Placenrtr (x989), IO, 331-344

Human Embryonal Extracts Modulate Placental Function in the First Trimester: Effects of Visceral Tissues upon Chorionic Gonadotropin and Progesterone Secretiona

EYTAN R. BARNEA$ ROBERT J. SIMON”~b,c & SHAHAR KOL

’ Supported in part by theJuvenile Diabetes Foundation and the Lady Davis Foundation to E.R.B.

b In partial fulfillment of the requirements to receive the degree of Doctor of Medicine, Faculty of Medicine, Technion, Israel Institute of Technology, Hasfa, Israel.

’ Current address: Division of Family Medicine, UCLA Center for Health Sciences, Los Angeles, California, U.S.A.

d Address for correspondence: Director, Feto-Placental Endocrinology Crnit Rappaport Research Institute, Technion, P.O.B., 9697, Hasfa, Israel

Paper accepted 23.1989

INTRODUCTION

First trimester pregnancy supports both the development of the functional placenta from the trophoblast, and organogenesis during embryonic development. During early differentiation of the blastocyst, a relationship between the developmental events in the trophoblast and the embryoblast is apparent. However, with the appearance of the extra-embryonic coelom and the eventual differentiation to the chorionic cavity, separation of the supporting placental struc- tures from the embryo occurs. The requirement of trophoblast development to maintain embryoblast support is well documented (Hertig, Rock and Adams, 1956). However, reliance of the trophoblast on embryonic development is not as well documented (Davies and Glasser, 1967). Available data supports the concept that pregnancy maintenance via hormone release from the corpus luteum is required in all mammalian species (Castracane and Goldzieher, 1986).

Responsibility for progesterone (P,) formation is assumed by the human placenta during weeks 7-9 of gestation but in other species this responsibility is shifted earlier (Goodman and Hodgen, 1979). The stimulus for this change is unknown, although it seems to be related to the presence of a living fetus. Levels of r7-hydroxyprogesterone decrease in pregnancies with blighted ovum, despite the presence of continued human chorionic gonadotropin (hCG) pro- duction (Aspillaga et al, 1983). Serum prolactin levels are also associated with viability of the conceptus and drop precipitously upon its death (Riss, Mick and Spona, 1980). Moreover, trophoblast obtained very shortly after embryonal demise grows poorly and has a low density of gonadotropin releasing hormone (GnRH) receptors (Belisle et al, 1987; Hustin and

0143-4004/89/~40331+ t4 $q.oo/o 0 1989 Baillike Tindall Ltd

332 Placenta (1989), Vol. IO

Gaspard, 1977). Similarly, in the sheep and the rat, placental lactogen levels drop upon fetal death (Robertson et al, 1984; Taylor et al, 1984).

In order to better define the embryo-placental relationship in the human, we have studied the effect of various embryonic visceral organs upon placental function, in vitro, as reflected by hCG and P, secretion into the media of trophoblast explant cultures grown in the presence of these extracts. We report that these embryonal extracts exert a significant effect upon tropho- blastic function, suggestive of their role in vivo.

MATERIALS AND METHODS

Embryonal extracts After obtaining the appropriate consent an elective pregnancy termination was performed in the first trimester of pregnancy in the Rambam Medical Center by using gentle suction curret- age. Embryos of 8-r I weeks were removed aseptically, and gestational ages were confirmed by last menstrual period and crown-rump measurements of the specimen itself. At that gestational age the error of such a measurement is only + / - 3 days. Only apparently normal embryos were used. This was determined by gross and microscopic examination of the specimen. Those that contained abnormal looking trophoblast, or embryos were not used. Embryonal organs were dissected of the whole embryo and positively identified utilizing a dissecting microscope, and were placed on ice. All subsequent steps were carried out at 4°C. Visceral organs, compris- ing lung, liver, kidney and adrenal, were extracted either in I ml of absolute ethyl alcohol or in I ml of sterile distilled deionized water by sonication for 60 set (Triacregler-Yatral GmbH D-7801, Dottringhen sonicator.) Particulate matter was sedimented by centrifugation at z 800 rpm for IO min. The supernatant fractions were removed and, in the case of the alcoholic extracts, were evaporated under nitrogen, and the sediment was resuspended in distilled water (alcohol/water). Several dilutions were prepared of each extract. Extracts were stored at - 2o°C until used.

