Synthesis and characterization of the multifunctional nanovectors
Various PNPs systems were prepared with the aim of demonstrating different points: selective cell internalization, tumor targeting and biodistribution, in vitro cytotoxicity, anti-tumor effect both in vivo and ex vivo. The common base for all the samples was the PLGA-b-PEG copolymer, able to generate stable nanoparticles’ emulsions. Thanks both to a very versatile formulation procedure, called water-in-oil-in-water double-emulsion, and to the lipo and hydro blocks of the polymer, it is easy to encapsulate molecules whether inside the ‘oil’ moiety, as the lipophilic dye BODIPY for visualization purposes in vitro, or inside the inner ‘hydro’ portion, as the Cis-Pt to assess the anti-tumor effect in and ex vivo. For the selective targeting of EGFR, the so differently loaded PNPs have been further conjugated onto the external surface with CL4 aptamer, or its scrambled sequence (SCR) with no affinity for EGFR [22–24, 31, 34] as control. Eventually, for following the nanovectors biodistribution in vivo, together with the tumor targeting, a near infrared Cy7 was covalently labelled onto the PNPs surface (see Methods and Supplementary Information for details, Figures S1-S3).
For a better understanding of the composition and final destination of the nanovectors, a schematic figure representing the samples is provided (Fig. 1).
The PNPs fabrication starts with the copolymer PLGA-b-PEG-COOH synthesis, following a procedure previously reported [17]. The nanoparticles emulsion is then made by means of the water-in-oil-in-water double-emulsion sonication method [38]; we entrapped Cis-Pt in the hydrophilic core of PNPs, obtaining a final water dispersible formulation. After purification of the obtained Cis-Pt@PNPs, the conjugation of the amino-terminated CL4 (or SCR) was performed via EDC chemistry, by exploiting the superficial residual carboxylic groups onto the PNPs surface, derived from the PEG chains. In order to demonstrate that there is an effective correspondence between the results obtained and the presence of a specific aptamer onto the surface of the nanoparticles, Cis-Pt@PNPs were also conjugated with SCR and therefore used as a negative control (indicated as Cis-Pt@PNPs-SCR).
Cis-Pt@PNPs-CL4 (or Cis-Pt@PNPs-SCR) were then purified and fully characterized by dynamic light scattering (DLS), which revealed particles with diameter equal to 91.2 ± 1.4 nm, a low polydispersity index (PDI = 0.290), and negative ζ-potential value of − 12.2 mV due to unreacted carboxylic acid groups. Overall, all types of PNPs formulations displayed size ranges suitable for IV administration for long-circulating sustained-release features. Size plays a key role in determining the PNPs’ fate upon administration to the body [39]: a diameter of less than 10 nm results in the removal of NPs by renal filtration, while, on the other hand, PNPs with hydrodynamic radii of 200 nm or more show a higher rate of clearance compared to the ones with lower size range [40]. For an effective penetration, the particle diameter should ideally be 70–200 nm as it will sustain longer circulation time and increased accumulation in the target site. Our nanovectors fully met these requirements.
Cis-Pt concentration was determined to be around 450 µM by Microwave Plasma Atomic Emission Spectrometry (MP-AES), a method for rapid and sensitive determinations of Pt concentration. The overall concentration of Cis-Pt@PNPs-CL4 in solution was measured by gravimetric analysis to be 7 mg/ml.
On the other hand, for performing fluorescence-based experiments (such as cell internalization of nanoparticles by confocal microscopy and tumor targeting and biodistribution in vivo by fluorescence reflectance imaging), we fabricated two additional kinds of nanovectors: the BODIPY@PNPs containing the green dye into the PNPs lipophilic portion, and the Cy7@PNPs, in which Cy7 was covalently conjugated to the carboxylic groups onto the nanoparticles’ external surface. Both the systems were also labelled with CL4 or SCR for targeting purposes, purified and completely characterized (see Supplementary Information, Figures S2 and S3). Table 1 exhibits the various properties of fabricated PNPs formulations.
