Sucker with a fat lip: The soft tissues underlying the viscoelastic grip of remora adhesion

Abstract

Remoras are fishes that attach to a broad range of hosts using an adhesive disc on their head that is derived from dorsal fin elements. Research on the adhesive mechanism of remoras has focused primarily on the skeletal components of the disc and their contribution to generating suction and friction. However, the soft tissues of the disc, such as the soft lip surrounding the bony disc and the muscles that control the bony lamellae, have been largely ignored. To understand the sealing mechanism of the disc, it is imperative to understand the tissue morphology and material properties of the soft lip. Here, we show that the soft lip surrounding the remora disc is comprised of discrete multilayered collagen, fat, and elastic tissues which we hypothesize to have specific roles in the viscoelastic sealing mechanism of the remora disc. The central, heavily vascularized fat and collagen layer are infiltrated by strands of elastic tissue and surrounded by crossed‐fiber collagen. A newly described jubilee muscle underneath the adhesive disc provides a mechanism for stopping venous return from the disc lip, thereby allowing it to become engorged and create a pressurized fit to the attachment substrate. Thus, the remora lip acts as a vascular hydrostat.

The soft tissue around the remora acts as a vascular hydrostat to create a viscoelastic seal. Blood flow is modulated by the jubilee muscle, which closes off drainage to the anterior cardinal sinus network.

1. INTRODUCTION

Remoras (Echeneidae) are fishes with a cosmopolitan distribution that adhere to a broad range of hosts with a wide variation in surface roughness and compliance, and maintain attachment under extreme depth, pressure, velocity, and drag conditions. Their remarkable attachment ability is a combination of friction and suction forces generated by their cranial adhesive disc, which evolved from dorsal fin spine elements (Britz and Johnson, 2012). Rotation of skeletal elements of the disc, specifically the pectinated lamellae, results in interaction between hundreds of tiny spinules and local asperities of the host surface to produce frictional forces that are calculated to be approximately 10 times greater than the shear forces experienced while attached to a swimming host (Fulcher and Motta, 2006; Beckert et al., 2015). The rotation of the lamellae while the outer fleshy lip of the disc is pressed against the host surface creates a relatively negative pressure space beneath the disc compared with ambient pressure, thus generating suction. Because the frictional forces of adhesion are an order of magnitude greater than the suction forces generated, if the remora disc were to fail under passive control conditions, it is hypothesized that this would be from pressure disequilibrium driving fluid seep through small crevices between the disc and host surface that would create a loss in suction and not as a result of drag (Beckert et al., 2016).

Most studies of the functional morphology of the remora adhesive disc have focused on the skeletal components of the disc and their contribution to friction and suction generation (Fulcher and Motta, 2006; Beckert et al., 2015). Histological sections of the disc lip were illustrated in the monograph by Houy (1910), but tissue types were not identified in the figures. Bargmann (1973) extended on Houy’s anatomical descriptions using a variety of staining techniques to identify collagen, fat, and elastic tissues within the remora disc lip. The disc lip is theorized to act as a viscoelastic seal (Beckert et al., 2016) but it is unknown how the tissues of the lip contribute to adhesive performance.

The structural and material properties of soft tissues are crucial to our understanding of biomechanical systems. Muscles, collagen, fat, and elastic tissues are continuously engaged during motion, each with different material properties and mechanical capabilities. To better answer how the remora creates an effective seal, it is important to know the three‐dimensional geometry and orientation of soft tissues supporting the adhesive disc. Herein, we describe the soft tissue anatomy of the remora disc lip in a biomechanical context to understand the relationship between material properties and functional performance.

2. METHODS

2.1. Sample collection

Remoras (Echeneis naucrates; total length = 22.8–29.7 cm) were maintained at the New Jersey Institute of Technology Department of Biological Sciences aquatic animal facility and euthanized following NJIT/Rutgers University IACUC protocol 17058‐A0‐R1.

