Disrupted trans-placental thyroid hormone transport in a human model for MCT8 deficiency
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (Review Commons)
Abstract
Article activity feed
-
Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Reply to the reviewers
The authors show expression of the thyroid hormone transporter MCT8 in the human placenta. The MCT8-inhibiting compound sylchristine reduces the transfer of T4 from the maternal to the fetal side and accordingly reduces T4 degradation by DIO3. Since maternal thyroid hormones are relevant for neurodevelopment before the onset of fetal thyroid gland function, disruption of MCT8 transport, as occurs in the Allan-Herdon-Dudley syndrome, may contribute to the neurodevelopment failure present in these patients.
The results of the transfer experiments are clear and support the authors' conclusions.
We thank the reviewer for this positive statement.
Minor …
Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Reply to the reviewers
The authors show expression of the thyroid hormone transporter MCT8 in the human placenta. The MCT8-inhibiting compound sylchristine reduces the transfer of T4 from the maternal to the fetal side and accordingly reduces T4 degradation by DIO3. Since maternal thyroid hormones are relevant for neurodevelopment before the onset of fetal thyroid gland function, disruption of MCT8 transport, as occurs in the Allan-Herdon-Dudley syndrome, may contribute to the neurodevelopment failure present in these patients.
The results of the transfer experiments are clear and support the authors' conclusions.
We thank the reviewer for this positive statement.
Minor comments:
- Line 71: please provide some references on the immaturity of the blood.brain barrier before 18 months. The endothelial cells may have tight junctions when the vessels sprout in the CNS. "Maturity" implies the full complement of the neurovascular unit, i.e., pericytes and astrocytes. So, please clarify this point, even if it does not contradict the experimental results showing the role of placental MCT8.
We acknowledge that there is debate when the human blood-brain barrier is regarded as mature. Tight junctions of endothelial cells are functional from week 14 in fetal development (Saili et al, DOI: 10.1002/bdr2.1180) and reach functionality comparable with adult blood-brain barrier from 18 weeks onwards (DOI: 10.1016/j.placenta.2016.12.005). A fully functional blood-brain barrier requires interaction with a range of cells, including pericytes, astrocytes, microglia and neurons (DOI: 10.1016/j.placenta.2016.12.005), which matures during the entire pregnancy (e.g. cortical astrocytes start to appear from 30 weeks onwards).
In our manuscript, we do not intend to overstate the relevance of our findings. Hence, in the revised manuscript, we changed the wording in lines 70-71 from ‘’mature’’ to ‘’functional’’. This avoids the discussion on when the human blood-brain barrier is mature, while conveying the message that the placental barrier is key in determining bioavailability of thyroid hormone for the fetal brain.
It is unclear if the partial effect of sylchristine on T4 transport means that MCT8 contribution is also partial and other transporters contribute.
We thank the reviewer for raising this point. Our in vitro data (Figure Expanded View 2) showed that at 10 µM concentration silychristin fully inhibits MCT8, agreeing with previous data by others (Johannes et al, DOI: 10.1210/en.2015-1933). As MCT8 is expressed at the apical membrane of the syncytiotrophoblasts which is in direct contact of maternal circulation, it can be inferred that T4 entering the placenta via MCT8 is fully inhibited. In our manuscript, we show that the application of silychristin on the maternal circulation leads to a 60% reduction of T4 accumulation at the fetal side, with the remaining 40% of fetal T4 corresponding to an absolute concentration of ~ 4 nM T4. Of note, we previously showed in the same placenta model that there is ~4 nM T4 endogenously present in the placenta (see Figure 3, DOI: 10.1089/thy.2022.0406). This endogenous placental T4 can be transferred to the fetal circulation; this latter process is not blocked by silychristin, which is only present in the maternal circulation. As adding silychristin results in only ~ 4 nM T4 appearing at the fetal side, equal to the endogenous concentration, it is likely that the contribution by other transporters is minimal.
To clarify this, we have added this in the Discussion of the revised manuscript (lines 117-120).
Lines 113-114. I missed the controls measuring TRIAC transport without sylchristine, or do the authors have strong reasons to assume that sylchristine does not affect TRIAC transport? If so, it should be stated.
