MUCUS UPDATE

Mucins and the esophagus

• The role of salivary constituents (mucins and bicarbonates) in the protection of the esophageal mucosa is largely documented. Thus, reflux esophagitis is often associated with impaired salivary secretion. Kinoshita et al (Saitama, Japan) studied the involvement of salivary mucins in the mechanisms of esophageal protection against acid and pepsin in the rat, an animal model which lacks submucosal mucus glands. The authors quantified the mucus present on the epithelium of the esophageal mucosa by alcian blue staining and undertook two series of experiments to illustrate the protective efficacy of this mucus. In vitro, they incubated the mucosa with 0.1N HCl solution containing pepsin, and observed a decrease of the adherent mucus and the release of cleaved peptides into the esophageal lumen; both effects were prevented by ecabet (0.3-3 mg/ml), an antipepsin agent with mucosal protective activity, and enhanced by N-acetyl-L cysteine (1% in the medium), a mucolytic agent. In vivo, they induced reflux esophagitis by ligating the pylorus and part of the stomach, and observed a rapid decrease of the adherent mucus, preceding the formation of hemorrhagic lesions of the mucosa; either ecabet (20 mg/kg) or omeprazole (10 mg/kg) prevented both mucus loss and epithelial lesions. These results confirm the importance of mucins, especially those from salivary origin, in the pre-epithelial protection of the esophagus against gastric reflux.

• Zbroch et al (Charlottesville & Kansas City, USA) recall, however, that 50 to 60% of patients with gastroesophageal reflux (GER), do not develop mucosal lesions. They state that mucosal integrity depends on the balance between "aggressive" factors of GER and "protective" mechanisms, linked to the presence of glycoconjugates (mucins) in esophageal mucus. In 10 patients with endoscopically negative GER disease, so-called negative-GERD, 39 healthy volunteers and 20 patients with GERD, the authors measured esophageal mucin secretion during a protocol mimicking the GERD scenario, i.e. by infusing the following solutions into the esophagus: saline, HCl, HCL plus pepsin, and saline again. They showed that, in both baseline conditions and after stimulation by HCl+pepsin, the secretion of mucins was significantly enhanced in negative-GERD compared with the other subjects. They conclude that the strong secretory response of the esophageal glands of negative-GERD patients enhanced the pre-epithelial barrier which may, therefore, absorb the acid and pepsin brought by the reflux, and protect the mucosa from injury.

• Marcinkiewicz et al of the same team (Kansas City, USA), emphasize the role of salivary mucus in the constitution of the pre-epithelial barrier of esophageal mucus. They collected salivary secretions using an intraesophageal catheter in the same patients as above, following a comparable protocol : baseline secretion, salivary secretion during mastication, during mechanical (balloon) stimulation, and during chemical (HCl+pepsin) stimulation of the esophagus. They showed that at all stages of the protocol, the secretion of glycoconjugates (mainly mucins) was significantly lower in GERD patients than in healthy volunteers, a fact which contributes to a lower efficiency of the pre-epithelial protective barrier. This impairment justifies therapeutic intervention targeting salivary function and/or mucus glands of the esophagus.

• To emphasize the role of salivary mucins, their colleagues Brown et al, (Kansas City, USA) recall that GERD and subsequent complications are less prevalent in African-Americans than in Caucasians. An explanation for this could be, according to the authors' hypothesis, better protective capacities of saliva in African-Americans. In healthy volunteers, indeed, the authors showed that the saliva of African-Americans (n=27) contained more mucins, proteins and non-bicarbonate buffer than that of Caucasians (n=39), in both basal conditions and during mastication; an additional increase in bicarbonate and TGFa was observed during mechanical or chemical stimulation of the esophagus. An enhanced secretion of protective salivary glycoconjugates (mucins) in African-Americans may, therefore, more efficiently prevent the development of esophageal mucosal pathologies and attached complications during episodes of GERD.

Mucins and stomach

Gastric cytoprotection

• The presence of a well constituted gel of resident mucus is indispensable to protect the gastric epithelium from back diffusion of acid to the mucosal surface. In contrast, in the gallbladder, epithelial cells are permanently exposed to high concentrations of biliary salts as the mucus remains in solution, while a thick mucus may lead to biliary stasis and the formation of sludge and stones. Why did mucins organize into an adherent gel on the gastric epithelium and remain in solution in bile? Cao et al (Boston, USA) investigated whether different pH conditions between both organs, could account for different physical properties between purified bovine gallbladder mucin and porcine gastric mucin. At pH 7, very few associations of mucins are observed; at pH 4, gastric mucins formed macromolecular associations whereas gallbladder mucins remained unchanged; at pH 2, gastric mucins formed a gel whereas gallbladder mucins remained in solution. Thus, gastric mucins display unique pH-sensitive properties which determine both mucus conformation and the transition from solution to gel; these pH-sensitive properties do not exist in gallbladder mucins.

