Acidic pH of the Golgi lumen is known to be crucial for correct glycosylation, transport and sorting of proteins and lipids during their transit through the organelle. To better understand why Golgi acidity is important for these processes, we have examined here the most pH sensitive events in N-glycosylation by sequentially raising Golgi luminal pH with chloroquine (CQ), a weak base. We show that only a 0.2 pH unit increase (20 microM CQ) is sufficient to markedly impair terminal alpha(2,3)-sialylation of an N-glycosylated reporter protein (CEA), and to induce selective mislocalization of the corresponding alpha(2,3)-sialyltransferase (ST3) into the endosomal compartments. Much higher pH increase was required to impair alpha(2,6)-sialylation, or the proximal glycosylation steps such as beta(1,4)-galactosylation or acquisition of Endo H resistance, and the steady-state localization of the key enzymes responsible for these modifications (ST6, GalT I, MANII). The overall Golgi morphology also remained unaltered, except when Golgi pH was raised close to neutral. By using transmembrane domain chimeras between the ST6 and ST3, we also show that the luminal domain of the ST6 is mainly responsible for its less pH sensitive localization in the Golgi. Collectively, these results emphasize that moderate Golgi pH alterations such as those detected in cancer cells can impair N-glycosylation by inducing selective mislocalization of only certain Golgi glycosyltransferases.

Glycans (i.e. oligosaccharide chains attached to cellular proteins and lipids) are crucial for nearly all aspects of life, including the development of multicellular organisms. They come in multiple forms, and much of this diversity between molecules, cells, and tissues is generated by Golgi-resident glycosidases and glycosyltransferases. However, their exact mode of functioning in glycan processing is currently unclear. Here we investigate the supramolecular organization of the N-glycosylation pathway in live cells by utilizing the bimolecular fluorescence complementation approach. We show that all four N-glycosylation enzymes tested (␤-1,2-N-acetylglucosaminyltransferase I, ␤-1,2-N-acetylglucosaminyltransferase II, 1,4-galactosyltransferaseI,and␣-2,6-sialyltransferaseI)formGolgi-localizedhomodimers. Intriguingly, the same enzymes also formed two distinct and functionally relevant heterodimers between the medial Golgi enzymes ␤-1,2-N-acetylglucosaminyltransferase I and ␤-1,2-N-acetylglucosaminyltransferase II and the trans-Golgi enzymes 1,4-galactosyltransferase I and ␣-2,6-sialyltransferase I. Given their strict Golgi localization and sequential order of function, the two heterodimeric complexes are probably responsible for the processing and maturation of N-glycans in live cells.

Aberrant secretion of lysosomal hydrolases such as (pro)cathepsin D (proCD) is a common phenotypic change in many human cancers. Here we explore the underlying molecular defect(s) and find that MCF-7 breast and CaCo-2 colorectal cancer cells that are unable to acidify their endosomal compartments secreted higher amounts of proCD than did acidification-competent cancer cell types. The latter secreted equivalent amounts of proCD only after dissipation of their organellar pH gradients with NH 4 Cl. Assessing the critical steps that resulted in proCD secretion revealed that the Golgi-associated sorting receptor for CD, i.e. the cationindependent mannose-6-phosphate receptor (MPR300), was aberrantly distributed in acidification-defective MCF-7 cells. It accumulated mainly in late endosomes and/or lysosomes as a complex with its ligand (proCD or intermediate CD), as evidenced by its co-localization with both CD and LAMP-2, a late endosome/lysosome marker. Our immunoprecipitation analyses also showed that MCF-7 cells possessed 7-fold higher levels of receptor-enzyme complexes than did acidification-competent cells. NH 4 Cl induced similar receptor redistribution into LAMP-2-positive structures in acidification-competent cells but not in MCF-7 cells. The receptor also recovered its normal Golgi localization upon drug removal. Based on these observations, we conclude that defective acidification results in the aberrant secretion of proCD in certain cancer cells and interferes mainly with the normal disassembly of the receptor-enzyme complexes and efficient receptor reutilization in the Golgi.

