- Email: [email protected]
Human malignant melanoma cells release a factor that inhibits the expression of smooth muscle α-actin
Journal of Dermatological Science 23 (2000) 170 – 177 www.elsevier.com/locate/jdermsci
Human malignant melanoma cells release a factor that inhibits the expression of smooth muscle a-actin M. Okamoto-Inoue a, J. Nakayama b, Y. Hori c, S. Taniguchi d,* a
Department of Dermatology, Faculty of Medicine, Kurume Uni6ersity, Kurume, Fukuoka, Japan b Department of Dermatology, Faculty of Medicine, Fukuoka Uni6ersity, Fukuoka, Japan c Aso Iizuka Hospital, Iizuka, Fukuoka, Japan d Angio-Research Di6ision, Department of Molecular Oncology and Angiology, Research Center on Aging and Adaptation, Shinshu Uni6ersity School of Medicine, 3 -1 -1 Asahi, Matsumoto 390 -8621, Japan Received 20 October 1999; received in revised form 29 November 1999; accepted 30 November 1999
Abstract Smooth muscle a-actin (SMA) is a cytoskeletal protein expressed in vascular smooth muscle cells, pericytes, hair follicle dermal sheaths and myofibroblasts which appear in the process of wound healing and tumor growth. To examine the effect of malignant melanoma on the expression of SMA in these non-neoplastic cells, we carried out immunohistochemical staining and a cell culture study. Conditioned medium prepared from a melanoma cell line M14 was incubated with a rat fibroblastic cell line 3Y1, which had been shown to express SMA. Human cells that had migrated from nevus tissue were also cultured either with or without M14 conditioned medium. Immuno-histochemical staining of human melanoma tissues suggested that the expression of SMA was low in the vicinity of the tumor as well as within the tumor nodules. The conditioned medium from melanoma, but not the medium from control non-neoplastic cells, suppressed the expression of SMA both in the 3Y1 cells and human cells that migrated from the nevus. Preincubation of the medium with anti-platelet-derived growth factor allowed 76% recovery of SMA expression. These data thus imply that melanoma cells release a platelet-derived growth factor-like substance which has a suppressive effect on the contractile elements in non-neoplastic cells. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Platelet-derived growth factor; Pericyte; Myofibroblast; Cytoskeletal protein
1. Introduction Abbre6iations: Ab, antibody; CM, conditioned medium; PDGF, platelet-derived growth factor; SMA, smooth muscle a-actin. * Corresponding author. Tel.: +81-263-372679; fax: + 81263-372724. E-mail address: [email protected] (S. Taniguchi).
Malignant tumor cells have been shown to produce several substances with biological activities, including autocrine growth factors [1], angiogenic factors [2], and enzymes which degrade matrix proteins [3]. Growth factors are also likely to have paracrine effects which modulate the proliferation
0923-1811/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 9 2 3 - 1 8 1 1 ( 0 0 ) 0 0 0 7 2 - 4
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
and matrix production of the stromal cells. However, little is known regarding the effects of malignant cells on cytoskeletal systems in the surrounding non-neoplastic cells. Actin is one of the major cytoskeletal proteins which plays an essential role in morphogenesis, movement, contraction and the attachment of cells. The disorganization of actin stress fibers is a common feature of malignant cells [4] and recent advances in molecular biology have shown that the introduction of a mutant b isoform of actin into cells increases tumorigenicity [5], while another type of variant b actin reduces invasiveness [6] and metastatic ability [7]. Thus, an altered expression of actin isoforms may lead to changes cellular agressiveness. We [8] have shown that melanoma tissues display less smooth muscle a-actin (SMA) compared with benign tissues, whereas the b- and g-cytoplasmic actins were not altered. Among the several isoforms of actin so far identified in benign human cells, SMA is expressed specifically in vascular smooth muscle cells, pericytes [9], a specific region of hair follicle dermal sheath [10], myoepithelial cells [11] and myofibroblasts in both wound healing [12] and tumors [11]. SMA has also been detected in embryonic, and fibroblastic cell lines such as the mouse 3T3 [13] and rat 3Y1 [14]. SMA expression is (i) tightly regulated during the cell cycle process, differentiation and transformation; (ii) expressed transiently in myofibroblasts during wound healing [12]; (iii) lost with viral transformation of 3Y1 and 3T3 cells [13,14]; decreased when aortic smooth muscle cells move into the G0/G1 during initial thickening after endothelial injury [15]; suppressed by the mitogen platelet-derived growth factor (PDGF) in cultured aortic smooth muscle cells [16 – 18]. Our immunohistochemical findings of melanoma tissue led us to examine the effect of a culture supernatant of melanoma cells on the expression of SMA in normal, SMA-positive cells. As a model of myofibroblast, we used a rat embryonic, and fibroblastic cell line 3Y1, which is a stable cell line and has been characterized to express SMA constitutively under the culture condition in 10% fetal bovine serum (FBS) [14].
171
2. Materials and methods
2.1. Immunohistochemistry Monoclonal anti-a-smooth muscle actin antibody was purchased from Sigma. Formalin-fixed, paraffin-embedded tissue sections of human malignant melanoma were stained using an avidin– biotin–alkaline phospatase system (Vectastain ABC-AP) and a substrate kit (Vector Red) as described by the manufacturer.
2.2. Cell culture The human melanoma cell line UCLA M14 (M14) has been described elsewhere [20]. Melanoma cells were cultured in Eagle’s minimum essential medium (MEM) supplemented with 10% fetal bovine serum (FBS) and 80 mg/ml Kanamycin. The medium was changed twice a week. Explant cultures of human fibroblastic cells were carried out from the excised tissue of pigmented nevi. A part of each tissue was then fixed in formalin for the histological diagnosis. For the culture, the rest of the tissue was rinsed three times in PBS containing 1000 U/ml penicillin and 800 mg/ml Kanamycin. The tissue was then cut into 1 mm3 pieces, and cultured in Petri dishes in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS and antibiotics. When migrated cells were grown in a subconfluent state, the cells were trypsinized and transferred. These human cells were then used in the second passages for the assays of SMA expression. Conditioned medium from subconfluent M14 cells (M14CM) filtered (0.22 mm Milex GV filter, Millipore Intertech, Bedford, MA) and were stored at −80°C until use. For controls, CM from benign cells was prepared, similarly from 3Y1 cells, 3T3 cells or human fibroblasts cultured in MEM supplemented with 10% FBS and antibiotics. In order to assess the effect of M14CM, the culture medium of 3Y1 cells in a growing phase was replaced with either 66% M14CM or control 66% CM in fresh medium and then the cells were cultured for the indicated period. Human fibroblastic cells cultured from human pigmented nevus
172
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
tissue were also treated with either M14CM or control CM for 48 h in the same way as the treatment of 3Y1. In some experiments, M14CM and control CM were pretreated for 30 min at 37°C either with or without anti-human PDGF Antibody (R&D systems, Minneapolis, MN) reactive with both PDGF-BB and AB. The final concentrations of anti-human PDGF antibody were 0, 37.5, 75, 150 mg/ml. Human platelet PDGF-AB (Wako Pure Chemical Industries, Ltd, Osaka, Japan) was diluted with PBS containing 2 mg/ml bovine serum albumin (BSA, fraction IV). Fifty microliters of PDGF solution (final concentration 5 ng/ml) or vehicle was then added to fresh medium and preincubated with anti-PDGF antibody (70, 35 or 0 mg/ml) for 30 min at 37°C. There are several reports that human PDGF plays biological functions on rat cells through the PDGF receptor [20]. Then 3Y1 cells were treated for 24 h with the above preincubated medium.
