Immunoreactivity of gastric ECL and A-like cells in fasted and fed rats and mice.

The oxyntic mucosa of rat and mouse stomach harbors histamine-producing ECL cells and ghrelin-producing A-like cells. The ECL cells are known to be active when the circulating gastrin levels are elevated in response to food intake. The A-like cells are the main source of circulating ghrelin. In response to starvation, the circulating ghrelin is elevated as a hunger signal. The aim of the present work was to study the correlation between the immunoreactivities and cellular activities of the ECL cells and A-like cells. Rats were either fed or fasted for 48 h and mice for 24 h. Immunohistochemical examination with antiserum against chromogranin A-derived fragment pancreastatin revealed both the ECL cells and the A-like cells without a difference between fasted and fed animals. Histamine was limited to the ECL cells with no significant difference between fasted and fed animals. Histidine decarboxylase (HDC) immunoreactivity occurred predominately in the ECL cells of the fed, but not fasted, animals in which the HDC enzymatic activity in the oxyntic mucosa was higher than in fasted animals. Ghrelin immunoreactivity was increased in terms of intensity, but not cell density in fasted animals. Thus, the immunoreactivities of ECL cells and A-like cells might be affected by starvation.     

Physiology of the ECL cells.

The enterochromaffin-like (ECL) cells of the oxyntic mucosa (fundus) of the stomach produce, store and secrete histamine, chromogranin A-derived peptides such as pancreastatin, and an unanticipated but as yet unidentified peptide hormone. The cells are stimulated by gastrin and pituitary adenylate cyclase activating peptide and suppressed by somatostatin and galanin. Choline esters and histamine seem to be without effect on ECL cell secretion. The existence of a gastrin-ECL cell axis not only explains how gastrin stimulates acid secretion but also may help to explore the functional significance of the ECL cells with respect to the nature and bioactivity of its peptide hormone. From the results of studies of gastrectomized/fundectomized and gastrin-treated rats, it has been speculated that the anticipated ECL-cell peptide hormone acts on bone metabolism.     

Hyperplastic proliferations of the ECL cells.

Enterochromaffin-like (ECL) cells are the dominant endocrine cell type in the oxyntic mucosa. Normally regarded as histamine-producing cells, they are exquisitely sensitive to the trophic action of gastrin and undergo a hyperplastic increase in a variety of hypergastrinemic conditions. A hyperplasia-neoplasia sequence of ECL-cell proliferations has been recently proposed, following the realization that increasingly severe degrees of ECL-cell hyperplasias over a period of several years can progress to ECL-cell carcinoids. Such carcinoids arising in patients with chronic hypergastrinemia differ both in their clinical and pathologic profiles from the sporadic carcinoids that occur in normogastrimenic individuals and, therefore, need to be distinguished from them. This distinction is particularly important for their clinical management, since antrectomy appears to be of benefit in ECL carcinoids of hypergastrinemic patients.     

Immunoreactivity of gastric ECL and A-like cells in fasted and fed rats and mice

The oxyntic mucosa of rat and mouse stomach harbors histamine-producing ECL cells and ghrelin-producing A-like cells. The ECL cells are known to be active when the circulating gastrin levels are elevated in response to food intake. The A-like cells are the main source of circulating ghrelin. In response to starvation, the circulating ghrelin is elevated as a hunger signal. The aim of the present work was to study the correlation between the immunoreactivities and cellular activities of the ECL cells and A-like cells. Rats were either fed or fasted for 48 h and mice for 24 h. Immunohistochemical examination with antiserum against chromogranin A-derived fragment pancreastatin revealed both the ECL cells and the A-like cells without a difference between fasted and fed animals. Histamine was limited to the ECL cells with no significant difference between fasted and fed animals. Histidine decarboxylase (HDC) immunoreactivity occurred predominately in the ECL cells of the fed, but not fasted, animals in which the HDC enzymatic activity in the oxyntic mucosa was higher than in fasted animals. Ghrelin immunoreactivity was increased in terms of intensity, but not cell density in fasted animals. Thus, the immunoreactivities of ECL cells and A-like cells might be affected by starvation.     

The biology and pathobiology of the ECL cells.

