Denham, J. et al. (2017) International Worm Meeting "An integrated neuro-mechanical model of C. elegans forward locomotion."

Ranner, T., Denham, J., Cohen, N.

[International Worm Meeting, 2017]

Behavioural responses in C. elegans can be observed through changes in locomotive patterns. It is therefore important to consider the role of the physics in computational models of the neural circuit for motor behaviours. We present a dynamical systems model of C. elegans forward locomotion in which the neural circuit is divided into a series of repeating identical units, coupled via posterior stretch receptor feedback. Each unit includes cholinergic, bistable B-type motor neurons (modelled after bistable RMD neurons, Mellem et al. 2008, Boyle et al. 2012), implicit GABAergic D-type motor neurons (Boyle et al. 2012) and nonlinear viscoelastic muscle forcing along the body (Boyle et. al. 2012). The neural model incorporates proprioceptive feedback as the mechanism for producing sustained oscillations. Integrating the neural model with a recent continuum mechanical body model (Cohen and Ranner 2017) provides this feedback and is also the method for incorporating environmental drag from the surrounding fluid, closing the neuronal-environmental loop. Biophysically realistic parameters are used to obtain sustained travelling waves in muscle activation which respond to changes in environmental viscoelasticity. To explore the pattern generation mechanism, we present results from bifurcation analysis performed in the isolated neural framework and in the fully integrated neuro-mechanical model. We show how these results are modulated by changes in the external drag and internal material properties of the passive and active body. References [1] Boyle JH, Berri S, Cohen N: Gait modulation in c. elegans: an integrated neuro-mechanical model. Frontiers in computational neuroscience 2012, 6:10. [2] Mellem JE, Brockie PJ, Madsen DM, Maricq AV: Action potentials contribute to neuronal signaling in C. elegans, Nature Neuroscience 2008, 11:865-867 [3] Cohen N, Ranner T: A new computational method for a model of C. elegans biomechanics: Insights into elasticity and locomotion performance, arXiv:1702.04988, 20

Ghosh, D. D. et al. (2015) International Worm Meeting "Neuroendocrine reinforcement of a dynamic multisensory decision."

Nitabach, M.N., Sanders, T., Cohen, N., Hong, S., Koelle, M.R., Chase, D.L., Ghosh, D. D.

[International Worm Meeting, 2015]

To navigate complex natural environments containing both dangerous and valuable items, animals must make economic decisions on the basis of information transduced by multiple senses. However, detailed underlying neural mechanisms of multisensory decision making remain poorly understood. Here we confronted worms with a multisensory decision in which the reward of food must be balanced with the threat of desiccation imposed by a hyperosmotic barrier intervening between the worm and a source of food odor. We find that this decision is modulated by food deprivation. To identify neural substrates underlying this decision, we focused on the RIM interneuron, which is advantageously positioned to transduce integrated multisensory information into locomotor outputs. Consistent with this hypothesis, we find that the activation of a neuropeptide receptor in RIM sets the balance of threat and reward in this decision, with greater receptor activation biasing the worm against crossing the dangerous barrier. Unexpectedly, however, RIM controls the decision not by synaptic signaling to the downstream premotor command circuit, but rather by extrasynaptic aminergic signaling directly onto the primary osmosensory neuron to tune its sensitivity. Additionally, our results suggest that this neuromodulator relay is suppressed in food deprived states, thereby providing a link between internal state, neural network activity, and decision making. Finally, to characterize the complex and dynamic interplay between neuromodulator activity, neuron state, and behavior in the decision making arena, we reproduced the paradigm in silico [1]. Computational modeling revealed how non-linear sensory integration in RIM modulates neuromodulator circuit activity to implement the decision. Taken together, these studies reveal a cellular and molecular mechanism for a dynamic multisensory decision. Intriguingly, our results identified organizational circuit principles conserved between mammalian and C. elegans multisensory decision making. Therefore our studies have broad implications for understanding principles underlying multisensory decision making in Metazoans.1Sanders, T., Ghosh, D.D., Nitabach, M.N., and Cohen, N. "Nonlinear sensory integration in C. elegans: a computational model." International C. elegans meeting.

Ranner, T. et al. (2017) International Worm Meeting "A new biomechanical model links material properties with locomotion behaviors and predicts that gait modulation is a form of speed optimisation in C. elegans."

Cohen, N., Ranner, T.

[International Worm Meeting, 2017]

An organism's ability to move freely is a fundamental behaviour across the animal kingdom. We present a biomechanical model of C. elegans locomotion together with a novel computational approach. We model the body as a flexible elastic shell, subject to muscle forcing along the body walls of the body. All parameters are grounded in behavioral, anatomical or physiological data. Our numerical approach allows us to solve for arbitrary body shapes. Our model replicates behaviours across a wide range of environments. We use it to study forward locomotion undulation gaits, linking between the animal's material properties and its performance across a range of viscoelastic environments. The model makes strong predictions on the viable range of the worm's Young's modulus and suggests that animals can control speed via the known mechanism of gait modulation that is observed across different media.

