Mechanisms of association of bacteria with mucosal surfaces

How the Interplay Between the Commensal Microbiota, Gut Barrier Integrity, and Mucosal Immunity Regulates Brain Autoimmunity

The intestinal barrier provides the host with a strong defense line against the external environment playing also a pivotal role in the crosstalk between the gut microbiota and the immune system. Notably, increasing lines of evidence concerning autoimmune disorders such as Multiple Sclerosis (MS) report an imbalance in both intestinal microbiota composition and mucosal immunity activation, along with an alteration of gut barrier permeability, suggesting this complex network plays a crucial role in modulating the course of autoimmune responses occurring in tissues outside the gut such as the central nervous system (CNS). Here, we review current knowledge on how gut inflammation and breakage of gut barrier integrity modulates the interplay between the commensal gut microbiota and the immune system and its role in shaping brain immunity.

Chemoreceptors of Escherichia coli CFT073 play redundant roles in chemotaxis toward urine

Community-acquired urinary tract infections (UTIs) are commonly caused by uropathogenic Escherichia coli (UPEC). We hypothesize that chemotaxis toward ligands present in urine could direct UPEC into and up the urinary tract. Wild-type E. coli CFT073 and chemoreceptor mutants with tsr, tar, or aer deletions were tested for chemotaxis toward human urine in the capillary tube assay. Wild-type CFT073 was attracted toward urine, and Tsr and Tar were the chemoreceptors mainly responsible for mediating this response. The individual components of urine including L-amino acids, D-amino acids and various organic compounds were also tested in the capillary assay with wild-type CFT073. Our results indicate that CFT073 is attracted toward some L- amino acids and possibly toward some D-amino acids but not other common compounds found in urine such as urea, creatinine and glucuronic acid. In the murine model of UTI, the loss of any two chemoreceptors did not affect the ability of the bacteria to compete with the wild-type strain. Our data suggest that the presence of any strong attractant and its associated chemoreceptor might be sufficient for colonization of the urinary tract and that amino acids are the main chemoattractants for E. coli strain CFT073 in this niche.

Unified theory of bacterial sialometabolism: how and why bacteria metabolize host sialic acids

Sialic acids are structurally diverse nine-carbon ketosugars found mostly in humans and other animals as the terminal units on carbohydrate chains linked to proteins or lipids. The sialic acids function in cell-cell and cell-molecule interactions necessary for organismic development and homeostasis. They not only pose a barrier to microorganisms inhabiting or invading an animal mucosal surface, but also present a source of potential carbon, nitrogen, and cell wall metabolites necessary for bacterial colonization, persistence, growth, and, occasionally, disease. The explosion of microbial genomic sequencing projects reveals remarkable diversity in bacterial sialic acid metabolic potential. How bacteria exploit host sialic acids includes a surprisingly complex array of metabolic and regulatory capabilities that is just now entering a mature research stage. This paper attempts to describe the variety of bacterial sialometabolic systems by focusing on recent advances at the molecular and host-microbe-interaction levels. The hope is that this focus will provide a framework for further research that holds promise for better understanding of the metabolic interplay between bacterial growth and the host environment. An ability to modify or block this interplay has already yielded important new insights into potentially new therapeutic approaches for modifying or blocking bacterial colonization or infection.

Preservative-mediated changes of the surface properties of Escherichia coli

The effectiveness of two commonly used preservatives, sodium benzoate and potassium disulfite, was evaluated in terms of their bactericidal activity and capacity to induce changes in the surface properties of Escherichia coli isolated from commercial food preserves. Preservative treatment over a five-week test period resulted in controlling the multiplication of these organisms and causing a decline in cell-surface hydrophobicity, hemagglutinating ability and adherence capacity to rat intestinal cells of E. coli isolates. A loss in motility was also exhibited.

