Download Sexual and seasonal plasticity in the emission of social electric
Document related concepts
no text concepts found
Transcript
Journal of Physiology - Paris 102 (2008) 272–278 Contents lists available at ScienceDirect Journal of Physiology - Paris j o u r n a l h o m e p a g e : w w w . e l s e v i e r. c o m / l o c a t e / j p h y s p a r i s Sexual and seasonal plasticity in the emission of social electric signals. Behavioral approach and neural bases Ana Silva a,b,*, Laura Quintana a, Rossana Perrone a, Felipe Sierra a,c a Departamento de Neurofisiología, Instituto de Investigaciones Biológicas Clemente Estable, Avda Italia 3318, 11600 Montevideo, Uruguay Laborat orio de Neurociencias, Facultad de Ciencias, Iguá 4225, 11400 Montevideo, Uruguay c Unidad Asociada Neurofisiología-IIBCE, Facultad de Ciencias, Avda Italia 3318, 11600 Montevideo, Uruguay b a r t i c l e i n f o Keywords: Electric fish Pacemaker nucleus CNS sexual dimorphism Glutamate CNS seasonal plasticity Neuroethology a b s t r a c t Behavior in electric fish includes modul ations of a stereotyped electric organ discharge (EOD) in addition to locomotor displays. Gymnotiformes can modul ate the EOD rate to produce signals that participate in different behaviors. We studied the reproductive behavior of Brachyhypopomus pinnicaudatus both in the wild and laboratory settings. During the breeding season, fish produce sexually dimorphic social electric signals (SES): males emit three types of chirps (distinguished by their duration and internal structure), and accelerations, whereas females interrupt their EOD. Since these SES imply EOD frequency modula tions, the pacemaker nucleus (PN) is involved in their generation and constitutes the main target organ to explore seasonal and sexual plasticity of the CNS. The PN has two types of neurons, pacemakers and relays, which receive modulatory inputs from pre-pacemaker structures. These neurons show an aniso tropic rostro-caudal and dorso-ventral distribution that is paralleled by different field potential wave forms in distinct portions of the PN. In vivo glutamate injections in different areas of the PN provoke different kinds of EOD rate modulations. Ventral injections produce chirp-like responses in breeding males and EOD interruptions in breeding females, whereas dorsal injections provoke EOD frequency rises in both sexes. In the non-breeding season, males and females respond with interruptions when stimu lated ventrally and frequency rises when injected dorsally. Our results show that changes of glutamate effects in the PN could explain the seasonal and sexual differences in the generation of SES. By means of behavioral recordings both in the wild and in laboratory settings, and by electrophysiological and phar macological experiments, we have identified sexual and seasonal plasticity of the CNS and explored its underlying mechanisms. © 2008 Elsevier Ltd. All rights reserved. 1. Introduction One of the major goals of Neuroscience is to unravel the bases of behavior, i.e., to understand how the brain receives and inte grates diverse information (environmental cues, social modula tions, and multiple changes of physiological states) to give rise to complex behaviors. This purpose requires the application of high technology for an intimate combination of behavioral and anatomo-functional studies in performing animals; and thus con tinues to be on the wish list of most behavioral scientists. Experi mental models provided by nature or artificially developed have, nonetheless, allowed important contributions, which are slowly paving the way to the ultimate understanding of behavior. Electric fish are advantageous models for the study of behav ior since any given behavior (agonistic, reproductive, parental care, etc.) includes modulations of the electric discharge (easily * Corresponding author. Address: Departamento de Neurofisiología, Instituto de Investigaciones Biológicas Clemente Estable, Avda Italia 3318, 11600 Montevideo, Uruguay. Tel.: +598 2 4875532; fax: +598 2 4875548. E-mail address: asilva@iibce.edu.uy (A. Silva). 0928-4257/$ - see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jphysparis.2008.10.016 recorded and reliably quantified) in addition to locomotor displays (Bass and Zakon, 2005; Rose, 2004). The electromotor system that governs the production of the electric organ discharge (EOD) is well known and involves a few groups of neurons devoted to the generation of this electromotor output (Bennett, 1971; Dye and Meyer, 1986). The EOD carries information with communicative social value signaling species and physiological state throughout lifetime. During breeding, more complex and precise information is encoded in the EOD: several species of Gymnotiformes show sexually dimorphic EODs used for mate recognition and selection (Hagedorn and Heiligenberg, 1985; Hopkins, 1974a,b; Hopkins et al., 1990; Kramer and Otto, 1988; Zakon and Dunlap, 1999); and individuals can transiently modulate EOD rate and waveform to produce social signals that participate in courtship and spawning (Bastian et al., 2001; Hagedorn, 1988; Hagedorn and Heiligenberg, 1985; Hopkins, 1974a,b; Kawasaki and