Figure 1.
AZI1 interacts with the BBSome through BBS4.
A) A silver stained gel from GFP IP sample of testis lysate. Lane 1 shows the wild-type (WT) precipitate and lane 2 is the precipitate of testis lysate from transgenic (TG) animals. The band labeled with an asterisk is the 120 KDa band identified as Azi1 by mass spectrometry. B) Western blot showing interaction of BBS4 with AZI1 and PCM1. FLAG and S tagged BBSome subunits were transfected in 293T cells and Co-IP was performed using FLAG agarose beads. The top blot shows efficient transfection of the BBS subunits detected by anti-FLAG antibody. The lower two blots were immunoblotted with AZI1 and PCM1 antibody, respectively. C) FLAG-BBS4 stable cell (293T) was co-immunoprecipitated with FLAG agarose beads. Input, supernatant and eluate were immunoblotted with, antibody against GAPDH to show that equal amounts were loaded, antibody against FLAG to show efficient precipitation of BBS4 by FLAG beads, and antibody against AZI1 to show amounts of AZI1 precipitated by BBS4 in our cell line. D) 293T cells stably expressing the FLAG tagged BBSome subunits BBS4, 5, 8, and 9 were co-IPed with FLAG agarose beads and immunoblotted with AZI1 antibody. Although all the subunits were able to precipitate AZI1, BBS4 showed the most efficient pull-down. E) Co-IP using AZI1 antibody was performed in 293T cells followed by Western blotting with different BBS antibody to show interaction of endogenous BBS proteins and AZI1.
Figure 2.
AZI1 co-localizes with BBS4 to the centriolar satellite.
A) AZI1 and PCM1 co-localize at the centrosome. IMCD3 cells were transfected with HA-AZI1 construct and stained with antibody against HA (green) and PCM1 (red). AZI1 and PCM1 co-localize at the centrosome, in the presence (row 1) as well as absence of cilia (row 2). γ-tubulin and acetylated α-tubulin staining were used to identify the basal body and cilia. B) BBS4 co-localizes with AZI1, which resembles its co-localization with PCM1. RPE-1 cells stably expressing GFP-BBS4 were used; anti-GFP antibody was used to detect BBS4 (green); anti-AZI1 (row 1) or anti-PCM1 (row 2) antibody is shown in red. C) Western blotting shows expression levels of BBS and other satellite proteins upon AZI1 knockdown. Three different siRNAs were used, and efficient and specific knockdown of AZI1 by all three siRNAs are shown. Antibody against various BBS and satellite proteins were used to show that loss of AZI1 does not cause significant differences in the expression of those proteins. GAPDH is used as a loading control. D) Centriolar satellite localization of AZI1 (red) is confirmed in RPE-1 cells. siRNA based depletion diminished AZI1 from centriolar satellites. Depletion of PCM1 also depletes AZI from satellites but AZI1 at core centriolar areas remain intact. Insets show enlarged view of the centrosomal region. Green is γ-tubulin and acetylated α-tubulin staining for basal body and cilia, respectively. Scale bar, 10 µm.
Figure 3.
BBS4 is part of a PCM1 dependent centriolar satellite complex.
293T cells were transfected with siRNA against CTRL, AZI1, and PCM1, and the cell lysates were separated by 10–40% sucrose gradient centrifugation and fractionation. Thirteen fractions, (from 7–19, 19 being the heaviest fraction) were run on an SDS-PAGE gel and analyzed by immunoblotting with antibodies against various BBS proteins, AZI1, and PCM1. Fractions marked red represent the peak for the BBSome. Fractions marked blue is the peak for centriolar satellite pool of BBS4, along with centrosomal proteins, AZI1 and PCM1.
Figure 4.
AZI1 knockdown reduces ciliogenesis but increases ciliary localization of the BBSome.
