Complexin-I is required for high-fidelity transmission at the endbulb of Held auditory synapse - PubMed (original) (raw)
Complexin-I is required for high-fidelity transmission at the endbulb of Held auditory synapse
Nicola Strenzke et al. J Neurosci. 2009.
Abstract
Complexins (CPXs I-IV) presumably act as regulators of the SNARE (soluble N-ethylmaleimide-sensitive factor attachment protein receptor) complex, but their function in the intact mammalian nervous system is not well established. Here, we explored the role of CPXs in the mouse auditory system. Hearing was impaired in CPX I knock-out mice but normal in knock-out mice for CPXs II, III, IV, and III/IV as measured by auditory brainstem responses. Complexins were not detectable in cochlear hair cells but CPX I was expressed in spiral ganglion neurons (SGNs) that give rise to the auditory nerve. Ca(2+)-dependent exocytosis of inner hair cells and sound encoding by SGNs were unaffected in CPX I knock-out mice. In the absence of CPX I, the resting release probability in the endbulb of Held synapses of the auditory nerve fibers with bushy cells in the cochlear nucleus was reduced. As predicted by computational modeling, bushy cells had decreased spike rates at sound onset as well as longer and more variable first spike latencies explaining the abnormal auditory brainstem responses. In addition, we found synaptic transmission to outlast the stimulus at many endbulb of Held synapses in vitro and in vivo, suggesting impaired synchronization of release to stimulus offset. Although sound encoding in the cochlea proceeds in the absence of complexins, CPX I is required for faithful processing of sound onset and offset in the cochlear nucleus.
Figures
Figure 1.
CPX I−/− mice are hearing impaired. A, B, ABR audiograms with average thresholds ± SEM of CPX I−/− (gray) and CPX I+/+ (black) littermate mice. The asterisks denote data points with significant threshold differences (*p < 0.05,**_p_ < 0.001, unpaired _t_ test). A**, Three- to 4-week-old CPX I−/− mice showed a significant but mild increase of ABR threshold in the midfrequency range. B, Hearing impairment of CPX I−/− mice was more prominent at 6–10 weeks of age [n (CPX I−/−) = 7; n (CPX I+/+) = 9]. The empty squares in B represent the mean ASSR threshold for a 12 kHz tone that was sinusoidally amplitude modulated using individually chosen modulation frequencies >400 Hz (see F). C, D, Average ABR waveforms of individual mice in response to 80 dB click stimuli in animals aged 3–4 weeks (C) [n (CPX I−/−) = 6; n (CPX I+/+) = 5] or 6–10 weeks (D**) [n (CPX I−/−) = 8; n (CPX I+/+) = 11]. The bold overlaid traces represent grand averages. E, ASSR, The mean modulation transfer functions (at 80 dB SPL) of 8-week-old CPX I−/− mice (n = 5) showed reduced ASSR amplitudes for all modulation frequencies when compared with the CPX I+/+ littermates (n = 5). F, Amplitude growth functions of ABR wave I and the ASSR peak (individually chosen modulation frequencies between 200 and 400 Hz) of 8-week-old CPX I−/− mice were shifted toward higher stimulation levels but showed comparable slopes as those of the CPX I+/+ littermates. G, DPOAEs at 2f1-f2 were recorded at different primary tone frequencies at stimulus levels of 60 dB and are displayed relative to the noise floor (n = 8 each). H, Growth functions of DPOAE amplitude (n = 7 each; displayed relative to the noise floor) in response to 10/12 kHz primary tones at varying levels. No significant differences in DPOAE levels across frequencies or amplitudes were observed between CPX I−/− and CPX I+/+ mice.
Figure 2.
Analysis of CPX mRNA expression in the cochlea. A, Nested RT-PCR after RNA extraction from brain, retina, and cochlea (manually dissected into modiolus, containing SGNs, and organ of Corti, containing the hair cells). Although CPXs I–IV were readily found in the retina, only CPXs I and II were detected in brain and cochlea after the first round of amplification. After the second round of amplification, we additionally found weak bands for CPX IV but no signal for CPX III in organ of Corti and modiolus. The product sizes of the first round of the PCR are 304, 422, 323, and 300 bp for CPXs I, II, III, and IV, and 127, 292, 202, and 222 bp for the nested PCR. B, The three top panels from the left to right show result of nested single-cell RT-PCR using the primers for CPXs I, II, and IV (PCR products as described in A): None of the investigated IHCs was positive for CPX I, II, or IV RNA. The three bottom panels show corresponding otoferlin cDNA-specific PCR product (217 bp) serving as the positive control of sampling and reverse transcription (detected in each case). No PCR products were observed in the negative controls. The faint bands present in lanes negative for CPX-specific bands and running faster than molecular weight marker represent primers initially added to the PCRs.
