1. Neuroscience
Download icon

Behavioral role of PACAP signaling reflects its selective distribution in glutamatergic and GABAergic neuronal subpopulations

  1. Limei Zhang  Is a corresponding author
  2. Vito S Hernandez
  3. Charles R Gerfen
  4. Sunny Z Jiang
  5. Lilian Zavala
  6. Rafael A Barrio
  7. Lee E Eiden  Is a corresponding author
  1. Department of Physiology, Faculty of Medicine, National Autonomous University of Mexico, Mexico
  2. Section on Molecular Neuroscience, National Institute of Mental Health, Intramural Research Program, United States
  3. Laboratory of Systems Neuroscience, National Institute of Mental Health, Intramural Research Program, United States
  4. Department of Complex Systems, Institute of Physics, National Autonomous University of Mexico (UNAM), Mexico
Research Article
  • Cited 0
  • Views 625
  • Annotations
Cite this article as: eLife 2021;10:e61718 doi: 10.7554/eLife.61718

Abstract

The neuropeptide PACAP, acting as a co-transmitter, increases neuronal excitability, which may enhance anxiety and arousal associated with threat conveyed by multiple sensory modalities. The distribution of neurons expressing PACAP and its receptor, PAC1, throughout the mouse nervous system was determined, in register with expression of glutamatergic and GABAergic neuronal markers, to develop a coherent chemoanatomical picture of PACAP role in brain motor responses to sensory input. A circuit role for PACAP was tested by observing Fos activation of brain neurons after olfactory threat cue in wild-type and PACAP knockout mice. Neuronal activation and behavioral response, were blunted in PACAP knock-out mice, accompanied by sharply downregulated vesicular transporter expression in both GABAergic and glutamatergic neurons expressing PACAP and its receptor. This report signals a new perspective on the role of neuropeptide signaling in supporting excitatory and inhibitory neurotransmission in the nervous system within functionally coherent polysynaptic circuits.

Introduction

Pituitary adenylate cyclase-activating peptide (PACAP) was first isolated from ovine hypothalamic tissue and characterized as a peptide which stimulates cyclic AMP elevation in rat anterior pituitary cells in culture (Miyata et al., 1989). PACAP binding to its receptors Vipr1, Vipr2 and, predominantly, PAC1 initiates cell signaling through multiple intracellular pathways (Harmar, 2001). PACAP acting at PAC1 is generally considered to engage Gαs, activating adenylate cyclase, with some isoforms also activating phospholipase C via Gαq, leading to multiple cellular responses including increased neuronal excitability (Pisegna and Wank, 1993; Spengler et al., 1993; Kawasaki et al., 1998; Emery et al., 2013; Jiang et al., 2017; Johnson et al., 2019). The PACAP/PAC1 signaling pathway has consistently been related to psychogenic stress responding, and potentiation of this pathway has been linked to psychopathologies including anxiety and PTSD in human (Ressler et al., 2011; Wang et al., 2013a; Mustafa et al., 2015). PACAP gene knock-out in the mouse results in decreased hypothalamo-pituitary-adrenal (HPA) axis activation after physical or psychogenic stress (Stroth and Eiden, 2010; Tsukiyama et al., 2011), and a hypoarousal behavioral phenotype in response to psychogenic stress (Lehmann et al., 2013; Mustafa et al., 2015; Jiang and Eiden, 2016). However, interactions within and among populations of PACAP- and PAC1-expressing neurons in brain circuits mediating behavioral responses to environmental stimulation remain to be understood. This is a critical step in integrative understanding of the functional significance of PACAP-PAC1 neurotransmission.

Exploration of PACAP-containing circuits in rodent CNS has been based on reports of the distribution of PACAP peptide and mRNA, and on expression from reporter genes under the control of a PACAP promoter transgene (Hannibal, 2002; Condro et al., 2016; Koves, 2016) or knocked-in to the PACAP gene itself (Krashes et al., 2014). Hannibal reported the anatomical distribution of PACAP projection fields and cell groups in rat CNS employing immunohistochemistry (IHC) and in situ hybridization (ISH), using radiolabeled riboprobes, in a rigorous study. However, due to the paucity of PACAP in cell bodies, dendrites, and axons compared to nerve terminals, peptide IHC has not provided more definitive PACAP chemoanatomical circuit identification in rodent brain. Similarly, ISH with radiolabeled riboprobes, while identifying PACAP-positive cell bodies, lacks the resolution to identify the co-transmitter phenotypes and precise microanatomical features of these cell groups. Thus, heterogeneity of PACAP-containing neurons within and between brain regions, both with respect to cell type and accurate regional boundaries could not be discerned. Nevertheless, an essential function for PACAP as a neurotransmitter within one or more brain behavioral circuits, and consistent with the cellular and post-synaptic actions of PACAP, has not yet emerged. A systematic analysis with accuracy at the level of cellular co-phenotypes, and with anatomical resolution to the level of sub-nuclei within CNS, is essential to complete this task.

To address these issues, we conducted a systematic analysis placing the PACAP>PAC1 signaling into anatomical and basic sensorimotor circuit contexts. We first describe in detail the overall topographical organization of expression of mRNAs encoding PACAP (Adcyap1) and its predominant receptor PAC1 (Adcyap1r1), and their co-expression with the small-molecule transmitters, glutamate, and GABA, using probes for the expression of mRNAs encoding the vesicular transporters VGLUT1 (Slc17a7), VGLUT2 (Slc17a6), and VGAT (Slc32a1), in mouse brain. We then examined the distribution of PACAP/PAC1 hubs within well-established sensory input-to-motor output pathways passing through the cognitive centers, within the context of glutamate/GABA neurotransmission. This systematic analysis has revealed several possible PACAP-dependent networks involved in sensory integration allowing environmental cues to guide motor output.

Results

We have studied Adcyap1 and Adcyap1r1 co-expression with Slc17a7, Slc17a6, and Slc32a1 in the mouse brain, with precise region and subfield identification, using a sensitive dual ISH (DISH) method. Figure 1; Figure 1—source data 1 (Adcyap1), Figure 2 (Adcyap1r1) show examples using this method, which unambiguously labels the co-expression of two mRNAs at the single-cell level for light microscopical examination. At the light microscopical level, facile low- and high- magnification switching allows detailed serial high-power images to be located in a global histological context for precise delineation of anatomical regions/subfields as well as their rapid photo-documentation. Single-cell co-expression of two mRNA targets can be clearly observed by light microscopy with both low and high magnification.

Examples of histological samples using the sensitive dual in situ hybridization (DISH) method that can label unambiguously the co-expression of two RNAs at single-cell level for light microscope examination.

(A and C) Adcyap1 (RNA coding for PACAP) co-expression with Slc17a6 (RNA coding for VGLUT2) in MEPO and BAC respectively. (B) High magnification of the MEPO region of (A). Double arrows indicate cells that co-express both mRNAs. This feature is better appreciated on the cells in which Adcyap1 is weakly expressed (weak staining), so the Slc17a6 staining can be clearly seen as independent dots. (D) Brain stem Koelliker Fuse (KF) nucleus of the parabrachial complex is another main PACAP-expressing nucleus. Adcyap1 was co-expressed intensely with Slc17a7 (RNA coding for VGLUT1). Double arrows indicate cells co-expressing both mRNAs. Panels E-H show two cerebellar regions, paraflocculus (E and F) and central (G and H) lobules, under low and high magnification, respectively, where the Adcyap1 expression was higher than rest of regions. Purkinje cells are the main GABAergic (expressing Slc32a1, RNA encoding VGAT) PACAP containing neurons, distributed in all regions of cerebellar cortex. Some GABAergic cells in granule cell layer of paraflocculus and central regions also co-expressed Adcyap1 and Slc32a1 (indicated with double pink arrowheads, see also SI-Fig.O1). In these two cerebellar regions, some granule cells also expressed Adcyap1 (indicated with single blue arrows). Nissl staining was used for counterstaining. Note: this figure contains excerpts from the more comprehensive figure supplement.

Figure 1—source data 1

Comprehensive DISH mapping of PACAP co-expression with VGLUT1, VGLUT2, and VGAT throughout mouse brain reveals an extensive distribution and diversity of cell types.

Adcyap1, PACAP-mRNA mapping within glutamatergic and GABAergic subpopulations of mouse brain. Slc17a7, Slc17a6, and Slc32a1: mRNAs encoding the VGLUT1, VGLUT2, and VGAT with respective color coding for the chromogens labeling the corresponding mRNAs. Double arrows indicate examples of colocalization of two probe mRNAs and single arrows indicate no-co-expression in given cells. Panels A–O: coronal and sagittal sections taken from the indicated Bregma or medio-lateral coordinates. The combination of mRNA probes is indicated with respective colors corresponding to the DISH method end products. The high-magnification panels are examples with relevant molecular features, of the regions labeled within low-magnification photomicrographs (see the abbreviations vide infra). Note that the high-magnifications photos are not from the same experiment of the low magnification. 3 v: third ventricle; ACA: anterior cingulate area; ac: anterior commissure; AHN: anterior hypothalamic nucleus; AI: agranular insular area; AONpv: anterior olfactory nucleus, postero-ventral; AUD: auditory areas; AVPV: antero-ventral periventricular nucleus; BAC: bed nucleus of anterior commissure; BSTpl: Bed nucleus of the stria terminalis, posterolateral; CA3v: ventral hippocampal formation, CA3; CBpj: purkinje layer of the cerebellum; CENT: central lobule of the cerebellum; COA: cortical amygdala; CSm: superior central nucleus raphe, medial part; CSl: superior central nucleus raphe, lateral part; DCO: dorsal cochlear nucleus; DR: Dorsal raphe nucleus; DMHa: dorsomedial nucleus of the hypothalamus, anterior part; ECT: ectorrinal area; ENT: entorrinal area; FL: flocculus; NLOT: nucleus of the lateral olfactory tract; gcl: granule cell layer; GU: gustatory area; IC: inferior colliculus; ICe: inferior colliculus external nucleus; IF: interfascicular nucleus of the raphe; IL: infralimbic area; IPN: interpeduncular nucleus; ISN: inferior salivatory nucleus; KF: Koelliker-Fuse subnucleus; LH: lateral habenula; ISN: inferior salivatory nucleus; LPO: lateral preoptic area; MD: mediodorsal nucleus of the thalamus; MEA: medial amygdala; MEPO: median preoptic nucleus; MGm: medial geniculate complex, medial part; MH: medial habenula; MM: medial mammillary nucleus; MOB: main olfactory bulb; MOp: primary motor area; MOs: supplemental motor area; MPO: medial preoptic area; MRN: midbrain reticular nucleus; NTS: nucleus of the tractus solitarius; opt: optic tract; ORB: orbital area; OV: vascular organ of lamina terminalis; PA: posterior amygdalar nucleus; PAG: periaqueductal gray; PB: parabrachial nucleus; PCG: contine central gray; PG: pontine gray; PH: posterior hypothalamic nucleus; PIR: piriform area; PL: prelimbic area; POL: posterior limiting nucleus of the thalamus; PP: peripeduncular nucleus; PPN: pedunculo pontine nucleus; PRN: pontine reticular nucleus; PSTN: parasubthalamic nucleus; PSV: principal sensory nucleus of the trigeminal nerve; RR: midbrain reticular nucleus, retrorubral area; RSPd: retrosplenial area dorsal; RSPv: retrosplenial area dorsal; PVH: periventricular hypothalamic nucleus; PVi: periventricular hypothalamic nucleus, intermediate part; RM: raphe magnus; SC: superior colliculus; SCm: superior colliculus, motor related; SFO: subfornical organ; SPF: subparafascicular nucleus; STN: subthalamic nucleus; SUBd: subiculum dorsal part; SGN: suprageniculate nucleus; SOC: superior olivary complex; SUM: supramammillary nucleus; TEa: Temporal association area; TM: tuberomammillary nucleus; VCO: ventral coclear nucleus; VISam: anteromedial visual área; VISp: primary visual area; VISal: anterolateral visual area; VISl: lateral visual area; VISC: visceral area; vHi: ventral hippocampus; VMH: ventromedial hypothalamic nucleus; VMHc: ventromedial hypothalamic nucleus, central part; VMHdm: ventromedial hypothalamic nucleus, dorsomedial part; VMHvl: ventromedial hypothalamic nucleus, ventrolateral part; VTA: ventral tegmental area; ZI: zona incerta.

https://cdn.elifesciences.org/articles/61718/elife-61718-fig1-data1-v2.pdf
Examples illustrating Adcyap1r1 (RNA encoding PAC1) co-expression with glutamatergic (Slc17a7-VGLUT1 and Slc17a6-VGLUT2 expressing) and GABAergic (Slc32a1-expressing) neurons in cortical and subcortical regions.

(A and A') Temporal hippocampal formation where the Adcyap1r1 was strongly expressed in the principal neurons (pyr: pyramidal layer and DGgcl: dentate gyrus granule cell layer) as well as the VGLUT1+ mossy cells in the hilar region. Double arrows show single cells co-expressing Adcyap1r1 and Slc17a7. (B) Single-cell Adcyap1r1 co-expression with Slc17a6 (double arrowheads) was observed in ACA and (C) in the BAC. Regarding the GABAergic neurons expressing Adcyap1r1, the structures in the striatum and pallidum hosted very intensely expressing structures. Panel (D) shows the LS and BST in its three anterior divisions, anteromedial (BSTam), antero-lateral (BSTal) and oval (BSToval), as well as caudate-putamen (CP) with strong Adcyap1r1 expression. Panel (E) shows high-magnification photomicrograph where green dots (Adcyap1r1, PAC1 labeling), are mostly overlapped with red staining (Slc32a1, VGAT expression). Double pink arrowheads indicate co-expression within a single cell and single green arrows indicate cells only expressing Adcyap1r1. Panel (F) shows the amygdaloid complex and neighboring regions where Adcyap1r1 was strongly expressed in the GABAergic cell populations; (G) High-magnification photomicrograph showing that the Adcyap1r1 is exclusively expressed in Slc32a1 (VGAT) expressing neurons in the CEAc, while in the BLA it was expressed in the sparsely distributed GABAergic neurons as in most of the non-VGAT expressing neurons. 1. zona incerta of hypothalamus; 2. lateral hypothalamic area; 3: reticular nucleus of the thalamus; 4. globus pallidus; 5. caudate-putamen; 6: central amygdalar nucleus, medial part (CEAm); 7: lateral part (CEAl); 8: capsular part (CEAc); 9: basolateral amygdalar nucleus (BLA) 10: medial amigdalar nucleus; 11: intercalated nucleus of the amygdala; 12: basomedial nucleus of the amygdala; 13: cortical amygdalar area; 14: dorsal endopiriform; 15: ventral endopiriform; 16: piriform area. Fiber tracts: sm: stria medullaris; ac: anterior commissure; ic: internal capsule; opt: optic tract. lv: lateral ventricle.

Comprehensive DISH mapping of Adcyap1 co-expression with Slc17a7, Slc17a6, and Slc32a1 throughout mouse brain reveals an extensive distribution and diversity of cell types

Table 1 describes the distribution, cell types and relative expression strength, within 180 identified Adcyap1 positive cell groups/subfields co-expressing vesicular transporters, organized hierarchically by grouping the regions according to their embryonic origins. These include 58 regions derived from cortical plate, 6 regions derived from cortical subplate, 9 regions within cerebral nuclei, 20 regions in thalamus, 2 regions in epithalamus, 26 regions in hypothalamus, 18 regions in midbrain, 16 regions in pons, 20 regions in medulla, and 5 regions in cerebellum. We semiquantitative scoring is specified in the Materials and method section. Briefly, our annotation criteria were the percentage of expressing cell/total Nissl-stained nuclei: '-', not observed; '+', weak (<20%); '++', low (20%–40%); '+++', moderate (40%–60%); '++++', intense (60–80%); '+++++', very intense (>80%). Functional neuroanatomy order and annotations are based on Allen Institute Mouse Reference Atlas (http://atlas.brain-map.org/). To compare with the previous comprehensive report for PACAP distribution in rat brain published in 2002 (Hannibal, 2002), a column containing the data published previously in rat is displayed. Most of the regions described as Adcyap1-expressing in the rat were also found positive in our study in mouse, albeit strength of expression in several regions differs substantially between the two rodent species. Eleven regions that were reported negative, labeled as ‘-’ from original publication, were found positive with this sensitive DISH method (indicated in the table). An additional 122 regions, which were not reported in detail in the previous paper (labeled in the table as ‘n/r’), were found to co-express Adcyap1 and either a glutamate or GABA vesicular transporter mRNA.

