Ned Lu ’16
Locomotion requires several sensory modalities, among them proprioception, which allows an organism to sense the spatial position of its body as well as the amount of force being applied by the body to the environment. We investigated the role of proprioception in mediating locomotor recovery after injury in Drosophila melanogaster. Specifically, we used a behavioral assay to assess whether recovery of clockwise/counterclockwise turning preference, or “circling bias,” after amputation of a limb is proprioception-dependent. Individual flies were placed in lidded circular arenas and allowed to explore while video recording tracked the animal. We then quantified the circling bias and characterized both the initial, acute response to amputation and subsequent long-term recovery over a four-day period. Interestingly, we observed a larger induced circling bias immediately after amputation in wild type flies compared to proprioceptive mutants. This differential response was further tested using the GAL4-UAS system, a powerful tool in Drosophila genetics. While the acute response was robust, the long-term recovery phenotypes were less predictable. These results suggest that the acute response to injury is vital as a signal to the animal that a perturbation has occurred, thus adding to the growing literature on the importance of proprioception in motor recovery.
40-60 word summary for website
We used a behavioral assay to assess whether recovery of clockwise/counterclockwise turning preference after amputation of a limb is proprioception-dependent in Drosophila melanogaster. These results suggest that the acute response to injury is vital as a signal to the animal that a perturbation has occurred, thus adding to the growing literature on the importance of proprioception in motor recovery.
Locomotion requires several sensory modalities, among them proprioception, which allows an organism to sense the position of its body parts in three-dimensional space as well as the amount of force being applied by the body to the environment. Mechanosensory neurons that innervate muscles, skin, exoskeleton and other tissues send proprioceptive information to the brain by transducing mechanical energy into an electrical neural signal. We investigated the role of proprioception in mediating locomotor plasticity after injury in Drosophila melanogaster. Specifically, we used a behavioral assay to assess whether recovery of clockwise/counterclockwise turning preference, or “circling bias,” after amputation of a limb is proprioception-dependent. Individual flies were placed in circular arenas and allowed to explore freely while video recording tracked the centroid of the animal during locomotion. We then quantified the circling bias and characterized both the initial, acute response to amputation and subsequent long-term recovery over a four day period. Interestingly, we observed a larger induced circling bias immediately after amputation in wild type flies compared to proprioceptive mutants. The robustness and cell-type specificity of the wild type and proprioceptive mutant phenotypes were tested using the GAL4-UAS system. While the acute response to amputation was robust, the long-term recovery phenotypes were less predictable. These results implicate proprioception in the acute response to injury, thus adding to the growing literature on the importance of proprioception in motor recovery.
The ability of an organism to move enables it to forage for food, escape predators, and find mates and suitable sites for reproduction. Given varied and often dangerous conditions, injuries during locomotion are common in nature. For example, wild dragonflies and bumblebees often experience wing damage due to repeated collisions with vegetation during flight, thus potentially reducing reproductive success and survival (Combes et al., 2010; Mountcastle et al., 2014). As a result, it is important to consider leg safety when analyzing possible locomotor strategies (Birn-Jeffery et al., 2014). Injuries are often unavoidable, and the ability to adapt and recover is crucial for survival. When locomotion is impaired, animals likely suffer a fitness cost, but plasticity (e.g., compensatory behaviors) may help mitigate this cost.
In the context of neuroscience, plasticity is the ability to modify the neural pathways and synapses of the nervous system in response to experience or injury. Plasticity is required for locomotor recovery in organisms, and this capability has been observed in humans who endure spinal cord damage and in rats (Harkema, 2001; Dietz et al., 2009; Ballerman and Fouad, 2006). After the amputation of a limb, mammals are able to recover and resume physical activity, as seen in three-legged dogs or cats. In humans, several medical interventions such as prosthesis after amputation or reconstructive surgery (Bosse et al., 2002) can help restore patients’ mobility with adequate training. Thus, the locomotor system is remarkably robust. Furthermore, the idea of plasticity after limb injury is an active research topic in robotics, as engineers are aiming to build legged robots that can recover locomotion after injury (Christensen et al., 2013; Cully et al., 2015).
Experiments from the de Bivort lab in Drosophila melanogaster demonstrate that the removal of a leg induces a turning bias away from the side of the amputation (e.g., induced counterclockwise bias after removal of the front right leg). Four day old wing-clipped flies were placed onto circular “island” arenas surrounded by water and allowed to walk freely for 2 hours. Their centroids were tracked and the clockwise/counterclockwise bias, or “circling bias,” was calculated. Remarkably, we found that fruit flies with an amputated leg are able to largely recover unbiased turning behavior with the remaining five legs 72 hours post-injury (Isakov & Buchanan et al., in press). The observed return to unbiased circling behavior shows that locomotion after injury may be plastic.