2.3. Immunofluorescent staining of cultured cells SMA in cultured cells was stained as described previously [14]. Briefly, the cells were cultured on glass coverslips overnight and fixed in ice-cold methanol for 5 min, dried and then incubated either with monoclonal anti-SMA (1:500) or with mouse control ascites fluid (Cederlane Laboratories, Ontario, Canada) for 30 min at room temperature. The cells were washed in PBS three times, incubated with FITC-labeled anti-mouse IgG (Daco, Glostrup, Denmark) (1:250) for 30 min, and then washed. The stained cells were mounted with glycerol and observed with a Zeiss Axiophoto microscope.
2.4. SDS –PAGE and immunoblotting The cultured cells in each dish were washed with PBS and harvested with a rubber policeman. The cells were collected in a 1.5-ml microcentrifuge tube and sonicated in PBS containing 1% sodium dodecyl sulphate (SDS). The protein concentration was measured with a BCA protein assay kit (Pierce, IL) using BSA as a standard. The samples were adjusted to a 5 mg/ml protein
concentration in SDS-sample buffer and then were separated on SDS–PAGE. The concentration of acrylamide was 10%. After electrophoresis the proteins were electroblotted to a nitrocellulose filter and stained with a 1:1000 dilution of monoclonal anti-SMA as described previously [8]. The relative intensities of the bands on the nitrocellulose filter were compared densitometrically using a Dual Wavelength Flying-spot scanner CS-9000 (Shimadzu Co., Kyoto, Japan) at a wavelength of 590 nm.
3. Results
3.1. Immunohistochemical staining of SMA The cells expressing SMA were screened in the melanoma tissue using immunohistochemical staining. Fig. 1 shows the staining of SMA in a representative tumor operated from the face of an 84-year-old man. The localization of the melanoma was recognized as a pigment-laden tumor mass (Fig. 1a). SMA was positive in the vessel walls of the capillaries, venules and arterioles surrounding the tumor. No myofibloblasts were seen in this case. No melanoma cells were stained with SMA. In the close vicinity of the tumor, the SMA in the vessel walls was often discontinuous and obscure. Within the tumor nodule, SMA was occasionally seen but the staining was faint. In a higher magnification (Fig. 1b), the capillaries were recognized within the tumor but the SMA in those capillaries was very faint.
3.2. Effect of M14CM on SMA le6el in 3Y1 cells The constitutive expression of SMA in 3Y1 was confirmed under the culture condition with 10% fetal bovine serum as reported previously [14] (data not shown). The incubation of M14CM for 24 h decreased the SMA in 3Y1 cells compared with that incubated with 3Y1CM (Fig. 2a). The conditioned medium from other benign cells such as 3T3 and human fibroblasts did not change the amount of SMA protein either (data not shown). Fig. 2b shows the time course of the suppression of SMA by M14CM. SMA expression was con-
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
173
stant until 12 h after incubation with M14CM and was inhibited completely SMA in the 3Y1 cells from 18 h up to 2 days. SMA expression started to recover 3 days after the incubation with M14CM. The control 3T3CM showed no change in the SMA level for 3 days.
Fig. 2. Immunoblot of SMA in 3Y1 cells after incubation with M14CM. (a) 3Y1 cells 24 h after incubation with control 3Y1 CM (lane 1) and M14CM (lane 2). (b) 3Y1 cells were extracted before (lane 1) or 3 h (lane 2), 6 h (lane 3), 12 h (lane 4), 18 h (lane 5), 24 h (lane 6), 2 days (lane 7) and 3 days (lane 8) after incubation with M14CM. Control 3Y1 cells were incubated with 3T3CM for 3 days (lane 9). Fifty micrograms of the extracted protein was applied in each lane.
3.3. Reco6ery of SMA by anti-PDGF Ab
Fig. 1. Immunohistochemical detection of SMA in malignant melanoma by avidin – biotin–alkaline phosphatase method. (a) Marginal area of a nodular melanoma; (b) tumor nodule, counterstained with hematoxylin. Arrows indicate faintly-positive vessel walls. Bars: (a) 100 mm; (b) 20 mm.
Because PDGF is a known suppresser against the expression of SMA in aortic smooth muscle cells [16–18], we examined the effect of preincubation of anti-PDGF blocking Ab with M14CM. As shown in Fig. 3a, SMA recovered to up to 76% of the control level by the preincubation with anti-PDGF antibody. On the other hand, the SMA-suppressing activity of the pure PDGF was blocked completely by the addition of the antibody (Fig. 3b).
174
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
3.4. Demonstration of SMA in human cells that migrated from the ne6us tissue Two types of the SMA-positive migrating from the nevus could be morphologically distinguished: one type consisted of large, polygonal cells with wide-spread cytoplasm and well-developed actin stress fibers, while the other type had a stellate or spindle-shaped cytoplasm (Fig. 4).
Fig. 4. Immunofluorescence staining of SMA in the cells migrated from human pigmented nevus tissue. Bar: 50 mm.
Fig. 5. Immunoblot analysis concerning effect of M14CM on SMA in the migrated fibroblastic cells cultured from human pigmented nevus tissue. Immunoblot. Each dish of primary culture was evenly transferred into two dishes, then one dish was added with 3Y1CM (lanes 1 and 3), while another was added with M14CM (lanes 2 and 4). Lanes 1 and 2: case 1; lanes 3 and 4: case 2.
Fig. 3. Effect of anti-PDGF antibody on the SMA-suppressive activity of M14CM and PDGF. M14CM (a) or PDGF (b) was preincubated with antibody and added to 3Y1 cells. a, bands of SMA on immunoblot were densitometrically quantified. (b) Immunoblot. Cells were incubated with: vehicle (lane 1), 5 ng/ml PDGF (lane 2), 5 ng/ml PDGF and 70 mg/ml antiPDGF antibody (lane 3), or 5 ng/ml PDGF and 35 mg/ml anti-PDGF antibody (lane 4).
