| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Unidad de Endocrinología Molecular, Instituto de Investigaciones Biomédicas, Consejo Superior de Investigaciones Científicas and Facultad de Medicina, Universidad Autónoma de Madrid (R.M.C., M.A., G.M.d.E.), 28029 Madrid, Spain; Academic Department of Obstetrics and Gynecology, Royal Free and University College London Medical School (E.J.), London, United Kingdom WCIE 6HX; Department of Clinical Chemistry, Academic Hopital Erasme (B.G., C.G.), and Institute of Interdisciplinary Research, Université Libre de Bruxelles (B.C.), B-1070 Brussels, Belgium
Address all correspondence and requests for reprints to: G. Morreale de Escobar, Ph.D., Instituto Investigaciones Biomédicas Alberto Sols, Arturo Duperier 4, 28029 Madrid, Spain. E-mail: . gmorreale{at}iib.uam.es
Abstract
Maternal hypothyroxinemia in early pregnancy is often associated with irreversible effects on neuropsychomotor development. To evaluate fetal tissue exposure to maternal thyroid hormones up to midgestation, we measured total T4 and free T4 (FT4), T3, rT3, TSH, and possible binding proteins in first trimester coelomic and amniotic fluids and in amniotic fluid and fetal serum up to 17 wk. Samples were obtained before interruption of maternal-fetal connections. The concentrations in fetal compartments of T4 and T3 are more than 100-fold lower than those in maternal serum, and their biological relevance for fetal development might be questioned. We found, however, that in all fetal fluids the concentrations of T4 available to developing tissues, namely FT4, reach values that are at least one third of those biologically active in their euthyroid mothers. FT4 levels in fetal fluids are determined by both their T4-binding protein composition and the T4 or FT4 in maternal serum. The binding capacity is determined ontogenically, is independent of maternal thyroid status, and is far in excess of the T4 in fetal fluids. Thus, the availability of FT4 for embryonic and fetal tissues would decrease in hypothyroxinemic women, even if they were euthyroid. A decrease in the availability of FT4, a major precursor of intracellular nuclear receptor-bound T3, may result in adverse effects on the timely sequence of developmental events in the human fetus. These findings ought to influence our present approach to maternal hypothyroxinemia in early pregnancy regardless of whether TSH is increased or whether overt or subclinical hypothyroidism is detected.
THYROID HORMONES are essential for fetal brain development, but the timing of onset of the different events that are thyroid hormone dependent has not yet been clearly defined, in particular for the first trimester of human development (1). This uncertainty persists despite the demonstration that nuclear TR, partially occupied by T3, are present in first trimester human fetal brain tissue (2, 3, 4). This suggests that thyroid hormone-mediated effects may already occur in the early fetus.
It is important to clarify this point. There is increasing evidence from epidemiological and experimental data (5) that first trimester maternal thyroid status is pivotal for the outcome of pregnancy (6) and for the neuropsychomotor development of the child (7, 8). Efforts to detect and prevent maternal hypothyroxinemia in early pregnancy appear fully justified (5). Indeed, neurodevelopment defects, including an increased probability of cerebral palsy, may be 150 times more frequent than those resulting from untreated congenital hypothyroidism.
More direct evidence obtained from human fetal fluids is still needed to better understand the maternal-fetal thyroid hormone relationships before the onset of fetal thyroid secretion. Available information up to midgestation is scarce, especially for fetal fluid samples obtained before interruption of maternal to fetal connections. During this period two distinct modes of maternal-fetal nutrient transfer occur sequentially. For most of the first trimester the human fetus is surrounded by two distinct fluid cavities separated from each other by a thin membrane (Fig. 1A
): the inner, or amniotic, cavity contains the fetus, and the outer, or exocoelomic, cavity separates the amniotic cavity from the placenta and contains the secondary yolk sac (9). The secondary yolk sac is directly connected to the fetal digestive tract and circulation. The investigation of early fetal fluids has demonstrated that the exocoelomic cavity is the site of important molecular exchanges between the mother and the fetus (10, 11, 12, 13). The coelomic fluid results from an ultrafiltrate of maternal serum with the addition of specific placental and secondary yolk sac bioproducts (10, 11, 12, 14). The exocoelomic cavity is clearly a physiological liquid extension of the early placenta that may act as a reservoir for nutrients needed by the developing fetus (13). We previously demonstrated that T4 (and possibly T3) is indeed present in coelomic fluid (CF) samples as early as 5.6 wk gestation (15). These preliminary data show that maternal thyroid hormones are potentially available to early fetal tissues, but raise new questions about whether the thyroid hormone concentrations found inside the gestational sac could be biologically relevant for the occupation of nuclear TR in the fetal brain.
|
). These major anatomical transformations modify considerably the spatial relationships between maternal tissue and developing fetus, and, consequently, the maternal-fetal exchange pathways. From 11–12 wk onward, maternal nutrients, including thyroid hormone, are transferred via the placenta directly into the fetal circulation. Thyroid hormone levels in fetal serum (FS) had only occasionally been measured during the same period (16). The aim of this study was to further investigate the distribution of thyroid hormones inside the first trimester exocoelomic and amniotic cavities, to determine thyroid hormone concentrations in the second trimester amniotic cavity and fetal circulation, and to quantify the concentrations of the T4 likely to be available to fetal tissues, namely those of free T4 (FT4), and interrelationships with maternal circulating iodothyronines. Materials and Methods
Subjects and samples
We have studied four series of samples. Series A and B consisted of matched samples of CF, amniotic fluid (AF), and maternal serum (MS) at 5.6–12 wk gestation. Series A corresponds to the 13 women presented in our previous report (15), all of whom were Caucasian. Series B comprised samples from 20 women collected in a different area of London; 12 were from ethnic minorities (7 black African and 5 Asian). The third series of samples (C) included matched samples of CF, AF, and MS at 7.6–11.6 wk (n = 15) and of AF, FS, and MS at 11.6–17.1 wk gestation (n = 28); 9 women were black African, and 4 were Asian. Series D consisted of first trimester CF and AF samples (n = 32) for which MS was not available. These samples were used for necessary verifications of some of the analytical procedures (HPLC; percent FT4 and percent FT3). All samples were collected before elective surgical termination of pregnancy under general anesthesia. Gestational age was determined from the first day of the last menstrual period and was confirmed by ultrasound measurement of the fetal crown-rump length. Written consent was obtained from each woman after receiving complete information on the procedure. The study included only uncomplicated pregnancies and was approved by University College London Hospitals committee on the ethics of human research.