2.2. Scanning electron microscopy (SEM)

The fleshy lip surrounding the adhesive disc (n _total = 3 individuals) was sampled for SEM (Figure 1). Frozen specimens were thawed, and the lip was dissected away from the rest of the specimen. The lip was then sectioned into three pieces that represented the anterior, middle, and posterior portions of the adhesive disc that were also used for histological analysis. These samples were placed in dH₂O for 24 hr before being fixed in a 2.5% glutaraldehyde solution. After complete fixation, samples were brought through an ethanol dehydration series to 100% EtOH. All samples were left in 100% EtOH overnight. The following day, samples were critical point‐dried and sputter‐coated with gold‐palladium alloy. All samples were imaged with a Joel 5000 Scanning Electron Microscope at 15 kV.

FIGURE 1.

Locations of histological and scanning electron microscope sampling. Dorsal view of remora cranial disc, showing anterior, middle, and posterior disc positions of sampling in both the transverse (fish right) and parasagittal (fish left) dimensions. ant, anterior; fl, fleshy lip; la, lamella; post, posterior; sp, spinules [Colour figure can be viewed at wileyonlinelibrary.com]

2.3. Histology

The lip from three remoras (ranging in size from 22.8 to 29.7 cm total length) was carefully dissected and used for histological examination. Eight samples from each lip were embedded for cross‐sectional, frontal, and parasagittal sectioning (n _total = 24 sections) from anterior, middle, and posterior locations on the lip (Figure 1).

Each sample was fixed in a 10% buffered formalin solution and then transferred through an ethanol dehydration series until 70% EtOH where it was stored. Evenly spaced sections of the lip (Figure 1) were dissected and infiltrated with Electron Microscopy Science JB‐4 Embedding media or embedded in paraffin after being fully dehydrated and the ethanol then removed with a clearing agent miscible in paraffin. Plastic samples were sectioned with either a glass knife (at 2.5 µm) or a diamond knife (at 1 µm) while paraffin samples were sectioned with a stainless‐steel knife (at 6 µm). Cross‐sections and parasagittal sections were taken at all locations sampled along the lip. Plastic sections were stained with Lee’s Basic Fuchsin and Methylene Blue, which allowed us to identify individual cell types based on their cellular morphology. Paraffin sections were stained with Mayer’s Hematoxylin and Eosin (Y), which allowed for further morphological description of collagen, or Verhoff’s stain to identify the presence of elastin fibers through a black dye.

2.4. Computed microtomography (CT) scanning

The head and disc of one remora were fixed in 10% formalin, stored in 70% EtOH, and µCT scanned in a Bruker Skyscan 1275 (Microphotonics, Inc.) located in the Otto York Bioimaging facility at the New Jersey Institute of Technology to visualize skeletal tissue. To visualize soft tissue, the specimen was then transferred to 3% phosphotungstic acid (PTA) for 3 weeks and scanned again at 125 µA, 80 kV, with 46 ms exposure time with a 1‐mm aluminum filter, resulting in a voxel size of 15 µm. Scanned images were reconstructed with nrecon reconstruction software (Bruker) and measurements made using mimics research suite v20.0 (Materialise, Leuven, Belgium).

2.5. Material properties

The soft tissue of the fleshy lip was assumed to exhibit both elastic and viscous characteristics similar to soft tissues in other organisms. Therefore, dynamic mechanical analysis (DMA) was employed to elucidate these characteristics. In DMA, a sinusoidal strain is applied to a sample and the resulting stress measured. From these measurements, storage and loss moduli are determined. The storage modulus represents a material’s ability to store energy elastically (similar to Young’s modulus in a solid), and the loss modulus is related to a material’s ability to dissipate energy (similar to viscosity in a fluid). DMA was performed on a 1‐cm³ section of tissue from the posterior region on the fleshy lip of a freshly euthanized Echeneis naucrates specimen. Uniaxial compression was applied in the frequency range 0.1–10 Hz with fixed strain amplitude at each frequency. The strain at each frequency was kept small (between 13 and 21 µm) to ensure the measurements remained in the linear viscoelastic region and to avoid crushing the sample. At least five compression cycles were recorded at each frequency with measurement times ranging from approximately 20 to 50 s each. Testing was carried out at room temperature (20°C).

