Enteroendocrine profile of α-transducin immunoreactive cells in the gastrointestinal tract of the European sea bass (Dicentrarchus labrax)

Abstract

In vertebrates, chemosensitivity of nutrients occurs through activation of taste receptors coupled with G protein subunits, including α-transducin (G_αtran) and α-gustducin (G_αgust). This study was aimed at characterizing the cells expressing G_αtran-immunoreactivity throughout the mucosa of the sea bass gastrointestinal tract. G_αtran immunoreactive cells were mainly found in the stomach, and a lower number of immunopositive cells were detected in the intestine. Some G_αtran immunoreactive cells in the stomach contained G_αgust immunoreactivity. Gastric G_αtran immunoreactive cells co-expressed ghrelin, obestatin and 5-hydroxytryptamine immunoreactivity. In contrast, G_αtran immunopositive cells did not contain somatostatin, gastrin/cholecystokinin, glucagon-like peptide-1, substance P, or calcitonin gene-related peptide immunoreactivity in any investigated segments of the sea bass gastrointestinal tract. Specificity of G_αtran and G_αgust antisera was determined by Western blot analysis, which identified two bands at the theoretical molecular weight of ~45 and ~40 kDa, respectively, in sea bass gut tissue as well as in positive tissue, and by immunoblocking with the respective peptide, which prevented immunostaining. The results of the present study provide a molecular and morphological basis for a role of taste related molecules in chemosensing in the sea bass gastrointestinal tract.

Introduction

The gustatory system plays a dominant role in the detection of dietary nutrients, sodium content and the acidity of foods as well as sensing the presence of potentially harmful substances (Sternini 2007; Behrens and Meyerhof 2011). The sense of taste enables animals to adapt to specific habitats (Oike et al. 2007; Ishimaru 2009; Barreiro-Iglesias et al. 2010). In vertebrates (Chandrashekar et al. 2000; Nelson et al. 2001, 2002; Zhao et al. 2003; Behrens and Meyerhof 2011), including fish (Ishimaru et al. 2005), two families of taste receptors (TRs), T1R and T2R, which detect complex tastes, have been cloned. TRs are G-protein-coupled receptors activated by different stimuli, including sweet and bitter substances, amino acids and nucleotides, which elicit a cascade of intracellular signals (Behrens and Meyerhof 2011). In mammals, TRs are abundantly expressed in taste buds, and interact with specific G α-subunits, including α-gustducin (G_αgust), which transmit gustatory signalling from the lingual epithelium to the sensory cortex in the brain (Ming et al. 1999; Margolskee 2002; Caicedo et al. 2003; Behrens and Meyerhof 2011). In addition to G_αgust, several G protein subunits have been identified, which are associated with TR signaling, Gα_i-2, Gα_i-3, Gα ₁₄, Gα ₁₅, Gα _q, Gα _s, and α-transducin (G_αtran) (Ruiz-Avila et al. 1995; Kusakabe et al. 1998). TRs and signalling molecules have been reported in the human and rodent gastrointestinal mucosa and pancreas (Höfer et al. 1996; Höfer and Drenckhahn 1998; Wu et al. 2002; Rozengurt et al. 2006), supporting the concept that there is more than a “taste” function for these molecules and that taste receptor-related molecules exert non-gustatory functions outside the mouth ( Sternini 2007; Behrens and Meyerhof 2011). G_αgust and G_αtran immunoreactivities have been localised to epithelial, predominantly endocrine, cells of the stomach and intestine of rodents (Höfer et al. 1996; Wu et al. 2002; Hass et al. 2007; Sternini 2007; Sutherland et al. 2007), pigs (Clavenzani et al. 2009; Mazzoni et al. 2013) and humans (Rozengurt et al. 2006; Steinert et al. 2011), including ghrelin, somatostatin, cholecystokinin, glucagon-like peptide-1 and peptide YY positive cells (Rozengurt et al. 2006; Sutherland et al. 2007; Clavenzani et al. 2009; Fujita et al. 2009; Moran et al. 2010; Janssen et al. 2011; Steinert et al. 2011; Mazzoni et al. 2013). Endocrine cells, which are distributed throughout the gastrointestinal tract (GIT) mucosa and pancreas, control digestive functions and contribute to regulate caloric intake and metabolism (Holmgren 1985; Plisetskaya and Mommsen 1996; Palmer and Greenwood-Van Meerveld 2001; Nelson and Sheridan 2006). Since fish taste buds express similar receptors and downstream signalling molecules as mammals (Yasuoka and Abe 2009), the aim of this study was to test whether the TR gustatory signalling protein, G_αtran is expressed in the sea bass gut and characterize the types of G_αtran immunoreactive (-IR) cells.

