Logo of jbcThe Journal of Biological Chemistry
J Biol Chem. 2017 Sep 22; 292(38): 15964–15973.
Published online 2017 Jul 20. doi: 10.1074/jbc.M117.800102
PMCID: PMC5612125
PMID: 28729422

O-Glycosylation modulates the stability of epidermal growth factor-like repeats and thereby regulates Notch trafficking

Hideyuki Takeuchi,‡,1,2 Hongjun Yu,§,1 Huilin Hao,‡ Megumi Takeuchi,‡ Atsuko Ito,‡ Huilin Li,§ and Robert S. Haltiwanger‡¶,3
Author information Article notes Copyright and License information Disclaimer

Associated Data

Supplementary Materials

Abstract

Glycosylation in the endoplasmic reticulum (ER) is closely associated with protein folding and quality control. We recently described a non-canonical ER quality control mechanism for folding of thrombospondin type 1 repeats by protein O-fucosyltransferase 2 (POFUT2). Epidermal growth factor-like (EGF) repeats are also small cysteine-rich protein motifs that can be O-glycosylated by several ER-localized enzymes, including protein O-glucosyltransferase 1 (POGLUT1) and POFUT1. Both POGLUT1 and POFUT1 modify the Notch receptor on multiple EGF repeats and are essential for full Notch function. The fact that POGLUT1 and POFUT1 can distinguish between folded and unfolded EGF repeats raised the possibility that they participate in a quality control pathway for folding of EGF repeats in proteins such as Notch. Here, we demonstrate that cell-surface expression of endogenous Notch1 in HEK293T cells is dependent on the presence of POGLUT1 and POFUT1 in an additive manner. In vitro unfolding assays reveal that addition of O-glucose or O-fucose stabilizes a single EGF repeat and that addition of both O-glucose and O-fucose enhances stability in an additive manner. Finally, we solved the crystal structure of a single EGF repeat covalently modified by a full O-glucose trisaccharide at 2.2 Å resolution. The structure reveals that the glycan fills up a surface groove of the EGF with multiple contacts with the protein, providing a chemical basis for the stabilizing effects of the glycans. Taken together, this work suggests that O-fucose and O-glucose glycans cooperatively stabilize individual EGF repeats through intramolecular interactions, thereby regulating Notch trafficking in cells.

Keywords: carbohydrate function, crystal structure, disulfide, glycosyltransferase, Notch protein, protein stability, O-fucose, O-glucose, O-glycosylation

Introduction

Glycosylation in the endoplasmic reticulum (ER)4 has long been associated with protein folding (1, 2). Oligosaccharyltransferase modifies newly synthesized proteins as they emerge in the lumen of the ER with N-glycans that are trimmed and recognized by calnexin/calreticulin chaperones, allowing a round of protein folding. Folded proteins move on, and unfolded proteins are recognized and re-glucosylated by the UDP-glucose glycoprotein glucosyltransferase (UGGT) to allow another folding cycle. In addition to this well-known pathway, we recently identified a non-canonical quality control pathway for folding of thrombospondin type 1 repeats (TSRs) (3). TSRs are small protein motifs (∼60 amino acids) typically found as tandem repeats in proteins (4). They are defined by six conserved cysteines forming three disulfide bonds. Most TSRs contain a consensus sequence for modification by protein O-fucosyltransferase 2 (POFUT2), a soluble enzyme in the ER lumen (5, 6). POFUT2 only modifies folded TSRs (6), and knockdown of POFUT2 in cells results in loss of secretion of multiple target proteins (3). The addition of O-fucose appears to stabilize the folded form of a TSR, accelerating its folding and secretion (3). Thus, in contrast to the classical N-glycan quality control pathway where UGGT recognizes an unfolded substrate, POFUT2 recognizes and modifies a folded structure.

Epidermal growth factor-like (EGF) repeats are also modified by O-glycans in the ER (7, 8). Similar to TSRs, EGF repeats are small protein motifs (∼40 amino acids) found as tandem repeats, also defined by six cysteines forming three disulfide bonds (9). Three O-glycans are added at specific sites on EGF repeats by three separate enzymes: protein O-glucosyltransferase 1 (POGLUT1) adds O-glucose to serine residues within the sequence C1XSX(P/A)C2 (where C1 and C2 are the first and second conserved cysteine of the EGF repeat); POFUT1 adds O-fucose to serine/threonine residues within the sequence C2XXXX(S/T)C3; and EGF-specific O-GlcNAc transferase (EOGT) adds O-GlcNAc to serine/threonine residues within the sequence C5XXGX(T/S)GXXC6 (10, 11). All three enzymes modify multiple EGF repeats in the extracellular domain of the Notch receptor (12,–14), and these modifications are essential for optimal Notch activity (15,–17). Human diseases result from inactivating mutations in these enzymes and appear to be mediated by reduced Notch function (17,–20). O-Glucose can be elongated by xylosyltransferases (GXYLT1/2 and XXYLT1) to form the trisaccharide Xylα1–3Xylα1–3Glc, and O-fucose can be elongated by Fringe enzymes (LFNG, MFNG, and RFNG) to ultimately form the tetrasaccharide Siaα2–6Galβ1–4GlcNAcβ1-3Fuc (21,–24). Elongation of O-fucose or O-glucose is known to modulate Notch activity (13, 25, 26).

POGLUT1, POFUT1, and EOGT are all localized to the ER and only modify properly folded EGF repeats (7, 8, 27), suggesting they may participate in a quality control pathway for the folding of EGF repeats similar to that of POFUT2 and TSRs. Nonetheless, the effects of eliminating these enzymes in model systems have shown a variety of effects on the trafficking of Notch to the cell surface, which seem to vary with cell type or system. Early studies in flies showed that the Drosophila form of POFUT1 (Ofut1) has chaperone activity that is independent of its fucosyltransferase activity and required for proper cell-surface expression of Notch (28). In contrast, Notch receptors, although non-functional, are expressed on the cell surface of mouse embryonic stem cells that lack Pofut1 (29). Other results in mice suggest loss of Pofut1 does result in Notch trafficking defects in certain cells (30). Similarly, elimination of the Drosophila form of POGLUT1 (Rumi) caused some Notch trafficking defects, whereas knockdown of Poglut1 in mouse cell lines did not reduce cell-surface Notch (8, 16). Elimination of EOGT in mammalian cells does not appear to have any effect on Notch trafficking (17).

A recent study in flies suggested that POGLUT1 and POFUT1 cooperate in folding and trafficking of the Notch receptor (31), raising the intriguing possibility that these enzymes may not function independently but work together, perhaps with additional cellular components, to fully fold the active Notch receptor. This would help to explain some of the variability seen in different systems described above. Here, we examine whether this is the case in mammalian cells. Using CRISPR/CAS9, we generated POGLUT1 and POFUT1 single or double knock-out cells to examine effects on Notch cell-surface expression. In addition, we used in vitro EGF stability assays to evaluate whether addition of O-glucose or O-fucose stabilizes EGF repeats similar to how O-fucose stabilizes TSRs. Finally, we solved the structure of an EGF repeat with a full O-glucose trisaccharide structure to reveal how the glycans interact with the underlying protein to offer a structural basis for the stabilizing effects of the glycan.

