sign in sign up
<<
Home , Fibroblast growth factor, TGF alpha

in vivo 19: 599-604 (2005)

Influence of Hepatocyte /Scatter Factor (HGF/SF) on Growth Factor-2 (FGF-2) Levels in External Auditory Canal Cholesteatoma (EACC) Cell Culture

RAMIN NAIM1, RAY C. CHANG2, HANEEN SADICK1 and KARL HORMANN1

1Department of Otolaryngology, Head and Neck Surgery, University Hospital Mannheim, D-68135 Mannheim, Germany; 2Department of Otolaryngology, University of Miami/Jackson Memorial Hospital, Miami, Florida, U.S.A.

Abstract. Background: In previous studies, we cited have been identified, including fibroblast circulatory disorders and hypoxia as etiological factors for the growth factor-a (aFGF), transforming growth factor-alpha formation of external auditory canal cholesteatoma (EACC) (TGF-alpha), TGF-beta, /scatter resulting in angiogenesis. Here, we investigate how the factor (HGF/SF), tumor necrosis factor-alpha (TNF-alpha), angiogenic factor hepatocyte growth factor/scatter factor angiogenin and -8 (IL-8) (3, 4). (HGF/SF) influences the level of another angiogenic factor Fibroblast growth factors (FGFs) are also considered FGF-2. Materials and Methods: After 16 to 72 hours of angiogenic factors, yet the exact relationship between FGF incubation with 20ng/ml HGF/SF, levels of VEGF in the and vascular development in normal and pathological tissue HGF/SF-treated and untreated culture was analyzed. We also has long remained elusive (5). FGF-2 is a member of the investigated the influence of HGF/SF (20-80ng/ml) on the FGF family, that comprises about nine members. FGF-2 concentration of FGF-2. Results: After 16 hours of incubation stimulates smooth muscle , , tissue with HGF/SF at 20ng/ml, FGF-2 was measured at 44.19pg/ml repair, and is increased in chronic inflammation (5). FGF-2 (control: 42.24pg/ml). After 72 hours, FGF-2 was at regulates the expression of several molecules including 23.41pg/ml (control: 14.83pg/ml). After 24 hours, 40ng/ml interstitial collagenase, urokinase type plasminogen HGF/SF showed the highest concentration of FGF-2. activator (uPA), plasminogen activator inhibitor (PAI-1), Conclusion: FGF-2 levels were initially elevated after treatment uPA receptor and (6). For a long time, FGF-2 was with HGF/SF, however, further incubation did not show any considered as the angiogenesis factor. This was challenged increase. We assume that HGF/SF released FGF-2 in the by the discovery of several other angiogenic factors, such as matrix but did not induce FGF-2 expression. The most the vascular endothelial growth factor (VEGF), HGF/SF effective HGF/SF concentration was 40ng/ml. and its receptor c-Met (7). HGF/SF was initially discovered as a secreted by Angiogenesis, the formation of new blood vessels within resulting in epithelial/cell mitogenesis, tissue, has been an exciting and promising new area of proliferation, morphogenesis and motogenesis (8, 9). research. During angiogenesis, endothelial cell growth, HGF/SF is a 100 kDa pleiotropic mitogenic growth factor migration and vascular tube formation are regulated by pro- and its role in establishing angiogenesis, organ regeneration, and anti-angiogenic factors, matrix-degrading proteases and development of the nervous system and tumor invasion cell– interactions (1). Matrix-bound emphasizes its important role in physiological and growth factors released by proteases and/or by angiogenic pathological processes (10). HGF/SF transduces signals via factors promote different diseases by enhancing endothelial its receptor, c-Met, a transmembrane tyrosine proto- migration and growth (2). Several potential regulators of oncogene (11). In the field of otolaryngology, the external auditory canal cholesteatoma (EACC) is a very rare pathological entity (12). EACC is usually found in the inferior or posterior part of the Correspondence to: Dr. med. Ramin Naim, Univ.-HNO-Klinik, bony external ear canal, and it is characterized by permanent Theodor-Kutzer-Ufer, 68135 Mannheim, Germany. Tel: +49 621 growth of the epithelial tissue (13). Establishing the 383 1600, Fax: +49 621 383 1972, e-mail: [email protected] heidelberg.de cholesteatoma, the epithelial layer is the so-called matrix, and the adjacent subepithelial tissue is called perimatrix. The Key Words: External auditory canal cholesteatoma, FGF-2, matrix contains hyperplastic epithelial tissue, which HGF/SF, culture, . destructively tends to grow into adjacent anatomical structures

