Distinct receptors for epidermal growth in cultured gastric smooth muscle cells
factor-urogastrone
S.-G. YANG AND M. D. HOLLENBERG Endocrine Research Group, Department of Pharmacology and Therapeutics and Department of Medicine, Faculty of Medicine, University of Calgary, Calgary, Alberta T2N 4N1, Canada
YANG,S.-G., AND M.D. HOLLENBERG.D~S~~T-K~ receptorsfor epidermal growth factor-urogastrone in cultured gastric smooth muscle ceLLs. Am. J. Physiol. 260 (Gastrointest. Liver Physiol. 23): G827-G834, 1991.-We describe a propagable cell strain from guinea pig gastric circular muscle (GCM), which we have characterized in terms of its smooth muscle phenotype and its binding and biological response (thymidine incorporation) to epidermal growth factor-urogastrone (EGF-URO) and transforming growth factor-a (TGF-cu). The binding of ‘““I-labeled EGF-URO to the GCM cells exhibited high affinity and an appropriate peptide specificity. A curvilinear Scatchard plot of the binding data indicated two classes of high-affinity binding sites (dissociation constants of 0.69 and 4.3 nM) and a maximal binding capacity of 24,000 sites/cell. Binding competition data demonstrated that the binding affinity of TGF-cu was greater than that of EGF-URO by a factor of 2. These relative binding affinities agreed with the two- to threefold greater potency of TGF-cu, compared with EGF-URO, for the stimulation of GCM thymidine incorporation. The relative order of binding affinity and biological potency (TGF-cu > EGF-URO) was distinct from the relative order of binding affinities (EGF-URO > TGF-cu) observed using guinea pig liver and human placental membrane preparations. We conclude that the cultured smooth musclederived GCM cells possess a receptor subtype that is in accord with contractile bioassay data obtained previously with intact gastric circular muscle strips. This receptor (TGF-c~ > EGFURO) appears distinct from the one previously characterized in nonmuscle tissues. epidermal growth factor-urogastrone receptor subtype; smooth muscle cultures; transforming growth factor-a
GROWTH FACTOR-urogastrone (EGF-URO), a 6-kDa polypeptide with mitogenic and acid-inhibitory activities, was originally isolated from male mouse submaxillary glands or from human urine on the basis of its ability to stimulate precocious eyelid opening in newborn mice or to inhibit gastric acid secretion in dogs (7, 14). EGF-URO is structurally closely related to transforming growth factor-a (TGF-a), a homologous polypeptide that can also bind with high affinity to the EGF-URO receptor (8, 25, 26, 34). It is recognized that EGF-URO and TGF-cu can exert their action on a wide variety of target cells, but their exact physiological role is not yet clear (e.g., seeRefs. 5,17). Because of our original observations (27), supported by the work of others (l), indicating that EGF-URO is a potent regulator of smooth muscle contractility, we have recently focused our attention on the actions of EGF-URO in a variety of vascular and nonEPIDERMAL
019% l&57/91
$1.50 Copyright
vascular smooth muscle systems (10, 11, 22, 27-30). In particular, work with a gastric muscle strip preparation from the guinea pig revealed that gastric circular smooth muscle possessesa receptor system for EGF-URO that is distinct from the one in other previously studied tissues. In this circular muscle (CM) preparation, the potency of TGF-cu is greater than that of EGF-URO, whereas the reverse is true for the longitudinal smooth muscle present in the same tissue, as well as for a variety of other EGF-URO target tissues (21, 29). Two hypotheses can be put forward to explain the unusual order of biological potencies for TGF-a and EGF-URO in the CM tissue: 1) there may be a difference in the order of receptor binding affinities for TGF-cu and EGF-URO in the CM tissue, compared with the relative affinities in other tissues, pointing to a distinct receptor subtype in the CM tissue; or 2) differences in receptor signaling pathways in the CM tissue compared with other tissues might in some manner reverse the biological potencies of TGF-cu and EGF-URO as measured by bioassay, even though their relative receptor binding affinities might be the same as those observed in other tissues (affinity of EGF-URO > TGF-cw). To distinguish between these two hypotheses at the biochemical level, it is essential to measure directly the relative binding affinities of EGF-URO and TGF-cu in the gastric circular muscle tissue. Thus the goal of the work we describe here was to measure the receptor binding of EGF-URO and TGF-cu in a CM preparation. In preliminary work with membranes isolated from CM muscle strips, it became apparent that the abundance of receptor was far too low for a meaningful biochemical analysis of the EGF-URO receptor by methods we have used successfully for other tissues (e.g., see Ref. 16). We therefore turned our attention to the culture of CM-derived smooth muscle cells, in the hope that the conditions of culture might lead to an increased expression of receptors. In such cultures, the receptor for EGFURO could be studied by approaches that we have used previously with human fibroblasts (e.g., see Ref. 20). In the work that we report here, we describe the characterization of the receptor for EGF-URO by ligand binding and bioassay (thymidine incorporation) methods in a cultured cell strain derived from guinea pig gastric circular smooth muscle (GCM). The GCM cells exhibited a smooth muscle phenotype in terms of their “hill-andvalley” growth pattern and their immunocytochemical reactivity with a muscle actin-specific monoclonal anti-
(~1 1991 the American
Physiological
Societ,y
G827
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
G828
GASTRIC
SMOOTH
MUSCLE
body (HHF35). Our results support the first of the two hypotheses outlined above, suggesting that the GCM cells possess a distinct receptor subtype for EGF-URO. MATERIALS
AND
METHODS
Peptides and other reagents. Murine EGF-URO was isolated from fresh-frozen submaxillary glands of testosterone-treated male mice (32). The polypeptide yielded a single peak when analyzed by reverse-phase liquid chromatography (C,, column) and yielded a single band upon gel electrophoresis. In a routine mitogenesis assay using human fibroblasts EGF-URO has a concentration causing a half-maximal response (EC& of -0.25 rig/ml (20). TGF-cu, a gift from Dr. M. E. Winkler, Genentech, San Francisco, was expressed in Escherichia coli as described (9), purified by Sephadex G-75 gel filtration followed by high-performance liquid chromatography (36). On a weight basis, TGF-cu has a specific activity of 0.55 ng of EGF-URO receptor equivalents/rig of TGF-a (36, 37). The TGF- CYused in this study was a sample from the same lot (PD-1) as used in our previous study (37). The concentrations of the stock solution of EGFURO were measured spectrophotometrically as done previously (ZO), using the formula (Ezlr, - E2& X 155 = pg/ ml. The concentration of TGF-a in the stock solution was determined by amino acid analysis. We have compared the activities of EGF-URO with TGF-cu on a weight or molar basis rather than on the basis of equal receptor occupancy (21,37). Fetal calf serum (FCS) was obtained from Flow Laboratories (Mississauga, Ontario); lz51- and [ methyl-“H] thymidine were from Amersham (Arlington Heights, IL). Oxytocin, dexamethasone, and [Arg8]vasopressin were obtained from Sigma (St. Louis, MO). All other chemicals were of reagent grade or better. Isolation of gastric circular smooth muscle and preparation of primary cell cultures. The standard isolation solution was of the following composition (in mM): 1.3 CaC12, 5.4 KCl, 0.4 KH2P04, 0.4 MgCl, . 6H20, 0.4 MgSOdo 7Hz0, 137 NaCl, 4.2 NaHCO,?, 0.34 Na2HP04= 7Hg0, 5.6 glucose, and 25 N-Z-hydroxyethylpiperazine1V’-2-ethanesulfonic acid (HEPES). The pH of the solution was adjusted to 7.4 with NaOH, and the solution was equilibrated with 95%02-5%CO,, sterilized by filtration with a 0.22-,um filter, and supplemented with antibiotics. Male guinea pigs were killed by a blow on the head, and the stomach was removed. The stomach was cut open, the contents were removed, and the tissue was washed twice in cold isolation solution. The gastric body was then cut off from the gastric fundus and antrum. The tissue was then put into a dissecting bath filled with isolation solution maintained at 4°C and aerated continuously with 95%0,-5%COz. The mesentery was trimmed, and the mucosa was removed with fine scissors. The two layers of the gastric body smooth muscle were then separated with forceps under a dissecting microscope. The CM was removed and washed three times with 50 ml of sterilized isolation solution and was then minced into -l-mm’ pieces. Freshly minced CM segments were centrifuged at 25 g for 5 min at room temperature. The segments were resuspended in growth medium compris-
