EXPERIMENTAL

CELL

RESEARCH

192,

l-9

(1991)

Transforming Growth Factor-(3 Receptor Profiles of Human and Murine Embryonic Palate Mesenchymal Cells KERSTIK.LINASK,MARINAD'ANGELO,AMY Department

of Anatomy,

Daniel

Baugh

Institute,

Jefferson

Medical

L. GEHRIS,ANDROBERTM.GREENE~ College,

Jefferson

University,

Philadelphia,

Pennsylvania

19107

dent growth of fibroblastic cells [ 11. What has emerged, however, is the concept of TGF-P as a multifunctional regulatory peptide, controlling both growth and differentiation in an ever increasing diversity of cell and tissue types (reviewed in [2,3]). Transforming growth factor-/3 (TGF-P) is a homodimeric peptide with a molecular mass of 25,000 Da that is synthesized as a larger inactive precursor. The precise method of physiological activation of the propeptide into a shorter, active peptide remains presently unknown. Almost total sequence homology exists among human, bovine, and porcine monomers with a single amino acid substitution existing in the mouse. Although expression of five different TGF-P mRNAs has been demonstrated during embryogenesis and tissue differentiation [4-81, TGF-Pl appears to be the most abundant form. The TGF-/3s bind at relatively high, but varying affinities, to three structurally distinct glycoproteins [9]. As detected by affinity-labeling of receptor complexes and polyacrylamide gel electrophoresis, these receptors have relative molecular weights of 65 kDa (type I), 8595 kDa (type II), and 250-350 kDa (type III). Receptor types I and II have higher affinity for TGF-@l than TGF-P2. The type III receptor has equal affinity for both TGF-01 and TGF-P2 [lo, 111. In most avian and mammalian cell types that have been analyzed for the presence of TGF-P receptors, all three types of receptors are usually present [9]. The biological functions of the family of TGFs-P are related in their ability to influence basic control processes of tissue development, particularly of mesodermally derived structures [ 121. TGFs-/3 have been shown to alter cell proliferation, acting as a growth inhibitor in most cell types [13]. This class of growth factors also influences angiogenesis [14], chondrogenesis [15], myogenesis [16], and various epithelial-mesenchymal interactions [17, 181. These growth factors also affect the expression of extracellular matrix molecules, such as fibronectin and collagen types I and VI [19]. Thus, TGF-P appears to be an important regulatory component of many embryological processes. Growth, differentiation, and cell signalling in the developing murine palate appear to be contingent upon the presence of various growth factors and hormones

Cell signalling in the developing mammalian palate appears to involve various growth factors and hormones. An important developmental role for the transforming growth factor-/3 (TGF-P) class of growth factors is suggested by the immunolocalization of TGF-81 in the palate during its ontogeny. This study examined the effects of TGF-/3 stimulation of, as well as TGF-/3 receptor profiles in, murine embryonic palate mesenchymal (MEPM) and human embryonic palate mesenchymal (HEPM) cells. Results showed that TGF-/31 (1 rig/ml) stimulated proliferation of HEPM cells and inhibited proliferation of MEPM cells in a dose-dependent manner. The time course of ‘2SI-TGF-/31 binding to specific receptors was determined by incubating cells in the presence of 170 pM ‘2SI-TGF-/31 for up to 4 h. In both cell types, at 37”C, the binding of ‘“‘I-TGF-/3 decreased linearly over 4 h, while at 4”C, binding increased with time of incubation. Incubation of both cell types at 4°C for 4 h, with increasing concentrations of ‘261-TGF-fil, resulted in binding which demonstrated saturation kinetics. Scatchard analyses revealed one class of receptors for HEPM (K 32.3 PM) and MEPM (K 26.3 PM). However, SDS-PAGE analyses of “‘1-TGFfi chemically crosslinked to specific receptor sites revealed that both cell types contained the types I (65,000 M,) and III (230,000 M,) TGF-/3 receptors while MEPM also contained the type II (86,000 M,) receptor. Binding studies further demonstrated the ability of plateletderived growth factor to transmodulate TGF-/3 binding. These results indicate that the HEPM cell line and primary cultures of MEPM cells, although obtained from palates at similar developmental stages, are dramatically different in their responsiveness to TGF-/l and have disparate TGF-fi receptor profiles. CC1991 Academic Press. Inc.

