0013-7227/92/1305-2476$03.00/O Endocrinology Copyright 0 1992 by The

Phenotypic Cells That Factor+l*

Vol. Printed

Endocrine Society

Alterations Overexpress

130, No. 5 in U.S.A.

in Fibroblasts and Fibrosarcoma Latent Transforming Growth

R. DANIEL BEAUCHAMP, HONG-MIA0 SHENG, DUNCAN A. MILLERS, RUSSETTE M. LYONS& AMIONEll, AND HAROLD L. MOSES

CHARLES GUILLERMO

Department of Surgery (R.D.B., H.-M.S.) and Human Biological Texas Medical Branch, Galveston, Texas 77550; and Department School of Medicine, Nashville, Tennessee 37232

C. BASCOMP, TORRE-

Chemistry (R.D.B.), The University of of Cell Biology, Vanderbilt University

2B cells that expressed higher levels of TGFpl also expressed high levels of c-sis and c-myc mRNAs and decreased TGFP2 and TGFP3 mRNAs in the same manner as parental AKR-2B cells that had been treated with active TGFPl. The transfected 1591 cells that overexpressed TGFBl bound less [1251]TGF@l than did parental 1591 cells, but after a mild acid wash demonstrated an increase in [‘251]TGFf11 binding. Our results suggest that these TGFBl-transfected fibroblast and fibrosarcoma cells have the capacity to activate TGFP; however, as very little activated TGFp is detected in the medium, it is hypothesized that these cells activate latent TGFPl and bind the activated TGF/31, thus acquiring a phenotype consistent with TGFfll-treated cells. (Endocrinology 130: 2476-2486, 1992)

ABSTRACT. Mouse embryo-derived AKR-2B fibroblasts and murine fibrosarcoma cells (the 1591 cell line) were transfected with a murine transforming growth factor+31 (TGFfll) cDNA under the transcriptional control of either the simian virus-40 early promoter or the cytomegalovirus promoter/enhancer. Selected clones secreted 2- to 4-fold more TGFP-competing activity into their media than the parental cell line or neomycin-transfected controls. The TGF/31 released into the cell-conditioned medium was latent. Despite the latency of the overexpressed TGFfll, the TGFfll-transfected cells exhibited phenotypic features of TGF/31-treated cells. When confluent, the TGF@ltransfected cells had the morphological characteristics of the parental cells that have been treated with active TGFPl. AKR-

T

HE BIOLOGICAL activities of transforming growth factor-p1 (TGF@l) are numerous and diverse (for review, see Ref. 1). While it is a potent growth inhibitor for most cell types, TGF/31 can stimulate the growth of fibroblastic cells in soft agar (2) and is mitogenie for selected connective tissue cells in monolayer (3, 4). The mitogenic effect of TGF/3 on monolayer cultures of connective tissue cells is probably an indirect one mediated through the induction of platelet-derived growth factor (PDGF) (4-6) and displays delayed kinetics compared to those of other mitogens. TGFPl also causes striking morphological changes in monolayer cul-

tures of mouse fibroblast cells (7). TGFPl is secreted by numerous cell types in culture and is released in an inactive, or latent, form (8, 9). TGFpl is synthesized as a 390-amino acid prepro form that contains a typical hydrophobic signal sequence of 29 amino acids, which is cleaved to yield pro-TGF@l (10, 11). The carboxyl-terminal 112 amino acid residues are cleaved from the N-terminal glycopeptide region at a dibasic cleavage site as predicted from analysis of cDNA clones coding for TGFPl (10-12). The carboxyl-terminal 112-amino acid peptide, linked to an identical 112-amino acid peptide by interchain disulfide bonds, forms the biologically active 25kilodalton (kDa) TGFPl molecule. Disruption of the disulfide linkages results in total loss of activity (13). Studies of recombinant TGFPl produced by transfected Chinese hamster ovary cells (14) reveal that the 25-kDa mature TGFP remains in a noncovalent association with the N-terminal glycopeptide portion of the TGFPl precursor as a llO-kDa latent complex (15). This latent form of TGF/31 cannot compete with mature active TGFP for cell surface receptors, but can be activated by methods that release the 25kDa homodimer from the precursor. Acidification (a process that disrupts

Received November 6, 1991. Address all correspondence and requests for reprints to: R. Daniel Beauchamp, M.D., Department of Surgery, University of Texas Medical Branch, Galveston, Texas 77550. * This work was supported by USPHS Grants CA-01309 (to R.D.B.), CA-48799 (to H.L.M.), and CA-42572 (to H.L.M.). t Present address: Procter & Gamble, P.O. Box 599, Cincinnati, Ohio 45201. $ Present address: Unigene Laboratories, Inc., 110 Little Falls Road, Fairfield, New Jersey 07004. § Present address: Genetic Therapy, Inc., 19 Firstfield Road, Gaithersburg, Maryland 20878. 1)Present address: Department of Internal Medicine, Baylor College of Medicine, Houston, Texas 77054. 2476

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

TGFpl

TRANSFECTION

the noncovalent interaction between the mature TGFP and the amino-terminal glycopeptide) has become a standard method of in vitro activation, and with a few exceptions, most of the studies of the biological effects of TGF/3 have used TGFp that is activated in this manner. Plasmin, a serine protease, can activate TGFP by cleavage within the amino-terminal glycopeptide, thereby releasing the mature carboxyl-terminal homodimer (15, 16), and may have a physiological role in TGFp activation (17). Previous studies (18) using anti-TGFP antibodies to inhibit the spontaneous anchorage-independent growth of the chemically transformed AKR-2B (AKR-MCA) cells suggested that TGFP can be activated and act in an autocrine fashion. To further examine the autocrine effects of TGFPl, we have selected clones of AKR-2B fibroblast cells that have been stably transfected with the entire coding region of the murine TGFpl cDNA under the transcriptional control of either the simian virus-40 (SV40) early promoter (pSVTGFfi1) or the cytomegalovirus (CMV) promoter-enhancer (pSV2NEOCMV-TGF/?l) and have also transfected the mouse fibrosarcoma cell line 1591 with pSVTGF/Il (1921). Overexpression of latent TGFPl coincides with several phenotypic changes in these cells, similar to those observed when untransfected cells are treated with purified active TGF/31. We conclude that the frbroblast and tibrosarcoma cells can activate and respond to endogenously produced TGFPl. Materials Cell culture and collection

and Methods of conditioned

medium

(CM)

