Notch deficiency implicated in the pathogenesis of congenital disorder of glycosylation IIc

  1. Hiroyuki O. Ishikawa*,
  2. Shunsuke Higashi†,
  3. Tomonori Ayukawa†,
  4. Takeshi Sasamura†,‡,
  5. Motoo Kitagawa§,
  6. Kenichi Harigaya§,
  7. Kazuhisa Aoki,
  8. Nobuhiro Ishida,
  9. Yutaka Sanai, and
  10. Kenji Matsuno*,†,‡,∥
  1. *Genome and Drug Research Center and †Department of Biological Science and Technology, Tokyo University of Science, Noda, Chiba 278-8510, Japan; ‡Precursory Research for Embryonic Science and Technology, Japan Science and Technology Corporation, Kawaguchi-shi, Saitama 332-0012, Japan; §Department of Molecular and Tumor Pathology, Chiba University Graduate School of Medicine, Chuo-ku, Chiba 260-8670, Japan; and Department of Biochemical Cell Research, The Tokyo Metropolitan Institute of Medical Science, Bunkyo-ku, Tokyo 113-8613, Japan

  1. Edited by Senitiroh Hakomori, Pacific Northwest Research Institute, Seattle, WA (received for review May 18, 2005)

 
Next Section

Abstract

Congenital disorder of glycosylation IIc (CDG IIc), also termed leukocyte adhesion deficiency II, is a recessive syndrome characterized by slowed growth, mental retardation, and severe immunodeficiency. Recently, the gene responsible for CDG IIc was found to encode a GDP-fucose transporter. Here, we investigated the possible cause of the developmental defects in CDG IIc patients by using a Drosophila model. Biochemically, we demonstrated that a Drosophila homolog of the GDP-fucose transporter, the Golgi GDP-fucose transporter (Gfr), specifically transports GDP-fucose in vitro. To understand the function of the Gfr gene, we generated null mutants of Gfr in Drosophila. The phenotypes of the Drosophila Gfr mutants were rescued by the human GDP-fucose transporter transgene. Our phenotype analyses revealed that Notch (N) signaling was deficient in these Gfr mutants. GDP-fucose is known to be essential for the fucosylation of N-linked glycans and for O-fucosylation, and both fucose modifications are present on N. Our results suggest that Gfr is involved in the fucosylation of N-linked glycans on N and its O-fucosylation, as well as those of bulk proteins. However, despite the essential role of N O-fucosylation during development, the Gfr homozygote was viable. Thus, our results also indicate that the Drosophila genome encodes at least another GDP-fucose transporter that is involved in the O-fucosylation of N. Finally, we found that mammalian Gfr is required for N signaling in mammalian cultured cells. Therefore, our results implicate reduced N signaling in the pathology of CDG IIc.

Congenital disorders of glycosylation (CDGs) are autosomal recessive disorders that are commonly associated with severe psychomotor and mental retardation (1). CDGs result from defects in N-linked oligosaccharides, and at least 13 CDGs have been reported (1). CDG IIc [OMIM (for On-line Mendelian Inheritance in Men) database accession no. 266265], also termed leukocyte adhesion deficiency type II, is a recessive syndrome characterized by growth and mental retardation and severe immunodeficiency with marked neutrophilia (2, 3). CDG IIc patients lack the carbohydrate epitopes Lewisx and sialyl-Lewisx. Neutrophils from these patients cannot roll on Selectins, the receptors for Lewisx, activated endothelial cells in vitro, or inflamed mesenteric microvessels in vivo, leading to impaired migration to sites of inflammation (4). Recently, the gene whose defect is responsible for CDG IIc was cloned by the complementation of cells derived from CDG IIc patients and was found to encode a GDP-fucose transporter (5, 6). The GDP-fucose transporter is a nucleotide sugar transporter, classified as belonging to solute carrier family 35 (7). The transporter is predicted to span the Golgi membrane 10 times, and it couples the import of GDP-fucose into the Golgi lumen with the export of guanosine triphosphate into the cytoplasm; in the Golgi, GDP-fucose is used by specific fucosyltransferases to add fucose to a variety of glycoproteins and glycolipids. Because fucosyltransferases use GDP-fucose, which is synthesized in the cytosol, as a fucose donor, the uptake of GDP-fucose into the Golgi is thought to be a critical step for the fucosylation event.

In CDG IIc patients, the oral administration of fucose is an effective therapy for their immunodeficiency (8, 9). In response to this treatment, the missing Selectin ligands on neutrophils are restored, and the high peripheral neutrophil counts are reduced to normal levels (8, 9). However, the mental retardation of CDG IIc patients cannot be cured through this therapy (8). This outcome suggested that a shortage of GDP-fucose in the Golgi lumen results in developmental defects responsible for at least some of the pathogenesis of CDG IIc. Recently, we and other groups showed that the O-fucosylation of Notch (N) catalyzed by protein O-fucosyltransferase 1 was essential for N signaling in Drosophila (10, 11) and mouse (12). The fucosylation of N-linked oligosaccharides is also found on Notch1 (13). Therefore, it is conceivable that a disruption of N signaling contributes to the developmental defects observed in CDG IIc patients. Recently, it has been broadly recognized that Drosophila can be an excellent model to study the functions of genes involved in human diseases (14). In this regard, in diseases associated with glycosylation disorders, many glycan structures, including O-fucose, are conserved between human and Drosophila (15). Moreover, the Drosophila genome encodes a set of enzymes involved in the metabolism of fucosylated glycan, including the GDP-fucose biosynthesis pathway enzymes, the GDP-fucose transporter, fucosidase, and fucosyltransferases (16). Therefore, Drosophila seems appropriate for studying the pathogenesis of CDG IIc and especially the possible involvement of N signaling. Here, we used newly generated Drosophila GDP-fucose mutant alleles to obtain evidence that a reduced uptake of GDP-fucose into the Golgi affects the activation of the N signaling.