Figure 3.
Distribution of CPX I and II protein in the cochlea. The inset of A sketches the afferent innervation of IHCs (white, nucleus stained red) and OHCs by type I (white) and type II (black) SGNs in the organ of Corti. Each of the type I SGNs possesses one peripheral axon, which forms one ribbon synapse with one IHC. Type II SGNs form synapses with several OHCs (not considered in this study). A, B, Representative projections of confocal sections of wild-type organs of Corti that were costained for CPX I/II (green) and CtBP2/RIBEYE (red) at different magnifications. The image represents a projection of confocal sections. The row of IHCs was devoid of a cytosolic CPX I/II immunoreactivity. CPX-positive fibers targeted the row of IHCs and some spot-like immunoreactivity was seen underneath the OHCs (A). CPX-positive fiber endings were juxtaposed to the RIBEYE-labeled ribbons at their contact with the IHCs. No overlap of CPX and RIBEYE immunofluorescence (B). C, Representative projection of confocal sections of a wild-type organ of Corti costained for CPX I/II (green) and synaptophysin (red; efferent synapse marker): juxtaposition rather than merge indicates that CPX resides within afferent SGNs. D, E, Cryosections of cochleae (left, middle panels) and retinas (right panels) of 17-d-old wild-type mice were processed as comparably as possible and stained on the same glass slide for CPX III (D) or IV (E) and otoferlin (IHC marker). The left and right panels show overlay of immunofluorescence for otoferlin (red) and the respective CPX (green). The middle panels show CPX immunofluorescence only. The arrowheads point to IHC in the cochlea sections. OPL, Outer plexiform layer; IPL, inner plexiform layer in the retina sections. The images were acquired with identical microscope settings. F, G, CPX I–II-labeled organs of Corti (projections of confocal sections) (F) and spiral ganglions (single confocal sections) (G) of WT (left panels), CPX I−/− (middle panels), and CPX II−/− (right panels) mice. In F, the colabeling for NF-200 is shown. Experiments were performed in parallel using identical conditions for staining and imaging. To illustrate the weak staining in CPX I−/− SGN somata (G), confocal microscope settings were chosen that resulted in some saturation in the images of wild-type and CPX II−/− SGNs.
Figure 4.
Normal presynaptic function of IHCs of 8-week-old CPX I KO mice. A, Representative Ca2+ current (I Ca 2+; bottom panel) and background-subtracted capacitance traces (C m; top panel) recorded from CPX I−/− and CPX I+/+ IHCs stimulated by a 20 ms step depolarization. B, Testing fast (responses to 5 and 20 ms) and sustained (50 ms stimuli) exocytosis in IHC of CPX I+/+ (n = 4 cells) and CPX I−/− (n = 5 cells) mice: C m increments and peak Ca2+ current amplitudes did not differ significantly. C, D, Representative projections of confocal sections obtained from organs of Corti from CPX I+/+ (C) and CPX I−/− (D) mice after staining with antibodies to CtBP2/RIBEYE (red) and glutamate receptors (GluR2/3; green). The line was drawn to illustrate the approximate position of one of the IHCs. Scale bar, 5 μm. E, Representative auditory nerve fiber recordings from CPX I+/+ (black) and CPX I−/− (red) mice illustrating sound-evoked (50 ms tone burst at CF; 30 dB re threshold; 200 repetitions) and spontaneous (50 ms silence) spikes with good (left) and acceptable (right) amplitudes (5 repetitions each). Recordings with inferior signal-to-noise ratios were excluded. F, Average PSTH ± SEM, bin width of 0.5 ms for CPX I−/− (red; n = 21), CPX I+/+ (black; n = 16), and C57BL/6 wild-type (gray; n = 13) mouse auditory nerve fibers for stimulation with 50 ms tone bursts at CF at 30 dB re threshold showing comparable spike rates for both genotypes. G, Thresholds at CF for CPX I−/− (red; n = 31), CPX I+/+ (black; n = 19), and C57BL/6 wild-type (gray; n = 26) auditory nerve fibers.
Figure 5.