Materials and Methods

Tissue preparation

Nine, non-sexed, 1 year-old European sea bass (Dicentrarchus labrax) were sampled from three tanks at the Laboratory of Aquaculture, Department of Veterinary Medical Science, University of Bologna, Cesenatico, Italy. The average weight and total length of the individuals were 234 ± 26 g and 26 ± 1 cm, respectively. Sea bass were sacrificed by anaesthetic overdose and segments of the GIT were harvested. The stomach, pyloric caeca and the intestine were isolated from each fish; the intestine was divided into cranial, middle and caudal segments. Tissues were either frozen for Western blot assay or fixed in 4% paraformaldehyde in phosphate buffer (0.1 M, pH 7.2) for 48 h at 4°C for immunohistochemistry. Fixed tissues were dehydrated in a graded series of ethanol and embedded in paraffin. Sections of 5 μm thickness were obtained and mounted on poly-L-lysine coated slides, and processed for immunohistochemistry.

Immunohistochemistry

Sections were processed for single and double labelling immunofluorescence. The following primary antisera (see details in Table 1) were used: G_αtran, G_αgust, ghrelin (GHR), 5-hydroxytryptamine (5-HT), obestatin (OB), somatostatin (SOM), gastrin/cholecystokinin (GAS/CCK), glucagon-like peptide-1 (GLP-1), calcitonin gene-related peptide (CGRP), and substance P (SP). Sections were deparaffinized, rehydrated and incubated in a humid chamber at room temperature with appropriate normal serum followed by the primary antibodies (2 days, at 4°C) and the appropriate secondary antibodies (1 hour at room temperature). For double labelling using antibodies raised in different species, sections were incubated with a mixture of primary antisera (e.g. G_αtran and SP or GAS/CCK) and immunoreactivities were visualized with secondary antibodies labelled with different fluorophores. Because the antibodies to G_αgust, SOM, OB, 5-HT, GLP-1 and CGRP were produced in the same species as the G_αtran antiserum, we utilised the procedure and appropriate specificity controls previously described by Takechi et al. (2008) to visualize more than one antigen. Sections were examined using a Zeiss Axioplan microscope and the images were recorded with a Polaroid DMC digital photocamera (Polaroid, Cambridge, MA, USA).

Table 1.

List of antibodies used in this study.

Primary antisera	Antigens	Code	Dilution	Supplier
Polyclonal rabbit anti-α-transducin	Gαt2 of bovine origin	sc-390	1:50	Santa Cruz
Polyclonal rabbit anti-α-gustducin	Gαgust of rat origin	sc-395	1:200	Santa Cruz
Polyclonal rabbit anti-somatostatin	Synthetic somatostatin15–28 conjugated to keyhole limpet hemocyanin	566	1:1000	INCSTAR
Monoclonal mouse anti-gastrin/cholecystokinin	Human gastrin/CCK C-terminus	GAS/CCK9303	1:1000	CURE Digestive Diseases Research Center, UCLA
Polyclonal goat anti-ghrelin	Human ghrelin	sc-10368	1:800	Santa Cruz
Polyclonal rabbit anti-obestatin	Human obestatin	—	1:1500	Prof. Rindi G.**
Polyclonal rabbit anti-5-hydroxytryptamine	Rat hypothalamus, raphe and spinal cord	20080	1:2000	INCSTAR
Polyclonal rabbit anti-glucagon-like peptide-1	Synthetic peptide: HDEFERHAEGTFTSDVSSY, corresponding to amino acids 92–110 of Human GLP 1.	Ab22625	1:1000	Abcam
Polyclonal rabbit anti-calcitonin gene-related peptide	Synthetic rat CGRP	IHC 6006	1:1000	Peninsula/Bachem
Monoclonal rat anti-substance P	Substance-P-BSA conjugate	10-S15	1:300	Fitzgerald
Secondary antisera
FITC-conjugated goat anti-rabbit IgG			1:600	Calbiochem
TRITC-conjugated donkey anti-goat IgG			1:400	Jackson
Alexa 594-conjugated goat anti-mouse IgG			1:400	Mol. Probes
Alexa 594-conjugated donkey anti-rat IgG			1:500	Mol. Probes
Biotin-conjugated goat anti-rabbit IgG			1:400	Vector
Texas Red®-conjugated streptavidin			1:2000	Vector

^**Kindly provided by Prof. Guido Rindi, Institute of Pathology, Catholic University of the Sacred Heart, Rome, Italy.

