Article Cite This: Mol. Pharmaceutics XXXX, XXX, XXX−XXX
pubs.acs.org/molecularpharmaceutics
Cyclic RGD-Peptide-Functionalized Polylipopeptide Micelles for Enhanced Loading and Targeted Delivery of Monomethyl Auristatin E Min Qiu, Xiuxiu Wang, Huanli Sun, Jian Zhang, Chao Deng,* and Zhiyuan Zhong* Biomedical Polymers Laboratory and Jiangsu Key Laboratory of Advanced Functional Polymer Design and Application, College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, People’s Republic of China
Mol. Pharmaceutics Downloaded from pubs.acs.org by UNIV OF SUNDERLAND on 09/27/18. For personal use only.
S Supporting Information *
ABSTRACT: Monomethyl auristatin E (MMAE) is an extremely potent peptide drug that is currently used in the form of antibody drug conjugates (ADCs) for treating different cancers. ADCs are, however, associated with low drug conjugation, immunogenicity, small scale production, and high costs. Here, cRGD-functionalized polylipopeptide micelles (cRGD-Lipep-Ms) were explored for enhanced loading and targeted delivery of MMAE to HCT-116 colorectal tumor xenografts. Interestingly, cRGD-Lipep-Ms achieved an MMAE loading content of 5.5 wt %, which was 55fold higher than that of poly(ethylene glycol)-b-poly(D,L-lactide) micelles. MMAE-loaded cRGD-Lipep-Ms (MMAE-cRGD-LipepMs) showed a small hydrodynamic size of 59 nm, minimal drug leakage in 10% FBS, and efficient uptake and superb antiproliferative activity in αvβ5-overexpressing HCT-116 tumor cells. Remarkably, MMAE-cRGD-Lipep-Ms displayed over 10-fold better toleration than free MMAE in mice and completely suppressed growth of HCT-116 colorectal tumor xenografts. These polylipopeptide micelles have appeared to be an attractive alternative to ADCs for targeted delivery of potent peptide drugs. KEYWORDS: polypeptide, micelles, monomethyl auristatin E, targeted delivery, cancer therapy
■
INTRODUCTION Monomethyl auristatin E (MMAE) is an extremely potent peptide drug and is often linked to a monoclonal antibody to treat different cancers.1−3 Currently, one MMAE antibody drug conjugates (ADCs), brentuximab vedotin (Adcetris) has been approved to treat relapsed Hodgkin lymphoma and systemic anaplastic large cell lymphoma.2,3 In addition, more than 30 MMAE-based ADCs are now under clinical trials to treat lymphoma, leukemia, and solid tumors and display potency up to 200 times of that of vinblastine.4,5 Nevertheless, ADCs are generally associated with low drug conjugation (drug-to-antibody ratio: 2−4), potential immunogenicity, small scale production, and high costs.6,7 In addition to ADCs, polymeric micelles represent another clinically validated platform for poorly water-soluble chemotherapeutic agents.8−12 As compared to ADCs, polymeric micelles provide a number of advantages such as easy manufacture, nonimmunogenicity, and high drug loading.13−18 In particular, polypeptide micelles are one of the most promising candidates due to their intrinsically versatile architecture and structure, which facilitate their interaction with therapeutic drugs via hydrophobic interactions, metal coordination, and electronic interactions, resulting in improved drug loading and superior stability.19−26 It is surprising to note © XXXX American Chemical Society
that given its clinical impact, there is no report on polymeric micelles for delivery of MMAE. Recently, we reported that polylipopeptide micelles and polymersomes developed from poly(ethylene glycol)-b-poly(α-aminopalmitic acid) (PEG− PAPA) allowed facile loading of docetaxel (DTX) and doxorubicin hydrochloride, leading to pronounced tumor growth suppression of B16F10 melanoma and A549 human lung tumor xenografts, respectively.27,28 In the present study, we aimed to evaluate whether cRGDfunctionalized polylipopeptide micelles (cRGD-Lipep-Ms) can be used for stable and efficient loading and targeted delivery of peptide drugs like MMAE. It was hypothesized that MMAE would be stably loaded into the core of cRGD-Lipep-Ms via the “like dissolves like” principle, i.e., lipophilic polypeptide core dissolves lipophilic peptide drugs (Scheme 1). Here, we used the cRGD peptide as a model targeting ligand that is known to strongly bind to αvβ3 and αvβ5 integrins.29−31 Several reports have shown that HCT-116 cells overexpress αvβ5 integrins.32,33 Remarkably, our results show that MMAEReceived: May 12, 2018 Revised: September 8, 2018 Accepted: September 13, 2018
A
DOI: 10.1021/acs.molpharmaceut.8b00498 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics
Scheme 1. Illustration of cRGD-Functionalized Polylipopeptide Micelles (cRGD-Lipep-Ms) That Achieve Enhanced and Stable Loading of MMAEa
Via the “like dissolves like” principle, i.e., lipophilic polypeptide core dissolves lipophilic peptide drug.
