Andrew French & Päivi Torkkeli
Dalhousie University, Canada

https://doi.org/10.36866/pn.99.34

The world of the typical spider is dominated by mechanical events such as vibrations generated by prey, predators or courting partners. Spiders also sense their own body position and movements. Consequently, they possess an array of sophisticated mechanosensory organs.

Spiders are arachnids, arthropod animals that lack wings or antennae, but have eight legs plus a body divided into two distinct regions. Like other arthropods (insects, crustaceans, millipedes and centipedes) arachnids have segmented bodies with jointed limbs. Their hard exoskeletons require frequent and major reconstruction as the animal grows and develops, they have open circulatory systems, and lack the oxygen-carrying hemoglobins or myoglobins of mammals, instead using copper-based hemocyanins. Considering all these similarities among the arthropods, what functional differences would give particular interest to spider physiology? Some are obvious, such as their ability to secrete extremely tough silks and lethal venoms, essential for the ecologically important function of spiders; capturing insects. It is less obvious that spiders’ lives depend on detection of mechanical signals; some generated by the insect prey and others by their partners during often elaborate courtship behavior (Foelix 2011). Because of this, spiders have highly developed mechanosensory systems. While spiders use other senses, including chemical senses and vision, particularly important for the impressive visual prey detection of jumping spiders, the archetypal spider image is the patient predator waiting for mechanical vibrations in the web that signal approaching prey.

Not all spiders build webs, some dwell in the ground or on leaves and subdue passing prey, but they all use mechanical senses for detecting prey and predators, as well as for proprioception and communication with potential mates (Foelix 2011). Detection of touch, stress and vibration are performed by an array of specialized mechanosensory organs (Barth 2002), and the unique and advantageous structures of some of these organs have made them important physiological models of mechanical sensation with widespread application to other animals, including humans (French & Torkkeli, 2012). One limitation to research in spider physiology has been a lack of the molecular and genetic information that is available for more common model species such as mouse and Drosophila. This situation is changing rapidly with sequenced genomes now available for several arachnids, including the medically and agriculturally important ticks and mites, and the first complete spider genomes (Sanggaard et al. 2014). Transcriptome sequencing has also been applied to spiders, including the Central American wandering spider, Cupiennius salei (French et al. 2014). This large species (leg span 10-20 cm, typical adult weight 5 g) has been used as a model of spider sensory physiology, behavior and development since the 1960s (Barth 2002), with breeding colonies in Europe and North America. For more information see
http://asf-pht.medicine.dal.ca/Pictures/Cupiennius_2003/Cupiennius.html

Mechanoreceptors and sensory transduction

Mechanotransduction is a process where a sensory neuron detects a mechanical stimulus, such as touch or vibration, and converts it to electrical signals that can propagate along the nerve toward the central nervous system. This process is thought to involve mechanosensitive ion channels in the sensory dendrites. Additional protein molecules probably tether these channels to extracellular matrix and to the cytoskeleton. This arrangement makes it difficult to isolate and investigate mechanotransduction channel proteins, and although several have been identified as good candidates, no complete mechanoreceptor structures have been identified in any animal model. This is partially due to the difficulty of performing physiological experiments on vertebrate mechanoreceptors because their sensory endings are small, often in difficult locations, and are at long distances from their cell bodies in the dorsal roots, cranial, or autonomic ganglia.

Spider mechanoreceptors

Thousands of touch detecting hairs cover the spider exoskeleton, and their legs have extremely flexible hair sensilla (trichobothria) that detect air movements. Research into both types of sensory organs in Cupiennius have provided a wealth of information about how sensory stimuli are detected and transformed into neural signals (Barth 2002). Spiders also have thousands of slit sensilla that detect cuticular strains and vibrations. Slit sensilla are unique to arachnids, with some functional similarities to the strain detecting campaniform sensilla of insects. They occur singly or in groups and are located in all body regions, but particularly near joints in the legs (Barth 2002). A compound lyriform slit sensillum VS-3 (Fig. 1) on the anterior side of spider leg patella provides a particularly useful mechanoreceptor model for research. The mechanosensory neurons are large enough to be seen during electrophysiological and optical experiments, and are accessible to pharmacological manipulations. Each VS-3 organ contains 7-8 slits, with a pair of large sensory neurons innervating each slit, allowing intracellular recording and voltage clamp during mechanical stimulation of the slits, so that receptor current, receptor potential and action potentials can all be observed (French & Torkkeli 2012).

