Mast cells are involved in the gastric hyperemic response to acid back diffusion via release of histamine

Acid back diffusion into the rat stomach mucosa leads to gastric vasodilation. We hypothesized that histamine, if released from the rat mucosa under such conditions, is mast cell derived and involved in the vasodilator response. Gastric blood flow (GBF) and luminal histamine were measured in an ex vivo chamber. Venous histamine was measured from totally isolated stomachs. Mucosal mast cells (MMC), submucosal connective tissue mast cells (CTMC), and chromogranin A-immunoreactive cells (CgA IR) were assessed morphometrically. After mucosal exposure to 1.5 M NaCl, the mucosa was subjected to saline at pH 5.5 (control) or pH 1.0 (H(+) back diffusion) for 60 min. H(+) back diffusion evoked a marked gastric hyperemia, increase of luminal and venous histamine, and decreased numbers of MMC and CTMC. CgA IR cells were not influenced. Depletion of mast cells with dexamethasone abolished (and stabilization of mast cells with ketotifen attenuated) both hyperemia and histamine release in response to H(+) back diffusion. GBF responses to H(+) back diffusion were attenuated by H(1) and abolished by H(3) but not H(2) receptor blockers. Our data conform to the idea that mast cells are involved in the gastric hyperemic response to acid back diffusion via release of histamine.

Pattern of distribution and prognostic value of angiogenesis in pancreatic duct carcinoma: a semiquantitative immunohistochemical study of 45 patients

In specimens obtained from resected pancreata, the intratumoral microvessel density (IMD), the proliferation rate of the neoplastic parenchymal cells, and their p53 protein expression were assessed. The sources of errors were great in the measurements of the IMD. This statement can be illustrated by the finding that when the IMD was calculated by manual counting in five areas of intense neovascularization (hot spot regions), using x200 and x400 magnifications, the numbers of microvessels per square millimeter were 65+/-23 and 106+/-8, respectively, which reflects a significant difference. Two patterns of microvessel distribution could be identified: one with hot spots only in the stroma (n = 19) and one in which the hot spots were located in areas of neoplastic parenchyma (including its stroma) (n = 26). The IMD was significantly greater in the latter group. There was no general correlation of neoplastic disease with the IMD. However, when a scoring system was used to assess the angiogenesis, hot spots in areas of neoplastic parenchyma were associated with a greater proliferation rate of the tumor cells, and with a short length of survival of the patients from their neoplastic disease.

Neuroendocrine differentiation in bronchial carcinomas of classic squamous-cell type: an immunohistochemical study of 29 cases applying the tyramide signal amplification technique

With regard to the cellular origin of bronchial squamous-cell carcinomas, there are some clinicopathologic and experimental data indicating a link between neuroendocrine (NE) bronchial tumors and the traditionally non-NE squamous-cell carcinomas. Against this background, 29 consecutively resected bronchial squamous-cell carcinomas were examined immunohistochemically (IHC) by means of the specific NE cell marker chromogranin A (CgA), using not only conventional IHC methods, but also the technique with increased sensitivity, offered by the tyramide signal amplification (TSA) procedure. Whereas none of the 29 tumors displayed CgA immunoreactive (IR) cells using the conventional IHC procedure, 10 were found to display a fine granular CgA IR in the neoplastic parenchymal cells using the TSA technique. This incidence is higher than previously reported. However, the CgA IR cells never formed any majority cell population of the neoplastic parenchyma; when present, most of them occurred as micronodules or larger confluent areas in the peripheral most undifferentiated parts of the carcinomatous sheets. Single CgA IR cells were detected only rarely in the spinocellular or keratinized areas. It can be speculated that the observations conform with the recently proposed hypothesis that there is a reservoir of NE progenitor cells in the bronchial mucosa capable of proliferation.