a
dialysis tube was directly submerged in dialysis medium (25 mL). The release studies were carried out under constant shaking at 37 °C. The released amount of Cy5-MMAE from Lipep-Ms at all time points was measured by fluorometry. Maximum-Tolerated Dose (MTD) Studies. The mice were handled under protocols approved by the Animal Care and Use Committee of Soochow University. Nude mice were intravenously (i.v.) injected with a single dose of MMAEcRGD-Lipep-Ms and free MMAE via the tail vein, and their body weights were measured daily for 10 days. The dosage of free MMAE was 0.2 mg/kg, and the dosages of MMAE-cRGDLipep-Ms were fixed at 1, 2, 3, and 4 mg of MMAE equiv./kg. The MTD was denoted as the highest dose at which non unacceptable toxicity (mortality, over 15% body weight loss, behavioral abnormality, etc.) was presented in the mice within the experimental period. In Vivo Pharmacokinetics and Biodistribution. Pharmacokinetics behavior was assessed by intravenous (i.v.) injection of MMAE labeled with Cy5 (Cy5-MMAE) and Cy5-MMAE-loaded micelles (Cy5-MMAE-cRGD-Lipep-Ms) to Balb/c mice at a Cy5-MMAE dosage of 2 nmol. At predetermined time points, blood samples were collected, and the amount of Cy5-MMAE was measured by fluorometry. The elimination half-life time (t1/2β) was calculated by fitting the experimental data using the exponential decay 2 model of Origin 9: y = A1 × exp(−x/t1) + A2 × exp(−x/t2) + y0 and then taking t1/2β = 0.693 × t2. Area under the curve was calculated using GraphPad Prism software. For in vivo biodistribution studies, the HCT-116 subcutaneous tumor-bearing mice were i.v. injected with Cy5-MMAEloaded micelles. The fluorescence images of mice were acquired using a near-infrared fluorescence imaging system (IVIS Lumina II) at predetermined time points. To further quantify the amount of Cy5-MMAE, the tumor blocks and major organs were homogenized in 0.5 mL of Triton X-100 (1%) with a homogenizer (IKA T25). Cy5-MMAE extracted using DMF was determined using a fluorometer.
cRGD-Lipep-Ms exhibit over 10-fold better toleration than free MMAE in mice and effective growth suppression of αvβ5 integrin-overexpressed HCT-116 colorectal tumor without inducing pronounced systemic toxicity.
■
EXPERIMENTAL SECTION Preparation of Peptide-Encapsulated cRGD-LipepMs. MMAE-cRGD-Lipep-Ms was obtained at a theoretical drug loading content (DLC) of 10 or 20 wt % by a solvent exchange method. Typically, a DMF solution of 20 mol % cRGD-PEG-b-PAPA (Mn = 6.0−4.2 kg/mol, Mw/Mn = 1.23) and 80 mol % mPEG-b-PAPA (Mn = 5.0−3.5 kg/mol, Mw/Mn = 1.14) at a polymer concentration of 5.0 mg/mL and a prescribed amount of MMAE were added dropwise to phosphate buffer (PB, 10 mM, pH 7.4) under stirring. The obtained solution was extensively dialyzed against PB buffer for 8 h to remove free drug and DMF. MMAE-loaded PEG− PDLLA (Mn = 5.0−5.0 kg/mol, Mw/Mn = 1.30) micelles were prepared through the same method except that DMSO was used instead of DMF. The amount of MMAE was determined by HPLC (Waters 1525) with UV detection at 220 nm. The mobile phase consisted of acetonitrile and KH2PO4 buffer (10 mM, pH 6.0) (4/6, v/v). Similarly, carfilzomib (CFZ) was encapsulated in cRGD-Lipep-Ms with a theoretical DLC of 5 to 20 wt %. The amount of CFZ was determined by HPLC with UV detection at 210 nm using a mobile phase of acetonitrile and KH2PO4 buffer (10 mM, pH 6.0) (6.5/3.5, v/ v). The DLC and drug loading efficiency (DLE) of cRGDLipep-Ms were calculated according to the following formulas DLC (wt.%) = (weight of loaded peptide/total weight of polymer and loaded peptide) × 100 DLE (%) = (weight of loaded peptide/weight of peptide in feed) × 100
In Vitro Release of MMAE. The release behavior of MMAE from Lipep-Ms was investigated in PB at 37 °C using a dialysis method, either with or without 10% FBS. Briefly, Cy5MMAE-cRGD-Lipep-Ms (0.5 mL) following transfer to a B
