stellate cells in vitro

Regulatory Peptides 90 (2000) 47–52 www.elsevier.com / locate / regpep

TGFb1 autocrine growth control in isolated pancreatic fibroblastoid cells / stellate cells in vitro a ¨ Marie-Luise Kruse a , *, Philipp B. Hildebrand a , Christian Timke a , Ulrich R. Folsch , Wolfgang E. Schmidt b a

Laboratory for Molecular Gastroenterology and Hepatology, 1 st Department of Medicine CAU Kiel, University of Kiel, Schittenhelmstr. 12, D-24105 Kiel, Germany b St. Josef-Hospital, University of Bochum, Gudrunstr.56, D-44791 Bochum, Germany Received 8 December 1999; received in revised form 16 February 2000; accepted 18 February 2000

Abstract TGFb1 is a multifunctional factor, controlling cellular growth and extracellular matrix production. Deletion of the TGFb1 gene in mice results in multiple inflammatory reactions. Targeted overexpression of TGFb1 in pancreatic islet cells leads to fibrosis of the exocrine pancreas in transgenic mice. In pancreatic fibrosis interstitial fibroblasts are primary candidates for production and deposition of extracellular matrix. Still, little is known about regulation of these cells during development of pancreatic disease. We established primary cell lines of pancreatic fibroblastoid / stellate cells (PFC) from rat pancreas. Investigation of rPFCs in vitro shows TGFb1 expression by RT-PCR analysis. Mature TGFb1 was detected in culture supernatants by immunoassay. Rat PFCs in culture possess both receptors TGFb receptor type I, and type II, necessary for TGFb1 signal transduction. Inhibition of TGFb1 activity by means of neutralizing antibodies interferes with an autocrine loop and results in a 2-fold stimulation of cell growth. So far, pancreatic fibroblastoid / stellate cells in vitro were known as a target of TGFb1 action, but not as a source of TGFb1. Our data indicate TGFb1 activity in rat pancreas extends beyond regulation of matrix production, but appears to be important in growth control of pancreatic fibroblastoid cells.  2000 Elsevier Science B.V. All rights reserved. Keywords: Pancreas; Stellate cells; TGFb; Autocrine function; Growth control

1. Introduction Pancreatic fibrosis is seen in chronic pancreatitis and pancreatic cancer. Fibrosis is characterized by replacement of functional tissue by connective tissue consisting mostly of densely packed extracellular matrix (ECM). Thus

Abbreviations: TGFb, transforming growth factor-b; rPFC, rat pancreatic fibroblastoid cells; RT-PCR, reverse trancsciption-polymerase chain reaction; SDS–PAGE, sodium dodecylsulfate–polyacrylamide gel electrophoresis *Corresponding author. E-mail address: [email protected] (M.-L. Kruse).

fibrosis appears to result from dysregulation of cells producing extracellular matrix. One of the factors assumed responsible for regulation of fibroblast cell function and matrix deposition is transforming growth factor-b (TGFb) [1–4]. In pancreas TGFb has been functionally investigated in experimental pancreatitis in rodents [5–7]. Investigation of TGFb expression in human pancreatitis was carried out mostly by immunohistochemistry [8–10]. Generally, expression of TGFb1 in pancreatic tissue is predominantly described in ductal and acinar cells. These findings implicate paracrine mechanisms for regulation of cells responsible for extracellular matrix (ECM) production, with parenchymal cells controlling fibroblasts. We have established long-term culture of pancreatic fibroblastoid / stellate cells from rat pancreas. Calling these

0167-0115 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0167-0115( 00 )00104-X

48

M.-L. Kruse et al. / Regulatory Peptides 90 (2000) 47 – 52

cells pancreatic fibroblastoid cells / stellate cells (PFC) pays respect to isolation and long-term culture [11]. While other investigators have analysed the effect of exogenously added TGFb on pancreatic stellate cells regarding production of extracellular matrix [12–16], we were mainly interested in the question of whether these cells themselves express TGFb. We describe here the production of biologically active TGFb1 by fibroblastoid / stellate cells from rat pancreas in vitro. Furthermore we provide evidence for an autocrine loop for TGFb1 activity derived from rPFCs. Inhibition of autocrine TGFb1 activity by neutralising antibodies results in growth stimulation in vitro, depending on cell density.

