Biol. Bull. 213: 307-315. (December 2007)
© 2007 Marine Biological Laboratory
Granular Chitin in the Epidermis of Nudibranch Molluscs
Rainer Martin1,*,
Sabine Hild1,
Paul Walther1,
Kerstin Ploss2,
Wilhelm Boland2 and
Karl-Heinz Tomaschko1,
1 Central Facility for Electron Microscopy, University of Ulm, Albert-Einstein-Allee 11, D 89069 Ulm, Germany
2 Max Planck Institute for Chemical Ecology, Hans-Knoell-Straße 8, D 07745 Jena, Germany
* To whom correspondence should be addressed. E-mail: rainer.martin{at}uni-ulm.de
 |
Abstract
|
|---|
Chitin is usually found in stiff extracellular coatings typified by the arthropod exoskeleton, and is not associated with the soft, flexible mollusc skin. Here, we show, however, that chitin in nudibranch gastropods (Opisthobranchia, Mollusca) occurs as intracellular granules that fill the epidermal cells of the skin and the epithelial cells of the stomach. In response to nematocysts fired by tentacles of prey Cnidaria, the epidermal cells of eolid nudibranchs (Aeolidacea) release masses of chitin granules, which then form aggregates with the nematocyst tubules, having the effect of insulating the animal from the deleterious action of the Cnidaria tentacles. Granular chitin, while protecting the animal, does not interfere with the suppleness and flexibility of the skin, in contrast to the stiffness of chitin armor. The specialized epidermis enables nudibranchs to live with and feed on Cnidaria.
|
Introduction
|
|---|
Chitin, a homopolymer of N-acetyl-D-glucosamine, is probably the most abundant biological material in the animal kingdom. The stiff chitin exoskeleton of arthropods may have been a decisive factor in the success of this phylum. Notwithstanding this success story, there appear to be major disadvantages of this construction. For example, the chitin armor has to be shed and renewed when the animal grows. Also, the organisms are exposed to predators in the early after-molt phase. Finally, for the sake of flexibility and mobility, the chitin plates have to be interrupted by numerous joints.
We report here on a different structural arrangement in molluscs, which may have overcome these disadvantages. Nudibranch slugs (Gastropoda, Opisthobranchia), especially the eolids (Nudibranchia, Aeolidacea), which have no shell, are exposed to nematocysts fired by their prey (Martin and Walther, 2002). They have a specialized skin, in which the epidermal cells contain masses of large vesicles that are each filled with a flattened oval granule (Trinchese, 1881; Henneguy, 1925; Edmunds, 1966; Schmekel and Wechsler, 1967; Porter and Rivera, 1980; Martin and Walther, 2003). An eolid epidermal cell has about 25 times the volume of a comparable cell of a sacoglossan species (Opisthobranchia, Sacoglossa) lacking these granules (Martin et al., 2007). Stomach epithelial cells, especially those of nudibranchs feeding on Cnidara, are also filled with similar granules (Henneguy, 1925; Graham, 1938; Martin et al., 2007). It has been suggested that the granules in the nudibranch epidermal cells have a protective function (Graham, 1938; Edmunds, 1966; Martin and Walther, 2003). We show that the basic structure of these granules—referred to as spindles—is chitin.
|
Materials and Methods
|
|---|
Specimens of Cratena peregrina (Gmelin, 1791) and Flabellina affinis (Gmelin, 1791) were collected by scuba diving near the islands of Giglio or Elba (Tuscany, Italy). The experiments with hydroid cnidophores and the extraction of spindles were carried out at the Institute for Marine Biology Dr. Claus Valentin at Giglio. Some animals were brought alive to the laboratory in Ulm.
