QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online June 29, 2007
Journal of Experimental Biology 210, 2526-2539 (2007)
Published by The Company of Biologists 2007
doi: 10.1242/jeb.003939

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Burgess, H. A. Articles by Granato, M. Search for Related Content PubMed PubMed Citation Articles by Burgess, H. A. Articles by Granato, M.

Modulation of locomotor activity in larval zebrafish during light adaptation

Harold A. Burgess and Michael Granato^*

Department of Cell and Developmental Biology, University of Pennsylvania School of Medicine, Philadelphia, PA 19104-6058, USA

View larger version (100K): [in this window] [in a new window]   Fig. 1. High throughput measurement of locomotor kinematics in zebrafish larvae. (A) Simultaneous tracking of multiple larvae. In this example, 24 larvae are tracked over 1000 ms (red), with position and curvature information measured every ms. Scale bar, 2.0 mm. (B) Primary measurements are position, orientation (Bi) and curvature (Bii). Quantitative kinematic descriptions of locomotion are derived from these measurements, yielding measures including C-bend angle (Bii), distance traveled (Biii) and trajectory (Biv). Note that this is a high-resolution image – movement analysis is performed on the lower quality images in A. (C) Four examples showing curvature across time (400 ms). The three lower examples demonstrate the smooth changes in curvature observed for active larvae, compared to the flat curvature function for a stationary fish. (D) Scatter analysis of stationary (red squares, N=175) and active larvae (blue squares, N=156) shows that active larvae can be distinguished from stationary larvae by the maximal signal power of the Fourier transform of the curvature function, together with the maximal three-point derivative of the function. (E) Comparison of automated and manual analysis of a new group of 800 events using the criteria established in D demonstrates that automatic analysis reliably distinguishes stationary and active larvae with 98% accuracy. Larvae moving at the beginning of the video recording are detected with 90% accuracy, being mistaken for larvae initiating movement 7% of the time and stationary larvae 3% of the time. Altogether, 96.8% (775/800) events are correctly recognized.

View larger version (60K): [in this window] [in a new window]   Fig. 2. Kinematic identification of the two most frequently observed elements of the larval locomotor repertoire, scoots and routine turns. (A) Example of a scoot, showing the low bend angle and forward trajectory of the larva. (B) Example of a routine turn, demonstrating the large bend angle and reorientation of the larva prior to forward swimming. (C) Histogram of bend amplitudes for 4199 movement events. The histogram was fitted as the sum of two Gaussians (solid black line: for peak 1, µ=16.9, =7.9; peak 2, µ=59.6, =20.1). (D) Scatter analysis of bend angles against bend amplitudes for 1681 movement episodes confirms spontaneous motor events do not form a behavioral continuum, but can be distinguished by selecting thresholds for bend amplitudes and angles. Red dotted line indicates the amplitude and head bend angle thresholds used to distinguish scoots from turns. (E–J) Kinematic analysis of the two types of movement events distinguished in D (672 scoots, 1009 turns) verifies that this method identifies motor patterns with distinct properties. Kinematic distributions for trajectory (E), displacement (F), head bend angle for the second sinusoid, equivalent to the `counterbend' for turns (G), bend amplitude for the counterbend (H), swim yaw (I) and swim rhythm (J) show highly significant differences (independent sample t-test with unequal variances, P<10–10).

View larger version (21K): [in this window] [in a new window]   Fig. 3. Locomotor responses to light and dark flash stimuli. (A) Transient increases in light elicit a sharp spike in turn initiations. Larvae were pre-adapted at 20 µW cm–2 white light and at time zero, tested with a 500 ms pulse of 200 µW cm–2 (open circles, N=8 groups) or maintained in constant illumination (closed circles, N=8 groups). Activity was measured in 400 ms windows at the indicated time points. A significant spike in turns was noted for the time window coinciding with the light flash (two-tailed t-test, P=0.0064) but not at any other time points. Scoot initiations were not significantly altered by the light-flash (data not shown). (B) Turn initiations (black circles) increase with the intensity of the light flash. Larvae were pre-adapted at 20 µW cm–2 before being tested with a series of 10 bright flashes at the indicated intensity levels, at 30 s intervals (N=5 sets of larvae for each intensity). A significant increase in the frequency of turn initiations compared to baseline levels was found for light flashes of >1 log unit above baseline illumination (*P<0.05). Scoot initiations (grey circles) in the same larvae were slightly depressed compared to baseline, but this only achieved significance at one intensity tested. (C) Transient decreases in light provoke an increase in turn initiations. Larvae were pre-adapted at 200 µW cm–2 and challenged with a 500 ms-long dark flash to 20 µW cm–2 at time zero (open circles, N=10 groups) or left in constant illumination throughout the experiment (closed circles, N=10 groups). Turns were significantly increased in the 500 ms window starting at the beginning of the dark flash (two-tailed t-test, P<10–10), but not at any other time point. Scoots were significantly reduced only in the time window corresponding to the dark-flash, most likely reflecting the huge increase in turns at that time. Scoot initiations were otherwise not affected (data not shown). (D) Larger reductions in illumination elicit more turn responses, without evoking scoots. Larvae were pre-adapted at 130 µW cm–2, then tested with a series of 10 dark flashes of the indicated magnitude (N=6 per intensity). Turn initiations (black circles) were significantly increased (*P<0.05) for dark flashes of around 1 log unit and greater whereas scoot initiations (grey circles) were reduced under the same conditions, likely as a result of the large number of larvae initiating turns.

