Effects of O-Linked Glycosylation on the Cell Surface Expression and Stability of Decay-accelerating Factor, a Glycophospholipid-anchored Membrane Protein

O-glycosylation is essential for cell surface expression of the transcobalamin receptor CD320

CD320 is a cell surface receptor that mediates vitamin B12 uptake in most tissues. To date, the mechanisms that regulate CD320 expression on the cell surface are not fully understood. In this work, we studied CD320 expression in transfected human embryonic kidney (HEK) 293 and hepatoma HepG2 cells. By glycosidase and trypsin digestion, monensin and brefeldin treatment, western blotting, flow cytometry, and lectin binding, we found that CD320 underwent N- and O-glycosylation and sialylation, resulting in a ∼70-kDa band that formed a high-molecular-weight complex on the cell surface. Site-directed mutagenesis altering Asn126, Asn195, and Asn213 residues, individually or together, abolished N-glycosylation in CD320 but did not block its intracellular trafficking and expression on the cell surface in HEK293 and HepG2 cells. In contrast, treatment of the cells with Ben-gal, a structural analog of GalNAc-α-1-O-Ser/Thr, inhibited O-glycosylation and cell surface expression of CD320 and decreased vitamin B12 uptake. Analysis of CD320 deletion mutants indicated that O-glycosylation sites in a Ser/Thr-rich region near the transmembrane domain were important for CD320 expression on the cell surface. These results reveal an important role of O-glycans, but not N-glycans, in the intracellular trafficking and cell surface expression of CD320, providing new insights into the cellular mechanisms in regulating CD320 function and vitamin B12 metabolism.

Complement Decay-Accelerating Factor is a modulator of influenza A virus lung immunopathology

Clearance of viral infections, such as SARS-CoV-2 and influenza A virus (IAV), must be fine-tuned to eliminate the pathogen without causing immunopathology. As such, an aggressive initial innate immune response favors the host in contrast to a detrimental prolonged inflammation. The complement pathway bridges innate and adaptive immune system and contributes to the response by directly clearing pathogens or infected cells, as well as recruiting proinflammatory immune cells and regulating inflammation. However, the impact of modulating complement activation in viral infections is still unclear. In this work, we targeted the complement decay-accelerating factor (DAF/CD55), a surface protein that protects cells from non-specific complement attack, and analyzed its role in IAV infections. We found that DAF modulates IAV infection in vivo, via an interplay with the antigenic viral proteins hemagglutinin (HA) and neuraminidase (NA), in a strain specific manner. Our results reveal that, contrary to what could be expected, DAF potentiates complement activation, increasing the recruitment of neutrophils, monocytes and T cells. We also show that viral NA acts on the heavily sialylated DAF and propose that the NA-dependent DAF removal of sialic acids exacerbates complement activation, leading to lung immunopathology. Remarkably, this mechanism has no impact on viral loads, but rather on the host resilience to infection, and may have direct implications in zoonotic influenza transmissions.

Exacerbated complement activation and immune deregulation are at the basis of several pathologies induced by respiratory viruses. Here, we report that complement decay-accelerating factor (DAF), which inhibits complement activation in healthy cells, increases disease severity upon influenza A virus (IAV) infection. Remarkably, DAF interaction with IAV proteins, hemagglutinin (HA) and neuraminidase (NA), resulted in excessive complement activation and recruitment of innate and adaptive immune cells, without affecting viral loads. Furthermore, we observed that viral NA directly cleaves DAF and promotes complement activation, providing a possible link between IAV-DAF interaction and pathology. Therefore, our results unveil a novel pathway that could modulate disease severity, which may help to understand the increased pathogenicity of zoonotic and pandemic IAV infections.

Glycans function as a Golgi export signal to promote the constitutive exocytic trafficking

Most proteins in the secretory pathway are glycosylated. However, the role of glycans in membrane trafficking is still unclear. Here, we discovered that transmembrane secretory cargos, such as interleukin 2 receptor α subunit or Tac, transferrin receptor, and cluster of differentiation 8a, unexpectedly displayed substantial Golgi localization when their O-glycosylation was compromised. By quantitatively measuring their Golgi residence times, we found that the observed Golgi localization of O-glycan–deficient cargos is due to their slow Golgi export. Using a superresolution microscopy method that we previously developed, we revealed that O-glycan–deficient Tac chimeras localize at the interior of the trans-Golgi cisternae. O-Glycans were observed to be both necessary and sufficient for the efficient Golgi export of Tac chimeras. By sequentially introducing O-glycosylation sites to ST6GAL1, we demonstrated that O-glycan's effect on Golgi export is probably additive. Finally, the finding that N-glycosylated GFP substantially reduces the Golgi residence time of a Tac chimera suggests that N-glycans might have a similar effect. Therefore, both O- and N-glycans might function as a generic Golgi export signal at the trans-Golgi to promote the constitutive exocytic trafficking.

Plasma membrane profiling during enterohemorrhagic E. coli infection reveals that the metalloprotease StcE cleaves CD55 from host epithelial surfaces

Enterohemorrhagic Escherichia coli (EHEC) is one of several E. coli pathotypes that infect the intestinal tract and cause disease. Formation of the characteristic attaching and effacing lesion on the surface of infected cells causes significant remodeling of the host cell surface; however, limited information is available about changes at the protein level. Here we employed plasma membrane profiling, a quantitative cell-surface proteomics technique, to identify host proteins whose cell-surface levels are altered during infection. Using this method, we quantified more than 1100 proteins, 280 of which showed altered cell-surface levels after exposure to EHEC. 22 host proteins were significantly reduced on the surface of infected epithelial cells. These included both known and unknown targets of EHEC infection. The complement decay–accelerating factor cluster of differentiation 55 (CD55) exhibited the greatest reduction in cell-surface levels during infection. We showed by flow cytometry and Western blot analysis that CD55 is cleaved from the cell surface by the EHEC-specific protease StcE and found that StcE-mediated CD55 cleavage results in increased neutrophil adhesion to the apical surface of intestinal epithelial cells. This suggests that StcE alters host epithelial surfaces to depress neutrophil transepithelial migration during infection. This work is the first report of the global manipulation of the epithelial cell surface by a bacterial pathogen and illustrates the power of quantitative cell-surface proteomics in uncovering critical aspects of bacterial infection biology.

