The raphe nuclei are distinct brain loci composed of groups of neurons located along the brainstem that have been implicated in many physiological functions such as the control of the sleep-wake cycle, motivation, pain, energy balance, and food intake (Berger et al., 2009; Pytliak et al., 2011; Schneeberger et al., 2019). One of these nuclei is the dorsal raphe nucleus (DR) which is located beneath the cerebral aqueduct and constitutes a collection of neurons with distinct morphology, projections, and neurochemical phenotypes (Adell et al., 2002). The DR sends neuronal projections to many forebrain structures, including a robust innervation to the hypothalamus, an important area that influences food intake (Muzerelle et al., 2016; Ren et al., 2019).

Several studies indicate that the DR has a pivotal role in feeding behavior (Stachniak et al., 2014; Anderberg et al., 2017; Bruschetta et al., 2020). Immunohistochemical studies revealed that food intake triggers neural activity in this nucleus (Wu et al., 2014). Moreover, optogenetic activation of specific GABAergic DR neurons has been shown to increase food intake, while activation of glutamatergic DR neurons suppresses feeding (Nectow et al., 2017). The DR is the main source of serotonin (5-HT) in the central nervous system, a neurotransmitter associated with satiety (Blundell and Latham, 1979; Blundell, 1991). Pharmacological approaches demonstrated that injection of 5-HT or 8-OH-DPAT, a 5-HT_1A receptor agonist, into the DR induces feeding in satiated rats. These effects were attributed to the activation of inhibitory DR 5-HT_1A somatodendritic autoreceptors, which may regulate 5-HT release (Hutson et al., 1986; Fletcher and Davies, 1990).

DR neural activity is also sensitive to endogenous catecholamines (Adell et al., 2002), receiving substantial noradrenergic input, especially from the commissural part of the nucleus of the solitary tract (A2) and the locus coeruleus (A6) (Peyron et al., 1996). High levels of mRNA for the α-1 adrenoceptors are present in the DR (Pieribone et al., 1994; Day et al., 1997) and in DR 5-HT neurons (Day et al., 2004). However, studies documented a only moderate presence of α-2 adrenoceptors in the DR (Unnerstall et al., 1985; Rosin et al., 1993; Talley et al., 1996) with no significant expression of α-2 adrenoceptor mRNA. These data suggest that these receptors are located presynaptically in noradrenergic terminals in this nucleus (McCune et al., 1993; Nicholas et al., 1993; Scheinin et al., 1994). Electrophysiological and microdialysis experiments disclosed that activation of α-1 adrenoceptors leads to an increase in local serotonin release and increase the firing rate of DR 5-HT neurons, while the activation of α-2 adrenoceptors leads to a decrease of serotonin release in this nucleus (Baraban and Aghajanian, 1980; Vandermaelen and Aghajanian, 1983; Bortolozzi and Artigas, 2003; Pudovkina et al., 2003). Moreover, lesions with DSP-4, a neurotoxin that impairs noradrenergic projections, abolishes the effects of local clonidine injection on 5-HT release, which suggests that, when administered into the DR, it acts predominantly on α-2 autoreceptors (Bortolozzi and Artigas, 2003).

In a recent study, we demonstrated that injection of α-2 agonist clonidine into the DR of satiated rats evoked hyperphagia (Flores et al., 2021). The feeding response induced by clonidine was similar to that found after noradrenaline or adrenaline injections into the DR, suggesting that this hyperphagia depends on α-2 adrenoceptors activation, while injection of a specific α-1 agonist did not affect food intake (Flores et al., 2021). Based on these previous data, we hypothesized that injection of α-adrenoceptor agonists into DR may also affect ingestive responses in fasted animals. To better understand the functional role of DR α-adrenoceptors in feeding behavior, this study aims to evaluate the effects of pharmacological manipulations of α-adrenergic agonists in the DR on food intake after fasting.

All rats included in statistical analyses (n = 112) had injection sites confirmed to be in the DR by histological analysis (Figures 1A,B).

In the present study, we investigated the effects of pharmacological manipulations of α-adrenergic agonists in the DR on food intake in fasted rats. Overall, we observed that acute injections of NA or AD into DR evoked reduced food intake in fasted rats. α-adrenoceptors are present in the DR, therefore our assumption is that the decrease in food consumption after NA or AD infusions might be due to the activation of these receptors. Strengthening this notion, injection of specific α-1 agonists PHE into DR of fasted rats decreased food intake similarly to response induced by NA or AD injection. The injection of specific α-2 agonist CLO into DR does not affect feeding, indicating that the activation of α-1 postsynaptic adrenoceptors in this nucleus has an inhibitory influence on feeding in fasted rats.