Antibody specificity

Specificity of G_αtran, G_αgust, GAS/CCK, GHR, OB, and GLP-1 antibodies has been assessed by Western blot and/or immunoblocking with the corresponding peptide (see details in Table 2). Specificity of the 5-HT, CGRP and SOM antibodies was previously demonstrated in the sea bass by preadsorption test (De Girolamo et al. 1999; Visus et al., 1996). The staining we obtained with the monoclonal SP antibody was completely overlapping with the immunostaining reported by Pederzoli et al. (2004) in the sea bass with a rabbit SP antibody (Cambridge Research Biochemical, U.K.), the specificity of which was verified by immunoblocking. Specificity of the secondary antibodies was assessed by omitting the primary antibodies.

Table 2.

Peptides used for absorption tests.

Peptide	Code	Concentration	Supplier
α-transducin	sc-390 P	20 μg peptide in 1 ml PBS	Santa Cruz Biotechnology, Santa Cruz, USA
α-gustducin	sc-395 P	20 μg peptide in 1 ml PBS	Santa Cruz Biotechnology, Santa Cruz, USA
Ghrelin*	sc-10368 P	20 μg peptide in 1 ml PBS	Santa Cruz Biotechnology, Santa Cruz, USA
Glucagon-likepeptide-1	ab50245	10−5 M	Abcam, Cambridge, UK

^*Ghrelin peptide was used for both anti-ghrelin and anti-obestatin antibody specificity, since obestatin belongs to the ghrelin peptide family.

Western blot

Sea bass brain, eye and stomach and mouse brain were collected, frozen in liquid nitrogen, and stored at −80°C. Tissues were homogenized into a sodium dodecyl sulfate (SDS) lysis solution (Tris-HCl 62.5 mM, pH 6.8; SDS 2%, 5% glycerol) with 0.1 mM phenylmethylsulfonylfluoride. Protein content of cellular lysates was determined by a Protein Assay Kit (TP0300; Sigma-Aldrich, St. Louis, MO). For Western blot using G_αgust and G_αtrans antibodies, 20 μg of proteins were separated on NuPage 4–12% bis-Tris Gel (Gibco-Invitrogen, Paisley, UK) for 50 minutes at 200V, then electrophoretically transferred onto a nitrocellulose membrane. For Western blot with the GAS/CCK antibody, 30 μg of proteins were separated on Novex 18% Tris-Glycine Gel (Gibco-Invitrogen, Paisley, UK) for 90 minutes at 125V, then electrophoretically transferred onto a nitrocellulose membrane. After blocking treatment, the membranes were incubated at 4°C overnight with anti-G_αgust (1:300), anti-G_αtrans (1:500) or anti-GAS/CCK antibody (1:1,000) in Tris-buffered saline-T20 (TBS-T20 20 mM Tris-HCl, pH 7.4, 500 mM NaCl, 0.1% T-20). Membranes were then washed with PBS-T20, and incubated with the secondary biotin-conjugated antibody and an anti-biotin horseradish peroxidase linked antibody (1:1,000). The blots were developed using chemiluminescent substrate (Super Signal West Pico Chemiluminescent Substrate, Pierce Biotechnology, Rockford, IL) according to the manufacturer’s instructions. The intensity of luminescent signal of the resulting bands was acquired by Fluor-STM Multimager using the Quantity One Software (Bio-Rad Laboratories, Hercules, CA).

Results

Antibody Specificity

Western blot analysis showed a major band at ~45 kDa in extracts from the sea bass gastric mucosa, brain and eye with the G_αtran antibody (theoretical molecular weight in human) and a unique band at ~40 kDa in extracts from the sea bass stomach and brain, and mouse brain (Fig. 1a, b) with the G_αgust antibody (theoretical molecular weight in human).

Fig. 1.

