Abstract
Currently, thyroid hormone status is predominantly determined by the measurement of serum thyroid-stimulating hormone and free thyroxine. Although it is assumed that serum thyroid hormone (TH) concentrations within the reference range represent euthyroidism, it is unknown whether this reflects euthyroidism in all tissues (e.g. brain, muscle, bone and liver). To date, no serum marker has been established for clinical use that represents TH status within tissues accurately. However, several biomarkers have been investigated and innovative techniques have been used to unravel new biomarkers. This review provides an overview of proposed serum biomarkers that reflect tissue TH status in humans. Furthermore, we discuss the feasibility of these serum markers in clinical practice.
Introduction
Thyroid hormone (TH) is essential for development, growth and metabolism; a large variety of cells and organs are affected by TH and dependent on TH for proper function. The main TH target organs are the liver, bone, kidney, intestine and heart, and key functions of these organs are affected by a shortage or excess of TH. In addition, several systems such as the complement and coagulation system are also dependent on TH. Thyroid dysfunction (both overt and subclinical thyroid disease) is one of the most common endocrine disorders with a prevalence of approximately 3.8% in Europe (Garmendia Madariaga et al. 2014). Currently, measuring serum thyroid-stimulating hormone (TSH) and/or free thyroxine (fT4) is leading in establishing TH status. However, it can be questioned whether both TSH and/or fT4 entirely capture TH status throughout the whole body. Furthermore, many patients with hypothyroidism are bound to TH medication (levothyroxine (L-T4)) whose dosage is, depending on the cause, based on TSH or fT4 concentrations measured in blood. Still, a substantial part of those patients experiences discomfort while serum TSH or fT4 concentrations are within the reference interval.
Serum TSH, as part of the hypothalamus–pituitary–thyroid axis, regulates the production of thyroxine (T4) and triiodothyronine (T3) by the thyroid gland. In this review, TH refers to both fT4, free triiodothyronine (fT3), total T4 and total T3 and specified if needed. The prohormone T4 is the most abundant TH in the circulation and fT4 is taken up by the cell via specific TH transporters. Within the cell, T4 is metabolized by TH converting enzymes (deiodinases) toward T3. T3 is the active hormone which predominantly exerts its action via binding to the TH receptor (TR). In clinical practice, it is assumed that serum TH concentrations reflect TH status in the tissues. However, previous research in both humans and rats indicates that serum TH concentrations may not always correspond with tissue TH status (Escobar-Morreale et al. 1995, Dumitrescu & Refetoff 2013). This might be due to changes in TH transport across the cell, TH metabolizing enzymes or T3-TR binding.
To investigate true TH action in an organ, a tissue biopsy is needed to measure tissue TH status or establish TH regulated pathways. Since this is not feasible in clinical practice and if feasible, TH tissue measurements are not validated in all tissues, there is a need for accurate serum markers representing (specific) tissue TH status. These serum markers are useful for several purposes. First, the determination of serum markers of tissue TH status might improve treatment. Ideally, these markers should give a better representation of euthyroidism in tissues than the currently used serum TSH and/or fT4 concentrations. Subsequently, treatment can be targeted to improve, for example, cardiovascular outcomes, osteoporosis, or overall well-being. It might help to understand why patients treated with L-T4 still experience discomfort even though TSH and/or fT4 concentrations are within the reference interval. Furthermore, TH therapy such as TR agonists, for treating other conditions than thyroid disorders (e.g. hepatosteatosis), could be monitored more easily using serum markers of specific tissue TH status. Finally, these markers could be used to improve diagnostics. Some thyroid disorders cannot be differentiated based on TSH and/or fT4 concentrations (such as resistance to TR (RTH) and TSH-secreting adenoma), in that case, biomarkers reflecting tissue TH status might be discriminating.
Over the past decade, several metabolic biomarkers have been associated with TH action (e.g. cholesterol, osteocalcin and sex hormone-binding globulin (SHBG)). Currently, innovative techniques using omics are used to examine not only one marker but determine multiple markers in order to unravel a panel of markers representing (tissue) TH status. Metabolomic profiling is a powerful tool for measuring the activity of metabolic tissues (Newgard 2017). Thousands of small molecule metabolites can be measured to represent the current metabolic state of specific cells, tissues or whole organisms (Smith et al. 2006). Targeted and nontargeted metabolomics can be distinguished. Targeted metabolomics is used when the specific concentration of a known metabolite is of interest. Nontargeted metabolomics are best suited to identify new metabolites affecting specific pathways. Metabolomics has been used so far to establish markers in different types of diabetes, cardiovascular diseases, fatty liver, non-alcoholic steatohepatitis (NASH), and TH status (Newgard 2017, Friedrich et al. 2020). Along with metabolomics, proteomic profiling has been used as well, where instead of metabolites, proteins are determined. Interestingly, associations with more classical markers of TH action previously described are also found when using these techniques.
Pathways in human TH metabolism
Tissue TH availability is determined by an interplay between serum TH levels, TH transporters, and cellular TH metabolizing enzymes, which is shown in a simplified manner in Fig. 1 and explained below. Once produced by the thyroid gland, TH is released into the bloodstream and transported to the tissues mostly bound to thyroid-binding globulin. Then, TH actively enters the cell via TH transporters. Monocarboxylate transporters (MCT8, MCT10), organic anion-transporting polypeptide (OATP1c1) and solute carrier family (SLC17A4) transporters are thought to facilitate this transport most efficiently (Friesema et al. 2006, 2008, van der Deure et al. 2008, Teumer et al. 2018). MCT8 has a better affinity for the transport of T3, where MCT10 and OATP1c1 transport predominantly T4 into the cells. Transporters are also involved in the transport of T3 out of the cell. Not all transporters are expressed in all tissues. MCT8 is expressed in neurons, the pituitary, liver, kidney, and adipose tissue, while MCT10 is expressed in the kidney, liver, muscle, macrophages, and hypothalamus. OATP1c1 is expressed in brain capillaries and therefore involved in the transportation of T4 over the blood–brain barrier and SLC17A4 is predominantly expressed in small intestinal and colonic epithelial cells, pancreas, liver and kidney.
Pathways in human TH metabolism. A simplified representation. A full color version of this figure is available at https://doi.org/10.1530/JOE-21-0364.
Citation: Journal of Endocrinology 253, 2; 10.1530/JOE-21-0364
In the cell, T4 and T3 are subject to pathways that activate or inactivate them. Three major peripheral pathways are described and are important in determining cellular TH status: deiodination, sulfation and glucuronidation. Other pathways (deamination and decarboxylation of the alanine side chain of TH, ether-link cleavage) also play a role in the TH metabolism, but their contribution is minor.
It is known that deiodination contributes greatly to the conversion of T4 into the biologically active T3 through outer-ring deiodination and to the degradation of T4 and T3 to respectively reversed T3 (rT3) and the inactive TH metabolite 3,3’-diiodothyronine (T2) through inner ring deiodination. Three types of deiodinases (DIO1, DIO2 and DIO3) are known and have distinct functions in the (in)activation of TH (van der Spek et al. 2017). DIO1 is able to perform both inner- and outer-ring deiodination. Its main role is the degradation of inactivated TH; however, DIO1 also plays a role in the conversion of T4 into the biologically active T3, especially in the liver. DIO2 has predominantly an activation role and is only capable of outer-ring deiodination and prefers deiodinating T4 into T3. The function of DIO3 is to inactivate TH, since it is only capable of inner-ring deiodination, DIO3 catalyzes the conversion of T3 into T2 and T4 into rT3. Interestingly, not all types are present in the same tissue. DIO1 is located in the liver, kidney, thyroid, and pituitary, whereas DIO2 is among others expressed in many brain areas, pituitary, brown adipose tissue, placenta, skeletal muscle, and macrophages. DIO3 is predominantly present in the placenta and neurons in the brain and plays a role in fetal development (Köhrle 2000). Genetic variation in deiodinases is common (Taylor et al. 2015) and previous research in deiodinase knock-out mice proved aberrant serum TH concentrations compared to wild-type ( WT) mice. In general, deiodinases contribute largely to TH tissue status and defects are of interest for the regulation of TH concentrations in specific tissues.
The pathway of sulfation interacts with the deiodination pathway just described. Sulfotransferases sulfate both T4 and T3. Sulfation of T4 (T4S) blocks the outer-ring deiodination, which disables the conversion of T4 into active T3 facilitated by DIO1. Subsequently, the only possible conversion, of T4S to rT3S, leads to an irreversible inactivation of TH. T4S is in fact a preferred substrate for DIO1 which subsequently leads to accelerated degradation of T4S into rT3S (Visser 1994). On the other hand, sulfated T3 (T3S) is reversible, which enables T3S to serve as T3 storage.
Another metabolic pathway, predominantly for T4, is glucuronidation. Glucuronidated T4 (T4G) can serve as a reservoir as well, predominantly in the gut, where β-glucuronidase can deconjugate T4G back to T4. Some medications (such as antiepileptic medication) are glucuronidation-inducing, which can lead to decreased serum T4 levels. Although this does not cause clinical hypothyroidism in initially euthyroid people, it can increase the need for a higher dosage of T4 supplementation in hypothyroid patients (Bianco & Kim 2013).
Once T4 is converted into its active form T3, TRs are primarily necessary to establish the actual effect of T3 either with or without (in)direct DNA binding. Furthermore, T3 can exert its effect independent of TRs via integrins (Flamant et al. 2017, Davis et al. 2021). Three TRs that bind T3 are known; TRα1, TRβ1 and TRβ2 which are encoded by the THRA and THRB genes. TRα1 is predominantly expressed in the brain, heart, spleen, intestine and bone, whereas TRβ1 is mostly expressed in the liver and TRβ2 in the pituitary, paraventricular nucleus of the hypothalamus, retina, and the inner ear. The effects of T3 can roughly be achieved via two pathways: the canonical or noncanonical pathway. The canonical pathway is the ‘classic’ pathway and reaches its effect through gene expression (Yen 2001). TRs bind to TH-responsive elements (TREs) and binding of T3 to TRs promotes coactivators which enhances gene transcription. In the absence of T3, corepressors are active, which reduces gene transcription. An additional pathway that does not require binding to DNA and TREs and where TRs rapidly activate intracellular second-messenger signaling pathways (e.g. PI3K pathway) without gene expression is called noncanonical. The study of Hönes et al. (2017) showed that several TH-dependent physiological effects stay intact in mice with total loss of the canonical pathway. This indicates that TR binding to DNA and TREs can be bypassed via the noncanonical pathway. It also indicates that T3 activates genes in the absence of the classic canonical pathway.
TH transporters, deiodinases and TRs are obviously of great importance to the availability of cellular TH and therefore the effect of TH. Changes in these components can lead to an imbalance between the available serum TH and cellular TH. A few conditions that highlight the necessity to be able to discriminate between serum TH concentrations and specific tissue TH concentrations will be discussed briefly. Mutations in the THRA and THRB genes cause distinct genetic diseases called RTHα and RTHβ. RTHα causes hypothyroidism in TRα-dominant tissues (intestine, bone, heart and brain), whereas serum TH concentrations are relatively normal. This complicates the diagnosis, while it is important to quickly recognize this disease because developmental consequences can be greatly reduced by early treatment with L-T4 (Dumitrescu & Refetoff 2013, van Gucht et al. 2016). An imbalance in TH concentration is also seen in non-thyroidal illness syndrome (NTIS). NTIS is characterized by low serum T4 and T3 concentrations and normal TSH concentrations. During acute inflammation, T3 availability in skeletal muscles is increased, whereas liver T3 concentrations seem to be reduced (Boelen et al. 2011) indicating a disbalance in TH concentrations between the circulation and a variety of organs. However, there are still many uncertainties and it might be of clinical importance to be able to determine tissue TH status during illness to consider appropriate treatment. Another example is the presence of a mutation in the gene coding for the TH transporter MCT8 leading to an x-linked disease which is associated with Allan–Herndon–Dudley syndrome. This disease presents with normal/slightly increased TSH, decreased serum fT4, increased serum T3 and decreased serum rT3. Tissue TH concentrations are, however, varying between different tissues. The MCT8 mutation causes the inability to transport T3 and T4 into cells that express MCT8. Cells that do not express MCT8 but a compensating transporter are subject to high T3 and, therefore, hyperthyroid, likewise cells that express MCT8 without compensating transporter are hypothyroid (Friesema et al. 2010). Determining TH status per tissue would be helpful in this situation.
Tissue-specific metabolic markers
Over the years, several (tissue-specific) markers have been distinguished which might correlate with human thyroid status. Nowadays, innovative techniques that have been developed enhance the ability to discover new biomarkers (Pietzner et al. 2018). In this review, we give an up-to-date overview of different markers per tissue in humans including the strengths and limitations (Table 1).