2. Materials and methods All chemicals were of highest analytical purity purchased from Sigma (Deisenhofen, Germany), Biomol (Hamburg, Germany) or Merck (Darmstadt, Germany). All tissue culture reagents were from Life Technologies (Karlsruhe, Germany), unless otherwise stated. Secondary antibodies were fluorescein (FITC)-coupled donkey antigoat antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA). Alkaline phosphatase-coupled anti-rabbit secondary antibodies were from New England Biolabs (Beverly, MA). Primary antibodies used were rabbit antiTGFb1, rabbit anti-TGFb receptor I, rabbit anti TGFb receptor II, and goat-anti TGFb receptor III from Santa Cruz Biotechnologies (Santa Cruz, CA).

2.1. Cell culture Isolation of rat pancreatic fibroblastoid cells was carried out by explantation of pancreatic tissue into culture dishes as described [11], a detailed description of the procedure and the characterisation of isolated cells and established cell lines will be given elsewhere (manuscript in preparation). For rat pancreatic fibroblastoid cells, culture dishes were coated with rat tail collagen (prepared by acid extraction of rat tail tendons [17]). Cells were routinely cultured in DMEM (Dulbecco’s modified Eagle’s medium) supplemented with 5% FCS (fetal calf serum; Biochrom, Berlin, Germany), 10% heat inactivated horse serum, 10 mM Hepes, and antibiotics (10 mg / ml streptomycin, 10 U / ml penicillin, 50 mg / ml gentamicin). Cells were incubated at 378C in a humid atmosphere under 5% CO 2 . For experimental procedures, cells were incubated in serum free medium (OptiMEM).

2.2. Indirect immunofluorescence staining Cells were plated onto glass coverslips in 12-well plates (Costar, Acton, MA) coated with extracellular matrix. Samples were fixed in ice cold methanol at 2 208C. For indirect immunofluorescence staining, fixed cells were

treated with Tris-buffered saline (20 mM Tris–HCl, 150 mM sodium chloride, pH 7.6) containing 0.2% glycine (TBS-glycine). Incubation with primary antibodies were carried out overnight at 48C in a moist chamber. Secondary antibodies were incubated at 378C for 60 min. Samples were mounted with slow fade mounting medium (Molecular Probes, Eugene, OR) and analysed using a Zeiss Universal microscope equipped for fluorescence microscopy.

2.3. RNA isolation and RT-PCR analysis RNA isolation from passaged cells was carried out using chaotropic homogenization buffers and silica matrix (RNeasy, Qiagen, Hilden, Germany), according to the manufacturer’s instructions. For reverse transcription, 2–3 mg of total cellular RNA were reverse transcribed using oligo-dT primers and pre-formulated reverse transcription mixture containing all necessary components for 1 h at 378C (Amersham-Pharmacia Biotech, Freiburg, Germany). For PCR amplification, 2 ml of reverse-transcribed DNA were combined with 2 nmol oligonucleotide mixture, and amplified using a pre-formulated TAQ polymerase mixture containing all necessary components (Life Technologies). Cycling conditions were adapted for different specific oligonucleotides. PCR-products were analysed on 10% polyacrylamide gels with Tris–borate–EDTA buffer. For TGFb1 amplification commercially available oligonucleotides were used (Clontech, Heidelberg, Germany). TGFb receptor type I oligonucleotides were designed from GenBank sequence L26110 at positions 69–88, 305–324 and 657–676. TGFb receptor type II primers were chosen from GenBank sequence L09653, positions 92–111, 453– 472, 734–753, oligonucleotides for betaglycan were from GenBank sequence M80784, positions 325–346, 856–879, ¨ 941–962 (Biometra,Gottingen, Germany). For amplification annealing temperature was set to 608C and 30 cycles of 45 s were carried out.

