0013-7227/90/1265-2343$02.00/0 Endocrinology Copyright© 1990 by The Endocrine Society
Vol. 126, No. 5 Printed in U.S.A.
Effects of Oxytocin on in Vitro Steroid Release of Midstage Small and Large Porcine Luteal Cells* LUTZ PITZEL, HUBERTUS JARRY, AND WOLFGANG WUTTKE Division of Clinical and Experimental Endocrinology, Department of Obstetrics/Gynecology, University of Gottingen, 3400 Gottingen, West Germany
ABSTRACT. Previously, we have demonstrated an inhibitory effect of oxytocin (OXT) on progesterone (P) and androstenedione (A) release of porcine luteal cell cultures. The present study examines whether OXT modulates P, A, or estradiol (E2) release of so-called small luteal cells (SLC) or of granulosaderived large luteal cells (LLC). To ensure clean Percoll-gradient separation of the 2 cell types, corpora lutea not older than 6 days were used. SLC, but not LLC, responded to human (h)CG (6 ng/ml) with increased P and A, but not E2, release. When OXT was added to the culture system, both basal as well as hCG-stimulated P release of SLC, but not of LLC, were dose dependently reduced. In contrast, E2 production of SLC and LLC was significantly stimulated by OXT whereas A release of SLC cultures, but not of LLC, was inhibited in response to OXT. In the presence of a specific OXT-antagonist, this inhibitory
effect of OXT on P release was abolished, indicating a specific receptor-mediated effect of OXT on porcine luteal cells. When E2 was added to the culture medium, a dose-dependent stimulatory effect on P release of SLC was demonstrated. The presence of the E2 receptor antagonist monohydroxy-tamoxifen in the culture system prevented the E2-induced increase of P release of SLC. E2 was able to counteract dose dependently the OXT-induced inhibition of P release in SLC cultures. These results suggest that OXT may have a dual function in young corpora lutea. The reduction of P and A production can be interpreted as a luteolytic effect of OXT. The simultaneous increase of E2 production, however, may also point to an indirect luteotropic effect since E2 was shown to stimulate luteal P release and to counteract OXT-induced inhibition of P release excessively. (Endocrinology 126: 2343-2349, 1990)
T
HE CORPORA lutea (CL) of mammalian species investigated so far consist of two different cell types with steroidogenic capacity (1-6). After ovulation follicular granulosa cells are believed to develop into so-called large luteal cells (LLC) which produce progesterone (P) and also a number of regulatory peptides including oxytocin (OXT) (7-10). In ovine corpora lutea (CL) these cells have few, if any, LH receptors (6). Hence, their P release is not modulated by LH upon luteinization. Follicular theca cells seem to form the so-called small luteal cells (SLC) which do not produce regulatory peptides but contain numerous LH receptors and react with increased P release upon LH stimulation (1, 2, 5, 6,11-13). Among the many regulatory peptides produced by large luteal cells, OXT is the best explored (14-22). Recently we demonstrated a dose-dependent inhibitory effect of OXT on in vitro P and androstenedione (A) release by porcine luteal cells obtained from young corpora lutea (23); whereas in vivo experiments in the same species point to a stimulatory effect of OXT on luteal P secretion (24).
Therefore, the aim of the present investigation was to clarify these apparently discrepant results. Under the assumption that OXT is produced by LLC we wanted to determine whether OXT acts on LLC or SLC in an autoor paracrine fashion on steroid release. The effect of OXT on luteal estrogen production has not been thoroughly studied. Estradiol (E2) production by porcine luteal cells is well established, and there is some evidence that it has luteotropic effects (25-27). Recently we published data indicating that OXT may stimulate in vitro E2 production by a mixed luteal cell culture although these data were statistically not significant (23). It is not known whether OXT affects E2 production in one of the two or in both steroidogenic cell types. There are indications that SLC may enlarge as the corpus luteum ages (28). Therefore, separation of SLC and LLC can best be achieved in relatively young CL. The present experiments were therefore performed with porcine SLC and LLC obtained from young CL.
Received October 6, 1989. Address requests for reprints to: Dr. L. Pitzel, Universitats-Frauenklinik, Division of Clinical and Experimental Endocrinology, RobertKoch-Strasse 40, D-3400 Gottingen, West Germany. * This work was supported by the Deutsche Forschungsgemeinschaft (Pi 116/2-4).
