Anatomical evidence for a ponto-septal pathway via the nucleus incertus in the rat

BR A IN RE S E A RCH 1 2 18 ( 20 0 8 ) 8 7 –9 6

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s

Research Report

Anatomical evidence for a ponto-septal pathway via the nucleus incertus in the rat☆ Vicent Teruel-Martía,⁎, Ana Cervera-Ferria , Angel Nuñezb , Alfonso Amador Valverde-Navarroa , Francisco Eliseo Olucha-Bordonaua , Amparo Ruiz-Tornera a

Department Anatomia y Embriología Humana, Facultad de Medicina, Universidad de Valencia, Avd. Blasco Ibañez, 15, 46010 Valencia, Spain Departamento de Anatomia, Histología y Neurociencia, Facultad de Medicina, Universidad Autónoma de Madrid, C/Arzobispo Morcillo, 4, 28029, Madrid, Spain

b

A R T I C LE I N FO

AB S T R A C T

Article history:

Hippocampal theta activity is involved in sensory–motor integration and constitutes a

Accepted 4 April 2008

functional basis for mnemonic functions. The medial septum–diagonal band of Broca (MS/

Available online 22 April 2008

DBv) is a key structure as pacemaker of the oscillation. In addition, some brainstem reticular structures are crucial for the activation of MS/DBv. Specifically, the nucleus reticularis pontis

Keywords:

oralis (RPO) is considered the most effective pontine site for eliciting theta rhythm. Nevertheless,

Hippocampus

its connection with the MS/DBv is not direct. A previous study by our group pointed out that the

Tegmentum

nucleus incertus (NI) could be considered as a relay in this multisynaptic pathway. From this

Brainstem

study, the stimulation of RPO increased the discharge rate of NI neurons in anesthetized rats and

Medial septum

the lesion of the NI suppressed the RPO-elicited hippocampal theta. Those findings suggested a

Diagonal band of Broca

projection from RPO to NI, although the existing literature did not support this hypothesis. In

Reticularis pontis oralis

order to clarify the dichotomy between the anatomical and the electrophysiological data, we performed a set of tracing studies. Anterograde tracer injections into RPO showed a profuse projection to NI. This connection was confirmed by retrograde tracer injections into NI. Injections of retrograde tracer in MS/DBv confirmed the intense NI-MS/DBv projection. Furthermore, simultaneous injections of anterograde and retrograde tracers into RPO and MS/ DBv respectively resulted in a high-correlated pattern of terminal-like fibers over labeled somata in the NI. This study provides the first anatomical evidence of a ponto-septal pathway via the NI that contributes to generation and modulation the hippocampal theta activity. © 2008 Elsevier B.V. All rights reserved.

☆ This study was supported by the Fondo de Investigaciones Sanitarias (FIS; ISCII) Research Program, PI061816 and FPU grant AP-20044126 (MEC). ⁎ Corresponding author. E-mail address: [email protected] (V. Teruel-Martí). Abbreviations: 4v, Fourth ventricle; ac, Anterior commissure; Aq, Aqueduct; Ba, Barrington nucleus; BDA, Fluorescein-labeled dextran biotin amine; CGB, Central gray pars beta; DBv, Diagonal band of Broca, vertical limb; DMT, Dorsomedial tegmental nucleus; DR, Dorsal raphe nucleus; dscp, Decussation of the superior cerebellar peduncle; DT, Dorsal tegmental nucleus; FG, Fluorogold; IP, Interpeduncular nucleus; LC, Locus coeruleus; LDT, Laterodorsal tegmental nucleus; LHA, Lateral hypothalamic area; LHb, Lateral habenula; LV, Lateral ventricle; mlf, Medial longitudinal fascicle; MR, Median raphe nucleus; MS, Medial septum; MS/DBv, Medial septum–diagonal band of Broca; NI, Nucleus incertus; PDT, Posterodorsal tegmental nucleus; PH, Posterior hypothalamic nucleus; PL, Paralemniscal nucleus; PMR, Paramedian raphe nucleus; PPT, Pedunculopontine tegmental nucleus; RPC, Nucleus reticularis pontis caudalis; RPn, Raphe pontis; RPO, Nucleus reticularis pontis oralis; rs, Rubrospinal tract; RtTg, Reticular tegmental nucleus; Sph, Sphenoidal nucleus; SUM, Supramammillary nucleus; VTg, Ventral tegmental nucleus; ZI, Zona incerta

0006-8993/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2008.04.022

88

1.

BR A IN RE S EA RCH 1 2 18 ( 20 0 8 ) 8 7 –96

Introduction

Theta oscillation is a periodic field potential recordable in hippocampal areas and related to its main functions (Hasselmo, 2005 for review). This synchronous activity is thought to participate in timing, encoding, consolidation and retrieval of memories (Axmacher et al., 2006; Buzsáki, 2002; Sakata, 2006). It is well established that the medial septum–diagonal band of Broca (vertical limb) complex (MS/DBv) is a key component of the hippocampal system. The rhythmic activity of MS/DBv neurons encodes hippocampal theta rhythm as a “pacemaker” input on the hippocampal formation (Petsche and Stumpf, 1962). The MS/DBv is under the influence of ascending brainstem systems. The reticular pontine region is the origin of an asõcending pathway to the caudal diencephalic region including posterior hypothalamic (PH) and supramammillary (SUM) nuclei, which in turn send projections to the MS/DBv and hippocampus. The medial septum region acts then as a node in these ascending multisynaptic pathways, sending inputs to the hippocampal formation (Bland, 2000; for review). Several studies indicate that the reticularis pontis oralis nucleus (RPO) is centrally involved in the activation hippocampal synchronization (Bland et al., 1994; Brazhnik et al., 1985; Nuñez et al., 1991; Vertes, 1981). Also in the pontine