Results

POFUT1 and POGLUT1 facilitate cell surface expression of Notch1 in an additive manner

To examine the effects of loss of O-fucose and O-glucose glycans on Notch1 functions, we genetically deleted POFUT1 or POGLUT1, which are solely responsible for the initial attachment of O-fucose or O-glucose to EGF repeats, respectively, in HEK293T cells. We previously reported that Notch1 is endogenously expressed on the cell surface of HEK293T cells, and that Notch1 produced in HEK293T cells is modified with O-fucose and O-glucose glycans at high stoichiometries at many EGF repeats (Fig. 1A) (13, 14). Successful bi-allelic mutations at the expected regions in POFUT1 or POGLUT1 were confirmed by genomic DNA sequencing (supplemental Fig. S1A). As expected, POFUT1 KO cells (clone KB8) or POGLUT1 KO cells (clone OD3) did not show any residual enzymatic POFUT1 or POGLUT1 activities, respectively, in in vitro assay with cell extracts (supplemental Fig. S1B). Flow cytometric analysis showed that the levels of endogenous Notch1 on the cell surface were reduced by a factor of 2 in either POFUT1 or POGLUT1 KO cells compared with the control wild-type cells (Fig. 1B). Western blot analysis showed that these single KO cells contained slightly higher levels of full-length Notch1 proteins than that of the wild-type cells (Fig. 1C). Because full-length Notch1 has not been processed by furin in the Golgi, this observation suggests an increased level of Notch1 in the ER in these cells. These results demonstrated that POFUT1 and POGLUT1 are required for efficient Notch1 trafficking from the ER to the cell surface in HEK293T cells.

An external file that holds a picture, illustration, etc. Object name is zbc0381773180001.jpg

POFUT1 and POGLUT1 function in an additive, non-redundant manner for efficient Notch1 cell-surface expression in HEK293T cells. A, schematic structure of the extracellular domain of Notch1. Each oval represents a single EGF repeat. EGF repeats with the consensus sequence for O-fucosylation (red) and/or O-glucosylation (blue) are indicated. B, flow cytometric analysis of cell-surface, endogenous Notch1 in HEK293T clones. Wild-type control (clone AB3), upper left panel; POFUT1 KO (clone KB8), upper right panel; POGLUT1 KO (clone OD3), lower left panel; POFUT1 and POGLUT1 double KO (clone 2D3), lower right panel. In each panel, the red line is anti-Notch1 antibody, and the black line is isotype-matched control antibody. The bar graph at the bottom shows difference in mean fluorescent intensity between anti-Notch1 and control for each clone from three independent measurements. The values indicate the mean ± S.E. C, Western blot analysis of clones analyzed in (B). Green channel, anti-Notch1; red channel, anti-tubulin. D, secretion assay with pSecTag-mN1(1–36)-MycHis6 expressed in WT HEK293T cells (clone AB3), POFUT1 KO (clone KB8), and POGLUT1 KO (clone OD3). EV, empty vector; F, co-transfection of POFUT1-MycHis6; G, co-transfection of POGLUT1-MycHis6. IgG was a secretion control. Red channel, anti-MYC. Green channel, anti-IgG. Top panels, medium (loaded based on volume). Bottom panels, cell lysate (loaded based on protein).

We next generated POFUT1 and POGLUT1 double KO cells using the above POFUT1 single KO cells. We confirmed their genotypes by genomic sequencing (supplemental Fig. S1A) and in vitro enzymatic assays (supplemental Fig. S1B). Double KO cells showed further reduction in the cell-surface expression of Notch1 compared with POFUT1 or POGLUT1 single KO cells (Fig. 1B). Of note, Western blot analysis showed increased full-length Notch1 proteins in the double KO cell lysates compared with the single KO cells (Fig. 1C). These results strongly suggested that both O-fucose and O-glucose glycans are required for proper Notch1 trafficking in HEK293T cells and that they function in a non-redundant, additive manner.

To demonstrate that the requirement for POFUT1 and POGLUT1 localizes to the EGF repeats in the extracellular domain of Notch1, we performed cell-based secretion assays with a Notch1 construct encoding EGF(1–36) (N1(1–36)) (Fig. 1D). N1(1–36) was expressed at comparable levels in cell lysates of all cells except the empty vector controls. In wild-type control cells, N1(1–36) was secreted to the culture media. In POFUT1 or POGLUT1 single KO cells, N1(1–36) was not secreted, whereas a control protein without EGF repeats, IgG, was unaffected. In POFUT1 KO cells, co-transfection of POFUT1 rescued this secretion defect, whereas co-transfection of POGLUT1 did not. Likewise, in POGLUT1 KO cells, co-transfection of POGLUT1 rescued the secretion defect, whereas co-transfection of POFUT1 did not. These results support the above notion that addition of O-fucose by POFUT1 and O-glucose by POGLUT1 to the EGF repeats in the Notch1 extracellular domain are both required for efficient protein folding and secretion in HEK293T cells.

O-Fucose and O-glucose stabilize EGF repeats independently

Our recent studies suggested that POFUT2 is required for secretion of TSR-containing proteins because addition of O-fucose stabilizes the folded TSRs in the ER (3). To evaluate whether O-fucose and O-glucose perform a similar function on EGF repeats, we adapted the unfolding assay used in the TSR study to measure the effects of the sugars on the stability of EGF repeats. We used the first EGF repeat from human coagulation factor 9 (hFA9) as a model because it contains similar numbers of amino acids as Notch EGF repeats, contains consensus sequences for both O-fucose and O-glucose modifications, and is endogenously modified with these types of O-glycans (32). As with TSRs, folded and denatured hFA9 EGF can be separated by reverse-phase HPLC, and monitoring UV absorbance allows us to quantitate the amounts of each species (supplemental Fig. S2). Dithiothreitol (DTT) unfolded hFA9 EGF by reducing the disulfide bonds, although it was not unfolded in the presence of guanidine hydrochloride (GdnHCl) alone (supplemental Fig. S2, A and C). GdnHCl accelerated the rate of unfolding by DTT (supplemental Fig. S2B). We also compared DTT sensitivities of several EGF repeats from hFA9 and mouse Notch1 (33) (supplemental Fig. S3). These EGF repeats showed differential sensitivities to DTT at a specific time point, suggesting different EGF repeats have distinct stabilities. Nonetheless, these results clearly indicated that this assay can be used to monitor the stability of EGF repeats in the presence of DTT.

To examine the effect of O-fucose and O-glucose modifications on the stability of EGF repeats, we modified hFA9 EGF with O-fucose and/or O-glucose glycans by in vitro glycosylation. Successful addition of O-glycans to the EGF repeat was confirmed by mass spectrometry (supplemental Table S1). The resulting glycosylated and unglycosylated EGF repeats were subjected to unfolding assays (Fig. 2A and supplemental Figs. S4 and S5). Addition of O-glucose monosaccharide slowed the unfolding of the EGF repeat compared with that of the unglycosylated EGF repeat. Exogenous free glucose did not show any effect on the unfolding of unglycosylated and Glc-EGF (data not shown), suggesting that the glucose attached covalently to the Ser in the O-glucose consensus stabilizes the folded EGF. We also examined whether elongation of O-glucose by xylose residues affects stability of EGF repeats (Fig. 2A, top row). Interestingly, addition of the first xylose canceled the stabilizing effect of O-glucose monosaccharide (i.e. it destabilized the EGF repeat). Addition of the second xylose stabilized the folding of EGF somewhat compared with Xyl-Glc-EGF.