Sanders, Tom et al. (2015) International Worm Meeting "Nonlinear sensory integration in C. elegans: a computational model."

Cohen, Netta, Ghosh, D.D., Sanders, Tom, Nitabach, M.N.

[International Worm Meeting, 2015]

Animal survival depends on a combination of often conflicting demands such as foraging and evading of dangers. To navigate effectively in such unknown and changing conditions, animals must continuously integrate over a variety of sensory cues, and adapt their decision making strategy in a context dependent manner. Here, we examine the neural control of a sensory integration task in the nematode C. elegans. The task involves an ASH-triggered aversive response to high osmolarity fructose and an AWA-triggered attractive response to diacetyl [1]. In the assay, worms are placed in the center of a ring of fructose; two drops of diacetyl are located outside the ring. We present a computational model, consisting of point worms, situated in a virtual arena that closely mimics this experimental assay, and endowed with a sensory motor pathway of two sensory neurons, a neural integration pathway and two motor programs (pirouettes and steering). A monoamine (PDF-2 and tyramine) modulation circuit involving RIM and ASH is overlaid on the synaptic circuit, in line with molecular data [1]. Model parameters were constrained by behavioral data for wild type and mutant animals for a range of stimulus concentrations. Based on our simulation results, we reject a null hypothesis of a linear sensory integration mechanism in RIM and present results that are consistent with the data for a sensory "coincidence detector" like process in RIM.[1] Ghosh, D.D., Sanders, T., Hong, S., Chase, D.L., Cohen, N., Koelle, M.R., and Nitabach, M.N. "Neuroendocrine reinforcement of a dynamic multisensory decision." International C. elegans meeting.

Denham, J.E. et al. (2019) International Worm Meeting "Signatures of central and peripheral control in an integrated neuromechanical model of C. elegans locomotion."

Denham, J.E., Cohen, N., Ranner, T.

[International Worm Meeting, 2019]

Animal neuromechanics describes the coordinated self-propelled movement of a body, subject to the combined effects of internal neural control and mechanical forces. Here we use a computational model to disentangle manifestations of neural and mechanical modulation of undulatory locomotion of C. elegans, with a focus on proprioceptively driven and central pattern generated neural control. We reveal a fundamental relationship linking body elasticity, internal viscosity and environmental drag in determining the dynamics of the body and demonstrate the manifestation of this relationship in the context of proprioceptively driven control. By modulating characteristics unique to proprioceptive neurons, we predict the signatures of internal gait modulation that contrast with the known signatures of externally or biomechanically modulated gait. We analyse the common and contrasting signatures of different models of proprioceptive and central pattern generated control. First, we show that, unlike central pattern generated control, proprioceptive feedback can suppress neuromechanical phase lags during undulatory locomotion. Second, we explore manipulations of the neural and mechanical systems and relate the results with recently published experimental findings.

Kalogeropoulou, E. et al. (2017) International Worm Meeting "Caenorhabditis elegans exhibits an increase in the probability of pirouettes and head swing amplitude in the absence of the SAA class of neurons."

Hope, I.A., Cohen, N., Kalogeropoulou, E.

[International Worm Meeting, 2017]

Many of the behaviours of C. elegans are mediated by the worm's ability to orient itself towards chemical gradients or other sensory cues. Two mechanisms of orientation have been observed in the worm: pirouettes and steering. Past studies have identified a number of neuron classes that contribute to navigation by linking neuronal laser ablations and observations of mutants to changes in body postures and locomotion statistics1,2. Recently, optogenetic manipulation has also shed light on the neuronal control of the worm's chemotactic behaviour3. SMB motor neurons have been postulated to be part of the navigation circuit. We have generated a chromosomally-integrated transgenic line (UL4230) where the SMBs are genetically ablated in early larvae. The line demonstrates a loopy phenotype similar to that previously observed following laser-ablation of SMBs1. We have also targeted another class of neuron, located close to and highly connected via synapses to the SMBs but not previously explored: the SAA interneurons. We initially used expression of GFP to confirm that the selected promoters drove expression in the location originally described. A combination of lad-2 and unc-42 promoters was used to genetically ablate the SAAs by targeted expression of the worm's caspase, as encoded by two distinct parts of the ced-3 gene, such that the intact enzyme is produced only in these neurons specifically4. Absence of GFP expression confirmed the specific and targeted ablation of the SAAs. The generated strains, with SMBs or SAAs ablated (UL4230 and UL4207), demonstrated a phenotype different to that of N2 in both spontaneous and evoked locomotion. Locomotion was evoked using a radial gradient of NH4Cl. Initial results suggest that both SMBs and SAAs suppress the probability of pirouettes and decrease the amplitude of sinusoidal undulations. Additional experiments are underway to explore the contributions of SAA and SMB, separately and together, to undulations and steering, and to link those to the pirouette initiation pathway. 1. Gray J. M., et al., (2005). A circuit for navigation in Caenorhabditis elegans. Proc. Natl. Acad. Sci. U.S.A, 102(9). 2. Iino Y., and Yoshida K. (2009). Parallel use of two behavioral mechanisms for chemotaxis in Caenorhabditis elegans. Journal of Neuroscience, 29(17). 3. Kocabas A., et al., (2012). Controlling interneuron activity in Caenorhabditis elegans to evoke chemotactic behaviour. Nature, 490(7419). 4. Chelur D. S., and Chalfie M. (2007). Targeted cell killing by reconstituted caspases. Proc. Natl. Acad. Sci., 104(7).

