Skip Navigation


Human Molecular Genetics Advance Access originally published online on April 27, 2005
Human Molecular Genetics 2005 14(12):1631-1639; doi:10.1093/hmg/ddi171
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
14/12/1631    most recent
ddi171v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Arboleda-Velasquez, J. F.
Right arrow Articles by Kosik, K. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arboleda-Velasquez, J. F.
Right arrow Articles by Kosik, K. S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2005. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oupjournals.org

CADASIL mutations impair Notch3 glycosylation by Fringe

Joseph F. Arboleda-Velasquez1,{dagger}, Raajit Rampal3,{dagger}, Erik Fung2, Diane C. Darland4, Min Liu5, Maria C. Martinez1, Christine P. Donahue1, Manuel F. Navarro-Gonzalez1, Peter Libby2, Patricia A. D'Amore4, Masanori Aikawa2, Robert S. Haltiwanger3 and Kenneth S. Kosik1,*

1Neurology Department and 2Leducq Center for Cardiovascular Research, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA 02115, USA, 3Department of Biochemistry and Cell Biology, Institute for Cell and Developmental Biology, State University of New York at Stony Brook, Stony Brook, NY 11794, USA, 4Department of Ophthalmology and Department of Pathology, Schepens Eye Research Institute, Harvard Medical School, Boston, MA 02114, USA and 5Laboratory for Drug Discovery in Neurodegeneration, Brigham and Women's Hospital and Harvard Medical School, Cambridge, MA, USA

* To whom correspondence should be addressed at current address: Neuroscience Research Institute, University of California Santa Barbara, Santa Barbara, CA 93106-5060, USA. Email: kosik{at}lifesci.ucsb.edu

Received February 19, 2005; Revised April 13, 2005; Accepted April 22, 2005


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Mutations in the NOTCH3 gene trigger adult-onset stroke and vascular dementia in patients with CADASIL (cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy). All CADASIL mutations described to date affect the epidermal growth factor-like (EGF-like) repeats located in the extracellular domain of the Notch3 receptor. These domains are also the target of sequential complex O-linked glycosylation mediated by protein O-fucosyltransferase 1 and Fringe. We investigated whether O-fucosylation or Fringe-mediated elongation of O-fucose on Notch3 is impaired by CADASIL mutations. Biochemical studies of a Notch3 fragment containing the first five EGF-like repeats of Notch3, including the mutational hot spot, showed that CADASIL mutations do not affect the addition of O-fucose but do impair carbohydrate chain elongation by Fringe. CADASIL changes also induced aberrant homodimerization of mutant Notch3 fragments and heterodimerization of mutant Notch3 with Lunatic Fringe itself. Together, these data suggest that Fringe plays a role in CADASIL pathophysiology.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is an autosomal dominant disorder caused by mutations in NOTCH3 (1Go). Degeneration of brain vessels result in recurrent ischemic events that often lead to adult-onset vascular dementia in patients with this disorder (1Go–3Go). The disease is characterized by the loss of vascular smooth muscle cells (SMC) and the extracellular accumulation of the Notch3 ectodomain and granular osmiophilic material in the vessels (4Go,5Go). However, the molecular mechanisms by which Notch3 mutations lead to vascular degeneration and stroke are unknown.

Notch3 belongs to a family of highly conserved single pass transmembrane receptors that regulate cell fate decisions during development of metazoans (6Go). Before reaching the plasma membrane, the immature mammalian Notch proteins are cleaved at an S1 site in the extracellular domain by furin-like convertase to form a mature heterodimeric receptor containing a 110–120 kDa fragment (NTM) with the transmembrane and intracellular segments, and an extracellular fragment (NEC) containing epidermal growth factor-like (EGF-like) repeats and Notch/Lin-12 repeats (7Go,8Go). Upon binding to Delta or Jagged/Serrate ligands expressed in adjacent cells, Notch undergoes a series of additional cleavage events that result in the translocation of the intracellular domain to the nucleus, where it regulates transcription of genes through interaction with the DNA binding protein C-promoter binding factor-1 (CBF-1) (9Go).

Most CADASIL mutations are missense changes that result in the addition or removal of cysteines from the EGF-like repeats of Notch3 (10Go). However, mutations resulting in small in-frame deletions within these modules have been reported (11Go,12Go). Changes in the number or the spacing of six highly conserved cysteine residues in an EGF-like repeat may interfere with the formation of disulfide bonds and disturb the tridimensional structure of this module (11Go). However, it is still unclear how such changes affect Notch function at the molecular level. Recent work indicates that rare CADASIL mutations located in the ligand-binding domain (EGF-like repeats 10 and 11) result in very early onset of stroke (13Go) and affect interaction of Notch3 receptor with ligands (14Go,15Go). However, most CADASIL mutations affect EGF-like repeats (1Go to 5Go) located away from the ligand-binding domain (10Go) and do not interfere with Notch3 signaling (14Go–17Go). In fact, mutations at such locations have been reported to impair S1 cleavage and delay membrane localization of Notch3 receptor (15Go,17Go). In contrast to other CADASIL mutations, the C542Y change located in EGF-like 13 almost completely abrogated Notch3 localization to the plasma membrane, thus affecting signaling (14Go). It is unclear whether Notch3 mutations lead to CADASIL through different molecular mechanisms or whether a common consequence of such changes is yet to be described.

