PMC

Associations of altered NRBP1 expression with human cancers. (A) Expression analysis of NRBP1 in human large-cell lung carcinoma, lung adenocarcinoma, T-cell prolymphocytic leukaemia, colorectal carcinoma, cutaneous follicular lymphoma, marginal zone B-cell lymphoma and diffuse large B-cell lymphoma compared to relevant normal tissues. *P<0.05, **P<0.01 and ***P<0.001. (B) Survival analysis of patients with lung adenocarcinoma expressing either high (green) or low (blue) NRBP1 expression levels.

Expression analysis of conditional Nrbp1 mice. (A) qRT–PCR analysis of the WNT target genes, Cd44, c-Myc and Ccnd1 in the intestine of cKO and C mice at 4–5 days post tamoxifen administration. (B) Immunohistochemical (i, ii) and in situ analysis (iii, iv) of small intestines from C and cKO mice at 7–9 days post tamoxifen treatment indicating loss of Nrbp1 causes an increase in (i) c-Myc ( × 400 magnification), (ii) Sox9 ( × 400 magnification), (iii) Mmp7 ( × 400 magnification) and (iv) Lgr5 ( × 200 magnification). (C) Immunohistochemical analysis of small intestines from C and cKO mice at 7–9 days post tamoxifen treatment show increased expression of (i) pErk ( × 200 magnification), but no obvious increase in (ii) nuclear Ctnnb1 signal ( × 400 magnification). *P<0.05, **P<0.01.

Nrbp1 is a tumour suppressor in vivo. (A) Survival analysis of cKO (Nrbp^floxRosa^CreERT2/+) and control mice (Nrbp^flox/+Rosa^CreERT2/+ and Nrbp^flox/floxRosa^+/+) after treatment with 1 mg tamoxifen per day for 4 days. Animals became moribund and were culled as a result of either an ‘early Nrbp1 loss phenotype’ (animals <10 days post treatment) or when they developed macro/microscopically visible tumour(s) (animals >10 days post treatment). (B) Table provides a summary of the tumour types found in the mice (some animals had more than one type of tumour). (C) H&E staining of tissues from cKO mice showing (i) widespread lymphoma in the liver ( × 100 magnification), (ii) invasive colorectal adenocarcinoma ( × 250 magnification) and (iii) high-grade angiosarcoma in the spleen ( × 250 magnification). (D) ISH analysis of Nrbp1 in a colorectal lesion from an Nrbp1^floxRosa^CreERT2/+ mouse shows (i) a tumour area left of the black line, and normal colonic tissue right of the black line ( × 25 magnification), with the lower box enlarged in (ii) and the upper box enlarged in (iii), showing reduced Nrbp1 expression levels in tumour tissue ( × 200 magnification). Immunohistochemical analysis of an invasive colorectal adenocarcinoma from an Nrbp1^floxRosa^CreERT2/+ mouse showing significant strongly positive staining with (iv) c-Myc ( × 400 magnification) and (vi) pErk ( × 200 magnification) antibodies, but (v) no increased nuclear Ctnnb1 signal ( × 400 magnification).

NRBP1 interacts with components of the Elongin BC E3 ubiquitin ligase complex. (A) Immunoprecipitation with anti-Flag antibody of mouse ES cells carrying either Flag-tagged NRBP1 cDNA expression plasmid (N) or Flag-tagged Phactr2 cDNA expression plasmid (P) as a control, followed by western blotting with one of two different anti-Sall4 antibodies (Ab1 and Ab2), showed coprecipitation of both Sall4 isoforms with Nrbp1 and not the control. In mouse ES cells, Sall4 has been shown to run at 85 and 140 kDa. (B) Immunoprecipitation of CUL5 and other members of the ubiquitination machinery with NRBP1 from HEK293 cells. (C) Western blot analysis with anti-Tsc22d2 antibody, anti-Sall4 and anti-β-Actin control antibody of intestinal tissue harvested at day 5 from cKO (Nrbp1^floxRosa^CreERT2/+) or C (control; Nrbp1^floxRosa^+/+) mice treated with 1 mg tamoxifen for 4 days. In somatic cells, Sall4 has been shown to run at 66 and 113 kDa. (D) Immunohistochemical analysis of Sall4 in the intestinal tissue from C (control; Nrbp1^floxRosa^+/+) and cKO (Nrbp1^floxRosa^CreERT2/+) mice at 4–5 days post tamoxifen treatment showing elevated levels of Sall4. (E) Transcriptional activation of TCF/LEF measured using the dual luciferase assay (TOP/FOP) in HCT-116 colon cancer cells transfected with a Sall4 cDNA plasmid (left panel) or an NRBP1 shRNA construct (middle panel). N refers to an empty vector control. Transcriptional activation of TCF/LEF was also measured by siRNA-mediated knockdown of NRBP1 in SW180 WNT reporter cells (right panel). Knockdown of CTNNB1 was used as a control. *P<0.05, **P<0.01, ***P<0.001.

