| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
U-159, INSERM Centre Paul Broca (C.L., M.T.B.-P., P.Z., S.C., P.D., E.P., J.E., R.G.), 75014, Paris, France; the Departments of Pediatrics and Neurology and Neurosurgery, McGill University (G.S.T.), Montréal, Québec, Canada H3A 2B4; the Department of Medical Anatomy, Panum Institute, University of Copenhagen (L.H.), DK2200 Copenhagen, Denmark; and Novartis Pharma (D.H.), CH 4002 Basel, Switzerland
Address all correspondence and requests for reprints to: Dr. Jacques Epelbaum, U-159, INSERM, Centre Paul Broca, 2ter rue d’Alésia, 75014 Paris, France. E-mail: epelbaum{at}broca.inserm.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Five different SRIH receptor subtypes (sst receptors), designated sst1 to sst5, have been cloned and characterized (7). All five isoforms are widely expressed in mammalian brain, although the functional role of the individual SRIH receptor subtypes is largely unknown. In the hypothalamus, in particular, the messenger RNAs (mRNAs) encoding for sst1 and sst2 receptor subtypes are highly concentrated (8, 9). Moreover, sst1 mRNA is expressed in hypophysiotropic SRIH neurons of the periventricular nucleus (10), whereas both subtypes are synthesized in GHRH-expressing arcuate nucleus neurons (11). Such a localization suggests that one or both of the sst1 and sst2 receptor subtypes may be involved in the control of GH secretion by SRIH.
To test this hypothesis, in the absence of sufficiently selective sst1 or sst2 receptor antagonists, we investigated the consequences of blocking sst1 or sst2 subtypes by using an oligodeoxynucleotide antisense approach. In a first series of in vitro experiments, we validated the efficacy of the anti-sst1 (AS-sst1) and anti-sst2 (AS-sst2) oligodeoxynucleotides on the previously characterized opposite effects of sst1 or sst2 activation on glutamate sensitivity of hypothalamic neurons in mouse primary cultures (12, 13). In a second series of in vivo experiments, we assessed the effects of icv infused antisense probes on the ultradian pulsatility of GH secretion in awake, freely moving rats.
In vitro results show that both antisense probes are efficient in selectively blocking SRIH-induced modulations of glutamate sensitivity. In contrast, changes in pulsatile GH secretion were only observed after icv infusion of the AS-sst1 probe. These observations favor a preferential role of the sst1 receptor subtype in the regulation of GH secretion by SRIH at the hypothalamic level.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In vitro experiments
Cell cultures. Hypothalamic neurons were grown as previously
described (14). Fetal hypothalami were dissected out from albino Swiss
mice (Janvier, Le Geneste Ste Isle, France) on day 16 of gestation,
pooled, and mechanically dissociated. The mouse model was chosen based
on previous experiments on the modulation of glutamate sensitivity by
sst1 and sst2 receptor agonists (13). Cells were plated in 35-mm petri
dishes, precoated with 5% FCS, at a density of 2.5 hypothalami/dish
(1.5 x 106 cells/dish), maintained in a 37
C, 7% CO2 humidified atmosphere and were treated
from 5 days in vitro (DIV) on with 1
µM cytosine arabinoside (Sigma,
St. Quentin-Fallavier, France). In such culture conditions, the glial
cell proportion is less than 15% of the total cell population (14).
ODNs were first added to the culture medium on 12 or 13 DIV, for 4
consecutive days, and, after medium replacement, on 16 or 17 DIV for 2
more days. ODN cell penetration was checked using fluorescein-labeled
ODNs (Eurogentec) incubated in the same conditions. Cells were recorded
from 18 DIV on when excitatory and inhibitory in vitro
connections are fully developed (15) and glutamate responsiveness to
SRIH stabilized (12).
Physiological recordings. Electrophysiological recordings
from hypothalamic neurons in vitro employed the whole cell,
tight seal recording configuration of the patch-clamp technique (16).
Patch-clamp electrodes (4–6 M
) were obtained by a two-stage pull on
a horizontal electrode puller (BB-CH Mecanex, Geneva, Switzerland) and
were filled with 120 mM KF, 3
mM KCl, 1 mM
CaCl2, 2 mM
MgCl2, 11 mM EGTA, 10
mM HEPES, and 2 mM
ATPNa2, at pH 7.35. The external patch recording
medium contained 140 mM NaCl, 5
mM KCl, 2 mM
CaCl2, 2 mM
MgCl2, 10 mM HEPES, 10
mM glucose, 0.5 µM
tetrodotoxin X (Latoxan, Rosans, France) at pH 7.35.
Glutamate-induced currents were recorded at room temperature (20–25 C)
with an Axopatch-1D patch-clamp amplifier (Axon, Foster City, CA).
Neurons were voltage-clamped to -80 mV, in the presence of 2
mM Mg2+. Experiments were
run on-line using an Axon TL-1 DMA interface and a 386/20 Tandon
computer. Delivery of command voltages and storage of current data into
data files were driven by the Axon Clampex software.