EGF-IJRO
RECEPTOR
ing Dulbecco’s modified Eagle’s medium (DMEM) supplemented with high glucose (24 mM), 5% (vol/vol) FCS, 4 mM L-glutamine but without antibiotics. Segments were plated in sterile 80-cm2 plastic flasks at a density of 2-3 pieces/cm” with 2 ml growth medium and were incubated at 37°C in a humidified atmosphere of 10% CO2 in room air. After the pieces had adhered (usually 2-3 days) cells were fed with 20 ml growth medium and were refed at weekly intervals until a substantial outgrowth of cells was observed (2-3 wk). When the outgrowth spread to cover at least 60% of the growth surface, the cells were subcultured with trypsin dissociation, using routine procedures for cell culture. The GCM cell strains obtained from individual animals were used for up to 10 generations of subculture at a time interval of 1 wk and at a passage ratio of 1:4 (each passage leads to 2 population doublings). For subcultures, the GCM cells were seeded at 2 x lo4 cell/ml in 80-cm2 (surface area) T flasks containing 20 ml DMEM and 5% (vol/vol) FCS. To avoid complications due to receptor variation that might be caused by different growth rates or different cell densities, only monolayers seeded at comparable cell densities and grown to confluence were used for our studies. Immunocytochemical staining of cultured GCM cells. HHF35, a muscle-specific monoclonal anti-actin antibody, was used to identify the cultured GCM cells by indirect immunofluorescent staining (13,23,35). HHF35 recognizes the o-actin of skeletal, cardiac, and smooth muscle and the y-actin of smooth muscle (23, 35). The cells for immunofluorescence staining were cultured on cover slips in multiwell plates. The cover slips were rinsed with Hanks’ balanced salt solution, treated with acetone at -20°C for 10 min and air dried. Indirect immunofluorescence was measured as described (13, 23). The cover slips of the cell cultures were incubated with HHF35 (l:l,OOO dilution, Enzo Diagnostics, New York, NY) at 21°C overnight, were washed with phosphatebuffered saline (PBS), and were then incubated with 0.1% bovine serum albumin (BSA)-Triton X-100 (100 pi/l) at 21°C for 5 min. The cover slips were washed again with PBS and incubated with fluorescence-conjugated goat anti-mouse immunoglobulin G (IgG; 1:64, Sigma, St. Louis, MO) at 21°C for 30 min. After the final wash, the cover slips were dehydrated, mounted on glass slides, and then photographed with a Zeiss III photomicroscope. Guinea pig liver plasma membrane preparations. Crude “microsomal” membranes were obtained from male guinea pig liver tissue by differential centrifugation according to the methods used routinely for isolating human placenta membranes in this laboratory (16). The isolation buffers and the cocktail of protease inhibitors were exactly as described previously (37). Binding assay. EGF-URO was labeled with 12’1 essentially as described previously (19), except that the 12’1EGF-URO was separated from the reaction mixture by chromatography on Sephadex G-10. This method of radiolabeling has been previously shown to yield ‘““I-EGFURO that is fully active in a thymidine incorporation bioassay (summarized in Ref. 19). The binding of “‘IEGF-URO to the GCM cells was performed on cell
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
GASTRIC
SMOOTH
MUSCLE
monolayers grown to confluence after subculture from T flasks into 3.5cm-diam Nunclon multidish trays (-120,000 cells/dish; GIBCO, Grand Island, NY). Binding was measured essentially as described previously (2, 18), except that the binding medium (DMEM containing 1 mg/ml BSA and 0.1 PM KI) was buffered at pH 7.4 with 20 mM HEPES. After equilibration, unbound radioactivity was washed from the cells with ice-cold PBS (pH 7.4) containing 1 mg/ml BSA and 0.1 PM KI. The washed cells were then solubilized in 1 ml of 1.0 M NaOH (60 min at room temperature), and the radioactivity was determined by crystal scintillation counting (efficiency -85%). Individual subcultures for the binding assays were seeded independently from stock T flasks and were grown in separate crops for separate binding assays. The binding of lz51-EGF-URO to the guinea pig liver plasma membranes and to human placental membranes was performed as described (16, 37), with a final reaction volume of 0.2 ml and 75 pg of membrane/assay tube. The specific binding was defined as the difference between the total amount of radioactivity bound minus the amount bound in the presence of a ?500-fold excess of unlabeled EGF-URO. In most experiments, wherein the concentration of radiolabeled EGF-URO was 70% of the labeled EGF-URO that had initially been bound to the cells at the start of the 37°C dissociation time course. Upon dissociation at 37°C intact ‘““I-EGF-URO released into the incubation medium (open circles, Fig. 3) matched exactly the reduction in cell surface-bound peptide (i.e., acid elutable; open triangles, Fig. 3). Only a small proportion of the radiolabel was either internalized by the cells (closed triangles, Fig. 3) or was resistant to precipitation by TCA (closed circles, Fig. 3). Thus we were able to conclude that even at 37°C the cellular uptake and degradation of ‘““I-EGFURO by GCM cells was very low, in contrast to previous studies that had been done with cultured non-smooth muscle cells (e.g., see Ref. 2). In view of the above findings, all subsequent binding assays were performed at 4°C with an equilibration time of 5 h. Binding-competition studies. Unlabeled EGF-URO and TGF-cu both competed efficiently for the binding of 12’1EGF-URO to the GCM cell monolayers, as illustrated in Fig. 4 and Table 1. In contrast, other disulfide-containing polypeptides like oxytocin and insulin did not compete
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
GASTRIC
CONCENTRATION
SMOOTH
MUSCLE
EGF-URO
G831
RECEPTOR
(nM)
EGF-URO and TGF-cu that maximally inhibited binding (50-100 rig/ml). In terms of binding inhibition, the con1.0 10 - centration of unlabeled EGF-URO that reduced lz51I I - EGF-URO binding by 50% (I&) was 5.7 t 0.6 rig/ml (means t SE for 3 independently conducted binding competition curves; 1.0 nM); in contrast, the IC& for - TGF-cu under identical conditions was 2.8 t 0.2 rig/ml (means t SE, n = 3; 0.5 nM). These two values of the IC&s for EGF-URO and TGF-cu obtained using three separately grown crops of cell monolayers were statistically different (P c 0.01). The binding inhibition assay yields reliable information about the relative affinities of unlabeled ligands for a receptor site (e.g., see Ref. 20) but does not necessarily yield information about the possible existence of multiple affinity sites (see equilibrium binding data below). Nonetheless, in the GCM cells, the binding competition experiments demonstrated that the order of binding affinities was TGF-cu > EGF-URO. j This order of binding affinities was retained over at least 1 I llll1ll I I1r11lll 1 II 111 six to seven GCM cell generations (passages 3-10) and 1.0 3.0 10 100 200 was observed for two cell strains independently isolated CONCENTRATION (rig/ml) from separate animals. FIG. 4. Competition-inhibition of ‘““I-EGF-URO binding to GCM In the guinea pig liver membrane binding assay, TGFcell monolayers by EGF-URO and TGF-cu. Multiple, independently cy and EGF-URO were also able to compete efficiently grown confluent cell monolayers were incubated for 5 h at 4°C with 6 for the binding of ‘““I-EGF-URO. However in this sysrig/ml of ““I-EGF-URO (190 cpm/pg) in the presence of increasing tern, under identical conditions of assay, the I&, for concentrations of unlabeled EGF-URO (0) and TGF-(U (A). Net amount of specifically bound radioactivity was then determined as outlined in EGF-URO was repeatedly smaller than the I& for TGFMATERIALS AND METHODS. Specific binding in the absence of competcy (Fig. 5); i.e., the order of binding affinities was EGFitor (1,930 & 60 cpm) was expressed as lOO%, and binding of “‘1-EGFURO at each concentration of unlabeled competitor was calculated relative to the 100% value. Each point represents average t SE of measurements made on 3 independently grown monolayers. Figure is representative of 5 independent experiments on separately grown crops of monolayers, using cell strains derived from 2 separate animals, between passages 3 and 10.
