BBS7 can enter the nuclear compartment. (A) Immunofluorescence staining of NIH3T3 with a high concentration of the BBS7 antibody in untreated cells (upper panel) and in cells treated with LMB (bottom panel). In both cases, BBS7 staining is shown in green in the left column, the merge with γ- and acetylated tubulin and DAPI is shown in the right column. (B) Immunofluorescence staining of NIH3T3 cells showing the formation of nuclear aggregates that are evident at normal concentrations of BBS7 antibody after treatment with NEM (red). The nucleus is stained with DAPI (blue) and the γ- and acetylated tubulin are shown in green. (C) Cell fractionation showing the presence of endogenous BBS7 in both cytoplasmic and nuclear compartments; antibodies against γ-tubulin and histone H3 were used as controls for each fraction. (D) Epifluorescence images of NIH3T3 cells expressing Myc–BBS7 wild-type (upper panel) and Myc–BBS7 NES (bottom panel) using an anti-Myc antibody (green). The graph shows the percentage of cells transfected with Myc–BBS7 wild-type, Myc–BBS7 NES and the BBS7 mutants H323R and T211I, showing either a cytoplasmic distribution with nuclear exclusion (categorized as cytoplasmic) versus cytoplasmic and nuclear staining. Values shown means ± s.d. (E) Cell fractionation after transfection with Myc–EV, Myc–BBS7 wild-type and Myc–BBS7 NES showing that Myc–BBS7 NES is enriched in the nucleus; antibodies against γ-tubulin and histone H3 were used as controls.

BBS proteins interact with RNF2. (A) Cytoplasmic yeast two-hybrid screen using BBS7 as bait resulted in the identification of RNF2. True positives, MAFB-MAFB (positive control) and BBS7-RNF2 are shown growing at the restrictive temperature of 37°C only in galactose (condition in which the prey is expressed). Negative controls include MAFB-Lamin C, Coll-MAFB and both empty vectors, EV-EV. (B) Co-immunoprecipitation assays showing that BBS7 interacts with RNF2 in mammalian cells. Left: Cell lysates of HEK293 cells transiently expressing constructs (as indicated) were used to immunoprecipitate with an anti-Myc antibody. Both immunoprecipitates and cell lysates were analyzed by SDS-PAGE with an anti-HA antibody. Right: Endogenous BBS7 interacts with endogenous RNF2. Lysates of untreated HeLa cells were immunoprecipitated with an anti-BBS7 antibody and purified rabbit IgGs as control, and western-blotted with a mouse monoclonal anti-RNF2 antibody. (C) BBS1, BBS2, BBS4, BBS5, BBS6, BBS8 and BBS10, but not an unrelated protein (Myc–Control), can also interact with RNF2 in mammalian cells.

Depletion of BBS proteins affects proteasome efficiency and RNF2 protein levels. (A) Western blots of cell extracts after transient inhibition of BBS7 and RNF2 with pSUPER vectors (using the empty vector, pSUPER EV, as control) and blotted with an anti-RNF2 antibody (upper panel) and an anti-BBS7 (lower panel). An anti-GAPDH antibody was used as a loading control. (B) RNF2 average band density quantification of the western blot (considering the three biological replicates), including the effect of the inhibition of BBS1, BBS2 and BBS4 (blots not shown). (C) Immunofluorescence image of HEK293 proteasome sensor line cells transfected with empty vector (left) and pSUPER BBS7 (right), showing the increased GFP signal in BBS7-depleted cells. (D) Quantification of the GFP signal measured by flow cytometry and expressed as fold change. Values shown means ± s.d.

BBS7 depletion alters the expression of RNF2 target genes. (A,B) Relative expression levels of RNF2 and 11 previously identified RNF2 target genes analyzed by semi-quantitative (A) and Real Time RT-PCR (B) in BBS7-depleted cells (pSUPER BBS7) relative to control cells (pSUPER EV). GAPDH and BBS7 were amplified as controls. The relative expression of each gene is shown as the fold change and the error bars represent the s.d. of replicates. The cross-bars in the y-axis indicate changes in the scale used.

In vivo misregulation of RNF2 target genes in zebrafish BBS morphants. (A–E) Embryos injected with bbs7, bbs4 or rnf2 MOs and stained by in situ hybridization at 30 hours post-fertilization using probes for bcl11a, six1b, pou3f2, hoxb8a and lef1. (A) Embryos injected with bbs7 and bbs4 MOs had significantly less staining of bcl11a than uninjected control embryos whereas co-injection of bbs7/4 with rnf2 rescued and even expanded the field of expression. (B) six1b show an altered pattern of expression with staining in the diencephalon in bbs morphants. (C) pou3f2 staining showed lower levels of expression in bbs7 and bbs4 morphants, with rescue of expression in double morphants of bbs7/rnf2 and bbs4/rnf2. (D) hoxb8a expression in bbs7, bbs4 and double morphants with rnf2 showed no significant difference in expression. (E) lef1 expression is reduced in morphant embryos.