Table 1
Distribution, cell types, and strength of main PACAPergic cell groups in mouse brain with comparison of rat brain reported by Hannibal, 2002*.
Cell group / sub-fieldHannibal, 2002Slc17a7
(VGLUT1)
Slc17a6
(VGLUT2)
Slc32a1
(VGAT)
Retina
Ganglion cell layer+-+++-
Cerebrum: Cortical plate
Olfactory area
Main olfactory bulb
Granular cell layer----
Inner plexiform layer----
Mitral cell layer+++++++
Outer plexiform layern.r.+++++++
Glomerular layer-+++
Periglomerular cellsn.r.+++
 Accessory olfactory bulb
Mitral cell layer++++++++
Glomerular layern.r.+++
Granular layern.r.+--
 Other olfactory areas
Ant olfactory n. lateral++++++++-
Ant olfactory n. medial++++++++-
Dorsal peduncular arean.r.+++++++
Taenia Tectan.r. +++--
Piriform area: Pir2n.r.+--
Piriform area: Pir3n.r.+++++-
N. lat. olfactory tract (NLOT)+++++++++++
Cortical amygdalar area (CoA)n.r.+++++-
 Hippocampal formation
Hippocampal region
Dorsal dentate gyrusn.r.---
Dorsal hippocampus CA1+---
Dorsal hippocampus CA2n.r.+--
Dorsal hippocampus CA3+---
Dorsal hilusn.r.++--
Ventral dentate gyrusn.r.---
Ventral CA3vvn.r.+++++--
Ventral hilusn.r.++--
Retrohippocampal regions
Entorhinal arean.r.++-
Parasubiculum++++++++-
Postsubiculum++++++++-
Presubiculumn.r.++++++-
Subiculumn.r.++++++-
 Isocortex§
Layer I+n.an.an.a
Layer II-II++n.an.an.a
Layer IV-n.an.an.a
Layer V++n.an.an.a
Layer VI+n.an.an.a
Agranular insular cortexn.r.++++++-
 Somatomotor areas
2ry motor area, layer 2–3n.r.+++--
2ry motor area layer 5n.r.+++++++
1ry motor area, layer 2–3n.r.+++--
1ry motor area, layer 5n.r.+++++++
 Orbital frontal cortex (OFC)
OFC 1n.r.++++-
OFC 2/3n.r.++++-
OFC 5n.r.+++--
 Prefrontal cortex (PFC)
Ant cingulate cortex (ACC):
ACC 2/3:
ACC 5:
n.r.
n.r.
++++
+++
+
+
-
-
Prelimbic (PL)
PL 2/3
PL 5
n.r.
n.r.
++++
+++
+
+
-
-
Infralimbic (IL)
IL 2/3
IL 5
n.r.
n.r.
++++
+++
+
+
-
-
Cell group / sub-fieldRat Hannibal
JCN, 2002
Slc17a7
(VGLUT1)
Slc17a6
(VGLUT2)
Slc32a1
(VGAT)
Prim somatosensory a. SSp,
SSp 1
SSp 2/3
SSp 4 (mouth)
SSp 5
SSp 6a
-
-
n.r.
-
-
+
+++
++
++
++
-
-
-
-
-
-
-
-
-
-
Gustatory areasn.r.++++--
Auditory arean.r.+++--
Visual arean.r.+++--
Visceral arean.r.+++--
Temporal association arean.r.+++--
Ectorhinal arean.r.+++--
Perirhinal arean.r.++--
Retrosplenial arean.r.++++--
Post parietal association arean.r.++++--
 Cortical subplate
Claustrumn.r.++-
Endopiriform nucleusn.r.++-
Lateral amygdalar nucleusn.r.++++++-
Post amygdalar nucleus (PA)n.r.+++++-
Basomedial amygdala-+ +-
Basolateral amygdala+++-
 Cerebral nuclei
 Striatum
Lateral septal nucleusn.r.++++-
Anterior amygdala arean.r.++-
Central amygdalar nucleus+--+
Medial amygdalar nucleus+++++++++
 Pallidum
Bed nucleus of Stria Terminalis (BST)+n.an.an.a
BST ovaln.r.-++
BST amn.r.-++++
BST dmn.r.-++
BST prn.r.+++++
Bed nucleus of anterior commissuren.r.++++++++++-
 Brain stem, inter-brain
 Thalamus
Somato-motor related
Subparafacicular nucleus, magnocellular partn.r.+++-
Subparafacicular arean.r.-+++-
Peripeduncular nucleusn.r.-+++-
Medial geniculate complexn.r.-+++-
Polymodal association cortex related
Lat. Posterior n. thaln.r.+++-
Post. Limiting nucleusn.r.+++-
Suprageniculate n.n.r.+++-
Anterodorsal n.-++++-
Anteromedial n.n.r.++++-
Parataenial n.n.r.++++-
Intermedial n.n.r.++-
Laterodorsal n.n.r.+++-
Centrolateral nn.r.-++-
Intermediodorsal n.n.r.+++-
Mediodorsal n.n.r.++++-
Pariventricular n.-++++-
Parateanial n.n.r.+++-
N. of reuniens-++++-
Posterior pretectal n.+++-+++-
Precommissural n.+++-+-
 Cell group / sub-fieldRat Hannibal JCN, 2002Slc17a7
(VGLUT1)
Slc17a6
(VGLUT2)
Slc32a1
(VGAT)
 Epithalamus
Medial habenula++++++++++++-
Lateral habenula++++-++++-
 Hypothalamus
Paraventricular n+-+-
Periventricular n+-++-
Anterodorsal preoptic n.n.r.-+-
Anteroventraln.r.-+++-
Dorsomedial n.+++-+++-
Median preoptic n.+++-++++-
Medial preoptic area+++-++-
Vascular organ of lamina terminalis+++-++++-
Posterodorsal preoptic n.n.r.-+-
Subfornical organ++++-++++-
Lateral preoptic arean.r.-++-
Anterior hyp. area++-++-
Premammillary n.n.r.-+++
Lateral mammillary n.++++-++++-
Medial mammillary n.--+++-
Supramammillary n.--+++
Median preoptic n.++-++-
Lateral hyp. area++-++-
Preparasubthalamic n.n.r.-+++-
Parasubthalamic n.n.r.-+++-
Subthalamic nucleus--+++++-
Retrochiasmatic arean.r.-+++-
Tuberomammillary nucleus--+++
Zona incerta+-++-
Ventromedial hyp. n++++-+++++-
Post. hypothalamic n.n.r.-+++-
 Midbrain
Sensorial related
Inf. colliculus (IC), central and external n.n.r.-++-
N. of the brachium of ICn.r.-++-
N. saculumn.r.-+-
Parabigeminal n.n.r.-+-
Midbrain trigeminal n.n.r.-++-
Motor related
Ventral tegmental arean.r.-++-
Midbrain reticular n.n.r.-+-
Superior colliculus, motor relatedn.r.-+++-
Periaqueductal grayn.r.-+++-
Cuneiform n.n.r.-++-
Edinger-Westphal n.n.r.-+-
Interfascicular n. Raphen.r.-++-
Behavior state related
Midbrain raphe nuclein.r.---
Pedunculopontine n.n.r.-++-
Dorsal n. raphen.r.---
Central linear n. raphen.r.-++-
Rostral linear n. raphen.r.---
Olivary pretectal nucleusn.r.-+++-
 Cell group/sub-fieldRat Hannibal JCN, 2002Slc17a7
(VGLUT1)
Slc17a6
(VGLUT2)
Slc32a1
(VGAT)
 Hindbrain
 Pons
Sensory related
N. lateral lemniscusn.r.-+-
Principal sensorial nucleus of trigeminal nerve--+-
Koelliker-Fuse subnucleusn.r.++++--
Parabrachial n. lateral div.n.r.-++++-
Parabrachial n, rest subfieldsn.r.-+++-
Superior olivary comp (lat)n.r.+--
Motor related
Tegmental reticular n.n.r.-+++-
Barrington’s nucleusn.r.-+++-
Dorsal tegmental n.n.r.-+-
Pontine grayn.r.-+++-
Pontine central grayn.r.-+-
Supratrigeminal nucleusn.r.-+-
Behavior state related
Locus Coerulus (state)++++++-
Laterodorsal tegmental n.+-+++-
Pontine reticular n.n.r.-+-
Superior central n. raphen.r.-++
 Medulla
N. tractus solitarii medial+++++++++-
N. tractus solitarii lateral+++-++++-
Hypoglossal (XII) n.-++-
Dorsal motor n. of the vagus nerve (X)++++++--
Dorsal cochlear n.+++++++-
Ventral cochlear n.n.r.++++-
Spinal n. trigeminaln.r.-++-
N. prepositusn.r.-++-
Inferior salivatory complexn.r.-++-
Facial motor n. (VII)n.r.-++-
N. ambiguus+++-++-
Magnocellular reticular n.n.r.-++-
Parapyramidal n.n.r.-++-
Spinal vestibular n.+++-++-
N. Xn.r.-+-
N. raphe magnus (state related)n.r.-++-
N. raphe pallidus (state related)n.r.-++-
N. raphe obscurus (state rel.)n.r.-++-
Cuneate n.-++++-
Inferior olivaryn.r.-++-
 Cerebellar cortex
Purkinje’s cells++--+++++
Golgi’s cellsn.r.--+
Granule cells-++--
 Cerebellar nuclei
Interposed n.+++--
Dentate n.n.r.-+-
  1. n.a.: not applicable.

    n.r.: not reported (blue color text refers to Hannibal, 2002) rat PACAPergic cell group and expression strength analysis.

  2. *Similar semiquantitative annotations are used here the percentage of expressing cell/total Nissl stained nuclei: '-', not detectable; '+', weak (<20%); '++', low (40–20%); '+++', moderate (60–40%); '++++', intense (80–60%); '+++++', very intense (>80%).

    † Functional neuroanatomy order and annotations are based on Allen Institute Mouse Reference Atlas.

  3. ‡ Circadian oscillating expression (Lindberg et al., 2019).

    § Isocortex expression was regionally evaluated.

  4. ¶ Dorsal half of the MHb which co-express Calb2 (RNA encoding calretinin).

    ** Prominent in lobules paraflocculus, central and uvula. Coincide with calretinin (Calb2) expression.

A whole brain mapping of Adcyap1expression with relevant brain regions/subfields co-expression features is presented in the Figure 1—source data1.

Mapping of Adcyap1r1 co-expression with Adcyap1, Slc17a7, Slc17a6, and Slc32a1 suggests the PACAP-PAC1 system can function in autocrine and paracrine modes

Adcyap1r1 expression was studied in 152 mouse brain regions. In Figure 3, panels A–F, we show the semi-quantitative expression levels of Adcyap1r1, based on microscopic observation as different intensities of blue shading. Adcyap1 expression was also symbolized with either red or green dots (VGAT vs VGLUT mRNA co-expression) in corresponding regions. Contrasting with the discrete expression of Adcyap1, Adcyap1r1 expression was diffuse and widespread. Adcyap1r1-positive cells co-expressed Slc17a7 in the temporal hippocampus (Figure 2 panels A and A'), anterior cingulate area (ACA, panel B) and bed nucleus of anterior commissure (BAC, panel C) and Scl32a1 in pallidum and striatum structures (Figure 2D–G and Table 2). Almost all the Adcyap1-expressing neurons we studied co-expressed Adcyap1r1 (Figure 3G, from ACA and H from medial preoptic nucleus, MEPO). Besides, most neurons neighboring Adcyap1-positive cells also expressed Adcyap1r1 (single arrows). These observations suggest that the PACAP/PAC1 pathway may use autocrine and paracrine mechanisms in addition to classical neurotransmission through axon innervation and transmitter co-release.

Figure 3 with 2 supplements see all
Adcyap1r1 expression assessment in relation to Adcyap1 expression suggests that the PACAP-PAC1 system uses autocrine, paracrine, and neuroendocrine modes for signal transduction.

(A–F) Mapping of Adcyap1r1 (RNA encoding Pac1) expression (symbolized by the intensity of blue shading) in six septo-temporal planes in relation to main PACAP containing brain nuclei and subfields, based on microscopic observations. Green and red dots represent Adcyap1 expressing neurons of glutamatergic (VGLUTs mRNA expressing) or GABAergic (VGAT mRNA expression) nature, respectively. Shaded regions with different blue intensity symbolize the strength of Adcyap1r1. For abbreviations see the corresponding table for abbreviations in Appendix 1. 1. MOB; 2. SFO; 3. MEPO; 4. OV; 5. PVT; 6. MD; 7. RE; 8. MBO; 9. SUM; 10. PH; 11. IF; 12. EW; 13. RL; 14. PAG; 15. DR.; 16 SCm; 17. IC; 18. AP; 19. NTS; 20. XII; 21. IO; 22. RPA; 23. RM; 24. GRN; 25. PRNr; 26. tegmental reticular n.; 27. PG; 28. IPN; 29. CLI 30. UVU; 31. ORB; 32. ACAd1; 32b. ACAd6; 33. RSP; 34 DP; 35. TTv; 36. LSr; 37. MS; 38. NDB; 39. LSc; 40. ACB; 41. OT; 42. ILA; 43. PL; 44. MOs; 45. MH; 46. BST.; 47. BAC; 48. MPO; 49. PVH; 50. SCH; 51. SO; 52a. DMH; 52b. VMH; 53. PVp; 54. VTA; 55. AM; 56. PT; 57. SCs; 58. PCG; 59. MV; 60. FN.; 61. CU; 62. SPIV.; 63. PB; 64. IRN; 65. LC; 66. SOC; 67. MARN; 68. MDRN; 69. MRN; 70. SN; 71. LM; 72. VLH; 73. LPO; 74. PHA; 75. AV; 76. AD; 77. RT; 78. PRC; 79. PF; 80. CP; 81. SI; 82. AON.; 83. AOB; 84. MOBgr; 85. MOBgl; 86: MOBml; 87.DG-gcl; 88. BSTov; 89. CBpj; 90. CBgcl; 91. CA3vv; 92. CA3v, 93. vhil; 94. MoV; 95. sV; 96. spV; 97. NLL; 98. PBG; 99. MG; 100. SPF; 101. PP; 102: ZI; 103. STN; 104. SUB; 105. MEApv; 106. PA; 107a. COAa; 107b. COAp; 108. NLOT; 109. AAA; 110. CEAm; 111. MEAad; 112. MEpd; 113. SI; 114. GPe; 115. GPi (entopeduncular nucleus); 116. FS; 117. EP; 118. CLA; 119. AI; 120. SS; 121. PTLp; 122. VIS; 123. DH; 124. POST; 125. PRE; 126. DCO; 127. VCO; 128. PAR; 129. PIR; 130. TT; 131. CEAc; 132. IA, 133. BMA; 134. CA2v; 135. LGv; 136. GU; 137. VISC; 138. ECT; 139. ENTl; 140. ENTm; 141. PAA; 142. TR; 143. BLA; 144. LA 145. CEAl 146. AMB; 147. OP; 148: PM.; 149. LGd; 150. IGL; 151. FL; 152. AN. Aq: aqueduct; och: optic chiasm; v4: forth ventricle; mlf: medial longitutinal fasciculus; cc: corpus callosum; vhc: ventral hippocampus commissure; fi/fx: fimbria/fornix; pyr: pyramidal layer; lot: lateral olfactory tract, mcp: middle cerebellar penducle; st: stria terminalis; opt: optic tract; ic: internal capsule; tb: trapezoid body; arb: arbor vidae. (G and H) Examples illustrating autocrine and paracrine features of PACAP-PAC1 signaling that Adcyap1r1 was expressed in PACAP containing (Adcyap1 expressing) neurons. (G) ACA in prefrontal cortex and (H) MEPO. Double arrowheads indicate co-expression and blue arrows indicate the Adcyap1r1 expressing neurons which are not Adcyap1 expressing but were adjacent to them.

Figure 3—source data 1

DISH mapping of Vipr1 co-expression with VGAT in selective brain regions.

Each panel from A to J show a low-magnification image of the coronal section analyzed, indicating with arrows the regions where the corresponding high-magnification photomicrographs were taken. The red signal corresponds to SLC32a1 (the mRNA for VGAT) and the green signal correspond to Vipr1 (the mRNA for VPAC1). The abbreviatures correspond to the Allen Brain Map and are indicated in the Figure 3—source data 5, where a comparison with the expression observed in the Vipr1 ISH experiments from Allen (73927619 and 77924538) and a semiquantitative analysis of the co-expression with VGAT mRNA was done. Scale bar: 2 mm for low-amplification and 50 µm for high-amplification photomicrographs.

https://cdn.elifesciences.org/articles/61718/elife-61718-fig3-data1-v2.pdf
Figure 3—source data 2

DISH mapping of Vipr2 co-expression with VGAT in selective brain regions.

Each panel from A to J show a low-magnification image of the coronal section analyzed, indicating with arrows the regions where the corresponding high-magnification photomicrographs were taken. The red signal corresponds to Slc32a1 (the mRNA for VGAT) and the green signal correspond to Vipr2 (the mRNA for VPAC2). The abbreviations correspond to the Allen Brain Map and are indicated in the Figure 3—source data 5, where a comparison with the expression observed in the Vipr2 ISH experiments from Allen (1104 and 1105) and a semiquantitative analysis of the co-expression with VGAT mRNA was done. Scale bar: 2 mm for low-amplification and 50 µm for high-amplification photomicrographs.

https://cdn.elifesciences.org/articles/61718/elife-61718-fig3-data2-v2.pdf
Figure 3—source data 3

Density of VipR1 mRNA expressing cells in the mouse brain: analysis of VGAT mRNA co-expression and comparison with data from Allen Brain Atlas.

https://cdn.elifesciences.org/articles/61718/elife-61718-fig3-data3-v2.docx
Figure 3—source data 4

Density of VipR2 mRNA expressing cells in the mouse brain: analysis of VGAT mRNA co-expression and comparison with data from Allen Brain Atlas.

https://cdn.elifesciences.org/articles/61718/elife-61718-fig3-data4-v2.docx
Figure 3—source data 5

Density distribution of PAC1 expressing cells in selective cortical regions.

https://cdn.elifesciences.org/articles/61718/elife-61718-fig3-data5-v2.docx

PACAP also binds to two other G protein–coupled receptors highly related to PAC1, called VPAC1 (Vipr1), and VPAC2 (Vipr2) (Harmar, 2001). In all the regions where Adcyap1r1 was expressed, the expression of mRNA for either or both VIP receptors (Vipr1 and Vipr2) was also found. To simplify this already extensive report, we present the data for these two receptors in Figure 3—source data 14 .

Distribution and glutamatergic/GABAergic vesicular transporter mRNA co-expression with Adcyap1 and Adcyap1r1 suggests a broad function for PACAP signaling in sensorimotor processing system(s)

Retina

Retinal ganglion cells (RGC) have been reported to express PACAP at various levels of abundance previously in the literature. In rat, Adcyap1 was reported at a low level (‘+') (Hannibal, 2002) within the RGC population. CD1 mice were reported to express PACAP in retina (Kawaguchi et al., 2010), and in Adcyap1 promoter-EGFP reporter mice, EGFP expression was reported to be low (‘+') (Condro et al., 2016). With the DISH method employed here, we found a higher percentage of RGCs co-expressing Adcyap1 and Slc17a6 than previously reported (Figure 1—source data 1A1). Expression levels of Adcyap1 oscillate daily from 50% to 80% with highest levels during subjective night (see Lindberg et al., 2019 for details).

Cerebral cortex: structures derived from cortical plate

Olfactory area

High levels of PACAP expression in the olfactory area have been previously reported (Hansel et al., 2001). Here, we report the subfields of olfactory area Adcyap1-expressing neurons in detail. In the main olfactory bulb (MOB, Figure 3B, area 1), Adcyap1 was intensely expressed in outer plexiform (OPL) and mitral layers. In OPL we observed the co-expression with Slc17a7, Slc17a6 and Slc32a1. (Table 1 and Figure 1—source data 1A2 and insets). Other cell types in the internal plexiform cell layers expressed Adcyap1 at low levels with mixed glutamate/GABA molecular signatures (see Table 1). In contrast, in the accessory olfactory bulb (AOB, Figure 3C, area 83), Adcyap1 was mainly expressed in the mitral cell layer with co-expression of Slc17a7, Slc17a6 (intense), and Slc32a1 (weak) mRNAs (Table 1). Other olfactory areas intensely expressing Adcyap1 and Slc17a7 were AON (layer 1, Figure 3C, area 82), TT (Figure 3C, area 34b), DPA (Figure 3C, area 34a), Pir (layer 3, Figure 3D and E, area 129), NLOT (layer, Figure 3D, area 108; SI Figure 1F). The COA (layer, Figure 3D, areas 107a and 107b) co-expressed Adcyap1 and Slc17a6 (Figure 1—source data 1F and F5).

Isocortex

PACAP’s role in isocortex has in general been little studied (Zhang and Eiden, 2019). Moderate expression of Adcyap1 was initially reported in the cingulate and frontal cortices, with lower concentrations found in other neocortical areas using radiolabeled riboprobe ISH (Mikkelsen et al., 1994). Hannibal subsequently reported that Adcyap1-expressing cells were observed mainly in layers 1–3 and layers 5–6, and PACAP‐IR nerve fibers in all layers of the cerebral cortex; however, no detailed information about the differential expression levels across cortical regions was presented (Hannibal, 2002).

In our study, we found intense expression of Adcyap1 in isocortex to be in the frontal pole of the telencephalon, including prelimbic, infralimbic and anterior cingulate and orbital area (Figure 3, A and B, areas 43, 42, 32a and 32b, and 31, respectively). Approximately 80% of the neuronal population of the layer 2 and layer 5 co-expressed Adcyap1 and Slc17a7. A significant population of Adcyap1-expressing cells in the layer 5 of prefrontal cortices co-expressed Slc17a6 or Slc32a1 (Table 1 and Figure 1—source data 1B and B1).

In the primary and secondary motor cortices (MOp and MOs, Figure 3B, area 44), Adcyap1 was found expressed in layer 2/3 and layer 5. This pattern was also observed in somatosensory, gustatory, auditory, visual, visceral, temporal association, ectorhinal, perirhinal, retrosplenial, and post-parietal association areas (see Table 1 for more cortical area expression and strength and Figure 1—source data 1A–K, low-magnification panels).

Adcyap1r1 expression in neocortex was widespread with more homogenous aspects concerning the different cortical areas, except that in the ACA and the entorhinal cortex layers 2–3 and layers 5–6 showed very intense expression levels (Figure 3, panels B and F, areas 32 and 32a, 139 and 140) and https://gerfenc.biolucida.net/images?selectionType=collection&selectionId=98. We sampled eight neocortex regions at two coronal levels, Bregma 0.14 mm and Bregma 1.7 mm, where we observed that more than 80% of neurons in layers 2–3 and layer 5 expressed Adcyap1r1 (Figure 3—source data 5). As approximately 20% of cortical neurons are GABAergic (Ascoli et al., 2008), we tested three of the main GABAergic cell types in these cortical regions, finding that in the selected cortical areas we sampled, all of somatostatin (Sst), parvalbumin (PV) and corticotropin releasing hormone (CRH) neurons co-expressed Adcyap1r1 (Figure 3—figure supplement 1 F,G,H).

Hippocampal formation

In the mouse dorsal (septal pole) hippocampal formation, in contrast to data obtained in rat (Figure 3—figure supplement 2) and from the PACAP-EGFP transgenic reporter mouse (Condro et al., 2016), we did not find Adcyap1-expressing cells in cell body layers of CA1, CA3, and DG, as previously reported (Hannibal, 2002; Condro et al., 2016). However, we report here the marked and selective expression of Adcyap1 in pyramidal neurons of the CA2 region (Table 1 and Figure 4C, left inset). Adcyap1r1 expression was observed to be low in CA subfields and Adcyap1r1 was selectively expressed in Slc32a1-expressing cells (Figure 3—figure supplement 1, A and B). In contrast, the DG-GCL had very intense expression level of Adcyap1r1, among all brain regions, in both Slc17a7- and Slc32a1-expressing cells (Figure 2A and Figure 3—figure supplement 1, A and C). In DG hilar region (polymorphic layer), we observed few cells co-expressed Adcyap1 and Slc17a7 (Figure 4A and D). These were mossy cells co-expressing calretinin mRNA (Calb2) (Figure 4B). The Adcyap1-expressing mossy cell quantity increased in the caudo-temporal direction. This population was also described in the previous reports (Hannibal, 2002; Condro et al., 2016), however, without identification of cell type, as reported here.

Ventral (temporal pole) hippocampus CA3 (CA3vv) contained a newly identified subset of pyramidal neurons distinguished by its molecular signatures of VGLUT1, PACAP and calretinin mRNA expression.