Distribution of CPX I and II protein in the cochlear nucleus. A, B, CPX I is expressed in the cochlear nucleus. Brainstem cryosections of 8-week-old mice were stained using antibodies against CPX I/II (green) and VGLUT1 (red). The VCN showed CPX I/II immunoreactivity in CPX I+/+ (A), but the staining was strongly reduced in CPX I−/− (B) sections. AN, Auditory nerve; CB, cerebellum; sptV, spinal tract of the trigeminal nerve; OPL, outer plexiform layer; IPL, inner plexiform layer. C, D, Higher-resolution confocal images of the auditory nerve entry zone into the cochlear nucleus of 8-week-old CPX I+/+ (C) and CPX I−/− (D) mice. In the wild type, CPX I/II immunofluorescent principal cells of the AVCN are engulfed by a ring-like array of CPX- and VGLUT1-positive spots (white arrowheads). In both CPX I+/+ and CPX I−/−, there were some CPX I/II-positive spots that were not costained with VGLUT1, presumably representing CPX II-expressing inhibitory synapses (gray arrowheads). The images in A,B and C,D were acquired at the same settings to illustrate the much-reduced CPX I/II immunofluorescence in the presynaptic and postsynaptic compartments of the CPX I−/− cochlear nucleus. E–J, Absence of CPX III and CPX IV in the cochlear nucleus. Brainstem cryosections of 8-week-old CPX I+/+ (E, G) and CPX I−/− (F, H) mice were stained using antibodies against CPX III (green; E, F) or CPX IV (green; G, H) and VGLUT1 (red). I, J, Cryosections of the retina of a CPX I−/− animal processed together with E–H and imaged with the same settings to demonstrate the absence of CPX III and IV immunofluorescence in the brainstem.
Figure 6.
CPX I has a positive role in synaptic transmission from spiral ganglion neurons to bushy cells in the AVCN—slice physiology. Data are from patch-clamp recordings in brain slices from CPX I+/+ or CPX I+/− (black) and CPX I−/− (gray) mice. A, Left, Representative EPSCs measured in CPX I+/+ and CPX I−/− slices. These EPSCs are also shown scaled to the same peak height, showing no effect of CPX I on EPSC time course. Right, Cumulative histograms of EPSC amplitude for 11 CPX I+/+ cells and 20 CPX I−/− cells. EPSCs are significantly smaller in CPX I−/− cells (see Results). B, Effect of CPX I on average mEPSC frequency (left) and amplitude (right). Data are shown as cumulative histograms of 25 CPX I+/+ and 21 CPX I−/− cells. Frequency is significantly lower, but amplitude is unchanged (see Results). C, CPX I−/− endbulbs have a higher paired-pulse ratio (PPR). Data are averages of six CPX I+/+ and nine CPX I−/− cells. The lines here and in D are fits to the data using a model of short-term plasticity at the endbulb (supplemental data 4, available at
as supplemental material), in which all model parameters were held constant, except for the initial probability of release. D, CPX I ablation affects short-term plasticity during high-frequency firing. The top traces are representative EPSCs in response to trains of 20 stimuli at 100 Hz in CPX I+/+ or CPX I−/− cells. The bottom graph summarizes average EPSC amplitudes throughout the train normalized to the amplitude of the first EPSC (EPSC1) for CPX I+/+ cells (n = 8) and CPX I−/− cells (top gray symbols; n = 18). To compare absolute amplitudes, the CPX I−/− data were scaled by the ratio of the average EPSC1 in CPX I−/− to the average EPSC1 in CPX I+/+, yielding the bottom gray symbols. This shows that EPSC amplitudes are similar after the second pulse in the train. E, Effect of CPX I on RRP (left) and initial P r (right). Data are shown as cumulative histograms for 8 CPX I+/+ cells and 17 CPX I−/− cells. RRP is not significantly different in CPX I−/−, but P r is significantly lower (see Results). F, Example traces and spike rasters showing effects of CPX I−/− on bushy cell firing in response to 200 Hz auditory nerve fiber stimulation. The vertical marks indicate stimulus times. The dashed line in the rasters indicates the start of the stimulus train. G, Effects of CPX I on bushy cell firing. Early in trains, the probability of spiking is greatly reduced (top). The latency (middle) and jitter (bottom) were also slightly greater. Data are averages of 7 CPX I+/− and 12 CPX I−/− cells. Error bars indicate SEM. The asterisks indicate points that are significantly different (p ≤ 0.05) between the two strains for at least two of the three frequencies analyzed (for details, see supplemental Table S1, available at
as supplemental material).
Figure 7.
