Increased cutaneous immunoreactive stem cell factor expression and serum stem cell factor level in systemic scleroderma

Journal of Dermatological Science 20 (1999) 72 – 78

Increased cutaneous immunoreactive stem cell factor expression and serum stem cell factor level in systemic scleroderma Chika Kihira, Hitoshi Mizutani *, Kunihiko Asahi, Hiroko Hamanaka, Masayuki Shimizu Department of Dermatology, Mie Uni6ersity, Faculty of Medicine, Tsu, Mie 5148507, Japan Received 30 September 1998; received in revised form 10 November 1998; accepted 10 November 1998

Abstract Skin hyperpigmentation and itching are characteristic findings in systemic sclerosis (SSC) patients. Stem cell factor (SCF, c-kit ligand) is a multifunctional cytokine which can promote melanocyte and mast cell development. We investigated the SCF expression histopathologically in normal and SSC skin, and compared the expression with the serum SCF levels measured with a specific enzyme-linked immunosorbent assay. The epidermal and dermal immunoreactive SCF expression was markedly higher in the forearm skin of edematous phase SSC patients than in that of normal subjects. Tissue SCF expression declined from the sclerotic phase to the atrophic phase, where it was close to the normal level. In contrast, the elevated serum SCF level seen in the edematous phase samples was further increased in the sclerotic phase samples. The serum SCF level decreased in the atrophic phase, but it still remained at a level higher than that of the normal controls. Itching and increase of dermal mast cell number are characteristic of edematous phase SSC, and are in bears a parallel to the presently observed dermal SCF expression profile. Pigmentation is significant in sclerotic phase SSC and lasts to the atrophic phase, which may correspond to the serum SCF level observed here. These results indicate a contribution of the fibroblast membrane integral SCF in dermal mast cell development, and of the soluble serum SCF to melanocyte activation in SSC. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: c-Kit; Steel factor; Pigmentation; Mast cell

1. Introduction * Corresponding author. Tel.: +81-59-231-5025; fax: +8159-231-5206. E-mail address: [email protected] (H. Mizutani)

Stem cell factor (SCF, c-kit ligand, steel factor, mast cell growth factor) is a growth factor for mast cell development, and is coded at the Sl locus in mouse chromosome 10 [1]. SCF was also

0923-1811/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 9 2 3 - 1 8 1 1 ( 9 8 ) 0 0 0 7 7 - 2

C. Kihira et al. / Journal of Dermatological Science 20 (1999) 72–78

cloned as a c-kit ligand for membrane tyrosine kinase-type oncogene: c-kit in W locus [2]. The Sl mutant mouse strain has an abnormality in the development of melanocytes, mast cells, erythrocytes and germ cells [3]. In humans, SCF is involved in hematopoietic cell development including that of mast cells and erythrocytes. Multifunctional SCF also affects human melanocyte growth and activation [4]. Generalized hyperpigmentation and itching of the early phase sclerosing skin lesion are characteristic clinical features of systemic sclerosis (SSC). These manifestations have been demonstrated histopathologically, as have an increase of mast cell number in lesional dermis [5] and epidermal melanin. Thus, it was hypothesized that SCF, a multifunctional growth factor for both mast cells and melanocytes, is involved in the manifestations of SSC. In the present study, we investigated the immunohistopathological expression of SCF in normal and SSC skin, and evaluated the serum SCF level in SSC samples, using specific enzyme-linked immunosorbent assay (ELISA).

73

at − 70°C until processing. In addition to SSC, mastocytoma (n= 2) was studied as positive control.

