Thyroid Hormone Metabolism in Action. We have some great abstracts that'll be presented today, spanning basic to clinical thyroid hormone action, and look forward to your participation. So remember, we'll have live question and answers. So as the presentations are going on, please put your questions into the Q&A session, and we'll have a few minutes after each presentation to go over the questions and share those. So look forward to that interaction. And we'll start with the first abstract, is Dr. Young-Wook Cho, from the Laboratory of Douglas Forrest at National Institutes of Health, NIDDK, in Bethesda, Maryland. The title of his abstract is, Transcriptional and Genomic Regulation of Pituitary Function by Thyroid Hormone Receptor Beta. I'd like to thank you, organizers, for giving me the opportunity to present my work. I don't have any disclosure. T-albeta has two isoforms, Beta-1 and Beta-2. Both isoforms are expressed in anterior pituitary. The anterior pituitary gland expresses following hormones, including thyroid stimulator hormone, which stimulates the thyroid gland to synthesize thyroid hormones. Growth hormone regulates growth, metabolism, body composition. T-albeta knockout mouse shows endocrine and sensory phenotype. In humans, T-albeta mutations are associated with thyroid hormone-resistant syndrome. To find out the T-albeta target gene, we use either wild-type or knockout mouse. We put the mouse in MMI drinking water for four weeks to make hypothyroid condition. I think T3 in last one week is enough to make a hypothyroid condition. After treatment, we check the pituitary hormone. Growth hormones are induced by T3 in wild-type mouse, not in knockout mouse. Not like growth hormone, TSH levels are elevated in hypothyroid condition and reduced in hypothyroid condition. We construct four or five RNA-seq libraries from each group. The red dot represents fourfold change in both T3-induced and T3-suppressed genes. The number of red dots in knockout mouse significantly reduced. That suggests T-albeta regulates most T3-response genes in pituitary. The question is, how does T-albeta regulate the pituitary gene at the transcriptional level? We generate a THRB-HAB mouse. This mouse produces T-albeta-HAB fusion protein, which are biotinylated by BLA. Now we can do chromatin affinity pull-down to define genome-wide T-albeta binding site in pituitary. Genome-wide T-albeta binding site in pituitary. C-terminal HAB tag does not affect T3-sensitive transactivation in reporter-safe system. Not like T-albeta knockout mouse, T-albeta-HAB mouse shows normal circulation level of thyroid hormone and TSH. This western block shows T-albeta-HAB protein stable in pituitary gland. T-albeta-HAB mouse to define the genome-wide T-albeta binding site in the pituitary. And we use a knockout wild-type mouse for the acetylation and methylation status. ATAC-seq, we are using the Sonon GFP-beta2 cream mouse. We divide the T-albeta-HAB peaks into three groups, A, B, and C. A group, de novo T-albeta peaks shows after 3-terminal T3. Constant binding peaks are in the B groups. We check the active enhancer mouse in the hypothyroid condition. You can clearly see in B group, their active enhancer mouse are increased, acetylation and monomethylation, and slightly increased in the A groups. And all of these groups are in the distal region, not the transcription star site. We are focusing on the A and B groups. We are further defined on the regulation site by the ATAC-seq. ATAC-seq, we can define the T3-induced chromatin opening site. So we overlap the two peaks, ATAC-seq and HAB binding site. We can see the 1725 peaks from the B groups. We can call the open chromatin constant beta2, T-albeta. And 456 from the A group, this is called open chromatin de novo peaks. Both groups contain de novo open chromatin region and a strong T-albeta binding site. Both have enhancer activation marks under the hypothyroid condition. So which group of the peaks are important for the target gene expression? We collect the associated gene from the both group of peaks. We have 1417 genes from associated with the open chromatin constant peaks, 231 genes from the open chromatin de novo peaks. This box plot shows red color from the 1417 genes increased by the hypothyroid condition. However, green does not. That shows T3-induced open chromatin site with T-albeta HAB peaks are critical for the gene regulation. This is an example. Growth hormones are well-known T3-induced gene. In the wild type, we can clearly see two-fold increase. In the knockout, however, it's not increased. That means growth hormone regulate by the T3 and T-albeta dependent manner. In growth hormone gene locus, we can find the two open chromatin constant peaks. One is close to the transcription start site, and the other one is a little far away, about 13 kb upstream region. Both contains enhancer marks, and both are also overlapped with the open chromatin region. Therefore, T-albeta binding sites are putative T3-sensitive regulated regions. We propose this model based on our current data. T3 is sitting on the putative site, putative enhancer site. After receiving the thyroid hormone, this site changes to the active enhancer region. It can increase of the monomethylation and acetylation. It also causes of the chromatin opening. Eventually, this site may regulate the T-albeta target genes. However, we don't know exactly mechanism how this site choose. It may require additional factors. I'd like to summarize my work. T-albeta regulates most T3-response genes, either T3-induced or T3-suppressed gene in the pituitary gland. T-albeta HIV model is a powerful tool to find the T-albeta binding site, even if in the tiny tissue, such as pituitary. T-albeta binding site disturb from the promoter, define the pituitary enhancer region. These pituitary enhancer regions are changed after receiving the T3. It will increase of the open chromatin and also increase of the active chromatin