A summary of the different markers per tissue.
| Tissue | Biomarker | Measurement | Conclusion | Studies |
|---|---|---|---|---|
| Liver | Lipids | Single measurement and MS/NMR-derived metabolomics | Increased in hyperthyroidism; effect independent of influencing factors; careful interpretation | Pietzner et al. 2017, Boumaza et al. 2019 |
| Cholesterol | Single measurement and MS/NMR-derived metabolomics | Increased in hypothyroidism and decreased in hyperthyroidism; effect independent of influencing factors; careful interpretation | Muls et al. 1982, Nishitani et al. 1990, Klausen et al. 1992, Arem et al. 1995, Kung et al. 1995, Hoppichler et al. 1995, Hayashi et al. 1996, O’Brien et al. 1997, Sundaram et al. 1997, Ozata et al. 1998, Martínez-Triguero et al. 1998, Paoli et al. 1998, Becerra et al. 1999, Erem et al. 1999, 2015, Tzotzas et al. 2000, Altinova et al. 2006, Ito et al. 2007, Pietzner et al. 2017, Boumaza et al. 2019 | |
| GPDH | Single measurement | Increased in hyperthyroidism; cannot be measured in blood | Lee et al. 1959, Lee & Lardy 1965, Okamura et al. 1981 | |
| Glycerol | NMR-metabolomics | Increased in hypothyroidism in one study; can be measured in blood; metabolic substrate of GPDH; research is still scarce | Piras et al. 2021 | |
| CD5L | Multi-omics analysis | Positively correlated with fT3 in hyperthyroid state; influences by other factors have to be investigated | Nock et al. 2020 | |
| SHBG | Single measurement | Positively associated with fT3 and fT4; helpful in prediction of outcome resmetirom on NASH; interferences should be taken into account | Dumoulin et al. 1995, Vierhapper et al. 1998, Brenta et al. 1999, Hampl et al. 2003, Heijboer et al. 2016, Ito et al. 2017 | |
| Ferritin | Single measurement | Increased in hyperthyroidism and decreased in hypothyroidism; large between-individual variation; strongly influenced by inflammation and iron status | Kubota et al. 1993, , Sachdeva et al. 2015 | |
| FGF-21 | Single measurement | Inconsistent outcomes in hyperthyroid and hypothyroid patients | Lee et al. 2013, Bonde et al. 2014, Xiao et al. 2015, 2019, Wang et al. 2016 | |
| Cardiovascular | SCUBE1 | Single measurement | Increased in both hypo- and hyperthyroidism | Bilir et al. 2016, Erem et al. 2016 |
| t-PA | Single measurement | Decreased in hyperthyroidism; inversely correlated with T3 and T4 in hyperthyroid state; markedly influenced by other factors | Li et al. 1998 | |
| PAI-1 | Single measurement | Positively correlated with TH in hyperthyroid state; markedly influenced by other factors | Li et al. 1998 | |
| vWF | Single measurement | Positively correlated with TH; markedly influenced by other factors | Dalton et al. 1987, Li et al. 1998, Stuijver et al. 2014 | |
| Coagulation proteins | Proteomic analysis | Relation between proteins involved in the coagulation cascade and TH; markedly influenced by other factors | Engelmann et al. 2015, Alfadda et al. 2018 | |
| ACE | Single measurement | Increased in hyperthyroidism and decreased in hypothyroidism; markedly influenced by other factors | Smallridge et al. 1983, Graninger et al. 1986, Lee et al. 1986, Reiners et al. 1988 | |
| (NT-pro)BNP | Single measurement | Positively correlated with fT3 and fT4 in hyperthyroid state; less clear in hypothyroidism; currently used to evaluate heart failure | Schultz et al. 2004, Christ-Crain et al. 2005, Özmen et al. 2007, Ertugrul et al. 2008, 2009, Ohba et al. 2020 | |
| Bone | Osteocalcin | Single measurement | Positively correlated with fT3 and fT4; keep interfering conditions in mind | Yoneda et al. 1988, Bergmann et al. 1989, Lee et al. 1990, Kojima 1992, Garnero et al. 1994, Lakatos et al. 1997, Akalin et al. 2002 |
| Bone alkaline phosphatase | Single measurement | Positively correlated with fT3 and fT4; no deviant outcomes in hypothyroidism | Garnero et al. 1994, Sabuncu et al. 2001, Akalin et al. 2002, Boruah et al. 2016, Ito et al. 2017 | |
| ICTP | Single measurement | Positively correlated with fT3 and fT4 in hyperthyroidism, not in hypothyroidism; proposed to serve as a biomarker to differentiate between TSH-secreting adenoma and RTHβ patients; currently an obsolete marker | Miyakawa et al. 1996, Loviselli et al. 1997, Nagasaka et al. 1997, Persani et al. 1997, Engler et al. 1999, Isaia et al. 2000, Loviselli et al. 2003 | |
| P1NP | Single measurement | Increased in hyperthyroid state and decreased in hypothyroid state; more research needed | Kužma et al. 2018, Mat Ali et al. 2020 | |
| TRAP-5b | Single measurement | Inconsistent outcomes in hyperthyroid and hypothyroid patients | Jódar et al. 1997, Minisola et al. 2002 | |
| Ca | Single measurement | Increased in hyperthyroidism; hypothyroidism did not cause deviant outcomes; strongly influenced by other conditions | Mosekilde & Christensen 1977, Kobe et al. 1999, Sabuncu et al. 2001, Kisakol et al. 2003, Dhanwal 2011 | |
| Skeletal muscle | CK | Single measurement | Inverse correlation with fT4 in hypothyroidism; literature in hyperthyroidism is scarce; markedly influenced by conditions causing cell degradation | Klein et al. 1980, Khaleeli & Edwards 1984, Suzuki et al. 1997, Beyer et al. 1998, Nakahama et al. 2001, Wan Nazaimoon et al. 2001, Hekimsoy & Kavalali Oktem 2005, Cheng et al. 2020, Hong et al. 2021 |
| Intestine | Intestinal alkaline phosphatase | Single measurement | Decreased in hypothyroid state; hyperthyroid state has not been investigated yet | Watson & Tuckerman 1971, Hodin et al. 1996, Malo et al. 2004 |
| Kidney | Creatinine | Single measurement and NMR-derived metabolomics | Increased in hypothyroidism and decreased in hyperthyroidism; already a general read out of kidney function | Aizawa et al. 1986, Ford et al. 1989, Shirota et al. 1992, Lafayette et al. 1994, Verhelst et al. 1997, Piras et al. 2017, 2020 |
| Pancreas | HOMA-IR | Single measurement | Inconsistent results in thyroid dysfunction and euthyroidism | De Pergola et al. 2007, Roos et al. 2007, Abdelwhap & Foda 2010, Ren et al. 2014, Wang et al. 2018, Bin Obead et al. 2021, Kazukauskiene et al. 2021, Kapadia et al. 2012, Srivastava et al. 2021, Owecki et al. 2006 |
| Non-organ specific markers | Cu | Single measurement | Positively correlated with fT4, T3 and T4; strongly influenced by other factors | Mittag et al. 2012, Jain 2014, Blasig et al. 2016 |
| Se | Single measurement | Inconsistent results | Kralik et al. 1996, Kucharzewski et al. 2003, Mittag et al. 2010, 2012, Blasig et al. 2016, Rasic-Milutinovic et al. 2017 | |
| Se/Cu ratio | Single measurement | Increased in RTHβ patients compared to patients with TSH-secreting pituitary adenoma; could serve as a biomarker to differentiate between these diagnoses or to evaluate TH status in RTHβ patients | Mittag et al. 2012 | |
| Acylcarnitines (ACs) | MS/NMR-derived metabolomics | In hyperthyroidism, medium-chain ACs are positively correlated with fT4; correlation with short and long-chain ACs is less well determined | Wong et al. 2013, Jourdan et al. 2014, Chng et al. 2016, Al-Majdoub et al. 2017, Pietzner et al. 2017, Lange et al. 2018 | |
| MDA | Single measurement | Increased in both hypo-and hyperthyroidism | Adali et al. 1999, Alicigüzel et al. 2001, Cetinkaya et al. 2005, Erdamar et al. 2008, Lassoued et al. 2010, Erem et al. 2015 | |
| IMA | Single measurement | Increased in both hypo-and hyperthyroidism | Reddy et al. 2017 | |
| GGAA | MS-derived metabolomics | Positively correlated with fT4; research is still scarce | Pietzner et al. 2017 | |
| PCs | MS/NMR-derived metabolomics | Positively correlated with fT4; research is still scarce | Jourdan et al. 2014, Al-Majdoub et al. 2017, Lange et al. 2018 | |
| LPs | MS-derived metabolomics | Inversely correlated with fT4; research is still scarce | Al-Majdoub et al. 2017, Pietzner et al. 2017 | |
| ADMA | Single measurement and MS-derived metabolomics | Positively correlated with fT4; also increased after adequate treatment for hyperthyroidism | Hermenegildo et al. 2002, Arikan et al. 2007, Gu et al. 2011, Ittermann et al. 2016, Pietzner et al. 2017 | |
| Complement proteins | Single measurement and proteomic analysis | Associations with TH were found, especially an inverse relation between fT3 and fT4 and C3; markedly influenced by other factors | Blanchin et al. 2003, Yu et al. 2006, Jafarzadeh et al. 2010, Pietzner et al. 2017, Alfadda et al. 2018 | |
| Phenylalanine | MS-derived metabolomics | Increased in hyperthyroidism; correlation in hypothyroidism has not been investigated yet | Chng et al. 2016 | |
| Tyrosine | MS/NMR-derived metabolomics | Increased in hyperthyroidism; correlation in hypothyroidism has not been investigated yet | Chng et al. 2016, Piras et al. 2017 | |
| Serine | NMR-derived metabolomics | Decreased in hyperthyroidism and increased in hypothyroidism; research is still scarce | Piras et al. 2017, 2020 | |
| Metabolic fingerprint | NMR-derived metabolomics | Significantly different metabolic profile in hypothyroid mice and mice with RTHα; profile might be used to discriminate between these conditions in humans as well. | Boumaza et al. 2019 |
MS, mass-spectrometry, NMR, nuclear magnetic resonance
Liver
The liver is an important TH target tissue with several metabolic functions being TH-dependent, such as lipid and carbohydrate metabolism and therefore regulating energy balance (Sinha et al. 2019). It expresses mainly TRβ1, although TRα is also expressed to a lesser degree and is heavily involved in thyroid metabolism via high expression of DIO1. Several liver-related conditions, like hypercholesterolemia and nonalcoholic fatty liver disease, have been associated with TH status (Sinha et al. 2019).
Lipids and cholesterol
THs are highly involved in the regulation of lipid and cholesterol metabolism via various mechanisms; T3 increases the transcription rate of genes encoding for enzymes involved in lipogenesis and beta-oxidation, and induces HMG-CoA reductase in the liver which in turn stimulates cholesterol synthesis (Cachefo et al. 2001, Sinha et al. 2019). Overall, hypothyroidism resulted in an increase in concentrations of LDL, HDL, total cholesterol, Apo B, and triglycerides, whereas hyperthyroidism was characterized by a decrease in predominantly LDL and total cholesterol concentrations (Nishitani et al. 1990, Kung et al. 1995, Sundaram et al. 1997, Erem et al. 1999, 2015, Altınova 2006). The other markers showed inconsistent results. Restoring euthyroidism resulted in changes of these markers toward concentrations comparable to healthy euthyroid controls (Muls et al. 1982, Klausen et al. 1992, Arem et al. 1995, Hoppichler et al. 1995, Kung et al. 1995, Hayashi et al. 1996, O’Brien et al. 1997, Martínez-Triguero et al. 1998, Ozata et al. 1998, Paoli et al. 1998, Becerra et al. 1999, Tzotzas et al. 2000, Ito et al. 2007, Erem et al. 2015). Therefore, a serum lipid and cholesterol profile has been shown to be a marker of TH action. Research in thyroidectomized patients using L-T4 showed that biochemical euthyroidism (defined by TSH within the reference range) while treated may be associated with increased cholesterol concentrations (Lee et al. 2019). This strengthens the assumption that cholesterol could be of interest as a biomarker to determine liver TH status. Innovative techniques confirmed the association between lipids, cholesterol and TH status. Analysis via untargeted NMR-derived metabolomics demonstrated that hypothyroid mice had decreased concentrations of VLDL and LDL and increased concentrations of unsaturated lipids and cholesterol compared to euthyroid mice (Boumaza et al. 2019). Mass spectrometry (MS)-derived metabolome analysis of hyperthyroid humans also revealed altered lipids and lipid-related compounds (free fatty acids, polyunsaturated fatty acids, acylcarnitines and lysophospholipids) compared to healthy euthyroid controls (Pietzner et al. 2017). A hyperthyroid state was characterized by increased concentrations of these lipids, while LDL, HDL and total cholesterol showed a decrease in the hyperthyroid state as shown previously. Thus, lipogenesis and cholesterol metabolism are highly influenced by TH. However, markers of the lipid and cholesterol profile are also markedly influenced by several conditions as BMI, cardiovascular diseases, diabetes mellitus and the use of statins. Nevertheless, the abovementioned studies found a correlation with TH even though most of the interfering conditions were excluded. Therefore, derivatives of lipid and cholesterol metabolism could be used as a biomarker to establish TH in the liver, although these markers should be interpreted with caution in patients with cardiovascular diseases and diabetes mellitus or using statins.
Glycerol
One of the first metabolic markers that has been linked to thyroid status is liver glycerol-3-phosphate dehydrogenase (GPDH) (Lee & Lardy 1965). GPDH reduces dihydroxyacetone phosphate into glycerol 3-phosphate, allowing the dephosphorylation of glycerol 3-phosphate into glycerol. Hyperthyroid rats showed a significant increase of liver GPDH indicating it is useful in measuring intrahepatic TH status (Lee et al. 1959, Lee & Lardy 1965, Okamura et al. 1981). It is suggested that the effect of T3 acts via coactivators and corepressors modulating GPDH activity (Gong et al. 1998). Although GPDH could be useful as an additional marker in assessing long-term changes in TH status, it cannot be measured in serum (Rauchová et al. 2011). Interestingly, Piras et al. (2021) recently found increased glycerol concentrations in hypothyroid patients using NMR metabolomics, which remained increased after treatment and restoration of TH concentrations. This suggests that liver TH status restores more slowly than serum TH concentrations. Literature regarding glycerol as a biomarker is scarce and more research is needed to establish a correlation between glycerol and TH status in the liver.
CD5 antigen-like (CD5L)
Recently, Nock et al. (2020) aimed to identify new biomarkers representative for TH function by using untargeted multi-omics analyses of plasma samples from thyrotoxic humans and mice (Engelmann et al. 2015, Pietzner et al. 2017). They showed that CD5L is a robust and promising marker for TH status since it correlates specifically well with serum fT3 concentrations. CD5L is a protein involved in the modulation of leukocyte function and belongs to the scavenger receptor cysteine-rich superfamily. Within the liver, it is mainly expressed in macrophages (Kupffer cells) and positively controlled by the transcription factor LXR, which is expressed in tissues with high metabolic activity (e.g. liver, adipocytes and macrophages) (Sanjurjo et al. 2015). Outside the liver, the main source of CD5L is macrophages. TH has an indirect effect on CD5L expression which appears to be TRβ-dependent, since TRβ knock-out mice do not show an increase in CD5L expression after T3 exposure. Macrophages express predominantly TRα and CD5L is thus proposed as a TRβ-dependent biomarker reflecting hepatic TH status. Due to its TRβ dependency, it might also be used to detect RTHβ patients. CD5L was already suggested as a biomarker for hepatocellular carcinoma and liver cirrhosis (Gray et al. 2009, Yamazaki et al. 2014). It is known that CD5L is easily detectable in serum, but literature also shows that CD5L concentrations differ between males and females and decrease with age (Yamazaki et al. 2014). In addition, inflammatory processes are proved to alter this biomarker and CD5L is known to be involved in several other diseases such as atherosclerosis, chronic kidney disease, chronic obstructive pulmonary disease, and cancer. In short, the study of Nock et al. (2020) showed that CD5L is an interesting candidate as a biomarker for hepatic TH function and it can also be used to identify patients with RTHβ. However, it needs to be investigated whether it is still distinctive in patients with different types of thyroid dysfunction and comorbidities. But since this marker could be used for both treatment and diagnostic purposes, it seems worthwhile investigating this biomarker.