Placental explants Ten first trimester (7-10 weeks gestational age from last menstrual period) placentae were obtained after elective pregnancy terminations. Tissue was rinsed several times in 0.9 per cent NaCl to remove all blood, followed by three rinses with culture medium (Dulbecco’s Modified Eagle Medium) (DMEM), (Beit Haemek, Israel,) containing penicillin 5 ooo U/ml, strepto- mycin 20 pug/ml and amphotericin B 50 pug/ml. Tissue was trimmed of supporting membranes and sharply cut into 1-2 mm size explants.

Cultures. Explants (4-20 per treatment group) were cultured in DMEM containing I per cent antibiotics and 0.5 per cent bovine serum albumin as previously reported (Barnea et al, 1986). Equivolume amounts of dilutions of tissue extracts of 0.005 per cent to 5 per cent were added to each culture. Cultures were incubated at 37°C for 24 h in 95 per cent air and 5 per cent CO, atmosphere. Dishes containing only vehicle served as controls. Additional culture dishes con- taining DMEM and tissue extracts, but lacking trophoblast explants were prepared to deter- mine possible interference of extracts with assay procedures. Following incubation, the media was removed and stored for assay at -20°C. Explant tissues were saved for protein content ana- lysis. Explant viability was confirmed by haematoxylin-eosin staining and progressive glucose consumption during the incubation period. The secretion of hCG and P, by explants increased

tlarnea et al: Human Embryonal Organs Effect on Placenta 333

linearly until 48 h. Each embryonal organ was tested on explants made of two to three different placentae. The detailed methodology was recently reported (Barnea and Kaplan, in press).

Superfusion studies .dn Accusyst S superfusion apparatus (Endotronics, St. Paul, MN), in addition to a multi- channel peristaltic pump (Ismatech, Chicago, IL) with multichannel digital flowmeter (ISCO, Omaha, NE) was utilized. Incubation chambers containing trophoblast explant cultures were perfused with DMEM containing 20 mM Hepes buffer (Sigma, St. Louis, MO.) with a pulse for only I min of the following (a) I ml/min pulses of aqueous embryonal lung tissue I: IOO extract added every 30 min; (b) I ml/min pulses of lung tissue extract plus I ml/min IOO pM gonadotropin releasing hormone agonist (GnRHa) (Sigma, St. Louis, MO) added every 30 min; (c) I ml/min of IOO pM GnRHa added every 30 min; (d) I ml/min Hepes DMEM alone (control). The experiments were performed for a period of 4 h at 37°C under 95 per cent air and 5 per cent CO, atmosphere at pH 7.4. Monitoring of oxygen and CO, content of the per- fusion medium was repeatedly performed using a blood-gas analyser (Radiometer, Copen- hagen) to ensure adequate gas exchange and pH control in the chambers, pC0, = 34 mmHg, ~0, = 164mmHg.

Dialysis Using an 8 ooo dalton exclusion bag, the aqueous extract of embryonal lung tissue was dialysed against an aqueous gradient solution for 24 h at 4°C. The dialysate was then recovered concen- trated and utilized for explant culture treatment.

Protein inactivation In order to inactivate protein putative compounds in the embryonal lung extract, the solution was treated with IO PM p-chloromercurobenzoate, a proteolytic inhibitor, or by heating at 57°C for 30 min prior to addition to the trophoblast cultures.

.&a)~. Placental secretion of hCG and P, into the media was measured by radioimmunoassay (Barnea et al, 1986). All assays for a specific placenta were carried out during the same assay. Visceral organ extracts hCG and P, content were below detection limit and they did not appear to interfere with the RIA procedure. Placental secretion of hCG following 24 h ranged from 50-5 ooo ng/mg protein, while P, secretion ranged from 2-20 ng/mg protein. Placental and visceral tissue protein content was measured by the method of (Lowry et al, 1951) using bovine serum albumin as the standard. Statistical analysis was performed by one-way ANOVA and Student’s t-test. Superfusion experiments were analysed by regression analysis and per- cent change using an Engineering one program on an IBM personal computer. p < 0.05 was considered statistically significant. Data were expressed as ng/mg trophoblast protein for both hCG and P,.

