- Email: [email protected]
Differential expression of transforming growth factor-α (TGF-α) and EGF receptor in transitional area of psoriatic epidermis
JOURMLOF
Dermatological Science Journal
of Dermatological
Science 7 (1994) 45-53
Differential expression of transforming growth factor-a (TGF-(x) and EGF receptor in transitional area of psoriatic epidermis Mari Higashiyama
*, Koji Hashimoto, Kunio Kunihiko Yoshikawa
Matsumoto,
Osaka Universitv School of Medicine, 2-2 Yamadagaoka, Suita-shi. Osaka 565, Japan (Received
15 July 1993: revision
received 17 August 1993; accepted 6 September 1993)
Abstract We investigated the localization of TGF-(Y and EGF receptor in psoriatic plaques, especially the clinicallydetermined transitional zone from uninvolved to involved psoriatic skin by immunohistochemistry. TGF-(Y was not increased in the transitional area compared with normal epidermis. It started to increase within the edge of psoriatic plaques. In contrast, EGF receptor increased up to the suprabasal layers even in the uninvolved skin adjacent to psoriatic plaques, compared with normal epidermis. Slight epidermal hyperplasia was also observed in the clinicallydetermined transitional area. This result suggests that the increase of EGF receptor precedes that of TGF-(r in psoriatic
plaque formation. Key words: TGF-CY; EGF receptor;
Psoriasis;
Epidermal
1. Introduction Transforming growth factor-a (TGF-a!) is a 50 amino acid peptide which is structurally homologous to EGF [ 1,2]. TGF-a binds to EGF receptor to exert its biological effects [3]. TGF-CY plays an important role in the regulation of keratinocyte function. It stimulates human keratinocyte growth and migration [4]. Interestingly, keratinocytes synthesize TGF-(Y, and addition of EGF or TGF-(r induces TGF-(r gene expression in keratinocyte * Corresponding
author.
hyperplasia
culture [5]. This suggests that an auto-induction mechanism of TGF-CY production exists in keratinocytes. TGF-a is also involved in the pathogenesis of hyperproliferative skin diseases such as psoriasis. Psoriasis is a common non-malignant skin disease showing epidermal hyperplasia. Overexpression of TGF-a in psoriatic epidermis has been reported [6-S]. We reported that TGF-a increased in psoriatic epidermis by about 5-fold more than normal epidermis [9]. However, to understand its role in vivo, it is essential to determine the localization of TGF-CL Since TGF-CY exerts its effects through the EGF receptor, the relationship be-
0 1994 Elsevier Science Ireland Ltd. All rights reserved. 0923-181 l/94/$07.00 SSDI 0923- I8 I 1(93)00252-V
46 tween TGF-CY and EGF receptor localization in psoriatic lesion and uninvolved adjacent skin might give us a clue to characterize the role of TGF-a in epidermal proliferation. In this study we showed that the increase of EGF receptor precedes that of TGF-(r in the edge of psoriatic plaques and uninvolved adjacent epidermis. 2. Materials and methods 2.1. Immunohistological staining The biopsies of typical psoriatic plaques and adjacent uninvolved skin were taken from five psoriatic patients who had not received treatment. These psoriatic lesions were supposed to be expanding ones since all the plaques got worse 1 week later by evaluation of size, thickness and scaling. Five normal skin samples were obtained by plastic reconstructive surgery. Skin samples were embedded in OTC compound with liquid nitrogen. Chicken monospecific anti-human TGF-(Y antibody (IgG: 15 mg/ml) was raised against recombinant human TGF-(Y and purified by TGF-(r affinity column. This antibody was kindly provided by Dr Joseph Delarco (Otsuka Pharmaceutical Co. Japan). In order to verify the immunoreactive specificity, preabsorbed antibody was used as a control. Briefly, anti-TGF-cr antibody (150 &ml) was mixed with recombinant TGF-a (100 pg/ml) at 37°C for 1 h and then incubated at 4°C overnight. After centrifugation at 10 000 x g, the supernatant was collected and used as TGF-a! preabsorbed antibody. Five-micrometer sections were placed on poly+lysine coated slides, and fixed with 4% paraformaldehyde-PBS for 10 min. The avidin-biotin peroxidase technique was used for detection of TGF-a. The sections were rinsed briefly with 0.5% Tween-PBS which was also used for further washing, and they were incubated with 10% normal goat serum for 30 min at room temperature. After a brief rinse, the sections were incubated with chicken anti-human TGF-CY antibody diluted in PBS/BSA (1: 1000) for 30 min at room temperature. The control sections were incubated with an equivalent amount of non-specific chicken IgG. After a brief rinse, the sections were incubated with a 1:250 dilution of biotinylated goat anti-chicken IgG (Vector Labs, Burlington, CA)
M. Hipshiynma et ul. /J. Drm~arol.Sci. 7 (1994) 45-53 for 30 min at room temperature, followed by extensive washing. Then, the tissue sections were incubated with avidin-peroxidase complex (Vecta Stain Elite kit, Vector Labs, Burlington, CA) for 30 min. The specific peroxidase substrate solution (3,3’-diaminobenzidine tetrahydrochloride with hydrogen peroxide) was added. The sections were washed with distilled water and counterstained in methyl green. The same procedure was performed for EGF receptor staining. Anti-EGF receptor antibody (Oncogene Science) at a 1:25 dilution in PBWBSA and biotinylated horse anti-mouse IgG (Vector Labs, Burlington, CA) were used as the first and second antibodies, respectively. 2.2. Western blotting Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed accor-
Fig. I. Recombinant TGF-or (lanes 1 and 3) and recombinant EGF (lane 2) were separated using 15% polyacrylamide gel. After SDS-PAGE. proteins were transferred onto nitrocellulose membrane. Lanes 1 and 2 were incubated with chicken anti-TGF-o antibody. and lane 3 was incubated with normal chicken IgG. Chicken anti-TGF-cy antibody reacted with TGFo. but not with EGF.
