|
|
|
|||||||
|
|||||||||
From the
Thyroid hormone 3,5,3`-triiodo-L-thyronine
(T
During vertebrate development thyroid hormones exert pleiotropic
effects on various tissues including the central nervous
system
(1) . In the chick and the rat, experimental
hypothyroidism causes impaired development of the central nervous
system, and in man hypothyroidism results in cretinism, with retarded
physical and mental growth
(1) . Thyroid hormones and in
particular the biologically most active form,
3,5,3`-triiodo-L-thyronine (T
In the
chick, as in other species, two related genes encode TRs, TR
Many morphological, biochemical, electrophysiological, and
behavioral defects result from hypothyroidism during development,
especially if it occurs during a hormone-sensitive period. Thyroid
hormone requirements have been implicated in differentiation of
neuronal and glial populations, affecting cell migration, myelination,
and synaptogenesis
(1, 17) . However, little is known
about specific target cells for hormone action nor of the roles of the
different receptors. In particular, there are virtually no data on the
requirements for T
To test the hypothesis we used the positive effect of thyroid
hormones on division of neuroblasts from embryonic chick optic lobe as
a model system to measure the actions of antisense
oligodeoxynucleotides (ODNs) directed specifically against chick
TR
[
Fertilized white Gallus domestic eggs (Haas, Kaltenhouse, France)
were incubated at 38 °C ± 0.5 °C. In one experimental
series, eggs were injected at embryonic day 5 with 100 µl of NaCl
(8.6‰) containing 0.9 MBq [
DNA synthesis was assayed at 2 days in vitro either by
scintillation counting or by determining the number of labeled cells.
For scintillation counting cells were seeded at 10
To determine the number of labeled cells, 10
As we had previously determined that lipospermine
in creases ODN uptake by neuroblasts and that when 1 µM
ODNs are applied over 95% is within the cell, we tested the effects of
the ODNs at concentrations <1 µM. ODNs with an 18-mer
randomized sequence of the same A, T, G, C content as AS
The data presented here show that T
The next question was that of the TR
responsible for mediating these effects. We employed an antisense
strategy to modulate differentially the expression of TR
We
applied antisense ODNs to neuroblasts in primary cultures in defined
SFM. We first verified that T
Our next
step was to modulate the expression of TR
The question arose as to whether the antisense TR
These observations suggest that
if low levels of TR
Taken together our results indicate
that T
Certain fundamental questions on TR roles during development remain:
what factors determine the distinctive expressions of the
In conclusion, these data suggest that TR
We thank Anne Marie Ableitner for densitometry
readings and Hanem Abdel Tawab for histology.
Volume 270,
Number 20,
Issue of May 19, pp. 12100-12108, 1995
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
Thyroid Hormone
Receptor Oligonucleotides (*)
) is required for normal brain development in
vertebrates. T acts through two classes of nuclear
receptors (TR
and TR
) that have distinct developmental
spatial and temporal distributions suggesting different functions
during neuronal development. One possibility is that TR, which is
expressed early in embryogenesis, is involved in neuroblast
proliferation. To test this hypothesis we used the embryonic chick
optic lobe, as we found that T stimulates
[H]thymidine incorporation in this tissue both
in vivo and in vitro during embryonic days 6-9.
We applied oligonucleotides (ODNs) against TR and TR to
primary cultures of chick optic lobes. By employing a cationic lipid
vector we could use very low ODN concentrations (<150 nM).
Antisense ODNs against TR significantly inhibited
[H]thymidine incorporation, whereas antisense
TR had no significant effect. However, both ODNs inhibited
expression of TRs, as they blocked transcription from a
T-activated reporter gene. Random ODNs used as controls had
no significant effect on [H]thymidine
incorporation or on T-dependent transcription. These
observations suggest that TR is implicated in neuroblast
proliferation and add credence to the hypothesis that the multiplicity
of nuclear receptors allows for specific actions of T during development.
),
()
exert most of their diverse effects via nuclear receptors
(TR)
(2) , the c-erbA proto-oncogenes, belonging to the
steroid hormone receptor superfamily (3-5). TRs are transcription
factors that modulate gene expression by binding as dimers to specific
DNA sequences known as T response elements, thereby
regulating the transcription of target genes
(6) .
