Am J Physiol Cell Physiol AJP: Heart and Circulatory Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Cell Physiol 284: C209-C219, 2003. First published September 11, 2002; doi:10.1152/ajpcell.00557.2001
0363-6143/03 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
284/1/C209    most recent
00557.2001v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kunz-Schughart, L. A.
Right arrow Articles by Knuechel, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kunz-Schughart, L. A.
Right arrow Articles by Knuechel, R.
Vol. 284, Issue 1, C209-C219, January 2003

Three-dimensional tissue structure affects sensitivity of fibroblasts to TGF-beta 1

Leoni A. Kunz-Schughart, Sabine Wenninger, Thomas Neumeier, Paula Seidl, and Ruth Knuechel

Institute of Pathology, University of Regensburg, D-93042 Regensburg, Germany


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Transforming growth factor-beta (TGF-beta ) is known to induce alpha -smooth muscle actin (alpha -SMA) in fibroblasts and is supposed to play a role in myofibroblast differentiation and tumor desmoplasia. Our objective was to elucidate the impact of TGF-beta 1 on alpha -SMA expression in fibroblasts in a three-dimensional (3-D) vs. two-dimensional (2-D) environment. In monolayer culture, all fibroblast cultures responded in a similar fashion to TGF-beta 1 with regard to alpha -SMA expression. In fibroblast spheroids, alpha -SMA expression was reduced and induction by TGF-beta 1 was highly variable. This difference correlated with a differential regulation in the TGF-beta receptor (TGFbeta R) expression, in particular with a reduction in TGF-beta RII in part of the fibroblast types. Our data indicate that 1) sensitivity to TGF-beta 1-induced alpha -SMA expression in a 3-D environment is fibroblast-type specific, 2) fibroblast type-independent regulatory mechanisms, such as a general reduction/loss in TGF-beta RIII, contribute to an altered TGFbeta R expression profile in spheroid compared with monolayer culture, and 3) fibroblast type-specific alterations in TGFbeta R types I and II determine the sensitivity to TGF-beta 1-induced alpha -SMA expression in the 3-D setting. We suggest that fibroblasts that can be induced by TGF-beta 1 to produce alpha -SMA in spheroid culture reflect a "premyofibroblastic" phenotype.

normal fibroblasts; tumor-derived fibroblasts; multicellular spheroid; transforming growth factor-beta 1; transforming growth factor-beta receptor; alpha -smooth muscle actin, ED-A fibronectin


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

THE RELATIONSHIP BETWEEN tumor cells and their heterologous peritumoral stroma has been of great interest to pathologists since the introduction of microscopic tissue imaging. However, in spite of the diagnostically relevant phenomenon of tumor-associated desmoplasia characterized by enhanced fibroblast proliferation/accumulation and a modified, collagen-rich extracellular matrix (ECM), fibroblasts were long presumed to be passive structural elements and were mainly investigated as a substrate of tumor cell invasion. Studies over the past 10-15 years have demonstrated that not only inflammatory and endothelial cells but also stromal fibroblasts may critically affect malignant growth and progression (for review, see Refs. 13, 14, and 37).

Recently, we established a spheroid coculture model of diverse breast tumor cell lines and fibroblast types to investigate the multiple regulatory feedback mechanisms between stromal fibroblasts and breast tumor cells in a well-defined three-dimensional (3-D) environment in vitro (12). We examined myofibroblast differentiation and, in particular, alpha -smooth muscle actin (alpha -SMA) expression as a representative potential functional "anomaly" of tumor-associated fibroblasts. Fibroblasts in spheroid culture were in general cell cycle arrested and immunonegative for alpha -SMA independent of their origin and also independent of their alpha -SMA expression profile in monolayer culture. We could show that some noninvasive tumor cell types such as T47D induced alpha -SMA expression in tumor-derived fibroblasts in the mesenchymal-epithelial contact zone. Others, e.g., SK-BR-3, induced alpha -SMA expression in the entire fibroblast population, which may be due to a diffuse infiltration of tumor cells in these cocultures. BT474 cells reflected a third group of tumor cells that was not capable of inducing alpha -SMA expression in any of the fibroblast cocultures investigated.

Interestingly, alpha -SMA expression in stromal fibroblasts was not only dependent on the interacting tumor cell type. Indeed, some fibroblasts types outgrown from breast tumor biopsy specimens showed an alpha -SMA-positive immunohistological staining after tumor cell contact, whereas normal skin fibroblasts were not induced by the same tumor cell types in 3-D culture (12). From these in vitro data, we concluded that some fibroblasts isolated from the reactive environment of breast lesions may exhibit a "premyofibroblastic" differentiation status that is conserved in vitro and accompanied by a higher sensitivity to alpha -SMA-inducing factors.

Transforming growth factor-beta (TGF-beta ) has been described as one of the most potent paracrine inducers of myofibroblast differentiation in vitro and in vivo (5, 27, 33, 36), and not only PDGF but also TGF-beta 1 was identified as playing a role in the establishment of tumor desmoplasia (19, 32). Therefore, the particular aims of the present study were 1) to evaluate whether TGF-beta 1 differentially induces alpha -SMA expression in fibroblasts of different origin according to their behavior in 3-D tumor-fibroblast coculture, 2) to verify whether the 3-D environment affects the alpha -SMA inducibility, and 3) to gain deeper insight into the regulatory mechanism associated with a potentially different behavior of fibroblasts in 2-D vs. 3-D culture. The experimental design involved immunohistochemical and Western blot analyses of three fibroblast types. The effect of TGF-beta 1 on alpha -SMA expression in monolayer and spheroid culture was documented, and the expression pattern of TGF-beta receptor (TGFbeta R) types I, II, and III was examined. In addition, effects of TGF-beta 1 on monolayer cell growth were documented, and the expression of the ED-A fibronectin (ED-A FN) splice variant was evaluated by immunohistochemistry in monolayer and spheroid cultures because ED-A FN was recently shown to be involved in the TGF-beta -induced myofibroblast differentiation process (8, 30, 31, 35).


    MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Fibroblast Types and Routine Cell Culture

PF1, PF28, and PF27 fibroblasts derived from fresh, resected specimens of invasive ductal breast carcinomas as detailed earlier (12). In brief, nontumor tissue was removed after frozen section diagnosis and tumor material was sliced (1-4 mm3) under sterile conditions following extensive washing. Tumor fragments were transferred into culture flasks and covered after attachment with DMEM containing 20% FCS, 25 mM glucose, 1% sodium pyruvate, 1% L-glutamine, 100 IU/ml penicillin, and 100 µg/ml streptomycin (Pan Biotech, Aidenbach, Germany). After sufficient outgrowth, fragments were removed and cells were passaged by using 0.05% trypsin and 0.02% EDTA in PBS (Pan Biotech) and transferred into DMEM with reduced glucose and serum content of 5 mM and 10%, respectively. Morphologically and immunohistochemically characterized stock cultures (12) were frozen with a cumulative population doubling (CPD) <= 25 in liquid N2 using 90% FCS and 10% DMSO and were subsequently recultured for the present study. Normal skin fibroblasts (N1) were provided by the Department of Clinical Chemistry, University of Regensburg, and cultured in complete DMEM containing 10% FCS. Preparation, storage, transfer, and recultivation were performed according to the protocol given above.

