This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hennemann, G. Articles by Visser, T. J. Search for Related Content PubMed PubMed Citation Articles by Hennemann, G. Articles by Visser, T. J.

Plasma Membrane Transport of Thyroid Hormones and Its Role in Thyroid Hormone Metabolism and Bioavailability

Department of Nuclear Medicine (G.H., M.d.J., E.P.K.) and Department of Internal Medicine (R.D., E.C.H.F., T.J.V.), Erasmus University Medical Center, 3015 GD Rotterdam, The Netherlands

Correspondence: Address all correspondence and requests for reprints to: Georg Hennemann, M.D., Ph.D., Vijverweg 32, 3062 JP Rotterdam, The Netherlands. E-mail: g{at}hennemann.demon.nl

	Abstract

Top Abstract I. Historical Introduction II. Binding of Thyroid... III. Transport of Thyroid... IV. Cellular Efflux of... V. Transport of Thyroid... VI. In Vivo Plasma... VII. Plasma Membrane Transport... VIII. Requirements for a... IX. Identification of Thyroid... X. Summary and Conclusions References

 
Although it was originally believed that thyroid hormones enter target cells by passive diffusion, it is now clear that cellular uptake is effected by carrier-mediated processes. Two stereospecific binding sites for each T₄ and T₃ have been detected in cell membranes and on intact cells from humans and other species. The apparent Michaelis-Menten values of the high-affinity, low-capacity binding sites for T₄ and T₃ are in the nanomolar range, whereas the apparent Michaelis- Menten values of the low-affinity, high-capacity binding sites are usually in the lower micromolar range. Cellular uptake of T₄ and T₃ by the high-affinity sites is energy, temperature, and often Na⁺ dependent and represents the translocation of thyroid hormone over the plasma membrane. Uptake by the low-affinity sites is not dependent on energy, temperature, and Na⁺ and represents binding of thyroid hormone to proteins associated with the plasma membrane. In rat erythrocytes and hepatocytes, T₃ plasma membrane carriers have been tentatively identified as proteins with apparent molecular masses of 52 and 55 kDa. In different cells, such as rat erythrocytes, pituitary cells, astrocytes, and mouse neuroblastoma cells, uptake of T₄ and T₃ appears to be mediated largely by system L or T amino acid transporters. Efflux of T₃ from different cell types is saturable, but saturable efflux of T₄ has not yet been demonstrated. Saturable uptake of T₄ and T₃ in the brain occurs both via the blood-brain barrier and the choroid plexus-cerebrospinal fluid barrier. Thyroid hormone uptake in the intact rat and human liver is ATP dependent and rate limiting for subsequent iodothyronine metabolism. In starvation and nonthyroidal illness in man, T₄ uptake in the liver is decreased, resulting in lowered plasma T₃ production. Inhibition of liver T₄ uptake in these conditions is explained by liver ATP depletion and increased concentrations of circulating inhibitors, such as 3-carboxy-4-methyl-5-propyl-2-furanpropanoic acid, indoxyl sulfate, nonesterified fatty acids, and bilirubin. Recently, several organic anion transporters and L type amino acid transporters have been shown to facilitate plasma membrane transport of thyroid hormone. Future research should be directed to elucidate which of these and possible other transporters are of physiological significance, and how they are regulated at the molecular level.

I. Historical Introduction

II. Binding of Thyroid Hormones to Isolated Plasma Membranes

A. %Binding kinetics

B. Analysis of binding protein(s)

III. Transport of Thyroid Hormones into Isolated Cells

A. Transport into hepatocytes

B. Transport into other cell types

C. Interactions of various compounds with thyroid hormone transport

IV. Cellular Efflux of Thyroid Hormones

V. Transport of Thyroid Hormone into Isolated Organs

A. Transport into the liver

B. Transport into other organs

VI. In Vivo Plasma Membrane Transport of Thyroid Hormones in Animals

A. Brain

B. Other organs

VII. Plasma Membrane Transport in Humans

A. Introduction

B. In starvation

C. In nonthyroidal illness

VIII. Requirements for a Regulatory Role of Plasma Membrane Transport in the Bioavailability of Thyroid Hormone

A. Specificity of plasma membrane transport

B. Absence of significant diffusion

C. Plasma membrane transport is subject to regulation

D. Transport is rate limiting for subsequent metabolism

IX. Identification of Thyroid Hormone Transporters

A. Organic anion transporters

B. Amino acid transporters

X. Summary and Conclusions

	I. Historical Introduction

Top Abstract I. Historical Introduction II. Binding of Thyroid... III. Transport of Thyroid... IV. Cellular Efflux of... V. Transport of Thyroid... VI. In Vivo Plasma... VII. Plasma Membrane Transport... VIII. Requirements for a... IX. Identification of Thyroid... X. Summary and Conclusions References

 
EARLY REPORTS ON uptake of thyroid hormones by cells and tissues of different species appeared in the early 1950s. For about two and a half decades it was assumed that the translocation of thyroid hormones over the plasma membrane of target cells was a process of simple diffusion. This assumption was based on the fact that thyroid hormones are lipophilic and, as the plasma membrane is constituted of a lipid bilayer, there seemed apparently no need to assume any other mechanism of translocation than that of diffusion. The belief in this concept was so strong that hardly any studies testing this assumption were performed in this period of time. The studies that were performed on thyroid hormone uptake by cells and tissues were predominantly directed at investigating the influence of temperature, pH, and extracellular thyroid hormone-binding proteins on the kinetics of this process. In the interpretation of the results of these studies, it was often taken for granted that thyroid hormones diffuse into the cells and that the driving force of this process is the concentration of the free hormone. This so-called "free hormone hypothesis" was formulated in 1960 by Robbins and Rall (1). They stated "that the free or diffusible thyroid hormone concentration in blood and extracellular tissues would determine the rate at which thyroid hormone is distributed to its loci of action and the rates at which it is degraded and excreted." As we will see in the following sections, this assumption is only partially correct. Plasma membrane translocation is a regulated process that is rate limiting for subsequent intracellular accumulation, action, and fate of the hormone. However, we will also see that, at least in vitro, the rate of uptake of thyroid hormones into the cell is determined not only by the efficacy of this plasma membrane translocation process but also by variations in the free hormone concentration in physiological and pathophysiological conditions. In vivo the situation is more complicated in that circulating inhibitors of thyroid hormone tissue uptake may be operative as well.

