|
Figure 1.
Rab11a binds to the C terminus of rhodopsin. A, Affinity chromatography followed by SDS-PAGE identified four small proteins binding to the C terminus of rhodopsin. Three of the four proteins eluted by the C-terminal peptide were identified by mass spectroscopy: ant2, rab11a, and PDE6δ. B, Western blot confirms that rab11a is pulled down by the C terminus of rhodopsin. C, Immunohistochemistry of rab11a in mouse retina shows a punctate distribution in the inner segment with some faint punctate staining in the outer segment. Green, Acetylated α-tubulin; red, rab11a; IB, immunoblot. Scale bar, 10 μm.
|
|
Figure 2.
Rab11a binding is dependent on the integrity of the C terminus of rhodopsin. Pull-down experiments using a rab11a antibody show that wild-type rhodopsin can be pulled down by rab11a from retinal extracts but not Q344X or rhodopsin-EGFP. I, Input; C, control IgG antibody wash; Ab, rab11 polyclonal antibody eluate; IP, immunoprecipitation; IB, immunoblot.
|
|
Figure 3.
Proximity ligation assay reveals rhodopsin: rab11a interactions in the inner and outer segments of mouse rods. PLA was performed using rhodopsin (B630N) and rab11a (71-5300, Life Technologies) antibodies. The fluorescent signal was quantified as a percentage of the total area of the inner and outer segments. There was a significant difference between wild-type rhodopsin and rhodopsinQ344X, as determined by Student's unpaired t test (p = 0.025). *p < 0.05. Red, rhodopsin: rab11a PLA; blue, wheat germ agglutinin (WGA). Scale bar, 10 μm.
|
|
Figure 4.
Rab11a binds rhodopsin directly and independently of nucleotide binding status. GST fusions of rab11a and mutants were assayed for their ability to pull down purified native rhodopsin. All rab11a mutants tested were bound to purified rhodopsin. Across repeated experiments, no rab11a mutant had an increased or decreased affinity for rhodopsin relative to wild-type rab11a. A, Western blot of purified rhodopsin pull-down experiments. B, Coomassie gel of purified GSTârab11a fusions. IB, immunoblot.
|
|
Figure 5.
Subcellular localization of EGFP-rab11a and mutants in transgenic X. laevis frog photoreceptors. A, Representative images of EGFP-tagged rab11a variants in X. laevis photoreceptors. EGFP-rab11a is present diffusely in the outer segment, as is the S25N mutant. The Q70L and N124I variants primarily localize in the inner segment. #, Putative recycling endosome localization; *, Golgi cells. Scale bar, 10 μm. B, Outer segment distributions of the rab11a constructs and soluble EGFP. Images were taken with higher-intensity settings to highlight the axoneme. EGFP, rab11aQ70L, and rab11aN124I are all present in an axonemal distribution. C, Rhodopsin staining among tadpoles expressing EGFP or EGFP-tagged rab11a constructs. Arrowheads indicate staining in the inner segment. D, The relative fluorescence of each mutant for the OS, IS, nucleus, and synapse was compared with that of EGFP-rab11a (n = 5â9 per mutant). The OS fluorescent signal was significantly increased for the S25N mutant and significantly decreased for the Q70L and N124I mutants (one-way ANOVA: F(3,22) = 18.34, p < 0.001; Tukey's HSD test: for S25N, p = 0.023; for Q70L, p = 0.018; for N124I, p = 0.026). The IS fluorescent signal was significantly increased for the Q70L mutant relative to rab11a (one-way ANOVA F(3,22) = 12.59, p < 0.001; Tukey's HSD test, p < 0.001). The nuclear fluorescent signal was significantly increased for the Q70L and N124I mutants relative to rab11a (F(3,22) = 21.40, p < 0.001; Tukey's HSD test: for Q70L, p = 0.04; for N124I, p < 0.001). No significant differences in synapse fluorescent signal were observed (F(3,22) = 3.20, p = 0.043; but for all Tukey's HSD test comparisons, p > 0.05). *p < 0.05, **p < 0.01, ***p < 0.001. Error bars represent the SEM. E, Dot blots showing the relative content of rab11a in mouse ROS preparation versus total retinal extract (RE). Green, EGFP; red, rhodopsin; blue, DRAQ5.
|
|
Figure 6.
