Construction and in vitro evaluation of pH-sensitive nanoparticles to reverse drug resistance of breast cancer stem cells

ATRA-g-PBAE prodrug polymer was obtained from previous experiments by our research group [9]. PBAE and ATRA-g-PBAE were respectively synthesized by michael and esterification reaction. The EDC·HCl and DMAP were added to the ATRA solution. Then, the PBAE solution was dropped into the mixture and reacted for 48 h. The purification process involved is the use of cold ether to precipitate excess ATRA. Then, to remove the EDC·HCl and DMAP, the fluid was transferred to a dialysis bag for dialysis for 24 h. The final product ATRA-g-PBAE was obtained by vacuum drying. Cy5.5-PBAE was prepared by the same method, only by replacing ATRA with Cy5.5.

SchB/AP NPs were prepared from the solvent volatilization method [20]. Briefly, a certain mass of ATRA-g-PBAE prodrug polymer carrier and SchB (20% carrier mass) were precisely weighed. The organic phase was mixtured by dissolving the carrier and SchB in acetone. According to the experimental design scheme, the organic phases with different carrier concentrations (mg/mL) and the feeding drug (SchB) ratio (%) were prepared. The above organic solution was slowly dripped into 15 mL of 0.5% poloxamer 188 aqueous phase. Then the solution continued stirring at 25 ℃ until the organic solvent evaporated. After filtering through a 0.22 μm filter membrane, the SchB/AP NPs solution with blue opalescent light was obtained. The preparation of C6/AP NPs is the same as the above, only C6 replaced SchB.

The factor and level of Central Composite Design Response Surface Model (CCD RSM) were determined according to the preliminary experiment. ATRA-g-PBAE carrier concentration (mg/mL) and the feeding drug (SchB) ratio (%) were set as A and B, respectively. Particle size (Y₁) and encapsulation rate (EE) (Y₂) were used as response parameters. The NPs were prepared according to the preparation method in Sect. 3.1 for other conditions. Meanwhile, the response surface figure was drawn with the results data, and the optimal scheme was predicted. The experimental design scheme is shown in Table 1.

1 mL of the prepared SchB/AP NPs solution was centrifuged at speed (10,000 rpm) for 60 min. The obtained supernatant was absorbed and methanol was added for demulsification. The solution was further demulsified by an ultrasonic cell crusher. Afterward, the demulsified solution was placed in a 5 mL volumetric bottle and filled with methanol to the scale. The liquid obtained by filtration through a 0.22 μm filter determined the SchB content as W_a. In addition, 1 mL prepared SchB/AP NPs solution was taken for all above operations except centrifugation, and SchB content was determined, denoted as W_b. The centrifugal precipitate above was taken and then its total mass was weighed after vacuum drying, denoted as W. All SchB contents were determined by high performance liquid chromatography (HPLC) (Shimadzu, 3100, Japan). The EE% and the drug loading capacity (LC)% were computed by the following formula:

In order to obtain the appearance of the SchB/AP NPs, the transmission electron microscope (TEM) (JEOL JEM, 4378, Japan) was used to observe the morphology. The specific operation method was as follows: a little SchB/AP NPs solution prepared above was added to the 200-mesh copper net by drops. After 3 min, the excess liquid was absorbed by filter paper. After drying at room temperature, the morphology of the NPs was characterised by TEM.

The particle size and size distribution of SchB/AP NPs were determined by the Marvin particle size analyzer (Marvin, 2900, US). One drop of the NPs solution was added to the size cell. Finally, the average size and particle size distribution were recorded. The same method was used to measure the zeta potential.

The optimal preparation of SchB/AP NPs was put into the given dialysis bags (Mw = 3,500). Moreover, the free SchB acted as the control group. Dialysis bags were transferred to 50 mL PBS solution containing 0.5% (m/w) polysorbate 80 at pH 5.5, 6.8, and 7.4. Water bath oscillations at 37 ℃ were used, samples were taken at different times. The sample solution was determined by the concentration of SchB by HPLC. The cumulative release was depicted and the cumulative release curve was plotted in accordance with following formula: where, E_r, V_e, C_i, V₀, C_n and m_drug, were the cumulative release of SchB, the displaced volume of PBS, the concentration at the ith sampling (µg/mL), release medium volume (mL), the concentration at the nth sampling (µg/mL) and SchB content in drug-loaded NPs (mg), respectively.