L-Glutamate (Sigma) was dissolved at
5 mM in external medium and applied by pressure
ejection at 8 psi for 20 msec, using a pneumatic picopomp (WPI,
Sarasota, Fl), from a 3- to 5-µm tip diameter pipette positioned
30–50 µm from the cell soma. Glutamate was delivered with a minimum
30-sec interval between two stimulations. Under these conditions,
receptor desensitization does not occur (12). Octreotide (0.1
µM; SMS 201–995, gift from Novartis,
Rueil-Malmaison, France) and CH-275 (0.1 µM;
gift from Carl Hoeger and Jean Rivier, San Diego, CA) were dissolved in
the external medium and perifused at 2 ml/min using a peristaltic pump
(Ismatec Bioblock, Illkirch, France).
Analyses were performed off-line with the Axon pClamp software package.
Individual values for glutamate sensitivity were estimated from the
amplitude of glutamate-induced peak current under control conditions or
in the presence of drugs. The difference between control and drugs is
referred to as glutamate sensitivity variation. All data are expressed
as the mean ± SEM, and statistical significance was
evaluated by Student’s t test (for pooled values), ANOVA
(for paired values during control vs. drug applications), or
2 test (for comparison between cell
populations).
RT-PCR. Quantification of mRNA levels was based on the use of synthetic standard RNA (cRNA) containing the sequences required for the amplification of each sst mRNA (17). Amplification was performed in an automatic thermocycler (Hybaid, Teddington, UK). Briefly, 1 µg of the total RNA of the cell cultures on 18 DIV was mixed with various dilutions of cRNA, and the samples were denatured and reverse transcribed for 90 min at 37 C. For each experiment, two types of controls were performed: 1) each sample was treated for RT reaction without enzyme to verify the absence of contaminating DNA in RNA samples; and 2) a control tube containing the RT mixture but without RNA was used to rule out contamination with foreign DNAs or RNAs. The detection limit of the method ranges from 100-1000 molecules of the amplified product (17).
The conditions of the amplification reaction were identical for the five receptor subtypes. One tenth of the RT reaction was amplified in presence of 25 pmol sense and reverse primers, 0.5 x 106 cpm of 5'-end 32P-labeled reverse primer, and 1.5 U Taq polymerase (Promega Corp., Charbonieres, France). The amplification was performed as follows: a hot start (94 C, 30 sec), followed by 29 cycles (94 C, 30 sec; 60 C, 60 sec; and 72 C, 30 sec), and an extension step at 72 C for 10 min. A control for external contamination was performed in each experiment (i.e. PCR mixture with no cDNA template). A 15-µl portion of the PCR reaction was electrophoresed on an 8% acrylamide gel. The gel was autoradiographed, and bands corresponding to the amplified products were cut and counted in a ß-scintillation counter (Pharmacia Biotech, Les Ulis, France). The linear regression of the curve was calculated, and the results were expressed as molecules per µg total RNA.
Oligonucleotides were commercially synthesized (Genset, Paris, France). Sense primer sequences were 5'-ATGGTGGCCCTCAAGGCCGG-3' (position 716, accession no. M81831), 5'-TCCTCTGGAATCCGAGTGGG-3' (position 711, accession no. M81832), 5'-TGTCAGTGGGTACAGCCACC-3' (position 738, accession no. M91000), 5'-TGCGGGCTGGCTGGCAACAA-3' (position 801, accession no. U26176), and 5'-CTGGGGCACCTGCAACCTGA-3' (position 728, accession no. L04535 and S53287) for sst1, sst2, sst3, sst4, and sst5, respectively. The common reverse primer sequence was 5'-CCGGTTGGCACAGCTGTTG-3' (seventh transmembrane domain, COM 1).
Single cell RT-PCR was performed as previously described (13). After recording, cytoplasm harvesting of patched cells was obtained by gentle suction applied to the pipette. The tip of the pipette was then broken in a PCR tube for RT. The resulting 15-µl sample was incubated for 1 h at 37 C to complete synthesis of single strand cDNAs. RT products were then kept at -80 C. For sst receptor multiplex PCR, two serial PCRs were performed. The first PCR reaction amplified simultaneously all five SRIH receptor subtypes, SRIH, and GHRH mRNAs. For sst receptors, common primers were used: a forward primer sequence COM 2 (5'-CGCTATCTGCCTGTGGTGCACCC-3') and the common reverse primer COM 1 (17). For GHRH mRNA amplification, the following primers were used: 5'-CACCTCGGAGGTCCCACCC-3' and 5'-CATGCTGTCTTCCTGG CGGC-3'. For SRIH mRNA amplification, the following primers were used: 5'-ATCGTC CTGGCTTTGGGC-3' and 5'-GCCTCATCTCGTCCTGCTCA-3'. cDNAs were then amplified in a 50-µl volume with 0.25 mM deoxy-NTPs, 250 pmol COM1 and COM2 [hot start, followed by five cycles (94 C, 30 sec; 45 C, 90 sec; and 72 C, 30 sec) and 27 cycles (94 C, 30 sec; 62 C, 60 sec; and 72 C, 30 sec)]. Selective amplification for each sst subtype was then performed during the second PCR, using the primers described above. The first 50 µl of PCR products were purified from contaminating primers using Nucleon microspin (Amersham Pharmacia Biotech, Rainham, UK). For each receptor subtype, 4 µl of the purified PCR products were amplified in a 50-µl volume with 25 pmol of each specific sense primer and 25 pmol COM2, 0.5 x 106 cpm 5'-end 32P-labeled reverse primer. Amplification began by a hot start, followed by 25 cycles (94 C, 30 sec; 60 C, 60 sec; and 72 C, 30 sec). A 10-µl sample of the PCR reaction was then electrophoresed on an 8% polyacrylamide gel (Bio-Rad Laboratories, Inc., Hercules, CA). After migration, the gel was dried and exposed overnight to X-OMAT film (Eastman Kodak Co., Marnes la Vallee, France). Bands corresponding to sst receptor subtypes and peptide-amplified products were characterized on the basis of their mol wt (205 and 217 bp, respectively, for sst1 and sst2, and 207 and 189 bp for SRIH and GHRH) and sequenced (Eurogentec, Seraing, Belgium). Sequencing revealed these amplified products to be identical to cloned mouse sst1 (accession no. M81831), sst2 (accession no. M81832), SRIH (accession no. X51468), and GHRH (accession no. M31658) sequences.
