Polysynaptic olfactory pathway to the ipsi- and contralateral entorhinal cortex mediated via the hippocampus

Neuroscience 130 (2005) 249 –258

POLYSYNAPTIC OLFACTORY PATHWAY TO THE IPSI- AND CONTRALATERAL ENTORHINAL CORTEX MEDIATED VIA THE HIPPOCAMPUS L. UVA AND M. DE CURTIS*

Key words: isolated guinea-pig brain, lateral olfactory tract, current source density, silicon probes.

Department of Experimental Neurophysiology, Istituto Nazionale Neurologico “Carlo Besta,” via Celoria 11, 20133 Milano, Italy

The olfactory input to the parahippocampal region supports essential brain functions that regulate spatial orientation, sensory recognition, learning and memory. The output generated by the neurons of the primary olfactory station, the olfactory bulbs, diffusely distributes via the lateral olfactory tract (LOT) into the piriform cortex and to neurons of the superficial layers of the rostral entorhinal cortex (EC), from where a projection to the hippocampal formation originates. The olfactory input to the limbic region has been extensively analyzed with anatomical techniques (van Groen and Wyss, 1990; Krettek and Price, 1977) and the presence of a polysynaptic projection from the olfactory bulbs to the ipsilateral EC and to the dentate gyrus of the hippocampus has been confirmed by in vivo electrophysiological recordings (Habets et al., 1980; Liu and Bilkey, 1997). The contralateral propagation of the olfactory input to the EC has never been analyzed; since anatomical studies (van Groen and Wyss, 1988; Laurberg, 1979) and in vivo electrophysiological recordings (Bartesaghi and Gessi, 1990; Bartesaghi et al., 1989; Bartesaghi and Gessi, 1986; Deadwyler et al., 1975) have shown commissural connections between in the hippocampus– EC system, a bilateral propagation of the olfactory input via the hippocampal formation can be predicted. We recently showed that repetitive electrical stimulation of the LOT at 2– 8 Hz induced a late-delay response at circa 60 ms in the caudal/medial portion of the EC that does not receive a direct olfactory projection. Optical imaging of voltage-dependent signals (Biella et al., 2003) and multielectrode mapping (Gnatkovsky et al., 2003) revealed that such a response has a delay consistent with a reentrant EC loop via the hippocampal formation. Nevertheless, the complete circuitry activated by the olfactory input was never described and is still unresolved. In the present study we directly analyzed the propagation patterns determined by olfactory tract stimulation by performing simultaneous laminar profile analysis in different hippocampal subfields and in the EC of the isolated guinea-pig brain preparation maintained in vitro by arterial perfusion (de Curtis et al., 1991, 1998; Muhlethaler et al., 1993). We definitely demonstrate that the LOT input generates a delayed response in the ipsilateral EC, mediated via a multisynaptic pathway that sequentially involves the lateral EC, the dentate gyrus, the CA3 and the CA1 regions. The commissural propagation of the olfactory input to the con-

Abstract—Interactions between olfactory cortices and the hippocampus support sensory discrimination and spatial learning functions. The olfactory input accesses the hippocampal formation via a polysynaptic pathway mediated by the lateral and rostral entorhinal cortex (EC). We recently demonstrated that following repetitive stimulation of the lateral olfactory tract (LOT) at 2– 8 Hz, a delayed response (onset at circa 60 ms) was evoked in the caudal portion of the EC, identified as medial EC, that does not receive a direct olfactory input. By performing simultaneous laminar profile analysis in the EC and in different hippocampal subfields, we conclusively demonstrate that the delayed EC response evoked by repetitive ipsilateral LOT stimulation is headed by the sequential activation of the dentate gyrus and the CA3/ CA1 subfields in the septal and temporal hippocampus. Repetitive stimulation of the contralateral LOT also induced an EC response that peaked at 76.28ⴞ2.42 ms (nⴝ15). Current source density analysis and time-delay analysis of simultaneous field potential laminar profiles performed from the EC and from DG, CA3 and CA1 hippocampal subfields suggested that the contralateral EC response is mainly carried by an intrahippocampal CA3–CA3 commissural pathway. Contralateral LOT stimulation also induced a later EC component (delay >100 ms) generated in the superficial layers, mediated either by local associative interactions or by extrahippocampal circuits. The opportunity to activate the ipsi- and contralateral olfactory pathways in the same experiment and to record field potentials profiles simultaneously in different structures of both hemispheres in the isolated guinea-pig brain confirms that this preparation is unique and is particularly suitable for investigating the system physiology of the limbic region. The present study demonstrates that patterned stimulation of the olfactory input that mimics sniffing patterns during odor discrimination induces a diffuse activation of both ipsi- and contralateral hippocampi and ECs. The findings contribute to the understanding the physiological mechanisms that underlie associative interactions between olfactory and nonolfactory cortical inputs converging into the mesial temporal region. © 2004 IBRO. Published by Elsevier Ltd. All rights reserved. *Corresponding author. Tel: ⫹39-02-239-4280; fax: ⫹39-02-70600775. E-mail address: [email protected] (M. de Curtis). Abbreviations: CSD, current source density; DG, dentate gyrus; EC, entorhinal cortex; GCL, granule cell layer; HF, hippocampal fissure; LD, Lamina dissecans; LOT, lateral olfactory tract; LUC, stratum lucidum; ML, molecular layer; ORI, stratum oriens; PCL, pyramidal cell layer; RAD, stratum radiatum.

