La COPROLOGIE sur le Web
Inflammatory diseases of the intestine.

Index de la page .

The research was on identifying genes coding for the various human intestinal mucins (the MUC 1 to 7 family) using molecular-biology techniques, regulating mucin synthesis and secretion under various pathological conditions such as chronic inflammatory diseases of the intestine (ulcerative colitis and Crohn's disease) or cancers (pancreas, stomach, colon, etc.), as well as mucin's involvement in gallstone formation, and its role in the prevention of gastric ulcers or Helicobacter pylori infection. To sum up, cytoprotection of the mucosa seems to be one of the main physiological functions of the digestive mucus.

OESOPHAGOPROTECTION


The protecting of the oesophageal mucosa from acid attack on gastric reflux. They underlined the importance of the saliva's buffer value and of oesophageal mucus pH in preventing the possible harmful effects of hydrochloric acid.

- The impact of chewing-stimuli (a "parafilm" ball) on the salivary secretion of mucins in patients with gastro-oesophageal reflux (with or without complications: strictures, Barrett's syndrome) and healthy volunteers. They showed that the chewing-stimulus multiplied not only the volume of saliva by two, but also the quantity of mucins produced, to a comparable degree both in healthy subjects and patients presenting a gastro-oesophageal pathology, complicated or otherwise. On the other hand, the oesophago-salivary reflex which varies the volume of saliva during mechanical and/or chemical stimulation of the oesophagus was seriously disturbed in patients with gastro-oesophageal reflux. Given the importance ascribed to the salivary components in oesophagoprotection and the deficit in salivary response to oesophageal stimuli in patients with reflux, the authors suggest the use of non-nutritional chewing gum as a complement to treating reflux oesophagitis.

- The importance of the mucous layer's hydrophobic character in maintaining the integrity of the oesophageal epithelium in cases of gastro-oesophageal reflux. It show that, in healthy subjects, the hydrophobicity of the oesophageal secretion obtained in response to HCl (pH 2.1) was significantly lower than that obtained by intraluminal instillation of physiological solution. The drop was even greater when HCl was instilled with pepsin (0.5 mg/ml; pH 2.1). Changes to the oesophageal secretions' hydrophobic properties could therefore play a pathogenic role in the development of mucosal lesions.

- It show a reflex increase in salivary mucin levels, largely from the parotid glands, after both mechanical (intubation or intraluminal distension) and chemical (HCl or HCl/pepsin) intraoesophageal stimulation, the latter representing the main nociceptive stimulus in gastro-oesophageal reflux. Similar trial among patients with severe reflux oesophagitis where they demonstrated serious functional deterioration of the oesophageal mechanoreceptors which regulate the rate of salivary mucin production, resulting in a lower production of mucins in response to a mechanical or chemical nociceptive stimulation.

- The importance of salivary pH which determines the pH at the surface of the oesophageal mucus. Because of its buffer capacity, salivary secretion plays a very active role against intraluminal hyperacidity resulting from gastro-oesophageal reflux. The saliva's pH following oesophageal stimulation by HCl or HCl/pepsin in asymptomatic subjects and in patients with reflux to greater or lesser degrees of seriousness. Among the latter, those who had the most serious attacks also exhibited a significant drop in their salivary pH. The authors concluded that salivary pH could be a key factor in certain complications associated with gastro-oesophageal reflux.

- Yang et al. (Iowa City, USA) (1636) studied the mediators and inhibitors of oesophageal mucus secretion induced experimentally by capsaicin - used as a model of nociceptive stimulus in the anaesthetised dog in vivo. They demonstrated that capsaicin (5 ml, 0.5% solution, intraoesophageal) increased oesophageal mucus secretion as much as substance P (1 µg/kg, IV) or bethanechol (30 µg/kg, IV). The effect of capsaicin was blocked by lidocaine (50 ml, 4%, intraoesophageal), suggesting that the mucosal secretory response to the nociceptive stimulus involved the afferent nerves. Confirmation of the hypothesis would come from the effects of capsaicin being blocked by CP 96 345, an antagonist of NK1 receptors of substance P, and by atropine, a muscarine antagonist. It therefore seems that oesophageal mucus production would, in the dog at least, be increased by antidromic stimulation of the afferent nerves, releasing substance P, an axon reflex of one or more cholinergic synapses.


THE STOMACH AND MUCUS Mucin expression and synthesis

- Ichikawa et al. (Sagamihara, Japan) studied the regulatory mechanisms of mucin biosynthesis, involving growth factors and/or stimulation of acid secretion, on an in vitro model of the gastric mucosa (corpus and antrum) in the Wistar rat. Mucin synthesis was measured by cultured explants in the presence of growth factors such as EGF, TGF-beta, HGF, or bFGF, or acid secretagogues such as histamine, carbachol or gastrin. They found that the two families of substances had different effects on mucin synthesis in the gastric corpus and antrum tissues. Not only EGF and HGF, but also histamine increased mucin synthesis in the corpus explant, while bFGF and TGF-beta had no effect. Carbachol stimulated mucin production in both regions, whereas the stimulant effect of gastrin was limited to the oxyntic region. They concluded that mucin synthesis in the two regions was governed by different regulatory mechanisms.

