Multiple O-glucosylation sites on Notch function as a buffer against temperature-dependent loss of signaling

Abstract

Mutations in Drosophila rumi result in a temperature-sensitive loss of Notch signaling. Rumi is a protein O-glucosyltransferase that adds glucose to EGF repeats with a C-X-S-X-P-C consensus sequence. Eighteen of the 36 EGF repeats in the Drosophila Notch receptor contain the consensus O-glucosylation motif. However, the contribution of individual O-glucose residues on Notch to the regulation of Notch signaling is not known. To address this issue, we carried out a mutational analysis of these glucosylation sites and determined their effects on Notch activity in vivo. Our results indicate that even though no single O-glucose mutation causes a significant decrease in Notch activity, all of the glucose residues on Notch contribute in additive and/or redundant fashions to maintain robust signaling, especially at higher temperatures. O-glucose motifs in and around the ligand-binding EGF repeats play a more important role than those in other EGF repeats of Notch. However, a single O-glucose mutation in EGF12 can be compensated by other O-glucose residues in neighboring EGF repeats. Moreover, timecourse cell aggregation experiments using a rumi null cell line indicate that a complete lack of Rumi does not affect Notch-Delta binding at high temperature. In addition, rumi fully suppresses the gain-of-function phenotype of a ligand-independent mutant form of Notch. Our data suggest that, at physiological levels of Notch, the combined effects of multiple O-glucose residues on this receptor allow productive S2 cleavage at high temperatures and thereby serve as a buffer against temperature-dependent loss of Notch signaling.

Fig. 1.

A 40 kb Notch genomic transgene (N^gt-wt) behaves similarly to an endogenous copy of Notch. (A) The Notch extracellular domain (Notch^ECD) with 36 EGF repeats, three LNR domains and heterodimerization (HD) and transmembrane (TM) regions. EGF repeats with an O-glucosylation motif are in blue. Corresponding exons encoding the Notch^ECD are marked beneath. (B) The Notch genomic region (exons in black) and the 40 kb Notch genomic transgene (N^gt-wt). (C-F) Gene dosage experiments at 30°C. (C) A wild-type adult female Drosophila wing. (D) The Confluens phenotype with the addition of one copy of the N^gt-wt transgene inserted at the VK22 docking site. Arrowheads mark extra wing veins. (E) A Notch^+/55e11 haploinsufficient female wing with thickened veins (arrowhead) and wing margin loss (arrow). This phenotype can be rescued with one copy of N^gt-wt (F). (G-I) A Notch^55e11/Y hemizygous male rescued with one copy of the N^gt-wt transgene inserted at VK22 at 30°C, showing normal legs (G, arrowhead marks the sex comb, arrows mark the leg joints), normal bristle pattern on the thorax (H) and normal wings (I).

Fig. 2.

The mutant Notch transgenes used in this study and a summary of results from the rescue studies. Blue or orange boxes represent EGF repeats with a wild-type O-glucosylation motif or with an S-to-A mutation, respectively. Each mutant Notch genomic transgene (N^gt-mut) was tested for its ability to rescue a null allele of Notch. The extent of microchaetae rescue by each transgene is indicated, from `++++' indicating a normal microchaetae pattern in the rescued flies to `– – – –' indicating an almost complete loss of microchaetae in the rescued flies. `Lethal' refers to animals that did not reach the pharate adult stage. rumi^–/– phenotypes are shown for comparison. When phenotypes at a particular temperature are variable, the less common phenotype is shown in parentheses.

Fig. 3.

O-glucose mutations in Notch cause bristle loss and leg abnormalities that recapitulate rumi mutations. (A-C′) Bristle and leg phenotypes of flies homozygous for the protein-null allele rumi^Δ26 (rumi^–/–). (D,D′) N^–/Y; N^gt-4_35/+ males show very severe bristle loss and leg joint abnormalities at 21-23°C. (E,E′) N^–/Y; N^gt-10_35/+ males show severe bristle loss and leg joint abnormalities at 21-23°C. (F,F′) N^–/Y; N^gt-10_20/+ males show some bristle loss and leg joint abnormalities at 21-23°C. Brackets mark leg shortening and/or leg joint abnormalities.

Fig. 4.