View larger version (66K): [in this window] [in a new window]   Fig. 4. Kinematic properties of turns initiated in response to changes in illumination. (A) Scatter analysis of turn kinematics for spontaneous routine turns (N=166, green), turns elicited by dark flashes (N=506, red) and short latency acoustic startle responses (N=269, blue). Dark-flash evoked turns form a distinct cluster, with bend angles exceeding those achieved by acoustic startle responses, but with much slower angular velocity. (B) Latency distribution for turns initiated during a 500 ms light flash (N=631, grey squares) or under constant illumination (N=244, dark squares). Turns peak around 200 ms after the increment in lighting. (C) Turns initiated in response to a 1000 ms dark flash (N=236, dark squares) have a longer latency, peaking 300 ms after the reduction in lighting. Turns initiated during constant illumination (N=63, light squares) show uniform distribution. (D–G) Comparison of movement kinematics for five motor patterns. Turns elicited by light flashes (`LF turn', N=111), routine turns (`Rout. turn', N=66), long latency acoustic startles (`LLC', N=96), short latency acoustic startles (`SLC', N=382) and turns elicited by dark flashes (`DF turn', N=104); values all means ± s.d. Light-flash turns were indistinguishable from routine turns for all kinematic parameters scored. Kinematics of long latency startles differed from routine turns for all kinematic parameters scored. Turns elicited by dark flashes were distinct from short latency acoustic startle responses. Although turn magnitude is extreme for dark flashes, turns are relatively slow, but occur over a protracted duration (two-tailed t-tests, **P<0.001). (H) After light extinction, larvae orient toward the area where the light was extinguished. Cones represent initial orientation of larvae, shading and numbers indicate rightward turn bias (where 100% signifies that all turns are made to the right, –100% means always left). Larvae initially facing the light with their right side show a strong right turn bias, while the opposite is true for larvae in the reverse orientation. Larvae oriented parallel to the direction of the light show no directional bias. (I) Example of a dark-flash turn, demonstrating the very large bend amplitude attained and 180° reorientation typical for these motor patterns.

View larger version (36K): [in this window] [in a new window]   Fig. 5. Large angle C-start responses to dark-flash stimuli are Mauthner independent. (A) Laser ablation of Mauthner cells. Cell bodies (asterisks) and axons (arrows) visible in control lesions (top) are absent in immunohistochemical staining 48 h after ablation (bottom), while fibers coursing around the Mauthner cells, including the axon cap (arrowheads) are unaffected. Scale bar, 20 µm. (B) High performance startle responses elicited by acoustic/vibratory stimuli in control larvae (`control', N=8) are completely abolished after bilateral Mauthner cell lesion (`ablation', N=6 larvae). Slower, long latency responses to the same stimulus are not affected by Mauthner cell lesions (data not shown). (C) Mauthner cell ablation does not impair dark-flash responsiveness in the same set of larvae, showing that O-bend responses to dark-flash stimuli are not initiated by the Mauthner cell startle circuitry (two-tailed t-test, P=0.89). Larvae were tested individually with a series of 10 dark flash stimuli from 65 µW cm–2 to darkness, at 1 min intervals. (D) O-bend kinematics during the dark-flash test in control larvae (`C') and ablated larvae (`A') are almost identical, arguing that the Mauthner cell is not involved in larval dark-flash responses. No statistical differences are present between any pair of kinematic values (two-tailed t-test, P>0.1 for all measures). Values are means ± s.d. of O-bend responses.