Heiligenberg, 1989; Silva, 2002; Zupanc, 2002). On the other hand, the cyclic nature of repro ductive behavior, particularly in Gymnotiformes from the temper ate zone, has been demonstrated by the assessment of electrical cues, some of which are also sex-dependent (Quintana et al., 2004; A. Silva et al. / Journal of Physiology - Paris 102 (2008) 272–278 Silva, 2002; Silva et al., 1999, 2002, 2003). On the whole, the elec tromotor system provides an outstanding model for the explora tion of the underlying neural circuits and neuroendocrinological features of behavioral plasticity. Our study closely integrates two approaches: (a) the behav ioral description of courtship electrical displays in the pulse-type weakly electric fish, Brachyhypopomus pinnicaudatus from the temperate zone; and (b) the electrophysiological analysis of the encephalic pacemaker nucleus (PN) that drives those displays. We therefore transit from behavior to the activity of about 100 neu rons in the medulla and provide new angles on understanding two current key issues: seasonal plasticity and sexual dimorphism of the CNS. 273 Chirp A Acceleration 50 ms 50 ms Interruption Chirp B 2. Social electric signals (SES) in the reproductive behavior of B. pinnicaudat us Gymnotiformes are nocturnal fish that live in freshwaters in a wide extension of the American continent (Kirschbaum, 1995). They show cyclic reproduction in the tropics, where the onset of breeding is triggered by changes in water conductivity due to the alternation of rainy and dry seasons (Kirschbaum, 1979, 1995); and also in the temperate region, where onset responds to sea sonal changes of water temperature and probably of photoperiod (Quintana et al., 2004; Silva et al., 2003). The precise temporal synchronization for a successful reproduction via external fertil ization is achieved by the use of the electric channel of commu nication. Therefore, courtship signaling in this particular group of fish includes distinctive and conspicuous electric displays. Several studies have elegantly demonstrated the role of electric signals in reproductive behavior in Gymnotiformes (Bastian et al., 2001; Hagedorn, 1988; Hagedorn and Heiligenberg, 1985; Hopkins, 1974a,b; Kawasaki and Heiligenberg, 1989; Zupanc, 2002). We study the reproductive behavior of the pulse-type gym notiform, B. pinnicaudat us (Hopkins, 1991), at the southern boundary of its continental distribution (Uruguay, 30–35°S) both in the wild and in laboratory settings. In the temperate region, B. pinnicaudatus breeds during late spring and summer (Novem ber–January) in association to high mean water temperature and asymmetric photoperiod (around 14 h L/10 h D) (Quintana et al., 2004; Silva et al., 2003). B. pinnicaudatus exhibits morphological and electrophysiol ogical sexual dimorphism during the breed ing season. The male is larger than the female, exhibits a broader caudal filam ent, and a longer EOD (Caputi et al., 1998; Hopkins et al., 1990; Silva et al., 1999). Individuals of B. pinnicaudatus aggre gate in dense breeding colonies in the wild (up to 20 individuals per m2) with sex ratios of adults highly skewed toward females (average 3:1, Miranda et al., 2008). The extreme sexual dimor phism exhibited by this species and the biased sex ratio support the hypothesis that B. pinnicaudatus is polygyn ous, whereas the differential use of space by males and females is consistent with an exploded lek breeding system (Miranda et al., 2008). During breeding, male–female dyads of B. pinnicaudatus pro duce social electric signals (SES) in addition to the regular emis sion of their biphasic EOD. Three different SES are recognized in reproductive context: chirps, accelerat ions, and interruptions, shown with representative examples in Fig. 1. The emission of SES always occurs during a characteristic nocturnal increase of EOD basal rate interpreted as a prerequisite for courtship displays (Silva et al., 2007a). Although most of the SES recorded occur dur ing nighttime, some of them have been observed during daytime (especially around the artificial sunset and sunrise). The SES are sexually dimorphic: males emit chirps and produce accelerations, whereas females interrupt their EOD (Fig. 1). In B. pinnicaudatus chirps can be defined as transient and brief changes in discharge pattern (ranging from 25 to 260 ms in dura 50 ms 50 ms Chirp & Interruption Chirp C 50 ms 50 ms Fig. 1. SES produced by male–female dyads of Brachyhypopomus pinnicaudatus during breeding. Dyads of freely moving fish were recorded in a recording station that mimics the natural habitat and allowed us to record electric signals simulta neously with visualization of locomotor displays. As described elsewhere (Silva et al., 2007a), the EODs of the dyad (female EODs are indicated by dots, and identified in an expanded time scale because they are shorter than male EODs) were recorded from two pairs of fixed electrodes attached to the tank walls. Representative recordings of the different types of SES: three types of chirps (A, B, and C) emitted by males, accelerations also emitted by males, and interruptions of the EOD pro duced by females. Voltage calib ration bars are not