0258-851X/2005 $2.00+.40 599 in vivo 19: 599-604 (2005) such as the bony canal and mastoid cells, in contrast to immunoassay technique. A monoclonal antibody specific for FGF-2 keratosis obturans (14). The treatment of choice remains total was pre-coated onto a microplate. Standards and samples were surgical removal of the chronic inflammatory tissue (15). In pipetted into wells and FGF-2 present was bound by the immobilized antibody. After washing away unbound substances, an previous immunohistochemical studies, we showed that the enzyme-linked polyclonal antibody specific for FGF-2 was added growth factors VEGF, FGF-2, HGF/FS and the receptor to the wells. Following a wash to remove unbound antibody- c-Met are up-regulated in EACC (16). Sudhoff et al. reported enzyme reagent, a substrate solution was added to the wells and that angiogenesis is a pivotal factor for the formation of middle color developed in proportion to the amount of FGF-2 bound in ear cholesteatoma (17). They showed an altered expression the initial step. The color development was stopped and the and distribution of VEGF, FGF-2, TGF-· and TGF-‚1 in intensity of the color measured. middle ear cholesteatoma in relation to middle ear mucosa and The cells were grown in 96-well plates (Part 890218) with 12 strips of 8 walls coated with a mouse monoclonal antibody against auditory meatal . Further, Adamczyk et al. showed, by FGF-2. After 16, 24, 48 and 72 hours of incubation with 20 ng/ml immunohistochemistry, that EACC has enhanced expression HGF/SF, the expression of the FGF-2 protein in the supernants of of TGF-·, another potent angiogenic factor (18). the HGF/SF-treated and untreated culture cell lines was analyzed. Many authors reported that growth of several tumors is Concurrently, cultured EACC were incubated with different levels induced by HGF/SF, c-Met, VEGF and FGF-2 (5, 10, 11). of HGF/SF (20-80 ng/ml) for 24 hours. The assay recognized all The aim of this study was to explore the effect of HGF/SF different forms of FGF-2 (Figure 1a). on the matrix tissue of EACC with respect to the Western blot analysis. For Western blot analysis, the protein extracts concentration of FGF-2. To the best of our knowledge, we of the cholesteatoma cell culture were separated in a Tris-glycine are the first to grow EACC in vitro by removing gel (Invitrogen, Carlsbad, CA, USA) and transferred onto a mesenchymal tissue including fibroblasts from purified polyvinylidene difluoride membrane. Western blot analysis was endothelial cell culture (19). We present data that HGF/SF performed, as described by Zhang (20). influences the FGF-2 concentration, which might be important in the pathogenesis of EACC. Statistical analysis. The statistical analysis was calculated according to Bortz et al. (21). This method enabled the calculation of the p-value (p<0.05) for our small collective and data. Materials and Methods Results Tissue collection, isolation and culture of human skin and cholesteatoma culture. All EACC cell cultures were used at passage 3. Incubation up to 72 hours