| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Biology Department and Molecular and Cellular Biology Program, University of Massachusetts, Amherst, Massachusetts 01003
Address all correspondence and requests for reprints to: Dr. R. Thomas Zoeller, Biology Department, Morrill Science Center, University of Massachusetts, Amherst, Massachusetts 01003. E-mail: tzoeller{at}bio.umass.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several lines of evidence support the concept that thyroid hormone, perhaps of maternal origin, exerts receptor-mediated effects on the fetal brain. For example, receptors for thyroid hormone (TRs) are expressed in the fetal brain before the onset of fetal thyroid function (18). Moreover, thyroid hormone of maternal origin reaches the fetal brain and is bound to the TR (7, 10, 19, 20, 21, 22). Considering that TRs are ligand-dependent transcription factors (23, 24), we recently initiated a study to identify genes expressed in the fetal rat brain before the onset of fetal thyroid function that are affected by changes in maternal thyroid hormone (25). The ultimate goal of this project is to identify the developmental processes affected by maternal thyroid hormone, the timing of thyroid hormone effects, and the mechanisms by which thyroid hormone exerts these effects.
One of the genes expressed in fetal cortex whose expression was affected by manipulation of maternal thyroid status was the gene encoding neuroendocrine-specific protein (NSP) (25). Because its expression is strongly correlated with neuronal differentiation (26), NSP may be an important mediator of thyroid hormone effects on brain development. Our previous work demonstrated that there are two NSP transcripts expressed in the fetal cortex, a 3.5-kb transcript designated NSP-A and a 1.5-kb transcript designated NSP-C (27, 28); the NSP-A transcript was selectively affected by thyroid hormone. However, we mapped the distribution of NSP expression in the fetal cortex and throughout development using a complementary RNA (cRNA) probe that cross-hybridized to the NSP-A and NSP-C transcripts. Therefore, we were unable to determine whether NSP-A and NSP-C exhibit different patterns of expression in the fetal brain. Moreover, we were unable to confirm that maternal hypothyroidism could alter NSP-A expression in the fetal brain. Therefore, we generated probes specific for NSP-A and NSP-C and now report that in the fetal cortex the thyroid hormone-responsive NSP-A is expressed exclusively in the proliferative ventricular zone, and that in the adult cerebellum, this transcript is expressed exclusively in Purkinje cells. In addition, we have confirmed that the expression of NSP-A is affected by maternal hypothyroidism.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the second experiment, timed pregnant female Sprague Dawley rats (n = 24; Zivic-Miller Laboratories, Inc.) arrived at our facility on G2 and were maintained on drinking water containing either 0.04% 6-(n)-propylthiouracil (PTU) (Sigma; n = 12) with 3% sucrose to reduce the bitterness associated with PTU or 3% sucrose alone (n = 12). The goitrogen PTU blocks both thyroid hormone synthesis and the conversion of T4 to T3 by type I 5'-deiodinase (29, 30). The two solutions were provided fresh daily for 14 days. On G15, all animals received a sc injection of either T4 (50 µg/kg BW in 100 µl; Sigma; n = 12) or 100 µl saline (n = 12) at both 1000 and 1800 h. This paradigm produced four groups of six litters each: control, control+T4, PTU+saline, and PTU+T4. At 1000 h on G16, all dams were killed as described above, and trunk blood was collected for measurement of serum T3, T4, and TSH. All fetuses were frozen intact as described above and stored at -80 C until they were sectioned for in situ hybridization.
To characterize the spatial and temporal changes in expression of NSP-A and NSP-C throughout normal development, nulliparous female Sprague Dawley rats (n = 5; Zivic-Miller Laboratories, Inc.) were maintained on rat chow and water ad libitum and mated as described above. Dams were killed at 1200 h on G15, G16, G18, and G21, and the fetuses were collected as described above. The remaining dam carried the pregnancy to term. The resulting pups (n = 1/postnatal day, P) were killed as described above at 1200 h on P3, P9, P14, and P19, and the intact head (P3–P9) or brains (P14–P19) were frozen on pulverized dry ice and stored at -80 C.