Low-magnification bright field coronal (A) and sagittal (C) whole-brain sections with ISH (RNAscope 2.5 High Definition (HD) Red Assay), showing the selective expression of mRNA of PACAP (Adcyap1) in a subset of CA3 of the temporo-ventral pole of hippocampus (red ovals circumscribed regions in A and C). Hilar mossy cells in the dorsal (A) and ventral (C) hippocampus, also Adcyap1 expressing, are circumscribed by blue ovals. (B) and inset show the corresponding coronal sections in low magnification of Calb2 (calretinin) and slc17a7 (VGLUT1) taken from Allen Brain Atlas (Ng et al., 2009) where CA3vv subset and hilar mossy cells are indicated with red and blue ovals, respectively). Right inset of C corresponds to calretinin mRNA expression in the same hippocampus sagittal squared region of C. Both the subset of CA3vv pyramidal neurons and the mossy cells co-expressed VGLUT1 mRNA (Slc17a7) and Adcyap1 (D). Left inset of C shows dorsal CA2 pyramidal layer expressed Adcyap1. The ‘trisynaptic-centric’ (E) vs ‘CA3-centric’ (F) view of hippocampal information processing, where the newly identified CA3vv subset of Adcyap1 and Calb2 containing glutamatergic neurons are presented in dark pink triangles and the mossy cells are in light pink circles. The Adcyap1r1 expressing granule cells (green) and interneurons (red) in the granule cell layer are symbolized with pink circle. Chartings were based on ventral pole of hippocampus (shaded region of atlas segment (Paxinos mouse brain), where this chemically distinct subset of CA3c pyramidal neurons was identified. Circuits were modified from Scharfman, 2007, with adaptation to the new finding from this study.

In the ventral (temporal pole) hippocampal formation, we identified two cell populations with Adcyap1 expression. One was the Slc17a7-expressing mossy cell population in the hilar region mentioned above, which was distributed from septo-dorsal to temporo-ventral hilus with increasing quantity (Figure 4C and Figure 1—source data 1J) Mossy cells are major local circuit integrators and they exert modulation of the excitability of DG granule cells (Scharfman and Myers, 2012; Sun et al., 2017). The DG granule cells strongly expressed Adcyap1r1 (Figure 3—figure supplement 1A and C). Glutamatergic hilar mossy cells of the dentate gyrus can either excite or inhibit distant granule cells, depending on whether they project directly to granule cells or to local inhibitory interneurons (Scharfman and Myers, 2012). However, the net effect of mossy cell loss on granule cell activity is not clear. Interestingly, dentate gyrus has a unique feature: there are two principal populations of glutamatergic cell type, the granule cells and the mossy cells. The former intensely expressed Adcyap1r1 and the latter intensely expressed Adcyap1, indicating that PACAP/PAC1 signaling may play a pivotal role for granule cell excitability.

A second population of Adcyap1-expressing neurons in ventral hippocampus was a subset of CA3 pyramidal neurons in the ventral tip, and have been addressed in the literature as CA3vv pyramidal neurons expressing the gene coch (Thompson et al., 2008; Fanselow and Dong, 2010). This population was previously photo-documented without comment in Figure 11J of the referenced report (Hannibal, 2002). This represents a distinct and novel group of pyramidal neurons in ventral CA3 (Figure 4A, C and D). These neurons strongly expressed Adcyap1, and co-expressed Slc17a7 (Figure 4, inset of B and D) as well as Calb2 (Figure 4B), with the rest of pyramidal neurons expressing Slc17a7 but neither Calb2 nor Adcyap1.

Retrohippocampal regions expressing Adcyap1 and either Slc17a7 or Slc17a6 were entorhinal area, prominently in the layer 5, parasubiculum, postsubiculum, presubiculum, and subiculum (Figure 3E and F, areas 139 and 140, 128, 124, 104). This latter region, subiculum, together with the pyramidal layer of dorsal CA1, CA2 and CA3, exhibited large differences in Adcyap1 expression strength between rat and mouse (Figure 3—figure supplement 2). Developmental studies of these regions, as well as extended amygdala, may indicate a recapitulation of phylogeny by development that is relevant to the evolution of PACAP neurotransmission across mammalian species (Zhang and Eiden, 2019).

Cerebral cortex: structures derived from cortical subplate

The subplate is a largely transient cortical structure that contains some of the earliest generated neurons of the cerebral cortex and has important developmental functions to establish intra- and extra-cortical connections (Bruguier et al., 2020). The concept of the subplate zone as a transient, dynamically changing and functional compartment arose from the combined application of functional and structural criteria and approaches (for a historical review see Judaš et al., 2010). Here, we adapt our classification from that of the Allen Brain Map (https://portal.brain-map.org/). Two noteworthy structures derived from the cortical subplate that expressed Adcyap1 and Adcyap1r1 are the claustrum (CLA) and the lateral amygdalar nucleus (LA).

The claustrum (CLA)

Owing to its elongated shape and proximity to white matter structures, the claustrum (CLA, Figure 3, panels D, E, F, area 118) is an anatomically well-defined yet functionally poorly described structure, once speculated to be the ‘seat of consciousness’ due to its extensive interconnections (Crick and Koch, 2005). CLA is located between the insular cortex and the striatum: it is a thin sheet of gray matter considered as a major hub of widespread neocortical connections (Bruguier et al., 2020). The CLA is recently reported to be required for optimal behavioral performance under high cognitive demand in the mouse (White et al., 2020). Consistent with recent work (White et al., 2018), rat CLA receives a dense innervation from the anterior cingulate cortex (ACA), one of the most prominent PACAP mRNA-expressing regions in frontal cortex, co-expressing Slc17a7 and Slc17a6 (Table 1 and Figure 3A, areas 32a and 32b, and G) and is implicated in top-down attention (Zhang et al., 2016b). The CLA interconnects the motor cortical areas in both hemispheres through corpus callosum (Smith and Alloway, 2010), where EGFP+ projections were reported in PACAP promoter-EGFP reporter mice (Condro et al., 2016). Expression of the neuropeptides somatostatin (SOM), cholecystokinin (CCK), and vasoactive intestinal polypeptide (VIP) has been reported in the rat CLA (Eiden et al., 1990). In mouse CLA, more than 80% of the neurons were Slc17a7- and Slc17a6-coexpressing and less that 20% of the neurons expressed Slc32a1. PACAP content in rodent CLA has not been reported. In our study, we observed 10–15% of glutamatergic cells of the CLA co-expressed Adcyap1 and almost 100% of cells expressed Adcyap1r1 (Figure 3D, area 118).

Endopiriform nucleus and amygdalar complex

The endopiriform nucleus (Figure 3E and F, area 117) and divisions of lateral (Figure 3F, area 144), basolateral (Figure 3F, area 143), basal medial (Figure 3E, area 133), and posterior amygdalar (Figure 3F, area 106, Figure 4C) nuclei are, from a phylogenetic point of view, olfactory structures (Groor, 1976) derived from cortical subplate which the main cell population is glutamatergic, co-expressing Slc17a7 and Slc17a6 (see Table 1). Adcyap1 was intensely expressed in the lateral (dorsal) amygdala (Figure 3F, area 144), anterior basomedial amygdala (Figure 3E, area 133), posterior amygdalar nucleus (area 106), and with low expression in the endopiriform nucleus (area 117) and basomedial amygdala posterior subnucleus (Figure 1—source data 1G) and weak expression in the basolateral amygdala (Figure 3F, area 143).

Structures derived from cerebral nuclei

The main structures expressing Adcyap1 in striatum were the lateral septum (LS, Figure 3B, areas 36 and 39) and medial amygdala (MEA, Figure 3D, areas 111 and 112) (SI Figure 1G). Most of these neurons coexpressed both Slc17a7 and Slc17a6 (see Table 1). In contrast to rat brain, where expression of Adcyap1 is prominent in central amygdala and intercalated cells (Figure 3—figure supplement 2, C), in the mouse Adcyap1-expressing cells were quite sparse in these structures (Figure 3—figure supplements 2C'). Adcyap1r1 was intensely expressed in these mainly Slc32a1-expressing structures (Figure 2D,E,F and G, Table 2).

In structures within pallidum, Adcyap1 was expressed, in order of abundance, in posterior, and anterior divisions of BST, and very weakly, in the oval nucleus (Table 1).

The very intense expression of Adcyap1r1 was observed on mainly GABAergic structures derived from cerebral nuclei. Figure 2D and G show examples illustrating Adcyap1r1- and Slc32a1-coexpressing neurons in some subcortical regions. Most of cells in the BST complex co-expressed Adcyap1r1 (Figure 2D and E). In BSTov, we tested the three main GABAergic cell types that co-express somatostatin (Sst), parvalbumin (Pvalb), and corticotropin releasing hormone (Crh) and found all the three types of neurons co-expressed Adcyap1r1 (Figure 3—figure supplement 1, I,J and K). Table 2 summarizes the distribution, cell types, and strength of expression of the main Adcyap1r1-expressing group in mouse cerebral nuclei.

Table 2
Distribution, cell types, and strength of main PAC1 expressing group in mouse cerebral nuclei (striatum and pallidum).
Cell group / sub-fieldSlc17a7
(VGLUT1)
Slc17a6
(VGLUT2)
Slc17a8
(VGLUT3)
Slc32a1 (VGAT)
Striatum
Caudoputamen
Nucleus accumbens---+++
Fundus of striadum---++
Olfactory tubercle---+
Lateral septum complex---++++
Medial amygdala (MeA)
MeAav-++++-+
MeApd-+-++++
Central amygdala (CeA)
CeA medial---+++
CeA lateral---+++
CeA capsular---++++
Anterior amygdala area---+++
Intercalated nucleus---++++
Cell group / sub-fieldSlc17a7
(VGLUT1)
Slc17a6
(VGLUT2)
Slc17a8
(VGLUT3)
Slc32a1 (VGAT)
Pallidum
Globus pallidum internal---++
Globus pallidum external---++
Globus pallidum ventral (VP)
Substantia innominata---++
Magnocellular nucleus---+++
Medial septal complex---+++
Bed nuclei stria terminalis (BNST)
BNSToval---++++
BNSTam---+++
BNSTdm---+++
BNSTpr---+++
Nucleus of Diagonal Band---++++
Bed nucleus of anterior commissure++--
  1. Semiquantitative annotations are used here the percentage of expressing cell/total Nissl stained nuclei: ‘-”, not detectable; '+”' weak (<20%); '++', low (40–20%); '+++', moderate (60–40%); '++++', intense (80–60%); '+++++', very intense (>80).

The bed nucleus of anterior commissure (BAC) is defined here, for the first time, as a major Adcyap1-expressing nucleus (Figure 1C, Figure 3B, area 47, Figure 5A, F, Si Figure 1E). BACs are bilateral triangular cell groups located on the dorsal corners of the chiasm of anterior commissure (ac) between the anterior part and posterior part (Papp et al., 2014). Most BAC neurons (90–95%) express Slc17a6 intensely (Figure 5B) and Slc17a7 weakly (Figure 5C) and around 5% express Slc32a1 (Figure 5D and inset). Calb2 is intensely expressed in BAC (Figure 5E), as in the case of the subset of CA3vv pyramidal neurons mentioned earlier (Figure 4A, C). The Adcyap1-expressing neurons were densely packed (Figure 1, C and Figure 5A). Almost all the Adcyap1-positive neurons co-expressed Slc17a6 mRNA (Figure 5F) and Slc17a7. Co-expression of Adcyap1 within the Slc32a1-positive neurons was not found in BAC (Figure 5G).

Bed nucleus of anterior commissure (BAC): a prominent PACAP containing glutamatergic nucleus chemo-anatomically identified.

(A) Two coronal sections at Bregma 0.02 mm and −0.10 mm of mouse brain showing Adcyap1 ISH (RNAscope 2.5 High Definition (HD) Red Assay) expressing BAC (ac: anterior commissure). Panels B–E are low-magnification photomicrographs taken from Allen Brain Atlas (Ng et al., 2009) showing the Slc17a6 (VGLUT2, (B)) Slc17a7 (VGLUT1, (C)) Slc32a1 (VGAT, (D) and inset), Calb2 (calretinin, (E)) expressed in BAC. The Adcyap1 expressing neurons were densely packed and co-expressed Slc17a6 (F) and we did not observe co-expression within the Slc32a1-expressing cells (G).

Brain stem

Interbrain

Thalamus

Adcyap is extensively expressed in thalamic nuclei (see Table 1), most prominently in suprafasicular nucleus (SPF, SI Figure 1I and I3), paraventricular nucleus of the thalamus (PVT, Figure 3A, area 5), medial geniculate complex (Figure 3D, area 99), the nucleus reuniens (Figure 3A, area 7), and the mediodorsal thalamic nucleus (MD) (Figure 3A, area 76) neurons co-expressing Slc17a6. The medial geniculate nucleus or medial geniculate body is part of the auditory thalamus and represents the thalamic relay between the inferior colliculus and the auditory cortex. The nucleus reuniens receives afferent input mainly from limbic and limbic-associated structures, mediating interactions between the hippocampus and medial prefrontal cortex important for spatial working memory (Griffin, 2015). It sends projections to the medial prefrontal cortex, the hippocampus, and the entorhinal cortex (Wouterlood et al., 1990; McKenna and Vertes, 2004), although there are sparse connections to many of the afferent structures as well. The prefrontal cortical-hippocampal connection allows regulation of neural traffic between these two regions with changes in attentiveness (Vertes et al., 2007) as well as in resilience to stress (Kafetzopoulos et al., 2018). All the thalamic nuclei that express Adcyap1 also express Adcyap1r1 (Table 1 and Figure 3A–E). The PVT and MD participate in many sensory information relays. In a recent study, their role in a key neural circuit for psychological threat-induced hyperthermia was reported (Kataoka et al., 2020). This circuit involves brain regions in the prefrontal pole, called the dorsal peduncular area (DP, Figure 3A and B, 34a, also called dorsal taenia tecta, TTd, a main Adcyap1-containing region mentioned in section The claustrum) that senses social stress and mediates increased body temperature in response to it (Lin, 2020). Neurons from the DP/TTd then project to and excite neurons in the dorsomedial hypothalamus (DMH, another Adcyap1-containing nucleus in hypothalamus, Table 1, Figure 3B, area 52a, and vide infra), which in turn sends neuronal projections to the rostral medullary raphé (rMR, also a Adcyap1-expressing nucleus, Table 1, Figure 3A, area 23, and description vide infra).

Epithalamus: habenula

Habenulae are bilateral triangular eminences of the stalk of the pineal gland, situated at the dorso-caudal end of the thalamus. Their medial divisions border the third ventricle. The habenula is considered as the relay hub where incoming signals from basal forebrain, including, diagonal band of Broca, lateral preoptic area, lateral hypothalamus, paraventricular nucleus, and entopeduncular nucleus, travel through the stria medullaris to habenula to be processed. The habenula then conveys the processed information to midbrain and hindbrain monoaminergic structures, such as ventral tegmental area, medial and dorsal raphe nuclei, and periaqueductal grey, through the fasciculus retroflexus. The habenula thus connects the cognitive-emotional basal forebrain to the modulatory monoaminergic area (Sutherland, 1982). Medial habenula (MHb, Figure 3B, area 45), was observed to express strongly Adcyap1 in the dorsal half, in cells which co-express Slc17a7 or Slc17a6 (Figure 1—source data 1G). In the lateral habenula (Figure 3B, area 78), the Adcyap1-positive neurons co-expressed Slc17a6, and were mainly located in the central nuclei of the lateral habenula (Figure 1—source data 1G), a region with rich input from hypothalamic peptidergic afferents including arginine vasopressin and orexin (Zhang et al., 2018). All those cells, both in lateral and medial habenula, co-expressed Calb2 (see Allen Brain Atlas for reference https://mouse.brain-map.org/experiment/show?id=79556662).

Hypothalamus

Using the sensitive DISH method, a total of 26 hypothalamic nuclei were found to express Adcyap1 (Table 1). Among the highest density Adcyap1-expressing cell clusters/nuclei (>80% of cells Adcyap1-positive) of hypothalamus are (numbers refer to Figure 3A, B, C): SFO (2, Figure 1—source data 1F), MEPO (3, Figure 1—source data 1C, D), OVLT (4, Figure 1—source data 1C), DMH (52a) and VMH (52b, Figure 1—source data 1G), STN (Figure 3D, 103, Figure 1—source data 1H), and lateral mammillary nucleus (71, Figure 1—source data 1I). All these Adcyap1-positive expressing cells co-expressed Slc17a6. Other hypothalamic regions listed in Table 1 had lower density of expression and were also Slc17a6-positive co-expressing, such as paraventricular hypothalamic nucleus (PVH, area 49, Figure 1—source data 1F). Slc32a1/Adcyap1-coexpressing cells were sparsely distributed mainly in the anterior hypothalamic area (AHA Figure 1—source data 1O3), supramammillary (Figure 1—source data 1O4), and tuberomammillary nuclei (Figure 1—source data 1O2).

Adcyap1r1 expression in hypothalamus was extensive (Figure 3A–C) and in fact ubiquitous. The nuclei with intense Adcyap1 expression mentioned above also had intense expression of Adcyap1r1. In addition, the PVH (49), SO (51), SCH (50), DMH (52a), arcuate hypothalamic, anterior hypothalamic nucleus, zona incerta (102, Figure 1—source data 1H), postero-lateral hypothalamic area (LH, area 74), periventricular hypothalamic nucleus posterior, dorsal premammillary nucleus and supramammillary nucleus medial (Figure 1—source data 1I and I10) exhibited strong Adcyap1r1 expression. Both Slc32a1- and Slc17a6-positive cells co-expressed Adcyap1r1.

Midbrain

We found moderate expression of Adcyap1 with Slc17a6 in the mainly sensory-related structures: inferior colliculus (IC, Figure 3A, B, C, area 17), nucleus of the brachium of IC, midbrain trigeminal nucleus, and sparse expression (between 20–40%) in parabigeminal nucleus (Table 1, Figure 3D, area 98). We found intense co-expression of Adcyap1 and Slc17a6 in motor related structures: superior colliculus, motor related subfield (SCm, Figure 3A–C, area 16, Figure 1—source data 1I, J, K) and periaqueductal gray (PAG, Figure 3A, area 14, Figure 1—source data 1I-K); moderate expression in ventral tegmental area (VTA, Figure 3B, area 54, Figure 1—source data 1I and I9) and Edinger-Westphal nucleus (Figure 3A, area 12). Adcyap1r1 was strongly expressed in the PAG and SC, although a moderate expression of this transcript was observed to be widespread in Slc17a6- and Slc32a1-expressing cells (Figure 3A–D).

Hindbrain

Pons

In sensory-related structures, we found intense Adcyap1 expression in the parabrachial complex (PBC, Figure 3C, area 63), in all its subfields, although it was more intense toward its lateral divisions, external to the superior cerebellar peduncles (scp, Figure 1D and Figure 1—source data 1K, L). The cells in those divisions, except the Koelliker-Fuse subnucleus (KF), were small cells mainly co-expressing Slc17a6. In contrast, the Adcyap1-positive cells in KF were bigger than the cells in the rest of the PBC divisions and strongly co-expressed Slc17a7 (Figure 1D). Other structures with moderate expression of Adcyap1 and Slc17a7 were the lateral division of SOC (Figure 3C, area 66 and Figure 1—source data 1L) and DCO and VCO (Figure 3E, area 126 and 127, Figure 1—source data 1K, M, M6). Lateral leminiscus nucleus and the principal sensorial nucleus of the trigeminal (Figure 3D, area 95, Figure 1—source data 1L) expressed Adcyap1 moderately in Slc17a6-expressing cells. In motor-related structures, we found moderate Adcyap1 co-expressed with Slc17a6 in the following structures: tegmental reticular nucleus, Barrington’s nucleus, dorsal tegmental nucleus, pontine grey (PG, Figure 3A, area 27, Figure 1—source data 1J), pontine reticular nucleus (Figure 3A, area 25, Figure 1—source data 1K, L), supratrigeminal nucleus (SUT, Figure 1—source data 1L), and superior central nucleus raphe (Table 1 and Figure 3A–C). In behavioral-state-related structures, we found intense Adcyap1 expression in locus coerulus (LC) neurons co-expressing Slc17a7 or Slc17a6. The laterodorsal tegmental nucleus also expressed intensely Adcyap1, co-expressed with Slc17a6. Pontine reticular nucleus and superior central nucleus of raphe expressed Adcyap1 with Slc17a6 in a moderate manner (Table 1).

Adcyap1r1 was strongly expressed in pons structures, which also intensely expressed Adcyap1, such as PG, LC, PBC, DTN regions. Otherwise, the expression was observed to be widespread in both glutamatergic and GABAergic cell types (Figure 3A–D).

Medulla

In the medulla oblongata, Adcyap1 was extensively expressed and generally in a sparse pattern. However, some of the nuclei showed strong-intense expression: the nucleus of tractus solitarius (NTS, Figure 3A, area 19), medial division (co-expressing mainly Slc17a6), the NTS lateral division (co-expressing mainly Slc17a6 and occasionally Slc17a7), and the dorsal and ventral cochlear nuclei (co-expressing Slc17a7 or Slc17a6). Details of other Adcyap1-positive neurons co-expressing Slc17a6 nuclei can be found in Table 1.

