- Email: [email protected]
Electrophysiological responses of interfascicular neurons of the rat anterior commissure to activation from the anterior olfactory nucleus, medial frontal cortex, and posterior nucleus of the amygdala
Brain Research 982 (2003) 288–292 www.elsevier.com / locate / brainres
Short communication
Electrophysiological responses of interfascicular neurons of the rat anterior commissure to activation from the anterior olfactory nucleus, medial frontal cortex, and posterior nucleus of the amygdala a,b ´ ´ ´ a , Gema Martınez ´ a, Miguel Condes-Lara , Carlos Paz c , Javier Rodrıguez Jimenez a ´ Guadalupe Martınez-Lorenzana , Jorge Larriva-Sahd a , * a
´ , Universidad Nacional Autonoma ´ ´ de Mexico , Campus Department of Developmental Biology and Neurophysiology, Instituto de Neurobiologıa ´ , Qro., Mexico UNAM-UAQ , Juriquilla, Apartado Postal 1 -1141, Zona Centro, C.P. 76001 Queretaro b ´ , Mexico ´ , D.F., Mexico Instituto Nacional de Psiquiatrıa c ´ y Neurocirugıa ´ , Tlalpan, Mexico, D.F., Mexico Department of Neurophysiology and Sleep Research Laboratory, Instituto Nacional de Neurologıa Accepted 23 May 2003
Abstract Interfascicular neurons (IFNs) of the anterior commissure (AC) include short-axon and projection types which receive inputs from commissural collaterals. Therefore, it was proposed that IFNs may play a role in processing nerve impulses arising from the forebrain and delivered by these collaterals [Brain Res. 931 (2002) 81–91]. To determine possible inputs from the forebrain to IFNs we performed extracellular recordings of 25 neurons from anesthetized adult rats. Short-latency evoked potentials in IFNs were elicited by electrical stimulation of the anterior olfactory, posterior amygdaloid nuclei (PA), and medial frontal cortex. The IFN responses showed three distinct patterns, namely, a single action potential (AP) followed by what appear to be spontaneous discharge; a burst of high-frequency APs, and a single AP followed by a period devoid of APs. The latter response which was elicited by stimulation of the PA, may be explained by an intervening inhibitory interneuron, perhaps GABAergic in nature. Finally, IFNs seem not to project back to any of these three forebrain areas, as we failed to demonstrate antidromic activation. 2003 Elsevier B.V. All rights reserved. Theme: Development and regeneration Topic: Cerebral cortex and limbic system Keywords: Neuron; Connectivity; Forebrain; Electrophysiology
Commissural fiber systems are considered to be an anatomical link between homonymous areas of the forebrain. In euplacental mammals these commissural fibers are grouped into three bundles: corpus callosum, forix, and ´ y anterior commissure (AC). Early last century, Ramon Cajal found that the rodent AC incorporates two distinct axonal bundles, a rostral, olfactory or bulbar part (rAC) and a caudal, temporal or sphenoidal part (cAC) [21]. In addition, the stria terminalis contributes a robust bundle which is incorporated into the cAC [5]. Subsequently, a number of studies using neuronal and tract tracing techniques have confirmed and extended the location of *Corresponding author. Fax: 15-442-234-0344. E-mail address: [email protected] (J. Larriva-Sahd). 0006-8993 / 03 / $ – see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016 / S0006-8993(03)03054-3
neurons projecting through the AC [1,4,6,9–13,16]. Despite the presence of neurons among the AC axons described in the mouse 20 years ago [24], the possibility that these neurons may be a site of integration of commissural nerve impulses had not been systematically explored. Recently, we have shown the presence of both short axon and projection neurons which, due to their location among the axonal fascicles of the AC, were termed interfascicular neurons (IFNs) [14]. Furthermore, IFNs are found in the AC of rats, gerbils, hamsters [15], and humans (J. LarrivaSahd, unpublished observations). Additionally, IFNs and their processes appear to be confined to the AC domain, receiving inputs from commissural collaterals [14,16]. It was latter demonstrated that IFNs fulfill the classical electrophysiological features of a neuron, as these cells
´ et al. / Brain Research 982 (2003) 288–292 M. Condes-Lara
exhibit spontaneous activity (i.e., action potentials; APs) whose frequency is modified by local application of electrical stimuli [3]. These observations, led us to suppose that IFNs represent a site of integration of commissural nerve impulses [3,14]. To explore possible inputs from the forebrain to IFNs, 12 Sprague–Dawley male rats of 12 weeks of age were used. The animals were kept under standard vivarium care, with free access to food and water. All animals were handled and killed in compliance with policies of ethical animal care of our institute. Each animal to be implanted was anesthetized with 3% halotane, 60% nitrous oxide, and 30% oxygen. Then, the animal was tracheotomized and held in a stereotaxic apparatus. Throughout the experiment, the electrocardiographic (ECG) activity was recorded, and the body temperature was kept constant at 38 8C by a hot pad. The skull was drilled and two silver electrodes were implanted for the electroencephalographic (EEG) recording. For stimulation, separated electrodes were placed in the anterior olfactory nucleus (AON) [21], medial prefrontal cortex, or in the posterior nucleus of the amygdala (PA) [1]. These electrodes consisted of two twisted stainless steel wires (0.008 in. diameter; 1 in.52.54 cm) insulated with Teflon (0.011 in. diameter) each one, with bared tips, which were separated by 0.1 mm distance. Each electrode was plugged to a connector, allowing electrical