In situ hybridization
Frozen tissues were sectioned at 12 µm in a cryostat
(Reichert-Jung Frigocut 2800N, Leica Corp., Deerfield,
IL). Frontal sections were collected from the cortex of one G16 fetus
per dam, and sagittal sections were collected from the brains of
animals in the developmental study. Sections were thaw-mounted onto
gelatin-coated microscope slides and stored at -80 C until
hybridization. In situ hybridization using the NSP-A RNA
probes was performed as described previously (25), except
that the hybridization buffer contained 2 x
106 cpm probe/slide. In situ
hybridization using the NSP-C oligodeoxynucleotide probe was performed
as described previously (31), except that sections were
immersed for 30 min in 4% formalin, and the hybridization buffer
contained 200 mM dithiothreitol and 200,000 cpm
probe/slide.
Probes
We cloned a fragment of the NSP-A transcript that does not
overlap with that of NSP-C by standard PCR methods using primers
designed to amplify a 202-bp region of NSP-A 1946–2147(1946–2147, accession no.
U17604). Briefly, the forward (5'-AAGGCCTGTGAACCTGAC-3') and reverse
(5'-GGCACGACCTCAGAACCAAG-3') NSP-A primers were synthesized by
Life Technologies, Inc., Custom Primers (Life Technologies, Inc., Gaithersburg, MD). The NSP-A fragment was
amplified from rat genomic DNA using 1 µM of each primer;
0.2 mM deoxy (d)-ATP, dCTP, dGTP, and dTTP; 1.0
mM MgCl2; and 5 U Taq DNA
polymerase (Life Technologies, Inc.). The PCR conditions
after an initial denaturing step at 91 C for 5 min were 91 C for 1.5
min, 60 C for 2 min, and 72 C for 2 min, for 40 cycles. After a final
extension at 72 C for 10 min, the PCR products were stored at 4 C. PCR
products were separated on a 1.2% agarose gel and purified using
GenElute spin columns according to the manufacturer’s instructions
(Sigma). The 202-bp NSP-A gene fragment was then ligated
into pCRII (Invitrogen, Carlsbad, CA), and its
authenticity was confirmed by sequence analysis using ABI
FS-Dye-Terminator chemistry (PE Applied Biosystems, Foster
City, CA). Both complementary and sense strand NSP-A RNA probes were
generated in vitro in the presence of 1 µg linearized
plasmid; 500 µM each of GTP, ATP, and CTP; and
12 µM UTP (UTP and
[33P]UTP at a molar ratio of 1:1). The NSP-A
complementary DNA (cDNA) was linearized with BamHI and
transcribed in the presence of T7 polymerase for cRNA synthesis; it was
linearized with EcoRV and transcribed in the presence of SP6
polymerase for sense strand RNA synthesis. In each case the DNA
template was removed by DNase digestion, and the RNA probe was purified
by standard phenol-chloroform extraction followed by two ethanol
precipitations.
Life Technologies, Inc. Custom Primers synthesized the NSP-C oligodeoxynucleotide. The sequence was complementary to bases 1918–1869 (accession no. L49143). For in situ hybridization, the oligodeoxynucleotide was 3'-end labeled in the presence of [33P]dATP (ICN Biomedicals, Inc., Costa Mesa, CA) using terminal deoxynucleotidyl transferase (Roche Molecular Biochemicals, Indianapolis, IN) as previously described (32). The oligodeoxynucleotide probe was purified by phenol-chloroform extraction followed by two ethanol precipitations.
Probe specificity was evaluated by Northern analysis (Fig. 2
). Total
RNA was extracted from G16 cortical tissue using the method of
Chomczynski and Sacchi (33). The RNA was electrophoresed
with RNA molecular weight standards (Life Technologies, Inc.) on a 1.2% agarose/6.5% formaldehyde gel. RNA was
transferred to a nylon Zeta-Probe membrane (Bio-Rad Laboratories, Inc., Hercules, CA) and cross-linked by baking. NSP-A cDNA was
labeled in both random primer labeling and nick translation reactions
in the presence of [32P]dCTP (ICN Biomedicals, Inc.) according to the manufacturer’s instructions
(Roche Molecular Biochemicals), and both types of probes
were combined for hybridization. The NSP-C oligodeoxynucleotide was
3'-end labeled as described above, except that the reaction was
performed in the presence of [32P]dCTP
(ICN Biomedicals, Inc.). Membranes were briefly
prehybridized, hybridized with either 500,000 (NSP-A cDNA) or 4 x
106 (NSP-C oligodeoxynucleotide) cpm probe/ml
hybridization buffer, and washed according to the manufacturer’s
instructions. The membranes were then apposed to a storage phosphor
screen (Molecular Dynamics, Inc., Sunnyvale, CA) for
either 6 days (NSP-A cDNA) or 23 h (NSP-C oligodeoxynucleotide).