We hypothesize that this form of locomotor plasticity depends on the proprioceptive system, which allows an organism to sense the position of its body parts as well as the amount of effort to employ during movements (e.g., walking). Mechanosensory neurons that innervate muscles, skin, exoskeleton and other tissues send proprioceptive information to the brain by transducing mechanical energy into electrical neural signals. For example, the stretching or compression of a muscle results in the opening of ion channels that send proprioceptive signals to the brain (Delmas et al., 2011). In humans, our proprioceptive sense allows us to touch our nose with our finger even when our eyes are closed. Proprioception is evaluated in sobriety tests, as alcohol intoxication impairs one’s proprioceptive sense (Downey et al., 2015). In addition, several proprioceptive tests are used to aid in the diagnosis of neurological disorders, including visual and tactile placing reflexes (Palmer, 1976). A deeper understanding of proprioception in fruit flies may facilitate better understanding of the proprioceptive sense in humans, as well as disorders that result from faulty proprioception.
A number of experimental and theoretical studies have demonstrated the importance of proprioception in locomotion in insects (Bässler, 1977; Borgmann et al., 2009; Mendes et al., 2013), nematodes (Wen et al., 2012; Mahadevan and Paoletti, 2014), mice (Akay et al., 2014) and cats (Lam and Pearson, 2001). Here, we use fruit flies to investigate whether proprioception plays an important role in recovery from injury in our locomotor paradigm.
One way to test the role of the proprioceptive system is to alter the function of genes mediating proprioception. Proprioceptive genes for which we have genetic tools include nanchung (nan) and inactive (iav), which encode transient receptor potential (TRP) channels. nan and iav are involved in mechanotransduction, converting mechanical energy into electrical signals. In insects, chordotonal organs serve as internal mechanoreceptors and are found at exoskeletal joints as well as between joints within limb and body segments (Field and Matheson 1998). The nan gene encodes a cation channel subunit and is expressed in the dendritic cilia of chordotonal organs (Kim 2003). nan mutants display an ‘uncoordinated’ phenotype as well as loss of hearing, hygrosensation (ability to detect humidity) and negative gravitaxis (upward movement) defects (Liu 2007). The iav gene is also expressed in chordotonal organs and is required for the discrimination between optimal and non-optimal temperatures (Kwon 2010). While nan and iav are essential for these phenotypes, it is unknown whether they are the mechanosensor itself or just part of the pathway.
Another way to test the function of proprioception is to modulate the activity of neurons expressing proprioceptive genes. Under the GAL4-UAS binary expression system, a powerful tool in Drosophila genetics, GAL4 lines for proprioceptive genes are crossed with effector lines with the Upstream Activating Sequence (UAS) acting as a promoter for a desired gene. GAL4, a yeast transcription factor, can be inserted into the genome with a cloned promoter or acquire an expression pattern by insertion into an enhancer region on a transposable element. GAL4 protein binds to UAS, its yeast target sequence. UAS can in turn be fused to effector genes such as Shibirets1 (Shi) or Tetanus toxin light chain (TNT) to drive the expression of the effector in specific cells.
Shibirets1 is a temperature-sensitive allele of dynamin, a GTPase responsible for endocytosis, that blocks synaptic vesicle recycling in neurons at restrictive temperatures (>30°C), thereby suppressing neurotransmission (Figure 1) (Kitamoto 2001). At permissive temperatures (<30°C), neurons function as normal. Shibirets1 is thus a powerful tool to study the functions of specific neurons because it allows for the comparison of behavior between animals with and without the function of specific circuit elements, while holding genetic background constant. In addition, the GAL4-driven expression of TNT in targeted cells blocks synaptic transmission and can result in behavioral defects, thereby shedding light on the role of specific neurons (Sweeney 1995). The efficiency of these effector genes depends on the properties of the targeted neurons. For example, Shibirets1 successfully impaired short-term olfactory memory when expressed in adult mushroom bodies, while TNT expression in the same cells did not lead to impairment (Thum 2006). Thus, it is important to consider the differential potencies and mechanisms of effector genes when using the GAL4/UAS system.
We aimed to investigate the role of proprioception in locomotive recovery after injury in Drosophila. The walking behavior of flies was observed both before and after an induced injury. After removal of a forelimb, recovery of unbiased turning behavior was assessed in wild type flies and proprioceptive mutants. In particular, we observed whether there is recovery to a significant biomechanical injury (the amputation of the right forelimb), and if that recovery requires proprioception. Though there has been research on freely walking Drosophila (Strauss and Heisenberg, 1990; Mendes et al., 2013; Berman et al., 2014), there is limited knowledge on walking and recovery after injury in insects. These authors observed that even immediately after amputation of a fly’s hind leg, changes in the fly’s behavior allowed it to continue walking but at a slower speed and with an altered gait. However, these studies only investigated the immediate results of an amputation of the hind leg in wild type flies. Here, we assess both acute and long-term recovery from injury over an extended period of time.
