Core–Shell Distinct Nanodrug Showing On-Demand Sequential Drug

May 29, 2019 - ...
0 downloads 0 Views 11MB Size
www.acsnano.org

Downloaded via BUFFALO STATE on July 18, 2019 at 11:55:10 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Core−Shell Distinct Nanodrug Showing OnDemand Sequential Drug Release To Act on Multiple Cell Types for Synergistic Anticancer Therapy Jinsheng Huang,†,‡,∥ Yongmin Xu,†,∥ Hong Xiao,†,‡ Zecong Xiao,† Yu Guo,§ Du Cheng,† and Xintao Shuai*,† †

PCFM Lab of Ministry of Education, School of Materials Science and Engineering, Sun Yat-sen University, Guangzhou 510275, China ‡ College of Chemistry and Materials Science, Jinan University, Guangzhou 510632, China § Department of General Surgery, The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou 510275, China S Supporting Information *

ABSTRACT: Among various inflammatory factors/mediators, autocrine and paracrine prostaglandin 2 (PGE2), which are abundant in various tumors, promote the proliferation and chemoresistance of cancer cells. Thus, eliminating the cytoprotective effect of PGE2 may strengthen the antitumor effect of chemotherapy. Chemo/antiinflammatory combination therapy requires the programmed activities of two different kinds of drugs that critically depend on their spatiotemporal manipulation inside the tumor. Here, a micellar polymeric nanosphere, encapsulating chemotherapeutic paclitaxel (PTX) in the core and conjugating anti-inflammatory celecoxib (CXB) to the shell through a peptide linker (PLGLAG), was developed. The PLGLAG linker was cleavable by the enzyme matrix metalloproteinase-2 (MMP-2) in the tumor tissue, causing CXB release and turning the negatively charged nanosphere into a positively charged one to facilitate PTX delivery into cancer cells. The released CXB not only acted on cyclooxygenase-2 (COX-2) to suppress the production of pro-inflammatory PGE2 in multiple cell types but also suppressed the expression of the anti-apoptotic Bcl-2 gene to sensitize cancer cells to chemotherapy, thus resulting in a synergistic anticancer effect of PTX and CXB. This study represents an example of using a surface chargeswitchable nanosphere with on-demand drug release properties to act on multiple cell types for highly effective chemo/ anti-inflammatory combination therapy of cancer. KEYWORDS: polymeric micelle, nanodrug, sequential drug release, multicell targeting, combination therapy

S

celecoxib (CXB), could inhibit not only the inflammatory COX-2/PGE2 pathway to influence the hallmarks of cancer but also the anti-apoptosis genes to reduce chemoresistance.1,8,9 Therefore, a combination therapy using a chemotherapeutic drug and a NSAID could be a powerful strategy to alter the characteristic pro-inflammatory environment of the tumor and sensitize cancer cells to chemotherapy.1,10 Although the combination therapy strategy using COX-2 inhibitors and chemotherapeutic drugs has initiated several clinical trials,11−14 anti-inflammatory drugs as adjuvants for chemotherapy in

olid tumors possess heterogeneous structures and a microenvironment regulated by a complex signaling pathway network, such that the combination therapy strategies using different drugs to act on multiple oncotargets have a better chance to improve the therapeutic outcome.1,2 It is well documented that chronic inflammation is one of the common features of cancer,3 where cyclooxygenase-2 (COX2) and a pro-inflammatory mediator, COX-2-derived prostaglandins 2 (PGE2), play critical roles in the maintenance of tumor viability, growth, metastasis, and angiogenesis.4,5 On the other hand, chemotherapy often induces the upregulation of COX-2, PGE2, and anti-apoptotic genes (e.g., Bcl-2), upon which the malignant cells acquire resistance to chemotherapeutic agents.6,7 Previous studies have shown that nonsteroidal anti-inflammatory drugs (NSAIDs), for example, © 2019 American Chemical Society

Received: March 19, 2019 Accepted: May 29, 2019 Published: May 29, 2019 7036

DOI: 10.1021/acsnano.9b02149 ACS Nano 2019, 13, 7036−7049

Article

Cite This: ACS Nano 2019, 13, 7036−7049

Article

ACS Nano

Figure 1. Preparation and in vivo performance of MMP-2-sensitive nanospheres carrying anti-inflammatory CXB and chemotherapeutic PTX (SN-CXB/PTX). Sequential drug release and surface charge switching to act on multiple cell types inside the tumor was expected.

