| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
NEUROENDOCRINOLOGY |
Department of Biological Sciences, University of Alberta, Edmonton, Alberta T6G 2E9, Canada
Address all correspondence and requests for reprints to: Dr. R. E. Peter, Department of Biological Sciences, University of Alberta, Edmonton, Alberta T6G 2E9, Canada. E-mail: dick.peter{at}ualberta.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
CART was initially isolated using PCR differential display as mRNA which levels increased in the rat striatum following acute administration of psychomotor stimulants such as cocaine and amphetamine (2). The amino acid sequence deduced from CART mRNA corresponded to the unknown peptide isolated from ovine hypothalamus by Spiess et al. (3). To date, cDNA encoding for CART has been cloned in rat (2), mouse (4), and human (5). The mammalian pre-pro-CART peptide has a 27 amino acid signal peptide, followed by a pro-CART protein. CART mRNA is found in two alternatively spliced forms, which results in the production of two pro-CART peptides, the long form, with 102 amino acid residues, and the short form, with 89 amino acid residues. Rodents possess the two forms, whereas only the short form is present in humans. In rodents, the shorter variant is more abundant than the long form (2, 6). Pro-CART has several potential cleavage sites (5), which are indicative of further processing of CART proteins into smaller peptides. To date, at least six CART peptides have been identified in mammals (7). The processing of CART peptides appears to be tissue specific (7, 8).
In addition to the mammalian brain, CART peptides are also found throughout the nervous system and in the spinal cord, pituitary, gut, pancreas, and adrenals (6, 7, 8, 9, 10, 11, 12, 13, 14, 15).
In mammals, CART peptides have a major role in the regulation of feeding. CART mRNA is expressed in hypothalamic areas implicated in the control of feeding behavior, such as the arcuate and paraventricular nucleus (2, 16, 17, 18). CART injections increase c-fos expression in the hypothalamus (19, 20), and CART neurons establish synaptic connections with neurons expressing other hypothalamic appetite-regulating peptides (21, 22). CART mRNA expression is lowered in fasting conditions (16, 23). Central injections of CART fragments of various lengths have been shown to cause a dose-dependent feeding inhibition in rodents (8, 16, 24, 25, 26, 27, 28, 29, 30, 31) and to decrease NPY-induced feeding (24, 32). CART (55–102), a naturally occurring fragment isolated from ovine hypothalamus (3), appears to be the most potent fragment. The numbers in the fragments derive from the predicted signal peptide cleavage site in the long form of CART (5). CART peptides appear to be regulated by the adipocyte hormone leptin. CART mRNA expression is greatly reduced in obese ob/ob rats (16, 23, 32) and anorexic anx/anx mice (33), both of which also have low circulating levels of leptin. Leptin treatment activates CART hypothalamic neurons (34) and increases CART mRNA expression in the hypothalamic arcuate nucleus (16, 35).
In goldfish, several appetite-related neuropeptides have been isolated (36). These include neuropeptide Y (37), corticotropin-releasing factor (38), cholecystokinin (39), and gastrin-releasing peptide (40). We recently showed that two human CART fragments, CART 62–76 and CART 55–102, act as potent satiety factors when centrally injected in goldfish (41), suggesting the presence and a physiological role of CART-like peptides in fish. Recent data also suggest the presence of CART peptide immunoreactivity in Atlantic salmon brain (42). To date, little is known about the structure of CART peptides in nonmammalian vertebrates.
In this study we have determined the nucleotide sequence of cDNAs encoding two forms of CART-like peptides in goldfish. We used RT-PCR, Northern blot, and slot blot analyses to examine the distribution and quantify CART mRNA in several tissues. We also examined the effects of fasting on CART mRNA expression in brain and its variations following a meal. To examine whether CART effects on food intake are regulated by leptin, we assessed feeding behavior in fish submitted to intracerebroventricular coinjections of CART and leptin and evaluated the effects of leptin treatment on CART gene expression in brain.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Preparation of RNA
Total RNA from whole brain, eye, liver, gastrointestinal tract,
kidney, gill, gonads, and muscle and various brain areas, including
olfactory bulbs and tracts, telencephalon and the preoptic region,
hypothalamus, optic tectum-thalamus, and posterior brain
(cerebellum-medulla), were extracted from fresh tissues by a
single-step acid guanidinium thiocyanate-phenol-chloroform extraction
method using Trizol RNA isolation reagent (Life Technologies, Inc./BRL, Gaithersburg, MD). Final RNA concentrations were
determined by optical density reading at 260 nm.