Sex hormone-binding globulin (SHBG)
SHBG as a marker for TH action has been described extensively (Olivo et al. 1970, Anderson 1974, Dumoulin et al. 1995, Vierhapper et al. 1998). SHBG is a binding protein contributing to the transportation of sex hormones to their target cells. TH increases hepatic gene expression of SHBG indirectly via hepatocyte nuclear factor 4 α (HNF4A)), a nuclear transcription factor, leading to increased serum concentrations of SHBG (Pugeat et al. 1996, Selva & Hammond 2009). In hyperthyroidism, serum SHBG concentrations are increased, whereas reduced SHBG concentrations are observed in hypothyroidism (Dumoulin et al. 1995, Vierhapper et al. 1998, Brenta et al. 1999, Hampl et al. 2003, Ito et al. 2017). Furthermore, SHBG concentrations normalize when thyroid status is restored and a significant positive correlation between SHBG and TH (fT4 and fT3) concentrations has been observed (Dumoulin et al. 1995, Hampl et al. 2003, Ito et al. 2017). Recently, SHBG concentrations in patients during treatment with a TR agonist, resmetirom, were shown to predict the effects on reduction of liver steatosis (Harrison et al. 2019). This shows the clinical applicability of this marker as a non-invasive marker of TH action. Furthermore, SHBG concentrations have been shown to discriminate between the thyrotoxic state in patients with a TSH-secreting adenoma and patients with RTH (Sarne et al. 1988, Beck-Peccoz et al. 1990). The correlation, especially in a hyperthyroid state, seems beyond question; however, there are multiple factors that interfere with SHBG concentrations as well, such as BMI, insulin resistance, age, use of oral contraceptives and pregnancy (Brenta et al. 1999, Heijboer et al. 2016). In conclusion, SHBG has been researched extensively, concentrations correlate well with serum TH concentrations and intrahepatic action and SHBG can be useful for the diagnosis of RTH and monitoring of therapy (e.g. resmetirom). However, the abovementioned interferences should be taken into account and need the appropriate attention before implementing this biomarker.
Ferritin
Back in the ’80s, serum ferritin was proposed as a marker indicating tissue TH status (Takamatsu et al. 1985). T3 regulates hepatic ferritin expression via TRs by modulating the iron-responsive element-binding activity of the iron regulatory protein (Leedman et al. 1996). Hyperthyroid patients presented increased serum ferritin concentrations, whereas hypothyroid patients showed decreased concentrations (Kubota et al. 1993, Sachdeva et al. 2015). Ferritin concentrations decreased after treatment for hyperthyroidism and increased after treatment for hypothyroidism to concentrations within the reference interval. However, serum ferritin concentrations were not significantly correlated with TH and literature points out that between-individual variation in ferritin concentrations is large (Takamatsu et al. 1985, Fischli et al. 2017). Moreover, ferritin is strongly influenced by common circumstances such as inflammation and iron depletion. Inflammation causes increased ferritin concentrations, whereas iron depletion due to, for example, intestinal or menstrual blood loss leads to decreased ferritin concentrations. These disadvantages combined outweigh the possibilities of ferritin as a TH biomarker.
Fibroblast growth factor 21 (FGF21)
The hormone fibroblast growth factor 21 (FGF21) is a part of the FGF superfamily and is predominantly released by hepatocytes (Li et al. 2013). FGF21 plays a role in the regulation of glucose, lipid and energy metabolism. TH metabolism is known to affect energy metabolism as well, so TH has been linked to FGF21 expression. It was shown that T3 directly induced hepatic FGF21 expression via PPARα-dependent mechanisms (Adams et al. 2010). In addition, increased FGF21 concentrations in hyperthyroid patients and decreased concentrations in hypothyroid patients have been observed, both improving after treatment (Xiao et al. 2015, 2018, Wang et al. 2016). However, inverse correlations as well as no correlation at all are also reported (Lee et al. 2013, Bonde et al. 2014). Furthermore, mice lacking FGF21 (FGF21 knock-out mice) and WT mice displayed similar effects of TH on hepatic metabolic action indicating that TH and FGF21 most likely have independent actions on energy metabolism (Domouzoglou et al. 2014, Zhang et al. 2015). Therefore, FGF21 seems unsuitable as a biomarker reflecting hepatic TH status.
Cardiovascular
TH affects the heart and skeletal muscles directly by improving myocardial relaxation and indirectly by causing peripheral vasodilatation (Vargas-Uricoechea and Sierra-Torres 2014). It is known that thyroid dysfunction is strongly correlated with cardiovascular diseases (Ochs et al. 2008, Flynn et al. 2010).
Signal peptide-CUB-EGF domain containing protein 1 (SCUBE1)
SCUBE1 is a cell surface glycoprotein, part of the epidermal growth factor (EGF) superfamily. It is highly expressed in platelets, enhances platelet–platelet aggregation and is present in thrombi (Tu et al. 2006). The study by Erem et al. (2016) showed that patients with (subclinical) hyperthyroidism had significantly elevated concentrations of SCUBE1 compared to healthy controls. It is known that (subclinical) hyperthyroidism is associated with hypercoagulability and is, therefore, thought to increase the risk of cardiovascular diseases in this group (Erem 2011). There was no difference in SCUBE1 concentrations between subclinical and overt hyperthyroidism. SCUBE1 concentrations decreased significantly after treatment with antithyroid drugs. The authors therefore suggested that SCUBE1 could be an interesting biomarker for platelet activation and inflammation in patients with (subclinical) hyperthyroidism, especially in the earlier stages. However, Bilir et al. (2016) showed that SCUBE1 concentrations are also increased in patients with hypothyroidism compared to healthy controls. SCUBE1 concentrations after treatment were not investigated in this study. The mechanism underlying the effect of TH on SCUBE1 concentrations has not been elucidated so far. Since both hypo- and hyperthyroidism are characterized by increased SCUBE1 concentrations, this biomarker will only be able to discriminate between a normal or abnormal thyroid tissue status and is therefore not suitable as a biomarker of tissue TH status.
Coagulation factors
Coagulation factors tissue plasminogen activator (tPA), plasminogen activator inhibitor (PAI-1) and von Willebrand factor (vWF) were associated with a hyperthyroid state (Li et al. 1998). It is hypothesized that TH influences the synthesis of endothelium-derived proteins, including tPA, PAI-1 and vWF via TRs that are expressed in endothelial cells (Dietrich et al. 1997, Burggraaf et al. 2001). Elbers et al. (2016) proved that the hypercoagulable state in hyperthyroidism is mediated through the TRβ. However, it is also hypothesized that T3 amplifies vWF secretion by affecting the adrenergic system (catecholamines acting via β2-adrenergic receptors) which regulates vWF secretion (Vischer & Wollheinn 1997). Hyperthyroid patients showed significantly lower concentrations of tPA antigen compared to healthy controls and these concentrations restored in the euthyroid state. A significant inverse relationship between TH (T3 and T4) and tPA antigen concentrations was found. PAI-1 and vWF concentrations were significantly increased in hyperthyroid patients compared to euthyroid controls and decreased to normal concentrations in the euthyroid situation. Both parameters showed a significant positive correlation with TH concentrations. Conversely, hypothyroid patients can present with acquired von Willebrand’s disease (VWD) with diminished vWF concentrations, which increased upon treatment (Dalton et al. 1987, Stuijver et al. 2014). This hypothyroid-associated acquired VWD is characterized by a decrease of vWF protein synthesis due to a lack of sufficient concentrations of T4 (Franchini 2004). tPA antigen and PAI-I concentrations have not been investigated in hypothyroid patients. Proteins involved in the coagulation cascade in relation to fT4 status were also found using mass spectrometry-derived proteomics and thereby endorse the relation between TH and coagulation (Engelmann et al. 2015, Alfadda et al. 2018). Nevertheless, the abovementioned clotting factors do not seem suitable as biomarkers to adequately evaluate TH status, since the literature on these biomarkers is limited, the relation in hypothyroid patients is not clear and other conditions also affect coagulation and endothelial dysfunction, such as insulin-dependent diabetes, hemodialysis or coronary occlusions (Paramo et al. 1985, Jensen et al. 1989, Nakayama et al. 1994). However, it is important to be aware that the thyroid state influences the process of blood coagulation (Elbers et al. 2018) and that adequate treatment normalizes aberrant concentrations of its parameters.
Angiotensin-converting enzyme (ACE)
In the 1980s, a few papers were published describing the relation between angiotensin-converting enzyme (ACE) and TH. T3 is known to induce ACE protein synthesis in pulmonary artery endothelial cells, but the exact mechanism by which T3 influences ACE remains unclear (Dasarathy et al. 1990). The studies showed that ACE concentrations were significantly higher in hyperthyroid patients compared to controls and significantly lower in hypothyroid patients compared to controls. Furthermore, ACE concentrations normalized after reaching euthyroidism (Smallridge et al. 1983, Graninger et al. 1986, Lee et al. 1986, Reiners et al. 1988). However, no further research has been done on this topic for the past several decades and as ACE is influenced by many factors (e.g. blood pressure regulation and salt management), we consider this biomarker as not feasible to be used in clinical practice.
Brain natriuretic peptide (BNP)
Thyroid dysfunction is associated with hemodynamic changes due to reduced heart contractility, increased cardiac output due to higher demands and even heart failure. Brain natriuretic peptide (BNP) is a marker released by the ventricular myocardium in response to volume expansion to monitor heart failure. Multiple studies showed that (NT-pro)BNP concentrations were significantly higher in hyperthyroid patients compared to hypothyroid and euthyroid patients (Schultz et al. 2004, Christ-Crain et al. 2005, Özmen et al. 2007, Ertugrul et al. 2008, Ohba et al. 2020). Adequate treatment of hyperthyroidism resulted in a significant decrease of serum (NT-pro)BNP concentrations (Ertugrul et al. 2009). Furthermore, a significant positive correlation was found between (NT-pro)BNP and fT3 and fT4 concentrations and a negative correlation was found between BNP and TSH concentrations. The situation in hypothyroidism is less clear. Özmen et al. (2007) and Christ-Crain et al. (2005) showed no difference in NT-proBNP concentrations between hypothyroid patients and controls, whereas Ertugrul et al. (2008) did find a correlation. TH exerts its function on BNP via a specific TRE in the BNP gene which is activated by T3 resulting in increased BNP gene transcription (Liang et al. 2003). (NT-pro)BNP is not suitable as a functional biomarker for heart TH status since no clear relationship in hypothyroid patients was found and (NT-pro)BNP is already used as a marker to evaluate heart failure independent of TH status. However, it should be kept in mind that elevated concentrations of (NT-pro)BNP without indications for heart failure could be a sign of hyperthyroidism.
Bone
Bones are sensitive to TH status which is clinically emphasized by the presence of osteoporosis in hyperthyroidism. TH acts predominantly via TRα in osteoblasts. Osteoclasts and chondrocytes also express TRα1, TRα2 and TRβ1 (Bassett & Williams 2003, Williams et al. 2008, Capelo et al. 2009). Furthermore, DIO3 is expressed in osteoclasts, osteoblasts and chondrocytes, whereas DIO2 is present in mature osteoblasts making it possible to regulate TH status at the tissue level. Research in thyroidectomized postmenopausal women using L-T4 showed that biochemical euthyroidism while treated may be associated with decreased bone mineral density (Ku et al. 2021). This suggests that markers involved in bone metabolism could be of interest to determine bone TH status.
Osteocalcin
Osteocalcin is a carboxyglutamic acid-containing osteoblast-specific protein and is secreted in the extracellular matrix of the bone. In the carboxylated form, it plays a role in calcium binding and bone formation, whereas in the uncarboxylated form, it reaches the blood circulation and acts as a hormone by exerting an effect on multiple organs/tissues throughout the body (Karsenty & Olson 2016). The synthesis and secretion of osteocalcin are stimulated by T3 via the expression of TRs in osteoblasts (Rizzoli et al. 1986, Gouveia et al. 2001, Siddiqi et al. 2002). Serum osteocalcin concentrations are lower in hypothyroid patients (Yoneda et al. 1988, Bergmann et al. 1989, Kojima et al. 1992) and higher in hyperthyroid patients compared to euthyroid controls (Yoneda et al. 1988, Lee et al. 1990, Kojima et al. 1992, Garnero et al. 1994, Lakatos et al. 1997, Akalin et al. 2002). Osteocalcin concentrations return to normal after adequate treatment, although the study by Kojima et al. (1992) showed that normalization could take up to 24 months, while TH concentrations in the blood were already restored. A positive correlation between fT4 and fT3 concentrations and osteocalcin was observed in most studies. Other conditions such as parathyroid disorders, vitamin D deficiency and osteoporosis are known to affect this marker as well, although some studies found the correlation with TH independent of these interfering conditions. Therefore, in most cases, osteocalcin can be used as a biomarker reflecting bone TH status and thus may be useful in optimizing treatment looking at bone-related complications of thyroid disease (e.g. osteoporosis).
Bone alkaline phosphatase
Bone alkaline phosphatase has been considered as another marker of TH status in bone. Bone alkaline phosphatase concentrations in blood are increased in hyperthyroid patients compared to euthyroid controls (Garnero et al. 1994, Sabuncu et al.2001, Akalin et al. 2002, Boruah et al. 2016, Ito et al. 2017) and a positive correlation was observed between bone alkaline phosphatase and fT3 and fT4 in both hypothyroid and hyperthyroid and euthyroid patients (Sabuncu et al. 2001, Boruah et al. 2016). Bone alkaline phosphatase concentrations eventually decreased after adequate treatment for hyperthyroidism, although normalization could take a few months. In contrast, hypothyroid patients did not show aberrant bone alkaline phosphatase concentrations when compared to euthyroid controls (Sabuncu et al. 2001, Boruah et al. 2016). Therefore, this biomarker is not feasible to evaluate bone TH status.
Collagen biomarkers
Carboxyterminal cross-linked telopeptide of type 1 collagen (ICTP) has been proposed as a marker indicating TH status as well. As previously mentioned, TH stimulates among others osteoclasts. During hyperthyroidism, osteoclasts are activated and secrete enzymes that resorb bone. In this process, fragments of collagen are released and incompletely digested. Since ICTP is a cross-link of such fragments, its concentration will increase. Several studies showed increased ICTP concentrations in hyperthyroid patients compared to euthyroid controls together with positive correlations between ICTP and fT3 and fT4 (Miyakawa et al. 1996, Loviselli et al. 1997, 2003, Nagasaka et al. 1997, Engler et al. 1999, Isaia et al. 2000). However, ICTP concentrations did not differ between hypothyroid patients and euthyroid controls (Miyakawa et al. 1996). ICTP has been proposed as a discriminative marker in distinguishing patients with TSH-secreting adenomas and RTH patients (Persani et al. 1997) and multiple guidelines recommend measuring this marker to discriminate between those conditions (Beck-Peccoz et al. 2013). However, the recommendation to use ICTP as a discriminative marker in patients with inappropriate secretion of TSH is based on one study describing 40 RTH patients and 10 patients with a TSH-secreting adenoma (Persani et al. 1997). Importantly, total amino-terminal peptide of procollagen type 1 (P1NP) is now commonly used as a bone marker instead of ICTP (Vasikaran et al. 2011). P1NP also correlates with thyroid status. P1NP is a precursor of collagen type 1 synthesized by osteoblasts and is considered a bone formation marker. P1NP concentrations increased in hyperthyroidism, whereas its concentrations diminished in hypothyroidism (Kužma et al. 2018, Mat Ali et al. 2020). Concentrations seem to normalize after reaching euthyroidism. However, literature regarding P1NP concentrations is still scarce, so its true relevance remains to be determined.