It is remarkable that, to the best of our knowledge, the first publication on thyroid hormone transport points to an energy-dependent uptake process (2). In this report, transport of T₃ into ascites carcinoma cells was inhibited by KCN, a metabolic blocker that suppresses ATP formation, indicating that energy is involved in the uptake mechanism. The authors of this study concluded that "this amino acid does not escape the cellular concentration process to which all other amino acids so far studied are subjected." This report apparently escaped attention and was "rediscovered" by Sorimachi and Robbins in 1978 (3).

In a review in 1957 (4), Robbins and Rall proposed that thyroid hormone action is a function of the free hormone in the blood. However, in view of the extremely low concentration of unbound T₄ in blood, they suggested that tissues are extraordinarily sensitive to thyroid hormone, or that T₄ has to be concentrated in target cells. This latter suggestion leaves open the possibility of an active transport process. On the basis of their studies using tissue slices at different incubation temperatures and metabolic activities, Freinkel et al. (5) concluded that the establishment of concentration differentials for T₄ between tissue slices and suspending media constitutes an equilibrium-binding phenomenon rather than an active transport. Hogness et al. (6) suggested that the higher concentration of T₄ and T₃ in rat diaphragm as compared with that in the incubation media was evidence for a true chemical binding. They did not consider the possibility of energy-dependent transport against a concentration gradient. Two groups of investigators, Beraud et al. (7) and Ingbar and Freinkel (8), were of the opinion that extra- and intracellular thyroid hormone binding-proteins govern transmembrane transfer of free diffusible hormone. In their studies of the uptake of T₄ and T₃ by rat diaphragm, Lein and Dowben (9) assumed that the kinetics of uptake they observed were based on diffusion into the tissue and subsequent binding of hormone to intracellular proteins. In his review on distribution and metabolism of thyroid hormone, Tata (10) suggested that the plasma membrane did not play an active role in the movement of free hormone from the vascular to the tissue compartments. Hillier (11) published a series of studies related to uptake and release of T₄ and T₃ in different organs. To our knowledge, he was the first to assess saturability of these processes. Studying the perfused rat heart, saturation of these processes could not be detected using free hormone concentrations ranging from 13 pM to 1.3 µM. As we will see below (Sections II and III), the highest concentration used is sufficient to saturate the high-affinity component of the uptake process detected in rat hepatocytes and many other cell types, although discrepancies have been described. One of the reasons why any saturation of the uptake mechanism might have escaped detection is that the conditions under which the studies were performed were not optimal to maintain intracellular ATP concentrations. This means that any energy-dependent, carrier-mediated process might have become undetectable. This possibility is in line with another observation from the same study (11), that thyroid hormone uptake was independent of changes in incubation temperature. In a follow-up study (12), Hillier concluded that extracellular thyroid hormone binding-proteins are an important factor determining the total amount of hormone taken up by the rat heart. Studying uptake and release of T₄ and T₃ in rat liver under similar "ATP-poor" conditions and using hormone concentrations up to 0.13 µM, he arrived at similar conclusions, in that uptake and release were temperature independent and that uptake was importantly influenced by extracellular hormone-binding sites (13). The assumption that thyroid hormones easily penetrate plasma membranes was strengthened by Hillier’s next studies (14) using liposomes prepared from egg-yolk lecithin. He reported that these membranes were readily permeable to T₄ and that the binding of both T₄ and T₃ to liposomes and to rat heart tissue is similarly dependent on pH.

In summary, until 1970 it was generally believed that thyroid hormones enter target cells by simple diffusion. This assumption was based on the fact that thyroid hormones are lipophilic and could therefore easily traverse the lipid-rich bilayer of the cell membrane. Transport of thyroid hormones into cells was envisaged to be mainly regulated by binding forces of extra- and intracellular thyroid hormone-binding proteins, directing the free moiety of thyroid hormone passively through the plasma membrane.

	II. Binding of Thyroid Hormones to Isolated Cell Membranes

Top Abstract I. Historical Introduction II. Binding of Thyroid... III. Transport of Thyroid... IV. Cellular Efflux of... V. Transport of Thyroid... VI. In Vivo Plasma... VII. Plasma Membrane Transport... VIII. Requirements for a... IX. Identification of Thyroid... X. Summary and Conclusions References

 
A. Binding kinetics
The earliest studies analyzing specificity of binding of thyroid hormones to plasma membranes of target cells were reported in 1975 by Tata (15) and in 1976 by Singh et al. (16). Although detecting saturability of binding of thyroid hormones to different cellular constituents, including plasma membranes, Tata questioned the biological relevance of these binding sites (15). Singh and his group studied inhibition of binding of T₃ and T₄ to intact hemoglobin-free erythrocyte membranes by thyroid hormone analogs (16). Specificity of binding was demonstrated for both T₄ and T₃ by structure-dependent inhibition by the analogs. The major finding of this study was that the avidity of erythrocyte membranes was greater for T₃ analogs than for T₄ analogs but was similar for L-T₃ and L-T₄.

Several reports concerned binding of thyroid hormones to plasma membranes of rat hepatocytes (17, 18, 19, 20). Pliam and Goldfine (17) reported on two binding sites for L-T₃, one with high affinity and low capacity and one with low affinity and high capacity. Mean apparent dissociation constant (K_d) values were 3.2 nM and 220 nM, respectively (Table 1). Similar values were found by others (18), who also reported on high- and low-affinity binding sites for L-T₄, with mean apparent K_d values of 0.57 nM and 23.8 nM, respectively, distinct from the T₃ binding sites (Table 1). Specific T₄ binding was inhibited by thiol-blocking agents and by proteases. L-T₄ was bound with high specificity regarding iodine substituents and alanine side chain modifications (20). Studies of L-rT₃ binding to rat hepatocyte membranes also revealed two binding sites, the high-affinity site being different from that of L-T₄ (21).

