0021-972X/90/7104-0923I02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society
Vol. 71, No. 4 Printed in U.S.A.
The Dual Effect of Epidermal Growth Factor upon Human Chorionic Gonadotropin Secretion by the First Trimester Placenta in Vitro EYTAN R. BARNEA, DORON FELDMAN, MARIELLE KAPLAN, AND DONALD W. MORRISH Feto-Placental Endocrinology Unit (E.R.B., D.F., M.K.), Rappaport Institute for Medical Research, and Department Ob/Gyn, Reproductive Endocrinology, Rambam Medical Center, Technion, Haifa, Israel; and Department of Medicine (D.W.M.), University of Alberta, Edmonton, Alberta, Canada T6G 2G3
ABSTRACT. Epidermal growth factor (EGF) is believed to play an important role in the regulation of placental function. We have examined the effect of EGF upon first trimester (7-10 gestational weeks) placental hCG secretion and cellular differentiation using both static (explants and isolated cells) and kinetic (superfusion of explants) culture methods. In superfused explants, short (1-4 min) pulses of EGF increased both the rate and amplitude of spontaneous pulsatility of hCG. The frequency increased from 3/h to 5/h, and the amplitude increased compared to the control channels as calculated by the area under the curve. This effect was dose dependent and the concentration of 50 ng/ml, which was the lowest dose tested, was the most
effective. In explants cultured for 24 h, EGF caused a 2-fold increase in hCG secretion, compared to control, P < 0.05. In two different dispersed trophoblastic cell cultures, EGF added daily for the first week caused a 180% increase in hCG secretion, P < 0.05. However, according to morphological criteria, i.e. light microscopy and vital staining, no significant effect upon the rate of differentiation to syncytiotrophoblast was observed in longterm cultures of one of these preparations. In conclusion, EGF plays a dual trophic role in stimulating hCG secretion in the first trimester. However, this effect is not dependent on cellular differentiation. (J Clin Endocrinol Metab 71: 923-928, 1990)
CG secretion follows a unique pattern after implantation: It increases exponentially until 5-6 weeks, slowly merges into a plateau phase at 8-10 weeks, gradually decreases at 11-13 weeks, and is subsequently maintained at a stable level until term (1). Despite extensive investigations, the factors involved in regulation of hCG secretion remain unknown. The vast majority of research conducted on the regulation of placental hCG was conducted at term, while studies done in the first trimester are limited. The role of epidermal growth factor (EGF) in the placenta was previously addressed (2-4). It leads to phosphorylation of membrane proteins (5) and acts as a mitogen (6) by increasing RNA and DNA synthesis. Current data are conflicting as to the effect of EGF upon hCG secretion: EGF was shown to increase hCG secretion at term (2), whereas others found either increase or no change in the first trimester (7). The effect of EGF upon trophoblastic differentiation in the first trimester is not known.
Most studies on hCG secretion used static culture techniques in which measurements can only be carried out after relatively long periods of time. Recently, a new method of placental superfusion was developed which allows real-time measurements of hCG secretion and provides a more physiologic approach to the study of hCG regulation. Using this model we found that the secretion of hCG is pulsatile (8). In the present study we have compared the effects of EGF upon first trimester placental hCG secretion, using both static and dynamic culture methods. Moreover, the effect of EGF upon cellular differentiation was examined.
Received December 14, 1989. Address correspondence and requests for reprints to: Eytan Barnea, M.D., F.A.C.O.G., Feto-Placental Endocrinology Unit, Rappaport Institute, Technion, P.O. Box 9697, Efron Street, Haifa Israel.
Placentas
h
Materials and Methods Chemicals EGF, trypsin, DNAase, HEPES, and soybean trypsin inhibitor were purchased from Sigma (St. Louis, MO). Dulbecco's modified Eagle's medium (DMEM) and fetal calf serum (FCS) were purchased from Beit Haemek Laboratories (Beit Haemek, Israel). MAIAclone kits for measurements of hCG were obtained from Serono (Rehovot, Israel). All other chemicals were of high analytical quality.
After obtaining appropriate consent, elective pregnancy terminations were carried out in the first trimester (7-10 weeks). 923
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 16:11 For personal use only. No other uses without permission. . All rights reserved.
924
BARNEA ETAL.
