Werner & Ingbar's The Thyroid: A Fundamental & Clinical Text, 9th Edition

5.Thyroglobulin Structure, Function, and Biosynthesis

Peter Arvan

Bruno Di Jeso

Thyroglobulin (Tg) is the most highly expressed protein in the thyroid gland. Its vastly higher abundance than other proteins in thyroid extracts tends to yield an exaggerated estimate of its expression relative to that of other thyroid proteins, because most Tg in the thyroid gland is in fact extracellular, having been secreted into the lumen of thyroid follicles, where it is stored as colloid. Nevertheless, even when extracellular storage is not considered, Tg is still the predominant thyroid protein, and is the most important one, because its encoded structure provides for the coupling reaction that underlies the biosynthesis of the active thyroid hormones thyroxine (T4) and triiodothyronine (T3) (see section on thyroid hormone synthesis in Chapter 4).

Tg has a number of remarkable features. First, it falls within the top 1% of all proteins in terms of the molecular mass of its monomeric structure (~330 kDa). Approximately 10% of this mass (i.e., ~30 kDa) consists of covalently bound carbohydrates, the result of posttranslational modifications as Tg passes along the intracellular secretory pathway of thyroid follicular cells (thyrocytes). Indeed, Tg has long served as a model protein for analysis of the carbohydrate side-chain structures that are formed during N-linked glycoprotein processing (1). In addition, the polypeptide backbone of Tg, amounting to ~90% of its total molecular mass (i.e., ~300 kDa), consists of a signal peptide sequence plus ~2750 residues (± 5 residues, depending upon the species) (2). The secreted polypeptide contains at least 66 tyrosine residues (in human Tg, with additional tyrosines in other species). The percentage of tyrosine residues in Tg that become iodinated normally varies with the dietary iodine supply, but it is not unusual for 20% of them to contain iodine, with three quarters of these as residues of monoiodotyrosine (MIT) and diiodotyrosine (DIT) that are not used for the formation of T4 and T3 (3). Thus, Tg protein serves as the primary internal reservoir of recycling iodine in the body, upon which future biosynthesis of thyroid hormones is based.


The appearance of Tg in evolution is associated with thyroid hormone biosynthesis. A colloid-filled follicular lumen surrounded by thyrocytes (presumably a response to decreasing availability of iodide as animals migrated away from the sea to a terrestial environment) and the Tg protein are present in all vertebrates (4). During mammalian embryonic thyroid folliculogenesis (5), as well as folliculogenesis in reconstituted mammalian thyrocyte cultures and in vivo models, intracellular lumens (membrane-bound internal cavities) form within thyrocytes. The cells then associate with one another, after which the intracellular lumens are externalized to form a central, apical extracellular space, which is a hallmark of Tg storage (6). The intracellular lumens have a close physical relationship with the Golgi complex through which newly synthesized Tg traverses (7), and iodination of Tg can occur within them (8), resulting in thyroid hormone formation [nevertheless, their quantitative contribution to overall iodination or hormonogenesis is probably insignificant, because the intracellular availability of hydrogen peroxide in vertebrates is limited (9)]. Although not well studied, some sea-dwelling prevertebrates have specialized pharyngeal epithelial cells that do not aggregate to form obvious follicles (perhaps such cells bear weak evolutionary relationships to salivary epithelial cells that exhibit basolateral iodide uptake and do not enclose an apical lumen), but do form Tg-like iodoproteins (10). Any such means to produce and secrete iodoproteins would not efficiently conserve iodide. Nevertheless, even in prevertebrates, formation of T4 and T3 is likely to be linked to a Tg evolutionary precursor (11), and within hagfish or lamprey eels, Tg-like protein is already present (12) along with demonstrable follicles. These findings suggest evolutionarily conserved roles for Tg protein in iodine storage (particularly in land-dwelling and freshwater organisms) and iodothyronine formation. Cloning and sequencing of Tg-like complementary DNA (cDNA) derived from prevertebrates could provide valuable new insights into the evolutionary origins of T4-forming sites and the domain structure of Tg molecules.


The functionality of the Tg messenger RNA (mRNA) was initially established when a 33S thyroidal mRNA fraction injected into Xenopus oocytes resulted in production of immunoreactive Tg protein in the eggs (13), while a similar 33S thyroidal mRNA fraction obtained from a strain of Dutch goats with congenital goiter did not result in production of detectable Tg protein (14). The complete Tg mRNA sequence was eventually deduced by sequencing of fragments of bovine Tg cDNA (2) and those of other species (Fig. 5.1), which allowed for many structural insights (15) and confirmed that there is only one Tg gene in mammals. The human Tg gene spans approximately 300 kB on the long arm of chromosome 8, of which only ~3% represents the exons [48 of them (16)], separated in many instances by large intronic sequences. Mutations within introns as well as imprecision in the mRNA splicing mechanism have led to the possibility of production of Tg splice variants with no change in Tg exon sequence (2,17,18,19). Alternative splicing is also promoted by increasing thyrotropin (TSH) stimulation (20). This might serve as an evolutionary adaptation to the high frequency of variations in the Tg coding sequence in the population (21), some of which might be less competent for hormonogenesis (22). Nevertheless, TSH action on its receptor is not an absolute requirement for Tg production (23), although there are requirements for the transcription factors TTF-1, TTF-2 (24), and Pax-8 (25), in conjunction with other factors (26). There are also reports of patients with congenital goiter whose thyrocytes may not maintain a sufficient steady-state level of Tg mRNA (27,28). While this could represent a class of transcriptional defects (29), active translation (polyribosome formation) is one of the critical features promoting mRNA stability. Therefore, poorly translated Tg mRNAs (from splice forms, or from exon sequence changes bearing frame-shifts and/or premature termination codons) are likely to correlate with low mRNA abundance of those forms.

FIGURE 5.1. The nucleotide and encoded 2746 amino-acid sequence of murine thyroglobulin (Tg) (described in single-letter code), derived from Genbank accession number AF076186. In addition, the location of selected point mutations in the coding sequence of Tg associated with thyroid pathology in several different species has been identified on this map by the normal amino acid presented in enlarged, bold type. In order, from amino to carboxyl terminus, this includes: a premature stop codon at position 277 in human Tg (260), a premature stop codon at position 296 in goat Tg (261), a premature stop codon at position 697 in bovine Tg (106), a Glu to His change in exon 10 in human (126,127), a Cys to Arg substitution at position 1263 in human Tg (124), a stop codon at position 1510 in human Tg (28), an Arg to His substitution at position 2223 in human Tg (123), a Leu to Pro substitution at position 2263 in murine Tg (112), and a Gly to Arg substitution at position 2320 in rat Tg (120,121).

The development of Tg cDNA expression vectors for mammalian cells (30) has allowed the study of full-length recombinant versions leading to Tg secretion from heterologous cells (31). Such an approach overlooks the complexity of Tg mRNA splicing, although it increases focus on the exon sequences. It is important to note that the Tg protein has never been studied by x-ray crystallography, because crystals of the pure protein have not been obtained, perhaps due to the numerous posttranslational modifications of Tg (see later in the chapter). Thus, insights into the internal structure of the Tg protein have been limited largely to primary sequence analysis—although negative-staining electron microscopy has demonstrated that a major form of Tg isolated from thyroid gland is dimeric (32).


The (12S) Tg monomer in all vertebrates is comprised mainly of four regions (Fig. 5.2), the first three containing distinct cysteine-rich repeats (2). Region I, found within the first ~1200 residues, contains 10 of the 11 so-called “Tg type 1 repeats” of roughly 60 amino acids each. Each repeat surrounds a central WCV (Trp-Cys-Val) sequence plus at least three additional cysteines (one expanding the motif to CWCV, in which each of the two Cys residues are partnered into different disulfide bonds), thereby shaping and stabilizing the domain. Type 1A repeats have a total of six cysteine residues in the domain (three disulfide bonds), which are also found in several other proteins, including equistatin and the thyropin family of cysteine protease inhibitors (33,34), entactin/nidogen (35), insulin-like growth factor-binding protein-6 (IGFBP-6) (36), and the carcinoma-associated antigen GA733–2 (37). Type 1B repeats have only four cysteine residues (38). The disulfide bond pattern Cys1-Cys2, Cys3-Cys4, Cys5-Cys6 prevails in all Tg type 1A repeats (37) and the sequence homology between different proteins bearing the repeat (Fig. 5.3) suggests that this domain has evolved with a common fold designed to be protease resistant and to function as a protease inhibitor/regulator (39). Region II (Fig. 5.2) contains the 11th Tg type 1 repeat, while region III contains the so-called type 3 repeat units that have little or no homology with other proteins in existing databases (2). Because cysteine side chains are internally located within the various Tg repeats, they form a series of domains internally “stapled together” (via disulfide bridges) while being tethered end-to-end with intervening peptide sequences. Consistent with such a view, limited digestion using a variety of proteases tends to cleave between the repeats, liberating them intact from the parent Tg molecule (40).

FIGURE 5.2. Sketch of the regional structure (below), including repeat domains (above), in the primary sequence of Tg. Note that the region and domain alignments are not drawn precisely to scale. As shown in the sketch (below), the amino-acid sequence of Tg begins with a cleavable signal peptide (open box), the conserved thyroxine hormonogenic site A (see text) (hatched box), the ten type 1 (A+B) repeat domain units contained in region I, the Cys-rich region II which comprises the 11th type 1 repeat, the region III with repeat units that do not closely match known sequences in the database, and the acetylcholinesterase-homology region followed by the hormonogenic site C (see text) (hatched box).

FIGURE 5.3. An alignment of polypeptide sequences containing thyroglobulin (Tg) type 1 repeats, showing the relative positioning of the conserved Cys residues. The last sequence shown is derived from human Tg.

The fourth and most carboxyl-terminal region, the acetylcholinesterase (AChE) region, (Fig. 5.2) Tg has 31% identity and 47% similarity to the entire length of acetylcholinesterase (41,42). All six cysteine residues involved in cholinesterase intrachain disulfide bonding are conserved in Tg (43), and it is likely that these two proteins share a common folding pattern in their respective tertiary structures.

Altogether, human (and bovine) Tg has 122 cytseines (mouse Tg has 121 Cys residues). When studied by nonreducing SDS-PAGE, the mobility of monomeric Tg increases dramatically, indicating that its molecular radius decreases progressively as a function of time after Tg biosynthesis (44), in comparison with the mobility of the same Tg molecules that have been fully “opened” (i.e., linearized) by reducing all disulfide bonds with dithiothreitol (Fig. 5.4). These results indicate that the vast majority of cysteine thiols are involved in intradomain disulfide pairing within the four major regions of Tg (45). That few unpaired Cys residues remain (46) may explain why Tg homodimers are not covalently associated during transit through the intracellular secretory pathway of thyrocytes (47,48). Nevertheless, extracellular Tg may begin to form intermolecular disulfide bridges when it reaches the oxidizing/iodinating environment of the follicular lumen, resulting in cross-linked dimers (19S), tetramers (27S), and higher order assemblies (49).

FIGURE 5.4. Kinetic resolution of the maturation of intrachain disulfide bonds of thyroglobulin (Tg). As disulfide bonds are formed within newly synthesized Tg and the structure of the various repeat domains is reorganized, the molecular radius of Tg decreases. When subjected to acrylamide gel electrophoresis under nonreducing conditions, this Tg penetrates further into the gel (lower band). When the Tg protein is intentionally denatured and all disulfide bonds are disrupted, and the protein subjected to electrophoresis under reducing conditions, the penetration of the newly synthesized Tg is unchanged at all times shown. (From Di Jeso B, Ulianich L, Pacifico F, et al. The folding of thyroglobulin in the calnexin/calreticulin pathway and its alteration by a loss of Ca2+ from the endoplasmic reticulum. Biochem J 2003;370:449, with permission.)



The initial stages of Tg biosynthesis are similar to those of other secretory glycoproteins. During translation, nascent Tg polypeptide chains, while still emerging from the ribosome-studded polyribosomes, are directed to the cytosolic side of the endoplasmic reticulum by virtue of the Tg signal peptide sequence that promotes association with the signal recognition particle and its receptor in the membranes of the endoplasmic reticulum (50). From there, the nascent Tg polypeptide is translocated into the lumen of the endoplasmic reticulum via the intramembrane Sec61 protein complex, with assistance from molecular chaperones on the lumenal side (51). After translocation of ~100 amino acids, the Tg signal peptide is excised by a specific protease (52), while ongoing translation is coupled to the translocation of the large remaining nascent polypeptide.

