HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 1, 407-415, January 4, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/1/407    most recent M108103200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mason, J. M. Articles by Tabibzadeh, S. Search for Related Content PubMed PubMed Citation Articles by Mason, J. M. Articles by Tabibzadeh, S. Social Bookmarking                      What's this?

Lefty Contributes to the Remodeling of Extracellular Matrix by Inhibition of Connective Tissue Growth Factor and Collagen mRNA Expression and Increased Proteolytic Activity in a Fibrosarcoma Model^*

James M. Mason, Hao-Peng Xu^§, Srinivasa K. Rao^¶, Andrew Leask, Michele Barcia, Jidong Shan^§, Robert Stephenson, and Siamak Tabibzadeh^**

From the ^** Department of Pathology,  Gene Therapy Vector Laboratory, Department of Research, and ^§ Department of Molecular Oncology, North Shore-Long Island Jewish Research Institute and New York University School of Medicine, Manhasset, New York 11030, ^¶ Long Island Jewish Medical Center, Long Island Campus for the Albert Einstein College of Medicine, New Hyde Park, New York 11042, and  FibroGen, Inc., South San Francisco, California 94080

Received for publication, August 22, 2001, and in revised form, September 20, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Homeostasis of the extracellular matrix (ECM) of tissues is regulated by controlling deposition and degradation of ECM proteins. The breakdown of ECM is essential in blastocyst implantation and embryonic development, tissue morphogenesis, menstrual shedding, bone formation, tissue resorption after delivery, and tumor growth and invasion. TGF- family members are one of the classes of proteins that actively participate in the homeostasis of ECM. Here, we report on the effect of lefty, a novel member of the TGF- family, on the homeostasis of extracellular matrix in a fibrosarcoma model. Fibroblastic cells forced to express lefty by retroviral transduction lost their ability to deposit collagen in vivo. This event was associated with down-regulation of the steady-state level of connective tissue growth factor that induces collagen type I mRNA. In addition, lefty transduction significantly decreased collagen type I mRNA expression and simultaneously increased collagenolytic, gelatinolytic, elastolytic, and caseinolytic activities in vivo by the transduced fibroblasts. These findings provide a new insight on the actions of lefty and suggest that this cytokine plays an active role in remodeling of the extracellular matrix in vivo.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Tissues have a supporting fibrovascular framework that undergoes constant remodeling. Generally, in normal tissues, a balance is reached between the formation and destruction of extracellular matrices (ECMs),¹ leading to a state of homeostasis. Acceleration of formation of ECM is a necessary event in some conditions, such as healing of wounds, repair of endometrium after menstruation, and embryonic development. In other physiologic conditions such as tissue breakdown in menstruating endometrium, endometrial tissue invasion by blastocyst, uterine tissue resorption after delivery, tissue morphogenesis, and bone resorption, the degradation of extracellular matrices supersedes the formation of stroma. On the other hand, uncontrolled destruction of ECM contributes to tumor invasion, and unabated deposition of ECM occurs in fibrotic conditions such as scleroderma, postsurgical adhesions, cirrhosis, glomerulosclerosis, idiopathic pulmonary fibrosis (Hamman-Rich syndrome), keloid, and post-burn scarring. For these reasons, there is a great interest in identifying the full complement of factors that contribute to the remodeling of ECM.

The formation and breakdown of ECM requires precisely coordinated and controlled timely expression and activation of cytokines as well as ECM proteins and a host of enzymes that degrade diverse cellular and extracellular matrix proteins. Growth-regulatory cytokines of the transforming growth factor  (TGF-) family are one of the few classes of proteins that provide the necessary signals required in the homeostasis of a fibrovascular stroma (1-3). TGF- is a major profibrogenic cytokine that promotes the proliferation of fibroblasts, enhances CTGF and collagen type I mRNA expression, and suppresses the degradation of extracellular matrices by a dual action that involves down-regulation of the expression of ECM proteases such as 72-kDa gelatinase and stimulation of protease inhibitors such as tissue inhibitor of metalloproteinase 1 and plasminogen activator inhibitor 1 (4-9). These actions of TGF- have been described under various physiologic and pathologic conditions such as normal wound healing and scar formation (7) and in a number of fibroproliferative conditions (10-17). TGF- also supports tumor growth and enhances the development of tumor stroma through increased proliferation of fibroblasts and enhancement of ECM deposition in cancers including fibrosarcoma (18-20). We recently described a new function for lefty, a novel member of the TGF- family, as a potent inhibitor of TGF- signaling in vitro (21). Lefty perturbs TGF- signaling by inhibiting the phosphorylation of Smad2 after activation of the TGF- receptor. Moreover, lefty inhibits the events that lie downstream from R-Smad phosphorylation including heterodimerization of R-Smads with Smad4, nuclear translocation of R-Smad-Smad4 complex, and downstream gene transcriptional activities. Lefty opposes the effect of TGF- on the expression of reporter genes for major cell cycle factors p21 and Cdc25. Smad3 and Smad4 both have domains that bind the 5'-TCTGAGAC-3' termed Smad binding element. Lefty inhibits the TGF--induced promoter activity driven by Smad binding element. Moreover, it was recently shown that the expression of CTGF, which mediates the actions of TGF- and induces proliferation of fibroblasts and collagen synthesis, is driven by Smad3 and Smad4 (22). Lefty is also capable of inhibiting the TGF--mediated promoter activity of CTGF. Thus, lefty provides a repressed state of TGF--responsive genes and participates in negative modulation of TGF- signaling by inhibition of phosphorylation of R-Smads (21). Based on these observations, it can be predicted that lefty might function in a manner opposite to that induced by TGF- in vivo. As the first step toward understanding the biologic activity of lefty, in this study, we introduced lefty⁺ fibroblastic tumor cells in vivo. Our results support the model in which lefty impairs the CTGF and collagen mRNA expression and deposition of collagen and drives degradation of ECM by an increased proteolytic action that is comprised of collagenolytic, gelatinolytic, and elastolytic activities. The overall effect of these activities leads to remodeling and significant shrinkage of extracellular matrix in the stroma.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- The full-length, 1.961-kilobase ebaf (lefty A) cDNA was derived from a human placental cDNA library (23). The materials used in these studies included an enhanced chemiluminescence system (Roche Molecular Biochemicals), polyvinylidene difluoride membranes (Bio-Rad), Kodak-OMAT films (Sigma), biotin-labeled goat anti-rabbit antiserum, avidin-biotin-complex reagent (Vector Laboratories, Burlingame, CA) and protein G plus agarose (Santa Cruz Biotechnology, Santa Cruz, CA). Casein, gelatin, and Coomassie Brilliant Blue R were obtained from Sigma. All other chemicals were from either Sigma-Aldrich Co. or Fisher Scientific (Pittsburgh, PA). The affinity-purified A353 polyclonal antibody used in this study was raised to a peptide at the C terminus of the lefty A/B protein (24). The athymic nu/nu mice were obtained from Charles River Laboratory (Wilmington, MA).