stimulation at each of these structures [19]. Another trephine was performed to record from the AC. After the surgical procedures were complete, the concentration of halotane was lowered to 1.5% and an injection of Pancuronium Bromide (10 mg / kg weight) was given to provoke further hypotonia. The ECG, EEG, and body temperature were monitored throughout the experiment. The recording unit consisted of glass pipettes filled with a 1 M aqueous solution of KCl and 4% pontamine blue. The electrode resistance ranged from 14 to 20 MV. Once the electrode reached the AC coordinates, it was guided in the dorsoventral direction with a micro-driver, until APs were recoded by an amplifier (WPI Dual Microprobe System). Then, cathodic current was applied to induce extracellular depolarization and to identify IFNs. A period of 3 min of spontaneous activity was recorded for each presumptive IFN, and this was considered as control or spontaneous neuron activity. To determine a possible antidromic activation, a device consisting of a sensor placed between the source of output signal and the trigger stimulator, allowing an incoming amplified nerve impulse to trigger the stimulator with a variable delay. This permitted one to apply a stimulus with a delay ranging from 3 to 20 ms. as determined from the latency of the short-latency response plus an estimated refractory period. This allowed a collision test [2,8] to be performed in every cell studied. Criteria used to define an antidromic response were: a uniform latency period, a collision with a spontaneous spike, and the ability of the cell to respond following a high frequency electrical stimulation. Upon induction of an
289
antidromic response, each cell was stimulated by trials of single square pulses (1 ms, ,100 mA, at 0.2 Hz) to search for ortodromic responses. After the recording session pontamine blue was delivered by iontophoresis with a continuous current source connected to the recording electrode (cathodic current 10–15 mA for 30 min). The position of the stimulation electrodes was determined by passing a 100 mA constant current for 5 s for each polarity. Finally, the animal was perfused through the left ventricle with 10% formalin dissolved in saline, the brain removed, frozen, and serially sectioned at 40 mm with a sliding microtome. The sections were counter-stained with safranin allowing the location of both neurons and electrodes to be determined. Only cells displaying a low-frequency baseline response (see above) and identified within the domain of the AC were considered to be IFNs (Fig. 1A). From a total of 99 neurons studied, 38 cells from eight animals were located in the AC as confirmed by light microscopic inspection (Fig. 1A). Using the antidromic stimulation protocol described here, we tested all 99 cells with the collision test, but none of them was activated by stimuli applied separately to the infralimbic cortex (IL), AON, or PA. From the 38 identified IFNs, 25 exhibited responses to electrical activation and 13 did not respond. Thus, only 25 IFNs with histological confirmation that the recording electrodes were within the AC, spontaneous activity of presumptive IFNs [3], and responses to electrical stimulation (see below) will be considered. Of these 25 neurons, two displayed short-latency responses to electrical activation from the AON, 11 from the IL, and 12 from the PA (Fig. 1B). Among the short-latency responding neurons three different response patterns were recorded: while one type of short-latency responding neuron show a single AP, other cells responded with a burst of high-frequency APs (Fig. 1B). This second pattern of firing was elicited in one cell after IL stimulation and in three cells after PA stimulation (Fig. 2A). Still another response, elicited by stimulation of the PA, was nearly the same as that recorded in the short-latency responding IFNs; however, after an initial single AP, this was characteristically followed by a suppression of the basal frequency discharge (Fig. 2B). This last pattern of neuronal discharge was observed in five of the 12 cells responding to PA stimulation. The injection of pontamine blue by inotophoresis allowed identification of both the trajectory of the recording electrode as well as the actual position of the recorded cells (Fig. 1A). The major contribution of the present study is the demonstration that the spontaneous activity of native neurons of the AC is influenced by nerve impulses generated in the forebrain and presumed to be traveling toward the opposite hemisphere. Although previous observations using tract-tracing techniques demonstrated that the IL [4], AON [13], and PA [1,6] send axons to the AC, it was not known if the axons arising from these areas
290
´ et al. / Brain Research 982 (2003) 288–292 M. Condes-Lara
Fig. 1. (A) Photomontage of the caudal part of the anterior commissure in a coronal section. The inset shows at higher magnification the site of recording evidenced by a pontamine blue-stained area. Notice the granular appearance of the dye inside the processes of an interfascicular neuron (arrows). The site of injection (1) is paler than the adjacent area. (B) Short latency single spike responses of interfascicular neurons. The histograms represent the latency periods of interfascicular neuron responses to stimuli in the anterior olfactory nucleus (AON); infralimbic (i.e., medial prefrontal) cortex (IL), and posterior nucleus of the amygdala (PA). Each bar represents the mean latency from 10 responses for each cell. In the right side there are two typical recordings of interfascicular neurons for each stimulated structure. Vertical bar5means and standard error.
