A nanoparticle-based combination chemotherapy delivery system for enhanced tumor killing by dynamic rewiring of signaling pathways

Abstract

Exposure to the EGFR (epidermal growth factor receptor) inhibitor erlotinib promotes the dynamic rewiring of apoptotic pathways, which sensitizes cells within a specific period to subsequent exposure to the DNA-damaging agent doxorubicin. A critical challenge for translating this therapeutic network rewiring into clinical practice is the design of optimal drug delivery systems. We report the generation of a nanoparticle delivery vehicle that contained more than one therapeutic agent and produced a controlled sequence of drug release. Liposomes, representing the first clinically approved nanomedicine systems, are well-characterized, simple, and versatile platforms for the manufacture of functional and tunable drug carriers. Using the hydrophobic and hydrophilic compartments of liposomes, we effectively incorporated both hydrophobic (erlotinib) and hydrophilic (doxorubicin) small molecules, through which we achieved the desired time sequence of drug release. We also coated the liposomes with folate to facilitate targeting to cancer cells. When compared to the time-staggered application of individual drugs, staggered release from tumor-targeted single liposomal particles enhanced dynamic rewiring of apoptotic signaling pathways, resulting in improved tumor cell killing in culture and tumor shrinkage in animal models.

Fig. 1. Characterization of the combination therapeutic–loaded liposomal system

(A) Cryogenic transmission electron micrograph of dual drug–loaded liposomes. Scale bar, 100 nm. (B) Schematic of dual loading of a small-molecule inhibitor (erlotinib, blue) into the hydrophobic, vesicular wall compartment and of a cytotoxic agent (doxorubicin, green) into the aqueous, hydrophilic interior.

Fig. 2. Evaluation of the dual drug–loaded liposomal system in vitro

(A) Drug release from dual drug–loaded liposomes in an excess volume of PBS (pH 7.4) at 37°C under agitation. (B) Comparative cytotoxicity of dual drug–loaded liposome relative to the single drug–loaded liposome in BT-20 (TNBC) and A549 (NSCLC) cell lines. This was measured by staining against cleaved caspase-3 and PARP. (C) Cleaved caspase-8 in BT-20 cells (top) and A549 cells (bottom) after addition of the indicated liposomes. In the gel images, red is actin and green is cleaved caspase-8. Quantification shown below stained gel images corresponds to relative signal of cleaved caspase-8 to actin. Data are presented as mean ± SEM of three experiments. (D) Dynamics of pERK in BT-20 cells (top) and A549 cells (bottom) after addition of the indicated liposomes. Data are representative of three experiments, and for quantification, pERK abundance was normalized to actin. (E) γH2AX formation in BT-20 cells (top) and A549 cells (bottom) after addition of the indicated liposomes. Data are representative of three experiments, and for quantification, γH2AX abundance was normalized to actin. D, doxorubicin only (single drug); DE, doxorubicin and erlotinib (dual drug).

Fig. 3. Decoration of combination therapeutic–loaded liposomes for targeted delivery

(A) Schematic of addition of DSPE-PEG_2K (0.5 mol % ratio) to minimize nonspecific protein binding, DSPE-PEG_2K^Cy5.5 (0.1 mol % ratio) for fluorescent tracking, and DSPE-PEG_5K-folate (0.5 mol % ratio) for cell-targeted delivery. (B) Cell uptake of the folate-targeted liposomes in BT-20 and A549 cells, visualized by confocal microscopy. Blue, nuclei labeled with DAPI; green, actin labeled with phalloidin-568 (Ph568); red, DSPE-PEG_2K^Cy5.5–labeled DFP liposomes. Bottom panels represent the fluorescence in each channel; top panels are merged images. (C) Cell-associated fluorescence measured by flow cytometry as a function of nanoparticle (NP) concentration in both cell lines after incubation with liposomes containing or lacking folate at 37°C for 1 hour. Top: Cy5.5 (λ_ex = 675 nm, λ_ex = 710 nm) corresponding to liposomal association for both targeted (DFP-Cy5.5, squares) and untargeted control (DP-Cy5.5, circles). Bottom: Amount of doxorubicin associated with the cells [doxorubicin fluorescence, (λ_ex = 480 nm, λ_ex = 560 nm)] for both targeted (DFP-Cy5.5, squares) and untargeted control (DP-Cy5.5, circles). Data are presented as mean ± SEM of three experiments.

