| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Medicine (J.K., S.L.W., S.-M.L., I.M.C., R.L.L., S.C.C.) and Pediatrics (S.-M.L., R.L.L., S.C.C.), Columbia University College of Physicians & Surgeons, New York, New York 10032
Address all correspondence and requests for reprints to: Judith Korner M.D., Ph.D., Department of Medicine, Division of Endocrinology, Columbia University College of Physicians & Surgeons, 630 West 168th Street, Black Building Room 905, New York, New York 10032. E-mail: jk181{at}columbia.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-MSH
(-MSH) action at several members of the melanocortin receptor family
(3). Expression of the Agouti gene in mice is normally
limited to the skin, where the gene product, Agouti signaling protein
(ASP), blocks the action of -MSH at MC1R, resulting in pheomelanin
synthesis (3). Ectopic overexpression of ASP in mice segregating for
dominant mutations of Agouti produces pleiotropic effects
including yellow coat color, hyperphagic obesity, hyperinsulinemia,
diabetes, and increased body length. These effects are presumably
caused by antagonism of -MSH action in both the periphery (at MC1R)
and in the hypothalamus (at MC4R) (4). The obesity-related (but not
coat color) phenotypes are recapitulated by deletion of the
Mc4r gene (5).
A gene with high sequence homology to Agouti, Agouti-related
protein (Agrp; 6) or agouti-related transcript
(Art; 7), was isolated from mouse and human expression
libraries by homology searches. Expression of Agrp is
limited primarily to the hypothalamus and the adrenal gland and is
increased in the hypothalamus of obese mice deficient in leptin
(Lepob/Lepob)
or its receptor
(Leprdb/Leprdb).
Ubiquitous overexpression of Agrp complementary DNA (cDNA)
in transgenic mice causes obesity without affecting coat pigmentation
(6, 8). Recombinant Agouti-related protein (AGRP) protein is a potent
inhibitor of -MSH action on the CNS-specific melanocortin receptors,
MC3R and MC4R, but has no apparent attenuating activity for -MSH at
the peripheral receptors, MC1R and MC2R (6, 9). These data suggest that
AGRP may be a physiological regulator of feeding through modulation of
-MSH action at central melanocortin receptors.
Further support for the role of the melanocortin system in the
regulation of energy homeostasis are studies showing a decrease in
Pomc gene expression in the hypothalamus of ad
libitum fed ob/ob and db/db mice and
food-restricted lean rodents (10, 11, 12, 13, 14, 15). We have previously shown that
Pomc gene expression and peptide products, including
-MSH, are also decreased in the medial basal hypothalamus (MBH) of
the obese leptin receptor-deficient
LA/N-faf/faf
rat (16). The
LA/N-faf/faf
rat (also known as Koletsky or corpulent, cp) is
characterized by an autosomal recessive phenotype that includes
early-onset severe obesity, insulin resistance, and strain-dependent
diabetes (17). This phenotype is attributable to a point mutation
resulting in a premature termination codon in the extracellular domain
of the leptin receptor (18, 19). In this report, we analyze the
expression of both Agrp and Pomc in the MBH of
LA/N (lean and obese) and Sprague Dawley (SD) rats in ad
libitum and food-restricted states.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cloning of partial cDNA for rat Agrp
Total RNA from rat MBH (LA/N strain) was reverse transcribed and
subjected to PCR using oligonucleotide primers from coding sequences
conserved between mouse and human AGRP genes within exon 2
(5'-AGGGCATCAGAAGGCCTGACCAGG-3') and exon 3
(5'-CTTGAAGAAGCGGCAGTAGCACGT-3'). The partial rat cDNA was then
subcloned into the pBluescript plasmid vector containing T3 and T7
promoters (Stratagene, San Diego, CA). One clone was
selected that bore no alterations in sequence from the primary
amplified cDNA. The sequence data has been submitted to GenBank,
accession number AF206017. Sequence alignments were performed with the
BLAST program.