These screens were scanned into a Storm 840 PhosphorImager at 200-µm
resolution and viewed using ImageQuant (Molecular Dynamics, Inc., Sunnyvale, CA).
|
Tissues that were used to characterize the spatial and temporal changes in expression of NSP-A and NSP-C transcripts throughout normal development were dipped in Kodak NTB-3 nuclear tract emulsion. These emulsion autoradiograms were developed in Dektol, fixed in Kodak fixer, and counterstained with methyl green (Sigma). Adjacent sections were counterstained with hematoxylin and eosin (Sigma).
RIA
Free T3 was measured according to the
manufacturer’s instructions using a T3 RIA kit
(ICN Biomedicals, Inc.). This assay was performed at
17.8% binding with detection limits of 0.69–17.5 pg/ml and an
intraassay variation of 17.9%. Total T4 was
measured according to the manufacturer’s instructions using a
T4 RIA kit (ICN Biomedicals, Inc.).
This assay was performed at 33.8% binding with detection limits of
2.0–20 µg/dl and an intraassay variation of 19.9%. Serum levels of
TSH were measured using [125I]rat TSH
(Covance Laboratories, Inc., Vienna, VA) and the double
antibody NIDDK RIA reagents, including RP-3 standards. This assay was
performed at 26.3% binding with detection limits of 1.0–30 ng/ml. All
samples were measured in duplicate in the same assay.
Statistical analysis
Outliers, defined as those values exceeding 1.5 interquartile
ranges from the upper and lower quartiles, were eliminated using a box
and whisker plot (Statistix, Analytical Software, Tallahassee, FL). A
two-way ANOVA was performed on hormone levels and imaging data using
the StatView statistical package (Abacus Concepts, Berkeley, CA), with
main effects of goitrogen treatment (MMI or PTU) and timing of acute
T4 exposure (MMI experiment) or
T4 injections (PTU experiment). The ANOVA was
followed by t tests between individual means.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
].
Free T3 levels were significantly elevated by
T4 injection in PTU-treated animals, but were
unaffected by T4 in euthyroid animals (Fig. 1).
However, total T4 [F
(1,19) 140.105; P < 0.001] was
elevated, and TSH was suppressed [F(1,17)
13453.487; P < 0.001] by the T4
injection in both PTU-treated and control dams.
|
). The NSP-A
complementary RNA probe hybridized to a single size class of RNA of
approximately 3.5 kb, whereas the NSP-C oligodeooxynucleotide
hybridized to a single size class of RNA of approximately 1.5 kb.
Quantitative analysis of film autoradiograms after in situ
hybridization of NSP-A revealed that chronic MMI
[F(1,18) 9.181; P < 0.008; Fig. 3A] or PTU
[F(1,19) 5.586; P < 0.03; Fig. 4] significantly increased NSP-A
expression in the G16 cortex. There was no effect of
T4 injection on NSP-A expression within the time
period examined. NSP-C expression was not affected by treatment in the
G16 cortex (Fig. 3B). Because MMI did not affect the expression of
NSP-C mRNA (25), we did not analyze the effect of PTU on
NSP-C expression.
