A tumor suppressor role of the Bub3 spindle checkpoint protein after apoptosis inhibition

Abstract

Most solid tumors contain aneuploid cells, indicating that the mitotic checkpoint is permissive to the proliferation of chromosomally aberrant cells. However, mutated or altered expression of mitotic checkpoint genes accounts for a minor proportion of human tumors. We describe a Drosophila melanogaster tumorigenesis model derived from knocking down spindle assembly checkpoint (SAC) genes and preventing apoptosis in wing imaginal discs. Bub3-deficient tumors that were also deficient in apoptosis displayed neoplastic growth, chromosomal aneuploidy, and high proliferative potential after transplantation into adult flies. Inducing aneuploidy by knocking down CENP-E and preventing apoptosis does not induce tumorigenesis, indicating that aneuploidy is not sufficient for hyperplasia. In this system, the aneuploidy caused by a deficient SAC is not driving tumorigenesis because preventing Bub3 from binding to the kinetochore does not cause hyperproliferation. Our data suggest that Bub3 has a nonkinetochore-dependent function that is consistent with its role as a tumor suppressor.

Figure 1.

Knockdown of Bub3 promotes tumorigenesis in Drosophila wing discs. (A and B) GFP expression in the adult (A) and wing disc (B). (C and D) Bub3 RNAi and GFP expression. Adult thorax (C) showing empty sockets and loss of bristles (arrows) and wing disc (D). Note that the wing disc is smaller than a wt disc. (E) Apoptosis in wing discs expressing Bub3 RNAi in the dorsal compartment. The dotted line separates the dorsal (bottom) from the ventral (top) sides of the wing disc. (F and G) Coexpression of Bub3 RNAi, DIAP1, and GFP in the adult thorax (F) and wing disc (G). (H and I) Expression of Bub3 RNAi, Drosophila p53R155H, and GFP. Adult thorax (H) and wing disc (I) showing a similar phenotype to A and B. (J and K) Larvae expressing GFP (top) and expressing Bub3 RNAi, p35, and GFP (bottom). (L) Wing disc dissected out of an L3 larvae expressing Bub3 RNAi, p35, and GFP. Note the hyperplasia. Transgenes in this study were driven by the apterous promoter. Bars: (B) 70 µm; (D, E, G, and I) 40 µm; (J) 100 µm; (K) 150 µm; (L) 80 µm.

Figure 2.

Bub3 knockdown induces aneuploidy and hyperplasia. (A–C) Karyotypic analysis of the hyperplastic wing discs expressing Bub3 RNAi, p35, and GFP. (A) wt mitotic cell karyotype where chromosomes I, II, II, and IV as well as the X and Y are shown. (B) Mitotic cell karyotype displaying aneuploidy. (C) Somatic chromosome (Chr) representation of aneuploidies found in hyperplastic wing discs (42 aneuploidies from a total of 63 metaphasic cells counted from three wing discs). (D–I) FACS analysis of dissociated ventral cells (E and H) and dorsal cells (F and I) from wing discs expressing Bub3 RNAi and GFP (D–F) and from wing discs expressing Bub3 RNAi, p35, and GFP (G–I). Note the increased number of aneuploid cells in I when compared with H and the higher percentage of GFP cells in the hyperplastic discs (83%; G) when compared with wing discs expressing Bub3 RNAi (38%; D). The data shown are from a single representative experiment out of three repeats (D–I). (J) Hyperplastic discs were transplanted into adult female hosts to assess their proliferative potential. Discs expressing GFP, GFP and p35, GFP and Bub3 RNAi, GFP and Bub3 RNAi, and p35 were implanted; the number of implants and the percentage of discs that grew are shown. Note that the implants grew extensively in the Bub3 RNAi p35 discs invading the host abdomen. Bars: (A and B) 5 µm; (D) 40 µm; (G) 80 µm.

Figure 3.

