Chagnaud BP et al. (2015), Spinal corollary discharge modulates motion sen...

XB-ART-51225

Nat Commun 2015 Sep 04;6:7982. doi: 10.1038/ncomms8982.

Show Gene links Show Anatomy links

Spinal corollary discharge modulates motion sensing during vertebrate locomotion.

Chagnaud BP , Banchi R , Simmers J , Straka H .

???displayArticle.abstract???
During active movements, neural replicas of the underlying motor commands may assist in adapting motion-detecting sensory systems to an animal's own behaviour. The transmission of such motor efference copies to the mechanosensory periphery offers a potential predictive substrate for diminishing sensory responsiveness to self-motion during vertebrate locomotion. Here, using semi-isolated in vitro preparations of larval Xenopus, we demonstrate that shared efferent neural pathways to hair cells of vestibular endorgans and lateral line neuromasts express cyclic impulse bursts during swimming that are directly driven by spinal locomotor circuitry. Despite common efferent innervation and discharge patterns, afferent signal encoding at the two mechanosensory peripheries is influenced differentially by efference copy signals, reflecting the different organization of body/water motion-detecting processes in the vestibular and lateral line systems. The resultant overall gain reduction in sensory signal encoding in both cases, which likely prevents overstimulation, constitutes an adjustment to increased stimulus magnitudes during locomotion.

???displayArticle.pubmedLink??? 26337184
???displayArticle.pmcLink??? PMC4569702
???displayArticle.link??? Nat Commun

Species referenced: Xenopus laevis
Genes referenced: gnl3

???attribute.lit??? ???displayArticles.show???