We thank the reviewer for raising this question. No human TRIAC transporters have been published and, hence, we cannot exclude the possibility that silychristin may inhibit TRIAC transport. However, we previously showed that human MCT8 does not induce TRIAC uptake (Figure 7, DOI: 10.1089/thy.2019.0009), indicating that TRIAC transport is MCT8 independent. In a previous study, we tested the specificity of silychristin for the thyroid hormone transporters expressed in human term placenta (Figure 2 in DOI: 10.1089/thy.2021.0503)). Silychristin potently inhibited MCT8 with an IC50 of 0.12 µM and at a much higher concentration (10 µM) it also inhibited OATP1A2 by ~40% but none of the other transporters. However, OATP1A2 does not transport TRIAC (unpublished data). Therefore, we feel that silychristin is unlikely to have relevant effects on other placental thyroid hormone transporters that may facilitate TRIAC transport. Hence, we did not include experiments of TRIAC transport in the absence of silychristin.
Our aim was to provide a proof-of-concept that TRIAC is very efficiently transported across human placenta when MCT8 is inhibited. Should the reviewer insist to perform TRIAC transport in the absence of silychristin, we would be happy to do so.
In the revised manuscript, we have included this point as a limitation in our study (lines 146-151).
Reviewer #1 (Significance (Required)):
This study confirms the presence of MCT8 in the human placenta and adds additional data to demonstrate its functionality with experiments using placental perfusion.
We are pleased to see that the reviewer agrees that our data are a relevant addition to the field.
Reviewer #2 (Evidence, reproducibility and clarity (Required)):
Short summary of findings:
The authors used silychristin to selectively block the thyroid hormone transporter MCT8 in perfused human placenta. This was done to model the MCT8 deficient placenta in the rare genetic condition called Allan Herndon Dudley Syndrome. They then showed that the thyroid hormone analogue TRIAC can still cross the placenta and may be a potential treatment to prevent some of the effects of thyroid hormone deficiency in the affected fetus.
Major comments:
The absence of binding proteins in the maternal circulation is an issue since protein bound thyroid hormones can also be taken up by trophoblasts. Additionally, the placenta produces and secretes transthyretin and albumin into the maternal circulation - was this taken into account?
We thank the reviewer for raising this point. We are aware of thyroid binding proteins such as transthyretin (TTR) (eg. DOI: 10.1016/j.placenta.2013.05.005; DOI: 10.1016/j.placenta.2012.01.006; DOI: 10.1210/jc.2009-0048). From such data, it has been established that the placenta secretes TTR. Also, it has been shown that the TTR-T4 complex can be internalized into Jeg3 cells. However, to our knowledge, there is no direct evidence showing that the TTR-T4 complex is transported across human placenta reaching the fetal circulation.
As we have mentioned in the response to the comment 2 of Reviewer 1, adding silychristin results in only ~ 4 nM T4 appearing at the fetal side, equal to the endogenous concentration present in the placenta. Therefore it is likely that the contribution by other transporters or transport mechanisms such as TTR-T4 is minimal.
A caveat to the abovementioned arguments is that our perfusion only lasted for 3 hours because longer perfusions will lead to loss of intactness of the placenta and, hence, less functionality. Therefore, it cannot be excluded that during longer exposures TTR might have a role.
Following this reviewer’s comment, we added this as a limitation to our model in the revised manuscript (lines 151-153).
Other studies have suggested that T4 cannot cross the placenta unless the type 3 deiodinase is blocked which differs from this study. This paper should be referenced and discussed.
We are aware of the study by Mortimer et al (DOI: 10.1210/jcem.81.6.8964859) that showed T4 could transport across human term placenta only when D3 was blocked by iopanoic acid. In our previous study (DOI: 10.1089/thy.2022.0406), we confirmed their findings in perfusion experiments with and without iopanoic acid on maternal-to-fetal T4 transfer (Figure 1), which we discussed in the discussion of our previous publication.
However, in that previous study we also found that transport of T4 in human term placenta is asymmetrical with fetal-to-maternal transfer being more rapid than maternal-to-fetal transfer. However, when adding albumin (BSA) to the fetal circulation (which was not done by Mortimer et al), we prevented re-uptake of T4 and were able to show fetal T4 accumulation in the absence of iopanoic acid. Therefore, we optimized the model by maintaining the physiological conditions in which the type 3 deiodinase is present.
