| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Physiology and Biophysics (P.D.F., D.G.R., R.A.S.) and Obstetrics and Gynecology (D.K.C., R.A.S.), Graduate Program in Neurobiology and Behavior (M.J.C.), and the Specialized Cooperative Centers Program in Reproduction Research (D.K.C., R.A.S.), University of Washington, Seattle, Washington 98195
Address all correspondence and requests for reprints to: Dr. Robert A. Steiner, Department of Physiology and Biophysics, Box 357290, University of Washington, Seattle, Washington 98195-7290. E-mail: steiner{at}u.washington.edu
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Although it would appear that leptin acts on the brain to affect feeding and reproduction, the neural circuitry mediating leptins action in the central nervous system has yet to be fully elucidated. To date, the majority of effort in mapping this circuitry has focused on the hypothalamus, an area of the brain known to be involved in the regulation of ingestive behaviors, metabolism, and reproduction (14). This is due to the abundant expression of Ob-R messenger ribonucleic acid (mRNA) in several hypothalamic nuclei (12, 15). Recently, we identified cells in the dorsal Raphe nucleus (DR) in pigtailed macaques (Macaca nemestrina) that express Ob-R mRNA (12), corroborating a similar observation in the rat reported by Elmquist et al. (15). Neurons located in the Raphe nuclei are the primary source of serotonin (5-HT) in the brain. Based on this observation and evidence that 5-HT has pronounced effects on both feeding and reproduction (16, 17), we postulated that 5-HT neurons in the brain stem are direct targets for leptins action. To test this hypothesis, we used double label in situ hybridization (ISH) to determine whether cells that express Ob-R mRNA also express the mRNA for the serotonin transporter (SERT), a highly specific marker for 5-HT neurons (18). Using computer imaging and quantitative analysis, we mapped the distribution of Ob-R mRNA in relation to SERT mRNA and report the numbers of cells that express both mRNAs in various divisions of the Raphe nuclei in the pigtailed macaque brain.
| Materials and Methods |
|---|
|
|
|---|
Tissue was obtained from adult female pigtailed macaques (Macaca nemestrina; 6.4 and 9.5 yr of age; n = 2) through the Tissue Distribution Program at the Regional Primate Research Center at the University of Washington. All procedures were approved by the animal care committee of the School of Medicine at the University of Washington, in accordance with the NIH Guide for the Care and Use of Laboratory Animals.
Tissue preparation
Animals were first sedated with ketamine and then overdosed with sodium pentobarbital. Approximately 20 min later, the brain was removed and dissected. A block of tissue containing the midbrain and pons was obtained by laying the brain ventral side up and making a coronal cut caudal to the mammillary bodies. Occipital cortex and portions of the cerebellum were removed, and the block was cut caudally at the pontine-medullary junction. Tissues were frozen immediately on dry ice and stored at -80 C. Cryostat sections were cut at 20 µm, thaw-mounted onto SuperFrost Plus slides (Fisher Scientific, Pittsburgh, PA), and stored at -80 C.
Riboprobe preparation
33P-Labeled Ob-R complementary RNA (cRNA) riboprobe. A construct was provided by Joseph Kuijper (ZymoGenetics, Inc., Seattle, WA), which contains a 380-bp fragment of complementary DNA (cDNA) coding for the human Ob-R inserted within Bluescript S/K (Stratagene, La Jolla, CA). This fragment is complementary to bases 10161396 (amino acids 275402) in the extracellular domain of the receptor. The cDNA was linearized with EcoRI (New England Biolabs, Inc., Beverly, MA), and the radiolabeled antisense cRNA probe was synthesized in vitro by inclusion of the following ingredients in a volume of 20 µL: 250 pmol [33P]UTP (NEN Life Science Products, Boston, MA); linearized cDNA (2 µg); T7 RNA polymerase (Roche Molecular Biochemicals, Indianapolis, IN); 1 x transcription buffer; 500 µmol/L ATP, CTP, and GTP; and ribonuclease (RNase) inhibitor (Stratagene, La Jolla, CA). Residual DNA template was digested by incubation with deoxyribonuclease (Stratagene), and this reaction was terminated by addition of 2 µL 0.5 mol/L ethylenediamine tetraacetic acid (pH 8.0). Finally, 1 µL yeast transfer RNA (10 mg/mL) was added as a carrier. The cRNA was separated from unincorporated nucleotides on a Quick Spin Sephadex G-50 column (Roche).
