Benjamin B. Rawdon

Gut endocrine cells in birds: an overview, with particular reference to the chemistry of gut peptides and the distribution, ontogeny, embryonic origin and differentiation of the endocrine cells.

This review deals with gut endocrine cells in birds. It focuses on both morphological and developmental aspects of these cells, which were included members of Pearses APUD series. They comprise many cell types, which, in birds as in mammals, produce serotonin and a range of regulatory peptides. The chemical structure of most avian gut peptides has been established. These peptides and their functions are outlined here. The types and distribution of avian gut endocrine cells are detailed and compared with the situation in mammals. In birds, ultrastructural work has been limited to certain types of gut endocrine cell and not as widely applied as in mammals. However, immunocytochemistry has found widespread application in studies on birds: the hatching chick and also the adult chicken and certain other species such as the quail and duck have been studied. Gut endocrine cells showing immunoreactivity for the following peptides/serotonin have been identified: somatostatin, pancreatic polypeptide (PP), peptide YY, glucagon, secretin, vasoactive intestinal peptide, gastrin, cholecystokinin (CCK), neurotensin, motilin, gastrin-releasing peptide, substance P, enkephalin and serotonin. The colocalization of different peptides (including chromogranins) and of peptides and serotonin in the same gut endocrine cells is reviewed: notable amongst such associations are glucagon with PP and gastrin/CCK with neurotensin in the same cells. On morphological grounds cells have been identified as endocrine in avian gut from at least 9 days of incubation. Immunocytochemical studies show the majority of the various types first to appear between 12 to 14 days of incubation, with substantial numbers being recorded from 17 days onwards. Experimental studies on chicken and quail embryos have determined the embryonic origin of gut endocrine cells: evidence is unequivocal that such cells arise from the endoderm, not the neural crest, other ectoderm or the mesoderm. Studies on avian embryos have also contributed to our knowledge of mechanisms controlling the differentiation of gut endocrine cells: evidence shows that gut mesenchyme plays an important role in provoking (or inhibiting) the development of gut endocrine cells and there are indications that the endocrine cell pattern in gut is established early and that an axially-derived factor may be important in this process. The kinds of genetic mechanism possibly involved are mentioned but full elucidation of the processes concerned is awaited. A better understanding of the formation of endocrine tumours of the gut should result from the findings.

Failure of insulin cells to develop in cultured embryonic chick pancreas: A model system for the detection of factors supporting insulin cell differentiation

SummaryLittle being known about factors necessary for insulin cell differentiation, we tested the chance observation that these cells were virtually absent from collagen gel cultures of embryonic avian pancreas in which the other pancreatic endocrine cells were numerous. Five-day dorsal buds stripped of their enveloping mesenchyme were embedded in gel and overlaid by a defined medium containing serum, then cultured for 7 days. Immunocytochemical evaluation showed a very low proportion of insulin cells. Substitution of the gel by a polyamino acid coating slightly increased the proportion. In an attempt to test for ability of insulin cell formation to recover, we transferred explants first cultured in collagen gel to polyamino-acid-coated dishes for a further 7 days. No improvement resulted. In controls grown for 14 days on a polyamino acid coating, insulin cells disappeared completely. We conclude that collagen gel does not support survival and differentiation of chick embryonic insulin cells and that the medium used is lacking in some essential factor(s). Determination of their identity should prove possible by exploitation of this model.

Distribution of serotonin-immunoreactive gut endocrine cells in chicks at hatching

Serotonin-immunoreactive, i.e. enterochromaffin (EC) cells were found to be widely distributed in the intestine of the newly hatched chick but sparse in the stomach, and being particularly abundant in the duodenum, upper ileum and rectum. Although in birds, as in mammals, EC cells are most abundant in the intestine, in the stomach they are far sparser than in mammals. Comparison of adjacent sections immunostained for serotonin and a peptide provided no evidence that EC cells in the hatching chick contain motilin or substance P, and that at least the great majority of bombesin-immunoreactive cells contain no serotonin: it is apparent that the mammalian pattern of distribution of peptides in EC cells does not occur in the chick, at least at hatching. Cross reaction of an antiserum to substance P with serotonin was discovered, suggesting the need for a review of existing evidence for co-localisation of this peptide with serotonin.