All cultures were kept in a humidified atmosphere with 5% CO2 in air at 37°C. Cell counts and cell volumes were routinely recorded with a Casy1 cell analyzer system (Schaerfe, Reutlingen, Germany) for culture quality assessment and to analyze cell growth kinetics as described earlier (16). Experiments were carried out with fibroblasts with a CPD of >30 and <80 to avoid cell senescence phenomena and to guarantee a relatively stable proportion of myofibroblasts in the untreated monolayer culture of about 10% in culture medium containing 10% FCS (12).

Spheroid Culturing

Multicellular spheroids (MCS) were cultured by using the liquid overlay technique (3), agarose-coated 96-well plates (100 µl of 1.5% agarose in serum-free DMEM per well; Sigma-Aldrich), and dissociated subconfluent monolayer fibroblasts. Fibroblast MCS were initiated by inoculating 4 × 103 N1 and 3 × 103 PF1 and PF28 fibroblasts, respectively, per well to reach a defined size of 300-350 µm (MCS volume: ~2 × 107µm3) after 3 days. MCS culturing was performed in supplemented DMEM containing 10% FCS under standard culture conditions. Mean spheroid sizes were routinely recorded by measuring two orthogonal diameters of 12-24 individual MCS quantified in an inverted microscope equipped with a calibrated reticule. Medium was renewed at day 3 and every 48 h thereafter.

Experimental Design and TGF-beta 1 Treatment Modalities of Monolayer Cultures

The effect of TGF-beta 1 on the following parameters was examined in fibroblast monolayer cultures: 1) the proportion of alpha -SMA-positive fibroblasts determined via immunohistochemical staining, 2) the alpha -SMA and TGFbeta R protein expression analyzed by Western blotting, and 3) the growth and cell volume kinetics. For 1 and 3, (1.5-2) × 103 fibroblasts were seeded per cm2 onto 16-well tissue culture glass slides (Nunc/Life Technologies, Karlsruhe, Germany) and into 24-well plates (Greiner Labortechnik, Frickenhausen, Germany), respectively. For 2, (4-5) × 103 cells/cm2 were placed into 100-mm culture dishes to reach confluence at the day of protein isolation. TGF-beta 1 concentrations ranged between 0.01 and 100 ng/ml and were given three times over a period of 4 days beginning 24 h after inoculation (days 1, 3, and 5). The TGF-beta 1 concentration(s) applied in 1 and 3 were defined according to the results obtained in 1. Cells were routinely cultured in supplemented DMEM containing 10% FCS. Each well of a 16-well chamber slide was fed with 100 µl while individual wells of 24-well plates were covered with 300 µl and 100-mm dishes with 7.5 ml of DMEM plus ingredients. For 1 and 2, cells were analyzed at day 6; cell growth was determined throughout confluence with culture medium routinely renewed every 48 h.

Experimental Design and TGF-beta 1 Treatment Modalities of Spheroid Cultures

Three days after initiation, MCS volume was monitored and spheroids of each fibroblast type were divided into three aliquots. From one aliquot, protein was isolated immediately to compare TGFbeta R expression in 2-D and 3-D cultures. The second portion was used as control and was fed with complete DMEM (see above) in parallel to the third aliquot that underwent TGF-beta 1 treatment. In accordance with the monolayer results, a TGF-beta 1 concentration of 10 ng/ml was applied at days 3, 5, and 7, followed by protein isolation for Western blot analysis of alpha -SMA (and TGFbeta R types I, II, and III) at day 8.

Immunohistochemistry

To determine the alpha -SMA-positive cell fraction in TGF-beta 1-treated vs. untreated N1, PF1, and PF28 fibroblast monolayer cultures, 16-well glass slides with cells were fixed in ice-cold acetone (10-20 min), air-dried, and stored at -20°C until use. Monolayer cells grown on conventional sterile glass slides were processed in the same way for immunohistochemical detection of fibronectin, in particular of the ED-A FN variant. MCS were shock frozen in liquid N2 using Jung tissue freezing medium (Leica, Nussloch, Germany) and stored at -80°C, and serial 5-µm frozen sections were prepared for further processing. Monoclonal antibodies against human alpha -SMA (1:50, final concentration: 1 µg/ml; Boehringer Mannheim, Germany), human fibronectin (clone 3E3, 1:20, final concentration: 5 µg/ml; Boehringer Mannheim), and human ED-A FN (clone Ist-9, 1:500, final concentration: 1 µg/ml; Biozol, Munich, Germany) were applied. Frozen tumor material (59 invasive ductal breast carcinomas) was processed similarly. Here, 5-µm parallel sections were routinely stained with monoclonal antibodies AS02 (1:75, final concentration: 2.7 µg/ml; Dianova, Hamburg, Germany), anti-cytokeratin 18 (1:50, final concentration: 0.4 µg/ml; Boehringer Mannheim), and anti-alpha -SMA. In 9/59 cases, normal breast epithelium was analyzed as an internal control. All primary antibodies were of mouse origin (IgG) and detected with the StreptABComplex/HRP Duet mouse/rabbit kit (Dako Diagnostika, Hamburg, Germany). 3,3'-Diaminobenzidine (DAB; Kem-En-Tec, Copenhagen, Denmark) was used for color development. Cell nuclei were counterstained with hematoxilin. Isotype controls revealed antibody binding specificity. The alpha -SMA-positive cell fraction in monolayer cultures (means ± SD) was determined by counting the anti-alpha -SMA-stained cells with a DAB-positive cytoplasm relative to the total number of cell nuclei in 4-8 separate samples for each treatment modality. Two-hundred to five-hundred cells were analyzed per sample by two independent investigators, and those individual data differing by no more than 20% were averaged. In a few cases, 1,000 cells were counted by each investigator to reduce the variance due to a variable cell concentration at different locations of the glass slides. Group differences were evaluated by using a two-tailed t-test for unpaired observations.

Western Blotting

Monolayer cultures and pellets of MCS were washed with PBS and lysed under addition of 24 mM Tris · HCl (pH 7.6), 1 mM EDTA, 1 mM PMSF, 1% DTT, and 1% SDS. Cell lysates were transferred into Eppendorf cups for a 30-min incubation on ice. Protein concentrations were determined via the BCA protein assay reagent kit (Pierce), adjusted by adding appropriate amounts of lysis buffer, subsequently mixed with 5× loading buffer (50 mM Tris · HCl, pH 6.8, 2% SDS, 0.1% bromphenol blue, 10% glycerin, and 5% beta -mercapthoethanol), and stored at -20°C.