Fig. 4. Biological performance of the folate-targeted liposomal system in vivo

(A) Biodistribution panel of folate-targeted liposomes containing no drug (tracked through Cy5.5 fluorescence) that were intravenously administered to BALB/c mice. In situ quantification (region identified with circle at 30 min after injection) of liver-associated nanoparticle fluorescence (normalized to injected dose) presented above each time point. (B) Circulation data displayed as percent injected dose (ID), on the basis of nanoparticle Cy5.5 fluorescence recovered in blood samples. Half-life calculated on the basis of a two-compartment model and presented as mean ± SEM. (C) Tumor visualization (left; visualized as firefly luciferase expressed in the xenografted cells) and nanoparticle visualization (right; visualized by Cy5.5 fluorescence) 30 days after injection of single 0.1-ml administration of folate-targeted empty liposomes (NFP-Cy5.5) to NCR nude mice bearing BT-20 or A549 xenografts on the hind flanks. The same animals are shown on the left and right images. See fig. S2 for tissue necropsy.

Fig. 5. Effect of dual drug– or single drug–loaded, folate-targeted liposomes on A549 and BT-20 tumor size

(A) A549-luciferase–expressing, xenograft-bearing NCR nude mice; n = 5, quantification representative of mean ± SEM (see fig. S5 for all five mice). (B) BT-20-luciferase–expressing, xenograft-bearing NCR nude mice; n = 5, quantification representative of mean ± SEM (see fig. S4 for all five mice). Tumor-imaging data for dual drug (DEFP, top), single drug (DFP, middle), and untreated control, along with luminescence quantification (reported as fold initial tumor luminescence, presented on a semi-log plot) corresponding to tumor size as a function of time, after a single administration of drug (1 mg/kg)–loaded liposomal formulations. Animals with tumor reaching 1 cm were sacrificed. An unpaired, two-tailed t test comparing the DFP and DEFP at the terminal 32-day time point shows statistical significance with P values of 0.0057 and 0.001 for treated A549 and BT-20 xenograft–bearing mice, respectively. A one-way analysis of variance (ANOVA) comparing all treatments for the duration of the experiment (all time points) was also performed for each xenograft cell line, and P values of 0.0024 and 0.0010 were obtained for A549 and BT-20 cells, respectively.

Fig. 6. Developing the RTK inhibitor–cytotoxic agent combination liposomal system as a platform for dual-drug delivery

(A) In vitro drug release from dual drug–loaded folate-targeted liposomal formulations. Top: Release from folate and PEG–containing liposomes (FP) with doxorubicin and the indicated RTK inhibitor. Bottom: Release from folate and PEG–containing liposomes (FP) with cisplatin and the indicated RTK inhibitor. Liposomes were incubated with PBS (pH 7.4) at 37°C under agitation in sink conditions. Data are presented as mean of triplicate experiments. (B) Cytotoxicity of dual drug–loaded (RTK and doxorubicin) liposomes compared with that of single drug–loaded (doxorubicin) liposomes against BT-20 and A549 cells. Data are presented as mean of triplicate experiments. (C) Cytotoxicity of dual drug–loaded (RTK and cisplatin) liposomes compared with that of single drug–loaded (cisplatin) liposomes against BT-20 and A549 cells. Data are presented as mean of triplicate experiments. D, doxorubi-cin; A, afatinib; E, erlotinib; G, gefitinib; L, lapatinib; C, cisplatin; F, folate; P, PEG.