Immediately upon entry into the endoplasmic reticulum, the linear Tg polypeptide begins to be converted into a native three-dimensional structure. This process is more extensive than coupled translation-translocation, in the sense that while a small peptide hormone precursor such as proinsulin achieves its native state after no more than a few minutes within the endoplasmic reticulum (53), for Tg this takes about an hour. During this process (54), many of the repeat domains are likely to be buried interiorly within the tertiary structure of Tg, based on the fact that it requires progressively increasing concentrations of reducing agents in vitro to gain access to these domains in the native structure (55). Few of the details of the Tg folding process are known, but the basic steps are likely to be similar to those of other secretory glycoproteins. For example, at an estimated translation rate of 6.5 amino acids per second, one can deduce that it should take ~7 minutes for the entire nascent Tg polypeptide to be delivered into the endoplasmic reticulum. In this case, many of the repeating units of region I of Tg will have been engaged in folding for several minutes before some of the downstream domains of Tg arrive within the lumen of the endoplasmic reticulum. Thus, initial folding of the type 1 repeats is likely to be independent of later downstream domains encoded in the Tg sequence. Along the same lines, when the domain of Tg homologous to AChE reaches the endoplasmic reticulum, its initial folding is likely to be independent from that of the N-terminal domains of Tg that are already beyond their earliest stages of folding. Quite possibly, regional folding may even be sequential (56).

The slowness with which nascent Tg buries its many cysteine side-chains while forming a compact structure (Fig. 5.4) suggests that there is ample opportunity for mispairing of Tg disulfides, and indeed there is evidence that some nascent Tg complexes have mispaired intermolecular disulfide bonds (54,57,58). These disulfide-linked complexes are apparently “on-pathway” in normal Tg folding, because their disappearance correlates with the appearance of Tg monomers (and later, dimers). Resolution of disulfide mispairings is likely to require the involvement of protein disulfide isomerase (PDI) or other members of the superfamily of endoplasmic reticulum oxidoreductases (59,60,61,62).

Additional molecular chaperones of the endoplasmic reticulum also play distinct roles in facilitating the early folding of nascent Tg. One member of the heat-shock protein-70 family, known as BiP (63), has been identified repeatedly in protein interactions with Tg both in vitro and within the endoplasmic reticulum (57,64,65,66,67,68,69). BiP is a glucose-regulated protein (GRP), i.e., it is transcriptionally and translationally up-regulated when the extracellular glucose concentration is low (70), when glycosylation is compromised by energy depletion (71) [such as in states of compromised blood flow (68)], or when misfolded Tg or other secretory proteins begin to accumulate in the endoplasmic reticulum for any other reason (such as alterations in redox environment or lumenal calcium concentrations) (72,73). GRP94, a member of the heat-shock protein-90 family that is located in the endoplasmic reticulum and regulated like BiP, also associates with nascent Tg (67). Neither BiP nor GRP94 actively participate in the process of folding Tg (31,74); rather they reversibly associate with newly synthesized Tg molecules, with preferential binding to short hydrophobic sequences, to prevent other inappropriate hydrophobic interactions. Merely by limiting the opportunity for “off-pathway” folding events, the association of the endoplasmic reticulum chaperones promotes “on-pathway” folding of Tg, thus improving the overall yield of native Tg. As this is crucial to the success of secretory protein production in all cells, several (but not all) endoplasmic-reticulum chaperone proteins are essential for life among eukaryotes (75), and the control of their expression is tightly regulated (76,77). Additional chaperones that interact with Tg during its maturation in the endoplasmic reticulum are still being discovered, such as GRP170 (69) and ERp29 (78).


Potential asparagine (N)-linked carbohydrate acceptor sites are encoded by the sequence Asn-Xxx-Ser/Thr (Xxx is any amino acid), each of which may accept transfer of a pre-assembled “core oligosaccharide” containing 14 monosaccharide units attached to the asparagine side chain. There are 20 such potential acceptor sites in the sequence of human Tg (and mouse Tg); on average, 16 of these sites are actually utilized by oligosaccharyl transferase (79). This varies between species and amongst the population of Tg molecules within a single species (80), e.g., bovine Tg has a lesser degree of N-glycosylation (81).

One critical link between N-glycosylation and Tg stability stems from the discovery that a family of endoplasmic-reticulum chaperones known as calreticulin and calnexin function as lectins, that is, they recognize sugars (82). Calreticulin is abundantly expressed in the lumen of the endoplasmic reticulum, while calnexin is an abundant integral membrane protein whose lumenal domain is highly homologous to calreticulin (83). The initially added N-linked core glycan has a (GlcNAc)2(Man)9(Glc)3 structure in which the nine mannose (Man) residues form a three-pronged branching structure, and all three glucose (Glc) residues are stacked at the terminus of one branch (Fig. 5.5). This is followed by stepwise removal of all three terminal glucose residues via the sequential actions of the endoplasmic-reticulum glucosidases I and II. During the short period when the core oligosaccharide structure has a single terminal glucose, it is a substrate for binding to calreticulin or calnexin. Both chaperone lectins have the same oligosaccharide-binding specificity, although only one lectin may be bound to the carbohydrate at any one time. Ultimately, removal of the final glucose residue destroys the calreticulin/calnexin binding site. However, if the domain in Tg (or other substrate glycoprotein) has not folded, it can be recognized by UDP-Glc glycoprotein glucosyl transferase (UGT) (Fig. 5.5), a critically important enzyme that adds a single glucose residue back to the same site on the core oligosaccharide (84,85). This in turn re-forms the calreticulin/calnexin binding site, so that chaperone association may proceed again, in a cyclical manner, until the domain has folded to the point that it is no longer a substrate for re-glucosylation by UGT (because UGT is active only on unfolded Tg and not on properly folded Tg) (86,87). Meanwhile, mannosidase activities of the endoplasmic reticulum initiate trimming of Man residues from the other oligosaccharide branches (1), an event that usually occurs in preparation for further modifications that will take place when Tg reaches the Golgi complex (see later in the chapter). The binding of Tg to calnexin under normal conditions is relatively modest (65), because some calnexin dissociation occurs cotranslationally (65,88) and also because calreticulin may be a more important lectin-chaperone for Tg (44).

FIGURE 5.5. Early N-linked carbohydrate-processing events on newly synthesized thyroglobulin (Tg). As the nascent Tg polypeptide enters the lumen of the endoplasmic reticulum, a high fraction of its encoded potential Asn-linked glycosylation acceptor sites bind covalently to a 14-unit oligosaccharide from a lipid-linked donor. Three terminal glucose residues (G) on one branch of the oligosaccharide structure are the site of attack of endoplasmic-reticulum glucosidases I and II. A monoglucosylated oligosaccharide is the site of binding of the endoplasmic reticulum lectin chaperones calreticulin or calnexin. These lectin chaperones are also associated with the thiol reductase ERp57, which assists in the formation/isomerization of Tg disulfide bonds, further facilitating Tg folding (see text). If a glycosylated Tg domain remains unfolded after the sole remaining glucose residue has been removed, this domain serves as a potential substrate for UDP-glycoprotein glucosyl transferase (UGT). UGT re-adds a single glucose residue to the carbohydrate structure, restoring its capacity to bind calreticulin/calnexin and associated ERp57. Ultimately, for Tg that is properly folded, mannose (M) trimming by mannosidases begins in the endoplasmic reticulum and continues in the Golgi complex. For Tg that is misfolded, mannose trimming by endoplasmic-reticulum mannosidase I precedes a process of endoplasmic reticulum-associated degradation that occurs via the ubiquitin-proteasome system (see text).

An additional special feature of calreticulin or calnexin binding is that it brings the folding of nascent Tg into proximity with ERp57, a thiol reductase member of the PDI superfamily that catalyzes formation/isomerization of disulfide bonds (89). Thus, ERp57 disulfide-isomerase function is also dependent upon the same early steps in N-linked carbohydrate processing. Because formation/isomerization of mispaired disulfide bonds is of great importance in the folding of newly synthesized Tg, the activity of ERp57 and the proper processing of the N-glycans of Tg (90) must be viewed as key steps ensuring an adequate rate and efficiency of this process. At the conclusion of the extended period of intrachain disulfide bond formation, which may be rate limiting in achieving proper regional domain structures, the tertiary structure of nascent Tg monomers is now suitable for homodimerization, and this proceeds before exit from the endoplasmic reticulum (54,57,64,91).

These considerations provide a molecular explanation for the established relationship between N-linked glycosylation and the stability of Tg, which in turn affects the efficiency of Tg iodination (92,93). In relation to this, perturbations of divalent cation concentrations in the secretory pathway, particularly those of Ca2+ (and perhaps Mn2+) may destabilize Tg structure (94). Some of this effect may be indirect, because calcium depletion from the endoplasmic reticulum alters carbohydrate processing (95,96,97), reduces Tg binding to calreticulin and calnexin (44), and greatly inhibits export of Tg from this compartment (91). This may be caused by a diminished rate of isomerization/formation of the Tg disulfide bonds needed for folding; under improper glycosylation conditions, increased residence time of newly synthesized Tg in the endoplasmic reticulum correlates with its prolonged association with BiP and GRP94 (44,57,69).


Even without global effects on the environment of the endoplasmic reticulum, mutations altering the amino acid sequence of Tg could yield phenotypes similar to those described earlier. Patients and animals with congenital goiter and hypothyroidism who have abnormal Tg molecules have been described (98,99). In some instances, the endoplasmic reticulum in the thyrocytes was dilated or swollen (100,101), with little or no stainable Tg in the colloid (102), but knowledge of the specific molecular defects at the genetic level was lacking (103). The finding of nonfunctional Tg mRNA was a first step in this direction (14). These disorders are inherited as autosomal recessive traits in virtually all humans and animals. The mating, without inbreeding, of phenotypically unaffected heterozygotes yielding affected compound heterozygous progeny is such a rare event that it may initially be confused with autosomal dominant inheritance (104) (see Chapters 20 and 48). Patients with these disorders may have normal or low levels of Tg, depending on how well their Tg mRNA can be translated (105,106).

Unlike the Tg splicing and frame-shift mutants that have been described (21), there are relatively few descriptions of single missense mutations (amino acid substitutions). Thus, what have been called “Tg synthesis defects” are not typically explained by mutations in the coding region of the Tg gene (22). Only a few particular examples of these defects are described here.

The homozygous cog mutation in mice results in the murine version of cretinism with microcephaly and hypomyelination (107). In addition, the animals are runts, with little or no detectable 19S Tg in their thyroid gland (108). Instead, a peak at 12S extends to above the 27S position, indicating defective dimerization and formation of abnormal Tg complexes, and very little Tg is iodinated (108,109,110). Serum TSH concentrations are very high, and there is massive thyrocyte hyperplasia and hypertrophy, with a swollen endoplasmic reticulum and increased expression of multiple molecular chaperones (111). The folding and dimerization defect, which is to some extent ameliorated at lower temperature (111), is the result of a single L2263P (leucine→proline) substitution in the AChE-homology domain of Tg (112). The misfolded mutant protein is primarily retained within the endoplasmic reticulum until the carbohydrate is trimmed by the endoplasmic-reticulum mannosidase I (112A), after which it is translocated back to the cytosol (113) for degradation by the ubiquitin-proteasome system.

A similar situation seems to prevail in dwarf (rdw/rdw) rats. These rats were initially thought to have growth hormone deficiency (114,115), but in fact their growth failure is secondary to congenital hypothyroidism (116,117). Although their serum TSH concentrations are not as high as those of cog/cog mice, the thyroid glands of rdw/rdw rats also have massive expansion of the endoplasmic reticulum (118) (Fig. 5.6), which contains high levels of multiple molecular chaperones (119). There is likely to be a similar defect in Tg folding and dimerization as in cog/cog mice, as the mutation in rat Tg proved to be due to a single G2320R (glycine→arginine) substitution (120), less than 60 residues away from the site of the cog mutation within the AChE-homology domain that blocks Tg secretion (121). The rdw/rdw rats have little thyroid gland enlargement, suggesting that accumulation of the mutant Tg protein may limit thyroid growth (122).

FIGURE 5.6. Electron photomicrograph a thyrocyte of a rdw/rdw rat, showing enlarged endoplasmic reticulum–derived vacuoles that overcrowd the remaining cytoplasm (arrow). Thyrocytes from some humans with congenital goiter due to defective Tg synthesis have a similar appearance (5,600× magnification). (From Sakai Y, Yamashina S, Furudate SI. Missing secretory granules, dilated endoplasmic reticulum, and nuclear dislocation in the thyroid gland of rdw rats with hereditary dwarfism. Anat Rec 2000;259:60, with permission.)