Tartrate-resistant acid phosphatase 5b (TRAP5b)
Tartrate-resistant acid phosphatase 5b (TRAP5b) is a specific marker of osteoclasts and reflects bone absorption (Halleen et al. 2000). Literature regarding this marker in relation to thyroid hormones is scarce and showed varying results because a decrease in TRAP5b concentrations was observed in both hyperthyroid patients compared to healthy controls and after treatment of hyperthyroidism (Jódar et al. 1997, Minisola et al. 2002). Therefore, this biomarker is not suitable to reflect bone TH status adequately.
Calcium (Ca)
Long-term exposure to T3 and T4 directly stimulates osteoclastic bone resorption (Mundy et al. 1976). Increased exposure to TH (e.g. during hyperthyroidism) thus causes an increase in serum Ca concentration (Mosekilde & Christensen 1977, Sabuncu et al. 2001, Dhanwal 2011). Ca concentrations normalized after adequate treatment (Sabuncu et al. 2001). Short-term exposure as well as subclinical hyperthyroidism and hypothyroidism did not show deviant Ca concentrations (Kobe et al. 1999, Sabuncu et al. 2001, Kisakol et al. 2003). Serum Ca is not considered as a reliable biomarker reflecting bone TH status because a hypothyroid state is not correlated with Ca concentrations and serum Ca is strongly influenced by other factors (e.g. parathyroid, intake, osteoporosis, and vitamin D deficiency).
Skeletal muscle
Creatine kinase (CK)
Several case reports described myopathy and a concordant increase in creatine kinase (CK) concentrations as the first expression of hypothyroidism (Burnett et al. 1994). Literature showed a significant increase in CK concentration in hypothyroid patients compared to euthyroid controls and a significant inverse correlation between CK and fT4 concentrations (Beyer et al. 1998, Hekimsoy & Kavalali Oktem 2005, Hong et al. 2021). After treatment for hypothyroidism, CK concentrations normalized (Klein et al. 1980, Khaleeli & Edwards 1984, Nakahama et al. 2001, Hekimsoy & Kavalali Oktem 2005). In hyperthyroidism, CK concentrations seemed to decrease, but literature is scarce (Wan Nazaimoon et al. 2001). Increased CK concentrations were also found in hyperthyroid patients receiving antithyroid therapy indicating a short period of hypothyroidism when therapy is initiated (Suzuki et al. 1997, Cheng et al. 2020). The mechanism remains unclear; it is unlikely that TH directly affects CK concentrations since high concentrations of T4 and T3 did not reduce CK concentrations in vitro (Smith 1976). Knowledge on CK and hyperthyroidism is limited, but CK concentrations might be useful to get an impression of TH status in skeletal muscle during hypothyroidism. However, it should be kept in mind that CK indicates cell degradation which is also seen in, for example, myocardial infarction and therefore lessens its specificity as a marker of TH status in skeletal muscle.
Intestine
TH exerts its effect on the intestine via the epithelial cells, the enterocytes (Ebert 2010). T3 affects the development of the intestinal epithelium and regulates the levels of specific brush-border enzymes via TRα1 (Plateroti et al. 2001). There is increasing evidence of the existence of a thyroid–gut axis (Knezevic et al. 2020), highlighting the connection between these two organs.
Intestinal alkaline phosphatase (IAP)
Intestinal alkaline phosphatase (IAP) is expressed and secreted by intestinal epithelial cells and its expression is highest in the duodenum. IAP’s primary function is to maintain intestinal homeostasis. T3 increased the transcription of IAP in both humans and rats (Hodin et al. 1992, 1996, Malo et al. 2004) and serum IAP concentrations were decreased in hypothyroid rats (Watson & Tuckerman 1971, Hodin et al. 1996, Malo et al. 2004). However, IAP concentration as a biomarker of TH status in the intestine has not been further investigated. It might be an interesting biomarker to evaluate intestinal TH status and gain more insight into intestinal symptoms in thyroid disorders, which could lead to better treatment.
Kidney
Another TH target organ is the kidney, which expresses DIO1. TH regulates the expression of ion channels and transporters via TRα1 directly (Li et al. 2002). Indirectly, decreased renal blood flow in hypothyroidism (due to decreased cardiac output, myocardial contractility and increased peripheral resistance) can cause an increase in creatinine and thus decrease of estimated glomerular filtration rate (eGFR) (Chou et al. 2011).
Creatinine
Hypothyroidism is in a large part of the patients characterized by a reversible reduction of GFR, whereas hyperthyroidism causes an elevation in GFR (Chou et al. 2011). This results in an elevation of the creatinine concentration in hypothyroidism and a reduction of creatinine concentration in hyperthyroidism (Aizawa et al. 1986, Ford et al. 1989, Lafayette et al. 1994, Shirota et al. 1992, Verhelst et al. 1997). This pattern is confirmed by NMR-metabolomic studies (Piras et al. 2017, 2021). Since creatinine is a general readout of kidney function, it cannot be used as a TH-specific biomarker.
Pancreas
Homeostasis model assessment for insulin resistance (HOMA-IR)
TH has direct effects on insulin production as T3 and TRα stimulate the proliferation of pancreatic islet cells and T3 is necessary for the transition of islets to glucose-responsive insulin-secreting cells (Furuya et al. 2010). Homeostasis model assessment for insulin resistance (HOMA-IR) is a model to assess insulin resistance based on fasting insulin and fasting glucose and has widely been investigated in relation to thyroid hormones and thyroid function. Literature reported both a negative association between HOMA-IR and both fT4 and fT3 in euthyroid participants (Roos et al. 2007, Abdelwhab & Foda 2010, Bin Obead et al. 2021), a positive association (Ren et al. 2014, Wang et al. 2018) as no association at all (De Pergola et al. 2007, Kazukauskiene et al. 2021). In both hypothyroid and hyperthyroid patients, increased levels of HOMA-IR were found compared to euthyroid controls (Kapadia et al. 2012, Srivastava et al. 2021), although no difference between hypothyroid patients and euthyroid controls was found as well (Owecki et al. 2006). In conclusion, HOMA-IR is a model assessing insulin resistance with possible associations with thyroid function but should not be used as a TH tissue biomarker.
Non-organ specific markers
Trace elements
Copper (Cu) and selenium (Se)
An increase in T3 and T4 is associated with an increase in Cu concentrations (Mittag et al. 2012, Blasig et al. 2016). Cu is a trace element and important for growth and development. Cu deficiency leads to anemia, growth defects, and susceptibility to infections. Research in mice showed that the liver is an important organ in Cu homeostasis which is affected by TH via the TRβ, the main TR in the liver (Flores-Morales et al. 2002, Cheng et al. 2010). TH downregulates competing intracellular Cu-binding proteins and enhances synthesis and export of hepatic ceruloplasmin. 95% of Cu in healthy human plasma is carried by the transport protein ceruloplasmin (Hellman & Gitlin 2002). T3 provides an increased expression of transporters ATP7A and ATP7B and ceruloplasmin and decreases the production of Cu-containing enzymes SOD1, Mtt1 and Mtt2 (Mittag et al. 2012). It is conceivable that ceruloplasmin might be correlated to TH, but no research has been performed yet regarding ceruloplasmin as a potential interesting biomarker. Blasig et al. (2016) showed that Cu concentrations were positively correlated with plasma T3 and T4 concentrations in children with congenital hypothyroidism and suggested that Cu status could serve as a diagnostic tool for monitoring the effect of TH substitution in these children or even at a more population-based scale. The study of Jain (2014) confirmed this correlation in the overall population for fT4 and TT4 in men and TT3 and TT4 in women. However, despite the fact that Cu concentrations are very stable and not susceptible to small changes, Cu concentrations are about 20% higher in males compared to females (Jain 2014) and other elements as cadmium can influence their concentrations. Furthermore, it should be taken into account that Cu is an acute phase reactant and its concentration increases during inflammation. Concluding, Cu does not contribute as a biomarker reflecting tissue TH status.
Another trace element that has been positively correlated to T3 concentrations is Se (Mittag et al. 2010, 2012). Se is required for the production of selenoproteins (DIO1, DIO2 and DIO3). Decreased Se concentrations together with decreased serum TH concentrations were observed in critical illness (Gärtner 2009). However, after recovery, T3 and T4 concentrations improved to normal, whereas Se concentrations remained low (Angstwurm et al. 2004). Furthermore, not all studies showed a positive correlation between Se and TH (Kralik et al. 1996, Kucharzewski et al. 2003, Blasig et al. 2016, Rasic-Milutinovic et al. 2017) and Se concentrations are influenced by inflammation. Therefore, its relevance as a TH biomarker is doubted.
The combination of Cu and Se has been studied to discriminate between RTHβ and TSH-secreting pituitary adenoma (Mittag et al. 2012). RTHβ is characterized by high serum fT3 and fT4 and normal/slightly elevated TSH and TSH-secreting pituitary adenoma can be presented with the same biochemical abnormalities. Based on decreasing Cu and increasing Se concentrations observed in RTHβ patients compared to healthy controls, the authors concluded that the Se/Cu ratio might be useful as a biomarker, since the Se/Cu ratio increased in RTHβ patients compared to patients with TSH-secreting pituitary adenoma. Furthermore, the Se/Cu ratio could be used to assess TH therapy in RTHα patients. Cu concentrations increased after initiation of therapy due to TRβ activation, whereas Se concentrations remained stable, indicative of TRα resistance (Moran et al. 2013). A decreased Se/Cu ratio could then indicate therapy response. Cu and Se concentrations are easily measured in serum and therefore potentially feasible. However, as mentioned before, both Cu and Se are influenced by other factors. To conclude, the Se/Cu ratio is potentially interesting as a biomarker in the diagnosis of RTH; however, its role should be handled with caution due to possible interferences.
Fatty acid metabolism
Acylcarnitines (ACs)
Acylcarnitines (ACs) play an important role in the transport of long-chain fatty acids (LCFA) across the mitochondrial membrane. LCFAs are acylated, resulting in the formation of acyl-coenzyme A (CoA). Carnitines interact with acyl-CoA under influence of carnitine palmitoyltransferase 1 (CPT-1) on the outer aspect of the inner mitochondrial membrane to form free CoA and acylcarnitine. ACs are transported across the inner mitochondrial membrane in exchange for carnitine. In the inner membrane, the acyl group is re-esterified with CoA by CPT2, resulting in acyl-CoA and carnitine. Acyl-CoA is converted into acetyl-CoA, which will go into the tricarboxylic acid cycle. It is known that TH affects β-oxidation and oxidative phosphorylation. T3 increases the expression of the CPT1 gene via the TRα resulting in elevated CPT1 concentrations, which enhances fatty acid β-oxidation (Mullur et al. 2014). The AC profile consists of short, medium and long-chain ACs. Short-chain ACs originate from glucose and amino acid metabolism, whereas medium and long-chain ACs are derived from fatty acid metabolism (Al-Majdoub et al. 2017). Many studies using metabolomics showed a correlation between TH status and ACs. In healthy men supplemented with L-T4 to mimic a hyperthyroid state, medium-chain ACs increased at the cost of short-chain ACs (Pietzner et al. 2017). Furthermore, hyperthyroid patients receiving treatment showed reduced concentrations of medium-chain, long-chain and total ACs after reaching euthyroidism in one study (Chng et al. 2016), whereas the study of Al-Majdoub et al. (2017) only found this reduction in medium-chain ACs. Short-chain ACs were elevated and long-chain ACs were unchanged. In euthyroid participants, positive associations between fT4 and ACs were found; however, no distinction was made between differences in chain length (Jourdan et al. 2014, Lange et al. 2018). In contrast, Wong et al. (2013) investigated AC profiles related to hypothyroidism, euthyroidism and hyperthyroidism and found no difference in these concentrations between the three different thyroid states. It is possible that due to their small sample size (six hypothyroid and six hyperthyroid patients) and lack of a control group (the euthyroid groups consisted of the treated hypo- and hyperthyroid group), outcomes are deviant. Overall, there is a correlation between ACs and fT4 concentrations, which indicates this profile could be of interest as a potential biomarker for establishing TH status. However, there are some discrepancies between the several studies. Previous literature showed that medium-chain ACs are quickly lowered after refeeding in the fasting state, while long-chain ACs remained stable (Ramos-Roman et al. 2012). Since the transition from hyperthyroidism to euthyroidism is characterized by decreased metabolic demand, this could explain that predominantly medium-chain ACs were altered and long-chain ACs remained stable. However, outcomes regarding the correlation of short and long-chain ACs differed between the studies. Most studies sampled blood in the morning in the fasting state, so this could not have affected the outcomes. It might be due to dietary habits (Kien et al. 2014). It is known that diet influences AC metabolism and since populations from different continents with different dietary habits were included, this could explain the altered results concerning chain length. Since multiple studies using metabolomics found a significant positive relationship between ACs and fT4 concentrations and especially medium-chain ACs were altered, ACs have potential as a biomarker reflecting TH status. They could be helpful in giving a better representation of euthyroidism and optimizing the treatment of thyroid dysfunction. Future studies should focus on evaluating the relevance of the different AC chain lengths in relation to TH status. Furthermore, research needs to prove whether TH can be related to one specific AC or a profile of (e.g. medium-chain) ACs would correlate better.
Oxidative stress
Hyperthyroidism and subclinical hyperthyroidism are associated with oxidative stress and, therefore, with elevated markers of lipid peroxidation. Free radicals, a result of oxidative stress, are able to react with free fatty acids, the major energy source of thyrocytes, resulting in lipid peroxidation. Malondialdehyde (MDA) is an important lipid peroxidation marker, which is produced in response to cell death in peripheral tissues. Significantly elevated concentrations of MDA were observed in patients with (subclinical) hyperthyroidism compared to euthyroid controls (Adali et al. 1999, Alicigüzel et al. 2001, Cetinkaya et al. 2005, Erdamar et al. 2008, Erem et al. 2015) and in patients with hyperthyroidism compared to subclinical hyperthyroidism (Erem et al. 2015). Alicigüzel et al. (2001) found a significantly positive correlation between MDA and serum T3 and T4 concentrations. Treatment of (subclinical) hyperthyroidism resulted in significantly decreased plasma MDA concentrations, although normalization was not reached in every study (Adali et al. 1999, Erdamar et al. 2008, Erem et al. 2015). Two studies investigated both hyper- and hypothyroidism and showed increased MDA concentrations in hypothyroid patients compared to controls as well (Erdamar et al. 2008, Lassoued et al. 2010).