TABLE
CONCENTRATION
(nM)
1. Binding competition studies E Addit
ion
None Oxytocin Insulin Dexamethasone Vasopressin EGF-URO EGF-URO TGF-CY
Concentration, K/ml
1 10 1 1 0.010 0.050 0.010
‘“‘I-EGF-URO pg/lO”
80
Bound, cells
44.0t1.2 42.4tl.O 44.3t2.6 41.7k1.5 31.4t1.4 22.1kl.O 14.7k1.2 16.5k1.4
Values are means t SE of measurements on 4 individually grown monolayers and are corrected for nonspecific binding. Under these conditions, the nonspecific binding of ““I-EGF-URO, measured in the presence of a 500-fold excess of unlabeled EGF-URO (3 pg/ml) was -8% of the total amount of 1’SI-EGF-URO bound. Individual confluent cell monolayers (3.5 cm diam) grown concurrently were rinsed and incubated for 5 h at 4°C with 6 rig/ml I”‘I-EGF-URO (230 cpm/pg) either in the presence or absence of the various added agents at the indicated concentrations in a final volume of 1 ml binding medium. Monolayers were then washed free of unbound “‘I-EGF and cellassociated radioactivity was measured as outlined in MATERIALS AND METHODS.
for binding, nor did dexamethasone (Table 1). Vasopressin, which has been shown to affect EGF-URO binding indirectly via a transmodulation mechanism in Al0 cell cultures (3), decreased ‘“51-EGF-UR0 binding slightly (Table 1) but did so only at very high peptide concentrations (1 pg/ml) compared with the concentrations of
W A
:
20
a
0 Ill
1.0
1
1
3.0
1
1 IIllll
I
I
I
l1l
100
10
CONCENTRATION
I
200
(rig/ml)
5. Competition-inhibition of I”‘I-EGF-URO binding to guinea pig liver plasma membranes by EGF-URO and TGF-cu. Crude “microsomal” plasma membranes (75 pg) were incubated with 6 rig/ml of lar,IEGF-URO (sp act 190 cpm/pg) in a final volume of 200 ~1 for 50 min at room temperature, and reaction was terminated by harvesting and washing membranes on 0.22~pm filters with cold PBS. Specific binding observed in the absence of competitor (1,840 & 150 cpm) was taken as 100% and specific binding observed in the presence of increasing concentrations of either EGF-URO (0) or TGF-tu (A) was expressed relative to the 100% value. Data points represent average k SE for estimates done in triplicate. Figure is representative of 4 independently conducted experiments.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
G832
GASTRIC
SMOOTH
MUSCLE
URO > TGF-AL In four independent experiments, the average I& values (*SE) were for EGF-URO 37 t 2.7 rig/ml (6 nM) and for TGF-cu 60 t 5.5 rig/ml (10 nM). The I& values for EGF-URO and TGF-cu were statistically different (P < 0.01). The same order of binding affinities (EGF-URO > TGF-c~) was also observed in a human placental membrane binding assay using the same set of reagents employed in all experiments (not shown). Although the relative values of the I& provide a reliable estimate of the relative affinities of EGF-URO and TGF-cu for the liver receptor, the absolute values of the IC50s must be interpreted with caution (19). Nonetheless, the data obtained with the liver membranes suggest that, in addition to having a peptide specificity that is the reverse of the one observed in the GCM cells, the liver membranes also possessa receptor with a much lower affinity for EGF-URO and TGF-cu compared with the GCM cell receptor affinity. The binding-competition data are summarized in Table 2. Equilibrium binding of “’ I-EGF- URO. The binding of ““I-EGF-URO to GCM cell monolayers as a function of increasing concentrations of radiolabeled peptide appeared saturable, as illustrated in Fig. 6, approaching a maximum at -200 pg/lO’ cells (-2 X lo4 sites/cell). Analysis of the binding data according to Scatchard (33) (inset, Fig. 6) yielded a curvilinear plot that indicated saturability of binding (x-intercept) and that could be interpreted in terms of two classes of binding sites with different affinities. By use of a two-site model the curvefitting was compatible with one high-affinity site [& = 0.67 nM; maximal binding (B,,,) = 60 pg/lO’ cells] and one lower affinity site (& = 4.3 nM; B,,, = 180 pg/106 cells). The total number of binding sites (high plus low affinity) obtained by Scatchard analysis of the binding data (2.4 x lo4 sites/cell) was in good accord with the maximal binding estimated visually from the binding curve in Fig. 6. Even though we were able to establish that ‘““I-labeled TGF-cu prepared in our laboratory bound efficiently to human placental membrane, using previously established methods (16), we were not able to obtain consistent measurements of the binding of “‘Ilabeled TGF-cu to the GCM monolayers. Thus it was not possible to do the same kind of binding experiments with ““I-TGF-cw as were done with ‘“‘I-EGF-URO. Thymidine incorporation assay. Concentration-response curves for the stimulation of thymidine incorporation in GCM cell monolayers by EGF-URO and TGFCYare shown in Fig. 7. The data were expressed as a percentage of the stimulation caused by the addition of 10% FCS. Under the same conditions of assay for both 2. Summary of binding-competition and bioassay data TABLE
Hinding
Assay
( IC:T,o, &ml)
Pept ide GCM
EGF-URO TGF-tu
5.7tO.6 2.8t0.2*
cells
Liver
membranes 37k2.7 6Ok5 t .t5”
Bioassay (EG,,,, rig/ml)
EGF-URO
RECEPTOR CONCENTRATION 5.0 I
1.0 I
(nM) 10 1
15 I
50 BOUND
20 CONCENTRATION
40
100 (pg/
1 O6
100
60 OF ’ *%EGF-URO
150 cek)
(rig/ml)
rltx. 6. Binding of ““I-EGF-URO to GCM cells as a function of lzf,IEGF-URO concentration. Multiple, independently grown confluent cell monolayers were rinsed and incubated for 5 h at 4°C with 1 ml binding medium containing increasing concentrations of I”‘I-EGF-URO (sp act 210 cpm/pg) either in the absence or presence of a 5OO-fold excess of unlabeled EGF-URO. Amount of cell-associated radioactivity was then determined as described in MATERIALS AND METHODS. Only specific binding is shown. Nonspecific binding of radiolabeled EGF-URO was EGF-URO) is clearly distinct from the order of binding affinities measured in this study in the guinea pig liver (Table 2) and human placental membrane preparations (EGF-URO > TGFcu), from the order of potencies observed in cultured fibroblast systems (36), and from the order of biological potencies of the two polypeptides (EGF-URO > TGF-a) in the guinea pig gastric longitudinal muscle strip bioassay reported previously (21). In terms of the two hypotheses outlined above to account for the unusual order of potencies of TGF-cu and EGF-URO in the circular muscle tissue, our data lead us to favor the first alternative, namely that the cultured GCM cells and CM muscle strips possess a novel EGF-URO receptor subtype that is distinct from the one present in the guinea pig liver, for which the relative peptide affinities (EGF-URO > TGF-cy) mirror the receptor commonly found in other mammalian tissues. A receptor with a higher affinity for TGF-cu compared with EGF-URO has been observed in chicken fibroblasts (24). It will be of much interest to determine whether the TGF-a-preferring receptor found in the guinea pig tissue is also present in circular muscle preparations from other mammalian species. The preference of the circular muscle-derived cells for TGF-0 is of particular interest in view of the finding of much higher levels of TGF- QI compared with EGF-URO in the gastrointestinal mucosa along the length of the gastrointestinal tract, with a marked predominance of TGF-a in the stomach and colon (6). Physiologically, the TGF-(U produced by the mucosal cells could, by a paracrine mechanism, potentially play a role in regulating CM contractility via the receptor described in this study. The GCM cell cultures that we describe here should provide a useful model system for studying the biochemical pathways whereby the distinct EGF-URO receptor subtype in gastric circular muscle triggers a cellular response. We thank Dr. J. Blay for helpful discussions. This work was supported by funds from the Medical Research Council of Canada (to M. D. Hollenberg). S.-G. Yang is the recipient of a Fellowship award from the Alberta Heritage Foundation for Medical Research. Address for reprint requests: M. D. Hollenberg, Dept. of Pharmacology and Therapeutics, Faculty of Medicine, University of Calgary, 3330 Hospital Dr. NW, Calgary, Alberta T2N 4N1, Canada. Received
30 April
1990; accepted
in final
form
24 February
1991.
REFERENCES 1. BERK, B. C., T. A. BROCK, R. C. WEBB, M. B. TAUBMAN, W. J. ATKINSON, M. A. GIMBRONE, JR., AND R. W. ALEXANDER. Epidermal growth factor, a vascular smooth muscle mitogen, induces rat aortic contraction. J. CLin. Invest. 72: 1083-1086, 1985. 2. BLAY, J., AND K. D. BROWN. Functional receptors for epidermal
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
G834
3.
4.
5. 6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
16.
17. 18.
19.
20.
21.
