Evaluation of cell penetrating peptide coated Mn:ZnS nanoparticles for paclitaxel delivery to cancer cells

This work aimed at formulating paclitaxel (PTX) loaded cell penetrating peptide (CPP) coated Mn doped ZnS nanoparticles (Mn:ZnS NPs) for improved anti-cancer efficacy in vitro and in vivo. The developed PTX loaded Mn:ZnS NPs with different CPPs (PEN, pVEC and R9) showed enhanced anti-cancer effect compared to bare PTX, which has been validated by MTT assay followed by apoptosis assay and DNA fragmentation analysis. The in vivo bio-distribution and anti-cancer efficacy was studied on breast cancer xenograft model showing maximum tumor localization and enhanced therapeutic efficacy with R9 coated Mn:ZnS NPs (R9:Mn:ZnS NPs) and was confirmed by H/E staining. Thus, R9:Mn:ZnS NPs could be an ideal theranostic nano-carrier for PTX with enhanced the rapeutic efficacy toward cancer cells, where penetration and sustainability of therapeutics are essential.

SCIEnTIfIC REPORTS | (2018) 8:1899 | DOI: 10.1038/s41598-018-20255-x potential for gene delivery applications 17 . Thus combining CPP and Mn:ZnS NPs could be beneficial for improved anti-cancer therapy as the Mn doping could enhance the ROS 7 mediated apoptosis along with the Paclitaxel (PTX), a first line chemoagnet against many cancers such as breast (metastatic) 18 , ovarian 19 and cervical cancers. Since UV fluoresced Mn:ZnS NPs has no live imaging capability, we further doped it with IR-780 dye to enhance the in vivo bio-imaging. Since the dye can be doped easily in the lattice space of Mn:ZnS NPs, the fluorescence of the dye can be retained for in vivo applications. Unlike the polymeric NPs, the Mn:ZnS NPs would render co-loading efficientlty as there might be minimum interaction beween PTX and IR-780 dye as the latter is inside the core space lattices while the former is attached on the CPP molecules on the shell.
Thus our current work focussed on modification of Mn:ZnS NPs for effective tumor cell penetration of partially soluble drugs like PTX for improving therapeutic outcome in breast cancer xenograft model along with in vivo imaging. The Mn:ZnS NPs were doped with IR-780 dye to have in vivo live imaging and coated with CPPs, maintaining their particle size in the optimum range for the PTX delivery. The basic research question we addressed in this study is: How can CPP modified IR-780:Mn:ZnS NPs act as an ideal theranostic agent for anticancer therapy in vitro and in vivo? We hypothesize that since Mn:ZnS NPs being cell illuminant and CPP, cell penetrable, they could improve the apoptosiss by PTX and ROS generated by the doped NPs, whereas the IR-780 in Mn:ZnS NPs core will enhance in vivo live imaging in cancer cells as strategized in Fig. 1.

