- Email: [email protected]
Stimulatory activity of serum on prolactin production by human decidua
Stimulatory activity of serum on prolactin production by human decidua ANTHONY A. LUCIANO ILA A. MASLAR WILLIAM F. KUSMIK DA:NIEL H. RIDDICK Farmington, Connecticut Fetal calf serum (FCS) enhances the synthesis and secretion of prolactin (PAL) by exptants of human decidua incubated in Gey's balanced salt solution. In order to further characterize the factors involved in the prolactin-stimulating activity (PSA) of serum, decidua was incubated in Gey's buffer alone and in buffer containing either FCS, bovine serum albumin (BSA), or human serum (HS). When buffer was supplemented with varying concentrations of FCS, a dose-related PSA was apparent between 0.1% and 10%. The PSA of FCS was not altered by either heating (85° C for 30 minutes), boiling, or dialysis. The protein fraction of FCS, precipitated with trichloroacetic acid, redissolved in dilute NaOH, and dialyzed against Dulbecco's phosphate buffer, demonstrated PSA comparable to that of whole serum. A .similar concentration of BSA added to Gay's buffer did not stimulate the production of PAL by decidua. The PSA of human serum obtained from euprolactinemic men and women was not different from that of FCS. These data indicate that FCS and HS contain a substance(s) (1) which stimulates the secretion of PAL by human decidua, (2) which is very stable, being resistant to dialysis, boiling, and denaturation by trichloroacetic acid, (3) and which is most likely a large polypeptide with a molecular weight greater than 12,000 or a smaller moiety tightly bound to a protein. (AM J. Oasret. GYNECOL. 138:665; 1980.)
ExPLANTS of human deciduaL 2 and of chorion with adherent decidua 3 • 4 synthesize and secrete immunoreactive prolactin (PRL). When decidua is incubated in oxygenated Gey's buffer, the amount of PRL produced is significantly increased if the salt solution is supplemented with fetal calf serum. 5• 6 In the present study the prolactin-stimulating activity (PSA) of fetal calf serum (FCS) was further investigated in order to partially characterize the factor(s) involved. The effect of human serum (HS) upon the in vitro production of PRL by decidua was also determined and compared to that of FCS. From the Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, The University of Connecticut School of Medicine. Presented at the Twenty-seventh Annual Meeting of the Society for Gynecologic lnvestigation, Denver, Colorado, March 19-22, 1980. Reprint requests: Anthony A. Luciano, M.D., Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology, University ofiowa, Iowa City, Iowa 52240. 0002-9378/80/220665+05$00.!50/0
©
19RO The C. V. Mosby Co.
Material and methods Fetal membranes were removed from the placenta and washed in normal saline solution immediately after delivery. The amnion was separated from the chorion and discarded. The decidua was dissected from the maternal surface of the chorion, minced with scissors, washed in cold, oxygenated Gey's buffer, 7 and aliquoted by weight (300 mg portions) into 50 ml Erlenmeyer flasks that contained 20 ml of incubation medium. Two flasks were frozen for subsequent analysis in order to determine the initial PRL content of the tissue. The remainder of the flasks were sampled, gassed with 95% oxygen and 5% carbon dioxide, tightly stoppered, and placed in a metabolic shaker maintained at 37° C. These flasks were sampled and regassed during a 24-hour incubation period. At the end of this time, the tissue in each flask was homogenized in the remaining incubation medium, and the supernatant was frozen to be assayed for PRL. PSA of FCS. In a series of experiments, samples of decidua were incubated in duplicate flasks that contained Gey's buffer or Gey's buffer supplemented with
665
666
Luciano et al.
Novc:-mber 15, 1980
Am.
J
Obstet. Gynecol.
Tissue + Medium
Incubation Medium
140 120 100 E ...... 80 Cl
c
...J
a:
60
Cl.
40 20 0 3
6
12
18
24
0
24
INCUBATION TIME (Hrs)
Fig. 1. Production of PRL by human decidua during a 24-hour incubation with (•--•) and without (o--o) fetal calf serum. Each point represents the mean± SEM of the average PRL values measured in samples of medium from duplicate flasks in a representative experiment. The bars represent the total amount of PRL present initially (solid bar) and after 24 hours' incubation with (hatched) and without (open) FCS in the medium.
