Exploring brain diversity in crustaceans: sensory systems of deep vent shrimps

Exploring the diversity of crustacean brains

Among the arthropods, crustaceans represent an important subgroup that displays a large diversity of sizes, morphologies, lifestyles, and life histories (for a review, see Schram, 2013). Specifically, members of the malacostracan crustaceans (Figure 1) have colonized habitats extending from the deepest ocean trenches and hydrothermal vents, through coastal, estuarine, to freshwater ecosystems. They also display a rich repertoire of complex behavioral patterns related to finding food, shelter, and mating partners; kin recognition and brood care; and orientation and homing. Complex social interactions include the establishment of dominance hierarchies, communal defensive tactics, the occupation of common shelters, and cooperative behavior during long-distance, offshore seasonal migration (reviewed by Breithaupt and Thiel, 2011; Derby and Thiel, 2014; Duffy and Thiel, 2007; Thiel and Walting, 2015). Their striking morphological, behavioral, and ecological diversity promotes this taxon as an interesting group of organisms for those neurobiologists interested in evolutionary transformation of brain structures and evolutionary diversification of neuronal circuits, specifically considering the antagonistic action of phylogenetic and functional constraints (Sandeman et al., 2014a, 2014b; Strausfeld, 2012). For example, crustaceans such as krill (Euphausia superba), which dominate the vast water bodies of the Southern oceans (Hempel, 1987), have evolved specific mechanisms of sensory-motor integration, which facilitates swimming in formation while schooling (Patria and Wiese, 2004). Cleaner shrimps, which are famous for their cleaning behavior of different species of coral reef fish and serve as iconic examples for communication across the vertebrate/invertebrate split (Urocadirella sp., Becker et al., 2005), display specific adaptation in their central mechanosensory pathways (Stenopus hispidus, Krieger et al., 2019). Fiddler crabs of the genus Uca are characteristic crustaceans of equatorial intertidal mud flats and serve as models to analyze crustacean visual ecology and the neuroethology of homing behaviors (Tomsic, 2016; Zeil and Hemmi, 2006).

Figure 1:

Phylogenetic relationships of malacostracan crustaceans (modified from Harzsch and Krieger, 2018; compiled after Richter and Scholtz, 2001; and Wirkner and Richter, 2010). The green asterisk show the phylogenetic position of the deep vent shrimp Rimicaris exoculata among Caridea.

We are interested in determining how the sensory landscape that the animals analyze in their various habitats is mirrored in the phenotype of the sensory systems and the morphology of the primary sensory brain areas, and against this background, we have previously studied the brains of a large diversity of malacostracan crustaceans (Figures 1 and 2). By describing the relative proportions of brain areas known to be involved in processing certain types of sensory input and comparing these proportions to other crustaceans’ brains may reflect how important these senses are for the animals to analyze their environment (Sandeman et al., 2014a, 2014b). Specifically, analyzing the nervous system of crustacean species from extreme habitats may provide deeper insights into the functional adaptations, which drive the diversification of crustacean brain structure. One obvious example comes from crustacean living in lightless habitats, an environmental factor that promotes the simplification of visual systems (Ramm and Scholtz, 2017; Stegner et al., 2015). Representatives of the malacostracan crustaceans have also invaded terrestrial habitats multiple times independently (for a review, see Hansson et al., 2011) and representatives of isopod crustaceans survive and reproduce not only in our European forests and gardens but also in desert salt pans of North Africa. Neurobiological studies suggest that terrestrial isopod crustaceans have failed to evolve aerial olfaction during their evolutionary diversification from marine ancestors (Harzsch and Hansson, 2008), whereas representatives of terrestrial hermit crabs have invested huge amounts of neuronal tissue in their olfactory pathway display a superb aerial sense of smell (Harzsch and Hansson, 2008; Krieger et al., 2010). Here, we explore what another example of crustaceans subjected to strong environmental constraints can teach us about the functional adaptations of crustacean sensory systems.

Figure 2:

Brain anatomy of malacostracans (modified from Sandeman et al., 2014a, 2014b). The subdivisions of the proto-, deuto-, and trito-cerebrum can be identified by their shape and color provided in the key. (A) Malacostraca. (B) Stomatopoda. (C) Eureptantia. (D) Amphipoda (Parhyale hawaiensis; from Wittfoth et al. 2019). (E) Stenopodidea (Stenopus hispidus; from Krieger et al., 2019). (F) Anomala (Birgus latro). (G) Isopoda (Saduria entomon). (H) Caridea (Rimicaris exoculata; from Machon et al., 2019).