An external file that holds a picture, illustration, etc. Object name is zbc0381773180002.jpg

O-Fucose and O-glucose glycans stabilize hFA9 EGF in an additive manner. In vitro unfolding assays were performed in the absence (A) or presence (B) of 5 mm calcium. The quantification of area under the curve values of folded (open circles) and unfolded (closed circles) forms of each glycoform, including unglycosylated hFA9 EGF repeat, was performed as in supplemental Fig. S2. Blue circle represents glucose; orange star, xylose; red triangle, fucose. The assays were performed independently at least three times with all glycoforms. The values indicate the mean ± S.E. See supplemental Figs. S4 and S5 for HPLC profiles used to generate these curves.

Next, we examined how the addition of O-fucose affected the stability of the hFA9 EGF repeat (Fig. 2A, bottom row). Addition of O-fucose also slowed the unfolding of the EGF repeat compared with that of the unglycosylated hFA9 EGF repeat. Addition of both O-fucose and O-glucose unfolded more slowly than either alone, indicating that O-fucose and O-glucose monosaccharides stabilize single EGF repeats in an additive manner. Addition of xyloses to the O-fucose/O-glucose-modified EGF had similar effects to that seen for the O-glucose alone.

hFA9 EGF repeat has a calcium-binding motif (34), and binding of a calcium ion stabilizes EGF repeats (35). We confirmed the stabilizing effect of calcium using our unfolding assay (compare Fig. 2B with 2A). We examined whether O-fucose or O-glucose glycans can further stabilize the hFA9 EGF repeat that has been stabilized by calcium. All EGF repeats in the presence of calcium unfolded more slowly than those in the absence of calcium (Fig. 2B compared with 2A). The O-glycan modifications showed similar effects on unfolding of hFA9 EGF repeat in the absence and presence of calcium ions (Fig. 2B compared with 2A).

Significance of the O-glucose monosaccharide at EGF28 on mouse Notch1 function

We previously reported that O-glucosylation at EGF28 is required for Delta-like 1-induced Notch1 signaling in NIH3T3 cell-based assays (14). In vitro enzymatic assays suggested that EGF28 of mouse Notch1 is a poorer substrate for POGLUT1, suggesting the efficiency of modification on EGF28 could vary depending on the levels of POGLUT1 in a particular cell (33). Our mass spectral approach revealed that the vast majority of O-glucose glycans attached to EGF28 of mouse Notch1 from NIH3T3 cells was a monosaccharide species (Fig. 3A and supplemental Fig. S6). These data indicated that the attachment of an O-glucose monosaccharide to mouse Notch1 EGF28 is important for its Delta-like 1-mediated activation.

An external file that holds a picture, illustration, etc. Object name is zbc0381773180003.jpg

O-Glucose stabilizes EGF28 of mouse Notch1. A, EICs of the ions corresponding to glycoforms of a peptide containing the O-glucose consensus sequence from EGF28 of mouse Notch1 produced in NIH3T3 cells. See supplemental Fig. S6 for mass spectra confirming the identity of the glycopeptide, 1063CDSAPCKNGGRCW1075. Black line, unglycosylated; red line, O-glucose monosaccharide, blue line, O-glucose disaccharide; green line, O-glucose trisaccharide. B, 20-min unfolding assays with various glycoforms of EGF28 from mouse Notch1. Reverse phase HPLC was performed to separate folded (F) and unfolded (U) forms. One representative data set is shown from the assays that were performed independently three times with all glycoforms. C, quantification of area under the curve values of folded (open bars) and unfolded (filled bars) forms of each glycoform from B. Assays were performed independently three times. The values indicate the mean ± S.E.

We then examined whether addition of O-glucose glycans to EGF28 of mouse Notch1 changes protein stability. Because EGF28 of mouse Notch1 does not have a calcium-binding motif, we performed a single time point unfolding assay in the absence of calcium. Because EGF28 was more resistant to DTT than hFA9 EGF repeat (supplemental Fig. S3), we optimized DTT concentration so that we could analyze the effects of O-glucose glycan modification (supplemental Fig. S7). EGF28 modified with an O-glucose monosaccharide was denatured more slowly than unglycosylated EGF28 (Fig. 3, B and C). As seen with hFA9 EGF repeat (Fig. 2), the first xylose destabilized EGF28, whereas the second xylose stabilized EGF28 (Fig. 3, B and C). Thus, O-glucose monosaccharides specifically at EGF28 may regulate Notch1 activation through modulating the stability of the EGF repeat.

Structural analysis of hFA9 EGF repeat modified with O-glucose glycans

To gain insight into the structural effects of O-glucose glycans on EGF repeats, we crystallized the hFA9 EGF repeat modified with O-glucose trisaccharide and determined its crystal structure to 2.2 Å resolution (Fig. 4 and supplemental Table S2). We found that the elongated O-glucose glycan occupied and filled up a deep surface groove of the EGF, and there were numerous intramolecular interactions between the glycan and EGF repeat (Fig. 4A). The O-glucose of the trisaccharide stacked against proline 55 and interacted with the main chain atoms of glutamine 50. The first xylose residue of the trisaccharide stacked against tyrosine 69 and formed the hydrogen bond with the main chain oxygen atom of asparagine 67. The second xylose stacked against tyrosine 69 and the side chains of serine 68 and lysine 80. This xylose further formed a H-bond with the main chain nitrogen atom of tyrosine 69. Proline 55 and tyrosine 69 formed a hydrophobic patch, which we previously identified as a structural signature for O-glucose addition by Rumi/POGLUT1 (36). Here we found this patch forms van der Waals interaction with all three sugar rings of the trisaccharide (Fig. 4, B and C). Sequence analysis of mouse Notch1 receptor showed that the hydrophobic patch is conserved in 15 of the 17 EGF repeats with O-glucose modifications (14); proline 55 in EGF9 and tyrosine 69 in EGF10 are each replaced by an alanine (Fig. 4D). Interestingly, the structure of trisaccharide-modified hFA9 is similar to its unmodified counterpart (PDB code 1EDM), with an r.m.s.d. of only 0.48 Å, indicating that there was no significant conformational change upon attachment of the O-glucose trisaccharide (Fig. 5A). Because the interaction between O-glucose trisaccharide and hFA9 involves either conserved residues such as proline 55 and tyrosine 69 or main chain atoms of non-conserved residues, we suggest that the mode of interaction observed here is likely applicable to many other EGF repeats with very different primary sequences such as those in mouse or human Notch receptors. Indeed, we observed similar stabilizing effects of O-glucose glycans on FA9 EGF repeat and mouse Notch1 EGF28 (Figs. 2 and ​and33).

An external file that holds a picture, illustration, etc. Object name is zbc0381773180004.jpg

Crystal structure of hFA9 EGF modified with an O-glucose trisaccharide reveals multiple intramolecular interactions. A and B, two views of the structure of hFA9 EGF covalently modified with Xylα1–3Xylα1–3Glcβ1. The protein structure is shown as blue schematic and the glycan as green sticks. Amino acids interacting with the glycan are also shown as sticks. Intramolecular interactions of the O-glucose trisaccharide with the underlying amino acids are shown as dashed lines for H-bonds and as curved and dashed lines for hydrophobic stacking. The boxed window in B shows 2Fo − Fc electron density map of the O-glucose trisaccharide in mesh view contoured at 1.2σ. C, O-glucose trisaccharide binds to a hydrophobic groove on the surface of hFA9. The hFA9 EGF is shown in electrostatic surface potential view. D, sequence alignment of hFA9 EGF with all 17 EGFs of mouse Notch1 with known O-glucose modifications (14). The hydrophobic region, corresponding to Pro-55 and Tyr-69 of hFA9 EGF (indicated by two red arrows) is conserved except for EGF9 and EGF10.