3.5. Effect of M14CM on SMA in the human migrated cells Fig. 5 shows the SMA levels in the human migrated cells after the addition of M14CM or control CM. Although the amount of SMA differs among the culture lots of the cells, M14CM
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
diminished SMA in each lot when compared with the paired control dish. 4. Discussion Our immunohistochemical study showed that main sources of SMA in malignant melanoma were vascular smooth muscle cells and pericytes, which was consistent with the report by Tsukamoto et al. [19]. In our staining data with primary tumors, fewer myofibroblasts were detected than in cases with breast carcinomas [21], ovarian tumors [22], uterine cervix cancer [23] and tumors of the salivary glands [24]. The staining also demonstrated a decreased expression of SMA in the pericytes within the tumor cell nest and in the area adjacent to the nest. These findings are also consistent with our reports on the immunohistochemical staining of malignant and benign tumors [25,26]. In support of these findings, our previous electrophoretic analysis [8] showed a relatively low amount of SMA in the tissue of malignant melanoma compared with that from nevus pigmentosus and other types of benign pigmented tissue. The decrease in SMA is not an indicate of less pericytes, because there were weakly positive cells within the tumor cell nests. However, it has been suggested that tumors have immature microvessels after neoangiogenesis and that those immature vessels lack SMA in pericytes [27]. These considerations therefore prompted us to carry out suppression experiments in vitro. In the present study, we showed the suppression of SMA in benign 3Y1 cells by a culture supernatant of melanoma. SMA in human cells cultured from the skin also decreased after treatment with M14CM. Such a suppression of SMA in the cultured cells was the result of an inhibitory factor for (1) the control CM prepared from either 3Y1 or human fibroblasts did not suppress SMA, (2) when a lower concentration of FBS is used in the medium, SMA expression is substantially enhanced[28,29], (3) the suppression was reversible and the expression of SMA gradually recovered after 3 days and (4) the suppression activity was blocked by the addition of anti-PDGF antibody. These findings thus indicate that SMA-suppressing activity exists in the M14CM.
175
Several human melanoma cell lines and melanoma in vivo have been shown to express A and B chains of PDGF [30], transforming growth factor (TGF-b), basic fibroblast growth factor, acidic fibroblast growth factor and keratinocyte growth factor [31]. Because the effects of PDGF on SMA in cultured aortic smooth muscle cells have been well-characterized [16–18], we thus expected anti-PDGF antibody to be a candidate of a blocking reagent against a factor which suppresses SMA. In our blocking experiments, this antibody completely blocked the suppressive activity of human PDGF while the activity of M14CM was blocked partially (76%). The partial block may be due to the heterogeneity in PDGF subunits contained in the M14CM. The human PDGF used in this experiment was AB heterodimer and the antibody raised against human PDGF demonstrated cross-reactivity to a BB homodimer with different affinity according to the manufacturer. Other cytokines in M14CM may also influence the suppression of SMA. For example, growth stimulation also diminishes the SMA content in vascular smooth muscle cells [15,16]. Although TGF-b by itself stimulates the SMA expression in serum-free systems [29,32], it can also enhance the effects of PDGF by inducing autocrine PDGF production [33]. Interferongamma [34] or other unknown factors which suppress SMA might also be included in M14CM. There remains a possibility that SMA-suppressing factor in M14CM is thus closely related but slightly different from PDGF. To characterize the suppressive activity, we used 3Y1 cells because 3Y1 is a stable, well-characterized cell line which constitutively expresses SMA during the growing phase [14]. As for human SMA-positive cells, we used migrated cells from nevus tissue because (1) the tissue contains interstitial cells surrounding the nests of cells of neural crest origin the same as melanoma tissue does, (2) the difference in the SMA content was obvious between the nevi and melanomas in our previous study [8], (3) the nevi are benign regions which can be obtained by surgical resections. The detected amounts of SMA in human migrated cells varied greatly among the culture lots and decreased after repeated passages (data not
176
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
shown), probably because of the different growth rates of the cells [35]. Therefore, these migrated cells were not suitable for carrying out large scale experiments. The population of the SMA-positive cells migrated from the nevi appeared to be heterogeneous. The morphology of the flat, large cells with well-developed actin stress fibers are characteristics of pericytes [36]. The stellate or spindle cells with SMA are similar to the myofibroblasts that appear in early cultures [35,37], although the origin of such myofibroblasts is still unknown. These two cell populations are also main SMApositive cells which have been reported in other tumors [11,21–26]. In addition to these two cell populations, there may be hair follicle dermal cells as SMA-positive cells [10]. Nevocytes are not likely to be a significant population in this culture system, because the growth of the nevocytes requires additional reagents or factors in the culture medium [38]. Perhaps the diversity of the SMA contents among the culture lots may result from such heterogeneity of the migrated cells. SMA has been suggested to be a contractile element because of its characteristic localization [9]. The SMA-suppressive factor(s) may help maintain the immature state of the vessels which have fragile structures and insufficient functions to control the blood supply. In addition, SMAsuppressive factor (s) from melanoma cells possibly inhibit capsule formation around the tumor by suppressing the contractions of myofibroblasts, and thus may be advantageous to the invasion [39]. It has been shown that melanoma cell lines have different degrees of cytokine expressions [31]. We have seen that a considerable diversity existed in the intensity of SMA staining among the patients (data not shown). Infiltrating cells such as lymphocytes and macrophages may have also influenced the SMA suppression by releasing PDGF [40]. The biological significance of these factors inducing the decrease in SMA with special reference to its influence on both tumor behavior and prognosis, however, remains to be investigated. In conclusion, melanoma cells can release factor(s) which alter the cytoskeletal system in sur-
rounding benign, stromal cells by suppressing SMA. Further investigations of these factors may help achieve a better understanding of tumor– stroma interactions.
Acknowledgements This work was supported by a grant from Scientific Research from the Ministry of Education, Science and Culture of Japan (109254104), a grant from the Lydia O’Leary Memorial Foundation.
References [1] Halaban R, Miglarese MR, Smicun Y, et al. Melanomas, from the cell cycle point of view. Int J Mol Med 1998;1:1419 – 25. [2] Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1995;1:27 – 31. [3] Duffy MJ. Proteases as prognostic markers in cancer. Clin Cancer Res 1996;2:613 – 8. [4] Friedman E, Verderame M, Winawer S, et al. Actin cytoskeletal organization loss in the benign-to-malignant tumor transition in cultured human colonic epithelial cells. Cancer Res 1984;44:3040 – 50. [5] Leavitt J, Bushar G, Kakunaga T, et al. Variations in expression of mutant b-actin accompanying incremental increases in human fibroblast tumorigenicity. Cell 1982;28:259 – 68. [6] Sadano H, Taniguchi S, Baba T. Newly identified type of b-actin reduces invasiveness of mouse B16-melanoma. FEBS Lett 1990;271:23 – 7. [7] Shimokawa-Kuroki R, Sadano H, Taniguchi S. A variant actin (bm) reduces metastasis of mouse B16 melanoma. Int J Cancer 1994;56:689 – 97. [8] Taniguchi S, Inoue M, Nakayama J, et al. Differential expression of smooth muscle a-like actin between benign and malignant human pigment tissues. Cancer Lett 1989;47:29 – 36. [9] Skalli O, Pelte M-F, Peclet M-C, et al. a-Smooth muscle actin, a differentiation marker of smooth muscle cells, is present in microfilamentous bundles of pericytes. J Histochem Cytochem 1989;37:315 – 21. [10] Reynolds AJ, Chaponnier C, Jahoda CAB, et al. A quantitative study of the differential expression of asmooth muscle actin in cell populations of follicular and non-follicular origin. J Invest Dermatol 1993;101:577 – 83. [11] Skalli O, Ropraz P, Trzeciak A, et al. A monoclonal antibody against a-smooth muscle actin: a new probe for smooth muscle differentiation. J Cell Biol 1986;103:2787– 96.