Proteins were separated by SDS-PAGE [10% polyacrylamide (PAA):bis-acrylamide (bis-AA) 38:1] with 5 mM Tris, 38.4 mM glycin, and 0.02% SDS as running buffer in a MiniproteanIII-electrophoresis system (Bio-Rad, Munich, Germany) at 120 V for 90 min. A routine semidry blotting technique (transfer buffer: 25 mM Tris, 150 mM glycin, 10% methanol; 2 h, 4-5 mA/cm2) was used to transfer protein to PVDF membrane (Boehringer). Membranes were blocked with 5% milkpowder in AP/T-buffer (0.1 M Tris · HCl, pH 7,4, 0.1 M NaCl, 2.5 mM MgCl2, and 0.05% Tween-20). Proteins were detected by indirect labeling using the monoclonal mouse anti-human alpha -SMA antibody (1:600, final concentration: 0.08 µg/ml; Boehringer), polyclonal rabbit antibodies against TGFbeta R types I and II (clone V-22, 1:50-1:100, final concentration: 2-4 µg/ml, and clone L-21, 1:100-1:200, final concentration: 1-2 µg/ml), or a polyclonal goat anti-TGFbeta R type III (clone C-20, 1:200, final concentration: 1 µg/ml; all from Santa Cruz Biotechnology, Santa Cruz, CA). Incubation was carried out for 1 h at room temperature except for the goat IgG that required incubation overnight at 4°C. After subsequent washing, horseradish peroxidase (HRP)-conjugated secondary anti-mouse, anti-rabbit, and anti-goat IgG (1:500-1:1,000; all from Dako Diagnostika) were applied for 1 h at 22°C. Peroxidase activity was recorded on a Hyperfilm-ECL (Amersham, Buckinghamshire, UK) using the Nowa-Western blotting detection kit (EnerGene, Regensburg, Germany). If possible, membranes were stripped in 0.1% glycin (pH 2.5) and stained for a different antibody. This allowed for a parallel detection of all antigens on one membrane if beta -actin or total actin was not stained as an internal protein control. Membranes were routinely stained with Coomassie blue. All experiments were performed at least three times. In some control experiments, actin (beta -actin and/or total actin-nonsmooth muscular) was determined as a protein control using monoclonal rabbit-anti-human or mouse-anti-human beta -actin antibodies (Sigma-Aldrich, Deisenhofen, Germany). In some cases, relative signal intensities were analyzed by densitometry. Unspecified chemicals and antibodies were obtained from Sigma-Aldrich or Merck (Darmstadt, Germany).


    RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

alpha -SMA Expression in Tumor-Associated Fibroblasts in Situ and in 3-D Coculture

It was described previously that some but not all breast cancer cells are capable of inducing alpha -SMA in fibroblasts in spheroid coculture (12). In addition, the distribution of alpha -SMA expression in fibroblast spheroids was shown to depend on the breast cancer cell line applied for coculturing. To evaluate whether this behavior reflects different breast tumor types or even different areas within one tumor and to validate that our model system reflects the in vivo situation, alpha -SMA distribution was analyzed in more than 50 desmoplastic breast cancer specimens using routine immunohistochemistry. In more than 90% of the cases showing myofibroblast differentiation, alpha -SMA was positive in fibroblasts adjacent to tumor cells and was negative in large tumor cell-free fibroblast cords, which is reflected in cocultures of T47 tumor cells and tumor-derived fibroblasts (Fig. 1). The more diffuse the distribution of tumor cells and fibroblasts, the larger the alpha -SMA-positive fibroblast fraction. This may be reflected in the SK-BR-3/PF1 coculture model (Fig. 1B).


View larger version (88K):
[in this window]
[in a new window]
  		Fig. 1.   Tumor cell induction of alpha -smooth muscle actin (alpha -SMA) in situ and in spheroid coculture. A: immunohistochemical staining of step sections of a scirrhous G3 breast carcinoma specimen using antibody AS02 (Dianova), which stains all fibroblasts in tumor tissues, as well as activated capillary endothelial cells, but not tumor cells; an anti-cytokeratin 18 antibody, which stains tumor but not stromal cells; and an anti-alpha -SMA antibody, which localizes alpha -SMA to the myofibroblasts. The figures labeled 1 and 2 are enlargements of the cancer section, which show 1) alpha -SMA-expressing fibroblasts only in stromal cells immediately adjacent to small tumor cell islets, whereas distant stromal cells are alpha -SMA negative, and 2) tumor cell area where alpha -SMA is expressed throughout the entire fibroblast population. B: cocultures of breast carcinoma cells and fibroblasts recapitulate the behavior noted in A. In the spheroid coculture (at left), tumor cells (T47D) and fibroblasts (PF1) stay separate and alpha -SMA is induced in a defined tumor-associated fibroblast population. Breast tumor cells (at right; SK-BR-3) invade the (PF1) fibroblast cultures, inducing alpha -SMA expression throughout the entire spheroid. Five-µm frozen sections; 3,3'-diaminobenzidine (DAB) for detection; hematoxilin counterstain; magnification, 250×.

alpha -SMA expression in tumor-associated fibroblasts was highly variable not only for different ductal invasive breast carcinomas but also within the individual tumor, reflecting its histomorphological heterogeneity (Fig. 1A). Thus application of different tumor cell lines in the coculture system may, to some extent, reflect the situation within one tumor. However, it remains unsolved why some fibroblast types, e.g., most tumor-derived fibroblast types, seem to be more sensitive to tumor-induced alpha -SMA expression in the coculture model than others, such as normal skin fibroblasts. Therefore, experiments to verify the impact of TGF-beta 1 on alpha -SMA expression were performed with two representative fibroblast types: N1 normal skin fibroblasts that were immunonegative for alpha -SMA in tumor-fibroblast cocultures and breast tumor-derived PF1 fibroblasts that clearly expressed alpha -SMA following contact with tumor cells in vitro (Fig. 1B) (12). PF28 cells were included as a second breast cancer-derived fibroblast type.

TGF-beta 1 Induced alpha -SMA Expression in 2-D Fibroblast Cultures

Immunohistochemical staining showed that all fibroblast types (N1 normal skin- and breast tumor-derived PF1 and PF28 fibroblasts) contained a proportion of 8-10% alpha -SMA-positive cells in the exponentially growing monolayer (Fig. 2, A and C) with a tendency to a higher proportion at confluence (Fig. 2B and data not shown). In exponential and confluent monolayer cultures, TGF-beta 1 induced alpha -SMA expression in all fibroblast types independent of their origin (Figs. 2 and 3). Both, exponentially growing N1 and PF1 fibroblasts, showed a significant increase in the alpha -SMA+ fibroblast fraction with increasing TGF-beta 1 concentrations. At 100 ng/ml TGF-beta 1, 70-80% of the fibroblasts clearly expressed alpha -SMA. PF28 fibroblasts were less sensitive and reached a maximum of about 30% alpha -SMA+ cells at 1 ng/ml TGF-beta 1.