Insulin-like growth factor-II in endocrine pancreatic tumours. Immunohistochemical, biochemical and in situ hybridization findings

In earlier studies a high-molecular-weight (HMW) insulin-like growth factor-II (IGF-II) peptide was identified in adult human pancreas and localized to the insulin-producing B-cells. This peptide has now been investigated in neoplastic insulin cells. Forty endocrine pancreatic tumours and 17 pancreatic adenocarcinomas of ductal type were included in the study. All cases were investigated with immunohistochemical techniques using antibodies to IGF-II, insulin, pro-insulin, glucagon, somatostatin, pancreatic polypeptide, gastrin and vasoactive intestinal peptide (VIP). Frozen tissue from nine tumours and two normal pancreatic glands was extracted, gel separated, and quantified using radioimmunoassay. The tumours were also investigated by in situ hybridization. IGF-II-immunoreactive cells were found in nearly all the 18 insulin-producing tumours (16/18), in a minority of the other endocrine tumours, but not in pancreatic adenocarcinomas. All extracts from the endocrine tumours showed varying amounts of IGF-II and had different molecular-weight forms. The immunohistochemical and radioimmunoassay findings are both based on immunological binding and were further confirmed by Northern blot and in situ hybridization. These results show that IGF-II is expressed in insulin-producing tumours as well as in pancreatic tumours producing other peptides, in contrast to normal pancreatic islets where IGF-II is found exclusively in insulin-producing cells.

[Immunohistochemical diagnosis by means of tyramide signal amplification]

Immunohistochemistry is used for in situ detection of proteins in histological slides and is now an important diagnostic tool. Due to several methodological and biological factors, conventional immunohistochemical procedures may sometimes have too low sensitivity, especially for tracing the histogenesis of malignant tumours. A few years ago, an amplification technique was introduced which greatly increased the sensitivity of some of the commonly used immunohistochemical methods. This technique permits the use of primary antibodies in significantly lower concentrations compared with the conventional methods. Alternatively, one can keep the antibody concentration unchanged and use the enhanced sensitivity to detect scarce proteins, which are not visualised by traditional immunohistochemical procedures. We present a brief description of the technique and show some examples of its use in the diagnosis of neuroendocrine carcinomas. Tyramide signal amplification might become an important supplement for the diagnosis and classification of malignant tumours.

Spontaneous ECLomas in cotton rats (Sigmodon hispidus): tumours occurring in hypoacidic/hypergastrinaemic animals with normal parietal cells

We have identified cotton rats with a high female-predominant occurrence of spontaneous gastric carcinomas localized to the oxyntic mucosa, classified as malignant enterochromaffin-like (ECL) omas. The present study was made to further characterize these ECLomas and surrounding oxyntic mucosa, both morphologically using histochemical and immunohistochemical methods, and for gene expression by northern blot analysis. Among eight female cotton rats, three had an irregularly thickened oxyntic mucosa, increased stomach weight and a high serum gastrin level. Histopathological examination showed adenomatous hyperplasia of the thickened oxyntic mucosa with areas of an invasive neoplastic tumour. Immunohistochemistry, using the general neuroendocrine cell marker chromogranin A (CgA) and the specific ECL cell marker histidine decarboxylase (HDC), showed a considerably increased ECL cell density. These ECL cells displayed active proliferation, with hyperplasia, dysplasia and neoplasia. Parietal cells were not found in the tumour tissue. Parietal cell density was only slightly reduced in the surrounding oxyntic mucosa. The antral mucosa was histopathologically normal with a normal number of gastrin-immunoreactive cells. Likewise, somatostatin-immunoreactive cells did not show any differences in the antral and oxyntic mucosa between rats with pathological and normal oxyntic mucosa. Northern blot analysis revealed increased expression of CgA and HDC mRNA in the thickened oxyntic mucosa, whereas H(+)/K(+) ATPase mRNA was similar in the oxyntic mucosa of those with thickened and normal oxyntic mucosa. Gastrin mRNA in the antral mucosa was high in animals with thickened oxyntic mucosa. Somatostatin mRNA expression was similar in the antral mucosa of control animals and animals with a thickened oxyntic mucosa. We conclude that the spontaneous gastric carcinoma occurring in female cotton rats is an ECLoma developing secondary to hypergastrinaemia due to reduced intragastric pH. The mechanism for reduced acidity is not known, but is not gastric atrophy.

