Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Apr 28;533(5):e70055. doi: 10.1002/cne.70055

Molecular and Morphological Circuitry of the Octopus Sucker Ganglion

Cassady S Olson 1,✉, Aashna Moorjani 2, Clifton W Ragsdale 2
PMCID: PMC12036646  PMID: 40293445

ABSTRACT

The octopus sucker is a profoundly complex sensorimotor structure. Each of the hundreds of suckers that line the octopus arm can move independently or in concert with one another. These suckers also contain an intricate sensory epithelium, enriched with chemotactile receptors. Much of the massive nervous system embedded in the octopus arm mediates control of the suckers. Each arm houses a large axial nerve cord (ANC), which features local enlargements corresponding to each sucker. There is also a sucker ganglion, a peripheral nervous element, situated in the stalk of every sucker. The structure and function of the sucker ganglion remain obscure. We examined the cellular organization and molecular composition of the sucker ganglion in Octopus bimaculoides. The sucker ganglion has an ellipsoid shape and features an unusual organization: the neuropil of the ganglion is distributed as a cap aborally (away from the sucker) and a small pocket orally (toward the sucker), with neuronal cell bodies concentrated in the space between. Using in situ hybridization, we detected positive expression of sensory (PIEZO) and motor (LHX3 and MNX) neuron markers in the sucker ganglion cell bodies. Nerve fibers spread out from the sucker ganglion, targeting the surrounding sucker musculature and the oral roots extending to the ANC. Our results indicate that the sucker ganglion is composed of both sensory and motor elements and suggest that this ganglion is not a simple relay for the ANC, but facilitates local reflexes for each sucker.

Keywords: cephalopod, invertebrate, muscular hydrostat, neural circuitry, neuroethology, Octopus bimaculoides, sensorimotor control

Each of the prehensile suckers on the octopus arm contains a sucker ganglion embedded in the sucker musculature. The function of this ganglion has long been disputed. We provide molecular evidence that the sucker ganglion comprises sensory and motor elements and demonstrate circuit pathways for local processing as well as connections to central nervous system. Thus, the sucker ganglion is well situated for control of sucker reflexes and integrative sensorimotor processing.

graphic file with name CNE-533-e70055-g007.jpg

1. Introduction

The octopus arm is a remarkable appendage, famously able to contort in near infinite degrees of freedom (Kennedy et al. 2020; Hochner et al. 2023; Olson and Ragsdale 2023). The hundreds of suckers that line each arm are equally remarkable. These suckers are vital for the survival of the octopus, allowing them to catch prey, distinguish food from other objects, pass food to and from the mouth, manipulate objects, build structures, and tend to their eggs (Altman 1971; Packard 1988; Grasso 2008; Hanlon and Messenger 2018; Bidel et al. 2022; Buresch et al. 2022). In support of these behaviors, and like the arms themselves, each sucker can be moved independently and in near infinite degrees of freedom (Kier and Smith 2002; Röckner et al. 2023). The suckers are additionally profound sensory structures, each containing an intricate sensory epithelium enriched with chemotactile and mechanoreceptors, allowing sensory detection of the environment even in the absence of vision (Wells 1964; Gutnick et al. 2020; van Giesen et al. 2020; Buresch et al. 2022; Kang et al. 2023).

The octopus arm houses a substantial nervous system, much of it dedicated to the control of the suckers (Young 1963; Graziadei 1971; Neacsu and Crook 2024; Olson et al. 2025). There is a central nerve center: the massive axial nerve cord (ANC), equivalent to a spinal cord, running down the center of each arm. Within the ANC, a neuronal cell body layer (CBL) wraps around the neuropil (NP) forming a horseshoe pattern on the oral, or sucker, side (Graziadei 1971; Olson et al. 2025). On the aboral side, or away from the sucker, there is a cerebrobrachial tract, comprising a pair of fiber tracts connecting the arms and the brain (Graziadei 1971; Zullo et al. 2019; Olson and Ragsdale 2023). The CBL and NP of the ANC can be further divided into an aboral brachial territory with connections to the arm and an oral sucker territory with connections to the sucker (Rowell 1963; Olson et al. 2025). Down the long axis of the arm, this oral sucker territory of the ANC forms local enlargements over each sucker (Rossi and Graziadei 1954; Olson et al. 2025; these enlargements are sometimes referred to as ganglia; see Young 1971; Nixon and Young 2003; Nödl et al. 2016; Neacsu and Crook 2024). From these enlargements, the ANC issues nerve fibers that spatially tile each sucker to create a topographic map (Olson et al. 2025).

The arm also harbors a peripheral nerve center for the sucker: the sucker ganglion (Graziadei 1971). Given the strong coupling between the ANC and each sucker, what does the sucker ganglion do? Some early authors postulated a primarily motor role, with the sucker ganglion modulating local reflexes for the sucker stalk (Guérin 1908). Others included a sensory role, with some sucker ganglion cell bodies being primary sensory cells (Rowell 1963; Graziadei and Gagne 1976). Lastly, others proposed a dual sensory and motor role (Martoja and May 1955; Graziadei 1965, 1971). Here, we investigated the role of the sucker ganglion by interrogating its molecular composition and circuitry with contemporary molecular and cellular methods.

2. Materials and Methods

2.1. Animals

Adult Octopus bimaculoides (California two‐spotted octopus) were wild caught and purchased from the field collection venture, Aquatic Research Consultants, operated by Dr. Chuck Winkler (San Pedro, CA). After overnight FedEx delivery, octopuses were housed individually in 20‐gallon artificial seawater (ASW) tanks, equipped with a carbon particle filter, an aquarium bubbler, and a UV light. Octopuses were fed daily with fiddler crabs, bivalves, or frozen shrimp. ASW was prepared by resuspending pharmaceutically pure sea salt (Tropic Marin “classic,” Wartenberg, Germany) in deionized water to a concentration of 33 g/L.

Adults were anesthetized in 4% ethanol/ASW (EtOH/ASW, n = 13) and trans‐orbitally perfused with 4% paraformaldehyde/phosphate‐buffered saline solution (PFA/PBS, pH 7.4) delivered via a peristaltic pump (Cole Parmer Masterflex) through a 21½ gauge needle (Becton Dickinson). Throughout the procedure, hematopoietic tissue, consisting of left and right white bodies, was targeted alternately and iteratively. Arms were stored in 4% PFA/PBS overnight at 4°C. These arms were then either cut into 2‐ to 4‐cm pieces for immediate processing or stored whole in diethyl pyrocarbonate (DEPC)‐treated PBS at 4°C.

These cephalopod experiments were performed in compliance with the EU Directive 2010/63/EU guidelines on cephalopod use, University of Chicago Animal Resources Center oversight, and the University of Chicago Institutional Animal Care and Use Committees (Fiorito et al. 2015; Lopes et al. 2017).

2.2. Tissue for Sections

Arm blocks were saturated with 30% sucrose/4%PFA/PBS at 4°C, rinsed with 30% sucrose/PBS, and infiltrated with 10% gelatin/30% sucrose/PBS for 1 h at 50°C. Tissue was embedded in 10% gelatin/30% sucrose/PBS, postfixed in 30% sucrose/4%PFA/PBS, and stored at −80°C. Serial arm sections were cut on a freezing microtome (Leica SM2000R) at 28‐ to 50‐µm thickness and collected in DEPC‐PBS. Sections were mounted and dried on charged, hydrophilic glass slides (TruBond380, Newcomer Supply, Middleton, WI) before storage at −80°C until further processing.

2.3. cDNA Synthesis and Cloning

For RNA extraction, dissected octopus central brain and arm tissues were flash‐frozen on dry ice and stored at −80°C. After homogenizing tissue with a micropestle, RNA was extracted with Trizol Reagent (Invitrogen) and phasemaker tubes (Invitrogen). cDNA was synthesized using the SuperScript III 1st‐strand cDNA kit (Invitrogen) following manufacturer instructions. RNase‐free water (Sigma–Aldrich) was used to dilute cDNA, which was stored at −20°C until use.