Materials and Methods For the present experiments only CL in the early phase of the midstage of estrous cycle were used. The cycle stage was assessed by the morphological appearance of CL (29). Luteal cells were isolated from ovaries of nonpregnant pigs as de-
2343
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 May 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
2344
OXYTOCIN EFFECT ON LUTEAL STEROID RELEASE
scribed earlier (23). The CL were dispersed by three successive incubation steps of 15 min at 37 C in Ham's F12 medium, containing 15 mM HEPES, 0.2% BSA, 0.2% collagenase (Worthington, Freehold, NJ) and 0.005% DNAse (Sigma, Munich, Germany). After collagenase treatment, dispersed luteal cells were filtered through a nylon gauze (70 /an) to remove undigested tissue debris. The harvested cells were pooled and, after washing in Ham's F12, layered on Percoll gradient (density, 1.07 g/ml in 0.15 M NaCl) to remove red blood cells. The enriched luteal cell population was subsequently washed and after centrifugation aspirated in 5 ml Ham's F12 medium, filtered to remove cell aggregates. Although this procedure resulted in a considerable loss of both cell types, no change of cell ratio between SLC and LLC was observed. Percoll solutions at concentrations of 40%, 20%, 15%, 10%, and 5% in 0.9% NaCl were sequentially overlayered, and the cell suspension was placed on this discontinuous gradient. The tubes were then centrifuged at 400 x g for 20 sec. Cells recovered from the interphases of 15% and 20% were combined and designated the SLC fraction while LLC were obtained from the 40% Percoll phase. The purity of each cell fraction was assessed by phase contrast microscope with an ocular micrometer. Small luteal cells were 10-20 iim in diameter whereas large luteal cells had diameters between 20-35 /an. The large cell fraction contained 8-15% (range of six independent cell preparations) SLC, whereas the SLC fraction was contaminated between 5-18% by large cells under the same experimental conditions. The ratio of SLC to LLC recovered after separation was around 5:1. Small and large cell fractions were washed twice with Ham's F12 and resuspended in culture medium after counting and determination of cell viability by trypan blue exclusion. Viable SLC and LLC were aliquoted such that each milliliter of culture medium contained 2 X 105 cells. Culture medium 199 (Boehringer, Mannheim, Germany) containing 14 mM HEPES, 17 mM NaHCO3, and 0.2% BSA was supplemented with 10% fetal calf serum (Boehringer), streptomycin (100 /ig/ml), gentamycin (50 Mg/ml), insulin (15.6 IU/liter), and dexamethasone (100 nM). One milliliter of cell suspension was added to 1 ml culture medium in 6 well plastic culture dishes (Costar, Tecnomara, Fernwald, Germany). The cell system was incubated at 37 C in a 5% CO2/air atmosphere; the medium was changed twice within the first 48 h, and steroids were measured after an additional 24 h of culture time. No differences in the viability of cultured luteal cells could be observed within 0-96 h of culture time. Substances to be tested were purchased from the following suppliers: hCG (Primogonyl 5000, Schering, Berlin, Germany); OXT (Bachem, Bubendorf, Switzerland); the OXT antagonist [(CH^6Tyr(Me)OVT], [l-(/S-mercapto-j8, 0-cyclopentamethylene propionic acid)2-(0-methyl)-tyrosine,8-ornithine] vasotocin (Peninsula Laboratories, Inc., London, U.K.); E2 (Serva, Heidelberg, Germany); monohydroxy-tamoxifen was a gift of Professor P. Jungblut (MPI f. Exp. Endokrinologie, Hannover, Germany). Stock solutions of tested substances were dissolved in culture medium and added when the cultures were set up and also each time the medium was changed. The following experiments were performed: I) The effect of
Endo • 1990 Voll26-No5
6 ng hCG/ml (5 mUI/ml) on P, A, and E2 production of SLC and LLC was tested. Ila) OXT (10~10 to 10~5 M) in combination with hCG or without hCG was added to the SLC and LLC cultures and the effect on P release was tested. lib) OXT (10~7 M) and the OXT-antagonist (2 x 10~6 M) were added to the SLC cultures. The effects of these substances and their combination on P production of SLC was tested. lie) The influence of OXT (10~7 M) on A and E2 release of SLC and LLC was examined. Ilia) The effect of E2 (10~7 to 10"5 M) on SLC and LLC progesterone production was tested in the presence and absence of hCG. Illb) The effect of the E2-antagonist monohydroxy-tamoxifen (10~8 M) on P release of SLC cultures was examined. IIIc) The influence of different E2 concentrations (10~7 to 10"5 M) on OXT-induced P inhibition in SLC cultures was tested. P was determined without extraction as described earlier (23). A and E2 were extracted by Sep-Pak C18 columns (Waters Associates, Milford, MA) from the culture medium. After absorption of steroids the columns were washed with 5 ml bidistilled water containing 0.1% acetic acid. Steroids were eluted with 2 ml methanol which was subsequently evaporated. The residue was dissolved in 200 /il hormone-free serum. The recovery for both hormones was greater than 95%, and no influence of the extraction procedure on the E2 (Radioassay Systems Laboratories, Carson, CA) and A (Diagnostic Systems Laboratories, Webster, TX) assay systems was found. Unless indicated otherwise, all data points are expressed as percent related means ± SEM of individual cultures. In each case, however, the 100% hormone concentrations are also given in absolute values. For each culture, parallel control incubations which represent the 100% values were performed. All data stem from at least three different experiments, each with triplicate cultures. To assess the statistical significance of treatment effects, analysis of variance followed by Duncan's multiple range test was employed.