In this work, we explored the effect of CADASIL mutations on post-translational O-linked glycosylation of Notch3. This particular type of modification has been shown to modulate Notch preference for its ligands Delta and Jagged/Serrate (18Go–21Go). EGF-like repeats located in the extracellular domain of Notch receptors undergo modification with O-fucose on serine or threonine residues in a process mediated by protein O-fucosyltransferase 1 (O-FucT-1) (22Go,23Go). Previous experiments indicate that O-FucT-1 glycosylates only properly folded EGF-like repeats and suggest a role for this enzyme in quality control (24Go–26Go). Several, but not all, of the fucosylated EGF-like repeats can then be modified by Fringe, which adds N-acetylglucosamine to the O-fucose to form a disaccharide (27Go–29Go). Other enzymes add galactose and finally sialic acid to form trisaccharide and tetrasaccharide structures (27Go). Differential localization of Fringe, of which mammals express three homologs named Manic, Radical and Lunatic, is thought to be the rate limiting step in the sequential glycosylation process (30Go). The glycosylation of many EGF-like repeats targeted by CADASIL mutations, which also contain the motif for O-linked glycosylation, may be impaired (Fig. 1A).



View larger version (26K):
[in this window]
[in a new window]
  		Figure 1. Notch3 EGF-like repeats are modified with O-fucose. (A) The localization of CADASIL mutations in relation to predicted O-linked fucosylation sites. Boxes indicate EGF-like repeats 1–34 of human Notch3. Gray color indicates the presence of a predicted O-linked fucosylation site in specific EGF-like repeats. The sites were identified using the motif C2X4–5(S/T)C3 [where X4–5 are any 4–5 amino acid residues, C2 and C3 are the second and third conserved cysteines of the EGF repeat and (S/T) is the O-fucosylation site] (29Go). The height of each box stands for the number of CADASIL mutations reported to occur within the EGF-like repeat. The location of some prevalent CADASIL mutations located in the vicinity of predicted glycosylation sites within the hot spot is indicated (R90C, R169C, C212S). All the predicted glycosylation sites in human Notch3 are present in the mouse homolog. (B) Wild-type or mutant Notch3 fragments were purified from culture media of stably transfected Lec1-CHO cells and analyzed by SDS–PAGE. The R91C, R170C and C213S mutations in mouse Notch3 correspond to the R90C, R169C and C212S CADASIL changes in human Notch3. Protein levels and incorporation of radiolabeled fucose were determined by western blot using anti-Myc antibody (top) and fluorography (bottom). The graph below the films indicates the ratio of protein amount over fucose incorporation as determined by densitometric scanning of films.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
A Notch3 fragment containing the CADASIL hot spot is modified by O-linked fucose
Glycosylation of Notch3 was studied directly by generating Lec1-Chinese hamster ovary (CHO) cell lines stably transfected with plasmids encoding the first five EGF-like repeats of mouse Notch3 and Lunatic Fringe. To analyze whether CADASIL mutations altered O-fucosylation, Notch3 fragments bearing the R91C, R170C or C213S CADASIL-like mutations were purified from media and analyzed by western blot (for protein levels) and fluorography (for O-fucose levels). In humans, EGF-like repeats 1–5 are encoded by exons 3 and 4 of the NOTCH3 gene where >60% of the reported CADASIL mutations cluster (10Go,31Go). Both the R170C and the C213S mutations are near predicted O-fucosylation sites (T174 and T212) and EGF-like repeats 4 and 5 of Notch3 are mutational hot spots (Fig. 1A).

Wild-type and CADASIL mutant constructs incorporated similar amounts of radiolabeled fucose on a per protein basis (Fig. 1B). Thus, CADASIL mutations located within the mutational hot spot of Notch3 do not alter the initial O-fucosylation reaction. Previous reports showed that O-FucT-1 preferentially glycosylates properly folded EGF-like repeats (24Go) and localizes to the endoplasmic reticulum (ER), suggesting a role for this enzyme in quality control (25Go,26Go). Therefore, it was surprising to find that CADASIL mutations, predicted to affect the disulfide bond structure of the EGF-like repeats, do not prevent O-fucosylation. Mutation of either O-fucose site independently (T174A or T212A) did not eliminate O-fucosylation, but mutation of both sites (T174A and T212A) caused a dramatic reduction in O-fucosylation, suggesting that both sites are modified with O-fucose (data not shown) (discussed subsequently).

Fringe elongates O-fucose residues on EGF-like repeat 4 and 5 of Notch3
To examine whether Fringe was capable of elongating O-fucose sites within the CADASIL hot spot, O-linked carbohydrates were released from purified protein and analyzed by gel filtration chromatography. Expression of the wild-type Notch3 fragment in control cells (pMirb) showed mainly monosaccharides with smaller amounts of di- and tetrasaccharides, indicating that endogenous Fringe in CHO cells can elongate O-fucose on this fragment and the amount of Fringe or the specific Fringe gene may affect the efficiency of the reaction (Fig. 2A). Co-expression of the Notch3 fragment with Lunatic Fringe resulted in a significant increase in the levels of elongated sugars (disaccharides, trisaccharides and tetrasaccharides) with a concomitant reduction in the levels of monosaccharide (Fig. 2A). Thus, O-fucose modifications within this Notch3 fragment can be elongated by Fringe and other glycosyltransferases. Elongated sugars appeared in the gel filtration profile of Notch3 fragments carrying mutations that abrogated individual glycosylation sites (T174A or T212A), indicating that both O-fucose residues are Fringe substrates (Fig. 2B and C). However, the efficiency and extent of elongation appears to be different for each EGF-like repeat. Monosaccharides and tetrasaccharides were the most predominant structures extracted from EGF 5 (represented by the T174A mutant, Fig. 2B), whereas disaccharides and tetrasacharides were the most abundant structures on EGF 4 (represented by the T212A mutant, Fig. 2C).