PCR primers were designed with MacVector software (version 12.6.0) or PrimerBlast from NCBI (Table 1). PCR reaction solutions were incubated using the T100 thermocycler from BioRad at 95°C for 5 min. Then, the solutions underwent 35–40 rounds of amplification cycles: 95°C for 30 s, 52–57°C for 45 s, and 72°C for 1 min. A final elongation step was performed at 72°C for 10 min.

TABLE 1.

Transcripts amplified for in situ hybridization.

Gene NCBI accession Forward primer Reverse primer
Obi SYT1 XM_014919324.1 5' CCT GGG ATG GAT ATG TCT GG ‐3' 5' GTT CGG AAG TCC CAA TAC GA‐3'
Obi CPLX1 XM_014919099.1 5' AAA ATG GCG GCA AGT GTA GG 3' 5' AAA GAG GTG AGG GGT TGT GC 3'
Obi MUNC18 XM_052972671.1 5' AGG ATC GGT CTC AGC TGT TG 3' 5' TTT CTT CGC CCT CCA TCT TGA 3'
Obi MNX XM_014912376.1 5' TGC ATT TGG TCA GGG TCA GC‐3' 5' TTT AGG CAA ACA GTT CGC TGG ‐3'
Obi LHX3 XM_052978610.1 5' GAA TTC GCC CTT CGA TTA CAA 3' 5' ACA GCT AAG CGT TGA AAC AG 3'
Obi NKX6 XM_014926771.2 5' GTC ATA GTT GAT AAA GAC GGC AAA C 3' 5' GTT CTT GTT TCT TTT TAG CTG ATG C 3'
Obi PIEZO XM_014926525.1 5' CTG GCC AGG CTT TGT CAA TC 3' 5' TCC ATT GGA TCT TCA TTA GTT G 3'
Obi DRGX XM_014929620.1 5' GGT TAT GGT GCA TGA TGG CG 3' 5' GCA CGT ACG GTC ATT TCT GC 3'
Obi FMRF XM_014929818.1 5' ACC AGT ATG CAA AGC CAA GC 3' 5' TCT CAT GTT TAT TGA GCA CGT CTG 3'
Obi VACHT XM_014912673.1 5' TGG ACA ACA TGC TTT ACA TGG ‐3' 5' ATG AAG CCC ATT TCC CAT TC ‐3'
Obi SLC5A7 XM_014929182.1 5' ATA TGG GCT GGT CGC AAG TC 3' 5' CTG GTT CCA GTC TGT GGA GG 3'
Obi VIAAT XM_014931765.1 5' CCG GTA AAA TAT TGG TGG AAT GTC TG ‐ 3' 5' AGT GTT GGT AGA AAT ATT TGT GAG G ‐ 3'
Open in a new tab

Sequences for the neurotransmitter genes GAD and CHAT were cloned by Dr. Rahul Parnaik using degenerate PCR. The sequence for the GAD cDNA matches 18–364 of XM_014922646. That for the CHAT matches 1085–2063 of XM_014912671.

PCR reaction products were ligated into a pGEM T‐Easy plasmid (Promega), and cloned inserts were Sanger sequenced by the University of Chicago Comprehensive Cancer Center DNA Sequencing Facility. To generate antisense templates, plasmids were linearized with SacII (New England Biolabs, Cat#: 50812058) or SpeI (New England Biolabs, Cat#: 50811989). Phenol–chloroform extraction was performed on the template. Antisense digoxigenin (DIG)‐labeled riboprobes (Sigma–Aldrich, Cat#: 11277073910) were transcribed with SP6 or T7 RNA polymerase (New England Biolabs). The residual template was digested at 37°C for 15 min with RNase‐free DNase I (Sigma–Aldrich). Riboprobes were ethanol‐precipitated and stored in 100 µL of DEPC‐H2O at −20°C until use.

2.4. In Situ Hybridization (ISH)

Slides of sectioned tissue were equilibrated to room temperature, postfixed in mailers in 4% PFA/PBS, and then washed in DEPC‐PBS. Following a 15‐min incubation at 37°C in proteinase K solution (Sigma–Aldrich; 1.45 µg proteinase K per milliliter of 100 mM Tris‐HCl [pH 8.0], 50 mM EDTA [pH 8.0]), slides were postfixed for 15 min in 4% PFA/PBS, rinsed in DEPC‐PBS, and acclimated to 72°C for 1 h in hybridization solution (50% formamide, 5× SSC, 1% SDS, 200 µg/mL heparin, 500 µg/mL yeast RNA). Slides were transferred to mailers with 1–2 mg antisense riboprobe in 15 mL of hybridization solution and incubated overnight at 72°C. The next day, slides were treated in preheated Solution X (50% formamide, 5× SSC, 1% SDS) at 72°C, washed in room temperature TBST (Tris‐buffered saline with 1% Tween 20), and blocked at room temperature for 1 h in 10% DIG buffer (Roche) in TBST. Meanwhile, anti‐DIG Fab fragments conjugated to alkaline phosphatase (Sigma–Aldrich, Cat#: 11093274910, RRID: AB_514497) were preabsorbed with octopus embryo powder in 1% DIG buffer in TBST for at least 1 h. Slides were incubated on a rocker overnight at 4°C in preabsorbed antibody diluted to a final concentration of 1:5000 in 10% DIG buffer in TBST.

The next day, slides were washed in TBST and moved into freshly prepared NTMT (100 mM Tris‐HCl [pH 9.5], 100 mM NaCl, 50 mM MgCl2, 1% Tween 20) for 10 min. To demonstrate alkaline phosphatase activity, the slides were incubated in NTMT containing nitro blue tetrazolium (NBT, 100 mg/mL in 70% dimethyl formamide/30% DEPC‐H20; Gold Biotechnology, St. Louis, MO) and 5‐bromo‐4‐chloro‐3‐indolyl phosphate (BCIP, 50 mg/mL in 100% dimethyl formamide; Gold Biotechnology). Color reaction was monitored for up to 5 days. After such time, slides were washed in TBST overnight, dehydrated through an ethanol series, cleared in Histoclear (National Diagnostics, Atlanta, GA), and coverslipped with Eukitt (Sigma–Aldrich).

ISH with no probe and a mouse probe were used as controls. No staining was observed in the no probe condition. The heterologous probe elicited labeling restricted to the edge of the sucker, as illustrated in Figure 1b,c,f.

FIGURE 1.

FIGURE 1

Open in a new tab

Geography of octopus arm and location of sucker ganglion. (a) Diagram of a transverse cross section of octopus arm. The major components of the arm nervous system are highlighted in blue. Key indicates transverse plane of section. Asterisk denotes acetabulobrachial muscles (ACBM). (b) Transverse arm section with in situ hybridization (ISH) for the neuronal gene complexin (CPLX1). Neuronal cell bodies in the axial nerve cord (ANC) and sucker ganglion (SG) are labeled. Scale bar: 100 µm. (c–f). Longitudinal sections through the octopus arm, as indicated by inset in c. In the longitudinal plane, the coordinate axes correspond to aboral–oral and proximal–distal, as indicated in f. (c) ISH of CPLX1 on a longitudinal section of octopus arm. There is an SG embedded in the ACBM for every sucker. Arrowheads point to the SG. Scale bar: 250 µm. (d, e) The SG is enriched with other pan‐neuronal markers. (d) MUNC18. (e) SYT1. Octopuses have many isoforms of SYT, so SYT1 may not label all neurons. Scale bars: 50 µm. (f) ISH of CPLX1 on a longitudinal section of octopus arm across many suckers. The location of the SG varies based on the location of the sucker. Scale bar: 500 µm. AC, acetabulum; ACBM, acetabulobrachial muscles; ANC, axial nerve cord; IMNC, intramuscular nerve cord; INF, infundibulum; SG, sucker ganglion.

2.5. Hematoxylin and Eosin (H&E) Staining

Slides were incubated in DEPC‐PBS for 4–6 h at 72°C to melt off gelatin. Then, slides were rehydrated in room temperature deionized water, incubated in Mayer's Hematoxylin (ScyTek, Cat#: HMM500) for 1 min, and rapidly rinsed in water to prevent overstaining. After a 15‐s incubation in a bluing solution (ScyTek, Cat#: BRT500), slices were dehydrated in an ethanol series (70%, 95%, and 100%) and placed in Eosin Y (ScyTek, Cat#: EYB500) for 1 min. Slides were rinsed in 100% EtOH, cleared in Histoclear, and mounted with Eukitt.