Binding of thyroid hormones to plasma membranes of other cell types and species was also reported. High-affinity binding sites for T₃ and T₄ in plasma membranes of rat kidney and testis were characterized by apparent K_d values in the low nanomolar range, whereas those of the low-affinity binding sites were in the high nanomolar range (Table 1). Specific binding sites for L-T₃ and L-T₄ could not be detected in rat spleen (18). In plasma membranes of human placenta, two specific L-T₃ binding sites were found with apparent K_d values of 2.0 nM and 18.5 µM (30). D-T₃, L-rT₃, L-T₄, and D-T₄ were less effective in displacing L-T₃ from both binding sites. In plasma membranes of a mouse neuroblastoma cell line, L-T₃ binding sites showed apparent K_d values of 8.4 nM and 7.3 µM, with lower affinity of both sites for D-T₃ (31).

B. Analysis of binding protein(s)
A series of publications by Cheng and co-workers (30, 32, 33, 34, 35) concerned the identification of T₃ and/or T₄-binding membrane proteins in different cell types by affinity-labeling techniques. In their experiments using human placenta (30), GH3 cells (32, 33), mouse Swiss 3T3 fibroblasts (33), and human A431 epitheloid carcinoma cells (33), the proteins were envisaged to be associated with the plasma membrane and to have a molecular mass between 55 (32, 33) and 65 kDa (30). Peptide mapping of the proteins labeled with N-bromoacetyl-[¹²⁵I]T₃ (BrAc[¹²⁵I]T₃) or BrAc[¹²⁵I]T₄ showed very similar patterns (33), indicating that the same protein was probably involved. Later immunocytochemical studies, using four different monoclonal antibodies against the 55-kDa thyroid hormone-binding protein, showed that this protein was loosely associated with the endoplasmic reticulum and nuclear envelope, although some association with the plasma membrane could not be excluded (34). In a later study by Kato et al. (35), this protein was shown to be identical to protein disulfide isomerase (PDI). This finding was confirmed by Horiouchi et al. (36), who detected both T₃-binding and PDI activity in a 55-kDa protein isolated from a plasma membrane-enriched beef liver fraction. Although some PDI may indeed be associated with the plasma membranes, most of this enzyme is located in the lumen of the endoplasmic reticulum (37). In contrast to the high reactivity of PDI toward BrAcT₃ and BrAcT₄, it shows only low affinity for underivatized T₃ and T₄ (38). Since, moreover, PDI is not an integral membrane protein (37, 38), it seems unlikely to be involved directly in plasma membrane transport of thyroid hormone.

Photoaffinity labeling of erythrocyte membranes with L-T₃ has identified a protein with an apparent molecular mass of 55 kDa (39). T₃ binding to this protein was critically dependent on the presence of phospholipids. Tryptophan but not leucine or D-T₃ competed with the L-T₃ binding site, indicating stereospecificity and a possible relationship with the amino acid transport system T (39). Using a monoclonal antibody that specifically inhibited uptake of T₃ in rat hepatocytes, a putative carrier protein was detected with an apparent molecular mass of 52 kDa (40). Affinity labeling of mouse neuroblastoma plasma membranes with BrAc[¹²⁵I]T₃ has detected a 27-kDa protein (31). Since the size of this protein is identical to that of the type I iodothyronine deiodinase, which is also readily labeled with BrAcT₃ (38), it is unlikely to be related to a thyroid hormone transporter.

In summary, the first studies showing specific binding of thyroid hormones to isolated cell membranes appeared in the mid-1970s. Most extensively studied were cell membranes from human and rat erythrocytes and rat hepatocytes. For each T₃ and T₄, two stereospecific binding sites were detected in these membranes; one with apparent K_d values in the lower nanomolar range, and the other in the (sub)micromolar range. Specific binding for both hormones was dependent on the reduced state of protein-SH groups. T₃-binding proteins have been identified in rat erythrocyte and hepatocyte membranes with apparent molecular masses of 55 and 52 kDa.

	III. Transport of Thyroid Hormones into Isolated Cells

Top Abstract I. Historical Introduction II. Binding of Thyroid... III. Transport of Thyroid... IV. Cellular Efflux of... V. Transport of Thyroid... VI. In Vivo Plasma... VII. Plasma Membrane Transport... VIII. Requirements for a... IX. Identification of Thyroid... X. Summary and Conclusions References

 
The first evidence, to our knowledge, that transport of thyroid hormones into intact cells is not a passive, but an energy-dependent, process was reported by Christensen et al. in 1954 (Ref. 2 ; see also Section I) but unfortunately temporarily escaped attention. It was not until 1976 that Rao et al. (41) and our laboratory (42, 43) in 1978 independently published the saturable and energy-dependent transport of T₃ and T₄ into rat hepatocytes. Since then a whole series of reports from different laboratories have confirmed carrier-mediated, mostly energy- and Na⁺-dependent transport of iodothyronines into a variety of cells from different species.

A. Transport into hepatocytes
In Table 2 the kinetics of thyroid hormone uptake by hepatocytes are summarized. In most studies two saturable processes have been discerned: a high-affinity, low-capacity and a low-affinity, high-capacity process (41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55). In the majority of the studies, the apparent K_m values of the high-affinity systems for T₄, T₃, or rT₃ uptake are in the nanomolar range (42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55). This process is thought to represent the translocation process across the plasma membrane as it is energy and temperature dependent (41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55). Studies testing the possible Na⁺ dependence of the high-affinity uptake of iodothyronines have produced controversial results in rats (44, 45, 46, 47, 50), confirmatory results in human hepatocytes (52), and negative results in trout hepatocytes (54, 55). The energy-, temperature-, and Na⁺-independent, low-affinity uptake process may represent binding of thyroid hormone to cell surface-associated proteins (45). T₄ and T₃ mutually inhibit their high-affinity uptake processes in rat hepatocytes, but kinetic analysis of these inhibitions indicates that T₃ and T₄ cross the plasma membrane by different pathways (47, 55). This finding was confirmed by others who found differences in the dependence of the T₃ and T₄ transport systems on the cell phase of the rat hepatocyte and on sodium butyrate stimulation (56). Preliminary results in rat hepatocytes suggest that rT₃ shares the same transport system with T₄ (48), but kinetic studies of plasma iodothyronine clearance in humans suggest different plasma-to-liver transfer mechanisms for rT₃ and T₄ (57), in line with different binding sites for rT₃ and T₄ in (rat) liver plasma membrane (21). In addition to the metabolic condition of hepatocytes in culture, in particular with regard to ATP concentration, the free T₄ concentration in the medium is also a determinant for the amount of hormone that is taken up by the cell and subsequently metabolized (58). Stereospecificity of T₃ and T₄ uptake has been demonstrated in rat and trout liver cells (51, 54, 55).