Incomplete clearance of apoptotic cells by core 1-derived O-glycan-deficient resident peritoneal macrophages

The core 1 β1,3-galactosyltransferase-specific molecular chaperon (Cosmc) is essential for the synthesis of the core 1 structure of mucin-type O-glycans. To clarify the physiological role of core 1-derived O-glycans in macrophages, we exploited the LysM-Cre transgene to generate a conditional Cosmc mutant allele (conditional Cosmc knockout; cKO) in myeloid cells. cKO mice developed normally with no gross phenotypic abnormalities or abnormal peripheral blood counts. Resident peritoneal macrophages (rpMacs) of cKO mice exhibited impaired engulfment of apoptotic cells but showed normal macrophage differentiation and counts. T-cell immunoglobulin and mucin domain-containing molecule 4 (Tim4) is a phosphatidylserine (PS) receptor expressed on rpMacs and possesses a heavily O-glycosylated domain. Tim4 tethers apoptotic cells through PS binding. Expression of the Tim4 transcript was unchanged in cKO rpMacs, whereas flow cytometric, Western and dot blot analyses revealed that Tim4 protein expression in cKO rpMacs was significantly lower than that in wild-type (WT) rpMacs. Moreover, the expression levels of other efferocytosis-related molecules, Mertk, Itgav and Itgb3, were normal in rpMacs. In addition, hypoglycosylated Tim4-FLAG fusion protein sufficiently recognized PS. These results demonstrated that core 1-derived O-glycan is required for Tim4-dependent normal efferocytosis and may contribute to the stable expression of the Tim4 glycoprotein.

SheddomeDB: the ectodomain shedding database for membrane-bound shed markers

BACKGROUND

A number of membrane-anchored proteins are known to be released from cell surface via ectodomain shedding. The cleavage and release of membrane proteins has been shown to modulate various cellular processes and disease pathologies. Numerous studies revealed that cell membrane molecules of diverse functional groups are subjected to proteolytic cleavage, and the released soluble form of proteins may modulate various signaling processes. Therefore, in addition to the secreted protein markers that undergo secretion through the secretory pathway, the shed membrane proteins may comprise an additional resource of noninvasive and accessible biomarkers. In this context, identifying the membrane-bound proteins that will be shed has become important in the discovery of clinically noninvasive biomarkers. Nevertheless, a data repository for biological and clinical researchers to review the shedding information, which is experimentally validated, for membrane-bound protein shed markers is still lacking.

RESULTS

In this study, the database SheddomeDB was developed to integrate publicly available data of the shed membrane proteins. A comprehensive literature survey was performed to collect the membrane proteins that were verified to be cleaved or released in the supernatant by immunological-based validation experiments. From 436 studies on shedding, 401 validated shed membrane proteins were included, among which 199 shed membrane proteins have not been annotated or validated yet by existing cleavage databases. SheddomeDB attempted to provide a comprehensive shedding report, including the regulation of shedding machinery and the related function or diseases involved in the shedding events. In addition, our published tool ShedP was embedded into SheddomeDB to support researchers for predicting the shedding event on unknown or unrecorded membrane proteins.

CONCLUSIONS

To the best of our knowledge, SheddomeDB is the first database for the identification of experimentally validated shed membrane proteins and currently may provide the most number of membrane proteins for reviewing the shedding information. The database included membrane-bound shed markers associated with numerous cellular processes and diseases, and some of these markers are potential novel markers because they are not annotated or validated yet in other databases. SheddomeDB may provide a useful resource for discovering membrane-bound shed markers. The interactive web of SheddomeDB is publicly available at http://bal.ym.edu.tw/SheddomeDB/ .

Polypeptide N-Acetylgalactosaminyltransferase 13 Contributes to Neurogenesis via Stabilizing the Mucin-type O-Glycoprotein Podoplanin

Mucin-type O-glycosylation is initiated by an evolutionarily conserved family of polypeptide N-acetylgalactosaminyltransferases (ppGalNAc-Ts). Previously, it was reported that ppGalNAc-T13 is restrictively expressed at a high level in the brain. Here we provide evidence for the critical role of ppGalNAc-T13 in neural differentiation. In detail, we show that the expression of ppGalNAc-T13 was dramatically up-regulated during early neurogenesis in mouse embryonic brains. Similar changes were also observed in cell models of neuronal differentiation by using either primary mouse cortical neural precursor cells or murine embryonal carcinoma P19 cells. Knockout of ppGalNAc-T13 in P19 cells suppressed not only neural induction but also neuronal differentiation. These effects are at least partly mediated by the mucin-type O-glycoprotein podoplanin (PDPN), as knockdown of PDPN led to a similar inhibition of neuronal differentiation and PDPN was significantly reduced at the posttranscriptional level after ppGalNAc-T13 knockout. Further data demonstrate that PDPN acts as a substrate of ppGalNAc-T13 and that the ppGalNAc-T13-mediated O-glycosylation on PDPN is important for its stability. Taken together, this study suggests that ppGalNAc-T13 contributes to neuronal differentiation through glycosylating and stabilizing PDPN, which provides insights into the regulatory roles of O-glycosylation in mammalian neural development.

Podoplanin requires sialylated O-glycans for stable expression on lymphatic endothelial cells and for interaction with platelets

O-glycosylation of podoplanin (PDPN) on lymphatic endothelial cells is critical for the separation of blood and lymphatic systems by interacting with platelet C-type lectin-like receptor 2 during development. However, how O-glycosylation controls endothelial PDPN function and expression remains unclear. In this study, we report that core 1 O-glycan-deficient or desialylated PDPN was highly susceptible to proteolytic degradation by various proteases, including metalloproteinases (MMP)-2/9. We found that the lymph contained activated MMP-2/9 and incubation of the lymph reduced surface levels of PDPN on core 1 O-glycan-deficient endothelial cells, but not on wild-type ECs. The lymph from mice with sepsis induced by cecal ligation and puncture, which contained bacteria-derived sialidase, reduced PDPN levels on wild-type ECs. The MMP inhibitor, GM6001, rescued these reductions. Additionally, GM6001 treatment rescued the reduction of PDPN level on lymphatic endothelial cells in mice lacking endothelial core 1 O-glycan or cecal ligation and puncture-treated mice. Furthermore, core 1 O-glycan-deficient or desialylated PDPN impaired platelet interaction under physiological flow. These data indicate that sialylated O-glycans of PDPN are essential for platelet adhesion and prevent PDPN from proteolytic degradation primarily mediated by MMPs in the lymph.