2.6. Picrosirius Red

The presence of connective tissue was examined with Picrosirius Red (Abcam, Cat#: ab150681). De‐gelatinized slides were rinsed three times with deionized water, then rehydrated for 1 min in deionized water. Sections were incubated in Picrosirius Red solution for 5 min, de‐stained with two quick rinses of acetic acid wash, dehydrated in 100% EtOH, and mounted with Eukitt.

2.7. Immunohistochemistry

Mouse monoclonal antibody 6–11B‐1 (1:500 dilution of ascites fluid; RRID: AB_477585, Sigma–Aldrich, Cat#: T6793) and mouse monoclonal antibody SMI‐31 (1:500 dilution; RRID: AB_2564641, BioLegend, Cat# 801601) were used to label neuronal processes in the octopus arm. Clone 6–11B‐1, which was isolated following immunization with sea urchin sperm flagella protein preparations, has been employed to label axon tracts in vertebrate and invertebrate nervous systems (Piperno and Fuller 1985; Chitnis and Kuwada 1990; Shigeno and Yamamoto 2002; Baratte and Bonnaud 2009). It recognizes an acetylated α‐tubulin (acTUBA) epitope found broadly but not universally across microtubules (Piperno and Fuller 1985; LeDizet and Piperno 1991). Clone SMI‐31 recognizes a phosphorylated epitope on neurofilament in mammals and neurofilament 220 in squid (Sternberger and Sternberger 1983; Grant et al. 1995; Grant and Pant 2016). Alexa Fluor 488 AffiniPure Goat Anti‐Mouse IgG (Jackson ImmunoResearch, Cat#: 115‐545‐003, RRID: AB_2338840) and Cy3 AffiniPure Donkey Anti‐Mouse IgG (Jackson ImmunoResearch, Cat#: 715‐165‐151, RRID: AB_2315777) were utilized as secondaries (1:500 dilutions).

For section immunohistochemistry, slides were washed in DEPC‐PBS containing 1% Tween 20 (PBST), incubated for 30 min in a proteinase K solution at 37°C (Sigma–Aldrich, Cat#: 03115828001; 19.4 µg proteinase K per milliliter of PBST), and then returned to room temperature with a postfix in 4% PFA/PBS for 15 min. Following washes in PBST, slides were blocked in 10% goat serum/PBST (Fisher Scientific, Cat#: 16210072) for 1 h and incubated in primary antibody diluted in 1% goat serum/PBST for 4 days at 4°C. Slides were washed in PBST for 3 × 30 min. Then, sections were incubated for 2 days at 4°C with secondary antibody along with 0.01 mg/mL 4′‐6‐diamidino‐2‐phenylindole (DAPI; Sigma–Aldrich, Cat#: D9542) to label cell nuclei fluorescently and either 0.2 µL/mL phalloidin‐iFluor 594 (Abcam, Cat#: ab176757) or phalloidin‐iFluor 488 (Abcam, Cat#: ab176753) to label filamentous actin (F‐actin). Slides were then washed with PBST and cover‐slipped with Fluoromount G (Southern Biotech, Birmingham, AL).

For whole‐mount immunohistochemistry, 0.5‐ to 1‐cm transverse slices and dissected sucker musculature were washed in PBST, dehydrated in a graded methanol series (25%, 50%, and 75% in PBST), and stored overnight at −20°C. Slices were rehydrated the next day in a graded methanol series (75%, 50%, and 25% in PBST), rinsed in PBST, incubated for 1 h at 37°C in a proteinase K solution (19.4 µg proteinase K per milliliter of PBST), and postfixed for 15 min in 4% PFA/PBS. Following washes in PBST, tissue was blocked for 1 h in 10% goat serum/PBST and incubated in primary antibody for 7 days at 37°C. Following a day of PBST washes, tissue was transferred to secondary antibody for 7 days at 4°C. Tissue was rinsed with PBST, washed with DEPC‐PBS, and postfixed for 4 days at 4°C. Blocks were rinsed in DEPC‐PBS and then cleared in a modified version of FRUIT with incubations in 35%, 40%, 60%, 80%, and 100% FRUIT, each for 24 h (Hou et al. 2015). Until imaging, slices were stored in 100% FRUIT at 4°C.

2.8. Imaging and Image Analysis

Lightfield and fluorescent‐labeled tissue was studied with a Zeiss Axioskop 50 upright microscope and a Leica MZ FLIII stereomicroscope, both outfitted with the Zeiss AxioCam digital camera and AxioVision 4.5 software system. Selected sections were also studied on a Leica SP5 Tandem Scanner Spectral 2‐photon confocal microscope (Leica Microsystems, Inc., Buffalo Grove, IL) or scanned by an Olympus VS200 Research Slide Scanner (Olympus/Evident, Center Valley, PA) with a Hamamatsu ORca‐Fusion camera (Hamamatsu Photonics, Skokie, IL). Whole mounts were imaged on a LaVision BioTec UltraMicroscope II (Miltenyi Biotec, Bergish Gladbach, Germany) run by ImSpector Pro v. 7_124 software (LaVision BioTec, Bielefeld, Germany). Collected images were corrected for contrast and brightness and false colored in FIJI (version 2.1.0/1.53c; National Institutes of Health [NIH]).

2.9. Proximal–Distal Analysis

Two series were created to examine sucker ganglia size changes along the proximal–distal axis of the arm, from the edge of crown webbing to tip. The first series consisted of three evenly spaced arm blocks from an arm on the right side (arm R1) of one animal sectioned longitudinally and stained with acTUBA. The second series consisted of three evenly spaced blocks from an arm on the left side (arm L3) of a second animal sectioned longitudinally and stained with H&E. The longitudinal width of each sucker ganglion along the proximal–distal axis was manually measured at the widest point using the line tool in FIJI (Figure 2a,b). The longitudinal widths of the sucker ganglion were then compared to the longitudinal widths of the corresponding suckers (SG width/sucker width). Sucker measurements are reported in Olson et al. (2025).

FIGURE 2.

FIGURE 2

Open in a new tab

Sucker ganglion width changes along the proximal–distal axis. (a, b) Example longitudinal sections through sucker ganglia (SG) used for measurements. Yellow line indicates location of measurement. Scale bars: 50 µm. (a) SG stained for acetylated α‐tubulin (acTUBA, cyan) and F‐actin (magenta). (b) SG stained with hematoxylin and eosin (H&E). (c) SG width decreases down the proximal (P)–distal (D) axis, error bars ± sem. (d) The width of the SG relative to sucker width increases along the P–D axis, error bars ± sem. acTUBA, acetylated α‐tubulin; H&E, hematoxylin and eosin; SG, sucker ganglion.

2.10. Nerve Analysis

Large‐scale nerve analysis was facilitated by segmenting two transversely imaged whole‐mount sucker ganglia, immunostained with acTUBA and SMI‐31, and two horizontally imaged whole‐mount sucker ganglia, immunostained with acTUBA. Nerve fibers connecting to the sucker ganglion were traced using Simple Neurite Tracer (SNT) in FIJI (Arshadi et al. 2021).

2.10.1. Nerve Fascicle Diameters

Using the path fitting feature in SNT, the radii of each node of the paths of the two horizontally imaged sucker ganglion were found (Arshadi et al. 2021). Then, using the SNT application programming interface (API) and a python script, the radii of the nodes corresponding to the first 300 µm of each path were extracted, smoothed, and averaged. The average radius for each path was multiplied by 2 to determine the diameter of the nerve fascicle. The distribution of nerve fascicle diameters was plotted as a histogram using Matplotlib (v3.7.3).

2.10.2. Nerve Root Distribution

The SNT API was used to find the 3D coordinates of the roots of the traces from the two transversely imaged sucker ganglia and from one horizontally imaged sucker ganglion (Arshadi et al. 2021). The coordinates of roots of the nerves were translated and rotated so that the center of all the roots was set as the origin. Subsequently, each root coordinate was normalized so that the vector from the origin to the tip had a magnitude of 1, creating a circle. The spatial distribution of roots from the three sucker ganglia was realized in the aboral–oral and external–internal plane with a colormap corresponding to the location of the root in that plane using Matplotlib. The spatial distribution of roots from the horizontally imaged sucker ganglion was also realized with a colormap corresponding to the diameter of the nerve fiber.

