We have developed multifunctional nanoparticles for codelivery of bortezomib and doxorubicin to synchronize their pharmacokinetic profiles and synergize their activities in solid tumor treatment, a need still unmet in the clinic. Micellar nanoparticles were formed by a spatially segregated, linear-dendritic telodendrimer containing three segments: a hydrophilic polyethylene glycol (PEG), a bortezomib-conjugating intermediate, and a dendritic doxorubicin-affinitive interior. Bortezomib-conjugated telodendrimers, together with doxorubicin, self-assembled into monodispersed micelles [NP(BTZ-DOX)] with small particle sizes (20–30 nm) for dual drug delivery. NP(BTZ-DOX) displayed excellent drug-loading capacity and stability, which minimized premature drug leakage and synchronized drug release profiles. Bortezomib release was accelerated significantly by acidic pH, facilitating drug availability in the acidic tumor microenvironment. Synergistic anticancer effects of combined bortezomib and doxorubicin were observed in vitro against both multiple myeloma and ovarian cancer cells. NP(BTZ-DOX) prolonged payload circulation and targeted tumors in vivo efficiently with superior signal ratios of tumor to normal organs. In vitro and in vivo proteasome inhibition analysis and biodistribution studies revealed decreased toxicity and efficient intratumoral bortezomib and doxorubicin delivery by nanoformulation. NP(BTZ-DOX) exhibited significantly improved ovarian cancer treatment in SKOV-3 xenograft mouse models in comparison with free drugs and their combinations, including bortezomib and Doxil. In summary, tumor-targeted and synchronized delivery system elicits enhanced anticancer effects and merits further development in the clinical setting. Cancer Res; 77(12); 3293–305. ©2017 AACR.

Ovarian cancers remain ongoing challenges mainly due to the development of drug resistance and the likely occurrence of cancer metastasis at the time of diagnosis (1). Relapse of disease is commonly seen for ovarian cancers after the primary treatment with platinum-based therapeutics, which will be further treated with different chemodrugs (2). Monotherapy through a single mechanism frequently shows limited efficacies due to intrinsic or acquired drug resistance in cancer treatment (3–6). Instead, drug combinations with different mechanisms of action can kill cancer cells synergistically and minimize the emergence of drug-resistant mutations (7). The development of novel and efficient drug combinations may improve ovarian cancer treatment, as well as for other solid tumor treatments. Proteasome inhibitors target many protein degradation pathways, providing rationale for the clinical use in combination therapy.

Bortezomib is a potent proteasome inhibitor that is approved for the treatment of multiple myeloma and other hematologic malignancies (8, 9). Bortezomib binds to the threonine residues in the active sites of the proteasome via boronic acid to block the degradation of ubiquitinated proteins (10–12). Proteasome inhibition regulates protein levels, which may sensitize or antagonize other drugs in cancer treatment. For example, bortezomib has been shown to antagonize microtubule-interfering drugs (e.g., paclitaxel) by inhibiting G2–M transition and MCL-1 degradation in both neuroblastoma (13) and ovarian cancer cells (14). In contrast, bortezomib is able to sensitize DNA-damaging agents, for example, doxorubicin and cisplatin, by inhibiting NF-κB activation (13).

Bortezomib-based combination chemotherapy has dominated multiple myeloma treatment in clinical practice (3). For example, bortezomib and Doxil combination yields significantly improved efficacy when compared with single reagent in multiple myeloma treatments (15). Preclinical studies have shown that free bortezomib is effective in several solid tumors, both in vitro and in vivo (16). However, a lack of therapeutic effect was observed in clinical trials utilizing bortezomib as a single agent (8) or in combination chemotherapy (16, 17) in treating solid tumors. This includes the trial using bortezomib in combination with Doxil to treat ovarian cancer (18). Although the reason is not clear for the difference between preclinical and clinical results, it is likely associated with the unfavorable and mismatched pharmacokinetic and biodistribution profiles of bortezomib and Doxil. Bortezomib can be taken up efficiently by red blood cells after intravenous administration (19), resulting in rapid clearance from plasma (∼5 minutes; refs. 19–21). In contrast, Doxil has a half-life of approximately 55 hours in humans with very slow drug release and limited intratumoral diffusion due to the large particle size (∼100 nm; ref. 22). Nanoparticle-based delivery systems could accommodate such variations in pharmacokinetics for tumor-targeted drug delivery, resulting in enhanced combination therapy (23–25).

Nanoparticle-mediated bortezomib delivery systems have been developed that utilize either physical encapsulation or reversible conjugation, demonstrating improved anticancer efficacy (26–32). Eliminating premature drug release from nanoparticles is critical to delivering a sufficient amount of drug to tumor cells via the enhanced permeability and retention (EPR) effect (33, 34). Both bortezomib and doxorubicin have relatively good solubility in aqueous solution. It poses challenges to control their release profile through physical encapsulation in nanoparticles, especially for small-sized micelles (20–30 nm) with less physical barrier for drug diffusion, which is preferred for intratumoral drug delivery over large polymeric nanoparticles (>100 nm; refs. 35, 36). Therefore, specific nanocarrier design is needed for efficient codelivery of these two drugs to reboot the efficacy of bortezomib and foster their synergism in ovarian cancer treatment.

Previously, we have developed a well-defined linear-dendritic telodendrimer as a versatile delivery vehicle (35, 37–40). The modular design and the precise telodendrimer synthesis enable rational nanocarrier engineering for either drug (41, 42) or protein/peptide (43, 44) delivery. Recently, we have demonstrated that drug binding affinity within a telodendrimer nanocarrier can be enhanced by decorating telodendrimers with drug-binding moieties identified via computational and experimental screenings (42). Bortezomib has a dipeptide structure lacking sufficient hydrophobicity for stable encapsulation in polymeric micelles. Notably, bortezomib can form a reversible boronate ester linkage efficiently for prodrug formation via coupling with cis-diols and be released either under acidic condition or in the presence of competing diols (45, 46). The acidic extracellular tumor microenvironments (pH 6.2–6.8; ref. 47) and the more acidic subcellular compartments, for example, late endosome and lysosome (∼pH 5.0), have been widely exploited in pH-triggered drug release (48). In this study, we rationally modify three-layered telodendrimer with both bortezomib-conjugating moiety (BCM) and doxorubicin-binding moiety (DBM) to load bortezomib and doxorubicin via reversible conjugation and affinitive encapsulation, respectively. Rhein (Rh) molecule as DBM will be introduced to strengthen doxorubicin affinity within nanocarrier by taking advantage of the pi-pi stacking. A collection of naturally occurring cis-diol and/or catechol-containing biomolecules, for example, caffeic acid (CaA), chlorogenic acid (ChA), and gluconic acid (GA), will be selected as BCMs. The bortezomib and doxorubicin coloaded nanocarrier is expected to enhance their accumulations in solid tumors and release both drugs preferably in tumor microenvironments to synergize their efficacy in ovarian cancer treatment.

See Supplementary Information for detailed information on Materials and spectroscopic characterization.

Model reactions of bortezomib conjugation with cis-diol/catechol-containing compounds

Bortezomib and cis-diol/catechol-containing compounds (CaA, ChA, and GA) were dissolved in DMSO at a concentration of 100 mmol/L. Bortezomib solution was mixed with equal volume of diol or catechol separately. The conjugation was carried out at 37°C for 18 hours. In situ proton nuclear magnetic resonance (1H-NMR) study for model compounds conjugation was conducted by mixing bortezomib (20 mmol/L) with ChA (20 mmol/L) solutions in DMSO-d6 at the same volume ratio.