GASTRIC
SMOOTH
MUSCLE
growth factor in an epithelial-cell line derived from the rat small intestine. Biochem. J. 225: 85-94, 1985. BLAY, J., AND M. D. HOLLENBERG. Heterologous regulation of EGF receptor function in cultured aortic smooth muscle cells. Eur. J. Pharmacol. Mol. Pharmacol. 172: 1-7, 1989. BROWN, K. D., AND D. M. BLAKELEY. Inhibition of the binding of ““I-labeled epidermal growth factor to mouse cells by a mitogen in goat mammary secretions. Biochem. J. 212: 465-472, 1983. CARPENTER, G., AND S. COHEN. Epidermal growth factor. Annu. Rev. Biochem. 48: 193-216, 1979. CARTLIDGE, S. A., AND J. B. ELDER. Transforming growth factor (Y and epidermal growth factor levels in normal human gastrointestinal mucosa. Br. J. Cancer 60: 657-660, 1989. COHEN, S. Isolation of a mouse submaxillary gland protein accelerating incisor eruption and eyelid opening in the newborn animal. J. Biol. Chem. 237: 1555-1562, 1962. DELARCO, J. E., AND G. J. TODARO. Growth factors from murine sarcoma virus-transformed cells. Proc. Natl. Acad. Sci. USA 75: 4001-4005, 1978. DERYNCK, R., A. B. ROBERTS, M. E. WINKLER, E. Y. CHEN, AND D. V. GOEDDEL. Human transforming growth factor-a precursor structure and expression in E. coli. Cell 38: 287-297, 1984. GAN, B. S., M. D. HOLLENBERG, K. L. MACCANNELL, K. LEDERIS, M. E. WINKLER, AND R. DERYNCK. Distinct vascular actions of epidermal growth factor-urogastrone and transforming growth factor-u. J. Pharmacol. Exp. Ther. 242: 331-337, 1987. GAN, B. S., K. L. MACCANNELL, AND M. D. HOLLENBERG. Epidermal growth factor-urogastrone causes vasodilatation in the anesthetized dog. J. Clin. Invest. 80: 199-206, 1987. GIMBRONE, M. A., JR., AND R. S. CONTRAN. Human vascular smooth muscle in culture. Lab. Invest. 33: 16-27, 1975. GOWN, A., AND A VOGEL. Monoclonal antibodies to intermediate filament proteins of human cells. Unique and cross-reacting antibodies. J. Cell Biol. 95: 414-424, 1982. GREGORY, H. Isolation and structure of urogastrone and its relationship to epidermal growth factor. Nature Lond. 257: 325-327, 1975. GUNTHER, S., R. W. ALEXANDER, W. J. ATKINSON, AND M. A. GIMBRONE. Functional angiotensin II receptors in cultured vascular smooth muscle cells. J. Cell Biol. 92: 289-298, 1982. HOCK, R. A., AND M. D. HOLLENBERG. Characterization of the receptor for epidermal growth factor-urogastrone in human placenta membranes. J. Biol. Chem. 255: 10731-10736,198O. HOLLENBERG, M. D. Epidermal growth factor-urogastrone: a polypeptide acquiring hormonal status. Vitam. Horm. 37: 69-110, 1979. HOLLENBERG, M. D., AND P. CUATRECASAS. Insulin and epidermal growth factor. Human fibroblast receptors related to deoxyribonucleic acid synthesis and amino acid uptake. J. Biol. Chem. 250: 3845-3853, 1975. HOLLENBERG, M. D., AND P. CUATRECASAS. Principles and techniques for the study of plasma membrane receptors related to hormone action. Methods Cancer Rex 12: 317-366, 1976. HOLLENBERG, M. D., AND H. GREGORY. Epidermal growth factorurogastrone: biological activity and receptor binding of derivatives. Mol. Pharmacol. 17: 314-320, 1980. HOLLENBERG, M. D., I. MURAMATSU, H. ITOH, P. PATEL, S.-G. YANG, AND K. LEDERIS. Contractile actions of epidermal growth factor-urogastrone in isolated smooth muscle preparations from guinea pig stomach: structure-activity relationships and comparison with the effects of human transforming growth factor-alpha. J. Pharmacol. Exp. Ther. 248: 384-390, 1989.
EGF-URO
RECEPTOR
22. ITOH, H., I. MURAMATSU, P. PATEL, M. D. HOLLENBERG, AND K. LEDERIS. Inhibition by anti-inflammatory agents of contraction induced by epidermal growth factor-urogastrone in isolated longitudinal smooth muscle strips from guinea pig stomach. Br. J. Pharmacol. 95: 821-829, 1988. 23. KAO, H. W., S. E. FINN, A. M. GOWN, J. LECHAGO, N. LACHANT AND W. J. SNAPE, JR. Cultured circular smooth muscle from the rabbit colon. In Vitro Cell. Dev. Biol. 24: 787-794, 1988. 24. LAX, I., A. JOHNSON, R. HOWK, J. SAP, F. BELLOT, M. WINKLER, A. ULLRICH, B. VENNSTROM, J. SCHLESSINGER, AND D. GIVOL. Chicken epidermal growth factor (EGF) receptor: cDNA cloning, expression in mouse cells, and differential binding of EGF and transforming growth factor alpha. Mol. Cell. Biol. 8: 1970-1978, 1988. 25. MASSAGUE, J. Epidermal growth factor-like transforming growth factor, isolation, chemical characterization, and potentiation by other transforming factors from feline sarcoma virus-transformed rat cells. J. Cell Biol. 258: 13606-13613, 1983a. 26. MASSAGUE, J. Epidermal growth factor-like transforming growth factor, interaction with epidermal growth factor receptors in human placenta membranes and A431 cells. J. Cell Biol. 258: 1361413620,1983b. 27. MURAMATSU, I., M. D. HOLLENBERG, AND K. LEDERIS. Vascular actions of epidermal growth factor-urogastrone: possible relationship to prostaglandin production. Can. J. Physiol. Pharmacol. 63: 994-999, 1985. 28. MURAMATSU, I., M. D. HOLLENBERG, AND K. LEDERIS. Modulation by epidermal growth factor-urogastrone of contraction in isolated canine helical mesenteric arterial strips. Can. J. Physiol. Pharmacol. 64: 1561-1565, 1986. 29. MURAMATSU, I., H. ITOH, K. LEDERIS, AND M. D. HOLLENBERG. Distinctive actions of epidermal growth factor-urogastrone in isolated smooth muscle preparations from guinea pig stomach: Differential inhibition by indomethacin. J. Pharmacol. Exp. Ther. 245: 625-631, 1988. 30. PATEL, P., H. ITOH, K. LEDERIS, AND M. D. HOLLENBERG. Contraction of guinea pig trachea by epidermal growth factor-urogastrone. Can. J. Physiol. Pharmacol. 66: 1308-1312, 1988. 31. ROSS, R., AND B. KARIYA. Morphogenesis of vascular smooth muscle in atherosclerosis and cell culture. In: Handbook of Physiology. The Cardiovascular System. Vascular Smooth Muscle. Bethesda, MD: Am. Physiol. Sot., 1980, sect. 2, vol. II, chapt. 3, p. 69-92. 32. SAVAGE, C. R., JR., AND S. COHEN. Epidermal growth factor and a new derivative. J. Biol. Chem. 247: 7609-7611, 1972. 33. SCATCHARD, G. The attraction of proteins for small molecules and ions. Ann. NY Acad. Sci. 51: 660-672, 1949. 34. TAM, J. P., H. MARQUARDT, D. F. ROSBERGER, T. W. WONG, AND G. J. TODARO. Synthesis of biologically active rat transforming growth factor I. Nature Lond. 309: 376-378, 1984. 35. TSUKADA, T., D. TIPPENS, D. GORDON, R. Ross, AND A. M. GOWN. HHF35, a muscle-actin-specific monoclonal antibody. I. Immunocytochemical and biochemical characterization. Am. J. Pathol. 126: 51-60, 1987. 36. WINKLER, M. E., T. BRINC;MAN, AND B. J. MARKS. The purification of fully active recombinant transforming growth factor CY produced in Escherichia coli. J. Biol. Chem. 29: 13838-13843, 1986. 37. YANG, S.-G., M. E. WINKLER, AND M. D. HOLLENBERG. Contribution of the C-terminal dipeptide of transforming growth factor alpha to its activity: biochemical and pharmacological profiles. Eur. J. Pharmacol. Mol. Pharmacol. 188: 289-300, 1990.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
factor-urogastrone
S.-G. YANG AND M. D. HOLLENBERG Endocrine Research Group, Department of Pharmacology and Therapeutics and Department of Medicine, Faculty of Medicine, University of Calgary, Calgary, Alberta T2N 4N1, Canada