marker. Eventually, it will increase of the T-albeta target gene expression. I'd like to thank you all my collaborators. Thank you for your attention, and I'm happy to receive questions. Well, thank you, Dr. Cho. That was an outstanding presentation. We'll start. Yunbo, she has a question. I'll pass on. Excellent work. In the open chromatin regions, did you measure total histone levels to see if histones were lost? If so, how many nucleosomes were lost? We didn't measure the nucleosome lost. We're just measuring the distance between the histones, not the exact measurement of the base pair. We only see the peaks from the ATAC-seq. So probably, I will look at how much of the histones are lost in that region. Sure, and I'll ask a question as others are coming in. Obviously, your search was based on thyroid receptor binding to DNA. Of course, there's some T3-induced activity that may be looser, indirect binding to DNA. As you interpret your studies, are there ways you might have seen non-direct DNA binding actions of TR? Yeah, actually, this is a good question. Someone suggests TSH-β are regulated by a kind of adhesion of the TR-β binding site. But unless I look at these regions, I try hard to find out the TR-β binding site in the TR-β close to region. But I'm not assuming that. We are probably looking at long-distance regions, too. So in our current model, TR-β HIV mouse model doesn't bring down the conclusion of this kind of attachment of the binding site. Thanks, and then Mitch Lazar has a question. Great talk. Can you comment on the motifs bound at binding sites near negative genes versus those that are positively regulated? So was there a difference in the motif of the binding sites for positive genes and genes that are down-regulated by T3? Yeah, that's kind of our interesting goal. Actually, I tried to look at what are the different transcription factors that are interacting with these down-regulated genes or unregulated genes. Basically, we are currently looking at some candidates. But still, we don't have any other clue which factors are involved with these two different regulations. And this is both down-regulated or unregulated genes. If you are looking at the close genes, both contain TR binding site in terms of our binding site. So we can clearly see the TR motif either way in negative regions or positive regions. But I'm not saying both are active enough or not. Actually, we can clearly see the positive regions correlated with a kind of gene regulation. But negative binding site or negative genes, I'm not sure how these genes are regulated still. But the motif is not different, the binding motif. No distinct pattern. No, actually, yeah. I tried to find out that way. Well, we saw which transcription factor motifs are different. There's nothing different, yeah. All right. Well, thanks so much. There's one more question, if you could address that, Dr. Cho, in the chat. And the Q&A may be answered too. We won't have time. We need to move on to the next presentation. But thanks so much. Thank you very much. All right. So we'll move on to the next presentation, abstract number two. This is Noelle Gillis, who works in the laboratory with Fran Carr at the University of Vermont in Burlington, Vermont. The title of her abstract is T3-Dependent Transcriptional Programming by Thyroid Receptor Beta in Thyroid Cells Requires Switch-Sniffed Chromatin Remodelers. Thank you for the introduction. And I'd also like to thank the Endocrine Society for inviting me to participate in this session. Today, I'm going to tell you about my project investigating T3-Dependent Transcriptional Programming by TR Beta in Thyroid Cells. I have no disclosures. So thyroid hormone receptors play key roles in maintaining cell identity and regulating metabolism and homeostasis in tissues throughout our body. The bimodal switch model, where TR is constitutively bound to chromatin and ligand binding allows for dissociation of co-repressors and recruitment of co-activators, has been the predominant model that we use to describe transcriptional regulation by TRs for many years. However, it's recently been appreciated that this model is highly oversimplified. Recent studies have proposed more nuanced models of TR interaction with chromatin. However, low endogenous expression of TRs and suboptimal commercially available antibodies have presented technical obstacles for those of us who wish to study genome-wide binding patterns of TRs. Although some groups have been able to overcome this technical hurdle, there are currently limited data sets available for TRs, especially when compared to other nuclear receptors. Therefore, we aim to profile genome-wide TR beta binding, accessibility, and changes in co-factor interactions in the presence and absence of T3 in normal human thyroid cells. We perform cut and run to generate genome-wide binding data for TR beta in normal thyroid cells in the presence and absence of 10 nanomolar T3. Cut and run offers several technical advantages over ChIP-seq, particularly lower starting inputs and a higher signal-to-noise ratio. With this method, we were able to identify just over 8,000 TR beta binding sites that we subcategorized with a direct overlap analysis, as shown in this Venn diagram. Unliganded binding sites are those that are only detected in the absence of T3. Ligand-independent binding sites are detected both in the presence and absence of T3. And liganded TR beta binding sites are those that are detected primarily when T3 is present. The cut and run signal at each of these three types of binding sites is shown in this heat map. We observe a loss of signal in the unliganded binding sites with T3 treatment, a relatively steady signal in the presence and absence of T3 at these ligand-independent binding sites, and a significant gain in signal when we treat with T3 at these liganded sites. Motif frequency analysis revealed that a significant proportion of all TR beta binding