View larger version (77K):
[in this window]
[in a new window]
  		Fig. 2.   Transforming growth factor (TGF)-beta 1 induction of alpha -SMA in 3 different fibroblast types in monolayer culture. A: representative immunohistochemical staining of alpha -SMA in untreated (control) and TGF-beta 1-treated (0.01/1/100 ng/ml) exponentially growing N1 skin fibroblasts using a specific monoclonal antibody, a two-step peroxidase technique, DAB for detection, and hematoxilin counterstain. B: representative immunohistochemical staining of alpha -SMA in control and TGF-beta 1-treated (10 ng/ml) confluent N1 skin fibroblasts. Staining details according to A. C: proportion of alpha -SMA-positive cells in untreated and TGF-beta 1-treated exponentially growing N1 skin and breast tumor-derived PF1 and PF28 fibroblasts in monolayer culture. TGF-beta 1 concentrations ranged from 0.01 to 100 ng/ml. All fibroblast types were inducible. * P < 0.05; ** P < 0.001 for TGF-beta 1-treated vs. control (0) values.



View larger version (42K):
[in this window]
[in a new window]
  		Fig. 3.   Effect of TGF-beta 1 on the expression of alpha -SMA in 3 different fibroblast types in monolayer culture and the respective TGFbeta R types I-III profile. Total nonmuscular or beta -actin signals and Coomassie blue-stained membranes are shown as controls. A: representative Western blot analysis showing alpha -SMA in normal skin- N1 and breast tumor-derived PF1 and PF28 fibroblasts in confluent monolayer cultures with or without TGF-beta 1 (10 ng/ml). All fibroblast types were inducible. B: representative Western blot analyses of TGFbeta R types I-III in normal skin- N1 and breast tumor-derived PF1 and PF28 fibroblasts in exponentially growing monolayer cultures with or without TGF-beta 1 (10 ng/ml). All TGFbeta R types are expressed by fibroblasts. The expression profile in the different fibroblast types did not essentially differ except for TGFbeta R type III which showed highest expression in the tumor-derived fibroblast type PF28.

The observation of an induction of alpha -SMA expression in fibroblast by 10 ng/ml TGF-beta 1 was confirmed for confluent monolayers in three independent Western blot analyses (Fig. 3A). The induction ranged between two- and threefold (n = 3 per fibroblast type) as verified by densitometric analysis but did not reflect the proportion of alpha -SMA-positive fibroblasts documented in Fig. 2. This phenomenon will be discussed below (see DISCUSSION, Technical Considerations).

TGF-beta 1 and Fibroblast Monolayer Growth

To evaluate the potential relationship between TGF-beta 1-induced alpha -SMA expression and alteration in cell growth, cell numbers of nonsenescent N1, PF1, and PF28 fibroblasts in DMEM containing 10% FCS with (1 and 10 ng/ml) and without TGF-beta 1 were recorded as a function of time in culture. The growth of N1 fibroblast cultures was unaffected by TGF-beta 1 with a cell doubling time of 42-44 h. Tumor-derived fibroblasts showed a different behavior. TGF-beta 1 treatment of PF1 fibroblast resulted in a clear growth delay as indicated by a reduced cell doubling time at days 1-7 (Fig. 4; 167 and 153 h, respectively, in the treated samples as opposed to 94 h in the control) and a reduction of the cell density per surface area at confluence. In contrast, in PF28 cultures, a slight decrease in the cell count per centimeters squared accompanying TGF-beta 1 incubation was only seen at days 1-5 and did not account for a reduced doubling time if day 7 was included in the calculation (cell doubling time: 108-113 h). In this fibroblast type, cell number per surface area rather increased after TGF-beta 1 treatment (>day 5) (Fig. 4). Supposedly, TGF-beta 1-induced alterations in cell growth and alpha -SMA expression are rather unrelated phenomena.


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4.   Effect of TGF-beta 1 on cell number per surface area as a function of time of 3 different exponentially growing fibroblast types in monolayer culture. Cell growth of normal skin- (N1, A)- and breast tumor-derived fibroblasts (PF1, B; PF28, C) as a function of time in culture with (1 and 10 ng/ml, respectively) and without TGF-beta 1 treatment at days 1, 3, and 5 (indicated by arrows). Medium was renewed every 48 h. Cell counts of 3 independent samples were determined and averaged for each time point. Exponentially growing, nonsenescent fibroblasts with a cumulative population doubling (CPD) of 30-80 were used. TGF-beta 1 differentially affected the growth of the fibroblast types investigated. Cell doubling times were determined for days 1-7 (indicated by the dashed frame).

In parallel, cell volume was determined throughout growth. No systematic and reproducible alteration in the cell volume of normal skin N1 and tumor-derived PF1 or PF28 fibroblasts following TGF-beta 1 treatment was observed. Average cell volumes ranged between 5,500-7,500 µm3 for N1 and 8,000-12,000 µm3 for PF1 and PF28 fibroblasts, respectively.

alpha -SMA and TGFbeta R Expression in 2-D and 3-D Fibroblast Cultures

Western blot analyses were performed to correlate the sensitivity of fibroblasts in monolayer and spheroid cultures to TGF-beta 1 induction of alpha -SMA with TGFbeta R expression. As indicated in Fig. 3B, untreated tumor-derived and normal skin fibroblasts in monolayer culture did not systematically differ in the distribution of the serine-threonine kinase TGFbeta R types I and II, and the receptor content was not consistently altered by the sequential treatment with 10 ng/ml TGF-beta 1 if three independent experiments and all three fibroblast types were taken into account. TGFbeta RI was always at the detection level in monolayer cultures and quantification was not feasible. TGFbeta RIII was reproducibly highest in PF28 fibroblasts in monolayer culture (see also Fig. 5).


View larger version (62K):
[in this window]
[in a new window]
  		Fig. 5.   Effect of TGF-beta 1 on the expression of alpha -SMA in 3 different fibroblast types in spheroid culture and the respective TGFbeta R types I-III profile in spheroid (S) compared with confluent monolayer (M) cultures. One of the corresponding Coomassie blue-stained membranes is shown. Western blot analyses of alpha -SMA and TGFbeta R types I-III in normal skin- N1 and breast tumor-derived PF1 and PF28 fibroblasts in S with and without TGF-beta 1 (10 ng/ml) as opposed to untreated M controls. Here, 50 µg total protein were applied per slot. A clear reduction in alpha -SMA expression in spheroids compared with monolayer cultures was shown. Induction of alpha -SMA by TGF-beta 1 was reproducibly detected in tumor-derived PF1 fibroblasts but was much lower in the tumor-derived PF28 fibroblast type and poor if not invisible in the N1 normal skin fibroblasts. Reduced expression of TGFbeta R type III was observed in all fibroblast types in S compared with M cultures. Arrows indicate a downregulation of TGFbeta R type II in skin fibroblasts N1 in S compared with M culture. Some but not all batches of antibodies (Santa Cruz Biotechnologies) against TGFbeta R type I showed a second band at ~65 to 75 kDa that was specific because its detection was completely inhibited by addition of TGFbeta R type I blocking peptide (data not shown). According to the manufacturer's information and to parallel staining against TGFbeta R type II, this band did not reflect cross-reaction with TGFbeta R type II.

