Lars-Göran Axelsson

Secreted enteric antimicrobial activity localises to the mucus surface layer

Objectives: The intestinal mucosa is constantly exposed to a dense and highly dynamic microbial flora and challenged by a variety of enteropathogenic bacteria. Antibacterial protection is provided in part by Paneth cell-derived antibacterial peptides such as the α-defensins. The mechanism of peptide-mediated antibacterial control and its functional importance for gut homeostasis has recently been appreciated in patients with Crohn’s ileitis. In the present study, the spatial distribution of antimicrobial peptides was analysed within the small intestinal anatomical compartments such as the intestinal crypts, the overlaying mucus and the luminal content. Methods: Preparations from the different intestinal locations as well as whole mouse small intestine were extracted and separated by reversed-phase high-performance liquid chromatography. Antibacterial activity was determined in extracts, and the presence of antimicrobial peptides/proteins was confirmed by N-terminal sequencing, mass spectrometry analysis and immunodetection. Results: The secreted antibacterial activity was largely confined to the layer of mucus, whereas only minute amounts of activity were noted in the luminal content. The extractable activity originating from either crypt/mucus/lumen compartments respectively (given as a percentage) was for Listeria monocytogenes, 48 (4)/44 (4)/8 (8); Enterococcus faecalis, 44 (10)/49 (3)/7 (7); Bacterium megaterium, 56 (4)/42 (3)/2 (1); Streptococcus pyogenes, 48 (4)/46 (3)/6 (6); Escherichia coli, 46 (4)/47 (3)/7 (7); and Salmonella enterica sv. Typhimurium, 38 (3)/43 (7)/19 (10). A spectrum of antimicrobial peptides was identified in isolated mucus, which exhibited strong and contact-dependent antibacterial activity against both commensal and pathogenic bacteria. Conclusion: These findings show that secreted antimicrobial peptides are retained by the surface-overlaying mucus and thereby provide a combined physical and antibacterial barrier to prevent bacterial attachment and invasion. This distribution facilitates high local peptide concentration on vulnerable mucosal surfaces, while still allowing the presence of an enteric microbiota.

Microbial-host interactions specifically control the glycosylation pattern in intestinal mouse mucosa.

Abstract. The glycosylation of the intestinal cell layer is thought to control several key functions of the gut such as vectorial transports, defence against microbial agents or immunological processes. It has been assumed that the gut microflora may modulate the glycosylation pattern of the intestinal cell layer. However, there is no direct evidence for this regulatory process. The first goal of this work was to establish the germ-free mice intestinal glycosylation baseline using a histochemical approach and a panel of ten lectins with defined glycan specificities to tissue sections prepared from various cellular compartments of the small and large intestine. Using this baseline, we have studied the contribution of the gut microflora on the carbohydrate composition of glycoconjugates of intestinal cells by comparing the germ-free and conventional mice glycosylation patterns. Analysis of the germ-free mice intestinal glycosylation baseline revealed that the expression of glycans depends on the proximodistal gradient (small to large intestine) and on the cell lineage (absorptive, goblet, crypt, and Paneth cells), indicating that mice are able to create and maintain a strict topological and cell lineage-specific regulation of glycosyltransferase expression. By comparing germ-free and conventional mice, we find that the gut microflora specifically modulates the gut glycosylation pattern, quantitatively as well as qualitatively by changing the cellular and subcellular distribution of glycans. This is the first report in mice to directly demonstrate the critical contribution of microflora to intestinal glycosylation, a key characteristic of the gut.

Indigenous microbes and their soluble factors differentially modulate intestinal glycosylation steps in vivo. Use of a "lectin assay" to survey in vivo glycosylation changes.

It has been shown that Bacteroides thetaiotaomicron, a representative member of the gut microflora, signals intestinal epithelial cells both in vivo and in vitro and modulate specific glycosylation processes that may mediate intestinal functions. However it is not known whether these modulations depend on the presence of live bacteria or may be elicited by soluble factors produced in vitro by this bacterium. We used lectins and an histochemical approach to survey tissue sections prepared from various cellular compartments of the small and large intestine of NRMI/KI mice grown under gnotobiotic conditions. We compared the results obtained with bacterial culture supernatant and live B. thetaiotaomicron to those obtained from germ-free mice or mice having a conventional microflora. This approach allowed us to conclude that (1) a small but specific number of glycan patterns were restored after treatment with bacterial culture supernatant and (2) the B. thetaiotaomicron associated mice restored a larger number of patterns, however, the complete conventional mice pattern must be a function of the whole microflora in the gut. The possibility to modulate this complex glycosylation pattern by introducing exogenous bacteria and bacterial products should be considered as a promising approach towards understanding the molecular basis of microbial-host interactions.

Abrogated lymphocyte infiltration and lowered CD14 in dextran sulfate induced colitis in mice treated with p65 antisense oligonucleotides.

Abstract.Background and aims: Dextran sulfate sodium (DSS) induced colitis exhibits a predominantly NF-κB dependent proinflammatory cytokine profile and shares similarities with human inflammatory bowel disease (IBD). Lamina propria macrophages of IBD patients display elevated levels of NF-κB p65. Knowing the role of NF-κB in IBD, we investigated the beneficial cellular mechanisms underlying the lasting effect of a single p65 antisense treatment in DSS-colitis mice. Methods: One local dose of p65 antisense oligonucleotides was administered in DSS colitis mice. Ten days later the mice were killed and examined at cellular and biochemical levels. The level of p65 in lamina propria cells was determined by electrophoretic mobility shift assay and by intracellular immunofluorescent staining of nuclear p65 levels, using laser scanning cytometer. Results: FACS analysis demonstrated a considerable drop in infiltrating lymphocytes and a drastic reduction in CD14+ cells in mice treated with p65 antisense oligonucleotides. Moreover, abrogation of inflammation extended all the way to the cecum in treated mice. Treatment was correlated with decreased DNA binding activity of NF-κB. Conclusions: Our data strongly support a model in which p65 antisense treatment possesses the capacity to disrupt the pathogenic autocrine loop propagated by NF-κB at the chronic phase of IBD.

Regulatory role of 5-HT and muscarinic receptor antagonists on the migrating myoelectric complex in rats

The 5-HT(3) and 5-HT(4) receptor antagonists alosetron and piboserod, and the muscarinic receptor antagonists PNU-171990A (2-(diisopropylamino)ethyl 1-phenylcyclopentanecarboxylate, hydrochloride) and PNU-174708A (2-(diisopropylamino)ethyl 1-phenylcyclohexanecarboxylate) were studied by electromyography, defining the migrating myoelectric complex (MMC) after i.v. administration in conscious rats. Alosetron prolonged the MMC cycle length from 16.6 to maximally 30.4 min at the dose 0.5 mg kg(-1). Piboserod promptly abolished MMC pattern and prolonged cycle length from 16.5 to >60 min at 0.5 mg kg(-1). PNU-171990A and PNU-174708A had no effect on basal cycle length up to a dose of 20 mg kg(-1). In controls, saline did not change the MMC pattern, while L-hyoscyamine at the same dose, 20 mg kg(-1), prolonged cycle length from 17.6 to 29.0 min. None of the drugs affected duration or propagation velocity of phase III of MMC. Blockade of 5-HT(4) receptors seems to exert a powerful inhibitory effect on motility, 5-HT(3) receptor blockade is less efficient and muscarinic receptor blockade has low efficacy.