Cells, Transfection, and Protein Preparation-- The GP+E86 fibroblastic cell line was maintained in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 10% heat-inactivated fetal bovine serum (Life Technologies, Inc.). For transduction, cells were seeded into 6-well plates (Falcon, Franklin Lakes, NJ) at a concentration of 1.3 × 10⁴ cells/ml and maintained in a CO₂ chamber at 37 °C for about 16 h. When cells reached 50% confluence, they were transduced with amphotropically packaged retroviral vectors, LG and LEIG, in presence of 8 mg/ml Polybrene as described previously (25).

Transient transfections were performed using plasmids pLG and pLEIG and calcium phosphate as described previously (25). Serum-free medium was collected 20-24 h after transfection, and the media were concentrated 12-fold using Centricon YM3 centrifugal filter devices (protein molecular size cut-off, 3 kDa; Amicon, Danvers, MA). Cells were lysed by the addition of 15 µl of Laemmli buffer. Protein concentration was determined by the Bio-Rad Protein Assay kit (Bio-Rad).

Affinity Purification of Lefty Protein-- Lefty proteins from conditioned culture media of cells stably transduced with lefty were affinity-purified as described recently (24). Protein concentration was determined by the Bio-Rad Protein Assay kit.

Fibroblast Cell Injections-- Animal studies were carried out after the approval of the institutional review board. For inoculation, cells in exponential growth phase were harvested by a brief exposure to 0.05% trypsin and 0.2% EDTA solution (w/v). The cell suspension was pipetted to produce a single-cell suspension. The cells were washed and resuspended in a serum-containing medium to the desired cell concentration. Cell viability was determined by trypan blue exclusion, and only single-cell suspensions of >90% viability were used. Cells were introduced to mice under anesthesia (90 mg/kg ketamine and 9 mg/kg Xylazaine). 5 × 10⁶ cells in a volume of 100 µl were injected into four to six different subcutaneous sites of the athymic (nu/nu) mice. At the termination of the study, tumors were resected from anesthetized mice. The animals were sacrificed on days 2, 14, and 21. About 50% of each tumor was flash-frozen in OCT in liquid nitrogen, and another 50% of each tumor was embedded in paraffin for preparation of paraffin sections. About 10% of the tumors removed on day 21 were processed for electron microscopy.

SDS-Polyacrylamide Gel Electrophoresis and Western Blotting-- SDS-polyacrylamide gel electrophoresis and Western blotting for lefty were carried out as described previously (24). Briefly, the conditioned media (12-15 µg protein/lane) were fractionated in a 12% gel together with prestained protein ladder (Life Technologies, Inc.) and subsequently blotted onto polyvinylidene difluoride membrane in a Mini-Trans-Blot apparatus (Bio-Rad). The blot was stained with the A353 affinity-purified rabbit anti-C terminus lefty A antibody (1-2 µg/ml). The secondary antibody was mouse anti-rabbit IgG-horseradish peroxidase (Santa Cruz Biotechnology). The specific bands were detected by chemiluminescence as described by the manufacturer. Silver staining was carried out according to the protocol provided by the manufacturer (Bio-Rad).

Immunohistochemcial Staining-- Immunohistochemical staining was performed according to the ABC procedure (26). Briefly, cryostat sections were fixed in 10% buffered formalin for 5 min and then washed in 0.1 M phosphate-buffered saline. Sections were incubated with primary antibody (1-2 µg/ml) followed by secondary biotin-labeled antibody (1-2 µg/ml). This was followed by incubation with avidin-biotin complex. Each incubation was for 30 min at room temperature, followed by a wash for 5 min in 0.1 M phosphate-buffered saline, pH 7.4. The sections were developed in diaminobenzidine-H₂O₂ and viewed at the light microscopic level without counterstain.

Transmission Electron Microscopy-- Tumor tissues were fixed in 2.5% phosphate-buffered glutaraldehyde, postfixed with osmium tetroxide and uranyl acetate, and then dehydrated in an ascending series of alcohol as described previously (27). After the 100% ethanol washes, cells and tissues were infiltrated with 50% ethanol/50% Epon for 30 min and infiltrated with 100% Epon for 20 h. Processed cells and tumor fragments were transferred to fresh 100% Epon and incubated at 56 °C for 48 h within plastic capsules to allow for polymerization of Epon. Thick sections were stained with toluidine blue, and thin sections were stained with lead acetate and examined with a JOEL transmission electron microscope.