High-pressure freezing and chemical fixation for transmission and scanning electron microscopy
Cerata of the slugs were prepared for transmission electron microscopy (TEM) at Ulm by high-pressure freezing, using a Wohlwend Compact 01 high-pressure freezing machine (Technotrade Inc., Manchester, NH, USA), and freeze substitution (Müller and Moor, 1984; Studer et al., 1989; Martin et al., 2007), or were chemically fixed at the field station on Giglio in fixatives prepared daily, consisting of 2% paraformaldehyde and 2% glutaraldehyde in filtered seawater, which had been mixed 1:1 with H2O, containing 0.35 mol l–1 sucrose, 0.17 mol l–1 NaCl, and 0.1 mol l–1 Na-cacodylate, pH 7.6. For scanning electron microscopy (scanning EM) they were critical-point-dried, mounted, and coated with platinum. The scanning electron microscopes used were a Zeiss DSM 962 and a Hitachi in-lens FE-S5200. For immunocytochemistry, the glutaraldehyde in the fixative was reduced to 0.5% and, after 45 min in the fixative, the specimens were stored in fixative diluted 1:10 with seawater. The specimens for TEM were osmicated, dehydrated, and embedded in Epon. The contrast of the spindles in sections was much improved by staining in a drop of 0.5% KMnO4 in water for 2 min, before treatment with lead citrate.
Isolation of spindles and boiling in KOH
The animals were killed by decapitation and immersed, without the heads, in a solution of 0.1 g sodium dodecyl sulfate and 10 µl Triton X-100 in 100 ml of H2O. They were shaken by hand for about 30 min, particulates were sedimented in a hand-driven desktop centrifuge, and the supernatant was centrifuged at 16, 060 x g for 10 min. The pellets with numerous spindles were rinsed several times in water, repelleted, frozen, and stored at –25 °C.
For the isolation of chitinous elements, the spindles and the heads of the slugs were boiled for 2 h in 10% KOH (w/v, in water), at 96 °C. The spindles were centrifuged at 16, 060 x g, the supernatant was discarded, and the samples were washed in water and repelleted several times.
Incubation of isolated spindles with enzymes
Pellets of isolated fresh spindles of Flabellina affinis were incubated in one of three ways: (1) with trypsin (from bovine pancreas, Sigma T-6567, 3.1 mg/ml) in 0.05 mol l–1 Tris/HCl buffer with 10 mmol l–1 CaCl3, pH 7.6, at 37 °C for 60 min; (2) with proteinase K (from Tritirachium album, Sigma P-2308, 1 mg/ml) in 0.05 mol l–1 Tris/HCl buffer with 50 mmol l–1 EDTA, pH 8.0, at 37 °C for 40 min; (3) or with chitinase (from Streptomyces griseus, Sigma C-6137, 2.5 mg/ml) in 0.1 mol l–1 phosphate buffer, pH 6.0, at room temperature, for 2 h. After incubation, the pellets were fixed in 2.5% glutaraldehyde in 0.1 mol l–1 cacodylate buffer and processed for TEM.
Immunocytochemistry
The antibody was raised in rabbits against purified and finely dispersed chitin from crab shells of 10 to 100 µm particle size (Spindler-Barth and Buss, 1997). Thick 500-nm sections of resin-embedded heads and cerata of the slugs were collected and attached by heat to glass slides. To expose the organic material, the resin was partially removed by sodium methoxide (Mayor et al., 1961), 1 min at room temperature, rinsed in methanol/benzene and acetone, and washed three times in 0.1 mol l–1 Tris/HCl buffer, pH 7.5. The sections were then preabsorbed with buffer containing 2% gelatin and 0.5 % bovine serum albumin, for 45 min at room temperature, and incubated overnight at 5 °C with anti-chitin, diluted 1:60 in Tris-buffered saline (3% NaCl in 0.1 mol l–1 Tris/HCl buffer, TBS) with 10% preabsorption buffer. They were then rinsed in TBS six times and incubated with a second antibody marked with CY2 dye, 1:50, at room temperature, in the dark. They were examined in a fluorescence microscope. The immunoreaction was controlled by omitting the anti-chitin antibody, and by preabsorption of the chitin antibody with triacetylchitotriose, which reduced the intensity but did not completely abolish the immunostaining. Pretreatment of the sections with chitinase also diminished the immunostaining.
Acidic hydrolysis of spindles, crab chitin, and standards
Spindles (ca. 0.1–0.7 mg) of Flabellina affinis or Cratena peregrina, or of authentic crab chitin (ca. 0.5 mg, Fluka) and arabitol (0.5 mg, as internal standard) were suspended in 0.10 ml of water to which 0.15 ml of concentrated hydrochloric acid (32%) was added. The mixture was heated to 60 °C for 24 h. To completely degrade residual spindle or crab polymers, more concentrated hydrochloric acid (0.05 ml, 37%) was added, and heating was continued until the solids disappeared (1 to 2 days at 60 °C). The hydrolysis of authentic N-acetyl-glucosamine required only 5 h of heating at 60 °C, followed by addition of more hydrochloric acid (50 µl, 37%) and heating to 60 °C for another 20 h. After cooling to room temperature, the mixture was evaporated to dryness by a stream of nitrogen. For gas chromatographic and mass spectroscopic analysis, the dry residue was silylated by treatment with 50 µl of MSTFA (N-Methyl-N-(trimethylsilyl)trifluoroacetamide) and 50 µl of pyridine. Silylation was completed by heating to 60 °C for 2 h. The mixture was diluted with dichloromethane (1:5, v:v) prior to analysis.