Preparation of messenger RNA (mRNA)
The MBH was dissected, as described previously (20), and
immediately homogenized in 1 ml cold homogenization buffer [10
mM Tris (pH 8.0), 3 mM
CaCl2, 2 mM
MgCl2, 0.5 mM dithiothreitol, 0.15%
Triton X-100] containing 0.3 M sucrose and 50 U of human
placental ribonuclease inhibitor. The homogenate was layered over 0.3
ml of homogenization buffer containing 0.4 M sucrose and
centrifuged at 2,500 x g, for 10 min at 4 C, to pellet
the cell nuclei. The supernatant was incubated with proteinase K and
extracted twice with phenolchloroform (1:1), and the cytoplasmic
RNA was precipitated with isopropanol. Total RNA was quantified by
spectrophotometry.
S1 Nuclease protection assay
RNA used to generate standard curves and
32P-labeled RNA probes were synthesized using
commercial transcription kits (Promega Corp., Madison,
WI). Sense and antisense 175-base Agrp RNAs were synthesized
from the plasmid vector containing T3 and T7 polymerase described
above. Sense and antisense Pomc RNAs were synthesized from
plasmid vectors containing the Sp6 promoter and a 538-bp
Pomc cDNA fragment spanning part of exon 3 (obtained from M.
Blum and J. L. Roberts, Mt. Sinai School of Medicine, New
York, NY; Ref. 21). Sense RNAs were quantified spectrophotometrically
and used to generate standard curves in the hybridization assays.
Components of the hybridization reaction included: 4 µg cytoplasmic
RNA in 5 µl H2O; 24.5 µl hybridization buffer
containing 80% formamide, 40 mM PIPES (pH
7.4), 400 mM NaCl, 1 mM
EDTA; 5 µg yeast RNA; and 32P-labeled riboprobe
(22). Reaction mixtures were heated at 85 C for 5 min, then incubated
for 16 h at 45 C. After hybridization, 300 µl S1 buffer [300
mM NaCl, 30 mM NaAc (pH
4.8), 3 mM ZnCl2, 20
µg/ml salmon sperm DNA] and 400–800 U S1 nuclease (Roche Molecular Biochemicals, Mannheim, Germany) were added and
incubated for 1 h at 56 C. Reaction mixtures were
phenol-chloroform extracted, precipitated, resuspended in 5
mM Tris (pH 7.5), 10 mM EDTA, and heated
for 5 min at 68 C before electrophoresis on a 4% nondenaturing
acrylamide gel. Protected bands were quantified by visualization with
an autoradiogram, followed by excision of the corresponding bands from
the gel and liquid scintillation counting (Packard 1500
Analyzer, Downers Grove, IL) and comparison with the standard
curve. Because the protected hybrids were smaller than the cellular
transcripts, 32P counts were normalized to the
full-length RNA species: 0.7 kb for full-length mouse Agrp
cytoplasmic RNA (6); 1.1 kb for full-length Pomc cytoplasmic
RNA (23). Results are presented as pg cytoplasmic RNA/µg total RNA.
The total mass of RNA isolated from the MBH of experimental and control
animals was not significantly different in any experiment.
In situ hybridizations with Agrp. In situ hybridizations were performed as described by Burke et al. (24). Brains were rapidly removed and frozen by immersion in isopentane on dry ice and then serially sectioned on a cryostat through the hypothalamus at 14 µm. For hybridization, sections were fixed by immersion in 4% paraformaldehyde 0.1 M phosphate buffer (pH 7.1) for 5 min, rinsed twice in PBS, and delipidated by successive immersion in increasing concentrations of ethanol and then chloroform. Sections were rehydrated and immersed in 2 x SSC, and then prehybridized at 50 C for 2 h with 1:1 formamide/prehybridization mix, consisting of 0.6 M NaCl, 10 mM Tris (pH 7.5), 0.04% Ficoll, 0.04% polyvinylpyrrolidone, 0.2% BSA, 2 mM EDTA, 1 mg/ml salmon sperm DNA, 1 mg/ml total yeast RNA, and 0.1 mg/ml yeast transfer RNA. Sections were then hybridized with 1:1 formamide/hybridization mix at 50 C for 16–18 h. Hybridization mix was similar to prehybridization mix but contained 20% dextran, 0.2 mg/ml salmon sperm DNA, 0.1 mg/ml total yeast RNA, and 35S-labeled Agrp cRNA probe at a final activity of approximately 12,000 cpm/ml. Riboprobes were synthesized as described above, except that 35S-uridine 5'-triphosphate, rather than 32P-uridine 5'-triphosphate, was used. After hybridization, brain sections were washed by immersion in 4 liters of 2 x SSC for 60 min at 40 C, followed by treatment with 30 µg/ml RNAase for 40 min at 37 C. Sections were then washed in 4 liters of 0.1 x SSC/0.05% sodium pyrophosphate/14 mM mercaptoethanol, dehydrated in ethanol, vacuum dried, and exposed in a cassette to Amersham Pharmacia Biotech (Piscataway, NJ) Hyperfilm b-max at -80 C for 3 weeks. The specificity of the hybridization was assessed by hybridization with sense riboprobe. The autoradiograms were analyzed on a Loats Associates image analysis system with Inquiry Software. 14C standards were included in each cassette to ensure that the optical densities recorded on the film from exposure to the labeled brain sections were within the linear response range, and to enable expression of optical densities in 14C-dpm equivalents. Exposure intensities were quantified by placing a fixed circular window over the arcuate nucleus and measuring the relative optical density. Background optical density was measured by placing the window over the VMN, an area devoid of AGRP neurons (7, 25, 26).
