Localization and distribution of CCK-8-, NPY-, Leu-ENK-, and Ghrelin- in the digestive tract of Prochilodus lineatus (Valenciennes, 1836)

: This study describes the histological characteristics and distribution of gastrointestinal tract endocrine cells (ECs) of Prochilodus lineatus (detritivorous fish) using immunohistochemical procedures. The digestive tract of P. lineatus was divided into seven portions: stomach (cardial and pyloric), pyloric caeca, and intestine (anterior, glandular, middle and posterior). A pool of specific antisera against cholecystokinin (CCK-8), -neuropeptide Y (NPY), -ghrelin (Ghre) and -leu-enkephalin (Leu-ENK) to identify ECs were used. According to the morphological characteristics of ECs, two different types were identified and classified as open or closed-type. The number of ECs varied throughout the gastrointestinal tract, though a high abundance was found in the anterior intestine and pyloric caeca. A large number of ECs immunoreactive to CCK-8 and NPY were recorded in the anterior, glandular and middle intestine. ECs immunopositive to Leu-ENK were distributed in the stomach and pyloric caeca. For Ghre, immunopositive ECs were restricted to the glandular intestine. The results of the present study indicate that P. lineatus presents an ECs distribution pattern with species-specific particularities. However, CCK showed a distribution similar to that of omnivores, which is possibly related to local signaling functions in order to achieve the correct digestion of the various organisms found in the detritus.


INTRODUCTION
The digestive tract (DT) of fish exhibit a diversity of morphological and functional characteristics, varying from short and simple to long and complex (Olsson 2011), and it is fundamentally related to the different environments, diets, and developmental states of the individuals (Angelescu & Gneri 1949, Wilson & Castro 2010. Regardless of these particularities, there exist different types of gastrointestinal neuropeptides synthesized by endocrine cells (ECs) located in the wall and epithelium of the digestive tract (Buddington & Krogdahl 2004, Holmgren & Olsson 2009). ECs are one of the largest endocrine systems in the body that participate in the mechanisms of control of the digestive processes, as well as in the peripheral signaling of food intake and energy homeostasis, similar to that observed in mammals (Lin et al. 2000, Canosa et al. 2005, Volkoff et al. 2005. Several studies demonstrated the occurrence and distribution of ECs through immunohistochemical techniques in the gastrointestinal tract of various fish species (Pan et al. 2000, Bosi et al. 2004, Ku et al. 2004, Çinar et al. 2006, Vigliano et al. 2011, Hernández et al. 2012, Pereira et al. 2015, Lin et al. 2017. Several neuropeptide distribution patterns were observed according to different gastrointestinal tract morphologies and feeding habits, as observed in carnivores (Bosi et al. 2004, Çinar et al. 2006, Pereira et al. 2015, omnivores (Kiliaan et al. 1992, Pan et al. 2000, and herbivores (Ku et al. 2004, Lin et al. 2017. Neuropeptides such as cholecystokinin (CCK), neuropeptide Y (NPY), leu-enkephalin (Leu-ENK), and ghrelin (Ghre) are synthesized by endocrine cells of the DT, and play a key role in nutritional homeostasis regulation. CCK is mainly synthesized in the DT and in the brain (Moran & Kinzig 2004), thereby intervening in both digestion processes and peripheral satiety signaling (anorexigenic) (Rubio et al. 2008, MacDonald & Volkoff 2009, Volkoff 2016. NPY is a peptide that presents a primary structure highly conserved among vertebrates, including fish (Jensen 2001). It also has important functions such as energy metabolism regulation, as well as digestive, reproductive, and immune processes, thus highlighting its important role in the regulation of eating behavior as an orexigenic factor (Volkoff et al. 2005, Volkoff 2006, MacDonald & Volkoff 2009, Zhou et al. 2013. Leu-ENK is found in the DT of different fish species (Pan et al. 2000, Vigliano et al. 2011, Hernández et al. 2012, Lin et al. 2017, and its function would be related to responses to inflammatory processes (Dezfuli et al. 2002(Dezfuli et al. , 2004. Ghrelin is known as an appetite-stimulating intestinal hormone and it is involved in multiple physiological functions, such as the regulation of food intake, growth, and reproduction (Kaiya et al. 2008).
Prochilodus lineatus a species widely distributed in Latin America and of great commercial importance to this region (Espinach Ros & Sánchez 2007). P. lineatus represents great productive potential for fish farming due to its good growth with foods of low protein content (Croux 1992). This species is strictly detritivorous, with several anatomical and physiological adaptations for the efficient collection and digestion of organic detritus (Bowen 1983). Previous studies described the morphological and histophysiological characteristics of the digestive tract, including intestine length, which exceeds several times the body length and exhibits a complex pattern of intestinal loops (Angelescu & Gneri 1949, Barbieri et al. 1998, Barbieri & Hernández-Blazquez 2002, numerous pyloric caeca (Angelescu & Gneri 1949), and glands in the midgut (Domitrovic 1983, Nachi et al. 1998). However, no records have reported to date on the occurrence of endocrine cells in the digestive tract of detritivorous fish.
This study aimed to determine the characteristics and distribution of some neuromodulators of the P. lineatus DT using immunohistochemical techniques. These results can provide useful information to improve our understanding of the relationship between the morphological and functional characteristics of the digestive tract and endocrine signaling mechanisms in a detritivorous species.