The mechanotransduction channels in VS-3 neurons are selective to Na⁺, impermeable to Ca²⁺ and open more easily at acidic pH. The receptor current through these channels is inhibited by the epithelial Na⁺ channel blocker amiloride. Most of the receptor current decays rapidly, with a time constant of ~6 ms, but there is also a smaller, slower component with a time constant of 100 ms. The single channel conductance estimated by steady noise analysis was ~7.5 pS, and there are only ~500 mechanotransduction channels per cell. Open probability of the mechanotransduction channels changed from near zero at rest to near unity just after a step stimulus. Most of these properties are quite different to the channels responsible for mechanotransduction in vertebrate auditory hair cells, but similar to channels found in some mammalian and nematode touch sensitive neurons. The spider mechanotransduction channels are particularly similar to the roundworm (Chaeorhabditis elegans) MEC channels that are members of the epithelial sodium channel/acid sensitive ion channel family (ENaC/ASiC) (French & Torkkeli 2012). Current research aims to identify and locate the Cupiennius mechanotransduction channel molecules and to understand their functional properties.

Modulation of spider mechanoreceptor neurons

Animals are bombarded by vast amounts of sensory information from their external and internal environments but only relevant information is delivered to the brain and used to modify behavior. Sensory organs detect stimuli in their own modality, but transmission of that information can be intensified, attenuated, or even completely blocked along the sensory pathways. Mechanosensory afferent nerve terminals entering the central nervous systems of all animals receive presynaptic, largely inhibitory, modulation but mechanoreceptors and pain receptors can also be modulated in the periphery, either by direct efferent innervation or by circulating neurochemicals. Well-known specialized examples are the fusimotor control of muscle spindles and the efferent innervation of hair cells in the inner ear.

One of the most fascinating properties of spider peripheral mechanoreceptors is the abundant and complex efferent innervation that reaches even the most peripherally located parts of the sensory neurons. Efferent neurons form synapses with the axons, cell bodies and dendrites of the mechanosensory neurons (Fig. 1; French & Torkkeli 2012). There are also synapses between neighboring efferent neurons and between efferent neurons and glial cells. These presynaptic sites have multiple types of synaptic vesicles and the efferent fibers are immunoreactive to antibodies against g-aminobutyric acid (GABA), glutamate, acetylcholine and octopamine. There is also evidence that more than one neurotransmitter can be released from a single efferent nerve. Each of the above mentioned transmitters modulates the sensitivity of VS-3 neurons by acting on a variety of ionotropic and metabotropic receptors, providing fine control of mechanosensation by tuning their sensitivity to different, behaviorally relevant vibration frequencies. VS-3 neurons provide an exceptional model of mechanoreceptor modulation because pharmacological agents can be applied during intracellular recordings and mechanical stimulation to investigate which agonists bind to these receptors, how this binding affects signal propagation within the cell, what types of signaling pathways are involved and how they can be inhibited.

The discovery of genes for several transmitter proteins in the C. salei transcriptome (French et al. 2014) is advancing research into sensory neuromodulation. The transcriptome contains genes for numerous adrenergic G-protein coupled receptors as well as ligand gated cys-loop receptor subunits, homologous to human and insect counterparts (French et al. 2014). Some of these receptors are already known to have a multifaceted role in modulating spider mechanoreception. Cys-loop receptors are targets of psychoactive drugs such as benzodiazepines and general anesthetics in humans as well as insecticides and antiparasitic agents, such as avermectins and neonicotenoids in invertebrates, with importance for human health and environment. Electrophysiological work has revealed differences in the ways that various transmitter receptor agonists modulate the sensitivity of VS-3 neurons. Even the same agonist can produce different effects in different neurons or with different stimulation. Future work will try to discover which receptors are located in the various spider mechanosensory neurons and how they modulate their functions.

Calcium-based sensory feedback in VS-3 neurons

When VS-3 neurons are mechanically stimulated their membrane potential depolarizes and they fire action potentials. Depolarization leads to opening of voltage activated Ca²⁺ channels that are widely distributed through all regions of the neurons. Ca²⁺-sensitive dyes have been used to investigate changes in intracellular Ca²⁺ concentration during mechanical stimulation, action potential firing and when neurotransmitters are applied. The effects of increased Ca²⁺ concentration were explored using controlled release of ‘caged’ Ca²⁺ molecules by flash photolysis, and were found to almost completely eliminate the mechanically-activated receptor current (Höger et al. 2010). The dose response relationship for receptor current reduction versus Ca²⁺ concentration indicated that two Ca²⁺ are needed to inhibit each mechanotransduction channel. This mechanism presumably provides a negative feedback control that limits the depolarization produced by mechanical stimulation. GABA application also caused a large increase in intracellular Ca²⁺ concentration, suggesting that this transmitter mediates its effect at least partially via Ca²⁺ and the feedback control mechanism.