sites contain TREs. However, the liganded independent binding sites as shown here in purple have a much higher frequency of full-length DR4 TREs compared to other types of binding sites. Interestingly, when we compared the distance to the nearest transcriptional start site of our three types of binding sites, this revealed that ligand-independent binding sites tend to be closer to transcriptional start sites than others. In order to map changes in chromatin accessibility in response to T3 in our model, we performed ATAC-seq at both 6 and 24 hours after treatment with T3. We observed striking increase in chromatin accessibility after T3 treatment for six hours, as shown in the light pink dots in the scatterplot. We observed an even more robust increase in chromatin accessibility, as shown in the dark pink dots in the scatterplot. Some sites were repressed by T3, as shown in the blue dots. When we look specifically at chromatin accessibility near TR-beta binding sites, some clear patterns emerge. Unligand-binding sites have very little T3-induced chromatin remodeling around them, while ligand-binding sites have T3-induced chromatin accessibility at 6 hours and 24 hours. Ligand-independent binding sites have even more striking increases in chromatin accessibility, both at 6 and 24 hours. When we quantify the distance between remodeled regions near those TR-beta binding sites, it becomes clear that remodeling around ligand-binding sites is much more focused, while remodeling around the ligand-independent sites stretches across broader regions. So to summarize what I've shared with you thus far, we have unligand-binding sites that have low TRE frequency and that tend to be found in distal regulatory regions with minimal chromatin remodeling activity. We have high DR4 TRE frequency in ligand-independent sites. They occur primarily in promoters, and they have broad remodeling activity. These ligand-independent sites have high TRE half-site frequency, tend to be split between promoters and distal regulatory regions, and have focused remodeling activity. To characterize the cofactor interactions around these different types of binding sites, we used a proximity labeling assay followed by mass spectrometry. Briefly, we created a TR-beta biotin ligase fusion construct by fusing a mini-turbo ID to the N-terminus of TR-beta and expressing it in our normal thyroid cells. After a brief incubation with biotin and T3, we collected protein lysates, isolated the biotinylated interacting proteins with streptavidin bead purification, and then performed mass spectrometry. In the volcano plot on the right, I'm showing all of the interacting proteins we identified in our screen. The dots that are colored in blue represent proteins that were differentially enriched in the presence of T3 and in the absence of T3. In the bubble chart below are the molecular function classifications of the interacting partners in each of these three categories. The size of the bubble is a function of the enrichment score, and the color is a function of the p-value. These functional classifications are consistent with what you might expect for interaction with a transcription factor. Interacting proteins that were lost with T3 treatment, for example, were classified as having transcription repressor activity, demethylase activity, and translation factor activity. Interacting proteins that remained stable included nucleosome remodeling, ATPase activity, and histone binding. Interacting proteins that were gained included proteins classified as transcription coactivators and RNA polymerase binding proteins. Upon closer examination of chromatin remodeling complexes that showed up in our screen, we noticed an interesting trend where tR-beta seemed to be preferentially recruiting different subspecies of switch SNF complexes in the presence and absence of T3. In this schematic, I'm showing two different species of switch SNF complex, BAF and pBAF, that are potentially being recruited by tR-beta. The proteins in gray are the core subunits that are present in both species, and the proteins in blue are BAF-specific subunits. Proteins in pink are pBAF-specific subunits. Proteins labeled with bold text were those that were actually identified in our screen, and those without the bold text are part of the complex that were not identified in our screen. The bar chart on the right, I'm showing you signal intensity from our mass spec data for each of the individual subunits that were differentially enriched. You'll notice that BAF-250, or arid 1A, in the blue bars was decreased with T3 treatment, while BAF-200, or arid 2, and BAF-180, or pBRM1, were increased in the presence of T3. BAF-57 was also slightly increased in the presence of T3. BRG1 was not differentially enriched. We then decided to perform cut and run for arid 1A to mark BAF complexes, pBRM to mark pBAF complexes, and of course, BRG1, which is present in both. We then examined the binding of each of these factors to tR-beta binding sites. In this box plot, we see that BRG1 cut and run tag density is stable at unliganded binding sites, increased with T3 at ligand-independent sites, and decreased with T3 at liganded sites. Arid 1A tag density is increased with T3 at unliganded sites, stable at ligand-independent sites, or increased with T3 at ligand-independent sites, and stabled at ligand-independent sites. BAF-180 is stable at unliganded sites. It's slightly increased with T3 at ligand-independent sites, and also stable at ligand-independent sites. Since unliganded tR-beta binding sites did not appear to have a significant remodeling activity around them, we decided to further divide our liganded and ligand-independent peaks based on whether they are near a promoter or in a distal regulatory region, and then examine those further. So arid 1A is recruited in a ligand-dependent manner to both tR-beta bound promoters and distal binding sites. PBRM seems to be recruited