Following the reviewer’s suggestion, we discussed this paper between lines 144-146.
How specific is the action of silychristin? Does it have effects on other thyroid hormone transporters that may facilitate TRIAC transport?
In a previous study, we tested the specificity of silychristin for the thyroid hormone transporters expressed in human term placenta (Figure 2 in DOI: 10.1089/thy.2021.0503). Silychristin potently inhibited MCT8 with an IC50 of 0.12 µM and at a much higher concentration 10 µM it also inhibited OATP1A2 by ~40%. However, OATP1A2 does not transport TRIAC (unpublished data). Therefore, we feel that silychristin is unlikely to have relevant effects on other placental thyroid hormone transporters that may facilitate TRIAC transport.
We have added discussion about this between lines 146-151.
Although TRIAC is able to cross the placenta, it is likely that it still would not be able to cross into the fetal brain making its use somewhat limited. Additionally, it has been suggested that TRIAC exposure may also be a neurodevelopmental risk ((Barez-Lopez et al., 2016) and (Yamauchi et al. 2022 TRIAC disrupts cerebral thyroid hormone action via a negative feedback loop and heterogenous distribution among organs. BioRXiv). This should be discussed.
We thank the reviewer for raising this important point. We would like to respectfully point out that TRIAC in different animal models for MCT8 deficiency has been able to restore abnormal brain development (DOI: 10.1210/me.2014-1135, DOI: 10.3390/ijms232415547, DOI: 10.1530/JOE-16-0323, DOI: 10.1242/dmm.027227).
Specifically, TRIAC administration between postnatal day 1 and 12 restored T3-dependent neural differentiation in the cerebral and cerebellar cortex in Mct8/Oatp1c1 double knockout mice, which represents a relevant mouse model recapitulating the neurological phenotype in patients with MCT8 deficiency (doi: 10.1210/me.2014-1135).
Barez-Lopez (2016) administered TRIAC to Mct8 single knock-out mice. As Oatp1c1 is a redundant T4 transporter in mice, brains of these animals are only mildly hypothyroid and do not recapitulate the severity of the phenotype seen in humans. Hence, we disagree with the conclusion of these authors as they utilized a non-optimal mouse model (DOI: 10.1210/jc.2012-3759). Yamauchi et al. (doi: 10.1016/j.isci.2023.107135) showed that TRIAC content in cerebral cortex did not increase after oral administration of TRIAC after postnatal day 21 in euthyroid and hypothyroid mice. Moreover, they utilized the same dose as T3 as comparison, while TRIAC should be dosed 10-times higher ( DOI: 10.1210/me.2014-1135). Using such a dose, it is very much understandable that TRIAC only affects the hypothalamus-pituitary-thyroid axis, but is insufficient to exert thyroid hormone action in the brain.
We would like to emphasize that there is >70-year experience with TRIAC in humans for other conditions. Neurotoxicity has never been observed. Currently, TRIAC is being studied in high dosages in young children with MCT8 deficiency. The study protocols have been approved by different Ethics Committees as well has been discussed with regulatory authorities.
In the revised manuscript, we have incorporated some information that there is sufficient data in different animal models showing that TRIAC is able to enter the brain.
The data for the control group was used in a previous publication - were the data collected at the same time? Ideally, the control vs silichrystin treated placental cotyledons should be from the same placental samples.
We agree with the reviewer that ideally the control and silychristin treated cotyledons should be matched from the same placenta. However, in practice, it is extremely difficult to realize this for many reasons. In our perfusion experiments, only ~30% of the perfusions succeeded as determined by the criteria of the quality controls (antipyrine and FITC-dextran). Moreover, using two cotyledons from one placenta is not feasible for different reasons (e.g. absence of two intact cotyledons due to damage during delivery; one cotyledon is intact during perfusion, whereas the other is not; one cotyledon has maternal antipyrine diffused to the fetal circulation whereas the other one is not. Therefore, due to practical obstacles and strict quality control criteria, it is not likely to obtain such data of these from different placentas.
Minor comments
T4 and TRIAC were prepared in 0.1N NaOH and silychristin in DMSO. Do NaOH or DMSO affect membrane transporters? Were vehicle controls used in the perfusion experiment?