Previous SectionNext Section

Materials and Methods

Fly Strains and cDNA Clones. The P-element insertion line EY01553 was from the Bloomington Drosophila Stock Center (Indiana University, Bloomington, IN). To excise the P-element, the P[Δ2–3] fly strain was used as a transposase source. Cyan fluorescent protein-endoplasmic reticulum (CFP-ER) (BD Biosciences, San Jose, CA), a GFP variant with an ER-retention (KDEL) signal, was used as an ER marker. Actin 5C-galactosidase-4 (Act5C-Gal4), hs-Gal4, and patched (ptc)-Gal4 flies were used as Gal4 drivers. All flies were raised at 18°C or 25°C. We used Canton-S as the wild type. We purchased cDNA clones RE40567 (Open Biosystems, Huntsville, AL) and PLACE1010321 (Helix Research Institute, Kisarazu, Japan), which encoded the Drosophila and human GDP-fucose transporter, respectively. Upstream activating sequence (UAS)-Gfr and UAS-human GDP-fucose transporter (HsGfr) contain the entire coding region of the Gfr and HsGfr cDNA, respectively. UAS-Gfr-HA and UAS-HsGfr-HA encoded the hemagglutinin (HA) sequence tag at the 3′ end of the Gfr and HsGfr cDNA, respectively. UAS-GfrR125C-HA and UAS-HsGfrR147C-HA carry the cDNAs corresponding to introduced amino acid substitutions that change R to C at amino acids 125 and 147, respectively.

Nucleotide Sugar Transport Assay. The Gfr-HA cDNA was subcloned into the copper-inducible expression vector pYEX-BX and transformed into Saccharomyces cerevisiae strain YPH500. The in vitro nucleotide sugar transport assay was performed essentially as described previously (17). Membrane fractions obtained by centrifugation at 100,000 × g for 1 h were used in the in vitro transport assay. The reaction was started by adding the membrane preparation (50 μg of protein) to the reaction mixture [0.8 M sorbitol/10 mM Tris·HCl (pH 7.0)/1 mM MgCl2/0.5 mM dimercaptopropanol, containing 1 mM 3H-labeled substrate (6,400 Ci/mol; 1 Ci = 37 GBq)] and incubating the mixture at 30°C for 1 min. To determine the rate of GDP-[3H]mannose transport, 10 μg of microsomal protein and GDP-mannose (3,200 Ci/mol) were used. The mixture was incubated at 30°C, diluted with 1 ml of ice-cold stop buffer [0.8 M sorbitol/10 mM Tris·HCl (pH 7.0)/1 mM MgCl2/1 mM nonradioactive nucleotide sugar] to stop the reaction, and poured onto a nitrocellulose filter (0.45 μm; Millipore, Bedford, MA). The radioactivity trapped on the filter was measured in a toluene-based scintillator.

Tissue Staining and in Situ Hybridization. Immunostaining and in situ hybridization were performed according to standard protocols. For the immunostaining, mouse anti-Wingless (Wg) (4D4, 1:250; Developmental Studies Hybridoma Bank, University of Iowa, Iowa City), mouse anti-120-kDa integral Golgi-membrane protein (7H6D7C2, 1:500; EMD Biosciences, San Diego, CA), rat anti-HA (3F10, 1:1,000; Roche Diagnostics), and rabbit anti-GFP (1:1,000; MBL, Nagoya, Japan) were used as the primary antibodies. Alexa Fluor 488-conjugated goat anti-rabbit IgG (1:500; Invitrogen, Carlsbad, CA), cyanine 3-conjugated goat anti-mouse IgG (1:500; Rockland, Gilbertsville, PA), and cyanine 5-conjugated goat anti-rat IgG (1:500; Rockland) were used as the secondary antibodies. Images were obtained with a confocal microscope (PASCAL; Zeiss, Oberkochen, Germany). For the in situ hybridization, antisense or sense RNA probes were prepared from the Gfr cDNA template, including the entire coding region by T3 and T7 RNA polymerase, respectively (Stratagene, La Jolla, CA). For lectin staining, FITC-conjugated Aleuria Aurantia lectin (AAL) (5 μg/ml; Seikagaku Corp., Tokyo) was used.

Western and Lectin Blotting Analyses. For Western blotting, a mouse anti-N antibody (C17.9C6; 1:1,000), a horseradish peroxidase-conjugated sheep anti-mouse Ig (1:5,000; Amersham Biosciences, Piscataway, NJ), and enhanced chemiluminescence reagents (Amersham Biosciences) were used. For lectin blotting, biotin-conjugated AAL (5 μg/ml; Seikagaku Corp.), Alexa Fluor 555-conjugated streptavidin (5 μg/ml; Invitrogen), and a laser scanner (Typhoon; Amersham Biosciences) were used.

Gfr Knockdown and Measurement of N Signaling Activation in Mammalian Cultured Cells. To knockdown the mouse Gfr gene (GenBank database accession no. AK030977), oligodeoxynucleotides that express short-hairpin (sh) RNAs were designed (5′-GCATGATAACCTTCAATAA-3′ as shGfr#1 and 5′-CAGATTAAGTGGAGTGATG-3′ as shGfr#5; B-Bridge, Sunnyvale, CA) and cloned into the pSUPER retro.puro vector (OligoEngine, Seattle). As a control, we used a vector that expresses shGFP (18). Retroviruses were produced and used to infect cells from the C2C12 murine myoblast cell line as described (19). Infected cells were selected with puromycin. To assess N signaling activity in the C2C12 derivatives, the cells were transfected with the pTP1-luc reporter (20) and a pRL-cytomegalovirus internal control vector (Promega) by using Lipofectamine (Invitrogen). Twenty-four hours later, we launched the coculture of the transfected cells with Nalm6 human B-cell acute lymphoblastic leukemia cells overexpressing Jagged1 or its vector-infected control lines (M.K., unpublished data). Forty-eight hours after the transfection, the luciferase assay was performed by using a dual luciferase assay kit (Promega) (21).