0306-4522/05$30.00⫹0.00 © 2004 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2004.08.042

249

250

L. Uva and M. de Curtis / Neuroscience 130 (2005) 249 –258

tralateral hippocampus and entorhinal region was also characterized. The results have been preliminarily reported in abstract form (Uva and de Curtis, 2003).

EXPERIMENTAL PROCEDURES Experiments were performed on 41 guinea-pig brains isolated in vitro according to the methods described in detail elsewhere (de Curtis et al., 1991, 1998; Muhlethaler et al., 1993). After careful and rapid isolation, the brain was perfused through the basilar artery with a complex saline solution (126 mM NaCl, 3 mM KCl, 1.2 mM KH2PO4, 1.3 mM MgSO4, 2.4 mM CaCl2, 26 mM NaHCO3, 15 mM glucose, 2.1 mM HEPES (N-2-Hydroxylethylpiperazine-N-zethanesulphoric acid) and 3% dextran M. W. 70,000), saturated with a 95% O2–5% CO2 gas mixture (pH 7.3). Recordings were performed at 32 °C. The

experimental protocol was reviewed and approved by the Committee on Animal Care and Use and by the Ethical Committee of the Istituto Nazionale Neurologico, according to the international guidelines on ethical use of animals. All efforts were made to minimize the number of animals used and their suffering. Extracellular laminar profiles were performed with linear multichannel silicon probes that featured i) 16 iridium recording sites separated by 50 ␮m or 100 ␮m on a single shaft (kindly provided by Jamille Hetke of the Center of Neural Communication Technology, University of Michigan, Ann Arbor, MI, USA); ii) 32 platinum recording sites at 50 ␮m (kindly provided by Peter Norlin from ACREO, Krista, Sweden). Silicon probes were inserted with a ventral approach into the lateral and medial EC and in different temporo-septal segments of the dentate gyrus (DG), the CA3 and the CA1 region of the hippocampus of one hemisphere. The probe orientation was calcu-

Fig. 1. Field potentials laminar profiles at different sites in the temporal (left) and septal (right) portions of the hippocampus performed with 16-channels silicon probes (the separation between the recording contacts is 100 ␮m). Clear potential reversals were observed at all sites. Laminar profiles were simultaneously recorded in the hippocampus and in the medial EC (m-EC) with two 16-channel probes inserted in the m-EC (lower profiles) and either in DG or in CA1/CA3. Dotted vertical lines mark DG and CA1 responses and presumed CA3 potentials (filled circles, triangles and squares, respectively). The borders of the layers in DG, CA1, CA3 and medial EC (m-EC) are indicated (GCL, granular cell layer; ML, molecular layer; OR, stratum oriens; PCL, pyramidal cell layer; RAD, stratum radiatum).

L. Uva and M. de Curtis / Neuroscience 130 (2005) 249 –258

251

lated on a guinea-pig stereotaxic atlas (Luparello, 1967) to reach the hippocampus as perpendicular as possible to the lamination of the target structure. Current source density (CSD) analysis was performed with 200 ␮m separation grid on laminar field potential profiles obtained by averaging 7 to 10 LOT-evoked responses (see Biella and de Curtis, 2000; de Curtis et al., 1994). Data were recorded with a 32-channel amplifier (Biomedical Engineering, Thornwood, NY, USA) and were digitized via an AT-MIO-64E3 National board (National Instrument, Milano, Italy). Data analysis was performed with custom-made software developed in Labview by Vadym Gnatkovsky. Twisted, teflon-insulated silver wires were utilized for bipolar stimulation of the LOT, ipsi- and contralateral to the hemisphere in which recordings were performed. LOT stimuli at an intensity just threshold for maximal amplitude responses in the lateral EC were delivered through Grass isolation units driven by a Grass S88 pulse generator. Repetitive stimulations at 2 Hz stimuli or twin stimuli with interstimulus delay of 150 –200 ms repeated at 0.6 – 1 Hz were effective in producing a delayed EC response. Recording and stimulating electrodes were positioned under direct visual control with a stereoscopic microscope. At the end of the electrophysiological experiment electrolytic lesions were performed by passing a 30 ␮A current for 30 s between the two deepest recording sites of the silicon probes. After brain fixation with 4% paraformaldehyde, the position of the electrodes was identified histologically on 75–100 ␮m coronal sections counterstained with thionine.