We dissolved T4 and TRIAC in 0.1 N NaOH and silychristin in DMSO and added them to the perfusion buffer at a 1000 times dilution. For NaOH, it is commonly used in transport assays. Such NaOH and DMSO dilutions did not affect thyroid hormone transport in COS1 cells; therefore we did not include vehicle control DMSO in perfusion experiments.
Were the silichrystin vs control samples matched from the same placentas?
The silychristin and control samples were not matched from the same placentas for the reasons mentioned above.
In Figure 1C, why is there an increase in TRIAC on the maternal side between the first and second time points?
As we sometimes observed in our perfusion experiments with other compounds, at t=0 min (the first time point), the buffer is not aerated and still heterogeneous, leading to differences in the measured concentrations of the TRIAC. Therefore we also included t=6 min (the second time point) to get a more accurate starting concentration.
In figure 1C, TRIAC movement from maternal to fetal side in silichrystin treated placenta is shown but there is no data from untreated placenta? TRIAC transport may be reduced but we cannot tell without a control to compare it to. This should be included.
We thank the reviewer for raising this question, which is similar to comment 3 of Reviewer 1. Hence, we would like to refer to our response to the comment 3 of Reviewer 1.
Reviewer #2 (Significance (Required)):
This study is interesting and adds to the current gaps in the knowledge of transplacental thyroid hormone transport. There is still very limited information around how thyroid hormones cross the placenta despite many groups working on this over the years. However, the manuscript would be more interesting if the placental transporter for TRIAC was identified. There are several clinical trials already underway looking at how TRIAC therapy may be useful in this condition and it may even be detrimental. If TRIAC can cross the placenta its use may still be problematic since others have shown that it cannot cross the into the MCT8 deficient brain from the circulation and must be delivered directly into the brain (Barez-Lopez et al., 2019). This is a small study and is fairly limited however it would be interesting to those, like me, with an interest in endocrinology, placental biology and pregnancy.
Barez-Lopez, S., Grijota-Martinez, C., Liao, X.H., Refetoff, S. and Guadano-Ferraz, A., 2019. Intracerebroventricular administration of the thyroid hormone analog TRIAC increases its brain content in the absence of MCT8, PLoS One. 14, e0226017.
Barez-Lopez, S., Obregon, M.J., Martinez-de-Mena, R., Bernal, J., Guadano-Ferraz, A. and Morte, B., 2016. Effect of Triiodothyroacetic Acid Treatment in Mct8 Deficiency: A Word of Caution, Thyroid. 26, 618-26.
We are pleased to see that the reviewer agrees that our data are a relevant addition to the field. We have alluded to the discussion in the field in the revised manuscript (lines 129-132).
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
Short summary of findings:
The authors used silychristin to selectively block the thyroid hormone transporter MCT8 in perfused human placenta. This was done to model the MCT8 deficient placenta in the rare genetic condition called Allan Herndon Dudley Syndrome. They then showed that the thyroid hormone analogue TRIAC can still cross the placenta and may be a potential treatment to prevent some of the effects of thyroid hormone deficiency in the affected fetus.
Major comments:
The absence of binding proteins in the maternal circulation is an issue since protein bound thyroid hormones can also be taken up by trophoblasts. …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
Short summary of findings:
The authors used silychristin to selectively block the thyroid hormone transporter MCT8 in perfused human placenta. This was done to model the MCT8 deficient placenta in the rare genetic condition called Allan Herndon Dudley Syndrome. They then showed that the thyroid hormone analogue TRIAC can still cross the placenta and may be a potential treatment to prevent some of the effects of thyroid hormone deficiency in the affected fetus.
Major comments:
The absence of binding proteins in the maternal circulation is an issue since protein bound thyroid hormones can also be taken up by trophoblasts. Additionally, the placenta produces and secretes transthyretin and albumin into the maternal circulation - was this taken into account?
Other studies have suggested that T4 cannot cross the placenta unless the type 3 deiodinase is blocked which differs from this study. This paper should be referenced and discussed.
How specific is the action of silychristin? Does it have effects on other thyroid hormone transporters that may facilitate TRIAC transport?