M. Higashiyama
et al. /J.
Dermatoi.
Sri.
7 (1994)
45-53
47
a
b
Fig. 2. Comparison of normal skin sections incubated with antibody to TGF-(Y (a), and preabsorbed antibody to TGF-cu (b) and antibody to EGF receptor(c). (a.b) In the epidermis. TGF-(Y was stained intensely through the upper several layers including the granular layer. In contrast. the basal cell layer showed no positive staining. In the dermis endothelial cells showed positive staining. Ant+TGF(Yantibody preabsorbed with TGF-(Y showed no positive staining. (c) EGF receptor was observed mainly in basal cells in the epidermis of normal skin as previously reported.
ding to the method of Laemmfi [lo]. Recombinant TGF-CY and EGF (Earth Pharmaceutical Co. Japan) were separated using 15% polyacrylamide gel. After SDS-PAGE, proteins were transferred
onto nitrocellulose membrane (Schleicher & Schuell, Dassel, Germany) by the method of Towbin et al. [l 11. TGF-CY was immunochemically detected using chicken anti-TGF-cr antibody, goat biotiny-
M. Higashi.vama et al. /J. Dermatol. Sri. 7 (1994) 4.5-53
Fig. 2. (continued)
lated anti-chicken IgG (Vector Labs, Burlington, CA) and peroxidase-conjugated avidin-biotin complex (Vector Labs, Burlington, CA). 3. Results To confirm the specificity of chicken anti-TGFcx antibody, the reactivity of this antibody was determined by Western blot. As shown in Fig. 1, this antibody reacted with TGF-a, but not with EGF. To investigate TGF-a localization in psoriatic skin, we first tested the immunoreactivity of this antibody with normal human skin. In the epidermis of normal skin, TGF-a! staining was intense through the upper several layers including the granular layer (Fig. 2a). In contrast, the basal cell layer did not show positive staining. Linear or dot staining pattern of TGF-a was observed mainly in the intercellular space, suggesting its localization on cell surface membrane. In the dermis, endothelial cells, eccrine duct and gland cells, and sebace-
ous glands showed positive staining. To further demonstrate that the immunostaining was specific for TGF-a, anti-TGF-cy antibody was preabsorbed with TGF-a as described in Section 2 above (Materials and methods). The preabsorbed antibody showed no positive staining (Fig. 2b). EGF receptor was observed mainly in the basal cell layer as previously reported (Fig. 2c) [ 121. It was localized on the cellular membrane also. Next, we examined the localization and distribution of TGF-(r and EGF receptor in the clinicallydetermined transitional area from uninvolved to involved skin as well as typical psoriatic plaques. In involved psoriatic epidermis TGF-a! was detected throughout the epidermis with the exception of the basal cell layer (Fig. 3a). The upper epidermis was stained more intensely than the lower epidermis in involved psoriatic skin. This observation was consistent in all five psoriatic plaques. In uninvolved skin adjacent to psoriatic plaque, the immunostaining pattern of TGF-a! was similar to that of normal epidermis (Fig. 3b). EGF
M. Higashiyama et al. /J.
Dermarol. Sci. 7 (1994) 45-53
49
a
b
epidermis from uninvolved to involved psoriatic skin. Clinical margin is shown by an Fig. 3 Expression I of TGF-ol in transitional of the involved psoriatic epidermis showed the presence of TGF-ol throughout the epidermis with the arrow (a) High m agnification except ion of the bmasal cell layer. The upper epidermis was stained more intensely than the lower epidermis. (b) High magnilicat ion zone showed the immunostaining pattern of TGF-a similar to normal epidermis. (c) Low magnification shot Ned of the transitional the whole skin.