- Okazaki et al. (Koshi, Japan) examined the in vitro inhibitory effects of dexamethasone (corticoid) on mucin biosynthesis and expression of the MUC1 gene by the human stomach. Assaying the mucins on samples of gastric mucosa taken from 6 subjects undergoing gastrectomy, and cultured with different concentrations of dexamethasone, showed that the corticoid had a dose-dependent inhibitory effect on not only mucin secretion, but also MUC1 gene expression. The result confirmed that one of the steroid's harmful effects on the gastric mucosa could be its inhibition of gastric mucus production.

- In humans, expression of a gene known as MDR1, coding for glycoprotein P (P-Gly) found in the membrane of various cells including liver tumour cells, is elicited by anticarcinogenic agents. Hall et al. (Houston, USA) studied the expression and role of P-Gly in phospholipid secretion by the rodent (mouse and rat) liver and gastric mucosa. They showed that P-Gly was expressed in the superficial cells of the gastric mucosa, i.e. those which secrete or are directly exposed to the acid, and not in the lamina propria or myocytes. The location points to a possible involvement of P-Gly in the secretion of gastroprotective phospholipids.

- Konda et al. (Tokyo, Japan) examined the role of the protease furine in the differentiation process of gastric mucosal cells from undifferentiated stem cells. They found that furine, an endopeptidase involved in the cleavage of protein precursors into active proteins, disappeared when a line of gastric surface mucous cells (GSM06) differentiated in culture. By transfecting either cDNA or the corresponding anti-sense, they found that cells expressing furine secreted less mucin and, vice versa, those not expressing it increased their mucus production. Furine could thus play a role in keeping cells in an undifferentiated state.

- Yamada et al. (Kyoto and Tokyo, Japan) showed that EGF (Epithelial Growth Factor) stimulated the proliferation and differentiation of gastric mucosa cells in culture into epithelial cells. On the other hand, on the same cell cultures, TGF-beta1 completely inhibited the EGF-elicited proliferation of mucus-secreting cells and synthesis of mucins. Inhibition by TGF-beta1 could act by eliminating the induction of early gene expression by EGF.

Gastroprotection

Work on gastroprotection concentrated more on the pharmacological and therapeutic aspects of ulcer treatment, but also on trying to explain the factors aggravating tissue attack.

- To compensate for the shortage of human in vitro models in long-term studies of gastric mucus secretion, Wagner et al. (Hanover, Germany) developed a primary culture system of epithelial cells isolated by collagenase/pronase from samples of human gastric mucosa. The culture, considerably enriched in mucus-producing cells (90 to 95%), stayed functional with no loss of secretory efficacy for 10 to 14 days. This, the authors demonstrated by stimulating mucus secretion by gastrin, prostaglandin E2 or carbachol. According to the authors, the study was the first to show that gastric mucus secretion in human cell cultures is regulated by gastrin.

- Synnerstad and Holm (Uppsala, Sweden) started from the premise that the internal pressure of gastric mucosa glands increases on stimulation of acid secretion. In the anaesthetised rat, high doses of omeprazole in long-term (400 µmoles/kg/day per os for 15-30 days) or acute treatment (40 µmoles/kg, IV), blocked acid secretion and raised the intraglandular pressure with no change to the thickness of the mucous gel adhering to the epithelial surface. Omeprazole would thus act on the mechanisms regulating the glands' hydrostatic pressure by limiting outward flux, and not by increasing the thickness of the adherent mucus.

- Systemic alkalosis or topical administration of prostaglandin E2 (PGE2) protects the gastric mucosa against luminal hyperacidity. Ranta-Knuutila et al. (Helsinki, Finland) tried to find out whether the protection was related to an increased secretion of HCO3- by the mucosal epithelium, the effect of which would be to render the pre-epithelial mucus alkaline and increase its buffer capacity. Rats treated with PGE2 and rats exhibiting HCO3 hyperalkalinaemia tolerated a luminal acidification resulting from an perfusion of HCl (0, 10, 30, 50 and 100 mM) significantly better than their controls. On the other hand, tissue resistance to the acid was low in the hyperalkalinaemic rats.

- Tanoue et al. (Long Beach and Irvine, USA and Fukuoka, Japan), after immunohistochemically labelling gastric mucosal samples from rats with portal hypertension, used an image analysis system to visualise the gastroprotective effects of teprenone against aggression of the mucosa by ethanol (8 ml/kg, intragastric). Teprenone is widely used in Japan for treating gastric ulcer and inflammation. It would seem that teprenone effectively protects the gastric mucosa from the effects of ethanol by acting on various mechanisms such as increasing the gastric blood flow, but also on the secretion of gastric mucus, growth factors (basic fibroblast growth factor) and prostaglandins.

