Article Images (0)
×![]()
featured_play_listSave article to folder
X
Create new folder
Full Text
PDF
Table of Contents
Abstract
ABSTRACT
1. Introduction
2. Material And Methods
4. Discussion
5. Conclusions
Acknowledgments
Funding
Declaration Of Competing Interest
Declaration Of Competing Interest
Graphical Abstract
Abstract
Visfatin is an adipokine with mediatory effects on inflammation. It is expressed at low levels in the pig stomach, but its role in the gastrointestinal (GI) tract is not well understood. This study explored visfatin expression and localization in the stomach and duodenum of piglets fed varying levels of polyphenols derived from olive mill waste extract, known for their antioxidant and immunomodulatory properties. Twenty-seven piglets were assigned to three dietary groups: control (commercial feed), low polyphenol (120 ppm), and high polyphenol (240 ppm) groups. After 14 days of feeding, samples from the glandular stomach and duodenum were collected from 13 piglets. Immunohistochemistry (IHC), digital image analysis (DIA) using QuPath software, and double-label immunofluorescence were performed to detect visfatin-positive cells and co-localize them with serotonin. Additionally, relative gene expression of visfatin was assessed via RT-qPCR. Visfatin-positive cells were identified in 5 out of 13 piglets, localized mainly in the basal portion of gastric and intestinal glands. The morphology of those cells was consistent with neuroendocrine cells and confirmed by co-localization of visfatin and serotonin. No significant differences were found in cell positivity or morphology between dietary groups or between tissues. However, visfatin transcript levels increased with the dose of polyphenolic extract. These findings suggest that dietary polyphenols may modulate visfatin gene expression in the GI tract. The study also highlights the value of digital anatomy for enhancing the accuracy and reproducibility of anatomical research. Further studies are needed to elucidate the functional role of visfatin transcript and protein in the porcine GI tract.
Journal Pre-proof Morphological digital assessment and transcripts of gastric and duodenal visfatin in growing piglets fed with increasing amounts of polyphenols from olive mill waste extract Daniele Marini, Maria Grazia Cappai, Elisa Palmioli, Gianni Battacone, Margherita Maranesi, Kamil Dobrzyń, Francesca Mercati, Cecilia Dall’Aglio PII: S0940-9602(24)00161-4 DOI: https://doi.org/10.1016/j.aanat.2024.152369 Reference: AANAT152369 To appear in: Annals of Anatomy Received date: 16 September 2024 Revised date: 14 November 2024 Accepted date: 3 December 2024 Please cite this article as: Daniele Marini, Maria Grazia Cappai, Elisa Palmioli, Gianni Battacone, Margherita Maranesi, Kamil Dobrzyń, Francesca Mercati and Cecilia Dall’Aglio, Morphological digital assessment and transcripts of gastric and duodenal visfatin in growing piglets fed with increasing amounts of polyphenols from olive mill waste extract, Annals of Anatomy, (2024) doi:https://doi.org/10.1016/j.aanat.2024.152369 This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2024 The Author(s). Published by Elsevier GmbH. Morphological digital assessment and transcripts of gastric and duodenal visfatin in growing piglets fed with increasing amounts of polyphenols from olive mill waste extract Daniele Marini a,b,*, Maria Grazia Cappai c, Elisa Palmioli a,d, Gianni Battacone e, Margherita Maranesi a,*, Kamil Dobrzyń f, Francesca Mercati a,§, Cecilia Dall’Aglio a,§ a Department of Veterinary Medicine, University of Perugia, Via San Costanzo 4, 06126 Perugia, Italy francesca.mercati@unipg.it; cecilia.dallaglio@unipg.it; b Department of Organismal Biology, Evolutionary Biology Centre, Uppsala University, Norbyvägen 18A, 752 36 Uppsala, Sweden c Department of Veterinary Medicine, University of Sassari, Italy mgcappai@uniss.it d Department of FISSUF, PhD Course in “Ethics of Communication, Scientific Research and Technological Innovation” Medical-Health Curriculum, University of Perugia, Piazza G. Ermini, 1, 06123 Perugia, Italy elisa.palmioli@dottorandi.unipg.it e Department of Agricultural Sciences, University of Sassari, Italy battacon@uniss.it f Faculty of Biology and Biotechnology, Department of Zoology, University of Warmia and Mazury in Olsztyn, Poland kamil.dobrzyn@uwm.edu.pl § These authors share last authorship Corresponding authors: Daniele Marini, Margherita Maranesi; Department of Veterinary Medicine, University of Perugia, Via San Costanzo 4, 06126 Perugia, Italy; daniele.marini@dottorandi.unipg.it; margherita.maranesi@unipg.it KEYWORDS: NAMPT, PBEF1, Adipokines, Digital Anatomy, QuPath software Jo rn al Pr epr oo f
ABSTRACT
Visfatin is an adipokine with mediatory effects on inflammation. It is expressed at low levels in the pig stomach, but its role in the gastrointestinal (GI) tract is not well understood. This study explored visfatin expression and localization in the stomach and duodenum of piglets fed varying levels of polyphenols derived from olive mill waste extract, known for their antioxidant and immunomodulatory properties. Twenty-seven piglets were assigned to three dietary groups: control (commercial feed), low polyphenol (120 ppm), and high polyphenol (240 ppm) groups. After 14 days of feeding, samples from the glandular stomach and duodenum were collected from 13 piglets. Immunohistochemistry (IHC), digital image analysis (DIA) using QuPath software, and double-label immunofluorescence were performed to detect visfatin-positive cells and co-localize them with serotonin. Additionally, relative gene expression of visfatin was assessed via RT-qPCR. Visfatin-positive cells were identified in 5 out of 13 piglets, localized mainly in the basal portion of gastric and intestinal glands. The morphology of those cells was consistent with neuroendocrine cells and confirmed by co-localization of visfatin and serotonin. No significant differences were found in cell positivity or morphology between dietary groups or between tissues. However, visfatin transcript levels increased with the dose of polyphenolic extract. These findings suggest that dietary polyphenols may modulate visfatin gene expression in the GI tract. The study also highlights the value of digital anatomy for enhancing the accuracy and reproducibility of anatomical research. Further studies are needed to elucidate the functional role of visfatin transcript and protein in the porcine GI tract.