and
TR
(3) . Both genes are expressed in the developing chick
brain (7), but as in the rat
(8, 9, 10) they
have distinct temporal and spatial profiles of expression
(11) ,
suggesting they have different developmental roles. Moreover, TRs bind
to the TRE sequences in target genes as homodimers or
heterodimers
(12, 13) . Heterodimers can be composed of
either TR and TR, or of a TR receptor with another class of
nuclear factor, particularly the 9-cis-retinoic acid
receptors, the RXRs (14, 15; for review, see Ref. 16). These multiple
possibilities generate great potential for diversity of signal
response. Cellular response depends on the expression of a given set of
receptors at a given time and the presence or absence of the respective
ligands. Interestingly, in the case of the T and the
retinoic acid systems, expression of different receptor classes and
production of ligands are both developmentally regulated processes.
and its actions during the proliferative
stages of neurogenesis. Given that TR is expressed during the
early mitotic phases of development, one hypothesis would be that this
receptor has a specific role to play in regulating entry into the cell
cycle.
or TR. T increased proliferation of
neuroblasts both in vivo and in vitro. However, we
tested the actions of the TRs in vitro, since an in vivo approach would require homogenous, highly efficient delivery of
the ODNs to the whole of the neuro-epithelium population, which is not
yet technically feasible. Moreover, the use of ODNs to modify gene
expression is limited by their low cellular uptake and their rapid
degradation by extracellular nucleases. To protect the ODNs and enhance
uptake, we complexed ODNs with a cationic lipid,
dioctadecylamidoglycylspermine (Transfectam, DOGS). DOGS
compacts the DNA and neutralizes the anionic charges carried by the ODN
that interfere with its crossing of the plasma membrane
(18) .
Some recent reports have shown that cationic lipids do improve uptake
of ODNs by cell lines
(19, 20) , but no demonstration of
their usefulness for introducing ODNs into primary neurons has
appeared. By using this approach we found that we could introduce ODNs
(at nanomolar concentrations) into primary cultures of embryonic
neurons prepared during the proliferative phase and examine their
biological effects. Our findings show that TR
is implicated in
neuronal proliferation, a result which conforms with the expression of
this gene during the early stages of development.
H]Thymidine Incorporation in Vivo
H]thymidine
([5`-H]thymidine, specific activity 488 GBq,
Amersham, Les Ulis, France) with or without T (3 µg).
This amount of T increases embryonic tissue levels of
T 5-fold, from 48 pg/g in saline-treated controls to 243
pg/g with T treatment (values determined by
radioimmunoassay on pools of six embryos).
()
Injections were made through the shell into the yolk sac.
On embryonic day 8 embryos were fixed in Carnoy's solution (1 h)
dehydrated through alcohol then embedded in paraffin. Coronary sections
(7 µm) mounted on gelatin-coated slides were left to dry (37
°C, 3 days) then deparaffinized in xylene (three times), followed
by 95% alcohol (three times), and dried overnight at room temperature.
Slides were placed in contact with H-sensitive film
(Hyperfilm-H, Amersham, Les Ulis, France) for 6 weeks. The
films were analyzed with a densitometer. In a second series of
experiments eggs were injected at embryonic day 5 with 100 µl of
saline (NaCl 8.6‰) ± T (3 µg). On
embryonic day 8, [H]thymidine (0.9 M Bq
in 100 µl of saline) was deposited on the extra-embryonic
circulation. Embryos were sacrificed 2 h later, fixed, and sections cut
as described above. Slides dipped in autoradiographic emulsion (NTB-2,
Kodak, Les Ulis, France) were developed 6 days later. Labeled cells in
the internal and external layers of the optic tectum were counted,
using a 100 objective on a Leitz microscope. For
standardization, a 90-µm wide strip within the ventrolateral area
of a median section (i.e. at the middle of the rostrocaudal
extent of the optic tectum) was counted for both layers.
Primary Neuroblast Cultures
Cells were prepared as described
(21) . Briefly, optic
lobes were dissected from chick embryos at embryonic day 6 (37.5
± 0.5 °C) and transferred to Petri dishes containing a
dissection medium composed of a 1:1 mixture of Dulbecco's
modified Eagle's medium and Ham's F12, supplemented with 25
mMD-glucose, 50 µg/ml penicillin, and 100
µg/ml streptomycin. After removing the meninges, tissue was minced
and mechanically dissociated. Dissociated cells were suspended in
serum-free medium (SFM, 21). For immunocytochemistry and ODN
incorporation, cells were seeded on glass coverslips at 4.10
cells/well. Dishes were pretreated for 30 mn with gelatine (0.25
mgml
), and overnight with
poly-D-lysine (M
70,000, 10
mgml
in 0.15 M sodium borate buffer,
pH 8.0). After rinsing with 0.1 M phosphate-buffered saline
(PBS) and dH
O, dishes were coated with a mixture of
Dulbecco's modified Eagle's medium-Ham's F12 (1:1)
supplemented with 10% fetal calf serum (FCS) for 2 h. Cultures were
grown at 37 °C in humid 93% air, 7% CO.