region, the median raphe nucleus (MR) is critical to the switching processes of activation/deactivation of theta rhythm (Kitchigina et al., 1999; Vertes and Kocsis, 1997) via its projections to MS/DBv, SUM and hippocampus (Acsady et al., 1996; Vertes et al., 1999). Previous studies have described the nucleus incertus (NI), in the dorsal tegmentum, as the origin of an ascending brainstem network mainly to basal forebrain areas (Goto et al., 2001; Olucha-Bordonau et al., 2003). Since anterograde tracing experiments have described MS/DBv, SUM, PH and MR as the main targets of the NI, a putative influence over theta synchronization was suggested (Olucha-Bordonau et al., 2003). In accordance with this possibility, our recent experiments in the urethane-anesthetized rat have shown that the NI is able to modify the hippocampal oscillation (Nunez et al., 2006). The study proved that electrical stimulation of NI induces an increase in theta rhythmicity in the hippocampus. Moreover, electrical lesions or muscimol infusion suppressed the hippocampal theta oscillation evoked by RPO stimulation. These results suggested that RPO and NI could be components of a common brainstem network, and perhaps the origin of the ascending pathway to the hippocampal “pacemaker” structures. However, the current hodological data does not give support to this hypothesis since no anatomical evidence has been found for a connection between the RPO and the NI. Only

Fig. 1 – Schematic drawings showing the most representative locations of the (A) fluorogold injections in the MS/DBv, (B) BDA or miniruby in RPO and (C) fluorogold in NI. Note that only injections with a minimal diffusion to adjacent areas were described. In each case the antero-posterior coordinates of the represented injections are indicated with respect to bregma. Fluorescence photomicrographs in (A′) MS/Dbv and (B′) RPO representative cases and (C′) bright-field photomicrographs of fluorogold injection stained with the DAB method. Scale bars, A′: 200 μm; B′ and C′: 500 μm. For abbreviations, see list.

BR A IN RE S E A RCH 1 2 18 ( 20 0 8 ) 8 7 –9 6

scarce ascending fibers of passage have been described in RPO perhaps originating in NI (Olucha-Bordonau et al., 2003) and no RPO labeled cells were found in the retrograde tracing study made by Goto et al. (2001). Although a contribution of the NI to theta-controlling circuitry has been previously suggested (Nuñez et al., 2006), the anatomical basis of the electrophysiological data is lacking. In the present study, we performed a set of tract tracing experiments to elucidate the connective substrate of the pontoseptal pathway.

2.

Results

Since the goal of the present study was to clarify the apparent contradiction between our previous electrophysiological results and the anatomical circuitry described in the literature, we focused on verifying a pathway from RPO to MS/DBv via NI. To achieve this goal, we first intended to reproduce the retrograde tracer injections in NI (n = 9) made by Goto et al. (2001) and compared the results, particularly at reticular levels, with anterograde tracer injections in RPO (n = 4). In addition, we performed retrograde tracer injections in MS/DBv (n = 6) to delimit the origin of the brainstem afferents to the medial septum complex. In order to visualize the multisynaptic pathways from RPO to MS/DBv, simultaneous injections of retrograde tracer in MS/DBv and anterograde tracer in RPO were made (n = 5). This description is a summary of the most prominent results and novel findings which help to clarify important pathways from RPO to MS/DBv. Four of the 24 cases have been described in detail, attending to the experimental divisions of this work. Moreover, the spillage of marker along the penetration track was virtually negligible in all the cases considered in the present description.

2.1.

89

were found and in most cases they seemed to cross the nuclei towards the dorsal periphery of the tegmentum. Along with the anterograde labeled fibers revealed by BDA or miniruby, the low molecular weight of miniruby also allowed some retrograde tracing. This allowed us to further verify the existence of a projection from NI to RPO by visualization of some retrograde labeled neurons in NI after miniruby injection in RPO. At the rostral levels of the NI, the laterodorsal tegmental nucleus (LDT) and dorsomedial tegmental nucleus (DMT) showed a less intense but significant density of fibers. Moderate terminal labeling was observed mainly in the ipsilateral ventral tegmental nucleus and its capsule and in the dorsal raphe nucleus (DR). The lateral and ventrolateral columns of the periaqueductal gray also showed a persistent pattern of scarce labeled axons. At the same levels, anterograde and scattered retrograde labeling was observed in the pedunculopontine nucleus (PPT) with extensions of terminal labeling in the cuneiform complex.

Miniruby and BDA injections in RPO

The single anterograde study was based on 4 cases with tracer injections mainly centered in the middle rostro-caudal extension of RPO (Fig. 1). In three of these cases the injection site was restricted to RPO without diffusion to neighboring areas. In all rostro-caudal levels studied a prominent pattern of anterogradely labeled fibers and terminal-like fibers was observed. Whereas a large number of labeled fibers crossed immediately to the contralateral RPO, a significant quantity of axons remained in the ipsilateral side, including the pars caudalis of the reticular pontine nucleus (RPC) and traveled through the reticular structure toward medullary areas. Although scattered fiber labeling was found entering in the medial longitudinal fascicle (mlf), bundles of axons appeared to cross the mlf to terminate in the dorsal tegmental area. Moreover, this group of axons was the most prominent labeling observed in pontine areas. Notably, caudally to RPO the NI displayed the most significant intensity of labeled fibers with evidence of terminal-like labeling (for a panoramic view see Fig. 2). Strongly labeled axons were accumulated in both subdivisions of the NI (pars compacta and pars dissipata) surrounding but not entering the dorsal and posterodorsal tegmental nucleus (DT and PDT, respectively). In these nuclei, little or no fibers

Fig. 2 – Fluorescence photomicrographs of the labeling at the NI level in the (A) miniruby injection in RPO and (B) FG injection in the MS/DBv. Intense anterograde labeling was found in both compacta and dissipata subdivisions of the NI, surrounding the DT. Note the absence of labeling in the DT. Additional fluorescence can be appreciated due to retrograde labeling of the miniruby in the NId (asterisk) or glial response (arrow). The most intense retrograde labeling in the pontine tegmentum after FG injections in MS/DBv was located in NI, LDT, LC and Ba. Scale bars, A: 100 μm; B: 400 μm. For abbreviations, see list.

