Quantitation and chemical coding of enteroendocrine cell populations in the human jejunum.

Recent studies reveal substantial species and regional differences in enteroendocrine cell (EEC) populations, including differences in patterns of hormone coexpression, which limit extrapolation between animal models and human. In this study, jejunal samples, with no histologically identifiable pathology, from patients undergoing Whipple's procedure were investigated for the presence of gastrointestinal hormones using double- and triple-labelling immunohistochemistry and high-resolution confocal microscopy. Ten hormones (5-HT, CCK, secretin, proglucagon-derived peptides, PYY, GIP, somatostatin, neurotensin, ghrelin and motilin) were localised in EEC of the human jejunum. If only single staining is considered, the most numerous EEC were those containing 5-HT, CCK, ghrelin, GIP, motilin, secretin and proglucagon-derived peptides. All hormones had some degree of colocalisation with other hormones. This included a population of EEC in which GIP, CCK and proglucagon-derived peptides are costored, and four 5-HT cell populations, 5-HT/GIP, 5-HT/ghrelin, 5-HT/PYY, and 5-HT/secretin cell groups, and a high degree of overlap between motilin and ghrelin. The presence of 5-HT in many secretin cells is consistent across species, whereas lack of 5-HT and CCK colocalisation distinguishes human from mouse. It seems likely that the different subclasses of 5-HT cells subserve different roles. At a subcellular level, we examined the vesicular localisation of secretin and 5-HT, and found these to be separately stored. We conclude that hormone-containing cells in the human jejunum do not comply with a one-cell, one-hormone classification and that colocalisations of hormones are likely to define subtypes of EEC that have different roles.

The effect of high-fat diet-induced metabolic disturbance on corneal neuroimmune features.

PURPOSE: The highly innervated cornea is susceptible to nerve loss secondary to systemic diseases such as diabetes and metabolic disturbances caused by high-fat diet. In this study, we characterize the effect of high-fat diet on the mouse corneal neuroimmune phenotype, including changes to corneal nerve density and resident immune cells, alongside the clinical assessment of corneal thickness and endothelial cell density. METHODS: Male C57Bl6/J mice, aged 10 weeks, were fed a high-fat diet (60 kcal% fat, 5.2 kcal/g) or control diet (10 kcal%, 3.8 kcal/g) for 16 weeks. At the study endpoint, metabolic parameters (HbA1c, weight, fasting glucose, body fat) were measured to confirm metabolic disturbance. Clinical imaging of the anterior segment was performed using optical coherence tomography to measure the corneal epithelial and stromal thickness. Corneal sensory nerves were visualized using flatmount immunostaining and confocal microscopy. The topographical distribution and density of sensory nerves (BIII-tubulin+), intraepithelial CD45+ and MHC- II+ cells, stromal macrophages (IBA1+CD206+) and endothelial cells (ZO-1+) were analysed using FIJI. RESULTS: High-fat diet mice had significantly higher blood HbA1c, higher body weight, a higher percentage of body fat and elevated fasting glucose compared to the control diet mice. Corneal epithelial and stromal thickness was similar in both groups. The sum length of the basal nerve plexus was lower in the central and peripheral cornea of mice fed a high-fat diet. In contrast, the sum length of superficial nerve terminals was similar between groups. Epithelial immune cell density was two-fold higher in the central corneas of high-fat diet mice compared to control diet mice. IBA1+CD206+ macrophage density was similar in the anterior stroma of both groups but was significantly higher in the posterior stroma of the peripheral cornea in the high-fat diet mice compared to controls. The percentage of nerve-associated MHC-II+ cells in the epithelium and stroma was higher in HFD mice compared to controls. Endothelial cell density was similar in the corneas of high-fat diet mice compared to controls. CONCLUSION: Together with corneal neuropathy, corneal immune cells in mice fed a high-fat diet were differentially affected depending on their topographical distribution and location within cornea, and appeared in closer proximity to epithelial and stromal nerves, suggesting a local neuroimmune disruption induced by systemic metabolic disturbance.

Distribution and co-expression patterns of specific cell markers of enteroendocrine cells in pig gastric epithelium.

