Asymmetrical Transport of Thyroxine Across Human Term Placenta

Abstract

Background:

Fetal development is crucially dependent on thyroid hormone (TH). Maternal-to-fetal transfer of TH is a prerequisite for fetal TH availability, particularly in the first half of pregnancy. The mechanisms of transplacental transport of TH, however, are yet poorly understood. We, therefore, investigated the TH transport processes across human placentas using an ex vivo perfusion system.

Methods:

Intact cotyledons from term placentas of uncomplicated pregnancies were cannulated within 30 minutes after delivery and the maternal and fetal circulations were re-established. One hundred nanomolar thyroxine (T4) was added to either the maternal or fetal circulation and perfusions run up to three hours during which samples were taken from both circulations at different time points. Variables included addition of iopanoic acid (IOP) to block activity of the deiodinase type 3 (D3) and bovine serum albumin (BSA) to trap released T4. T4 and 3,3′,5′-triiodothyronine concentrations in the perfusates were measured by radioimmunoassays.

Results:

Maternal-to-fetal transfer was slow, with T4 barely detectable in the fetal circulation unless D3 was blocked by IOP. Fetal T4 was detected after three hours perfusion (10.6 ± 0.6 nM) when BSA (34 g/L) was added in the fetal circulation to trap the released T4. In contrast, fetal-to-maternal transfer of T4 was rapid and maternal T4 increased to 43.6 ± 5.5 nM.

Conclusions:

Maternal-to-fetal T4 transport is limited, whereas fetal-to-maternal transport is rapid indicating that T4 transport across human term placenta is an asymmetrical process. With the high D3 activity, our observations are compatible with a protective role of the placental barrier. Future studies should reveal how the placenta exerts its gatekeeper function in ensuring optimal TH passage to the fetus.

Introduction

Thyroid hormone (TH) is essential for fetal brain development.¹ Since the fetal thyroid gland is not fully functional until gestational weeks 18 to 20, fetal development in the first half of pregnancy largely depends on maternal TH supply.² Even mild alterations in maternal TH concentrations have been associated with a worse neuropsychological performance of the child.^3

–6 This underscores the importance of the adequate supply of TH to the fetus through transplacental transport.

TH transfer across human term placenta was first demonstrated in vivo by Vulsma et al in newborns with fetal thyroid agenesis or total dyshormogenis as 35–70 nM thyroxine (T4) (25–50% of the normal concentrations) was detected in cord blood.⁷ However, when Mortimer and coworkers studied maternal-to-fetal T4 transfer in human term placenta in an ex vivo perfusion system, they found that maximally 4.1 pM T4 was transferred to the fetal circulation when perfused with 150 nM maternal T4 for 6 hours.⁸

Only when the deiodinase type 3 (D3) was inhibited, they measured 10 nM T4 in the fetal circulation.⁸ These seemingly discordant observations indicate that although maternal-to-fetal T4 transport occurs during pregnancy, the transport process itself is slow, at least in term placenta, with extensive inactivation of T4 before it can reach the fetal circulation. However, placental transport is expected to be a bidirectional process, yet little is known about fetal-to-maternal T4 transport.

To further elucidate the T4 transplacental transport processes, we used an ex vivo placental perfusion model and studied T4 transfer across human term placentas in both directions. We found that human term placentas are capable of bidirectional T4 transfer with slow maternal-to-fetal transfer but rapid fetal-to-maternal transfer. Our findings indicate that during late gestation, the placenta mainly serves as a protective barrier to prevent fetal thyrotoxicosis.

Methods

Reagents

Antipyrine, fluorescein isothiocyanate (FITC)-dextran, T4, bovine serum albumin (BSA), and antifoaming A concentrate were purchased from Sigma Aldrich (Zwijndrecht, The Netherlands). Iopanoic acid (IOP) was purchased from Sterling Winthrop and Sigma Aldrich and the blood collection tubes were from BD Vacutainer (Franklin Lakes, New Jersey). IOP was resuspended in 0.1 N NaOH (57.1 mg/0.2 mL) to help dissolving before adding to the perfusion buffer (described in section Placenta perfusion experiments) with a final concentration of 0.5 mM.