2.10.3. Nerve Targets

The 3D spatial coordinates, roots, and tips of the nerves from the horizontally imaged and transversely imaged sucker ganglia, both stained with acTUBA, were found using the SNT API (Arshadi et al. 2021). The average trajectory of each nerve was computed by creating a vector from the root of the nerve to the center of the process (Figure 8i). The unit vector for each average trajectory was found. The distribution of the average trajectories for the transversely imaged sucker ganglion was realized in the aboral–oral and external–internal planes with a colormap corresponding to nerve root locations using Matplotlib. The distribution of the average trajectories for the horizontally imaged sucker ganglion was realized in the proximal–distal and external–internal planes with a colormap corresponding to nerve target location in that plane.

FIGURE 8.

FIGURE 8

Open in a new tab

Distribution of sucker ganglion nerve fibers. (a) Maximum projection of a whole mount of a sucker stained for acetylated α‐tubulin (acTUBA, cyan). Many fascicles enter and exit the sucker ganglion (SG). Scale bar: 250 µm. (b) Distribution of fascicle diameters. Diameters range from 6 to 45 µm, with the peak in the 11–20 µm range. (c) Transverse section through the SG stained for acTUBA (cyan) and F‐actin (magenta). Location of nerve fiber exits is denoted with asterisk. Scale bar: 50 µm. (d) Distribution of fiber exits around the SG (n = 3 ganglia, 2 stained for acTUBA, 1 stained for SMI‐31) plotted in the transverse plane. There is a gap corresponding to the neuropil (NP) cap (***) and two gaps corresponding to a protrusion of the cell body region (**). Color key by location of root in the aboral–oral and external–internal planes. (e–g) The largest nerve fibers exit aborally. (e) Horizontal maximum projection of the aboral portion of the SG stained for acTUBA. Red arrow heads indicate large nerve bundles. (f) Horizontal maximum projection of the oral portion of the sucker ganglion stained with acTUBA. No large nerve bundles are present. Scale bars: 100 µm. (g) Distribution of fiber exits around the SG (n = 1 ganglion, stained for acTUBA) plotted in the transverse plane. Color key corresponds to size of fascicles. The largest fiber bundles exit aborally, corresponding to the portion between the NP cap (***) and cell body protrusion (**). (h) Example of SG nerves (blue) segmented out of a transverse whole mount slice stained for acTUBA. Red traces depict the axial nerve cord oral roots. (i) Schematic of the average trajectory, which is a vector created from the nerve exit point to the center of the processes. (j) Distribution of average trajectories in the transverse plane (n = 1 ganglion, stained for acTUBA). Color key corresponds to nerve root location in the aboral–oral, external–internal plane. Asterisks depict aboral gap (***) and lateral gaps (**) as in d and g. Nerve targets correspond to nerve exit. Key illustrates the transverse orientation. (k) Distribution of average trajectories in the horizontal plane (n = 1 ganglion, stained for acTUBA). Color key corresponds to nerve target location in the horizontal plane. There is a bias away from the proximal–distal axis. Key illustrates the horizontal orientation. acTUBA, acetylated α‐tubulin; CB, cell bodies; ExS, external side; InS, internal side; NP cap, neuropil cap; SG, sucker ganglion.

3. Results

We confirmed the presence and position of the sucker ganglion in O. bimaculoides with ISH for pan‐neuronal markers (Figure 1b–e). As in other species, the sucker ganglion is embedded in the acetabulobrachial muscles (ACBM) of the sucker stalk aboral, or deep, to the acetabulum of each sucker (Figure 1b,f). Because of its position within the sucker musculature, the sucker ganglion is a mobile structure. As reflected in Figure 1f, the sucker ganglion moves with the sucker as the sucker muscles contract.

Suckers decrease in width along the proximal–distal axis of the arm. The sucker ganglion likewise decreases in width, though it maintains its ellipsoid shape down the long axis of the arm (Figure 2c). We asked if this decrease is isometric with the decrease in sucker width. We found instead an allometric relationship: ganglia from the distal region of the arm were larger relative to the width of their respective sucker than ganglia found in the intermediate or proximal regions (Figure 2d). This difference could be due to a developmental constraint on the size of the circuit or underlying functional differences across suckers located at different arm positions.

3.1. Internal Composition of the Sucker Ganglion

The octopus sucker ganglion has an unusual structure (Graziadei 1971; Video S1). Instead of neuronal cell bodies surrounding a central NP (Richter et al. 2010), the sucker ganglion NP is predominantly arranged as a cap on the aboral surface (acTUBA labeling; Figure 3d). A dense cell body region is located just orally to this NP cap, that is, a step closer to the sucker (Figure 3e,f,h–k). This cell body region is composed of small cell bodies situated in a basket of larger cell bodies (Figure 3e,f). At the junction of the NP cap and cell body region is an inner belt (Figure 3d,e). This inner belt is composed of circumferentially running fiber bundles with large cell bodies lining its outer edge (Figure 3d,e,h–k). In addition, connective tissue identified with Picrosirius Red staining forms a distinct plate between the NP cap and cell body region, separating these two territories (Figure 4). On the oral side of the sucker ganglion, that is, closest to the sucker, there is a small pocket of NP and an additional belt of circumferential fibers (“outer belt”; Figure 3f,g). Unlike the inner belt, however, the outer belt appears devoid of neuronal cell bodies (Figure 3f,g).

FIGURE 3.

FIGURE 3

Open in a new tab

Internal neural organization of the sucker ganglion. (a–c) The axes for a sucker: proximal–distal, external–internal, and aboral–oral. (a) An en face, horizontal cartoon of the suckers showing that the suckers are arranged in two offset rows. In the horizontal plane, each sucker has a proximal (P)–distal (D) axis, and an internal side (InS)–external side (ExS) axis. The P–D axis corresponds to the P–D axis for the arm as whole (i.e., from head to arm tip). The InS corresponds to the side of the sucker that hugs the midline of the arm. The ExS corresponds to the side of the sucker that is situated on the outer edge of the arm. (b) A transverse cartoon of a sucker. In the transverse plane, each sucker has an aboral–oral axis and an ExS–InS axis. The aboral side is the side further from the sucker; the oral side is the side closer to the sucker. (c) A cartoon of a transverse section through the sucker ganglion (SG). Like the whole sucker, the SG has an aboral–oral axis and an ExS–InS axis in the transverse plane. (d–g) Organization of the SG demonstrated by acetylated α‐tubulin (acTUBA, cyan) immunostaining. Horizontal virtual slices produced by fusing multiple 1‐µm confocal sections from aboral (d) to oral (g). See Video S1 for a 3D rendering of this data set. (d) Most aborally is a neuropil (NP) cap and an internal belt (in.belt) that is composed of circumferential fibers and large cell bodies. (e) The in.belt is situated aboral to the cell body (CB) region. (f) Oral to the CB territory is an outer belt (out.belt), composed of circumferentially running fiber bundles. (g) On the oral side of the SG is a pocket of oral NP (oNP). Inset highlights the slight asymmetry of the horizontal sections through the SG. (h–k) Longitudinal sections from external (h) to internal (k) through the SG prepared for in situ hybridization with CPLX1 demonstrating the location of neuronal cell bodies. Neurons are located in the in.belt and are dense throughout the full extent of CB region. Scale bars: 50 µm. acTUBA, acetylated α‐tubulin; CB, cell bodies; D, distal; ExS, external side; in.belt, internal belt; InS, internal side; NP cap, neuropil cap; oNP, oral neuropil; out.belt, outer belt; P, proximal.

FIGURE 4.

FIGURE 4

Open in a new tab

Connective tissue splits the sucker ganglion. (a, b) Picrosirius Red staining of a longitudinal section through the sucker ganglion (SG). Red denotes the presence of connective tissue. (a) Brightfield image depicting the full extent of the SG. Black arrowheads point to connective tissue plate (CTP) between the cell body (CB) region and neuropil (NP) cap. (b) Polarized light image of the same field, highlighting the red labeling of connective tissue. White arrowheads point to CTP. Scale bars: 50 µm. CB, cell bodies; CTP, connective tissue plate; NP cap, neuropil cap.