Abbreviations: Visual neuropils: La, lamina; Me, medulla; Lo, lobula; LoP, lobula plate neuropil. Protocerebrum: HN (HN1, HN2), hemiellipsoid body (subdivisions thereof); TM, terminal medulla; AMPN, anterior medial protocerebral neuropil; PMPN, posterior medial protocerebral neuropil; PB, protocerebral bridge; CB, central body; LL, lateral lobe. Deutocerebrum: OL, olfactory lobe; LAN, lateral antennular (antenna 1) neuropil; OGTN, olfactory globular tract neuropil; AcN, accessory lobe; MAN, median antennular (antenna 1) neuropil. Tritocerebrum: AnN, antennal (antenna 2) neuropil; TN, tegumentary neuropil; Receptors, chiasmata and tracts: R, retina; CH1, first visual chiasm; CH2, second visual chiasm; OGT, olfactory globular tract. Note that brain schemes are aligned according to their neuroaxis (not to their arrangement within the cephalothorax) and are not equally scaled.

Introduction to vent shrimp sensory ecology

Since their discovery in the late 1980s, shrimps from deep hydrothermal vents have sparked interest among biologists because of their particular habitat and lifestyle. In addition to the high hydrostatic pressure and the absence of sunlight that characterize their environment, vent shrimps live in the surrounding areas of black smoker hydrothermal vents (Figure 3A, B). Hydrothermal fluids are hot (up to 350 °C) and rich in potentially toxic chemicals (Charlou et al., 2000, 2002, 2010). A key adaptation of the Rimicaris exocualta shrimps is their symbiotic association with chemoautotrophic bacteria that provide energy to their hosts by oxidizing the fluid chemicals. Hence, in light of their habitat, the fairly extreme conditions they must cope with, and their lifestyle, these vent shrimps are fascinating models to study biological adaptations to the environment.

Figure 3:

Swarms and morphology of the deep vent shrimp Rimicaris exoculata. A,B: Swarms of thousands of shrimp are surrounding the walls of black smoker hydrothermal vents at the TAG vent site (3600 m depth), Mid-Atlantic Ridge (©IFREMER/Nautile6000, BICOSE 2018 cruise). C: Dorsal view of the cephalothorax showing voluminous gill chambers covered by the branchiostegites, dorsal eyes (namely the ocular plate) with two elongated retinae fused in the anterior region, and sensory appendages (antennae 1 and 2). A1l, lateral flagellum of antenna 1; A1m, medial flagellum of antenna 1; A2, antenna 2; bs, branchiostegite; ocp, ocular plate. Scale bar = 5 mm. Modified from Machon et al. (2019).

An important question in vent shrimp ecology is to understand the sensory mechanisms that the animals use for orientation in their dark environment, to select a suitable habitat, to find food, to detect congeners, or to locate active smokers to supply their symbiotic bacteria with chemicals. Because they colonize the vicinity of black smokers, it has been suggested that they might exploit abiotic factors of the hydrothermal fluid as orientation cues (Sarrazin et al., 1999; Segonzac et al., 1993). The commonly considered factors are the fluid chemicals, temperature, and thermal radiation emitted by the hot fluids. The shrimp may have evolved specific abilities to detect these stimuli, and the related senses could play a major role in orientation within the habitat.

Therefore, analyzing the architecture of the peripheral and central sensory systems provides new insights into the sensory biology of vent shrimps (Figures 3 and 4), knowledge that is essential for learning their ecology and long-term evolutionary adaptations. Comparative studies with shallow water relatives are especially relevant in this context to reveal sensory adaptations to the habitat and related evolutionary processes (Derby and Weissburg, 2014).

Figure 4:

Brain architecture of the deep vent shrimp Rimicaris exoculata. 3D Reconstructions (A, C) and schematic representation (B) of the brain and neuropils, viewed from a dorsal, slightly anterior direction (A, B) and from the left (C). In C, the open white arrow point towards anterior of the body axis. The clusters of cell somata are not shown. The 3D reconstructions are based on an image stack obtained by serial sectioning of paraffin-embedded material. A1Nv, antenna 1 nerve; A2Nv, antenna 2 nerve; AnN, antenna 2 neuropil; CB, central body; ENv, eye nerve; HN, hemiellipsoid body neuropil; La, lamina; LAN, lateral antenna 1 neuropil; Lo, lobula; MAN, median antenna 1 neuropil; Me, medula; MPN, median protocerebral neuropil; Ob, onion bodies; og, olfactory glomerulus; ON, olfactory neuropil; PT, protocerebral tract; PNT, projection neuron tract; TM, terminal medulla neuropil; TN, tegumentary neuropil; TNv, tegumentary nerve. Scale bar = 500 µm. Modified from Machon et al. (2019).