terminate in IFNs or simply pass through the AC. The modification of the spontaneous activity of IFNs in response to stimuli in any of these three areas favors the existence of a direct input to IFNs from these three sites. While a polysynaptic interaction cannot be excluded, the following observations support the existence of a monosynaptic projection. First, the relatively short latency periods elapsing between stimulation and action potentials recoded in IFNs (Fig. 1B), match those documented in the so-called ‘early response’ that is recorded at the surface of
the cerebral cortex in response to an electrical stimulus applied to the opposite hemisphere [22], and is abolished by interruption of the AC [18,22]. Second, injections of tract-tracing substances confined to either PA (Phaseolus vulgaris leucoagglutinin) [1,6], AON [horseradish peroxidase (HRP)] [13], or IF (wheat germ agglutinin–horseradish peroxidase) [4] visualized axons arising from either nuclear or cortical projection neurons. In fact, in these studies the relatively short periods of time between the injection of tracers and sacrifice (48 to 72 h) precludes the
Fig. 2. (A) Responses recorded in one identified interfascicular neuron by single pulse stimulations from the posterior nucleus of the amygdala. The left side shows the spontaneous activity. Following stimulation (doted line) there is a short latency period followed by a burst of high-frequency action potentials. (B) Responses of an interfascicular neuron to stimulation of the posterior nucleus of the amygdala. a, shows a single recording; b, corresponds to a raster display of 30 responses; c, post-stimulus time histogram of the same cell. Note that after the first action potential there is a period of inhibition followed by a gradual recovery of action potentials.
´ et al. / Brain Research 982 (2003) 288–292 M. Condes-Lara
labeling of second order neurons (i.e., transynaptical labeling) [17]. Finally, in the newborn [11] and adult rat [9,12], HRP applied to the AC reaches the projection neurons of the cerebral cortex by retrograde transport. The information outlined here suggests that the activity of IFNs is modulated by nerve impulses arising from the forebrain, conduced by commissural axons, and delivered to IFNs by commissural axon collaterals [4,14]. The identification of a set of IFNs responding to stimulation of the posterior amygdala with a single spike followed by a period of quiescence mimics the behavior of neuronal systems with inhibitory recurrences. In fact, a very similar response has been documented in a variety of forebrain and spinal cord sites. For instance, electrical stimuli that activate some projection neurons of the ventral spinal cord [7], ventrolateral [20] and lateral geniculated [23] nuclei of the thalamus yield a single AP which may be followed by an interval of cell discharges. It is assumed that this kind of response is mediated by an axon recurrent provided by the projection neuron that activates a GABAergic inhibitory interneuron, which, in turn, blocks the discharge of the former cell. The striking similarity of our recordings to those of the aforementioned examples, together with the presence of an IFN population that expresses mRNA of the enzyme glutamic acid decarboxylase (J. Larriva-Sahd, unpublished observation), suggests that activation of these cells by axon collaterals of projection IFNs may in turn trigger GABAergic-mediated inhibitory presynaptic potentials, possibly a short-axon IFN, accounting for the inhibitory response described here. The observation of such an inhibitory response, coupled with the presence of short-axon neurons within the AC [16], suggests a local mechanism of modulation within the domain of the AC. The possible site(s) of projection of IFN axons remains as an important, unresolved issue that is important to clarify the possible functional involvement of the IFN circuitry. From the recordings obtained so far, it seems unlikely that projection IFNs send axons back to the PA, AON and IL, as we failed to provoke antidromic collision [8]. Whichever the projection site(s) of IFNs is (are), we may conclude that these cells receive and integrate nerve impulses delivered by commissural collaterals. In conclusion, the present study adds fundamental aspects to the understanding of IFN circuitry. First, action potentials may be evoked in IFNs by stimuli applied to three widespread cortical and olfactory areas whose axons deliver presynaptic excitatory potentials (PSEPs) via commissural collaterals, the only IFN input documented thus far [3,14,15]. Second, IFNs do not seem to project back to neither AON, IL, or PA. Third, the patterns of IFN firing to cortical and olfactory PSEPs yield three broad responses; among these, the one spike-silent type of response elicited by PA stimulation is perhaps mediated by an inhibitory interneuron, possibly GABAergic in nature.
291
Acknowledgements The authors thank Drs. Larry W. Swanson and Alan Watts for the use of their in-situ hybridization collection. We thank Drs. Carlos Valverde and Dorothy Pless for ´ Garcıa ´ Servın, ´ reviewing the manuscript, and Dr. Martın Head of the Vivarium for his valuable help. Supported by CONACyT, Grant V40286 to J.L.-S.