included since EOD amplitude depends on the relative position of freely moving fish with respect to the recording electrodes. tion) characterized by an increase in EOD rate (with intrachirp EOD rate ranging from 170 to 1000 Hz), decrease in EOD ampli tude (ranging from 0% to 80% of basal EOD amplitude), and dis tortion of EOD waveform (Kawasaki and Heiligenberg, 1989; Silva et al., 2007b). In contrast to chirps, accelerations last longer, show a smaller increase of EOD rate, and minimum changes of the EOD waveform. We propose that EOD interruptions should be consid ered as SES even though they constitute a cease in the emission of an active and otherwise permanent signal. The communica tive value of the interruption is strongly suggested because only females interrupt their EOD in male–female interactions, most interruptions occur when the dyad is close together (less than 5 cm between partners), and many of them are temporally correlated to male chirping (Fig. 1, Macadar and Silva, 2007). There are three types of chirps distinguished by their duration and internal structure produced by males in male–female interac tions. This suggests a more complex and sophisticated communi cative value for this kind of signal. Fig. 1 shows electric recordings 274 A. Silva et al. / Journal of Physiology - Paris 102 (2008) 272–278 of different examples of chirp types: type A chirps are long twophase signals (with an initial phase of maximal frequency increase and minimal amplitude, and a second phase with spindle-like amplitude modulations); type B chirps are short increases of EOD rate with decremental EOD amplitude; and type C chirps look like a type B chirp followed by its mirror image. Three selected param eters (chirp duration, time to minimal intra-chirp EOD amplitude, and symmetry index; see Fig. 2 legend for details) were enough to Table 1 Classification matrix showing percentages of correct classific ation of chirp types A, B, and C in clusters 1, 2, and 3, respectively; and the accuracy of clusters 1, 2, and 3 in representing chirp types A, B, and C, respectively. Cluster Type A chirp Type B chirp Type C chirp Accuracy (%) 1 2 3 Correct classification 44 7 16 65.7% 0 52 3 94.5% 0 5 49 90.74% 100 81.25 72.06 Total 67 55 54 1.0 Symmetry inde x 0.8 0.6 0.4 0.2 300 250 200 Du 150 rat ion 100 (m 50 s) 0 0 20 e to Tim 40 60 80 0.0 100 e itud l mp la ima ) (ms min 3 Chirp A Chirp B Chirp C 2 Factor 2 1 0 -1 -2 -3 -3 -2 -1 0 1 Factor 1 2 3 Fig. 2. Discrimination between different types of male chirps. Three parameters were used to achieve the characterization of male chirps during breeding: (i) chirp duration (measured from first intra-chirp EOD to the next basal EOD); (ii) time to minimal amplitude of intra-chirp EOD (in percentage of chirp duration); and (iii) symmetry index (defined as the ratio of amplitudes between the last intra-chirp EOD and the first intra-chirp EOD). (a) 3D representation of chirp types classified by a trained observer: A (red), B (blue), and C (green) showing that chirps can be acceptably discriminated by these parameters. (b) Plot of factor scores of the cases after principal components analysis (PCA) with respect to Factor 1 and Factor 2. Correlations between factors and variables: Factor 1, chirp duration (0.896), time to minimal amplitude (0.796), symmetry index (¡0.916); Factor 2, chirp duration (0.317), time to minimal amplitude (¡0.604), symmetry index (¡0.214). The distri bution of cases in three clusters is represented as shaded areas (pink for cluster 1, light blue for cluster 2, and light green for cluster 3) and was compared with our visual classification into three types of chirps (represented as dots: red for type A; blue for type B; and green for type C). discriminate among chirp types as represented in Fig. 2a. More over, a principal components analysis (PCA) and ulterior k-means cluster analysis was performed using the same selected parame ters in a sample of 176 chirps (see Fig. 2 legend for details). This allowed us to identify three types of chirps (shaded areas in Fig. 2b, Table 1) in accordance with our original visual classific ation done by a trained observer (dots in Fig. 2b, Table 1). This procedure was carried out using raw parameters measured directly on elec trophysiological recordings; and though preliminary, it is strong enough to discriminate among chirp types. During breeding nights, hundreds of chirps are emitted in an irregular temporal pattern, usually in bouts, which are in turn grouped throughout the night. For example, in one breeding dyad (out of 10 analyzed), the male emitted 828 chirps: 380 type A chirps (45.9%), 248 type B chirps (29.95%), and 200 type C chirps (24.15%). Different types of chirps predominate at different moments of the night: type A chirps are mainly observed during the evening and early night, whereas types B and C prevail close to sunrise (unpub lished data). There are outstanding seasonal differences in the emission of SES. Male accelerations and female interruptions are never observed during the non-breeding season. The number of chirps emitted by males is more than 10-fold higher during the breed ing season. In breeding dyads, males