Ascending Control of Arousal and Motivation: Role of Nucleus Incertus and Its Peptide Neuromodulators in Behavioural Responses to Stress S

Ascending Control of Arousal and Motivation: Role of Nucleus Incertus and its Peptide Neuromodulators in Behavioural Responses to Stress S. Ma*† and A. L. Gundlach*†‡ *Neuropeptides Division, The Florey Institute of Neuroscience and Mental Health, Melbourne, VIC, Australia. †Florey Department of Neuroscience and Mental Health, The University of Melbourne, Melbourne, VIC, Australia. ‡Department of Anatomy and Neuroscience, The University of Melbourne, Melbourne, VIC, Australia.

Journal of Arousal is a process that involves the activation of ascending neural pathways originating in Neuroendocrinology the rostral pons that project to the forebrain through the midbrain reticular formation to pro- mote the activation of key cortical, thalamic, hypothalamic and limbic centres. Established modulators of arousal include the cholinergic, serotonergic, noradrenergic and dopaminergic networks originating in the pons and midbrain. Recent data indicate that a population of largely GABAergic projection neurones located in the nucleus incertus (NI) are also involved in arousal and motivational processes. The NI has prominent efferent connections with distinct hypothalamic, amygdalar and thalamic nuclei, in addition to dense projections to key brain regions associated with the generation and pacing of hippocampal activity. The NI receives strong inputs from the prefrontal cortex, lateral habenula and the interpeduncular and median raphe nuclei, suggesting it is highly integrated in circuits regulating higher cognitive behaviours (hippocampal theta rhythm) and emotion. Anatomical and functional studies have revealed that the NI is a rich source of multiple peptide neuromodulators, including relaxin-3, and may mediate extra- hypothalamic effects of the stress hormone corticotrophin-releasing factor, as well as other key modulators such as orexins and oxytocin. This review provides an overview of earlier studies and highlights more recent research that implicates this neural network in the integration of Correspondence to: Dr S. Ma and Dr A.L. Gundlach, arousal and motivated behaviours and has begun to identify the associated mechanisms. Future The Florey Institute of Neuroscience research that should help to better clarify the connectivity and function of the NI in major and Mental Health, 30 Royal Parade, experimental species and humans is also discussed. Parkville, Victoria 3052, Australia (e-mail: sherie.ma@ﬂorey.edu.au Key words: nucleus incertus, arousal, stress, relaxin-3, motivation, corticotrophin-releasing and factor, orexin, oxytocin andrew.gundlach@ﬂorey.edu.au). doi: 10.1111/jne.12259

neural inputs and projection targets of these subregions are quite Introduction similar (4,5). In the mouse (6) and macaque (2), the neuroanatomi- The nucleus incertus (NI) was first identified in the human brain by cal definition of NIc and NId are less clear and the descriptions George Streeter in 1904 (1), who coined the name, which is Latin provided here will refer to the nucleus as a whole. for ‘uncertain’ nucleus, based on its unknown function. In the non- In terms of connectivity, although there are some similarities in human primate (2) and other mammals examined to date (3), the the neural projections of NI and caudal dorsal raphe (DRc) neuro- NI consists of a midline, bilateral cluster of large, multipolar neurones with regard to the initial course of axons/fibres through the nes in the central grey which lie immediately posterior to the dorsal pontine reticular core, their terminal fields in forebrain regions such raphe nucleus (DR) and are bordered laterally by the dorsal teg- as the septum, nucleus accumbens and hippocampus are quite dis- mental nucleus and ventrally by the medial longitudinal fascicular tinct (4). Therefore, although reports in the literature have often white matter tracts. In the rat, there are two distinct subregions of mistakenly regarded the NI as a caudal extension of the DRc or the NI based on cytoarchitecture - midline columns of compact even as other brain structures (7–9), it is important that the dis- neurones termed NI pars compacta (NIc) and lateral wings of distinct anatomical characteristics and functional roles of the NI persed neurones termed the NI pars dissipata (NId), although the should be recognised and evaluated separately. 458 S. Ma and A. L. Gundlach

RSC mPFC Hip SC

LHb Thal

MS DRC 5-HT NI

MR LHorexin IPN RPO MCH

Fig. 1. Major neural inputs and outputs of the nucleus incertus. Schematic representation of the major neural inputs to and outputs from the nucleus incertus (NI) in the rat brain (see text and Table 1) (4,5). Putative glutamatergic inputs are indicated in green, inhibitory inputs in red, and unknown in blue; and major ascending NI output pathways assessed by anterograde tract-tracing studies are indicated in yellow. DRC, caudal part of dorsal raphe nucleus; Hip, hippocampus; IPN, interpeduncular nucleus; LH, lateral hypothalamus; LHb, lateral habenula; MCH, melanin-concentrating hormone; mPFC, medial prefrontal cortex; MR, median raphe nucleus; MS, medial septum; NI, nucleus incertus; RPO, reticularis pontis oralis; RSC, retrosplenial cortex; SC, superior colliculus; 5-HT, 5-hydroxytryptamine.

In rat brain, the major neural inputs to NI (Fig. 1 and Table 1) field, and continues through to the IPN and rostrally into the hypo- arise from cortical prelimbic (and to a lesser extent the infralimbic), thalamus, innervating various hypothalamic regions via the medial anterior cingulate, medial and ventral orbital areas, generally from forebrain bundle, including lateral and medial mammillary and su- pyramidal neurones of layer V (4). Other major inputs arise from pramammillary nuclei, and the lateral hypothalamic and lateral pre- neurones in medial regions of the lateral habenula (9), which are optic areas. A branch of these fibres courses dorsally to innervate densely innervated by dopamine and neurotensin fibres (10,11) and the mediodorsal, centromedial, centrolateral and paraventricular express serotonin-2c receptors (12). Neurones in the medial part of thalamic regions. A substantial group of fibres ascend through the the lateral habenula also provide projections to the intermediate hypothalamus to innervate the nucleus of the diagonal band and part of the interpeduncular nucleus (IPN) (13) and median raphe MS, and pass through the fornix and fimbria to innervate the dorsal nucleus (MR) (9) and may play a role in the maintenance of rapid- and ventral hippocampus, as well as the amygdala. Finally, a fibre eye movement (REM) sleep (14). Notably, neurones of the IPN and bundle continues rostrally from the hypothalamus to innervate the MR also provide dense inputs to the NI (4), suggesting there are infralimbic, prelimbic and anterior cingulate cortex. Some of these strong bidirectional (reciprocal) neural circuits between the NI, lat- fibres extend through the midline cingulum bundle, providing inner- eral habenula, MR and IPN. vation of the anterior cingulate and posterior retrosplenial cortices Moderate neural inputs to the rat NI arise from inhibitory neuro- (4,5) (Fig. 1 and Table 1). The cell types innervated by NI remain to nes in the medial septum (MS) and diagonal band (4,15) and, be characterised, although recent histological studies have started because the septum is a principal target of NI projections (4,5), this to assess this in the densely innervated MS, and indicate that NI input may provide an inhibitory feedback regulation of a ponto- terminations are located on both cholinergic neurones and inhibi- septohippocampal network (16). Other neural inputs stem from the tory neurones expressing parvalbumin and calbindin (19), including lateral preoptic and lateral hypothalamic areas, particularly the peri- those that provide further projections to the dorsal hippocampus, fornical region (4,17), periaqueductal grey (PAG) and DRc, although suggesting broad modulatory actions of the septohippocampal net- cell types are unknown (4). In the brainstem, small and medium work. neurones within a region immediately caudal to the laterodorsal By contrast to the detailed analysis in the rat, there have not tegmental nucleus are also retrogradely labelled from the NI, which been any equivalent focused studies of the connectivity of the NI is a source of cholinergic neurones interacting with habenular and in the mouse, which is the other major experimental species being meso-dopaminergic circuits underlying motivated behaviours (18). used in related functional studies of the NI and its peptide neuro- Furthermore, NI neurones also appear to send projections to their modulators. Although the NI is recognised in major anatomical contralateral equivalent, suggesting the existence of a local inhibi- atlases of the mouse brain (20) (also termed ‘nucleus O’ and ‘cen- tory microcircuit (4). tral grey alpha’) (21) and several histological studies (6,22), the dis- With regard to NI outputs, anterograde tracing studies in the rat tribution of neural projections from and inputs to the mouse NI demonstrate that NI ascending projections are long-range and arise has not been described, apart from some data available in the pub- as two major fibre bundles (4,5) (Fig. 1 and Table 1). The first effer- lic access Allen Brain Institute Atlas (http://www.brain-map.org), ent projection group courses through the pontine periventricular which reveals, for example, a strong innervation of the NI region region to innervate the DR, PAG and superior colliculus. Another from the prelimbic cortex. In light of recent advances in viral-based efferent group courses into the MR, generating a dense terminal methods for neural tract tracing (23,24), their application in this