1. Introduction
Visfatin is a multifunctional protein with diverse roles in metabolism, inflammation, cellular homeostasis and diseases. Historically, its human gene was originally discovered as PBEF (pre-B-cell colony enhancing factor), which has been reported to be a growth factor for B lymphocyte precursors development (Samal et al. 1994). The gene encoding this human protein is currently known as NAMPT (nicotinamide phosphoribosyltransferase) gene, and the protein NAMPT (P43490 history in UniProt Database; The UniProt Consortium 2022). In 2005, Fukuhara and colleagues (2005) reidentified the NAMPT/PBEF with the name “visfatin” describing it as an adipocytokine secreted by the visceral adipose tissue. Structurally, NAMPT/PBEF/visfatin consists of 491 amino acids with molecular mass of 52-55 kDa in humans, mammalian models and pigs (Chen et al. 2007; The UniProt Consortium 2022). The molecule can exert its pleiotropic activities via an enzymatic or non-enzymatic activity (Revollo et al. 2007; Li et al. 2008; Colombo et al. 2022; Semerena et al. 2023). This protein has been categorized and differently named by its type of activity and localization at sub-organismal or organismal level: (i) the intracellular nicotinamide phosphoribosyltransferase (iNAMPT), and (ii) the extracellular nicotinamide phosphoribosyltransferase (eNAMPT) called also Pre-B cell colony-Enhancing Factor (PBEF) or visfatin. iNAMPT is a regulator of intracellular nicotinamide adenine dinucleotide (NAD) levels, it is found in the cytoplasm, nucleus and mitochondria, and influence cell metabolism, apoptosis and oxidative stress response (reviewed by Garten et al. 2015). When the protein is secreted into the extracellular space, it is called eNAMPT, PBEF or visfatin and has multifaceted functions (reviewed by Semerena et al. 2023): it is a potential ectoenzyme, a proangiogenic factor, a mediator of inflammation and an adipokine. eNAMPT potentially catalyzes extracellular NAD biosynthesis, although its efficacy outside cells remains debated (Semerena et al. 2023). eNAMPT significantly contributes to angiogenesis by stimulating proliferation and migration of capillaries, inducing the production vascular endothelial growth factors and upregulating proteases involved in remodeling of extracellular matrix (Adya et al. 2007; Ezzati-Mobasere et al. 2020). This molecule also takes part in the J ur n l P re -p ro of inflammatory response as pro-inflammatory cytokine, leading to the production of several interleukins and TNFα, and activating inflammatory signaling pathways (Lago et al. 2007; Li et al. 2008; Moschen et al. 2007). At the same time, it can be considered an alarmin or Damage-Associated Molecular Pattern (DAMP) able to recruit immune cells via chemotaxis and promote phagocytosis (Moschen et al. 2007; Colombo et al. 2022). Adipocytokines (or adipokines) are a heterogeneous group of bioactive molecules produced mainly by adipocytes. These molecules are involved in numerous metabolic activities including metabolism, energy homeostasis, inflammatory processes, immune function, control of oxidative stress, reproduction (Fasshauer & Blüher, 2015). Considering the aforementioned points, it becomes evident that eNAMPT/PBEF/visfatin is inherently classified within the adipokines category. It was firstly grouped within the adipokines by Fukuhara et al. (2005), that found visfatin to be secreted by adipocytes in vitro and exerting insulin-mimetic effects lowering the glucose plasma levels in mice. Increase in serum levels of visfatin was observed parallel to visceral fat but not subcutaneous in both mice and humans, and its circulating levels and expression increased under obesity conditions (Fukuhara et al. 2005). While the latter study suggested direct activation of the insulin receptor (Fukuhara et al. 2005), others indicate inconsistencies and lack of reproducibility, leaving the topic still highly debated and uncertain (Fukuhara et al. 2007; Revollo et al. 2007). Eventually, the receptor for visfatin remains unidentified. For clarity, the NAMPT/PBEF/visfatin gene, transcript or protein will hereinafter be referred to simply as “visfatin”. Visfatin seems to be involved in various physiological processes in swine, notably in energy homeostasis, inflammation, and reproduction. In pigs (Sus scrofa domestica), visfatin transcripts and protein were found in all the studied organs, specifically in pituitary, liver, heart, kidney, stomach, lung, uterus, ovary, skeletal muscle, perirenal fat and subcutaneous fat tissues (Palin et al. 2008). This protein is locally produced in the pituitary gland, where its expression influenced by the hormonal environment characteristic of different reproductive states (Szymanska et al. 2023): during the estrous cycle and early pregnancy, its transcripts are expressed in both the adenohypophysis and neurohypophysis. In the porcine anterior pituitary, visfatin protein colocalize with several pituitary tropic hormones and its secretion during the estrous cycle stimulated by gonadotropin-releasing hormone (GnRH), FSH, LH, and insulin (Szymanska et al. 2023). Visfatin may regulate reproductive functions by modulating the anterior pituitary transcriptome during the mid-luteal phase of the oestrous cycle, affecting gene expression and alternative splicing events (Dobrzyn et al. 2024). Additionally, visfatin significantly impacts luteal cells during implantation by modulating the expression of proteins associated with adipogenesis and ovarian turnover, likely being involved in steroid hormones synthesis and metabolism of prostaglandins (Kopij et al. 2024). Moreover, visfatin expression in the corpus luteum depends on the hormonal status related to the phase of the estrous cycle or early pregnancy (Mlyczyńska et al. 2024). Visfatin enhances lipid uptake and lipogenesis in porcine adipocytes by upregulating the expression of lipoprotein lipase, fatty acid synthase, and IL-6 (Yang et al 2010). However, it seems to be not involved in swine fattening (Barbe et al. 2020), and cannot be considered as maker of fat accumulation since the highest visfatin transcripts levels were found to be associated with the leaner pigs (Palin et al. 2008). Instead, visfatin mRNA levels increased in stromal-vascular cells treated with TNFα, indicating that the visfatin response might be modulated by inflammatory markers also in pigs (Palin et al. 2008). In rodent models, parental administration of visfatin has been shown to modulate the inflammatory response in intestinal mucosal tissues under lipopolysaccharide (LPS) stress by enhancing mucosal immunity and integrity — stabilizing tight junction proteins and modulating pro-inflammatory cytokines such as TNF-α, IL-1β, and IL-6 in mice (Pang et al. 2021) and reducing caspase-3 activation, a marker of apoptosis, in intestinal mucosal cells in rats (Zhou et al. 2017). The knowledge on the visfatin transcript presence in the gastrointestinal (GI) tract of the porcine model is limited to the stomach, where it is expressed at relatively low levels (Palin et al. 2008). Information is absent on its presence in the rest of the gastrointestinal (GI) tract, and insights about its role in the digestive system are scarce. Jo ur n l P re -p ro of Polyphenols are plant secondary metabolites that have gathered increasing interest for their health benefits when included in the diet. These chemical compounds are well-known for their antioxidant and immunomodulatory properties (Han et al. 2007). Several studies report that long-term intake of diets rich in plant polyphenols provides humans with defence against the onset of cancers, cardiovascular diseases, diabetes, osteoporosis, and neurodegenerative disorders (reviewed by Pandey & Rizvi 2009). Indeed, polyphenols can hinder biochemical processes induced by tumor promoters, induce apoptosis in cancerous cells, and suppress the growth of tumor cells (Han et al. 2007). Moreover, polyphenols have significant effects in human and animal reproduction (e.g. Pasquariello et al. 2020, Guelfi et al. 2023). Studies conducted in piglets have shown that dietary polyphenols can have multiple positive effects on the gastrointestinal function and morphology also of extra-enteral organs, like salivary glands and relative secretions in pigs (Cappai et al. 2010; 2013; 2014) or else improving antioxidant capacity (Silva-Guillen et al. 2020; Varricchio et al. 2019) and enhancing the intestinal microbiota along with resistance to pathogens (Fiesel at al. 2014; Verhelst et al. 2014). They influence intestinal morphology by increasing the size of colon crypts and reducing the activation of gut-associated lymphoid tissue (GALT), suggesting improved nutrient absorption and a potential immuno-preventive effect (Sehm et al. 2007). Polyphenol rich diet, such as those from acorns, can accumulate in animal tissues (Alldritt et al. 2019); also, supplementation of aqueous oregano extract in adult pigs, promotes antioxidant activity and improves the production of glycoconjugates involved in direct and indirect defense, thus improving the protection of the intestinal mucosa (Dall’Aglio et al. 2020b; Mercati et al. 2020). This study aimed to highlight the presence of visfatin in the stomach and duodenum of weaning pigs and to characterize its expression and localization in piglets fed varying levels of polyphenolic additions. Specifically, the research explored the impact of increasing dietary additions of olive mill wastewater extracts on visfatin expression and localization in the glandular stomach and duodenum cells of weaning piglets using molecular, immunohistochemical, and digital image analysis techniques.