References [1] N.S. Canteras, R.B. Simerly, L.W. Swanson, Connections of the posterior nucleus of the amygdala, J. Comp. Neurol. 324 (1992) 143–179. ˜ ´ ´ ´ [2] M. Condes-Lara, I. Omana-Zapata, M. Leon-Olea, M. SanchezAlvarez, Dorsal raphe neuronal responses to thalmic centralis lateralis and medial prefrontal cortex electrical stimulation, Brain Res. 499 (1989) 141–144. ´ ´ ´ [3] M. Condes-Lara, G. Martınez-Cabrera, G. Martınez-Lorenzana, J. Larriva-Sahd, Electrophysiological evidence that a set of interfascicular cells of the rat anterior commissure are neurons, Neurosci. Lett. 323 (2002) 121–124. ´ ´ [4] M. Condes-Lara, E. Talavera-Cuevas, J. Larriva-Sahd, G. MartınezLorenzana, Different wheat germ agglutinin–horseradish peroxidase labeling in structures related to the development of amygdaline kindling in the rat, Neurosci. Lett. 299 (2001) 13–16. [5] J.S. de Olmos, W.R. Ingram, The projection field of the stria terminalis in the rat brain. An experimental study, J. Comp. Neurol. 146 (1972) 303–334. [6] H.-W. Dong, G.D. Petrovich, L.W. Swanson, Topographic of projections from amygdala to bed nuclei of the stria terminalis, Brain Res. Rev. 38 (2001) 192–246. [7] J.C. Eccles, P. Fatt, K. Koketsu, Cholinergic and inhibitory synapses in a pathway from motor axon collaterals to motoneurones, J. Physiol. (London) 126 (1954) 524–562. [8] J.H. Fuller, J.S. Schlag, Determination of antidromic excitation by the collision test: problems of interpretation, Brain Res. 112 (1976) 283–298. [9] E.M. Granger, R.B. Masterton, K.K. Glendenning, Origin of the interhemispheric fibers in acallosal opossum (with comparison to callosal origins in the rat), J. Comp. Neurol. 241 (1985) 82–98. ˜ ´ F. Escobar del Rey, G. Morrale-Escobar, G.M. [10] A. Guadano-Ferraz, Innocenti, P. Berbel, The development of the anterior commissure in normal and hypothyroid rats, Dev. Brain Res. 81 (1994) 293–308. [11] E. Hedin-Pereira, D. Uziel, R. Lent, Bicommissural neurons in the cerebral cortex of developing hamsters, Neuroreport 3 (1992) 873– 876. [12] J.A. Horel, D.J. Stelzner, Neocortical projections of the rat anterior commissure, Brain Res. 220 (1981) 1–12. [13] M.L. Jouandet, V. Hartensein, Basal telencephalic origins of the anterior commissure, Exp. Brain Res. 50 (1983) 183–192. ´ ´ [14] J. Larriva-Sahd, M. Condes-Lara, G. Martınez-Cabrera, A. VarelaEchevarria, Histological and ultrastructural characterization of interfascicular neurons in the rat anterior commissure, Brain Res. 931 (2002) 81–91. ´ [15] J. Larriva-Sahd, G. Martınez, R. Paredes, I. Huerta, A. Varela, Characterization of a novel interfascicular neuron and neuropil in the anterior commissure of four classes of rodentia: a rapid-Golgi and ultrastructural study, Abstr. Soc. Neurosci. 26 (2000) 835.9A. [16] A.H.M. Lohman, The anterior olfactory lobe of the guinea pig. A descriptive and experimental anatomical study, Acta Anat. 53 (1963) 9–109.
292
´ et al. / Brain Research 982 (2003) 288–292 M. Condes-Lara
[17] M. Mesulam, Tracing Neuronal Connections with Horseradish Peroxidase, Wiley, New York, 1982. [18] L.R. Nelson, R.A. Lende, Interhemispheric responses in the opossum, J. Neurophysiol. 28 (1965) 189–199. [19] G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, London, 1998. [20] D.P. Purpura, T.L. Frigyesi, J.G. McMurtry, T. Scarf, Synaptic mechanisms in thalamic regulation of cerebello–cortical projection activity, in: D.P. Purpura, M.D. Yahr (Eds.), The Thalamus, Columbia University Press, New York, 1966, pp. 153–172. ´ y Cajal (Ed.), [21] V´ıas nacidas de la corteza esfenoidal, in: S. Ramon
Textura del Sistema Nervioso Central del Hombre y los Vertebrados, ´ de Nicolas ´ Moya, 1904. Vol. 2, Imprenta y Librerıa [22] L.T. Rutledge, T.T. Kennedy, Brain stem and cortical interactions in the interhemispheric delayed response, Exp. Neurol. 4 (1961) 470– 483. [23] S.M. Sherman, Ch. Koch, Thalamus, in: G.M. Shepherd (Ed.), The Synaptic Organization of the Brain, Oxford University Press, New York, 1990, pp. 246–278. [24] R.R. Sturrock, Neurons in the mouse anterior commissure. A light microscopic, electron microscopic and autoradiographic study, J. Anat. 123 (1977) 751–762.