Ischemia-modified albumin (IMA) is a modified form of albumin due to oxidative damage and has first been reported as a marker for cardiac ischemia. However, IMA turned out to be elevated in non-cardiac diseases as well. A recent meta-analysis of 12 studies showed a significant increase in IMA concentrations in hyperthyroid patients compared to euthyroid controls. The same was observed in hypothyroid patients and no significant difference in IMA concentrations was found before and after treatment. IMA levels correlated negatively with T4, fT4 and fT3 concentrations in hypothyroid patients and positively with fT4 and fT3 concentrations in hyperthyroid patients (Reddy et al. 2017). Since both hyper- and hypothyroidism resulted in increased MDA and IMA concentrations, we can conclude that these biomarkers are not specific to act as a marker for tissue TH status. However, oxidative stress remains associated with hyperthyroidism, which is also demonstrated by (non-targeted) metabolome and proteome analysis, revealing several biomarkers indicating augmented defense against oxidative stress (Jourdan et al. 2014, Al-Majdoub et al. 2017, Pietzner et al. 2017, Lange et al. 2018). Concentrations of γ-glutamyl amino acids (GGAA) were positively correlated with fT4 concentrations, whereas phosphatidylcholines (PCs) and lysophospholipids (LPs) showed a negative correlation with fT4 concentrations. GGAA is part of the γ-glutamyl cycle (GGC), a pathway for metabolizing glutathione and plays a role in detoxification. TH is thought to stimulate the GGC in its antioxidant effect. Furthermore, research in rats showed that T3 induces the hepatic activation of a transcriptional factor (Nrf2) of the major cellular defense mechanism against oxidative stress (Romanque et al. 2011, Cornejo et al. 2013). Decreased PCs and LPs together with increased fT4 concentrations are most likely the result of enhanced lipid oxidation, which strengthens the relation of TH with oxidative stress. Although several biomarkers indicate that (defense against) oxidative stress seems to be related to TH, research is still too scarce to conclude these are all adequate biomarkers determining TH status.
Asymmetric dimethylarginine (ADMA)
Untargeted mass-spectrometry proteomics revealed a positive correlation of fT4 with asymmetric dimethylarginine (ADMA) (Pietzner et al. 2017). ADMA is produced as a by-product of systemic proteolysis, which is triggered by TH excess (Müller & Seitz 1984). It favors hypertension by directly impairing nitric oxide (NO)-dependent vasodilatation and, therefore, seems to increase the risk of cardiovascular diseases. This correlation with thyroid function is supported by previous literature (Hermenegildo et al. 2002, Arikan et al. 2007, Gu et al. 2011, Ittermann et al. 2016). However, ADMA concentrations were also elevated in well-treated patients with former hyperthyroidism and hypothyroidism, which diminishes its clinical application as a marker for TH status.
Complement proteins
The complement system is an important component of innate immunity. Complement proteins are predominantly produced by B-lymphocytes in response to infectious and inflammatory processes. However, also thyrocytes are able to produce complement proteins, which suggests a relation between complement proteins and thyroid hormones (Blanchin et al. 2003). Proteomic analyses revealed associations between thyroid state and several proteins of the complement system (Pietzner et al. 2017, Alfadda et al. 2018), namely complement factor H-related protein 5, factor H and complement factor C2, C3 and C4a. This association has been proposed in previous literature (Yu et al. 2006, Jafarzadeh et al. 2010) and specifically the inverse association between fT3 and fT4 with C3 was confirmed. However, other conditions (e.g. inflammation, multiple autoimmune diseases) increase the concentration of different complement proteins and the complement system compasses a broad range of proteins, which complicates the selection of proteins associated with TH. Therefore, more literature is needed to establish the value of complement proteins in determining TH status and they can currently not serve as feasible biomarkers evaluating TH status.
Amino acids
A variety of studies using (un)targeted metabolomics found multiple alterations in several amino acid concentrations depending on TH status. Tyrosine concentrations were elevated in hyperthyroid patients compared to euthyroid controls, as well as phenylalanine, a precursor of tyrosine (Chng et al. 2016, Piras et al. 2017). During the synthesis of TH, iodination of tyrosyl residues occurs, which might be a reason that increased TH production in hyperthyroid patients resulted in elevated concentrations of tyrosine and phenylalanine. Serine concentrations were decreased during hyperthyroidism and increased during hypothyroidism and remained increased after adequate treatment of hypothyroidism (Piras et al. 2017, 2021). The decrease in serine concentrations is thought to be explained by increased uptake of serine for gluconeogenesis purposes in hyperthyroid patients while hypothyroidism is characterized by a hypometabolic state explaining the increased concentrations of serine.
Metabolic fingerprint
Another approach to look at this massive amount of data is using profiles to distinguish between different conditions. Boumaza et al. (2019) used untargeted NMR-based metabolomics to discriminate mice with resistance to TRα (RTHα), due to mutations created in the THRA gene encoding for TRα, from hypothyroid mice and control mice. The study showed no specific biomarker to be discriminative for RTHα, hypothyroidism and euthyroidism in mice but the metabolic fingerprint of the RTHα mice was significantly different compared to control mice and ‘normal’ hypothyroid mice. In distinguishing RTHα and hypothyroid mice, especially differences in the lipid and cholesterol metabolism were detected. Therefore, using the whole metabolic profile, metabolomics might also be useful in humans to diagnose RTHα more easily and distinguish this rare disease from ‘normal’ hypothyroidism. It should be further elucidated how cut-off values for these metabolic fingerprints are shaped. However, the approach of discriminating between conditions upon profiles using OMICs has already been performed in humans and might be promising and worthwhile to investigate more extensively for this purpose (Friedrich et al. 2020).
Conclusion and recommendations
This review addressed the need for new biomarkers in determining tissue TH status. We summarized several biomarkers that, in the current literature, have been related to TH status in a variety of tissues and systems and can contribute to a diagnosis. Most of the biomarkers discussed are by themselves not considered suitable to serve as adequate biomarkers to indicate TH status. However, a few of them are promising (Fig. 2).
Overview of the promising biomarkers that have been related to TH status in the liver, intestine, bone and throughout the body. A full color version of this figure is available at https://doi.org/10.1530/JOE-21-0364.
Citation: Journal of Endocrinology 253, 2; 10.1530/JOE-21-0364
Derivatives of lipid and cholesterol metabolism could be used as biomarkers to establish TH status in the liver, although appropriate attention should be paid to conditions that influence these pathways. Glycerol is proposed as a biomarker to represent liver TH status in thyroid disorders. GPDH showed a strong correlation with TH status but cannot be measured in blood. Glycerol, which can be measured in blood, is a substrate of GPDH and could therefore be an attractive alternative (Piras et al. 2021). Recently, CD5L has been proposed as a TRβ-dependent biomarker for hepatic TH status although only in the hyperthyroid state (Nock et al. 2020). A few limitations of this marker have been discussed previously. Notably, inflammatory processes are suspected to influence this marker; it might therefore be interesting to assess whether CD5L could still function as a reflection of hepatic TH status in autoimmune thyroid diseases. SHBG can be used as a non-invasive hepatic biomarker to monitor the treatment outcome of TH therapy in patients with NASH (Harrison et al. 2019). Furthermore, osteocalcin and P1NP might be interesting indicators of bone TH status. Available research regarding IAP as a biomarker of intestinal TH status seems promising and the biomarker can be measured in serum via electrophoresis. Although not organ-specific, the Se/Cu ratio could serve as a biomarker to evaluate treatment effect in RTHα and RTHβ patients, since TSH is not always the correct indicator for a sufficient TH concentration in all tissues (Mittag et al. 2012). Several studies using metabolomics found a positive association between ACs and TH status in the hyperthyroid state making an AC profile or specific ACs of interest as an overall marker indicating tissue TH status (Jourdan et al. 2014, Chng et al. 2016, Al-Majdoub et al. 2017, Pietzner et al. 2017, Lange et al. 2018).
Biomarkers can also be useful in clinical practice to discriminate between thyroid disorders. CD5L might be useful in diagnosing patients with RTHβ, due to its TRβ-dependent action. SHBG has been proposed as a useful tool to discriminate between RTH and TSH-secreting adenoma (Sarne et al. 1988, Beck-Peccoz et al. 1990). Cu and Se were mentioned as predictors of TH status but have never really gained a foothold as official biomarkers. However, the Se/Cu ratio also seems promising in order to differentiate between RTHβ and TSH-secreting pituitary adenoma (Mittag et al. 2012). Another interesting approach was the use of a metabolic profile to distinguish between RTHα and hypothyroidism in the diagnostic process instead of a single biomarker, although this discriminative effect still has to be confirmed in humans (Boumaza et al. 2019).
The goal of this review was to assess potential tissue TH biomarkers in serum, since it is not feasible in humans to assess tissue TH status using a biopsy. In this review, most studies described the correlation between the proposed biomarker and TH status based on a hypo- or hyperthyroid state compared to a (treated) euthyroid state. Certainly, the response of the proposed biomarker upon TH fluctuation is important in exploring a possible correlation, although it does not necessarily prove that the biomarker indicates an altered tissue TH status in a patient who is biochemically euthyroid. Studies that investigate tissue TH biomarkers in patients before and after thyroidectomy due to thyroid carcinoma, and thus comparing a euthyroid state without and without L-T4, might be better to detect alterations at the tissue level when serum TH is considered euthyroid. Hitherto, the few studies that investigated this specific patient group confirmed the proposed correlation with cholesterol and bone biomarkers. Therefore, future research in thyroidectomized patients after thyroid carcinoma might be of interest to detect other promising biomarkers.
Currently, not only OMIC techniques are used to map tissue TH status, but also a TH action indicator mouse (THAI) model which is an interesting and alternative way to study the tissue TH status (Mohácsik et al. 2018). This animal model visualizes TH signaling in vivo via luciferase activity, allowing many different tissues to be examined under a variety of experimental conditions. Although this model is not relevant for daily clinical practice, it illustrates that there is a need for knowledge about tissue TH status in relation to systemic TH concentrations.
In summary, a variety of old and new biomarkers are available for establishing tissue TH status and using these for treatment and diagnostic purposes. The proposed biomarkers of interest all have in common that further exploration is needed to determine their value as a tissue-specific biomarker for TH status. Recently described and of potential interest are ACs. In the future, a thyroid panel combining several specific biomarkers fully capturing thyroid status in serum and most target tissues might be worthwhile investigating. To achieve that goal, more research is first needed to validate the aforementioned biomarkers of interest, followed by the composition of such a thyroid panel using these biomarkers.
Declaration of interest
Anita Boelen is a Senior Edior on the JOE/JME joint editorial board. Anita Boelen was not involved in the review or editorial process for this paper, on which she is listed as an author. The other authors have nothing to disclose.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
References
Abdelwhab S & Foda O 2010 Relation between thyroid hormones and insulin resistance in hemodialysis patients. Kidney 19 66–72. (https://doi.org/10.1007/s00596-009-0126-x)
Adali M, İnal-erden M, Akalin A & Efe Bn 1999 Effects of propylthiouracil, propranolol, and vitamin E on lipid peroxidation and antioxidant status in hyperthyroid patients. Clinical Biochemistry 32 363–367. (https://doi.org/10.1016/s0009-9120(99)00024-7)
Adams AC, Astapova I, Fisher FM, Badman MK, Kurgansky KE, Flier JS, Hollenberg AN & Maratos-Flier E 2010 Thyroid hormone regulates hepatic expression of fibroblast growth factor 21 in aPPARα-dependent manner. Journal of Biological Chemistry 285 14078–14082. (https://doi.org/10.1074/jbc.C110.107375)
Aizawa T, Hiramatsu K, Ohtsuka H, Kobayashi M, Koizumi Y, Miyamoto T, Niwa A & Yamada T 1986 An elevation of BUN/creatinine ratio in patients with hyperthyroidism. Hormone and Metabolic Research 18 771–774. (https://doi.org/10.1055/s-2007-1012432)
Akalin A, Colak O, Alatas O & Efe B 2002 Bone remodelling markers and serum cytokines in patients with hyperthyroidism. Clinical Endocrinology 57 125–129. (https://doi.org/10.1046/j.1365-2265.2002.01578.x)
Alfadda AA, Benabdelkamel H, Masood A, Jammah AA & Ekhzaimy AA 2018 Differences in the plasma proteome of patients with hypothyroidism before and after thyroid hormone replacement: a proteomic analysis. International Journal of Molecular Sciences 19 88. (https://doi.org/10.3390/ijms19010088)
Alicigüzel Y, Özdem SN, Özdem SS, Karayalçin U, Siedlak SL, Perry G & Smith MA 2001 Erythrocyte, plasma, and serum antioxidant activities in untreated toxic multinodular goiter patients. Free Radical Biology and Medicine 30 665–670. (https://doi.org/10.1016/s0891-5849(00)00509-8)
Al-Majdoub M, Lantz M & Spégel P 2017 Treatment of Swedish patients with graves’ hyperthyroidism is associated with changes in acylcarnitine levels. Thyroid 27 1109–1117. (https://doi.org/10.1089/thy.2017.0218)
Altınova AE, Törüner FB, Aktürk M, Bukan N, Çakır N, Ayvaz G & Arslan M 2006 Adiponectin levels and cardiovascular risk factors in hypothyroidism and hyperthyroidism. Clinical Endocrinology 65 530–535. (https://doi.org/10.1111/j.1365-2265.2006.02628.x)
Anderson DC 1974 Sex-hormone binding globulin. Clinical Endocrinology 3 69–96. (https://doi.org/10.1111/j.1365-2265.1974.tb03298.x)
Angstwurm MW, Schopohl J & Gaertner R 2004 Selenium substitution has no direct effect on thyroid hormone metabolism in critically ill patients. European Journal of Endocrinology 151 47–54. (https://doi.org/10.1530/eje.0.1510047)
Arem R, Escalante DA, Arem N, Morrisett JD & Patsch W 1995 Effect of L-thyroxine therapy on lipoprotein fractions in overt and subclinical hypothyroidism, with special reference to lipoprotein(a). Metabolism: Clinical and Experimental 44 1559–1563. (https://doi.org/10.1016/0026-0495(95)90075-6)
Arikan E, Karadag CH & Guldiken S 2007 Asymmetric dimethylarginine levels in thyroid diseases. Journal of Endocrinological Investigation 30 186–191. (https://doi.org/10.1007/BF03347423)
Bassett JH & Williams GR 2003 The molecular actions of thyroid hormone in bone. Trends in Endocrinology and Metabolism 14 356–364. (https://doi.org/10.1016/s1043-2760(03)00144-9)
Becerra A, Bellido D, Luengo A, Piédrola G & De Luis DA 1999 Lipoprotein(a) and other lipoproteins in hypothyroid patients before and after thyroid replacement therapy. Clinical Nutrition 18 319–322. (https://doi.org/10.1016/s0261-5614(98)80031-9)
Beck-Peccoz P, Roncoroni R, Mariotti S, Medri G, Marcocci C, Brabant G, Forloni F, Pinchera A & Faglia G 1990 Sex hormone-binding globulin measurement in patients with inappropriate secretion of thyrotropin (Ist) – evidence against selective pituitary thyroid-hormone resistance in nonneoplastic Ist. Journal of Clinical Endocrinology and Metabolism 71 19–25. (https://doi.org/10.1210/jcem-71-1-19)
Beck-Peccoz P, Lania A, Beckers A, Chatterjee K & Wemeau JL 2013 2013 European Thyroid Association guidelines for the diagnosis and treatment of thyrotropin-secreting pituitary tumors. European Thyroid Journal 2 76–82. (https://doi.org/10.1159/000351007)
Bergmann P, Dediste A, Demeester-Mirkine N, Deconinck I & Corvilain J 1989 Serum bone Gla protein (BGP) in primary hypothyroidism before and during treatment with thyroid hormones. Hormone and Metabolic Research 21 47–48. (https://doi.org/10.1055/s-2007-1009146)
Beyer IW, Karmali R, DeMeester-Mirkine N, Cogan E & Fuss MJ 1998 Serum creatine kinase levels in overt and subclinical hypothyroidism. Thyroid 8 1029–1031. (https://doi.org/10.1089/thy.1998.8.1029)
Bianco A & Kim B 2013 Nondeiodinative metabolism of iodothyronines. In Werner and Ingbar’s the Thyroid: A Fundamental and Clinical Text. Eds Bravermen LE, Cooper DS. Lippincott Williams & Wilkins.