In vivo experiments
Animals. Four weeks before sampling experiments, young adult
male Wistar rats (Charles River Laboratories, Inc.,
Elbeuf, France), weighing 100–150 g, were housed individually in a
room with controlled temperature (22–24 C) and illumination (12-h
light, 12-h dark schedule, with lights on at 0100 h). They had
free access to food and water and were regularly handled and weighed to
minimize stress effects. Food intake was monitored daily.
Surgery and experimental procedures. Nine days before the experiment, a chronic icv cannula (Alzet Brain Infusion Kit, Alza Corp., Palo Alto, CA) was inserted into the lateral ventricle of the brain under pentobarbital anesthesia (35 mg/kg BW, ip), according to the procedure described by Pellegrini et al. (18). The following coordinates (19) were used: anterior, -0.8 mm to the bregma; lateral, -1.5 mm to the midline; height, -3.8 mm to the bregma, with the incisor bar set at -3.3 mm below the interaural line. The cannula was then secured to the skull of the animal with stainless steel screws and dental cement and connected to a polyethylene tube catheter and a miniosmotic pump (Alzet 2002, Alza Corp.) introduced under the dorsal skin of the animal. ODNs or saline were administered at a delivery rate of 0.5 µl/h after a 4-h priming at 37 C in sterile saline. The flow moderator of the pump was connected with the polyethylene catheter containing 48 µl saline, which was administrated during the first 4 days in all animals. A small air bubble was inserted to separate the saline solution from an additional 60 µl saline or ODNs to be delivered for an additional 120 h (108 h before and during the entire GH-sampling period).
Experiments were performed in freely moving rats. Two days before the experiments, an indwelling cannula was inserted into the right atrium under ether anesthesia as previously described (20). On the day of the experiments, 2 h before the sampling period, the distal extremity of the cannula was connected to a polyethylene catheter filled with 25 IU/ml heparinized saline. Blood samples (0.25 ml) were withdrawn every 15 min from 0900–1800 h and centrifuged immediately. Red blood cells were resuspended in saline and reinjected at the next sampling to attenuate hemodynamic modifications. Plasma was stored at -20 C until GH assay.
At the end of the sampling period, animals were either decapitated for autoradiography and in situ hybridization or deeply anesthetized with sodium pentobarbital (Sanofi Pharmaceuticals, Inc., Toulouse, France; 80 mg/kg, ip) and perfused with 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, for immunohistochemistry.
Radioautography of SRIH-binding sites. Brains were removed, frozen in isopentane (-40 C for 30 sec), and kept at -70 C. Serial 16-µm cryostat sections at the level of the periventricular (PeV; from interaural 7.4 to 6.8) and the arcuate (ARC; from interaural 6.2 to 5.4) nuclei, according to the atlas of Paxinos and Watson (19), were mounted on poly-L-lysine-coated slides and stored at -70 C until use. Microscopic examination of toluidine blue-stained sections allowed for appropriate verification of icv cannula placement and anatomical match across animals.
Monoiodo-Tyr0-D-Trp8-SRIH ([125I]SRIH; SA, 780 Ci/mmol) was prepared as previously described (21). Sections were preincubated at room temperature for 30 min in 0.05 M Tris-HCl buffer (pH 7.4) containing 0.32 M sucrose, 0.5% BSA, and 5 10-5 M bacitracin. A 60-min incubation was then carried out in the same medium containing 5 mM MgCl2 in the presence of [125I]SRIH (0.4 nM). Under these conditions, [125I]SRIH binding accounts for the sst2/sst3/sst5 receptor subtypes [IC50 in CHO (sst1–4) and COS (sst5) cells: hsst2, 0.2 nM; hsst3, 0.5 nM; hsst5, 0.5 nM; vs. hsst1, 2.0 nM; hsst4, 3.2 nM; Hoyer, D., unpublished data). Adjacent sections were incubated in the presence of 1 µM SRIH-14 to determine nonspecific binding, which amounted to less than 5% of the total binding. Specific binding was calculated as the difference between total and nonspecific binding (four sections per rat in both groups).