- Following the stimulation of acid secretion, caffeine can cause gastric lesions. On the other hand, nothing is known as to the potential effects of caffeine on mucus secretion by the gastric epithelial cells. Hamada et al. (Tokyo and Osaka, Japan) examined the effects of caffeine on JR1, a mucus-producing epithelial cell line derived from human gastric carcinoma. They observed that caffeine selectively counters the increase of intracellular calcium induced by agents like acetylcholine, histamine or ATP, probably by blocking the hydrolysis of PIP2 into IP3, resulting in a drop of mucus production. It would thus seem that caffeine induces mucosal lesions, not only by increasing acid secretion, but also by decreasing gastric mucus production.

- Working on the intracellular level (Golgi preparations with fetuin, a positive substrate, and creatine kinase, a negative substrate), Slomiany et al. (Newark, USA) investigated the mechanism of gastric mucosal lesion by ethanol. They showed that, under ethanol, modified mucins with different substituents in their carbohydrate chains were synthesised. They also observed substantial changes to the glycosylation of proteins other than mucins which could, in cases of chronic alcoholism, have serious consequences on the immunological and inflammatory responses.

- Iaquinto et al. (Pisa, Italy) demonstrated that the gastroprotective effects of 4-methylpyrazole (4-MP), known to be an alcohol dehydrogenase inhibitor, were due neither to cessation of acid secretion in the stomach, nor to biosynthesis of prostaglandins by the gastric mucosa. The authors therefore put forward the hypothesis, and confirmed experimentally in the rat, that the compound's mode of action could involve stimulating secretion of the mucus adhering to the gastric mucosa. The effectiveness of 4-MP in gastroprotection would thus be related to an increased secretion of mucus, yet another illustration of its cytoprotective function.

- Takahashi et al. (Kyoto, Japan) underlined that not all gastroprotective effects seemed to involve the gastric mucus itself. For example, the anti-ulcer protective effects of leminoprazole (NC-1300-O-3), a proton-pump inhibitor, on rabbit gastric mucosa cell culture attacked by indometacine did not seem to be due to action on the mucus and/or prostaglandins E2, but to involve the synthesis of specific proteins making the cells resistant to the harmful effects of indometacine.

- On the other hand, under basal conditions on the same cell model (Takahashi et al., Kyoto, Japan), leminoprazole was able to stimulate mucus synthesis and secretion directly, independent of its effects on the proton-pump, the stimulant effect being mediated by the nitric oxide produced by a constitutive nitric oxide synthase.

- Tanaka et al. (Los Angeles, USA) studied the possible relations between how gastrin and prostaglandins work in gastroprotection. Pentagastrin strengthens the mucosal defence barrier in vivo, mainly by increasing the thickness of the adherent mucous gel. These effects are reproduced by exogenous prostaglandins. Consequently, the authors blocked the synthesis of endogenous prostaglandins by indometacine and measured the intracellular pH and thickness of the gastric mucus after pentagastrin. They showed that indometacine did not affect the increase in acid secretion due to pentagastrin, but did partially prevent the increase in mucus thickness. They thus demonstrated that strengthening the mucous barrier by pentagastrin only partially resulted in the synthesis of endogenous prostaglandins.

- The same team studied the effects of the ranitidine-bismuth citrate complex (GR 122,311X, Glaxo) on the permeability of the adherent mucous gel. They measured the initial level of acidification in the rat gastric mucosa, indicating the mucus's permeability to acid. They showed that GR 122,311X made the mucus more impermeable to acid, almost certainly due to the bismuth binding with the mucins.

- Bismuth binding with mucus in the gastric wall was also examined by O'Brien et al. (Cincinnati, USA). The bonding was very fast (< 1 mn) and stable (> 2 h) on a tissue model treated by bismuth subsalicylate in an Ussing chamber, essentially involving the metabolite bismuth oxychlorhydrate. The observations could be used in the therapeutic applications of bismuth in treating dyspepsia and/or eradicating Helicobacter pylori.

- Lichtenberger et al. (Houston, USA) studied the effects of naproxen, a non-salicylated NSAID, on the hydrophobicity of the gastric mucosa, and more particularly its molecular interactions with surface phospholipids such as DPPC (dipalmitoyl phosphatidylcholine). It would seem that the ability of NSAIDs to cause lesions of the gastro-intestinal tract could be related to their properties of rapid association with phospholipids on the mucosal surface.