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177 [12] Darby I, Skalli O, Gabbiani G. a-Smooth muscle actin is transiently expressed by myofibroblasts during experimental wound healing. Lab Invest 1990;63:21–9. [13] Leavitt J, Gunning P, Kedes L, et al. Smooth muscle a-actin is a transformation-sensitive marker for mouse NIH 3T3 and Rat-2 cells. Nature 1985;316:840–2. [14] Okamoto-Inoue M, Taniguchi S, Sadano H, et al. Alteration in expression of smooth muscle a-actin associated with transformation of rat 3Y1 cells. J Cell Sci 1990;96:631 – 7. [15] Clowes AW, Clowes MM, et al. Arterial smooth muscle cells in vivo: relationship between actin isoform expression and mitogenesis and their modulation by heparin. J Cell Biol 1988;107:1939–45. [16] Blank RS, Thompson MM, Owens GK. Cell cycle versus density dependence of smooth muscle a actin expression in cultured rat aortic smooth muscle cells. J Cell Biol 1988;107:299 – 306. [17] Corjay MH, Tompson MM, Lynch KR, et al. Differential effect of platelet-derived growth factor versus serum-induced growth on smooth muscle a-actin and non-muscle b-actin mRNA expression in cultured rat aortic smooth muscle cells. J Biol Chem 1989;264:10501–6. [18] Blank RS, Owens GK. Platelet-derived growth factor regulates actin isoform expression and growth state in cultured rat aortic smooth muscle cells. J Cell Physiol 1990;142:635 – 42. [19] Tsukamoto H, Mishima Y, Hayashibe K, et al. a-Smooth muscle actin expression in tumor and stromal cells of benign and malignant human pigment cell tumors. J Invest Dermatol 1992;98:116–20. [20] Tsuchida T, Ravindranath MH, Saxton RE, et al. Gangliosides of human melanoma: altered expression in vivo and in vitro. Cancer Res 1987;47:1278–81. [21] Sappino A-P, Skalli O, Jackson B, et al. Smooth muscle differentiation in stromal cells of malignant and non-malignant breast tissues. Int J Cancer 1988;41:707–12. [22] Czernobilsky B, Shezen E, Lifschitz-Mercer B, et al. smooth muscle actin (a-SM actin) in normal human ovaries, in ovarian stromal hyperplasia and in ovarian neoplasms. Virchows Arch B Cell Pathol 1989;57:55–61. [23] Cintorino M, de Marco EB, Leoncini P, et al. Expression of a-smooth-muscle actin in stromal cells of the uterine cervix during epithelial neoplastic changes. Int J Cancer 1991;47:843 – 6. [24] Leoncini P, Cintorino M, Vindigni C, et al. Distribution of cytoskeletal and contractile proteins in normal and tumour bearing salivary and lacrimal glands. Virchows Arch A Pathol Anat Histopathol 1988;412:329–37. [25] Kobayashi H, Tsuruchi N, Sugihara K, et al. Expression of a-smooth muscle actin in benign or malignant ovarian tumors. Gynecol Oncol 1993;48:308–13. [26] Inoue M, Nakayama J, Urabe A, et al. Malignant schwannoma in a case of type 1 neurofibromatosis with decreased immunoreactivity of smooth muscle a-actin in tumor vessels. J Dermatol Sci 1994;8:157–61.
177
[27] Schlingemann RO, Rietveld FJR, Kwaspen F, et al. Differential expression of markers for endothelial cells, pericytes, and basal lamina in the microvasculature of tumors and granulation tissue. Am J Pathol 1991;138:1335– 47. [28] Owens GK, Loeb A, Gordon D, et al. Expression of smooth muscle-specific a-isoactin in cultured vascular smooth muscle cells: relationship between growth and cytodifferentiation. J Cell Biol 1986;102:343 – 52. [29] Desmouliere A, Geinoz A, Gabbiani F, et al. Transforming growth factor-b1 induce a-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol 1993;122:103 – 11. [30] Barnhill RL, Xiao M, Graves D, et al. Expression of platelet-derived growth actor (PDGF)-A, PDGF-B and the PDGF-alpha receptor, but not the PDGF-beta receptor, in human malignant melanoma in vivo. Br J Dermatol 1996;135:898 – 904. [31] Albino AP, Davis BM, Nanus DM. Induction of growth factor RNA expression in human malignant melanoma: markers of transformation. Cancer Res 1991;51:4815 – 20. [32] Ronnov-Jessen L, Petersen OW. Induction of a-smooth muscle actin by transforming growth factor-b1 in quiescent human breast gland fibroblasts. Lab Invest 1993;68:696 – 707. [33] Leof EB, Proper JA, Goustin A, et al. Induction of c-sis mRNA and activity similar to platelet-derived growth factor by transforming growth factor b: a proposed model for indirect mitogenesis involving autocrine activity. Proc Natl Acad Sci USA 1986;83:2453 – 7. [34] Hansson GK, Hellstrand M, Rymo L, et al. Interferon g inhibits both proliferation and expression of differentiation-specific a-smooth muscle actin in arterial smooth muscle cells. J Exp Med 1989;170:1595– 608. [35] Van de Berg JS, Rudolph R, Woodward M. Comparative growth dynamics and morphology between cultured myofibroblasts from granulation wounds and dermal fibroblasts. Am J Pathol 1984;114:187 – 200. [36] Okamoto-Inoue M, Nakayama J, Taniguchi S, et al. Vascularity in neurofibromas demonstrated by immunostaining of smooth muscle a-actin. In: Ishibashi Y, Hori Y, editors. Tuberous Sclerosis and Neurofibromatosis: Epidemiology, Pathopysiology, Biology and Management. Amsterdam: Elsevier, 1990:93 – 9. [37] Fool TH, Naylor IL, Timmons MJ, et al. Intracellular actin as a marker for myofibroblasts in vitro. Lab Invest 1992;67:727 – 33. [38] Morelli JG, Yohn JJ, Zekman T, et al. Melanocyte movement in vitro: role of matrix proteins and integrin receptors. J Invest Dermatol 1993;101:605 – 8. [39] Barsky SH, Gopalakrishna R. Increased invasion and spontaneous metastasis of B16 melanoma with inhibition of the desmoplastic response in C57 BL/6 mice. Cancer Res 1987;47:1663 – 7. [40] Shaw RF, Doherty DE, Ritter AG, et al. Adherence-dependent increase in human monocyte PDGF(B) mRNA is associated with increases in c-fos, c-jun, and EGF2 mRNA. J Cell Biol 1990;111:2139– 48.