The enterochromaffin-like (ECL) cells represent the predominant endocrine cell population in the acid-producing part of the stomach of both experimental animals and man. These cells actively produce and store histamine in addition to an anticipated but as yet unidentified peptide hormone and are under the control of gastrin. An acute gastrin stimulus causes exocytosis of the cytoplasmic granules/vesicles (and release of histamine and activation of the histamine-forming enzyme, histidine decarboxylase), while a more sustained gastrin stimulus causes first hypertrophy and then hyperplasia of the ECL cells in the rat (at most, a fivefold increase in the cell number). These effects can be demonstrated following infusion of gastrin or following an increase in the concentration of circulating gastrin of endogenous origin. The growth of the ECL cells reflects an accelerated self-replication rate. As studied in the rat, the self-replication rate is accelerated quite soon after induction of hypergastrinemia (blockade of acid secretion), the rate is maximally elevated within two weeks and then declines to control values at ten and 20 weeks despite the sustained hypergastrinemia. Lifelong hypergastrinemia in rats is associated not only with ECL-cell hyperplasia but also with an increased incidence of ECL-cell carcinoids. Recently, we could show that alpha-fluoromethylhistidine, which is a suicide inhibitor of histidine decarboxylase, effectively depletes the ECL cells of histamine and that the histamine-depleted ECL cells respond to gastrin with hyperplasia in a manner identical to normal ECL cells. Other factors beside gastrin seem to participate in the control of ECL-cell function and proliferation. Although exogenous somatostatin is known to suppress the activity of the ECL cells, we have failed to obtain evidence that the somatostatin cells in the oxyntic mucosa play a role in the physiological control of the ECL cells. The vagus, however, is important for the ability of the ECL cells to respond to gastrin. This conclusion is based on the observation that vagal denervation suppresses the hyperplastic response of the ECL cells to gastrin. Porta-cava shunting, on the other hand, greatly enhances the responsiveness of the ECL cells to gastrin. The mechanism behind this effect is unknown.     

Intracellular signal transduction during gastrin-induced histamine secretion in rat gastric ECL cells

Activation of G_qprotein-coupled receptors usually causes a biphasic increase inintracellular calcium concentration ([Ca²⁺]_i)that is crucial for secretion in nonexcitable cells. In gastric enterochromaffin-like (ECL) cells, stimulation with gastrin leads to aprompt biphasic calcium response followed by histamine secretion. Thisstudy investigates the underlying signaling events in this neuroendocrine cell type. In ECL cells, RT-PCR suggested the presence of inositol 1,4,5-trisphosphate receptor (IP₃R) subtypes1-3. The IP₃R antagonist 2-aminoethoxydiphenyl borateabolished both gastrin-induced elevation of[Ca²⁺]_i and histamine release. Thapsigarginincreased [Ca²⁺]_i, however, without inducinghistamine secretion. In thapsigargin-pretreated cells, gastrinincreased [Ca²⁺]_i through calcium influxacross the plasma membrane. Both nimodipine and SKF-96365 inhibitedgastrin-induced histamine release. The protein kinase C (PKC) activatorphorbol 12-myristate 13-acetate induced histamine secretion, an effectthat was prevented by nimodipine. In summary, gastrin-stimulatedhistamine release depends on IP₃R activation andplasmalemmal calcium entry. Gastrin-induced calcium influx wasmediated by dihydropyridine-sensitive calcium channels that appear tobe L-type channels activated through a pathway involving activation of PKC.

     

Ontogeny of ECL cells in the rat.

ECL cells produce histamine and chromogranin A, and are restricted to the oxyntic mucosa of the stomach. ECL cell ontogeny has been studied in some detail in the rat. Using histidine decarboxylase immunostaining, the first ECL cells can be demonstrated at embryonic day 17. Immunoreactive histamine and chromogranin A appear one day later. At embryonic day 20, the vesicular monoamine transporter type 2 is also present in the ECL cells. Neonatally the ECL cell proliferation is slow; however, one to three weeks postnatally there is a rapid growth of ECL cells to populate the basal half of the glands. Gastrin is known to be an important stimulator of ECL cell activity and growth in the adult rat. As revealed in recent mouse gene knock out models gastrin does not seem to play a role in the early ECL cell differentiation and development.     

Natural history, clinicopathologic classification and prognosis of gastric ECL cell tumors.