Patients and placentas

The study received exemption for approval from the local institutional medical ethics committee according to the Dutch Medical Research with Human Subjects Law (MEC-2017-418). All patients gave written consent before donating their placentas. Healthy placentas of uncomplicated singleton pregnancies were collected immediately after delivery (through cesarean section) at Erasmus University Medical Center, Rotterdam, The Netherlands. Placentas with maternal viral infections (HIV, hepatitis B, Zika, and SARS-CoV-2), maternal diabetes, or fetal congenital abnormalities observed on ultrasound were excluded.

Placental perfusion experiments

The perfusion model was set up as described before (Figs. 1A and 2A).⁹ Perfusion buffer consisted of Krebs–Henseleit buffer (118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 25 mM NaHCO₃, and 8.3 mM glucose), supplemented with 2.5 IU/L heparin LEO and aerated with 95% O₂/5% CO₂).

FIG. 1.

Maternal-to-fetal T4 transfer. (A) Illustration of placenta perfusion experiments of maternal-to-fetal T4 transfer, in the absence or presence of IOP (added in the maternal reservoir) or BSA (added in the fetal reservoir). Antipyrine (added in the maternal reservoir) and FITC-dextran (added in the fetal reservoir) were used as markers of quality control. *If applicable. The placenta and the circulation buffers were maintained at 37°C. (B, C) T4 (B) and rT3 (C) concentrations in the perfusates determined by radioimmunoassay. N = 4 placentas for perfusions in the absence of IOP and BSA (control group), n = 2 for perfusions with IOP (+IOP group) and n = 3 for perfusions with BSA (+BSA group). Data are presented as mean ± SD. Unpaired two-tailed t-test was used for statistical analysis to compare T4 or rT3 concentrations in different groups at t = 180 minutes. **0.001 < p ≤ 0.01. (D–F) Schematic illustration of changes of T4 and rT3 concentrations in maternal-to-fetal T4 perfusion experiments. (D) In the absence of IOP and BSA; (E) in the presence of IOP; (F) in the presence of BSA. BSA, bovine serum albumin; D3, the deiodinase type 3; FITC, fluorescein isothiocyanate; IOP, iopanoic acid; rT3; 3,3′,5′-triiodothyronine; T4, thyroxine; ns, not significant.

FIG. 2.

Fetal-to-maternal T4 transfer. (A) Illustration of placenta perfusion experiments of fetal-to-maternal T4 transfer, in the absence or presence of BSA (added in the maternal reservoir). *If applicable. (B, C) T4 (B) and rT3 (C) concentrations in the perfusates determined by radioimmunoassay. Data are presented as mean ± SD from three placentas. Unpaired two-tailed t-test was used for statistical analysis to compare T4 or rT3 concentrations in different groups at t = 180 minutes. *0.01 < p < 0.05. (D, E) Schematic illustration of changes of T4 and rT3 concentrations in fetal-to-maternal T4 perfusion experiments in the absence (D) or presence (E) of BSA in the maternal circulation.

Placentas were cannulated within 30 minutes after delivery. The fetal circulation was established by cannulating an artery and its corresponding vein. Next, the fetal flow was gradually adjusted from 1 to 6 mL/min and maintained afterward. The maternal circulation was established by inserting four cannulas into the intervillous space of the corresponding cotyledon and the outflow was collected into a chamber beneath and brought back to the maternal reservoir. The maternal flow was kept constant at 12 mL/min. After running as open circulations for ∼45 minutes to wash out the blood of the placenta, both maternal and fetal circulations were closed at the start of the experiment and the buffers were switched to 200 mL fresh perfusion buffers containing the substances as described hereunder.

Antipyrine was used as a positive marker for the sufficient overlap between the maternal and fetal circulations as it is able to diffuse across the placental barrier. It was added to the maternal buffer with a final concentration of 110 mg/mL. FITC-dextran (molecular weight 40 kDa) was used as a marker of integrity of the capillary bed. It was added to the fetal buffer with a final concentration of 39.5 mg/mL.

To study maternal-to-fetal T4 transfer, 200 μL of 100 μM T4 (final concentration 100 nM) and, if applicable, 57.1 mg IOP (final concentration 0.5 mM) were added to the maternal buffer. If applicable, 6.8 g BSA (final concentration 34 g/L) was added to the fetal buffer (Fig. 1A).

To study fetal-to-maternal T4 transfer, 200 μL of 100 μM T4 was added to the fetal buffer and 5.8 g BSA (final concentration 29 g/L) to the maternal buffer (if applicable) (Fig. 2A). In case of addition of BSA, antifoaming A concentrate was applied to the top edge of the cylinders to prevent excessive foaming.