50
M. Higashiyama et al. /J.
Dermatol. Sci. 7 (1994) 45-53
Fig. 3. (continued)
receptor was detected up to the upper epidermis in involved psoriatic skin (Fig. 4a) as previously reported [12]. The focal intensity of anti-EGF receptor staining was prominent in the basal cell layer and lower portion of rete ridge, and significantly reduced in the upper portion of epidermis. This pattern was in clear contrast to that of TGF-CL In the clinically-determined transitional area from uninvolved to involved shown by the arrows (Fig. 3b,c), TGF-(r staining was not increased although the increase of EGF receptor in the basal and suprabasal layer was clearly detected (Fig. 4b,c). Slight epidermal hyperplasia was also observed in the clinically-determined transitional area (Figs. 3b and 4b). The increase of TGF-a staining was observed within the edge of psoriatic plaques. It should be noted that EGF receptor increased even in the uninvolved area adjacent to psoriatic plaques (Fig. 4b). Taken together, the increase of EGF receptor expression preceded that of TGF-(r in the epidermis of psoriatic lesion. 4. Discussion This study aimed to elucidate the localization of TGF-(r in normal and psoriatic epidermis in comparison with that of EGF receptor. Since TGF-a exerts its effect through binding to EGF receptor,
the relationship between the localization of TGF-a and EGF receptor is important to understand their roles in the skin. Our data demonstrated that EGF receptor increased to the mid-layer even in the adjacent uninvolved and transitional epidermis of psoriatic plaques, preceding the increase of TGF(Y. The same phenomenon was observed in the wound healing model [ 131. Therefore, the increase of EGF receptor preceding that of TGF-c~ could be a general phenomenon in epidermal hyperplasia. Coffey et al. (1987) reported the autocrine mechanism of TGF-a in human keratinocyte growth [5], in which the binding of TGF-a to EGF receptor enhances further production of TGF-CL Our observation may indicate the necessity of preceding increase of EGF receptor for autocrine production of TGF-CX, resulting in epidermal hyperplasia. Several groups have reported the immunohistochemical localization of TGF-(r in epidermis. However, the results are not consistent with each other. One group showed cytoplasmic staining [14], while other workers reported membranous staining [7]. There are further discrepancies in reports on TGF-a! distribution in epidermis: basalar TGF-(Y immunostaining in normal epidermis and whole epidermal staining in psoriatic skin [7], immunostaining throughout normal epidermis and intense suprabasalar but not basalar immuno-
M. Higashiyama el al. /J.
Dermatol. Sri. 7 (1994) 4.5-53
51
a
Fig. 4. Expression of EGF receptor in transitional epidermis from uninvolved to involved psoriatic skin. Clin ical margin is shown by an arrow. (a) High magnification of the involved psoriatic skin the presence of EGF receptor throughout the thickened epi dermis. (b) High magnification of the transitional zone showed the appearance of EGF receptor in the suprabasal cell layers differel nt from normal epidermis. (c) Low magnification showed the whole skin.
52
Fig. 4. (continued)
reactivity in psoriatic epidermis [14], intense subcornea1 and faint basalar TGF-(Y immunostaining in psoriatic epidermis, and only basalar staining in normal epidermis [8]. We observed predominantly membranous staining pattern of TGF-(r, and subcornea1 and suprabasal immunoreactivity in both normal and psoriatic epidermis, although psoriatic epidermis showed more intense immunostaining. These discrepancies may reflect the difference of antibodies or immunostaining methods such as fixation. The result of in situ hybridization study also seems at odds [8]. The localization of TGF-(Y mRNA was reported to be in the subcorneal layer of psoriatic epidermis. However, the present study showed that protein expression was deeper into the epidermis of psoriatic lesions. The distribution of TGF-a and EGF receptor were different. In normal epidermis, EGF receptor was predominantly located in the basal layer, whereas TGF-a! was detected mainly in the upper epidermis, and not detected in the basal layer. Since TGF-a is supposed to facilitate keratinocyte growth, the coexistence of TGF-a and EGF receptor seems likely. However, our observation is inconsistent with this hypothesis. Recently, Finzi et al. (1992) reported that stronger immunostaining of TGF-(II in differentiated cells of appendageal tumors and the association of TGF-a with squamous cell differentiation [ 151. Furthermore, the presence of TGF-(r mRNA was not confirmed in the basal layer by in situ hybridization. In this
case, some other EGF-like growth factor such as amphiregulin [ 161 could be involved, mainly in epidermal keratinocyte growth in vivo. Recently, these investigators demonstrated that amphiregulin mRNA is elevated in psoriatic epidermis compared with non-lesional epidermis in normal skin [17]. Another possibility is that TGFQ could be synthesized but released rapidly from the cell surface after cleavage of the extracellular domain by elastase-like proteinase [18]. The antibody used in this study was raised against the recombinant mature TGF-(r and recognizes the mature form only. Therefore, it is possible that TGF-ol is synthesized much more in the basal layer than the upper epidermis, but released more quickly from the cell membrane. Two groups reported that non-cleavable pro-TGF-a is less potent than soluble TGF-(r as an activator of EGF receptor [19,20]. This observation may support the second possibility. In this case, the C-terminal portion of pro-TGF-cr remains through the plasma membrane. The antibody reacting with the C-terminal of pro-TGF-a could give us the answer to this important question.