Results
Synthesis, characterization, and PTX formulations with CPP coated Mn:ZnS NPs. Different CPP modified Mn:ZnS NPs have been prepared using a simple aqueous route in which CPP could act as a stabilizing and surface passivating agent. The CPP:Mn:ZnS NPs had almost spherical shapes, with size ranging from 100-150 nm, as visualized in the bio-TEM ( Fig. 2B-D). Unlike the Mn:ZnS NPs ( Fig. 2A), the CPP modified NPs showed distinct contrast in the bio-TEM, whereas the Mn:ZnS NPs appeared to have a darker core, while the CPP modification rendered a cloudy surface appearance.
In the case of the PEN modified NPs, the surface was completely covered compared to the other CPP molecules (Fig. 2B). The pVEC bound onto the Mn:ZnS NPs with an irregular covering (Fig. 2C). R9 showed entirely different attachment on the core, with maximum R9 residues attached on the side of Mn:ZnS NPs (Fig. 2D).
The DLS analysis showed slightly increased particle size, likely due to the swelling of CPP chains in aqueous solutions. The size varied from 130-230 nm for the control Mn:ZnS NPs, whereas the PEN: Mn:ZnS NPs showed 140-280 nm and pVEC modification had 140-240 nm (Fig. 2E). The R9:Mn:ZnS NPs showed 110-240 nm size by the DLS (Fig. 2E). The SEM and DLS analyses confirmed that even after PTX loading, the CPP modified NPs could retain their size in the optimum range for effective drug delivery (Fig. S1). The zeta values were almost same except for the PTX:R9:Mn:ZnS NPs which could be due to their better PTX loading (Fig. 2F). The PL (photo luminescence) studies confirmed obvious changes in PL intensity before and after CPP modification(Supplementary Fig. S2E-H and M-P). The PL intensity increased significantly higher with the R9 modified NPs among the samples (Supplementary Fig. S3A). The CPP binding efficiency on the Mn:ZnS NPs was maximum for PEN (~80%), followed by pVEC (~40%) and R9 (~20%) respectively (Fig. 2G). On the other hand, PTX encapsulation efficiency (EE) was ~100% with R9 modified NPs followed by Mn:ZnS NPs (~82%), PEN (~53%), pVEC (~20%). (Fig. 2H) The possible PTX interaction with CPP modified NPs could be via the anchoring mechanism ( Supplementary Fig. S3B) 20 .
In vitro cyto-compatibility and cellular trafficking of CPP coated Mn:ZnS NPs. As shown in   In vitro PTX release. It was done in two different pHs (5 and 7.4). The pH 5 was selected as it is the endosomal pH 21 whereas pH 7.4 is the normal pH of the blood stream. As shown in the Supplementary Fig. S5, the PTX was released much faster in pH 5 within 48 h, whereas under the neutral pH, the release was extended upto 2 days resulting in 100% PTX release. The PEN:Mn:ZnS NPs have shown slow release and ~80% release was there within 7 days in acidic pH, while ~40% release was there in 7 days under pH 7.4. The pVEC:Mn:ZnS NPs had fast PTX release (almost 100%) within 3 h at pH 5, whereas the neutral pH extended the release 3 days till 100%. On the other hand, the PTX was released slowly from R9 modified NPs, with only ~60% release at pH5 in 7 days. The release % was significantly low as ~30% under neutral condition after 7 days suggesting that the R9 could release PTX more controllably.    Fig. S13). This could be associated with the effective stabilization of CPP coating on PTX:Mn:ZnS NPs. However, control PTX:Mn:ZnS NPs started degrading after day 21 (Fig. 5A,B), and colloidal stability indices were not in detection range (Fig. 5C), indicating their poor serum stability.
In vivo biodistribution and preclinical therapeutic evaluation using 4T1 breast tumor xenograft model. The IR-780 modified NPs showed similar particle size with unmodified NPs (Supplementary Fig. S14).
The fluorescence intensity of IR-780 loaded Mn:ZnS NPs was much higher than control IR-780 (Supplementary Fig. S15A and B). Compared to the IR-780 dye, (which retained only for 3 days after i.v. injection), the R9/ Mn:ZnS NPs showed high retention in tumor even after 7 days. (Fig. 6A) This was confirmed by ex vivo imaging. (Fig. 6B-G) In vivo efficacy studies revealed maximum tumor efficacy for PTX:R9:Mn:ZnS NPs (Fig. 7A,B) (Fig. 7C). Additionally, as shown in Fig. 7D-O, there were no significant pathological changes in any of the major organs (Fig. 7R). This suggests that these formulations might be good candidates for bio-applications. Further, Fig. 7R clearly indicates that the treatment did not affect the body weight of animals, assuring nontoxic behavior, as observed in the H and E staining studies.