In humans, single Tg amino-acid substitutions involved in congenital hypothyroid goiter have been reported only rarely. One mutation was in the AChE-homology domain (123); others have involved replacement of a conserved cysteine residue. The homozygous C1263R mutant resulted in only mild hypothyroidism, indicating that the mutant Tg protein can be iodinated and serve as a substrate for hormonogenesis (124). Given the importance of intrachain disulfide bonding within Tg (Fig. 5.4), folding defects associated with replacement of single Cys residues are easy to rationalize.

Simple goiter without hypothyroidism has also been associated with production of defective Tg (125), such as the missense mutation resulting in a glutamine to histidine substitution within exon 10 of the Tg gene (126,127). A variant type of adenomatous goiter, without hypothyroidism, has been attributed to the C1995S (cysteine→serine) mutation (128). By analogy to the cog/cog mice, which as adults have normal serum T4 concentrations (108), humans in whom the Tg is only slightly defective may, by increasing the total number of thyrocytes contributing to hormonogenesis, compensate for the reduced efficiency of hormonogenesis within individual thyrocytes, so that euthyroidism is achieved at the expense of goiter.


As Tg migrates to the Golgi complex, roughly two-thirds of the N-linked core glycans become modified to a “complex glycan” structure (129). This includes additional trimming of mannose residues from the core glycan branches by alpha-mannosidase-II and endomannosidase (130), followed by addition to these branches of other carbohydrates (131), including N-acetylglucosamine, galactose, fucose, sialic acid (132,133), or sulfate (134,135) or phosphate (136,137). [Independently of this, sulfation and phosphorylation may also occur directly on the Tg polypeptide backbone (136,138,139)].

In addition, human Tg has been reported to undergo “O-linked” glycosylation; in this case, the carbohydrate is attached to the side chain of serine or threonine (1). Such O-linked carbohydrate has various structures, including typical glycosaminoglycan chains that may also be sulfated (140,141).

Knowledge of these additional post-translational modifications of Tg has helped to identify normal and abnormal Tg maturation at defined points along the secretory pathway. As an example, both protein and carbohydrate sulfation are thought to occur primarily within the trans-Golgi network (142). Therefore, perturbed secretion of Tg that had been radiolabeled with 35S-methionine might represent delayed export from the endoplasmic reticulum, whereas perturbed secretion of Tg that had been radiolabeled with 35S-sulfate might represent a delay in a late Golgi or post-Golgi step of the Tg secretory pathway. A similar investigative tool involves in vitro digestion of Tg with endoglycosidase H (endo H), which will cleave N-linked oligosaccharides representative of the initial core carbohydrate structure within the endoplasmic reticulum. As Tg contains carbohydrates amounting to ~30kDa of its molecular mass, endo H digestion of Tg that has been confined to the endoplasmic reticulum compartment results in a substantial decrease in molecular mass, observed as an increased mobility of the Tg band by SDS-PAGE. By contrast, when the mannose-trimmed core oligosaccharide becomes modified in the Golgi complex by addition of N-acetylglucosamine (the first of the “complex sugars” to be added), that oligosaccharide can no longer be digested by endo H. Consequently, on endo H digestion, Tg that has migrated to the Golgi compartment loses much less molecular mass, corresponding to only the ~one-third of N-linked core glycans that did not become modified to a “complex glycan” structure. The ratio of this endo H-resistant Tg to endo H-sensitive Tg in the steady state, as well as the rate of acquisition of endo H-resistance in kinetic studies, are both useful indicators of the efficiency of Tg export from the endoplasmic reticulum. Typically, in congenital goitrous hypothyroidism due to a mutation in Tg that affects its folding, the Tg protein never acquires endo H-resistance (Fig. 5.7).

FIGURE 5.7. Thyroglobulin (Tg) glycans normally acquire resistance to digestion with endoglycosidase H upon exit from the endoplasmic reticulum, but defective Tg that does not migrate to the Golgi complex remains persistently endo H-sensitive. A: Newly synthesized Tg in normal (Nl) murine thyroid tissue comprises both post-Golgi, endo H-resistant Tg (R) and pre-Golgi, endo H-sensitive Tg (S). In thyrocytes from cog/cog mice the newly synthesized Tg does not acquire endo H-resistance, being comprised entirely of endo H-sensitive molecules. B: Consistent with its inability to exit the endoplasmic reticulum, newly-synthesized Tg from cog/cog thyrocytes in primary culture disappears from the cells over a one-day period (by a process of endoplasmic reticulum-associated degradation) without being detectably secreted into the extracellular space. (From Kim PS, Hossain SA, Park YN, et al. A single amino acid change in the acetylcholinesterase-like domain of thyroglobulin causes congenital goiter with hypothyroidism in the cog/cog mouse: a model of human ER storage diseases. Proc Natl Acad Sci U S A 1998;95:9909, with permission).

The biological functions of the aforementioned post-translational modifications of Tg glycans (or polypeptide) in the Golgi complex are not clear. One hypothesis is that carbohydrate modifications may control the intracellular targeting of Tg in later stages of the intracellular secretory or endocytotic pathways. For example, Tg that is secreted to the apical side of thyrocytes (i.e., the lumenal side) may have in some of its carbohydrate side chains the lysosomal recognition marker mannose-6-phosphate (143). If so, endocytotic degradation of Tg, which is essential for hormonogenesis, might be directed initially by the well-known mannose 6-phosphate receptors (144); however, this hypothesis has not been confirmed. A second hypothesis is that biosynthetic sulfation (or phosphorylation) of tyrosyl residues could alter the capacity of Tg for subsequent iodination and hormonogenesis (145,146). Such a view is not easily reconciled with the very low degree of tyrosine sulfation or phosphorylation in Tg (134), although, theoretically, these ions could be displaced from Tg at the time of iodination or coupling (147). There are other hypotheses (some described below), but the general view of these Golgi-type post-translational modifications is that they have further structural effects on Tg, such as facilitating multimerization (148), and surface charge effects that increase the solubility of Tg (48).

TSH clearly regulates the carbohydrate processing of Tg, affecting both the number of N-linked glycosylation sites that are initially utilized and the subsequent modifications of those glycans to a complex structure (135,149,150). This may include an increase in the number of oligosaccharides bearing galactose and sialic acid, as well as alterations of sulfation. Perhaps the most interesting of these changes is in respect to the domain structure of Tg. Specifically, in region I (the first ~1200 amino acids) of bovine Tg there are 6 sequential N-linked glycans, and these bear complex or hybrid-type oligosaccharides (at positions 91, 464, 476, 835, 928, and 1121) (81). The situation appears essentially identical in human Tg (79). Thus, a TSH-stimulated increase in the number of complex oligosaccharide units is likely to have a more profound structural effect on the C-terminal half of the Tg molecule. Future studies should focus on the last two or three N-glycosylation sites of Tg [i.e., positions 2104 and 2277 of bovine Tg and positions 2275 and 2562 of human Tg, which are ordinarily high mannose units (79,81)], because changes in glycosylation near the C-terminus of Tg might create structural alterations that would increase the exposure of the C-terminal iodination site and thereby help to account for the increase in production of T3 that results from increased TSH secretion or the action of TSH receptor-stimulating antibodies in patients with Graves' thyrotoxicosis.


Thyrocytes are highly polarized epithelial cells with distinct apical and basolateral regions (151). The secretion of newly synthesized proteins by these cells is directed to both regions (152,153), but Tg is transported almost completely through the apical region into the follicular lumen (154). The apical targeting and storage of Tg is known to go awry in Graves' disease, thyroiditis, and many thyroid cancers. Still, very little is known about the molecular mechanisms of sorting that allow for normal Tg targeting to the apical surface of the cells for iodination, even while other proteins are selectively targeted basolaterally (155,156,157,158,159,160). The apical secretory pathway of polarized epithelial cells may be created in the trans-Golgi network by the creation and merging (in the plane of the membrane) of lipid-protein microdomains known as rafts, which serve as a platform upon which apical secretory proteins can be ferried (161). These rafts are especially enriched in glycosphingolipids. Two key features of the rafts are that they are not solubilized at 4°C in the presence of 1% Triton X-100 (unlike other membranes as well as a large fraction of cellular proteins that are soluble in the presence of this non-ionic detergent), and because they are lipid-rich they rise to the top of a sucrose-flotation gradient. Several newly synthesized apical-plasma-membrane proteins are recovered in the raft fraction (162). The raft fraction has been used as a source of antigen to identify novel proteins that may be involved mechanistically in membrane transport to the apical cell surface. One such protein, VIP17 (163), also known as MAL (164), is a membrane protein whose expression in some epithelial cells is correlated with the efficiency of apical-plasma-membrane protein delivery (165). MAL is expressed in human thyrocytes both at the mRNA and protein level (where it is restricted to the apical zone of thyroid follicles), as well as in some thyroid-cell culture lines such as FRT, in which it is recovered in the raft fraction (166). These features raise the possibility of Tg transport via a raft pathway from the Golgi complex to the thyroid follicular lumen (163).

In PC Cl3 thyrocytes, the steady-state pool size of Tg in the Golgi complex or in post-Golgi compartments is small, as compared with that in the endoplasmic reticulum. Nevertheless, from the endo H-resistant Tg pool (i.e., Tg having passed through the Golgi complex), most is recovered in Triton X-100 insoluble rafts, in spite of the fact that secreted Tg is completely soluble in Triton X-100 (167). The endo H-resistant portion of recombinant Tg expressed in heterologous epithelial cells is also recovered in a raft fraction, and depletion of MAL from these cells inhibits apical secretion of Tg (168). These findings suggest that Tg may be actively selected for apical delivery by association with one or more other raft components. Although the mechanism of this selection is unknown, it may be mediated by one or more novel lectins (169). One such lectin isolated from the raft fraction, known as VIP36, specifically binds N-acetylgalactosamine (170,171), raising again the idea that carbohydrate modifications of Tg occurring in the Golgi could serve as recognition signals for post-Golgi protein trafficking. One potentially serious problem with this model is that treatment of thyrocytes with various inhibitors of carbohydrate processing (e.g., castanospermine) does not prevent the apical secretion of Tg (90,172). Additionally, several basolaterally secreted thyroid proteins also have N-linked glycans (55,160), and it remains to be proven that carbohydrate processing of these proteins in the Golgi is appreciably different from that of apically secreted glycoproteins.

The sheer abundance of Tg is likely to stoichiometrically overwhelm the expression level of any putative Tg sorting receptor for apical secretion. However, if Tg were to undergo higher-order self-assembly while still within the Golgi complex (which presently is merely speculation) (148), such a mechanistic problem might be circumvented. In fact, such a mechanism appears to exist in other peptide hormone-producing endocrine cells that must package high concentrations of a subset of newly synthesized protein products into secretory granules (173). Tg has certain features that suggest it might be incorporated into classical secretory granules, including the presence in thyrocytes of Tg secretory compartments that have been described as apical vesicles or secretory granules (174,175). In addition, Tg exocytosis is to some extent acutely stimulated by TSH both in vivo (176) and in primary thyrocyte cultures (156,157,158,160,177). Moreover, in filter-polarized thyrocyte monolayers, certain heterologous (recombinant) peptide hormones are also secreted apically (153). The recent observations that an increasing number of classical secretory granule proteins also may utilize the raft mechanism for their Golgi-based sorting makes such a possibility plausible for Tg as well (172,173,174,175,176,177,178,179,180,181,182).


After exocytosis, especially after stimulation with TSH (175,183), Tg that has fully matured within the secretory pathway undergoes iodination in the lumen of the thyroid follicles. It is nearly impossible to study iodination and hormonogenesis in human Tg in vivo. While this can be done in vivo in animals (especially rodents), there are likely to be differences in Tg structure (e.g., glycosylation) and iodination among different species. Normally (as described above), only a subset of the tyrosine residues of Tg becomes iodinated (184), and the fraction that is iodinated is further diminished in iodide-poor environments (see section on iodine deficiency in Chapter 11).

It is important to understand how the iodination sites are selected, because it has long been recognized that iodination of Tg provides a much higher yield of the hormonally active iodothyronines, T4 and T3, than iodination of any other protein, and that most of the specificity of the iodination sites (and subsequent coupling to form iodothyronines) is somehow encoded in the structure of Tg itself (13).