Interestingly, in a previous study opposite feeding responses were induced by injection of α-adrenoceptors agonists into the DR of satiated rats. In the fed state, AD, NA, or CLO injection increased food intake while PHE treatment did not change feeding behavior (Flores et al., 2021). Based on these findings, the hyperphagic effect was attributed to the inhibition of NA release in DR noradrenergic terminals by the activation of α-2 presynaptic auto-receptors. This activation removes a possible endogenous α-1 adrenergic stimulatory tone on DR serotonergic neurons of fed animals, leading to a decrease in 5-HT release in projection areas, which could favor ingestive behaviors (Flores et al., 2021). In fact, 5-HT release decreases after CLO injection into DR, and lesion with DSP-4 abolishes these effects (Bortolozzi and Artigas, 2003). These data are, in part, corroborated by the experiments in the present work. The hypophagia caused by PHE injection into the DR of fasted rats is comparable to hypophagia induced by NA or AD injections. On the other hand, injection of CLO does not change food intake in these rats, suggesting that the action of AD or NA is mediated by α-1 adrenoceptors in this case. Due to this difference in feeding depending on whether the animal is fed or not, we believe the intensity of this endogenous noradrenergic activity, mediated by α-1 adrenoceptors in the DR, seems to decline in fasted rats.

Several studies demonstrated that 5-HT acts as a satiety signal in hypothalamic nuclei, such as the arcuate and the paraventricular nuclei, as well as other areas such as the parabrachial nucleus and nucleus of the solitary tract (Voigt and Fink, 2015). Stimulation of DR neurons increases extracellular 5-HT levels in the hypothalamus (De Fanti et al., 2000) and manipulation of adrenoceptor activity in the DR induces FOS expression in discrete populations of arcuate and paraventricular nucleus neurons (Flores et al., 2021). Thus, it is possible that activation of DR α-1 adrenoceptors by PHE results in 5-HT release in fasted animals, since facilitatory control of 5-HT release is attributed to these receptors (Bortolozzi and Artigas, 2003). In agreement with this suggestion, a study demonstrated that PHE injections into the median raphe nucleus (MR), another major serotonergic cell group with α-adrenoceptors (Adell and Artigas, 1999), also evoked hypophagia in fasted rats (Ribas et al., 2012). Also, serotonergic activity is low in food restricted rats and, especially in the DR, food restriction decreases the optical density of 5-HT positive neurons when compared to fed rats (Haider and Haleem, 2000; Kang et al., 2001). Therefore, our hypothesis is that in fasted animals the effect of AD or NA injection into the DR is mainly mediated by α-1 adrenoceptor, stimulating 5-HT release in the projection areas and consequently decreasing food intake (Figure 4). However, further experiments using adrenergic antagonists in fasted and fed rats are necessary to better understand the role of these DR adrenergic receptors in feeding behavior.

DR is a neurochemically heterogeneous structure containing distinct clusters of 5-HT neurons and several other differentially distributed major neurotransmitters and neuropeptides (Calizo et al., 2011). Some DR subregions display large proportions of GABAergic, dopaminergic, glutamatergic and neurons with a mixed glutamatergic/serotonergic phenotype (Hioki et al., 2010; Calizo et al., 2011; Soiza-Reilly and Commons, 2011). It has been reported that ICV injection of AD 20 nmol evoked serotonergic and non-serotonergic neuronal activation in the DR (Flores et al., 2018). In a recent study conducted by Nectow et al. (2017), treatments that enhance GABAergic tone within DR lead to an increase in food intake, while activation of DR glutamatergic neurons decreases feeding. α-1 adrenoceptors have been reported to be robustly expressed only in 5-HT DR neurons and in a small population of DR GABAergic neurons that expresses the type α-1b receptor (Day et al., 2004). There was no evidence, until the present study, that α-adrenoceptors are expressed in glutamatergic DR neurons. However, the functional data presented in this work does not support activation through GABAergic α-1b receptors, since these neurons stimulate food intake when activated. These observations further support the idea that decreased food intake involves the participation of 5-HT DR neurons.