3. RESULTS

3.1. Overview of the lip surrounding the remora adhesive disk

The lip surrounding the remora adhesive disc can be separated into five distinct regions based on cellular morphology, orientation, and proportions of different tissues. The regions are defined as followed and are used below when describing specific aspects of the lip morphology (Figure 2a).

• Region I: Epidermal layer on the dorsal surface. This refers solely to the stratified epithelium and epidermal structures embedded in this layer, such as papillae and push‐rod receptor complexes (Figure 3). Structures that begin in the epidermis but project into and/or through dermal layers are described here and in the dermal regions in which they appear.

• Region II: The dermal layer is located directly beneath Region I. This layer is primarily composed of longitudinally oriented collagen fibers across the horizontal plane of the lip. Depending on the anterior to posterior position, large interstitial channels were observed. In more anterior regions, we find a thin section of densely packed collagen fibers oriented in a crossed‐fiber pattern. This layer of collagen is oriented in the horizontal plane across the width of the lip. Numerous nerve fibers are oriented dorsoventrally and innervate dermal and epidermal sensory organs.

• Region III: This central layer of the lip varies in cellular composition from anterior to posterior. Here, we find multiple forms of collagen, elastin, fat cells, and connective tissues. Large, kinked collagen bundles oriented dorsoventrally through the lip are the hallmark of this region.

• Region IV: The layer is ventral to Region III and is composed of crossed‐fiber collagen. Just as in Region II the fibers are layered in the horizontal plane and are not organized dorsoventrally like the collagen fibers in Region III. Some nerve bundles may be found throughout this region. Skeletal muscle attaches to this region at the most posterior surface of the lip.

• Region V: Similar in composition to Region I, this is the epidermal layer on the ventral surface of the lip. There are no major structural differences between Regions I and V except that skeletal muscle extending from the axial body of the remora attaches to this surface on the posterior surface. Muscle fibers do not extend the full width of the lip but converge where the lip and adhesive disk meet.

FIGURE 2.

Schematic of a parasagittal section through the remora lip. (a) Schematic of overall lip histology. The lip is composed of five discrete layers composed of collagen, fat, and elastin. The third layer is the most complex and is formed of kinked collagen bundles (light pink), elastic fibers (black lines), and disorganized collagen tissues (pink cells with black nuclei) surrounded by a sea of fat, connective tissue and interstitial space (dark purple). (b) Zoom in of push‐rod receptor complex found in Region I. (c) Zoom in of frontal view of crossed‐fiber oriented collagen. Double headed‐arrow symbol designated a section through the tissue. C&D are slices of (a) in a frontal view as opposed to the parasagittal view in (a, b). (d) A 1‐µm coronal section through the remora lip to visualize the crossed‐fiber collagen. Sample is embedded in plastic and stained with Lee’s Basic Fuchsin and Methylene Blue. Scale set to 50 µm [Colour figure can be viewed at wileyonlinelibrary.com]

FIGURE 3.

Epithelial topography of the remora disc lip. SEM displaying the epithelial topography of the papillae and push‐rod receptors around the circumference of the disc lip. White arrows identify push‐rod receptors (Cohen et al., 2020)

3.2. Histology of the anterior lip

Region I, the most dorsal epidermal layer, is composed of a stratified epithelium studded with fields of papillae (Figure 4a,b,c, arrows). Each papilla is innervated by a single nerve and supported by a highly cellular cartilage (Figure 4b, star). Exposed in Region I but penetrating through dermal Regions II–IV are several push‐rod receptors (Figure 4d box, 4e). Each push rod receptor is a dome‐shaped sensory structure supported by cellular cartilage and innervated by three separate nerves (Figure 4e; Cohen et al., 2020). There is no sign of goblet cells or additional sensory structures such as taste buds on the surface of the dorsal lip or embedded in Region I.

FIGURE 4.