View larger version (17K):
[in this window]
[in a new window]
  		Figure 2. Fringe elongates O-fucose residues on EGF-like repeats 4 and 5 of Notch3. Radiolabeled ([6-3H]fucose) O-linked carbohydrates were released from the purified wild-type or mutant N3–EGF 1–5 using beta-elimination and analyzed using gel filtration chromatography. (A) The height of each peak (y-axis in c.p.m. of radioactivity) indicates the abundance of O-fucose-containing tetra-, tri-, di- and monosaccharides in wild-type Notch3 fragment purified from culture medium of cells stably expressing Lunatic Fringe-AP (LFng) or control cells (stably transfected with empty pMirb plasmid). Percentage of total radioactivity in each saccharide species is reported in the table below the graph. (B) Gel filtration chromatography analysis of carbohydrates extracted from Notch3 fragments carrying the T174A glycosylation mutation in EGF-like 4. In this experiment, the abrogation of the O-fucose site on EGF-like 4 allowed us to study the elongation of the other O-fucose site located on EGF-like 5. (C) Chromatograph of sugars extracted from a Notch3 fragment carrying the T212A mutation in EGF-like 5. This mutation removes the glycosylation site from EGF-like 5 allowing us to study the elongation of the O-fucose site located on EGF-like 4. Samples shown in both (B) and (C) were obtained from cells expressing LFng.

 
CADASIL mutations located within the hot spot impair glycosylation by Fringe
Fringe elongation of O-fucose was studied in the presence of CADASIL mutations located near glycosylation sites. The R170C and C213S CADASIL mutations caused a significant reduction in the levels of Fringe elongated sugars with a consequent increase in the levels of monosaccharide species when compared with that of the wild-type (Fig. 3A and B). Such reduction was more evident when studying the glycosylation profile of constructs carrying both a CADASIL-like and a glycosylation mutation in different EGF-like repeats. Fucose elongation was reduced in the double mutant construct carrying the R170C/T212A changes and was almost completely abrogated in the construct carrying the T174A/C213S mutations (Fig. 3C and D). Interestingly, a mutation located outside those EGF-like repeats that contain glycosylation sites also had reduced levels of elongated sugars (Fig. 3E). A Notch3 fragment carrying the R91C CADASIL-like mutation in EGF-like 2, which does not contain a glycosylation site, failed to undergo efficient Fringe-mediated elongation of the O-fucose. To confirm these findings, Notch3 fragments were purified from cells after incubation with radioactive glucosamine and analyzed by western blot and fluorography (to measure Fringe-mediated glycosylation). Consistent with the former experiment, the Notch3 fragments carrying the R91C, R170C or C213S mutation incorporated less glucosamine on a per protein basis when compared with that of the wild-type construct (Fig. 3F) Thus, CADASIL mutations located within the mutational hot spot of Notch3 impair glycosylation events subsequent to the initial O-fucosylation, specifically those mediated by Fringe. Surprisingly, the CADASIL mutations do not need to be in the same EGF-like repeat as the O-fucosylation site to affect Fringe recognition.



View larger version (32K):
[in this window]
[in a new window]
  		Figure 3. CADASIL mutations in Notch3 impair Fringe mediated elongation of O-fucose. (A–E) Graphs show the gel filtration chromatography profile of sugars extracted from wild-type and mutant Notch3 fragments. Glycosylation sites in EGF-like repeats 4 (T174) and 5 (T212) of mouse Notch3 are also present in the human Notch3 homolog. Percentage of total O-fucose in each saccharide species is reported in the table below each graph. (F) Wild-type or mutant Notch3 fragments were purified from culture media of transfected Lec1-CHO cells and analyzed by SDS–PAGE. Protein levels and incorporation of radiolabeled glucosamine were determined by western blot using anti-Myc antibody (top) and fluorography (bottom).

 
CADASIL mutations induce aberrant dimerization of Notch3 fragments
Unpaired cysteines from EGF-like repeats carrying CADASIL mutations are, in principle, available to mediate abnormal interactions with other proteins or Notch3 itself through disulfide bonds. To study this possibility, purified Notch3 fragments were analyzed by SDS–PAGE in reducing and non-reducing conditions (with or without beta-mercaptoethanol, BME). Under non-reducing conditions, the wild-type construct was resolved as an individual band of ~25 kDa (Fig. 4A). In contrast, a construct carrying the C213S CADASIL-like mutation was resolved as two bands of 25 and 50 kDa (Fig. 4A). Wild-type or mutant proteins were purified from culture media of COS-7 cells expressing Lunatic Fringe (LFng-V5) to determine a possible effect of this enzyme in the formation of higher molecular weight structures. In addition to the 25 and 50 kDa bands, SDS–PAGE analysis of the C213S mutant protein revealed another band of ~65 kDa in the presence of Lunatic Fringe (Fig. 4A). The 50 and 65 kDa bands were also resolved by SDS–PAGE analysis of total cell lysate in non-reducing conditions (Fig. 4B). In every case, the higher molecular weight conformations were disassembled by incubation of the purified protein preparations or cell lysates with reducing agent (Fig. 4A and B). The 50 kDa band likely represents homodimers formed by two mutant proteins bound through intermolecular disulfide bonds: such dimers are present in purified preparations of Notch3 fragments, are twice as big as the monomer, and are SDS/EDTA-resistant and BME/dithiothreitol (DTT)-labile (Fig. 4A and B) (data not shown). Reblotting of the membranes after stripping detected Lunatic Fringe in the 65 kDa band suggesting that in addition to homodimers, the C213S CADASIL mutation induces the formation of heterodimers containing Notch3 and Fringe itself (Fig. 4B). The proposed heterodimers are present in purified preparations of Notch3 and total lysate of cells expressing Fringe, have a molecular weight of 65 kDa corresponding to the expected size for a complex between mutant Notch3 (25 kDa) and Fringe (40 kDa), and, similar to the mutant Notch3 homodimers, are SDS/EDTA-resistant and BME/DTT-labile (Fig. 4A and B) (data not shown). Previous genetic analysis of Fringe mutants in Drosophila suggested that this enzyme may function as a dimer (32Go). In keeping with this, Lunatic Fringe was resolved as two bands of 40 and 80 kDa under non-reducing conditions and as a single band under reducing conditions (Fig. 4B). Thus, Fringe dimerization may be mediated by the formation of intermolecular disulfide bridges. In support of this interpretation, we note that Drosophila Fringe, the three human isoforms of Lunatic Fringe and human Manic are all predicted to have an unpaired cysteine residue likely to be involved in the formation of Fringe homodimers and Fringe/mutant Notch3 heterodimers. Unlike Lunatic Fringe, O-FucT-1, which has an even number of cysteine residues, was resolved as a single band by SDS–PAGE under non-reducing conditions and did not form aberrant complexes with mutant Notch3 fragments (data not shown). The R91C and the R170C CADASIL-like mutations also induced the formation of higher molecular weight structures resolved by SDS–PAGE under non-reducing conditions (Fig. 4C). Thus, aberrant dimerization of Notch3 with proteins containing an unpaired cysteine is likely to be a common feature in CADASIL.



