OOl~~-i~~7/92/1~111-O~~7SO~.OO/O Enducrmology Copyright (c‘ 1992 by The Endocrme

Vol. 131, No. 1 Printed in U.S.A.

Soaety

The Rat Placenta from the Mother Thyroid Status*

and the Transfer of Thyroid Hormones to the Fetus. Effects of Maternal

ROSA CALVO, MARIA JEStiS OBREGON, GABRIELLA MORREALE DE ESCOBAR Unidad de Endocrinologia Investigaciones Cientificas

FRANCISCO

ESCOBAR

Molecular, Instituto de Investigaciones Biomidicas, and Facultad de Medicina, Universidad Autbnoma

DEL REY,

AND

Consejo Superior de de Madrid, Madrid, Spain

ABSTRACT We have studied the effects of maternal thyroid status on the effectiveness of the rat placenta near term as a barrier for the transfer of T, and T:, to the fetus. Dams were given methimazole to minimize the fetal contribution to the T, and T:r pools, so that the iodothyronines found in the conceptus are ultimately of maternal origin. The dams were infused with saline, or with T, or TI1 at doses ranging from 2.327.8 nmol Tq and from 0.77-20.7 nmol T:,/lOO g BW per day. A group of normal pregnant dams (C) was included. At 21 days of gestation T,, T:,, and rT:$ were measured by RIA in maternal and fetal plasma, and in maternal and fetal sides of the placenta. The total fetal extrathyroidal Tq and T:, pools were also determined. The dose-related changes

in T,, T:,, and rT:% levels in the placenta confirm the presence of both inner and outer ring iodothyronine deiodinase activities, and suggest increasing accumulation of the iodothyronines. Despite this, fetal extrathyroidal T, and TR increase progressively in T4-infused groups as a function of maternal circulating Tq levels. Fetal extrathyroidal TR increases progressively in Trl-infused groups as a function of maternal plasma T:?. There was no evidence that the net maternal contribution of T, or T, would be proportionally less when the maternal pools became very high. It was concluded that the rat placenta is only a limited barrier for the transfer of T, and T:! to the fetus. (Endocrinology 131: 357-365, 1992)

T

effective barrier, preventing maternal iodothyronines from reaching the fetus. This would be a consequence both of poor permeability of the fetal membranes and of active deiodinating mechanism(s) (2, 6, 10, 13, 14). Despite the existence of such mechanism(s),however, some maternal T4 and T3 are transferred from the mother to the fetus, before and after onset of fetal thyroid function, both in rat and man (15-23). In those caseswhere the fetal thyroid is impaired, transfer of maternal thyroxine plays a crucial role in the protection of the fetal brain from TX deficiency up to birth (19, 23). It appeared of interest to evaluate the role of the placenta near term as a barrier to the thyroid hormones, and the possible effects of the maternal thyroid status on its effectiveness. The placenta might well act as a more effective barrier when maternal circulating levels of T4 and T3 become too high. Net maternal transfer was assessedby measuring the amounts of T4 and T3 found in the placenta and fetus, after minimizing the fetal contribution to these pools by blocking fetal thyroid function. For this we used dams given methimazole (1-methyl-mercapto-imidazole-2-thiol, MMI), an antithyroid drug which crossesthe placenta (24) and blocks both fetal and maternal thyroid function, and MMItreated dams on constant infusions of T4 or T3, as well as a group of age-paired normal pregnant rats. Three different experiments were performed to assessthe reproducibility of the results and to study the effects of a wide range of T4 and T3 infusion doses. The concentrations of T4 and T3 were measured in the maternal and fetal sides of the placenta, obtained near term, and compared with changesin maternal and fetal plasma and in the fetal extrathyroidal T4 and T3

here is abundant evidence that the placenta is an active site for the deiodination of both T3 and Tq. The placentas of the rat, guinea pig, and man have high tyrosyl ring deiodinating (5 D) activities (l-lo), T3 being the preferred substrateover Tq, and the latter over rT,. It has been proposed (i.e. Ref. 10) that T3 is rapidly cleared and deiodinated by the fetal side of the placenta, that is, by the layers of the placental disc away from the endometrium. This would contribute to the low T3levels in the fetal circulation, whereas deiodination of T4 to rT3 would contribute to the high rTj levels. Placental 5 D activity is not affected by thyroid dysfunction or fasting (9). Human and rat placentas also have type II 5’ (phenolic ring) deiodinase(5’ D) activity (1, 11, 12), generating T3 from T.,. This activity was found in the maternal side of the placenta, that is, the basal placental disc adjacent to the endometrium. It increasesin situations of T4 deprivation, and could potentially defend intracellular and/or circulating T3 pools in situations of mild to moderate hypothyroxinemia (12). The net outcome of the different placental deiodinating activities on the amounts of maternal T, and T3 “sequestered” by the placenta and those actually transferred to the fetus have not, however, been assesseddirectly in vivo. It has been proposed that the placenta acts as a very Received November 18, 1991. Address all correspondence and requests for reprints to: Dr. G. Morreale de Escobar, Unidad de Endocrinoloaia Molecular, Institute de Investigaciones BiomPdicas, Facultad de Me&ina, Arzobispo Morcillo 4, 28029.Madrid, Spain. * This work has been carried out under a research grant (PM85-0005) from Direccibn General Interministerial de Ciencia y Tecnologia (DGICYT, Spain).

357

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 July 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.

THYROID

358

STATUS

AND

pools. In some experiments the concentration of rT3 was also measured. As will be seen, results show that the rat placenta is not a complete barrier for T4 and TX, and does not protect the fetus from excessive transfer of both iodothyronines when these reach supraphysiological levels in the maternal circulation.