0.1 %, 1%, 10%, or 20% of FCS (Microbiological Association). The PSA of the FCS was determined by comparing the amount of PRL produced in the enriched medium with that produced in buffer alone. Additional experiments included duplicate flasks in which bovine serum albumin (BSA) (Sigma· Chemical Co.) was added to Gey's buffer in order to provide a concentration of nonspecific protein equivalent to the protein content of 10% FCS in Gey's buffer. The PSA of BSA was compared to that of FCS. In a second series of experiments, FCS was treated as follows: ( 1) Heat-labile proteins were denatured. FCS was either heated for 30 minutes in a water bath maintained at 85° C or boiled for 5 minutes. (2) Smallmolecular weight moieties were removed. Aliquots of 50 ml of FCS were dialyzed at 4o C for 48 hours against 8 L of Dulbecco's phosphate buffer. 8 Cellulose dialysis bags (Thomas Scientific Co.) with a pore diameter of 48 A were used in order to retain substances with a molecular weight of :2:: 12,000. (3) The acid-precipitable protein fraction was isolated. The proteins present in FCS were precipitated by the addition of lO% trichloroacetic acid (TCA). After centrifugation for 20 minutes at 28,000 X g, the supematant was removed. The pellet (precipitated protein fraction) was dissolved in 6N NaOH and deionized water to the original volume (50 ml). This solution was then dialyzed for 48
hours against 8 L of phosphate buffer in order to remove the TCA and to restore the pH to 7.2. Each preparation of treated FCS was tested for PSA in several experiments, each with the use of decidua obtained from a different set of membranes. Decidual tissue was incubated in duplicate or triplicate flasks that contained Gey's buffer or Gey's buffer supplemented with either 10% treated FCS or 10% untreated FCS. PSA was determined by comparing the amount of PRL produced in the presence of treated FCS with that produced in Gey's buffer alone. PSA of HS. Serum was obtained from euprolactinemic male and female volunteers of reproductive age. In each case the serum PRL concentration was <20 ng/ml, and the final concentration ofPRL in Gey's buffer supplemented with 10% HS was <2 ng/ml. The protein fraction of HS was precipitated with saturated ammonium sulfate (SAS) to a final concentration of 50% SAS. After centrifugation for 20 minutes at 28,000 X g, the supernatant was removed, and the pel" let (the precipitated protein fraction) was dissolved in deionized water to the original volume of 25 ml. The solution was then dialyzed in 8 L of phosphate buffer for 48 hours to remove the ammonium sulfate ions and restore the original pH of 7.2. The PSA was determined by comparing the amount of PRL produced by the decidual tissue incubated in buffer that contained
Serum stimulation of decidual prolactin
Volume 138 Nu mber 6
667
Incubation Medium
350
E
'Ol 1:
..J
a:
-*
601 40
I
300 250 200
(/)
a.. 150 100
20
0..
50 3
6
12
0
24
INCUBATION TIME (Hrs) Fig. 2. Accumulation of PRL in Gey's buffer alone (•--•) and in Gey's buffer supplemented with bovine serum albumin (X--X). Each point represents the mean± SEM of the average PRL values measured in samples of medium from duplicate flasks in a representative experiment.
either 10% HS or 10% redissolved protein fraction of HS. with that produced in Gey's buffer alone. In these experiments, samples of decidua were also incubated in Gey's buffer supplemented with 10% FCS, so that the PSA of HS could be compared to that of FCS. PRL and protein assays. Concentrations of PRL in HS, incubation medium, and tissue homogenate supernatants were measured by radioimmunoassay based on the technique of Sinha and associates. 9 Human prolactin (HPr-V-LS2) and rabbit antisera (HPr-V-LS2 and HPr- V-LS3) were provided by the National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, Bethesda, Maryland. Radioimmunoassay data were analyzed by computer program based on the model developed by Burger and associates. 10 All samples from each experiment were analyzed in duplicate in the same assay; the mean intra-assay coefficient of variation was 5.7%. Concentrations of protein in the enriched incubation medium (Gey's buffer supplemented with serum or serum proteins) were determined by the method of Lowry and associates. 11 Statistical analysis. Data were analyzed by the twotailed Student's t test, wi th significance defined as p < 0.05. Results
Fig. l shows the pattern of accumulation of PRL in the medium during a 24-hour incubation of equal amounts of decidua from the same patient in Gey's buffer alone and in Gey's buffer containing iO% FCS. The medium that contained 10% FCS had significantly higher concentrations of PRL than Gey's buffer alone
0.1
1.0
10.0
FCS CONCENTRATION IN MEDIUM (%) Fig. 3. PSA of different concentrations of FCS present in the medium during a 24-hour incubation. Each bar represents the mean and SEM of eight different experiments.
VAl D
-c(
C/)
a..
Gey 's + 10% FCS Gey's + 10% Heated FCS Gey's + 10% Dialysed FCS
300
250
200 150 100
50 Fig. 4. PSA of FCS heated or dialyzed'prior to addition to the incubation medium. The mean ± SEM of four experiments is shown.
at all sampling intervals (P < 0.05). Concentrations of PRL were also determined in the homogenates of tissue plus medium initially and after 24 hours' incubation. The PRL content of the homogenates after incubation in Gey's buffer was three times greater than that present prior to incubation. Homogenates of decidua incubated in Gey's buffer supplemented with 10% FCS contained nearly six times as much PRL as was present prior to incubation. Thus, not only was the amount of PRL released into the medium increased by the presence of FCS, but the total amount of PRL produced was increased, as well (P < 0.0 1). Since FCS might
668
Luciano et al.