It has been shown that individual, uninjured flies exhibit idiosyncratic clockwise/counterclockwise circling bias in an open, circular arena, but that a population is on average unbiased (Buchanan et al., 2015). We used automated video analysis to characterize the circling bias of wild type and genetically manipulated flies during recovery from leg amputation. Using the powerful genetic tools available in Drosophila (e.g., the GAL4-UAS system), we drove the tissue- and cell-specific expression of effector genes such as Shibirets1 and TNT in neurons expressing pertinent transgenes, like nan-GAL4 and iav-GAL4. These effectors alter the physiology of the targeted neurons, and thus we aim to test whether these proprioceptive neurons are necessary for post-injury locomotor plasticity.
This thesis investigates the role of proprioception in mediating locomotor plasticity after injury in Drosophila melanogaster. Specifically, we will 1) use a circling bias behavioral assay to determine the effect of an induced leg injury on locomotor behavior; 2) assess whether recovery of exploratory locomotor circling bias is proprioception-dependent; 3) implicate proprioception in the acute response to injury using the GAL4-UAS system.
Fly strains and care
Flies were housed on modified CalTech medium in 25°C incubators with a 12/12 h light-dark cycle. Canton S (CS) flies were used as the wild type strain. Flies mutant for nanchung and inactive were obtained from the Bloomington Drosophila Stock Center (nan36a BDSC #24902 and iav3621 BDSC #24768; Bloomington, IN, USA). All experimental flies were 4-10 days post-eclosion and female. Behavioral experiments were recorded in an environmental room set at 23°C, 40% humidity, with the room lights off.
A tray of 27 2-inch circular, arenas was fabricated from clear acrylic cut with a laser engraver (epilog, Golden, CO, USA) (Figure 2A). The arena floors were sanded, biasing the flies to walk on the floor. Clear, acrylic lids prevented flies from escaping and were coated with Sigmacote to prevent flies from walking in an inverted position. Arenas were uniformly illuminated from below by an array of LEDs (5500K, LuminousFilm, Shreveport, LA, USA) covered by a diffuser fabricated from two sheets of 1/8″ thick clear acrylic frosted on both sides by sanding (Figure 2B). The tray of arenas was imaged from above by a 2MP digital camera (Logitech, Newark, CA, USA, and Point Grey, Richmond, BC, Canada), and the X-Y position of individual flies’ centroids were identified and tracked by custom written software written in LabView (National Instruments, Austin, TX, USA).
Circling bias experiments
4-10 day old individual flies were placed into the lidded-arenas of the tray. The animals were allowed to walk freely for 1 hour while their centroids were tracked and recorded. Flies were removed and anesthetized with CO2. The right foreleg was amputated at the tibia-femur joint and the flies were tested again 1 hour, 24 hours, 48 hours, 72 hours and 96 hours post-injury. The direction of motion was inferred as the vector angle between centroids of successive frames.
Centroids were acquired in real time using custom LabVIEW scripts. Image analysis was implemented using Matlab2012a with the Image Processing Toolbox and Statistics Toolboxes (The MathWorks, Inc., Natick, MA, USA). The script for determining locomotion circling bias was also implemented in Matlab. All other analyses were performed using the statistical software R3.0.3 (R Foundation for Statistical Computing, 2014).
Amputation of a limb induces circling bias
First, in order to determine how an induced leg injury affects locomotor behavior, we investigate path-level behavior of adult wild type Drosophila before and after amputation of the right foreleg. Individual flies are allowed to explore circular arenas (Figure 3A), for 1 hour, and the X-Y positions of their centroids are tracked and recorded. In general, all observed paths are composed of segments that are predominantly clockwise or counterclockwise, with little inward or outward locomotion (Figure 3B). After the pre-amputation recording, the right foreleg is removed, and flies are recorded in the arena 1 hour, 24 hours, 48 hours, 72 hours and 96 hours post-amputation. In order to quantitatively assess the average clockwise/counterclockwise circling bias, we measure the weighted average direction of the tangential component of the velocity relative to the center of the arena (Figure 3C). We call this mu “for mean” score (μ) (Buchanan et al., 2015). A score of μ = 0 indicates perfectly unbiased locomotion (flies moving clockwise and counterclockwise to the same extent), while -1 > μ > 0 reflects an overall clockwise bias and 0 < μ < 1 corresponds to an overall counterclockwise bias. Histograms of the magnitude of the circumferential component of motion are used to visualize circling bias and calculate μ (Figure 3D).
Wild type flies exhibit a change from paths composed of roughly equal portions of clockwise and counterclockwise segments (Figure 4A) to highly biased walking in the direction opposite to the leg that is removed immediately after amputation (Figure 4B). We find that wild type Canton S (CS) flies are unbiased before amputation (μ = -0.029, s.e.m. = 0.022) and exhibit counterclockwise bias (μ = 0.217, s.e.m. = 0.026) immediately following amputation of the right forelimb (Figure 5). However, only partial recovery (μ = 0.132, s.e.m. = 0.024) is observed 96 hours after amputation (Figure 5), rather than the full recovery observed in previous experiments using island arenas (Isakov & Buchanan et al., in press).