Materials Experimental

and Methods

design

Female rats of a Wistar strain were used. They were mated with n Irma1 males, the day of appearance of a vaginal plug being considered as day 0 of gestation (17). They were housed, treated, and killed under recommended humane conditions. Three different experiments were performed over a 3-yr period, which included transfer of the rat breeding colony to another building with new housing conditions. Exp A. The pregnant dams were divided into eight different groups comprising 3 dams each: seven of the eight groups were given 0.02% MM1 as drinking water, starting on the morning of the 14th day of gestation (MM1 dams). The 8th group (C dams) did not receive MMI. At 15 days of gestation, all the dams were implanted with osmotic minipumps delivering at a constant rate either infusion solvent (dams of the C group and of one MM1 group), T, (three MM1 + T, groups) or T, (three MM1 + T3 groups). The T, doses were 2.32, 3.09, and 4.63 nmol T, (1.8, 2.4, and 3.6 fig Tq, respectively)/100 g body weight (BW) per day; the T3 doses were 0.77, 2.30, and 6.90 nmol T3 (0.5, 1.5, and 4.5 pg T3, respectively)/100 g BW per day. The T4 and T3 doses indicated above are referred to 100 g BW at 15 days of gestation, when the pumps were prepared and implanted. The amount of hormone delivered daily remained constant until the end of the experiment, but the dose per 100 g BW was decreasing progressively, as the weight of the dams plus concepta increased from 193 + 5 to 275 f 7 g between 15 and 21 days of gestation. Thus, the doses received by the dams at 21 days of gestation, expressed per 100 g final BW, were 70% of those reported above. The dams were killed at 21 days of gestation, and perfused as previously described in detail (17, 19). We used a total of 24 dams and their 223 fetuses. The mean number of fetuses per litter for Exp A was 9.7. Exp B. To confirm the main findings from Exp A, a second experiment was carried out involving one group of C and one of MMI-treated dams, and MMI-treated dams receiving three different doses of T, and one dose of T3. The doses were 3.09, 6.18, and 9.27 nmol T, (2.4, 4.8, and 7.2 rg, respectively), and 2.30 nmol T3 (1.5 fig)/100 g BW per day, as determined at 15 days of gestation, when the dams weighed 250 + 8 g. At 21 days of gestation they weighed 335 + 5 g, and therefore the doses received by the dams at 21 days of gestation, expressed per 100 g final BW, were 70% of those reported above. A total of 22 dams and their 243 fetuses were used, with the mean number of fetuses per dam being 11. Exp C. A third experiment was performed to study the effects of higher T, and T) infusion doses than used for Exp A and B. This experiment involved the use of one group of C and one of MMI-treated dams, of four groups of MMI-treated dams receiving different doses of T4 and three groups receiving T3. The doses were 3.09, 6.18, 9.27, and 27.80 nmol T, (2.4, 4.8, 7.2, and 21.6 pg, respectively)/100 g BW per day, and 2.30, 6.90, and 20.70 nmol T3 (1.5, 4.5, 13.5 pg)/lOO g BW per day, as determined at 15 days of gestation, when the dams weighed 278 f 3 g. At 21 days of gestation they weighed 316 + 5 g. A total of 30 dams and their 324 fetuses were used, with the mean number of fetuses per litter being 10.8.

Continuous

infusion

of T4 and T,,

We used Alzet 2ML2 osmotic minipumps (Alza, Palo Alto, CA), implanted under the dorsal skin (19). The pumps delivered 5 @l/h for 14 days. The minipumps implanted into the C and MM1 dams contained only the infusion solvent, namely phosphosaline buffer, pH = 7.4, containing 5% serum from thyroidectomized rats. Tq or T3 in free acid form (Sigma Chemical Co., St. Louis, MO) were dissolved in a small

PLACENTAL

Tq, TY, rT:i

volume of 0.05 N NaOH and then diluted previously described in detail (19, 20).

Preparations

with

infusion

Endo. Vol131.

1992 No 1

solvent

as

of samples

Maternal plasma, liver, heart, and brain were excised. The fetuses were bled, separated from the placenta, and immediately placed on ice. The fetal (F) and maternal (M) sides of each placenta (F-placenta and M-placenta, respectively) were separated with blunt forceps, and stored frozen for the present study. We have here defined the maternal side of the placenta as the placental disc adjacent to the endometrium and the fetal placenta as the labyrinthine placenta, namely the layers of the placental disc away from the endometrium. The thyroid, adhering to the trachea, was obtained under the dissecting microscope, and kept frozen. The liver, brain, heart, lung, and interscapular brown adipose tissue pads were dissected out. These tissues and the rest of the fetus (carcass) were weighed, frozen rapidly on dry ice, and used for the determination of T, and T3 concentrations in parallel studies (23, 25).

Determination

of T4, T:,, and rT:, concentrations

Two to three F- or M-placentas were pooled to obtain one value of the T, or T, concentration. Six such pools, obtained equally from different dams, were processed for each experimental groups. All samples corresponding to the same type of tissue (i.e. F- or M-placenta) were extracted and processed in the same analytical run, as described elsewhere in detail (19, 20, 23). The maternal plasmas (M-plasmas) were processed individually, the fetal plasmas (F-plasmas) were pooled. In brief, homogenization in methanol is followed by extraction in chloroform-methanol, back-extraction into an aqueous phase, and purification of this phase through Bio-Rad (Richmond, CA) AG 1 X 2 resin columns. The iodothyronines were eluted with 70% acetic acid, which is then evaporated to dryness. RIA buffer is added, and the samples are submitted to highly sensitive RIAs for the determination of T, and T3, the limits of sensitivity being 3.2 fmol T4 and 2.3 fmol TJtube. Crossreactivities for the T, RIA and T) RIA were as recently reported (26). Each sample is processed in duplicate or triplicate at two or more dilutions. Results are then calculated using individual recovery data obtained after addition of [‘3’I]Td and [“‘IIT during the initial homogenization process. The amounts of tracers added are such that the radioactivities carried over into the RIA tubes are too low to interfere with the determinations, representing less than 2.5% of the radioactivity added as labeled antigen. For the determination of rT3, M- and F-placentas were extracted after addition of [‘251]rT3 as recovery tracer. rT, was measured in M- and Fplasma directly, as described by van der Heide et al. (27), using 50 and 25 ~1 samples, respectively, in duplicate. The antibody used was a highly specific one, kindly supplied by Dr. D. Van der Heide (Department of Physiology, University of Wageningen, The Netherlands), and used at a final dilution of 1:20,000. High specific activity [‘3’I]T,, [‘*‘I]T,, [1251]T3, and [“‘I]rT, (3000 lCi/ fig) were synthesized in our laboratory, as already described (15, 17, 19, 20), using 13’1 or ‘*‘I from Amersham International (Amersham, England) or New England Nuclear (Boston, MA) and TX, 3, 5-T> or 3, 3’.Tz as substrates, respectively. They were used as recovery tracers and/or labeled antigens in the highly specific RIAs indicated above.