A unique visual system

The first anatomical observations of vent shrimp specimens had led to the reassessment of the general belief that deep sea animals were entirely blind (Van Dover et al., 1989). Vent shrimps possess highly modified eyes, which, in Rimicaris exoculata, consist of large ocular plates located at the anterodorsal region of the cephalothorax (Figures 3C and 5A). Their retinal structure consists of a smooth cornea that covers a layer of photoreceptive rhabdoms, under which reflective cells of the tapetum are located (Figure 5B). Several authors have shown that the rhabdoms are hypertrophied, which would maximize the absorption of light (Chamberlain, 2000; Jinks et al., 1998; Machon et al., 2019; Nuckley et al., 1996; O'Neill et al., 1995). However, in the specimen presented here, the rhabdoms are degenerated, which appears to be a consequence of a dramatic deterioration of the retina following the intense light exposure during sampling, as shown by Herring et al. (1999). In addition, the dioptric apparatus, which is a characteristic of typical crustacean compound eyes, is lacking, indicating that these eyes cannot form images. Functional rhodopsin-like visual pigments were also identified in high quantities in the retina of R. exoculata (Van Dover et al., 1989) (Figure 5C). At the central nervous level, as in other malacostracans, vent shrimp present a suite of retinotopic visual neuropils (lamina, medulla, and lamina) (Figure 5D) that are strongly reduced in size and that are fused with the median brain (Figures 4 and 5A), coinciding with the absence of eyestalks (Machon et al., 2019).

Figure 5:

Visual system of the deep vent shrimp Rimicaris exoculata. A: Drawing showing the location of the visual organs and their connection to the brain (dorsal view; from Van Dover et al., 1989). B: Histological section of the retina. C: Bleaching difference spectrum of visual pigment in R. exoculata (solid trace) that maximally absorbs at 500 nm, typical of rhodopsin, and creates a new pigment at 367 nm, which corresponds to retinaloxime. The same features are seen in the difference spectrum of the frog rhodopsin (dashed line) (from Van Dover et al., 1989). D: Histological frontal section of the visual neuropils (from Machon et al., 2019). Scale bars = 100 µm. (2–3), cell cluster 2–3; br, brain; dR, degenerated rhabdoms; ENv, eye nerve; La, lamina; Lo, lobula; Me, medulla; ocp, ocular plate; pc, pigment cell; T, tapetum.

Because selective pressure favors the reduction of unsolicited nervous tissues (Klaus et al., 2013; Moran et al., 2015; Niven and Laughlin, 2008), the presence of visual neuropils, together with a seemingly functional retina, is indicative of an effective visual system in vent shrimp, which implies that a light signal does occur in their environment. The prominent hypothesis is that vent shrimp might have evolved a highly sensitive retina, to the detriment of spatial resolution, to detect the very dim light of the thermal radiation emitted by the hot (up to 350 °C) hydrothermal fluids as they exit the black smoker (Pelli and Chamberlain, 1989; Van Dover and Fry, 1994; Van Dover et al., 1988, 1996).

A common olfactory system

Asking how the shrimps can detect active vent areas, biologists had previously proposed that they locate black smokers by chemotaxis, reacting to specific chemical compounds of the hydrothermal fluids (Segonzac et al., 1993). This hypothesis was soon supported by electrophysiological recordings of concentration-dependent sulfide sensitivity from the first antenna nerves of R. exoculata (Figure 6A), and the authors suggested that these shrimp may present enhanced olfactory abilities to detect naturally occurring sulfide gradients in the near field of the vents (Jinks et al., 1998; Renninger et al., 1995). The complexity of the animal's peripheral and central olfactory pathways may reflect these functional demands in that their chemical senses evolved to compensate for the underperformance of the visual system.