with 20 ng HGF/SF. After 16 hours All cells were obtained from 5 patients undergoing reconstructive of incubation with HGF/SF, FGF-2 was measured at 44.19 ear canal surgery at the Department of Otorhinolaryngology of the University of Mannheim, Germany. (Prior to surgery, written pg/ml. In untreated EACC (control), the level of FGF-2 was consent was obtained from all patients to take tissue samples of the 42.24 pg/ml. After 24 hours, the treated EACC culture resected cholesteatoma. This study was approved by the Ethics showed 46.36 pg/ml, while the control was 36.52 pg/ml. committee of the Faculty of Clinical Medicine, Mannheim, After 48 hours, the difference between the treated EACC University of Heidelberg, Germany). After removal of connective and control was less than 13% (24.79 pg/ml and 28.42 pg/ml, tissue, skin and cholesteatoma samples were cut into small pieces respectively). Cells incubated for 72 hours with HGF/SF had and incubated in a trypsin solution (0.25% trypsin in phosphate- 23.41 pg/ml FGF-2, whereas in untreated EACC culture buffered saline, PBS) overnight at 4ÆC. For primary culture of keratinocytes and cholesteatoma cells, the suspension of epidermal 14.83 pg/ml was measured. The proliferation assay, based cells was added onto mitomycin-treated (23.9 ÌM) human fibroblast on determination of total protein (Bradford), showed a slow monolayers and cultured in FAD2-medium (Dulbecco’s Modified growth of the EACC culture (311 Ìg/ml to 414 Ìg/ml). Eagle Medium and Ham’s F12 in a 3:1 ratio supplemented with fetal Comparing total protein content and the fraction of FGF-2 calf serum, adenine, , triiodothyronin hydrocortisone, at different intervals of time, the concentration of FGF-2 , cholera toxin and penicillin/streptomycin decreased in untreated EACC; however, HGF/SF-incubated at 37ÆC in a 10% CO atmosphere). On reaching subconfluency, the 2 EACC culture showed an increase of FGF-2 protein feeder layer was removed by incubation with 0.02% ethylenediamine tetraacetic acid (EDTA) in PBS for 4 min at 37ÆC and the concentration as a fraction of total protein content. There cholesteatoma cells were further cultured in keratinocyte growth was a decrease of FGF-2 after 48 hours, which stagnated up medium (KGM, Clonetics, San Diego, CA, USA) without serum. to 72 hours after incubation with HGF/SF. The untreated Cells were passaged by trypsinization (0.1% trypsin and 0.02% culture showed a continuing decrease of FGF-2 EDTA dissolved in PBS, 5 min, 37ÆC). Cholesteatoma cells of concentration (Figure 1b). passage 3 were used for all experiments. The immunohistochemical staining showed a strong reactivity of the EACC matrix to FGF-2, whereas, HGF/SF FGF-2-ELISA. We used the Quatikine Human FGF-2 Immunoassay from R&D Systems (Minneapolis, MN, USA) to was only positive in the mesenchymal perimatrix. Western determine the exact concentration of FGF-2 in the cell culture blot analysis for HGF/SF showed a clear band in EACC supernants. The assay employed the quantitative sandwich enzyme compared to normal auditory skin.

