INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

A VARIETY OF neurotransmitter systems appear to be involved in the pathogenesis of stress-induced gastric mucosal injury, including dopamine, norepinephrine, 5-hydroxytryptamine, thyrotropin-releasing hormone, and neurotensin systems (3, 14, 21). Neurotensin is a tridecapeptide, originally isolated from bovine hypothalamus, that is expressed in both the central nervous system and the distal small bowel (4). In addition to functioning as a neurotransmitter or neuromodulater, neurotensin plays an important role in the central nervous system, affecting gastrointestinal function such as stimulating growth of various gastrointestinal tissues (10), inhibiting gastric acid secretion (38), and maintaining gastric mucosal blood flow during cold water restraint (CWR) (37).

A number of studies have shown that both the intracerebroventricular administration of neurotensin and direct administration of neurotensin into specific brain regions of rats attenuate gastric mucosal injury produced by CWR (16, 25, 26). The precise protective mechanism of exogenous neurotensin, administered centrally, is not known. Previous work has demonstrated that direct injection of neurotensin into the nucleus accumbens, a terminal region of the mesolimbic dopamine system, has the same inhibitory effect on gastric acid secretion (34, 38) as intracerebroventricular neurotensin, suggesting that neurotensin's protective effect is mediated through interactions within the mesolimbic dopamine system (34).

At least four lines of evidence support the notion that neurotensin is closely involved in the mesolimbic dopamine system (20): 1) neurotensin-like immunoreactivity and neurotensin receptors have been detected in all brain structures containing dopaminergic cell bodies and terminals, including the ventral tegmental area and nucleus accumbens, both part of the mesolimbic system (18); 2) in situ hybridization studies have shown that the dopamine D3 receptor mRNA is expressed in the nucleus accumbens, where its distribution matches that of neurotensin mRNA (7); 3) administration of neurotensin into the ventral tegmental area increases release of dopamine in the nucleus accumbens (19); 4) intracerebroventricular neurotensin enhances the metabolism of dopamine in the nucleus accumbens (27). Interactions between neurotensin and the mesolimbic dopaminergic system are complex. For example, D2 and D3 dopamine receptors have opposite effects on the expression of neurotensin mRNA in the nucleus accumbens of rats (7).

The neurotensin receptor is a member of the G protein-coupled receptor family (29). Chronic treatment of rats with the nonpeptide neurotensin receptor antagonist SR-48692 causes an increase in both the number of neurotensin binding sites in whole brain membrane homogenates and in neurotensin receptor mRNA levels in the ventral mesencephalon (1). These observations suggest that neurotensin receptors are upregulated following their pharmacological blockade in vivo and that endogenous neurotensin regulates neurotensin receptor mRNA levels.

It is not known whether stressful conditions, such as CWR, influence the expression of neurotensin or neurotensin receptor in vivo. Because the administration of neurotensin into the central nervous system protects rats against gastric mucosal injury produced by CWR, it was hypothesized that CWR decreases endogenous neurotensin concentration in the central nervous system and that this decrease in neurotensin, in turn, results in increased neurotensin receptor expression. The present studies were undertaken to test these hypotheses.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Male Sprague-Dawley rats (Charles River, Wilmington, MA) weighing 250-300 g were used. Animals were housed in a temperature- and humidity-controlled environment with a 12:12-h light-dark cycle and free access to food and water and were fasted for 18 h before each experiment. Different groups of rats were used for each experimental protocol.

CWR-induced gastric injury. This protocol has been approved by the Institutional Animal Care and Use Committee in this institution. Rats were lightly anesthetized and stapled into wire screen restraints and then submersed to the neck in 20°C water for the specified time. After CWR, the rats were immediately killed by decapitation. The stomachs were removed and perfused intraluminally with 4% formaldehyde solution. Gastric mucosal lesions were evaluated by an unbiased observer who measured the length and width of each lesion using a ×3.5 magnifier. Each lesion was considered to be an ellipse, and the surface area was calculated using the formula area = length × width × /4 (34).