Cloning of cDNA by reverse transcription (RT) and rapid
amplification of cDNA ends (RACE)
To isolate the 3'-end of cDNA, two partially degenerated
primers, P1 (GCCGGNGAGCARTGYGCNGT) and P2 (GGGAAGYTNTGYGAYTGYCC) were
designed on the basis of the putative biological active (C-terminal)
portion of CART peptides in the sequences of mammalian CART (primers
shown in Fig. 1C
). Total RNA was reverse
transcribed with dT-AP
(GGCCACGCGTCGACTAGTAC(T)17) using SuperScript II
reverse transcriptase (Life Technologies, Inc.). The
3'-end of cDNA was amplified by two rounds of PCR, with adaptor primer
(AP, GGCCACGCGTCGACTAGTAC) and P1, and AP and P2, respectively, using a
Robocycler 40 temperature cycler (Stratagene, La Jolla,
CA). PCR products were separated by agarose gel electrophoresis, and
the band of desired size was excised and purified using Geneclean II
kit (Bio 101, La Jolla, CA). The desired PCR products were
then subcloned using the pGEM-T vector system (Promega Corp., Madison, WI). Plasmid DNA containing the DNA insert was
purified by an alkaline lysis method (43). DNA sequence
analyses were carried out on a PE Applied Biosystems
automated sequencer (Perkin-Elmer Corp., Norwalk, CT)
according to the manufacturer’s protocol. Both strands of cloned DNA
were sequenced in opposite directions using T7 and SP6 sequencing
primers that flank the inserted DNA.
|
Cloning of genomic DNA
Genomic DNA was isolated from gut tissue using Trizol reagent
according to instructions of the manufacturer. DNA was then amplified
by two rounds of PCR using P5 and P9 (GCAGAGGACGCTTGCTGTGA) and P8 and
P10 (GTTGCGCGCTAAATTCAGAACC). PCR products were then purified, cloned,
and sequenced as described above.
Detection of mRNA expression in different tissues
RT-PCR assay and Southern blot analysis. Samples of
total RNA were treated with RQ1 DNase (Promega Corp.) to
ensure noncontamination with genomic DNA. Four micrograms total RNA
were reverse transcribed into cDNA with dT-adapter primer using
SuperScript II reverse transcriptase (Life Technologies, Inc.). A PCR amplification was carried out for 35 cycles with
the primers Y1 (CCATGGAGAGCTCCAAACTC) and Y2 (TCTTGACCCTTTCCTGATGG) or
F1 (TCTGATCTGCTTGTTGACCG) and F2 (GTTTCGTCTGCAGCTTTTCC) (primers shown
in Fig. 1C
). Twenty five microliters PCR reaction mixture were then
fractionated in a 1.5% agarose gel, blotted onto Hybond-N membrane
(Amersham Pharmacia Biotech, Little Chalfont,
Buckinghamshire, UK) by capillary transfer and fixed by baking at 80 C
for 2 h. The membranes were prehybridized at 65 C for 1 h in
a hybridization solution containing 0.5 M
NaHPO4 (pH 7.2), 7% SDS, 1 mM EDTA
(pH 8.0), and 1% BSA (fraction V, Sigma, St. Louis,
MO). The membranes were then transferred into fresh hybridization
solution to which labeled probe was added. The probes (P5/P9 and
P8/P10) were synthesized by PCR using P5 and P9 (GCAGAGGACGCTTGCTGTGA)
and P8 and P10 (GTTGCGCGCTAAATTCAGAACC) and 32P
labeled by a random primer method (T7 Quick Prime;
Pharmacia, Uppsala, Sweden). Hybridization was carried
out overnight at 65 C. The membranes were subsequently washed with a
series of stringent washes (0.04 M
NaHPO4, 1% SDS, 1 mM EDTA) and
exposed to a phosphoimaging screen for 96 h. As a negative
control, PCRs were performed in the absence of cDNA to examine the
cross-contamination of samples.