Dynamically varying biomaterial properties – the metatarsal vibration receptor

Arthropod cuticles are variable and complex materials. The basic building block is chitin, a polymer of N-acetylglucosamine, but the hardness, elasticity and other physical properties are varied by the addition of proteins and other molecules in a process called sclerotisation to provide the range from soft, flexible joint membranes to the hard materials of fangs and carapaces. Spider cuticle typically consists of many layers of chitin, whose organization and thicknesses also affect the biomechanical properties. Recent work on the spider metatarsal vibration detector provides a striking illustration of dynamically changing material properties contributing to physiology.

The metatarsal vibration detector is another compound lyriform slit-sense organ, located on the tarsal-metatarsal joint and called HS-10 (Fig. 2). Substrate vibration is directly transmitted to the tarsus, as the most distal leg segment, and then used to compress the 21 slits via a cuticular cap structure at the end of the metatarsus. The organ is extremely sensitive to vibration; movements in the nanometer range at frequencies up to 1000 Hz elicit action potentials in the sensory neurons. However, behavioral and physiological measurements show that the HS-10 organ also responds to lower frequencies ranging from 0.1 – 40 Hz if the movements are significantly larger, in the range of 10 – 100 µm. This transition from high to low frequency sensitivity occurs at about 40 Hz, and probably allows the organ to be used for proprioceptive feedback of rotation in the tarsal-metatarsal joint (Barth 2002).

How can a delicate sensory structure cope with such a large range of displacements without damage? Experiments using a wide range of biomechanical measurement techniques, including atomic force microscopy, X-ray microtomography and scanning acoustic microscopy, have demonstrated that the structure of the cuticular metatarsal cap is crucial (Fig. 2). The material of the cap, which transmits tarsal movements to the closely adjacent slits, transitions between rubbery and glassy states depending on a range of factors that include temperature, vibration frequency and hydration state (Erko et al. 2015). Large, low frequency movements find the cap in a rubbery, compressive state that strongly attenuates pressure from the tarsus, while higher frequency movements cause transition of the cap to a more rigid, glassy state that can transmit them without significant attenuation (Fig. 2). These mechanical properties of the cap depend critically on the chemical composition and layered structure of the cuticle to create the correct level of hydration in the normal temperature range that the animal experiences.

Conclusions

This essay has only covered a few of the recent findings in the sensory physiology of spiders, and much clearly remains to be learned. The emphasis on mechanical processes reflects the particular lifestyle of these animals, with its reliance on mechanical signaling for major behavioral modes, but the information being gathered promises to be important for understanding the physiology of other species, including humans. Spider mechanoreceptors have also provided inspiration for technical advances, such as creation of an artificial vibration sensor based on slit sensilla with ultrahigh sensitivity that can detect physiological signals (Kang et al. 2014), so contributions to prosthetics and robotics may also arise in the future.

References

Barth FG (2002). A spider’s world. Senses and behavior. Berlin Heidelberg New York: Springer-Verlag.

Erko M, Younes-Metzler O, Rack A et al. (2015). Micro- and nano-structural details of a spider’s filter for substrate vibrations: relevance for low-frequency signal transmission. J R Soc Interface 12, 20141111.

Foelix RF (2011). Biology of spiders. 3rd Ed. New York, NY: Oxford University Press.

French AS, Torkkeli PH (2012). Sensory receptors and mechanotransduction. In: Cell Physiology Sourcebook. Ed: Sperelakis N. San Diego, London, Boston, New York, Sydney, Tokyo, Toronto, Academic Press. 633-647.

French AS, Li AW, Meisner S, Torkkeli PH (2014). Upstream open reading frames and Kozak regions of assembled transcriptome sequences from the spider Cupiennius salei. Selection or chance? Gene 539, 203-208.

Höger U, Torkkeli PH, French AS (2010). Feedback modulation of transduction by calcium in a spider mechanoreceptor. Eur J Neurosci 32, 1473-1479.

Kang D, Pikhitsa PV, Choi YW et al. (2014). Ultrasensitive mechanical crack-based sensor inspired by the spider sensory system. Nature 516, 222-226.

Sanggaard KW, Bechsgaard JS, Fang X et al. (2014). Spider genomes provide insight into composition and evolution of venom and silk. Nat Commun 5, 3765.