Table 1
PNP formulation
|
PNPs characteristics
|
|
Average size (nm)
|
PDI
|
Zeta potential (mV)
|
Drug/Dye concentration
|
Dry matter amount (mg/ml)
|
Cis-Pt@PNPs
|
90.3 ± 0.3
|
0.254
|
-21.4
|
460.78 µM (0.138 µg/µl)
|
7
|
Cis-Pt@PNPs-CL4
|
91.1 ± 1.4
|
0.290
|
-12.2
|
422.70 µM (0.127 µg/µl)
|
7
|
Cis-Pt@PNPs-SCR
|
118.5 ± 20
|
0.290
|
-11.1
|
455.18 µM (0.136 µg/µl)
|
7
|
BODIPY@PNPs
|
97.6 ± 0.3
|
0.142
|
0.08
|
166 µM
|
8
|
BODIPY@PNPs-CL4
|
131.9 ± 0.2
|
0.185
|
-17.9
|
13.6 µM
|
5
|
BODIPY@PNPs-SCR
|
142.2 ± 1.5
|
0.185
|
-23.4
|
9.7 µM
|
5
|
Cy7@PNPs
|
90.5 ± 0.5
|
0.218
|
-24.9
|
104.16 µM
|
4
|
Cy7@PNPs-CL4
|
107.9 ± 0.3
|
0.350
|
-20.6
|
106.8 µM
|
6
|
Cy7@PNPs-SCR
|
104.2 ± 2.7
|
0.350
|
-20.6
|
105.96 µM
|
6
|
In vitro targeting by aptamer-conjugated nanovectors
To assess whether the CL4 aptamer is able to specifically target the nanovectors to EGFR-positive TNBC cells and enhance their intracellular uptake, we exploited the BODIPY-loaded PNPs decorated with CL4 (BODIPY@PNPs-CL4) or SCR (BODIPY@PNPs-SCR) as a negative control. MDA-MB-231 cells, which represent a well-established model for aggressive TNBC [41–43] and express abundant levels of EGFR [24, 33] (Figure S4A), were incubated with fluorescent nanovectors for different times (from 30 to 60 min) at 37°C and visualized by confocal microscopy (Fig. 2a). As shown, the signal associated with BODIPY@PNPs-CL4 was clearly visible in the cytoplasm at 30 min and further increased in a time-dependent manner. Conversely, a very weak signal was detected with SCR-decorated nanovectors only starting at 50 min, which remains unchanged for up to 60 min of incubation, resulting about 10-fold lower than the signal associated with CL4-targeted nanovectors. Furthermore, in agreement with the high efficacy of aptamer targeting toward EGFR-positive BT-549 TNBC cells [24 and Figure S4A], BODIPY@PNPs-CL4, but not BODIPY@PNPs-SCR, were internalized into these cells as well (Fig. 2b). Finally, further confirming the specificity of the EGFR aptamer [19–21, 24], BODIPY@PNPs-CL4 could distinguish MDA-MB-231 cells from EGFR-depleted MDA-MB-231 cells obtained by a CRISPR/Cas9 approach (Fig. 2c and Figure S4A), demonstrating the excellent selective internalization of CL4-conjugated nanoparticles.
Altogether, these data clearly indicate that the CL4 aptamer specifically delivers PNPs to EGFR-positive TNBC cells, strongly enhancing their intracellular uptake.
In vitro cytotoxicity of aptamer-conjugated and cisplatin-loaded nanovectors
In order to assess the ability of CL4 aptamer-conjugated and Cis-Pt-loaded PNPs to specifically kill EGFR-positive cells, MDA-MB-231 cells were incubated for 72 h with increasing concentrations of the drug, both free and entrapped in PNPs, ranging from 0.1 to 30 µM, and cell viability was determined using an MTT assay. As shown (Fig. 3a and d), Cis-Pt loaded in the nanoparticles, either unconjugated or conjugated with SCR, was almost 2.5-fold more cytotoxic than free drug (P < 0.01, both), with half maximal inhibitory concentration (IC50) values of 10.23 ± 0.54 µM and 8.93 ± 1.29 µM for Cis-Pt@PNPs and Cis-Pt@PNPs-SCR, respectively, and 24.91 ± 3.24 µM for free Cis-Pt. Notably, the conjugation of CL4 aptamer significantly increased the cytotoxicity of the Cis-Pt-loaded nanoparticles that displayed an IC50 of 1.96 ± 0.42, thus approximately 12-fold lower than that of free drug (P < 0.0001) and 5-fold lower than that of Cis-Pt@PNPs-SCR or Cis-Pt@PNPs (P < 0.05). Very limited basal toxicity of unloaded PNPs conjugated to either CL4 or SCR aptamers was observed up to 1 mg/ml (Figure S5), the maximal carrier concentration used in cytotoxicity studies with Cis-Pt loading. These data confirm that PNPs are extremely safe and atoxic [31]. Furthermore, as expected based on the high specificity of the EGFR aptamer, superimposable cell viability curves were obtained upon incubation of CL4-targeted PNPs