Degradation of spindle polymers by chitinase and N-acetyl-glucosaminidase
Spindles (ca. 0.1–0.7 mg) and arabitol as an internal standard (10% of the amount of the biopolymer) were suspended in phosphate buffer (1.0 ml, 50 mmol l–1 ) adjusted to pH 6.0. Chitinase (1 U, from Streptomyces griseus, Sigma C-6137) and N-acetyl-glucosaminidase (1 U, from Jack beans, Sigma A-2264, suspension in 2.5 mol l–1 ammonium sulfate) were added, and hydrolysis was achieved with gentle shaking at 37 °C for 7 days. Solids were removed by centrifugation, and 0.2 ml of the clear solution was evaporated to dryness by a stream of nitrogen. For gas chromatography and mass spectroscopic analysis, the dry residue was silylated as above.
Gas chromatography and mass spectroscopic analysis of spindle hydrolysates
The silylated monomers of the degraded spindles were analyzed by combined gas liquid chromatography and mass spectroscopy, using a Trace GC connected to a Trace MS detector (both from Thermo Finnigan, San Jose, CA, USA). Compounds were separated on an EC-5 column (15 m x 0.25 mm i.d., 0.25 µm film thickness; Alltech, Deerfield, IL, USA). Helium at a flow rate of 1.5 ml/min served as carrier gas, and a split mode injection (1:10) was employed. The GC injector, transfer line, and ion source were set at 220 °C, 280 °C, and 280 °C, respectively. Spectra were taken in the total-ion-scanning (TIC) mode at 70 eV. Compounds were eluted under programmed conditions starting at 120 °C (2-min isotherm), followed by heating at 10 °C min–1 to 180 °C, then at 5 °C min–1 to 210 °C, and finally at 10 °C min–1 to 280 °C maintained for 5 min; the injected amount was 1 µl.
Confocal Raman microscopy
Raman spectra were performed using a confocal Raman microscope (WITec GmbH, Ulm, Germany) equipped with a NdYag laser with a wavelength of 532 nm for excitation. An Olympus 100x (numerical aperture = 0.95) objective was employed for all measurements. Raman spectra were recorded on 20-µm-thick films of spindles extracted from Cratena peregrina, which had been boiled in KOH. Spindle suspensions in water from three different preparations were left drying on glass slides. As a reference, a 50-µm-thick film of crab carapace chitin (Fluka, cat. no. 22718) was prepared. The films were analyzed by taking individual spectra with an integration time of 5 s. Due to the different film thicknesses, the spectra obtained from the spindles and the chitin reference had different count rates, so the spectra were normalized, setting the alkane peak (C-C, C-H) at 2850 – 2960 cm–1, for comparisons.
FT-IR spectroscopy
Spectra were recorded on a Bruker Equinox IFS 55, Fourier-transform-infrared spectrometer. Spindle material of Flabellina affinis was measured as a KBr pellet; to avoid partial deacetylation, the material had not been pretreated with KOH. The scan range was adjusted from 4000 to 600 cm–1 with a spectral resolution of 0.5 cm–1. Spectra were averaged from 32 scans of background and sample. The spectra of crab chitin and spindle material were normalized (intensity minimum = 100% transmission, and maximum intensity = 0% transmission).
|
Results
|
|---|
High-pressure frozen and freeze-substituted samples of the eolid skin most clearly revealed the in vivo structure of the epidermal cells with their large vesicles, each filled with a spindle. The sagitally or parasagitally cut spindles appeared as sticks with caps at both ends (Fig. 1). The spindles can be extracted in SDS and Triton-X-100 and isolated (Fig. 2A). They were biconcave, oval granules of 5.13 µm length (+0.18 SEM, n = 33; Flabellina affinis), with irregular filaments spanning the cavities (Fig. 2B). The spindles resisted boiling in 10% KOH for 2 h (Fig. 2C). High-resolution scanning EM images showed a fibrillar ultrastructure in KOH-boiled spindles (Fig. 2D), the fibrils measuring about 12 nm in diameter. Isolated undischarged nematocysts in the spindle preparations, presumably extracted from the cnidosacs of the cerata, did not resist boiling in KOH.