To test endogenous T4 release from the term placentas, only the quality control molecules antipyrine and FITC-dextran were applied to the perfusion buffer. Neither T4 nor other compounds were added in the experiments.

One milliliter samples were collected from the reservoirs into the blood collection tubes during three hours perfusion at indicated time points. The samples were centrifuged at 1680 g for 10 minutes and the supernatants were collected as perfusates and stored at −20°C.

Quality controls of the placental perfusion experiments

The concentrations of antipyrine and FITC-dextran in the perfusates were measured as described previously.⁹ A fetal-to-maternal (F/M) ratio of antipyrine lower than 0.75 at t = 180 minutes indicates that one or more cannulas were not cannulated into the intervillous space of the corresponding cotyledon leading to insufficient overlap between maternal and fetal circulations. A maternal-to-fetal (M/F) ratio of FITC-dextran higher than 0.03 at t = 180 minutes indicates compromised intactness of the placental barrier. Therefore, the perfusion experiments were excluded from further analysis when the F/M ratio of antipyrine is <0.75 or the M/F ratio of FITC-dextran >0.03 at t = 180 minutes.

Radioimmunoassays

Total T4 and 3,3′,5′-triiodothyronine (rT3) concentrations were measured in technical duplicates using radioimmunoassays as described previously.¹⁰ The quantification range for T4 is from 1.3 to 321.8 nM and that for rT3 is from 0.02 to 15.36 nM. The intra-assay coefficients of variation (CVs) were 2–8% for T4 and 3–4% for rT3. The inter-assay CVs were 5–10% for T4 and 9–16% for rT3.¹⁰

Statistical analysis

The data are presented as mean ± SD of three experiments (one placenta/experiment) unless indicated otherwise. GraphPad Prism 8.4.0 was used for data analysis and an unpaired two-tailed t-test was used to test for statistical significance analysis of the differences in T4 and rT3 concentrations at the end of the perfusion. p-Values <0.05 were considered as significant.

Results

Maternal-to-fetal T4 transfer

To study maternal-to-fetal T4 transfer, 100 nM T4 was added to the maternal reservoir. After 3 hours perfusion, T4 in the maternal circulation decreased from 68.4 ± 11.2 to 47.7 ± 10.4 nM, although T4 remained below the limit of detection in the fetal circulation (Fig. 1B). Instead, maternal and fetal rT3 increased and reached 5.7 ± 1.0 and 0.8 ± 0.1 nM, respectively, by the end of perfusion (Fig. 1C).

T4 entering the maternal side of the placenta may be converted to rT3 by D3. Therefore, we blocked D3 activity by adding 0.5 mM IOP to the maternal reservoir. In the presence of IOP, up to 2.7 ± 0.4 nM T4 was detected in the fetal circulation, while rT3 concentrations remained low in maternal (0.6 ± 0.07 nM) and fetal (0.1 ± 0.01 nM) circulations throughout perfusion (Fig. 1B, C). Thus, after inhibition of D3 activity, intact maternal T4 was transported to the fetal circulation.

Next, we examined whether the lack of circulating TH binding proteins in our model could explain the limited maternal-to-fetal T4 transfer. Thus, to mimic the concentrations found in maternal and cord blood, 29 g/L BSA was added in the maternal reservoir and 34 g/L BSA in the fetal reservoir. When BSA was added to both circulations, we hardly detected T4 in the fetal circulation (Supplementary Fig. S1). This is most likely because under these conditions, the vast majority of T4 will be bound to BSA. This reduces the amount of free T4 available for transport in the maternal circulation to such levels that the rate of transport will be too slow for T4 to accumulate in the fetal circulation within the time of perfusion.

When BSA was only added to the fetal reservoir, the decrease in maternal T4 was not significantly different from perfusions without BSA (69.8 ± 17.8 to 42.4 ± 5.4 nM with BSA vs. 68.4 ± 11.2 to 47.7 ± 10.4 nM without BSA). However, fetal T4 increased steadily to 10.6 ± 0.6 nM by the end of perfusion (Fig. 1B), indicating that maternal T4 was transferred to and accumulated in the fetal circulation. In addition, maternal rT3 concentrations at the end of perfusion did not differ (3.8 ± 0.5 nM with BSA vs. 5.7 ± 1.0 nM without BSA), but fetal rT3 concentrations were significantly higher when BSA was present in the fetal circulation (4.4 ± 1.2 nM with BSA vs. 0.8 ± 0.1 nM without BSA) (Fig. 1C).