Table 1. Major Neural Inputs and Outputs of the Rat Nucleus Incertus. Table 1 (continued)

Input to Innervation by Input to Innervation by Brain region nucleus incertus nucleus incertus Brain region nucleus incertus nucleus incertus

Cerebral cortex Rhomboid nucleus À ++ Anterior cingulate cortex ++++ ++ Zona incerta À ++ Claustrum À ++ Basal ganglia Infralimbic prefrontal cortex ++ ++ Nucleus accumbens À + Orbital prefrontal cortex ++++ + Ventral tegmental area À +++ Prelimbic prefrontal cortex ++++ ++ Brainstem Retrosplenial cortex +++ Dorsal raphe nucleus À ++++ Hippocampus – dorsal Interpeduncular nucleus ++++ ++++ CA1 À + Median raphe nucleus ++++ ++++ CA2 À + Nucleus incertus (contralateral) +++ +++ CA3 À ++ Nucleus prepositus À +++ Dentate gyrus À + Periaqueductal grey ++ ++++ Fimbria À +++ Pontine reticular nucleus +++ ++ Hippocampus – ventral Superior colliculus À ++ CA1 À ++ CA2 À + A summary of data from Goto et al. (4), Olucha-Bordonau et al. (5,19) and CA3 À +++ Teruel-Marti et al. (88). Scale: ++++, very dense; +++, dense; ++, moderate; Dentate gyrus À ++++ +, sparse; À, not present. Subiculum À ++++ Amygdala Amygdalohippocampal area À ++++ context would be an excellent way to document these details; and Basolateral nucleus À + would greatly assist the design and interpretation of functional Basomedial nucleus À ++ studies in relevant transgenic mouse models. Central nucleus À + Medial nucleus À + Septum Chemoarchitecture: NI is a rich source of peptide Bed nucleus stria terminalis À ++ neuromodulators and a primary source of relaxin-3 À +++ Lateral septum The heterogeneous chemoarchitecture of the NI in the rat and Medial septum ++ ++++ mouse is better characterised than in other species. For example, in Nucleus of diagonal band ++ ++++ the rat, NI neurones are distinguished from adjacent DR neurones Hypothalamus Anterior hypothalamic nucleus À ++ by their lack of serotonin content and their expression of the GABA Arcuate nucleus À + synthesising enzyme, glutamic acid decarboxylase (GAD) (25) and Dorsomedial nucleus À ++++ expression of mRNA and immunoreactivity for several neuropep- Lateral hypothalamic area/medial +++ ++++ tides, including cholecystokinin (CCK) (5,26), neurotensin, a ranaten- forebrain bundle sin-like peptide (27) (subsequently identiﬁed as neuromedin-B) Lateral mammillary nucleus À ++++ (NMB) (28) and particularly relaxin-3 (20,29,30); a review is also Lateral preoptic area + +++ provided elsewhere (3). NI neurones in the rat also express high ÀÀ Medial mammillary nucleus levels of the calcium-binding proteins, calbindin and calretinin, but À +++ Medial preoptic area not parvalbumin (26,31). Paraventricular hypothalamic À ++ Although the neurochemical anatomy of the NI is less formally nucleus characterised in the mouse than the rat (3), data are available in Periventricular nucleus À +++ Posterior hypothalamic nucleus ++ +++ the Allen Institute Brain Atlas (20,30) suggesting both similarities Supramammillary nucleus ++ ++++ and differences in the neurochemical phenotype of NI neurones in Ventromedial hypothalamic nucleus À + these two major experimental species. In a previous survey (3), we Thalamus used this resource to identify the presence in mouse of mRNA Centromedial nucleus À +++ encoding marker proteins for the major transmitters, GABA and Lateral habenula ++++ ++ glutamate (i.e. GAD65/67 and the vesicular glutamate transporters- Mediodorsal nucleus À +++ 2 and -3) (VGLUT2/3) and mRNAs encoding multiple calcium-bind- À ++ Medial habenula ing proteins and several neuropeptides noted earlier in the rat, À +++ Paraventricular nucleus including relaxin-3, CCK, NMB and dynorphin (Fig. 2). Reuniens nucleus À ++ A particular feature of the NI is its high expression of the neu- (continued) ropeptide, relaxin-3 (Fig. 2) (32,33). In the rat, the NI is the pri-

(A) (B)

RLN3

(C)(D)

VGLUT2 VGLUT3

(E)(F)

GAD67 GAD65

(G)(H)

NMB CRF-BP

(I)(J)

pDyn CCK

Fig. 2. Examples of peptide and transmitter-related enzyme and transporter mRNA species present in mouse nucleus incertus (NI). Expression of peptide, transporter and enzyme mRNA species detected by nonradioactive, digoxigenin-based in situ hybridisation histochemistry in mouse NI and available from the Allen Institute for Brain Science website (http://www.alleninstitute.org). Images are (A) Nissl and schematic of NI and surrounding structures; and in near-adjacent, coronal level sections, mRNA encoding (B) relaxin-3 (RLN3), (C) vesicular glutamate transporter-2 (VGLUT2), (D) VGLUT3, (E) glutamate decarboxylase-67 (GAD67), (F) GAD65, (G) neuromedin B (NMB), (H) corticotrophin-releasing factor binding protein (CRF-BP), (I) pro-dynorphin (pDyn; some moderately labelled neurones are apparent under high-power view) and (J) cholecystokinin (CCK). Images adapted from Allen Institute for Brain Science (19,29). For further information, see also Ryan et al. (3). AQ, cerebral aqueduct; B, Barrington’s nucleus; CENT3mo, lobule III molecular layer; DTN, dorsal tegmental nucleus; KF, Koelli- ker-Fuse subnucleus; LC, locus ceruleus; lc, parabrachial nucleus central lateral part; ld, parabrachial nucleus dorsal lateral part; le, parabrachial nucleus external lateral part; lv, parabrachial nucleus ventral lateral part; MEV, midbrain trigeminal nucleus; mlf, medial longitudinal fascicle; mtv, midbrain tract of the trigeminal; NI, nucleus incertus; PB, parabrachial nucleus; PBmm, parabrachial nucleus medial medial part; PCG, pontine central gray; scp, superior cerebellar peduncles; sctv, ventral spinocerebellar tract; SLD, sublaterodorsal nucleus; SUT, supratrigeminal nucleus.