RESULTS

Fig. 2. Upper panel: Field potential laminar profile obtained with a 16-channel silicon probe inserted in the DG (filled circle) and CA1 (filled triangle). Recording sites were separated by 100 ␮m on the probe shaft. Middle panel: the CSD profile derived from the field potential profile above; positive and negative deflections correspond to sinks and source respectively. Lower panel: CSD contour plot of the sinks (continuous lines) and sources (dotted lines). The DG sink is localized in the ML and the polysynaptic CA1 sink is in the RAD, where Schaeffer collateral fibers terminate. As in Fig. 1, filled circles and triangles point at DG and CA1 potentials/sinks, respectively. The position of CSD sinks/sources was reconstructed after histological verification of the electrode track shown on the histological section on the right of the contour plot. CSD iso-current lines⫽0.03 mV/mm2. GCL, granule cell layer; ML, molecular layer; PCL, pyramidal cell layer; RAD, stratum radiatum.

The first objective of the study was to evaluate whether the delayed response around 60 ms observed in the EC after LOT stimulation is preceded by the sequential activation of the trisynaptic loop within the hippocampus. Repetitive LOT stimulation ipsilateral to the hemisphere in which recordings were performed entrained both the septal and temporal portions of the hippocampus. As shown in the experiment illustrated in Fig. 1, ipsilateral LOT stimuli generated local responses in both septal and temporal regions of the DG (filled circles) and CA1 (filled triangles), as indicated by the presence of a potential reversal in the laminar profiles recorded with 16-channel silicon probes. DG and CA1 responses preceded the potential simultaneously recorded in the medial EC with a second 16channel silicon probe (lower profiles in Fig. 1; see Table 1). Laminar profile recordings from the CA3 subfield could be performed successfully in the septal portion of the hippocampus in two experiments only (filled square in the middle profile in Fig. 1; see Discussion). To further localize the generators of the field potentials, CSD analysis of laminar profiles was performed and the position of the recording silicon probes was reliably reconstructed on histological coronal brain sections, as described in the Experimental Procedures. Fig. 2 shows a field potential laminar profile and the relative CSD contour plot (lower panel) performed in the septal DG (filled circle) and CA1 (filled triangle) with a 16-channel probe (100 ␮m leads separation). No clear sink in CA1/CA3 was detected at a delay comparable with that of the DG response. The position of the CSD sinks was reconstructed from the histological identification of the electrode track illustrated on the right of the contour plot. These findings demonstrate i) that the entire septo-temporal extent of the hippocampus is acti-

252

L. Uva and M. de Curtis / Neuroscience 130 (2005) 249 –258

Table 1. Onset and peak amplitude delays of the ispi- and contralateral responses evoked in the hippocampus and EC by LOT stimulation Ipsilateral stimulation

DG Septal Temporal CA1 Septal Temporal m-EC

Contralateral stimulation

Onset

Peak

n

32.67⫾0.96 31.52⫾1.64

41.17⫾1.25 40.59⫾0.81

19 28

Septal Temporal

64.5⫾2.02 61.8⫾7.12

77.25⫾3.40 74.25⫾6.56

4 6

49.87⫾1.86 52.01⫾1.91

57.20⫾2.74 55.89⫾1.95

12 22

Septal Temporal

57.42⫾1.70 58.82⫾1.52

67.28⫾2.23 66.45⫾1.60

9 9

59.96⫾1.67

66.51⫾2.73

25

Deep Superficial

69.71⫾3.32 100.10⫾3.75

76.28⫾2.42 120.87⫾5.30

15 16

vated by the ipsilateral LOT stimulus and ii) that the hippocampal activation precedes the delayed EC response.

Onset

Peak

n

We further analyzed the time dependence of the development of the hippocampal and the delayed EC poten-

Fig. 3. Repetitive LOT stimulation ipsi- (upper part) and contralateral (lower part) to the recorded hemisphere induced sequential activation of the hippocampal–EC pathway. The scheme of the electrode placement is illustrated in the inset. Two electrodes were utilized, one located in the medial EC (lower traces) and the other in a position in which potentials generated in DG and CA1 were recorded (upper traces). Traces were separated by 10 s during continuous repetitive LOT stimulation. A progressive entrainment of DG (filled circles, left traces), CA1 (filled triangles, middle traces) and m-EC (asterisks, right traces) was observed during ipsi-lateral LOT stimulation. CA1 and m-EC, but not DG, were activated in sequence upon contralateral LOT stimulation.