Total amounts

of T4, T:,, and rT:,

The total pools of T, or T3 in the M- and F-placenta were calculated from the mean concentrations and the mean weights of the tissue. The total extrathyroidal T, and T) pools were calculated by addition of the contents in each organ (liver, lung, brain, brown adipose tissue) plus the carcass and blood contents, with the total amounts of iodothyronines in each organ being calculated from the mean T, or T3 concentration and the mean organ weight. For the blood, mean plasma T, and T, concentrations were used, a 50% hematocrit value being assumed and a blood volume equal to 16.8% of the body weight (28).

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 July 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.

STATUS

THYROID

AND

PLACENTAL

T.,, Tzi, rT:,

EXPERIMENT M-plasma

50

359

A M- placenta

F-plasma

F-DlaCenta

40 30 20

1 rn z

10 0

0

0

a

Dams:

c

b

c

a

b

0

c

MMI

c

T4 -MM,

-

T3

T4

T4

T3

-MM,

-MhlI

-

MMI

a

T3

T4 -MM,

-

T3 -

FIG. 1. Exp A. Concentrations of T, and TZi in M- and F-plasma and in M- and F-placentas obtained at 21 days of gestation from normal (C) and MMI-treated (MMI) dams, as well as from MMI-treated dams on constant infusions of T, or T+ Bars represent mean values (~sEM). The a, b, and c doses of T, were 2.32, 3.09, and 4.63 nmol (1.8, 2.4, and 3.6 pg T,, respectively)/100 g BW per day; the a, b, and c doses of TX were 0.77, 2.30, and 6.90 nmol (0.5, 1.5 and 4.5 pg T:,, respectively)/100 g BW per day, referred to the BW at 15 days of gestation and 70% of this value at 21 days. The horizontal dashed lines correspond to the height of the bars for the C group. Asterisks on top of the data bars identify the statistically significant differences with respect to the C group.

TABLE

1. Exp C: mean (k SEM) values of the concentrations Concentration Dose of T, or T:, nmo1/100 g BW

Groups Ch MM1 MM1 MM1 MM1 MM1 MM1 MM1 MM1

+ + + + + + +

T,-b T,-d T,-e T,-f T:,-b T:,-c Ta-d

3.09 6.18 9.27 27.80 2.30 6.90 20.71

(T,) (T,) (T,) (T,) (Tz) (T:,) (Tn)

(2.69) (5.38) (8.07) (24.19) (2.00) (6.18) (27.80)

M-plasma (nmol/l) 19.9 3.6 37.3 38.7 52.8 81.1 2.2 2.2 2.7

f 1.2 f 0.5’ Z?Z4.1’ f 2.7’ + 5.8’ t 27.3’ + 0.1’ f 0.1’ + 0.1’

M-placenta (Pmolk!) 16.3 5.5 33.0 52.4 66.8 94.3 2.3 2.3 2.1

f + + + + k k f +

2.2 0.3’ 3.1’ 5.9’ 26.2’ 14.2’ 0.6’ 0.2 0.2’

of T, and T:, in the M-

and F-plasma

and in the M and F sides of the placenta

of T,

Concentration

F-placenta (Pmwd 5.2 1.9 13.1 17.9 19.5 44.8 0.9 0.9 0.8

t + + + + + + f f

0.2 0.2 1.4’ 0.7’ 0.6” 4.6’ 0.1’ 0.1’ 0.1”

F-plasma (nmolfl) 8.2 0.8 3.8 4.4 5.3 13.0 0.36 0.41 0.43

+ f + f + + + k +

0.6 0.1’ 0.4’ 0.2’ 0.5’ 1.5’ 0.01’ 0.01’ 0.03’

M-plasma (nmol/l) 0.92 0.35 0.95 0.92 1.53 4.15 2.69 6.49 10.65

+ + f + + + + + +

M-placenta (pmolk)

0.09 0.07’ 0.10 0.17 0.26 0.96 0.17’ 2.10’ 2.92’

0.68 0.38 0.77 1.37 1.93 2.72 1.11 3.90 9.14

+ + + + f + + + f

0.03 0.03’ 0.12’ 0.11’ 0.10’ 0.21’ 0.07’ 0.30” 0.46’

of T:% F-placenta (pmolk)

0.29 0.10 0.30 0.38 0.56 1.45 0.95 5.01 9.25

a The infusion doses are given as the daily amount of T, or T:l/lOO g BW at 15 days of gestation. The numbers given to the daily dose at 21 days of gestation (see Exp C, Materials and Methods). * The circulating levels in age-paired nonpregnant normal rats were 49.6 + 3.2 nmol T,/l and 1.41 + 0.12 nmol TJ. ’ Identifies statistically significant differences (P < 0.05) with respect to the C group.

Statistical

analysis

Mean values (&EM) are given. Data from the eight (Exp A), six (Exp B), or nine (Exp C) groups were submitted to one-way analysis of variance, after testing for homogeneity of variance using Bartlett’s procedure for groups of unequal size. Square root or logarithmic transformations usually ensured homogeneity of variance when this was not found with the raw data. Significance of differences between groups was assessed using the protected least significant difference test, and considered significant when P is less than 0.05. All these calculations were performed as described by Snedecor and Cochran (29). Whenever it is stated that the TI, T3, or rT) concentration in a tissue of a given group is increased, or decreased, as compared to another group, it is implied that the difference between the mean values was statistically significant.

Results None of the treatments (MM1 for 7 days, with or without concomitant infusion of T4 or T3) altered the number of

+ f + f f + k k -t

0.01 0.01’ 0.03 0.02’ 0.06’ 0.05’ 0.07’ 0.45’ 1.00’

F-plasma (nmol/l) 0.097 0.051 0.071 0.074 0.102 0.198 0.096 0.295 0.742

in parentheses

f 0.006 + 0.004’ f 0.005 f 0.007 Z!Z0.007 + 0.009’ f 0.011 t 0.033’ + 0.063’ correspond

fetuses per litter. No treatment-dependent effects were served in the weight of the placentas and fetuses. Effects of treatments

on the concentrations

ob-

of T4 and Tii

The changes in the concentrations of T, and T3 in the Mand F-plasma and in the M- and F-sides of the placenta are shown in Figs. 1 and 2 for Exp A and B, and in Table 1 for Exp C. Figure 2 also shows the corresponding changes in rT3 levels. Treatment of the pregnant dams with MM1 resulted in decreased concentrations of T, and T3 in all samples, whether M or F, as expected from previous experiments (19, 20, 23). The infusion with increasing doses of T, resulted in doserelated increases of both the T., and T3 concentrations in Mand F-plasma and in M- and F-placentas. The T, and T3 levels continued to increase, although not always linearly,

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 July 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.