Patients were on no medication and were nonsmokers. Tissues collected were immediately placed on ice in sterile 0.9% NaCl solution. Explant preparation: The method was previously published (9). Briefly, explants 50-70 mg wet weight were dissected out and rinsed in 1% antibiotic solutions (penicillin, 10,000 U/ml; streptomycin, 10 ng/m\; amphotericin B, 25 ^g/ml). For culture, explants were placed in DMEM media with or without EGF. Twelve replicates per test agent or control were plated at 37° C in an atmosphere of 95% air and 5% CO2. After 24 h of incubation, the media were collected and stored at —20 C until hCG assay. The tissue was saved for protein analysis. In the preliminary experiments we found that hCG secretion is linear up to 48 h. Therefore in subsequent experiments, EGF effect was tested until 24 h at the half-maximal increase point. Cell culture Two methods were used as previously published with some modifications (10-12). In the first, after rinsing in cold 0.9% NaCl solution, followed by multiple rinsing in DMEM, the membranes were trimmed and the tissue was incubated with 0.25% trypsin-EDTA solution containing DNAse 200 Kunitz U/ml for 15 min with gentle agitation at 37° C. At the end of the incubation period, the cells were filtered through a sterile gauze. Soybean trypsin inhibitor (Sigma, St. Louis, MO) was added to the recovered cell suspension to block trypsin activity. Then, 3 ml trophoblastic cell suspension were mixed with 7 ml DMEM containing 10% FCS. Cells were separated by centrifugation at 500 X g X 10 min at room temperature and the supernatant was discarded. For culture, cells were resuspended in media composed of DMEM, 5% FCS and 1% garamycin 10 /ig/ml. Cells were counted by a hemocytometer and in each dish approximately 100,000 cells were plated and incubated at 37 C in an atmosphere of 95% air and 5% CO2. EGF at concentrations of 50-200 ng/ml was added after 24 h of culture, and the media were collected 48 h after the start of the experiment. Subsequently EGF was added and the media were collected daily for 7 days. Thereafter the media were collected every 3 days until the culture was terminated after 2-3 months. The media were stored at —20 C until assay. In the second method (11, 12) intact first trimester placenta without membranes was incubated for 8- to 10-min periods in 0.25% trypsin-10 n\/m\ DNAse I at 37° C in bicarbonate buffer. Cells were collected from the fourth to the eighth trypsinization, pooled, centrifuged at 200 x g for 7 min, and plated on 60-mm Petri dishes (Corning, Corning, NY) at 1-2 X 106 cells per dish in 10% FCS-DMEM. After 2 h, medium was changed to serum-free DMEM. From days 1-9 10 ng/ml EGF (Sigma) were added to test plates and 0.001% BSA (final concentration) to controls. Medium was collected daily (days 1-7) and frozen at -20 C until assay. Cytospins of dispersed cells were stained by the immunoperoxidase-avidin-biotin complex method as previously described (12) for vimentin, a macrophage monocyte marker (8243), and cytokeratin (Becton-Dickinson). The details of the histochemical studies were reported previously (10). Studies of differentiation were carried out using cultures prepared by the first method (10). Briefly, at different times during culture, the media were removed from control and EGF-treated dishes and the cells were fixed. Subsequently, they
JCE & M • 1990 Vol 7, • No 4
were stained with trypan blue and observed at light microscopy for morphological changes using previously established criteria (11). Stained dishes were blinded and analyzed for at least three microscopical fields in three different dishes from each control and EGF-treated cells. Superfusion culture The method of placental superfusion was recently reported (8). Briefly, a superfusion apparatus (Accusyst, Endotronics, St. Paul, MN) with a multichannel peristaltic pump and fraction collector (model 272, ISCO, Durham, NC) were used to study the short-term dynamics of hCG secretion. The explants (200-300 mg wet wt) were placed into the culture chambers and a HEPES (18 mM) DMEM solution was washed through in an atmosphere of 5% CO2 and 95% air at 37° C. Experiments were conducted for a 120-min period; a 1-ml sample from the effluent was collected every 2.4 min for hCG measurements. In each experiment, one channel served as control and four served as experimental channels. At given intervals, a 1-4-min pulse of EGF was given through a peristaltic pump equipped with a digital flowmeter (ISMATECH DD, Chicago, IL). Assay /3-hCG assay was carried out with a commercially available RIA kit with an intraassay variability of 1.7% (MAIA CLONE, Serono Diagnostics) or by intact hCG RIA as described (11, 12). All measurements were carried out in the same assay. Protein content was determined according to the Lowry protein assay procedure (13). Statistics Statistical analysis was performed using one way analysis of variance (ANOVA) and Student's t test, and P < 0.05 was considered as statistically significant. In superfusion experiments, the area under a graph was determined using Simpson's rule in a personal computer. This program is able to calculate a function that best fit the series of points we are giving, and then to determine the area under the curve using the integral of the function. Data of superfusion experiments represents a single channel run in duplicate in four different placentas yielding similar results. Data in explants and cells shows mean and SEM of 612 individual placentas. In each placenta 6-12 dishes were run.