Oligodeoxynucleotide Uptake and Stability
[P]ODN Incorporation and
Stability-[
P]ODN incorporation and
stability were measured according to Wickstrom et
al.(22) . AS
was 5`-labeled with
[-P]ATP (Amersham, Les Ulis, France) by
using T4 polymerase kinase (Amersham). For each point 2
10
counts/min of 5`-labeled ODN (1 pmol of labeled ODN and
10 µM carrier unlabeled ODN complexed with a three times
charge excess of DOGS or 10 µM carrier unlabeled ODN) was
added to 4.10 cells. After a further 12 h, labeled ODN was
recovered from cells and culture medium and radioactivity determined by
scintillation counting. To evaluate ODN stability, aliquots of cell
extracts and culture medium were electrophoresed in a denaturing 20%
polyacrylamide gel and exposed to X-OMAT film (Kodak, Les Ulis, France)
for 24 h.
Intracellular Localization of Rhodamine-linked ODN
For
fluorescence microscopy, 1 µM AS (final
concentration) labeled in 3` with rhodamine was complexed with DOGS (at
a three times charge excess, see below) and added to the cells. After 6
h, cells were fixed in 4% formaldehyde PBS for 30 mn. After rinsing
twice with 0.1 M PBS and distilled water, coverslips were
mounted in Moviol (Hoechst) and visualized by fluorescence, Nomarski
optics, and confocal microscopy.
ODNs
The translation start regions of cTR and cTR mRNA
were chosen as targets to block cTRs protein expression (see Refs. 3
and 7 for cDNA sequences). ODNs, purified on SDS-polyacrylamide gel
electrophoresis, were purchased from Genosys (United Kingdom) and
Eurogentec (Belgium). Control ODNs with an 18-mer randomized sequence
of the same A, T, G, C content as antisens (AS) and antisens
(AS) were, respectively, designated as random
(RD) and random (RD). To follow ODN incorporation and
intracellular distribution, AS was labeled in 3` with rhodamine
(Eurogentec). Sequences were as follows: AS,
5`-GCTGGGCTTCTGTTCCAT-3`; RD, 5`-GTCGGCGTTCTGTTATCG-3`; AS,
5`-ATATACCCTGACATACTG-3`; RD, 5`-ATATACCTCCAGATAGTC-3`.
Plasmids
The T3RE-LUC construct contains a synthetic ODN sequence
encoding a palidromic TRE in SV-luciferase
(23) . It was kindly
provided by Dr. C. K. Glass, San Diego.
Transfections
The lipid used was DOGS. DOGS (135 µM
DOGS/
µM ODN) was incubated in 150 mM
NaCl with ODN for 10 min at room temperature. The mixture was added
directly to the cells and incubated together for a given time. For
plasmid transfection, cultures were transfected according to published
methodology
(21) . DNA was complexed with DOGS (1 µl 4
mM DOGS/µg DNA) in 100 µl of 0.015 M NaCl,
then diluted into the appropriate volume of SFM. After 1 h the
transfection medium was replaced with fresh SFM. Transfection
efficiency is high and very regular
(21) . Thus, there has never
been any need to normalize for transfection efficiency in any cells so
far transfected by this method. [5`-
H]Thymidine Incorporation Assay in Vitro
cells/well in 24-well plates. At 1 day in vitro,
neuroblasts were labeled with 37 kBq of
[5`-H]thymidine (specific activity 488 GBq,
Amersham, Les Ulis, France) for 24 h. ODNs (25-150 nM,
complexed with three time charge excess of DOGS, see above) were added
at 24 h. At 48 h, cells were rinsed twice with SFM and lysed with 1%
sodium dodecyl sulfate. Homogenates were counted in a scintillation
counter.
cells/well were plated in 6-well plates. The number of labeled
cells represents the proportion of cells incorporating
[H]thymidine in the nucleus, reflecting DNA
synthesis. Cultures were incubated with 37 kBq of
[5`-H]thymidine from 1 to 2 days in
vitro. ODNs (150 nM, complexed with three times charge
excess of DOGS, see above) were added at 24 h. At 2 days in
vitro, cells were processed for autoradiography. Briefly, cells
fixed in 4% formaldehyde PBS (30 mn), were dehydrated in ethanol,
dipped into NTB2 emulsion (Kodak, Les Ulis, France) at 43 °C,
dried, and exposed in the dark at 4 °C for 8 days. Slides were
developed in Kodak DEKTOL, fixed in Kodak UNIFIX, and washed in water.