2.2.1. Western blotting samples The sample preparation and Western blot procedure were reported previously [7]. After the removal of fatty tissue and epidermis, the dermis of SSC and normal control samples were split into small pieces by scalpel and treated with 0.5% Triton X-100 (Sigma, St. Louis, MO) in Tris-buffered saline consisting of 0.05 M Tris–HCl and 0.1 M NaCl, pH 7.5, for 1 h at 4°C, and then centrifuged at 15 000 rpm for 10 min at 4°C. The supernatant was collected and stored at − 20°C until use. The normal epidermal sheet was shaved with a keratotome in plastic surgery operations. Normal human keratinocytes and SV40 transformed keratinocytes (RHEK) were cultured as reported previously [7]. Split epidermal sheets and cultured cells were immersed in SDS-sample buffer and stored at −20°C until use. 2.3. Anti-SCF antibodies

2. Materials and method

2.1. Patients Thirty-three female SSC patients aged 25 – 75 years (mean 48.1 years), diagnosed as having SSC by the American Rheumatism Association criteria, were investigated. They were divided into three groups clinically; edematous/early phase (n = 3), sclerotic/classic phase (n = 26) and atrophic/late phase (n=4) [6]. Twenty-four age-matched female healthy volunteers, aged 25 – 70 years (mean 45.2 years), were also investigated as normal controls.

2.2. Samples Skin biopsy specimens were obtained from the forearm and fingers of the SSC patients (edematous phase, n= 3; sclerotic phase, n =8; atrophic phase, n = 4) and normal subjects after informed consent. Samples were divided into two pieces, and one piece was processed for paraffin-embedded sections. The other piece was snap-frozen in OCT compound (Miles, Elkhalt, IN) for frozen sections, and stored

The primary antibodies used for immunohistochemical studies were polyclonal rabbit anti-SCF peptide antibody (SCF-89: IBL, Maebashi, Japan) and sheep anti-human SCF antibody (Genzyme, Cambridge, MA), and monoclonal mouse anti human SCF (Genzyme, Cambridge, MA). The immunogen of the polyclonal rabbit anti-human SCF is an extracellular domain (89–103) synthetic peptide, and the immunogen of the others are recombinant human SCF (rh-SCF).

2.4. Immunohistochemical study The deparaffinized sections were blocked with diluted normal horse serum without enzyme treatment, and 100 ml of diluted anti-SCF antibodies were applied to each sections. The final concentrations of antibodies were determined by preliminary experiment (rabbit anti-human SCF peptide antibody: 0.01 mg/ml; sheep anti-human SCF antibody: 0.01 mg/ml; mouse monoclonal anti-human SCF antibody: 0.02 mg/ml). Bound antibody was detected by the avidin-biotin peroxidase com-

74

C. Kihira et al. / Journal of Dermatological Science 20 (1999) 72–78

plex method (ABC-Elite kit, Vector, Burlingame, CA) and 3-amino-9-ethylcarbazole (AEC) as a chromogen. We used a biopsy specimen of a mastocytoma as a positive control. As a negative control, samples were stained without primary antibody. From the results of preliminary experiments, mainly rabbit anti SCF antibody (SCF-89) was used as the primary antibody, because of its high intensity and low background. More than three slides were studied in each specimens.

2.5. ELISA for serum SCF We measured the serum SCF levels of the 24 age-matched normal controls and 33 SSC patients with a SCF specific sandwich ELISA kit (Quantikine human SCF Immunoassay, R&D, Minneapolis, MN). Ninety-six-well plates were coated with murine monoclonal antibody against SCF. One hundred microliters of samples and series diluted SCF standard were incubated with 100 ml of a buffered protein base in each well at room temperature for 2 h. After washing, they were incubated with 200 ml of horseradish peroxidaselabeled anti-SCF polyclonal antibody for 2 h. The bound antibodies were detected with tetrametylbenzidine substrate. Absorbance at 450 nm was measured with an ELISA reader (Bio-Rad 350, Richmond, CA).