Fetal-to-maternal T4 transfer

The difference of maternal-to-fetal T4 transfer between the conditions in the absence and presence of BSA in the fetal circulation could imply the reuptake of fetal T4 and subsequent reverse transport into the placenta. Therefore, we explored fetal-to-maternal transfer by adding T4 to the fetal reservoir (Fig. 2A). Fetal T4 decreased rapidly from 59.8 ± 12.1 to 8.2 ± 2.1 nM after 3 hours perfusion and maternal T4 increased to 43.6 ± 5.5 nM (Fig. 2B). It suggests that the placenta is able to clear excessive T4 efficiently from the fetal circulation.

When we added BSA to the maternal circulation to capture any T4 transferred to the maternal side, fetal T4 decreased to 6.4 ± 1.7 nM after 3 hours perfusion and maternal T4 increased to 53.5 ± 3.6 nM (Fig. 2B). However, this was not significantly different from conditions wherein no BSA was added to the maternal circulation. Fetal rT3 concentrations were significantly reduced in the presence compared with absence of BSA in the maternal circulation (Fig. 2C), possibly due to shorter retention of T4 in placenta leading to less conversion by D3. Taken together, fetal-to-maternal T4 transfer in human term placentas is a rapid process, even when T4 is not trapped by BSA in the maternal circulation.

T4 release from the human term placenta

Finally, we examined whether endogenous T4 present in collected placentas was also preferentially transferred to the maternal side. For this, we perfused placentas without addition of exogenous T4 and BSA. T4 in the maternal circulation reached 3.8 ± 1.2 nM after 3 hours perfusion, while fetal T4 was not detectable (Fig. 3A). Maternal rT3 increased to 0.8 ± 0.4 nM and fetal rT3 only increased to 0.05 ± 0.03 nM (Fig. 3B). Together, this suggests that also endogenous T4 in the placentas is preferentially transported to the maternal side.

FIG. 3.

Endogenous T4 release from human term placenta. T4 (A) and rT3 (B) concentrations in the perfusates from placenta perfusions without adding exogenous T4. T4 (A) and rT3 (B) concentrations in the perfusates determined by radioimmunoassay. Fetal T4 was not detectable at all time points. Data are represented as mean ± SD from three placentas.

Discussion

The main finding in our ex vivo human placental model is the presence of bidirectional T4 transplacental transport. We found that maternal-to-fetal T4 transfer in human term placentas is limited, which is in agreement with the observations by Mortimer et al.⁸ However, we are the first to report that fetal T4 is able to transfer to the maternal side rapidly and this process is not depending on the presence of the TH binding protein BSA.

In the absence of any exogenous TH binding proteins, we found that little T4 reaches the fetal circulation unless D3 activity is blocked by IOP, which is in agreement with previous observations.⁸ However, when the TH binding protein BSA is added to the fetal circulation, maternal T4 accumulates in the fetal circulation. This suggests that T4 is transferred from maternal to fetal with BSA preventing T4 being reabsorbed by the placenta. Our findings of rapid fetal-to-maternal T4 transfer and placental endogenous T4 preferentially transferring to the maternal circulation also support the idea of the placenta reabsorbing T4 from the fetal circulation, even at low T4 concentrations.

The results of maternal-to-fetal T4 transfer indicate that the activity of the deiodinases may affect the T4 transplacental transfer. According to Stulp et al, mRNA expression of the deiodinase type 2 (D2) and D3 is present in human placentas with gestational age of 13 weeks, 31 weeks, and term whereas the deiodinase type 1 (D1) is absent.¹¹ In our perfusion experiments with T4, no significant conversion of T4 to T3 in placental perfusates was found as we measured negligible T3 in the fetal circulation in a subset of perfusates by radioimmunoassay (data not shown). This is in agreement with a previous study that shows at all gestational ages that D2 activity is ∼200 times lower than that of D3 and falls to virtually 0 at term.¹²

Though D3 activity relative to protein or DNA decreases with increasing gestational age, the total D3 activity in the human placenta increases.¹² This implies that more maternal T4 may be transferred to the fetal side during earlier gestational age when the total D3 activity is lower than at term and when the fetus is more dependent on maternal TH supply. In contrast, D3 converting T4 to rT3 in placentas during late pregnancy, when the fetal thyroid gland is functioning, releases iodide that may serve as a supply for TH synthesis in the fetus.