While denatured Tg can be readily iodinated in vitro, it is no longer competent for hormonogenesis (185,186), indicating that there are stringent steric requirements within Tg that allow thyroid peroxidase to catalyze efficiently not only iodination but also formation of T4 and T3 from the iodotyrosyl residues (13). By contrast, T4 and T3 synthesis in vitro can be catalyzed not only by thyroid peroxidase but also by myeloperoxidase (187) and lactoperoxidase (188), although the efficiency of hormonogenesis and maintenance of the integrity of Tg structure varies with the conditions used. Provided certain critical domains of Tg are in the native state, iodination of isolated (iodine-poor) Tg in vitro probably provides a relatively faithful estimation of the iodotyrosyl selectivity that occurs in vivo (189). Using this methodology, efforts have been directed to explore the hypothesis that hormonogenic sites occur at limited and fixed positions in the Tg protein, involving discrete donor and acceptor sites that have relatively precise physical and spatial relationships. For identification of most sites of iodination and hormonogenesis, there has been good agreement between studies of Tg iodinated in vivo and in vitro. Site identification has rested largely on the sequencing of peptides after limited proteolytic digestion of Tg to find iodotyrosines and iodothyronines and their neighboring amino acids, in order to identify the iodopeptide location within the context of the complete Tg sequence. Because the coupling reaction involves iodophenyl transfer from a donor MIT or DIT residue to an acceptor DIT residue, this method is relatively straightforward for locating the acceptor residues, but it will not identify donor iodotyrosine residues. For donor residue identification, it is necessary to identify the dehydroalanine residues that remain within the polypeptide chain after transfer of the iodophenyl moiety to a DIT residue (190,191,192,193,194).

The conclusion of these studies is that, from the many tyrosines in the Tg molecule, three specific tyrosine residues located near the amino and carboxyl ends of the protein are the primary acceptor sites for hormonogenesis, including positions 5 (“site A”), 2554 (“site B”) and 2747 (“site C”) of human Tg, respectively (195). Site A (Tyr5) is the predominant hormonogenic site. It is the most conserved between species, the first to become iodinated in some species, and normally is reserved for formation of T4 (196,197,198,199,200,201). Site B (Tyr2554) is the second most highly utilized site for T4 formation, comprising one-fourth to one-fifth of the T4 content of Tg (3). Site C (position 2747) tends to be favored for T3 formation [with even greater predominance in certain rodent species (197)], although it may account for as much as one-seventh of T4 formation in human Tg (3). In addition to these three sites there are four ancillary hormonogenic sites, termed sites “D” (position 1291), “G” (position 2568), “N” (position 685), and “R” (equivalent to position 632 although not present in human Tg) that may be favored in certain species under certain conditions, but which tend to be more variable than the three primary iodination sites (3).

The variability of specific tyrosine sites for iodination and hormonogenesis is likely to reflect at least three features: tyrosine exposure on the outer surface of Tg; the amino acids flanking the iodination sites, which may provide a biochemical basis favoring iodination and coupling; and the physical proximity of paired “donor” and “acceptor” iodotyrosines, thereby promoting the coupling reaction. It seems plausible that positioning tyrosines near the ends of the protein is likely to increase their surface exposure, although a definitive statement awaits three-dimensional structural analyses of Tg. As regards flanking information, in vitro iodination of human Tg reveals three patterns of neighboring amino acids that favor both early tyrosine iodination as well as T4 formation: Asp/Glu-Tyr; Ser/Thr-Tyr-Ser; and Glu-X-Tyr (195). With respect to proximity of iodotyrosyl pairs, increasing evidence suggests that the donor sites may be closely positioned within the same regions of Tg as their paired acceptor sites (202). Evidence from in vitro iodination of full-length bovine Tg suggests that Tyr130 is an important outer ring donor for T4 formation at Tyr5 (194). Although not completely physiological (203), coupling between Tyr5 and Tyr130 can also be achieved by in vitro iodination of a fragment containing only the first 171 residues of human Tg (192), which is facilitated by N-glycosylation in this domain (204). Mutagenic replacement of 4 of the 5 other tyrosines in the vicinity with phenylalanine (at positions 29, 89, 97, and 192) does not dminish the content of T4 generated in vitro in such a fragment (205). T4 formation has also been demonstrated in a fragment of rat Tg just large enough to encode iodination sites B and C (206). These findings increase the likelihood that the primary donor and acceptor tyrosine sites are near each other within the same monomer of the Tg homodimer.


The state of the Tg in the colloid may vary, depending upon the levels of TSH and iodination (207,208,209). At the time of exocytosis, Tg is actively iodinated and hormone formation is occurring (210). The proximity of these newly completed Tg molecules to the surface of the apical plasma membrane makes them most readily available for endocytotic internalization. Especially with TSH stimulation (or in Graves' disease), the newly-secreted and iodinated molecules contribute maximally to the pool of iodinated Tg that provides most of the soon-to-be-secreted T4 and T3; this has come to be known as the “last come, first served” hypothesis (211,212).

Tg also undergoes further structural changes upon iodination and storage in colloid (213). As reviewed above, Tg initially forms noncovalently bound homodimers. To the extent that higher order noncovalent Tg complexes are formed (148), they tend to be quantitatively dissociated under the conditions of sample preparation used for the analysis of oligomerization of cellular Tg by sucrose gradient or native gel electrophoresis. Indeed, noncovalently bound Tg homodimers are relatively easily dissociated by heating, changes of ionic strength, or in the presence of increasing concentrations of various detergents. On the other hand, upon TSH stimulation of Tg secretion into colloid (214), and especially in conjunction with iodination, Tg dimers (215) and higher order oligomers (216) are stabilized. The stability of tetrameric Tg (27S by sucrose velocity gradient) is directly correlated with increasing iodine content (3). This could be due to direct structural modification of Tg by iodide, but more likely is due to secondary structural events within the oxidative environment of the colloid; specifically, interchain covalent cross linking primarily via disulfide bonds (49) and secondarily via N-epsilon-amino lysine cross linking (217) and formation of dityrosine bridges (218,219). Such iodide-rich, multimeric Tg is curiously devoid of T4 and T3 (220), possibly because hormone-rich iodopeptides have already been released from the remaining Tg protein by proteolysis (221). In any event, SDS-PAGE analysis of colloidal Tg under reducing conditions reveals the presence of many iodinated Tg fragments, including a variety of overlapping N-terminal fragments that include the primary “site A” for T4formation (189). The generation of these and other Tg fragments within the colloidal space has been attributed both to direct peptide bond cleavage during the oxidative process of iodination (221,222,223), as well as possible cleavage by secreted proteases (224,225,226) with discrete and distinct cleavage sites. Most of these cleavage fragments are likely to remain associated within the parent molecule under nonreduced conditions because of the extensive disulfide bonded structure of Tg.


Tg has the potential to interact with a number of additional proteins, not only within the endoplasmic reticulum but also in later compartments of the secretory pathway. Tg interacts with receptor proteins in the apical plasma membrane, followed by endocytotic internalization. In addition, there may be a sufficiently high concentration of soluble, newly iodinated Tg molecules that an ample number should be available for fluid-phase endocytosis, requiring no receptors, after which they are directed to lysosomes for complete proteolytic digestion, including liberation of T4 and T3 (227). In this case, the high-affinity apical receptors for Tg could be used for apical recycling of Tg, presumably for the purpose of increasing the ultimate efficiency of its iodination (228); for transcytosis to the basolateral (bloodstream) side of the cell (229) [a process which is less easily linked to normal thyroid hormonogenesis (230)]; or for selective uptake of a subset of colloidal Tg molecules en route to lysosomes (231). Given the very large size of Tg, it is not surprising that several peptides derived from the Tg molecule have the capability for binding to surface molecules of thyrocytes and other cells (232). The more difficult question is the physiological importance of such binding (233).

N-acetylglucosamine receptors (228,234,235) may be important in endocytotic routing of poorly-iodinated Tg molecules (236), although recently this notion was modified to suggest modulation or recognition of protein determinants on extracellular Tg (237) by the cell-surface protein disulfide isomerase (PDI) (238). Alternatively, cell-surface PDI-like activity might function in conjunction with the oxidizing environment of the follicular lumen to promote disulfide-linked multimerization of Tg (239); or extracellular PDI might (in conjunction with other extracellularly located chaperones) work oppositely, to solubilize multimeric colloidal Tg (240).

Megalin is another endocytotic receptor for Tg. Initially, it was was thought that megalin-mediated Tg endocytosis might be involved in thyroid hormone synthesis (241). Later studies suggested that Tg does not dissociate from megalin at the low pH of endosomes (242), but rather that megalin binding to Tg facilitates apical-to-basal transcytosis of Tg (243). Addition of receptor-associated protein (RAP) inhibits Tg binding to megalin (241), presumably by a direct interaction between RAP and Tg (244). Tg also contains short heparin-binding sequences, and heparin treatment inhibits Tg binding by megalin (245,246), which may signify a role for one or more heparan-sulfate proteoglycans in the internalization process (247). These experiments have been mostly conducted in cultures of rat thyroid FRTL5 cells, but megalin is expressed in normal thyroid tissue. In rats with aminotriazole-induced goiters, the thyroid content of megalin is increased, suggesting TSH-dependent regulation (243).

Tg binding to the surface of FRTL5 cells has been reported to alter the expression of several thyroid differentiation markers (248,249,250). This effect may be mediated by apical binding of Tg (251), specifically, via an apical version of the asialoglycoprotein receptor (252). Nevertheless, as Tg displays to the cell surface many potential binding sites, it is not clear whether this effect of Tg binding is actually restricted to this receptor, restricted to thyrocytes (253), or requires that Tg be present under native conditions (254). Overall, these studies are intriguing, but need to be pursued further to determine whether Tg is a physiological regulator of the expression of markers of thyroid differentiation in vivo.


Ultimately, Tg is exposed to cathepsins within lysosomes to liberate T4 and T3 from the Tg backbone. Early studies of the endocytotic degradation of Tg were done in collagenase-digested porcine thyroid follicles using [125I]Tg (255). Although confounded by the possibility of Tg binding to the basolateral surface (256), Western blotting with anti-Tg of lysosomes isolated from the thyrocytes revealed slightly smaller-than-normal dimeric Tg which dissociated into fragments upon SDS-PAGE under reducing conditions (257), consistent with early limited proteolysis. Indeed, extracts of thyroid lysosomes can themselves generate such fragments from pure Tg (258). It is difficult to establish the fraction of such fragments that are generated before endocytosis (224,225) versus the fraction generated within the endosome-lysosome system, especially because lysosomes themselves may, under stimulation, undergo exocytosis (226).

In vitro analyses have established that cysteine proteinases mediate proteolytic processing of Tg (3). Individual Tg type 1 repeats are themselves substrates for lysosomal cysteine proteases (33), even as they may function as inhibitors of these enzymes (34). Studies in mice with genetic deficiency of various cathepsins (B, K, or L) have revealed impaired proteolysis of Tg in mice deficient for cathepsin B or cathepsin L, but not cathepsin K-deficient mice (259). Of particular interest, cathepsin K(-/-)/L(-/-) double-mutant mice had low serum free T4 concentrations, suggesting that either or both of these enzymes may participate in liberation of T4 from the Tg backbone.


The authors acknowledge funding from NIH DK40344 and from Ministero dellâ Universitê Ricerca Scientifica Grant No. 2002063745. The authors also thank Drs. E. Consiglio, S. Formisano, Y. N. Park, and P. Kim for helpful discussions and support.


1. Spiro M. Synthesis and processing of thyroglobulin carbohydrate units. In: MC Eggo, GN Burrow, eds. Thyroglobulin: the prothyroid hormone. New York: Raven Press, 1985;103.

2. Mercken L, Simons MJ, Swillens S, et al. Primary structure of bovine thyroglobulin deduced from the sequence of its 8,431-base complementary DNA. Nature 1985;316:647.

3. Dunn JT, Dunn AD. Thyroglobulin: chemistry, biosynthesis, and proteolysis. In: LE Braverman, RD Utiger, eds. Werner & Ingbar's the thyroid: a fundamental and clinical text. Philadelphia: Lippincott Williams & Wilkins, 2000:91.

4. Brisson A, Marchelidon J, Lachiver F. Comparative studies on the amino acid composition of thyroglobulins from various lower and higher vertebrates: phylogenetic aspect. Comp Biochem Physiol B Biochem Mol Biol 1974;49:51.

5. Shepard TH. Development of the human fetal thyroid. Gen Comp Endocrinol 1968;10:174.

6. Yap AS, Stevenson BR, Keast JR, et al. Cadherin-mediated adhesion and apical membrane assembly define distinct steps during thyroid epithelial polarization and lumen formation. Endocrinology 1995;136:4672.

7. Alluchon-Gerard MJ. Morphogenese ultrastructurale et differenciation fonctionnelle du follicule thyroidien de la Rousette. Arch Anat Microsc Morphol Exp 1979;68:43.

8. Ericson L. Intracellular lumens in thyroid follicle cells of thyroxine-treated rats. J Ultrastruct Res 1979;69:297.

9. Nilsson M. Iodide handling by the thyroid epithelial cell. Exp Clin Endocrinol Diabetes 2001;109:13.

10. Suzuki S, Kondo Y. Demonstration of thyroglobulin-like iodinated proteins in the branchial sac of tunicates. Gen Comp Endocrinol 1971;17:402.