Roles for trafficking and O-linked glycosylation in the turnover of model cell surface proteins

Proteins targeted to the plasma membrane (PM) of cells are degraded at different rates. Sorting motifs contained within the cytoplasmic domains of transmembrane proteins, post-translational modifications (e.g. ubiquitination), and assembly into multiprotein or protein-lipid complexes all may affect the efficiency of endocytosis and recycling and influence the delivery to degradative compartments. Using the SNAP-tag labeling system, we examined the turnover of a model PM protein, the α chain of the interleukin-2 receptor (Tac). The surface lifetimes of SNAP-Tac fusions were influenced by their mode of entry into cells (clathrin-dependent versus clathrin-independent), their orientation in the PM (transmembrane versus glycosylphosphatidylinositol-anchored), and ubiquitination in their cytosolic domains. In addition, shedding of SNAP-Tac into the medium was greatly influenced by its O-linked glycosylation status. For a number of PM proteins, delivery to lysosomes and ectodomain shedding represent distinct parallel mechanisms to determine protein half-life.

Age- and sex-related differences in extra-hepatic low-density lipoprotein receptor

To determine whether differences in LDLr behavior in extra-hepatic tissues and whether extra-hepatic receptors could differentially contribute to cholesterol homeostasis under physiological conditions, we evaluated the presence and regulation of LDLr from both a gender and an aging perspective. We used the brain cortex, the gastrocnemius, and the heart ventricle of 3- and 12-month-old male and female rats. We observed a protein decrease of total LDLr in 12-month-old female rat brains that was completely restored by 17-β estradiol treatment. In the gastrocnemius, LDLr accumulates in the skeletal muscle in both male and female aged rats as a precursor probably due to a glycosylation impairment. In the heart, no modifications were observed in either older rats or rats of a specific gender. These data highlight a tissue-specific dysregulation of LDLr that is age- and gender-dependent.

Identification of the decay-accelerating factor CD55 as a peanut agglutinin-binding protein and its alteration in non-small cell lung cancers

PURPOSE

Peanut agglutinin (PNA) recognizes tumor-associated carbohydrates. In this study, we aimed to identify the core protein harboring PNA-binding sugars in the human lung and to explore the relationship with the pathology of primary non-small cell lung cancers (NSCLC).

EXPERIMENTAL DESIGN

PNA lectin blotting was used to detect PNA-binding proteins in the microsomal fraction of lung tissue from 24 patients with NSCLC. The 55- to 65-kDa core peptide PNA-binding protein was characterized by enzymatic treatment and identified by immunoprecipitation and affinity chromatography. The expression level and increase in size of the 55- to 65-kDa PNA-binding protein/decay-accelerating factor (DAF) were compared between normal and tumor regions of the tumor tissue by Western blotting and quantitative PCR.

RESULTS

The 55- to 65-kDa PNA-binding protein was observed in human lung. This was a glycosylphosphatidylinositol-anchored membrane protein carrying O-linked carbohydrates. This core protein was identified as DAF, one of the complementary regulatory proteins. DAF was enlarged to 65 to 75 kDa in NSCLC tumor lesions due to sialylation in the sugar moiety. At the transcription level, DAF levels were significantly lower in tumor regions, suggesting its down-regulation in NSCLC cells.

CONCLUSIONS

DAF was identified as a new PNA-binding protein in the human lung. The down-regulation and heavy sialylation of DAF was associated with pathology in NSCLC, and these alterations make this protein a potential marker for NSCLC.

Molecular cloning and characterization of novel splicing variants of human decay-accelerating factor

Decay-accelerating factor (DAF) is one of the complement regulatory proteins. Two isoforms of DAF have been identified in humans. In this study, we isolated novel cDNAs encoding five isoforms of DAF from the human lung, which were generated by insertion of new exonic sequences. RT-PCR revealed that all isoforms were expressed in almost all tissues tested, although the expression patterns and levels differed among the tissues. Transfection of isoform vDAF1, 2, and 3 cDNAs into CHO cells showed that these molecules are soluble forms secreted after glycosylation. Isoform vDAF4 and vDAF5 cDNAs included a part of and the entire intron 7 sequence, respectively, and the transfection of vDAF4 cDNA produced a large, glycosylated, membrane-bound form. These results suggest that more than seven isoforms of human DAF are involved in the regulation of complement activation under physiological conditions through their specific structures and localization.

Stable expression of chimeric anti-CD3 receptors on mammalian cells for stimulation of antitumor immunity

Expression of CD80 or CD86 costimulatory molecules on tumor cells can produce rejection of immunogenic but not poorly immunogenic tumors. We have previously shown that anti-CD3 single-chain antibodies expressed on the surface of cells can directly activate T cells. We therefore investigated whether anti-CD3 "receptors" could enhance CD86-mediated rejection of poorly immunogenic tumors. Expression of anti-CD3 receptors on cells was increased by introduction of membrane-proximal "spacer" domains containing glycosylation sites between the single-chain antibody and the transmembrane domain of the chimeric receptors. Removal of glycosylation sites in the spacer reduced surface expression due to increased shedding of chimeric receptors from the cell surface. Induction of T-cell proliferation by anti-CD3 receptors did not correlate with the expression level of chimeric protein, but rather depended on the physical properties of the spacer. Anti-CD3 receptors effectively induced T-cell cytotoxicity, whereas coexpression with CD80 or CD86 was required for generating T-cell proliferation and IL-2 secretion. Although expression of CD86 did not significantly delay the growth of poorly immunogenic B16-F1 tumors, expression of anti-CD3 receptors with CD86 produced complete tumor rejections in 50% of mice and induced significant protection against wild-type B16-F1 tumor cells. Our results show that spacer domains can dramatically influence the surface expression and the biological activity of chimeric antibody receptors. The strong antitumor activity produced by anti-CD3 receptors and CD86 on tumor cells indicates that this strategy may be beneficial for the gene-mediated therapy of poorly immunogenic tumors.