MATERIALS AND METHODS
In this study, six healthy adult specimens of P. lineatus without sex distinction (average weight and standard length: 152 ± 18.60 g, 195 ± 11.25 mm, respectively) collected from the Northeast Institute of Ichthyology, Faculty of Veterinary Sciences (Corrientes, Argentina) were used. After euthanasia with an overdose of benzocaine (100 ppm), tissue samples were taken from stomach (cardial and pyloric), pyloric caeca and intestine (anterior, glandular, middle and posterior) (Domitrovic 1983, Barbieri et al. 1998, Nachi et al. 1998). The procedures adopted with the animals in this research were in accordance with the ethical principles of animal experimentation, and approved according to protocol N° 0033 by the Ethics and Biosafety Committee of School of Veterinary Sciences of the Northeast National University (UNNE) of Argentina.

Light microscopy and immunohistochemistry
Samples were fixed in Bouin's solution (12 h) and then embedded in paraffin wax after processing in a graded ethanol series. Microtome sections (1-3 µm thick) were collected on slides pretreated with silane (3-amino-propyltriethoxysilane; Sigma Chemical, St Louis, MO, USA), allowed to dry overnight and then de-waxed and hydrated. To assess digestive structures by light microscopy, sections were then stained with haematoxylin and eosin (H&E). For immunohistochemistry, all incubations were performed in a humid chamber with primary antibody for 16-20 h at 4 °C, and all washing procedures consisted of three successive 5 min immersions in 0.1 M phosphate-buffered saline (PBS; 8 mM Na 2 HPO 4 , 3 mM NaH 2 PO 4 , 150 mM NaCl). Endogenous peroxidase activity was blocked by incubation in peroxidase blocking solution (3% H 2 O 2 in PBS) for 30 min, and after a rinse in PBS, the sections were treated with 3% skim milk powder for 15 min to block non-specific antibody binding. Subsequently, the samples were incubated with the primary antibodies listed in Table I, washed with PBS, and incubated with biotinylated secondary antibody followed by streptavidin peroxidase conjugate (CytoScan TM HRP Detection System, Cell Marque) both at room temperature for 20 min. Finally, the sections were treated with DAB chromogen (3-3' diaminobenzidine), then immersed in deionized water to stop the reaction, counterstained with haematoxylin, dehydrated, and coverslipped. As positive control, sections of pig intestine were used. On the other hand, negative control slides were sections in which the primary antibody was replaced by PBS.
The recorded values were obtained from the total number of endocrine cells, manually counting the cross sections independent of each portion of the DT and for each antibody, and reported as average values from 1000 μm of intestinal perimeter (Hall & Bellwood 1995). The number of endocrine cells of each section was classified into the following grades: no detected (−), 1 ~ 10 cells (+), 11 ~ 20 cells (++), 21 ~ 30 cells (+++); more than 30 cells (++++). The images were obtained using a Leica DM500 microscope with Leica ICC50 digital camera equipped with an image analysis system: Leica Application Suite 3.4. 1.