Although TRIAC is able to cross the placenta, it is likely that it still would not be able to cross into the fetal brain making its use somewhat limited. Additionally, it has been suggested that TRIAC exposure may also be a neurodevelopmental risk ((Barez-Lopez et al., 2016) and (Yamauchi et al. 2022 TRIAC disrupts cerebral thyroid hormone action via a negative feedback loop and heterogenous distribution among organs. BioRXiv). This should be discussed. The data for the control group was used in a previous publication - were the data collected at the same time? Ideally, the control vs silichrystin treated placental cotyledons should be from the same placental samples.
Minor comments
T4 and TRIAC were prepared in 0.1N NaOH and silychristin in DMSO. Do NaOH or DMSO affect membrane transporters? Were vehicle controls used in the perfusion experiment?
Were the silichrystin vs control samples matched from the same placentas?
In Figure 1C, why is there an increase in TRIAC on the maternal side between the first and second time points? In figure 1C, TRIAC movement from maternal to fetal side in silichrystin treated placenta is shown but there is no data from untreated placenta? TRIAC transport may be reduced but we cannot tell without a control to compare it to. This should be included.
Significance
This study is interesting and adds to the current gaps in the knowledge of transplacental thyroid hormone transport. There is still very limited information around how thyroid hormones cross the placenta despite many groups working on this over the years. However, the manuscript would be more interesting if the placental transporter for TRIAC was identified. There are several clinical trials already underway looking at how TRIAC therapy may be useful in this condition and it may even be detrimental. If TRIAC can cross the placenta its use may still be problematic since others have shown that it cannot cross the into the MCT8 deficient brain from the circulation and must be delivered directly into the brain (Barez-Lopez et al., 2019). This is a small study and is fairly limited however it would be interesting to those, like me, with an interest in endocrinology, placental biology and pregnancy.
Barez-Lopez, S., Grijota-Martinez, C., Liao, X.H., Refetoff, S. and Guadano-Ferraz, A., 2019. Intracerebroventricular administration of the thyroid hormone analog TRIAC increases its brain content in the absence of MCT8, PLoS One. 14, e0226017.
Barez-Lopez, S., Obregon, M.J., Martinez-de-Mena, R., Bernal, J., Guadano-Ferraz, A. and Morte, B., 2016. Effect of Triiodothyroacetic Acid Treatment in Mct8 Deficiency: A Word of Caution, Thyroid. 26, 618-26.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
The authors show expression of the thyroid hormone transporter MCT8 in the human placenta. The MCT8-inhibiting compound sylchristine reduces the transfer of T4 from the maternal to the fetal side and accordingly reduces T4 degradation by DIO3. Since maternal thyroid hormones are relevant for neurodevelopment before the onset of fetal thyroid gland function, disruption of MCT8 transport, as occurs in the Allan-Herdon-Dudley syndrome, may contribute to the neurodevelopment failure present in these patients. The results of the transfer experiments are clear and support the authors' conclusions.
Minor comments:
- Line 71: please provide …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
The authors show expression of the thyroid hormone transporter MCT8 in the human placenta. The MCT8-inhibiting compound sylchristine reduces the transfer of T4 from the maternal to the fetal side and accordingly reduces T4 degradation by DIO3. Since maternal thyroid hormones are relevant for neurodevelopment before the onset of fetal thyroid gland function, disruption of MCT8 transport, as occurs in the Allan-Herdon-Dudley syndrome, may contribute to the neurodevelopment failure present in these patients. The results of the transfer experiments are clear and support the authors' conclusions.
Minor comments:
- Line 71: please provide some references on the immaturity of the blood.brain barrier before 18 months. The endothelial cells may have tight junctions when the vessels sprout in the CNS. "Maturity" implies the full complement of the neurovascular unit, i.e., pericytes and astrocytes. So, please clarify this point, even if it does not contradict the experimental results showing the role of placental MCT8.
- It is unclear if the partial effect of sylchristine on T4 transport means that MCT8 contribution is also partial and other transporters contribute.
- Lines 113-114. I missed the controls measuring TRIAC transport without sylchristine, or do the authors have strong reasons to assume that sylchristine does not affect TRIAC transport? If so, it should be stated.
Significance
This study confirms the presence of MCT8 in the human placenta and adds additional data to demonstrate its functionality with experiments using placental perfusion.
-
