No abstract

SummaryVery little is known about the biological functions of pili that have recently been found to be expressed by important Gram-positive pathogens such as Corynebacterium diphtheriae, Streptococcus agalacticae, S. pneumoniae and S. pyogenes. Using various ex vivo tissue and cellular models, here we show that pili mediate adhesion of serotype M1 S. pyogenes strain SF370 to both human tonsil epithelium and primary human keratinocytes, which represent the two main sites of infection by this human-specific pathogen. Mutants lacking minor pilus subunits retained the ability to express cell-surface pili, but these were functionally defective. In contrast to above, pili were not required for S. pyogenes adhesion to either immortalized HEp-2 or A549 cells, highlighting an important limitation of these extensively used adhesion/invasion models. Adhering bacteria were internalized very effectively by both HEp-2 and A549 cells, but not by tonsil epithelium or primary keratinocytes. While pili acted as the primary adhesin, the surface M1 protein clearly enhanced adhesion to tonsil, but surprisingly, had the opposite effect on adhesion to keratinocytes. These studies provide clear evidence that S. pyogenes pili display an adhesive specificity for clinically relevant human tissues and are likely to play a critical role in the initial stages of infection.

No abstract

We investigated the binding of four lectins to the follicle-associated epithelium (FAE) overlying fixed mouse small intestinal Peyer's patches to identify M-cell-specific surface markers. Wheat germ agglutinin and peanut agglutinin displayed heterogeneous staining patterns, binding most avidly to the intestine goblet cells. In contrast, the lectins Ulex europaeus 1 (UEA 1) and Psophocarpus tetragonolobus (winged bean; WBA) were almost exclusively M-cell specific. When confocal laser scanning images of tissues stained with fluorescein isothiocyanate (FITC)-conjugated UEA1 or WBA were compared with the appearance of the same tissues under the scanning electron microscope (SEM), UEA1 strongly stained 97.2% (106/109) of M-cells, 0.6% (3/516) enterocytes, and 0% (0/28) goblet cells, whereas WBA stained 100% (83/83) M-cells, 1.7% (6/361) enterocytes, and 5.3% (1/19) goblet cells. The M-cell specificity of the lectin binding was further demonstrated by localization of horseradish peroxidase (HRP)-conjugated lectins under the transmission electron microscope (TEM). This is the first demonstration of carbohydrates in the glycocalyx of M-cells that are not expressed elsewhere on the FAE surface. These carbohydrates not only provide a means to identify mouse M-cells by LM but may also contribute to the occurrence of specific interactions between microorganisms and the M-cell apical membrane.

No abstract