Discussion
Optimum particle size is important for efficient drug delivery applications 22 . All the synthesized CPP modified Mn:ZnS NPs showed spherical shapes and the possible interaction could be via the metal-polymer interaction 23 . The CPP modified Mn:ZnS NPs showed improved fluorescence as well. The high fluorescence with the PTX loaded Mn:ZnS NPs might be due to the higher anchoring of PTX on the Mn:ZnS core (Supplementary Fig. S3B). This could lend Mn:ZnS NPs greater stability, thereby high fluorescence among the other samples 24 . The PTX anchoring could actually favor surface passivation of the Mn:ZnS NPs to have better fluorescence 25 . The high PTX EE of R9 with low surface passivation capability could be attributed to strong metal:polymer interaction with the PTX molecules. In addition, the exposed core in the case of R9/Mn:ZnS NPs could additionally favor anchoring of PTX onto it, giving more space for PTX binding.
The interaction between CPPs and Mn:ZnS could be either via the strong metal to ligand interaction or by the electrostatic interaction between Mn:ZnS and CPPs. It is known that the amino acid sequence in CPPs is electron rich, allowing donation of electrons to the vacant d-shells in the Mn:ZnS leading to strong bonding between  26 . The difference in the CPP binding on the Mn:ZnS could be attributed to the difference in amino acid sequence and electron density in it. The PEN has amino acid sequence of RQIKIWFQNRRMKWKK, which may have enhanced electron density, and thus the interaction would be higher than that of pVEC and R9, respectively.
The cytocompatibility as well as the cellular assessment showed that the developed formulations were nontoxic, thus favoring applications for in vivo animal studies. The higher uptake of R9 modified NPs in all the tested cell lines could be attributed to their arginine abundant amino acid sequence, which could enable selective entry into cancer cells 27 . The improved anti-cancer efficacy of the PTX:R9:Mn:ZnS NPs could be due to the higher intracellular localization capability. In the case of PTX, it is highly recommended to have an intracellular availability of PTX to show its potency against cancer cells. PTX is one of several cytoskeletal drugs that can specifically target tubulin. PTX can affect the mitotic spindle assembly, chromosome segregation, and cell division 28 . The R9:Mn:ZnS NPs has benefits of maximum PTX loading capability (Supplementary Table S1) which might be enhanced in an acidic environment, lending higher toxicity to the treated cancer cells. The in vitro PTX release showed controlled and sustained release from R9:Mn:ZnS NPs which could be related to its high anti-tumor efficacy in vivo. Additionaly the doped ZnS NPs could increase the ROS generation thereby improving the apoptosis along with the PTX 7 . Unlike the other CPP modified samples, the PTX:R9:Mn:ZnS NPs showed a very different stability pattern, retaining stability even up to 70 days. The particle size and colloidal satiability indices were maintained throughout the 70 days, as presented in Fig. 5aA-C. This high stability could assure its in vivo therapeutic efficacy than other samples.
In vivo live imaging is critical for drug delivery systems.Unlike the undoped ZnS NPs, Mn:ZnS NPs has many advantages such as improved cytocompatibility, enhanced ROS generation etc, however it cannot be used for live imaging as its fluorescence is in the UV range. Obviously the tissue penetration of UV light is very poor compared to the NIR light 29 . Infact practical use of UV light for live imaging is hazardous as UV radiation is classified as a "complete carcinogen" because it is both a mutagen and a non-specific damaging agent and has properties of both a tumor initiator and a tumor promoter 30 . Thus with the inherent fluorescence of Mn:ZnS NPs, it is difficult to do the live imaging in the animal models. Therefore we used IR-780 dye which was doped in the NPs where it can stay well in the space lattice of NPs without undergoing any degradation or leaching. Additionally, since the IR-780 is doped in the core material there would be minimum interaction with the co-laoded PTX as they are attached to the CPP shell. This was seen in the bio distribution study as most of the R9:Mn:ZnS NPs CPP could retain in the tumor even after a week than other samples. This could be associated with its enriched arginine composition that facilitate intratumoral entry 31 . Since PTX has poor solubility, the R9:Mn:ZnS NPs may enhance its tumor localization to improve efficacy. In addition, the R9 coated Mn:ZnS NPs were nontoxic to the internal organs as evident in the histopathology analysis (Fig. 7P). The enhanced in vivo theranostic property of PTX:R9:Mn:ZnS NPs could be due to the better tumor localization capability, improved ROS generation thereby better apoptois along with the PTX effect. Eventhough there have been few reports regarding the applicability of Mn:ZnS NPs for drug delivery, most of them are confined to in vitro studies without any pre-cliniacl analysis on animal models. The studies by Yu et al., (2013) showed that their Mn:ZnS NPs could only be attached on the surface of the tested cells than being taken up inside the cells which might reduce the therapeutic outcome 32 . Another study by Zhao et al., (2017) used PTX loaded Mn:ZnS NPs lacks the therapeutic analysis 12 . To the best of our knowledge no paper has been reported with CPP functionalization for IR-780:Mn:ZnS NPs to have simultaneous imaging and therapy, thus our current study could open up new channels for using IR-780 doped Mn:ZnS NPs for cancer drug delivery appications.

