|Home | About | Journals | Submit | Contact Us | Français|
Angiotensin II (AngII) stimulation of water and NaCl intake is a classic model of the behavioural effects of hormones. In vitro studies indicate that the AngII type 1 (AT1) receptor stimulates intracellular pathways that include PKC and MAP kinase activation. Previous studies support the hypotheses that PKC is involved in AngII-induced water, but not NaCl intake and that MAP kinase plays a role in NaCl consumption, but not water intake, after injection of AngII. The present experiments test these hypotheses using central injections of AngII in the presence or absence of a PKC inhibitor or a MAP kinase inhibitor. Pre-treatment with the PKC inhibitor chelerythrine attenuated AngII-induced water intake, but NaCl intake was unaffected. In contrast, pre-treatment with U0126, a MAP kinase inhibitor, had no effect on AngII-induced water intake, but attenuated NaCl intake. These data support the working hypotheses and significantly extend our earlier findings and those of others. Perhaps more importantly, these experiments demonstrate the remarkable diversity of peptide receptor systems and add support for the surprising finding that intracellular signalling pathways can have divergent behavioural relevance.
Determining specific behavioural consequences of intracellular signalling pathways is an important problem in behavioural neuroscience. A degree of progress has been made for a number of behavioural phenomena including food intake, learning and memory, drug addiction, and reproductive behaviours (Diamond & Gordon, 1997; Abel & Lattal, 2001; Etgen et al., 2001; Bjorbaek & Kahn, 2004). Angiotensin II (AngII) stimulation of water intake has been a powerful model for measuring the effects of hormones on behaviour, but the intracellular signalling pathways involved have received only minimal attention. In addition to the reliable increases in water intake, AngII stimulates consumption of concentrated NaCl solution, although the amount tends to vary between experiments and is far more sensitive to a number of variables including the concentration of NaCl provided and activity of other hormone systems [for review see (Daniels & Fluharty, 2004)].
In vitro studies have provided a wealth of information about the intracellular signalling pathways engaged by AngII receptors. The AngII type 1 (AT1) receptor, which is largely responsible for the behavioural effects of AngII (Fluharty, 2002), is a prototypical G protein-coupled receptor that acts through Gq to generate IP3 and activated PKC (de Gasparo et al., 2000). In addition to this more traditionally described pathway, AT1 receptor activation results in phosphorylation of MAP kinase family members (de Gasparo et al., 2000) and studies using receptor mutagenesis demonstrated that this activation can occur without Gq coupling (Hines et al., 2003). Additional in vitro experiments using a double isoleucine-substituted AngII analogue (Sar1, Ile4, Ile8-AngII; abbreviated as SII) further demonstrated the separation of these pathways. Specifically, AT1 transfected cells treated with SII had increased levels of phosphorylated p44/42 MAP kinase without any increases in IP3 levels (Holloway et al., 2002; Wei et al., 2003; Miura et al., 2004; Daniels et al., 2005). Thus, it is becoming clear that the activation of G protein-mediated pathways and MAP kinase family members can occur in parallel, rather than in series.
In spite of the wealth of data from in vitro experiments, far less is known about the role of these signalling pathways in behaviours stimulated by AngII. Nevertheless, some progress has been made in this direction. Fleegal and Sumners (2003) demonstrated that AngII activates PKC in rat brain and that PKC activation contributes to AngII-induced water intake. The data described are compelling with respect to water intake, but the experiments performed were not designed to examine a potential role of PKC in AngII-induced NaCl intake. More recently, we have showed that AngII and SII, which each stimulate MAP kinase in vitro, increased activated MAP kinase in rat brain (Daniels et al., 2005). In addition to showing that the effects of these peptides in brain were consistent with the in vitro models, the experiments showed that SII antagonized AngII-induced water intake, but stimulated NaCl intake similar to that seen by AngII-treated positive control animals (Daniels et al., 2005). Taken together, these findings provided the foundation for the hypotheses proposed previously (Daniels et al., 2007) that the Gq/PLC/IP3/PKC limb of AngII receptor signalling mediates increases in water intake and that activation of MAP kinase stimulates NaCl intake. The present experiments represent the first direct test of these hypotheses.
