Tumor selective uptake of drug-nanodiamond complexes improves therapeutic outcome in pancreatic cancer

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is one of the leading causes of cancer-related deaths and novel treatment approaches are urgently needed. Here we show that poly(ethylene glycol)-functionalized nanodiamonds loaded with doxorubicin (ND-PEG-DOX) afforded a considerable improvement over free drug in an orthotopic pancreatic xenograft model. ND-PEG-DOX complexes were also superior to free DOX in 3-dimensional (3D) tumor spheroids of PDAC. ND-PEG showed no cytotoxicity towards macrophages, and histopathological analysis showed no abnormalities of major organs upon in vivo administration of ND-PEG-DOX. These results provide evidence that ND-mediated drug delivery may serve as a means of improving the therapeutic outcome in PDAC.

Graphical abstract legend

We investigated the therapeutic potential of poly(ethylene glycol)-functionalized nanodiamonds (NDs) loaded with doxorubicin (ND-PEG-DOX) using an array of in vitro and in vivo models of pancreatic ductal adenocarcinoma (PDAC). Our results showed that ND-PEG-DOX were more effective than free drug in 3D models of PDAC, but not when cells were grown in 2D. We also demonstrated tumor-selective uptake of ND-PEG-DOX in vivo and we documented a superior antitumor activity when compared to free drug.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) or pancreatic cancer is the fourth leading cause of cancer-related death.¹ Most patients present with an advanced stage of disease and a dismal prognosis.^2,3 Gemcitabine, a nucleoside analogue of deoxycytidine, has typically been used as the standard first-line therapy for PDAC, but therapeutic responses have been disappointing.⁴ Combination chemotherapy, e.g., gemcitabine plus nab-paclitaxel, has recently overtaken gemcitabine as the standard-of-care for patients with pancreatic cancer, but this has also met with limited success.⁵ Similarly, the use of FOLFIRINOX (irinotecan, 5-fluorouracil, oxaliplatin, and leucovorin) for the treatment of metastatic and locally advanced disease has increased, but toxicities in most cases have been a limiting issue.⁶ Several targeted therapies have been evaluated as treatments for metastatic PDAC in combination with standard-of-care treatments, but most have failed to progress to the clinic.^7,8 There is, therefore, an urgent need for new therapeutic approaches with which to combat this devastating disease.

Considering the current challenges of targeted therapies in terms of minimizing toxicities and improving efficacies, reports from several groups including ours suggest the potential of nanomedicine strategies for targeted drug delivery in cancer patients.^9-11 However, despite tremendous potential, cancer nanomedicine is also faced with considerable challenges regarding the design of the nanomaterials, as well as an incomplete understanding of tumor biology and the nano-bio-interface.¹² Notwithstanding, nanoparticle carriers with appropriately tailored physicochemical properties, which can negotiate biological barriers and safely deliver their cargo (i.e., drug, or siRNA) have the potential to improve therapeutic efficacy in cancer patients.^13,14 Nanodiamonds (NDs) are endowed with several unique properties that make them promising for biomedical applications including a chemically inert core and unique facet-specific surface electrostatic potentials.^15,16 Recent studies have disclosed that NDs may be used to overcome drug resistance in murine models of liver and breast cancer through delivery and sustained release of chemotherapeutic drugs (doxorubicin, epirubicin).¹⁷-¹⁹ NDs were also used to deliver CpG oligonucleotides for sustained immunostimulation in preclinical models of melanoma and breast cancer.²⁰ Furthermore, evidence was provided that NDs are well-tolerated in non-human primates at clinically relevant doses.²¹ However, the potential application of NDs for the treatment of PDAC has not been evaluated. We focused on PDAC as there is currently no efficient therapy. Our results show that PEG-functionalized NDs can be used to deliver anticancer drugs in PDAC with superior therapeutic outcome as compared to the free drug.

Methods

Nanomaterials and other reagents

Nanodiamonds (ND) with different surface modifications (i.e., ND-NH₂, ND-COOH, ND-PEG) were provided by PlasmaChem GmbH (Berlin, Germany). The primary particle size of the NDs is between 20 and 50 nm according to the manufacturer. The chemotherapeutic drug doxorubicin was purchased from LC Laboratories (Woburn, MA).