Telodendrimer synthesis

The telodendrimers were synthesized via solution-phase peptide condensation reactions following the typical procedure reported in our previous publications (35, 40). The procedure is detailed in Supplementary Information.

Bortezomib–telodendrimer formation and dual-drug loading

Bortezomib was initially reacted with telodendrimer to form bortezomib–telodendrimer conjugate. Dual-drug loaded nanocarriers with different ratios of bortezomib and doxorubicin were formulated by varying the amount of bortezomib-conjugated telodendrimer and doxorubicin. A detailed procedure is described in Supplementary Information.

In vitro drug release

The releases of bortezomib and doxorubicin from nanocarriers were performed using a dialysis method. A 300 μL nanoformulation in PBS was placed in dialysis tubing (MWCO = 3,500 Da) and immersed in a 40 mL release medium of PBS (pH 7.4) or acetate buffer (pH 5.5) at 37°C. Two microliters of aliquots from dialysis tubing were sampled at predetermined intervals. Doxorubicin concentration was determined using fluorescence qualification at excitation/emission of 488/599 nm. The amount of bortezomib was detected by fluorescence using alizarin as a fluorescent reporter (49, 50). The quantitative bortezomib titration by alizarin was validated for both bortezomib–mannitol and bortezomib–telodendrimer conjugates. Because of interference in fluorescence between doxorubicin and alizarin in bortezomib measurements, the release profile of bortezomib was conducted using bortezomib-conjugated nanoformulations without doxorubicin loaded. Three microliters of bortezomib–telodendrimer sampled were treated with 2 μL mannitol (10 mg/mL in DMSO) overnight followed by the addition of 25 μL alizarin solution (1 mg/mL in DMSO) for another 6 hours. The alizarin fluorescence was examined at excitation/emission = 485/625 nm.

Cell lines and animals

H929 MM and SKOV-3 ovarian cancer cell lines were purchased from ATCC (2009) without further authentication. The reference cancer cells were expanded and cryopreserved in liquid nitrogen until use. Both cell lines used were within 6–8 passages and subcultured less than three passages. H929 cell was cultured in RPMI1640 medium and SKOV-3 cell was maintained in McCoy 5A medium, supplemented with 10% FBS, 100 U/mL penicillin G, and 100 μg/mL streptomycin at 37°C using a humidified 5% CO2 incubator. Female athymic nude mice (Nu/Nu strain), 5–6 weeks age, were purchased from The Jackson Laboratory. All animals were kept under pathogen-free conditions in accordance with Association for Assessment and Accreditation of Laboratory Animal Care guidelines and allowed to acclimatize for at least 4 days prior to any experiments. All animal experiments were performed in compliance with institutional guidelines following protocols approved by the Animal Use and Care Administrative Advisory Committee. SKOV-3 cells (7 × 106) in a 100-μL mixture of PBS and Matrigel (1:1, v/v) without FBS were injected subcutaneously into the flanks of nude mice to form nodules.

In vitro efficacy and synergistic effects

The cytotoxicities of unloaded and drugs loaded nanoparticles were studied using H929 and SKOV-3 cells and cell viability was tested via MTS assays. Cells were seeded in a 96-well plate at the cell densities of 8 × 103 cells per well. After overnight incubation, the cells were treated with different formulations at serial concentrations. After incubation, CellTiter 96 aqueous cell proliferation reagent was added to each well according to the manufacturer's instructions. Untreated cells served as negative controls. Results were obtained as the average cell viability of triplicate experiments calculated by a formula of [(ODtreat − ODblank)/(ODcontrol − ODblank) × 100%].

The median-effect analysis proposed by Chou and Talalay (51) were used to evaluate synergistic drug combinations in vitro. This method assesses the drug–drug interaction in terms of combination index (CI), which is based on the relationship between concentration and response. The CI was used to evaluate synergy between doxorubicin and bortezomib against H929 and SKOV-3 cells in vitro. CI analysis was performed by CompuSyn software. Values of CI < 1, CI = 1, and CI > 1 indicate synergy, additivity, and antagonism, respectively.

In vitro proteasome inhibition studies

20S Proteasome Activity Assay Kit (EMD Millipore) was used to measure proteasome activity. A total of 6 × 105 cells/well (H929 and SKOV-3) were placed in a 6-well dish. Free bortezomib and doxorubicin formulation (BTZ+DOX) and BTZ-DOX coloaded PEG5kChA4-Rh4 nanoformulation [NP(BTZ-DOX)] were added at 10 ng/mL bortezomib and 100 ng/mL doxorubicin equivalent concentrations to their respective wells and incubated at 37°C for 1 and 18 hours, respectively. Negative control groups were incubated with PBS. After incubation, cells were washed twice with cold PBS and lysed in 60 μL lysis buffer (50 mmol/L HEPES, 5 mmol/L EDTA, 150 mmol/L NaCl, 1% Triton X-100, pH 7.5) on ice for 30 minutes. The lysate was centrifuged at 20,000 × g for 15 minutes at 4°C. Proteasome activity assays were carried out by the addition of 10 μL of the lysate, 10 μL of proteasome substrate (Suc-LLVY-AMC, 0.5 mmol/L), and 80 μL of the assay buffer [25 mmol/L HEPES, pH 7.5, 0.5 mmol/L EDTA, 0.05% NP-40, and 0.001% SDS (w/v)] into a 96-well plate with incubation at 37°C for 1 hour. Proteasome activity was assessed via fluorescence spectroscopy at excitation/emission = 380/460 nm (BioTek Synergy H1).

Optical animal imaging

Nude mice bearing human SKOV-3 ovarian cancer xenografts (approximately 500 mm3) were randomized into two groups (3 mice per group). DiD [a hydrophobic near-infrared (NIR) cyanine dye] was encapsulated in the nanocarrier together with doxorubicin and bortezomib [DiD-NP(BTZ-DOX)] at a mass ratio of 4:1:0.1 (DiD/DOX/BTZ). One-hundred microliters of DiD-NP(BTZ-DOX) solution was filtered with a 0.22-μm filter to sterilize the solution before injection. An equal amount of DiD in ethanol solution was diluted with PBS, mixed with BTZ+DOX and intravenously injected. Mice were anesthetized with isoflurane and optically imaged at designed time points using an IVIS 200 (PerkinElmer) with the excitation/emission at 625/700 nm. After 72 hours, the animals were sacrificed and all the major organs and tumors excised for ex vivo imaging to determine the in vivo biodistribution of nanoparticles. The associated fluorescence intensities were determined by Living Image software (Caliper Life Sciences) using operator-defined regions of interest (ROI) measurements.

Molecular basis for in vivo anticancer effects

SKOV-3 bearing mice were distributed into three groups of 3 mice and were treated intravenously with PBS, BTZ+DOX, and NP(BTZ-DOX) formulation, respectively on days 0 and 4 at a dose of 0.5 mg/kg bortezomib and 5 mg/kg doxorubicin equivalents. The mice were sacrificed on day 5, and the tumors were excised for the proteasome activity measurement and apoptosis analysis, respectively. Tumors were homogenized in 200-μL lysis buffer (100 mmol/L Tris, pH 7.8, 150 mmol/L NaCl, 1% Triton-X100) per 50 mg tissue for 30 minutes on ice. The homogenate was centrifuged at 20,000 × g for 15 minutes at 4°C. The supernatant was removed and centrifuged again to ensure complete removal of any precipitate. The proteasome activity was measured following the same protocol described above for the in vitro proteasome inhibition study. The same tumor lysates were analyzed by Western blot analysis to compare the protein ubiquitination levels in different groups. Gel loading buffer was added to the clarified lysates and incubated at 37°C for 30 minutes, subjected to SDS-PAGE and transferred to nitrocellulose for Western blotting and probed with anti-ubiquitin (a gift from Dr. Martin Obin, Tufts University, Medford, MA) and anti-Grp94 (Stressgen #SPA-850) following the protocol described previously (52).