L. Uva and M. de Curtis / Neuroscience 130 (2005) 249 –258

253

Fig. 4. Propagation of activity evoked by ipsi- and contra-lateral LOT stimulation in the hippocampus–EC. (A) Simultaneous recordings were performed in DG, CA1 and m-EC following both ipsi- and contralateral LOT stimulation. (B) Potentials recorded in CA1/CA3 and in m-EC in a different experiment. DG, CA3 and CA1 potentials are marked by filled circle, square and triangle, respectively. Note the absence of DG activation in response to contralateral LOT stimulation.

tial during the repetitive LOT stimulation. Neuronal activity was recorded from the same position in one hemisphere and stimulation was performed from either ipsi- or contralateral LOT. Single shock stimulation of the LOT induced a response restricted to the lateral EC ipsilateral to the stimulus exclusively. The progressive entrainment of sequential synaptic stations during repetitive stimulation along the EC– hippocampal–EC pathway (see Table 1) was demonstrated by performing simultaneous extracellular recordings in the EC and in each of the following subfields of the hippocampus: DG (n⫽47), CA3 region (n⫽2) and CA1 region (n⫽34). As illustrated in the representative experiment in Fig. 3, simultaneous recordings at the border between DG and CA1 and in the medial EC (m-EC) showed an isolated response in the DG at the onset of repetitive ipsilateral LOT stimulation (filled circles; left traces in Fig. 3). Within a few seconds a potential in the CA1 region appeared (filled triangles; middle traces) and increased in amplitude to reach a threshold value, above which a delayed response in the EC was observed (asterisks). Contralateral LOT stimulation was also effective in invading CA1 of the opposite hemisphere. CA1 responses (see Table 1), but not DG responses, were consistently observed before the appearance of the delayed EC potential (lower panel in Fig. 3). The above data suggest that the delayed EC potentials induced by olfactory stimulation are boosted by the previ-

ous activation of the ipsi- and contralateral CA1 region. Next we analyzed in details the hippocampal circuit that sustains the contralateral EC response. In 10 of 20 experiments, contralateral LOT stimulation induced small amplitude responses in the septal and temporal DG that peaked at 77.25⫾3.40 ms (n⫽4) and 74.25⫾6.56 ms (n⫽6), respectively. Neither CA1 nor EC activation followed contralateral DG responses. In the experiments in which a DG response was observed, the active sinks evoked by ipsiand contralateral LOT stimuli co-localized in the outer molecular layer (n⫽10; Fig. 5, upper panel). As mentioned above, contralateral CA1 potentials were consistently observed before contralateral EC activation, even in experiments in which no DG response was observed (Fig. 4, lower traces). Contralateral LOT stimulation diffusely activated CA1 region with peak delays of 67.28⫾2.23 ms (n⫽9) and 66.54⫾1.60 ms (n⫽9) in septal and temporal portions, respectively (see Table 1). In Fig. 4B a sharp potential, possibly located in CA3 (filled square) was observed in response to both ipsi- (upper traces) and contralateral LOT stimuli (lower traces). CSD laminar analysis demonstrated that the sinks responsible for the CA1 response to ipsi- and contralateral LOT stimuli localized at the same depth in the stratum radiatum (n⫽9; Fig. 5, lower panel). Contralateral LOT stimulation induced delayed EC potentials with different complexities (see Table 1). In Fig. 6

254

L. Uva and M. de Curtis / Neuroscience 130 (2005) 249 –258

Fig. 5. Depth co-localization of current sinks in DG (upper panel) and CA1 (lower panel) following ipsi-and contralateral LOT stimulation CA1. On the left, superimposed traces obtained with 16-channel silicon probe are shown. The electrode position was not changed during ipsi- and contralateral stimulation. The CDS contour plots are illustrated on the right. The borders of DG and CA1 layers are indicated (GCL, granular cell layer; ML, molecular layer; OR, stratum oriens; PCL, pyramidal cell layer; RAD, stratum radiatum). CSD iso-current lines⫽0.01 mV/mm2.

typical EC responses recorded at identical sites following ipsi- (upper profiles in A) and contralateral LOT stimulation (lower profiles in A) are illustrated. Contralateral LOT responses showed a similar reversal pattern, with a delay of circa 10 ms (peak delay at 76.28⫾2.42 ms; n⫽15) in comparison to the response to ipsilateral LOT stimuli. In 16 experiments contralateral stimulation also induced a response with a longer delay (lower right profile; 100.10⫾3.75 ms). CSD analysis demonstrated that the early EC response evoked by contralateral LOT stimulation correlated with a current sink in deep layers (left panel in Fig. 6B), whereas the late potential was associated with a sink in superficial layers (asterisk in the right panel in Fig. 6B). Finally, contralateral LOT stimulation also induced small amplitude responses in the PPC and in the rostral part of the lateral EC, possibly mediated by the anterior commissure (not shown). The peak delay of the lateral EC response to contralateral LOT stimulation jittered between 35 and 60 ms.