Previous SectionNext Section

Results

Gfr, an Ortholog of the Human GDP-Fucose Transporter, Transports GDP-Fucose. The Drosophila CG9620 gene was identified as a putative ortholog of the human GDP-fucose transporter, and here we designated it as Gfr, for Golgi GDP-fucose transporter (5, 6, 22). The predicted Gfr protein has 337 aa with 10 putative transmembrane domains (22). The amino acid sequences of the human and Drosophila proteins show 47% identity and 65% similarity. Recently, CG9620 was demonstrated to restore fucosylation in cells from a patient with CDG IIc, indicating that the CG9620 product serves as a functional GDP-fucose transporter (22). However, the specificity of the GDP-fucose transportation through Gfr remained to be determined. Therefore, we first tested the activity of Gfr in transporting various nucleotide sugars, using a yeast in vitro system. Microsomes were prepared from transformants carrying vectors with and without the Gfr cDNA insert, and their ability to transport nucleotide sugars was investigated (Fig. 1A). The expression of Gfr-HA, a wild-type Gfr tagged with a HA polypeptide at the C terminus, was confirmed by Western blotting with an anti-HA monoclonal antibody (data not shown). Fig. 1 A shows that the Gfr transport of GDP-fucose was more than twice as efficient as the GDP-fucose transport in the control, and it did not transport CMP-sialic acid, UDP-galactose, UDP-N-acetylgalactosamine, UDP-glucuronic acid, UDP-N-acetylglucosamine, UDP-glucose, or GDP-mannose. The time course of GDP-fucose uptake by membrane vesicles with or without Gfr expression was also studied (Fig. 1B). In membrane vesicles from Gfr-expressing cells, uptake of the nucleotide sugar initially proceeded rapidly for about 30 s and continued slowly for the next 30 s. In contrast, the uptake of GDP-fucose by the membrane vesicles from control cells exhibited only the slow phase of uptake for the entire 1 min. These data further confirmed that Gfr was involved in the uptake of GDP-fucose by membrane vesicles. The apparent K m value of Gfr could not be estimated because the vesicles seemed to have an endogenous background activity (data not shown).

Fig. 1.
View larger version:
Fig. 1.

The Drosophila CG9620 gene encodes a Golgi GDP-fucose transporter. (A) The transport activity of CMP-sialic acid (CMP-Sia), GDP-fucose (GDP-Fuc), UDP-galactose (UDP-Gal), UDP-N-acetylgalactosamine (UDP-GalNAc), UDP-glucuronic acid (UDP-GlcA), UDP-N-acetylglucosamine (UDP-GlcNAc), UDP-glucose (UDP-Glc), and GDP-mannose (GDP-Man) into S. cerevisiae vesicles transfected with Gfr (solid bars) or mock transfected (open bars). Values are the mean ± SEM from duplicate experiments. (B) Time course of GDP-fucose uptake in vitro. Membrane vesicles from Gfr-expressing yeast cells (solid squares) or cells not expressing Gfr (open circles) were incubated with GDP-[3H]fucose for the indicated time periods. Values are the mean ± SEM from duplicate experiments. (C–F) Subcellular localization of Gfr. UAS-Gfr-HA and UAS-CFP-ER were driven by hs-Gal4, and, after a 30 min heat shock at 37°C, the wing disc of the third-instar larva was isolated and triple-stained with anti-HA (blue in C), anti-Golgi 120-kDa protein (red in D), and anti-GFP (green in E). (G–J) In situ hybridization of Drosophila embryos using an antisense (G and I) or sense (H and J) probe for Gfr RNA at stage 3 (G and H) and stage 11 (I and J).


The human GDP-fucose transporter is reportedly localized to the Golgi (5). Therefore, we next investigated the subcellular localization of Gfr in vivo in wing disc cells, using Gfr-HA. Gfr-HA colocalized with a Golgi marker, the Golgi 120-kDa protein, but not with an ER marker CFP-ER (Fig. 1 C–F). This result is consistent with a previous observation that the CG9620 products localize to the Golgi in cultured mammalian cells (22). Our in situ hybridization analysis revealed that the expression of Gfr is ubiquitous at stages 3 and 11 (Fig. 1 G–J). Strong expression of Gfr was detected from stages 1 to 17 (data not shown).

Gfr-Null Mutants Exhibit Wing Phenotypes That Indicate a Partial Reduction of N Signaling. We and other groups demonstrated that the O-fucosylation of epidermal growth factor-like repeats in the extracellular domain of N is essential for the binding between N and its ligands (10, 11, 23). Therefore, N signaling could be disturbed in Gfr mutants, because Gfr is apparently involved in the transportation of GDP-fucose into the Golgi, which is essential for the catalytic action of O-fucosyltransferase. To address this issue directly, we generated a mutation of Gfr in Drosophila. We used an approach involving imprecise excision of a P-element. Derivative strains that lost a P-element were established from EY01553 in which the P-element was inserted 5′ of the Gfr gene. Genomic lesions were determined by PCR, and we obtained two deletion lines, designated Gfr 1 and Gfr 2. Genomic sequence analysis revealed that both Gfr 1 and Gfr 2 lacked most of the Gfr gene locus, including a putative initiation codon (Fig. 2A). Therefore, these alleles are likely to be null mutations. These Gfr alleles were homozygous viable but exhibited a small terminal delta in veins III and IV of the adult wing at 25°C (Fig. 2C, and data not shown). Given that O-fut1, which encodes the protein O-fucosyltransferase 1, is an essential gene, these results suggest that the Drosophila genome encodes redundant GDP-fucose transporter activities (11, 23).