Intratumoral distribution and tumor pathology studies

Tumors were collected from euthanized mice, freshly embedded in Tissue-Tek O.C.T. compound, and stored at −80°C. Tumor tissue was sectioned to a thickness of 16 μm on a microtome-cryostat for fluorescence imaging of DiD biodistribution in tumor sites. For bioactivity studies, tumor samples were cut into 10-μm sections and fixed with 4% paraformaldehyde and stained with hematoxylin and eosin (H&E) for pathology analysis. Levels of apoptosis in tumors treated with different formulations were detected via TUNEL staining using an In Situ Cell Death Detection Kit, POD (Roche) following the manufacturer's instructions. Nuclei were counterstained with DAPI and tumor apoptosis signals were detected under a laser scanning fluorescence microscope (Leica).

In vivo anticancer efficacy

We implanted 6 nude mice per group with human ovarian SKOV-3 cancer cells with the consideration of possible failure in tumor growth. Nude mice bearing SKOV-3 xenograft tumors (approximately 100–150 mm3 in volume) were randomly divided into six groups (n = 5–6 per group), including control (PBS), bortezomib (0.5 mg/kg), BTZ+DOX (0.5 mg/kg bortezomib and 5 mg/kg doxorubicin equivalents), BTZ+Doxil (0.5 mg/kg bortezomib and 5 mg/kg doxorubicin equivalents), NP(BTZ-DOX) (0.5 mg/kg bortezomib and 5 mg/kg doxorubicin equivalents), and NP(BTZ-DOX) (0.8 mg/kg bortezomib and 8 mg/kg doxorubicin equivalents). Treatments were intravenously administered via tail vein injection on days 0, 4, and 8 for a total of three treatments. Animal body weight and tumor volume were monitored over time. Seven days after the last treatment, approximately 100 μL of blood was collected via tail bleeding for blood counts. The tumor sizes were measured with electronic calipers, and volumes calculated using the following formula: V = (shortest diameter2 × longest diameter)/2. Animals were sacrificed when tumor volume exceeded 1,500 mm3, the tumor became necrotic, or body weight loss > 20% was observed.

Statistical analysis

Data are presented as means ± SD, unless otherwise specified. All statistical analyses were performed using Student t test for comparison of two groups, one-way ANOVA for multiple groups. The level of statistical significance was set with P < 0.05 considered significant. IC50 values were calculated from cell viability curves by fitting a dose–response model using sigmoidal function with variable Hillslope. Pharmacokinetic data were fitted to a two-compartment model. Sample size estimation for in vivo efficacy study was based on our previous studies (45, 46). We targeted an effect size of two in the treatment groups, resulted from an anticipated 30% reduction in tumor volume and an estimated SD of 15% of tumor inhibition at the end of experiment. Using two-sample t test, a sample size of 5 per group will give an 80% power to detect the targeted effect size of 2 at a significance level of 0.05, in both tumor growth inhibition and survival analysis, which reported the time of tumor volume exceeding 1,500 mm3. Animal survival data were analyzed descriptively using the Kaplan–Meier method and the survival differences between two groups were assessed using the Student t-test.

To codeliver bortezomib and doxorubicin with the ultimate goal of synergizing their anticancer effects by synchronizing drug accumulations in tumor sites, we designed a series of multifunctional and well-defined three-layered telodendrimers (Fig. 1A), consisting of a linear polyethylene glycol (PEG) block, a cis-diol–containing intermediate layer for bortezomib conjugation and a doxorubicin binding interior layer.

Model reactions of bortezomib conjugation with BCMs

The reactivities of BCMs with bortezomib in the formation of boronate were evaluated initially before their conjugation onto telodendrimers to demonstrate the feasibility for telodendrimer modification. The explicit formations of bortezomib conjugates with CaA, ChA, and GA were confirmed by MALDI-TOF MS and displayed as sodium adducts of molecular ion peaks (Fig. 1B). A bivalent conjugate of bortezomib with ChA was detected at m/z 1073.674, which is correlated with the chemical structure of ChA (Fig. 1A) containing both cis-diol and catechol functionalities. 1H-NMR analysis provides further spectrometric evidence for the conjugation between bortezomib and diols (Supplementary Fig. S1). The disappearance of characteristic phenol resonances indicates the success complexation of catechol with boronic acid into ChA–BTZ conjugate. The association constant of diols for bortezomib chelation was measured to be 370, 290, and 34 M−1 for ChA, GA, and CaA, respectively (Supplementary Fig. S2; refs. 50, 53). The results show that ChA has the highest affinity for bortezomib, which is in agreement with literature reports that the catechols have greater reactivity than cis-diols in reacting with boronic acid (30, 53). Consequently, these biocompatible compounds were chosen as BCMs to functionalize telodendrimers to validate the methodology for nanocarrier design and optimize bortezomib conjugation.

Codelivery telodendrimer synthesis

The teloendrimers were constructed using solution peptide chemistry with the structural scaffold shown in Fig. 1A. The oligolysine-based scaffold is spatially segregated into two dendritic domains for dual-drug loading. A flexible oligo-ethylene glycol spacer is placed between the two domains to reduce steric hindrance. Telodendrimers were synthesized from PEG5k-NH2 (∼5 kDa); the stepwise synthetic pathway is detailed in Supplementary Fig. S3. The oligolysine components were synthesized by the coupling of the orthogonally protected lysine via Boc- or Fmoc-protecting chemistries. The efficient amide bond formation allows the construction of telodendrimers with ease and efficiency. The chemical structures of all the intermediates in telodendrimer synthesis have been confirmed by 1H-NMR with the clear signal assignments and accurate peak integration (Supplementary Figs. S4–S10). Molecular weight determination by MALDI-TOF MS reveals the maintained narrow polydispersity and accurate mass increase according to the intermediate chemical structures throughout the telodendrimer synthesis (Supplementary Fig. S11). The functionalizations of telodendrimers with BCMs were carried out at the final step through DIC/NHS chemistry. The codelivery telodendrimers bearing DBM (Rh) and BCMs (GA, CaA, and ChA) are denoted as PEG5kGA4-Rh4, PEG5kCaA4-Rh4, and PEG5kChA4-Rh4 (chemical structures are shown in Fig. 1A).