Intact motility as a Salmonella typhi invasion-related factor

Invasiveness of Salmonella typhi was investigated. At first, we introduced Tn5 into the chromosome of a wild-type S. typhi strain, GIFU 10007, and screened the independent Tn5 insertion mutants for noninvasive (Inv-) strains. During the first half of this work, we obtained 4 Inv- strains from 1,338 independent Tn5 mutants. The four were either nonflagellate (Fla-), nonmotile (Mot-), or nonchemotactic (Che-). We then isolated more Fla-, Mot-, or Che- mutants and examined the invasiveness of these mutants. Sixty-three spontaneous or Tn5 insertion motility mutants, i.e., Fla-, Mot-, or Che-, were independently isolated from the wild-type strain GIFU 10007; all of them were noninvasive. Motile revertants isolated from some of these mutants showed the same invasiveness as the parent strain. P22-mediated transductional crosses were carried out between some of the motility mutants (as the recipients) and the Fla- reference strains of S. typhimurium with known deletion sites on the genome (as the donors). The mutational sites of the S. typhi mutants were assigned almost evenly to the three flagellar gene regions (regions I, II, and III) of S. typhimurium. The invasiveness of the motile recombinants obtained from the transduction assays was examined. The restoration of intact motility resulted in the restoration of invasiveness. Thus, we conclude that intact motility is an invasion-related factor of S. typhi. The relationship of Vi antigen to the invasiveness of S. typhi was also studied. Vi-negative mutants with intact motility remained invasive, whereas all 63 Inv- spontaneous or Tn5 mutants were Vi positive. Therefore, Vi antigen was not related to the invasiveness of S. typhi.

Colonization of the streptomycin-treated mouse large intestine by a human fecal Escherichia coli strain: role of adhesion to mucosal receptors

Escherichia coli F-18, a normal fecal isolate, was previously shown to be an excellent colonizer of the streptomycin-treated CD-1 mouse large intestine, whereas E. coli F-18col-, a derivative of E. coli F-18 that no longer makes the E. coli F-18 colicin, was shown to be a poor mouse colonizer. It was also shown that E. coli F-18 bound two to three times more soluble colonic mucus protein than did E. coli F-18col- and that a major receptor in CD-1 mouse colonic mucus was a 50.5-kilodalton glycoprotein. In the present investigation, an additional E. coli F-18 colonic mucus glycoprotein receptor (66 kilodaltons) and three cecal mucus glycoprotein receptors (94, 73, and 66 kilodaltons) were identified. Numerous colonic and cecal brush border protein receptors specific for E. coli F-18 were also identified. Furthermore, E. coli F-18col- was found to bind to the same mucus and brush border receptors as E. coli F-18, although to a far lesser extent. Adhesion of both E. coli F-18 and F-18col- was inhibited by D-mannose and alpha-methyl-D-mannoside, and both strains were shown to bind specifically to the mannose moiety of a mannose-bovine serum albumin glycoconjugate, although again E. coli F-18col- bound to a lesser extent. Finally, both E. coli F-18 and F-18col- were shown to be piliated. The possible role of pilus mediated adhesion in E. coli F-18 colonization of the streptomycin-treated mouse large intestine is discussed.

Colonization of the streptomycin-treated mouse large intestine by a human fecal Escherichia coli strain: role of growth in mucus

The relative colonizing abilities of Escherichia coli F-18, isolated from the feces of a healthy human, and E. coli F-18col-, a strain derived from it which does not make the E. coli F-18 colicin, were studied. In a previous report, it was shown that when each strain was fed individually to streptomycin-treated mice, at approximately 10(10) CFU per mouse, each colonized the large intestine at between 10(7) and 10(8) CFU/g of feces indefinitely. However, when simultaneously fed to mice, although E. coli F-18 colonized at about 10(8) CFU/g of feces, E. coli F-18col- dropped to a level of 10(3) CFU/g of feces within 3 to 5 days. In the present investigation, we show that when given enough time to establish a state of colonization, E. coli F-18col- persists in feces in high numbers despite subsequent challenge by E. coli F-18. Therefore, a major defect in the ability of E. coli F-18col- to colonize in the presence of E. coli F-18 appears to be in initiating that state. In addition, when mucus was scraped from the cecal wall and, without further treatment, was inoculated with E. coli F-18 or F-18col-, both strains grew well. However, when cecal mucus was inoculated with both strains simultaneously, E. coli F-18 grew far more rapidly than E. coli F-18col-. Moreover, neither strain grew in cecal luminal contents. Together, these data suggest the possibility that both E. coli F-18 and F-18col- must grow in mucus to colonize the streptomycin-treated mouse large intestine, that E. coli F-18col- is eliminated by E. coli F-18 because it does not grow in mucus as well as E. coli F-18, and that E. coli F-18col- can resist elimination by E. coli F-18 if it is allowed enough time to establish itself within the mucus layer.