Semiquantitative RT-PCR
cDNA synthesis was performed with 2 µg MBH RNA using
ribonuclease-H-reverse transcriptase (Superscript II, Life Technologies, Inc. Rockville, MD) and an anchored
oligo-dT primer (Not PA, Promega Corp.). cDNA templates,
the equivalent of 20 ng total RNA, were amplified with Agrp-
and Actin-specific primers (spanning at least 1 intron) and
Taq DNA polymerase for 28 (Agrp) or 20 cycles
(Actin) in the presence of
35S-deoxy-ATP. These conditions were previously
determined to be within the linear range of amplification and will
measure relative message levels. The products were size-fractionated on
4% polyacrylamide gels, dried, and exposed to x-ray film. Bands
corresponding to the Agrp and Actin cDNAs were
quantified with a scanning densitometer (Bio-Rad Laboratories, Inc., Hercules, CA; GS700) and QuantOne
software (Bio-Rad Laboratories, Inc.).
Plasma insulin and glucose
Plasma insulin concentrations were measured with an RIA kit
using mouse insulin as a standard (Sensitive Insulin RIA Kit,
Linco, St. Charles, MO). Plasma glucose concentrations were
measured with a Glucometer Elite (Bayer Corp., Pittsburgh, PA).
Statistical analysis
Relative chemical concentrations of mRNA are reported as mean
values (pg/µg total RNA) ± SEM. For experiments
using solution hybridization and in situ hybridization,
significant differences between groups were determined by ANOVA,
followed by the Bonferroni/Dunn all-means test when more than two
groups were compared. Statistical analysis of RT-PCR was performed by
the Mann-Whitney test (one-sided with significance set at
P 
0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Within the amino terminus of the rat
sequence, there is a 3-amino-acid deletion, compared with the mouse
sequence. This region of the protein is not required for antagonism of
-MSH (6). The highly conserved cysteine-rich motif in the carboxyl
terminus is maintained, with the exception of a 2-amino-acid difference
and a single amino acid deletion (Fig. 1; shaded box).
|
). In
these dissections, the total RNA content (12.11 ± 81
vs. 12.10 ± 41 ug,) and the absolute amount
of Agrp mRNA in the MBH (5.54 ± 0.59 vs.
5.54 ± 0.62 pg) were also not significantly different in the
obese rats, compared with lean controls. Because the absence of an
increase in Agrp expression in the obese
(faf/faf)
LA/N rat was not the predicted result (levels of Agrp mRNA
are increased in the MBH of mice homozygous for inactivating mutations
of Lepr; Refs. 7, 27), the anatomical localization
of Agrp RNA in the rat brain was analyzed by in
situ hybridization using the same rat probe. Similar to the mouse,
Agrp RNA expression was restricted to the arcuate nucleus
(Fig. 3). However, in support of the
results with solution hybridization, there was no significant
difference in levels of Agrp expression (mean optical
densities expressed in 14C-dpm equivalents and
measured in the linear range) in the MBH between lean (258 ± 33;
n = 4) and obese
faf/faf
(304 ± 40; n = 3) rats (P = 0.4). There was
also no significant regional difference in Agrp RNA
expression when the rostral and caudal arcuate were analyzed
separately. Semiquantitative RT-PCR was performed to further confirm
these results. Again, no difference in hypothalamic Agrp
mRNA expression was detected between lean and obese
faf/faf
rats, whereas db/db mice exhibited a 3-fold increase,
compared with lean controls (Fig. 4).
|
|
|
and 6A).
