ABCG2, a novel antigen to sort luminal progenitors of BRCA1^- breast cancer cells

To identify novel antigens that can improve the power of the CD44/CD24 antigenic phenotype in order to isolate TICs, we measured the expression of 28 surface antigens reported to be essential for cell adhesion, migration, apoptosis, cell signaling or stemness (Table 1 and Additional file 1: Table S1) in two basal A/basal-like cell lines, namely BT20 and HCC1937 (BRCA1^-/-) and the basal B/claudin-low Hs578T cell line. Figure 1 shows the expression of these antigens, as assessed by flow-cytometry, in the CD44⁺/CD24⁺ and CD44⁺/CD24^-/low cell subpopulations. Particularly, we determined the ratio between the percentage of cells positive for each antigen in the two cell subpopulations of each cell line (Figure 1b). No significant differences were observed in the expression of the examined antigens between the CD44⁺/CD24⁺ and CD44⁺/CD24^-/low cell subpopulations in the BT20 and Hs578T cell lines (ratio ± 1), while several of them were significantly enriched in the CD44⁺/CD24⁺ population of HCC1937, including the stemness markers CD10, CD133 and CD338/ABCG2 [25, 31, 32]. Evaluation of the mean fluorescence intensity (MFI) of each surface marker in the CD44⁺/CD24⁺ and CD44⁺/CD24^-/low cell subpopulations of the HCC1937 cell line, demonstrated that CD338 is expressed at a higher level in the CD24⁺ than in CD24^- cell subpopulation (Additional file 2: Figure S1).

Given the high differential of CD338 (ratio >3, Figure 1), we explored the link between CD24 and CD338 expression. To this aim, we gated the HCC1937 CD338^high and CD338^- cell subpopulations, and measured CD24 expression. As shown in Figure 2, the MFI was 7-fold higher in CD338^high cells (panel a, red events) than in CD338^- cells (panel a, blue events; mean ± SEM: 5,200.0 ± 916.5 and 1,100.0 ± 404.1, p < 0.05). This finding is consistent with a overlap between CD24⁺ and CD338^high cells.

If CD338 is a reliable marker of TICs, culture conditions reported to induce enrichment of stem cells and progenitors should be associated with an increase in CD338 expression. To address this issue, we performed mammosphere formation assays in which we ran three successive culture/dissociation passages in ultra-low adherent conditions, and measured the percentage of CD338-expressing cells. The percentage of CD338^high cells in mammospheres was 4.9-fold higher than that in the whole cell line when cultured in adherent conditions (mean ± SEM: 8.3% ± 0.5 and 1.7% ± 0.1 respectively, p < 0.0001). Furthermore, the MFI of CD338 cells, which indicates its expression level, was 3.8-fold higher in mammosphere-derived cells than in adherent cells. This observation supports the assumption that CD338-positive cells display some stem cell-like properties (Figure 3). Strikingly, CD338-positive cells in ultra-low adherent conditions were the only cell subpopulation to retain CD326/EpCAM and CD49f/α6-integrin expression (Figure 3f), which is an antigenic phenotype assigned to luminal progenitors.

If CD338 is an antigenic marker of TICs in the HCC1937 cell line, the CD338^high cellular subpopulation would be expected to generate mammospheres when cultured in ultra-low adherent conditions, whereas CD338^- cells should be devoid of stemness properties. To test this hypothesis, we sorted by flow-cytometry (Additional file 3: Figure S2) three distinct populations that differ in the expression of CD338: a CD338^high population expressing CD338 at high level (1% of cells), a CD338^low population expressing CD338 at low level (79% of cells), and a CD338^- population negative for CD338 (20% of cells). The three sorted populations were grown for two days in standard adherent conditions to allow them to recover from the sorting procedure before testing their stemness properties in a mammosphere formation assay. As expected, after two serial passages, only CD338-expressing cells gave rise to mammospheres (Figure 4a).

We next assessed the transformation potential of the three subpopulations in a soft-agar colony assay. As shown in Figure 4b, CD338-positive cells, CD338^high and CD338^low, displayed a significantly higher transformation potential than CD338^- cells. The few colonies observed in CD338^- cells probably reflect contamination of the cell population with CD338^low cells during the cell sorting (Additional file 4: Figure S3). There were no significant differences between CD338^high and CD338^low populations in either the mammosphere or the and colony formation assay. Assays were invariably performed shortly after reseeding sorted cells. Notably, after several days in culture, CD338^high cells gave rise to a heterogenous CD338^high and CD338^low population, which suggests parenting between these cells (Figure 5). Both CD338^high and CD338^low cells, but not CD338^- cells, displayed stemness properties and transformation potential.

To strengthen our conclusions, we next assessed the consequences of ABCG2 depletion on stemness properties of HCC1937 breast cancer cells by performing mammospheres formation assays. In line with our expectations, the knockdown of ABCG2, achieved through RNA interference, leads to a significant decrease of stemness potentials (Figure 6).