Conclusions
Different CPP coated Mn: ZnS NPs were made to evaluate their potency as drug delivery and imaging agents against cancer cells. The size of the developed NPs would be ideal for in vivo applications. The in vitro studies including PL spectroscopy suggest that the fluorescence has been improved with the CPP modification, and the NPs were well localized in all the tested cell lines. Among the CPP modified PTX encapsulated NPs, the R9 modified PTX loaded NPs were found to be more cytotoxic to the cancer cells, which has been proven by MTT, live/ dead, DNA fragmentation, and apoptosis assay. The bio-distribution analysis showed that the R9 modified IR-780 doped Mn:ZnS NPs accumulated significantly higher than other CPP modified NPs, which was validated with ex vivo imaging. The in vivo analysis showed highest anticancer potency for PTX: R9/Mn:ZnS NPs compared to the other samples without any toxicity to the internal organs. These preliminary results suggest that the PTX:R9/ Mn:ZnS NPs may be a better PTX delivery agent against cancer cells, compared to the other CPP modified NPs.

Materials and Methods
Materials. CPPs (PEN, pVEC, R9, and their FITC modified versions) were custom made by Peptron Company, Republic of Korea. ZnCl 2 , MnSO 4 were purchased from Junsei Chemical Co. Limited, South Korea. Na 2 S was purchased from Sigma Aldrich, USA. (PEN, pVEC and R9). The Mn:ZnS NPs were synthesized according to previous protocols with slight modifications 8 . Briefly, 0.1 M ZnCl 2 was made in 4 mL of distilled water (pH was adjusted to 5, inorder to dissolve ZnCl 2 properly) and then mixed with an equal volume of 0.01 M MnSO 4 solution (pH 7). To this solution, a 0.1 M Na 2 S was added drop-wise under continuous probe sonication (~2 min) untill there was white opalescent coloration (indicative of Mn:ZnS NPs formation) followed by two times centrifugation at 13,000 rpm for 30 min, to separate remnant solutes from the NPs. For surface passivation with CPPs, the CPPs were made at 0.01 mM concentration. Each of the CPPs were directly added into the preformed Mn:ZnS NPs (1 mg/mL) and surface passivated for about 30 min followed by centrifugation at 13,000 rpm (30 min) to remove the unwanted solutes.

Synthesis of Mn:ZnS and surface passivation with CPPs
CPP binding efficiency of modified Mn:ZnS NPs. Detailed protocls have been given in File S1. Long term stability studies of NPs. 10% FBS was used for this. The samples were made as per the literature 33,34 . Precisely, 0.5 mL samples were mixed with 1 mL of 10% FBS, and incubated at 37 °C for predetermined time intervals. The samples were collected after each time for DLS/Zeta/PDI determination.

Synthesis of PTX
In vivo studies. All animal experimental protocols and methods complied with the principles of Laboratory and Animal Care established by the National Society for Medical Research and were approved by the Korea Advanced Institute of Science and Technology (KAIST) on Use and Care of Animals. Details of in vivo studies have been given in Supplementary File S6. The protocols for Histopathology was done according to the previous work 35,36 . Analytical determinations. Refer Supplementary File S7.
Statistical analysis. Statistical significance of differences values were calculated by a two-tailed Student's t-test. For multiple comparisons, an ANOVA analysis was performed. A value P < 0.05 was considered to be statistically significant.