The structural integrity and molecular weight distribution of the resultant codelivery telodendrimers were characterized by MALDI-TOF MS. Similar to the PEG-NH2 precursor, monomodal and narrowly dispersed molecular weight distributions were detected with molecular peaks centered at m/z of 7,780, 7,640, and 8,210 for PEG5kGA4-Rh4, PEG5kCaA4-Rh4, and PEG5kChA4-Rh4 telodendrimers, respectively (Fig. 1C). As shown in Table 1, the observed molecular weights are close to the theoretical values calculated based upon the PEG 5000 precursor. The slight discrepancies in molecular weight detected by MALDI-TOF MS can be attributed to the entanglement of high-molecular weight polymers, which are likely to hinder the desorption of ionized species. The chemical composition of codelivery telodendrimers were further confirmed using 1H-NMR spectroscopy. A representative 1H-NMR spectrum for the telodendrimer PEG5kChA4-Rh4 is shown in Supplementary Fig. S12. Well-resolved resonances of the olefin protons in chlorogenic moieties at 6.21–6.26 and 7.46–7.49 ppm, aromatic protons at 6.77, 6.98, and 7.04 ppm, and phenolic protons at 9.12 and 9.54 ppm were observed. The integration values of these signature peaks match closely to the predicted values relative to the characteristic methoxy proton of PEG at 3.25 ppm, highlighting the well-defined structure of the telodendrimer. In addition, the apparent peak broadening in the telodendrimer compared with the small molecule (Supplementary Fig. S12) is indicative of reduced chain mobility after the covalent attachment to the telodendrimer. 1H-NMR spectra for PEG5kGA4-Rh4 and PEG5kCaA4-Rh4 are shown in Supplementary Figs. S13 and S14.

BTZ–telodendrimer conjugation

The presence of cis-diols/catechol on the designed telodendrimers provides an efficient anchor for bortezomib conjugation via reversible boronate bond. In this study, the drug-loading contents (DLC) for bortezomib were set at 10% in nanotherapeutics production. For BTZ–telodendrimer prodrug formation, a 10% DLC gives a molar ratio of telodendrimer to drug of approximately 1:2. The progress of the coupling reaction was monitored by alizarin dye assay and UV-Vis absorbance of unbounded bortezomib at 270 nm. The results revealed that the bortezomib coupling reaction completed after 8-hour incubation at 37°C with an initial bortezomib concentration of 20 mg/mL (Supplementary Fig. S15). Bortezomib conversions are estimated to be 86%, 70%, and 94% for GA-, CaA-, and ChA-containing telodendrimers, respectively. Conjugation efficiency was then quantified using 1H-NMR by comparing the signal integration of methoxy proton on the PEG chain at 3.25 ppm with dimethyl proton on bortezomib ranging from 0.74 to 0.93 ppm (Supplementary Figs. S16–S18). A summary of loading efficiencies is presented in Table 1. Bortezomib drug loading efficiency (DLE) reflected by NMR study is determined to be 89% ± 2%, 80% ± 5%, and 94% ± 4% for PEG5kGA4-Rh4, PEG5kCaA4-Rh4, and PEG5kChA4-Rh4, respectively. Alizarin dye assay further confirmed that approximately 10% wt bortezomib content was presented in the final telodendrimer prodrug (Supplementary Fig. S19). The overall conjugation efficiency is correlated with the density and reactivity of the cis-diol/catechol groups as confirmed by molecular bindings study (Supplementary Fig. S2). In each molecule of ChA and GA, there are two pairs of cis-diol/catechol moieties available for bortezomib attachment, which ultimately leads to more effective conjugation. CaA with only one reactive site shows lower coupling efficiency in prodrug formation.

Codelivery telodendrimer nanocarrier characterization

The self-assembly behavior of telodendrimers was characterized by critical micelle concentration and dynamic light scattering (DLS) measurements (Supplementary Fig. S20 and Table 1). The high encapsulation capacity of doxorubicin by codelivery telodendrimers is attainable due to the strong interactions between doxorubicin and rhein through pi-pi stacking and hydrophobic interactions (42). The DLE of doxorubicin at 10 wt% feeding ratio were determined to be 93% ± 2%, 96% ± 2%, and 97% ± 1% for PEG5kGA4-Rh4, PEG5kCaA4-Rh4, and PEG5kChA4-Rh4, respectively (Table 1). Bortezomib-conjugated telodendrimers self-assembled with doxorubicin into monodispersed micelles with small sizes ranging from 20 to 40 nm for all telodendrimer platforms, as evident by both DLS and TEM studies (Fig. 2A–D). Representative TEM images showed that the empty micelle formed by PEG5kChA4-Rh4 are elongated micelles in shape (Fig. 2C), which is correlated with pi-pi stacking of Rh moieties observed in our previous studies (42). After dual-drug loading, a morphology transformation from a rod-like shape to a spherical structure was observed, suggesting the disruption of long-range molecular stacking (Fig. 2D).

In vitro drug release

The stabilized doxorubicin encapsulation at the core region of micelles is further confirmed by the sustained doxorubicin release from micelles as compared with free doxorubicin (Fig. 2E). Over 24-hour dialysis, 50% to 70% of doxorubicin was released from telodendrimer nanoformulations, DOX-PEG5kGA4-Rh4, DOX-PEG5kCaA4-Rh4, and DOX-PEG5kChA4-Rh4, which were significantly slower than free doxorubicin (100% release at 8 hours). In contrast, less than 20% of doxorubicin can be released from Doxil within 24 hours and no significant drug release was observed afterwards, which may limit the drug availability in cancer treatment. The difference in release rates from three telodendrimer nanoformulations is likely attributed to the varying hydrophobicity of diol/catechol functionalities at the intermediate layer between the doxorubicin-affinitive core and the PEG shielding corona. Relatively more hydrophilic GA shows lower capability to sustain doxorubicin, leading to slightly faster release; while more hydrophobic CaA and ChA interfaces create additional barrier to prevent the doxorubicin diffusion. A modest increase in doxorubicin release rate from DOX-PEG5kChA4-Rh4 nanoformulation was observed at pH 5.5, due to the increased solubility of doxorubicin (Fig. 2E).

The release profiles in Supplementary Fig. S21 indicate that bortezomib release from telodendrimer conjugates are significantly slower compared with the bortezomib–mannitol (free drug formulation) in PBS (pH 7.4) at 37°C. The boronate ester is sensitive to both acidic pH and the cis-diol containing sugar, for example, glucose. The bortezomib releases from telodendrimer conjugates were found to be dual stimuli-responsive as designed. About 39% of bortezomib was released within 24-hour incubation at pH 7.4, which was slightly increased to 48% in the presence of an elevated glucose concentration (50 mmol/L; Fig. 2F). Bortezomib release was significantly accelerated under acidic condition (pH 5.5), showing 75% of bortezomib release after 24 hours. The pH-triggered bortezomib release was further enhanced by the presence of glucose, which indicates the preferred bortezomib release from nanoformulations on-demand at tumor microenvironments and in the acidic lysosome after tumor cell uptake.

Biocompatibility of nanocarriers

The in vitro toxicities of the designed telodendrimers were evaluated in terms of hemolytic property and cytotoxicity. All three telodendrimers show negligible in vitro hemolytic activity toward red blood cells after 24-hour incubation at concentrations ranging from 10 to 1,000 μg/mL (Supplementary Fig. S22A). In vitro cell culture studies revealed noncytotoxicity of telodendrimers with concentrations ranging from 0.32 to 625 μg/mL as determined by MTS assay (Supplementary Fig. S22B). We further examined the potency of bortezomib–telodendrimer conjugates in H929 MM cells. The results demonstrated that the bortezomib–telodendrimer conjugates maintained drug potency, having IC50 values of 0.49, 3.49, and 1.74 ng/mL for BTZ-PEG5kGA4-Rh4, BTZ-PEG5kCaA4-Rh4, and BTZ-PEG5kChA4-Rh4, respectively. The slightly elevated IC50 values as compared with that of BTZ–mannitol (0.23 ng/mL) could be attributed to the sustained bortezomib release from the nanoparticles. PEG5kChA4-Rh4 exhibits superior properties in drug loading, release profile and in vitro efficacy, and therefore was selected for further studies.