3. Results

3.1. Immunohistochemical expression of SCF in the normal and scleroderma skin 3.1.1. Normal skin The polyclonal rabbit anti-h-SCF extracellular domain peptide antibody (SCF-89) reacted intensely with suprabasal keratinocytes, but showed limited reactivity to the basal layer cells. Immunoreactive SCF was detected in the cytoplasm of keratinocytes. These antibodies also reacted at hair follicles, sparsely distributed dermal interstitial cells and endothelial cells (Fig. 1, Table 1). Both the polyclonal sheep rh-SCF antibody and monoclonal mouse rh-SCF antibody could detect immunoreactive SCF in the epidermis. They reacted diffusely through the epidermal layers with very high background (data not shown). Therefore, the following results were from experiments using anti-h-SCF peptide antibody (SCF-89). 3.1.2. Scleroderma skin Distribution of immunoreactive SCF was significantly enhanced in the edematous phase samples. Interstitial cells stained with anti-SCF antibody were increased in the dermis, and its immunoreactivity was more intense compared with that of the normal controls (Fig. 2). The

2.6. Western blotting Samples were analyzed with immunoblotting as reported previously.[7] Briefly, the samples electrophoresed in 15% sodium dodecyl sulfate polyacrylamide gel were electrically transferred onto a nitrocellulose membrane. After blocking with 3% skim milk, the membrane was incubated with anti-h-SCF peptide antibody for 1 h as primary antibody. After washing, the bound antibody was detected with alkaline phosphatase conjugated anti-rabbit IgG antibody and Western blue developing system (Promega, Madison, WI).

2.7. Statistical analysis Mann–Whitney U-test and Fisher’s exact test were used to compare the group data. Probability values less than 5% were considered significant.

Fig. 1. Immunoreactive SCF distributed mainly in the cytoplasm of epidermal keratinocytes. SCF reactivity is marked in the spinous layer cells, but basal keratinocytes were scarcely stained. In dermis, SCF was detected in the dermal interstitial cells and dermal vessels (anti-human SCF antibody (SCF-89) staining with AEC, bar: 50 mm).

C. Kihira et al. / Journal of Dermatological Science 20 (1999) 72–78

75

Table 1 Immunoreactive SCF distribution and intensity in normal and SSC skin Normal

Epidermis Dermal interstitial cells (Cell number) Endothelial cells

Suprabasal + + +

immunoreactivity of epidermal keratinocytes was also increased in the SSC samples, but the distribution of SCF was the same as that of the normal controls. In the sclerotic phase samples, the SCFpositive dermal interstitial cell number was markedly decreased (Fig. 3). The immunoreactivity of the epidermis also decreased. In the atrophic phase SSC samples, the immunoreactivity of epidermal SCF was decreased, close to the normal control level (Fig. 4). Very few dermal interstitial cells were scarcely reactive to the anti-SCF antibody.

3.2. Serum SCF le6el The serum SCF level of SSC patients was 1574.49 56.5 pg/ml (mean9S.E.M.), which was significantly higher than that of normal control (1201.3940.5 pg/ml, P B0.001, Fig. 5). The serum SCF level in the edematous phase patients was slightly elevated at 1305.09 125.0 pg/ml. In the sclerotic phase patients, it was significantly elevated up to 1619.5 959.6 pg/ml (PB 0.001). The SCF levels in the atrophic phase serum samples were significantly decreased from the sclerotic phase (1270.0959.6 pg/ml, P B 0.05), which was still higher than the level in normal controls. In relation to treatment, the serum SCF levels of the systemic corticosteroid-treated SSC patients (1539.89 68.2 pg/ml) was slightly lower than that of the corticosteroid non-treated SSC patients (1629.19 99.9 pg/ ml) with no significant difference. There was also no significant difference between groups categorized by autoantibodies (anti-Scl-70 antibody, antiU1-RNP antibody, anti-centromere antibody) and limited and diffuse SSC (data not shown).

