Calcium/calmodulin-dependent kinases can regulate the TSH expression in the rat pituitary.

PURPOSE: The endocrine secretion of TSH is a finely orchestrated process controlled by the thyrotropin-releasing hormone (TRH). Its homeostasis and signaling rely on many calcium-binding proteins belonging to the "EF-hand" protein family. The Ca2+/calmodulin (CaM) complex is associated with Ca2+/CaM-dependent kinases (Ca2+/CaMK). We have investigated Ca2+/CaMK expression and regulation in the rat pituitary. METHODS: The expression of CaMKII and CaMKIV in rat anterior pituitary cells was shown by immunohistochemistry. Cultured anterior pituitary cells were stimulated by TRH in the presence and absence of KN93, the pharmacological inhibitor of CaMKII and CaMKIV. Western blotting was then used to measure the expression of these kinases and of the cAMP response element-binding protein (CREB). TSH production was measured by RIA after time-dependent stimulation with TRH. Cells were infected with a lentiviral construct coding for CaMKIV followed by measurement of CREB phosphorylation and TSH. RESULTS: Our study shows that two CaM kinases, CaMKII and CaMKII, are expressed in rat pituitary cells and their phosphorylation in response to TRH occurs at different time points, with CaMKIV being activated earlier than CaMKII. TRH induces CREB phosphorylation through the activity of both CaMKII and CaMKIV. The activation of CREB increases TSH gene expression. CaMKIV induces CREB phosphorylation while its dominant negative and KN93 exert the opposite effects. CONCLUSION: Our data indicate that the expression of Ca2+/CaMK in rat anterior pituitary are correlated to the role of CREB in the genetic regulation of TSH, and that TRH stimulation activates CaMKIV, which in turn phosphorylates CREB. This phosphorylation is linked to the production of thyrotropin.

Differential expression of IL-8 and IL-8 receptors in benign, borderline and malignant ovarian epithelial tumours.

INTRODUCTION: Ovarian Cancer is the leading cause of death from gynecological malignancy. The poor prognosis is mainly due to presentation at a late stage and poor response to therapy. Much research is needed to identify diagnostic and prognostic biomarkers as well as therapeutic targets for ovarian cancer. Interleukin-8 is expressed by many tumour types and is known to have mitogenic, motogenic and angiogenic effects on tumour cells. AIMS: The aim of this study was to investigate the expression of IL-8 and IL-8 receptors (IL-8RA and IL-8RB) in different histological subtypes of ovarian tumours, as potential prognostic biomarkers in ovarian tumours. MATERIALS AND METHODS: Immunohitochemistry was used to study the expression of IL-8 and IL-8 receptors in 115 ovarian tumours including 21 benign tumours, 25 borderline tumours and 69 carcinomas of serous, clear cell, endometrioid and mucinous types. The correlation of expression profile, tumour type, stage, and progression free survival and overall survival was statistically analysed. RESULTS: IL-8 and IL-8 receptors were expressed in all types of tumours with variable intensity and subcellular distribution. There was a statistically significant correlation between levels of expression and tumour stage and tumour type, being mostly significant in serous tumours. No correlation with patient progression free survival or overall survival was found. CONCLUSION: This is the first study investigating the expression of IL-8 and IL-8 receptors using immunohistochemistry in different types of ovarian tumours, including benign and borderline tumours. IL-8 and IL-8RA are potential prognostic biomarkers and therapeutic targets in ovarian cancer, particularly in ovarian serous carcinoma.

Estrogen receptor beta in breast cancer.

Estrogen is essential for normal growth and differentiation in the mammary gland. It also supports growth of approximately 50% of primary breast cancers. For this reason, removal of estrogen or blocking of its action with the anti-estrogen, tamoxifen, is the main treatment for estrogen receptor alpha (ERalpha)-positive tumors. In 1996, when oncologists became aware of a second ER, ERbeta, there was some doubt as to whether this receptor would be of importance in breast cancer because the clinical consensus was that responsiveness to tamoxifen is related to the presence of ERalpha in breast cancer. Today we know that ERalpha and ERbeta have distinct cellular distributions, regulate separate sets of genes and can oppose each other's actions on some genes. We also know that ERbeta is widely expressed in both the normal and malignant breast and that there are proliferating cells in the breast which express ERbeta. In this review we summarize what is known about ERbeta in breast cancer and examine the possibility that ERbeta-selective ligands may well represent a useful class of pharmacological tools with a novel target, namely proliferating cells expressing ERbeta.