RESULTS

Lung Effect on KG. Alcohol/water extracts: Trophoblast explant cultures grown in the presence of 1/200 diluted extracts of 9 week embryonal lung tissue caused up to a fivefold increase in hCG production, an effect which was not seen with more concentrated extracts (Figure I).

Water extracts: The 8 week I/IOO diluted lung extract produced a significant twofold de-

334 Placenta (1989), Vol. 10

250 r --- _--

0 -0 IO weeks

0-0 9 weeks

”

Control

*p<o.o5 vs. control

0.05 0.005

Lung extract dilution

Figure I. The effect of alcohol/water extracts of fetal lung upon hCG secretion by placental explants. Both the 9- and IO-week extract had a significant stimulatory effect upon hCG secretion.

crease in hCG production. Trophoblast cultures grown in the presence of the water extract which had undergone dialysis to remove compounds with molecular weight of < 8 ooo dalton, eliminated the drop in hCG secretion obtained with the untreated extracts. Similarly, incuba- tions made with I/IOO diluted lung extracts, pretreated with (IO pM) p-chloromercurobenzoate or heating for 30 min at 57°C eliminated the inhibitory effect of the lung extract upon tropho- blast hCG production (Figure 2).

Pulsatile effects of lung extracts. Figure 3 shows the significant inhibitory effect of I min pulses

140

120

100 .E z! g 00

F 2 60

:: = 40

20

0 Control lung IO/4

ext. PCMB

-- Dialysis

lung (8 ZOO MW

- Treatment

*p<O.O5 vs. control

Figure .z. The effect of the 8 week water extract of fetal lung diluted at I/IW upon hCG secretion. The untreated water extract reduced signifiantly hCG secretion. Heat inactivation for 3omin at 57”C, treatment with (IO~M) p-chloro- mercurobenzoate (PCh4B) and dialysis with a 8ooo dalton exclusion limit eliminated the obtained effect with the untreated water extract.

Burnea et al: Human Embryonal Organs Effect on Placenta 335

. -0 Control 0 0 Lung

ol I I I I I I I I I I I I I I I ICY I2 34 5 6 7 8 9 IO II I2 I3 I4 I5 I6 17 I8 19

A A A Lung Lung LWg

Sample number

Figure 3, The effect of the 8-week I/IOO diluted water lung extract pulses for I min upon hCG secretion in superfusion of placental explants. Pulses are denoted with a * sign). Compared to controls (black dots) I min pulses given at 30,60 and go min significantly decreased the pulsatile release of hCG. Sample collection of effluent was carried out every 6 min. Further details are described in the material and methods section.

of r/100 diluted water extracts of 8-week lung upon pulsatile hCG secretion in the placental superfusion system. Concomitant addition of a I min pulse of a potent GnRHa at IOO PM,

which is very effective in stimulating pulsatile hCG release (data not shown), failed to affect the noted inhibitory effect seen with the lung extract alone (Figure 4).

Effect on Pd. Alcohol/water extracts: This extract produced a twofold decrease in P4 production levels by the explant cultures at 9 weeks while the opposite was noted at IO weeks (Figure 5).

Water extraction had no significant effect upon P, secretion (data not shown).

Kidney E&t on KC. Alcohol/water extracts: Diluted extracts of fetal kidney stimulated, inhibited and had no effect at 9, IO and I I weeks, respectively. The water extracted kidney of I I weeks had a significant inhibitory effect upon hCG secretion (Figure 6).

Effect on P4. Alcohol/water extracts: P, production was significantly decreased in the presence of 9-week fetal kidney extract. At I I weeks, however it had a significant stimulatory effect upon P, secretion (Figure 7).

The water extracted kidney had no significant effect upon P, secretion (data not shown).

Adrenal E&t on KG. Alcohol/water extracts: A significant decrease in hCG production by the explant cultures was obtained in presence of 1/20 extract of 9 week embryonal adrenal tissue. Tissue extract from IO week embryo however, had no effect (Figure 8).