Human malignant melanoma cells release a factor that inhibits the expression of smooth muscle a-actin M. Okamoto-Inoue a, J. Nakayama b, Y. Hori c, S. Taniguchi d,* a
Department of Dermatology, Faculty of Medicine, Kurume Uni6ersity, Kurume, Fukuoka, Japan b Department of Dermatology, Faculty of Medicine, Fukuoka Uni6ersity, Fukuoka, Japan c Aso Iizuka Hospital, Iizuka, Fukuoka, Japan d Angio-Research Di6ision, Department of Molecular Oncology and Angiology, Research Center on Aging and Adaptation, Shinshu Uni6ersity School of Medicine, 3 -1 -1 Asahi, Matsumoto 390 -8621, Japan Received 20 October 1999; received in revised form 29 November 1999; accepted 30 November 1999
Abstract Smooth muscle a-actin (SMA) is a cytoskeletal protein expressed in vascular smooth muscle cells, pericytes, hair follicle dermal sheaths and myofibroblasts which appear in the process of wound healing and tumor growth. To examine the effect of malignant melanoma on the expression of SMA in these non-neoplastic cells, we carried out immunohistochemical staining and a cell culture study. Conditioned medium prepared from a melanoma cell line M14 was incubated with a rat fibroblastic cell line 3Y1, which had been shown to express SMA. Human cells that had migrated from nevus tissue were also cultured either with or without M14 conditioned medium. Immuno-histochemical staining of human melanoma tissues suggested that the expression of SMA was low in the vicinity of the tumor as well as within the tumor nodules. The conditioned medium from melanoma, but not the medium from control non-neoplastic cells, suppressed the expression of SMA both in the 3Y1 cells and human cells that migrated from the nevus. Preincubation of the medium with anti-platelet-derived growth factor allowed 76% recovery of SMA expression. These data thus imply that melanoma cells release a platelet-derived growth factor-like substance which has a suppressive effect on the contractile elements in non-neoplastic cells. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Platelet-derived growth factor; Pericyte; Myofibroblast; Cytoskeletal protein
1. Introduction Abbre6iations: Ab, antibody; CM, conditioned medium; PDGF, platelet-derived growth factor; SMA, smooth muscle a-actin. * Corresponding author. Tel.: +81-263-372679; fax: + 81263-372724. E-mail address: [email protected] (S. Taniguchi).
Malignant tumor cells have been shown to produce several substances with biological activities, including autocrine growth factors [1], angiogenic factors [2], and enzymes which degrade matrix proteins [3]. Growth factors are also likely to have paracrine effects which modulate the proliferation
0923-1811/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 9 2 3 - 1 8 1 1 ( 0 0 ) 0 0 0 7 2 - 4
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
and matrix production of the stromal cells. However, little is known regarding the effects of malignant cells on cytoskeletal systems in the surrounding non-neoplastic cells. Actin is one of the major cytoskeletal proteins which plays an essential role in morphogenesis, movement, contraction and the attachment of cells. The disorganization of actin stress fibers is a common feature of malignant cells [4] and recent advances in molecular biology have shown that the introduction of a mutant b isoform of actin into cells increases tumorigenicity [5], while another type of variant b actin reduces invasiveness [6] and metastatic ability [7]. Thus, an altered expression of actin isoforms may lead to changes cellular agressiveness. We [8] have shown that melanoma tissues display less smooth muscle a-actin (SMA) compared with benign tissues, whereas the b- and g-cytoplasmic actins were not altered. Among the several isoforms of actin so far identified in benign human cells, SMA is expressed specifically in vascular smooth muscle cells, pericytes [9], a specific region of hair follicle dermal sheath [10], myoepithelial cells [11] and myofibroblasts in both wound healing [12] and tumors [11]. SMA has also been detected in embryonic, and fibroblastic cell lines such as the mouse 3T3 [13] and rat 3Y1 [14]. SMA expression is (i) tightly regulated during the cell cycle process, differentiation and transformation; (ii) expressed transiently in myofibroblasts during wound healing [12]; (iii) lost with viral transformation of 3Y1 and 3T3 cells [13,14]; decreased when aortic smooth muscle cells move into the G0/G1 during initial thickening after endothelial injury [15]; suppressed by the mitogen platelet-derived growth factor (PDGF) in cultured aortic smooth muscle cells [16 – 18]. Our immunohistochemical findings of melanoma tissue led us to examine the effect of a culture supernatant of melanoma cells on the expression of SMA in normal, SMA-positive cells. As a model of myofibroblast, we used a rat embryonic, and fibroblastic cell line 3Y1, which is a stable cell line and has been characterized to express SMA constitutively under the culture condition in 10% fetal bovine serum (FBS) [14].
171
2. Materials and methods
2.1. Immunohistochemistry Monoclonal anti-a-smooth muscle actin antibody was purchased from Sigma. Formalin-fixed, paraffin-embedded tissue sections of human malignant melanoma were stained using an avidin– biotin–alkaline phospatase system (Vectastain ABC-AP) and a substrate kit (Vector Red) as described by the manufacturer.
2.2. Cell culture The human melanoma cell line UCLA M14 (M14) has been described elsewhere [20]. Melanoma cells were cultured in Eagle’s minimum essential medium (MEM) supplemented with 10% fetal bovine serum (FBS) and 80 mg/ml Kanamycin. The medium was changed twice a week. Explant cultures of human fibroblastic cells were carried out from the excised tissue of pigmented nevi. A part of each tissue was then fixed in formalin for the histological diagnosis. For the culture, the rest of the tissue was rinsed three times in PBS containing 1000 U/ml penicillin and 800 mg/ml Kanamycin. The tissue was then cut into 1 mm3 pieces, and cultured in Petri dishes in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS and antibiotics. When migrated cells were grown in a subconfluent state, the cells were trypsinized and transferred. These human cells were then used in the second passages for the assays of SMA expression. Conditioned medium from subconfluent M14 cells (M14CM) filtered (0.22 mm Milex GV filter, Millipore Intertech, Bedford, MA) and were stored at −80°C until use. For controls, CM from benign cells was prepared, similarly from 3Y1 cells, 3T3 cells or human fibroblasts cultured in MEM supplemented with 10% FBS and antibiotics. In order to assess the effect of M14CM, the culture medium of 3Y1 cells in a growing phase was replaced with either 66% M14CM or control 66% CM in fresh medium and then the cells were cultured for the indicated period. Human fibroblastic cells cultured from human pigmented nevus
172
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
tissue were also treated with either M14CM or control CM for 48 h in the same way as the treatment of 3Y1. In some experiments, M14CM and control CM were pretreated for 30 min at 37°C either with or without anti-human PDGF Antibody (R&D systems, Minneapolis, MN) reactive with both PDGF-BB and AB. The final concentrations of anti-human PDGF antibody were 0, 37.5, 75, 150 mg/ml. Human platelet PDGF-AB (Wako Pure Chemical Industries, Ltd, Osaka, Japan) was diluted with PBS containing 2 mg/ml bovine serum albumin (BSA, fraction IV). Fifty microliters of PDGF solution (final concentration 5 ng/ml) or vehicle was then added to fresh medium and preincubated with anti-PDGF antibody (70, 35 or 0 mg/ml) for 30 min at 37°C. There are several reports that human PDGF plays biological functions on rat cells through the PDGF receptor [20]. Then 3Y1 cells were treated for 24 h with the above preincubated medium.