Proprioceptive mutants demonstrate differential locomotor response to injury
Next, in order to characterize the mechanistic basis of the observed locomotor plasticity, we tested proprioceptive mutant strains in the circling bias behavioral assay. Nanchung and Inactive are TRP ion channels that are co-expressed in the fly’s proprioceptive organs, which include the chordotonal organs of the femur, wing and other joints. These channels are required for normal locomotion and hearing (Kim et al., 2003; Gong et al., 2004). We hypothesize that disrupting proprioceptive feedback prevents a fly from demonstrating long-term recovery of unbiased circling post-injury. Similar to wild type flies, flies mutant for nanchung (nan) and inactive (iav) show little clockwise/counterclockwise bias while exploring the arena pre-amputation (nan: μ = -0.009, s.e.m. = 0.015; iav: μ = -0.022, s.e.m. = 0.014) (Figure 5). Interestingly, both nan and iav mutants display a counter-clockwise bias (nan: μ = 0.105, s.e.m. = 0.015; iav: μ = 0.077, s.e.m. = 0.018) immediately post-amputation that is significantly lower than the induced bias observed in wild type flies (μ = 0.217, s.e.m. = 0.026) (Figure 5). nan mutants exhibit only partial recovery (μ = 0.050, s.e.m. = 0.017) 96 hours after amputation rather than full recovery to baseline (Figure 5). iav mutants demonstrate minimal recovery (μ = 0.063, s.e.m. = 0.016) (Figure 5). The offspring of wild type and nan mutants, CS x nan, show the wild type phenotype, characterized by the large counterclockwise bias immediately post-amputation (μ = 0.199, s.e.m. = 0.033) followed by partial recovery (μ = 0.134, s.e.m. = 0.029) (Figure 5). Thus, wild type flies exhibit a greater response to injury compared to proprioceptive mutants immediately after amputation. The recovery phenotype over the following four days is less clear, since partial recovery is observed in both wild type flies and nan mutants. Full recovery of unbiased circling behavior is not observed in any of the genotypes in the assay, although more complete recovery was seen in wild type, but not proprioceptive mutants in alternative, earlier configuration experiments (Isakov & Buchanan et al., in press).
The acute response to injury is likely proprioception-dependent
In order to further characterize a fly’s response and subsequent recovery to injury, we used the GAL4-UAS system to test the robustness and cell-type specificity of the wild type and proprioceptive defect phenotypes. The GAL4-driven expression of TNT in neurons blocks synaptic transmission and can result in behavioral defects, thereby shedding light on the role of specific neurons (Sweeney 1995). In nan>TNT animals, nan-expressing neurons are presumably silenced. Consistent with the nan proprioceptive mutant phenotype, nan>TNT flies show a small response immediately after amputation (μ = 0.075, s.e.m. = 0.019) (Figure 6A). Interestingly, nan>TNT flies demonstrate a larger counterclockwise bias four days after injury (μ = 0.149, s.e.m. = 0.032) (Figure 6A), indicating a lack of recovery of unbiased circling behavior and potential deterioration of locomotor control, or even an exacerbation of the effects of injury over time. The parental lines, nan-GAL4 and UAS-TNT, were tested as controls. Consistent with the wild type phenotype, UAS-TNT flies show a large response immediately after amputation (μ = 0.182, s.e.m. = 0.019) and partial recovery (μ = 0.126, s.e.m. = 0.022) (Figure 6A). However, nan-GAL4 flies exhibit a small response (μ = 0.086, s.e.m. = 0.021) and no recovery of circling bias (μ = 0.091, s.e.m. = 0.022) (Figure 6A). In iav>TNT flies, a large response is observed (μ = 0.186, s.e.m. = 0.026), inconsistent with the phenotype observed in iav proprioceptive mutants (Figure 6B). The parental lines show a large response (iav-GAL4: μ = 0.202, s.e.m. = 0.020; UAS-TNT: μ = 0.182, s.e.m. = 0.019) and minimal recovery (iav-GAL4: μ = 0.160, s.e.m. = 0.032; UAS-TNT: μ = 0.126, s.e.m. = 0.022) (Figure 6B).
With inconclusive results from the TNT experiments, we turned our attention to a different effector gene to characterize the wild type and proprioceptive defect phenotypes in response to injury. Shibirets1 is a temperature-sensitive allele of dynamin that blocks vesicle recycling (endocytosis) from neurons above 30°C, effectively silencing them. At lower temperatures, the neuron functions as normal (Kitamoto 2001). Control shibirets1 experiments were performed at 23°C. In the chronic shibirets1 experiments, flies were amputated and recorded at room temperature and stored at 30°C to determine the effect of inhibiting vesicular release in proprioceptive neurons on the long-term response to injury (Figure 7A). We hypothesize that the initial response to amputation will be normal (large μ Post Day 0) but recovery on subsequent days will be impaired by the silencing of proprioceptive neurons. Both nan>shi chronic and iav>shi chronic strains show a large response 1 hour after amputation (nan>shi chronic: μ = 0.174, s.e.m. = 0.023; iav>shi chronic: μ = 0.229, s.e.m. = 0.028) (Figure 8A, 8B). As expected, the nan>shi control and iav>shi control lines, which were always stored at 23°C, also show a large response immediately after amputation as well (nan>shi control: μ = 0.225, s.e.m. = 0.026; iav>shi control: μ = 0.205, s.e.m. = 0.024) (Figure 8A, 8B). Inconsistent with our hypothesis, both nan>shi chronic and iav>shi chronic strains partial recovery by Day 4 (nan>shi chronic: μ = 0.102, s.e.m. = 0.027; iav>shi chronic: μ = 0.095, s.e.m. = 0.021) (Figure 8A, 8B).