2.4. Western blot analysis For Western blot analysis, cells were scraped into 1 ml of 10 mM Tris–HCl, pH 7.6, containing proteinase inhibitors and sonicated. 80–100 mg of protein were analysed by 12.5% preparative reduced SDS–PAGE and transferred onto ImmobilonE membrane (Millipore, Eschwege, Germany) by wet blot technique. Membranes were cut into strips and blocked in 5% dry milk in TBS. Antibodies were incubated overnight and immune complexes were detected using alkaline phosphatase coupled secondary antibodies and chemiluminescence detection (New England Biolabs).

2.5. Detection of mature TGFb in culture supernatants For detection and quantitation of mature TGFb1 in

M.-L. Kruse et al. / Regulatory Peptides 90 (2000) 47 – 52

culture supernatants, confluent layers of cells were washed twice with PBS and incubated for 24 h in serum free medium. Media were harvested and centrifuged to collect cellular debris. The resulting supernatants were analysed using a sandwich immunoassay detecting mature TGFb1 according to the manufacturer’s protocol (Promega, Madison, WI). Optical density (OD) was determined in triplicates for a range of dilutions from 1:40 to 1:1600. Experiments were carried out with different rat pancreatic fibroblastoid cell lines for confirmation of production of mature TGFb1.

2.6. Inhibition of TGFb activity For inhibition of TGFb activity, cells were plated onto 96-well plates at different densities. For low density, 250 cells per well were plated, for high density, 1000 cells per well were plated. After 24 h of culture, cells were washed with PBS and switched to serum-free medium for another 24 h. For inhibition of TGFb activity, cells were then incubated in serum-free medium containing 1 mg / ml antipan TGFb antibody (Sigma). Controls were carried out by either omitting the antibody, and as a positive control 10% FCS was used as a growth stimulus. After 24 h of incubation in the presence of neutralising antibody, MTS reagent (Promega, Madison, WI) was added and colour development was monitored every 30 min for 4 h using a Dynatech MR 5000 plate reader. Optical density (OD) was determined in triplicates and three different experiments were carried out on subsequent passages of cells. Changes above 20% were considered significant.

3. Results We have investigated TGFb production and activity in rat pancreatic fibroblastoid / stellate cells (rPFCs) grown on

49

collagen matrix. The cells were positive for vimentin, a marker for mesenchymal origin. Furthermore, these cells express a-smooth muscle actin (aSMA) and desmin as described before [11].

3.1. Investigation of TGFb1 expression TGFb1 expression was first analysed by RT-PCR. Fig. 1a shows that TGFb1 mRNA is expressed in rat PFCs in vitro. Presence of messenger RNA for TGFb receptors type I and II and III was also shown by RT-PCR as seen in Fig. 1b. Expression at mRNA level might not correlate with protein expression, so we further tested for protein expression by Western blot analysis on cell lysates of rPFCs cultured on collagen. As seen in Fig. 2a, protein expression for TGFb receptor type I and II is detected at their appropriate molecular weights of about 55 and 70 kDa, respectively, using commercially available antibodies against the precursor forms. TGFb receptor type III or betaglycan was found all over the cell surface by indirect immunofluorescence staining (Fig. 2b). To test for the mature form of TGFb1, thus activity of TGFb1, produced by pancreatic fibroblastoid / stellate cells, we performed a sandwich immunoassay detecting mature TGFb1 in culture supernatants of rPFCs. Fig. 3 shows the amount of mature TGFb1 per ml conditioned medium, revealing a concentration 1.8 ng / ml TGFb1 produced in 24 h by rat pancreatic fibroblastoid cells at confluency.

3.2. Functional investigation of TGFb activity To test for biological activity of TGFb1 secreted by rPFCs, cells were incubated with neutralising anti-pan TGFb antibodies. Fig. 4 shows that treatment of rat PFCs with TGFb-neutralising antibody results in about 100% stimulation of growth at low cell density. At higher cell density, near confluency, treatment with neutralising anti-

Fig. 1. (A) TGFb1 expression in rat pancreatic fibroblastoid cells as detected by RT-PCR (lane 1). For controls, nested PCR (lane 2) of the primary product using an internal primer was carried out, as well as amplification of RNA without reverse transcription (lane 3), to rule out genomic contamination. (B) Expression of TGFb-receptors type I, II, and III at mRNA level. All receptors are present in similar amounts at mRNA level. Lane 1, primary PCR amplification product; lane 2, nested PCR product; lane 3, RNA amplification control.