600 Naim et al: HGF/SF and FGF-2

Figure 1. a. Experimental overview: A) Incubation of EACC cell culture with 20ng/ml HGF/SF and measurement of FGF- 2 levels up to 72 hours. B) Incubation of EACC cell culture with different levels of HGF/SF (20-80 ng/ml) and measurement of FGF-2 levels after 24 hours. b. FGF-2 levels of incubated EACC cell culture (black) and control (grey). c. Interpretation of the results: possible release of FGF-2 storage induced by HGF/SF or by another protein, which is influenced by HGF/SF.

601 in vivo 19: 599-604 (2005)

Incubation with different levels of HGF/SF for 24 hours. After was no increase of FGF-2 levels at any time. According to 24 hours, in the untreated culture FGF-2 was measured at Bortz et al., the difference between incubated EACC cells 36.52 ng/ml. Incubation with 20 ng/ml HGF/SF resulted in and control was statistically significant (p<0.05) after 24 and 43.27 ng/ml of FGF-2 (16% difference from control), while 72 hours (21). This observation can be interpreted as a 40 ng/ml HGF/SF showed a concentration of 47.47 ng/ml of possible release of FGF-2 storage induced by HGF/SF or by FGF-2 (23% difference from control). Sixty and 80 ng/ml another protein, which is influenced by HGF/SF. It can be HGF/SF resulted in 41.78 and 45.6 ng/ml, respectively concluded that there is a link, which connects the signalling (17.5% and 20%). pathways of FGF-2 and HGF/SF (Figure 1c). However, the mechanism of FGF-2 release remains to be further Discussion investigated. FGF-2 has no signalling sequence and is not released by classic signal sequence pathway (5). It is In recent studies, we discussed that hypoxia is possibly the assumed that FGF-2 is released from cells through passive main etiological factor for the formation of EACC (16). We processes such as cell death, wounding, or chemical injury showed that expression of different angiogenic factors (25). In the EACC culture, none of these releasing factors (VEGF, c-Met, HGF/SF and FGF-2) was increased in are evident. Consequently, the levels of FGF-2 were EACC specimens. We also showed that HGF/SF decreased after 72 hours. We assume that HGF/SF significantly increased the levels of VEGF in cultured EACC represents a releasing factor. Still, with the EACC culture (19). This study was designed to elucidate the significant co- model we could not determine the connection between expression of FGF-2 and HGF/SF in EACC, which was HGF/SF and FGF-2 release. observed in recent studies (16). Onimaru et al. demonstrated Totani et al. suggested that endothelial cells exposed to that FGF-2 directly stimulated HGF mRNA without the leukocytes release FGF-2 by an undefined mechanism (26). requirement of new protein synthesis in mesenchymal tissue. In EACC, chronic inflammation and migration of leukocytes They suggested that there was a regulational mechanism of in the matrix consisting of epithelial tissue, but also into the HGF expression via FGF-2 independent of the presence of sub-epithelial tissue, is typical (12). The observation of Totani hypoxia (22). HGF/SF is a mesenchymal-derived protein et al. supports the idea that FGF-2 may play an important with strong influence on adjacent epithelial tissue (10). The role during the formation of EACC. goal of this study was to elucidate the converse, viz., the Incubating the EACC culture with different levels of effect of HGF/SF on FSF-2. Our hypothesis was that HGF/SF (20-80 ng/ml) showed different responses of FGF-2 HGF/SF influences the concentration of FGF-2 in the release. The difference between the FGF-2 levels in cultured EACC without mesenchymal cells. untreated EACC culture (36.52 ng/ml) and incubation with Our results presented a strong decrease of FGF-2 levels 20 ng/ml HGF/SF (46.36 ng/ml) was 16%. A significant in untreated EACC culture (from 42.24 ng/ml after 16 hours difference was found at 40 ng/ml (23%, p<0.05), which to 14.83 ng/ml after 72 hours). Obviously, there was no decreased when incubated with 60 ng/ml HGF/SF (17%). stimulus for maintaining the levels of FGF-2. However, we This observation shows that the incubating levels of HGF/SF recently showed that levels of another angiogenic factor, are important for the outcome of FGF-2 concentration in VEGF, significantly increased with time in untreated EACC the EACC culture. Between 20 and 80 ng/ml, 40 ng/ml seems culture (19). to be the most effective concentration for increasing FGF-2 HGF/SF and FGF-2 seem to induce tissue proliferation (p<0.05). Gille et al. reported that HGF/SF induced VEGF angiogenesis by different pathways. Wajih et al. showed that secretion within 30 minutes and, thus, they concluded that angiostatin significantly inhibited proliferation of human up-regulation seemed to be independent of de novo protein umbilical vein endothelial cells (HUVECs) induced by HGF synthesis. They also suggested that HGF/SF may directly (23). In contrast, angiostatin did not inhibit VEGF - or increase VEGF levels independently of new protein FGF-2 - induced signalling events on HUVEC proliferation. synthesis (27). We assume that the initial increase of FGF-2 However, Lai et al. showed that clonal lines of HSulf-1- can be compared to the observation of Gille et al. We also expressing 012SCC attenuated the activation of assume that higher levels of HGF/SF might inhibit, by a ERK/-activated protein kinase (MAPK) signalling different signalling pathway, the release of FGF-2. Wang et mediated by FGF-2 and HGF, respectively (24). The results al. showed that, at high levels of HGF, cell apoptosis was of Wajih and Lai suggest that FGF-2 and HGF/SF mediate promoted (28). HGF was reported to releases apoptotic through common and different signalling pathways. factors allowing apoptosis to occur (29). Further studies are In our cholesteatoma culture model, there was an necessary to reveal the linkage between FGF-2 and HGF- increase of FGF-2 levels after 24-hour incubation with induced apoptosis. This might explain our results that HGF/SF. However, there was a decrease after 48 hours with measured a decrease of FGF-2 levels at higher HFG/SF subsequent stagnation of FGF-2 levels. In the control, there concentrations.