View larger version (113K):
[in this window]
[in a new window]
|
Figure 1. A high-pressure cryo-fixed epidermal cell of Flabellina affinis with a layer of microvilli-like processes embedded in mucus (mv); the nucleus (nu); and many spindles (sp) in large vesicles. When the spindles are incised sagittally, they appear as sticks with two caps (sp*).
|
|
View larger version (149K):
[in this window]
[in a new window]
|
Figure 2. Isolated spindles in light (A, C), and scanning electron (B, D) micrographs: the samples are fresh (A), chemically fixed (B), and boiled in KOH (C, D). High-resolution scanning EM reveals the fibrillar structure (D).
|
|
In the heads of the slugs, the radula with the rows of teeth also resisted boiling in KOH, as did the expanded cuticles lining the pharynx, the radula pouch, and the esophagus (Fig. 3A). In TEM sections it was clear that these cuticles were the extracellular linings of the epithelia. An abrupt change was apparent at the transitional junction between the esophagus and the stomach, at which point the stomach epithelial cells, lacking a cuticle, were filled with spindles (Fig. 3B) (see Martin and Walther, 2003, fig. 7a; Martin et al., 2007, fig. 8a). Both the KOH-boiled radula teeth and cuticles in high-power scanning EM revealed a compact organized fibrillar ultrastructure (Fig. 3C, D), in the same range of diameters, as was the case with the KOH-boiled spindles (Fig. 2D).
View larger version (110K):
[in this window]
[in a new window]
|
Figure 3. Cuticles and spindles in the alimentary tract. (A) The head of a specimen of Cratena peregrina after boiling in KOH: the radula (ra), as well as the cuticles lining the pharynx (ph), the radula pouch (rp), and the esophagus (es), resisted KOH. (B) The transition of the esophagus (es) into the stomach (st) (large arrow): the esophagus epithelium is lined by an extracellular cuticle (arrowheads), which is absent in the stomach epithelium. The stomach epithelial cells are filled with vesicles (small arrows, inset). (C, D) High-resolution scanning EM of the radula (C) and the cuticle of the radula pouch (D) after boiling in KOH, showing the fibrillar nature of these structures.
|
|
View larger version (31K):
[in this window]
[in a new window]
|
Figure 7. (A) The Raman spectra of the reference crab chitin (a) and the spindles of Cratena peregrina (b). (B) A list of characteristic Raman shifts (Galat and Popowicz, 1978).
|
|
View larger version (35K):
[in this window]
[in a new window]
|
Figure 8. FT-IR spectra from crab chitin and spindles from Flabellina affinis. Relevant absorption characteristics for chitin are also found in the spindle material.
|
|
We tested for the effects of enzymes on spindle structure. In TEM sections of pellets, the spindles appeared as sticks with caps or as irregular crescents (Martin et al., 2007: Fig. 3b). Isolated, fresh spindles incubated with proteinase K (Fig. 4A), or trypsin did not exhibit structural alterations when compared to untreated spindles. Also, treatment with mercaptoethanol, which in extracts subjected to electrophoresis yielded two distinct protein bands, did not obviously affect the fine structure of the spindles. A distinct structural effect, however, was discerned after short (2-h) incubations with chitinase. The spindles were significantly shorter (Student's t test, 2.27 µm +0.9 SEM, n = 36, versus 2.77 µm +0.9 SEM, n = 66; P < 0.001) than untreated spindles and appeared collapsed (Fig. 4B), indicating that the basic spindle structure, as seen in TEM, is chitin. Undischarged nematocysts, which occurred randomly in these pellets, were not obviously altered in their ultrastructure after chitinase treatment.