YANG,S.-G., AND M.D. HOLLENBERG.D~S~~T-K~ receptorsfor epidermal growth factor-urogastrone in cultured gastric smooth muscle ceLLs. Am. J. Physiol. 260 (Gastrointest. Liver Physiol. 23): G827-G834, 1991.-We describe a propagable cell strain from guinea pig gastric circular muscle (GCM), which we have characterized in terms of its smooth muscle phenotype and its binding and biological response (thymidine incorporation) to epidermal growth factor-urogastrone (EGF-URO) and transforming growth factor-a (TGF-cu). The binding of ‘““I-labeled EGF-URO to the GCM cells exhibited high affinity and an appropriate peptide specificity. A curvilinear Scatchard plot of the binding data indicated two classes of high-affinity binding sites (dissociation constants of 0.69 and 4.3 nM) and a maximal binding capacity of 24,000 sites/cell. Binding competition data demonstrated that the binding affinity of TGF-cu was greater than that of EGF-URO by a factor of 2. These relative binding affinities agreed with the two- to threefold greater potency of TGF-cu, compared with EGF-URO, for the stimulation of GCM thymidine incorporation. The relative order of binding affinity and biological potency (TGF-cu > EGF-URO) was distinct from the relative order of binding affinities (EGF-URO > TGF-cu) observed using guinea pig liver and human placental membrane preparations. We conclude that the cultured smooth musclederived GCM cells possess a receptor subtype that is in accord with contractile bioassay data obtained previously with intact gastric circular muscle strips. This receptor (TGF-c~ > EGFURO) appears distinct from the one previously characterized in nonmuscle tissues. epidermal growth factor-urogastrone receptor subtype; smooth muscle cultures; transforming growth factor-a
GROWTH FACTOR-urogastrone (EGF-URO), a 6-kDa polypeptide with mitogenic and acid-inhibitory activities, was originally isolated from male mouse submaxillary glands or from human urine on the basis of its ability to stimulate precocious eyelid opening in newborn mice or to inhibit gastric acid secretion in dogs (7, 14). EGF-URO is structurally closely related to transforming growth factor-a (TGF-a), a homologous polypeptide that can also bind with high affinity to the EGF-URO receptor (8, 25, 26, 34). It is recognized that EGF-URO and TGF-cu can exert their action on a wide variety of target cells, but their exact physiological role is not yet clear (e.g., seeRefs. 5,17). Because of our original observations (27), supported by the work of others (l), indicating that EGF-URO is a potent regulator of smooth muscle contractility, we have recently focused our attention on the actions of EGF-URO in a variety of vascular and nonEPIDERMAL
019% l&57/91
$1.50 Copyright
vascular smooth muscle systems (10, 11, 22, 27-30). In particular, work with a gastric muscle strip preparation from the guinea pig revealed that gastric circular smooth muscle possessesa receptor system for EGF-URO that is distinct from the one in other previously studied tissues. In this circular muscle (CM) preparation, the potency of TGF-cu is greater than that of EGF-URO, whereas the reverse is true for the longitudinal smooth muscle present in the same tissue, as well as for a variety of other EGF-URO target tissues (21, 29). Two hypotheses can be put forward to explain the unusual order of biological potencies for TGF-a and EGF-URO in the CM tissue: 1) there may be a difference in the order of receptor binding affinities for TGF-cu and EGF-URO in the CM tissue, compared with the relative affinities in other tissues, pointing to a distinct receptor subtype in the CM tissue; or 2) differences in receptor signaling pathways in the CM tissue compared with other tissues might in some manner reverse the biological potencies of TGF-cu and EGF-URO as measured by bioassay, even though their relative receptor binding affinities might be the same as those observed in other tissues (affinity of EGF-URO > TGF-cw). To distinguish between these two hypotheses at the biochemical level, it is essential to measure directly the relative binding affinities of EGF-URO and TGF-cu in the gastric circular muscle tissue. Thus the goal of the work we describe here was to measure the receptor binding of EGF-URO and TGF-cu in a CM preparation. In preliminary work with membranes isolated from CM muscle strips, it became apparent that the abundance of receptor was far too low for a meaningful biochemical analysis of the EGF-URO receptor by methods we have used successfully for other tissues (e.g., see Ref. 16). We therefore turned our attention to the culture of CM-derived smooth muscle cells, in the hope that the conditions of culture might lead to an increased expression of receptors. In such cultures, the receptor for EGFURO could be studied by approaches that we have used previously with human fibroblasts (e.g., see Ref. 20). In the work that we report here, we describe the characterization of the receptor for EGF-URO by ligand binding and bioassay (thymidine incorporation) methods in a cultured cell strain derived from guinea pig gastric circular smooth muscle (GCM). The GCM cells exhibited a smooth muscle phenotype in terms of their “hill-andvalley” growth pattern and their immunocytochemical reactivity with a muscle actin-specific monoclonal anti-
(~1 1991 the American
Physiological
Societ,y
G827
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
G828
GASTRIC
SMOOTH
MUSCLE
body (HHF35). Our results support the first of the two hypotheses outlined above, suggesting that the GCM cells possess a distinct receptor subtype for EGF-URO. MATERIALS
AND
METHODS
Peptides and other reagents. Murine EGF-URO was isolated from fresh-frozen submaxillary glands of testosterone-treated male mice (32). The polypeptide yielded a single peak when analyzed by reverse-phase liquid chromatography (C,, column) and yielded a single band upon gel electrophoresis. In a routine mitogenesis assay using human fibroblasts EGF-URO has a concentration causing a half-maximal response (EC& of -0.25 rig/ml (20). TGF-cu, a gift from Dr. M. E. Winkler, Genentech, San Francisco, was expressed in Escherichia coli as described (9), purified by Sephadex G-75 gel filtration followed by high-performance liquid chromatography (36). On a weight basis, TGF-cu has a specific activity of 0.55 ng of EGF-URO receptor equivalents/rig of TGF-a (36, 37). The TGF- CYused in this study was a sample from the same lot (PD-1) as used in our previous study (37). The concentrations of the stock solution of EGFURO were measured spectrophotometrically as done previously (ZO), using the formula (Ezlr, - E2& X 155 = pg/ ml. The concentration of TGF-a in the stock solution was determined by amino acid analysis. We have compared the activities of EGF-URO with TGF-cu on a weight or molar basis rather than on the basis of equal receptor occupancy (21,37). Fetal calf serum (FCS) was obtained from Flow Laboratories (Mississauga, Ontario); lz51- and [ methyl-“H] thymidine were from Amersham (Arlington Heights, IL). Oxytocin, dexamethasone, and [Arg8]vasopressin were obtained from Sigma (St. Louis, MO). All other chemicals were of reagent grade or better. Isolation of gastric circular smooth muscle and preparation of primary cell cultures. The standard isolation solution was of the following composition (in mM): 1.3 CaC12, 5.4 KCl, 0.4 KH2P04, 0.4 MgCl, . 6H20, 0.4 MgSOdo 7Hz0, 137 NaCl, 4.2 NaHCO,?, 0.34 Na2HP04= 7Hg0, 5.6 glucose, and 25 N-Z-hydroxyethylpiperazine1V’-2-ethanesulfonic acid (HEPES). The pH of the solution was adjusted to 7.4 with NaOH, and the solution was equilibrated with 95%02-5%CO,, sterilized by filtration with a 0.22-,um filter, and supplemented with antibiotics. Male guinea pigs were killed by a blow on the head, and the stomach was removed. The stomach was cut open, the contents were removed, and the tissue was washed twice in cold isolation solution. The gastric body was then cut off from the gastric fundus and antrum. The tissue was then put into a dissecting bath filled with isolation solution maintained at 4°C and aerated continuously with 95%0,-5%COz. The mesentery was trimmed, and the mucosa was removed with fine scissors. The two layers of the gastric body smooth muscle were then separated with forceps under a dissecting microscope. The CM was removed and washed three times with 50 ml of sterilized isolation solution and was then minced into -l-mm’ pieces. Freshly minced CM segments were centrifuged at 25 g for 5 min at room temperature. The segments were resuspended in growth medium compris-
EGF-IJRO
RECEPTOR