The SchB/AP NPs were placed in serum contained DMEM at 37 ℃ to investigate the stability of the NPs. The mixture was shaken in a shaker, and 1 mL of the sample was aspirated at 0, 6, 12, 36, 24 and 48 h for particle size determination. The particle size was measured according to the particle size determination method in Sect. 3.4.

The SchB was measured by HPLC fitted with a Prominence LC-20AD at 220 nm. The experimental result was detected on a Diamonsil C18 column (4.6 mm × 250 mm, 5 μm). The mobile phase A and B were respectively 15% phosphoric acid aqueous solution and 85% methanol. Moreover, the flow rate was 0.8 mL /min. Measured method of SchB concentration section “3.3 and 3.5” was according to those conditions above.

In vitro cytotoxicity was detected by MTT. MS cells were seeded in 96 wells at a density of 7 × 10³ cells per well and incubated at 37 ℃ for 24 h at 5% CO₂ in an incubator. The free solution of the drug was a blank control group. The pharmaceutical solution was taken into certain wells in the experimental group. Cell culture grade 0.05% DMSO was added to the wells to aid dissolution. Each concentration was made three parallel while leaving blank wells. After that, the cell culture plates were placed in the incubator for 48 h. Finally, the absorbance value was detected at 490 nm with a multifunctional microplate reader (JD-SY96A, JingDao, China). The IC₅₀ at which the reversal agent SchB was added to certain wells was denoted as IC₅₀ (A/S). The Reversal Factor (RF) was calculated according to the following formula:

To verify the cellular uptake performance of the carriers of the NPs, Cy5.5-g-PBAE NPs were used to replace ATRA-g-PBAE NPs. Besides, to testify the cellular uptake ability of SchB /AP NPs, C6/AP NPs were used instead SchB /AP NPs. The NPs preparation method was referred to the Part 3.1. MS cells were digested and seeded in 6-well cell culture plates at a density of 2.5 × 10⁴. C6/AP and Cy5.5-g-PBAE NPs were diluted in the serum-free medium that were added to the wells and incubated for 0.5, 2 and 4 h, respectively. Pre-chilled PBS was added at the end of the incubation to stop cell uptake. PBS washed the cells twice to remove drug interference and then placed them under a fluorescence microscope (Laika, 5699, Germany) to observe cell uptake. Finally, flow cytometry (Beckman, CytoFLEX, China) was used to determine the intensity of cells fluorescence at different times.

Coverslips were placed in 6-well plates, and MS cells were seeded on them at 10⁵ cells/well. After overnight culture, 2 mL Cy5.5-g-PBAE NPs solution was added into each well and incubated for 0.5, 1 and 2 h. Intracellular lysosomes were labeled with 200 µL Lyso-Tracker Green lysosome fluorescent probe. After 30 min of incubation, cells were fixed with 4% paraformaldehyde, and placed under confocal laser scanning microscope (CLSM) (Olympus, FV3000, Germany) to observe the intracellular drug delivery of MS cells.

The experiment was repeated 3 times and the results are reported as mean ± SD (standard deviation). The experimental design and data processing of CCD-RSM was completed by Design Export 11. The analysis of experimental data was made using GraphPad Prism 11. The Student’ s t-test between 2 groups or more was with ANOVA test when more than 3 groups. Significance at *p < 0.05 or **p < 0.01 was acceptable.