Clinical and histopathological tumour progression in ECL cell carcinoids ("ECLomas")

AIMS

The aims of this study were to illustrate the malignant potential of gastric enterochromaffin-like (ECL) cell carcinoids (ECLomas) associated with hypergastrinemia, and the gradual neoplastic progression of such tumours. In addition, we examined whether the tyramide signal amplification (TSA) technique could visualize immunohistochemical (IHC) neuroendocrine (NE) features in the dedifferentiated neoplastic ECL cells which were not detected by conventional methods.

METHODS

Conventional histopathological and IHC methods for visualizing ECL cells and cell proliferation were used in addition to the TSA technique.

OBSERVATIONS

Our patient was followed for 5 years. During that period, her ECLoma displayed all the signs of classical tumour progression, ultimately with the appearance of metastases in the regional lymph nodes, the liver and the skin. The neoplastic ECL cells became progressively dedifferentiated with an increasing number of Ki-67 immunoreactive (IR) cell nuclei. In addition, there was a substantial decrease in argyrophil and IR NE cells that could be visualized by conventional methods. By applying the TSA technique, however, the number of IR tumour cells increased considerably.

CONCLUSIONS

ECLomas secondary to hypergastrinemia should be closely followed for signs of clinical and histopathological tumour progression. Such ECLomas deserve early, active, radical surgical treatment. The TSA technique is a valuable tool for visualizing the characteristic IHC features in dedifferentiated NE cells.

Pre- and postnatal testosterone administration induces proliferative epithelial lesions with neuroendocrine differentiation in the dorsal lobe of the rat prostate

BACKGROUND

Androgens are implicated in the pathogenesis of prostatic carcinoma. We have elucidated the role of pre- and postnatal testosterone administration in the occurrence of proliferative lesions as well as neuroendocrine (NE) cells in the rat prostatic complex.

METHODS

Female rats were given a single dose of 9 mg testosterone enantate i.m. on day 15 of pregnancy; it gave a high testosterone exposure to the fetus in the early organogenetic period of the rat prostatic complex. One group of the male offspring was followed without further testosterone treatment; a second group received testosterone only in the pubertal period; a third group was given testosterone from puberty and throughout life (46 weeks). These groups were compared to parallel groups (1A-1C) of male offspring without a testosterone supplement in pregnancy.

RESULTS

The serum testosterone concentrations in the rats receiving testosterone were significantly higher than those of control rats. Histopathologically, the testosterone-induced proliferative lesions, mainly hyperplastic, were almost exclusively located in the dorsal lobe. Chromogranin A-immunoreactive (CgA-IR) cells were rarely found normally, but occurred more often in the proliferative lesions (P < 0.001).

CONCLUSIONS

The incidence of proliferative lesions in rats exposed to testosterone only in puberty was comparable to the incidence found in those rats receiving testosterone in puberty and throughout life. This finding may have clinical implications for young athletes, who use testosterone as an anabolic drug. The occurrence of CgA-IR cells increased in proliferative lesions in the dorsal lobe of the rat prostatic complex.

Neuroendocrine (ECL cell) differentiation of spontaneous gastric carcinomas of cotton rats (Sigmodon hispidus)

BACKGROUND AND PURPOSE

Female inbred cotton rats develop adenocarcinomas in the oxyntic mucosa. Since a female preponderance is typical for enterochromaffin-like (ECL) cell tumors, we examined such tumors for ECL cells. Gastrin plays a decisive role in ECL cell tumorigenesis, so blood gastrin concentration and gastric mucosal pH were measured.

METHODS

The stomachs from six female cotton rats (6 to 8 months old) were studied histologically, and at euthanasia, gastric mucosal pH was determined. Euthanasia was performed on 15 other female cotton rats of similar age for determination of blood gastrin values by radioimmunoassay (RIA) and gastric mucosal pH. Rats were classified macroscopically to have normal or thick oxyntic mucosa, with or without tumor.