and untargeted PNPs, both unconjugated and decorated with SCR, onto MDA-MB-231 EGFR-KO cells (Fig. 3b and d). Interestingly, these cells had a response to Cis-Pt treatment comparable to that of parental cells (Fig. 3d), but a higher proliferative potential, as assessed by clonogenic growth rate analysis (Figure S4B), thus suggesting the occurrence of compensatory pathways to EGFR silencing. Then, we performed a short 40-min incubation with Cis-Pt, free or loaded into PNPs, followed by washing and 3 days recovery period. Under these experimental conditions, viability of MDA-MB-231 cells was much less affected by free Cis-Pt (IC50 value of 290.40 ± 19.23 µM) with respect to the 72-h continuous exposure to the drug (Fig. 3c and d). Importantly, IC50 of free Cis-Pt decreased of about 28-fold (P < 0.0001) when the drug was incapsulated into the CL4-equipped PNPs (Cis-Pt@PNPs-CL4, IC50 10.24 ± 1.43), thus indicating that Cis-Pt@PNPs-CL4 strongly improved drug accumulation in cancer cells. Moreover, Cis-Pt@PNPs-CL4 were almost 3-fold more effective than either Cis-Pt@PNPs or Cis-Pt@PNPs-SCR (P < 0.05) (Fig. 3c and d), further confirming the rapid internalization of EGFR aptamer-driven nanoparticles and more efficient killing of TNBC cells than under passive delivery conditions.
Next, we verified that Cis-Pt@PNPs-CL4 are also more efficient in reducing viability of BT-549 cells than free Cis-Pt and control Cis-Pt@PNPs-SCR nanovectors (Fig. 3e). Indeed, even at low drug concentration, at which control nanovectors and free drug did not affect cell viability, the targeted nanovectors caused significant cancer cells killing (approximately 50% inhibition at 2 µM-concentration, P < 0.0001).
Moreover, in order to evaluate whether the 72-h treatment of MDA-MB-231 cells with free Cis-Pt (10 µM) and Cis-Pt-loaded nanovectors decorated either with CL4 or SCR (Cis-Pt-concentration, 2 µM) induced apoptosis, the cell death was measured by flow cytometry analysis after Annexin-V/PI dual staining. As shown, (Fig. 3f), while free Cis-Pt and Cis-Pt@PNPs-SCR had no significant effect, Cis-Pt@PNPs-CL4 treatment significantly increased the apoptotic cell proportion from ~ 13% (Mock treatment) to ~ 56 % (Cis-Pt@PNPs-CL4).
Overall, these data demonstrate the ability of Cis-Pt@PNPs-CL4 to efficiently concentrate Cis-Pt in EGFR-positive TNBC cells and kill them, furthermore by discriminating target cells from non-target cells that do not express EGFR.
In vivo tumor targeting by aptamer-conjugated nanovectors
To evaluate the in vivo tumor targeting efficiency of CL4-conjugated nanovectors, Cy7@PNPs-CL4 and its non-targeting variants, Cy7@PNPs-SCR and unconjugated Cy7@PNPs, were administered intravenously via tail vein (Cy7-concentration, 5 nmol/100 µl injection) into MDA-MB-231 tumor bearing nude mice and non-invasive imaging was performed over 24 h by FRI. As shown (Fig. 4a), Cy7@PNPs-CL4-associated fluorescence signal was readily detected in tumors as early as 30 min after injection, persisted at 1h and then steadily decreased as the time was prolonged up to 24 h, but nonetheless remained higher than both unconjugated Cy7@PNPs and SCR-conjugated PNPs, which were poorly detected in tumors at all experimental time-points. Figure 4b showed that, within groups, only in Cy7@PNPs-CL4 treated mice the corrected signal intensity (SI) at 30 min, 1 h and 3 h was significantly higher than 24 h (P < 0.0001, P < 0.0001 and P = 0.0086, respectively) and at 30 min and 1 h was significantly higher than 3 h (P < 0.01). No other significant differences were detected in the other groups. Moreover, between groups, only Cy7@PNPs-CL4 treated mice had a significantly higher corrected SI compared to both Cy7@PNPs and Cy7@PNPs-SCR treated mice at 30 min (P = 0.022 and P = 0.036, respectively), at 1 h (P = 0.029 and P = 0.041, respectively), at 3 h (P = 0.037, both) and at 24 h (P = 0.019 and P = 0.026, respectively).