Histology of the anterior remora lip. (a) Parasagittal section through the remora lip showing layers I–IV. Arrow points to sensory papilla extending from the epidermal layer. Bold arrowhead highlights nerves throughout the lip. Inset zooms in on the cellular morphology of Region III. Arrowhead points to matrix of fibroblasts and collagen fibers and asterisks denotes large, kinked collagen bundles. (b) Sensory papilla supported by cell‐rich cartilage (star) and innervated by a single nerve (bold arrowhead). (c) SEM of dorsal remora lip surface showing the distribution of sensory papillae (arrows). (d) Coronal section through the anterior remora lip. Arrow points to sensory papilla, bold arrowhead points to nerves. Asterisk shows skeletal muscles insertion on Regions IV and V. (e) Inset showing cellular anatomy of push‐rod receptor. (f) Inset showing the orientation of the crossed‐fiber collagen bundles that make up Region IV and portions of Region II. (g) Parasagittal section of remora lip embedded in paraffin, sectioned at 5 µm and stained with Verhoeff’s solution to identify elastin fibers (black). Elastin is found clustered in Region III. Scale bars: (a,d) 200 µm; (b,c) 100 µm; (e,f) 20 µm; (g) 50 µm. Panels (a, b, d, e, and f) embedded in plastic, sectioned at 2 µm, and stained with Lee’s Basic Fuchsin and Methylene Blue [Colour figure can be viewed at wileyonlinelibrary.com]

Region II is composed of loosely connected and longitudinally oriented connective tissue and collagen (Figure 4a). There is an abundance of nerves in this region (Figure 4a,b,d, bold arrowhead) and supportive cellular cartilages supporting the base of push rod receptors. In the center of Region II is an area of much more densely packed connective tissue (Figure 4d,f) overlying a section of loosely organized collagen. The collagen in this dense region is oriented in a crossed‐fiber pattern across the width of the lip (Figure 4d,f).

Region III is the most complex and regionalized of all five regions (Figure 4a), characterized by three types of collagen, fat cells, interstitial channels, elastin, and nerves. The more dorsal and ventral portions of Region III are composed of dense irregular connective tissue formed by a matrix of fibroblasts and collagen fibers (Figure 4a inset, arrowhead). This region of connective tissue is packed tightly together with no extracellular matrix or space among adjacent fibers. The center of Region III is composed of thicker, kinked collagen bundles orientated dorsoventrally (Figure 4a inset, asterisk). The fibers range in length but are substantially longer than all other tissues in this region and link the dorsal and ventral edges of Region III. Elastin fibers are abundant in Region III (Figure 4g, black fibers). Fat cells separate the long‐kinked collagen bundles (Figure 4a, ‘f’ in g). Finally, the base and central nervous structure of the push rod receptor complexes are embedded through this Region.

Region IV, abutting the ventral extreme of Region III, is composed of densely packed collagen fibers oriented in a crossed‐fiber pattern (Figure 4a,f for reference of pattern). The collagen creates a lattice‐like structure similar to a rope made from helically or biaxially braided fibers. Unlike a rope where the helical bundles wrap around and create a closed cylindrical form, the crossed‐fiber collagen is oriented in the horizontal plane across the width of the lip. Region IV is thicker than Region II but not as thick as Region III at the anterior portion of the lip. Region V is the epidermal layer on the ventral face of the lip and is structurally similar to Region I. There is some indication that fewer papillae are present, but this was not consistent across specimens and could not be reliably quantified.

3.3. Histology of the middle lip

The middle portion includes the highest density of fat cells. Regions I and II are similar in morphology to the most anterior portion of the lip. In more posterior sections, the band of dense connective tissue is directly adjacent to Region III.

The middle of Region III is distinctly different from that in more anterior sections. First, there is a substantial increase in the density of fat cells between kinked collagen fibers (Figure 5a, arrows). Additionally, we find an increase in interstitial spaces between adjacent collagen bundles (Figure 5b). There are many nerves in the ventral edge of Region III extending from more anterior regions.

FIGURE 5.