2. T₄ transport. T₄ transport into intact cells has been less well studied than T₃ transport (Table 3). The most probable explanation for this, at least in liver cells, is the greater requirement of an optimal energy charge of the cells under study for transport of T₄ than for uptake of T₃. This is explained by the much steeper slope of the relationship between cellular ATP concentration and the rate of T₄ (and rT₃) transport in hepatocytes than that of the relationship between ATP and T₃ transport (Fig. 1) (46). Even a small decrease in cellular ATP concentration results in a major reduction in T₄ (and rT₃) transport but only slightly affects T₃ uptake. This may also be the reason why some authors could not observe specific, energy-dependent transport of T₄ in liver cells (44, 92). Others (93) did find saturable but energy-independent uptake not only of T₄ but also of T₃ in rat hepatocytes under far from optimal cellular ATP conditions. In other cell types, such as erythrocytes, rat neonatal cardiac myocytes, rat brain cells, pituitary cells, and fibroblasts, some laboratories observed that, in contrast to T₃, T₄ was apparently taken up by diffusion only or not at all, whereas other laboratories did find (stereo)specific, mostly energy-dependent T₄ uptake in the same cell types (Table 3). It is not known whether these discrepancies are related to the different energy requirements of the T₄ and T₃ transport processes as mentioned above or due to other factors such as the use of different techniques.

View larger version (20K): [in this window] [in a new window]   Figure 1. Uptake of T3 (•), rT3 (), and T4 () vs. ATP concentration in rat hepatocytes preincubated with different concentrations of glucose or fructose. [Reproduced with permission from E.P. Krenning et al.: FEBS Lett 140:229–233, 1982 (48 ).]

a. Erythrocytes.
In rat erythrocytes, the aromatic amino acids tryptophan (Trp), phenylalanine (Phe), and tyrosine (Tyr) competitively inhibited T₃ transport, while transport of Trp was similarly inhibited by T₃, D-T₃, T₄, and thyronine (T₀) (94). N-ethylmaleimide (NEM) irreversibly inhibited Trp and T₃ transport, and both ligands protected each others transport from inactivation by this compound. These data indicated common or closely linked transport systems for T₃ and for aromatic amino acids, i.e., the system T amino acid transporter, at least in erythrocytes (94). Similar results were obtained for binding of T₃ and Trp to rat erythrocyte membranes (28). Further studies suggested a common carrier for T₃ and Trp, which also facilitates countertransport such that the uphill transport of T₃ is driven by heteroexchange with intracellular aromatic amino acids (95). Evidence for uptake of T₃ by the system T amino acid transporter or a closely linked transporter was also obtained using human and trout erythrocytes (62, 67). No such relationship was found between T₄ and system T amino acid transport in trout erythrocytes (67).

b. Other cell types.
In rat hepatocyte sinusoidal membrane vesicles, Trp transport occurs via a NEM-resistant (system T) and a NEM-sensitive (system L) pathway, and T₃ and T₄ mainly inhibit Trp transport via system T (96). The inhibitory activity of T₃ and T₄ is dependent on the thyroid status of the donor rat, i.e., decreasing in the order hyperthyroid > euthyroid > hypothyroid. T₃ and T₄ share the same stereospecific uptake carrier in the rat pituitary (68, 69), and the potent inhibition of T₃ and T₄ uptake by leucine (Leu) suggests the involvement of amino acid transport system L (70). This system was also found to participate in T₃ and T₄ transport in mouse neuroblastoma cells (74) and in T₃ transport in rat astrocytes (97). In Ehrlich ascites cells, the neutral amino acids Phe, -aminoisobutyric acid, and cycloleucine did not compete with transport of T₄, indicating that the system A, L, and ASC amino acid pathways were not involved (98). In rat hepatocytes, participation of the amino acid transport system A in uptake of T₃ and T₄ was ruled out (51, 99). A weak interaction was found between uptake of system L and T amino acids and uptake of T₃ in human JAR choriocarcinoma cells (100).