A) RPE-1 cells were transfected with siRNA against BBS4, PCM1, and AZI1. Approximately 500 cells per sample were counted. B) AZI1 depletion increases the ciliary localization of BBS9 compared to the control knockdown. Red staining represents BBS9, and cilia are stained with acetylated α-tubulin. Nuclei are stained blue with DAPI. C) Graph showing a significant increase in cells with ciliary BBS9 upon AZI1 knockdown by different siRNAs. D) BBS8 (red) is used as a BBSome marker to confirm increased ciliary localization of the BBSome upon AZI1 knockdown. Cilia (green) in the insets of figures B and D are slightly shifted to show ciliary localization of BBS proteins. E) The graph shows a significant increase in the number of ciliated cells with BBS8 or BBS9 upon AZI1 knockdown, and decrease in ciliary BBS8 or BBS9 localization upon AZI1 overexpression. Approximately 250 ciliated cells were counted in control as well as AZI1 knockdown culture of RPE-1 cells. All data are presented in mean +/− SEM. Significance is calculated using Student's t-test for C and E, and one way ANOVA for A. P<0.05 is considered significant for each analysis.
Figure 5.
AZI1 knockdown increases ciliary localization of BBS4.
A) RPE-1 cells expressing GFP-BBS4 were depleted of AZI1, and the number of cells with ciliary GFP (BBS4) was counted. BBS4 is stained with GFP (green), Acetylated α-tubulin was used to detect cilia (red). B) Graph showing significant increases in ciliary localization of BBS4 upon AZI1 knockdown. C) Depletion of AZI1 in BBS3 and BBS5 depleted cells restores ciliary BBSome localization. RPE-1 cells were transfected with siRNA as indicated, and BBS9 (red) localization was analyzed. Cilia (green) in the insets of figures A and C are slightly shifted to show ciliary localization of BBS proteins. D) Cilia containing BBS9 at different conditions were counted and presented graphically. All data are presented in mean +/− SEM. Significance is calculated using the Student's t-test for C and E, and one way ANOVA for A. P<0.05 is considered significant for each analysis.
Figure 6.
Azi1 (Cep131) knockdown causes defects in KV formation, melanosome trafficking and vision defects in zebrafish.
Embryos were untreated or injected with 125 µM Bbs4 MO or 150 µM cep131 MO as previously described [28], [29], [31], [33]. A) azi1, as well as bbs4 morpholino injection causes a mild to severe body curvature phenotype in the 48 hpf larva. Images indicate examples of normal, mild or severe phenotypes. B) Quantification of body curvature. More of the azi1 morphants have severe body curvature than bbs4 morphants. C) Micrographs show KVs from MO injected and uninjected 8–10 somite stage embryos. D) bbs4 MO caused KV abnormalities in 30% of embryos, and azi1 MO caused 35% abnormal KV compared with 12% in control embryos (P<0.001 Fisher's exact test). E) bbs4 MO and azi1 MO delayed melanosome retrograde transport to 4.04+/−0.43 min, and 3.03+/−0.12 min, respectively compared to the control MO, in which the complete transport occurs within 1.92+/−0.04 minutes. F) Vision assays were performed on the morphants by observing their response to dark/light cues as previously described [33], [35]. For each fish the vision response tested 5 times. For C and D data are presented mean +/− SEM, P<0.01 by one-way ANOVA with Tukey post-test. The number of embryos used is presented in the figure.
Figure 7.
A model showing centrosomal complex of BBS4, BBSome complex formation, and its ciliary localization.
A) 1. Partial BBSome complex (without BBS4) arrives near the centrosome, where BBS4 is part of a satellite complex. At the centrosome BBS4 is incorporated into the BBSome and a stable holo-BBSome complex is formed, which is trafficked to cilia while the rest of the satellite complex remain at the centrosome. A potential rate-limiting factor of BBSome entry into cilia is availability of the BBSome complex that can enter cilia. B) Loss of AZI1 weakens the interaction of BBS4 with the satellite complex, and more BBSome complex is available for entry into cilia.