~
:-lovember 15, 19HO Am . J. Obstet. Gynecol.
Gey's + 10% FCS
-
Buffer Alone
D
Gey's + Redissolved Protein from 10% FCS
D
Gey's + 10% FCS
-
Gey's
~ Gey's
300 200
< (/')
150
0.
100
50
HS protein
--
300 250
?fi 200 <(
en
Fig. 5. PSA (mean ± SEM for six experiments) of buffer alone compared to that of buffer containing FCS or the redissolved protein fraction from FCS.
+ 10% redissolved
350
250
-
+ 10% HS
a..
150 100 50
0 _.____._ _.___ simply provide a nonspecific protein substrate that resulted in an apparent stimulation of PRL synthesis, a similar concentration of albumin was added to Gey's buffer in place of 10% FCS. In this instance, no significant difference in the amount of PRL released during incubation was observed when compared to Gey's buffer alone (Fig. 2). The effect of increasing concentrations of FCS on total PRL in the tissue plus medium after a 24-hour period of incubation is shown in Fig. 3. Because of the large variability in total secretion of PRL by decidua obtained from different membranes, the data from several experiments (n = 8) are expressed as the percentage increase in PRL produced by the decidua incubated in varying concentrations of FCS compared to that produced in Gey's buffer alone. Minimal PSA was observed when the concentration of FCS in Gey's buffer was 0. 1% or less. PSA was present at FCS concentrations of 1% and was increased still further at concentrations of 10% (P < 0.05). No additional PSA was observed at concentrations greater than 10%. To further characterize the chemical nature of the PRL-stimulating factor or factors present in FCS, the effect of heating and dialysis of FCS prior to the addition of it to Gey's buffer was investigated (Fig. 4). Since no difference was noticed in the PSA of heated versus boiled FCS, the data from these experiments were pooled. Neither heating nor dialysis significantly altered the PSA of FCS. The protein content of FCS ranged between 30 and 40 mg/ml. In order to examine the effect of FCS proteins on PSA, sufficient TCA was added to FCS to effectively precipitate most of the measurable protein. Less than 0.05 mg/ml protein remained in the supernate after this treatment. Fig. 5 shows the PSA of
Fig. 6. PSA (mean ± SEM for 3 experiments) of FCS, HS. and redissolved protein fraction from HS.
buffer alone, redissolved proteins precipitated from FCS, and untreated FCS when added to Gey's buffer containing equal amounts of decidua and incubated for 24 hours . The PSA of the redissolved protein fraction was not significantly different from that of untreated FCS. Thus, PSA resides in the protein fraction of FCS precipitated by TCA. If the PSA of FCS were of possible physiologic significance, it should also be present in HS, as well. To investigate this possibility, parallel experiments were done in which the decidua was incubated in Gey's buffer alone, Gey's buffer that contained 10% FCS, and Gey's buffer that contained either 10% HS or 10% redissolved proteins precipitated from HS. The results of these experiments are shown in Fig. 6.
Commenis There is now substantial evidence that PRL in amniotic fluid is derived, in large part, from PRL synthesized and released by human decidua. This evidence includes: ( 1) the lack of correlation between maternal, fetal, and amniotic fluid levels of PRL under both physiologic and pharmacologic conditions 6 • 12- 15 ; (2) a direct and highly significant correlation between the levels of PRL in amniotic fluid and both the content of PRL in the decidual tissue and its ability to synthesize and release PRL in vitro throughout gestation' 6 ; and (3) the rapid transport of decidual PRL across the chorion in vitro when compared to the transport of radiolabeled PRL added to the medium bathing the fetal membranes. 17 Virtually nothing is known about the factors that
Volume 1:~8 Number 6
control the synthesis and release of PRL by human decidua, although there is evidence that the control mechanisms may be significantly different from those of the pituitary gland. For example, ( 1) bromocriptine, dopamine, and thyrotropin-releasing hormone (TRH), potent modulators of pituitary production of PRL, have no effect on the synthesis and release of PRL by human decidua, 18 • 19 as well as no effect on PRL in amniotic fluid. 1'1 Since FCS is the only substance reported to date which apparently stimulates the production of PRL by decidual tissue, it seemed important to further characterize the substance or substances causing the stimulatory effect. The initial characterization of this factor(s) reported here would suggest that it is either a largeprotein moiety or smaller compound tightly bound to a protein. PSA is contained in the TCA-precipitable pro-
Serum stimulation of decidual prolactin
669
tein fraction of FCS and is unaffected by heating or dialysis. Woolf and associates 20 • 21 have descrilx·d a factor in HS with physical characteristics simibr to those described in this report which increast>d the production of PRL by dispersed rat pituitary cells in \itro. f"he factor in HS was twice as potent as FCS in their system, whereas the PSA of FCS and HS on human decidua appeared to be similar. It is possible that tim f~Ktor(s) could represent a common step in the control of production of PRL by pituitary cells and decidua in vitro. but a more general and unassociated effect ~> also possible. We 'Wish to express our appreciation to ~1s. SallY Kuslis for her technical assistance and to \Is. Darlene Havanec and Ms. Diane Schafer for their help in the preparation of this manuscript.