DISCUSSION The olfactory input targets layer II and III neurons of the rostral EC either directly or via the piriform cortex (Krettek and Price, 1974; Haberly and Price, 1977; Habets et al., 1980; Kosel et al., 1981; Wouterlood and Nederlof, 1983; Room et al., 1984; Biella and de Curtis, 1995; Liu and Bilkey, 1997; Chapman and Racine, 1997). Superficial layer neurons, in turn, send off axons to the hippocampal formation (Wilson, 1978; Habets et al., 1980), from where a feedback projection to the EC originates (for review see Witter et al., 1989; Witter, 1993). We recently confirmed with electrophysiological and imaging techniques that the LOT input induces an early direct activation of the rostral portion of the EC, followed by a delayed response that generates extracellular current sinks in both superficial and deep layers of the medial part of the EC (Biella et al., 2003; Gnatkovsky et al., 2003). No simultaneous activation of the CA1 and the DG was observed in our experiments. In

L. Uva and M. de Curtis / Neuroscience 130 (2005) 249 –258

addition, because of the problems mentioned below (see next paragraph) we could not detect any direct EC–CA3 response, as previously reported (Habets et al., 1980; Wouterlood and Nederlof, 1983; Liu and Bilkey, 1997). Since EC–CA1/CA3 and EC–DG inputs are mediated by neurons located respectively in layer III and layer II of the EC (see Witter, 1993), our findings suggest that the olfactory input mediated by the activation of LOT preferentially targets neurons in EC layer II, since the largest sink we observed was located in the ML of the DG. The delayed EC component has been suggested to depend on the previous activation of the hippocampal formation (Biella and de Curtis, 2000). The present study definitely demonstrates that the delayed EC response induced by repetitive

255

stimulation of the LOT is preceded by the activation of the entire septo-temporal extent of the hippocampus along a pathway that sequentially involves the DG, CA3 and CA1. The scheme in Fig. 7 summarizes the presumed circuit responsible for the hippocampal/entorhinal propagation of the olfactory input. Only in two experiments we could identify a clear potential component in CA3 by laminar analysis (see Fig. 1). The difficulties in localizing a CA3 response probably relate to the fact that DG fibers terminate on both apical and basal dendrites of CA3 neurons (Swanson et al., 1978). The simultaneous supra-pyramidal and infra-pyramidal input carried by mossy fibers, indeed, could hinder the detection of a clear potential reversal in CA3 because of the

Fig. 6. (A) Laminar profiles performed in the m-EC with 16-channel silicon probes in response to stimuli delivered with ipsi- (upper traces) and contralateral (lower traces) LOT electrodes. Couples of upper/lower profiles were recorded at the same m-EC site. (B) Correlation between potentials and CSD sink/sources performed on the two profiles on the right in A. Continuous and dotted line represent sinks and sources, respectively. The asterisks mark the late EC response generated in the superficial layers at ⬎100 ms delay. CSD iso-current lines in B⫽0.015 mV/mm2. The borders of the layers of m-EC are shown. LD, lamina dissecans.

256

L. Uva and M. de Curtis / Neuroscience 130 (2005) 249 –258

Fig. 7. Scheme of the hypothetical circuit responsible for the ipsi- and contralateral EC projections of the olfactory input mediated via the hippocampus (see text). LOT stimulation is illustrated on the left. Dotted lines represent commissural fibers. Question marks are positioned on pathways that could not be identified in the present study, responsible for the late, contra-lateral, superficial EC response.