• Chu et al (Los Angeles & Baltimore, USA) used confocal microscopy to assess the "protective" effect of mucus on pH at the surface of gastric epithelium. The gastric mucosa of 24 anesthetized rats was exposed luminal side up under a confocal microscope (Zeiss LSM410) and superfused with Krebs buffer at different acid pHs, containing 10 µM of a pH-sensitive dye (CI-NERF). In basal conditions, mucus thickness was found very variable (0 to 270 µm; mean: 70 µm) and not correlated with the pH at the epithelial surface which remained close to 4 when the superfusion fluid was at pH 3 or 5. In contrast, when bicarbonate secretion was stimulated by dimethyl-prostaglandin E2, the alkalinization (pH 5,5) of the epithelium was better maintained under a thick layer of mucus (> 50 µm) than under a thinner one (pH 4,5 for 0 to 25 µm). So, mucus thickness could play an important role in the conservation of the relative alcalinization of gastric epithelium induced by bicarbonate.

• Septic shock is often accompanied by gastric ulcers and bleeding for reasons that are not totally understood. The LPS-treated rat (E. coli lipopolysaccharide) is an animal model for investigations on septic shock, in which slowing down of gastric emptying and liquid accumulation in the stomach have been described. Dial et al, (Houston, USA) tried to identify both the origin and the mechanism of this accumulation of fluid, and to determine whether it was cytoprotective or harmful response of the organ to the infectious agent. They observed that LPS (E. coli 0111:B4; 1 mg/kg, intraperitoneally) induced a nine-fold increase in gastric fluid volume and rose 17 to 40 times concentrations of proteins, mucins, and myeloperoxidase (an enzyme marker of neutrophil infiltration, and also in this particular case a marker of plasma extravasation). The data showed that fluid accumulation resulted, indeed, from delayed gastric emptying, but also and mainly from mucus secretion and plasma extravasation following the destruction of the hydrophobic barrier of gastric epithelium. It cannot be concluded, however, whether LPS-induced response mechanisms were cytoprotective (mucus secretion) or harmful (destruction of pre-epithelial layer).

• In rats, where diabetes was experimentally induced by an intravenous injection of streptozotocin (65 mg/kg), the gastric mucosa is exquisitely sensitive to ulcerogenic stimuli, especially in fasting animals. (Kinoshita et al, Saitama, Japan) addressed the consequences of fasting on the integrity of gastric mucosa and mucin secretion. They observed that a 24 hour fast did not provoke neither epithelial injury nor changes in mucins in control rats. In contrast, a significant reduction of total glycoproteins in gastric mucus was observed before the formation of gastric lesions in diabetic rats, of which 70% were affected since the 9th hour of fasting. That mucus plays a key role is highlighted by the fact that gluthation (1 g/kg; 2 ip injections), which counteracted the reduction of glycoproteins also prevented epithelial injury.

• The same pathophysiological model was used by Kinoshita et al, (Saitama, Japan) to evaluate the influence of cold exposure (5°C) upon the integrity of gastric mucosa. No alterations were noticed on both rectal temperature and macroscopic features of the mucosa in fasted control rats. In contrast, in diabetic rats, fasting resulted in a complete loss of mucins in the mucus gel and the formation of mucosal lesions; these later were enhanced and became hemorrhagic after exposure to cold. In addition, a marked hypothermia was observed in these rats, whose gastrinemia, as well as gastric acid and pepsin amounts were dramatically increased. Investigations aimed at determining the mechanisms involved in the effects of cold exposure showed that vagotomy, or treatment with either cimetidine, YM-022 (a gastrin antagonist), or ecabet (a pepsin antagonist), prevented these effects. It is concluded that hypothermia enhances vagal activity and gastrin levels, therefore increasing the amounts of acid and pepsin, resulting in gastric lesions in the fasting diabetic rats. Starvation-induced depletion of mucus glycoproteins further aggravates the lesions by increasing the vulnerability of the mucosa to acid and pepsin.