As internal control of the RT step, PCR amplification was carried out for 35 cycles of 94 C for 1 min, 50 C for 1 min, and 73 C for 1 min with a pair of primers, A1 (CTACTGGTATTGTGATGGACTCCG) and A2 (TCCAGACAGAGTATTTGCGCTCAG), designed on the basis of ß- actin partial cDNA sequence in goldfish (39).
Northern blot
Fourty micrograms total RNA were denaturated in a
formaldehyde/formamide denaturing buffer (42% formamide, 8%
3-(N-morpholino)propanesulfonic acid, 15% formaldehyde)
at 65–70 C for 15 min. RNA was then separated by electrophoresis on a
formaldehyde-1.5% agarose gel. RNA was transferred to Hybond-N
membrane by capillary transfer with 20x saline sodium citrate.
Membranes were fixed by baking at 80 C for 2 h and hybridized with
either Y1/Y2 or F1/F2 probes. Probe labeling and hybridization were
carried out as for Southern blots.
Slot blots
Ten micrograms total RNA samples from individual fish were
denaturated in a denaturing solution (42% formamide, 1.6%
3-(N-morpholino)propanesulfonic acid, 2 M
formaldehyde) at 65 C for 15 min. Samples were then blotted onto
Hybond-N nylon membrane (Amersham Pharmacia Biotech) by
vacuum suction on a slot blot apparatus (Bio-Rad Laboratories, Inc.). The membranes were fixed by baking at 80 C for 2 h.
Membranes were hybridized with either Y1/Y2 or F1/F2 probes. Probe
labeling and hybridization were carried out as for Southern blots.
Intracerebroventricular (ICV) injections
Brain ICV injections were performed following procedures
described by Peter and Gill (44). Briefly, following deep
anesthesia, a three-sided flap was cut in the roof of the skull using a
dentist drill equipped with a circular saw. The flap was folded to the
side, exposing the brain. Fish were then placed in a specially designed
stereotaxic apparatus. The needle of a 5-µl microsyringe was
stereotaxically placed in the preoptic region of the brain third
ventricle according to coordinates (+ 1.0 M, D 1.2) taken
from the stereotaxic atlas of the goldfish brain (44).
Following injection of 2 µl test solution, the needle was withdrawn
and the space in the cranial cavity filled with teleost physiological
saline (45). The skull flap was put back in place and
secured by surgical thread. Fish were then returned to their tanks and
normally recovered from anesthesia within 2–5 min.
Fish were injected with different doses of human CART (55–102) (American Peptide Co., Inc., Sunnyvale, CA) and recombinant murine leptin (PeproTech, Inc., Rocky Hill, NJ) in teleost physiological saline. A stock solution of recombinant murine leptin was made in fish physiological solution acidified with 5 µl 0.1 N HCl, and neutralized with 5 µl 0.1 N NaOH. A stock solution of CART was made in saline. Stocks were aliquoted, stored at -20 C, and subsequently diluted in physiological saline.
Observational experiments
Fish were tested in random order in terms of treatment and days.
Forty-eight hours before experimentation, two fish were moved into an
observation tank and starved. For each experiment, two fish were
injected and observed for feeding behavior and food consumption. An
approximately 4% bw ration of pellets per fish was administered 15 min
after injection. Experiments were carried out at the regular feeding
time to which the fish had been adapted (0800 h). Behavioral
observations and measurement of food consumption commenced upon entry
of pellets into the tank and were made for 1 h. Observations were
divided into 15-min periods. Feeding behavior and exploring behavior
were monitored. Feeding behavior was assessed by counting the number of
feeding acts. A "feeding act" was defined as an occasion when a
fish approached a pellet. A feeding act was complete when the fish
consumed the pellet or incomplete when the fish either engulfed a
pellet and then spat it out or bumped it with a closed mouth. Food
consumption was converted to milligrams of food consumed/wet bw per
time feeding based on the mean pellet weight fed to fish.
The total number of acts (TA) was defined as the sum of the number of feeding acts and nonfeeding acts. The number of nonfeeding acts reflected locomotor/searching behavior and was assessed by counting the times fish mouthed, picked up, and spat gravel, or bumped any object in the tank (air stone, wall) or their tank mate.