Short communication
Electrophysiological responses of interfascicular neurons of the rat anterior commissure to activation from the anterior olfactory nucleus, medial frontal cortex, and posterior nucleus of the amygdala a,b ´ ´ ´ a , Gema Martınez ´ a, Miguel Condes-Lara , Carlos Paz c , Javier Rodrıguez Jimenez a ´ Guadalupe Martınez-Lorenzana , Jorge Larriva-Sahd a , * a
´ , Universidad Nacional Autonoma ´ ´ de Mexico , Campus Department of Developmental Biology and Neurophysiology, Instituto de Neurobiologıa ´ , Qro., Mexico UNAM-UAQ , Juriquilla, Apartado Postal 1 -1141, Zona Centro, C.P. 76001 Queretaro b ´ , Mexico ´ , D.F., Mexico Instituto Nacional de Psiquiatrıa c ´ y Neurocirugıa ´ , Tlalpan, Mexico, D.F., Mexico Department of Neurophysiology and Sleep Research Laboratory, Instituto Nacional de Neurologıa Accepted 23 May 2003
Abstract Interfascicular neurons (IFNs) of the anterior commissure (AC) include short-axon and projection types which receive inputs from commissural collaterals. Therefore, it was proposed that IFNs may play a role in processing nerve impulses arising from the forebrain and delivered by these collaterals [Brain Res. 931 (2002) 81–91]. To determine possible inputs from the forebrain to IFNs we performed extracellular recordings of 25 neurons from anesthetized adult rats. Short-latency evoked potentials in IFNs were elicited by electrical stimulation of the anterior olfactory, posterior amygdaloid nuclei (PA), and medial frontal cortex. The IFN responses showed three distinct patterns, namely, a single action potential (AP) followed by what appear to be spontaneous discharge; a burst of high-frequency APs, and a single AP followed by a period devoid of APs. The latter response which was elicited by stimulation of the PA, may be explained by an intervening inhibitory interneuron, perhaps GABAergic in nature. Finally, IFNs seem not to project back to any of these three forebrain areas, as we failed to demonstrate antidromic activation. 2003 Elsevier B.V. All rights reserved. Theme: Development and regeneration Topic: Cerebral cortex and limbic system Keywords: Neuron; Connectivity; Forebrain; Electrophysiology
Commissural fiber systems are considered to be an anatomical link between homonymous areas of the forebrain. In euplacental mammals these commissural fibers are grouped into three bundles: corpus callosum, forix, and ´ y anterior commissure (AC). Early last century, Ramon Cajal found that the rodent AC incorporates two distinct axonal bundles, a rostral, olfactory or bulbar part (rAC) and a caudal, temporal or sphenoidal part (cAC) [21]. In addition, the stria terminalis contributes a robust bundle which is incorporated into the cAC [5]. Subsequently, a number of studies using neuronal and tract tracing techniques have confirmed and extended the location of *Corresponding author. Fax: 15-442-234-0344. E-mail address: [email protected] (J. Larriva-Sahd). 0006-8993 / 03 / $ – see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016 / S0006-8993(03)03054-3
neurons projecting through the AC [1,4,6,9–13,16]. Despite the presence of neurons among the AC axons described in the mouse 20 years ago [24], the possibility that these neurons may be a site of integration of commissural nerve impulses had not been systematically explored. Recently, we have shown the presence of both short axon and projection neurons which, due to their location among the axonal fascicles of the AC, were termed interfascicular neurons (IFNs) [14]. Furthermore, IFNs are found in the AC of rats, gerbils, hamsters [15], and humans (J. LarrivaSahd, unpublished observations). Additionally, IFNs and their processes appear to be confined to the AC domain, receiving inputs from commissural collaterals [14,16]. It was latter demonstrated that IFNs fulfill the classical electrophysiological features of a neuron, as these cells
´ et al. / Brain Research 982 (2003) 288–292 M. Condes-Lara
exhibit spontaneous activity (i.e., action potentials; APs) whose frequency is modified by local application of electrical stimuli [3]. These observations, led us to suppose that IFNs represent a site of integration of commissural nerve impulses [3,14]. To explore possible inputs from the forebrain to IFNs, 12 Sprague–Dawley male rats of 12 weeks of age were used. The animals were kept under standard vivarium care, with free access to food and water. All animals were handled and killed in compliance with policies of ethical animal care of our institute. Each animal to be implanted was anesthetized with 3% halotane, 60% nitrous oxide, and 30% oxygen. Then, the animal was tracheotomized and held in a stereotaxic apparatus. Throughout the experiment, the electrocardiographic (ECG) activity was recorded, and the body temperature was kept constant at 38 8C by a hot pad. The skull was drilled and two silver electrodes were implanted for the electroencephalographic (EEG) recording. For stimulation, separated electrodes were placed in the anterior olfactory nucleus (AON) [21], medial prefrontal cortex, or in the posterior nucleus of the amygdala (PA) [1]. These electrodes consisted of two twisted stainless steel wires (0.008 in. diameter; 1 in.52.54 cm) insulated with Teflon (0.011 in. diameter) each one, with bared tips, which were separated by 0.1 mm distance. Each electrode was plugged to a connector, allowing electrical stimulation at each of these structures [19]. Another trephine was performed to record from the AC. After the surgical procedures were complete, the