Mucus and Helicobacter pylori

• Based on earlier experiments showing that water extracts of Helicobacter pylori (Hp) prevented the formation of the mucus gel in the corpus of the rat stomach, Atuma et al (Stockholm & Gotheborg, Sweden) tried to further investigate mucus gel accumulation in the corpus and antrum of rats chronically infected with Hp. Thus, rats were infected with a suspension of Hp (strain Hel73; VacA positive) given by gastric gavage. Four months later, rats were anesthetized and mucus thickness measured directly on the gastric mucosa mounted luminal side up under a microscope; the lower firm layer of mucus gel was differentiated from the upper layer of accumulated mucus (which was gently removed by suction with a micropipette before measurements). The results indicate that, compared to healthy controls, chronic infection with Hp significantly reduced the total thickness of mucus and that of the lower gel in both gastric corpus and antrum. They also show that mucus accumulation in response to a 20 min exposure of the mucosa to acid (pH 1) was halved in infected rats (42 µm) compared with controls (100 µm). Overall, chronic infection with Hp diminishes the protection of gastric epithelium and more specially the ability of the antrum to accumulate an upper layer of protective mucus in response to acid.

• There is little doubt about the involvement of cytokines in the pathogenicity of Hp, but experimental data are scarce. For instance, Takahashi et al, (Kyoto, Japan) used primary cultures of rabbit gastric epithelial cells to investigate the effects of cytokines on mucus secretion in the presence or absence of a lysate of Hp (NCTC 11637). In basal conditions, they observed that interleukin (IL)-1ß and IL-6 stimulated mucin secretion, with IL-1 being more potent than most mucus secretagogues. In contrast, IL-7, IL-8, IL-10, interferon (INF) g and TNFa had no effect. In the presence of Hp lysate, basal secretion of mucus was significantly reduced and was no long stimulated by any of the cytokines; besides, INFg augmented the Hp lysate-induced inhibitory effects. The authors suggest that IL-1 and IL-6 stimulate mucus secretion by a healthy stomach, whereas INFg inhibits mucus secretion by the Hp-infected stomach.

• The methods of transport to laboratories of gastric mucus samples intended for Hp identification is apparently not a hot issue. Nevertheless, it actually was an issue for Windsor et al, (Perth, Australia), probably worried by the dimension of their continent-wide country and the huge problems arising from long distances to hospital centers and laboratories. The authors' specific purpose was to validate a transportation method for samples using dry cotton swabs where gastric mucus was transferred after wiping the antral mucosa with a cytology brush; Hp conservation in samples was tested by PCR amplification of a specific cDNA probe. Mucus samples were collected during routine endoscopy in 27 patients (12 Hp-positive, 11 Hp-negative, 4 excluded; CLO-test and antral mucosa histology), transferred in duplicate to cotton swabs and sent to the laboratory within 36 hours. The authors note that all samples which were Hp-positive at the initial examination, remained positive after transportation and conservation at different temperatures (-85°C, 4°C, 23°C, 37°C) and times (up to 3 weeks). They suggest, therefore, that mucus samples for Hp identification can be sent onto dry cotton swabs through regular mail.

Mucus and the gallbladder

Mucin secretion and et gallbladder cytoprotection

• Biliary crystals induce inflammation of the gallbladder wall and release of inflammatory cytokines, especially interleukin (IL)-1. Cheung et al, (Chicago & Richmond, USA) investigated whether IL-1 induces mucin secretion. Using cultures of human gallbladder epithelial cells, the authors indeed showed that IL-1a (10 and 50 pg/ml), just like PGE2, stimulated mucin secretion and significantly enhanced the thickness of the mucus layer; this effect is mediated by specific IL-1 receptors, since it was inhibited by IL-1ra (10 ng/ml), a selective IL-1 receptor antagonist. This study suggests that inflammatory cytokines most probably contribute to the huge amounts of mucus present into the gallbladder during gallstone formation.

• Just like IL-1, TNFa (tumor necrosis factor a) is a major inflammatory cytokine. Cheung et al, Chicago & Richmond, USA) adapted the above hypothesis, and suggested now that TNFa could be the relevant cytokine involved in the formation of gallstones. Again, just like IL-1, they showed that TNFa (10 and 50 pg/ml) as well as PGE2 (0.1 mM), stimulates mucin secretion and increases 4-fold the thickness of the mucus layer. The authors conclude that TNFa is a potential inducer of the hypersecretion of mucins by the epithelial cells of the inflammatory and/or gallstone-bearing gallbladder.