2. Material and methods
2.1. Animal experimentation and sample collection Twenty-seven weanling piglets from six different litters of a local farm in Sardinia, raising Sardo breed pigs, were assigned to three feeding groups. All feeding groups were consistent per number (n.9), age (4 weeks) and body weight of animals (range at start: 7.73 – 8.04 kg). At weaning, piglets were moved to three stalls with slatted floors and close to each other, to allow eye contact among the groups. All animals were housed according to the dietary group. The feeding protocol started from the 28th day of life (dof), with apelleted diet offered for 14 days to all piglets. The commercial feed for weaned piglets represented the basal diet. Its main nutrient composition was: dry matter (DM) content of 91.9% of the feed as fed, crude protein (CP) at 20.4% in DM, and gross energy (GE) density of 19.0 MJ/kg of DM. The basal diet was offered to the control group (C). A polyphenolic extract dosed at 120 ppm and 240 ppm was added to the basal diet and offered to S1 group and S2 group, respectively. The extract was obtained from olive mill wastewater and included directly in the formulation, and the predominant polyphenols were tyrosol, hydroxytyrosol and verbascoside. Relatively low doses of supplementation were chosen to minimize the risk that the addition of polyphenolrich material might negatively impact piglet feed intake. The animals were regularly slaughtered after 14 days of feeding (42nd dof). The production performance parameters measured/calculated included live weight and average daily gain. Samples from the proper gastric region of the glandular stomach and cranial portion of duodenum were collected from 13 randomly selected piglets (C: 5; S1: 4; S2: 4) and a part of the tissue was fixed in 10% neutral buffered formalin, while a part was stored at -80°C. All piglets were slaughtered in an authorized abattoir under the supervision of the veterinary for the local health authority, for human consumption according to the Council Regulation (EC) No. 1099/2009, which focuses on protecting animals Jo u na l P re -p ro f during the slaughter process under law n.333/98 (Council Directive 93/119/EC of 22 December 1993) as specified by Annex C of Section II. 2.2. Morphological staining and immunohistochemistry Formalin fixed tissues were routinely embedded in paraffin, and 5 µm sections with a transversal orientation of the stomach and duodenum were cut and mounted on SuperFrost® Plus adhesion microscope slides (Bio Optica Milano, IT). The slides underwent in duplicate to chromogenic immunohistochemistry (IHC) for the porcine visfatin analysis (Dall’Aglio et al. 2020a; Maranesi et al. 2024). Briefly, after deparaffination and rehydration the slides were pre-treated using heat mediated antigen retrieval (3 cycles of 5 min at 700 w in the microwave with sodium citrate buffer, pH 6.3). The latters underwent endogenous peroxidase blocking (H2O2 3% in distilled water per 10 min) and were successively incubated with normal goat serum (1:10 – 30 min) to block nonspecific binding. The primary polyclonal antibody against porcine visfatin (ab233294, Abcam, UK) was employed at 1:100 PBS dilution and incubated for 1,5 h at room temperature. A section per slide was used as negative control omitting the primary antibody. Interspersed with PBS washes, the slides were incubated with the secondary biotinylated goat anti-rabbit IgG antibody (1:200) and treated with the avidin–biotin complex. The chromogenic reaction was obtained using diaminobenzidine, and each slide development lasted for 20 sec. Only one of two replicates underwent counterstain with hematoxylin, and the slides were routinely mounted with 0.16 mm Ø coverslips. A Nikon Eclipse E800 microscope (Nikon Corp., Tokyo, Japan) with a digital camera (Nikon Dxm 1200 digital camera) was used to inspect and photomicrograph the sections. 2.3. Digital-Image Analysis The Leica Aperio AT2 whole-slide scanner (magnification: 20X, 0.75 Numerical Aperture, 2X optical magnifier) was utilized to scan counterstained slides where clear DAB-positive cells were identified. The image analysis was performed using QuPath open source software platform (version 0.4.3 – Bankhead et al. 2017). After scanning, all IHC whole slide images (WSI – pixel size: 0.2526 X 0.2526 µm) underwent a visual quality evaluation to confirm that the data did not contain any out-of-focus artefacts, substantial variations in background level due to white-balancing errors, or notable variations in hematoxylin staining. Two different QuPath projects (glandular stomach and duodenum) were created and all the WSI were set as ImageType 'BRIGHTFIELD_H_DAB'. For each project all the WSI were pre-processed using the “estimate stain vectors” function in QuPath, estimating the stain vectors from a single slide and applied to all the others; to estimate the DAB stain vector an annotation on a representative zone with DAB-positive cells was created, and the vectors auto-detected (Min OD channel 0.05; Max total OD 2; Ignore extrema 2%); the hematoxylin values were directly extrapolated from an annotation on a representative zone within the negative section on the slide; the background values were directly extrapolated from an annotation on a representative blank zone. Only a positive section for each WSI was used to select the region of interest (ROI). ROIs included the tunica mucosa (excluding the muscularis mucosae) and were annotated using the “Wand” and “Brush” tools. Annotations (a single ROI per section – Fig. 1) included as much of the readable and parsable section as possible, and districts facing the lumen were excluded from the selection. Moreover, regions affected by tissue-processing and staining artefacts, such as foldings and cracks, were manually excluded from the annotation process or recalibrated after the following steps (annotation reshaping). Once the ROIs were highlighted, the QuPath’s positive cell count function (watershed cell detection – Malpica et al. 1997) was used on smaller annotations or whole ROIs, and the parameters were adjusted and refined for each project (glandular stomach and duodenum) in order to optimize and obtain the accurate detections. Jo ur na l P re -p ro of Figure 1: Representative sections of whole slide images (WSI) showing annotated regions of interest (ROIs) in the tunica mucosa of the glandular stomach (A; sample 3) and the duodenum (B; sample 8), with object classifier mask applied (epithelial cells in light blue; extra-epithelial cells in red). Areas facing the lumen and regions with artefacts were excluded from the ROI. Scale bars: 2 mm (A); 1 mm (B). A positive cell detection with 3 levels of intensity threshold positivity (1+; 2+; 3+) was used. The final parameters for each organ were: Glandular stomach: Detection image hematoxylin OD; requested pixel size 0.3 µm; background radius 0 µm; use opening by reconstruction true; median filter radius 0.25 µm; sigma 1.25 µm; minimum cell area 13 µm2; maximum cell area: 100 µm2; maximum background intensity 2; split by shape true; exclude DAB (membrane staining) true; cell expansion 4 µm; include cell nucleus true; smooth boundaries true; make measurements