View larger version (142K):
[in this window]
[in a new window]
|
Figure 4. The effect of enzymes on the structure of isolated spindles of Flabellina affinis: (A) after incubation with proteinase K, they do not look different from untreated spindles; (B) after incubation in chitinase, they were smaller and distorted. Transmission electron microscopy.
|
|
An antibody, which was raised in rabbits against crab chitin, revealed that the radula teeth (Fig. 5A), and the cuticles of the head alimentary tract (Fig. 5B), as well as the spindles in epidermal cells (Fig. 5C) were immunoreactive. This is evidence for a similar antigenic component in the radula, the cuticles of the head alimentary tract, and the intracellular spindles.
View larger version (162K):
[in this window]
[in a new window]
|
Figure 5. An antibody raised against crab chitin is immunoreactive in the radula (A), in the cuticula of the esophagus (B), and in the spindles of cerata epithelial cells (C) of Flabellina affinis. Fluorescence microscopy.
|
|
More information on the molecular composition of the spindles was obtained by hydrolysis with strong acid (HCl) or by enzymatic degradation with chitinase and N-actetyl-glucosaminidase. The hydrolysis with concentrated hydrochloric acid proceeded smoothly and showed glucosamine (2-amino-2-desoxy-D-glucopyranose) as the only major product. The same compound was obtained by hydrolysis of authentic chitin from crabs or from N-acetyl-glucosamine. Silylated glucosamine from the hydrolysis of spindles, crab chitin, and N-acetyl-glucosamine eluted as two peaks with almost identical spectra. Reference and degraded spindle material displayed identical retention times and fragments of comparable intensity at m/z 434 (10%), m/z 304 (35%), m/z 216/217 (35%), m/z 203 (100%), m/z 147 (35%), and m/z 73 (55%). Enzymatic hydrolysis of spindle material and crab chitin with chitinase and N-acetyl-glucosaminidase proceeded slowly, but coincidently yielded N-acetyl-glucosamine along with some unidentified side products. Analogous to silylated glucosamine, the silylated N-acetylglucosamine eluted also as two well-separated peaks with almost identical spectra (Fig. 6) coinciding with the retention time and the fragments of the authentic reference. Spindles of both Flabellina affinis and Cratena peregrina were examined; specific differences were not evident.
View larger version (20K):
[in this window]
[in a new window]
|
Figure 6. Mass spectrum of silylated N-acetyl-glucosamine obtained from enzymatic degradation of spindles from Flabellina affinis. Coincident spectra were obtained from silylation of N-acetyl-glucosamine or from the degradation product of crab chitin.
|
|
The Raman spectrum in Figure 7 corresponds to the spectrum of the reference crab chitin in many characteristic peaks. The same holds for the FT-IR spectrum of spindle material and crab chitin (Fig. 8). Both samples generated similar transmission spectra consistent with previously published data (Pearson et al., 1960; Brugnerotto et al., 2001; Yamaguchi et al., 2005). Since the absorption around the reference band at 2900 cm–1 was of similar intensity in the range from 1000 to 1200 cm–1, the spindle sample was still largely acetylated.
Spindles are shed upon attack of nematocysts
Cnidophores are the long defense tentacles of the prey hydroid Eudendrium racemosum (Cnidaria, Hydrozoa), which are covered with holotrichous isorhiza nematocysts. Contact of cnidophores with cerata of Flabellina affinis or Cratena peregrina elicited a massive discharge of nematocysts (Martin and Walther, 2002, 2003). In a marginal area, where a single nematocyst tubule had impacted a ceras, the effect of the nematocysts was visible in greater detail. In the scanning EM image of Figure 9A, the tubule adhered to the ceras surface and left a cleft-like furrow with holes as it traversed along the skin. In other areas the ceras skin had apparently erupted to expose spindles that were extruded from the epidermal cells (Fig. 9B). Where numerous tubules adhered to the ceras, there were also numerous spindles (Fig. 9C). The spindles appeared to be hooked by the spines of the tubules, the spines and the spindle filaments functioning like a hook-and-loop (Velcro) fastener (Fig. 9D, E). When we observed living specimens, the cnidophores first adhered to the cerata; after several minutes they detached, and the cerata were liberated from the cnidophores (not shown).