The Adcyap1r1 expression in medulla is similar to pons, a widespread pattern with intense expression in NTS divisions, and other nuclei, which expressed Adcyap1 (Figure 3A–D).

Cerebellum

In the cerebellum, Adcyap1 was expressed in all Purkinje cells, which co-expressed Slc32a1 (Figure 1, Figure 3A–F and Figure 1—source data 1M, M1-M4, O and O1). Figure 1, panels E-H show two cerebellar regions, paraflocculus (E and F) and central (G and H), in low and high magnification, respectively, where Adcyap1 expression was higher than other cerebellar lobules. Purkinje cells, distributed in all regions of cerebellar cortex, are the most prominent population of GABA/PACAP co-expressing neurons of the brain. Some GABAergic cells in the granule cell layer of paraflocculus and central regions also co-expressed Adcyap1 (indicated with double pink arrowheads in Figure 1E–H), although we could not identify whether these were, Golgi, Lugaro or globular cells (Simat et al., 2007).

In these two cerebellar regions, some granule cells also expressed Adcyap1 (indicated with single blue arrows, Figure 1E–H). In deep cerebellar nuclei, few Adcyap1-expressing cells were found in fastigial, interposed and dentate nuclei, with the former two co-expressing Slc17a7 and the latter co-expressing Slc17a6 (Table 1). Adcyap1r1 expression here was more limited than in other brain regions analyzed above. Adcyap1 was mainly expressed in the Slc32a1-expressing Purkinje cells and sparsely expressed in Slc17a7- and Slc17a6-expressing neurons in the deep cerebellar nuclei (Figure 3A–F).

PACAP→PAC1 signaling within sensory and behavioral circuits

Here, we analyze the chemo-anatomical aspects of PACAP/PAC1 mRNA expression using the results described above, but putting them into basic sensory circuit wiring maps, as well as in behavioral state and survival instinctive brain longitudinal structures, especially the hypothalamic hubs, based on existing classification schema (Swanson, 2012; Sternson, 2013; Swanson et al., 2016; Zimmerman et al., 2017; Swanson, 2018). Based on PACAP mRNA expression in these proposed sensory/behavioral circuits, we have addressed consequences of PACAP deficiency on neuronal activation and behavioral output in a mouse model of predator odor exposure and defensive behavior.

PACAP-PAC1 co-expression in forebrain sensory system

Thirst circuit for osmotic regulation

As shown above, Adcyap1 was intensely expressed in all peri/para ventricular structures directly related to thirst and osmotic regulation (Figure 7). These structures include SFO, OVLT, MEPO, and PVH (vide supra). Other hypothalamic nuclei intrinsically related to osmotic control and anticipatory drinking are SO and SCH, which were intensely Adcyap1r1-expressing.

The SFO is an embryonic differentiation of the forebrain roof plate, in a dorsal region between the diencephalon (interbrain, thalamus) and the telencephalon (endbrain) (Swanson and Cowan, 1979; Anderson et al., 2001). This nucleus lacks a normal blood-brain barrier, and so its neurons are exposed directly to peptide hormones in the blood. One such hormone is angiotensin II, whose blood levels are elevated upon loss of body fluid due to dehydration or hemorrhage. Hence, the SFO is a humorosensory organ that detects hormone levels in the circulation to control drinking behavior and body water homeostasis. The SFO is situated immediately dorsal to the third ventricle and contains intermingled populations of glutamatergic (VGLUT2-PACAP-PAC1) and GABAergic (VGAT-PAC1) neurons with opposing effects on drinking behavior. Optogenetic activation of SFO-GLUT neurons stimulates intensive drinking in hydrated mice, whereas optogenetic silencing of SFO-GLUT neurons suppresses drinking in dehydrated mice (Zimmerman et al., 2017). By contrast, optogenetic activation of SFO-GABA neurons suppresses drinking in dehydrated mice (Bichet, 2018). SFO-GLUT projections to the median preoptic nucleus (MEPO) and OVLT drive thirst, whereas SFO-GLUT projections to the ventrolateral part of the bed nucleus of the stria terminalis (BSTvl) promote sodium consumption (Zimmerman et al., 2017). SFO-GLUT projections to the paraventricular (PVH) and supraoptic (SO) nuclei of the hypothalamus Anderson et al., 2001 have not yet been functionally annotated with cell-type specificity, but classic models suggest that these projections mediate secretion of arginine vasopressin (AVP) and, in rodents, oxytocin (OXT) into the circulation by posterior pituitary (PP)-projecting magnocellular neurosecretory cells (MNNs). Recent studies also demonstrated that these MNNs possess ascending projections innervating limbic structures such as amygdala, hippocampus, lateral habenula, and lateral hypothalamus (Hernández et al., 2015; Hernández et al., 2016; Zhang et al., 2021) in a cell-type specific manner (Zhang and Hernández, 2013; Zhang et al., 2016a; Zhang et al., 2018). When these are activated, their central collaterals can exert motivational effect on exploration and drinking behavior.

Thirst and AVP release are regulated not only by the classical homeostatic, interosensory plasma osmolality negative feedback (through SFO as a humorosensory organ), but also by novel, exterosensory, anticipatory signals (Gizowski et al., 2016). These anticipatory signals for thirst and vasopressin release converge on the same homeostatic neurons of circumventricular organs that monitor the composition of the blood. Acid-sensing taste receptor cells (which express polycystic kidney disease 2-like one protein) on the tongue that were previously suggested as the sour taste sensors also mediate taste responses to water. Recent findings obtained in humans using blood oxygen level-dependent (BOLD) signals demonstrating that the increase in the lamina terminalis (LT) BOLD signal observed during an infusion of hypertonic saline is rapidly decreased after water intake well before any water absorption in blood. This is relevant in the context of this paper since the MEPO of the hypothalamus has been shown to mediate this interesting phenomenon, integrating multiple thirst-generating stimuli (Allen et al., 2017; Gizowski and Bourque, 2017); however, functionally annotated cell-type specific circuitry has not been clarified. Very intense expression of Adcyap1 was observed in MEPO. Together, these observations open new possibilities to further understand the role of PACAP-PAC1 signaling within this nucleus for homeostatic and allostatic control.

Information about plasma sodium concentration enters the circuit through specialized aldosterone-sensitive neurons in the NTS, an intense PACAP-expressing nucleus in the medulla, that expresses 11β-hydroxysteroid dehydrogenase type 2 (NTS-HSD2 neurons) (Zimmerman et al., 2017; Bichet, 2018), which promote salt appetite and project to the LC and PBN, both intense Adcyap1-expressing and BSTvl, moderate Adcyap1-expressing and intense Adcyap1r1-expressing (Figure 6A, basic circuit presented based on above literature).

Mapping the spatial distribution of PACAP-PAC1 hubs within glutamate/GABA context in relevant sensory circuits in mice.

For abbreviations also see the corresponding table in Appendix 1. (A) Thirst and salt appetite-related pathways with PACAP-PAC1 glutamatergic / GABAergic signaling noted. The main figure is the enlargement of color-shaded areas of the box in the inset at the upper right, projected against a midsagittal section of mouse brain. Blue shaded area symbolizes the hypothalamus and pink shaded area the hindbrain. (B) Olfactory pathway. The projection neurons from the OB send their axons to the different structures of the olfactory cortex, among them AON, TT, OT, PIR, the amygdaline complex (LA, BLA, CEA, BST), ENT, and nLOT). (C) Visual pathway and circadian circuit for brain states. II: optic nerve; PVN: paraventricular nucleus; LHA: lateral hypothalamic area; LG: lateral geniculate nuclei; IGL: intergeniculate leaflet; M/LHb: medial and lateral habenula; OP: olivary pretectal nucleus, SC: superior colliculus; Vis: visual area; PTL: parietal association area; RSP: retrosplenial area; MBSR: midbrain behavioral state related (pedunculopontine nucleus, substantia nigra, midbrain raphe nuclei; PBSR: pons behavioral state related (locus coeruleus, superior central nucleus of raphe, pontine reticular nucleus). (D) Auditory pathway. VIII: cochlear nerve; DCO/VCO: dorsal and ventral cochlear nuclei; MNTB: medial nucleus of the trapezoid body; SO: superior olivary complex; NLL: Nucleus of the lateral lemniscus; IC: Inferior colliculus; MG: medial geniculate complex; AC: Auditory cortex; AAC: associate auditory cortex; TE: temporal association area; Hipp: hippocampus; LHA: lateral hypothalamic area; PVN: hypothalamic paraventricular nucleus; IL: infralimbic cortex; PL: prelimbic cortex. (E) Gustative pathway. V, VII, IX, X: represent trigeminal, facial, glossopharyngeal and vagus nerves respectively. NTS: Nucleus of the solitary tract (nucleus of tractus solitarius); NA: nucleus ambiguus; PBN: Parabrachial nucleus; VPM: ventropostero medial nucleus of the thalamus; BLA: Basolateral amygdala; CeA: central Amygdala; BST: bed nucleus of the stria terminalis; LHA: lateral hypothalamic area; RF: reticular formation.

Olfactory pathway

In the main olfactory system, input from olfactory sensory neurons reaches the main olfactory bulb (MOB), and the axons of the projection neurons in MOB travel to the anterior olfactory nucleus (AON), piriform cortex (Pir), and amygdala for additional processing (Wacker et al., 2011). This multi-step pathway brings olfactory information to be processed in multiple areas of the cerebral cortex, including the anterior AON and Pir. The AON, a cortical area adjacent to the olfactory bulb, is part of the main olfactory pathway. A parallel system, the accessory olfactory system, brings information, for instance that conveyed by pheromones, from the vomeronasal organ into the accessory olfactory bulb, which innervates the MEA, BST, and cortical amygdala (Figure 10). Although it was originally assumed that only the accessory olfactory system processed pheromonal and other socially relevant odors, more recent evidence suggests that social information is processed by both pathways (Wacker and Ludwig, 2012).

The olfactory system appears specially to use PACAP/PAC1 as one of its main modes of co-transmission (see section The claustrum), especially within the cell population in the outer plexiform layer of the MOB and in the mitral cell layer of the AOB, in which some cells were observed to co-express Adcyap1 and Slc17a7, Slc17a6, or Slc32a1 (Figure 1—source data 1A2) and Figure 1A–F.

Visual pathway and the circadian circuits for brain states

There is a vast literature on the visual system and we only touch on selected hubs here to emphasize the PACAP-PAC1 signaling role for visual information processing (Figure 6C). PACAP-PAC1 in the visual system was first discovered via identification of the PACAP immunopositive RGCs, soon after the discovery of PACAP itself (Hannibal et al., 1997), which project to the SCH. Here, using DISH method, we report Adcyap1 co-expressed with Slc17a6 in retina ganglion cells. Adcyap1 expression intensity followed a circadian oscillation mode (see section Retina). The RGC sends visual information through its axons forming the optic nerve, chiasm and optical tract and accessory optical tract, with its first offshoot to SCH for brain state modulation. The SCH was found intensely expressing Adcyap1r1 in both Slc32a1- and Slc17a6-expressing neurons widely and homogenously distributed, at variance with the observation in rat that PAC1 is expressed mainly in the ventral part of the SCH (Hannibal et al., 1997). Leaving the optic chiasm, the optic tract courses latero-caudally and emits a second offshoot, the accessory optical tract, splits off and courses to the midbrain, where it ends in three terminal nuclei, that is medial, lateral and dorsal terminal nuclei. They play an important role in controlling eye movements and are thus parts of the motor system and were all Adcyap1-expressing. The main optic tract continues on to end in the lateral geniculate complex (LG) of the thalamus, after giving off collaterals to the superior colliculus (SC) of the midbrain and to the olivary pretectal nucleus (OP). The dorsal part of the LG then projects to primary visual cortex (Vis), whereas the OP is involved in visual reflex, and the superior colliculus has two main roles: projecting to the motor system, and projecting to secondary visual cortical area via thalamus (Swanson, 2012). All above optic nerve/tract/accessory tract targeted structures were observed to have intense expression of Adcyap1r1. The visual cortex, parietal associated area, and retrosplenial area intensely expressed Adcyap1 (Table 1 and Figure 1—source data 1I, J, K and Figure 1A–F).

Adcyap1r1 was strongly expressed in SCH (Figure 6C and Figure 7), which projects in turn to several Adcyap1r1-expressing regions, which control brain state (sleep-wake cycle), such as midbrain behavioral state related structures (MBSR: pendunculopontine, midbrain raphe nuclei, all Adcyap1-expressing) and pons behavioral state related (PBSR: locus coeruleus, superior central nucleus of raphe, pontine reticular nucleus, all Adcyap1-expressing) (Figure 1A–F).

Presence of PACAP-PAC1 in major cells groups associated with behavioral state (left) and behavioral control system and hypothalamic instinctive survival system (right column).

Left column: critical nodes for behavioral state symbolized by dark gray shaded objects, modified from Swanson, 2012. In the longitudinal cell group-column of brain stem the key neurotransmitters are annotated. ACH, acetylcholine; CRH, corticotropin-releasing hormone; DA, dopamine; ENK, encephalin: GABA, gamma-amino butyric acid; GAL, galanin; GLUT, glutamate; H/O, hypocretin/orexin; HIST, histamine; MCH, melanin-concentrating hormone; NE, norepinephrine; 5HT, serotonin. Right column: hypothalamic survival circuit that consisted of discrete hypothalamic regions contain interoceptors for a variety of substances and have neuronal afferences from primary sensory systems to control the secretory and instinctive motor outputs. The rectangle in the midline represents the neuroendocrine motor zone for secretion of hypophysiotropic hormones, which include thyrotropin-releasing hormone, corticotropin-releasing hormone, growth hormone-releasing hormone, somatostatin, gonadotropin-releasing hormone, dopamine, neurotensin. SFO, subfornical organ; OVLT, organum vasculosum of lamina terminalis; MnPO, preoptic nucleus; AHN, anterior hypothalamic area; PVH, paraventricular hypothalamic nucleus; VMH, ventromedial hypothalamic nucleus; LH, lateral hypothalamic area; MBO, mammillary body (for general reference see Swanson, 2012).

PACAP-PAC1 co-expression in the central processing of ganglion cell sensory systems

The sensory ganglion cells were observed to express PACAP shortly after its discovery (Sundler et al., 1996). PACAP mRNA and peptide levels are increased in sensory ganglion cells within 1 day following axotomy, suggesting possible roles in neuronal protection, differentiation, nerve fiber outgrowth, and/or restoration of perineuronal tissue upon neuronal damage. The sensory ganglion cells send their axons to primary sensory nuclei in the dorsal medulla, which include: (a) auditory system, which ends in the cochlear nuclei (Figure 6D); (b) vestibular system, which ends in the vestibular nuclei (not included in this analysis); (c) gustatory system which ends in the rostral nucleus of tractus solitarius (NTS, Figure 6E); and (d) vagal/glossopharyngeal visceroceptive system, which ends in the caudal nucleus of the NTS (not included in this analysis). The special sensory nuclei in the medulla are all derived in the embryo from a highly differentiated, dorsal region of the primary hindbrain vesicle, the rombic lip (Figure 5.14 of Swanson, 2012) – they were all observed to co-express Adcyap1 and Slc17a6 from very intense (NTS) to moderate levels (DCN and VCN) (see Table 1, Figure 1A–F, Figure 1—source data 1M5-M6 and website).

Here, we analyze the PACAP/PAC1 participation in the auditory and gustatory central processing circuits.

Auditory pathway

Auditory information, for example the spectrum, timing, and location of sound, is analyzed in parallel in the lower brainstem nuclei, that is, dorsal and ventral cochlear nuclei (DCN and VCN), medial nucleus of trapezoid body (MNTB), superior olivary complex (SO), and nuclei of the lateral lemniscus (NLL). These hindbrain structures (NLL and SO) project to the inferior colliculus (IC, in midbrain) through both the excitatory and inhibitory inputs to IC. From IC, the information travels toward thalamus via relay in medial geniculate complex (MG), to reach the auditory cortex, associate auditory cortex, temporal association area; hippocampus, which subsequently projects to prefrontal cortices (Ono and Ito, 2015). Main neuronal structures that are relevant to the auditory pathway are shown in the schematic diagram of Figure 6D. The glutamatergic population of these structures were all found to co-express Adcyap1, Slc17a6 and/or Slc176a7, as well as Adcyap1r1 (Figure 3A–F).

The mainly Sc321a1-expressing structures located in subcortical regions which strongly co-expressed Adcyap1r1 (Figure 2D–G and Table 2), include, for instance, BST and the CeA with its three subdivisions (Figure 6D, structures symbolized in pink-GABAergic, and blue outline–PAC expressing) participate in cognitive and emotional auditory information, obtaining auditory sensory input from brain stem>midbrain>thalamic MG, which project to lateral amygdala (LA, a main PACAP-containing nucleus derived from cortical plate) which send input to the cognitive centers through basal lateral amygdala (BLA, Figure 2F and G).

Gustatory pathway

Chemicals (taste substances) in foods detected by sensory cells in taste buds distributed in the oropharyngeal epithelia are recognized as tastes in the gustatory cortex (GC), including granular insular cortex (GI) and agranular insular cortex (AI), also called dysgranular insular cortex (DIC), which is the primary gustatory cortex (Figure 6E). Between the peripheral sensory tissue and the GC, many neuronal hubs participate in the gustatory information processing, beginning with the geniculate ganglion (GG) of the facial nerve, cranial nerve CN-VII which receive taste stimuli from the anterior 2/3 of the tongue; the petrosal ganglion (PG) of the glossopharyngeal nerve, CN-XI, which receives the taste stimuli from the posterior third of the tongue; and nodose ganglion (NG), also called the inferior ganglion of the vagus nerve (CN-X, which innervates the taste buds of the epiglottis). The lingual nerve, a branch of mandibular nerve, from trigeminal nerve (CN-V), also innervates the anterior 2/3 portion of tongue. All these ganglia express PACAP as their co-transmitter and this expression is potentiated during injury (Sundler et al., 1996). Taste information from sensory ganglia of CN V, VII, IX, and X project to the rostral NTS of the medulla, which intensely expressed Adcyap1 and Adcyap1r1. NTS neurons send taste information to parabrachial nucleus (PB, which also intensely expressed Adcyap1 and Adcyap1r1) and then to the parvocellular division of the ventral posteromedial nucleus VPM of the thalamus, which projects to GC, as well as to the cognitive centers, central amygdala (CEA), bed nucleus of stria terminalis (BNST) and lateral hypothalamus (LH) which also reciprocally innervate the PB (Halsell, 1992). In Figure 6 panel E, the basic wiring of taste circuit was based on the literature (Halsell, 1992; Carleton et al., 2010), and modified with PACAP-PAC1 signaling annotated (Figure 1A–F).

PACAP-PAC1 signaling in major cell groups associated with behavioral state control system and hypothalamic instinctive survival system

In the course of a day, brain states fluctuate, from conscious awake information-acquiring states to sleep states, during which previously acquired information is further processed and stored as memories (Tukker et al., 2020). Anatomically and chemically distinct neuronal cell groups stretching from medulla, pons, midbrain, through the hypothalamus to the cerebral nuclei all participate in modulation of behavioral state (Swanson, 2012). We have briefly referred to sleep and waking behavioral states in analysis of the hypothalamic SCH in the visual/photoceptive pathways that project to midbrain and hindbrain behavioral state structures broadly defined. In Figure 7, left column we complemented the model of behavioral state cell groups and chemical signatures previously presented (Swanson, 2012), adding PACAP-PAC1 as informed by the current study (Figure 1A–F). Three behavioral state-related glutamatergic structures in medulla, that is raphé magnus, raphé pallidus, and raphé obscurus are not represented in the left column in recognition of the original figure design of the author (Figure 9. 5, Swanson, 2012).

In the right column of Figure 7, we present the main hypothalamic survival and locomotor control hubs with PACAP/PAC1 signaling (Figure 1A–F) noted. These hubs are SFO, OVLT, MEPO, AHN, PVN, VMH, LH, MBO, and periventricular neuroendocrine motor zone. Recent evidence suggests that these cell groups control the expression of motivated or goal-oriented behaviors, such as drinking, feeding, sex, aggression, fear, foraging behaviors (Figure 7, color-filled rectangles represent the correlative behaviors with anatomical structure based on the literature [Sternson, 2013]). The rostral segment of this behavior control column has controllers for the basic classes of goal-oriented ingestive, reproductive and defensive behaviors, common to all animals, where the caudal segment has the controllers for exploratory behavior used to obtain any goal object (Swanson, 2012).

Other relevant structures more caudal in this longitudinal brain stem column are the reticular part of SN (not labeled in Figure 7), which is involved in the control of orientating movements of the eyes and head via projecting to SC, the VTA, Adcyap1-expressing, which together with ACB (intensely Acyap1r1-expressing) and STN (intensely Adcyap1-expressing), form the hypothalamic locomotor region.