Aberrant O-glycosylation inhibits stable expression of dysadherin, a carcinoma-associated antigen, and facilitates cell-cell adhesion

Recently, we identified dysadherin, a novel carcinoma-associated glycoprotein, and showed that overexpression of dysadherin in human hepatocarcinoma PLC/PRF/5 cells could suppress E-cadherin-mediated cell-cell adhesion and promote tumor metastasis. The present study shows evidence that dysadherin is actually O-glycosylated. This was based on a direct carbohydrate composition analysis of a chimera protein of an extracellular domain of dysadherin fused to an Fc fragment of immunoglobulin. To assess the importance of O-glycosylation in dysadherin function, dysadherin-transfected hepatocarcinoma cells were cultured in a medium containing benzyl-alpha-GalNAc, a modulator of O-glycosylation. This treatment facilitated homotypic cell adhesion among dysadherin transfectants accompanied with morphological changes, indicating that the anti-adhesive effect of dysadherin was weakened. Modification of O-glycan synthesis also resulted in down-regulation of dysadherin expression and up-regulation of E-cadherin expression in dysadherin transfectants but did not affect E-cadherin expression in mock transfectants. Structural analysis of O-glycans released from the dysadherin chimera proteins indicated that a series of O-glycans with core 1 and 2 structures are attached to dysadherin, and their sialylation is remarkably inhibited by benzyl-alpha-GalNAc treatment. However, sialidase treatment of the cells did not affect calcium-dependent cell aggregation, which excluded the possibility that sialic acid itself is directly involved in cell-cell adhesion. We suggest that aberrant O-glycosylation in carcinoma cells inhibits stable expression of dysadherin and leads to the up-regulation of E-cadherin expression by an unknown mechanism, resulting in increased cell-cell adhesion. The carbohydrate-directed approach to the regulation of dysadherin expression might be a new strategy for cancer therapy.

Role of the membrane-proximal O-glycosylation site in sorting of the human receptor for neurotrophins to the apical membrane of MDCK cells

We have analyzed the respective roles of the stalk and/or the O-glycosylation sites in apical sorting by producing partially deleted mutants in this region of the human receptor for neurotrophins (P75(NTR)). The mere presence of O-glycosylations was not sufficient for efficient delivery to the apical surface since changing the stalk domain of P75(NTR) for the heavily O-glycosylated stalk from human decay-accelerating factor led to random distribution of the chimera. The presence of O-glycosylations, however, was a prerequisite for exit from the ER and protection from intracellular cleavage since a P75(NTR) containing the non O-glycosylated stalk of the human placental alkaline phosphatase was not transported to the cell surface but was cleaved and secreted from the basolateral side. Deletion of the membrane-proximal part of the stalk showed a more dramatic reversal of polarity of P75(NTR) than the deletion of the distal part. Furthermore, moving the first putative O-glycosylation site (T216) two amino acids away from the membrane resulted in a loss of apical polarity of P75(NTR), suggesting that an important clue for apical sorting resides in this part of the stalk. This loss of apical polarity paralleled a loss of association of P75(NTR) mutants with Lubrol rafts. These data indicate that the position of O-glycans in the proximal part of the stalk domain of P75(NTR) is crucial for apical sorting and may regulate association with apical rafts.

Transmodulation of cell surface regulatory molecules via ectodomain shedding

Cell responses to exogenous stimuli often result in a rapid decrease of cell surface density of a wide range of diverse regulatory proteins, receptor and adhesion molecules in particular. This decrease may occur in a ligand-dependent fashion (down-regulation), following endocytosis and degradation by lysosomal proteases, or by down-modulation, where molecules are targeted by endoproteases directly on cell surface. These proteases are recruited by trans-modulating agents, different from ligand, which act via their own receptors and the related intracellularly-generated signals. Endoproteolytic activity determines the release of large portions (shedding) of substrate proteins, called ectodomains, which are usually not ligand-bound, and therefore represent biologically-active molecules. Ectodomain shedding is involved in a number of pathophysiological processes, such as inflammation, cell degeneration and apoptosis, and oncogenesis. Common features of the process, such as the involvement of protein kinase C and of transmembrane metalloproteases, have been identified. In this review, we summarize basic concepts on down-modulation and ectodomain shedding, and provide an update of the issue with respect to: (i) new entries to the list of molecules found involved in the process; (ii) current views about the upstream control of shedding, i.e. the pathways linking the signals triggered by the trans-modulating agents to the activation of endoproteolytic activity on the cell surface.

Polymorphic expression of decay-accelerating factor in human colorectal cancer

BACKGROUND

We have previously shown that expression of decay-accelerating factor (DAF), a complement regulatory protein, is enhanced immunohistochemically on the luminal surface of cancer glands in human colorectal cancer and is detected in stool specimens of patients with colorectal cancer. The amount of DAF present in the stools might be influenced by the stability of DAF on the cell surface which is regulated by biochemical properties such as glycosylation of the protein. In the present study, to help elucidate the mechanism for the release of DAF from human colorectal cancers, we biochemically analyzed DAF expression by western and northern blotting by using surgically resected specimens of colorectal cancers.

METHODS

Surgically resected colorectal cancer tissues were obtained from 10 patients. Expression of DAF was determined by western and northern blotting, and glycosylation of DAF protein was analyzed with glycosidase digestion.

RESULTS

Northern blot analysis demonstrated that the expression of DAF mRNA in colorectal cancer was enhanced two- to threefold compared with normal tissues. In western blotting, expression of DAF protein in the cancer tissue was increased, and heterogeneity in the apparent molecular weight of DAF was observed among patients. When o-linked sugars were removed, this heterogeneity of DAF size diminished.

CONCLUSIONS

The polymorphic expression of DAF in colorectal cancer is likely to reflect variability in the o-glycosylation of the protein. We speculate that this variability could affect the stability of DAF on the surfaces of cancer cells and, in turn, the amount of DAF shed into the stools of colorectal cancer patients.

