Targeting CXCR1 on breast cancer stem cells: signaling pathways and clinical application modelling

Abstract

In breast cancer it has been proposed that the presence of cancer stem cells may drive tumor initiation, progression and recurrences. IL-8, up-regulated in breast cancer, and associated with poor prognosis, increases CSC self-renewal in cell line models. It signals via two cell surface receptors, CXCR1 and CXCR2. Recently, the IL-8/CXCR1 axis was proposed as an attractive pathway for the design of specific therapies against breast cancer stem cells. Reparixin, a powerful CXCR1 inhibitor, was effective in reducing in vivo the tumour-initiating population in several NOD/SCID mice breast cancer models, showing that the selective targeting of CXCR1 and the combination of reparixin and docetaxel resulted in a concomitant reduction of the bulk tumour mass and CSC population. The available data indicate that IL-8, expressed by tumour cells and induced by chemotherapeutic treatment, is a key regulator of the survival and self-renewal of the population of CXCR1-expressing CSC. Consequently, this investigation on the mechanism of action of the reparixin/paclitaxel combination, was based on the observation that reparixin treatment contained the formation of metastases in several experimental models. However, specific data on the formation of breast cancer brain metastases, which carry remarkable morbidity and mortality to a substantial proportion of advanced breast cancer patients, have not been generated. The obtained data indicate a beneficial use of the drug combination reparixin and paclitaxel to counteract brain tumour metastasis due to CSC, probably due to the combined effects of the two drugs, the pro-apoptotic action of paclitaxel and the cytostatic and anti-migratory effects of reparixin.

Keywords: preclinical MRI.

Figure 1. Mammospheres formation and characterization from MDA-MB231: breast tumor stem cell marker enrichment, with respect to the starting cell line, such as ABCG2 and ALDHA1, CXCR1 and CXCR2, evalutated by cytofluorimetry, is reported in (A)

In (B) the increase of the stem cell population in the mammospheres is confirmed by the Aldefluor assay, measuring ALDHA1 activity. On purified mammospheres CD44 was also measured by cytofluorimetry (C) Data are mean ± SE of three different experiments. **p ≤ 0.005; ***p ≤ 0.0005. In (D) immunofluorescence for ABCG2 and ALDHA1, CXCR1 and CXCR2 in purified mammospheres. Bar = 25 μm.

Figure 2. Cell viability: mammosphere viability (A) in control and treated conditions

Data are expressed as percentage of the respective control. In (B) cytofluorimetric apoptosis detection in purified mammospheres in control and treated conditions. Data are mean ± SE of three different experiments. *p ≤ 0.05; **p ≤ 0.005; ***p ≤ 0.0005. R: reparixin; P: paclitaxel; R + P: reparixin + paclitaxel.

Figure 3. Tumorsphere, number, size, proliferation and microfilament/fokal adhesion organization: In (A-C) for mammospheres number, and for single cells, data are expressed as X-fold of the respective control, while volume is expresed in μm3

Data are mean ± SE of three different experiments. *p ≤ 0.05; **p ≤ 0.005; ***p ≤ 0.0005; +, p≤ 0.05. In (D) CFSE staining in control and treated mammospheres. Bar = 25 μm. In (E) falloidin-TRITC decoration of microfilaments in control and treated mammospheres. Bar = 25 μm. In (F) a representative western blotting and relative densitometric analysis, expressed as relative units, for p-FAK and β-catenin in control and treated mammospheres. Data are mean ± SE of three different experiments. *p ≤ 0.05. C: control: R: reparixin 10 μM; P: paclitaxel 5 nM; R + P: reparixin 10 μM + paclitaxel 5 nM.

Figure 4. In (A) cytofluorimetric analyisis of cell cycle on control and treated mammospheres

Data are mean ± SE of three different experiments. ***p ≤ 0.0005. In (B) a representative western blotting and relative densitometric analysis, expressed as relative units, for cyclin B1 in control and treated mammospheres. Data are mean ± SE of three different experiments. **p ≤ 0.005. In (C) immunofluorescence analysis for cyclin B1 in control and treated mammospheres, Bar = 25 μM. C: control: R: reparixin 10 μM; P: paclitaxel 5 nM; R + P: reparixin 10 μM + paclitaxel 5 nM.

Figure 5. Representative western blotting and relative densitometric analysis, expressed as relative units, for PI3K, p-AKT, p27, NFKB, p-EGFR in control and treated mammospheres

Data are mean ± SE of three different experiments. *p ≤ 0.05; ***p ≤ 0.0005. p-FOXO3A was analyzed by immunolocalization of the inactive phosphorylated form. Bar = 50 μm. C: control: R: reparixin 10 μM; P: paclitaxel 5 nM; R + P: reparixin 10 μM + paclitaxel 5 nM.

Figure 6. Effect of anti-CXCR1 and anti-IL8 neutralization on mammospheres

In (A) cytofluorimetric analysis of cell cycle in the presence of the specific antibody administered alone or in combination with paclitaxel (P). Data are mean ± SE of three different experiments. ***p ≤ 0.0005. In (B) representative western blotting and relative densitometric analysis, expressed as relative units, for cyclin B1 and p-FAK in the same experimental conditions. Data are mean ± SE of three different experiments. *p ≤ 0.05; **p ≤ 0.005; ***p ≤ 0.0005.

Figure 7. Metastasis formation in nude mice injected with MDA-MB231 cells in the two different experimental models used: the early metastatic model (T14) and a claimed brain tumor occurrence model (T21), both examined by MRI and measured by ROI segmentation

Each group consisted of at least 11 animals. Data are mean ± SE. *p ≤ 0.05; **p ≤ 0.01. For each group the top row of each histogram reports representative MRI images, showing ROIs utilized for quantification in control and treated conditions are Bar = 2 mm. VTOT is expressed in μl.

Figure 8. Representative MRI pictures of T21 vehicle and treated animals and HE staining and immunolocalization of PCNA in the same areas identified in the MRI images (red inset)

MRI: Bar = 1,7 mm; HE: Bar = 400 μm; PCNA: Bar = 50 μm.