mary source of this approximately 6-kDa neuropeptide, which is relaxin-3-immunoreactive ﬁbres are complementary in rat and synthesised by approximately 2000 neurones, all of which are mouse brain (6,26,38). GABAergic (26), with another approximately 2000 relaxin-3 The NI in the rat exhibits prominent expression of one of the expressing neurones located in smaller populations in the pontine two receptors for the stress peptide, corticotrophin-releasing factor raphe nucleus, medial PAG and a region dorsal to the substantia (CRF), namely CRF receptor-1 mRNA (CRF-R1) (39,40). In fact, all nigra (34). The cognate receptor for relaxin-3, RXFP3 (also known relaxin-3 neurones in the rat co-express CRF-R1, although they do as GPCR135, SALPR) (35,36), is a Gi/o-protein-coupled receptor, not encompass the entire CRF-R1 positive population (41). In the and its activation produces inhibition of intracellular cAMP accu- mouse, the NI also expresses moderate levels of CRF-R1 (20,42) mulation and extracellular signal-regulated kinase 1/2 activation and, notably, high levels of the CRF-binding protein mRNA (Fig. 2) (35,37). In line with this ligand-receptor pairing, the comparative (20,43), consistent with a regulatory effect of CRF signalling on distributions of RXFP3 mRNA and protein (binding sites) and neurones in the region in mice and rats (further discussed below).

(A) (B) LC V4

DTN NI Dorsal pons

(C)(D)(E) 4V DTg 4V DTg 4V DTg NIc

NId

µm 20 um NeuN CRF-R1 20 NeuN/CRF-R1

(F) CRF CRF CRF 30 mV

–120 250 s

Fig. 3. Corticotrophin-releasing factor (CRF) induces neural activation of nucleus incertus (NI). (A) Fos induction in the NI, as indicated by Fos-immunostaining observed at 2 h after i.c.v. injection of 1 lg CRF. (B) Dense CRF-R1 mRNA expression in the NI visualised by in situ hybridisation; reproduced from Bittencourt and Sawchenko (40) with permission. (C) Immunostaining for neurones with neuronal nuclear antigen (NeuN) and (D) CRF-R1 in the NI (E, merged images) indicate that CRF-R1-positive neurones represent a signiﬁcant population in the NI. (arrows), though not all neurones express CRF-R1 (arrowheads). (F) In vitro current-clamp recording demonstrating the effect of three consecutive CRF applications (scale bar = 100 nM) and illustrating the sustained sensitivity and lack of desensitisation of an NI relaxin-3 neurone to CRF activation; reproduced from Ma et al. (41) with permission. 4V and V4, 4th ventricle; CRF, corticotropin- releasing factor; CRF-R1, corticotropin-releasing factor receptor-1; DTg and DTN, dorsal tegmental nucleus; LC, locus ceruleus; NI, nucleus incertus; NIc, nucleus incertus pars compacta; NId nucleus incertus pars dissipata.

The NI region in the rat and mouse also expresses receptors for Activation of NI neurones by neurogenic stressors and a range of ‘stress’, ‘arousal’ and ‘neuroendocrine factors’ (reflected CRF: modulation of stress and motivated behavioural either by the presence of mRNA and/or binding sites and/or protein responses immunoreactivity), including orexin (44,45), melanin-concentrating hormone (46), NMB (47–49), oxytocin (50), ghrelin (20,51) and sero- Based on substantial anatomical and functional evidence (39,40,53), the NI is a highly ‘stress-reactive’ nucleus (3). Existing data suggest tonin (5-HT1A) (52), although the precise nature of the impact of these diverse signalling systems on NI activity has not been that NI neurones can integrate signals mediated by CRF-containing reported. It has been reported, however, that the serotonergic input and other forebrain neural inputs (41,54) and/or CRF reaching them to NI serves as a negative regulator of relaxin-3 expression via from the CSF (40). Notably, unlike DR neurones, which express both CRF-R1 and CRF-R2 (55), the NI appears to only express CRF-R1 direct actions on 5-HT1A receptors (52), with the serotonergic input likely emanating from the DRc and/or other adjacent serotonergic (40,42,56). Thus, NI neurones (including all relaxin-3 neurones in nuclei. The effects of CRF on NI neural firing have also been inves- the area) only exhibit depolarisation in response to CRF, which is tigated (see below). RXFP3 mRNA and binding sites are also long-lasting and non-desensitising (41) (Fig. 3). This differs from expressed in the NI of rat and mouse (6,26), which may reflect the serotonergic and non-serotonergic DR neurones that display differ- presence of an ipsilateral to contralateral NI connection observed in ential, dose-dependent responses that are rapidly desensitised (55). neural tracing studies (4). In the anaesthetised rat, NI relaxin-3 neurones increased their firing