The handling and care of experimental animals conformed to the regulations provided by the NIH Guide for the Care and Use of Laboratory Animals, and the experimental protocols were approved by the Institutional Animal Care and Use Committee of the University at Buffalo, SUNY. Adult, male Sprague Dawley rats weighing between 225–250 g were obtained from Charles River Laboratories (Wilmington, MA). Animals were single housed in hanging wire cages with food (2018 global rodent diet, Harlan Teklad, Madison, WI) and water available ad libitum in a temperature- and humidity-controlled environment with a 12 h light/dark cycle
Animals were anesthetized using a combination of ketamine (70 mg/kg, i.m.) and xylazine (5 mg/kg, i.m.) before being fixed in a stereotaxic frame and implanted with 26 ga guide cannulae (Plastics One, Roanoke, VA) aimed at the lateral ventricle. The coordinates (0.9 mm caudal to Bregma, 1.4 mm from midline, and 1.8 mm ventral to dura) were chosen to allow an internal injection cannula to extend beyond the guide into the ventricular space. The cannulae were fixed in place with dental cement and bone screws and the animals were allowed to recover before verification procedures were performed.
Proper placement of cannulae and responsiveness to AngII were confirmed using a single injection of 10 ng AngII (Bachem Biosciences, King of Prussia, PA) in 1 μl tris-buffered saline (TBS), delivered through a 33 ga injection cannula attached by water-filled PE tubing to a Hamilton syringe. Multiple lengths of cannulae were used until each animal responded to the injection by consuming at least 5 ml of water within 30 min. Animals that failed to respond after three separate verification attempts, each with an injection cannula of a different length, were excluded from the experiment. Animals were habituated to mock injection procedures and were given access to water and 1.5% NaCl overnight before the onset of each experiment. Each experiment began during the first several hours of the light portion of the light:dark cycle.
Water and 1.5% NaCl were made available in separate 50 ml water bottles, each marked with 1 ml graduations, 24 h before the onset of the experiment. The concentration of saline (1.5%) was selected to maintain consistency with previous experiments (Daniels et al., 2005) and to help ensure a modest salt ingestion following AngII administration. Animals with proper cannula placement were randomly assigned to one of four groups and subsequently injected with 0.1 ng, 1.0 ng, 10 ng, or 100 ng AngII in 1 μl TBS into the lateral ventricle. Immediately after injection, animals were returned to their home cages and water and 1.5% NaCl intakes were recorded at 5, 10, 15, 30, and 60 min.
Graduated bottles containing water or 1.5% NaCl were made available 24 h before the onset of the experiment. Animals were randomly assigned to one of four treatment groups using a traditional 2×2 experimental design in which access to fluids was removed and animals were pre-treated by icv injection of 100 μM chelerythrine (obtained from EMD Chemicals, San Diego, CA and from Biomol International, Plymouth Meeting, PA) or vehicle (2 μl DMSO). Fifteen minutes after the pre-treatment, animals received a second icv injection of AngII (10 ng) or vehicle (1 μl TBS). After the injection of vehicle or AngII, water and 1.5% NaCl were provided in separate bottles and intake of each was measured 5, 10, 15, 30, 60, and 180 min as well as 24 h after the second injection (AngII or vehicle). The experiment was performed twice, each using different animals and chelerythrine obtained from different vendors. Each experiment included 3–4 rats per group. The dose and timing of the chelerythrine application were chosen to correspond best to the experiments reported in Fleegal and Sumners (2003).
The experiment was performed using a similar experimental design as that described for Experiment 2 above, except that a new set of animals was pre-treated with 1 mM U0126 (Promega, Fitchburg, WI) or vehicle (2 μl DMSO) 20 min before injection of AngII (10 ng) or vehicle (1 μl TBS). This experiment was performed once using 3–4 animals per treatment group and the dose and timing of the U0126 injection were based on pilot studies and were within the range of concentration and pre-treatment times used for icv injections in previous studies (e.g., Kuroki et al., 2001; Sutton et al., 2005).
Data were analyzed using one-way or mixed design two-way ANOVA (SPSS version 14.0; SPSS Inc., Chicago, IL) to evaluate between-subjects effects of the drug treatments while accounting for the repeated time-measures for each animal when necessary. Statistical significance was defined as p<0.05 and differences were further analyzed by post hoc testing procedures when necessary as described in the Results section.