Comparison of the receptor expression of fibroblasts in monolayer and spheroid cultures, taking into account at least three Western blot analyses for each TGFbeta R type, demonstrated the following.

TGFbeta R type I. The TGFbeta RI level in N1 and PF1 fibroblasts in spheroid cultures is potentially higher than in the corresponding monolayers, which does not correlate with the respective discrepancy in the alpha -SMA inducibility. However, in 3-D culture, TGFbeta RI is highest in PF1 fibroblasts that are characterized by the most prominent alpha -SMA induction by TGF-beta 1; PF28 and N1 fibroblasts did not reproducibly differ in their TGFbeta RI content in spheroids. TGF-beta 1 treatment of fibroblast spheroids is (frequently but not always) accompanied by an increase in the TGFbeta RI level.

TGFbeta R type II. This serine-threonine kinase receptor is reduced in N1 normal skin fibroblasts when grown in spheroid culture. This reduction is moderate but reproducible and in contrast to PF1 and PF28 tumor-derived fibroblasts with constant TGFbeta RII levels. As a result, TGFbeta RII expression in spheroids is lower in N1 than in PF1 or PF28 fibroblasts.

TGFbeta R type III. TGFbeta RIII is downregulated in 3-D cultures of all fibroblast types and is undetectable in the Western blot analyses of spheroid cultures with and without TGF-beta 1 treatment. Representative Western blots demonstrating expression of alpha -SMA and TGF-beta receptors in monolayer vs. TGF-beta 1-treated and untreated spheroid cultures are shown in Fig. 5.

To verify our observations on TGFbeta R types I and II, we have examined an additional breast carcinoma-derived fibroblast type (PF27) that shows reduced but clear expression of alpha -SMA in spheroid culture compared with the respective monolayer but is not induced by exposure to TGF-beta 1 (Fig. 6). These experiments were only performed in duplicate and human recombinant TGF-beta 1 was applied. In both experiments, tumor-derived PF27 fibroblasts behaved analogously to N1 normal skin fibroblasts: loss of sensitivity in 3-D compared with confluent 2-D culture correlated with a reduction in the TGFbeta RII. In parallel, TGFbeta RI was enhanced.


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 6.   Expression of alpha -SMA and TGFbeta R types I and II in a "TGF-beta noninducible" tumor-derived fibroblast type. Western blot analyses of alpha -SMA and TGFbeta R types I and II in tumor-derived PF27 fibroblasts in S with and without TGF-beta 1 (10 ng/ml) as opposed to confluent untreated M controls. Here, 50 µg of total protein were applied per slot. alpha -SMA was not induced in S culture by TGF-beta 1 treatment and TGFbeta R type II was downregulated in S compared with M cultures, in contrast to TGFbeta R type I, which showed an inverse expression profile. This experiment was performed in duplicate; the respective Coomassie blue-stained membrane is shown as control. Parallel staining of total actin indicates that actins (alpha /beta and/or gamma ) are in general suppressed in 3-dimensional (3-D) compared with 2-D culture of PF27 fibroblasts. This seems to be a general phenomenon in fibroblasts, is also indicated in beta -actin staining (data not shown), and will be further evaluated in the near future.

Fibronectin/ED-A Fibronectin Distribution in 2-D and 3-D Fibroblast Cultures

All fibroblasts produce a dense ECM in monolayer and spheroid culture. As documented earlier, fibronectin is one of the major ECM compounds in spheroids of all fibroblast types, including normal skin fibroblasts N1 (12). Immunohistochemical staining for the oncofetal ED-A FN variant showed that monolayer fibroblasts of normal skin and breast tumor origin in general express and secrete ED-A FN. There was no difference between the various types of fibroblasts based on semiquantitative immunohistochemical evaluation (data not shown). However, if fibroblast spheroids highly positive for fibronectin were stained with the ED-A FN-specific antibody, a striking difference between N1 and tumor-derived PF1 and PF28 fibroblasts was observed. N1 spheroids showed only poor staining, whereas spheroids of tumor-derived PF1 and PF28 stained strongly (Fig. 7A). The fibroblast type-specific difference in the ED-A FN distribution/expression was preserved in spheroid cocultures of fibroblasts of different origin and breast tumor cell lines, as documented in Fig. 7B. No induction of ED-A FN in fibroblasts following tumor cell contact was observed.


View larger version (75K):
[in this window]
[in a new window]
  		Fig. 7.   Expression of the oncofetal ED-A fibronectin splice variant in 3 different fibroblast types. Bar = 100 µm. A: representative immunohistochemically stained 5-µm frozen sections of fibroblast S (N1 = normal skin fibroblasts; PF1, PF28 = breast tumor-derived fibroblasts) using a monoclonal antibody against the ED-A fibronectin splice variant, DAB for detection, and hematoxilin counterstain. B: 5-µm frozen sections of tumor cell-fibroblast S cocultures (T47D/N1; T47D/PF1) stained for ED-A fibronectin according to A.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Myofibroblastic Phenotype in Vivo and in Vitro

Myofibroblasts were first described by G. Majno and G. Gabbiani and were mainly investigated in chronic inflammatory diseases and during wound healing (for review, see Refs. 20, 25, 29, and 31). The expression of alpha -SMA has become one of the most reliable markers for myofibroblast differentiation (for review, see Refs. 7, 25, 26, 29, 31) and is a consistent feature in the desmoplastic reaction of tumors, including ductal breast cancers. As described earlier, primary fibroblasts from normal dermal tissue show poor alpha -SMA expression (34), and those isolated from normal breast contain a significantly lower proportion of myofibroblasts than those isolated from malignant tissues (6% vs. 60%) (28). However, with increasing time in culture, the alpha -SMA-positive cell fraction in tumor-derived fibroblasts is considerably reduced, reaching values as low as those determined in normal skin fibroblast cultures (12). This may result from an instability of the myofibroblastic phenotype (23) and/or a slower growth rate of myofibroblasts, as opposed to undifferentiated fibroblasts in vitro (34). Also, it has to be taken into account that fibroblasts in culture are removed from the heterologous tumor micromilieu. Loss of tumor-specific cell-cell, cell-matrix, and paracrine interactions may cause myofibroblast dedifferentiation. This hypothesis is strengthened by our observation that various tumor cells are capable of reinducing alpha -SMA expression in some tumor-derived fibroblast types in 3-D coculture. In contrast, normal skin fibroblasts that had not seen a tumor environment a priori could not be induced in the model system. This indicated a persistent "reactive" environment to be required for tumor-associated myofibroblastic differentiation. In addition, the existence of a premyofibroblastic phenotype with a higher susceptibility to factors inducing alpha -SMA expression was hypothesized.