In the acute shibire experiments, flies were amputated at 30°C and recorded at 23°C to investigate the effect of inhibiting vesicular release in proprioceptive neurons on the acute response to injury (Figure 7B). Only pre-amputation and one hour post-amputation videos were recorded to determine the effect of silencing proprioceptive neurons on the immediate response to injury. If proprioception is necessary for the acute response to injury, we expect that the initial response to amputation will be small, as seen in the nan and iav proprioceptive mutant response. Consistent with our hypothesis, both nan>shi acute and iav>shi acute flies show unbiased circling pre-amputation (nan>shi acute: μ = -0.002, s.e.m. = 0.028; iav>shi acute: μ = 0.009, s.e.m. = 0.027) and a small response immediately post-amputation (nan>shi acute: μ = 0.100, s.e.m. = 0.020; iav>shi acute: μ = 0.123, s.e.m. = 0.032) (Figure 9A, 9B). As expected, the nan and iav acute controls (Figure 7B), which were amputated at room temperature, demonstrate a large response post-amputation (nan>shi control: μ = 0.199, s.e.m. = 0.017; iav>shi control: μ = 0.217, s.e.m. = 0.018) (Figure 9C, 9D).
To assess which lines show an acute response to injury that is significantly different from the others, we use one-way ANOVA with post-hoc Tukey-Kramer HSD (honest significant difference) test (Figure 10). The one-way ANOVA confirms that one or more of the lines tested are significantly different (P < 0.001). We then use the Tukey-Kramer HSD method to identify which pairs of genotypes are significantly different from each other. Wild type flies show a significantly larger acute response compared to the proprioceptive mutants, nan and iav. While the nan>shi and iav>shi acute controls show the wild type response, their responses are not significantly greater than the responses of the nan>shi acute and iav>shi acute experimental flies. However, the nan>shi acute and iav>shi acute flies are very similar in magnitude, and statistically indistinguishable from the iav and nan mutants. The lack of significant difference from their respective nan>shi and iav>shi controls may be due to smaller sample sizes of the nan>shi acute and iav>shi acute lines (N = 54, 52 and N = 54, 49, respectively). Thus, wild type and acute shibire controls show a larger circling bias response following amputation compared to the proprioceptive mutants and shibire flies with temperature-induced silencing of nan– and iav-expressing neurons. While we find no evidence of a chronic role for proprioception in this assay, we show that proprioception is likely involved in the acute response to injury.
Our results provide evidence of the importance of proprioception in the acute response to injury using the model organism Drosophila melanogaster. We see that flies initially spend equal portions of time exploring in clockwise and counterclockwise directions. After amputation of the right foreleg, wild type flies exhibit a strong counterclockwise bias. By contrast, nanchung and inactive mutants show a significantly smaller counterclockwise bias immediately post-amputation.
This result can be interpreted in a number of ways. First, the differential acute response could be due to differences in activity level between proprioceptive mutants and wild type flies. If wild type flies are less active and make fewer turns, then a larger circling bias might result if there are not as many turns made to counterbalance the induced bias. For example, a sluggish amputated fly may make only a few counterclockwise turns and stop moving, resulting in a large mu score. To address overall activity level as a potentially confounding factor, we use total distance traveled as a proxy for activity level. We find that wild type flies walk a significantly longer distance than proprioceptive mutants (Figure 11A), thus indicating that activity level is not the major determinant of acute circling bias response.
Second, mutants may be better suited to adapt to leg injury because they do not rely on proprioception and are able to overcompensate by enhancing other sensory systems. However, this explanation is unlikely because proprioceptive mutants fail to show a higher survival rate than wild type flies (Figure 11B). Rather, the mortality rate of mutants is likely higher than wild type flies, suggesting that lack of proprioception is unlikely to be an adaptive trait involved in locomotor recovery after injury.
We propose that a salient, acute response to injury is vital as a signal to the animal that a perturbation has occurred. An injured fly that “recognizes” the life-threatening injury via the induced circling bias is more effectively able to adapt its locomotion with the disability. By contrast, mutants lacking feedback from their appendages regarding the force applied by and location of their limbs fail to show a large acute response to injury and, consequently, may fail to adapt with the handicap. Moreover, the larger turn bias in wild type animals may represent an adaptive state, like “favoring” the injured leg, and in the absence of proprioception flies cannot engage this state. A correlate in humans may be limping, in which there is a deviation from the normal gait pattern usually caused by physical trauma (Naranje et al., 2015).