true; threshold score compartment Cytoplasm DAB OD mean; single threshold false; threshold 1+ 0.26; threshold 2+ 0.285; threshold 3+ 0.31. Duodenum: Detection image hematoxylin OD; requested pixel size 0.3 µm; background radius 0 µm; use opening by reconstruction true; median filter radius 0.25 µm; sigma 1.25 µm; minimum cell area 13 µm2; maximum cell area: 100 µm2; maximum background intensity 2; split by shape true; exclude DAB (membrane staining) true; cell expansion 5 µm; include cell nucleus true; smooth boundaries true; make measurements true; threshold score compartment Cytoplasm DAB OD mean; single threshold false; threshold 1+ 0.25; threshold 2+ 0.275; threshold 3+ 0.3. Each annotation (ROI) was classified as "Region*" and the “Positive cell detection” function ran automatically by mean of scripts on the whole project in QuPath (see Supplementary Note S1). After using the cell detection algorithm, the training of the machine learning based QuPath’s object classifier was applied to discriminate epithelial cells and extra-epithelial cells (i.e. all detected cells not classified as epithelial in the mucosa). Briefly, using the function “Train object classifier” function, a random trees classifier (random forest) was actively trained on all WSI in each project. Annotations on single detections or clusters of epithelial cells and extra-epithelial cells were traced, reviewed and corrected until a well performing cell classification mask was obtained. The variable importance of each parameter used to build both object classifiers were assessed via the “show log” function after the classification and are reported in the Supplementary Note S2. The quality control of the cell segmentation and object classifier was performed by a morphologist. Furthermore, a correction of positive cells classified as “extra-epithelial” but “epithelial” cells was carried out by annotating the formers as “epithelial” via carefully checking every detected cell through “show detection measurements” in each slide, and implementing the object classifier (negative mining). Cells that were not Jo ur na l P re -p ro of re-classified as “epithelial” and “positive” were considered false positives and/or artifactual, and, instead of leaving them under the classification of extra-epithelium, were deleted (excluded from the dataset). In similar manner, a quality control of “epithelial“ and “positive” cells with deletion of artifactual ones via “show detection measurements” was performed. For each analysed positive section coming univocally from a single WSI, organ (glandular stomach or duodenum) and individual within a group were exported or calculated the following values: Epithelial cells (%): percentage of detected cells classified as epithelial; Extra-epithelial cells (%): percentage of detected cells classified as extra-epithelial; Percentage of positive cells calculated based on the total detected cells or only on epithelial cells (%); H-score for total detected cells or epithelial cells only: quantitative metric aiming to explain the percentage and intensity of positive cells in a ROI. It is calculated as (1 × percentage of “1+” positive cells [weak staining]) + (2 × percentage of “2+” positive cells [moderate staining]) + (3 × percentage of “3+” positive cells [strong staining]). The resulting score falls within the range of 0 (all cells are negative) to 300 (all cells are “3+” positives) (McCarty et al. 1985). 2.4. Double-label immunofluorescence and Colocalization analysis A double-label localization of visfatin with serotonin was carried out in glandular stomach and duodenal tissues from one individual per group showing clear DAB-positive cells to visfatin (i.e., individuals 1 [C], 7 [S1], and 11 [S2] – Palmioli et al. 2021; 2023; Maranesi et al. 2024). Briefly, the formalin-fixed and paraffinembedded (FFPE) tissue sections were heated at 60°C for 2 hours and then treated with xylene and gradually rehydrated with different ethanol concentrations until distilled water. The sections were subjected to antigen retrieval using a Tris-based solution (Antigen Unmasking Solution, Tris-based, Vector Laboratories, Burlingame, CA, USA) in a microwave (3 cycles of 5 min at 700 w), followed by washing with acetone (20 min, -20°C) and incubation with NH4Cl (30 min, RT). All subsequent steps were performed in a humidity chamber. The tissues were permeabilized for 10 min with Triton X-100 solution (BioChemika, Sigma-Aldrich, Darmstadt, DE) and incubated for 1.5 h with Fish Serum Blocking Buffer (ThermoScientific, Rockford, USA). Successively, the sections were incubated with primary antibodies (Polyclonal Rabbit Anti-Visfatin antibody, 1:500, ab233294, Abcam, UK; monoclonal mouse anti-human serotonin antibody, 1:500, M0758, DakoCytomation, Glostrup, Denmark) and left overnight at RT. Interposed by 5 min PBS washes, fluorescent secondary antibodies (goat anti-rabbit IgG Alexa Fluor 555 antibody, 1:1000, A-21428, Invitrogen, Thermo Fisher Scientific, USA; donkey anti-mouse IgG Alexa Fluor 488 antibody, 1:1000, ab150105, Abcam, UK) were applied for 1.5 h. Prior to mounting the sections with DAPI solution (Fluoroshield with DAPI, SIGMA, St. Louis, USA), they were washed for 20 min with Sudan black solution (Sudan Black B, Sigma-Aldrich, Darmstadt, DE). Negative control sections were treated with antigen retrieval solution and omitting the primary antibodies. The microphotographs used for immunofluorescence analysis were obtained from three consecutive sections per sample, with one field acquired per section at 400x magnification using an Olympus BX51 Fluorescence Microscope. Images were captured using a Nikon DS-Qi2 microscope digital camera and NISElements D software provided by Nikon Europe B.V. Colocalization analysis of visfatin with serotonin for both glandular stomach and duodenum was conducted using the FIJI version of ImageJ software (version 1.53t; Schindelin et al. 2012) and the JACoP plugin (Bolte & Cordelières, 2006; Maranesi et al. 2024). To distinguish signal from background and minimize noise and artifacts, an intensity threshold (IT) was applied to each channel in every image. The reported colocalization metrics included Pearson's correlation coefficient (PC) denoted by the "r" value, Manders' coefficients (MCs) with the specified ITs represented by M1 (fraction of the 519 nm wavelength [green channel – Alexa Fluor 488] overlapping the 565 nm wavelength [red channel Jo ur na l P re -p ro of – Alexa Fluor 555]) and M2 (fraction of the 565 nm wavelength overlapping the 519 nm wavelength), as well as the Overlap coefficient (OC) represented by the "r," "k1," and "k2" values. 