Sections were placed under x-ray film (Amersham Pharmacia Biotech) for 4 days, and films were developed in Dektol (Eastman Kodak Co.). Binding was quantitated by reference to iodinated standards (prepared from brain paste in the presence of different concentrations of the tracer) with the help of a computer-assisted image analyzer using a video camera and the RAG program (Biocom, Les Ulis, France), which allows the conversion of optical densities into radioactivity units.
Immunohistochemistry. Serial 30-µm sections from the rostral PeV (bregma, from -1.0 to -1.4 mm) and the median eminence (bregma, from -2.5 to -2.9 mm) were processed for sst1 immunohistochemistry using fluorophore-labeled tyramide as previously described and validated (10, 22). Briefly, sections were rinsed in 0.1 M Tris buffer saline, pH 7.4 (TBS) containing 0.05% Tween-20 and preincubated for 30 min in 10% normal goat serum (NGS) in TBS. Sections were then incubated overnight at room temperature with AS-sst1 (10) diluted 1:10,000 in TBS containing 0.5% NGS and 0.3% Tween-20. Sections were then rinsed in TBS containing 0.05% Tween-20 and incubated for 45 min in biotinylated goat antirabbit IgG (Vector Laboratories, Inc., Burlingame, CA) diluted 1:300 in TBS containing 3% NGS. They were then incubated for 45 min in avidin-biotinylated horseradish peroxidase complex (Vector Laboratories, Inc., Burlingame, CA). Sections were finally incubated for 10 min in a 1:100 fluorescein tyramide solution (TSA-Direct, NEN Life Science Products, Boston, MA), rinsed in TBS, mounted, and coverslipped.
Sections were analyzed by confocal laser scanning microscopy using a TCS-4D confocal imaging system equipped with an argon-krypton ion laser (Leica Corp., Heidelberg, Germany).
In situ hybridization. In situ hybridization was
carried out as described previously (23). The original plasmids were
provided by Dr. Graeme Bell, Howard Hugues Medical Institute,
University of Chicago (Chicago, IL). Single stranded antisense RNA
probes were generated from constructed mouse sst1 and sst2
complementary DNA expression vectors, pGEM-3Z (Promega Corp., Madison, WI). The mouse sst1 probe is a 413-bp
BanII-AccI fragment of the gene that encodes
amino acids 214–352, and the mouse sst2 probe is a 458-bp
BstEII-XbaI fragment of the cDNA that encodes
amino acids 254–369, the stop codon, and 107 nucleotides of the
3'-flanking/untranslated region. To obtain sst1 and sst2 antisense
probes, cDNA templates were produced by linearization of the vector
with EcoRI and transcription with the Gemini II system
(Promega Corp.) using SP6 RNA polymerase and
5'-[
-35S]UTP (NEN Life Science Products). The final probe specific activity was approximately
2.2 x 109 dpm/mg.
Sections were postfixed for 10 min at room temperature in a 4% paraformaldehyde solution and rinsed in 1 M phosphate buffer (three times, 10 min each time), then in H2O (1 min). They were incubated in 0.25% anhydrous acetic acid in 0.1 M triethanolamine for 10 min and in 0.1 M Tris-glycine for 30 min and washed three times in a double concentration of SSC (standard saline citrate). After dehydratation in ascending concentrations of ethanol, they were prehybridized by immersion for 2 h at room temperature in 4 x SSC containing 1 x Denhardt’s and 10 mM mercaptoethanol. Hybridization was run overnight at 60 C in the hybridization solution (1% sarcosyl, 50% deionized formamide, 10% dextran sulfate, 10 mM dithiothreitol, 4 x SSC, 1 x Denhardt’s, and 10 mM mercaptoethanol) containing the 35S-labeled antisense probe (2 x 106 cpm/ml). The following day, sections were immersed in 4 x SSC at room temperature for 60 min, washed three times in 4 x SSC, incubated in a solution of 4 x SSC containing 20 µl/ml ribonuclease A at 37 C for 30 min, rinsed in decreasing concentrations of SSC, incubated in 0.1 x SSC containing 0.25% dithiothreitol at 60 C for 30 min, dehydrated, dried, and coated by dipping in RPN40 LM1 emulsion (Amersham Pharmacia Biotech). The exposure time was 15 weeks. Autoradiograms were developed in Dektol (Eastman Kodak Co.), stained with toluidine blue, and coverslipped.
Sections were visualized at x500 magnification under fluorescent epiillumination (BH2 microscope, Olympus Corp., Tokyo, Japan). Grain counting was performed with a Biocom 200 image analyzer (Les Ulis, France) using the computer-based image analysis system (RAG 200), which allows for rapid estimation of grain numbers over neuronal perikarya. An internal calibration curve was recorded for each section, measuring the mean quantity of light reflected by a known number of grains according to the procedure described by Bisconte et al. (24). Labeled neurons were identified by toluidine blue under brightfield illumination and delineated on the screen, and the quantity of light reflected in the area was measured under epiillumination.