The role of proprioception in the chronic response to injury is less clear. We used a high-throughput assay (27 flies per tray) suitable for future experiments involving extensive screens for circuit mapping of the proprioceptive system. We were unable to replicate the island arena experiments that demonstrate robust recovery of unbiased circling bias in wild type flies 72 hours post-injury (Isakov & Buchanan et al., in press). There are several reasons why we might not have observed full recovery in the modified high-throughput assay. First, it is important to note that the island experiments were limited to a relatively small sample size for all test groups (9 < N < 30) due to experimental limitations (e.g., the water created difficulties with the setup). Second, flies in the island experiments were wing-clipped, and this perturbation may have resulted in more efficient tuning mechanisms for walking in the absence of flight. In addition, the size of the arenas, the surrounding water, and the lids are all aspects that changed between the 27-arena tray assay and the island assay. Moreover, differences in genetic background of the experimental lines likely contribute to the variation in chronic recovery phenotypes observed.
This study points to a number of avenues for future work. One direction is to perform similar behavioral experiments in other animals, such as larger hexapods or tetrapods. In addition, investigating the role of proprioception in the acute response to injury in larger animals will elucidate whether our findings are generalizable to other species. The role of proprioception in the chronic response to injury warrants further investigation. Animals could be monitored for a longer period of time to better understand long-term plasticity.
A future goal, which requires a robust behavioral assay demonstrating proprioception-dependent recovery of locomotion, is to map out the neural circuitry underlying proprioception. Circuit mapping requires mining of the FlyLight database to identify candidate neurons based on morphology. Specifically looking at the prothoracic ganglion of Drosophila for neural activity in this region may indicate the transfer of proprioceptive information from the right forelimb. Any “hits” are neurons that may be involved in the proprioceptive system, and GAL4 lines for these neurons can be screened in the circling bias assay. Candidates showing the proprioceptive mutant phenotype are likely to be involved in the proprioceptive circuit. Any discoveries in neural circuitry can be extremely beneficial to further understand proprioception of not only flies but other organisms as well. By discovering the role that proprioception plays in locomotor plasticity, we aim to further understand how animals, and humans, generate adaptive behaviors and achieve high levels of locomotor performance even after physical damage.
Akay T, Tourtellotte WG, Arber S, Jessell TM (2014) Degradation of mouse locomotor pattern in the absence of proprioceptive sensory feedback. Proc. Natl. Acad. Sci. 111(47), 16877-16882.
Ballermann M, Fouad K (2006) Spontaneous locomotor recovery in spinal cord injured rats is accompanied by anatomical plasticity of reticulospinal fibers. Eur. J. Neurosci. 23(8), 1988-1996.
Bässler U (1977) Sensory control of leg movement in the stick insect Carausius morosus. Biol. Cybern. 25(2), 61-72.
Berman GJ, Choi DM, Bialek W, Shaevitz JW (2014) Mapping the stereotyped behaviour of freely moving fruit flies. J. R. Soc. Interface 11(99), 20140672.
Birn-Jeffery AV, Hubicki CM, Blum Y, Renjewski D, Hurst JW, Daley MA (2014) Don’t break a leg: running birds from quail to ostrich prioritise leg safety and economy on uneven terrain. J. Exp. Biol. 217(21), 3786-3796.
Borgmann A, Hooper SL, Buschges A (2009) Sensory feedback induced by front-leg stepping entrains the activity of central pattern generators in caudal segments of the stick insect walking system. J. Neurosci. 29(9), 2972-2983.
Bosse MJ, MacKenzie EJ, Kellam JF, Burgess AR, Webb LX, Swiontkowski MF, Sanders RW, Jones AL, McAndrew MP, Patterson BM (2002) An analysis of outcomes of reconstruction or amputation after leg-threatening injuries. New Engl. J. Med. 347(24), 1924-1931.
Buchanan SM, Kain JS, de Bivort BL (2015) Neuronal control of locomotor handedness in Drosophila. Proc Natl Acad Sci U S A 112:6700-5.
Christensen DJ, Schultz UP, Stoy K (2013) A distributed and morphology-independent strategy for adaptive locomotion in self-reconfigurable modular robots. Robot. Auton. Syst. 61(9), 1021-1035.
Combes SA, Crall JD, Mukherjee S (2010) Dynamics of animal movement in an ecological context: dragonfly wing damage reduces flight performance and predation success. Biol Lett 6(3): 426-9.
Cully A, Clune J, Tarapore D, Mouret JB (2015) Robots that can adapt like animals. Nature 521(7553), 503-507.
Delmas P, Hao J, Rodat-Despoix L (2011) Molecular mechanisms of mechanotransduction in mammalian sensory neurons. Nat Rev Neurosci. 140(12): 139–53.