INTRODUCTION Transforming growth factors-p were originally defined by their ability to stimulate anchorage-indepen’ To whom reprint requests should Anatomy, Thomas Jefferson University, phia, PA 19107.

Thomas

be addressed at Department of 1020 Locust Street, Philadel-

1

0014.4827/91$3.00

Copyright All

tights

8

1991 of reproduction

by Academic

Press, Inc.

in any

reserved.

form

2

LINASK

[ZO, 211. An important developmental role for the TGF/!I class of growth factors is suggested by the immunolocalization of TGF-61 in the palate during ontogenesis [17], as well as by the extensive body of literature indieating an important role of TGF-P in the stimulation of extracellular matrix components [ 191 which play a critical role in normal palatal ontogeny. The study presented here was undertaken to compare and contrast murine embryonic palate mesenchymal (MEPM) cells and a cell line of human embryonic palate mesenchymal (HEPM) cells in their responsiveness to TGF-01 as determined by cell proliferation and receptor profiles and characteristics. The human embryonic palate mesenchyme cell line (CRL 1486) was established by the American Type Culture Collection (Rockville, MD) and is presumed to represent the undifferentiated fibroblast-like cells from embryonic palatal shelves that were obtained from a single human abortus at the time of palatal shelf elevation, but prior to epithelial contact [22]. This is analogous to the developmental stage from which MEPM cells were obtained in setting up primary cultures for this study. Results of the present study indicate that HEPM and MEPM cells, which ostensibly represent a similar population of cells, differ significantly in their TGF-0 receptor profiles as well as their response to TGF-p stimulation. METHODS Cells and culture methods. Pregnant female mice were killed by cervical dislocation on Day 13 of gestation. Embryos were dissected free from uteri and extraembryonic membranes in calcium-magnesium-free phosphate buffered saline (CMF-PBS), pH 7.2. Palatal shelves were dissected from the embryos, minced, and dissociated with 0.025% trypsin, 0.01% EDTA (GIBCO 1:250) in CMF-PBS for 10 min at 37°C with constant agitation. The action of trypsin was inhibited by adding cold (4°C) Opti-minimum essential medium (Opti-MEM; GIBCO Laboratories, Life Technologies, Inc., Grand Island, NY) containing 5% fetal bovine serum (FBS). Cells were dispersed, washed, and concentrated by centrifugation at 50g for 15 min. The resultant cell pellet was resuspended in a known amount of OptiMEM containing 5% FBS. Cell number was determined with the aid of a hemocytometer and primary cultures of MEPM cells were initiated by seeding cells into 35-mm tissue culture dishes at a density of 2.5 X lo4 cells/cm* (Culture Day 0). HEPM cells, an established cell line (CRL 1486), were purchased at passage 7 from the American Type Culture Collection (Rockville, MD) and stored in liquid nitrogen. Cells were plated into 35.mm tissue culture dishes (Falcon) at an initial density of 5 X lo3 cells/cm’ (Culture Day 0). Both cell types were grown in Opti-MEM containing Earles salts and 25 mM Hepes buffer, supplemented with 2 mlM glutamine, fetal bovine serum (5%), streptomycin (150 @g/ml), and penicillin (150 units/ml). Cultures were maintained at 37°C in an atmosphere of 5% CO,/95% air. Binding assays. Binding assays were carried out on subconfluent (6 days of incubation) and confluent cells (11 days of incubation) as previously described [20]. Briefly, cell monolayers were washed with binding buffer (128 mM NaCI, 5 mM KCI, 1.2 mM CaCI,, 1.2 mM MgSO,, 50 mM Pipes (pH 7.5)), containing 5 mg/ml BSA. Intact cells were incubated for 1 h at 37°C in the presence of 1.0 ml of binding