90 2.2.

BR A IN RE S EA RCH 1 2 18 ( 20 0 8 ) 8 7 –96

Fluorogold injections in NI

In order to assess the projection from RPO to NI not yet described but suggested by the activation of NI neurons after RPO stimulation, we obtained 9 cases (n = 9) with retrograde tracer injections in the dorsomedial tegmental area mainly centered in the NI. The selected coordinates attempted to avoid the possible diffusion of tracer into the DR and DT. The passage of the mlf usually limited the ventral diffusion of the tracer to the nearby raphe pontis (RPn), DMT or reticular tegmental nucleus (RtTg). In five of these cases the tracer site was limited to the NI with lesser diffusion into the adjacent areas. Injections located in each of the neighboring structures served as controls. Overall, the afferent projections of NI neurons corresponded to those previously described (Goto et al., 2001) except for the findings at reticular levels. The main sources of afferences were localized in MR and paramedian raphe nucleus (PMR), interpeduncular nucleus (IP), and lateral habenula (LHb) (data not shown).

The systematic study of pontine levels (Fig. 3) revealed a consistent projection from reticular pontine nuclei, including RPO. Labeled neurons were found in the rostro-caudal and dorso-ventral, extent of the RPO, preferentially at the ipsilateral part. The scattered distribution of RPO neurons meant that the visualization of all labeled cells in each section was only possible by a detailed survey the entire thickness of each section. At the same levels, labeled neurons were also present, in PPT and surrounding the mlf and rubrospinal tract (rs). A significant number of labeled neurons also appeared in RtTg, the ventral reticular adjacent to RPO. Consistent with previous studies, retrograde filled neurons were also found in the MS/DBv, mostly in the diagonal band area, and in the SUM, PH, lateral hypothalamic area (LHA), zona incerta (ZI) and surrounding the mamillary tract. In the tegmentum labeling was also seen in DR, which cannot be excluded as being just the result of diffusion of the tracer injection.

Fig. 3 – (A–B) Schematic drawings showing the retrograde labeled neurons (black dots) at rostral pontine levels after fluorogold injections in NI. The diagrams correspond to the detailed study at 40× objective. The injection of the retrograde tracer in NI resulted in the labeling of neurons located into the RPO. As represented, labeled somata were also found at these levels in MR, PMR, PPT, RtTg, PL nuclei and surrounding the mlf and rs. (C) Photomicrograph showing sparse labeled neurons in RPO. (E) Photomicrograph detail of retrogradely labeled neurons in RPO.

BR A IN RE S E A RCH 1 2 18 ( 20 0 8 ) 8 7 –9 6

2.3.

Fluorogold injections in MS/DBv

The distribution of neurons that project to the MS/DBv from NI was examined using Fluorogold as retrograde tracer. To confirm the boundaries of this projection, six animals received tracer injections confined to the medial septal area, with deposits centered mainly in the nucleus of the medial septum (MS; n = 2) with diffusion to the diagonal band of Broca, vertical

91

limb (DBv; n = 3) or centered in the DBv (n = 1). The medial septal region was well suited to the use of small and localized iontophoretic injections of tracers. In all cases, a very similar pattern of retrograde labeled cells was observed (Fig. 2). A dense retrograde cell labeling was found in both subdivisions of the NI (NIc and NId). The neighboring areas, including locus coeruleus (LC), LDT and the Barrington nucleus (Ba), presented a sparser cell labeling. A significant number of labeled cells

Fig. 4 – Labeling in the NI after simultaneous fluorogold and miniruby injections in MS/DBv and RPO, respectively at rostral (A) and caudal (D) levels of the NI. Note the high correlation of the anterograde and retrograde labeling in the NI and the relative absence of labeling in DT. Detailed photomicrographs (100×) of NIc (B–C) and NId (E–F). Scale bars: 200 μm (A), 100 μm (D), 10 μm (B–C, E–F).

92

BR A IN RE S EA RCH 1 2 18 ( 20 0 8 ) 8 7 –96

was observed around the mlf and invading the immediate ventral zone. Notably, only the ventral part at the most caudal area of DR showed numerous labeled cells, a zone previously identified as the rostral pole of the NIc (Olucha-Bordonau et al., 2003) and clearly continuous with more caudal labeling in our material. In addition, labeled somata were not seen in the reticular pontine nuclei, confirming no direct connections between RPO and medial septal area. Of particular importance, consistent retrograde labeling (data not shown) was found in SUM, PH, LHb and MR, with a pattern of projections quite comparable to that previously described in the literature (see Discussion).

2.4.