Journal of Neuroendocrinology, 2015, 27, 457–467 © 2015 British Society for Neuroendocrinology 462 S. Ma and A. L. Gundlach frequency in response to i.c.v. infusion of 1–3 lg CRF, whereas a in the NI, which was associated with elevated plasma corticoste- subpopulation of relaxin-3 negative neurones exhibited decreased rone levels and low expression of CRF mRNA in the PVN (65). Fur- firing (41). Notably, only excitation of NI neurones in response to thermore, in diet-induced obese rats relaxin-3 mRNA levels in the CRF (10–600 nM) was observed in recordings made in 250-lm NI were significantly greater than those in rats that were deemed coronal brain slices maintained in vitro (41). This is consistent with obesity-resistant, but following food deprivation, obesity-resistant the possibility that central CRF actions also lead to indirect inhibi- rats exhibited greater relaxin-3 expression than diet-induced obese tion of a population of NI neurones that are non-CRF-R1 and non- rats (66). relaxin-3 positive; and further studies are required to identify the Acute administration of relaxin-3 peptide into the cerebral ven- phenotype and function of this population. In this regard, some of tricle or direct injection into the PVN is capable of inducing feed- the other neuropeptide or calcium-binding protein markers of NI ing in rats (67) via actions on hypothalamic RXFP3 (68–70). neurones may prove useful. Moreover, chronic RXFP3 agonist secretion within the hypothala- The effect of direct NI stimulation on the activity of the hypo- mus increased both daily food intake and body weight of adult thalamus-pituitary-adrenal (HPA) axis or non-HPA (extra-hypotha- male rats, which was associated with significantly attenuated lamic) CRF neurones has not been examined, although the central hypothalamic oxytocin and vasopressin mRNA levels (71). In mice, administration of relaxin-3 has been reported to regulate the activ- however, central RXFP3 agonist administration is not orexigenic, ity of hypothalamic CRF neurones. Intracerebroventricular infusion although i.c.v. injection of an RXFP3 antagonist can inhibit normal of relaxin-3 in rats resulted in elevated c-fos and CRF mRNA dark phase feeding, feeding following a mild food deprivation expression, as well as Fos-immunoreactivity in CRF neurones within stress, and food anticipatory behaviours that develop following the paraventricular nucleus of the hypothalamus (PVN) (57), which restricted food access (72). Thus, within the hypothalamus, relaxin- was associated with increased plasma levels of adrenocorticotrophic 3/RXFP3 signalling underlies stress-related and motivated feeding hormone (57,58). behaviour. Induction of neurogenic stress in rats, using, for example, a Relaxin-3/RXFP3 signalling is also associated with other moti- forced-swim paradigm, is associated with increased relaxin-3 mRNA vated and stress-sensitive behaviours, such as alcohol self-adminis- in the NI, which is largely CRF-R1-dependent (53). Notably, of the tration and seeking, and relapse behaviour following abstinence various neurogenic stressors that have been shown to activate the (73). An alcohol-preferring rat strain (iP rat) has been reported to NI (by quantification of Fos-immunoreactivity) (3), the NI of rats display an elevated stress/CRF response reflected by enhanced EEG exposed to an environment previously paired with a predator (i.e. a activity following central CRF infusion and decreased brain CRF lev- cat) exhibit decreased Fos immunoreactivity (59). However, immobil- els (74), and infusion of a RXFP3-specific antagonist into the lateral isation stress in rats, produced using a range of methods (restraint cerebral ventricle or directly into the bed nucleus of the stria termi- tube (60,61), fixed lying on their back (61), or limb restraint with nalis (BNST) significantly attenuated lever pressing for alcohol and partial water submersion (34), strongly activates the NI. Thus, NI cue- and stress-induced reinstatement of lever pressing, although activity appears related to ‘active locomotor components’ of there was no significant effect on responding for sucrose (73). In behavioural stress responses and its inactivity appears related to these rats, levels of relaxin-3 mRNA in the NI positively correlated ‘active immobility’. with levels of alcohol and sucrose intake (75). However, it is noted Behavioural studies in rats with lesions of NI neurones are con- that NI projections to the BNST are quite weak (4,5), suggesting the sistent with this idea. Rats in which CRF-R1-positive NI neurones relaxin-3 innervation to this area may also derive from other were selectively lesioned using a CRF-conjugated saporin toxin, relaxin-3 neurone groups. Together, these findings suggest relaxin-3 exhibited significantly prolonged freezing behaviour following the inputs to key hypothalamic and limbic circuits regulate orexigenic presentation of a cue previously associated with footshock and dur- and goal-directed (motivated) behaviours, and are capable of inte- ing the 5 min after cue presentation offset, although there were no grating stress-related external and internal information to modulate impairments in conditioned learning (62). Similarly, electrolytic these behaviours. lesion of the NI has been reported to significantly delay the extinc- However, despite these intriguing findings, the primary involve- tion of conditioned fear behaviour, whereby rats exhibit signifi- ment of the NI relaxin-3 network and/or any impact of the other cantly more freezing during conditioned cue presentations (63). relaxin-3 neurone populations in these various effects have not been Because NI activity appears related to active locomotor components demonstrated directly. Thus, we postulate that NI and associated of behavioural responses, these findings suggest ablation of the NI relaxin-3 neurotransmission may be involved in the initiation and/or results in impairment of active locomotor responses that would maintenance of active motivated behaviour, such as transition from normally be associated with active aversion/withdrawal from a freezing to fight or flight response to elude threat, or stress-induced potential threat and/or during the extinction process, which feeding and reward seeking. It is our goal, therefore, to test these involves new learning that a cue is no longer associated with a ideas in further studies using existing receptor-selective agonists threat (64). and antagonists and novel viral vector-based pharmacogenetic and In terms of relaxin-3 function, recent studies using a rat model optogenetic methods, which can selectively activate or silence neural of binge feeding, observed that female (but not male) rats exhibited pathways (76,77), aiming to dissect the involvement of NI neurone hyperphagic responses to chronic restraint stress (1 h restraint once populations in various complex behavioural responses in awake, a week for 7 weeks) and significantly increased relaxin-3 expression freely-moving rats and mice (78) (as discussed below).

Functions of NI neural networks: activation of the Thus, NI/relaxin-3 septal projections regulate the activity of MS via septohippocampal system and promotion of hippocampal actions on multiple cell groups and MS inhibitory neurones project- theta rhythm and associated cognition ing to the NI (15) form a bidirectional neural circuit regulating hippocampal activity and theta-related cognition. The NI appears to Hippocampal theta rhythm refers to distinct oscillations of 4–12 Hz act as neural node capable of integrating neurogenic stress and detectable in the electroencephalogram of all mammals, in which CRF signalling and the septohippocampal network, to modulate hip- temporal aspects of brain rhythms and their behavioural correla- pocampal theta rhythm and associated arousal and cognitive pro- tions are highly conserved (79). Specifically, theta rhythm has been cesses. Moreover, an NI innervation of the neighbouring lateral implicated in several neural functions and behaviours, including septum (LS) suggests related actions on emotional fear, such as memory consolidation, arousal states such as exploratory behaviour those associated with oxytocin (90), and inhibition of motivated and REM sleep, as well as the control of anxiety and mood (80–83). feeding behaviour (91), which remain to be examined. Hippocampal theta activity is strongly modulated by a septohippo- Taken together, current anatomical and functional data suggest campal system, whereby ‘pacemaker’ GABAergic and cholinergic the NI is centrally positioned in a brainstem network involved in neurones in the MS are capable of generating and pacing this brain the control of behavioural activation, a hypothesis first proposed in rhythm (16,84,85). With its distinct innervation of various brain tar- 2001 by Goto et al. (4) based on neuroanatomical connectivity. gets important in neural circuitry underlying hippocampal theta, Such behavioural arousal is consistent with current functional find- particularly the MS, it was proposed that the NI was a modulator ings that have revealed correlations between NI activity and of the septohippocampal network (86). In anaesthetised rats, elec- relaxin-3 levels with behavioural activity, particularly in relation to trical stimulation of the NI rapidly induced hippocampal theta oscil- stress and its firing coherence with hippocampal theta activity lations, whereas electrolytic lesion of (or muscimol injection into) (41,92). A major population of NI efferents target the septum and the NI resulted in attenuated brainstem-induced hippocampal theta ventral hippocampus, and moderately target the dorsal hippocam- rhythm following stimulation of the reticularis pontis oralis (RPO) pus and amygdala, via a projection through the fimbria-fornix. In (86). Interestingly, the RPO and other brainstem structures do not addition to the mnemonic roles of the hippocampus and theta send direct projections to the MS or other subcortical regions rhythm, this circuit also functions to modulate interrelated emo- involved in pacing theta rhythm (87), but do project strongly to the tional and social behaviour. For example, the lateral septum is an NI (88), suggesting that the NI is a relay node for connecting the essential negative regulator of lordosis/sociosexual (93) and anxi- brainstem with forebrain theta-pacing regions. Relaxin-3 neurones ety-like behaviour (94), which is also strongly modulated by the in the NI display significantly coherent firing during theta oscilla- ventral hippocampus (95). Recent studies also indicate that a spe- tions, whereas non-relaxin-3 neurones do not, which supports a cific subclass of interneurones in the prefrontal cortex, which is a functional link between NI and its relaxin-3 containing neurones primary NI input area and receives moderate NI projections, medi- with this important brain activity (41). ate changes in female sociosexual behaviour in response to oxyto- These findings raise equally important questions about the role cin (96). Thus, there is growing evidence this intriguing brainstem of NI neurones that display a different phenotype, and further nucleus, which signals through a variety of neurotransmitters and studies are required of the overlap or segregation of other peptide peptide neuromodulators, including the highly-conserved peptide modulators in the NI such as NMB, dynorphin and CCK, with relaxin-3, is a regulator of complex behaviour associated with relaxin-3 expression, as well as their functional involvement at the responses to stress, perceived threat and appropriate behavioural level of the NI and/or any specific or convergent target brain planning and selection. regions. Such investigations at the level of the NI are feasible, whereas the abundance throughout the brain of additional neurone populations expressing these various other peptides make the latter Current understanding of the NI is not so uncertain but task more challenging. However, there are powerful new methods there is certainly more to learn based on the use of cell-specific viral vectors that would allow dif- After more than 10 years of ‘steady-paced’ research by a small ferent groups of NI peptide neurones to be controlled in functional number of interested groups on the characteristics of the nucleus behavioural and mechanistic studies (76). incertus (‘uncertain’ nucleus) primarily in rodents (5,29,66,97,98), In the MS, relaxin-3 containing projections from the NI contact we have a far better idea of its potential functional importance various pacemaker cell types, including choline acetyltransferase, under normal physiological conditions in adult rats and mice. How- parvalbumin and GAD67-positive neurones that project to the hip- ever, the studies completed so far, driven primarily by interests in pocampus, as well as inhibitory calbindin- and calretinin-positive the biological roles of CRF and relaxin-3 signalling systems neurones (19). Functionally, local infusion of RXFP3 agonist or (41,57,58), have served to emphasise that, in a similar way to com- antagonist into the MS significantly enhanced or attenuated hippo- parable neural networks, such as the orexin and monoamine sys- campal theta power in awake rats, respectively, and infusion of tems, which act to integrate external and internal signals about the RXFP3 antagonist into the MS resulted in dose-dependent decreases environment and internal metabolic and homeostatic status, via in spatial working memory performance in rats (89). Similarly, in sensory, neuroendocrine and autonomic transmission/inputs, there anaesthetised rats, infusion of RXFP3 antagonist into the MS signif- is much we still need to learn about the ‘heterogeneous’ NI. Future icantly attenuated RPO-induced hippocampal theta power (89). studies need to address the specific roles of different populations