In spite of the numerous dose-response and time course analyses of AngII-induced water and NaCl intake, data specific to 1.5% NaCl intake are less common. Given our use of this concentration of saline previously (Daniels et al., 2005) and our desire to compare the present and past results, we determined the dose-response relationship between AngII and the ingestion of water and 1.5% NaCl by treating rats with a single injection of 0.1, 1, 10, or 100 ng AngII (n=6–8 per group). As shown in Figure 1, AngII increased both water and saline intake, with subtle differences between the two fluids. Analysis of the total fluid consumed during the testing period revealed a statistically significant between-subjects effect of AngII dose on water intake (Kruskal-Wallis One Way ANOVA, H=21.44, df=3, p<0.001), NaCl intake (Kruskal-Wallis One Way ANOVA, H=17.58, df=3, p<0.001), and total fluid intake (One Way ANOVA, F3,27=29.0, p<0.001). Differences between animals receiving 0.1 ng and those receiving each other dose were further analyzed using t-tests at individual times with Bonferroni-adjusted critical values. These tests confirmed that animals given 1 ng AngII did not differ from those given 0.1 ng in their intake at any time. Differences between the other doses (10 and 100 ng) and 0.1 ng were detected as early as the earliest measure of water intake and total fluid intake, but intake of NaCl was slightly delayed and differences were not detected until the 10 min measurement.
To determine if PKC plays a role in the water and 1.5% NaCl intake observed after icv injection of AngII, we pre-treated animals with the PKC inhibitor chelerythrine (2 μl of a 100 μM solution) or vehicle (DMSO) 15 min before AngII (10 ng) or vehicle (1 μl TBS). Measurements of water and 1.5% NaCl intake, using a two-bottle test, replicated the findings of Fleegal and Sumners (2003) by showing reduced water intake in the presence of the PKC inhibitor (Fig 2A) and importantly extended these findings by showing no effect of the inhibitor on NaCl intake (Fig 2B). Like the response to AngII in the first experiment, the majority of intake occurred within the early portion of the testing period with negligible additional intake beyond the first hour. Statistical tests on water intake using a mixed-design ANOVA revealed effects of chelerythrine (F1,25=4.986, p=0.035), AngII (F1,25=162.378, p<0.001), and an interaction of chelerythrine and AngII (F1,25=6.028, p=0.021). Post hoc tests using the Tukey method indicate that the difference between vehicle- and chelerythrine-pre-treated groups occurred at the 30 min and later time points. Analysis of the NaCl intake, however, failed to reveal an effect of PKC inhibition. This analysis revealed a main effect of AngII (F1,25=38.74, p<0.001), but failed to find a main effect of chelerythrine (F1,25=0.052, p=0.82) or a significant interaction between chelerythrine and AngII (F1,25=0.18, p=0.68). The differences in water intake were not, however, sufficient to create a detectable difference in total fluid intake (Fig 2C). Specifically, a mixed-design two way ANOVA revealed a significant between-subjects effect of AngII (F1,25=197.2, p<0.001), without an effect of chelerythrine (F1,25=1.96, p=0.174) or an interaction (F1,25=1.96, p=0.174).
To test the role of p44/42 MAP kinase in the fluid intake stimulated by AngII, we used a separate set of rats that also were implanted with lateral ventricular cannulae and screened for proper placement as described above. Rats (n=3–4 per group) were pre-treated with icv injection of vehicle (DMSO) or the MAP kinase inhibitor U0126 (2 μl of a 1 mM solution) 20 min before injection of vehicle (1 μl TBS) or AngII (10 ng). As found in our earlier experiments, the majority of AngII-induced intake occurred within the early times measured, with negligible intake after the first hour (Fig 3). In contrast to the effects of chelerythrine on water intake, rats pre-treated with U0126 did not differ from animals given AngII alone in their water intake (Fig 3A), suggesting that MAP kinase is not required for AngII-stimulated water intake. Analysis of the water intake data confirmed a main effect of AngII (F1,9=36.39, p<0.001), but failed to find a main effect of U0126 (F1,9=1.36, p=0.273) or an interaction effect (F1,9=1.03, p=0.337). Also in contrast to the experiments using chelerythrine, rats pre-treated with U0126 drank markedly less NaCl than rats given AngII alone (Fig 3B). Analysis of these data confirmed significant main effects of U0126 (F1,9=19.95, p=0.002) and AngII (F1,9=81.08, p<0.001), as well as a significant interaction (F1,9=18.06, p=0.002). Post hoc comparisons using the Tukey method revealed that the divergence between vehicle- and U0126-pretreated rats appeared by the first measure (5 min) and remained in each subsequent measure of NaCl intake. In this experiment, the magnitude of the difference in 1.5% NaCl intake was sufficient to produce overall differences in total fluid intake (Fig 3C). These differences were confirmed by a mixed-design two way ANOVA that revealed main effects of U0126 (F1,9=13.2, p=0.005) and AngII (F1,9=343.3, p<0.001), as well as an interaction (F1,9=13.2, p=0.005).