Fig. 2.
View larger version:
Fig. 2.

Gfr mutations showed Notch-like phenotypes. (A) Genomic organization of the Gfr locus. The transcription unit of Gfr is shown as boxes, and the predicted coding region is shaded black. The regions 3′ to the P-element EY01553 insertion site are deleted in Gfr 1 (≈0.8 kb) and Gfr 2 (≈1.1 kb). (B and C) Adult wing phenotypes at 25°C. Veins III (L3) and IV (L4) show a small terminal delta in homozygous Gfr 1 (C). (D–G) Adult wings of the following genotypes obtained at 18°C: (D) wild type; (E) EY01553/EY01553;(F) Gfr 1/Gfr 1 (G) Gfr 1/Exel6147. Note that the wing phenotypes of Gfr at 18°C are more severe than at 25°C. (H and J) Phenotypes of the Gfr 1 mutant were rescued by transgenes of Gfr-HA (H) and human GDP-fucose transporter-HA (HsGfr-HA; J) cDNAs, but GfrR125C-HA (I) and HsGfrR147C-HA (K) cDNAs, which have a point mutation found in CDG IIc patients, did not rescue them. These results are also described in detail in Table 1.


We found that these Gfr mutations were temperature sensitive. Homozygotes of Gfr showed a wing-notch phenotype at 18°C, which is also observed in the wing of N mutants (Fig. 2F) (24). We speculated that the redundant activity of the GDP-fucose transporter becomes a limiting factor for N signaling at 18°C. Because of the higher sensitivity, we used flies cultured at 18°C unless otherwise noted. This phenotype was not observed in homozygotes of EY01553 or its revertants at 18 or 25°C (Fig. 2E, and data not shown). In addition, the same phenotype was observed in transheterozygotes of Gfr 1 and Gfr 2 (data not shown), Gfr 1 and Df(3R)Exel6147 (Fig. 2G), a deficiency that uncovered the Gfr locus, and Gfr 2 and Df(3R)Exel6147 (data not shown). We found that the phenotypes observed in Gfr 1/Gfr 1, Gfr 2/Gfr 2, Gfr 1/Gfr 2, Gfr 1/Df(3R)Exel6147, and Gfr 2/Df(3R)Exel6147 were essentially the same (Fig. 2 F and G, and data not shown), suggesting that Gfr 1 and Gfr 2 are null alleles, which is consistent with the nature of their genomic lesions.

Gfr 1 mutant phenotypes were rescued by the ubiquitous expression of wild-type and HA-tagged versions of Gfr and HsGfr, driven by Act5C-Gal4 (Fig. 2 H and J; Table 1). Thus, the Gfr gene is an ortholog of HsGfr, as suggested previously, and is responsible for the phenotypes of the Gfr 1 mutant (Fig. 2 C and F). We also found that mutant Gfr and HsGfr, GfrR125C and HsGfrR147C, which have an amino acid substitution that is found in CDG IIc patients, failed to rescue Gfr 1 under the same conditions, suggesting that CDG IIc is associated with a loss-of-function mutation of HsGfr (Fig. 2 I and K; Table 1) (5, 6).

View this table:
Table 1. Phenotypes of Gfr homozygotes are rescued by transgenes of Gfr and its human ortholog

Reduction of N Signaling in the Gfr Mutant Occurs Downstream of Ligand Presentation but Upstream of Receptor Activation. We and Okajima et al. (25) have generated mutants of the GDP-mannose 4,6-dehydratase (Gmd) gene, which encodes an essential enzyme for GDP-fucose biosynthesis in Drosophila (T.S., H.O.I., K. Noda, E. Miyoshi, N. Taniguchi, and K.M., unpublished data). If Gfr plays a crucial role in the transportation of GDP-fucose into the Golgi lumen, it is conceivable that a Gmd mutant might enhance the phenotypes of Gfr. Indeed, the wing of Gmd H78/+; Gfr 1/Gfr 1 showed a significantly more severe wing phenotype than Gfr 1/Gfr 1 (compare Figs. 2F and 3D), although Gmd H78/+ showed a wild-type wing (data not shown). Consistent with this, the expression of Wg at the dorsoventral compartment boundary of the wing disc, which depends on the activation of N signaling (26), was reduced significantly in the wing disc of Gfr 1/Gfr 1 and Gmd H78/+; Gfr 1/Gfr 1 (Fig. 3, compare A–C). These results provide evidence suggesting that Gfr acts as a transporter for GDP-fucose in vivo and is partially required for N signaling, at least in some developmental contexts. We also found that Gfr 1 genetically interacted with mutant forms of the genes involved in N signaling. For example, Delta (Dl), which encodes a ligand for N, and Hairless, a negative regulator of N signaling, were a dominant enhancer and suppressor of Gfr, respectively (data not shown). Together, our results further suggest that N signaling deteriorates in the Gfr mutants. Moreover, we performed epistatic analyses to identify the step of N signaling at which Gfr is required. Ectopic expression of Dl along the anteroposterior compartment boundary induces Wg expression ectopically (Fig. 3E). This effect of Dl was suppressed in the Gfr mutant (Fig. 3F). In contrast, ectopic Wg expression induced by the activated form of N under the same conditions was not suppressed by the Gfr mutation (Fig. 3, compare G and H). These results indicate that Gfr acts downstream of ligand presentation but upstream of receptor activation in N signaling.