THYROID

360

EXPERIMENT M-plasma

F-Plasma

STATUS AND PLACENTAL

T4, Ty3,rT:,

F-placenta

z/‘j.rzz-J

0

1992 No 1

EXPERIMENT C fTT?Egg

B M-placenta

Endo. Vol131.

10

T4 dose

30 0

20

infused

into

dams

: nmols

10

/ 100

20

30

g BW /day

0.8 06 0.4 02 0

c MM,

bde 5 --MMC

b T3

0

5 T3 --MMC

FIG. 2. Exp B. Concentrations of T,, T:r, and rT:i in M- and F-plasma, and in the M- and F-placentas obtained at 21 days of gestation from normal (C) and from MMI-treated (MMI) dams, as well as from MMItreated dams on constant infusions of T, or Ta. Bars represent mean values (SEM). The b, d, and e doses of Tq were 3.09, 6.18, and 9.27 nmol (2.4, 4.8, and 7.2 pg T,, respectively)/100 g BW per day; the b dose of T:, was 2.30 nmol (1.5 fig T3)/100 g BW per day, referred to the BW at 15 days of gestation. The doses at 21 days of gestation are 70% of these values. The horizontal lines and asterisks have the meaning indicated in the legend to Fig. 1.

even with the highest T, infusion doses (Exp C), which raised Tq and T3 levels in the M-plasma 4-fold above the levels found in the normal (C) dams. When increasing doses of T3 were infused into the dams, the concentration of T3 increased progressively in all samples, even at the highest infusion dose (Exp C), which increased T3 in M-plasma 12-fold above C levels. On the contrary, plasma and placental levels of Tq either remained unchanged with respect to the values found for the MM1 group, or decreased further. The changes observed for the concentrations of T., and T3 were not of equal magnitude in different samples, or for each iodothyronine. This is illustrated for Exp C in Fig. 3. In order to compare the degree of the changes, the T4 and T3 concentrations in the M- or F-plasma, M- or F-placenta were referred to the respective values obtained in samples from C animals. This was found necessary in order to compare the slopes of the dose-response curves, because the concentrations of each iodothyronine were quite different in M- and F-plasma, or M- and F-placenta. The two upper panels show the changes in the concentration of T4 in samples from the T,-infused dams, the intermediate panels the concentration of T3 in samples from the same animals, and the two bottom panels the concentration of T3 in samples from T3-infused dams.’ The slopes of the dose-response curves for T4 and T3 in Mplasma are steeper than for F-plasma: the highest doses of T, and T3 raised the T4 and T3 levels in M-plasma 4- and 12fold, as compared to those of C dams, whereas the levels in F-plasma increased only 2- and 7-fold, respectively. The slopes of the dose-response curves for F-plasma T, and T3 were about half those for M-plasma in Exp A and B as well. The slopes in the M- and F-placenta were either the same as ’ The changes of the T4 levels in samples from the T,-infused dams are not shown. As already indicated, they either were the same as those from the saline-infused MMI-treated dams, or lower.

0

IO

20

T4 dose

10

infused

20

into

dams

30

: nmols

0

/ 100

g BW i day

T3 dose

infused

into

dams

: nmols

/ 100 g BW i day

30

FIG. 3. Exp C. This figure shows the degree of the changes in the concentration of T, or T:! in M- and F-plasma (left panels) and in Mand F-placenta (right panels), as a function of the dose of T, or of TZi infused into the dams. To compare the degree of change for each iodothyronine in different M and F samples, data were transformed into percentages of the corresponding value for C rats (open square and horizontal dashed lines). The closed circles shown on the vertical axis (at a dose = 0) correspond to the MM1 group. Vertical bars show the SEM, not visible if smaller than the size of the symbol used for the mean value.

those in M-plasma, or steeper. The sharpest increase was found for the concentration of T3 in the F-placenta of T3infused dams. T4 and T,%concentration

gradients

The T4 and T3 concentrations in the F-placenta of Tqinfused dams were much lower than in M-placenta and Mplasma, being reduced to 20-25% of the levels in M-plasma; they decreased further in F-plasma, to 6-9% of the levels in M-plasma. In the T3-infused dams, the concentrations of T3 in the F-placenta were reduced to 60-75% of levels in Mplacenta, and decreased further in F-plasma to 6-10% of T3 levels in M-plasma. Thus, in the T,-infused dams there is a steep concentration gradient for T4 between the M and F sides of the placenta, and a further steep gradient between the F-placenta and Fplasma. The major concentration gradient for T3 was not between the M and F sides of the placenta, but between the F-placenta and the F-plasma. Net maternal pools

contribution

to fetal and placental

T4 and Ti3

The amounts of T, and T3 found in M- and F-placentas, and the mean fetal extrathyroidal pools of T4 and T3 calcu-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 July 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.

THYROID

STATUS

AND

lated as described in Materials and Methods. The fetal extrathyroidal and the F-placental T, pool are of combined maternal and fetal origin in the C animals, that of the Mplacenta being entirely of maternal origin (22). In fetuses from MMI-treated dams, however, the contribution by the fetal thyroid is negligible (22) and the iodothyronine pools in the conceptus are entirely of maternal origin. Thus, if data from C fetuses and placentas are excluded, the total Tq and T3 pools represent the net maternal contribution per fetus or individual placenta: T, would be derived directly as such from the maternal circulation. T3 could be derived from the mother either directly as such from the maternal circulation, or could be generated locally from T, of maternal origin. The data for Exp C are shown in Fig. 4, plotted against the Tq or T3 levels in the M-plasma. As may be seen (panels A, B, and C, respectively), the amounts of T, of maternal origin found in the fetuses and placentas increased in proportion to the maternal thyroxinemia. The three lower panels show the corresponding results for T3. The net maternal contribution to fetal T3 (panel D) and to F-placental T3 (panel F) increased in proportion to the maternal T3 levels, and were closely fitted by the same functions, whether the dams were infused with T3 or with Tq. For M-placentas (panel E), however, the data fall on curves which are distinct for dams

EXPERIMENT fetus

0

20

40

60

C

M-placenta

80 100

0

20

40

60

80 100

F-placenta

0

20

40

60

80 100

Maternal plasma Tq(nmols / I)