Double labeling experiments

Double labeling experiments were undertaken in an attempt to obtain more accurate identification and description of the complete pathway from RPO to the medial septal area via the NI. In this group of experiments a total of five (n = 5) rats received double-tracer injections with an anterograde and retrograde marker in RPO and MS/DBv, respectively. These injections supported the results of the single tracer experiments. Moreover, it was possible to obtain a very high correlation between the location of the fibers from RPO and the somata projecting to MS/DBv at NI levels (Fig. 4). A detailed study at 100× optical magnification allowed the visualization of terminal-like fibers originated in RPO colocalizing with retrograde labeled NI neurons (Fig. 4). Two types of putative axo-somatic or proximal axodendritic contacts could be seen in the labeled neurons of the NI: terminal-like labeling between anterograde labeled boutons and retrogradely filled somata; fibers en passant with several boutons in the same labeled fiber over the labeled somata. Double labeling was also found in SUM and PH. Although the pattern of terminal-like fibers was similar, the projection from RPO was less intense in these areas than at the levels of the dorsal tegmentum.

3.

SUM and MR also pointed to the NI as a part of a brainstem influence over the control of theta rhythm (Olucha-Bordonau et al., 2003). Numerous early studies report that theta oscillation can be evoked by electrical or chemical stimulation of the brainstem reticular formation (Klemm, 1972; Macadar et al., 1974; Nuñez et al., 1991; Robinson and Vanderwolf, 1978). More importantly, at the pontine tegmentum, the ventral region of the RPO has been described as the most effective brainstem site for eliciting the hippocampal theta rhythmicity (Nuñez et al., 1991; Vertes, 1981). In the urethane-anesthetized rat, electrophysiological recordings corroborated the involvement of the NI in the synchronization of theta rhythm in the hippocampus (Nunez et al., 2006). From this study, electrical stimulation of NI evokes hippocampal theta activity. Furthermore, electrical stimulation of RPO increases the firing rate of NI neurons with orthodromic activation and electrical lesions or muscimol infusion in NI suppressed the theta rhythm induced by RPO stimulation. Therefore, NI seems to be able to generate theta activity in the hippocampus and RPO-elicited theta depends on its integrity. The efferent projections of the NI to the MS/ DBv could be critical in this effect. However, the mentioned functional findings do not allow us to distinguish between either monosynaptic or multisynaptic pathway in the RPO-NI projection. The present study provides a connective substrate that could explain the effect of RPO over NI neurons and justify the integration of the NI into the brainstem circuit underlying the control of theta oscillation (see Fig. 5).

Discussion

The present results demonstrate the existence of a direct projection from RPO to NI, which is consistent with a complementary pathway from NI to MS/DBv, and in accordance with recent electrophysiological data (Nuñez et al., 2006), indicates that this neuronal circuit is involved in the brainstem activation of hippocampal theta rhythm. The NI has been described as a group of neurons located in the dorsal tegmental area just caudal to the DR and surrounding the DT (Goto et al., 2001). NI has been elsewhere referred either as part of the central gray (Paxinos and Watson, 1986) or as nucleus O (Paxinos, 1985). In the present study, we have followed the NI nomenclature used by the most recent descriptions (Goto et al., 2001; Olucha-Bordonau et al., 2003). According to these works, the main efferents of the NI include the MS/DBv, SUM, PH and MR, while the more dense described afferents are MR, the IP and LHb. The present results confirmed these connections. The existence of this complex network suggested that NI could modulate the level of behavioral activation (Goto et al., 2001). The connections with MS/DBv,

Fig. 5 – Schematic diagram showing the proposed integration of the NI in the classic ponto-septal circuitry (arrow lines) involved in the control of the activation of the hippocampal oscillation in theta states. Note the represented direct connections (neuron symbology) from RPO to neurons in NI which project to the septal areas; a putative pathway complementary to the RPO-SUM-MS/DBv circuit, which could explain the previous electrophysiological results. Moreover, the NI would be integrated inside the network including MR and SUM over hippocampal activity. These projections could be also influenced by RPO input.

BR A IN RE S E A RCH 1 2 18 ( 20 0 8 ) 8 7 –9 6

The injection of the anterograde tracer into RPO revealed an intense and direct projection from RPO to NI. We have observed that RPO neurons, mainly located in the ventral region, provide a substantial input to both divisions of the NI in all its rostro-caudal extension. Moreover, the current data showed another dorsal tegmental nucleus near the NI receiving significant inputs from RPO. We have been able to confirm a projection from RPO to the Ba and caudal half of the LDT, also referred to as retrolaterodorsal tegmental nucleus (Goto et al., 2001). To confirm this connectivity, injections of the retrograde tracer FG were performed into NI, in an attempt to reproduce the experiments made by Goto et al. (2001), which ruled out a projection from RPO. Restricted injections in the NI resulted in the labeling of scarce somata in all the extension of the RPO, although more consistently at the ventral region of the nucleus. These reticular locations agree to the previously described as the most effective theta evoking sites at RPO (Nuñez et al., 1991; Vertes, 1981). The scattering of the labeled neurons in this reticular area made their detection and visualization difficult and they only were noticeable by examining carefully all the thickness of the sections. This might also be the cause for the gap in the existing literature about a projection from RPO to NI. The described results confirm the existence of a direct projection from RPO to NI. These data support that the NI could suppose an important relay in the brainstem-septal pathway, in accordance with the electrophysiological study. In support of these data, the simultaneous double tract tracing experiments carried out in the present study showed a profuse pattern of terminal-like fibers from RPO over NI neurons projecting to the MS/DBv. These results are not direct evidence about the detailed nature of this type of synaptic contacts and can only provide a primary description. These findings suggest the NI as a critical nucleus in the RPO-MS/DBv pathway. The MS/DBv is a crucial site to pacemaking theta rhythmicity in the field potential recorded in the hippocampal formation (Stewart and Fox, 1984a,b; Barrenechea et al., 1995). However, there is no evidence for a direct projection from RPO to the MS/DBv complex and, so, it has been proposed that there exists a multisynaptic pathway (Kirk et al., 1996) through the caudal diencephalic nuclei (SUM, PH). Based on early results (Martin et al., 1985), the SUM was nominated as the “best candidate” for the “intervening cell group” between the pontine reticular formation and MS/DBv (Vertes, 1988). Supporting this proposed circuit, several studies have been able to demonstrate that SUM represents an important relay between the RPO and MS/DBv in the elicitation of theta (Borhegyi et al., 1998; Haglund et al., 1984; Kiss et al., 2000; Leranth and Kiss, 1996; Maglóczky et al., 1994; Vertes and Martin, 1988; Vertes, 1988; Vertes and McKenna, 2000). According to our results, the anterograde tracer injections into RPO revealed fibers also at hypothalamic levels, in SUM and PH. Neurons labeled with FG and thus projecting to MS/ DBv with terminal-like fibers from the RPO were also observed in SUM and PH (data not shown), and confirm previously described pathways to the ultimate pacemaker MS/DBv. These projections were less dense than those revealed in the dorsal tegmentum, and could provide new evidence of the projection between RPO and caudal diencephalon.