In addition to the effects observed on food intake, AD, NA and PHE significantly decreased water intake. Generally, water intake occurs in conjunction with food intake, i.e., animals often drink fluids during or right after a meal (Kissileff, 1969; Mecawi et al., 2015). Indeed, in this study water and food intake after AD and NA microinjections were strongly correlated. However, intra-DR PHE administration affected water intake in the smaller dose used, which was not sufficient to affect food intake. This result indicates that activation of α-1 receptors within the DR may inhibit water intake irrespective of food intake. Previous studies have shown that electrolytic lesion of the DR or depletion of 5-HT synthesis induce water intake in rats. This increase in water intake is accompanied by decreased urinary volume, and several endocrine alterations that culminate with water retention, indicating participation of DR serotoninergic systems in hydromineral balance (Reis et al., 1994). Additionally, acute administration of a 5-HT1a agonist, which decreases endogenous 5-HT release, potently induced water intake in rats (Fonseca et al., 2009). These studies indicate that activation of DR 5-HT neurons decreases water intake. Thus, the effects of PHE administration to decrease water intake further supports the idea that α-1 activation increases the activity of DR 5-HT neurons that, in turn, affects the ingestive behaviors in rats.

Binding studies also demonstrated that NA shows some affinity for dopaminergic receptors, with a low to moderate potency to bind and activate D4-class receptors (Lanau et al., 1997; Newman-Tancredi et al., 1997; Sánchez-Soto et al., 2016). In contrast, some authors report low to moderate levels of D2 receptor expression in DR 5-HT neurons (Dawson et al., 1986; Mansour et al., 1990; Meador-Woodruff et al., 1991; Ferré and Artigas, 2006). Despite the possibility of NA acting on D2 receptors, the similarity between the responses evoked after NA injection and the specific α-1 agonist PHE, support the hypothesis that the effects of NA are mediate by α-1 receptors. Additionally, although some activity via D2 receptors is possible, NA shows higher affinity for α and β adrenoceptors; consequently, the probability of NA acting at these receptors is greater. Furthermore, to our knowledge, there is no evidence that DR dopaminergic receptors affect feeding behavior in rats.

Manipulations that interfere with the serotonergic system can potentially influence a variety of behaviors in addition to food intake (Lucki, 1998). However, DR injections of adrenergic agonists only modified immobility behavior. The neurochemical mechanism by which AD or NA may reduce food intake during food deprivation can be attributed to an anticipation of satiety signals, a result consistent with the postulated inhibitory role for 5-HT in controlling eating behavior (Hutson et al., 1986; Blundell, 1991; Leibowitz and Alexander, 1998; Wirtshafter, 2001). Therefore, these results indicate that the treatments used may have anticipated the behavioral sequence of satiety, which is characterized by the increase in immobility after food consumption (Halford et al., 1998), thus inducing the end of the meal. Adrenergic agonist injections into DR decreased feeding duration and this response has been linked to changes in mechanisms that end the meal (Ritter and Epstein, 1975; Blundell, 1986). Several studies have documented increased neuronal activation in A2/C2 catecholaminergic neurons in the brainstem in response to anorexic peptides such as cholecystokinin (Rinaman et al., 1993; Blevins et al., 2003). As previously mentioned, the DR receives robust NTS noradrenergic input (Peyron et al., 1996). Based on these findings, peripheral satiety signals from gut could induce satiety via an increase in the activity of NTS noradrenergic neurons that innervate the DR (Figure 4). However, further studies using specific chemogenetic or optogenetic approaches are required to better understand the neural circuits involved.

In conclusion, the data presented in this study indicate that activation of α-1 receptors in the DR reduces food intake in hungry animals, while activation of α-2 does not affect hunger-induced food intake. Interestingly, these effects differ from those observed in ad libitum fed rats, in which α-2 activation induces food intake. Taken together, these results suggest that an endogenous release of adrenalin/noradrenalin by DR neurons mediates satiety in fed rats, while in the overnight fasted rats the intensity of this endogenous noradrenergic activity mediated by α-1 adrenoceptors seems to decline.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by the Ethics Committee for the use of Experimental Animals (CEUA) of the Federal University of Santa Catarina, SC, Brazil (CEUA protocol: PP0075).

RF and MP conceived the study and designed the experiments. RF performed all the experiments with the participation of RS and JA-R in the behavioral tests. RD-S and IR-S contributed to data analysis. RF wrote the manuscript with input from all other authors and revision. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.