Finally, slides were lightly stained for cell bodies and nucleus using
eosin, hematoxylin, and coverslipped with DPX (Electron Microscopic
Sciences). Positive and negative cells in seven or eight random areas
were scored for silver granules at 100 magnification on each
well until 500-1000 cells were counted. The percentage of positive
cells was then calculated.
Luciferase Assay
Luciferase activity was measured using the Promega Luciferase
assay system according to the manufacturer's instructions using a
Tropix luminometer (MGM Instruments, Hamden, CT)
calibrated against a tritium standard. Light emitted was measured over
10 s.
Immunocytochemistry
TR Labeling
Cultures were fixed at 3 days in
vitro with ethanol, rinsed in 0.1 M PBS, supplemented
with 5% FCS and 0.05% sodium azide (PBS/FCS/NaA), then incubated with a
1:100 dilution of a polyclonal antiserum against cTRs (24). Controls
for the cTRs antibody were incubated in preimmune rabbit serum. After
incubation overnight at room temperature, cells were washed (three
times) in PBS/FCS/NaA and incubated (1 h, room temperature) with a
1:100 dilution of biotinylated anti-rabbit Ig (Amersham, Les Ulis,
France). After three rinses cells were incubated (15 mn) with
fluorescein-conjugated streptavidin (1:100, Amersham). Coverslips were
mounted in Moviol.
Neurofilament Labeling
Cultures at 4 days in
vitro were fixed and treated as above, then incubated with a 1:5
dilution of a monoclonal antibody against the 200-kDa neurofilament
subunit (Boehringer Mannheim). After incubation overnight at 4 °C,
cells were washed and incubated for 1 h (20 °C) with anti-mouse Ig
linked to fluorescein (1:100, Amersham).
Statistical Analysis
Student's t test was used to calculate
differences between means. Differences were considered significant when
p 
0.05.
T
To test the effect of
increasing T Increases Proliferation in the Optic
Tectum of 8-Day-old Chick Embryos levels on neuroblast proliferation, we
injected 3 µg of T into the yolks of 5-day-old chick
embryos and assessed the incorporation of
[H]thymidine at embryonic day 8.
[H]thymidine incorporation was found in the inner
and external layers of the optic tectum of both control and treated
embryos in both experiments (Fig. 1, a-c). Densitometry
showed that T injection at day 5 significantly increased
[H]thymidine incorporated into both inner and
external layers of the optic tectum (p < 0.05 in each
case).
Figure 1:
T treatment increases
[H]thymidine incorporation in vivo.
a, optical densities of internal and external layers of the
optic tectums of control and T-treated chick embryos
sacrificed at embryonic day 8. At embryonic day 5, embryos were
injected with physiological saline (100 µl containing 0.9 MBq
[H]thymidine). The T-treated embryos
also received 3 µg of T (dissolved in the 100 µl of
saline with the thymidine). Means ± S. E. are given, n = 2 embryos/group. Six readings per embryo were made.
b, coronary sections of optic lobes from control (b,
left) and T-treated chick embryos (b,
right) sacrificed at 8 days of incubation. At day 5 of
incubation, control embryos were injected with saline (100 µl), and
T-treated embryos received 3 µg of T (in
100 µl of saline). All embryos were injected with 0.9 MBq
[H]thymidine 2 h before sacrifice. I,
internal layer; E, external layer. Arrows show the
few labeled in cells in the external layer. Bar = 500
µm. c, numbers of radiolabeled cells in the internal and
external layers of optic lobes from control and T-treated
chick embryos sacrificed at 8 days of incubation. Experimental methods
as for b. Means ± S.E. are given, n = 6
embryos/group. ** = p < 0.01 (Student's t test).
Applying [H]thymidine just prior to
sacrifice allowed us to assess the number of cells synthetizing DNA
within a 2-h period. When the number of cells labeled with
[H]thymidine was counted in the inner and
external layers of optic tectum from saline- and T-treated
embryos, we found similar results to those obtained by densitometry. As
seen in Fig. 1, b and c, most labeled cells
were in the inner zone. T significantly increased numbers
of cells labeled in this inner zone as compared to the saline controls
(p < 0.01).
T
Our aim in these experiments was to use an
antisense approach to evaluate the contributions of different TRs to
neuroblast proliferation. Given the wide distribution of the neurogenic
precursors throughout the neuroepithelium it is apparent that efficient
and homogenous delivery of ODNs to such an extensive area is currently
technically unfeasible. Thus, we applied ODNs to neurogenic precursors
maintained in primary culture, checking first that T Increases Proliferation of Neurogenic
Precursors in Vitro increased [H]thymidine incorporation under
these conditions (Fig. 2, a and b).