Results Effect of EGF on hCG secretion Explants in superfusion . Data presented in Fig. 1, A and B, are representative of 10 different experiments. One minute pulses of EGF stimulated spontaneous pulsatile hCG secretion; the pulse frequency increased from 3/h to 5/h, and pulse amplitude increased as calculated by the area under the curve. The maximal effect was noted at 50 ng/ml EGF. At 100 ng/ml only a 2-fold increase in the amplitude was noted. Longer pulses of EGF (up to 4 min) did not have a stimulatory effect greater than that
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 16:11 For personal use only. No other uses without permission. . All rights reserved.
ROLE OF EGF IN REGULATION OF PLACENTAL FUNCTION
250
o CONTROL • TREATMENT II : 1 min pulse of EGF 50 n g / m l
A — A CONTROL • — • EGF 50ng/ml
1
z
925
200
o 150,
a: a.
EGF 50 n g / m l
50-
i
o a
30:. 10:. 0
30
180
TIME (min.)
160
minutes 50 &—a .-.
4 0 • •
CONTROL
o — o EGF 100ng/ml
o
eg CL
FIG. 2. Effect of continuous administration of EGF 50 ng/ml for 90 minutes on placental explants in superfusion system. There was a significant decrease after a first increase. After cessation of continuous administration 1-min pulse of EGF 50 ng/ml had only a mild stimulatory effect.
30-LJ O
20-E o o
300 •-
CD CONTROL C3DEGF 100ng/ml EfflEGF 200ng/ml
200 ••
o en
a>
TIME'(min.)
_
V)
_
a
a
a
3 Q.
FIG. 1. The stimulatory effect of 1-min pulses of 50 ng/ml (A) and 100 ng/ml (B) EGF upon spontaneous pulsatile hCG secretion by placental explants in superfusion. The 50 ng/ml concentration was more effective in stimulating hCG secretion as calculated by the area under the curve (see Materials and Methods section for further detail). Data represent results obtained from 4 different placentas in a total of 10 individual experiments.
observed with the 1-min pulses (data not shown). The effect of continuous administration of 50 ng/ml of EGF was examined in superfusion (Fig. 2). Results show that after a significant increase the first few minutes, pulsatile hCG secretion decreased to a level below control. This inhibition continued throughout the remainder of superfusion, however, the response to 1-min pulse of EGF was maintained after continuous EGF administration was stopped. However, this response of EGF was lower than the one observed without the continuous administration of EGF. Explants in static short term cultures Data presented in Fig. 3 is mean ± SEM of a total of eight individual experiments. Overall in the two doses
LJ CL O O
100--
0 FlG. 3. The stimulatory effect of EGF upon hCG secretion by placental explants incubated for 24 h. *, P < 0.05. Data represents mean ± SEM of a total of 8 individual placentas with 12 replicates each per test compound or vehicle-treated controls.
tested, 50 and 100 ng/ml (final concentration), EGF caused a significant 157% and 169% increase, respectively, in hCG secretion as compared to control (P < 0.05). Isolated cells in long-term cultures a, First culture method (10). Data represented in Fig. 4 are mean ± SEM of a total of 12 individual experiments. Overall, EGF stimulated hCG secretion by isolated trophoblastic cells (P < 0.05). This was noted in both 100 and 200 ng/ml EGF. In all cultures hCG secretion decreased in the media. However, the addition of EGF prevented this decline in a significant manner, P < 0.05. This stimulatory effect of the growth factor was noted after 24 h of culture, and remained significant until the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 16:11 For personal use only. No other uses without permission. . All rights reserved.
BARNEA£TAL.
