| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Unidade de Amilóide, Instituto de Biologia Molecular e Celular (J.A.P., R.F., M.J.S.), 4150–180 Porto, Portugal; Instituto Superior de Ciências da Saúde-Norte (J.A.P.), 4585-116 Gandra, Portugal; Unidad de Endocrinologia Molecular, Instituto de Investigaciones Biomédicas Alberto Sols, Consejo Superior de Investigaciones Científicas and Universidad Autónoma de Madrid (G.M.D.E.), 28029 Madrid, Spain; Medical Research Council Clinical Sciences Center, Mammalian Neurogenesis Laboratory, Hammersmith Hospital (V.E.), W12 ONN London, United Kingdom; Institute of Cancer Research, Columbia University (M.G.), New York, New York 10032; and Instituto de Ciências Biomédicas Abel Salazar (M.J.S.), 4099-003 Porto, Portugal
Address all correspondence and requests for reprints to: Dr. Joana Almeida Palha, Instituto de Biologia Molecular e Celular, Rua do Campo Alegre 823, 4150–180 Porto, Portugal. E-mail: japalha{at}ibmc.up.pt
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TTR is the major plasma carrier of thyroid hormones in rodents (7) and is the main thyroid hormone-binding protein in CSF of both rodent and humans (8). The high concentration of TTR in the CSF suggested a major role for TTR in mediating T4 transfer from the blood into the brain across the blood-choroid-plexus-CSF barrier (9, 10, 11, 12, 13).
Thyroid hormones are essential for normal mammalian brain physiology and are particularly critical during development (14, 15, 16). Moreover, the adult brain is sensitive to thyroid hormones beyond the time of the so-called critical period of brain development. More than half of the T3 (the biologically active thyroid hormone) in the cerebral cortex arises from local deiodination of T4 by the enzymatic action of type 2 iodothyronine 5'-deiodinase (D2) (17, 18). In hypothyroidism, the brain compensates for the low levels of circulating hormones by increasing D2 activity (19). T4 or T3 can regulate brain D2 through both pre- and posttranslational mechanisms (20). Among the molecular events known to be controlled by thyroid hormones in the adult brain is the transcription of the RC3 gene, whose product is a brain-specific, calmodulin-binding protein kinase C substrate that has been implicated in postsynaptic events (21). A thyroid hormone-responsive element has been found in the human RC3 gene homolog (22). Previous analysis of thyroid hormone metabolism markers (23) revealed that TTR-null mice are hypothyroxinemic, but euthyroid (24). In addition, kinetic studies using [125I]T4 showed that TTR is not essential for T4 uptake by the brain and other tissues (25). However, lower total T4 levels were detected in the TTR-null whole brain. In the present work we use the TTR-null mouse strain to study further the involvement of TTR in the maintenance of thyroid hormone homeostasis in the brain.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In vitro choroid plexus labeling
Pools of choroid plexuses from three wild-type or TTR-null mice
were incubated in 300 µl leucine-free medium (DMEM, Life Technologies, Inc., Paisley, UK) containing 15 µCi
[14C]leucine (Amersham Pharmacia Biotech, Uppsala, Sweden; SA, 319 mCi/mmol). After 3-h
incubation in 5% CO2 at 37 C, cell pellets were
resuspended in medium containing 0.1% Tween. Radioactive proteins
present in both secretion media and cells were separated on SDS-PAGE
gels and analyzed by autoradiography. As a standard we used a
preparation containing 125I-radiolabeled TTR and
retinol-binding protein.
T4 binding to CSF proteins
Native gel electrophoresis. Aliquots of 13 µl pooled CSF
and 5 µl serum were incubated for 30 min at room temperature with 12
µl [125I]T4 (DuPont,
Wilmington, DE; 5700 µCi/µg) diluted 4-fold in 200
mM glycine and 130 mM sodium acetate, pH 8.6
(7) and subjected to electrophoresis on 8% native polyacrylamide gels,
using the same glycine-acetate as gel and running buffer, for 3–5 h at
40 mA at room temperature. The migration of albumin was determined in a
serum sample incubated with bromophenol blue run on a different gel
lane. Radioactive proteins were visualized by autoradiography 4 h
after exposure.
