Expansion of prostate epithelial progenitor cells after inflammation of the mouse prostate

Abstract

Prostatic inflammation is a nearly ubiquitous pathological feature observed in specimens from benign prostate hyperplasia and prostate cancer patients. The microenvironment of the inflamed prostate is highly reactive, and epithelial hyperplasia is a hallmark feature of inflamed prostates. How inflammation orchestrates epithelial proliferation as part of its repair and recovery action is not well understood. Here, we report that a novel epithelial progenitor cell population is induced to expand during inflammation. We used sphere culture assays, immunofluorescence, and flow cytometry to show that this population is increased in bacterially induced inflamed mouse prostates relative to naïve control prostates. We confirmed from previous reports that this population exclusively possesses the ability to regrow entire prostatic structures from single cell culture using renal grafts. In addition, putative progenitor cells harvested from inflamed animals have greater aggregation capacity than those isolated from naïve control prostates. Expansion of this critical cell population requires IL-1 signaling, as IL-1 receptor 1-null mice exhibit inflammation similar to wild-type inflamed animals but exhibit significantly reduced progenitor cell proliferation and hyperplasia. These data demonstrate that inflammation promotes hyperplasia in the mouse prostatic epithelium by inducing the expansion of a selected epithelial progenitor cell population in an IL-1 receptor-dependent manner. These findings may have significant impact on our understanding of how inflammation promotes proliferative diseases such as benign prostatic hyperplasia and prostate cancer, both of which depend on expansion of cells that exhibit a progenitor-like nature.

Keywords: hyperplasia; inflammation; interleukin-1; progenitor; prostate.

Fig. 1.

Inflammation increases the proportion of epithelial cells capable of forming large spheres in the mouse prostate but not the total sphere number. Isolated epithelial cells from inflamed and control mice were grown in anchorage-independent conditions for 21 days at a density of 10,000 cells/10 ml culture. After 21 days, the number of cell groups that met the criteria for being called a sphere (>3 cells in diameter) were counted and their size was determined by microscopy (n = 6). A: the total number of spheres per 10,000 cells did not significantly increase with inflammation. B: the number of spheres 5–20 cells in diameter was increased 3-fold. C: the number of very large (>20 cells in diameter) spheres increased 10-fold. D: the average diameter of spheres cultured from inflamed prostates increased significantly compared with controls. E: example spheres from control prostates. F: example of a very large sphere (>10 cells in diameter) from an inflamed prostate as quantified in Fig. 1C. *P < 0.05 inflamed vs. control (n = 6, ANOVA).

Fig. 2.

Larger spheres exhibit adhesive qualities in vitro. A–D: fluorescent images of spheres isolated from control and inflamed prostates from green fluorescent protein (GFP)- and dtTomato-red fluorescent protein (RFP)-expressing mice and cultured in anchorage-independent conditions for 7 days. A: spheres cultured from control GFP-expressing mice. B and C: spheres cultured from inflamed GFP-expressing (B) and dtTomato-RFP-expressing (C) mice. D: spheres cultured from a 1:1 mix of cells from GFP- and dtTomato-RFP-expressing mice, illustrating spheres with cells from multiple mice. E: the majority of spheres cultured contained cells from both mice, indicating cell-cell aggregation, and all spheres with a diameter ≥8 cells had both GFP and RFP components.

Fig. 3.

Inflammation increases the number of four-marker (Lin⁻, Sca-1⁺, CD44⁺, CD133⁺, c-Kit⁺) putative progenitor cells in the prostate. A: isolated epithelial cells from control and inflamed mouse prostates from days 1 to 5 of inflammation were analyzed by flow cytometry for cells positive for the four-marker panel (n = 6). The number of four-marker cells was significantly increased 3 days postinflammation. *P < 0.05 vs. control (n = 6, one-way ANOVA). B: flow cytometry data comparing the expression of CD49F (an alternate progenitor cell marker) in the four-marker population relative to that in total epithelial cells. C: correlation data representing the number of four-marker cells plotted with the number of hematopoietic Lin⁺ cells in the same prostate, as an indicator of inflammation severity. Four-marker cells correlated with Lin⁺ cell number, suggesting that four-marker cell abundance correlates with inflammation severity. Pearson correlation coefficient = 0.957, n = 10 inflamed prostates and 13 control prostate. #Four-marker cells from a control mouse that did not meet the criteria of being noninflamed (<20% Lin⁺ cells/total viable cells). D: sorted four-marker cells have greater sphere-forming capacity than total epithelial cells do, but those isolated from inflamed prostates have no greater sphere-forming capacity than those from control prostates (n = 4). PEPC, primary epithelial prostate cells.

Fig. 4.

Four-marker cells have the capacity to form prostatic structures in a renal graft from a single cell. Four-marker cells were isolated into single cells by flow sorting and grown into spheres in anchorage-independent conditions. Spheres were then combined with urogenital mesenchymal cells from dtTomato-expressing mice and implanted under the renal capsule. A–C: hematoxylin and eosin-stained sections of an example graft that grew into a prostatic structure at ×10 (A), ×20 (B), and ×40 (C) magnification. D: immunofluorescence imaging demonstrating prostatic structure-expressing probasin (red) off the luminal epithelium and dtTomato (green) in a fibroblast population.

Fig. 5.

c-Kit⁺ cells expand from the basal to luminal compartment during inflammation. A: immunofluorescence image (c-Kit, green; pan-cytokeratin, red; yellow arrows denote dual-stained cells) demonstrating a rare c-Kit+ cell in the basal layer of a control prostatic duct. B and C: c-Kit⁺ cells expand beyond the basal layer during 2 and 3 days of inflammation into the intermediate and luminal layers. D: calculations for expanded c-Kit⁺ cells, as measured by immunofluorescence and flow cytometry, demonstrating the consistency of the results. Four ×20 fields were counted per specimen for each single data point, and there were 6 prostates in each group. *P < 0.05 vs. control prostates (by ANOVA). n = 4 for immunofluorescence and 6 for flow cytometry. E and F: costaining of c-Kit⁺ (green) and bromodeoxyuridine (BrdU)⁺ (red) in control (E) and 2-day inflamed (F) prostates indicating that half of the c-Kit⁺ cells were also proliferative (BrdU⁺), whereas we did not observe BrdU⁺/c-Kit⁺ cells in control prostates. Green arrows indicate c-Kit⁺ BrdU⁻ cells; magenta arrows indicate BrdU⁺/c-Kit⁺ cells.

Fig. 6.

Mice lacking intact IL-1 receptor (IL-1R) exhibit decreased expansion of putative progenitor cells. A: the inflammation-induced c-Kit⁺ population was attenuated by loss of IL-1Rs, as a percentage of total epithelial cells. *P < 0.05, wild-type vs. IL-1R1^−/− mice. n = 6. B: the 3-day inflamed four-marker cell population was attenuated in IL-1R1^−/− mice, as a percentage of epithelial cells. *P < 0.05, inflamed vs. control tissues; #P < 0.05, wild-type vs. IL-1R1^−/− inflamed tissues. n = 5–8.