Delayed release is enhanced in CPX I−/− endbulbs. A, Example voltage-clamp traces in CPX I+/+ and CPX I−/− bushy cells, showing increased delayed release after 333 Hz stimulation. The insets are magnified fivefold for the time period after the train. The horizontal calibration applies to all traces in A and D. B, Average time course of delayed release in the first 100 ms after the end of 20 pulse, high-frequency trains. The symbols are as in Figure 6, D and G. Data are averages of eight (CPX I+/+) or nine (CPX I−/−) experiments. C, Total delayed release integrated over the first 100 ms after a train. Delayed release increases with stimulation frequency. Delayed release in CPX I−/− endbulbs is significantly greater than in CPX I+/+ for all frequencies (p < 0.01). D, Example current-clamp traces in CPX I+/− and CPX I−/− bushy cells showing increased delayed release, as well as misplaced spikes (inverted triangle). The dashed line corresponds to a membrane voltage of −60 mV. E, Average number of misplaced spikes, measured over the first 40 ms after the train. No misplaced spikes were detected in any CPX I+/− endbulbs (n = 7 cells), and CPX I−/− endbulbs showed considerable variability (n = 11 cells). Misplaced spikes were significantly elevated for 100 and 333 Hz cases (p < 0.05). Error bars indicate SEM.
Figure 8.
CPX I has a role in synaptic transmission from spiral ganglion neurons to neurons of the ventral cochlear nucleus—in vivo experiments. A, Sound thresholds of AVCN neurons (bushy and multipolar neurons) at their CF: threshold elevation in CPX I-deficient mutants. B, Dot raster displays of CPX I+/+ and CPX I−/− bushy cells in response to 100 ms tone bursts (at CF and 80 dB SPL, 50 stimulus repetitions: see schematic representation in the top panel). The example cells were chosen to represent cases of narrow FSLs distributions (right column) and broad FSL distributions (left column). The FSL distributions are shown in E, top panel. Bottom panel, Average PSTH of CPX I+/+ (black) and CPX I−/− (gray) bushy cells (bin width, 1 ms). The pronounced onset activity seen in the initial peak in the CPX I+/+ cell was reduced and delayed in the CPX I−/− cells (left arrow). Moreover, activity of many CPX I−/− cells outlasted the stimulus offset, which is in stark contrast to the poststimulus reduction in spontaneous rate typically observed in CPX I+/+ (right arrow). C, Spontaneous and sound-evoked firing rates of CPX I−/− and CPX I+/+ bushy cells were recorded before, during, and after a 100-ms-long stimulus. Average rates were calculated for the whole stimulation time of 100 ms, onset rates were averaged over 10 ms beginning at the respective FSL of the unit, adapted rates were acquired during the plateau phase of the PSTH (50–100 ms), and offset rates were averaged over 10 ms after the end of the stimulus. Error bars indicate SEM. D, Distribution of FSLs for CPX I−/− and CPX I+/+ bushy cells: averaged from all cells in each groups after normalization for the number of trials: delayed mean latency and broader distribution in CPX I−/− bushy cells (bin width, 2 ms; each cell contributes the same area to the histogram regardless of the number of trials). E, Examples of FSL distributions from individual bushy cells (top panels) and corresponding simulated distributions (bottom panels) created to match experiments. Wild-type cells (open bars) in the top panels are chosen as typical examples from the group of cells with very high FSL jitter (left) and very low FSL jitter (right) (see Results). The model was used to approximate the experimental data: the sharp distribution was reproduced when assuming subthreshold input from 20 SGNs (for details, see supplemental material, available at
). The broader distribution could be reproduced assuming input from only two SGNs, each of which could drive AP generation in the bushy cell. The gray bars in the bottom panel result from models with identical input but reduced initial release probability at the endbulb of Held. This resembles the effects of CPX I ablation and leads in the model to FSL distributions that resemble experimental findings on mutant bushy cells (gray bars, top panels). F, Analysis of FSLs in responses to sounds of 80 dB SPL: increased latency and jitter in VCN units of CPX I−/− mice.
Figure 9.
Flow chart of the computational modeling. The model consists of several modules, each mimicking the behavior of a part of the auditory pathway, including inner hair cell synaptic transmission and triggering of action potentials in SGNs (step 1), synaptic plasticity at the endbulb of Held (step 2), and the synaptically driven action potential generation in bushy cells (step 3). The figures show the output of the individual steps for a single run of an example configuration: three SGNs with a low spontaneous rate and one SGN with a high spontaneous rate driving one wild-type bushy cell. EPSCamp was set to 3.2 times the threshold for action potential initiation at rest (i.e., suprathreshold fibers).