The AKR-2B (and a clone 84A) cell line was derived from mouse embryonic mesenchymal tissue and has been previously characterized (22). The 1591 cells are derived from a fibrosarcoma tumor line from UV-exposed C3H mice (19, 20). The 1591 cells expressvery little TGF/31, and TGFP2 and TGFP3 mRNAs are undetectable (21). Mouse embryo-derived AKR2B fibroblasts and the transfectants were grown in McCoy’s 5A medium containing 5% fetal bovine serum (FBS; Armour Pharmaceutical Co., Kankakee, IL). The 1591Re cells (19) and transfected clones were grown in Dulbecco’s Modified Eagle’s Medium supplementedwith 10% FBS. Serum-free media conditioned by the cell cultures were collected as previously described(21). CM werethen stored at 4 C for usewithin 2 weeks or were stored at -20 C for later assays.To activate latent TGFp in the CM, aliquots were acidified to a pH of 1.5 with 12 N HCl and incubated for 1 h at 4 C, then reneutralized with 10 N NaOH to pH 7.2. This acid-treated reneutralized medium is referred to as acid-activated CM. Untreated CM is referred to as neutral CM. For assayson epithelial cells, either a cloned cell line of BALB/c mouse keratinocytes (BALB/MK) or a mink lung epithelial cell line (CCL-64, American Type Culture Collection, Rockville, MD) was used. BALB/MK cells were grown in Minimum Essential Medium containing 0.05 mM

OF FIBROBLASTS

2417

calcium supplementedwith 8% dialyzed FBS (Hazelton ResearchProducts, Inc., Lenexa, KS) and epidermalgrowth factor (4 rig/ml; Collaborative Research, Inc., Waltham, MA). The CCL-64 cells were cultured in McCoy’s 5A medium supplemented with 10% FBS. Plasm&

used for transfection

The pSVTGFP1 plasmidwaspreparedby introduction of the 1.6-kilobase(kb) mouseTGF@l cDNA (12) into the EcoRI site of pKCR3 (23). The construct (Fig. 1A) has the SV40 early promotor and origin of replication, portions of rabbit &globin exons 2 and 3 that are not translated, a splice site, and a rabbit P-globin polyadenylation signal. TGFPl is under the transcriptional control of the SV40 promoter. The pSV2NEOCMV-TGFpl plasmid wasprepared by ligating the mouse TGFPl cDNA into the EcoRI site of pSV2NEOCMV. This placesthe TGF@l cDNA under the transcriptional regulation of the CMV enhancer/promoter (24). This plasmid contains the neomycin resistancegeneunder the transcriptional control of the SV40 early promoter and in the oppositeorientation to the CMV TGFPl. The neomycin resistance gene provided the selectablemarker in all experiments. In cotransfection experiments, neomycin resistancewas conferred by either the pRSVNE0 or the pZipNE0 (25) plasmids. Transfection

of cell

Exponentially growing cells were trypsinized and seededat 5 x lo5 cells/lo-cm plate. The ARK-2B cells were incubated overnight in 10 ml McCoy’s 5A mediumwith 5% FBS, and the 1591 Re cells were grown in Dulbecco’sModified Eagle’s Medium supplementedwith 10% FBS. The medium was replaced with fresh medium, and cells were transfected by calcium phosphate-DNA coprecipitation (26,27). They weretransfected with either pRSVNE0 (2 pg) with 18 rg carrier salmonsperm DNA or pRSVNE0 (2 pg) and pSVTGFP1 (18 rg). In separate experiments, transfections were performed using the pSV2NEOCMV-TGFpl vector (20 pg). After 6 h, the cellswere shockedwith 15%glycerol for 90 set (28). The cellswere gently washedwith medium and then incubated for 24 h before selection of stable transformants was begun. Antibiotic-resistant clones were selected in G418 medium. Colonies were then randomly picked, amplified, and assayedfor TGFP-competing activity in a RRA. Further subcloningwasaccomplishedby the limiting dilution method. Transfection and cloning of the 1591 cells were describedpreviously (21). Northern

blot analysis

RNA was extracted from cell cultures by the method of Schwabet al. (29). Oligo(dT)-selectedpolyadenylated [poly(A)] RNA was separatedby electrophoresisin 1.2% agarose-formaldehyde gels,and Northern blotting wasperformed, as previously described(30). The TGFPl probe is a 974-basepair(bp) SmaI fragment derived from the mousecDNA (12). The TGF/32 riboprobe correspondsto nucleotides 1511-1953of the murine TGF/32 cDNA clone (31). The TGFP3 riboprobe corresponds to nucleotides831-1440of the murine TGFP3 cDNA (32). The c&s riboprobe correspondsto the 2.0-kb PstI-AccI fragment derived from the human c-sis cDNA clone pSM-1 (33) sub-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

TGF/31 TRANSFECTION

2478

A

BomHl

m

SV40

Promoler

-

TCFBI

Rabbit

O-globin

0

AMP’

B

OF FIBROBLASTS

FIG. 1. Expression of TGF@l in transfected and control AKR-2B cells. A, Map of pSVTGFP1 construct. The mouse TGFPl cDNA was ligated into the EcoRI site of pKCR3. B, Northern analysis demonstrating that the Bl.l C4 cells contain and express the transfected construct. Total cellular poly(A) RNA was collected from AKR-PB, NC 6.1, and B1.l C4 cells that had reached confluenceby continuedgrowth in McCoy’s 5A medium supplemented with 5% fetal calf serum. In addition, AKR-2B cells were treated for 48 h with TGF@l (0.1 and 10 ng/ ml) before extraction of poly(A) RNA. Northern blot analysis was performed, and each lane contains 2 rg poly(A) RNA that was hybridized to a murine “P-labeled TGFPl cDNA, the 32P-labeled rabbit figlobin cDNA EcoRI/XhoI insert from the pSVTGF@l construct (see A), or a 32P-labeled rat lB15 cDNA.

cloned into pGEM 2. The mousec-myc probe correspondsto exons 2 and 3 and was derived from the pSVc-myc plasmid (34). The rabbit @globin cDNA probe was the 527-bp EcoRI/ XhoI fragment correspondingto the 3’-region of exon III (Fig. 1A). The human type I plasminogenactivator inhibitor (PAI-

l

1992 No 5

1) probe was a 3.0-kb EcoRI insert (35). All blots were hybridized with cyclophilin cDNA probe (lB15) (36) to determine intactness of RNA and to normalize gel loading and transfer, becausecyclophilin mRNA levels are not altered by TGFPl treatment. The cDNA probes were radiolabeledby the random primer method (37), and the riboprobes were labeled by in uitro transcription, as previously described(38). Hybridizations and posthybridization washes were performed at 43 C for cDNA probes and at 65 C for riboprobesin 5 X SSC, 50%formamide, 1 X Denhardt’s solution, 0.1%sodium dodecyl sulfate (SDS), 50 pg/ml polyadenosine,and 250 pg/ml salmonsperm DNA. Three consecutive 20-min posthybridization washeswere performed at the sametemperature as hybridization in 0.1 x SSC-0.1% SDS. After washeswere completed, the filters were exposed to x-ray film with an intensifying screen at -70 C. Autoradiographic signal intensity was determined by laser densitometry. r251]TGF/31