Inducing CIN by knockdown of CENP-E does not drive tumorigenesis. (A and B) CENP-E knockdown in the adult thorax (A) and in the wing disc (B). Note the missing bristles in the thorax (arrows). (C) Apoptosis in CENP-E knocked down wing discs. Note that there are more mitotic cells in the dorsal than in the ventral domain. The dotted line separates the dorsal (bottom) from the ventral (top) sides of the wing disc. (D) Mitotic cells on a colchicine-treated wing disc with CENP-E depleted on the dorsal domain. (E) Immunohistochemistry with a CENP-E antibody revealing the CENP-E protein depletion in the dorsal domain of the wing disc. (F and G) Quantification of CENP-E levels in wing discs with CENP-E depleted on the dorsal domain (F) and in wt discs (G). Immunofluorescence intensity was assessed in cells from the dorsal or ventral side independently and shown in a scatter plot with the mean (large horizontal lines) and SD (small horizontal bars). AU, arbitrary unit. (H) Wing disc dissected out of an L3 larvae expressing CENP-E RNAi, p35, and GFP. Note the mild hyperplasia. (I–N) FACS analysis of dissociated ventral (J and M) and dorsal cells (K and N) from wing discs expressing CENP-E RNAi and GFP (I–K) and expressing CENP-E RNAi together with p35 and GFP (L–N). Note the aneuploid cells in K and N. Note that the proportion of GFP versus non-GFP cells is comparable between I (71%/26%) and L (75%/21%), indicating that both compartments grew at a similar rate. The data shown are from a single representative experiment out of three repeats (I–N). (O) Size measurements for discs expressing GFP; GFP and Bub3 RNAi; GFP, Bub3 RNAi, and p35; and GFP, CENP-E, and p35 at the end of the third instar larval stage. For comparison, the percentage of ventral over dorsal (% V/D) volume is also shown. Note that Bub3 RNAi p35 wing discs are larger than the wing discs expressing CENP-E RNAi and p35. (P) Wing discs expressing CENP-E RNAi or CENP-E RNAi with p35 were implanted into adult female hosts to assess their proliferative potential. Error bars represent SD in F, G, and O. Bars: (B–E and I) 40 µm; (H and L) 70 µm.

Figure 4.

Tumorigenesis is not driven by a loss of kinetochore-associated Bub3 but by depletion of the Bub3 cytoplasmic pool. (A and B) Expression of nsl1 RNAi in the adult thorax (A) and in the wing imaginal disc (B). Note the lack of bristles in A (arrows) and the morphology of the dorsal compartment of the wing disc (B). (C) Mitotic cells on a colchicine-treated wing imaginal disc with nsl1 depleted on the dorsal domain. (D) Apoptosis in nsl1 knocked down wing discs. Note more apoptotic cells in the dorsal than in the ventral domain. In C and D, the dotted lines separate the dorsal (bottom) from the ventral (top) sides of the wing disc. (E–X) SAC components failed to bind kinetochores of cells depleted of nsl1. (E–G) wt wing disc. nsl1 protein in dorsal (F) and ventral cells (G). (I–K, M–O, Q–S, and U–W) Knockdown of nsl1 in the dorsal compartment. nsl1 protein in dorsal (F and J) and ventral cells (G and K). (M–O) Bub3 protein in dorsal (N) and ventral cells (O). (Q–S) BubR1 protein in dorsal (R) and ventral cells (S). (U–W) Mad1 protein in dorsal (V) and ventral cells (W). Centromere identifier (CID) labels the centromeres. (H, L, P, T, and X) Quantification of kinetochores analyzed. Note that a high percentage of kinetochores in dorsal cells depleted of nsl1 failed to show localization of SAC proteins. (Y) Wing disc dissected out of an L3 larvae expressing nsl1 RNAi, p35, and GFP. Note that this disc is not hyperplastic. (Z) Wing disc simultaneously expressing nsl1 RNAi, Bub3 RNAi, p35, and GFP. Note that this disc is hyperplastic. (A′) Wing discs expressing nsl1 RNAi, nsl1 RNAi and p35, and nsl1, Bub3 RNAi, and p35 were implanted into adult female hosts to assess their proliferative potential. Note that only discs expressing nsl1, Bub3 RNAi, and p35 grew extensively and invaded the host abdomen. Bars: (B–D) 40 µm; (E, I, M, Q, and U) 10 µm; (F, G, J, K, N, O, R, S, V, and W) 5 µm; (Y) 70 µm; (Z) 80 µm.

Figure 5.

Knockdown of SAC genes BubR1 or Mad2 when apoptosis is inhibited drive tumorigenesis. (A and D) Knockdown of BubR1 (A) Mad2 (D) in the adult thorax. Note the missing bristles (arrows). (B and E) Apoptosis in wing discs expressing BubR1 RNAi (B) or Mad2 RNAi (E). Note more apoptosis in the dorsal than in the ventral domain. In B and E, the dotted lines separate the dorsal (bottom) from the ventral (top) sides of the wing disc. (C and F) Wing disc dissected out of an L3 larvae expressing BubR1 RNAi (C) or Mad2 RNAi (F) together with p35 and GFP in the dorsal domain of the wing disc. Note the hyperplasia. (G) Hyperplastic wing discs were transplanted into adult female hosts to assess their proliferative potential. Discs expressing GFP and BubR1 RNAi; GFP, BubR1 RNAi, and p35; GFP and Mad2 RNAi; and GFP, Mad2 RNAi, and p35 were implanted; the number of implants and the percentage of discs that grew are shown. The implanted BubR1 RNAi p35– and Mad2 RNAi p35–expressing discs grew extensively and invaded the host abdomen. Bars: (B and E) 40 µm; (C and F) 80 µm.