3.2. Molecular Characterization of the Sucker Ganglion Cell Bodies

With ISH, we investigated the molecular composition of the cell body region of the sucker ganglion, first examining sensory and motor neuron markers (Jessell 2000; Coste et al. 2010; Nomaksteinsky et al. 2013). We found positive expression for two motor neuron makers, MNX and LHX3, and the mechanosensitive ion channel marker PIEZO (Figure 5a–c). PIEZO expression is found in scattered cells throughout the cell body territory, whereas MNX and LHX3 labeling lies at its lateral edges and in its center. PIEZO expression was also present in a subset of cells in the inner belt (Figure 5a). Interestingly, we failed to detect clear expression for DRGX, the dorsal root ganglia homeobox gene, and NKX6, a transcription factor marking motor neurons, despite strong labeling for both of these markers in the ANC and DRGX‐positive cells in the ACBM (Figure 5g–j; Olson et al. 2025). Hence, while the sucker ganglion contains expression for both sensory and motor markers, its molecular profile differs from that of the ANC.

FIGURE 5.

FIGURE 5

Open in a new tab

Gene expression patterns in the cell body territory of the sucker ganglion. (a–c) The sucker ganglion (SG) is enriched with sensory and motor neuron markers. (a) In situ hybridization (ISH) for the mechanosensory ion channel marker PIEZO. Inset demonstrates PIEZO labeling in the inner belt (in.belt). (b) ISH for LIM homeobox 3 transcription factor (LHX3), a motor neuron marker. (c) ISH for motor neuron and pancreas homeobox transcription factor (MNX). (d–f) Presysnaptic markers for cholinergic neurons label a subset of neurons in the cell body region. (d) ISH for choline acetyl transferase (CHAT). (e) ISH for vesicular acetylcholine transporter (VACHT). (f) ISH for solute carrier family 5 member 7 (SLC5A7). Scale bars: 50 µm. (g–j) The molecular profile of the SG differs from that of the ANC. No positive labeling was detected in the SG for (g) NK6 homeobox transcription factor (NXK6), a motor neuron marker, or (i) dorsal root ganglia homeobox transcription factor (DRGX), despite positive labeling in the ANC (h, NKX6) or ACBM (j, DRGX). Scale bars for g–i: 50 µm. Scale bar for j: 20 µm. (k–n) Markers of inhibitory neurotransmission, (k) glutamate acid decarboxylase (GAD) and (m) vesicular inhibitory amino acid transporter (VIAAT), do not show detectable labeling in the SG, although the ANC (l, GAD) and ACBM (n, VIAAT) demonstrate cellular labeling readily. Scale bars for k–m: 50 µm. Scale bar for n: 20 µm. All sections longitudinal. ACBM, acetabulobrachial muscles; ANC, axial nerve cord; CB, cell bodies; in.belt, inner belt; NP cap, neuropil cap.

Graziadei (1965) reported the existence of muscle receptors in the ACBM around the sucker ganglion. With acTUBA immunostaining, we identified many cells embedded in the ACBM that issue processes to nerve fascicles connected to the sucker ganglion (Figure 6). Some of these cells are multipolar, with many processes extending into the local musculature (Figure 6b,c). Some cells cluster together in groups (Figure 6d). Others are located close to the sucker ganglion itself (Figure 6e). We also found that the ACBM are enriched with expression of PIEZO and DRGX, indicating that these cells may be sensory in nature (Figure 5j; PIEZO not shown). This suggests the sucker ganglion handles sensory input from cells within the ganglion itself as well as from the local musculature.

FIGURE 6.

FIGURE 6

Open in a new tab

Cells in the local musculature feed back to the sucker ganglion. (a) Maximum projection of the sucker ganglion (SG) and surrounding musculature in the horizontal plane stained for acetylated α‐tubulin (acTUBA, cyan). Arrowheads point to a cell body in the musculature feeding back to SG. Scale bar: 100 µm. (b–e) Close‐up images of cells in musculature stained with acTUBA. (b, c) Cells exhibit a complex morphology. Yellow arrowheads point to processes that extend into the musculature; white arrowheads indicate cell bodies. Arrows point in the direction of the SG. Scale bars: 25 µm. (d) Some cells are located in clusters. White arrowheads indicate cell bodies. Arrow points in the direction of SG. Scale bar: 50 µm. (e) Some cells are located close to the SG. White arrowheads indicate cell bodies. Arrow points in the direction of the SG. Scale bar: 25 µm. ACBM, acetabulobrachial muscles; acTUBA, acetylated α‐tubulin; in.belt, inner belt; SG, sucker ganglion.

We interrogated the sucker ganglion with ISH for other neural subtype markers. The sucker ganglion is enriched for expression of presynaptic cholinergic neuron markers (VACHT/SLC18A3, SLC5A7, and CHAT). Unlike in the ANC, where we found expression for these cholinergic neuron markers throughout the entirety of the CBL, VACHT, SLC5A7, and CHAT expression in the sucker ganglion is found at the lateral edges of the ganglion and the center, similar to the labeling pattern of the motor neuron markers (Figure 5d–f). Glutamate decarboxylase (GAD) and vesicular inhibitory amino acid transporter (VIAAT) serve as markers of fast GABAergic inhibitory neurons. We failed to detect expression for either of these inhibitory neuron markers in the sucker ganglion in experiments presenting positive labeling in the ANC or the ACBM (Figure 5k–n). Rapid ganglionic inhibition therefore would need to be mediated by another neurotransmitter, such as acetylcholine, or by extra‐ganglionic input from the ANC or from VIAAT‐positive cells seen scattered within the sucker musculature (Figure 5n).

The expression pattern of FMRF, which encodes a neuromodulatory neuropeptide, was striking in the sucker ganglion. In the ANC, it is broadly and abundantly expressed (Figure 7a; see also Winters‐Bostwick et al. 2024). In the sucker ganglion, however, it is very restricted (Figure 7b). Moreover, the distribution of labeling is asymmetric within the ganglion, with labelling found only on the external side, or the side that corresponds to the outer edge of the sucker (Figure 7c; see Figure 3a). This asymmetry in molecular identity could underlie an asymmetry in function.

FIGURE 7.

FIGURE 7

Open in a new tab

Asymmetric gene expression of FMRF in the sucker ganglion. (a) Transverse section of the axial nerve cord with in situ hybridization (ISH) for FMRF. FMRF expression is found abundantly throughout the cell body layer. Inset indicates the location of the panel image. (b) Longitudinal section through the sucker ganglion (SG) with ISH for FMRF. Arrowhead point to FMRF labeling of a small subset of cells. (c) Horizontal section through the SG with FMRF ISH. FMRF labeling is restricted to the external side of the SG, as indicated by arrowheads. ANC, axial nerve cord; CB, cell bodies; CBL, cell body layer; ExS, external side of sucker; InS, internal side of sucker; in.belt, inner belt; NP, neuropil; NP cap, neuropil cap. Scale bars: 50 µm.

3.3. Nerve Fiber Organization

Many fiber fascicles issue from the sucker ganglion (Figure 8a). These bundles vary in diameter from 6 to 45 µm, with the bulk falling in the 11–20 µm range (Figure 8b). We examined the location of the fascicles around the perimeter of the sucker ganglion. We found that the roots are primarily distributed along the lateral sides and oral underside, forming a U‐shape (Figure 8c,d). There are notable gaps in this pattern, with a prominent gap corresponding to the NP cap, and two bilaterally symmetric gaps corresponding to a bulge in the cell body territory (Figure 8c,d). The largest nerve bundles depart from the aboral portion of the ganglion (Figure 8e,f). This territory falls in between the aboral and lateral gaps (Figure 8g). These results suggest a regionalization along the aboral–oral axis within the ganglion based on root location.

