Tumor Necrosis Factor-α-Induced Apoptosis in a Human Keratinocyte Cell Line (HaCaT) Is Counteracted by Transforming Growth Factor-α

EXPERIMENTAL CELL RESEARCH ARTICLE NO.

228, 334–340 (1996)

0333

Tumor Necrosis Factor-a-Induced Apoptosis in a Human Keratinocyte Cell Line (HaCaT) Is Counteracted by Transforming Growth Factor-a JEANNETTE REINARTZ, MICHAEL J. BECHTEL,

AND

MICHAEL D. KRAMER1

University Institute for Immunology, Laboratory for Immunopathology, Im Neuenheimer Feld 305, 69120 Heidelberg, Germany

The integrity of the human epidermis is guaranteed by a regulated balance of proliferation, differentiation, and physiologic cell death of its main cellular constituent, the epidermal keratinocyte. Physiologic cell death is known as apoptosis and has been recognized as an active regulatory mechanism, complementary to, but functionally opposite of, proliferation. The regulators of the delicate balance between cell death and proliferation are only partially understood in human keratinocytes. Transforming growth factor-a (TGF-a) has been identified as a positive regulator of proliferation and growth, while tumor necrosis factor-a (TNFa) induces apoptosis. Both mediators are thought to influence epidermal keratinocytes under various physiological and pathophysiological conditions. In the current study we have begun to investigate potential regulatory interactions between these two mediators in the human keratinocyte cell line HaCaT. We have found that, when the HaCaT cells were sensitized by the translation inhibitor cycloheximide, TNF-a induced apoptosis, as evidenced by nuclear disintegration, DNA fragmentation (‘‘DNA laddering’’), and the appearance of soluble DNA/histone complexes. Moreover, we found that the induction of apoptosis was reduced by preincubation of the cells with TGF-a. The protective effect of TGF-a was abrogated by translation inhibition, indicating that it depended on de novo protein synthesis. Moreover, the protective effect was not accompanied by a reduced surface expression of TNF receptor molecules. We postulate that TNF-a-induced apoptosis in HaCaT cells is counteracted by constitutively produced suppressors of apoptosis, the synthesis of which can be downregulated by inhibition of translation and upregulated by the cytokine TGF-a. q 1996 Academic Press, Inc.

INTRODUCTION

The human epidermis is a continuously renewing tissue. Its integrity is guaranteed by a regulated balance

of cell proliferation, cell differentiation, and cell death [1]. In recent years physiologic cell death, known as apoptosis, has been recognized as an active regulatory mechanism, complementary to, but functionally opposite of, proliferation. In human keratinocytes the regulators of the delicate balance between cell death and proliferation are only partially understood [2]. Transforming growth factor-a (TGF-a)2 induces growth and proliferation of human keratinocytes [for review see 3]. Its activity is transduced via a receptor that is shared with epidermal growth factor (EGF) and amphiregulin [for review see 4]. TGF-a is synthesized by keratinocytes in vitro, has been detected in the normal human epidermis [5, 6], and is elevated in the epidermis under certain pathophysiological conditions in vivo [7, 8], suggesting that it may be involved in the regulation of keratinocyte functions. On the other hand, apoptosis can be induced by cytokines, in particular tumor necrosis factor-a (TNF-a) [9]. It appears that the sensitivity of a given cell to TNF-a-induced apoptosis is not simply correlated with the number of TNF receptors (TNF-R) [10–12], but is rather dependent on cellular characteristics as yet not identified. An essential determinant of TNF-a activity on different cell types is the balance between apoptosisinducing and apoptosis-preventing factors, which are both apparently subject to regulation by exogenous mediators [for review see 1, 13], such as hormones and cytokines. Blocking the synthesis of protective proteins by translation inhibitors sensitizes many, but not all [14, 15], cell types to TNF-a-induced apoptosis. TNFa is locally released in skin wounds and inflammatory lesions [16, 17]; the inflammatory cell infiltrate is thought to be a major source of this cytokine [18]. TNFa may therefore influence epidermal keratinocytes in vivo. In the current study we have begun to investigate potential regulatory interactions between the apoptosis-in2

1

To whom correspondence and reprint requests should be addressed. Institut fu¨r Immunologie der Universita¨t Heidelberg, Laboratorium fu¨r Immunpathologie, Im Neuenheimer Feld 305, 69120 Heidelberg, Germany. Fax: (49)6221/564030.

