Polymeric micelles for enhanced lymphatic drug delivery to treat metastatic tumors

Contents lists available at ScienceDirect

Journal of Controlled Release journal homepage: www.elsevier.com/locate/jconrel

Polymeric micelles for enhanced lymphatic drug delivery to treat metastatic tumors Lei Qin a,1, Fayun Zhang a,1, Xiaoyan Lu d,1, Xiuli Wei a, Jing Wang a, Xiaocui Fang a, Duanyun Si b, Yiguang Wang a, Chunling Zhang a, Rong Yang c, Changxiao Liu b,⁎, Wei Liang a,⁎⁎ a

Protein & Peptide Pharmaceutical Laboratory, Institute of Biophysics, Chinese Academy of Sciences, Beijing, China Tianjin State Key Laboratory of Pharmacokinetics and Pharmacodynamics, Tianjin Institute of Pharmaceutical Research, Tianjin, China National Center for Nanoscience and Technology, Chinese Academy of Sciences, Beijing, China d Pharmaceutical Informatics Institute, College of Pharmaceutical Sciences, Zhejiang University, Hangzhou 310058, China b c

a r t i c l e

i n f o

Article history: Received 26 November 2012 Accepted 7 July 2013 Available online 14 July 2013 Keywords: Lymphatic drug delivery Polymeric micelles Extravasation Biodistribution Metastatic tumor

a b s t r a c t Polymeric micelles have been proven to be a promising nano-sized system for drug delivery. Understanding its in vivo behaviors at the whole body, tissue and cellular levels is critical for translating this drug delivery system into clinical practice. In this study, the 14.5 nm micelles made of polyethylene glycol-phosphatidylethanolamine (PEG-PE) for delivery of doxorubicin and vinorelbine were investigated. Using confocal and two-photon microscopy imaging of live mice or tissue sections, we observed that after systemic administration, the fluorescently labeled PEG-PE micelles encapsulating doxorubicin migrated through blood vessels in entirety into the interstitial tissue, collected by lymphatic vessels, and accumulated in lymph nodes. Importantly, encapsulated drugs such as vinorelbine (Nanovin), preferentially accumulate in lymph nodes when compared to the free drugs. Moreover, the in vivo bioluminescent imaging showed that Nanovin significantly reduced lymph node metastasis rate (P b 0.05) in 4 T1-luc2 murine breast tumor bearing mice. Finally, we observed that Nanovin enhanced antitumor activity against primary tumors and lung metastases while having low toxicity in various 4 T1 tumor models. This study suggests that PEG-PE micelle is a promising drug delivery system for the treatment of lymphatic metastases, and may also have important applications in other lymphatic system-related diseases. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Metastasis affects approximately 90% of cancer patients, yet developing effective therapeutic interventions has proven to be elusive [1,2]. Cancer cells escape from the primary tumor site to secondary organs via two major routes, blood vessels and lymphatic vessels. In fact, the lymphatic dissemination is found to precede hematogenous dissemination in many types of cancers, including melanoma, breast, prostate, colon and lung cancers [3]. Evidence of tumor cells in a lymph node is often the first indicator of cancer spread. Also, lymph node metastasis is correlated with an increased risk of distant metastasis and poor clinical outcome [4]. Therefore, preventing or inhibiting lymph node metastasis is critical for improving the outcome of patients. Abbreviations: LN, lymph node; Vin, vinorelbine; Dox, doxorubicin; ICG, indocyanine green; Axi, axillary; Ing, inguinal; Mes, mesentery; i.v, intravenous; PEG-PE, distearoylsn-glycero-3-phosphoethanolamine-N-[methoxy (polyethylene glycol)-2000]. ⁎ Correspondence to: C. Liu, Tianjin State Key Laboratory of Pharmacokinetics and Pharmacodynamics, Tianjin Institute of Pharmaceutical Research, Tianjin 300193, China. Tel.: + 86 22 23006870; fax: +86 22 23006860. ⁎⁎ Correspondence to: W. Liang, Institute of Biophysics, Chinese Academy of Sciences, 15 Datun Road, Chaoyang District, Beijing 100101, China. Tel.: +86 10 64889861; fax: +86 10 64845388. E-mail addresses: [email protected] (C. Liu), [email protected] (W. Liang). 1 Lei Qin, Fayun Zhang and Xiaoyan Lu contributed equally to this work. 0168-3659/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jconrel.2013.07.005

Conventional chemotherapeutic drugs fail to enter the lymphatic system effectively due to dose-limiting toxicities or failure to access the lymphatics at adequate concentrations via intravenous (i.v.) injection. Encapsulating chemotherapeutic drugs into nanocarriers may improve drug accumulation in the lymphatic system, if their particle size and administration route are appropriate. Lymphatic uptake requires an optimal size between 10 nm to 100 nm [5], while the nanocarriers need to be small enough to extravasate from blood vessels when administrated intravenously. For this reason, the delivery of drug into regional lymphatic system is mostly facilitated by local injection of colloidal particles, such as liposomes [6–8], emulsion [9] and nanoparticles [10,11]. The polymer-conjugated protein drug SMANCS in lipiodol formulation was found to be delivered to lymph nodes after tumorfeeding artery injection [12–15]. However, there are limited studies regarding lymphatic drug delivery system via conventional intravenous injection. Intravenously injected drug delivery systems have better compliance and suitability for multiple dose administration compared with local injected drug delivery systems. Among novel drug delivery systems, polymeric micelles are emerging as a promising platform for drug delivery. Polymeric micelles are nanosized supramolecular assemblies of amphiphilic polymers that possess a core-shell type architecture. The hydrophobic core of micelles is a drug reservoir, and the shell is a hydrophilic corona that provides a

NANOMEDICINE

Journal of Controlled Release 171 (2013) 133–142

NANOMEDICINE

134

L. Qin et al. / Journal of Controlled Release 171 (2013) 133–142

protective interface between the core and the external environment [16]. Importantly, alterations in the composition of the constituent polymers influence their characteristics including micelle size and shape, core-drug compatibility, drug loading capacity and release, and stability, enabling manipulation of the encapsulated drug's pharmacokinetic profile, tissue and cellular level distribution [17–19]. The polyethylene glycol-phosphatidylethanolamine (PEG-PE)-based polymeric micelle is one of the most renowned micelle for small molecule anti-cancer drug delivery. Above the critical micelle concentration (CMC), PEG-PE spontaneously self-assembles to form micelles. The flexible PEG serves as a hydrophilic shell, and the PE serves as a hydrophobic core to admit drugs. To achieve specific delivery, a full understanding of the fate of polymeric micelle-based dug formulations is needed. However, knowledge of many points related to micelle fate, such as stability in vivo [20], capability of extravasation [21,22], transport within tissue and transport at cellular level [19,23], remains limited. Previously we have shown that PEG-PE micelles can effectively encapsulate doxorubicin [24]. Micelle-encapsulated doxorubicin showed much better antimetastatic activity than free (unencapsulated) doxorubicin through i.v. injection. However, the mechanism underlying the enhanced antimetastatic activity for this micelle-based drug delivery system remains unclear. We hypothesize that the micelle encapsulation changes the in vivo behaviors of chemotherapeutic drugs. Micelle encapsulation may cause more drug accumulation in lymph nodes, which can be sanctuary sites of metastatic tumor cells. To test this hypothesis, we used imaging techniques to investigate the in vivo behaviors of PEG-PE micelles alone or with their cargos, and further confirmed the antimetastatic property of this delivery system by encapsulating chemotherapeutic drug vinorelbine. 2. Materials and methods 2.1. Drugs and reagents 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (PEG2000-DSPE), 1,2-distearoyl-sn-glycero-3phosphoethanolamine-N-[poly(ethylene glycol)2000-N′-carboxyfluorescein] (CF-PEG-PE), hydrogenated phos-phatidylcholine (HSPC), cholesterol, and 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolaminen-(lissamine rhodamine B sulfonyl) (rhodamine-PE) were purchased from Avanti Polar Lipids. Doxorubicin hydrochloride and Vinorelbine tartrate were provided by Hisun Pharmaceutical Co. Ltd (Zhejiang, China) and Minsheng Corp (Zhejiang, China), respectively. Indocyanine green (ICG) was from Sigma.