11. Thorndyke MC. Evidence for a ‘mammalian’ thyroglobulin in endostyle of the ascidian Styela clava. Nature 1978;271:61.

12. Suzuki S, Gorbman B, Rolland M, and et al. Thyroglobulins of cyclostomes and an elasmobranch. Gen Comp Endocrinol 1975; 26:59.

13. Lissitzky S. Biosynthesis and secretion of thyroglobulin. Ann Endocrinol (Paris) 1981;42:363.

14. de Vijlder JJ, van Ommen GJ, van Voorthuizen WF, et al. Non-functional thyroglobulin messenger RNA in goats with hereditary congenital goiter. J Mol Appl Genet 1981;1:51.

15. Malthiery Y, Marriq C, Berge-Lefranc JL, et al. Thyroglobulin structure and function: recent advances. Biochimie 1989;71:195.

16. Mendive FM, Rivolta CM, Vassart G, et al. Genomic organization of the 3′ region of the human thyroglobulin gene. Thyroid 1999;9:903.

17. Ieiri T, Cochaux P, Targovnik HM, et al. A 3′ splice site mutation in the thyroglobulin gene responsible for congenital goiter with hypothyroidism. J Clin Invest 1991;88:1901.

18. Bertaux F, Noel M, Lasmoles F, et al. Identification of the exon structure and four alternative transcripts of the thyroglobulin-encoding gene. Gene 1995;156:297.

19. Targovnik HM, Rivolta CM, Mendive FM, et al. Congenital goiter with hypothyroidism caused by a 5′ splice site mutation in the thyroglobulin gene. Thyroid 2001;11:685.

20. Graves PN, Davies TF. A second thyroglobulin messenger RNA species (rTg-2) in rat thyrocytes. Mol Endocrinol 1990;4:155.

21. van de Graaf SA, Ris-Stalpers C, Pauws E, et al. Up to date with human thyroglobulin. J Endocrinol 2001;170:307.

22. van de Graaf SA, Cammenga M, Ponne NJ, et al. The screening for mutations in the thyroglobulin cDNA from six patients with congenital hypothyroidism. Biochimie 1999;81:425.

23. Marians RC, Ng L, Blair HC, et al. Defining thyrotropin-dependent and -independent steps of thyroid hormone synthesis by using thyrotropin receptor-null mice. Proc Natl Acad Sci U S A 2002;99:15776.

24. Damante G, Tell G, Di Lauro R. A unique combination o f transcription factors controls differentiation of thyroid cells. Prog Nucleic Acid Res Mol Biol 2001;66:307.

25. Pasca di Magliano M, Di Lauro R, Zannini M. Pax8 has a key role in thyroid cell differentiation. Proc Natl Acad Sci U S A 2000;97:13144.

26. Mascia A, Nitsch L, Di Lauro R, et al. Hormonal control of the transcription factor Pax8 and its role in the regulation of thyroglobulin gene expression in thyroid cells. J Endocrinol 2002; 172:163.

27. Targovnik H, Propato F, Varela V, et al. Low levels of thyroglobulin messenger ribonucleic acid in congenital goitrous hypothyroidism with defective thyroglobulin synthesis. J Clin Endocrinol Metab 1989;69:1137.

28. Targovnik, HM, Medeiros-Neto G, Varela V, et al. A nonsense mutation causes human hereditary congenital goiter with preferential production of a 171-nucleotide-deleted thyroglobulin ribonucleic acid messenger. J Clin Endocrinol Metab 1993;77:210.

29. Gonzalez-Sarmiento R, Corral J, Mories MT, et al. Monoallelic deletion in the 5′ region of the thyroglobulin gene as a cause of sporadic nonendemic simple goiter. Thyroid 2001;11:789.

30. van de Graaf SA, Pauws E, de Vijlder JJ, et al. The revised 8307 base pair coding sequence of human thyroglobulin transiently expressed in eukaryotic cells. Eur J Endocrinol 1997;136:508.

31. Muresan Z, Arvan P. Thyroglobulin transport along the secretory pathway. Investigation of the role of molecular chaperone, GRP94, in protein export from the endoplasmic reticulum. J Biol Chem 1997;272:26095.

32. Berg G, Ekholm R. Electron microscopy of low iodinated thyroglobulin molecules. Biochim Biophys Acta 1975;386:422.

33. Pungercic G, Dolenc I, Dolinar M, et al. Individual recombinant thyroglobulin type-1 domains are substrates for lysosomal cysteine proteinases. Biol Chem 2002;383:1809.

34. Galesa K, Pain R, Jongsma MA, et al. Structural characterization of thyroglobulin type-1 domains of equistatin. FEBS Lett 2003;539:120.

35. Durkin ME, Chakravarti S, Bartos BB, et al. Amino acid sequence and domain structure of entactin. Homology with epidermal growth factor precursor and low density lipoprotein receptor. J Cell Biol 1988;107:2749.

36. Neumann GM, Bach LA. The N-terminal disulfide linkages of human insulin-like growth factor-binding protein-6 (hIGFBP-6) and hIGFBP-1 are different as determined by mass spectrometry. J Biol Chem 1999;274:14587.

37. Chong JM, Speicher DW. Determination of disulfide bond assignments and N-glycosylation sites of the human gastrointestinal carcinoma antigen GA733–2 (CO17–1A, EGP, KS1–4, KSA, and Ep-CAM). J Biol Chem 2001;276:5804.

38. Mercken L, Simons MJ, De Martynoff G, et al. Presence of hormonogenic and repetitive domains in the first 930 amino acids of bovine thyroglobulin as deduced from the cDNA sequence. Eur J Biochem 1985;147:59.

39. Guncar G, Pungercic G, Klemencic I, et al. Crystal structure of MHC class II-associated p41 Ii fragment bound to cathepsin L reveals the structural basis for differentiation between cathepsins L and S. EMBO J 1999;18:793.

40. Gentile F, Salvatore G. Preferential sites of proteolytic cleavage of bovine, human, and rat thyroglobulin. The use of limited proteolysis to detect solvent exposed regions of the primary structure. Eur J Biochem 1993;218:603.

41. Swillens S, Ludgate M, Mercken L, et al. Analysis of sequence and structure homologies between thyroglobulin and acetylcholinesterase: possible functional and clinical significance. Biochem Biophys Res Commun 1986;137:142.

42. Mori N, Itoh N, Salvaterra PM. Evolutionary origin of cholinergic macromolecules and thyroglobulin. Proc Natl Acad Sci U S A 1987;84:2813.

43. MacPhee-Quigley K, Vedvick TS, Taylor P, et al. Profile of the disulfide bonds in acetylcholinesterase. J Biol Chem 1986;261: 13565.

44. Di Jeso B, Ulianich L, Pacifico F, et al. The folding of thyroglobulin in the calnexin/calreticulin pathway and its alteration by a loss of Ca2+ from the endoplasmic reticulum. Biochem J 2003;370:449.

45. Veneziani BM, Giallauria F, Gentile F. The disulfide bond pattern between fragments obtained by the limited proteolysis of bovine thyroglobulin. Biochimie 1999;81:517.

46. Andreoli M, Sena L, Edelhoch H, et al. The noncovalent subunit structure of human thyroglobulin. Arch Biochem Biophys 1969;134:242.

47. Schneider AB, Bornet H, Edelhoch H. The effects of low temperature on the conformation of thyroglobulin. J Biol Chem 1971;246:2672.

48. Tarutani O, Kondo T, Shulman S. Properties of carbohydrate-stripped thyroglobulin. III. Solubility characteristics of thyroglobulin. Biochim Biophys Acta 1977;492:284.

49. Berndorfer U, Wilms H, Herzog V. Multimerization of thyroglobulin (TG) during extracellular storage: isolation of highly cross-linked TG from human thyroids. J Clin Endorinol Metab 1996;81:1918.

50. Stroud RM, Walter P. Signal sequence recognition and protein targeting. Curr Opin Struct Biol 1999;9:754.

51. Sanders SL, Whitfield KM, Vogel JP, et al. Sec61p and BiP directly facilitate polypeptide translocation into the ER. Cell 1992;69:353.

52. Rutkowski T, Lingappa VR. Membrane targeting of proteins. In: Cell Biology www.ergito.com. 2002, Chapter 2.

53. Huang XF, Arvan P. Intracellular transport of proinsulin in pancreatic b-cells: structural maturation probed by disulfide accessibility. J Biol Chem 1995;270:20417.

54. Kim PS, Arvan P. Folding and assembly of newly synthesized thyroglobulin occurs in a pre-Golgi compartment. J Biol Chem 1991;266:12412.

55. Prabakaran D, Kim PS, Dixit VM, et al. Oligomeric assembly of thrombospondin in the endoplasmic reticulum of thyroid epithelial cells. Eur J Cell Biol 1996;70:134.

56. Netzer WJ, Hartl FU. Recombination of protein domains facilitated by co-translational folding in eukaryotes. Nature 1997; 388:343.

57. Kim P, Bole D, Arvan P. Transient aggregation of nascent thyroglobulin in the endoplasmic reticulum: relationship to the molecular chaperone, BiP. J Cell Biol 1992;118:541.

58. Kim PS, Kim KR, Arvan P. Disulfide-linked aggregation of thyroglobulin normally occurs during nascent protein folding. Am J Physiol 1993;265:C704.

59. Mazzarella RA, Srinivasan M, Haugejorden SM, et al. ERp72, an abundant luminal endoplasmic reticulum protein, contains three copies of the active site sequences of protein disulfide isomerase. J Biol Chem 1990;265:1094.

60. Rupp K, Birnbach U, Lundstrom J, et al. Effects of CaBP2, the rat analog of ERp72, and of CaBP1 on the refolding of denatured reduced proteins. Comparison with protein disulfide isomerase. J Biol Chem 1994;269:2501.

61. Molinari M, Helenius A. Glycoproteins form mixed disulphides with oxidoreductases during folding in living cells. Nature 1999;402:90.

62. Anelli T, Alessio M, Mezghrani A, et al. ERp44, a novel endoplasmic reticulum folding assistant of the thioredoxin family. EMBO J 2002;21:835.

63. Munro S, Pelham HR. An hsp70-like protein in the ER: identity with the 78kd glucose-regulated protein and immunoglobulin heavy chain binding protein. Cell 1986;46:291.

64. Kim PS, Arvan P. Hormonal regulation of thyroglobulin export from the endoplasmic reticulum of cultured thyrocytes. J Biol Chem 1993;268:4873.

65. Kim PS, Arvan P. Calnexin and BiP act as sequential molecular chaperones during thyroglobulin folding in the endoplasmic reticulum. J Cell Biol 1995;128:29.

66. Nigam SK, Goldberg AL, Ho S, et al. A set of endoplasmic reticulum proteins possessing properties of molecular chaperones includes Ca2+ binding proteins and members of the thioredoxin superfamily. J Biol Chem 1994;269:1744.

67. Kuznetsov G, Chen LB, Nigam SK. Several endoplasmic reticulum stress proteins, including ERp72, interact with thyroglobulin during its maturation. J Biol Chem 1994;269:22990.

68. Kuznetsov G, Bush KT, Zhang PL, et al. Perturbations in maturation of secretory proteins and their association with endoplasmic reticulum chaperones in a cell culture model for epithelial ischemia. Proc Natl Acad Sci U S A 1996;93:8584.

69. Kuznetsov G, Chen LB, Nigam SK. Multiple molecular chaperones complex with misfolded large oligomeric glycoproteins in the endoplasmic reticulum. J Biol Chem 1997;272:3057.

70. Kaufman RJ, Scheuner D, Schroder M, et al. The unfolded protein response in nutrient sensing and differentiation. Nat Rev Mol Cell Biol 2002;3:411.

71. Spiro RG, Spiro MJ, Bhoyroo VD. Studies on the regulation of the biosynthesis of glucose-containing oligosaccharide-lipids. J Biol Chem 1983;258:9469.

72. Lee AS. Mammalian stress response: induction of the glucose-regulated protein family. Curr Opin Cell Biol 1992;4:267.

73. Leonardi A, Vito P, Mauro C, et al. Endoplasmic reticulum stress causes thyroglobulin retention in this organelle and triggers activation of nuclear factor-kappa B via tumor necrosis factor receptor-associated factor 2. Endocrinology 2002;143: 2169.

74. Muresan Z, Arvan P. Enhanced binding of the molecular chaperone, BiP, slows thyroglobulin export from the endoplasmic reticulum. Mol Endocrinol 1998;12:458.