334

0014-4827/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

AID

ECR 3325

/

6i15$$$$41

Abbreviations used: CHX, cycloheximide; HaCaT, human epidermal keratinocyte cell line; PAI-2, plasminogen activator inhibitor type-2; TGF-a, transforming growth factor-a; TNF-a, tumor necrosis factor-a; TNF-R, TNF receptor(s); EGF, epidermal growth factor; DAPI, 4*,6-diamidine-2*-phenylindole-dihydrochloride.

10-04-96 11:50:17

eca

AP: Exp Cell

335

TGF-a COUNTERACTS TNF-a-INDUCED APOPTOSIS

ducing mediator TNF-a and the growth-promoting cytokine TGF-a. Primary cultures of normal human keratinocytes are a heterogenous mixture of cells, containing a fraction of actively growing cells as well as a fraction of cells committed to terminal differentiation, which results in variable responses to regulators of proliferation and growth. We have therefore used the well-characterized cloned human keratinocyte cell line HaCaT for our studies. The HaCaT cells, established from keratinocytes of adult trunk skin [19], behave like normal keratinocytes in many biochemical and functional aspects studied so far [20]. As a first step we have determined the requirements for TNF-a-induced apoptosis in the HaCaT cell line. Moreover, we explored whether apoptosis is counteracted by TGF-a, a positive regulator of keratinocyte growth. Our results indicate that in HaCaT cells TNFa-induced apoptosis is counteracted by apoptosis-preventing factors (‘‘suppressors of apoptosis’’), the synthesis of which can be blocked by inhibition of translation. Moreover, these suppressors are apparently upregulated by the cytokine TGF-a. MATERIALS AND METHODS Materials. TNF-a (No. 1371843) was obtained from Boehringer (Mannheim, Germany) and TGF-a (No. ARM 17005) from Amersham (Braunschweig, Germany). The cell death detection ELISA (No. 1544 675) and 4*,6-diamidine-2*-phenylindole-dihydrochloride (DAPI; No. 236 276) were obtained from Boehringer. The monoclonal anti human TNF receptor antibody (No. 1995-01) was obtained from Genzyme (Ru¨sselsheim, Germany) and the isotype-identical monoclonal control antibody (No. 0571) from Immunotech (Marseille, France). The peroxidase-labeled goat anti-mouse IgG antibody (No. 115-035-071) was from Dianova (Hamburg, Germany). Buffer salts and detergents of analytical grade were from Merck (Darmstadt, Germany). DME medium and all medium supplements were from Seromed (Berlin, Germany). Cell culture and cytokine treatment. HaCaT cells were cultured at 377C in a humidified atmosphere of 7% CO2 using DME medium supplemented with 2 mM L-glutamine and 10% (v/v) heat-inactivated fetal calf serum. Confluent cultures were taken for the experiments. The cells were treated with different doses of TNF-a (0.1–20 ng/ml) and 10 mg cycloheximide/ml, with TNF-a alone, with cycloheximide alone, or left untreated for 24 h. Additionally, the cultures were preincubated with TGF-a for 24 h before the addition of TNF-a and cycloheximide. Viability and cytotoxicity assay. HaCaT cells were seeded in 96well plates at a density of 3 1 105 cells/ml. The cellular viability was determined by staining the cells with methylene blue. The keratinocytes were fixed with 2.5% (v/v) glutardialdehyde for 15 min at room temperature and subsequently washed with 0.1 M borate buffer, pH 8.5. The cells were then stained with methylene blue (1 mg/ml) in borate buffer for 15 min at room temperature. After repeated washes with deionized water, the plates were air-dried and the blue color was extracted from the cells by using 0.1 N HCl for 10 min. The intensity of the blue color is equivalent to the number of cells stained and can be quantified directly by absorbance measurement at 595 nm using an ELISA reader. DNA staining. HaCaT cells were seeded on glass coverslips at a density of 3 1 105 cells/ml. Twenty-four hours after the addition of TNF-a/cycloheximide, the cultures were washed with methanol and