2.5. RNA extraction and quantitative reverse transcription PCR (RT-qPCR) Relative gene expression of visfatin from glandular stomach and duodenum was realized following already described methods (Dobrzyn et al. 2018; Kaminski et al. 2021). Briefly, samples previously stored at -80°C were homogenized using Omni-μU (Analytical Control, Cinisello Balsamo, Milano, Italy) in 1 ml of the solution supplied with TRIzol, following the manufacturer's instructions. The concentration and quality of total RNA were then evaluated with Nanodrop 2000c spectrophotometer (ThermoFisher, Carlsbad, CA, USA). Each RNA sample was applied to GenTegra® RNA Screw Cap Tubes (GenTegra, Pleasanton, CA, USA) for dry storage and, after shipping at room temperature, it was recovered and reconstituted with molecular biology grade water following the manufacturer's instructions. One microgram of the RNA was then converted into cDNA using the Omniscript RT Kit (Qiagen, Germany) and oligo(dt)15 primers (Roche, Germany) at 37 °C for 1 hour (total volume 20 µl), followed by termination at 93 °C for 5 minutes. Quantitative real-time PCR (qPCR) was carried out on an AriaMx Real-Time PCR System (Agilent Technologies, USA) using Power SYBR Green Master Mix (Applied Biosystems Inc., USA). Specific primer pairs for visfatin, cyclophilin (PPIA) and β-actin (ACTB) genes were used, with PPIA and ACTB serving as reference genes. The qPCR mix contained 20 ng of cDNA, forward and reverse primers, Power SYBR Green PCR Master Mix (12.5 μl Applied Biosystems, USA), and RNase-free water up 20 μl of final volume. Negative controls contained nuclease free water instead of the cDNA template. The RT-qPCR conditions, primers sequences and concentration for each target gene are detailed in Table 1. The amplification was performed in duplicates, and the melting curve analysis validated the specificity of the amplification. The relative gene expression level was determined using the comparative cycle threshold method (2−ΔΔCT; Livak and Schittgen 2001) and normalized with the geometric mean of PPIA and ACTB reference gene expression levels. All non-template control reactions showed Ct values below the detection threshold. Table 1. Characteristics of primers and conditions used to study the gene expression in porcine glandular stomach and duodenum. Gene Primers sequences Reaction Conditions Primer (nM) Accession number (GenBank) References visfatin F: 5′-CCA GTT GCT GAT CCC AAC AAA3′ R: 5′-AAA TTC CCT CCT GGT GTC CTATG-3′ Activation (1): 50 °C, 2 min; Activation (2): 95 °C, 10 min; 40 cycles of: denaturation (95 °C, 15 s) and annealing (60 °C, 1 min) 300 NM_001031793.2 Szeszko et al. 2016, Kaminski et al. 2021 PPIA F: 5′-GCA CTG GTG GCA AGT CCA T-3′ R: 5′-AGG ACC CGT ATG CTT CAG GA3′ Activation (1): 50 °C, 2 min; Activation (2): 95 °C, 10 min; 40 cycles of: denaturation (95 °C, 15 s) and annealing (60 °C, 1 min) 300 AY266299.1 Dobrzyn et al. 2018 ACTB F: 5′-ACA TCA AGG AGA AGC TCT GCT ACG -3′ Activation: 95 °C, 10 min; 40 cycles of denaturation (95 °C, 15 s), annealing (61 500 U07786.1 Dobrzyn et al. 2018 Jo ur al Pr epr oo f R: 5′-GAG GGG CGA TGA TCT TGA TCT TCA -3′ °C, 1 min) and elongation (70°C, 1 min) PPIA: cyclophilin (PPIA); ACTB: β-actin. 2.6. Statistical analysis Statistical analysis was performed with JASP (v0.18.3) and RStudio (v2023.03.1+446) software. In case of grouping, data are presented as means ± standard error of mean (SEM). For statistical significance the alpha level was set at 0.05. Contingency tables were analyzed using the χ2 test to determine the independence of frequencies between factors, specifically the treatment groups (C; S1; S2) and the absence/presence of IHC positivity in GI tract cells. Data obtained through digital image analysis (DIA) of slides featuring DAB-positive visfatin cells in IHC (i.e. epithelial cells (%); extra-epithelial cells (%); percentage of positive cells on total detected cells (%); H-score of total cell detected; positive epithelial cells (%); H-score of epithelial cells) were used to assess differences between the C group and groups receiving polyphenolic supplementation (S1 + S2 aggregated together for this analysis and referred as S in the text) in the glandular stomach and in the duodenum. The non-parametric Mann-Whitney test was employed for the latter statistical analysis. Same data were employed to assess differences between the glandular stomach and the duodenum via Student Ttest following the respect of assumption checked through test of normality (Shapiro-Wilk) and test of equality of variances (Levene's). A multivariate approach was used to assess morphological trends of epithelial cells positive and negative to visfatin in both glandular stomach and duodenum. To perform this principle component analysis (PCA), QuPath’s morphometrical values from each epithelial cell were used as response variables (i.e. Nucleus..Area, Nucleus..Perimeter, Nucleus..Circularity, Nucleus..Max.caliper, Nucleus..Min.caliper, Nucleus..Eccentricity, Cell..Area, Cell..Perimeter, Cell..Circularity, Cell..Max.caliper, Cell..Min.caliper, Cell..Eccentricity and Nucleus.Cell.area.ratio). Tidyverse, factoextra, vegan and ggplot2 R packages were used. Data were scaled and centred and the two dimensions with the highest percentage of explained variances were chosen (Supplementary Fig. S1 A, B). Differences between groups in the production performance parameters and in the gene expression data were analyzed by one-way ANOVA followed by Duncan's post hoc test and p values < 0.05 were considered as statistically significant. These data were previously checked for the assumptions of normality (Shapiro–Wilk test) and homogeneity of variances (Levene’s test). 3. Results 3.1. Production performance Immediately before treatment, the piglets weighed 2.71 ± 0.17 kg, increasing to 11.6 ± 0.28 kg by the end of the experiment. No significant effect of the treatments was observed on overall weight gain during the administration period or on weight at slaughter (ANOVA, p > 0.05). 3.2. Qualitative immunohistochemistry and morphology Visfatin-positive cells have been detected in five out of 13 individuals. The immunostaining was typically cytoplasmic and highlighted cells mostly located at the basal portion of gastric and intestinal glands (Fig. 2 A, C). All the positive cells belonged to epithelial tissue; they were mostly oval or irregular in shape in the glandular stomach, while in the duodenum were mostly triangular to flask-shaped, with a broad base that narrows toward the apical surface or were relegated on the basal membrane (Fig. 2 B, D). Cells were Jo ur na l P epr oo f considered to have a DAB-positive reaction when the signal was clearly more intense compared to the surrounding epithelial cells, regarded as aspecific background. The slides considered negative showed the background, and no detectable signal was observed in the negative control sections (data not shown). There was not a conserved trend between experimental groups, but animals with positive cells have been identified in all of them (Table 2). Moreover, piglets in which visfatin was present showed always immunopositivity in both glandular stomach and duodenum (Table 2). A different pattern of immunoreaction has not been observed among positive individuals. No dependence between experimental groups and presence/absence of visfatin (χ² = 0.536, p = 0.764) was highlighted. 7 S1 + + 8 S1 + + 9 S1 - - 10 S2 - - 11 S2 + + 12 S2 - - 13 S2 - - GS: Glandular stomach; D: Duodenum; +: immunostaining present; -: immunostaining absent. 