Because the tumorigenic potential is specific to TICs, we assumed that, when xenografted into immunocompromised mice, HCC1937 cell-generated tumors would be enriched in CD338-expressing cells. To address this point, 2 × 10⁶ and 4 × 10⁵ HCC1937 cells were injected into the left and right flanks of NOD/SCID mice, respectively (Figure 7a). Invariably, cells induced tumor formation with a delay depending on the number of cells injected (100%, n = 5). Tumors were excised, digested to single cell suspensions and analyzed by flow-cytometry. The percentages of CD338^high cells were determined in viable (SYTOX^-), human (HLA-ABC⁺), epithelial (EpCAM⁺)-gated tumor-derived cells (Figure 7b). The CD338^high subpopulation was significantly enriched in both tumors (38.8% ± 1.1) compared with the parental cell line (Figure 7a). To compare the tumorigenic potential of the CD24⁺ and CD24^- cell subpopulations, we sorted the two populations and immediately xenografted them into mice. The same number of cells (assays were performed with either 5 × 10⁴ or 5 × 10⁵ cells) were injected into the flanks of five mice. All injections invariably led to tumor growth and there were no obvious differences between the CD24⁺ and CD24^- cell subpopulations. Moreover, no difference in tumor growth was detected between CD24 sorted cells and the unsorted HCC1937 cell line (data not shown). While the CD24⁺ and CD24^- cell populations displayed a similar tumorigenic potential, CD24⁺-derived tumors had a higher percentage of CD338^high cells than CD24^--derived tumors (60.2% ± 3.4 versus 42.5% ± 0.5) (Figure 7c). This observation strengthens the link between CD24 and CD338 expression.

The human breast cancer cell lines Hs578T and BT-20 were provided by American Type Culture Collection (Rockville, MD, USA) and cultured in DMEM, 10% FBS. The HCC1937 cell line was from the American Type Culture Collection, and was cultured in IMDM medium (Invitrogen), 20% FBS (Gibco).

shRNA ABCG2 lentiviral particles were generated through co-transfection of 293 T cells with 4 different shRNA pLKO.1 (4 different shRNAs GCAACAACTATGACGAATCAT, CCTTCTTCGTTATGATGTTTA,GCTGTGGCATTAAACAGAGAA,CCTGCCAATTTCAAATGTAAT, Sigma Aldrich), pCMV R8.91 (gag-pop-Tat-Rev) and phCMVG-VSVG (env) expression constructs using the calcium phosphate precipitation technique. Infections were performed as described in [64]. Infected cells were selected with puromycin (1 μg/ml) before being plated in ultra-low adherent conditions or plated on soft agar. Inhibition of ABCG2 expression was confirmed by qRT-PCR.

For mammosphere generation, HCC1937 cells were seeded in 96-well ultra-low attachment plates (Corning, New York, NY, USA) at the concentration of 1,000 cells/well for first-generation mammospheres, and at 100 cells/well for subsequent passages. Cells were grown in a serum-free mammary epithelial growth medium (MEBM, Lonza, Verviers, Belgium) supplemented with B27 (Invitrogen, Carsbal, CA, USA), 20 ng/ml EGF, 20 ng/ml bFGF and 4 μg/ml heparin (Sigma, St. Louis, MO, USA). Mammospheres were collected by gentle centrifugation (1,000 rpm, 5 minutes), and enzymatically (5 min in trypsin/EDTA) and mechanically (26 Gauge needle) dissociated between each step and for further analysis.

Plates were coated with 0.75% low-melting agarose (Lonza) obtained by mixing equal volumes of 1.5% agar and 2× growth medium (IMDM). Cells were enzymatically dissociated, resuspended in growth medium and counted. An overlaid suspension of cells in 0.45% low-melting agarose was obtained by mixing equal volumes of 0.9% agar and 2 × IMDM with cells (5 × 10⁴ cells/well in 6-well plates). Plates were incubated for 3–4 weeks at 37°C and colonies were counted under microscope.

All animal experiments were conducted in accordance with accepted standards of animal care and in agreement with a protocol established with our Ethics Committee. The study has been approved by the Institutional Review Board “CEINGE-Biotecnologie Avanzate”. To evaluate in vivo tumorigenicity, unsorted HCC1937 or sorted cell subpopulations were resuspended in media and Matrigel (1:1; BD Biosciences), and injected into the left and right flanks of 4-week-old NOD/SCID mice (C. River laboratories). To evaluate the tumorigenic potential of the unsorted cell line, 2 × 10⁶ and 4 × 10⁵ of HCC1937 cells were injected into the left and right flanks of mice respectively (n = 5). To compare the tumorigenic potentials of the CD24⁺ and CD24^- cell populations, 5 × 10⁴ or 5 × 10⁵ cells were injected into the left and right flanks of NOD/SCID mice respectively (n = 5). Tumor formation was assessed and measured once a week. Animals were killed when the tumor reached the size of 1.5 cm. Tumor tissues were minced into < 1 mm pieces, dissociated in an enzymatic solution consisting of collagenase type 1 (1.5 mg/ml, Sigma), penicillin/streptomycin 20%, amphotericin 1% and DNase (1 mg/ml, Roche), and incubated at 37°C for 60 min with gentle agitation. The single cell suspensions were analyzed by flow-cytometry after staining with appropriate antibodies.