Vocal-Acoustic Pathways in a Teleost Fish
Total Page:16
File Type:pdf, Size:1020Kb
The Journal of Neuroscience, July 1994, 74(7): 4025AO39 Vocal-Acoustic Pathways in a Teleost Fish Andrew H. Bass,‘J Margaret A. Marchaterre,is2 and Robert Baker3 ‘Section of Neurobiology and Behavior, Cornell University, Ithaca, New York 14853, 2Bodega Marine Laboratory, University of California, Bodega Bay, California 94923, and 3Department of Physiology and Biophysics, New York University Medical Center, New York, New York 10016 Many teleost fish generate acoustic signals for vocal com- of marine fishes,sound is generatedby the contraction of paired munication by the synchronized, high-frequency contraction skeletal, sonic musclesattached to the lateral walls of the swim of skeletal, sonic muscles. In midshipman, eight groups of bladder (SM, Fig. lA,B). From a comparative, evolutionary brainstem neurons were distinguished after biocytin appli- perspectivewith tetrapods, it is appropriate to classify the acous- cation to the sonic nerve that, we propose, represent the tic signalsofthese fish as vocalizations. Tetrapods uselaryngeal, entire vocal motor circuit. Biocytin-filled terminals were ubiq- syringeal, and tongue musculatureto vocalize. Thesesonic mus- uitous within all areas containing labeled neurons and, to- cles share a number of traits with those of batrachoidid fishes gether with ultrastructural evidence, suggested a serial, including the production of species-and sex-typical acoustic transneuronal transport at synaptic sites between at least signals,differentiation as skeletal musclesoriginating from oc- three neuronal groups. The most intensely labeled neurons cipital somites, and innervation by homologous sets of moto- were positioned in the caudal brainstem and included a pre- neurons: syringeal and tongue musclesare innervated by the viously characterized pacemaker-motoneuron circuit and a hypoglossalmotor nucleus, which is considered a homolog of newly recognized ventral medullary nucleus that itself gave the batrachoidid sonic motor nucleus (review: Bass, 1989). rise to extensive commissural and lateral brainstem bundles In the plainfin midshipman (Povichthys notutus), a represen- linking the pacemaker circuitry to the rostra1 brainstem. Five tative batrachoidid, the sonic motor nucleus (SMN, Fig. 1C) additional groups formed a column rostrally within the medial has a midline position at the caudal end of the brainstem and brainstem adjacent to eighth nerve (octaval)-recipient nuclei innervates the ipsilateral muscle via occipital nerve roots (OC, largely presumed to be acoustic. This column extended dor- Fig. 1C). So far, extensive intracellular recording and staining sally up to the ventricular cell layer and as far anterior as studies in midshipman have identified individual pacemaker midbrain isthmal levels. The best-defined group was in the neurons, which lie adjacent to the sonic motor nucleus(PN, Fig. octaval efferent nucleus that directly innervates the sacculus 1C), as the only source of direct synaptic input onto sonic mo- that is considered the auditory division of the inner ear. Sac- toneurons (Bassand Baker, 1990). Pacemakerneurons set the cular afferents and neurons throughout the medial column dischargefrequency ofthe motoneurons, which then determines were also filled after biocytin application to the saccular the contraction rate ofthe musclesand, in turn, the fundamental nerve. This vocal-acoustic network overlaps low-threshold, frequency of vocalizations. Individual pacemaker neurons in- electrical stimulation sites in the rostra1 brainstem that elicit nervate both motor nuclei, consistent with the hypothesis that vocalizations. The medial column must therefore be the or- their role is to synchronize motoneuron firing that leads to the igin of the descending pathway controlling activation of the simultaneous contraction of both musclesat the fundamental vocal pacemaker circuitry and likely forms the basis for frequency. Thus, there is a direct relationship betweenthe rhyth- acoustically elicited vocalizations. We suggest this network, mic, patterned output of a brainstem pacemakercircuit and the together with input from the pacemaker circuitry, is also the physical attributes of species-and sex-typical vocalizations. origin of a vocal-related, corollary discharge to acoustic nu- One long-term goal of vocal motor studiesin both anamniotes clei. Direct links between vocal and acoustic brain regions and amniotes has been to identify the neurons controlling the are thus traits common to aquatic and terrestrial vertebrates. rhythmic discharge of the vocal pacemaker circuitry. While a [Key words: brainstem, pacemaker circuit, vocalization, number of studies in batrachoidid fishes had used electrical acoustic, corollary discharge] stimulation