Dynamic light scattering (DLS)

The hydrodynamic diameter was determined from number-based distributions using a Malvern Zetasizer Nano ZS (laser, angle = 173°) at a ND of 50 μg/mL. Samples were prepared in deionized water, PBS or DMEM/RPMI cell culture medium and sonicated for 1 min prior to the measurements. The ζ potential determinations were performed using folded capillary cells in automatic mode. Stability of the samples was measured at various time-points after incubation of NDs in DMEM/RPMI supplemented or not with 10% FBS.

Fourier transform infrared spectroscopy (FTIR)

FTIR spectra of NDs were measured using JASCO- FT/IR-4600 spectrometer (JASCO Instrument Corp., USA). ND powders (1 mg) were placed on the diamond chamber and the spectra were recorded immediately. A signal from a blank chamber was subtracted as a background.

Endotoxin test (LAL assay)

Endotoxin contamination was evaluated using the endpoint chromogenic Limulus Amebocyte Lysate (LAL) assay (Lonza), as described previously.²² The three NDs were all found to be endotoxin-free (i.e., endotoxin level below 0.5 EU/mL) (data not shown).

Pancreatic cancer cell lines

BxPC3 and PANC-1 cell lines were obtained from the American Type Culture Collection (ATCC). 6741 is an annotated primary pancreatic cancer cell line developed in our laboratory (Mayo Clinic). BxPC3/PANC-1 and 6741 cells were expanded and maintained in DMEM and DMEM-F12 media, respectively, supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100 mg/ml streptomycin at 37°C in a humidified 5% CO₂ incubator. 3D spheroids from BxPC3 and PANC-1 cells were grown as described previously.²³ Briefly, BxPC3 or PANC-1 cells were seeded onto round-bottom non-tissue culture treated 96 well-plates (Falcon) at a concentration of 2500 cells/well in 100 μL phenol red-free DMEM-F12 medium, containing 10% FBS and supplemented with 20% methyl cellulose stock solution (Sigma-Aldrich) and grown in 5% CO₂ at 37°C.

Macrophage culture and toxicity assay

The human myeloid leukemia cell line, PLB-985 (generous gift of Dr. Mary Dinauer, Indiana University Medical Center, Indianapolis, IN) was used as a model to evaluate in vitro toxicity of the NDs. Cells were maintained as described in the appendix. Cell viability following exposure to NDs was determined using the Alamar Blue^® assay (ThermoFisher Scientific) as described previously.²⁴ Quantification of cytokine (TNF-α) release was performed by LUMINEX^® assay.²⁴ Refer to the appendix for experimental details.

Drug absorption and release

DOX was loaded onto PEGylated NDs according to previously reported procedures.¹⁷ Refer to the appendix for details. The adsorption of DOX on NDs was confirmed by measuring the absorbance of DOX at 480 nm using a SHIMADZU-UV spectrometer. Similarly, the amount of unbound DOX was obtained by measuring the absorbance of DOX remaining in the supernatant after centrifuging 2 h at 14.000 rpm. The amount of DOX complexed with NDs was then calculated by subtracting the amount of unbound DOX. To determine drug release from NDs, 100 μL of ND-PEG-DOX was added to 1 mL of buffer solutions with different pH in 1.5 ml micro centrifuge tubes (7 tubes for each pH) and incubated at 37 °C. After 0, 2, 4, 6, and 12, 24 and 48 h, one tube for each pH was centrifuged at 14000 rpm for 2 h and DOX in the supernatant was measured by fluorescence. The percent of DOX released was calculated by multiplying the ratio of DOX released in supernatant and DOX initially adsorbed onto the NDs by 100. The percent release of DOX was plotted against time for each pH to obtain DOX release curves.