2. Drugs and other chemicals. As shown in Table 4, a variety of compounds has been demonstrated to inhibit thyroid hormone uptake in different cells. Despite their widely different properties, the inhibitory activity of most of these substances is suggested to be based on competition because of structural similarity with thyroid hormone (19, 48, 51, 53, 101, 103, 104, 105, 106, 107, 111, 112). The antiarrhythmic drug amiodarone is also known to inhibit binding of T₃ to its nuclear receptors on the basis of structural similarity (114). The concentration of amiodarone shown to inhibit uptake of thyroid hormone in rat hepatocytes was 1 µM, which is similar to therapeutical serum levels in humans (114). However, since in serum, amiodarone is primarily bound to albumin that circulates at a concentration of 4% but was used in the hepatocyte incubations at a concentration of 1%, the free amiodarone concentrations obtained in vitro may be higher than in treated humans. Nevertheless, in vivo kinetic data in patients treated with amiodarone also show decreased net tissue uptake of thyroid hormone (115). This decrease can be explained by inhibition of thyroid hormone transport into tissues and/or by inhibition of thyroid hormone binding to intracellular proteins. Cholecystographic agents usually reach serum concentrations between 100 and 700 µM in humans (116) and were tested in vitro (at lower albumin levels) at concentrations between 10 and 100 µM (48). These agents not only inhibit thyroid hormone transport into rat hepatocytes, supposedly on the basis of molecular structural similarity (19, 48), but also displace T₄ from the human liver in vivo (117). The non-bile acid cholephils, sulfobromophthalein, bilirubin, and indocyanine green, also inhibit thyroid hormone transport and binding in rat hepatocytes on the basis of structural similarity (19, 51). Diphenylhydantoin, the nonsteroidal antiinflammatory phenylanthranilic acids, flufenamic acid, meclofenamic acid, and mefenamic acid, and the structurally related compounds, 2,3-dimethyldiphenylamine and diclofenac, all competitively inhibit rat hepatocyte and pituitary uptake of thyroid hormone (51, 107, 111, 113). Analysis of the structure-activity relationship for inhibition of T₃ uptake in rat hepatocytes by the phenylanthranilic acids demonstrated that inhibitory potency was highly dependent on the hydrophobicity of the inhibitor (107). Phloretin, a glucose transporter inhibitor that is structurally related to thyroid hormones, competitively inhibited T₃ uptake into human HepG2 hepatocarcinoma cells (53). Many of the inhibitors of thyroid hormone uptake discussed here also interact competitively with thyroid hormone-binding sites on serum proteins and nuclear T₃ receptors (51, 107, 113, 118). Amiodarone, cholecystographic agents, and bilirubin have been shown to interact with deiodinases (114, 119). The benzodiazepine drugs do not interact with nuclear T₃-binding sites, but inhibit T₃ uptake in different cell types from human and rat origin (Table 4) by competing for the T₃ carrier without being transported themselves (104). The structure-activity relationships were studied for inhibition of T₃ uptake in HepG2 cells by benzodiazepine and thyromimetic compounds. The results of these studies, along with computer-assisted molecular modeling techniques, predicted a "tilted crossbow" conformation of the inhibitor for interaction with the iodothyronine transporter (105).

Little information is available about stimulatory factors of thyroid hormone uptake in vitro. The histamine H1 receptor antagonist, telemastine, and phenobarbital enhance the specific, energy-dependent uptake of T₄ in rat hepatocytes but not in hepatocytes from guinea pig or beagle dog (120). The exact mechanism of this induction in rat hepatocytes is unknown but appears to be a primary effect on the plasma membrane transport system. Telemastine did not influence T₃ uptake in rat hepatocytes, underscoring the functional difference in the uptake systems of T₃ and T₄ in the liver (121).

In summary, transport of T₄ and T₃ has been studied extensively in human, rat, and trout hepatocytes. For both T₄ and T₃, high-affinity, low-capacity and low-affinity, high-capacity uptake processes have been identified. The high-affinity processes have apparent K_m values in the nanomolar range and represent the translocation of the hormones over the plasma membrane. This transport is temperature, energy, and Na⁺ dependent, and rate limiting for subsequent hormone metabolism. T₄ and T₃ mutually inhibit their high-affinity uptake processes, but they are transported by different carriers. The low-affinity processes represent binding to cell surface-associated proteins and are not involved in transport. High-affinity, energy-dependent T₃ transport systems similar to those in hepatocytes have also been identified in many other cell types, although their Na⁺ dependence varies. T₄ transport has been less well studied in other cell types, and results are variable, possibly because of its greater requirement for an optimal energy charge of the cells.

T₃ uptake in different cells (rat erythrocytes, pituitary cells, astrocytes, and mouse neuroblastoma cells) is inhibited by Trp, Phe, Tyr, and/or Leu, suggesting the involvement of system L or T amino acid transporters. A large variety of chemicals (Table 4) inhibit cellular uptake of thyroid hormones on the basis of structural similarity or by decreasing the cellular energy charge. Alternatively, inhibition is mediated by a decrease in the Na⁺ gradient over the plasma membrane, or by other as yet unknown mechanisms. The inhibitory activities of amino acids and other compounds are in the concentration range observed in humans and may interfere with in vivo tissue uptake of thyroid hormone.

	IV. Cellular Efflux of Thyroid Hormones

Top Abstract I. Historical Introduction II. Binding of Thyroid... III. Transport of Thyroid... IV. Cellular Efflux of... V. Transport of Thyroid... VI. In Vivo Plasma... VII. Plasma Membrane Transport... VIII. Requirements for a... IX. Identification of Thyroid... X. Summary and Conclusions References

 
Efflux of thyroid hormones has been studied in a number of cell types from different species, i.e., hepatocytes (122, 123, 124), erythrocytes (60, 61, 64, 125, 126), placenta cells (84, 127, 128), pituitary cells (129), FRTL-5 thyroid cells (130), NIH-3T3 cells (130), thymocytes (90), lymphocytes (131), and Ehrlich ascites cells (98).

We reported on absence of energy dependence of T₃ and T₄ efflux from cultured rat hepatocytes (122). Cellular efflux consisted of two components, representing release of hormone bound to the outer cell surface and of intracellularly located hormone. We also observed a lack of saturability of T₃ efflux after loading of rat hepatocytes using free T₃ concentrations up to 54 nM (122). However, further results suggested saturation of T₃ efflux after loading of the cells using a free T₃ concentration of 1.5 µM. Others also observed saturability of T₃ efflux, by both T₃ and T₄, from a poorly differentiated rat hepatoma cell line (HTC) (123). The same authors also demonstrated that verapamil inhibited thyroid hormone efflux from these cells as well as from isolated rat hepatocytes, cardiomyocytes, and fibroblasts (123). Furthermore, they observed increased verapamil-inhibitable T₃ efflux from HTC cells adapted for resistance to a permeable bile ester (HTC-R cells). The authors suggested that the carrier protein involved in export of thyroid hormone is related to the family of the multidrug resistance-related ABC transporters as these membrane proteins are overexpressed in HTC-R cells (123). The same group also found verapamil inhibition of T₃ efflux from FRTL-5 thyroid cells and NIH-3T3 cells (130). Others assessed T₄ and T₃ efflux from multidrug-resistant pituitary tumor cells but did not find kinetics to be different from control pituitary tumor cells (129). Neither was any effect detected by verapamil on thyroid hormone efflux in both cell types. Possible saturability of thyroid hormone efflux was not tested by these authors (129).