produce over 500 chirps per night, whereas in non-breeding dyads, chirps are rare (Silva et al., 2007b). These unusual non-breeding chirps are longer and exhibit a lower intra-chirp EOD frequency than breeding chirps, which precludes their classification into any of the above-described cate gories of male breeding chirps. Although beyond the scope of this study, we cannot but discuss the meaning of SES as courtship signals. In most mating systems, the males display signals to attract possible mates while females choose among signalers (Bateman, 1948; Andersson, 1994). Male chirps can be interpreted as traditional courtship displays, speciesspecific signals that favor reproductive isolation, and as conspicuous displays to synchronize the actions of males and females towards mating. Differences in the number and types of chirps emitted by males allow us to hypothesize that male quality is related to chirp ing activity, and therefore important for sexual selection. Some sexual selection models envisage male signaling to be costly in order to be a reliable indicator of the quality of potential mates for choosy females (Grafen, 1990). It is not obvious how chirping activ ity correlates with male quality. Is chirping activity risky for the male? It is still a matter of debate how a conspecific or an electro sensory predator interprets chirping in terms of the electric image evoked. It may be interpreted as an electric silence due to the abrupt decrease in EOD amplitude or as a conspicuous electrical exhibition. Moreover, the metabolic cost of chirping activity is also unclear. Although the EOD constitutes 11–22% of the 24 h energy budgets of males and 3% of energy budgets for females (Salazar and Stoddard, 2008), there are no estimations of the energetic cost of intra-chirp EODs in comparison to normal discharge pattern. With respect to the interpretation of female interruptions as SES, A. Silva et al. / Journal of Physiology - Paris 102 (2008) 272–278 the question to be posed is: why do choosy females have to signal at all? Interruption may be acceptance signals, submissive ones, or active interacting signals necessary to synchronize spawning in these nocturnal fish with external fertilization. Summarizing our behavioral observations, distinctive SES play a role in the reproductive behavior of B. pinnicaudatus. As court ship signals from cyclic reproductive temperate fish, SES show seasonal changes and are sexually dimorphic. Since all SES imply EOD rate modul ations, the medullary pacemaker nucleus (PN) is ultimately involved in their generat ion and becomes an interesting target organ to explore seasonal and sexual plasticity of the CNS. D 100 µm The electrogenic capacity of Gymnotiformes relies initially on the activity of the medullary PN, an endogenous oscillator which triggers each EOD by a single command pulse. The PN has tradi tionally been described as having two neuron types: pacemaker neurons that fire in synchrony and generate the rhythm, and neu rons that relay the pacemaker command pulse to the electromo torneurons in the spinal cord. Except in the family Apteronotidae, these electromotorneurons, in turn, innervate the electric organ and trigger its discharge. (Bennett, 1971; Dye and Meyer, 1986; Kennedy and Heiligenberg, 1994). In most Gymnotiformes, the PN has large relay cells, which reside on the ventral midline of the medulla (Szabo and Enger, 1964), and small pacemaker cells, half the size of relay cells, and more dorsally located. The connection between pacemaker and relay cells is supported by mixed synapses (Bennett et al., 1967; Elekes and Szabo, 1981). Recently there has been a novel interneu ron described in Apteronotus leptorhynchus, which may be elec trotonically coupled with relay and pacemaker cells (Smith et al., 2000; Turner and Moroz, 1995). EOD rate modulations, as chirps and interruptions, arise from descendent inputs to the PN in several gymnotiform species includ ing the genus Brachyhypopomus (Kawasaki et al., 1988; Kawasaki and Heiligenberg, 1990; Keller et al., 1991). With only the two neuronal types organized in a rather sim ple circuit with exclusive feed forward connections, the PN is remarkable in its capacity to produce not only steady rhythmic fir ing, but also outputs with distinct temporal dynamics. We set out to explore the functional anatomy of the PN in B. pinnicaudatus, which is critic al in understanding and further exploring the gener ation mechanisms of SES. 3.1. Anatomo-functional organization of the PN in B. pinnicaudatus As described in other Gymnotiformes, relay cells are ventral and pacemaker neurons are dorsal in the PN of B. pinnicaudatus (Fig. 3). Relay cells cover a wide range of sizes (more than 25 lm in diam eter), whereas pacemaker neurons are smaller (less than 25 lm in diameter). There is no evidence of other neuronal types in the PN, although parvalbumin labeled fibers are present, apparently orig inating from neighboring neurons (Quintana et al., 2007). In addi tion to the expected dorso-ventral neuron distribution, there was a clear rostro-caudal anisotropic distribution, as pacemaker cells are distinctly grouped in the rostral two thirds of the