AID

ECR 3325

/

6i15$$$$42

10-04-96 11:50:17

incubated with DAPI (1 mg/ml methanol; 15 min, 377C). After being washed with methanol, the cultures were mounted in PBS/glycerine. Nuclear morphology was assessed by fluorescence microscopy (Zeiss microscope; filter combination: BP 365/FT 395/LP 397). Photomicrographs were taken with an integrated camera and Kodak T-Max 400 ASA black and white film. DNA fragmentation analysis. HaCaT cells were grown in 24-well plates at a density of 3 1 105 cells/ml. Twenty-four hours after the addition of TNF-a/cycloheximide, the cells were detached, washed twice in PBS, resuspended in 1 ml lysis buffer (25 mM EDTA, 10 mM Tris, 100 mM NaCl, 0.5% SDS, and 1 mg proteinase K/ml), and incubated at 377C for 18 h. Samples were extracted twice with equal volumes of phenol/chloroform and incubated with RNase A (0.2 mg/ ml) for 1 h at 377C. Total DNA was ethanol-precipitated, resuspended in TE buffer, and electrophoresed (5 mg/lane) in 1% agarose gels containing ethidium bromide (1 mg/ml). The degree of chromosomal degradation was visualized by UV illumination. Alternatively, DNA fragmentation was measured by quantitation of cytosolic oligonucleosome-bound DNA by using a so-called cell death detection ELISA (Boehringer Mannheim, Germany) according to the manufacturer’s instructions. Briefly, the cytosolic fraction (15,000g supernatant) of approximately 103 HaCaT cells was used as antigen source in a sandwich ELISA with a primary anti-histone antibody coated to the microtiter plate and a secondary anti-DNA antibody coupled to peroxidase. From the absorbance values, the percentage of fragmentation in comparison to the controls was calculated. Cell ELISA for detection of the 55-kDa TNF-R. HaCaT cells were seeded in 96-well flat-bottom microtiter plates at a density of 0.5, 1, and 2 1 105 cells/well and incubated in the absence or presence of TGF-a (10 ng/ml) for 24 h. The cultured cells were then analyzed for the surface expression of the 55-kDa TNF-R in a so-called ‘‘cell ELISA’’: The cells were washed and fixed with 2% paraformaldehyde for 15 min at room temperature. Unspecific binding sites were blocked with 10% normal goat serum (NGS) in PBS. Afterward, the cells were incubated with the monoclonal anti-human TNF-R antibody (2 mg/ml in 5% NGS) or an isotype-identical monoclonal control antibody (2 mg/ml in 5% NGS). Subsequently, the cells were incubated with peroxidase-labeled goat anti-mouse IgG antibodies (1:5000 in 5% NGS). Bound antibodies were detected by incubation with o-phenylenediamine (1 mg/ml) and H2O2 (1 ml/ml). The reaction was stopped by adding 1.3 N H2SO4 and the reaction product was quantified by measuring absorbance at 492 nm.

RESULTS

TNF-a Is Cytotoxic for CHX-Sensitized HaCaT Cells We examined the cytotoxicity of TNF-a on HaCaT cells over a period of 24 h. Viability was never reduced by more than 13% compared to the medium control (Fig. 1). The data demonstrate that HaCaT cells are resistant to the direct cytotoxic effects of TNF-a. However, when the TNF-a treatment was performed in the presence of the protein synthesis inhibitor CHX, the cells were susceptible to TNF-a-induced toxicity (Fig. 1). The number of cells was reduced by 70% in the presence of 20 ng TNF-a / 10 mg CHX/ml. TNF-a Induces DNA Fragmentation in CHXSensitized HaCaT Cells Next, we asked whether cytolysis of HaCaT cells upon TNF-a/CHX treatment is due to apoptosis. Since

eca

AP: Exp Cell

336

REINARTZ, BECHTEL, AND KRAMER

FIG. 1. Cytotoxicity of TNF-a in CHX-sensitized HaCaT cells. The cells were incubated in DME medium in the presence of CHX (10 mg/ml), TNF-a (20 ng/ml), or a combination thereof (0.1–20 ng TNF-a/ml and 10 mg CHX/ml; closed bars) for 24 h. Afterward, the cells were fixed and stained by using methylene blue. Cell-associated methylene blue was extracted and quantified by absorbance measurement at 595 nm. The data shown are mean values and SD of four replicate determinations.

nuclear disintegration and degradation of chromosomal DNA are cardinal features of apoptosis, we analyzed the nuclear morphology and the DNA organiza-

tion of TNF-a/CHX-treated cells. When TNF-a/CHXtreated cells were examined by fluorescent microscopy after DAPI staining, an increase in the number of cells with nuclear condensation and fragmentation was apparent (nuclear ‘‘blebbing’’ (Fig. 2C) and nuclear disintegration (Fig. 2D)). In control cells incubated in the presence of TNF-a or CHX alone we observed a regular chromatin staining pattern characteristic of viable cells (Figs. 2A and 2B). To explore whether TNF-a/CHX treatment resulted in DNA fragmentation, we analyzed cellular DNA by agarose gel electrophoresis (Fig. 3). DNA obtained from TNF-a/CHX-treated cells was not randomly cleaved but, rather, showed a regularly spaced ladder pattern of fragmentation in multiples of nucleosome-sized (185 bp) subunits (Fig. 3, lane 4). Moreover, the DNA fragmentation caused by TNF-a/ CHX was concentration-dependent, as determined by an ELISA specific for cytosolic histone-bound DNA (Fig. 4). The determinations revealed an increase in cytosolic DNA/histone complexes in the presence of TNF-a (twice as much as in the medium (DMEM) control). This observation might be explained by the higher sensitivity of the ELISA in comparison to the determination of the cellular viability by methylene blue staining. Taken together, the data demonstrated that TNF-a/CHX-treated cells exhibited the known indicators of apoptosis.