3.3. Digital image analysis The whole slide images scanned and analysed belonged to 5 visfatin immuno-positive individuals (Table 2). The positive cell detection (Fig. 3) and the object classifier (Fig. 1, 3) ran in two different projects, one for each organ. Jo ur na l P re -p ro of Figure 3: Identical fields from the glandular stomach (A, C, E) and the duodenum (B, D, F) displaying unmasked tissue (A, B), positive cell detection (C, D) and object classifier (E, F) applied on selected annotations within WSI (visfatin immuno-positive slides counterstained with hematoxylin) from QuPath projects. Panels C and D show detected negative cells (nucleus and cytoplasm periphery circled in blue) and positive cells graded by intensity (1+: yellow; 2+: orange; 3+: red). Panels E and F represent the trained object classifiers distinguishing epithelial cells (outlined in light blue) from extra-epithelial cells (outlined in red); note that the object classifiers are built on the positive cell detection layer, with positive cells rendered in darker shades (examples indicated by arrows). Scale bars: 50 µm. Jo ur na l P re -p ro of The values of epithelial cells (%), extra-epithelial cells (%), percentage of positive cells on total detected cells (%), H-score of total cell detected, percentage of positive on epithelial cells (%) and H-score of epithelial cells resulted from DIA of ROIs from glandular stomach are reported in Table 3, and the ones from duodenum are reported in Table 4. No statistical difference was found between individuals from C group and experimental groups (S) in both glandular stomach and duodenum (Mann-Whitney test, p > 0.05) for all the resulting values. Similarly, there was no statistical difference in epithelial/extra-epithelial composition and cell positivity scores between glandular stomach and duodenum (Student T-test, p > 0.05). Table 3. Results from Digital Image Analysis of visfatin immuno-positive slides from the glandular stomach. Individual Group Epithelial cells (%) Extraepithelial cells (%) Total detected cells: positive (%) Total detected cells: Hscore Epithelial cells: positive (%) Epithelial cells: Hscore 1 C 51.98 48.02 0.37 0.50 0.48 0.58 3 C 60.29 39.71 0.81 1.12 1.29 1.74 7 S1 32.78 67.22 0.01 0.01 0.03 0.03 8 S1 62.70 37.30 1.89 2.69 2.86 4.03 11 S2 46.70 53.30 0.21 0.27 0.40 0.52 Grouped C + S 50.89 ± 5.36 49.11 ± 5.36 0.66 ± 0.33 0.92 ± 0.48 1.01 ± 0.51 1.38 ± 0.72 C: control group; S1: treatment group fed with polyphenolic extract dosed at 120 ppm; S2: treatment group fed with polyphenolic extract dosed at 240 ppm; C + S: control and treatment groups clustered together per investigated tissue. Table 4. Results from Digital Image Analysis of visfatin immuno-positive slides from the duodenum. Individual Group Epithelial cells (%) Extraepithelial cells (%) Total detected cells: positive (%) Total detected cells: Hscore Epithelial cells: positive (%) Epithelial cells: Hscore 1 C 57.89 42.11 0.44 0.73 0.76 1.24 3 C 45.82 54.18 0.27 0.51 0.58 1.1 7 S1 55.84 44.16 0.91 1.73 1.62 3.10 8 S1 45.32 54.68 0.37 0.72 0.82 1.59 11 S2 49.93 50.07 0.71 1.21 1.41 2.39 Grouped C + S 50.96 ± 2.56 49.04 ± 2.56 0.54 ± 0.12 0.98 ± 0.22 1.04 ± 0.2 1.88 ± 0.38 C: control group; S1: treatment group fed with polyphenolic extract dosed at 120 ppm; S2: treatment group fed with polyphenolic extract dosed at 240 ppm; C + S: control and treatment groups clustered together per investigated tissue. The 13 variables used to assess morphological trends via a multivariate approach belonged to 315602 observations (epithelial cells) in the glandular stomach and to 264334 in the duodenum. The PCA plots (Fig. 4; S1 E, F) were built on the first two dimensions (PC1 and PC2), explaining together 70.34% of the variance in the glandular stomach and 68.05% in the duodenum (see Fig. S1 A, B). Loading matrix showing the Jo ur na l P re -p ro of contributions of each original variable to PC1 and PC2 are reported in Supplementary Note S3 and Fig. S1 C, D. Plots (Fig. 4; S1 E, F) did not show any clustering of negative or positive cells, but they were similarly influenced by the variables in both organs. Geometric dimension metrics such as area, perimeter and minimum/maximum calliper lied on the positive side of PC1 in both organs. Shape symmetry indices had a very strong influence from the second dimension, and positioned on its two different extremities (i.e. eccentricity on the positive side of PC2 and circularity on the negative side), as expected from near-tocomplementary measures. Nucleus/Cell area ratio was highly associated with cell eccentricity in the duodenum. This relationship was less pronounced in the glandular stomach. Figure 4: Principal Component Analysis of 13 morphometrical values from each epithelial cells in glandular stomach (A) and duodenum (B), constructed based on the first two principal components (PC1 and PC2). Negative cells are depicted as pink circles, while positive cells are represented by light blue triangles. Scaling factor of loading matrix (represented by arrows) was adjusted to 12.5 for explanatory purposes. 3.4. Double-label immunofluorescence and Colocalization analysis Immunofluorescence revealed the localization of both visfatin and serotonin secreting cells at the basal portion of the gastric and intestinal glands (Fig. 5). The major part of visfatin-positive cells secreted serotonin, whereas a more abundant number of visfatin-positive cells did not co-occur with serotonin (Fig. 5, arrows). No detectable signal was observed in the negative control sections (data not shown). Colocalization analysis for serotonin with visfatin in the glandular stomach using ITs of 31 (green channel, i.e. serotonin) and 100 (red channel, i.e. visfatin) resulted in r=0.544 (PC), M1=0.019 (MC), M2=0.012 (MC), r=0.943 with k1=2.628 and k2=0.338 (OC). The same analysis for the duodenum using ITs of 31 (green channel, i.e. serotonin) and 57 (red channel, i.e. visfatin) resulted in r=0.781 (PC), M1=0.018 (MC), M2=0.019 (MC), r=0.959 with k1=1.346 and k2=0.683 (OC). Jo ur na l P re -p ro of Figure 5: Immunofluorescence staining of visfatin and serotonin in glandular stomach (A, B, C) and duodenum (D, E, F) of piglets. A, D: visfatin (red – Alexa Fluor 555); B, E: serotonin (green - Alexa Fluor 488); C, F: Merged channels with DAPI (blue) nuclear counterstain; arrows point cells positive only to visfatin and not cooccurring with serotonin. Scale bars: 10 µm. 3.5. RT-qPCR analysis Visfatin expression in the glandular stomach and duodenum of C piglets was the lowest among the groups and significantly differed from groups S1 and S2 in the stomach. In the duodenum visfatin expression of C group was statistically different only when compared to S2 (Fig. 6). In the stomach, visfatin mRNA levels in groups S1 and S2 were not significantly different, whereas in the duodenum, expression from group S2 was significantly higher compared to S1 (Fig. 6). Jo ur na l P re -p ro of Figure 6: visfatin gene expression in the piglets' glandular stomach (A) and duodenum (B) analyzed via RTqPCR. The x-axis represents the experimental groups (C, S1, S2), while the y-axis represents arbitrary units of visfatin mRNA normalized to the mean expression of reference genes. Results are presented as means ± SEM. Bars labelled with different superscripts indicate significant differences (determined by one-way ANOVA at p < 0.05 followed by Duncan's post hoc test at p < 0.05).