DOI: 10.1021/acs.molpharmaceut.8b00498 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics In Vivo Antitumor Efficacy. The in vivo anticancer activity of MMAE-loaded Lipep-Ms was evaluated using a subcutaneous HCT-116 human colon cancer tumor model. The tumor-bearing mice were divided into five groups (six mice/group) and i.v. injected with MMAE-cRGD-Lipep-Ms (0.1 and 0.2 mg of MMAE equiv./kg), MMAE-Lipep-Ms (0.1 mg of MMAE equiv./kg), free MMAE (0.1 mg/kg), and PBS. All formulations were administered via the tail vein every 4 days (four injections in total). The volume of the tumor blocks was determined using the following formula: V = LW2/2 (L: the length of tumors; W: the width of tumors). The relative tumor volume and body weights were normalized by their initial ones. Statistical Analysis. Data are presented as mean ± SD. The difference between groups was assessed using the one-way analysis of variance (ANOVA). *p < 0.05 was considered significant, and **p < 0.01 and ***p < 0.001 were considered highly significant.
interactions between the peptide drugs and polypeptide backbone. The drug−carrier compatibility and miscibility were reported to play critical roles on drug loading and drug delivery in vivo.34−36 It should be noted, however, that most current clinical ADCs displayed a low average drug-to-antibody ratio (DAR) of 2−4,5 which equals a low loading content of 0.95−1.90 wt %. Enhanced loading of MMAE in Lipep-Ms would greatly reduce the amount of vehicles and dosing volume of nanomedicines. High drug loading as an important property for nanomedicines has been actively pursued.37−39 Notably, cRGD-Lipep-Ms showed also a high loading of carfilzomib (CFZ, a tetrapeptide drug), achieving a CFZ loading content of 13.4 wt % (Table S1), which was over 10fold higher than the typically reported value for polymeric micelles and liposomes.40−42 As revealed by DLS and TEM, MMAE-cRGD-Lipep-Ms had a small size of ca. 59 nm (Figure 1a,b), which was close to that of blank micelles, further signifying the strong intermolecular interaction between MMAE and PAPA segments in the micellar core. MMAEloaded nontargeted Lipep-Ms (MMAE-Lipep-Ms) obtained from PEG-b-PAPA alone exhibited a similar hydrodynamic size (Figure S1). Interestingly, MMAE-cRGD-Lipep-Ms exhibited an absence of burst release and less than 9.2% release of MMAE in 48 h under physiological conditions (pH 7.4, 37 °C), no matter with or without 10% FBS (Figure 1c), indicating that MMAE-cRGD-Lipep-Ms is robust and has low drug leakage. These polylipopeptide micelles have appeared to be an appropriate nanoplatform for efficient and stable loading of peptide drugs. Cellular Uptake and Antiproliferative Activity of MMAE-cRGD-Lipep-Ms. Cy5-MMAE was used to visualize the cellular uptake behaviors of micelles in αvβ5-overexpressing HCT-116 tumor cells. Confocal studies displayed strong and widespread Cy5 fluorescence inside HCT-116 cells following incubation with Cy5-MMAE-cRGD-Lipep-Ms (Figure 2a). Though Cy5 fluorescence was clearly detected in the cells treated with nontargeted Cy5-MMAE-Lipep-Ms and free Cy5MMAE, the intensity of fluorescence was comparably weaker. Flow cytometry also showed remarkable internalization of Cy5-MMAE-cRGD-Lipep-Ms (Figure 2b), which was over 2fold higher than the nontargeted Cy5-MMAE-Lipep-Ms and free Cy5-MMAE controls (Figure S2). Similar improvement of cellular uptake was also observed for Cy5-MMAE-cRGDLipep-Ms in αvβ3-overexpressing A549 lung cancer cells (Figure S3), indicating that MMAE-cRGD-Lipep-Ms has a comparable affinity toward αvβ3- and αvβ5-overexpressing cancer cells. The cRGD peptide has been reported to promote
■
RESULTS AND DISCUSSION Preparation and In Vitro Drug Release of MMAEcRGD-Lipep-Ms. MMAE is a small amphipathic peptide drug, which renders its physical loading into nanocarriers a great challenge. Interestingly, our results showed that cRGD-LipepMs co-self-assembled from PEG-b-PAPA and cRGD-PEG-bPAPA (4/1, mol/mol) and achieved decent MMAE loading contents of 3.3 and 5.5 wt % at theoretical MMAE loading contents of 10 and 20 wt %, respectively (Table 1). In sharp Table 1. Characterization of MMAE-Loaded cRGD-LipepMs and PEG−PDLLA Micelles DLC (wt %) micelles
theory
determineda
DLE (%)
MMAE-cRGDLipep-Ms
10 20 10 20
3.3 5.5 0.04 0.10
31.2 23.5 0.4 0.4
MMAE−PEGPDLLA
sizeb (nm)
PDIb
± ± ± ±
0.15 0.17 0.19 0.17
58 59 54 56
1.2 1.4 0.3 0.4
a
Determined by HPLC. bDetermined by DLS.
contrast, PEG-b-PDLLA micelles which are under investigation in clinical trials, revealed trivial MMAE loading levels of 0.04 and 0.10 wt % under otherwise the same conditions. The greatly enhanced loading of MMAE in cRGD-Lipep-Ms is likely via the “like dissolves like” principle, i.e., lipophilic polypeptide core dissolves lipophilic peptide drug through the strong intermolecular hydrogen bond and hydrophobic
Figure 1. Characterization of MMAE-cRGD-Lipep-Ms. (a) Hydrodynamic size determined by DLS. (b) TEM photograph. (c) In vitro drug release in PB with or without 10% FBS (n = 3). C
DOI: 10.1021/acs.molpharmaceut.8b00498 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics
Figure 2. Cellular uptake and antiproliferative activity of MMAE-cRGD-Lipep-Ms. CLSM images (a) and flow cytometry (b) of HCT-116 cells after 4 h of incubation with Cy5-MMAE-loaded micelles or free Cy5-MMAE. (Cy5-MMAE concentration: 2 nM). Scale bars, 25 μm. (c) MTT assays of MMAE-loaded micelles or free MMAE toward HCT-116 cells. The cells following 4 h of incubation with drug were cultured in a fresh medium for 44 h.
uptake of different polymeric micelles in αvβ3 and αvβ5 integrin-overexpressing cancer cells like U87MG glioma tumor cells, B16 melanoma cells, and A549 lung cancer cells.28,43−45 In contrast, no difference in cellular uptake was discerned for Cy5-MMAE-cRGD-Lipep-Ms and Cy5-MMAELipep-Ms in αvβ5/αvβ3 negative MCF-7 cells (Figure S4), supporting the active targeting effect of cRGD-Lipep-Ms. The cellular uptake of MMAE-cRGD-Lipep-Ms was significantly reduced for both HCT-116 and A549 cells when changing the temperature from 37 to 4 °C (Figures S2 and S3). The nontargeted MMAE-Lipep-Ms showed reduced cellular uptake, though to a less extent, at 4 °C, as pinocytosis is also energy-dependent. The fact that no significant difference in cellular uptake was observed for MMAE-cRGD-Lipep-Ms and MMAE-Lipep-Ms at 4 °C, indicating that receptormediated endocytosis does not play a role for MMAE-cRGDLipep-Ms at low temperature. Similar results were also observed for αvβ3 (cRGD) and low-density lipoprotein (Angiopep) mediated cellular uptake of polymeric micelles.45,46 MTT assays demonstrated that blank polylipopetide micelles (0.1−1.0 mg/mL) with or without a cRGD ligand were noncytotoxic to HCT-116 cells (Figure S5). In contrast, MMAE-cRGD-Lipep-Ms was highly potent to HCT-116 cells with a half-maximal inhibitory concentration (IC50) of 7.10 nM, which was about 4.2- and 1.9-fold lower than that of free MMAE and MMAE-Lipep-Ms, respectively (Figure 2c). Improved cytotoxicity was also shown for MMAE-cRGDLipep-Ms in αvβ3-overexpressing A549 cells (Figure S6). This
significantly enhanced antitumor potency of MMAE-cRGDLipep-Ms confirms that it can be efficiently taken up by HCT116 and A549 cells, and MMAE is quickly released into the cytoplasm. In Vivo Pharmacokinetics, Tolerability, and Biodistribution of MMAE-cRGD-Lipep-Ms. The pharmacokinetics studies displayed a prolonged circulation time of Cy5-MMAEcRGD-Lipep-Ms in healthy Balb/c mice with an elimination half-life (t1/2β) of 2.33 h (Figure 3a). In contrast, free Cy5MMAE was quickly eliminated following injection. Cy5MMAE-cRGD-Lipep-Ms revealed 67.3-fold higher AUC than free Cy5-MMAE, supporting that cRGD-Lipep-Ms can significantly improve blood retention of MMAE. Notably, in vivo near-infrared fluorescence imaging showed clearly that cRGD-Lipep-Ms efficiently delivered Cy5-MMAE to the HCT-116 tumor (Figure 3b). Strong Cy5-MMAE fluorescence was observed in the tumor at 12 and 24 h postinjection. In comparison, mice treated with nontargeted Cy5-MMAELipep-Ms showed comparably weaker tumor Cy5 fluorescence at all time points, though rather strong tumor accumulation was also observed at 24 h postinjection. We further quantified Cy5-MMAE in the HCT-116 tumor and major organs at 6 h postinjection. The results corroborated an enhanced Cy5MMAE tumor accumulation of 4.7%ID/g for Cy5-MMAEcRGD-Lipep-Ms, which was approximately 2.3-fold and 5.3fold higher than that for Cy5-MMAE-Lipep-Ms and free Cy5MMAE, respectively (Figure 3c). Therefore, cRGD-Lipep-Ms D
DOI: 10.1021/acs.molpharmaceut.8b00498 Mol. Pharmaceutics XXXX, XXX, XXX−XXX
Article
Molecular Pharmaceutics
Figure 3. In vivo pharmacokinetics, biodistribution, and tolerability assays of MMAE-cRGD-Lipep-Ms. (a) Pharmacokinetics behavior of Cy5MMAE-cRGD-Lipep-Ms and free Cy5-MMAE in healthy Balb/c mice (Cy5-MMAE dosage: 2 nmol/mouse) (n = 3). (b) In vivo fluorescence photographs of HCT-116 subcutaneous tumor-bearing mice treated with Cy5-MMAE-cRGD-Lipep-Ms or Cy5-MMAE-Lipep-Ms (Cy5-MMAE dosage: 2 nmol/mouse). (c) Quantification of Cy5-MMAE accumulated in major organs and tumor at 6 h postinjection of Cy5-MMAE-cRGDLipep-Ms, Cy5-MMAE-Lipep-Ms, or free Cy5-MMAE (n = 3, *p < 0.05). (d) MTD assays of MMAE-cRGD-Lipep-Ms and free MMAE in nude mice.
given at a dosage of 0.1 mg/kg. Figure 4a shows that MMAEcRGD-Lipep-Ms at 0.1 mg of MMAE equiv./kg induced significantly better tumor inhibition than the nontargeted MMAE-Lipep-Ms and free MMAE. Increasing the MMAEcRGD-Lipep-Ms dosage to 0.2 mg of MMAE equiv./kg led to nearly complete inhibition of tumor progression. Similarly, Adcetris as an MMAE antibody conjugate was reported to induce effective tumor regression in Karpas 299 anaplastic large cell lymphoma (ALCL) and L540cy non-Hodgkin lymphoma (HD) tumor models.48 Figure 4b showed that no significant body weight loss was observed for all treatment groups, implying that these treatment schemes are welltolerated. Figure 4c exhibited that the MMAE-cRGD-LipepMs group had the smallest tumor blocks, corroborating that MMAE-cRGD-Lipep-Ms induced the best antitumor effect. Further analyses on the tumor inhibition rate (TIR) revealed that MMAE-cRGD-Lipep-Ms at a dosage of 0.2 mg of MMAE equiv./kg gave a high TIR of 93% (Figure 4d). At a dose of 0.1 mg of MMAE equiv./kg, MMAE-cRGD-Lipep-Ms afforded a TIR of 78%, which was remarkably higher than that with MMAE-Lipep-Ms (48%) or free MMAE (26%). The most prominent adverse effect of MMAE is neutropenia.51 The hematological analyses showed that the counted levels of neutrophils (NEU), white blood cells (WBC), serum urea (UREA), and liver enzyme aspartate aminotransferase (AST) were in the normal ranges (Figure 5), indicating that MMAEcRGD-Lipep-Ms induced negligible systemic toxicity.
is able to markedly improve the circulation time and tumor accumulation of MMAE. Dose-limiting toxicity is a major problem for clinical use of MMAE.47 To evaluate their safety, we investigated the maximum-tolerated dose (MTD) of MMAE-cRGD-Lipep-Ms in nude mice. Intriguingly, MMAE-cRGD-Lipep-Ms displayed a remarkably decent MTD of over 2 mg of MMAE equiv./kg, which was over 10-fold higher than that of free MMAE (