PN. Relay cells are present ventrally in the rostral part, and occupy the entire caudal pole of the PN (Fig. 3). In vivo electrophysiological field potential recordings of the PN show a compound field potential phase locked with each EOD, preceding it by approxim ately 5 ms (Fig. 3). In dorso-ventral pen etrations of the PN, the field potential waveform can be separated in an early component that reaches its maximal amplitude in dorsal sites reflecting the pacemaker neuron activity; and a late C R 1 mV 4 ms RF 3. The pacemaker nucleus (PN): the ultimate target of EOD rate modulations 275 EOD V RF EOD Fig. 3. Anatomo-functional organization of the PN. Center: Sagital section (60 lm) of the PN with Triple Stain of Cajal-Gallego, with overlaying schematic drawing. Pacemaker neurons are small, dorsal, and mainly in the rostral region (enclosed in the dotted line). Relay neurons are present along the ventral PN, and make up the whole caudal end of the PN (enclosed in the dashed line). To the right and left of the PN the field potential waveforms of the indicated sites are shown. Dotted lines show the alignment of the relay field potential peak (RF) and the negative peak of the EOD. Overall maximum pacemaker (early) component of the field potential is recorded in the rostral penetration. In the penetration 200 lm caudal to this rostral track, the relay component reaches its overall maximum. R: rostral, C: caudal, D: dorsal, V: ventral. component, which reaches its maximal amplitude more ventrally resulting from the synchronized activity of relay cells. The profile of these dorso-ventral penetrations in rostral and caudal tracks are different, in correspondence to what has been described his tologically. Overall, maximum pacemaker component is recorded in rostral tracks, whereas relay components are larger in ventral and caudal sites (Fig. 3). Dye marking during the field potential recordings confirm that the maximum pacemaker component is recorded when the electrode is close to the pacemaker neuron population whereas the maximum relay component is recorded in the proximity of caudal relay population (data not shown). In summary, the histological and electrophysiol ogical results presented in this study show a complex citoarquitectural organi zation of the PN of B. pinnicaudatus with anisotropic neuronal dis tribution not only in the dorso-ventral axis but also in the rostrocaudal one. These topographic differences within the PN might give support to its plastic functionality. It is generally accepted that the PN only controls the discharge rate whereas the descendent pathway and the electric organ transform each command pulse into a complex spatio-temporal pattern of electric currents that are transferred to the water around the fish (Caputi, 1999; Caputi et al., 2005). However, some reports suggest the PN may also be involved in shaping the EOD. Studies of retrograde tracing from electromotorneurons in Gymnotus, suggest a somatotopic organi zation within the PN (Ellis and Szabo, 1980). Intracellular stimula tion of different relay cells in Gymnotus activates distinct regions of the electric organ (Lorenzo et al., 1993), and transient local inac tivation of certain regions of the ventral PN affects the EOD wave form (M. Borde, personal communication). 3.2. Distinctive SES can be evoked by glutamate injection in particular portions of the PN As mentioned above, the EOD rate modulations that B. pinnicaudatus display in different behavioral contexts, originate from pre-pacemaker nuclei, which provide modulatory inputs to the PN. In particular, chirps are produced by a glutamatergic input to the PN from diencephalic nuclei, whereas interruptions are gen erated by glutamatergic input from sublemniscal nuclei (Kawasaki and Heiligenberg, 1990; Kennedy and Heiligenberg, 1994). Elec trophysiological experiments have shown that different classes of glutamate receptors mediate the generation of different EOD rate 276 A. Silva et al. / Journal of Physiology - Paris 102 (2008) 272–278 modulations. It has been proposed that chirps are mediated by the activation of AMPA-kainate receptors upon the relay cells whereas interruptions depend upon the activation of NMDA receptors on the same cells (Kawasaki and Heiligenberg, 1990). Injections of anterograde tracers to the pre-pacemaker diencepahlic nuclei, which modul ate the PN to produce chirps, reveal terminals scat tered freely within the PN. Simil ar injections to brainstem nuclei, which modulate the PN to produce EOD interruptions, yield termi nals, clustered around the relay cell somata. It thus appears that the two inputs to the relay cells are spatially segregated (Kennedy and Heiligenberg, 1994). We began the analysis of the mechanisms responsible for sex ual and seasonal behavioral differences in the production of SES by focusing on the PN, and its response to glutamate. In vivo electro physio logical experiments were performed in breeding males and females (n = 38), and in non-breeding adults (n = 13). The correla tion between anatomy and field potential waveform was used to chemically stimulate certain areas of the PN in fish that had pre viously been behaviorally observed