Clathrin-mediated endocytosis of MUC1 is modulated by its glycosylation state

MUC1 is a mucin-like type 1 transmembrane protein associated with the apical surface of epithelial cells. In human tumors of epithelial origin MUC1 is overexpressed in an underglycosylated form with truncated O-glycans and accumulates in intracellular compartments. To understand the basis for this altered subcellular localization, we compared the synthesis and trafficking of various glycosylated forms of MUC1 in normal (Chinese hamster ovary) cells and glycosylation-defective (ldlD) cells that lack the epimerase to make UDP-Gal/GalNAc from UDP-Glc/GlcNAc. Although the MUC1 synthesized in ldlD cells was rapidly degraded, addition of GalNAc alone to the culture media resulted in stabilization and near normal surface expression of MUC1 with truncated but sialylated O-glycans. Interestingly, the initial rate of endocytosis of this underglycosylated MUC1 was stimulated by twofold compared with fully glycosylated MUC1. However, the half-lives of the two forms were not different, indicating that trafficking to lysosomes was not affected. Both the normal and stimulated internalization of MUC1 could be blocked by hypertonic media, a hallmark of clathrin-mediated endocytosis. MUC1 endocytosis was also blocked by expression of a dominant-negative mutant of dynamin-1 (K44A), and MUC1 was observed in both clathrin-coated pits and vesicles by immunoelectron microscopy of ultrathin cryosections. Our data suggest that the subcellular redistribution of MUC1 in tumor cells could be a direct result of altered endocytic trafficking induced by its aberrant glycosylation; potential models are discussed. These results also implicate a new role for O-glycans on mucin-like membrane proteins entering the endocytic pathway through clathrin-coated pits.

Activation of lymphocytes by anti-CD3 single-chain antibody dimers expressed on the plasma membrane of tumor cells

Activation of cytotoxic T cells without MHC restriction was attempted by expressing single-chain antibodies (scFv) against CD3 on the surface of tumor cells. A chimeric protein consisting of a scFv of mAb 145.2C11, the hinge-CH2-CH3 region of human IgG1, and the transmembrane and cytosolic domains of murine CD80 formed disulfide-linked dimers on the plasma membrane of cells and specifically bound lymphocytes. Anti-CD3 scFv dimers expressed on the cell surface induced CD25 (IL-2 receptor alpha-chain) expression and proliferation of splenocytes. CT26 tumor cells engineered to express surface scFv dimers (CT26/2C11) also induced potent lymphocyte cytotoxicity with or without addition of exogenous IL-2. Splenocytes activated by CT26/2C11 cells also killed wild-type CT26 cells, indicating that activated splenocytes could kill bystander tumor cells. Immunization of BALB/c mice with irradiated CT26/2C11 cells did not protect against a lethal challenge of CT26 cells, suggesting that systemic immunity was not induced. However, the growth of CT26 tumors containing 50% CT26/2C11 cells was significantly retarded compared with CT26 tumors, whereas CT26/2C11 tumors did not grow in syngeneic mice. These results suggest that expression of anti-CD3 scFv dimers on tumors may form the basis for a novel therapeutic strategy for tumors that exhibit defects in antigen processing or presentation. Gene Therapy (2000) 7, 339-347.

O-Glycosylation of Axl2/Bud10p by Pmt4p is required for its stability, localization, and function in daughter cells

Cells of the yeast Saccharomyces cerevisiae choose bud sites in a manner that is dependent upon cell type: a and alpha cells select axial sites; a/alpha cells utilize bipolar sites. Mutants specifically defective in axial budding were isolated from an alpha strain using pseudohyphal growth as an assay. We found that a and alpha mutants defective in the previously identified PMT4 gene exhibit unipolar, rather than axial budding: mother cells choose axial bud sites, but daughter cells do not. PMT4 encodes a protein mannosyl transferase (pmt) required for O-linked glycosylation of some secretory and cell surface proteins (Immervoll, T., M. Gentzsch, and W. Tanner. 1995. Yeast. 11:1345-1351). We demonstrate that Axl2/Bud10p, which is required for the axial budding pattern, is an O-linked glycoprotein and is incompletely glycosylated, unstable, and mislocalized in cells lacking PMT4. Overexpression of AXL2 can partially restore proper bud-site selection to pmt4 mutants. These data indicate that Axl2/Bud10p is glycosylated by Pmt4p and that O-linked glycosylation increases Axl2/ Bud10p activity in daughter cells, apparently by enhancing its stability and promoting its localization to the plasma membrane.

The role of O-linked sugars in determining the very low density lipoprotein receptor stability or release from the cell

The very low density lipoprotein receptor is a member of the low density lipoprotein receptor supergene family for which two isoforms have been reported, one lacking and the other containing an O-linked sugar domain. In order to gain insight into their functionality, transient and stable transformants separately overexpressing previously cloned bovine variants were analyzed. We report evidence that the variant lacking the O-linked sugar domain presented a rapid cleavage from the cell and that a large amino-terminal very low density lipoprotein receptor fragment was released into the culture medium. As only minor proteolysis was involved in the other very low density lipoprotein receptor variant, the clustered O-linked sugar domain may be responsible for blocking the access to the protease-sensitive site(s). To test this hypothesis, a mutant Chinese hamster ovary cell line, ldlD, with a reversible defect in the protein O-glycosylation, was used. The instability of the O-linked sugar-deficient very low density lipoprotein receptor on the cell surface was comparable to that induced by the proteolysis of the variant lacking the O-linked sugar domain. Moreover, our data suggest that the O-linked sugar domain may also protect the very low density lipoprotein receptor against unspecific proteolysis. Taken together, these results indicate that the presence of the O-linked sugar domain may be required for the stable expression of the very low density lipoprotein receptor on the cell surface and its absence may be required for release of the receptor to the extracellular space. The exclusive expression of the variant lacking the O-linked sugar domain in the bovine aortic endothelium opens new perspectives in the physiological significance of the very low density lipoprotein receptor.

Putative O-glycosylation sites and a membrane anchor are necessary for apical delivery of the human neurotrophin receptor in Caco-2 cells

We have expressed the human neurotrophin receptor p75 (p75(NTR)) in the intestinal epithelial cell line Caco-2 as a model to study intracellular transport and subcellular sorting signals in intestinal cells. p75(NTR) was localized at the apical membrane of Caco-2 cells and reached this membrane mainly via an indirect pathway. Apical localization, intracellular routing, and basolateral to apical transcytosis were not affected by truncation of the cytoplasmic domain or replacement of the transmembrane domain by a glycosyl phosphatidylinositol anchor. Removal of membrane anchoring resulted in basolateral secretion of the ectodomain of p75(NTR) in Caco-2 cells but in apical secretion in Madin-Darby canine kidney (MDCK) cells. Substitution of potential O-glycosylation sites present in the stalk of p75(NTR) led to intracellular cleavage and secretion of the ectodomain into the basolateral medium both in Caco-2 and MDCK cells. These results suggest that the stalk of p75(NTR) carries an apical sorting information that is recognized efficiently by Caco-2 cells only when attached to the membrane. This apical sorting information is linked to the presence of predicted O-glycosylation sites in that region. These putative O-glycosylation sites also play a role in the regulation of p75(NTR) transport to the cell surface and in the prevention of rapid degradation by cleavage of the stalk domain.