336 Placenta (1989), Vol. 10

0-a Control

0 0 Treatment

13 14 15 16 17 18 I9 A

Lung A

Lung + GnRH

Sample number

Ltng

Figure 4. The effect of I/IOO diluted pulses of water extracted lung extract for I min upon hCG secretion in superfu- sion of placental explants (pulses are denoted with - sign). Compared to controls, (black dots) a significant suppression was noted. This was not overcome by concomitant addition of too PM of GNRH (a), a strong simulator of hCG secre- tion Sample collection was carried out every 6 min. Further details are described in the materials and methods section.

0 0 IO weeks

0-0 9 weeks

x

0: Control 0.05 0.005 0.0005

Lung extract dilution

*pCO.O5 vs. control

Figure 5, The effect of various dilutions of alcohol/water extract of lung upon placental explants P, secretion. Signific- ant inhibition of P, was noted at o weeks. The effect became stimulatory at IO weeks.

Ham-a it al: Human Embryonal Organs Effect on Placenta

m Control 9 weeks 17 0.05 dilution

* m 0.005 dilution

W 0.01 dilution

*p<O.O5 vs. control

Kidney tissue extract

337

Figure 6. The effect of alcohol/water and water extracts of kidney at various gestational ages upon placental explants hCG secretion. The alcohol/water extract had an inhibitory effect in 0.05 and 0.005 dilutions, while the water extracts had a significant inhibitory effect at 0.01 dilutions.

300

1

250 1

0 0 9 weeks *

l --@ 9 weeks i A- - n II weeks

,a / , / /

* , /

,L- ’

/

,

0' Control

*p<O.O5 vs. control

0.05 0.005

Kidney extract dilution

Ftgure 7. The effect of alcohol/water kidney extracts upon placental explants P, secretion. At o weeks a significant in- hibitory effect was noted. The effect at I r weeks was however stimulatory.

338 Placenta (1989), Vol. 10

Em Control

0 Treatment

9 weeks IO weeks IO weeks I I weeks

Alcohol

I I weeks

Water

*p’p<O.O5 vs. cantral

Adrenal tissue dilution

Figure 8. The effect of various gestational age adrenal extracts upon placental explants hCG secretion. The alcohol/ water extracts effect varied in the gestational weeks (w) studied. The I r-week water extract had however, a significant inhibitory effect.

Water extracts: The extracted adrenal diluted at I/IOO had a significant inhibitory effect upon hCG secretion at I I weeks (Figure 8).

Effect on P4. Alcohol/water extracts: A twofold increase in P, production by trophoblast explant cultures was seen under the influence of 9 week embryonal adrenal extract. This effect was more pronounced with the IO weeks sample (Figure 9).

Water extract: The water extracted adrenal had no effect upon P, secretion (data not shown).

Liver Efict on KG. Both alcohol/water and water extracts had no significant effect upon hCG secre- tion into the media (data not shown).

Efict on P4. Alcohol/water extracts of 9- and ro-week embryonal liver produced up to a fivefold stimulation of P, production (Figure IO). Water extracts, however, had no effect.

Protein content The protein content of the visceral tissue sediments after centrifugation of the alcohol/water and water based extracts is given in Table I. Protein content of the alcohol and water extracts supranatant were below detection limit.

DISCUSSION

Our data indicate that human embryonic visceral organs have a significant effect upon placen- tal function as reflected by hCG and P, secretion by placental explants. This effect was seen

Barnea PI al: Human Embryonal Organs Effect on Placenta 339

250 c

0 0 IO weeks

@---a 9 weeks

0’ Control

*p<o.o5 vs. Control

0.05 0.005 Adrenal extract dilution

Figure y. The stimulatory effect of alcohol/water adrenal extracts upon placental explants P, secretion. Highest effect was note at high dilutions of the extracts.

% 400 L z 1

8 x 300 -

E z ; 200 -

,o

z 100 -

0' Control

*pcO.O5 vs. control

0.05 0.005

Liver extract dilution

FIgwe IO. The stimulatory effect of alcohol/water liver extracts upon placental explants P, secretion. A significant stimulatory effect was noted with both 9 and IO weeks extracts.