50

M.-L. Kruse et al. / Regulatory Peptides 90 (2000) 47 – 52

Fig. 2. (A) Expression of TGFb-receptors at protein level. (A) TGFb receptor types I and II are shown by Western blot analysis at their appropriate molecular weights of about 55 and 75 kDa, respectively. (B) TGFb receptor type III or betaglycan expression detected by indirect immunofluorescence staining. Betaglycan is found at the plasma membrane.

growth control by cell–cell interaction or contact inhibition.

4. Discussion

Fig. 3. TGFb1 production by pancreatic fibroblastoid cells: for comparison, the TGFb1 content of standard growth medium is shown at a concentration of about 1 ng / ml. Production of TGFb1 was determined using serum-free medium, which does not contain TGFb1. During 24 h of incubation, rPFCs at confluency produce about 1.8 ng / ml of mature TGFb1.

bodies had no significant effect on cellular growth above control. To exclude the possibility of exhaustion of antibody, a higher dose of 5 mg / ml anti-TGFb antibody was used, but did not show any effect (data not shown). Thus, growth of rat PFCs is tightly controlled in vitro. This tight control of cellular growth was also detected during routine culture of cells. Confluent cell layers were kept in culture over several days without destruction of the monolayer by focal growth or enhanced apoptosis. Stimulation of starved cells at high density with 10% FCS did not yield significant growth responses (data not shown), arguing for

TGFb is a multifunctional growth / differentiation factor which has been investigated extensively in pancreas, and was mainly found in parenchymal cells by in situ hybridisation and immunohistochemistry [5,8–10]. A major target of TGFb1 activity are cells producing extracellular matrix [3,4]. Detection of immunoreactivity for TGFb1 or its mRNA suggests parenchymal expression and paracrine activity. We were most interested, whether isolated rat pancreatic fibroblastoid cells / stellate cells (rPFC) were able to produce TGFb1 in vitro. We detected TGFb1 mRNA expression by RT-PCR analysis. As TGFb1 is synthesized and secreted as an inactive precursor, the question arose whether rPFC were capable of activating the precursor form to yield the biologically active form. We thus used a sandwich immunoassay that detects TGFb1 only in its mature, biologically active conformation. Production of mature TGFb1 by rPFCs was detected at about 1.8 ng / ml in 24 h. This finding demonstrates that rPFCs not only synthesize and secrete TGFb1, but are also capable of generating the biologically active form of TGFb1. TGFb receptor expression was shown at mRNA and protein level. Co-expression of TGFb receptor type I and type II is essential for TGFb activity [18–20], while betaglycan (TGFb type III receptor) is believed to represent a pericellular reservoir of TGFb1, due to binding of the molecule [21], thus all necessary elements for autocrine activity were found. TGFb regulates production of ex-

M.-L. Kruse et al. / Regulatory Peptides 90 (2000) 47 – 52

51

Fig. 4. Inhibition of TGFb activity by neutralising antibodies. Cells were plated at different density (250 and 1000 cells per well, respectively) and their response to TGFb inhibition recorded by conversion of chromogenic substrate (left, means6S.D., n 5 3). Expression of measurements as percent of control (right) reveals a 2-fold stimulation of growth at low cell density.

tracellular matrix and metalloproteinases, controlling connective tissue deposition [2–4]. In experimental pancreatitis, it has been shown that inhibition of TGFb activity by systemically applied anti-TGFb antibodies resulted in reduced matrix production in cerulein pancreatitis in rats [5]. Systemic administration of TGFb1 in mice after repeated courses of pancreatitis led to enhanced matrix production [6]. These data strongly argue for TGFb1 as a major factor in control of extracellular matrix production. Others have shown that treatment of isolated pancreatic fibroblast-like cells or pancreatic stellate cells in vitro with exogenously added TGFb1 positively regulates collagen and fibronectin expression [12,14,15]. However, these studies did not address the question of endogenous TGFb1 production and possible autocrine functions. We were most interested as to whether rPFCs produce TGFb1 and its involvement in growth control. To test for autocrine activity, that is, a function for TGFb1 in the cell supernatant upon rPFCs themselves, inhibition experiments were carried out. Addition of neutralising antibodies against TGFb at 100-fold molar excess, captured mature TGFb isoforms, which resulted in stimulation of cellular growth of rat PFCs. So far, TGFb activity in pancreas has been viewed mainly as a paracrine system, due to immunological detection of the factor in parenchymal cells. We have presented evidence that pancreatic fibroblastoid cells / stel-