Results Influence of hCG on P, A, and E2 release by SLC and LLC
Figure 1 shows absolute hormone concentrations under control conditions and the mean percent-related data of experiments under hCG-stimulated conditions. While hCG stimulates P and A but not E2 production significantly in SLC, no effect of hCG was observed in LLC. Effect of OXT on basal and hCG-stimulated P, A, and E2 release by SLC and LLC Figure 2 details that OXT (10~10 to 10~5 M) inhibits P release from SLC dose dependently under basal as well as hCG-stimulated conditions. The inhibitory effect became significant (P < 0.05) at a dose of 10~8 M OXT. Concentration needed for inhibiting half-maximal P release (ED50) in response to OXT was 10~7 M. Figure 3 details that OXT had no significant effect on P release from SLC during the initial 48 h incubation period. During the following 24 h (between 48 h and 72
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 May 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
OXYTOCIN EFFECT ON LUTEAL STEROID RELEASE
SLC
LLC
360 300 -
100%"
7«0n«P/w«H
Influence of E2 on P release by luteal cell cultures
200 160 100 60 -
0
360100%-
300 -
1.6n»A/W«H
100%-7.2ngA/w«H
260 200 160O
stimulated P release of small cells was abolished, while the antagonist itself did not change the hormone release.
100%-3766 ngP/w*N
260
A O * Q.
2345
10060
E2 has a dose-dependent stimulatory effect on P release of porcine SLC (Table 2). At a concentration of 10~7 M it increased P release about 2-fold, this is similar to the effect of 6 ng/ml hCG. An E2 concentration of 10~5 M was significantly more potent to stimulate P release than the hCG stimulus. This stimulatory effect of E2 on P release was also observed in hCG-stimulated SLC cultures. In LLC cultures E2 has no significant effect on P release. E2-induced P release can be prevented by coincubation of SLC with the E2- receptor antagonist monohydroxy-tamoxifen (10~8 M) (Table 2). In SLC cultures the addition of different E2 concentrations (10~7 to 10~5 M) can counteract OXT induced inhibition of P release excessively (Table 3).
0
Discussion £
360-
2
300
S
2s
g
200 -
100%-69pgE2/w*N
100%-72pgE2/w»H
°-
A
180
O
100
w
60
hCG(6ng/ml)
•
•
FIG. 1. In vitro effect of hCG on P, A, and E2 release by porcine SLC and LLC. Luteal cells were cultured for 72 h, and steroid release of the last 24 h of culture time was determined. hCG (6 ng/ml) was present during the entire incubation time. Data represent percent-related means ± SEM of at least three different experiments with triplicate cultures. The absolute concentrations representing the 100% values for P, A, and E2 are also given. *, P < 0.05 vs. basal.