Reverse Transcription-Polymerase Chain Reaction-- Total RNA was isolated from cells and tumors using RNeasy, a commercially available kit (Qiagen, Valencia, CA). Briefly, samples of each tumor were excised and homogenized with a polyron. The homogenates were spun, and supernatants were loaded onto a Qiagen RNeasy column. RNA was eluted with 50 µl of diethylpyrocarbonate-water, treated with DNA-free DNase, and quantified using a ribogreen kit. Equal amounts of RNA were transcribed into cDNA using omniscript and an oligo(dT) primer, as described by the manufacturer (PerkinElmer Life Sciences). Reverse transcription-polymerase chain reaction was carried out using the following primers: (a) murine COL1A1, ATGTTCAGCTTTGTGGACCTCCGG (forward) and CCTTGGGCCTTGGGGGCCAG (reverse); and (b) GAPDH GGTCATCCCTGAGCTGAACG (forward) and TTCGTTGTCATACCAGGAAA (reverse). Quantitative real-time PCR was performed according to the manufacturer's protocol (Roche) using the following primers: (a) COL1A1, TGGAAGAGCGGAGAGTAC (forward) and GCGCAGGAAGGTCAGCTG (reverse); (b) CTGF, TGACTGCCCCTTCCCGAGAA (forward) and TCTTCCAGTCGGTAGGCAGCTAGG (reverse); and (c) GAPDH, GGTCGGTGTGAACGGATTTGG (forward) and GCCGTGGGTAGAGTCATACTGGAAC (reverse).

Colorimetric Determination of Proteolytic Activity-- For determination of collagenolytic, gelatinolytic, and elastolytic activities, a colorimetric assay was used as described recently (28). Culture media or tissue lysates were incubated with succinylated collagen, gelatin, and elastin or with all three as substrates. The reactions were carried out at 37 °C. After 30 min of incubation, TNBSA (50 ml; 0.03% in 50 mM sodium borate, pH 8.5; Pierce) was added, and the optical density was measured at A₄₅₀ after 20 min at room temperature. Collagenase type I (216 units/mg; Calbiochem), gelatinase, and elastase were used as positive controls. Blank reactions (negative control) contained all components except the substrate.

Casein and Gelatin Zymography-- Casein and gelatin zymographies were carried out as described previously (29). Briefly, proteins extracted from tumors using Triton X-100 were mixed with Laemmli sample buffer in the absence of reducing agents. After 15 min of incubation at 37 °C, the proteins (10 µg protein/lane) were separated in 10% Tris-glycine gel with 0.1% gelatin incorporated as a substrate. The casein zymographies were carried out using 12% Tris-glycine gel with -casein incorporated as substrate. To remove SDS, gels were renatured after electrophoresis for 30 min at room temperature with gentle agitation. Gels were stained with Coomassie Blue R-250 for 30 min and destained with 10% acetic acid and 30% methanol. Zones of protein lysis appeared as clear bands within a blue background. Positive control consisted of trypsin type IX in casein zymography and collagenase type VII in the gelatin zymographies. Equal loading was verified by running identical gels using 10 µg protein/lane that were then stained with Coomassie Blue stain.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Expression and Release of Lefty Proteins by Lefty⁺ Cells-- In this study, we investigated the functional in vivo effects of human lefty A after transducing the fibroblastic cell line GP+E86 with retroviral particles. We constructed two retroviral vectors: (a) LG, a control vector enabling cells to express GFP, and (b) LEIG, a vector that induces the expression of both GFP and lefty A (Fig. 1). The success of these transduction experiments was assessed by analysis of GFP fluorescence. The GFP-positive cells were cloned by cell sorting using a cell sorter (Becton Dickinson). The sorted cells were maintained under routine culture conditions in Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal bovine serum. The majority of clonal cells transduced with LG and LEIG viral particles were fluorescent (Fig. 2, A and B). The GFP cDNA was driven by the retroviral long terminal repeat in the LG vector and by the weaker IRES in the LEIG vector (Fig. 1). After transduction, the LG-transduced cells exhibited an intense green fluorescence (Fig. 2A), and the LEIG-transduced cells had a dimmer fluorescence (Fig. 2B).

View larger version (7K): [in this window] [in a new window]   Fig. 1.   LG and LEIG retroviral vectors. Plasmid LX-lefty A was double-digested with SphI and XhoI to generate a 6155-bp fragment. Oligonucleotides NS204 (5'-AAAGATATCGCATGCCCTCTCCCTCCCCCCCCCCTAAC-3') and NS205 (5'-TTTGATATCCTCGAGTTACTTGTACAGCTCGTCCATGCC-3') were used as PCR primers with plasmid pIRES-eGFP (CLONTECH). PCR amplification generated a 1338-bp SphI/XhoI IRES-eGFP fragment that was cloned into the LX-lefty A plasmid to generate the plasmid LX-lefty A-IRES-eGFP (pLEIG). This retroviral vector plasmid was used to generate retroviral vector particles from GP+E86 cells. To create the control retroviral vector expression plasmid, pIRES-eGFP plasmid (LG vector), plasmid LX2, was digested with HindIII/BamHI. This digestion generated a 4782-bp fragment that contained the 5' and 3' Moloney murine leukemia virus long terminal repeats flanking the retroviral packaging signal and a multicloning site. Plasmid pIRES-eGFP was digested with HindIII/BclI, and the 774-bp fragment containing the IRES and eGFP sequences was ligated into the 4782-bp fragment to generate plasmid pLG. This vector was used as a control.

