br These results agreed with those obtained with NP
These results agreed with those obtained with NP intracellular up-take and suggest that PLGA NPs enhance cellular internalisation of PTX, which is a notable advantage over the free drug. Interestingly, the in-tracellular PTX concentration remained constant in lung tumor cell
lines following a peak of drug accumulation after the first hour of ex-posure. By contrast, normal lung cell lines showed a very pronounced decrease after the first hour of incubation (6 × 10−4 μg PTX/μg pro-teins) reaching a low drug concentration at the end of the exposure time (2 × 10−4 μg PTX/μg proteins at 4 h). Leiva et al.  also showed that Tripalm-PTX-NP favoured PTX entry in A549 and MCF7 cells, sig-nificantly increasing the accumulation of PTX at 2 h and 4 h compared to free PTX. Xu et al.  observed accumulation of NR in the peri-nuclear region after just 1 min of exposure to PLGA-NR NPs. Similar results were obtained with micelles of poly(ε-caprolactone) or PCL loaded with FITC-PTX at 2 h compared to free FITC-PTX in LL2 cells. These results were corroborated by fluorescence imaging and in-tracellular HPLC measurements of PTX . In our case, the use of the PLGA NPs meant we observed significant diﬀerences after periods as short as 30 min compared to free PTX.
3.7. PLGA nanoparticle modulation of drug resistance
In order to determine the ability of PLGA NPs to inhibit the me-chanism of drug resistance mediated by P-gp, a retention rhodamine test (P-gp substrate) was carried out. We used the resistant HCT-15 cancer cell line, with high P-gp expression, and the sensitive T84 cell line (see Methods). Verapamil was used as a positive control of P-gp inhibition. As shown in Fig. S7, pre-incubation of resistant POM 1 with blank PLGA NPs showed a rhodamine profile similar to those observed in verapamil non-pre-treated cells, suggesting that the NPs were unable to bypass the resistance mechanism of P-gp.
3.8. Characterisation of cancer stem cells and proliferation assays
Despite the lack of evidence of interference with the P-gp resistance mechanism, PLGA-PTX NPs were assayed against resistant CSCs from A549 characterised as being radio- and chemoresistant. Tumorspheres from A549 CSCs (see Methods) were observed from the first days of testing (Fig. 6A). These CSCs had a similar morphology to that observed by other authors who have used similar isolate assays with lung tumor lines [31,45,46]. The presence of specific CSC markers in tumorspheres was assessed by quantitative real-time PCR analysis (Fig. S8). The four
Fig. 6. Cancer stem cells proliferation assays. A, Representative inverted phase-contrast microscopy images of A-549 CSCs from day 1 (cellular aggregation) to day 14 (cellular compact aggregates around 300 μm); Magnification, 4 × . B, Cytotoxicity of free PTX and PLGA-PTX NPs in CSCs after 72 h of incubation (1–100 nM). Percentage of cellular viability (CV(%)) represented the mean value ± SD of octuplicates cultures. *Data with significant diﬀerences between PLGA-PTX and free PTX treatment (p < 0,05).
markers studied for the stem cell phenotype were overexpressed, as described in other studies [45,47]. Exposure of the isolated and char-acterised CSCs to PLGA-PTX NPs and free PTX demonstrated a sig-nificant resistance to both drug formulations in comparison with A549 cancer cells (see “proliferation assay”). As shown in Fig. 6B, the concentration of PTX and PLGA-PTX NPs used against CSCs (1–100 nM) did not reach the PTX IC50. The blank PLGA NPs did not present toxicity for the CSCs. Interestingly, the lowest doses of PLGA-PTX NPs induced a significantly greater antitumor eﬀect against CSCs (around 10%) compared to free PTX. These results may prove relevant when attempting to obtain a more selective therapy against CSCs that are associated with resistance, recurrence and me-tastasis in lung cancer patients . In pursuing this objective, Swa-minathan et al.  functionalized PLGA NPs loaded with PTX with the anti-CD33 antibody, an overexpressed marker in CSCs. More recently, Wang et al.  developed NPs associated with doxorubicin and ir-inotecan and functionalised with hyaluronic acid (HA) (HA-PFP-DC) in order to treat CD44 of CSCs. Compared to the HA-free NPs, these na-noformulations reduced the proliferation of CSCs derived from prostate and breast cancers by up to 66.3 and 419.1-fold, respectively.