RESULTS
According to the morphological characteristics, two types of endocrine cells located between the enterocytes of the intestinal epithelium were observed. Open-type ECs exhibit an elongated shape and are wider in the zone occupied by the nucleus, and exhibits a granular content in the supranuclear cytoplasm ( Figure 1). Closedtype ECs are round in shape and located in the basal region of the epithelium (Figure 2). In addition, some nerve cells with an irregular contour defining their typical stellate shape with cytoplasmic projections were observed. The distribution of each EC type exhibited high variation along the digestive tract (Table II).

CCK-immunoreactive endocrine cells
Immunoreactivity to CCK was detected in cells of the epithelial layer of the pyloric caeca and anterior intestine (Figure 1a, b). The highest density of CCK-IR open-type ECs was observed in the pyloric caeca, while the number of ECs decreased from anterior toward the caudal segments of the gut, being absent in posterior intestine (Table II). Notably, no CCK-IR ECs were found in the stomach of P. lineatus.

NPY-immunoreactive endocrine cells
Large amounts of NPY-IR ECs were detected, mainly in the pyloric caeca, but also in the anterior, glandular, and middle portions of the intestine (Table II) no NPY-IR ECs were found in other portions of the DT, such as the stomach and posterior intestine.

Leu-ENK-immunoreactive endocrine cells
In the epithelial layer of the cardial and pyloric stomach, only open-type Leu-ENK-IR ECs were detected, while closed-type endocrine cells were observed in the pyloric caeca (Table  II) (Figure 2a, b). Moreover, immunoreaction to Leu-ENK was observed in nerve cells surrounding the gastric glands (Figure 2c, e), as well as in neurons of the myenteric plexus. Leu-ENK IR neurons presented an irregular outline defining their typical stellated shape with cytoplasmic projections (Figure 2d, f ). Nerve fibers distributed in the lamina propriasubmucosa and muscle layers throughout the DT also showed immunoreactivity to Leu-ENK antisera.

Ghre-immunoreactive endocrine cells
Ghre-IR open-type ECs were restricted to the glandular intestine (Figure 2g), and exhibited a triangular shape with large secretory granules distributed throughout the cell (Figure 2h). These cells were found in the base and middle portion of the glands.