ET

AI,.

hu ffer

and subsequently equilibrated for an additional 20 min at either 4 or 37°C. After equilibration, binding buffer was aspirated and the appropriate concentrations of ‘2”I-laheled (85.7 MCi/Fg) purified porcine platelet TGF-01 (R&D Systems, Inc., Minneapolis, MN) in 1 ml of binding buffer were added for specified time periods. Nonspecific binding was determined in the presence of a 60.fold excess of crude porcine TGF-fl (R&D Systems, Inc.) for each concentration. Incubations were terminated by removing binding buffer and washing monolayers four times with ice-cold binding buffer. Monolayers were then solubilized for 15 min with 0.5 ml of 1% Triton X-100 in isotonic saline, pH 7.4, and scraped from the dish. Dishes were subsequently rinsed with 0.5 ml of isotonic saline which was added to the solubilized cell layer fraction. Bound radioactivity was determined in a TM analytic No. 1190 gamma counter. Nonspecific hinding, measured in the presence of excess cold TGF+, was subtracted from total binding to obtain specific binding. Keccptor characterization by afinity labeling. Cells to be affinitylabeled with ““I-TGF-61 were removed from tissue culture dishes by preincubation for approximately 15 min at 37°C in the presence of affinity buffer containing 128 mM NaCl, 5 mM KCl, 25 mM Pipes, pH 7.5, 5 mM glucose, 6 mM EDTA. At the end of the preincubation period cells still attached were scraped from dishes and suspended cells collected by centrifugation at 1OOOg for 5 min. Pelleted cells were rinsed twice with affinity buffer and incubated with 0.5 ml of affinity buffer containing 150 pM ‘251-TGF-fi1 (118 bCi/pg). Cells were incubated for 2 h at 4°C with shaking. Cells were then centrifuged 1 min at 13,OOOg at 4°C. Radioactive buffer was removed and cells were rinsed twice with ice-cold afhnity buffer. Subsequently, cells were resuspended with 0.5 ml affinity buffer containing 5 ~1 of DMSO and 0.25 mM disuccinimidyl suherate (Pierce Chem., Co., Rockford, IL). The crosslinking reaction (15 min at 4°C) was stopped by dilution with 0.25 M sucrose, 10 mM Tris, pH 7.4, and 1.0 m&f EDTA, followed by centrifugation and a rinse with dilution buffer. The cell pellet was solubilized in a solution of 1% (v/v) Triton X-100, 10 mM Tris, pH 7.0,l mM EDTA containing 0.1 mM PMSF, pepstatin, and leupeptin. A 40.min incubation was followed by centrifugation at 13,OOOg at 4°C for 15 min. Detergent soluble and insoluble fractions were analyzed in the presence or absence of excess cold TGF-fl by SDS-PAGE. Receptor characterization was carried out on both confluent and subconfluent cells. Electrophoresis and autoradiography. Gel electrophoresis was carried out as described by Hames [24] utilizing both 5 and 10% polyacrylamide gels. Gels were subsequently stained with Coomassie blue and subjected to autoradiography at ~70°C.

RESULTS Effects of TGF-/31 on Cell Proliferation The cell morphology of HEPM and MEPM cells is strikingly different. HEPM cells tend to be larger than MEPM cells. This is reflected in HEPM cells being seeded at 40,000 cells per 35-mm dish in comparison to 225,000 cells for MEPM cells in order to have similar growth curves. Upon reaching confluency, HEPM cells exhibit an elongated form and a parallel orientation relative to each other (Fig. 1A). MEPM cells are irregularly shaped, smaller, and rounder without a noticeable organization (Fig. 1B). TGF-@l showed opposite effects on cell proliferation for the HEPM cell line and MEPM primary cultures. MEPM cells grown in the presence of 5% fetal bovine serum (FBS) had a generation time of 28.8 h. Addition

TGF-/I’

FIG. 1. Cell morphology of HEPM and MEPM (10th passage, 11th day in culture). (R) Confluent (primary culture, 11th day in culture).