Localization of 7B2, neuromedin B, and neuromedin U in specific cell types of rat, mouse, and human pituitary, in rat hypothalamus, and in 30 human pituitary and extrapituitary tumors.

The distributions of three novel peptides, 7B2, neuromedin B, and neuromedin U, in rat, mouse, and human pituitaries, rat hypothalamus, and 30 human pituitary tumors were investigated with immunocytochemistry. Immunoreactivity for 7B2 was present in rat, mouse, and human gonadotropes, in intermediate lobe cells and posterior lobe nerve fibers in rats and mice, in rat hypothalamus (particularly in the median eminence), and in eight human pituitary gonadotropinomas. In gonadectomized rats, larger, more numerous LH beta- and 7B2-immunoreactive gonadotropes were seen than in controls. Extractable 7B2-like immunoreactivity was elevated but not significantly so in gonadectomized rat pituitaries [males: castrated, 37.4 +/- 4.3 (mean +/- SE); controls, 26.9 +/- 4.3; females: ovariectomized, 27.2 +/- 2.7; controls, 19.1 +/- 2.2 pmol/gland]. Neuromedin B immunoreactivity was found in normal rat and mouse thyrotropes and weakly in "thyroidectomy" cells in hypothyroid rats, in which extractable pituitary neuromedin B was significantly depleted (thyroidectomized, 87.0 +/- 14.0; methimazole-treated, 82.0 +/- 11.4; control, 230.7 +/- 25.6 fmol/gland). Hyperthyroid rat pituitaries showed increased TSH beta and neuromedin B immunoreactivities and neuromedin B content (TRH-treated, 385.2 +/- 30.2; T4-treated, 352.6 +/- 20.2; control, 230.7 +/- 25.6 fmol/gland). Neuromedin U immunoreactivity occurred in corticotropes of all species, in rat and mouse intermediate lobe, and throughout the rat hypothalamus, with immunoreactive cell bodies in the arcuate nucleus. Neuromedin U-immunoreactive cells were present in six of six human pituitary and five of six human extrapituitary corticotropinomas. In adrenalectomized rats, corticotropes were larger and more numerous than in controls, but extractable anterior pituitary neuromedin U-like immunoreactivity was not raised (adrenalectomized, 3.30 +/- 0.45; control, 3.32 +/- 0.27 pmol/gland). Our findings suggest that 7B2, neuromedin B, and neuromedin U may be involved in pituitary function.

Identification of bombesin-immunoreactive cells in rat, human, and other mammalian pituitaries, their ontogeny and the effect of endocrine manipulations in the rat.

Bombesin and gastrin-releasing peptide are homologous peptides which have biological activity in mammals. The distribution of bombesin immunoreactivity in rat, guinea pig, cat, dog, pig, cow, monkey, and human pituitaries was investigated using immunocytochemistry with various different antisera. Polyclonal antisera identified bombesin-immunoreactive (IR) cells in the anterior pituitaries of all species except monkey and human, although positive nerves were present in the human posterior lobe. In contrast, a monoclonal antibody demonstrated bombesin-IR cells in anterior and intermediate lobe (or equivalent) of all species. Both types of antibodies identified the anterior pituitary cells as somatotrophs, which may be significant because bombesin and related peptides influence pituitary growth hormone secretion. Differences in bombesin immunoreactivity were seen in male and female rats, with males having more positive cells, and females showing more intense immunoreactivity in those cells which were positive. Ontogenetic studies in rats revealed that bombesin-IR cells were first seen at birth. The effect of estrogen on bombesin-IR cells was studied using ovariectomized and estrogen-treated female rats. Estrogen treatment decreased very significantly the number of bombesin-IR cells, compared with controls, whereas ovariectomy increased significantly the frequency of bombesin-IR cells, so that the staining pattern began to resemble that seen in normal male rats. No bombesin-IR cells were detected in pituitaries from thyroidectomized rats. These results suggest that immunoreactive bombesin/gastrin-releasing peptide in the pituitary is modulated by endocrine status and this peptide may be involved in paracrine interactions in this tissue.

Localization of immunoreactivity for calcitonin gene-related peptide in the rat anterior pituitary during ontogeny and gonadal steroid manipulations and detection of its messenger ribonucleic acid.