View larger version (68K): [in this window] [in a new window]   Fig. 2.   Localization of lefty and GFP proteins in the lefty and lefty+ clonal cells. A and B, GP+E86 cells transduced with LG (A) and LEIG retroviral particles (B) were trypsinized, and cell pellets were deposited on slides and examined by fluorescence microscopy. C and D, LG (C)- and LEIG (D)-transduced cells were grown over polyglycolic acid mesh fibers, and the cryostat sections of meshes frozen in OCT were immunostained for lefty using the A353 antibody.

The ability of the transduced cells to express lefty was first analyzed by immunostaining. The LG- and LEIG-transduced GP+E86 cells were deposited on glass slides and immunostained using A353, an affinity-purified polyclonal antibody to lefty (24). Whereas the LG-transduced cells failed to show positive immunoreactivity, the LEIG-transduced cells expressed lefty (data not shown). The cells were also grown on polyglycolic acid fibers, and 7 days after culture, the meshes were frozen in OCT medium (30). Cryostat sections of cells grown over the fibers were then immunostained for lefty. The LG-transduced GP+E86 cells adherent to polyglycolic acid failed to exhibit positive staining, whereas the LEIG-transduced cells showed a strong staining throughout their cytoplasm (Fig. 2, C and D).

Following these observations, the ability of LEIG-transduced cells to secrete lefty proteins into the culture medium was evaluated. The conditioned media of LG- and LEIG-transduced cells were subjected to Western blotting after a 20-40-fold concentration in Centricon devices with a molecular cutoff of 10,000 kDa. The blots were probed with A353 antibody. We recently reported that lefty protein is secreted as a 42-kDa precursor and two cleaved 28- and 34-kDa proteins (24). Lefty proteins were purified from the culture media of LEIG-transduced cells (24). When subjected to silver staining, affinity-purified lefty proteins of LEIG-transduced cells showed the presence of 42-, 34-, and 28-kDa lefty proteins (Fig. 3, lane 1). Moreover, the affinity-purified material was subjected to gel electrophoresis followed by Western blotting. Staining showed the presence of 42-, 34-, and 28-kDa forms of lefty (Fig. 3). The secreted lefty proteins were biologically active because they induced mitogen-activated protein kinase (24). The immunofluorescence and secretion of lefty into the culture medium were routinely monitored. All cells maintained their fluorescence, and the LEIG-transduced cells continued to secrete lefty into the medium for a period of >1 year.

View larger version (41K): [in this window] [in a new window]   Fig. 3.   Purification of lefty from culture media of lefty+ clonal cells. Lefty proteins were purified from culture media of the LEIG-transduced cells as described recently (21). 1 µg of the purified material was subjected to silver staining (lane 1). 200 ng of the purified protein were subjected to Western blotting and stained without (lane 2) and with A353 antibody to lefty (lane 3). Size is shown in kDa.

The cell morphology of the GP+E86 cells, as assessed with a phase-contrast microscope, was not different in the LEIG-transduced cells as compared with the LG-transduced cells. In both cultures, cells had spindle to stellate appearance. These findings show that lefty does not significantly impact the morphologic phenotypes of transduced cells in two-dimensional cultures.

Lefty Leads to Shrinkage of ECM in Vivo-- To gain an insight on the deposition of extracellular matrix by lefty, the LG- and LEIG-transduced cells were introduced subcutaneously, and the tumors were removed on days 2, 14, and 21. Examination of Hematoxylin and eosin-stained sections of both tumors on these days showed a sarcomatous growth that was comprised of spindle cells forming the zebra pattern, typical of fibrosarcomas (Fig. 4). However, individual or clusters of fibroblasts were separated by an abundant extracellular matrix deposited both in the central and peripheral regions of tumors derived from LG-transduced cells (Fig. 4A). The fluorescence microscopy of these lesions showed the highly fluorescent green fibers typical of collagen (Fig. 4B). In marked contrast, the sections of the tumors derived from LEIG-transduced cells showed highly compact and tightly adherent cells (Figs. 4C and 5). Very little intervening extracellular matrix and collagen could be observed between these cells (Fig. 4, C and D).

View larger version (146K): [in this window] [in a new window]   Fig. 4.   Deposition of ECM in the tumors derived from lefty and lefty+ cells. 5 × 106 LG- and LEIG-transduced cells were injected into subcutaneous sites in each athymic nude mouse. Skin and tissues underlying the injection sites were removed and frozen in OCT medium on day days 2, 14, and 21. Hematoxylin and eosin-stained representative sections of the tumors removed on day 21 were visualized under light microscopy and are shown. B and D are negative images of the same fields shown in A and B visualized by fluorescence microscopy. A and B, LG tumor; C and D, LEIG tumor. ECM is seen in Fig. 1 as homogeneous pink sheets between other tumor components (solid arrows). Dashed arrows in B and D point to the fluorescent collagen fibers between the tumor cells.

View larger version (154K): [in this window] [in a new window]   Fig. 5.   Deposition of ECM in the tumors derived from lefty and lefty+ cells. Tumors removed on day 21 and shown in Fig. 4 were processed for visualization at the ultrastructural level. A, LG tumor; B, LEIG tumor. Bars, 5 µm. Inset, high magnification of the box showing collagen fibers. Bar, 500 nm.

To quantitate the amount of collagen deposited in tumors, sections were stained with trichrome, which, in view of its affinity, casts a blue color onto collagen fibers. Whereas collagen fibers were detected in large amounts at both the center and periphery of the LG tumors (Fig. 6, A and C), the LEIG tumors exhibited a paucity of these fibers at both the center and peripheral areas (Fig. 6, B and C).