Western blot analysis showing α-gustducin (a), α-transducin (b) and cholecystokinin (c) immunoreactive bands in sea bass tissue extract. (a): α-gustducin antibody detects a single immunoreactive band near the theoretical molecular weight ~40 kDa in sea bass brain and gastric mucosa (lanes 1–2 respectively) and in mouse brain (lane 3); sea bass and mouse brain served as positive controls. (b): α-transducin antibody detects a major immunoreactive band at the theoretical molecular weight ~45 kDa in sea bass gastric mucosa, brain and eye (lanes 1, 2 and 3 respectively); the brain and eye served as positive control. (c): cholecystokinin monoclonal antibody visualizes a weak, single immunoreactive band close to the theoretical molecular weight of ~15kDa in sea bass intestinal mucosa

Different molecular forms of CCK have been described deriving from enzymatic cleavage of a precursor peptide of 115 AA (UNIPROT P06307) so the expected molecular weight of CCK is between 4 and 12 kDa. In our blot analysis, we identified a faint band near the theoretical molecular weight of ~15 kDa (Fig. 1c). We were unable to identify the smallest form probably because of the very low amount of each component present in the tissue.

Preadsorption of G_αgust, G_αtrans, GLP-1, OB and GHR antisera prevented immunostaining with each antiserum (not shown) confirming tissue staining specificity. The lack of immunostaining of sea bass retina incubated with the anti-G_αgust antibody (not shown) confirms that the G_αgust antibody used in this study does not recognize rod or cones transducins and provides additional support to the tissue specificity of this antibody.

Distribution of G_αtrans cells in the sea bass gut

In the stomach, G_αtran-IR cells were counted in 54 randomly selected fields (0.28 mm² each) with a 40X objective lens, for a total area of 15.1 mm². Since the intestinal mucosa differs morphologically from the stomach for the presence of folds, the number of G_αtran-IR cells in the intestine was evaluated in 200 randomly selected folds for a more accurate representation of cell density in these regions of the GIT. Values were expressed as mean ± standard error mean (SEM). The GIT of the sea bass consists of a siphonal stomach, numerous pyloric caeca and a relatively short intestine. G_αtran-IR cells were detected in the stomach and intestine, but not in the pyloric caeca. Intense immunolabelling was observed in the basal portion of the gastric gland and in the epithelial lining of the intestinal mucosal folds. G_αtran-IR cells showed homogenously labelled cytoplasm, with an unlabelled nucleus and an elongated (“bottlelike”) shape (Fig. 2a, e). These cells were characterised by two thin cytoplasmic processes, the first extending up to the endoluminal surface of the mucosa and the second projecting down to the basal lamina. These features indicate that these cells correspond to “open-type” enteroendocrine cells (EECs) (Höfer et al. 1999; Sternini 2008). The average number of G_αtran-IR cells in the stomach (Fig. 2a, c, e, g) was 15.7 ± 2.2, while in the cranial and middle-caudal portions of the intestine there were 3–4 IR cells/200 folds and 1–2 IR cells/200 folds, respectively. Furthermore, in the stomach, most G_αtran-IR cells co-expressed 5-HT (Fig. 2a, b), OB (Fig. 2c, d) or GHR (Fig. 2e, f). Only a few G_αtran-IR cells colocalized with 5-HT in the intestine. Colocalization between G_αtran- and G_αgust-IRs was observed in some cells in the gastric mucosa (Fig. 2g, h). In contrast, none of the G_αtran-IR cells contained SOM (Fig. 3g, h), GAS/CCK, CGRP, SP, and GLP-1-IR in the stomach and intestine. SOM, SP, CGRP, GAS/CCK, (Fig. 3a, b, c, d) and GLP-1 labelled cells were observed intermingled with unlabelled epithelial cells in the GIT mucosa. In addition, CGRP (Fig. 3c, e), GAS/CCK (Fig. 3f), SOM and SP antibodies labelled nerve fibres running either singly or in small fascicles in the submucosal and muscular layers, with some GAS/CCK and CGRP positive neuronal cell bodies detected only in the muscular layer (Fig. 3e, f).

Fig. 2.

Representative images of sea bass gastric mucosa. α-transducin-immunoreactivity (G_αtran-IR) (a, c, e and g) colocalised with 5-hydroxytryptamine (5-HT) (b), obestatin (OB) (d), ghrelin (GHR) (f) and α-gustducin (G_αgust) (h) (arrows) immunoreactivity in the basal portion of the gastric gland and in the epithelial lining of the mucosal folds. Arrowheads in a, b, c and d indicate 5-HT- and OB-IR (b and d) cells negative for G_αtran (a and c). a–f scale bars = 30μm; g-h scale bars: 20μm

Fig. 3.