Figure 6:

Olfactory system of the deep vent shrimp Rimicaris exoculata. A: Recordings from a bundle of nerve fibers in the lateral antenna 1 showing massed action potential generation in response to the application of different concentrations of sulphide (from Renninger et al., 1995). B–D: Electron microscopy images of the lateral flagellum of antenna 1 (B) bearing the aesthetasc olfactory sensilla (C), often associated with a bacterial coverage (D) (from Zbinden et al., 2017, 2018). E: Sagittal histological section of the olfactory neuropil. The white open arrow points towards anterior of the body axis. F: Horizontal immunohistological section of the olfactory neuropil, triple-labeled for synapsin immunoreactivity (SYN, magenta), allatostatin-like immunoreactivity (AST, green) and a nuclear marker (NUC, cyan). F’ shows an enlargement view of the olfactory glomeruli (E and F from Machon et al., 2019). (9–11), cell cluster 9–11, A1, antenna 1; aes, aesthetasc; b, bacteria; bG, base region of the glomerulus; cG, cap region of the glomerulus; lF, lateral foramen; mF, medial foramen; ON, olfactory neuropil; scG, subcap region of the glomerulus. Scale bars B = 500 µm, C, E, F, F’ = 100 µm, D = 20 µm.

However, a comparative study of the first antenna and the specialized olfactory aesthetasc sensilla (Figure 6B, C) of R. exoculata, three other vent shrimp species, and a closely related shallow-water species revealed no specific adaptation between hydrothermal and coastal species regarding the chemosensory organs morphology (Zbinden et al., 2017). Additionally, recent electrophysiological recordings from the antennae in the vent shrimp Mirocaris fortunata and a related member of the Caridea, the shallow-water shrimp Palaemon elegans, showed that sulfide detection is not specific to the vent species (Machon et al., 2018). Hence, no features indicative of a more sophisticated olfactory performance have been identified from the peripheral system of vent shrimp thus far. Nevertheless, a noticeable aspect of vent shrimp is the dense coverage of their antennae and aesthetasc sensilla by bacteria (Figure 6D), which were identified to be similar to known chemoautotrophic sulfur oxidizers and may thus influence the chemosensory system in several ways, but their specific roles are unknown yet (Zbinden et al., 2018).

Regarding the sensory centers, a thorough description of the brain neuroanatomy in R. exoculata allowed to search for a differential investment in the olfactory neuropils that might reflect an enhanced olfactory performance (Machon et al., 2019). These neuropils are lobe-shaped and composed of olfactory glomeruli that are radially arranged around the periphery of a non-synaptic core (Figure 6E, F) and subdivided into three regions (Figure 6F) as observed in several decapod taxa (Harzsch and Krieger, 2018). Structural features that may be linked to the efficiency of the olfactory system were compared between R. exoculata and other malacostracan species, showing that the olfactory neuropils are not overly hypertrophied in the vent species. Overall, the structural complexity of the olfactory system does not suggest that R. exoculata presents adaptations to the specific chemosensory landscape at vents, and olfaction is probably not a particularly dominant sensory modality in vent shrimp.

The hemiellipsoid body, a higher integrative brain center

Higher integrative centers in the malacostracan brain provide the neuronal substrate for more sophisticated processing and receive input exclusively from second- or higher-order neurons but not from any primary sensory afferents. Interneurons within such centers typically respond to the stimulation of several different sensory systems (reviewed by Sandeman et al., 2014a, 2014b). In the malacostracan brain, the (bilaterally paired) complex of hemiellipsoid body and terminal medulla (HE/MT) is one of these higher integrative centers. It is targeted by the axons of the olfactory projection neurons as output pathway of the olfactory system and also receives input from the visual neuropils (for reviews, see Derby and Weissburg, 2014; Harzsch and Krieger, 2018; Schmidt, 2016). Substantial morphological modifications and changes related to the relative proportion of types of input occurred during the evolutionary elaboration of the HE/MT complex (reviewed, e. g., by Harzsch and Krieger, 2018; Machon et al. 2019). Nevertheless, recent studies suggest that, despite many morphological differences, the MT/HE complex of crustaceans and the iconic mushroom bodies of insects share common architectural, physiological, and neurochemical features, suggesting a homology of their very basic neuronal circuitry (e. g., Maza et al., 2016; Wolff et al., 2012, 2017; Wolff and Strausfeld, 2015; Strausfeld and Sayre et al. 2020, Sayre and Strausfeld 2019). Considering such basal anatomical similarities of the crustacean hemiellipsoid body and insect mushroom body, Wolff et al. (2017) suggested an involvement of both structures in place memory. Furthermore, because of its close anatomical association with the olfactory system as target of the projection neuron tract, evolutionary (Sullivan and Beltz, 2001, 2004), and functional considerations (Harzsch and Krieger, 2018; Sandeman et al., 2014a, 2014b; Strausfeld, 2012) have focused on the possible roles of these centers in higher-order olfactory processing and have suggested that the structural elaboration and size of hemiellipsoid bodies largely mirror the importance of the central olfactory pathway in a given brain (e. g., Harzsch and Hansson, 2008; Harzsch and Krieger, 2018; Krieger et al., 2010).