Gallstone formation and gallbladder inflammation

• Eder et al (Boston, USA) were also interested by biliary TNFa and investigated whether this cytokine induced not only the secretion but also the synthesis of mucins. As their Chicago colleagues, the authors showed that TNFa (50 ng/ml) caused a two-fold increase in mucin secretion by gallbladder epithelial cells in primary culture and cell line culture. This increase was abolished by pre-incubation with neutralizing anti-TNFa antibodies, but not by gene transcription inhibitors. In contrast, TNFa (10 to 100 ng/ml) did not alter levels of MUC5B and MUC3 gene transcripts. Consequently, treatment with TNFa appears to provoke the release of pre-synthesized mucins, but has no effect on the synthesis of new mucins. Other constituents in bile are likely to be involved in the control of the sustained production of mucins which precedes gallstone formation.

• Kano et al, (Tokyo et Ibaraki, Japan) compared the expression levels of phospholipase A2 (PLA2) and mucin genes in gallstone patients chronically treated with ursodeoxycholic acid (UDCA) and in untreated patients. They found that long-term treatment with UDCA resulted in a significant drop of total mucins, but did not modify the expression levels of MUC3, MUC5B and MUC6 genes, coding for the major mucins found in the gallbladder of these patients. The efficacy of UDCA in the treatment of gallstone disease can, therefore, be related to its ability to specifically inhibit the release but not the synthesis of gallbladder mucins.

• Lipid peroxidation, as assessed by malondialdehyde (MDA) assay, increases in parallel with mucin secretion in lithogenic bile. Because UDCA has been shown to reduce the lithogenicity of bile, von Ritter et al (Munich, Germany), wondered whether UDCA actually decreased biliary lipid peroxidation and/or mucin secretion. The study was carried out in 19 gallstone patients treated with UDCA or placebo during 10 days before cholecystectomy. The data showed that lipid peroxidation increased significantly more in the placebo group that in the UDCA-treated group. In addition, the increase in biliary mucin secretion in the placebo group was significantly higher than that induced in the bile of UDCA-treated patients. So, UDCA reduces lipid peroxidation and mucus secretagogue properties of the gallbladder, and both mechanisms could well account for the beneficial action of UDCA in cholesterol gallstone patients.

• In the inflammatory gallbladder, trefoil peptides are co-expressed and co-secreted with the mucins involved in the formation of cholesterol crystals and gallstones. Their own role, however, remains obscure. Nunes et al, (Boston, USA) used a standard bile (reference CSI 1.2) on which they measured the appearance time and number of cholesterol crystals formed in the presence of the trefoil hSP (human spasmolytic polypeptide), alone or in combination with purified human and bovine mucins (0.5 mg/ml). They found that hSP alone (0.1, 1 et 10 µg/ml) modestly, but concentration-dependently, increased the number of crystals at 6 days, indicating nucleogenic properties comparable to those of mucins. In contrast, if hSP was added to bile in combination with mucins, then the appearance of the first crystal was accelerated and the number of crystal formed was dramatically increased; for instance, when hSP was added at 10 µg/ml, the first crystals appeared after 1.5 day in the presence of mucins vs 5 days in the absence of mucins; in addition, the number of crystals formed at 6 days is higher than 1000 for hSP+mucins vs only 3 for hSP alone. The pronucleating properties of both trefoil peptides and mucins are not simply additive; they both act in a really synergistic fashion resulting in a dramatic enhancement of the lithogenic properties of bile.