RECEPTOR

PROFILES

OF

cells in uitro. (A) Confluent MEPM cells are irregularly

of exogenous TGF-01 (0.01-1.0 rig/ml) resulted in a dose-dependent increase in generation time (Table IA). Hence, TGF-@l had an inhibitory effect on MEPM cell growth. In contrast, HEPM control cells exhibited a generation time of 36.5 h and with increasing concentrations of exogenously added TGF$l, a decrease in generation time was noted (Table 1B). Therefore, TGF-/31 stimulated proliferation of HEPM cells. Incubation of HEPM cells, with TGF-@l neutralizing antibody (R&D Systems, Inc.) that blocks the binding of the growth factor to its receptor, resulted in growth inhibition relative

PALATE

CELLS

3

HEPM cells show an elongated form and parallel orientation shaped, small, and round with no noticeable organization

to control cells incubated in 5% serum-supplemented medium (data not shown). This most likely reflected the antibody blocking any TGF-P activity present in the serum. Receptor Profiles Radioiodinated TGF-fll was crosslinked to its receptors in both MEPM and HEPM cells and gel electrophoresis was carried out using both 5 and 10% polyacrylamide gels. The 5% gel gave better resolution of the

4

LINASK

TABLE

A. TGF-/3

B. TGF-/3 5% 1.0 0.1 0.01

inhibition

FBS control rig/ml TGF-fi rig/ml rig/ml

FBS control rig/ml TGF-fl rig/ml rig/ml

of MEPM

% Change

cell growth

28.8 52.8* 36.0* 32.4’ stimulation

of HEPM 36.5 26.0* 29.8* 21.5*

AL.

Time Course and Temperature

Generation time in hours*

Treatment

5% 1.0 0.1 0.01

1

ET

0 +s3 +25 +13 cell growth“ 0 -41.0 -18.4 -29.8

a Both MEPM and HEPM were cultured as described under Methods and incubated in the presence of increasing concentrations of TGF-01. Note that MEPM cells exhibit a dose-related inhibition of cell proliferation while HEPM cells exhibit a dose-related stimulation of cell proliferation. * Generation time, defined as the time required for the number of cells in the population to exactly double during exponential growth, was determined by direct cell counts and calculated using the formula 2.3 log,, (N,lN,) = k(T, - TJ. Solving for this formula, where N represents the number of cells at time T, yields a growth constant, k, for a specific population of cells which represents the percentage increase in cell number. After a time interval equal to the generation time has elapsed, the ratio N,IN, is equal to 2; thus 2.3 log 2 = kT or 0.693 = kT. Solving for this formula then yields a generation time, T, in hours. * P < 0.05. The slope of the growth curve during exponential growth was compared to control by one-way analysis of variance (ANOVA).

higher molecular weight receptor (230K-265K). However, under these conditions the 65K band ran at the buffer front (Fig. 2A) where band spreading of various components caused bands to become indistinct and overlap. For this reason a 10% gel (Fig. 2B) was run to better resolve this class of receptor. Electrophoretic profiles of subconfluent and confluent cultures of both cell types were similar. HEPM and MEPM cells, however, presented different receptor profiles. Under nonreduced conditions in 5 and 10% gels (Figs. 2A and 2B) the type III receptor predominated in HEPM cells showing a diffuse band at approximately M, 265,000 and a barely detectable class of receptors at M, 65,000. MEPM cells contained type III (M, 230,000), type II (M, 86,000), and type I (n/i, 65,000) receptors. The type III receptor on these two cell populations showed heterogeneity in that the relative mobilities were slightly different, possibly due to the degree of glycosylation [25]. Type II receptors were not apparent for the HEPM cell type. In Fig. 2B the heavily labeled band at 25,000 M, most likely represents the dimeric form of radioiodinated TGF-P run under nonreduced condition. This band migrated at 12,500 M, (monomeric form) under reduced conditions, seen in lanes 3 and 4 (Fig. 2B).