late cells produce TGFb1 in its mature form and respond in an autocrine fashion in vitro. The question arises, how these in vitro data transfer into the in vivo situation. In normal rat pancreas, under normal physiological conditions, masses of acinar cells are accompanied by few fibroblasts, embracing whole acini with long cellular extensions. This compares to a low density growth situation. Furthermore, in normal pancreas, growth of fibroblasts, as well as matrix production, needs to be tightly controlled. Uncontrolled proliferation and / or matrix production would result in organ fibrosis. Our findings suggest that in normal pancreas TGFb1, might inhibit growth of fibroblasts when present in small numbers (read: low density), thus stabilising steady-state conditions. Furthermore, capability of rPFCs to activate the latent form was demonstrated, thus activation of TGFb1 delivered in a paracrine fashion would be ensured, as well as autocrine control. Transgenic mice, overexpressing TGFb1 in islet cells [22] develop fibrosis of the exocrine pancreas. Mice with TGFb1 null mutations clearly underline the role of TGFb1 in maintenance of steady-state conditions, as pancreatic inflammation was found as a result of TGFb1 depletion [23,24]. Regulation of expression / activation levels of TGFb1 appears to be the major switch between beneficial and pathological activity of this growth / differentiation factor. One of the questions arising from genetically engineered animals, as well as several other

52

M.-L. Kruse et al. / Regulatory Peptides 90 (2000) 47 – 52

models, and human pancreatic disease, is whether proliferation of fibroblastoid cells and subsequent enhanced matrix production is the primary event, or vice versa. An answer to this question might be of therapeutic importance, and in vitro investigation of pancreatic fibroblastoid cells / stellate cells could provide a clue to the riddle.

Acknowledgements Parts of the work described here are results from the doctoral thesis of PBH and CT.

References [1] Roberts A, Sporn MB, Assoian RK, Smith JS, Roche NS, Wakefield LM, Heine UI, Liotta LA, Falanga V, Kehr JH, Fauci AS. Transforming growth factor type b: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc Natl Acad Sci USA 1986;83:4167–71. [2] Braun L, Mead JE, Panzica M, Mikumo R, Bell GI, Fausto N. Transforming growth factor b mRNA increased during liver regeneration: a possible paracrine mechanism of growth regulation. Proc Natl Acad Sci USA 1988;85:1539–43. [3] Ignotz RA, Endo T, Massague J. Regulation of fibronectin and type I collagen mRNA levels by transforming growth factor-b. J Biol Chem 1987;14:6443–6. [4] Penttinen RP, Sentaro K Bornstein P. Transforming growth factor b increases mRNA for matrix proteins both in the presence and absence of changes in mRNA stability. Proc Natl Acad Sci USA 1988;85:1105–8. ¨ ¨ [5] Gress TM, Muller-Pillasch F, Elsasser H-P, Bachem MG, Ferrara C, Weidenbach H, Lerch M, Adler G. Enhancement of transforming growth factor b1 expression in the rat pancreas during regeneration from caerulein-induced pancreatitis. Eur J Clin Invest 1994;24:679– 85. [6] Menke A, Yamaguchi H, Gress TM, Adler G. Extracellular matrix is reduced by inhibition of transforming growth factor b1 in pancreatitis in the rat. Gastroenterology 1997;113:295–303. [7] Van Laethem J-L, Robberecht P, Resibois A, Deviere J. Transforming growth factor b promotes development of fibrosis after repeated courses of acute pancreatitis in mice. Gastroenterology 1996;110:576–82. [8] Slater SD, Williamson RCN, Foster CS. Expression of transforming growth factor b1 in chronic pancreatitis. Digestion 1995;56:237–41. [9] Van Laethem J-L, Deviere J, Resibois A, Rickaert F, Vertongen P, Ohtani H, Cremer M, Miyazono K, Robberecht P. Localization of transforming growth factor b1 and its latent binding protein in human chronic pancreatitis. Gastroenterology 1995;108:1873–81. ¨ [10] Friess H, Zhao L, Riesle E, Uhl W, Brundler A-M, Horvath L, Gold ¨ LI, Korc M, Buchler MW. Enhanced expression of TGF-bs and their receptors in human acute pancreatitis. Ann Surg 1998;227(1):95– 914.