Bilir B, Soysal-Atile N, Ekiz Bilir B, Yilmaz I, Bali I, Altintas N, Baykiz D, Aydin M & Guldiken S 2016 Evaluation of SCUBE-1 and sCD40L biomarkers in patients with hypothyroidism due to Hashimoto’s thyroiditis: a single-blind, controlled clinical study. European Review for Medical and Pharmacological Sciences 20 407–413.(PMID: https://pubmed.ncbi.nlm.nih.gov/26914113/)
Bin Obead M, Al Badr N, AlSedairy S, Albrahim T, Alhussain M, Aljurbua T, Arzoo S & Alqahtani W 2021 Association between thyroid hormone levels with insulin resistance in obese euthyroid women with metabolic syndrome in Saudi Arabia: a cross-sectional study. Progress in Nutrition 23 e2021088.(https://doi.org/10.23751/pn.v23i1.10825)
Blanchin S, Estienne V, Durand-Gorde JM, Carayon P & Ruf J 2003 Complement activation by direct C4 binding to thyroperoxidase in Hashimoto’s thyroiditis. Endocrinology 144 5422–5429. (https://doi.org/10.1210/en.2003-0918)
Blasig S, Kühnen P, Schuette A, Blankenstein O, Mittag J & Schomburg L 2016 Positive correlation of thyroid hormones and serum copper in children with congenital hypothyroidism. Journal of Trace Element in Medicine and Biology 37 90–95. (https://doi.org/10.1016/j.jtemb.2016.05.007)
Boelen A, Kwakkel J & Fliers E 2011 Beyond low plasma T3: local thyroid hormone metabolism during inflammation and infection. Endocrine Reviews 32 670–693. (https://doi.org/10.1210/er.2011-0007)
Bonde Y, Breuer O, Lütjohann D, Sjöberg S, Angelin B & Rudling M 2014 Thyroid hormone reduces PCSK9 and stimulates bile acid synthesis in humans. Journal of Lipid Research 55 2408–2415. (https://doi.org/10.1194/jlr.M051664)
Boruah P, Baruah AJ, Hajong R, Nath CK, Barman B, Chutia H & Sarma K 2016 A study to assess the validity of estimation of serum Ostase level in hyperthyroid and hypothyroid cases. Journal of Clinical and Diagnostic Research 10 BC08–BC11. (https://doi.org/10.7860/JCDR/2016/19750.8441)
Boumaza H, Markossian S, Busi B, Rautureau GJP, Gauthier K, Elena-Herrmann B & Flamant F 2019 Metabolomic profiling of body fluids in mouse models demonstrates that nuclear magnetic resonance is a putative diagnostic tool for the presence of thyroid hormone receptor α1 mutations. Thyroid 29 1327–1335. (https://doi.org/10.1089/thy.2018.0730)
Brenta G, Schnitman M, Gurfinkiel M, Damilano S, Pierini A, Sinay I & Pisarev MA 1999 Variations of sex hormone-binding globulin in thyroid dysfunction. Thyroid 9 273–277. (https://doi.org/10.1089/thy.1999.9.273)
Burggraaf J, Lalezari S, Emeis JJ, Vischer UM, de Meyer PH, Pijl H & Cohen AF 2001 Endothelial function in patients with hyperthyroidism before and after treatment with propranolol and thiamazol. Thyroid 11 153–160. (https://doi.org/10.1089/105072501300042820)
Burnett JR, Crooke MJ, Delahunt JW & Feek CM 1994 Serum enzymes in hypothyroidism. New Zealand Medical Journal 107 355–356.(PMID: https://pubmed.ncbi.nlm.nih.gov/8078622/)
Cachefo A, Boucher P, Vidon C, Dusserre E, Diraison F & Beylot M 2001 Hepatic lipogenesis and cholesterol synthesis in hyperthyroid patients. Journal of Clinical Endocrinology and Metabolism 86 5353–5357. (https://doi.org/10.1210/jcem.86.11.7981)
Capelo LP, Beber EH, Fonseca TL & Gouveia CHA 2009 The monocarboxylate transporter 8 and L-type amino acid transporters 1 and 2 are expressed in mouse skeletons and in osteoblastic MC3T3-E1 cells. Thyroid 19 171–180. (https://doi.org/10.1089/thy.2008.0120)
Cetinkaya A, Kurutas EB, Buyukbese MA, Kantarceken B & Bulbuloglu E 2005 Levels of malondialdehyde and superoxide dismutase in subclinical hyperthyroidism. Mediators of Inflammation 2005 57–59. (https://doi.org/10.1155/MI.2005.57)
Cheng SY, Leonard JL & Davis PJ 2010 Molecular aspects of thyroid hormone actions. Endocrine Reviews 31 139–170. (https://doi.org/10.1210/er.2009-0007)
Cheng Y, Sun Z, Zhong D, Ren L & Yang H 2020 Cross-sectional investigation of serum creatine kinase concentration in Graves disease patients treated with oral antithyroid drugs. Medicine 99 e22898. (https://doi.org/10.1097/MD.0000000000022898)
Chng CL, Lim AY, Tan HC, Kovalik JP, Tham KW, Bee YM, Lim W, Acharyya S, Lai OF & Chong MF et al.2016 Physiological and metabolic changes during the transition from hyperthyroidism to euthyroidism in Graves’ disease. Thyroid 26 1422–1430. (https://doi.org/10.1089/thy.2015.0602)
Chou KM, Chiu SY-H, Chen CH, Yang NI, Huang BY & Sun CY 2011 Correlation of clinical changes with regard to thyroxine replacement therapy in hypothyroid patients – focusing on the change of renal function. Kidney and Blood Pressure Research 34 365–372. (https://doi.org/10.1159/000328324)
Christ-Crain M, Morgenthaler NG, Meier C, Müller C, Nussbaumer C, Bergmann A, Staub JJ & Müller B 2005 Pro-A-type and N-terminal pro-B-type natriuretic peptides in different thyroid function states. Swiss Medical Weekly 135 549–554. (https://doi.org/2005/37/smw-11119)
Cornejo P, Vargas R & Videla LA 2013 Nrf2-regulated phase-II detoxification enzymesand phase-III transporters are induced by thyroidhormone in rat liver. BioFactors 39 514–521. (https://doi.org/10.1002/biof.1094)
Dalton R, Savidge G, Matthews K, Dewar M, Kernoff P, Greaves M & Preston F 1987 Hypothyroidism as a cause of acquired von Willebrand’s disease. Lancet 329 1007–1009.(https://doi.org/10.1016/S0140-6736(87)92272-0)
Dasarathy Y, Stevens J, Lanzillo JJ & Fanburg BL 1990 Elevation of angiotensin converting enzyme in bovine endothelial cells quantitated by an elisa. Life Sciences 47 883–889. (https://doi.org/10.1016/0024-3205(90)90602-n)
Davis PJ, Mousa SA & Lin HY 2021 Nongenomic actions of thyroid hormone: the integrin component. Physiological Reviews 101 319–352. (https://doi.org/10.1152/physrev.00038.2019)
De Pergola G, Ciampolillo A, Paolotti S, Trerotoli P & Giorgino R 2007 Free triiodothyronine and thyroid stimulating hormone are directly associated with waist circumference, independently of insulin resistance, metabolic parameters and blood pressure in overweight and obese women. Clinical Endocrinology 67 265–269. (https://doi.org/10.1111/j.1365-2265.2007.02874.x)
Dhanwal DK 2011 Thyroid disorders and bone mineral metabolism. Indian Journal of Endocrinology and Metabolism 15 S107–S112. (https://doi.org/10.4103/2230-8210.83339)
Dietrich JB, Kuchler-Bopp S, Boutillier S, Ittel ME, Reeber A, Zaepfel M, Delaunoy JP & Vincendon G 1997 Expression of thyroid hormone receptors alpha and beta-1 messenger RNAs in human endothelial cells. The T3 hormone stimulates the synthesis of the messenger RNA of the intercellular adhesion molecule-1. Cellular and Molecular Biology 43 1205–1212.(PMID: https://pubmed.ncbi.nlm.nih.gov/9489946/)
Domouzoglou EM, Fisher FM, Astapova I, Fox EC, Kharitonenkov A, Flier JS, Hollenberg AN & Maratos-Flier E 2014 Fibroblast growth factor 21 and thyroid hormone show mutual regulatory dependency but have independent actions in vivo. Endocrinology 155 2031–2040. (https://doi.org/10.1210/en.2013-1902)
Dumitrescu AM & Refetoff S 2013 The syndromes of reduced sensitivity to thyroid hormone. Biochimica et Biophysica Acta 1830 3987–4003. (https://doi.org/10.1016/j.bbagen.2012.08.005)
Dumoulin SC, Perret BP, Bennet AP & Caron PJ 1995 Opposite effects of thyroid hormones on binding proteins for steroid hormones (sex hormone-binding globulin and corticosteroid-binding globulin) in humans. European Journal of Endocrinology 132 594–598. (https://doi.org/10.1530/eje.0.1320594)
Ebert EC 2010 The thyroid and the gut. Journal of Clinical Gastroenterology 44 402–406. (https://doi.org/10.1097/MCG.0b013e3181d6bc3e)
Elbers LP, Moran C, Gerdes VE, van Zaane B, Meijers J, Endert E, Lyons G, Chatterjee VK, Bisschop PH & Fliers E 2016 The hypercoagulable state in hyperthyroidism is mediated via the thyroid hormone β receptor pathway. European Journal of Endocrinology 174 755–762. (https://doi.org/10.1530/EJE-15-1249)
Elbers LPB, Fliers E & Cannegieter SC 2018 The influence of thyroid function on the coagulation systemand its clinical consequences. Journal of Thrombosis and Haemostasis 16 634–645. (https://doi.org/10.1111/jth.13970)
Engelmann B, Bischof J, Dirk AL, Friedrich N, Hammer E, Thiele T, Führer D, Homuth G, Brabant G & Völker U 2015 Effect of experimental thyrotoxicosis onto blood coagulation: a proteomics study. European Thyroid Journal 4 119–124. (https://doi.org/10.1159/000381769)
Engler H, Oettli RE & Riesen WF 1999 Biochemical markers of bone turnover in patients with thyroid dysfunctions and in euthyroid controls: a cross-sectional study. Clinica Chimica Acta: International Journal of Clinical Chemistry 289 159–172. (https://doi.org/10.1016/s0009-8981(99)00173-4)
Erdamar H, Demirci H, Yaman H, Erbil MK, Yakar T, Sancak B, Elbeg S, Biberoğlu G & Yetkin I 2008 The effect of hypothyroidism, hyperthyroidism, and their treatment on parameters of oxidative stress and antioxidant status. Clinical Chemistry and Laboratory Medicine 46 1004–1010. (https://doi.org/10.1515/CCLM.2008.183)
Erem C 2011 Thyroid disorders and hypercoagulability. Seminars in Thrombosis and Hemostasis 37 17–26. (https://doi.org/10.1055/s-0030-1270067)
Erem C, Değer O, Bostan M, Orem A, Sönmez M, Ulusoy S & Telatar M 1999 Plasma lipoprotein (a) concentrations in hypothyroid, euthyroid and hyperthyroid subjects. Acta Cardiologica 54 77–81.(PMID: https://pubmed.ncbi.nlm.nih.gov/10378018/)
Erem C, Suleyman AK, Civan N, Mentese A, Nuhoglu İ, Uzun A, Ersoz HO & Deger O 2015 Ischemia-modified albümin and malondialdehyde levels in patients with overt and subclinical hyperthyroidism: effects of treatment on oxidative stress. Endocrine Journal 62 493–501. (https://doi.org/10.1507/endocrj.EJ14-0542)
Erem C, Civan N, Coskun H, Mentese A, Suleyman AK, Altay DU, Akgul Z & Deger O 2016 Signal peptide-CUB-EGF domain-containing protein 1 (SCUBE1) levels in patients with overt and subclinical hyperthyroidism: effects of treatment. Clinical Endocrinology 84 919–924. (https://doi.org/10.1111/cen.12955)
Ertugrul DT, Gursoy A, Sahin M, Unal AD, Pamuk B, Berberoglu Z, Ayturk S, Tutuncu NB & Demirag NG 2008 Evaluation of brain natriuretic peptide levels in hyperthyroidism and hypothyroidism. Journal of the National Medical Association 100 401–405. (https://doi.org/10.1016/s0027-9684(15)31272-4)
Ertugrul DT, Yavuz B, Ata N, Yalcin AA, Kucukazman M, Algul B, Dal K, Akin KO, Deveci OS & Canbolat N et al.2009 Decreasing brain natriuretic peptide levels after treatment for hyperthyroidism. Endocrine Journal 56 1043–1048. (https://doi.org/10.1507/endocrj.k09e-159)
Escobar-Morreale HF, Obregón MJ, Escobar del Rey F & Morreale de Escobar G 1995 Replacement therapy for hypothyroidism with thyroxine alone does not ensure euthyroidism in all tissues, as studied in thyroidectomized rats. Journal of Clinical Investigation 96 2828–2838. (https://doi.org/10.1172/JCI118353)
Fischli S, von Wyl V, Trummler M, Konrad D, Wueest S, Ruefer A, Heering K, Streuli R, Steuer C & Bernasconi L et al.2017 Iron metabolism in patients with Graves’ hyperthyroidism. Clinical Endocrinology 87 609–616. (https://doi.org/10.1111/cen.13450)
Flamant F, Cheng SY, Hollenberg AN, Moeller LC, Samarut J, Wondisford FE, Yen PM & Refetoff S 2017 Thyroid hormone signaling pathways: time for a more precise nomenclature. Endocrinology 158 2052–2057. (https://doi.org/10.1210/en.2017-00250)
Flores-Morales A, Gullberg H, Fernandez L, Ståhlberg N, Lee NH, Vennström B & Norstedt G 2002 Patterns of liver gene expression governed by TRβ. Molecular Endocrinology 16 1257–1268. (https://doi.org/10.1210/mend.16.6.0846)
Flynn RW, Bonellie SR, Jung RT, MacDonald TM, Morris AD & Leese GP 2010 Serum thyroid-stimulating hormone concentration and morbidity from cardiovascular disease and fractures in patients on long-term thyroxine therapy. Journal of Clinical Endocrinology and Metabolism 95 186–193. (https://doi.org/10.1210/jc.2009-1625)