Effects

Figures 3A and 3B illustrate the kinetics of ‘“‘I-TGF/Yl specific binding to cell surface receptors at 4 and at 37°C. At 4°C MEPM cells showed half-maximal binding of radioiodinated TGF-fil at 0.5 h of incubation and maximal binding at 1 h. Binding remained relatively level up to 4 h of incubation (Fig. 3B). At 37°C MEPM cells showed maximal binding by 15 min of incubation with a progressive decrease in binding to nonspecific background levels after 4 h of incubation. Similar results were obtained with HEPM cells in that saturation of receptors was obtained during a 4-h incubation period at 4°C with maximal binding reached by 3 h (Fig. 3A). Half-maximal binding, at 1 h, took twice as long to reach for HEPM cells relative to MEPM. At 37°C a progressive decrease in binding was observed for HEPM cells over a 4-h period. The greatest decrease in binding occurred between 2 and 3 h of incubation. Results of each experiment were based on the average of triplicate determinations. Experiments on HEPM and MEPM cells were repeated twice with similar results. Afinity

and Numbers

of TGF-fi Receptors

Binding isotherms obtained with various concentrations of iz5I-TGF-01 indicated that saturation of TGF-0 receptors at 4”C, after 4 h of incubation, is achieved at 125 and 150 pM in HEPM (Fig. 4A) and MEPM (Fig. 4B) cells, respectively. Scatchard analysis of lz51-TGF-/3 binding data resulted in a rectilinear plot suggesting the presence of a single class of high-affinity binding sites in both cell types. The Kilodalton values deduced from the slope of Scatchard plots in Figs. 4A and 4B were 26 pM for HEPM cells and 32 pM for MEPM cells. The abscissa intercept indicated the presence of approximately 300,000 receptors/cell for HEPM and 350,000 receptors/cell for MEPM cells. Dissociation Kinetics Incubation with ‘251-TGF-/Y for 0.5 h at 37°C resulted in substantial binding (Fig. 3) to both HEPM and MEPM cells. Subsequent dissociation (Table 2) after 150 min at 37°C resulted in 82% dissociation of lz51TGF-P from MEPM and 100% dissociation from HEPM cells. After 150 min of incubation at 4”C, comparatively little labeled TGF-P was dissociated from either cell type (28% from HEPM and 34% from MEPM). After incubation for 3.5 h at 37”C, little TGF-6 remained bound to MEPM cells (Fig. 3). Of what remained, 65% dissociated from MEPM cells and only nonspecific background levels were detectable on HEPM after 150 min. After binding for 3.5 h at 4°C conditions under which high levels of receptor-ligand binding activity are seen (Fig. 3), followed by dissociation for 150 min under similar conditions, 65% of the