SSC phase Edematous

Sclerotic

Atrophic

Suprabasal ++ ++   +

Suprabasal ++ 9 ¡ 9

Suprabasal + 9 ¡ 9

3.3. Western blotting The extract of SSC dermis showed an intense anti-SCF antibody reactive band at 35 kD, which is compatible to the long membrane bound form SCF. However, no clear band was detected in the lane of the normal dermis extract. The samples from shaved normal epidermis showed an immunoreactive single band at the molecular weight 19 kD, compatible to soluble rh-SCF (Fig. 6). SV40 transformed differentiated keratinocyte cells (RHEK) expressed a stronger band at the same molecular weight; however, the undifferentiated normal human keratinocytes cultured under low calcium conditions scarcely expressed an immunoreactive band.

Fig. 2. The immunoreactivity of SCF was notably increased in the specimen of edematous phase forearm specimen of SSC patients. Dermal spindle-shaped cells and endothelial cells were stained clearly. Spinous layer cells of epidermis were stained intensely, but basal cells were weakly stained (anti-human SCF antibody (SCF-89) staining with AEC, bar: 50 mm).

76

C. Kihira et al. / Journal of Dermatological Science 20 (1999) 72–78

Fig. 3. The specimen of established sclerotic phase skin had moderately decreased SCF immunoreactivity compared with the edematous phase samples. The number of the dermal spindle-shaped stained cells was less than that of the edematous phase (anti-human SCF antibody (SCF-89) staining with AEC, bar: 50 mm).

4. Discussion The importance of SCF (c-kit ligand) in melanogenesis, and hematopoiesis (especially mast cell development) has been well-recognized in the murine W locus [8] and Sl locus mutation [9], and in the c-kit mutation in human piebaldism [10]. SSC is a systemic inflammatory autoimmune disease, which expresses intensive hyperpigmentation and itching as major clinical manifestations. Because of the unknown etiology of SSC, the pathogenesis of these manifestations is also not yet clear. Since no incontinentia pigmenti histologica

Fig. 4. In the atrophic phase SSC samples, the immunoreactivity of epidermis and dermal spindle-shaped cells was far lower than that of the sclerotic phase specimens (anti-human SCF antibody (SCF-89) staining with AEC, bar: 50 mm).

Fig. 5. The serum SCF level was significantly increased in the SCC samples (P B0.001), and was maximally elevated in the sclerotic phase samples (P B0.001). The SCF level in the atrophic phase samples was low, close to normal level.

has been observed in SSC, hyperpigmentation mainly due to the intraepidermal deposition of over-produced melanin, and melanocyte activation. On the other hand, the itching in SSC has been speculated to be due to the increased chemical mediators discharged from dermal mast cells in the lesional dermis. In the active phase of SSC, the dermal mast cell counts were increased in proliferating dermal collagenous tissue [5]. Of the various growth factors, SCF, granulocyte macrophage colony stimulating factor, fibroblast growth factors, and endothelin are known melanocyte growth and activating factors [4,11], and SCF, interleukin (IL)-3, IL-4, IL-9, and IL-10 are mast cell growth and activation factors [12]. Thus, we hypothesized that SCF is a bipotential

Fig. 6. Lane 1: rh-SCF; lane 2: keratotome-shaved normal skin; lane 3: cultured normal human keratinocytes; lane 4: SV40 transformed keratinocytes cell line (RHEK); lane 5: SSC dermis extract; and lane 6: normal dermis extract. Stained with anti-human SCF antibody (SCF-89).

