Table 1.
Human phototransduction KEGG pathway (hsa04744) molecular homologs used for L. stagnalis blastp query.
Table 2.
Sequence data for rhodopsin proteins use for generation of the maximum-likelihood phylogenetic tree.
Fig 1.
The pathways for sensory information between the L. stagnalis visual system and the central ring ganglia provides a map to understanding visual processing in gastropod mollusks.
(A) Depiction of the external anatomy of L. stagnalis, illustrating anatomical features involved in sensory and visual perception (B) Photographed preparation of the internal optic sensory system illustrated by the left eye and left optic nerve connected to the central ring ganglia, via the intermediate n. labialis nerve, facilitating some afferent connections to the cerebral ganglia. Scale bar is indicated (scale bar = 0.5mm). (C) Schematic depiction of reported direct afferent projections to the central ring ganglia reported from the retina of both eyes in L. stagnalis specifically to the cerebral ganglia (blue dashed line), through the cerebral commissure (purple dashed line), to the bilateral statocysts (green dashed line), to the ipsilateral parietal ganglia (grey dashed lines) and to the visceral ganglia (orange dashed line). (D) Schematic depiction of select reported efferent (ascending/descending) projections from corresponding retinal afferent terminations in the central ring ganglia. Indicated are the cerebral-statocyst connection (blue dashed line), the crossing cerebral commissure connection to the retina and n. tentacularis (purple dashed line), from the statocysts to the ipsilateral pedal ganglia (green dashed line), from the parietal ganglia to the ipsilateral cerebral ganglia and the right pleural ganglia (grey dashed lines) and from the visceral ganglia to the left pleural ganglia and through the n. intestinalis to the digestive system (orange dashed line). Dots at the end of each projection indicate terminating points for each projection. Anatomical feature of the tentacles (yellow chevrons), eyes (red chevrons), optic nerves (blue chevrons), and cerebral ganglia (green chevrons) are noted. Ret.- retina, cc- cerebral commissures, LCeG- left cerebral ganglia, LPeG- left pedal ganglia, LPaG- left parietal ganglia, LPlG- left pleural ganglia, RCeG- right cerebral ganglia, RPeG- right pedal ganglia, RPaG- right parietal ganglia, RPlG- right pleural ganglia, VG- visceral ganglia, St.- statocyst, nl- n. labialis, nt- n. tentacularis, ni- n. intestinalis, npdi- n. pallialis dexter internus, npdi- n. pallialis dexter externus.
Fig 2.
Morphological characterization of the L. stagnalis eye illustrates the presence of key anatomical features involved in phototransduction and visual perception.
(A) Hematoxylin and eosin-stained sections of the L. stagnalis eye depict the general histology and structure of the dermal and ocular tissues. (B) Thionin stained section of the L. stagnalis eye depicts the dense nuclei staining in the somatic layer of retina. (C) Immunofluorescence staining of the L. stagnalis eye illustrating the general distribution of the cell nuclei and rhodopsin positive cells in the photopigment and somatic retinal layers of the eye and the peripheral dermal tissue. Hoechst-2293 positive nuclei are indicated in blue, and rhodopsin positive cells are stained with Anti-Octopus rhodopsin in green. Photopigment positive dermal photoreceptor cells (yellow chevrons), rhabdomeric membranes (grey chevrons) and somatic photopigment cells (white chevrons) in the retina are noted, and the lens is denoted ‘L’. Sections—7μm, Scale bars—50 μm.
Fig 3.
Electron microscopy of L. stagnalis retinal ultrastructure reveals detailed organization of the layered retina and distinct retinal subtypes.
Electron microscopic images of (A) longitudinal cross sections of the L. stagnalis retina show, from anterior to posterior, lens (l), microvillar layer (ml), pigment layer (pl), the somatic layer (sl), (B) and the neural layer (nl). (C) Higher magnification electron micrographs of the photoreceptor sensory cell (sc) and pigment cell (pg) distribution within the microvilli and pigment layers of the retina are shown, with the apical projections (ap) and microvilli (mv) indicated. Pigment granules (pgs) of the pigment layer are labelled. Directionality from the anterior (A) to posterior (P) portions of the eye are indicated. Sections—70 nm, Scale bars: A = 5 μm, B = 5 μm, C = 1 μm.
Fig 4.
Development of a neurobehavioral protocol using DeepLabCut to assess snail phototaxis response in vivo.
(A) Diagram of the experimental phototaxis arena where snails are tracked by an overhead infrared RaspberryPi camera while performing locomotion. (B) Pictorial representation of phototaxis testing protocol example frames from each of the three phases—acclimation, dark, and focal light- and the recording time for each phase. Noted is the consistent positioning of the snail on the opposite end of the focal light source at the start of each phase. (C) Schematic drawing and representative image collected from RaspberryPi recordings. Snail labelling of the head region (purple dot), top of the shell (yellow dot), bottom of the shell (orange dot) and shell apex (red dot) are done in the DeepLabCut GUI [26]. (D) General pipeline for DeepLabCut analysis of snail phototaxis behavior. Videos of snail phototaxis (20 fps) were acquired on RaspberryPi computers with recording infrared cameras and extracted from the computer for analysis through DeepLabCut. 50 frames from each video were extracted and critical body parts were labelled in the DeepLabCut GUI. Once labelling was completed, labelled frames were used to train the DeepLabCut machine learning model neural network on the placement of the labelled body parts, with the apex being the most consistent label throughout all videos, to learn the animals’ movement during phototaxis testing. Videos were then analyzed and the coordinates for each labelled body part were extracted to determine various parameters of animal movement.
Fig 5.
Assessing phototaxis abilities in vivo reveals that most animals exhibit patterns of positive phototaxis.