CF and AF samples were obtained by trans-vaginal puncture under sonographic guidance (10, 12, 13). The first 0.2 ml of each fluid was discarded to decrease the risk of contamination by maternal blood. Fetal blood samples of 2.0–5.0 ml were aspirated by ultrasonographically guided intracardiac puncture (13). The samples were collected in heparinized tubes, centrifuged within 60 min of collection. In all cases, maternal venous blood was obtained during the surgical procedure. FS and MS samples were stored at -20 C until assayed; CF and AF were stored at -70 C.
Determination of T4, T3, and rT3 in fetal fluids
Aliquots (400 µl) from each CF sample were taken to determine T4 and rT3 concentrations individually. The rest was used to measure the T3 concentrations and the percent FT4, for which 1–4.5 ml are needed. With AF, 200-µl aliquots were sufficient to evaluate the percent FT4, but 3–6 ml were needed for the assays of T4 and T3. If necessary, CF or AF samples were pooled according to equivalent gestational age. The iodothyronines were determined using highly sensitive and specific RIAs after extensive extraction and purification with methods previously described in detail (17, 18) with modifications to improve recovery. Briefly, [131I]T4 and [125I]T3 or [131I]T4 and [125I]rT3 were added to each sample as internal tracers for recovery calculations. The samples were extracted with a 2.5-fold volume of ethanol, and the ethanol extracts were further purified on AG 1 x 2 resin columns (Bio-Rad Laboratories, Inc., Richmond, CA), from which the iodothyronines were eluted with 70% acetic acid, pooled, evaporated to dryness, and extensively counted for the calculation of individual recoveries. RIA buffer was then added, and the iodothyronine contents were determined by RIAs in duplicate at two dilutions. The limits of detection are 3.2 fmol T4, 1.0 pmol T3, and 1.5 pmol rT3/tube.
Verification of the presence of thyroid hormones in coelomic fluid by HPLC
CF samples were extracted twice with ethanol-0.5 N HCl (10:0.5, vol/vol). After centrifuging at 2000 rpm for 10 min, the supernatants were mixed and neutralized with 200 µl ammonia and passed through Dowex AG 1 x 2 columns as described above. The iodothyronines were eluted with 70% acetic acid and evaporated to dryness. The residue was dissolved in 100 µl 0.3 N ammonia and transferred to a Chromspher 5 C8 100 x 3-mm column (Chrompack, Middleburg, The Netherlands) after addition of [131I]T4 and [125I]T3. T4 and T3 were separated with an isocratic gradient using 55% methanol and 45% 0.015 M ammonium acetate, pH 3.5–4, at a rate of 0.6 ml/min. The fractions corresponding to the labeled iodothyronines were evaporated to dryness, and 200 µl RIA buffer were added. The RIAs were performed as described above.
Percentages of free T4 and T3 in fetal fluids
The percent FT4 was determined in undiluted samples of CF and AF, with slight modifications to the method described by Mendel et al. (19). A 5-µl aliquot of high specific activity [125I]T4 (
350,000 cpm for CF and 70,000 cpm for AF) was added to 195 µl sample and incubated at room temperature for 1 h. A 180-µl aliquot of each labeled CF or AF sample was ultrafiltered using Microcon 10 microconcentrators (Amicon Division, W.R. Grace Co., Beverly, MA) and centrifuged for 30 min at 3000 rpm. A measured volume of each ultrafiltrate and of the initial tracer was added to 0.2 ml bovine serum and precipitated with 2.5 ml 10% trichloroacetic acid; the pellet was counted after being washed twice with the trichloroacetic acid solution. The radioactivity in the pellet was calculated as a percentage of the tracer initially added. Each sample was processed in duplicate. The percent FT4 and the total T4 determined by RIA were used to calculate the concentrations of FT4.
The percent FT3 was determined in some undiluted CF samples using the same procedure as that used to calculate the percent FT4, but only approximately 100,000 cpm high specific activity [125I]T3 were added.
Drugs and reagents
T4, T3, and 3,5-diiodothyronine (3,5-T2) were obtained from Sigma (St. Louis, MO). rT3 and 3',3-T2 were obtained from Henning Berlin GmbH \|[amp ]\| Co. (Berlin, Germany), AG 1 x 2 resin was purchased from Bio-Rad Laboratories, Inc. High specific activity [125I]T3, [125I]T4, [125I]rT3, and [131I]T4 (3000 µCi/µg) were synthesized by our laboratory as previously described (17).
Other assays
Hemoglobin content was measured in 10-µl aliquots of CF and AF (20). To assess the amounts of T4 and T3 from maternal blood that might have contaminated CF or AF samples, we used the T4 and T3 concentrations in MS and assumed a value of 140 g/liter for hemoglobin and a 50% hematocrit.