Figure 2:
T increases
[H]thymidine incorporation in neuroblasts in
vitro.Primary cultures of neuroblasts from optic lobes
of 6-day-old chick embryos were cultured in the presence of increasing
concentrations of T (added immediately after plating) and
[H]thymidine (37 kBq) added after 24 h. After a
further 24 h, cells were either lysed and total radioactivity counted
(a) or fixed and the percentage of labeled cells counted after
autoradiography (b). Means ± S.E. are given, n = 6/group. * = p < 0.05 (Student's
t test).
A
dose-dependent response to T was found
(Fig. 2a). A significant effect over controls (no
T added) was seen at 1 nM, with maximum effect at
10 nM T. Increasing T concentrations
to 1 µM, a pharmacological level, decreased
[H]thymidine incorporation as compared to that
seen at 100 nM T. Significant effects of T on neuroblast proliferation were found by assessing both the
total amount of [H]thymidine incorporated into
the cultures (Fig. 2a) and the percentage of cells
labeled on histological examination (Fig. 2b). In both
cases, the maximum T-dependent increase in proliferation
was approximately 30%.
Use of Lipospermine Increases Efficiency of ODN Delivery
to Embryonic Neuroblasts
A major problem with ODN delivery is
their instability due to nucleases in the intra- and extracellular
compartments. Compaction with a cationic lipid (DOGS) should increase
uptake and reduce exposure to extracellular nucleases.
Fig. 3b, shows that complexing ODNs with a three time
charge excess of DOGS increased the percentage of ODNs found in the
intracellular medium 12 h after application from <2% when applying
ODNs alone, to >95% when complexed with lipid. Moreover, gel
electrophoresis of intracellular and extracellular extracts showed that
if ODNs were applied without DOGS, as the majority remained in the
extracellular compartment, they were completely degraded 12 h after
application and no signal was seen in the intracellular compartment
(Fig. 3a, lane 4). In contrast, when ODNs are
applied with DOGS, no signal is found in the extracellular compartment,
and the signal from the intracellular medium shows a single band
migrating at the level of an 18-mer (Fig. 3a, lane
3).
Figure 3:
Uptake of oligonucleotides in the presence
and absence of DOGS. a, autoradiogram of denaturing
polyacrylamide gel of nucleic acids extracted from intra- and
extracellular medium of cells treated with P-labeled
oligonucleotide (10 µM) either alone or complexed with
DOGS. B and X indicate, respectively, the migration
levels of bromphenol blue and xylene (comigration with 28-mer ODNs and
8-mer ODNs, respectively). b, percentage of
P-labeled oligonucleotide in the intracellular and
extracellular compartments of cultures treated with oligonucleotide (10
µM), supplied either alone or complexed with DOGS. For
both this experiment and that depicted in a, ODNs were applied
18 h after plating the cells and cells lysed 12 h later. Means ±
S.E. are given, n = 6
To determine whether the ODNs applied with DOGS were
actually within the cells and not just adhering to the membrane, we
used a rhodamine-labeled ODN and examined its distribution in cultures
exposed to ODN alone and ODN complexed with a three times charge excess
of DOGS. Fluorescent optic microscopy (Fig. 4a) and
confocal microscopy (data not shown) showed that when the ODNs were
applied with the lipospermine vector, a strong signal was detected in
the cytoplasm. However, in the cultures exposed to ODNs applied alone
no signal could be detected.
Figure 4:
a, intracellular distribution of a
rhodamine-labeled oligonucleotide. ODNs were applied complexed (1
µM) with DOGS. Cells were exposed to ODNs 18 h after
plating, then fixed after a further 6 h. Two representative fields are
shown. Cells were examined with Nomarski optics (left,
upper and lower) and appropriate filters for
detecting rhodamine fluorescence (right, upper and
lower). In cultures where ODNs were applied without DOGS, no
fluorescent signal was found (data not shown). Bar =
2.5 µm. b, characterization of neuronal cultures. Cultures
maintained for 4 days in vitro were fixed and
immunocytochemistry carried out for neurofilament protein. A homogenous
population of neurofilament protein-positive cells is revealed. Bar = 150 µm.
Antisense ODNs against TR
We next tested the effects of antisense
ODNs against different TRs on the T Inhibit Neurogenic
Precursor Proliferation-dependent proliferation
of embryonic neuroblasts. First, we confirmed that the culture
conditions gave rise to a wholly neuronal population. As shown in
Fig. 4b, neurofilament immunocytochemistry carried out
on cultures at 4 days in vitro showed a homogenous population
of neuronal cells.
and
AS were used as controls and were, respectively, designated as
RD and RD. We found that 150 nM AS inhibited
[H]thymidine incorporation by 46%
(Fig. 5a). This inhibition was significant (p < 0.05). Application of the same amount of AS inhibited
proliferation by approximately 30%, but this decrease was not
significant. Both the random oligonucleotides had small,
non-significant effects on neuroblast proliferation
(Fig. 5a). In some experiments the sense controls were
used instead of random sequences, and again these ODNs did not have any
significant effect on proliferation (data not shown).