93

An anterograde projection has been previously described between NI and SUM-PH nuclei (Olucha-Bordonau et al., 2003). The experiments of NI lesion (described above) suggested that the RPO-caudal diencephalon pathway is ineffective in theta generation through RPO stimulation without the involvement of NI. This is consistent with the idea of a robust connection from RPO to NI, as documented by our findings that it could complement the RPO-MS/DBv projection through SUM. Thus, the brainstem origin of the ascending pathway should include NI and RPO. Finally, it is necessary to highlight that the NI presents a complex pattern of direct and strong connections over “desynchronizing” brainstem components of the hippocampal theta modulatory system (Vertes, 2005), i.e. the inhibitory MR (Goto et al., 2001; Olucha-Bordonau et al., 2003). This connection might imply that the extinction of theta activity induced by RPO stimulation after NI lesions could be due not only to the projection to MS/DBv, but to a coordinated effect over these structures. The neurochemical nature of the connections of the NI neurons is not well known. Previous studies have described that both NI and RPO are included in the brainstem cholinergic system. In both nuclei, chemical stimulation of its neurons with cholinergic agonists (carbachol), in order to elicit hippocampal theta oscillations, has been able to demonstrate its cholinoceptive nature (Nuñez et al., 1991, 2006). Neuropeptides, such as neurotensin, ranatensin and cholecystokinin has been described as markers in NI neurons (Jennes et al., 1982; Chronwall et al., 1985; Olucha-Bordonau et al., 2003). Recently, some data have shown that NI neurons specifically express the peptide relaxin-3 (RLX3; Bathgate et al 2002; Burazin et al 2002), colocalizing with GABAergic somata (Tanaka et al., 2005; Ma et al., 2007). RLX3-immunoreactive fibers have also been shown in MS/DB and SUM (Ma et al., 2007). Since RLX3 activates molecular cascades linked to ERK phosphorylation (van der Westhuizen et al., 2007) and these molecular pathways have been related to mediate LTP processes (Lynch, 2004), RLX3 could be involved in LTP. Whether this peptide does have any effect in theta activity and memory consolidation remains to be elucidated. Further studies about the neurotransmission of the NI neurons are also necessary. In conclusion, we have performed anatomical tracer experiments to clarify and to better define the brainstem system involved in hippocampal theta modulation. This work helps to reconcile the apparent dichotomy between the recent electrophysiological and anatomical data. In the scenario where the RPO seems to be a good candidate to explain the origin of the brainstem generation of the hippocampal theta rhythmicity, the recent findings reveal new pontine sites that could participate in the reticular formation pathway to the basal forebrain involved in the generation of theta activity. The classical definition of a multisynaptic pathway between RPO and MS/DBv through SUM and PH might be extended by considering the presence of the NI, according to our main finding showing the presence of a RPO-NI-MS/DBv pathway (Fig. 5). Regarding the connectivity of the NI, including as main afferences the MR, SUM and MS/DBv, and its relevance over the theta generation, our purpose is that NI can be a key brainstem structure in the influence over hippocampal

94

BR A IN RE S EA RCH 1 2 18 ( 20 0 8 ) 8 7 –96

synchronization. Thus, an accumulating body of evidence suggests that the ponto-septal pathway via the NI could play a meaningful role in the brainstem generation of the hippocampal theta rhythm. Further research is necessary to detail its nature and functional significance.

4.

Experimental procedures

4.1.

Subjects

Twenty-four adult Sprague–Dawley rats (Blackthorn, Bicester, England) of both sexes, weighting 200–350 g, were used in the experiments. All surgical procedures and animal care were performed according to the animal care guidelines of the European Council (86/609/EEC) and approved by the Ethics Committee of the University of Valencia prior to the onset of the experiments. All efforts were made to minimize the number of animals used and their suffering. Rats were housed at a thermoregulated environment in a 12 h light/dark cycle, with ad libitum access to food and water.

4.2.

Surgery

Animals were anesthetized with a mixture of Ketamine 50– 60 mg/kg i.p. (Imalgen, 0.05 g/ml; Rhône Mérieux, Lyon, France) and Xylazine 10 mg/kg i.p. (Xilagesic, 20 mg/ml; Lab. Calier, Barcelona, Spain). Anesthetized rats were placed onto a stereotaxic frame (David Kopf Instruments, Tujunga, USA) and their body temperature was kept at 37 °C with an isothermal heating pad. Trephine holes were drilled in the skull based on stereotaxic coordinates according to the atlas of Paxinos and Watson (1986). Due to the small size of the NI, the interindividual variability in its localization was high. A correction in the coordinates was made to increase the accuracy, using the ratio between the experimental and the theoretical distance from bregma to interaural. However, the coordinates specified below correspond to the atlas references without correction.