We also examined how the trajectories of the nerve fibers vary according to their origin or root location (Figure 8h,i). We found the distribution of trajectories follows the distribution of the roots (Figure 8d,j). The nerves issuing from the aboral territory between the aboral and lateral gaps extend aborally in the direction of the aboral ACBM and the oral roots leading to the ANC (Figure 8h,j). Those issuing from the oral territory extend laterally and orally in the direction of the lateral ACBM, the sucker's acetabulum, and the ANC oral roots targeting the sensory epithelium of the sucker (Figure 8h,j). These results establish a spatial regionalization to nerve fiber trajectories along the aboral–oral axis of the sucker ganglion.

In the ANC, there is an asymmetry in nerve fibers targeting the sucker: those issued from the external side of the ANC cover more territory of the sucker compared with those issued from the internal side (Olson et al. 2025). We asked if such an asymmetry in nerve fiber distribution is present in the sucker ganglion by examining the average trajectories along the external–internal axis (see Figure 3a). We did not detect a similar kind of asymmetry in the sucker ganglion nerve fibers, with the external and internal sides similarly covered (Figure 8j,k). We found instead a bias away from the proximal–distal axis and toward the external–internal axis, proposing a variation in function between the proximal–distal and external–internal planes (Figure 8k). These results also demonstrate a divergence between sucker ganglion and ANC nerve fiber distributions, signifying that the sucker ganglion has processing roles distinct from those of the ANC.

4. Discussion

The sucker ganglia are an exceptional feature of the octopus nervous system. First, across the eight arms of an adult O. bimaculoides, there are estimated to be over 1000 ganglia. Second, as noted by Graziadei (1971) and confirmed in the results of this study, they have an inverted structure. Third, as we document here, they have a rich internal complexity, an aboral inner belt with neuronal cell bodies, an oral outer belt devoid of neurons, and a connective tissue plate subdividing the ganglion as a whole (Figure 9a).

FIGURE 9.

FIGURE 9

Open in a new tab

Summary of sucker ganglion internal complexity and nerve fiber connections. (a) Cartoon depicting internal organization of the sucker ganglion (SG). The inner belt (red; in.belt), outer belt (orange; out.belt), and nerves (blue) connecting to the SG are highlighted. (b) Schematic depicting nerve connections. Oral nerves from the axial nerve cord are green. Nerves from the SG that target the oral nerves are pink. Nerves from the SG that target local musculature are blue. AC, acetabulum; ACBM, acetabular brachial muscles; CB, cell bodies; CTP, connective tissue plate; in.belt, inner belt; INF, infundibulum; NP cap, neuropil cap; oNP, oral neuropil; out.belt, outer belt; SG, sucker ganglion.

Our molecular investigations showed positive expression for both motor and sensory neuron markers, indicating a dual sensory and motor role for the sucker ganglion. The molecular profile of the sucker ganglion, however, is less complex than that of the ANC, consistent with a divergence in function between these two structures. We failed to detect positive expression of DRGX within the ganglion, although both the CBL of the ANC and the ACBM are enriched with DRGX‐expressing cells (Olson et al. 2025). Vertebrate mammalian studies suggest that DRGX is expressed in nociceptive sensory cells (Chen et al. 2001; Rebelo et al. 2006). Primary nociceptive cell bodies for the sucker, therefore, may be situated within the ANC or the sucker musculature but not in the sucker ganglion. Our positive detection of PIEZO within the sucker ganglion and ACBM supports Graziadei's (1965) observation of muscle receptors. This molecular evidence suggests that the sucker ganglion carries proprioceptive information for the sucker.

The presence of both sensory and motor cells suggests that the sucker ganglion is well suited to modulate local reflexes for the sucker without input from the ANC. Indeed, the sucker ganglion extends nerve fibers to the local musculature, and cells within these muscles feed back to the sucker ganglion. Classic literature has also suggested the presence of sensory–motor cells extending from the sucker ganglion: a single cell body with one sensory projection that might detect muscle stretch and one motor projection that could activate muscle (Martoja and May 1955). Future work is needed to investigate this interesting possibility.

Our data strongly imply functional differences in sensory–motor processing across the three cardinal axes for a sucker: the aboral–oral, external–internal, and proximal–distal axes (see Figure 3). The sucker ganglion sends nerve fibers in both the oral and aboral directions toward the oral roots issued from the ANC to the sucker epithelium. The connections orally suggest that the sucker ganglion may also exchange information with the sensory epithelium. The connections aborally, combined with the presence of large nerve bundles, suggest that the sucker ganglion sends information to and receives information from the ANC. Hence, the sucker ganglion is situated for both local reflexes and relays to the ANC (Figure 9b).

There are also biases of nerve fibers in the external–internal and proximal–distal axes, with a prejudice toward the external–internal axis and away from the proximal–distal axis. Along with asymmetric FMRF gene expression in the external–internal axis, this could underlie preferential sensorimotor activity along the external–internal axis. Further studies of the distribution of specific chemoreceptor subtypes around the sucker's sensory epithelium could elucidate any biases in sensory transduction, and an examination of individual sucker movements along the cardinal axes could uncover biases in motor action. Interestingly, many behaviors utilizing multiple suckers, such as passing food along the arm or grasping an object, require the coordination of the movements of individual suckers along the proximal–distal axis (Wells 1964; Altman 1968; Sivitilli et al. 2022). We did not detect any direct connections between adjoining sucker ganglia, so the ANC most certainly handles the connections needed for concerted motor action between suckers (Rowell 1963; Graziadei 1971; Olson et al. 2025). The ANC may also rely on its own separate circuits for control of coordinated motor actions between suckers or instead coordinate the activities of the sucker ganglia. Functional studies are needed to assess the relative roles of the ANC and the sucker ganglion for motor control of the suckers.

Structure and function of peripheral ganglia vary greatly across phyla (Tauc 1967). Vertebrate sympathetic autonomic ganglia, for example, are traditionally thought of as relay stations for faithful transmission of central nervous system signals to effectors (Tauc 1967; Nauta and Feirtag 1986; Karemaker 2017). Our data suggest that the sucker ganglia do not have an analogous function as either a sensory or motor relay, but rather exhibit substantial internal integrative circuitry. A comparison to other peripheral ganglia in cephalopods is also informative. Cephalopods have a pair of stellate ganglia located in the mantle, most famously known for the giant fiber system in squid (Young 1936, 1972, 1936, 1972; Hodgkin and Huxley 1939; Lund 1971; Wang and Ragsdale 2019). These ganglia are implicated in control of mantle contractions for respiration and escape as well as adaptive coloration both through direct relays from the brain and local ganglionic processing (Young 1938; Wilson 1960; Lund 1971; Bühler et al. 1975; Gonzalez‐Bellido et al. 2014, Gonzalez‐Bellido et al. 2018). Structurally, the stellate ganglia have a clear regionalization along the dorsal–ventral axis in the size of the cells and the structure of the NP (Lund 1971; Young 1972). This is reminiscent of the regionalization we report in the sucker ganglion along the aboral–oral axis. However, unlike the sucker ganglion, the stellate ganglia have a typical ganglionic structure of cell bodies wrapping NP in its center (Lund 1971; Young 1972). The stellate ganglia also do not have an equivalent outer or inner belt as the sucker ganglia do. Thus, even in the context of other cephalopod ganglia, the structure of the octopus sucker ganglion is strikingly unique.

5. Conclusion

The sucker ganglion is embedded in the sucker musculature, strategically situated between the ANC and the acetabulum of the sucker. Previous anatomical studies have proposed, in turn, a sensory, a motor, or a combined sensory and motor role for the ganglion (Guérin 1908; Martoja and May 1955; Rowell 1963; Graziadei 1965; Graziadei and Gagne 1976). Our molecular analysis does point to the sucker ganglion having populations of both sensory and motor neurons. Having intermixed sensory and motor neurons does not, however, guarantee an integrative role for the ganglion. It is our cellular analysis demonstrating a complex internal organization and a suite of fiber projections independent of those of the ANC to the sucker apparatus (Figure 9) that lead us to conclude that the sucker ganglion has specialized and unique integrative roles in sucker sensory–motor neuronal processing.

Conflicts of Interest

The authors declare no conflicts of interest.

Peer Review

The peer review history for this article is available at https://publons.com/publon/10.1002/cne.70055.