preferentially to tR-beta bound promoters and not to distal binding sites. And finally, BIRG1 seems to be recruited to both types of binding sites in the presence and absence of T3. Interestingly, when we take a look at the remodeling happening around these sites, we see that there's a much greater degree of change in chromatin accessibility at 6 and 24 hours near tR-beta bound promoters than at distal binding sites. So in summary, we've added some complexity to the model for tR interacting with chromatin. At these unliganded sites that have low TRE frequency and tend to be found in distal regulatory regions, there's minimal remodeling activity, and they recruit BIRG1 potentially to prime binding sites for receptor binding. These ligand-independent sites have high tR4 DRE frequency. They tend to be found near promoters and have broad remodeling activity. They also seem to recruit both BAF and PBAF complexes to these types of sites. The ligand-independent tR-beta binding sites have high TRE half-site frequency, tend to be split between promoters and distal regulatory regions. They tend to have more focused remodeling activity, and they recruit BAF complexes primarily. With that, I have some acknowledgments. I'd like to thank my lab and, of course, my mentor, Dr. Fran Carr, my co-mentor, Dr. Seth Fritz, and members of his lab for support with bioinformatics and genomics analysis, and, of course, our genomics core at the University of Vermont for sequencing all my samples and our funding. And I can take any questions. And here are my key references. Well, thank you so much. It looks like we've already got some questions. And I'll actually start it by asking, and I'm sure you've said this, but what's your cutoff with respect to what you're calling distal and those that are close to the promoter, approximately, in terms of spacing? Yeah, so for spacing, for the analysis that was sort of more towards the end of the presentation, I defined a promoter site as within a kilobase of the transcriptional start site. And anything that was greater than 1 kb away, I considered distal. Great. Thanks. So Brian Hogan says, very nice work. I may have missed it, but where were your normal thyrocytes? So use the thyroid cell line. I actually had a similar question. So he's asking, did you compare normal? Were these primary human or other cells or a transformed cell line? Did you also examine this in thyroid cancer cells? So maybe talk about that. Oh, yeah. So these experiments were done in the NTHY or a cell line. So that's an immortalized thyroid epithelial cell line. So we did not do that in a primary thyrocyte, although that would have been very cool. And we haven't done this yet in thyroid cancer cells. Expression of TR beta is even lower in thyroid cancer cells normally than what you get in a thyrocyte. So it would be difficult to do the cut and run, but it's certainly something we'd love to give a try. All right. Yehuda Shabtai, great work. In the TR binding proteins, have you found RXR, N-core, or P300 complexes? So the typical co-repressors and co-activators. Yes, we did identify all of those. And it was good to see as a sanity check that those were there. But we decided to push forward with something that might be a little more out of the mainstream. But I will ask and add to that question, then was there a difference in N-core, the typical co-activators and co-repressors, with the more distal versus proximal sites relative to the enrichment for those? I did not do cut and run on those sites. So I didn't follow up on the mass spectrometry with genomics on those, since they had been done many times over by other labs. But we did see a decrease in enrichment of the N-core complex with T3 treatment, as you might expect. And we saw an increase in the P300 complex, as you might expect. So all of those were checking that our experiment had worked, as you might expect. OK, and another one from Yungbo Shi, excellent talk. I wonder if you've analyzed BRG binding to individual genes, especially those with strong TREs. I was a little surprised to see very little effect by T3 on overall binding by remodeling proteins to target genes, considering the big effect on remodeling of T3. So any specific genes, individual genes you've looked at relative to the BRG binding? I haven't paid close attention to what's going on nearby individual genes, I'll admit. And I was also surprised by that result at first, that there wasn't a whole lot of change in enrichment near TR binding sites of BRG1 specifically, that rather the change in enrichment seemed to be with the other members of the complexes. But there were actually some paper, there was more than one paper coming from Trevor Archer's lab that suggested that for GR, BRG1 can act as a pioneer factor to prime a binding site for GR. And I suspect that maybe there's a similar mechanism going on there, where somehow BRG1 is able to get to those binding sites ahead of TR and prime the binding site. And I'll ask, in this cell line, we heard the previous talk related to pituitary, and we know growth hormones T3 responsive. Are there index genes in the thyroid doesn't have a lot of well-described T3 responsive genes? I mean, you got a lot in your assay, but are there any markers? I guess DI Dase 1 is upregulated. Any genes you looked at in your cell line that you knew is T3 responsive as kind of a marker of the gene expression? Yeah, we're beginning to pay closer attention now to what actually came in our RNA-seq in terms of pathways. Because you're right, there haven't been a whole lot of profiling of T3 response in normal thyroid cells. So we don't have a lot to go on in terms of prior literature about that. But we do see some of the classic thyroid differentiation type genes like deodonase 1, they all go up. And a lot of the pathways that come out are related to metabolism and reactive oxygen species and things like that. And I think Dr. Haugen will collaborate with you