- Teprenone is an anti-ulcer agent which has been shown to protect the gastric mucosa against lesions produced by irritants. Hirakawa et al. (Tokushima, Japan) have developed a cell culture system in which cell growth can be blocked and induction of mucin synthesis suppressed. They thus demonstrated that the cytoprotective effects of teprenone act via the heat-shock proteins secreted by the gastric epithelial cells, independently of mucin synthesis.

- The protective efficacy of antacids on the gastric mucosa is due to their ability to neutralise HCl, although other mechanisms of action have also been suggested recently. By comparing hydrotalcit with Talcid (a polymer suspension of the former), Gruebel et al. (Boston, USA) demonstrated that the polymers used in antacid formulations could play a major part in gastroprotection and the remission of ulcers, due either to their own properties (comparable visco-elasticity to that of mucins), or by interaction with the gastric mucins (increasing their viscosity).

Mucus and Helicobacter pylori (Hp)

- Taha et al. (Glasgow, United Kingdom), showed significant variations in the concentration of copper, zinc and iron ions in the gastric epithelium and in the mucus according to the presence or not of Hp in 52 patients. For example, the copper/nitrogen concentration ratio in the gastric epithelium rose from 0.01 during Hp infection to 0.16 after its eradication. The same ratio dropped from 0.12 to 0 in the mucus. Under the same conditions, the relative concentrations of iron and zinc did not change significantly. Hp gastritis is thus associated with a relative deficiency of copper (known, like zinc, to have anti-inflammatory properties) in the gastric epithelium, probably by the metal's diffusion into the mucus.

- Flemström et al. (Newcastle, United Kingdom) wanted to ascertain whether the mucous layer adhering to the gastroduodenal mucosa was a barrier to high-molecular-weight toxins migrating from the lumen to the surface of the duodenal epithelium. On an in situ preparation of anaesthetised rat duodenum, they observed that perfusing a suspension of Hp or cholera toxin (CT) only elicited an increase of bicarbonated secretion if the adherent mucous gel had been eliminated by papain beforehand. If the mucous gel was intact, Hp or CT were virtually incapable of reaching the epithelial surface.

- Degradation of soluble mucins by Hp could be a major factor in the development of gastritis and gastroduodenal ulcers. Byrd et al. (Detroit, USA) compared the purified soluble mucins from the gastric juices of patients infected (Hp+) or not (Hp-) with Hp. They showed that there was no reduction in the concentration, carbohydrate content or size of the soluble mucins in the Hp+ patients. However, the same mucins bound more than those of the Hp- patients to the monoclonal antibodies PS10 (pancreatic cancer) and GS2 (gastric cancer), which recognise non-sialile carbohydrate epitopes on the mucins.

- It has been shown that one result of ingesting ethanol is a reduction in thickness of the adherent mucous gel. Since Hp mainly attaches itself beneath the mucous layer, Sarma et al. (Newington and Farmington, USA) investigated the effects of regular ingestion of alcohol on frequency of Hp infection. The frequency of Hp infection was determined (CLO test and histology) on biopsies of 55 gastritic patients, classed into groups according to degree of alcohol consumption: a significant reduction in Hp frequency was observed in patients who drank the most. Ethanol thus seemed to alter the gastric surface's milieu making it unsuitable for Hp colonisation.

- Tanaka et al. (Los Angeles, USA) showed that indometacine increased the bismuth concentration in the mucus of rats treated with ranitidine-bismuth citrate (GR 122,311X). Their objective was to maintain sufficient antibacterial concentrations to eradicate Hp, without having to perform multiple administrations. Under indometacine, the concentration of bismuth fixed onto the gastric mucus was ten times higher than in the control untreated by the anti-inflammatory. The effect could be the result of inhibition of pepsin activity by the bismuth and stimulation of mucus synthesis by indometacine. Combining GR122,311X with an NSAID would therefore prove very useful in eradicating Hp, without antibiotics being added.

- Logan et al. (Nottingham, USA) sought to understand why Hp only infected gastric type epithelia. They therefore studied the chemotactic response of Hp to gastric mucus and acid. To do so, they used isolates of H. pylori and H. mustelae in culture, and had them rise inside capillary tubes containing chemotaxins. All isolates tested exhibited optimal chemotactism towards slightly acidified mucus (pH 5.7-6.0) which, given the conditions in vivo, could contribute to Hp's gastric epithelial tropism.

- Lastly, Gibson et al. (Washington, USA and Lausanne, Switzerland) established a cell line, HAE, which they describe as an excellent cell model for studying Hp's pathobiology. The HAE cells, obtained by collagenase digestion of human gastric biopsies, were immortalised by introducing the gene coding for the antigen T SV40. They secrete a mucus consisting essentially of neutral and some acid mucins. When Hp was added to the culture medium, it very quickly adhered to the HAE cells and specifically inhibited their growth.