The main organs and tumors of the mice were harvested 24 h after PNPs injection for evaluation of biodistribution by ex vivo FRI imaging (Fig. 5). Consistent with in vivo imaging results, Cy7@PNPs-CL4 showed greater intra-tumor accumulation than unconjugated PNPs and SCR-conjugated PNPs (Fig. 5a). Indeed, the normalized mean SI of tumors was significantly different between groups, being in Cy7@PNPs-CL4 group approximately 4.7- and 2.6-fold higher than in Cy7@PNPs-SCR and Cy7@PNPs groups, respectively (P = 0.01 both; Fig. 5b). Conversely, there was no significant difference in normalized mean SI between groups for liver, spleen, kidneys, heart, and lungs (Fig. 5a and 5b). As expected, a high fluorescence signal was observed in the liver and kidneys of all groups accordingly to the detoxification function and the elimination routes for biodegradable nanoparticles of liver and kidney, respectively [44], whereas not substantial accumulation was seen in other organs such as lung, heart and spleen (Fig. 5a).
Further, confirming the in vitro findings, Cy7@PNPs-CL4 were able to discriminate MDA-MB-231 tumors from MDA-MB-231 EGFR-KO-derived tumors (Figure S6), thus indicating the targeting selectivity of the nanovectors for EGFR-positive tumors.
Taken together, these results clearly demonstrate tumor-specific uptake of Cy7@PNPs-CL4 and thus the feasibility of CL4 aptamer-conjugated PNPs as an efficient delivery vehicle in a targeted manner.
In vivo antitumor efficacy of aptamer-conjugated and cisplatin-loaded nanovectors
Given the excellent tumor-specific targeting of CL4-PNPs, we next investigated the in vivo antitumor efficacy of the Cis-Pt-loaded nanovectors in xenograft models of MDA-MB-231 tumor. To this aim, tumor-bearing mice were injected intravenously with free Cis-Pt, Cis-Pt@PNPs-CL4 or Cis-Pt@PNPs-SCR as control nanovectors at day 0, 2, 5, 7, 9, 13, 16, 19, 21. Mice treated with DPBS served as control (Ctrl). Tumor growth was monitored over time (up to 23 days). In order to appreciate the potential effectiveness of our targeted chemotherapy, Cis-Pt@PNPs-CL4 and untargeted Cis-Pt@PNPs-SCR were administered at 0.6 mg/kg-Cis-Pt concentration and their effects compared with that of the same dosage of free Cis-Pt. This dosage was at least 5-fold lower than the minimum reference dosage of Cis-Pt (repeated iv injection of 5 − 3 mg/kg) used to study the antitumor activity in mice bearing MDA-MB-231 subcutaneous xenografts [45, 46]. In this experimental condition (suboptimal concentration of the drug), treatments with either free Cis-Pt or Cis-Pt@PNPs-SCR resulted in a slightly but significantly delayed tumor growth compared to Ctrl treatment (P = 0.0015 and P = 0.018, respectively; Fig. 6a). However, they displayed similar response because of no statistical differences between them, indicating that the non-targeted nanoparticle delivery system was not able to significantly improve the free Cis-Pt efficacy. Conversely, the use of CL4-PNPs as a delivery vehicle for Cis-Pt, resulted in a highly significant inhibitory performance on tumor growth compared to free Cis-Pt (P = 0.0004) and Cis-Pt@PNPs-SCR (P = 0.0009) (Fig. 6a).
This indicates that the presence of the EGFR aptamer on the PNP surface was able to concentrate the Cis-Pt payload at the tumor site, thereby enhancing the anticancer activity of the drug. Importantly, the treatment was well tolerated in vivo, with no significant bodyweight loss and behavioral change in the treated mice throughout the entire study (Fig. 6b).
Next, immunoblot analyses performed on tumor lysates showed a more efficient inhibition of ERK1/2 phosphorylation in the tumors from mice treated with Cis-Pt@PNPs-CL4 than those treated with Cis-Pt@PNPs-SCR or free Cis-Pt (Fig. 6c). Finally, in agreement with in vitro findings, the inhibiting effect of Cis-Pt@PNPs-CL4 was accompanied by a strong activation of caspase-3, a hallmark of apoptosis (Fig. 6d), thus confirming the better efficacy of Cis-Pt when delivered by the EGFR aptamer-conjugated nanovectors.