Effects of tri-iodothyronine (T3), insulin, insulin-like growth factor I (IGF-I) and transforming growth factor beta1 (TGFβ1) on the proportion of insulin cells in cultured embryonic chick pancreas

Abstract With a view to ultimately identifying factors involved in the development of pancreatic insulin cells, we have cultured dorsal pancreatic buds from 5-day chick embryos on a basement membrane matrix (Matrigel) in a serum-free medium supplemented with selected factors. The endodermal components of the buds were freed of almost all the mesenchyme so as to eradicate as much as possible of this source of some such factors. In 7-day cultures, insulin and glucagon cells were demonstrated immunocytochemically; numbers of insulin cells were expressed as a percentage of insulin plus glucagon cell counts. Our standard medium contained insulin. Addition of tri-iodothyronine to this medium did not increase the proportion of insulin cells, but in combination with raised concentrations of glucose and essential amino acids it improved somewhat the marked increase previously recorded for these nutrient conditions. Omission of insulin from the standard medium greatly reduced the proportion of these cells; substitution of insulin by insulin-like growth factor I increased the proportion considerably more than did insulin. To test for an overall effect of growth factors, explants were cultured in standard medium on Matrigel containing reduced amounts of growth factors: the proportion of insulin cells proved to be increased over that reached on normal Matrigel. The suspicion that transforming growth factor β1, a component of Matrigel, might act to reduce the proportion of insulin cells was tested and found to be correct. It is suggested that the different factors studied here may affect either or both of proliferation and determination in the differentiation pathway of insulin vis-à-vis glucagon cells.

DEVELOPMENT OF EMBRYONIC CHICK INSULIN CELLS IN CULTURE: BENEFICIAL EFFECTS OF SERUM-FREE MEDIUM, RAISED NUTRIENTS, AND BIOMATRIX

SummaryA previous finding that insulin cells do not survive or differentiate in explants of embryonic avian pancreas cultured in collagen gel with a serum-containing medium has provided a model system for identification of conditions favorable for development of these cells. To this end, we here modify the substrate and the medium. The epithelial component of dorsal pancreatic buds of 5-d chick embryos was cultured for 7 d on Matrigel in serum-containing and in serum-free medium, the latter incorporating insulin, transferrin, and selenium, Endocrine cell types were distinguished by immunocytochemistry; insulin cell counts were expressed as a proportion of insulin plus glucagon cells. With serum-containing medium, Matrigel stimulated a significant increase in this proportion as compared with collagen gel—3.1% as against 0.2%; the serum-free medium further increased this proportion to 17.3%. Raising the level of essential amino acids approximately fivefold increased the latter figure somewhat (to 18.9%), but it was more than doubled (to 37.4%) by raising the glucose concentration from 10 mM to 20 mM. Raising the levels of amino acids and glucose simultaneously yielded a lesser increase (to 31.8%). Some cultures grown in collagen gel and serum-containing medium for 7 d were transferred to Matrigel and serum-free medium for a further 7 d. Insulin cell development recovered, indicating that progenitor cells had survived and were stimulated to develop by the improved conditions. This study indicates that components of the biomatrix and the medium (in particular, a raised glucose concentration) are important for the survival and differentiation of embryonic insulin cells.

Early development of the gut: new light on an old hypothesis.

This review deals with the early development of the gut. It draws largely on information provided from the study of avian embryos. Evidence that concerns the early determination of the regional fate of the endoderm and mesoderm of the gut is reviewed. Gut endoderm can undergo a limited degree of differentiation from a remarkably early age when cultured in the absence of mesoderm and there is evidence that points to the establishment of a pre‐pattern in the early mesoderm before the genes responsible for patterning in gut are active. Initially, at least at cranial levels, those parts of the mesoderm and endoderm that are in contact are not those parts that will ultimately be in apposition; the consequence of this for any signalling between these layers is considered. In the light of the above information, the probable role of mesenchyme in gut development is re‐examined.

Can a non-gut mesenchyme support differentiation of gut endocrine cells?

