The evaluation of intestinal trophism, mainly the mucosal layer, is an important issue in various conditions associated with injury, atrophy, recovery, and healing of the gut. The integrity of the mucosal barrier and the speed of either proliferation or apoptosis are other common variables of relevance for many investigative studies.
To make these investigations objective and measurable, cellular turnover or morphometric parameters such as villus height, crypt depth, and enterocyte height are frequently used. Focusing on fish, histological analysis of the digestive system is considered a good indicator of the nutritional status of health.
The intestine and liver are the most important organs in digestion and the absorption of nutrients from food and, therefore, the monitoring of these organs is considered necessary. Intestinal villi of mammals’ and birds’ digestive systems fit the Euclidean shape (a cylinder-like structure) and allow the employment of the classical morphometric parameters. The morphological peculiarities of the fish intestine that distinguish it from the mammalian intestine, including the lack of distinct crypt compartments and the presence of longitudinal folding of the mucosa instead of villi, do not permit the use of this classical approach. The aim of the present study was to evaluate the kinetics of the proliferation and apoptosis of enterocytes by immunohistochemistry and subsequently, to assess the complexity of intestinal mucosa by fractal dimension analysis in Solea solea fed different experimental diets.
Immunohistochemistry
Anti-proliferating cell nuclear antigen (PCNA) antibody was employed for detection of the proliferation rate of enterocytes. Sections were deparaffinised for 30 min in Solvent Plus (Carlo Erba, Reagents S.r.l., Italy) and hydrated in a graded series of alcohols. Subsequently, endogenous peroxidase activity was blocked with 3% hydrogen peroxide in distilled water for 30 min at room temperature. Sections were then treated in citrate buffer (pH 6.0) for 10 min in a microwave oven (750 W) for antigen retrieval. Nonspecific binding was blocked by incubating the sections with 5% normal goat serum and 1% bovine serum albumin (BSA) in PBS buffer (blocking solution) for 1 h at room temperature and then sections were incubated overnight at 4°C with mouse monoclonal PCNA antibody (clone PC10, Cell Signaling Technology, Danvers, MA) diluted 1:4000 in blocking solution. A streptavidin-biotin-peroxidase polymeric complex (Super- Picture kit peroxidise, Zymed® Lab, San Francisco, USA)was used to visualise antibody-antigen interaction and applied on the sections for 10 min at room temperature. The immunohistochemical reaction was developed using a 3,3’-diaminobenzidine (DAB) solution (Sigma-Aldrich Corporate, St. Louis, MO). The slides were counterstained with haematoxylin, dehydrated through alcohol, and cleared in xylene before mounting.
Terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) assay
The TUNEL method was used to detect the apoptosis rate of enterocytes. The ApopTag® Peroxidase In Situ Apoptosis Detection Kit (Millipore Corporation, Billerica, MA, USA) was employed. Sections were deparaffinised for 30 min in Solvent Plus (Carlo Erba Reagents S.r.l., Italy) and hydrated in a graded series of alcohols. Treatment with proteinase K (20 μg/ml) in PBS for 15 min was performed. Subsequently, endogenous peroxidase activity was blocked with 3% hydrogen peroxide in distilled water for 5 min at room temperature. Equilibration Buffer was then applied on the sections for 3 min and finally, they were incubated with Working Strength TdT Enzyme for 1 h at 37°C in a humid chamber. After washing with Stop/Wash Buffer for 10 min, the secondary antibody anti-digoxigenin peroxidase conjugate was applied on the sections for 30 min at room temperature. After washing, the sections were subjected to the same protocol as used for PCNA.
Image analysis and cell counting
From each stained section, photomicrographs of five fields per gastrointestinal tract at 20× magnification were captured using a Nikon Digital Sight SD-MS camera (Nikon Corporation, Japan) connected to an optical microscope. Images were then processed using ImageJ 1.46 software, which is freely downloadable. Acquired photographs of PCNA- and TUNEL-stained slides were in 24 bit jpg format, with image resolution of 2560×1920 pixels.
PCNA-positive nuclei of enterocytes were counted using the plugin color segmentation. The original image is divided into colour channels, which are manually chosen by the operator as DAB-stained positive nuclei, haematoxylin-stained negative nuclei, cytoplasm of enterocytes, and white background. The software automatically calculates the percentage of area occupied by each chosen colour and, specifically, the percentage of area occupied by PCNA-positive (DAB-stained) nuclei.
In order to eliminate false positivity of non-epithelial cells (erythrocytes, macrophages, lymphocytes, and fibroblasts), the lamina propria had been manually erased using the software Adobe Photoshop CS5 (Adobe Systems Inc, San Jose, CA). For apoptotic cell count, the following findings were considered to represent apoptosis: (a) marked condensation of chromatin and cytoplasm (apoptotic cells); (b) cytoplasmic fragments with or without condensed chromatin (apoptotic bodies); and (c) intra- and extracellular chromatin (apoptotic micronuclei). TUNEL-positive cells were counted manually in order to eliminate the nonspecific signals due to the application of this method on intestinal sections, and then the plugin color segmentation was employed only to calculate the percentage of area occupied by the other colours (haematoxylin-stained negative nuclei, cytoplasm of enterocytes, and white background).