Supporting information

Supplemental Video 1: 3D‐rendering of the sucker ganglion. Video depicts 3D view of the sucker ganglion stained with acetylated α‐ mtubulin (acTUBA) and presented in Figure 3. Note that in 3D, the inner and outer belts can clearly be seen to be spatially segregated.

Download video file (4.9MB, mp4)

Acknowledgments

We thank Dr. Chuck Winkler of Aquatic Research Consultants for providing us with octopuses, and Dr. Caroline Albertin, Dr. Rahul Parnaik, and Ms. Natalie Grace Schulz for cDNAs. We thank Ms. Schulz for the Picrosirius Red staining protocol and for comments. Imaging was performed at the University of Chicago Integrated Light Microscopy Core (RRID: SCR_019197). We extend special thanks to Dr. Christine Labno and Mr. Khalil Rodriguez for their invaluable assistance. We thank Ms. Caroline Miller for assistance with whole mount image processing. This work was supported by the NIH UF1NS115817 award (C.W.R.).

Funding: This study was funded by National Institutes of Health (Grant number UF1 NS115817).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Altman, J. 1968. “The Nervous Control of Arm and Buccal Movements in Octopus vulgaris.” Doctoral thesis, University College London.
  2. Altman, J. 1971. “Control of Accept and Reject Reflexes in the Octopus.” Nature 229: 204–206. [DOI] [PubMed] [Google Scholar]
  3. Arshadi, C. , Günther U., Eddison M., Harrington K. I. S., and Ferreira T. A. 2021. “SNT: A Unifying Toolbox for Quantification of Neuronal Anatomy.” Nature Methods 18: 374–377. [DOI] [PubMed] [Google Scholar]
  4. Baratte, S. , and Bonnaud L. 2009. “Evidence of Early Nervous Differentiation and Early Catecholaminergic Sensory System During Sepia officinalis Embryogenesis.” Journal of Comparative Neurology 517: 539–549. [DOI] [PubMed] [Google Scholar]
  5. Bidel, F. , Bennett N. C., and Wardill T. J. 2022. “ Octopus Bimaculoides' Arm Recruitment and Use During Visually Evoked Prey Capture.” Current Biology 32: 4727–4733.e3. [DOI] [PubMed] [Google Scholar]
  6. Bühler, A. , Froesch D., Mangold K., and Marthy H.‐J. 1975. “On the Motor Projection of the Stellate Ganglion in Octopus vulgaris .” Brain Research 88: 69–72. [DOI] [PubMed] [Google Scholar]
  7. Buresch, K. C. , Sklar K., Chen J. Y., et al. 2022. “Contact Chemoreception in Multi‐Modal Sensing of Prey by Octopus.” Journal of Comparative Physiology A 208: 435–442. [DOI] [PubMed] [Google Scholar]
  8. Chen, Z.‐F. , Rebelo S., White F., et al. 2001. “The Paired Homeodomain Protein DRG11 Is Required for the Projection of Cutaneous Sensory Afferent Fibers to the Dorsal Spinal Cord.” Neuron 31: 59–73. [DOI] [PubMed] [Google Scholar]
  9. Chitnis, A. , and Kuwada J. 1990. “Axonogenesis in the Brain of Zebrafish Embryos.” Journal of Neuroscience 10: 1892–1905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Coste, B. , Mathur J., Schmidt M., et al. 2010. “Piezo1 and Piezo2 Are Essential Components of Distinct Mechanically Activated Cation Channels.” Science 330: 55–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fiorito, G. , Affuso A., Basil J., et al. 2015. “Guidelines for the Care and Welfare of Cephalopods in Research – A Consensus Based on an Initiative by CephRes, FELASA and the Boyd Group.” Laboratory Animals 49: 1–90. [DOI] [PubMed] [Google Scholar]
  12. Gonzalez‐Bellido, P. T. , Scaros A. T., Hanlon R. T., and Wardill T. J. 2018. “Neural Control of Dynamic 3‐Dimensional Skin Papillae for Cuttlefish Camouflage.” iScience 1: 24–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gonzalez‐Bellido, P. T. , Wardill T. J., Buresch K. C., Ulmer K. M., and Hanlon R. T. 2014. “Expression of Squid Iridescence Depends on Environmental Luminance and Peripheral Ganglion Control.” Journal of Experimental Biology 217: 850–858. [DOI] [PubMed] [Google Scholar]
  14. Grant, P. , and Pant H. C.. 2016. “Compartment‐Specific Phosphorylation of Squid Neurofilaments.” Methods in Enzymology 568: 615–633. https://linkinghub.elsevier.com/retrieve/pii/S0076687915005510. [DOI] [PubMed] [Google Scholar]
  15. Grant, P. , Tseng D., Gould R. M., Gainer H., and Pant H. C. 1995. “Expression of Neurofilament Proteins During Development of the Nervous System in the Squid Loligo pealei .” Journal of Comparative Neurology 356: 311–326. [DOI] [PubMed] [Google Scholar]
  16. Grasso, F. W 2008. “Octopus Sucker‐Arm Coordination in Grasping and Manipulation.” American Malacological Bulletin 24: 13–23. [Google Scholar]
  17. Graziadei, P 1965. “Muscle Receptors in Cephalopods.” Proceedings of the Royal Society B: Biological Sciences 161: 392–402. [Google Scholar]
  18. Graziadei, P 1971. “The Nervous System of the Arms.” In The Anatomy of the Nervous System of Octopus vulgaris, edited by Young J. Z., 45–61. Clarendon Press. [Google Scholar]
  19. Graziadei, P. P. C. , and Gagne H. T. 1976. “Sensory Innervation in the Rim of the Octopus Sucker.” Journal of Morphology 150: 639–679. [DOI] [PubMed] [Google Scholar]
  20. Guérin, M. J 1908. “Contribution a L'étude Des Systèmes Cutané, Musculaire et Nerveux De L'appareil Tentaculaire Des Céphalopodes.” Archives De Zoologie Expérimentale Et Générale 8: 1–179. [Google Scholar]
  21. Gutnick, T. , Zullo L., Hochner B., and Kuba M. J. 2020. “Use of Peripheral Sensory Information for Central Nervous Control of Arm Movement by Octopus vulgaris .” Current Biology 30: 4322.e3–4327.e3. [DOI] [PubMed] [Google Scholar]
  22. Hanlon, R. T. , and Messenger J. B. 2018. Cephalopod Behavior. 2nd ed. Cambridge University Press. [Google Scholar]
  23. Hochner, B. , Zullo L., Shomrat T., Levy G., and Nesher N. 2023. “Embodied Mechanisms of Motor Control in the Octopus.” Current Biology 33: R1119–R1125. [DOI] [PubMed] [Google Scholar]
  24. Hodgkin, A. L. , and Huxley A. F. 1939. “Action Potentials Recorded From Inside a Nerve Fibre.” Nature 144: 710–711. [Google Scholar]
  25. Hou, B. , Zhang D., Zhao S., et al. 2015. “Scalable and DiI‐Compatible Optical Clearance of the Mammalian Brain.” Frontiers in Neuroanatomy 9: 19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Jessell, T. M 2000. “Neuronal Specification in the Spinal Cord: Inductive Signals and Transcriptional Codes.” Nature Reviews Genetics 1: 20–29. [DOI] [PubMed] [Google Scholar]
  27. Kang, G. , Allard C. A. H., Valencia‐Montoya W. A., et al. 2023. “Sensory Specializations Drive Octopus and Squid Behaviour.” Nature 616: 378–383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Karemaker, J. M 2017. “An Introduction Into Autonomic Nervous Function.” Physiological Measurement 38: R89–R118. [DOI] [PubMed] [Google Scholar]
  29. Kennedy, E. B. L. , Buresch K. C., Boinapally P., and Hanlon R. T. 2020. “Octopus Arms Exhibit Exceptional Flexibility.” Scientific Reports 10: 20872. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kier, W. M. , and Smith A. M. 2002. “The Structure and Adhesive Mechanism of Octopus Suckers.” Integrative and Comparative Biology 42: 1146–1153. [DOI] [PubMed] [Google Scholar]
  31. LeDizet, M. , and Piperno G. 1991. “Detection of Acetylated α‐Tubulin by Specific antibodies.” Methods in Enzymology 196: 264–274. https://linkinghub.elsevier.com/retrieve/pii/007668799196025M. [DOI] [PubMed] [Google Scholar]