Pathogenicity and virulence: another view

The concepts of pathogenicity and virulence have governed our perception of microbial harmfulness since the time of Pasteur and Koch. These concepts resulted in the recognition and identification of numerous etiological agents and provided natural and synthetic agents effective in therapy and prevention of diseases. However, Koch's postulates--the premier product of this view--place the onus of harmfulness solely on the microbial world. Our recent experiences with polymicrobic and nosocomial infections, legionellosis, and acquired immunodeficiency syndrome point to the host as the major determinant of disease. The principles of parasitism, enunciated by Theobold Smith, approximate more accurately the disturbances of the host-parasite equilibrium we designate as infection. Many complex attributes of microbial anatomy and physiology have been obscured by our dependency on the pure-culture technique. For example, bacterial attachment organelles and the production of exopolysaccharides enable microorganisms to interact with mammalian glycocalyces and specific receptors. In addition, selection, through the use of therapeutic agents, aids in the progression of environmental organisms to members of the intimate human biosphere, with the potential to complicate the recovery of patients. These factors emphasize further the pivotal significance of host reactions in infections. Parasitism, in its negative aspects, explains the emergence of "new" infections that involve harm to more than host organs and cells: we may encounter subtler infections that reveal parasitic and host cell nucleic acid interactions in a form of genomic parasitism.

In vivo colonization of the mouse large intestine and in vitro penetration of intestinal mucus by an avirulent smooth strain of Salmonella typhimurium and its lipopolysaccharide-deficient mutant

The relative abilities of an avirulent Salmonella typhimurium strain with wild-type lipopolysaccharide (LPS) character, SL5319, and a nearly isogenic LPS-deficient mutant, SL5325, to colonize the large intestines of streptomycin-treated CD-1 mice in vivo and to penetrate colonic mucus in vitro were studied. Previously it had been shown that, when fed simultaneously to streptomycin-treated mice (approximately 10(10) CFU each), the S. typhimurium strain with wild-type LPS colonized at 10(8) CFU/g of feces indefinitely, whereas the LPS-deficient mutant dropped within 3 days to a level of only 10(4) CFU/g of feces. In the present investigation, when SL5325 was allowed to colonize for 8 days before feeding mice SL5319 or when it was fed to mice simultaneously with an Escherichia coli strain of human fecal origin (10(10) CFU each), both strains colonized indefinitely at 10(7) CFU/g of feces. Moreover, when the wild-type and LPS-deficient mutant strains were fed to mice simultaneously in low numbers (approximately 10(5) CFU each) the strains survived equally well in the large intestines for 8 days, after which the LPS-deficient mutant was eliminated (less than 10(2) CFU/g of feces), whereas the wild-type colonized at a level of 10(7) CFU/g of feces. In addition although both strains were able to adhere to mucus and epithelial cell preparations in vitro, the wild-type strain was shown to have greater motility and chemotactic activity on CD-1 mouse colonic mucus in vitro and to more rapidly penetrate and form a stable association with immobilized colonic mucosal components in vitro. Based on these data, we suggest that the ability of an S. typhimurium strain to colonize the streptomycin-treated mouse large intestine may, in part, depend on its ability to penetrate deeply into the mucus layer on the intestinal wall and subsequently, through growth, colonize the mucosa.

Adherence of Salmonella typhimurium to small-intestinal enterocytes of the rat

The adherence of radiolabeled Salmonella typhimurium to freshly isolated enterocytes of rats was studied. The results established that type 1 fimbriated strains adhered in significantly higher numbers than did related nonfimbriated strains. Adherence was inhibited by D-mannose and methyl alpha-D-mannoside. Results of kinetic studies indicated that adherence was biphasic; the number of bacteria that adhered per enterocyte remained constant for approximately 20 min and then increased rapidly under the assay conditions. The second phase was associated with structural damage to the enterocytes. The addition of chloramphenicol did not prevent the initial attachment of bacteria to enterocytes but did prevent the second phase. Viable and nonviable bacterial cells adhered to enterocytes, but only viable bacteria were destructive. Freshly isolated enterocytes (trypan blue impermeable) and enterocytes stored overnight (trypan blue permeable) were infected by viable S. typhimurium in a similar manner, suggesting that metabolic activity of the host cell was of less consequence than metabolic activity of the bacterial cells. A model for the role of mannose-sensitive fimbriae as a virulence factor is proposed.