REFERENCES l. Riddick, D. H., Luciano, A. A., Kusmik, W. F., and MasJar, I. A.: De novo synthesis of prolactin by human decidua, Life Sci. 23:1913, 1978. 2. Golander, A., Hurley, T .. Barrett, J., and Handwerger, S.: Synthesis of prolactin by human decidua in vitro, ]. Endocrinol. 82:263, 1979. 3. Colander, A., Hurley, T., Barrett,j., Hizi, A., and Handwerger, S.: Prolactin synthesis by human chorion-decidual tissue: A possible source of prolactin in the amniotic fluid, Science 202:311, 1978. 4. Healy, D. L., Burger, H. G., and Muller, H. K.: Hypothesis: Placental membranes nroduce nrolactin. Mol. Cell. , ' Endocrinol. 11:1, 1978. ' 5. Riddick, D. H., and Kusmik, W. F.: Decidua: A possible source of amniotic fluid prolactin, AM. ]. 0BSTET. GYNECOL. 127:187, 1977. 6. Riddick, D. H., Luciano, A. A., Kusmik, W. F., and MasJar, 1. A.: Evidence for a non-pituitary source of amniotic fluid prolactin, Fertil. Steril. 31:35, 1979. 7. Porterfield,]. C.: A simple plaque-inhibition test for the study of arthropod-borne viruses, Bull. W. H. 0. 22:373, 1960. 8, Dulbecco, R., and Vogt, M.: Plaque formation and isolation of pure lines with poliomyelitis viruses, J. Exp. Med. 99:167, 1954. 9. Sinha, Y. N., Seiby, P. W., Lewis, V. J., and Vanderlaan, W. P.: A homologous radioimmunoassay for human prolactin, T. Clin. Endocrinol. Metab. 36:509, 1973. 10. Burger, H. G., Lee, V. W. K., and Rennie, G. C.: A generalized computer program for the treatment of data from competitive protein-binding assays including radioimmunoassays,]. Lab. Clin. Med. 80:302, 1972. 1l. Lowry, 0. H., Rosebrough, N. ]., Farr, A. L., and Randall, R. J.: Protein measurement with the Folin phenol reagent,]. Bioi. Chern. 193:265, 1951. 12. Friesen, H., Hwang, P., Guyda, H., Tolis, G., Tyson,]., Meyers, R.: A radioimmunoassay for human prolactin: Physiological, pathological and pharmacological factors which affect the secretion of prolactin, in Boyns, A. R., and Griffith, K., editors: Prolactin and Carcinogenesis, Cardiff, 1972, Alpha Omega Publishing, p. 64.
13. Bigazzi, M., Ronga, R., Lancranjan, I., et al.: A pregnancr in an acromegalic woman during bromocriptine treatment: Effects on growth hormone and prolactin in the maternal, fetal, and amniotic comportments . .J. Clin. Endocrinol. Metab. 48:9, 1979. 14. Josimovich_. .J• Wei~s, .G., and Hu.tchinson, D.: S~mrces and depos1t1on of p1tu1tary prolacun In maternal Circula-
tion, amniotic fluid, fetus and placenta in the pregnant rhesus monkey, Endocrinology 94: 1364, 197-t 15. Josimovich, J ., Merisko, K., and Bocceiia, L.: Amniotic prolactin control over amniotic and fetal extracellular fluid water and electrolytes in the rhesus monkev, Endocrinology 100:564, 1977. 16. Rosenberg, S. M., Maslar, I. A., and Riddick, D. H.: Capacity of human decidua to produce prolactin in vitro throughout late gestation, in Scientific Abstracts of the Twenty-seventh Annual Meeting of the Society for Gynecologic Investigation, Denver, Colorado. March 19-22, 1980. (Abst. No. 317.) 17. Riddick, D. H., and Maslar, I. A.: The transport of prolactin by human fetal membranes, in Scientific Abstracts of the Twenty-seventh Annual Meeting of the Society for Gynecologic Investigation, Denver, Colorado. March 19-22, 1980. (Abstract No. 275.) 18. Golander, A., Barrett, J., Hurley, T., Barry, S., and Handwerger, S.: Failure of bromocriptine, dopamine, and thyrotropin-releasing hormone to affect prolactin secretion by human decidual tissue in vitro, f. Clin. Endocrinol. Metab. 49:787, 1979. 19. Bigazzi, M., Pollicino, G., and Nardi, E.: Is human de· cidua a specialized endocrine organ~ J. Clin. Endocrinol. Metab. 49:847, 1979. 20. Woolf, P. D., and LeTourneau, K.: Characterization of a pituitary cell culture system derived from pregnant rats, Am.]. Physiol. 236:E239, 1979. 2l. Woolf, P. D., LeTourneau, K., and Livingston, .J. N .: Prolactin stimulatory activity of human serum, in Scientific Abstracts of the Sixty-first Annual Meeting of The Endocrine Society, June, 1979. (Abstract No. 346.)