on thyroid cancer. So that'll be, especially these are growth pathways. I think that could be very interesting. All right, thank you so much. I think we'll move on. All right, the next abstract number three, the effects of energy restriction on thyroid hormone dynamics. This is a clinical project. Jocelyn Tessa Tonloo and her colleagues at the National Institutes of Environmental Health Sciences in NIH and in the research program of Janet Hall. And we look forward to this presentation. Thank you. Thank you all for joining me today. And to the Endocrine Society for inviting me to present. Authors have no disclosures. Before diving into our research project, I wanted to take some time to provide you with some background on the effects of energy restriction on thyroid hormone dynamics. Intermittent energy restriction or IER has gained popularity for weight loss and other potential health benefits. Definitions of IER can be quite fluid, but include periodic fasting, alternate day fasting, or repeated dieting for several days among others. Two studies employing total caloric restriction for two and a half to eight days have shown a reduction in TSH and total T3, an increase in reverse T3, but have shown no effects on total T4 and free T4. However, there's still no information on the effects of more moderate caloric restriction as it's more commonly used by women on thyroid function. And no previous studies have used liquid chromatography mass spectrometry, which has the advantage of increased specificity for measuring thyroid hormones. So depicted here is the hypothalamic-pituitary-thyroid or HPT axis, where you can see on the left-hand side that the thyroid-releasing hormone travels from the hypothalamus into the anterior pituitary where it stimulates production of TSH, the thyroid-stimulating hormone. TSH then travels through the bloodstream to the thyroid gland where it stimulates production of T4 and T3. T4 can be converted to T3, and both of these hormones travel through the bloodstream to distant organs to regulate metabolic rate. The conversion of T4 to T3 will be later covered, but it's important to remember that T4 and T3 have a central control on thyroid hormones through a negative feedback on the hypothalamus and pituitary. On the right-hand side is the thyroid gland depicted, where you see that the thyroid gland secretes large amounts of T4 and smaller amounts of the more biologically active T3. Both of these hormones travel through the bloodstream with the help of the thyroid-binding globulin. And as I mentioned before, T4 can be converted to T3 using the D1 and D2 isoform of deiodinases. T3 and T4 can also be converted to the biologically inactive or first T3 using the D3 isoform of deiodinases. Cortisol and growth hormone both have a peripheral control on thyroid hormones by inhibiting deiodinase activity. So our hypothesis in this study was that thyroid hormone dynamics will adapt to even modest reductions in energy availability to maintain homeostasis through energy conservation. So in this slide, I'm going to take you through the methods used in our study. Our study compared a mutual energy availability intervention of 45 kilocalories per kilogram lean body mass per day meant to produce no weight gain or weight loss to a deficient energy availability of 20 kilocalories per kilogram lean body mass per day. Both of these interventions occurred over a five-day period. As you can see through our study schema here, the blue lines represent the pre-ovulatory surges in LH across different menstrual cycles, and the red boxes represent menses. You can also see that our study spanned across four menstrual cycles, beginning with the recruitment cycle and intervention occurring in cycle two and three. Information collected in cycle one was used to predict menses in the following cycles, along with when diet intervention should begin. On the fifth day of the diet intervention, participants reported to the clinical research unit of the NIEHS where they had their blood frequently sampled for thyroid hormones, cortisol, growth hormones, and CBG. Liquid chromatography mass spectrometry was used for thyroid hormone assays except TSH and CBG. So now looking at some of the results from our study, we have here a demographics table with numbers given as mean plus or minus standard deviation. You see that we had fairly young women with a good racial and ethnic distribution take part of our study. But more importantly down here in the physiologic characteristics section, you see that our participants experienced a 1.5 kilogram decrease in weight loss due to diet intervention, along with a decrease in body mass index. Lean body mass and percent body fat were unaffected. So now looking at some of the results from our frequent sampling, linear mixed models were used for analysis of frequent sampling data, while parity tests were used for analysis that only included times to zero and eight. Following data slides will have the similar format where each study began at 8 a.m. following an overnight fast and you can see down here, the x-axis will always represent the time in hours and the y-axis will always represent the hormone values. The first two dashed lines will represent the administration of breakfast and lunch meant to either satisfy NEA or DEA. And the third line will represent the administration of a snack, an NEA snack that was used at both NEA and DEA. These blue lines will always represent NEA, while these red lines will always represent DEA. As they're expected changes in thyroid hormones across the day, these P values will always represent the differences between NEA and DEA intervention. In instances where the hormone was only sampled at the beginning and end of the frequent sampling day, bar graphs were used, but although bar graphs were used, the color schemes remained the same and the P value also signifies the difference in intervention. So now looking at TSH, you see that