MUCUS AND HEPATIC DUCTS


Mucus and lithogenesis

- The bile of patients with calcified cholesterol stones is not very acid. Although the reason for this lack of acidity is not known, it is known that calcification by CaCO3 makes the stones insensitive to all acid therapy. Tomida et al. (Tsukuba and Ibaraki, Japan) looked at the hypothesis whereby high bile pH could be due to hypersecretion of mucus glycoproteins which, by an increased buffer effect, would prevent normal acidification of the bile. Be this as it may, their research showed that the mucin levels of patients with calcified or non-calcified cholesterol stones, or even with bile-pigment stones did not differ. Similarly, no correlation was found between bile pH and mucin concentration. Consequently, it would seem that insufficient acidification of the bile is due to the gastric mucosa not secreting enough acid.

- Shiffman et al. (Richmond, USA) considered the hypothesis of ursodesoxycholic acid (UDCA) having an inhibitory action on the concentration of mucins in the bile. UDCA is acknowledged to dissolve cholesterol stones and, moreover, to prevent their formation during the rapid weight loss process. The authors treated 21 patients from a group of subjects with symptoms of chronic cholelithiasis, confirmed by echography, with UDCA (300 mg/day, for 15-30 days). A significant drop in bile mucin concentration was observed in the patients, very probably due to a decrease in mucin secretion by the gallbladder epithelium caused by the UDCA.

- Working on canine gallbladder epithelial cell cultures (DG8E), Klinkspoor et al. (Amsterdam, Netherlands and Seattle, USA) studied the mechanisms by which bile salts, cholesterol crystals and phospholipids could regulate the secretion of mucins. They concluded that it was indeed the bile salts and not the cholesterol, cholesterol crystals or the phospholipids, which stimulated the secretion of mucins by the gallbladder. Here, the hydrophobic salts were more powerful stimulants than the hydrophilic. Unlike prostaglandin E2, bile salts did not stimulate the synthesis of mucins by an increase in intracellular cAMP. Their mechanism of action remains to be determined.

- Eder et al. (Munich, Germany) considering the lithogenic role of mucins secreted by the bile, examined the physiopathological mechanism behind the induction of mucin hypersecretion observed in patients with lithiasis. Adding lithogenic bile from these patients to cultured human gallbladder epithelium cells, they found a significant increase (2 to 3 times more than the control) in mucins produced. Adding bile-pigment stones to the same preparations only elicited a slight increase in mucin secretion. Nevertheless, the authors put forward no hypothesis as to what specific ingredient of lithogenic bile is responsible for the mucosal reaction.

- The American prairie dog (Cynomys sp.) has the particularity of developing lithiases very soon after being fed on a cholesterol-supplemented diet. It is therefore a useful model for studying cholelithiases. Fisk et al. (Nashville, USA) devised a culture technique for Cynomys gallbladder epithelial cells allowing them to examine the role of mucins in the development of gallstones on the cellular and molecular levels by immunohistochemical methods. Working on the same species, Johnston et al. (Baltimore, USA) (169) researched the possible role of the non-mucus glycoproteins which very soon appear in Cynomys bile when its diet is cholesterol rich. They suggest that these particular glycoproteins could be major early pathogenic factors in the lithogenic process.

- Lechene de la Porte et al. (Marseille, France and Amsterdam, Netherlands) examined the location of protein molecules inside mixed gallstones (made up of pigments, calcium salts and cholesterol crystals). They located the mucins inside the stones using specific monoclonal antibodies. The specific arrangement of mucins demonstrated their structural role in stone formation. Thus, the calcium salts and pigments deposit onto a protein matrix (APF/CBP and aminopeptidase-N) linked to the mucins, and cholesterol crystallises onto the mucins directly.

- Rege and Prystowsky (Chicago, USA) examined the inflammatory properties of bile in dogs with pigment lithiases induced experimentally by a methionine-deficient diet. The diet caused inflammation of the gallbladder and hypersecretion of bile mucus. One millilitre of this bile caused inflammation of the gallbladder of healthy guinea pigs, and the effect persisted after ultracentrifuging the bile to eliminate crystals and stones. The bile supernatant thus contained soluble factors able to elicit inflammation of the gallbladder wall.


MUCUS AND INFLAMMATION OF THE GALLBLADDER


- Epitope CA19-9 is a tumoral marker whose serum concentrations increase in patients affected with gastro-intestinal tumours. Although CA19-9 is naturally present in the healthy human bile, it is not in the colon or pancreas. Von Ritter et al. (Munich, Germany) investigated why this epitope was present in the bile. Having revealed mucins with the epitope on epithelial cell cultures of normal human gallbladder, they concluded that the rise in serum epitope concentration observed in inflammation of the bile tract could be due to the mucins passing from the bile to the blood.