dendritic scattering of the underlying sink generators. The involvement of CA3 was indirectly supported by the delay analysis of the responses in DG and CA1. Both onset and peak delays in DG and CA1 were separated by circa 15–20 ms; this value is compatible with the existence of an interposed synapse between DG and CA1. The delay of the CA3 response detected in the two mentioned experiments corroborates this assumption (see Figs. 1 and 4). Interestingly, stimulation of the LOT also activated the hippocampus/EC of the contralateral hemisphere. In our experimental conditions contralateral responses were observed in the DG, in CA1 and in the EC. In principle, either EC–DG or commissural DG–DG pathway could be responsible for the contralateral DG response. The former possibility is the most likely, since, in line with tracing observations (Steward and Scoville, 1976), both ipsi- and contralateral sinks observed in our experiments co-localized in the outer molecular layer; anatomical data demonstrate a more proximal termination of the commissural DG–DG fiber system on the dendrites of granule cells (Blackstad, 1956; Golarai and Sutula, 1996). Our findings exclude the possibility that the DG–DG commissural pathway is responsible for the contralateral propagation of activity to the EC, since i) contralateral DG responses showed latencies longer than CA1 responses and equal to the EC potentials recorded in the same hemisphere, ii) delayed EC responses to contralateral LOT stimuli were observed also in the absence of contralateral DG potentials, and iii) the amplitude of DG responses evoked by contralateral LOT stimulation was consistently smaller than that necessary to induce a CA1 response in response to ipsilateral LOT stimulation. An anatomical study demonstrates that the DG–DG commissural projection in the guinea-pig is weak in comparison to the rat and the monkey (van Groen and Wyss, 1988). In vivo studies in the guinea-pig by Bartesaghi and Gessi (1986, 1990) and Bartesaghi et al. (1989) also showed that a massive activation of CA3 and CA1 is achieved without a previous activation of the DG after the

stimulation of ventral psalterium. Thus, while DG is essential for the activation of the ipsilateral entorhinal– hippocampal loop, commissural DG–DG projections do not substantially contribute to inter-hemispheric EC activity propagation. The approximately 10 ms delay between the ipsi- and the contralateral CA1 responses could be due to propagation of activity along different contingents of commissural fibers. The largest and preeminent commissural pathway connects CA3 regions of the two hemispheres (Gottlieb and Cowan, 1973; Deadwyler et al., 1975; Laurberg, 1979). A fiber contingent connect CA3 with the contralateral CA1 (Laurberg, 1979), whereas no CA1–CA1 commissural system was revealed with anatomical and physiological techniques (Deadwyler et al., 1975). Moreover, CA3–CA1 commissural fibers diffusely terminate in the radiatum– oriens layers (Laurberg, 1979) while ipsilateral CA3–CA1 fibers preferentially terminate the stratum radiatum within the same hemisphere (Amaral and Witter, 1989; Swanson et al., 1978; Laurberg, 1979; Witter, 1993). Since we demonstrated that the CA1 sinks generated by ipsiand contralateral LOT stimulation co-localized in the same position in the stratum radiatum, a contribution of the CA3– contralateral CA1 pathway to the commissural propagation of the activity should be excluded. Therefore, our experiments suggest that the CA3–CA3 path is the most likely path responsible for the commissural propagation within the hippocampus. This hypothesis could not be directly verified, because of the difficulties discussed above to unequivocally localize CA3. The possibility that anterior commissural fiber systems in the olfactory cortex or in the amygdala (Haberly and Price, 1978) enforce propagation to the contralateral EC can also be excluded. Such a projection, indeed, generates small amplitude responses in the most rostral portion of the contralateral EC with a delay equal or shorter than the one observed in contralateral CA1 (Gnatkowsky, Uva and de Curtis, unpublished observations).

L. Uva and M. de Curtis / Neuroscience 130 (2005) 249 –258

A late onset EC component associated to a sink in the superficial layers was observed during contralateral EC stimulation. Its long delay (⬎100 ms from the LOT stimulus) suggests that it is mediated by a polysynaptic pathway. Several elements play against the possibility that such a potential is carried by the commissural EC–EC fiber system (Laurberg, 1979; Blackstad, 1956; Sorensen, 1985; Kohler, 1986). First, the neurons of origin of the contralateral EC projection are located in layers II and III. These layers are only weakly activated by the hippocampal output (Gnatkovsky et al., 2003) and, therefore, are not likely to generate an output large enough to induce the late contralateral response. Secondly, the commissural propagation from the rostral EC directly activated by the olfactory input occurs between 20 and 30 ms, while the late contralateral EC response was observed at circa 100 ms, a latency that is not compatible with a single synapse propagation. The same consideration also stands for the time lag between the delayed ipsilateral EC response (circa 66 ms) and the contralateral late EC response. The most likely sources responsible for the superficial-layer contralateral EC potential reside either in local associative EC interactions or in pathways independent from the activation of the intra-hippocampal and intra-temporal networks that have been considered. The present study demonstrates that the hippocampus and the EC of the guinea-pig are bilaterally activated by repetitive LOT stimulation at a particular frequency range (1–2 Hz or slow theta determined by twin stimuli) that in mammals mimics sniffing patterns observed during odor discrimination (Bressler, 1987; Laurent et al., 2001; Chaput, 2000) and odour-driven oscillatory activity in olfactory bulbs (Macrides and Chorover, 1972; Macrides et al., 1982; Chaput, 2000; Margrie and Schaefer, 2003) and piriform cortex (Wilson, 1998). It is tempting to speculate that large amplitude oscillations generated in the primary olfactory cortex during sniffing are effective in diffusely activating the hippocampus/EC region bilaterally, therefore promoting associative interactions between olfactory signals and non-olfactory cortical inputs converging in the mesial temporal region. Acknowledgments—Multi-channel silicon probes were kindly provided by Jamille Hetke of the Center for Neural Communication Technology (sponsored by NIH NIBIB grant P41-RR09754) and by Peter Norlin of ACREO (Krista, Sweden). The study was supported by the European Community grant VSAMUEL (IST 199910073). L.U. was partially supported by the Mariani Foundation and by the Italian Ministry of Health. We would like to thank Renata Bartesaghi for helpful discussion and valuable comments.