Dietz V, Grillner S, Trepp A, Hubli M, Bolliger M (2009) Changes in spinal reflex and locomotor activity after a complete spinal cord injury: a common mechanism? Brain 132(8), 2196-2205.
Downey LA, Hayley AC, Porath-Waller AJ, Boorman M, Stough C (2015) The Standardized Field Sobriety Tests (SFST) and measures of cognitive functioning. Accid Anal Prev 86: 90-98.
Field LH, Matheson T (1998) Chordotonal Organs of Insects. Advances in Insect Physiology 27: 3.
Gong Z, Son W, Chung YD, Kim J, Shin DW, McClung CA, Lee Y, Chang DJ, Kaang BK (2004) Two interdependent TRPV channel subunits, inactive and Nanchung, mediate hearing in Drosophila. J. Neurosci. 24(41), 9059-9066.
Harkema SJ (2001) Neural plasticity after human spinal cord injury: application of locomotor training to the rehabilitation of walking. Neuroscientist 7(5), 455-468.
Hughes, GM (1952) The co-ordination of insect movements 1. The walking movements of insects. J. Exp. Biol. 29(2), 267-285.
Isakov A, Buchanan SM, Sullivan B, Ramachandran A, Chapman JKS, Lu E, Mahadevan L, de Bivort B (2016) Recovery of locomotion after injury in Drosophila depends on proprioception. In press.
Kain J, Stokes C, Gaudry Q, Song X, Foley J, Wilson R, de Bivort B (2013) Leg-tracking and automated behavioural classification in Drosophila. Nat. Commun. 4, 1910.
Kasuya J, Ishimoto H, Kitamoto T (2009) Neuronal mechanisms of learning and memory revealed by spatial and temporal suppression of neurotransmission using shibire, a temperature-sensitive dynamin mutant gene in Drosophila melanogaster. Front Mol Neurosci. 2:11.
Kim J, Chung YD, Park DY, Choi S (2003) A TRPV family ion channel required for hearing in Drosophila. Nature 424: 81-4.
Kitamoto T (2001) Conditional modification of behavior in Drosophila by targeted expression of a temperature-sensitive shibire allele in defined neurons. J Neurobiol 47(2): 81-92.
Kwon Y, Shen WL, Shim H, Montell C (2010) Fine Thermotactic Discrimination between the Optimal and Slightly Cooler Temperatures via a TRPV Channel in Chordotonal Neurons. J Neurosci 30: 10465-71.
Lam T, Pearson KG (2001) Proprioceptive modulation of hip flexor activity during the swing phase of locomotion in decerebrate cats. J. Neurophysiol. 86(3), 1321-1332.
Liu L, Li Y, Wang R, Yin C (2007) Drosophila hygrosensation requires the TRP channels water witch and nanchung. Nature 450: 294-8.
Mahadevan L, Paoletti P (2014) A proprioceptive neuromechanical theory of crawling. Proc. Roy. Soc. Lond. B Bio. 281(1790), 20141092.
Mendes CS, Bartos I, Akay T, Marka S, Mann RS (2013) Quantification of gait parameters in freely walking wild type and sensory deprived Drosophila melanogaster. Elife 2, e00231.
Mountcastle AM, Combes SA (2014) Biomechanical strategies for mitigating collision damage in insect wings: structural design versus embedded elastic materials. J Exp Biol 217: 1108-15.
Naranje S, Kelly DM, Sawyer JR (2015) A Systematic Approach to the Evaluation of a Limping Child. Am Fam Physician 92(10): 908-16.
Palmer AC (1976) Introduction to Animal Neurology, 2nd Edition. Blackwell Scientific, Oxford.
Strauss R, Heisenberg M (1990) Coordination of legs during straight walking and turning in Drosophila melanogaster. J Comp Physiol [A] 167: 403–412.
Sweeney ST, Broadie K, Keane J, Niemann H, O’Kane CJ (1995) Targeted expression of Tetanus Toxin Light Chain in Drosophila specifically eliminates synaptic transmission and causes behavioral defects. Neuron 14: 341–351.
Thum AS, Knapek S, Rister J, Dierichs-Schmitt E, Heisenberg M, Tanimoto H (2006) Differential potencies of effector genes in adult Drosophila. J Comp Neurol 498(2): 194-203.
Wen Q, Po MD, Hulme E, Chen S, Liu X, Kwok SW, Gershow M. Leifer AM, Butler V, Fang-Yen C (2012) Proprioceptive coupling within motor neurons drives C. elegans forward locomotion. Neuron 76(4), 750-761.
