Table of Contents

  1. Introduction
  1. Aims
  2. Methods
  3. Results and Discussion
  1. Conclusions and Future Directions
  2. References

1. Introduction

Eight-legged locomotion is one of the most ancient travelling modes on land. Spiders exhibit a diverse locomotion repertoire, moving at slow and fast speeds, forwards, backwards, sideways, and even jumping and climbing, all at similar speeds and maintaining high locomotor performance (Biancardi et al. 2011). They can voluntarily lose legs when threatened, a phenomenon called autotomy, which has shown to have negative effects on locomotor performance (Wrinn and Uetz 2008). At the costs of autotomy on locomotor performance, a trade-off with survival and other benefits should exist, as autotomy has been maintained in several lineages across evolutionary time. Perhaps one of the additional benefits of autotomy conservation comes from the way they compensate for it by adjusting kinematic patterns and neural control mechanisms. Rhythmic motor behaviors such as walking are controlled by Central Pattern Generators (CPGs), which are centrally located neural circuits capable of generating rhythmic and repetitive outputs even in the absence of sensory feedback (Dickinson 2006; Katz 2016). CPGs evidence behavioral flexibility, which suggests that on spiders, learning from previous autotomized limbs, can play an important role on increasing locomotor performance after subsequent limb autotomizations and regeneration. Therefore, the main goal of the first phase of this project is twofold: 1) determine the effects of limb autotomy on running performance and kinematics and, 2) determine how gait is adjusted after limb autotomy. We hypothesize running performance will decrease after first autotomy. These findings will be determinant for the second phase of the project; gait analyses after leg regeneration and second autotomy, to determine if long-term learning plays an important role in spider gait adaptation.

1.1 What is Autotomy?

  • Voluntary shed of one or more appendages
lizard tail autotomy spider leg autotomy

1.2 How spiders walk

  • Stereotyped alternating tetrapod gait
  • Two sets of 4 legs move in phase
walking spider 
embed_youtube("GtHzpX0FCFY")
stereotyped alternating tetrapod gait

1.3 Central Pattern Generators

  • Centrally located neural circuits
  • No need of sensory feedback –> brain
  • Produce rythmic and repetititve movements

2. Aims

1. Determine the effects of limb autotomy on running performance and basic kinematics parameters.

2. Determine how gait is adjusted after limb autotomy.

3. Methods

For the running trials, individuals of guatemalan tiger rump tarantula (Davus pentaloris), were used. These spiders are excellent runners, autotomize their legs and regrow them in approximately 43 days, and molt frequently. Spiders were obtained from the pet trade. Only recently hatched spiders were obtained in order to assure no autotomizations events had occured.

tiger rump tarantula

The following is the experimental protocol for the focus of the present project.

4. Results and Discussion

Running Speed

ANCOVA to test the effects of treatment on mean velocity controlling for body length

mod1_vel <- lm(UtotMean ~ length + Treatment, data = mean_vel)
anova(mod1_vel)
## Analysis of Variance Table
## 
## Response: UtotMean
##           Df  Sum Sq Mean Sq F value    Pr(>F)    
## length     1  2737.8 2737.81 16.0981 0.0001176 ***
## Treatment  1   661.1  661.08  3.8871 0.0514774 .  
## Residuals 98 16666.9  170.07                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
mean_vel %>%
  ggplot(aes(x = Treatment, y = UtotMean)) +
  geom_boxplot(aes(color = Treatment), show.legend = FALSE) +
  ylab("Mean Velocity (cm/s)") +
  theme_dark()

p-value = 0.0514

Fail to reject null hypothesis

No significant differences in running speed between control and 2 leg autotomy

Running performance does not decrease after loosing 2 legs!!!