RSM has been widely used in process optimization schemes, including the preparation of nanoparticles. As a means of experimental design, CCD is often used in conjunction with RSM (CCD RSM) to optimize the formulation of NPs preparation. CCD RSM was used to optimize the preparation of SchB /AP NPs. The experimental data were fitted with multiple linear and second-degree polynomial regression, and the results were as follows:

The response surface in Fig. 1 was obtained by fitting the equation. Figure 1a is the change in particle size caused by factors A (carrier concentration) and B (feeding drug ratio), and Fig. 1b is the change of EE caused by factors A and B. It can be seen Fig. 1a that particle size gradually decreases and then increases with the increase of A and B. As shown in Fig. 1b, EE increased first and then decreased with the rise in values of A and B. The reason is that after the concentration of the carrier was too high, the organic phase was opposed to dispersing evenly in the water phase, and agglomeration occurred [21]. As the feeding drug ratio was too high, the carriers could not carry more SchB, resulting in the decline of EE. Thus, carrier concentration and feeding drug ratio should be maintained at a medium level to obtain lower particle size and higher EE. Finally, according to the software prediction, the optimal conditions are as follows: carrier concentration was 22.78 mg/mL, and feeding drug ratio was 19.32%. The optimal particle size is 146.5 nm and the EE is 73.89%. The best prescription verification results of SchB/AP NPs prepared by the optimal scheme were as follows: the particle size was 135.1 ± 0.96 nm, the LC was 5.01 ± 0.15%, the EE was 80.68 ± 1.15%. The actual result is similar to the size and EE predicted by the software.

Typically, the size of nanoparticles is designed to be around 100–200 nm. This is due to the presence of the enhanced permeability and retention effect (EPR), which means that the capillary epithelial cells near the tumor cells are not tightly arranged, resulting in pore sizes between 100 and 200 nm, while the capillary vessels in normal tissues are extremely tightly arranged [22]. Therefore, the nanoparticles will be selectively enriched in the tumor area.

As shown in Fig. 2a, the TEM of SchB/AP NPs could be seen from the figure that the NPs were spherical-likeand monodisperse. The result showed that the hydrophobic ATRA-g-PBAE could condense into spherical form under the influence of hydrophobic action, and the stronger the hydrophobic action was, the smaller the particle size was. Monodisperse nanoparticles indicate that there is no aggregation between particles, which is extremely important for the distribution of nanoparticles in vivo [23]. Figure 2b shows the particle size distribution, which showed a particle size of 134.97 ± 0.25 nm. The results of the two characterization methods are consistent and similar to the size obtained by the optimal scheme. Thus, the SchB/AP NPs prepared by the optimal scheme have good pharmaceutical properties.

Sufficiently small particle size is beneficial to NPs to be passively targeted to the tumor and enriched within the tumor under the EPR effect [24]. The EPR effect relies on the high vascular permeability and low lymphatic outflow of solid tumors to achieve passive targeting and long-term retention of nanoparticles at the tumor site.

To realize the study of drug release by NPs in tumor cells, pH 7.4, 6.8 and 5.5 were selected to simulate the pH of normal physiology in vivo, tumor microenvironment and the lysosomes, respectively [13, 25]. The release curves of SchB/AP NPs in different pH release media were shown in Fig. 2c. Free SchB was rapidly released from the dialysis bag to the medium solution within 12 h, indicating that the dialysis bag had minimal effect on the release of SchB. Within 48 h, the release of SchB reached 80% at pH 5.5. At pH 6.8, SchB release was finally 30%.

The pH of the lysosomes of tumor cells is 5.5, and experimental results show that the drug is rapidly released at this acidity. PBAE is a pH-sensitive polymer, which can undergo strong protonation under acidic lysosomal conditions [26]. At the same time, the protonation causes abundant chloride ions and water molecules to flow into the lysosome, resulting in osmotic pressure increasing and e lysosome rupturing [27]. Accordingly, after the nanoparticles are protonated in the lysosome, the released drugs SchB can be quickly dispersed into the cytoplasm and take effect.

The stability of NPs in blood circulation is the primary condition for the passive targeting of NPs to tumor sites through EPR effects. Drugs are expected to have more excellent stability in the blood to maintain a longer duration of action in the body. To investigate the stability of the SchB/AP NPs, the NPs were placed in the medium containing serum for a total of 48 h. In Fig. 2d, the size of the NPs changed little within 48 h. This was indicated that SchB/AP NPs had weak interaction with serum proteins and showed a stable state.