Four sections corresponding to the levels of the PeV and the ARC were analyzed in each rat. Data were expressed as frequency distribution of grains per cell. Cells were considered labeled if the grain density was at least 3 times the background level.
GH enzymoimmunoassay (EIA). Plasma and pituitary GH concentrations were measured in duplicate by EIA as previously described (25). GH values were reported in terms of rat GH RP-2. The sensitivity of the EIA is 1 ng/ml. The intra- and interassay coefficients of variation were below 7%.
Statistical analysis. GH pulse analysis was performed using the Cluster program (26), setting the t value to 4.1 to maintain false positive rates under 1%. Cluster size was set to one prepulse and one postpulse nadir value. False positive error for pulse detection was 7%. Values are given as the mean ± SEM, and statistical analysis was performed by ANOVA and post-hoc comparisons using StatView 4.5 software (Abacus Concepts, Palo Alto, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, the
proportions of GHRH neurons coexpressing sst1 and sst2 mRNAs were
identical, and only 1 of 19 GHRH neurons was devoid of sst mRNAs. In
contrast, sst1 mRNAs were more often present than sst2 in
SRIH-expressing neurons, of which a significant number (7 of 30) did
not contain any sst mRNAs. Finally, 5 of 12 cells that did not contain
either peptide also did not express sst receptor mRNAs. In this group,
the remaining 7 cells expressed slightly more sst1 (5 of 7) than sst2
(3 of 7) mRNAs.
|
Thus, optimal labeled ODN internalization was obtained between 24 and 72 h of incubation. In view of these results, the protocol of ODN incubation was set so as to obtain the highest efficiency of penetration before electrophysiological recordings, i.e. a first incubation for 96 h from 12–13 to 16–17 DIV and a second incubation for 48 h from 16–17 to 18–19 DIV.
Quantitative RT-PCR. Quantitative expression of the five sst
receptor subtypes and efficiency of the antisense treatments were
analyzed by quantitative RT-PCR on two independent cultures on 18 DIV
(Fig. 1). In control experiments, sst1
and sst2 mRNA levels were 8–10 times higher than those of sst3 and
sst4 mRNAs and 30 times higher than sst5 mRNA, respectively. Both
AS-sst1 and AS-sst2 treatments selectively and significantly reduced by
76% and 64%, respectively, the expression of sst1 and sst2 mRNAs
without significantly modifying sst3 and sst4 mRNA levels. The sst5
mRNA levels were increased after AS-sst2 treatment. RS-sst1 and RS-sst2
treatments were ineffective, with the exception of a slight, but
significant, increase in sst2 mRNA expression under RS-sst2
treatment.
|
ODN incubation did not induce cytotoxicity. Resting potentials were not altered in the presence of ODNs [-46.2 ± 3.4 mV (n = 32) vs. -48.9 ± 1.19 mV (n = 45); from Ref. 13 ]. Modifications in input resistances or spontaneous activities were not detected after ODN treatments (data not shown).
Mean glutamate-induced peak current, estimated from the whole cell population, reached 224 ± 14 pA (n = 122). The mean peak current value within each series of treatments was not different from that in either the whole cell population or the nontreated group, indicating that the different ODN treatments did not modify glutamate sensitivity [no treatment, 231 ± 30 pA (n = 40); AS-sst1, 226 ± 23 pA (n = 26); AS-sst2, 210 ± 27 pA (n = 28); RS-sst1, 273 ± 50 pA (n = 12); RS-sst2, 190 ± 32 pA (n = 16)].
Recorded neurons processed for single cell RT-PCR were distributed as follows: AS-sst1 (25 cells), AS-sst2 (23 cells), RS-sst1 (10 cells), and RS-sst2 (14 cells). The presence of sst1 and/or sst2 mRNAs was correlated with the effects of the sst1 agonist CH-275 and/or the sst2 agonist octreotide on glutamate sensitivity in 94% of the studied neurons (68 of 72); no signal was detected for the 4 remaining cells despite variations in glutamate sensitivity in the presence of SRIH agonists.
CH-275 increased glutamate sensitivity in 7 of 19 tested cells (37%),
whereas octreotide decreased it in 9 of 24 tested neurons (38%).
Glutamate-induced currents were increased by 41.9 ± 15.8% under
CH-275 (n = 7 paired values; P < 0.05) and
decreased by 22.2 ± 3.4% under octreotide (n = 9 paired
values; P < 0.01; Table 2).
|
1) or RS-sst2 (data not shown), CH-275-
or octreotide-induced modulations of glutamate sensitivity still
persisted and were not significantly modified compared with the
nontreated series for either the mean amplitude variations of the
glutamate response in the presence of either agonists or the
percentages of responsive cells in both cases.
|
1), the variation in the glutamate response under
CH-275 was suppressed (mean variation, -1.6 ± 1.3%; n =
14). The percentage of octreotide-sensitive neurons (50%, 8 of 16) and
the octreotide-induced decrease (-31.0 ± 5.4%; n = 8) in
glutamate sensitivity were not significantly increased (32% and 40%
increases, respectively). AS-sst1 treatment totally abolished sst1 mRNA
detection at the single cell level (0 of 25 cells), but sst2 mRNAs were
present in 32% of the studied neurons (8 of 25), a percentage not
statistically different from those obtained in the RS-sst1 (40%) or
RS-sst2 (29%) groups.