An external file that holds a picture, illustration, etc. Object name is zbc0381773180005.jpg

O-Glucose and O-fucose glycans are located on opposite faces of EGF repeats. A, superimposition of hFA9 EGF repeat modified with an O-glucose trisaccharide (this study) with unmodified hFA9 EGF repeat (green, PDB code 1EDM), rat Notch1 EGF12 modified with an O-glucose monosaccharide (magenta, PDB code 4XLW) and disaccharide (yellow, PDB code 5UK5), and human Notch1 EGF12 with O-fucose monosaccharide (orange, PDB code 4CUD) and disaccharide (cyan, PDB code 4D0E). B–G, all structures superimposed in A are shown here individually as surface views to highlight the surface groove and the opposite locations of O-glucose and O-fucose glycans.

Previous studies have revealed structures of EGF modified by O-fucose and O-glucose glycans (37,–39). We superimposed the current structure of trisaccharide-modified hFA9 with an O-glucose monosaccharide-modified or an O-glucose disaccharide-modified EGF12, and we found that they align well with an r.m.s.d. of 0.83 and 0.65 Å, respectively (Fig. 5A). It is clear that O-glucose saccharides, mono-, di-, or tri-saccharide, all fold back onto the surface groove of the EGF, rather than protruding to the solvent. The trisaccharide forms a larger interface with underlying surface groove than that of the mono-and disaccharide (Fig. 5, A–D). We further compared the above-described structures with the structure of EGF12 either unmodified or modified with an O-fucose monosaccharide or an O-fucose disaccharide (Fig. 5, A and E–G) (40), and we found that O-fucose glycans are on the opposite side of the protein from where the O-glucose glycans attach. This result supports our cell biological and biochemical observation that the effects of O-fucose and O-glucose on EGF repeats are independent and therefore additive.

Discussion

In this study, we showed that POFUT1 and POGLUT1 function in an additive, non-redundant manner for efficient cell-surface expression of endogenous Notch1 and secretion of the Notch1 extracellular domain in HEK293T cells. We also showed that simple addition of O-fucose and O-glucose glycans stabilizes a single EGF repeat in an additive, non-redundant manner as measured using in vitro unfolding assays. Taken together, these data suggest that POFUT1 and POGLUT1 participate in a novel ER quality control pathway for the folding and stabilization of EGF repeats similar to what we have previously shown for POFUT2 and TSRs (3). In addition, our structural analysis of the hFA9 EGF repeat modified with an O-glucose trisaccharide by X-ray crystallography indicated that the O-glucose trisaccharide lies flat on the surface of the EGF repeat and is highly ordered, associating with underlying amino acids, providing a rationale for the observed stabilization of the EGF repeat.

Our results are consistent with the cooperative effects of POFUT1 and POGLUT1 on Notch trafficking recently reported in the fly system (31). Addition of xylose to O-glucose has been reported to reduce the level of cell-surface Notch in Drosophila (25). Analysis of Drosophila Notch produced in S2 cells suggests that the majority of O-glucose glycans are the Xyl-Glc disaccharide with very little trisaccharide (12). Thus, the effect of xylose on cell-surface expression is consistent with our result that addition of a single xylose to O-glucose destabilizes EGF repeats (Fig. 2, A and B). A more recent report from the Matsuno and co-workers (41) suggested that addition of xylose to O-glucose was partially redundant with O-fucose in Notch trafficking and function, inconsistent with our stabilization data. A potential explanation for this difference was also presented in this report, that addition of xylose to O-glucose mainly affects transcytosis of Notch from the apical plasma membrane to the adherens junctions, thereby also affecting cell-surface levels of Notch (41). Although it is likely that the destabilization caused by addition of xylose to O-glucose mainly affects trafficking to the plasma membrane, it is possible that destabilization also affects the conformation of Notch, which may be important for transcytosis. Alternatively, xylose may affect transcytosis independently of destabilization. A recent study also demonstrated that xylosylation of the O-glucose glycans of Drosophila Notch decreased trans- but not cis-Delta ligand binding (26). Further work on xylosylation is clearly needed to better understand its effects on Notch. Thus, many of the Notch trafficking defects observed in the fly mutants may be explained by loss of stabilizing effects of these glycans on EGF repeats. The cooperative/additive effect of adding glycans to EGF repeats on stabilization also suggests that there are multiple mechanisms to efficiently stabilize and fold an EGF repeat. Thus, the differences in Notch trafficking seen in distinct systems lacking either POFUT1 or POGLUT1 (28,–30) could well be attributed to differences in levels of these enzymes or other chaperones/quality control systems for EGF folding in each distinct cell system.

Our structural data suggest that the mechanism for EGF stabilization by glycans is the multiple interactions with underlying amino acids. Significantly, addition of the O-glucose trisaccharide did not alter the conformation of the EGF repeat. We had previously shown that addition of O-fucose monosaccharide or disaccharide to EGF12 of Notch1 also does not induce conformational change of the EGF repeat, but similar to what we observed here, the O-fucose glycan has multiple interactions with underlying amino acids (40). Similar results have been reported by other investigators (37, 39). These glycans are highly ordered on the surface of EGF repeats and seem to be “born to stabilize” EGF repeats. Because EGF repeats share a unique structural fold, modulations of EGF stability by O-glycans revealed in this study are likely important for Notch and other targets of POFUT1 and POGLUT1. Highly ordered O-linked sugar modifications may also be important for molecular recognition because O-linked sugars may work as “surrogate amino acids” in a given protein structure. For example, O-fucose glycans attached to EGF8 and EGF12 of Notch1 interact with the underlying amino acids and form a part of the Notch ligand-binding surface (37,–40). O-Glucose glycans on Notch are not required for Notch–ligand binding (8, 16), which is supported by recent co-crystallized structures of Notch1 with ligands (37, 38). Instead, O-glucose glycans apparently conceal the exposed hydrophobic region of EGF repeats that may otherwise non-specifically bind to unrelated molecules or Notch itself (Fig. 4). Indeed, the accumulation of Notch in POGLUT1-null HEK293T cells may be partly attributed to aggregation of Notch1 via hydrophobic interactions. Of note, our recent structural analysis of POGLUT1 indicated that the hydrophobic region formed by tyrosine 69 and proline 55 of hFA9 EGF repeat is specifically recognized by POGLUT1/Rumi (36).

It is not yet clear why there are so many EGF repeats in the Notch extracellular domain, and why O-linked glycan modifications are so conserved. In the Notch receptor activation cascade, ligand binding to Notch followed by ligand endocytosis into signal-sending cells generates a pulling force that is required for Notch receptor activation in signal-receiving cells (42). Alterations in the rigidity of individual EGF repeats may play an important role in regulation of generation of mechanical force necessary for Notch activation. Our data suggest the alterations in O-glycosylation of EGF repeats may provide a mechanism to regulate the rigidity of the Notch extracellular domain.