All data obtained were used to calculate indexes as follows:
Image analysis and fractal dimension (FD) calculation
Eighty photomicrographs of H&E-stained slides of intestinal tracts were obtained in blinded fashion at 4× magnification using a Nikon Digital Sight SD-MS camera connected to an optical microscope. Images were then processed using ImageJ 1.46 software: manual segmentation and thresholding were performed blindly by a second operator to extract the one-pixel wide outline of the intestinal mucosal interface (outline) (Additional file 1). The FD calculation was performed with the Box Counting method using the specific command in ImageJ and setting boxes of 2, 3, 4, 6, 8, 1216, 32, 64, 128, 256, 512, 1024, and 2048 pixels. Image resolution was 2560×1920 pixels and format was 24 bit TIF for the original acquired images of H&E-stained slides; image resolution was 2560×1920 pixels and format was 8 bit Gif for the acquired photographs of outlines.
Statistical analysis
All data obtained were analysed by the software Statistica 8 (StatSoft Inc., Tulsa, Oklahoma). Normal distribution was tested by Shapiro-Wilk. If normality criteria were met, one way ANOVA with post-hoc Fisher LSD test were run; if normality criteria were not met, Kruskal-Wallis multiple comparison were run. Significance was set as p < 0.05.
Results
Histomorphological evaluation of all intestinal segments did not show signs of degeneration or inflammation. PCNA labelling revealed a brown, intense, and homogeneous nuclear staining of enterocytes mainly in the basal area and along the intestinal folds. TUNEL labelling revealed a brown, intense staining of scattered single condensed nuclei or nuclear fragments.
PCNA index was not normally distributed (n = 108; SW W= 0.96290; p = 0.00418). PCNA index in enterocytes showed a significant decrease in parallel with the increase in MM (n = 108; H= 13.64672; p = 0.0034). With respect to the intestinal tract, PCNA index did not show significant differences (n = 108; H = 4.298455; p = 0.1166). Apoptotic index was not normally distributed (n = 108; SW W= 0.45922; p = 0.00000) and did not show significant differences with respect to the diets (n = 108; H = 3.037620; p = 0.3859). With respect to the intestinal tract, apoptotic index was significantly higher in the posterior segment compared with the anterior and intermediate segments (n = 108; H = 21.21437; p = 0.0000). FD was normally distributed (n = 80; Shapiro- Wilk W= 0.97847; p = 0.19599) and showed a significant decrease with a diet that was high in MM (n = 80; F (3.76) = 3.9997; p = 0.01063)). With respect to the intestinal tract, FD did not show significant differences (n = 80; current effect: F(2.77) = 1.3182; p = 0.27359). Linear regression analysis between apoptotic index (independent variable) and FD (dependent variable) showed a statistically significant inverse relationship (Beta = -0.337225; p = 0.002528). Linear regression analysis between PCNA index (independent variable) and FD (dependent variable) did not show a significant correlation (Beta = 0.164612; p = 0.131582).
Discussion
In the present study, the histomorphological evaluation of all three intestinal segments did not show signs of degeneration or inflammation, suggesting that MM could be considered a valid substitute for FM in the diet of sole. Immunohistochemistry with PCNA antibody and the TUNEL method have been widely used in fish species to demonstrate the renewal of enterocytes in studies ranging from experimental diet to environmental and toxicological assessments. Focusing on the TUNEL method for apoptosis assessment, it has been widely demonstrated that it can produce an intrinsic nonspecific labelling. This does not represent a background staining but a binding with DNA fragments produced by endogenous cytoplasmic endonuclease that is physiologically present, or by antigen retrieval pretreatment. In the present study, the non-specific TUNEL labelling was cytoplasmic, patchy, and hazy, with intense peripheral membranous positivity. Despite these apparent limitations, the TUNEL method coupled to H&E-stained sections enabled us to univocally identify cells undergoing apoptosis.
PCNA index showed a decrease in enterocytes proliferation with a diet with major inclusion of MM (MM75) with respect to the control diet (MM0), supporting the hypothesis of an hyperplastic adaptive response induced by a diet that deviates from the feeding habits of this species (a benthonic organism). On the other hand, apoptotic index was not influenced by the diet, probably because dietetic changes were not sufficient to induce a proapoptotic response compared with xenobiotics or hormonal stimuli.
Moreover, the results obtained demonstrated a relationship between reduced enterocytes proliferation and lessened complexity of mucosal folding, expressed by a lower FD, in the diet with the highest MM content. The biological interpretation of this result can be that diets containing high MM (MM75) are more similar to the natural feeding of sole. On the contrary, the control diet (MM0), which is usually administered in this reared species, is probably far from an optimal formulation; this could have triggered an hyperplastic adaptive response of the intestinal mucosa, supported by a higher proliferation index, with the aim of increasing the absorptive surface. In linear regression analysis, an inverse and significant correlation between FD and apoptosis index emerged, confirming that a higher apoptosis rate corresponded to more simple mucosal folding even if there was no significant difference in apoptosis index in relation to the diet.
This study demonstrated the usefulness of FD calculation as an alternative method to immunohistochemistry for the detection of intestinal trophism; FD permits a more rapid assessment of histological sections, reducing the costs of reagents and time.
September 2014