5. Acknowledgement This work was supported by Grant-in Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan 03454274.
hf. Higashiyama et al. /J. 6.
Dermatol. Sri. 7 (1994) 45-53
53
References I
2
3
4
5
6
7
8
9
10
Marquardt H, Hunkapiller MW, Hood LE et al.: Transforming growth factors produced by retrovirus transformed rodent libroblasts and human melanoma cells: amino acid sequence homology with epidermal growth factor. Proc Nat1 Acad Sci 80: 4684-4688, 1983. Massagut J: Epidennal growth factor-like transforming growth factor. II. Interactions with epidermal growth factors in human placenta membranes and A43 I cells. J Biol Chem 258: 13614-13620, 1983. Carpenter G, Stoscheck CM, Preston YA et al.: Antibodies to the epidermal growth factor block the biological activities of sarcoma growth factor. Proc Nat1 Acad Sci 80: 5627-5630, 1983. Barrandon Y, Green H: Cell migration is essential for sustained growth of keratinocyte colonies: the roles of transforming growth factor-o and epideimal growth factor. Cell 50: 1131-1137, 1987. Coffey RJ, Derynck R, Wilcox JN et al.: Production and auto-induction of transforming growth factor in human keratinocytes. Nature 328: 817-820, 1987. Elder JT. Fisher GJ, Lindquist PB et al.: Overexpression of transforming growth factor-o in psoriatic epidermis, Science 243: 811-814, 1989. Gottlieb AB, Chang CK, Posnett DN et al.: Detection of transforming growth factor-o in normal, malignant and hyperproliferative human keratinocytes. J Exp Med 167: 670-675, 1988. Turbitt ML, Akhurst RJ, White SI et al.: Localization of elevated transforming growth factor-alpha in psoriatic epidermis. J Invest Dermatol 95: 229-232, 1990. Higashiyama M, Matsumoto K. Hashimoto K et al.: Increased production of transforming growth factor-o in psoriatic epidermis. J Dermatol 18: 117-l 19, 1991. Laemmli UK: Cleavage of structural proteins during the
II
12
13
14
15
16
17
18 19
20
assembly of the head of bacteriophage T4. Nature 227: 1970. 680-685, Towbin H, Staehelin T, Gordon J: Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Nat1 Acad Sci 76: 4350-4354, 1979. Nanney LB, Stoscheck CM, Magid M et al.: Altered (“‘I) epidermal growth factor binding and receptor distribution in psoriasis: J Invest Dermatol 86: 260-265. 1986. Stoscheck CM, Nanney LB. King Jr., LB: Quantitative determination of EGF-R during epidermal wound healing. J Invest Dermatol 99: 645-649. 1992. Finzi E, Harkins R, Horn T: TGF-or is widely expressed in differentiated as well as hyperproliferative skin epitheliurn. J Invest Dermatol 96: 328-333, 1991. Finzi E, Ho T, Anhalt G, Hawkins W et al.: Localization of transforming growth factor-o in human appendageal tumors. Am J Path01 141: 643-653. 1992. Cook PW. Mattox PA, Keeble WW et al.: A heparin sulfate-regulated human keratinocyte autocrine factor is similar or identical to amphiregulin. Mol Cell Biol 11: 2547-2557, 1991. Cook PW, Pittelkow MR. Keeble WN et al.: Amphiregulin mRNA is elevated in psoriatic epidermis and gastrointestinal carcinomas. Cancer Res 52: 3224-3227, 1992. Derynck R: Transforming growth factor-o. Cell 54: 593-595, 1988. Sharon TW, Lisa FW, Bryan KM et al.: The TGF-(Y precursor expressed on the cell surface binds to the EGF receptor on adjacent cells, leading to signal transduction. Cell 56: 495-506, 1989. Rainer B, Patricia BL, Nagashima M et al.: Transmembrane TGF-(U precursors activate EGF/TGF-(Y receptors. Cell 56: 691-700, 1989.