RESULTS

Among the six cotton rats studied histologically, two 6-month-old rats had normal and two others had thick gastric mucosa, whereas two 8-month-old rats had thick mucosa with tumors. The ECL cells were markedly hyperplastic in all rats with thick mucosa, and ECL cells were found in the neoplastic parenchyma. All cotton rats with normal-appearing gastric mucosa had pH <2.5, whereas 14 rats with thick mucosa had pH >3.1 and hypergastrinemia.

CONCLUSIONS

Gastrin may play a major role in ECL cell hyperplasia and, perhaps, in adenocarcinoma genesis.

Prognosticating tools in primary neuroendocrine (Merkel-cell) carcinomas of the skin: histopathological subdivision, DNA cytometry, cell proliferation analyses (Ki-67-immunoreactivity) and NCAM immunohistochemistry. A clinicopathological study in 25 patients

Histopathologically, 18 of our patients had classical Merkel-cell carcinomas (MCC); seven had neuroendocrine (NE) carcinomas with features different from MCC, here called "aberrant MCC". These patients showed a progressive neoplastic disease with a fatal outcome in four of them. The cytometric DNA distribution pattern of the tumor cell nuclei of all the aberrant MCCs was found to be of the aneuploid type. By contrast, the neoplastic disease of the majority of patients with classical MCC ran a milder course; a fatal outcome occurred in only one of them. Here, the DNA ploidy pattern was of the euploid (diploid or tetraploid) type in eight cases and of the aneuploid type in another eight. Our recently described "proliferation cell index" (PCI), based on nuclear immunoreactivity (IR) with the proliferation "marker" antigen Ki-67, was significantly lower in those five MCCs of the classical "DNA-diploid" type than in the seven "DNA-aneuploid" ones. These five patients presented a mild neoplastic disease; only one had a local recurrence and none had metastases. Otherwise, neither the PCI values nor the NCAM IR of the MCC cells were found to be of any prognostic significance.

Blood flow to the pancreatic islet parenchyma of the Atlantic hagfish (Myxine glutinosa)

The islet organ of the Atlantic hagfish, a cyclostome, phylogenetically is the most original islet parenchyma. It is well equipped with blood vessels, but lacks nerves. The aim of the present study was to evaluate the relative proportions between regional blood flows in these animals with a microsphere technique and with laser-Doppler flowmetry. As regards the results obtained in the kidneys, gills, and islet organ, the data were essentially the same found using both methods. Approximately 20% of the injected microspheres were found in the kidneys; when corrected for weight, the gill bodies and the islet organ were also found to have the same blood perfusion. The blood flow to the brain and that of the gut and the liver were only approximately 30% and 10%, respectively, of that to the kidney. Topical application of the nitric oxide donor, sodium nitroprusside failed to affect the islet blood flow. Apart from this observation, this non-innervated "primitive" islet parenchyma did not differ from that of the amply innervated mammalian islets with regard to their high basal blood perfusion.

Ultrastructural localization of insulin-like growth factor-2 (IGF-2) to the secretory granules of insulin cells: a study in normal and diabetic (GK) rats

By using biochemical and light-microscopical techniques, insulin-like growth factor-2 (IGF-2) has recently been found in adult pancreas, co-localized immunohistochemically with insulin in the islet B-cells. The purpose of this study was to trace IGF-2 immunoreactivity (IR) at the ultrastructural level in normal and diabetic Goto-Kakizaki (GK) rats. Using a pre-embedding technique and immuno-gold-silver staining, IGF-2 antibody binding was localized exclusively to the halo of a subset of secretory beta-granules in normal rats. Insulin IR occurred more frequently in the granules. GK rats had, in addition to normal-looking islets, some islets with irregular shape and an increased amount of fibrous tissue, so-called "starfish-shaped" islets. In these, beta-granules were usually found, but most of the B-cells were also occupied by large, usually electron-translucent vesicles, some resembling crinophagic bodies, i.e., the sign of intracellular degradation of secretory granules. In starfish-shaped islets, IGF-2 IR was localized to the halo of beta-granules, as in GK islets with normal appearance. Occasionally, IGF-2 IR was also found in the cytoplasm and even in adjacent fibroblasts. Insulin IR was restricted to beta-granules. Because the lysosomes have IGF-2 receptors, the presence of IGF-2 peptide in secretory granules could explain why some granules are guided to lysosomes for degradation.