The rapid fetal-to-maternal T4 transfer suggests that the placenta protects the fetus from thyrotoxicosis not only by limiting maternal-to-fetal T4 transfer, but also by rapidly removing excessive T4 in the fetal circulation. This would suggest that at term, the healthy fetus mostly controls its own TH supply with limited dependence on the maternal supply. Only in the absence of fetal TH production substantial amounts of maternal T4 may still be transferred to the fetus until an equilibrium, although with lower fetal T4, is reached, as seen in the case of fetal thyroid agenesis and dyshormogenesis.⁷

The rapid fetal-to-maternal T4 transfer also implies that at the fetal side, probably in the endothelial cells of the blood vessels, the first layer of cells that encounters fetal T4 expresses efficient T4 transporter(s). In addition, potent T4 transporter(s) for T4 influx at the basolateral membrane and for T4 efflux at the apical membrane are necessary to achieve this rapid process. It deserves further investigation which TH transporter(s) are involved. Furthermore, this finding also implies that the TH metabolites generated in the placenta or produced by the fetus may be transferred to the maternal circulation to accumulate to detectable concentrations, for example, the fetal derived compound W, a potential biomarker of the status of the fetal thyroid.¹³

Though placental perfusion model has been used for studies of transport and metabolism of other hormones such as parathyroid hormone, cortisol, cortisone, and serotonin in human term placenta,^14,15 there are several limitations of using this model to study TH transplacental transport.

First, we only used human term placentas. Although it is of interest to investigate TH transfer across placentas during early pregnancy when the fetus is fully dependent on maternal supply, rare availability of placentas from early and healthy pregnancy and practical unfeasibility to cannulate such placentas restricted us from selecting such placentas for perfusion. Second, we did not add TH binding proteins to the maternal reservoir during maternal-to-fetal perfusion or to the fetal reservoir during fetal-to-maternal perfusion, because addition of BSA may result in limited free fraction of T4 and IOP, like many medicines used in perfusion studies.¹⁶ The lack of TH binding proteins may lead to super-physiological concentrations of the free fraction of T4.

Third, to ensure the intactness and functionality of the placentas in vitro, we only perfused the placentas for three hours and this duration may not be long enough for significant T4 transfer. A previous study by Dussault and Coulombe estimated that <1% of maternally infused ¹²⁵I- T4 is transferred to the fetus after 3 hours infusion in near-term rats,¹⁷ whereas Morreale de Escobar et al showed that 17.5% of fetal T4 derives from transplacental transfer during late gestation after 10 days infusion.¹⁸ We speculate that longer maternal T4 exposure ensures more T4 being transferred to the fetus.

In summary, we provide evidence for the first time that T4 transport can be bidirectional and it is asymmetrical with the directions in human term placenta. Our findings suggest that human term placenta protects the fetus from excessive TH action. Future studies should reveal the key players regulating this transport process.

Footnotes

Acknowledgments

We thank all the patients participated in this study and their willingness to donate their placentas. We thank the gynecologist Sam S. Schoenmarker and the assistants in the operation rooms who helped with placenta logistics. We thank Barbara M. van der Linden and Lotte W. Voskamp for obtaining informed consents from the patients. We are grateful for the assistance in antipyrine measurements provided by the department of pharmacy at Erasmus MC.

Authors' Contributions

Z.C., M.E.M., W.E.V., and R.P.P. designed the study. Z.C., M.B., R.I.N., E.H., and L.T. Performed the perfusion experiments. Z.C. And S.L. Performed the radioimmunoassays and processed the experimental data. Z.C. Analyzed the data. Z.C., M.E.M., W.E.V., and R.P.P. wrote the article. All authors critically reviewed, revised, and approved the article.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study is funded by a Vidi grant (016.176.331) from the Netherlands Organization for Scientific Research to R.P.P. This research is funded by the EU Horizon 2020 program, ATHENA project, grant number 825161, which is gratefully acknowledged. This publication reflects only the authors' view, and the European Commission is not responsible for any use that may be made of the information it contains.