Figure 1. Spatially and temporally restricted suppression of neurotransmission using the UAS-shits1 transgene. A GAL4 driver specific to neuronal subsets is crossed to the UAS-shits1 line. Progeny ectopically expressing shits1 in GAL4-positive neurons are raised at permissive temperatures. When the temperature is shifted from permissive to restrictive, the shits1 product (temperature-sensitive dynamin) is rapidly inactivated and synaptic vesicle recycling is interrupted. As a result, the GAL4-positive neurons are depleted of synaptic vesicles and synaptic transmission is blocked. Behavioral consequences of spatial and temporal suppression of neurotransmission can be observed in free-moving animals. The shits1 product regains its activity and synaptic vesicles are restored immediately after the animals are returned to permissive temperatures. (Taken from Kasuya et al., 2009)
Figure 2. Circling bias assay experimental setup. (A) Tray of circular arenas. (B) Rig consists of an LED board, 2 sheets to diffuse light, and a camera above. The tray is placed on the top diffuser sheet.
Figure 3. Individual flies exhibit circling bias in an open arena. (A) An open arena assay for exploratory locomotion. Individual flies were placed in circular arenas and allowed to walk freely for 1 hour. Their position was tracked. (B) Example path data collected from an individual fly. (C) For each data point, a circling score is calculated by subtracting the fly’s direction of motion (d) from its angular position (θ) in radians. This gives the circumferential component of motion, with π/2 indicating clockwise motion (CW), 3π/2 indicating counterclockwise motion (CCW), π indicating walking straight into the center of the arena, and 0 indicating walking straight out from the center. (D) Histogram of circling scores for a strongly CW-biased fly (red), a CCW-biased fly (blue), and a relatively unbiased individual (black). μ is the averaged signed circumferential component of motion (velcircum/speed); 1 corresponds to purely CCW motion, -1 corresponds to purely CW motion, and 0 corresponds to unbiased circling. (Taken from Buchanan et al., 2015)
Figure 4. Induced circling bias after leg injury. Histograms of clockwise/counterclockwise circling behavior of representative wild type flies (A) pre-amputation and (B) 1 hour post-amputation. Relatively unbiased turning is seen pre-amputation (roughly symmetrical peaks). The asymmetry immediately post-amputation corresponds to the induced counter-clockwise bias. Each line represents an individual fly. Average μ value indicated.
Figure 5. Differential response to injury in proprioceptive mutants. Time series plot of mean circling bias (μ) before and after amputation. Shaded regions indicate ± 1 s.e.m. for CS wild type (+/+) (N = 157, 153, 141, 129, 125, 110), nan (nan/nan) (N = 181, 181, 163, 133, 105, 89), iav (iav/iav) (N = 135, 130, 119, 103, 95, 87), and CS x nan (nan/+) (N = 66, 66, 63, 59, 58, 53). Full recovery of unbiased circling (μ = 0 on Day 4) is not observed in any of the genotypes.
Figure 6. Tetanus neurotoxin light chain (TNT) experiments. Time series plot of mean circling bias (μ) before and after amputation. Shaded regions indicate ± 1 s.e.m. for (A) nan>TNT (N = 68, 64, 58, 54, 50, 44), nan-GAL4 (N =108, 105, 76, 70, 55, 48), UAS-TNT (N = 81, 74, 59, 56, 54, 49) and (B) iav>TNT (N = 81, 78, 77, 76, 73, 69), iav-GAL4 (N = 79, 74, 67, 65, 63, 59), UAS-TNT (N = 81, 74, 59, 56, 54, 49).
Figure 7. Chronic and acute shibire experiments. Schematic of (A) chronic and (B) acute time course. Control experiments were done always at 23°C.
Figure 8. Chronic inhibition of nan– and iav-expressing neurons shows no evidence for role of proprioception in long-term recovery. Time series plot of mean circling bias (μ) pre-and post-amputation for (A) nan>shi chronic (N = 80, 78, 73, 70, 68, 66), nan>shi control (N = 78, 75, 67, 64, 63, 63), (B) iav>shi chronic (N = 81, 77, 74, 66, 63, 62), iav>shi control (N = 81, 77, 76, 74, 71, 70). Shaded regions indicate ± 1 s.e.m.
Figure 9. Acute inhibition of nan– and iav-expressing neurons results in smaller induced circling bias. Plots of circling bias (μ) before and after amputation for (A) nan>shi acute (N = 54, 52), (B) iav>shi acute (N = 54, 49), (C) nan>shi control (N = 78, 75), and (D) iav>shi control (N = 81, 77). Lines represent individual flies. Red line indicates average μ.
Figure 10. The acute response to injury is likely proprioception-dependent.
Mu scores pre-amputation (blue) and 1 hour post-amputation (red). The nan>shi and iav>shi lines tested at room temperature display the wild type phenotype (large acute response to injury) and serve as controls for the acute shibire experiments. nan>shi acute and iav>shi acute demonstrate the proprioceptive mutant phenotype (small acute response to injury) seen in nan and iav. Statistically significant differences between groups indicated by different letters above bars, one-way ANOVA, P < 0.001, and Tukey-Kramer HSD test, P < 0.05. Error bars are ± 1 s.e.m.
Figure 11. Activity level and survival rate by genotype. (A) Total distance traveled by genotype. Error bars indicate ± 1 s.e.m. (B) Survival curves for wild type (blue, N = 157), nan (red, N = 181) and iav (green, N = 135) flies.