Journal of Neuroendocrinology, 2015, 27, 457–467 © 2015 British Society for Neuroendocrinology 464 S. Ma and A. L. Gundlach of NI neurones under ‘normal physiological’ conditions, as well as 7 Hayakawa T, Zyo K. Comparative cytoarchitectonic study of Gudden’s under experimental conditions that can help reveal key functions tegmental nuclei in some mammals. J Comp Neurol 1983; 216: 233–244. associated with specific behavioural states, gender and/or hormonal 8 Fuchs E, Wasmuth JC, Flugge G, Huether G, Troost R, Beyer J. Diurnal variation of corticotropin-releasing factor binding sites in the rat brain status and/or in pathological models or conditions that reflect the and pituitary. Cell Mol Neurobiol 1996; 16: 21–37. clinical pathology observed in neurodegenerative and psychiatric 9 Quina LA, Tempest L, Ng L, Harris JA, Ferguson S, Jhou TC, Turner EE. diseases. This is an exciting prospect because there are a large Efferent pathways of the mouse lateral habenula. J Comp Neurol range of new and powerful methods available to investigators for 2015; 523: 32–60. probing further aspects of the anatomy, connectivity, chemoarchi- 10 Jennes L, Stumpf WE, Kalivas PW. Neurotensin: topographical distribu- tecture, regulation and physiological function of the NI and its tion in rat brain by immunohistochemistry. J Comp Neurol 1982; 210: extensive neural network. These include pharmacogenetic and opto- 211–224. 11 Phillipson OT, Pycock CJ. Dopamine neurones of the ventral tegmentum genetic methods for the activation and silencing of neurones, using project to both medial and lateral habenula. Some implications for ha- designer receptors exclusively activated by designer drugs benular function. Exp Brain Res 1982; 45: 89–94. (DREADDs), and channelrhodopsins/halorhodopsins, respectively, 12 Aizawa H, Kobayashi M, Tanaka S, Fukai T, Okamoto H. Molecular char- which could be used in combination with relevant transgenic acterization of the subnuclei in rat habenula. J Comp Neurol 2012; mouse lines (99–101) (e.g. RXFP3- or relaxin-3-Cre mice) and viral 520: 4051–4066. vectors (77,102). Studies of this type are underway in our labora- 13 Kim U. Topographic commissural and descending projections of the ha- tory and it is hoped that other systems neuroscience laboratories benula in the rat. J Comp Neurol 2009; 513: 173–187. 14 Aizawa H, Cui W, Tanaka K, Okamoto H. Hyperactivation of the haben- interested in the control of behavioural state and complex behav- ula as a link between depression and sleep disturbance. Front Hum iours will also conduct studies of NI networks to assist the growth Neurosci 2013; 7: 826. of knowledge in the field and the identification of associated trans- 15 Sanchez-Perez AM, Arnal-Vicente I, Santos FN, Pereira CW, ElMlili N, lational and therapeutic opportunities. Sanjuan J, Ma S, Gundlach AL, Olucha-Bordonau FE. Septal projections to the nucleus incertus in the rat: bidirectional pathways for modulation of hippocampal function. J Comp Neurol 2015; 523: 565–588. Acknowledgements 16 Vertes RP, Kocsis B. Brainstem-diencephalo-septohippocampal systems controlling the theta rhythm of the hippocampus. Neuroscience 1997; Research by the authors reviewed in this article was supported by an Aus- 81: 893–926. tralian Biomedical Fellowship from the National Health and Medical 17 Goto M, Canteras NS, Burns G, Swanson LW. Projections from the sub- Research Council of Australia and a Commonwealth of Australia Endeavour fornical region of the lateral hypothalamic area. J Comp Neurol 2005; Fellowship (SM); a Research Fellowship and project grants from the National 493: 412–438. Health and Medical Research Council of Australia, a grant from the Pratt 18 Lammel S, Lim BK, Ran C, Huang KW, Betley MJ, Tye KM, Deisseroth K, and Besen Family Foundations, and a Brain & Behavior Research Foundation Malenka RC. Input-specific control of reward and aversion in the ven- (USA) NARSAD Independent Investigator Award (ALG); and by the Victorian tral tegmental area. Nature 2012; 491: 212–217. Government Operational Infrastructure Support Programme (The Florey Insti- 19 Olucha-Bordonau FE, Otero-Garcia M, Sanchez-Perez AM, Nunez A, tute of Neuroscience and Mental Health). Ma S, Gundlach AL. Distribution and targets of the relaxin-3 innervation of the septal area in the rat. J Comp Neurol 2012; 520: 1903– Received 12 November 2014, 1939. revised 12 January 2015, 20 Allen Institute for Brain Science. Allen Brain Atlas [Internet]. Available accepted 15 January 2015 from: http://www.brain-map.org. 21 Paxinos G, Franklin BJ. The Mouse Brain in Stereotaxic Coordinates. Boston, MA: Elsevier Academic Press, 2004. References 22 Yoshida M, Takayanagi Y, Inoue K, Kimura T, Young LJ, Onaka T, Nishi- mori K. Evidence that oxytocin exerts anxiolytic effects via oxytocin 1 Streeter GL. Anatomy of the floor of the fourth ventricle. Am J Anat receptor expressed in serotonergic neurons in mice. J Neurosci 2009; 1903; 2: 299–313. 29: 2259–2271. 2 Ma S, Sang Q, Lanciego JL, Gundlach AL. Localization of relaxin-3 in 23 Taylor SR, Badurek S, Dileone RJ, Nashmi R, Minichiello L, Picciotto brain of Macaca fascicularis: identification of a nucleus incertus in MR. GABAergic and glutamatergic efferents of the mouse ventral teg- primate. J Comp Neurol 2009; 517: 856–872. mental area. J Comp Neurol 2014; 522: 3308–3334. 3 Ryan PJ, Ma S, Olucha-Bordonau FE, Gundlach AL. Nucleus incer- 24 Weissbourd B, Ren J, DeLoach KE, Guenthner CJ, Miyamichi K, Luo L. tus – an emerging modulatory role in arousal, stress and memory. Presynaptic partners of dorsal raphe serotonergic and GABAergic neu- Neurosci Biobehav Rev 2011; 35: 1326–1341. rons. Neuron 2014; 83: 645–662. 4 Goto M, Swanson LW, Canteras NS. Connections of the nucleus incer- 25 Ford B, Holmes CJ, Mainville L, Jones BE. GABAergic neurons in the rat tus. J Comp Neurol 2001; 438: 86–122. pontomesencephalic tegmentum: codistribution with cholinergic and 5 Olucha-Bordonau FE, Teruel V, Barcia-Gonzalez J, Ruiz-Torner A, Valv- other tegmental neurons projecting to the posterior lateral hypothala- erde-Navarro AA, Martinez-Soriano F. Cytoarchitecture and efferent mus. J Comp Neurol 1995; 363: 177–196. projections of the nucleus incertus of the rat. J Comp Neurol 2003; 26 Ma S, Bonaventure P, Ferraro T, Shen P-J, Burazin TCD, Bathgate RAD, 464: 62–97. Liu C, Tregear GW, Sutton SW, Gundlach AL. Relaxin-3 in GABA projec- 6 Smith CM, Shen P-J, Banerjee A, Bonaventure P, Ma S, Bathgate RAD, tion neurons of nucleus incertus suggests widespread influence on Sutton SW, Gundlach AL. Distribution of relaxin-3 and RXFP3 within forebrain circuits via G-protein-coupled receptor-135 in the rat. Neuro- arousal, stress, affective, and cognitive circuits of mouse brain. J Comp science 2007; 144: 165–190. Neurol 2010; 518: 4016–4045.