FIG. 2. Nuclear fragmentation in CHX-sensitized HaCaT cells caused by TNF-a. Cells were seeded on coverslips and incubated in the presence of 20 ng TNF-a/ml (A), 10 mg CHX/ml (B), or a combination thereof (C and D) for 24 h. The cells were fixed, stained with DAPI, and analyzed by fluorescence microscopy. Arrowheads indicate nuclear blebbing (C) and nuclear fragmentation (D).

AID

ECR 3325

/

6i15$$$$42

10-04-96 11:50:17

eca

AP: Exp Cell

TGF-a COUNTERACTS TNF-a-INDUCED APOPTOSIS

FIG. 3. DNA fragmentation in CHX-sensitized HaCaT cells caused by TNF-a. An example of agarose gel electrophoresis and ethidium bromide staining of DNA extracted from cells cultured in medium (lane 1) or in the presence of TNF-a (20 ng/ml; lane 2), CHX (10 mg/ml; lane 3), or a combination thereof (lane 4). S, DNA molecular weight marker III (Boehringer Mannheim).

TGF-a Counteracts TNF-a/CHX-Induced Apoptosis We asked whether TNF-a/CHX-induced apoptosis could be prevented by pretreatment of the HaCaT cells with TGF-a. HaCaT cells were preincubated with different concentrations of TGF-a for 24 h and then exposed to TNF-a and CHX. TGF-a pretreatment caused a concentration-dependent resistance to TNF-a-mediated apoptosis: the TNF-a/CHX-induced appearance of cytosolic DNA/histone complexes, as measured in the ELISA, was É60% lower in cells pretreated with TGFa (10 ng/ml) than in cells precultured in the absence

FIG. 4. Concentration-dependent induction of DNA fragmentation in CHX-sensitized HaCaT cells by TNF-a. The cells were incubated in DME medium in the presence of CHX (10 mg/ml), TNF-a (20 ng/ml), or a combination thereof (0.1–20 ng TNF-a/ml and 10 mg CHX/ml; closed bars) for 24 h. DNA fragmentation was determined by quantifying the amount of oligonucleosome-bound DNA in the 15,000g supernatant of cell lysates. The data shown are mean values and SD of four replicate determinations.

AID

ECR 3325

/

6i15$$$$43

10-04-96 11:50:17

337

FIG. 5. Pretreatment of the HaCaT cells with TGF-a reduces TNF-a/CHX-induced DNA fragmentation. HaCaT cells were grown in graded concentrations of TGF-a (0–10 ng/ml; closed bars) for 24 h. Afterward, the cells were exposed to CHX (10 mg/ml) and TNF-a (20 ng/ml). Control cells were incubated in medium, CHX alone (10 mg/ml), or TNF-a alone (20 ng/ml). DNA fragmentation was determined by quantifying the amount of oligonucleosome-bound DNA in the 15,000g supernatant of cell lysates. The data shown are mean values and SD of four replicate determinations.

of TGF-a. The resistance-inducing effect of TGF-a was dose-dependent (Fig. 5). The Protective Effect of TGF-a Requires Protein Synthesis and Is Not Caused by the Reduced Surface Expression of the 55-kDa TNF-a Receptor To explore whether the protective activity of TGF-a required active protein synthesis, HaCaT cells were preincubated with TGF-a in the absence or presence of the protein synthesis inhibitor CHX. Compared to the medium control, CHX decreased the number of cells that survived the TNF-a/CHX treatment (Fig. 6). In contrast, TGF-a caused an increase in the number of cells that survived the TNF-a/CHX treatment. The increase was counteracted when the TGF-a pretreatment was performed in the presence of CHX. The findings indicated that the protective effect of TGF-a depends on the de novo synthesis of protective proteins. In separate experiments we found that HaCaT cells—like normal human epidermal keratinocytes [12]—expressed the 55-kDa TNF-R, but not the 75kDa TNF-R (data not shown). We asked whether the protective effect of TGF-a was caused by a reduced surface expression of the 55-kDa TNF-R. HaCaT cells were seeded at three different densities (0.5, 1, and 2 1 105 cells/well) in 96-well plates and incubated in the absence or presence of TGF-a (10 ng/ml) for 24 h. The cultures were then analyzed by a so-called cell ELISA using specific monoclonal antibodies against the 55kDa TNF-R. The relative levels of surface-expressed