|
|
). Both
transcripts were clearly detected on G15, the earliest time point
examined, and they increased in abundance progressively through
postnatal day 19. Film autoradiograms of frontal sections examined by
in situ hybridization on G16 (Fig. 3, A and B) indicated
that NSP-A and NSP-C transcripts were expressed in different parts of
the cortex at this time. This was confirmed by analysis of the liquid
emulsion-coated slides of G16 sections (Fig. 6). The emulsion autoradiograms
demonstrated that NSP-A mRNA is selectively expressed in the
proliferative ventricular zone of the G16 cortex. The abundance of
NSP-A expression in this layer diminishes by G18 when cortical
proliferation is nearly complete and the ventricular zone disappears
(35). As development proceeds, NSP-A expression becomes
expressed in the outer layers of the cortex (Figs. 6 and 5) and even
takes on a laminar appearance during postnatal development. In
contrast, NSP-C appears to be expressed exclusively in the outer,
intermediate zone of the cortex on G16 (Fig. 6) and, as development
proceeds, exhibits robust and continuous expression in these outer
layers. The film autoradiograms also indicated that NSP-A and NSP-C are
differentially expressed in the nasal epithelium, olfactory bulb and
cerebellum (Fig. 5). We further characterized the expression of NSP-A
and NSP-C in the developing cerebellum (Fig. 7). NSP-A was exclusively expressed in
cerebellar Purkinje cells throughout development, whereas NSP-C was
expressed in granule cells, but was absent from Purkinje cells.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We previously reported that the NSP-A transcript was expressed in the intermediate zone of the fetal cortex (25). However, our previous conclusion was based on results using a cRNA probe that hybridized to both NSP-A and NSP-C mRNAs. Apparently, this probe gave us misleading results, because the transcript encoding NSP-C is much more abundant than that encoding NSP-A. Therefore, the signal for the NSP-C transcript in liquid emulsion was robust before the signal for NSP-A was visible, which gave us the impression that both NSP transcripts were expressed in the intermediate zone. The preparation and use of specific probes for NSP-A and NSP-C allowed us to clearly determine that NSP-A is exclusively expressed in the ventricular zone of the fetal cortex. We also found previously that a single injection of T4 significantly reduced the abundance of NSP-A mRNA in the G16 cortex (25). This conclusion was based on Northern analysis and in situ hybridization. However, we also observed a trend for NSP-C to be reduced by T4 that may have contributed to the reduction in NSP-A/C we observed in the in situ hybridization. The use of probes specific for the two NSP transcripts in the present experiments precludes the possibility of this kind of cross-contamination. Using these specific probes, we now find no evidence for an effect of T4 on NSP-A expression within 24–36 h. The ability of a single injection of T4 to reduce cellular levels of a specific mRNA within a given time will necessarily depend on the overall degree to which T4 reduces its transcription and the half-life of the mRNA in the cell. Therefore, it is likely that a single injection of T4, regardless of dose, does not suppress NSP-A expression in the G16 cortex to the extent that in situ hybridization can resolve it within 24 h. This does not imply, however, that a single injection of T4 does not affect gene expression in the G16 cortex, because we previously showed that Oct-1 expression was significantly elevated within 12 h of the T4 injection (25). Clearly, the logistical issues surrounding the measurement of gene induction differ from those surrounding the measurement of gene repression.
The two experiments we now describe differed in the timing of goitrogen treatment in that MMI was initiated 2 weeks before the females were mated, whereas PTU was initiated in pregnant females on G2 (see Materials and Methods). Considering that NSP-A expression was affected by the two treatments to the same extent, the present results suggest that goitrogen treatment initiated before conception does not produce more severe effects on gene expression in the G16 fetus than goitrogen treatment initiated after conception. This is important in part because it may indicate that hypothyroidism before pregnancy does affect the course of brain development. However, it is also important because it represents a refinement in our model of gestational hypothyroidism and the consequences on gene expression in and development of the fetal brain. This model allows us to focus on the effects of maternal thyroid status on fetal brain development despite the fact that both MMI and PTU can cross the placenta (39, 40). Our reasoning is that fetal thyroid function does not begin in the rat until G17 (37), so the only source of thyroid hormone to the fetus on G16 is the maternal system. Therefore, this model allows us to study the effects of maternal hypothyroidism on fetal brain development without the confounding influence of effects on the fetal thyroid.
The developmental significance of the present findings will require in part that we identify the specific subpopulation of neuroblasts that selectively express NSP-A. The ventricular zone of the fetal cortex is a pseudostratified germinal epithelium containing cells that are actively dividing and cells that have stopped dividing and are beginning to differentiate (35, 41, 42, 43, 44). Therefore, it will be important to determine whether NSP-A is expressed in actively proliferating neuroblasts or in those neurons that have begun to differentiate but have not yet left the ventricular zone. Neuroendocrine-specific proteins are anchored in the membrane of the endoplasmic reticulum (45, 46, 28) and may be important for neuronal differentiation and axonal guidance (47). Thus, it is possible that the extended N-terminus of NSP-A (27) confers a function to this protein that is important in cortical neurogenesis. Although the magnitude of the effect of maternal goitrogen treatment on NSP-A expression was relatively small, it is likely that this effect is an underestimation of the actual effect. NSP-A-positive cells are quite restricted in their distribution, producing a very narrow signal on film that saturates quickly.