After analyzing the above presence of PACAP and its receptor PAC1 in sensory-cognitive to motor output, a question emerged: what would be the behavioral and neuronal activation consequences if PACAP signaling were deficient?

PACAP knockout impairs predator odor salience processing via reducing neuronal activation and vesicular transporter expression in key PACAP-PAC1 nuclei

To assess the behavioral implications of PACAP action within one of these circuits, we examined the effect of knockout of PACAP expression in brain on defensive behavior initiated via olfaction, using a predator odor paradigm (Figure 8A and B). Using a modified open-field box with a lidded container with cat urine litter, and wild type (WT) and PACAP-deficient (KO) C57Bl/6N mice (Hamelink et al., 2002), we assessed the defensive locomotion patterns during cat odor exposure, which include displacement pattern in the four quadrants of the box; approach to the stimulus (urine-containing cat litter) with complete retreats and incomplete retreats; and periods of immobility (freezing behavior). DISH for fos, Slc17a7, Slc17a6, and Slc32a1 expression in the two experimental groups was performed and analyzed post hoc.

PACAP-deficient mice exhibited aberrant predator odor response and defensive behavior.

(A and B) Schematics of behavioral procedure using a modified open field test (OFT) to study defensive behavior during predator odor exposure, with typical behaviors of wild type (A: freezing) and PACAP KO (B, wandering, hyperactive and jumping) symbolized. (Ai and Bi) representative 2D movement tracking in the OFT of WT (Ai) and OK (Bi) mice, with dashed lines symbolizing complete (green lines) /incomplete (red lines) approach/sniff > retreat cycles. (Aii and Bii) 2D movement heat-map, WT, n = 9 mice; KO, n = 9 mice. (C) Pixel analysis reflecting total distance traveled in each of the four quadrants, being the Q1 the container located quadrant (clockwise numbering). The pixel numbers of per quadrant (Q) per mouse were compared that revealed a significant increase only in Q1 (WT vs KO, 26500 ± 3288 vs. 43640 ± 3921, *p<0.05. (D) Numbers of freezing, complete sniff>retreat or partial sniff>retreat episodes were assessed using Student t-test. For freezing (n = 8) and complete sniff>retreat (n = 9) behaviors, KO mice showed significant reduction WT:14.29 ± 2.18 vs KO: 3.18 ± 0.64; and WT: 5.89 ± 0.84 vs KO: 1.33 ± 0.17; for partial sniff>retreat behavior (n = 9) KO mice showed significant increase (WT: 0 vs KO: 2 ± 0.41). (E) Number of cells expressing fos mRNA 45 min after the predator odor test were quantified in a 0.0314 mm2 area and statistic significant differences between KO and WT were determined using a multiple t-test and the Bonferroni-Dunn method. There was a significant decrease in the expression of fos in the following regions of KO mice: olfactory bulb mitral layer ‘OB ml’ (WT: 44.4 ± 2.60 vs KO: 19.43 ± 1.25); olfactory bulb granule layer ‘OB gl’(WT: 21.43 ± 0.84 vs KO: 7.29 ± 0.52); medial amygdala ‘MeA’ (WT: 12 ± 0.44 vs KO: 5 ± 0.76); central amygdala ‘CeA’ (WT: 6.71 ± 0.57 vs KO: 2.71 ± 0.47); medial septum ‘MS’ (WT: 5.71 ± 0.42 vs KO: 3.71 ± 0.29); lateral septum ‘LS’ (WT: 14 ± 0.62 vs KO: 5.14 ± 0.56); bed nucleus of stria terminalis ‘BST’ (WT: 11.43 ± 0.57 vs KO: 3.57 ± 0.37); ventromedial hypothalamus ‘VMH’ (WT: 11.86 ± 0.91 vs KO: 4.86 ± 0.50); infralimbic cortex ‘IL’ (WT: 15.71 ± 0.714 vs KO: 9.86 ± 0.74); prelimbic cortex ‘PL’ (WT: 19.71 ± 0.97 vs KO: 11.43 ± 1.23); cingulate cortex ‘CG’ (WT: 24.28 ± 0.92 vs KO: 11.57 ± 1.15) and parabrachial nucleus ‘PB’ (WT: 30.71 ± 2.82 vs KO: 7.29 ± 1.02) and granule cell layer of cerebellum ‘CBgcl’ (WT: 13.14 ± 0.96 vs KO: 8 ± 0.76). In contrast, the number of fos nuclei was augmented in two motor related areas in KO animals related to WT: motor cortex, layer IV ‘MC IV’ (WT: 4.43 ± 0.65 vs KO: 26.71 ± 1.54) and motor cortex, layer V (WT: 5.28 ± 0.57 vs KO: 33.57 ± 1.43). (F and G) example of reduced fos expression pattern in central amygdala in KO mice and (H and I) augmented fos expression in primary motor cortex of KO mice. Scale bars: 200 µm for low-magnification and 20 µm for high-magnification insets. Statistic significant differences are depicted as ***p<0.001, **p<0.01, *p<0.05. Note that during the freezing analysis, two outliers, one in each experimental group were discarded because being 45 and 24 (counts), with 5 times and 11 times higher than the standard deviations for the corresponding groups, that is WT:14.29 ± 6.17 (mean ± SD) vs KO: 3.18 ± 1.80 (mean ± SD). This exclusion was based on criteria explained in NIH Rigor and Reproducibility Training course, https://www.nigms.nih.gov/training/documents/module4-sample-size-outliers-exclusion-criteria.pdf and http://www.itl.nist.gov/div898/handbook/prc/section1/prc16.htm.

Cat urine triggered purposeful movement such as sniff>retreat in WT subjects (Figure 8Ai, green dashed lines indicate complete sniff>retreat cycles). In contrast, the movement patterns exhibited by KO subjects consisted in sniff>retreat cycles of short distances and the mice often returned to the container location area (Figure 8Bi, red dashed lines indicate incomplete/short sniff>retreat cycles). PACAP-deficient mice had significant reduction of complete retreats (p<0.001) and significant augmentation of incomplete retreats (p<0.001) (Figure 8D). Figure 8Aii and Bii show the X-Y dimension movement heat-maps for each group. Movement analysis by quadrants showed a significant increase in pixel values of the KO group in the quadrant where the odorant container was located (Figure 8C, p<0.05, C1). It is worth mentioning that movement in Z dimension (repetitive jumping), characteristic of PACAP-deficient mice (Hashimoto et al., 2001) was not represented in the X-Y dimension heat maps, and investigating the phenomenon is beyond the scope of the study. KO mice also exhibited significant reduction in freezing behavior (Figure 8D, p<0.001). To test whether or not the behavior of PACAP-deficient mice could be explained in whole or part by hypo- or anosmia, a standard test for olfaction was carried out. Wild-type and PACAPko mice were tested for the ability to retrieve a buried cookie as described in Materials and methods. Retrieval times for wild-type and PACAP-deficient mice were not significantly different (wild-type retrieval latency 38.0 ± 10.4 s; PACAPko retrieval latency 44.5 ± 12 s, p=0.7058, t-test, n = 6/group, each group comprising three male and three female mice approximately 8 weeks of age). We conclude that PACAP-deficient and wild-type C57Bl6/N mice were equally able to locate a food source using olfaction.

The Fos expression associated with the predator odor exposure behavioral test was significantly reduced, in PACAP-deficient mice, in olfactory areas, the mitral layer and glomerular layers, as well as higher predator-odor processing centers, for instance, the amygdaloid complex MEA and CEA, the LS, the endbrain ACA, PL and IL cortical areas, the BSTov; the hindbrain PB complex, which provides the main upstream PACAP signaling source especially for BSTov and CEA (http://connectivity.brain-map.org/) for the above cognitive centers, and the hypothalamic VMH, which control the aggressive behavior (Figure 8E–G), as well as the cerebellar cortex in which we found reduced fos expression in the granule cell layer of the ansiform lobule, which is reported to influence limb movement control (Manni and Petrosini, 2004; Zhu et al., 2006). In contrast, KO mice had increased Fos expression, compared to wild type, in the motor cortex layer VI and V (Figure 8E,H and I).

We assessed fos expression in the PACAP/PAC1 system including both glutamatergic and GABAergic neurons in response to odorant exposure, using DISH technique and Slc17a7, Slc17a6, and Slc32a1 probes to identify the glutamatergic and GABAergic neurons. Surprisingly, we observed a sharp reduction in the abundance of three vesicular transporters in the main PACAP-containing nuclei we described vide supra (Figure 9). This reduction was observed both as reduced abundance of the ISH staining puncta at the single-cell level and the density of expressing cells in the given region (Figure 9E).

PACAP-deficient (KO) mice showed significant down-regulation of vesicular transporters for glutamate and GABA in regions where Adcyap1 or Adcyap1r1 were strongly expressed in WT mice.

(A–D) Examples of in situ hybridization using RNAscope method showing down-regulation in KO mice (A’, B’, C’, D’) of Slc17a7 (VGLUT1 mRNA) in anterior cingulate area (ACA) (panels As), Slc17a6 (VGLUT2 mRNA) in hypothalamic subtalamic nucleus (STN) (Panels B) and of Slc32a1 (VGAT mRNA) in lateral septum (LS) and bed nucleus of stria terminalis, oval subnucleus (BSTov) (panels Cs), and cerebellar cortex, the ansiform lobule's, Purkinje’s cells (panels Ds). Note that the feature of reduced expression of Slc32a1 at both single-cell and cell density levels, can be clearly observed (arrows) in the Purkinje cells. (E) Number of cells expressing Slc32a1 (orange shading), Slc17a7 (blue shading) and Slc17a6 (green shading) were quantified in a 0.0314 mm2 area and statistically significant reductions of KO compared to WT were detected septum (WT: 39.67 ± 1.23 vs KO: 8.67 ± 0.49); BST (WT = 23.83 ± 1.35 vs KO = 5.17 ± 0.48); ansiform lobule of cerebellum (WT: 21.83 ± 1.19 vs KO: 1.83 ± 0.4); anterior cingulate area (WT: 39.67 ± 1.23 vs KO: 15.67 ± 1.05) and subthalamic nucleus (WT: 26.17 ± 1.78 vs KO: 4.67 ± 0.76), while in the reticular thalamic nucleus, a region that did not contain Adcyap1 expressing cells, no significant difference was detected (negative control). Statistic differences are depicted as ***p<0.001 and ns: not significant. Scale bars: A and A’: 300 µm; Ai and Ai’: 100 µm; Bs and Cs: 20 µm; D and D’: 500 µm; D1 and D1’: 100 µm.

Discussion

Here, we describe in detail the overall topographical organization of expression of mRNA encoding PACAP, and that of its predominant receptor PAC1, and their co-expression with the small-molecule transmitters, glutamate and GABA, using probes for mRNA encoding the vesicular transporters VGLUT1/2 and VGAT, as obligate markers for glutamatergic and GABAergic phenotypes, respectively, in mouse brain. A highly sensitive dual in situ hybridization method (RNAscope 2.5HD duplex detection) was used and corroborated with Allen Brain Atlas ISH data throughout brain. We report glutamatergic or GABAergic identity of 181 Adcyap1-expressing cell groups as indicated by co-expression of corresponding vesicular transporters. All Adcyap1-expressing neurons studied, and most of their neighboring cells, co-express mRNA encoding PAC1 (Adcyap1r1), indicating that the PACAP/PAC1 pathway, besides using classical neurotransmission through axon innervation and transmitter release, likely employs autocrine and paracrine mechanisms for signal transduction as well.

With the information obtained by PACAP/PAC1 expression and co-localization with small-molecule transmitters, we explored the hypothesis that the broad brain distribution of PACAP may reflect a more specific physiological function at a systems level. We examined the distribution of PACAP/PAC1 expression within specific sensory input-to-motor output pathways passing through the cognitive centers mentioned in these studies, in which most of the hubs analyzed used PACAP>PAC1 signaling within the context of glutamate/GABA neurotransmission. This systematic analysis has revealed several possible PACAP-dependent networks involved in the highest levels of motor control,that is the hypothalamic pattern initiator and controller system, intermediate (or intervening) levels of somatosensory information processing, and the complex cognitive and behavioral state control for behavior.

Newly identified PACAP-expressing cell groups in mouse brain suggesting circuit-level function

The high-resolution DISH experiment and analysis showed that although the hypothalamus has the highest group-density of Adcyap1-expressing subpopulations in the brain, PACAP should not be considered as mainly a hypothalamic neuropeptide. Our data revealed newly identified Adcyap1-expressing neuronal populations with mainly Slc17a7 co-expression, derived from both cortical plate and brain stem. Interestingly, most of these neurons co-expressed Calb2, the mRNA transcript encoding the calcium binding protein calretinin. The prominent regions/nuclei of this group include: (1) regions derived from the cortical plate, that is MOB, AOB, AON, TT, PIR, NLOT, COAa, related directly to olfactory processing, (2) regions derived from cortical subplate, that is hippocampus CA3vv subset of pyramidal neurons in the ventral pole, dentate gyrus mossy cells, and posterior amygdalar nucleus and bed nucleus of anterior commissure; (3) brain stem structures that is pontine grey, Koelliker-Fuse of parabrachial complex, nucleus of lateral lemniscus within pons; nucleus of tractus solitarius, dorsal and ventral vestibular nucleus, and superior olivary complex lateral part.

Bed nucleus of anterior commissure (BAC): a prominent yet chemoanatomically ill-defined PACAP-expressing nucleus

We described this nucleus as a newly identified major Adcyap1-expressing nucleus (section Structures derived from cerebral nuclei; Figure 5). Regarding the phylogenetic classification of the BAC, some authors have argued that “the septum (within striatum) is divided into the lateral, medial, and posterior septum (LS, MS and PS, respectively); and the PS is further subdivided into the triangular septum (TS) and the bed nucleus of the anterior commissure (BAC)" (Risold, 2004). However, the Allen Brain Map classifies the BAC as a pallidum structure (https://portal.brain-map.org/). Although literature on BAC connectivity is sparse, LS is identified as one of the inputs to the BAC (Swanson et al., 2016) and the medial habenula is one of the reported target region, with implication in the control of anxiety and fear responses (Yamaguchi et al., 2013), the connectivity and function of this glutamatergic-PACAP nucleus is largely elusive. Hence, the chemical identification of this nucleus opens new opportunities for generation of animal models using optogenetic/chemogenetic tools to discern the role of this structure within behavioral circuit(s).

A novel transcriptomically distinct pyramidal subpopulation in ventral hippocampal CA3c is well-placed for modulation of the predator threat response

The hippocampus is typically described in the context of the tri-synaptic circuit. The tri-synaptic circuit is composed of three sequential glutamatergic synapses: perforant path axons of layer II neurons in entorhinal cortex project to the outer two-thirds of the dentate gyrus molecular layer, the location of the distal granule cell dendrites; mossy fiber axons of granule cells project to proximal dendrites of area CA3 pyramidal cells; and the Schaffer collateral axons of CA3 pyramidal cells project to stratum radiatum of CA1, where the apical dendrites of area CA1 pyramidal cells are located (Amaral and Witter, 1989; Figure 4E). However, this trisynaptic circuit seems insufficient to explain the ventral hippocampus observations about the CA3c pyramidal Slc17a7/Adcyap1 co-expressing neurons. Hence, this is a surprising finding since, as discussed by Scharfman, 2007, ventral CA3 may be a point of entry that receives information, which needs to be ‘broadcast,’ such as for stress responding, whereas the dentate gyrus may be a point of entry that receives information with more selective needs for hippocampal processing (Scharfman, 2007). It has been reported that the CA3c pyramidal cells possess collaterals that project in the opposite direction to the tri-synaptic circuit, ‘back’ to the dentate gyrus, by either direct innervation of the mossy cells and GABAergic interneurons in the hilus, or to the granule cell layer glutamatergic and GABAergic neurons (Scharfman, 2007). Those targeted cell types strongly expressed Adcyap1r1 (Figure 2A and A'). A hypothetical circuit modified from the literature (Scharfman, 2007), including this newly identified subset of CA3c PACAP-expressing cells, is presented here (Figure 4F). The hippocampal formation has been suggested to act not as a unitary structure, but with the dorsal (septal pole, DH) and ventral (temporal pole, VH) portions performing different functions (Moser and Moser, 1998). Their argument is based on three data sets. First, prior anatomical studies indicated that the input and output connections of the dorsal hippocampus (DH) and ventral hippocampus (VH) are distinct (Swanson and Cowan, 1977). Second, spatial memory appears to depend on DH not VH (Moser et al., 1995). Third, VH lesions, but not DH lesions, alter stress responses and emotional behavior (Henke, 1990). Here, we contribute with a fourth element, which is the subset of CA3 pyramidal neurons, located in the temporo-ventral pole of the hippocampal formation, transcriptomically distinct from the rest of CA3 pyramidal neurons, based on Slc17a7, Adcyap1 and Calb2 expression, which latter two elements are absent in the septo-dorsal pole.

The VH has attracted attention in odor studies due to dense reciprocal connections to the MeA and to other amygdalar nuclei such as cortical nucleus that receives input directly from the main olfactory system (Scalia and Winans, 1975; Figure 6). The VH also project to AOB and the piriform cortex, a major target of the MOB (Shipley and Adamek, 1984). It was reported that rats with VH lesions exhibited deficits in freezing and crouching when exposed to cat odor (Pentkowski et al., 2006). Another study in mice exposed to coyote urine showed that VH lesions impaired avoidance and risk assessment behaviors (Wang et al., 2013b). In the experiments reported here, PACAP-deficient mice, with no neurons expressing Adcyap1 in CA3vv of VH, yielded similar behaviors, suggesting this cell subpopulation may contribute to optimal predator odor processing using PACAP as co-transmitter to coordinate the predator threat response.

The autocrine/paracrine and neuroendocrine nature of PACAP>PAC1 signaling and glutamate and GABA vesicular transporter expression

The analysis of our data showed an impressive amount of Adcyap1 and Adcyap1r1 coexpression at both single-cell and regional levels (Figure 3), suggesting that the PACAP/PAC1 pathway uses autocrine and paracrine mechanisms in addition to classical neurotransmission through axon innervation and transmitter co-release (Hökfelt et al., 1984; Zhang and Eiden, 2019). That is, besides sending PACAP-containing axons to innervate regions where PAC1 is strongly expressed, PACAP may be released through soma and dendrites, to bind PAC1 expressed on the same and neighboring cells, to prime the neuron for optimum function (Leng, 2018). We still know relatively little about the possible functional interactions and control of release of PACAP>PAC1 signaling in this aspect. However, the observation reported in this study about down-regulation of vesicular transporter mRNAs in PACAP knockout mice, in regions/cell populations which were normally Adcyap1-expressing in wild-type mice, provides one possible example of such a function for PACAP. Activity-dependent regulation of both glutamate and GABA vesicular transporter synthesis and membrane insertion has been reported (De Gois et al., 2005; Erickson et al., 2006; Doyle et al., 2010). PACAP signaling most commonly leads to a net increase in neuronal excitability through modulation of intrinsic membrane currents and transiently increasing intracellular calcium concentration (Johnson et al., 2019). Decreased vesicular transporter mRNA expression would be expected at the cellular level to decrease transmitter quantal size, thus decreasing excitatory/inhibitory postsynaptic currents (EPSCs and IPSCs, respectively) (Billups, 2005), and is consistent with our behavior data. PACAP absence in KO mice resulted in a behavioral hypoarousal response to moderate predator-odor stimulus. The vesicular transporter mRNA loss observed at RNA level upon complete PACAP deficiency, revealed in predator odor-exposed mice, may reflect a modulatory role for PACAP, in wild-type animals, on vesicular transporter mRNA (and protein) abundance within a more restricted, but functionally relevant, range depending on the level of PACAP release and autocrine signaling.