Experimental procedures

CRISPR/Cas9-mediated genome editing

HEK293T cells (ATCC) were grown in DMEM–high glucose supplemented with 10% bovine calf serum and 1% penicillin and streptomycin. Gene-specific gRNAs were inserted into CAG-Cas9–2A-GFP vector (43). The gRNA sequence for targeting POFUT1 exon 1 was 5′-CAGTAGAGCAGGTAACCGGC-3′. The sequence for targeting POGLUT1 exon 1 was 5′-GGTGGGCTAGCTCGCCGCTT-3′. The expression vector (2 μg) was transfected in HEK293T cells grown in 10-cm dishes using PEI. Single cell sorting was performed in 96-well plates using the MoFlo XDP Cell Sorter (Beckman Coulter) at the Center for Tropical and Emerging Global Diseases Cytometry Resource Laboratory at the University of Georgia. Successful gene editing was confirmed by genomic PCR with Q5 polymerase (New England Biolabs) and DNA sequencing. For establishing double knock-out cells, POGLUT1 KO cells (clone OD3) were transfected with CAG-Cas9–2A-GFP vector encoding the gRNA targeting POFUT1 exon 1. After a 24-h transfection, GFP-positive cells were sorted in 96-well plates. Successful gene editing was confirmed by genomic PCR with Q5 polymerase and DNA sequencing.

Analysis of endogenous Notch1 expression by flow cytometry and Western blotting

Notch1 cell-surface expression in HEK293T cells was analyzed using an Accuri6 flow cytometer (BD Biosciences). HEK293T cells (wild-type, POFUT1 null, POGLUT1 null, or double null) were washed with Hanks' balanced salt solution supplemented with 1% BSA, 1 mm CaCl2, and 0.02% NaN3 (FACS buffer) and then incubated with 16 μg/ml phycoerythrin-conjugated anti-human Notch1 (BioLegend, clone MHN1–519) or phycoerythrin-conjugated mouse IgG1 (BioLegend, clone MOPC-21) as control on ice for 1 h. After washing with 1 ml of FACS buffer twice, the cells were analyzed by flow cytometry. Endogenous Notch1 proteins in the whole-cell lysates were analyzed by Western blotting. Cell lysates were prepared with RIPA buffer containing protease inhibitors. Protein concentrations in cell lysates were determined by using the protein assay kit (Thermo Fisher Scientific) with BSA as a standard. Twenty μg of total proteins were loaded onto 10% SDS-PAGE, transferred to nitrocellulose membrane, and blocked with 5% milk in TBS containing 0.1% Tween 20 (TBS-T) at room temperature for 1 h. The membrane was incubated with rabbit anti-Notch1 antibody (Cell Signaling Technology, clone D1E11, 1:2,000) or mouse anti-tubulin antibody (Sigma, clone DM1A, 1:2,000) at 4 °C overnight. After washing with TBS-T three times for 10 min, the membrane was incubated with IRDye 800-conjugated goat anti-rabbit antibody (LI-COR 1:10,000) or IRDye 680-conjugated goat anti-mouse IgG antibody (LI-COR, 1:10,000) at room temperature for 1 h. After the membrane was washed with TBS-T for 10 min six times, and the bands were visualized using an Odyssey System (LI-COR).

Secretion assays with Notch1 extracellular domain

The plasmids used for secretion assays were pSecTag2c-Notch1 EGF(1–36)-Myc-His6 (44), pcDNA4-POGLUT1-Myc-His6 (45), and pSecTag2c-POFUT1-Myc-His6 (7). A plasmid encoding human IgG (3) was used as secretion control. HEK293T cells (1 × 106) were seeded in 6-well dishes in DMEM, 10% calf serum and transiently transfected with 1 μg/well pSecTag-Notch1 EGF(1–36)-Myc-His6 (or empty vector), 0.5 μg/well plasmids for POGLUT1 or POFUT1 (or corresponding empty vector), and 0.5 μg/well IgG plasmid using PEI. After 4–5 h, the medium was changed to 1 ml of Opti-MEM I. The cells were cultured for another 3 days. Culture media samples were collected, and cell pellets were lysed as described above. Samples were loaded onto 10% SDS-PAGE, transferred to nitrocellulose membrane, and blocked with 5% milk in TBS containing 0.1% Tween 20 (TBS-T) at room temperature for 1 h. The membrane was incubated with anti-Myc antibody (9E10) diluted in 5% milk in TBS-T at 4 °C overnight. After washing with TBS-T for 10 min three times, the membrane was incubated with IRDye 680-conjugated goat anti-mouse IgG antibody (LI-COR 1:10,000) and IRDye 800-conjugated goat anti-human IgG antibody (LI-COR, 1:10,000) diluted in TBS-T at room temperature for 1 h. After the membrane was washed with TBS-T for 10 min six times, the bands were visualized using an Odyssey System (LI-COR).

Preparation of EGF repeats and glycosyltransferases

EGF repeats were expressed in Escherichia coli and purified as described previously (33). The cDNA encoding EGF1 of human factor IX and EGF28 of mouse Notch1 was cloned into a pET20b(+) expression vector. BL21 (DE3) E. coli was transformed with the expression vectors. Protein expression was induced with 0.4 mm isopropyl 1-thio-β-d-galactopyranoside at the time when the A600 values of the E. coli cell suspensions reached 0.5–0.6, and the E. coli were cultured in the shaker at 200 rpm at 20 °C overnight. The E. coli pellet was suspended in 50 mm Tris-HCl, pH 8.0, with 200 mm PMSF and sonicated. After centrifuging, the soluble fractions were applied to Ni-NTA-agarose (Qiagen) affinity chromatography, and after washing, the bound proteins were eluted with TBS containing 250 mm imidazole. Further purification of the EGF repeat was performed by reverse phase HPLC (1200 Series, Agilent Technologies). A preparatory C18 reverse phase column (Protein and Peptide C18, 10 × 250 mm, Vydac) was eluted with a linear gradient from 10 to 90% solvent B (0.1% TFA in 80% acetonitrile in water) in solvent A (0.1% TFA in water) at a flow rate of 2 ml/min for 60 min. Eluates were monitored for absorbance at 214 nm.

Recombinant glycosyltransferases (POGLUT1-MycHis6, GXYLT1-MycHis6, XXYLT1-MycHis6, and His6-POFUT1) were expressed in HEK293T cells and purified from medium by Ni-NTA-agarose (Qiagen) affinity chromatography as described previously (33). Enzyme concentrations were determined by 10% SDS-PAGE followed by Coomassie staining with BSA as a standard.

Preparation of glycosylated EGF repeats

In vitro glycosylation and purification of glycosylated EGF repeats were performed as described previously (33). For addition of O-glucose, 10–20 μm EGF repeats were incubated with 200 μm UDP-Glc and recombinant POGLUT1 in 50 mm Tris-HCl, pH 6.8, and 10 mm MnCl2. After full modification of EGF repeats with O-glucose was confirmed by mass spectrometry, 200 μm UDP-Xyl and recombinant GXYLT1 were added to the reaction mixture to synthesize O-glucose disaccharides. Recombinant XXYLT1 was added subsequently to make the O-glucose trisaccharides. For addition of O-fucose monosaccharide, 10–20 μm EGF repeats were incubated with 200 μm GDP-Fuc and recombinant POFUT1 in 50 mm Tris-HCl, pH 6.8, and 10 mm MnCl2. All products were purified by reverse phase HPLC, and their molecular weights were confirmed by mass spectrometry.