Efflux of T₃ from rat erythrocytes was found to be a saturable process that is stimulated by aromatic amino acid countertransport, much as T₃ uptake is stimulated by counter efflux of aromatic amino acids (61, 64). Efflux of T₄ from these cells occurred apparently by diffusion as is the case with T₄ and rT₃ efflux from human JAR choriocarcinoma cells, while also in these latter cells efflux of T₃ is saturable (84, 128). No inhibitory effect on thyroid hormone efflux by neutral system A, L, and ASC amino acids was observed in Ehrlich ascites cells (98). In many of the in vitro studies discussed in this section, it has been shown that thyroid hormone-binding proteins, including T₄-binding globulin (TBG), transthyretin (TTR), albumin, and lipoproteins have a permissive effect on efflux of thyroid hormones, probably by facilitating diffusion of thyroid hormone through the water layer around the cell (122, 124, 126).

In summary, efflux of T₃ from rat hepatocytes, cardiomyocytes, and fibroblasts has shown to be a saturable but energy-independent process. The efflux carriers in these cells may be related to the multidrug resistance-related ABC transporter family. In rat erythrocytes, T₃ efflux is also saturable and is stimulated by aromatic amino acid counter transport. Neither T₄ efflux from these cells nor T₄ and rT₃ efflux from human JAR choriocarcinoma cells was found to be saturable, in contrast to the saturable efflux of T₃. Little is known about the role of efflux mechanisms in the regulation of intracellular hormone concentrations.

	V. Transport of Thyroid Hormone into Isolated Organs

Top Abstract I. Historical Introduction II. Binding of Thyroid... III. Transport of Thyroid... IV. Cellular Efflux of... V. Transport of Thyroid... VI. In Vivo Plasma... VII. Plasma Membrane Transport... VIII. Requirements for a... IX. Identification of Thyroid... X. Summary and Conclusions References

 
Transport of thyroid hormones into perfused organs isolated from animals has been extensively studied. The advantage of studying an isolated organ is that its function can be evaluated without interference from other influences in the intact organism. Compared with experiments using isolated cells, the study of intact organs better represents the function of the tissues in vivo, although conditions are still appreciably different from the (patho)physiological situation. The results of thyroid hormone uptake studies using perfused, isolated organs from different species will be discussed in this section.

A. Transport into the liver
Transport of thyroid hormones into the intact liver has been mostly studied using organs isolated from rats. In 1979, Jenning et al. (132) reported on the effect of starvation on T₃ production from T₄ taken up by the perfused rat liver. They found that the reduced T₃ production was not caused by impaired deiodination of T₄ to T₃ in the liver but by reduced transport of T₄ into the liver, underlining the regulatory role of transport of thyroid hormone in subsequent hormone metabolism (132). One of the explanations that these authors mentioned was that T₄ uptake was inhibited by decreased activity of a "specific" transport system. We extended these studies to T₃ and also found inhibition of T₃ uptake in the intracellular compartment of livers from fasted vs. normally fed rats perfused with medium lacking glucose, insulin, and cortisol (133). This inhibition was reverted to normal by a 30-min preperfusion of fasted livers with medium containing a combination of glucose, insulin, and/or cortisol but not by the individual additions. On the basis of these results, we explained the diminished T₃ uptake by a decrease in cellular ATP induced by fasting, which was restored by preperfusion with energy-rich medium (133). Further studies using fructose in the perfusate to (transiently) lower cellular ATP stores in the rat liver showed a parallel decrease in T₄ uptake in the intracellular compartment of the liver, thus underscoring the regulatory role of the energy charge of the cell in the transport process (Fig. 2, Ref. 134). Similar to the results in cultured rat hepatocytes, we found that, in addition to the energy state of the liver, the free hormone concentration in the perfusion medium determined the amount of hormone taken up by the intracellular compartment of the liver (135). Studies using livers from amiodarone-treated animals indicated that transport of T₄, but not of T₃, was inhibited (136), in agreement with hepatocyte studies (47, 55) showing that T₄ and T₃ are transported differently across the liver plasma membrane. Efflux of T₃ from the isolated perfused trout liver was stimulated by addition of T₄, epinephrine, or TSH to the perfusion medium, and efflux of T₄ was stimulated by addition of T₄ to the medium. The stimulating effect of extracellular thyroid hormone on efflux of T₄ and T₃ may be caused by inhibition of reuptake, stimulation of an exchange mechanism, and/or displacement of hormone from intracellular binding sites (137, 138). However, the stimulation of T₃ efflux by epinephrine and TSH remains unexplained.

View larger version (18K): [in this window] [in a new window]   Figure 2. T4 liver uptake (in % dose) in rat livers during glucose (•) and glucose/fructose () perfusion. [Reproduced with permission from M. de Jong et al.: Am J Physiol 266:E768-E775, 1994 (134 ).]

Incubation of whole soleus muscle isolated from rats showed stereospecific, energy- and Na⁺-dependent uptake of T₃, but T₄ uptake was considered to be a diffusion process (49, 144). Addition of insulin to the incubation medium stimulated T₃ uptake but did not affect T₄ uptake (145). T₃ uptake in the perfused rat heart showed a saturable process with an apparent K_m value of 80 µM (146). This value is about 1 order of magnitude higher than the apparent K_m values obtained in in vitro studies using isolated cardiomyocytes (Tables 2 and 3). This difference may be explained by the fact that T₃ uptake in the perfused rat heart was determined after a single capillary passage that proceeds within seconds and differs fundamentally from techniques in which initial uptake rates in cells are measured over a period of minutes. The question is if the former method represents uptake of the ligand by the cardiomyocytes, since this assumes that the hormone has already passed the endothelium after such a short time lapse. Another explanation, of course, is that the experiments using cultured cells provide data that are more remote from the in vivo situation than data obtained from isolated organ studies. In contrast to the rat liver (132), fasting did not decrease uptake of T₄ by the isolated perfused rat kidney, but T₄ uptake was decreased in kidneys of diabetic rats (147, 148).