Localization of calcitonin gene-related peptide (CGRP) expression in the rat anterior pituitary and its changes during ontogeny and after gonadal steroid manipulations were studied by immunocytochemistry, RIA, and in situ hybridization. Colocalization studies and the combined use of immunocytochemistry and in situ hybridization revealed that CGRP immunoreactivity is localized mainly in gonadotropes and alpha- and beta-CGRP messenger RNAs were detected in CGRP-immunoreactive cells. Immunoreactivity for CGRP also was detected in nerve fibers and colocalized with substance P immunoreactivity. Cells immunoreactive to CGRP antiserum were first detected in fetal rats at gestational day 18, and the incidence considerably increased between postnatal days 5 and 14. CGRP immunoreactivity was low in control adults of both sexes and in pregnant and ovariectomized females but increased in lactating, estrogen-supplemented ovariectomized and high-dose estrogen-treated females, and in high-dose estrogen-treated and castrated males. Testosterone supplement suppressed the effect of castration on CGRP immunoreactivity in males. Quantities of extractable immunoreactive CGRP under conditions of estrogen manipulation corresponded well to the immunocytochemical findings (females: controls, 96.4 +/- 13.1 fmol/gland; ovariectomized, 107.6 +/- 19.2; high-dose estrogen-treated, 212 +/- 23.0; estrogen-supplemented ovariectomized, 680 +/- 42.1). The present study suggests that pituitary CGRP is synthesized and stored in gonadotropes, is modulated by gonadal steroids, and may have a functional link with gonadotropins.

Adrenocorticotropin and lipotropin secretion by dispersed cell cultures of a human corticotropic adenoma: effect of hypothalamic extract, arginine vasopressin, hydrocortisone, and serotonin.

Basal and modulated secretion of ACTH and lipotropin (LPH) by cultures of trypsin-dispersed cells of a biopsy of a human corticotropic adenoma have been examined. ACTH secretion was detectable throughout the period of culture (13 days) but declined steadily from an initial production rate of 238 +/- 124 ng/3 X 10(5) cells/12 h. The time course of secretion showed a slower phase over the first 4 h, with increases up to 12 h. An extract of rat stalk median eminence caused a significant (P less than 0.005) dose-dependent increase in both ACTH and LPH secretion during 30 min. The patterns of response for ACTH and LPH were very similar; both exhibited a decline in the basal release of peptide subsequent to the period of stimulation. The addition of hydrocortisone (0.2 micrograms/ml) did not suppress basal ACTH secretion during 30 min but significantly (P less than 0.05) inhibited stimulation produced by rat stalk median eminence extract. Arginine vasopressin (dose range, 1-9 ng/ml) significantly (P less than 0.025) stimulated both ACTH and LPH secretion during 30 min. The patterns of response were again very similar. Serotonin (dose range, 0.01-10 micrograms/ml) did not affect ACTH secretion during incubations of 30 min to 4 h. The results obtained with the cell cultures of a human corticotropic cell adenoma concur with in vivo findings of incomplete autonomy of secretion, parallel secretion of ACTH and LPH in response to provocative stimuli, and suppression by corticosteroids. The technique has potential for exploring the cellular mechanisms controlling secretion by human corticotropic adenomas as well as the nature of the hormones produced.

The interaction of CGRP and adrenomedullin with a receptor expressed in the rat pulmonary vascular endothelium.

An abundant, seven trans-membrane domain receptor related to the calcitonin receptor has been studied by a number of groups without identification of its ligand. A recent report claimed that the receptor was a type 1 CGRP receptor (Aiyar et al J. Biol. Chem. 271 11325-11329 (1996)). We have studied the equivalent rat sequence in transfected cells. When expressed in 293 cells the receptor interacts with CGRP and adrenomedullin with KD values of 1.2 nM for CGRP and 11 nM for adrenomedullin. Both ligands cause an elevation of intracellular cAMP with EC50 values of 4 nM and 20 nM respectively and these effects are inhibited by the antagonist CGRP8-37. The receptor is expressed at high levels in the pulmonary vascular endothelium. Both the pharmacological data and the localisation are consistent with the conclusion that the orphan receptor is a type J CGRP receptor. However, when expressed in COS-7 cells, no receptor activity could be demonstrated suggesting that 293 cells contain a factor necessary for functional receptor expression.