View larger version (78K): [in this window] [in a new window]   Fig. 6.   Deposition of collagen in tumors derived from lefty and lefty+ cells. Sections of day 21 tumors derived from LG (A)- and LEIG (B)-transduced cells were stained with trichrome to visualize the collagen fibers. Solid arrows in A point to the collagen fibers and surrounding tumor cells. Dashed arrows in A point to muscle fibers invaded by the tumor cells. Arrows in B point to tumor cells without visible surrounding collagen fibers. Bar, 250 µm. C, the amount of collagen was quantitated in trichrome-stained sections by morphometric analysis using a Leitz Aristoplan microscope fitted with a Spot 2 digital camera and a digital image analysis system (MetaView/MetaMorph System; Universal Imaging Corp., West Chester, PA). The mean total area/pixel density of the trichrome-positive areas was measured per microscopic field, at ×4 magnification. Three separate, randomly selected fields for each sample were analyzed in both the center and peripheral regions of tumors. The values shown are the means ± S.D. of triplicate measurements in tumor sections.

Examination of the tumors at the ultrastructural level substantiated the light microscopic findings. Abundant extracellular matrix and deposits of collagen fibers were readily seen in the LG tumors, whereas cells in LEIG tumors were tightly coherent, and little extracellular matrix and intervening collagen containing stroma was present (Fig. 5). Taken together, these findings show that lefty leads to the shrinkage of extracellular matrix and significantly reduces the amount of collagen deposited by fibroblastic cells. We considered that two events might underlie these observations: decreased synthesis of ECM proteins and increased degradation. These considerations were tested by the following series of studies.

Lefty Suppresses the Steady-state Level of CTGF and Collagen Type I mRNA in Vivo-- Because the amount of collagen deposited in tumors derived from lefty-transduced cells was significantly reduced, we considered the possibility that this effect might be due to down-regulation of transcription of collagen type I mRNA. To show this, total RNA was extracted from the LG- and LEIG-transduced cells grown in vitro and from the tumors derived from subcutaneous inoculation of these cells into athymic mice. Equal amounts of total RNA from cells and tumors were subjected to reverse transcription followed by PCR using a pair of oligonucleotides to murine collagen type I. There was no significant difference in the expression of collagen I mRNA levels in the LG- and LEIG-transduced cells maintained in vitro (Fig. 7A). Collagen I mRNA was also readily detected in the tumors derived from LG-transduced cells (Fig. 7A). However, there was a marked reduction of collagen type I mRNA in the tumors derived from LEIG-transduced cells (Fig. 7A). These findings show that lefty inhibits the collagen type I mRNA in vivo but not in vitro.

View larger version (45K): [in this window] [in a new window]   Fig. 7.   CTGF and collagen mRNA expression in lefty- and lefty+-transduced cells and tumors derived from lefty and lefty+ cells. LG- and LEIG-transduced cells were injected into subcutaneous sites in each athymic nude mouse. The RNA was extracted from the LG- and LEIG-transduced cells cultured in vitro and tumors removed on day 21. A, 2 µg of each RNA sample were subjected to reverse transcription using avian myeloblastosis virus reverse transcriptase. One-eighth of the resulting cDNAs of each sample was PCR amplified using a pair of mouse primers specific to collagen type 1 (top panel) and to GAPDH (bottom panel). The amplified products were resolved in a 2% agarose gel. Lanes 1 and 8, 100-bp molecular weight (MW) markers. Sources of RNA included: Lane 2, bovine meniscus used as a positive control (+Cont); Lane 3, in vitro cultured LG-transduced GP+E86 cells; Lane 4, in vitro cultured LEIG-transduced GP+E86 cells; Lane 5, LG-transduced GP+E86 cells grown as tumors in vivo; Lane 6, LEIG-transduced GP+E86 cells grown as tumors in vivo; Lane 7, negative control (Cont, no DNA added). The size of the amplified collagen (327 bp) and GAPDH (294 bp) bands corresponded to those expected. The 300- and 400-bp markers are marked. B, 500 ng of RNA from each tumor sample was subjected to real-time quantitative PCR using a pair of mouse primers specific to CTGF and collagen type 1 and GAPDH. Values shown are the means ± S.D. of four determinations.

CTGF, a member of the CCN (CTGF/Cyr61/NOV) family of growth regulators, is a secreted cysteine-rich, heparin-binding, 38-kDa protein. CTGF is considered to be an immediate early growth responsive gene and a downstream mediator of TGF- actions in fibroblasts (31). CTGF induces chemotaxis in mesenchymal cells and promotes proliferation of fibroblasts and collagen synthesis in a number of in vitro and in vivo models (32-35). Moreover, because of overexpression in human cancers, CTGF is considered to be part of the molecular pathways that lead to the formation of tumor stroma by TGF- (36-39). CTGF appears to be a mediator of fibrotic reactions in a host of fibroproliferative diseases and in the paraneoplastic condition pseudo-scleroderma, which develops in some patients with lung cancer (36, 40). Following the earlier observations, we carried out quantitative real-time reverse transcription-polymerase chain reaction to determine the amount of both CTGF and collagen type I mRNA in the same tumor tissues (Fig. 7B). These studies confirmed the results of reverse transcription-polymerase chain reaction and showed a 2.8-fold reduction of CTGF mRNA and a 4.7-fold reduction in the expression of collagen type I mRNA in the tumors derived from lefty⁺ cells as compared with the control tumors. These results are consistent with the histologic data on reduced deposition of collagen in tissue sections of tumors and suggest that such reduction is, at least in part, the result of reduced CTGF and collagen mRNA transcription.

Lefty Suppresses ECM Accumulation by Enhanced Proteolytic Activities-- To gain an insight on the biological role of lefty in degradation of extracellular matrices, the proteins derived from tumors removed on day 21 were subjected to casein and gelatin zymographies. This analysis showed that proteins derived from lefty⁺ and not lefty fibroblastic tumors degrade heat-denatured casein and gelatin (Fig. 8). Notably, five distinct bands were detected in the gelatin zymographies and at least four bands were detected in casein zymographies of proteins derived from tumors derived from lefty⁺ cells. Similar bands were not detected in the gelatin or casein zymographies of the proteins from tumors derived from lefty fibroblasts (Fig. 8).