eca

AP: Exp Cell

338

REINARTZ, BECHTEL, AND KRAMER

TNF-R varied among different cell densities: the highest level of surface-expressed TNF-R was observed at the lowest cell density (Fig. 7). However, cells preincubated in the absence (Fig. 7, open bars) or presence (Fig. 7, closed bars) of TGF-a displayed similar levels of TNF-R expression. DISCUSSION

Apoptosis is a mechanism of physiological cell death that serves to delete cells from the body during both development and homeostasis [21, 22]. In the human epidermis apoptosis appears as an important mechanism for controlling the size of its major cell population, the epidermal keratinocytes [1, 2]. Apoptosis is characterized by the breakdown of cells into membrane-surrounded apoptotic bodies containing intact cytoplasmic organelles and nuclear fragments [23]. At the biochemical level the fragmentation of cellular DNA into oligonucleosome-sized particles is a characteristic hallmark [24]. Studies of exogenous modulation of apoptosis suggest that this process can be induced by hormones and cytokines [9, 25, 26]. For most cell types, it remains to be explored whether apoptosis results from the induction of proapoptotic intracellular effector molecules, the activation of a preexisting latent death program, or the elimination of intracellular suppressors of apoptosis [27, 28]. In the present study we have analyzed the requirements for TNF-a-induced apoptosis in the human kera-

FIG. 6. The protective effect of TGF-a against TNF-a/CHX treatment depends on active protein synthesis. HaCaT cells were preincubated in medium alone (DMEM) or in the presence of CHX (10 mg/ ml), TGF-a (10 ng/ml), or a combination thereof (10 ng TGF-a/ml and 10 mg CHX/ml; closed bars) for 24 h. Afterward, the cells were exposed to the combination of TNF-a/CHX (20 ng TNF-a/ml and 10 mg CHX/ml) for 24 h. Then, the cells were fixed and stained by methylene blue. Cell-associated methylene blue was extracted and quantified by absorbance measurement at 595 nm. The data shown are mean values and SD of four replicate determinations.

AID

ECR 3325

/

6i15$$$$43

10-04-96 11:50:17

FIG. 7. The protective effect of TGF-a against TNF-a/CHX treatment is not due to a reduction of TNF-R surface expression. HaCaT cells were seeded in 96-well flat-bottom microtiter plates at a density of 0.5, 1, and 2 1 105 cells/well and incubated in the absence (open bars) or presence (closed bars) of TGF-a (10 ng/ml) for 24 h. The cultured cells were then analyzed for the surface expression of the 55-kDa TNF-R by using specific monoclonal antibodies in a so-called ‘‘cell ELISA’’ (for details see Materials and Methods). Afterward, the cell number in the microtiter plates was determined by methylene blue staining and a standard curve derived from HaCaT cells that had been seeded at defined concentrations into 96-well flat-bottom microtiter plates. The level of TNF-R expression was normalized to the level in HaCaT cells seeded at a density of 1 1 105 cells/well and cultured in the absence of TGF-a (Å1). The data shown are mean values and SD of four replicate determinations.

tinocyte line HaCaT. It appeared that HaCaT cells are inherently resistant to the apoptosis-inducing effect of TNF-a. However, apoptosis could be induced when the HaCaT cells were simultaneously exposed to the protein synthesis inhibitor CHX. The finding suggests that the production of intracellular proteins inhibiting cell death was blocked by CHX, which then allowed TNF-ainduced apoptosis to proceed. Under in vivo conditions translational arrest could result from metabolic, toxic, or immunological damage to the epidermal keratinocytes. Thus, a combination of TNF-a with such environmental hazards could induce apoptosis in epidermal keratinocytes. Furthermore, the finding that CHX promotes apoptosis indicates that effector molecules responsible for apoptosis were unaffected by translation inhibition and could be set into operation by TNF-a. This is in contrast to hepatocytes in which TNF-a-inducible DNA fragmentation depended on protein synthesis, as concluded from the protective effect of cycloheximide [29]. TNF-a exerts its biological activities via binding to specific cell surface receptors present on almost all cells analyzed, including keratinocytes [12]. Two distinct TNF-R molecules with molecular masses of 55 and 75 kDa, respectively, have been identified and cloned [for