vinorelbine tartrate, or indocyanine green (ICG) were prepared by a one-step self-assemble method as previously described [25]. Briefly, each drug was dissolved in Milli-Q water, and mix with PEG-PE micelle stock solution at their optimal molar ratios (doxorubicin: PEG-PE = 0.5:1, vinorelbine: PEG-PE = 0.25:1, ICG: PEG-PE = 0.01:1). The mixtures were incubated at 60 °C for 30 min to allow for drug encapsulation. After incubation, the prepared drug-loaded micelles were purified by passage through a 0.22 μm polyethersulfone syringe membrane, and the flow-through part was collected for further study. The encapsulation efficiency of drugs was determined after removing the free drug via ultrafiltration [25]. The drug concentration in micelles was measured by high-performance liquid chromatography (HPLC). The doublelabeled drug-loaded micelles were made using a similar procedure by incubating fluorescent drug doxorubicin and CF-labeled micelles. Labeled liposome was prepared by dissolving HSPC, cholesterol and rhodamine-PE in chloroform at a molar ratio of 55:45:1. Chloroform was removed, and the formed lipid film was hydrated by PBS at 60 °C for 1 h to form the liposomal suspension. This suspension was extruded by the Extruder with 100 nm polycarbonate membrane (Whatman) to become the uniform liposome with a diameter of 100 nm. 2.4. Two photon microscopy (TPM) To study the extravasation of micelles, mice were i.v. injected with fluorescence labeled micelles or liposomes or free fluorescent dyes as described in the Supplementary methods. Imaging was performed using a multiphoton laser scanning microscope (Olympus' Fluoview FV1000-MPE, Japan). The excitation wavelength was set to 810 nm, and the filter BA495-540 was used for CF labeled PEG-PE fluorescence collection, the filter BA570-625 was used for doxorubicin fluorescence collection. The images around blood vessels in the dorsal skin of live nude mice were continuously collected for 1 min at 1, 2 and 3 h intervals after injection. The fluorescence intensity was quantified using Image J 1.43 m. 2.5. Confocal microscopy and fluorescence molecular tomography

Female BALB/c mice or immunodeficient BALB/c nu/nu mice were purchased from the Vital River Laboratory Technology Co. Ltd. (Beijing, China), and treated in compliance with Animal Care and Use Committee of Institute of Biophysics guidelines. The mouse mammary carcinoma cell line 4 T1 and the human melanoma tumor cell line MDA-MB435S were purchased from American Type Culture Collection (ATCC). The luciferase expressing cell line 4 T1-luc2 was purchased from Caliper Life Science. 4 T1 and 4 T1-luc2 cells were grown in RPMI 1640 medium and MDA-MB-435S cells were grown in L15 medium. All cell lines were cultured in media supplemented with 10% fetal bovine serum (Invitrogen) and maintained at 37 °C in a 5% CO2, 95% humidity incubator.

For confocal microscopy studies, female athymic BALB/c nu/nu mice were injected i.v. with CF-labeled micelles alone or CF-labeled micelles encapsulating doxorubicin, and allowed to circulate for 60 min. To study the lymphatic transportation of micelles, rhodamine-labeled liposomes were i.v. administered 3 min before observation to serve as a blood pool marker. Mice were sacrificed. Skin samples were removed, fixed and scanned by confocal microscope (Fluor View FV500, Olympus) [26]. To study the interstitial tissue and lymph node accumulation, mice were autopsied 60 min after the administration of CF-labeled micelles encapsulating doxorubicin injection. Lymph nodes and small intestines were harvested. Lymph nodes were frozen in OCT and 10-μm sections were cut with a Cryostat microtome (Leica CM3050S). The mesentery was prepared by stretching out a loop of the small intestine on a slide. Specimens were air dried and examined [27]. Full angle fluorescence molecular tomography system was used to evaluate the specific lymph node accumulation of micelleencapsulated indocyanine green (Nano-ICG). Mice were given an i.v. injection of Nano-ICG (3 mice per group), and were sacrificed at 1 and 3 h after injection. Ex vivo spectral fluorescence images from lymph nodes were obtained using laser with excitation/emission: 780 nm/ 840 nm [28,29].

2.3. Preparation of micelles and liposomes

2.6. Tumor models

The empty PEG-PE micelle stock solution was prepared by dispersing PEG-PE in Milli-Q water at a concentration of 14 mM. To label the micelles, carboxyfluorescein covalently conjugated PEG-PE (CF-PEG-PE) was mixed with unlabeled PEG-PE at 2.5:100 molar ratio to form the CF-labeled micelles. Micelles encapsulating doxorubicin hydrochloride,

In all animal experiments, mice were randomly divided into three groups: the control group, the free vinorelbine treatment group and the micelle-encapsulated vinorelbine (Nanovin) treatment group. Unless otherwise indicated, these groups were treated intravenously with PBS, 5 mg/kg free vinorelbine, or 5 mg/kg Nanovin (based on

2.2. Animals and cells

vinorelbine content) once weekly for 3 weeks. The free vinorelbine or Nanovin (vinorelbine: PEG-PE = 1:10, w/w) solution containing 0.5 mg/mL or 1 mg/mL vinorelbine was used to treat mice. To detect spontaneous lymph node metastasis, 4 T1-luc2 cells were injected into the left fourth mammary fat pad of 6–8 week old female BALB/c mice. Mice were randomly divided into three groups and received the above treatment on day 6. Lymph node metastasis was evaluated by bioluminescent signals of 4 T1-luc2 cells in inguinal and axillary lymph nodes on day 35. To detect spontaneous lung metastasis, 4 T1 cells were injected into the left fourth mammary fat pad of mice. Tumors were allowed to develop for 4 days and Nanovin (10 mg/kg), free vinorelbine (10 mg/kg) or PBS was given to the mice once weekly for 2 weeks. The lung metastasis was assessed by weight and lung tumor nodules in the lung on day 31. In the tumor surgery model, 4 T1 cells were injected into the left fourth mammary fat pad of mice. The primary tumors were surgically removed on day 8, when the spontaneous metastases had already developed. On the next day, mice were treated once weekly for 2 weeks. The anti-metastatic effect was reflected by life span of the mice. In the direct metastasis model, 4 T1 was injected intravenously to the mice. The treatments were started on the next day. On day 24, the lung metastasis was evaluated as described above. To evaluate primary tumor growth, MDA-MB435S cells were injected into the left mammary fat pad of female athymic BALB/c nu/nu mice. Twenty days later, the mice started to receive different treatments once weekly for 5 weeks. The tumor volumes and life spans of mice were recorded. Additional methods regarding this method are described in the Supplementary Materials and Methods. 2.7. Systemic toxicity The acute toxicity produced by Nanovin or free vinorelbine was tested in non-tumor-bearing mice. These mice were administered once intraperitoneally at doses of 10, 20.8, 30, 43.2 and 62.2 mg/kg (10 mice per dose). The control group was given saline alone. The lethal dose causing death to 50% of the population (LD50) was calculated using the Bliss probit method. For myelosuppression study, 2 × 105 4 T1 cells were subcutaneously injected into the fat pad area of mice. On the day following cell implantation, mice received Nanovin, free vinorelbine or PBS i.v., once weekly for 3 weeks. The tumor-bearing mice receiving PBS injection served as controls, whereas the tumorfree mice which did not receive any treatments were called normal group. The mice were sacrificed on day 17 after 4 T1 cell implantation, and the right leg bones were harvested and washed with 4 mL 3% acetic acid. Bone marrow cells were counted under a microscope. To detect any venous irritation caused by the drugs, Nanovin (0.2 mL, 10 mg/kg) or free vinorelbine (0.2 mL, 10 mg/kg) was injected into the ear vein of New Zealand White rabbits. The erythema at the injection site was photographed. To examine the toxicity of free vinorelbine and Nanovin to T cells, 4 × 105 T cells from spleens or lymph nodes of female BALB/c mice were plated into 96-well plates and allowed to grow for 24 h. The cells were then exposed to a series of concentrations of free vinorelbine or Nanovin for 72 h, and the viability of cells was measured using the methylthiazoltetrazolium (MTT) method, as previously describe [30]. The percentage of survival cells relative to that of untreated cells was estimated from data of three individual experiments. 2.8. Statistical analysis Statistical tests for data analysis included Student's t test, Wilcoxon signed rank test, Mann–Whitney test, Fisher's exact test and Log rank test. Statistical analyses were performed using SPSS version 12.0 or GraphPad version 5.0. All tests were two-sided, and P values of less than 0.05 were considered statistically significant.