Total T4 in MS was measured by RIA, and free T4 in MS and FS was measured by a two-step RIA method (Gammacoat kits provided by Clinical Assays, INCSTAR, Stillwater, MN), T3 was determined using the Amerlex RIA (Amersham Pharmacia Biotech, Little Chalfont, UK), rT3 was determined using an RIA kit (Tickland, Liege, Belgium), FT3 in MS was determined with the Dynotest FT3 kit from Brahms Diagnostica GmbH (Berlin, Germany), TSH in MS and FS was determined by immunoradiometric assay, and T4-binding globulin (TBG) was determined by RIA, with commercial kits (Behring, Marburg, Germany). The coefficients of variation were less than 10 for T3, T4, and FT4; less than 5% for TSH; and less than 10% for TBG and rT3.
Transthyretin (TTR) concentrations in MS and CF were determined by immunonephelometry, using antihuman transthyretin Ab (Behring) on the Behring nephelometer analyzer. The concentrations of albumin, 
1-, 2-, ß-, and 
-globulins in CF were quantified by densitometry after electrophoresis (14) with a 5% coefficient of variation.
Intact hCG was measured using a solid phase RIA based on the principle of the sandwich assay (RIA-gnost hCG, Behring). The sensitivity and intraassay coefficient of variations of this test were 1 IU/liter and less than 5%, respectively.
To test likely interferences in the TSH RIA when samples contained very high concentrations of hCG, we obtained purified hCGß (provided by Dr. A. F. Parlow from National Hormone and Pituitary Programs, NIDDK, NIH). Native hCG obtained from pregnant human urine was supplied by Sigma (St. Louis, MO).
Statistical analysis
The data were analyzed SPSS 6.1 (SPSS, Inc., Chicago, IL). All values are expressed as the mean ± SE. One-way ANOVA, followed by the least significant difference post-hoc test, were used to compare mean values of different series of data. Partial correlation analysis was performed, and Pearson’s correlations coefficients were calculated; P < 0.05 was considered statistically significant. Most of the variables measured in the present study were positively correlated with gestational age, and partial correlation analyses disclosed that most of the correlations between variables were spurious, as only a few persisted after correcting for gestational age. Unless stated otherwise, only those correlations that were independent of gestational age are considered here.
Results
Gestational age and maternal thyroid function
Table 1 summarizes the results obtained in the MS from series A–C. Series C was further divided into groups C1 (first trimester) and C2 (second trimester). First trimester levels of T4 and TSH were the same in all groups, but T4 was higher in second trimester samples. FT4 levels were lower in groups B, C1, and C2 compared with A, whereas T3 levels were higher. The T3/T4 molar ratio was increased in all groups compared with group A. FT3 (not shown in Table 1) was only measured in series C and increased between the first and second trimester MS from 6.32 ± 0.25 to 7.49 ± 0.23 pmol/liter (P = 0.002), as did the FT3/FT4 molar ratio, from 0.305 ± 0.017 to 0.427 ± 0.016 (P < 0.001).
|
Verification by HPLC of the presence of thyroid hormones in CF
The peaks of both iodothyronines, obtained by RIA, coincided with the peaks of in vitro added labeled T3 and T4 (Fig. 2), thus confirming the specificity of results obtained by RIA in the purified CF.
|
Table 2 shows the comparison between T4, FT4, T3, and rT3 concentrations in CF and AF samples from all series (A–D). No significant differences were found between the concentrations of T4 or T3 in first trimester CF from different series; rT3 was higher in the C series and was 2- to 4-fold higher than T4 in all series. The T3 levels are still tentative estimates, as larger volumes of a greater number of CF samples are needed for a more reliable assessment. Nevertheless, T3 levels appeared to be 30–50 times lower than T4 and 100 times lower than rT3. T4, T3, and rT3 were also found in those AF samples in which the volume was sufficient to attempt their determination. The total concentrations of all three iodothyronines were lower in AF than in age-paired first trimester CF samples, but FT4 levels were higher. Both T4 and rT3 concentrations were higher in the second trimester AF samples. AF volumes required for the determination of T3 were large, excluding determinations in series C and D.
|
rT3 levels correlated with T4 in both CF and AF (P < 0.001). No correlations were found in CF or AF between T3 and either T4 or rT3, but the available T3 data were limited (see above). Correlations of the different variables between the two cavities were not found.
The percent FT4 and FT4 concentrations were measured in CF and AF from series B–D. Initial results obtained with samples from series B indicated that the CF and AF percent FT4 values were much higher than those in MS. Despite the more than 150-fold difference in total T4 between CF, or AF, and MS, the concentrations of FT4 were of the same order of magnitude and should, therefore, be potentially measurable with FT4 kits used for human serum. We repeatedly attempted this with several commercial kits [Dynotest FT4, Brahms Diagnostica GmbH, Amerlex MAB FT4, Ortho-Clinical Diagnostics (Amersham Pharmacia Biotech), Clinical Assays Gammacoat two-step FT4, INCSTAR Corp. (Stillwater, MN), and FT4 by equilibrium dialysis (Nichols Institute Diagnostics, San Juan Capistrano, CA)] and concluded they cannot be used to measure FT4 in CF and AF, because the standards in these kits mimic the protein content of human sera, which is markedly different from that of CF and AF. Results, including those obtained with the dialysis kit, were not only unreliable when using CF or AF, but also when using ultrafiltrates of MS.