1Blti

Endo. Voll30

binding

experiments

After washing the cells five times with Modified Eagle’s Medium and 10% FBS, they were incubated for 2 h at 22 C in HEPES binding buffer (1) with 0.25 ng *251-labeled TGF/31 in a l-ml total volume. Nonspecific binding was determined by the addition of 50 rig/ml pure porcine TGFPl (R & D Systems, Inc., Minneapolis, MN). [‘251]TGF/31comigrateswith unlabeled TGF@l on SDS-polyacrylamide gels and retains more than 90% bioactivity, as demonstrated by the ability to stimulate AKR-2B (84A) cells to form colonies in soft agar (39) (Lyons, R. M., unpublished data). Total binding is routinely -7-15% of the total counts per min added.Nonspecific binding wasless than 25% of the total [1251]TGFblbound. For some experiments, a mild 3-min acid wash (150 mM sodiumchloride, 0.1% acetic acid, and 0.1% BSA) wasemployed before ligand binding. This procedure removes endogenously produced TGF/3 that may occupy receptor sites and has been describedpreviously (40). This experiment was performed to detect occupiedbinding sites on 1591cells. TGFP bioassays

Soft agar assaysusing AKR-2B (clone 84A) cells were performed as previously described (2, 41). Briefly, baselayers of 0.8% agarosein McCoy’s 5A medium supplementedwith 10% FBS were prepared in 35mm culture dishes.Upper layers of 0.4% agarosein McCoy’s 5A medium contained 10% FBS, 7.5 x lo3 cells, and the CM after the indicated treatments, as indicated. In parallel experiments, acid-activated CM was incubated with anti-TGF@ immunoglobulin G (100 Fg/ml) overnight at 4 C before use in soft agar assays.The number of colonieslarger than 50 pm that developedafter 7-10 days was quantitated using an Omnicon III image analyzer (Bausch & Lomb, Inc., Rochester,NY). Selectedmediawere alsotested for the ability to inhibit [3H] thymidine incorporation in mousekeratinocytes (BALB/MK) or mink lung epithelial cells (CCL-64) growing at logarithmic rates in monolayer, aspreviously described(7, 42, 43).

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

TGFPl

TRANSFECTION

Results Expression of TGFP by AKR-2B

OF FIBROBLASTS A

cells and transfectants

We have previously shown that the 1591 cells that were stably transfected with pSVTGFP1 express the transfected gene in abundance, as demonstrated by probing with either mouse TGFPl cDNA or the EcoRI/XhoI rabbit fl-globin cDNA fragment (21). This is in contrast to the very low level of TGFpl expression in parental 1591 cells. Similarly, Northern blot analysis revealed that there is much greater level of TGFPl mRNA in the TGF@l-transfected AKR-2B cell clone designated B1.l C4 than in either the AKR-2B parental cells or the RSVNEO-transfected (NC 6.1) controls (Fig. 1B). There were two major mRNA sizes for TGFPl from the B1.l C4 cells compared to the single 2.5-kb signal in control cells. The smaller signal is similar to the size of the endogenous mouse TGFPl mRNA (12). The larger mRNA may be due to alternative processing or polyadenylation. Probing with the rabbit @-globin EcoRI/XhoI cDNA revealed expression of rabbit fl-globin mRNA only in B1.l C4 cells, which indicated that the increased TGFpl expression was due to the transfected gene. The TGFPl mRNA sizes in the B1.l C4 cells precisely matched those observed upon probing with rabbit pglobin cDNA fragment. Five additional G418-resistant clones were selected from the same cotransfection experiment; however, none of these clones exhibited increased TGF/31 mRNA levels, nor did they express rabbit /3globin mRNA (data not shown), and these clones were not studied further. Acid-activated CM from the TGFpl-transfected (B1.l C4) cells exhibited increased TGFp bioactivity compared to that in neomycin-transfected control cells (NC 6.1) or parental AKR-2B cells by several different bioassays (Fig. 2A). Acid-activated CM from B1.l C4 cells had greater inhibitory activity on [3H]thymidine incorporation in BALB/MK cells than did control medium (Fig. 2A). The increased release of latent TGFp by the B1.l C4 cells was also confirmed by demonstrating that acidactivated CM (10% concentration) from the B1.l C4 cells could inhibit [3H]thymidine incorporation in proliferating BALB/MK cells by 50%, whereas the same concentration of acid-activated CM from NC 6.1 or AKR2B cells resulted in less than 10% inhibition of DNA synthesis. Neutral CM had no effect at any of the concentrations tested. The TGF/3 bioactivity was latent and was not detectable unless the medium was transiently acidified, as described in Materials and Methods. There was less than 0.1 rig/ml TGFP bioactivity in undiluted crude neutral CM from AKR-BB, NC 6.1, and B1.l C4 cells (data not shown). Analysis of the media conditioned by selected clones of the 1591 cells that were stably transfected with pSVTGFP1 revealed a similar abun-

F g

2479

70000, 60000-

2 50000m B 40000b E : 30000z 200002 7 ,I

- AKR-26 -NC61

10000-

- I31 1 c4

01 0

200

100 ConditIoned

-01

.l

medium

1

TGFpl CondItIoned

.01

.l

added

10

100

(rig/ml) media

(~1)

1 TGFpl Conditioned

10 @g/ml) media

(pl)

1000

100

1000

(pl)

FIG. 2. Transfected cells release more TGFp bioactivity. A, Increasing concentrations of acid-activated CM (50-200 ~1) from AKR-BB, NC 6.1, and B1.l C4 cells were added to proliferating BALB/MK cells and incubated for 22 h. At this time, the cells were incubated with 2.0 &i/ ml [3H]thymidine for 2 h, and DNA synthesis was determined from trichloroacetic acid-precipitable counts. Equal concentrations of neutral CM did not contain detectable BALB/MK inhibitory activity (data not shown). B, Increasing concentrations of TGFpl or acid-activated CM from AKR-2B (2B AA), NC 6.1 (6.1 AA), or B1.l C4 (C4 AA) cells were added to AKR-2B (clone 84A) cells in soft agar and incubated at 37 C for 7-10 days. At this time, colonies larger than 50 pm were counted. Neutral CM from any of the clones did not stimulate growth in soft agar of the AKR-2B (clone 84A) cells (data not shown). C, TGF@l and acid-activated B1.l C4 cell CM stimulated anchorageindependent growth of the 84A tibroblasts. These effects were inhibited by anti-TGFpl antibody 282 (100 pg/ml), TGFPl plus antibody, or B1.l C4 CM plus antibody. The soft agar assays were performed as described in Materials and Methods.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