Figure 5:
Effect of oligonucleotides directed
against chick TR or TR on TH-dependent mitosis. a,
cultures were incubated with 37 kBq of
[5`-H]thymidine from day 1 to day 2 in
vitro. ODNs (150 nM, complexed with three times charge
excess of DOGS) were added at 24 h. At 48 h in vitro cells
were processed for autoradiography. Labeled cells were counted with the
[times 100 objective of a Leitz microscope. b, dose
response: at 1 day in vitro, neuroblasts were labeled with
[H]thymidine and varying concentrations of
antisense TR (AS) or a control, random ODN
containing the same proportion of bases as the antisense
(RD). ODNs were complexed with a three times charge
excess of DOGS (see ``Experimental Procedures''). After 24 h
cells were rinsed twice with culture medium and lysed with 1% SDS.
Homogenates were counted in a scintillation counter. Means ±
S.E. are given, n = 6/group. Both experiments were
repeated three times; representative results from one experiment are
shown. * = p < 0.05 (Student's t test).
Antisense TR
Next we carried out a dose-response study
using TR Inhibits Neuroblast Proliferation in a
Dose-dependent Manner antisense ODN concentrations between 25 and 150
nM. AS significantly and maximally inhibited (p < 0.05) proliferation of the neurogenic precursors at 100
nM producing 30% inhibition of
[H]thymidine incorporation.
(Fig. 5b). The control ODN, RD, had no significant
effect on proliferation at any of the concentrations used
(Fig. 5b).
Oligonucleotides against Both TR
To examine whether both AS sequences
(AS and TR Inhibit
T3-dependent Transcription and AS) were functional, we used a cotransfection assay,
where first we transfected in the cells a reporter construct containing
a TRE linked to the luciferase reporter gene, and 1 h later transfected
in the antisense ODNs or their respective controls. Cells were exposed
to T (10 nM) for a further 24 h and then assayed
for luciferase activity. As shown in Fig. 6, addition of T increased transcription 1.45-fold in controls, and both antisense
ODNs inhibited this T-dependent transcription from the TRE
construct (p < 0.05, in both cases as compared to
controls). Transfection of the random controls had no significant
effect on T-dependent transcription.
Figure 6:
Both As and AS inhibit
T-dependent transcription. Transcription from a
TRE-luciferase was measured in the absence of ODN (controls,
Ct) or presence of ODNs directed against chick TR
(AS) or TR (AS) or random
(RD) controls containing the same proportion of bases as
AS (RD) or AS (RD). Means
± S.E. are shown, n = 9/group, The results are
combined from three separate experiments. * = p <
0.05 (Student's t test).
Antisense TR
To examine whether the AS Inhibit Expression of TR Receptor
Protein ODNs were functional in
decreasing the levels of TR protein, we used immunocytochemistry
as the levels of protein expression were insufficient for
immunoblotting. Cultures were grown in the presence of T and treated with lipospermine-complexed ODNs 36 h after plating.
After a further 24 h, cells were fixed and immunocytochemistry carried
out employing an antibody specific for c-erbA- raised in
rabbit
(24) . We found expression of a nuclear-located protein in
most of the cells in cultures exposed to 0.15 µM RD
(Fig. 7a). A similar nuclear reaction was revealed in
control cultures not treated with ODNs (data not shown). In contrast,
in cultures treated with 0.15 µM AS, a nuclear signal
was much reduced (Fig. 7b). Cultures treated with
preimmune serum showed no nuclear reaction (data not shown).
Figure 7:
Antisense ODNs against TR inhibit
expression of thyroid hormone receptors. Immunocytochemistry against
c-erb-A was performed on cultures treated with a random (RD)
ODN (a) or an ODN against chick TR (b). Note
reduced nuclear signal in b. Bar = 2,5
µm.