Berri, Stefano et al. (2009) International Worm Meeting "An integrated model of C. elegans locomotion: from swimming to crawling."

Berri, Stefano, Boyle, Jordan H., Cohen, Netta, Tassieri, Manlio, Hope, Ian A.

[International Worm Meeting, 2009]

C. elegans is capable of coordinated locomotion both when swimming in water and when crawling on an agar surface, two behaviors with distinct kinematics. By analyzing the worm''s locomotion in a range of fluids with increasing visco-elasticity, we were able to demonstrate that swimming and crawling are merely two snapshots out of a continuum of locomotory behaviors that are achieved by a modulation of a single gait[1]. This finding suggests that a single neural mechanism underlies this entire range of behaviors. We developed an integrated model of the worm''s forward locomotion that consists of a ventral cord nervous system, muscles, a body and an environment. The neural model consists of excitatory B-like bistable neurons that receive sensory feedback (mediated, in the model, by posteriorly directed stretch receptors on B class motorneurons) as well as inhibition from D-type neurons. Muscles receive both excitatory and contralateral inhibitory inputs and control the shape of a worm, instantiated by a physical model of the C. elegans body, embedded in a model of the visco-elastic environment. The integrated model can generate and coordinate oscillations, and captures the entire swim-crawl transition as the properties of the environment are changed. The neuromuscular control of the locomotion is modulated solely by the stretch receptor input which in turn varies with the external physics. No parameters are changed in the neuromuscular control. The model does not require a central pattern generator (or distinct neural mechanism) in the head, and can start from arbitrary initial body shape. Tests of the model are presented in virtual knockout and laser ablation simulation experiments, as well as in complex environments such as microfluidic "artificial dirt" chips and irregular granular media, which the worm is likely to encounter in its natural habitat. We are now using the model to gain insight into various uncoordinated phenotypes, including those that have a stronger effect in liquid than on agar. [1] Berri S, Boyle JH, Tassieri M, Hope IA and Cohen N. 2009. "Forward locomotion of the nematode C. elegans is achieved through modulation of a single gait." HFSP journal, In press.

Vidal-Gadea A G et al. (2010) Neuronal Development, Synaptic Function and Behavior, Madison, WI "Biogenic amines mediate transition between crawling and swimming in C. elegans"

Brauner M, Gottschalk A, Erbguth K, Kressin L, Vidal-Gadea A G, Topper S, Young L, Pierce-Shimomura J T

[Neuronal Development, Synaptic Function and Behavior, Madison, WI, 2010]

A wide variety of animals must quickly adjust their pattern of locomotion to successfully navigate through different environmental niches. Selection and execution of the appropriate locomotory pattern is therefore paramount to survival. Although C. elegans is capable of performing many adaptive behaviors, it has been controversial whether forward crawling and swimming represent distinct gait-like forms of locomotion or the modulation of a single form of locomotion [1-3]. Biogenic amines have been shown to mediate the transition between gait-like forms of locomotion across taxa as diverse as sea slugs, leeches, lampreys and humans. We previously reported that C. elegans crawls and swims with distinct kinematics and different patterns of muscle activity [2]. We now combine quantitative behavioral analysis, optogenetic tools and neuronal ablation to show that C. elegans uses biogenic amines to switch between crawling and swimming in a gait-like manner. As in other invertebrates, we find that serotonin mediates the smooth transition from crawling to swimming in C. elegans. Serotonin is further required to inhibit motor behaviors (e.g. foraging and pharyngeal pumping) during swimming that normally only accompany crawling. Mirroring the role of dopamine in other invertebrates, C. elegans uses dopamine to successfully initiate and maintain crawling when emerging from liquid. Over 600 million years of separate evolution notwithstanding, the highly conserved role played by biogenic amines such as dopamine and serotonin across taxa attests to how vital their function is to adaptive strategies for locomotion. Korta J, Clark DA, Gabel CV, Mahadevan L, Samuel AD. J. Exp. Bio. 2007 210:2383-9.Pierce-Shimomura JT, Chen BL, Mun JJ, Ho R, Sarkis R, McIntire SL. PNAS. 2008 105:20982-7.Berri S, Boyle JH, Tassieri M, Hope IA, Cohen N. HSFP J. 2009 3:186-93.