2.3. Immunofluorescent staining of cultured cells SMA in cultured cells was stained as described previously [14]. Briefly, the cells were cultured on glass coverslips overnight and fixed in ice-cold methanol for 5 min, dried and then incubated either with monoclonal anti-SMA (1:500) or with mouse control ascites fluid (Cederlane Laboratories, Ontario, Canada) for 30 min at room temperature. The cells were washed in PBS three times, incubated with FITC-labeled anti-mouse IgG (Daco, Glostrup, Denmark) (1:250) for 30 min, and then washed. The stained cells were mounted with glycerol and observed with a Zeiss Axiophoto microscope.
2.4. SDS –PAGE and immunoblotting The cultured cells in each dish were washed with PBS and harvested with a rubber policeman. The cells were collected in a 1.5-ml microcentrifuge tube and sonicated in PBS containing 1% sodium dodecyl sulphate (SDS). The protein concentration was measured with a BCA protein assay kit (Pierce, IL) using BSA as a standard. The samples were adjusted to a 5 mg/ml protein
concentration in SDS-sample buffer and then were separated on SDS–PAGE. The concentration of acrylamide was 10%. After electrophoresis the proteins were electroblotted to a nitrocellulose filter and stained with a 1:1000 dilution of monoclonal anti-SMA as described previously [8]. The relative intensities of the bands on the nitrocellulose filter were compared densitometrically using a Dual Wavelength Flying-spot scanner CS-9000 (Shimadzu Co., Kyoto, Japan) at a wavelength of 590 nm.
3. Results
3.1. Immunohistochemical staining of SMA The cells expressing SMA were screened in the melanoma tissue using immunohistochemical staining. Fig. 1 shows the staining of SMA in a representative tumor operated from the face of an 84-year-old man. The localization of the melanoma was recognized as a pigment-laden tumor mass (Fig. 1a). SMA was positive in the vessel walls of the capillaries, venules and arterioles surrounding the tumor. No myofibloblasts were seen in this case. No melanoma cells were stained with SMA. In the close vicinity of the tumor, the SMA in the vessel walls was often discontinuous and obscure. Within the tumor nodule, SMA was occasionally seen but the staining was faint. In a higher magnification (Fig. 1b), the capillaries were recognized within the tumor but the SMA in those capillaries was very faint.
3.2. Effect of M14CM on SMA le6el in 3Y1 cells The constitutive expression of SMA in 3Y1 was confirmed under the culture condition with 10% fetal bovine serum as reported previously [14] (data not shown). The incubation of M14CM for 24 h decreased the SMA in 3Y1 cells compared with that incubated with 3Y1CM (Fig. 2a). The conditioned medium from other benign cells such as 3T3 and human fibroblasts did not change the amount of SMA protein either (data not shown). Fig. 2b shows the time course of the suppression of SMA by M14CM. SMA expression was con-
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
173
stant until 12 h after incubation with M14CM and was inhibited completely SMA in the 3Y1 cells from 18 h up to 2 days. SMA expression started to recover 3 days after the incubation with M14CM. The control 3T3CM showed no change in the SMA level for 3 days.
Fig. 2. Immunoblot of SMA in 3Y1 cells after incubation with M14CM. (a) 3Y1 cells 24 h after incubation with control 3Y1 CM (lane 1) and M14CM (lane 2). (b) 3Y1 cells were extracted before (lane 1) or 3 h (lane 2), 6 h (lane 3), 12 h (lane 4), 18 h (lane 5), 24 h (lane 6), 2 days (lane 7) and 3 days (lane 8) after incubation with M14CM. Control 3Y1 cells were incubated with 3T3CM for 3 days (lane 9). Fifty micrograms of the extracted protein was applied in each lane.
3.3. Reco6ery of SMA by anti-PDGF Ab
Fig. 1. Immunohistochemical detection of SMA in malignant melanoma by avidin – biotin–alkaline phosphatase method. (a) Marginal area of a nodular melanoma; (b) tumor nodule, counterstained with hematoxylin. Arrows indicate faintly-positive vessel walls. Bars: (a) 100 mm; (b) 20 mm.
Because PDGF is a known suppresser against the expression of SMA in aortic smooth muscle cells [16–18], we examined the effect of preincubation of anti-PDGF blocking Ab with M14CM. As shown in Fig. 3a, SMA recovered to up to 76% of the control level by the preincubation with anti-PDGF antibody. On the other hand, the SMA-suppressing activity of the pure PDGF was blocked completely by the addition of the antibody (Fig. 3b).
174
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
3.4. Demonstration of SMA in human cells that migrated from the ne6us tissue Two types of the SMA-positive migrating from the nevus could be morphologically distinguished: one type consisted of large, polygonal cells with wide-spread cytoplasm and well-developed actin stress fibers, while the other type had a stellate or spindle-shaped cytoplasm (Fig. 4).
Fig. 4. Immunofluorescence staining of SMA in the cells migrated from human pigmented nevus tissue. Bar: 50 mm.
Fig. 5. Immunoblot analysis concerning effect of M14CM on SMA in the migrated fibroblastic cells cultured from human pigmented nevus tissue. Immunoblot. Each dish of primary culture was evenly transferred into two dishes, then one dish was added with 3Y1CM (lanes 1 and 3), while another was added with M14CM (lanes 2 and 4). Lanes 1 and 2: case 1; lanes 3 and 4: case 2.
Fig. 3. Effect of anti-PDGF antibody on the SMA-suppressive activity of M14CM and PDGF. M14CM (a) or PDGF (b) was preincubated with antibody and added to 3Y1 cells. a, bands of SMA on immunoblot were densitometrically quantified. (b) Immunoblot. Cells were incubated with: vehicle (lane 1), 5 ng/ml PDGF (lane 2), 5 ng/ml PDGF and 70 mg/ml antiPDGF antibody (lane 3), or 5 ng/ml PDGF and 35 mg/ml anti-PDGF antibody (lane 4).