- Inflammation of the gallbladder mucosa results in hypersecretion of mucins, which in turn promotes the lithogenic properties of bile. Eder et al. (Munich, Germany and Seattle, USA) demonstrated that the polynuclear neutrophils activated during the inflammation process stimulated mucin secretion by gallbladder epithelial cells in the dog. The inhibitory effects of catalase and superoxide dismutase in the model indicated that the hydroxyl radicals (active metabolites secreted by the polynuclear neutrophils) could mediate the hypersecretion.


INTESTINAL OR COLON MUCINS

Synthesis and regulation - Trefoil peptides

- ITF (intestinal trefoil peptide : intestinal) and SP (spasmolytic peptide : antral) "trefoil peptides" are a recently recognised family of epithelial regulatory peptides associated with mucins. They are found widespread in gastro-intestinal tissues and are implicated in protecting and repairing the gastric mucosa.

- The ITF and SP "trefoil peptides" are conserved and potently expressed in the gastro-intestinal region during development. Using radio-immunological and immunohistochemical techniques, Taupin et al. (Melbourne, Australia) quantified expression of the two peptides in the rat between the 17th day of foetal life (gestation: 22 days) and weaning (16 to 27 days post-partum). They demonstrated the presence of appreciable quantities of "trefoil peptides" long before the morphological or cytochemical appearance of traces of mucins or gastrin. The trefoil peptides' concentration peak would thus be just before weaning.

- IGFBP-4 is the binding protein for cytokine IGF which stimulates the proliferation of intestinal mucosa in vivo and numerous cell lines in vitro. Human colon carcinoma line HT29, for example, normally expresses IGFBP-4. By transfecting in these cells, Corkins et al. (Omaha, USA) (1401) caused protein IGFBP-4 to superexpress (multiplication by 2-3). They found that a) the rate of proliferation of transfected HT29 cells was significantly higher than that of non-transfected cells, and that b) their MUC3 mRNA production increased (360%) compared with controls, whereas the MUC2 mRNA rate did not change.

- Ogata et al. (Boston, USA) examined the promoter region of the gene coding for the ITF trefoil peptide specific to colon and small intestine goblet cells in the rat. They identified a 37-nucleotide sequence (-153 to -116) which seemed determinant for ITF gene expression in these cells.

- Genes MUC1 to MUC7 have been described in man. Among them, MUC2 and MUC3 have been cloned from intestinal cDNA libraries. MUC2 is the commonest mucin in the colon. Van Klinken et al. (Amsterdam, Netherlands) studied the expression of MUC2 and MUC3 mRNA and protein synthesis in the human small intestine. They showed that both genes were expressed there and resulted in the synthesis and secretion of well-characterised mucins. Nevertheless, a third form of mucin, as yet unidentified, of approx. 500 kDa was also revealed.

- Hunninghake et al. (Minneapolis and Athens, USA) tried to ascertain whether there was a range of mucin genes in the mouse. By hybridising and cloning, they identified a specific colon mucin different to that which expresses itself in the mouse stomach.

- Cohen et al. (Cincinnati and St Louis, USA) studied the role of guanyline on rat and human intestinal cell lines (Caco-2). They concluded that, in vivo, the function of the mechanisms implicated by guanyline on the intestinal cell level was to ensure hydration of the mucins secreted by the goblet cells.

- Previous papers of Castagliuolo et al. (Boston, USA) have demonstrated that stress due to immobilisation elicited mucin synthesis in the rat colon, along with an increase of colon motricity and prostaglandin synthesis. They now examined the roles of the nervous system, neurotensin and substance P on the effects of stress on the colon. They blocked the effects by a ganglioplegic, hexamethonium. More specifically, they inhibited the production of mucins and PGE2 elicited by stress by means of either neurotensin receptor antagonists (SR 48692), or antineurotensin antibody. Substance P did not seem to be implicated. They raised questions as to the existence of neural pathways involving neurotensin in the colon mucosa's response to stress.

- Until now, it was thought that the cholera toxin (CT) elicited mucin secretion indirectly via the enteric nervous system. Epple et al. (Berlin, Germany) wished to check the assertion by testing the effects of CT on highly differentiated cell cultures of human colon carcinoma (HT29/B6). Mucin secretion was increased in these cells by a factor of 2.6 by CT and by a factor of 8 by forskoline compared to non-stimulated controls. The authors also demonstrated that it was indeed mucins synthesised de novo, unlike what is observed with carbachol on the same preparations, which only elicits the release of small quantities of probably new-formed mucins. There is thus a direct mechanism whereby the cholera toxin stimulates mucin synthesis, independent of the autonomic nervous system.

- Keratocyte growth factor (KGF) is a mitogen specific to epithelial cells. It has been well described for the skin, but its role in the intestine remains obscure. Zech et al. (Torrance, USA) investigated the possible involvement of KGF in restoring the integrity of the colon mucosa after lesion, cell proliferation and mucus production in the rat colon. KGF (5 mg/kg) distinctly improved the condition of the mucosal tissue aggressed by trinitrobenzene sulphonic acid (0.1 to 5 mg/kg), promoting both the proliferation of epithelial cells and the production of colon mucus.