to elicit vocalizations at sitesdistant from the motor nucleus, the positions of neuronal groups forming a descending Teleost fishesgenerate acoustic communication signalsimpor- pathway were unknown (review: Bass and Baker, 1991). As tant to both their survivorship and reproductive success(re- mentioned above, intracellular studies have so far identified views: Bass, 1990, 1992). Among Batrachoidiformes, an order individual pacemaker neurons as providing the sole input to sonic motoneurons. Given the limitations of single-cell studies in delineating the complete extent and location of neuronswith- Received Aug. 18, 1993; revised Dec. 16, 1993; accepted Dec. 21, 1993. in a circuit, a potential transneuronal tracer was sought to both This work was supported from the National Science Foundation (BNS8708559, map the entire complement of pacemakerneurons and identify 902 1563) and the National Institutes of Health (NS 13742). Thanks to the Bodeaa Marine Laboratory for logistical support, Dr. h. Weiser’ for assistance with tie their afferents. Biocytin, a complex of biotin and lysine, was figures, and Drs. H. Baker, M. Braford Jr., and C. McCormick for reading and chosen as a candidate transneuronal tracer for a number of commenting on the manuscript. Correspondence should be addressed to Dr. Andrew H. Bass, Section of Neu- reasons,foremost among which were its low molecular weight robiology and Behavior, Mudd Hall, Cornell University, Ithaca, NY 14853. (372), diffusion throughout the soma-dendritic-axonal com- Copyright 0 1994 Society for Neuroscience 0270-6474/94/144025-15$05.00/O partments of neurons, and high affinity for avidin, which made 4026 Bass et al. * Vocal-Acoustic Circuit in Teleost Fish Figure 1. Overview of vocal motor system of midshipman. A, Line drawing of a lateral view of a plainfin midshipman (Porichthys notutus) showing the position of the swim bladder (SB) and attached sonic muscles (SM) at the level of the pelvic fin (P). B, Line drawing of a ventral view ofa male; the testes(r) have been deflected to show the paired sonic muscles (SA4) attached to the swim bladder’s walls. C, A schematic reconstruction superimposed on a dorsal view of the midshipman’s brain showing the positions of the eight groups of biocytin-filled neurons labeled following biocytin labeling of the sonic nerve at the level of the swim bladder (BC*, B). The approximate anterior-posterior extent of the line drawings presented in Figures 4, 6, and 8 is indicated. AC, anterior canal ampulla; C, cerebellum; DT, dorsal tegmental group; HC, horizontal canal ampulla; HP, hindbrain paraventricular group; ZP, isthmal paraventricular group; M, midbrain; OB, olfactory bulb; OL, olfactory nerve; OC, occipital nerves; ON, optic nerve; PLLN, posterior lateral line nerve; PN, pacemaker neurons; SA, saccular otolith; SE, statoacoustic efferent nucleus; SMN, sonic motor nucleus; SP, spinal cord; T, telencephalon; V, trigeminal nerve; VM, ventral medullary nucleus; VT, ventral tegmental group; VIII, eighth nerve; IX, glossopharyngeal nerve; I”, vagus nerve. Scale bar, 1 mm. it optimally suited for use with a variety of immunocytochem- cal systems (Suga and Shimozawa, 1974; Bell, 1989; Metzner, ical markers (Horikawa and Armstrong, 1988). Surprisingly, 1993). Finally, as reasoned here, if the entire complement of biocytin labeling of the midshipman’s sonic nerve at the level neurons in a vertebrate circuit can be delineated by a single of the swim bladder delineated the full extent of the pacemaker application of a tracer to its most peripheral element, a motor circuitry and, more impressively, demonstrated a linkage be- or sensory nerve, then this technique has obvious advantages tween that circuitry and acoustic regions ofthe rostra1 brainstem over most other tract-tracing methqdologies, especially for small, extending up to midbrain levels. A vocal-acoustic linkage was nonmammalian vertebrates. The comprehensive mapping sug- further implied when, after biocytin labeling of the eighth nerve, gested here by biocytin provides an immediate guide to the which in,nervates the inner ear, transneuronal transport resulted causal analysis of a neuronal network, like a vocal motor circuit, in the filling of neurons that were also labeled via the sonic determining the execution of a specific behavior, such as vo- nerve. calizations. Comparisons with neurophysiological studies indicate this Portions of these results were reported earlier (Bass et al., vocal-acoustic network must include the pathway for descend- 1992). ing control of the vocal pacemaker circuit, which, in part, could be activated by primary or higher-order acoustic inputs. We Materials and Methods suggest that this vocal-acoustic circuit also includes a corollary This study included both adult and juvenile midshipman fish.