Whole chromosome instability caused by Bub1 insufficiency drives tumorigenesis through tumor suppressor gene loss of heterozygosity

Abstract

Genetic alterations that promote chromosome missegregation have been proposed to drive tumorigenesis through loss of whole chromosomes containing key tumor suppressor genes. To test this unproven idea, we bred Bub1 mutant mice that inaccurately segregate their chromosomes onto p53(+/-), Apc(Min/+), Rb(+/-), or Pten(+/-) backgrounds. Bub1 insufficiency predisposed p53(+/-) mice to thymic lymphomas and Apc(Min/+) mice to colonic tumors. These tumors consistently lacked the nonmutated tumor suppressor allele but had gained a copy of the mutant allele. In contrast, Bub1 insufficiency had no impact on tumorigenesis in Rb(+/-) mice and inhibited prostatic intraepithelial neoplasia formation in Pten(+/-) mice. Thus, Bub1 insufficiency can drive tumor formation through tumor suppressor gene loss of heterozygosity, but only in restricted genetic and cellular contexts.

Figure 1

(A) Kaplan-Meier overall survival curves of p53^+/−, Bub1^+/H/p53^+/−, Bub1^+/−/p53^+/−, Bub1^H/H/p53^+/−, Bub1^−/H/p53^+/−, Bub1^+/H, Bub1^+/−, Bub1^H/H and Bub1^−/H mice. *p < 0.05 compared to p53^+/− mice (Log rank test). Mice in these cohorts were monitored for 600 days. Animals represented in the curves are mice that died or that had to be sacrificed due to ill health or overt tumors. Cohorts of Bub1^+/−, Bub1^H/H and Bub1^−/H mice are as described (Jeganathan et al., 2007). (B) Kaplan-Meier tumor-free survival curves of mice presented in (A). Animals that died without tumors were censored from the data. We note that the percentage of p53^+/− animals dying without tumors was similar to that previously reported for this strain (Jacks et al., 1994). *p < 0.05 compared to p53^+/− mice (Log rank test). Legend is as in (A). (C) Tumor incidence and average tumor latency of mice shown in (A). *p < 0.05 versus p53^+/− mice (Fisher’s exact test for incidence; Log rank test for latency). **These cohorts were monitored for 1000 days, instead of 600 days. (D) Kaplan-Meier curves showing the thymic lymphoma-free survival of mice presented in (A). *p < 0.05 compared to p53^+/− mice (Log rank test). Legend is as in (A). (E) Same as (D) with sarcoma incidence plotted. Legend is as in (A). We note that for groups heterozygous for p53, after 500 days no individual group had more than 10 animals at risk for developing neoplasia. The large steps that appear during late age are because individual tumors that occurred in these small sample size groups reflect larger proportional changes. Furthermore, note that all Bub1^H/H/p53^+/− and Bub1^−/H/p53^+/− mice in the study died before 600 days. Therefore, the curves for these animal groups in (B), (D) and (E) do not extend to 600 days. Tick marks indicate censored animals in curves plotted in (B), (D) and (E).

Figure 2. p53Bub1^−/H/p53^+/− and Bub1^H/H/p53^+/− mice

(A) p53 status of lymphomas from Bub1^−/H/p53^+/− mice analyzed by PCR. Genomic DNA samples prepared from thymus of 6-week-old p53^+/+, p53^+/− and p53^−/− mice were used as PCR controls. (B) Analysis of the p53 status of lymphomas shown in (A) using Southern blotting as described (Jacks et al., 1994). Genomic DNA samples prepared from thymus of 6-week-old p53^+/+ and p53^+/− mice were used as controls. (C) Spectral karyotype analysis of lymphoma cells harvested from a representative Bub1^−/H/p53^+/− mouse. Shown are examples of cells (from the same tumor) that: (1) are karyotypically normal; (2) have 40 chromosomes but are karyotypically abnormal; (3 and 4) have minor numerical abnormalities (upper right and lower left); or, (5) exhibit combined numerical and structural chromosome changes (lower right). About 10–11% of tumor cells of three independent lymphomas had structural chromosome changes, which is to be expected for cells that lack p53, a key regulator of genome maitenance. Note that all cells have two copies of chromosome 11 (see red squares).

Figure 3. p53 LOH in lymphomas of Bub1^−/H/p53^+/− mice occurred by loss of chromosome 11 harboring the p53⁺ locus

Haplotypes of lymphomas and matched tail (normal) tissue from Bub1^−/H/p53^+/− mice.