INTESTINAL CYTOPROTECTION


- Although mainly dealing with bicarbonate secretion in the duodenum, Safsten et al. (Uppsala, Sweden) demonstrated that endogenous benzodiazepins (endozepins) play a major part in regulating mucus secretion intended to protect the duodenal wall against gastric hydrochloric acid.

- Sababi et al. (Uppsala, Sweden) studied the effects of lumen acidity, endogenous prostanoids and nitrogen monoxide (NO) on mucus and bicarbonate production in the anaesthetised rat duodenum. They demonstrated that secretion of duodenal mucus was strongly stimulated by HCl (10 mM, isotonic), by prostanoids and, to a lesser extent, by NO. On the other hand, bicarbonate secretion was stimulated by acid and endogenous prostanoids, but inhibited by NO.

- Following a reminder that prostaglandins' cytoprotective properties in the intestine are partly due to the release of active phospholipids resulting in increased hydrophobicity of the intestinal mucosa, Lugea et al. (Barcelona, Spain) showed that, in the rat, the non-steroid anti-inflammatory drugs (NSAIDs), indometacine (2 and 5 mg/kg, SC) and diclofenac (5 and 10 mg/kg, SC) inhibited the formation of phospholipid surfactant at the surface of the duodenal mucosa and thus reduced the cytoprotective hydrophobic properties. These observations contribute towards a better understanding of NSAIDs' ulcerogenic mechanisms.

- The same authors (Lugea et al., Barcelona, Spain) also researched cytoprotective mechanisms independent of prostaglandins which implicated the stimulation of phospholipid surfactant synthesis. Working on the rat, they used an agent which stimulates pneumocytes to synthesise and release pulmonary surfactant: ambroxol (40 mg/kg, intraduodenal). Ambroxol elicited phosphatidylcholine secretion by a mechanism independent of prostaglandins (resistant to indometacine). This increased the resistance of the duodenal mucosa to experimental aggression by hydrochloric acid (1 ml, 0,4 M, intraduodenal).


MUCUS ET MICI

In ulcerative colitis (UC), mucins may be deteriorated, thus affecting the mucous barrier's cytoprotective function. Are mucins the cause or the consequence of UC?

- Young et al. (New York, USA) reviewed the prevalence of the antigen sialosyl-Tn (STn) in human UC. Since STn expression occurs earlier than dysplasia may be identified, they stressed the value of studying the antigen alongside monitoring the colitis by endoscopy.

- The above conclusion was added to by Karlen et al. (Stockholm, Sweden and New York, USA) who studied STn expression in patients suffering from chronic UC. They observed that the antigen only expressed itself in these patients when the UC was compounded by dysplasia or cancer of the colon.

- Pladdet et al. (Amsterdam, Netherlands) considered mucin genes as risk markers of predisposition to UC. Gene MUC2, for example, would be a risk marker of genetic predisposition to ulcerative colitis. The presence of these mucins in the colon of patients suffering from UC would thus not be the result of the disorder but a predisposing factor.

- IL-11 is an interleukin whose anti-inflammatory properties now seem well established. Pfeiffer et al. (Blacksburg, USA) tested 312 rats for the effects of various doses of rhIL-11 (recombinant human) on experimental TNBS (trinitrobenzene sulphonic acid) colitis. They observed that IL-11 increased mucus production by the colon, significantly attenuated the increase in myeloperoxydase (MPO) enzyme activity elicited by the TNBS and decreased the ulceration index of the colon mucosa. The anti-inflammatory effects of IL-11 would thus be partly related to the cytokine's stimulation of mucus production.

- Weiss et al. (New York, USA), by in situ hybridisation, studied the expression of MUC2's and MUC3's messenger RNAs in the intestinal tissues in the constitutive, tumoral or inflammatory state. MUC2 was expressed in the goblet cells across the entire thickness of the mucosa in both the small intestine and the colon. MUC3 was especially expressed by differentiated cells in the villi (goblet and columnar cells) and surface epithelium. Expression of genes MUC2 and MUC3 was kept during intestinal inflammation (Crohn and ulcerative colitis) but decreased significantly in cancers of the colon.

- MUC2 is the main colon mucin secreted in UC. Tytgat et al. (Amsterdam, Netherlands) isolated mucins from UC patients and compared them with those identified from normals. It appeared that MUC2 was still the main mucin biosynthesised in the colon, in both the healthy subject and the patient suffering from UC. In addition, there seemed to be no quantitative differences in the mucin concentrations between controls and UCs, although this remains to be confirmed.