Isolation of brain nuclei. The brain was rapidly removed, placed in a brain blocker (David Kopf Instruments, Tujunga, CA), and cut into coronal sections. The nucleus accumbens, prefrontal cortex, ventral tegmental area, and hypothalamus were dissected from the sections on a cold plate as described by Heffner and colleagues (13). In brief, to obtain the prefrontal cortex, a coronal cut was made ~3 mm caudal of the frontal pole, and the olfactory tubercule on the ventral surface was removed, leaving ~50-60 mg of tissue. A second coronal cut was made ~1 mm rostral to the optic chiasm, and the nucleus accumbens was removed from this coronal slab by trimming the cortical tissue medially and ventrally, making a horizontal cut through the anterior commissure and a vertical cut through the lateral extent of the nucleus accumbens. Samples of nucleus accumbens weighed ~20 mg. The hypothalamus was dissected by taking a vertical slice between the optic chiasm anteriorly and the mammillary bodies posteriorly and then taking a parasagittal cut through the perihypothalamic sulcus and a horizontal cut below the anterior commissure. The ventral tegmental area was dissected from a coronal slab formed by coronal cuts ~5.5 and 7.0 mm rostral to the optic chiasm. A horizontal cut is made across the brain stem at the level of the rhinal sulcus and dorsal edge of the substantia nigra. The nigral tissue and telencephalic tissue are removed by a ventromedial cut to the base of the brain stem. The remaining tissue medial to this cut contains the ventral tegmental area.

Radioimmunoassay for measurement of neurotensin. For measurement of neurotensin concentration, the tissue was weighed, frozen on dry ice, and stored at 70°C. For the extraction of neurotensin (28), 10 volumes of distilled water were added to the frozen tissue and heated for 10 min in a boiling water bath. The mixture was homogenized, an equal volume of 6% acetic acid was added, and the mixture was heated for a further 10 min. The mixture was cooled and centrifuged at 10,000 g for 10 min. The supernatant was studied for neurotensin concentration and the pellet for protein determination. For radioimmunoassay, all samples were prediluted at least 1:10 in 1% bovine serum albumin-borate buffer. Neurotensin radioimmunoassay kits were purchased from IncStar (Stillwater, MN). This neurotensin antiserum is directed toward the COOH-terminal and crossreacts 100% with neurotensin-(113), 1.2% with neurotensin-(813), and less than 0.002% with eledosin, physalamine, bombesin, and substance P. The detection limit in tissue was 3 pg/tube (15 pg/ml). The recovery rate was 108-119% and the intra-assay coefficient of variation was 8%. The concentration of neurotensin was expressed as nanograms per milligram protein.

Assay of neurotensin binding to brain membranes. Neurotensin binding assays were performed according to the method of Goedert (12). Fresh tissue was used for the neurotensin binding studies. The specified brain regions were homogenized in 10 volumes of cold 50 mM tris(hydroxymethyl)aminomethane (Tris) · HCl, pH 7.4, using a Polytron (Brinkman Instruments, Westbury, NY) at setting 7 for 15 s. After 30 min centrifugation at 4°C at 50,000 g, the pellet was resuspended in 10 volumes of cold 50 mM Tris · HCl and incubated at 37°C for 30 min to remove endogenous neurotensin. After 20 min centrifugation at 50,000 g, the pellet was resuspended in binding buffer (50 mM Tris · HCl, pH 7.4, containing 0.1% bovine serum albumin, 40 mg/ml bacitracin, and 1 mM EDTA) to yield a concentration equivalent to 10 mg original tissue/ml buffer.

One hundred microliters of freshly prepared brain membranes in a total volume of 1 ml, various concentrations of [3,11-tyrosyl-3,5-³H] neurotensin-13 (New England Nuclear, Boston, MA; saturation experiments), or 2 nM [³H]neurotensin and various amounts of unlabeled neurotensin-(113) (Cambridge Research Biochemicals; displacement experiments) were incubated at 25°C for 15 min. Nonspecific binding was defined as the radioactivity bound in the presence of excess (1 µM) unlabeled neurotensin. At the end of incubation, the tubes were quickly transferred to ice and the mixtures rapidly filtered under vacuum through GF/B glass fiber filters (Whatman, Maidstone, UK) that had been pretreated for at least 3 h with 0.2% polyethylenimine in water. Each filter was washed with 20 ml cold 50 mM Tris buffer, pH 7.4. Radioactivity was determined by liquid scintillation counting, and protein concentrations were determined using the Bio-Rad protein assay (Bio-Rad, Hercules, CA). The maximum number of neurotensin binding sites (B_max) and half-maximal inhibition (IC₅₀) values were formulated from least-squares analysis of Scatchard plots, using the Lundon program. All experiments were performed in duplicate using membranes pooled from two brains.

Oligonucleotides used for amplification and hybridization. Oligonucleotide primers and probes were synthesized based on published sequences for rat neurotensin (17), rat neurotensin receptor (29), and rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (11). Their sequences were neurotensin, 5'-CTTCAGCTGGTGTGCCTGAC-3' (sense, nucleotides 270-290) and 5'-GAGTATGTAGGGCCTTCTGGG-3' (antisense, nucleotides 720-740), which amplify a 470-base pair (bp) fragment; neurotensin receptor, 5'-CATCTGGGTACACCATCCC-3' (sense, nucleotides 384-402) and 5'-AGTCCACTGTTCATCCGAG-3' (antisense, nucleotides 1002-1020), which amplify a 636-bp fragment; for GAPDH, 5'-CAAGATTGTCAGCAATGCAT-3' (sense, nucleotides 511-530) and 5'-CTTGATGTCATCATACTTGGC-3' (antisense, nucleotides 856-836), which amplify a 346-bp fragment. The following oligonucleotides served as amplification product-specific probes: for the neurotensin fragment, 5'-CCTCCTGAATTATCTCCCAGTG-3' (nucleotides 603-62); for the neurotensin receptor fragment, 5'-CCTGGATGACGACCTTGACAGTGGC-3' (nucleotides 693-717); for the GAPDH fragment, 5'-GCTGTGGGCAAGGTCATCCCAGAGCTGAAC-3' (nucleotides 725-754).

Reverse transcription-polymerase chain reaction analysis. Tissue for RNA isolation was immediately frozen in liquid nitrogen and stored at 70°C. Total RNA was isolated using the UltraSpec II system (24). Reverse transcription (RT) of the RNA and subsequent polymerase chain reaction (PCR) amplifications were carried out in separate reaction tubes. For RT, 2 µg total RNA was reverse transcribed using 150 U Moloney murine leukemia virus reverse transcriptase (Promega, Madison, WI) in a total volume of 25 µl containing 500 ng poly(dT)₂₀ oligonucleotide primer, buffer provided by the vendor, 10 mM 1,4-dithiothreitol, and 1 µl RNAsin (Promega). All reagents were combined and then incubated as follows: 25°C, 10 min; 42°C, 60 min; 95°C, 5 min. For PCR amplification, in all cases, 2.5 µl RT mixture was used as template in 100 µl total volume. Amplification was carried out in a Robocycler (Stratagene, La Jolla, CA). Cycle conditions were as follows: neurotensin (95°C, 30 s; 60°C, 30 s; 72°C, 30 s), 32-35 cycles; neurotensin receptor (94°C, 1 min; 57°C, 1 min; 72°C, 2 min), 32-35 cycles; GAPDH (95°C, 30 s; 55°C, 45 s; 72°C, 30 min), 30-32 cycles. In each case, a single extension cycle (72°C, 7 min) was added after cycling. Amplification products were identified by both size- and gene-specific hybridization.

Southern analysis of RT-PCR products. Aliquots of PCR products (10 µl) were electrophoresed on a 1.5% agarose gel. DNA transfer to Zeta-probe membrane (Bio-Rad) was carried out using a vacuum apparatus (Pharmacia Biotech, Piscataway, NJ). Gels were depurinated with 0.2 N HCl for 30 min and transferred in 1 N NaOH for 60 min. The membrane was rinsed briefly in 2× SSC (1× SSC is 0.15 M NaCl and 0.015 M sodium citrate, pH 7.0), air dried, and the DNA immobilized by either baking at 80°C for 1 h or ultraviolet cross-linking. Blots were prehybridized for 1 h at 65°C in 5× Denhart's containing 100 µl/ml salmon sperm DNA. Blots were then hybridized at 45°C for 2 h in a solution containing 1× Denhart's, 300 mM NaCl, 60 mM Tris, pH 8.0, 1% sodium dodecyl sulfate (SDS), 2 mM EDTA, 100 µg/ml salmon sperm DNA with 1 × 10⁵ counts/min (cpm)/ml ³²P-end-labeled oligonucleotide probe. Blots were washed twice for 5 min at room temperature in solution 1 (300 mM NaCl, 60 mM Tris, pH 8.0, 2 mM EDTA). Blots were exposed to X-ray film, and the autoradiographs were then scanned using a laser densitometer (Molecular Dynamics) to obtain relative levels of the amplification products. Levels of neurotensin and neurotensin receptor mRNA were normalized to GAPDH mRNA.

Statistics. All results are given as means ± SE. Comparisons between groups were made using Student's t-test for unpaired data. P < 0.05 was considered to represent statistical significance.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of CWR on gastric mucosal integrity. Gastric mucosal injury, expressed as the area of epithelial ulceration in square millimeters, is shown in Table 1. After 1 h CWR the area of mucosal injury was 14.43 mm² (P < 0.05) and had increased to 21.71 mm² (P < 0.05) after 2 h CWR compared with 0 mm² in control animals.

Effect of CWR on neurotensin concentration. Neurotensin concentrations in the nucleus accumbens, prefrontal cortex, and hypothalamus after 1 and 2 h of CWR are shown in Table 2. There was a significant (P < 0.05) 41% decrease in neurotensin concentration in the nucleus accumbens after 1 h CWR compared with control animals. After 2 h CWR the neurotensin concentration had returned toward control values in the nucleus accumbens. In contrast, neurotensin concentrations did not change in the prefrontal cortex or hypothalamus during 1 or 2 h of CWR. In control animals the concentration of neurotensin was highest in the hypothalamus and lowest in the prefrontal cortex, with the nucleus accumbens concentration falling between that in the hypothalamus and prefrontal cortex. These concentrations are relatively similar to the neurotensin concentrations for these regions reported by others (8, 9).

Effect of CWR on neurotensin mRNA. The levels of neurotensin and neurotensin receptor mRNA were measured to determine whether the changes in neurotensin concentration and neurotensin binding (see below) in the nucleus accumbens during CWR resulted from changes of expression of the neurotensin and neurotensin receptor genes. Total RNA was isolated from nucleus accumbens, ventral tegmental area, prefrontal cortex, and hypothalamus. Because there were small amounts of tissue available for study, RT-PCR was used to examine the relative levels of neurotensin and neurotensin receptor mRNAs in different brain regions under the various experimental conditions. The cycling conditions for each primer pair described in MATERIALS AND METHODS produced a single amplification product that was visible in both ethidium bromide-stained agarose gels and on autoradiograms after transfer and gene-specific hybridization. Figure 1 shows autoradiograms of neurotensin, neurotensin receptor, and GAPDH fragments amplified from cDNA reverse transcribed from total RNA isolated from the nucleus accumbens of three groups of three rats that were subjected to 0, 1, or 2 h CWR in one representative experiment. These and similar autoradiograms of RT-PCR products from the nucleus accumbens and other brain regions were scanned to provide the data on the relative levels of neurotensin and neurotensin receptor mRNA that are presented in Fig. 2. At least five more cycles of amplification were required from cDNA reverse transcribed from prefrontal cortex RNA to produce levels of neurotensin and neurotensin receptor amplification products comparable to those obtained from nucleus accumbens, ventral tegmental area, and hypothalamus RNA. The increased cycle number reflects the low level of both neurotensin and neurotensin receptor mRNA in the prefrontal cortex compared with the other brain regions studied (8, 9). For the housekeeping gene GAPDH the same number of cycles produced comparable amounts of product from the four regions studied. Neurotensin and neurotensin receptor mRNA levels were normalized to GAPDH mRNA.

View larger version (48K): [in this window] [in a new window]   Fig. 1.   Southern hybridization of reverse transcription-polymerase chain reaction (RT-PCR) products derived from total RNA isolated from the nucleus accumbens of 3 groups of 3 rats subjected to 0, 1, or 2 h cold water restraint (CWR) in 1 experiment. RNA was reverse transcribed and amplified with gene-specific oligonucleotide primers. NT, neurotensin; NTR, neurotensin receptor; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. Sizes of the amplification products are indicated in base pairs. These and similar autoradiograms were scanned to generate the data shown in Fig. 2.

View larger version (49K): [in this window] [in a new window]   Fig. 2.   A: relative concentration of neurotensin mRNA in various brain regions. RNA isolated from rat brain regions was reverse transcribed and amplified as described in MATERIALS AND METHODS. Amplification products were transferred to membrane, hybridized with a neurotensin-specific probe, and the resultant autoradiograms subjected to laser densitometry. * P < 0.01, compared with CWR 0 h. B: relative concentration of neurotensin receptor mRNA in various brain regions. Sample preparation was as described for A but using a neurotensin receptor-specific probe for hybridization. * P < 0.05, compared with CWR 0 h. In these graphs comparisons of relative levels of neurotensin mRNA and neurotensin receptor mRNA during CWR are made within brain region, not between brain regions. Samples were normalized to GAPDH mRNA concentration. NACB, nucleus accumbens; VTA, ventral tegmental area; PFC, prefrontal cortex; HP, hypothalamus. Number of animals in each group is shown in the corresponding column.

The mean level of neurotensin mRNA decreased very significantly to 54% of the mean control value (P < 0.001) in the nucleus accumbens during 1 h CWR and returned toward control levels after 2 h CWR (Fig. 2A), in parallel with the changes in neurotensin concentration that occurred under the same conditions. A large, although not statistically significant, decrease in neurotensin mRNA concentration in the prefrontal cortex was observed after 1 h CWR. The significance of this decrease is unknown. Neurotensin mRNA concentrations were unchanged in the ventral tegmental area and hypothalamus.

Effect of CWR on neurotensin binding to brain membranes. [³H]neurotensin bound to whole brain membranes in a specific and saturable manner (Fig. 3A). Scatchard plot analysis of [³H]neurotensin binding showed a single set of sites with a dissociation constant (K_d) of 3.28 nM (Fig. 3B). This value is similar to that previously reported for high-affinity neurotensin binding sites (12). Table 3 shows the B_max and IC₅₀ measured in various brain regions during CWR. In displacement experiments, where the unlabeled competing ligand is the same as the labeled ligand, the IC₅₀ is equal to the K_d. The IC₅₀ was not significantly altered by CWR in any of the brain regions studied. In contrast, the number of binding sites (B_max) in the nucleus accumbens increased from control values by 43% after 1 h CWR and significantly (P < 0.05) increased by 76% after 2 h CWR, compared with control values. There was a large but not statistically significant decrease in neurotensin binding capacity in the prefrontal cortex after 2 h CWR. The significance of this decrease is unknown. There was no significant change in the B_max in the hypothalamus.

View larger version (12K): [in this window] [in a new window]   View larger version (10K): [in this window] [in a new window]   Fig. 3.   Saturation isotherm (A) and Scatchard (B) plot of the specific binding of [3H]neurotensin to rat brain membranes. Whole brain membranes were incubated with increasing concentrations of [3H]neurotensin and bound and unbound (B/F) [3H]neurotensin separated as described in MATERIALS AND METHODS. Specific binding was that binding not displaced by 1 µM unlabeled neurotensin. Each point represents the mean of at least 3 separate experiments.

Effect of CWR on neurotensin receptor mRNA. The level of neurotensin receptor mRNA significantly (P < 0.01) decreased in the nucleus accumbens by 51% after 1 h and 50% (P < 0.02) after 2 h of CWR (Fig. 2B). The neurotensin receptor probe used recognizes only the high-affinity neurotensin receptor and would not hybridize to the low-affinity neurotensin receptor recently cloned from mouse brain (23). Both neurotensin and neurotensin receptor mRNA concentrations were unchanged in the ventral tegmental area, prefrontal cortex, and hypothalamus. The neurotensin and neurotensin receptor mRNA concentration varied widely in the prefrontal cortex. This variability probably reflects the limitation of semiquantitative RT-PCR for detection of very low level messages, as is the case for neurotensin and neurotensin receptor mRNA in the prefrontal cortex.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

It has been shown that administration of neurotensin into cerebrospinal fluid of rats or directly into the nucleus accumbens or ventral tegmental area protects the gastric mucosa against injury produced by CWR by reducing gastric acid secretion, enhancing endogenous prostaglandin biosynthesis, and maintaining mucosal blood flow (16, 25, 37). Pretreatment with the dopamine receptor antagonist haloperidol blocks the gastroprotective effect of intranucleus accumbens neurotensin (34). The nucleus accumbens is a terminal field for the mesolimbic dopaminergic fibers from the ventral tegmental area. These observations led to the hypotheses that endogenous neurotensin neurotransmission is reduced in the mesolimbic system during CWR and that the decrease in neurotensin will produce a compensatory increase in neurotensin binding capacity or in the affinity of neurotensin for its receptor within the mesolimbic system. Therefore the effect of CWR on neurotensin concentration and neurotensin binding in the nucleus accumbens, a key region of the mesolimbic system, was determined, and the effect of CWR on the levels of neurotensin and neurotensin receptor mRNA was examined to ascertain whether changes in neurotensin concentration and binding result from changes in neurotensin and neurotensin receptor gene expression. The prefrontal cortex, a terminal area of the mesocortical dopamine system that contains relatively little neurotensin, and the hypothalamus, which contains the highest neurotensin concentration in the brain, were used as controls (8, 9). The interpretation of data relating to neurotensin and neurotensin receptor mRNA differs because the fates of their translation products are not the same. Translatable mRNA resides in the cell bodies of neurons where protein synthesis occurs. Neurotensin is synthesized from its mRNA in the cell body and is then rapidly transported to the axon terminal where it is stored in vesicles before release into the synaptic cleft. The result is that neurotensin measured in a specific brain region was synthesized in distant cell bodies and transported to that region, whereas neurotensin synthesized in the specific region is quickly transported out to other terminal fields. In contrast, neurotensin receptors are inserted into the membranes of the cell body, dendrites, and axon of the synthesizing neuron. Therefore, neurotensin receptors detected in a specific brain region reflect, at least in part, synthesis from neurotensin receptor mRNA in that brain region.

CWR resulted in a 41% decline in neurotensin concentration in the nucleus accumbens after 1 h, and after 2 h neurotensin concentration had returned toward control levels. Neurotensin mRNA concentration in the nucleus accumbens decreased proportionately during 1 h CWR and returned toward the control value after 2 h of stress but did not change in the other brain regions examined. These observations suggest that CWR specifically reduces the expression of the neurotensin gene in the nucleus accumbens. Axon terminals from neurotensin-synthesizing neurons of the A10 cell group of the ventral tegmental area that project to the nucleus accumbens also contain neurotensin in storage vesicles. Therefore, to determine whether changes in neurotensin gene expression in the ventral tegmental area contributed to the decreased neurotensin concentration in the nucleus accumbens, the effect of CWR on neurotensin mRNA concentration in the ventral tegmental area was also examined. CWR did not affect the level of neurotensin mRNA in this region. An additional mechanism that could explain the decreased neurotensin concentration is increased neurotensin degradation in the nucleus accumbens. Neurotensin secretory vesicles undergo regulated exocytosis from axon terminals into the synapse, and the secreted neurotensin is rapidly catabolized by membrane-bound peptidases (2, 6). The decreased neurotensin concentration in the nucleus accumbens after 1 h of stress may reflect acute stress-induced release of neurotensin from terminals in the nucleus accumbens followed by rapid degradation; the recovery after 2 h may reflect replenishment of depleted neurotensin stores by continued synthesis in the ventral tegmental area, where neurotensin mRNA levels were unchanged, without further neurotensin release. Recently Wagstaff and colleagues (33) showed that decreased and increased nucleus accumbens neurotensin content caused by systemic treatment with a dopamine D2 receptor agonist and antagonist, respectively, was associated with increased and decreased extracellular neurotensin concentration, indicating altered neurotensin secretion. A similar observation that tissue neurotensin was increased by dopamine D2 receptor stimulation and decreased by dopamine D2 blockade in the striatum and nucleus accumbens as rapidly as 1 h after a single intraperitoneal dose of agonist or antagonist suggested that the changes in tissue neurotensin probably reflected altered neurotensin release (30). The decreased concentration of neurotensin mRNA in the nucleus accumbens would result in reduced neurotensin concentration "downstream" in its terminal fields, unless CWR also affects the translation of neurotensin mRNA. The nucleus accumbens is a region where complex interactions occur between motor and visceroendocrine systems and nucleus accumbens neurons project to many different brain regions. Specific regions to which the neurotensin neurons project have not yet been identified. A previous study of the effect of stress on gastric mucosal injury found that neurotensin concentration was increased in the ventral tegmental area and unchanged in the nucleus accumbens and other brain regions (9). These results differ from those reported here; the model of stress used, however, footshock over 20 min, is very different from the CWR model used in the present study, which may explain these different observations.

To evaluate the functional significance of decreased neurotensin concentration in the nucleus accumbens during CWR, specific [³H]neurotensin binding to nucleus accumbens membranes from CWR-exposed rats was measured. Neurotensin binding reflects the presence of specific neurotensin receptors. The number of neurotensin binding sites (B_max) increased in the nucleus accumbens during CWR without change in their neurotensin-binding affinity (IC₅₀). In contrast, CWR decreased the concentration of neurotensin receptor mRNA in this region. One interpretation of the data reported here that the number of neurotensin receptors increased despite decreased neurotensin receptor gene expression, is that an intracellular pool of neurotensin receptors exists that can be rapidly translocated to the cell surface in response to appropriate stimuli. It has been established that neurotensin receptors can be downregulated after exposure to neurotensin by a mechanism involving internalization of neurotensin-neurotensin receptor complexes within endosome-like organelles (32). Recruitment or upregulation of integral membrane proteins to the cell membrane from intracellular stores has been described in a variety of systems. These include rapid upregulation of glucose transporters in muscle in response to insulin (15), translocation of the adhesion molecule _IIb3 integrin to the cell membrane of protein kinase C-stimulated metastatic melanoma cells (31), and exocytic upregulation of collagen receptors in HeLa cells during substrate adhesion and spreading (22). Of particular pertinence to the present study are the observations of Yamada and colleagues (36) who reported rapid downregulation of neurotensin receptors in murine neuroblastoma clone N1E-115 cells in response to neurotensin. Removal of excess neurotensin by extensive washing followed by 37°C incubation resulted in recovery of neurotensin binding sites on the cell surface within 20 min. Neurotensin receptor mRNA level was not measured in this study, but the rapid time course of neurotensin receptor upregulation is suggestive of recruitment from intracellular sites. We hypothesize that rats exposed to CWR have a rapid release of neurotensin stored in the nucleus accumbens followed by its degradation in the synaptic cleft, quickly resulting in decreased extracellular neurotensin concentration. Neurotensin receptors upregulate in response to the decreased neurotensin concentration. Replenishment of nucleus accumbens neurotensin occurs after 2 h, although relatively slowly because de novo synthesis of neurotensin in distant sites followed by axonal transport to the nucleus accumbens is required. The continued increase in neurotensin binding sites after 2 h suggests that the extracellular neurotensin concentration in the synaptic cleft remains low despite the apparent replenishment of tissue neurotensin. The half-life of the neurotensin receptor is not known. The data presented in this manuscript suggest that the half-life is at least several hours because the decreased neurotensin receptor mRNA concentration is not reflected in decreased neurotensin binding in the 2-h period over which these experiments were conducted. Recruitment of existing intracellular neurotensin receptors to the cell surface may have overridden the effect of depressed neurotensin receptor gene expression. An alternative explanation for the data that cannot be ruled out at this time is that "inactive" neurotensin receptors are in some way activated during CWR. Rapid decrease of neurotensin receptor mRNA concentration in rat colon in response to stress has also been reported recently (5). In that study, in situ hybridization of rat colon tissue showed ~40% decrease in neurotensin receptor mRNA labeling intensity after 30 min of immobilization stress. A report has appeared in which increased neurotensin binding was accompanied by increased neurotensin receptor mRNA concentration (1). Intraperitoneal injection of the nonpeptide neurotensin receptor antagonist SR-48692 increased the number of neurotensin binding sites in whole brain homogenates and also increased neurotensin receptor mRNA level in rat substantia nigra and, to a lesser extent, in the ventral tegmental area (1, 35). These studies employed chronic treatment with SR-48692 over a period of 1 or 2 wk, however, a very different scenario from the rapid upregulation of binding sites reported here. There was no change in neurotensin receptor mRNA after acute treatment with SR-48692 in these studies (35).

In summary, these observations indicate that CWR causes a decline in neurotensin concentration in the nucleus accumbens that may be due to decreased neurotensin gene expression and/or increased neurotensin release and degradation. Reduced endogenous neurotensin concentration is associated with rapid upregulation of neurotensin receptors, perhaps by translocation of neurotensin receptors from intracellular sites to the cell surface. These findings support the hypothesis that reduced neurotensin concentration in the nucleus accumbens or perhaps the entire mesolimbic system, in vivo, contributes to stress-induced gastric mucosal injury and provides a basis for the protective effect of centrally administered neurotensin. A number of important questions remain to be addressed, such as how rapidly after exposure to CWR does neurotensin concentration in the nucleus accumbens decrease, what effector is produced or activated in response to CWR that causes neurotensin mRNA and neurotensin receptor mRNA concentration to decrease, and what is the final step in the chain of events initiated by decreased mesolimbic neurotensin concentration in response to CWR that results in gastric mucosal injury? CWR-induced stress provides a useful model in which to study the link(s) between events occurring in the brain and in the periphery. It would be of interest to determine whether these observed biochemical changes in the nucleus accumbens also occur in other experimental conditions of stress.

Perspectives