Representative images illustrating different subpopulations of EEC cells in sea bass gastric and intestinal mucosa. Somatostatin (SOM), substance P (SP), calcitonin gene-related peptide (CGRP), cholecystokinin (CCK) and α-transducin (G_αtran) (a–h) labelled cells in the stomach (a, arrows) and intestinal mucosa (b, c, d, g, arrows and h, arrowhead), respectively. CGRP and CCK immunolabelled nerve fibers (arrowheads) were observed in the stomach (e) and in the submucosal layer of the intestine (c and f). Some labelled CGRP (e) and CCK (f) neuronal cell bodies (arrows) were identified in the muscular layer. Arrow in g indicates a G_αtran-IR cell negative for SOM (h) in the intestinal mucosa and arrowheads in h point to SOM immunolabelled cells. a, b, e, and h scale bars = 30 μm; c, d scale bars = 60 μm; f scale bars = 20 μm

Discussion

Our data provide evidence for the presence of G_αtran immunolabelled cells in the sea bass GIT and their EEC nature as indicated by the co-expression with GHR, 5-HT or OB, which are markers of distinct subpopulations of EECs. Taste transduction in vertebrates is mediated by specialised receptors organised in groups of cells, which form the taste buds (Chandrashekar et al. 2000; Nelson et al. 2001; 2002; Zhao et al. 2003; Behrens and Meyerhof 2011). The molecular mechanisms through which sweet, L-amino acid (umami), and bitter tastes signal from the tongue to the sensory cortex have been clarified by the discovery of TRs (McLaughlin et al. 1992; Hoon et al. 1999; Lindemann, 2001), which activate Gα-subunits, including G_αtran and G_αgust, to transmit different tastes (Margolskee 2002; Ruiz-Avila 1995; Ming et al. 1999; Caicedo et al. 2003). G_αgust is the best characterized Gα protein associated with bitter taste transmission, however, the findings that G_αgust^−/− mice retain sensitivity to bitter substances, imply that other Gα-subunits, including G_αtran, contribute to signalling bitter substances (Margolskee 2002; Ruiz-Avila et al. 1995; He et al. 2002). The discovery of G_αtran and G_αgust in the GIT of different species from the mouse (Hass et al. 2007; Sutherland et al. 2007; Wu et al. 2002), rat (Höfer et al. 1996) and pig (Clavenzani et al. 2009; Moran et al. 2010; Mazzoni et al. 2013) to human (Rozengurt et al. 2006; Steinert et al. 2011), supports the involvement of TRs in chemosensory mechanisms elicited by luminal contents in different species.