602 Naim et al: HGF/SF and FGF-2

In summary, the increased expression of FGF-2 in the 15 Anthony PF and Anthony WP: Surgical treatment of external presence of HGF/SF in EACC tissue specimens and their auditory canal cholesteatoma. Laryngoscope 92(1): 70-75, 1982. co-expression in EACC tissue suggest a possible role in 16 Naim R, Riedel F, Gotte K, Bran G, Sadick H, Gossler U and Hormann K: Co-expression of different angiogenic factors in establishing EACC. Abrogation of FGF-2 and HGF/SF external auditory canal cholesteatoma. Acta Otolaryngol 124(5): signalling pathways is a possible avenue for preventing the 563-568, 2004. growth of EACC. 17 Sudhoff H, Dazert S, Gonzales AM, Borkowski G, Park SY, Baird A, Hildmann H and Ryan AF: Angiogenesis and Acknowledgements angiogenic growth factors in middle ear cholesteatoma. Am J Otol 21(6): 793-798, 2000. We would like to thank Petra Prohaska and Mrs. Herbst for their 18 Adamczyk M, Sudhoff H and Jahnke K: Immunohistochemical excellent technical assistance. This work was in part supported by investigations on external auditory canal cholesteatomas. Otol the research Fund of the Faculty of Medicine Mannheim, Neurotol 24(5): 705-708, 2003. University of Heidelberg, Germany. 19 Naim R, Sadick H, Bayerl C, Riedel F, Schäfer C and Hormann K: Hepatocyte growth factor/scatter factor (HGF/SF) and References VEGF in human external auditory canal cholesteatoma (EACC) cell culture. Int J Mol Med, 2004 (in press). 20 Zhang YW, Su Y, Volpert OV and Vande Woude GF: 1 Rosen EM and Goldberg ID: Regulation of angiogenesis by Hepatocyte growth factor/scatter factor mediates angiogenesis scatter factor. EXS 79193-208, 1997. through positive VEGF and negative 2 Tarnawski A: Molecular mechanisms of ulcer healing. Drug regulation. Proc Natl Acad Sci USA 100(22): 12718-12723, News Perspect 13(3): 158-168, 2000. 2003. 3 Folkman J and Shing Y: Control of angiogenesis by and 21 Bortz J, Lienert GA and Boehnke K: Verteilungsfreie other sulfated polysaccharides. Adv Exp Med Biol 313: 355-364, Methoden in der Biostatistik. Berlin 2000: Springer-Verlag. 1992. 22 Onimaru M, Yonemitsu Y, Tanii M, Nakagawa K, Masaki I, 4 Risau W: Mechanisms of angiogenesis. Nature 386(6626): 671- Okano S, Ishibashi H, Shirasuna K, Hasegawa M and Sueishi 674, 1997. K: -2 transfer can stimulate 5 Bikfalvi A, Han ZC and Fuhrmann G: Interaction of fibroblast hepatocyte growth factor expression irrespective of hypoxia- growth factor (FGF) with megakaryocytopoiesis and mediated downregulation in ischemic limbs. Circ Res 91(10): demonstration of FGF receptor expression in megakaryocytes 923-930, 2002. and megakaryocytic-like cells. Blood 80(8): 1905-1913, 1992. 