Fig. 3.
View larger version:
Fig. 3.

Gfr is involved in N signaling. (A) Wg expression was detected in the dorsoventral compartment boundary of the wild-type wing disc. In the Gfr 1 homozygote wing imaginal disk, the Wg expression was reduced (B), and wing nicking was observed in the adult wing (see Fig. 2F). (C and D) In double mutants heterozygous for Gmd H78 and homozygous for Gfr 1, the Wg expression was largely absent (C), and the adult wing phenotypes of Gfr 1 were enhanced (D). (E and F) Ectopic expression of Dl along the anteroposterior boundary driven by ptc-Gal4 resulted in the ectopic induction of Wg (E). This effect was suppressed in the Gfr 1 background (F). (G and H) Expression of a constitutively active form of N (Nact) driven by ptc-Gal4 induced Wg expression ectopically (G). This effect was not altered by the Gfr 1 background (H).


Gfr Is Partially Required for the O-Fucosylation of N. The shortage of GDP-fucose in the Golgi of Gfr mutants seemed to affect N signaling. However, this reduction of N signaling could be attributed to a defect in O-fucosylation and/or fucosylation of the N-glycans of N. Therefore, we attempted to determine which fucose modification of N is affected in the Gfr mutants. It is thought that the O-fucosylation of N is abolished in the O-fut1 mutant (11, 23). Fringe (Fng) encodes another glycosyltransferase, β1,3 N-acetylglucosaminyltransferase, which adds Glc-NAc specifically to the O-linked fucose of N and modulates the binding between N and its ligands (27, 28). Fng functions exclusively in N signaling (27, 28). The ectopic expression of Fng under the control of the ptc-Gal4 driver resulted in the ectopic induction of Wg as well as suppression of the endogenous Wg expression at the dorsoventral compartment boundary of the wing disc (compare Figs. 3A and 4A) (27). This ectopic Wg expression resulted in the formation of ectopic wing margin bristles (Fig. 4B) (27). However, these effects of Fng were suppressed in the Gfr mutant at 18°C (Fig. 4 C and D). Conversely, the wing phenotypes associated with the ectopic expression of Fng were essentially the same in wild-type flies and the Gfr 1 homozygote at 25°C (data not shown). These results indicate that Fng function is reduced in the Gfr mutants, suggesting that Gfr is partly required for the O-fucosylation of N at 18°C.

Fig. 4.
View larger version:
Fig. 4.

Loss of Gfr function affects the O-fucosylation of N. (A–D) The ectopic expression of Fng driven by ptc-Gal4 induced the ectopic expression of Wg through N activation (A) and caused the formation of ectopic margin bristles in the adult wing (arrowhead in B). (C and D) However, the wing of UAS-fng22A/ptc-Gal4; Gfr 1/Gfr 1 did not show the ectopic Wg expression (C)orthe extra bristles (D).


Fucosylation of N-Linked Glycans Is Significantly Reduced in the Gfr Mutant. AAL is defined as an l-fucose lectin that recognizes α1,3- and α1,6-linked fucose residues (22, 29). In fibroblasts from a CDG IIc patient, the staining of AAL was strongly reduced compared with control cells (5, 6). Therefore, we compared AAL staining between wild-type flies and the Gfr mutant. In wild-type wing discs, AAL staining was detected ubiquitously (Fig. 5A). In contrast, AAL staining was strongly reduced in the wing disc of the Gfr 1 homozygote at 18°C and 25°C (Fig. 5B, and data not shown). Furthermore, the reduction of AAL staining in the Gfr mutant was restored by the ectopic overexpression of Gfr-HA or HsGfr-HA driven by ptc-Gal4 but not by their CDG IIc mutant forms (Fig. 5 C and D, and data not shown). Next, we tested whether the fucose modification of N was reduced in the Gfr mutants. The total protein extracts of imaginal discs and the central nervous system from wild-type and Gfr 1 mutant larvae were subjected to SDS/PAGE and stained with AAL. Lectin blotting revealed that the reactivity to AAL was reduced in the lysate of the Gfr mutant, indicating that the fucose modification of N-linked glycans added to bulk proteins was virtually abolished at 25°C (Fig. 5E).

Fig. 5.
View larger version:
Fig. 5.

The Gfr mutation affects the fucosylation of N-linked glycans. (A–D) AAL staining (green) of the wing imaginal discs of wild-type (A), Gfr 1/Gfr 1 (B), ptc-Gal4/UAS-Gfr-HA; Gfr 1/Gfr 1 (C), and ptc-Gal4/UAS-GfrR125C-HA; Gfr 1/Gfr 1 (D). Gfr-HA expression was shown by anti-HA (magenta). (E) Lectin blotting with AAL. Equal amounts of wild-type (wt) and homozygous Gfr 1 lysates from third-instar larvae were separated by 7.5% SDS/PAGE. (F) The fucosylation of N was decreased in the Gfr mutant. (Upper) Lysates prepared from third-instar wild-type (wt) and Gfr 1 larvae were immunoprecipitated (IP) with a monoclonal anti-N antibody (C17.9C6) and blotted with C17.9C6 (anti-N) or AAL. (Lower) Total lysates from the wild-type and Gfr larvae were blotted with C17.9C6 (anti-N) and anti-tubulin, as a loading control.


The N protein was immunoprecipitated from the lysates of wild-type larvae or larvae homozygous for Gfr 1, and the fucosylated form of N was detected by AAL. We found that the fucose modification of N was significantly lower at 25°C (Fig. 5F). Together, these results suggest that the fucosylation of N-linked glycans on N and its O-fucosylation were diminished in the Gfr mutants.