Figure 4. Colon tumor formation dramatically increases in Apc^Min/+ mice when Bub1 levels decline

(A) Small intestine and colon tumor incidences in mutant mice of the indicated genotypes. Bub1 mutant strains not bred onto an Apc^Min/+ background did not develop any intestinal tumors. *p < 0.05 compared to Apc^Min/+ mice (Fisher’s exact test). (B) Small intestine tumor multiplicity. Color codes are as indicated in (A). Data are mean ± SEM. (C) Colon tumor burden. Color codes are as indicated in (A). Data are mean ± SEM. *p < 0.05 compared to Apc^Min/+ mice (unpaired t test). (D) Mean size of colon tumors. Data are mean ± SEM. *p < 0.05 compared to Apc^Min/+ mice (unpaired t test). (E) Colonic tumors in Apc^Min/+ and Bub1^−/H/Apc^Min/+ mice (Top). Colons were stained with methylene blue. Hatched yellow lines surround individual colon tumors. (Bottom) Hematoxylin-eosin stained sections of colon tumors from Apc^Min/+ and Bub1^−/H/Apc^Min/+ mice. Bars represent 2 mm.

Figure 5. Apc LOH in colon tumors from Bub1^−/H/Apc^Min/+ mice

(A) Chromosome counts on splenocytes of 3-month-old wild-type, Bub1^+/−, Bub1^−/H, Apc^Min/+, Bub1^+/−/Apc^Min/+ and Bub1^−/H/Apc^Min/+ mice. (B) PCR-based screening for Apc LOH in colon tumors from Bub1^−/H/Apc^Min/+. Tail DNA samples from wild-type and Apc^Min/+ mice were used as PCR controls. Note that colon tumors from Bub1^−/H/Apc^Min/+ mice consistently show preferential amplification of the Apc^Min allele, indicating loss of the Apc⁺ allele. (C) Interphase FISH image of a 5 µm colon tumor section of a Bub1^−/H/Apc^Min/+ mouse hybridized to both centromeric (red) and telomeric (green) chromosome probes. Quantification of the chromosome 18 copies in colon cancer cells and flanking normal colon epithelial cells. FISH signals of one hundred cells were counted per section. Scale bar is 5 µm.

Figure 6. Tumorigenesis in Rb^+/− mice is not impacted by Bub1 insufficiency

(A) Kaplan-Meier tumor-free survival curves of Rb^+/−, Bub1^−/H/Rb^+/−, Bub1^−/H and wild-type mice. Curves of Rb^+/− and Bub1^−/H/Rb^+/− mice are not significantly different from each other (Log rank test). (B) Tumor spectrum of mice represented in (A). There were no statistically significant changes in tumor spectrum between any of the mouse cohorts (Fisher’s exact test). (C) PCR-based screening for Rb LOH in pituitary tumors from Rb^+/− and Bub1^−/H/Rb^+/− mice. As controls we used DNA samples from flanking normal brain tissue and brain DNA samples of wild-type and Bub1^−/H mice. (D) Western blot analysis of normal brain (N) and pituitary tumors (T) of Rb^+/− and Bub1^−/H/Rb^+/− animals. Note that Bub1, but not the Bub1-related protein BubR1, was expressed at reduced levels in both normal and tumor tissues of Bub1^−/H/Rb^+/− mice. Tubulin was used as a loading control. (E) Chromosome counts on splenocytes of 5-month-old Rb^+/−, Bub1^−/H/Rb^+/− and Bub1^−/H mice.

Figure 7. Low Bub1 decreases high-grade PIN lesions in Pten^+/− males

(A) At 6 months, Pten^+/− males have significantly more PIN II and III lesions than age-matched Bub1^−/H/Pten^+/− mice. *p = 0.0017; **p = 0.0186 (unpaired t tests comparing similar grade PIN lesions). Data represented are mean ± SD. PIN IV multiplicity was similar for Pten^+/− and Bub1^−/H/Pten^+/− males. (B) PIN II lesions of 6-month-old Pten^+/− and Bub1^−/H/Pten^+/− males analyzed for Pten by immunohistochemistry. Pten was either undetectable (shown) or barely detectable (not shown) in PIN II lesions of both mouse strains. Normal prostate epithelium of a Pten^+/− male was used as a positive control for Pten staining. Scale bar is 100 µm. (C) Pten^+/− and Bub1^−/H genotypes strongly synergize in causing chromosome missegregation in splenocytes. Chromosome counts were performed on splenocytes of 5-month-old mice of the indicated genotypes. (D) Analysis of dead cells in PIN lesions of Pten^+/− and Bub1^−/H/Pten^+/− animals (at 6 months). (Top) High resolution image of a hematoxylin and eosin stained section of a PIN II lesion from a Bub1^−/H/Pten^+/− male showing cells with highly condensed pyknotic nuclei with halo like appearance. (Bottom) Quantification of pyknotic nuclei in various PIN lesions of Pten^+/− and Bub1^−/H/Pten^+/− animals. N = 270 PIN II, 100 PIN III and 12 PIN IV lesions per genotype. Data presented are mean ± SD. *p < 0.01 and **p < 0.05 (unpaired t test). Scale bar represents 10 µm. (E) Analysis of cell proliferation in various PIN lesions of 6-month-old Pten^+/− and Bub1^−/H/Pten^+/− males by immunohistochemistry for PCNA (n = 3 mice per genotype). Scale bar is 100 µm.