TGF-beta 1 Induced alpha -SMA Expression and Myofibroblastic Phenotype

TGF-beta , in particular TGF-beta 1, has been described as a potent paracrine inducer of myofibroblast differentiation. An effect of TGF-beta has been documented in the formation of granulation tissue, during wound repair and scar formation (4, 6, 8, 17, 39), further in the fibroblast differentiation process induced as foreign body reaction to biomaterials, e.g., during capsule formation (10), and also in the development of tumor-associated myofibroblasts and desmoplasia (5, 27, 33, 36). As a result, a model for TGF-beta -dependent fibroblast-myofibroblast modulation has been introduced recently. Here, TGF-beta activates resident fibroblasts to synthesize and organize an extracellular matrix (ECM) scaffold containing the fibronectin splice variant ED-A, an effect that was described in normal fibroblasts more than 10 years ago (1) and was shown more recently to be necessary but not sufficient for the TGF-beta -induced myofibroblastic phenotype (8, 30, 31, 35).

Our data show that TGF-beta 1-induced alpha -SMA expression in fibroblasts in vitro is critically affected by the culture conditions and, in particular, by a tissue-like 3-D environment. In monolayer culture, alpha -SMA expression is induced in all fibroblast types independent of their origin. This behavior correlates with a comparable expression pattern of the TGF-beta receptors and the ED-A FN variant in the different fibroblast types. In contrast, cultivation of fibroblast spheroids is accompanied by a considerable reduction in the alpha -SMA expression in all fibroblast types and in a reduced sensitivity to TGF-beta 1, in particular in normal skin N1 fibroblasts but also in one of the two tumor-derived fibroblast types (PF28) studied in detail.

alpha -SMA, TGFbeta R Types I and II, and ED-A FN in Normal Skin Fibroblasts

The considerable reduction in alpha -SMA expression and the almost complete loss of sensitivity to TGF-beta 1-induced alpha -SMA expression in N1 fibroblasts in 3-D compared with 2-D culture is accompanied by 1) a moderate decrease in the TGFbeta RII expression and 2) a low production of the ED-A FN splice variant. The reduced level of TGFbeta RII, a serine-threonine kinase responsible for TGF-beta 1 binding, and respective activation of TGFbeta RI (40) to induce intracellular signaling (9) is likely to affect the sensitivity of these fibroblasts to TGF-beta 1. Induction of alpha -SMA and collagen type I by TGF-beta was shown recently to depend on an ED-A FN-derived permissive outside-in signaling (30, 31). The low level in this oncofetal fibronectin variant reflects the in vivo situation of normal adult tissues (11) and may also be involved in a reduced alpha -SMA inducibility in the 3-D setting. We interpret ED-A FN produced by N1 monolayer cultures (data not shown) as a potential cell culture "artifact." Such features, similarly to alpha -SMA inducibility, may not reflect the in vivo characteristics of these fibroblasts. ED-A FN is frequently found in the desmoplastic reaction in tumors. The retained production of high levels of ED-A FN in fibroblasts in spheroid culture indicates that 3-D fibroblast cultures reflect the in vivo situation more adequately than conventional 2-D systems. This was found similarly for tumor cells with respect to morphological and physiological characteristics as complex cell-to-cell and cell-to-matrix interactions (15, 21, 22).

alpha -SMA, TGFbeta R Types I and II, and ED-A FN in Tumor-Derived Fibroblasts

Tumor-derived PF28 fibroblasts showed a reduced expression and TGF-beta 1 inducibility of alpha -SMA in spheroid culture. The levels of expression of ED-A FN and TGFbeta RII are comparable to those in PF1 fibroblasts that are highly inducible. This finding indicates that alterations in ED-A FN and TGFbeta RII expression may not be the exclusive limiting factor for TGF-beta -induced alpha -SMA expression. Here, a reduced level of TGFbeta RI may account for the poor induction of alpha -SMA by TGF-beta 1. PF1 fibroblasts that also expressed high levels of ED-A FN showed the most efficient induction of alpha -SMA by TGF-beta 1. This induction correlated with a level of TGFbeta RI that was higher than in the respective monolayer culture and an unaltered TGFbeta RII level.

The tumor-derived fibroblast type PF27 behaved analogous to normal skin fibroblasts, with loss of sensitivity correlating with a reduction in TGFbeta RII. These results indicate that regulation of alpha -SMA in normal skin- and breast tumor-derived fibroblasts does not necessarily differ. Also, it is in accordance with the observation that only fibroblast subpopulations, but not all stromal fibroblasts in breast tumors, may show myofibroblast differentiation (Fig. 1) and supports the hypothesis of a premyofibroblastic phenotype.

In a model of myofibroblastic differentiation presented by Serini and Gabbiani (31), TGF-beta is released as a paracrine inducer from platelets and macrophages as a consequence of activation through granulocyte macrophage colony-stimulating factor (38). Our previous observations with spheroid cocultures of tumor cells and fibroblasts showed that tumor-associated induction of alpha -SMA expression in fibroblasts does not require immune cell commitment (12). Data presented further indicate that TGF-beta and/or the activation of the TGF-beta signal transduction pathway via TGFbeta R types I and II play a potential role in this process.

TGF-beta 1-Induced alpha -SMA Expression and the TGFbeta R Type III

The considerable reduction/loss of TGFbeta RIII in fibroblasts in 3-D culture is to be stressed because it shows for the first time that this receptor is not required for tumor-associated and TGF-beta 1-induced alpha -SMA expression in fibroblasts. As a consequence, TGF-beta 2, which can only interact with the type II receptor following binding to and mediation through TGFbeta RIII, is unlikely to be involved in the mechanism of tumor-induced myofibroblast differentiation, although both TGF-beta 2 and TGFbeta RIII are implicated in an endothelium-myofibroblast differentiation process (2, 24). Preliminary experiments performed in medium containing 0.1% serum indicate that the reduction in TGFbeta RIII expression may in general accompany cell growth/cycle arrest. However, this hypothesis requires further examination in particular with regard to the tumor-induced stimulation of fibroblast proliferation supposed to result in tumor desmoplasia.