ing Dulbecco’s modified Eagle’s medium (DMEM) supplemented with high glucose (24 mM), 5% (vol/vol) FCS, 4 mM L-glutamine but without antibiotics. Segments were plated in sterile 80-cm2 plastic flasks at a density of 2-3 pieces/cm” with 2 ml growth medium and were incubated at 37°C in a humidified atmosphere of 10% CO2 in room air. After the pieces had adhered (usually 2-3 days) cells were fed with 20 ml growth medium and were refed at weekly intervals until a substantial outgrowth of cells was observed (2-3 wk). When the outgrowth spread to cover at least 60% of the growth surface, the cells were subcultured with trypsin dissociation, using routine procedures for cell culture. The GCM cell strains obtained from individual animals were used for up to 10 generations of subculture at a time interval of 1 wk and at a passage ratio of 1:4 (each passage leads to 2 population doublings). For subcultures, the GCM cells were seeded at 2 x lo4 cell/ml in 80-cm2 (surface area) T flasks containing 20 ml DMEM and 5% (vol/vol) FCS. To avoid complications due to receptor variation that might be caused by different growth rates or different cell densities, only monolayers seeded at comparable cell densities and grown to confluence were used for our studies. Immunocytochemical staining of cultured GCM cells. HHF35, a muscle-specific monoclonal anti-actin antibody, was used to identify the cultured GCM cells by indirect immunofluorescent staining (13,23,35). HHF35 recognizes the o-actin of skeletal, cardiac, and smooth muscle and the y-actin of smooth muscle (23, 35). The cells for immunofluorescence staining were cultured on cover slips in multiwell plates. The cover slips were rinsed with Hanks’ balanced salt solution, treated with acetone at -20°C for 10 min and air dried. Indirect immunofluorescence was measured as described (13, 23). The cover slips of the cell cultures were incubated with HHF35 (l:l,OOO dilution, Enzo Diagnostics, New York, NY) at 21°C overnight, were washed with phosphatebuffered saline (PBS), and were then incubated with 0.1% bovine serum albumin (BSA)-Triton X-100 (100 pi/l) at 21°C for 5 min. The cover slips were washed again with PBS and incubated with fluorescence-conjugated goat anti-mouse immunoglobulin G (IgG; 1:64, Sigma, St. Louis, MO) at 21°C for 30 min. After the final wash, the cover slips were dehydrated, mounted on glass slides, and then photographed with a Zeiss III photomicroscope. Guinea pig liver plasma membrane preparations. Crude “microsomal” membranes were obtained from male guinea pig liver tissue by differential centrifugation according to the methods used routinely for isolating human placenta membranes in this laboratory (16). The isolation buffers and the cocktail of protease inhibitors were exactly as described previously (37). Binding assay. EGF-URO was labeled with 12’1 essentially as described previously (19), except that the 12’1EGF-URO was separated from the reaction mixture by chromatography on Sephadex G-10. This method of radiolabeling has been previously shown to yield ‘““I-EGFURO that is fully active in a thymidine incorporation bioassay (summarized in Ref. 19). The binding of “‘IEGF-URO to the GCM cells was performed on cell
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
GASTRIC
SMOOTH
MUSCLE
monolayers grown to confluence after subculture from T flasks into 3.5cm-diam Nunclon multidish trays (-120,000 cells/dish; GIBCO, Grand Island, NY). Binding was measured essentially as described previously (2, 18), except that the binding medium (DMEM containing 1 mg/ml BSA and 0.1 PM KI) was buffered at pH 7.4 with 20 mM HEPES. After equilibration, unbound radioactivity was washed from the cells with ice-cold PBS (pH 7.4) containing 1 mg/ml BSA and 0.1 PM KI. The washed cells were then solubilized in 1 ml of 1.0 M NaOH (60 min at room temperature), and the radioactivity was determined by crystal scintillation counting (efficiency -85%). Individual subcultures for the binding assays were seeded independently from stock T flasks and were grown in separate crops for separate binding assays. The binding of lz51-EGF-URO to the guinea pig liver plasma membranes and to human placental membranes was performed as described (16, 37), with a final reaction volume of 0.2 ml and 75 pg of membrane/assay tube. The specific binding was defined as the difference between the total amount of radioactivity bound minus the amount bound in the presence of a ?500-fold excess of unlabeled EGF-URO. In most experiments, wherein the concentration of radiolabeled EGF-URO was 70% of the labeled EGF-URO that had initially been bound to the cells at the start of the 37°C dissociation time course. Upon dissociation at 37°C intact ‘““I-EGF-URO released into the incubation medium (open circles, Fig. 3) matched exactly the reduction in cell surface-bound peptide (i.e., acid elutable; open triangles, Fig. 3). Only a small proportion of the radiolabel was either internalized by the cells (closed triangles, Fig. 3) or was resistant to precipitation by TCA (closed circles, Fig. 3). Thus we were able to conclude that even at 37°C the cellular uptake and degradation of ‘““I-EGFURO by GCM cells was very low, in contrast to previous studies that had been done with cultured non-smooth muscle cells (e.g., see Ref. 2). In view of the above findings, all subsequent binding assays were performed at 4°C with an equilibration time of 5 h. Binding-competition studies. Unlabeled EGF-URO and TGF-cu both competed efficiently for the binding of 12’1EGF-URO to the GCM cell monolayers, as illustrated in Fig. 4 and Table 1. In contrast, other disulfide-containing polypeptides like oxytocin and insulin did not compete
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
GASTRIC
CONCENTRATION
SMOOTH
MUSCLE
EGF-URO
G831
RECEPTOR
(nM)
EGF-URO and TGF-cu that maximally inhibited binding (50-100 rig/ml). In terms of binding inhibition, the con1.0 10 - centration of unlabeled EGF-URO that reduced lz51I I - EGF-URO binding by 50% (I&) was 5.7 t 0.6 rig/ml (means t SE for 3 independently conducted binding competition curves; 1.0 nM); in contrast, the IC& for - TGF-cu under identical conditions was 2.8 t 0.2 rig/ml (means t SE, n = 3; 0.5 nM). These two values of the IC&s for EGF-URO and TGF-cu obtained using three separately grown crops of cell monolayers were statistically different (P c 0.01). The binding inhibition assay yields reliable information about the relative affinities of unlabeled ligands for a receptor site (e.g., see Ref. 20) but does not necessarily yield information about the possible existence of multiple affinity sites (see equilibrium binding data below). Nonetheless, in the GCM cells, the binding competition experiments demonstrated that the order of binding affinities was TGF-cu > EGF-URO. j This order of binding affinities was retained over at least 1 I llll1ll I I1r11lll 1 II 111 six to seven GCM cell generations (passages 3-10) and 1.0 3.0 10 100 200 was observed for two cell strains independently isolated CONCENTRATION (rig/ml) from separate animals. FIG. 4. Competition-inhibition of ‘““I-EGF-URO binding to GCM In the guinea pig liver membrane binding assay, TGFcell monolayers by EGF-URO and TGF-cu. Multiple, independently cy and EGF-URO were also able to compete efficiently grown confluent cell monolayers were incubated for 5 h at 4°C with 6 for the binding of ‘““I-EGF-URO. However in this sysrig/ml of ““I-EGF-URO (190 cpm/pg) in the presence of increasing tern, under identical conditions of assay, the I&, for concentrations of unlabeled EGF-URO (0) and TGF-(U (A). Net amount of specifically bound radioactivity was then determined as outlined in EGF-URO was repeatedly smaller than the I& for TGFMATERIALS AND METHODS. Specific binding in the absence of competcy (Fig. 5); i.e., the order of binding affinities was EGFitor (1,930 & 60 cpm) was expressed as lOO%, and binding of “‘1-EGFURO at each concentration of unlabeled competitor was calculated relative to the 100% value. Each point represents average t SE of measurements made on 3 independently grown monolayers. Figure is representative of 5 independent experiments on separately grown crops of monolayers, using cell strains derived from 2 separate animals, between passages 3 and 10.