In vitro drug uptake

Cells were seeded at a density of 10,000 cells per well in 96-well plates. After incubation for 18-24 h, fresh medium containing free DOX or ND-PEG-DOX (at the equivalent DOX concentration 2.5 μg/mL) was added to the cells. After 4 h of incubation cells were treated with 2.5 μg/mL Hoechst dye solution for another 30 min. Then, cell media were discarded and cells were carefully washed three times with PBS. Finally, 100 μL of PBS was added to each well and images were captured using a Zeiss LSM 880 confocal microscope with excitation at 361 and 485 nm for DAPI and DOX, respectively.

In vitro cytotoxicity assay

The cytotoxicity of free DOX and ND-PEG-DOX was evaluated by colorimetric MTS assay. PANC-1 and 6741 cells were seeded in 96-well plates at a density of 5,000 cells per well for 24 h. Then the medium was replaced with a fresh medium containing free DOX and ND-PEG-DOX (the equivalent DOX concentrations were set to be 0.312, 0.625, 1.25, 2.5, 5 and 10 μM, respectively). After incubation for 72 h, MTS assays were performed according to the manufacturer’s instructions. The absorbance at 490 nm was recorded using a Spectramax i3x microplate reader (Biotech). The in vitro cell viability in PDAC cells grown as 3D spheroids was assessed by using the CellTiter-Glo® 3D Cell Viability Assay (Promega), an assay that has been specifically developed for 3D spheroids.

Transmission electron microscopy (TEM)

Cells incubated with ND or ND-PEG-DOX were examined by TEM in order to determine cellular uptake and subcellular localization of the NDs; refer to the appendix for details.

In vivo and ex vivo fluorescence imaging

Male NSG mice (6-8 weeks old) were purchased from Charles River Laboratories and housed in the institutional animal facilities. All animal work was performed under protocols approved by Mayo Clinic Institutional Animal Care and Use Committee. Fluorescence snapshots of mice were acquired after administration of DOX and ND-PEG-DOX using an in vivo imaging system (In-Vivo Xtreme™) (Bruker Corp) at 6, 24 and 48 h. At 48 h post-injection, the mice were sacrificed and their organs such as liver, lung, spleen, kidneys, and tumor were harvested for imaging, semiquantitative biodistribution and tumor accumulation analysis. Consult appendix for experimental details.

In vivo antitumor efficacy

Male NSG mice (6-8 weeks old) were purchased from Charles River Laboratories and housed in the institutional animal facilities. All animal studies were performed under protocols approved by Mayo Clinic Institutional Animal Care and Use Committee. To establish the tumor model, approximately 1.0×10⁶ GFP-Luciferase transfected 6741 cells suspended in 100 μL PBS containing 20% of matrigel were slowly injected orthotopically into the pancreas of the mice. The tumors were allowed to grow for three weeks and then randomly divided into four groups (n=5) for the treatment with ND-PEG-DOX (10 mg/kg of DOX equivalent) or free DOX (10 mg/kg DOX mice) or ND-PEG 2x/wk for three weeks by i.p. injection. The mice were imaged for luciferase activity to ensure similar tumor growth among the treatment groups before the start of treatment and also confirmed by palpation. PBS was used as vehicle control. After three weeks of treatment, mice were sacrificed and tumors were harvested, measured with slide calipers and weighed. Tumor volume was calculated using an equation: V = 1/2 × a× b² mm³, where a is the largest diameter, and b is the smallest diameter.

Histological analysis

Tumors were harvested and fixed in neutral buffered 10% formalin at room temperature for 24 h prior to embedding in paraffin and sectioning. Sections were deparaffinized and subjected to staining with hematoxylin and eosin (H&E) and Ki67, according to the manufacturers’ instructions. For Ki67 staining, stable diaminobenzidine was used as a chromogen substrate and sections were counterstained with hematoxylin. Slides were digitized with Aperio AT2 slide scanner (Leica) and analyzed with imagescope software (Leica).

Western blot analysis

Protein expression in harvested tissues was analyzed by western blot. The following antibodies were used: β-actin, Ki-67, P-53 (SC-126) AKT and P-AKT (SC-7985) and HRP–conjugated secondary antibodies (Santa Cruz Biotechnology), BCL2 (15071S) and ABCG2 (4477S) (Cell Signaling Technology). For further experimental details, refer to appendix.