Ford HC, Lim WC, Chisnall WN & Pearce JM 1989 Renal function and electrolyte levels in hyperthyroidism: urinary protein excretion and the plasma concentrations of urea, creatinine, uric acid, hydrogen ion and electrolytes. Clinical Endocrinology 30 293–301. (https://doi.org/10.1111/j.1365-2265.1989.tb02238.x)
Franchini M 2004 Hemostasis and thyroid diseases revisited. Journal of Endocrinological Investigation 27 886–892. (https://doi.org/10.1007/BF03346287)
Friedrich N, Pietzner M, Engelmann B, Homuth G, Führer D, Brabant G, Wallaschofski H & Völker U 2020 Screening for new markers to assess thyroid hormone action by OMICs analysis of human samples. Experimental and Clinical Endocrinology and Diabetes 128 479–487. (https://doi.org/10.1055/a-1144-2636)
Friesema ECH, Kuiper GGJM, Jansen J, Visser TJ & Kester MHA 2006 Thyroid hormone transport by the human monocarboxylate transporter 8 and its rate-limiting role in intracellular metabolism. Molecular Endocrinology 20 2761–2772. (https://doi.org/10.1210/me.2005-0256)
Friesema ECH, Jansen J, Jachtenberg JW, Visser WE, Kester MHA & Visser TJ 2008 Effective cellular uptake and efflux of thyroid hormone by human monocarboxylate transporter 10. Molecular Endocrinology 22 1357–1369. (https://doi.org/10.1210/me.2007-0112)
Friesema ECH, Visser WE & Visser TJ 2010 Genetics and phenomics of thyroid hormone transport by MCT8. Molecular and Cellular Endocrinology 322 107–113. (https://doi.org/10.1016/j.mce.2010.01.016)
Furuya F, Shimura H, Yamashita S, Endo T & Kobayashi T 2010 Liganded thyroid hormone receptor-α enhances proliferation of pancreatic β-cells. Journal of Biological Chemistry 285 24477–24486. (https://doi.org/10.1074/jbc.M109.100222)
Garmendia Madariaga A, Santos Palacios S, Guillén-Grima F & Galofré JC 2014 The incidence and prevalence of thyroid dysfunction in Europe: a meta-analysis. Journal of Clinical Endocrinology and Metabolism 99 923–931. (https://doi.org/10.1210/jc.2013-2409)
Garnero P, Vassy V, Bertholin A, Riou JP & Delmas PD 1994 Markers of bone turnover in hyperthyroidism and the effects of treatment. Journal of Clinical Endocrinology and Metabolism 78 955–959. (https://doi.org/10.1210/jcem.78.4.8157727)
Gärtner R 2009 Selenium and thyroid hormone axis in critical ill states: an overview of conflicting view points. Journal of Trace Element in Medicine and Biology 23 71–74. (https://doi.org/10.1016/j.jtemb.2009.01.001)
Gong DW, Bi S, Weintraub BD & Reitman M 1998 Rat mitochondrial glycerol-3-phosphate dehydrogenase gene: multiple promoters, high levels in brown adipose tissue, and tissue-specific regulation by thyroid hormone. DNA and Cell Biology 17 301–309. (https://doi.org/10.1089/dna.1998.17.301)
Gouveia CH, Schultz JJ, Bianco AC & Brent GA 2001 Thyroid hormone stimulation of osteocalcin gene expression in ROS 17/2·8 cells is mediated by transcriptional and post-transcriptional mechanisms. Journal of Endocrinology 170 667–675. (https://doi.org/10.1677/joe.0.1700667)
Graninger W, Pririch KR, Speiser W, Deutsch E & Waldhäusl WK 1986 Effect of thyroid hormones on plasma protein concentrations in man. Journal of Clinical Endocrinology and Metabolism 63 407–411. (https://doi.org/10.1210/jcem-63-2-407)
Gray J, Chattopadhyay D, Beale GS, Patman GL, Miele L, King BP, Stewart S, Hudson M, Day CP & Manas DM et al.2009 A proteomic strategy to identify novel serum biomarkers for liver cirrhosis and hepatocellular cancer in individuals with fatty liver disease. BMC Cancer 9 271. (https://doi.org/10.1186/1471-2407-9-271)
Gu LQ, Zhao L, Zhu W, Li FY, Zhang MJ, Liu Y, Liu JM, Ning G & Zhao YJ 2011 Relationships between serum levels of thyroid hormones and serum concentrations of asymmetric dimethylarginine (ADMA) and N-terminal-pro-B-type natriuretic peptide (NT-proBNP) in patients with Graves’ disease. Endocrine 39 266–271. (https://doi.org/10.1007/s12020-011-9436-7)
Halleen JM, Alatalo SL, Suominen H, Cheng S, Janckila AJ & Väänänen HK 2000 Tartrate-resistant acid phosphatase 5b: a novel serum marker of bone resorption. Journal of Bone and Mineral Research 15 1337–1345. (https://doi.org/10.1359/jbmr.2000.15.7.1337)
Hampl R, Kancheva R, Hill M, BicÏíková M & Vondra K 2003 Interpretation of sex hormone-binding globulin levels in thyroid disorders. Thyroid 13 755–760. (https://doi.org/10.1089/105072503768499644)
Harrison SA, Bashir MR, Guy CD, Zhou R, Moylan CA, Frias JP, Alkhouri N, Bansal MB, Baum S & Neuschwander-Tetri BA et al.2019 Resmetirom (MGL-3196) for the treatment of non-alcoholic steatohepatitis: a multicentre, randomised, double-blind, placebo-controlled, phase 2 trial. Lancet 394 2012–2024. (https://doi.org/10.1016/S0140-6736(19)32517-6)
Hayashi H, Mizushima N, Yoshinaga H, Kawamitsu H, Matsuda S, Tanoue M, Kurosawa H, Fujimoto H & Amakawa T 1996 The relationship between lipoprotein(a) and low density lipoprotein receptors during the treatment of hyperthyroidism. Hormone and Metabolic Research 28 384–387. (https://doi.org/10.1055/s-2007-979821)
Heijboer AC, Savelkoul E, Kruit A, Endert E & Blankenstein MA 2016 Inaccurate first-generation testosterone assays are influenced by sex hormone–binding globulin concentrations. Journal of Applied Laboratory Medicine 1 194–201. (https://doi.org/10.1373/jalm.2016.020065)
Hekimsoy Z & Kavalali Oktem IK 2005 Serum creatine kinase levels in overt and subclinical hypothyroidism. Endocrine Research 31 171–175. (https://doi.org/10.1080/07435800500371706)
Hellman NE & Gitlin JD 2002 Ceruloplasmin metabolism and function. Annual Review of Nutrition 22 439–458. (https://doi.org/10.1146/annurev.nutr.22.012502.114457)
Hermenegildo C, Medina P, Peiró M, Segarra G, Vila JM, Ortega J & Lluch S 2002 Plasma concentration of asymmetric dimethylarginine, an endogenous inhibitor of nitric oxide synthase, is elevated in hyperthyroid patients. Journal of Clinical Endocrinology and Metabolism 87 5636–5640. (https://doi.org/10.1210/jc.2002-020905)
Hodin RA, Chamberlain SM & Upton MP 1992 Thyroid hormone differentially regulates rat intestinal brush border enzyme gene expression. Gastroenterology 103 1529–1536. (https://doi.org/10.1016/0016-5085(92)91174-3)
Hodin RA, Shei A, Morin M & Meng S 1996 Thyroid hormone and the gut: selective transcriptional activation of a villus-enterocyte marker. Surgery 120 138–143. (https://doi.org/10.1016/s0039-6060(96)80280-7)
Hönes GS, Rakov H, Logan J, Liao XH, Werbenko E, Pollard AS, Præstholm SM, Siersbæk MS, Rijntjes E & Gassen J et al.2017 Noncanonical thyroid hormone signaling mediates cardiometabolic effects in vivo. PNAS 114 E11323–E11332. (https://doi.org/10.1073/pnas.1706801115)
Hong K, Linton JA, Kim S, Song Y & Shin H 2021 Analysis of thyroid hormones and sarcopenia: association of creatine kinase and thyroid hormones (thyroid-stimulating hormone, triiodothyronine, and free thyroxine): a cross-sectional study. JCSM Clinical Reports 6 36–52. (https://doi.org/10.1002/crt2.29)
Hoppichler F, Sandholzer C, Moncayo R, Utermann G & Kraft HG 1995 Thyroid hormone (fr4) reduces lipoprotein(a) plasma levels. Atherosclerosis 115 65–71. (https://doi.org/10.1016/0021-9150(94)05500-i)
Isaia GC, Roggia C, Gola D, Stefano MD, Gallone G, Aimo G, Ardissone P & Mussetta M 2000 Bone turnover in hyperthyroidism before and after thyrostatic management. Journal of Endocrinological Investigation 23 727–731. (https://doi.org/10.1007/BF03345061)
Ito M, Arishima T, Kudo T, Nishihara E, Ohye H, Kubota S, Fukata S, Amino N, Kuma K & Sasaki I et al.2007 Effect of levo-thyroxine replacement on non-highdensity lipoprotein cholesterol in hypothyroid patients. Journal of Clinical Endocrinology and Metabolism 92 608–611. (https://doi.org/10.1210/jc.2006-1605)
Ito M, Miyauchi A, Hisakado M, Yoshioka W, Ide A, Kudo T, Nishihara E, Kihara M, Ito Y & Kobayashi K et al.2017 Biochemical markers reflecting thyroid function in athyreotic patients on levothyroxine monotherapy. Thyroid 27 484–490. (https://doi.org/10.1089/thy.2016.0426)
Ittermann T, Bahls M, Atzler D, Friedrich N, Schwedhelm E, Böger RH, Felix SB, Völzke H & Dörr M 2016 L-arginine derivatives are associated with the hyperthyroid state in the general population. Thyroid 26 212–218. (https://doi.org/10.1089/thy.2015.0385)
Jafarzadeh A, Poorgholami M, Izadi N, Nemati M & Rezayati M 2010 Immunological and hematological changes in patients with hyperthyroidism or hypothyroidism. Clinical and Investigative Medicine 33 E271–E279. (https://doi.org/10.25011/cim.v33i5.14352)
Jain RB 2014 Thyroid function and serum copper, selenium, and zinc in general U.S. population. Biological Trace Element Research 159 87–98. (https://doi.org/10.1007/s12011-014-9992-9)
Jensen T, Feldt-Rasmussen B, Bjerre-Knudsen J & Deckert T 1989 Features of endothelial dysfunction in early diabetic nephropathy. Lancet 333 461–463.(https://doi.org/10.1016/S0140-6736(89)91365-2)
Jódar E, Muñoz-Torres M, Escobar-Jiménez F, Quesada M, Luna JD & Olea N 1997 Antiresorptive therapy in hyperthyroid patients: longitudinal changes in bone and mineral metabolism. Journal of Clinical Endocrinology and Metabolism 82 1989–1994. (https://doi.org/10.1210/jcem.82.6.4026)
Jourdan C, Linseisen J, Meisinger C, Petersen AK, Gieger C, Rawal R, Illig T, Heier M, Peters A & Wallaschofski H et al.2014 Associations between thyroid hormones and serum metabolite profiles in an euthyroid population. Metabolomics 10 152–164. (https://doi.org/10.1007/s11306-013-0563-4)
Kapadia KB, Bhatt PA & Shah JS 2012 Association between altered thyroid state and insulin resistance. Journal of Pharmacology and Pharmacotherapeutics 3 156–160. (https://doi.org/10.4103/0976-500X.95517)
Karsenty G & Olson EN 2016 Bone and muscle endocrine functions: unexpected paradigms of inter-organ communication. Cell 164 1248–1256. (https://doi.org/10.1016/j.cell.2016.02.043)
Kazukauskiene N, Podlipskyte A, Varoneckas G & Mickuviene N 2021 Insulin resistance in association with thyroid function, psychoemotional state, and cardiovascular risk factors. International Journal of Environmental Research and Public Health 18 3388. (https://doi.org/10.3390/ijerph18073388)
Khaleeli AA & Edwards RH 1984 Effect of treatment on skeletal muscle dysfunction in hypothyroidism. Clinical Science 66 63–68. (https://doi.org/10.1042/cs0660063)
Kien CL, Bunn JY, Stevens R, Bain J, Ikayeva O, Crain K, Koves TR & Muoio DM 2014 Dietary intake of palmitate and oleate has broad impact on systemic and tissue lipid profiles in humans. American Journal of Clinical Nutrition 99 436–445. (https://doi.org/10.3945/ajcn.113.070557)
Kisakol G, Kaya A, Gonen S & Tunc R 2003 Bone and calcium metabolism in subclinical autoimmune hyperthyroidism and hypothyroidism. Endocrine Journal 50 657–661. (https://doi.org/10.1507/endocrj.50.657)
Klausen IC, Nielsen FE, Hegedüs L, Gerdes LU, Charles P & Faergeman O 1992 Treatment of hypothyroidism reduces low-density lipoproteins but not lipoprotein(a). Metabolism: Clinical and Experimental 41 911–914. (https://doi.org/10.1016/0026-0495(92)90176-b)
Klein I, Mantell P, Parker M & Levey GS 1980 Resolution of abnormal muscle enzyme studies in hypothyroidism. American Journal of the Medical Sciences 279 159–162. (https://doi.org/10.1097/00000441-198005000-00004)
Knezevic J, Starchl C, Tmava Berisha A & Amrein K 2020 Thyroid-gut-axis: how does the microbiota influence thyroid function? Nutrients 12 1769. (https://doi.org/10.3390/nu12061769)
Kobe N, Takamatsu J, Ito M, Sakane S & Ohsawa N 1999 Acute and early effects of triiodothyronine administration on serum markers of bone and mineral metabolism. Endocrine 10 53–56. (https://doi.org/10.1385/ENDO:10:1:53)