In another experiment, 40 µl CSF were applied on a blue-Sepharose (Amersham Pharmacia Biotech) column (300-µl volume) equilibrated in 30 mM sodium phosphate, pH 7.0. The bound material was eluted in 0.5 M sodium thiocyanate in 30 mM sodium phosphate, pH 7.0. Both the bound- and unbound fractions were concentrated on Microcon 10 (Amicon, Beverly, MA) and then incubated with [125I]T4 and analyzed by electrophoresis as described above.
Chromatography on Biogel P-6 DG. Seventeen microliters of
pooled CSF were incubated for 4 h at 4 C with
105 cpm
[125I]T4 (DuPont; SA,
1250 µCi/µg) in TNE buffer (100 mM Tris, 100
mM NaCl, and 1 mM EDTA, pH 8.0; 200-µl final
volume). Samples were applied to 1-ml Biogel P-6 DG columns
(Bio-Rad Laboratories, Inc., Hercules, CA) (27) in TNE
buffer immediately after loading 200 µl TNE/20% sucrose and were
centrifuged for 1 min at 1000 rpm. The columns were then washed with
200 µl TNE, and the eluates, containing the protein-bound
[125I]T4 were counted in
a 
-counter. Free
[125I]T4 remained bound
to the gel, allowing calculation of the percentage of protein-bound
[125I]T4 in CSF. A blank
consisted of 106 cpm
[125I]T4 in TNE buffer;
the eluate of this preparation is considered background
radioactivity.
T4 and T3 tissue
content
Animals were anesthetized and perfused via the vena cava with 5
ml PBS containing 11 U heparin/ml, followed by 2 ml PBS; the portal
vein was used as an exit for the perfusate. Cortex, cerebellum,
hippocampus, and choroid plexus were dissected and frozen in dry ice.
Samples were kept at -70 C until use. T4 and
T3 contents were determined by RIA after
extraction from tissue (28).
Northern blot analysis for RC3
RNA was isolated from brain using the Ultraspec RNA Isolation
System (Biotecx Laboratories, Inc., Houston, TX),
and 20-µg RNA samples were subjected to electrophoresis and blotted
on nylon membranes using standard procedures. Probes were labeled using
a random primer labeling kit (Stratagene, La Jolla,
CA).
A rat RC3 complementary DNA (cDNA) probe was used to detect RC3 messenger RNA (mRNA) (29). Prehybridization was performed for 5 h at 42 C in 50% formamide, 3 x SSC (standard saline citrate), 5 x Denhardt’s solution, 0.1% SDS, and 100 µg/ml salmon sperm DNA, and hybridization was performed under the same conditions for 20 h after the addition of 1 x 106 cpm probe/ml solution. Washes were performed at room temperature in 1 x SSC-0.1% SDS. A signal was observed after 6 h.
The membrane was normalized with cyclophilin message, a ubiquitous mRNA (30), using a rat cyclophilin cDNA probe. Prehybridization was for 5 h at 39 C in 53% formamide, 0.75 M sodium chloride, 25 mM PIPES (pH 6.8), 25 mM EDTA, 5 x Denhardt’s solution, 0.1% SDS, and 100 µg/ml salmon sperm DNA, and hybridization was performed under the same conditions for 20 h after the addition of 1 x 106 cpm probe/ml solution. Washes were performed at room temperature in 2 x SSC-0.1% SDS and in 0.2 x SSC-0.2% SDS at 60 C. Dr. J. G. Sutcliffe (The Scripps Research Institute, La Jolla, CA) provided both the RC3 and cyclophilin cDNAs.