O-glucose residues on EGF10-15 of Notch show a combination of additive and redundant functions. (A,A′) At 30°C, N^–/Y; N^gt-10_15/+ male flies show a severe loss of bristles and shortened legs with severe joint defects. (B,B′) At 25°C, N^–/Y; N^gt-10_15/+ males show a mild loss of bristles and normal legs. (C,C′) At 21-23°C, N^gt-10_15 fully rescues the lethality and phenotypes of Notch null mutants. (D,D′) The leg defects and microchaetae loss of the N^–/Y; N^gt-10_15/+ males at 30°C are significantly improved by the addition of a second copy of the N^gt-10_15 transgene (compare with A,A′). (E-F′) N^–/Y; N^gt-10,13_15/+ (E,E′) and N^–/Y; N^gt-12/+ (F,F′) males show a normal bristle pattern and normal legs at 30°C. (G,G′) N^–/Y; N^{gt-10,12,14,15}/+ males show mild bristle loss and abnormal leg joints at 30°C. (H,H′) N^gt-12,14,15 fully rescues the phenotypes of a Notch null allele at 30°C. Brackets (A′,D′,G′) mark leg shortening and leg joint abnormalities.

Fig. 5.

O-glucose mutations decrease the activity of Notch in a temperature-dependent manner. Wings of adult male flies with wild-type Notch on the X chromosome and one or two copies of N^gt-wt (A-B′′) or two copies of N^gt-mut (C-D′′) transgenes inserted at the VK22 docking site on the second chromosome. (A-A′′) One copy of N^gt-wt results in a Confluens phenotype (extra vein, arrowheads) at 18-30°C. (B-B′′) Two copies of N^gt-wt cause an enhancement of the Confluens phenotype at 18-30°C. (C-C′′) At 18°C, two copies of N^gt-10_15 show a Confluens phenotype comparable to that caused by two copies of N^gt-wt (compare B and C). The amount of extra vein tissue gradually decreases as the temperature is increased from low (18°C) to high (30°C). (D-D′′) The extra vein phenotype caused by two copies of N^gt-10_35 at 18°C and 23°C is much milder than that caused by N^gt-wt and N^gt-10_15 (compare with B,B′ and C,C′). At 25°C (D′) and 30°C (D′′), almost no Confluens phenotype is observed.

Fig. 6.

O-glucose mutations do not alter the level of Notch expression or its traffic to the cell surface. Confocal images of wing imaginal discs of third instar Drosophila larvae raised at 30°C. Nuclear GFP (green) marks MARCM clones of the N^54l9 allele. (A-D′) N^54l9/54l9 MARCM clones are generated in the absence (A,A′) or presence (B-D′) of one copy of wild-type (B,B′) or mutant (C-D′) Notch transgenes and stained with anti-NICD antibody (red in A-D, gray in A′-D′). (A,A′) Lack of Notch staining in the clones indicates that N^54l9 is a protein-null allele. Notch proteins with mutations in EGF10-15 (C,C′) or EGF10-35 (D,D′) are expressed at levels, comparable to wild-type Notch (B,B′). (E-J) Surface expression of Notch. (F) A projection of three consecutive apical optical sections; (H,J) single apical optical sections; (E,G,I) basolateral optical sections from the same datasets as F,H,J, respectively.

Fig. 7.

Complete loss of Rumi does not affect the binding of endogenous Notch to Delta at the surface of neighboring cells. (A) Relative mRNA level of E(spl)m3 as measured by qRT-PCR in control and rumi^–/– Drosophila cells cultured at either room temperature (RT, 21-23°C) or 32°C. ^*P<0.04, ^**P<0.00035. Error bars indicate s.e.m. (B) Cell aggregation assays between rumi^–/– cells (cultured at room temperature or 32°C) and S2-Delta (S2-Dl) or S2 cells. Images of co-cultures at 0, 1, 3 and 6 minutes.

Fig. 8.

Loss of rumi suppresses the gain-of-function phenotype resulting from the overexpression of Notch^ΔLNR. (A-F′) Confocal images of wing imaginal discs of third instar Drosophila larvae raised at 30°C. Nuclear GFP (green) marks MARCM clones of wild type (A-B′), Delta Serrate (Dl^– Ser^–) double mutant (C-D′) or rumi^– mutant (E-F′) overexpressing Notch^ΔLNR. Discs are stained with anti-Wingless (Wg) antibody (red in A-F, gray in A′-F′). (A,C,E) Projections of multiple optical sections; (B,D,E) higher magnifications of A,C,E, as projections of several sections.