View larger version (75K): [in this window] [in a new window]   Fig. 6. Locomotor activity is regulated by light. (A) The frequency of spontaneous turn (closed circles) and scoot (open circles) initiations are significantly greater in more intense light. Larvae (N=10 groups per light level) were adapted to each light level for 30 min before testing. For each group, a series of twenty 400 ms video recordings was made under constant conditions at the indicated light level and average activity computed. Regression lines for activity versus log(intensity) are shown (turns, r2=0.49; scoots, r2=0.51). (B) Both turn (Bi) and scoot (Bii) initiations show a gradual reduction during dark adaptation (DA). Larvae were pre-adapted to 400 µW cm–2 white light for at least 3 h before being subjected to sudden darkness (open circles, N=10 groups) or were maintained under constant illumination (CI; closed circles, N=10 groups). At each time point, a series of twenty 400 ms video recordings was made and the average activity computed for each group. Two-way ANOVA for group and time after light extinction revealed significant group by time interaction for both scoots (F(5,107)=3.3, P=0.009) and turns (F(5,107)=5.4, P<0.001). (C) Both turn (Ci) and scoot (Cii) initiations rapidly increase after larvae maintained in constant darkness are suddenly switched to bright light. Within 15 min of illumination with 230 µW cm–2 (light-adapted, LA; open circles, N=10 groups), both turn and scoot initiations reach levels similar to larvae maintained in bright light for several hours (for example, constant illumination groups in B). Control larvae maintained in constant darkness (CI, closed circles, N=10 groups) continue to show low levels of locomotor activity. Two-way ANOVA for group and time for time points after the onset of illumination revealed a significant main effect of group for both scoots (F(1,107)=216, P<0.001) and turns (F(1,107)=482, P<0.001), and a significant group by time interaction for turns (F(5,107)=7.9, P<0.001). (D) Ultraradian light:dark cycles of 1 h each demonstrate that photic input directly modulates activity levels in larvae. Larvae were monitored over a 24 h period (consisting of 12 cycles), with a series of twenty 400 ms video recordings taken every 10 min (offset from the beginning of each transition by 5 min). During light cycles, larvae were exposed to constant 60 µW cm–2, while during dark cycles (shaded brown) larvae were maintained in darkness. (Di) The initiation frequency of both scoots and turns closely follows the light:dark cycle periodicity. Orange broken curves show functions estimated by performing non-linear regression according to the model: activity=b1+b2*sin(2*time/b3+b4), such that b3 is the periodicity of the function (see text). (Dii) Mean initiation frequency for scoots (open circles) and turns (closed circles) for each time point during a 2 h period averaged over all 12 cycles. (E) Turn initiations show a transient increase for 5 s following the switch to sustained darkness (Ei, open circles, N=30 groups) compared to larvae maintained in constant illumination (200 µW cm–2, closed circles, N=30 groups). No change in turn initiations occurs over the next 7 min. Immediately after light extinction, scoot initiations (Eii) are slightly reduced; however, after 60 s, scoots show a transient but highly significant increase above baseline levels. For each group of 400 ms recordings were collected at the indicated time points (*P<0.05, t-test versus constant light). (F) Behavioral light adaptation begins 60 s after dark-adapted embryos are exposed to bright light (140 µW cm–2, open circles, N=20 groups). After an initial spike in turns (Fi) elicited by the abrupt change in illumination, there is a lag of approximately 1 min in which turn initiations remain at similar levels to larvae maintained in constant darkness (closed circles, N=20 groups). Thereafter turn initiations rapidly climb to light-adapted levels. Scoot initiations (Fii) show a similar pattern, with an acute spike following light onset, a lag phase of 60 s, then a rapid increase to normal light-adapted levels (*P<0.05, t-test versus constant dark).

View larger version (12K): [in this window] [in a new window]   Fig. 7. Effect of light and dark adaptation on dark-flash responses. (A) Kinetics of behavioral dark adaptation assessed by responsiveness to dark flashes. Light-adapted larvae were placed in darkness, then tested at a single time point after the onset of dark adaptation (N=5 groups each time point). A test consisted of restoring the original level of illumination (200 µW cm–2) and assessing responsiveness to a series of 5 dark flash stimuli of 500 ms duration, with 30 s intervals between stimuli. No change in dark-flash responsiveness is seen after 3 min of dark adaptation; however, exposure to longer periods of constant darkness rapidly reduces dark-flash responsiveness so that by 30 min, responsiveness to dark flashes is almost completely lost. (B) Kinetics of behavioral light adaptation assessed by responsiveness to dark flashes. Responsiveness to dark flashes develops slowly after dark-adapted larvae are exposed to constant bright light, reaching a maximum 20 min after the beginning of light adaptation. After the onset of illumination (400 µW cm–2), groups (N=9) were tested with a 1000 ms long dark flash every 2 min. Video recordings were taken during the dark flash to measure O-bend responses. (C) Light-adapted larvae adjust quickly to increases in illumination. Larvae were pre-adapted at 10 µW cm–2 for at least 3 h, then shifted to 100 µW cm–2 for the indicated intervals before being tested with a 1000 ms dim flash back to 10 µW cm–2 (N=11 groups for each time point). After just 1 s of increased illumination, 25% of larvae respond to dim flashes with O-bend responses. Larvae reach maximal levels of responsiveness (70% of larvae, see Fig. 3D) to dim flashes after 100 s of sustained illumination.