Supplementary Material

Supplementary Figure S1

References

Oppenheimer

, Schwartz

. Molecular basis of thyroid hormone-dependent brain development. Endocr Rev, 1997; 18(4):462–475; doi: 10.1210/edrv.18.4.0309

Patel

, Landers

, Li

, et al. Delivery of maternal thyroid hormones to the fetus. Trends Endocrinol Metab, 2011; 22(5):164–170; doi: 10.1016/j.tem.2011.02.002

Korevaar

, Muetzel

, Medici

, et al. Association of maternal thyroid function during early pregnancy with offspring IQ and brain morphology in childhood: A population-based prospective cohort study. Lancet Diabetes Endocrinol, 2016; 4(1):35–43;, doi: 10.1016/S2213-8587(15)00327-7

Jansen

, Korevaar

TIM

, Mulder

, et al. Maternal thyroid function during pregnancy and child brain morphology: A time window-specific analysis of a prospective cohort. Lancet Diabetes Endocrinol, 2019; 7(8):629–637; doi: 10.1016/S2213-8587(19)30153-6

Pop

, Kuijpens

, van Baar

, et al. Low maternal free thyroxine concentrations during early pregnancy are associated with impaired psychomotor development in infancy. Clin Endocrinol (Oxf), 1999; 50(2):149–155; doi: 10.1046/j.1365-2265.1999.00639.x

, Shan

, Teng

, et al. Abnormalities of maternal thyroid function during pregnancy affect neuropsychological development of their children at 25–30 months. Clin Endocrinol (Oxf), 2010; 72(6):825–829; doi: 10.1111/j.1365-2265.2009.03743.x

Vulsma

, Gons

, de Vijlder

. Maternal-fetal transfer of thyroxine in congenital hypothyroidism due to a total organification defect or thyroid agenesis. N Engl J Med, 1989; 321(1):13–16; doi: 10.1056/NEJM198907063210103

Mortimer

, Galligan

, Cannell

, et al. Maternal to fetal thyroxine transmission in the human term placenta is limited by inner ring deiodination. J Clin Endocrinol Metab, 1996; 81(6):2247–2249; doi: 10.1210/jcem.81.6.8964859

Hitzerd

, Broekhuizen

, Mirabito Colafella

, et al. Placental effects and transfer of sildenafil in healthy and preeclamptic conditions. EBioMedicine, 2019; 45:447–455; doi: 10.1016/j.ebiom.2019.06.007

10.

Hume

, Simpson

, Delahunty

, et al. Human fetal and cord serum thyroid hormones: Developmental trends and interrelationships. J Clin Endocrinol Metab, 2004; 89(8):4097–4103; doi: 10.1210/jc.2004-0573

11.

Stulp

, de Vijlder

, Ris-Stalpers

. Placental iodothyronine deiodinase III and II ratios, mRNA expression compared to enzyme activity. Mol Cell Endocrinol, 1998; 142(1–2):67–73; doi: 10.1016/s0303-7207(98)00118-x

12.

Koopdonk-Kool

, de Vijlder

, Veenboer

, et al. Type II and type III deiodinase activity in human placenta as a function of gestational age. J Clin Endocrinol Metab, 1996; 81(6):2154–2158; doi: 10.1210/jcem.81.6.8964844

13.

, Huang

, Ho

, et al. Compound W, a 3,3′-diiodothyronine sulfate cross-reactive substance in serum from pregnant women—A potential marker for fetal thyroid function. Pediatr Res, 2007; 61(3):307–312; doi: 10.1203/pdr.0b013e318030d17d

14.

Balabanova

, Lang

, Wolf

, et al. Placental transfer of parathyroid hormone. J Perinat Med, 1986; 14(4):243–250; doi: 10.1515/jpme.1986.14.4.243

15.

Mathiesen

, Bay-Richter

, Wegener

, et al. Maternal stress and placental function; ex vivo placental perfusion studying cortisol, cortisone, tryptophan and serotonin. PLoS One, 2020; 15(6):e0233979; doi: 10.1371/journal.pone.0233979

16.

Schneider

, Albrecht

, Ahmed

, et al. Ex vivo dual perfusion of an isolated human placenta cotyledon: Towards protocol standardization and improved inter-centre comparability. Placenta, 2022; 126:83–89; doi: 10.1016/j.placenta.2022.05.003

17.

Dussault

, Coulombe

. Minimal placental transfer of L-thyroxine (T4) in the rat. Pediatr Res, 1980; 14(3):228–231; doi: 10.1203/00006450-198003000-00010

18.

Morreale de Escobar

, Calvo

, Obregon

, et al. Contribution of maternal thyroxine to fetal thyroxine pools in normal rats near term. Endocrinology, 1990; 126(5):2765–2767; doi: 10.1210/endo-126-5-2765