Identification of central projections from amylin-activated neurons to the lateral hypothalamus

Abstract

The ability of the pancreatic hormone amylin to inhibit food intake relies on a direct activation of the area postrema (AP). This activation is synaptically transmitted to the nucleus of the solitary tract (NTS), the lateral parabrachial nucleus (LPB), the central amygdaloid nucleus (Ce) and the lateral bed nucleus of stria terminalis (BSTL). Interestingly, neurons of the rostro-dorsal lateral hypothalamic area (dLHA), which are activated during fasting, are inhibited by peripheral amylin, although they lack amylin receptors. Using the retrograde tracer cholera toxin-B (Ctb) we analyzed whether the dLHA receives neuronal projections from amylin-activated brain areas. The anterograde tracer biotinylated dextran-amine (BDA) was used to confirm the projections and to identify further neuronal pathways potentially involved in amylin signaling. We identified dense projections from the amylin activated neurons in the LPB and sparse projections from the NTS to the dLHA. LPB fiber efferents were found in close proximity to dLHA nuclei activated by 24 h of fasting. The AP and the Ce showed no projections to the dLHA. Dense efferents were also observed from the LPB to other hypothalamic areas, namely to the ventromedial, dorsomedial, paraventricular and arcuate nuclei. This study provides neuroanatomical evidence that among the amylin activated areas, the LPB provides the strongest input to the dLHA, thus it may mediate the amylin-induced inhibition of the dLHA.

Introduction

The maintenance of energy homeostasis is a fundamental requirement for body function. Part of the system controlling food intake and body weight relies on peripheral satiation signals. One of these signals is the peptide hormone amylin, which is co-secreted with insulin from pancreatic beta-cells in response to food intake (Butler et al., 1990, Ogawa et al., 1990) and is considered a physiological satiation signal (Lutz et al., 1995, Lutz, 2005). Chronic amylin administration leads to a sustained reduction in food intake due to a decrease in average meal size and subsequently to a reduction in body weight gain in rats (Lutz et al., 2001). Amylinomimetic agents were shown to reduce eating and decrease body weight in obese human subjects (Aronne et al., 2007, Smith et al., 2007). These mimetics also improve blood glucose profiles in diabetic patients (Weyer et al., 2001, Hollander et al., 2004).

The area postrema (AP) of the brainstem, which lacks a functional blood–brain barrier, mediates amylin's anorectic action. The important role of the AP has been demonstrated in lesion studies showing that amylin's anorectic effect is blunted in AP lesioned rats (Lutz et al., 1998, Lutz et al., 2001). A direct action of amylin on AP neurons has been confirmed in electrophysiological studies, in which amylin exerted excitatory effects (Riediger et al., 2001).

Immunohistochemical studies using the immediate early gene product c-Fos as a marker of neuronal activation showed that peripherally applied amylin not only activates the AP, but also the nucleus of the solitary tract (NTS), the lateral parabrachial nucleus (LPB), the central amygdaloid nucleus (Ce) and the lateral subdivisions of the bed nucleus of the stria terminalis (BSTL) (Rowland et al., 1997, Riediger et al., 2004). These brain structures are part of the central gustatory/enteroceptive systems and are mutually interconnected and linked to the hypothalamus, the main integrative center for the control of energy balance (Saper, 2002). Since the amylin-induced c-Fos response in the NTS, LPB and Ce was markedly attenuated in AP-lesioned rats, activation of these neurons seems to be synaptically mediated and secondary to an action of amylin on AP neurons (Rowland and Richmond, 1999, Riediger et al., 2004).

It has been shown that amylin affects hypothalamic centers involved in the control of eating. Amylin and its receptor agonist, salmon calcitonin, decrease the mRNA expression of orexigenic peptides orexin and melanin concentrating hormone (MCH) in the lateral hypothalamic area (LHA) (Barth et al., 2003). In a recent study, we identified a particular group of fasting-activated neurons of unknown neurochemical phenotype that are arranged as a cluster in a defined rostro-dorsal subregion of the LHA, which is termed dLHA in the current work (Riediger et al., 2004). The fasting-induced c-Fos expression in dLHA neurons was attenuated both by refeeding and also by injecting amylin in rats that had no access to food (Riediger et al., 2004). Since amylin administration alone was sufficient to partly reverse the fasting-induced dLHA activation, exogenously applied amylin seems to mimic the ingestion of food, as far as the inhibitory impact on dLHA neurons is concerned. A direct inhibitory influence of amylin on these dLHA neurons is unlikely because the entire LHA is devoid of amylin binding sites (Beaumont et al., 1993, Sexton et al., 1994, van Rossum et al., 1994). Therefore, we hypothesize that the dLHA receives input from amylin-activated neurons of the AP-NTS-LPB-Ce-BSTL axis.

Using retrograde (cholera toxin B, Ctb) and anterograde (biotinylated dextran-amine, BDA) tracing approaches in adult rats, we sought to investigate whether amylin-activated neurons project to the dLHA. To test this we correlated amylin- and fasting-induced c-Fos expression with tracer labeling resulting from local tracer microinjection into the investigated brain nuclei.