h of culture time) and later (these data are not shown), both basal and hCG-stimulated P release was significantly reduced by OXT. Figure 4 shows that OXT (10~7 M) reduced not only the P but also the A release from SLC cultures under basal as well as hCG-stimulated conditions. In contrast to these results E2 release of SLC was significantly stimulated by OXT. No effect on OXT on P and A release of LLC was observed but as in SLC OXT also induced E2 release from these cells. In each case hCG was without effect on this OXT-stimulated E2 release. In order to test the specifity of this inhibitory effect of OXT on P production of SLC cultures, experiments with an OXT-antagonist were performed. Table 1 shows that in the presence of [d(CH2)5Tyr(Me)OVT] (2 x 10~6 M) the inhibitory effect of OXT on basal as well as hCG-
Using a modification of the method of Ohara et al. (30) for human luteal cells it is possible to obtain a useful separation of the two distinct steroidogenic luteal cell populations and to establish that porcine SLC and LLC have functional differences in response to hCG and to OXT. It is interesting to note that spontaneous as well as hCG stimulated P release from SLC decreases with increasing incubation time although hCG responsiveness of these cells increases during this time. This is in agreement with results from mixed luteal cell cultures in the porcine (23, 31) as well as in other species (32, 33). The reason for this spontaneous fall of P secretion in luteal cell cultures is unclear. Based on identical cell numbers per culture well, basal P secretion rates were estimated to be 5- to 7-fold higher in LLC than in SLC. This is in agreement with data published for ovine (34) and bovine (35) luteal cells and also for porcine luteal cells obtained from pregnant sows (1). Our finding that porcine SLC respond to hCG with significantly increased P release whereas LLC under the same experimental conditions respond not at all to hCG, is in agreement with results published for luteal cells of other species (34, 35). It is interesting and new that hCG acts also in SLC to stimulate A. Theoretically the availability of higher precursor concentrations for E2 synthesis should also increase E2 production. As this could not be observed it is possible that A is too far diluted in the culture medium to serve effectively as precursor for E2 synthesis. In earlier in vitro experiments we demonstrated that porcine luteal cells responded to OXT with a dosedependent inhibition of P release under basal as well as hCG-stimulated conditions (23). The present experi-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 May 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
2346
OXYTOCIN EFFECT ON LUTEAL STEROID RELEASE
Endo • 1990 Vol 126 • No 5
100
•o
100% SLC - 536 ng P/well 100% LLC - 2853 ng P/well
so FIG. 2. In vitro effect of OXT on P release by porcine SLC and LLC. Cells were cultured for 72 h, and P release of the last 24 h of culture time was determined. hCG (6 ng/ml) and OXT (1CT10 to 10"6 M) were present during the entire incubation time. Data represent percent-related means ± SEM of at least three experiments with triplicate cultures. OXT at 10~8 M and higher concentrations was significantly (P < 0.05) inhibitory to both basal and hCGstimulated P release.
70 -
o
s CO
•o so 40 30 20 10 -
A - LCC basal n • 6 * - SLC basal n - 6 - 9 • • SLC hCQ n • 6 - 9 -7
-10
log OXT (M) 360 -
incubation tint* 48 h 1OO%- 749ngP/woll
***
300 -
FIG. 3. In vitro effect of OXT on P release by porcine SLC at different incubation times. P release was measured between 24 to 48 h and 48 to 72 h of culture time. Data represent percent-related means ± SEM of three experiments with triplicate cultures. *, P < 0.05 vs. basal; • , P < 0.05 vs. hCG stimulated.
P4 0.06va.hCQ
Incubation tlma 72 h
•* ©
260
8
200
2
180
o c£ %#>
* P * 0.06 va. control •
C
o
n-9
10
100%-427ngP/w«ll
°~
Q.
60 -
hCG (6 ng/ml) OXT (100 nM) ments extend this observation and allow the conclusion that for this effect only SLC are the target cells for OXT. A concentration of 10~8 M OXT is effective to inhibit P release significantly under basal as well as hCG-stimulated conditions. Results from Fig. 3 indicate that OXT became most efficient to decrease P release of SLC after 48 h of incubation time. Therefore we chose to measure effects of OXT on steroidogenesis between 48 and 72 h of incubation time. SLC cultures responded to OXT also with a significant inhibition of basal and hCG-stimulated
A release. As OXT is produced by LLC (7-10, 36, 37) the inhibitory effect on P and A release of SLC represents a paracrine luteolytic effect. The specificity of the OXT-induced inhibitory effect on P release of SLC is demonstrated by the fact that coincubation of OXT and [d(CH2)5Tyr(Me)OVT], a specific OXT antagonist, completely abolished the decrease in P release. This is not due to a toxic effect as we showed previously by measurement of DNA concentrations that OXT influenced neither plating efficiency nor cell via-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 May 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
OXYTOCIN EFFECT ON LUTEAL STEROID RELEASE
§
360
•
300
•
260
75 •
200
o
100
£ 0.