View larger version (85K): [in this window] [in a new window]   Fig. 8.   Proteolytic activity in tumors derived from lefty and lefty+ cells. The proteins derived from the tumors derived from lefty and lefty+ cells were subjected to gelatin (A) and casein (B) zymographies. Arrowheads point to the areas of dissolution of gelatin in the background of Coomassie Blue-stained gel. MW, molecular weight standards. Pos Cont, positive control consisting of collagenase type VII (A) and trypsin type IX (B).

Because of their highly hydrophobic nature, the elastic fibers are quite resistant to proteolysis (41). Under normal conditions, these fibers undergo a minimal turnover (42), but under certain physiologic conditions such as menstruation (43), various pathologic states such as aortic aneurysm (44), and carcinomas (45), they undergo extensive proteolysis. Because lefty was maximally expressed in endometrium in the background of menstrual bleeding (27), is overexpressed in solid human tumors (46) and is involved in the development of cardiovascular system (47), elastin fibers, like collagen, might also be a target for proteolysis induced by lefty. To gain additional insight into the biological role of lefty in the degradation of extracellular matrix and to independently validate the zymographic findings, we first assessed the effect of culture media of lefty⁺ and lefty cells on degradation of collagen, gelatin, and elastin. Elastin fibers, which confer resilience to tissues, are comprised of cross-linked monomers of tropoelastin. Cells were incubated overnight in serum-free media. These media were concentrated ~40-fold in Centricon 10,000-kDa devices. The effect of these concentrated media on lysis of collagen, gelatin, and elastin was simultaneously quantitated by a colorimetric method that measures the lysis of substrates. This analysis failed to show significantly different dissolution of these substrates by the concentrated tissue culture media of LG- and LEIG-transduced cells (Fig. 9).

View larger version (11K): [in this window] [in a new window]   Fig. 9.   Proteolytic activity in lefty and lefty+ cells in vitro. The combined proteolytic activity of the culture media of LG- and LEIG-transduced cells was quantitated by using collagen, gelatin- and elastin together as substrate for lysis. The amount of lysed substrates was quantitated by the colorimetric method as described in the text.

In contrast to the in vitro findings, there was a significant increase in the total proteolytic activities in the proteins derived from LEIG-transduced tumors as compared with the proteins derived from LG-transduced tumors when all three substrates were used (data not shown). We then individually tested the effect of lefty on degradation of collagen, gelatin, and elastin. This analysis revealed about a 13.7-fold increase in collagenolytic activity in tumors derived from lefty⁺ cells as compared with those that were derived from lefty cells (Fig. 10). Moreover, 22-fold increased lytic activity for gelatin and 13.7-fold increased elastolytic activity were also noted in the tumors derived from lefty⁺ clonal cells (Fig. 10). Insoluble elastin fibers are degraded by serine (48, 49) and cysteine proteinases (50) as well as by several matrix metalloproteases including the 92- and 72-kDa gelatinases, macrophage metalloelastase, and matrilysin (51-53). It was recently shown that treatment of fibroblasts with tropoelastin or with heterogenic peptides leads to the expression of MMP-1 and MMP-3, which have collagenase activity (54). These finding suggest that inhibition of elastase activity might reduce collagenase activity as well. To test such a possibility and to identify whether the elastase activity induced by lefty belongs to the serine protease or other non-serine proteinases, the collagenolytic and elastolytic assays were performed in presence of serine protease inhibitor ₁-antitrypsin. The elastolytic activity induced by lefty could be significantly (3-fold decrease) inhibited by ₁-antitrypsin (Fig. 11). However, this inhibition did not cause a significant shift in collagenolytic activity in the presence of ₁-antitrypsin (Fig. 11).

View larger version (11K): [in this window] [in a new window]   Fig. 10.   Collagenolytic, gelatinolytic, and elastolytic activities in tumors derived from lefty and lefty+ cells. The collagenolytic (A), gelatinolytic (B), and elastolytic (C) activity of the proteins from tumors derived from LG- and LEIG-transduced cells was quantitated by using collagen, gelatin, and elastin as substrate for lysis. The amount of lysed substrates was quantitated by the colorimetric method as described in the text.

View larger version (15K): [in this window] [in a new window]   Fig. 11.   Inhibition of collagenolytic and elastolytic activities in tumors derived from lefty and lefty+ cells by serine proteinase inhibitor 1-antitrypsin. The collagenolytic (A) and elastolytic (B) activity of the proteins derived from tumors derived from LG- and LEIG-transduced cells was quantitated by using collagen and elastin as substrate for lysis in the absence () and presence (+) of 1-antitrypsin. The amount of lysed substrates was quantitated by the colorimetric method as described in the text.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Lefty is normally expressed in mesenchymal cells and fibroblasts. In the mouse embryo, lefty-1 gene is expressed in mesodermal cells, and in humans, it is highly expressed in endometrial stromal fibroblasts immediately before the shedding of the endometrial tissue (23). Lefty is expressed at a low level in other tissues such as colon, ovary, testis, and pancreas (46). Lefty is also overexpressed in certain forms of human cancer (46). Currently, there is no information about the biological consequences of lefty gene expression in normal or malignant tissues in vivo. In this report, we showed a new function for lefty by stable expression of lefty in fibroblastic tumor cells and analysis of the effect of lefty on deposition of ECM. Introduction of lefty into these tumor-producing cells significantly impaired the formation of ECM in vivo by a dual mechanism of action that involved inhibition of CTGF and collagen mRNA synthesis and increased degradation of ECM proteins.