TGFPl

2480

TRANSFECTION

dance of latent, but very little active, TGFP competing activity in the medium (21). Acid-activated B1.l C4 CM (515% concentration) also caused a concentration-dependent increase in the growth of AKR-2B (clone 84A) in soft agar, whereas medium from control cells (AKR-2B or NC 6.1) did not stimulate the growth of 84A cells in soft agar (Fig. 2B), an effect that could be blocked by coincubation with anti-TGFP antibodies (Fig. 2C). Neutral CM from all clones failed to stimulate soft agar colony formation (data not shown). Morphology of parental and transfected cells

The first indication that the TGF/31-overexpressing cells could potentially activate and respond to endogenously produced latent TGF@l was based upon the pronounced alteration in morphology of the B1.l C4 cells (Fig. 3). Clone NC 4.1 (and NC 6.1, not shown) and the parental AKR-2B cells became flattened when confluent and, within 24 h of exposure to TGF/31 (10 rig/ml), underwent a distinctive change in morphology. DoseControl

TGFpl(10

r&ml)

AKRBB

NC 4.1

FIG. 3. Morphological comparison of SV40-TGFpl-transfected cells. AKR-2B, NC 6.1, and B1.l C4 cells were seeded onto sterile coverslips in 60-mm dishes containing McCoy’s 5A medium supplemented with 5% fetal calf serum and grown until confluent. At this time, TGFPl (10 rig/ml) was added directly to the cultures and incubated for 24 h. After fixation, the cells were photographed at x200. Bar = 50 pm.

OF FIBROBLASTS

Endo. Voll30.

1992 No 5

response studies (not shown) showed that the complete morphological transformation required at least 0.3 ng/ ml active TGF/31. In contrast, the confluent B1.l C4 cells always had the appearance of TGFP-treated AKR-2B cells, and their morphology was not changed upon exposure to exogenous active TGF/?l (Fig. 3). Similarly, active TGF/31 caused a characteristic morphological change in parental 1591 cells, and clones that overexpressed the transfected TGFPl had a morphological appearance identical to that of parental cells that had been treated with active TGFPl (data not shown). When sparsely plated, it was not possible to detect differences in morphology by light microscopy. Suramin is a polyanionic compound that can inhibit the binding of TGFp (and other growth regulatory polypeptides) to AKR-2B cell surface receptors and can inhibit growth factor-stimulated DNA synthesis in AKR2B cells (44). A conversion of morphology to the control appearance occurred when B1.l C4 cells were exposed to suramin (Fig. 4). Expression of TGFpl -responsive genes in parental and transfectant AKR-2B cells

Leof et al. (5) have previously shown that serumdeprived quiescent AKR-2B cells stimulated with TGF/31 exhibit an induction of c-sis expression within 20 min of TGFPl treatment, followed by an induction of c-fos (detectable at 4 h) and a concomitant induction of c-myc expression, which peaks 8-12 h after TGFPl exposure. In the present study, confluent AKR-2B cells growing in complete medium expressed c-myc, but there was a much greater level of c-myc expression in the B1.l C4 transfectants that overexpress TGF@l, comparable to the level of expression observed in AKR-2B cells after 48-h exposure to TGFPl (10 rig/ml; Fig. 5). The doses of TGFPl used to treat AKR-2B cells were based upon the amount of TGFP-competing activity in neutral CM (0.1 rig/ml) and an amount (10 rig/ml) that has maximal biological effects on AKR-2B cells. Treatment was carried out for 48 h to determine the effects of long term treatment of AKR-2B cells with TGFfil under the assumption that the transfected cells are continuously exposed to higher levels of active TGF/31. B1.l C4 cells expressed high levels of c-&s mRNA, as did the TGFpl-treated (10 ng/ ml) AKR-2B cells, whereas the untreated NC 6.1 cells and the control AKR-2B cells (untreated or treated with 0.1 rig/ml TGFPl) expressed very low (or undetectable) levels of c-sis mRNA (Fig. 5). A characteristic action of TGFPl on AKR-2B cells is the induction of PAIexpression (45). PAImRNA levels were essentially undetectable in parental AKR-2B cells, but were abundant in TGFpl-treated cells and TGF@l-transfected cells (Fig. 5). Treatment with 0.1 ng/ ml TGF/31 (the amount of TGFP bioactivity detected in

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

TGFPl

TRANSFECTION

OF FIBROBLASTS

2481

c-myc

c-eis

PAI-

lB15

FIG. 4. Suramin treatment of B1.l C4 cells. The B1.1 C4 cells were grown to confluence in 35-mm plates. Either vehicle (McCoy’s 5A medium; A) or suramin (0.5 mM; B) was added to the cultures 24 h before photomicroscopy at x200. Bar = 50 pm.

neutral CM from all of the clones) was not sufficient to induce any of these typical responses to TGFPl. Additional transfection experiments were performed using the pSV2NEOCMV-TGF@l plasmid in order to obtain more AKR-2B clones expressing increased TGF@l. When transiently acidified CM from five of these clones was assayed for TGFp bioactivity by the ability to inhibit [3H]thymidine incorporation into trichloroacetic acid-precipitable material in mink lung epithelial cells, it was found that three of five clones released into their medium approximately twice the TGF/3 bioactivity per lo6 cells as the control cells (Table 1). Clones of AKR-2B cells that were stably transfected with pSV2NEOCMV-TGFfil and exhibited increased expres-

FIG. 5. Expression of TGFj31-inducible genes is increased in the SV40TGF+31-transfected AKR-2B cells. Poly(A) RNA was isolated and purified from AKR-BB, NC 6.1, and B1.l C4 cells, as previously described. Northern analysis was performed, and each lane represents 2 pg poly(A) RNA that was hybridized to either 32P-labeled human csis RNA or murine c-myc, human PAI-1, or rat lB15 cDNA probes, as described in Materials and Methods.

sion of TGFPl were similar in morphology to the B1.l C4 clone (see Fig. 3), whereas cells transfected with pRSVNE0 appeared no different from parental cells (data not shown). Four of five clones that were selected after transfection with the pSV2NEOCMV-TGFpl plasmid exhibited increased mRNA species that hybridized to the TGFPl cDNA probe (Fig. 6, lanes 4, 5, 6, and 8) compared to controls (Fig. 6, lanes l-3). c-sis mRNA was detectable only in those clones with increased TGF@l mRNA and