60
O
360 -
J S
300 260
«
200
o
100
2
80
Vol. 126, No. 5 Printed in U.S.A.
Effects of Oxytocin on in Vitro Steroid Release of Midstage Small and Large Porcine Luteal Cells* LUTZ PITZEL, HUBERTUS JARRY, AND WOLFGANG WUTTKE Division of Clinical and Experimental Endocrinology, Department of Obstetrics/Gynecology, University of Gottingen, 3400 Gottingen, West Germany
ABSTRACT. Previously, we have demonstrated an inhibitory effect of oxytocin (OXT) on progesterone (P) and androstenedione (A) release of porcine luteal cell cultures. The present study examines whether OXT modulates P, A, or estradiol (E2) release of so-called small luteal cells (SLC) or of granulosaderived large luteal cells (LLC). To ensure clean Percoll-gradient separation of the 2 cell types, corpora lutea not older than 6 days were used. SLC, but not LLC, responded to human (h)CG (6 ng/ml) with increased P and A, but not E2, release. When OXT was added to the culture system, both basal as well as hCG-stimulated P release of SLC, but not of LLC, were dose dependently reduced. In contrast, E2 production of SLC and LLC was significantly stimulated by OXT whereas A release of SLC cultures, but not of LLC, was inhibited in response to OXT. In the presence of a specific OXT-antagonist, this inhibitory
effect of OXT on P release was abolished, indicating a specific receptor-mediated effect of OXT on porcine luteal cells. When E2 was added to the culture medium, a dose-dependent stimulatory effect on P release of SLC was demonstrated. The presence of the E2 receptor antagonist monohydroxy-tamoxifen in the culture system prevented the E2-induced increase of P release of SLC. E2 was able to counteract dose dependently the OXT-induced inhibition of P release in SLC cultures. These results suggest that OXT may have a dual function in young corpora lutea. The reduction of P and A production can be interpreted as a luteolytic effect of OXT. The simultaneous increase of E2 production, however, may also point to an indirect luteotropic effect since E2 was shown to stimulate luteal P release and to counteract OXT-induced inhibition of P release excessively. (Endocrinology 126: 2343-2349, 1990)
T
HE CORPORA lutea (CL) of mammalian species investigated so far consist of two different cell types with steroidogenic capacity (1-6). After ovulation follicular granulosa cells are believed to develop into so-called large luteal cells (LLC) which produce progesterone (P) and also a number of regulatory peptides including oxytocin (OXT) (7-10). In ovine corpora lutea (CL) these cells have few, if any, LH receptors (6). Hence, their P release is not modulated by LH upon luteinization. Follicular theca cells seem to form the so-called small luteal cells (SLC) which do not produce regulatory peptides but contain numerous LH receptors and react with increased P release upon LH stimulation (1, 2, 5, 6,11-13). Among the many regulatory peptides produced by large luteal cells, OXT is the best explored (14-22). Recently we demonstrated a dose-dependent inhibitory effect of OXT on in vitro P and androstenedione (A) release by porcine luteal cells obtained from young corpora lutea (23); whereas in vivo experiments in the same species point to a stimulatory effect of OXT on luteal P secretion (24).
Therefore, the aim of the present investigation was to clarify these apparently discrepant results. Under the assumption that OXT is produced by LLC we wanted to determine whether OXT acts on LLC or SLC in an autoor paracrine fashion on steroid release. The effect of OXT on luteal estrogen production has not been thoroughly studied. Estradiol (E2) production by porcine luteal cells is well established, and there is some evidence that it has luteotropic effects (25-27). Recently we published data indicating that OXT may stimulate in vitro E2 production by a mixed luteal cell culture although these data were statistically not significant (23). It is not known whether OXT affects E2 production in one of the two or in both steroidogenic cell types. There are indications that SLC may enlarge as the corpus luteum ages (28). Therefore, separation of SLC and LLC can best be achieved in relatively young CL. The present experiments were therefore performed with porcine SLC and LLC obtained from young CL.
Received October 6, 1989. Address requests for reprints to: Dr. L. Pitzel, Universitats-Frauenklinik, Division of Clinical and Experimental Endocrinology, RobertKoch-Strasse 40, D-3400 Gottingen, West Germany. * This work was supported by the Deutsche Forschungsgemeinschaft (Pi 116/2-4).