Mass spectrometry

EGF repeats were injected to Q-Exactive Orbitrap mass spectrometer (Thermo Fisher Scientific) equipped with an Easy nano-LC 1000 HPLC system with Accucore C4 reverse phase column (Thermo Fisher Scientific). Separation was performed with a binary gradient solvent system that consists of solvent A (0.1% formic acid in water) and solvent B (80% acetonitrile and 0.1% formic acid in water) with a constant flow rate of 300 nl/min. Spectra were recorded with a resolution of 70,000 in the positive polarity mode over the range of m/z 400–2,000 and an automatic gain control target value was 1 × 106. Raw data files were analyzed using Proteome Discoverer version 2.1 (Thermo Fisher Scientific).

Mass spectral analysis of mouse Notch1 produced in NIH3T3 cells was performed as described previously (13). NIH3T3 cells were transiently transfected with pSecTag2c-Notch1 EGF(1–36)-Myc-His6 and purified from the medium using Ni-NTA-agarose (Qiagen) affinity chromatography. Purified mouse Notch1 EGF(1–36)-Myc-His6 was in-gel digested by chymotrypsin. Chymotryptic digests were analyzed by nano-LC-MS/MS using an Agilent 6130 mass spectrometer as described (13). Extracted ion chromatograms (EICs) were performed by searching for the ions corresponding to the doubly and triply charged glycoforms of the peptide, 1063CDSAPCKNGGRCW1075, from EGF28 of mouse Notch1. The predicted m/z for the doubly charged ions corresponding to the unglycosylated, O-glucose monosaccharide, disaccharide, or trisaccharide glycoforms used for EIC searches were 784.9, 865.9, 931.9, or 997.9, respectively.

Unfolding assay

HPLC-based unfolding assays were performed essentially as described (3). EGF proteins (5–10 μm) were denatured in 100 mm Tris-HCl, pH 7.4, containing 0.6 m GdnHCl, 2 mm dithiothreitol (DTT), and 10 mm EDTA at room temperature for various times. The unfolding reaction was stopped by addition of 2% (v/v) trifluoroacetic acid (TFA). The samples were analyzed by reverse phase HPLC, and the absorbance at 214 nm wavelength was monitored. Quantification of areas under the curve was performed by using Agilent software.

X-ray crystallography

Lyophilized hFA9 EGF repeat with O-glucose trisaccharide modification was solubilized to 20 mg/ml in 10 mm HEPES and 10 mm CaCl2. Hanging-drop diffusion method was used to produce initial cluster crystals in mother liquor containing 0.1 m Tris, pH 8.5, 15% glycerol, and 1.6 m (NH4)2SO4 at 20 °C. Single crystals with good diffraction quality were obtained by seeding optimization with 5% 2,2,2-trifluoroethanol as additive. Well solution with increasing glycerol concentrations up to 30% was used as the cryo-protectant for flash-freezing crystals in liquid nitrogen. X-ray diffraction data were collected at the APS LRL-CAT at 0.9793-Å wavelength. Diffraction images were processed and scaled in XDS (46) and Mosflm (47). The structure was solved by molecular replacement with MOLREP (48) using hFA9 EGF structure (PDB code 1EDM) as the search model, followed by one round of automatic refinement in REFMAC (49) in CCP4 suite (50). O-Glucose trisaccharide was modeled based on the well-defined density present at this stage. Then, several additional rounds of refinement were carried out in REFMAC followed by manual adjustment in COOT (51). Water molecules were added at the last step of refinement. The crystallographic statistics for data collection and refinement are presented in supplemental Table S2.

Accession codes

The coordinate and the structure factor files of hFA9-Glc-Xyl-Xyl were deposited in PDB with accession number 5VYG.

Author contributions

H. T., H. Y., H. L., and R. S. H. designed the research. H. T. and H. H. performed cell biological experiments. H. T., M. T., and A. I. performed biochemical and mass spectral experiments. H. T. and M. T. prepared O-glucosylated EGF proteins for X-ray crystallography. H. Y. crystallized O-glucosylated EGF proteins. H. T., H. Y., H. L., and R. S. H. analyzed the data and wrote the paper. All authors contributed to finalizing the manuscript and approved the final version of the manuscript.

Supplementary Material

Supplemental Data:

Acknowledgments

We thank Drs. Steve Dalton and Amar M. Singh at University of Georgia for assistance with CRISPR/Cas9 knockouts, Julie Nelson at Center for Tropical and Emerging Global Diseases, Cytometry Shared Resource Laboratory at University of Georgia for assistance with cell sorting, and Dr. Shinako Kakuda for assistance with mass spectral analysis of EGF28. We also thank members and rotation students of the Haltiwanger laboratory and the Li laboratory for helpful discussion and critical comments on this manuscript. This research used resources of the Advanced Photon Source, a United States Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357. Use of the Lilly Research Laboratories Collaborative Access Team (LRL-CAT) beamline at Sector 31 of the Advanced Photon Source was provided by Eli Lilly Co., which operates the facility.

This work was supported by National Institutes of Health Grants GM061126 (to R. S. H.) and AG029979 (to H. L.). The authors declare that they have no conflicts of interest with the contents of this article. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

This article was selected as one of our Editors' Picks.

The atomic coordinates and structure factors (code 5VYG) have been deposited in the Protein Data Bank (http://wwpdb.org/).

This article contains supplemental Tables S1–S2 and Figs. S1–S7.

4The abbreviations used are:

ER
endoplasmic reticulum
UGGT
UDP-glucose glycoprotein glucosyltransferase
TSR
thrombospondin type 1 repeat
Ni-NTA
nickel-nitrilotriacetic acid
PDB
Protein Data Bank
EIC
extracted ion chromatogram
EOGT
EGF-specific O-GlcNAc transferase
r.m.s.d.
root mean square deviation
GdnHCl
guanidine hydrochloride.