To verify that the ICV procedure itself did not influence feeding, food intake was assessed for control fish submitted to either anesthesia alone or sham operations and compared with saline-treated animals.
Stressed animals were easily detected because they displayed characteristic behavioral signs, such as rapid opercular movements, lowering of the fins, or decreased locomotor activity. These fish did not feed and were not taken into consideration in the study.
Postprandial variations of CART mRNA expression
Five experimental groups, each containing six to eight fish,
were acclimatized as described above. Fish were gonadal recrudescent
(December to March). Preliminary studies showed no significant
difference in CART mRNA expression between male and female groups (data
not shown) in a given brain region and at a given sampling time.
Consequently, both sexes were used in the experiments. On experimental
day, a control (time zero) group was killed at the scheduled feeding
time. Brain tissue was sampled for total RNA extraction and
determination of mRNA levels by slot blot analysis. Of the remaining
four groups, two were fed with the normal ration and two remained
unfed. One group of fed and one group of unfed fish were killed and
sampled for brain tissue 2 and 6 h following food
administration.
Effects of food deprivation on CART mRNA expression
Two experimental groups, each containing 8–12 fish, were
acclimatized as described above, before food deprivation. Control fish
were fed a normal daily ration at 2% bw (0800 h). The experimental
group was food deprived for 96 h. This fasting period was chosen
as, in goldfish, 72–96 h fasting induces a significant increase in
food intake (unpublished observations) and variations in the expression
of appetite-related neuropeptides, such as NPY (47). All
fish were killed at the end of the food deprivation period, and brain
tissues were collected for total RNA extraction and determination of
mRNA levels by slot blot analysis.
Effects of leptin injections on CART-induced inhibition of
feeding
Fish were submitted to ICV injections and observed for 1 h
for assessment of food intake and feeding behavior. As controls, fish
were submitted to ICV injections of either saline or vehicle (fish
physiological solution acidified with 5 µl 0.1 N HCl, and
neutralized with 5 µl 0.1 N NaOH). Because no significant
difference in food intake was found between saline- and vehicle-treated
fish, data from the two treatments were pooled into one group and used
for comparison with all other treatment groups. Fish were ICV injected
with saline/vehicle (n = 15), leptin at 1 (n = 7) or 5 ng/g
(n = 8), 5 ng/g CART (n = 12), or coinjected with 5 ng/g CART
and 1 ng/g leptin (n = 7) or 5 ng/g CART and 1 ng/g leptin (n
= 8). The dose of 5 ng/g CART was chosen because it is within the range
of optimal doses for inhibition of food intake (41).
Effects of leptin injections on CART mRNA expression in the
brain
CART gene expression was determined in four brain regions
(telencephalon-preoptic region, hypothalamus, optic tectum, and
olfactory bulbs) 2 and 6 h after ICV injection of saline and
6 h after ICV injection of 10 ng/g leptin. Fish were ICV injected,
returned to their tank, and allowed to recover for 5–10 min. Fish were
then given food pellets and observed for 1 h to verify that the
fish behaved according to the treatment. Fish were then killed 2 or
6 h following food administration and sampled for brain tissues. A
control group was injected with saline and killed and sampled for brain
tissue 2–3 min after feeding. Total RNA was prepared and subjected to
slot blot analysis.
Data analysis and statistics
Analysis of the sequences was performed using SignalP v1.1
software from the Center for Biological Sequence Analysis Web site
(http://www.cbs.dtu.dk/). Sequence comparisons were performed
using MAP and ClustalW v1.8 from the Baylor College of Medicine Search
Launcher Web site (http://searchlauncher.bcm.tmc.edu/).
In gene expression studies, the hybridization signals were scanned using PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA) and quantified using ImageQuant software (Molecular Dynamics, Inc.). The mRNA levels for each CART form were expressed as a ratio of CART mRNA to ß-actin mRNA levels (internal control) and normalized as a percentage of the mRNA levels from the control (time 0) group.