View larger version (34K):
[in this window]
[in a new window]
  		Figure 4. CADASIL mutations induce aberrant interactions of Notch3. Notch3 fragments purified from culture medium of transiently transfected COS-7 cells (A and C) or total cell lysates (B) in loading buffer (2% SDS) were analyzed by SDS–PAGE (10%) in non-reducing or reducing conditions (10% BME). (A) The C213S mutation induces the formation of Notch3 homodimers resolved as a 50 kDa band under non-reducing conditions (left). In contrast, only the monomeric form of the C213S mutant construct is detected under reducing conditions (right). (B) The Notch3 fragment carrying the C213S mutation forms both homodimers of 50 kDa and heterodimers of 65 kDa reactive to both Notch3 (top left) and Lunatic Fringe antibodies (top right) under non-reducing conditions. Lunatic Fringe monomers (40 kDa) and homodimers (80 kDa) are resolved by SDS–PAGE in the absence of BME (top right). Addition of reducing agent to the protein samples prior SDS–PAGE results in disruption of both homodimers and heterodimers (bottom) (C) SDS–PAGE analysis of R91C and R170C mutant constructs reveal the presence of Notch3 homodimers under non-reducing conditions. Anti-Myc and Anti-V5 antibodies (Invitrogen) were used to detect Notch3 fragments and Lunatic Fringe, respectively.

 
Glycosylation mutations mimic a maturation defect caused by CADASIL mutations
Previous work showed that CADASIL mutations located in the ligand-binding domain [C428S (14Go) and C455R (15Go)] but not those located within the mutational hot spot (R90C, R133C, R141C, R170C, C183R, C212S) affect interaction with the ligands (14Go–17Go). Instead, mutations located in the hot spot appear to result in reduced levels of mature S1-cleaved receptor or cause trafficking defects (R133C, R141C and C183R) (15Go,17Go). Thus, we investigated whether glycosylation mutations reproduce some of the maturation and trafficking defects found in CADASIL mutants located in the mutational hot spot. CADASIL mutations R170C or C213S in Notch3 constructs (full-length fused to GFP, N3–GFP) decreased S1-cleaved Notch3 receptor, whereas constructs carrying individual mutations in the glycosylation sites at either EGF-like 4 or 5 did not (Fig. 5A). However, abrogation of both glycosylation sites within the mutational hot spot (T174A and T212A) or the removal of a predicted glycosylation site within the ligand-binding domain of Notch3 (T446A) resulted in a decrease in the S1-cleaved fragment when compared with that of the wild-type (Fig. 5A).



View larger version (39K):
[in this window]
[in a new window]
  		Figure 5. Glycosylation mutations in Notch3 mimic a maturation defect present in CADASIL mutants. (A) COS-7 cells were transiently transfected with wild-type or mutant versions of N3–GFP construct, lysed and analyzed directly by SDS–PAGE as total protein extract. The anti-GFP antibody detected the C-terminus tagged full-length protein (~310 kDa) and the intracellular S1-cleaved fragment (NTM, 120 kDa). Anti-{alpha}-tubulin antibody was used to show protein load (B-5-1-2, Sigma). (B) The lysate of biotinylated COS-7 cells expressing N3–GFP or N3–GFP/Lunatic Fringe (C) was either subjected to avidin D pull down prior to analysis by SDS–PAGE (P) or run directly as total protein extract (T). Lane N shows cells precipitated with avidin D without prior biotinylation. The anti-mouse Notch3 extracellular domain antibody recognized both the full-length N3–GFP (~310 kDa) and the extracellular fragment (NEC, ~210 kDa) proteins in cells transfected with wild-type or mutant Notch3 constructs. The presence of Notch3 extracellular domain in the biotinylated sample after precipitation indicated plasma membrane localization of all the glycosylation and CADASIL mutants studied. The ratio of biotinylated receptor versus total amount of receptors was similar between the wild-type and the mutant constructs (0.21±0.04 for wild-type Notch3, 0.18±0.009 for CADASIL mutants and 0.23±0.05 for glycosylation mutants) as determined by densitometric scanning of films.