The stimulation of water and NaCl intake by AngII is a classic model of endocrine regulation of behaviour. We used this model to examine the role of intracellular signalling pathways in the observed responses to AngII. Previous studies using in vitro and in vivo preparations provided evidence that AngII receptor-mediated MAP kinase activation can occur without Gq-mediated stimulation of the PLC/IP3/PKC pathway (Hines et al., 2003; Daniels et al., 2005). These studies also showed that treatment with the AngII analogue SII, which activates MAP kinase without stimulating the PLC/IP3/PKC pathway in vitro, led to NaCl intake, but did not produce the water intake typically observed after AngII is administered to laboratory rats. Taken together with data reported in Fleegal and Sumners (2003), we proposed the working hypothesis that AngII-induced water intake involves PKC whereas MAP kinase activation plays a role in AngII-induced NaCl intake (Daniels et al., 2007). The present experiments provide the first direct test of each part of this hypothesis. Accordingly, we confirmed that 10 ng was a moderate, but effective dose of AngII for both water and 1.5% NaCl intake and subsequently used this dose of AngII in the presence or absence of a PKC or MAP kinase inhibitor to stimulate fluid intake. Within the respective experiments, rats pre-treated with the PKC inhibitor chelerythrine drank less water than AngII-injected control animals and rats pre-treated with the MAP kinase inhibitor U0126 drank less 1.5% NaCl than control animals injected with AngII alone. AngII-induced intake of 1.5% NaCl and water were not, however, affected by chelerythrine or U0126, respectively. Thus, the present data strongly support the working model.
Although the data were in agreement with the working hypothesis, some caveats must be considered. As is true for any experiment in which the expected result is a decrease in behaviour, it is important to rule out the possibility that the observed decreases result from a general malaise or other deleterious effect of the independent variable. In the present experiments, pre-treatment with the PKC inhibitor was associated with lower levels of AngII-induced water intake. It is unlikely, however, that this effect was due to a global suppressive effect because the PKC inhibitor had no effect on saline intake in the same animals. A similar argument can be made for the experiments with the MAP kinase inhibitor. Specifically, animals given AngII and the MAP kinase inhibitor drank markedly less 1.5% NaCl than the animals injected with AngII alone in the same experiment, but levels of water intake were no different between the groups. Thus, it is unlikely that the observed differences were due to general malaise or a more global reduction in behaviour.
It is important to note that there was variability in the level of baseline water and 1.5% NaCl intakes between the experiments presented here. This variability in baseline intake should not be a source of great concern and is, indeed, why each experiment contained the appropriate positive (AngII alone) and negative (vehicle only and inhibitor only) control groups. Furthermore, the variability in fluid intake described here is no greater than observed between experiments in other reports, including the report by Fleegal and Sumners (2003) upon which these studies were based. Indeed, water intake in the various experiments reported in Fleegal and Sumners (2003) varies from approximately 6 ml to 12 ml. The source of this variance in the present and other findings is unclear, but one possibility describing the variability here is that the timing of the vehicle pre-treatment (DMSO) in the third experiment somehow facilitated the actions of AngII on saline intake, without having a similar effect when given at a different time point. It is difficult, however, to speculate on why this effect would be selective to the saline intake, but lead to no apparent difference in water intake. In spite of the differences in baseline intake between the experiments, it remains clear that animals pre-treated with the MAP kinase inhibitor U0126 (delivered in DMSO) 20 min before injection of AngII drank markedly less saline than animals pre-treated with DMSO before injection of AngII. It is equally apparent that these groups of animals did not differ in their water intake. Thus, it is reasonable to conclude that MAP kinase inhibition reduced AngII-induced intake of 1.5% NaCl without affecting water intake.