In vitro cellular uptake

H929 MM cells were incubated with dual-drug loaded PEG5kChA4-Rh4, NP(BTZ-DOX), as well as the free drug combination (BTZ+DOX), and imaged using confocal microscopy to investigate the overall cellular internalization. Free doxorubicin rapidly translocated into the nucleus with little fluorescence detected in the cytoplasm, even with brief incubation for 30 minutes (Supplementary Fig. S23A). In sharp contrast, doxorubicin in nanoformulation showed exclusive distribution in cytoplasm within 30 minutes (Supplementary Fig. S23A) and gradual translocation into the nucleus after 2-hour incubation (Fig. 2G). In addition, cells treated with the NP(BTZ-DOX) had stronger colocalization of doxorubicin within the lysosomal compartments (stained by lysotracker green) compared with a low level of colocalization for BTZ+DOX treatment, suggesting the endocytotic pathway for nanocarrier uptake. In addition, the cellular uptake of NP(BTZ-DOX) was founded to be hindered at 4°C, indicating the energy dependent process for nanoparticle uptake. As for BTZ+DOX, cell uptake at 4°C was not affected (Supplementary Fig. S23A). Further cell lysis and drug extraction analysis quantitatively reveals that the cell uptake profile of the nanoformulation is temperature and concentration dependent (Supplementary Fig. S23B–S23E).

In vitro synergistic effect in cancer treatment

The systemic toxicity and in vivo efficacy largely relies on the optimal therapeutic ratios loaded and the amount of drugs delivered to tumor sites. As such, we first assessed the in vitro synergistic effect of BTZ-DOX coloaded by our nanoparticles in killing cancer cells. In agreement with other studies (54), results show that bortezomib could markedly enhance the H929 MM cell killing at subtoxic concentrations of doxorubicin. Cell viability analysis demonstrated that the IC50 value for doxorubicin in H929 cells was 229.3 ng/mL in the absence of bortezomib or 7.33 ng/mL in combination with bortezomib at a 1:4 molar ratio (Fig. 3A and B). Similarly, NP(BTZ-DOX) at different BTZ/DOX drug ratios were tested in SKOV-3 ovarian cancer cells (Fig. 3C and D). We assessed the synergistic effect of bortezomib and doxorubicin using a whole cell killing panel and analysis via the Chou–Talalay method (51). The CIs were observed to be generally less than 1 at different BTZ/DOX ratio, indicating synergistic effects of the two drugs. Only at high percentage of cell killing range, low BTZ/DOX ratio, for example, 1:4 and 1:10, showed antagonism in both cell lines, which however showed significant synergism at lower concentrations (Fig. 3E and F). The CI at 50% cancer cell inhibition (CI50) were detected to be significantly smaller than 1 for a wide range of BTZ/DOX ratios from 1:1 to 1:10, indicating their strong synergism in killing both cancer models (Fig. 3E and F and Supplementary Table S1).

Biodistribution and pharmacokinetic profile

Near-infrared fluorescence (NIRF) imaging was utilized as a noninvasive method to monitor real-time tissue distribution and tumor accumulation of nanocarriers in vivo. DiD, a NIRF dye, was coloaded into NP(BTZ-DOX) micelles to probe in vivo biodistribution of nanoparticles. The in vivo whole-body fluorescent imaging showed that DiD-labeled NP(BTZ-DOX) micelles gradually accumulated at the SKOV-3 tumor xenografts throughout the 72 hours period after tail vein injection. In contrast, weak tumor fluorescence was observed in mice in the control group injected with free drugs (DiD–BTZ+DOX; Fig. 4A). At 72-hour postinjection, tumors and other major organs were harvested for ex vivo NIRF imaging to compare the tissue distribution of formulations. As shown in Fig. 4B, DiD–NP(BTZ-DOX) micelles were mainly accumulated in tumors with more than four-fold higher intensity than that in the vital organs, for example, liver, lung, spleen, and kidney. Mice treated with DiD–BTZ+DOX showed the majority accumulations in spleen and lung with lower accumulation in the tumor (Fig. 4C). During in vivo imaging, blood samples were collected and DiD fluorescent signal were measured to compare the pharmacokinetic profiles. As shown in Fig. 4D, nanoformulation exhibits significantly prolonged circulation times with the 7.5-fold increase of AUC and eight-fold increase in half-life. Following ex vivo imaging, the distribution of DiD signals in tumor slices were observed under fluorescent microscope. DiD signal was observed throughout tumor tissue slices treated with DiD–NP(BTZ-DOX), whereas signal was hardly detectable in tumor treated with DiD–BTZ+DOX (Fig. 4E).

Proteasome inhibition and tumor apoptosis

In vitro proteasome inhibition assay was performed on H929 and SKOV-3 cells after incubation with bortezomib-based combination formulation at 10 ng/mL bortezomib and 100 ng/mL doxorubicin equivalent concentrations (Fig. 5A). The proteasome activity was inhibited by the bortezomib treatments in both BTZ+DOX and NP(BTZ-DOX) formulations. After 1-hour incubation, the lysate from cells exposed to NP(BTZ-DOX) showed significantly higher proteasome activity compared with BTZ+DOX counterpart. Additional incubation (18 hours, 37°C) with NP(BTZ-DOX) elicited more proteasome inhibition. The kinetically slower proteasome inhibition by NP(BTZ-DOX) was expected due to sustained BTZ release. The reduced proteasome inhibition for NP(BTZ-DOX) in vitro implies the potentially reduced systemic off-target toxicity of the nanotherapeutics, which is associated with the spike of the free drug concentration in circulation. Therefore, it is promising to reduce toxicity and increase the tolerable dosages of nanoformulations in cancer treatment. Accordingly, the maintained efficacy of NP(BTZ-DOX) in cell culture as shown in Fig. 3 indicated that sustained drug release at tumor sites would not compromise the anticancer efficacy of nanoformulations.

We next investigated the ability of bortezomib and doxorubicin combination in inducing in vivo proteasome inhibition at tumor sites. Nude mice bearing SKOV-3 ovarian cancers were treated with PBS, BTZ+DOX, and NP(BTZ-DOX), respectively, on day 0 and day 4. On day 5, mice were sacrificed and tumors were harvested. Half of each tumor was homogenized and proteins were extracted for measurement of proteasome activity and substrate ubiquitination. As shown in Fig. 5B, proteasome activity was significantly reduced in the tumors treated with the nanoformulation compared with PBS control and BTZ+DOX treatment (P < 0.05). Western blot analysis (Fig. 5C) revealed the accumulation of ubiquitinated species in both bortezomib-treated groups, demonstrating directly that the processing of ubiquitinated substrates is inhibited by bortezomib. However, only slightly increased or comparable ubiquitination was observed for NP(BTZ-DOX) compared with BTZ+DOX, despite that tumor proteasome activity were not inhibited by free drug combination (Fig. 5B). This may be because of relatively slow degradation of ubiquitinated proteins after the restoration of proteasome activity from bortezomib inhibition. Importantly, significant reduced tumor cell density and treatment-induced tumor necrosis were observed in tumors treated with NP(BTZ-DOX) in H&E staining (Fig. 5D, top row). Furthermore, the TUNEL staining demonstrated the enhanced tumor apoptosis in the tumors treated with NP(BTZ-DOX) in comparison with both PBS and BTZ+DOX treatment (Fig. 5D, bottom row).