To compare variations of feeding parameters and postprandial and postleptin treatment variations of CART mRNA levels for a given brain area, statistical analyses were conducted using ANOVA followed by a pairwise Student-Newman-Keuls multiple comparison test. A t test was used to compare CART mRNA levels in fed and fasted fish in a given brain area. Significance was considered at P < 0.05. Data are expressed as mean ± SEM.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
The CART form I precursor cDNA consists of 578 bp, comprising a
5'-untranslated region (47 bp), an open reading frame (351 bp), a stop
codon (TGA), and a 3'-untranslated region (177 bp), including two
possible polyadenylation signals (AATAAA and ATTAAA) and a poly(A) tail
(Fig. 1A). The CART form II precursor cDNA consists of 654 bp,
comprising a 5'-untranslated region (55 bp), an open reading frame (360
bp), a stop codon (TGA), and a 3'-untranslated region (239 bp),
including two possible polyadenylation signals (ATTAAA and AATAAA) and
a poly(A) tail (Fig. 1B). The deduced amino acid sequence shows that
goldfish CART form I and form II are part of a 117 and a 120 amino
acids precursor, respectively. Potential signal peptide cleavage sites
occur in the precursor after amino acid residue 25 (M) in form I and
(A) in form II (Fig. 1).
Both goldfish CART precursors also contain several potential enzyme cleavage recognition sites. These include two dibasic sites (Lys-Arg and Lys-Lys) at positions 71–72/78–79 and 68–69/75–76 for CART I and CART II, respectively.
To confirm the CART cDNA sequences, PCR of genomic DNA was performed to obtain partial gene sequences for CART. Comparison of goldfish cDNA and genomic DNA confirmed the sequence obtained by RACE and showed that CART mRNA is encoded by three exons. Two introns are located within the coding region at positions 54 (102 bp between nucleotides encoding L and L) and 82 (132 bp between nucleotides encoding T and C) for form I and 57 (174 bp between nucleotides encoding L and L) and 84 (304 bp between nucleotides encoding M and C) for form F (data not shown, GenBank accession numbers AY033816 and AY033817).
Tissue distribution of CART mRNA
Northern blot. Northern blot analysis revealed the presence of
single bands of approximately 700 and 800 bp for CART I and CART II
mRNAs, respectively (Fig. 2). Bands were
seen in total RNA from the olfactory bulb, hypothalamus,
telencephalon-preoptic region, and optic tectum for form I, and
hypothalamus, optic tectum, and telencephalon-preoptic region for form
II. No signal was generated in the gut and ovary for either CART
forms.
|
) that were detected in Southern blot
analysis. CART I mRNA was detected in the brain, eye, gill, kidney,
ovary, and testis but not in the muscle, liver, and gut (Fig. 3). CART
II mRNA was detected in the brain, eye, kidney, testis, and ovary. Both
CART I and II mRNAs were detected in all brain regions and in the
pituitary (Fig. 3). No products were detected from the negative
control. PCR amplification of the cDNA samples using a specific primer
set of goldfish ß-actin (A1 and A2) was used as internal control of
the RT step and revealed a band of 600 bp. The ß-actin mRNA was
detected in all tissues and brain samples examined, verifying the
quality of the mRNA and DNA samples.
|
Using slot blot with specific probes for each CART mRNA, both CART
forms were detected in the brain and ovary but not in the GI tract
(Fig. 4). Within the brain, both forms I
and II were detected in the optic tectum, telencephalon-preoptic
region, hypothalamus, and olfactory bulbs but not in the posterior
brain and pituitary (Fig. 4, A and B). The olfactory bulbs exhibited
the highest level of CART I mRNA, approximately 5-fold higher than that
in the hypothalamus (Fig. 4A). Form II was predominant in the optic
tectum (Fig. 4B).
|
Postprandial CART mRNA expression
As revealed by slot blot analysis, CART I mRNA expression
showed no postprandial variation in either the optic tectum or the
telencephalon-preoptic region. CART I mRNA expression in these regions
in both fed and unfed fish was identical to that of control (time zero)
fish at 2 and 6 h after the scheduled feeding time (Fig. 5A). In the hypothalamus and olfactory
bulbs, CART I mRNA levels of fed fish were significantly higher than
that of control fish and unfed fish 2 h following food
administration. At 6 h, CART I hypothalamic levels in unfed fish
were lower than that of control 0 h fish but similar to that
of fed 6 h fish. In the olfactory bulbs, CART I mRNA levels in
6 h unfed fish were significantly lower than that of both
0 h control and 6 h fed fish.
|
).