926
JCE & M • 1990 Vol 7, • No 4
i ou -
< X (J O
x x
I
10075-
t *
xxx;
*
250-
1
LI
X X X X
3 xX
I
T I
X
X X
X X
X X X X X X X X
x
* *
X
X
X
xX X X X X
0
:xxx:
50-
O O
IQOQ)Lbh z U U n g / m l
*
X X X
K X
cr LJ Q_
CD CONTROL
BHEGF 100ng/ml
MXXX
LLJ
125-
*
f
xxxx:
FIG. 4. The stimulatory effect of EGF upon hCG secretion by ITC (isolated trophoblastic cells) in long-term culture. Compared to control, there was a significant increase in hCG secretion within 24 h of the addition of EGF. This stimulation was maintained when EGF was added daily until the fifth day. *,P < 0.05. Data represent results mean ± SEM of 12 different placentas, each run with 12 replicates per test compound or vehicle-treated controls.
LJ O
X X X X X
X
S
T
x
T
x
TIME (DAYS)
fifth day. Thereafter, hCG levels decreased in both cases and became similar on the seventh day. No significant differences in hCG secretion were noted between 100 and 200 ng/ml EGF concentrations. hCG remained detectable in culture for up to 14 days. Second culture method (12). Figure 5 shows ABC-immunoperoxidase staining of cultured cells. Over 98% of cells were cytokeratin positive with minimal fibroblast and macrophage contamination. Unlike term placenta (12), a few preparations contained higher percentages (1015%) of macrophages and fibroblasts. These were inadequate preparations and were discarded. The mean hCG response to EGF of six separate preparations is shown in Fig. 6. EGF-stimulated cells demonstrated higher hCG secretion in each experiment as compared to controls (P < 0.05) by t test and ANOVA of area under the curves. The increase in integrated hCG secretion induced by EGF in all experiments was 189 ± 29% (mean ± SEM). Effect of EGF upon cellular differentiation In these experiments the effect of EGF upon the differentiation of isolated cells using one of the culture methods (10) was examined over a period of 3 months. This was carried out through morphological examination by light microscopy and photography of trypan blue stained cells at the beginning of culture and at subsequent frequent intervals throughout the experiment (data not shown). The rate of differentiation into syncytiotrophoblast did not increase in the EGF-tested cultures, in spite of the significant increase noted in hCG secretion found in the second day of culture.
Discussion In this study, the use of different culture systems has enabled us to demonstrate a dual effect of EGF upon
hCG secretion: an acute stimulation of secretion, and a delayed effect upon biosynthesis. In superfused explants, within 10 min of administration, short pulses of EGF stimulated spontaneous pulsatile hCG secretion, increasing both the pulse amplitude and frequency, indicating that EGF has effects upon rapidly released stored hCG. In static cultures using two methods, stimulatory effect of EGF were noted by 24 h, the time of first medium collection. The persistence of increased hCG secretion in static cultures for up to 9 days in EGF-treated cells (Figs. 3, 6) implies new biosynthesis, as culture medium was changed daily. Tissue stores would be unlikely to be adequate to supply this amount of hormone, as previous work has shown that 85% of newly synthesized hCG is secreted within 4 h (14), and that there are only limited intracellular hCG storage granules (15). In long-term studies of morphology using one of the static dispersed cell culture systems (10), no morphologic change was noted in association with the EGF-induced hCG secretion, suggesting that this effect is not necessarily coupled with enhanced differentiation. In this cell system addition of EGF also did not change long-term hCG secretion and cellular viability. In both treated and untreated cell cultures, after 14 days, hCG was not detectable in the media. In these cultures, however, the trophoblast remained viable even after 2 weeks as evidenced by the continued secretion of estradiol, progesterone, and pregnancy associated plasma protein A into media (Barnea, E. R. and P. Bischof, unpublished observations). In both static dispersed cell systems, approximately the same magnitude (180%) of hCG increase after EGF stimulation was noted. Few other studies have examined the effects of EGF in first trimester trophoblast. Maruo et al. (3) found that EGF stimulated hCG and free ahCG secretion 24-48 h
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 16:11 For personal use only. No other uses without permission. . All rights reserved.
ROLE OF EGF IN REGULATION OF PLACENTAL FUNCTION
927
FlG. 5. Immunoperoxidase-ABC staining of first trimester trophoblast. a, Positive cytoplasmic staining for cytokeratin. X1250; b, negative staining for vimentin, X800; c, negative staining for macrophages, X1250; d, negative control, X 1250.
O — 0 Control 1600-
• — «EGF
1200.1
•
600-
•
l/
1/
400 • ~> J
0
1
r
Vol. 71, No. 4 Printed in U.S.A.