In vivo anticancer treatment

Based on the bortezomib dose in animal treatments reported in literatures (26, 28), SKOV-3 ovarian cancer xenograft models were treated with different formulations intravenously at the equivalent bortezomib dose of 0.5 mg/kg and doxorubicin dose of 5 mg/kg on days 0, 4, and 8. Having demonstrated the reduced proteasome inhibition in cell culture in vitro as shown in Fig. 5A, NP(BTZ-DOX) is expected to be more tolerable than free drugs in mice. Therefore, a 60% escalation in dose for NP(BTZ-DOX) was injected in BALB/c mice to test toxicity, in comparison with BTZ+DOX, for three treatments on the same schedule (day 0, 4, and 8). No significant body weight losses were found in mice treated with NP(BTZ-DOX) at 0.8/8 mg/kg. In contrast, BTZ+DOX showed body weight loss near 20% at 0.8/8 mg/kg dose (Supplementary Fig. S24). Given the decreased in vivo toxicity by nanoformulation, an elevated dose at 0.8/8 mg/kg were tested for NP(BTZ-DOX) in parallel with other treatments. As shown in Fig. 6A, body weights of mice in all treatment groups experienced modest body weight losses of <15%, which were fully recovered after day 20, indicating the tolerable toxicity associated with the treatments (Fig. 6A). Blood count analysis performed on day 7 after the third administration revealed the normal counts for all treatment groups (Supplementary Table S2). The tumor volumes in the SKOV-3 tumor bearing mice were continuously measured to monitor the tumor growth (Fig. 6B). Free drug bortezomib and BTZ+DOX treatments delayed tumor growth in comparison with PBS control group. The combination of bortezomib with a nanoformulation of Doxil exhibited improved tumor inhibition, which is likely due to the sustained drug release of Doxil. Our combinational nanoformulation significantly inhibited tumor growth at the identical dose level compared with other groups. With the increased dose level at 0.8/8 mg/kg, NP(BTZ-DOX) dramatically improved anticancer efficacy. The animal survival data in Fig. 6C revealed that NP(BTZ-DOX) significantly extended the mean survival time to 52 days at the same dose level in comparison to PBS (24 days, P < 0.001), bortezomib (30 days, P = 0.001), BTZ+DOX (37 days, P = 0.005), and BTZ+Doxil (42 days, P = 0.016). Noticeably, NP(BTZ-DOX) treatment showed a prolonged mean survival of 58 days at a dose of 0.8/8 mg/kg BTZ/DOX. The enhanced anticancer effects in these SKOV-3 ovarian cancer xenograft models by the optimized NP(BTZ-DOX)s are attributed to their improved tumor targeting, reduced systemic toxicity and the synchronized drug availability at tumor sites.

The application of multiple agents in vivo is complicated by their independent pharmacokinetics, biodistribution, and off-target effects. The unsatisfying clinical observation for bortezomib combination therapy in solid tumor treatment, including the bortezomib and Doxil combination, has been postulated to result from diverse pharmacokinetics of payloads. A nanoparticle-based codelivery system is promising to deliver drug combinations to tumor sites in a defined temporal and spatial manner at a ratiometric dose. The integration of multiple drugs into one vehicle is critical for yielding synergism in cancer treatment. The encapsulation of bortezomib and doxorubicin in a nanocarrier is challenging due to their distinct chemical structures and physical properties. In our approach, bortezomib and doxorubicin were loaded in functionally segregated telodendrimer nanocarriers via reversible conjugation and affinitive physical encapsulation, respectively, through the rational design of telodendrimers. The stability of a nanoformulation determines the in vivo performance and ultimate treatment efficacy. Polymer micelles with small particle sizes (<50 nm) are preferred for intratumoral drug delivery (55). However, polymer micelles featuring dynamic assemblies are generally associated with fast or even burst drug release when administered systemically. Thus, a strategy to control micelle size and stability is to increase the affinity between the drug and polymer within the core of the micelle. Rhein was introduced to the telodendrimer as a doxorubicin binding moiety via pi-pi stacking, exhibiting strong affinity in doxorubicin loading and stabilization. In an alternative approach, bortezomib was facilely conjugated onto telodendrimers through the reversible boronate ester bond to synchronize drug release with doxorubicin. As a result, such nanoformulation sustained the drug release of both drugs in similar profiles (Fig. 2), which are otherwise dramatically mismatched for Doxil and bortezomib (Fig. 2E and F). It is promising to synchronize their pharmacokinetics and drug distributions in vivo to synergize their activities in cancer treatment. In addition, both bortezomib and doxorubicin can respond to acidic pH to release from the nanocarrier more efficiently, which enables on-demand drug release within the tumor microenvironment.

Telodendrimers have a well-defined architecture and chemical structure via precise peptide chemistry as evidenced by NMR and MALDI-TOF MS analysis. In this study, the nanocarrier displayed attractive physiochemical characteristics as a bortezomib-based anticancer drug codelivery vehicle, including accommodation for different drug combination ratio, small particle size, high loading capacity, sustained drug release, and excellent stability and biocompatibility. The optimized nanocarrier is able to synchronize drug release profiles and increase tumor availability of bortezomib and Doxil for solid tumor treatment. The in vitro evaluations demonstrated that the combined delivery of bortezomib and doxorubicin by our nanoformulation fosters their synergism in treating MM and ovarian cancer in cell culture.

Furthermore, in vivo studies revealed that NP(BTZ-DOX) prolonged systemic circulation and delivered payloads to tumor sites efficiently. The in vitro and in vivo proteasome inhibition, tumor targeting/distribution and tumor apoptosis analysis strongly correlated with the improved anticancer effects and reduced side effects by the optimized combination nanotherapy. The molecular and pathological analysis evidenced the enhanced level of apoptosis in tumors treated with combination therapeutics delivered by nanoparticle, especially as it pertains to bortezomib activity. Given the reduced systemic side effects, increased tolerated dosage, targeted drug delivery, and synchronized drug residency, NP(BTZ-DOX) exhibits significantly enhanced anticancer effects in ovarian cancer treatment. Although the tumor microenvironments in human patients are different from the xenografted tumors, the improved pharmacokinetics and biodistribution of NP(BTZ-DOX) can be foreseen based on their optimized physiochemical properties, which is promising to reignite the activity of bortezomib in solid tumor treatments and merit further development in the clinical setting.

No potential conflicts of interest were disclosed.

Conception and design: L. Wang, D. Wang, J. Luo

Development of methodology: L. Wang, J. Luo

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L. Wang, C. Shi, F.A. Wright, D. Guo, R.J.H. Wojcikiewicz

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L. Wang, C. Shi, F.A. Wright, D. Wang, R.J.H. Wojcikiewicz, J. Luo

Writing, review, and/or revision of the manuscript: L. Wang, X. Wang, D. Wang, R.J.H. Wojcikiewicz, J. Luo

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): D. Guo

We greatly appreciate Prof. Stephan Wilkens (SUNY UMU) for the assistance in TEM analysis and Prof. Golam Mohi (SUNY UMU) for the help in blood cell analysis.