We recently demonstrated that lefty does not inhibit the promoter activity of CTGF on its own but is able to impart a significant inhibition on CTGF promoter activity induced by TGF- (21). The results reported here extend these earlier in vitro observations and show that CTGF mRNA expression is significantly down-regulated in vivo in the tumors derived from lefty⁺ fibroblasts. These data allowed the prediction that the reduced expression of CTGF in lefty⁺ tumors would lead to reduced steady-state level of collagen type I mRNA. Our findings clearly demonstrated the validity of this prediction and showed that both the collagen type I mRNA and accumulation of collagen and extracellular matrix in lefty⁺ tumors were significantly reduced. The extent of deposition of extracellular matrix and collagen was greatly diminished in both the central and leading edge of the tumors transduced with lefty as compared with their control counterparts. These latter tumors showed highly sclerotic centers, and a significant amount of collagen fibers was present at the peripheral borders of such tumors. These findings support the view that lefty regulates the deposition of extracellular matrix and collagen in the stroma. The reduced collagen in the ECM could be attributed, to some extent, to the decreased steady-state level of CTGF and collagen I mRNA.

Besides regulation of synthesis, deposition of ECM proteins is controlled by their rate of degradation. The breakdown of extracellular matrix is essential in embryonic development, tissue morphogenesis, menstrual shedding, bone formation, tissue resorption after delivery, blastocyst implantation, and tumor growth and invasion. This breakdown requires precisely coordinated and controlled timely expression and activation of cytokines and a host of enzymes that degrade diverse cellular and ECM proteins. We showed that lefty significantly induces a proteolytic cascade in vivo that is comprised of collagenolytic, gelatinolytic, elastolytic, and caseinolytic activities. One possible explanation for these effects is that lefty itself has enzymatic activity. Lefty is a member of the TGF- family of molecules, and another member of this family, bone morphogenetic protein-1, which is also known as procollagen C-proteinase, has been shown to possess metalloprotease activity (23, 55, 56). Bone morphogenetic protein-1 cleaves the C-terminal propeptides of procollagen types I, II, and III. Lefty shows a very low homology to bone morphogenetic protein-1 and lacks the highly conserved cysteine switch sequence (PRCG(V/N)PD) and the catalytic domain with a zinc binding motif (HEXXHXXGXXH) (57) that are shared by all matrixin family members (21). It was recently shown that treatment of fibroblasts with tropoelastin or with heterogenic peptides obtained after organo-alkaline or leukocyte elastase hydrolysis of insoluble elastin induced high expression of pro-MMP-1 and pro-MMP-3 (54). However, lefty lacks the VGVAPG or the consensus sequence GXXPG, which was found to be the minimal domain required for binding to elastin receptor on fibroblasts to trigger signals that lead to the expression of MMP-1 and MMP-3 (54). Moreover, lefty failed to induce enhanced enzymatic activity for collagen or elastin in vitro. These findings argue against the possibility that lefty possesses an enzymatic activity that directly causes cleavage of collagen, gelatin, or elastin.

A second possible likely scenario is that lefty induces proteolytic activity in fibroblasts or other cell types present in vivo or acts as an inhibitor for those cytokines such as TGF- that suppress the activation of enzyme(s) that cause degradation of ECM proteins (58-60). Among the possible enzymes that lefty actions might target are serine proteinases, soluble or membrane-bound MMPs, or fibroblast activation protein. Fibroblast activation protein is a cell membrane serine prolyl oligopeptidase and gelatinase that acts as a dual-specificity dipeptidyl peptidase and collagenase and is expressed by sarcomas and fibroblastic cells in areas of active tissue remodeling (57, 61, 62). Another group of candidates are members of the ADAM family that have recognized roles in proteolysis of extracellular matrix components (63). The serine proteinase inhibitor ₁-antitrypsin significantly inhibited the elastase activity induced by lefty but did not inhibit the collagenase activity induced by lefty. These findings show that lefty is capable of inducing the activity of both the serine and non-serine proteinase class of enzymes. The serine protease family is one of the oldest and largest multigene families either secreted or sequestered in the membrane and is well positioned to interact with other proteins that contain extracellular serine protease domains (64). This family includes proteases with elastolytic activity, but elastin is also degraded by enzymes that are found in thiol, aspartic, and metallo enzymes (65).

Of the known MMPs, the 92- and 72-kDa mouse and human macrophage metalloelastase and matrilysin all degrade insoluble elastin. MMPs, also known as matrixins, are a family of highly homologus zinc metalloenzymes that collectively digest virtually all extracellular matrix proteins and constituents of basal lamina including collagen, gelatin, elastin, entactin, laminin, and fibronectin (57, 66). Thus far, at least 22 soluble and 6 different membrane-type MMPs have been reported (57, 67, 68). Some members of this family, such as 72-kDa gelatinase A (MMP-2), 92-kDa gelatinase B (MMP-9), and stromelysins have broad substrate specificity and digest several ECM proteins including collagen, denatured collagen (gelatin), fibronectin, and laminin, whereas others, such as the newly described uterus-specific endometase, cleave type I gelatin but do not digest collagens, laminin, elastin, or -casein (62, 66, 69). Collagenase 2 (MMP-8), which is predominantly expressed in the postpartum and involuting uterus, has preferential activity against type I collagen (70). Once activated, some members of this family, such as membrane-type MMPs, activate other members of the MMP family such as MMP-2 (71-73). Strikingly, stromelysin-1 (MMP-3) proteolytically activates at least five other members of the MMP family including interstitial collagenase-1 (MMP-1), matrilysin (MMP-7), neutrophil collagenase-2 (MMP-8), 92-kDa type IV collagenase (MMP-9), and collagenase-3 (MMP-13), suggesting that stromelysin-1 holds a special "upstream" role in ECM degradation and remodeling (74-79). For this reason, we examined the expression of MMP-3 in lefty⁺ and lefty tumors by immunoblotting. Although we failed to show an increased expression of MMP-3 in the lefty⁺ tumors (data not shown), incubation of endometrial explants with recombinant Escherichia coli lefty caused a significant increase in MMP-3 expression.² These findings show that the effect of lefty on MMP expression may be contextual.