27 Chronwall BM, Pisano JJ, Bishop JF, Moody TW, O’Donohue TL. Bio- mRNA expression in the rat brain and pituitary. Proc Natl Acad Sci chemical and histochemical characterization of ranatensin immunore- USA 1994; 91: 8777–8781. active peptides in rat brain: lack of coexistence with bombesin/GRP. 40 Bittencourt JC, Sawchenko PE. Do centrally administered neuropeptides Brain Res 1985; 338: 97–113. access cognate receptors? An analysis in the central corticotropin- 28 Moody TW, Korman LY, O’Donohue TL. Neuromedin B-like peptides in releasing factor system. J Neurosci 2000; 20: 1142–1156. rat brain: biochemical characterization, mechanism of release and 41 Ma S, Blasiak A, Olucha-Bordonau FE, Verberne AJ, Gundlach AL. Heter- localization in synaptosomes. Peptides 1986; 7: 815–820. ogeneous responses of nucleus incertus neurons to corticotrophin- 29 Smith CM, Ryan PJ, Hosken IT, Ma S, Gundlach AL. Relaxin-3 systems releasing factor and coherent activity with hippocampal theta rhythm in the brain - the first 10 years. J Chem Neuroanat 2011; 42: in the rat. J Physiol 2013; 591: 3981–4001. 262–275. 42 Justice NJ, Yuan ZF, Sawchenko PE, Vale W. Type 1 corticotropin-releasing 30 Lein ES, Hawrylycz MJ, Ao N, Ayres M, Bensinger A, Bernard A, Boe AF, factor receptor expression reported in BAC transgenic mice: implications Boguski MS, Brockway KS, Byrnes EJ, Chen L, Chen TM, Chin MC, for reconciling ligand-receptor mismatch in the central corticotropin- Chong J, Crook BE, Czaplinska A, Dang CN, Datta S, Dee NR, Desaki AL, releasing factor system. J Comp Neurol 2008; 511: 479–496. Desta T, Diep E, Dolbeare TA, Donelan MJ, Dong HW, Dougherty JG, 43 Potter E, Behan DP, Linton EA, Lowry PJ, Sawchenko PE, Vale WW. The Duncan BJ, Ebbert AJ, Eichele G, Estin LK, Faber C, Facer BA, Fields R, central distribution of a corticotropin-releasing factor (CRF)-binding Fischer SR, Fliss TP, Frensley C, Gates SN, Glattfelder KJ, Halverson KR, protein predicts multiple sites and modes of interaction with CRF. Proc Hart MR, Hohmann JG, Howell MP, Jeung DP, Johnson RA, Karr PT, Ka- Natl Acad Sci USA 1992; 89: 4192–4196. wal R, Kidney JM, Knapik RH, Kuan CL, Lake JH, Laramee AR, Larsen 44 Greco MA, Shiromani PJ. Hypocretin receptor protein and mRNA KD, Lau C, Lemon TA, Liang AJ, Liu Y, Luong LT, Michaels J, Morgan JJ, expression in the dorsolateral pons of rats. Brain Res Mol Brain Res Morgan RJ, Mortrud MT, Mosqueda NF, Ng LL, Ng R, Orta GJ, Overly 2001; 88: 176–182. CC, Pak TH, Parry SE, Pathak SD, Pearson OC, Puchalski RB, Riley ZL, 45 Marcus JN, Aschkenasi CJ, Lee CE, Chemelli RM, Saper CB, Yanagisawa Rockett HR, Rowland SA, Royall JJ, Ruiz MJ, Sarno NR, Schaffnit K, M, Elmquist JK. Differential expression of orexin receptors 1 and 2 in Shapovalova NV, Sivisay T, Slaughterbeck CR, Smith SC, Smith KA, the rat brain. J Comp Neurol 2001; 435: 6–25. Smith BI, Sodt AJ, Stewart NN, Stumpf KR, Sunkin SM, Sutram M, Tam 46 Saito Y, Cheng M, Leslie FM, Civelli O. Expression of the melanin-con- A, Teemer CD, Thaller C, Thompson CL, Varnam LR, Visel A, Whitlock centrating hormone (MCH) receptor mRNA in the rat brain. J Comp RM, Wohnoutka PE, Wolkey CK, Wong VY, Wood M, Yaylaoglu MB, Neurol 2001; 435: 26–40. Young RC, Youngstrom BL, Yuan XF, Zhang B, Zwingman TA, Jones AR. 47 Wada E, Way J, Lebacq-Verheyden AM, Battey JF. Neuromedin B and Genome-wide atlas of gene expression in the adult mouse brain. Nat- gastrin-releasing peptide mRNAs are differentially distributed in the rat ure 2007; 445: 168–176. nervous system. J Neurosci 1990; 10: 2917–2930. 31 Paxinos G, Carrive P, Wang H, Wang P-Y. Chemoarchitectonic Atlas of 48 Merali Z, Graitson S, Mackay JC, Kent P. Stress and eating: a dual role the Rat Brainstem. San Diego, CA: Academic Press, 1999. for bombesin-like peptides. Front Neurosci 2013; 7: 193. 32 Bathgate RAD, Samuel CS, Burazin TCD, Layfield S, Claasz AA, Reyto- 49 Boughton CK, Patel SA, Thompson EL, Patterson M, Curtis AE, Amin A, mas IG, Dawson NF, Zhao C, Bond C, Summers RJ, Parry LJ, Wade JD, Chen K, Ghatei MA, Bloom SR, Murphy KG. Neuromedin B stimulates Tregear GW. Human relaxin gene 3 (H3) and the equivalent mouse the hypothalamic-pituitary-gonadal axis in male rats. Regul Pept 2013; relaxin (M3) gene. Novel members of the relaxin peptide family. J Biol 187: 6–11. Chem 2002; 277: 1148–1157. 50 Vaccari C, Lolait SJ, Ostrowski NL. Comparative distribution of vaso- 33 Burazin TCD, Bathgate RAD, Macris M, Layfield S, Gundlach AL, Tregear pressin V1b and oxytocin receptor messenger ribonucleic acids in brain. GW. Restricted, but abundant, expression of the novel rat gene-3 (R3) Endocrinology 1998; 139: 5015–5033. relaxin in the dorsal tegmental region of brain. J Neurochem 2002; 82: 51 Mani BK, Walker AK, Lopez Soto EJ, Raingo J, Lee CE, Perello M, 1553–1557. Andrews ZB, Zigman JM. Neuroanatomical characterization of a growth 34 Tanaka M, Iijima N, Miyamoto Y, Fukusumi S, Itoh Y, Ozawa H, Ibata Y. hormone secretagogue receptor-green fluorescent protein reporter Neurons expressing relaxin 3/INSL 7 in the nucleus incertus respond to mouse. J Comp Neurol 2014; 522: 3644–3666. stress. Eur J Neurosci 2005; 21: 1659–1670. 52 Miyamoto Y, Watanabe Y, Tanaka M. Developmental expression and 35 Liu C, Eriste E, Sutton S, Chen J, Roland B, Kuei C, Farmer N, Jornvall serotonergic regulation of relaxin 3/INSL7 in the nucleus incertus of H, Sillard R, Lovenberg TW. Identification of relaxin-3/INSL7 as an rat brain. Regul Pept 2008; 145: 54–59. endogenous ligand for the orphan G-protein coupled receptor 53 Banerjee A, Shen PJ, Ma S, Bathgate RAD, Gundlach AL. Swim stress GPCR135. J Biol Chem 2003; 278: 50754–50764. excitation of nucleus incertus and rapid induction of relaxin-3 expres- 36 Bathgate RAD, Ivell R, Sanborn BM, Sherwood OD, Summers RJ. Inter- sion via CRF1 activation. Neuropharmacology 2010; 58: 145–155. national Union of Pharmacology LVII: recommendations for the 54 Huang Q, Timofeeva E, Richard D. Regulation of corticotropin-releasing nomenclature of receptors for relaxin family peptides. Pharmacol Rev factor and its types 1 and 2 receptors by leptin in rats subjected to 2006; 58: 7–31. treadmill running-induced stress. J Endocrinol 2006; 191: 179–188. 37 van der Westhuizen ET, Werry TD, Sexton PM, Summers RJ. The relaxin 55 Kirby LG, Freeman-Daniels E, Lemos JC, Nunan JD, Lamy C, Akanwa A, family peptide receptor 3 activates extracellular signal-regulated kinase Beck SG. Corticotropin-releasing factor increases GABA synaptic activ- 1/2 through a protein kinase C-dependent mechanism. Mol Pharmacol ity and induces inward current in 5-hydroxytryptamine dorsal raphe 2007; 71: 1618–1629. neurons. J Neurosci 2008; 28: 12927–12937. 38 Sutton SW, Bonaventure P, Kuei C, Roland B, Chen J, Nepomuceno D, 56 Van Pett K, Viau V, Bittencourt JC, Chan RK, Li HY, Arias C, Prins GS, Lovenberg TW, Liu C. Distribution of G-protein-coupled receptor (GPCR) Perrin M, Vale W, Sawchenko PE. Distribution of mRNAs encoding CRF 135 binding sites and receptor mRNA in the rat brain suggests a role receptors in brain and pituitary of rat and mouse. J Comp Neurol for relaxin-3 in neuroendocrine and sensory processing. Neuroendocri- 2000; 428: 191–212. nology 2004; 80: 298–307. 57 Watanabe Y, Miyamoto Y, Matsuda T, Tanaka M. Relaxin-3/INSL7 regu- 39 Potter E, Sutton S, Donaldson C, Chen R, Perrin M, Lewis K, Sawchenko lates the stress-response system in the rat hypothalamus. J Mol Neu- PE, Vale W. Distribution of corticotropin-releasing factor receptor rosci 2010; 43: 169–174.