Cellular morphology of the middle remora lip. (a) Parasagittal section of Regions III and IV highlighting cellular changes in the proportion of collagen from more anterior regions. Arrows point to fat cells. (b) Cross‐section through Region III of the lip highlighting the interstitial channels found throughout. (c) Inset zooms in on Region IV to show the crossed‐fiber collagen bundles. (d) Middle portion of the lip is heavily innervated. n = nerves. (a, c) 100 µm; (b) 50 µm; (d) 20 µm. Panel (a & b) sectioned at 2.5 µm, panel (c) sectioned at 1.5 µm. All three panels are from samples embedded in plastic and stained with Lee’s Basic Fuchsin and Methylene Blue [Colour figure can be viewed at wileyonlinelibrary.com]

Region IV is composed solely of crossed‐fiber collagen oriented in the same direction as the fibers found in Region II that are densely packed along the entirety of this region (Figure 5a,c). Major nerve bundles are seen here, passing out of the lip and likely towards the central nervous system (Figure 5a, ‘n’ in d,). There is no significant difference in the histology of Region V at the middle lip compared with that seen in more anterior sections.

3.4. Histology of the posterior lip

There is distinct regionalization in the arrangements and proportions of tissues within and across the five regions of the remora lip. In the posterior portions of the lip, Region IV is the largest and Region III is the thinnest compared with more anterior sections. Region I is more or less identical from anterior to posterior except that there are noticeable differences in the density of push‐rod receptor complexes.

Region II is very thin, and the ring of dense connective tissue found between layers of loose connective tissue is now directly adjacent to Region III (Figure 6a, arrow). In‐between Regions I and II, there is a series of evenly spaced interstitial channels (Figure 6b ‘ch’). There is a noticeable decrease in the density of fibroblasts and extracellular matrix at the dorsal and ventral edges of Region III (Figure 6b,c). The kinked collagen bundles of Region II are more densely packed than more anterior sections, with few or no fat cells between adjacent bundles.

FIGURE 6.

Cellular morphology of the posterior remora lip. (a) Parasagittal section showing Regions I–V. Arrow points to loose connective tissue composing Region II, star highlights large ventral channel. (b) Parasagittal section cut more medially highlighting the interstitial channels (ch) studded throughout Region II. (c) Inset shows changes to Region III; collagen bundles are packed more tightly with extracellular space. (d) From panel (a) highlighting a large channel that extends throughout the ventral surface of the lip. (a) 200 µm; (b) 100 µm; (c) 200 µm. All panels from samples embedded in plastic, sectioned at 2 µm, and stained with Lee’s Basic Fuchsin and Methylene Blue [Colour figure can be viewed at wileyonlinelibrary.com]

Region IV is the thickest layer at the posterior lip. The crossed‐fiber collagen at the posterior lip attach to the kinked bundles of Region II. Like the middle and anterior sections, we find numerous nerves running the length of the lip. There is an additional large interstitial channel found running through the posterior lip consistent with morphology seen in the PTA‐stained CT scan (Figure 6a,d, ‘star’). Attaching to Region IV are skeletal muscle fibers extending from the axial musculature of the body of the remora. The skeletal muscle fibers attach to the collagen directly. There are no notable differences in the morphology of Region V as compared with the rest of the lip.

3.5. Micro‐computed tomography of the remora lip

Micro‐computed tomography (µCT) of the muscle of the remora disc enabled visualization of contrast‐stained soft tissues (Figure 7). Three groups of erector muscles, medial, lateral, and central (mer, ler, cer, purple) were identified: the mer originates anterolaterally on the medial aspect of bony ventral prominence at base of lamellae and inserts posteromedially on the spinous process of the interneural ray and skull roof, the ler originates anteromedially on the lateral aspect of bony ventral prominence at the base of lamellae and inserts posterolaterally on the cranial roof, and the cer extend ventrally from the ler anterior to neurocranium and ventrally from the mer posterior to neurocranium. Medial and lateral depressor (mdep and ldep, respectively, yellow) muscle fibers are oriented dorsoventrally; they originate on the roof of the neurocranium and insert onto the ventral surface of the posterolateral aspect of the intercalary plates. The anterior cardinal sinus (acs, blue) separates the erectors and depressors through the anterior two‐thirds of the disc. Marginalis muscles (m, red; Houy, 1910) originate on the anteroventral aspect of the intercalary bone and insert into the proximal lip via a collagenous sheath. Not previously described are two bilateral anterior compressor muscles (comp, orange) which are situated in the center of the ring‐shaped orbital sinuses with fibers oriented in a dorsoventral direction. While the axial muscles of the body do not interact with the disc, the supracarinalis muscles (sc, green), which would ordinarily be attached to the posterior margin of the dorsal fin spines from which the disc is derived, insert into the midbody of the disc and extend caudally along the dorsal midline to the anterior fin ray of the dorsal fin.