PACAP-deficient mice show hyperlocomotion and abnormal gait, as well as bouts of repetitive jumping (data not shown, but see Hashimoto et al., 2001, Gaszner et al., 2012, Hattori et al., 2012). In 1930, Hinsey, Ranson and McNattin reported a visionary experimental result, done in cats and rabbits about the rôle of the hypothalamus in locomotion (Hinsey, 1930). Quoting Swanson’s interpretation of this early experiment: “when the central nervous system is transected roughly between the mesencephalon and diencephalon, the animals displayed no spontaneous locomotor behavior. They remain immobile until stimulated. On the other hand, animals with transection roughly between diencephalon and telencephalon (or upon complete removal of the cerebral hemispheres, and the thalamus) display considerable spontaneous behavior. In fact, they can be hyperactive when the transection is a bit caudal, but cannot spontaneously eat, mate, or defend themselves. Notably, these last three functions are preserved at a primitive level when the transection is just slightly more rostral". This evidence, combined with the selective lesions or stimulations of the hypothalamus, suggests that the ventral half of the diencephalon (hypothalamus) contains neural mechanisms regulating setpoints for locomotor and other classes of motivated behavior’ (Figure 8.9, Swanson, 2012). As reported here, the major cell groups in the ventral diencephalon, associated with behavioral state (Figure 7, left column) and behavioral control (Figure 7, right column) all use PACAP/PAC1 signaling for co-transmission. Hence, movement pattern/initiation disorders would be expected in PACAP-deficient animals. Moreover, we found marked reductions in Slc32a1 in Purkinje cells (PC) of the ansiform lobule of the cerebellum, as well as significant reduction in Fos expression in the granule cells (Figure 9 panels Ds), the main excitatory input of the PC (Manni and Petrosini, 2004). The reduced VGAT availability in PC in the lateral lobules of the cerebellum, involved mainly in coordinating the movement patterns of the limbs (Manni and Petrosini, 2004, Sakayori et al., 2019), necessarily weakens the inhibitory influence on the excitatory output of the deep cerebellar nuclei which make ascending projections to several forebrain and midbrain structures, including hypothalamus (Zhu et al., 2006). In Figure 10, we propose a model based on the analysis of this study to provide a rationale for hyperlocomotion and hypoarousal behavior observed in PACAP- deficient mice. Under normal conditions, the hypothalamic motor initiator/controller centers (Figure 7, longitudinal PACAP cell groups in hypothalamus) are modulated, for instance, by direct excitatory (e.g. from deep cerebellar nuclei, DCN), indirect excitatory (through disinhibition of GABAergic projections from BNST, CEA to GABAergic neuronal circuits of hypothalamus) and inhibitory (from lateral septum) influences (Swanson, 2012; Figure 10A). We showed in this study that PACAP deficiency caused sharp downregulation of both vesicular glutamate and GABA transporters in the key nuclei that in wild-type mice Adcyap1 is expressed, such as cerebellar Purkinje cells that send inhibitory input to DCN; the prefrontal cortex, that with reduced VGLUT1 expression (Figure 9), could result in a deficient activation of striatum and pallidum cognitive structures, that send GABAergic input to the hypothalamic centers, through direct inhibition or indirect excitation through disinhibition mechanisms, for locomotor patterning and initiation. Unbalanced excitation/inhibition occurring in this hypothalamic region may result in hyperexcitation to motor pathways ending in primary motor cortex (Figure 8, higher fos expression in primary motor cortex in KO subjects), which command the spinal motor neurons for muscle activities (Figure 10B).

Proposed model to interpret how PACAP deficiency influence the olfactory information salience processing to impact motor output, bases on the analysis of this study.

(A) Schematic representations of the mouse peripheral and central olfactory pathways to cognitive centers and motor higher control centers in brain stem under normal condition. PACAP-Pac1 glutamatergic/GABAergic signaling is symbolized by colors. (B) PACAP deficiency weakens the corresponding neuronal hubs inducing ultimately unbalanced excitation/inhibition in sensory-cognitive and motor pattern initiator and controller centers, which may underlie the hyperactivity and attention deficit to salient olfactory stimulus during predator odor exposure test. OSN: olfactory sensorial neurons; VNO: vomeronasal organ; MOB: main olfactory bulb; AOB: accessory olfactory bulb; AON, anterior olfactory n.; TT: taenia tecta; OT: olfactory tubercle; DP: dorsal peduncular area; Pir: piriform cortex; NLOT: n. lateral olfactory tract; EC: entorhinal cortex; PFC: prefrontal cortex; COApm: corticoamygdalar, posteromedial; MEA and CEA: medial and central amygdala; LS: lateral septum; BNST: bed nucleus of stria terminalis; SCm: superior colliculus, motor related; IC: inferior colliculus; DCN: deep cerebellar nuclei; GC: granule cells; thal VL: thalamic ventrolateral n.; i: inhibition; d: disinhibition; e: excitation.

Materials and methods

Key resources table
Reagent type (species) or resourceDesignationSource or referenceIdentifiersAdditional information
Strain, strain background (Mus musculus), both sexesWildtype C57Bl/6NIn-house breeding program in accordance with NIH guidelines and standards and housed in cages containing3–5 male siblings.C57Bl/6N
Strain, strain background (Mus musculus), both sexesPACAP KO mice C57Bl/6NHamelink et al Proc Natl Acad Sci, 2002, 99(1):461–6, PMID:11756684
NIH in-house breeding program
C57Bl/6N Adcyap1-/-
Commercial assay or kitRNAscope detection assay probes2020 Advanced Cell Diagnostics, Inc405911
409561
319171
416331
319191
316921
502231
465391
404631
421931
316091
Adcyap1
Adcyap1r1
Slc17a6
Slc17a7
Slc32a1
Fos
Vipr1
Vipr2
Sst
Pvalb
Crh
Commercial assay or kitRNAscope 2.5 HD Duplex detection kit
Detection reagent-red
2020 Advanced Cell Diagnostics, Inc----------
322500
322360
Software, algorithmMATLAB scriptsThis paperScriptFor motion heatmaps

Mice

The following male mouse strain/line were used in this study: C57BL/6N and C57BL/6N PACAP deficient, age 8–10 weeks, generated through an in-house breeding program in accordance with NIH guidelines and standards. Generation of PACAP knock-out mice have been described previously (Hamelink et al., 2002). All experiments were approved by the NIMH Institutional Animal Care and Use Committee (ACUC) and conducted in accordance with the NIH guidelines.

In situ hybridization (ISH): RNAscope single and duplex (DISH) procedure for PACAP mRNA co-distribution mapping

Request a detailed protocol

Mice (n = 5) were deeply anesthetized with isoflurane, decapitated and whole brains were removed, embedded in OCT medium and rapidly frozen on dry ice. Serial brain sections (12 μm thick) were cut on a cryostat (Leica CM 1520), adhered to SuperFrost Plus slides (ThermoScientific). RNA probes were designed and provided by Advanced Cell Diagnostics (Hayward, CA): Mn-Adcyap1 (gene encoding PACAP), Mn-Adcyapr1r (gene encoding PAC1), Mn-Slc17a7 (gene encoding VGLUT1), Mn-Slc17a6 (gene encoding VGLUT2), Mn-Slc32a1 (gene encoding VIAAT, also called VGAT), Mn-Vipr1 (gene encoding VPAC1), Mn-Vipr2 (gene encoding VPAC2), Mn-Sst (gene encoding somatostatin), Mn-Pvalb (gene encoding palvalbumin), Mn-Crh (gene encoding corticotropin releasing hormone), and Mn-Fos (gene encoding cFos). All staining steps were performed following the protocols provided by the manufacturer for chromogenic detection of mRNA in fresh frozen tissue samples. Stained slides were examined with both light microscope with digital camera and were image-captured with 20X objective with ZEISS Axio Scan (Carl Zeiss Microscopy, Thornwood, NY from Systems Neuroscience Imaging Resource, NIMH-IRP, NIH) and images for sections from each animal were organized and converted to TIF files with BrainMaker (MBF Bioscience, Williston, VT).

Chemo-anatomical analysis and circuit identification

Request a detailed protocol

Anatomical nomenclature and regional delineation were done according to Allen Mouse Brain Atlas (http://www.allenbrainatlas.org) corroborated with Paxinos and Franklin's the Mouse Brain in Stereotaxic Coordinates. The Adcyap1/Adcyap1r1 distribution mapping and semi-quantification and co-expression with glutamate/GABA transporter mRNAs were corroborated with the Allen brain atlas https://mouse.brain-map.org/. For semiquantitative scoring we used criteria similar to previous literature reports (Hannibal, 2002). Our annotation criteria were: the number of expressing cells as a percentage of total Nissl stained nuclei: '-', not observed; '+', weak (<20%); '++', low (20–40%); '+++', moderate (40–60%); '++++', intense (80–60%); '+++++', very intense (>80%). Functional neuroanatomy order and annotations are based on Allen Institute Mouse Reference Atlas (http://atlas.brain-map.org/). For schematic circuit chartings of Figures 6, 7 and 10 we used Swanson’s rodent flat maps downloaded from http://larrywswanson.com/?page_id=164. Sensorimotor circuit hubs and connectivity of Figures 6 and 10 are based on literature cited in the results section as well as consulting the website https://sites.google.com/view/the-neurome-project/connections/cerebral-nuclei?authuser=0.

Behavioral experiments: predator odor

Subjects

C57BL/6N wild-type and PACAP-deficient male mice, age 8–10 weeks, n = 9 (N = 18) were used. They were group-housed (4/cage) in a room kept on a controlled light-dark cycle (light on 7:00 am and off 7:00 pm) under constant humidity and temperature conditions.

Odor stimuli

Request a detailed protocol

Cat urine material was collected from domestic cat litter, where multiple male and female cats used the same box to urinate.

Modified open-field box for odor exposure and fos expression assessment

Request a detailed protocol

A custom‐made wooden box (28 x 28 x 28 cm, with a sliding glass lid) was used (see Figure 8A). The box was positioned inside a low-noise suction hood, then in a lidded container, cat urine containing litter material was introduced into the box (Figure 8A). After introducing the subject, the lid of container was opened through a string and the box was closed to avoid the odor to escape. Video recording of the animal behavior was made during a 10-min period, after which animals were returned to its home cage for 30 min and then euthanized by cervical dislocation and brains rapidly removed and processed for dual in situ hybridization (DISH) using the RNAscope 2.5 HD Duplex Assay to assess the expression of fos mRNA within Adcyap1-, Adcyap1r1-, Slc17a7-, Slc17a6-, and Slc32a1‐positive cells.

Behavior assessment

Request a detailed protocol

Mouse behavior during the 10 min of cat urine exposure was recorded by an overhead camera. Behavioral scoring was performed off-line with blind analysis (performed by ECG, medical student in laboratory rotation, see acknowledgement). Freezing behavior was assessed in the first 5 min lapse, giving a score every 5 s when the subject exhibited immobility with piloerection. Complete cycle of sniff>retreat was defined as the movement of approach the containing and immediately retreat beyond the quadrant I (QI, Figure 8) while incomplete sniff > retreat cycles were defined when the subject approached the container and stayed in the proximity or retreat only a short distance within the same quadrant (Q1). Mice spatial displacement maps were produced on-screen by the analyst with aid of the software PowerPoint > shape format > curve tool (Microsoft Office). The experimental arena was represented in a slide with a square containing 16 small squares. The flask containing the cat litter is symbolized with a filled circle. The computer aided manual drawing was visually guided by the video recording under the strict criterion that the analyst makes a click in the corresponding position of the map for each end of a lineal movement of the mouse. The jumping behavior is only symbolized with ‘zig-zags’. Four of mice exhibited this behavior and due to the aim of this study – the place preference assessment – these subjects were discarded for the Heat map. PowerPoint traces where skeletonized with ImageJ and movement traces from n = 5 animals were used. The 2D movement heat-map was produced with MATLAB R2016b, for each time the mouse passed through a point in space, the value of 1 was assigned. Using the MATLAB ‘color dispersion function’ (PSF) and ‘color map’ functions, a heat map was constructed. The trajectories of the five mice in each group were superimposed, then using the 'bwarea' function, a count of the pixel quantity per quadrant was performed, being Q1 the container located quadrant (clockwise numbering). Thereby, each pixel in the image had a different value and color according to the number of times the mouse was at that point. The number of pixels per quadrant for mice in the KO group was compared with mice in the WT group following one-way ANOVA.

Behavioral experiments: olfaction

Request a detailed protocol

A standard test for olfaction as described by Yang and Crawley, Current Protocols in Neuroscience 8.24.1–8.24-12, July 2009 was carried out in wild-type and PACAP-deficient mice. Briefly, mice were individually housed under standard light-dark conditions for a period of four days. Subsequently, a cookie (Chocolate Teddy Grahams—Nabisco) was placed in each cage at mid-day on two successive days, consumption was confirmed, and all food was removed on the afternoon of the second day. The following day, mice were acclimated to the testing room for 1 hr, and testing initiated by placing mice in a fresh cage (test cage) with bedding 3 cm deep, allowing exploration for 5 min; removal to a fresh cage while a cookie was buried 1 cm deep at one end of the test cage, and placement at the opposite end to the buried cookie in the test cage. Time for mouse to retrieve the buried cookie and hold the cookie in its forepaws and begin to eat was recorded.

Assessment for Fos and vesicular transporter expression 

Request a detailed protocol

The counting for Fos and vesicular transporters expressed in cells was done on a computer screen connected to a digital camera mounted over a light microscope. The region of interest (ROI) was centered through observation using the microscope oculars and 20x objective and projected to a large computer screen through a digital camera. A fixed square equivalent to 0.0314 mm2 for the magnification/computer enlargement was pre-fixed and moved to the ROI choosing a region with more or less homogenous cell population. Positive cells within the square were counted. We chose two sections of the same region from each mouse (n = 3) for each region's assessment. The means were obtained averaging the six numbers and statistics were performed as described (vide infra). The counting was done by an experimenter blind to genotype.

Data analysis

Request a detailed protocol

GraphPad Prism 7.0 was used to perform Student t-tests and one-way ANOVA for evaluation of statistical differences between groups, levels are indicated as follows: *p<0.05, **p<0.01, ***p<0.001 Unless otherwise indicated, values are reported as mean ± SEM.

Appendix 1

Abbreviations

AbbreviationStructure nameNum. Fig.3
 AAAanterior amygdalar area130
 ACAd1anterior cingulate area, dorsal layer 132
 ACAd6anterior cingulate area, dorsal part, layer 632 b
 ACBnucleus accumbens40
 ADanterodorsal nucleus of the thalamus76
 AIagranular insular area119
 AManteromedial nucleus55
 AMBnucleus ambiguus146
 ANcerebellar cortex, ansiform lobule, granule cell layer152
 AOBaccessory olfactory bulb83
 AONanterior olfactory nucleus82
 AParea postrema18
 AVanteroventral nucleus of the thalamus75
 BACbed nucleus of anterior commissure47
 BLAbasolateral amygdalar nucleus143
 BMAbasomedial amygdalar nucleus133
 BSTbed nucleus of stria terminalis88
 BSTovbed nucleus of stria terminalis, oval nucleus46
 CA2vventral hippocampus, CA2134
 CA3vhippocampal formation, ventral CA3c91
 CA3vvVentral tip of ventral CA3 field92
 CBgclgranular layer of cerebellum90
 CBpjPurkinje cell layer of cerebellum89
 CEAccentral amygdalar nucleus, capsular part131
 CEAIcentral amygdalar nucleus, lateral part145
 CEAmcentral amygdalar nucleus, medial part110
 CLAclaustrum118
 CLIcentral linear nucleus raphe29
 COAacortical amygdala anterior part107 a
 COApcortical amygdala posterior part107 b
 CPcaudoputamen80
 CUcuneate nucleus61
 DCOdorsal cochlear nucleus126
 DG-sgdentate gyrus, granule cell layer87
 DHdorsal hippocampus proper123
 DMHdorsomedial nucleus of the hypothalamus52 a
 DPdorsal peduncular área (TTd)34 a
 DRdorsal nucleus raphe15
 ECTectorhinal area138
 ENTlentorhinal area, lateral part139
 ENTmentorhinal area, medial part140
AbbreviationStructure nameNum. Fig.3
 EPendopiriform nucleus117
 EWEdinger-Westphal nucleus12
 FLcerebellar cortex, flocculus, granule cell layer151
 FNfastigial nucleus of the cerebellum60
 FSfundus of striatum116
 GPeglobus pallidus, external segmet114
 GPiglobus pallidus, internal segment (entopeduncular nucleus)115
 GRNgigantocellular reticular nucleus24
 GUgustatory areas136
 IAintercalated amygdalar nucleus132
 ICinferior colliculus17
 IFnucleus raphe, interfascicular11
 IGLintergeniculate leaflet of the lateral geniculate complex150
 ILAinfralimbic area42
 IOinferior olivary complex21
 IPNinterpeduncular nucleus28
 IRNintermediate reticular nucleus64
 LAlateral amygdalar nucleus144
 LClocus coeruleus65
 LGddorsal part of the lateral geniculate complex149
 LGvventral part of the lateral geniculate complex135
 LMlateral mammillary nucleus71
 LPOlateral preoptic area73
 LSclateral septal nucleus caudal39
 LSrlateral septal nucleus rostral36
 MARNmagnocellular reticular nucleus67
 MBOmammillary body8
 MDmediodorsal nucleus of the thalamus6
 MDRNmedullary reticular nucleus68
 MEAadmedial amydgala, anterodorsal part111
 MEApdmedial amygdala, posterodorsal part112
 MEApvmedial amydgala, posteroventral part105
 MEPOmedian preoptic nucleus3
 MGmedial geniculate complex99
 MHmedial habenula45
 MOBmain olfactory bulb1
 MOBglmain olfactory bulb, glomerular layer85
 MOBgrmain olfactory bulb, granule layer84
 MOBmlmain olfactory bulb, mitral layer86
 MOssecondary motor area44
 MoVmotor nucleus of trigeminal nerve94
AbbreviationStructure nameNum. Fig.3
 MPOmedial preoptic area48
 MRNmidbrain reticular nucleus69
 MSmedial septal nucleus37
 MVmedial vestibular nucleus59
 NDBdiagonal band nucleus38
 NLLnucleus of lateral lemniscus97
 NLOTnucleus of the lateral olfactory tract108
 NTSnucleus of the tractus solitarius19
 OPolivary pretectal nucleus147
 ORBorbital area31
 OTolfactory tubercle41
 OVvascular organ of the lamina terminalis4
 PAposterior amygdalar nucleus106
 PAApiriform-amygdalar area141
 PAGperiaqueductal gray14
 PARparasubiculum128
 PBparabrachial nucleus63
 PBGparabigeminal nucleus98
 PCGpontine central grey58
 PFparafascicular nucleus79
 PGpontine grey27
 PHposterior hypothalamic nucleus10
 PHAposterior hypothalamic area74
 PIRpiriform area129
 PLprelimbic area43
 PMpremammillary nucleus148
 POSTpostsubiculum124
 PPperipeduncular nucleus101
 PRCprecommisural nucleus78
 PREpresubiculum125
 PRNrpontine reticular nucleus25
 PSVprincipal sensory nucleus of trigeminal nerve95
 PTparataenial nucleus56
 PTLpposterior parietal association areas121
 PVHparaventricular hypothalamic nucleus49
 PVpperiventricular hypothalamic nucleus, posterior part53
AbbreviationStructure nameNum. Fig.3
 PVTparaventricular nucleus of the thalamus5
 REnucleus of reuniens7
 RLrostral linear nucleus raphe13
 RMnucleus raphe magnus23
 RPAnucleus raphe pallidus22
 RSPretrosplenial area33
 RTreticular nucleus of the thalamus77
 rVtrigeminal reticular nucleus26
 SCHsuprachiasmatic nucleus50
 SCmsuperior colliculus, motor related16
 SCssuperior colliculus, sensory related57
 SFOsubfornical organ2
 SIsubstantia innominata81, 113
 SNsubstantia nigra70
 SOsupraoptic nucleus51
 SOCsuperior olivary complex66
 SPFsubparafascicular nucleus100
 SPIVspinal vestibular nucleus62
 SpVspinal nucleus of trigeminal96
 SSsomatosensory areas120
 STNsubthalamic nucleus103
 SUBsubiculum104
 SUMsupramammillary nucleus9
 TRpostpiriform transition area142
 TTtaenia tecta34b
 TTvtaenia tecta ventral35
 UVUuvula of cerebellum30
 VCOventral cochlear nucleus127
 vHilusventral hilus93
 VISvisual areas122
 VISCvisceral area137
 VLHventrolateral hypothalamic nucleus72
 VMHventromedial hypothalamic nucleus52 b
 VTAventral tegmental area54
 XIIhypoglossal nucleus20
 ZIzona incerta102

Color indicate embryological origin according to Allen Brain Atlas Common Coordinate Framework as follows:

cortical platesubcortical platecerebral nucleiinterbrain
midbrainhindbraincerebellum

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
    What is the function of the claustrum?
    1. FC Crick
    2. C Koch
    (2005)
    Philosophical Transactions of the Royal Society B: Biological Sciences 360:1271–1279.
    https://doi.org/10.1098/rstb.2005.1661
  11. 11
  12. 12
  13. 13
  14. 14
    GPCR-Gs signaling to ERK is controlled by the cAMP-sensing guanine nucleotide exchange factor NCS/Rapgef2 in neuronal and endocrine cells
    1. A Emery
    2. MV Eiden
    3. T Mustafa
    4. LE Eiden
    (2013)
    Science Signaling 6:ra51.
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
    Inputs and Outputs of the Amygdala: What the Amygdala Is Trying to Tell the Rest of the Brain
    1. P Groor
    (1976)
    Springer.
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
  43. 43
    Distribution of PACAP in the Mammalian Nervous System
    1. K Koves
    (2016)
    In: T. A Reglodi, D cham, editors. Pituitary Adenylate Cyclase Activating Polypeptide — PACAP. Switzerland: springer. pp. 179–203.
  44. 44
  45. 45
  46. 46
  47. 47
  48. 48
  49. 49
  50. 50
  51. 51
  52. 52
  53. 53
  54. 54
  55. 55
  56. 56
  57. 57
  58. 58
  59. 59
  60. 60
  61. 61
  62. 62
    The Septal Region, the Rat Nervous System
    1. PY Risold
    (2004)
    Academic Press.
  63. 63
  64. 64
  65. 65
  66. 66
  67. 67
  68. 68
  69. 69
  70. 70
  71. 71
  72. 72
  73. 73
  74. 74
  75. 75
  76. 76
    Brain Architecture
    1. LW Swanson
    (2012)
    Oxford: Oxford University Press.
  77. 77
  78. 78
  79. 79
  80. 80
  81. 81
  82. 82
  83. 83
  84. 84
  85. 85
  86. 86
  87. 87
  88. 88
  89. 89
  90. 90
  91. 91
  92. 92
  93. 93
  94. 94
  95. 95
  96. 96
  97. 97
  98. 98
  99. 99
  100. 100

Decision letter

  1. Ronald L Calabrese
    Senior Editor; Emory University, United States
  2. Rebecca Shansky
    Reviewing Editor; Northeastern University, United States
  3. Tim James Viney
    Reviewer; University of Oxford, United Kingdom
  4. Gilliard Lach
    Reviewer; University of Edinburgh, United Kingdom

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

The reviewers find this manuscript to be of significant value to the field, as it provides a comprehensive and nicely presented map of PACAP and PAC1 expression across the brain, including novel and detailed information on the cell types in which expression was found.