This work was financially supported by the NIH grant NIBIB 1R21EB019607 to J. Luo, New York State Health Department Peter T. Rowley Breast Cancer project DOH01-Rowley-2015-00067 to J. Luo, and Napi Family Research Awards to J. Luo.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Kim
A
,
Ueda
Y
,
Naka
T
,
Enomoto
T
. 
Therapeutic strategies in epithelial ovarian cancer
.
J Exp Clin Cancer Res
2012
;
31
:
14
.
2.
Foley
OW
,
Rauh-Hain
JA
,
del Carmen
MG
. 
Recurrent epithelial ovarian cancer: an update on treatment
.
Oncology
2013
;
27
:
288
94
,
98
.
3.
Gatti
L
,
Zuco
V
,
Zaffaroni
N
,
Perego
P
. 
Drug combinations with proteasome inhibitors in antitumor therapy
.
Curr Pharm Des
2013
;
19
:
4094
114
.
4.
Gottesman
MM
. 
Mechanisms of cancer drug resistance
.
Annu Rev Med
2002
;
53
:
615
27
.
5.
Longley
D
,
Johnston
P
. 
Molecular mechanisms of drug resistance
.
J Pathol
2005
;
205
:
275
92
.
6.
Gillet
J-P
,
Gottesman
MM
. 
Mechanisms of multidrug resistance in cancer
.
Methods Mol Biol
2010
;
596
:
47
76
.
7.
Woodcock
J
,
Griffin
JP
,
Behrman
RE
. 
Development of novel combination therapies
.
N Engl J Med
2011
;
364
:
985
7
.
8.
Yang
CH
,
Gonzalez-Angulo
AM
,
Reuben
JM
,
Booser
DJ
,
Pusztai
L
,
Krishnamurthy
S
, et al
Bortezomib (VELCADE) in metastatic breast cancer: pharmacodynamics, biological effects, and prediction of clinical benefits
.
Ann Oncol
2006
;
17
:
813
7
.
9.
Moreau
P
,
Richardson
PG
,
Cavo
M
,
Orlowski
RZ
,
San Miguel
JF
,
Palumbo
A
, et al
Proteasome inhibitors in multiple myeloma: 10 years later
.
Blood
2012
;
120
:
947
59
.
10.
Adams
J
,
Kauffman
M
. 
Development of the proteasome inhibitor Velcade™(Bortezomib)
.
Cancer Invest
2004
;
22
:
304
11
.
11.
Bedford
L
,
Lowe
J
,
Dick
LR
,
Mayer
RJ
,
Brownell
JE
. 
Ubiquitin-like protein conjugation and the ubiquitin–proteasome system as drug targets
.
Nat Rev Drug Discov
2011
;
10
:
29
46
.
12.
Orlowski
RZ
,
Kuhn
DJ
. 
Proteasome inhibitors in cancer therapy: lessons from the first decade
.
Clin Cancer Res
2008
;
14
:
1649
57
.
13.
Rapino
F
,
Naumann
I
,
Fulda
S
. 
Bortezomib antagonizes microtubule-interfering drug-induced apoptosis by inhibiting G2/M transition and MCL-1 degradation
.
Cell Death Dis
2013
;
4
:
e925
.
14.
Bruning
A
,
Burger
P
,
Vogel
M
,
Rahmeh
M
,
Friese
K
,
Lenhard
M
, et al
Bortezomib treatment of ovarian cancer cells mediates endoplasmic reticulum stress, cell cycle arrest, and apoptosis
.
Invest New Drugs
2009
;
27
:
543
51
.
15.
Orlowski
RZ
,
Nagler
A
,
Sonneveld
P
,
Blade
J
,
Hajek
R
,
Spencer
A
, et al
Randomized phase III study of pegylated liposomal doxorubicin plus bortezomib compared with bortezomib alone in relapsed or refractory multiple myeloma: combination therapy improves time to progression
.
J Clin Oncol
2007
;
25
:
3892
901
.
16.
Milano
A
,
Iaffaioli
RV
,
Caponigro
F
. 
The proteasome: a worthwhile target for the treatment of solid tumours?
Eur J Cancer
2007
;
43
:
1125
33
.
17.
Huang
Z
,
Wu
Y
,
Zhou
X
,
Xu
J
,
Zhu
W
,
Shu
Y
, et al
Efficacy of therapy with bortezomib in solid tumors: a review based on 32 clinical trials
.
Future Oncol
2014
;
10
:
1795
807
.
18.
Parma
G
,
Mancari
R
,
Del Conte
G
,
Scambia
G
,
Gadducci
A
,
Hess
D
, et al
An open-label phase 2 study of twice-weekly bortezomib and intermittent pegylated liposomal doxorubicin in patients with ovarian cancer failing platinum-containing regimens
.
Int J Gynecol Cancer
2012
;
22
:
792
800
.
19.
Wheat
LM
,
Kohlhaas
SL
,
Monbaliu
J
,
De Coster
R
,
Majid
A
,
Walewska
RJ
, et al
Inhibition of bortezomib-induced apoptosis by red blood cell uptake
.
Leukemia
2006
;
20
:
1646
9
.
20.
Attar
EC
,
De Angelo
DJ
,
Supko
JG
,
D'Amato
F
,
Zahrieh
D
,
Sirulnik
A
, et al
Phase I and pharmacokinetic study of bortezomib in combination with idarubicin and cytarabine in patients with acute myelogenous leukemia
.
Clin Cancer Res
2008
;
14
:
1446
54
.
21.
Ogawa
Y
,
Tobinai
K
,
Ogura
M
,
Ando
K
,
Tsuchiya
T
,
Kobayashi
Y
, et al
Phase I and II pharmacokinetic and pharmacodynamic study of the proteasome inhibitor bortezomib in Japanese patients with relapsed or refractory multiple myeloma
.
Cancer Sci
2008
;
99
:
140
4
.
22.
Unezaki
S
,
Maruyama
K
,
Hosoda
J-I
,
Nagae
I
,
Koyanagi
Y
,
Nakata
M
, et al
Direct measurement of the extravasation of polyethyleneglycol-coated liposomes into solid tumor tissue by in vivo fluorescence microscopy
.
Int J Pharm
1996
;
144
:
11
7
.
23.
Hu
C-MJ
,
Zhang
L
. 
Nanoparticle-based combination therapy toward overcoming drug resistance in cancer
.
Biochem Pharmacol
2012
;
83
:
1104
11
.
24.
Goldman
A
,
Kulkarni
A
,
Kohandel
M
,
Pandey
P
,
Rao
P
,
Natarajan
SK
, et al
Rationally designed 2-in-1 nanoparticles can overcome adaptive resistance in cancer
.
ACS Nano
2016
;
10
:
5823
34
.
25.
Kratz
F
,
Warnecke
A
. 
Finding the optimal balance: Challenges of improving conventional cancer chemotherapy using suitable combinations with nano-sized drug delivery systems
.
J Control Release
2012
;
164
:
221
35
.
26.
Shen
S
,
Du
X-J
,
Liu
J
,
Sun
R
,
Zhu
Y-H
,
Wang
J
. 
Delivery of bortezomib with nanoparticles for basal-like triple-negative breast cancer therapy
.
J Control Release
2015
;
208
:
14
24
.
27.
Thamake
SI
,
Raut
SL
,
Gryczynski
Z
,
Ranjan
AP
,
Vishwanatha
JK
. 
Alendronate coated poly-lactic-co-glycolic acid (PLGA) nanoparticles for active targeting of metastatic breast cancer
.
Biomaterials
2012
;
33
:
7164
73
.
28.
Swami