The Dual Effect of Epidermal Growth Factor upon Human Chorionic Gonadotropin Secretion by the First Trimester Placenta in Vitro EYTAN R. BARNEA, DORON FELDMAN, MARIELLE KAPLAN, AND DONALD W. MORRISH Feto-Placental Endocrinology Unit (E.R.B., D.F., M.K.), Rappaport Institute for Medical Research, and Department Ob/Gyn, Reproductive Endocrinology, Rambam Medical Center, Technion, Haifa, Israel; and Department of Medicine (D.W.M.), University of Alberta, Edmonton, Alberta, Canada T6G 2G3
ABSTRACT. Epidermal growth factor (EGF) is believed to play an important role in the regulation of placental function. We have examined the effect of EGF upon first trimester (7-10 gestational weeks) placental hCG secretion and cellular differentiation using both static (explants and isolated cells) and kinetic (superfusion of explants) culture methods. In superfused explants, short (1-4 min) pulses of EGF increased both the rate and amplitude of spontaneous pulsatility of hCG. The frequency increased from 3/h to 5/h, and the amplitude increased compared to the control channels as calculated by the area under the curve. This effect was dose dependent and the concentration of 50 ng/ml, which was the lowest dose tested, was the most
effective. In explants cultured for 24 h, EGF caused a 2-fold increase in hCG secretion, compared to control, P < 0.05. In two different dispersed trophoblastic cell cultures, EGF added daily for the first week caused a 180% increase in hCG secretion, P < 0.05. However, according to morphological criteria, i.e. light microscopy and vital staining, no significant effect upon the rate of differentiation to syncytiotrophoblast was observed in longterm cultures of one of these preparations. In conclusion, EGF plays a dual trophic role in stimulating hCG secretion in the first trimester. However, this effect is not dependent on cellular differentiation. (J Clin Endocrinol Metab 71: 923-928, 1990)
CG secretion follows a unique pattern after implantation: It increases exponentially until 5-6 weeks, slowly merges into a plateau phase at 8-10 weeks, gradually decreases at 11-13 weeks, and is subsequently maintained at a stable level until term (1). Despite extensive investigations, the factors involved in regulation of hCG secretion remain unknown. The vast majority of research conducted on the regulation of placental hCG was conducted at term, while studies done in the first trimester are limited. The role of epidermal growth factor (EGF) in the placenta was previously addressed (2-4). It leads to phosphorylation of membrane proteins (5) and acts as a mitogen (6) by increasing RNA and DNA synthesis. Current data are conflicting as to the effect of EGF upon hCG secretion: EGF was shown to increase hCG secretion at term (2), whereas others found either increase or no change in the first trimester (7). The effect of EGF upon trophoblastic differentiation in the first trimester is not known.
Most studies on hCG secretion used static culture techniques in which measurements can only be carried out after relatively long periods of time. Recently, a new method of placental superfusion was developed which allows real-time measurements of hCG secretion and provides a more physiologic approach to the study of hCG regulation. Using this model we found that the secretion of hCG is pulsatile (8). In the present study we have compared the effects of EGF upon first trimester placental hCG secretion, using both static and dynamic culture methods. Moreover, the effect of EGF upon cellular differentiation was examined.
Received December 14, 1989. Address correspondence and requests for reprints to: Eytan Barnea, M.D., F.A.C.O.G., Feto-Placental Endocrinology Unit, Rappaport Institute, Technion, P.O. Box 9697, Efron Street, Haifa Israel.
Placentas
h
Materials and Methods Chemicals EGF, trypsin, DNAase, HEPES, and soybean trypsin inhibitor were purchased from Sigma (St. Louis, MO). Dulbecco's modified Eagle's medium (DMEM) and fetal calf serum (FCS) were purchased from Beit Haemek Laboratories (Beit Haemek, Israel). MAIAclone kits for measurements of hCG were obtained from Serono (Rehovot, Israel). All other chemicals were of high analytical quality.
After obtaining appropriate consent, elective pregnancy terminations were carried out in the first trimester (7-10 weeks). 923
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 16:11 For personal use only. No other uses without permission. . All rights reserved.
924
BARNEA ETAL.