Localization of human pituitary hormones by multiple immunoenzyme staining procedures using monoclonal and polyclonal antibodies.

Mouse monoclonal and rabbit polyclonal antibodies to human pituitary hormones were applied together to sections of normal and neoplastic human pituitary tissue. Binding sites were revealed with species-specific immune reagents combined with various enzymes (peroxidase, alkaline phosphatase, and beta-D-galactosidase). The enzymes were developed separately to give differently colored end-products. Where two hormones were present in the same cell, a mixed color was produced. Up to four hormones could be immunostained in a single section. Multiple immunoenzymatic staining has great potential for the analysis of plural antigen production by single cells and relationships between cells producing different antigens.

High expression of CD23 in the proliferation centers of chronic lymphocytic leukemia in lymph nodes and spleen.

Chronic lymphocytic leukemia (CLL) is a B-cell neoplasm composed of a heterogeneous mixture of cells, including small lymphocytes, prolymphocytes, and large transformed cells; these last cells appear to represent the proliferating compartment. CLL cells express, in addition to B cell markers, the transmembrane receptor CD23. CD23 functions as the receptor for IgE and also appears to play a role in controlling the growth and proliferation of lymphocytes. Its level of expression among the different cells in CLL has not been examined. In this study, we show that CD23 expression is much higher in the large transformed CLL cells than in the small lymphoid population. This may provide an explanation for the observed correlation between a circulating CD23 cleavage product (soluble CD23) and prognosis in CLL. In addition, we have shown that proliferation in splenic CLL occurs preferentially in the white pulp zones, even in cases in which both the white and red pulp are extensively infiltrated.

False positive immunoreaction in products of conception.

False positive immunostaining for cytomegalovirus in products of conception was revealed using an avidinbiotinylated peroxidase complex. The cause was shown to be endogenous biotin. The use of a non-avidin-biotin method avoided the problem.

Silver enhancement of polymerised diaminobenzidine: increased sensitivity for immunoperoxidase staining.

Unambiguous identification of lymphocytes is sometimes difficult because of weak immunostaining of the cell membrane immunoglobulins. A simple method of intensifying the diaminobenzidine (DAB) peroxidase reaction was therefore devised. Paraffin wax sections of formalin fixed tonsils and lymphomas were digested with trypsin and immunostained for kappa and lambda light immunoglobulin chains and CD3 antigen by various peroxidase linked detection systems. After reaction with hydrogen peroxide and DAB the sections were immersed in methenamine silver solution at 60 degrees C for three to seven minutes. The light brown stain on the cell membranes of the mantle zone lymphocytes became dark brown and the stronger stain of the plasma cells became black. Mantle zone B lymphocytes and CD3 positive T lymphocytes were precisely outlined even at low magnification and the lymphomas were easily classified as monoclonal or polyclonal. At high magnification, staining was clearer than with the immunogold-silver stain. Cryostat and paraffin wax sections of other tissues immunostained for various antigens showed similar intensification. Silver methenamine provides an easy means of increasing the sensitivity and visual impact of an immunoperoxidase/DAB reaction in any preparation.

Inappropriately low serum GH in an acromegalic: lysosomal involvement in intracellular hormone degradation.

A patient with galactorrhea, amenorrhae and severe acromegaly was found to have a large pituitary adenoma. In view of the severity of the disease the serum growth hormone (GH) level (22.5 mlU/liter) was considered inappropriately low. Tissue from the adenoma was obtained during successful treatment with interstitial irradiation. (90Yttrium). Trypsin-dispersed biopsy cells in culture for 12 days secreted low amounts of GH compared to the same number of adenoma cells from 5 other unselected acromegalics or a normal pituitary. No other hormones were secreted in culture. Immunocytochemical staining was positive only with GH antisera but showed low intracellular content. This was confirmed by direct analysis of the tumor tissue which showed the GH content to be only 20% of that found in 5 normal pituitaries and 4% of that found in 8 other adenomas from acromegalics. Electron microscopy showed a striking appearance, with GH secretory granules that were sparse in number, smaller than usual, and in the main arranged around numerous intracellular profiles with double membranes and low electron density that were tentatively identified as autophagic vacuoles (secondary lysosomes). Subcellular fractionation showed the distribution of the radioimmunoassayable GH in the gradient to be coincident with the peak of the lysosomes whereas in 2 other acromegalics the GH peak was clearly separated from the lysosomes. We conclude that the simultaneous appearance in our patient of the relatively low serum GH together with a large tumor and severe acromegaly can be explained biochemically by the striking finding of crinophagy - disposal of hormone secretory granules within the somatotroph cells themselves.