concentration of halotane was lowered to 1.5% and an injection of Pancuronium Bromide (10 mg / kg weight) was given to provoke further hypotonia. The ECG, EEG, and body temperature were monitored throughout the experiment. The recording unit consisted of glass pipettes filled with a 1 M aqueous solution of KCl and 4% pontamine blue. The electrode resistance ranged from 14 to 20 MV. Once the electrode reached the AC coordinates, it was guided in the dorsoventral direction with a micro-driver, until APs were recoded by an amplifier (WPI Dual Microprobe System). Then, cathodic current was applied to induce extracellular depolarization and to identify IFNs. A period of 3 min of spontaneous activity was recorded for each presumptive IFN, and this was considered as control or spontaneous neuron activity. To determine a possible antidromic activation, a device consisting of a sensor placed between the source of output signal and the trigger stimulator, allowing an incoming amplified nerve impulse to trigger the stimulator with a variable delay. This permitted one to apply a stimulus with a delay ranging from 3 to 20 ms. as determined from the latency of the short-latency response plus an estimated refractory period. This allowed a collision test [2,8] to be performed in every cell studied. Criteria used to define an antidromic response were: a uniform latency period, a collision with a spontaneous spike, and the ability of the cell to respond following a high frequency electrical stimulation. Upon induction of an
289
antidromic response, each cell was stimulated by trials of single square pulses (1 ms, ,100 mA, at 0.2 Hz) to search for ortodromic responses. After the recording session pontamine blue was delivered by iontophoresis with a continuous current source connected to the recording electrode (cathodic current 10–15 mA for 30 min). The position of the stimulation electrodes was determined by passing a 100 mA constant current for 5 s for each polarity. Finally, the animal was perfused through the left ventricle with 10% formalin dissolved in saline, the brain removed, frozen, and serially sectioned at 40 mm with a sliding microtome. The sections were counter-stained with safranin allowing the location of both neurons and electrodes to be determined. Only cells displaying a low-frequency baseline response (see above) and identified within the domain of the AC were considered to be IFNs (Fig. 1A). From a total of 99 neurons studied, 38 cells from eight animals were located in the AC as confirmed by light microscopic inspection (Fig. 1A). Using the antidromic stimulation protocol described here, we tested all 99 cells with the collision test, but none of them was activated by stimuli applied separately to the infralimbic cortex (IL), AON, or PA. From the 38 identified IFNs, 25 exhibited responses to electrical activation and 13 did not respond. Thus, only 25 IFNs with histological confirmation that the recording electrodes were within the AC, spontaneous activity of presumptive IFNs [3], and responses to electrical stimulation (see below) will be considered. Of these 25 neurons, two displayed short-latency responses to electrical activation from the AON, 11 from the IL, and 12 from the PA (Fig. 1B). Among the short-latency responding neurons three different response patterns were recorded: while one type of short-latency responding neuron show a single AP, other cells responded with a burst of high-frequency APs (Fig. 1B). This second pattern of firing was elicited in one cell after IL stimulation and in three cells after PA stimulation (Fig. 2A). Still another response, elicited by stimulation of the PA, was nearly the same as that recorded in the short-latency responding IFNs; however, after an initial single AP, this was characteristically followed by a suppression of the basal frequency discharge (Fig. 2B). This last pattern of neuronal discharge was observed in five of the 12 cells responding to PA stimulation. The injection of pontamine blue by inotophoresis allowed identification of both the trajectory of the recording electrode as well as the actual position of the recorded cells (Fig. 1A). The major contribution of the present study is the demonstration that the spontaneous activity of native neurons of the AC is influenced by nerve impulses generated in the forebrain and presumed to be traveling toward the opposite hemisphere. Although previous observations using tract-tracing techniques demonstrated that the IL [4], AON [13], and PA [1,6] send axons to the AC, it was not known if the axons arising from these areas
290
´ et al. / Brain Research 982 (2003) 288–292 M. Condes-Lara
Fig. 1. (A) Photomontage of the caudal part of the anterior commissure in a coronal section. The inset shows at higher magnification the site of recording evidenced by a pontamine blue-stained area. Notice the granular appearance of the dye inside the processes of an interfascicular neuron (arrows). The site of injection (1) is paler than the adjacent area. (B) Short latency single spike responses of interfascicular neurons. The histograms represent the latency periods of interfascicular neuron responses to stimuli in the anterior olfactory nucleus (AON); infralimbic (i.e., medial prefrontal) cortex (IL), and posterior nucleus of the amygdala (PA). Each bar represents the mean latency from 10 responses for each cell. In the right side there are two typical recordings of interfascicular neurons for each stimulated structure. Vertical bar5means and standard error.