Dermatological Science Journal
of Dermatological
Science 7 (1994) 45-53
Differential expression of transforming growth factor-a (TGF-(x) and EGF receptor in transitional area of psoriatic epidermis Mari Higashiyama
*, Koji Hashimoto, Kunio Kunihiko Yoshikawa
Matsumoto,
Osaka Universitv School of Medicine, 2-2 Yamadagaoka, Suita-shi. Osaka 565, Japan (Received
15 July 1993: revision
received 17 August 1993; accepted 6 September 1993)
Abstract We investigated the localization of TGF-(Y and EGF receptor in psoriatic plaques, especially the clinicallydetermined transitional zone from uninvolved to involved psoriatic skin by immunohistochemistry. TGF-(Y was not increased in the transitional area compared with normal epidermis. It started to increase within the edge of psoriatic plaques. In contrast, EGF receptor increased up to the suprabasal layers even in the uninvolved skin adjacent to psoriatic plaques, compared with normal epidermis. Slight epidermal hyperplasia was also observed in the clinicallydetermined transitional area. This result suggests that the increase of EGF receptor precedes that of TGF-(r in psoriatic
plaque formation. Key words: TGF-CY; EGF receptor;
Psoriasis;
Epidermal
1. Introduction Transforming growth factor-a (TGF-a!) is a 50 amino acid peptide which is structurally homologous to EGF [ 1,2]. TGF-a binds to EGF receptor to exert its biological effects [3]. TGF-CY plays an important role in the regulation of keratinocyte function. It stimulates human keratinocyte growth and migration [4]. Interestingly, keratinocytes synthesize TGF-(Y, and addition of EGF or TGF-(r induces TGF-(r gene expression in keratinocyte * Corresponding
author.
hyperplasia
culture [5]. This suggests that an auto-induction mechanism of TGF-CY production exists in keratinocytes. TGF-a is also involved in the pathogenesis of hyperproliferative skin diseases such as psoriasis. Psoriasis is a common non-malignant skin disease showing epidermal hyperplasia. Overexpression of TGF-a in psoriatic epidermis has been reported [6-S]. We reported that TGF-a increased in psoriatic epidermis by about 5-fold more than normal epidermis [9]. However, to understand its role in vivo, it is essential to determine the localization of TGF-CL Since TGF-CY exerts its effects through the EGF receptor, the relationship be-
0 1994 Elsevier Science Ireland Ltd. All rights reserved. 0923-181 l/94/$07.00 SSDI 0923- I8 I 1(93)00252-V
46 tween TGF-CY and EGF receptor localization in psoriatic lesion and uninvolved adjacent skin might give us a clue to characterize the role of TGF-a in epidermal proliferation. In this study we showed that the increase of EGF receptor precedes that of TGF-(r in the edge of psoriatic plaques and uninvolved adjacent epidermis. 2. Materials and methods 2.1. Immunohistological staining The biopsies of typical psoriatic plaques and adjacent uninvolved skin were taken from five psoriatic patients who had not received treatment. These psoriatic lesions were supposed to be expanding ones since all the plaques got worse 1 week later by evaluation of size, thickness and scaling. Five normal skin samples were obtained by plastic reconstructive surgery. Skin samples were embedded in OTC compound with liquid nitrogen. Chicken monospecific anti-human TGF-(Y antibody (IgG: 15 mg/ml) was raised against recombinant human TGF-(Y and purified by TGF-(r affinity column. This antibody was kindly provided by Dr Joseph Delarco (Otsuka Pharmaceutical Co. Japan). In order to verify the immunoreactive specificity, preabsorbed antibody was used as a control. Briefly, anti-TGF-cr antibody (150 &ml) was mixed with recombinant TGF-a (100 pg/ml) at 37°C for 1 h and then incubated at 4°C overnight. After centrifugation at 10 000 x g, the supernatant was collected and used as TGF-a! preabsorbed antibody. Five-micrometer sections were placed on poly+lysine coated slides, and fixed with 4% paraformaldehyde-PBS for 10 min. The avidin-biotin peroxidase technique was used for detection of TGF-a. The sections were rinsed briefly with 0.5% Tween-PBS which was also used for further washing, and they were incubated with 10% normal goat serum for 30 min at room temperature. After a brief rinse, the sections were incubated with chicken anti-human TGF-CY antibody diluted in PBS/BSA (1: 1000) for 30 min at room temperature. The control sections were incubated with an equivalent amount of non-specific chicken IgG. After a brief rinse, the sections were incubated with a 1:250 dilution of biotinylated goat anti-chicken IgG (Vector Labs, Burlington, CA)
M. Hipshiynma et ul. /J. Drm~arol.Sci. 7 (1994) 45-53 for 30 min at room temperature, followed by extensive washing. Then, the tissue sections were incubated with avidin-peroxidase complex (Vecta Stain Elite kit, Vector Labs, Burlington, CA) for 30 min. The specific peroxidase substrate solution (3,3’-diaminobenzidine tetrahydrochloride with hydrogen peroxide) was added. The sections were washed with distilled water and counterstained in methyl green. The same procedure was performed for EGF receptor staining. Anti-EGF receptor antibody (Oncogene Science) at a 1:25 dilution in PBWBSA and biotinylated horse anti-mouse IgG (Vector Labs, Burlington, CA) were used as the first and second antibodies, respectively. 2.2. Western blotting Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed accor-
Fig. I. Recombinant TGF-or (lanes 1 and 3) and recombinant EGF (lane 2) were separated using 15% polyacrylamide gel. After SDS-PAGE. proteins were transferred onto nitrocellulose membrane. Lanes 1 and 2 were incubated with chicken anti-TGF-o antibody. and lane 3 was incubated with normal chicken IgG. Chicken anti-TGF-cy antibody reacted with TGFo. but not with EGF.
