Cytokine levels in nasal and lower airways in young cystic fibrosis (CF) patients were compared with those in controls. Nasal (NLF) and bronchoalveolar (BALF) lavage fluids were obtained from children with or without CF who were undergoing bronchoscopy for clinical indications. In NLF, neither inflammatory cells nor cytokine concentrations differed between patients and controls. However, interleukin (IL)-8 levels in infected BALF from children with CF were markedly elevated compared with levels in infected and uninfected controls, even after standardization of IL-8 concentrations to bacterial counts. BALF IL-6 was modestly elevated in infected CF patients compared with uninfected but not infected controls; IL-10 did not differ among the groups. NLF and BALF IL-8 levels were not significantly correlated. Excessive airway inflammation in early CF thus appears to be confined to the lower respiratory tract, and IL-8 levels are markedly increased in children with CF compared with control children with a bacterial infection of the lower airways.

Prostate cancer progression is associated with up-regulation of sialyl-T antigen produced by β-galactoside α-2,3-sialyltransferase-1 (ST3Gal1) but not with core 2-associated polylactosamine despite expression of core 2 N-acetylglucosaminyltransferase-L (C2GnT-L/GCNT1). This property allows androgen-refractory prostate cancer cells to evade galectin-1 (LGALS1)-induced apoptosis, but the mechanism is not known. We have recently reported that Golgi targeting of glycosyltransferases is mediated by golgins: giantin (GOLGB1) for C2GnT-M (GCNT3) and GM130 (GOLGA2)-GRASP65 (GORASP1) or GM130-giantin for core 1 synthase. Here, we show that for Golgi targeting, C2GnT-L also uses giantin exclusively while ST3Gal1 employs either giantin or GM130-GRASP65. In addition, the compact Golgi morphology is detected in both androgen-sensitive prostate cancer and normal prostate cells, but fragmented Golgi and mislocalization of C2GnT-L are found in androgen-refractory cells as well as primary prostate tumors (Gleason grade 2–4). Furthermore, failure of giantin monomers to be phosphorylated and dimerized prevents Golgi from forming compact morphology and C2GnT-L from targeting the Golgi. On the other hand, ST3Gal1 reaches the Golgi by an alternate site, GM130-GRASP65. Interestingly, inhibition or knockdown of non-muscle myosin IIA (MYH9) motor protein frees up Rab6a GTPase to promote phosphorylation of giantin by polo-like kinase 3 (PLK3), which is followed by dimerization of giantin assisted by protein disulfide isomerase A3 (PDIA3), and restoration of compact Golgi morphology and targeting of C2GnT-L. Finally, the Golgi relocation of C2GnT-L in androgen-refractory cells results in their increased susceptibility to galectin-1-induced apoptosis by replacing sialyl-T antigen with polylactosamine.

A high-efficiency, nonviral gene transfer protocol employing cationic liposome plus a receptor ligand is described. The delivery of the beta-galactosidase (beta-Gal) gene (pCMVlacZ) by lipofectin plus transferrin can achieve 98-100% transfection of HeLa cells as compared to 3-4% by lipofectin alone. A dose-dependent gene transfer was observed between 1 and 16 micrograms transferrin, and maximal transfection efficiency was obtained at > or = 16 micrograms transferrin. The expression of beta-Gal activity in 100% transfected cells decreased progressively with each passage and returned to the baseline value after six passages, indicating that the DNA delivered was only transiently expressed. The amount of DNA delivered to the cells by lipofectin plus transferrin was approximately two times that obtained by lipofectin, which in turn was two times that by transferrin or without lipofectin and transferrin. In addition, DNA can form complexes with lipofectin and transferrin. These results suggest that transferrin enhances gene transfer and expression in the presence of lipofectin by further facilitating the entry of DNA into the cells through the lipofectin-DNA-transferrin complex. The enhancement of liposome-mediated gene transfer efficiency and expression by transferrin varies with different cationic liposomes. The four different liposomes examined show the following relative transfection efficiency: transfectin > lipofectACE > > DC-cholesterol > > lipofectAMINE.

The Golgi apparatus undergoes morphological changes under stress or malignant transformation, but the precise mechanisms are not known. We recently showed that nonmuscle myosin IIA (NMIIA) binds to the cytoplasmic tail of Core 2 N-acetylglucosaminyltransferase mucus-type (C2GnT-M) and transports it to the endoplasmic reticulum for recycling. Here, we report that Golgi fragmentation induced by brefeldin A (BFA) or coatomer protein (β-COP) knockdown (KD) in Panc1-bC2GnT-M (c-Myc) cells is accompanied by the increased association of NMIIA with C2GnT-M and its degradation by proteasomes. Golgi fragmentation is prevented by inhibition or KD of NMIIA. Using multiple approaches, we have shown that the speed of BFA-induced Golgi fragmentation is positively correlated with the levels of this enzyme in the Golgi. The observation is reproduced in LNCaP cells which express high levels of two endogenous glycosyltransferases-C2GnT-L and βgalactoside α2,3 sialyltransferase 1. NMIIA is found to form complexes with these two enzymes but not Golgi matrix proteins. The KD of both enzymes or the prevention of Golgi glycosyltransferases from exiting endoplasmic reticulum reduced Golgi-associated NMIIA and decreased the BFAinduced fragmentation. Interestingly, the fragmented Golgi detected in colon cancer HT-29 cells can be restored to a compact morphology after inhibition or KD of NMIIA. The Golgi disorganization induced by the microtubule or actin destructive agent is NMIIA-independent and does not affect the levels of glycosyltransferases. We conclude that NMIIA interacts with Golgi residential but not matrix proteins, and this interaction is responsible for Golgi fragmentation induced by β-COP KD or BFA treatment. This is a novel non-enzymatic function of Golgi glycosyltransferases.