View larger version (166K):
[in this window]
[in a new window]
|
Figure 9. Scanning EM (A, B, C, E) of the ceras surface of Flabellina affinis after contact with a cnidophore of Eudendrium racemosum. (A) A tubule of a holotrichous isorhiza nematocyst (arrows) grazed the ceras and left a fissured track within the skin (arrowheads). (B) Spindles (sp) are thrust through skin eruptions and liberated. (C) An area with many nematocyst tubules in which free spindles mix with the tubules. (D) A light micrograph of a discharged small microbasic eurytele nematocyst (arrow), presumably from a cnidosac, with several free spindles (arrowheads) attached to the tubule. (E) A spindle (sp) attached to the spines (arrowheads) of a holotrichous isorhiza nematocyst tubule.
|
|
|
Discussion
|
|---|
Chitin appears to occur in eolid nudibranchs in three different organs: first, in the radula teeth; second, in cuticles of the head alimentary tract, which form the inner lining of the pharynx, the radula pouch, and the esophagus; and third, in intracellular granules of the epidermal cells of the skin and the gut epithelium, the spindles.
Gastropod radula teeth have previously been shown to include chitin (Peters, 1972; Salwini-Plawen and Nopp, 1974; Peters and Latka, 1986). For the identification of chitin in the other structures, we rely on the following facts: (1) They resisted boiling in KOH (Figs. 2C, D); few biological materials resist this procedure—the major examples are cellulose and chitin. (2) After treatment with KOH, the radula, as well as the cuticles and the spindles, showed a fibrillar fine structure (Figs. 2D; 3C, D). (3) The three structures were immunoreactive with an anti-chitin antibody (Fig. 5). (4) Chitinase had a destructive effect on the isolated spindles, while proteinase K and trypsin did not affect the spindle structure, as seen in TEM (Fig. 4). This indicates that the effects found after chitinase treatment were not due to peptidases contaminating the chitinase. (5) Degradation experiments with concentrated acid (HCl) afforded glucosamine, and degradation with the enzyme pair chitinase and N-acetyl-glucosaminidase yielded N-acetyl-glucosamine, the characteristic monomer of chitin (Fig. 6). (6) The Raman (Fig. 7) and FT-IR (Fig. 8) spectra of the spindles were almost identical with the spectrum of crab chitin. The close resemblance of the absorption shape between 1600 and 1000 cm–1 suggests a comparable degree of N-acetylation for both samples (Brugnerotto et al., 2001).
The three different chitinous structures in the head, gut, and skin of eolids show that the chitin-synthesizing cells handle the polysaccharide in different ways. In the radula and the cuticles, chitin is secreted and deposited extracellularly, whereas the vesicles containing the spindles remain in the synthesizing cells (Porter and Rivera 1980). At the junction of the esophageal epithelium and the stomach epithelium, there is a distinct transition between the two modes of chitin expression. Chitin is secreted and deposited as a cuticle in the esophagus and then appears as intracellular inclusions of granular chitin in the stomach epithelial cells, a change that occurs abruptly from one cell to the next. One may hypothesize that intracellular trafficking of chitin in the secretory pathway is truncated in stomach epithelial cells so as to block secretion of the polymeric polysaccharide. The only other report of granular intracellular chitin that we are aware of concerns cytoplasmic granules in granulocytes of prawns and lobsters (G. G. Martin et al., 2003). These granules were thought to prevent dissemination of pathogens, when foreign material is being phagocytosed/encapsulated.
Spindles apparently serve a protective function against the deleterious effects of nematocysts fired by prey and co-habiting Cnidaria (Graham, 1938; Edmunds, 1966; Martin and Walther, 2003; Martin et al., 2007). In the eolids Flabellina affinis and Cratena peregrina, as shown here and previously (Martin and Walther, 2002), when cnidophores of the hydroid Eudendrium racemosum come in contact with the cerata of the slugs, they elicit a discharge of masses of nematocysts from the cnidophores. After local lysis of the slug's skin, masses of spindles are released and form aggregates with the everted tubules of the nematocysts (Martin and Walther, 2003). The holotrichous isorhiza nematocysts of Eudendrium racemosum cnidophores do not have a hard, mechanically penetrating, proximal apparatus of stylets; instead, they have long flexible tubules covered regularly with short spines. Thus, in the absence of stiff mechanical devices, the disruptive effects of the nematocyst tubules on the slug's skin appear to be due to the action of toxins. It is possible that the lytic effect of the toxin is caused by a phospholipase A (Weber et al., 1987). When the masses of spindles liberated from the epidermal cells engage the nematocyst tubules, the meshwork of tubules detach as a whole matrix from the cerata. Macroscopically, the cnidophores first stick to the cerata, and then detach from them some minutes later. Thus, the vital tissues below the basal lamina are preserved. The free spindles attached to the everted nematocyst tubules are reminiscent, for example, of the scales of butterfly wings, which are released on contact with the sticky silk of a spider web, liberating the butterfly.