• Eicosapentanoic acid (EPA), major constituent of fish oil, possess antilithogenic properties in different animal models, but its mechanism of action has not been elucidated. Molecular similarities between EPA and arachidonic acid (AA) which is strongly involved in the intracellular cascade leading to mucin release, have driven Shimizu et al, (Ibaraki, Japan & Seattle, USA) to investigate the involvement of EPA in the secretion of mucins by the gallbladder epithelium. The authors incubated canine gallbladder epithelial cells with fatty acids, including EPA and AA, both as free acids or phosphatidylcholine, and measured the secretion of mucins and PGE2 in the culture medium after either short-term (4 hours) or long-term (10 days) incubation. In both cases, AA (100 µM) stimulated PGE2 and mucin secretion, whereas EPA (100 µM) and the other free fatty acids as well, were without effect. In contrast, in cultures pretreated with EPA (30 µM, 10 days) responses to AA were abolished. It is suggested that the beneficial effects of EPA in the prevention of gallstone formation in vivo are indirect and involve the inhibition of the metabolic cascade of arachidonic acid which leads to PGE2 synthesis and mucin release.
• Gallbladder bile from gallstone patients contains a non-soluble sediment or "sludge" consisting of a suspension of cholesterol monohydrate and calcium bilirubinate granules, intimately embedded in mucin. Lechène de la Porte et al (Marseille, France & Munich, Germany) hypothesized that this "sludge" is an early stage in the process of gallstone formation and, consequently, investigated the distribution pattern of mucins and proteins in the "sludge" and in cholesterol crystals. Bile from gallstone patients was ultra-centrifuged (100.000 x g for 1 hour); the insoluble pellet was then prepared for analysis. The distribution patterns of mucins and proteins in this sediment appeared to be very close to those found in the central core of gallbladder cholesterol crystals. This "sludge" could, therefore, well represent an initial step of the crystallization process; better knowledge of this sediment appears to be useful for better understanding of the mechanisms leading to gallstone formation.

Mucus, small intestine and colon

Cytoprotection

• Jean-Claude Cuber's team in Lyon (France), in collaboration with Australian researchers, (Moro et al, Lyon & Jouy-en-Josas, France & Melbourne, Australia) investigated the mechanisms involved in the release of the trefoil peptide ITF (intestinal trefoil factor) into the gut lumen. The authors measured ITF immunoreactivity in luminal and portal effluents in a vascularly (superior mesenteric artery) perfused (Krebs, with or without mucin secretagogues) isolated colonic loop (10 cm) in rats. They found no ITF immunoreactivity in the luminal effluent both under baseline conditions and after intra-arterial infusion of betanechol, VIP, IL-1ß and PGE2. In contrast, if the effluent was treated with di-thio-treitol (DTT; 10 mM), then a constitutive ITF secretion was detected (673±59 ng for 30 min; n=24). The mucus secretagogues used in this study enhanced ITF secretion in the luminal effluent (betanechol: 3358±652 ng, n=13; bombesine: 4903±216 ng, n=16) but never in the portal effluent. It is the first time that a secretion rate of ITF is reported for the rat colon, both under baseline conditions and after stimulation by secretagogues of which the peptide bombesine appeared particularly active. Finally, ITF appears to be linked by di-sulfure bridges to other constituents of the luminal effluent, and most probably to mucins.

• Daniel Podolsky's team in Boston is well known for its expertise in the field of trefoil peptides and addressed an issue related to the role of these peptides in celiac disease (CD). Thus, Ciacci et al (Naples, Italy & Boston, USA) (#1268) hypothesized that dietary gluten could impair intestinal production of trefoil peptides and mucins, and induce the chronic lesions of the gut epithelium observed in CD. Using specific antibodies, they measured changes of ITF secretion in jejunal goblet cells in 30 CD patients, including 10 on a gluten-rich diet (untreated CD) and 20 on gluten-free diet (treated CD), and 6 healthy volunteers. Mucin expression was found normal, whereas ITF expression was very low, in both untreated patients and in those of the treated CD who still had non-complete mucosal recovery. In contrast, treated CD patients with complete mucosal recovery expressed ITF and mucins at a similar level as healthy volunteers. The authors suggest that the low expression of ITF during celiac disease could reflect a gluten-induced injury or an over-consumption secondary to chronic mucosal damage.

Synthesis and regulation

• MUC2 is the predominant mucin of the resident mucus-layer of the human colon. Tytgat's team in Amsterdam, which has works on mucin involvement in inflammatory bowel diseases, previously demonstrated that MUC2 secretion was deficient in ulcerative colitis. In a very precise study, using immunology and molecular biology techniques, van Klinken et al, Amsterdam et Rotterdam, Pays Bas) characterized the predominant mucin in the colon of mice, an animal model frequently used for the study of colitis. They identified a predominant colonic mucin very close to human MUC2, which they called murine Muc2. They cloned a 1 kb cDNA fragment from murine colonic mRNA by RT-PCR (Reverse Transcription-Polymerase Chain Reaction) using human MUC2 pimers; this cDNA fragment showed 75% homology with the 5' region of human MUC2 cDNA sequence.