TSH decreases across the day, but more importantly, TSH also decreases due to deficient energy availability. So now looking at TBG here on the right-hand side, we actually observe no effect of diet intervention on TBG, nor did we observe an effect of time of day. So now looking at T4, we have total T4 here on the left-hand side, where you can see that total T4 decreases across the day, but more importantly, total T4 increases due to deficient energy availability. Looking at free T4 here on the right, we also observed an increase in free T4 due to diet intervention, but this was only at fasting measures. There was no effect on free T4 due to time of day at either intervention. So now looking at T3, we have total T3 here on the left, where you can see total T3 decreases across the day and also decreases due to diet intervention. Looking at free T3 here on the right, you see that free T3 decreases due to diet intervention, but only at fasting measures. You also see that there was a decrease in free T3 across the day, only at the NEA intervention. So now looking at reverse T3, you see that reverse T3 increases due to diet intervention here. Growth hormone and cortisol were both sampled. With growth hormone here on the left, you can see that growth hormone decreases across the day, but there was no effect of diet intervention on growth hormone. Similar results were observed with cortisol, where you can see that cortisol also decreases across the day, but has no effect of diet intervention. So what conclusions can be drawn from this study? Using sensitive and specific liquid chromatography mass spectrometry for measuring thyroid hormones, we have shown that in response to five days of modest energy restriction, there is a decrease in TSH, along with a decrease in total T3 and free T3. There is also an increase in reverse T3, along with an increase in total T4 and free T4. As also depicted here on the right-hand side, you see that less active thyroid hormones increase, as seen with T4 and reverse T3 here, at the expense of more biologically active hormones, as seen with T3. All of these changes occurring in the absence of changes in TBG, growth hormone, or cortisol. So in conclusion, thyroid hormone dynamics adapt to even modest reductions in energy availability in women to maintain energy homeostasis using both central and peripheral mechanisms predicted to reduce metabolic rate. Here are key references used in this presentation. I would now like to acknowledge my co-authors on this abstract, participants, nursing, and laboratory staff at the NIEHS-CRU. Thank you so much for a very clear presentation. I look forward to questions. I'll get started, and I believe, okay, we got some questions. Okay, we got some questions. I'll start with going through those. So Joseph Curley, transthyretin, an important thyroid hormone distributor protein is known to be affected by energy restriction. Any data on this with respect to thyroid hormone concentration changes different affinities of T4 versus T3 to transthyretin? So you looked at TBG, but this is now the smaller fraction thyroid-binding transthyretin. Did you measure that or look at that? So we did not measure that or look at that, but so I'm not too sure the extent. So with TBG, we did measure TBG, but we also anticipated that we wouldn't have significant changes in TBG because TBG is a protein that changes with prolonged undernutrition. So with this also being a protein, I wonder if we would have seen significant changes there. Yeah, and I believe this is a shorter half, more change with nutrition, the TBPA or transthyretin. So that may be one you want to look at. I think it does change more often than TBG. Mary Mariosch, why is free T3 unchanged at eight hours? So I guess he's looking at the data. There was some variation in the time. So free T3 was unchanged in DEA. So yeah, free T3 was unchanged in DEA across eight hours, but it was changed in NEA. So we believe that maybe this was due to the diet intervention as well. So we haven't fully looked at why we had that change in free T3 in DEA, why we had the change we observed in free T3 in NEA versus DEA, but we did see that change in NEA. And Francesco, Chelsea, great talk. Were you able to measure energy expenditure and correlate with T3? So, you know, metabolic rate or energy expenditure? Yes, we did measure resting metabolic rate, but we have not done correlations as far as T3, like correlations between the resting metabolic rate and T3 yet, but we will be working on those analyses as well. Great, did you look at prolactin levels from Martha Hoyvas, yes? Oh, we have not measured prolactin levels, but definitely would have been helpful as well. All right, and then Brian Haugen, great talk. TSH and T3 reduction with increased reverse T3 makes sense. What do you think the mechanism of increased total T4 and free T4 is? So our hypothesis is that seeing that we had no change in growth hormone or cortisol, we believe that there are other factors such as possibly leptin or insulin or TSH that could also be affecting deiodinase activity. So T4 is converted to T3 through deiodinase reduction one and two. So we believe that there is something, a mechanism other than growth hormone and cortisol that could be affecting that change. We're still looking at why we observed that increase in T4 and free T4, but we do have that hypothesis. All right, and then Joseph Curley also asks, any data on free fatty acids? So the severe nonparietal illness has elevated free fatty acids, this kind of mini NTI, did you see high free fatty acids or what could that be? So we don't have data on the free fatty acids, but yes, that also again would have been good to look at. We, I did read something from, I did read a paper from Dr. Bianco that did look at the administration of bile acids on D2 or deiodinase activity. So that was really helpful as far as trying to understand coming up with possible hypotheses for the