and recorded in male–female dyads during both the breeding and non-breeding seasons. Based upon the histological studies, dorso-ventral penetrations were performed in rostral and caudal areas of the PN. Glutamate was injected in the dorsal PN, where the pacemaker component of the field potential was larger, and in the ventral PN, where the relay component was maximal. Glutamate injections provoke different kinds of EOD modulations depending on the site of injection, the sex of the animal, and its reproductive state. As shown in Fig. 4, glutamate stimulation in the PN of breeding males produce chirplike responses (in 12 out of 18 experiments) most clearly in ventral areas of rostral tracks, and EOD interruptions in ventral areas of caudal tracks (in 18 out of 18 experiments). Chirp-like responses to glutamate fit the definition of chirp (see Section 2) in terms of dura tion, rate increase, and amplitude distortion. In contrast, breeding females interrupt their EOD when glutamate is injected in ventral areas (20 out of 20 experim ents) regardless of the rostro-caudal position (Fig. 4). During breeding, males and females generate EOD frequency rises when stimulated in dorsal areas of the PN (data not shown). During the non-breeding season, both males and females generate EOD frequency rises when stimulated in dorsal areas of the PN and EOD interruptions when stimulated in ventral areas of the PN (13 out of 13 experiments, data not shown). ROSTRAL CAUDAL Our results show that the seasonal and sexual differences in the generation of SES are explained, at least in part, by changes in the response to glutamate of the PN. In vivo and histological data sug gest that the caudal area of the PN, solely composed by relay cells, remains unchanged; whereas the rostral area, namely where relay neurons are ventral to pacemaker neurons, is subject to modifica tion. Chirps have been shown to occur in relay cells via the activa tion of AMPA receptors (Kawasaki and Heiligenberg, 1990). Results from in vitro preparations of the PN of B. pinnicaudatus, indicate that AMPA reliably induces neuronal chirping-like activity when injected in the ventral PN of breeding males (Macadar et al., 2007). Therefore, the seasonal and sexual differences in the response to glutamate may be due to differential expression (specific locali zation or density) of AMPA receptors on a certain group of relay cells. This surprisingly narrows down our search for targets that respond to seasonal and sexual factors to a few dozen neurons in the PN. Particularly, within these few relay cells, we can postulate that changes in distribution of AMPA receptors will be more likely Wild 50 ms Behavioral Recording Station 50 ms Glutamate induced MALE 500 ms FEMALE 50 ms Fig. 4. Sexual dimorphism in the response of the ventral PN to glutamate during breeding. Head-to-tail EOD recordings from anesthetized (pentobarbital, 25 lg/g) breeding male (upper traces) and female (lower traces). Glutamate (10 mM) was injected by pressure in different portions of the PN (identified by their field potential waveforms). Glutamate injections in the ventral PN, where the relay component of the field potential recording is maximal, provoke different effects in males and females. In rostral sites (left recordings), the male responded with chirp-like EOD modulations whereas the female interrupted its EOD. In caudal sites (right record ings), both the male and the female interrupted their EOD. Fig. 5. Type B chirps recorded with different experimental approaches. The EODs of freely moving male–female dyads (smaller female EODs indicated by dots) were recorded both in the wild (from 2 pairs of fixed electrodes placed in a restricted area within the Laguna el Tigre, Uruguay, 33°189S, 54°359W), and in the labora tory recording station (see Fig. 1) during breeding. Hundreds of spontaneous type B chirps are observed both in the wild and in the recording station during breeding nights (representative examples displayed). Furthermore, chirp-like responses can be evoked by glutamate injections in distinct regions of the PN (see Fig. 4 for meth odological details). A. Silva et al. / Journal of Physiology - Paris 102 (2008) 272–278 to occur on their dendritic trees based on tract-tracing studies by Kennedy and Heiligenberg (1994). These interpretations open sev eral interesting aspects to explore, from the influence of androgens on the PN to the role of seasonal changes of glutamate activity in both free-swimming dyads and physiol ogical preparations. 