We found in the present study that NSP-A mRNA is expressed in several
brain areas that differ from those in which NSP-C is expressed
throughout development. For example, the film autoradiograms indicate
that NSP-A is expressed in restricted regions of the nasal epithelium,
olfactory bulb, and cerebellum compared with NSP-C (compare E with F,
and G, I, and K with H, J, and L in Fig. 5). We further characterized
the differential distribution of NSP-A and NSP-C expression in the
cerebellum (Fig. 7) and found that during postnatal development, NSP-A
expression is always restricted to Purkinje cells, and NSP-C expression
is always absent from Purkinje cells. This may be particularly
important because neonatal hypothyroidism permanently affects the
maturation of Purkinje cells by decreasing arborization and numbers of
dendritic spines (48). Because the Purkinje cells are the
major efferent neurons from the cerebellar cortex (49),
these morphological effects of hypothyroidism may contribute to the
motor deficits associated with neonatal hypothyroidism (50, 51). The selective expression of NSP-A in Purkinje cells and its
regulation by thyroid hormone (Dowling, A. L. S., in
preparation) clearly suggest that NSP-A may play a role in the effects
of Purkinje cell morphology and on animal behavior.
The present findings demonstrate that NSP-A exhibits a different pattern of expression in the developing brain compared with that of NSP-C. This is especially important in the early period of cortex development, when NSP-A is selectively expressed in the germinal ventricular zone. In addition, NSP-A is selectively expressed in cerebellar Purkinje cells. Finally, the present findings demonstrate that induction of maternal hypothyroidism with MMI or PTU affects NSP-A expression in the proliferative zone of the fetal cortex before the onset of fetal thyroid function. It is important to recognize that this observation does not allow us to conclude whether thyroid hormone exerts a direct effect on NSP-A expression. One can imagine several indirect mechanisms by which thyroid hormone could affect NSP-A expression in the fetal cortex. However, it is possible that thyroid hormone from the maternal system can reach the neuroblasts of the fetal cortex, activate the thyroid hormone receptor, and directly repress NSP-A expression. Thyroid hormone from the maternal circulation can cross the placenta and gain access to fetal tissues (52, 53, 20). In addition, the ß1 thyroid hormone receptor is selectively expressed in the proliferative zone of the fetal cortex on G16 (18). Moreover, considering that maternal hypothyroidism affects the expression of NSP-A, but not that of NSP-C, and that thyroid hormone appears to suppress NSP-A expression, but enhance Oct-1 expression, in the same tissue (25), it appears that thyroid hormone exerts specific effects on gene expression in the fetal brain.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received July 19, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and ß-thyroid hormone receptor mRNAs,
including the ß2-subtype, in the developing
mammalian nervous system. J Neurosci 12:2288–2302[Abstract]
-1 thyroid hormone
receptor in fetal rat brain. Alcohol Clin Exp Res 22:2111–2117[Medline]
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]
|
|
|
|
 |
|
 R. A. Sinha, A. Pathak, V. Mohan, S. Bandyopadhyay, L. Rastogi, and M. M. Godbole Maternal Thyroid Hormone: A Strong Repressor of Neuronal Nitric Oxide Synthase in Rat Embryonic Neocortex Endocrinology, September 1, 2008; 149(9): 4396 - 4401. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Auso, R. Lavado-Autric, E. Cuevas, F. E. del Rey, G. Morreale de Escobar, and P. Berbel A Moderate and Transient Deficiency of Maternal Thyroid Function at the Beginning of Fetal Neocorticogenesis Alters Neuronal Migration Endocrinology, September 1, 2004; 145(9): 4037 - 4047. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Calvo, E. Jauniaux, B. Gulbis, M. Asuncion, C. Gervy, B. Contempre, and G. Morreale de Escobar Fetal Tissues Are Exposed to Biologically Relevant Free Thyroxine Concentrations during Early Phases of Development J. Clin. Endocrinol. Metab., April 1, 2002; 87(4): 1768 - 1777. [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 |
, euthyroid dams (no MMI); 
, hypothyroid dams (MMI). Sense
controls produced no detectable signal (not shown). V, Ventricle; Cx,
cortex. a, Significantly different from euthyroid animals receiving no
injection (P < 0.05). b, Significantly different
from euthyroid animals receiving T4 injection at 2100
h on G14 (P < 0.05).