TABLE
CONCENTRATION
(nM)
1. Binding competition studies E Addit
ion
None Oxytocin Insulin Dexamethasone Vasopressin EGF-URO EGF-URO TGF-CY
Concentration, K/ml
1 10 1 1 0.010 0.050 0.010
‘“‘I-EGF-URO pg/lO”
80
Bound, cells
44.0t1.2 42.4tl.O 44.3t2.6 41.7k1.5 31.4t1.4 22.1kl.O 14.7k1.2 16.5k1.4
Values are means t SE of measurements on 4 individually grown monolayers and are corrected for nonspecific binding. Under these conditions, the nonspecific binding of ““I-EGF-URO, measured in the presence of a 500-fold excess of unlabeled EGF-URO (3 pg/ml) was -8% of the total amount of 1’SI-EGF-URO bound. Individual confluent cell monolayers (3.5 cm diam) grown concurrently were rinsed and incubated for 5 h at 4°C with 6 rig/ml I”‘I-EGF-URO (230 cpm/pg) either in the presence or absence of the various added agents at the indicated concentrations in a final volume of 1 ml binding medium. Monolayers were then washed free of unbound “‘I-EGF and cellassociated radioactivity was measured as outlined in MATERIALS AND METHODS.
for binding, nor did dexamethasone (Table 1). Vasopressin, which has been shown to affect EGF-URO binding indirectly via a transmodulation mechanism in Al0 cell cultures (3), decreased ‘“51-EGF-UR0 binding slightly (Table 1) but did so only at very high peptide concentrations (1 pg/ml) compared with the concentrations of
W A
:
20
a
0 Ill
1.0
1
1
3.0
1
1 IIllll
I
I
I
l1l
100
10
CONCENTRATION
I
200
(rig/ml)
5. Competition-inhibition of I”‘I-EGF-URO binding to guinea pig liver plasma membranes by EGF-URO and TGF-cu. Crude “microsomal” plasma membranes (75 pg) were incubated with 6 rig/ml of lar,IEGF-URO (sp act 190 cpm/pg) in a final volume of 200 ~1 for 50 min at room temperature, and reaction was terminated by harvesting and washing membranes on 0.22~pm filters with cold PBS. Specific binding observed in the absence of competitor (1,840 & 150 cpm) was taken as 100% and specific binding observed in the presence of increasing concentrations of either EGF-URO (0) or TGF-tu (A) was expressed relative to the 100% value. Data points represent average k SE for estimates done in triplicate. Figure is representative of 4 independently conducted experiments.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
G832
GASTRIC
SMOOTH
MUSCLE
URO > TGF-AL In four independent experiments, the average I& values (*SE) were for EGF-URO 37 t 2.7 rig/ml (6 nM) and for TGF-cu 60 t 5.5 rig/ml (10 nM). The I& values for EGF-URO and TGF-cu were statistically different (P < 0.01). The same order of binding affinities (EGF-URO > TGF-c~) was also observed in a human placental membrane binding assay using the same set of reagents employed in all experiments (not shown). Although the relative values of the I& provide a reliable estimate of the relative affinities of EGF-URO and TGF-cu for the liver receptor, the absolute values of the IC50s must be interpreted with caution (19). Nonetheless, the data obtained with the liver membranes suggest that, in addition to having a peptide specificity that is the reverse of the one observed in the GCM cells, the liver membranes also possessa receptor with a much lower affinity for EGF-URO and TGF-cu compared with the GCM cell receptor affinity. The binding-competition data are summarized in Table 2. Equilibrium binding of “’ I-EGF- URO. The binding of ““I-EGF-URO to GCM cell monolayers as a function of increasing concentrations of radiolabeled peptide appeared saturable, as illustrated in Fig. 6, approaching a maximum at -200 pg/lO’ cells (-2 X lo4 sites/cell). Analysis of the binding data according to Scatchard (33) (inset, Fig. 6) yielded a curvilinear plot that indicated saturability of binding (x-intercept) and that could be interpreted in terms of two classes of binding sites with different affinities. By use of a two-site model the curvefitting was compatible with one high-affinity site [& = 0.67 nM; maximal binding (B,,,) = 60 pg/lO’ cells] and one lower affinity site (& = 4.3 nM; B,,, = 180 pg/106 cells). The total number of binding sites (high plus low affinity) obtained by Scatchard analysis of the binding data (2.4 x lo4 sites/cell) was in good accord with the maximal binding estimated visually from the binding curve in Fig. 6. Even though we were able to establish that ‘““I-labeled TGF-cu prepared in our laboratory bound efficiently to human placental membrane, using previously established methods (16), we were not able to obtain consistent measurements of the binding of “‘Ilabeled TGF-cu to the GCM monolayers. Thus it was not possible to do the same kind of binding experiments with ““I-TGF-cw as were done with ‘“‘I-EGF-URO. Thymidine incorporation assay. Concentration-response curves for the stimulation of thymidine incorporation in GCM cell monolayers by EGF-URO and TGFCYare shown in Fig. 7. The data were expressed as a percentage of the stimulation caused by the addition of 10% FCS. Under the same conditions of assay for both 2. Summary of binding-competition and bioassay data TABLE
Hinding
Assay
( IC:T,o, &ml)
Pept ide GCM
EGF-URO TGF-tu
5.7tO.6 2.8t0.2*
cells
Liver
membranes 37k2.7 6Ok5 t .t5”
Bioassay (EG,,,, rig/ml)
EGF-URO
RECEPTOR CONCENTRATION 5.0 I
1.0 I
(nM) 10 1
15 I
50 BOUND
20 CONCENTRATION
40
100 (pg/
1 O6
100
60 OF ’ *%EGF-URO
150 cek)
(rig/ml)
rltx. 6. Binding of ““I-EGF-URO to GCM cells as a function of lzf,IEGF-URO concentration. Multiple, independently grown confluent cell monolayers were rinsed and incubated for 5 h at 4°C with 1 ml binding medium containing increasing concentrations of I”‘I-EGF-URO (sp act 210 cpm/pg) either in the absence or presence of a 5OO-fold excess of unlabeled EGF-URO. Amount of cell-associated radioactivity was then determined as described in MATERIALS AND METHODS. Only specific binding is shown. Nonspecific binding of radiolabeled EGF-URO was EGF-URO) is clearly distinct from the order of binding affinities measured in this study in the guinea pig liver (Table 2) and human placental membrane preparations (EGF-URO > TGFcu), from the order of potencies observed in cultured fibroblast systems (36), and from the order of biological potencies of the two polypeptides (EGF-URO > TGF-a) in the guinea pig gastric longitudinal muscle strip bioassay reported previously (21). In terms of the two hypotheses outlined above to account for the unusual order of potencies of TGF-cu and EGF-URO in the circular muscle tissue, our data lead us to favor the first alternative, namely that the cultured GCM cells and CM muscle strips possess a novel EGF-URO receptor subtype that is distinct from the one present in the guinea pig liver, for which the relative peptide affinities (EGF-URO > TGF-cy) mirror the receptor commonly found in other mammalian tissues. A receptor with a higher affinity for TGF-cu compared with EGF-URO has been observed in chicken fibroblasts (24). It will be of much interest to determine whether the TGF-a-preferring receptor found in the guinea pig tissue is also present in circular muscle preparations from other mammalian species. The preference of the circular muscle-derived cells for TGF-0 is of particular interest in view of the finding of much higher levels of TGF- QI compared with EGF-URO in the gastrointestinal mucosa along the length of the gastrointestinal tract, with a marked predominance of TGF-a in the stomach and colon (6). Physiologically, the TGF-(U produced by the mucosal cells could, by a paracrine mechanism, potentially play a role in regulating CM contractility via the receptor described in this study. The GCM cell cultures that we describe here should provide a useful model system for studying the biochemical pathways whereby the distinct EGF-URO receptor subtype in gastric circular muscle triggers a cellular response. We thank Dr. J. Blay for helpful discussions. This work was supported by funds from the Medical Research Council of Canada (to M. D. Hollenberg). S.-G. Yang is the recipient of a Fellowship award from the Alberta Heritage Foundation for Medical Research. Address for reprint requests: M. D. Hollenberg, Dept. of Pharmacology and Therapeutics, Faculty of Medicine, University of Calgary, 3330 Hospital Dr. NW, Calgary, Alberta T2N 4N1, Canada. Received
30 April
1990; accepted
in final
form
24 February
1991.
REFERENCES 1. BERK, B. C., T. A. BROCK, R. C. WEBB, M. B. TAUBMAN, W. J. ATKINSON, M. A. GIMBRONE, JR., AND R. W. ALEXANDER. Epidermal growth factor, a vascular smooth muscle mitogen, induces rat aortic contraction. J. CLin. Invest. 72: 1083-1086, 1985. 2. BLAY, J., AND K. D. BROWN. Functional receptors for epidermal
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
G834
3.
4.
5. 6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
16.
17. 18.
19.
20.
21.