4. From behavior to the PN In its simplest definition neuroethology is the study of the neu ral mechanisms underlying behavior (Pflüger and Menzel, 1999). Neuroethology is an approach to understanding the neural control of behavior integrating at least two levels: the identification of the neural circuits involved and how information is represented and processed in them; and, ultimately, how evolution shaped behav iors and their neural solutions (Rose, 2004). In the words of C.D. Hopkins, “the ethological approach of neuroethology emphasizes the causation, the development, the evolution, and the function of behavior and neuroethologists seek to understand this in terms of neural circuits”. Although electric fish have been traditional model systems in neuroethology, few studies have been able to combine field and labor atory approaches. We have been able to identify sexual and seasonal plasticity of the central nervous system and to initiate the exploration of underlying mechanisms by integrating behavioral recordings both in the wild and in laboratory settings, and in vivo electrophysiological and pharmacological approaches. Male chirps recorded in behavioral labor atory settings are indistinguishable from those recorded in the wild (for example, type B chirps shown in Fig. 5). In addition, equivalent chirp-like responses are recorded after in vivo glutamate injection in the breeding male PN (for example, type B-like chirp shown in Fig. 5). In vitro, glutamate and AMPA induce rate modulations in the male PN that also resemble type B chirps (Macadar et al., 2007). In conclusion, we believe this study contributes to the integra tive approach of neuroethology. Moreover, the strong match of diverse data originated from behavior in the natural environment to the activity of a small group of medullary neurons, reinforces the advantages of this model in the exploration of the neural bases of behavior. Acknowledgments We thank Omar Macad ar for extensive revision of the man uscript and helpful comments. We are especially thankful to Gonzalo de Armas for his help in the solution and running of the PCA and cluster analysis. We also thank Maira Colacce for her advice on multivariate statistics. This study was partially supported by PEDECIBA and DICYT (PDT 043). Collections and experimental procedures were performed under the guidelines and approval of our local ethical committee (Comisión Honor aria de Experimen tación Animal, Universidad de la República, “Uso de animales en experimentación, docencia e investigación Universitaria”, CDC Exp 4332/99, Diario Oficial N 25467, Feb. 21/00). References Andersson, M., 1994. Sexual Selection. Princeton University Press, Princeton, NJ. Bass, A.H., Zakon, H.H., 2005. Sonic and elect ric fish: at the crossroads of neuroet holo g y and behavi oral neuroe nd oc rin olo gy. Horm. Behav. 48, 360–372. Bastian, J., Schniederjan, S., Nguyenkim, J., 2001. Arginine vasotocin modulates a sexually dimorphic communication behavior in the weakly electric fish Apteronotus leptorhynchus. J. Exp. Biol. 294, 1909–1923. Bateman, A.J., 1948. Intra-sexual selection in Drosophila. Heredity 2, 349–368. Bennett, M.V.L., 1971. Electric organs. In: Hoar, W.S., Randall, D.J. (Eds.), In: Fish Physiology, Vol. V. Academic Press, London, UK, pp. 347–491. Bennett, M.V.L., Pappas, G.D., Giménez, M., Nakajima, Y., 1967. Physio logy and ultra structure of electrotonic junctions. IV. Medullary electromotor nuclei in gym notid fish. J. Neurophysiol. 30, 236–300. 277 Caputi, A.A., Silva, A., Macadar, O., 1998. The EOD of Brachyhypopomus pinnicauda tus: the effect of environmental variables on waveform generation. Brain Behav. Evol. 52, 148–158. Caputi, A.A., 1999. The electric organ discharge of pulse gymnotiforms: the trans formation of a simple impulse into a complex spatio-temporal electromotor pattern. J. Exp. Biol. 202, 1229–1241. Caputi, A.A., Carlson, B.A., Macadar, O., 2005. Electric organs and their control. In: Bullock, T., Hopkins, C., Fay, R., Popper, N. (Eds.), Electroreception. Springer, New York, NY, pp. 410–451. Dye, J.C., Meyer, J.H., 1986. Central control of the electric organ discharge of a weakly electric fish. In: Bullock, T.H., Heiligenberg, W. (Eds.), Electroreception. Wiley, New York, pp. 72–102. Elekes, K., Szabo, T., 1981. Synaptology of the command (pacemaker) nucleus in the brain of the weakly electric fish, Sternarchus (Apteronotus) albifrons. Neurosci ence 6, 443–460. Ellis, D.B., Szabo, T., 1980. Identification of different cells types in the command (pacemaker) nucleus of several gynotiform species by retrograde transport of horseradish peroxidase. Neuroscience 5, 1917–1929. Grafen, A., 1990. Biological signals as handicaps. J. Theor. Biol. 144, 517–546. Hagedorn, M., 1988. Ecology and behavior of a pulse-type electric fish Hypopomus occidentalis (Gymnotiformes, Hypopomidae) in a fresh-water stream in Pan ama. Copeia 2, 324–335. Hagedorn, M., Heiligenberg, W., 1985. Court and spark: electric signals in the court ship and mating of gimnotoid fish. Anim. Behav. 33, 254–265. Hopkins, C.D., 1974a. Electric communication: functions in the social behavior of Eigenmannia virescens. Behaviour 50, 270–305. Hopkins, C.D., 1974b. Electric communication in the reproductive behavior of Sternopygus macrurus (Gymnotoidei). Z. Tierpsychol. 35, 518–535. Hopkins, C.D., Comfort, N.C., Bastian, J., Bass, A.H., 1990. Functional analysis of sex ual dimorphism in an electric fish, Hypopomus pinnicaudatus, Order Gymnoti formes. Brain Behav. Evol. 35, 350–367. Hopkins, C.D., 1991. Hypopomus pinnicaudatus (Hypopomidae) a new species of gymnotiform fish from French Guiana. Copeia 1, 151–161. Kawasaki, M., Maler, L., Rose, G.J., Heiligenberg, W., 1988. Anatomical and func tional organization of the prepacemaker nucleus in gymnotiform electric fish: the accommodation of two behaviors in one nucleus. J. Comp. Neurol. 