D2 activity measurements
Hippocampus and cortex samples, pooled from three animals, and
individual cerebellum samples were homogenized in 470, 1000, and 350
µl 10 mM Tris-HCl (pH 7.2) and 0.25 M
sucrose, respectively; and kept at -70 C until use. A previously
described procedure (31) was used, with minor modifications. The
200-µl final mix consisted of 50 mM potassium phosphate
(pH 7.2), 1 mM propylthiouracil, 1 mM EDTA, 10
mM dithiothreitol, 8 nM
T4, 2 µM T3,
800 µg homogenate protein, and 105 cpm
[125I]T4 (SA, 1250
µCi/µg; DuPont) previously purified on LH-20 Sephadex
chromatography (32). Each sample, run in duplicate or triplicate, was
incubated for 60 or 90 min at 37 C. Two types of blanks were used, some
in which incubation was stopped at time zero by adding 100 µl BSA (70
mg/ml) and 500 µl 10% trichloroacetic acid, and others in which the
homogenate was boiled before the addition of the labeled hormone and
then incubated for 60 or 90 min at 37 C. The reaction was stopped by
adding 100 µl BSA (70 mg/ml) and 500 µl 10% trichloroacetic acid.
Samples were centrifuged for 5 min at 3500–4500 rpm. Iodine produced
in the reaction was separated by Sephadex LH-20 chromatography (32).
Radioactivity was measured by -spectrometry.
Protein concentrations were determined using a protein assay (Bio-Rad Laboratories, Inc.) with albumin as standard.
Statistics
Group values, expressed as the mean ± SD, were
compared by Student’s t test. P < 0.05 was
considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
indicates that despite an overload of
TTR-null secretion medium protein, no TTR was present in the choroid
plexus secretion medium of the TTR-null mouse (lane 2 vs.
lane 1). Identification of the TTR band in the wild-type choroid plexus
secretion medium (lane 1) and cells (lane 3) was based on the migration
of a TTR standard (lane 4). TTR was absent in choroid plexus cells and
CSF of the TTR-null mice (data not shown). RT-PCR experiments further
confirmed the absence of TTR transcription in the TTR-null choroid
plexus (data not shown).
|
shows that in the wild-type mouse, TTR
and albumin were the major
[125I]T4-binding proteins
(lane 3). The TTR-null mouse serum showed less
[125I]T4 binding at the
albumin front, which corresponds to TTR mobility, and an additional
band in the TBG region of the gel (lane 4). The appearance of the TBG
band in the TTR-null serum did not correspond to increased TBG
concentration, but, rather, to the shift of the binding toward TBG in
the absence of TTR (24). In wild-type CSF,
[125I]T4 labeling
occurred in the region where TTR and albumin overlapped and appeared as
a single band (lane 1). In TTR-null CSF (lane 2),
[125I]T4 was recovered
entirely in a band corresponding to free
[125I]T4 (lane 5).
|
Because in the experimental conditions used, TTR and albumin cannot be
resolved, we took advantage of the ability of blue-Sepharose to bind
albumin to discriminate between TTR and albumin in wild-type CSF. The
blue-Sepharose-bound and -unbound fractions of wild-type or TTR-null
CSF were incubated with
[125I]T4 and
electrophoresed under native conditions. Figure 2B
shows that no
binding of [125I]T4
occurred to the blue-Sepharose-bound fraction in wild-type CSF (lane
2), indicating the absence of albumin. On the other hand,
[125I]T4 bound to the
blue-Sepharose-unbound wild-type CSF fraction (lane 1) that contained
TTR. These results imply that TTR is the only
T4-binding protein in wild-type CSF. The
blue-Sepharose-bound (lane 4) and -unbound (lane 3) fractions from the
TTR-null CSF did not bind
[125I]T4, as all labels
were recovered in the free
[125I]T4
region (lane 5). These observations further suggest the absence of
T4-binding protein in TTR-null CSF.
