N-glycosylated proteins are involved in efficient internalization of Klebsiella pneumoniae by cultured human epithelial cells.

Klebsiella pneumoniae obtained from patients with urinary tract infections is able to invade cultured human epithelial cells. The internalization process is dependent upon both microfilaments and microtubules. To better understand the interaction of these invasive bacteria with the host cell receptor(s), bladder, lung, and ileocecal epithelial cells were infected with K. pneumoniae in the presence of various lectins possessing multiple glycan specificities. It was found that the N-acetylglucosamine (GlcNAc)-specific lectins concanavalin A, Datura stramonium agglutinin, and wheat germ agglutinin significantly inhibited the invasion of K. pneumoniae into these cells but did not interfere with the internalization of an invasive strain of Salmonella typhimurium. Conversely, internalization of K. pneumoniae but not S. typhimurium was also significantly inhibited when the bacteria were pretreated with GlcNAc or chitin hydrolysate, a GlcNAc polymer, prior to the gentamicin invasion assay. Other carbohydrates such as glucose, galactose, mannose, fucose, and N-acetylneuraminic acid had no inhibitory effects on K. pneumoniae uptake. Furthermore, internalization of K. pneumoniae but not S. typhimurium by HCT8 cells was also significantly inhibited when eukaryotic protein glycosylation was interrupted by tunicamycin or when host N-linked surface glycans were removed by pretreatment with N-glycosidase F. These studies suggest that a N-glycosylated protein receptor is involved in the internalization of K. pneumoniae by human epithelial cells in vitro. The results also indicate that internal GlcNAc residues might be a carbohydrate component of the receptor.

Induction of micronuclei by mitomycin C and colchicine in the marine mussel Mytilus galloprovincialis.

The frequencies of micronuclei (MNi) in the gill cells of the mussel Mytilus galloprovincialis were determined over a long period (up to 28 days) following a 48-h treatment with colchicine. The frequency of MNi at the end of treatment was significantly higher than in controls and 24 h later it had increased even more. After this period, the frequency of MNi rapidly declined until a plateau level was reached on day 2-3, which was significantly higher than the control baseline level, and persisted until the end of the experiment (28th day). In the same cell system we previously reported a persistence of an increased frequency of MNi after treatment with mitomycin C (MMC) (Majone et al., 1987). In order to establish the origin of MNi, the difference between their size distribution in MMC- and colchicine-treated animals was determined at the end of treatment as well as during the plateau phase. The difference was statistically significant (P less than 0.001) at the end of treatment, the MNi induced by MMC being smaller than those induced by colchicine. However, the difference during the plateau phase was not statistically significant. Human CREST antikinetochore fluorescent antibodies reacted with chromosome centromeres of Mytilus and were applied to gill cells at the end of a 48-h treatment with MMC or colchicine. About 60% of the MNi induced by colchicine but only 30% of those produced by MMC reacted positively with the fluorescent antibodies. This result indicates that the majority of MNi observed at the end of a 48-h treatment with MMC or colchicine originate, respectively, from acentric chromosome fragments and from whole lagging chromosomes.