2017 Nov 15
A Circuit for Detection of Interaural Time Differences in the Nucleus Laminaris of Turtles
Item in Clipboard
A Circuit for Detection of Interaural Time Differences in the Nucleus Laminaris of Turtles
J Exp Biol
The physiological hearing range of turtles is approximately 50-1000 Hz, as determined by cochlear microphonics ( Wever and Vernon, 1956a). These low frequencies can constrain sound localization, particularly in red-eared slider turtles, which are freshwater turtles with small heads and isolated middle ears. To determine if these turtles were sensitive to interaural time differences (ITDs), we investigated the connections and physiology of their auditory brainstem nuclei. Tract tracing experiments showed that cranial nerve VIII bifurcated to terminate in the first-order nucleus magnocellularis (NM) and nucleus angularis (NA), and the NM projected bilaterally to the nucleus laminaris (NL). As the NL received inputs from each side, we developed an isolated head preparation to examine responses to binaural auditory stimulation. Magnocellularis and laminaris units responded to frequencies from 100 to 600 Hz, and phase-locked reliably to the auditory stimulus. Responses from the NL were binaural, and sensitive to ITD. Measures of characteristic delay revealed best ITDs around ±200 μs, and NL neurons typically had characteristic phases close to 0, consistent with binaural excitation. Thus, turtles encode ITDs within their physiological range, and their auditory brainstem nuclei have similar connections and cell types to other reptiles.
Binaural hearing; Interaural time difference; Sound localization; Turtle.
© 2017. Published by The Company of Biologists Ltd.
Conflict of interest statement
Competing interestsThe authors declare no competing or financial interests.
Rostrocaudal location of the auditory brainstem nuclei. Schematic boundaries of the auditory nuclei in transverse section. The section containing only nucleus angularis (NA) is most rostral; the section containing only the nucleus magnocellularis (NM) is most caudal. N8, cranial nerve VIII. Scale bar,100 μm.
Nucleus magnocellularis neurons project bilaterally to the nucleus laminaris. (A) Cresyl Violet-stained transverse section of the acoustic tubercle and the nucleus magnocellularis (NM), outlined in black. (B) Retrogradely labelled NM neurons in rostral NM following injection of neurobiotin into the contralateral acoustic tubercle. NL, nucleus laminaris. (C) Retrogradely labelled NM neuron with an ascending dendrite in the cranial nerve VIII tract. (D) Neurobiotin-labelled NM axons in the internal arcuate tract at the midline, just ventral to the fourth ventricle. All scale bars, 100 μm.
The nucleus laminaris is primarily composed of vertically oriented, bitufted neurons. (A) Cresyl Violet-stained transverse section through the NL (the dashed white line outlines the NL). (B,C) Retrogradely labelled NL neurons after neurobiotin injection into the torus semicircularis (TS). (D–F) Retrogradely labelled NL neurons after horseradish peroxidase injection into the TS. The arrow in E indicates the ventral bend in lateral NL. (F) Bitufted dendritic morphology in retrogradely labelled NL neurons magnified from D. Scale bar, 20 μm. Inset, polar plot of dendritic orientation reveals bitufted organization (values in deg) in both dorsal and ventral dendrites [ N=22 neurons from the medial lamina region (grey), N=17 from the ventral bend of the lateral region of NL (black)]. Each dendrite has a value of one on the unit circle, with dendritic orientation plotted with respect to the lamina of NL cell bodies. All scale bars, 100 μm (except in F).
Summary of auditory circuits. (A) Labelled nucleus magnocellularis (NM; above) and nucleus laminaris (NL; below) neurons from a young turtle, drawn from a single transverse section. Rapid Golgi method (black neurons) with dotted outlines showing positions of unlabelled neurons. The grey line outlines the medial acoustic tubercle. These nuclei are shown in schematic form in B. Scale bar, 100 µm. (B) Circuit diagram summarizing the results of tract tracing experiments. The black arrows mark the auditory nerve input to first-order NM and NA. The blue arrows show the projections of the NM and NL. The superior olive (SO) receives input from NA and NL (not shown). N8, cranial nerve VIII. The red arrow shows the path from the NA to the midbrain. This diagram shows the acoustic tubercle, which is defined as the portion of the brainstem that contains NA, NM and NL.
Physiological properties of NM neurons. (A) Representative constant intensity (80 dB) frequency–rate curves for three NM neurons with best frequencies (BFs) between 200 and 500 Hz. (B) Rate intensity functions of the neurons shown in A, measured at BF, as well as a lower threshold example (grey ×, BF=355 Hz). BFs for symbols in A and B: triangle, 226 Hz; diamond, 424 Hz; square, 475 Hz). (C) Threshold measured at BF in NM single units with ABR audiogram (Christensen-Dalsgaard et al., 2012) and in vivo cochlear nucleus data for comparison (Manley, 1970).