0

2

4

6

81012

0

2

4

6

8 1012

0

2

4

6

81012

Maternal plasma T3(nmols / I) l

MMI

0 MMI + T4

0 MMI + T3

4. Exp C. The three upper panels show the net maternal contribution of T, to a single fetus, M-placenta and F-placenta of a litter, as a function of the T, levels in the maternal circulation, with the mean number of fetuses per litter being 10.8. The three lower panels show the corresponding data for T:s, as a function of the maternal circulating T:%levels. Data for the MMI-treated group are shown as filled circles, MMI-treated dams on T4 as open circles and MMI-treated dams on T, as open squares. The asterisk in panel A corresponds to the amount of Tq contributed by the mother to C fetuses, calculated as 18% (22) of the fetal extrathyroidal T, pool, which was 14.54 pmol T,/fetus. The horizontal and uertical dashed lines correspond to the concentrations of Tq or Tgi in fetuses and dams from the C group, the arrow indicates the circulating T, or Ta levels in the plasma from the age-paired nonpregnant normal rats. Data in panels A, B, and C are closely fitted by linear or quadratic polynomial functions (? = 0.993, f2 = 0.983, and r2 = 0.983, respectively). So are data in panels D and F (r’ = 0.989 and r2 = 0.986, respectively). In panel E data corresponding to T3-infused dams are fitted by a quadratic polynomial function (r* = 0.998), whereas those for the TI-treated groups are fitted by a logarithmic function (r* = 0.921).

PLACENTAL

T,, T:i, rT:,

361

infused with T4 as compared to those infused with T3: at a similar level of T3 circulating in the M-plasma, the amount of T3 found in M-placentas is clearly higher when these are obtained from T,-infused dams as compared to M-placentas from T3-infused mothers. These differences tend to decrease at the higher levels of T3 in M-plasma. Although not shown, findings for Exp A were quite superimposable and those of Exp B quite similar. Comparison

of results from Exp A, B, and C

The net amounts of T, and T3 contributed by the mother to the fetus are plotted in Fig. 5 against the maternal circulating levels of T4 and TX. There were unexplained differences in the concentrations of the iodothyronines between the C groups from Exp A, 8, and C.2 After referring the T, and T3 concentrations in different samples to the respective C values for each experiment, however, the data from all three experiments were fitted by the same quadratic functions. There was a progressive increase in fetal extrathyroidal T, and T3 pools as the M-plasma T, and T3 levels increased. There was no evidence that the fetal T, and T3 pools were nearing a plateau. On the contrary, the maternal contribution to the fetal extrathyroidal T4 and T3 pools appeared to increase proportionally more as the maternal circulating levels reached the higher values. Changes in the concentrations

of rT:,

The concentrations of rT3 in the M- and F-plasma and in the M- and F-sides of the placenta are shown in Fig. 2 for Exp B. In the C animals, the rT, levels in M-plasma were much lower than those of TQ, and about half those of TX. On the contrary, rTj levels in the M- and F-placenta were higher than the T3 concentrations, although lower than the placental T4 levels. The rT3 levels in the fetal plasma were also higher than the T3 levels, and actually higher than the maternal circulating rT3 concentrations, contrary to findings for the circulating levels of T4 and TJ. In the T,-infused animals, the rT3 level was l-2% of the T4 level in M-plasma, 22-25% in the M-placenta, 20-50% in the F-placenta, 27-40% in the Fplasma. The molar ratios of rT, to T4 in the placenta are about 5 times higher than the T3 to Tq ratios. The rT3 concentrations in M- and F-plasma and M- and F-placentas increased with increasing infusion doses of Tl, as described also for the concentrations of Tq. On the contrary, when TX was infused the concentrations of rTj were lower than those of the saline-infused animals; in these animals, Tq levels decreased as well. Similar results were obtained for the TJinfused animals of Exp C (data are not shown). The Tq concentrations in the M- and F-placenta increased with increasing T4 infusion doses. The rTg concentrations, on the contrary, reached a plateau at M-plasma T4 levels between 20-30 nmol/l. This is already apparent for the F’ Other investigators have also reported differences in circulating Tq and TZ levels between experiments performed with pregnant rats at 21 days of gestation. Thus, El Zaheri et al. (30) reported in the same study mean values for plasma Tq in normal pregnant rats ranging from 1.8 + 0.3 to 3.6 f 0.3 Fg/dl, and for plasma T, ranging from 25.5 + 1.8 to 87.4 t 4.2 ng/dl.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 July 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.

THYROID

362

STATUS

AND

PLACENTAL

EXPERIMENTS

0

100 200 300 400 M-PLASMA T4 as % Of c value

500

0

100

200 300 400 M-PLASMA T4 as % Of c value

Endo. 1992 Vol 131 *No 1

Tq, T:i, rT:,

A, B and C

500

0

400 600 M-PLASMA T3 as % Of c value

1200

FIG. 5. The fetal extrathyroidal T, and T:, pools, as determined for Exp A, B, and C, are plotted against the maternal circulating levels of each iodothyronine. Data from the individual experiments are transformed into percentages of the respective C values. The asterisk indicates the maternal contribution to fetal extrathyroidal T, for C animals, calculated as 18% of the total fetal extrathyroidal T, pool (22). The horizontal and vertical dashed lines correspond to the T, or Tii levels which are found in fetuses and dams from the C group, the arrow indicates the circulating T, or T:, levels in the plasma from age-paired nonpregnant normal rats from Exp C. Data shown in panel I are from the MMI, MM1 + T,, and MM + T:, groups and represent the direct maternal T, contribution to fetal Tq. Data in panel II are from the MM1 and MM1 + Tq groups and correspond to the fetal extrathyroidal T:, generated from maternal T,, either in the fetus or in the mother. Panel III corresponds to the data from MM1 and MM1 + T:, groups and represents the direct maternal contribution of T:, to the fetus. Data from the three experiments were fitted by the same quadratic polinomial functions (r2 = 0.956, r’ = 0.935, and r’ = 0.995, respectively, for panels I, II, and III).

placenta with the highest dose of T, used for Exp B (see Fig. 2), and became apparent for the M-placenta when results from Exp C were considered (not shown). Even with the highest T4 dose used for the latter experiment, M-placenta rTj concentrations did not exceed those shown in Fig. 2 for Exp B. The slopes of the dose response curves for the Tq and rTj levels in M- and F-plasma and in M- and F-placenta are shown in Fig. 6, where data are plotted as in Fig 3 for Exp C. This illustrates that the slope for the increase in rT3 levels in F-plasma is about twice that for the M-plasma, in contrast to the slopes for circulating T4 and T3 where the opposite occurs (see Fig. 3). It also illustrates that at the two lower T, infusion doses the increase in rTJ concentrations both in the M- and F-placenta are not only much greater than for the M- and F-plasma, but also greater than for placental T4. At higher infusion doses this difference decreases. The changes in T3 (not shown) were similar to those reported for Exp A and C.