Statistics

Data were plotted as mean values ± SD. The probability of significant differences between groups was analyzed by independent-samples t-test. p<0.05 (*) and p<0.01 (**) were considered as statistically significant and highly statistically significant, respectively.

Results

Cytotoxicity assessment of NDs alone

Prior to testing of NDs for drug delivery, we evaluated the nanoparticles for their potential cytotoxicity using the human macrophage-differentiated PLB-985 cell line. To this end, a comparison was made between NDs with different surface functionalization (i.e., ND-NH2, ND-COOH, ND-PEG) and cells were cultivated in the presence or absence of 10% FBS in order to evaluate the potential impact of the protein corona. We employed a macrophage cell line because macrophages are the first line of cellular defense against foreign pathogens or particles,²⁵ and we used a cell line instead of primary cells in order to reduce the variability that may occur between individual donors. The three NDs showed no cytotoxicity at doses up to 25 μg/mL, irrespective of whether the cells were grown in the absence (Figure 1A) or presence (Figure 1B) of FBS. Dose-dependent cytotoxicity was observed at higher doses (i.e., 50 and 100 μg/mL) when cells were grown without serum, but there was no cytotoxicity even at 100 μg/mL in cells grown in medium with FBS (Figure 1A, B). No secretion of pro-inflammatory TNF-α was observed when cells were exposed to NDs (25 μg/mL) in the presence or absence of 10% FBS (data not shown). LPS was included as a positive control. DLS measurements performed on NDs in RPMI-1640 medium showed that NDs were stabilized and the ζ potential was equalized for all NDs in the presence of serum (Supplementary Table S1).

Figure 1.

Cytotoxicity assessment of NDs. NDs with different surface modifications (i.e., ND-NH₂, ND-COOH, ND-PEG) were evaluated for potential cytotoxicity using the human macrophage-differentiated cell line, PLB-985. Differentiation of cells was confirmed by detection of surface markers and on the basis of morphological changes. Cells were exposed to NDs for 24 h in the absence (A) or presence (B) of 10% FBS. Cell viability was monitored using the Alamar Blue^® assay. Data shown are mean values ± S.D. of three independent experiments each performed in triplicate. *p<0.05 by Student’s t-test.

Characterization of ND-PEG and ND-PEG-DOX

Because the PDAC cell lines are grown in DMEM, DLS measurements of ND-PEG and ND-PEG-DOX were also conducted in water, PBS, DMEM and DMEM containing 10% FBS, as reported in Table 1. Notably, the size of either ND-PEG or ND-PEG-DOX increased significantly in high salt medium (eg., PBS or DMEM) compared to that in water. This is not unexpected, since there are multiple instances in the literature that demonstrate a similar increase in size of nanomaterials in high salt medium that may be attributed to aggregation of the individual particles.^26,27 On the other hand, the presence of 10% serum seemed to reduce aggregation in high salt medium presumably as a result of the stabilization of individual particles by formation of a protein corona on the particle surface (Table 1). The hydrodynamic size and zeta potential of ND-PEG and ND-PEG-DOX in DMEM supplemented with 10% FBS are illustrated in Figure 2A and 2B. Although there was no significant change in size between ND-PEG and ND-PEG-DOX, the ζ potential shifted towards neutral from a strongly negative value for ND-PEG alone indicating successful conjugation of the drug (discussed further below). Furthermore, the size and polydispersity index of ND-PEG and ND-PEG-DOX did not change significantly after 72 h incubation in DMEM containing 10% FBS suggesting a highly stable dispersion of the NDs in physiologically relevant conditions (Figure 2 C-D).

Table 1.

Hydrodynamic size of ND-PEG and ND-PEG-DOX.