The data provided here are consistent with a primary role of the AT1 receptor, rather than the type II (AT2) receptor, in the ingestive responses stimulated by AngII. Although we do not discount the possibility of more subtle influences through the AT2 receptor, the inhibitors used here do not likely target any activity of this receptor subtype. Specifically, in vitro studies clearly show that AT1 receptors couple to Gq, leading to PLC activation and subsequent formation of IP3 and PKC. It is also clear that AT1 receptor ligands can stimulate MAP kinase phosphorylation (activation) (Sadoshima et al., 1995). In contrast, the AT2 receptor does not activate Gq, but more likely couples to Gi, with no apparent connection to PLC (Mukoyama et al., 1993; Hansen et al., 2000). There is, however, a known effect of AT2 receptor stimulation on MAP kinase, but AT2 receptors generate decreases in activated MAP kinase, not the increases associated with the AT1 receptor (Huang et al. 1996; Horiuchi et al., 1998). These differences and the direction of the observed effects make it far more likely that the inhibitors used here target the AT1 receptor and less likely that the results are due to disruption of AT2 receptor signalling.
Although the present experiments represent an important step toward understanding the role of two intracellular signalling pathways in the separable ingestive behaviours stimulated by AngII, the data do not provide direct information about the anatomical specificity of the observed effects. Nevertheless, several lines of evidence collectively provide clues regarding likely sites of action. A parsimonious explanation points to a specific set of structures known to be involved in the drinking response to AngII and known to express AT1 receptors because structures that express AT1 receptors are the most likely behaviourally relevant target of endogenous or exogenous AngII (Beresford & Fitzsimons, 1992; Sakai et al., 1994; Sakai et al., 1995; Weisinger et al., 1997; Li et al., 2003) and because these receptors couple to the pathways targeted in the present experiments. The subfornical organ (SFO) and organum vasculosum of the lamina terminalis (OVLT) are two sensory circumventricular organs that have been highlighted as targets of AngII and are critical for the observed ingestive responses (Buggy & Johnson, 1978; Simpson et al., 1978; McKinley et al., 1990; Morris et al., 2002). Both of these structures have reciprocal neural connections with the median preoptic nucleus (MnPO) (Camacho & Phillips, 1981; Miselis, 1981; Lind et al., 1982; Saper & Levisohn, 1983), which also expresses AngII receptors (Lenkei et al., 1997). Moreover, lesions placed in the MnPO inhibit AngII-induced drinking and pressor responses (Ployngam & Collister, 2007). In addition to the endocrine action of AngII at the circumventricular organs, it is also likely that endogenous AngII acts as a neurotransmitter at the paraventricular nucleus of the hypothalamus (PVN) (Bains et al., 1992; Ferguson & Wall, 1992; Li & Ferguson, 1993; Ferguson et al., 2001), but there is less evidence that these actions are required for the ingestive responses to AngII. Given that each of these structures lies near or on the walls of the ventricles, the injected inhibitors could easily reach these structures and alter their response to AngII. It is also likely, however, that the responsive circuits within the brain use other transmitters or neuromodulators that act through receptors coupling to the same intracellular signalling pathways targeted here. As such, it is difficult to construct a complete list of involved structures that may employ PKC or MAP kinase to generate the ingestive responses to AngII, but it is highly probable that any such list would include the OVLT and SFO.
The present findings highlight the complexity of behavioural regulation by endocrine systems and call attention to a novel mechanism through which a single peptide can influence multiple behaviours. This mechanism provides additional diversity to ligand-selective signalling, which has been observed by a number of receptors including those for dopamine, substance P, opioids, cholecystokinin, serotonin, and gonadotropin-releasing hormone (GnRH) (Kenakin, 2003). Like these receptor systems and the receptors for AngII (Daniels et al., 2005), modification of the ligand results in the selective activation of one intracellular signalling pathway or the other. The GnRH receptor provides an interesting example in which several natural forms of the ligand exist, providing diversity of function (Millar, 2005). The mechanism revealed by the present findings, however, differs from anything described previously because here a single ligand appears to produce multiple behaviours, presumably through a single receptor type, by stimulating divergent intracellular pathways. As such, these data may provide important information relevant to the role of other receptor systems in various behaviours and certainly extend our knowledge about the regulation of behaviour by AngII.
The authors are grateful to Dr. Craig Colder (University at Buffalo) for providing statistical expertise and consultation. Ethan Gable, Martha Adams, and Etana Berger provided technical support and Dr. Lucy Faulconbridge (University of Pennsylvania) provided assistance with pilot studies that led to the current experiments.
Support for these experiments was provided by the National Institutes of Health through awards DK-73800 (D.D.) and DK-52018 (S.J.F.).