5

TGF-B RECEPTOR PROFILES OF PALATE CELLS

-MW B

L K -MW 224.3

.zOOK

,106K

K

71.8

K

28.5

K

15.3

K

’ 71K

LANES LANES

I

2

3

4

FIG. 2. Autoradiograms of 5 and 10% polyacrylamide gel electrophoresis of iz51-TGF-fi1 crosslinked to its receptors. The position and molecular weights of protein standards in a parallel lane are indicated on the right side of gels. (A) Autoradiogram of 5% gel run under nonreduced conditions. HEPM (lanes l-3): Lanes 1 and 2 reflect 60 and 30 (~1 of protein loaded per well. respectively. Two bands marked with arrowheads at M, 265,000 and M, 65,000 are detectable, signifying the presence of type III and type I TGF-fi receptor classes. Lane 3 (+) shows ligand-receptor binding in the presence of a 60.fold excess of unlabeled crude TGF-0 added in binding assay. No bands or decreased binding is evident indicating specificity of TGF-p binding. MEPM (lanes 4-6): Lanes 4 and 5 demonstrate 60 and 30 ~1 of protein loaded per well, respectively. Three distinct bands (see arrowheads) are noticeable: Type III (M, 230,000), type II (M, 86,000), and type I (M, 65,000). Lane 6 (+) shows ligand-receptor binding in the presence of a 60.fold excess of unlabeled crude TGF-fl added in binding assay. No bands or decreased binding is evident signifying specificity of binding. (B) Autoradiogram of 10% gel. Arrows point to reported molecular weights of TGF-P receptor classes I (M, 65,000), II (M, S6,000), and III (M, approximately 260,000). HEPM (lane 1): Under nonreduced conditions, one distinct band at M, 260,000 (type III receptor class) and a barely detectable band at M, 65,000 (type I) are apparent. The heavily labeled band at M, 25,000 most likely represents the unbound radioiodinated dimeric TGF-01 and M, 12,500, the monomeric form. MEPM (lane 2): Under nonreduced conditions, three bands are observeable migrating with molecular weights corresponding to the three receptor classes of TGF-/31, type III (M, 260,000), type II (M, 86,000), and type I (M, 65,000). The heavily labeled band at M, 25,000 corresponds to the molecular weight of dimeric TGF-@l that remained unbound and the band at M, 12,500 represents the monomeric form of the growth factor. HEPM and MEPM, lanes 3 and 4, respectively, run under reduced (r) conditions with the addition of P-mercaptoethanol. The receptor profiles are the same as seen in lanes 1 and 2.

labeled TGF-P dissociated from MEPM cells (52% from HEPM cells). Binding was therefore reversible for both cell types and showed similar patterns of dissociation. Specificity Specific binding of radioiodinated TGF-Pl occurred in both cell populations in the presence of unlabeled TGF-P, epidermal growth factor (EGF), basic fibroblast growth factor (FGF), insulin-like growth factor II (IGFII), transforming growth factor-a (TGF-a), and platelet-derived growth factor (PDGF). Results are presented in Table 3. Binding appeared to be specific for TGF-0. Only PDGF was able to compete with lz51-TGFpl for binding sites in both cell populations. Percentage specific binding is expressed as percentage of binding relative to binding to control cells, i.e., cells incubated without the addition of excess unlabeled TGF-@. DISCUSSION

A major biological effect of TGF-P is growth inhibition [26]. This has been demonstrated in T-lympho-

cytes [27], endothelial cells [28], and megakaryocytic and erythroid precursors [29]. In our study TGF-Pl also had an inhibitory effect on proliferation of primary cultures of MEPM cells. Growth stimulatory responses to TGF-/3 have also been demonstrated in fibroblast cells or fibroblast-like mesenchymal cells [26]. Mitogenic stimulation appears to be a delayed response, suggesting that TGF-fl acts indirectly. The data suggest that the mitogenicity may be mediated by TGF-P’s induction of c-sis and PDGF and stimulation of PDGF-inducible genes [30, 311. Whether the mitogenic effect of TGF-P on the HEPM, as reported here, acts in a similar manner remains to be determined. Transforming growth factor-p affects extracellular matrices either by increasing synthesis of matrix molecules such as glycosaminoglycans [32, 331, fibronectin, and collagen [16, 341 or by stimulating the induction of protease inhibitors, e.g., plasminogen activator inhibitor [35-371. Thus, TGF-/3 treatment may result in accumulation of extracellular matrix components both via increased synthesis and/or decreased degradation. The extracellular matrix has been shown to influence cell

6

LINASK

7000

ET

AL.

9000

A

k

--.