	Figure 1. Locomotor-related neural activity in vestibular nerve efferent neurons in Xenopus tadpoles.(a–d) Episodes of spontaneous fictive swimming in semi-isolated in vitro preparations (a), recorded as multiple-unit impulse discharge (b–d) in the left (ipsilateral) and right (contralateral) ventral roots (i-vr and c-vr, respectively; black traces) of spinal segment 14 together with the central cut portion of the left anterior vestibular (VIIIth) nerve branch (AVN, red trace). The initial discharge at episode onset () and subsequent regular (*) vr bursting (shaded areas in b) are shown on an extended timescale in c and d, respectively. After mostly tonic firing at swim episode onset (c), the AVN activity develops into rhythmic bursting occurring in phase with locomotor bursts in the ipsilateral vr (red dashed lines in d). (e) Different preparation showing coincident burst coupling between ipsilateral vr11 and the posterior vestibular nerve (PVN) branch (blue dashed lines) during an episode of fictive swimming. (f) Polar plot quantifying the phase relationship between the i-vr/AVN and i-vr/PVN activity shown in d and e; AVN (red area) and PVN bursts (blue area) are approximately in phase (angle towards 0°) with the i-vr burst rhythm. Calibration bars: 5 s in b, 1 s in c, 0.2 s in d and e.
	Figure 2. Ca2+ imaging of morphologically identified mechanosensory efferent neurons (ENs) during fictive locomotion.(a–d) Confocal reconstructions of hindbrain whole mounts after combined application of Alexa Fluor 546 (red) and 488 (green) dextran to the AVN and ALLN (a,c,d) and to the AVN and PLLN, respectively (b) showing afferent axonal projections and locations of EN somata in rhombomeres (r) 4 and 6. Note that the longitudinal and rhombomere-specific transverse fibres (blue in a) were visualized by 612 nm illumination. The inset in a shows r4 and r6 ALLN ENs (green) in relation to segmental boundaries (dashed lines). (c) AVN (red) and ALLN (green) ENs in r4 at higher magnification; note the double-labelled neurons in yellow (arrowheads). (d) ALLN ENs in r4 and r6 that extend dendrites (arrowheads) across the midline (ml). (e,f) Imaging of Ca2+ transients in ENs of semi-isolated preparations (e) following retrograde loading of cell bodies with Calcium Green-1 dextran from the anterior vestibular nerve (AVN). Ca2+ transients (f) were recorded simultaneously in several ENs (colour-coded cells and traces) during episodes of evoked (red ) and spontaneous (green ) ventral root/myotomal locomotor burst activity (black traces). (g,h) Correlation between Ca2+ dynamics, measured as the overall response half-width (g) and corresponding locomotor episode duration (h) in 32 cells during 4–7 swimming episodes per monitored cell (n=160). (i) Plot of Ca2+-response durations (measured as half-width; see colour-coded, normalized transients in inset) of 25 EN pairs during 4–7 locomotor episodes per cell (n=114). Red lines in h and i represent linear regression. Calibration bars: 0.1 mm in a and b and inset in a; 25 μm in c and d; 15 μm in f; 5 s for traces in f and inset in i; 0.2 s in g.
	Figure 3. Parameter representation of locomotor activity in EN discharge.(a) Recordings of AVN (red) and c-vr15 activity (black) during swimming episodes of increasing length in the same preparation. (b) Pooled data plot showing matching episode durations of vestibular (red, n=7 preparations) and lateral line (blue, n=7 preparations) EN versus vr discharge (68 episodes; black line: linear regression). (c) Recordings of AVN (red) and c-vr14 activity (grey) with corresponding integrals of intra-burst firing rates (pink and black traces) during sequences of strong (left) and weak fictive swimming (right) within the same episode (compare colour-shaded areas). During strong swimming (left), additional spikes occurred in the AVN (blue ) in phase with the contralateral vr. (d) Group data plot showing a close correlation between the magnitudes of vestibular (red circles, n=6 preparations) and lateral line (blue circles, n=6) EN versus vr burst integrals (313 burst cycles; black line: linear regression,). (e) Recording of AVN (red trace) and i-vr13 activity (grey trace) with respective integrals of intra-burst firing rates (pink and black traces) during a swim episode where vr bursting changed spontaneously from a slower (2.1 Hz) to a faster (4.4 Hz) rhythm. (f) Group plot showing a close correlation between vestibular (red, n=6) and lateral line (blue, n=6) EN versus vr burst frequencies (101 cycles of 5–10 bursts per episode). Note that any biphasic EN burst patterns were omitted from this analysis (black line: linear regression). (g) Recording of AVN (red) and c-vr15 activity (grey) with corresponding firing rate integrals (pink and black) during a swimming episode in which the single-phase vr-EN coupling (red in g) followed a pattern of EN activity occurring in time with the vr bursts on both sides (red and blue * in g). (h) Box and whisker plots showing that the biphasic EN activity (blue) occurred with vr bursts of significantly larger relative magnitude (**P≤0.001; Mann–Whitney U-test) than during monophasic EN-vr coupling. Number (n) of preparations is indicated in b,d,f and h. Vertical calibration bars in c,e and g indicate a discharge rate of 100 spikes per second.