To confirm the absence of [125I]T4 binding in TTR-null CSF, we chromatographed CSF after incubation with [125I]T4 on Biogel P-6 DG. For wild-type mice, about 60% of the total [125I]T4 were recovered in the protein-bound fraction. In contrast, only approximately 4% of the total [125I]T4 were recovered in the protein-bound fraction of the TTR-null CSF, a value equivalent to the blank. Thus, both procedures indicated that TTR-null CSF contains no T4-binding protein.
T4 and T3 tissue
content
As shown in Table 1, no
differences in T4 and T3
levels were found between wild-type and the TTR-null mutant cortex,
cerebellum, and hippocampus. In TTR-null choroid plexus,
T4 and T3 levels were
significantly reduced, amounting to 14% and 48% of wild-type values,
respectively.
|
|
. No difference between these
groups was observed, indicating that the brain was euthyroid with
respect to this criterion.
|
). Further measurements of deiodinase
activity in synaptosomal fractions of both cortex and hippocampus
likewise revealed no differences between TTR-null and wild-type mice
(data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T4 binds only to TTR in the CSF of normal mice (8, 11). Using two independent methods, we show that T4 is found exclusively unbound in TTR-null mice. Thus, no other CSF protein replaces TTR in the transport of T4.
T4 has been suggested to reach the brain through transport of the serum free T4 fraction across the blood-brain barrier (34) as well as by association with TTR through the blood-choroid-plexus-CSF barrier (10, 12, 35). T4 bound to CSF TTR could be delivered into different brain regions. Thus, after iv injection of [125I]T4, radioactivity first appeared in the choroid plexus, then in the CSF, and later in the cortex and striatum of the brain (9, 10).
We previously showed that TTR is not absolutely required for T4 to reach the brain (24). However, we also found that whole brain total T4 levels were decreased in the TTR-null mice (25). We report here that the T4 levels in cortex, hippocampus, and cerebellum are normal. The decrease in T4 levels previously measured in whole brain can be accounted for by the very marked reduction of T4 levels in the choroid plexus of the TTR-null mice.
The TTR-bound fraction represents the major T4 pool in the wild-type choroid plexus. Our results indicate that T4 can enter the choroid plexus and, presumably, pass on into the CSF in the absence of TTR. Thus, T4 entry into the choroid plexus may derive from the free T4 pool. In fact, serum free T4 is normal in the TTR-null sera (24).
Schreiber et al. (36) proposed that circulating T4-binding proteins act as distributors of the hormone within the tissues. They also suggested that, given the redundancy of T4-binding proteins (namely, T4-binding globulin, albumin, and TTR), the absence of one protein would be compensated for by the others. This concept of a protein network may apply for serum (24). It cannot, however, apply for the blood-choroid-plexus-CSF barrier, as we found no T4-binding protein in the CSF of TTR-null mice. Our results indicate that CSF TTR is not required for normal maintenance of thyroid hormone in brain. Whether the distribution of the hormone in the brain is equivalent in the TTR-null and wild-type mice is under investigation.
It has been suggested that one fifth of the T4 present in the brain may originate from TTR-bound T4 transported from the choroid plexus to the CSF (11). Our observations do not support the idea that TTR constitutes a major route of T4 transfer into the brain parenchyma under normal physiological conditions (9, 10). A direct measure of T4 levels in the CSF of wild-type and TTR-null mice is, unfortunately, not feasible, due to the large amounts of CSF required for the analysis.
To evaluate further the thyroid hormone status of the TTR-null mouse brain, we measured parameters known to be affected in altered thyroid states. Thyroid hormones regulate gene expression in the central nervous system, as in other tissues, through transcriptional activation (15). Therefore, we investigated the levels of steady state messenger RNA for RC3, a gene whose expression is decreased in adult hypothyroidism (21). Consistent with an euthyroid status, the levels of RC3 messenger RNA were normal in the TTR-null brain.
