Breast cancer stromal fibroblasts promote the generation of CD44+CD24- cells through SDF-1/CXCR4 interaction

Abstract

Background: Breast cancer stem cells (BCSCs) have been recently identified in breast carcinoma as CD44+CD24- cells, which exclusively retain tumorigenic activity and display stem cell-like properties. Using a mammosphere culture technique, MCF7 mammosphere cells are found to enrich breast cancer stem-like cells expressing CD44+CD24-. The stromal cells are mainly constituted by fibroblasts within a breast carcinoma, yet little is known of the contributions of the stromal cells to BCSCs.

Methods: Carcinoma-associated fibroblasts (CAFs) and normal fibroblasts (NFs) were isolated and identified by immunohistochemistry. MCF7 mammosphere cells were co-cultured with different stromal fibroblasts by a transwell cocultured system. Flow cytometry was used to measure CD44 and CD24 expression status on MCF7. ELISA (enzyme-linked immunosorbent assay) was performed to investigate the production of stromal cell-derived factor 1 (SDF-1) in mammosphere cultures subject to various treatments. Mammosphere cells were injected with CAFs and NFs to examine the efficiency of tumorigenity in NOD/SCID mice.

Results: CAFs derived from breast cancer patients were found to be positive for alpha-smooth muscle actin (alpha-SMA), exhibiting the traits of myofibroblasts. In addition, CAFs played a central role in promoting the proliferation of CD44+CD24- cells through their ability to secrete SDF-1, which may be mediated to SDF-1/CXCR4 signaling. Moreover, the tumorigenicity of mammosphere cells with CAFs significantly increased as compared to that of mammosphere cells alone or with NFs.

Conclusion: We for the first time investigated the effects of stromal fibroblasts on CD44+CD24- cells and our findings indicated that breast CAFs contribute to CD44+CD24- cell proliferation through the secretion of SDF-1, and which may be important target for therapeutic approaches.

Figure 1

Mammosphere cells contained subpopulations of cells expressing prospective BCSC markers. (A) FACS analysis to measure CD44 and CD24 expression of cells derived from MCF7 monolayer cultures (left) or primary mammospheres (right), which were cultured in suspension for six days. The expression of CD44⁺CD24^- in mammosphere cells was (7.9 ± 0.8%), compared with (1.9 ± 0.1%) for the monolayer culture cells, P < 0.01. A minimum of 10,000 events were collected per sample. (B) qRT-PCR showed that Notch2 and β-catenin mRNA expression in mammosphere cells were at higher levels by around 4.0 and 3.1 fold than that in monolayer cells, respectively, P <0.01. The data were provided as the mean ± SD. Each experiment was performed three times.

Figure 2

Immunohistochemistry of NFs and CAFs. (A) Phase images of primary cultures of stromal fibroblasts isolated from invasive ductal carcinomas (right) and stromal fibroblasts from normal breast tissue (left), original magnification × 100. (B) CAFs (right) were positive for α-SMA staining, while NFs (left) were negative.

Figure 3

Mammosphere cells were cocultured with different stromal fibroblasts and flow cytometry was used to measure CD44 and CD24 expression. (A) Mammosphere cells (1 × 10⁵ cells/dish) cocultured with different stromal fibroblasts (1 × 10⁵ cells/dish) using transwells for six days, and mammosphere cells cocultured with CAFs (middle) had the highest MFE (13.5 ± 1.2%), compared with monoculture mammosphere cells (left) (8.1 ± 0.7%), P < 0.01. (B) Flow cytometry analysis to measure CD44 and CD24 expression of cells derived from monoculture mammosphere cells and cocultured mammosphere cells. The expression of CD44⁺CD24^- in monoculture mammosphere cells (left) was (17.2 ± 2.3%). Compared to monoculture mammosphere cells, the expression of CD44⁺CD24^- in cocultured mammosphere cells with CAFs (middle) was (21.4 ± 1.8%), P < 0.05, and the expression of CD44⁺CD24^- in cocultured mammosphere cells with NFs (right) was (8.7 ± 0.9%), P < 0.01. The data were provided as the mean ± SD. Each experiment was performed three times.

Figure 4

The SDF-1 protein expression in cocultured medium of mammosphere cells with CAFs and NFs. The SDF-1 protein level in the medium conditioned by the CAFs was (426.4 ± 30.6) (pg/ml) (middle), compared to the levels produced by mammosphere cells alone (283.6 ± 35.1) (pg/ml) (left), P <0.05. The cocultured medium of mammosphere cells with NFs (right) showed a far lower level of SDF-1(52.9. ± 13.1) (pg/ml) secretion when compared with mammosphere cells alone, P <0.01. The SDF-1 level was measured three times in each experiment.

Figure 5

Mammosphere cells and monolayer cells were cultured in the presence of 1 μg/ml AMD3100 for 48 h. qRT-PCR showed that CXCR4 mRNA expression in mammosphere cells was 3.9 fold higher than that in monolayer cells, (P <0.01), and AMD3100 could significantly down-regulate it in both when treated for 48 h.

Figure 6

Mammosphere cells were cocultured with different stromal fibroblasts with the administration of AMD3100 and flow cytometry was used to measure CD44 and CD24 expression. (A) Mammosphere cells were cocultured with different stromal fibroblasts with the administration of AMD3100 (1 μg/ml) for six days. As a result, MFE in monoculture mammosphere cells (left), cocultured mammosphere cells with CAFs (middle) and NFs (right) was significantly reduced to (1.6 ± 0.1%), (2.3 ± 0.2%) and (1.5 ± 0.2%), respectively. (B) Flow cytometry analysis was used to measure CD44 and CD24 expression of cells derived from mammosphere cells. The expression of CD44⁺CD24^- in monoculture mammosphere cells (left), cocultured mammosphere cells with stromal CAFs (middle) and NFs (right) was (2.2 ± 0.3%), (4.4 ± 0.8%) and (2.7 ± 0.3%), respectively. The data were provided as the mean ± SD. Each experiment was performed three times.