Figure 1.
(A) The setup combines a classic electrophoresis box (∼20 cm long) with a video camera and a LED ring to record images of nematodes moving at the surface of an agar gel. The gel is flat and has walls (in grey) to prevent buffer inflow in the electrotaxis area. (B) Velocity distribution during an electrotactic event and evolution with time of the velocity and the orientation of the trajectory of a single worm performing electrotaxis. This shows that during electrotaxis, a single worm moves steadily in a relatively constant direction. (C) The corresponding trajectory is relatively straight and has an angle θ of 15° with the electric field orientation. The characteristic sinusoidal shape of the nematode crawling gait can be observed, indicating that the worm is moving by generating a rearward flexural wave on its body.
Figure 2.
Electrotaxis and directed locomotion.
(A) Trajectories obtained from several distinct experiments done with 10–15 worms are displayed on the same graph. Directed locomotion by electrotaxis (orange, N = 130) is observed for a difference of potential of 120 V, while without any electric field, trajectories are randomly oriented (blue, N = 146). (B) From those trajectories, one can extract a spatio-temporal diagram of the density of nematode (graded in orange or blue intensity) at the surface of the gel. Electrotaxis leads to a directed spreading at the surface of the gel. Note that even in a synchronized population of worms there is a large variability in velocity. (C) Orientations of the trajectory are mostly parallel to the electric field (orange), though they vary from one worm to another. It is not known yet how the electrotaxis orientation is set by worms without electrotaxis, trajectories do not exhibit any preferred orientation (blue). (D) The histograms of parallel and perpendicular velocity with (orange, v// = 140±90 µm/s) or without (blue, v// = 1±70 µm/s) an electric field. The measured velocities may depend on environmental conditions such as the presence or absence of nutrients. This is why worms were systematically rinsed in M9 buffer before their transfer. Similarly, the poro-elastic properties and humidity of the agar gel can affect the worms velocity. It is therefore recommended to run a control with wild-type worms to set a reference speed.
Figure 3.
Comparative analysis of mutant worms and chronological aging effects on forced locomotory abilities.
(A) acr-16 (v// = 80 µm/s±40 µm/s, N = 28), unc-29 (v// = 35 µm/s±34 µm/s, N = 26) and unc-29;acr-16 mutants (v// = 15 µm/s±19 µm/s, N = 22) exhibit reduced velocity when compared to the control population (N2, v// = 110 µm/s±50 µm/s, N = 28) in successive electrotactic runs. Errors are computed as standard deviations. (B) Velocity histograms for aging populations (cf. D). (C) The histograms of v//for mutant worms are significantly different (p<0.05, Fisher test). (D) Populations of worms show a decrease of the average velocity as they get older, from the 1st day (D1) to the 8th day (D8). Here, the average parallel velocity at Day 1, <vD1> = 120 µm/s is taken as a reference. Number of worms: D1/N = 17; D3/N = 15; D7/N = 6. The normalized average velocity is indicated by a vertical line on the histograms (B, C).
Figure 4.
(A) Principle of population sorting. (B) Sorting in action. We conducted a sorting experiment with a mix of 15 wild-type and 15 dbl-1 worms. The number of wild-type worms (orange) and dbl-1 mutant worms (blue) are shown as a function of time and space. We divided the observed area into 5 slices of equal size and computed the number of worms of each strain at different time points (every 2 minutes). Progressively, the wild-type worms separate from the initial mix. The final strip contains only wild-type worms, while, the 2nd and 3rd stripes contain only dbl-1 mutant. The experiment was repeated three times. See also Movie S4.
Figure 5.
(A) We numerically computed the histograms of the distance traveled by a fictitious population of 1000 worms assuming a Gaussian velocity distribution for each worm with an average parallel velocity given by v0(1+gnoise(µ)) and a standard deviation of 50 µm/s. The resulting population has 1000 worms with normally distributed averaged velocity and displays larger intra-population variability for larger µ. We then compared two populations with different v0. Wild-type worms (orange) are moving at <v+> = 200 µm/s and slow worms (blue) are moving at <v−> = 100 µm/s. (B) The same principle allows to compute the proportion of wild-type worms (v = 200 µm/s) in the collect area (>4 cm from start) as a function of the ratio of the average velocity of the two populations. As expected, population with similar dynamics and intra population variability of velocities (larger µ) decrease the sorting efficiency. (C) Using the experimental data displayed on Fig. 3 we computed the distribution of worms position at time τ = 100s. Calling f1(x) and f2(x) the two position distributions, the fraction of population 1 over population 2 as a function of the distance traveled is given by .When starting from a fictitious 50% mix of wild-type and any of the mutant strains acr-16, unc-29 or unc-29;acr-16 (see text), the sub-population is quickly enriched in wild-type worms (faster worms) as the distance of capture increases. Ideally, capturing worms as far as possible from the starting point ensure a perfect sorting. (D,E,F) However, since not all worms move at their maximum velocity during electrotaxis, there is a tradeoff between the degree of separation and the total number of worms that can be captured. Population densities decrease with the distance to start.