The major portion of T3 in the brain is locally produced by 5'-deiodination of T4 by D2 (17, 18). To guarantee a sufficient supply of T3, the activity of D2 increases in hypothyroidism, whereas D2 activity is decreased in hyperthyroidism (37, 38). In the adult brain, D2 mRNA is expressed in the cerebellar cortex, hippocampus, caudate-putamen, thalamus, and cerebellum (39). We reported previously that despite decreased whole brain total T4 levels in the TTR-null mice, their normal whole brain T3 levels were not accounted for by increased D2 activity (24). However, as regional differences may have been masked in the analysis of the whole brain, we performed regional D2 analyses. No differences were found in D2 activity in the cortex, cerebellum, and hippocampus of the TTR-null compared with wild-type mice. In addition, T3 levels in the cortex, cerebellum, and hippocampus were equivalent in wild-type and TTR-null mice. The greatly reduced T3 levels found in the TTR-null choroid plexus are probably a direct consequence of the low total T4 levels found in this epithelium. The presence of deiodinases in the choroid plexus has not been reported. Taken together, our data provide strong evidence that the brain is euthyroid in the absence of TTR.
The normal free T4 serum levels in the TTR-null mice (24) and the data presented here support the free hormone hypothesis for T4 tissue uptake. In this hypothesis, the biological activity of hormones is a function of their unbound (free), rather than protein-bound, concentration (40).
We propose that under normal physiological conditions, TTR has a storage role for T4 in blood, choroid plexus, and CSF. This function may be relevant under conditions of increased hormone demand. In the absence of TTR as a reservoir in the choroid plexus and CSF, T4 exit from the CSF to the cerebral and, ultimately, the general circulation might be accelerated.
In summary, the data reported using TTR knockout mice bear on the role of TTR in thyroid economy in brains of normal mice. Because no other T4-binding protein replaces TTR in the CSF and because a normal brain T4 content (aside from the choroid plexus) is observed, we conclude that the blood-choroid-plexus-CSF barrier plays no essential role in entry or maintenance of T4 in the brain. If the TTR-null mouse possesses compensatory mechanisms for brain CSF transport and distribution in the absence of TTR, our studies to date have not revealed them. To rule out the possibility that TTR might account for some quantitative or qualitative aspects of thyroid hormone transfer, we are investigating the distribution of radiolabeled T4 and T3 in the brains of TTR-null and wild-type mice.
|
Footnotes |
|---|
Received March 1, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. Diamanti-Kandarakis, J.-P. Bourguignon, L. C. Giudice, R. Hauser, G. S. Prins, A. M. Soto, R. T. Zoeller, and A. C. Gore Endocrine-Disrupting Chemicals: An Endocrine Society Scientific Statement Endocr. Rev., June 1, 2009; 30(4): 293 - 342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Kassem, R. Deane, M. B. Segal, and J. E. Preston Role of transthyretin in thyroxine transfer from cerebrospinal fluid to brain and choroid plexus Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2006; 291(5): R1310 - R1315. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J C Sousa, G M. de Escobar, P Oliveira, M J Saraiva, and J A Palha Transthyretin is not necessary for thyroid hormone metabolism in conditions of increased hormone demand J. Endocrinol., November 1, 2005; 187(2): 257 - 266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sampaolo, A. Campos-Barros, G. Mazziotti, S. Carlomagno, V. Sannino, G. Amato, C. Carella, and G. Di Iorio Increased Cerebrospinal Fluid Levels of 3,3',5'-Triiodothyronine in Patients with Alzheimer's Disease J. Clin. Endocrinol. Metab., January 1, 2005; 90(1): 198 - 202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Montero-Pedrazuela, J. Bernal, and A. Guadano-Ferraz Divergent Expression of Type 2 Deiodinase and the Putative Thyroxine-Binding Protein p29, in Rat Brain, Suggests that They Are Functionally Unrelated Proteins Endocrinology, March 1, 2003; 144(3): 1045 - 1052. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
