Anatomical evidence for binaural processing in the descending octaval nucleus of the toadfish (Opsanus tau)

Anatomical adventures in the fish auditory medullaa)

This paper provides an overview of my work on the central auditory system of fish. It focuses on my comparative analyses of a nucleus that receives input from the inner ear, the descending nucleus, and more specifically on that part of the descending nucleus supplied by the otolith end organs, the dorsal descending nucleus. I begin by summarizing my initial work on the bowfin, Amia calva, and go on to explain the importance of taking a comparative approach to understanding ancestral and specialized anatomical and putative functional characteristics of the dorsal descending nucleus in modern bony fishes, the teleosts.

A journey through the field of fish hearinga)

My interest in fish bioacoustics was ignited more than 50 years ago and resulted in a zigzag time travel between various interesting problems that were unsettled at the time. The present paper gives a brief overview of the main topics I have worked on in the field of fish hearing, i.e., auditory function of the swim bladder, directional hearing, function of the lateral line system, and infrasound sensitivity. Rather than being a comprehensive review of these issues, the paper is autobiographical and limited. The aim is to show young scientists that experimental science can be exciting, diverse, and rewarding-and open doors to a rich collegial network, collaboration, and friendships.

Bone conduction pathways confer directional cues to salamanders

Sound and vibration are generated by mechanical disturbances within the environment, and the ability to detect and localize these acoustic cues is generally important for survival, as suggested by the early emergence of inherently directional otolithic ears in vertebrate evolutionary history. However, fossil evidence indicates that the water-adapted ear of early terrestrial tetrapods lacked specialized peripheral structures to transduce sound pressure (e.g. tympana). Therefore, early terrestrial hearing should have required nontympanic (or extratympanic) mechanisms for sound detection and localization. Here, we used atympanate salamanders to investigate the efficacy of extratympanic pathways to support directional hearing in air. We assessed peripheral encoding of directional acoustic information using directionally masked auditory brainstem response recordings. We used laser Doppler vibrometry to measure the velocity of sound pressure-induced head vibrations as a key extratympanic mechanism for aerial sound reception in atympanate species. We found that sound generates head vibrations that vary with the angle of the incident sound. This extratympanic pathway for hearing supports a figure-eight pattern of directional auditory sensitivity to airborne sound in the absence of a pressure-transducing tympanic ear.

Tracing of Afferent Connections in the Zebrafish Cerebellum Using Recombinant Rabies Virus

The cerebellum is involved in some forms of motor coordination and learning, and in cognitive and emotional functions. To elucidate the functions of the cerebellum, it is important to unravel the detailed connections of the cerebellar neurons. Although the cerebellar neural circuit structure is generally conserved among vertebrates, it is not clear whether the cerebellum receives and processes the same or similar information in different vertebrate species. Here, we performed monosynaptic retrograde tracing with recombinant rabies viruses (RV) to identify the afferent connections of the zebrafish cerebellar neurons. We used a G-deleted RV that expressed GFP. The virus was also pseudotyped with EnvA, an envelope protein of avian sarcoma and leucosis virus (ALSV-A). For the specific infection of cerebellar neurons, we expressed the RV glycoprotein (G) gene and the envelope protein TVA, which is the receptor for EnvA, in Purkinje cells (PCs) or granule cells (GCs), using the promoter for aldolase Ca (aldoca) or cerebellin 12 (cbln12), respectively. When the virus infected PCs in the aldoca line, GFP was detected in the PCs’ presynaptic neurons, including GCs and neurons in the inferior olivary nuclei (IOs), which send climbing fibers (CFs). These observations validated the RV tracing method in zebrafish. When the virus infected GCs in the cbln12 line, GFP was again detected in their presynaptic neurons, including neurons in the pretectal nuclei, the nucleus lateralis valvulae (NLV), the central gray (CG), the medial octavolateralis nucleus (MON), and the descending octaval nucleus (DON). GFP was not observed in these neurons when the virus infected PCs in the aldoca line. These precerebellar neurons generally agree with those reported for other teleost species and are at least partly conserved with those in mammals. Our results demonstrate that the RV system can be used for connectome analyses in zebrafish, and provide fundamental information about the cerebellar neural circuits, which will be valuable for elucidating the functions of cerebellar neural circuits in zebrafish.