  32. Lopes, V. M. , Sampaio E., Roumbedakis K., et al. 2017. “Cephalopod Biology and Care, a COST FA1301 (CephsInAction) Training School: Anaesthesia and Scientific Procedures.” Invertebrate Neuroscience 17, no. 3: 8. [DOI] [PubMed] [Google Scholar]
  33. Lund, R. D 1971. “Stellate Ganglion.” In The Anatomy of the Nervous System of Octopus vulgaris, edited by Young J. Z., 621–640. Clarendon Press. [Google Scholar]
  34. Martoja, R. , and May R. M. 1955. “Les Cellules Neuro‐sensorieeless et le Ganglion Sous‐acétabulaire dans L'appereil Brachial Du Céphalopode Octopus vulgaris .” Comptes Rendus Hedbomadaires Des Seances De L'academie Des Sciences 240: 2440–2442. [PubMed] [Google Scholar]
  35. Nauta, W. J. H. , and Feirtag M. 1986. Fundamental Neuroanatomy. W. H. Freeman and Company. [Google Scholar]
  36. Neacsu, D. , and Crook R. J. 2024. “Repeating Ultrastructural Motifs Provide Insight Into the Organization of the Octopus Arm Nervous System.” Current Biology 34: 4767.e2–4773.e2. [DOI] [PubMed] [Google Scholar]
  37. Nixon, M. , and Young J. Z. 2003. The Brains and Lives of Cephalopods. Oxford University Press. [Google Scholar]
  38. Nödl, M.‐T. , Kerbl A., Walzl M. G., Müller G. B., and De Couet H. G. 2016. “The Cephalopod Arm Crown: Appendage Formation and Differentiation in the Hawaiian Bobtail Squid Euprymna Scolopes.” Frontiers in Zoology 13: 44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Nomaksteinsky, M. , Kassabov S., Chettouh Z., et al. 2013. “Ancient Origin of Somatic and Visceral Neurons.” BMC Biology 11: 53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Olson, C. S. , and Ragsdale C. W. 2023. “Toward an Understanding of Octopus Arm Motor Control.” Integrative And Comparative Biology 63, no. 6: 1277–1284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Olson, C. S. , Schulz N. G., and Ragsdale C. W. 2025. “Neuronal Segmentation in Cephalopod Arms.” Nature Communications 16: 443. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Packard, A 1988. “The Skin of Cephalopods (Coleoids): General and Special Adaptations.” In Form and Function, edited by Trueman E. R. and Clarke M. R., 37–67. Elsevier. https://linkinghub.elsevier.com/retrieve/pii/B9780127514116500102. [Google Scholar]
  43. Piperno, G. , and Fuller M. T. 1985. “Monoclonal Antibodies Specific for an Acetylated Form of a‐Tubulin Recognize the Antigen in Cilia and Flagella From a Variety of Organisms.” The Journal of Cell Biology 101: 2085–2094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Rebelo, S. , Chen Z.‐F., Anderson D. J., and Lima D. 2006. “Involvement of DRG11 in the Development of the Primary Afferent Nociceptive System.” Molecular and Cellular Neuroscience 33: 236–246. [DOI] [PubMed] [Google Scholar]
  45. Richter, S. , Loesel R., Purschke G., et al. 2010. “Invertebrate Neurophylogeny: Suggested Terms and Definitions for a Neuroanatomical Glossary.” Frontiers in Zoology 7: 29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Röckner, J. L. , Arellano M. D., and Zullo L. 2023. “Sucker Attachment and Detachment Patterns in Octopus vulgaris .” In Biomimetic and Biohybrid Systems. Living Machines 2023. Lecture Notes in Computer Science. Vol. 14157, edited by Meder F., Hunt A., Margheri L., Mura A., and Mazzolai B., 266–280. Springer. https://link.springer.com/10.1007/978‐3‐031‐38857‐6_20. [Google Scholar]
  47. Rossi, F. , and Graziadei P. 1954. “Nouvelles Recherches sur Le Systeme Nerveux Du Bras Des Cephalopodes Avec Des Methodes Specifiques pour Le Tissu Nerveux.” Acta Anatomica 22: 202–215. [PubMed] [Google Scholar]
  48. Rowell, C. H 1963. “Excitatory and Inhibitory Pathways in the Arm of Octopus .” Journal of Experimental Biology 40: 257–270. [Google Scholar]
  49. Shigeno, S. , and Yamamoto M. 2002. “Organization of the Nervous System in the Pygmy Cuttlefish, Idiosepius paradoxus ortmann (Idiosepiidae, Cephalopoda).” Journal of Morphology 254: 65–80. [DOI] [PubMed] [Google Scholar]
  50. Sivitilli, D. M. , Smith J. R., and Gire D. H.. 2022. “Lessons for Robotics From the Control Architecture of the Octopus.” Frontiers in Robotics and AI 9: 862391. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Sternberger, N. H. , and Sternberger L. A.. 1983. “Neurotypy: The Heterogeneity of Brain Proteins.” Annals of the New York Academy of Sciences 420: 90–99. [DOI] [PubMed] [Google Scholar]
  52. Tauc, L 1967. “Transmission in Invertebrate and Vertebrate Ganglia.” Physiological Reviews 47: 521–593. [DOI] [PubMed] [Google Scholar]
  53. van Giesen, L. , Kilian P. B., Allard C. A. H., and Bellono N. W. 2020. “Molecular Basis of Chemotactile Sensation in Octopus.” Cell 183: 594.e14–604.e14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wang, Z. Y. , and Ragsdale C. W. 2019. “Cephalopod Nervous System Organization.” In Oxford Research Encyclopedia of Neuroscience. Oxford University Press. https://oxfordre.com/neuroscience/view/10.1093/acrefore/9780190264086.001.0001/acrefore‐9780190264086‐e‐181. [Google Scholar]
  55. Wells, M. J 1964. “Tactile Discrimination of Surface Curvature and Shape by the Octopus.” Journal of Experimental Biology 41: 433–445. [Google Scholar]
  56. Wilson, D. M 1960. “Nervous Control of Movement in Cephalopods.” Journal of Experimental Biology 37: 57–72. [Google Scholar]
  57. Winters‐Bostwick, G. C. , Giancola‐Detmering S. E., Bostwick C. J., and Crook R. J.. 2024. “Three‐Dimensional Molecular Atlas Highlights Spatial and Neurochemical Complexity in the Axial Nerve Cord of Octopus Arms.” Current Biology 34: 4756–4766.e6. [DOI] [PubMed] [Google Scholar]
  58. Young, J. Z 1936. “The Giant Nerve Fibres and Epistellar Body of Cephalopods.” Journal of Cell Science S2‐78, no. 311: 367–386. [Google Scholar]
  59. Young, J. Z 1938. “The Functioning of the Giant Nerve Fibres of the Squid.” Journal of Experimental Biology 15: 170–185. [DOI] [PubMed] [Google Scholar]
  60. Young, J. Z 1963. “The Number and Sizes of Nerve Cells in Octopus .” Proceedings of the Zoological Society of London 140: 229–254. [Google Scholar]
  61. Young, J. Z 1971. The Anatomy of the Nervous System of Octopus vulgaris. Clarendon Press. [Google Scholar]
  62. Young, J. Z 1972. “The Organization of a Cephalopod Ganglion.” Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 263: 409–429. [DOI] [PubMed] [Google Scholar]
  63. Zullo, L. , Eichenstein H., Maiole F., and Hochner B. 2019. “Motor Control Pathways in the Nervous System of Octopus vulgaris Arm.” Journal of Comparative Physiology A 205: 271–279. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Video 1: 3D‐rendering of the sucker ganglion. Video depicts 3D view of the sucker ganglion stained with acetylated α‐ mtubulin (acTUBA) and presented in Figure 3. Note that in 3D, the inner and outer belts can clearly be seen to be spatially segregated.

Download video file (4.9MB, mp4)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from The Journal of Comparative Neurology are provided here courtesy of Wiley

RESOURCES