Pilus-mediated binding of bovine enterotoxigenic Escherichia coli to calf small intestinal mucins

In this study we show that the adhesion to mucus of the enterotoxigenic Escherichia coli strains responsible for diarrhea in calves involves a bacterium-mucin recognition phenomenon in which the bacterial pili and specific mucus receptors carried by the glycoproteins (2,000 to 400 kilodalton) play a major role. An adhesion maximum was observed at a pH of less than 6 (4.75 to 5.25). The sialic acids and galactose appeared to be at least partly responsible for the attachment of K99 pili, whereas F41 pili preferentially recognized desialylated receptors. The attachment of different strains of E. coli characterized by the presence of the three main pili, K99, F41, and FY, known to be responsible for the binding of enterotoxigenic E. coli to the intestinal epithelium of the calf, was studied using Scatchard and Hill analyses. The attachment mechanism of bacteria carrying K99 pili showed positive cooperativity. FY and F41 pili recognized independent receptor sites, the first on sialylated mucus and the second on sialidase-treated mucus. Moreover, F41 pili were found to bind the native mucus according to a negative cooperativity phenomenon. Finally, the recognition sites carried by bacterial pilins may be saturated by some animal glycoprotein glycans which are therefore adhesion inhibitors.

Ecology of Candida albicans gut colonization: inhibition of Candida adhesion, colonization, and dissemination from the gastrointestinal tract by bacterial antagonism

Antibiotic-treated and untreated Syrian hamsters were inoculated intragastrically with Candida albicans to determine whether C. albicans could opportunistically colonize the gastrointestinal tract and disseminate to visceral organs. Antibiotic treatment decreased the total population levels of the indigenous bacterial flora and predisposed hamsters to gastrointestinal overgrowth and subsequent systemic dissemination by C. albicans in 86% of the animals. Both control hamsters not given antibiotics and antibiotic-treated animals reconventionalized with an indigenous microflora showed significantly lower gut populations of C. albicans, and C. albicans organisms were cultured from the visceral organs of 0 and 10% of the animals, respectively. Conversely, non-antibiotic-treated hamsters inoculated repeatedly with C. albicans had high numbers of C. albicans in the gut, and viable C. albicans was recovered from the visceral organs of 53% of the animals. Examination of the mucosal surfaces from test and control animals indicated further that animals which contained a complex indigenous microflora had significantly lower numbers of C. albicans associated with their gut walls than did antibiotic-treated animals. The ability of C. albicans to associate with intestinal mucosal surfaces also was tested by an in vitro adhesion assay. The results indicate that the indigenous microflora reduced the mucosal association of C. albicans by forming a dense layer of bacteria in the mucus gel, out-competing yeast cells for adhesion sites, and producing inhibitor substances (possibly volatile fatty acids, secondary bile acids, or both) that reduced C. albicans adhesion. It is suggested, therefore, that the indigenous intestinal microflora suppresses C. albicans colonization and dissemination from the gut by inhibiting Candida-mucosal association and reducing C. albicans population levels in the gut.

Adhesion of a human fecal Escherichia coli strain to mouse colonic mucus

Escherichia coli F-18 isolated from the feces of a healthy human is an excellent colonizer of the CD-1 mouse colon. In the present investigation, adhesion of E. coli F-18 to CD-1 mouse colonic mucus and bovine serum albumin (BSA), immobilized on polystyrene, was studied. Adhesion of E. coli F-18 to mucus was two- to sixfold greater than to either BSA or polystyrene. E. coli F-18 lipopolysaccharide specifically blocked adhesion of E. coli F-18 to mucus and mimicked adhesion of E. coli F-18 to mucus, BSA, and polystyrene. Purified capsule also blocked adhesion of E. coli F-18 to mucus, but this inhibition was found to be entirely nonspecific. The specific E. coli F-18 receptor in mucus appeared to be a glycoprotein, containing sugars normally found in mucins and having a maximum molecular weight of between 1.25 X 10(5) and 2.5 X 10(5).