ExPLANTS of human deciduaL 2 and of chorion with adherent decidua 3 • 4 synthesize and secrete immunoreactive prolactin (PRL). When decidua is incubated in oxygenated Gey's buffer, the amount of PRL produced is significantly increased if the salt solution is supplemented with fetal calf serum. 5• 6 In the present study the prolactin-stimulating activity (PSA) of fetal calf serum (FCS) was further investigated in order to partially characterize the factor(s) involved. The effect of human serum (HS) upon the in vitro production of PRL by decidua was also determined and compared to that of FCS. From the Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, The University of Connecticut School of Medicine. Presented at the Twenty-seventh Annual Meeting of the Society for Gynecologic lnvestigation, Denver, Colorado, March 19-22, 1980. Reprint requests: Anthony A. Luciano, M.D., Department of Obstetrics and Gynecology, Division of Reproductive Endocrinology, University ofiowa, Iowa City, Iowa 52240. 0002-9378/80/220665+05$00.!50/0
©
19RO The C. V. Mosby Co.
Material and methods Fetal membranes were removed from the placenta and washed in normal saline solution immediately after delivery. The amnion was separated from the chorion and discarded. The decidua was dissected from the maternal surface of the chorion, minced with scissors, washed in cold, oxygenated Gey's buffer, 7 and aliquoted by weight (300 mg portions) into 50 ml Erlenmeyer flasks that contained 20 ml of incubation medium. Two flasks were frozen for subsequent analysis in order to determine the initial PRL content of the tissue. The remainder of the flasks were sampled, gassed with 95% oxygen and 5% carbon dioxide, tightly stoppered, and placed in a metabolic shaker maintained at 37° C. These flasks were sampled and regassed during a 24-hour incubation period. At the end of this time, the tissue in each flask was homogenized in the remaining incubation medium, and the supernatant was frozen to be assayed for PRL. PSA of FCS. In a series of experiments, samples of decidua were incubated in duplicate flasks that contained Gey's buffer or Gey's buffer supplemented with
665
666
Luciano et al.
Novc:-mber 15, 1980
Am.
J
Obstet. Gynecol.
Tissue + Medium
Incubation Medium
140 120 100 E ...... 80 Cl
c
...J
a:
60
Cl.
40 20 0 3
6
12
18
24
0
24
INCUBATION TIME (Hrs)
Fig. 1. Production of PRL by human decidua during a 24-hour incubation with (•--•) and without (o--o) fetal calf serum. Each point represents the mean± SEM of the average PRL values measured in samples of medium from duplicate flasks in a representative experiment. The bars represent the total amount of PRL present initially (solid bar) and after 24 hours' incubation with (hatched) and without (open) FCS in the medium.