/

--*

4%

6

6000

4000 -

: 5000

3000

, 0

i

,

, 1

,

, , 2 TIME, hrs

, 3

,

, 4

4000

5 TIME, hrs

FIG. 3. Kinetics of “%TGF-01 specific binding to cell surface receptors at 4°C and at 37°C over a 4-h incubation period. (A) MEPM: At 37°C MEPM cells showed peak binding by 15 min of incubation with a progressive decrease in binding to nonspecific background levels after 4 h of incubation. At 4°C peak binding occurred at 3 h of incubation and remained relatively level through the 4-h incubation period. (B) HEPM: At 37°C a progressive decrease in binding was observed over a 4-h period. At 4°C peak binding was reached after 1 h and binding remained high at 4 h.

morphology [38]. TGF-/3’s stimulation of cell surface receptors, as integrin [39,40], that recognize extracellular matrix components and that provide a transmembrane link with the cytoskeleton, indicates that TGF-P has the ability to affect cell morphology and adhesion. Preliminary studies indicate that TGF-P stimulates fibronectin synthesis in both MEPM and HEPM cells and that there are distinct differences in the appearance of extracellular fibronectin fibrils (K. K. Linask, unpublished). Whether these differences influence MEPM or HEPM cell behavior is yet to be determined. Data presented in the current study demonstrate different profiles of the three known receptor classes for TGF-/31 on HEPM and MEPM cells. HEPM showed a predominance of the class III receptors with a barely detectable level of the lower affinity receptor, type I (M, 65,000). In contrast, all three classes of receptors were distinctly detectable on MEPM cells, with type II and type III classes of receptors predominating. These three receptors have been identified as being structurally distinct glycoproteins [ 10,231. Affinity constants appear to be in the picomolar range for a number of cell types [ 131, including, as shown in the present study, embryonic palate mesenchymal cells. The type III receptor has been shown to be the major TGF-/3 receptor class for many mammalian and avian cells, including fibroblasts and epithelial cells. It appears to be a membrane proteoglycan containing heparan and chondroitin sulfate glycosaminoglycans [lo]. The latter property may underlie the apparent heterogeneity of this receptor, resulting in its migration as a diffuse band during SDS-PAGE. An interesting aspect of the present study is that the HEPM cell line with a barely detectable level of type I

receptors showed a growth stimulatory response to TGF-Pl. Primary cultures of MEPM cells with detectable levels of all three receptor classes demonstrated growth inhibition at similar TGF-61 concentrations. It may be speculated that occupancy of receptor classes I and/or II at specific concentration levels may modulate the response elicited by TGF-/31’s binding to the higher affinity type III receptor. It had been shown previously that all three receptor types have similar affinity for TGF-Pl [lo]. Receptor types I and II, however, have lower affinity for TGF-P2, while the type III receptors display similar affinities for TGF-Bl and -p2 [lo]. Functionality of the different receptor types, as well as the different forms of endogenous TGF-fi associated with palatal mesenchymal cells, remains to be clarified. Binding of lz51-TGF-/31 to both HEPM and MEPM cell populations demonstrated time and temperature dependence, saturability, and reversibility. The time course of binding of 12”1-TGF-p to both types of palatal cells at 37 and 4°C showed maximal cell-associated radioactivity after 0.5-l h of incubation. At 37”C, maximal binding was followed by a decrease of cell-associated radioactivity down to background levels during a 4-h period. This decrease was not observed at 4°C. This decrease in cell-associated radioactivity at 37°C could be due to a loss of cell surface receptors by down-regulation of the receptors and bound ligand. Indeed, NRK fibroblasts have been shown to internalize the labeled ligand, as well as down-regulate its receptors, thus resulting in a decrease in the cell-associated ligand [41]. Incubation of both palate mesenchymal cell populations with increasing concentrations of ‘251-TGF-fll result,ed saturation ki_ .~~._ .~ in ~~. bindinp ~~~ .~~~~_which demonstrated

TGF-6

RECEPTOR

PROFILES

OF PALATE

7

CELLS

~.