4.3.

Single and dual fluorescent tracer microinjection

Tracers were iontophoretically delivered through glass micropipettes with an inner tip diameter of 20–25 μm, made in a pipette puller from Narishige (Japan). The pipette was retracted 100–200 μm after lowing to desired depth, before injecting the tracer to create a reservoir region. Reflux was avoided by performing the removal of the pipette 5 min after finishing injection and using a negative current during the extraction. Rats were then sutured and allowed to recover. Rats were divided into 4 groups depending of the tracer received. The first group had only microinjection of anterograde tracer into RPO; the second group received retrograde tracer injection into NI; the third group had microinjection of retrograde tracer into MS/DBv; and the fourth group received unilateral injections of anterograde tracer and retrograde tracer into RPO and MS/DBv, respectively (see Fig. 1). The anterograde tracers used were either fluorescein-labeled dextran biotin amine (BDA; Invitrogen-Molecular Probes, Eugene, Oregon, USA) or dextran tetramethylrhodamine and biotin (miniruby; Invitrogen-Molecular Probes), 10% diluted in

0.05 M Tris buffer saline (TBS), pH 7.6, and injected by iontophoresis by a current of + 5 μA (7 s on/7 s off, 6–9 min) at RPO levels (7.7–8.3 mm posterior to bregma, 1.2–1.5 mm lateral, 7.4–7.8 mm ventral). In retrograde tracing experiments we used hydroxystilbamidine methanesulfonate (Fluorogold, FG; Invitrogen-Molecular Probes) as fluorescent retrograde tracer, 5% diluted in acetate buffer, pH 4. The current used was +3 μA, (2 s on/2 s. off). The duration of the injection was 5 min into NI (9.5– 9.8 mm posterior to bregma, 0–0.1 mm lateral, 7.3–7.8 mm ventral) and 8–10 min into MS/DBv (0.5–1 mm anterior to bregma, 0–0.2 mm lateral, 7.5–7.6 mm ventral). The optimal tracer diffusion was guaranteed by a previous set of trials changing the conditions of time and current.

4.4.

Fixation and tissue preparation

After 7 to 9 days of survival, the animals were anesthetized with a lethal dose of sodium pentobarbital (100 mg/kg, 20%; Dolethal, Vetoquinol, Madrid, Spain) and transcardially perfused with 500 ml of 0.1% heparinized saline (0.9%, pH 7) followed by 500 ml of fixative composed of 4% paraformaldehyde (Sigma-Aldrich, St. Louis, MO), 0.2% glutaraldehyde (Sigma-Aldrich) and 15% saturated picric acid (Sigma-Aldrich) in 0.1 M phosphate buffer (PB), pH 7.4. The brain was postfixed in the head for 1 h and then removed from the skull and stored in 0.1 M PB (pH 7.2) containing 0.05% sodium azide (SigmaAldrich). Coronal sections were obtained with a vibratome (Leica VT-1000 M, (Leica Microsystems, Cambridge, UK) and collected in the same solution before processing.

4.5.

Immunohistochemical labeling

The brain sections with single FG, BDA or miniruby injection were processed adapting the tract tracing method of Veenman et al. (Veenman et al., 1992). Briefly, endogenous peroxidase was blocked by incubating sections for 30 min at room temperature (RT) with 0.3% H2O2. After rinsing several times, tissues were incubated in avidin–biotin complex (Vectastain ABC Elite kit; Vector Laboratories, Burlingame, CA) for 90 min at room temperature (RT). Then the sections were washed several times in PB 0.1 M and revealed in Tris/HCl at pH 8 with 0.015% of the chromagen 3,3′-diaminobenzidine tetrahidroclorure (DAB) enhanced with NiNH4SO4 and activated with 0.04% H2O2 or in Tris–HCl at pH 7.6 with 0.05% DAB without NiNH4SO4 enhancement. In double-tracer injection experiments, the protocol used was similar. The tissue was successively processed first for BDA or miniruby cytochemistry and then for FG immunochemistry. The staining of the anterograde tracers was carried out as described above, but in this case the buffer was Tris–HCl at pH 8 saline with 0.1% Triton X100 (TBS-Tx). The sections were incubated in ABC complex for 90 min and then successively washed in TBS-Tx and in Tris–HCl at pH 8. BDA or miniruby were revealed in Tris–HCl at pH 8 enhanced with NiNH4SO4 and DAB (0.015%). The reaction was stopped by two rinses in Tris–HCl at pH 8 and one in PB 0.1 M and then collected in PB 0.1 M containing 0.05% sodium azide. In the FG immunohistochemistry, after several rinses, the tissue was preincubated in 3% normal goat serum (NGS) 0.1% TBS-Tx at pH 8 for 1 h and then incubated in the same medium containing antibody anti-FG (24 h, RT) made in rabbit (1:10 000,

BR A IN RE S E A RCH 1 2 18 ( 20 0 8 ) 8 7 –9 6

Sigma-Aldrich). After several washes, the tissue was incubated in biotinylated goat immunoglobulin G anti-rabbit (1:200; Vector Laboratories) with 3% NGS. Finally, the sections were treated with ABC complex (90 min, RT) and revealed in Tris–HCl at pH 7.6 containing 0.05% DAB without NiNH4SO4 enhancement and activated with H2O2 0.04%. The detention and collection method was carried out as described above.