C. elegans let-60 (n1046gf) RNAi kinase screen and mammalian in vitro transformation analysis. (A) Muv phenotype of let-60 (n1046gf) worms after RNAi-mediated knockdown of 656 worm kinase genes. The percentage of animals with a multivulva phenotype is shown. (B) We took the top 2% of genes and performed a second round of screening to identify four robust enhancers of the Muv phenotype. (C) RNAi against H37N21.1 causes excessive LET-60 signalling in vulval cells. egl-17::cfp transgenic worms showing ectopic expression of CFP in P5.p descendants after one cell division (P5.px), which indicates excessive LET-60 signalling. Also shown is the normal CFP expression in the P6.p descendants (P6.px). We observed ectopic expression at a rate of 18% (n=23). Top image is light field and bottom image is CFP expression. (D) Histogram of colony numbers from transformants derived from NIH3T3 cells cotransfected with an Nrbp1 shRNA vector (1–4) and pBabe-Ras^V12 vector. pBabe-Ras^V12 vector alone (R) and empty pBabe vector (N) were used as controls. (E) Western blot analysis of Nrbp1 knockdown in NIH3T3 cells. (F) Histogram of transformants after BJ-ET-st p53^kd, p16^kd cells were transfected with NRBP1 human shRNA vectors (1–6) or pRS empty vector control (N). (G) BJ-ET-st p53^kd, p16^kd transformed colonies transfected with human NRBP1 shRNA vector 3 were picked, cultured and whole-cell lysates collected for use in pull-down assays using GST- Raf1-BSD. Lysates were separated and visualised using a pan-RAS antibody (L—whole cell lysate, GTP—positive control, GDP—negative control, N—untransfected control cell pull-down, 3—shRNA trasformant pull-down). Data are represented as mean±s.d.; *P<0.05, **P<0.01. All shRNA hairpin sequences are available in the .

Generation and characterisation of Nrbp1 knockout mice. (A) Schematic representation of the Nrbp1 locus, along with targeted, conditional and knockout Nrbp1 alleles. The PGK–neomycin selection cassette (neo) is flanked by Frt sites (striped triangles). Open triangles indicate LoxP sites for conditional deletion of exons 5–11. (B) Confirmation of correct targeting was performed by Southern blot analysis of BamHI-digested DNA; The wildtype band for both 5′- and 3′-probes was 11.6 kb and the targeted band was 3.8 and 4.3 kb, respectively. (C) Subsequent genotyping was performed by PCR, using either primers ‘b/c’ to detect the wild-type and conditional allele, or primers ‘a/c’ to detect the knock out allele. (D) Survival curve of Nrbp1^flox, Rosa^CreERT2/+ (cKO) mice and controls after intraperitoneal dosing with 3 mg tamoxifen per day for 3 days. (E) H&E staining of the SI (small intestine) and LI (large intestine) from C (control; Nrbp1^+/−Rosa^CreERT2/+ or Nrbp1^flox/+Rosa^CreERT2/+) and cKO (Nrbp1^flox/floxRosa^CreERT2/+ or Nrbp1^flox/−Rosa^CreERT2/+) mice at 5–7 days post tamoxifen treatment showing crypt elongation by primitive looking cells ( × 200 magnification). Inset: cKO intestines show evidence of crypt fission. (F) cKO mouse intestines had more crypts per field of view than C intestines at 4–5 days post dosing. (G) (i) ISH analysis using probe specific to Nrbp1 in the SI (small intestine) and LI (large intestine) of C mice (control; × 400 magnification) and the SI of cKO mice ( × 200 magnification) at 5 days post tamoxifen treatment, showing reduction of Nrbp1 expression levels in cKO tissue. (G) (ii) Western blot analysis of intestinal tissue harvested from C and cKO mice at 5 days post treatment with 1 mg tamoxifen for 4 days shows a reduction of Nrbp1 protein levels. β-Actin levels used as a control. (H) Histological and immunohistological analysis showing cKO mouse intestines had fewer goblet cells (as determined by Alcian blue staining) and an altered distribution of Paneth cells (as determined by lysozyme staining) compared to C mice at 4–5 days post dosing. *P<0.05. (I) cKO mice had more proliferating cells (as determined by Ki67 immunopositivity) than C intestines at 4–5 days post dosing. Data are represented as mean±s.d.; *P<0.05. (J) Histological and immunohistochemical analysis of the small intestines of 7–8-week-old Nrbp1^−/cLgr5-EGFP-IRES-CreER^T2+ and Nrbp1^−/+Lgr5-EGFP-IRES-CreER^T2+ control mice at 4–5 post dosing with 1 mg tamoxifen for 3 days. Data are represented as mean +/− s.d. (n=6 per genotype). Stem cell-specific deletion of Nrbp1 resulted in abnormal paneth cell localisation and granulisation, and abnormal goblet cell production ( × 200 magnification; haematoxylin and eosin, H&E; anti-lysozyme antibody, LY; Alcian blue, AB). **P<0.01, ***P<0.001.