Site-specific core 1 O-glycosylation pattern of the porcine submaxillary gland mucin tandem repeat. Evidence for the modulation of glycan length by peptide sequence

The sequence-specific O-linked core 1 ([R1, R2]-beta-Gal(1-3)-alpha-GalNAc-O-Ser/Thr) glycosylation pattern has been quantitatively determined for 30 of the 31 Ser/Thr residues in the 81-residue porcine submaxillary gland mucin tandem repeat. This was achieved by Edman amino acid sequencing of the isolated tandem repeat after selective removal of non-C3-substituted, peptide-linked GalNAc residues by periodate oxidation and subsequent trimming of the remaining oligosaccharides to peptide-linked GalNAc residues by mild trifluoromethanesulfonic acid/anisole treatment. The sequencing reveals 61% (range, 12-95%) of the peptide alpha-N-acetylgalactosamine (GalNAc) residues to be substituted by core 1 chains, a value in agreement with the carbon-13 NMR analysis of the native mucin. Residues with the lowest C3 substitution were typically clustered in regions of sequence with the highest densities of (glycosylated) serine or threonine. This suggests that the porcine beta3-Gal, core 1, transferase is sensitive to peptide sequence and/or neighboring core GalNAc glycosylation in vivo, in keeping with earlier in vitro enzymatic glycosylation studies (Granovsky, M., Blielfeldt, T., Peters, S., Paulsen, H., Meldal, M., Brockhausen, J., and Brockhausen, I. (1994) Eur. J. Biochem. 221, 1039-1046). These results demonstrate that the O-glycan structures in mucin domains are not necessarily uniformly distributed along the polypeptide core and that their lengths can be modulated by peptide sequence. The data further suggest that hydroxyamino acid spacing may contribute to the regulation of glycan length, thereby, providing a mechanism for maintaining an optimally expanded, protease resistant, mucin conformation.

Defectively N-glycosylated and non-O-glycosylated aminopeptidase N (CD13) is normally expressed at the cell surface and has full enzymatic activity

In order to study the effects of the absence of O-glycosylation and modifications of N-glycosylation on a class II membrane protein, pig and human aminopeptidase N (CD13) were stably expressed in the ldl(D) cell line. This cell line carries a UDP-Gal/UDP-GalNAc-epimerase deficiency which blocks the conversion of glucose into galactose derivatives. Thus it is possible in the ldl(D) cell line to selectively block O-glycosylation by the omission of N-acetylgalactoseamine from the culture medium and to alter N-glycosylation by the omission of galactose. In this way selectively altered glycosylated forms of the glycoprotein aminopeptidase N can be synthesized and the effects of altered glycosylation can be studied. It is demonstrated that aminopeptidase N carries "mucin-type" O-glycans and that this is predominantly located in the stalk, which connects the catalytic headgroup to the membrane anchor. Normally glycosylated aminopeptidase N is present in the plasma membrane of the ldl(D) cells. This is also the case for the non-O-glycosylated and defectively N-glycosylated forms. This is in line with the finding that the intracellular transport APN is unaffected by the absence of O-glycosylation or by changes in N-glycosylation as the various glycosylated forms of aminopeptidase N are normally converted from the high-mannose form to the complex glycosylated form. Enzymatic activity is not influenced by the changes in glycosylation.

Kidney N-acetylgalactosamine (GalNAc)-1-phosphate kinase, a new pathway of GalNAc activation

A new enzyme that phosphorylates GalNAc at position 1 to form GalNAc-alpha-1P was purified approximately 1275-fold from the cytosolic fraction of pig kidney, and the properties of the enzyme were determined. The kinase is quite specific for GalNAc as the phosphate acceptor and is inactive with GlcNAc, ManNAc, glucose, galactose, mannose, GalN, and GlcN. This enzyme is clearly separated from galactokinase by chromatography on phenyl-Sepharose. The GalNAc kinase has a pH optimum between 8.5 and 9.0 and requires a divalent cation in the order Mg2+ > Mn2+ > Co2+, with optimum Mg2+ concentration at approximately 5 mM. The enzyme was most active with ATP as the phosphate donor, but slight activity was observed with ITP, acetyl-P, and phosphoenolpyruvate. Enzyme activity was highest in porcine and human kidney and porcine liver, but was low in most other tissues. Cultured HT-29 cells also had high activity for this kinase. The purified enzyme fraction was incubated with azido-[32P]ATP, exposed to UV light, and run on SDS gels. A 50-kDa protein was labeled, and this labeling showed saturation kinetics with increasing amounts of the probe and was inhibited by unlabeled ATP. Although the most purified GalNAc kinase preparation still had two bands that labeled with ATP, maximum labeling of the 50-kDa protein, but not the 66-kDa band, was coincident with maximum GalNAc kinase activity on a column of DEAE-Cibacron blue. On Sephacryl S-300, the native enzyme has a molecular mass of 48-51 kDa, indicating that the active kinase is a monomer. The product of the reaction was characterized as GalNAc-alpha-1-P by various chemical procedures.

Structural requirements for the ectodomain cleavage of human cell surface macrophage colony-stimulating factor

One form of human macrophage colony-stimulating factor (CSF-1(256), M-CSFalpha) is a member of a restricted set of cell surface transmembrane proteins, which is selected to undergo proteolytic ectodomain cleavage. To determine the substrate requirements for this cleavage, we have constructed a series of mutations in the cytoplasmic tail, transmembrane domain, and juxtamembrane region of CSF-1(256) and stably expressed the mutated genes in NIH 3T3 cells. Our results demonstrate that membrane association of the CSF-1 precursor is required for cleavage of its growth factor ectodomain and furthermore that the juxtamembrane region Pro161-Gln162-Leu163-Gln164-Glu165 (PQLQE) (residues 161-165 of the ectodomain) is an essential determinant of cell surface CSF-1(256) cleavage and that the cleavage site is partially sequence-specific. Furthermore, a mechanism of steric hindrance, which likely involves interference with protease accessibility, is postulated to explain the observed decreases in the cleavage efficiency in certain CSF-1 mutants. Finally, our results strongly suggest that the CSF-1 ectodomain is cleaved at or very near the cell surface by a membrane-associated proteolytic system.