75. Rose MD, Misra LM, Vogel JP. KAR2, a karyogamy gene, is the yeast homolog of the mammalian BiP/GRP78 gene. Cell 1989;57:1211.

76. Kaufman RJ. Orchestrating the unfolded protein response in health and disease. J Clin Invest 2002;110:1389.

77. Harding HP, Calfon M, Urano F, et al. Transcriptional and translational control in the mammalian unfolded protein response. Annu Rev Cell Dev Biol 2002;18:575.

78. Sargsyan E, Baryshev M, Szekely L, et al. Identification of ERp29, an endoplasmic reticulum lumenal protein, as a new member of the thyroglobulin folding complex. J Biol Chem 2002;277:17009.

79. Yang SX, Pollock HJ, Rawitch AB. Glycosylation in human thyroglobulin: location of the N-linked oligosaccharide units and comparison with bovine thyroglobulin. Arch Biochem Biophys 1996;327:61.

80. Franc JL, Mallet B, Lanet J, et al. The number of oligosaccharides borne by porcine thyroglobulin is variable. Endocrinology 1994;134:885.

81. Rawitch AB, Pollock HG, Yang SX. Thyroglobulin glycosylation: location and nature of the N-linked oligosaccharide units in bovine thyroglobulin. Arch Biochem Biophys 1993;300: 271.

82. Helenius A, Trombetta ES, Hebert DN, et al. Calnexin, calreticulin and the folding of glycoproteins. Trends Cell Biol 1997; 7:193.

83. Wada I, Imai S, Kai M, et al. Chaperone function of calreticulin when expressed in the endoplasmic reticulum as the membrane-anchored and soluble forms. J Biol Chem 1995;270: 20298.

84. Parodi AJ, Mendelzon DH, Lederkremer GZ. Transient glucosylation of protein-bound Man9GlcNAc2, Man8GlcNAc2, and Man7GlcNAc2 in calf thyroid cells. A possible recognition signal in the processing of glycoproteins. J Biol Chem 1983; 258:8260.

85. Trombetta SE, Ganan SA, Parodi AJ. The UDP-Glc-glycoprotein glucosyltransferase is a soluble protein of the endoplasmic reticulum. Glycobiology 1991;1:155.

86. Sousa MC, Ferrero-Garcia MA, Parodi AJ. Recognition of the oligosaccharide and protein moieties of glycoproteins by the UDP-Glc:glycoprotein glucosyltransferase. Biochemistry 1992; 31:97.

87. Sousa M, Parodi AJ. The molecular basis for the recognition of misfolded glycoproteins by the UDP-Glc:glycoprotein glucosyltransferase. EMBO J 1995;14:4196.

88. Chen W, Helenius J, Braakman I, et al. Cotranslational folding and calnexin binding during glycoprotein synthesis. Proc Natl Acad Sci U S A 1995;92:6229.

89. Oliver JD, van der Wal FJ, Bulleid NJ, et al. Interaction of the thiol-dependent reductase ERp57 with nascent glycoproteins. Science 1997;275:86.

90. Franc JL, Giraud A, Lanet J. Effects of deoxymannojirimycin and castanospermine on the polarized secretion of thyroglobulin. Endocrinology 1990;126:1464.

91. Di Jeso B, Pereira R, Consiglio E, et al. Demonstration of a Ca2+ requirement for thyroglobulin dimerization and export to the Golgi complex. Eur J Biochem 1998;252:583.

92. Eggo MC, Burrow GN. Glycosylation of thyroglobulin—its role in secretion, iodination, and stability. Endocrinology 1983; 113:1655.

93. Mallet B, Lejeune PJ, Baudry N, et al. N-glycans modulate in vivo and in vitro thyroid hormone synthesis. Study at the N-terminal domain of thyroglobulin. J Biol Chem 1995;270: 29881.

94. Formisano S, Di Jeso B, Acquaviva R, et al. Calcium-induced changes in thyroglobulin conformation. Arch Biochem Biophys 1983;227:351.

95. Choudhury P, Liu Y, Bick RJ, et al. Intracellular association between UDP-glucose:glycoprotein glucosyltransferase and an incompletely folded variant of alpha1-antitrypsin. J Biol Chem 1997;272:13446.

96. Di Jeso B, Formisano S, Ulianich L. Perturbation of cellular calcium delays the secretion and alters the glycosylation of thyroglobulin in FRTL-5 cells. Biochem Biophys Res Commun 1997;234:133.

97. Di Jeso B, Formisano S, Consiglio E. Depletion of divalent cations within the secretory pathway inhibits the terminal glycosylation of complex carbohydrates of thyroglobulin. Biochimie 1999;81:497.

98. Riesco G, Bernal J, Sanchez-Franco F. Thyroglobulin defect in a human congenital goiter. J Clin Endocrinol Metab 1974; 38:33.

99. Monaco F, Grimaldi S, Dominici R, et al. Defective thyroglobulin synthesis in an experimental rat thyroid tumor: iodination and thyroid hormone synthesis in isolated tumor thyroglobulin. Endocrinology 1975;97:347

100. Michel-Bechet M, Cotte G, Codaccioni JL, et al. Ultrastructure thyroidienne et perturbations biochimiques de l'hormonogenese. Acta Anat (Basel) 1969;73:389.

101. Medeiros-Neto G, Targovnik H, Knobel M, et al. Qualitative and quantitative defects of thyroglobulin resulting in congenital goiter. Absence of gross gene deletion of coding sequences in the TG gene structure. J Endocrinol Invest 1989;12:805.

102. Medeiros-Neto G, Kim PDS, Yoo SE, et al. Congenital hypothyroid goiter with deficient thyroglobulin. Identification of an endoplasmic reticulum storage disease (ERSD) with induction of molecular chaperones. J Clin Invest 1996;98:2838.

103. Lissitzky S, Torresani J, Burrow GN, et al. Defective thyroglobulin export as a cause of congenital goiter. Clin Endocrinol (Oxf) 1975;4:363.

104. Baas F, Bikker H, van Ommen GJ, et al. Unusual scarcity of restriction site polymorphism in the human thyroglobulin gene. A linkage study suggesting autosomal dominance of a defective thyroglobulin allele. Hum Genet 1984;67:301.

105. Cabrer B, Brocas H, Perez-Castillo A, et al. Normal level of thyroglobulin messenger ribonucleic acid in a human congenital goiter with thyroglobulin deficiency. J Clin Endocrinol Metab 1986;63:931.

106. Ricketts MH, Simons MJ, Parma J, et al. A nonsense mutation causes hereditary goitre in the Afrikander cattle and unmasks alternative splicing of thyroglobulin transcripts. Proc Natl Acad Sci U S A 1987;84:3181.

107. Sugisaki T, Beamer WG, Noguchi T. Microcephalic cerebrum with hypomyelination in the congenital goiter mouse (cog). Neurochem Res 1992;17:1037.

108. Adkison LR, Taylor S, Beamer WG. Mutant gene-induced disorders of structure, function and thyroglobulin synthesis in congenital goitre (cog/cog) in mice. J Endocrinol 1990;126: 51.

109. Basche M, Beamer WG, Schneider AB. Abnormal properties of thyroglobulin in mice with inherited congenital goiter (cog/cog). Endocrinology 1989;124:1822.

110. Fogelfeld L, Harel G, Beamer WG, et al. Low-molecular-weight iodoproteins in the congenital goiters of cog/cog mice. Thyroid 1992;2:329.

111. Kim PS, Kwon OY, Arvan P. An endoplasmic reticulum storage disease causing congenital goiter with hypothyroidism. J Cell Biol 1996;133:517.

112. Kim PS, Hossain SA, Park YN, et al. A single amino acid change in the acetylcholinesterase-like domain of thyroglobulin causes congenital goiter with hypothyroidism in the cog/cog mouse: a model of human ER storage diseases. Proc Natl Acad Sci U S A 1998;95:9909.

112A. Tokunaga F, Brostrom C, Kiode T, et al. Endoplasmic reticulum (ER)-associated degradation of misfolded N-linked glycoproteins is suppressed upon inhibition of ER mannosidase I. J Biol Chem 2000;275:40757.

113. Hosokawa N, Wada I, Hasegawa K, et al. A novel ER alpha-mannosidase-like protein accelerates ER-associated degradation. EMBO Rep 2001;2:415.

114. Koto M, Sato T, Okamoto M, et al. rdw rats; a new hereditary dwarf model in the rat. Jikken Dobutsu 1988;37:21.

115. Umezu M, Fujimura T, Sugawara S, et al. Pituitary and serum levels of prolactin (PRL), thyroid stimulating hormone (TSH) and serum thyroxine (T4) in hereditary dwarf rats (rdw/rdw). Jikken Dobutsu 1993;42:211.

116. Ono M, Harigai T, Furudate S. Pituitary-specific transcription factor Pit-1 in the rdw rat with growth hormone- and prolactin-deficient dwarfism. J Endocrinol 1994;143:479.

117. Umezu M, Kagabu S, Jiang J, et al. Evaluation and characterization of congenital hypothyroidism in rdw dwarf rats. Lab Anim Sci 1998;48:496.

118. Sakai Y, Yamashina S, Furudate SI. Missing secretory granules, dilated endoplasmic reticulum, and nuclear dislocation in the thyroid gland of rdw rats with hereditary dwarfism. Anat Rec 2000;259:60.

119. Oh-Ishi M, Omori A, Kwon JY, et al. Detection and identification of proteins related to the hereditary dwarfism of the rdw rat. Endocrinology 1998;139:1288.

120. Hishinuma A, Furudate S, Oh-Ishi M, et al. A novel missense mutation (G2320R) in thyroglobulin causes hypothyroidism in rdw rats. Endocrinology 2000;141:4050.

121. Kim PS, Ding M, Menon S, et al. A missense mutation G2320R in the thyroglobulin gene causes non-goitrous congenital primary hypothyroidism in the WIC-rdw rat. Mol Endocrinol 2000;14:1944.

122. Kim PS, Arvan P. Endocrinopathies in the family of endoplasmic reticulum (ER) storage diseases: disorders of protein trafficking and the role of ER molecular chaperones. Endocr Rev 1998;19:173.

123. Caron P, Moya CM, Malet D, et al. Compound heterozygous mutations in the thyroglobulin gene (1143delC and 6725G—>A [R2223H]) resulting in fetal goitrous hypothyroidism. J Clin Endocrinol Metab 2003;88:3546.

124. Hishinuma A, Kasai K, Masawa N, et al. Missense mutation (C1263R) in the thyroglobulin gene causes congenital goiter with mild hypothyroidism by impaired intracellular transport. Endocr J 1998;45:315.

125. Ohyama Y, Hosoya T, Kameya T, et al. Congenital euthyroid goitre with impaired thyroglobulin transport. Clin Endocrinol (Oxf) 1994;41:129.

126. Corral J, Martin C, Perez R, et al. Thyroglobulin gene point mutation associated with non-endemic simple goitre. Lancet 1993;341:462.

127. Perez-Centeno C, Gonzalez-Sarmiento R, Mories MT, et al. Thyroglobulin exon 10 gene point mutation in a patient with endemic goiter. Thyroid 1996;6:423.

128. Hishinuma A, Takamatsum J, Ohyama Y, et al. Two novel cysteine substitutions (C1263R and C1995S) of thyroglobulin cause a defect in intracellular transport of thyroglobulin in patients with congenital goiter and the variant type of adenomatous goiter. J Clin Endocrinol Metab 1999;84:1438.

129. Ronin C, Fenouillet E, Hovsepian S, et al. Biosynthesis of thyroglobulin carbohydrate chains. In: Eggo MC, Burrow GN, eds. Thyroglobulin: the prothyroid hormone. New York: Raven Press, 1985:95.

130. Zuber C, Spiro MJ, Guhl B, et al. Golgi apparatus immunolocalization of endomannosidase suggests post-endoplasmic reticulum glucose trimming: implications for quality control. Mol Biol Cell 2000;11:4227.

131. Arima T, Spiro MJ, Spiro RG. Studies on the carbohydrate units of thyroglobulin. J Biol Chem 1972;247:1825.

132. Yamamoto K, Tsuji T, IrimuraT, et al. The structure of carbohydrate unit B of porcine thyroglobulin. Biochem J 1981;195: 701.

133. Grollman EF, Saji M, Shimura Y, et al. Thyrotropin regulation of sialic acid expression in rat thyroid cells. J Biol Chem 1993;268:3604.

134. Spiro MJ, Spiro RG. Biosynthesis of sulfated asparagine-linked complex carbohydrate units of calf thyroglobulin. Endocrinology 1988;123:56.