Decision letter after peer review:

Thank you for submitting your article "Behavioral role of PACAP reflects its selective distribution in glutamatergic and GABAergic neuronal subpopulations" for consideration by eLife. Your article has been reviewed by two peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Ronald Calabrese as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Tim James Viney (Reviewer #1); Gilliard Lach (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

We would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, when editors judge that a submitted work as a whole belongs in eLife but that some conclusions require a modest amount of additional new data, we are asking that the manuscript be revised to either limit claims to those supported by data in hand, or to explicitly state that the relevant conclusions require additional supporting data.

Summary:

The manuscript of Zhang and colleagues studied the expression of PACAP and PAC1 mRNA in inhibitory and excitatory neurons in the entire mouse brain by using dual ISH method. Additionally, a behavioural test is carried out to provide a functional role for PACAP/PAC1 on olfaction and defensive behaviour followed by cFos examination of selected brain regions to indicate the role of PACAP and PAC1 in such behavioural outputs. The reviewers believe that this is a valuable and important study on PACAPergic brain regions in mice, especially relating to the hypothalamus, but would benefit from a major reorganisation to improve the presentation of data, and further quantitative criteria to strengthen the observations. Please address the considerable comments detailed in this letter in your revised submission.

Revisions for this paper:

Reviewer #1:

1) To appeal to the more general readership of eLife, the paper would benefit from a reorganisation, especially when referring to figures and tables. There are a very large number of abbreviations. A list near the beginning of the manuscript would help the reader, and would also shorten the figure legends and improve readability/flow. For the non-expert, some areas should be labelled/highlighted separately or provide more information in the figures, e.g. “ACA and the entorhinal cortex” one has to search the figure legend, find the number then search the figure panels to find the location of these brain regions. Abbreviations and brain region names should be consistent, e.g. ACC is used in text, but ACA in figure and legend. Unless mistaken, Table S1 is not mentioned in the text. Figure 9 is first mentioned in the Discussion. Since these are valuable data, refer to this figure in the main Results section in terms of the knockout. Figure 1—figure supplement 1 is very informative, but requires a lot of searching to find the panel that is referred to in the text. In Figure S1-7/7-M, panels M1-4 are identical to Figure 1E-H and the scale bar in M3 is different to 1G.

2) In several places there are anecdotal statements and it is not clear about the reproducibility of the results. The methods for quantification (including those mentioned in table legends) should be included in Materials and methods.

For animals, please check and state the total number of mice and rats used in the study, and whether EGFP mice were also used.

For c-fos experiments, how were these cells counted, how many sections per mouse, what was the section thickness, how were the values calculated (mean, absolute numbers). Was fos counting done blind to genotype?

Was there variation between animals in terms of expression levels/strength? Case/animal numbers in figures would help. It is not clear what is meant throughout by statements such as “strongest”. Is this by density in cells or number/intensity of puncta? For example, section 3.1, retina. What is meant by “higher percentage than previously reported”? Is this referring to both previous reports in mice? Also see Engelund et al. Cell Tissue Res 2010. How many samples and/or mice were examined and how were ganglion cells counted?

Similarly, cortical expression in different layers, how were the values of 80% obtained? Again, “highest expression level of PAC1 among all brain regions” is a strong claim, how was this quantified? Subsection “The claustrum (CLA)”, need references/evidence for observations of mouse claustrum percentages. “more than 90%”. “the highest expression of PACAP was observed in the MnPO”.

In terms of the olfactory pathways, is there evidence of co-transmission or is this a hypothesis?

Some claims will need careful revision. E.g. in the Figure 5 legend, the last sentence contradicts the main text.

The finding that 100% of the 3 GABAergic subpopulations expressed PAC1 is a big claim, yet there is no quantification to back this up. How many brain regions were examined, how many mice, sections, counted cells etc.? If it just refers to primary somatosensory cortex, was it all or some layers?

Table 2 (also applies to parts of Table 1), do blank areas of the table mean not examined? Or should there be “-“ in these areas? For example, the medial septal complex contains vglut2 expressing cells but the corresponding row/column is blank.

There is the claim that PACAP mRNA was not found in cell body layers, but in Table 1 it is reported that there is weak expression in VGLUT1+ cells. Since VGLUT1 cells are in the pyramidal cell layer, this seems contradictory. It would be helpful to have a higher power image of CA1 (as for rat in Figure S2). Could expression outside this layer be in subpopulations of GABAergic neurons? Were these examined (blank in Table 1)? DG is also missing from Table 1.

PAC1 expression. Subsection “Hippocampal formation” claims it is selective for VGAT cells. But there are clear examples of VGAT- cells in Figure S3B expressing PAC1. What are these?

3) Suggestion about paracrine/autocrine signalling. Is there is evidence in literature for such a role? This seems speculative without immunohistochemical evidence. Hannibal 2002, carried out at both the protein and mRNA levels, showed axon terminals in multiple regions. Can these be mapped to the regions that express PAC1 in mice? Is there any evidence or could the authors comment on the existence of presynaptic PACAP receptors? Expression of PAC1 mRNA does not imply that the cell would express the protein exclusively along its somatodendritic membrane. “Classical” neurotransmission presumably could occur in PACAP/PAC1 rich regions via local axons in addition to long-range axons.

4) The observation of PACAP in part of temporal CA3, which the authors refer to as CA3c, has in fact previously been defined as CA3vv, corresponding to the coch expressing domain (see Thompson et al., 2008, Fanselow and Dong, 2010). PACAP may indeed be an additional marker along with calretinin for this principal cell subpopulation, and they may want to revise their model or refer to these earlier papers.

5) PACAP KO. Some clarification would be welcome in terms of animal cohorts. Please state the experimental unit (i.e. n=9 mice/group). In D, the freezing data show only 8 mice, was one pair excluded due to lack of freezing in an animal, as for jumping mice in C? In Ai, Aii, Bi, Bii, does this show the traces for the total time?

In the separate experiment, was n=3 a separate cohort of mice or from the N=18 total as stated in the Materials and methods? Is the n=3 per group or total mice? This may require an increase sample size for this claim, or show quantification/statistical test. For this test, were experimenters also blind to the genotype? The last sentence is difficult to follow.

For the behavioural tests, please include details about whether the wooden boxes, room and experimenter were familiar to the mice before the test (which could affect variability), whether mice were tested at the same time of day, and if KO and WT animals were housed together.

In the Discussion, can the authors comment on or provide evidence of possible developmental changes / compensatory mechanisms occurring in the KO animals.

Reviewer #2:

Part 1: the PACAP/PAC1 characterization is well designed and executed. The result description is lengthy and sometimes confusing. Figures and tables (including the supplementary information) are clear and informative. The authors decide to do not show Vipr1/Vipr2 data, which should be reconsidered for publication. Overall, this part of the manuscript represents a nice piece of work and surely will be very helpful to whom wish to work with PACAP/PAC1.

Part 2: I think this part is the critical one in this manuscript. Starting from section 4, it uses the part 1 of the manuscript to review the literature and build a neuronal circuit with PACAP/PAC1 that makes for behavioural processes. It is literally a review inside the Results section. The schematic figures are interesting but also quite speculative regarding brain signalling since the authors did not performed any experiment to investigate the pathway of PACAP and the literature is scarce. Moreover, the role of Vip receptors were completely neglected here.

https://doi.org/10.7554/eLife.61718.sa1

Author response

Revisions for this paper:

Reviewer #1:

1) To appeal to the more general readership of eLife, the paper would benefit from a reorganisation, especially when referring to figures and tables. There are a very large number of abbreviations. A list near the beginning of the manuscript would help the reader, and would also shorten the figure legends and improve readability/flow. For the non-expert, some areas should be labelled/highlighted separately or provide more information in the figures, e.g. “ACA and the entorhinal cortex” one has to search the figure legend, find the number then search the figure panels to find the location of these brain regions. Abbreviations and brain region names should be consistent, e.g. ACC is used in text, but ACA in figure and legend.

We thank the reviewer for the appreciation and careful and professional criticism to benefit clarity in presenting our results to the readership of eLife.

The inconsistencies of term usage (produced by using Allen brain maps and the Paxino's and Swanson atlas) have been corrected and now we unified the nomenclature usage to Allen Brain maps (http://mouse.brain-map.org/). The related difficulties on the aspect of abbreviations vs full anatomical region's names mentioned by Reviewer #1 have been emended by inserting a table of nomenclatures number and embryological origin color-coded, according to the Allen Brain Atlas Common Coordinate Framework. The table is now ordered in alphabetical order, with full region's abbreviations being the first column from the left, which are the ones used in the entire text and the figure legends, in order to facilitate the search while saving space. Besides, we made the number codes column for Figure 3.

Unless mistaken, Table S1 is not mentioned in the text.

Actually, it was mentioned in the first submitted version. We have made some corrections on this table: a) in the title we added the word "selective", i.e. "Density distribution of PAC1 expressing cells in selective cortical regions; b) we unified the abbreviations of the brain regions studied to be consistent with the new nomenclature table in the main text (however listed in the table caption since it is an independent file); c) we explain the reason why those regions were chosen for our sampling in the table legends; d) the table is now called Figure 3—source data 3, due to the addition of two more tables for Vipr1 and Vipr2.

Figure 9 is first mentioned in the Discussion. Since these are valuable data, refer to this figure in the main Results section in terms of the knockout.

We are embarrassed for this omission that occurred during versions passing among authors and changing from word (.docx) version to Overleaf software for LaTex for eLife template, where an old version was already uploaded there, that in some moment this paragraph was missed unwilling and also the text was moved upwards, above its corresponding subtitle "2" (now returned to its corresponding place). We sincerely apologize to the reviewers for not realizing this before the first submission, which might have cause confusions. These are all emended. The paragraph referring to Figure 9 is added at the very end of the Results section:

"We assessed fos expression in the PACAP/PAC1 system including both glutamatergic and GABAergic co-expressing neurons in response to odorant exposure, using DISH technique and VGLUT1, VGLUT2 and VGAT mRNAs probes to identify the glutamatergic and GABAergic neurons. Surprisingly, we observed a sharp reduction in the abundance of three vesicular transporters in the main PACAP containing nuclei we described (Figure 9). This reduction was observed both as reduced abundance of the ISH staining puncta at the single cell level and the density of expressing cells in the given region (Figure 9 E)."

Figure 1—figure supplement 1 is very informative, but requires a lot of searching to find the panel that is referred to in the text. In Figure S1-7/7-M, panels M1-4 are identical to Figure 1E-H and the scale bar in M3 is different to 1G.

We thank the reviewer once more for this careful examination. We have revised the entire text aiming to make the citations more precise. We have added the number ID of regions from the figure 3, to the text when they were mentioned. We would also like to explain that we kept the duplicated panels thinking that the Figure 1—figure supplement 1, which is a comprehensive mapping by itself, could be used independently for readers for anatomical reference. The Figure 1E-H were taken from this figure as examples of the DISH technique demonstration. In other words, they are the same pictures but aiming to show different aspects in two different places; hence, for SI-Figure 1's comprehensiveness we would like to keep them in both files. We have added a sentence in the figure legend of Figure 1 “Note: this figure contains excerpts from the more comprehensive Figure 1—figure supplement 1”.

We thank the reviewer for pointing out the inconsistence of the scale bars – the M3 was wrongly labelled as "100µm" and is now corrected to "250µm". We have also made a thorough revision/edition of this figure, all the panels, mainly adding the missing region labels and correcting the wrongly located arrows and completed the figure captions with all the abbreviations defined in accordance with the abbreviation table of the main text.

2) In several places there are anecdotal statements and it is not clear about the reproducibility of the results. The methods for quantification (including those mentioned in Table legends) should be included in Materials and methods.

We have corrected anecdotal statements as requested. The deleted segments are listed as appendix of this document.

We have added the following sentences in the Materials and method section, as well as the beginning of the "Results" sections and have unified the descriptions in the whole text using the terms "weak", "low", "moderate", "intense" and "very intense" (labelled in italics) only, to give a semiquantitative notion for each given description.

"For semi-quantitative scoring we used similar criteria from literature (Hannibal, 2002). Concretely, we used the annotation criteria as the following: the percentage of expressing cell/total Nissl stained nuclei: “-”, not observed; “+”, weak (<20%); “++”, low (20%-40%); “+++”, moderate (40%-60%); “++++”, intense (60%-80%); “+++++”, very intense (>80%)".

For animals, please check and state the total number of mice and rats used in the study, and whether EGFP mice were also used.

We have now added clearly in Materials and methods the details for quantification for both figures and tables and numbers of animals used. This study did not include EGFP mice (the sentence in the text was a reference to the work of Condro et al. and is now so indicated). Text yext refers to the total number of mice used for distribution studies by DISH analysis.

For c-fos experiments, how were these cells counted, how many sections per mouse, what was the section thickness, how were the values calculated (mean, absolute numbers). Was fos counting done blind to genotype?

We thank the reviewer for raising this question which we have missed to specify in the previous version. The following paragraph is now added to the text:

"The counting for fos and transporters expressed cells was done on a computer screen connected to a digital camera mounted over a light microscope. The region of interest (ROI) was centred through observation using the microscope oculars and 20x objective and projected to a large computer screen through a digital camera. A fixed square equivalent to 0.0314mm2 for the magnification/computer enlargement was pre-fixed and moved to the ROI choosing a region with more or less homogenous cell population. Positive cells within the square were counted. We chose 2 sections of the same region from each mouse (n=3) for each region's assessment. The means were obtained averaging the 6 numbers and statistics were performed as described (vide infra). The counting was done by an experimenter blind to genotype (see acknowledgement)."

The thickness of the section was 12 µm as specified in the text.

Was there variation between animals in terms of expression levels/strength? Case/animal numbers in figures would help. It is not clear what is meant throughout by statements such as “strongest”. Is this by density in cells or number/intensity of puncta? For example, section 3.1, retina. What is meant by “higher percentage than previously reported”? Is this referring to both previous reports in mice? Also see Engelund et al. Cell Tissue Res 2010. How many samples and/or mice were examined and how were ganglion cells counted?

These definitions such as "strongest", "moderate" were always referring the number of cells against the total of Nissls stained nuclei as we first mentioned in the table legend of Table 1 in the first version and not referred the number of puncta. We have implemented this in the text added to the Results and Materials and methods sections and we avoided the usage of "strong" or "strongest". Instead, we are now using "intense" and "very intense" to be in accordance with the quantification criteria we stated.

See above for your concern on case/animal numbers.

The retina data was reported in a previous publication from Eiden's group and collaborators that we cited in this study only for the sake of completeness. We cited the full reference.

Similarly, cortical expression in different layers, how were the values of 80% obtained? Again, “highest expression level of PAC1 among all brain regions” is a strong claim, how was this quantified? Subsection “The claustrum (CLA)”, need references/evidence for observations of mouse claustrum percentages. “more than 90%”. “the highest expression of PACAP was observed in the MnPO”.

See the reply above: we have unified the description to only use the terms "weak", "low", "moderate", "intense" and "very intense" (labelled in italics), which correspond to semiquantitative percentage stated in the beginning of the Results and in the Materials and methods sections.

In terms of the olfactory pathways, is there evidence of co-transmission or is this a hypothesis?

We thank the reviewer again for pointing out this particular anecdotal-style mention. Here, we referred to co-transmission of PACAP/PAC1 with VGAT or VGLUT1/VGLUT2/VGLUT3 that we have described in the previous section. However, the glutamate/GABA co-transmission, a point not intrinsically related to this report is also reported in Table 1. For instance, the neurons in the outer plexiform layer of the main olfactory bulb (MOB), besides expressing PACAP mRNA, expressed mRNAs of VGLUT1(+++), VGLUT2 (+++) and VGAT (+). We cannot assure that all the mRNAs could co-localize within the same cell, as for the DISH method detection. We did test that some of cells co-expressed a given VGLUT and VGAT. It is interesting to note that vasopressin mRNA had similar co-expression pattern as PACAP in these cells (see Zhang et al., JNE 2020), which suggest co-expression of these two neuropeptides, together with small neurotransmitters, similar to the case in the magnocellular neurons of the hypothalamic paraventricular nucleus extensively reported.

We modified the first sentence of the paragraph referred by adding: "The olfactory system appears specially to use PACAP/PAC1 as one of its main modes of co-transmission (see section 3. 2. 1), especially within the cell population in the outer plexiform layer of the MOB and in the mitral cell layer of the AOB, in which some cells were observed to co-express VGLUT1/VGLUT2 and VGAT". We deleted the long sentence which did not belong to this result section.

Some claims will need careful revision. E.g. in the Figure 5 legend, the last sentence contradicts the main text.

We thank the reviewer for noting this discrepancy/typographical error. In the Figure 5 legend we meant to say "… Slc17a7 (F), and (not "but") we did not (not "also") observe co-expression within the Slc32a1 cells (G)". It is now corrected. The panel was aimed to show "no-co-expression" which was inserted under that panel.

The finding that 100% of the 3 GABAergic subpopulations expressed PAC1 is a big claim, yet there is no quantification to back this up. How many brain regions were examined, how many mice, sections, counted cells etc.? If it just refers to primary somatosensory cortex, was it all or some layers?

We thank the reviewer for pointing out this paragraph that lacks specifications. We have amended the paragraph. The added words to limit our conclusion are labelled in bold:

"PAC1 mRNA expression in neocortex was widespread with more homogenous aspects concerning the different cortices, except that in the ACA and the entorhinal cortex layers 2-3 and layers 5-6, which showed very intense expression levels (Figure 3, panels B and F) and https://gerfenc.biolucida.net/images?selectionType=collection&selectionId=98. We sampled eight neocortex regions, at two coronal levels, Bregma 0.14mm and Bregma 1.7mm, where we observed that more than 80% of neurons in layers 2-3 and layer 5 expressing PAC1 mRNA (Figure 3—source data 3). As approximately 20% of cortical neurons were GABAergic (Petilla Interneuron Nomenclature, Ascoli et al., 2008), we tested the three main GABAergic cell types in these cortical regions, finding that all of somatostatin (Sst), parvalbumin (PV)and corticotropin releasing hormone (CRH) neurons in the regions we sampled co-expressed PAC1 (Figure 3—figure supplement 3F, G, H)".

We replaced the expression "100%" with "all", which we meant the VGAT neurons, in the fields we examined, all co-expressed PAC1 (see the SI-Tab. 3 caption). In most of the cases there were less than 10 VGAT neurons within the ROI of (0.03mm2), so using a percentage for such small numbers was just not adequate. We are sorry for not having considered this aspect that made confusions.

Table 2 (also applies to parts of Table 1), do blank areas of the table mean not examined? Or should there be “-“ in these areas? For example, the medial septal complex contains vglut2 expressing cells but the corresponding row/column is blank.

Corrected this aspect in all the tables and made definitions for abbreviations. We thank the reviewer for pointing this out.

There is the claim that PACAP mRNA was not found in cell body layers, but in Table 1 it is reported that there is weak expression in VGLUT1+ cells. Since VGLUT1 cells are in the pyramidal cell layer, this seems contradictory. It would be helpful to have a higher power image of CA1 (as for rat in Figure S2). Could expression outside this layer be in subpopulations of GABAergic neurons? Were these examined (blank in Table 1)? DG is also missing from Table 1.