Technical Considerations and Future Directions

Growth curves and alpha -SMA-positive cell fractions were recorded for exponentially growing fibroblast monolayer cultures. These data imply that the regulation of alpha -SMA expression and cell growth by TGF-beta 1 are unrelated processes in the 2-D environment. Fibroblasts in spheroid culture do not proliferate. Therefore, we verified that alpha -SMA is also induced in confluent monolayer cultures (Fig. 2B) and performed all Western blot analyses with fibroblasts that had reached confluence at the day of protein isolation. alpha -SMA-positive cells in confluent monolayer cultures were not counted because the cytoplasmic immune reactions could not clearly be assigned to the respective cell nuclei. As a consequence, the data shown in Fig. 2C representing the increase of the alpha -SMA+ fractions in exponentially growing fibroblast induced by 10 ng/ml TGF-beta 1 may not correlate with the alpha -SMA induction rate for confluent monolayers shown in the Western blot analyses (Fig. 3A). Differences in the cellular alpha -SMA content per cell and per unit protein, an increase alpha -SMA-positive fraction, and/or a decreased alpha -SMA inducibility in confluent as opposed to exponentially growing fibroblasts may explain this discrepancy.

We have avoided cell senescence, which accompanies long-term cell quiescence in fibroblast monolayer cultures. However, future investigation is needed to show whether long-term cell cycle-arrested fibroblasts in 2-D culture still differ from those in a 3-D environment with regard to TGF-beta 1-induced alpha -SMA expression. To interpret these data, one needs to consider that cell cycle in fibroblasts may be differentially regulated in 2-D and 3-D culture as indicated by LaRue et al. (18), who showed a divergent regulation of diverse cyclin-dependent kinases and their inhibitors in a rat fibroblast model.

No striking effect of TGF-beta 1 exposure on fibroblast spheroid size was observed. However, studies to evaluate proliferative activity and viability of different fibroblasts in spheroid coculture with tumor cells and in spheroid monocultures with TGF-beta 1 are recommended to verify whether the model system not only reflects late stages but also the onset of tumor desmoplasia.

Conclusions

Our data clearly show that the TGF-beta receptor expression differs in 2-D and 3-D fibroblast cultures. In addition to fibroblast type-independent differences, fibroblast type-specific modifications in the receptor expression profile that correlate with a highly variable sensitivity of these fibroblasts to TGF-beta 1-induced alpha -SMA expression in the 3-D environment were recorded. This variability was not present in the respective confluent monolayer cultures, indicating that quiescent 3-D fibroblast cultures better reflect the in vivo behavior of different fibroblast phenotypes than the 2-D culture system. We hypothesize that fibroblasts with a high susceptibility to TGF-beta 1 in 3-D culture represent a premyofibroblastic phenotype.


    ACKNOWLEDGEMENTS

We thank F. Van Rey and M. Hoffmann for excellent technical assistance.


    FOOTNOTES

This work was supported by the Deutsche Forschungsgemeinschaft (Grants Ku 917/2-1 to 2-4) and by the Bayerische Staatsministerium für Wissenschaft, Forschung, und Kunst.

Address for reprint requests and other correspondence: L. A. Kunz-Schughart, Institute of Pathology, Univ. of Regensburg, Franz-Josef-Strauss-Allee 11, D-93053 Regensburg, Germany (E-mail: leoni.kunz-schughart{at}med.uni-regensburg.de).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

First published September 11, 2002;10.1152/ajpcell.00557.2001

Received 5 December 2001; accepted in final form 3 September 2002.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Balza, E, Borsi L, Allemanni G, and Zardi L. Transforming growth factor beta regulates the levels of different fibronectin isoforms in normal human cultured fibroblasts. FEBS Lett 228: 42-44, 1988[ISI][Medline].

2.   Brown, CB, Boyer AS, Runyan RB, and Barnett JV. Requirement of type III TGF-beta receptor for endocardial cell transformation in the heart. Science 283: 2080-2082, 1999[Abstract/Free Full Text].

3.   Carlsson, J, and Yuhas JM. Liquid overlay culture of cellular spheroids. In: Spheroids in Cancer Research, edited by Acker H, Carlsson J, Durand R, and Sutherland RM.. Berlin: Springer-Verlag, 1984, p. 1-23.

4.   Desmouliere, A. Factors influencing myofibroblast differentiation during wound healing and fibrosis. Cell Biol Int 19: 471-476, 1995[ISI][Medline].

5.   Desmouliere, A, Geinoz A, Gabbiani F, and Gabbiani G. Transforming growth factor-beta 1 induces alpha-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol 122: 103-111, 1993[Abstract/Free Full Text].

6.   Diamond, JR, van Goor H, Ding G, and Engelmyer E. Myofibroblasts in experimental hydronephrosis. Am J Pathol 146: 121-129, 1995[Abstract].

7.   Gabbiani, G. Some historical and philosophical reflections on the myofibroblast concept. Curr Top Pathol 93: 1-5, 1999[Medline].

8.   George, J, Wang SS, Sevcsik AM, Sanicola M, Cate RL, Koteliansky VE, and Bissell DM. Transforming growth factor-beta initiates wound repair in rat liver through induction of the EIIIA-fibronectin splice isoform. Am J Pathol 156: 115-124, 2000[Abstract/Free Full Text].

9.   Gold, LI. The role for transforming growth factor-beta (TGF-beta) in human cancer. Crit Rev Oncog 10: 303-360, 1999[ISI][Medline].

10.   Khouw, IM, van Wachem PB, Plantinga JA, Vujaskovic Z, Wissink MJ, de Leij LF, and van Luyn MJ. TGF-beta and bFGF affect the differentiation of proliferating porcine fibroblasts into myofibroblasts in vitro. Biomaterials 20: 1815-1822, 1999[ISI][Medline].

11.   Kornblihtt, AR, Pesce CG, Alonso CR, Cramer P, Srebrow A, Werbajh S, and Muro AF. The fibronectin gene as a model for splicing and transcription studies. FASEB J 10: 248-257, 1996[Abstract].

12.   Kunz-Schughart, LA, Heyder P, Schroeder J, and Knuechel R. A heterologous 3-D coculture model of breast tumor cells and fibroblasts to study tumor-associated fibroblast differentiation. Exp Cell Res 266: 74-86, 2001[ISI][Medline].

13.   Kunz-Schughart, LA, and Knuechel R. Tumor-associated-fibroblasts (part II): functional impact on tumor tissue. Histol Histopathol 17: 623-637, 2002[ISI][Medline].

14.   Kunz-Schughart, LA, and Knuechel R. Tumor-associated fibroblasts (part I): active stromal participants in tumor development and progression? Histol Histopathol 17: 599-621, 2002[ISI][Medline].

15.   Kunz-Schughart, LA, Kreutz M, and Knuechel R. Multicellular spheroids: a three-dimensional in vitro culture system to study tumour biology. Int J Exp Pathol 79: 1-23, 1998[ISI][Medline].