 
Membrane proteins were labeled with biotin and analyzed by western blots to determine whether impaired proteolytic maturation of Notch3 resulted in reduced cell surface localization. Constructs carrying glycosylation mutations or CADASIL mutations localized to the plasma membrane of transiently transfected COS-7 cells as did the wild-type (Fig. 5B). Furthermore, Lunatic Fringe over-expression had no substantial effect on the levels of wild-type or mutant Notch3 receptor present at the plasma membrane (Fig. 5C). Together, these findings show that defective glycosylation of Notch3 may contribute to the maturation impairment found in CADASIL mutants but imply that mechanisms other than Fringe-mediated glycosylation regulate membrane localization of the receptor.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
In this work, we showed that CADASIL mutations located within the mutational hot spot reduce Fringe-mediated elongation of O-fucose and induce aberrant dimerization of Notch3 fragments (Figs 3 and 4). In support of a functional interaction between Notch3 and Fringe homologs, we found co-expression of these proteins in SMC from the human adult vasculature (Arboleda-Velasquez et al. unpublished data). On the basis of our data, we propose that CADASIL mutations cause a change in the tertiary structure or aggregation state of Notch3, resulting in impairment of both elongation of O-linked glycosylation and proteolytic maturation. Such structural changes may explain how mutations located away from the sites of glycosylation and cleavage could affect modification of Notch3 by Fringe and Furin, respectively. Furthermore, these conformational changes of the mutant Notch3 receptor preferentially appear to affect modifications like Fringe glycosylation (27Go) and Furin processing that are thought to occur in the Golgi apparatus (8Go,27Go) whereas O-fucosylation, which appears to occur in the ER (25Go,26Go), is preserved. Both impaired glycosylation and aberrant dimerization of mutant Notch3 may contribute to the abnormal accumulation of Notch3 ectodomain in the brain vasculature of CADASIL patients (5Go).

An alternative model is that Fringe-mediated elongation of O-fucose may serve as a tag to indicate correct folding of the EGF-like repeats and license Notch proteins to undergo proteolytic maturation. The decreased levels of elongated sugars in the Notch3 receptor carrying a CADASIL mutation would indicate inappropriate folding and result in inefficient cleavage by Furin. The link between proteolytic processing of Notch3 and its glycosylation state is strengthened by the observation that mutations in glycosylation sites decreased S1-processing of Notch3 (Fig. 5A).

In mammals, proteolytic maturation of Notch receptors is an important regulatory mechanism that precedes translocation to the plasma membrane. Furthermore, our biotinylation experiments are in agreement with the previous evidence showing that most of the Notch3 receptor present at the plasma membrane is cleaved. However, impaired proteolytic maturation of mutant Notch3 did not result in reduced levels of receptor present at the plasma membrane (Fig. 5B). Persistent membrane localization of mutant Notch3 in the context of defective S1-cleavage is consistent with a model where decreased delivery of mutant Notch3 to the plasma membrane is counterbalanced by impaired clearance of the receptor from the same compartment. In fact, as mentioned earlier, the Notch3 extracellular domain was reported to accumulate in the plasma membrane of SMC in CADASIL patients (5Go).

CADASIL mutations affect glycosylation of distant EGF-like repeats suggesting that a defect in Fringe-mediated chain elongation is a common mechanism by which all CADASIL mutations may operate (Fig. 3E). By removing or adding cysteines, a typical CADASIL mutation could induce the formation of relatively stable non-native intramolecular disulfide bonds between neighboring EGF-like repeats that may interfere with Notch3 recognition by Fringe. This model is consistent with previous studies showing that individual domains within a multidomain protein do not fold independently and sequentially from the N- to the C-terminus, but instead fold by way of ‘collapsed’ intermediates that contain non-native disulfide bonds (33Go,34Go). Such intermediates are normally resolved by dynamic reshuffling of disulfide bonds until individual modules acquire the native structure (33Go). However, in the presence of CADASIL mutations, the non-native bonds could be preserved and may interfere with the folding of neighboring domains.

In an alternative model, we postulate that CADASIL mutations induce the formation of intermolecular disulfide bonds and dimerization of the mutant proteins. This dimerization phenomenon could make the glycosylation sites inaccessible to Fringe even if the CADASIL mutations are located away from the target sites. In support of this model, we found that CADASIL-like mutations located within the mutational hot spot induced aberrant homodimerization of a Notch3 fragment and the formation of Notch3/Lunatic Fringe heterodimers (Fig. 4). In fact, the formation of heterodimers may render Fringe inactive and contribute to the glycosylation defect of mutant Notch3.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Cloning
Plasmids encoding the intracellular (N3IC-HA) and the extracellular (JJ5) domains of mouse Notch3 (mNotch3) were the gift of Dr Urban Lendahl. PCR was utilized to amplify the region of mNotch3 that encodes for the intracellular domain of the protein (amino acids 1664–2318) using N3IC-HA as a template. The product of the PCR reaction was subcloned into PEGFP-N1 (Clontech) plasmid using the NheI and HindIII sites to obtain a fusion of EGFP to the C-terminal region of mNotch3 intracellular domain (N3ICD–GFP). To make a full-length mNotch3 fused to EGFP (N3–GFP), we took the coding sequence of mNotch3 spanning the extracellular and transmembrane domains from JJ5 plasmid using EcoRV sites and this fragment was ligated into the N3ICD–GFP construct lacking the overlapping region. The coding region of EGF-like repeats 1–5 of mNotch3 was amplified from the JJ5 plasmid using primers containing the HindIII and ApaI restriction sites at the 5' and the 3' ends, respectively. After restriction enzyme digestion, the amplicon was subcloned into a pSecTag2A backbone, N3–EGF 1–5, (Invitrogen). Mutagenesis of Notch3 constructs was performed using Quickchange multisite-directed mutagenesis kit (Stratagene). The pcDNA3-Lunatic plasmid containing the coding sequence of mouse Lunatic Fringe (mL-Fringe) was the kind gift of Dr Sean Egan. The coding region of mL-Fringe was amplified from pcDNA3-Lunatic plasmid using PCR and subcloned into the pcDNA4 backbone (Invitrogen) to obtain a C-terminal V5 tagged version of the enzyme (LFng-V5). In each cloning step that involved PCR, the product was sequenced entirely. All the restriction enzymes were from New England Biolabs.