• This study emphasizes the potential usefulness of transgenic murine models of Muc2 gene deficiency, in the study of the role of MUC2 in human colonic inflammatory pathology.

Mucus and IBD

Experimental inflammation

• Lipopolysaccharide (LPS) present in the intestine of E. coli -colonized animals has been shown to stimulate the release of cytoprotective mucins. The study by Enss et al, (Hannover, Germany) was aimed at verifying these effects of LPS and, further, at highlighting a possible involvement of certain types of macrophages in mucin secretion. The authors administered LPS (E. coli O55:B5; 7 mg/g body weight) by gastric gavage to germ-free mice from two different strains: one (NMRI strain) possess LPS-sensitive macrophages; the second (C3H/HeJ) only possesses LPS-sensitive macrophages. They measured the mucins released into the cecum and colon (amount, composition). In baseline conditions, both strains produced similar amounts of mucins, but on stimulation by LPS, NMRI cells produce much more mucins, and of higher molecular weight, than C3H/HeJ. It was concluded that LPS-sensitive macrophages are of paramount importance for the mucus-releasing effect, yet cytoprotective, of LPS.

• Inflammatory cytokines are among the mediators potentially involved in this secretory action of LPS. Thus, the same authors (Enss et al, Hannover, Germany) tested the effects of cytokines on the expression of mucins. A human colon cell line (LS180; American Type Culture Collection) was incubated with IL-1, IL6 and TNFa (1 to 10 mg/ml) for 1 to 16 hours. Thereafter, expression of MUC2, MUC5AC, MUC5B and MUC6 mRNAs was assessed by semi-quantitative RT-PCR, and mucins were characterized and quantitated. The authors observed that all three cytokines tested reduced the expression of MUC2 and increased the expression of MUC5AC, MUC5B and MUC6 in the cultures; IL-1 and IL-6 especially increased the release of mucin proteins (+43% and +33% respectively after 8 hour incubation), whereas TNFa, not only dramatically increased proteins (+69%) but also stimulated the glycosylation of the glucidic chain of mucins (+53%). This study brings to light a differential regulation of mucin secretion by different inflammatory cytokines; one of these, namely TNFa, induces the release of mucins with different degrees of glycosylation, which may imply functional alterations of these mucins during colonic inflammation.

• Ulcerative colitis is, indeed, characterized by important lesions of colonic epithelium and generalized impairment of colonic function whose mechanism is poorly understood. Renes et al, (Amsterdam & Rotterdam, The Netherlands) used a rat model of experimental colitis (7% dextran sodium sulfate (DSS) in drinking water) to study the regulation of colonic epithelial genes during an inflammatory process close to human ulcerative colitis. The authors were more specially interested by MUC2 and intestinal trefoil factor (ITF) genes, as markers of goblet cells. The noticed that MUC2 production, a major constituent of colonic mucus gel, was well maintained during DSS-induced inflammation. In contrast, the expression of ITF was greatly reduced. Since ITF plays a major role in epithelial repair, the reduced ITF expression implies a lower resistance to inflammation of the colonic epithelium.

Ulcerative colitis

• Transdermal nicotine is effective for relief of active ulcerative colitis (UC), but its mechanism of action remains unknown. In order to explain this clinical action of nicotine, Louvet et al, (Lille, France & Rochester MN, USA) tested two hypotheses: nicotine can modify both the quality and quantity of mucus, and/or the expression of colonic mucins. Sixty four non-smoking UC patients were assigned to two groups: 31 were treated with transdermal nicotine (22 mg/day) et 33 with placebo for 4 weeks. Colonic biopsies were collected at the onset and at the end of the treatment; inflammatory cytokines (IL-1, IL-6, IL-8, TNFa), immuno-regulatory cytokines (IL-2, INFg, IL-4), and mRNAs for MUC1 to 6 were quantified. Twelve out of 31 patients treated with nicotine, and 3 out of 33 patients receiving placebo, responded positively. In contrast, mucins and cytokines were not modified, with the exception of IL-8 which dropped significantly in those of the nicotine-treated patients whose condition was improved. So, it looks as if mucins were not implicated in the beneficial clinical response to nicotine in the treatment of ulcerative colitis.