M. Higashiyama
et al. /J.
Dermatoi.
Sri.
7 (1994)
45-53
47
a
b
Fig. 2. Comparison of normal skin sections incubated with antibody to TGF-(Y (a), and preabsorbed antibody to TGF-cu (b) and antibody to EGF receptor(c). (a.b) In the epidermis. TGF-(Y was stained intensely through the upper several layers including the granular layer. In contrast. the basal cell layer showed no positive staining. In the dermis endothelial cells showed positive staining. Ant+TGF(Yantibody preabsorbed with TGF-(Y showed no positive staining. (c) EGF receptor was observed mainly in basal cells in the epidermis of normal skin as previously reported.
ding to the method of Laemmfi [lo]. Recombinant TGF-CY and EGF (Earth Pharmaceutical Co. Japan) were separated using 15% polyacrylamide gel. After SDS-PAGE, proteins were transferred
onto nitrocellulose membrane (Schleicher & Schuell, Dassel, Germany) by the method of Towbin et al. [l 11. TGF-CY was immunochemically detected using chicken anti-TGF-cr antibody, goat biotiny-
M. Higashi.vama et al. /J. Dermatol. Sri. 7 (1994) 4.5-53
Fig. 2. (continued)
lated anti-chicken IgG (Vector Labs, Burlington, CA) and peroxidase-conjugated avidin-biotin complex (Vector Labs, Burlington, CA). 3. Results To confirm the specificity of chicken anti-TGFcx antibody, the reactivity of this antibody was determined by Western blot. As shown in Fig. 1, this antibody reacted with TGF-a, but not with EGF. To investigate TGF-a localization in psoriatic skin, we first tested the immunoreactivity of this antibody with normal human skin. In the epidermis of normal skin, TGF-a! staining was intense through the upper several layers including the granular layer (Fig. 2a). In contrast, the basal cell layer did not show positive staining. Linear or dot staining pattern of TGF-a was observed mainly in the intercellular space, suggesting its localization on cell surface membrane. In the dermis, endothelial cells, eccrine duct and gland cells, and sebace-
ous glands showed positive staining. To further demonstrate that the immunostaining was specific for TGF-a, anti-TGF-cy antibody was preabsorbed with TGF-a as described in Section 2 above (Materials and methods). The preabsorbed antibody showed no positive staining (Fig. 2b). EGF receptor was observed mainly in the basal cell layer as previously reported (Fig. 2c) [ 121. It was localized on the cellular membrane also. Next, we examined the localization and distribution of TGF-(r and EGF receptor in the clinicallydetermined transitional area from uninvolved to involved skin as well as typical psoriatic plaques. In involved psoriatic epidermis TGF-a! was detected throughout the epidermis with the exception of the basal cell layer (Fig. 3a). The upper epidermis was stained more intensely than the lower epidermis in involved psoriatic skin. This observation was consistent in all five psoriatic plaques. In uninvolved skin adjacent to psoriatic plaque, the immunostaining pattern of TGF-a! was similar to that of normal epidermis (Fig. 3b). EGF
M. Higashiyama et al. /J.
Dermarol. Sci. 7 (1994) 45-53
49
a
b
epidermis from uninvolved to involved psoriatic skin. Clinical margin is shown by an Fig. 3 Expression I of TGF-ol in transitional of the involved psoriatic epidermis showed the presence of TGF-ol throughout the epidermis with the arrow (a) High m agnification except ion of the bmasal cell layer. The upper epidermis was stained more intensely than the lower epidermis. (b) High magnilicat ion zone showed the immunostaining pattern of TGF-a similar to normal epidermis. (c) Low magnification shot Ned of the transitional the whole skin.