Patients were on no medication and were nonsmokers. Tissues collected were immediately placed on ice in sterile 0.9% NaCl solution. Explant preparation: The method was previously published (9). Briefly, explants 50-70 mg wet weight were dissected out and rinsed in 1% antibiotic solutions (penicillin, 10,000 U/ml; streptomycin, 10 ng/m\; amphotericin B, 25 ^g/ml). For culture, explants were placed in DMEM media with or without EGF. Twelve replicates per test agent or control were plated at 37° C in an atmosphere of 95% air and 5% CO2. After 24 h of incubation, the media were collected and stored at —20 C until hCG assay. The tissue was saved for protein analysis. In the preliminary experiments we found that hCG secretion is linear up to 48 h. Therefore in subsequent experiments, EGF effect was tested until 24 h at the half-maximal increase point. Cell culture Two methods were used as previously published with some modifications (10-12). In the first, after rinsing in cold 0.9% NaCl solution, followed by multiple rinsing in DMEM, the membranes were trimmed and the tissue was incubated with 0.25% trypsin-EDTA solution containing DNAse 200 Kunitz U/ml for 15 min with gentle agitation at 37° C. At the end of the incubation period, the cells were filtered through a sterile gauze. Soybean trypsin inhibitor (Sigma, St. Louis, MO) was added to the recovered cell suspension to block trypsin activity. Then, 3 ml trophoblastic cell suspension were mixed with 7 ml DMEM containing 10% FCS. Cells were separated by centrifugation at 500 X g X 10 min at room temperature and the supernatant was discarded. For culture, cells were resuspended in media composed of DMEM, 5% FCS and 1% garamycin 10 /ig/ml. Cells were counted by a hemocytometer and in each dish approximately 100,000 cells were plated and incubated at 37 C in an atmosphere of 95% air and 5% CO2. EGF at concentrations of 50-200 ng/ml was added after 24 h of culture, and the media were collected 48 h after the start of the experiment. Subsequently EGF was added and the media were collected daily for 7 days. Thereafter the media were collected every 3 days until the culture was terminated after 2-3 months. The media were stored at —20 C until assay. In the second method (11, 12) intact first trimester placenta without membranes was incubated for 8- to 10-min periods in 0.25% trypsin-10 n\/m\ DNAse I at 37° C in bicarbonate buffer. Cells were collected from the fourth to the eighth trypsinization, pooled, centrifuged at 200 x g for 7 min, and plated on 60-mm Petri dishes (Corning, Corning, NY) at 1-2 X 106 cells per dish in 10% FCS-DMEM. After 2 h, medium was changed to serum-free DMEM. From days 1-9 10 ng/ml EGF (Sigma) were added to test plates and 0.001% BSA (final concentration) to controls. Medium was collected daily (days 1-7) and frozen at -20 C until assay. Cytospins of dispersed cells were stained by the immunoperoxidase-avidin-biotin complex method as previously described (12) for vimentin, a macrophage monocyte marker (8243), and cytokeratin (Becton-Dickinson). The details of the histochemical studies were reported previously (10). Studies of differentiation were carried out using cultures prepared by the first method (10). Briefly, at different times during culture, the media were removed from control and EGF-treated dishes and the cells were fixed. Subsequently, they
JCE & M • 1990 Vol 7, • No 4
were stained with trypan blue and observed at light microscopy for morphological changes using previously established criteria (11). Stained dishes were blinded and analyzed for at least three microscopical fields in three different dishes from each control and EGF-treated cells. Superfusion culture The method of placental superfusion was recently reported (8). Briefly, a superfusion apparatus (Accusyst, Endotronics, St. Paul, MN) with a multichannel peristaltic pump and fraction collector (model 272, ISCO, Durham, NC) were used to study the short-term dynamics of hCG secretion. The explants (200-300 mg wet wt) were placed into the culture chambers and a HEPES (18 mM) DMEM solution was washed through in an atmosphere of 5% CO2 and 95% air at 37° C. Experiments were conducted for a 120-min period; a 1-ml sample from the effluent was collected every 2.4 min for hCG measurements. In each experiment, one channel served as control and four served as experimental channels. At given intervals, a 1-4-min pulse of EGF was given through a peristaltic pump equipped with a digital flowmeter (ISMATECH DD, Chicago, IL). Assay /3-hCG assay was carried out with a commercially available RIA kit with an intraassay variability of 1.7% (MAIA CLONE, Serono Diagnostics) or by intact hCG RIA as described (11, 12). All measurements were carried out in the same assay. Protein content was determined according to the Lowry protein assay procedure (13). Statistics Statistical analysis was performed using one way analysis of variance (ANOVA) and Student's t test, and P < 0.05 was considered as statistically significant. In superfusion experiments, the area under a graph was determined using Simpson's rule in a personal computer. This program is able to calculate a function that best fit the series of points we are giving, and then to determine the area under the curve using the integral of the function. Data of superfusion experiments represents a single channel run in duplicate in four different placentas yielding similar results. Data in explants and cells shows mean and SEM of 612 individual placentas. In each placenta 6-12 dishes were run.