Although the pig is an accepted model species for human digestive physiology, no previous study of the pig gastric mucosa and gastric enteroendocrine cells has investigated the parallels between pig and human. In this study, we have investigated markers for each of the classes of gastric endocrine cells, gastrin, ghrelin, somatostatin, 5-hydroxytryptamine, histidine decarboxylase, and PYY cells in pig stomach. The lining of the proximal stomach consisted of a collar of stratified squamous epithelium surrounded by gastric cardiac glands in the fundus. This differs considerably from human that has only a narrow band of cardiac glands at its entrance, surrounded by a fundic mucosa consisting of oxyntic glands. However, the linings of the corpus and antrum are similar in pig and human. Likewise, the endocrine cell types are similar and similarly distributed in the two species. As in human, gastrin cells were almost exclusively in the antrum, ghrelin cells were most abundant in the oxyntic mucosa and PYY cells were rare. In the pig, 70% of enterochromaffin-like (ECL) cells in the antrum and 95% in the fundus contained 5-hydroxytryptamine (5-HT), higher proportions than in human. Unlike the enteroendocrine of the small intestine, most gastric enteroendocrine cells (EEC) did not contain colocalised hormones. This is similar to human and other species. We conclude that the pig stomach has substantial similarity to human, except that the pig has a protective lining at its entrance that may reflect the difference between a pig diet with hard abrasive components and the soft foods consumed by humans.

Cellular and sub-cellular localisation of oxyntomodulin-like immunoreactivity in enteroendocrine cells of human, mouse, pig and rat.

We use a monoclonal antibody against the C-terminal of oxyntomodulin (OXM) to investigate enteroendocrine cells (EEC) in mouse, rat, human and pig. This antibody has cross-reactivity with the OXM precursor, glicentin (Gli) but does not recognise glucagon. The antibody stained EEC in the jejunum and colon of each species. We compared OXM/Gli immunoreactivity with that revealed by antibodies against structurally related peptides, GLP-1 and glucagon and against GIP and PYY that are predicted to be in some EEC that express OXM/Gli. We used super-resolution to locate immunoreactive vesicles. In the pancreas, OXM/Gli was in glucagon cells but was located in separate storage vesicles to glucagon. In jejunal EEC, OXM/Gli and GIP were in many of the same cells but often in separate vesicles, whereas PYY and OXM/Gli were commonly colocalised in the same storage vesicles of colonic EEC. When binding of anti-GLP-1 to the structurally related GIP was removed by absorption with GIP peptide, GLP-1 and OXM/Gli immunoreactivities were contained in the same population of EEC in the intestine. We conclude that anti-OXM/Gli is a more reliable marker than anti-GLP-1 for EEC expressing preproglucagon products. Storage vesicles that were immunoreactive for OXM/Gli were almost always immunoreactive for GLP-1. OXM concentrations, measured by ELISA, were highest in the distal ileum and colon. Lesser concentrations were found in more proximal parts of small intestine and pancreas. Very little was in the stomach. In EEC containing GIP and OXM/Gli, these hormones are packaged in different secretory vesicles. Separate packaging also occurred for OXM and glucagon, whereas OXM/Gli and PYY and OXM/Gli and GLP-1 were commonly contained together in secretory vesicles.

Development of the aganglionic colon following surgical rescue in a cell therapy model of Hirschsprung disease in rat.

Patients with Hirschsprung disease lack enteric ganglia in the distal colon and propulsion of colorectal content is substantially impaired. Proposed stem cell therapies to replace neurons require surgical bypass of the aganglionic bowel during re-colonization, but there is inadequate knowledge of the consequences of bypass. We performed bypass surgery in Ednrb-/- Hirschsprung rat pups. Surgically rescued rats failed to thrive, an outcome reversed by supplying electrolyte- and glucose-enriched drinking water. Histologically, the bypassed colon had normal structure, but grew substantially less in diameter than the functional region proximal to the bypass. Extrinsic sympathetic and spinal afferent neurons projected to their normal targets, including arteries and the circular muscle, in aganglionic regions. However, although axons of intrinsic excitatory and inhibitory neurons grew into the aganglionic region, their normally dense innervation of circular muscle was not restored. Large nerve trunks that contained tyrosine hydroxylase (TH)-, calcitonin gene-related peptide (CGRP, encoded by Calca or Calcb)-, neuronal nitric oxide synthase (nNOS or NOS1)-, vasoactive intestinal peptide (VIP)- and tachykinin (encoded by Tac1)-immunoreactive axons occurred in the distal aganglionic region. We conclude that the rescued Ednrb-/- rat provides a good model for the development of cell therapies for the treatment of Hirschsprung disease.