	Figure 4. Spinal origin and trajectory of ascending pathways mediating locomotor corollary discharge signalling to ENs.(a–i) Episodes of spontaneous fictive swimming in the same in vitro preparation (a,d,g) recorded from the right (contralateral, c) vr16 (black trace) and the central stump of the left AVN (b,e and h: red, blue and green traces, respectively) in control (a–c) after midbrain removal (d–f) and then after a right obex hemisection (g–i) The corresponding polar plots in c, f and i show that the out-of-phase (contralaterally timed) vr-EN coupling remained largely unaffected by the two lesions. However, although the additional synchronous (ipsilaterally timed) EN activity (b,c) persisted after midbrain removal (e,f), it disappeared after obex hemisection (pink arrowheads in h and pink area/arrowhead in i). (j–l) Episodes of spontaneous swimming activity in a different semi-isolated preparation (j) recorded simultaneously from right vrs 3, 10 and 16 (respectively, black, dark-grey and light-grey traces) and the central stump of the left PLLN (red trace). PLLN bursting coupled with locomotor bursts in all three cord segments (k) disappeared (l; red *) when CPG burst activity ceased spontaneously in vr3 and vr10, but persisted in vr16. Calibration bars: 0.2 s in b,e,h,k and l.
	Figure 5. Locomotor corollary discharge influence on lateral line sensory encoding.(a–c) Simultaneous recordings of right vr15 (black traces) and en passant recordings of afferent fibres in the left PLLN (red traces) of a semi-isolated preparation with intact neuromast–lateral line nerve connectivity (a) in the absence (b) and presence (c) of constant sensory stimulation by Ringer flow along the skin surface. During episodes of fictive swimming (shaded areas in b and c; see black vr trace), tonic PLLN activity was initially abolished (* in b and c), but regained lower levels of firing as the locomotor activity progressed and eventually returned to resting discharge levels before episode termination. (d,e) Plots illustrating a decrease (d, red) or absence of locomotor influence (e, blue) on the firing of individual lateral line afferents during CPG activity. (f) Histograms (error bars represent s.e.m.) showing significant average decreases in firing rates of the recorded afferent population (number indicated in each bar) during the initial phase (I) and throughout the entire (E) duration of fictive swimming episodes. P≤0.001; *P≤0.0001 (Wilcoxon signed-rank test). (g,h) Distinguishable afferent and efferent fibre activity in a single PLLN (g) recorded en passant from one branch (PLLN1, red trace) connected to lateral line hair cells and from a second transected branch (PLLN2, blue trace) of the same nerve (h). During locomotor-related efferent activity (shaded area; indicated by the multiple-unit discharge in PLLN2 and spiking in the single small unit (arrow) in PLLN1), ongoing afferent firing (large spikes in PLLN1) was virtually suppressed throughout the swim episode. Horizontal calibration bars: 5 s in b and c and 1 s in h.
	Figure 6. Locomotor corollary discharge influence on vestibular sensory encoding.(a–h) Recordings of right vr12 (black traces in b and f), the left ALLN (green trace in f) and en passant recordings of afferent fibres in the left AVN (red traces in b and f) in semi-isolated preparations with intact inner ear hair cell–vestibular nerve connectivity (a,e) during sinusoidal (1 Hz, ±60° s−1) horizontal-axis roll motion (a) or vertical-axis head rotations (e) imposed by a two-axis turntable (Tpos). During fictive swimming (shaded areas in b and f), the firing of some vestibular afferent fibres was attenuated (b–d), but facilitated in others (f–h). (c,d,g,h) Histogram (black bars) depicting the mean afferent firing rate modulation (responses of fibres recorded in b and f) over a single cycle of turntable motion (dashed lines) in the absence (c,g) and presence (d,h) of locomotor CPG activity; also plotted are the respective population averages (red curves) of fibres with decreasing (c,d) and increasing firing (g,h) during swimming activity. (i) Plots of individual mean firing rate alterations during motion stimulation (green—increase, red—decrease and black—no change) before (control) and during swimming activity. Grey and red boxes with whisker plots show the distributions of the average firing rates in the two conditions. NS, not significant. (j,k) Averaged response of all recorded afferent fibres (±s.e.m., shaded areas in j; n=22) over a single cycle of turntable motion (dashed grey line) before (black plot) and during (red plot) locomotor CPG activity (j). (k) Averaged extent (error bars represent s.e.m.) of firing rate modulation before (control) and during fictive swimming. (k) **P≤0.001 (Wilcoxon signed-rank test). Horizontal calibration bar in b and f: 1 s.
	Figure 7. Schematics summarizing the functional impact of locomotor corollary discharge on lateral line and vestibular mechanosensory signalling.(a) Connectivity between spinal locomotor CPG circuitry, mechanosensory efferent neurons (ENs) and the sensory periphery along with afferent projections to central vestibular and lateral line nuclei. (b,c) Influence of locomotor activity conveyed by ENs to lateral line (b) and vestibular (c) hair cells/afferent innervation on stimulus signal encoding. The afferent responsiveness to either constant water motion (b—left) or cyclic head rotation (c—left) is substantially attenuated (b,c—right) during concomitant locomotor CPG activation.