Digoxigenin-labeled SERT cRNA riboprobe. A 253-bp cDNA fragment of the human SERT gene, which was generated by PCR using the full-length human SERT cDNA as a template (19, 20), was provided by Dr. Cynthia Bethea (Oregon Regional Primate Research Center, Beaverton, OR). The cDNA was linearized with NotI (New England Biolabs, Inc.), and the cRNA probe was transcribed in vitro in a volume of 40 µL containing linearized cDNA (4 µg), 1 x DIG RNA labeling mix containing digoxigenin-11-UTP (Roche), SP6 RNA polymerase (Roche), 1 x transcription buffer, and RNase inhibitor. Residual DNA template was digested by incubation with deoxyribonuclease, and this reaction was terminated by addition of 2 µL 0.5 mol/L ethylenediamine tetraacetic acid (pH 8.0). The cRNA was separated from unincorporated nucleotides on a Quick Spin Sephadex G-50 column. The optimal concentration of the SERT riboprobe was determined empirically.
Double label ISH
To identify cells that contain both Ob-R and SERT mRNA, we performed double label ISH using slight modifications of previously published protocols (12, 21). Briefly, the tissue was fixed with 4% paraformaldehyde, acetylated, and delipidated. Next, prehybridization solution, consisting of 25% transfer RNA (10 mg/mL) in hybridization buffer, was applied to the sections, which were then covered with silane-coated glass coverslips and incubated in humid chambers for 2 h at 55 C. The sections were rinsed in 2 x SSC (standard saline citrate), dehydrated with ethanol, and air-dried. Next, hybridization buffer containing freshly denatured 33P-labeled Ob-R cRNA probe (6.8 x 105 cpm/µL), digoxigenin-labeled SERT cRNA probe, and transfer RNA (1.7 mg/mL) was applied to the tissue. Slides were covered with silane-coated glass coverslips and incubated in humid chambers overnight at 57 C. The following day, the tissue was treated with RNase A and washed under conditions of increasing stringency, including two successive 30-min washes at 60 C in 0.1 x SSC. The tissue was then blocked for 60 min with 2% normal sheep serum, incubated with antidigoxigenin fragments conjugated to alkaline phosphatase (Roche) diluted 1:1000 for 3 h at 37 C, and kept at 4 C overnight. The next day, slides were incubated in a chromagen reaction with 4-nitro blue tetrazolium-chloride (340 µg/mL; Roche), 5-bromo-4-chloro-3-indolyl phosphate (175 µg/mL; Roche), and levamisole (240 µg/mL) until cells were clearly visible at the light microscopic level. The slides were then dipped briefly in 70% ethanol, air-dried, and coated in 3% parlodion dissolved in isoamyl acetate. After air-drying, the slides were dipped in NTB-3 emulsion (Eastman Kodak Co., Rochester, NY), exposed for 11 days, and developed.
Control experiments
The identity and integrity of the 33P-labeled Ob-R cRNA probe were verified by PAGE against known standards. The integrity of the digoxigenin-labeled SERT cRNA probe was verified by agarose gel electrophoresis. The results of sense controls with these probes have been documented previously (12, 19).
Analysis
Coexpression of SERT and Ob-R mRNAs within the same cells was
analyzed in a manner similar to a previously published protocol
(12). The brainstem Raphe nuclei were divided into several
regions along the rostral to caudal axis using the Template Atlas of
the Primate Brain as a guide
(http://rprcsgi.rprc.washington.edu/
atlas), and representative
sections were chosen from each animal for each region. The most rostral
section contained the caudal linear Raphe nucleus (CLi; Fig. 1A
). More caudally, the Raphe nuclei were
divided into the DR and median Raphe (MR) by the medial longitudinal
fasciculus. The DR and MR were further subdivided into rostral (Fig. 1B
) and caudal (Fig. 1C
) portions by the appearance medially of the
cerebellum and the fourth ventricle. Thus, a representative section
from each of five areas (CLi, rostral DR, rostral MR, caudal DR, and
caudal MR) was chosen per animal for analysis. Atlas plates were chosen
to represent these regions, and the distribution of Ob-R mRNA- and SERT
mRNA-containing cells was transferred to these plates (Fig. 1
).