Passage of Salmonella enteritidis and Salmonella thompson through chick ileocecal mucosa

The passage of Salmonella enteritidis and S. thompson across the cecal mucosa has been visualized in an electron microscope study with the freshly hatched chick as a model. The uptake of salmonellae by macrophages took place in the cecal lumen; the macrophages became abnormal and often ruptured to release organisms back into the lumen. The entry of bacteria into the epithelial cells was associated with a series of pathological changes, beginning with the appearance of active Golgi apparatus and the production of a variety of lysosomal vesicles. Salmonellae became sequestered within lysosomes but were unaffected by the presence of hydrolytic enzyme. Epithelial cell death was related to particularly large numbers of bacteria. Fragments of invaded epithelial cells, especially those undergoing cell death, contributed to the cytoplasmic debris and released further salmonellae into the lumen. Bacteria were never observed in large numbers below the basement membrane, and there was no significant pathology in the lamina propria tissue. Wandering cells, identified as macrophages and containing the bacteria, were observed spanning the epithelial and lamina propria regions through breaks in the basement membrane. It is suggested that the passage of bacteria from the epithelium to the lamina propria is primarily the result of capture and transport within host macrophages.

Cell-free released components of Streptococcus sanguis inhibit human platelet aggregation

To study the role of surface components in the selective binding and aggregation of platelet-rich plasma (PRP) by strains of viridans streptococci, we treated the binding, aggregation strain Streptococcus sanguis I 2017-78 by sonication or trypsinization. Morphologically identifiable electron-dense fibrils were released from the cell wall, apparently from an inner electron-dense layer, under conditions that left cells intact. These controlled conditions were determined to cause submaximal loss in adhesion to platelet ghosts and PRP aggregation by treated, washed S. sanguis. Soluble components were recovered from the controlled sonic or L-(tosylamido 2-phenyl)ethyl chloromethyl ketone-trypsin treatments. Each showed dose-response inhibition of aggregation when preincubated with PRP before challenge with fresh, untreated S. sanguis. The time to onset of PRP aggregation was inhibited by 50% with 0.2 mg of TPCK-trypsin peptides or 1.0 mg of the sonicate per ml per 2 X 10(8) platelets. Components of both preparations were immunologically cross-reactive, but lipoteichoic acid was not a major antigen of either. By weight, the TPCK-trypsin peptides were virtually all protein; the sonicate residues identified were about 50% protein and 7% hexose. Each was a complex mixture of components as shown by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. More than 8 TPCK-trypsin peptides and 16 sonicate components were so identified. In contrast, at least four or five components from either preparation were recognized as surface determinants by a rabbit antiserum to whole homologous microbes. Platelet-binding ligands of S. sanguis could be among these determinants.

Contribution of adhesion to bacterial persistence in the mouse urinary tract

The affinity of uropathogenic Escherichia coli to kidneys and bladders of experimentally infected mice was shown to be determined in part by the adhesive properties of the infecting bacteria. Mice were infected with various pairwise combinations of two homogeneic sets of bacteria: (i) mutants derived from a human pyelonephritis E. coli isolate which were selected to express either or both adhesins specific for globoseries glycolipid receptors or for "mannosides"; and (ii) transformants of a normal fecal isolate which harbored recombinant plasmids encoding the genes for one or the other adhesin or which harbored only the vector plasmid. The relative efficiency of survival of the strains to be compared was evaluated in each animal by plating on selective media of samples of homogenized kidneys and bladders taken 24 h after intravesical inoculation. The presence of adhesins specific for globoseries glycolipid receptors, which mediate the in vitro mannose-resistant attachment to human and mouse uroepithelial cells, enhanced bacterial recovery from both kidneys and bladders of infected animals. The addition to the infecting strain of adhesins binding mannoside residues further improved bacterial recovery from the bladder, but not from the kidney. The mutants and transformants with adhesins binding only mannosides were recovered in higher numbers from the bladder than those expressing adhesins specific for the globoseries glycolipids only. There was apparent selection in vivo decreasing expression of mannoside binding adhesins in the kidneys, but not in the bladders, of animals infected with the mutant expressing both types of adhesins. Regardless of adhesive properties, the mutants of the pyelonephritis isolate were recovered in significantly higher numbers than the fecal isolate with adhesins encoded on recombinant plasmids. We conclude that the adhesive properties in part determine the localization and retention of bacteria in the mouse urinary tract. However, the addition of adhesins to a commensal E. coli strain was not sufficient to confer colonization capacity comparable to that of a pyelonephritis strain.