Junyi
Chen
ab,
Yadan
Zhang
a,
Zhao
Meng
a,
Lei
Guo
a,
Xingyi
Yuan
ab,
Yahan
Zhang
a,
Yao
Chai
ab,
Jonathan L.
Sessler
*b,
Qingbin
Meng
*ade and
Chunju
Li
*bc
aState Key Laboratory of Toxicology and Medical Countermeasures, Beijing Institute of Pharmacology and Toxicology, Beijing 100850, P. R. China. E-mail: nankaimqb@sina.com
bDepartment of Chemistry, Center for Supramolecular Chemistry and Catalysis, Shanghai University, Shanghai 200444, P. R. China. E-mail: sessler@cm.utexas.edu
cKey Laboratory of Inorganic–Organic Hybrid Functional Material Chemistry, Ministry of Education, Tianjin Key Laboratory of Structure and Performance for Functional Molecules, College of Chemistry, Tianjin Normal University, Tianjin 300387, P. R. China. E-mail: cjli@shu.edu.cn
dKey Laboratory of Advanced Energy Materials Chemistry (Ministry of Education), College of Chemistry, Nankai University, Tianjin, 300071, China
eKey Laboratory of Natural Resources and Functional Molecules of the Changbai Mountain, Affiliated Ministry of Education, College of Pharmacy, Yanbian University, Yanji, Jilin, 133002, China
First published on 3rd June 2020
Most cancer chemotherapy regimens rely on the use of two or more chemotherapeutic agents. However, achieving the best possible dosing of the individual drugs can be challenging due to differences in metabolism, uptake, and clearance among other factors. Here we describe a supramolecular strategy for achieving drug delivery in which the loading ratio of two active components is easily defined. Specifically, we report the formation of aggregates comprised of self-assembled amphiphiles between carboxylatopillar[6]arene (CP6A) and an oxaliplatin (OX)-type Pt(IV) prodrug (PtC10). The association constant (Ka) for the underlying host–guest interaction at pH 7.4 ((1.16 ± 0.03) × 104 M−1) is an order of magnitude higher than at pH 5.0 ((1.73 ± 0.15) × 103 M−1). A second chemotherapeutic, doxorubicin (DOX), may be encapsulated in the resulting vesicles (PtC10⊂CP6A) to give a supramolecular combination chemotherapeutic system DOX@PtC10⊂CP6A. Drug release studies served to confirm that PtC10 and DOX are released in acidic environments. Support for a synergistic antiproliferative effect relative to PtC10 + DOX came from cellular studies of DOX@PtC10⊂CP6A using the human liver hepatocellular carcinoma (HepG-2) cell line. In vivo studies revealed that DOX@PtC10⊂CP6A is not only able to retard tumor growth efficiently but also reduce drug-related toxic side effects in BALB/c nude mice bearing HepG-2 subcutaneous tumor xenografts. These favorable findings are attributed to the formation of a ternary complex that benefits from an enhanced permeability and retention (EPR) effect in vivo while allowing for the pH-based release of PtC10 and DOX at the tumor site.
DDS currently being used for combination chemotherapy include liposomes,9–11 polymeric nanoparticles,12–14 organic–inorganic hybrid materials,15–17 among other approaches.18–20 As detailed below, we propose a novel supramolecular combination chemotherapy system that permits the co-delivery of two recognized chemotherapeutics, namely oxaliplatin (OX) (in the form of a Pt(IV) prodrug) and doxorubicin (DOX). The present strategy is attractive in that drug release is triggered efficiently as the result of pH responsive host–guest interactions.21–26
OX, a diaminocyclohexane analogue of cisplatin, constitutes the third platinum drug approved by the US FDA.27–29 In spite of its better tolerability compared to other Pt(II) compounds, including cisplatin and carboplatin, OX still suffers from low selectivity and dose-limiting side effects.30–32 Nevertheless, OX in combination with fluorouracil,33,34 capecitabine,35,36 doxorubicin37,38 and other agents,39,40 is either used clinically or the subject of ongoing clinical studies. For example, OX and DOX are used in combination for the management of patients suffering from hepatocellular carcinoma.41,42 Although not yet benefiting from FDA approval, Pt(IV) complexes have attracted attention recently as potential Pt(II) prodrugs. As a generally rule, Pt(IV) species are less reactive (labile) than the corresponding Pt(II) congeners, which reduces concerns involving systemic toxicity. These prodrugs are thought to undergo reduction to release an active platinum(II) species that then mediates an antitumor cytotoxic effect through inter alia binding to DNA.43–45
In the present study, a supramolecular amphiphilic complex derived from carboxylatopillar[6]arene (CP6A) and an OX-based Pt(IV) prodrug (PtC10) were used to construct nano-scale aggregates that encapsulate DOX effectively. The resulting supramolecular combination chemotherapy system (DOX@PtC10⊂CP6A) was found to target tumor tissues passively through an enhanced permeability and retention (EPR) effect. However, they were then seen to collapse in the lower pH lysosomal environment after cellular uptake to release the two drugs, PtC10 and DOX (Scheme 1). Evidence for a synergistic antitumor effect was then seen. This work thus serves to highlight the inherent promise of smart supramolecular self-assembled amphiphiles as DDS for combination chemotherapy.