• Adrian Allen's team is a recognized authority in the field of cytoprotection of intestinal epithelium by the adherent mucus gel. Thus, Strugala et al (Newcastle/Tyne, United Kingdom) stated that the efficacy of this mucus barrier depends on both the integrity and thickness of the layer. The aim of their study was to measure the thickness of colonic mucus layer and to identify pathophysiological alterations during inflammatory bowel disease. Colonic biopsies were collected from patients with UC or Crohn disease, and volunteers with an apparently normal colonic mucosa. The layer of adherent mucus was found to be relatively thin (20 to 65 µm with an oral to aboral gradient from cecum to rectum) but firm and continuous on the normal colonic mucosa. It appeared much thinner (10 à 30 µm) and disrupted on UC biopsies. In contrast, in Crohn's disease biopsies, the mucus layer was thicker (198 µm) and continuous.

• The synthesis of the predominant mucin of the human colon, MUC2, has been shown to be reduced during episodes of ulcerative colitis, thus contributing to the weakening of the resident mucus layer. A new study by van Klinken et al (Amsterdam & Rotterdam, The Netherlands) specifies the alterations of both secretion and biosynthesis of this mucin during ulcerative colitis. The authors performed 4 colonic biopsies in 10 patients with active ulcerative colitis, 6 with colitis in remission and 11 healthy volunteers; two of these biopsies were incubated with [³⁵S]methionine/cysteine and the other two with [³⁵S]sulfate, metabolic radiomarkers of the synthesis of different types of mucins. No differences were observed in secretion rates of total mucins and [³⁵S]methionine/cysteine radiolabeled mucins between the three groups. In contrast, the secretion rate of sulfated mucins was increased twofold in active UC biopsies, compared with colitis in remission and healthy controls. A relative increase of sulfated MUC2 mucins in colonic mucus may, therefore, indicate the presence of an active ulcerative colitis.

• Numerous studies suggest that ulcerative colitis may have an autoimmune etiology, but none so far has firmly identified the corresponding auto-antigens. Ohara et al, (Sagamihara, Japan) (#3092) immunized rabbits with subcutaneous injections of purified colonic mucins (1 mg emulsion in Freund's complete adjuvant). After a few weeks, they collected serum from these immunized rabbits, and found the presence of anti-mucin antibodies. In parallel, they performed histochemistry and revealed inflammatory cell infiltration and colonic mucosal erosion in these rabbits. To summarize, colonic mucins induce anti-mucin antibodies and the immunized animals develop colitis. Consequently, the authors who had already found similar results in immunized rats, conclude that colonic mucins could be at least one of the auto-antigenes of ulcerative colitis.

Irritable bowel syndrome

• Mucus-rich diarrheas are frequent in patients with irritable bowel syndrome (IBS). It has been shown that the peptide neurotensin and its receptors were involved in colonic mucin secretion in rats during experimental stress in vivo, or in human colonic cell culture in vitro . Castagliuolo et al, (Boston, USA) verified the reality of this effect of neurotensin in 7 patients with IBS and 8 asymptomatic volunteers. They measured similar neurotensin concentrations in colonic biopsies from both groups; in contrast, densities of neurotensin receptors were four-fold higher in IBS patients than in healthy subjects. IBS pathophysiology appears, therefore, to involve an overexpression of neurotensin receptors in the human colonic mucosa.

Mucus and cancers

Cancer of the esophagus

• Barrett's metaplasia is the primary risk factor for esophageal adenocarcinoma. Although histochemical studies demonstrated the presence of acidic mucins in the columnar mucosa which characterizes Barrett's esophagus, the different types of mucins actually secreted by the metaplastic epithelium have not been specified. The aims of Byrd et al, (Detroit, USA) were to ascertain by immunochemistry if MUC2 (intestinal type ) and MUC6 (gastric type) mucin genes were expressed in Barrett's esophagus and different esophageal adenocarcinomas. Monoclonal (for MUC2 gene detection) and polyclonal (for MUC6) antibodies were incubated with esophageal tissue samples collected from 13 patients with gastro-esophageal reflux (GER), 26 Barrett's esophagus, 17 esophageal adenocarcinomas and 12 other epithelial carcinomas of the esophagus. These later, as well as GER samples, showed no staining for MUC2 and MUC6. In contrast, 81% of Barrett's esophagus showed both MUC2 (highly expressed in goblet cells) and MUC6 (in glands), vs 4% for MUC2 alone and 15% for MUC6 alone. These percentages were dramatically altered in adenocarcinoma: MUC2 and 6 were simultaneously present in 29% of samples, while 35% contained MUC2 alone and 29% MUC6 alone. So, aberrant expression of these two mucin genes seems to be an early signal in the development of esophageal carcinoma.