23 Wajih N and Sane DC: Angiostatin selectively inhibits signaling 6 Vasir B, Reitz, P, Xu, G, Sharma, A, Bonner-Weir S and Weir by hepatocyte growth factor in endothelial and smooth muscle GC: Effects of diabetes and hypoxia on gene markers of cells. Blood 101(5): 1857-1863, 2003. angiogenesis (HGF, cMET, uPA and uPAR, TGF-alpha, TGF- 24 Leung DW, Cachianes G, Kuang WJ, Goeddel DV and Ferrara beta, bFGF and Vimentin) in cultured and transplanted rat N: Vascular endothelial growth factor is a secreted angiogenic islets. Diabetologia 43(6): 763-77, 2000. mitogen. Science 246(4935): 1306-1309, 1989. 7 Leung KH, Pippalla, V, Kreutter A and Chandler M: 25 McNeil PL, Muthukrishnan L, Warder E and D'Amore PA: Functional effects of FGF-13 on human fibroblasts, dermal Growth factors are released by mechanically wounded microvascular endothelial cells, and aortic smooth muscle cells. endothelial cells. J Cell Biol 109(2): 811-822, 1989. Biochem Biophys Res Commun 250(1): 137-142, 1998. 26 Totani L, Piccoli A, Pellegrini G, Di Santo A and Lorenzet R: 8 Giacobini P, Giampietro C, Fioretto M, Maggi R, Cariboni A, Polymorphonuclear leukocytes enhance release of growth Perroteau I and Fasolo A: Hepatocyte growth factor/scatter factors by cultured endothelial cells. Arterioscler Thromb 14(1): factor facilitates migration of GN-11 immortalized LHRH 125-132, 1994. . Endocrinology 143(9): 3306-3315, 2002. 27 Gille J, Khalik M, Konig V and Kaufmann R: Hepatocyte 9 Stoker M, Gherardi E, Perryman M and Gray J: Scatter factor growth factor/scatter factor (HGF/SF) induces vascular is a fibroblast-derived modulator of epithelial cell mobility. permeability factor (VPF/VEGF) expression by cultured Nature 327(6119): 239-242, 1987. keratinocytes. J Invest Dermatol 111(6): 1160-5, 1998. 10 Nakamura T, Nawa K, Ichihara A, Kaise N and Nishino T: 28 Wang X, DeFrances MC, Dai Y, Pediaditakis P, Johnson C, Purification and subunit structure of hepatocyte growth factor Bell A, Michalopoulos GK and Zarnegar R: A mechanism of from rat . FEBS Lett 224(2): 311-316, 1987. cell survival: sequestration of Fas by the HGF receptor Met. 11 Bottaro DP, Rubin JS, Faletto DL, Chan AM, Kmiecik TE, Mol Cell 9(2): 411-421, 2002. Vande Woude GF and Aaronson SA: Identification of the 29 Naim R, Shen T, Riedel F, Bran G, Sadick H and Hormann K: hepatocyte as the c-met proto-oncogene Regulation of apoptosis in external auditory canal product. Science 251(4995): 802-804, 1991. cholesteatoma by hepatocyte growth factor/scatter factor. ORL, 12 Naim R and Linthicum FH Jr: External auditory canal 2004 (in press). cholesteatoma. Otol Neurotol 25(3): 412-413, 2004. 13 Holt JJ: Ear canal cholesteatoma. Laryngoscope 102(6): 608- 613, 1992. 14 Naim R, Shen, T, Linthicum Jr F and Hormann K: Pathology of the external auditory chanal cholesteatoma (EACC). Received September 21, 2004 Laryngoscope (in press). Accepted December 9, 2004

603