GASTRIC
SMOOTH
MUSCLE
growth factor in an epithelial-cell line derived from the rat small intestine. Biochem. J. 225: 85-94, 1985. BLAY, J., AND M. D. HOLLENBERG. Heterologous regulation of EGF receptor function in cultured aortic smooth muscle cells. Eur. J. Pharmacol. Mol. Pharmacol. 172: 1-7, 1989. BROWN, K. D., AND D. M. BLAKELEY. Inhibition of the binding of ““I-labeled epidermal growth factor to mouse cells by a mitogen in goat mammary secretions. Biochem. J. 212: 465-472, 1983. CARPENTER, G., AND S. COHEN. Epidermal growth factor. Annu. Rev. Biochem. 48: 193-216, 1979. CARTLIDGE, S. A., AND J. B. ELDER. Transforming growth factor (Y and epidermal growth factor levels in normal human gastrointestinal mucosa. Br. J. Cancer 60: 657-660, 1989. COHEN, S. Isolation of a mouse submaxillary gland protein accelerating incisor eruption and eyelid opening in the newborn animal. J. Biol. Chem. 237: 1555-1562, 1962. DELARCO, J. E., AND G. J. TODARO. Growth factors from murine sarcoma virus-transformed cells. Proc. Natl. Acad. Sci. USA 75: 4001-4005, 1978. DERYNCK, R., A. B. ROBERTS, M. E. WINKLER, E. Y. CHEN, AND D. V. GOEDDEL. Human transforming growth factor-a precursor structure and expression in E. coli. Cell 38: 287-297, 1984. GAN, B. S., M. D. HOLLENBERG, K. L. MACCANNELL, K. LEDERIS, M. E. WINKLER, AND R. DERYNCK. Distinct vascular actions of epidermal growth factor-urogastrone and transforming growth factor-u. J. Pharmacol. Exp. Ther. 242: 331-337, 1987. GAN, B. S., K. L. MACCANNELL, AND M. D. HOLLENBERG. Epidermal growth factor-urogastrone causes vasodilatation in the anesthetized dog. J. Clin. Invest. 80: 199-206, 1987. GIMBRONE, M. A., JR., AND R. S. CONTRAN. Human vascular smooth muscle in culture. Lab. Invest. 33: 16-27, 1975. GOWN, A., AND A VOGEL. Monoclonal antibodies to intermediate filament proteins of human cells. Unique and cross-reacting antibodies. J. Cell Biol. 95: 414-424, 1982. GREGORY, H. Isolation and structure of urogastrone and its relationship to epidermal growth factor. Nature Lond. 257: 325-327, 1975. GUNTHER, S., R. W. ALEXANDER, W. J. ATKINSON, AND M. A. GIMBRONE. Functional angiotensin II receptors in cultured vascular smooth muscle cells. J. Cell Biol. 92: 289-298, 1982. HOCK, R. A., AND M. D. HOLLENBERG. Characterization of the receptor for epidermal growth factor-urogastrone in human placenta membranes. J. Biol. Chem. 255: 10731-10736,198O. HOLLENBERG, M. D. Epidermal growth factor-urogastrone: a polypeptide acquiring hormonal status. Vitam. Horm. 37: 69-110, 1979. HOLLENBERG, M. D., AND P. CUATRECASAS. Insulin and epidermal growth factor. Human fibroblast receptors related to deoxyribonucleic acid synthesis and amino acid uptake. J. Biol. Chem. 250: 3845-3853, 1975. HOLLENBERG, M. D., AND P. CUATRECASAS. Principles and techniques for the study of plasma membrane receptors related to hormone action. Methods Cancer Rex 12: 317-366, 1976. HOLLENBERG, M. D., AND H. GREGORY. Epidermal growth factorurogastrone: biological activity and receptor binding of derivatives. Mol. Pharmacol. 17: 314-320, 1980. HOLLENBERG, M. D., I. MURAMATSU, H. ITOH, P. PATEL, S.-G. YANG, AND K. LEDERIS. Contractile actions of epidermal growth factor-urogastrone in isolated smooth muscle preparations from guinea pig stomach: structure-activity relationships and comparison with the effects of human transforming growth factor-alpha. J. Pharmacol. Exp. Ther. 248: 384-390, 1989.
EGF-URO
RECEPTOR
22. ITOH, H., I. MURAMATSU, P. PATEL, M. D. HOLLENBERG, AND K. LEDERIS. Inhibition by anti-inflammatory agents of contraction induced by epidermal growth factor-urogastrone in isolated longitudinal smooth muscle strips from guinea pig stomach. Br. J. Pharmacol. 95: 821-829, 1988. 23. KAO, H. W., S. E. FINN, A. M. GOWN, J. LECHAGO, N. LACHANT AND W. J. SNAPE, JR. Cultured circular smooth muscle from the rabbit colon. In Vitro Cell. Dev. Biol. 24: 787-794, 1988. 24. LAX, I., A. JOHNSON, R. HOWK, J. SAP, F. BELLOT, M. WINKLER, A. ULLRICH, B. VENNSTROM, J. SCHLESSINGER, AND D. GIVOL. Chicken epidermal growth factor (EGF) receptor: cDNA cloning, expression in mouse cells, and differential binding of EGF and transforming growth factor alpha. Mol. Cell. Biol. 8: 1970-1978, 1988. 25. MASSAGUE, J. Epidermal growth factor-like transforming growth factor, isolation, chemical characterization, and potentiation by other transforming factors from feline sarcoma virus-transformed rat cells. J. Cell Biol. 258: 13606-13613, 1983a. 26. MASSAGUE, J. Epidermal growth factor-like transforming growth factor, interaction with epidermal growth factor receptors in human placenta membranes and A431 cells. J. Cell Biol. 258: 1361413620,1983b. 27. MURAMATSU, I., M. D. HOLLENBERG, AND K. LEDERIS. Vascular actions of epidermal growth factor-urogastrone: possible relationship to prostaglandin production. Can. J. Physiol. Pharmacol. 63: 994-999, 1985. 28. MURAMATSU, I., M. D. HOLLENBERG, AND K. LEDERIS. Modulation by epidermal growth factor-urogastrone of contraction in isolated canine helical mesenteric arterial strips. Can. J. Physiol. Pharmacol. 64: 1561-1565, 1986. 29. MURAMATSU, I., H. ITOH, K. LEDERIS, AND M. D. HOLLENBERG. Distinctive actions of epidermal growth factor-urogastrone in isolated smooth muscle preparations from guinea pig stomach: Differential inhibition by indomethacin. J. Pharmacol. Exp. Ther. 245: 625-631, 1988. 30. PATEL, P., H. ITOH, K. LEDERIS, AND M. D. HOLLENBERG. Contraction of guinea pig trachea by epidermal growth factor-urogastrone. Can. J. Physiol. Pharmacol. 66: 1308-1312, 1988. 31. ROSS, R., AND B. KARIYA. Morphogenesis of vascular smooth muscle in atherosclerosis and cell culture. In: Handbook of Physiology. The Cardiovascular System. Vascular Smooth Muscle. Bethesda, MD: Am. Physiol. Sot., 1980, sect. 2, vol. II, chapt. 3, p. 69-92. 32. SAVAGE, C. R., JR., AND S. COHEN. Epidermal growth factor and a new derivative. J. Biol. Chem. 247: 7609-7611, 1972. 33. SCATCHARD, G. The attraction of proteins for small molecules and ions. Ann. NY Acad. Sci. 51: 660-672, 1949. 34. TAM, J. P., H. MARQUARDT, D. F. ROSBERGER, T. W. WONG, AND G. J. TODARO. Synthesis of biologically active rat transforming growth factor I. Nature Lond. 309: 376-378, 1984. 35. TSUKADA, T., D. TIPPENS, D. GORDON, R. Ross, AND A. M. GOWN. HHF35, a muscle-actin-specific monoclonal antibody. I. Immunocytochemical and biochemical characterization. Am. J. Pathol. 126: 51-60, 1987. 36. WINKLER, M. E., T. BRINC;MAN, AND B. J. MARKS. The purification of fully active recombinant transforming growth factor CY produced in Escherichia coli. J. Biol. Chem. 29: 13838-13843, 1986. 37. YANG, S.-G., M. E. WINKLER, AND M. D. HOLLENBERG. Contribution of the C-terminal dipeptide of transforming growth factor alpha to its activity: biochemical and pharmacological profiles. Eur. J. Pharmacol. Mol. Pharmacol. 188: 289-300, 1990.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (150.216.068.200) on January 15, 2019.