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

2482 TABLE

TGF/31 TRANSFECTION

OF FIBROBLASTS

Endo. Vol130*No5

1. TGFfi bioactivity in AKR-2B cells transfected with CMV-TGFfil Transfected clones (pSVZNEOCMV-TGF@l)

TGF/3 bioactivity (ng/ ml. lo6 cells)

CB4

CB6

CBlO

CBll

4.88 + 0.06”

4.36 k 0.15”

4.75 f 0.14”

3.46 + 0.17

Controls CB12 2.87 f 0.29

AKR-2B 2.21 * 0.29

NC51 2.35 f 0.39

Serum-free CM were collected from loo-mm tissue culture plates. Cell counts were obtained from each plate. The TGFp bioactivity was estimated by assaying for the ability to inhibit [3H]thymidine incorporation in CCL-64 cells. The activities in 10,50, and 100 ~1 CM were compared to a TGF61 standard curve, and the amount of TGFB activity (mean f SEM of triplicate determinations) was expressed as nanograms per ml/IO6 cells. oP < 0.01 compared to controls, by analysis of variance.

12345678 4

28STGFBl 18S9Y ,, 28Sc - sis 18S-

FIG. 6. Overexpression of TGF@l and c-sis in AKR-2B cells transfected with pSVZNEOCMV-TGFBl. Separate clones, designated CB 4, 6,10,11, and 12 (lanes 4-8; see Table l), were selected in G418 medium after transfection with the pSVZNEOCMV-TGF@l plasmid. Clones NC 5.1 and NC 6.1 (lanes 2 and 3) were selected after transfection with RSVNEO, and AKR-2B parental cells (lane 1) were not transfected. Each lane contains 3 pg poly(A) RNA, and hybridizations were performed with murine TGF61 cDNA, human c-si.s riboprobe, or the lB15 cDNA after labeling with 32P.

was particularly abundant in those clones (CB4, CB6, and CBlO) that released more TGFp activity into their medium. We have previously reported that the pSVTGF@l-transfected 1591 cells that overexpress TGFPl also have higher fibronectin mRNA levels, similar to the parental 1591 cells after treatment with TGFPl (10 rig/ml) for 24 h (46). TGF/31 treatment of AKR-2B cells leads to downregulation of TGFP2 and TGFP3 mRNA levels (47). In the present study, AKR-2B and NC 6.1 cells expressed readily detectable levels of TGFP2 and TGFP3. After a 48-h exposure to TGFPl(l0 rig/ml), there was a decrease in TGF/32 and TGFP3 expression in control cells, similar to the decreased levels of expression observed in untreated B1.l C4 cells (Fig. 7). Receptor occupation by endogenously produced TGFfl The 1591 cells and their TGFPl transfected clones (1591-N-TGF-1-4) were chosen for these studies because

TGFf32

TGFfl3

FIG. 7. TGFBl-repressible genes are down-regulated in the SV40TGF/31-transfected AKR-2B cells. Poly(A) RNA was isolated and purified from AKR-BB, NC 6.1, and B1.1 C4 cells, as previously described. Each lane contains 2 pg poly(A) RNA that was hybridized to either a 32P-labeled murine TGFfi2 riboprobe or a 3ZP-labeled murine TGF@3 riboprobe.

the parental cells expressed very little TGFPl, and TGF/32 and TGFP3 transcripts were essentially unde-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

TGFpl

TRANSFECTION

tectable by Northern blot analysis (21). The 1591-NTGF/31 clones released 3-10 ng TGFp-competing activity/ml. lo5 cells into their medium, but less than 0.3 ng/ ml was detected as the active form (46). A mild acid wash can dislodge TGFpl from its receptors and uncover receptors occupied by endogenous ligand (40). When binding was examined after either PBS or mild acid wash, there was no difference noted in [1251]TGF/31 binding to the parental 1591 cells; however, in the transfected clones that overexpressed TGFPl, there was fewer [‘251]TGF@1binding sites if the cells were washed with Modified Eagle’s Medium only. After the acid wash, there was greater capacity for [‘251]TGF/31 binding (Fig. 8). This suggests that receptor sites in the TGFpl-producing clones were occupied by an endogenously produced ligand, and that after removal of the ligand by acid wash, these binding sites were available to bind exogenously added [1251]TGF/31. Discussion To study the potential autocrine actions of TGFPl in cultured mouse fibroblast (AKR-2B) and mouse fibrosarcoma (1591) cells, increased expression of TGFPl has been achieved by stable transfection and selection of G418-resistant clones. The cDNA with the entire TGFpl-coding region was used, so that the TGF/31 precursor would be expressed, processed normally, and presumably released in its native latent form, as others have 4l

Control

Experiment

1

Experiment

2

8. Decreased binding of [‘*‘I]TGFPl in cells that produce more TGFbl. 1591, control cells, or various 1591 TGFP-producing clones (1591-N-TGF-1, -2, -3, and -4) were grown to confluency in monolayer cultures, harvested, washed three times, and then recultured at 2 X W/24-well plate. Eighteen hours later, the cells were washed three times with Modified Eagle’s Medium. At this point, the cells were treated with either acid (W) or PBS (IS) for 3 min. The cells were washed five times with Modified Eagle’s Medium-10% FBS, and [lz51] TGF(31 binding to the cells was determined, as described in Materials and Methods. Exp 1 represents data pooled from two different TGFPproducing clones, 1591-N-TGF-1 and -2, while Exp 2 represents data pooled from TGFP-producing clones, 1591-N-TGF-3 and -4. All TGFPproducing clones make 3-10 ng TGF@/105 cells.ml CM. Control data are pooled from Exp 1 and 2. The data are expressed as the mean counts per min f SEM. FIG.