Recent studies on the fish genome failed to detect a gene encoding an ortholog of the mammalian G_αgust gene (Ohmoto et al., 2011), however, Oka and Korsching (2011) showed 80% homology between G_αgust and other G proteins, and Sarwal et al. (1996) reported a high homology of Gα gene between mammals and the puffer fish Fugu rubripes, with an identical intron/exon structure throughout the coding regions. Several lines of evidence support the tissue specificity of the G_αtran and G_αgust antibodies used in our study, including Western blot and immunoblocking experiments. Furthermore, the lack of G_αgust immunostaining in retina sections confirms that the G_αgust antibody does not detect transducins and provides further support of tissue staining specificity. In addition, Zhang et al. (2006) found G_αgust immunofluorescence in the barbells of yellow catfish (Pelteobagrus fulvidraco) with the same rabbit polyclonal G_αgust antibody we used in the present study. However, we cannot exclude the possibility that the gustducin antibody stains others G-alpha proteins in the sea bass gut, thus the term G_αgust immunolabeling should be regarded as “G_αgust like-immunoreactivity”. Finally, it should be pointed out that our transducin antibody does not differentiate between the two molecular forms of this protein, G_αtranS and G_αtranL, as shown by Muradov et al. (2008), who reported the expression of both these transducins in long and short photoreceptors of lamprey (Petromyzon marinus) with the same rabbit polyclonal G_αtran antibody used in our study. The observation that only some cells showed both G_αtran and G_αgust immunoreactivities in the sea bass GIT is in agreement with previous reports in mammals showing only partial colocalization of these two Gα-proteins (unpublished, personal observation) and a differential distribution in some regions (Wu et al. 2002). G_αtran and G_αgust mediate signals initiated by tastants acting at both families of TRs, the T1Rs and the T2Rs (Wong et al. 1996; Ruiz-Avila et al. 2001; He W et al 2002; Caicedo et al, 2003). G_αtran and G_αgust could serve different chemosensitive modalities depending upon the luminal content and according to the receptor subtype being stimulated, which would be consonant with the report that different T2Rs exert their function through the activation in vitro of distinct Gαi related forms (Sainz et al. 2007). Our findings that G_αtran immunoreactivity was localized to distinct subsets of EECs expand previous data in other animals species (Rozengurt et al. 2006; Sutherland et al. 2007; Moran et al. 2010; Janssen et al. 2011, Steinert et al. 2011; Mazzoni et al. 2013). EECs have been reported in the stomach and intestine of several fish species (Holmgren et al. 1982; Reinecke et al. 1997; Ku et al. 2004; Bermúdez et al. 2007; Manning et al. 2008), including the sea bass, where 5-HT-, SOM- and GHR-IR EECs have been described (Visus et al. 1996; Ferrando et al. 2009a; Terova et al. 2008). Our study has shown GHR-IR cells in the gastric mucosa, many of which contain G_αtran-IR. Our results are consistent with data from Terova et al. (2008), who observed high levels of GHR gene expression in the sea bass stomach. The colocalization of G_αtran and GHR-IRs is in line with data reported by Janssen et al. (2011) showing that 98% of the G_αtran-IR cells co-express GHR in mouse stomach, a finding of special interest as it indicates that this cell immunophenotype is conserved through evolution. GHR might have a role in regulating food intake in response to fasting and re-feeding in sea bass (Terova et al. 2008). Moreover, OB, an anorexigenic peptide derived from the GHR precursor, has been reported to counteract GHR effects on energy homeostasis and gastrointestinal function (Zhang et al. 2005). Furthermore, an OB encoding sequence has been recently identified in the black sea bream (Yeung et al. 2006) and in Atlantic halibut (Manning et al. 2008). Thus, our results, together with previous observations, suggest a morphological link between chemical sensing and food intake.

Our findings that the GIT mucosa of the sea bass contains GAS/CCK, OB, 5-HT, CGRP, SOM, GLP-1 and SP immunoreactive EECs extend previous knowledge on the distribution of bioactive messengers in the fish alimentary tract (Elbal et al. 1988; Beorlegui et al. 1992; Barrenechea et al. 1994; Groff and Youson 1997; Reinecke et al. 1997; Al-Mahrouki and Youson 1998; Domeneghini et al. 2000; Bosi et al. 2004; Ku et al. 2004; Pederzoli et al. 2004; Bosi et al. 2005a, b; Nelson and Sheridan 2006; Bermúdez et al. 2007). CGRP-, GAS/CCK-, SP- and SOM-IRs were also identified in nerve processes in the submucosal and muscular layer, and labelled neuronal cell bodies were observed within the muscular layer. Some studies have detailed peptide- and serotonin-containing innervation in fish. Bermúdez et al. (2007) have demonstrated 5-HT-, SP- and CGRP-, but not CCK-IR nerve fibers in the submucosal and muscular layers in the turbot Scophthalmus maximus; similar results were obtained by Bosi et al. (2005a) in chubs Leuciscus cephalus. Moreover, Pederzoli et al. (2004) observed SP-IR neurons in sea bass GIT. The presence of a peptidergic and serotonergic neural network, in addition to the EECs expressing the same signalling molecules, provides support for a link between chemosensory and neuronal systems, which could control GIT physiology in fish via integrated neuro-endocrine mechanisms.

In conclusion, our study demonstrates that G proteins involved in chemosensory transmission are expressed in the sea bass GIT enteroendocrine system. Nutrients may elicit the release of different bioactive messengers (mainly peptides), which directly, or via neural reflexes, contribute to the control of GIT functions and nutrient intake of this fish. Taste-related molecules might represent the initial molecular events involved in chemosensing processes. A better understanding of the mechanisms involved in luminal chemosensitivity in the fish may provide a new basis for feeding formulations to be applied in aquaculture.