50
M. Higashiyama et al. /J.
Dermatol. Sci. 7 (1994) 45-53
Fig. 3. (continued)
receptor was detected up to the upper epidermis in involved psoriatic skin (Fig. 4a) as previously reported [12]. The focal intensity of anti-EGF receptor staining was prominent in the basal cell layer and lower portion of rete ridge, and significantly reduced in the upper portion of epidermis. This pattern was in clear contrast to that of TGF-CL In the clinically-determined transitional area from uninvolved to involved shown by the arrows (Fig. 3b,c), TGF-(r staining was not increased although the increase of EGF receptor in the basal and suprabasal layer was clearly detected (Fig. 4b,c). Slight epidermal hyperplasia was also observed in the clinically-determined transitional area (Figs. 3b and 4b). The increase of TGF-a staining was observed within the edge of psoriatic plaques. It should be noted that EGF receptor increased even in the uninvolved area adjacent to psoriatic plaques (Fig. 4b). Taken together, the increase of EGF receptor expression preceded that of TGF-(r in the epidermis of psoriatic lesion. 4. Discussion This study aimed to elucidate the localization of TGF-(r in normal and psoriatic epidermis in comparison with that of EGF receptor. Since TGF-a exerts its effect through binding to EGF receptor,
the relationship between the localization of TGF-a and EGF receptor is important to understand their roles in the skin. Our data demonstrated that EGF receptor increased to the mid-layer even in the adjacent uninvolved and transitional epidermis of psoriatic plaques, preceding the increase of TGF(Y. The same phenomenon was observed in the wound healing model [ 131. Therefore, the increase of EGF receptor preceding that of TGF-c~ could be a general phenomenon in epidermal hyperplasia. Coffey et al. (1987) reported the autocrine mechanism of TGF-a in human keratinocyte growth [5], in which the binding of TGF-a to EGF receptor enhances further production of TGF-CL Our observation may indicate the necessity of preceding increase of EGF receptor for autocrine production of TGF-CX, resulting in epidermal hyperplasia. Several groups have reported the immunohistochemical localization of TGF-(r in epidermis. However, the results are not consistent with each other. One group showed cytoplasmic staining [14], while other workers reported membranous staining [7]. There are further discrepancies in reports on TGF-a! distribution in epidermis: basalar TGF-(Y immunostaining in normal epidermis and whole epidermal staining in psoriatic skin [7], immunostaining throughout normal epidermis and intense suprabasalar but not basalar immuno-
M. Higashiyama el al. /J.
Dermatol. Sri. 7 (1994) 4.5-53
51
a
Fig. 4. Expression of EGF receptor in transitional epidermis from uninvolved to involved psoriatic skin. Clin ical margin is shown by an arrow. (a) High magnification of the involved psoriatic skin the presence of EGF receptor throughout the thickened epi dermis. (b) High magnification of the transitional zone showed the appearance of EGF receptor in the suprabasal cell layers differel nt from normal epidermis. (c) Low magnification showed the whole skin.
52
Fig. 4. (continued)
reactivity in psoriatic epidermis [14], intense subcornea1 and faint basalar TGF-(Y immunostaining in psoriatic epidermis, and only basalar staining in normal epidermis [8]. We observed predominantly membranous staining pattern of TGF-(r, and subcornea1 and suprabasal immunoreactivity in both normal and psoriatic epidermis, although psoriatic epidermis showed more intense immunostaining. These discrepancies may reflect the difference of antibodies or immunostaining methods such as fixation. The result of in situ hybridization study also seems at odds [8]. The localization of TGF-(Y mRNA was reported to be in the subcorneal layer of psoriatic epidermis. However, the present study showed that protein expression was deeper into the epidermis of psoriatic lesions. The distribution of TGF-a and EGF receptor were different. In normal epidermis, EGF receptor was predominantly located in the basal layer, whereas TGF-a! was detected mainly in the upper epidermis, and not detected in the basal layer. Since TGF-a is supposed to facilitate keratinocyte growth, the coexistence of TGF-a and EGF receptor seems likely. However, our observation is inconsistent with this hypothesis. Recently, Finzi et al. (1992) reported that stronger immunostaining of TGF-(II in differentiated cells of appendageal tumors and the association of TGF-a with squamous cell differentiation [ 151. Furthermore, the presence of TGF-(r mRNA was not confirmed in the basal layer by in situ hybridization. In this
case, some other EGF-like growth factor such as amphiregulin [ 161 could be involved, mainly in epidermal keratinocyte growth in vivo. Recently, these investigators demonstrated that amphiregulin mRNA is elevated in psoriatic epidermis compared with non-lesional epidermis in normal skin [17]. Another possibility is that TGFQ could be synthesized but released rapidly from the cell surface after cleavage of the extracellular domain by elastase-like proteinase [18]. The antibody used in this study was raised against the recombinant mature TGF-(r and recognizes the mature form only. Therefore, it is possible that TGF-ol is synthesized much more in the basal layer than the upper epidermis, but released more quickly from the cell membrane. Two groups reported that non-cleavable pro-TGF-a is less potent than soluble TGF-(r as an activator of EGF receptor [19,20]. This observation may support the second possibility. In this case, the C-terminal portion of pro-TGF-cr remains through the plasma membrane. The antibody reacting with the C-terminal of pro-TGF-a could give us the answer to this important question.