Production of stably transfected cell lines
CHO Lec1 cells stably transfected with either mL-Fringe-AP or with empty pMirb vector were developed and provided by the laboratory of Dr Pamela Stanley. The N3–EGF 1–5 constructs (wild-type, CADASIL-like mutants or glycosylation-site mutants) were transfected into Lec1/LFng or Lec1/pMirb cells using Geneporter (Gene Therapy Systems) according to the manufacturer's recommendations. Stable transfectants were selected by incubating cells lines with 250 µg/ml of Zeocin (Invitrogen).

Analysis of O-fucose structures
Cell lines were labeled with [6-3H]fucose or [6-3H]glucosamine as previously described (29Go). Purification of N3–EGF 1–5 fragments, analysis of O-fucose saccharide structures via alkali-induced ß-elimination and gel filtration chromatography were carried out as described earlier (27Go,35Go). Analysis of O-fucose or glucosamine incorporation via fluorography and western blot using anti-Myc antibody was carried out as previously described (29Go).

Western blot analysis of Notch3 fragments
N3–EGF 1–5 proteins were purified from culture medium of COS-7 cells transiently transfected with 8 µg of wild-type or mutant Notch3 constructs and 2 µg of LFng-V5 or control vector using Lipofectamine 2000 (Invitrogen). Purified proteins or total cell lysate in RIPA buffer [50 mM Tris–Cl (pH 8.0), 150 mM NaCl, 1% NP40, 0.5% DOC, 0.1% SDS, 5 mM EDTA and 10 mM iodoacetamide] were prepared in reducing (loading buffer containing 2% SDS and 10% BME or DTT 100 mM) or non-reducing conditions (loading buffer containing 2% SDS) prior SDS–PAGE.

Biotinylation of cell surface proteins
COS-7 cells in 60 mm dishes were transfected with 8 µg of wild-type or mutant N3–GFP and 2 µg of LFng-V5 or control vector using Lipofectamine 2000. Forty-eight hours after transfection, the cells were washed with HBSS (pH 8.0, supplemented with calcium chloride, 2.5 mM) and incubated with 2.5 mM EZ-link sulfo-NHS-Biotin (Pierce) in the same solution for 20 min on ice or lysed for direct western blot analysis as described subsequently: two washes followed by 15 min incubation of the cells in HBSS supplemented with glycine (100 mM) quenched the biotinylation reaction. Cells were lysed in RIPA buffer supplemented with protease inhibitor cocktail (Roche) and briefly sonicated. To purify biotinylated proteins, 90% of the lysate was incubated overnight with 300 µL of agarose-bound avidin D (Vector). The beads were washed with RIPA buffer without EDTA three times and boiled in SDS sample buffer to elute the protein. Proteins from biotinylated and non-biotinylated precipitated samples along with straight lysates (2%) were analyzed by SDS–PAGE. Anti-mouse Notch3 antibody (R&D, 1/300 dilution in 5% milk–TBST) was used to detect the full-length receptor and the extracellular domain of the receptor. Anti-GFP antibody (Chemicon) was used to detect the full-length protein and the intracellular domain of the protein.


   ACKNOWLEDGEMENTS
 
We thank Urban Lendahl, Sean Egan and Pamela Stanley for providing us with reagents that made this work possible. We acknowledge Spyros Artavanis-Tsakonas and Angeliki Louvi for critically reviewing this manuscript and Katie Maggard, Christopher Hill, Suresh Jasti and Angel Maldonado for technical assistance. We acknowledge the Colombian families with CADASIL and Francisco Lopera for inspiring this project. This work was supported by an award from the American Heart Association. J.F.A.-V. was supported in part by the ‘Young Minds in CNS Award’ from AstraZeneca. This work was also supported in part by grants to P.L. (HL-34636 and Fondation Leducq), P.D. (EY 05318) and R.S.H. (GM 61126).

Conflict of Interest statement. None declared.


   FOOTNOTES
 
{dagger} The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Joutel, A., Corpechot, C., Ducros, A., Vahedi, K., Chabriat, H., Mouton, P., Alamowitch, S., Domenga, V., Cecillion, M., Marechal, E. et al. (1996) Notch3 mutations in CADASIL, a hereditary adult-onset condition causing stroke and dementia. Nature, 383, 707–710.[CrossRef][Medline]

  2. Joutel, A., Corpechot, C., Ducros, A., Vahedi, K., Chabriat, H., Mouton, P., Alamowitch, S., Domenga, V., Cecillion, M., Marechal, E. et al. (1997) Notch3 mutations in cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), a Mendelian condition causing stroke and vascular dementia. Ann. NY Acad. Sci., 826, 213–217.[Web of Science][Medline]

  3. Dichgans, M., Mayer, M., Uttner, I., Bruning, R., Muller-Hocker, J., Rungger, G., Ebke, M., Klockgether, T. and Gasser, T. (1998) The phenotypic spectrum of CADASIL: clinical findings in 102 cases. Ann. Neurol., 44, 731–739.[CrossRef][Web of Science][Medline]

  4. Ruchoux, M.M., Guerouaou, D., Vandenhaute, B., Pruvo, J.P., Vermersch, P. and Leys, D. (1995) Systemic vascular smooth muscle cell impairment in cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy. Acta Neuropathol. (Berl), 89, 500–512.[Medline]