DISCUSSION
Our aim in these experiments was to analyze T effects on neuroblast proliferation and to determine the type of
TR implicated in the response. We used neuroblasts from the chick
embryo optic lobe, and our analysis was carried out during embryonic
days 6-8 (or its equivalent in vitro). This window of
time was chosen for four reasons. First, it is a period during which
mitosis in the optic lobe is still occurring, albeit at a lower rate
than during days 4-6
(25, 26, 27) . Second,
it is the earliest stage when embryonic T levels can be
experimentally increased by injecting T into the yolk
without inducing embryonic mortality.()
Third, it
is the earliest stage at which pure neuronal cultures can be
prepared
(28) . Fourth, TR is first detected in the
embryonic chick brain at this stage
(7) . Indeed this observation
of TR in a number of tissues, including the optic lobe
(11) during early chick embryogenesis, coupled with the finding
that TR is expressed in a more restricted fashion during later in
development, led Forrest and co-workers
(7, 11) to
suggest that different TRs might have distinct functions in
development. Similar patterns of TR expression, with TR appearing
well before TR are found during development of the rat (10) and
Xenopus(29, 30) . Taken together, these
observations provide the basis for the hypothesis that T has a role to play in the early stages of embryogenesis, and that
TR is implicated in T-dependent responses occurring at
this time.
does
affect neurogenesis. T-treated embryos showed increased
proliferation of neuroblasts as assessed by
[H]thymidine incorporation (Fig. 1,
a-c), and this corroborates an early report
(31) showing that treatment of premetamorphic Xenopus tadpoles with T increased the number of mitotic
figures in the spinal cord.
and
TR. Even though TR mRNA has not been detected in the chick
brain before embryonic day 16
(7) , we thought it relevant to
test both TR and TR, since mRNA levels are not necessarily
closely correlated with protein levels. Moreover, this seems to be
particularly true for TR in neuronal context. Strait and
co-workers
(32) found that the molar concentrations of TR
protein levels exceeded those of the corresponding mRNA 350-fold in the
rat cerebellum. Thus, it remains possible that TR pro-tein may be
present even when the mRNA is undetectable by hybridization.
stimulated proliferation of
neuroblasts in these conditions (Fig. 2, a and
b). We chose this approach, rather than an in vivo methodology, as the number of cells to be targeted in vivo is very large and it is currently technically unfeasible to
deliver ODNs to the whole population of the neuro-epithelium germinal
layer. Delivery of ODNs to a small part of the neuro-epithelium would
result in ambiguous results and unreliable interpretations. However,
the in vitro approach employing pure cultures of neuroblasts
has advantages: it allows a large percentage of embryonic neurons to be
transfected (up to 70%, 21) and allows one to determine if the effects
are direct or indirect. The data presented here suggest that the effect
of T is exerted directly on the neuronal precursor cells
and that T does not require other cell types to affect
neuronal proliferation. Two arguments bolster this hypothesis: first,
it occurred in the primary cultures of optic lobe cells that develop
into virtually 100% neuronal cultures as shown by neurofilament
immunochemistry (Fig. 4b), and second, most of the cells
in these cultures express TRs (Fig. 7a).
and TR in these
embryonic neuronal precursor cells. We used a cationic lipid to
vectorize the ODNs, as this method resulted in high uptake and
stability of the ODNs in the intracellular compartment for up to 12 h
post-transfection ( Fig. 3and Fig. 4). This enabled us to
use very low ODN concentrations (<150 nM) on the embryonic
neuroblasts, thus reducing toxicity and nonspecific effects. Antisense
ODNs (150 nM) directed against the chick TR significantly
inhibited neuroblast proliferation by up to 45%
(Fig. 5a), whereas AS and random ODNs based on the
TR and the TR sequences used had no significant effect.
Moreover, the inhibition of [H]thymidine
incorporation seen with AS was dose dependent, with a maximal
effect occurring at 100 nM (Fig. 5b). This
demonstration provides a physiological role for TR in early
neurogenesis that correlates with expression of the mRNA for this
receptor in the brain of the 6-8-day-old chick embryo
(7) .
chosen was
functional in inhibiting the production of TR protein. To address
this problem we used a cotransfection approach where we cotransfected a
TRE-reporter construct with the different ODNs to be tested, AS,
AS, and their respective randomized-sequence controls. Application
of AS and AS reduced T-dependent transcription
(Fig. 6) showing both AS constructions to have a biological
effect. However, in the case of AS, T-dependent
transcription was completely abolished, whereas with AS a residual
stimulation of transcription in the presence of T was seen,
possibly due to the action of TR, most probably the more abundant
receptor. Thus, even though blocking expression of TR did not
affect neuroblast proliferation, it did diminish T3-dependent
transcription from a consensus TRE.
are present they are not required for the
T-dependent stimulation of the mitotic response observed
in vitro and that the response is mediated through TR.