Quantification of TUNEL staining and OS length among the rab11a constructs. EGFP-rab11a-expressing tadpoles and nontrangenic controls (data not shown) were stained for TUNEL, to detect dying cells, and with WGA (data not shown), to measure outer segment length. A, Representative images of EGFP-rab11a tadpole sections stained for TUNEL. B, EGFP-rab11aN124I-expressing tadpoles had a significant increase in TUNEL-positive cells per section (n = 5 per group; one-way ANOVA: F(4,20) = 6.383, p = 0.002; Tukey's HSD test: for N124I compared with nontransgenic animals, p = 0.003; for all other comparisons to nontransgenic animals, p > 0.05). These animals also had significantly decreased OS length relative to nontransgenic controls (n = 6â8 per group; one-way ANOVA: F(4,29) = 6.01, p = 0.001; Tukey's HSD test: for N124I, p = 0.035; for all other comparisons to nontransgenic animals, p > 0.05). *p < 0.05, **p < 0.01. C, TUNEL-positive processes (arrow) projecting from the ONL into the inner retina were observed in EGFP-rab11aN124I-expressing tadpoles. Although TUNEL-positive processes from inner retinal cells could be observed in all transgenic animals and nontransgenic controls, the only TUNEL-positive processes extending from nuclei in the ONL were found in those animals expressing EGFP-rab11aN124I. Green, EGFP; red, TUNEL; blue, DRAQ5. Error bars represent the SEM. Arrows depict regions in the inner segment putatively containing Golgi apparatus. Scale bar, 10 μm.
|
|
Figure 7.
Expression of hairpin against rab11a leads to ectopic process formation, reduction in outer segment length, and increased photoreceptor death. A, Representative images of tadpoles expressing various constructs. Yellow, mVenus; green, EGFP-rab11a; red, TUNEL; blue, DRAQ5. B, Representative dot blots from whole tadpole eye showing rab11a expression changes with hairpin expression. Expression was normalized to tubulin (data not shown). Hairpin-expressing tadpoles had a significant 35.5% reduction in total eye rab11a (n = 5â8 per group; one-way ANOVA, F(3,22) = 4.038, p = 0.02; Tukey' HSD test for nontransgenic versus hairpin, p = 0.038. C, Quantification of TUNEL-positive cells and OS length. Rods expressing a hairpin against firefly luciferase had normal morphology. Rods expressing a hairpin against rab11a had shorter outer segments and significantly increased numbers of TUNEL-positive cells per section. Coexpression of EGFP-rab11a resistant to the hairpin rescued the effects. TUNEL-positive cells (n = 8â14 per group; one-way ANOVA, F(3,39) = 6.031, p = 0.002; Tukey's HSD test: for rab11a hairpin versus nontransgenic animals, p = 0.001; for rab11a hairpin versus luciferase hairpin, p = 0.009; for rab11a hairpin versus rescue, p = 0.048; all other comparisons, p > 0.05). OS length (n = 6â14 per group): one-way ANOVA: F(3,34) = 6.858, p = 0.001; Tukey's HSD test: for rab11a hairpin versus nontransgenic animals, p = 0.006; for rab11a hairpin versus luciferase hairpin, p = 0.017; for rab11a hairpin versus rescue, p = 0.002; all other comparisons, p > 0.05. D, TUNEL-positive processes (arrow) projecting from the ONL into the inner retina were observed in rab11a hairpin-expressing tadpoles. Yellow, mVenus. E, Rhodopsin staining of tadpoles expressing hairpin constructs. Arrowheads indicate inner segment or nuclear layer staining of rhodopsin. Yellow, mVenus; red, rhodopsin; blue, DRAQ5. Scale bars, 10 μm. *p < 0.05, **p < 0.01, ***p < 0.001.
|
|
Figure 8.
Schematic representation of the rab11a recycling endosome in rod photoreceptors. Rab11a in the GDP-bound form associates with rhodopsin in the IS near the Golgi and also in the outer segment disks. Rab11a in the GTP-bound form is associated with rhodopsin around the centrosome and extending up the axoneme. The rab11a that remains associated with disks is in the GDP-bound form.
|