References [+] :

Angelaki, Vestibular system: the many facets of a multimodal sense. 2008, Pubmed

	Figure 1. Locomotor-related neural activity in vestibular nerve efferent neurons in Xenopus tadpoles.(a–d) Episodes of spontaneous fictive swimming in semi-isolated in vitro preparations (a), recorded as multiple-unit impulse discharge (b–d) in the left (ipsilateral) and right (contralateral) ventral roots (i-vr and c-vr, respectively; black traces) of spinal segment 14 together with the central cut portion of the left anterior vestibular (VIIIth) nerve branch (AVN, red trace). The initial discharge at episode onset () and subsequent regular (*) vr bursting (shaded areas in b) are shown on an extended timescale in c and d, respectively. After mostly tonic firing at swim episode onset (c), the AVN activity develops into rhythmic bursting occurring in phase with locomotor bursts in the ipsilateral vr (red dashed lines in d). (e) Different preparation showing coincident burst coupling between ipsilateral vr11 and the posterior vestibular nerve (PVN) branch (blue dashed lines) during an episode of fictive swimming. (f) Polar plot quantifying the phase relationship between the i-vr/AVN and i-vr/PVN activity shown in d and e; AVN (red area) and PVN bursts (blue area) are approximately in phase (angle towards 0°) with the i-vr burst rhythm. Calibration bars: 5 s in b, 1 s in c, 0.2 s in d and e.
	Figure 2. Ca2+ imaging of morphologically identified mechanosensory efferent neurons (ENs) during fictive locomotion.(a–d) Confocal reconstructions of hindbrain whole mounts after combined application of Alexa Fluor 546 (red) and 488 (green) dextran to the AVN and ALLN (a,c,d) and to the AVN and PLLN, respectively (b) showing afferent axonal projections and locations of EN somata in rhombomeres (r) 4 and 6. Note that the longitudinal and rhombomere-specific transverse fibres (blue in a) were visualized by 612 nm illumination. The inset in a shows r4 and r6 ALLN ENs (green) in relation to segmental boundaries (dashed lines). (c) AVN (red) and ALLN (green) ENs in r4 at higher magnification; note the double-labelled neurons in yellow (arrowheads). (d) ALLN ENs in r4 and r6 that extend dendrites (arrowheads) across the midline (ml). (e,f) Imaging of Ca2+ transients in ENs of semi-isolated preparations (e) following retrograde loading of cell bodies with Calcium Green-1 dextran from the anterior vestibular nerve (AVN). Ca2+ transients (f) were recorded simultaneously in several ENs (colour-coded cells and traces) during episodes of evoked (red ) and spontaneous (green ) ventral root/myotomal locomotor burst activity (black traces). (g,h) Correlation between Ca2+ dynamics, measured as the overall response half-width (g) and corresponding locomotor episode duration (h) in 32 cells during 4–7 swimming episodes per monitored cell (n=160). (i) Plot of Ca2+-response durations (measured as half-width; see colour-coded, normalized transients in inset) of 25 EN pairs during 4–7 locomotor episodes per cell (n=114). Red lines in h and i represent linear regression. Calibration bars: 0.1 mm in a and b and inset in a; 25 μm in c and d; 15 μm in f; 5 s for traces in f and inset in i; 0.2 s in g.
	Figure 3. Parameter representation of locomotor activity in EN discharge.(a) Recordings of AVN (red) and c-vr15 activity (black) during swimming episodes of increasing length in the same preparation. (b) Pooled data plot showing matching episode durations of vestibular (red, n=7 preparations) and lateral line (blue, n=7 preparations) EN versus vr discharge (68 episodes; black line: linear regression). (c) Recordings of AVN (red) and c-vr14 activity (grey) with corresponding integrals of intra-burst firing rates (pink and black traces) during sequences of strong (left) and weak fictive swimming (right) within the same episode (compare colour-shaded areas). During strong swimming (left), additional spikes occurred in the AVN (blue ) in phase with the contralateral vr. (d) Group data plot showing a close correlation between the magnitudes of vestibular (red circles, n=6 preparations) and lateral line (blue circles, n=6) EN versus vr burst integrals (313 burst cycles; black line: linear regression,). (e) Recording of AVN (red trace) and i-vr13 activity (grey trace) with respective integrals of intra-burst firing rates (pink and black traces) during a swim episode where vr bursting changed spontaneously from a slower (2.1 Hz) to a faster (4.4 Hz) rhythm. (f) Group plot showing a close correlation between vestibular (red, n=6) and lateral line (blue, n=6) EN versus vr burst frequencies (101 cycles of 5–10 bursts per episode). Note that any biphasic EN burst patterns were omitted from this analysis (black line: linear regression). (g) Recording of AVN (red) and c-vr15 activity (grey) with corresponding firing rate integrals (pink and black) during a swimming episode in which the single-phase vr-EN coupling (red in g) followed a pattern of EN activity occurring in time with the vr bursts on both sides (red and blue * in g). (h) Box and whisker plots showing that the biphasic EN activity (blue) occurred with vr bursts of significantly larger relative magnitude (**P≤0.001; Mann–Whitney U-test) than during monophasic EN-vr coupling. Number (n) of preparations is indicated in b,d,f and h. Vertical calibration bars in c,e and g indicate a discharge rate of 100 spikes per second.