Results Effect of EGF on hCG secretion Explants in superfusion . Data presented in Fig. 1, A and B, are representative of 10 different experiments. One minute pulses of EGF stimulated spontaneous pulsatile hCG secretion; the pulse frequency increased from 3/h to 5/h, and pulse amplitude increased as calculated by the area under the curve. The maximal effect was noted at 50 ng/ml EGF. At 100 ng/ml only a 2-fold increase in the amplitude was noted. Longer pulses of EGF (up to 4 min) did not have a stimulatory effect greater than that
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 16:11 For personal use only. No other uses without permission. . All rights reserved.
ROLE OF EGF IN REGULATION OF PLACENTAL FUNCTION
250
o CONTROL • TREATMENT II : 1 min pulse of EGF 50 n g / m l
A — A CONTROL • — • EGF 50ng/ml
1
z
925
200
o 150,
a: a.
EGF 50 n g / m l
50-
i
o a
30:. 10:. 0
30
180
TIME (min.)
160
minutes 50 &—a .-.
4 0 • •
CONTROL
o — o EGF 100ng/ml
o
eg CL
FIG. 2. Effect of continuous administration of EGF 50 ng/ml for 90 minutes on placental explants in superfusion system. There was a significant decrease after a first increase. After cessation of continuous administration 1-min pulse of EGF 50 ng/ml had only a mild stimulatory effect.
30-LJ O
20-E o o
300 •-
CD CONTROL C3DEGF 100ng/ml EfflEGF 200ng/ml
200 ••
o en
a>
TIME'(min.)
_
V)
_
a
a
a
3 Q.
FIG. 1. The stimulatory effect of 1-min pulses of 50 ng/ml (A) and 100 ng/ml (B) EGF upon spontaneous pulsatile hCG secretion by placental explants in superfusion. The 50 ng/ml concentration was more effective in stimulating hCG secretion as calculated by the area under the curve (see Materials and Methods section for further detail). Data represent results obtained from 4 different placentas in a total of 10 individual experiments.
observed with the 1-min pulses (data not shown). The effect of continuous administration of 50 ng/ml of EGF was examined in superfusion (Fig. 2). Results show that after a significant increase the first few minutes, pulsatile hCG secretion decreased to a level below control. This inhibition continued throughout the remainder of superfusion, however, the response to 1-min pulse of EGF was maintained after continuous EGF administration was stopped. However, this response of EGF was lower than the one observed without the continuous administration of EGF. Explants in static short term cultures Data presented in Fig. 3 is mean ± SEM of a total of eight individual experiments. Overall in the two doses
LJ CL O O
100--
0 FlG. 3. The stimulatory effect of EGF upon hCG secretion by placental explants incubated for 24 h. *, P < 0.05. Data represents mean ± SEM of a total of 8 individual placentas with 12 replicates each per test compound or vehicle-treated controls.
tested, 50 and 100 ng/ml (final concentration), EGF caused a significant 157% and 169% increase, respectively, in hCG secretion as compared to control (P < 0.05). Isolated cells in long-term cultures a, First culture method (10). Data represented in Fig. 4 are mean ± SEM of a total of 12 individual experiments. Overall, EGF stimulated hCG secretion by isolated trophoblastic cells (P < 0.05). This was noted in both 100 and 200 ng/ml EGF. In all cultures hCG secretion decreased in the media. However, the addition of EGF prevented this decline in a significant manner, P < 0.05. This stimulatory effect of the growth factor was noted after 24 h of culture, and remained significant until the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 16:11 For personal use only. No other uses without permission. . All rights reserved.
BARNEA£TAL.