Inhibitory Neural Circuits in the Mammalian Auditory Midbrain

The auditory midbrain is the critical integration center in the auditory pathway of vertebrates. Synaptic inhibition plays a key role during information processing in the auditory midbrain, and these inhibitory neural circuits are seen in all vertebrates and are likely essential for hearing. Here, we review the structure and function of the inhibitory neural circuits of the auditory midbrain. First, we provide an overview on how these inhibitory circuits are organized within different clades of vertebrates. Next, we focus on recent findings in the mammalian auditory midbrain, the most studied of the vertebrates, and discuss how the mammalian auditory midbrain is functionally coordinated.

Evolution of Sound Source Localization Circuits in the Nonmammalian Vertebrate Brainstem

The earliest vertebrate ears likely subserved a gravistatic function for orientation in the aquatic environment. However, in addition to detecting acceleration created by the animal's own movements, the otolithic end organs that detect linear acceleration would have responded to particle movement created by external sources. The potential to identify and localize these external sources may have been a major selection force in the evolution of the early vertebrate ear and in the processing of sound in the central nervous system. The intrinsic physiological polarization of sensory hair cells on the otolith organs confers sensitivity to the direction of stimulation, including the direction of particle motion at auditory frequencies. In extant fishes, afferents from otolithic end organs encode the axis of particle motion, which is conveyed to the dorsal regions of first-order octaval nuclei. This directional information is further enhanced by bilateral computations in the medulla and the auditory midbrain. We propose that similar direction-sensitive neurons were present in the early aquatic tetrapods and that selection for sound localization in air acted upon preexisting brain stem circuits like those in fishes. With movement onto land, the early tetrapods may have retained some sensitivity to particle motion, transduced by bone conduction, and later acquired new auditory papillae and tympanic hearing. Tympanic hearing arose in parallel within each of the major tetrapod lineages and would have led to increased sensitivity to a broader frequency range and to modification of the preexisting circuitry for sound source localization.

What the Toadfish Ear Tells the Toadfish Brain About Sound

Of the three, paired otolithic endorgans in the ear of teleost fishes, the saccule is the one most often demonstrated to have a major role in encoding frequencies of biologically relevant sounds. The toadfish saccule also encodes sound level and sound source direction in the phase-locked activity conveyed via auditory afferents to nuclei of the ipsilateral octaval column in the medulla. Although paired auditory receptors are present in teleost fishes, binaural processes were believed to be unimportant due to the speed of sound in water and the acoustic transparency of the tissues in water. In contrast, there are behavioral and anatomical data that support binaural processing in fishes. Studies in the toadfish combined anatomical tract-tracing and physiological recordings from identified sites along the ascending auditory pathway to document response characteristics at each level. Binaural computations in the medulla and midbrain sharpen the directional information provided by the saccule. Furthermore, physiological studies in the central nervous system indicated that encoding frequency, sound level, temporal pattern, and sound source direction are important components of what the toadfish ear tells the toadfish brain about sound.

Does the magnocellular octaval nucleus process auditory information in the toadfish, Opsanus tau?