REFERENCES Amaral DG, Witter MP (1989) The three-dimensional organization of the hippocampal formation: a review of anatomical data. Neuroscience 31:571–591. Bartesaghi R, Gessi T (1986) Hippocampal output to the subicular complex: an electrophysiological study. Exp Neurol 92:114 –133. Bartesaghi R, Gessi T (1990) Electrophysiological analysis of the hippocampal output to the presubiculum. Neuroscience 37: 335–345.

257

Bartesaghi R, Gessi T, Sperti L (1989) Electrophysiological analysis of the hippocampal projections to the entorhinal area. Neuroscience 30:51– 62. Biella G, de Curtis M (1995) Associative synaptic potentials in the piriform cortex of the isolated guinea-pig brain in vitro. Eur J Neurosci 7:54 – 64. Biella G, de Curtis M (2000) Olfactory inputs activate the medial entorhinal cortex via the hippocampus. J Neurophysiol 83: 1924 –1931. Biella GR, Gnatkovsky V, Takashima I, Kajiwara R, Iijima T, de Curtis M (2003) Olfactory input to the parahippocampal region of the isolated guinea pig brain reveals weak entorhinal-to-perirhinal interactions. Eur J Neurosci 18:95–101. Blackstad TW (1956) Commissural connections of the hippocampal region in the rat, with special reference to their mode of termination. J Comp Neurol 105:417–537. Bressler SL (1987) Relation of olfactory bulb and cortex: I. Spatial variation of bulbocortical interdependence. Brain Res 409:285–293. Chapman AC, Racine RJ (1997) Converging inputs to the entorhinal cortex from the piriform cortex and medial septum: facilitation and current source density analysis. J Neurophysiol 78:2602–2615. Chaput MA (2000) EOG responses in anesthetized freely breathing rats. Chem Senses 25:695–701. de Curtis M, Biella G, Buccellati C, Folco G (1998) Simultaneous investigation of the neuronal and vascular compartments in the guinea pig brain isolated in vitro. Brain Res Prot 3:221–228. de Curtis M, Biella G, Forti M, Panzica F (1994) Multifocal spontaneous epileptic activity induced by restricted bicuculline ejection in the piriform cortex of the isolated guinea pig brain. J Neurophysiol 71:2463–2475. de Curtis M, Pare D, Llinás RR (1991) The electrophysiology of the olfactory-hippocampal circuit in the isolated and perfused adult mammalian brain in vitro. Hippocampus 1:341–354. Deadwyler SA, West JR, Cotman CW, Lynch G (1975) Physiological studies of the reciprocal connections between the hippocampus and entorhinal cortex. Exp Neurol 49:35–57. Gnatkovsky V, Biella GR, de Curtis M(2003) Interactions between parahippocampal subfields in the isolated guinea pig brain. Society for Neuroscience, 33rd Annual Meeting, Abstract 719:7, 85. Golarai G, Sutula T (1996) Bilateral organization of parallel and serial pathways in the dentate gyrus demonstrated by current-source density analysis in the rat. J Neurophysiol 75:329 –342. Gottlieb DI, Cowan WM (1973) Autoradiographic studies of the commissural and ipsilateral association connection of the hippocampus and detentate gyrus of the rat: I. The commissural connections. J Comp Neurol 149:393– 422. Haberly LB, Price JL (1977) The axonal projection patterns of the mitral and tufted cells of the olfactory bulb in the rat. Brain Res 129:152–157. Haberly LB, Price JL (1978) Association and commissural fiber systems of the olfactory cortex of the rat: II. Systems originating in the olfactory peduncle. J Comp Neurol 181:781– 807. Habets AMM, Lopes Da Silva FH, Mollevanger WJ (1980) An olfactory input to the hippocampus of the cat: field potential analysis. Brain Res 182:47– 64. Kohler C (1986) Intrinsic connections of the retrohippocampal region in the rat brain: II. The medial entorhinal area. J Comp Neurol 246:149 –169. Kosel KC, Van Hoesen GW, West JR (1981) Olfactory bulb projection to the parahippocampal area of the rat. J Comp Neurol 198:467– 482. Krettek JE, Price JL (1974) A direct input from the amygdala to the thalamus and the cerebral cortex. Brain Res 67:169 –174. Krettek JE, Price JL (1977) Projections from the amygdaloid complex and adjacent olfactory structures to the entorhinal cortex and the subiculum in the rat and cat. J Comp Neurol 172:723–752. Laurberg S (1979) Commissural and intrinsic connections of the rat hippocampus. J Comp Neurol 184:685–708.