The strong stability of SchB/AP NPs in serum may be because the NPs were not adsorbed by the proteins in serum and maintained a stable nanostructure. The stability of the nanoparticles ensures that the injected nanoparticles do not accumulate or precipitate in the blood circulation, thus maintaining their properties and drug activity. In addition, stable nanoparticles can achieve multiple functions: long circulation in the body, better-targeting ability, toxic side effects reduction, and therapeutic effect improvement [28].

The inhibitory effects of the free SchB group and SchB/AP NPs on the proliferation of MS cells were shown in Table 2. The inhibitory effect of SchB/AP NPs (IC₅₀: 10.54 ± 1.29 µg/mL) on MS cells was slightly smaller than that of free SchB (IC₅₀: 7.31 ± 1.12 µg/mL).

This difference was explained by in vitro release results: the different cytotoxic effects of NPs and free groups were caused by the incomplete release of SchB in NPs within 48 h. Even though free SchB is highly toxic, free drugs are difficult to dissolve in water and lack an effective target site for accumulation in the body. In this case, the drug cannot effectively achieve the toxic effects of MS cells. It is worth noting that the IC₅₀ (13.43 ± 0.65 µg/mL) of MCF-7 cells administered with AP NPs was significantly increased compared with the IC₅₀ in MS cells previously reported by our research group. The drug resistance of MS cells is many times stronger than that of MCF-7 cells, and strong drug resistance is an important reason for the reduction of toxicity [29].

Li et al. used Vitamin E derivatives as drug resistance reversal agents to treat drug-resistant cancer cells together with chemotherapy drugs. They found that chemotherapy drugs were less toxic in resistant cancer cells than in normal cancer cells. Studies have found that the VE derivatives, while inhibiting the activity of drug-resistant cells, can reduce drug resistance and play a stronger inhibitory role [30]. Therefore, after loading SchB, NPs have great potential to reverse MDR and enhance the effect of ATRA on Pin1 inhibition.

The uptake of Cy5.5-g-PBAE NPs and C6/AP NPs by MS cells was detected by fluorescence microscopy. In Fig. 3a and b, the fluorescence intensity was low within 0.5 h, and the amount of drug in the cell was minimal. When incubation time reached 2 h, enhanced red fluorescence intensity could be observed due to partial drug entry into cells. At 4 h, the fluorescence intensity in the cell was strengthened, and the amount of drug entering the cell increased. This proved that drug intake of MS cells was a time-dependent behavior with the extension of incubation time.

To demonstrate that the encapsulated drug can be transported into the cell through the carrier, the uptake of C6/AP NPs at different times was detected. C6 is similar to SchB in hydrophobic properties and can replace SchB as an encapsulated fluorescent drug. The green fluorescence intensity in MS cells also increased with the extension of incubation time.

Compared with free C6, nanoparticle-encapsulsated C6 showed excellent uptake performance. Li et al. incubated cells with free C6 and loaded C6 nanoparticles respectively, and found that weak fluorescence of free C6 could be observed only after 4 h, while the NPs group showed obvious green light within 1 h [31]. Nanocarriers can transfer hydrophobic molecules into cells more rapidly and release molecules into the cytoplasm with pH sensitivity. Compared with drug molecules, nano-loaded drug particles can increase the accumulation of local drug concentrations in tumor cells, increase bioavailability, and thus have higher therapeutic potential.

Fig. 3c, weak yellow fluorescence appeared in 0.5 h. This indicated that the color of Lyso-Tracker Green coincidated with Cy5.5. At 2 h, the red fluorescence increased significantly, and the yellow fluorescence was more obvious. However, the intracellular yellow fluorescence was significantly weaker in MS cells treated with chloroquine. Chloroquine is a drug that disrupts the acidic environment of the lysosome, thereby raising the pH of the lysosome. Therefore, chloroquine interfered with the release of acid-responsive drugs in lysozyme.

The results of the above uptake experiments have proved that blank NPs can carry SchB for cell uptake. The ingested NPs enter the early endosome and then the lysosome. Since PBAE is pH-sensitive, protonation of PBAE will cause lysosome rupture and further escape release [32].