When cultures were treated with AS-sst2 (Fig. 2C1), octreotide-induced
variation in the glutamate response was totally suppressed (mean
variation, 0.7 ± 0.9%; n = 16). The percentage of
CH-275-sensitive neurons (53%, 10 of 19) and the CH-275-induced mean
increase (+55.6 ± 26.6%; n = 10) in glutamate sensitivity
were not significantly increased (43% and 33% increases,
respectively). AS-sst2 treatment totally abolished the detection of
sst2 mRNA at the single cell level (0 of 24 cells), but sst1 mRNAs were
present in 43% of studied neurons (10 of 23), an expression rate
similar to that previously published (44%) (13) and to that found with
RS-sst1 (40%) or RS-sst2 (36%) treatments (Table 2).
Effects of AS-sst1 and AS-sst2 icv infusion on GH secretory
profiles in male rats
Exp I. In the first series of experiments, male rats received
an icv infusion of saline, AS-sst1, or AS-sst2 at the rate of 2
nmol/0.5 µl·h for 120 h. Although surgery caused a small and
temporary weight loss, antisense treatments did not significantly alter
body weight gain and daily food intake compared with those in
saline-infused controls. All animals had recovered their initial
weights at the time of the blood sampling (data not shown). There were
no observable behavioral effects of any of the treatments.
Administration of AS-sst1, but not AS-sst2, resulted in an overall
decrease in pulsatile GH secretion compared with saline treatment (Fig. 3). In AS-sst1-treated animals, mean
plasma GH levels during 9 h, and GH pulse amplitudes were
significantly reduced [mean 9 h GH: saline (n = 7), 46
± 3 ng/ml; AS-sst2 (n = 5), 39 ± 4 ng/ml; AS-sst1 (n =
5), 27 ± 2 ng/ml; P < 0.001 vs.
saline and P < 0.05 vs. AS-sst2; pulse
amplitude: saline, 125 ± 10 ng/ml; AS-sst2, 110 ± 18 ng/ml;
AS-sst1, 69 ± 6 ng/ml; P < 0.01 vs.
saline and P < 0.05 vs. AS-sst2]. In the
same animals, GH nadir and pulse frequency were not significantly
modified (nadir: saline, 23 ± 3 ng/ml; AS-sst2, 22 ± 4;
AS-sst1, 19 ± 2; P > 0.05; number of pulses:
saline, 4.7 ± 0.4; AS-sst2, 4.0 ± 0.6; AS-sst1, 3.6 ±
0.5; P > 0.05).
|
). After
AS-sst1 treatment, the levels of binding sites remained unchanged in
all regions.
|
).
Pituitary GH content was not modified by AS-sst1 infusion [saline
(n = 5), 162 ± 14 ng/µg protein; scrambled ODNs (n =
5), 153 ± 25 ng/µg proteins; AS-sst1 (n = 6), 170 ±
9 ng/µg proteins].
|
, sst1
immunoreactivity corresponded to its previously described distribution
(10). In control rats, strongly labeled neurons and nerve fibers were
found along the border of the third ventricle, mostly at the ventral
rim of the third ventricle (Fig. 5A). Strongly immunoreactive nerve
fibers, endowed with large swellings, were apparent in the external
layer of the median eminence, whereas the internal layer was devoid of
immunostaining (Fig. 5C). In AS-sst1 treated rats, only sparsely
labeled perikarya were observed in the PeV, and only sparse and weakly
labeled axon-like terminals and nerve fibers were visualized in both
the PeV and the external layer of the median eminence (Fig. 5, B and
D).
|
). In contrast, the number of cells
expressing sst2 mRNA was increased by 100% (94 ± 9
vs. 47 ± 16 in saline controls; P <
0.01), with a slight shift of the frequency distribution of grain
counts to higher values (Fig. 6B). In ARC from AS-sst1-treated animals,
sst1 mRNA expression was not modified, but sst2 mRNA levels were
increased (from 32 ± 2 to 53 ± 5 cells/section;
P < 0.05). Scrambled ODN infusion did not affect the
number of cells expressing sst1 or sst2 mRNAs or the frequency
distribution of grain counts in both nuclei compared with the effect of
saline infusion.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Distribution of sst mRNAs in GHRH and SRIH hypothalamic neurons
We first determined the distribution of SRIH receptor mRNAs
and GHRH- and SRIH-expressing cells in mouse hypothalamic cell
cultures, as previous data were mostly obtained in the rat (9, 10, 11, 23, 28). We confirmed in vitro that sst1 and sst2 represented
the major SRIH receptors as observed by in situ
hybridization in mouse hypothalamus (8, 29). Moreover, we
found that most GHRH neurons expressed sst1 and sst2 mRNAs in an
equivalent proportion, whereas SRIH neurons expressed either no sst or
slightly more sst1 than sst2. The percentage of neurons expressing
either SRIH or GHRH mRNAs is high in the in vitro mouse
hypothalamic model. This might be due to a neuronal selection along the
time of the culture in defined medium. For instance, no GnRH-containing
neurons were ever observed in such cultures, whereas TRH neurons were
numerous (Faivre-Bauman, A., and C. Loudes, personal communication).