- In ulcerative colitis, the mucolytic activity of colon proteinases is high, resulting in a thinning out and discontinuity of the adherent mucous gel. Rankin et al. (Newcastle, USA) looked at the interaction of a polyacrylic acid, Carbopol 934P, with the mucus colon. Patients suffering from UC, treated with the product for 4 weeks (400 mg, 3 times a day), showed a distinct improvement in the sigmoidoscopic symptoms, and a significant drop in colon protease activity. The polyacrylic acids would thus seem to prevent mucolysis by preventing proteases from breaking down the pre-epithelial protective barrier.

- Pitcher et al. (Cambridge, USA) studied the role of sulphur-reducing bacteria (SRB) in the colon. SRB produce sulphites which generate inflammation in subjects genetically predisposed to colon disorders. They demonstrated that UC patients had a higher prevalence of SRB in the colon than healthy volunteers, and a higher level of free sulphates, probably from enzyme desulphatation of the mucus by faecal bacteria.


MUCUS AND CANCER
Gastric cancer

- Schmidt et al. (Hanover, Germany) used highly differentiated gastric cancer cell lines (HM02 and HS746T) which produce phospholipid-enriched mucus. Under normal conditions, the phospholipids are transported in the blood by lipoproteins. The authors checked the hypothesis according to which the synthesis and secretion of mucus phospholipids are regulated by apolipoproteins. The lines used secreted mucus containing phospholipids and expressed apoE but not apoA-I, apoB-48 or apoB-100 unlike the intestinal or hepatic cells. The intracellular role of apoE remains to be determined.

- Working on stomach cancer, Yong et al. (Osaka and Nagoya, Japan) studied the deterioration of human mucins occurring alongside malignant transformations identified by the expression of antigens Tn, sialyl Tn and T in cancer of the colon. The authors identified a new antigen (F1 alpha) associated with the mucins during experimental research on 141 patients with gastric cancer. The F1 alpha antigen expressed itself very early on, at the same time as carcinogenesis in the gastric epithelial cells.

Cancer of the Pancreas

- Raoul et al. (Rennes and Kremlin-Bicêtre, France) emphasise the prognostic value of determining the type of mucin (sialomucin, sulphomucin, etc.) encountered in, for example, Vater's ampulla carcinoma. The identification can be a useful complement and back up the well-established classical criteria for prognosis of this type of carcinoma.

- CAM17.1 is a murine antimucin antibody which binds to the sialyl I group of blood antigens. Yiannakou et al. (Liverpool, United Kingdom) demonstrated that the antibody can be used as serum tracer for detecting mucins released into the blood circulation by a pancreatic carcinoma.

Cancer of the Colon

- The mucins secreted by colon tumoral tissues have been characterised by Cho et al. (San Francisco, USA) on cell lines derived from colon adenocarcinomas (HM3) and colloid cancers (C1a) using cDNA probes coding for MUC1, 2, 3, 5 and 6. Colloid cancer cells secrete more MUC2-type mucins than adenocarcinomas, and present a higher level of antigen expression associated with short-side-chain mucins (CHO) such as sialyl-Tn, T and Tn.

- Yeh et al. (San Francisco, USA) examined the possible alterations in expression of mucin structural peptides and corresponding carbohydrate antigens on colorectal carcinomas (Duke's cancer) compared to metastases of lymph nodes. It would seem that expression of the mucins, which is specific to each tissue, is deregulated in cancerous tissues.

- In animal models, mucin production by cancerous colon cells is often related to the presence of hepatic metastases. Bresalier et al. (New York, USA) examined the relations between the structure of mucins' carbohydrate moieties and their adhesive properties. It is thus the carbohydrate chains, truncated at the mucin's body, which ensure cancerous colon cells adhere to the basement membranes. On the other hand, it is the long glycoprotein chains that are needed for colonising the liver, by means of interactions between the selectins and the sinusoid endothelium.

- Patients with pancreatic carcinoma produce killer T cells and antibodies which recognise certain peptide regions of the polypeptide core of MUC1-type mucins. Ho et al. (San Francisco, USA) evaluated the accessibility of MUC1-type peptide regions of intracellular, surface or secreted mucins on a pancreatic cancer cell line (SW1990). They observed that the highly glycosylated MUC1 peptide regions were accessible, eventhough they are normally protected by molecules of sialic acid.

- The MUC2 human gene codes for a mucin specific to goblet cells in the colon and small intestine. Gum et al. (San Francisco, USA) isolated and sequenced a genomic DNA clone of some 8000 base pairs on the 5' end region of gene MUC2, allowing them to characterise the MUC2 gene promoter in detail. The site is understood to include the first 500 base pairs starting from the transcription initiation site and consist in 63% G+C base pairs. The sequence isolated contains a TATA box located 24 bases above the initiation site. The research is geared towards producing transgenic mice expressing the MUC2 promoter as a model for studying various cell types in vivo.

Index de la page .