Materials and Methods For the present experiments only CL in the early phase of the midstage of estrous cycle were used. The cycle stage was assessed by the morphological appearance of CL (29). Luteal cells were isolated from ovaries of nonpregnant pigs as de-
2343
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 May 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
2344
OXYTOCIN EFFECT ON LUTEAL STEROID RELEASE
scribed earlier (23). The CL were dispersed by three successive incubation steps of 15 min at 37 C in Ham's F12 medium, containing 15 mM HEPES, 0.2% BSA, 0.2% collagenase (Worthington, Freehold, NJ) and 0.005% DNAse (Sigma, Munich, Germany). After collagenase treatment, dispersed luteal cells were filtered through a nylon gauze (70 /an) to remove undigested tissue debris. The harvested cells were pooled and, after washing in Ham's F12, layered on Percoll gradient (density, 1.07 g/ml in 0.15 M NaCl) to remove red blood cells. The enriched luteal cell population was subsequently washed and after centrifugation aspirated in 5 ml Ham's F12 medium, filtered to remove cell aggregates. Although this procedure resulted in a considerable loss of both cell types, no change of cell ratio between SLC and LLC was observed. Percoll solutions at concentrations of 40%, 20%, 15%, 10%, and 5% in 0.9% NaCl were sequentially overlayered, and the cell suspension was placed on this discontinuous gradient. The tubes were then centrifuged at 400 x g for 20 sec. Cells recovered from the interphases of 15% and 20% were combined and designated the SLC fraction while LLC were obtained from the 40% Percoll phase. The purity of each cell fraction was assessed by phase contrast microscope with an ocular micrometer. Small luteal cells were 10-20 iim in diameter whereas large luteal cells had diameters between 20-35 /an. The large cell fraction contained 8-15% (range of six independent cell preparations) SLC, whereas the SLC fraction was contaminated between 5-18% by large cells under the same experimental conditions. The ratio of SLC to LLC recovered after separation was around 5:1. Small and large cell fractions were washed twice with Ham's F12 and resuspended in culture medium after counting and determination of cell viability by trypan blue exclusion. Viable SLC and LLC were aliquoted such that each milliliter of culture medium contained 2 X 105 cells. Culture medium 199 (Boehringer, Mannheim, Germany) containing 14 mM HEPES, 17 mM NaHCO3, and 0.2% BSA was supplemented with 10% fetal calf serum (Boehringer), streptomycin (100 /ig/ml), gentamycin (50 Mg/ml), insulin (15.6 IU/liter), and dexamethasone (100 nM). One milliliter of cell suspension was added to 1 ml culture medium in 6 well plastic culture dishes (Costar, Tecnomara, Fernwald, Germany). The cell system was incubated at 37 C in a 5% CO2/air atmosphere; the medium was changed twice within the first 48 h, and steroids were measured after an additional 24 h of culture time. No differences in the viability of cultured luteal cells could be observed within 0-96 h of culture time. Substances to be tested were purchased from the following suppliers: hCG (Primogonyl 5000, Schering, Berlin, Germany); OXT (Bachem, Bubendorf, Switzerland); the OXT antagonist [(CH^6Tyr(Me)OVT], [l-(/S-mercapto-j8, 0-cyclopentamethylene propionic acid)2-(0-methyl)-tyrosine,8-ornithine] vasotocin (Peninsula Laboratories, Inc., London, U.K.); E2 (Serva, Heidelberg, Germany); monohydroxy-tamoxifen was a gift of Professor P. Jungblut (MPI f. Exp. Endokrinologie, Hannover, Germany). Stock solutions of tested substances were dissolved in culture medium and added when the cultures were set up and also each time the medium was changed. The following experiments were performed: I) The effect of
Endo • 1990 Voll26-No5
6 ng hCG/ml (5 mUI/ml) on P, A, and E2 production of SLC and LLC was tested. Ila) OXT (10~10 to 10~5 M) in combination with hCG or without hCG was added to the SLC and LLC cultures and the effect on P release was tested. lib) OXT (10~7 M) and the OXT-antagonist (2 x 10~6 M) were added to the SLC cultures. The effects of these substances and their combination on P production of SLC was tested. lie) The influence of OXT (10~7 M) on A and E2 release of SLC and LLC was examined. Ilia) The effect of E2 (10~7 to 10"5 M) on SLC and LLC progesterone production was tested in the presence and absence of hCG. Illb) The effect of the E2-antagonist monohydroxy-tamoxifen (10~8 M) on P release of SLC cultures was examined. IIIc) The influence of different E2 concentrations (10~7 to 10"5 M) on OXT-induced P inhibition in SLC cultures was tested. P was determined without extraction as described earlier (23). A and E2 were extracted by Sep-Pak C18 columns (Waters Associates, Milford, MA) from the culture medium. After absorption of steroids the columns were washed with 5 ml bidistilled water containing 0.1% acetic acid. Steroids were eluted with 2 ml methanol which was subsequently evaporated. The residue was dissolved in 200 /il hormone-free serum. The recovery for both hormones was greater than 95%, and no influence of the extraction procedure on the E2 (Radioassay Systems Laboratories, Carson, CA) and A (Diagnostic Systems Laboratories, Webster, TX) assay systems was found. Unless indicated otherwise, all data points are expressed as percent related means ± SEM of individual cultures. In each case, however, the 100% hormone concentrations are also given in absolute values. For each culture, parallel control incubations which represent the 100% values were performed. All data stem from at least three different experiments, each with triplicate cultures. To assess the statistical significance of treatment effects, analysis of variance followed by Duncan's multiple range test was employed.