EXPERIMENT 600 , 0 M-Plasma l F-Plasma

9

: 400 I

B

1200 1000 1 I

A M-Placenta . F-Placenta

8001

I

!fj==-J;-~ 0

2

T4 dose

4

6

infused

8

10

into

dams

0 : nmols

2

4 / 100

6

S

10

g BW i day

FIG. 6. Exp B. This figure shows the degree of the changes in the concentration of T, or rTn in M- and F-plasma (left panels) and in Mand F-placenta (right panels), as a function of the dose of T, or T3 infused into the dams. Transformation of the concentration values and meaning of symbols, vertical bars, and dashed lines are the same as in Fig. 3.

Discussion The aim of the present study was to assess the effectiveness of the rat placenta near term as a barrier to the materno-fetal transfer of the natural iodothyronines, T, and T?, and the possible effects of maternal thyroid status. The experimental approach we have used for this involves the use of a goitrogen, MMI, to block fetal thyroid function, and the infusion into the dams of either T, or TX. The fetal contribution to fetal and placental T4 and T3 pools would be minimized, and any thyroid hormone found in these compartments would be maternal in origin. The effectiveness of MM1 in this respect has been repeatedly confirmed by determining the total T4 and T3 contents in the fetal thyroid (19, 20, 23). The use of MM1 might, however, affect the results per se. This does not appear to be likely: MM1 treatment of the dams did not consistently affect 5 D activity of the placenta in five studies (9). The net maternal contribution of T4 in normal rats was similar to that transferred in MMI-treated dams on a physiological dose of T, (22). The T, contributed to the extrathyroidal T4 pool of the C fetuses from Exp A and C would be 2.59 and 2.60 pmol/fetus, respectively, considering that at 21 days of gestation 18% of the extrathyroidal T, pool of normal fetuses is of maternal origin (22). These values were fitted by the same curves as the data from dams receiving MMI. Thus, the present experimental model appears relevant for the maternal transfer of T4, irrespective of treatment with MMI. We have extrapolated this conclusion to the transfer of T3. The T4 found in placentas and fetuses from MM1 dams infused with T4 is mostly derived from the mother. The T3 would also ultimately be of maternal origin, although with the present approach it is not possible to define how much T3 is generated from T4 and how much is derived directly from the maternal thyroid. Secretion of T3 by the maternal thyroid is likely to be negligible in MMI-treated dams infused with T4 or TX, as their plasma TSH is suppressed (20, 30). Thus, most of the T3 found in the placentas and fetuses from T4-infused dams would appear to be generated from Tl, although we do not know how much T3 is generated in the

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 July 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.

THYROID

STATUS

AND

maternal compartment, and then transferred to the conceptus, and how much is generated in the different compartments of the conceptus itself. In the case of TX-infused dams, practically all of the T3 found in the placenta and fetuses would be derived as such from the maternal compartment, as T, levels were even lower than for saline-infused animals, and generation of T3 from T, would contribute minimally to the findings. The amounts of maternal T, and T3 which are found in the fetus are related to the respective maternal Tq and T, levels, although the relationship is not lineal, and is better fitted by a quadratic polynomial function. There is no evidence that at supraphysiological maternal T, and T3 levels the net contribution to fetal pools would increase proportionally less. If anything, it appears to increase more than the maternal circulating T, and T3 levels, although we cannot exclude that the shape of the curves might change at higher doses than those used for the present study. Present results confirm that transfer of T4 from the mother to the fetus is not free, but limited (19, 20, 22, 23). The dose of T, required to normalize the extrathyroidal T, and T3 pools of the fetus as a whole3 increases maternal plasma T, to a level which is 35fold that of pregnant rats (C group), and 1.5-fold that of nonpregnant adult females (31). In groups infused with T3 a similar relationship exists between the extrathyroidal fetal T3 pool and the M-plasma T3. These T3infused dams and their fetuses are far from normal, as their T, pools are very low. Their tissues are thus deprived of regulatory mechanisms involving local generation of T3 from T‘l. The changes in the placenta suggest that T, and T3 accumulate in this organ, either in the M or F sides of the placenta, or in both, and to a greater degree than would be expected from the increase in M- and F-plasma Tq and T3. This suggests either a high and increasing uptake of T., and T3 from the maternal circulation, or other as yet undefined changes leading to an increasing accumulation of T4 and T3 in the placenta. Whichever the mechanism involved, it would obviously decrease the amounts of T, and T3 reaching the fetus. The rTJ levels in the M- and F-plasma of C animals confirm those reported by El Zaheri et al. (30), who described higher rT3 levels in the fetal as compared to the maternal circulation and found that the rT3 in the rat amniotic fluid was dependent on the maternal T4 supply. The present results show that the rT3 level in the placenta is related to the maternal T4 supply as well. A similar correlation between placental rTg and maternal T4 has been described by Yoshida et al. (32) in women near delivery. Our present data show that at lower Tq infusion doses, the increase in rT3 in the M- and F-placenta is greater than expected from the change in the M-plasma T, levels, suggesting a very high uptake of rT3 from the maternal plasma and/or local generation of rT3 from Tq in the placenta. The latter is more likely, considering that high 5 D activities 3 This does not mean that T, and TS levels in all individual fetal tissues reach normal (C) levels with the same dose of Tq: T3 in the fetal brain, for instance, reaches normal values with a maternal T4 dose which is much lower than needed for normal liver or lung T3 (19, 20, 23 and R. Calvo, M. J. Obregon, F. Escobar de1 Rey, and G. Morreale de Escobar, manuscript in preparation).