A series of 50 gastric endocrine tumors classified according to Rindi et al. [1] comprised 12 small cell neuroendocrine carcinomas (NEC) and 38 ECL cell carcinoids, of which 22 associated with type A chronic atrophic gastritis (A-CAG), eight with hypertrophic gastropathy due to combined Multiple Endocrine Neoplasia and Zollinger/Ellison syndrome (MEN/ZES), and eight sporadic. Variables found to predict tumor malignancy were: size > 2 cm, > 2 mitoses and > 130 Ki67 positive cells/10 high power fields (HPF), grade 2 or 3 histology, angioinvasion, p53 protein nuclear accumulation, and the presence of a single tumor. None of these factors increased significantly the predicting ability of tumor classification itself, although grade 2 + 3 shows 100 percent negative predictive value and Ki67 and angioinvasion 100 percent positive predictive value. When the mostly non-malignant A-CAG and MEN-ZES tumors were analysed against the mostly malignant sporadic and NEC tumors, a positive predictive value of 90 percent and a negative predictive value of 93 percent was obtained. Investigation of a larger tumor series is under way with the aim to develop an optimal model for prognostic evaluation of gastric endocrine tumors.     

Role of gastrin in the development of gastric mucosa,ECL cells and A-like cells in newborn and young rats

Histamine-producing ECL cells and ghrelin-producing A-like cells are endocrine/paracrine cell populations in the acid-producing part of the rat stomach. While the A-like cells operate independently of gastrin, the ECL cells respond to gastrin with mobilization of histamine and chromogranin A (CGA)-derived peptides, such as pancreastatin. Gastrin is often assumed to be the driving force behind the postnatal development of the gastric mucosa in general and the ECL cells in particular. We tested this assumption by examining the oxyntic mucosa (with ECL cells and A-like cells) in developing rats under the influence of YF476, a cholecystokinin-2 (CCK(2)) receptor antagonist. The drug was administered by weekly subcutaneous injections starting at birth. The body weight gain was not affected. Weaning occurred at days 15-22 in both YF476-treated and age-matched control rats. Circulating gastrin was low at birth and reached adult levels 2 weeks after birth. During and after weaning (but not before), YF476 greatly raised the serum gastrin concentration (because of abolished acid feedback inhibition of gastrin release). The weight of the stomach was unaffected by YF476 during the first 2-3 weeks after birth. From 4 to 5 weeks of age, the weight and thickness of the gastric mucosa were lower in YF476-treated rats than in controls. Pancreastatin-immunoreactive cells (i.e. all endocrine cells in the stomach) and ghrelin-immunoreactive cells (A-like cells) were few at birth and increased gradually in number until 6-8 weeks of age (control rats). At first, YF476 did not affect the development of the pancreastatin-immunoreactive cells, but a few weeks after weaning, the cells were fewer in the YF476 rats. The ECL-cell parameters (oxyntic mucosal histamine and pancreastatin concentrations, the histidine decarboxylase (HDC) activity, the HDC mRNA levels and serum pancreastatin concentration) increased slowly until weaning in both YF476-treated and control rats. From then on, there was a further increase in the ECL-cell parameters in control rats but not in YF476 rats. The postnatal development of the ghrelin cells (i.e. the A-like cells) and of the A-like cell parameters (the oxyntic mucosal ghrelin concentration and the serum ghrelin concentrations) was not affected by YF476 at any point.We conclude that gastrin affects neither the oxyntic mucosa nor the endocrine cells before weaning. After weaning, CCK(2) receptor blockade is associated with a somewhat impaired development of the oxyntic mucosa and the ECL cells. While gastrin stimulation is of crucial importance for the onset of acid secretion during weaning and for the activation of ECL-cell histamine formation and secretion, the mucosal and ECL-cell growth at this stage is only partly gastrin-dependent. In contrast, the development of the A-like cells is independent of gastrin at all stages.     

ECL cell morphology.