3.5. Effect of M14CM on SMA in the human migrated cells Fig. 5 shows the SMA levels in the human migrated cells after the addition of M14CM or control CM. Although the amount of SMA differs among the culture lots of the cells, M14CM
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
diminished SMA in each lot when compared with the paired control dish. 4. Discussion Our immunohistochemical study showed that main sources of SMA in malignant melanoma were vascular smooth muscle cells and pericytes, which was consistent with the report by Tsukamoto et al. [19]. In our staining data with primary tumors, fewer myofibroblasts were detected than in cases with breast carcinomas [21], ovarian tumors [22], uterine cervix cancer [23] and tumors of the salivary glands [24]. The staining also demonstrated a decreased expression of SMA in the pericytes within the tumor cell nest and in the area adjacent to the nest. These findings are also consistent with our reports on the immunohistochemical staining of malignant and benign tumors [25,26]. In support of these findings, our previous electrophoretic analysis [8] showed a relatively low amount of SMA in the tissue of malignant melanoma compared with that from nevus pigmentosus and other types of benign pigmented tissue. The decrease in SMA is not an indicate of less pericytes, because there were weakly positive cells within the tumor cell nests. However, it has been suggested that tumors have immature microvessels after neoangiogenesis and that those immature vessels lack SMA in pericytes [27]. These considerations therefore prompted us to carry out suppression experiments in vitro. In the present study, we showed the suppression of SMA in benign 3Y1 cells by a culture supernatant of melanoma. SMA in human cells cultured from the skin also decreased after treatment with M14CM. Such a suppression of SMA in the cultured cells was the result of an inhibitory factor for (1) the control CM prepared from either 3Y1 or human fibroblasts did not suppress SMA, (2) when a lower concentration of FBS is used in the medium, SMA expression is substantially enhanced[28,29], (3) the suppression was reversible and the expression of SMA gradually recovered after 3 days and (4) the suppression activity was blocked by the addition of anti-PDGF antibody. These findings thus indicate that SMA-suppressing activity exists in the M14CM.
175
Several human melanoma cell lines and melanoma in vivo have been shown to express A and B chains of PDGF [30], transforming growth factor (TGF-b), basic fibroblast growth factor, acidic fibroblast growth factor and keratinocyte growth factor [31]. Because the effects of PDGF on SMA in cultured aortic smooth muscle cells have been well-characterized [16–18], we thus expected anti-PDGF antibody to be a candidate of a blocking reagent against a factor which suppresses SMA. In our blocking experiments, this antibody completely blocked the suppressive activity of human PDGF while the activity of M14CM was blocked partially (76%). The partial block may be due to the heterogeneity in PDGF subunits contained in the M14CM. The human PDGF used in this experiment was AB heterodimer and the antibody raised against human PDGF demonstrated cross-reactivity to a BB homodimer with different affinity according to the manufacturer. Other cytokines in M14CM may also influence the suppression of SMA. For example, growth stimulation also diminishes the SMA content in vascular smooth muscle cells [15,16]. Although TGF-b by itself stimulates the SMA expression in serum-free systems [29,32], it can also enhance the effects of PDGF by inducing autocrine PDGF production [33]. Interferongamma [34] or other unknown factors which suppress SMA might also be included in M14CM. There remains a possibility that SMA-suppressing factor in M14CM is thus closely related but slightly different from PDGF. To characterize the suppressive activity, we used 3Y1 cells because 3Y1 is a stable, well-characterized cell line which constitutively expresses SMA during the growing phase [14]. As for human SMA-positive cells, we used migrated cells from nevus tissue because (1) the tissue contains interstitial cells surrounding the nests of cells of neural crest origin the same as melanoma tissue does, (2) the difference in the SMA content was obvious between the nevi and melanomas in our previous study [8], (3) the nevi are benign regions which can be obtained by surgical resections. The detected amounts of SMA in human migrated cells varied greatly among the culture lots and decreased after repeated passages (data not
176
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177
shown), probably because of the different growth rates of the cells [35]. Therefore, these migrated cells were not suitable for carrying out large scale experiments. The population of the SMA-positive cells migrated from the nevi appeared to be heterogeneous. The morphology of the flat, large cells with well-developed actin stress fibers are characteristics of pericytes [36]. The stellate or spindle cells with SMA are similar to the myofibroblasts that appear in early cultures [35,37], although the origin of such myofibroblasts is still unknown. These two cell populations are also main SMApositive cells which have been reported in other tumors [11,21–26]. In addition to these two cell populations, there may be hair follicle dermal cells as SMA-positive cells [10]. Nevocytes are not likely to be a significant population in this culture system, because the growth of the nevocytes requires additional reagents or factors in the culture medium [38]. Perhaps the diversity of the SMA contents among the culture lots may result from such heterogeneity of the migrated cells. SMA has been suggested to be a contractile element because of its characteristic localization [9]. The SMA-suppressive factor(s) may help maintain the immature state of the vessels which have fragile structures and insufficient functions to control the blood supply. In addition, SMAsuppressive factor (s) from melanoma cells possibly inhibit capsule formation around the tumor by suppressing the contractions of myofibroblasts, and thus may be advantageous to the invasion [39]. It has been shown that melanoma cell lines have different degrees of cytokine expressions [31]. We have seen that a considerable diversity existed in the intensity of SMA staining among the patients (data not shown). Infiltrating cells such as lymphocytes and macrophages may have also influenced the SMA suppression by releasing PDGF [40]. The biological significance of these factors inducing the decrease in SMA with special reference to its influence on both tumor behavior and prognosis, however, remains to be investigated. In conclusion, melanoma cells can release factor(s) which alter the cytoskeletal system in sur-
rounding benign, stromal cells by suppressing SMA. Further investigations of these factors may help achieve a better understanding of tumor– stroma interactions.
Acknowledgements This work was supported by a grant from Scientific Research from the Ministry of Education, Science and Culture of Japan (109254104), a grant from the Lydia O’Leary Memorial Foundation.
References [1] Halaban R, Miglarese MR, Smicun Y, et al. Melanomas, from the cell cycle point of view. Int J Mol Med 1998;1:1419 – 25. [2] Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1995;1:27 – 31. [3] Duffy MJ. Proteases as prognostic markers in cancer. Clin Cancer Res 1996;2:613 – 8. [4] Friedman E, Verderame M, Winawer S, et al. Actin cytoskeletal organization loss in the benign-to-malignant tumor transition in cultured human colonic epithelial cells. Cancer Res 1984;44:3040 – 50. [5] Leavitt J, Bushar G, Kakunaga T, et al. Variations in expression of mutant b-actin accompanying incremental increases in human fibroblast tumorigenicity. Cell 1982;28:259 – 68. [6] Sadano H, Taniguchi S, Baba T. Newly identified type of b-actin reduces invasiveness of mouse B16-melanoma. FEBS Lett 1990;271:23 – 7. [7] Shimokawa-Kuroki R, Sadano H, Taniguchi S. A variant actin (bm) reduces metastasis of mouse B16 melanoma. Int J Cancer 1994;56:689 – 97. [8] Taniguchi S, Inoue M, Nakayama J, et al. Differential expression of smooth muscle a-like actin between benign and malignant human pigment tissues. Cancer Lett 1989;47:29 – 36. [9] Skalli O, Pelte M-F, Peclet M-C, et al. a-Smooth muscle actin, a differentiation marker of smooth muscle cells, is present in microfilamentous bundles of pericytes. J Histochem Cytochem 1989;37:315 – 21. [10] Reynolds AJ, Chaponnier C, Jahoda CAB, et al. A quantitative study of the differential expression of asmooth muscle actin in cell populations of follicular and non-follicular origin. J Invest Dermatol 1993;101:577 – 83. [11] Skalli O, Ropraz P, Trzeciak A, et al. A monoclonal antibody against a-smooth muscle actin: a new probe for smooth muscle differentiation. J Cell Biol 1986;103:2787– 96.