In nudibranchs, cells filled with spindles were found especially in surface areas exposed to nematocysts, such as the cerata, the rhinophores, the lips, or the non-retractile gill-like organs (Martin et al., 2007). The spindles in the stomach epithelial cells (Henneguy, 1925; Graham, 1938; Martin et al., 2007), presumably also protect the stomach epithelium against nematocysts. There is evidence for discharge of nematocysts from a food source in the stomach lumen (Harris, 1973; Martin and Walther, 2002). The chitin granules in the stomach epithelial cells, unlike a cuticular plating, do not prevent peristaltic movements, expansion or relaxation of the stomach.
Different defensive strategies have been shown to be operative in the Nudibranchia. Especially in the dorids (Nudibranchia, Doridacea), there are hard and rough protective surfaces with spicules (Kress, 1981), alone or in combination with repugnant and toxic skin secretions of substances extracted from the food and either modified or synthesized de novo (Cimino and Ghiselin, 1999). Mucus functioning as a protective shield—or even species-specific mucous compounds employed as agents that inhibit discharge of nematocysts of the specific prey—were shown to protect eolids when feeding on sea anemones (Mauch and Elliott, 1997; Greenwood et al., 2004).
Cushion-like or sandbag-like cells filled with chitin granules in the skin and stomach epithelium seem to offer many advantages compared to the rigid chitin exoskeleton of arthropods. There is no need for periodic molts, as in the case of the exoskeleton of arthropods. Damage in the skin of eolids is repaired by cell proliferation in a fast, locally circumscribed regeneration process (Martin and Walther, 2003). While the slugs without shells are able to attack Cnidaria and devour their tentacles—indeed, even those of a Portuguese-man-of-war siphonophore (Thompson and Bennett, 1969)—they also elegantly creep, move, bend, and swim with a flexible skin, waving their rhinophores and cerata on contact with their prey's tentacles. Their specialized skin enabled them to invade an aversive and highly toxic biotic niche with abundant food.
|
Acknowledgments
|
|---|
We thank Drs. Claus Valentin and Iris Schmidt at the Giglio Institute for Marine Biology for the excellent working conditions, and Prof. E. Koenig (University at Buffalo) for critically reviewing the manuscript. We thank Dipl. Chem. Andrea Fiedler for the FT-IR spectra.
|
Footnotes
|
|---|
Received 14 March 2007; accepted 31 July 2007.
Current address: Akademie für medizinische Berufe, Universitätsklinikum, Schlossstr. 38, D 89079 Ulm, Germany. 
|
Literature Cited
|
|---|
Brugnerotto, J., J. Lizardi, F. M. Goycoolea, W. Argüelles-Monal, J. Desbrières, and M. Rinaudo. 2001. An infrared investigation in relation with chitin and chitosan characterization. Polymer 42:3569–3580.[ISI]
Cimino, G., and M. T. Ghiselin. 1999. Chemical defense and evolutionary trends in biosynthetic capacity among dorid nudibranchs (Mollusca: Gastropoda: Nudibranchia). Chemoecology 9:187–207.
Edmunds, M. 1966. Protective mechanisms in the Eolidacea (Mollusca Nudibranchia). J. Linn. Soc. Lond. Zool. 47:27–71.
Galat, A., and J. Popowicz. 1978. Study of the Raman scattering spectra of chitins. Bull. Acad. Pol. Sci. Sci. Ser. Biol. II 26:519–524.
Graham, A. 1938. The structure and function of the alimentary canal of aeolid molluscs, with a discussion on their nematocysts. Trans. R. Soc. Edinb. 59:267–307.
Greenwood, P. G., K. Gary, A. Hunter, and M. Jennings. 2004. Adaptable defense: a nudibranch mucus inhibits nematocyst discharge and changes with prey type. Biol. Bull.206:113–120.[Abstract/Free Full Text]
Harris, L. G. 1973. Nudibranch associations. Pp. 219–315 in Current Topics in Comparative Pathobiology, Vol. 2, T. C. Cheng, ed. Academic Press, Baltimore.