OF FIBROBLASTS

2483

previously demonstrated when expressing recombinant TGFPl (14, 48-50). At this point, we do not know if the clones used in the present study produce the latent TGF@l-binding protein (LTBP) that is a component of the secreted large latent TGFPl complex demonstrated in platelets, human erythroleukemia (HEC) cells, and human fibroblasts (51, 52). The function of the LTBP is not yet clear, but it does not bind and inactivate mature TGFfil (51), nor is it required for production of recombinant latent TGF@l by COS-1 cells (53, 54). As expected, in the present study the TGFP released into the medium conditioned by the transfected cells was latent, and no more TGFP bioactivity was detectable in neutral (or nonactivated) transfectant CM than in the neutral CM from the parental cells until the medium was transiently acidified. Preincubation of the acid-treated CM with the TGF&neutralizing antibodies blocked the ability of the CM to stimulate soft agar growth of the AKR (clone 84A) cells. Thus, by several different bioassays of TGFP activity, we have shown that some of the transfected clones produced considerably more TGFp activity than the control cells. The TGF/3 secreted by the transfected cells is latent, as was TGFP that was produced by nontransfected cells that produced detectable TGFp bioactivity. Similarly, when overexpression of human TGFPl was induced by introduction of a retroviral vector into rat kidney fibroblasts (NRK-49F) and rat kidney epithelial cells (NRK-49E), most of the TGFp was released in a latent form (55). Interestingly, there appeared to be a greater increase in TGF/31 mRNA levels compared to the modest 2- to 3fold increase in secreted latent TGFpl bioactivity. Similar results were obtained with the transfected 1591 cells, as reported previously (46). There are several possible explanations for these observations: 1) there may be translational or posttranslational regulation that prevents production of the fully processed latent TGFp; 2) the TGFpl may not be entirely secreted and may reside intracellularly; 3) the excess TGFp protein may be shunted into intracellular degradative pathways; 4) the TGFp may become activated and bind to intracellular receptors in an “intracrine” manner; 5) the secreted TGFPl may be largely bound in the extracellular matrix or by other binding proteins; or 6) the presence or absence of LTBP may affect the processing, release, or activation of TGFfll. Further studies are required to resolve these possibilities. The finding that acid wash can unmask binding sites in TGF@l overproducing 1591-N-TGF cells suggests that the endogenously produced TGF/3 has bound to these cell surface receptors. Such receptor binding could occur either before or after the receptors appear on the cell surface, but once on the cell surface, the binding can be dislodged by acid washing.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

2484

TGFPl

TRANSFECTION

Active TGFPl causes a striking morphological alteration of quiescent AKR-2B cells in monolayer (2, 7). Similarly, the 1591 cells used in the present study undergo a characteristic change in morphology after exposure to TGFPl. In the absence of exogenous TGF/3 treatment, the TGF@l-transfected cells have the same appearance as TGFP-treated parental cells when they are near confluency. Expression of TGFpl-inducible genes by the TGF/31-transfected AKR-2B and 1591 cells, the down-regulation of TGFP2 and TGFP3 mRNAs in the transfected AKR-2B cells, and the evidence for high receptor occupancy in the TGF@-transfected 1591 cells are all consistent with the hypothesis that the TGFPltransfected cells were not only expressing more latent TGFfil, but were activating and responding to it as an autocrine factor. The observed morphological changes are probably related to the regulation of extracellular matrix components (56-58) and cell adhesion receptors (59, 60) by TGF/31, but further studies are required to confirm this. Sato and Rifkin (17) recently observed that coculture of bovine endothelial cells with pericytes leads to TGFp activation. The diverse biological effects of TGFP activation are blocked by inhibitors of plasmin, and these results provide evidence for a physiological mechanism of TGFp activation involving plasmin. Furthermore, Torre-Amione et al. (21,46) have recently demonstrated that the highly immunogenic mouse fibrosarcoma cell line (1591) transfected with pSVTGFP1 resulted in overproduction and release of latent TGFP activity into the medium. The TGF/31-overexpressing clones of the 1591 cells retained the cytolytic T-lymphocyte target antigen, but stimulated little or no antigen-specific cytolytic Tlymphocyte responses in vitro or in L&O. This action could not be reproduced by merely using untreated cell supernatants in which the TGFp activity is latent, but appeared to require contact between the fibrosarcoma cells and the lymphocytes. Overexpression of TGFpl by the transfected fibrosarcoma cells allowed them to escape the immune surveillance mechanism, and tumors grew in syngeneic animals under conditions that ordinarily lead to immunologically mediated regression of tumors derived from the parental (nontransfected) fibrosarcoma cells (21). Shipley et al. (3) previously demonstrated that TGFPl stimulates DNA synthesis in quiescent ARK-2B cells after a prolonged prereplicative interval of approximately 24 h compared to the 12-h prereplicative interval required for epidermal growth factor, PDGF, and fibroblast growth factor. The mitogenic effect of TGFp on AKR2B cells appears to be mediated through the induction of c-sis (PDGF B-chain) (5), which induces a state of competence in these cells, thus enabling them to progress through the cell cycle. In quiescent fibroblasts, PDGF

OF FIBROBLASTS

Endo. Voll30.

1992 No 5

rapidly induces expression of c-fos and c-myc and a number of other “early event” genes that lead to DNA synthesis (61-63). The constitutively increased expression of c&s along with the increased c-myc that we observed in the TGFpl-transfected AKR-2B cells suggest that these cells may have acquired a state of competence without exogenously added growth factors. In summary, the TGFpl-transfected cells that were studied release 2- to 3-fold more latent TGFP bioactivity than control cells. Despite this increase in latent TGFP, there appeared to be no increase in active TGFP in transfectant cell-conditioned medium compared to that in controls. This level of TGFP activity in neutral CM was not sufficient to cause the typical TGFpl-induced phenotypic alterations in AKR-2B cells. We, therefore, conclude that there is activation of the latent TGF/31 either at the cell surface or intracellularly; however, the activated TGF/31 probably binds rapidly to the high affinity cell surface receptors, thus being unavailable for detection in cell-conditioned medium. This activated receptor-associated TGFpl results in numerous phenotypic alterations. Acknowledgments

We wish to thank Mary Aakre, Jan Winkle, and Wilfred Saito for their excellent technical assistance. We thank Kelly Lee for secretarial assistance. References 1. Massague J 1987 Identification of receptor for type-0 transforming growth factor. Methods Enzymol 146:174-195 2. Moses HL, Branum EB, Proper JA, Robinson RA 1981 Transforming growth factor production by chemically transformed cells. Cancer Res 41:2842-2848 3. Shipley GD, Tucker RF, Moses HL 1985 Type /3 transforming growth factor/growth inhibitor stimulates entry of monolayer cultures of AKR-2B cells into S phase after a prolonged prereplicative interval. Proc Nat1 Acad Sci USA 82:4147-4151 4. Battegay EJ, Raines EW, Seifert RA, Bowen-Pope DF, Ross R 1990 TGF-(3 induces bimodal proliferation of connective tissue cells via complex control of an autocrine PDGF loop. Cell 63:515524 5. Leof EB, Proper JA, Goustin AS, Shipley GD, DiCorleto PE, Moses HL 1986 Induction of c-si.s mRNA and activity similar to plateletderived growth factor by transforming growth factor 8: a proposed model for indirect mitogenesis involving autocrine activity. Proc Nat1 Acad Sci USA 83:2453-2457 6. Soma Y, Grotendorst GR 1989 TGF-(3 stimulates primary human skin fibroblast DNA synthesis via an autocrine production of PDGF-related peptides. J Cell Physiol 140:246-253 7. Shipley GD, Childs DB, Volkenant ME, Moses HL 1984 Differential effects of epidermal growth factor, transforming growth factor, and insulin on DNA and protein synthesis and morphology in serum-free cultures of AKR-2B cells. Cancer Res 44:710-716 8. Lawrence DA, Pircher R, Kryceve-Martinerie C, Jullien P 1984 Normal embryo fibroblasts release transforming growth factors in a latent form. J Cell Physiol 121:184-188 9. Pircher R, Jullien P, Lawrence DA 1986 P-Transforming growth factor is stored in human blood platelets as a latent high molecular weight complex. Biochem Biophys Res Commun 136:30-37 10. Derynck R, Jarrett JA, Chen EY, Eaton DH, Bell JR, Assoian RK,