Similar articles
- Endbulb synaptic depression within the range of presynaptic spontaneous firing and its impact on the firing reliability of cochlear nucleus bushy neurons.
Wang Y, Ren C, Manis PB. Wang Y, et al. Hear Res. 2010 Dec 1;270(1-2):101-9. doi: 10.1016/j.heares.2010.09.003. Epub 2010 Sep 17. Hear Res. 2010. PMID: 20850512 Free PMC article. - Synaptic transmission at the cochlear nucleus endbulb synapse during age-related hearing loss in mice.
Wang Y, Manis PB. Wang Y, et al. J Neurophysiol. 2005 Sep;94(3):1814-24. doi: 10.1152/jn.00374.2005. Epub 2005 May 18. J Neurophysiol. 2005. PMID: 15901757 Free PMC article. - Role of GluA3 AMPA Receptor Subunits in the Presynaptic and Postsynaptic Maturation of Synaptic Transmission and Plasticity of Endbulb-Bushy Cell Synapses in the Cochlear Nucleus.
Antunes FM, Rubio ME, Kandler K. Antunes FM, et al. J Neurosci. 2020 Mar 18;40(12):2471-2484. doi: 10.1523/JNEUROSCI.2573-19.2020. Epub 2020 Feb 12. J Neurosci. 2020. PMID: 32051325 Free PMC article. - Molecularly and structurally distinct synapses mediate reliable encoding and processing of auditory information.
Wichmann C. Wichmann C. Hear Res. 2015 Dec;330(Pt B):178-90. doi: 10.1016/j.heares.2015.07.008. Epub 2015 Jul 15. Hear Res. 2015. PMID: 26188105 Review. - The auditory hair cell ribbon synapse: from assembly to function.
Safieddine S, El-Amraoui A, Petit C. Safieddine S, et al. Annu Rev Neurosci. 2012;35:509-28. doi: 10.1146/annurev-neuro-061010-113705. Annu Rev Neurosci. 2012. PMID: 22715884 Review.
Cited by
- DMXL2 Is Required for Endocytosis and Recycling of Synaptic Vesicles in Auditory Hair Cells.
Peng H, Wang L, Gao Y, Liu H, Lin G, Kong Y, Xu P, Liu H, Yuan Q, Liu H, Song L, Yang T, Wu H. Peng H, et al. J Neurosci. 2024 Sep 18;44(38):e1405232024. doi: 10.1523/JNEUROSCI.1405-23.2024. J Neurosci. 2024. PMID: 39147590 Free PMC article. - Complexin has a dual synaptic function as checkpoint protein in vesicle priming and as a promoter of vesicle fusion.
López-Murcia FJ, Lin KH, Berns MMM, Ranjan M, Lipstein N, Neher E, Brose N, Reim K, Taschenberger H. López-Murcia FJ, et al. Proc Natl Acad Sci U S A. 2024 Apr 9;121(15):e2320505121. doi: 10.1073/pnas.2320505121. Epub 2024 Apr 3. Proc Natl Acad Sci U S A. 2024. PMID: 38568977 Free PMC article. - Probing the role of the C2F domain of otoferlin.
Chen H, Fang Q, Benseler F, Brose N, Moser T. Chen H, et al. Front Mol Neurosci. 2023 Dec 12;16:1299509. doi: 10.3389/fnmol.2023.1299509. eCollection 2023. Front Mol Neurosci. 2023. PMID: 38152587 Free PMC article. - In-depth proteome analysis of brain tissue from Ewsr1 knockout mouse by multiplexed isobaric tandem mass tag labeling.
Jung JW, Kim H, Park J, Woo J, Jeon E, Lee G, Park M, Kim S, Seo H, Cheon S, Dan K, Lee J, Ryu H, Han D. Jung JW, et al. Sci Rep. 2023 Sep 14;13(1):15261. doi: 10.1038/s41598-023-42161-7. Sci Rep. 2023. PMID: 37709831 Free PMC article. - Complexins: Ubiquitously Expressed Presynaptic Regulators of SNARE-Mediated Synaptic Vesicle Fusion.
López-Murcia FJ, Reim K, Taschenberger H. López-Murcia FJ, et al. Adv Neurobiol. 2023;33:255-285. doi: 10.1007/978-3-031-34229-5_10. Adv Neurobiol. 2023. PMID: 37615870
References
- Brose N. For better or for worse: complexins regulate SNARE function and vesicle fusion. Traffic. 2008;9:1403–1413. - PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Molecular Biology Databases
Research Materials
Miscellaneous