FIGURE 7.

Micro‐computed tomography (µCT) of the remora head soaked in phosphotungstic acid. (a) Cross‐sectional image of muscles of disc at location of dashed line in (b,c). Scale bar: 1 cm. (b) Dorsal view of disc musculature, with skeletal elements of disc removed. (c) Ventral view of disc musculature and vasculature. Scale bars: 5 mm. (d) 3D reconstruction of µCT scan of ventral side of disc with muscle origins and insertions superimposed. Scale bar: 1 cm. mer, medial erector; ler, lateral erector; cer, central erector, purple; acs, anterior cardinal sinus, blue; mdep, medial depressor; ldep, lateral depressor, yellow; ju, jubilee, pink; m, marginalis, red; comp, anterior compressor muscles, orange; sc, supracarinalis, green; drad, distal radial bone; ical, intercalary bone; inr, interneural ray; lam, lamella; lp, lamellar process; all bones are tan [Colour figure can be viewed at wileyonlinelibrary.com]

The ultimate pectinated lamella (upl; Figure 8) is a single hollow‐chambered bone at the posterior of the disc, embedded caudally within the proximal margin of the soft tissue lip. The anterior of the ultimate pectoral lamella contains a foramen through which a vein passes from a small sinus; multiple smaller foramina are found on the posterior end of the ultimate pectinate lamella providing passage for smaller veins into the dorsal and ventral margins of the soft tissue lip. Vessels entering the ultimate pectoral lamella via the small sinus originate from the caudal end of the anterior cardinal sinus (acs, Figure 7c), indicating that they are venous in nature. Surrounding these veins as they enter the anterior foramen of the ultimate pectinated lamella is a circular formation of dorsoventrally arranged muscles encapsulated within their own epimysium and thus discrete from surrounding disc musculature. We have named these previously undescribed muscles the jubilee muscles (ju, pink, Figure 8b‐d) because, like a jubilee clamp that is used to tighten around tubing for closure, we believe that their contraction would constrict the veins passing through their center given the isovolumetric property of muscles.

FIGURE 8.

Micro‐computed tomography of the remora disc and vascular hydrostat control mechanism after soaking in phosphotungstic acid (PTA). (a) Dorsal view of posterior skeletal elements of remora disc. (b) Lateral view of midsagittal section of posterior disc after treatment with PTA. (c, d) Anterior and lateral views, respectively, of three‐dimensional reconstruction of the jubilee muscle and veins that control the vascular hydrostatic mechanism of the remora disc lip. Scale bars: 1 cm. ju, jubilee, magenta; veins, blue; upl, ultimate pectinated lamella, tan; ch, chamber [Colour figure can be viewed at wileyonlinelibrary.com]

3.6. Material properties of the remora disc lip

The measured storage and loss moduli of the remora disc fleshy lip are shown in Figure 9. The storage modulus was in the range of 4–14 kPa, which is similar to that of a gel or soft polymer foam, and the loss modulus was of similar magnitude to the storage modulus. This indicates that the fleshy lip tissue is both highly compliant and energy‐dissipating.

FIGURE 9.

Storage and loss moduli of remora lip tissue. Blue, measured storage modulus (closed circles); red, measured loss modulus (open circles) [Colour figure can be viewed at wileyonlinelibrary.com]

4. DISCUSSION

4.1. Morphological summary

An advantage to using µCT scanning to dissect the remora soft tissue digitally is that it allowed us to identify several muscle groups that were misidentified in previous literature; this is not surprising given the complex architecture of the multiple bundles of interwoven erectors and depressors and the inherent difficulty in dissecting them. The erector muscles identified here were consistent with previous literature; however, what were previously identified as depressor muscles were determined to be the marginalis muscles (Fulcher and Motta, 2006). The lateral and medial depressors, compressor muscles, jubilee muscles, and insertion of the supracarinalis muscles into the disc were not identified in previous literature.