terminate in IFNs or simply pass through the AC. The modification of the spontaneous activity of IFNs in response to stimuli in any of these three areas favors the existence of a direct input to IFNs from these three sites. While a polysynaptic interaction cannot be excluded, the following observations support the existence of a monosynaptic projection. First, the relatively short latency periods elapsing between stimulation and action potentials recoded in IFNs (Fig. 1B), match those documented in the so-called ‘early response’ that is recorded at the surface of
the cerebral cortex in response to an electrical stimulus applied to the opposite hemisphere [22], and is abolished by interruption of the AC [18,22]. Second, injections of tract-tracing substances confined to either PA (Phaseolus vulgaris leucoagglutinin) [1,6], AON [horseradish peroxidase (HRP)] [13], or IF (wheat germ agglutinin–horseradish peroxidase) [4] visualized axons arising from either nuclear or cortical projection neurons. In fact, in these studies the relatively short periods of time between the injection of tracers and sacrifice (48 to 72 h) precludes the
Fig. 2. (A) Responses recorded in one identified interfascicular neuron by single pulse stimulations from the posterior nucleus of the amygdala. The left side shows the spontaneous activity. Following stimulation (doted line) there is a short latency period followed by a burst of high-frequency action potentials. (B) Responses of an interfascicular neuron to stimulation of the posterior nucleus of the amygdala. a, shows a single recording; b, corresponds to a raster display of 30 responses; c, post-stimulus time histogram of the same cell. Note that after the first action potential there is a period of inhibition followed by a gradual recovery of action potentials.
´ et al. / Brain Research 982 (2003) 288–292 M. Condes-Lara
labeling of second order neurons (i.e., transynaptical labeling) [17]. Finally, in the newborn [11] and adult rat [9,12], HRP applied to the AC reaches the projection neurons of the cerebral cortex by retrograde transport. The information outlined here suggests that the activity of IFNs is modulated by nerve impulses arising from the forebrain, conduced by commissural axons, and delivered to IFNs by commissural axon collaterals [4,14]. The identification of a set of IFNs responding to stimulation of the posterior amygdala with a single spike followed by a period of quiescence mimics the behavior of neuronal systems with inhibitory recurrences. In fact, a very similar response has been documented in a variety of forebrain and spinal cord sites. For instance, electrical stimuli that activate some projection neurons of the ventral spinal cord [7], ventrolateral [20] and lateral geniculated [23] nuclei of the thalamus yield a single AP which may be followed by an interval of cell discharges. It is assumed that this kind of response is mediated by an axon recurrent provided by the projection neuron that activates a GABAergic inhibitory interneuron, which, in turn, blocks the discharge of the former cell. The striking similarity of our recordings to those of the aforementioned examples, together with the presence of an IFN population that expresses mRNA of the enzyme glutamic acid decarboxylase (J. Larriva-Sahd, unpublished observation), suggests that activation of these cells by axon collaterals of projection IFNs may in turn trigger GABAergic-mediated inhibitory presynaptic potentials, possibly a short-axon IFN, accounting for the inhibitory response described here. The observation of such an inhibitory response, coupled with the presence of short-axon neurons within the AC [16], suggests a local mechanism of modulation within the domain of the AC. The possible site(s) of projection of IFN axons remains as an important, unresolved issue that is important to clarify the possible functional involvement of the IFN circuitry. From the recordings obtained so far, it seems unlikely that projection IFNs send axons back to the PA, AON and IL, as we failed to provoke antidromic collision [8]. Whichever the projection site(s) of IFNs is (are), we may conclude that these cells receive and integrate nerve impulses delivered by commissural collaterals. In conclusion, the present study adds fundamental aspects to the understanding of IFN circuitry. First, action potentials may be evoked in IFNs by stimuli applied to three widespread cortical and olfactory areas whose axons deliver presynaptic excitatory potentials (PSEPs) via commissural collaterals, the only IFN input documented thus far [3,14,15]. Second, IFNs do not seem to project back to neither AON, IL, or PA. Third, the patterns of IFN firing to cortical and olfactory PSEPs yield three broad responses; among these, the one spike-silent type of response elicited by PA stimulation is perhaps mediated by an inhibitory interneuron, possibly GABAergic in nature.
291
Acknowledgements The authors thank Drs. Larry W. Swanson and Alan Watts for the use of their in-situ hybridization collection. We thank Drs. Carlos Valverde and Dorothy Pless for ´ Garcıa ´ Servın, ´ reviewing the manuscript, and Dr. Martın Head of the Vivarium for his valuable help. Supported by CONACyT, Grant V40286 to J.L.-S.