Gastric cancer

• Intestinal metaplasia of the gastric mucosa is a major risk of developing an adenocarcinoma, and is often associated with a chronic infection by Helicobacter pylori.. Looking for a relevant animal model, Dangler et al, (Cambridge et Boston, USA) (#2864) undertook a compared histochemical study of the metaplastic gastric mucosa of three strains of mice (BALB/c, C3H/H3J, C57BL/6) infected for 19 or 41 weeks by Helicobacter felis. Only the C57BL/6 strain presented a severe gastritis associated with metaplastic epithelium in response to chronic infection. This epithelium was highly stained by alcian blue at pH 2.5, indicating the presence of sialomucins, typical phenotype of intestinal mucins. Some metaplastic mucosal cells were stained by alcian blue at pH 1.0, indicating the presence of colonic type sulfomucins. Consequently, this animal model which mimicks the main features of intestinal metaplasia of gastric epithelium, will be very useful for studying the mechanisms involved in the transition from gastric metaplasia to carcinoma in humans.

• An alteration of cytoprotective mucin glycoproteins may have a decisive influence on gastric epithelium invasion by cancer cells and further development of metastasis. Kim et al, (Seoul, South Korea & San Francisco, USA) studied the expression of 5 types of mucin genes (MUC 1, 2, 3, 5 et 6) in relation with the depth of invasion in early gastric cancer, i.e. limited to the mucosa and/or the submucosa of the stomach, or in the presence of lymph node metastases. In 73 samples of early gastric cancer, they found MUC1 (18% of cases), MUC2 (36%), MUC3 (27%), MUC 5/6 (38%). MUC1, 2 and 3 were frequently associated with deeper tumors; in contrast MUC 5/6 was less frequent in deep cancers than in more superficial cancers. Consequently, mucin gene expression appears to be markedly altered in early gastric cancer and this alteration influences depth of invasion.

Cancer of the pancreas

• Mucinous cystadenomas (MCA) are premalignant lesions of the pancreas which should be differentiated from serous cystadenomas or pseudocysts which are not malignant. Because of the specific presence of mucins in these lesions, Freedman et al, (Boston, USA) (#609) proposed that anti-mucin antibodies could be efficient tools for the diagnosis. Surgical samples of cystic lesions of the pancreas from 20 patients were studied. Four MCA and 7 mucinous cystadenocarcinomas (MCAC) were found. Among tested antibodies, only the anti-MUC1 antibody identified a specific protein in 6 out of 7 MCAC samples.

• The anti-MUC1 antibody can therefore be of benefit in the diagnosis of lesions based on cystic fluid collected by endoscopic or percutaneous drainage.

Colon cancer

• Investigations on colorectal cancer (CRC) seldom addresses the histomorphometry of the "mucocellular layer" at the epithelial surface. This is quite surprising, because this layer is a snapshot of shed cellular elements transiently entrapped in the mucus, which represents a mine of information on cancerous pathogenicity. Ahlquist et al, (Rochester MN, USA) aimed at identifying constitutional differences in the mucocellular layer between CRC and normal mucosa, in order to identify possible tumor markers to be used in stool screening. Their results confirmed their expectations: they observed that the mucus of cancerous tissues was highly infiltrated with inflammatory and epithelial cell debris, and contained a large amount of soluble tumor-associated antigens. This study suggests that both qualitative and quantitative differences between contents of the mucocellular layer are so marked that they should be of direct relevance for coproscopic diagnosis.

• It is well known that the expression of MUC2 is correlated to the metastatic capacity of human colonic cancer cells, but the constitutive presence of this mucin in the normal colon forbids its use as a tumoral marker. Usually, the existence of cancer is revealed by the aberrant expression of a mucin in a particular tissue. Byrd et al, (Detroit, USA) tried to identify MUC5AC, which is normally expressed in the gastric epithelium, in 15 samples of primary colonic cancer and associated liver metastases. No immunohistochemical staining by anti-MUC5 antibody was observed for the normal-appearing parts of the adjacent colonic mucosa (0/8); in contrast, 12/15 primary tumors and 5/5 liver metastases expressed MUC5AC. In situ hybridization confirmed that MUC5AC gene expression was limited to tumor cells and metastases. Thus, the ectopic presence of MUC5AC in the colon is a relevant marker of tumor progression.

   

Index de la page .
coproweb@free.fr