	Figure 4. Spinal origin and trajectory of ascending pathways mediating locomotor corollary discharge signalling to ENs.(a–i) Episodes of spontaneous fictive swimming in the same in vitro preparation (a,d,g) recorded from the right (contralateral, c) vr16 (black trace) and the central stump of the left AVN (b,e and h: red, blue and green traces, respectively) in control (a–c) after midbrain removal (d–f) and then after a right obex hemisection (g–i) The corresponding polar plots in c, f and i show that the out-of-phase (contralaterally timed) vr-EN coupling remained largely unaffected by the two lesions. However, although the additional synchronous (ipsilaterally timed) EN activity (b,c) persisted after midbrain removal (e,f), it disappeared after obex hemisection (pink arrowheads in h and pink area/arrowhead in i). (j–l) Episodes of spontaneous swimming activity in a different semi-isolated preparation (j) recorded simultaneously from right vrs 3, 10 and 16 (respectively, black, dark-grey and light-grey traces) and the central stump of the left PLLN (red trace). PLLN bursting coupled with locomotor bursts in all three cord segments (k) disappeared (l; red *) when CPG burst activity ceased spontaneously in vr3 and vr10, but persisted in vr16. Calibration bars: 0.2 s in b,e,h,k and l.
	Figure 5. Locomotor corollary discharge influence on lateral line sensory encoding.(a–c) Simultaneous recordings of right vr15 (black traces) and en passant recordings of afferent fibres in the left PLLN (red traces) of a semi-isolated preparation with intact neuromast–lateral line nerve connectivity (a) in the absence (b) and presence (c) of constant sensory stimulation by Ringer flow along the skin surface. During episodes of fictive swimming (shaded areas in b and c; see black vr trace), tonic PLLN activity was initially abolished (* in b and c), but regained lower levels of firing as the locomotor activity progressed and eventually returned to resting discharge levels before episode termination. (d,e) Plots illustrating a decrease (d, red) or absence of locomotor influence (e, blue) on the firing of individual lateral line afferents during CPG activity. (f) Histograms (error bars represent s.e.m.) showing significant average decreases in firing rates of the recorded afferent population (number indicated in each bar) during the initial phase (I) and throughout the entire (E) duration of fictive swimming episodes. P≤0.001; *P≤0.0001 (Wilcoxon signed-rank test). (g,h) Distinguishable afferent and efferent fibre activity in a single PLLN (g) recorded en passant from one branch (PLLN1, red trace) connected to lateral line hair cells and from a second transected branch (PLLN2, blue trace) of the same nerve (h). During locomotor-related efferent activity (shaded area; indicated by the multiple-unit discharge in PLLN2 and spiking in the single small unit (arrow) in PLLN1), ongoing afferent firing (large spikes in PLLN1) was virtually suppressed throughout the swim episode. Horizontal calibration bars: 5 s in b and c and 1 s in h.
	Figure 6. Locomotor corollary discharge influence on vestibular sensory encoding.(a–h) Recordings of right vr12 (black traces in b and f), the left ALLN (green trace in f) and en passant recordings of afferent fibres in the left AVN (red traces in b and f) in semi-isolated preparations with intact inner ear hair cell–vestibular nerve connectivity (a,e) during sinusoidal (1 Hz, ±60° s−1) horizontal-axis roll motion (a) or vertical-axis head rotations (e) imposed by a two-axis turntable (Tpos). During fictive swimming (shaded areas in b and f), the firing of some vestibular afferent fibres was attenuated (b–d), but facilitated in others (f–h). (c,d,g,h) Histogram (black bars) depicting the mean afferent firing rate modulation (responses of fibres recorded in b and f) over a single cycle of turntable motion (dashed lines) in the absence (c,g) and presence (d,h) of locomotor CPG activity; also plotted are the respective population averages (red curves) of fibres with decreasing (c,d) and increasing firing (g,h) during swimming activity. (i) Plots of individual mean firing rate alterations during motion stimulation (green—increase, red—decrease and black—no change) before (control) and during swimming activity. Grey and red boxes with whisker plots show the distributions of the average firing rates in the two conditions. NS, not significant. (j,k) Averaged response of all recorded afferent fibres (±s.e.m., shaded areas in j; n=22) over a single cycle of turntable motion (dashed grey line) before (black plot) and during (red plot) locomotor CPG activity (j). (k) Averaged extent (error bars represent s.e.m.) of firing rate modulation before (control) and during fictive swimming. (k) **P≤0.001 (Wilcoxon signed-rank test). Horizontal calibration bar in b and f: 1 s.
	Figure 7. Schematics summarizing the functional impact of locomotor corollary discharge on lateral line and vestibular mechanosensory signalling.(a) Connectivity between spinal locomotor CPG circuitry, mechanosensory efferent neurons (ENs) and the sensory periphery along with afferent projections to central vestibular and lateral line nuclei. (b,c) Influence of locomotor activity conveyed by ENs to lateral line (b) and vestibular (c) hair cells/afferent innervation on stimulus signal encoding. The afferent responsiveness to either constant water motion (b—left) or cyclic head rotation (c—left) is substantially attenuated (b,c—right) during concomitant locomotor CPG activation.