Phase-locking properties of NM neurons. (A) Vector strengths (VS) of single-unit NM neurons; inset: example phase histogram of a NM neuron, BF=150 Hz, VS=0.88. (B) NM post-stimulus time histogram (PSTH) with primary-like discharge pattern, BF=251 Hz. (C) Dot raster plot from 65 repetitions of stimulus to a NM neuron, BF=180 Hz, VS=0.97. (D) PSTH from the same NM neuron in as in C; inset: interval histogram from same unit.
Physiological properties of NL neurons. (A) Response thresholds measured at BF for all monaural (NM; open circles, N=161) and binaural (NL; filled circles, N=18) single units recorded. (B) Vector strengths measured at BF in single units (open circles) and multi-units (filled circles). (C) PSTH of a single NL unit at its BF (400 Hz). (D) Dot raster plot from 65 repetitions of the stimulus in C.
Turtle NL neurons are sensitive to interaural time difference. (A) Response of a NL single unit to varying ITD at 400 Hz. (B) Raster plot from the same unit. (C) ITD responses recorded in NL as a function of stimulus frequency (BF=425 Hz). Interaural delay curves plot the response of this unit for a range of frequencies from 375 to 500 Hz. Equation for characteristic delay (CD) is y=−0.000136 x+0.017; r²=0.97. (D) Period histograms of an ITD-sensitive neuron in response to monaural contralateral (black) and ipsilateral (grey) stimulation at 400 Hz. The difference between ipsilateral (left, black, mean phase 0.220, VS=0.81) and contralateral (right, grey, mean phase 0.218, VS=0.77) phase was −5 µs, while binaural stimulation yielded a best ITD of −44 µs. Inset: distribution of characteristic phase (CP) for all NL single units yields a mean CP of −0.01±0.08 ( N=43), showing that the inputs to the cell at the CD are in phase. (E) Computation of CD for a single unit in left-hand side NL, BF=450 Hz. Each curve was normalized and fitted with a cosine for the nine frequencies tested, to yield a CD of −84 µs. Inset: best IPD (in cycles) determined for each of the curves shown in E, as a function of stimulation frequency (Hz); the continuous line shows the linear regression. The slope of the line represents the CD, and the y-intercept the characteristic phase (CP); y=−0.000084 x+0.012; r²=0.15. (F) If laminaris neurons are excited by input from each side, and act as coincidence detectors, then differences between left and right mean phase should predict best ITD. A regression line was drawn for all data ( y=1.0132 x+56.89; r²=0.62).
The majority of ITDs fall within the predicted physiological range. (A) Section of a response of an NL single unit to varying ITD at 500 Hz (grey curve) and 550 Hz (black curve), measured at 25 µs intervals, with five repetitions (±s.e.m.). The grey rectangle (±150 μs) designates the likely physical range available to a large red-eared slider. (B) Distribution of all best ITDs, shown as if measured from the right-hand side NL, 50 μs bins. (C) Distribution of calculated CDs from all NL single units, as if measured from the right-hand side NL. The grey area (±150 μs) designates the likely physical range available to a turtle, and the black line shows the function (best ITD=0.125/BF) (Grothe et al., 2010).
All figures (9)
A circuit for detection of interaural time differences in the brain stem of the barn owl.
J Neurosci. 1990 Oct;10(10):3227-46. doi: 10.1523/JNEUROSCI.10-10-03227.1990.
J Neurosci. 1990.
2213141 Free PMC article.
Coincidence detection by binaural neurons in the chick brain stem.
J Neurophysiol. 1993 Apr;69(4):1197-211. doi: 10.1152/jn.19220.127.116.117.
J Neurophysiol. 1993.
Change in the coding of interaural time difference along the tonotopic axis of the chicken nucleus laminaris.
Front Neural Circuits. 2015 Aug 20;9:43. doi: 10.3389/fncir.2015.00043. eCollection 2015.
Front Neural Circuits. 2015.
26347616 Free PMC article.
The analysis of interaural time differences in the chick brain stem.
Physiol Behav. 2005 Oct 15;86(3):297-305. doi: 10.1016/j.physbeh.2005.08.003. Epub 2005 Oct 3.
Physiol Behav. 2005.
16202434 Free PMC article.
Neuronal specializations for the processing of interaural difference cues in the chick.
Front Neural Circuits. 2014 May 9;8:47. doi: 10.3389/fncir.2014.00047. eCollection 2014.
Front Neural Circuits. 2014.
24847212 Free PMC article.
Inhibitory Neural Circuits in the Mammalian Auditory Midbrain.
J Exp Neurosci. 2018 Dec 12;12:1179069518818230. doi: 10.1177/1179069518818230. eCollection 2018.
J Exp Neurosci. 2018.
30559596 Free PMC article.
Research Support, N.I.H., Extramural
Brain Stem / physiology
Sound Localization / physiology