PLACENTAL

T,, Tii, rT:)

363

have been reported for this organ (l-10). The rT3 levels are not, however, increasing linearly with increasing T4 in Mplasma or placenta, a plateau being reached at about 12 nmol rT3/g in the M-placenta, at about 4 nmol rT3/g in the Fplacenta. This suggests a regulatory role of maternal T4 on the local production of rT3 from Tq, possibly mediated through increasing levels of T3, an inhibitor of 5 D. Emerson et al. (9), however, did not find an effect of the maternal T, supply on placental 5 D activity as measured in vitro. This does not necessarily disagree with present results, as we lack information on the relative roles in viva of the concentrations of enzyme and cofactors, substrates and inhibitors, in determining the final concentrations of rT3 in the tissue. Our results strongly suggest local generation of T3 from T, in the maternal side of the placenta: at comparable levels of T3 in M-plasma, the amounts of T3 found in the M-placenta are higher in T4-infused than in T3-infused dams. In the latter animals T3 can only be derived from the maternal circulation. The excess of locally produced T3 over the T3 derived from plasma appears to decrease at higher T, doses. These findings are in conceptual agreement with the reports by Kaplan and Shaw (11) and Hidal and Kaplan (12), who found high 5’D II activities in cells of the basal zone of the human and rat placenta, which decreased when the concentration of Tq in the incubation medium was increased. Hidal and Kaplan (12) suggested that the high 5’D II activity at low T4 concentrations “could potentially defend intracellular, and/or circulating, T3 pools in pathological states of mild to moderate hypothyroxinemia.” From our present data it appears more likely that this mechanism defends the intracellular supply of T3 in the M-side of the placenta itself. Hidal and Kaplan (12) speculated that 5’D II would help provide T3 for metabolic functions in the placenta, a suggestion which is supported by recent findings. The c-eubA/3 thyroid hormone receptor gene is expressed in the human placenta (33) and the nuclear receptor has been characterized in human trophoblastic cells (34, 35). Thus, a biological effect of thyroid hormones in the placenta is likely. Treatment of pregnant rats with high doses of T3 decreases placental glycogen content near term (36). Maruo et al. (37) have shown several dose-dependent actions of Tq and T3 in human trophoblast early in pregnancy, including increased in vitro secretion of progesterone, estradiol-17/3, human CG (hCG), hCG-cu, hCG-@, and human placental lactogen. A thyroid hormone response element has been described in the human placental lactogen-B promoter by Voz et al. (38) in the GC rat pituitary cell line. High levels of transthyretin (TTR) messenger RNA are found in the yolk sac endodermal layer of the developing rats until term (39, 40). Very low levels were also expressed in the placenta. The TTR synthesized in the yolk sac is secreted toward the fetal compartment, high levels being detected in the exocoelomic fluid surrounding the fetus (40). The synthesis and secretion of TTR by the yolk sac may function as part of a carrier mechanism across the yolk sack membrane similar to choroid plexus TTR mediated transport of Tq across the blood-cerebrospinal fluid barrier (41). In rodents the yolk sac persists throughout gestation (40) and it

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 July 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.

364

THYROID

STATUS

AND

is possible that not all of the thyroid hormones of maternal origin found in the fetus near term have reached it through the placenta. In summary: In the rat the net amounts of T4 and T3 contributed by the mothers to the placental and fetal extrathyroidal T, and T3 pools increase in proportion to the maternal levels. Increasing the maternal T, pool increases both T4 and TX. On the contrary, increasing maternal T3 only ameliorates the placental and the extrathyroidal T3 of the fetus as a whole, but leaves both the mother and fetus deprived of local generation of T3 from T,. This would be especially dangerous for the fetal brain, highly dependent on deiodination of T, to T3 via 5’D II for its supply of T3 (20, 23, 42). Although an important fraction of the T4 and T3 is retained and metabolized in the placenta, the amounts actually reaching the fetus should not be minimized as physiologically irrelevant: normal extrathyroidal T, and T3 pools of the fetus as a whole are ensured by increasing the maternal T4 levels 3.5-fold with respect to those of normal dams. Such doses were well tolerated by the MMI-treated dams, as far as assessed from litter size and fetal weight. Lower doses are sufficient for normal fetal brain T3 (23), and fetal pituitary and plasma GH (25). Present results support previous reports indicating active placental 5 D deiodination of T4 to rT,, as well as 5’D deiodination of T4 to T3 in the basal side. Despite active metabolization of the iodothyronines and their accumulation, the placenta acts only as a limited barrier to the transfer of iodothyronines. Within the range of T, and T3 infusion doses used for the present study the placenta does not prevent an increase of the fetal T, and T3 pools to levels which are higher than normal when the maternal T4 and T3 pools are excessive. Acknowledgments We thank Socorro Duran and Maria Jesus Presas for invaluable technical assistance in the hormone determinations, and Prof. D. Van der Heide for the supply of antiserum for the determination of rTg by RIA.

References 1. Banovac K, Bzik L, Tislaric D, Sekso M 1980 Conversion of thyroxine to triiodothyronine and reverse triiodothyronine in human placenta and fetal membranes. Horm Res 12:253-259 2. Roti E, Fang SL, Green K, Emerson CH, Braverman LE 1981 Human placenta is an active site of thyroxine and 3,3’,5 triiodothyronine tyrosyl ring deiodination. J Clin Endocrinol Metab 53:498501 3. Roti E, Fang SL, Braverman LE, Emerson CH 1982 Rat placenta is an active site of inner ring deiodination of thyroxine and 3,3’,5 triiodothyronine. Endocrinology 110:34-37 4. Roti E, Braverman LE, Fang SL, Alex S, Emerson CH 1982 Ontogenesis of placental inner ring thyroxine deiodinase and amniotic fluid 3,3’,5’ triiodothyronine concentrations in the rat. Endocrinology 111:959-963 5. Cooper E, Gibbens M, Thomas CR, Lowy C, Burke CW 1983 Conversion of thyroxine to 3,3’,5’ triiodothyronine in the guinea pig placenta: in vim studies. Endocrinology 112:1808-1815 6. Roti E, Gnudi A, Braverman LE 1983 The placental transport, synthesis and metabolism of hormones and drugs which affect thyroid function. Endocr Rev 4:131-149

