Figure 2. Development of a stiffness gradient in the Xenopus embryo brain precedes axon turning.

(a) Time-lapse stiffness maps obtained from a tiv-AFM experiment, showing outlines of fluorescently labelled OT axons (blue) and processed AFM-based stiffness maps (colour maps) overlaid on images of the brain. Colour maps encode the apparent elastic modulus K, a measure of tissue stiffness, assessed at an indentation force F = 10 nN. The time in minutes on each frame is taken from the timestamp of the first measurement in each successive stiffness map; the corresponding overlaid fluorescence images were obtained simultaneously. (b) Visualisation of fold-changes in brain tissue stiffness from one time point to the next, based on the interpolated and smoothed data shown in Figure 2a. Colour scale encodes the fold-change in K at each location on the stiffness map, expressed relative to the values at the previous time point, with the exception of t = 0 min, where all values were set to 1.Tissue stiffness changes throughout the time course, with large changes already occurring between ~40–80 min after the start of the experiment. (c) Plot of mean re-scaled values for the stiffness gradient (orange) and OT turn angle (blue). Stiffness values were binned to match the time points of the developmental stages at which cell body densities were assessed. Dashed lines denote linear fits (R² = 0.99). (d) Boxplots of the extrapolated appearance times of the stiffness gradients and the onset of OT axon turning, relative to the start time of tiv-AFM measurements, with ladder plots for individual embryos overlaid (grey circles/dashed lines). Extrapolations are based on linear fits to the re-scaled data for individual animals (Figure 2c). Stiffness gradients appear significantly earlier than the onset of axon turning (p=0.03, paired Wilcoxon signed-rank test). (e) Scatterplot showing the time delay between extrapolated onsets of stiffness gradients and axon turning, calculated for individual animals. The average delay of 18 min is indicated by the blue line. Boxplots show median, first, and third quartiles; whiskers show the spread of the data; ‘×’ indicates the mean. Error bars denote standard error of the mean. *p<0.05. AFM measurement resolution, 20 µm; all scale bars, 100 µm. N denotes number of animals.

Figure 2—figure supplement 1. Data processing for tiv-AFM experiments.

(a–c) Same Xenopus embryo shown in Figure 2a–b. (a) Inverted epifluorescence raw images of OT axons (black) acquired during tiv-AFM measurements, labelled with membrane-bound GFP under the control of the cell type-specific ath5 promoter, showing turn angles (overlaid red lines). (b) Overlaid tiv-AFM-based colourmaps (‘stiffness maps’), encoding raw values of apparent elastic moduli K, with OT axons outlined in blue. Black squares denote points where AFM data were not analysable. (c) Final stiffness maps used for mechanical gradient quantification. To generate these maps, missing K values were interpolated and data were smoothed in x-, y-, and t-dimensions using an algorithm based on the discrete cosine transform (cf. Figure 1d; for more details see Materials and methods). Regions of interest used to calculate stiffness gradients immediately in front of the advancing OT axons are overlaid on the stiffness maps (green squares). (d) Boxplot of absolute K values obtained in rostral and caudal ROIs (pooled for all N = 6 embryos at all time points), binned over the same time ranges shown in Figure 2c (i.e. 0–60 min – stage 33/34 equivalent; 60–120 min – stage 35/36 equivalent; 120–180 min – stage 37/38 equivalent). Rostral stiffness is already significantly higher than caudal at 0–90 min (p=0.002, Mann-Whitney test), and continues to increase rapidly over the time course, along with the consequent stiffness difference between the two ROIs. Boxplots show median, 1 st, and third quartiles; whiskers show the spread of the data; red crosses denote outliers; ‘o’ indicates the mean. *p<0.05; **p<0.01; AFM measurement resolution, 20 µm; all scale bars, 100 µm.

Figure 2—figure supplement 2. Iterated AFM measurements of brain in vivo do not affect OT outgrowth.

(a, b) Representative images of control (a) and AFM-manipulated (b) embryo brains (outlined with black dashed lines), with OTs labelled using DiI (white, outlined with blue dashed lines). Overall OT morphology appears similar in both conditions. Magenta dashed lines indicate turn angle. (c, d) Quantification of OT elongation (c) and turn angle (d) in control and AFM-manipulated brains. There is no significant difference between the two conditions for either metric (Wilcoxon rank-sum test; p=0.97 and 0.84 respectively). All scale bars: 100 µm. Boxplots show median, first, and third quartiles; whiskers show the spread of the data; ‘o’ indicates the mean (excluding outliers); red crosses denote outliers where applicable. N denotes number of animals.

Figure 2—figure supplement 3. Stereotypy of brain stiffness changes relative to OT development.

Montages from time-lapse AFM measurements in two additional embryos. (a, c) Overlaid tiv-AFM-based stiffness maps for each embryo, encoding interpolated and smoothed values of apparent elastic moduli K. OT axon outlines are shown in blue. In both animals, a similar mechanical gradient arose in the region where OT axons would turn, again due mostly to increasing tissue stiffness of tissue rostral to the axons (cf. Figure 2a, b). (b, d) Time-lapse montages for the same embryos shown in (a) and (c) respectively, with colourmaps encoding fold-changes in K at each location on the stiffness map, expressed relative to the values obtained at t = 0 min. Brain tissue becomes stiffer over time, with largest changes rostral to the growing OT axons (cf. control embryo in Figure 4f). AFM measurement resolution: 20 µm; all scale bars, 100 µm.