Henneguy, L. F. 1925. Contribution à l'histologie des nudibranchs. Arch. Anat. Microsc. 21:400–468.
Kress, A. 1981. A scanning electron microscope study of notum structures in some dorid nudibranchs (Gastropoda: Opisthobranchia). J. Mar. Biol. Assoc. UK61:177–191.
Martin, G. G., C. Castro, N. Moy, and N. Rubin. 2003. N-acetyl-D-glucosamine in crustacean hemocytes: possible functions and usefulness in hemocyte classification. Invertebr. Biol. 122:265–270.
Martin, R., and P. Walther. 2002. Effects of discharging nematocysts when an eolid nudibranch feeds on a hydroid. J. Mar. Biol. Assoc. UK 82:455–462.
Martin, R., and P. Walther. 2003. Protective mechanisms against the action of nematocysts in the epidermis of Cratena peregrina and Flabellina affinis (Gastropoda, Nudibranchia). Zoomorphology 122:25–35.[ISI]
Martin, R., K.-H. Tomaschko, and P. Walther. 2007. Protective skin structures in shell-less marine gastropods. Mar. Biol. 150:807–817.
Mauch, S., and J. Elliott. 1997. Protection of the nudibranch Aeolidia papillosa from nematocyst discharge of the sea anemone Anthopleura elegantissima. Veliger 40:148–151.
Mayor, H. D., J. C. Hampton, and B. Rosario. 1961. A simple method for removing the resin from epoxy-embedded tissue. J. Cell Biol. 9:909–910.[Free Full Text]
Müller, M., and H. Moor. 1984. Cryofixation of suspensions and tissues by propane jet freezing and high pressure freezing. Pp. 6–9 in Proceedings of the 42nd Annual Meeting of the Electron Microscopy Society of America, M. Bailey, ed. San Francisco Press, San Francisco.
Pearson, F. G., R. H. Marchessault, and C. Y. Liang. 1960. Infrared spectra of crystalline polysaccharides. J. Polym. Sci.43:101–116.
Peters, W. 1972. Occurrence of chitin in Mollusca. Comp. Biochem. Physiol. 41B: 541–550.
Peters, W., and I. Latka. 1986. Electron microscopic localization of chitin using colloidal gold labelled with wheat germ agglutinin. Histochemistry 84: 155–160.
Porter, K. J., and E. R. Rivera. 1980. The Golgi apparatus in epidermal mucoid and ellipsoid-vacuolate cells of Aeolidia papillosa and Coryphella rufibranchialis (Nudibranchia). Protoplasma 102:217–233.[ISI]
Salwini-Plawen, L. von, and H. Nopp. 1974. Chitin bei Caudofoveata (Mollusca) and die Ableitung ihres Radulaapparates. Z. Morphol. Tiere 77:77–86.
Schmekel, L., and W. Wechsler. 1967. Elektronenmikroskopische Untersuchungen über Struktur und Entwicklung der Epidermis von Trinchesia granosa (Gastropoda, Opisthobranchia). Z. Zellforsch. 77:95–114.[ISI][Medline]
Spindler-Barth, M., and U. Buss. 1997. ELISA for determination of chitin and chitosan. Pp. 9–13 in Chitin Handbook, R. A. A. Muzzarelli and M. G. Peter, eds. Atec, Grottammare, Italy.
Studer, D., M. Michel, and M. Müller. 1989. High pressure freezing comes of age. Scanning Microsc. Suppl. 3:253–269.[Medline]
Thompson, T. E., and I. Bennett. 1969. Physalia nematocysts: utilized by molluscs for defense. Science166:1532–1533.
Trinchese, S. 1881. Aeolididae e famiglie affini. Atti Real. Accad. Lincei Mem. Classe Sci. Fis. Mat. Nat. 10:1–142.
Weber, J., M. Klug, and P. Tardent. 1987. Some physical and chemical properties of purified nematocysts of Hydra attenuata Pall. (Hydrozoa, Cnidaria). Comp. Biochem. Physiol. 88B:855–862.
Yamaguchi, Y., T. T. Nge, A. Takemura, N. Hori, and H. Ono. 2005. Characterization of uniaxially aligned chitin film by FT-IR spectroscopy. Biomacromolecules 6:1941–1947.[ISI][Medline]
TOP
Abstract
Introduction
