The abundant production of mucus by various epithelial tissues of the eel covering both internal and external body surfaces is well known, and has predominantly protective functions. In fish species, digestive mucosubstances, which are predominantly composed of glycoconjugates, are probably implicated in other physiological processes including lubrication, increasing digestive efficiency, promotion of macromolecular absorption, buffering of intestinal fluid, prevention of proteolytic damage to the epithelium and defense against bacteria and other pathogens. In addition, an osmotic function is especially important in fish, in the form of binding and transportation of water and various ions. Thus, fish gastrointestinal epithelial surfaces and their secreted glycoconjugates have a fundamental role in mediating relationships between the external environment and the body.

In the present study, we focussed on epithelial cells that synthesize glycoconjugates in the gastrointestinal tract of the adult European eel (Anguilla anguilla L.), in order to characterize glycoconjugate secretion and identify possible correlations with specific functional roles of the alimentary canal. The histochemical analysis of glycoconjugates was achieved by applying conventional histochemical reactions. In addition, the presence of various sugar residues in the gut glycoconjugates was investigated by the use of a panel of biotinylated lectins. The importance of glucosidic residues for structural and functional properties of glycoprotein moieties is well known.

The present study is one of a series of investigations of glycoconjugates in the alimentary tract of fish, such as Ictalurus melas, Sparus aurata, Acipenser transmontanus, Solea solea and pantex, a hybrid sparid fish. Other authors have described the histochemical nature of mucous cells of the gut of other fish species, highlighting remarkable differences between species.

Ten adult female Anguilla anguilla individuals were obtained in April, and were used in the present study. Body length was about 40 cm (“silver eel” stage). Fish were killed with an overdose of MS222 (Sandoz, Milan, Italy) anaesthesia, and the alimentary tract specimens were collected immediately afterwards. Several samples of the oesophagus (proximal, medium and distal), stomach (cardiac, fundic and pyloric zones) and intestine (proximal, near the pyloric valve, and distal) were fixed for 24 h at 4 °C in one of the following fixatives: (i) 10% neutral buffered formaldehyde; (ii) 10% formol-calcium; (iii) B4G fixative (6% mercuric chloride and 0.1% glutaraldehyde in 1% sodium acetate), according to Kantani-Matsumoto and Kataoka (1989). Samples were then paraffin-embedded after dehydration in a graded series of ethanol. Dewaxed sections (4 ?m-thick) obtained from samples fixed in all the three fixatives were stained with haematoxylin and eosin (H&E) as well as with Azan trichromic stain to show general morphology. Sections obtained from samples fixed in B4G fluid were treated with Lugol’s iodine prior to staining.

Sections obtained from tissue fixed in 10% neutral buffered formaldehyde were stained with:

• diastase/periodic acid Schiff (D/PAS) to demonstrate neutral glycoconjugates and exclude the presence of glycogen.

• Alcian Blue 8GX pH 2.5/periodic acid-Schiff (AB/PAS) staining, to demonstrate both acidic glycoconjugates (blue) and periodate-reactive vicinal diols (purple).

• high iron diamine/Alcian Blue 8GX pH 2.5 (HID/AB) staining, to differentiate sulphated (brownish-black) from carboxylated, nonsulphated (blue stained) acidic glycoconjugates.

• Astra Blue pH 2.5 (Astra Blue 6GLL; Bioptica, Milan, Italy) to visualize the sialylated glycoconjugates (azure stain).

Removal of terminal sialic acids in order to confirm their presence in carboxylated nonsulphated glycoconjugates was performed utilizing neuraminidase (sialidase) with or without previous KOH treatment (saponification). Sections were incubated at 37 °C for 18 h in a 0.8 IU/ml solution of neuraminidase (Neu) from Clostridium perfringens (Sigma, Milan, Italy) in 0.1 M sodium acetate buffer, pH 5.5, containing 10 mM CaCl₂. Saponification was performed by immersing sections in a 0.5% solution of KOH in 70% ethanol for 15 min at room temperature prior to enzymatic digestion. The presence of sialic acid in carboxylated nonsulphated glycoconjugates was determined by loss (or strong reduction) of AB reactivity after KOH-Neu-AB/PAS treatment.

Sections obtained from tissue fixed in 10% formol-calcium were used for further histochemical identification of sialylated glycoconjugates. Glycoconjugates containing sialic acid O-acylated at different carbon positions were identified applying the following staining methods:

• KOH/PAS; saponification with 0.5% KOH in 70% ethanol for 30 min at room temperature was performed to deacetylate sialic acid residues and was followed by PAS staining.

• PBT (periodate-borohydride-technique)/KOH/PAS; PAS staining of preexisting vicinal diols was abolished by periodate oxidation followed by sodium borohydride reduction. KOH/PAS staining was subsequently obtained as a red colour by saponification followed by standard PAS staining.

• PATS (periodic acid-thionin-Schiff)/KOH/PAS to demonstrate PAS-positive glycoconjugates in blue, whereas other KOH/PAS-positive glycoconjugates appear in purple, with combinations being violet.

After hydration, sections were incubated in a 0.3% solution of H₂O₂ in methanol for 30 min at room temperature to inhibit endogenous peroxidase activity. Sections were then washed in 0.1 M phosphate-buffered saline (PBS), pH 7.6, and incubated with biotinylated lectins, at a concentration of 10 ?g/ml in 10 mM HEPES buffer, pH 7.5, for 2 h at room temperature. Sections were then washed again in PBS and subsequently treated with an Avidin–Biotin–peroxidase Complex (ABC; Vector) for 1 h at room temperature. After washing in PBS, sites of lectin binding were visualized by treating sections for 5–6 min at room temperature with a freshly-prepared solution of 0.25 mg/ml 3,3?-diaminobenzidine tetrahydrochloride (DAB; Sigma) and 0.003% H₂O₂ in PBS. Subsequently, sections were rinsed with tap water, cleared and mounted in Eukitt (Bioptica). Negative controls for lectin binding included: (i) incubation in the buffer medium without the respective lectin, (ii) omission of ABC treatment, (iii) incubation in the presence of lectins to which the respective hapten sugars were added at a concentration of 0.2 M. All these controls completely abolished all staining. As positive controls, sections of mammal (bovine, swine) alimentary canals were tested in parallel, and the expected positive results were obtained in mucous cell populations. In addition, as sialic acid is commonly linked to the specific sugar recognized by DBA lectin, a further control was performed to remove sialic acids before staining with DBA lectin. This was performed by incubating adjacent sections for 18 h at 37 °C in a solution of sodium acetate buffer 0.1 M, pH 5.5, containing 0.1 IU/ml sialidase (neuraminidase from Clostridium perfringens; Sigma) and 10 mM CaCl₂ prior to staining with DBA lectin. The removal of sialic acids was confirmed in adjacent sections by the lack of staining with AB.