340 Placenta (1989), Vol. IO

Table I. Protein content of the embryonal visceral organs used for extraction. For incubations the supranatants obtained after centrifugation 2 800 rpm for IO min were

used.

Organ Week

lung 8 lung 9 lung IO

adrenal 9 adrenal IO

adrenal II kidney 9 kidney IO

kidney II liver 9 liver IO

mg protein/ml

0.7 I., 2.6 0% I .5 0.3-4 0.8-1.1 I.,? 0.2--o. I 0.6-0.4 1.1 6.3 4.8 9.1 3.8

with very dilute solutions of the extracts, which is a strong indication of their potency. While the extract nature of the active factors obtained in the extracts is currently unknown, the method of extraction utilizing either alcohol/water or water alone to obtain crude extracts of the fetal organs could result in solutions which may contain proteins, peptides, amino acids, carbohydrates, and/or small lipids. The results obtained with the water extract of embryonic lung tissue, indicate that the involved compounds contain proteins predominantly.

Effects of the extracts on hCG and P, secretion were either concordant or divergent. Some organs have shown apparent ontogenetic changes, altering their effect on trophoblast hormone production with advancing gestational age. Since we have used trophoblast from 7-10 gesta- tional weeks, and no intra-experimental differences in effects by the various extracts were seen when using the various aged trophoblasts, the data obtained probably do not result from differ- ences in trophoblast development.

During pregnancy, hCG secretion increases almost linearly until 8 weeks, plateaus at 9-10

weeks; and thereafter declines until term. This pattern is obligatory for normal trophoblastic development. In fact, if circulating hCG levels continue to increase beyond IO weeks gestation, trophoblastic neoplasia (i.e. molar pregnancy) must be suspected (Delfs, 1957). Conversely, a premature decrease in circulating hCG level generally indicates early pregnancy failure (Aspil- laga, 1983). Cause(s) for this decline in hCG secretion is (are) not clear at present. Earlier studies suggested that the placenta functions in an independent manner during early gestation (Davies and Glasser, 1967). Our data provide indirect evidence that human embryonic visceral tissues may control trophoblastic hormonal secretion.

The microenvironment in which the trophoblast develops is comprised of the fetus, the chorion and the mother. This environment may, in fact, be responsible for modulation of pla- cental activities through sequential elaboration of various stimulatory and depressant sub- stances. These influences may act in concert with other factors secreted by the decidua and corpus luteum (Riddick et al, 1978; Csapo, Pulkkinen and Wiest, 1973). Hustin and Gaspard (1977) have shown the poor viability of trophoblast grown in explant cultures after removal from failed pregnancies with very recent fetal death. Additionally, hCG levels have been found to be very low in tubal pregnancies where the fetus has been reabsorbed and only trophoblast tissue can be identified (Schermers, 1984).

The alcohol/water extraction of fetal lung stimulated hCG production at 9 weeks and inhibited

Rlrrnea et al: Human Embryonal Organs Effect on Placenta 34’

P, secretion. The stimulatory effect on hCG secretion seen at IO weeks became less prominent, while the alcohol/water extract effect upon P, secretion became at this point stimulatory. This may indicate that ontogenetic changes take place in fetal control systems modifying placental function at the end of the first trimester. Since the fetus continuously swallows amniotic fluid, production of active factors by the contiguous fetal lung may be an efficient method for transfer of factors to the microenvironment of the trophoblast. Recent literature has shown the lung to he a rich source of various active substances (Assali, 1968).

The diluted water extract of eight week lung tissue significantly supressed hCG production, as compared to the alcohol/water extract of the 9 week tissue. Removal of compounds of 8 ooo or greater molecular weight (MW) from the water extract abolished the inhibitory effect and even caused a slight stimulatory effect (Figure 2). A possible explanation is that two separate active compounds may be present, a stimulatory which is alcohol soluble, (perhaps a small pro- tein) and a depressor substance which is water soluble and is < 8 ooo MW. Since the lung extract is completely inactivated by heating at 57”C, a process which would denature proteins, or by addition of p-chloromercurobenzoate (a proteolytic inhibitor), the protein nature of the inhibitor is probable. Even very short (pulsatile) exposure to water extracted lung has a pro- nounced inhibitory effect on pulsatile hCG secretion. This effect is not eliminated by addition of GnRHa, a known stimulator of hCG in the superfusion system (data not shown) (Barnea and Kaplan, 1989).