58 McGowan BM, Minnion JS, Murphy KG, Roy D, Stanley SA, Dhillo WS, 75 Ryan PJ, Krstew EV, Sarwar M, Gundlach AL, Lawrence AJ. Relaxin-3 Gardiner JV, Ghatei MA, Bloom SR. Relaxin-3 stimulates the neuro- mRNA levels in nucleus incertus correlate with alcohol and sucrose endocrine stress axis via corticotrophin-releasing hormone. J Endocri- intake in rats. Drug Alcohol Depend 2014; 140: 8–16. nol 2014; 221: 337–346. 76 Betley JN, Sternson SM. Adeno-associated viral vectors for mapping, 59 Ribeiro-Barbosa ER, Canteras NS, Cezario AF, Blanchard RJ, Blanchard monitoring, and manipulating neural circuits. Hum Gene Ther 2011; DC. An alternative experimental procedure for studying predator-related 22: 669–677. defensive responses. Neurosci Biobehav Rev 2005; 29: 1255–1263. 77 Urban DJ, Roth BL. DREADDs (Designer Receptors Exclusively Activated 60 Cullinan WE, Herman JP, Battaglia DF, Akil H, Watson SJ. Pattern and by Designer Drugs): chemogenetic tools with therapeutic utility. Annu time course of immediate early gene expression in rat brain following Rev Pharmacol Toxicol 2015; 55: 399–417. acute stress. Neuroscience 1995; 64: 477–505. 78 Ma S. Nucleus incertus and relaxin-3: An emerging modulatory role in 61 Senba E, Matsunaga K, Tohyama M, Noguchi K. Stress-induced c-fos arousal, stress and memory in The 8th International Congress of Neu- expression in the rat brain: activation mechanism of sympathetic path- roendocrinology. Sydney, Australia: 2014: 5. way. Brain Res Bull 1993; 31: 329–344. 79 Buzsaki G, Logothetis N, Singer W. Scaling brain size, keeping timing: 62 Lee LC, Rajkumar R, Dawe GS. Selective lesioning of nucleus incertus evolutionary preservation of brain rhythms. Neuron 2013; 80: 751– with corticotropin releasing factor-saporin conjugate. Brain Res 2014; 764. 1543: 179–190. 80 Vertes RP. Brainstem control of the events of REM sleep. Prog Neuro- 63 Pereira CW, Santos FN, Sanchez-Perez AM, Otero-Garcia M, Marchioro biol 1984; 22: 241–288. M, Ma S, Gundlach AL, Olucha-Bordonau FE. Electrolytic lesion of the 81 Vertes RP. Hippocampal theta rhythm: a tag for short-term memory. nucleus incertus retards extinction of auditory conditioned fear. Behav Hippocampus 2005; 15: 923–935. Brain Res 2013; 247: 201–210. 82 McNaughton N, Gray JA. Anxiolytic action on the behavioural inhibition 64 Quirk GJ, Mueller D. Neural mechanisms of extinction learning and system implies multiple types of arousal contribute to anxiety. J Affect retrieval. Neuropsychopharmacology 2008; 33: 56–72. Disord 2000; 61: 161–176. 65 Lenglos C, Mitra A, Guevremont G, Timofeeva E. Sex differences in the 83 Stujenske JM, Likhtik E, Topiwala MA, Gordon JA. Fear and safety effects of chronic stress and food restriction on body weight gain and engage competing patterns of theta-gamma coupling in the basolater- brain expression of CRF and relaxin-3 in rats. Genes Brain Behav 2013; al amygdala. Neuron 2014; 83: 919–933. 12: 370–387. 84 Simon AP, Poindessous-Jazat F, Dutar P, Epelbaum J, Bassant MH. Fir- 66 Lenglos C, Mitra A, Guevremont G, Timofeeva E. Regulation of expres- ing properties of anatomically identified neurons in the medial septum sion of relaxin-3 and its receptor RXFP3 in the brain of diet-induced of anesthetized and unanesthetized restrained rats. J Neurosci 2006; obese rats. Neuropeptides 2014; 48: 119–132. 26: 9038–9046. 67 McGowan BM, Stanley SA, Smith KL, White NE, Connolly MM, Thomp- 85 Hangya B, Borhegyi Z, Szilagyi N, Freund TF, Varga V. GABAergic neu- son EL, Gardiner JV, Murphy KG, Ghatei MA, Bloom SR. Central relaxin- rons of the medial septum lead the hippocampal network during theta 3 administration causes hyperphagia in male Wistar rats. Endocrinol- activity. J Neurosci 2009; 29: 8094–8102. ogy 2005; 146: 3295–3300. 86 Nunez A, Cervera-Ferri A, Olucha-Bordonau F, Ruiz-Torner A, Teruel V. 68 Shabanpoor F, Akhter Hossain M, Ryan PJ, Belgi A, Layfield S, Kocan Nucleus incertus contribution to hippocampal theta rhythm generation. M, Zhang S, Samuel CS, Gundlach AL, Bathgate RA, Separovic F, Wade Eur J Neurosci 2006; 23: 2731–2738. JD. Minimization of human relaxin-3 leading to high-affinity analogues 87 Vertes RP, Martin GF. Autoradiographic analysis of ascending projec- with increased selectivity for relaxin-family peptide 3 receptor (RXFP3) tions from the pontine and mesencephalic reticular formation and the over RXFP1. J Med Chem 2012; 55: 1671–1681. median raphe nucleus in the rat. J Comp Neurol 1988; 275: 511–541. 69 Haugaard-Kedstrom LM, Shabanpoor F, Hossain MA, Clark RJ, Ryan PJ, 88 Teruel-Marti V, Cervera-Ferri A, Nunez A, Valverde-Navarro AA, Olucha- Craik DJ, Gundlach AL, Wade JD, Bathgate RA, Rosengren KJ. Design, Bordonau FE, Ruiz-Torner A. Anatomical evidence for a ponto-septal synthesis, and characterization of a single-chain peptide antagonist for pathway via the nucleus incertus in the rat. Brain Res 2008; 1218: the relaxin-3 receptor RXFP3. J Am Chem Soc 2011; 133: 4965–4974. 87–96. 70 Kuei C, Sutton S, Bonaventure P, Pudiak C, Shelton J, Zhu J, Nepomuc- 89 Ma S, Olucha-Bordonau FE, Hossain MA, Lin F, Kuei C, Liu C, Wade JD, eno D, Wu J, Chen J, Kamme F, Seierstad M, Hack MD, Bathgate RA, Sutton SW, Nunez A, Gundlach AL. Modulation of hippocampal theta Hossain MA, Wade JD, Atack J, Lovenberg TW, Liu C. R3(BDelta23 27)R/ oscillations and spatial memory by relaxin-3 neurons of the nucleus I5 chimeric peptide, a selective antagonist for GPCR135 and GPCR142 incertus. Learn Mem 2009; 16: 730–742. over relaxin receptor LGR7: in vitro and in vivo characterization. J Biol 90 Guzman YF, Tronson NC, Jovasevic V, Sato K, Guedea AL, Mizukami H, Chem 2007; 282: 25425–25435. Nishimori K, Radulovic J. Fear-enhancing effects of septal oxytocin 71 Ganella DE, Callander GE, Ma S, Bye CR, Gundlach AL, Bathgate RA. receptors. Nat Neurosci 2013; 16: 1185–1187. Modulation of feeding by chronic rAAV expression of a relaxin-3 pep- 91 Mitra A, Lenglos C, Timofeeva E. Inhibition in the lateral septum tide agonist in rat hypothalamus. Gene Ther 2013; 20: 703–716. increases sucrose intake and decreases anorectic effects of stress. Eur 72 Smith CM, Chua BE, Zhang C, Walker AW, Haidar M, Hawkes D, J Neurosci 2014; Nov 24. doi: 10.1111/ejn.12798. [Epub ahead of Shabanpoor F, Hossain MA, Wade JD, Rosengren KJ, Gundlach AL. Cen- print]. tral injection of relaxin-3 receptor (RXFP3) antagonist peptides reduces 92 Cervera-Ferri A, Guerrero-Martinez J, Bataller-Mompean M, Taberner- motivated food seeking and consumption in C57BL/6J mice. Behav Cortes A, Martinez-Ricos J, Ruiz-Torner A, Teruel-Marti V. Theta syn- Brain Res 2014; 268: 117–126. chronization between the hippocampus and the nucleus incertus in 73 Ryan PJ, Kastman HE, Krstew EV, Rosengren KJ, Hossain MA, Churilov L, urethane-anesthetized rats. Exp Brain Res 2011; 211: 177–192. Wade JD, Gundlach AL, Lawrence AJ. Relaxin-3/RXFP3 system regulates 93 Tsukahara S, Kanaya M, Yamanouchi K. Neuroanatomy and sex differ- alcohol-seeking. Proc Natl Acad Sci USA 2013; 110: 20789–20794. ences of the lordosis-inhibiting system in the lateral septum. Front 74 Ehlers CL, Chaplin RI, Wall TL, Lumeng L, Li TK, Owens MJ, Nemeroff Neurosci 2014; 8: 299. CB. Corticotropin releasing factor (CRF): studies in alcohol preferring 94 Chee SS, Menard JL, Dringenberg HC. Behavioral anxiolysis without and non-preferring rats. Psychopharmacology 1992; 106: 359–364. reduction of hippocampal theta frequency after histamine applica-