258

L. Uva and M. de Curtis / Neuroscience 130 (2005) 249 –258

Laurent G, Stopfer M, Friedrich RW, Rabinovich MI, Volkovskii A, Abarbanel HD (2001) Odor encoding as an active, dynamical process: experiments, computation, and theory. Annu Rev Neurosci 24:263–297. Liu P, Bilkey DK (1997) Parallel involvement of perirhinal and lateral entorhinal cortex in the polysynaptic activation of hippocampus by olfactory inputs. Hippocampus 7:296 –306. Luparello TJ (1967) Stereotaxic atlas of the forebrain of the guinea pig. Basel: S. Krager. Macrides F, Chorover SL (1972) Olfactory bulb units: activity correlated with inhalation cycles and odor quality. Science 175:84 – 87. Macrides F, Eichenbaum HB, Forbes BW (1982) Temporal relationship between sniffing and the limbic theta rhythm during odor discrimination reversal learning. J Neurosci 2:1705–1717. Margrie TW, Schaefer AT (2003) Theta oscillation coupled spike latencies yield computational vigour in a mammalian sensory system. J Physiol 546:363–374. Muhlethaler M, de Curtis M, Walton K, Llinás R (1993) The isolated and perfused brain of the guinea pig in vitro. Eur J Neurosci 5:915–926. Room P, Groenewegen HJ, Lohman AH (1984) Inputs from the olfactory bulb and olfactory cortex to the entorhinal cortex in the cat: I. Anatomical observations. Exp Brain Res 56:488 – 496. Sorensen KE (1985) Projection of the entorhinal area to the striatum, nucleus accumbens and cerebral cortex of the guinea pig. J Comp Neurol 238:308 –322. Steward O, Scoville SA (1976) The cells of origin of entorhinal afferents to the hippocampus and fascia dentata of the rat. J Comp Neurol 169:347–370.

Swanson LW, Wyss JM, Cowan WM (1978) An autoradiographic study of the organization of intrahippocampal association pathways in the rat. J Comp Neurol 181:681–716. Uva L, de Curtis M(2003) Ipsilateral and contralateral LOT stimulation activate the entorhinal-hippocampal in the isolated guinea pig brain. Society for Neuroscience, 33rd Annual Meeting, Abstract 719:6, 85. van Groen T, Wyss JM (1988) Species differences in hippocampal commissural connections: studies in rat, guinea pig, rabbit, and cat. J Comp Neurol 267:322–334. van Groen T, Wyss JM (1990) Extrinsic projections from area CA1 of the rat hippocampus: olfactory, cortical, subcortical, and bilateral hippocampal formation projections. J Comp Neurol 302:515– 528. Wilson DA (1998) Synaptic correlates of odor habituation in the rat anterior piriform cortex. J Neurophysiol 80:998 –1001. Wilson RC, Steward O (1978) Polysynaptic activation of the dentate gyrus of the hippocampal formation: an olfactory input via the lateral entorhinal cortex. Exp Brain Res 33:523–534. Witter MP (1993) Organization of the entorhinal-hippocampal system: a review of current anatomical data. Hippocampus 3:33– 44. Witter MP, Groenewegen HJ, Lopes da Silva FH, Lohman AH (1989) Functional organization of the extrinsic and intrinsic circuitry of the parahippocampal region. Prog Neurobiol 33:161–253. Wouterlood FG, Nederlof J (1983) Terminations of olfactory afferents on layer II and III neurons in the entorhinal area: degenerationGolgi-electron microscopic study in the rat. Neurosci Lett 36:105–110.

(Accepted 18 August 2004)