Effect of food restriction on CART mRNA expression
As revealed by slot blot analysis, food restriction for 4 d
induced a decrease in expression of CART I mRNA in the
telencephalon-preoptic region, hypothalamus, and olfactory bulbs but
not in the optic tectum (Fig. 6). Under
the same conditions, CART II mRNA expression was decreased in the
olfactory bulbs but not in the telencephalon, hypothalamus, and optic
tectum.
|
). Doses of 5 ng/g CART caused a
significant decrease in food intake after 60 min. The cumulative food
intake in fish cotreated with 1 ng/g leptin and 5 ng/g CART was similar
to that of fish treated with 5 ng/g CART. The cumulative food intake in
fish cotreated with 5 ng/g leptin and 5 ng/g CART was lower than that
of fish treated with 5 ng/g CART. Both groups of leptin/CART-treated
fish had levels of feeding lower than that of the saline group.
|
). The number
of TAs in leptin-treated fish was similar to that in control fish. Fish
treated with 5 ng/g CART displayed more TAs than control fish. In fish
cotreated with 5 ng/g CART and 1 ng/g leptin, the number of TAs was
similar to both CART- and saline-treated fish. Fish cotreated with 5
ng/g CART and 5 ng/g leptin had a number of TAs similar to that of
saline-treated fish.
Effects of leptin ICV on CART gene expression
Slot blot analysis revealed that, in the optic tectum, there were
no significant differences in CART I (Fig. 8A) or CART II (Fig. 8B) mRNA levels in
saline-treated fish, at either 2 or 6 h after injection, compared
with 0 h fish. At 6 h, leptin-treated fish had CART I mRNA
levels that were similar to 0 h fish but higher than 6 h
saline-treated fish. At 6 h, leptin-treated fish had CART II mRNA
levels that were similar to both 0 and 6 h saline-treated
fish.
|
, A and B), there were no
significant differences in either CART I or II mRNA levels in any group
at 0, 2, or 6 h.
In the hypothalamus (Fig. 8A), at 2 h after feeding, CART I mRNA
levels were increased in saline-treated fish, compared with 0 h
fish. At 6 h, CART I mRNA levels in saline-treated fish were
similar to that of 0 h fish. At 6 h, leptin-treated fish had
CART I mRNA levels similar to that of the leptin-injected fish at
2 h and significantly higher than that of 0 and 6 h
saline-treated fish. There were no significant differences in CART II
hypothalamic mRNA levels in any group at 0, 2, or 6 h.
In the olfactory bulbs, at 2 h after feeding, levels of CART I
mRNA were higher than that of 0 h fish, whereas levels of CART II
mRNA were similar to that of 0 h fish (Fig. 8, A and B). At 6
h, levels of both CART I and II mRNA in saline-treated fish were
similar to that of 0 h fish. At 6 h, leptin-treated fish
displayed levels of both CART I and II mRNAs that were similar to
levels in 0 and 6 h saline-treated fish.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
In rodents, CART mRNA is found in two alternatively spliced forms, resulting in the production of two proteins, the long and short forms. The long form, which is present in rodents but not in humans, originates from the translation of the last 29 nucleotides located at the end of the first exon (2). The short variant is more frequent than the long form (2, 4, 6). As in humans, goldfish do not appear to have splice variant forms.
Goldfish CART precursors are about 120 amino acids (aa), consisting of
a short signal peptide (25 aa) and the mature peptide of approximately
100 aa. They present a high homology with their mammalian counterparts.
The two goldfish pro-CART forms present 70% similarity, whereas the
two mature forms are 76% homologous. Comparison of amino acid
sequences with known sequences of other species show that for both
pro-CART and mature CART, there is a 40–50% homology between both
goldfish and mammalian forms (Fig. 9). In
the C-terminal portions of the peptide sequences (last 46 aa), there is
an 85% homology between the two goldfish forms and 70–80% similarity
between goldfish and mammalian CART. Variability occurs in the
N-terminal portion of the peptide rather than its C-terminal end
region, which is not surprising, because the C-terminal region is
thought to be the biologically active segment of the CART. Because the
portion (55–102) is highly homologous to mammalian CART (55–102), it
is not surprising that central injection of human CART (55–102) in
goldfish has pronounced effects on feeding (41). Cysteine
residues located in the same position in mammalian and goldfish CART
peptides suggest the presence of disulfide bonds and an identical
tertiary structure of the protein in the two species. This would
account for the conservation of physiological activity of the human
form in goldfish.
|
The genomic organization of the goldfish is tetraploid so that multiple forms of CART are not surprising. The two CART forms isolated in this study appear too different to originate from the same gene. In addition to these two forms, the presence of other forms of CART peptide variants in goldfish is not to be ruled out.