Kidney The alcohol/water extracted 9 week fetal kidney stimulated hCG and decreased P, production (Figures 6 and 7). At I I weeks however, it did not affect hCG but stimulated P, production. The water extracted and kidney’s effect on hCG was inhibitory. These data are compatible with the in vivo findings which report that the 1-1 I weeks gestational age coincide with the known plateau and drop seen in hCG production and with the increase in P, production. While hCG has been shown to be produced by the fetal kidney (McGregor, Kuhn and Jaffe, 1983). this did not appear to be a significant factor in this study. The kidney has not been shown to be the site of P, precursors so this would not explain the results obtained in this study. When using the fetal kidney as the source of specific stimulatory or inhibitory factors, easy access to the amniotic fluid, and the trophoblastic microenvironment, via the fetal urine, is assured.

Adrenal The alcohol/water fetal adrenal extract showed either an inhibitory effect or no effect upon hCG production, while, at the same time, stimulating production of P, at all gestational ages. This may be due to the production of increasing amounts of P, precursors such as pregnenolone sulfate. Conversely, the water extract had an obvious inhibitory effect. (Haning et al, 1983) re- ported that the fetal adrenal contains an inhibitor of hCG production, possibly DHEA. Our data are in accord with this possibility, however since the active compound(s) were water and not alcohol soluble makes the suggestion that the compound in question is a steroid unlikely. An examination of the fetal adrenal extract for P, precursors and/or for DHEA may clarify the nature of these compounds. The lack of effect upon P, secretion by the water soluble extract suggests that the stimulator may be a steroid, though a tenfold dilution of the extract further increased P, secretion, instead of reducing it as expected by a steroid precursor (Figure 9).

Liver The alcohol/water and water extracts of fetal liver had no significant effect upon hCG produc-

342 Placenta (1989), Vol. IO

tion. In contrast, alcohol/water but not water extracts studied showed a significant stimulatory effect on P, production (Figure IO). Since the liver is a rich source of cholesterol, it could pro- vide the necessary substate for P, production by the trophoblastic enzyme systems (SSCside chain cleavage). However, the earliest time during embryogenesis that the fetal liver produces cholesterol, or whether the liver could be a source for cholesterol formation already at IO weeks is unknown. The fact that alcohol extract was converted to a water-soluble extract for utiliza- tion in our study makes it less likely that cholesterol was contained in the final solution. Furthermore, the liver has been shown to serve primarily as a hemopoietic organ during this stage of embryogenesis and would be an unlikely source of lipid, peptides and other nitrogen- based compounds. In the event that the cholesterol synthesis system is not yet operative at this stage of development, a separate stimulatory factor for P, production by the trophoblast must be sought.

Our data agree with previous studies on the regulatory role of the embryo in animal models, these studies have shown that, in early pregnancy, the embryo controls various placental and luteal functions (Robertson et al, 1984; Taylor et al, 1984). This control, however, becomes less evident as gestation progresses and, as seen in fetectomy studies in the primate, the absence of the fetus during second trimester or later does not appear to impair placental function (Albrecht et al, 1979). A similar pattern could be envisaged in the human where, once embryo- genesis is completed and hCG secretion levels begins to decline and local P, secretion rises, the placenta becomes more independent. A process that would become even more prominent toward term. Stimulation of P, synthesis, with concomitant suppression of hCG release, was apparent with some of the tissue extracts studied (lung, liver and adrenal). These findings correlate with previous studies in which P, has been shown to antagonize hCG synthesis (Maruo et al, 1986). While other investigators have studied mediators which influence hCG or P, secretion by placental cultures, this study relates this effect to the human embryo; more specifically, we suggest that these placental modulators emanate from individual embryonal organs during unique periods of embryogenesis. A possible role for the embryo in control of the placenta is suggested. Further investigations will be required to substantiate this hypo- thesis. Furthermore, in patterns of changing activity, shifts from suppressant to stimulant effects were demonstrated for specific fetal organs, suggesting an ontogenetically evolved con- trol relationship between the developing fetus and the placenta during early gestation. The active components in the tissue extracts were not specifically identified during this study and remain to be elucidated in future work, although dialysis separation and heat inactivation of lung extract did suggest that they are proteinaceous. Relationship of these extracts to pre- viously demonstrated mediators of the feto-placental system, including GnRH whose effect was blocked, CAMP or epidermal growth factor (Ulloa-Aguirre, 1986), is uncertain and remains to be clarified.