Increased amounts of a high molecular weight insulin-like growth factor II (IGF-II) peptide and IGF-II messenger ribonucleic acid in pancreatic islets of diabetic Goto-Kakizaki rats

Insulin-like growth factor II (IGF-II), a member of the insulin family, regulates cell growth and differentiation. The IGF-II gene is localized close to the insulin gene in man and rat. IGF-II peptide binds weakly to the insulin receptor and exerts insulin-like effects on the blood glucose level. We studied IGF-II in endocrine pancreas in an animal model of noninsulin-dependent diabetes mellitus, the Goto-Kakizaki (GK) rat. At the age of 2 months, these rats have structural islet changes, with fibrosis and irregular configuration, so-called starfish-shaped islets. Immunohistochemical investigation revealed IGF-II immunoreactivity in the beta-cells in both GK and control rats. Pancreatic extraction, followed by size separation using gel chromatography, disclosed a high mol wt form of IGF-II in all animals, and RIA measurements revealed a considerably larger amount of the IGF-II peptide in the 2-and 6-month-old GK rats than in the 1-month GK and control rats. In situ hybridization of 3-month-old GK rats showed increased IGF-II messenger RNA expression in the starfish-shaped islets of GK rats than in the islets with normal structure in both diabetic and control animals. The reason for the increased amount of IGF-II is unclear. As the animals are diabetic before the islet changes occur, it might be a compensatory effect in response to hyperglycemia, but could also be a cause of the islet fibrosis.

A technique for automatic/interactive assessment of the proliferating fraction of neoplastic cells in solid tumors. A methodological study on the Ki-67 immunoreactive cells in human mammary carcinomas, including a comparison with the results of conventional S-phase fraction assessments by means of DNA cytometry

Based on a computerized microscopy technique, a method has been devised which allows the practising pathologist to easily and rapidly assess quantitatively the relative number of actively proliferating neoplastic parenchymal cells in a tumor nodule. Our method has been tested on a series of 20 conventionally formalin-fixed and paraffin-embedded female mammary adenocarcinomas, using immunoreactivity with the MIB-1 monoclonal antibody against the cell proliferation antigen Ki-67. The values of the proportion of the MIB-1 immunoreactive cell nuclei were compared with those obtained DNA-cytometrically for the fraction of cells in the S-phase; a good correlation was found, although the MIB-1 values were consistently somewhat higher. A prerequisite for a success of the method was, of course, to achieve standardization of the MIB-1 immunostaining technique. By making simple adjustments of it, it could actually be improved to such an extent that almost the same color calibration and thresholding setup could be used. The measuring technique could be either interactive or automatic. The total number of immunoreactive and non-immunoreactive nuclei, as well as the total nuclear area of both cell types were registered in a computerized device. The data were accumulated sequentially for each measure field. To investigate the reproducibility of the immunostaining, two slides of each case were stained on different occasions. Each slide was measured three times; systemically randomly in the x- and y-axis-directions as well as in the subjectively defined histopathologically "most proliferative" area of the tumor. The values obtained were in good agreement with each other and obviously gave some valuable and objective supplementary pieces of information to that of the conventional clinical and histopathologic assessment of the degree of aggressiveness of a malignant neoplasm.