|
0.5% of the total number of SERT mRNA-expressing
cells). Signal to background ratios (SBRs) for individual cells were
calculated as previously described (22). For each animal,
the number of counted cells with a SBR greater than a particular value
(e.g. 0, 1, 2, etc.) was calculated and then
converted to a percentage of the total counted cells (cumulative
frequency curve). The percentages at each SBR interval were averaged
across animals to produce means and SEMs. The
percentage of cells deemed to be double labeled was compared among
anatomical regions at a SBR criterion of 5 by one-way ANOVA with
repeated measures. | Results |
|---|
|
|
|---|
The majority of SERT mRNA-containing cells were found in the CLi, DR,
and MR. The locations of these cells at different levels of the macaque
brain stem are illustrated in Fig. 1
. SERT mRNA-containing neurons with
overlying clusters of silver grains, indicating the presence of Ob-R
mRNA (Fig. 2
), were found in all three
Raphe nuclei. Measurements of SBRs for Ob-R mRNA in SERT mRNA-positive
cells are displayed as a cumulative density function in Fig. 3A
. The ordinate value of each point
represents the percentage of neurons with a SBR greater than a
particular value (abscissa). This graph depicts the
percentage of SERT mRNA-containing neurons considered to be double
labeled for Ob-R mRNA at any threshold (SBR) level. Figure 3B
presents
the percentage of SERT mRNA-containing cells that also express Ob-R
mRNA with a SBR greater than 5 for each anatomical region. Using a SBR
of 5 as an arbitrary threshold, the percentage of double labeled cells
varied considerably by anatomical region (P < 0.005,
by ANOVA). In the rostral DR, few SERT mRNA-positive cells had
overlying clusters of silver grains, as reflected by the rapid decline
in the cumulative density curve for this area (Fig. 3B
). In contrast, a
high percentage of cells in both the caudal DR and caudal MR expressed
the mRNAs for both SERT and Ob-R (Fig. 3
, A and B). An intermediate
level of double labeling was found in the CLi and rostral MR (Fig. 3B
).
|
|
| Discussion |
|---|
|
|
|---|
The serotonin transporter is used by serotonergic neurons to clear 5-HT from the synaptic junction. In the rat, SERT mRNA expression in the Raphe nuclei is found exclusively in neurons that are immunoreactive for either tryptophan hydroxylase (the rate-limiting enzyme in 5-HT synthesis) or 5-HT itself (18, 24). We cannot dismiss the possibility that some of the SERT mRNA-expressing cells may be glia, but the literature suggests that this is unlikely (25). The probe we used to detect Ob-R mRNA was targeted to a portion of the gene encoding the extracellular domain of the receptor; thus, in our double labeling experiment, we could not distinguish between the long form (signaling) and the short form (nonsignaling) of the receptor. However, subsequent experiments using a riboprobe specific to Ob-Rb revealed that this isoform is present in the Raphe (our unpublished observation), making it likely that 5-HT neurons actually express Ob-Rb. We have provided an estimate of the percentage of SERT/Ob-R mRNA-expressing neurons as a function of SBR, allowing one to gauge the amount of coexpression based on different criteria.
The projections of 5-HT neurons have been studied in a variety of mammalian species, including nonhuman primates. Based on the patterns of development and connections in the adult monkey brain, the 5-HT system can be divided into two subsystems (26). The rostral subsystem (CLi, DR, and MR) projects expansively to the forebrain, whereas the caudal subsystem (magnus, obscurus, and pallidus Raphe nuclei) projects to the brainstem and spinal cord. Among the nuclei of the rostral subsystem, we observed a pronounced difference in the amount of Ob-R/SERT mRNA double labeling between the rostral and caudal Raphe nuclei. This distinction may have functional implications for leptins actions on 5-HT neurons. Tract-tracing studies (using both antero- and retrograde tracing) involving the Raphe in rats indicate that there are differential projections to the forebrain from rostral and caudal portions of the DR (27, 28). A study of the projections of rostral vs. caudal MR found that although the MR projects to regions of the forebrain distinct from those innervated by the DR, there appear to be no major rostrocaudal differences in MR projections (29). The implications of the differential expression of Ob-R mRNA in the primate Raphe nuclei remain unknown, but may reflect leptins targeting of specific afferents to the forebrain.