tion in the lateral septum of rats. Hippocampus 2014; 24:615– 99 Atasoy D, Betley JN, Li WP, Su HH, Sertel SM, Scheffer LK, Simpson JH, 627. Fetter RD, Sternson SM. A genetically specified connectomics approach 95 Allsop SA, Vander Weele CM, Wichmann R, Tye KM. Optogenetic applied to long-range feeding regulatory circuits. Nat Neurosci 2014; insights on the relationship between anxiety-related behaviors and 17: 1830–1839. social deficits. Front Behav Neurosci 2014; 8: 241. 100 Sasaki K, Suzuki M, Mieda M, Tsujino N, Roth B, Sakurai T. Pharmaco- 96 Nakajima M, Gorlich A, Heintz N. Oxytocin modulates female sociosex- genetic modulation of orexin neurons alters sleep/wakefulness states ual behavior through a specific class of prefrontal cortical interneu- in mice. PLoS ONE 2011; 6: e20360. rons. Cell 2014; 159: 295–305. 101 Ting JT, Feng G. Recombineering strategies for developing next genera- 97 Tanaka M. Relaxin-3/insulin-like peptide 7, a neuropeptide involved in tion BAC transgenic tools for optogenetics and beyond. Front Behav the stress response and food intake. FEBS J 2010; 277: 4990–4997. Neurosci 2014; 8: 111. 98 Smith CM, Walker AW, Hosken IT, Chua BE, Zhang C, Haidar M, Gund- 102 Steinberg EE, Christoffel DJ, Deisseroth K, Malenka RC. Illuminating lach AL. Relaxin-3/RXFP3 networks: an emerging target for the treat- circuitry relevant to psychiatric disorders with optogenetics. Curr Opin ment of depression and other neuropsychiatric diseases? Front Neurobiol 2014; 30: 9–16. Pharmacol 2014; 5: 46.