Neuropeptides and 5-HT immunoreactivity in the gastric nerves of the dogfish (Scyliorhinus stellaris).

The gastric autonomic innervation of the dogfish was examined for regulatory peptides and serotonin by immunochemical techniques. Bouin's-fixed, paraffin-embedded or benzoquinone-fixed frozen sections were used for light microscopical immunocytochemistry and glutaraldehyde-fixed resin-embedded sections for electron microscopical immunocytochemistry. Bombesin-, somatostatin-, gastrin/cholecystokinin-, substance P-, peptide histidine isoleucine-, vasoactive intestinal peptide- and serotonin-immunoreactive nerves were found in all layers of the stomach wall. Bombesin and vasoactive intestinal peptide-containing nerves were identified at ultrastructural level. Radioimmunoassay of acetic acid extracts of tissue confirmed the presence of immunoreactivity for bombesin, somatostatin, substance P, peptide histidine isoleucine and vasoactive intestinal peptide. Reverse phase high performance liquid chromatography indicated that the peptides identified were broadly similar to their mammalian counterparts.

The Splenic White Pulp in Chronic Lymphocytic Leukaemia: A Microenvironment Associated with CR2 (CD21) Expression, Cell Transformation and Proliferation.

Splenectomy specimens from 8 cases of chronic lymphocytic leukaemia (CLL) were examined. The infiltrate in the red pulp consisted predominantly of small lymphocytes. In contrast the predominant cells in the white pulp were pro-lymphocytes and para-immunoblasts. The Ki67 marker, which identifies cells in growth phase, was also concentrated among the transformed cells in the white pulp. Staining with monoclonal antibodies for CD21, which identifies the CR2 receptor, showed that the cells in the white pulp were strongly positive in contrast to those in the red pulp which were weak or negative. These findings suggest that factors present in the white pulp promote transformation and proliferation of CLL cells.

Failure of ultrasonic vibration as a means of antigen retrieval in routine diagnostic immunocytochemistry.

It was recently reported that certain antigens can be retrieved from paraffin sections of formalin-fixed tissues by ultrasonication using either an expensive and sophisticated ultrasonic cell disrupter probe (cytokeratins 13 and 16) or an inexpensive and generally available ultrasonic cleaning bath (bcl-1). We wished to investigate the routine use of the latter method and therefore tried to retrieve from various tissues 11 antigens that usually require heat-mediated retrieval in citrate buffer. We applied ultrasonic vibration for periods of 30 seconds to 1.5 minutes in a cleaning bath containing citrate buffer or water, with and without the addition of heat, or for 1 minute in hot citrate buffer after microwaving for 10 minutes in the same buffer. Although a slight effect of ultrasound was noted for a few antigens, in no case did the immunostaining reach the level achieved after standard microwave heating in citrate buffer. We conclude that, under the conditions we used, ultrasonic antigen retrieval cannot be used for immunocytochemistry in a routine histopathology laboratory.

Gastrin-releasing peptide gene products in midtrimester human fetal lung with and without maternal smoking history during pregnancy.

A preliminary morphological study on human fetal lungs with positive maternal smoking history demonstrated alterations of the neuroepithelial bodies (NEBs). We studied human fetal lung tissue between the gestational ages of 12 weeks and 19 weeks, comprising 12 cases with a smoking history during pregnancy (Group 1) and eight cases without a smoking history during pregnancy (Group 2). We demonstrated, by immunocytochemistry, the presence of gastrin-releasing peptide (GRP) gene products: GRP 14-27 in all 20 cases, and C terminal peptide of pro-GRP (C-flanking peptide) in 17 cases. Quantification of the neuroepithelial cells (NECs) was made by computer-enhanced image analysis using the Context Vision system, expressing 1) the total stained areas of the NECs per unit area of section and 2) the total staining areas of the NECs per unit area of airway epithelium, measured as the area of cytokeratin immunoreactivity in an adjacent section. The results show no statistically significant difference between groups 1 and 2 for either GRP 14-27 or C-flanking peptides. The apparent lack of influence of maternal smoking during pregnancy on the expression of GRP gene products in the NECs could be a reflection of inherently reduced reactivity of the cells during the gestation period studied. However, a larger series is needed before any conclusions can be made. Alternatively, the adverse effects of smoking might be reflected during the canalicular phase of lung development; an increased immunoreactivity appears to be present during that period. The expression of pro-GRP gene products in the pulmonary NECs of older fetuses and neonates with maternal smoking history during pregnancy requires further study.