Results Influence of hCG on P, A, and E2 release by SLC and LLC
Figure 1 shows absolute hormone concentrations under control conditions and the mean percent-related data of experiments under hCG-stimulated conditions. While hCG stimulates P and A but not E2 production significantly in SLC, no effect of hCG was observed in LLC. Effect of OXT on basal and hCG-stimulated P, A, and E2 release by SLC and LLC Figure 2 details that OXT (10~10 to 10~5 M) inhibits P release from SLC dose dependently under basal as well as hCG-stimulated conditions. The inhibitory effect became significant (P < 0.05) at a dose of 10~8 M OXT. Concentration needed for inhibiting half-maximal P release (ED50) in response to OXT was 10~7 M. Figure 3 details that OXT had no significant effect on P release from SLC during the initial 48 h incubation period. During the following 24 h (between 48 h and 72
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 May 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
OXYTOCIN EFFECT ON LUTEAL STEROID RELEASE
SLC
LLC
360 300 -
100%"
7«0n«P/w«H
Influence of E2 on P release by luteal cell cultures
200 160 100 60 -
0
360100%-
300 -
1.6n»A/W«H
100%-7.2ngA/w«H
260 200 160O
stimulated P release of small cells was abolished, while the antagonist itself did not change the hormone release.
100%-3766 ngP/w*N
260
A O * Q.
2345
10060
E2 has a dose-dependent stimulatory effect on P release of porcine SLC (Table 2). At a concentration of 10~7 M it increased P release about 2-fold, this is similar to the effect of 6 ng/ml hCG. An E2 concentration of 10~5 M was significantly more potent to stimulate P release than the hCG stimulus. This stimulatory effect of E2 on P release was also observed in hCG-stimulated SLC cultures. In LLC cultures E2 has no significant effect on P release. E2-induced P release can be prevented by coincubation of SLC with the E2- receptor antagonist monohydroxy-tamoxifen (10~8 M) (Table 2). In SLC cultures the addition of different E2 concentrations (10~7 to 10~5 M) can counteract OXT induced inhibition of P release excessively (Table 3).
0
Discussion £
360-
2
300
S
2s
g
200 -
100%-69pgE2/w*N
100%-72pgE2/w»H
°-
A
180
O
100
w
60
hCG(6ng/ml)
•
•
FIG. 1. In vitro effect of hCG on P, A, and E2 release by porcine SLC and LLC. Luteal cells were cultured for 72 h, and steroid release of the last 24 h of culture time was determined. hCG (6 ng/ml) was present during the entire incubation time. Data represent percent-related means ± SEM of at least three different experiments with triplicate cultures. The absolute concentrations representing the 100% values for P, A, and E2 are also given. *, P < 0.05 vs. basal.