M. Okamoto-Inoue et al. / Journal of Dermatological Science 23 (2000) 170–177 [12] Darby I, Skalli O, Gabbiani G. a-Smooth muscle actin is transiently expressed by myofibroblasts during experimental wound healing. Lab Invest 1990;63:21–9. [13] Leavitt J, Gunning P, Kedes L, et al. Smooth muscle a-actin is a transformation-sensitive marker for mouse NIH 3T3 and Rat-2 cells. Nature 1985;316:840–2. [14] Okamoto-Inoue M, Taniguchi S, Sadano H, et al. Alteration in expression of smooth muscle a-actin associated with transformation of rat 3Y1 cells. J Cell Sci 1990;96:631 – 7. [15] Clowes AW, Clowes MM, et al. Arterial smooth muscle cells in vivo: relationship between actin isoform expression and mitogenesis and their modulation by heparin. J Cell Biol 1988;107:1939–45. [16] Blank RS, Thompson MM, Owens GK. Cell cycle versus density dependence of smooth muscle a actin expression in cultured rat aortic smooth muscle cells. J Cell Biol 1988;107:299 – 306. [17] Corjay MH, Tompson MM, Lynch KR, et al. Differential effect of platelet-derived growth factor versus serum-induced growth on smooth muscle a-actin and non-muscle b-actin mRNA expression in cultured rat aortic smooth muscle cells. J Biol Chem 1989;264:10501–6. [18] Blank RS, Owens GK. Platelet-derived growth factor regulates actin isoform expression and growth state in cultured rat aortic smooth muscle cells. J Cell Physiol 1990;142:635 – 42. [19] Tsukamoto H, Mishima Y, Hayashibe K, et al. a-Smooth muscle actin expression in tumor and stromal cells of benign and malignant human pigment cell tumors. J Invest Dermatol 1992;98:116–20. [20] Tsuchida T, Ravindranath MH, Saxton RE, et al. Gangliosides of human melanoma: altered expression in vivo and in vitro. Cancer Res 1987;47:1278–81. [21] Sappino A-P, Skalli O, Jackson B, et al. Smooth muscle differentiation in stromal cells of malignant and non-malignant breast tissues. Int J Cancer 1988;41:707–12. [22] Czernobilsky B, Shezen E, Lifschitz-Mercer B, et al. smooth muscle actin (a-SM actin) in normal human ovaries, in ovarian stromal hyperplasia and in ovarian neoplasms. Virchows Arch B Cell Pathol 1989;57:55–61. [23] Cintorino M, de Marco EB, Leoncini P, et al. Expression of a-smooth-muscle actin in stromal cells of the uterine cervix during epithelial neoplastic changes. Int J Cancer 1991;47:843 – 6. [24] Leoncini P, Cintorino M, Vindigni C, et al. Distribution of cytoskeletal and contractile proteins in normal and tumour bearing salivary and lacrimal glands. Virchows Arch A Pathol Anat Histopathol 1988;412:329–37. [25] Kobayashi H, Tsuruchi N, Sugihara K, et al. Expression of a-smooth muscle actin in benign or malignant ovarian tumors. Gynecol Oncol 1993;48:308–13. [26] Inoue M, Nakayama J, Urabe A, et al. Malignant schwannoma in a case of type 1 neurofibromatosis with decreased immunoreactivity of smooth muscle a-actin in tumor vessels. J Dermatol Sci 1994;8:157–61.
177
[27] Schlingemann RO, Rietveld FJR, Kwaspen F, et al. Differential expression of markers for endothelial cells, pericytes, and basal lamina in the microvasculature of tumors and granulation tissue. Am J Pathol 1991;138:1335– 47. [28] Owens GK, Loeb A, Gordon D, et al. Expression of smooth muscle-specific a-isoactin in cultured vascular smooth muscle cells: relationship between growth and cytodifferentiation. J Cell Biol 1986;102:343 – 52. [29] Desmouliere A, Geinoz A, Gabbiani F, et al. Transforming growth factor-b1 induce a-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol 1993;122:103 – 11. [30] Barnhill RL, Xiao M, Graves D, et al. Expression of platelet-derived growth actor (PDGF)-A, PDGF-B and the PDGF-alpha receptor, but not the PDGF-beta receptor, in human malignant melanoma in vivo. Br J Dermatol 1996;135:898 – 904. [31] Albino AP, Davis BM, Nanus DM. Induction of growth factor RNA expression in human malignant melanoma: markers of transformation. Cancer Res 1991;51:4815 – 20. [32] Ronnov-Jessen L, Petersen OW. Induction of a-smooth muscle actin by transforming growth factor-b1 in quiescent human breast gland fibroblasts. Lab Invest 1993;68:696 – 707. [33] Leof EB, Proper JA, Goustin A, et al. Induction of c-sis mRNA and activity similar to platelet-derived growth factor by transforming growth factor b: a proposed model for indirect mitogenesis involving autocrine activity. Proc Natl Acad Sci USA 1986;83:2453 – 7. [34] Hansson GK, Hellstrand M, Rymo L, et al. Interferon g inhibits both proliferation and expression of differentiation-specific a-smooth muscle actin in arterial smooth muscle cells. J Exp Med 1989;170:1595– 608. [35] Van de Berg JS, Rudolph R, Woodward M. Comparative growth dynamics and morphology between cultured myofibroblasts from granulation wounds and dermal fibroblasts. Am J Pathol 1984;114:187 – 200. [36] Okamoto-Inoue M, Nakayama J, Taniguchi S, et al. Vascularity in neurofibromas demonstrated by immunostaining of smooth muscle a-actin. In: Ishibashi Y, Hori Y, editors. Tuberous Sclerosis and Neurofibromatosis: Epidemiology, Pathopysiology, Biology and Management. Amsterdam: Elsevier, 1990:93 – 9. [37] Fool TH, Naylor IL, Timmons MJ, et al. Intracellular actin as a marker for myofibroblasts in vitro. Lab Invest 1992;67:727 – 33. [38] Morelli JG, Yohn JJ, Zekman T, et al. Melanocyte movement in vitro: role of matrix proteins and integrin receptors. J Invest Dermatol 1993;101:605 – 8. [39] Barsky SH, Gopalakrishna R. Increased invasion and spontaneous metastasis of B16 melanoma with inhibition of the desmoplastic response in C57 BL/6 mice. Cancer Res 1987;47:1663 – 7. [40] Shaw RF, Doherty DE, Ritter AG, et al. Adherence-dependent increase in human monocyte PDGF(B) mRNA is associated with increases in c-fos, c-jun, and EGF2 mRNA. J Cell Biol 1990;111:2139– 48.