0.1 %, 1%, 10%, or 20% of FCS (Microbiological Association). The PSA of the FCS was determined by comparing the amount of PRL produced in the enriched medium with that produced in buffer alone. Additional experiments included duplicate flasks in which bovine serum albumin (BSA) (Sigma· Chemical Co.) was added to Gey's buffer in order to provide a concentration of nonspecific protein equivalent to the protein content of 10% FCS in Gey's buffer. The PSA of BSA was compared to that of FCS. In a second series of experiments, FCS was treated as follows: ( 1) Heat-labile proteins were denatured. FCS was either heated for 30 minutes in a water bath maintained at 85° C or boiled for 5 minutes. (2) Smallmolecular weight moieties were removed. Aliquots of 50 ml of FCS were dialyzed at 4o C for 48 hours against 8 L of Dulbecco's phosphate buffer. 8 Cellulose dialysis bags (Thomas Scientific Co.) with a pore diameter of 48 A were used in order to retain substances with a molecular weight of :2:: 12,000. (3) The acid-precipitable protein fraction was isolated. The proteins present in FCS were precipitated by the addition of lO% trichloroacetic acid (TCA). After centrifugation for 20 minutes at 28,000 X g, the supematant was removed. The pellet (precipitated protein fraction) was dissolved in 6N NaOH and deionized water to the original volume (50 ml). This solution was then dialyzed for 48
hours against 8 L of phosphate buffer in order to remove the TCA and to restore the pH to 7.2. Each preparation of treated FCS was tested for PSA in several experiments, each with the use of decidua obtained from a different set of membranes. Decidual tissue was incubated in duplicate or triplicate flasks that contained Gey's buffer or Gey's buffer supplemented with either 10% treated FCS or 10% untreated FCS. PSA was determined by comparing the amount of PRL produced in the presence of treated FCS with that produced in Gey's buffer alone. PSA of HS. Serum was obtained from euprolactinemic male and female volunteers of reproductive age. In each case the serum PRL concentration was <20 ng/ml, and the final concentration ofPRL in Gey's buffer supplemented with 10% HS was <2 ng/ml. The protein fraction of HS was precipitated with saturated ammonium sulfate (SAS) to a final concentration of 50% SAS. After centrifugation for 20 minutes at 28,000 X g, the supernatant was removed, and the pel" let (the precipitated protein fraction) was dissolved in deionized water to the original volume of 25 ml. The solution was then dialyzed in 8 L of phosphate buffer for 48 hours to remove the ammonium sulfate ions and restore the original pH of 7.2. The PSA was determined by comparing the amount of PRL produced by the decidual tissue incubated in buffer that contained
Serum stimulation of decidual prolactin
Volume 138 Nu mber 6
667
Incubation Medium
350
E
'Ol 1:
..J
a:
-*
601 40
I
300 250 200
(/)
a.. 150 100
20
0..
50 3
6
12
0
24
INCUBATION TIME (Hrs) Fig. 2. Accumulation of PRL in Gey's buffer alone (•--•) and in Gey's buffer supplemented with bovine serum albumin (X--X). Each point represents the mean± SEM of the average PRL values measured in samples of medium from duplicate flasks in a representative experiment.
either 10% HS or 10% redissolved protein fraction of HS. with that produced in Gey's buffer alone. In these experiments, samples of decidua were also incubated in Gey's buffer supplemented with 10% FCS, so that the PSA of HS could be compared to that of FCS. PRL and protein assays. Concentrations of PRL in HS, incubation medium, and tissue homogenate supernatants were measured by radioimmunoassay based on the technique of Sinha and associates. 9 Human prolactin (HPr-V-LS2) and rabbit antisera (HPr-V-LS2 and HPr- V-LS3) were provided by the National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, Bethesda, Maryland. Radioimmunoassay data were analyzed by computer program based on the model developed by Burger and associates. 10 All samples from each experiment were analyzed in duplicate in the same assay; the mean intra-assay coefficient of variation was 5.7%. Concentrations of protein in the enriched incubation medium (Gey's buffer supplemented with serum or serum proteins) were determined by the method of Lowry and associates. 11 Statistical analysis. Data were analyzed by the twotailed Student's t test, wi th significance defined as p < 0.05. Results
Fig. l shows the pattern of accumulation of PRL in the medium during a 24-hour incubation of equal amounts of decidua from the same patient in Gey's buffer alone and in Gey's buffer containing iO% FCS. The medium that contained 10% FCS had significantly higher concentrations of PRL than Gey's buffer alone
0.1
1.0
10.0
FCS CONCENTRATION IN MEDIUM (%) Fig. 3. PSA of different concentrations of FCS present in the medium during a 24-hour incubation. Each bar represents the mean and SEM of eight different experiments.
VAl D
-c(
C/)
a..
Gey 's + 10% FCS Gey's + 10% Heated FCS Gey's + 10% Dialysed FCS
300
250
200 150 100
50 Fig. 4. PSA of FCS heated or dialyzed'prior to addition to the incubation medium. The mean ± SEM of four experiments is shown.
at all sampling intervals (P < 0.05). Concentrations of PRL were also determined in the homogenates of tissue plus medium initially and after 24 hours' incubation. The PRL content of the homogenates after incubation in Gey's buffer was three times greater than that present prior to incubation. Homogenates of decidua incubated in Gey's buffer supplemented with 10% FCS contained nearly six times as much PRL as was present prior to incubation. Thus, not only was the amount of PRL released into the medium increased by the presence of FCS, but the total amount of PRL produced was increased, as well (P < 0.0 1). Since FCS might
668
Luciano et al.
~
:-lovember 15, 19HO Am . J. Obstet. Gynecol.
Gey's + 10% FCS
-
Buffer Alone
D
Gey's + Redissolved Protein from 10% FCS
D
Gey's + 10% FCS
-
Gey's
~ Gey's
300 200
< (/')
150
0.
100
50
HS protein
--
300 250
?fi 200 <(
en
Fig. 5. PSA (mean ± SEM for six experiments) of buffer alone compared to that of buffer containing FCS or the redissolved protein fraction from FCS.