Using immunohistochemistry at the conventional light, confocal and electron microscopic levels, we have demonstrated that rat stomach ECL cells store histamine and pancreastatin in granules and secretory vesicles, while histidine decarboxylase occurs in the cytosol. Furthermore the ECL cells display immunoreactivity for vesicular monoamine transporter type 2 (VMAT-2), synaptophysin, synaptotagmin III, vesicle-associated membrane protein-2, cysteine string protein, synaptosomal-associated protein of 25 kDa, syntaxin and Munc-18. Using electron microscopy in combination with stereological methods, we have evidence to suggest the existence of both an exocytotic and a crinophagic pathway in the ECL cells. The process of exocytosis in the ECL cells seems to involve a class of proteins that promote or participate in the fusion between the granule/vesicle membrane and the plasma membrane. The granules take up histamine by VMAT-2 from the cytosol during transport from the Golgi zone to the more peripheral parts of the cells. As a result, they turn into secretory vesicles. As a consequence of stimulation (e.g., by gastrin), the secretory vesicles fuse with the cell membrane to release their contents by exocytosis. The crinophagic pathway was studied in hypergastrinemic rats. In the ECL cells of such animals, the secretory vesicles were found to fuse not only with the cell membrane but also with each other to form vacuoles. Subsequent lysosomal degradation of the vacuoles and their contents resulted in the development of lipofuscin bodies.     

Diagnosis and treatment of ECL cell tumors.

The diagnosis of ECL-omas is easy to perform. In patients with Zollinger-Ellison syndrome (ZES), ECL-omas are almost always observed in the setting of multiple endocrine neoplasia type I. In patients without ZES, the first step is to discard non-gastrin-related sporadic ECL-omas whose prognosis is poor. By contrast, prognosis of ECL-omas in patients with ZES or chronic atrophic gastritis is good. Metastases are rare, and tumor-related deaths are exceptional. In both conditions, ECL-omas measuring less than 1 cm should be treated by endoscopic polypectomy and survey. Treatment modalities (surgery, endoscopic polypectomy) for larger tumors are still discussed. The impact of endoscopic ultrasonography on the therapeutic decision has not yet been evaluated. Considering the good prognosis of these tumors, aggressive surgery could be limited to selected patients. Multicentric studies should be undertaken to determine the best treatment modalities.     

A gene expression fingerprint of mouse stomach ECL cells

Many of the endocrine cells in the stomach are poorly characterized with respect to physiological significance. In some cases, the anticipated hormone has not yet been identified. Global gene expression analysis of mouse stomach was performed in an attempt to identify the ECL-cell peptide/protein. Specific functional activation (omeprazole-induced hypergastrinaemia) was used as a tool to generate a gene expression fingerprint of the ECL cells. The proposed fingerprint includes 14 genes, among them six are known to be expressed by ECL cells (=positive controls), and some novel ones, which are likely to be ECL-cell-related. The known ECL-cell-related genes are those encoding histidine decarboxylase, chromogranin A and B, vesicular monoamine transporter 2, synaptophysin, and the cholecystokinin-B receptor. In addition, the fingerprint included five genes, which might be involved in the process of secretion and three ESTs with unknown function. Interestingly, parathyroid hormone-like hormone (Pthlh) was identified as a candidate ECL-cell peptide hormone.     

Functionally impaired, hypertrophic ECL cells accumulate vacuoles and lipofuscin bodies

ECL cells in the oxyntic mucosa of stomach control gastric acid secretion by mobilizing histamine in response to gastrin. They respond to gastrin also with hypertrophy and hyperplasia. ECL cells exhibit functional impairment upon long-term gastrin stimulation. The impairment is manifested in a gradual decline of the activity of the histamine-forming enzyme per individual ECL cell and in a failure of gastrin to mobilize histamine. The mechanism behind this impairment is unknown. In the present study, rats were treated with the proton pump inhibitor pantoprazole for 45 days to induce sustained hypergastrinemia. The ECL cells were isolated from normogastrinemic and hypergastrinemic rats and size-separated from other mucosal cells by the elutriation technique. The total ECL cell number was twofold higher in hypergastrinemic rats than in normogastrinemic rats, and most of the cells appeared in elutriation fractions where large cells predominate. The ECL cells of the different fractions were analyzed by quantitative electron microscopy. Normal-sized ECL cells from hypergastrinemic rats displayed a reduced number of secretory vesicles (probably because of degranulation) compared with normal-sized ECL cells from normogastrinemic rats. Hypertrophic ECL cells from hypergastrinemic rats had an unchanged number of secretory vesicles, supporting the view that such cells fail to respond to gastrin with degranulation. Although both normal-sized and hypertrophic ECL cells from hypergastrinemic rats contained vacuoles, those in the hypertrophic ECL cells were larger and more numerous. In addition, hypertrophic ECL cells were found to contain numerous, prominent lipofuscin bodies which are the presumed end product of crinophagia. Conceivably therefore, large vacuoles and lipofuscin bodies cause functional impairment of the hypertrophic ECL cells.     