Co-localisation of substance P-, bombesin- and peptide histidine isoleucine (PHI)-like peptides in gut endocrine cells of the dogfish Scyliorhinus stellaris.

The gastrointestinal tract of cartilaginous fishes, like that of higher vertebrates, is known to contain endocrine cells and nerves immunoreactive for a wide variety of peptides, some of which have been structurally characterised. Since we have found that substance P-, bombesin- and peptide histidine isoleucine-like immunoreactivities are similarly distributed in the endocrine cells of the dogfish pyloric stomach, we have tried to establish whether any of these peptides are co-localised. The cells were compared in thin serial sections with both light- and electron microscopical immunocytochemistry. Double immunolabelling was also used to show two immunoreactive peptides in the same tissue section. Further characterisation of the immunoreactivity was attempted by preabsorbing the antibodies with various peptides or synthetic fragments of peptide molecules. Immunoreactivity for all three peptides was frequently present in the same cells, whereas antibodies to other peptides such as gastrin and somatostatin marked different cells. Electron microscopy indicated that all the secretory granules in three morphologically different cell types reacted with antibodies to all three peptides. Dual localisation of unrelated peptides in endocrine cells or nerves is established in many cases, but triple localisation is as yet unusual. The immunoreaction for bombesin-like peptides is different in endocrine cells and nerves, indicating that dogfish bombesin may be present in two forms, in agreement with biochemical evidence.

Rapid flow carbon dioxide laparoscopy disperses cancer cells into the peritoneal cavity but not the port sites in a new rat model.

BACKGROUND: The role of carbon dioxide (CO2) in the pathogenesis of tumor recurrence after laparoscopy remains controversial. Using a new rat model, we studied the effect of different CO2 flow rates on the dispersal of free cancer cells. METHODS: A novel model of desufflation without trocar was developed, and 55 Fischer rats were randomized into three flow groups: group A (rapid, 0.67 l/min; n = 20), group B (slow, 0.44 l/min; n = 20), and group C (gasless, n = 15). We vented CO2 via a portless surgical valve that filtered cells. After the abdominal wall had been suspended to create space, half of the animals in each group (nonrecovery) received 7.5 x 10(6) immunolabeled rat colon cancer cells (RCC2) intraperitoneally, whereas the other half (recovery) received 7.5 x l0(6) viable RCC2 before insufflation or gasless laparoscopy. Nonrecovery animals were killed after 20 l of insufflation. Parietal peritoneal and port-site specimens were examined for RCC2 by fluorescence microscopy (FM) and flow cytometry (FC). The recovery animals were killed at 4 weeks for evidence of wound recurrence. RESULTS: Nine of 10 nonrecovery animals in A had RCC2 on FM or FC, as compared with 2 animals in each of the nonrecovery groups B and C (p = 0.018, Fisher's exact test). Two of the nine animals in group A also had RCC2 in their portless valves. Two recovery (A) animals developed wound recurrence as compared with none in the other groups (p = 0.315). CONCLUSION: In this model, rapid CO2 flow dispersed free cancer cells into the peritoneal cavity, but not into the port sites, thus supporting a role for CO2 in the intraperitoneal dispersal of free cancer cells, but not in wound recurrence.

The oesophagus and stomach of the African elephant: a histological, immunocytochemical and immunofluorescence study.

Histological, immunocytochemical and immunofluorescence methods were employed to study the oesophagus and stomach of the elephant. The histological findings were in line with the situation in monogastric species like swine and man. In the mucosa of the stomach, endocrine cells were immunoreactive to gastrin, somatostatin, chromogranin A and serotonin. Nerve cells immunoreactive to somatostatin, bombesin, VIP, PHI and CGRP were detected in the submucosal and myenteric plexus of the stomach. In the stomach, the absence of glucagon cells and the presence of endocrine cells immunoreactive to PYY, are in contrast to the situation in mammals and need further investigation. Small gastric ulcers were observed in some of the specimens.