results we observed in our study. Great, thanks so much. And I believe this is the smallest change in food intake or provocation for these changes in T3 that you've shown. So very nice study. So we'll move on to the next and final talk. Thank you, Jocelyn. All right, and the final talk, abstract four, is from University of Pennsylvania, Dr. Paige Mieslick and her colleagues in the research program of ANCAPOLA at University of Pennsylvania. The title of her talk is Physiologic Effects of Levothyroxine and Lyothyronine in Older Individuals with Persistent Subclinical Hypothyroidism, a Randomized Double Blind Crossover Study. Thank you. Thank you very much for that introduction. I appreciate the opportunity to share our data today. I have nothing to disclose. Our group is interested in subclinical hypothyroidism. So I'd like to orient ourselves to the clinical scenario in which a patient comes into clinic and we check their TSH. It could be low, normal, or high. If it's high, we can then check the thyroid hormone levels. And if that's low, then they have overt hypothyroidism. But in the scenario where the thyroid hormone level is normal, these patients are considered subclinically hypothyroid. It's a bit of a misnomer, however, because they may not have any clinical symptoms and it's purely a biochemical diagnosis. Our group is interested in older individuals that are 70 or over with thyroid peroxidase antibody negative, persistent subclinical hypothyroidism, with a TSH of 4.5 to 19.9, and a normal free T4. Our initial goal with this study was to examine the pituitary thyroid axis with dynamic TRH stimulation testing before and after LT4, LT3 therapy. And this data will be presented in poster format at this conference. Our additional goal was to examine whether there are differential responses in end organ function after LT4 and LT3 therapy. If we look at the randomized controlled trials of LT4 therapy in older adults with subclinical hypothyroidism, such as the TRUST trial, what we see is that they found no difference between LT4 and placebo for thyroid symptoms, quality of life, fatigue, muscle strength, bone density, bone turnover markers, cardiac function, blood pressure, or BMI. We can also consider more mechanistic studies that also evaluated physiologic endpoints on LT4 and LT3. This is a study by Dr. Chelley's group of 14 men and women with previous thyroidectomy and overt hypothyroidism. And when looking at physiologic endpoints when their TSH had reached the lower half the normal reference range of 0.5 to 1.5, they found no difference between LT4 and LT3 in a range of parameters such as quality of life, metabolic and cardiac parameters. But what they did see was lower weight after LT3 and lower cholesterol, LDL cholesterol and ApoB after LT3 in comparison to LT4. We conducted a randomized double-blind crossover study in patients over 70 years of age with TPO antibody negative status and persistence of clinical hypothyroidism. Once persistence of clinical hypothyroidism was established they came into the clinic for a first visit with TRH stimulation and a range of physiologic outcomes and assessments that I'll speak about in a minute. They then were randomized to either LT4 or LT3. They had their TSH levels checked every four weeks and their dose was titrated until they reached a goal TSH of 0.5 to 1.5. At that point, they came in for a second visit with all the same assessments. Following that visit too, they had a four week washout so that we didn't overlap the LT4 and LT3 therapy. They then proceeded to go on to the alternate thyroid therapy. So either LT4, LT3. And again, had their TSH levels checked every four weeks and titrated until they reached that lower half of the normal reference range. And they then came in for a final visit three. The thyroid hormone replacement was given via customized dose tailored capsules that were made possible through the investigational pharmacy. There was three times a day dosing and the titration protocol was a standardized protocol that was the same for each participant. The physiologic outcomes that we measured were weight and body composition by DEXA, resting energy expenditure, bone mineral content and bone mineral density, lipids, vital signs, SF36 survey for quality of life, thyroid symptoms, and the digit symbol substitution test or DSST for cognition. In terms of statistical analysis, we use paired T tests for group comparisons and P less than 0.05 was considered the threshold for statistical significance. We screened 76 participants. 59 did not meet study eligibility. And the most common reasons were that they had a normal TSH on repeat testing when we went to confirm persistence or they had TPO antibody positivity or both. Three participants declined. In the end, we had 14 enroll. One withdrew, however, due to non-adherence with the TID dosing. And we included 13 in the final analysis. Here are the baseline characteristics of our cohort. The mean age was 77 with a range of 71 to 84. There were four females and nine males. All were Caucasian. The screening TSH mean was six with a range of 4.5 to 10.5 with a normal free T4. With a normal free T4. When the patients came in for visit one, that was considered their baseline. And the baseline TSH mean for the group was 4.84 with a range of 2.92 to 6.67. So we could see that the mean dropped a little bit from the initial screening and pre-screening TSH levels. And at that baseline visit one, they had normal free T4 and total T3 levels. When we looked at the dose and duration of thyroid hormone throughout the study, we found that six participants were randomized to the LT4 treatment first in the sequence. We found that the mean LT4 dose per day was 105 micrograms with a range of 60 to 195. This was a mean dose of 1.4 micrograms per kilogram per day. And the mean dose of LT3 was 34 micrograms with a range of 16 to 47, equating to a mean dose of 0.46 micrograms per kilogram per day. They