C. Kihira et al. / Journal of Dermatological Science 20 (1999) 72–78

growth factor for mast cell and melanocytes in SSC. As we expected, the SSC skin lesions had increased immunoreactive SCF expression. Both the dermal and epidermal SCF were maximally expressed in the edematous phase, and this tissue SCF expression decreased in accordance with the clinical course, and reached to the normal control level in the atrophic phase. In addition to tissue, the serum SCF level of SSC was also significantly increased compared with that of normal control. However, serum SCF level was moderately elevated at the edematous phase, but further elevated at the sclerotic phase. In the atrophic phase, it returned to that of the edematous phase, but it still remained at a level higher than that of the normal controls. Thus, there was a discrepancy between the tissue and serum SCF profiles in SSC. SCF has two isoforms, the long and short membrane-bound forms [3]. Because of the absence of the processing sequence in the short form, soluble from SCF has been speculated to be shed from long membrane bound form [13]. Unlike bone marrow SCF, the biological role of cutaneous SCF has not yet been determined. The dermal mast cells increase in the early phase of SSC, and decrease to the normal level at the late phase [5], which is similar to the tissue SCF expression profile obtained in the present study. In addition, the Western blot results suggest that dermal SCF is the long membrane-bound form. The long membrane-bound form is indispensable in hematopoietic cell development, and affects the adhesion to mast cells [3]. SSC dermal mast cells proliferate between fibrous tissues. Recently, the mast cell chymase has been reported as a potent SCF solubilizing enzyme [14]. Therefore, it is likely that dermal SCF contributes to de novo mast cell proliferation in a paracrine system. The epidermal SCF was also increased in the present SSC samples. The inter-cellular distribution of SCF in mastocytoma [15] has been suggested. In the present study, the epidermal SCF of SSC and mastocytoma samples were distributed with an intra-cytoplasmic pattern the same as that found by Grabbe et al.’s report [16]. The samples in the present study did not show epidermal cell lysis nor free immunoreactive SCF deposition in the upper dermis; therefore, a direct effect of

77

intracytoplasmic epidermal SCF on the dermal mast cells in the SSC patients was unlikely. Keratinocytes SCF was reported as a non-alternative spliced form at the polymerase chain reactionbased messenger RNA level [17], but this has not been well-characterized biochemically and biologically. In the present study, epidermal SCF migrated at 19 kD, which is a far smaller molecular weight than that of dermal SCF and is close to that of the non-glycosylated mature form of SCF. It is likely that epidermal SCF is physiologically different from dermal SCF. Serum SCF has been speculated to be enzymatically shed products from the long membrane bound form [13]. However, the source of serum SCF and the biological roles of serum SCF are not well understood. In contrast to tissue SCF, the serum high SCF level observed in the present study persisted even at the late atrophic phase of SSC. The pigmentation of SSC becomes remarkable at the sclerotic phase and persists until the atrophic phase. It is likely that serum SCF affects melanocytes. Grichnik et al. [18] reported local hyperpigmentation at soluble recombinant SCF subcutaneous injection sites, but not an increase of mast cells. This may indicate that the membrane-bound form SCF affects mast cells, and the soluble form targets melanocytes in vivo. However, SSC is a systemic inflammatory autoimmune disease, so we cannot eliminate the possibility that unknown hematopoietic stimuli up-regulated the serum SCF level. The regulation of serum SCF is still unclear. In relation to treatment, the serum SCF level showed some decrease in the corticosteroid-treated patients compared with non-treated patients, but not significantly so different. In our experience, the hyperpigmentation of some systemic corticosteroid-treated SSC patients improves clinically. Since several factors are known for growth and activation of mast cells (IL-3, IL-4, IL-9 and IL-10) and melanocytes (IL-1a, endothelin, heparin and histamine), SCF can not account for hyperpigmentation and itching in SSC. However, the present data suggests that SCF has a pathogenic role in part in the local and systemic manifestations of SSC.

78

C. Kihira et al. / Journal of Dermatological Science 20 (1999) 72–78

Acknowledgements This work was supported in part by grants from Japanese Ministry of Education (HM 08670962) and Ministry of Health and Welfare (HM).