Our analysis of the soft tissue lip encircling the remora adhesive disc revealed several potential mechanisms involved in adhesion not previously described for remoras. First, the papillae on the epithelium of the lip may be important for creating a tight hold against a host surface. Increased surface area allows for an increase in contact surface against the local asperities of rough surfaces. The papillae may then add a frictional component to adhesion, in the same way that manufactured gripping surfaces often bear small bumps. Similar papillae are noted in the adhesive organ of the rheophilic sisorid catfish Pseudocheneis sulcatus (Das and Nag, 2009). Alternatively, spaces between the papillae could provide a network of interconnected fluid spaces connected to the area under the disc, allowing the sub‐ambient pressure differential created under the disc to be extended through the lip contact surface, thereby creating a pressure differential throughout the entire contact area of the disc, similar to the mechanism in octopus infundibulum (Kier and Smith, 2002). Similar small projections have been noted on the suckers of clingfish and lumpsuckers (Green and Barber, 1988; Wainwright et al., 2013). The papillae in the remora lip are about 20–30 µm in diameter (Cohen et al., 2020), as compared with 3–4 µm in octopus (Kier and Smith, 2002) and 250 µm in clingfish (Wainwright et al., 2013). However, the papillae of both clingfish and lumpsuckers are more tile‐shaped and made of nu,merous microvilli which may indicate a different kind of surface interaction driving adhesion for those species.

Second, the tissues of the remora lip create a multi‐layered sponge capable of different functions dependent on collagen, elastin, and fat composition and orientation. The cellular morphology of Region III could permit the lip to conform tightly to any surface against which it was pressed, and the crossed‐fiber collagen bundles found in both Regions II and IV may act to resist forces in any given direction.

Third, there is no skeletal muscle within the lip surrounding the adhesive disc, rather, the marginalis muscle is only found attaching to the ventral aspect of the proximal perimeter of the lip. Contraction of those muscles would increase the tension on the lip surface that is not in contact with the attachment surface, potentially acting to peel it back. Therefore, the muscles that insert into the lip are not a part of the sealing mechanism itself but are likely important for detaching from a host. The newly described compressor muscles surrounding the anterior foramen of the disc presumably pull the anterior disc and skull close together when contracted. This action would thereby streamline the anterior body, reducing drag while attached to a swimming host. It may also provide an additional method to modulate pressure under the disc if fluid is retained or expelled from the anterior foramen by compressor muscle actuation.

Finally, discovery of a novel jubilee muscle bundle surrounding venous return from the lip provides a mechanism by which blood may be trapped within the lip tissue, causing it to become engorged, thus changing its tensile properties. A wide‐mesh network of capillaries running throughout the disc, which was also described previously by Bargmann (1973), provides the blood flow for the vascular hydrostat. Potential tissue damage by increased pressure from fluid buildup near the jubilee muscles is mediated by exiting the lip through the chamber within the thin‐walled ultimate pectinated lamella bone.

4.2. Functional role of the soft tissue

To truly understand the role of the disc lip during remora–host attachment we must consider the tension dynamics of the soft tissues. The biases of collagen fiber orientation will necessarily affect the deformation, function, and modularity of the lip (Fratzl, 2008). Collagen is a complex material. Aside from being resilient it is easily able to increase energy dissipation under deformation, in part due to its helical structure (Buehler, 2006). We find the greatest density of collagen bundles and elastin in Region III. We suspect that the kinked bundles found throughout Region III are Type 1 collagen; this type of collagen is specifically known for its cell‐to‐cell adhesions and is the main component in most extracellular networks (Kielty et al., 1993; Ikoma et al., 2003). Under load, the kinked fibers are pulled straight, likely resisting vertical forces from separating the remora from its hosts. The kinking of collagen fibers that were observed may be responsible for the lower modulus at small strains, whereas the fibers exhibited higher modulus behavior when pulled taut. Alternatively, the crossed‐fiber collagen found on the dorsal and ventral faces of the lip, when under tension, likely aid in lateral and compressive forces during attachment. Material testing found that the lip is both highly compliant and energy‐dissipating. Compliance allows the soft spongy tissue of the lip to deform precisely along any surface asperities, thereby creating a perfect seal when pressed against a host surface. The energy‐dissipating elastic properties of the fleshy lip would provide tension that may result in gripping of the surface roughness, as well as damp forces pulling on the disc lip.