stayed on thyroid hormone for quite some time. They had a mean of 431 days and that excludes the one month for washout. They had a mean of 200 days on LT4 and 231 days on LT3. When we looked at body composition via DEXA and also via weight by scale, what we found was that weight, total mass and fat mass were lower after each therapy. And we also found that weight, total mass and fat mass were lower after LT3 than after LT4. There were no differences in lean mass or body fat. We looked at resting energy expenditure and found that there was no difference in resting energy expenditure after each therapy or in comparing the differences between LT4 and LT3. On DEXA, we found that there was no difference in bone mineral content or bone mineral density after each therapy and no difference after LT4 versus after LT3. We did see differences, however, in the lipid profile with total and LDL cholesterol being lower after LT3 than after LT4. And triglycerides were lower after LT3 compared to baseline, but no difference when comparing LT3 versus LT4. There were no differences in HDL. There were no differences in vital signs including systolic and diastolic blood pressure, heart rate and rectal temperature, as well as no differences in the SF-36 questionnaire, thyroid symptom scores or DSST. Both therapies were well tolerated and no adverse events. In conclusion, close to full replacement doses of LT4 and LT3 were required to lower TSH to 0.5 to 1.5. And for most outcomes, while there was no difference between LT4 and LT3 use, however, the number of participants was small. Participants lost weight and fat mass after each treatment with a greater reduction after LT3. We also found that triglycerides were lower after LT3 while total and LDL cholesterol reduction did in fact differ between LT3 and LT4. And both treatments were well tolerated. In terms of implications for the field, I found it interesting that the suppression of the majority of endogenous thyroid function is required to achieve TSH levels in that lower half of the normal range in these older patients with TPO antibody negative persistent mild subclinical hypothyroidism. Our findings do support differences in physiologic responses to LT4 compared to LT3 at the doses we use and are consistent with previous reports. And finally, it's important to note that our study was designed to test mechanistic hypotheses, not to test therapeutic options for subclinical hypothyroidism. Thank you very much. Thank you, Dr. Meisler for a very clear presentation. And I'll start with the questions. I don't see any yet. I hope Dr. Celli is still there and we'll ask some, but as we contrast your results with his, obviously I think this is the oldest published group that's gotten T3 up to age 80, and I salute you for doing that. Can you contrast? So the results seem generally similar to the younger thyroidectomy patients, but how would you contrast it with those results from Dr. Celli's earlier study you quoted? Thank you, Dr. Brent. Yes, I mean, one of the nice things that we saw not presented here today was that they responded similarly to the TRH simulation testing. So there, and we found that that pharmaco equivalence in that dynamic effect, and we also saw the one to three ratio into T4 to T3. And the other piece, even though it's a small sample size, what I think also gave some support when we looked at the data was that the differences that we're seeing in Dr. Celli's physiologic testing was in the weight and the liver profile, I believe the LDL, and those are the areas that we saw as well. All right, so Carrie Mariosh asks, so you specifically looked at TPO negative patients. Did you measure anti-TG antibodies as some with Hashimoto's or TPO negative, but TG antibody positive? No, we did not. All right. Just the TPO excreting, and that if they were positive, they didn't come into the study. All right, and Julia Carlip asks, thank you, Dr. M, is there a hypothesis to explain why it took to use full replacement doses, suppress endogenous function completely to get to lower TSH? You know, it's a good question. I think that when you look at the TRH testing data, you can see that the TSH response to TRH is suppressed on LT3 or LT4, and maybe you really do have to suppress full endogenous function to get there. And, but, you know, I think it also comes back to thinking about whether the elevated TSH levels in these antibody negative patients is HPA axis, seven point abnormality, or issues with the TSH bioactivity versus issues at the level of the thyroid. Sure. Dr. Celli is listening. He asks, did you look at echocardiograms, especially with those older patients? Yes, we did. We did Doppler echocardiogram, as well as arterial tonometry, and that data will be, will take some time to be evaluated and come out later. Okay. And he congratulates you on the study. Maribel Aranda again, for how long the patients had subclinical hypothyroidism? So you define the group, but how long, do you know the, was the average duration of subclinical? I don't have the average duration, you know, but we did make sure they had the persistence, for at least four weeks in between. And I think from what I remember about the data, many of them, the initial pre-screening TSH that we looked at were, you know, were even earlier. All right. So I think- More than eight weeks, I mean. All right. Thank you so much, Dr. Maislik. There's a few more questions. If you can answer those in the Q and A, we're getting the signal that we're at the end and running over. So I want to thank all the participants for an outstanding session. Please stay on. There's more program to go. So please stay on and participate. Thank you so much. And thanks to the Endocrine Society.

thyroid hormone metabolism

transcriptional regulation

genomic regulation

pituitary function

thyroid hormone receptor beta

energy restriction

thyroid hormone dynamics

physiologic effects

levothyroxine

liothyronine