References

1. Braakman I., and Hebert D. N. (2013) Protein folding in the endoplasmic reticulum. Cold Spring Harb. Perspect. Biol. 5, a013201. [PMC free article] [PubMed] [Google Scholar]
2. Xu C., and Ng D. T. (2015) Glycosylation-directed quality control of protein folding. Nat. Rev. Mol. Cell Biol. 16, 742–752 [PubMed] [Google Scholar]
3. Vasudevan D., Takeuchi H., Johar S. S., Majerus E., and Haltiwanger R. S. (2015) Peters plus syndrome mutations disrupt a noncanonical ER quality-control mechanism. Curr. Biol. 25, 286–295 [PMC free article] [PubMed] [Google Scholar]
4. Adams J. C., and Tucker R. P. (2000) The thrombospondin type 1 repeat (TSR) superfamily: diverse proteins with related roles in neuronal development. Dev. Dyn. 218, 280–299 [PubMed] [Google Scholar]
5. Luo Y., Koles K., Vorndam W., Haltiwanger R. S., and Panin V. M. (2006) Protein O-fucosyltransferase 2 adds O-fucose to thrombospondin type 1 repeats. J. Biol. Chem. 281, 9393–9399 [PubMed] [Google Scholar]
6. Luo Y., Nita-Lazar A., and Haltiwanger R. S. (2006) Two distinct pathways for O-fucosylation of epidermal growth factor-like or thrombospondin type 1 repeats. J. Biol. Chem. 281, 9385–9392 [PubMed] [Google Scholar]
7. Luo Y., and Haltiwanger R. S. (2005) O-Fucosylation of Notch occurs in the endoplasmic reticulum. J. Biol. Chem. 280, 11289–11294 [PubMed] [Google Scholar]
8. Acar M., Jafar-Nejad H., Takeuchi H., Rajan A., Ibrani D., Rana N. A., Pan H., Haltiwanger R. S., and Bellen H. J. (2008) Rumi is a CAP10 domain glycosyltransferase that modifies Notch and is required for Notch signaling. Cell 132, 247–258 [PMC free article] [PubMed] [Google Scholar]
9. Campbell I. D., and Bork P. (1993) Epidermal growth factor-like modules. Curr. Opin. Struct. Biol. 3, 385–392 [Google Scholar]
10. Haltom A. R., and Jafar-Nejad H. (2015) The multiple roles of epidermal growth factor repeat O-glycans in animal development. Glycobiology 25, 1027–1042 [PMC free article] [PubMed] [Google Scholar]
11. Takeuchi H., and Haltiwanger R. S. (2014) Significance of glycosylation in Notch signaling. Biochem. Biophys. Res. Commun. 453, 235–242 [PMC free article] [PubMed] [Google Scholar]
12. Harvey B. M., Rana N. A., Moss H., Leonardi J., Jafar-Nejad H., and Haltiwanger R. S. (2016) Mapping sites of O-glycosylation and fringe elongation on Drosophila Notch. J. Biol. Chem. 291, 16348–16360 [PMC free article] [PubMed] [Google Scholar]
13. Kakuda S., and Haltiwanger R. S. (2017) Deciphering the fringe-mediated Notch code: identification of activating and inhibiting sites allowing discrimination between ligands. Dev. Cell 40, 193–201 [PMC free article] [PubMed] [Google Scholar]
14. Rana N. A., Nita-Lazar A., Takeuchi H., Kakuda S., Luther K. B., and Haltiwanger R. S. (2011) O-Glucose trisaccharide is present at high but variable stoichiometry at multiple sites on mouse Notch1. J. Biol. Chem. 286, 31623–31637 [PMC free article] [PubMed] [Google Scholar]
15. Shi S., and Stanley P. (2003) Protein O-fucosyltransferase I is an essential component of Notch signaling pathways. Proc. Natl. Acad. Sci. U.S.A. 100, 5234–5239 [PMC free article] [PubMed] [Google Scholar]
16. Fernandez-Valdivia R., Takeuchi H., Samarghandi A., Lopez M., Leonardi J., Haltiwanger R. S., and Jafar-Nejad H. (2011) Regulation of mammalian Notch signaling and embryonic development by the protein O-glucosyltransferase Rumi. Development 138, 1925–1934 [PMC free article] [PubMed] [Google Scholar]
17. Sawaguchi S., Varshney S., Ogawa M., Sakaidani Y., Yagi H., Takeshita K., Murohara T., Kato K., Sundaram S., Stanley P., and Okajima T. (2017) O-GlcNAc on NOTCH1 EGF repeats regulates ligand-induced Notch signaling and vascular development in mammals. eLife 6, e24419. [PMC free article] [PubMed] [Google Scholar]
18. Li M., Cheng R., Liang J., Yan H., Zhang H., Yang L., Li C., Jiao Q., Lu Z., He J., Ji J., Shen Z., Li C., Hao F., Yu H., and Yao Z. (2013) Mutations in POFUT1, encoding protein O-fucosyltransferase 1, cause generalized Dowling-Degos disease. Am. J. Hum. Genet. 92, 895–903 [PMC free article] [PubMed] [Google Scholar]
19. Basmanav F. B., Oprisoreanu A. M., Pasternack S. M., Thiele H., Fritz G., Wenzel J., Grösser L., Wehner M., Wolf S., Fagerberg C., Bygum A., Altmüller J., Rütten A., Parmentier L., El Shabrawi-Caelen L., et al. (2014) Mutations in POGLUT1, encoding protein O-glucosyltransferase 1, cause autosomal-dominant Dowling-Degos disease. Am. J. Hum. Genet. 94, 135–143 [PMC free article] [PubMed] [Google Scholar]
20. Servián-Morilla E., Takeuchi H., Lee T. V., Clarimon J., Mavillard F., Area-Gómez E., Rivas E., Nieto-González J. L., Rivero M. C., Cabrera-Serrano M., Gómez-Sánchez L., Martínez-López J. A., Estrada B., Márquez C., Morgado Y., et al. (2016) A POGLUT1 mutation causes a muscular dystrophy with reduced Notch signaling and satellite cell loss. EMBO Mol. Med. 8, 1289–1309 [PMC free article] [PubMed] [Google Scholar]
21. Sethi M. K., Buettner F. F., Ashikov A., Krylov V. B., Takeuchi H., Nifantiev N. E., Haltiwanger R. S., Gerardy-Schahn R., and Bakker H. (2012) Molecular cloning of a xylosyltransferase that transfers the second xylose to O-glucosylated epidermal growth factor repeats of notch. J. Biol. Chem. 287, 2739–2748 [PMC free article] [PubMed] [Google Scholar]
22. Sethi M. K., Buettner F. F., Krylov V. B., Takeuchi H., Nifantiev N. E., Haltiwanger R. S., Gerardy-Schahn R., and Bakker H. (2010) Identification of glycosyltransferase 8 family members as xylosyltransferases acting on O-glucosylated notch epidermal growth factor repeats. J. Biol. Chem. 285, 1582–1586 [PMC free article] [PubMed] [Google Scholar]
23. Rampal R., Li A. S., Moloney D. J., Georgiou S. A., Luther K. B., Nita-Lazar A., and Haltiwanger R. S. (2005) Lunatic fringe, manic fringe, and radical fringe recognize similar specificity determinants in O-fucosylated epidermal growth factor-like repeats. J. Biol. Chem. 280, 42454–42463 [PubMed] [Google Scholar]
24. 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., and Vogt T. F. (2000) Fringe is a glycosyltransferase that modifies Notch. Nature 406, 369–375 [PubMed] [Google Scholar]
25. Lee T. V., Sethi M. K., Leonardi J., Rana N. A., Buettner F. F., Haltiwanger R. S., Bakker H., and Jafar-Nejad H. (2013) Negative regulation of notch signaling by xylose. PLoS Genet. 9, e1003547. [PMC free article] [PubMed] [Google Scholar]
26. Lee T. V., Pandey A., and Jafar-Nejad H. (2017) Xylosylation of the Notch receptor preserves the balance between its activation by trans-Δ and inhibition by cis-ligands in Drosophila. PLoS Genet. 13, e1006723. [PMC free article] [PubMed] [Google Scholar]
