated by ASK1-⌬N but not the negative control SB202474. In SB202190-treated cells, the levels of transglutaminase-1, loricrin, and involucrin mRNA declined to 0.13, 0.02, and 0.38-fold compared with the controls, respectively. Because the inhibitors did not affect the level of ASK1-⌬N protein expression, the suppressive effect of p38 MAP kinase inhibitors on the induction of transglutaminase-1 and involucrin mRNA was caused by the blockade of p38 MAP kinase. Therefore, p38 MAP kinase is necessary for the downstream signal transduction of ASK1-induced differentiation.
Protein kinase C (PKC) is an intracellular signal transduction molecule that regulates keratinocyte differentiation (20,21), and epidermal keratinocytes express ␣, ␦, ⑀, , and isoforms of PKC (20, 28 -31). We examined whether PKC isoforms were involved in the ASK1 signaling cascade. Activation of PKC isoforms was determined by analyzing the subcellular distribution of PKC isoforms using Western blotting. The re-

FIG. 1. The morphological change of keratinocytes with Ad-ASK1-⌬N.
After infection with Ad-ASK1-⌬N or Ad-␤-gal at an MOI of 5 or 50, normal human keratinocytes were cultured for 48 h. The morphological change was observed under phase-contrast microscopy. Apoptotic cells were stained with TUNEL. distribution of PKC from the soluble fraction to a particulate fraction is a useful indicator of PKC activation (32). However, the expression of ASK1-⌬N did not affect the subcellular distribution of PKCs (data not shown), indicating that PKCs were not activated by ASK1-⌬N. Therefore, PKCs are not localized in the downstream signaling of the ASK1-p38 MAP kinase cascade. A recent study showed that PKC regulates involucrin FIG. 4. Activation of the involucrin promoter by ASK1-⌬N. Keratinocytes cultured to 2 ϫ 10 5 in 6-well plates were transfected with 1 g of involucrin reporter plasmid (pINV-Luc) and 0.5 g of pRL-TK (Renilla luciferase) as a standard, using FuGENE6. After 24 h, the cells were infected with Ad-ASK1-⌬N or Ad-␤-gal at an MOI of 5 and cultured for an additional 24 h. Luciferase activity was measured using the Dual-Luciferase reporter assay system. Transfection was performed in triplicate. The relative luciferase activity was calculated by normalizing to the Renilla luciferase activity. Statistical analysis was performed using Student's t test. * statistically significant (p Ͻ 0.0001). promoter activity via p38 MAP kinase (33) as well as ASK1. Therefore, the ASK1-p38 MAP kinase cascade may cross-talk with the PKC cascade at the level of p38 MAP kinase. NF-B is another candidate for a molecular regulator of keratinocyte differentiation (34 -37). The epidermis of mice lacking the inhibitor of B kinase ␣ (IKK␣) shows abnormal differentiation (35). In normal epidermis, NF-B is localized in the cytoplasm of basal cells and then translocates to the nucleus in suprabasal cells, suggesting a possible role for commitment to differentiate (34). In contrast to NF-B, ASK1 protein is found in the upper epidermis and has a different distribution from that of NF-B. Furthermore, the expression of ASK1-⌬N does not activate NF-B as analyzed with a gel shift mobility assay (data not shown), indicating that NF-B is not involved in the ASK1-induced keratinocyte differentiation. One suggested role for NF-B is that it prevents premature apoptosis before the final step by regulating the expression of the antiapoptotic molecules TRAF1, TRAF2, c-IAP1, and c-IAP2 (37).
On the other hand, ASK1 is an apoptosis inducer and appears in the late stages of differentiation in vivo. Therefore, NF-B and ASK1 have distinct roles in the regulation of keratinocyte differentiation.
Although, ASK1 has been identified as an apoptosis inducer (5), we have shown that ASK1 induces keratinocyte differentiation. We also found that introducing ASK1 induced apoptosis. However, it required 10-fold higher (MOI of 50) levels of ASK1-⌬N expression than those required for differentiation (MOI of 5). These results suggest that ASK1 has dual physiological functions in keratinocytes; weak or strong activation of ASK1 may lead to differentiation or apoptosis of keratinocytes, FIG. 5. Involvement of p38 MAP kinase in ASK1-induced differentiation. A, the time course for the JNK and p38 MAP kinase activities after infecting keratinocytes with Ad-ASK1-⌬N. The JNK and p38 MAP kinase activities are shown as phospho-c-Jun and phospho-ATF-2, respectively. Keratinocytes were infected with Ad-ASK1-⌬N or Ad-␤-gal at an MOI of 5 and cultured for the indicated period. p38 MAP kinase was immunoprecipitated with a rabbit antibody to p38 MAP kinase and protein G-Sepharose. The resulting immunoprecipitate was then incubated with ATF-2 fusion protein at 30°C for 30 min in the presence of 200 M ATP. JNK was precipitated from the cell lysates with c-Jun fusion protein bound to glutathione-Sepharose beads and incubated with 100 M ATP at 30°C for 30 min. The phosphorylation of ATF-2 at Thr-71 and c-Jun at Ser-63 was detected by Western blotting using phospho-specific ATF-2 or c-Jun. Equal amounts of JNK and p38 MAP kinase were precipitated as shown by Western blot analysis using antibody to JNK and p38 MAP kinase. B, effect of p38 MAP kinase inhibitors on ASK1-induced differentiation. After infection with Ad-ASK1-⌬N at an MOI of 5, keratinocytes were cultured for 24 h in the presence of p38 MAP kinase inhibitors, SB202190 and SB203580 at a concentration of 200 M. Transglutaminase-1, loricrin, and involucrin mRNA and ASK1 protein were analyzed by RNase protection assay and Western blotting, respectively. SB202474 is a negative control. The intensity of each band was quantified using NIH Image, referring to the signal of the control as one unit. respectively. The biological activity of ASK1 depends on the cell type and conditions. In the pheochromocytoma cell line PC12, moderate expression of ASK1-⌬N induces neuronal differentiation and survival rather than apoptosis (38). Thus, ASK1 is not only an apoptosis inducer but also mediates a wide range of cellular functions.
In continuously self-renewing tissues, such as the gastrointestinal tract, epithelial cells are shed by terminal differentiation and apoptosis. Epidermal keratinocytes also differentiate and are ultimately shed from the epidermis after cell death. However, the morphology of cells dying in terminal differentiation is different from that of cells undergoing pathological apoptosis induced by ultraviolet light and Fas. This physiological cell death is suggested to be a specialized form of apoptosis (39). In ASK1-induced apoptosis, involucrin and transglutaminase-1 are strongly induced before apoptosis (MOI of 50, data not shown), whereas ultraviolet B irradiation does not enhance the expression of transglutaminase-1 mRNA (data not shown). Therefore, the two apoptosis mechanisms are different. Because ASK1 induces apoptosis with differentiation markers, and its expression is localized in the upper epidermis, ASK1 may be involved in the mechanism of apoptosis in terminal differentiation.