Distribution of CART mRNA in goldfish brain and peripheral
tissues
We found CART mRNA throughout the goldfish brain and in the
pituitary. The mRNAs of the two goldfish CART forms show a different
distribution within the brain. Form I mRNA is most abundant in the
olfactory bulb and hypothalamus, whereas form II mRNA shows highest
levels in the optic tectum-thalamus. In the posterior brain and
pituitary, mRNAs of both forms were detected by RT-PCR but not by
Northern blot or slot blot, suggesting a low expression in these
regions. The distribution of form I is similar to that found in
mammals, in which CART peptides and mRNA are present throughout the
brain, including various regions of the hypothalamus, forebrain,
midbrain, cortex, and cerebellum (2, 6, 11, 18, 48). CART
mRNA and peptides have also been shown in rat pituitary (6, 11, 12). In salmon, CART-immunoreactive material has been reported
in the preoptic nucleus and laminar nucleus (42). The
presence of CART mRNA in goldfish hypothalamus and pituitary suggests a
role of these peptides in endocrine regulation of fish. The presence of
CART mRNA in the hypothalamus also concurs with the role of CART in
regulating feeding in goldfish (41). Notably, the highest
levels of CART I mRNA were found in the olfactory bulbs of goldfish.
High mRNA expression in goldfish olfactory bulbs have previously been
shown for other satiety peptides, such as cholecystokinin (39, 49), corticotropin-releasing factor (38), and
tachykinins (50). This might be explained by the fact
that, in goldfish, the olfactory bulbs play an important role in the
regulation of feeding because the hypothalamus is connected to the
olfactory system and olfactory tract lesions affect feeding behavior
(51). In mammals, CART immunoreactivity is found in the
olfactory bulbs, retinal ganglion cells, and spinal cord, suggesting
that CART peptides are involved in peripheral sensory information
processing (6).
The different expression pattern of CART II mRNA suggests that this peptide has a distinct role from that of CART I. CART II is expressed mainly in the brain, and it is present only at low levels in peripheral tissues. Within the brain, it is predominant in the optic tectum-thalamus. High levels of CART have also been demonstrated in human thalamus and other sensory-related brain regions (48). A putative role of CART peptides in the transmission/reception of olfactory and visual information in the brain may account for the disturbances in motor activity, and the seizures and anxiety-like behavior induced by central injection of CART peptides (26, 29, 31, 41, 52, 53).
RT-PCR analysis did not reveal the presence of detectable CART mRNA in goldfish liver and muscle, which is in agreement with studies in mammals (2). Messenger RNA encoding the CART precursor was not detected in the gastrointestinal tract of the goldfish. This is also consistent with what is known in mammals: even though stomach and intestine contain CART immunoreactivity (6, 11, 13, 14), CART mRNA has never been detected in the gastrointestinal tract (2, 11), suggesting that CART peptides are synthesized in other tissues and transported to the gut, perhaps via sensory fibers or the vagus nerve. Indeed, CART immunoreactivity is found in vagal efferent fibers, often colocalized with cholecystokinin receptors, suggesting that CART peptides may mediate cholecystokinin satiety effects (54).
CART mRNA was detected in the kidney, gills, and gonads of goldfish. To our knowledge, this is the first report of the presence of CART peptides in peripheral tissues of fish. In rats, CART immunoreactivity or mRNA is not present in the kidneys or respiratory tract (2, 11) but has been reported in the adrenal medulla. Because in teleost fish, the adrenal medulla intermingles with the interrenal tissue, the presence of CART mRNA in goldfish kidney might be a consequence of a contamination with adrenal tissue.