In contrast, Pomc mRNA within the same dissections was
decreased by 25% in fasted vs. fed rats: 0.51 ± 0.04
vs. 0.68 ± 0.03 pg/µg total RNA (P=0.006;
Fig. 6B).
|
|
) and showed a 231% increase in
Agrp mRNA expression, compared with fed controls showing a
4.1 ± 0.7% gain in body weight: 0.74 ± 0.11 vs.
0.32 ± 0.03 pg Agrp mRNA/µg total RNA
(P = 0.002; Fig. 7A).
However, there was no significant difference in Agrp mRNA
levels between fasted and fed obese
(faf/faf;
mean weight, 350 g) LA/N rats exhibiting a 12.1 ± 0.6%
decrease and a 4.7 ± 0.7% increase in body weight, respectively:
0.41 ± 0.05 vs. 0.36 ± 0.05 pg/µg total RNA
(P = 0.5; Fig. 7A). Again, no difference was
detected in Agrp expression in fed obese LA/N rats, compared
with fed lean controls. After the 48-h fast, there was a significant
decline in plasma insulin levels in both lean and obese rats (Table 1;
P < 0.0001). Insulin levels were significantly higher
in the fed and food-restricted obese LA/N rats, compared with their
respective lean controls. Plasma glucose levels also declined after the
48-h fast in both lean and obese rats.
|
|
). There was also a significant decrease in Pomc mRNA
levels in food-restricted lean LA/N rats, compared with lean LA/N rats
fed ad libitum: 0.34 ± 0.05 vs. 0.50
± 0.07 pg/µg total RNA (P = 0.03; Fig. 7B). No
change, however, was detected in Pomc mRNA levels in 48-h
food-restricted obese LA/N rats, compared with obese rats fed ad
libitum: 0.25 ± 0.02 vs. 0.24 ± 0.03
pg/µg total RNA (P = 0.9; Fig. 7B). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because the faf mutation is a null variant (28), it is unlikely that the lack of an increase in basal Agrp expression in faf/faf animals is caused by suppression from partial leptin activity at the mutant rat leptin receptor, as might be the case with the Leprfa mutation (though we did not examine this possibility). Thus, in the rat, the normal Agrp response to fasting may involve leptin receptor-mediated phenomena, but factors in addition to the leptin-leptin receptor axis must also be important for the regulation of Agrp gene expression. If leptin were the sole regulator of Agrp expression, then levels of Agrp mRNA would be increased in the hypothalamus of fed leptin receptor-deficient rats. It has also been shown that factors other than leptin-leptin receptor are involved in the regulation of hypothalamic Agrp gene expression in mice: in db/db mice, Agrp gene expression is further increased upon fasting, although the animals are leptin insensitive (27); and in streptozotocin-induced diabetic mice, fasting stimulates hypothalamic Agrp mRNA, whereas plasma leptin levels remain unchanged (29).
So important to survival is the drive to eat, that there likely exist redundant mechanisms to ensure that this drive is maintained. Orexigenic factors other than AGRP, such as neuropeptide Y (NPY), which is coexpressed with AGRP in the same neurons of the medial arcuate nucleus (30), may contribute to the obesity syndrome in the faf rat. In fact, NPY peptide levels are significantly increased in the arcuate nucleus of ad libitum fed obese faf/faf rats compared with lean controls (31). Intracerebroventricular administration of insulin suppresses the fasting-induced increase in NPY gene expression in the arcuate nucleus of lean, but not obese, Zucker (fa/fa) rats (32). Obese faf/faf rats are hyperinsulinemic (see Results and Refs. 33, 34) and also have increased plasma concentrations of corticosterone (16). The effects of insulin and corticosterone on Agrp expression are unknown, but elevated levels of these hormones may have different effects on Agrp expression in faf/faf rats, compared with db/db mice. Species differences in gene regulation, presumably caused by differences in the function of regulatory elements, are known to occur in other physiological circumstances. For example, mice show a 20-fold increase in circulating leptin in late stages of pregnancy, whereas rats and humans only show a modest 2-fold increase (35).