DISCUSSION
The occurrence and distribution of ECs was analyzed in the DT of several fish species, which showed considerable variation in morphology and physiology independent of trophic habit (Vigliano et al. 2011, Pereira et al. 2015, Lin et al. 2017). This could indicate that variations found among species would be the result of adaptations to environmental conditions and nutritional requirements (Olsson et al. 2011). However, several studies attempted to generalize the regional distribution patterns of ECs. In this sense, Rønnestad et al. (2007) proposed a distribution model of CCK ECs in fish larvae related to the macroscopic anatomy of the digestive tract. In this model, species with straight gut would have a distribution pattern of CCK ECs throughout the gut, whereas in larvae with rotated gut, CCK ECs would be highly concentrated in the anterior segment of the intestine. Therefore, in species with rotated gut, CCK-producing cells would exhibit a strategic distribution related to the capture of chemical signals of food coming from the stomach in order to achieve a highly functional feedback control of the digestive process. In P. lineatus (fish with coiled gut), we observed a distribution pattern similar to fish with rotated gut. This was similar to the pattern observed in other species with coiled guts, such as Carassius auratus In the present study, the high density of CCK-IR cells observed in the anterior intestine and pyloric caecum would represent a key location, since this region is strongly related to diffuse acinar pancreatic tissue (Sverlij et al. 1993, Rotta 2003. In this place, ECs would release CCK in response to intraluminal nutrients, consequently stimulating the release of pancreatic digestive enzymes towards the intestinal lumen (Einarsson et al. 1997). In addition, this is valid when considering the existence of a retrograde peristalsis mechanism (Rønnestad et al. 2000) favoring the filling of pyloric caeca with chyme and mixing with the digestive secretions of this region (Gräns & Olsson 2011). Thereby, our results are logical when considering the biological role of CCK on pancreatic enzymes release, gallbladder contraction, gastrointestinal motility stimulation, and gastric emptying inhibition (Volkoff et al. 2005, Gorissen et al. 2006, Rønnestad et al. 2017). In addition, the distribution of CCK ECs in P. lineatus exhibits similarity to omnivorous species. This could be related to the diversity of organisms consumed by the detritivores, where phytoplankton, zooplankton, benthic microand macroflora, necton macroflora, coprogenic material, and organic allochthonous material are the main food source of this species (Gneri & Angelescu 1951, Sverlij et al. 1993.
In fish, NPY is widely distributed in the CNS (Rodríguez-Gómez et al. 2001, Pérez Sirkin et al. 2013, Hosomi et al. 2014) and in the digestive system (Vigliano et al. 2011, Pereira et al. 2015, Lin et al. 2017). In the CNS, hypothalamic neurons are the main site of NPY production, which play a key role in increasing food consumption (López-Patiño et al. 1999, Silverstein & Plisetskaya 2000, Kiris et al. 2007, whereas the DT is the main peripheral NPY producer organ and exerts primarily inhibitory effects on secretion, intestinal motility, and blood flow (Uesaka et al. 1996, Shahbazi et al. 2002, Gomez et al. 2012, and induces immune activation or suppression (Carpio et al. 2007, Farzi et al. 2015. In contrast to CCK, the regional distribution of NPY in DT shows minor variations between species. Several studies mentioned that NPY-IR ECs were primarily observed in the anterior intestine and pyloric caeca of fish, as seen in carnivorous (Al-Mahrouki & Youson 1998, Çinar et al. 2006, Min et al. 2009, Vigliano et al. 2011, Pereira et al. 2015, omnivorous (Al-Mahrouki & Youson 1998, Pereira et al. 2017), or herbivorous (Lin et al. 2017). In the present study, NPY-IR ECs were localized in the epithelial mucosa throughout the intestine, with the exception of the rectal portion. The highest reactivity was observed in the anterior intestine and pyloric caeca, being similar to that observed in most fish species. Moreover, the regional distribution of NPY in the DT of P. lineatus would be related to local signaling functions and peripheral monitoring for appetite hypothalamic regulating center (Vigliano et al. 2011, Babichuk & Volkoff 2013, Pereira et al. 2015, Hernández et al. 2018. Leu-ENK is a pentapeptide associated with the regulation of the inflammatory process, as well as the modulation of intestinal peristalsis (Radulovic et al. 1996). In addition, previous studies reported that fish infected with parasites showed increased immunostaining of Leu-ENK in affected areas, revealing a possible role in the modulation of the inflammatory process (Dezfuli et al. 2002, 2004, Bosi et al. 2005. Previous studies described the occurrence of closed-type or open-type Leu-ENK ECs distributed in the epithelium of the DT, as well as in the neuronal bodies or nerve fibers of the DT wall. In C. chanos (Lin et al. 2017), closed-type Leu-ENK-IR ECs were found in the pyloric caeca, whereas open-type ECs were observed in the anterior intestine region. In Cyprinus carpio, Ctenopharyngodon idellus, Mylopharyngodon piceus (Pan et al. 2000), Odontesthes bonariensis (Vigliano et al. 2011), and R. quelen (Hernández et al. 2012), only open-type Leu-ENK IR ECs were observed, and these were distributed in the epithelial layer of the intestine. In the present study, closed-type Leu-ENK-IR ECs were observed in the stomach and pyloric caeca, as well as in nerve fibers of the myenteric plexus. This distribution could be related to the immunomodulatory action necessary in strategic sites of the digestive tract, and with the peristalsis modulation that would help to displace food through the long intestinal tract characteristic of this species.
The hormone ghrelin is considered an orexigenic factor that is highly conserved among vertebrates (Kaiya et al. 2008 (Ji et al. 2015) was observed in the posterior intestine. In the present study, ghrelin immunoreactivity was only observed in the glandular cells of the glandular intestine. Likely, this distribution could be related to the biological function of ghrelin in appetite regulation. However, the observed variations in ghrelin expression would exhibit species-specific distribution patterns (Feng et al. 2013).
The present study provides valuable information regarding the distribution of different neuromodulators of the digestive tract in P. lineatus, which show some similarity with the distribution pattern of neuropeptides found in omnivorous fish. However, the regional variation observed among the distribution patterns of neuropeptides in species with the same trophic habit is likely the result of environmental changes and physiological adaptations, affecting the dietary behavior of fish in evolutionary terms (Volkoff 2016, Soengas et al. 2018).