SUMMARY

We investigated the effect of human first trimester fetal visceral organ extracts upon placental function, as evidenced by secretion of human chorionic gonadotropin and progesterone in explant cultures of 7-10 weeks gestational age trophoblast. Certain alcohol/water or water extracted embryonal tissues, in highly diluted solutions (I : IOC+I : 2 ooo final) had significant effects upon secretion of both hormones. The alcohol/water extract produced an opposite effect. This inhibitory effect was also seen when pulses of the water extract were added to the superfused trophoblast. Both heat inactivation at 57°C and treatment with IO~LM p-chloro-

Barnea et al: Human Embryonal Organs Effect on Placenta 343

mercurobenzoate eliminated the inhibitory effect, which suggests that the compounds in ques- tion are proteinaceous. Also dialysis with exclusion of less than 8 ooo daltons abolished the in- hibitory effect seen with the untreated water extract. The alcohol/water lung extract at 9 weeks inhibited, while at IO weeks it stimulated P, secretion. The water extracted lung had no effect upon P, secretion. The alcohol/water extract of kidney had no consistent effect upon hCG secretion. The water extract effect was inhibitory. The effect of alcohol/water extract at 9 weeks upon P, secretion was inhibitory while at I I weeks it was stimulatory. The water extract had no effect upon P, secretion. Both alcohol/water and water extracted adrenal inhibited hCG secretion. The alcohol/water extract also increased P, secretion, while the water extract had no effect. The alcohol/water and water extract liver had no effect upon hCG secretion. The effect of alcohol/water upon P, secretion was markedly stimulatory, while the water extract had no effect. ln conclusion, fetal visceral organs in very dilute concentrations have a significant effect upon placental hormonal secretion in vitro. In case of the lung, the active compound(s) appears to be a protein with a molecular weight of less than 8000 daltons. The role of the human embryo in modulating early trophoblastic function is suggested.

REFERENCES

Albrecht, E. D., Haskins, A. L. & Pepe, G. J. (1979) Th e influence of fectectomy at midgestation upon the serum concentrations of progesterone, estrone, and estradiol in baboons. Endocrinology, 107,766-779.

Aspillaga, M. O., Whittaker, P. G., Grey, C. E. & Lind, T. (1983) Endocrinological events in early pregnancy failure. AmericanJournal of Obstetrics and Gynecology, 147,903p909.

Assali, N. S. (1968) The fetus and neonate. In Biology of Gestation Vol. 2 (Ed.) Assali, N. S. pp. 25-38. London: Academic Press.

Barnea, E. R. & Kaplan M. Spontaneous, GNRH induced and progesterone inhibited pulsatile secretion of hCG placenta in vitro in the first trimester. Journal of Clinical Endocrinology and Metabolism (in press).

Barnea, E. R., Lavy, G., Fakih, H. & DeCherney, A. H. (r986a) The role of ACTH in placental steroidogenesis. Placenta, 7, 307- 3 Is.

Bamea, E. R., Oelsner, G., Benveniste, R., Romero, R. & DeCherney, A. H. (t986b) Progesterone, estradiol and alpha hCG secretion in ectopic pregnancies. Journal of Clinical Endocrinology and Metabolism, 62,529-533.

Belisle, S., Lehoux, J., Bellabarba, D., Gallo-Payet, N. & Guevin, W. (1987) Dynamics of LHRH binding to human term placental cells from normal and anencephalic gestations. Molecular and Cellular Endocrinology, 49, 197~ 203.