5. Acknowledgement This work was supported by Grant-in Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan 03454274.
hf. Higashiyama et al. /J. 6.
Dermatol. Sri. 7 (1994) 45-53
53
References I
2
3
4
5
6
7
8
9
10
Marquardt H, Hunkapiller MW, Hood LE et al.: Transforming growth factors produced by retrovirus transformed rodent libroblasts and human melanoma cells: amino acid sequence homology with epidermal growth factor. Proc Nat1 Acad Sci 80: 4684-4688, 1983. Massagut J: Epidennal growth factor-like transforming growth factor. II. Interactions with epidermal growth factors in human placenta membranes and A43 I cells. J Biol Chem 258: 13614-13620, 1983. Carpenter G, Stoscheck CM, Preston YA et al.: Antibodies to the epidermal growth factor block the biological activities of sarcoma growth factor. Proc Nat1 Acad Sci 80: 5627-5630, 1983. Barrandon Y, Green H: Cell migration is essential for sustained growth of keratinocyte colonies: the roles of transforming growth factor-o and epideimal growth factor. Cell 50: 1131-1137, 1987. Coffey RJ, Derynck R, Wilcox JN et al.: Production and auto-induction of transforming growth factor in human keratinocytes. Nature 328: 817-820, 1987. Elder JT. Fisher GJ, Lindquist PB et al.: Overexpression of transforming growth factor-o in psoriatic epidermis, Science 243: 811-814, 1989. Gottlieb AB, Chang CK, Posnett DN et al.: Detection of transforming growth factor-o in normal, malignant and hyperproliferative human keratinocytes. J Exp Med 167: 670-675, 1988. Turbitt ML, Akhurst RJ, White SI et al.: Localization of elevated transforming growth factor-alpha in psoriatic epidermis. J Invest Dermatol 95: 229-232, 1990. Higashiyama M, Matsumoto K. Hashimoto K et al.: Increased production of transforming growth factor-o in psoriatic epidermis. J Dermatol 18: 117-l 19, 1991. Laemmli UK: Cleavage of structural proteins during the
II
12
13
14
15
16
17
18 19
20
assembly of the head of bacteriophage T4. Nature 227: 1970. 680-685, Towbin H, Staehelin T, Gordon J: Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Nat1 Acad Sci 76: 4350-4354, 1979. Nanney LB, Stoscheck CM, Magid M et al.: Altered (“‘I) epidermal growth factor binding and receptor distribution in psoriasis: J Invest Dermatol 86: 260-265. 1986. Stoscheck CM, Nanney LB. King Jr., LB: Quantitative determination of EGF-R during epidermal wound healing. J Invest Dermatol 99: 645-649. 1992. Finzi E, Harkins R, Horn T: TGF-or is widely expressed in differentiated as well as hyperproliferative skin epitheliurn. J Invest Dermatol 96: 328-333, 1991. Finzi E, Ho T, Anhalt G, Hawkins W et al.: Localization of transforming growth factor-o in human appendageal tumors. Am J Path01 141: 643-653. 1992. Cook PW. Mattox PA, Keeble WW et al.: A heparin sulfate-regulated human keratinocyte autocrine factor is similar or identical to amphiregulin. Mol Cell Biol 11: 2547-2557, 1991. Cook PW, Pittelkow MR. Keeble WN et al.: Amphiregulin mRNA is elevated in psoriatic epidermis and gastrointestinal carcinomas. Cancer Res 52: 3224-3227, 1992. Derynck R: Transforming growth factor-o. Cell 54: 593-595, 1988. Sharon TW, Lisa FW, Bryan KM et al.: The TGF-(Y precursor expressed on the cell surface binds to the EGF receptor on adjacent cells, leading to signal transduction. Cell 56: 495-506, 1989. Rainer B, Patricia BL, Nagashima M et al.: Transmembrane TGF-(U precursors activate EGF/TGF-(Y receptors. Cell 56: 691-700, 1989.