135. Desruisseau S, Franc JL, Gruffat D, et al. Glycosylation of thyroglobulin secreted by porcine cells cultured in chamber system: thyrotropin controls the number of oligosaccharides and their anionic residues. Endocrinology 1994;134:1676.

136. Consiglio E, Acquaviva AM, Formisano S, et al. Characterization of phosphate residues on thyroglobulin. J Biol Chem 1987;262:10304.

137. Sakurai S, Fogelfeld L, Ries A, et al. Anionic complex-carbohydrate units of human thyroglobulin. Endocrinology 1990;127: 2056.

138. Herzog V. Secretion of sulfated thyroglobulin. Eur J Cell Biol 1986;39:399.

139. Spiro M J, Gorski KM. Studies on the posttranslational migration and processing of thyroglobulin: use of inhibitors and evaluation of the role of phosphorylation. Endocrinology 1986; 119:1146.

140. Schneider AB, McCurdy A, Chang T, et al. Metabolic labeling of human thyroglobulin with [35S]sulfate: incorporation into chrondroitin 6-sulfate and endoglycosidase-F-susceptible carbohydrate units. Endocrinology 1988;122:2428.

141. Fogelfeld L, Schneider AB. Inhibition of chondroitin sulfate incorporation into human thyroglobulin by p-nitrophenyl-b-D-xylopyranoside. Endocrinology 1990;126:1064.

142. Baeuerle PA, Huttner WB. Tyrosine sulfation is a trans-Golgi-specific protein modification. J Cell Biol 1987;105:2655.

143. Herzog V, Neumuller W, Holzmann B. Thyroglobulin, the major and obligatory exportable protein of thyroid follicle cells, carries the lysosomal recognition marker mannose-6-phosphate. EMBO J 1987;6:555.

144. Kasper D, Dittmer F, von Figura K, et al. Neither type of mannose 6-phosphate receptor is sufficient for targeting of lysosomal enzymes along intracellular routes. J Cell Biol 1996;134: 6150623.

145. Nlend MC, Cauvi D, Venot N, et al. Sulfated tyrosines of thyroglobulin are involved in thyroid hormone synthesis. Biochem Biophys Res Commun 1999;262:193.

146. Nlend MC, Cauvi D, Venot N, et al. Thyrotropin regulates tyrosine sulfation of thyroglobulin. Eur J Endocrinol 1999;141: 61.

147. Venot N, Nlend MC, Cauvi D, et al. The hormonogenic tyrosine 5 of porcine thyroglobulin is sulfated. Biochem Biophys Res Commun 2002;298:193.

148. Shifrin S, Consiglio E, Kohn LD. Effect of the complex carbohydrate moiety on the structure of thyroglobulin. J Biol Chem 1983;258:3780.

149. Ronin C, Fenouillet E, Hovsepian S, et al. Regulation of thyroglobulin glycosylation. J Biol Chem 1986;261:7287.

150. Di Jeso B, Liguoro D, Ferranti P, et al. Modulation of the carbohydrate moiety of thyroglobulin by thyrotropin and calcium in Fisher rat thyroid line-5 cells. J Biol Chem 1992;267:1938.

151. Chambard M, Verrier B, Gabrion J, et al. Polarization of thyroid cells in culture: evidence for the basolateral localization of the iodide “pump” and of the thyroid-stimulating hormone receptor-adenyl cyclase complex. J Cell Biol 1983;96:1172.

152. Mauchamp J, Chambard M, Verrier B, et al. Epithelial cell polarization in culture: orientation of cell polarity and expression of specific functions, studied with cultured thyroid cells. J Cell Sci Suppl 1987;8:345.

153. Prabakaran D, Ahima RS, Harney JW, et al. Polarized targeting of epithelial cell proteins in thyrocytes and MDCK cells. J Cell Sci 1999;112:1247.

154. Bjorkman U, Ekholm R, Elmqvist LG, et al. Induced unidirectional transport of protein into the thyroid follicular lumen. Endocrinology 1974;95:1506.

155. Ericson LE. Exocytosis and endocytosis in the thyroid follicle cells. Mol Cell Endocrinol 1981;22:1.

156. Chambard M, Mauchamp J, Chabaud O. Synthesis and apical and basolateral secretion of thyroglobulin by thyroid cell monolayers on permeable substrate: modulation by thyrotropin. J Cell Physiol 1987;133:37.

157. Chambard M, Depetris D, Gruffat D, et al. Thyrotrophin regulation of apical and basal exocytosis of thyroglobulin by porcine thyroid monolayers. J Mol Endocrinol 1990;4:193.

158. Arvan P, Lee J. Regulated and constitutive protein targeting can be distinguished by secretory polarity in thyroid epithelial cells. J Cell Biol 1991;112:365.

159. Desruisseau-Gonzalvez S, Delori P, Gruffat D, et al. Polarized secretion of tissue-plasminogen activator in cultured thyroid cells. Vitro Cell Dev Biol-Animal 1993;29:161.

160. Prabakaran D, Kim PS, Kim KR, et al. Polarized secretion of thrombospondin is opposite to thyroglobulin in thyroid epithelial cells. J Biol Chem 1993;268:9041.

161. Scheiffele P, Roth MG, Simons K. Interaction of influenza virus haemagglutinin with sphingolipid-cholesterol membrane domains via its transmembrane domain. EMBO J 1997;16: 5501.

162. Danielsen EM. Involvement of detergent-insoluble complexes in the intracellular transport of intestinal brush border enzymes. Biochemistry 1995;34:1596.

163. Cheong KH, Zacchetti D, Schneeberger EE, et al VIP17/MAL, a lipid raft-associated protein, is involved in apical transport in MDCK cells. Proc Natl Acad Sci U S A 1999;96: 6241.

164. Puertollano R, Alonso MA. MAL, an integral element of the apical sorting machinery, is an itinerant protein that cycles between the trans-Golgi network and the plasma membrane. Mol Biol Cell 1999;10:3435.

165. Puertollano R, Martin-Belmonte F, Millan J, et al. The MAL proteolipid is necessary for normal apical transport and accurate sorting of the influenza virus hemagglutinin in Madin-Darby canine kidney cells. J Cell Biol 1999;145:141.

166. Martin-Belmonte F, Kremer L, Albar JP, et al. Expression of the MAL gene in the thyroid: the MAL proteolipid, a component of glycolipid-enriched membranes, is apically distributed in thyroid follicles. Endocrinology 1998;139:2077.

167. Martin-Belmonte F, Alonso MA, Zhang X, et al. Thyroglobulin is selected as luminal protein cargo for apical transport via detergent-resistant membranes in epithelial cells. J Biol Chem 2000;275:41074.

168. Martin-Belmonte F, Arvan P, Alonso MA. MAL mediates apical transport of secretory proteins in polarized epithelial Madin-Darby canine kidney cells. J Biol Chem 2001;276: 49337.

169. Scheiffele P, Peranen J, Simons K. N-glycans as apical sorting signals in epithelial cells. Nature 1995;378:96.

170. Fiedler K, Parton RG, Kellner R, et al. VIP36, a novel component of glycolipid rafts and exocytic carrier vesicles in epithelial cells. EMBO J 1994;13:1729.

171. Fiedler K, Simons K. Characterization of VIP36, an animal lectin homologous to leguminous lectins. J Cell Sci 1996;109:271.

172. Franc JL, Hovsepian S, Fayet G, et al. Inhibition of N-linked oligosaccharide processing does not prevent the secretion of thyrogloblulin. A study with swainsonine and deoxynorjirimicin. Eur J Biochem 1986;157:225.

173. Arvan P, Zhang BY, Feng L, et al. Lumenal protein multimerization in the distal secretory pathway/secretory granule. Curr Opin Cell Biol 2002;14:448.

174. Novikoff AB, Novikoff PM, Ma M, et al. Cytochemical studies of secretory and other granules associated with the endoplasmic reticulum in rat thyroid epithelial cells. In: Ceccarelli B, Clementi F, Meldolesi J, eds. Advances in cytopharmacology. New York: Raven Press, 1974:349.

175. Bjorkman U, Ekholm R, Ericson LE. Effects of thyrotropin on thyroglobulin exocytosis and iodination in the rat thyroid gland. Endocrinology 1978;102:460.

176. Yi X, Yamamoto K, Shu L, et al. Effects of propyithiouracil (ptu) administration on the synthesis and secretion of thyroglobulin in the rat thyroid gland: a quantitative immuno-electron microscopic study using immunogold technique. Endocr Path 1997;8:315.

177. Desruisseau S, Alquier C, Depetris D, et al. Hormonal regulation of some steps of thyroglobulin synthesis and secretion in bicameral cell culture. J Cell Physiol 1994;160:336.

178. Dhanvantari S, Loh YP. Lipid raft association of carboxypeptidase E is necessary for its function as a regulated secretory pathway sorting receptor. J Biol Chem 2000;275:29887.

179. Blazquez M, Thiele C, Huttner WB, et al. Involvement of the membrane lipid bilayer in sorting prohormone convertase 2 into the regulated secretory pathway. Biochem J 2000;349:843.

180. Blazquez M, Docherty, Shennan KI. Association of prohormone convertase 3 with membrane lipid rafts. J Mol Endocrinol 2001;27:107.

181. Schmidt K, Schrader M, Kern HF, et al. Regulated apical secretion of zymogens in rat pancreas. Involvement of the glycosylphosphatidylinositol-anchored glycoprotein GP-2, the lectin ZG16p, and cholesterol-glycosphingolipid-enriched microdomains. J Biol Chem 2001;276:14315.

182. Zhang CF, Dhanvantari S, Lou H, et al. Sorting of carboxypeptidase E to the regulated secretory pathway requires interaction of its transmembrane domain with lipid rafts. Biochem J 2003; 369:453.

183. Bjorkman U, Ekholm R. Accelerated exocytosis and H2O2 generation in isolated thyroid follicles enhance protein iodination. Endocrinology 1988;122:488.

184. Dunn JT. The amino acid neighbors of thyroxine in thyroglobulin. J Biol Chem 1970;245:5954.

185. Turakulov I, Saatov T, Babaev TA, et al. Synthesis of iodoamino acids during in vitro thyroglobulin iodination in different states of its molecule. Biokhimiia 1976;41:1004.

186. Maurizis JC, Marriq C, Rolland M, et al. Thyroid hormone synthesis and reactivity of hormone-forming tyrosine residues of thyroglobulin. FEBS Lett 1981;132:29.

187. Taurog A, Dorris ML. Myeloperoxidase-catalyzed iodination and coupling. Arch Biochem Biophys 1992;296:239.

188. Magnusson RP, Taurog A, Dorris ML. Mechanism of thyroid peroxidase- and lactoperoxidase-catalyzed reactions involving iodide. J Biol Chem 1984;259:13783.

189. Dunn JT, Kim PS, Dunn AD. Favored sites for thyroid hormone formation on the peptide chains of human thyroglobulins. J Biol Chem 1982;257:88.

190. Palumbo G. Thyroid hormonogenesis. Identification of a sequence containing iodophenyl donor site(s) in calf thyroglobulin. J Biol Chem 1987;262:17182.

191. Ohmiya Y, Hayashi H, Kondo T, et al. Location of dehydroalanine residues in the amino acid sequence of bovine thyroglobulin. J Biol Chem 1990;265:9066.

192. Marriq C, Lejeune PJ, Venot N, et al. Hormone formation in the isolated fragment 1–171 of human thyroglobulin involves the couple tyrosine 5 and tyrosine 130. Mol Cell Endocrinol 1991;81:155.

193. Gentile F, Ferranti P, Mamone G, et al. Identification of hormonogenic tyrosines in fragment 1218–1591 of bovine thyroglobulin by mass spectrometry. J Biol Chem 1997;272:639.

194. Dunn AD, Corsi CM, Myers HE, et al. Tyrosine 130 is an important outer ring donor for thyroxine formation in thyroglobulin. J Biol Chem 1998;273:25223.

195. Dunn JT, Dunn AD. The importance of thyroglobulin structure for thyroid hormone biosynthesis. Biochimie 1999;81:505.

196. Dunn JT, Anderson PC, Fox JW, et al. The sites of thyroid hormone formation in rabbit thyroglobulin. J Biol Chem 1987; 262:16948.

197. Fassler CA, Dunn JT, Anderson PC, et al. Thyrotropin alters the utilization of thyroglobulin's hormonogenic sites. J Biol Chem 1988;263:17366.

198. Gavaret JM, Deme D, Nunez J, et al. Sequential reactivity of tyrosyl residues of thyroglobulin upon iodination catalyzed by thyroid peroxidase. J Biol Chem 1977;252:3281.

199. Palumbo G, Gentile F, Condorelli GL, et al. The earliest site of iodination in thyroglobulin is residue number 5. J Biol Chem 1990;265:8887.