The value of cytometric DNA analysis as a prognostic tool in neuroendocrine neoplastic diseases

In several traditionally non-endocrine, common, human, neoplastic diseases, it has become well established during the last few years, that cytometric analyses of the DNA distribution pattern of the nuclei of tumour cells can be an excellent supplement to the conventional prognostic tools, (such as clinical staging and histopathologic malignancy assessments). When analogous studies of the value of DNA analysis by means of flow cytometry and/or image cytometry are made in neuroendocrine (NE) neoplastic diseases, the ensuing results often become rather disappointing. Thus, clear-cut aneuploid DNA histograms can be found in the neoplastic cell nuclei of clinically and histopathologically completely benign NE adenomas (and even hyperplastic nodules). In contrast, highly aggressive NE carcinomas not seldom reveal themselves to be composed of tumour cells with nuclei, displaying an euploid, i.e. normal, DNA pattern. Statements of this kind have been based on the results of comprehensive investigations in several laboratories, analysing such NE tumours as insulomas/insular carcinomas, bronchial/gastrointestinal carcinoids, phaeochromocytomas, paragangliomas, neuroblastomas, adenomas of the anterior pituitary gland, parathyroid adenomas, medullary carcinoma of the thyroid and Merkel-cell tumours of the skin. Thus, the prognostic value of the cytometric DNA ploidy pattern of the nuclei of neoplastic parenchymal cells is definitely lower in NE tumours than in most of the traditionally non-endocrine carcinomas and sarcomas. Data from published and unpublished series of these kinds of NE tumours, and those of prostatic and breast carcinomas with NE differentiation, are given. By means of a new, consecutive double staining technique, it was shown that in idiopathic nesidioblastosis, the hyperinsulinism is caused by beta cells with a nuclear DNA ploidy pattern of euploid type. By the same technique, it can be shown that in the pathogenesis of the hypergastrinaemia-induced ECL-cell carcinoids of the stomach, a switch from an euploid to an aneuploid nuclear DNA distribution pattern occurs in the ECL-cells when they pass from a state of hyperplasia to that of a genuine neoplasia. In neuroblastomas, a triploid (i.e. aneuploid) DNA pattern is part of an algorithm capable of predicting a 96% survival rate, whereas a diploid/tetraploid (i.e. euploid) DNA pattern predicts a 0% survival.

Preserved beta-cell density in the endocrine pancreas of young, spontaneously diabetic Goto-Kakizaki (GK) rats

The Goto-Kakizaki (GK) rat represents a spontaneous animal model of non-insulin-dependent diabetes mellitus (NIDDM) characterized by impaired glucose-stimulated insulin release from the pancreatic beta cells. To study whether an alteration in their islet beta-cell numbers occurs in parallel with the impairment of insulin secretion in this model, the relative volume density of beta cells was determined by means of conventional point sampling in immunostained 4-microns-thick sections of the pancreata from 8-week-old GK rats. The pancreata of nondiabetic Wistar rats were used as control parenchyma. In the GK pancreata the majority of islets was found to have a normal structure; only a few of the islets demonstrated an irregular shape (starfish-shaped islets) with fibrosis. The relative volume of the total endocrine parenchyma was found to be 2.0 +/- 0.6% (mean +/- SEM) of the whole pancreatic parenchyma in GK rats. In the control rats the corresponding value was 2.3 +/- 0.5%. The islet beta-cell density was also similar in GK and control rat islets, amounting to 75.2 +/- 8.5 and 66.9 +/- 6.6%, respectively. Thus, the total relative volume of beta cells was 1.5 +/- 0.5% in GK rats and 1.6 +/- 0.4% in controls. In conclusion, the density of beta cells is preserved in the pancreata of the young, diabetic GK rats, suggesting the absence of a causal relationship between the relative pancreatic beta-cell volume and the impaired glucose-induced insulin secretion in this NIDDM animal model.