In summary, uptake of T₄ and T₃ is decreased in isolated livers from fasted vs. fed rats perfused with the same "energy-poor" medium. Changing the perfusate to an energy-rich medium restores uptake in 30 min, suggesting restoration of cellular ATP. Perfusion of fed livers with fructose results in a lowering of cellular ATP and a parallel decrease in thyroid hormone uptake. Analysis of transport in livers from amiodarone-treated rats showed that in the intact liver, T₃ and T₄ are also taken up by different mechanisms. Apart from the cellular energy charge, the free and not the protein-bound fraction of thyroid hormone determines the amount of hormone taken up by the cellular compartment of the liver. Uptake of T₄ and T₃ in isolated rat or sheep choroid plexus was found to be nonsaturable by some investigators but partly saturable by others. Saturable transport of T₃, but not of T₄, was observed in the isolated rat soleus muscle. Saturable T₃ transport was also found in the perfused rat heart.

	VI. In Vivo Plasma Membrane Transport of Thyroid Hormones in Animals

Top Abstract I. Historical Introduction II. Binding of Thyroid... III. Transport of Thyroid... IV. Cellular Efflux of... V. Transport of Thyroid... VI. In Vivo Plasma... VII. Plasma Membrane Transport... VIII. Requirements for a... IX. Identification of Thyroid... X. Summary and Conclusions References

 
To assess plasma membrane transport of thyroid hormones to different organs in vivo, animals were injected with tracer amounts of labeled hormones after which entry of hormones into the isolated organs was analyzed.

A. Brain
Several questions related to transport of thyroid hormone to the brain have been addressed. One aspect is whether entry of thyroid hormone into brain proceeds via a passive process or via a carrier-mediated mechanism. When dogs were injected intravenously with tracer T₄, allowing entry in the brain via the blood-brain barrier (BBB) and the CSF, brain uptake was saturable under conditions of T₄ loading, indicating that transport occurred via a carrier-mediated process (149). In mice, transport of T₃ into the brain was saturable but, under the conditions of the experiment, no saturation of T₄ transport was observed. Efflux of both T₃ and T₄ from the brain appeared to proceed by a carrier-mediated mechanism (150).

Another point of interest is to what extent transport through the BBB and the choroid plexus-CSF barrier (CP-CSFB) contributes to overall brain uptake of thyroid hormone. To investigate this, rats were injected either intravenously or intrathecally with radioactive thyroid hormones. When administered intravenously, hormones have access to the brain via both the BBB and the CP-CSFB. However, hormone injected intrathecally represents entry into brain cells via the CP-CSFB. After injection of radioactive hormones via these two routes and subsequent autoradiography of the brain, distribution of thyroid hormone over brain areas could be documented as well as the contribution of the BBB and the CP-CSFB to brain accessability (151, 152, 153). These studies demonstrated that T₃ and T₄ enter the brain mainly via the BBB for distribution throughout the brain, but that localization in the ependymal cells and in the circumventricular organs occurs via the CP-CSFB. In contrast, rT₃ is excluded by the BBB but has limited access to the brain via the CP-CSFB (Fig. 3).

View larger version (29K): [in this window] [in a new window]   Figure 3. Routes of iodothyronine transport between blood and brain. According to autoradiographic results, rT3 crosses the CP-CSFB but not the BBB, whereas T3 and T4 cross both BBB and CP-CSFB. [Derived from Ref. 153 .]

B. Other organs
The liver is another organ that has been studied in animals for plasma membrane transport of thyroid hormone. Pardridge et al. published a series of in vivo studies in the rat (for review see Ref. 160). From their studies the authors concluded that thyroid hormone delivery to the liver "occurs via the free intermediate mechanism, i.e., protein-bound hormone debinding is an obligatory intermediate step in the transport process." Although they found that transport of T₄ into rat brain via the BBB is a saturable process, they could not find saturability of plasma membrane transport in rat liver, and suggested that this occurred via passive diffusion. The authors used for their studies a single capillary pass technique for analysis of initial kinetics of transport (160). The model used by Pardridge et al. and their interpretation of the data were strongly contested (161, 162). The main criticism concerned the rate-limiting role in the transport process that was attributed to the dissociation of hormone from serum binding proteins. No such role could be envisaged, both on theoretical and experimental basis, by these opponents. Others documented hepatic uptake in mice, injected in vivo with radioactive T₃, using autoradiography (163). Excess unlabeled T₃ resulted in 90% inhibition of liver uptake of labeled T₃. Time sequence autoradiographic analysis showed that the plasma membrane is initially labeled before internalization of T₃ occurs (163). These results clearly document in vivo specific binding of T₃ to the liver plasma membrane as an initial step to internalization of the hormone. In vivo injection of rats with radiolabeled T₄ and subsequent measurement of uptake in heart and lung tissue, isolated at different time intervals, showed that T₄ transport in these organs was also saturable, in accordance with a carrier-mediated transport mechanism (164).

In summary, brain entry of T₄ in dogs appears to proceed via a carrier-mediated mechanism. This was also found for brain uptake of T₃, but not of T₄, in the mouse. It was further shown in the rat that T₃ and T₄ mainly enter the brain via the blood-brain barrier for distribution throughout the brain, and via the CP-CSF barrier for restricted distribution in circumventricular areas. Although it has been envisaged for a long time that TTR expressed in the choroid plexus plays an essential role in the transport of thyroid hormones into the brain, total lack of the protein in TTR knock-out mice has no effect on concentrations of plasma free thyroid hormones and TSH or on tissue thyroid hormone status. In vivo studies have shown saturable T₃ uptake into rat liver and saturable T₄ uptake into mouse lung and heart.