Different TRs have been shown to play distinct roles in regulating
target gene transcription in defined cellular
contexts
(21, 33) . Indeed, structural differences
between the and the forms provide a biochemical basis for
differential transcriptional activities. Such structural differences in
the chick TR and TR include variations in their N termini
sequences
(7) that engender various possibilities for
phosphorylation by Ser/Thr kinases of this part of TR
(34) but not TR. Other differences lie in the DNA-binding
domain in a region of the second zinc finger that could be involved
both in determining target gene specificity
(35) and in dimer
formation on TREs
(36) .
does have a role to play in stimulating
proliferation of neurogenic precursors during neurogenesis and that
TR, which is expressed during this period, is implicated in
mediating this response. These observations broaden the perspective in
which TRs have previously been thought to function so as to include the
early formation of brain regions, as well as the later maturation
phases, well known to be T-dependent
(37) . They also
demonstrate that TR function in early embryogenesis is independent of
the embryonic thyroid, as neuronal proliferation occurs before the
onset of endogenous thyroid function (at about embryonic day 11 in the
chick, 38). This implies that the T present in the yolk and
embryo is sufficient to activate the receptors. Significant amounts of
ligand are available at this stage (39 and this report). It is probable
that in other species where TR is expressed before endogenous
thyroid activity, maternal supplies of hormone are sufficient for
activating TR-dependent transcription during early development.
and
genes and what will be the consequences on development if these
genes are blocked in the whole animal? An important area for future
work will be the manipulation of these genes, first through mutations
introduced by homologous recombination and the production of transgenic
animals, and second, through modulation of expression during
development in species (such as amphibians) where T has
dramatic effects and in which it is now possible to introduce exogenous
genes
(40) . More specifically concerning the role of TR, it
will be interesting to see if this proto-oncogene is also implicated in
other phenomena occurring during neurogenesis. Since certain
proto-oncogenes, particularly c-myc(41) , affect both
proliferation and apoptosis, it will pertinent to examine if TR is
implicated in both these intimately related processes during
neurogenesis.
provides a specific contribution to transducing the mitotic promoting
effects of T during neurogenesis. They add credence to the
hypothesis that multiplicity of nuclear receptors allows for specific
actions of T in defined cellular contexts. Differential
control of TR expression during development could provide a mechanism
through which one extracellular signal can be linked to a variety of
specific cellular responses.
,
3,5,3`-triiodo-L-thyronine; TRE, thyroid hormone response
element; ODNs, oligodeoxynucleotides; DOGS,
dioctadecylamidoglycylspermine; SFM, serum-free medium; PBS,
phosphate-buffered saline; FCS, fetal calf serum.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
ABSTRACT
This article has been cited by other articles:
|
|
 |
|
  A. Nicot, V. Lelievre, J. Tam, J. A. Waschek, and E. DiCicco-Bloom Pituitary Adenylate Cyclase-Activating Polypeptide and Sonic Hedgehog Interact to Control Cerebellar Granule Precursor Cell Proliferation J. Neurosci., November 1, 2002; 22(21): 9244 - 9254. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Guissouma, S. M. Dupre, N. Becker, E. Jeannin, I. Seugnet, B. Desvergne, and B. A. Demeneix Feedback on Hypothalamic TRH Transcription Is Dependent on Thyroid Hormone Receptor N Terminus Mol. Endocrinol., July 1, 2002; 16(7): 1652 - 1666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Forrest Editorial: Twists in the Tail--Change-of-Function Mutations in Thyroid Hormone Receptors Endocrinology, July 1, 2002; 143(7): 2466 - 2468. [Full Text] [PDF]
|
|
|
|
 |
|
 L. Calza, M. Fernandez, A. Giuliani, L. Aloe, and L. Giardino Thyroid hormone activates oligodendrocyte precursors and increases a myelin-forming protein and NGF content in the spinal cord during experimental allergic encephalomyelitis PNAS, February 20, 2002; (2002) 52704499. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. Trentin, M. Alvarez-Silva, and V. Moura Neto Thyroid hormone induces cerebellar astrocytes and C6 glioma cells to secrete mitogenic growth factors Am J Physiol Endocrinol Metab, November 1, 2001; 281(5): E1088 - E1094. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ben-Hur, B. Rogister, K. Murray, G. Rougon, and M. Dubois-Dalcq Growth and Fate of PSA-NCAM+ Precursors of the Postnatal Brain J. Neurosci., August 1, 1998; 18(15): 5777 - 5788. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Calza, M. Fernandez, A. Giuliani, L. Aloe, and L. Giardino Thyroid hormone activates oligodendrocyte precursors and increases a myelin-forming protein and NGF content in the spinal cord during experimental allergic encephalomyelitis PNAS, March 5, 2002; 99(5): 3258 - 3263. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |