Slow disease progression in a C57BL/6 pten-deficient mouse model of prostate cancer

Abstract

Prostate-specific deletion of Pten in mice has been reported to recapitulate histological progression of human prostate cancer. To improve on this model, we introduced the conditional ROSA26 luciferase reporter allele to monitor prostate cancer progression via bioluminescence imaging and extensively backcrossed mice onto the albino C57BL/6 genetic background to address variability in tumor kinetics and to enhance imaging sensitivity. Bioluminescence signal increased rapidly in Pten(p-/-) mice from 3 to 11 weeks, but was much slower from 11 to 52 weeks. Changes in bioluminescence signal were correlated with epithelial proliferation. Magnetic resonance imaging revealed progressive increases in prostate volume, which were attributed to excessive fluid retention in the anterior prostate and to expansion of the stroma. Development of invasive prostate cancer in 52-week-old Pten(p-/-) mice was rare, indicating that disease progression was slowed relative to that in previous reports. Tumors in these mice exhibited a spontaneous inflammatory phenotype and were rapidly infiltrated by myeloid-derived suppressor cells. Although Pten(p-/-) tumors responded to androgen withdrawal, they failed to exhibit relapsed growth for up to 1 year. Taken together, these data identify a mild prostate cancer phenotype in C57BL/6 prostate-specific Pten-deficient mice, reflecting effects of the C57BL/6 genetic background on cancer progression. This model provides a platform for noninvasive assessment of how genetic and environmental risk factors may affect disease progression.

Figure 1

Bioluminescence imaging (BLI) of prostate cancer progression in C57/Luc/Pten^p−/− mice reveals biphasic expansion of the prostatic epithelium. A: Bioluminescent overlay images from the IVIS100 imaging system demonstrate increasing bioluminescence intensity in C57/Luc/Pten^p+/− and Pten^p−/− mice from 6 to 32 weeks. B: Quantitation of bioluminescence signal demonstrates that bioluminescence increase is rapid in C57/Luc/Pten^p−/− mice from 3 to 11 weeks, but then is slow from 11 to 52 weeks. Bioluminescence intensity also increases in C57/Luc/Pten^p+/− mice, compared with Pten^p+/+ mice, from 20 weeks. The bioluminescence signal is expressed as log photon flux (photons/second) from the region of interest surrounding the prostate. The square represents bioluminescence signal from nonserially imaged C57/Luc/Pten^p−/− mice. C: Immunofluorescence for Ki-67 demonstrates that bioluminescence intensity is correlated with proliferation in the prostatic epithelium. The 6-week-old C57/Luc/Pten^p−/− mice exhibit increased proliferation rates, compared with rates at 25 weeks. Examples of Ki-67-positive cells are marked by arrows. Scale bar = 100 μm. D: Quantitation of Ki-67-positive cells (% of total DAPI-positive epithelial cells) reveals a significant reduction in proliferation from 6 weeks to 25 weeks in C57/Luc/Pten^p−/− mice. ***P < 0.001, Student's t-test.

Figure 2

Magnetic resonance imaging reveals progressive increases in prostate growth attributable to excessive fluid retention in anterior prostate (AP) and stromal responses. C57/Luc/Pten^p−/− mice aged 3 months (n = 3), 6 months (n = 3), and 12 months (n = 3) were subjected to MRI to assess bulk prostate growth. A: Sagittal MRI images demonstrate progressive increases in prostate growth from 3 to 12 months of age (dashed line). B: Quantitation of MRI indicates significant increase in prostate volume (left), but BLI does not (right). C: Postmortem analyses revealed considerable fluid retention in AP from 12-month-old C57/Luc/Pten^p−/− mice, compared with age-matched C57/Luc/Pten^p+/+ mice (dashed lines, top row; imaging captured by digital camera). H&E and Masson's trichrome staining demonstrate florid stromal response in AP from 52-week-old C57/Luc/Pten^p−/− mice. Scale bar = 100 μm. **P < 0.01, ***P < 0.001, one-way analysis of variance.

Figure 3

Histological disease progression in C57/Luc/Pten mice. C57/Luc/Pten^p−/−, Pten^p+/−, and Pten^p+/+ mice were analyzed for pathological changes to the prostatic epithelium via histology. A: Hematoxylin and eosin staining of Pten^p+/+, Pten^p+/−, and Pten^p−/− mice demonstrates the development of pathological changes to the epithelial and stromal compartments in Pten^p+/− and Pten^p−/− mice. Arrowheads mark mPIN lesions, asterisks mark stromal response, the double dagger marks loss of gland architecture, and the arrow marks invasive lesion in a 52-week-old Pten^p−/− animal. Scale bar = 100 μm. B: Quantitative analysis of mPIN lesions in individual prostate lobes of Pten^p+/− and Pten^p−/− mice demonstrates that lobular progression to mPIN is asynchronous. No data were collected for AP of 52-week-old Pten^p−/− (asterisk). For numbers of animals, see Table 1.

Figure 4

Stromal changes in prostates from C57/Luc/Pten^p−/− mice. A: Masson's trichrome stain in C57/Luc/Pten^p+/+ and Pten^p−/− mice demonstrates increased collagen deposition in 25-week-old Pten^p−/− prostates. B: H&E stain reveals infiltrating inflammatory cells (arrows) in prostate lobes of 25-week-old C57/Luc/Pten^p−/− mice. Scale bars = 100 μm.

Figure 5

C57/Luc/Pten^p−/− mice exhibit persistence of a castrate-resistant population of epithelial cells, but fail to relapse to a growth phase. C57/Luc/Pten^p−/− mice were monitored for response to hormone deprivation therapy via BLI and histopathology. A: Quantitation of bioluminescence signal demonstrates that after castration at 11 weeks (arrow), bioluminescence signal in Pten^p−/− mice remains constant above initial values at 3 weeks (dotted horizontal line), compared with Pten^p+/− mice, which display progressively decreasing signal intensity. B: Images of whole prostate sections demonstrate that castrated Pten^p−/− prostates are significantly smaller than intact 52-week-old Pten^p−/− prostates. Castrated Pten^p−/− prostates at 35 and 59 weeks after castration are of similar size. Scale bar = 1 mm. C: Ki-67 staining reveals that castrated Pten^p−/− prostates proliferate significantly less than intact 6-week-old Pten^p−/− prostates. D: Quantitation of chromogranin A (CgA) staining reveals no significant difference in numbers of CgA-positive cells in castrated versus intact Pten^p−/− prostates. ***P < 0.001, Student's t-test.

Figure 6

Recruitment of CD45⁺Gr-1⁺CD11b⁺ cells, which express Arg-1 and iNOS and are T-cell suppressive, in prostates of C57/Luc/Pten^p−/− mice. Prostates from 4-month-old C57/Luc/Pten^p−/− (n = 6) and Pten^p+/+ mice (n = 6) were analyzed for infiltrating CD45⁺ and CD45⁺Gr-1⁺CD11b⁺ cells by fluorescence-activated cell sorting and were assessed for Arg-1 and iNOS expression and for their ability to inhibit T-cell proliferation in vitro. A: Absolute numbers of infiltrating CD45⁺ cells are increased in DP, AP, and VP in Pten^p−/− C57/Luc mice. B: Absolute numbers of CD45⁺CD11b⁺ and Gr-1⁺ cells are increased in DP, AP, and VP of C57/Luc/Pten^p−/− mice. C: Arg-1 and iNOS are up-regulated in prostate CD45⁺Gr-1⁺CD11b⁺ cells. ND, not detected. D: CD45⁺Gr-1⁺CD11b⁺ cells from prostates of C57/Luc/Pten^p−/− mice are more potent suppressors of T-cell proliferation in vitro, compared with the two other genotypes.