PLACENTAL

T,. T,. rT,,

Endo - 1992 Vol131.No1

7. McCann UD, Shaw EA, Kaplan MM 1984 Iodothyronine deiodination reaction types in several tissues; effects of age, thyroid status and glucocorticoid treatment. Endocrinology 114:1513-1521 8. Yoshida K, Suzuki M, Sakurada T, Kitaoka H, Kaise N, Kaise K, Fukazawa H, Nomura T, Yamamoto M, Saito S, Yoshinaga K 1984 Changes in thyroxine monodeiodination in rat liver, kidney and placenta during pregnancy. Acta Endocrinol (Copenh) 107:495499 9. Emerson CH, Bambini G, Alex S, Castro MI, Roti E, Braverman LE 1988 The effect of thyroid dysfunction and fasting on placenta inner ring deiodinase activity in the rat. Endocrinology 122:809-816 10. Emerson CH 1989 Role of the placenta in fetal thyroid homeostasis. In: Delange F, Fisher DA, Glinoer D (eds) Research in Congenital Hypothyroidism. NATO AS1 Series, Series A: Life Sciences, Plenum Press, New York, vol 161:31-42 5’ deiodination 11. Kaplan MM, Shaw EA 1984 Type II iodothyronine by human and rat placentas in vitro. J Clin Endocrinol Metab 59:253-257 12. Hidal J, Kaplan MM 1985 Characteristics of thyroxine 5’ deiodination-in cultured human placental cells. J Clin Invest 76:947-955 13. Fisher DA, Klein AK 1981 Thvroid develoument and disorders of thyroid function in the newborn. N Engl J Med 304:702-712 14. Fisher DA 1986 The unique endocrine milieu of the fetus. J Clin Invest 78:603-611 15. Obregon MJ, Mall01 J, Pastor R, Morreale de Escobar G, Escobar de1 Rey F 1984 L-thyroxine and 3,5,3’ triiodo-L-thyronine in rat embryos before onset of fetal thyroid function. Endocrinology 114:305-307 16. Woods RJ, Sinha A, Ekins R 1984 Uptake and metabolism of thyroid hormones by the rat fetus in early pregnancy. Clin Sci 67x359-363 17. Morreale de Escobar G, Pastor R, Obregon MJ, Escobar de1 Rey F 1985 Effects of maternal hypothyroidism on the weight and thyroid hormone content of rat embryonic tissues, before and after onset of fetal thyroid function. Endocrinology 117:1890-1900 18. Bernal J, Pekonen F 1984 Ontogenesis of the nuclear 3,5,3’ triiodothyronine receptor in the human fetal brain. Endocrinology 114:677-679 19. Morreale de Escobar G, Obregon MJ, Ruiz de Oiia C, Escobar de1 Rey F 1988 Transfer of thyroxine from the mother to the fetus near term: effects on brain 3,5,3’ triiodothyronine deficiency. Endocrinology 122:1521-1531 20. Morreale de Escobar G, Obregon MJ, Ruiz de Ona C, Escobar de1 Rey F 1989 Comparison of maternal to fetal transfer of 3,5,3’ triiodothyronine versus thyroxine in rats, as assessed from the 3,5,3’ triiodothyronine levels in fetal tissues. Acta Endocrinol (Copenh) 120:490-489 T, Gons MH, de Vijlder J 1989 Maternal-fetal transfer of 21. Vulsma thyroxine in congenital hypothyroidism due to a total organification defect or thyroid agenesis. N Engl J Med 321:13-16 22. Morreale de Escobar G, Calvo R, Obregon MJ, Escobar de1 Rey F 1990 Contribution of maternal thyroxine to fetal thyroxine pools in normal rats near term. Endocrinology 126:2765-2767 23. Calvo R, Obregon MJ, Ruiz de Ona C, Escobar de1 Rey F, Morreale de Escobar G 1990 Congenital hypothyroidism, as studied in rats: crucial role of maternal thyroxine (T4), but not of 3,5,3’ triiodothyronine (T3) in the protection of the fetal brain. J Clin Invest 86:889-899 B, Brownlie BEW, McKay Hart D, Horton PW, Alex24. Marchant ander WD 1977 The placental transfer of propylthiouracil, methimazole and carbimazole. J Clin Endocrinol Metab 45:1187-1193 25. Morreale de Escobar G, Calvo R, Escobar de1 Rey F, Obregon MJ Differential effects of T4 and T3 on fetal rat growth (GH) and thyrotrophic (TSH) hormones. Program of the 19th Annual Meeting of the European Thyroid Association, Hannover, Germany, 1991. Ann Endocrinol (Paris) 52:37 (Abstract) de Escobar G, Calvo RM, Escobar de1 26. Ruiz de Ona C, Morreale Rey F, Obregon MJ 1991 Thyroid hormone and 5’-deiodinase in the rat fetus late in gestation. Effects of maternal hypothyroidism. Endocrinology 128:422-432 27. Van der Heide D, Ende-Visser Ml’ 1980 Thyroxine, triiodothyronine and reverse triiodothyronine in the plasma of rats during the

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 July 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.

THYROID 28. 29. 30.

31.

32.

33.

34.

35.

STATUS AND PLACENTAL

first three months of life. Acta Endocrinol (Copenh) 93:448-454 Barker JN 1966 Fetal and neonatal cerebral blood flow. Am J Physiol 210:897-902 Snedecor GW, Cochran WG 1980 Statistical Methods, ed 7. Iowa State University Press, Ames, IA El Zaheri M, Vagenakis AG, Hinerfeld L, Emerson CH, Braverman LE 1981 Maternal thyroid function is the major determinant of amniotic fluid 3,3’,5’-triiodothyronine in the rat. J Clin Invest 67:1126-1133 Calvo R, Obregon MJ, Ruiz de Ofia C, Ferreiro 8, Escobar de1 Rey F, Morreale de Escobar G 1990 Thyroid hormone economy in pregnant rats near term: a “physiological” animal model of nonthyroidal illness? Endocrinology 127:10-16 Yoshida K, Suzuki M, Sakurada T, Takahashi T, Furuhashi N, Yamamoto M, Saito S, Yoshinaga K 1987 Measurement of thyroid hormone concentrations in human placenta. Horm Metab Res 19:130-133 Weinberger C, Thomson CC, Ong ES, Lebo R, Gruol DJ, Evans RM 1986 The cerbA gene encodes a thyroid hormone receptor. Nature 324:641-646 Ashitaka Y, Maruo M, Takeuchi Y, Nakayama H, Mochizuki M 1988 3,5,3’-triiodo-L-thyronine binding sites in nuclei of human trophoblastic cells. Endocrinol Jpn 35:197-206 Nishii H, Ashitaka Y, Maruo M, Mochizuki M 1989 Studies on the nuclear 3,5,3’-triiodo-L-thyronine binding sites in cytotropho-

36. 37.

38.

39.

40.

41.

42.

T,, T:l, I-T: