INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE INSULAR CORTEX (IC) receives inputs from multiple sensory systems, including taste, and from receptors that monitor body fluid and electrolyte balance and blood pressure (6, 8, 18, 26, 27, 36, 37). This input is relayed though a network involving the nucleus tractus solitarius (NTS), parabrachial complex, and ventromedial thalamus (1, 27). The outputs of the IC are directed to the amygdala, nucleus accumbens, hypothalamus, and pontine and medullary structures (29).

Behavioral studies implicate the IC in sodium taste aversion (4, 5, 9, 15, 28). Lesions of IC result in increased ingestion of sodium in sodium preference tests, possibly by increasing the palatability of salt solutions in animals with damage to the IC (2). However, the specific role of IC in the palatability of sodium ingestion is not established. Analysis of brain c-fos expression after sodium depletion and the subsequent ingestion of sodium has identified several brain areas that likely serve as the neural network mediating salt appetite behavior (10, 12, 21, 24, 33, 34). These areas include structures of the lamina terminalis [subfornical organ (SFO), organum vasculosum of the lamina terminalis (OVLT), median preoptic nucleus], the hypothalamus (supraoptic nucleus, paraventricular nucleus), the extended amygdala, and several midbrain and hindbrain areas (lateral parabrachial nucleus, area postrema, NTS). The IC is directly connected with several of these areas and may represent a key cortical component in salt appetite. The present work uses immunohistochemical detection of Fos immunoreactivity (Fos-ir) to examine activity in the cortical components of the neural network subserving sodium appetite after sodium depletion in animals that have fluids available for rehydration and in animals that do not. Furthermore, the distribution of c-fos expression in insular subregions was examined. The IC is divisible into regions along the anterior-posterior axis and into different zones distinguished by the density of granular cells in layers II and IV. The anterior IC, which has granular, dysgranular, and agranular zones, projects to motor cortex and the nucleus accumbens (29). Caudal to this are two sensory-related insular areas, the posterior and parietal IC, both with granular, dysgranular, and agranular zones. The posterior IC receives input from the gustatory thalamus and parabrachial complex and projects to the anterior IC and the extended amygdala (29, 30). Their different connections suggest potentially different functional roles for each subregion in salt appetite.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Male Sprague-Dawley derived rats weighing 280-300 g were purchased from Harlan (Indianapolis, IN). They were housed individually in hanging wire cages for at least 1 wk before experimentation. Purina Rat Chow, tap water, and 0.3 M NaCl were available ad libitum except during the experimental periods. Room lights were on for 12 h/day, and temperature was controlled at 23°C.

Sodium depletion. Sodium depletion was accomplished according to previously published protocols from our laboratory (31-33). Rats were weighed and injected with the diuretic furosemide (10 mg/kg body wt sc; Abbott Laboratories, North Chicago, IL) to produce diuresis and natriuresis. About 5 min later, the animals were injected with the angiotensin-converting enzyme (ACE) inhibitor captopril (5 mg/kg body wt sc; SQ 14,225, Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ) to partially block ACE within the systemic circulation. This procedure results in volume contraction and small reductions in arterial blood pressure that stimulate renin secretion and formation of ANG II in the brain, thereby rapidly stimulating water and sodium ingestion (31, 32).

Treatment. The rats were tested in the home cage. Access to chow was blocked, and water bottles and saline tubes were removed. One group of animals (n = 4) was depleted of body water and sodium as outlined above. Starting 90 min later, they were provided with water and a 0.3 M NaCl solution from glass burets with sipper tubes. Fluid was available for 2 h. Another group of animals (n = 4) was similarly depleted but did not receive access to water and saline for drinking in order to control for the effects of rehydration. A third group (n = 4) had access to water and saline for 2 h but was not depleted of water and sodium in order to control for the effects of fluid depletion. Instead, this group received two subcutaneous injections of 0.9% NaCl spaced 5 min apart and had access to water and saline starting 90 min later. At the end of testing, rats were injected with an overdose of pentobarbital sodium (100 mg/kg ip; Nembutal, Abbott Laboratories, North Chicago, IL) and perfused transcardially for subsequent immunohistochemical detection of Fos.

Animal perfusion and histology. Staining for c-fos activation was performed after the completion of the experimental procedures. The rats were anesthetized with pentobarbital sodium (100 mg/kg ip) and perfused transcardially with ~100 ml of normal saline followed by 300 ml of paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.2). The brains were removed, fixed in the same solution overnight, and then stored in PB containing 20% sucrose, at 4°C. Coronal sections (40 µm) were cut using a freezing microtome.

Immunohistochemistry. Free-floating sections were placed in a solution of 5% normal goat serum in PB for 1 h to reduce the background. Then the sections were incubated overnight at room temperature with a Fos antibody raised in rabbits (Santa Cruz; 1:2,000 in PB). The primary antibody was then reacted with a biotin-labeled anti-rabbit immunoglobulin (Vector; 1:200 in PB) and finally with avidin-biotin peroxidase complex (Vector Elite Kit; 1:200 in PB). The peroxidase label was detected using 3,3'-diaminobenzidine hydrochloride (dissolved in PB with 0.02% hydrogen peroxide) as chromogen. After that the tissue sections were mounted on gelatin-coated slides, air dried, dehydrated, cleared in xylene, and placed under coverslips with DePeX mounting medium (BDH Laboratory Supplies, Poole, UK). Cells were considered Fos positive if they displayed a brown reaction product in the nucleus.

Quantitative analysis. Sections were digitized with an Optronics CCD video camera system in auto exposure mode connected to a Leitz Aristoplan microscope using a ×10 objective lens. Photomicrographs were produced by capturing the image and editing with software (Adobe Photoshop). The images were not retouched, and only the contrast and brightness were adjusted digitally. Quantification of the number of Fos-ir cells in the IC was obtained under constant illumination, in selected areas of sections, using the Image (NIH) program. Densely labeled Fos-ir cells were visualized in the computer monitor and were discriminated using the contrast threshold function of the program. For purposes of comparison, we counted the number of Fos-ir cells in structures of the lamina terminalis (i.e., SFO and OVLT) in the same animals. Representative sections containing the greatest coronal cross-sectional areas of the SFO and OVLT for each animal were examined under the microscope, and the number of Fos-ir cells was counted by hand.

Data analysis. The data were analyzed using two-way or three-way ANOVA. Planned comparisons were made with Fisher's least significant difference (LSD) tests when the global F ratio was significant at the P < 0.05 level. The data are expressed as means ± SE. The number of Fos-ir-positive nuclei in each subregion of IC was determined by counting Fos-ir-positive nuclei in coronal sections through IC. Six sections were used for determination of Fos-ir-positive nuclei in each subregion, the section level relative to bregma being determined by comparison with sections in a standard atlas (25). The range of levels (~2.7 mm in front of bregma to 2.7 mm behind) sampled the anterior IC (anterior to 1 mm in front of bregma), and the posterior IC (1 mm in front of bregma to 2 mm behind). The gustatory cortical representation is located between bregma and 1 mm in front. Treatment was the between-subjects factor, and subregion and bregma level were the within-subjects factors. The number of Fos-ir-positive nuclei in the lamina terminalis was determined from one representative section for each area (SFO or OVLT). Treatment was the between-subjects factor, and area was the within-subjects factor.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Behavior. Animals that were not depleted of body fluids drank no water or concentrated saline solution during the 2 h of fluid access (Table 1). Animals depleted of body fluids and with ad libitum access to drinking fluids demonstrated robust thirst and salt appetite responses by drinking significant amounts of water and saline in 2 h.

Global analysis of c-fos in IC. There were significant main effects of treatment [F(2,9) = 542.60; P < 0.05], subregion [F(2,18) = 29.45; P < 0.05], bregma level [F(5,45) = 87.35; P < 0.05], and a significant treatment × subregion × bregma level interaction [F(20,90) = 5.81; P < 0.05]. Levels of Fos-ir were lowest in each subregion of IC in nondepleted (i.e., fluid replete) animals (Fig. 1). Levels of Fos-ir were significantly increased in all subregions in depleted rats that were not allowed to drink. Levels of Fos-ir were further significantly increased in all subregions in depleted rats that drank significant amounts of water and saline. Thus levels of Fos-ir were lowest in rats that presumably were water and sodium replete and in normal water and sodium balance, were higher in depleted rats, and were highest in depleted rats that ingested fluids. The agranular subregion had the most Fos-ir-positive nuclei, the dysgranular subregion had significantly fewer Fos-ir-positive nuclei, and the granular cortex had the least number of Fos-ir-positive nuclei. In all subregions of IC, Fos-ir-labeled cells were located mainly in the deep layers of IC, with a few Fos-ir-positive cells located in the superficial layers.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Average number of Fos-immunoreactive (ir)-positive nuclei from all sections taken through the agranular, dysgranular, and granular subregions of insular cortex for fluid-replete rats with fluid access, fluid-depleted rats without fluid access, and fluid-depleted rats with fluid access. Values are means ± SE. * Significantly different from fluid-replete rats with fluid access; ** significantly different from both fluid-replete rats with fluid access and fluid-depleted rats without fluid access.

Analysis of subregions. In all subregions, fluid-replete rats tended to have low levels of Fos-ir in the most anterior and posterior sections, with significantly increased levels in at least one of the middle sections.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Number of Fos-ir-positive nuclei at different levels from bregma in the granular subregion of insular cortex for fluid-replete rats with fluid access, fluid-depleted rats without fluid access, and fluid-depleted rats with fluid access. Values are means ± SE; + Significantly different from fluid-replete rats with fluid access; * significantly different from both fluid-replete rats with fluid access and fluid-depleted rats without fluid access.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Number of Fos-ir-positive nuclei at different levels from bregma in the dysgranular subregion of insular cortex for fluid-replete rats with fluid access, fluid-depleted rats without fluid access, and fluid-depleted rats with fluid access. Values are means ± SE. + Significantly different from fluid-replete rats with fluid access; * significantly different from both fluid-replete rats with fluid access and fluid-depleted rats without fluid access.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Number of Fos-ir-positive nuclei at different levels from bregma in the agranular subregion of insular cortex for fluid-replete rats with fluid access, fluid-depleted rats without fluid access, and fluid-depleted rats with fluid access. Values are means ± SE. + Significantly different from fluid-replete rats with fluid access; * significantly different from both fluid-replete rats with fluid access and fluid-depleted rats without fluid access.

View larger version (115K): [in this window] [in a new window]   Fig. 5.   Photomicrographs of coronal sections of rat insular cortex showing the pattern of Fos-ir at 2 bregma levels (1.7 mm anterior and 2.6 mm posterior to bregma, respectively) after the experimental treatments. A and B: fluid-replete rats with fluid access. C and D: fluid-depleted rats with 2-h fluid access. E and F: fluid-depleted rats without fluid access. Calibration bar, 100 µm. AI, agranular insular cortex; DI, dysgranular insular cortex; GI, granular insular cortex.

Analysis of c-fos in SFO and OVLT. There was a significant main effect of treatment [F(2,9) = 31.64, P < 0.05]. Fluid-replete rats (i.e., sham-depletion condition) had minimal levels of Fos-ir in either the SFO or the OVLT (Fig. 6). Rats depleted of fluids had significantly increased levels of Fos-ir in both SFO and OVLT, and there were no differences in Fos-ir levels in these nuclei between those rats without access to fluids and those with access to fluids. There was no main effect of brain area [i.e., SFO vs. OVLT, F(1, 9) = 0.05; P > 0.05] and no significant interaction effect [i.e., brain area × treatment, F(2,9) = 0.13; P > 0.05].

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Number of Fos-ir-positive nuclei in the subfornical organ (SFO) and organum vasculosum of the lamina terminalis (OVLT) for fluid-replete rats, fluid-deleted rats without fluid access, and fluid-depleted rats with fluid access. Values are means SE. * Significantly different from fluid-replete rats with fluid access.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study reveals significant differences in the pattern of c-fos expression of cells across regions of the IC after acute depletion of body water and sodium and the subsequent ingestion of fluids. The data show that the pattern of Fos staining in the three subregions of IC differs according to hydration level and fluid ingestion. Fluid-replete animals had weak-to-moderate levels of Fos-ir in the three subregions of IC differing in granule cell density. Animals depleted of body fluids, but without access to fluids for rehydration, had significantly elevated levels of Fos-ir in all subregions of IC. This elevated level of cellular activity likely reflects visceral activity related to sensing aspects of hypovolemia and/or hypotension (22, 31). Animals that drank water and saline after fluid depletion had significantly increased levels of Fos-ir throughout the IC compared with both fluid-replete control animals and depleted animals not allowed to rehydrate with fluids after sodium depletion. This additional increased Fos-ir likely reflects consequences of ingesting water and sodium after dehydration, possibly including the taste of the ingested fluids and postingestional factors related to the volume and tonicity of the ingested fluids. Last, there was greater Fos-ir activity in rostral sections compared with caudal sections of all subregions of IC. The greater cellular activity in the anterior sections may indicate a greater role for the rostral areas in consummatory responses to dehydration and fluid ingestion after sodium depletion-induced appetite, at least at the time point sampled in this experiment.

Under conditions of body fluid deficits, multiple inputs derived from humoral and neural receptors converge in key areas of the brain where the information is integrated. The present work extends findings from previous studies (10, 12, 20, 21, 23, 24, 33, 34) using c-fos expression as a marker of neuronal activity during sodium depletion and the induction of sodium appetite. These previous studies examined changes in the number of Fos-ir cells of structures such as the lamina terminalis, area postrema, raphe nuclei, lateral parabrachial nucleus, and extended amygdala. These structures are components of a neural network that participates in generating behaviors that maintain body fluid and cardiovascular homeostasis. By examining changes in the pattern of c-fos activation of cells in the IC, the present work adds cortical elements to these pathways subserving body fluid and cardiovascular homeostasis.

Levels of c-fos in the SFO and OVLT were examined to provide a comparison of the effects of the experimental treatments on c-fos expression in other brain areas. As was observed in IC, fluid-replete rats had low levels of c-fos in the SFO and OVLT. Similarly, sodium depletion dramatically increased c-fos levels in the SFO and OVLT. These two observations agree well with those of previous work (33) in which low levels of c-fos were obtained in the SFO and OVLT of sham-depleted animals, and much higher levels of c-fos were observed in these structures on fluid depletion. Unlike the observations in IC, however, fluid ingestion did not cause further increases in c-fos levels in the SFO and OVLT. Rather, c-fos levels in the SFO and OVLT of depleted rats that drank fluids were not statistically different from those of depleted rats not allowed access to fluids.

The water and sodium ingestion induced by the experimental procedure employed here are dependent on the actions of the endocrine renin-angiotensin system. Fluid depletion induces increased Fos-ir-positive cells in the SFO and OVLT, the area postrema, the NTS, and lateral parabrachial nucleus (33). The increase in the OVLT and SFO during fluid depletion is prevented by pharmacological blockade of the renin-angiotensin system (24, 33). Furthermore, two studies have shown that the increased number of Fos-ir cells in the SFO (20, 34) and OVLT (34) induced by sodium depletion is reversed on ingestion of hypertonic saline. We did not observe decreased Fos-ir in the SFO or OVLT after fluid ingestion in our depleted rats. There are notable differences between the two previous studies and our present study. Both previous studies involved overnight depletion protocols, while ours was acute. Fos-ir levels were ascertained at different times after fluid ingestion in the studies. The previous studies found significant reductions in Fos-ir at 3 h (34) and 12 h (20) h after the start of ingestion. However, we looked at 2 h after the start of ingestion and failed to find a reduction in Fos-ir. There may not have been enough time in our study for the effects of fluid ingestion to manifest changes in Fos-ir levels in the SFO and OVLT.

Rather than being decreased, Fos-ir levels in IC were increased on ingestion of water and sodium by sodium-depleted animals. Increased c-fos activity in IC after water and sodium intake may be the result of convergent inputs from taste receptors, blood pressure/volume receptors, and chemoreceptors (16, 18, 19). It is known that afferent activity due to oropharyngeal stimulation induces responses, such as taste sensation, orofacial reflex, mastication, digestion, swallowing, respiration, digestive secretion (salivary, gastric, and pancreatic juices), insulin secretion, and cardiovascular responses (17). Some combination of these stimuli could act together during the ingestion of sodium solution to induce the observed increases in the c-fos expression of the IC.

The greatest activation of IC occurred in the posterior IC after water and sodium ingestion by dehydrated rats. This area includes the cortex along the rhinal sulcus running from ~2 mm anterior to bregma to 2 mm posterior to bregma (i.e., the caudal end of the claustrum) and consisting of granular, dysgranular, and agranular regions (7, 29, 38). Fluid depletion accompanied by water and sodium ingestion increases c-fos activation in all subregions, particularly in anterior sections of the granular and dysgranular areas. These contain the gustatory and viscerosensory cortical areas. Therefore, c-fos expression observed after ingestion of water and sodium could be due to the taste properties of these fluids. Furthermore, the act of drinking involves activation of mechanoreceptors, thermoreceptors, and chemoreceptors located in the oropharyngeal regions that are excited during tongue movements and swallowing (35). These sense modalities could also alter neuronal activity in the IC during fluid ingestion.

Postingestive factors, including oral-pharyngeal-esophageal and gastric cues, have been shown to induce c-fos immunoreactivity in neural regions such as the dorsal nucleus of the vagus, NTS, and the area postrema (13, 14). It is possible that cues related to gastric filling and changes in osmolality due to tonicity of the ingested fluid were responsible for the increased c-fos expression in IC on fluid ingestion.

The anterior portion of the IC is the main target of cortical projections from the posterior IC (29). The anterior IC projects to jaw and face regions of the motor cortex and to the nucleus accumbens (3, 30). Increased numbers of Fos-ir cells were seen in this region (level 2.7 in Figs. 3-4) in depleted animals allowed access to fluids. The increase of c-fos expression in the premotor interface could be associated with the generation of stereotypical orofacial and forelimb movements (11).

In summary, this study showed different patterns of cellular activation within subdivisions of the IC after fluid depletion and after subsequent ingestion of water and saline solution. The presence of Fos-ir cells in the IC after water and sodium depletion and water and sodium ingestion provides evidence that this portion of the cortex is involved in regulation of body fluid homeostasis. The increased cortical Fos-ir in fluid-depleted animals likely reflects stimulation by afferent visceral information related to hypovolemia. The additional increased Fos-ir of fluid-depleted animals that subsequently ingested water and sodium implicates both taste and/or postingestional factors. Furthermore, the functional-morphological results of this study suggest that the cells activated during sodium depletion and sodium intake constitute an independent neuronal group along the IC. These cells respond differently in relation to the physiological sodium state of the rats.