The mechanical relationship between collagen and interstitial fluid is largely unknown (Stylianopoulos and Barocas, 2007). However, the movement of fluid throughout soft structures is vital for viscoelastic seals. In the remora lip, we find complex matrices of various types of collagen interspersed with large interstitial channels. The combination of these morphologies with the additive knowledge of the mechanics of collagen leads us to believe that the remora lip acts as a hydrostatically mediated viscoelastic seal.

4.3. Soft tissue contribution to remora adhesion

A tight seal against the host surface is crucial for reducing seep (Beckert et al., 2016); seep resulting from the pressure differential below the disc as compared with ambient pressure will cause equalization of pressure and loss of suction adhesion. We hypothesize that the remora lip is effective in reducing seep and enhancing adhesion using a hydrostatic gripping mechanism. When a remora comes into contact with its host, it presses the disc against the host body to expel the majority of fluid from beneath the disc (Beckert et al., 2015). This results in the soft spongy tissue of the disc lip conforming to the local roughness and curvature of the attachment surface. The spongy region of the lip is a series of interstitial channels through which blood flow would increase the volume of the lip under compression. The large cranial veins beneath the remora disc (Flammang and Kenaley, 2017) provide blood to the lip via passage through the jubilee muscle and into the chamber within the ultimate pectinated lamella. Increasing the volume of fluid within the disc lip results in increased stiffness of the lip that is maintained by the elastic tension of the elastin fibers resisting deformation. In addition, the crossed‐fiber collagen of Regions II and IV serve to maintain the integrity of the hydrostat. The combined effect is that the engorged lip generates tension across the contact surface thereby increasing grip force.

Few biological vascular hydrostats have been studied in detail. Within vertebrates, most hydrostatic mechanisms are muscular in nature (e.g. tongues, elephant trunk); the mammalian penis is the only other known vertebrate structure to use blood flow as a method of mechanically stiffening soft tissue to perform a behavioral function (Kier, 2012). Already a model of exaptation with bony elements derived from dorsal fin spines and the cranial vein hydraulic differential mechanism within the disc, it is perhaps no surprise that the remora disc lip is also an elaborate remodeling of locally available tissues. Therefore, the remora adhesive disc remains not only a new method to understand tissue viscoelasticity and vascular hydrostats, but a window into the evolution of functional novelty.

CONFLICT OF INTEREST

The authors declare no competing interests.

AUTHOR CONTRIBUTIONS

K.E.C. contributed to the design of the study, acquisition of data, data analysis/interpretation, drafting of the manuscript, critical revision of the manuscript and approval of the article. C.H.C. contributed to acquisition of data, critical revision of the manuscript and approval of the article. L.P.H. contributed to the design of the study, acquisition of data, data analysis/interpretation, critical revision of the manuscript and approval of the article. M.B. contributed to acquisition of data, data analysis/interpretation, critical revision of the manuscript and approval of the article. J.H.N. contributed to data analysis/interpretation, critical revision of the manuscript and approval of the article. B.E.F. contributed to the concept and design of the study, acquisition of data, data analysis and interpretation, drafting of the manuscript, critical revision of the manuscript and approval of the article.

ETHICAL APPROVAL

All study animals were handled humanely and ethically following New Jersey Institute of Technology/Rutgers University IACUC protocol 17058‐A0‐R1.

Cohen KE, Crawford CH, Hernandez LP, Beckert M, Nadler JH, Flammang BE. Sucker with a fat lip: The soft tissues underlying the viscoelastic grip of remora adhesion. J. Anat. 2020;237:643–654. 10.1111/joa.13227

DATA AVAILABILITY STATEMENT

Data are available upon request.