Castracane, V. D. & Goldzieher, J. W. (1986) Timing of the luteal-placental shift in the baboon (Papio cyno- cephalus). Endocrinology, r18,50&509.

Csapo, A. I., Pulkkinen, M. 0. & Wiest, W. G. (1973) Effects of luteectomy and progesterome replacement in early pregnant patients. AmericanJournal of Obstetrics and Gynecology, I 15,759-768.

Davies, J. & Glasser, S. R. (1967) Light and electron microscopic observations on a human placenta 2 weeks after fetal death. AmericanJournal of Obstetrics and Gynecology, 98, I I I I-I I 22.

Delfs, E. (1957) Qpantitative chorionic gonadotropin: prognostic value in hydatidiform mole and choriocarcinoma. Obstetrics and Gynecology, 9, 1-7.

Goodman, A. L. & Hodgen, G. D. (1979) Corpus luteum<onceptus-follicle relationships during the fertile cycle in rhesus monkeys: pregnancy maintenance despite early luteal removal.3ournal of Clinical Endocrinology and Meta- bolism, 49,903+09.

Haning, Jr., R. V., Choi, L., Cure& L. B., Henderson, P. A., Kiggens, A. J., Leifheit, T. L., Olson, R. W., Schwabe, M. G., Schwartz, D. B. & Traver, M. (1983) Interrelationships among human hCG, 17 beta-estradiol, progesterone, and estriol in maternal serum: evidence for an inhibitory effect of the fetal adrenal on secretion of hCG. Journal of Clinical Endocrinology and Metabolism, 56, I 188-1196.

Hertig, A. T., Rock, J. & Adams, E. C. (1956) A description of 34 human ova within the first 17 days of develop- ment. AmericanJournal of Anatomy, 98,435-439.

Hustin, J. & Gaspard, U. (1977) Comparison of histological changes seen in placental tissue cultures and in placentae obtained after fetal death. BritishJournal of Obstetrics and Gynecology, 84, zrc+216.

Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) Protein measurements with the folin phenol reagent. Journal of Biological Chemistry 193,265~269.

Maruo, T., Matsuo, H., Ohtani, T., Hoshina, M. & Mochizuki, M. (1986) Differential modulation of subunit messenger ribonucleic acid levels and hCG secretion by progesterone in normal placenta and choriocarcinoma cultured in vitro. Endocrinology, 119,855-859.

344 Placenta (r989), Vol. 10

McGregor, W. G., Kuhn, R. W. SC Jaffe, R. B. (1983) Biologically active chorionic gonadotropin: synthesis by the human fetus. Science 220,30&310.

Riddick, D. H., Luciano, A. A., Kusnik, W. F. SC Maslar, I. A. (1978) De now synthesis of prolacdin by human decidua. Life Sciences, 23, 1913-1916.

Riss, P., Mick, R. & Spona, J. (1980) First trimester serum prolactin levels in normal and complicated pregnancies. Gynecologic and Obstetric Investigation, II, t 13-t 18.

Robertson, M. C., Owens, R. E., McCoshen, J. A. & Friesen, H. C. (1984) Ovarian factor inhibit and fetal factors stimulate the secretion of rat placental lactogen. Endocrinology, 1q,22-28.

Schenners, J. P. (1984) Ertopic pregnancy: A morphologtc and endocrine s&y. pp. r-87. Enoffsetdrukkenj los- Naarden, Amsterdam: Bock.

Taylor, M. J., Jenkin, G., Robinson, J. S. & Friesen, H. C. (1984) Effect of intrauterine death and fetectomy on ovine placental lactogen production. Research Veterinary Sciences, 35,22-3 I.

Ulloa-Aguirre, A., August, A. M., Golos, T. G., Kao, L., Sakuragi, N., Klinman, H. J. & Strauss III, J. F. (1986) 8-bromo-adenosine 3, 5 monophosphate regulates expression of chorionic gonadotropin and fibroectin in human cytotrophoblasts.3ournal of Clinical Endocrinology and Metabolism, 64, 1002~-1009.