+ 10% redissolved
350
250
-
+ 10% HS
a..
150 100 50
0 _.____._ _.___ simply provide a nonspecific protein substrate that resulted in an apparent stimulation of PRL synthesis, a similar concentration of albumin was added to Gey's buffer in place of 10% FCS. In this instance, no significant difference in the amount of PRL released during incubation was observed when compared to Gey's buffer alone (Fig. 2). The effect of increasing concentrations of FCS on total PRL in the tissue plus medium after a 24-hour period of incubation is shown in Fig. 3. Because of the large variability in total secretion of PRL by decidua obtained from different membranes, the data from several experiments (n = 8) are expressed as the percentage increase in PRL produced by the decidua incubated in varying concentrations of FCS compared to that produced in Gey's buffer alone. Minimal PSA was observed when the concentration of FCS in Gey's buffer was 0. 1% or less. PSA was present at FCS concentrations of 1% and was increased still further at concentrations of 10% (P < 0.05). No additional PSA was observed at concentrations greater than 10%. To further characterize the chemical nature of the PRL-stimulating factor or factors present in FCS, the effect of heating and dialysis of FCS prior to the addition of it to Gey's buffer was investigated (Fig. 4). Since no difference was noticed in the PSA of heated versus boiled FCS, the data from these experiments were pooled. Neither heating nor dialysis significantly altered the PSA of FCS. The protein content of FCS ranged between 30 and 40 mg/ml. In order to examine the effect of FCS proteins on PSA, sufficient TCA was added to FCS to effectively precipitate most of the measurable protein. Less than 0.05 mg/ml protein remained in the supernate after this treatment. Fig. 5 shows the PSA of
Fig. 6. PSA (mean ± SEM for 3 experiments) of FCS, HS. and redissolved protein fraction from HS.
buffer alone, redissolved proteins precipitated from FCS, and untreated FCS when added to Gey's buffer containing equal amounts of decidua and incubated for 24 hours . The PSA of the redissolved protein fraction was not significantly different from that of untreated FCS. Thus, PSA resides in the protein fraction of FCS precipitated by TCA. If the PSA of FCS were of possible physiologic significance, it should also be present in HS, as well. To investigate this possibility, parallel experiments were done in which the decidua was incubated in Gey's buffer alone, Gey's buffer that contained 10% FCS, and Gey's buffer that contained either 10% HS or 10% redissolved proteins precipitated from HS. The results of these experiments are shown in Fig. 6.
Commenis There is now substantial evidence that PRL in amniotic fluid is derived, in large part, from PRL synthesized and released by human decidua. This evidence includes: ( 1) the lack of correlation between maternal, fetal, and amniotic fluid levels of PRL under both physiologic and pharmacologic conditions 6 • 12- 15 ; (2) a direct and highly significant correlation between the levels of PRL in amniotic fluid and both the content of PRL in the decidual tissue and its ability to synthesize and release PRL in vitro throughout gestation' 6 ; and (3) the rapid transport of decidual PRL across the chorion in vitro when compared to the transport of radiolabeled PRL added to the medium bathing the fetal membranes. 17 Virtually nothing is known about the factors that
Volume 1:~8 Number 6
control the synthesis and release of PRL by human decidua, although there is evidence that the control mechanisms may be significantly different from those of the pituitary gland. For example, ( 1) bromocriptine, dopamine, and thyrotropin-releasing hormone (TRH), potent modulators of pituitary production of PRL, have no effect on the synthesis and release of PRL by human decidua, 18 • 19 as well as no effect on PRL in amniotic fluid. 1'1 Since FCS is the only substance reported to date which apparently stimulates the production of PRL by decidual tissue, it seemed important to further characterize the substance or substances causing the stimulatory effect. The initial characterization of this factor(s) reported here would suggest that it is either a largeprotein moiety or smaller compound tightly bound to a protein. PSA is contained in the TCA-precipitable pro-
Serum stimulation of decidual prolactin
669
tein fraction of FCS and is unaffected by heating or dialysis. Woolf and associates 20 • 21 have descrilx·d a factor in HS with physical characteristics simibr to those described in this report which increast>d the production of PRL by dispersed rat pituitary cells in \itro. f"he factor in HS was twice as potent as FCS in their system, whereas the PSA of FCS and HS on human decidua appeared to be similar. It is possible that tim f~Ktor(s) could represent a common step in the control of production of PRL by pituitary cells and decidua in vitro. but a more general and unassociated effect ~> also possible. We 'Wish to express our appreciation to ~1s. SallY Kuslis for her technical assistance and to \Is. Darlene Havanec and Ms. Diane Schafer for their help in the preparation of this manuscript.