  5. Joutel, A., Andreux, F., Gaulis, S., Domenga, V., Cecillon, M., Battail, N., Piga, N., Chapon, F., Godfrain, C. and Tournier-Lasserve, E. (2000) The ectodomain of the Notch3 receptor accumulates within the cerebrovasculature of CADASIL patients. J. Clin. Invest., 105, 597–605.[Web of Science][Medline]

  6. Lardelli, M., Dahlstrand, J. and Lendahl, U. (1994) The novel Notch homologue mouse Notch 3 lacks specific epidermal growth factor-repeats and is expressed in proliferating neuroepithelium. Mech. Dev., 46, 123–136.[CrossRef][Web of Science][Medline]

  7. Blaumueller, C.M., Qi, H., Zagouras, P. and Artavanis-Tsakonas, S. (1997) Intracellular cleavage of Notch leads to a heterodimeric receptor on the plasma membrane. Cell, 90, 281–291.[CrossRef][Web of Science][Medline]

  8. Logeat, F., Bessia, C., Brou, C., LeBail, O., Jarriault, S., Seidah, N.G. and Israel, A. (1998) The Notch1 receptor is cleaved constitutively by a furin-like convertase. Proc. Natl Acad. Sci. USA, 95, 8108–8112.[Abstract/Free Full Text]

  9. Selkoe, D. and Kopan, R. (2003) Notch and Presenilin: regulated intramembrane proteolysis links development and degeneration. Annu. Rev. Neurosci., 26, 565–597.[CrossRef][Web of Science][Medline]

  10. Joutel, A., Vahedi, K., Corpechot, C., Troesch, A., Chabriat, H., Vayssiere, C., Cruaud, C., Maciazek, J., Weissenbach, J., Bousser, M.G. et al. (1997) Strong clustering and stereotyped nature of Notch3 mutations in CADASIL patients. Lancet, 350, 1511–1515.[CrossRef][Web of Science][Medline]

  11. Dichgans, M., Ludwig, H., Muller-Hocker, J., Messerschmidt, A. and Gasser, T. (2000) Small in-frame deletions and missense mutations in CADASIL: 3D models predict misfolding of Notch3 EGF-like repeat domains. Eur. J. Hum. Genet., 8, 280–285.[CrossRef][Web of Science][Medline]

  12. Mazzei, R., Conforti, F.L., Lanza, P.L., Sprovieri, T., Lupo, M.R., Gallo, O., Patitucci, A., Magariello, A., Caracciolo, M., Gabriele, A.L. et al. (2004) A novel Notch3 gene mutation not involving a cysteine residue in an Italian family with CADASIL. Neurology, 63, 561–564.[Abstract/Free Full Text]

  13. Arboleda-Velasquez, J.F., Lopera, F., Lopez, E., Frosch, M.P., Sepulveda-Falla, D., Gutierrez, J.E., Vargas, S., Medina, M., Martinez De Arrieta, C., Lebo, R.V. et al. (2002) C455R notch3 mutation in a Colombian CADASIL kindred with early onset of stroke. Neurology, 59, 277–279.[Abstract/Free Full Text]

  14. Joutel, A., Monet, M., Domenga, V., Riant, F. and Tournier-Lasserve, E. (2004) Pathogenic mutations associated with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy differently affect Jagged1 binding and Notch3 activity via the RBP/JK signaling Pathway. Am. J. Hum. Genet., 74, 338–347.[CrossRef][Web of Science][Medline]

  15. Peters, N., Opherk, C., Zacherle, S., Capell, A., Gempel, P. and Dichgans, M. (2004) CADASIL-associated Notch3 mutations have differential effects both on ligand binding and ligand-induced Notch3 receptor signaling through RBP-Jk. Exp. Cell. Res., 299, 454–464.[CrossRef][Web of Science][Medline]

  16. Haritunians, T., Boulter, J., Hicks, C., Buhrman, J., DiSibio, G., Shawber, C., Weinmaster, G., Nofziger, D. and Schanen, C. (2002) CADASIL Notch3 mutant proteins localize to the cell surface and bind ligand. Circ. Res., 90, 506–508.[Abstract/Free Full Text]

  17. Karlstrom, H., Beatus, P., Dannaeus, K., Chapman, G., Lendahl, U. and Lundkvist, J. (2002) A CADASIL-mutated Notch 3 receptor exhibits impaired intracellular trafficking and maturation but normal ligand-induced signaling. Proc. Natl Acad. Sci. USA, 99, 17119–17124.[Abstract/Free Full Text]

  18. Irvine, K.D. (1999) Fringe, Notch, and making developmental boundaries. Curr. Opin. Genet. Dev., 9, 434–441.[CrossRef][Web of Science][Medline]

  19. Hicks, C., Johnston, S.H., diSibio, G., Collazo, A., Vogt, T.F. and Weinmaster, G. (2000) Fringe differentially modulates Jagged1 and Delta1 signalling through Notch1 and Notch2. Nat. Cell Biol., 2, 515–520.[CrossRef][Web of Science][Medline]

  20. Haines, N. and Irvine, K.D. (2003) Glycosylation regulates Notch signalling. Nat. Rev. Mol. Cell Biol., 4, 786–797.[Web of Science][Medline]

  21. Haltiwanger, R.S. and Lowe, J.B. (2004) Role of glycosylation in development. Annu. Rev. Biochem., 73, 491–537.[CrossRef][Web of Science][Medline]

  22. Moloney, D.J., Shair, L.H., Lu, F.M., Xia, J., Locke, R., Matta, K.L. and Haltiwanger, R.S. (2000) Mammalian Notch1 is modified with two unusual forms of O-linked glycosylation found on epidermal growth factor-like modules. J. Biol. Chem., 275, 9604–9611.[Abstract/Free Full Text]