Mammalian Gfr Is Required for N Signaling in Mammalian Cells. To investigate whether Gfr function is required for N signaling in mammals as well, we tried using RNA interference to knockdown the Gfr in C2C12 murine myoblast cells that express Notch1 (20). Retrovirus vector-mediated expression of two kinds of shRNAs against mouse Gfr (shGfr#1 and shGfr#5) resulted in a reduction of AAL staining compared with cells expressing the control shGFP (Fig. 6, compare A and B, and data not shown). Concomitant with the staining, N signaling activation induced by Jagged1 (a ligand of N) was significantly reduced in cells expressing either shGfr compared with the control cells (Fig. 6C). Similar results were obtained in experiments involving Delta1 (another ligand of N) (data not shown). These results indicate that mammalian Gfr is required for N signaling in mammals.

Fig. 6.
View larger version:
Fig. 6.

Knockdown of mammalian Gfr affects N signaling in mammalian cells. (A and B) AAL staining of C2C12 cells expressing shGFP (A) or shGfr#1 (B). (Scale bar, 20 μm.) (C) N signaling activity in shGFP-, shGfr#1-, and shGfr#5-expressing cells as measured by luciferase assay. The C2C12-derived cells were cocultured with x-ray-inactivated cells expressing Jagged1 or control cells (control). Values are the relative luciferase activities normalized to the mean activity of the shGFP-expressing cells. The error bars indicate the SDs (n = 3).


Previous SectionNext Section

Discussion

The GDP-fucose transporter is a multitransmembrane protein that transports GDP-fucose into the Golgi lumen from the cytoplasm, in which it is synthesized (5, 6). Its activity is absent in CDG IIc individuals (5, 6). In this report, we showed that N signaling was reduced in flies with mutant Gfr, a Drosophila ortholog of the human GDP-fucose transporter gene. Furthermore, mammalian Gfr was also required for ligand-dependent activation of N signaling in mammalian cultured cells. Therefore, our finding implies that a defect in N signaling is responsible for the pathogenesis of CDG IIc, at least in part. Interestingly, diseases caused by defects in N signaling components, such as Alagille syndrome (OMIM database accession no. 118450) and CADASIL (for cerebral autosomal-dominant arteriopathy w ith subcortical infarcts and leukoencephalopathy) (OMIM database accession no. 125310), are occasionally associated with mental retardation. Alagille syndrome is an autosomal-dominant disorder characterized by growth and mental retardation (30, 31). Positional cloning studies revealed that Alagille syndrome is caused by mutations in the Jagged1 gene, the human homolog of Drosophila Serrate, one of the N ligands (32, 33). CADASIL is an autosomal-dominant vascular disorder associated with migraine with aura, mood disorders, recurrent subcortical ischemic strokes, progressive cognitive decline, dementia, and premature death (34, 35). CADASIL is caused by mutations in the Notch3 gene (36). Therefore, our results imply that a reduction of N signaling may be responsible for the mental retardation observed in CDG IIc patients. Moreover, in view of the fact that N signaling regulates lymphocyte development and function in mammals, it is highly probable that a deficiency of N signaling is responsible for the immunodeficiency associated with CDG IIc (37).

GDP-fucose is essential for the terminal fucosylation of N-linked glycans and O-fucose, and both fucose modifications are reported to occur on mammalian Notch1; these modifications are thought to be conserved in Drosophila (13). In CDG IIc fibroblasts, the bulk addition of fucose as a terminal modification of N-linked glycans is severely diminished, whereas bulk protein O-fucosylation is not affected (13). However, we demonstrated here that the reduction of N O-fucosylation accounts for at least a portion of the defects seen in the Gfr mutants, although it is difficult to evaluate the contribution of N N-glycan fucosylation to these defects at this time. Our results suggest that, in the Gfr 1 mutant, the fucosylation of N-glycans drastically declined but the effect on O-fucosylation was subtle. These distinct sensitivities to the reduction of GDP-fucose may be accounted for by previously published results that the fucosylation of N-glycans requires a higher concentration of GDP-fucose than O-fucosylation (13). These results also suggest that at least one other GDP-fucose transporter, which is sufficient for N O-fucosylation, is present in Drosophila. It was reported that the majority of GDP-fucose transport activity is found in the Golgi rather than in the ER, in which the O-fucosylation of N occurs (38). However, given that a low concentration of GDP-fucose is sufficient for O-fucosylation (13), a putative GDP-fucose transporter that has low GDP-fucose transport activity might be present in the ER (39). We speculate that the cold-sensitive activity of this GDP-fucose transporter is rate limiting in the Gfr 1 mutant. In this mutant, the wing nicking occurred in different regions of the wing than in the N heterozygote. The wing nick in the Gfr 1 mutant was often in the anterior and posterior regions of the wing margin, whereas nicking occurs mostly in the wing tip in the N mutant. These observations may suggest that another GDP-fucose transporter is more active in the center of the wing pouch. Notably, the symptoms of CDG IIc are partially suppressed by the oral administration of fucose, which is made into GDP-fucose through the salvage pathway, indicating the presence of a GDP-fucose transporter activity in these patients (8, 9). Therefore, a GDP-fucose transporter, which is required for O-fucosylation, may be conserved between Drosophila and mammals.

We found that AAL staining was not enhanced by the overexpression of Gfr in the wild-type wing imaginal disc and that Wg expression was not altered in these discs, suggesting that the overexpression of Gfr did not affect N signaling (data not shown). We also found that Gfr overexpression driven by various Gal4 drivers did not cause any detectable adult phenotype (data not shown). This result raises the possibility that gene therapy involving the introduction of Gfr transgenes into somatic cells might be used to cure CDG IIc, because the likelihood of side effects is probably low.