926
JCE & M • 1990 Vol 7, • No 4
i ou -
< X (J O
x x
I
10075-
t *
xxx;
*
250-
1
LI
X X X X
3 xX
I
T I
X
X X
X X
X X X X X X X X
x
* *
X
X
X
xX X X X X
0
:xxx:
50-
O O
IQOQ)Lbh z U U n g / m l
*
X X X
K X
cr LJ Q_
CD CONTROL
BHEGF 100ng/ml
MXXX
LLJ
125-
*
f
xxxx:
FIG. 4. The stimulatory effect of EGF upon hCG secretion by ITC (isolated trophoblastic cells) in long-term culture. Compared to control, there was a significant increase in hCG secretion within 24 h of the addition of EGF. This stimulation was maintained when EGF was added daily until the fifth day. *,P < 0.05. Data represent results mean ± SEM of 12 different placentas, each run with 12 replicates per test compound or vehicle-treated controls.
LJ O
X X X X X
X
S
T
x
T
x
TIME (DAYS)
fifth day. Thereafter, hCG levels decreased in both cases and became similar on the seventh day. No significant differences in hCG secretion were noted between 100 and 200 ng/ml EGF concentrations. hCG remained detectable in culture for up to 14 days. Second culture method (12). Figure 5 shows ABC-immunoperoxidase staining of cultured cells. Over 98% of cells were cytokeratin positive with minimal fibroblast and macrophage contamination. Unlike term placenta (12), a few preparations contained higher percentages (1015%) of macrophages and fibroblasts. These were inadequate preparations and were discarded. The mean hCG response to EGF of six separate preparations is shown in Fig. 6. EGF-stimulated cells demonstrated higher hCG secretion in each experiment as compared to controls (P < 0.05) by t test and ANOVA of area under the curves. The increase in integrated hCG secretion induced by EGF in all experiments was 189 ± 29% (mean ± SEM). Effect of EGF upon cellular differentiation In these experiments the effect of EGF upon the differentiation of isolated cells using one of the culture methods (10) was examined over a period of 3 months. This was carried out through morphological examination by light microscopy and photography of trypan blue stained cells at the beginning of culture and at subsequent frequent intervals throughout the experiment (data not shown). The rate of differentiation into syncytiotrophoblast did not increase in the EGF-tested cultures, in spite of the significant increase noted in hCG secretion found in the second day of culture.
Discussion In this study, the use of different culture systems has enabled us to demonstrate a dual effect of EGF upon
hCG secretion: an acute stimulation of secretion, and a delayed effect upon biosynthesis. In superfused explants, within 10 min of administration, short pulses of EGF stimulated spontaneous pulsatile hCG secretion, increasing both the pulse amplitude and frequency, indicating that EGF has effects upon rapidly released stored hCG. In static cultures using two methods, stimulatory effect of EGF were noted by 24 h, the time of first medium collection. The persistence of increased hCG secretion in static cultures for up to 9 days in EGF-treated cells (Figs. 3, 6) implies new biosynthesis, as culture medium was changed daily. Tissue stores would be unlikely to be adequate to supply this amount of hormone, as previous work has shown that 85% of newly synthesized hCG is secreted within 4 h (14), and that there are only limited intracellular hCG storage granules (15). In long-term studies of morphology using one of the static dispersed cell culture systems (10), no morphologic change was noted in association with the EGF-induced hCG secretion, suggesting that this effect is not necessarily coupled with enhanced differentiation. In this cell system addition of EGF also did not change long-term hCG secretion and cellular viability. In both treated and untreated cell cultures, after 14 days, hCG was not detectable in the media. In these cultures, however, the trophoblast remained viable even after 2 weeks as evidenced by the continued secretion of estradiol, progesterone, and pregnancy associated plasma protein A into media (Barnea, E. R. and P. Bischof, unpublished observations). In both static dispersed cell systems, approximately the same magnitude (180%) of hCG increase after EGF stimulation was noted. Few other studies have examined the effects of EGF in first trimester trophoblast. Maruo et al. (3) found that EGF stimulated hCG and free ahCG secretion 24-48 h
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 16:11 For personal use only. No other uses without permission. . All rights reserved.
ROLE OF EGF IN REGULATION OF PLACENTAL FUNCTION
927
FlG. 5. Immunoperoxidase-ABC staining of first trimester trophoblast. a, Positive cytoplasmic staining for cytokeratin. X1250; b, negative staining for vimentin, X800; c, negative staining for macrophages, X1250; d, negative control, X 1250.
O — 0 Control 1600-
• — «EGF
1200.1
•
600-
•
l/
1/
400 • ~> J
0
1
r