Control of the complement system

The complement system has developed a remarkably simple but elegant manner of regulating itself. It has faced and successfully dealt with how to facilitate activation on a microbe while preventing the same on host tissue. It solved this problem primarily by creating a series of secreted and membrane-regulatory proteins that prevent two highly undesirable events: activation in the fluid phase (no target) and on host tissue (inappropriate target). Also, if not checked, even on an appropriate target, the system would go to exhaustion and have nothing left for the next microbe. Therefore, the complement enzymes have an intrinsic instability and the fluid-phase control proteins play a major role in limiting activation in time. The symmetry of the regulatory process between fluid phase and membrane inhibitors at the C4/C3 step of amplification and convertase formation as well as at the MAC steps are particularly striking features of the self/nonself discrimination system. The use of glycolipid anchored proteins on membranes to decay enzymes and block membrane insertion events is unlikely to be by chance. Finally, it is economical for the cofactor regulatory activity to produce derivatives of C3b that now specifically engage additional receptors. Likewise, C1-Inh leads to C1q remaining on the immune complex to interact with the C1q receptor. Thus the complement system is designed to allow rapid, efficient, unimpeded activation on an appropriate foreign target while regulatory proteins intervene to prevent three undesirable consequences of complement activation: excessive activation on a single target, fluid phase activation, and activation on self.

Structure of Saccharomyces cerevisiae alpha-agglutinin. Evidence for a yeast cell wall protein with multiple immunoglobulin-like domains with atypical disulfides

alpha-Agglutinin of Saccharomyces cerevisiae is a cell wall-associated protein that mediates cell interaction in mating. Although the mature protein includes about 610 residues, the NH2-terminal half of the protein is sufficient for binding to its ligand a-agglutinin. alpha-Agglutinin20-351, a fully active fragment of the protein, has been purified and analyzed. Circular dichroism spectroscopy, together with sequence alignments, suggest that alpha-agglutinin20-351 consists of three immunoglobulin variable-like domains: domain I, residues 20-104; domain II, residues 105-199; and domain III, residues 200-326. Peptide sequencing data established the arrangement of the disulfide bonds in alpha-agglutinin20-351. Cys97 is disulfide-bonded to Cys114, forming an interdomain bond between domains I and II. Cys202 is bonded to Cys300, in an atypical intradomain disulfide bond between the A and F strands of domain III. Cys227 and Cys256 have free sulfhydryls. Sequencing also showed that at least two of three potential N-glycosylation sites with sequence Asn-Xaa-Thr are glycosylated. At least one of three Asn-Xaa-Ser sequences is not glycosylated. No residues NH2-terminal to Ser282 were O-glycosylated, whereas Ser282, and all hydroxy amino acid residues COOH-terminal to this position were modified. Therefore O-glycosylated Ser and Thr residues cluster in the COOH-terminal region of domain III, and the O-glycosylation continues into a Ser/Thr-rich sequence that extends from domain III to the COOH-terminal of the full-length protein.

Alloantigenic recognition of artificial glycosyl phosphatidylinositol-anchored HLA-A2.1

Alloantigen presentation by GPI-reanchored variants of the human class I MHC molecule HLA-A2.1 was studied in human cellular systems. To this end, we generated chimeric coding sequences for two GPI-modified HLA-A2.1 heavy chain derivatives. In these chimeras, the coding sequence for the HLA-A2.1 heavy chain was fused in-frame to alternative overlapping sequences from the 3'-end of human DAF containing the GPI-modification signal sequence. The encoded polypeptides HLA-A2.1:DAF-S and HLA-A2.1:DAF-L differed by 53 amino acids of additional DAF sequence in the latter. Both were detected on stably transfected C1R cell surfaces by HLA-A2.1-specific mAb, and their GPI-modification was confirmed by PI-PLC enzymatic cleavage. Immunoprecipitation analysis of surface-biotinylated C1R transfectants revealed heterodimeric association for both HLA-A2.1:DAF-L and HLA-A2.1:DAF-S heavy chains with beta 2m. Alloantigenic stimulation by, and cytotoxic recognition of, both HLA-A2.1:DAF-S/C1R and HLA-A2.1/CIR cells was observed; however, HLA-A2.1:DAF-L/C1R cells could not serve as allostimulators or allotargets. These findings establish that polymorphic human class I MHC molecules can function, when artificially GPI-reanchored, as alloantigenic targets. Moreover, the data suggest that the sequence bridging the HLA-A2 extracellular domain and the membrane can influence alloantigenic presentation.

Differences between the sugar moieties of liver- and bone-type alkaline phosphatases: a re-evaluation

We re-evaluated the differences between the sugar moieties of liver and bone alkaline phosphatases (ALPs). Sialic acid was added to ALP sugar moieties by alpha 2,3- or 2,6-sialyltransferase treatment of the asialo-form ALP (neuraminidase-treated ALP). Asialo-bone ALP was converted to a liver-like ALP by the 2,6-sialyltransferase treatment. The resulting liver-like ALP was less susceptible to neuraminidase than non-treated bone ALP, but was still labile to heat exposure at 56 degrees C like non-treated bone ALP. However, after the O-linked sugar moiety had been released by additional treatment with O-glycanase the liver-like ALP became more heat stable at 56 degrees C, like non-treated liver ALP. Non-treated liver ALP reacted specifically with anti-liver ALP monoclonal antibody, and non-treated bone ALP reacted with both anti-liver and anti-bone ALP antibodies. The asialo-bone ALP still reacted with anti-bone ALP antibody, whereas the asialo-form liver ALP showed little, if any, reaction with anti-liver and anti-bone ALP antibodies. Neuraminidase and O-glycanase-treated bone ALP reacted less with anti-bone ALP antibody. After O-glycanase treatment, bone ALP molecules deprived of an O-linked sugar moiety had a molecular size and heat stability similar to liver ALP. The difference between liver and bone ALP molecules may be due not only to their manner of sialic acid linkage but also to the attachment of the O-linked sugar moiety.