	VII. Plasma Membrane Transport in Humans

Top Abstract I. Historical Introduction II. Binding of Thyroid... III. Transport of Thyroid... IV. Cellular Efflux of... V. Transport of Thyroid... VI. In Vivo Plasma... VII. Plasma Membrane Transport... VIII. Requirements for a... IX. Identification of Thyroid... X. Summary and Conclusions References

 
A. Introduction
In healthy individuals, about 80% of plasma T₃ is produced outside the thyroid gland, the remaining 20% being secreted directly by the thyroid (165). In the extrathyroidal pathway, T₃ is produced by outer ring deiodination of T₄, and in this process the type I deiodinase in the liver (and kidneys) plays an important role (165, 166). Another organ that may be involved in this pathway in humans is skeletal muscle, expressing the type II deiodinase that also catalyzes the conversion of T₄ to T₃ (167). To reach the intracellular T₃-producing enzymes, T₄ must cross the plasma membrane of these tissues. It has been established in rats that the extent to which nuclear receptor-bound T₃ is derived from plasma T₃ and from local T₃ production from T₄ varies among the tissues. Thus, for instance, nuclear T₃ in cerebral cortex is derived for 80% from local conversion of T₄, in pituitary for 50%, in skeletal muscle for 40%, and in liver for only 5% (168, 169). In other words, for exertion of biological activity by nuclear T₃, both T₄ and T₃ must cross the plasma membrane of target cells. It follows that the activity of these transport processes may have an important influence on the regulation of the biological activity of thyroid hormone. Although the exact contribution of the different sources of nuclear T₃ in human tissues is unknown, it will also depend to varying degrees on plasma membrane transport of T₃ and its precursor T₄.

Many reports have dealt with the measurement of thyroid hormone distribution and metabolism in humans. However, few of these are concerned with analysis of unidirectional transport of thyroid hormones into tissues. To study regulation of biological processes, it is in general necessary to analyze these under circumstances of perturbation of the physiological steady state. This is certainly also true for the study of the regulation of thyroid hormone transport into tissues. Both in starvation and in so-called nonthyroidal illness (see Section VII.C), plasma T₃ production is decreased. As the diminution in plasma T₃ production may be substantial and thyroidal secretion of T₃ contributes only little to total plasma T₃, the main cause of this diminution in T₃ production must consequently be located in the extrathyroidal pathway. Both starvation and nonthyroidal illness have been used as models to study regulation of thyroid hormone penetration into target tissues. Two possibilities have been suggested to be responsible for the lowered T₃ production in these situations, i.e., a decrease in outer ring deiodinase activity in plasma T₃-producing tissues and/or a decrease of T₄ transport into these tissues as substrate for T₃ production. There is evidence in animals, but not in humans (170), that outer ring deiodination is indeed lowered in starvation and in nonthyroidal illness, but this aspect will not be further discussed here. For further orientation, the reader is referred to Ref. 171 . In this section we will discuss plasma membrane transport of thyroid hormones in human tissues both in starvation and in nonthyroidal illness.

B. In starvation
In caloric deprivation, as in nonthyroidal illness (see Section VII.C), abnormalities in serum thyroid function parameters are invariably present. The most constant and thus characteristic abnormality is a low serum T₃ concentration; hence the term "low T₃ syndrome" for this entity. Serum T₄ and TSH are usually normal, whereas serum rT₃ is usually elevated (for a review see Ref. 172). To our knowledge the first published study that was primarily designed to evaluate unidirectional transport of thyroid hormones into tissues before and during caloric deprivation in man was published in 1986 by our laboratory (170). In this study T₄ and T₃ kinetics were studied using a three-pool model of thyroid hormone distribution and metabolism in 10 obese but otherwise healthy subjects before dieting and while on a 240-kcal diet. During caloric restriction, unidirectional transport of T₄ and T₃ into the rapidly equilibrating tissues (liver) was decreased by 50% and 25%, respectively, when corrected for changes in free hormone concentration. The decrease in plasma T₃ production amounted to 42%, about equaling the reduction in T₄ transport into the liver. T₄-to-T₃ conversion rate decreased by an insignificant 8%. Therefore, the lowered T₃ production during caloric deprivation is largely, if not fully, explained by a decrease of T₄ entry into T₃-producing tissues. The fasting-induced decrease in liver T₄ transport may be explained, at least in part, by a decrease in the energy charge of liver cells. This explanation is based on at least two points. First, it has been shown that starvation leads to ATP depletion of the liver as assessed by ³¹P-magnetic resonance spectroscopy (173). Second, tissue T₄ transport was much more affected by caloric deprivation than transport of T₃, similar to findings of T₄ and T₃ transport in cultured rat hepatocytes deficient in ATP (Ref. 48 and Fig. 1). To further substantiate the effect of the intracellular ATP concentration on hepatic T₄ uptake in vivo in humans, liver T₄ uptake was measured in four healthy human volunteers, using T₄ tracer plasma kinetics, before and after an intravenous bolus injection of fructose, which is known to transiently decrease liver ATP levels. Obviously, hepatic ATP could not be measured, but fructose was found to induce an increase in serum lactic acid and uric acid concentrations, reflecting a decrease in liver ATP. After fructose administration there was a temporary decrease in liver T₄ uptake that normalized after fructose was metabolized and hepatic ATP concentrations were restored, as reflected by the normalization of serum lactic acid and uric acid levels (134). In contrast to the transient effect of fructose, transport of T₄ into the liver remained suppressed when the same subjects were studied on a calorie-restricted diet (Fig. 4). As will be discussed in Section VII.C, NEFAs that circulate in increased concentrations during caloric restriction have an additional inhibitory effect on T₄ uptake by the liver.

View larger version (16K): [in this window] [in a new window]   Figure 4. Computed kinetics of T4 uptake into the rapid equilibrating pool (REP, representing largely liver) in four obese volunteers, before () and during (•) caloric deprivation and after intravenous fructose (). [Derived from Refs. 134 and 170 .]

C. In nonthyroidal illness
NTI may be defined as any acute or chronic illness, not related to the thyroid gland, that is accompanied by an abnormal pattern of thyroid function parameters. Other terms that are synonymously being used are the "low T₃ syndrome" because serum T₃ is invariably low in NTI, and the "euthyroid sick syndrome" because patients are usually clinically euthyroid despite the low serum T₃ and sometimes also low T₄ levels. With an increase in severity of disease there is a progressive decrease in serum T₃ and, in most diseases, an increase in serum rT₃ that eventually plateaus. Serum T₄ is usually normal but may be slightly increased in mild disease and lowered in critical illness (Fig. 5). Serum TSH is usually normal but may be depressed in severe illness (175, 176). Many studies of thyroid hormone distribution and turnover kinetics in patients with NTI have been reported (for reviews see Refs. 171, 175, 176, 177). In general, they show that T₄ production rates are normal, except in severe illness when it is decreased, but that T₄ transport into tissues is decre