Holbrook, R.I. et al. (2017) International Worm Meeting "Novel 3D locomotion, body postures and exploration behaviour in Caenorhabditis elegans."

Holbrook, R.I., Salfelder, F., Ranner, T., Cohen, N.

[International Worm Meeting, 2017]

How animals move through their natural habitats is fundamental to understanding their biology. We possess considerable knowledge of Caenorhabditis elegans locomotion and movement behaviour from extensive investigations of these animals living on the 2D surface of an agar plate. However, C. elegans in the wild live in the complex-structured 3D environment of decaying vegetation, and we have very limited information about how C. elegans behaves in their natural habitats. This raises concerns that C. elegans locomotion data collected in 2D may be incomplete and unrepresentative, severely limiting our models and predictions about their neural control, biomechanics and movement. One reason for the gap in our knowledge is the difficulty in recording accurate 3D movements of animals at the scale of C. elegans (~1 mm) with high spatial and temporal resolution. At the macro scale, conventional 3D imaging relies on multiple camera views with overlapping focus regions, but lenses do not have sufficient magnification to image at the C. elegans scale. Conventional microscopes have a very narrow depth of focus so cannot be used in a multiple camera system. Instead, they have to employ Z-stacking to image static bodies in 3D, making them unsuitable for imaging a freely moving body in real-time. To overcome these obstacles, we built a 3D imaging system using three telecentric lenses each with a 7- 21 mm depth of focus (depending on the magnification) attached to three 4.2 megapixel cameras. This system allows us to image the worm moving freely through up to 9.2 cm3. This volume is calibrated using photogrammetry to solve for the lens distortion and camera geometry. A novel image analysis algorithm produces accurate reconstructions of the 3D posture, in the form of a smooth midline, and location of the worm within the volume. We recorded individual worms moving freely through a range of viscoelastic fluids corresponding to different concentrations of gelatine in M9 buffer solution. All recordings were between 25 and 45 Hz. Across a wide range of viscoelasticities, we found that the body postures of the worm are more commonly 3D than 2D. Specifically, we report novel 3D body postures exhibited by C. elegans, including a helicoid locomotion gait and a "lasso" turning posture. Furthermore, the trajectories of the worms are more commonly 3D than 2D. These kinematic and performance data inform a new integrated and 3D neuromechanical locomotion model.

Yuval, O. et al. (2019) International Worm Meeting "Characterization of C. elegans motor behaviour in 3D environment?s."

Ranner, T., Yuval, O., Salfelder, F., Cohen, N.

[International Worm Meeting, 2019]

In its natural habitat, C. elegans maneuvers through 3D environments with complex physical and chemical properties. However, to date, the worm's locomotion has been studied almost exclusively in quasi-planar settings, such as on an agar surface. Observing how the worm moves in more natural environments may unravel new behaviors that may allow us to push forward our understanding of their underlying neural and biomechanical control mechanisms. Understanding the interplay between the biomechanics and neural control in C. elegans requires a characterization of the anatomy, mechanics and neurobiology. We adopted an integrated approach, combining 3D imaging of freely-moving animals with quantitative behavioral characterization and neural and biomechanical modeling. Using an experimentally-grounded computation model, we capture observed behaviors with a view to making testable predictions. We recorded freely-moving worms in different gelatin concentrations in M9 buffer. Our 3D imaging system consists of three telecentric lenses attached to 4.2MP cameras. Worms move freely through a large cube with 2-21mm depth of focus and recorded at 25-45 Hz. The imaged volume is calibrated using photogrammetry to solve for the lens distortion and camera geometry. A novel image analysis algorithm produces accurate reconstructions of the 3D posture, and location of the worm within the volume. We use a worm-reference coordinate system to find a low-dimensional representation of non-planar body postures to quantify their evolution continuously over time. Across a wide range of gelatin concentrations, we observe transitions between planar and non-planar postures and link postural dynamics to worm speed, trajectory and possible behavioral strategies. We have developed a 3D biomechanical model that integrates neural and mechanical control to generate locomotion dynamics in different environments. In this model, the body of the worm is modeled as a continuum viscoelastic shell with internal pressure, subject to the balance of internal (e.g. muscle activation) and external (environmental) forces, which we solve for the midline of the worm. The muscles, which are attached to the cuticle, provide the link between neural control and body deformation that can give rise to bending in both the dorso-ventral and left-right directions, captured in our anatomically grounded coordinate system. The model was parameterized and validated using 2D scenarios, before adding three dimensional components.