Previous SectionNext Section

Acknowledgments

We thank the Bloomington Stock Center (Indiana University, Bloomington) and the Developmental Studies Hybridoma Bank (University of Iowa, Iowa City) for flies and antibodies. The UAS-CFP-ER and UAS-fng22A strains, C2C12 cells, and pTP1-luc were kind gifts from A. Satoh, K. Irvine, M. Noda, and L. Strobl, respectively. We also thank T. Kuba, M. Kawakita, M. Inaki, Y. Murakami, T. Honjo, N. Tanaka, S. Goto, N. Taniguchi, and Y. Nagai for their technical advice and critical comments. This work was supported by grants-in-aid from the Japanese Ministry of Education, Culture, Sports, and Science (to K.M.) and grants from Precursory Research for Embryonic Science and Technology, Japan Science and Technology Corporation (to K.M.).

Previous SectionNext Section

Footnotes

  • ↵ ∥ To whom correspondence should be addressed at: Department of Biological Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan. E-mail: matsuno{at}rs.noda.tus.ac.jp.

  • Author contributions: H.O.I., T.S., M.K., K.H., K.A., N.I., Y.S., and K.M. designed research; H.O.I., S.H., T.A., M.K., and K.A. performed research; H.O.I., S.H., T.A., T.S., M.K., K.H., K.A., N.I., Y.S., and K.M. analyzed data; and H.O.I., M.K., and K.M. wrote the paper.

  • Conflict of interest statement: No conflicts declared.

  • This paper was submitted directly (Track II) to the PNAS office.

  • Abbreviations: AAL, Aleuria Aurantia lectin; Act5C, actin 5C; CDG, congenital disorder of glycosylation; CFP, cyan fluorescent protein; Dl, Delta; ER, endoplasmic reticulum; Fng, Fringe; Gal4, galactosidase-4; Gfr, Golgi GDP-fucose transporter; Gmd, GDP-mannose 4,6-dehydratase; HA, hemagglutinin; HsGfr, human GDP-fucose transporter; N, Notch; ptc, patched; sh, short-hairpin; UAS, upstream activating sequence; Wg, Wingless.