Our findings show that lefty acts in a manner that is distinctly opposite to the effects of TGF- on fibroblasts in vivo. For several reasons, an attractive model is one in which the in vivo actions of lefty on CTGF and collagen mRNA expression, as well as reduced accumulation and increased degradation of ECM proteins, might be mediated by inhibition of the signaling of TGF-. First, there is a close correlation between the expression of TGF- and the extent of fibrosis in human adenocarcinomas that exhibit central fibrosis (79). The overexpression of TGF- in PANC-1 pancreatic carcinoma cells led to desmoplasia in experimental tumors in the pancreas of athymic mice (80). In contrast to these effects of TGF-, in our fibrosarcoma model, the central fibrosis seen in the control tumors was abated by introduction of lefty into the tumor cells. Similarly, the deposition of collagen at the peripheral regions of tumors was distinctly reduced by the use of lefty⁺ cells. Second, lefty inhibits diverse gene transcriptional activities that are driven by TGF- (21). CTGF is transcriptionally up-regulated by TGF- by a pathway that recruits Smad3 and Smad4, making it a likely target for lefty actions (22). Consistent with this, lefty inhibited the CTGF promoter activity induced by TGF- in vitro (21) and reduced CTGF mRNA expression in vivo. Our studies showed that lefty is capable of down-regulating collagen type I mRNA, one of the best-characterized TGF--induced genes (81, 82), in fibroblastic tumors. Also, the effects of lefty were similar to other approaches that inhibit the TGF- activities such as administration of soluble betaglycan, antisense TGF-, and antibody to TGF-, which all led to decreased accumulation of extracellular matrix (83-86). Finally, TGF- suppresses the expression and activation of enzymes that degrade ECM proteins (4-9), whereas lefty exerts an effect that is distinctly opposite to these actions of TGF-. Collectively, these arguments suggest that lefty might act in vivo by inhibiting the TGF- action, thereby contributing to decreased accumulation of ECM in vivo.

In conclusion, the findings presented here suggest that lefty is a member of a family of proteins that participate in the homeostasis and remodeling of the ECM deposited by fibroblastic cells. In addition, the findings from this study are significant because they provide a broader understanding that lefty regulates the expression of CTGF and collagen mRNA, deposition of ECM and collagen, and degradation of ECM proteins in vivo. A challenging task would be to identify the in vivo pathways through which lefty regulates CTGF and collagen mRNA expression and characterize the enzymatic cascade that lefty utilizes to cause the remodeling of ECM by increasing proteolytic activities. Such an insight would be quite pertinent to a wide variety of physiologic conditions such as menstrual shedding, blastocyst implantation and embryogenesis, wound repair, repair of endometrium after menstruation, and to human diseases that are associated with aberrant fibroblast proliferation as well as development of tumor stroma.

	ACKNOWLEDGEMENTS

We thank Dr. Peter Mento for help with the morphometric analysis of collagen in trichrome-stained sections and Ana Kuth for preparation of histological sections.

	FOOTNOTES

^* This work was supported by a grant from Lexon Inc. (to S. T.).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.

To whom correspondence should be addressed: Frontiers in Bioscience, P. O. Box 160, Searingtown, NY 11507. Tel.: 516-484-2831; Fax: 516-484-2831; E-mail: tabibzadeh@bioscience.org.

Published, JBC Papers in Press, September 25, 2001, DOI 10.1074/jbc.M108103200

² K. Osteen and S. Tabibzadeh, unpublished data.

	ABBREVIATIONS

The abbreviations used are: ECM, extracellular matrix; TGF-, transforming growth factor ; CTGF, connective tissue growth factor; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PCR, polymerase chain reaction; GFP, green fluorescence protein; IRES, internal ribosomal entry site; MMP, matrix metalloprotease; bp, base pair(s).

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				H. Xu, J. Shan, V. Jurukovski, L. Yuan, J. Li, and K. Tian TSP50 Encodes a Testis-Specific Protease and Is Negatively Regulated by p53 Cancer Res., February 1, 2007; 67(3): 1239 - 1245. [Abstract] [Full Text] [PDF]

				P. Bonniaud, M. Kolb, T. Galt, J. Robertson, C. Robbins, M. Stampfli, C. Lavery, P. J. Margetts, A. B. Roberts, and J. Gauldie Smad3 Null Mice Develop Airspace Enlargement and Are Resistant to TGF-{beta}-Mediated Pulmonary Fibrosis J. Immunol., August 1, 2004; 173(3): 2099 - 2108. [Abstract] [Full Text] [PDF]

				P. B. Cornet, C. Picquet, P. Lemoine, K. G. Osteen, K. L. Bruner-Tran, S. Tabibzadeh, P. J. Courtoy, Y. Eeckhout, E. Marbaix, and P. Henriet Regulation and Function of LEFTY-A/EBAF in the Human Endometrium. mRNA EXPRESSION DURING THE MENSTRUAL CYCLE, CONTROL BY PROGESTERONE, AND EFFECT ON MATRIX METALLOPROTEINASES J. Biol. Chem., November 1, 2002; 277(45): 42496 - 42504. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/1/407    most recent M108103200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mason, J. M. Articles by Tabibzadeh, S. Search for Related Content PubMed PubMed Citation Articles by Mason, J. M. Articles by Tabibzadeh, S. Social Bookmarking                      What's this?