CART immunoreactive fibers have been reported in the vas deferens and epididymis of male rats (55). However, Douglass et al. (2) failed to detect CART mRNA in either testis or ovaries. It is possible that CART mRNA was not detected in gonads owing to the poor sensitivity of the method used (i.e. Northern blot). CART peptides appear to have some role in reproduction because differences in expression of CART have been reported in rat amygdala (56) and CART might influence the pituitary-gonadal axis (57, 58). The significance of the presence of CART and its biological function in goldfish gonads requires further investigation.
Influence of nutritional status and leptin injection on CART mRNA
expression
Postprandial variations of CART I mRNA expression were seen in
goldfish hypothalamus and olfactory bulbs, whereas CART II expression
showed no postprandial variations. Levels of CART I mRNA increased
2 h following a meal and returned to basal levels (levels at meal
time) after 6 h. Similar variations in mRNA expression have been
reported in goldfish for orexigenic peptides, such NPY (59, 60) and anorexigenic peptides such as tachykinin
(49) and cholecystokinin (48) mRNAs. The
existence of a feeding-related pattern in CART I expression further
indicates a role of this peptide in regulation of food intake in
goldfish.
Fasting induced significant reduction in levels of CART I mRNA. These results are consistent with what is known in mammals. Hypothalamic levels of CART mRNA are decreased in hypoleptimic states, such as seen in obese ob/ob or anorexic anx/anx mice, and after prolonged fasting (16, 33, 34, 35). CART II expression was little affected by fasting. A small but significant decrease was observed only in the olfactory bulbs. Taken together with the lack of postprandial variations in CART II mRNA expression, our results indicate that CART II is not involved in the regulation of feeding in goldfish. It is possible, however, that a more sensitive method than slot blot analysis might be needed to detect small variations in CART II expression associated with feeding behavior. It is also possible that more pronounced changes in nutritional status (e.g. a food deprivation of several weeks) may be necessary to trigger detectable variations in CART II expression.
Our results demonstrate that leptin at low ICV doses, that do not significantly affect feeding when given alone, enhance the inhibitory effects of CART peptide on feeding behavior of goldfish. In mammals (35, 61) and reptiles (62), peripheral and central administration of leptin induces a significant decrease in food intake. In fish, data on leptin remain contradictory. Indirect evidence points to the presence of a leptin-like molecule in fish. Immunoreactivity to an antibody against mammalian leptin was demonstrated in eel (63), six species of bony fish (64), and lamprey (65). Furthermore, the amount of leptin-like immunoreactivity appears to increase in fasted fish and be correlated with percentage of body fat (64). However, Baker et al. (66) failed to show any effect of ip injection of human leptin in immature coho salmon; following 2–4 wk of continuous injection of leptin via implanted osmotic pumps, fish showed no variations in growth, body weight, or energy storage. Our results show that, although leptin alone at low doses has no effects on food intake of goldfish, it regulates the action of at least one appetiteregulating peptide (i.e. CART). These are not only short-term effects, as seen in the rapid effects in feeding behavior (30 min), but also long-term effects because leptin influences CART mRNA expression in the brain. This is the first report of a functional action of a leptin-like molecule in fish. The apparent discrepancy between our results and the findings of Baker et al. (66) might be simply owing to the difference in the animal used (immature salmon vs. mature goldfish) and in the methodology (peripheral vs. central administration, long-term vs. short-term effects, measure of food intake vs. measure of body weight). In rodents, peripheral administration of recombinant leptin has been shown to increase CART mRNA levels in the arcuate nucleus (16, 18) and to induce Fos expression in hypothalamic CART neurons. In situ hybridization histochemistry studies show that hypothalamic CART neurons express leptin-receptor mRNA, indicating that leptin directly acts on CART neurons in distinct nuclei of the rat hypothalamus (18, 60). It is noteworthy that CART appears to be a mediator of the effects of leptin not only in the control of appetite but also in the hypothalamic control of the pituitary-gonadal axis (57, 58). Such a role of CART peptides in the reproductive function of goldfish has yet to be investigated.
|
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
This work was supported by Grant A6371 from the Natural Sciences and Engineering Research Council (NSERC) of Canada (to R.E.P.).
Abbreviations: aa, Amino acids; bw, body weight; CART, cocaine- and amphetamine-regulated transcript; ICV, intracerebroventricular; RACE, rapid amplification of cDNA ends; RT, reverse transcription; TA, total number of acts.
Received May 21, 2001.
Accepted for publication August 6, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