SummaryThis experiment was designed to find out if endoderm lacks an intrinsic ability to give rise to gut endocrine cells, and, if not, whether differentiation of endocrine cells can be supported by mesenchyme from a source outside the digestive tract. Heterospecific combinations of proventricular endoderm and flank mesenchyme from chick and quail embryos at 3.25–4 days of incubation were grown as chorio-allantoic grafts to a final incubation age of 21 days. Re-associated proventricular endoderm and mesenchyme served as controls. The proventricular endoderm induced some smooth muscle in the flank mesenchyme but the latter did not support as advanced glandular morphogenesis as did proventricular mesenchyme. Nevertheless, endocrine cells differentiated in experimental as in control grafts and at similar frequencies. The various types were distinguished immunocytochemically by their contained peptides; the range of types found was specific for the proventriculus. Hence it is concluded not only that the particular non-gut mesenchyme used does support differentiation of gut endocrine cells, but also that the determination of the progenitors of endocrine cells, and the selection of the range of types destined to differentiate in a particular part of the digestive tract under normal circumstances, occurs early in development — before 3.25 days of incubation in the case of the proventriculus.

Morphogenesis of gut and distribution of the progenitors of gut endocrine cells at cranial somite levels of the chick embryo.

The primary objective of this study was to establish the distribution of the progenitors of selected gut endocrine cell types at cranial somite levels. In addition, analysis of the material has provided new information about the location of the presumptive territories of certain gut regions and of the pancreas. Narrow transverse strips of full‐thickness blastoderm two or three somites in length were excised at the levels of somites 1 to 5 of 8.5‐ to 18‐somite chick embryos and cultured as chorioallantoic grafts to an age equivalent to 20 days of incubation. The grafts were analysed by immunocytochemistry, and their morphology was evaluated. Individual grafts exhibited up to five different types of gut morphology, including those of oesophagus, proventriculus, gizzard, pyloric region, small intestine, and pancreas. The morphologic survey yielded new information about the location, extent, or both, of the territories of the pyloric region, the small intestine, and the pancreas. In general, the progenitors of gut endocrine cell types identified were those expected for the different morphologic regions: in only a few instances were ectopic endocrine cell types detected. The available evidence points to the progenitors of bombesin/gastrin‐releasing peptide cells being located cranial to somite 5 at the stages studied. Based on the morphology and the proportion of insulin cells, the development of pancreas in grafts appeared compromised compared with grafts of the intact dorsal pancreatic bud: this may relate to the likely exclusion of dorsal pancreatic bud mesoderm from the graft area. The results show that presumptive small intestinal endoderm in grafts can differentiate in the absence of homologous (i.e., small intestinal) mesoderm: this accords with the view that the primary source of positional information in the gut is in the endoderm.

Can gastric endoderm change the regionally specific inducing ability of presumptive small intestinal mesoderm

This study was designed to establish the source of gut mesoderms ability to induce regional pattern in the endoderm. The most obvious possibility is induction by the endoderm through epithelial‐mesenchymal interaction. To test this experimentally, reciprocal quail/chick combinations were prepared of early proventricular endoderm (that is already known to be regionally determined) and presumptive small intestinal mesoderm. The combinations were cultured for 7 days to allow for ‘programming’ of the mesoderm by the endoderm. After removal of the proventricular endoderm the mesoderm was combined with young gizzard endoderm. It is known that gizzard endoderm can be provoked to develop in either a proventricular or a small intestinal direction by association with the appropriate mesoderm. Thus, by combining intestinal mesoderm ‘programmed’ by association with proventricular endoderm with gizzard endoderm, the subsequent differentiation of the gizzard endoderm would indicate whether or not the inducing ability of the intestinal mesenchyme had been altered. In addition to such experimental grafts, three types of control graft were prepared. The results of the experiment, based on the morphology of the grafts and the immunocytochemical analysis of selected endocrine cell types, showed that in the majority of cases the gizzard endoderm developed the features of small intestine, not those of proventriculus. This indicates that at the stages studied, endoderm does not act to program mesoderm with which it is associated. If this does occur, it must take place at an earlier stage, i.e., before the time of explantation of the presumptive small intestinal mesoderm (1.25 days of incubation).