Previous Section
 

References

  1. ↵
    Freeze, H. H. (2002) Biochim. Biophys. Acta 1573 , 388-393.pmid:12417423
    Medline
  2. ↵
    Etzioni, A., Frydman, M., Pollack, S., Avidor, I., Phillips, M. L., Paulson, J. C. & Gershoni-Baruch, R. (1992) N. Engl. J. Med. 327 , 1789-1792.pmid:1279426
    MedlineISI
  3. ↵
    Hirschberg, C. B. (2001) J. Clin. Invest. 108 , 3-6.pmid:11435449
    CrossRefMedlineISI
  4. ↵
    Becker, D. J. & Lowe, J. B. (1999) Biochim. Biophys. Acta 1455 , 193-204.pmid:10571012
    Medline
  5. ↵
    Luhn, K., Wild, M. K., Eckhardt, M., Gerardy-Schahn, R. & Vestweber, D. (2001) Nat. Genet. 28 , 69-72.pmid:11326279
    CrossRefMedlineISI
  6. ↵
    Lubke, T., Marquardt, T., Etzioni, A., Hartmann, E., von Figura, K. & Korner, C. (2001) Nat. Genet. 28 , 73-76.pmid:11326280
    CrossRefMedlineISI
  7. ↵
    Ishida, N. & Kawakita, M. (2004) Pflügers Arch. 447 , 768-775.
    CrossRefMedlineISI
  8. ↵
    Marquardt, T., Luhn, K., Srikrishna, G., Freeze, H. H., Harms, E. & Vestweber, D. (1999) Blood 94 , 3976-3985.pmid:10590041
    Abstract/FREE Full Text
  9. ↵
    Hidalgo, A., Ma, S., Peired, A. J., Weiss, L. A., Cunningham-Rundles, C. & Frenette, P. S. (2003) Blood 101 , 1705-1712.pmid:12406889
    Abstract/FREE Full Text
  10. ↵
    Okajima, T. & Irvine, K. D. (2002) Cell 111 , 893-904.pmid:12526814
    CrossRefMedlineISI
  11. ↵
    Sasamura, T., Sasaki, N., Miyashita, F., Nakao, S., Ishikawa, H. O., Ito, M., Kitagawa, M., Harigaya, K., Spana, E., Bilder, D., et al. (2003) Development (Cambridge, U.K.) 130 , 4785-4795.
    Abstract/FREE Full Text
  12. ↵
    Shi, S. & Stanley, P. (2003) Proc. Natl. Acad. Sci. USA 100 , 5234-5239.pmid:12697902
    Abstract/FREE Full Text
  13. ↵
    Sturla, L., Rampal, R., Haltiwanger, R. S., Fruscione, F., Etzioni, A. & Tonetti, M. (2003) J. Biol. Chem. 278 , 26727-26733.pmid:12738772
    Abstract/FREE Full Text
  14. ↵
    Bier, E. (2005) Nat. Rev. Genet. 6 , 9-23.pmid:15630418
    MedlineISI
  15. ↵
    Haltiwanger, R. S. & Lowe, J. B. (2004) Annu. Rev. Biochem. 73 , 491-537.pmid:15189151
    CrossRefMedlineISI
  16. ↵
    Roos, C., Kolmer, M., Mattila, P. & Renkonen, R. (2002) J. Biol. Chem. 277 , 3168-3175.pmid:11698403
    Abstract/FREE Full Text
  17. ↵
    Aoki, K., Ishida, N. & Kawakita, M. (2003) J. Biol. Chem. 278 , 22887-22893.pmid:12682060
    Abstract/FREE Full Text
  18. ↵
    Ui-Tei, K., Naito, Y., Takahashi, F., Haraguchi, T., Ohki-Hamazaki, H., Juni, A., Ueda, R. & Saigo, K. (2004) Nucleic Acids Res. 32 , 936-948.pmid:14769950
    Abstract/FREE Full Text
  19. ↵
    Fujita, Y., Kitagawa, M., Nakamura, S., Azuma, K., Ishii, G., Higashi, M., Kishi, H., Hiwasa, T., Koda, K., Nakajima, N. & Harigaya, K. (2002) FEBS Lett. 528 , 101-108.pmid:12297287
    CrossRefMedlineISI
  20. ↵
    Minoguchi, S., Taniguchi, Y., Kato, H., Okazaki, T., Strobl, L. J., Zimber-Strobl, U., Bornkamm, G. W. & Honjo, T. (1997) Mol. Cell. Biol. 17 , 2679-2687.pmid:9111338
    Abstract
  21. ↵
    Kuroda, K., Tani, S., Tamura, K., Minoguchi, S., Kurooka, H. & Honjo, T. (1999) J. Biol. Chem. 274 , 7238-7244.pmid:10066785
    Abstract/FREE Full Text
  22. ↵
    Luhn, K., Laskowska, A., Pielage, J., Klambt, C., Ipe, U., Vestweber, D. & Wild, M. K. (2004) Exp. Cell Res. 301 , 242-250.pmid:15530860
    CrossRefMedlineISI
  23. ↵
    Okajima, T., Xu, A. & Irvine, K. D. (2003) J. Biol. Chem. 278 , 42340-42345.pmid:12909620
    Abstract/FREE Full Text
  24. ↵
    de Celis, J. F. & Garcia-Bellido, A. (1994) Mech. Dev. 46 , 109-122.pmid:7918096
    CrossRefMedlineISI
  25. ↵
    Okajima, T., Xu, A., Lei, L. & Irvine, K. D. (2005) Science 307 , 1599-1603.pmid:15692013
    Abstract/FREE Full Text
  26. ↵
    Rulifson, E. J. & Blair, S. S. (1995) Development (Cambridge, U.K.) 121 , 2813-2824.
    Abstract
  27. ↵
    Moloney, D. J., Panin, V. M., Johnston, S. H., Chen, J., Shao, L., Wilson, R., Wang, Y., Stanley, P., Irvine, K. D., Haltiwanger, R. S. & Vogt, T. F. (2000) Nature 406 , 369-375.pmid:10935626
    CrossRefMedline
  28. ↵
    Bruckner, K., Perez, L., Clausen, H. & Cohen, S. (2000) Nature 406 , 411-415.pmid:10935637
    CrossRefMedline
  29. ↵
    Kochibe, N. & Furukawa, K. (1980) Biochemistry 19 , 2841-2846.pmid:7397108
    CrossRefMedline
  30. ↵
    Alagille, D., Odievre, M., Gautier, M. & Dommergues, J. P. (1975) J. Pediatr. 86 , 63-71.pmid:803282
    CrossRefMedlineISI
  31. ↵
    Krantz, I. D., Piccoli, D. A. & Spinner, N. B. (1997) J. Med. Genet. 34 , 152-157.pmid:9039994
    Abstract
  32. ↵
    Oda, T., Elkahloun, A. G., Pike, B. L., Okajima, K., Krantz, I. D., Genin, A., Piccoli, D. A., Meltzer, P. S., Spinner, N. B., Collins, F. S. & Chandrasekharappa, S. C. (1997) Nat. Genet. 16 , 235-242.pmid:9207787
    CrossRefMedlineISI
  33. ↵
    Li, L., Krantz, I. D., Deng, Y., Genin, A., Banta, A. B., Collins, C. C., Qi, M., Trask, B. J., Kuo, W. L., Cochran, J., et al. (1997) Nat. Genet. 16 , 243-251.pmid:9207788
    CrossRefMedlineISI
  34. ↵
    Sourander, P. & Walinder, J. (1977) Acta Neuropathol. 39 , 247-254.pmid:906807
    CrossRefMedline
  35. ↵
    Chabriat, H. & Bousser, M. G. (2003) Adv. Neurol. 92 , 147-150.pmid:12760177
    Medline
  36. ↵
    Joutel, A., Corpechot, C., Ducros, A., Vahedi, K., Chabriat, H., Mouton, P., Alamowitch, S., Domenga, V., Cecillion, M., Marechal, E., et al. (1996) Nature 383 , 707-710.pmid:8878478
    CrossRefMedline
  37. ↵
    Radtke, F., Wilson, A., Mancini, S. J. & MacDonald, H. R. (2004) Nat. Immunol. 5 , 247-253.pmid:14985712
    CrossRefMedlineISI
  38. ↵
    Puglielli, L. & Hirschberg, C. B. (1999) J. Biol. Chem. 274 , 35596-35600.pmid:10585436
    Abstract/FREE Full Text
  39. ↵
    Luo, Y. & Haltiwanger, R. S. (2005) J. Biol. Chem. 280 , 11289-11294.pmid:15653671
    Abstract/FREE Full Text
« Previous | Next Article »Table of Contents

This Article

  1. Published online before print December 12, 2005, doi: 10.1073/pnas.0504115102 PNAS December 20, 2005 vol. 102 no. 51 18532-18537
  1. AbstractFree
  2. Figures Only
  3. » Full Text
  4. Full Text (PDF)

Classifications