4. Discussion
The findings of this study provide significant insights into the role of visfatin in the gastrointestinal tract of pigs, particularly in the context of dietary polyphenol supplementation. The presence of visfatin in the stomach and duodenum, and the modulation of its transcripts by polyphenolic compounds derived from olive mill wastewater, expands our understanding on the localization and potential functions of this adipokine in non-adipose tissues. Visfatin plays crucial roles in the regulation of energy homeostasis, inflammation, and key processes related to reproduction in pigs, including the modulation of protein expression in the pituitary gland and luteal cells during the estrous cycle and early pregnancy (Szymanska et al. 2023; Kopij et al. 2024; Mlyczyńska et al. Jo ur na l P re -p ro of 2024). It also influences lipid metabolism in porcine adipocytes by regulating the glucose uptake (Kopij et al. 2024). Despite that, this adipokine is not a marker of lipidic accumulation (Palin et al. 2008) and it has not a role in porcine fattening (Barbe et al. 2020). Instead, visfatin is potentially modulated by inflammation in pigs, and it should be a suitable marker to investigate during anti- and pro-inflammatory challenges (e.g. Pang et al. 2021; Zhou et al. 2017). Information on visfatin role in GI tract of swine is very scanty and limited to the gene expression in the stomach (Palin et al. 2008). This study demonstrates that epithelial cells at the basal portion of the gastric and intestinal glands of piglets produce visfatin at the cytoplasmic level. These DAB-stained cells were clearly distinguishable from the rest of the surrounding epithelial cells of the mucosa via qualitative and semiquantitative methods, with the possibility to measure the optic density of their cytoplasmatic immunostaining. In the glandular stomach positive cells were oval or irregular, while in the duodenum they were triangular or flask-shaped, as corroborated by PCA results. The morphology of these cells is consistent with enteroendocrine cells, whose primary function is to secrete signaling substances (Sternini et al. 2008). The colocalization of most visfatin-positive cells with serotonin-producing cells confirms the secreting nature of these cells. The partial colocalization of the two cell morphotypes is a situation already described for other adipokines (Maranesi et al. 2024; Palmioli et al. 2023; Cinti et al. 2000; Cammisotto et al. 2005) and can be explained by the presence of heterogeneous enteroendocrine cell populations of the GI tract (Sternini et al. 2008). In this context, visfatin can act on distant targets through the bloodstream as other peptides secreted by enteroendocrine cells (reviewed in Sternini et al. 2008) or locally by autocrine and paracrine mechanism as hypothesized by Sethi and Vidal-Puigand (2005) and described for other adipokines (Dall’Aglio et al. 2021; Mercati et al. 2019a; Mercati et al. 2019b; Mercati et al. 2018). However, the receptor for visfatin is not still known. DIA of WSI was performed on the mucosa of the five individuals showing clearly visfatin-positive cells. The different parameters of the QuPath’s positive cell detection function between the two tissues were the cell expansion and the DAB intensity thresholds. A discrimination of the epithelial cells from all the other tissue cells (extra-epithelial cells) was carried out for each tissue via the separate training of two QuPath’s object classifiers. The aim of this approach was to characterize the tunica mucosa by the identification of the DAB-positive cells and to secondarily perform a further characterization via cell classification. The resulting quantitative values – percentages and metrics – permitted an objective comparison of the selected ROIs. Nevertheless, the epithelial/extra-epithelial composition and the cell positivity were not significantly diverse among glandular stomach or duodenum or within treatment groups. This result suggests that polyphenolic supplementation does not influence the expression of visfatin protein revealed by chromogenic IHC in our selected samples. The qualitative assessment of IHC has been carried out in all the samples from the 13 individuals. Interestingly, the five piglets with clearly positive cells exhibited them in both the glandular stomach and the duodenum. The other eight individuals never showed cells with a chromogenically intense reaction. Despite the limited statistical power, there was no significant association between the experimental groups and the presence or absence of visfatin. Positive cells were found sporadically across all experimental groups, indicating that the protein expression might not be influenced by the polyphenolic supplementation. Additionally, the consistent immunopositivity in both glandular stomach and duodenum across positive individuals, as well as the characteristic immunostaining and localization of the positive cells, suggest a uniform distribution pattern of visfatin that is independent of experimental variables. The diverse visfatin immunostaining, with some cells being clearly DAB positive and others negative, could be attributed to intrinsic biological variability, environmental factors or, although minimized, sample handling, technical variability and sensitivity of the technique (Fothergill and Furness, 2018). Lastly, the intensity of the DABpositive reaction being used as a criterion might have introduced variability in interpretation, especially if the staining considered background was lightly-to-moderately intense. The cells considered DAB-positive were distinctly more intense than the surrounding epithelial cells, which were considered for analytic purposes as nonspecific background (Fig. 2, 3 A, B). On the other hand, it cannot be excluded that these epithelial cells Jo ur al Pr epr of expressed lower, but still present, levels of “visfatin”. In fact, the intracellular form of “visfatin”, better known as iNAMPT, is known to be a regulator of NAD levels in all the cellular compartments (Garten et al. 2015). Conversely, the actual visfatin or eNAMPT is the protein secreted in the extracellular space (Semerena et al. 2023). It is possible that the clearly DAB-positive enteroendocrine cells secrete the eNAMPT, while the other epithelial cells maintain relatively high levels of iNAMPT that are detectable by the IHC analysis. Moreover, in our qualitative IHC analysis can be noted that muscular and connective tissue do not present the light intensity found in the epithelial cells, while a similar intensity can be found in the nervous tissue (see Fig. 2A). Further characterization of the visfatin expression is warranted to unveil this possibility, but this potential explanation can justify the different trends between the quantification of visfatin transcripts (RT-qPCR) and the semi-quantification of visfatin protein (DIA) after experimental polyphenolic addition. The gene expression of visfatin significantly changed across experimental conditions in both tissues. In the glandular stomach, visfatin mRNA levels were found to be the lowest in the C group, which significantly differed from both the S1 and S2 groups. This indicates that the inclusion of polyphenolic compounds in the diet may upregulate visfatin gene expression in the stomach. The absence of a significant difference between S1 and S2 suggests that the effect of polyphenols on visfatin expression in the stomach likely reaches a plateau, with no additional increase observed at the higher supplementation level. The duodenal visfatin expression presented a slightly different pattern. The C group exhibited significantly lower visfatin mRNA levels compared to S2, but not to S1, indicating a more pronounced effect of the higher polyphenol dose on visfatin expression in the duodenum. Moreover, visfatin expression in S2 was significantly higher than in S1, suggesting a dose-dependent response to polyphenol supplementation in this part of the GI tract. This tissuespecific pattern could be attributed to the distinct physiological environments and cellular compositions – comprising enteroendocrine cell populations – of these regions, which may interact differently with dietary components or their metabolites. Future studies should include direct measurements of visfatin protein levels to corroborate the mRNA data, as gene expression does not always correlate with protein abundance due to post-transcriptional