Also, one might postulate that colocalizations with other peptides and
neurotransmitters can occur and will have to be defined further.
Assessment of AS-sst1 and AS-sst2 selectivity
In vitro, as previously reported (30), ODNs migrated
first to the cytoplasm and reached the nucleus after 12 h in all
neurons and some underlying glial cells, the latter displaying less
intense staining. The efficacy of AS-sst1 and AS-sst2 treatments was
checked by quantitative RT-PCR. Selective antisense treatment
significantly decreased sst1 and sst2 mRNA levels by 76% and 64%,
respectively; in keeping with similar results reported on rat pituitary
dopamine D2 (31) or superior cervical ganglion M1 muscarinic
receptor (32) mRNAs. By single cell RT-PCR, the percentage of
neurons expressing sst1 or sst2 mRNAs in RS-ssts ODN-treated cultures
was not different from that under control conditions (13), nor was the
number of cells expressing sst1 mRNA after AS-sst2 treatment or sst2
mRNA after AS-sst1 treatment. Thus, control applications did not modify
the expression of sst mRNAs at the single neuron level. In marked
contrast, sst1 or sst2 mRNAs were not detected after AS-sst1 or AS-sst2
treatment, respectively. This null expression may appear at variance
with the quantitative RT-PCR data presented above, but is likely to
reflect a differing sensitivity of the two techniques. Indeed, the very
low level of mRNA present in one neuron does not allow for accurate
quantification. Therefore, single cell RT-PCR is only indicative of a
decrease in sst mRNA content/neuron to an undetectable level compared
with that in nontreated cells.
In vivo, the efficacy of AS-sst receptor treatment was first assessed by radioligand binding studies. The penetration of ODNs appeared dependent on the distance from the ventricle, as AS-sst2 treatment decreased [125I]SRIH binding by only 9% in a distant structure, such as the basolateral amygdala vs. 46% in PeV and 31% in ARC nuclei. This ligand is sst2 selective, given the hardly detectable hypothalamic expression of sst3 and sst5 receptors as quantified herein or as visualized by in situ hybridization (28, 33). The decreases reached in PeV and ARC are similar to data obtained for other receptors: 35% reduction in NMDA-R1 (34), 48% reduction in dopamine D2 (35), and 30% reduction in serotonin 5-HT6 (36) receptor binding. With respect to sst1 receptors, we intended to use as a selective radioligand [125I]CH-288, a tyrosinated derivative of CH-275 (37), but its affinity was too weak to allow for an accurate quantitative evaluation (data not shown). We thus used an immunocytochemical approach, which indicated a marked decrease in the sst1 receptor protein both at the level of cell bodies in the PeV and nerve fibers in this nucleus as well as in the external layer of the median eminence. We finally used an in situ hybridization approach to quantify the effects of AS-sst1 ODN treatment on sst1 and sst2 mRNA levels. After saline or scrambled ODN infusion, both hypothalamic cell number and frequency distribution of grain counts for sst1 or sst2 mRNAs were similar. In contrast, after AS-sst1 infusion, the number of cells expressing sst1 mRNA was decreased by 53% in PeV, and the overall distribution of grain counts was shifted to significantly lower values, thus attesting to the efficiency of the treatment.
Taken together, these observations provide evidence of a potent and selective blockade of sst1 and sst2 receptors after AS-sst1 and AS-sst2 treatments in vitro as in vivo.
Effects of AS-sst1 or AS-sst2 treatment on SRIH modulations of
glutamate sensitivity in hypothalamic neurons
AS-sst1 treatment selectively suppressed the CH-275-induced
increase in glutamate responsiveness and detection of sst1 mRNA, at the
single cell level, whereas AS-sst2 selectively abolished the
octreotide-induced decrease in glutamate sensitivity and sst2 mRNA
expression. These observations are in keeping with the differential
modulation of glutamate sensitivity in hypothalamic neurons in culture
by sst1 or sst2 receptor subtypes (13). Although sst5 mRNA expression
was never detected at the single cell level, an increase in sst5 mRNA
signal was observed by quantitative RT-PCR, after AS-sst2 treatment.
Given that the sst2 and sst5 receptor subtypes belong to the same
SRIH-1 receptor family and that both can be activated by octreotide
(33), this modest increase may also represent a compensatory mechanism
to the blockade of sst2 receptors. However, the increase in sst5 mRNA
did not reach the residual value of sst2 mRNA expression, as measured
after AS-sst2 treatment, and octreotide-induced modulation of the
glutamate response was not found either.