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

TGFpl

11. 12. 13.

14.

15. 16. 17. 18. 19. 20.

21.

22. 23.

24.

25. 26. 27. 28. 29.

30.

TRANSFECTION

Roberts AB, Sporn MB, Goeddel DV 1985 Human transforming growth factor-p complementary DNA sequence and expression in normal and transformed cells. Nature 316:701-705 Sharples K, Plowman GD, Rose TM, Twardzik DR, Purchio AF 1987 Cloning and sequence analysis of simian transforming growth factor-p cDNA. DNA 6239-244 Derynck R, Jarrett JA, Chen EY, Goeddel DV 1986 The murine transforming growth factor-p precursor. J Biol Chem 261:43774379 Assoian RK, Komoriya A, Meyers CA, Miller DM, Sporn MB 1983 Transforming growth factor-8 in human platelets. Identification of a major storage site, purification, and characterization. J Biol Chem 2587155-7160 Gentry LE, Lioubin MN, Purchio AF, Marquardt H 1988 Molecular events in the processing of recombinant type 1 pre-pro-transforming growth factor beta to the mature polypeptide. Mol Cell Biol 84162-4168 Lyons RM, Gentry LE, Purchio AF, Moses HL 1990 Mechanism of activation of latent recombinant transforming growth factor 01 by plasmin. J Cell Biol 110:1361-1367 - Lvons RM. Keski-Oia J. Moses HL 1988 Proteolvtic activation of latent transforming” growth factor-p from fibroblast-conditioned medium. J Cell Biol 106:1659-1665 Sato Y, Rifkin DB 1989 Inhibition of endothelial cell movement by pericytes and smooth muscle cells: activation of a latent TGFpllike molecule by plasmin during coculture. J Cell Biol 109:309-315 Keski-Oja J, Lyons RM, Moses HL 1987 Immunodetection and modulation of cellular growth with antibodies against native transforming growth factor+l. Cancer Res 47:6451-6458 Urban JL, Kripke ML, Schreiber H 1986 Stepwise immunologic selection of antieenic variants during tumor growth. J Immunol 137:3036-3041 Stauss HJ, Linsk R, Fischer A, Watts S, Banasiak D, Haberman A. Clark I. Forman J. McMillan M, Schreiber H, Goodenow RS 1986 Isolation of the’ MHC genes encoding the tumour-specific class I antigens expressed on a murine fibrosarcoma. J Immunogenet 13:101-111 Torre-Amione G, Beauchamp RD, Koeppen H, Park BH, Schreiber H, Moses HL, Rowley DA 1990 A highly immunogenic tumor transfected with a murine transforming growth factor type 01 cDNA escapes immune surveillance. Proc Nat1 Acad Sci USA 87:1486-1490 Moses HL, Proper JA, Volkenant ME, Wells DJ, Getz MJ 1978 Mechanism of growth arrest of chemically transformed cells in culture. Cancer Res 382807-2812 Matrisian LM, Bowden GT, Krieg P, Furstenberger G, Briand JP, Leroy P, Breathnach R 1986 The mRNA coding for the secreted protease transin is expressed more abundantly in malignant than in henien tumors. Proc Nat1 Acad Sci USA 83:9413-9417 DorschlHasler K, Keil GM, Weber F, Jasin M, Schaffner W, Koszinowski UH 1985 A long and complex enhancer activates transcription of the gene coding for the highly abundant immediate early mRNA in murine cytomegalovirus. Proc Nat1 Acad Sci USA 82:8325-8329 Cepko CL, Roberts BE, Mulligan RC 1984 Construction and applications of a highly transmissible murine retrovirus shuttle vector. Cell 37:105311062 Graham FL. van der Eb AJ 1973 A new techniaue for the assay of infectivity of human adenovirus 5 DNA. Virology 52:456-467 ” Wigler M, Pellicer A, Silverstein S, Axe1 R 1978 Biochemical transfer of single-copy eucaryotic genes using total cellular DNA as donor. Cell 14725-731 Mellon P, Parker V, Gluzman Y, Maniatis T 1981 Identification of DNA sequences required for transcription of the human (Alglobin gene in a new SV40 host-vector system. Cell 27:279-288 Schwab M, Alitalo K, Varmus HE, Bishop JM 1983 A cellular oncogene (c-Ki-ras) is amplified, overexpressed, and located with karyotypic abnormalities in mouse adrenocortical tumour cells. Nature 303:497-501 Thomas PS 1980 Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc Nat1 Acad Sci USA 77:5201-5205