IGF-2-like peptides are present in insulin cells of the elasmobranchian endocrine pancreas: an immunohistochemical and chromatographic study

Evidence for the presence of peptides, related to insulin-like growth factor 2 (IGF-2), has been obtained in the endocrine pancreas of the elasmobranchian species Raja clavata, the sting ray. By radioimmunoassay, IGF-2-like immunoreactivity was detected in Raja pancreas extract. Further characterization of this activity by acid gel chromatography revealed two distinct peaks of IGF-2-like immunoreactivity with apparent molecular weights of approximately 8.2 kDa and 4.5 kDa. Using the same IGF-2 antibody as well as antisera specific for mammalian IGF-1, insulin, glucagon, somatostatin and pancreatic polypeptide in double immunofluorescence studies, IGF-2-like immunoreactivity was located exclusively in insulin-immunoreactive cells. In contrast, IGF-1-like immunoreactivity was mainly observed in somatostatin- and glucagon-immunoreactive cells. A varying proportion (0-70%) of insulin-immunoreactive cells, however, displayed both IGF-1- and IGF-2-like immunoreactivity. Absorption studies indicated that the IGF-2-like peptides in Raja are different from mammalian and submammalian insulin and mammalian IGF-1, but similar to mammalian IGF-2. Thus, IGF-2-like peptides seem to occur during evolution as early as the phylogenetic development of the elasmobranchians. Furthermore, the results indicate a particularly conservative evolution of the islet IGF-2 system.

Phylogeny and ontogeny of the neuroendocrine cells of the gastrointestinal tract

The results of both phylogenetic and ontogenetic investigations of the evolution of the disseminated NE cells of the GIT have shown that they are members of the large NE system, consisting not only of the GI-NE cells but also of the neurons of the central and peripheral nervous system with their nerve fibers (the peptidergic nervous system) and of the classic, solid endocrine glands. The evolutionary studies also have shown that these three major parts of the NE system are closely interrelated to each other. The most original part is obviously the neuronal one, occurring already in the most primitive animals (the coelenterates). The next step in the evolution of the NE system is the appearance of NE cells of open type in the mucosa of the alimentary tract. Such gut NE cells are present in the most highly developed invertebrates, both Protostomian and Deuterostomian, and they persist and become even more diversified in the vertebrates, including humans. The presence of GEP-NE glands of classic, solid type seems to be a feature restricted to the true vertebrate animals. The earliest vertebrates (the jawless fish and the cartilaginous fish) often offer the best pieces of evidence for the manner in which the parenchyma of such a GEP-NE gland, notably the islets of Langerhans, is formed from disseminated NE cells of open type in a mucosa, in this case that of the gut.

The branching of insulin-like growth factor 1 and insulin: an immunohistochemical analysis during phylogeny

The co-existence of insulin-like growth factor 1 (IGF-1) with the classical islet hormones insulin (INS), glucagon (GLUC), somatostatin (SOM) and pancreatic polypeptide (PP) in the endocrine pancreas of representative species of cyclostomes (Myxine glutinosa), cartilaginous fish (Raja clavata, Squalus acanthias) and bony fish (Cottus scorpius, Carassius auratus, Cyprinus carpio, Anguilla anguilla) was studied by the use of monoclonal and polyclonal antisera and the double immunofluorescence technique. In all species investigated, IGF-1-like-immunoreactive cells were found in the endocrine pancreas, however, in varying localization. In Myxine glutinosa, all INS-immunoreactive cells and some of the SOM-immunoreactive cells contained IGF-1-like-immunoreactivity. In Raja and Squalus, only a minority of the INS-immunoreactive cells also displayed IGF-1-like-immunoreactivity. The majority of the IGF-1-like-immunoreactivity was observed in SOM- and in GLUC-immunoreactive cells. Different results were obtained in bony fish. In Cottus, in the Brockmann bodies and the small islets IGF-1-like- and INS-immunoreactivities co-existed to 100%. In contrast, in the other bony fish studied IGF-1-like-immunoreactivity was not observed in INS-immunoreactive cells: in Cyprinus, IGF-1-like-immunoreactivity was found in GLUC-, PP- and SOM-immunoreactive cells and in Carassius and Anguilla, in SOM-immunoreactive cells only. Thus, in all bony fish species with the exception of Cottus, IGF-1 and insulin display a distinct cellular distribution, similar to that of mammals. The present results, thus, may indicate that the branching of IGF-1 and insulin has occurred at the phylogenetic level of bony fish.