Pathogenesis of ECL cell tumors in humans.

In ECL cell tumors developed in the setting of hypergastrinemic conditions (ECL cell carcinoids type 1 and 2), hypergastrinemia is the dominant agent acting as a promoter in all steps (hyperplasia-dysplasia-neoplasia) of the tumorigenic sequence. In contrast, it apparently lacks transforming properties as shown by the absence of ECL cell carcinoids in patients exposed to hypergastrinemia alone, i.e., those with sporadic Zollinger-Ellison syndrome. The potential transforming factors include: the allelic loss of the MEN-1 suppressor gene in the genetically predisposed MEN-1 patients, an alteration that may induce ECL cell tumors even in the absence of hypergastrinemia; the still unknown factor(s) associated with atrophic corporal gastritis; agents whose role in the induction of human ECL cell tumors is still unclarified, such as basic Fibroblast Growth Factor, human Chorionic Gonadotropin-alpha and Transforming Growth Factor-alpha; and agents having a favoring role on the ECL exposure to mitogens such as BCL-2. No information is currently available on the pathogenesis of gastrin-independent, sporadic ECL cell carcinoids (type 3) or of gastric neuroendocrine carcinomas.     

Calbindin 28 kDa in endocrine cells of known or putative calcium-regulating function. Thyro-parathyroid C cells, gastric ECL cells, intestinal secretin and enteroglucagon cells, pancreatic glucagon, insulin and PP cells, adrenal medullary NA cells and some pituitary (TSH?) cells

The distribution of calbindin in some endocrine glands (thyroid, parathyroid, ultimobranchial body, pituitary and adrenals) and in the diffuse endocrine cells of the gut and pancreas has been investigated immunohistochemically using an antiserum raised against the 28 kDa calbindin from chicken duodenum. The identity of calbindin-immunoreactive cells in a number of avian and mammalian species was ascertained by comparison with hormone-reactive cells in consecutive sections or by double immunostaining of the same section with both calbindin and hormone antibodies. Calcitonin-producing C cells of the mammalian and avian thyroid, parathyroid or ultimobranchial body, PP, glucagon and insulin cells of the mammalian and avian pancreas, enteroglucagon cells of the avian intestine, secretin cells of the mammalian duodenum, histamine-producing ECL cells of the mammalian stomach, as well as noradrenaline-producing cells of the adrenal medulla and some (TSH?) cells of the adenohypophysis were among the calbindin-immunoreactive cells. Although some species variability has been observed in the intensity and distribution of the immunoreactivity, especially in the pancreas and the gut, a role for calbindin in the mechanisms of calcium-mediated endocrine cell stimulation or of intracellular and extracellular calcium homeostasis is suggested.     

Effect of cholecystokinin-2 receptor blockade on rat stomach ECL cells

The ECL cells in the oxyntic mucosa of rat stomach produce histamine and chromogranin A-derived peptides such as pancreastatin. The cells respond to gastrin via cholecystokinin-2 (CCK2) receptors. A CCK2 receptor blockade was induced by treatment (for up to 8 weeks) with two receptor antagonists, YM022 and YF476. Changes in ECL-cell morphology were examined by immunocytochemistry and electron microscopy, while changes in ECL cell-related biochemical parameters were monitored by measuring serum pancreastatin and oxyntic mucosal pancreastatin, and histamine concentrations, and histidine decarboxylase (HDC) activity. The CCK2 receptor blockade reduced the ECL-cell density only marginally, if at all, but transformed the ECL cells from slender, elongated cells with prominent projections to small, spherical cells without projections. The Golgi complex and the rough endoplasmic reticulum were diminished. Secretory vesicles were greatly reduced in volume density in the trans Golgi area. Circulating pancreastatin concentration and oxyntic mucosal HDC activity were lowered within a few hours. Oxyntic mucosal histamine and pancreastatin concentrations were reduced only gradually. The CCK2 receptor blockade was found to prevent the effects of omeprazole-evoked hypergastrinaemia on the ECL-cell activity and density. In conclusion, gastrin, acting on CCK2 receptors, is needed to maintain the shape, size and activity of the ECL cells, but not for maintaining the ECL-cell population.