We thank the reviewer again for this careful examination and we apologise again for not revising carefully the table contents. Specifically, those 4 "+" referred to by the reviewer, corresponded to the column of Hannibal, 2002. This has been corrected. The shared-first authors of this work, LZ and VSH, returned to revise each region and their semiquantitative scores and we found, in dorsal hippocampus, there are clear PACAP mRNA expression in the CA2 region. Since this is an important discovery, we have scored "+++" in CA2 and inserted photomicrographs in the Figure 5C as insets. This sentence was added to the text: "However, we report here the marked and selective expression of PACAP in pyramidal neurons of the CA2 region (Figure 4C insets) and in the hilus (vide infra)".

We have also made the following modifications to Table 1: (1) the MOB two of the strong PACAP expressing cell groups were wrongly identified as periglomerular cells – they are corrected as outer plexiform cells and periglomerular cells with different expressing densities; (2) five missing regions were addressed, i. e. dorsal DG and ventral DG, parasubthalamic nucleus and posterior hypothalamic nucleus, and pontine central grey; (3) some expression strengths were corrected (highlighted in the labelled version); (4) we have simplified Table 1 by removing the column for VGLUT3 expression of which there were only a few instances.

PAC1 expression. Subsection “Hippocampal formation” claims it is selective for VGAT cells. But there are clear examples of VGAT- cells in Figure S3B expressing PAC1. What are these?

The term “selective expression” of PAC1 mRNA in VGAT-positive cells in CA subfields in the first version is meant to convey that most PAC1-expressing cells in CA subfields are VGAT-positive, while the (far more numerous; presumptively excitatory) neurons are mainly non-PAC1-positive. It is evident that there were two cells in the referred field that expressed PAC1 but clearly not VGAT that we have not identified. By the way, this figure and panel is now Figure 3—figure supplement 4 because of the addition of Vipr1 and Vipr2 that have been called S3 and S4.

3) Suggestion about paracrine/autocrine signalling. Is there is evidence in literature for such a role? This seems speculative without immunohistochemical evidence. Hannibal, 2002, carried out at both the protein and mRNA levels, showed axon terminals in multiple regions. Can these be mapped to the regions that express PAC1 in mice? Is there any evidence or could the authors comment on the existence of presynaptic PACAP receptors? Expression of PAC1 mRNA does not imply that the cell would express the protein exclusively along its somatodendritic membrane. “Classical” neurotransmission presumably could occur in PACAP/PAC1 rich regions via local axons in addition to long-range axons.

According to modern neuroendocrinology the autocrine, paracrine mechanisms are defined by the co-expression of a given neuropeptide and its receptor(s) or expression of its receptor in the adjacent cells, since the soma-dendritic release of neuropeptides are widely documented (for recent reviews see Leng, 2018; Colin H Brown, Mike Ludwig, Jeffrey G Tasker, Javier E Stern. Somato-dendritic vasopressin and oxytocin secretion in endocrine and autonomic regulation. J. Neuroendocrinol. 2020 Jun;32(6):e12856. doi: 10.1111/jne.12856. Epub 2020 May 14.). In our case, we suggested the autocrine/paracrine mechanisms for PACAP/PAC1 signalling only at DISH level, by showing that PACAP and PAC1 mRNAs were expressed withing the same cells and their adjacent cells, as we showed in Figure 3. Please note that this notion does not exclude that the same peptide and its receptor(s) can use the neurocrine mechanisms (secretion from synaptic cleft), even within the same cell. We just wanted to emphasise that we are aware of all those possibilities, but we consider it is merited the demonstration, at DISH level, the notion for autocrine/paracrine mechanisms for PACAP/PAC1 signalling.

4) The observation of PACAP in part of temporal CA3, which the authors refer to as CA3c, has in fact previously been defined as CA3vv, corresponding to the coch expressing domain (see Thompson et al., 2008, Fanselow and Dong, 2010). PACAP may indeed be an additional marker along with calretinin for this principal cell subpopulation, and they may want to revise their model or refer to these earlier papers.

We now reference the detailed genomic “map” of mouse hippocampus of Thompson et al., 2008, as reviewed by Fanselow and Dong, 2010, to put in context our observation of PACAP-positive neurons of CA3c as likely CA3vv, although we have not co-stained with Coch mRNA to confirm this. We thank the reviewer for helping us to give this point the attention it deserves.

5) PACAP KO. Some clarification would be welcome in terms of animal cohorts. Please state the experimental unit (i.e. n=9 mice/group). In D, the freezing data show only 8 mice, was one pair excluded due to lack of freezing in an animal, as for jumping mice in C? In Ai, Aii, Bi, Bii, does this show the traces for the total time?

The subjects for behavioural assessment (n=9) were all from the same cohort (performed in March 2019, during a study visit that the two first authors lead the experiments, performed in NIMH, with duly permissions from our institutions). The experiments for the double odour stimulus (n=3, performed in June 2019 in NIMH, led by LZ and VH) and for the olfaction cookie retrieval test (n=3 for each gender, N=12, performed oct 2020) were from other cohorts. Besides, due to the incomplete documentation of the experiment of June 2019 (LZ and VH visiting NIMH) and the current difficulty to repeat this ancillary experiment, we have deleted the mention of former experiment (n=3, double amount of odour stimulus produced higher freezing behaviour in both groups) and replaced with the new experiment of cookie retrieval.

The reason for the n=8 in behavioural assessment analysis for freezing is because that two outliers, one in each experimental group, were excluded. We have added the clarification at the end of the figure legend of Figure 8: "Note that during the freezing analysis, two outliers, one in each experimental group were discarded because being 45 and 24 (counts), with 5 times and 11 times higher than the standard deviations for the corresponding groups, i.e. WT:14.29 ± 6.17 (mean ± SD) vs KO: 3.18 ± 1.80 (mean ± SD). This exclusion was based on criteria explained in NIH Rigor and Reproducibility Training course, https://www.nigms.nih.gov/training/documents/module4-sample-size-outliers-exclusion-criteria.pdf and http://www.itl.nist.gov/div898/handbook/prc/section1/prc16.htm."

In the separate experiment, was n=3 a separate cohort of mice or from the N=18 total as stated in the Materials and methods? Is the n=3 per group or total mice? This may require an increase sample size for this claim, or show quantification/statistical test. For this test, were experimenters also blind to the genotype? The last sentence is difficult to follow.

In light of the importance to establish that PACAP-deficient mice are capable of detecting odors, we performed a more explicit simple behavioral assessment of olfaction in these mice (Yang and Crawley, Current Protocols in Neuroscience 8.24.1-8.24-12, July 2009). We have added the following description in place of our previous description of olfaction-testing in wild-type versus PACAPko mice: “An equal number of male and female mice (n=6 wild-type mice, 3 each male and female, and n=6 PACAPko mice, 3 each male and female) were tested for the ability to retrieve a buried cookie according to the procedure described by Yang and Crawley, Current Protocols in Neuroscience 8.24.1-8.24-12, July 2009. Briefly, mice were individually housed under standard light-dark conditions for a period of four days. Subsequently, a cookie (Chocolate Teddy Grahams—Nabisco) was placed in each cage at 11:45 a.m. (i.e. halfway through the day of the standard 12:12 light/dark cycle). The following day, cookie consumption was confirmed, and a second cookie placed in each cage, consumption confirmed by end of day, and all food removed at approximately 4 p.m. The next day, mice were transported to the testing room, with cages re-labeled for the blinded observer at 10:45 a.m. One hour later, testing was initiated by placing mice in a fresh cage (test cage) with bedding 3 cm deep. Mice were allowed to explore the cage for 5 min; removed to a fresh cage while a cookie was buried 1 cm deep at one end of the test cage, and the mouse placed at the opposite end to the buried cookie in the test cage. Time for mouse to retrieve the buried cookie and hold the cookie in its forepaws and begin to eat was recorded. Mice were tested in pairs in a biosafety hood containing two pairs of test/holding cages at a time; testing was completed by 1:45 p.m. (3 hours after initial transport of mice to testing room). Retrieval times for wild-type and PACAPko mice were not statistically significantly different (wild-type retrieval latency 38.0 +/- 10.4 seconds; PACAPko retrieval latency 44.5 +/- 12 seconds, means +/- s.e.m., n=6/group).

We believe that the use of a separate cohort of mice, using a standard test for olfaction, clarifies the issue of whether or not anosmia or hypo-osmia might complicate the interpretation of the experiments described for response to predator odor: we conclude that it does not.

For the behavioural tests, please include details about whether the wooden boxes, room and experimenter were familiar to the mice before the test (which could affect variability), whether mice were tested at the same time of day, and if KO and WT animals were housed together.

All behavioural experiments were performed in the Institutes animal facility's experiment room, next to their usual residential room. The mice were housed 4-5 per cage. The wooden box was located within a functioning hood with low noise level and dim light. The experimental subjects were introduced for the first time to the environment (no previous habituation). All were tested between 10am-2pm of the same day.

In the Discussion, can the authors comment on or provide evidence of possible developmental changes / compensatory mechanisms occurring in the KO animals.

We cannot presently comment on whether developmental changes or compensatory mechanisms occur in PACAP KO mice: we have embarked upon detailed transcriptomic changes in PACAPko compared to WT mice to address this issue. To do this, we have developed several lines of conditional PACAP knock-out mice and are breeding them with various Cre-driver lines to assess (a) basal transcriptome differences and (b) whether these are affected by PACAP ablation at different developmental stages. This very interesting question will be the subject of future reports in which we will reference this initial report, and ascribe transcriptomic changes noted here either to “constitutive”/developmental effects of PACAP loss, or acute effects contingent upon neuronal activation as evidenced by fos expression. At this time, we can say that for PACAPergic neurons activated by predator odour exposure, there is down-regulation, in a regiospecific way, of either VGAT (e.g. ansiform lobule of cerebellum) or VGluT1 (e.g. PFC) in fos-activated PACAPergic neurons. We hypothesize this effect as occurring as a result of PACAP release in these areas, but cannot specify the mechanisms of regulation until we have completed the developmental studies sketched out above.

Reviewer #2:

Part 1: the PACAP/PAC1 characterization is well designed and executed. The result description is lengthy and sometimes confusing. Figures and tables (including the supplementary information) are clear and informative. The authors decide to do not show Vipr1/Vipr2 data, which should be reconsidered for publication. Overall, this part of the manuscript represents a nice piece of work and surely will be very helpful to whom wish to work with PACAP/PAC1.

We first would like to thank the reviewer for the very careful and detailed revision and criticisms, which helped us to improve the quality of this manuscript. As for the first concern, please kindly see our replies to Reviewer 1 and the appendix listing the deleted segments that were not indispensable and rendered the previous version "lengthy and sometimes confusing".

For Vipr1 and Vipr2 expression mapping, upon receiving your request, we have added two tables and two figures (photo galleries) to the supplementary information to provide a more comprehensive description, that we think they can nicely complement the information presented in the main manuscript. However, we also added these two lines in the new version: " To simplify this already extensive report, we present the data for these two receptors in Figure 3—figure supplements 1 and 2 and source data 1 and 2. " and deleted the specific region description from the previous version: "Vipr1 (VPAC1) expression was widespread, like PAC1 mRNA, in cerebral cortex, hippocampal formation (prominently in the mossy cells), structures derived from cerebral subplate and cerebral nuclei, as well as hypothalamus. The cerebellar cortex in the flocculus and deep cerebellar nuclei moderately expressed, and the Purkinje cells strongly expressed Vipr1; Vipr2 (VPAC2) was observed to be strongly and selectively expressed in the MOB, the mitral and granule layers, the BNSTov, CEAc, lateral division, the SCN, ventral anterior, posterior, posterior medial and lateral geniculate nuclei of thalamus, hypothalamic preoptic area, suprachiasmatic nucleus, inferior, midbrain inferior colliculus, interpeduncular nucleus, periaqueductal gray, and superior colliculus, , hindbrain dorsal tegmental nucleus, cranial nerve nuclei III, V nucleus of the lateral lemniscus, pontine reticular nucleus, superior olivary complex, nucleus raphé pontis, and paragigantocellular reticular nucleus in medulla the nucleus of the trapezoid body and the facial motor nuclei showed a high expression, and in the cerebellum the paraflocculus and flocculus granule and molecular layer showed high expression of VipR2 (SI table 1 and 2 and SI-Figure 2 and 3)."

We mention here that for the DISH experiment we used the channel 1 for either Vipr1 or Vipr2 probe(s) (weak blueish green punctate labelling which was very sensitive to dryness of the sample and more prone to fade with time) and VGAT mRNA probe in channel 1 (strong red labelling) so the blue signal can only be seen at high magnification. The photo-galleries presented in the Figure 3—figure supplement 1 and Figure 3—figure supplement 2, contained reactively high-resolution pictures but with high files sizes which may not be allowed to attach to the eventual publication. Hence, we plan to offer to provide the original files to interested readers.

Part 2: I think this part is the critical one in this manuscript. Starting from section 4, it uses the part 1 of the manuscript to review the literature and build a neuronal circuit with PACAP/PAC1 that makes for behavioural processes. It is literally a review inside the Results section. The schematic figures are interesting but also quite speculative regarding brain signalling since the authors did not performed any experiment to investigate the pathway of PACAP and the literature is scarce. Moreover, the role of Vip receptors were completely neglected here.

The main aim of the study was to place PACAP/PAC1 signalling within the context of the glutamate/GABA co-transmission in two scenarios, anatomical/embryological and basic sensorimotor circuits, to reveal its conspicuous role. We have added a short paragraph at the end of the Introduction aiming to better explain this design: " To address these issues, we conducted a systematic analysis placing the PACAP>PAC1 signalling into anatomical and basic sensorimotor circuit contexts. We first describe in detail the overall topographical organization of expression of PACAP mRNA and its predominant receptor PAC1 mRNA, and their co-expression with the small-molecule transmitters, glutamate and GABA, using VGLUT1, VGLUT2, and VGAT mRNAs, in mouse brain. We then examined the distribution of PACAP/PAC1 hubs within well-established sensory input-to-motor output pathways passing through the cognitive centers, within the context of glutamate/GABA neurotransmission. This systematic analysis has revealed several possible PACAP-dependent networks involved in sensory integration allowing environmental cues to guide motor output."

The so called "part 2" is a key part of this work because it provides a link between the anatomical findings and possible roles from system biology perspective and also toward identification of physiological factors, which could further up- or down-regulate the strength of this PACAP/PAC1 signalling pathway.

We also would like to state that the circuits we presented in this manuscript are all well-established basic circuits that most can be found in textbooks of physiology and classical literature, that it was not our aim to make any additions and to discuss the connectivity of any circuit components. They only serve the purpose for the present study as a functional Atlas / background, in resonance with the Allen Mouse Brain Atlas or Paxinos Mouse Atlas we used in the anatomical part, to project the PACAP-PAC1 hubs, to allow the functional significance of this peptide together with its receptor(s) and molecular signatures and anatomical locations to emerge at a systems level.

https://doi.org/10.7554/eLife.61718.sa2

Article and author information

Author details

  1. Limei Zhang

    1. Department of Physiology, Faculty of Medicine, National Autonomous University of Mexico, Mexico City, Mexico
    2. Section on Molecular Neuroscience, National Institute of Mental Health, Intramural Research Program, Bethesda, United States
    Contribution
    Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing - review and editing
    Contributed equally with
    Vito S Hernandez
    For correspondence
    limei@unam.mx
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-7422-5136
  2. Vito S Hernandez

    Department of Physiology, Faculty of Medicine, National Autonomous University of Mexico, Mexico City, Mexico
    Contribution
    Conceptualization, Data curation, Software, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - review and editing
    Contributed equally with
    Limei Zhang
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-1486-1659
  3. Charles R Gerfen

    Laboratory of Systems Neuroscience, National Institute of Mental Health, Intramural Research Program, Bethesda, United States
    Contribution
    Resources, Data curation, Software, Writing - review and editing
    Competing interests
    No competing interests declared
  4. Sunny Z Jiang

    Section on Molecular Neuroscience, National Institute of Mental Health, Intramural Research Program, Bethesda, United States
    Contribution
    Data curation, Investigation, Visualization, Writing - review and editing
    Competing interests
    No competing interests declared
  5. Lilian Zavala

    Department of Physiology, Faculty of Medicine, National Autonomous University of Mexico, Mexico City, Mexico
    Contribution
    Software, Formal analysis, Visualization, Writing - review and editing
    Competing interests
    No competing interests declared
  6. Rafael A Barrio

    1. Section on Molecular Neuroscience, National Institute of Mental Health, Intramural Research Program, Bethesda, United States
    2. Department of Complex Systems, Institute of Physics, National Autonomous University of Mexico (UNAM), Mexico, Mexico
    Contribution
    Formal analysis, Validation, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-0987-0785
  7. Lee E Eiden

    Section on Molecular Neuroscience, National Institute of Mental Health, Intramural Research Program, Bethesda, United States
    Contribution
    Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Methodology, Project administration, Writing - review and editing
    For correspondence
    eidenl@nih.gov
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-7524-944X

Funding

Consejo Nacional de Ciencia y Tecnología (CB238744)

  • Limei Zhang

National Institute of Mental Health (NIMH-IRP-1ZIAMH002386)

  • Lee E Eiden

Universidad Nacional Autónoma de México (IN216918)

  • Limei Zhang

Universidad Nacional Autónoma de México (G1200121)

  • Limei Zhang
  • Rafael A Barrio

Consejo Nacional de Ciencia y Tecnología (CB283279)

  • Rafael A Barrio

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We apologize to colleagues whose contributions may have been overlooked in citation of the relevant literature for this report. We thank Manuel Hernández for producing the modified open-field box for the predator odor test, to students Enrique C Guerra, Anil K Verma, Sean Sweat and Mario A Zetter for technical assistance and to Angel Fermín Barrio-Zhang for the drawing of the experimental design. LZ was a Fulbright visiting scholar to NIMH-IRP, NIH. LZ and RAB were on sabbatical stay hosted by LEE (NIMH), supported by PASPA-DGAPA-UNAM fellowships. LZ also thank Peter Somogyi and Rafael Lujan for hosting her academic research visits and discussions when part of this manuscript was developed. We thank Peter Somogyi and Rafael Lujan for critical reading and comments on an early version of this manuscript.

Grants: UNAM-DGAPA-PAPIIT-IN216918, IG200121 (LZ) and CONACYT-CB-238744 (LZ), CB-283279 (RB) and NIMH-IRP-1ZIAMH002386 (LEE).

Ethics

Animal experimentation: This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All experiments were approved by the NIMH Institutional Animal Care and Use Committee (ACUC, LCMR-08) and conducted in accordance with the NIH guidelines.

Senior Editor

  1. Ronald L Calabrese, Emory University, United States

Reviewing Editor

  1. Rebecca Shansky, Northeastern University, United States

Reviewers

  1. Tim James Viney, University of Oxford, United Kingdom
  2. Gilliard Lach, University of Edinburgh, United Kingdom

Publication history

  1. Received: August 5, 2020
  2. Accepted: January 18, 2021
  3. Accepted Manuscript published: January 19, 2021 (version 1)
  4. Version of Record published: February 10, 2021 (version 2)

Copyright

This is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.

Metrics

  • 625
    Page views
  • 135
    Downloads
  • 0
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Neuroscience
    Daniela Saderi et al.
    Research Article Updated

    Both generalized arousal and engagement in a specific task influence sensory neural processing. To isolate effects of these state variables in the auditory system, we recorded single-unit activity from primary auditory cortex (A1) and inferior colliculus (IC) of ferrets during a tone detection task, while monitoring arousal via changes in pupil size. We used a generalized linear model to assess the influence of task engagement and pupil size on sound-evoked activity. In both areas, these two variables affected independent neural populations. Pupil size effects were more prominent in IC, while pupil and task engagement effects were equally likely in A1. Task engagement was correlated with larger pupil; thus, some apparent effects of task engagement should in fact be attributed to fluctuations in pupil size. These results indicate a hierarchy of auditory processing, where generalized arousal enhances activity in midbrain, and effects specific to task engagement become more prominent in cortex.

    1. Neuroscience
    Pratish Thakore et al.
    Research Article

    Cerebral blood flow is dynamically regulated by neurovascular coupling to meet the dynamic metabolic demands of the brain. We hypothesized that TRPA1 channels in capillary endothelial cells are stimulated by neuronal activity and instigate a propagating retrograde signal that dilates upstream parenchymal arterioles to initiate functional hyperemia. We find that activation of TRPA1 in capillary beds and post-arteriole transitional segments with mural cell coverage initiates retrograde signals that dilate upstream arterioles. These signals exhibit a unique mode of biphasic propagation. Slow, short-range intercellular Ca2+ signals in the capillary network are converted to rapid electrical signals in transitional segments that propagate to and dilate upstream arterioles. We further demonstrate that TRPA1 is necessary for functional hyperemia and neurovascular coupling within the somatosensory cortex of mice in vivo. These data establish endothelial cell TRPA1 channels as neuronal activity sensors that initiate microvascular vasodilatory responses to redirect blood to regions of metabolic demand.