Köhrle J 2000 The deiodinase family: selenoenzymes regulating thyroid hormone availability and action. Cellular and Molecular Life Sciences 57 1853–1863. (https://doi.org/10.1007/PL00000667)
Kojima N, Sakata S, Nakamura S, Nagai K, Takuno H, Ogawa T, Matsui I, Sarui H & Miura K 1992 Serum concentrations of osteocalcin in patients with hyperthyroidism, hypothyroidism and subacute thyroiditis. Journal of Endocrinological Investigation 15 491–496. (https://doi.org/10.1007/BF03348786)
Kralik A, Eder K & Kirchgessner M 1996 Influence of zinc and selenium deficiency on parameters relating to thyroid hormone metabolism. Hormone and Metabolic Research 28 223–226. (https://doi.org/10.1055/s-2007-979169)
Ku EJ, Yoo WS, Lee EK, Ahn HY, Woo SH, Hong JH, Chung HK & Park JW 2021 Effect of TSH suppression therapy on bone mineral density in differentiated thyroid cancer: a systematic review and meta-analysis. Journal of Clinical Endocrinology and Metabolism 106 3655–3667. (https://doi.org/10.1210/clinem/dgab539)
Kubota K, Tamura J, Kurabayashi H, Shirakura T & Kobayashi I 1993 Evaluation of increased serum ferritin levels in patients with hyperthyroidism. Clinical Investigator 72 26–29. (https://doi.org/10.1007/BF00231112)
Kucharzewski M, Braziewicz J, Majewska U & Góźdź S 2003 Copper, zinc, and selenium in whole blood and thyroid tissue of people with various thyroid diseases. Biological Trace Element Research 93 9–18. (https://doi.org/10.1385/BTER:93:1-3:9)
Kung AW, Pang RW, Lauder I, Lam KS & Janus ED 1995 Changes in serum lipoprotein(a) and lipids during treatment of hyperthyroidism. Clinical Chemistry 41 226–231. (https://doi.org/10.1093/clinchem/41.2.226)
Kužma M, Vaňuga P, Binkley N, Ságová I, Pávai D, Blažíček P, Kužmová Z, Jackuliak P, Vaňuga A & Killinger Z et al.2018 High serum fractalkine is associated with lower trabecular bone score in premenopausal women with Graves’ disease. Hormone and Metabolic Research 50 609–614. (https://doi.org/10.1055/a-0633-2814)
Lafayette RA, Costa ME & King AJ 1994 Increased serum creatinine in the absence of renal failure in profound hypothyroidism. American Journal of Medicine 96 298–299. (https://doi.org/10.1016/0002-9343(94)90158-9)
Lakatos P, Foldes J, Horvath C, Kiss L, Tatrai A, Takacs I, Tarjan G & Stern PH 1997 Serum interleukin-6 and bone metabolism in patients with thyroid function disorders. Journal of Clinical Endocrinology and Metabolism 82 78–81. (https://doi.org/10.1210/jcem.82.1.3641)
Lange T, Budde K, Homuth G, Kastenmüller G, Artati A, Krumsiek J, Völzke H, Adamski J, Petersmann A & Völker U et al.2018 Comprehensive metabolic profiling reveals a lipid-rich fingerprint of free thyroxine far beyond classic parameters. Journal of Clinical Endocrinology and Metabolism 103 2050–2060. (https://doi.org/10.1210/jc.2018-00183)
Lassoued S, Mseddi M, Mnif F, Abid M, Guermazi F, Masmoudi H, Feki AE & Attia H 2010 A comparative study of the oxidative profile in Graves’ disease, Hashimoto’s thyroiditis, and papillary thyroid cancer. Biological Trace Element Research 138 107–115. (https://doi.org/10.1007/s12011-010-8625-1)
Lee YP & Lardy HA 1965 Influence of thyroid hormones on L-a-glycerophosphate dehydrogenases and other dehydrogenases in various organs of the rat. Journal of Biological Chemistry 240 1427–1436. (https://doi.org/10.1016/S0021-9258(18)97593-9)
Lee YP, Takemori AE & Lardy H 1959 Enhanced oxidation of α-glycerophosphate by mitochondria of thyroid-fed rats. Journal of Biological Chemistry 234 3051–3054. (https://doi.org/10.1016/S0021-9258(18)69723-6)
Lee DS, Chung JK, Cho BY, Koh CS & Lee M 1986 Changes of serum angiotensin-converting enzyme activity during treatment of patients with Graves’ disease. Korean Journal of Internal Medicine 1 104–112. (https://doi.org/10.3904/kjim.1986.1.1.104)
Lee MS, Kim SY, Lee MC, Cho BY, Lee HK, Koh CS & Min HK 1990 Negative correlation between the change in bone mineral density and serum osteocalcin in patients with hyperthyroidism. Journal of Clinical Endocrinology and Metabolism 70 766–770. (https://doi.org/10.1210/jcem-70-3-766)
Lee Y, Park YJ, Ahn HY, Lim JA, Park KU, Choi SH, Park DJ, Oh BC, Jang HC & Yi KH 2013 Plasma FGF21 levels are increased in patients with hypothyroidism independently of lipid profile. Endocrine Journal 60 977–983. (https://doi.org/10.1507/endocrj.ej12-0427)
Lee YK, Lee H, Han S, Jung H, Shin DY, Nam KH, Chung WY & Lee EJ 2019 Association between thyroid-stimulating hormone level after total thyroidectomy and hypercholesterolemia in female patients with differentiated thyroid cancer: a retrospective study. Journal of Clinical Medicine 8 1106. (https://doi.org/10.3390/jcm8081106)
Leedman PJ, Stein AR, Chin WW & Rogers JT 1996 Thyroid hormone modulates the interaction between iron regulatory proteins and the ferritin mRNA iron-responsive element. Journal of Biological Chemistry 271 12017–12023. (https://doi.org/10.1074/jbc.271.20.12017)
Li Y, Chen H, Tan J, Wang X, Liang H & Sun X 1998 Impaired release of tissue plasminogen activator from the endothelium in Graves’ disease – indicator of endothelial dysfunction and reduced fibrinolytic capacity. European Journal of Clinical Investigation 28 1050–1054. (https://doi.org/10.1046/j.1365-2362.1998.00381.x)
Li X, Misik AJ, Rieder CV, Solaro RJ, Lowen A & Fliegel L 2002 Thyroid hormone receptor α1 regulates expression of the Na+/H+ exchanger (NHE1). Journal of Biological Chemistry 277 28656–28662. (https://doi.org/10.1074/jbc.M203221200)
Li H, Zhang J & Jia W 2013 Fibroblast growth factor 21: a novel metabolic regulator from pharmacology to physiology. Frontiers of Medicine 7 25–30. (https://doi.org/10.1007/s11684-013-0244-8)
Liang F, Webb P, Marimuthu A, Zhang S & Gardner DG 2003 Triiodothyronine increases brain natriuretic peptide (BNP) gene transcription and amplifies endothelin-dependent BNP gene transcription and hypertrophy in neonatal rat ventricular myocytes. Journal of Biological Chemistry 278 15073–15083. (https://doi.org/10.1074/jbc.M207593200)
Loviselli A, Mastinu R, Rizzolo E, Massa GM, Velluzzi F, Sammartano L, Mela Q & Mariotti S 1997 Circulating telopeptide type I is a peripheral marker of thyroid hormone action in hyperthyroidism and during levothyroxine suppressive therapy. Thyroid 7 561–566. (https://doi.org/10.1089/thy.1997.7.561)
Loviselli A, Rizzolo E, Mastinu R, Velluzzi F, Secci G, Taberlet A & Mariotti S 2003 High serum circulating telopeptide type I in multinodular goiter. Hormone and Metabolic Research 35 377–381. (https://doi.org/10.1055/s-2003-41361)
Malo MS, Zhang W, Alkhoury F, Pushpakaran P, Abedrapo MA, Mozumder M, Fleming E, Siddique A, Henderson JW & Hodin RA 2004 Thyroid hormone positively regulates the enterocyte differentiation marker intestinal alkaline phosphatase gene via an atypical response element. Molecular Endocrinology 18 1941–1962. (https://doi.org/10.1210/me.2003-0351)
Martínez-Triguero ML, Hernández-Mijares A, Nguyen TT, Muñoz ML, Peña H, Morillas C, Lorente D, Lluch I & Molina E 1998 Effect of thyroid hormone replacement on lipoprotein(a), lipids, and apolipoproteins in subjects with hypothyroidism. Mayo Clinic Proceedings 73 837–841. (https://doi.org/10.4065/73.9.837)
Mat Ali MH, Tuan Ismail TS, Wan Azman WN, Yaacob NM, Yahaya N, Draman N, Wan Mohamed WMI, Abdullah MS, Ibrahim HA & Wan Nik WNFH et al.2020 Comparison of vitamin D levels, bone metabolic marker levels, and bone mineral density among patients with thyroid disease: a cross-sectional study. Diagnostics 10 1075. (https://doi.org/10.3390/diagnostics10121075)
Minisola S, Dionisi S, Pacitti MT, Paglia F, Carnevale V, Scillitani A, Mazzaferro S, De GS, Pepe J & D’Erasmo E et al.2002 Gender differences in serum markers of bone resorption in healthy subjects and patients with disorders affecting bone. Osteoporosis International 13 171–175. (https://doi.org/10.1007/s001980200009)
Mittag J, Behrends T, Hoefig CS, Vennström B & Schomburg L 2010 Thyroid hormones regulate selenoprotein expression and selenium status in mice. PLoS ONE 5 e12931. (https://doi.org/10.1371/journal.pone.0012931)
Mittag J, Behrends T, Nordstrom K, Anselmo J, Vennström B & Schomburg L 2012 Serum copper as a novel biomarker for resistance to thyroid hormone. Biochemical Journal 443 103–109. (https://doi.org/10.1042/BJ20111817)
Miyakawa M, Tsushima T & Demura H 1996 Carboxy-terminal propeptide of type 1 procollagen (P1CP) and carboxy-terminal telopeptide of type 1 collagen (1CTP) as sensitive markers of bone metabolism in thyroid disease. Endocrine Journal 43 701–708. (https://doi.org/10.1507/endocrj.43.701)
Mohácsik P, Erdélyi F, Baranyi M, Botz B, Szabó G, Tóth M, Haltrich I, Helyes Z, Sperlágh B & Tóth Z et al.2018 A transgenic mouse model for detection of tissue-specific thyroid hormone action. Endocrinology 159 1159–1171. (https://doi.org/10.1210/en.2017-00582)
Moran C, Schoenmakers N, Agostini M, Schoenmakers E, Offiah A, Kydd A, Kahaly G, Mohr-Kahaly S, Rajanayagam O & Lyons G et al.2013 An adult female with resistance to thyroid hormone mediated by defective thyroid hormone receptor α. Journal of Clinical Endocrinology and Metabolism 98 4254–4261. (https://doi.org/10.1210/jc.2013-2215)
Mosekilde L & Christensen MS 1977 Decreased parathyroid function in hyperthyroidism: interrelationships between serum parathyroid hormone, calcium-phosphorus metabolism and thyroid function. Acta Endocrinologica 84 566–575. (https://doi.org/10.1530/acta.0.0840566)
Müller MJ & Seitz HJ 1984 Thyroid hormone action on intermediary metabolism. Part III. Protein metabolism in hyper- and hypothyroidism. Klinische Wochenschrift 62 97–102. (https://doi.org/10.1007/BF01738699)
Mullur R, Liu YY & Brent GA 2014 Thyroid hormone regulation of metabolism. Physiological Reviews 94 355–382. (https://doi.org/10.1152/physrev.00030.2013)
Muls E, Blaton V, Rosseneu M, Lesaffre E, Lamberigts G & de Moor P 1982 Serum lipids and apolipoproteins A-I, A-II, and B in hyperthyroidism before and after treatment. Journal of Clinical Endocrinology and Metabolism 55 459–464. (https://doi.org/10.1210/jcem-55-3-459)
Mundy GR, Shapiro JL, Bandelin JG, Canalis EM & Raisz LG 1976 Direct stimulation of bone resorption by thyroid hormones. Journal of Clinical Investigation 58 529–534. (https://doi.org/10.1172/JCI108497)
Nagasaka S, Sugimoto H, Nakamura T, Kusaka I, Fujisawa G, Sakuma N, Tsuboi Y, Fukuda S, Honda K & Okada K et al.1997 Antithyroid therapy improves bony manifestations and bone metabolic markers in patients with Graves’ thyrotoxicosis. Clinical Endocrinology 47 215–221. (https://doi.org/10.1046/j.1365-2265.1997.2401045.x)
Nakahama H, Sakaguchi K, Horita Y, Sasaki O, Nakamura S, Inenaga T & Takishita S 2001 Treatment of severe hypothyroidism reduced serum creatinine levels in two chronic renal failure patients. Nephron 88 264–267. (https://doi.org/10.1159/000046000)
Nakayama M, Yamada K, Yamamoto Y, Yokoyama K, Nakano H, Kubo H, Shigematsu T, Kawaguchi Y & Sakai O 1994 Vascular endothelial dysfunction in patients on regular dialysis treatment. Clinical Nephrology 42 117–120.(PMID: https://pubmed.ncbi.nlm.nih.gov/7955573/)
Newgard CB 2017 Metabolomics and metabolic diseases: where do we stand? Cell Metabolism 25 43–56. (https://doi.org/10.1016/j.cmet.2016.09.018)
Nishitani H, Okamura K, Noguchi S, Inoue K, Morotomi Y & Fujishima M 1990 Serum lipid levels in thyroid dysfunction with special reference to transient elevation during treatment in hyperthyroid Graves’ disease. Hormone and Metabolic Research 22 490–493. (https://doi.org/10.1055/s-2007-1004953)
Nock S, Johann K, Harder L, Wirth EK, Renko K, Hoefig CS, Kracke V, Hackler J, Engelmann B & Rauner M et al.2020 CD5L constitutes a novel biomarker for integrated hepatic thyroid hormone action. Thyroid 30 908–923. (https://doi.org/10.1089/thy.2019.0635)