We reverted to the ultrafiltration method described above after having excluded likely sources of error. The possibility was considered that the T4 contained in 350,000 cpm high specific activity [125I]T4, added to 195 µl CF or AF, might have raised the concentration of T4 in the sample beyond its binding capacity and result in an overestimation of the percent FT4. This was investigated with eight different samples of CF to which one half or one fourth the usual amount of [125I]T4 was added or to which the usual amount of tracer was added plus increasing amounts of T4, ranging from 0.09–33 nmol/liter. No significant correlation was found between the percent FT4 values and the concentration of added T4 up to the addition of 7 nmol T4/liter. At higher concentrations of added T4, the percent FT4 started increasing in several samples, but in others it remained the same even after the addition of up to 33 nmol T4/liter. The capacity of the first trimester AF samples to bind added T4 without altering the measured percent FT4 was less than that of the CF, and for this reason the [125I]T4 added to AF samples was reduced to approximately 70,000 cpm. AF samples from the second trimester were assayed as described for CF, because of the sudden increase in protein content and T4 binding capacity compared with those of the first trimester AF.
Figure 3 shows the total T4 concentration, the percent FT4, and the FT4 concentration in matched samples of CF and AF. The mean percent FT4 in CF was much higher than that in MS (0.42% vs. 0.012–0.028%), and even higher (21.9%) in first trimester AF. There was a sharp decrease in the percent FT4 (to 0.38%), between the end of the first and the beginning of the second trimester, which corresponds to the period when the fetus starts producing urine (10, 11, 14). In both extraembryonic cavities the total T4 concentration changed inversely to that of percent FT4. The resulting concentrations of FT4 in the different compartments were buffered and much more similar to each other than the very different concentrations of total T4. In both compartments FT4 correlated with T4 (r = 0.86; P < 0.001 for CF and r = 0.55; P = 0.014 for AF). The FT4 concentrations of both compartments were also correlated (r = 0.84; P = 0.005).
|
TTR was present in CF, and its concentration increased with gestational age (r = 0.82; P < 0.001), with values ranging from 0.30 fmol/liter at 7 wk to about 0.56 fmol/liter between 9–11 wk. The concentrations of 1-, 2-, and -globulin in CF increased significantly with gestational age, from 15 to 67 mg/dl for 1-globulin, from 8 to 45 mg/dl for 2-globulin, and from 31 to 105 mg/dl for -globulin. Changes in the concentrations of all of these proteins were gestational age dependent and independent of maternal thyroid status. Total T4 also increased with gestational age, and a positive correlation was found between CF T4 and TTR (r = 0.62; P = 0.01) and between T4 and 1-, 2-, and -globulin concentrations. However, only the correlation between T4 and 1-globulin persisted after adjusting for gestational age (r = 0.74; P < 0.01).
Contrary to the increase in hCG concentrations occurring in MS during the first trimester, ranging in the present series from 16,665–175,500 IU/liter between 5 and 10 wk gestation, the hCG concentrations decreased in CF from 385,000 to 81,000 IU/liter. No correlation persisted between T4 and hCG after correction for gestational age in either CF or MS.
Thyroid hormones in fetal serum
Individual data for T4, FT4, T3, TBG, and TSH in FS are shown as a function of advancing gestation in Fig. 4. T4 and T3 were undetectable using commercial kits in all but a few of the samples from older fetuses. Both, however, could be measured, after extraction and purification of larger volumes of serum, using the method described for CF and AF. Mean values (±SE) were 6.88 ± 0.72 nmol/liter for total T4 and 0.78 ± 0.18 nmol/liter for total T3 at a mean gestational age of 14.5 ± 0.3 wk. There was insufficient FS to measure the percent FT4 by the same ultrafiltration method as that used for CF and AF, but there was sufficient FS for the determination of FT4 using a commercial kit tested for human sera, including samples with very low TBG concentrations. The mean FT4 concentration was 5.69 ± 0.40 pmol/liter, a value not significantly different from that found for second trimester AF samples (Fig. 3). TBG in FS was undetectable (<0.091 fmol/liter) up to 15 wk, after which the mean value for six samples was 0.12 ± 0.01 fmol/liter. FT4 in FS correlated with the T4 concentration in the same fluid (r = 0.90; P < 0.001) and T3 with T4 (r = 0.89; P < 0.001) and FT4 (r = 0.68; P = 0.017).
|
), and did not correlate with T4, T3, or FT4. We had not anticipated these concentrations of TSH in the serum of the human fetus before the primary capillary plexus between the hypothalamus and the median eminence develops at 18–20 wk. TSH in FS was higher than that in MS; no correlation was found between TSH in FS and MS. We tested the possibility that despite the high specificity of the TSH Ab described by the manufacturers of the TSH kits, the very high hCG concentrations in the FS and MS might cross-react with it, and this might result in an overestimation of the true TSH concentration. This possibility was excluded; we added up to 400,000 IU/liter of both hCGß and native hCG to several samples of human serum without any effect on the resulting TSH value. Comparisons between extraembryonic fluids and fetal serum
As the T4 and T3 concentrations in CF and AF are referred to the volume of these fluids, we compared them with those of the corresponding volume of the fluid in the fetal circulation, namely, blood, not serum. The concentrations in blood have been calculated from those in serum by assuming a 50% hematocrit. The concentrations of T4 or T3 in CF, AF, and fetal blood were plotted simultaneously against gestational age. These plots showed that, with development, there was a progressive and continuous increase in total T4 and T3 in the embryonic and fetal fluids bathing the yolk sac and fetal tissues. The data fitted exponential functions of age, with r = 0.89 for total T4 and r = 0.87 for total T3 (P < 0.001 for both). On the contrary, as shown in Fig. 5, FT4 concentrations in the embryonic and fetal fluids remained fairly constant.