eca

AP: Exp Cell

TGF-a COUNTERACTS TNF-a-INDUCED APOPTOSIS

339

review see 30]. Studies on the roles of these two recep- HaCaT cell line. Moreover, it has recently been demtors in the fibroblast cell line L929 have shown that onstrated that epidermal keratinocytes can be driven TNF cytotoxicity, as in most other cell lines, is medi- into programmed cell death by ultraviolet-B irradiaated via the 55-kDa receptor [31]. In separate experi- tion [40] or a combination of interferon-g and antiments we found that HaCaT cells—like normal epider- Fas antibodies [41, 42]. Most notably, UV-B-induced mal keratinocytes [12]—express the 55-kDa but not apoptosis was partially inhibitable by anti-TNF-a anthe 75-kDa TNF-R. Thus, in HaCaT cells as well as in tibodies, indicating that this cytokine is partially innormal keratinocytes the effects of TNF-a are appar- volved in this type of apoptosis induction [40]. In view ently mediated via the 55-kDa TNF-R. of our present findings it is tempting to speculate that TNF-a is cytostatic for normal keratinocytes and UV-B irradiation would provide two signals that tomay thus play a role in inhibiting epidermal prolifera- gether lead to apoptosis: first, the downregulation of tion [3]. The mediator induces a wide variety of genes apoptosis-preventing factors by inhibition of cellular [30, 32–34], including the proto-oncogene c-myc [35]. protein synthesis and second, the induction of cell This is of particular interest, since the MYC protein death via release of TNF-a. This matter, however, has been implicated in the control of both cell prolifera- must await further detailed studies. tion and apoptosis [36, 37]. However, the intracellular When considering the significance of our findings it processes following receptor–ligand interactions that is important to ask whether and to what extent HaCaT lead to TNF-a-induced inhibition of cell growth and cells behave like normal keratinocytes. The HaCaT cell apoptosis remain to be explored. line has been established from a culture of trunk keraTGF-a is present in normal keratinocytes and has tinocytes and has retained a relatively high capacity of both structural and functional homology to EGF. It epidermal differentiation [19]. Apparently, HaCaT binds to the same cell surface receptor as EGF and cells duplicate the proliferating cell pool of a normal amphiregulin (EGF receptor) [4]. The EGF receptor has epidermis, as indicated, e.g., by the expression of basal been localized in normal human skin to basal and para- keratins (K5/K14) and the adhesion molecule ICAM-1, basal keratinocytes. The collective data suggest that which in the normal epidermis are both restricted to EGF receptor ligands may play a complex role in nor- the basal proliferating compartment [12, 20]. Theremal skin growth, development, and function. In the fore, we postulate that the regulatory interactions elulesional epidermis of psoriasis, a hyperproliferative cidated in HaCaT cells may apply to the proliferating epidermal disorder, TGF-a is overexpressed and EGF subpopulation of normal epidermal keratinocytes. Of receptors are present in all viable layers. Thus, the course one important difference between normal epiTGF-a/EGF receptor system is a candidate for the initi- dermal keratinocytes and HaCaT cells is that the latter ation or maintenance of epidermal hyperproliferation are immortalized and have gained the capacity for auin this disease [38]. tonomous growth in vitro. Whether this is linked with In the present report we have demonstrated that pre- the constitutive expression of apoptosis-preventing factreatment of HaCaT cells with TGF-a renders the cells tors is tempting to speculate on but clearly must await less susceptible to TNF-a/CHX-induced apoptosis (Fig. further experimentation. 5). Basically this finding indicates that the HaCaT The identity of intracellular suppressors of apoptosis cell’s phenotype is altered either by decreased synthe- is beginning to be elucidated. There are several candisis of an apoptosis-promoting factor(s) or by an in- date molecules, such as mitochondrial superoxide discreased synthesis of apoptosis-preventing factors. We mutase [43], the proto-oncogene bcl-2 [22, 28], and a favor the latter hypothesis, since TGF-a treatment of proteinase inhibitor, the so-called plasminogen activathe HaCaT cells in the presence of CHX did not de- tor inhibitor type-2 (PAI-2) [44, 45]. With respect to crease but increased their sensitivity to TNF-a/CHX- epidermal keratinocytes the latter is of particular ininduced apoptosis (Fig. 6). Moreover, the protective ef- terest, since PAI-2 is expressed in vivo in the normal fect of TGF-a is apparently not due to a reduced surface epidermis in a zone that is associated with keratinocyte expression of the 55-kDa TNF-R (Fig. 7). Although the differentiation and apoptosis [46]. Moreover, PAI-2 is level of TNF-R surface expression varied between Ha- observed in the lesional epidermis of psoriasis vulgaris CaT cells at different densities, there was no difference [46] and lupus erythematodes [47], epidermal diseases between untreated and TGF-a-treated cells. The fact which are both associated with an increase in the epithat surface expression of TNF-R was higher at lower dermal keratinocyte population. At present we are excell densities is reminiscent of previous similar findings ploring which of these known protective factors is subwith the epithelial tumor cell line HeLa [39]. ject to regulation by TGF-a and to what extent the Basically we have shown that apoptosis can be mod- respective factor(s) is involved in the regulation of ulated in a human keratinocyte line by factors, i.e. apoptosis in HaCaT cells. Understanding the mechaTNF-a and TGF-a, that may act in vivo also on epider- nisms that regulate apoptosis and identifying individmal keratinocytes, the physiological ancestors of the ual control points of the apoptotic pathway in keratino-