OF FIBROBLASTS

2485

31. Miller DA, Lee A, Pelton RW, Chen EY, Moses HL, Derynck R 1989 Murine transforming growth factor-82 cDNA sequence and expression in adult tissues and embryos. Mol Endocrinol 3:11081114 32. Miller DA, Lee A, Matsui Y, Chen EY, Moses HL, Derynck R 1989 Complementary DNA cloning of the murine transforming growth factor-@3 (TGFfl3) precursor and the comparative expression of TGF/33 and TGFPl messenger RNA in murine embryos and adult tissues. Mol Endocrinol 3:1926-1934 33. Josephs SF, Ratner L, Clarke MF, Westin EH, Reitz MS, WongStaal F 1984 Transforming potential of human c-sis nucleotide sequences encoding platelet-derived growth factor. Science 225:636-639 34. Land H, Parada JT, Weinberg RA 1983 Tumorigenic conversion of primary embryo fihroblasts requires at least two cooperation oncogenes. Nature 304:596-602 35. van Zonneveld A-J, Curriden SA, Loskutoff DJ 1988 Type 1 plasminogen activator inhibitor gene: functional analysis and glucocorticoid regulation of its promoter. Proc Nat1 Acad Sci USA 85:5525-5529 36. Danielson PE, Forss-Petter S, Brow MA, Calavetta L, Douglass J, Milner RJ, Sutcliffe JG 1988 plB15: a cDNA clone of the rat mRNA encoding cyclophilin. DNA 7:261-267 37. Taylor JM, Illmensee R, Summers J 1976 Efficient transcription of RNA into DNA by avian sarcoma virus _polymerase. Biochim _ Biophys Acta 442:324-330 38. Melton DA. Kriee PA. Rebasliati MR. Maniatis T. Zinn K. Green MR 1984 Efficieit in vitro iynthesis’ of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promoter. Nucleic Acids Res 12:7035-7056 39. Tucker RF, Branum EL, Shipley GD, Ryan RJ, Moses HL 1984 Specific binding to cultured cells of ‘251-labeled type fl transforming growth factor from human platelets. Proc Nat1 Acad Sci USA 81:6757-6761 40. Coffey RJ, Goustin AS, Soderquist AM, Shipley GD, Wolfshohl J, Carpenter G, Moses HL 1987 Transforming growth factor LYand /3 expression in human colon cancer lines: implications for an autocrine model. Cancer Res 47:4590-4594 41. Childs CB, Proper JA, Tucker RF, Moses HL 1982 Serum contains a platelet-derived transforming -- growth factor. Proc Nat1 Acad Sci USA 79:5312-5316 42. Coffey Jr RJ, Sipes NJ, Bascom CC, Graves DR, Pennington CY, Weissman BE, Moses HL 1988 Growth modulation of mouse keratinocytes by transforming growth factors. Cancer Res 48:15961602 43. Howe PH, Cunningham MR, Leof EB 1990 Inhibition of mink lung epithelial cell proliferation by transforming growth factor-j3 is coupled through a pertussis-toxin-sensitive substrate. Biochem J 266537-543 44. Coffey Jr RJ, Leof EB, Shipley GD, Moses HL 1987 Suramin inhibition of growth factor receptor binding and mitogenicity in AKR-PB cells. J Cell Phvsiol 132:143-148 45. Laiho M, Saksela 0, Andreasen PA, Keski-Oja J 1986 Enhanced production and extracellular deposition of the endothelial-type plasminogen activator inhibitor in cultured human lung tihroblasts by transforming growth factor-@. J Cell Biol 103:2403-2410 46. Torre-Amione G, Koeppen H, Beauchamp RD, Keenan M, Rowley DA 1990 TGF@ and escape from immune surveillance: production and activation of TGFP by tumor cells. In: Dinarello CA, Kluger MJ, Powanda MC, Oppenheim JJ (eds) The Physiological and Pathological Effects of Cytokines. Wiley-Liss, New York, pp 357361 47. Bascom CC, Wolfshohl JR, Coffey Jr RJ, Madisen L, Webb NR, Purchio AR, Derynck R, Moses HL 1989 Complex regulation of transforming growth factor fll, 82, and 83 mRNA expression in mouse fibroblasts and keratinocytes by transforming growth factors Bl and 82. Mol Cell Biol9:5508-5515 48. Wakefield LM, Smith DM, Broz S, Jackson M, Levinson AD, Sporn MB 1989 Recombinant TGF-pl is synthesized as a twocomponent latent complex that shares some structural features with the native platelet TGF-pl complex. Growth Factors 1:203218

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.

2486

TGF@l

TRANSFECTION

49. Sha X, Brunner AM, Purchio AF, Gentry LE 1989 Transforming growth factor /31: importance of glycosylaytion and acidic proteases for processing and secretion. Mol Endocrinol3:1090-1098 50. Gentrv LE. Webb NR. Lim GJ. Brunner AM. Ranchalis JE. Twardzik DR, Lioubin MN, Marquardt H, Purchio AF 1987 Type 1 transforming growth factor beta: amplified expression and secretion of mature and precursor polypeptides in Chinese hamster ovary cells. Mol Cell Biol 7:3418-3427 51. Kanzaki T, Olofsson A, Moren A, Wernstedt C, Hellman U, Miyazono K, Claesson-Welsh L, Heldin C-H 1990 TGF-/31 binding protein: a component of the large latent complex of TGF-Pl with multiple repeat sequences. Cell 61:1051-1061 52. Miyazono K, Olofsson A, Colosetti P, Heldin C-H 1991 A role of the latent TGF-@l-binding protein in the assembly and secretion of TGF-81. EMBO J 10:1091-1101 53. Sha X, Yang L, Gentry LE 1991 Identification and analysis of discrete functional domains in the pro region of pre-pro-transforming growth factor beta 1. J Cell Biol 114827-839 LE, Nash BW 1990 The pro domain of pre-pro-transform54. Gentry ing groth factor 01 when independently expressed is a functional binding protein for the mature growth factor. Biochemistry 296851-6857 55. McGeady ML, Arthur PM, Seidman M 1990 Development of a retroviral vector for inducible expression of transforming growth

OF

FIBROBLASTS

Endo. 1992 Vol 130. No 5

factor 81. J Virol64:3527-3531 56. Ignotz RA, Endo R, Massague J 1987 Regulation of fibronectin and type I collagen mRNA levels by transforming factor-B. -- growth J BiolChem 262:6443-6446 57. Raghow R, Postlethwaite AE, Keski-Oja J, Moses HL, Kang AH 1987 Transforming growth factor-p increases steady state levels of type I procollagen and fibronectin messenger RNAs-post-transcriptionally in cultured human dermal fibroblasts. J Clin Invest 79:1285-1288 58. Ruoslahti E, Yamaguchi Y 1991 Proteoglycans as modulators of growth factor activities. Cell 64:867-869 59. Ignotz RA, Massague J 1987 Cell adhesion protein receptors as targets for transforming growth factor-8 action. Cell 51:189-197 60. Heino J, Massague J 1989 Transforming growth factor-/3 switches the pattern of integrins expressed in MG-63 human osteosarcoma cells and causes a selective loss of cell adhesion to laminin. J Biol Chem 264:21806-21811 61. Cochran BH, Reffel AC, Stiles CD 1983 Molecular cloning of gene sequences regulated by platelet-derived growth factor. Cell 33:939-

- -.

947

62. Greenberg ME, Ziff EB 1984 Stimulation of 3T3 cells induces transcription of the c-fos proto-oncogene. Nature 311:433-438. R, Bravo R, Burckhardt J, Curran T 1984 Induction of c63. Muller fos gene and protein by growth factors precedes activation of cmyc. Nature 312:716-720

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 21:56 For personal use only. No other uses without permission. . All rights reserved.