16.   Kunz, LA, and Mueller-Klieser W. Influences of recombinant human tumor necrosis factor-alpha on growth and oxygen consumption of MCF-7 and HT29 cells. Cell Physiol Biochem 1: 214-225, 1991.

17.   Lanning, DA, Diegelmann RF, Yager DR, Wallace ML, Bagwell CE, and Haynes JH. Myofibroblast induction with transforming growth factor-beta1 and beta3 in cutaneous fetal excisional wounds. J Pediatr Surg 35: 183-187, 2000[ISI][Medline].

18.   LaRue, KE, Bradbury EM, and Freyer JP. Differential regulation of cyclin-dependent kinase inhibitors in monolayer and spheroid cultures of tumorigenic and nontumorigenic fibroblasts. Cancer Res 58: 1305-1314, 1998[Abstract/Free Full Text].

19.   Löhr, M, Schmidt C, Ringel J, Kluth M, Müller P, Nizze H, and Jesnowski R. Transforming growth factor-beta1 induces desmoplasia in an experimental model of human pancreatic carcinoma. Cancer Res 61: 550-555, 2001[Abstract/Free Full Text].

20.   Meister, P. [Myofibroblasts. Review of outlook]. Pathologe 19: 187-193, 1998[ISI][Medline].

21.   Mueller-Klieser, W. Three-dimensional cell cultures: from molecular mechanisms to clinical applications. Am J Physiol Cell Physiol 273: C1109-C1123, 1997[Abstract/Free Full Text].

22.   Mueller-Klieser, W. Tumor biology and experimental therapeutics. Crit Rev Oncol Hematol 36: 123-139, 2000[ISI][Medline].

23.   Oda, D, Gown AM, Vande BJ, and Stern R. Instability of the myofibroblast phenotype in culture. Exp Mol Pathol 52: 221-234, 1990[ISI][Medline].

24.   Petroll, WM, Jester JV, Bean JJ, and Cavanagh HD. Myofibroblast transformation of cat corneal endothelium by transforming growth factor-beta1, -beta2, and -beta3. Invest Ophthalmol Vis Sci 39: 2018-2032, 1998[Abstract/Free Full Text].

25.   Powell, DW, Mifflin RC, Valentich JD, Crowe SE, Saada JI, and West AB. Myofibroblasts. I. Paracrine cells important in health and disease. Am J Physiol Cell Physiol 277: C1-C9, 1999[Abstract/Free Full Text].

26.   Powell, DW, Mifflin RC, Valentich JD, Crowe SE, Saada JI, and West AB. Myofibroblasts. II. Intestinal subepithelial myofibroblasts. Am J Physiol Cell Physiol 277: C183-C201, 1999[Abstract/Free Full Text].

27.   Rønnov-Jessen, L, and Petersen OW. Induction of alpha-smooth muscle actin by transforming growth factor-beta 1 in quiescent human breast gland fibroblasts. Implications for myofibroblast generation in breast neoplasia. Lab Invest 68: 696-707, 1993[ISI][Medline].

28.   Rønnov-Jessen, L, Van Deurs B, Nielsen M, and Petersen OW. Identification, paracrine generation, and possible function of human breast carcinoma myofibroblasts in culture. In Vitro Cell Dev Biol 28: 273-283, 1992.

29.   Schürch, W. The myofibroblast in neoplasia. Curr Top Pathol 93: 135-148, 1999[Medline].

30.   Serini, G, Bochaton-Piallat ML, Ropraz P, Geinoz A, Borsi L, Zardi L, and Gabbiani G. The fibronectin domain ED-A is crucial for myofibroblastic phenotype induction by transforming growth factor-beta1. J Cell Biol 142: 873-881, 1998[Abstract/Free Full Text].

31.   Serini, G, and Gabbiani G. Mechanisms of myofibroblast activity and phenotypic modulation. Exp Cell Res 250: 273-283, 1999[ISI][Medline].

32.   Shao, ZM, Nguyen M, and Barsky SH. Human breast carcinoma desmoplasia is PDGF initiated. Oncogene 19: 4337-4345, 2000[ISI][Medline].

33.   Sieuwerts, AM, Klijn JG, Henzen-Logmand SC, Bouwman I, Van Roozendaal KE, Peters HA, Setyono-Han B, and Foekens JA. Urokinase-type-plasminogen-activator (uPA) production by human breast (myo) fibroblasts in vitro: influence of transforming growth factor- beta(1) [TGF beta(1)] compared with factor(s) released by human epithelial-carcinoma cells. Int J Cancer 76: 829-835, 1998[ISI][Medline].

34.   Vande, BJ, Rudolph R, Poolman WL, and Disharoon DR. Comparative growth dynamics and actin concentration between cultured human myofibroblasts from granulating wounds and dermal fibroblasts from normal skin. Lab Invest 61: 532-538, 1989[ISI][Medline].

35.   Vaughan, MB, Howard EW, and Tomasek JJ. Transforming growth factor-beta1 promotes the morphological and functional differentiation of the myofibroblast. Exp Cell Res 257: 180-189, 2000[ISI][Medline].

36.   Wang, QP, Escudier E, Roudot-Thoraval F, Abd-Al S, I, Peynegre R, and Coste A. Myofibroblast accumulation induced by transforming growth factor-beta is involved in the pathogenesis of nasal polyps. Laryngoscope 107: 926-931, 1997[ISI][Medline].

37.   Wernert, N. The multiple roles of tumour stroma. Virchows Arch 430: 433-443, 1997[ISI][Medline].

38.   Xing, Z, Tremblay GM, Sime PJ, and Gauldie J. Overexpression of granulocyte-macrophage colony-stimulating factor induces pulmonary granulation tissue formation and fibrosis by induction of transforming growth factor-beta 1 and myofibroblast accumulation. Am J Pathol 150: 59-66, 1997[Abstract].

39.   Yokozeki, M, Baba Y, Shimokawa H, Moriyama K, and Kuroda T. Interferon-gamma inhibits the myofibroblastic phenotype of rat palatal fibroblasts induced by transforming growth factor-beta1 in vitro. FEBS Lett 442: 61-64, 1999[ISI][Medline].

40.   Zimmerman, CM, and Padgett RW. Transforming growth factor beta signaling mediators and modulators. Gene 249: 17-30, 2000[ISI][Medline].

Am J Physiol Cell Physiol 284(1):C209-C219
0363-6143/03 $5.00 Copyright © 2003 the American Physiological Society


This article has been cited by other articles:


Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
M. Pho, W. Lee, D. R. Watt, C. Laschinger, C. A. Simmons, and C. A. McCulloch
Cofilin is a marker of myofibroblast differentiation in cells from porcine aortic cardiac valves
Am J Physiol Heart Circ Physiol, April 1, 2008; 294(4): H1767 - H1778.
[Abstract] [Full Text] [PDF]