Clinical aspects of glycoprotein biosynthesis

Glycoproteins are widely distributed among species in soluble and membrane-bound forms, associated with many different functions. The heterogenous sugar moieties of glycoproteins are assembled in the endoplasmic reticulum and in the Golgi and are implicated in many roles that require further elucidation. Glycoprotein-bound oligosaccharides show significant changes in their structures and relative occurrences during growth, development, and differentiation. Diverse alterations of these carbohydrate chains occur in diseases such as cancer, metastasis, leukemia, inflammatory, and other diseases. Structural alterations may correlate with activities of glycosyltransferases that assemble glycans, but often the biochemical origin of these changes remains unclear. This suggests a multitude of biosynthetic control mechanisms that are functional in vivo but have not yet been unraveled by in vitro studies. The multitude of carbohydrate alterations observed in disease states may not be the primary cause but may reflect the growth and biochemical activity of the affected cell. However, knowledge of the control mechanisms in the biosynthesis of glycoprotein glycans may be helpful in understanding, diagnosing, and treating disease.

Biosynthesis and glycosylation of serine/threonine-rich secreted proteins from Toxocara canis larvae

Toxocara canis infective stage larvae continually produce excretory-secretory (TES) glycoproteins in long-term in vitro culture. The kinetics of synthesis and secretion were studied by metabolic labelling with radioactive [35S]methionine, [14C]serine and [14C]threonine. Maximal incorporation rates required overnight pre-incubation of parasites in medium depleted of the appropriate amino acid. Larvae rapidly incorporated isotope into their somatic tissues, but there was a minimum delay of 10 h before secretion of labelled antigens. Labelling with [14C]serine and [14C]threonine demonstrated a relative abundance of these amino acids in the major surface/secreted glycoproteins of this nematode (TES-32 and 120). Pulse-chase experiments suggested that TES-120 may be derived from a 58 kDa precursor, reflecting extensive posttranslational glycosylation. Inhibition of N-glycosylation with tunicamycin and digestion with N-glycanase provided evidence of N-glycosylation in the lower molecular weight ES components (TES-32, 55 and 70). These agents had no effect on the higher molecular weight components (TES-120 and 400) implying that for these molecules glycosylation is predominantly O-linked. The largest ES component (TES-400) was unusual, in incorporating serine and threonine but not methionine, and by exhibiting increased apparent molecular weight following pronase digestion; it is suggested that this molecule is a proteoglycan.

Molecular cloning of a novel member of the immunoglobulin gene superfamily homologous to the polymeric immunoglobulin receptor

The CMRF35 monoclonal antibody recognizes a cell membrane antigen present on the surface of monocytes, neutrophils, a proportion of peripheral blood T and B lymphocytes and lymphocytic cell lines. Initial studies with CMRF35 suggested an unusual pattern of serological reactivity which did not correspond to any of the known leukocyte differentiation antigen clusters. We describe here the cloning and sequencing of a cDNA encoding the CMRF35 antigen by means of expression in COS cells and immunoselection with the CMRF35 monoclonal antibody. The cDNA encodes a novel integral membrane glycoprotein of 224 amino acids that represents a new member of the immunoglobulin (Ig) gene superfamily. The molecule comprises (a) a single extracellular Ig variable domain remarkably similar to the Fc receptor for polymeric IgA and IgM, (b) a membrane-proximal domain containing a high proportion of proline, serine and threonine residues that was predicted to be heavily O-glycosylated, (c) an unusual transmembrane anchor that contained a glutamic acid and a proline residue and (d) a short cytoplasmic tail. Transcripts encoding the CMRF35 protein were detected in early monocytic cell lines, in peripheral blood T cells and in some B lymphoblastoid cell lines, confirming the results of immunocytological staining. However, the level of CMRF35 expression on peripheral blood T cells was shown to decrease in response to mitogenic stimulation. The likelihood that the CMRF35 antigen shares a common evolutionary ancestor with the poly Ig Fc receptor is discussed.

The E3-14.5K integral membrane protein of adenovirus that is required for down-regulation of the EGF receptor and for prevention of TNF cytolysis is O-glycosylated but not N-glycosylated

The adenovirus E3-14.5K protein is a cytoplasmic integral membrane protein that functions in concert with the E3-10.4K protein to down-regulate the epidermal growth factor receptor and to prevent tumor necrosis factor cytolysis in adenovirus-infected cells. The 14.5K protein migrates as multiple bands in SDS-PAGE, indicating that it undergoes post-translational modification. The 14.5K protein is known to be phosphorylated on serine. We show here that 14.5K can be metabolically labeled with [3H]glucosamine, that the label is labile to alkali, and that the SDS-PAGE band pattern is simplified in a cell line that is defective in O-glycosylation. Thus, 14.5K is O-glycosylated, probably at a single site in the NH2-terminal lumenal domain. The protein was not metabolically labeled with [3H]mannose, and its SDS-PAGE band pattern was not affected by tunicamycin treatment in vivo or endo F treatment in vitro; thus, 14.5K is not N-glycosylated. There was no evidence that the 10.4K protein is glycosylated, and the 10.4K protein was not required for glycosylation of 14.5K. Virtually all 14.5K molecules appear to contain the core disaccharide Gal beta 1-3GalNAc alpha 1-Ser/Thr which is commonly found on mucin-type O-glycoproteins, and neuraminidase digestion experiments indicated that this disaccharide contains terminal sialic acid.

Membrane cofactor protein

MCP serves to down-regulate the activation of complement on host tissue. It performs this function by serving as a cofactor for the factor I-mediated cleavage of C3b and C4b. MCP is most likely an intrinsic regulator, i.e., it primarily protects its home cell. The wide tissue distribution of MCP mirrors this critical function of host cell protection. With the exception of erythrocytes, every cell and tissue examined expresses this protein. MCP is represented as two broad heterogeneous bands on SDS-PAGE with M(r)s of 51,000-58,000 and 59,000-68,000. The quantity of each form expressed is inherited in an autosomal codominant fashion. In most cells and cell lines, four isoforms of MCP predominate and arise by alternative splicing of a single MCP gene. All forms possess four repeating modules of--60 aminoacids, an area enriched in serines, threonines, and prolines [(STP), probable site of O-linked glycosylation], a short area of unknown function, a transmembrane domain, and a cytoplasmic tail. The isoforms differ, however, in the length and composition of the STP region and in the cytoplasmic tail. Alternative splicing of a single exon within the STP region determines the protein phenotype. Alternative splicing at the COOH_terminus gives rise to two distinct cytoplasmic tails. The biological significance of these structural variations in the STP and cytoplasmic tail regions is being investigated.