|
Comparison of hormones in fetal fluids with maternal thyroid status
Most of the gestational age-related changes in parameters of maternal thyroid hormone status of the present study agree with those already described by others for early pregnancy (23). Thus, TBG increased from 5 to 17 wk (r = 0.65; P < 0.001). Circulating T4 (Table 1) also increased with advancing gestation (r = 0.34; P = 0.011) and was higher in the second trimester compared with the first (P = 0.001). Changes in total T3 were similar, with second trimester values higher than those in the first trimester (P < 0.001). There was no appreciable difference in TSH between the trimesters. FT4 was higher in the first trimester compared with the second (Table 1) and negatively correlated to gestational age (r = -0.40; P = 0.013). In the present study, however, FT3 concentrations and FT3/FT4 ratios increased from the first to the second trimester. In the first trimester, T4 in CF was positively correlated to T4 in MS (r = 0.61; P < 0.001) and in the second trimester T4 in FS was correlated to FT4 in MS (r = 0.57; P = 0.14), but T3 in FS was not correlated to MS T3 or FT3. Neither T4 nor FT4 in AF correlated with these variables in MS in either the first or second trimester.
Figure 5 shows individual values of T4 and FT4 for series C, which comprised not only samples of CF and AF, but also those of fetal and maternal blood. These concentrations were derived from the serum values, assuming a 50% hematocrit. The highest T4 concentrations in fetal blood were at least 10-fold lower than those in maternal blood and 100-fold lower, or more, in CF and AF compared with maternal blood. In contrast, the differences between FT4 in the fetal and maternal compartments were much smaller than those observed for total T4. The mean FT4 concentration in CF (see Fig. 3) reached 45% of the mean maternal blood value (9.4 pmol FT4/liter blood, derived from 18.4 pmol FT4/liter serum), and 53% of the reference blood value for the general population (7.64 pmol FT4/liter blood, derived from 15.3 pmol/liter serum). The mean FT4 in AF was similar to that in maternal blood, and the mean FT4 in fetal blood reached 43% of the maternal value (51% of that in the general euthyroid adult population).
Discussion
Total T4 concentrations increase steadily during the first half of pregnancy in both maternal and fetal compartments. T3 concentrations in fetal fluids are at least 10-fold lower than those of T4, whereas rT3 in CF and AF is higher than that in the maternal circulation. T4 in fetal fluids is directly related to the thyroxinemia of euthyroid mothers, whereas no correlations were found for T3 or rT3 between the maternal and fetal compartments. Findings are consistent with present concepts of maternal-fetal transfer of molecules (13). During the first trimester maternal T4 is transferred into the exocoelomic cavity and subsequently into the fetal gut and circulation via the secondary yolk sac; during the second trimester, it is transferred directly into fetal blood.
Synthesis of TTR by the yolk sac epithelium (14, 15, 25) then facilitates transfer of FT4 from the CF into the yolk sac, and its early synthesis by choroid plexus epithelium (26) enhances the transfer of T4 into cerebral structures. The composition of the AF is more likely to reflect fetal metabolism of the iodothyronines, although limited exchanges between the two fluid cavities occur during the first trimester (13). The marked change in AF T4 between the first and second trimesters is probably linked to the onset of glomerular filtration and direct excretion of fetal proteins into this cavity (10).
The present results neither confirm nor exclude maternal T3 transfer into first trimester CF. T3 in the second trimester FS is not correlated to T3 or FT3 in MS, but to the T4 and FT4 concentrations in the same fetal compartment, suggesting that the T3 and FT3 in early fetal fluids are not transferred from the mother, but result from metabolism of the maternal T4 that reached fetal tissues.
The low concentrations in fetal fluids of T3, the high concentrations of rT3, and the high molar rT3/T4 ratios suggest that iodothyronines transferred from maternal to fetal fluids in early pregnancy undergo the same inactivating processes that are characteristic of later fetal development (27, 28, 29, 30, 31, 32, 33, 34, 34, 35, 36). These are attributed to high activities of inner ring 5'-iodothyronine deiodinase (D-III), an inner ring deiodinating isoenzyme with preference for T3 over T4 as substrate, and of sulfo-transferases, producing T4 and T3 sulfates. High D-III activities have been described in the decidua, most placental cellular types and membranes, and many fetal tissues. Also present in placenta and some fetal tissues (i.e. brain, brown adipose tissue, and pituitary) is the 5'D-II isoform, which permits local intracellular generation of T3 from T4. 5'D-I, an important source of the systemic T3 needed for the supply of intracellular T3 to other tissues (i.e. liver, lung, and kidney), appears much later during perinatal development.
D-III activity in uterus and placenta constitutes an important barrier for the transfer of maternal T4 and T3 to the fetal compartment. D-III in placental membranes and fetal tissues further inactivates the iodothyronines escaping the barrier and is responsible, together with high sulfo-transferase activities, for the low T3 and high rT3 levels characteristic of fetal compartments. Studies in animal models and humans show that placental 5D-III activity is not affected by maternal thyroid hormone status (29, 37); a decrease in the amount of substrate, namely maternal circulating T4 and T3, results in lower concentrations of the hormones in fetal fluids. This conclusion is strongly supported by the positive correlations between T4 in fetal fluids and circulating T4 or FT4 levels in their euthyroid mothers.
The T4 in fetal fluids that is actually available to tissues, namely FT4, is much higher than expected in adults with comparably low circulating T4. FT4 in fetal fluids reaches concentrations similar to those exerting biological effects in adults. Such concentrations may well be biologically relevant for the early developing brain, as T3-occupied nuclear receptors are already found (2, 3, 34). Our results reconcile apparently contradictory observations, namely, that maternal thyroid hormones are needed by the early developing brain, but at the same time they are prevented from reaching the fetal compartment during a period of development when the mother is their only source. Present findings show that, if such a barrier were not effective and T4 and T3 in fetal and maternal fluids were the same, the FT4 and FT3 available to developing tissues could be excessive and potentially harmful.