A high-molecular IGF-2 immunoreactive peptide (pro-IGF-2?) in the insulin cells of the islets of Langerhans in pancreas of man and rat

Histopathologically normal pancreatic parenchyma from 12 adult men and women, as well as that from 14 adult rats (Sprague-Dawley and Wistar strains), were investigated immunohistochemically with a mouse monoclonal antibody, raised against recombinant human pro-IGF-2. The antiserum showed no crossreactivity with insulin; IGF-1 had 0.1% of the reactivity of IGF-2. The immunohistochemical observations were checked by means of a radioimmunoassay (RIA), based on the same antibody, of an extract of a sample of one of the human pancreatic glands. Analogous investigations for insulin were made in parallel, using polyclonal insulin antisera. A high-molecular (12 kDa) IGF-2-like peptide was found in the islets of Langerhans, being localized to the insulin cells. These cells were identified as beta-cells by immunohistochemistry with insulin antisera on adjacent paraffin sections. From observations made by means of acid-gel-chromatography, the peptide was tentatively supposed to represent either pro-IGF-2, or a partially processed form of it.

Occurrence of members of the insulin superfamily in central nervous system and digestive tract of protochordates

Antisera specific for mammalian insulin-like growth factor 1 (IGF-1) and mammalian insulin and the double immunofluorescence technique were used for this study. IGF-1-like-immunoreactivity was localized in entero-endocrine cells in the gastro-intestinal tract of the protochordates Ciona intestinalis and Branchiostoma lanceolatum. Some of the specimens also showed IGF-1-like-immunoreactive (-IR) perikarya and fibers in the central nervous system. Whilst in rat endocrine pancreas, IGF-1-IR and insulin-IR occurred in different cell populations, in Ciona and Branchiostoma the vast majority of entero-endocrine cells and central neurons were IGF-1-like- +insulin-IR. A minor portion exhibited IGF-1-like-IR alone. For further characterization of the IGF-1-like-IR material, in Ciona intestinalis, peptides related to IGF-1 were identified by radioimmunoassay and gel chromatography. In accordance with the immunohistochemical results, IGF-I-like-IR was detected both in cerebral ganglion and in gastro-intestinal tract. Using acid gel chromatography, in Ciona gastro-intestinal tract the IGF-1-like-IR was found to occur in two peaks, with apparent molecular weights of approximately 16 kDa and 3 kDa. Absorption studies with insulin- and IGF-related peptides, with crude extracts and the peak material obtained after gel chromatography, indicated that the IGF-1-like peptides in Ciona are different from mammalian insulin and IGF-1. The findings are in accordance with the presence of a common insulin/IGF precursor molecule in protochordates.

Identification of IGF-1 receptors in primitive vertebrates

This is the first report of the existence of insulin-like growth factor (IGF-1) receptors in three representatives of lower vertebrates: the osteichtyes, chondrichtyes and cyclostomi. Competitive binding studies and affinity labelling of brain membranes from Cottus scorpius (sea scorpion), Raja clavata (ray) and Myxine glutinosa (atlantic hagfish) identified a mammalian type 1 or IGF-1 receptor by its binding specificity and the molecular size of its alpha-subunit. IGF-1 and IGF-2 are almost equally potent in displacing receptor-bound 125I-IGF-1 or 125I-IGF-2, and the proteins labeled with both tracers have a molecular size of 100,000-120,000 under reducing conditions. There was no evidence for the presence of a mammalian type 2 or IGF-2/mannose 6-phosphate receptor in brains of Cottus, Raja or Myxine. In all three species the binding of 125I-IGF-1 and 125I-IGF-2 was significantly higher in brain compared with liver and gastrointestinal tract, and the IGF-1 receptor could only be identified with certainty in Raja liver. It is concluded that the brain of three lower vertebrates express mammalian IGF-1 receptors, whereas IGF-2-mannose 6-phosphate receptors could not be detected.