Effects of AS-sst1 or AS-sst2 treatment on pulsatile GH secretion
in vivo
Although hypothalamic [125I]SRIH binding
sites were decreased after AS-sst2 treatment, the icv infusion of
AS-sst2 ODN did not modify the GH secretory pattern. This result
appears surprising in light of previous data suggesting that
hypothalamic sst2 receptors may be involved in the control of GH
secretion. Indeed, [125I]SRIH-binding sites
(38, 39) and sst2 mRNAs (11) have been visualized on GHRH ARC neurons.
Furthermore, sst2 knockout mice appear refractory to GH negative
feedback on arcuate hypothalamic neurons (40). However, in the latter
study GH still increased Fos immunoreactivity in PeV neurons, in
keeping with the well described GH stimulation of SRIH in PeV neurons
(18, 41, 42). The 31% decrease in ARC
[125I]SRIH binding observed after AS-sst2
infusion might be insufficient to influence GH secretion. However, this
seems unlikely, because comparable antisense-induced decreases in other
receptors led to physiological effects (18, 34, 35, 36, 43). Finally, ARC
[125I]SRIH binding is not restricted to GHRH
neurons. It is also present on POMC- and neuropeptide Y-synthesizing
neurons, but not on SRIH interneurons (44). Similarly, as shown
previously in the rat (11) and confirmed in the mouse herein, the
distribution of sst2 mRNA-containing neurons in the hypothalamus is
also not restricted to GHRH neurons. Such a multiplicity of expression
sites might be reflected in the lack of effect of AS-sst2 treatment on
pulsatile GH secretion.
In contrast, the in vivo infusion of AS-sst1 led to a marked decrease in GH pulse amplitude and overall mean plasma GH levels compared with those in saline-, AS-sst2-, or scrambled ODN-infused controls. In parallel, AS-sst1 treatment induced a decrease in PeV sst1 mRNA levels. Interestingly, sst2 mRNA levels were increased in both PeV and ARC. The finding of an AS-sst1-induced increase in sst2 mRNA levels suggests that SRIH, through sst1 receptor activation, may inhibit sst2 receptor expression. However, this was not reflected in the [125I]SRIH binding experiments. Thus, it is still unclear whether the marked changes in sst2 mRNA levels are reflected at the level of functional sst2 receptors in the hypothalamus. On the other hand, physiological data suggest that intrahypothalamic sst1 receptors are related to GH levels; although hypophysectomy reduces both sst1 and sst2 expression in ARC cells, only sst1 mRNA levels are restored by GH replacement (23). Moreover, in keeping with the well known sexual dimorphism of GH secretion, there is a marked sex-related difference in sst1, but not sst2, expression in the rat arcuate nucleus; both the number and labeling density of sst1 mRNA-expressing cells are 2- to 3-fold greater in males than in females (45). On the other hand, there is pharmacological (46) and physiological (45) evidence implicating the sst2 receptor subtype in GH regulation at the level of the pituitary. Taken together, these findings suggest a dual brain/pituitary sst1/sst2 receptor involvement in the control of GH secretion. The present findings provide further support for the idea that sst1 receptor interactions are involved in the intrahypothalamic control of pulsatile GH secretion.
The sst1 mRNA is coexpressed in PeV SRIH neurons projecting to the median eminence and appears mainly confined to terminals (10). SRIH is able to repress its own synthesis (47) and release (48) from these neurons. Multiple effects of sst1 receptor activation might be postulated. These effects might be inhibitory on SRIH PeV neurons, suggesting an autoregulatory function of the sst1 receptor as previously proposed (10). In addition, SRIH may directly modulate the activity of GHRH ARC neurons through sst1 receptors (11). This would account for the decreased amplitude of GH secretory pulses after AS-sst1 infusions. In keeping with this hypothesis, SRIH-induced inhibition of SRIH release from hypothalamic neurons (49, 50) and central SRIH-induced activation of GHRH release in vivo (4) were previously reported. At any rate, the effects of sst1 receptor activation must take into account the positive modulation of aminohydroxy-5-methyl-4-isoazole-propionate/kainate receptors as shown herein. Indeed, glutamate is the major excitatory neurotransmitter regulating the hypothalamic neuronal network (51). Similar modulations of fast neurotransmitter actions by hypothalamic neuropeptides such as neuropeptide Y or orexins (52) have recently been demonstrated.
Thus, an integrative scheme of SRIH actions in controlling pituitary GH secretion is now emerging. A body of evidence strongly suggests that sst1 receptors are preferentially involved in the intrahypothalamic network, whereas sst2 (46) and possibly sst5 (53, 54) receptor subtypes appear more important for the direct inhibition of GH release from pituitary somatotrophs. Such a model provides a framework in which it will be possible to integrate the effects of novel intervening substances, such as ligands of the GH secretagogue receptor or leptin, which are likely to also be involved in the complex functioning of the intrahypothalamic GH oscillator.
|
Footnotes |
|---|
2 C.L. and M.T.B.-P. contributed equally to this work.
Received September 20, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. M. Luque and R. D. Kineman Gender-Dependent Role of Endogenous Somatostatin in Regulating Growth Hormone-Axis Function in Mice Endocrinology, December 1, 2007; 148(12): 5998 - 6006. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