AID

ECR 3325

/

6i15$$$$43

10-04-96 11:50:17

eca

AP: Exp Cell

340

REINARTZ, BECHTEL, AND KRAMER

cytes will likely have an impact on future strategies for therapeutic intervention in epidermal diseases. The authors are indebted to Ms. K. Neumann and Ms. S. Jobstmann for expert technical assistance. This work has been financially supported by the Deutsche Forschungsgemeinschaft (Kr 931/3-1) and the Forschungszentrum Karlsruhe (PUG U 95001).

REFERENCES 1. Haake, A. R., and Polakowska, R. R. (1993) J. Invest. Dermatol. 101, 107–112. 2. Polakowska, R. R., Piacentini, M., Bartlett, R., Goldsmith, L. A., and Haake, A. R. (1994) Dev. Dyn. 199, 176–188. 3. Stoof, T. J., Boorsma, D. M., and Nickoloff, B. J. (1994) in The Keratinocyte Handbook (Leigh, I. M., Lane, E. B., and Watt, F. M., Eds.), pp. 365–399, Cambridge Univ. Press, Cambridge, UK. 4. Massague´, J. (1990) J. Biol. Chem. 265, 21393–21396. 5. Coffey, R. J., Derynck, R., Wilcox, J. N., Bringman, T. S., Goustin, A. S., Moses, H. L., and Pittelkow, M. R. (1987) Nature 328, 817–820. 6. Gottlieb, A. B., Chang, C. K., Posnett, D. N., Fanelli, B., and Tam, J. P. (1988) J. Exp. Med. 167, 670–675. 7. Elder, J. T., Klein, S. B., Tavakkol, A., Fisher, G. J., Nickoloff, B. J., and Voorhees, J. J. (1990) J. Invest. Dermatol. 95, 7S–9S. 8. Finzi, E., Harkins, R., and Horn, T. (1991) J. Invest. Dermatol. 96, 328–332. 9. Laster, S. M., Wood, J. G., and Gooding, L. R. (1988) J. Immunol. 141, 2629–2634. 10. Brockhaus, M., Schoenfeld, H. J., Schlaeger, E. J., Hunziker, W., Lesslauer, W., and Loetscher, H. (1990) Proc. Natl. Acad. Sci. USA 87, 3127–3131. 11. Dembic, Z., Loetscher, H., Gubler, U., Pan, Y. C., Lahm, H. W., Gentz, R., Brockhaus, M., and Lesslauer, W. (1990) Cytokine 2, 231–237. 12. Trefzer, U., Brockhaus, M., Loetscher, H., Parlow, F., Kapp, A., Schopf, E., and Krutmann, J. (1991) J. Invest. Dermatol. 97, 911–916. 13. Sen, S. (1992) Biol. Rev. Camb. Philos. Soc. 67, 287–319. 14. Ruggiero, V., Latham, K., and Baglioni, C. (1987) J. Immunol. 138, 2711–2717. 15. Woods, K. M., and Chapes, S. K. (1993) J. Leuk. Biol. 53, 37– 44. 16. Luger, T. A., and Schwarz, T. (1990) J. Invest. Dermatol. 95, 100S–104S. 17. Barker, J. N. W. N., Mitra, R. S., Griffiths, C. E. M., Dixit, V. M., and Nickoloff, B. J. (1991) Lancet 337, 211–214. 18. Riches, D. W. H. (1994) in The Molecular and Cellular Biology of Wound Repair (Clark, R. A. F. and Henson, P. M., Eds.), pp. 213–239, Plenum, New York. 19. Boukamp, P., Petrusevska, R. T., Breitkreuz, D., Hornung, J., Markham, A., and Fusenig, N. E. (1988) J. Cell Biol. 106, 761– 771.

