Short photoperiod-induced decrease of histamine H3 receptors facilitates activation of hypothalamic neurons in the Siberian hamster

Abstract

Nonhibernating seasonal mammals have adapted to temporal changes in food availability through behavioral and physiological mechanisms to store food and energy during times of predictable plenty and conserve energy during predicted shortage. Little is known, however, of the hypothalamic neuronal events that lead to a change in behavior or physiology. Here we show for the first time that a shift from long summer-like to short winter-like photoperiod, which induces physiological adaptation to winter in the Siberian hamster, including a body weight decrease of up to 30%, increases neuronal activity in the dorsomedial region of the arcuate nucleus (dmpARC) assessed by electrophysiological patch-clamping recording. Increased neuronal activity in short days is dependent on a photoperiod-driven down-regulation of H3 receptor expression and can be mimicked in long-day dmpARC neurons by the application of the H3 receptor antagonist, clobenproprit. Short-day activation of dmpARC neurons results in increased c-Fos expression. Tract tracing with the trans-synaptic retrograde tracer, pseudorabies virus, delivered into adipose tissue reveals a multisynaptic neuronal sympathetic outflow from dmpARC to white adipose tissue. These data strongly suggest that increased activity of dmpARC neurons, as a consequence of down-regulation of the histamine H3 receptor, contributes to the physiological adaptation of body weight regulation in seasonal photoperiod.

Fig. 1.

Neuronal activation, indicated by c-fos expression, in the dmpARC correlates with a LD to SD shift in photoperiod. A, Autoradiography of c-fos antisense riboprobe in situ hybridization to Siberian hamster brain sections housed in LD and SD photoperiod, respectively. Arrow indicates dmpARC. B, Emulsion-coated sections of LD and SD brain sections, respectively, enlarged to show the hybridization signal in the region of the dmpARC (arrow). C, Immunohistochemistry on perfused LD and SD. Siberian hamster brain sections with a c-Fos antibody (dark brown) in the region of the dmpARC. Arrow indicates example of immunostained c-Fos positive neuron. D, Analysis of c-fos expression over the course of a 24-h light-dark cycle. Shown are the values of c-fos expression in a SD light-dark cycle. No difference in the expression was observed in a 24-h LD light-dark cycle. E, Graph plotting the inverse relationship of c-fos expression to photoperiod changes in testis weight and body weight. Scale bars (B and C), 20 μm.

Fig. 2.

Retrograde labeling of arcuate nucleus neurons after injection of PRV into adipose tissue fat pads. A–E, Immunostained PRV-infected neurons through the arcuate nucleus. Subdivisions are indicated. Insert (A) is a dark-field image of an in situ hybridization performed for histamine H3 receptor to illustrate the location of the dmpARC and the relative locations of the labeled histamine H3 receptor-labeled neurons with PRV-infected neurons. The location defined as the dmpARC is outlined by a dotted line. F, A higher magnification image of the area boxed in image (B) showing PRV infected neurons in the dmpARC. 3V, Third ventricle; LP, lateral posterior arcuate nucleus; MP, medial posterior arcuate nucleus; dmp, dorsomedial posterior arcuate nucleus; DM, dorsomedial nucleus; VMH, ventral medial nucleus; Arc, arcuate nucleus.

Fig. 3.

Increased neuronal activity in dmpARC neurons correlates with a LD to SD shift in photoperiod. A, Continuous records from two neurons illustrating spontaneous action potential firing is significantly lower in LD vs. SD photoperiods. B, Superimposed electrotonic potentials evoked in response to current pulse injection of variable amplitude (not shown) in LD (top) and SD (bottom) photoperiod revealed increased activity in SD neurons was associated with an increase in input resistance, suggesting closure of ion channels in SD vs. LD. C, Corresponding voltage-current relations. Note the increased slope in the SD dmpARC neuron, indicating a higher input resistance (open circles) than the corresponding neuron in LD (closed squares). D, Summary overview of the averaged data for firing rate and membrane resistance showing the significant increase in these membrane properties in SD vs. LD photoperiods. ***, Significance at P < 0.001).

Fig. 4.

H3 receptor-mediated signaling regulates neuronal activity of dmpARC neurons in LD but not SD photoperiods. A, Continuous records from two dmpARC neurons showing the H3 receptor agonist imetit-induced inhibition in a LD dmpARC neuron (top) but was without effect on a SD dmpARC neuron (bottom). B, Continuous record showing the H3 receptor antagonist/partial agonist clobenpropit-induced excitation of a LD dmpARC neuron (top) and was without effect in SD dmpARC neurons (bottom). The downward deflections in membrane potential in A (top) and B (bottom) are the result of repetitive rectangular-wave constant current injections (0.2Hz, −10 pA, 0.5 sec) used to monitor changes in membrane resistance. C, Bar charts summarizing the effects of imetit (filled bars) and clobenpropit (diagonal bars) on firing rate (left) and membrane resistance (right) in LD hamsters. Imetit significantly decreased firing rate and input resistance, whereas clobenpropit increased firing rate and input resistance, suggesting mechanisms mediated through opening and closing of ion channels, respectively. *, P < 0.05; **, P < 0.005.

Fig. 5.

H3 receptors in LD dmpARC neurons modulate neuronal activity via a chloride-dependent mechanism. A, Samples of a continuous record showing superimposed electrotonic potentials evoked in response to current injection in the absence (top) and presence of imetit (bottom). Right, Corresponding voltage-current relations (control, closed squares; imetit, open circles). B, Samples of a continuous record showing superimposed electrotonic potentials evoked in response to current injection in the absence (top) and presence of clobenpropit (bottom). Right, Corresponding voltage-current relations (control, closed squares; clobenpropit, open circles). Note plots in both A and B intersect around −60 mV, close to the reversal potential for chloride ions under our recording conditions. C, Top trace, depolarization of dmpARC neurones by clobenpropit; middle trace, the application of a combination of bicuculline and 2-OH-saclofen mimicked the clobenpropit-induced membrane depolarization previously observed in the same neurons (top trace). Subsequent application of clobenpropit in the presence of bicuculline and 2-OH-saclofen failed to induce a membrane response. Bottom trace Upon washout of bicuculline and 2-OH-saclofen, the clobenpropit responsiveness was restored. Shown is a representative trace (one of three) of responses to drug applications performed in a dmpARC neuron.