In the brain of R. exoculata, an inconspicuous and moderately developed olfactory neuropils (see previous section A common olfactory system) contrast with disproportionally large hemiellipsoid bodies (Figure 4). Because visual input also plays a minor role, Machon et al. (2019) suggested that the impressive hemiellipsoid body of R. exoculata may fulfill functions in addition to higher-order sensory processing in that they perhaps serve as the neuronal basis for a sophisticated place memory. For survival in the extreme, lightless habitat of R. exoculata, an excellent place memory may be essential for avoiding the dangerously hot vent chimneys and memorizing emission sites of hydrothermal fluids rich in those chemicals on which their endosymbiont bacteria depend (Machon et al. 2018). To test this hypothesis, other representatives of the taxon Alvinocarididae should serve as a model because behavioral experiments with R. exoculata are technically challenging because they need to be carried out in the pressurized aquariums. Other vent shrimp species such as M. fortunata also display pronounced hemiellipsoid bodies (see next section Conclusions and perspectives), can be maintained at atmospheric pressure, and therefore are more suited for behavioral observations.

Conclusions and perspectives

The case of the vent shrimp discussed here provides new insights into aspects of the evolutionary transformation of crustacean brains and their associated sensory organs. Layered visual neuropils are present within R. exoculata brains with an arrangement similar to that of phylogenetically related shallow water shrimps with fully developed compound eyes, although much smaller. The compound eyes of R. exoculata ancestors, during the evolutionary diversification of this group, were transformed to a flattened but seemingly functional retina without notable spatial resolution but most likely can detect the very dim light of the thermal radiation (Pelli and Chamberlain, 1989; Van Dover and Fry, 1994; Van Dover et al., 1988, 1996). The fact that, despite these major modifications of sensory input, the number and principal arrangement of the three visual neuropils remain conserved is remarkable and suggest a selective pressure acting to retain this arrangement. Concerning the HE/MT complex, the brain of R. exoculata has taught us that the structural elaboration of these neuropils may be less dependent on olfactory input than we previously thought and instead points to other functions that were previously less in the focus of crustacean neurobiologists (Machon et al., 2019).

Future studies in the field of research exposed in this paper should, in addition to qualitative aspects of brain structure, pay more attention to numerical and volumetric aspects. For example, determining across-species variations in numbers of olfactory sensory neurons and olfactory interneurons as well as glomerular numbers and volumes can be instructive for discussing functional aspects of crustacean olfactory systems such as the wiring logic from receptor to glomerulus (Harzsch and Krieger, 2018). We expect that determining neuropil volumes and neuronal numbers of other sensory systems and comparing these across species will be essential as a basis for new insights into brain adaptations related to ecological complexity of the habitat, lifestyle, and locomotion, and perhaps also biological processes such as invasiveness, sexual dimorphisms, sociality, and aging. As for crustaceans from extreme habitats, the related vent shrimp species from the Mid-Atlantic Ridge within the Alvinocarididae, which include Rimicaris chacei, M. fortunata, and Alvinocaris markensis (Gebruk et al., 2000), should be suitable for such comparisons, although they are less numerous than R. exoculata and are, therefore, even more difficult to sample. Preliminary results suggest that all these species present an overall similar brain pattern as in R. exoculata (Figure 7). Other vent crustaceans, such as the hydrothermal crab Segonzacia mesatlantica, could be also good models for future investigations (Charmantier-Daures and Segonzac, 1998; Matabos et al., 2015).

Figure 7:

Brain architecture of other Alvinocarididae vent shrimp species. A: Rimicaris chacei. B: Mirocaris fortunata. C: Alvinocaris markensis. Abbreviations: Ob, organ of Bellonci. Visual neuropils: La, lamina; Me, medulla; Lo, lobula. Protocerebrum: HN hemiellipsoid body; TM, terminal medulla; MPN, medial protocerebral neuropil; CB, central body. Deutocerebrum: OL, olfactory lobe; og, olfactory glomerulus; LAN, lateral antennular (antenna 1) neuropil; MAN, median antennular (antenna 1) neuropil. Tritocerebrum: AnN, antennal (antenna 2) neuropil; TN, tegumentary neuropil. Note that brain schemes are aligned according to the neuroaxis (not to their arrangement within the cephalothorax) and are not equally scaled.