20. Breitkreutz, D., Boukamp, P., Ryle, C. M., Stark, H. J., Roop, D. R., and Fusenig, N. E. (1991) Cancer Res. 51, 4402–4409. 21. Schwartzman, R. a., and Cidlowski, J. A. (1993) Endocrine Rev. 14, 133–151. 22. Wyllie, A. H. (1993) Br. J. Cancer 67, 205–208. 23. Ellis, R. E., Yuan, J. Y., and Horvitz, H. R. (1991) Annu. Rev. Cell Biol. 7, 663–698. 24. Duke, R. C., Chervenak, R., and Cohen, J. J. (1983) Proc. Natl. Acad. Sci. USA 80, 6361–6365. 25. Wyllie, A. H., Kerr, J. F., and Currie, A. R. (1980) Int. Rev. Cytol. 68, 251–306. 26. Colombel, M., Olsson, C. A., Ng, P. Y., and Buttyan, R. (1992) Cancer Res. 52, 4313–4319. 27. Gerschenson, L. E., and Rotello, R. J. (1992) FASEB J. 6, 2450– 2455. 28. Vaux, D. L. (1993) Proc. Natl. Acad. Sci. USA 90, 786–789. 29. Leist, M., Gantner, F., Bohlinger, I., Germann, P. G., Tiegs, G., and Wendel, A. (1994) J. Immunol. 153, 1778–1788. 30. Beyaert, R., and Fiers, W. (1994) FEBS Lett. 340, 9–16. 31. Vanhaesebroeck, B., Decoster, E., Van Ostade, X., Van Bladel, S., Lenaerts, A., Van Roy, F., and Fiers, W. (1992) J. Immunol. 148, 2785–2794. 32. Dixit, V. M., Green, S., Sarma, V., Holzman, L. B., Wolf, F. W., O’Rourke, K., Ward, P. A., Prochownik, E. V., and Marks, R. M. (1990) J. Biol. Chem. 265, 2973–2978. 33. Sarma, V., Wolf, F. W., Marks, R. M., Shows, T. B., and Dixit, V. M. (1992) J. Immunol. 148, 3302–3312. 34. Camussi, G., Albano, E., Tetta, C., and Bussolino, F. (1991) Eur. J. Biochem. 202, 3–14. 35. Manchester, K. M., Heston, W. D., and Donner, D. B. (1993) J. Biochem. 290, 185–190. 36. Askew, D. S., Ashmun, R. A., Simmons, B. C., and Cleveland, J. L. (1991) Oncogene 6, 1915–1922. 37. Shi, Y., Glynn, J. M., Guilbert, L. H., Cotter, T. G., Bissonnette, R. P., and Green, D. R. (1992) Science 257, 212–214. 38. Nickoloff, B. J., Mitra, R. S., Elder, J. T., Fisher, G. J., and Voorhees, J. J. (1989) Br. J. Dermatol. 121, 161–174. 39. Pocsik, E., Mihalik, R., Ali, O. F., and Aggarwal, B. B. (1994) J. Cell Biochem. 54, 453–464. 40. Schwarz, A., Bhardwaj, R., Aragane, Y., Mahnke, K., Riemann, H., Metze, D., Luger, T. A., and Schwarz, T. (1995) J. Invest. Dermatol. 104, 922–927. 41. Takahashi, H., Kobayashi, H., Hashimoto, Y., Matsuo, S., and Iizuka, H. (1995) J. Invest. Dermatol. 105, 810–815. 42. Sayama, K., Yonehara, S., Watanabe, Y., and Miki, Y. (1994) J. Invest. Dermatol. 103, 330–334. 43. Wong, G. H. (1995) Biochim. Biophys. Acta 1271, 205–209. 44. Kumar, S., and Baglioni, C. (1991) J. Biol. Chem. 266, 20960– 20964. 45. Dickinson, J. L., Bates, E. J., Ferrante, A., and Antalis, T. M. (1995) J. Biol. Chem. 270, 27894–27904. 46. Lyons-Giordano, B., Loskutoff, D., Chen, C.-S., Lazarus, G., Keeton, M., and Jensen, P. J. (1994) Histochemistry 101, 105–112. 47. Bechtel, M. J., Schaefer, B. M., and Kramer, M. D. (1996) Br. J. Dermatol. 134, 411–419.

Received May 22, 1996 Revised version received August 6, 1996

AID

ECR 3325

/

6i15$$$$43

10-04-96 11:50:17

eca

AP: Exp Cell