References [1] N.S. Canteras, R.B. Simerly, L.W. Swanson, Connections of the posterior nucleus of the amygdala, J. Comp. Neurol. 324 (1992) 143–179. ˜ ´ ´ ´ [2] M. Condes-Lara, I. Omana-Zapata, M. Leon-Olea, M. SanchezAlvarez, Dorsal raphe neuronal responses to thalmic centralis lateralis and medial prefrontal cortex electrical stimulation, Brain Res. 499 (1989) 141–144. ´ ´ ´ [3] M. Condes-Lara, G. Martınez-Cabrera, G. Martınez-Lorenzana, J. Larriva-Sahd, Electrophysiological evidence that a set of interfascicular cells of the rat anterior commissure are neurons, Neurosci. Lett. 323 (2002) 121–124. ´ ´ [4] M. Condes-Lara, E. Talavera-Cuevas, J. Larriva-Sahd, G. MartınezLorenzana, Different wheat germ agglutinin–horseradish peroxidase labeling in structures related to the development of amygdaline kindling in the rat, Neurosci. Lett. 299 (2001) 13–16. [5] J.S. de Olmos, W.R. Ingram, The projection field of the stria terminalis in the rat brain. An experimental study, J. Comp. Neurol. 146 (1972) 303–334. [6] H.-W. Dong, G.D. Petrovich, L.W. Swanson, Topographic of projections from amygdala to bed nuclei of the stria terminalis, Brain Res. Rev. 38 (2001) 192–246. [7] J.C. Eccles, P. Fatt, K. Koketsu, Cholinergic and inhibitory synapses in a pathway from motor axon collaterals to motoneurones, J. Physiol. (London) 126 (1954) 524–562. [8] J.H. Fuller, J.S. Schlag, Determination of antidromic excitation by the collision test: problems of interpretation, Brain Res. 112 (1976) 283–298. [9] E.M. Granger, R.B. Masterton, K.K. Glendenning, Origin of the interhemispheric fibers in acallosal opossum (with comparison to callosal origins in the rat), J. Comp. Neurol. 241 (1985) 82–98. ˜ ´ F. Escobar del Rey, G. Morrale-Escobar, G.M. [10] A. Guadano-Ferraz, Innocenti, P. Berbel, The development of the anterior commissure in normal and hypothyroid rats, Dev. Brain Res. 81 (1994) 293–308. [11] E. Hedin-Pereira, D. Uziel, R. Lent, Bicommissural neurons in the cerebral cortex of developing hamsters, Neuroreport 3 (1992) 873– 876. [12] J.A. Horel, D.J. Stelzner, Neocortical projections of the rat anterior commissure, Brain Res. 220 (1981) 1–12. [13] M.L. Jouandet, V. Hartensein, Basal telencephalic origins of the anterior commissure, Exp. Brain Res. 50 (1983) 183–192. ´ ´ [14] J. Larriva-Sahd, M. Condes-Lara, G. Martınez-Cabrera, A. VarelaEchevarria, Histological and ultrastructural characterization of interfascicular neurons in the rat anterior commissure, Brain Res. 931 (2002) 81–91. ´ [15] J. Larriva-Sahd, G. Martınez, R. Paredes, I. Huerta, A. Varela, Characterization of a novel interfascicular neuron and neuropil in the anterior commissure of four classes of rodentia: a rapid-Golgi and ultrastructural study, Abstr. Soc. Neurosci. 26 (2000) 835.9A. [16] A.H.M. Lohman, The anterior olfactory lobe of the guinea pig. A descriptive and experimental anatomical study, Acta Anat. 53 (1963) 9–109.
292
´ et al. / Brain Research 982 (2003) 288–292 M. Condes-Lara
[17] M. Mesulam, Tracing Neuronal Connections with Horseradish Peroxidase, Wiley, New York, 1982. [18] L.R. Nelson, R.A. Lende, Interhemispheric responses in the opossum, J. Neurophysiol. 28 (1965) 189–199. [19] G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, London, 1998. [20] D.P. Purpura, T.L. Frigyesi, J.G. McMurtry, T. Scarf, Synaptic mechanisms in thalamic regulation of cerebello–cortical projection activity, in: D.P. Purpura, M.D. Yahr (Eds.), The Thalamus, Columbia University Press, New York, 1966, pp. 153–172. ´ y Cajal (Ed.), [21] V´ıas nacidas de la corteza esfenoidal, in: S. Ramon
Textura del Sistema Nervioso Central del Hombre y los Vertebrados, ´ de Nicolas ´ Moya, 1904. Vol. 2, Imprenta y Librerıa [22] L.T. Rutledge, T.T. Kennedy, Brain stem and cortical interactions in the interhemispheric delayed response, Exp. Neurol. 4 (1961) 470– 483. [23] S.M. Sherman, Ch. Koch, Thalamus, in: G.M. Shepherd (Ed.), The Synaptic Organization of the Brain, Oxford University Press, New York, 1990, pp. 246–278. [24] R.R. Sturrock, Neurons in the mouse anterior commissure. A light microscopic, electron microscopic and autoradiographic study, J. Anat. 123 (1977) 751–762.