Scheme 1 Chemical structures of CP6A and PtC10, schematic illustration of the preparation of DOX@PtC10⊂CP6A, and the proposed mechanism for drug release. |
Aqueous mixtures of PtC10⊂CP6A produced using [CP6A]/[PtC10] = 1:2 per the above, exhibited a notable Tyndall effect (Fig. 2a). Such a finding provides support for the existence of nanoparticles. Transmission electron microscopy (TEM) images proved consistent with the formation of hollow supramolecular vesicles with diameters ranging from 50 nm to 90 nm (Fig. 2c and S11a†). Results obtained from dynamic laser scattering (DLS) measurements were consistent with an average particle size of 91.3 nm (Fig. 2b). The thickness of the outer wall of the PtC10⊂CP6A nanoparticle was about 6 nm, as inferred from TEM studies (Fig. 2c). This value is consistent with the bilayer molecular length of PtC10⊂CP6A simulated by Chem3D (Fig. S12†). The resulting supramolecular vesicles were predicted to possess a bilayer structure with two hydrophilic carboxylate shell layers, as well as a core layer containing the hydrophobic alkyl chains. This prediction reflects an appreciation that the cyclohexyl group present in PtC10 would be bound within the cavity of the CP6A receptor as the result of solvatophobic interactions. The CP6A moiety would then serve as a hydrophilic head group, while the ancillary alkyl ligands would serve as hydrophobic tails. The net result is a set of tadpole-like self-assembled amphiphiles (Scheme 1). The zeta potential of PtC10⊂CP6A was determined to be −30.6 mV (Fig. S13†), leading us to suggest that electrostatic repulsion could facilitate the stabilization of supramolecular vesicles when CP6A is combined with PtC10 at pH = 7.4.
As noted above, lowering the pH from 7.4 to 5.0 served to decrease the interaction between PtC10 and CP6A. To probe whether the resulting PtC10⊂CP6A vesicles also displayed pH responsiveness, the solution pH was adjusted to 5.0. This led to a loss of the Tyndall effect, which reappeared when the pH was readjusted back to 7.4 (Fig. 2a). This switching is taken as evidence that mixing PtC10 and CP6A at a pH of 7.4 followed by time-dependent aggregation leads to formation of pH-responsive supramolecular vesicles.
Adding DOX to an aqueous solution of PtC10⊂CP6A led to a change from colorless to light red (Fig. S14†). This was taken as a preliminary indication that DOX was successfully encapsulated into PtC10⊂CP6A vesicles to form a two-drug construct DOX@PtC10⊂CP6A. Incorporation of DOX into vesicles was accompanied by changes in the zeta potential. Specifically, after loading, the zeta potential decreased from −30.6 to −27.3 mV, an effect ascribed to uptake of the positively charged DOX (Fig. S15†). Changes in the morphology and size distribution of the presumed DOX@PtC10⊂CP6A DDS were observed, as inferred from TEM and DLS measurements. As shown in Fig. 2b, d and S10b,† upon treatment with DOX the hollow vesicles ascribed to PtC10⊂CP6A were replaced by ca. 100 nm diameter nanoparticles with dark interiors. DLS studies gave an average diameter of 122 nm for the presumed DOX@PtC10⊂CP6A constructs. Importantly, no change in the size of the DOX@PtC10⊂CP6A ensembles was seen when they were allowed to stand in double-distilled water for 3 days (Fig. S16†); this was taken as evidence of their high stability, at least under these conditions.