~.

60

2&

~

do

TGF-I31 CONCENTRATION,

0

pM

0 I

\

OOA 0

2

4

6

8

5 B

0

100

cells

300

200

TGF-I31 CONCENTRATION,

fmol/lO

FIG. 4. Confluent monolayers of HEPM or MEPM cells in 35-mm culture dishes were incubated various amounts of unlabeled TGF-/!I to achieve the final concentrations indicated on graph. The amount concentration was determined. Each concentration was done in triplicate. Scatchard analysis of binding isotherm and Scatchard plot of data. (B) MEPM: Binding isotherm and Scatchard plot of data.

netics. Scatchard analysis of binding data suggested a model consistent with a single class of binding sites of high affinity with relatively equal concentrations of binding sites in both cell types. That Scatchard analysis did not indicate more than one class of receptors may reflect the predominance of the higher affinity type III receptor class. MEPM cells do, however, show pronounced amounts of types I and II receptors as observed from SDS-PAGE. Furthermore, Scatchard analysis (Fig. 4B) may be interpreted as two intersecting lines suggesting at least two distinct classes of binding sites.

400

pM

for 4 h at 4°C with iz51-TGF-8 plus of TGF-fl specifically bound for each is seen in insets. (A) HEPM: Binding

Unlabeled TGF-P (0.1 and 0.5 rig/ml) was able to compete with radioiodinated TGF-Pl for specific binding sites on both cell populations. In contrast, similar concentrations of EGF, basic FGF, IGF-II, and TGF-a did not compete with radioiodinated TGF-/31 for binding to cell surface receptors. PDGF (0.1 rig/ml) was, however, able to compete for binding sites with TGF$l on both HEPM and MEPM cells. A similar decrease of 12”1-TGF-p binding in the presence of PDGF has been reported in NRK fibroblasts [41]. Many growth factors show transmodulation or differential regulation of re-

8

IJNASK

ceptor-ligand binding. For example, TGF-P has been shown to regulate the expression of PDGF receptors [42]. EGF receptors have also been shown to be transmodulated by PDGF [43], FGF [44], and TGF-fl [45]. None of these, however, bind directly to the receptor being regulated. Such receptor transmodulation may serve as a partial explanation of growth factor synergisms. Whether PDGF is actually involved in modulating TGF-P receptors on palatal cells and influencing subsequent biological activity remains to be determined. Our present studies indicate that the HEPM cell line, as compared to primary cultures of MEPM cells, exhibits markedly different responsiveness to TGF-61 and displays disparate TGF-P receptor profiles. HEPM cells (CRL 1486) have been used for several years to study mechanisms of cleft palate induction by glucocorticoids [46], effects of EGF stimulation [22], and effects of teratogens [47]. The HEPM cell line has also generated interest among reproductive toxicologists for its possible use as an in vitro teratogenicity screening assay [48]. It has been shown that these cells have a normal female complement of human chromosomes that remained reasonably stable up to passage 14 and that no significant changes occur in cell growth characteristics and karyotype during this period [48]. However, data presented in this communication as well as observations that CAMP-mediated cellular events in MEPM and HEPM cells and CAMP-dependent protein kinase profiles are dramatically different (R. M. Greene, unpub-

TABLE Percentage

2

Dissociation after ““I-TGF$l MEPM and HEPM Cells

Binding

to

Dissocation” conditions (%I) 37°C

Binding conditions 3.5 h, 37°C 3.5 h, 4°C 0.5 h, 37°C

15

HEPM MEPM HEPM MEPM HEPM MEPM

min 65

58 6T 64

4°C

15Omin

15

min b

100 65

100 82

150min

34 34 24 13

ET

TABLE Specificity

Treatment TGF-p1 EGF FGF

applicable.

basic

IGF-II TGF-