The present work offers evidence for colocalization of SERT and Ob-R mRNAs, suggesting that some of leptins effects to suppress feeding may be mediated through alteration of serotonergic activity. It is well established that the 5-HT system influences feeding, as evidenced in part by the ability of 5-HT reuptake inhibitors such as fenfluramine and dexfenfluramine to inhibit food intake in primates (30, 31). Mice that are null mutants for the 5-HT1B or 5-HT2C receptor subtype are less sensitive to the anorectic/satiating effects of fenfluramine/dexfenfluramine, implicating these receptor subtypes in mediating the effects of these drugs on feeding (32, 33), although only the 5-HT2C knockout mouse displays an obesity phenotype (34). Hypothalamic areas in which direct microinjection of serotonergic agents affect food intake in rats include the paraventricular (PVN) and ventromedial (VMN) nuclei (35). These two nuclei also show increased expression of the immediate-early gene product c-Fos after injection of the 5-HT reuptake inhibitor fluoxetine (36). The PVN and VMN are also implicated in leptins action on feeding. The protein or mRNA for Ob-R is known to be expressed in these nuclei in rats (15, 37), and leptin injection induces expression of c-Fos in the VMN and PVN of fasted rats (38). Whether leptin activates the VMN/PVN indirectly (through 5-HT release) as well as directly (through Ob-R in the hypothalamus) remains to be elucidated.
5-HT may also play a role in mediating leptins stimulatory effects on the reproductive axis. Most experiments investigating the effects of 5-HT on reproductive function have been performed in rats, where it seems that 5-HT can be either stimulatory or inhibitory to LH release depending on the circulating steroid milieu (17). There is also evidence implicating 5-HT in the regulation of sexual behavior as well as gonadotropin secretion in humans (39, 40). A direct effect of 5-HT on GnRH secretion is suggested by the observations that serotonergic fibers terminate in the proximity of GnRH perikarya/fibers (41). To date, little is known about the neural circuitry mediating leptins effects on the reproductive system. The question of whether 5-HT plays a role in mediating leptins effects on reproduction can only be answered after further experimentation.
In summary, the present results demonstrate coexpression of Ob-R and SERT mRNAs in neurons in the brainstem of the macaque. Although it has been established that leptin acts on the brain to influence feeding and reproduction, the detailed neural circuitry mediating these effects has been only partially elucidated, and the present work now extends the list of known targets for leptins action to include 5-HT neurons in the brainstem of the monkey.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Present address: Ligand Pharmaceuticals, Inc., 10275 Science Center
Drive, San Diego, California 92121. ![]()
3 These authors contributed equally to this work. ![]()
Received June 2, 2000.
Revised September 25, 2000.
Accepted October 3, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
O. O. Yevtushenko, S. J. Cooper, R. O'Neill, J. K. Doherty, J. V. Woodside, and G. P. Reynolds Influence of 5-HT2C receptor and leptin gene polymorphisms, smoking and drug treatment on metabolic disturbances in patients with schizophrenia The British Journal of Psychiatry, June 1, 2008; 192(6): 424 - 428. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Wade, P. Juneja, A. W. MacKay, J. Graham, P. J. Havel, L. H. Tecott, and E. H. Goulding Synergistic Impairment of Glucose Homeostasis in ob/ob Mice Lacking Functional Serotonin 2C Receptors Endocrinology, March 1, 2008; 149(3): 955 - 961. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. A. Clark, S. M. J. MohanKumar, B. S. Kasturi, and P. S. MohanKumar Effects of central and systemic administration of leptin on neurotransmitter concentrations in specific areas of the hypothalamus Am J Physiol Regulatory Integrative Comp Physiol, February 1, 2006; 290(2): R306 - R312. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. D. Sullivan, L. C. Howard, A. H. Clayton, and S. M. Moenter Serotonergic Activation Rescues Reproductive Function in Fasted Mice: Does Serotonin Mediate the Metabolic Effects of Leptin on Reproduction? Biol Reprod, June 1, 2002; 66(6): 1702 - 1706. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Cunningham, J. M. Scarlett, and R. A. Steiner Cloning and Distribution of Galanin-Like Peptide mRNA in the Hypothalamus and Pituitary of the Macaque Endocrinology, March 1, 2002; 143(3): 755 - 763. [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 |