The size range for DOX@PtC10⊂CP6A was considered to augur well for potential biological use. Previous studies have shown that particles in the range of tens to hundreds nanometers often display favorable pharmacokinetic characteristics. For instance, they typically accumulate within tumor tissues as the result of an EPR effect; this, in turn, can increase the therapeutic efficiency and reduce the toxic side effects of nanoparticle-based treatment protocols.9–11 Accordingly, efforts were made to explore further whether DOX@PtC10⊂CP6A could be used to effect the co-delivery of the two bound drugs (DOX and PtC10) to cancer cells in vitro and to tumor targets in vivo.
High-performance liquid chromatography (HPLC) was used to determine accurately the concentration of the two drug components present in DOX@PtC10⊂CP6A. First, appropriate calibration curves were derived (Fig. S17a and b†). Using these curves and the peak area obtained from diluted samples, the encapsulation efficiency of PtC10 and DOX was calculated to be 83.8% and 25.8%, respectively. This corresponds to a molar ratio of 3.25. Fortuitously, this matches well the relative dosages in clinical use, namely OX = 130 mg m−2 and DOX = 60 mg m−2 or a molar ratio of 3.17. Synergy effects matching those seen for OX + DOX were thus expected for DOX@PtC10⊂CP6A.
Prior to carrying out biological tests with DOX@PtC10⊂CP6A, we sought to test whether OX and DOX would be released as the pH was lowered. As noted above, the carboxylate moieties of CP6A are partially protonated under acidic conditions. This leads to a weakening of the interaction between CP6A and PtC10. and effective release of the two components (DOX and PtC10).
It is well known that lysosomes are acidic organelles, typically characterized by a pH of 5.0 or lower.50,51 The drug release behavior of DOX@PtC10⊂CP6A was thus investigated at pH 5.0. As shown in Fig. 3a, approximately 7% of the bound DOX was released from DOX@PtC10⊂CP6A over the course of 24 h when this construct was placed in a dialysis bag at pH 7.4 and allowed to equilibrate. In contrast, a cumulative release of about 80% within 24 h was seen at pH 5.0. Similar release behavior was seen for PtC10; at pH 7.4 the cumulative release of PtC10 was only about 6% over the course of 24 h but about 70% at a pH of 5.0 under otherwise identical conditions (Fig. 3b).
Fig. 3 Time dependent release of (a) DOX and (b) PtC10 from DOX@PtC10⊂CP6A in PBS at different pH (mean ± SD, n = 3). Concentrations at any given time point were determined by HPLC. See the ESI for details.† |
Fluorescence spectroscopy was used to investigate further the release behavior (Fig. S18†). No fluorescence signal could be observed for aqueous pH 7.4 mixtures of PtC10⊂CP6A over the spectral range corresponding to the DOX-based emission. A weak fluorescence signal was seen for DOX@PtC10⊂CP6A, a finding ascribed to the encapsulation of DOX. Upon adjusting the solution pH to 5.0, the fluorescence intensity of DOX was enhanced. This increase is taken as evidence that DOX@PtC10⊂CP6A undergoes pH-dependent vesicle collapse with concomitant release of DOX. In control studies, DOX@PtC10⊂CP6A was treated with Triton X-100 so as to achieve the complete release of DOX thus allowing comparisons with the release thought to be triggered by lowering the pH. Taken together, these results lead us to suggest that DOX@PtC10⊂CP6A may have a role to play as a pH-responsive co-delivery system.
The cellular uptake and intracellular drug release features of DOX@PtC10⊂CP6A were then investigated using confocal laser scanning microscopy (CLSM) and the HepG-2 cell line. 4′,6-Diamidino-2-phenylindole (DAPI) was used to stain the cell nucleus blue. Upon incubation with DOX@PtC10⊂CP6A for 2 h, a weak red fluorescence ascribed to DOX was observed in HepG-2 cells (Fig. 4c). These red dots were largely colocalized with DAPI. Moreover, the fluorescence intensity increased as the incubation time increased. This is as expected for a DDS system that releases DOX in a time-dependent manner. The presumed cellular internalization was further studied using flow cytometry (Fig. 4d and S20†). The uptake of DOX@PtC10⊂CP6A in HepG-2 cells was reduced to less than 35% upon incubation at 4 °C, as measured by flow cytometry. This finding lends credence to the conclusion that the observed internalization is mediated primarily by an energy-dependent endocytosis process. Negligible uptake inhibition was seen upon co-incubation with 5-(N-ethyl-N-isopropyl) amiloride (EIPA), a macropinocytosis inhibitor. This leads us to suggest that there is minimal involvement of this potential uptake pathway.
We also investigated the effect of sucrose and chlorpromazine (CP). Both agents act as inhibitors of clathrin-coated vesicle formation. Previous studies have served to confirm that nanoparticles between 100 and 200 nm in diameter, the size of DOX@PtC10⊂CP6A, are subject to clathrin-mediated endocytosis.54,55 It was found that the cellular uptake of DOX@PtC10⊂CP6A as inferred from flow cytometry was reduced in a statistically meaningful way in the presence of sucrose or CP. The effect of the lysosome function inhibitor, ammonium chloride (AC), was also tested. It was found that the uptake of DOX@PtC10⊂CP6A into HepG-2 cells was inhibited by about 50% in the presence of AC. This result was taken as evidence that DOX@PtC10⊂CP6A enters HepG-2 cells in part through the lysosomes.
The normalized tumor weight was assessed for the various groups (Fig. 5c). The average tumor weight of the glucose control group (0.51 ± 0.08 g) was 55% higher than that of the OX group (0.32 ± 0.07 g) or DOX group (0.32 ± 0.05 g). The tumor weight of the OX + DOX and DOX@PtC10⊂CP6A groups were 0.20 ± 0.05 g and 0.15 ± 0.09 g, respectively. Thus, a large and statistically significant reduction in tumor regrown was seen for these two group relative to the control or OX and DOX alone.
Immunohistochemical analyses, including hematoxylin and eosin (H&E) staining and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assays, were used to assess the anti-tumor efficiency of the various groups (Fig. 5e). Imaging of H&E-stained tumor tissue from the control group revealed spindle shapes and unbroken nuclei, features that are characteristic of rapid tumor growth. As compared to the control group, images from both the OX + DOX and DOX@PtC10⊂CP6A groups revealed a loss of nuclei. Furthermore, treatment with either OX + DOX or DOX@PtC10⊂CP6A was found to induce a greater level of TUNEL-positive cells. These results were taken as further evidence that both free OX + DOX or DOX@PtC10⊂CP6A were effective at mediating an antitumor response.
In clinical use, both OX and DOX can induce a number of toxicity-related side reactions, including nausea, vomiting, thrombocytopenia, leukopenia, and in the case of DOX, cardiotoxicity. In the present study, the change in body weight of tumor-bearing mice after administration was taken as a surrogate for acute systematic toxicity. As shown in Fig. 5b, mice treated with free OX suffered a barely significant body weight loss from 21.7 ± 1.2 to 18.0 ± 1.2 g. The body weight of the DOX group decreased from 20.3 ± 0.8 to 15.6 ± 1.0 g. Mice administrated OX + DOX suffered a body weight loss from 20.7 ± 1.2 to 12.8 ± 1.4 g. In contrast, almost no body weight changes were observed for the mice treated with DOX@PtC10⊂CP6A (non-statistically significant decrease from 20.8 ± 1.6 to 20.1 ± 1.6 g). These favorable findings were taken as support for the proposition that DOX@PtC10⊂CP6A acts as a supramolecular DDS system and enhances the free drug concentration at the tumor tissue as the result of an EPR effect and site specific release of OX and DOX. In normal biological environments a large Ka is expected to favor PtC10⊂CP6A formation and preclude substantial drug loss or leakage. However, in the acidic environment of lysosomes and solid tumors, dissociation of PtC10⊂CP6A and release of DOX and PtC10 is expected to occur. The above whole animal studies are fully consistent with this design expectation.
Further support for the low toxicity inferred for DOX@PtC10⊂CP6A came from histological analyses of major organ slices, including those of the heart, liver, spleen, lung and kidney of the mice used in the above studies (Fig. S21†). Compared with the control group, severe splenic toxicity and notable cardiotoxicity was observed in the OX + DOX group. Evidence of characteristic inflammation and necrosis in splenocytes and cytoplasmic relaxation in cardiomyocytes was also seen. In contrast, treatment with DOX@PtC10⊂CP6A induced much lower organ toxicity as inferred from the corresponding histological analyses. On this basis, we propose that the supramolecular DDS combination chemotherapy strategy embodied in DOX@PtC10⊂CP6A can be used decrease the undesirable side effects of OX and DOX while maintaining good antitumor efficacy. To the extent this favorable augury translates into the clinic, it is expected to provide a significant and salutary benefit for patients.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/d0sc01756f |
This journal is © The Royal Society of Chemistry 2020 |