The unexpected similarity of the concentrations of FT4 in maternal and fetal fluids is related to major qualitative and quantitative differences in thyroid hormone-binding proteins. TBG in first trimester fetal fluid is not detectable, and its concentration in early FS is very low. TTR (38) and some of the T4-binding proteins of the 1-globulin fraction, such as 1-antitrypsin, 1-antichymotrypsin, and 1-fetoprotein (39, 40, 41), are present in CF at low concentrations. Thus, the proportion of maternal T4 that reaches the fetal fluid cavities and is found in the free form, i.e. the percent FT4, is much higher than that in adults; the resulting FT4 concentrations in fetal and maternal fluids become similar. The protein composition of fetal fluids is defined ontogenically and is not affected by maternal thyroid status. As the thyroid hormone-binding capacity in each fluid is far in excess of the total concentrations of iodothyronines, the FT4 in the fetal fluid compartments is determined by their total T4. The latter, in turn, is correlated to circulating maternal T4 (first trimester) or FT4 (second trimester); a decrease in maternal T4 or FT4 thus results in a concordant decrease in FT4 in fetal fluids and, consequently, in tissues.
T4 and T3 have been found in different tissues between 9 and 18 wk gestation (2, 3, 34), providing direct evidence that FT4 (and possibly FT3) reaching fetal fluids actually enters tissues; the concentration of T3 in the cortex at 13 wk gestation may actually reach 50–60% of the values reported for adults (42), with 20–34% thyroid hormone nuclear receptor occupancy by T3 (43). These estimates are approximate and do not take into account that different tissues and organs or cell types within the same organ have specific temporal patterns of development, and their requirements for T3 may vary widely. This would be of special importance early in life, when either a deficient or an excessive supply of hormone would result in the premature or delayed onset of biological events and an irreversible loss of opportunity for orderly development.
The ontogenetically programmed expression of 5'D-II and 5D-III is an important mechanism to achieve tightly controlled concentrations of T3 in cerebral structures at different stages of development. It is frequently observed during development even in tissues other than the brain and in mammalian and nonmammalian species (34, 44, 45), overcoming the fact that these are all exposed to the same concentrations of FT4 or FT3 in fetal fluids. The importance of the availability of T4 over T3 becomes evident, as T4 is the initial substrate of these regulatory cascades that are determined ontogenically at cell-specific developmental stages. Relatively high T3 levels and nuclear receptor occupancy by T3 have indeed been found in the early human cortex (2, 3, 43) despite very low FT3 in fetal fluids, but present results show that FT4 is relatively normal and could provide enough substrate for generation of sufficient T3 by 5'D-II deiodination. This enzyme is already present in the cortex at 11–14 wk gestational age (46). Intracellular T3 concentrations in other fetal tissues, such as liver, heart, and lung, however, were still undetectable at 18 wk gestation (2, 3, 43). These tissues depend mostly on systemic T3 generated by the 5'D-I, an enzyme that is only expressed much later, near term or postnatally.
In the rat, maternal T4 is the principal source of intracellular T3 in the early developing brain (47, 48). Moreover, its local generation from T4 of maternal origin protects the fetal brain from T3 deficiency in cases of fetal thyroid failure, because cerebral 5'D-II activity increases markedly in response to a minor decrease in T4 (49, 50), compensating for the decreased availability of substrate by generating more T3. These findings in rat fetuses correspond to a period of brain development that occurs during the second half of human pregnancy. Whether they are operative in earlier stages of pregnancy has not been fully established. Karmarkar et al. (46) found that 5'D-II activity in the human cerebral cortex at 11–14 wk gestation did not respond to a decrease in maternal T4. This suggests that down-regulation of 5'D-II activity by T4 is delayed with respect to the timing of onset of its expression. A response was detected later, at 15–18 wk, but appeared to be inadequate, because T3 in the cortex remained low. Therefore, in the absence of a compensatory regulatory response of cerebral 5'D-II early in development, a decrease in maternal T4 would ultimately result in a lower intracellular T3 concentration.
Biological effects of thyroid hormone have not yet been identified in early human fetal brain, but recent studies show that maternal hypothyroidism early in pregnancy results in the inappropriate expression of specific genes in the fetal rat brain (51, 52) and in irreversible alterations of the migration of cells in the cortex and hippocampus (53). The latter are also observed in the progeny from iodine-deficient rat dams (54) that are hypothyroxinemic but not hypothyroid. A very short period of goitrogen-induced mild hypothyroxinemia early in pregnancy is also sufficient to induce both abnormal migration and permanent neurological damage (i.e. increased audiogenic seizure susceptibility) (55). Alterations of migration are easily visualized by a significant increase in heterotopic cells in the subcortical white matter. This increase has been observed in human fetuses from an iodine-deficient area (56), suggesting that human cerebral cytoarchitecture and the establishment of normal circuits are also affected by early maternal hypothyroxinemia; in humans the two main waves of migration of cells in the neocortex occur before midgestation, with peaks at 8–10 and 12 wk gestation (57).
In all FS samples we examined in this study TSH values were relatively high and did not correlate with maternal TSH, confirming a previous report with fewer samples (16). Studies on sera from normal and anencephalic human fetuses have shown that TSH bioactivity is greatly increased with respect to that circulating in the mothers (58, 59), confirming that it is not of maternal origin. Some of the results suggest the existence of sources of fetal TSH that are not under hypothalamic neuroendocrine control. In the rat the brain is one of these sources (60). A TSH receptor has been found in early human fetal brain and human astrocytes in primary culture (61). This receptor mediates extrathyroidal cAMP-independent biological effects of TSH, among which is, interestingly, the stimulation of 5'D-II in astroglial cells (62).