Previous work on auditory processing in Opsanus tau has focused on the descending octaval nucleus; however, the magnocellular octaval nucleus receives similar inputs from the otolithic endorgans. The purpose of this study was to assess whether cells in any of the three subdivisions of the magnocellular nucleus respond to auditory frequencies and encode sound source direction. Extracellular recording sites were chosen based on anatomical landmarks, and neurobiotin injections confirmed the location of auditory sites in subdivisions of the magnocellular nucleus. In general, the auditory cells in M2 and M3 responded best to frequencies at or below 100 Hz. Most auditory cells responded well to directional stimuli presented along axes in the horizontal plane. Cells in M3 (not M2) also responded to lateral line stimulation, consistent with otolithic endorgan and lateral line inputs to M3. The convergence of auditory and lateral line inputs in M3, the lack of Mauthner cells in this species, and previous evidence that the magnocellular nucleus does not contribute to ascending auditory pathways suggest to us that the large cells of M3 may play a role in rapid behavioral responses to particle motion stimuli in oyster toadfish.

Gamma-aminobutyric acid is a neurotransmitter in the auditory pathway of oyster toadfish, Opsanus tau

Binaural computations involving the convergence of excitatory and inhibitory inputs have been proposed to explain directional sharpening and frequency tuning documented in the brainstem of a teleost fish, the oyster toadfish (Opsanus tau). To assess the presence of inhibitory neurons in the ascending auditory circuit, we used a monoclonal antibody to GABA to evaluate immunoreactivity at three levels of the circuit: the first order descending octaval nucleus (DON), the secondary octaval population (dorsal division), and the midbrain torus semicircularis. We observed a subset of immunoreactive (IR) cells and puncta distributed throughout the neuropil at all three locations. To assess whether contralateral inhibition is present, fluorescent dextran crystals were inserted into dorsal DON to fill contralateral, commissural inputs retrogradely prior to GABA immunohistochemistry. GABA-IR somata and puncta co-occurred with retrogradely filled, GABA-negative auditory projection cells. GABA-IR projection cells were more common in the dorsolateral DON than in the dorsomedial DON, but GABA-IR puncta were common in both dorsolateral and dorsomedial divisions. Our findings demonstrate that GABA is present in the ascending auditory circuit in the brainstem of the toadfish, indicating that GABA-mediated inhibition participates in shaping auditory response characteristics in a teleost fish as in other vertebrates.

Physiological evidence for binaural directional computations in the brainstem of the oyster toadfish, Opsanus tau (L.)

Comparisons of left and right auditory input are required for sound source localization in most terrestrial vertebrates. Previous physiological and neuroanatomical studies have indicated that binaural convergence is present in the ascending auditory system of the toadfish. In this study, we introduce a new technique, otolith tipping, to reversibly alter directional auditory input to the central nervous system of a fish. The normal directional response pattern (DRP) was recorded extracellularly for auditory cells in the first-order descending octaval nucleus (DON) or the midbrain torus semicircularis (TS) using particle motion stimuli in the horizontal and mid-sagittal planes. The same stimuli were used during tipping of the saccular otolith to evaluate changes in the DRPs. Post-tipping DRPs were generated and compared with the pre-tipping DRPs to ensure that the data had been collected consistently from the same unit. In the DON, ipsilateral or contralateral tipping most often eliminated spike activity, but changes in spike rate (+/-) and DRP shape were also documented. In the TS, tipping most often caused a change in spike rate (+/-) and altered the shape or best axis of the DRP. The data indicate that there are complex interactions of excitatory and inhibitory inputs in the DON and TS resulting from the convergence of binaural inputs. As in anurans, but unlike other terrestrial vertebrates, binaural processing associated with encoding the direction of a sound source begins in the first-order auditory nucleus of this teleost.

Directional and frequency response characteristics in the descending octaval nucleus of the toadfish (Opsanus tau)

This study is a continuation of a long-term investigation of the auditory circuit in the oyster toadfish, Opsanus tau. Input from the auditory periphery projects to the ipsilateral descending octaval nucleus (DON). Ipsilateral and contralateral DONs project to the auditory midbrain, where a previous study indicated that both frequency tuning and directional sharpening are present. To better understand the transformation of auditory information along the auditory pathway, we have examined over 400 units in the DON to characterize frequency and directional information encoded in the dorsolateral division of the nucleus. Background activity was primarily low (<10 spikes/s) or absent. The maximum coefficient of synchronization was equivalent to the periphery (R = 0.9) and substantially better than in the midbrain. The majority of DON units (79%) responded best to stimulus frequencies of 84-141 Hz and were broadly tuned. DON cells retain or enhance the directional character of their peripheral input (s); however, characteristic axes were distributed in all quadrants around the fish, providing further evidence that binaural computations may first occur in the DON of this species.