In the rat, as in the mouse, AGRP probably provides a tonic orexigenic
signal by antagonizing MC3R and MC4R. However, there is a noteworthy
difference between the mouse and rat amino sequence in the highly
conserved cysteine-rich C-terminal domain of the AGRP and ASP proteins.
The Cys-110 to Cys-117 octapeptide loop of human AGRP mimics the
conformation of -MSH, and is critical in mediating its antagonism of
-MSH action (36). There is 100% homology of amino acid sequence in
this region between mouse, human, and bovine AGRP, but two conservative
substitutions and one amino acid deletion are detected in the rat
sequence (Fig. 1). These changes may have functional significance,
although the rat sequence does maintain the RFF111–113 triplet
presumably most critical in determining the binding affinity of AGRP to
the melanocortin receptors (36). The deletion of several amino acids in
the N-terminal of rat AGRP (Fig. 1) probably does not have functional
significance, because this portion of the protein is not required for
antagonism of -MSH (6, 37). The differences in the rat AGRP sequence
are authentic, given that only a single PCR product was obtained from
total RNA of rat MBH. Also, genomic clones were obtained that
corresponded to the cDNA, with intron/exon junctions observed in both
mice and humans (data not shown).
In the same experiments, we also quantified the levels of hypothalamic
Pomc mRNA and showed that its expression is decreased in
ad libitum fed obese
faf/faf
rats, compared with lean controls, and in food-restricted lean LA/N and
SD rats, compared with fed controls. We have previously shown that
levels of the anorexigenic peptide, -MSH, are also decreased in the
hypothalamus of obese
faf/faf
rats (16). Because both POMC and AGRP fiber tracts project to the
same brain regions, it is likely that some of the biological effects of
leptin depend on the competitive interactions of -MSH with AGRP at
the same melanocortin receptors (26, 38). Although Agrp mRNA
levels are not changed in the fed obese
faf/faf
rats, compared with fed lean LA/N controls, the ratio of
Pomc to Agrp mRNA is decreased by 57%. It would,
therefore, be of interest to also measure hypothalamic AGRP peptide
levels in these animals, because the peptide ratio of -MSH to AGRP
may be an important modulator of food intake.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 1, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  V. Tolle and M. J. Low In Vivo Evidence for Inverse Agonism of Agouti-Related Peptide in the Central Nervous System of Proopiomelanocortin-Deficient Mice Diabetes, January 1, 2008; 57(1): 86 - 94. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Goto, H. Arima, M. Watanabe, M. Hayashi, R. Banno, I. Sato, H. Nagasaki, and Y. Oiso Ghrelin Increases Neuropeptide Y and Agouti-Related Peptide Gene Expression in the Arcuate Nucleus in Rat Hypothalamic Organotypic Cultures Endocrinology, November 1, 2006; 147(11): 5102 - 5109. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. D. Morrison, G. J. Morton, K. D. Niswender, R. W. Gelling, and M. W. Schwartz Leptin inhibits hypothalamic Npy and Agrp gene expression via a mechanism that requires phosphatidylinositol 3-OH-kinase signaling Am J Physiol Endocrinol Metab, December 1, 2005; 289(6): E1051 - E1057. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R G P Denis, C Bing, S Brocklehurst, J A Harrold, R G Vernon, and G Williams Diurnal changes in hypothalamic neuropeptide and SOCS-3 expression: effects of lactation and relationship with serum leptin and food intake J. Endocrinol., October 1, 2004; 183(1): 173 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Baran, E. Preston, D. Wilks, G. J. Cooney, E. W. Kraegen, and A. Sainsbury Chronic Central Melanocortin-4 Receptor Antagonism and Central Neuropeptide-Y Infusion in Rats Produce Increased Adiposity by Divergent Pathways Diabetes, January 1, 2002; 51(1): 152 - 158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Hwa, L. Ghibaudi, J. Gao, and E. M. Parker Central melanocortin system modulates energy intake and expenditure of obese and lean Zucker rats Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2001; 281(2): R444 - R451. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Fekete, J. Kelly, E. Mihaly, S. Sarkar, W. M. Rand, G. Legradi, C. H. Emerson, and R. M. Lechan Neuropeptide Y Has a Central Inhibitory Action on the Hypothalamic-Pituitary-Thyroid Axis Endocrinology, June 1, 2001; 142(6): 2606 - 2613. [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 |