276, 113–131. Kawasaki, M., Heiligenberg, W., 1989. Distinct mechanisms of modul ation in a neuronal oscillator generate different signals in the electric fish Hypopomus. J. Comp. Physiol. A 165, 731–741. Kawasaki, M., Heiligenberg, W., 1990. Different classes of glutamate receptors and GABA mediate distinct modulations of a neuronal oscillator, the medullary pacemaker of a gymnotiform electric fish. J. Neurosci. 10, 3896–3904. Keller, C.H., Kawasaki, M., Heiligenberg, W., 1991. The control of pacemaker modula tions for social communication in the weakly electric fish Sternopygus. J. Comp. Physiol. A 169, 441–450. Kennedy, G., Heiligenberg, W., 1994. Ultrastructural evidence of GABA-ergic inhibi tion and glutamatergic excitation in the pacemaker nucleus of the gymnotiform electric fish, Hypopomus. J. Comp. Physiol. A 174, 267–280. Kirschbaum, F., 1979. Reproduction of the weakly electric fish Eigenmannia virescens (Rhamphichthydae, Teleostei) in captivity. Behav. Ecol. Sociobiol. 4, 331–355. Kirschbaum, F., 1995. Taxonomy, zoogeography and general ecology of South Amer ican knifefishes (Gymnotiformes). In: Moller, P. (Ed.), Behavior, History, Electric fishes. Chapman and Hall, London, UK, pp. 446–464. Kramer, B., Otto, B., 1988. Female discharges are more electrifying: spontaneous preference in the electric fish, Eigenmannia (Gymnotiformes, Teleostei). Behav. Ecol. Sociobiol. 23, 55–60. Lorenzo, D., Sierra, F., Silva, A., Macadar, O., 1993. Spatial distribution of the med ullary command signal within the electric organ of Gymnotus carapo. J. Comp. Physiol. A 173, 221–226. Macadar, O., Lescano, C., Lorenzo, D., 2007. Sex-specific social electric signals induced in vitro in a weakly electric fish. Abstract in “8th International Con gress of Neuroethology”, 234. Macadar, O., Silva, A., 2007. Comunicación eléctrica en peces sudamericanos del orden gymnotiformes. Rev. Lat. Am. Psicol. 39, 31–45. Miranda, M., Silva, A.C., Stoddard, P.K., 2008. Use of space as an indicator of social behavior and breeding systems in the gymnotiform electric fish Brachyhypopomus pinnicaudatus. Env. Biol. Fish., in press, doi:10.1007/ s10441-008-9358-2. Pflüger, H.J., Menzel, R., 1999. Neuroethology, its roots and future. J. Comp. Physiol. A 185, 389–392. Quintana, L., Silva, A., Berois, N., Macadar, O., 2004. Temperature induces gonadal maturation and affects electrophysiol ogical sexual maturity indicators in Brachyhypopomus pinnicaudatus from a temperate climate. J. Exp. Biol. 207, 1843–1853. Quintana, L., Pouso, P., Macadar, O., 2007. Neuroanatomical basis of social electric signals. Abstract in “8th International Congress of Neuroethology”, 111. Rose, G., 2004. Insights into neural mechanisms and evolution of behaviour from electric fish. Nat. Neurosci. Rev. 5, 943–951. Salazar, V., Stoddard, P.K., 2008. Sex differences in energetic costs explain sexual dimorphism in the circadian rhythm modulation of the electrocommunication signal of the gymnotiform fish Brachyhypopomus pinnicaudatus. J. Exp. Biol. 211, 1012–1020. Silva, A., 2002. Efectos de la temperat ura en Gymnotiformes de la zona temp lada. PhD Dissertation. Montevideo, Uruguay: PEDECIBA-Universidad de la República. 278 A. Silva et al. / Journal of Physiology - Paris 102 (2008) 272–278 Silva, A., Quintana, L., Galeano, M., Errandonea, P., Macadar, O., 1999. Water tem perat ure sensitivity of the EOD waveform in Brachyhypopomus pinnicaudatus. J. Comp. Physiol. A 185, 187–197. Silva, A., Quintana, L., Ardanaz, J.L., Macadar, O., 2002. Environmental and hormonal influences upon EOD waveform in gymnotiform pulse fish. J. Physiol. Paris 96, 473–484. Silva, A., Quintana, L., Galeano, M., Errandonea, P., 2003. Biogeography and breeding in Gymnotiformes from Uruguay. Environ. Biol. Fish. 66, 329–338. Silva, A., Perrone, R., Macadar, O., 2007a. Environmental, seasonal, and social modu lations of basal activity in a weakly electric fish. Physiol. Behav. 90, 525–536. Silva, A., Perrone, R., and Macadar, O., 2007b. Sexual and seasonal plasticity in the emission of social electric signals. I. Behavioral approach. Abstract in “Electro sensory Systems: Satellite Meeting to the 8th International Congress of Neu roethology”, 15. Smith, G.T., Lu, Y., Zakon, H.H., 2000. Parvocells: a novel interneuron type in the pacemaker nucleus of a weakly electric fish. J. Comp. Neurol. 423, 427–439. Szabo, T., Enger, P., 1964. Pacemaker activity of the medullary nucleus controlling electric organs in high-frequency gymnotid fish. J. Comp. Physiol. A 49, 285– 300. Turner, R.W., Moroz, L.L., 1995. Localiz ation of nicotinamide adenine dinucleo tide phosphate-diaphorase activity in electrosensor y and electromotor sys tems of a gymnotiform teleo st, Apteronotus leptorhynchus. J. Comp. Neurol. 356, 261–274. Zakon, H.H., Dunlap, K.D., 1999. Sex steroids and communication signals in electric fish: a tail of two species. Brain Behav. Evol. 54, 61–69. Zupanc, G.K.H., 2002. From oscillators to modulators: behavioral and neural con trol of modulations of the electric organ discharge in the gymnotiform fish, Apteronotus leptorhynchus. J. Physiol. Paris 96, 459–472.