Sharpening of directional responses along the auditory pathway of the oyster toadfish, Opsanus tau

Our previous studies have shown that the peripheral auditory system of the toadfish encodes the direction of a sound source. Here, we compare directional responses of peripheral saccular afferents, cells in the descending octaval nucleus (DON) of the medulla, and the torus semicircularis (TS) of the midbrain. Recording locations in the brain were labeled with neurobiotin to confirm the site. To compare directional responses among cells, we calculated an index [sharpening ratio (SR)] that weights the relative strength of responses to the best direction for that cell and to the adjacent stimulus angles tested. Unsharpened saccular afferents tend to have a cosinusoidal directional response pattern (DRP) with an expected SR of 0.87. In DON, more than 60% of the cells exhibited directional sharpening (defined as SR <0.8). In TS, more than 80% of the cells exhibited directional sharpening. We conclude that directional auditory sharpening first occurs in DON and some additional sharpening occurs in the ascending pathway to the midbrain, particularly in azimuth. The sharpening of directional selectivity is likely to be an important component of the neural computations underlying directional hearing.

Fiber connections of the central nucleus of semicircular torus in cyprinids

Fiber connections of the presumed auditory portion of the semicircular torus or the central nucleus (TSc) as well as other central and peripheral auditory structures were studied by tract-tracing methods in carp and goldfish. Major ascending projections to the TSc originated from the descending octaval nucleus (DO) and medullary secondary octaval population (SO). Toropetal neurons of the DO were within the saccular terminal zone. A small number of toropetal DO neurons might receive inputs from other inner ear and lateral line endorgans as well. Fibers from the DO terminated in a deep zone of the TSc, while those from the SO in both deep and superficial zones. Afferents to the TSc also arose from the rhombencephalic reticular formation, anterior octaval nucleus, isthmic reticular nucleus, perilemniscular nucleus, medial pretoral nucleus, anterior tuberal nucleus, and central posterior thalamic nucleus. The TSc projected ascending fibers to the medial pretoral nucleus, anterior tuberal nucleus, central posterior thalamic nucleus, and the preglomerular complex (anterior preglomerular nucleus, the caudomedial region of lateral preglomerular nucleus, and a medial zone of medial preglomerular nucleus). These parts of preglomerular complex appear structurally continuous sharing common hodological and cytoarchitectonic features, and thus might be regarded as a single neuronal population. Abundant descending pathways were also noted in the present study. Of particular note is the medial pretoral nucleus, which received fibers from diencephalic auditory nuclei. The nucleus gave rise to indirect descending pathways to a medial zone of the cerebellar crest where dendrites of crest cells in the SO ramify.

Directional selectivity and frequency tuning of midbrain cells in the oyster toadfish, Opsanus tau

Single-unit recordings were made from areas in the midbrain (torus semicircularis) of the oyster toadfish. We evaluated frequency tuning and directional responses using whole-body oscillation to simulate auditory stimulation by particle motion along axes in the horizontal and mid-sagittal planes. We also tested for bimodality in responses to auditory and hydrodynamic stimuli. One recording location in each animal was marked by a neurobiotin injection to confirm the recording site. Recordings were made in nucleus centralis, nucleus ventrolateralis, and the deep cell layer. Most units were frequency-selective with best frequencies between 50 and 141 Hz. Suppression of activity was apparent in 10% of the cells. Bimodality was common, including inhibition and suppression of background activity by auditory or hydrodynamic stimulation. The majority of the cells were directionally selective with directional response patterns that were sharpened compared with those of primary saccular afferents. The best directional axes were arrayed widely in spherical space, covering most azimuths and elevations. This representation is adequate for the computation of the motional axis of an auditory stimulus for sound source localization.