[11] Kruse M-L, Hildebrand PB, Timke C, Foelsch UR, Schmidt WE. TGFb autocrine growth regulation in rat pancreatic fibroblastoid cells (rPFC) (Abstract). Gastroenterology 1999;116(4):A1141. [12] Kato Y, Inoue H, Fujiyama Y, Bamba T. Morphological identification of and collagen synthesis by periacinar fibroblastoid cells cultured from isolated rat pancreatic acini. J Gastroenterol 1996;31:565–71. [13] Saotome T, Inoue H, Fujimiya M, Fujiyama Y, Bamba T. Morphological and immunocytochemical identification of periacinar fibroblast-like cells derived from human pancreatic acini. Pancreas 1997;14(1):373–82. [14] Apte MV, Haber PS, Applegate TL, Norton ID, McCaughan GW, Korsten M, Pirola RC, Wilson JS. Periacinar stellate shaped cells in rat pancreas: identification, isolation, and culture. Gut 1998;43:128– 33. [15] Bachem MG, Schneider E, Groß H, Weidenbach H, Schmid RM, ¨ Menke A, Siech M, Beger H, Grunert A, Adler G. Identification, culture, and characterization of pancreatic stellate cells in rats and humans. Gastroenterology 1998;115:421–32. [16] Apte MV, Haber PS, Darby SJ, Rodgers SC, McCaughan GW, Korsten MA, Pirola RC, Wilson JS. Pancreatic stellate cells are activated by proinflammatory cytokines: implications for pancreatic fibrogenesis. Gut 1999;44:534–41. [17] Paul J. In: Cell and tissue culture, 5th ed, Edinburgh, London, New York: Churchill Livingstone, 1975. [18] Lin HY, Wang X-F, Ng-Eaton E, Weinberg RA, Lodish HF. Expression cloning of the TGF-b type II receptor. A functional transmembrane serine / threonine kinase. Cell 1992;68:775–85. [19] Wrana JL, Attisano L, Carcamo J, Zentella A, Doody J, Laiho M, Wang, Massague J. TGFb signals through a heterotrimeric protein kinase receptor complex. Cell 1992;71:1003–14. [20] Ebner R, Chen R-H, Shum L, Lawler S, Zioncheck TF, Lee A, Lopez AR, Derynck R. Cloning of a type I TGF-b receptor and its effect on TGF-b binding to the type II receptor. Science 1993;260:1344–8. [21] Andres JL, Stanley K, Cheifetz S, Massague J. Membrane-anchored and soluble forms of betaglycan, a polymorphic proteoglycan that binds transforming growth factor-b. J Cell Biol 1989;109(6):3137– 45. [22] Sanvito F, Nichols A, Herrera P-L, Huarte J, Wohlwend A, Vassali J-D, Orci L. TGF-b1 overexpression in murine pancreas-induced chronic pancreatitis and, together with TNF-a, triggers insulindependent diabetes. Biochem Biophys Res Commun 1995;217(3):1279–86. [23] Shull MM, Ormsby I, Kier AB, Pawlowski S, Diebold RJ, Yin M, Allen R, Sidman C, Proetzel G, Clvin D, Annunziata N, Doetschmann T. Targeted disruption of the mouse transforming growth factor-b1 gene results in multifocal inflammatory disease. Nature 1992;359:693–9. [24] Kulkarni AB, Huh C-G, Becker D, Geiser A, Lyght M, Flanders KC, Roberts AB, Sporn MB, Ward JM, Karlsson S. Transforming growth factor b 1 null mutation in mice causes excessive inflammatory response and early death. Proc Natl Acad Sci USA 1993;90:770–4.