Circadian integration of glutamatergic signals by little SAAS in novel suprachiasmatic circuits

Abstract

Background: Neuropeptides are critical integrative elements within the central circadian clock in the suprachiasmatic nucleus (SCN), where they mediate both cell-to-cell synchronization and phase adjustments that cause light entrainment. Forward peptidomics identified little SAAS, derived from the proSAAS prohormone, among novel SCN peptides, but its role in the SCN is poorly understood.

Methodology/principal findings: Little SAAS localization and co-expression with established SCN neuropeptides were evaluated by immunohistochemistry using highly specific antisera and stereological analysis. Functional context was assessed relative to c-FOS induction in light-stimulated animals and on neuronal circadian rhythms in glutamate-stimulated brain slices. We found that little SAAS-expressing neurons comprise the third most abundant neuropeptidergic class (16.4%) with unusual functional circuit contexts. Little SAAS is localized within the densely retinorecipient central SCN of both rat and mouse, but not the retinohypothalamic tract (RHT). Some little SAAS colocalizes with vasoactive intestinal polypeptide (VIP) or gastrin-releasing peptide (GRP), known mediators of light signals, but not arginine vasopressin (AVP). Nearly 50% of little SAAS neurons express c-FOS in response to light exposure in early night. Blockade of signals that relay light information, via NMDA receptors or VIP- and GRP-cognate receptors, has no effect on phase delays of circadian rhythms induced by little SAAS.

Conclusions/significance: Little SAAS relays signals downstream of light/glutamatergic signaling from eye to SCN, and independent of VIP and GRP action. These findings suggest that little SAAS forms a third SCN neuropeptidergic system, processing light information and activating phase-shifts within novel circuits of the central circadian clock.

Figure 1. Little SAAS localizes to the central SCN of rat and mouse.

Patterns of neuropeptide immunoreactivity in rat and mouse are compared in mid-SCN coronal section. (A, D) Schematic of previously established neuropeptide expression , , in sub-compartments (shading) and somata (circles): AVP, orange; VIP, blue; GRP, green. (B,C,E,F) Photomicrographs of immunohistochemistry demonstrate that little SAAS localization (green) in rat mid-SCN (B,C) is similar to that observed in mouse SCN (E,F). In both species, little SAAS-positive somata are primarily in the central sub-compartment, although punctuate expression extends more broadly. Little SAAS (green) is shown double-labeled with arginine vasopressin, AVP (B, red), vasoactive intestinal peptide, VIP (C, red), calbindin, CalB (E, red) or gastrin releasing peptide, GRP (F, red). Representative higher magnification (inset, F) of little SAAS (green) and GRP (red) shows significant somatic co-localization (yellow). Each SCN is delineated by an outer dashed line in the shape of an oval (rat) or teardrop (mouse). Localization of strong immunoreactivity for little SAAS is marked by an inner circle, which corresponds to previous descriptions of the central SCN. Images are single optical sections (B,F, z = 0.11µm; F inset, z = 0.09 µm). OX, optic chiasm; SCN, suprachiasmatic nucleus; 3V, third ventricle. Scale bar = 50 µm.

Figure 2. Little SAAS expression spans across VIP and GRP cytoarchitectonic regions of rat SCN.

Confocal tile-scan images of coronal tissue sections from rostal-to-caudal quadrants (vertical panels) reveal that little SAAS localizes to VIP- and GRP-expressing sub-compartments in all quadrants. SCN are triple immunostained for VIP (white; A,F,K,P), GRP (green; B,G,L,Q), and little SAAS (red; C,H,M,R). Merge of images in the three channels across each row (D,I,N,S) reveals co-localizations: VIP/little SAAS (pink), VIP/GRP (light green), and GRP/little SAAS (yellow). Enlargements of these merged images appear in supplementary material (Figs. S2, S3, S4, S5). Schematic representations of each merge image (E, J, O, T) mark the location of each stained soma in each optical slice: VIP (blue circles), GRP (green circles), little SAAS (red circles), VIP/little SAAS co-expressing (pink diamonds), and GRP/little SAAS co-expressing (yellow diamonds). Gray arrowheads in B indicate non-specific staining of particulate introduced during tissue mounting. Whereas VIP localizes primarily to the more ventromedial region, GRP occupies the more central SCN, as reported previously. Some overlap with VIP exists along the medial edge of VIP localization, whereas laterally there is considerable overlap with GRP. Non-somatic immunostaining marks peptide localized in cellular processes and puncta. Coronal sections (40 µm) are from colchicine-treated rat. Images in each row are the same optical plane (z = 0.11 µm). Approximate image coordinates provided in I, N, and S are relative to the rostral-most image (D, designated as +0 µm). 3V, third ventricle; OX, optic chiasm. Scale bar (in S) = 100 µm.

Figure 3. Light-induced c-FOS colocalizes with little SAAS in subset of neurons in the central SCN.