27. Sakaidani Y., Nomura T., Matsuura A., Ito M., Suzuki E., Murakami K., Nadano D., Matsuda T., Furukawa K., and Okajima T. (2011) O-Linked N-acetylglucosamine on extracellular protein domains mediates epithelial cell–matrix interactions. Nat. Commun. 2, 583. [PubMed] [Google Scholar]
28. 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 [PubMed] [Google Scholar]
29. Stahl M., Uemura K., Ge C., Shi S., Tashima Y., and Stanley P. (2008) Roles of Pofut1 and O-fucose in mammalian Notch signaling. J. Biol. Chem. 283, 13638–13651 [PMC free article] [PubMed] [Google Scholar]
30. Okamura Y., and Saga Y. (2008) Pofut1 is required for the proper localization of the Notch receptor during mouse development. Mech. Dev. 125, 663–673 [PubMed] [Google Scholar]
31. Ishio A., Sasamura T., Ayukawa T., Kuroda J., Ishikawa H. O., Aoyama N., Matsumoto K., Gushiken T., Okajima T., Yamakawa T., and Matsuno K. (2015) O-Fucose monosaccharide of Drosophila Notch has a temperature-sensitive function and cooperates with O-glucose glycan in Notch transport and Notch signaling activation. J. Biol. Chem. 290, 505–519 [PMC free article] [PubMed] [Google Scholar]
32. Harris R. J., and Spellman M. W. (1993) O-Linked fucose and other post-translational modifications unique to EGF modules. Glycobiology 3, 219–224 [PubMed] [Google Scholar]
33. Takeuchi H., Kantharia J., Sethi M. K., Bakker H., and Haltiwanger R. S. (2012) Site-specific O-glucosylation of the epidermal growth factor-like (EGF) repeats of Notch: efficiency of glycosylation is affected by proper folding and amino acid sequence of individual EGF repeats. J. Biol. Chem. 287, 33934–33944 [PMC free article] [PubMed] [Google Scholar]
34. Handford P. A., Baron M., Mayhew M., Willis A., Beesley T., Brownlee G. G., and Campbell I. D. (1990) The first EGF-like domain from human factor IX contains a high-affinity calcium binding site. EMBO J. 9, 475–480 [PMC free article] [PubMed] [Google Scholar]
35. Rao Z., Handford P., Mayhew M., Knott V., Brownlee G. G., and Stuart D. (1995) The structure of a Ca2+-binding epidermal growth factor-like domain: its role in protein–protein interactions. Cell 82, 131–141 [PubMed] [Google Scholar]
36. Yu H., Takeuchi H., Takeuchi M., Liu Q., Kantharia J., Haltiwanger R. S., and Li H. (2016) Structural analysis of Notch-regulating Rumi reveals basis for pathogenic mutations. Nat. Chem. Biol. 12, 735–740 [PMC free article] [PubMed] [Google Scholar]
37. Luca V. C., Jude K. M., Pierce N. W., Nachury M. V., Fischer S., and Garcia K. C. (2015) Structural biology. Structural basis for Notch1 engagement of Delta-like 4. Science 347, 847–853 [PMC free article] [PubMed] [Google Scholar]
38. Luca V. C., Kim B. C., Ge C., Kakuda S., Wu D., Roein-Peikar M., Haltiwanger R. S., Zhu C., Ha T., and Garcia K. C. (2017) Notch-Jagged complex structure implicates a catch bond in tuning ligand sensitivity. Science 355, 1320–1324 [PMC free article] [PubMed] [Google Scholar]
39. Hayakawa S., Koide R., Hinou H., and Nishimura S. (2016) Synthetic human NOTCH1 EGF modules unraveled molecular mechanisms for the structural and functional roles of calcium ions and O-glycans in the ligand-binding region. Biochemistry 55, 776–787 [PubMed] [Google Scholar]
40. Taylor P., Takeuchi H., Sheppard D., Chillakuri C., Lea S. M., Haltiwanger R. S., and Handford P. A. (2014) Fringe-mediated extension of O-linked fucose in the ligand-binding region of Notch1 increases binding to mammalian Notch ligands. Proc. Natl. Acad. Sci. U.S.A. 111, 7290–7295 [PMC free article] [PubMed] [Google Scholar]
41. Matsumoto K., Ayukawa T., Ishio A., Sasamura T., Yamakawa T., and Matsuno K. (2016) Dual roles of O-glucose glycans redundant with monosaccharide O-fucose on Notch in Notch trafficking. J. Biol. Chem. 291, 13743–13752 [PMC free article] [PubMed] [Google Scholar]
42. Gordon W. R., Zimmerman B., He L., Miles L. J., Huang J., Tiyanont K., McArthur D. G., Aster J. C., Perrimon N., Loparo J. J., and Blacklow S. C. (2015) Mechanical allostery: evidence for a force requirement in the proteolytic activation of Notch. Dev. Cell 33, 729–736 [PMC free article] [PubMed] [Google Scholar]
43. Berger R. P., Sun Y. H., Kulik M., Lee J. K., Nairn A. V., Moremen K. W., Pierce M., and Dalton S. (2016) ST8SIA4-dependent polysialylation is part of a developmental program required for germ layer formation from human pluripotent stem cells. Stem Cells 34, 1742–1752 [PMC free article] [PubMed] [Google Scholar]
44. 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 [PubMed] [Google Scholar]
45. Takeuchi H., Fernández-Valdivia R. C., Caswell D. S., Nita-Lazar A., Rana N. A., Garner T. P., Weldeghiorghis T. K., Macnaughtan M. A., Jafar-Nejad H., and Haltiwanger R. S. (2011) Rumi functions as both a protein O-glucosyltransferase and a protein O-xylosyltransferase. Proc. Natl. Acad. Sci. U.S.A. 108, 16600–16605 [PMC free article] [PubMed] [Google Scholar]
46. Kabsch W. (2010) XDS. Acta Crystallogr. D Biol. Crystallogr. 66, 125–132 [PMC free article] [PubMed] [Google Scholar]
47. Battye T. G., Kontogiannis L., Johnson O., Powell H. R., and Leslie A. G. (2011) iMOSFLM: a new graphical interface for diffraction-image processing with MOSFLM. Acta Crystallogr. D Biol. Crystallogr. 67, 271–281 [PMC free article] [PubMed] [Google Scholar]
48. Vagin A., and Teplyakov A. (2010) Molecular replacement with MOLREP. Acta Crystallogr. D Biol. Crystallogr. 66, 22–25 [PubMed] [Google Scholar]
49. Pannu N. S., Murshudov G. N., Dodson E. J., and Read R. J. (1998) Incorporation of prior phase information strengthens maximum-likelihood structure refinement. Acta Crystallogr. D Biol. Crystallogr. 54, 1285–1294 [PubMed] [Google Scholar]
50. Emsley P., Lohkamp B., Scott W. G., and Cowtan K. (2010) Features and development of Coot. Acta Crystallogr. D Biol. Crystallogr. 66, 486–501 [PMC free article] [PubMed] [Google Scholar]
51. Winn M. D., Ballard C. C., Cowtan K. D., Dodson E. J., Emsley P., Evans P. R., Keegan R. M., Krissinel E. B., Leslie A. G., McCoy A., McNicholas S. J., Murshudov G. N., Pannu N. S., Potterton E. A., Powell H. R., et al. (2011) Overview of the CCP4 suite and current developments. Acta Crystallogr. D Biol. Crystallogr. 67, 235–242 [PMC free article] [PubMed] [Google Scholar]
Articles from The Journal of Biological Chemistry are provided here courtesy of American Society for Biochemistry and Molecular Biology