  23. Wang, Y., Shao, L., Shi, S., Harris, R.J., Spellman, M.W., Stanley, P. and Haltiwanger, R.S. (2001) Modification of epidermal growth factor-like repeats with O-fucose. Molecular cloning and expression of a novel GDP-fucose protein O-fucosyltransferase. J. Biol. Chem., 276, 40338–40345.[Abstract/Free Full Text]

  24. Wang, Y. and Spellman, M.W. (1998) Purification and characterization of a GDP-fucose:polypeptide fucosyltransferase from Chinese hamster ovary cells. J. Biol. Chem., 273, 8112–8118.[Abstract/Free Full Text]

  25. Okajima, T., Xu, A., Lei, L. and Irvine, K.D. (2005) Chaperone activity of protein O-fucosyltransferase 1 promotes Notch receptor folding. Science, 307, 1599–1603.[Abstract/Free Full Text]

  26. Luo, Y. and Haltiwanger, R.S. (2005) O-fucosylation of notch occurs in the endoplasmic reticulum. J. Biol. Chem., 280, 11289–11294.[Abstract/Free Full Text]

  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. et al. (2000) Fringe is a glycosyltransferase that modifies Notch. Nature, 406, 369–375.[CrossRef][Medline]

  28. Bruckner, K., Perez, L., Clausen, H. and Cohen, S. (2000) Glycosyltransferase activity of Fringe modulates Notch–Delta interactions. Nature, 406, 411–415.[CrossRef][Medline]

  29. Shao, L., Moloney, D.J. and Haltiwanger, R. (2003) Fringe modifies O-fucose on mouse Notch1 at epidermal growth factor-like repeats within the ligand-binding site and the Abruptex region. J. Biol. Chem., 278, 7775–7782.[Abstract/Free Full Text]

  30. Cohen, B., Bashirullah, A., Dagnino, L., Campbell, C., Fisher, W.W., Leow, C.C., Whiting, E., Ryan, D., Zinyk, D., Boulianne, G. et al. (1997) Fringe boundaries coincide with Notch-dependent patterning centres in mammals and alter Notch-dependent development in Drosophila. Nat. Genet., 16, 283–288.

  31. Donahue, C.P. and Kosik, K.S. (2004) Distribution pattern of Notch3 mutations suggests a gain-of-function mechanism for CADASIL. Genomics, 83, 59–65.[CrossRef][Web of Science][Medline]

  32. Correia, T., Papayannopoulos, V., Panin, V., Woronoff, P., Jiang, J., Vogt, T.F. and Irvine, K.D. (2003) Molecular genetic analysis of the glycosyltransferase Fringe in Drosophila. Proc. Natl Acad. Sci. USA, 100, 6404–6409.

  33. Jansens, A., van Duijn, E. and Braakman, I. (2002) Coordinated nonvectorial folding in a newly synthesized multidomain protein. Science, 298, 2401–2403.[Abstract/Free Full Text]

  34. Zanuttin, F., Guarnaccia, C., Pintar, A. and Pongor, S. (2004) Folding of epidermal growth factor-like repeats from human tenascin studied through a sequence frame-shift approach. Eur. J. Biochem., 271, 4229–4240.[Web of Science][Medline]

  35. Moloney, D.J., Lin, A.I. and Haltiwanger, R.S. (1997) The O-linked fucose glycosylation pathway. Evidence for protein-specific elongation of O-linked fucose in Chinese hamster ovary cells. J. Biol. Chem., 272, 19046–19050.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 NEJMHome page
S. D. Brass, E. E. Smith, J. F. Arboleda-Velasquez, W. A. Copen, and M. P. Frosch
Case 12-2009 -- A 46-Year-Old Man with Migraine, Aphasia, and Hemiparesis and Similarly Affected Family Members
N. Engl. J. Med., April 16, 2009; 360(16): 1656 - 1665.
[Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
M. Stahl, K. Uemura, C. Ge, S. Shi, Y. Tashima, and P. Stanley
Roles of Pofut1 and O-Fucose in Mammalian Notch Signaling
J. Biol. Chem., May 16, 2008; 283(20): 13638 - 13651.
[Abstract] [Full Text] [PDF]

Home page
 GlycobiologyHome page
Y. Chigira, T. Oka, T. Okajima, and Y. Jigami
Engineering of a mammalian O-glycosylation pathway in the yeast Saccharomyces cerevisiae: production of O-fucosylated epidermal growth factor domains
Glycobiology, April 1, 2008; 18(4): 303 - 314.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
E. Fung, S.-M. T. Tang, J. P. Canner, K. Morishige, J. F. Arboleda-Velasquez, A. A. Cardoso, N. Carlesso, J. C. Aster, and M. Aikawa
Delta-Like 4 Induces Notch Signaling in Macrophages: Implications for Inflammation
Circulation, June 12, 2007; 115(23): 2948 - 2956.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
R. Rampal, A. S. Y. Li, D. J. Moloney, S. A. Georgiou, K. B. Luther, A. Nita-Lazar, and R. S. Haltiwanger
Lunatic Fringe, Manic Fringe, and Radical Fringe Recognize Similar Specificity Determinants in O-Fucosylated Epidermal Growth Factor-like Repeats
J. Biol. Chem., December 23, 2005; 280(51): 42454 - 42463.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
R. Rampal, J. F. Arboleda-Velasquez, A. Nita-Lazar, K. S. Kosik, and R. S. Haltiwanger
Highly Conserved O-Fucose Sites Have Distinct Effects on Notch1 Function
J. Biol. Chem., September 16, 2005; 280(37): 32133 - 32140.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
14/12/1631    most recent
ddi171v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Arboleda-Velasquez, J. F.
Right arrow Articles by Kosik, K. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Arboleda-Velasquez, J. F.
Right arrow Articles by Kosik, K. S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?