(A–B) Confocal tile-scan images of SCN-containing coronal tissue sections from a rat receiving no light exposure vs. 400-lux light at CT 14 for 1 h. Light induction of c-FOS (green) is within the region of little SAAS staining (red) in the ventrolateral core SCN. High magnification optical section (z = 0.09 µm) reveals c-FOS (nuclear) and little SAAS (cytoplasmic) within the same cell. (C) Cell counting from mid-SCN sections found 23.7±5.3% of c-FOS-positive neurons also are little SAAS-positive, while 48.2±1.7% of little SAAS-positive neurons are also c-FOS positive. Tissue section thickness = 40 µm; optical image depth = 0.11 µm; 3V, third ventricle; OX, optic chiasm. Scale bar, A–B (shown in B) = 100 µm; scale bar, B inset = 10 µm.

Figure 4. Little SAAS is not expressed in mouse melanopsin-positive retinal ganglion cells or rat optic nerve.

(A–C) Confocal micrographs of male melanopsin-GFP transgenic mouse retina. (A) GFP-immunoreactivity (green) is localized to melanopsin-expressing cell bodies of the ganglion cell layer (GCL). (B) Little SAAS staining (red) is not observed above non-specific background levels in the same layer of retina. (C) Overlay of both channels shows no co-expression of little SAAS and melanopsin-GFP. (D–E) Mass spectrometry analysis of peptides in rat optic nerve. (D) Representative peptide profile of the optic nerve as detected by MALDI-TOF MS in extracts of individual nerve samples. Major peak at m/z 1784.9 corresponds to myelin basic protein (MBP_s). Other truncated forms of MBP_S verified by tandem MS are labeled. (E) Tandem mass spectrometry by MALDI-TOF/TOF confirms identity of the 1784.9 peak as acetylated N-terminus fragment of MBP_S. GCL, ganglion cell layer of the retina; INL, inner nuclear layer of the retina; IPL, inner plexiform layer of the retina; ME, median eminence; ONL, outer nuclear layer of the retina; OPL, outer plexiform layer of the retina; scale bar, C = 100 µm.

Figure 5. Little SAAS acts downstream of NMDAR and parallel to VIP/GRP signaling during early night.

(A, I) SCN in a coronal brain slice displays characteristic mid-subjective daytime peak at CT 7 in spontaneous rhythm of neuronal firing rate. (B, I) The competitive NMDAR antagonist, APV (100 µM), does not alter phase. (C, I) Application of little SAAS following pre-treatment with APV results in phase delay. (D, I) Application of little SAAS alone causes a similar phase delay. (E, I) Firing rate rhythm in control SCN slice with peak at CT 7. (F, I) Application of glutamate (10 mM, 1µL drop) at CT 14 causes phase delay. (G, I) Application of little SAAS antiserum (#2768,undiluted) does not significantly alter SCN phasing. (H, I) Little SAAS antiserum blocks the phase delay induced by glutamate. (I) Mean data demonstrate that whereas the NMDAR antagonist, APV, does not block little SAAS action, little SAAS antiserum blocks the phase delaying effect of glutamate, but does not affect not the phase delay stimulated by the GRP/VIP cocktail. n≥3 for all groups; **, p≤0.001 vs. control, one-way ANOVA with Bonferroni post-hoc test.

Figure 6. Effects of little SAAS on SCN phasing are independent of VPAC2/BB2 signaling.

Microdrop application to the SCN at CT 14 of little SAAS (1 or 100 nM) or VIP/GRP (100 nM each) induces significant phase delay compared with controls (p≤0.01). Pre-incubation with a cocktail of VPAC2 receptor antagonist [4Cl-D-Phe,Leu]VIP (50 mM)/BB2 receptor antagonist PD176252 (50 mM) does not alter SCN phasing (0.04±0.29 h) vs. controls. The antagonist cocktail does not alter the phase delay induced by little SAAS (−1.94±0.43) but fully blocks the phase shift induced by a VIP/GRP (−0.13±0.20 h). n≥3 for all groups; **, p≤0.01 vs. control; $, p≤0.01 vs. 100 nM VIP/GRP cocktail; #, p≤0.05 vs. 100 nM VIP/GRP cocktail, one-way ANOVA w/Tukey-Kramer post-hoc test.

Figure 7. Model of light signal processing in the rat SCN.

(A) Schematic of general functional organization within the rat SCN (coronal section, mid-SCN). SCN regions (ventral, dark blue; central, light blue; and dorsomedial, purple) denote functionally distinct regions that receive and integrate different signals (central/ventral) and relay this information locally and to the dorsomedial SCN. These regions also extend efferents to nearby hypothalamus. (B) Peptidergic signal processing within the SCN. (Bi) Glutamate released from RHT terminals activates NMDA receptors, which are expressed on neurons throughout the SCN (Bii) Signal transduction downstream of NMDAR activation leads to c-FOS induction and peptide release in a subset of neurons. Intra-SCN signaling is mediated by VIP, GRP and/or little SAAS peptides. (Biii) VIP and GRP transmit signals via VPAC2 and BB2 receptors, respectively. Little SAAS functions independently of VPAC2 or BB2 receptor signaling, and is predicted to function through (Biv) an unidentified cell surface receptor, and (Bv) downstream signal-processing events.