In summary, we here show that first trimester fetal tissues are exposed to concentrations of FT4 in the same range as those available to adult tissues. These concentrations depend ultimately on the circulating maternal levels of T4 or FT4. Even in euthyroid women, a decrease in this maternal supply results in lower concentrations of FT4 in fetal fluids and, consequently, of T4 available to the developing brain. We here propose that this T4 would determine the amount of T3, generated locally by 5'D-II deiodination, that ultimately binds to thyroid hormone nuclear receptors in different brain structures and triggers biological responses early in gestation in orderly sequence. Ensuring maternal T4 levels appropriate for the stage of pregnancy, regardless of whether TSH is increased or whether overt or subclinical hypothyroidism is detected, may decrease the risk of neurodevelopmental defects in the child (5). A major cause of maternal hypothyroxinemia worldwide is an inadequate iodine intake, which is easily avoided by encouraging pregnant women to take iodine supplements from early pregnancy even in countries such as the U.S. (1).
Acknowledgments
We are grateful to Ms. Socorro Durán and María Jesús Presas for their invaluable technical assistance. We have been very much helped by written discussions with Drs. Carole Spencer and Jerald C. Nelson from Quest Diagnostics, Inc./Nichols Institute Diagnostics (San Juan Capistrano, CA) in understanding principles regarding the measurement of FT4 with kits commercialized for human serum. We are also grateful to the Spanish dealers, Atom, Isaza, Sorin, and Nuclear Ibérica, who kindly provided some FT4 kits free of charge.
Footnotes
This work was supported by Grant 99/0852 from Fondo de Investigaciones Sanitarias and Grant 08.5/0059.1/2000 from Comunidad Autónoma de Madrid (Spain).
Abbreviations: AF, Amniotic fluid; CF, coelomic fluid; 5'D-I, type I 5'-iodothyronine deiodinase; 5'D-II, type II 5'-iodothyronine deiodinase; 5D-III, inner ring iodothyronine deiodinase; FS, fetal serum; FT3, free T3; FT4, free T4; MS, maternal serum; rT2, 3,3'-diiodo-L-thyronine; TBG, T4-binding globulin; TTR, transthyretin.
Received September 10, 2001.
Accepted January 14, 2002.
References
- foetoprotein, 1-antitrysin and transferrin by the human yolk sac. Nature 228:995–997[CrossRef][Medline]
-fetoprotein. Specificities of the phenylbutazone-binding and warfarin-binding sites. Biochem J 239:451–458[Medline]
This article has been cited by other articles:
 |
|
 |
|
  H.-L. Liu, L. T. Lam, Q. Zeng, S.-q. Han, G. Fu, and C.-c. Hou Effects of drinking water with high iodine concentration on the intelligence of children in Tianjin, China J. Public Health Med., March 1, 2009; 31(1): 32 - 38. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Mayor-Lynn, H. J. Rohrs III, A. C. Cruz, J. H. Silverstein, and D. Richards Antenatal Diagnosis and Treatment of a Dyshormonogenetic Fetal Goiter J. Ultrasound Med., January 1, 2009; 28(1): 67 - 71. [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Yonkers and A. B. Ribera Sensory Neuron Sodium Current Requires Nongenomic Actions of Thyroid Hormone During Development J Neurophysiol, November 1, 2008; 100(5): 2719 - 2725. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Moleti, V. P. Lo Presti, M. C. Campolo, F. Mattina, M. Galletti, M. Mandolfino, M. A. Violi, G. Giorgianni, D. De Domenico, F. Trimarchi, et al. Iodine Prophylaxis Using Iodized Salt and Risk of Maternal Thyroid Failure in Conditions of Mild Iodine Deficiency J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2616 - 2621. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Olivieri, E. Medda, S. De Angelis, H. Valensise, M. De Felice, C. Fazzini, I. Cascino, V. Cordeddu, M. Sorcini, M. A. Stazi, et al. High Risk of Congenital Hypothyroidism in Multiple Pregnancies J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3141 - 3147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Subsection Reports J. Clin. Endocrinol. Metab., August 1, 2007; 92(8_suppl): s8 - s47. [Full Text] [PDF]
|
|
|
|
|
|
M. Abalovich, N. Amino, L. A. Barbour, R. H. Cobin, L. J. De Groot, D. Glinoer, S. J. Mandel, and A. Stagnaro-Green Management of Thyroid Dysfunction during Pregnancy and Postpartum: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., August 1, 2007; 92(8_suppl): s1 - s47. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Szinnai, L. Lacroix, A. Carre, F. Guimiot, M. Talbot, J. Martinovic, A.-L. Delezoide, M. Vekemans, S. Michiels, B. Caillou, et al. Sodium/Iodide Symporter (NIS) Gene Expression Is the Limiting Step for the Onset of Thyroid Function in the Human Fetus J. Clin. Endocrinol. Metab., January 1, 2007; 92(1): 70 - 76. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Negro, G. Formoso, T. Mangieri, A. Pezzarossa, D. Dazzi, and H. Hassan Levothyroxine Treatment in Euthyroid Pregnant Women with Autoimmune Thyroid Disease: Effects on Obstetrical Complications J. Clin. Endocrinol. Metab., July 1, 2006; 91(7): 2587 - 2591. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. E. Pedraza, M.-J. Obregon, H. F. Escobar-Morreale, F. Escobar del Rey, and G. M. de Escobar Mechanisms of Adaptation to Iodine Deficiency in Rats: Thyroid Status Is Tissue Specific. Its Relevance for Man Endocrinology, May 1, 2006; 147(5): 2098 - 2108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Y Chan, M. H Andrews, R. Lingas, C. J McCabe, J. A Franklyn, M. D Kilby, and S. G Matthews Maternal nutrient deprivation induces sex-specific changes in thyroid hormone receptor and deiodinase expression in the fetal guinea pig brain J. Physiol., July 15, 2005; 566(2): 467 - 480. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Rajatapiti, M. H. A. Kester, R. R. de Krijger, R. Rottier, T. J. Visser, and D. Tibboel Expression of Glucocorticoid, Retinoid, and Thyroid Hormone Receptors during Human Lung Development J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 4309 - 4314. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Negro, T. Mangieri, L. Coppola, G. Presicce, E. C. Casavola, R. Gismondi, G. Locorotondo, P. Caroli, A. Pezzarossa, D. Dazzi, et al. Levothyroxine treatment in thyroid peroxidase antibody-positive women undergoing assisted reproduction technologies: a prospective study Hum. Reprod., June 1, 2005; 20(6): 1529 - 1533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Simpson, F. L. R. Williams, C. Delahunty, H. van Toor, S.-Y. Wu, S. A. Ogston, T. J. Visser, R. Hume, and with collaboration from the Scottish Preterm Thyro Serum Thyroid Hormones in Preterm Infants and Relationships to Indices of Severity of Intercurrent Illness J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1271 - 1279. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Vermiglio, V. P. Lo Presti, M. Moleti, M. Sidoti, G. Tortorella, G. Scaffidi, M. G. Castagna, F. Mattina, M. A. Violi, A. Crisa, et al. Attention Deficit and Hyperactivity Disorders in the Offspring of Mothers Exposed to Mild-Moderate Iodine Deficiency: A Possible Novel Iodine Deficiency Disorder in Developed Countries J. Clin. Endocrinol. Metab., December 1, 2004; 89(12): 6054 - 6060. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Auso, R. Lavado-Autric, E. Cuevas, F. E. del Rey, G. Morreale de Escobar, and P. Berbel A Moderate and Transient Deficiency of Maternal Thyroid Function at the Beginning of Fetal Neocorticogenesis Alters Neuronal Migration Endocrinology, September 1, 2004; 145(9): 4037 - 4047. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Santiago-Fernandez, R. Torres-Barahona, J. A. Muela-Martinez, G. Rojo-Martinez, E. Garcia-Fuentes, M. J. Garriga, A. G. Leon, and F. Soriguer Intelligence Quotient and Iodine Intake: A Cross-Sectional Study in Children J. Clin. Endocrinol. Metab., August 1, 2004; 89(8): 3851 - 3857. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Hume, J. Simpson, C. Delahunty, H. van Toor, S. Y. Wu, F. L. R. Williams, and T. J. Visser Human Fetal and Cord Serum Thyroid Hormones: Developmental Trends and Interrelationships J. Clin. Endocrinol. Metab., August 1, 2004; 89(8): 4097 - 4103. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. T. Zoeller Editorial: Local Control of the Timing of Thyroid Hormone Action in the Developing Human Brain J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3114 - 3116. [Full Text] [PDF]
|
|
|
|
|
|
M. H. A. Kester, R. Martinez de Mena, M. Jesus Obregon, D. Marinkovic, A. Howatson, T. J. Visser, R. Hume, and G. Morreale de Escobar Iodothyronine Levels in the Human Developing Brain: Major Regulatory Roles of Iodothyronine Deiodinases in Different Areas J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3117 - 3128. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Jauniaux, T. Cindrova-Davies, J. Johns, C. Dunster, J. Hempstock, F. J. Kelly, and G. J. Burton Distribution and Transfer Pathways of Antioxidant Molecules inside the First Trimester Human Gestational Sac J. Clin. Endocrinol. Metab., March 1, 2004; 89(3): 1452 - 1458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. E. Kempers, D. A. van Tijn, A. S. P. van Trotsenburg, J. J. M. de Vijlder, B. M. Wiedijk, and T. Vulsma Central Congenital Hypothyroidism due to Gestational Hyperthyroidism: Detection Where Prevention Failed J. Clin. Endocrinol. Metab., December 1, 2003; 88(12): 5851 - 5857. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. P. Carvalho Modulation of Uterine Iodothyronine Deiodinases--A Critical Event for Fetal Development? Endocrinology, October 1, 2003; 144(10): 4250 - 4252. [Full Text] [PDF]
|
|
|
|
|
|
K. Poppe, D. Glinoer, H. Tournaye, P. Devroey, A. van Steirteghem, L. Kaufman, and B. Velkeniers Assisted Reproduction and Thyroid Autoimmunity: An Unfortunate Combination? J. Clin. Endocrinol. Metab., September 1, 2003; 88(9): 4149 - 4152. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chan, S. Kachilele, E. Hobbs, J. N. Bulmer, K. Boelaert, C. J. McCabe, P. M. Driver, A. R. Bradwell, M. Kester, T. J. Visser, et al. Placental Iodothyronine Deiodinase Expression in Normal and Growth-Restricted Human Pregnancies J. Clin. Endocrinol. Metab., September 1, 2003; 88(9): 4488 - 4495. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Raber, P. Nowotny, E. Vytiska-Binstorfer, and H. Vierhapper Thyroxine treatment modified in infertile women according to thyroxine-releasing hormone testing: 5 year follow-up of 283 women referred after exclusion of absolute causes of infertility Hum. Reprod., April 1, 2003; 18(4): 707 - 714. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) or later, up to 17 wk (
). Values are the mean ± SE. T4 and FT4 are expressed in the same units to permit direct comparisons between both variables. An asterisk discloses statistically significant differences between the first trimester and early second trimester AF samples. CF can no longer be obtained after 12 wk gestation, and only first trimester data are shown.