clinical studies are usually administered as free form, making it hard to achieve a desirable synergistic anticancer effect and was even reported to likely lower the anticancer activity of chemotherapeutic drugs due to two drug interactions.15 Additionally, the long-term administration of free NSAIDs may result in severe side effects.16 For combination drug therapy in vivo, a two-in-one codelivery is advantageous over the separate delivery of two drugs in many aspects, including simplified pharmacokinetics, less carrier usage, and better synergistic effect,17−20 which has made the design of nanocarriers for codelivering two different kinds of drugs a hot research topic in recent years. Particularly, nanodrugs based on biodegradable polymers such as aliphatic polyesters and poly(amino acids) have drawn great attention because of their superior performances in controlled drug release, improved biosafety, and easily achievable tumor targeting via convenient chemical modifications.21,22 Because both cancer cells and other stromal cells in the tumor chronic inflammatory environment may be activated to secrete PGE2 to promote tumor growth and chemoresistance,9,23−25 autocrine PGE2 of cancer cells and paracrine PGE2

of stromal cells should be both inhibited by CXB to sensitize cancer cells to paclitaxel (PTX) chemotherapy. In this context, the development of codelivery systems with on-demand sequential drug release properties inside the tumor is extremely important for cancer combination drug therapy. To date, although the well-programmed intracellular release of two drugs (e.g., siRNA and chemotherapeutic drug) acting on the same cancer cells has been intensively investigated,26,27 research on the intratumoral sequential release of two drugs acting on multiple cell types (e.g., both cancer cells and stromal cells) has been very rare.21,22 Moreover, polymeric nanocarriers for the intratumoral sequential release of one chemotherapeutic drug and another NSAID have not been reported yet. On the other hand, the surface charge-switching strategy has shown great potential to improve the tumortargeting drug delivery efficiency of nanocarriers.28 Unlike the tumor-targeting ligand modification strategy that may have adverse effects on the pharmacokinetics of nanodrugs (e.g., shortened circulation time) by altering their size and surface chemistry,29,30 this strategy could make the nanodrugs negatively charged in the bloodstream to prolong circulation 7037

DOI: 10.1021/acsnano.9b02149 ACS Nano 2019, 13, 7036−7049

Article

ACS Nano

Figure 2. Characterizations of the polymer and nanosphere. (a) Prepolymers and CXB-conjugated peptide for the click reaction. (b) Final copolymer for nanosphere preparation. (c) TEM imaging showing the distinct core−shell structure of nanosphere before MMP-2-triggered CXB release, because the sodium salt of CXB increased the electron density of the shell for high TEM visibility. (d) TEM only observed the core of the nanosphere after CXB release, due to the loss of sodium salt in the shell. (e) Nanospheres were negatively charged before MMP2-triggered CXB release but (f) positively charged after CXB release according to ζ potential measurements. (g) HPLC traces of the CXBconjugated peptide with 10 nM MMP-2 treatment at different times in TCNB solution (pH 7.4). The MMP-2-sensitive peptide (peak I) was cleaved gradually by MMP-2 to release CXB-GGPLG (peak II), whereas the MMP-2-insensitive peptide (peak III) remained intact to show one peak unaffected by MMP-2. Scale bars in TEM images represent 100 nm.

delivery into cancer cells. The combinative chemo/antiinflammatory anticancer potency of MMP-2-sensitive nanospheres was explored.

but positively charged inside the tumor to facilitate cancer cell uptake through a tumor microenvironment-responsive structural design.31,32 It may also be a feasible means to improve the drug delivery efficiency in cancer combination therapy. Herein, we describe a polymer-based nanodrug showing ondemand sequential drug release and surface charge switching to act on multiple cell types for synergistic chemo/antiinflammation therapy of cancer (Figure 1). An amphiphilic triblock copolymer of CXB-peptide/mPEG-grafted poly(Lglutamate) (PPLG-g-(CXB-peptide&mPEG)), polyethylene glycol (PEG), and poly(ε-caprolactone) (PCL), abbreviated as PPLG-g-(CXB-peptide&mPEG)-PEG-PCL (PCxbP), was synthesized and then self-assembled into a nanosphere with distinct core−shell structure. The chemotherapeutic paclitaxel (PTX) was physically encapsulated into the core via hydrophobic interaction, while the anti-inflammatory celecoxib (CXB) was conjugated to the shell through a peptide (PLGLAG) linker sensitive to matrix metalloproteinase-2 (MMP-2) overexpressed in solid tumors. The PLGLAG chain could be cleaved inside tumor tissue, which was expected to result in the release of CXB acting on multiple cell types to regulate tumor-associated inflammation and meanwhile a surface charge switching of nanosphere to facilitate PTX

RESULTS AND DISCUSSION Preparation of Polymers and Nanospheres. As peptides with the PLGLAG sequence are sensitive to the MMP-2 being enriched in solid tumors,33,34 a triblock copolymer using the GGPLGLAGG peptide to conjugate CXB was synthesized to prepare MMP-2-sensitive nanocarriers. A triblock copolymer using inert GGLALGPGG peptide to conjugate CXB was also synthesized to prepare MMP-2-insensitive nanocarriers as a negative control. The details to synthesize prepolymers and final amphiphilic triblock copolymers [PPLG-g-(CXB-peptide&mPEG)]-PEG-PCL (PCxbP) via multistep reactions are described in the Supporting Information. Characterizations using 1H NMR, FTIR, UV−vis, and GPC confirmed the successful syntheses of the designed polymers (Figure 2a,b and Figures S1−S11, Supporting Information). The amphiphilic triblock copolymer was composed of a hydrophobic PCL block with a molecular weight of 4 kDa and a hydrophilic [PPLG-g-(CXBpeptide&mPEG)]-PEG diblock with a molecular weight of 7038

DOI: 10.1021/acsnano.9b02149 ACS Nano 2019, 13, 7036−7049

Article

ACS Nano

Figure 3. FRET, ζ potential and drug release analyses of nanospheres in vitro. (a) Fluorescence spectra and (b) surface charges of MMP-2sensitive nanospheres (SN-CXB/FITC/RhoB) in solution containing 10 nM MMP-2. The MMP-2 was set at 10 nM to mimic the tumor microenvironment. (c) Fluorescence spectra and (d) surface charges of MMP-2-insensitive nanospheres (InN-CXB/FITC/RhoB) in solution containing 10 nM MMP-2. (e) CXB release of nanospheres in mouse serum, liver homogenate, and pH 7.4 solutions with or without MMP-2 (means ± SD, n = 3). (f) PTX release of MMP-2-sensitive nanospheres (SN-CXB/PTX) after CXB release at pH 5.0 and different lipase concentrations mimicking lysosomal conditions (means ± SD, n = 3).

sphere, that is, with a spherical morphology and a nanosize of approximately 130 ± 4 nm (Figure 2c). The nanosphere size, as determined by TEM, was slightly smaller than that measured by dynamic light scattering (DLS) (130 ± 4 nm vs 140 ± 5 nm), likely because drying the sample for TEM measurement caused particle shrinkage. Unlike most core− shell structural polymeric micelles whose hydrophilic shells (e.g., PEG) can hardly be seen under TEM,2,17,37 the hydrophilic shell of our micellar nanosphere was highly manifested due to the incorporation of CXB sodium salt with high electron density. The core−shell distinct nanosphere has a shell thickness of 28 ± 2 nm and a core diameter of 75 ± 3 nm according to TEM analysis. Acid−base titration showed that the sulfonamide structure (−SO2NHCO−) conjugating CXB possesses a pKa of 4.5, indicating that it was completely ionized to form CXB sodium salt (−SO2N−Na+CO−) at pH 7.4 (Figure S12b, Supporting Information). Consequently, the nanospheres showed a negative surface charge (ζ = −14.8 ± 1.1 mV), as determined by ζ potential measurement (Figure 2e). The introduction of SS allowed the formation of CXB-free nanoparticles with similar shell properties (ζ potential and MMP-2 sensitivity) comparable to that of CXB-bearing nanoparticles. The pKa values of CXB sodium of CXB-bearing nanoparticles (SNCXB) and sodium carboxylate of the CXB-free nanoparticles (SN) were very close (pKa 4.5 vs 4.8, Figure S12b, Supporting Information). Additionally, the same grafting densities of CXB sodium and carboxylate sodium endowed the SN-CXB/PTX and SN-PTX with similar ζ potentials (Table S3, Supporting Information). Nanoparticles with negative surface charges were found to show a long circulation time, enhancing tumor accumulation in previous studies.31,32,38,39 Indeed, the size of MMP-2-sensitive nanospheres in PBS containing 10% fetal bovine serum (FBS) or in mouse plasma remained almost unchanged over 48 h at 37 °C, implying a high stability and no protein binding in the bloodstream (Figure S13a−d,

42.1 kDa (Table S1, Supporting Information). The CXB content of [PPLG-g-(CXB-peptide&mPEG)]-PEG-PCL was 10.2 wt %, and the PEGylation extent of the PPLG-g-(CXBpeptide&mPEG) block was 66.7%, corresponding to a PEG content of 47.5 wt % in the final copolymer (Table S2, Supporting Information). Instead of the CXB-terminated peptides, sodium succinate (SS)-terminated peptides with or without MMP-2 sensitivity were used to prepare the carrier PPLG-g-(SS-peptide&mPEG)-PEG-PCL containing no CXB (Scheme S10, Supporting Information). Next, the amphiphilic triblock copolymers were selfassembled into micelles in aqueous solution, and PTX was encapsulated in the micelle core via hydrophobic interaction with PCL. Thus, MMP-2-sensitive nanospheres (SN) and MMP-2-insensitive nanospheres (InN) carrying a single drug or two drugs were obtained. As shown in Table S2 (Supporting Information), the CXB loading contents of SN-CXB and InNCXB were 10.2% and 10.5%, respectively. Meanwhile, the PTX loading contents of SN-PTX, InN-PTX, SN-CXB/PTX, and InN-CXB/PTX were 8.9%, 9.2%, 8.6%, and 9.0%, respectively. Because both PCL and PTX possess high hydrophobicity,35,36 a strong hydrophobic interaction might have facilitated a relatively high PTX loading content and efficiency. Their particle sizes and ζ potentials are summarized in Table S3 (Supporting Information). The MMP-2-sensitive and MMP-2insensitive micelles showed approximately the same particle size and ζ potential, implying that the amino acid sequence of the peptide linker had no effect on the copolymer selfassembly. Moreover, the copolymer showed a low critical micellization concentration (CMC) of 28.2 μg/mL (Figure S12a, Supporting Information), indicating high stability of the nanospheres against dilution in the bloodstream, which may be due to crystallization,36 chain entanglement, and the high hydrophobicity of PCL. Under transmission electron microscopy (TEM), the MMP2-sensitive micelle appeared to be a highly uniform nano7039

DOI: 10.1021/acsnano.9b02149 ACS Nano 2019, 13, 7036−7049

Article

ACS Nano

solution were not affected by MMP-2 digestion (Figure 3c,d) because MMP-2 was unable to cleave the insensitive peptide linker for FITC/CXB release; thus, FRET between FITC and RhoB exists all the time. Quantitative determination of CXB release from nanospheres in solution (pH 7.4) containing 10 nM MMP-2 showed consistent results. CXB release from InN-CXB/PTX was not detected regardless of whether the solution contained 10 nM MMP-2. By contrast, CXB release from SN-CXB/PTX and SN-CXB appeared much different (Figure 3e and Figure S14a, Supporting Information). CXB release from SN-CXB/ PTX and SN-CXB was not detected when MMP-2 was absent in the solution. However, CXB release was rapid in solution containing 10 nM MMP-2, reaching a plateau of approximately 88% cumulative release for SN-CXB/PTX within just 8 h (Figure 3e). Thus, 8 h was regarded as the time for the nanospheres to complete the CXB release. It was noted that the time to complete the CXB release seemed longer than that needed to complete surface charge switching upon MMP-2 digestion (Figure 3b), likely because it took extra time for CXB to diffuse out of the dialysis bag. The MMP-2-triggered CXB release from SN-CXB/PTX was in line with the results of DLS, TEM, and HPLC analyses (Figure 2c−g). Because the MMP-2 concentration is below 0.5 nM in serum,41,42 the CXB release from the nanosphere was evaluated in serum and buffered solution containing 0.5 nM MMP-2. As shown in Figure 3e, CXB release was