135

3. Results 3.1. Micelles extravasate from blood vessels and accumulate in the interstitial tissue The PEG-PE micelles, and CF-labeled PEG-PE micelles were prepared by the self-assembly of amphiphilic PEG-PE polymers or carboxyfluorescein covalently conjugated PEG-PE polymers (CF-PEG-PE). TEM showed that the micelles were spherical-shaped and uniform in size (Supplementary Fig. 1). And the average diameter of these micelles is 14.5 nm as previously determined by DSL [25]. The CMC of PEG-PE is ~10−5 M [31,32]. By incubating these micelles with fluorescent drug doxorubicin, we successfully encapsulated doxorubicin into micelles, with an encapsulation efficiency of 99.4% at 1:2 drug:PEG-PE molar ratio [24]. In vitro release profile of the micelle-encapsulated doxorubicin at pH 7.0 showed low and sustained drug release [25]. The PEG-PE micelles encapsulating doxorubicin demonstrated to be more stable than PEG-PE micelles alone in the presence of serum proteins in vitro [33]. Here, we investigated the in vivo extravasation and stability of the micelles. Real time in vivo two-photon microscopy (TPM) was used for 4-hour imaging of i.v. injected CF-labeled micelles or micelles containing fluorescence derived from doxorubicin or 100 nm rhodamine-labeled liposomes in blood vessels up to a depth of 500 μm in the skin of live mice. To observe the micelle in vivo process, a device was used to fix the skin of live nude mouse (Supplementary Fig. 2). When we injected CF-labeled micelles and rhodamine-labeled liposomes separately into the same mouse, the 14.5 nm of micelles, seen as green perivascular fluorescence, extravasated from blood vessels and later became diffuse and spread throughout the skin interstitium (Supplementary Fig. 3). However, red perivascular fluorescence from 100 nm liposomes was not observed throughout the 4-hour observation period, indicating that there was little or no extravasation of liposomes [34]. To clarify whether micelles remain intact in vivo, even after extravasation, we performed an experiment to image CF-labeled micelles encapsulating doxorubicin in blood vessels after i.v. injection. The fate of micelles in vivo depends largely on the in vivo stability of micelles, but the processes underlying micelle fate are unclear [35]. We found that CF-PEG-PE and doxorubicin-derived fluorescence were completely colocalized both on the inside and outside of blood vessels throughout the 4-hour observation period (Fig. 1A–C). The quantification of fluorescence derived from CF-PEG-PE and doxorubicin showed that the fluorescent intensity of both dyes inside the blood vessels decreased gradually while the fluorescent intensity outside the blood vessels increased gradually throughout the time course of the experiment (Fig. 1E and F). However, when free doxorubicin and CF-labeled micelles were injected separately, colocalization was not observed and the green to red fluorescent ratio changed over time both inside and outside of the blood vessel (Supplementary Fig. 4). These results suggest that micelles remain intact before and after extravasation, and are transported effectively to tissues from the circulatory system. Micelle extravasation and stability were further confirmed by confocal imaging of mouse mesentery. Clear colocalization of the two dyes both inside and outside of the blood vessels was also observed (Fig. 1D). 3.2. Micelles transport through the lymphatic system and accumulate in lymph nodes We expect that, after extravasation, micelles will be collected in the lymphatic system. Confocal microscopy was used to track the fate of micelles in mouse mesentery. We found that 60 min after injection of CF-labeled micelles encapsulating doxorubicin, the extravasated micelles and their cargo were transported around cells, and the cargos released from micelles could barely be seen in the interior of the cells (Fig. 2A). This further indicates that most extravasated micelles remained intact. Unlike the vascular endothelium, the endothelium of

NANOMEDICINE

L. Qin et al. / Journal of Controlled Release 171 (2013) 133–142

NANOMEDICINE

136

L. Qin et al. / Journal of Controlled Release 171 (2013) 133–142

Fig. 1. Extravasation and interstitial accumulation of micelles. Mice were i.v. injected with micelles containing CF-PEG-PE and doxorubicin. Images were acquired from the skin of the nude mouse using the TPM (A–C) or from the mesentery of the mouse using the confocal microscopy (D) in two separate experiments. (A-C) The extravasation of micelles across the blood vessel into the surrounding skin tissue at 1 h (A), 2 h (B) and 3 h (C) after injection. CF-PEG-PE and doxorubicin showed almost a complete colocalization during the period. (E and F) Quantification of fluorescence derived from both CF-PEG-PE and doxorubic inside (E) as well as outside (F) of the blood vessel. R represents the fluorescence intensity ratio between CF-PEG-PE to doxorubicin. Data are given as mean ± SEM. (D) The colocalization of CF-PEG-PE and doxorubicin in the mesentery 60 min after injection. Inset, higher magnified view. Dox, doxorubicin. Scale bars, 50 μm.

lymphatic capillaries is highly permeable because of discontinuities between individual endothelial cells. These gaps can be larger than a micrometer in diameter, large enough to readily admit micelles [36]. Therefore, most of the extravasated intact micelles may enter the lymphatic vessels rather than go back to the blood vessels. Lymphatic localization of the extravasated micelles was studied by confocal microscopy using a sequence injection method [26]. Intravenously administered CF-labeled micelles were allowed to circulate for 60 min. Rhodamine-

labeled liposomes were injected three minutes before skin was sampled to label the blood vessels. Thus, blood vessels should be labeled by both dyes. If we see vessels that are labeled only with CF, that indicates the micelles have been accumulated in lymphatics. Consistent with what we expect, the lymphatic vessels observed were filled with CF-labeled micelles (Fig. 2C and Supplementary Fig. 5). In addition, using LYVE-1 antibody to specifically recognize lymphatic vessels, we detected the injected CF-labeled micelles encapsulating doxorubicin. Confocal imagings showed a colocalization of CF-PEG-PE and doxorubicin in lymphatic vessels (Supplementary Fig. 6). When CF-labeled micelles and free doxorubicin were injected separately into the blood vessel (CF-labeled micelles were injected 5 min before doxorubicin), only fluorescence-derived from CF-PEG-PE was observed in lymphatic vessels, while the free doxorubicin exhibited a different accumulation pattern (Supplementary Fig. 7). These results further confirmed that the micelle-encapsulated doxorubicin remained intact even after they extravasated from blood vessels and were collected by lymphatic system. Next, we sought to demonstrate whether micelles can reach lymph nodes from the periphery. Lymph nodes were isolated 60 min after injection of CF-labeled micelles encapsulating doxorubicin and their cross sections were examined by confocal microscopy. We observed that the two dyes localized in completely different cellular compartments: green fluorescence from CF-PEG-PE micelles were detected only in the plasma membrane, while red fluorescence from doxorubicin was detected in the cytoplasm and the cell nucleus (Fig. 2B). This indicates that micelles can be transported to the lymph node from the periphery, and encapsulated doxorubicin is delivered into the cells, while CF-PEG-PE is detained in the plasma membrane. Collectively, these experiments indicate that micelles are transported to lymph nodes through the lymphatic system via extravasation. We then investigated whether micelles can deliver other drugs to lymph nodes and compared the behaviors between micelleencapsulated drugs and free drugs. Indocyanine green (ICG), an agent for evaluating hepatic function, and vinorelbine, a vinca alkaloid anti-tumor agent, were encapsulated into PEG-PE micelles respectively [37]. The results showed that lymph nodes from mice treated with micelle-encapsulated ICG (Nano-ICG) or free ICG both displayed high fluorescent intensity after 1 h (Fig. 2D, upper). However, fluorescence was only seen in lymph nodes from mice administered Nano-ICG after 3 h (Fig. 2D, lower). ICG is unstable and loses its fluorescence rapidly in solution [38]. Micelle encapsulation reduced ICG clearance from the lymph nodes, indicating that micelles readily deliver ICG to lymph nodes and enhance ICG stability. Next, we tested the bio-distribution of micelle-encapsulated vinorelbine (Nanovin) in mice. The results showed dynamic changes in drug concentrations in lymph nodes (Fig. 2E). The accumulation of Nanovin in lymph nodes at 24 h after administration was 4.7 fold higher than that of free vinorelbine (P b 0.01). Drug concentrations in other tissues were not significantly different between Nanovin and free vinorelbine treatments at each time point, except in the bone marrow. Free vinorelbine was 2.6 fold higher than Nanovin at 12 h in the bone marrow (P b 0.01) (Supplementary Fig. 8). This observation further confirmed fluorescent images of lymph nodes using ICG-labeled micelles (Fig. 2D). A model illustrating micelle transportation to lymph nodes via extravasation is shown in Supplementary Fig. 9. 3.3. Vinorelbine encapsulated in micelles prevent lymphatic metastasis After we confirmed that micelles enhanced delivery of drugs to lymph nodes, we then asked whether encapsulation of drugs into micelles might improve their activity against lymph node metastasis. Vinorelbine, an anticancer mitotic inhibitor, has been used for the treatment of metastatic breast cancer and non-small cell lung cancer [39]. The encapsulation efficiency of micelle-encapsulated vinorelbine (Nanovin) is 99.8% at 1:4 drug:PEG-PE molar ratio [40]. Similar to the release of doxorubicin from micelles, Nanovin also showed a low and

137

Fig. 3. Effect of Nanovin on lymph node metastases. (A) Ex vivo bioluminescence imaging of lymph node metastases. The 4 T1-luc2 tumor bearing mice were treated with 5 mg/kg Free Vin, 5 mg/kg Nanovin or PBS once weekly for 3 weeks. Five weeks after tumor implantation, mice were sacrificed and left axillary LN (ipsilateral), right axillary LN (contralateral) and right inguinal LN (contralateral) were excised and imaged. (B) The percentage of mice with detectable LN metastasis. P b 0.05 for Nanovin relative to Control (Fisher's exact test). (C) The total number of left axillary LN, right axillary LN or right inguinal LN that had detectable metastasis. n = 13 for control and Free Vin group, n = 16 for Nanovin group.

Fig. 2. Transportation of micelles via lymphatics and accumulation of the encapsulated drugs in lymph nodes. Confocal images were acquired from mesentery (A) and lymph node sections (B) of mice 60 min after i.v. injection of micelles containing CF-PEG-PE and doxorubicin, or from the skin (C) of mice that received both an i.v. injection of CF-labeled micelles 60 min before observation and an i.v. injection of rhodaminelabeled liposomes immediately before observation. (A) The complete colocalization of CF-PEG-PE and doxorubicin in the interstitial tissue of the mice mesentery. Inset, higher magnified view. (B) Different localization pattern of CF-PEG-PE and doxorubicin in the LN. (C) Increased accumulation of CF-PEG-PE in lymphatic vessels. The blood vessels were double-stained, and the lymphatic vessels were single-stained (green). (D) Ex vivo spectral fluorescence images of LNs 1 h and 3 h after micelle-encapsulated ICG (Nano-ICG) or free ICG injection. (E) Vinorelbine concentrations in LNs at indicated times after i.v. administration of micelle-encapsulated vinorelbine (Nanovin) or free vinorelbine (Free Vin). Drug concentrations in LNs were measured by HPLC-MS at different time points after injection. Data are given as mean ± SEM; n = 5/group. **P b 0.01 (Student's t test). Mes, mesentery; Ing, inguinal; Axi, axillary; Scale bars, 50 μm.

sustained in vitro drug release (Supplementary Fig. 10) [25]. Using bioluminescent imaging, we investigated the anti-metastatic efficacy of Nanovin in a high luciferase expressing 4 T1-luc2 murine breast cancer model. 4 T1 cell line is characterized by high metastases through both blood and lymphatic vessels [41,42]. We orthotopically implanted these 4 T1-luc2 cells into the left mammary fat pad of BALB/c mice, and excised axillary and inguinal lymph nodes (LNs) to evaluate metastasis on day 35. The results showed that Nanovin significantly reduced

the incidence of lymph node metastasis as compared with control mice (P b 0.05), while free vinorelbine treatment did not (P N 0.05). The number of mice with detectable lymph node metastases was only one of 16 (6.3%) in the Nanovin group, whereas those in the control and free vinorelbine groups were 6 of 13 (46.2%) and 4 of 13 (30.8%), respectively (Fig. 3A and B). Correspondingly, the total number of LNs with detectable metastasis was much less in the Nanovin group (1 positive LN) than that in the control group (9 positive LNs) and the free vinorelbine group (10 positive LNs) (Fig. 3C). The mean bioluminescence value of all positive LNs in the Nanovin group was also much lower than that in the control group (49-fold) and the free vinorelbine group (76-fold), indicating less tumor burden in the positive LN of the Nanovin treated mouse. As seen in Fig. 3A, ipsilateral axillary LN metastases appeared in all LN positive mice of all three groups, whereas contralateral axillary LN metastases and contralateral inguinal LN metastases were much less frequent and did not appear in the Nanovin group. Notably, one mouse in the free vinorelbine treatment group showed extensive lymph node metastases. These results are consistent with previous findings that ipsilateral axillary LN is the draining lymph node of breast cancer, and indicate that distant lymph node metastasis is rare at the time of necropsy [43]. Taken together, our data strongly demonstrate that encapsulation of drugs into micelles significantly prevent outgrowth of tumor cells metastasizing spontaneously to LNs.

NANOMEDICINE

L. Qin et al. / Journal of Controlled Release 171 (2013) 133–142

NANOMEDICINE

138

L. Qin et al. / Journal of Controlled Release 171 (2013) 133–142

Finally, we performed surgical removal of 4 T1 primary tumors to investigate the effect of Nanovin on tumor metastases. Primary tumors were removed 8 days after tumor implantation, when spontaneous metastases had already occurred. The mice treated with Nanovin showed a significant increase in survival compared with free vinorelbine (P b 0.05; Fig. 4C). The proportion of survivors at the end of the experiment on day 100 was higher in Nanovin group (4/10) than that in the free vinorelbine group (1/10) and the PBS group (0/10). 3.5. Vinorelbine encapsulated in micelles inhibit the primary tumor growth

Fig. 4. Effect of Nanovin on lung metastases. (A) Mice receiving i.v. injection of 4 T1 cells were treated with 5 mg/kg Free Vin, 5 mg/kg Nanovin or PBS once weekly for 3 weeks. Tumor weight in the lungs (upper panel), and number of metastatic nodules in the lungs were calculated 24 days after tumor injection (lower panel). Data are given as mean ± SEM; n =12/group. *P b 0.05; **P b 0.01; ***P b 0.001 (Student's t test). (B) Mice receiving subcutaneous injection of 4 T1 cells were treated with 10 mg/kg of Free Vin, Nanovin, or PBS once weekly for 2 weeks. Lungs were removed on day 31. Mean tumor weight in the lungs (upper panel), and number of metastatic nodules in the lungs were calculated (lower panel). Data are given as mean ± SEM; n = 6/group. **P b 0.01; ***P b 0.001 (Student's t test). (C) The life span of Nanovin-treated mice. The primary tumors were removed by surgery on day 8. Mice were treated once weekly for 2 weeks with 5 mg/kg Free Vin, 5 mg/kg Nanovin or PBS after surgery. n = 10/group. P b 0.05 for Nanovin relative to Free Vin (log-rank test).

3.4. Vinorelbine encapsulated in micelles diminish the outgrowth of metastatic tumor cells to lung We further examined whether Nanovin could effectively suppress metastatic outgrowth of breast cancer to the lung in three different 4 T1 metastatic models, since lungs are the main metastatic organs except lymph nodes in 4 T1 tumors [42]. In the 4 T1 direct metastasis model, BALB/c mice received an i.v. injection of 4 T1 cells. The metastasisinhibitory effect of Nanovin was reflected in a significant tumor weight reduction in the lung as compared with control (P b 0.001) and free vinorelbine treatments (P b 0.05) at the end of the experiment (Fig. 4A, upper). To quantify the effect of Nanovin on tumor cell colonization in the lung, the number of metastatic nodules was counted. Mice treated with Nanovin had 2-fold fewer metastases in the lung than controls (P b 0.001), and had 1.5-fold fewer metastases than those treated with free vinorelbine (P b 0.01), indicating that Nanovin effectively suppressed outgrowth of lung metastases (Fig. 4A, lower). In the 4 T1 spontaneous metastasis model, 4 T1 cells were injected into the mammary fat pad to induce primary tumors. This model shares many characteristics with human breast cancer, particularly its ability to spontaneously metastasize to the lung. The Nanovin treated mice showed a reduced lung tumor burden as compared with controls (P = 0.07) at the end of the experiment (Fig. 4B, upper). In the study of tumor cell colonization, the number of tumor nodules in the lungs was ~3-fold fewer than controls (P b 0.001; Fig. 4B, lower).

We also asked whether incorporation of drugs into micelles could improve their antitumor efficacy. We studied the primary tumor inhibition effect of Nanovin in MDA-MB435s and 4 T1-bearing mice. The MDA-MB-435S-bearing BALB/c nu/nu mice were injected i.v. with free vinorelbine, Nanovin or PBS once a week for 5 weeks. On day 91, the ratio of the mean tumor volume of treated animals (T) to that of the untreated control group (C), T/C, was 0.53 for free vinorelbine and 0.09 for Nanovin (Fig. 5A and B). Primary tumors completely regressed in two mice given Nanovin treatment. However, tumor growth in mice treated with free vinorelbine was only slower than that in untreated control mice, and regression was not observed. Extensive metastases in the liver, lung and spleen were observed in all mice that did not survive after treatment. The mice treated with Nanovin showed a significant increase in survival compared with free vinorelbine (P b 0.05; Fig. 5C). The median survival of mice treated with Nanovin was more than 200 d (5/6), whereas that of mice treated with free vinorelbine or PBS was 180.3 days and 140.8 days, respectively. The primary tumor growth of 4 T1 murine mammary carcinoma was also significantly decreased in mice treated with free (P b 0.05) and micelle-encapsulated vinorelbine (P b 0.001) as compared with control mice on day 31, and the T/C was 0.77 for free vinorelbine and 0.52 for Nanovin (Fig. 5D and E). The TUNEL staining of 4 T1 tumor tissues showed a significant increase in the number of apoptotic tumor cells in Nanovin-treated tumors, indicating that more drug was available in the Nanovin-treated tumors (Fig. 5F). 3.6. Micelle encapsulation reduces vinorelbine toxicity Vinorelbine treatment results in phlebitis, myelosuppression in humans [44]. We therefore compared the toxicity of free vinorelbine and Nanovin. The LD50 values for Nanovin and free vinorelbine were 21.7 mg/kg (95% CI = 18.0 to 26.1) and 12.7 mg/kg (95% CI = 8.6 to 18.8), respectively, indicating that Nanovin was significantly less toxic than free vinorelbine (P b 0.05; Supplementary Table 1). In the study of drug-induced irritation, we found that Nanovin completely eliminated venous irritation induced by peripheral infusion (Fig. 6A). We also examined the myelosuppression caused by vinorelbine. The results showed that Nanovin treated mice were associated with a negligible decrease in the number of bone marrow cells compared with the normal group, whereas free vinorelbine treated mice had a significant drop in the number of bone marrow cells (P b 0.05) compared with the normal group (Fig. 6B). This was concordant with the drug distribution in bone marrows (Supplementary Fig. 8). Taken together, Nanovin showed lower toxicity than free vinorelbine. In addition, Nanovin did not increase the cytotoxicity of vinorelbine to T cells in vitro (Fig. 6C). 4. Discussion In the present study, we have demonstrated that PEG-PE micelles with a diameter of 14.5 nm can extravasate from blood vessels to interstitial tissue in an intact form, and then are captured by lymphatic system. Utilizing this property of PEG-PE micelles, we selectively delivered chemotherapeutic drugs into lymphatic system in which micelles maintain a sustained high drug concentration to kill the resident tumor

139

Fig. 5. Effect of Nanovin on primary tumor growth. (A) The tumor volume of MDA-MB-435S-bearing BALB/c nu/nu mice treated with 5 mg/kg Free Vin, 5 mg/kg Nanovin or PBS i.v. once a week for 5 weeks. Data are given as mean ± SEM; n = 6/group. **P b 0.01 compared with controls at corresponding time points (Student's t test). (B) Representative photos of MDA-MB-435S tumors on day 91. (C) The life span of MDA-MB-435S-bearing mice. P b 0.05 for Nanovin relative to Free Vin (log-rank test). (D) Mice bearing 4 T1 primary tumors were treated with 10 mg/kg Free Vin, 10 mg/kg Nanovin or PBS i.v. once a week for 2 weeks. Tumor volume was measured by caliper. Data are given as mean ± SEM; n = 10/group. *P b0.05, **P b0.01 when compared with controls at corresponding time points (Student's t test). (E) Representative photos of 4 T1 primary tumors on day 31. (F) The apoptotic effect of Nanovin in 4 T1 tumor sections was evaluated by TUNEL assay. Tumor cells undergoing apoptosis are shown in brown.

NANOMEDICINE

L. Qin et al. / Journal of Controlled Release 171 (2013) 133–142

NANOMEDICINE

140

L. Qin et al. / Journal of Controlled Release 171 (2013) 133–142

Fig. 6. The toxicity of Nanovin. (A) Nanovin reduces venous irritation. Representative photos of rabbit ear-rim auricular veins following injections of 10 mg/kg Nanovin or 10 mg/kg Free Vin. n = 5/group. (B) The myelosuppression by Nanovin. 4 T1 tumor-bearing mice were given Nanovin or Free Vin at a dose of 5 mg/kg once weekly for 3 weeks following tumor inoculation. After killing the mice on day 17, the bone marrow cells from one leg were counted. Data are given as mean ± SEM; n = 3 for the normal group; n = 8/group for the other groups; *P b 0.05 (Student's t test). (C) Nanovin does not increase the cytotoxicity of vinorelbine to normal T cells in vitro. T cells separated from spleen were treated with Free Vin or Nanovin for 72 h, the cell viability was calculated as percentage between the optical densities (OD) of the treated cells and that of the untreated controls. Data are represented as a mean of three individual experiments.

cells. Indeed, vinorelbine encapsulated in the micelles significantly reduced tumor metastases both in the lymph node and in the lung, contributing to increased survival compared to the unencapsulated vinorelbine in mice bearing highly metastatic tumor. Recent lymphatic targeting delivery systems are almost exclusively administered via local routes, in particular, via subcutaneous route [45]. Delivery of nanoparticles to lymphatic system via venous injection is limited mainly by two factors: first, nanoparticles without surface modification of PEG are cleared rapidly from the bloodstream by the mononuclear phagocyte system, which in turn decreases the specific targeting effect [46]; second, the vascular endothelial barrier in most tissues prevents the large sized nanoparticles (N6 nm) from extravasating [47]. Subcutaneous administration of nanoparticles can overcome these problems, although they retain some limitations. First, they are unable to kill the distant dissemination of tumor cells. Moreover, although nanoparticles have the potential to protect the surrounding tissue against the toxicity of the encapsulated anti-tumor drugs, the continuously released drug may cause severe local tissue damage [45]. Here, we show that i.v. administrated PEGylated micelles. The i.v. injection can avoid the local tissue damage. More importantly, this system administration can deliver drugs through the whole body, thus it may be helpful for treating distant metastases. Transportation of agents to lymphatics via i.v. injection mainly depends on agent structures. Ferumoxtran-10 is a commercial magnetic resonance imaging (MRI) agent consisting of iron oxide crystalline core coated with dextran for the differentiation of metastatic and nonmetastatic lymph nodes in various primary malignancies [48]. This nanoparticle with a core diameter of 4–6 nm and a mean diameter of 39 nm [49] can be taken up by every normal lymph node after i.v. injection. A recent study also showed that PEGylated dendrimers with a core diameter of 5.6 nm and a mean diameter of 13.4 nm had a potential for direct transfer from the system circulation to the lymphatics [50]. Based

on these results, we infer the nanoparticle structure, especially that the core diameter is critical for extravasation. Nanoparticles with appropriate size (relative rigid cores in diameters of 4–6 nm) and structure (core-shell structure, the shells are much flexible) can be extravasated from vasculature into the interstitial space, from which they are transported to lymph nodes by way of lymphatic vessels. In our study, PEG-PE micelles with a hydrophobic core in diameter of 4.5 nm and a mean diameter 14.5 nm have been demonstrated to be able to transport from blood vessels to lymph nodes through the lymphatic system via extravasation. Our finding is consistent with previous observations as described above [51]. However, the mechanisms of the extravasation of polymeric micelles remain to be clarified. Several mechanisms may explain the transport of micelles from blood to tissues. One possibility is that it occurs by mechanisms similar to those involved in the translocation of leukocytes through the vascular endothelium, such as the formation of transcellular channels or interendothelial junctions [52,53]. The size and shape of polymeric micelles are sensitive to changes in the environment, such as temperature or applied mechanical forces [54]. The primary mechanism of transport of micelles from the blood into the interstitium is diffusion through intercellar gaps. Although it is not clear what forces drive the extrusion of micelles, both haemodynamic pressure and microvascular constriction could be involved. The ability of a micelle system to function as a solubilizer or a true carrier depends on the stability of the micelle in vivo and drug retention in the presence of blood components. It is well known that amphiphilic copolymers are thermodynamically driven to self-assemble into micelles when above the critical micelle concentration (CMC). Below the CMC, the copolymer disperses in solution as single copolymer chain. In vitro serum stability of micelles has been demonstrated by several researchers [55–57]. However, there is a significant challenge to achieve adequate in vivo stability of the micelle system, due to dilution by blood when micelles are administrated intravenously. Here, we studied the in vivo stability of one of the most renowned micelle called PEG-PE micelles, and demonstrated that PEG-PE micelles can remain kinetically stable throughout 4 hour observation following i.v. administration. Besides, Foerster Resonance Energy Transfer (FRET) experiment also suggested that micelles are stable in vivo (data not shown). The high kinetic stability of these PEG-PE micelles is probably due to the hydrophobic PE core. Our previous studies have shown that both doxorubicin and vinorelbine can be tightly packaged by PEG-PE, possibly due to the amphiphilic nature of the drugs and PEG-PE molecules and their specific structures [24,25]. These results indicate that drug itself is an influencing factor on micelle stability. We used three different tumor models (i.v. injected tumor cells for direct metastasis to the lung, subcutaneously implanted tumor cells and surgical removal of primary tumors) to study the metastatic process, allowing modulation of early steps in the complex metastatic process of intravasation, and evaluation of the end result of distant organ metastasis. These three animal models have different characteristics. In the direct metastasis model, tumor cells injected i.v. directly reside in the lung, where they may grow [58]. In spontaneous metastasis model, the subcutaneous tumors have the advantages of providing access to rich lymphatic beds, allowing evaluation of specific aspects of this metastasis [59]. Surgical removal of primary tumors has the advantage of being comparative to the clinical situation where the primary breast tumor is surgically removed and metastatic foci remain intact [41,60]. In each of these three tumor models, when compared with the unencapsulated vinorelbine, the encapsulated vinorelbine treatment resulted in a significant reduction of tumor weight in the lung and/or the number of metastatic lung nodules, which in turn increased the life span of mice bearing tumors. The enhanced permeability and retention (EPR) effect is the tumor tissue accumulation property of macromolecular drug delivery system, which is due to the leaky blood vessels and lack of effective lymphatic drainage in tumor tissues [12,15]. In this study, there was no significant difference between the retention of Nanovin and free vinorelbine in

tumor tissues (Supplementary Fig. 8D), which might be due to the relatively small size of micelles and the injection time [61]. Researchers have reported that macromolecular drugs (≥100 nm) allow them to preferentially accumulate in solid tumors by EPR effect [62], thus reducing normal tissue toxicity and adverse effects [12,15]. However, the large size of the nanoparticles seems to prevent the drug from delivering throughout the entire tumor in sufficient concentration [63]. To resolve this problem, Wong et al. developed a multistage system of nanoparticles. Their 100-nm nanoparticles will “shrink” to 10-nm nanoparticles after they extravasate from the tumor vasculature. The shrunken nanoparticles (10 nm) can readily diffuse throughout the tumor's interstitial space [64]. Recently, this group found that small (~12 nm) nanomedicines had superior tumor penetration and thus were ideal for cancer therapy [65]. In this study, we found the 14.5 nm PEG-PE micelles could extravasate from blood vessels into the interstitial space (Figs. 1D, 2A). In agreement with the previous studies described above, our 14.5 nm micelles had relative low EPR effect and high penetration in tumor tissue. Our previous study demonstrated PEG-PE encapsulation accelerated the tumor cell uptake of the drug, and caused more cytotoxicity to cancer cells [24,40,66]. We speculated that, instead of increasing total drug concentration (intracellular and intercellular) in tumor tissue, the micelle encapsulation increased the intracellular drug concentration through elevated penetration, and caused more tumor cells to die as shown in Fig. 5F. Our results suggest that PEG-PE micelles may provide an enhanced system for lymphatic drug delivery with a little change to drug exposure in systemic circulation. This drug delivery system may be most suitable for the treatment of lymphatic and circulatory system-related diseases, such as metastatic tumors. In addition to treating tumors, our drug delivery system may also be used to treat other immune related diseases such as HIV [67].

Acknowledgments This work was supported by the State Key Development Plan Project (2011CB707705) and the National Nature Sciences Foundation of China (81173012). The authors thank Y. Teng and J. Hao for the assistance with two photon microscopy and cryosectioning.

Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.jconrel.2013.07.005.

References [1] G.P. Gupta, J. Massague, Cancer metastasis: building a framework, Cell 127 (2006) 679–695. [2] S.N. Markovic, L.A. Erickson, R.D. Rao, R.H. Weenig, B.A. Pockaj, A. Bardia, C.M. Vachon, S.E. Schild, R.R. McWilliams, J.L. Hand, S.D. Laman, L.A. Kottschade, W.J. Maples, M.R. Pittelkow, J.S. Pulido, J.D. Cameron, E.T. Creagan, Malignant melanoma in the 21st century, part 1: epidemiology, risk factors, screening, prevention, and diagnosis, Mayo Clin. Proc. 82 (2007) 364–380. [3] Y. Xie, T.R. Bagby, M.S. Cohen, M.L. Forrest, Drug delivery to the lymphatic system: importance in future cancer diagnosis and therapies, Expert Opin. Drug Deliv. 6 (2009) 785–792. [4] S.L. Chen, D.M. Iddings, R.P. Scheri, A.J. Bilchik, Lymphatic mapping and sentinel node analysis: current concepts and applications, CA Cancer J. Clin. 56 (2006) 292–309, (quiz 316-297). [5] M.A. Swartz, The physiology of the lymphatic system, Adv. Drug Deliv. Rev. 50 (2001) 3–20. [6] V.I. Kaledin, N.A. Matienko, V.P. Nikolin, Y.V. Gruntenko, V.G. Budker, Intralymphatic administration of liposome-encapsulated drugs to mice — possibility for suppression of the growth of tumor-metastases in the lymph-nodes, J. Natl Cancer Inst. 66 (1981) 881–887. [7] Z.Q. Yan, C.Y. Zhan, Z.Y. Wen, L.L. Feng, F. Wang, Y. Liu, X.K. Yang, Q. Dong, M. Liu, W.Y. Lu, LyP-1-conjugated doxorubicin-loaded liposomes suppress lymphatic metastasis by inhibiting lymph node metastases and destroying tumor lymphatics, Nanotechnology 22 (2011).

141

[8] C.K. Kim, J.H. Han, Lymphatic delivery and pharmacokinetics of methotrexate after intramuscular injection of differently charged liposome-entrapped methotrexate to rats, J. Microencapsul. 12 (1995) 437–446. [9] J.S. Karajgi, S.P. Vyas, A lymphotropic colloidal carrier system for diethylcarbamazine: preparation and performance evaluation, J. Microencapsul. 11 (1994) 539–545. [10] S.T. Reddy, A.J. van der Vlies, E. Simeoni, V. Angeli, G.J. Randolph, C.P. O'Neil, L.K. Lee, M.A. Swartz, J.A. Hubbell, Exploiting lymphatic transport and complement activation in nanoparticle vaccines, Nat. Biotechnol. 25 (2007) 1159–1164. [11] B. Lu, S.B. Xiong, H. Yang, X.D. Yin, R.B. Chao, Solid lipid nanoparticles of mitoxantrone for local injection against breast cancer and its lymph node metastases, Eur. J. Pharm. Sci. 28 (2006) 86–95. [12] N. Ohtsuka, T. Konno, Y. Miyauchi, H. Maeda, Anticancer effects of arterial administration of the anticancer agent SMANCS with lipiodol on metastatic lymph nodes, Cancer 59 (1987) 1560–1565. [13] H. Maeda, SMANCS and polymer-conjugated macromolecular drugs: advantages in cancer chemotherapy, Adv. Drug Deliv. Rev. 46 (2001) 169–185. [14] H. Maeda, Y. Matsumura, Tumoritropic and lymphotropic principles of macromolecular drugs, Crit. Rev. Ther. Drug Carrier Syst. 6 (1989) 193–210. [15] K. Yamasaki, T. Konno, Y. Miyauchi, H. Maeda, Reduction of hepatic metastases in rabbits by administration of an oily anticancer agent into the portal vein, Cancer Res. 47 (1987) 852–855. [16] V.P. Torchilin, Recent advances with liposomes as pharmaceutical carriers, Nat. Rev. Drug Discov. 4 (2005) 145–160. [17] H.M. Aliabadi, M. Shahin, D.R. Brocks, A. Lavasanifar, Disposition of drugs in block copolymer micelle delivery systems: from discovery to recovery, Clin. Pharmacokinet. 47 (2008) 619–634. [18] F. Alexis, E. Pridgen, L.K. Molnar, O.C. Farokhzad, Factors affecting the clearance and biodistribution of polymeric nanoparticles, Mol. Pharm. 5 (2008) 505–515. [19] R. Savic, L. Luo, A. Eisenberg, D. Maysinger, Micellar nanocontainers distribute to defined cytoplasmic organelles, Science 300 (2003) 615–618. [20] Y.H. Bae, H. Yin, Stability issues of polymeric micelles, J. Control. Release 131 (2008) 2–4. [21] R.K. Jain, Delivery of molecular and cellular medicine to solid tumors, Adv. Drug Deliv. Rev. 46 (2001) 149–168. [22] K. Hori, M. Nishihara, M. Yokoyama, Vital microscopic analysis of polymeric micelle extravasation from tumor vessels: macromolecular delivery according to tumor vascular growth stage, J. Pharm. Sci. 99 (2010) 549–562. [23] A.I. Minchinton, I.F. Tannock, Drug penetration in solid tumours, Nat. Rev. Cancer 6 (2006) 583–592. [24] N. Tang, G. Du, N. Wang, C. Liu, H. Hang, W. Liang, Improving penetration in tumors with nanoassemblies of phospholipids and doxorubicin, J. Natl. Cancer Inst. 99 (2007) 1004–1015. [25] Y. Wang, R. Wang, X. Lu, W. Lu, C. Zhang, W. Liang, Pegylated phospholipidsbased self-assembly with water-soluble drugs, Pharm. Res. 27 (2010) 361–370. [26] H. Dafni, T. Israely, Z.M. Bhujwalla, L.E. Benjamin, M. Neeman, Overexpression of vascular endothelial growth factor 165 drives peritumor interstitial convection and induces lymphatic drain: magnetic resonance imaging, confocal microscopy, and histological tracking of triple-labeled albumin, Cancer Res. 62 (2002) 6731–6739. [27] C.A. Kunder, A.L. St John, G. Li, K.W. Leong, B. Berwin, H.F. Staats, S.N. Abraham, Mast cell-derived particles deliver peripheral signals to remote lymph nodes, J. Exp. Med. 206 (2009) 2455–2467. [28] D. Wang, X. Liu, J. Bai, Analysis of fast full angle fluorescence diffuse optical tomography with beam-forming illumination, Opt. Express 17 (2009) 21376–21395. [29] F. Liu, X. Liu, D. Wang, B. Zhang, J. Bai, A parallel excitation based fluorescence molecular tomography system for whole-body simultaneous imaging of small animals, Ann. Biomed. Eng. 38 (2010) 3440–3448. [30] T. Mosmann, Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays, J. Immunol. Methods 65 (1983) 55–63. [31] A.N. Lukyanov, Z. Gao, L. Mazzola, V.P. Torchilin, Polyethylene glycol-diacyllipid micelles demonstrate increased acculumation in subcutaneous tumors in mice, Pharm. Res. 19 (2002) 1424–1429. [32] T. Wang, V.A. Petrenko, V.P. Torchilin, Paclitaxel-loaded polymeric micelles modified with MCF-7 cell-specific phage protein: enhanced binding to target cancer cells and increased cytotoxicity, Mol. Pharm. 7 (2010) 1007–1014. [33] X. Wei, Y. Wang, W. Zeng, F. Huang, L. Qin, C. Zhang, W. Liang, Stability influences the biodistribution, toxicity, and anti-tumor activity of doxorubicin encapsulated in PEG-PE micelles in mice, Pharm. Res. 29 (2012) 1977–1989. [34] F. Yuan, M. Leunig, S.K. Huang, D.A. Berk, D. Papahadjopoulos, R.K. Jain, Microvascular permeability and interstitial penetration of sterically stabilized (stealth) liposomes in a human tumor xenograft, Cancer Res. 54 (1994) 3352–3356. [35] H. Chen, S. Kim, W. He, H. Wang, P.S. Low, K. Park, J.X. Cheng, Fast release of lipophilic agents from circulating PEG-PDLLA micelles revealed by in vivo forster resonance energy transfer imaging, Langmuir 24 (2008) 5213–5217. [36] A. Supersaxo, W.R. Hein, H. Steffen, Mixed micelles as a proliposomal, lymphotropic drug carrier, Pharm. Res. 8 (1991) 1286–1291. [37] T. Okouneva, B.T. Hill, L. Wilson, M.A. Jordan, The effects of vinflunine, vinorelbine, and vinblastine on centromere dynamics, Mol. Cancer Ther. 2 (2003) 427–436. [38] V. Saxena, M. Sadoqi, J. Shao, Indocyanine green-loaded biodegradable nanoparticles: preparation, physicochemical characterization and in vitro release, Int. J. Pharm. 278 (2004) 293–301. [39] A. Romero, M.G. Rabinovich, C.T. Vallejo, J.E. Perez, R. Rodriguez, M.A. Cuevas, M. Machiavelli, J.A. Lacava, M. Langhi, L.R. Acuna, S. Amato, R. Barbieri, C. Sabatini, B.A. Leone, Vinorelbine as first-line chemotherapy for metastatic breast-carcinoma, J. Clin. Oncol. 12 (1994) 336–341.

NANOMEDICINE

L. Qin et al. / Journal of Controlled Release 171 (2013) 133–142

NANOMEDICINE

142

L. Qin et al. / Journal of Controlled Release 171 (2013) 133–142

[40] X.Y. Lu, F.Y. Zhang, L. Qin, F.Y. Xiao, W. Liang, Polymeric micelles as a drug delivery system enhance cytotoxicity of vinorelbine through more intercellular accumulation, Drug Deliv. 17 (2010) 255–262. [41] C.J. Aslakson, F.R. Miller, Selective events in the metastatic process defined by analysis of the sequential dissemination of subpopulations of a mouse mammary tumor, Cancer Res. 52 (1992) 1399–1405. [42] B.A. Pulaski, S. Ostrand-Rosenberg, Mouse 4 T1 breast tumor model, Curr. Protoc. Immunol. (2001) 1–11, (Chapter 20, Unit 20.2). [43] A.E. Giuliano, P.S. Dale, R.R. Turner, D.L. Morton, S.W. Evans, D.L. Krasne, Improved axillary staging of breast cancer with sentinel lymphadenectomy, Ann. Surg. 222 (1995) 394–399, (discussion 399 – 401). [44] S.A. Johnson, P. Harper, G.N. Hortobagyi, P. Pouillart, Vinorelbine: an overview, Cancer Treat. Rev. 22 (1996) 127–142. [45] C. Oussoren, G. Storm, Liposomes to target the lymphatics by subcutaneous administration, Adv. Drug Deliv. Rev. 50 (2001) 143–156. [46] S. Cai, Q.H. Yang, T.R. Bagby, M.L. Forrest, Lymphatic drug delivery using engineered liposomes and solid lipid nanoparticles, Adv. Drug Deliv. Rev. 63 (2011) 901–908. [47] J. Rosenecker, W. Zhang, K. Hong, J. Lausier, P. Geppetti, S. Yoshihara, D. Papahadjopoulos, J.A. Nadel, Increased liposome extravasation in selected tissues: effect of substance P, Proc. Natl. Acad. Sci. U. S. A. 93 (1996) 7236–7241. [48] M.G. Harisinghani, J. Barentsz, P.F. Hahn, W.M. Deserno, S. Tabatabaei, C.H. van de Kaa, J. de la Rosette, R. Weissleder, Noninvasive detection of clinically occult lymph-node metastases in prostate cancer, N. Engl. J. Med. 348 (2003) 2491–2499. [49] C.W. Jung, P. Jacobs, Physical and chemical properties of superparamagnetic iron oxide MR contrast agents: ferumoxides, ferumoxtran, ferumoxsil, Magn. Reson. Imaging 13 (1995) 661–674. [50] L.M. Kaminskas, J. Kota, V.M. McLeod, B.D. Kelly, P. Karellas, C.J. Porter, PEGylation of polylysine dendrimers improves absorption and lymphatic targeting following SC administration in rats, J. Control. Release 140 (2009) 108–116. [51] M. Rafi, H. Cabral, M.R. Kano, P. Mi, C. Iwata, M. Yashiro, K. Hirakawa, K. Miyazono, N. Nishiyama, K. Kataoka, Polymeric micelles incorporating (1,2-diaminocyclohexane) platinum (II) suppress the growth of orthotopic scirrhous gastric tumors and their lymph node metastasis, J. Control. Release 159 (2012) 189–196. [52] W.A. Muller, Leukocyte-endothelial-cell interactions in leukocyte transmigration and the inflammatory response, Trends Immunol. 24 (2003) 327–334. [53] B. Engelhardt, H. Wolburg, Mini-review: transendothelial migration of leukocytes: through the front door or around the side of the house? Eur. J. Immunol. 34 (2004) 2955–2963.

[54] C.C. Huang, J.P. Ryckaert, H. Xu, Structure and dynamics of cylindrical micelles at equilibrium and under shear flow, Phys. Rev. E 79 (2009). [55] T. Musacchio, V. Laquintana, A. Latrofa, G. Trapani, V.P. Torchilin, PEG-PE micelles loaded with paclitaxel and surface-modified by a PBR-ligand: synergistic anticancer effect, Mol. Pharm. 6 (2009) 468–479. [56] T. Miller, R. Rachel, A. Besheer, S. Uezguen, M. Weigandt, A. Goepferich, Comparative investigations on in vitro serum stability of polymeric micelle formulations, Pharm. Res. 29 (2012) 448–459. [57] J. Lu, S.C. Owen, M.S. Shoichet, Stability of self-assembled polymeric micelles in serum, Macromolecules 44 (2011) 6002–6008. [58] D.K. Hsu, F.T. Liu, Regulation of cellular homeostasis by galectins, Glycoconj. J. 19 (2004) 507–515. [59] Y. He, K. Kozaki, T. Karpanen, K. Koshikawa, S. Yla-Herttuala, T. Takahashi, K. Alitalo, Suppression of tumor lymphangiogenesis and lymph node metastasis by blocking vascular endothelial growth factor receptor 3 signaling, J. Natl. Cancer Inst. 94 (2002) 819–825. [60] B.A. Pulaski, M.J. Smyth, S. Ostrand-Rosenberg, Interferon-gamma-dependent phagocytic cells are a critical component of innate immunity against metastatic mammary carcinoma, Cancer Res. 62 (2002) 4406–4412. [61] S. You, L. Zuo, W. Li, Optimizing the time of Doxil injection to increase the drug retention in transplanted murine mammary tumors, Int. J. Nanomedicine 5 (2010) 221–229. [62] S.D. Perrault, C. Walkey, T. Jennings, H.C. Fischer, W.C. Chan, Mediating tumor targeting efficiency of nanoparticles through design, Nano Lett. 9 (2009) 1909–1915. [63] R.K. Jain, Delivery of molecular and cellular medicine to solid tumors, J. Control. Release 53 (1998) 49–67. [64] C. Wong, T. Stylianopoulos, J. Cui, J. Martin, V.P. Chauhan, W. Jiang, Z. Popovic, R.K. Jain, M.G. Bawendi, D. Fukumura, Multistage nanoparticle delivery system for deep penetration into tumor tissue, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 2426–2431. [65] V.P. Chauhan, T. Stylianopoulos, J.D. Martin, Z. Popovic, O. Chen, W.S. Kamoun, M.G. Bawendi, D. Fukumura, R.K. Jain, Normalization of tumour blood vessels improves the delivery of nanomedicines in a size-dependent manner, Nat. Nanotechnol. 7 (2012) 383–388. [66] J. Wang, Y. Wang, W. Liang, Delivery of drugs to cell membranes by encapsulation in PEG-PE micelles, J. Control. Release 160 (2012) 637–651. [67] T. Schacker, The role of secondary lymphatic tissue in immune deficiency of HIV infection, AIDS 22 (Suppl. 3) (2008) S13–S18.