REFERENCES l. Riddick, D. H., Luciano, A. A., Kusmik, W. F., and MasJar, I. A.: De novo synthesis of prolactin by human decidua, Life Sci. 23:1913, 1978. 2. Golander, A., Hurley, T .. Barrett, J., and Handwerger, S.: Synthesis of prolactin by human decidua in vitro, ]. Endocrinol. 82:263, 1979. 3. Colander, A., Hurley, T., Barrett,j., Hizi, A., and Handwerger, S.: Prolactin synthesis by human chorion-decidual tissue: A possible source of prolactin in the amniotic fluid, Science 202:311, 1978. 4. Healy, D. L., Burger, H. G., and Muller, H. K.: Hypothesis: Placental membranes nroduce nrolactin. Mol. Cell. , ' Endocrinol. 11:1, 1978. ' 5. Riddick, D. H., and Kusmik, W. F.: Decidua: A possible source of amniotic fluid prolactin, AM. ]. 0BSTET. GYNECOL. 127:187, 1977. 6. Riddick, D. H., Luciano, A. A., Kusmik, W. F., and MasJar, 1. A.: Evidence for a non-pituitary source of amniotic fluid prolactin, Fertil. Steril. 31:35, 1979. 7. Porterfield,]. C.: A simple plaque-inhibition test for the study of arthropod-borne viruses, Bull. W. H. 0. 22:373, 1960. 8, Dulbecco, R., and Vogt, M.: Plaque formation and isolation of pure lines with poliomyelitis viruses, J. Exp. Med. 99:167, 1954. 9. Sinha, Y. N., Seiby, P. W., Lewis, V. J., and Vanderlaan, W. P.: A homologous radioimmunoassay for human prolactin, T. Clin. Endocrinol. Metab. 36:509, 1973. 10. Burger, H. G., Lee, V. W. K., and Rennie, G. C.: A generalized computer program for the treatment of data from competitive protein-binding assays including radioimmunoassays,]. Lab. Clin. Med. 80:302, 1972. 1l. Lowry, 0. H., Rosebrough, N. ]., Farr, A. L., and Randall, R. J.: Protein measurement with the Folin phenol reagent,]. Bioi. Chern. 193:265, 1951. 12. Friesen, H., Hwang, P., Guyda, H., Tolis, G., Tyson,]., Meyers, R.: A radioimmunoassay for human prolactin: Physiological, pathological and pharmacological factors which affect the secretion of prolactin, in Boyns, A. R., and Griffith, K., editors: Prolactin and Carcinogenesis, Cardiff, 1972, Alpha Omega Publishing, p. 64.
13. Bigazzi, M., Ronga, R., Lancranjan, I., et al.: A pregnancr in an acromegalic woman during bromocriptine treatment: Effects on growth hormone and prolactin in the maternal, fetal, and amniotic comportments . .J. Clin. Endocrinol. Metab. 48:9, 1979. 14. Josimovich_. .J• Wei~s, .G., and Hu.tchinson, D.: S~mrces and depos1t1on of p1tu1tary prolacun In maternal Circula-
tion, amniotic fluid, fetus and placenta in the pregnant rhesus monkey, Endocrinology 94: 1364, 197-t 15. Josimovich, J ., Merisko, K., and Bocceiia, L.: Amniotic prolactin control over amniotic and fetal extracellular fluid water and electrolytes in the rhesus monkev, Endocrinology 100:564, 1977. 16. Rosenberg, S. M., Maslar, I. A., and Riddick, D. H.: Capacity of human decidua to produce prolactin in vitro throughout late gestation, in Scientific Abstracts of the Twenty-seventh Annual Meeting of the Society for Gynecologic Investigation, Denver, Colorado. March 19-22, 1980. (Abst. No. 317.) 17. Riddick, D. H., and Maslar, I. A.: The transport of prolactin by human fetal membranes, in Scientific Abstracts of the Twenty-seventh Annual Meeting of the Society for Gynecologic Investigation, Denver, Colorado. March 19-22, 1980. (Abstract No. 275.) 18. Golander, A., Barrett, J., Hurley, T., Barry, S., and Handwerger, S.: Failure of bromocriptine, dopamine, and thyrotropin-releasing hormone to affect prolactin secretion by human decidual tissue in vitro, f. Clin. Endocrinol. Metab. 49:787, 1979. 19. Bigazzi, M., Pollicino, G., and Nardi, E.: Is human de· cidua a specialized endocrine organ~ J. Clin. Endocrinol. Metab. 49:847, 1979. 20. Woolf, P. D., and LeTourneau, K.: Characterization of a pituitary cell culture system derived from pregnant rats, Am.]. Physiol. 236:E239, 1979. 2l. Woolf, P. D., LeTourneau, K., and Livingston, .J. N .: Prolactin stimulatory activity of human serum, in Scientific Abstracts of the Sixty-first Annual Meeting of The Endocrine Society, June, 1979. (Abstract No. 346.)