4.6.

Microscopy, image acquisition and data analysis

Verification of the injections sites was performed with light microscope Nikon Eclipse E600 (Japan), equipped with the appropriate fluorescence filters: UV-2A (EX 330–380, DM 400, BA 420) for the FG, B-2A (EX 450–490, DM 505, BA 520) for the BDA and G-2A (EX 510–560, DM 575, BA 590) for the miniruby. Sections were mounted onto gelatin-coated slides, air dried and covered with glycerol for its visualization and analysis. Revealed sections with DAB were mounted as described, dehydrated in alcohol, cleared in xylene, coverslipped with DPX mountant for histology and viewed using standard light microscopy (Nikon Eclipse E600). Images were acquired using Nikon DMX-2000 camera and sent to a PC with ACT-1 acquisition software (Nikon). The images were exported to Adobe Photoshop CS2, which was used to compensate for levels, brightness and contrast and to convert to grayscale. In the cases of retrograde tracer injections into NI, when studying the reticular levels, somata distribution was analyzed in detail at 40× throughout all the thickness of the sections and exporting the results to 4× and 10 photomicrograph images in Adobe Photoshop CS2.

Acknowledgments We thank R. Muñoz Izquierdo for her technical support and both J. Cervera and H. Salvo for their help with the nuances of the language.

REFERENCES

Acsady, L., Arabadzisz, D., Katona, I., Freund, T.F., 1996. Topographic distribution of dorsal and median raphe neurons with hippocampal, septal and dual projection. Acta Biol. Hung. 47, 9–19. Axmacher, N., Mormann, F., Fernández, G., Elger, C.E., Fell, J., 2006. Memory formation by neuronal synchronization. Brain Res. Rev. 52 (1), 170–182. Bathgate, R.A.D., Samuel, C.S., Burazin, T.C.D., Layfeld, S., Claasz, A.A., Reytomas, I.G., Dawson, N.F., Zhao, C., Bond, C., Summers, R.J., Parry, L.J., Wade, J.D., Tregear, G.W., 2002. Human relaxin gene 3 (H3) and the equivalent mouse relaxin (M3) gene. Novel members of the relaxin peptide family. J. Biol. Chem. 277, 1148–1157. Barrenechea, C., Pedemonte, M., Nuñez, A., García-Austt, E., 1995. In vivo intracellular recordings of medial septal and diagonal band of Broca neurons: relationships with theta rhythm. Exp. Brain Res. 103, 31–40. Bland, B.H., 2000. The medial septum: node of the ascending brainstem hippocampal synchronizing pathways. In: Numan, Robert (Ed.), The Behavioral Neuroscience of the Septal Region. Springer-Verlag, NY, pp. 115–145.

95

Bland, B.H., Oddie, S.D., Colom, L.V., Vertes, R.P., 1994. Extrinsic modulation of medial septal cell discharges by the ascending brainstem hippocampal synchronizing pathway. Hippocampus 4, 649–660. Borhegyi, Z., Maglóczky, Z., Acsády, L., Freund, T.F., 1998. The supramammillary nucleus innervates cholinergic and GABAergic neurons in the medial septum–diagonal band of Broca complex. Neuroscience 82, 1053–1065. Brazhnik, E.S., Vinogradova, O.S., Karanov, A.M., 1985. Frequency modulation of neuronal theta-bursts in rabbit's septum by low-frequency repetitive stimulation of the afferent pathways. Neuroscience 14, 501–508. Burazin, T.C.D., Bathgate, R.A.D., Macris, M., Layfeld, S., Gundlach, A.L., Tregear, G.W., 2002. Restricted, but abundant, expression of the novel rat gene-3 (R3) relaxin in the dorsal tegmental region of brain. J. Neurochem. 82, 1553–1557. Buzsáki, G., 2002. Theta oscillations in the hippocampus. Neuron 33, 325–340. Chronwall, B.M., Skirboll, L.R., O'Donohue, T.L., 1985. Demonstration of a pontine-hippocampal projection containing a ranatensin-like peptide. Neurosci. Lett. 53, 109–114. Goto, M., Swanson, L.W., Canteras, N.S., 2001. Connections of the nucleus incertus. J. Comp. Neurol. 438, 86–122. Haglund, L., Swanson, L.W., Kohler, C., 1984. The projection of the supramammillary nucleus to the hippocampal formation: an immunohistochemical and anterograde transport study with the lectin PHA-L in the rat. J. Comp. Neurol. 229, 171–185. Hasselmo, M.E., 2005. What is the function of hippocampal theta rhythm?—linking behavioral data to phasic properties of field potential and unit recording data. Hippocampus 15, 936–949. Jennes, L., Stumpf, W.E., Kalivas, P.W., 1982. Neurotensin: topographical distribution in rat brain by immunohistochemistry. J. Comp. Neurol. 210, 211–224. Kirk, I.J., Oddie, S.D., Konopacki, J., Bland, B.H., 1996. Evidence for differential control of posterior hypothalamic, supramammillary, and medial mammillary theta-related cellular discharge by ascending and descending pathways. J. Neurosci. 16, 5547–5554. Kiss, J., Csáki, A., Bokor, H., Shanabrough, M., Leranth, C., 2000. The supramammillo-hippocampal and supramammillo-septal glutamatergic/aspartatergic projections in the rat: a combined [3H]D-aspartate autoradiographic and immunohistochemical study. Neuroscience 97, 657–669. Kitchigina, V.F., Kudina, T.A., Kutyreva, E.V., Vinogradova, O.S., 1999. Neuronal activity of the septal pacemaker of theta rhythm under the influence of stimulation and blockade of the median raphe nucleus in the awake rabbit. Neuroscience 94, 453–463. Klemm, W.R., 1972. Effects of electric stimulation of brain stem reticular formation on hippocampal theta rhythm and muscle activity in unanesthetized, cervical- and midbrain-transected rats. Brain Res. 41, 331–344. Leranth, C., Kiss, J., 1996. A population of supramammillary area calretinin neurons terminating on medial septal area cholinergic and lateral septal area calbindin-containing cells are aspartate/glutamatergic. J. Neurosci. 16, 7699–7710. Lynch, M.A., 2004. Long-term potentiation and memory. Physiol. Rev. 84, 87–136. Ma, S., Bonaventure, P., Ferraro, T., Shen, P.J., Burazin, T.C.D., Bathgate, R.A.D., Liu, C., Tregear, G.W., Sutton, S.W., Gundlach, A.L., 2007. Relaxin-3 in GABA projection neurons of nucleus incertus suggests widespread influence on forebrain circuits via G-protein-coupled receptor-135 in the rat. Neuroscience 144 (1), 165–190. Macadar, A.W., Chalupa, L.M., Lindsley, D.B., 1974. Differentiation of brain stem loci which affect hippocampal and neocortical electrical activity. Exp. Neurol. 43, 499–514. Maglóczky, Z., Acsády, L., Freund, T.F., 1994. Principal cells are the postsynaptic targets of supramammillary afferents in the hippocampus of the rat. Hippocampus 4, 322–334.