Brainstem acoustic areas in the marine catfish, Arius felis

The marine catfish Arius felis produces low frequency sounds for communication and obstacle detection. It was hypothesized that the utriculus of the inner ear might play an important role in these behaviors. In the current study, brainstem acoustic areas were studied to reveal possible neuroanatomical specializations in utricular processing areas. The first-order octaval nuclei in Arius were identical in number, anatomical characteristics, and organization of saccular, lagenar, and utricular inputs to previous reports of these features in Ictalurus, a closely related species of catfish that does not exhibit the specialized acoustic behaviors present in Arius. Similarly, injections of neural tracer in the acoustic midbrain (nucleus centralis) of Arius revealed afferent and retrograde pathways almost identical to those previously reported in Ictalurus. It is suggested that areas within the primary and higher-order octaval nuclei that utilize utricular input in acoustic processing are likely identical in Arius and Ictalurus. Two sets of higher-order connections in Arius differ from those in Ictalurus. First, Arius apparently lacks the direct input from the anterior octaval nucleus to nucleus centralis reported in Ictalurus. Second, in Arius nucleus centralis projects bilaterally to a strip of neurons positioned ventral to the ventral boundary of the torus semicircularis. This projection is apparently absent in Ictalurus and in the related species Carassius (goldfish), but has been previously reported in Porichthyes, a sound-producing species belonging to a different teleost taxon.

Acoustic nuclei in the medulla and midbrain of the vocalizing Gulf toadfish (Opsanus beta)

The organization of the descending and secondary octaval nuclei in the hindbrain of the Gulf toadfish, Opsanus beta, was revealed following the injection of biotin compounds into a physiologically identified auditory region of the torus semicircularis. The results show retrogradely-filled neurons mainly in a dorsomedial division of the descending octaval nucleus, and dorsal and ventral divisions of a secondary octaval nucleus; minor labeling also appeared in dorsolateral and rostromedial intermediate divisions of the descending nucleus. The pattern identified is consistent with that reported in other teleosts, including both vocal and non-vocal species, and clarifies earlier reports of the organization of hindbrain octaval nuclei in toadfish and the closely related midshipman fish.

Midbrain acoustic circuitry in a vocalizing fish

The mapping of auditory circuitry and its interface with vocal motor systems is essential to the investigation of the neural processing of acoustic signals and its relationship to sound production. Here we delineate the circuitry of a midbrain auditory center in a vocal fish, the plainfin midshipman. Biotin injections into physiologically identified auditory sites in nucleus centralis (NC) in the torus semicircularis show a medial column of retrogradely filled neurons in the medulla mainly in a dorsomedial division of a descending octaval nucleus (DO), dorsal and ventral divisions of a secondary octaval nucleus (SO), and the reticular formation (RF) near the lateral lemniscus. Biotin-filled neurons are also located at midbrain-pretectal levels in a medial pretoral nucleus. Terminal fields are identified in the medulla (ventral SO, RF), isthmus (nucleus praeeminentialis), midbrain (nucleus of the lateral lemniscus, medial pretoral nucleus, contralateral NC, tectum), diencephalon (lateral preglomerular, central posterior, and anterior tuber nuclei), and telencephalon (area ventralis). The medial column of toral afferent neurons is adjacent to and overlapping the positions of DO and SO neurons shown previously to be linked to the vocal pacemaker circuitry of the medulla. Midshipman are considered "hearing generalists" because they lack the peripheral adaptations of "specialists" that enhance the detection of the pressure component of acoustic signals. Whereas the results indicate a general pattern of acoustic circuitry similar to that of specialists, they also show central adaptations, namely, a vocal-acoustic interface in DO and SO related to this species' vocal abilities.