A
,
Reagan
MR
,
Basto
P
,
Mishima
Y
,
Kamaly
N
,
Glavey
S
, et al
Engineered nanomedicine for myeloma and bone microenvironment targeting
.
Proc Natl Acad Sci
2014
;
111
:
10287
92
.
29.
Su
J
,
Chen
F
,
Cryns
VL
,
Messersmith
PB
. 
Catechol polymers for pH-responsive, targeted drug delivery to cancer cells
.
J Am Chem Soc
2011
;
133
:
11850
3
.
30.
Ashley
JD
,
Stefanick
JF
,
Schroeder
VA
,
Suckow
MA
,
Kiziltepe
T
,
Bilgicer
B
. 
Liposomal bortezomib nanoparticles via boronic ester prodrug formulation for improved therapeutic efficacy in vivo
.
J Med Chem
2014
;
57
:
5282
92
.
31.
Liu
R
,
Guo
Y
,
Odusote
G
,
Qu
F
,
Priestley
RD
. 
Core–shell Fe3O4 polydopamine nanoparticles serve multipurpose as drug carrier, catalyst support and carbon adsorbent
.
ACS Appl Mater Interfaces
2013
;
5
:
9167
71
.
32.
Zuccari
G
,
Milelli
A
,
Pastorino
F
,
Loi
M
,
Petretto
A
,
Parise
A
, et al
Tumor vascular targeted liposomal-bortezomib minimizes side effects and increases therapeutic activity in human neuroblastoma
.
J Control Release
2015
;
211
:
44
52
.
33.
Matsumura
Y
,
Maeda
H
. 
A new concept for macromolecular therapeutics in cancer chemotherapy: mechanism of tumoritropic accumulation of proteins and the antitumor agent smancs
.
Cancer Res
1986
;
46
:
6387
92
.
34.
Maeda
H
,
Wu
J
,
Sawa
T
,
Matsumura
Y
,
Hori
K
. 
Tumor vascular permeability and the EPR effect in macromolecular therapeutics: a review
.
J Control Release
2000
;
65
:
271
84
.
35.
Luo
J
,
Xiao
K
,
Li
Y
,
Lee
JS
,
Shi
L
,
Tan
YH
, et al
Well-defined, size-tunable, multifunctional micelles for efficient paclitaxel delivery for cancer treatment
.
Bioconjug Chem
2010
;
21
:
1216
24
.
36.
Lee
H
,
Fonge
H
,
Hoang
B
,
Reilly
RM
,
Allen
C
. 
The effects of particle size and molecular targeting on the intratumoral and subcellular distribution of polymeric nanoparticles
.
Mol Pharm
2010
;
7
:
1195
208
.
37.
Cai
L
,
Xu
G
,
Shi
C
,
Guo
D
,
Wang
X
,
Luo
J
. 
Telodendrimer nanocarrier for co-delivery of paclitaxel and cisplatin: a synergistic combination nanotherapy for ovarian cancer treatment
.
Biomaterials
2015
;
37
:
456
68
.
38.
Li
Y
,
Xiao
K
,
Luo
J
,
Xiao
W
,
Lee
JS
,
Gonik
AM
, et al
Well-defined, reversible disulfide cross-linked micelles for on-demand paclitaxel delivery
.
Biomaterials
2011
;
32
:
6633
45
.
39.
Xiao
K
,
Li
Y
,
Luo
J
,
Lee
JS
,
Xiao
W
,
Gonik
AM
, et al
The effect of surface charge on in vivo biodistribution of PEG-oligocholic acid based micellar nanoparticles
.
Biomaterials
2011
;
32
:
3435
46
.
40.
Xiao
K
,
Luo
J
,
Fowler
WL
,
Li
Y
,
Lee
JS
,
Xing
L
, et al
A self-assembling nanoparticle for paclitaxel delivery in ovarian cancer
.
Biomaterials
2009
;
30
:
6006
16
.
41.
Huang
W
,
Wang
X
,
Shi
C
,
Guo
D
,
Xu
G
,
Wang
L
, et al
Fine-tuning vitamin E-containing telodendrimers for efficient delivery of gambogic acid in colon cancer treatment
.
Mol Pharm
2015
;
12
:
1216
29
.
42.
Shi
C
,
Guo
D
,
Xiao
K
,
Wang
X
,
Wang
L
,
Luo
J
. 
A drug-specific nanocarrier design for efficient anticancer therapy
.
Nat Commun
2015
;
6
:
7449
.
43.
Wang
X
,
Bodman
A
,
Shi
C
,
Guo
D
,
Wang
L
,
Luo
J
, et al
Tunable lipidoid-telodendrimer hybrid nanoparticles for intracellular protein delivery in brain tumor treatment
.
Small
2016
;
12
:
4185
92
.
44.
Wang
X
,
Shi
C
,
Zhang
L
,
Bodman
A
,
Guo
D
,
Wang
L
, et al
Affinity-controlled protein encapsulation into sub-30 nm telodendrimer nanocarriers by multivalent and synergistic interactions
.
Biomaterials
2016
;
101
:
258
71
.
45.
Brooks
WL
,
Sumerlin
BS
. 
Synthesis and applications of boronic acid-containing polymers: from materials to medicine
.
Chem Rev
2016
;
116
:
1375
97
.
46.
Li
Y
,
Xiao
W
,
Xiao
K
,
Berti
L
,
Luo
J
,
Tseng
HP
, et al
Well-defined, reversible boronate crosslinked nanocarriers for targeted drug delivery in response to acidic pH values and cis-diols
.
Angew Chem Int Ed Engl
2012
;
51
:
2864
9
.
47.
Tannock
IF
,
Rotin
D
. 
Acid pH in tumors and its potential for therapeutic exploitation
.
Cancer Res
1989
;
49
:
4373
84
.
48.
Lee
ES
,
Gao
Z
,
Bae
YH
. 
Recent progress in tumor pH targeting nanotechnology
.
J Control Release
2008
;
132
:
164
70
.
49.
Kato
J
,
Li
Y
,
Xiao
K
,
Lee
JS
,
Luo
J
,
Tuscano
JM
, et al
Disulfide cross-linked micelles for the targeted delivery of vincristine to B-cell lymphoma
.
Mol Pharm
2012
;
9
:
1727
35
.
50.
Springsteen
G
,
Wang
B
. 
Alizarin Red S. as a general optical reporter for studying the binding of boronic acids with carbohydrates
.
Chem Commun
2001
;
17
:
1608
9
.
51.
Chou
T-C
. 
Drug combination studies and their synergy quantification using the Chou-Talalay method
.
Cancer Res
2010
;
70
:
440
6
.
52.
Xu
Q
,
Farah
M
,
Webster
JM
,
Wojcikiewicz
RJ
. 
Bortezomib rapidly suppresses ubiquitin thiolesterification to ubiquitin-conjugating enzymes and inhibits ubiquitination of histones and type I inositol 1,4,5-trisphosphate receptor
.
Mol Cancer Ther
2004
;
3
:
1263
9
.
53.
Springsteen
G
,
Wang
B
. 
A detailed examination of boronic acid–diol complexation
.
Tetrahedron
2002
;
58
:
5291
300
.
54.
Mitsiades
N
,
Mitsiades
CS
,
Richardson
PG
,
Poulaki
V
,
Tai
Y-T
,
Chauhan
D
, et al
The proteasome inhibitor PS-341 potentiates sensitivity of multiple myeloma cells to conventional chemotherapeutic agents: therapeutic applications
.
Blood
2003
;
101
:
2377
80
.
55.
Torchilin
VP
. 
Micellar nanocarriers: pharmaceutical perspectives
.
Pharm Res
2007
;
24
:
1
16
.