Protein folding and quality control in the endoplasmic reticulum are critical processes for which our current understanding is far from complete. Here we describe the functional characterization of a new human 27.7-kDa protein (ERp27). We show that ERp27 is a two-domain protein located in the endoplasmic reticulum that is homologous to the non-catalytic b and b domains of protein disulfide isomerase. ERp27 was shown to bind ⌬-somatostatin, the standard test peptide for protein disulfide isomerase-substrate binding, and this ability was localized to the second domain of ERp27. An alignment of human ERp27 and human protein disulfide isomerase allowed for the putative identification of the peptide binding site of ERp27 indicating conservation of the location of the primary substrate binding site within the protein disulfide isomerase family. NMR studies revealed a significant conformational change in the b-like domain of ERp27 upon substrate binding, which was not just localized to the substrate binding site. In addition, we report that ERp27 is bound by ERp57 both in vitro and in vivo by a similar mechanism by which ERp57 binds calreticulin.The formation of native disulfide bonds in the endoplasmic reticulum (ER) 2 is a complex, but essential, process in the biogenesis of many proteins, which pass beyond the carefully regulated environment of the cell. The folding process can occur via multiple parallel pathways within a protein (see for example Ref. 1), and this process can be catalyzed in vivo by different biological molecules. For example, there appears to be at least three parallel pathways for oxidation (disulfide bond formation) in substrate proteins, direct oxidation by Ero1 (2-4) or flavin-dependent sulfhydryl oxidases (5, 6), oxidation catalyzed by proteins belonging to the thioredoxin superfamily (for reviews see Refs. 7-9), and oxidation by low molecular weight compounds such as oxidized glutathione (10, 11).The first class of enzymes reported to be involved in disulfide bond formation was the protein disulfide isomerase (PDI) family (7,9,12). Although the founding member of this family, PDI, is sufficient in vitro, when combined with a physiological glutathione redox buffer, to refold reduced protein substrates or to isomerize substrates with incorrect disulfide bonds (13-15), it is clear that in vivo there exists a large family of enzymes. There is evidence that the exact make-up of this family is species-dependent, with, as an extreme example, the yeast Saccharomyces cerevisiae having five reported family members (16), whereas humans have at least seventeen (9), not all of which have been characterized yet. Although the family name suggests that all of the members of the family are able to catalyze isomerization reactions in folding protein substrates, this is not the case. For example, one sub-group of the family, which includes ERp28/29 (17) and Drosophila wind protein (18), lacks both cysteines from the CXXC active site motif that are essential for the catalysis of dithiol-disulfide exchange rea...

Carcinoembryonic antigen (CEA, ceacam5) is an important tumor-associated antigen with reported roles, e.g., in immunological defense, cell adhesion, cell survival and metastasis. Its overexpression in cancer cells is known to involve transcriptional activation of the CEA gene, but the underlying molecular details remain unclear. Here, we show that hypoxia and intracellular alkalinization, 2 factors commonly found in solid tumors, increase CEA protein expression in breast (MCF-7) and colorectal (CaCo-2 and HT-29) cancer cells. The increase was comparable (2-3-fold) to that observed in colorectal carcinomas in vivo. CEA promoter analyses further revealed that this upregulation involves a known binding site for HIF-1 transcription factor (5 0 -ACGTG-3 0 ) within one of the CEA promoter's positive regulatory elements (the FP1 site; the E-box). Accordingly, deletion or targeted mutagenesis of this motif rendered the CEA promoter unresponsive to hypoxia. Our chromatin immunoprecipitation data confirmed that endogenous HIF-1a binds to the CEA promoter in hypoxic cells but not in normoxic cells. Moreover, overexpression of the hypoxia-inducible factor (HIF-1a) was sufficient to increase CEA protein expression in the cells. In contrast, c-Myc, which is known to bind to the overlapping E-box, did not potentiate HIF-1a-induced CEA expression. CEA overexpression in vivo was also found to coincide with the expression of carbonic anhydrase IX, a well-known hypoxia marker. Collectively, these results define CEA as a hypoxia-inducible protein and suggest an important role for the tumor microenvironmental factors in CEA overexpression during tumorigenesis. ' 2007 Wiley-Liss, Inc.Key words: neoplasia; carcinoembryonic antigen expression; hypoxia; pH homeostasis Carcinoembryonic antigen (CEA, ceacam5), originally identified about 40 years ago, 1 is a widely used tumor marker and also currently a target for vaccine-based immunotherapy. 2,3 It is a 180-200 kDa glycoprotein attached to the apical membrane of epithelial cells via its C-terminal glycosylphosphatidylinositol (GPI) anchor. It has been assigned multiple functions, including a role in immunological defense, cell signaling and cell adhesion. When overexpressed, it has also been shown to inhibit apoptosis, anoikis, cell polarity and differentiation, 3-9 and to promote anchorage-independent growth and tumor formation in nude mice. 10 CEA expression commences during the early fetal life and continues thereafter at low levels mainly in the epithelial cells of the gastrointestinal tract, cervix, sweat glands and the prostate. 3 It is re-expressed at higher levels (2-6-fold) in many cancers. 2,11,12 Previous studies 13 have shown that this upregulation does not involve gene amplification nor its rearrangements. Rather, CEA protein levels were found to correlate with its mRNA levels in the cells and tissues examined, suggesting that CEA overexpression in cancer cells involves transcriptional activation of the CEA gene. However, the underlying molecular details have remained uncle...