Abstract 911 Poster Session IV, Tuesday, 5/4 (poster 131)

Previous studies indicate that the C. albicans INT1 gene is linked to filament formation, epithelial adhesion (HeLa cells), and mouse virulence. Because the intestine is a frequent portal of entry for systemic candidiasis, experiments were designed to clarify the effect of INT1 on C. albicans adherence to human enterocytes, namely Caco-2 cells. Using 24-well dishes and medium containing 15% serum, 106 confluent Caco-2 cells were incubated with C. albicans CAF2 (INT1/INT1), CAG1 (INT1/int1), CAG3 (int1/int1), or CAG5 (int1/int1+INT1). Adherence was determined by an ELISA assay: washed enterocytes were fixed, incubated with rabbit polyclonal anti-candidal serum followed horse radish peroxidase-conjugated anti-rabbit IgG and developed. Following 1 h incubation of 107 CAF2, CAG1, CAG3, or CAG5, ODs (avg±SE, n=14) were 2.7±0.3, 3.1±0.4, 2.3±0.3, 3.0±0.4, respectively; inocula of 106 (n=14) resulted in 1.2±0.3, 2.1±0.4, 1.6±0.2, 2.4±0.4, respectively; inocula of 105 (n=8) resulted in 0.2±0.1, 0.3±0.1, 0.2±0.1, 0.4±0.1, respectively. Statistical analysis indicated that adherence was similar for the four C. albicans strains (ANOVA), and was dependent on the inoculum size (r=0.6-0.7, P<0.01). Using light microscopy of Wright-Giemsa stained cultures, numbers of adherent candida appeared similar for the four strains, but precise quantitation was not possible due to C. albicans clumping; CAF2 readily formed filaments in this serum-containing medium, while filamentation of the other strains was noticeably reduced. Using low voltage (2kV) scanning electron microscopy, CAF2 germ tubes were easily located, while CAG1, CAG3, and CAG5 were predominantly yeast forms. The distal end of many germ tubes appeared intimately entwined with enterocyte microvilli, and in some instances, elongated microvilli completely engulfed the distal end of the germ tube. Observations of stereo pairs (10°tilt) confirmed that C. albicans adherence was often mediated exclusively by the distal germ tube. Clumps of C. albicans were readily observed, although only a portion of the candidal cells in these clumps actually contacted the enterocyte surface, suggesting that assays such as ELISA might artifactually quantify C. albicans adherent to each other as well as those adherent to the epithelial surface. Thus, INT1 did not appear to quantitatively affect C. albicans adherence to cultured enterocytes, although the tendency of C. albicans to adhere to itself might confound data interpretation; disruption of INT1 noticeably interfered with germ tube formation, and ultrastructural observations indicated that germ tubes might penetrate the enterocyte surface by a mechanism involving intimate association of the distal end of the germ tube with enterocyte apical microvilli.