200. Rawitch AB, Chernoff SB, Litwer MR, et al. Thyroglobulin structure-function. The amino acid sequence surrounding thyroxine. J Biol Chem 1983;258:2079.

201. Rawitch AB, Litwer MR, Gregg J, et al. The isolation of identical thyroxine containing amino acid sequences from bovine, ovine and porcine thyroglobulins. Biochem Biophys Res Comm 1984;118:423.

202. Erregragui K, Prato S, Miquelis R, et al. Antigenic mapping of human thyroglobulin–topographic relationship between antigenic regions and functional domains. Eur J Biochem 1997;244:801.

203. Xiao S, Pollock HG, Taurog A, et al. Characterization of hormonogenic sites in an N-terminal, cyanogen bromide fragment of human thyroglobulin. Arch Biochem Biophys 1995;320:96.

204. Baudry N, Lejeune PJ, Niccoli P, et al. Dityrosine bridge formation and thyroid hormone synthesis are tightly linked and are both dependent on N-glycans. FEBS Lett 1996;396:223.

205. den Hartog MT, Sijmons CC, Bakker O, et al. Importance of the content and localization of tyrosine residues for thyroxine formation within the N-terminal part of human thyroglobulin. Eur J Endocrinol 1995;132:611.

206. Asuncion M, Ingrassia R, Escribano J, et al. Efficient thyroid hormone formation by in vitro iodination of a segment of rat thyroglobulin fused to Staphylococcal protein A. FEBS Lett 1992;297:266.

207. Anderberg B, Enestrom S, Gillquist J, et al. Protein composition in single follicles, homogenates and fine-needle aspiration biopsies from normal and diseased human thyroid. Acta Endocrinol (Copenh) 1981;96:328.

208. Smeds S. On the distribution of thyroglobin and larger iodoproteins in single rat thyroid follicles. Pflugers Arch 1977; 372:145.

209. Smeds S, Anderberg B. Change of the protein composition of the thyroid colloid during treatment with propylthiouracil and thyroxine: a microgel electrophoretic study of single rat thyroid follicles. Biochim Biophys Acta 1978;542:47.

210. Ekholm R. Biosynthesis of thyroid hormones. Int Rev Cytol 1990;120:243.

211. Schneider PB. Thyroidal iodine heterogeneity: “last come, first served” system of iodine turnover. Endocrinology 1964; 74:973.

212. Matsukawa S, Hosoya T. Process of iodination of thyroglobulin and its maturation. II. Properties and distribution of thyroglobulin labeled in vitro or in vivo with radioiodine, 3H-tyrosine, or 3H-galactose in rat thyroid glands. J Biochem (Tokyo) 1979;86:199.

213. Saboori AM, Rose NR, Bresler HS, et al. Iodination of human thyroglobulin (Tg) alters its immunoreactivity. I. Iodination alters multiple epitopes of human Tg. Clin Exp Immunol 1998; 113:297.

214. Dunn JT, Ray SC. Changes in thyroglobulin structure after TSH administration. J Biol Chem 1975;250:5801.

215. Matsukawa S, Hosoya T. Process of iodination of thyroglobulin and its maturation. I. Properties and distribution of thyroglobulin labeled with radioiodine in pig thyroid slices. J Biochem 1979;85:1009.

216. Frati L, Bilstad J, Edelhoch H, et al. Biosynthesis of the 27S thyroid iodoprotein. Arch Biochem Biophys 1974;162:126.

217. Saber-Lichtenberg Y, Brix K, Schmitz A, et al. Covalent cross-linking of secreted bovine thyroglobulin by transglutaminase. FASEB J 2000;14:1005.

218. Herzog, V, Berndorfer U, Saber Y. Isolation of insoluble secretory product from bovine thyroid: extracellular storage of thyroglobulin in covalently cross-linked form. J Cell Biol 1992; 118:1071.

219. Leonardi A, Acquaviva R, Marinaccio M, et al. Presence of dityrosine bridges in thyroglobulin and their relationship with iodination. Biochem Biophys Res Communun 1994;202:38.

220. Baudry N, Lejeune PJ, Delom F, et al. Role of multimerized porcine thyroglobulin in iodine storage. Biochem Biophys Res Commun 1998;242:292.

221. Dunn JT, Kim PS, Dunn AD, et al. The role of iodination in the formation of hormone-rich peptide from thyroglobulins. J Biol Chem 1983;258:9093.

222. Kim PS, Dunn JT, Kaiser DL. Similar hormone-rich peptides from thyroglobulins of five vertebrate classes. Endocrinology 1984;114:369.

223. Duthoit C, Estienne V, Delom F, et al. Production of immunoreactive thyroglobulin C-terminal fragments during thyroid hormone synthesis. Endocrinology 2000;141:2518.

224. Brix K, Lemansky P, Herzog V. Evidence for extracellularly acting cathepsins mediating thyroid hormone liberation in thyroid epithelial cells. Endocrinology 1996;137:1963.

225. Metaye T, Kraimps JL, Goujon JM, et al. Expression, localization, and thyrotropin regulation of cathepsin D in human thyroid tissues. J Clin Endocrinol Metab 1997;82:3383.

226. Linke M, Jordans S, Mach L, et al. Thyroid stimulating hormone upregulates secretion of cathepsin B from thyroid epithelial cells. Biol Chem 2002;383:773.

227. Marino M, McCluskey RT. Role of thyroglobulin endocytic pathways in the control of thyroid hormone release. Am J Physiol 2000;279:C1295.

228. Bastiani P, Papandreou MJ, Blanck O, et al. On the relationship between completion of N-acetyllactosamine oligosaccharide units and iodine content of thyroglobulin: a reinvestigation. Endocrinology 1995;136:4204.

229. Herzog V. Transcytosis in thyroid follicle cells. J Cell Biol 1983; 97:607.

230. Romagnoli P, Herzog V. Transcytosis in thyroid follicle cells: regulation and implications for thyroglobulin transport. Exp Cell Res 1991;194:202.

231. Montuori N, Pacifico F, Mellone S, et al. The rat asialoglycoprotein receptor binds the amino-terminal domain of thyroglobulin. Biochem Biophys Res Comm 2000;268:42.

232. Siffroi-Fernandez S, Delom F, Nlend MC, et al. Identification of thyroglobulin domain(s) involved in cell-surface binding and endocytosis. J Endocrinol 2001;170:217.

233. Hatipoglu BA, Schneider AB. Selective endocytosis of thyroglobulin: a review of potential mechanisms for protecting newly synthesized molecules from premature degradation. Biochimie 1999;81:549.

234. Miquelis R, Courageot J, Jacq A, et al. Intracellular routing of GlcNAc-bearing molecules in thyrocytes: selective recycling through the Golgi apparatus. J Cell Biol 1993;123:1695.

235. Blanck O, Perrin C, Mziaut H, et al. Molecular cloning, cDNA analysis, and localization of a monomer of the N-acetylglucosamine-specific receptor of the thyroid, NAGR1, to chromosome 19p13.3–13.2. Genomics 1994;21:18.

236. Thibault V, Blanck O, Courageot J, et al. The N-acetylglucosamine-specific receptor of the thyroid: purification, further characterization, and expression patterns on normal and pathological glands. Endocrinology 1993;132:468.

237. Mziaut H, Bastiani P, Balivet T, et al. Carbohydrate and protein determinants are involved in thyroglobulin recognition by FRTL 5 cells. Endocrinology 1996;137:1370.

238. Mezghrani A, Courageot J, Mani JC, et al. Protein-disulfide isomerase (PDI) in FRTL5 cells. pH-dependent thyroglobulin/PDI interactions determine a novel PDI function in the post-endoplasmic reticulum of thyrocytes. J Biol Chem 2000; 275:1920.

239. Klein M, Gestmann I, Berndorfer U, et al. The thioredoxin boxes of thyroglobulin: possible implications for intermolecular disulfide bond formation in the follicle lumen. Biol Chem 2000;381:593.

240. Delom F, Mallet B, Carayon P, et al. Role of extracellular molecular chaperones in the folding of oxidized proteins. Refolding of colloidal thyroglobulin by protein disulfide isomerase and immunoglobulin heavy chain-binding protein. J Biol Chem 2001;276:21337.

241. Zheng G, Marino M, Zhao J, et al. Megalin (gp330): a putative endocytic receptor for thyroglobulin (Tg). Endocrinology 1998; 139:1462.

242. Marino M, Lisi S, Pinchera A, et al. Targeting of thyroglobulin to transcytosis following megalin-mediated endocytosis: evidence for a preferential pH-independent pathway. J Endocrinol Invest 2003;26:222.

243. Marino M, Zheng G, Chiovato L, et al. Role of megalin (gp330) in transcytosis of thyroglobulin by thyroid cells. A novel function in the control of thyroid hormone release. J Biol Chem 2000;275:7125.

244. Marino M, Chiovato L, Lisi S, et al. Binding of the low density lipoprotein receptor-associated protein (RAP) to thyroglobulin (Tg): putative role of RAP in the Tg secretory pathway. Mol Endocrinol 2001;15:1829.

245. Lisi S, Pinchera A, McCluskey RT, et al. Binding of heparin to human thyroglobulin (Tg) involves multiple binding sites including a region corresponding to a binding site of rat Tg. Eur J Endocrinol 2002;146:591.

246. Marino M, Zheng G, McCluskey RT. Megalin (gp330) is an endocytic receptor for thyroglobulin on cultured fisher rat thyroid cells. J Biol Chem 1999;274:12898.

247. Marino M, Pinchera A, McCluskey RT, et al. Megalin in thyroid physiology and pathology. Thyroid 2001;11:47.

248. Suzuki K, Lavaroni S, Mori A, et al. Autoregulation of thyroid-specific gene transcription by thyroglobulin. Proc Natl Acad Sci U S A 1998;95:8251.

249. Suzuki K, Mori A, Lavaroni S, et al. In vivo expression of thyroid transcription factor-1 RNA and its relation to thyroid function and follicular heterogeneity: identification of follicular thyroglobulin as a feedback suppressor of thyroid transcription factor-1 RNA levels and thyroglobulin synthesis. Thyroid 1999;9:319.

250. Royaux IE, Suzuki K, Mori A, et al. Pendrin, the protein encoded by the Pendred syndrome gene (PDS), is an apical porter of iodide in the thyroid and is regulated by thyroglobulin in FRTL-5 cells. Endocrinology 2000;141:839.

251. Suzuki K, Mori A, Saito J, et al. Follicular thyroglobulin suppresses iodide uptake by suppressing expression of the sodium/iodide symporter gene. Endocrinology 1999;140:5422.

252. Ulianich L, Suzuki K, Mori A, et al. Follicular thyroglobulin (TG) suppression of thyroid-restricted genes involves the apical membrane asialoglycoprotein receptor and TG phosphorylation. J Biol Chem 1999;274:25099.

253. Sellitti DF, Suzuki K, Doi SQ, et al. Thyroglobulin increases cell proliferation and suppresses Pax-8 in mesangial cells. Biochem Biophys Res Comm 2001;285:795.

254. Huang SS, Cerullo MA, Huang FW, et al. Activated thyroglobulin possesses a transforming growth factor-beta activity. J Biol Chem 1998;273:26036.

255. Rousset B, Selmi S, Alquier C, et al. In vitro studies of the thyroglobulin degradation pathway: endocytosis and delivery of thyroglobulin to lysosomes, release of thyroglobulin cleavage products—iodotyrosines and iodothyronines. Biochimie 1989; 71:247.

256. Gire V, Kostrouch Z, Bernier-Valentin F, et al. Endocytosis of albumin and thyroglobulin at the basolateral membrane of thyrocytes organized in follicles. Endocrinology 1996;137:522.

257. Rousset B, Selmi S, Bornet H, et al. Thyroid hormone residues are released from thyroglobulin with only limited alteration of the thyroglobulin structure. J Biol Chem 1989;264: 12620.

258. Dunn AD, Crutchfield HE, Dunn JT. Proteolytic processing of thyroglobulin by extracts of thyroid lysosomes. Endocrinology 1991;129:3073.

259. Friedrichs B, Tepel C, Reinheckel T, et al. Thyroid functions of mouse cathepsins B, K, and L. J Clin Invest 2003;111:1733.

260. van de Graaf SA, Ris-Stalpers C, Veenboer GJ, et al. A premature stop codon in thyroglobulin messenger RNA results in familial goiter and moderate hypothyroidism. J Clin Endocrinol Metab 1999;84:2537.

261. Veenboer GJ, de Vijlder JJ. Molecular basis of the thyroglobulin synthesis defect in Dutch goats. Endocrinology 1993;132:377.