[9]

[10]

References [1] Copeland NG, Gilbert DJ, Cho BC, Donovan PJ, Jenkins NA, Cosman D, Anderson D, Lyman SD, Williams DE. Mast cell growth factor maps near the steel locus on mouse chromosome 10 and is deleted in a number of steel alleles. Cell 1990;63:175–83. [2] Matsui Y, Zsebo KM, Hogan BL. Embryonic expression of a haematopoietic growth factor encoded by the Sl locus and the ligand for c-kit. Nature 1990;347:667–9. [3] Flanagan JG, Chan DC, Leder P. Transmembrane form of the kit ligand growth factor is determined by alternative splicing and is missing in the Sld mutant. Cell 1991;64:1025 – 35. [4] Funasaka Y, Boulton T, Cobb M, et al. c-Kit-kinase induces a cascade of protein tyrosine phosphorylation in normal human melanocytes in response to mast cell growth factor and stimulates mitogen-activated protein kinase but is down-regulated in melanomas. Mol Biol Cell 1992;3:197 – 209. [5] Hawkins RA, Claman HN, Clark RA, Steigerwald JC. Increased dermal mast cell populations in progressive systemic sclerosis: a link in chronic fibrosis? Ann Int Med 1985;102:182 – 6. [6] LeRoy EC. Scleroderma. In: Kelley WN, editor. Textbook of Rheumatology. Philadelphia, PA: Saunders, 1981:1217 – 8. [7] Mizutani H, Hayashi T, Nouchi N, Ohyanagi S, Hashimoto K, Shimizu M, Suzuki K. Functional and immunoreactive thrombomodulin expressed by keratinocytes. J Invest Dermatol 1994;103:825–8. [8] Duttlinger R, Manova K, Chu TY, Gyssler C, Zelenetz AD, Bachvarova RF, Besmer P. W-sash affects positive

.

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

and negative elements controlling c-kit expression: ectopic c-kit expression at sites of kit-ligand expression affects melanogenesis. Development 1993;118:705 – 17. Brannan CI, Lyman SD, Williams DE, Eisenman J, Anderson DM, Cosman D, Bedell MA, Jenkins NA, Copeland NG. Steel – Dickie mutation encodes a c-kit ligand lacking transmembrane and cytoplasmic domains. Proc Natl Acad Sci USA 1991;88:4671 – 4. Giebel LB, Spritz RA. Mutation of the KIT (mast/stem cell growth factor receptor) protooncogene in human piebaldism. Proc Natl Acad Sci USA 1991;88:8696 – 9. Krasagakis K, Garbe C, Zouboulis CC, Orfanos CE. Growth control of melanoma cells and melanocytes by cytokines. Recent Results Cancer Res 1995;139:169 – 82. Rottem M, Hull G, Metcalfe DD. Demonstration of differential effects of cytokines on mast cells derived from murine bone marrow and peripheral blood mononuclear cells. Exp Hematol 1994;22:1147 – 55. Langley KE, Bennett LG, Wypych J, Yancik SA, Liu XD, Westcott KR, Chang DG, Smith KA, Zsebo KM. Soluble stem cell factor in human serum. Blood 1993;81:656 – 60. Longley BJ, Tyrrell L, Ma Y, Williams DA, Halaban R, Langley K, Lu HS, Schechter NM. Chymase cleavage of stem cell factor yields a bioactive, soluble product. Proc Natl Acad Sci USA 1997;94:9017 – 21. Longley B Jr., Morganroth GS, Tyrrell L, Ding TG, Anderson DM, Williams DE, Halaban R. Altered metabolism of mast-cell growth factor (c-kit ligand) in cutaneous mastocytosis. New Engl J Med 1993;328:1302– 7. Grabbe J, Welker P, Dippel E, Czarnetzki BM. Stem cell factor, a novel cutaneous growth factor for mast cells and melanocytes. Arch Dermatol Res 1994;287:78 – 84. Morita E, Lee DG, Sugiyama M, Yamamoto S. Expression of c-kit ligand in human keratinocytes. Arch Dermatol Res 1994;286:273 – 7. Grichnik JM, Crawford J, Jimenez F, Kurtzberg J, Buchanan M, Blackwell S, Clark RE, Hitchcock MG. Human recombinant stem-cell factor induces melanocytic hyperplasia in susceptible patients. J Am Acad Dermatol 1995;33:577 – 83.