h of culture time) and later (these data are not shown), both basal and hCG-stimulated P release was significantly reduced by OXT. Figure 4 shows that OXT (10~7 M) reduced not only the P but also the A release from SLC cultures under basal as well as hCG-stimulated conditions. In contrast to these results E2 release of SLC was significantly stimulated by OXT. No effect on OXT on P and A release of LLC was observed but as in SLC OXT also induced E2 release from these cells. In each case hCG was without effect on this OXT-stimulated E2 release. In order to test the specifity of this inhibitory effect of OXT on P production of SLC cultures, experiments with an OXT-antagonist were performed. Table 1 shows that in the presence of [d(CH2)5Tyr(Me)OVT] (2 x 10~6 M) the inhibitory effect of OXT on basal as well as hCG-
Using a modification of the method of Ohara et al. (30) for human luteal cells it is possible to obtain a useful separation of the two distinct steroidogenic luteal cell populations and to establish that porcine SLC and LLC have functional differences in response to hCG and to OXT. It is interesting to note that spontaneous as well as hCG stimulated P release from SLC decreases with increasing incubation time although hCG responsiveness of these cells increases during this time. This is in agreement with results from mixed luteal cell cultures in the porcine (23, 31) as well as in other species (32, 33). The reason for this spontaneous fall of P secretion in luteal cell cultures is unclear. Based on identical cell numbers per culture well, basal P secretion rates were estimated to be 5- to 7-fold higher in LLC than in SLC. This is in agreement with data published for ovine (34) and bovine (35) luteal cells and also for porcine luteal cells obtained from pregnant sows (1). Our finding that porcine SLC respond to hCG with significantly increased P release whereas LLC under the same experimental conditions respond not at all to hCG, is in agreement with results published for luteal cells of other species (34, 35). It is interesting and new that hCG acts also in SLC to stimulate A. Theoretically the availability of higher precursor concentrations for E2 synthesis should also increase E2 production. As this could not be observed it is possible that A is too far diluted in the culture medium to serve effectively as precursor for E2 synthesis. In earlier in vitro experiments we demonstrated that porcine luteal cells responded to OXT with a dosedependent inhibition of P release under basal as well as hCG-stimulated conditions (23). The present experi-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 May 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
2346
OXYTOCIN EFFECT ON LUTEAL STEROID RELEASE
Endo • 1990 Vol 126 • No 5
100
•o
100% SLC - 536 ng P/well 100% LLC - 2853 ng P/well
so FIG. 2. In vitro effect of OXT on P release by porcine SLC and LLC. Cells were cultured for 72 h, and P release of the last 24 h of culture time was determined. hCG (6 ng/ml) and OXT (1CT10 to 10"6 M) were present during the entire incubation time. Data represent percent-related means ± SEM of at least three experiments with triplicate cultures. OXT at 10~8 M and higher concentrations was significantly (P < 0.05) inhibitory to both basal and hCGstimulated P release.
70 -
o
s CO
•o so 40 30 20 10 -
A - LCC basal n • 6 * - SLC basal n - 6 - 9 • • SLC hCQ n • 6 - 9 -7
-10
log OXT (M) 360 -
incubation tint* 48 h 1OO%- 749ngP/woll
***
300 -
FIG. 3. In vitro effect of OXT on P release by porcine SLC at different incubation times. P release was measured between 24 to 48 h and 48 to 72 h of culture time. Data represent percent-related means ± SEM of three experiments with triplicate cultures. *, P < 0.05 vs. basal; • , P < 0.05 vs. hCG stimulated.
P4 0.06va.hCQ
Incubation tlma 72 h
•* ©
260
8
200
2
180
o c£ %#>
* P * 0.06 va. control •
C
o
n-9
10
100%-427ngP/w«ll
°~
Q.
60 -
hCG (6 ng/ml) OXT (100 nM) ments extend this observation and allow the conclusion that for this effect only SLC are the target cells for OXT. A concentration of 10~8 M OXT is effective to inhibit P release significantly under basal as well as hCG-stimulated conditions. Results from Fig. 3 indicate that OXT became most efficient to decrease P release of SLC after 48 h of incubation time. Therefore we chose to measure effects of OXT on steroidogenesis between 48 and 72 h of incubation time. SLC cultures responded to OXT also with a significant inhibition of basal and hCG-stimulated
A release. As OXT is produced by LLC (7-10, 36, 37) the inhibitory effect on P and A release of SLC represents a paracrine luteolytic effect. The specificity of the OXT-induced inhibitory effect on P release of SLC is demonstrated by the fact that coincubation of OXT and [d(CH2)5Tyr(Me)OVT], a specific OXT antagonist, completely abolished the decrease in P release. This is not due to a toxic effect as we showed previously by measurement of DNA concentrations that OXT influenced neither plating efficiency nor cell via-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 May 2015. at 17:05 For personal use only. No other uses without permission. . All rights reserved.
OXYTOCIN EFFECT ON LUTEAL STEROID RELEASE
§
360
•
300
•
260
75 •
200
o
100
£ 0.
60
O
360 -
J S
300 260
«
200
o
100
2
80