96

BR A IN RE S EA RCH 1 2 18 ( 20 0 8 ) 8 7 –96

Martin, G.F., Vertes, R.P., Waltzer, R., 1985. Spinal projections of the gigantocellular reticular formation in the rat. Evidence for projections from different areas to laminae I and II and lamina IX. Exp. Brain Res. 58, 154–162. Nuñez, A., Cervera-Ferri, A., Olucha-Bordonau, F., Ruiz-Torner, A., Teruel, V., 2006. Nucleus incertus contribution to hippocampal theta rhythm generation. Eur. J. Neurosci. 23, 2731–2738. Nuñez, A., Andrés, I.d., García-Austt, E., 1991. Relationships of nucleus reticularis pontis oralis neuronal discharge with sensory and carbachol evoked hippocampal theta rhythm. Exp. Brain Res. 87, 303–308. Olucha-Bordonau, F.E., Teruel, V., Ruiz-Torner, A., Barcia-González, J., Valverde-Navarro, A.A., Martínez-Soriano, F., 2003. Cytoarchitecture and efferent projections of the nucleus incertus of the rat. J. Comp. Neurol. 464, 62–97. Paxinos, G., 1985. The Rat Nervous System. Academic Press, Sydney. Orlando. Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates. Academic Press, Sydney. New York. Petsche, H., Stumpf, C., 1962. The origin of theta-rhythm in the rabbit hippocampus. Wien. Klin. Wochenschr. 74, 696–700. Robinson, T.E., Vanderwolf, C.H., 1978. Electrical stimulation of the brain stem in freely moving rats: II. Effects on hippocampal and neocortical electrical activity, and relations to behavior. Exp. Neurol. 61, 485–515. Sakata, S., 2006. Timing and hippocampal theta in animals. Rev. Neurosci. 17 (1–2), 157–162. Stewart, M., Fox, S.E., 1989a. Two populations of rhythmically bursting neurons in rat medial septum are revealed by atropine. J. Neurophysiol. 61, 982–993.

Stewart, M., Fox, S.E., 1989b. Firing relations of medial septal neurons to the hippocampal theta rhythm in urethane anesthetized rats. Exp. Brain Res. 77, 507–516. Tanaka, M., Iijima, N., Miyamoto, Y., Fukusumi, S., Itoh, Y., Ozawa, H., Ibata, Y., 2005. Neurons expressing relaxin 3/INSL 7 in the nucleus incertus respond to stress. Eur. J. Neurosci. 21, 1659–1670. van der Westhuizen, E., Werry, T.D., Sexton, P.M., Summers, R.J., 2007. The relaxin family peptide receptor 3 activates extracellular signal-regulated kinase 1/2 through a protein kinase C-dependent mechanism. Mol. Pharmacol. 71 (6), 1618–1629. Veenman, C.L., Reiner, A., Honig, M.G., 1992. Biotinylated dextran amine as an anterograde tracer for single- and double-labeling studies. J. Neurosci. Methods 41, 239–254. Vertes, R.P., 1981. An analysis of ascending brain stem systems involved in hippocampal synchronization and desynchronization. J. Neurophysiol. 46, 1140–1159. Vertes, R.P., 1988. Brainstem afferents to the basal forebrain in the rat. Neuroscience 24, 907–935. Vertes, R.P., 2005. Hippocampal theta rhythm: a tag for short-term memory. Hippocampus 15, 923–935. Vertes, R.P., Martin, G.F., 1988. Autoradiographic analysis of ascending projections from the pontine and mesencephalic reticular formation and the median raphe nucleus in the rat. J. Comp. Neurol. 275, 511–541. Vertes, R.P., Kocsis, B., 1997. Brainstem-diencephaloseptohippocampal systems controlling the theta rhythm of the hippocampus. Neuroscience 81, 893–926. Vertes, R.P., McKenna, J.T., 2000. Collateral projections from the supramammillary nucleus to the medial septum and hippocampus. Synapse 38, 281–293. Vertes, R.P., Fortin, W.J., Crane, A.M., 1999. Projections of the median raphe nucleus in the rat. J. Comp. Neurol. 17, 555–582.