Stride Frequency and Stride Length

Stride Frequency

ANCOVA to test the effects of treatment on mean Stride Frequency controlling for individual

mod1_SF <- lm(meansFreq ~ ID + Treatment, data = mean_SF)
anova(mod1_SF)
## Analysis of Variance Table
## 
## Response: meansFreq
##           Df  Sum Sq Mean Sq F value Pr(>F)
## ID         7  33.344  4.7634  0.7801 0.6099
## Treatment  1   0.003  0.0034  0.0006 0.9813
## Residuals 25 152.645  6.1058
mean_SF %>%
  ggplot(aes(x = Treatment, y = meansFreq)) +
  geom_boxplot(aes(color = Treatment), show.legend = FALSE) +
  ylab("Mean Stride Frequency (Hz)") +
  theme_dark()

p-value = 0.9813

Fail to reject null hypothesis

No significant differences in stride frequency between control and 2 leg autotomy

Spiders do not seem to change their stride frequency after loosing 2 legs!!!

Stride Length

ANCOVA to test the effects of treatment on mean Stride Length controlling for individual

mod1_SL <- lm(meansLen ~ ID + Treatment, data = mean_SL)
anova(mod1_SL)
## Analysis of Variance Table
## 
## Response: meansLen
##           Df Sum Sq Mean Sq F value Pr(>F)
## ID         7 126.97 18.1386  1.1949 0.3430
## Treatment  1   0.40  0.4019  0.0265 0.8721
## Residuals 24 364.32 15.1799
mean_SL %>%
  ggplot(aes(x = Treatment, y = meansLen)) +
  geom_boxplot(aes(color = Treatment), show.legend = FALSE) +
  ylab("Mean Stride Length (BL)") +
  theme_dark()

p-value = 0.8721

Fail to reject null hypothesis

No significant differences in stride length between control and 2 leg autotomy

Spiders do not seem to change their stride length after loosing 2 legs!!!

WHY???

Gait Adjustments

Because they are switching their stereotyped alternating tetrapod gait, to a stable alternating tripod, as shown below

alternating tatrapod gait legs 1 and 5 autotomized
modified tripod gait

Assuming numbers close to \(\pi\) represent legs moving out of phase and numbers close to zero, legs moving in phase, the plot below shows the mean for the phasing of legs 4-6 are closer to \(\pi\), which means that spiders are tending to adopt the modified tripod gait, in which legs 4-6, which in the tetrapod gait are in phase, are now out of phase–> closer to \(\pi\) (Wilshin et al. forthcoming).

rel_phasing %>%
  ggplot() +
  geom_boxplot(aes(x = Treatment, y = absRel46))

5. Conclusions and Future Directions

* Running performance and kinematic parameters do not seem to be affected by two leg autotomies.

* Spiders seem to be switching their gait effectively to a modified tripod gait after loosing 2 legs.

* After spiders regrow their legs, a second autotomy will be performed to determine if they adjust their gait faster, which could potentially indicate learning from previous experience by means of their Central Pattern Generators.

6. References

  1. Biancardi, C. M., C. G. Fabrica, P. Polero, J. F. Loss, and A. E. Minetti. 2011. Biomechanics of Octopedal Locomotion: Kinematic and Kinetic Analysis of the Spider Grammostola Mollicoma. Journal of Experimental Biology 214(20):3433-42.

  2. Dickinson, P. S. 2006. Neuromodulation of Central Pattern Generators in Invertebrates and Vertebrates. Current Opinion in Neurobiology 16(6):604-14.

  3. Katz, P. S. 2016. Evolution of Central Pattern Generators and Rhythmic Behaviours. Philosophical Transactions of the Royal Society B: Biological Sciences 371:20150057.

  4. Wrinn, K. M. and Uetz G. W. 2008. Effects of Autotomy and Regeneration on Detection and Capture of Prey in a Generalist Predator. Behavioral Ecology 19(6):1282-88.

  5. Wilshin, S, P.S. Shamble, K.J. Hovey, R. Harris, A.J. Spence, S.T. Hsieh. Limping following limb loss increases locomotor stability. Accepted with minor revisions at Journal of Experimental Biology.