Representative travel trajectory plots of snails who (A) exhibit weak light sensitive behaviors and (A’) strong light sensitive behaviors, where the boundaries by which absolute trajectory length was calculated are noted. (B) Chart depicting the total number/ percentage of animals within the testing cohort (n = 29) who reached the focal light area during the focal light phase, exhibiting strong light sensitivity (n = 23, 79.3%) and those who did not enter the focal light area during the focal light phase, therefore exhibiting weak light sensitivity phototaxis (n = 6, 20.7%). Comparative plots of (C) mean absolute trajectory length travelled and (D) mean absolute trajectory speed during the dark and focal light phases of testing for strong light sensitive (n = 23) and weak light sensitive (n = 6) animals. For panels (C, D), unpaired two-way ANOVA and Tukey’s multiple comparisons tests were performed and statistical significance, if any, were reported. (* = p < 0.05; ** = p < 0.01; *** = p < 0.005, **** = p < 0.001; n.s. = not significant; two-way ANOVA with Tukey’s multiple comparison). +/-SEM for all groups are noted.
Fig 6.
The conservation patterns in the phototransduction pathway provides insights on mollusk light sensing modalities.
(A) Phenotypic patterns of critical components of visual system structure and function across eukaryotic animals are depicted in this phylogenetic tree rooted in choanoflagellates as the most divergent group of eukaryotes with photo-sensitive properties. While the evolution of photo opsins and the phototaxis behavior appears to predate the evolution of the canonical nervous system in Ctenophora, Cnidaria, and Bilateria, the presence of sensory eye organ and an organized retina appear to be exclusively found in Bilateria. Evident is the evolution of canonical retina and nervous systems in Mollusca and other protostomes, as well as the presence of phototaxis behaviors in bilaterians. Of note, this tree supports the hypothesis that ctenophores are the most early diverging animals at the base of Metazoans, not poriferans (i.e., sponges). (B) Average TPM expression level analysis of the L. stagnalis vertebrate and invertebrate phototransduction pathway homologs, mined from the L. stagnalis CNS (i.e., central ring ganglia) transcriptome [27] reveals reduced expression of both invertebrate and vertebrate phototransduction genes in the CNS of the snail. Reference genes glyceraldehyde-3-phosphate dehydrogenase (GAPDH), beta-actin (ACTb), and beta-tubulin (TUBB) were included in this study. Mean and +/-SEM for all mined genes are noted.
Fig 7.
In silico characterization of L. stagnalis RHO reveals a unique phylogenetic placement despite its canonical GPCR rhodopsin structure.
(A) Kyte-Dootlittle plot depicting transmembrane (>1) and cytoplasmic/extracellular (<1) spanning regions of L. stagnalis rhodopsin (RHO) reveals seven hydrophobic regions corresponding to seven transmembrane spanning alpha-helices. (B) Phylogenetic tree of various vertebrate and invertebrate rhodopsin proteins from model organisms critical to vision research, inferred by IQ tree using the LG+F+I+G4 model. Node support values resulting from 1,000 ultra-fast bootstrap replicates are indicated, and the scale bar indicates the number of amino acid substitutions per site. Phylogenetic subgroups of rhodopsins engaged in G-protein coupled mechanisms through Gt-, Gq- and Go-protein interactions are noted, revealing that L. stagnalis rhodopsin does not cluster with the canonical vertebrate and invertebrate rhodopsin and may function through unique G-coupled mechanisms.
Fig 8.
Evolutionary conservation of critical amino acid residues to rhodopsin’s light-sensitive function in vertebrate rhodopsins are present in gastropod mollusks.
Protein sequence muscle alignment of rhodopsin alpha-helices, identified through PROMALS3D, reveals deep conservation of rhodopsin proteins among representative animals in animal phyla commonly referenced in vision sciences. Amino acids residues involved in arrestin binding (green chevrons), comprising the major retinal ligand binding pocket of rhodopsin are shown (black chevrons), rhodopsin GPCR activation/inactivation (red chevrons) and G-protein binding (orange chevrons), and additional amino acids that cooperate with those involved in lining the ligand binding pocket but are preferentially involved in photoactivation of 11-cis-retinal are indicated (blue chevrons) are indicated. Of note are the conserved NPxxY and D(E)RY motifs, which are involved in rhodopsin’s conformational change in response to retinal photoisomerization and the ability of rhodopsin to reach the active conformation, respectively.
Fig 9.
AlphaFold2 predicted structure of L. stagnalis rhodopsin.
(A) Multicolored AlphaFold2-predicted secondary structure of the putative L. stagnalis rhodopsin with the N- and C- termini of the structure located within the extracellular and cytoplasmic regions, respectively, indicated. Also depicted on this structure are the predicted seven transmembrane and one cytoplasmic helix (TH1-TH8) noted. The L. stagnalis structure surface was colored to indicate (B) Electrostatic surface potential and (C) hydrophobicity, with the minimum, mean, and maximum scores for these calculations and scales pertaining to these parameters are noted.
Fig 10.
Predicted structural elements of L. stagnalis rhodopsin depict invertebrate specific structural features.
Alphafold2-predicted L. stagnalis rhodopsin structure aligned to the (A) resolved Bos taurus rhodopsin (PDB: 1U19) in its native state with bound 11-cis-retinol depicts the shorter transmembrane alpha-helix 5 (TH5) in B. taurus when compared to L. stagnalis. The alignment between L. stagnalis rhodopsin and (B) resolved T. pacificus rhodopsin (PDB: 2Z73) in its inactive state when bound to 11-cis-retinol to L. stagnalis rhodopsin. Root mean squared deviation (RSMD) values are noted.