regulation (Liu, 2008, Gry et al. 2009, de Sousa Abreu et al. 2009, Schwanhäusser et al. 2011). Additionally, expanding the sample size would enhance the robustness of the findings and allow for more definitive conclusions. The utilization of DIA in anatomical studies using WSI holds substantial promise, despite the lack of statistically significant findings in this research. DIA offers precise and reproducible measurements of tissue and cellular structures, which can significantly enhance the accuracy of histological and morphometrical evaluations, as currently done in the field of digital pathology (e.g. Apaolaza et al. 2021; Jiang et al. 2021; Smits et al. 2023; Zhang et al. 2023). In our study, DIA enabled detailed visualization and quantification of visfatin expression and localization in the glandular stomach and duodenum cells of weaning piglets. This technology's ability to process large volumes of images and extract quantifiable data efficiently can greatly benefit anatomical research, especially when dealing with complex tissue architectures and subtle expression patterns. The analysis on the double-label immunofluorescence for serotonin with visfatin in the glandular stomach and the duodenum reveals differences in the degree of colocalization and in the nature of their spatial relationships. There is moderate positive correlation (PC) between the molecules in the glandular stomach, whereas in the duodenum the correlation is stronger. This variation might be explained by the distinct expression patterns of serotonin and serotonin-related genes/proteins in the tissues analyzed. In fact, in humans these levels are higher in the duodenum than in the antrum of the stomach (Van Lelyveld et al. 2007; Busslinger et al. 2021), and a similar trend might be exhibited by porcine tissues (Jung et al. 2018; Wiarda et al. 2023). The high OC in both tissues emphasizes the robust overall co-occurrence between the two channels, where the green channel (serotonin) is contributing more to the overlap (i.e. k1 value). This is likely due to the fact that serotonin is not secreted by all the neuroendocrine cells of the GI tract (see Oberg 1998; Palmioli et al. 2021), and that it overlaps more the visfatin-positive cells and not vice versa. Visfatin seems to Jo ur na l P re -p ro f be secreted by a more widespread enteroendocrine population of cells compared to serotonin. However, while MCs indicate relatively small fractions of overlap for each channel individually, the high values of the PC and OC with its components suggest substantial overall colocalization, with specific characteristics unique to each tissue type. A colocalization analysis of visfatin with a protein expressed in a wider manner by neuroendocrine cells such as INSM1, synaptophysin and chromogranin A (see Möller et al. 2024) would give more comprehensive results to characterize visfatin secretion in GI tract. Nonetheless, the revealed morphology of the visfatin-positive cells at the chromogenic reaction, as well as the partial co-occurrence with serotonin at the fluorogenic reaction currently suggest that most of the visfatin-producing enteroendocrine cells belong to the subset of enterochromaffin-like cells in the stomach and to enterochromaffin cells in the duodenum (see Håkanson et al. 1994; Kuramoto et al. 2007; Penkova et al. 2010; Rezzani et al. 2022). The application of principal component analysis (PCA) to our dataset of epithelial cells from the glandular stomach and duodenum aimed to uncover potential morphological trends associated with immunohistochemical visfatin expression. Despite the first two principal components captured a substantial portion of the variance, the PCA did not demonstrate distinct morphological clustering between visfatinpositive and visfatin-negative cells, suggesting that epithelial morphological parameters are not influenced by visfatin production or secretion. However, limitations in our cell segmentation method – such as the expansion of nuclear areas into adjacent regions which may not precisely capture exact cell shapes – could have contributed and influenced the PCA results. In both organs, the nuclei and the (epithelial) cells tend to have high variability in their geometric dimensions. In duodenum, cells with a higher Nucleus/Cell Area Ratio may exhibit higher eccentricity values, suggesting elongation or irregularity in cell shape. Moreover, in duodenum fewer cells appear to have circular shape compared to those with an eccentric shape, a pattern not observed in the glandular stomach. Nonetheless, even if this pattern should be attributed to the inherent characteristics of the tissues, it could also have been influenced by the different cell expansion parameters used during cell detection (4 μm in the glandular stomach and 5 μm in the duodenum).
5. Conclusions
This study investigates the complex role of visfatin in the gastrointestinal tract of pigs and its modulation by dietary polyphenols. While polyphenolic supplementation appears to influence visfatin gene expression, its impact on protein levels requires further investigation. The colocalization analysis permitted to reveal the enteroendocrine nature of visfatin-positive cells, which may be involved in the regulation of gastrointestinal activity. Despite not significant, the digital image analysis provided valuable insights into oxyntic mucosa and duodenal cells morphology. “Digital anatomy”, as an emerging field paralleling the advancements in digital pathology, offers a transformative potential for research in histology and morphology by utilizing whole slide imaging to provide precise, reproducible measurements and detailed visualizations, thereby enhancing the accuracy and efficiency of tissue and cellular studies. Additional research is needed to fully understand the functions and regulatory mechanisms of visfatin in nonadipose tissues, particularly in response to dietary interventions.
Acknowledgments
We are grateful to Prof Paolo Giovenale and Dr Nicola Becchetti for the support in digitizing of the slide. Moreover, we acknowledge Dr Paola Coliolo and Dr Gabriele Scattini for their technical assistance. Jo rn al Pr -p ro of
Funding
KD was supported by the Polish National Science Center research grant OPUS 16 [grant number: 2018/31/B/NZ9/00781]. CRediT authorship contribution statement Marini D.: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data Curation, Visualization, Writing – Original draft preparation. Cappai M.G.: Resources, Investigation. Palmioli. E.: Validation, Investigation. Battacone G.: Resources, Investigation, Formal analysis. Maranesi M.: Investigation, Writing – Review & Editing. Dobrzyn K.: Validation, Formal Analysis, Investigation, Visualization, Writing – Review & Editing. Mercati F.: Conceptualization, Resources, Methodology, Investigation, Supervision, Project Administration, Writing – Review & Editing. Dall’Aglio C.: Conceptualization, Resources, Methodology, Investigation, Supervision, Project Administration, Writing – Review & Editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work, the authors used ChatGPT by OpenAI to improve language and readability. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper
Graphical abstract
Highlights Visfatin is expressed in enteroendocrine cells of piglet stomach and duodenum. Dietary polyphenols modulate visfatin gene expression but not protein levels. Colocalization analysis reveals visfatin is produced by serotonin-secreting cells. WSI digital image analysis is a powerful tool to study normal cell morphology. Jo ur na l P re -p ro of
Article Images (0)
×![]()
Bioz is the world’s most comprehensive AI search engine for scientific experimentation. The Bioz search engine offers researchers billions of data-driven product, technique, and protocol recommendations. Bioz taps into the latest advances in Artificial Intelligence (AI) – including Natural Language Processing (NLP), Machine Learning (ML) and Generative AI to mine and structure hundreds of millions of pages of complex and unstructured scientific papers. By harnessing the power of AI, Bioz provides researchers with an unprecedented amount of summarized scientific experimentation knowledge right at their fingertips, ultimately helping to speed up drug discovery and the rate of success in finding cures for diseases. Bioz is used by over 15 Million researchers from over 17,000 different universities and companies in 196 countries.