	Water		PBS		DMEM		DMEM + 10% FBS
	size (nm)	PDI	size (nm)	PDI	size (nm)	PDI	size (nm)	PDI
ND-PEG	50 ± 2	0.25 ± 0.01	830 ± 121	0.23 ± 0.01	912 ± 75	0.3 ± 0.04	68 ± 10	0.22 ± 0.01
ND-PEG-DOX	44 ± 5.5	0.27 ± 0.01	962 ± 140	0.24 ± 0.02	1348 ± 117	0.3 ± 0.09	76 ± 14	0.22 ± 0.04

Figure 2.

Physical characterization of ND and ND-DOX. (A) Hydrodynamic size distribution and (B) zeta potential of ND-PEG and ND-PEG-DOX in DMEM containing 10% FBS. Stability analysis of ND-PEG and ND-PEG-DOX by monitoring (C) hydrodynamic size and (D) polydispersity index (PDI) for 72 h in DMEM containing 10% FBS.

We also characterized the lyophilized ND-PEG-DOX complexes along with ND-PEG or DOX by FTIR spectroscopy. The FTIR spectrum of ND-PEG-DOX contained signature peaks from both ND-PEG and DOX (Figure 3A). Similarly, the UV-Vis absorption spectrum of ND-PEG-DOX demonstrated the characteristic peak of DOX at 500 nm in addition to the appearance of the ND-PEG like spectrum (Figure 3B). These data clearly suggest binding of DOX with ND-PEG, as reported previously for non-PEGylated NDs.¹⁷

Figure 3.

DOX binding and release profile of ND-DOX. Confirmation of binding of doxorubicin (DOX) on ND-PEG by (A) FTIR and (B) UV-Visible absorbance analysis. Representative peaks are indicated with arrows. (C) Drug release profile in different media with varying pH. (D) TEM images of the primary PDAC derived cell line 6741 incubated with ND-PEG and ND-PEG-DOX, respectively. Internalization of NDs could be clearly observed (marked by black arrows) in cells exposed to ND-PEG and ND-PEG-DOX.

Efficiency of ND loading and release of DOX

The amount of DOX loaded into ND-PEG particles was calculated following the method described previously.¹⁷ We found ~75% of the added DOX to be complexed with the ND-PEG leaving ~25% of the DOX as free drug in the supernatant when an initial loading ratio of 5:1 of ND-PEG:DOX was used. The release kinetics of DOX from ND-PEG-DOX in various pH media revealed a pH-dependent release of DOX from ND-PEG-DOX (Figure 3C). There was a very limited amount of DOX release (~10% and ~20% up to 48 h at pH 10 and 7.4, respectively) in high pH media whereas the release kinetics became faster with decreasing pH. Approximately, ~65%, ~80% and ~95% of the loaded DOX was released in pH 6.5, 5 and 2 respectively. We also evaluated ND-NH₂ and ND-COOH with respect to drug loading and drug release and noted that drug loading efficiency was reduced for ND-NH₂ when compared to ND-COOH and ND-PEG, and we found that there was no appreciable degree of DOX release for ND-COOH at different pH values while drug release was observed for ND-PEG and ND-NH₂ at low pH (Supplementary Figure S1A-B). Taken together, ND-PEG presented the most favorable combination of drug loading and drug release profiles of the three NDs tested.

In vitro cellular uptake of ND-PEG-DOX

We then incubated ND-PEG or ND-PEG-DOX with the primary 6741 cell line for 4 h and visualized the internalized nanoparticles by using TEM. As shown in Figure 3D, ND-PEG and ND-PEG-DOX were internalized and displayed some aggregation. Most of the particles aggregated close to the cell membrane, but some accumulation was seen in vesicles in the cytoplasm (Figure 3D). To analyze whether ND-PEG-DOX is capable of delivering DOX into pancreatic cancer cells, we studied cellular uptake of DOX in two PDAC-derived cell lines, the commercially available PANC-1 cell line and the primary cell line 6741 isolated from a patient-derived xenograft, and compared with the uptake of free DOX. DOX uptake was determined by analyzing its characteristic fluorescence signal. ND-PEG-DOX showed significantly more uptake when compared to free DOX in both cell lines (Figures 4A-B), thus demonstrating the efficacy of ND-PEG as a drug delivery agent. We quantified the amount of DOX in BxPC3 and PANC-1 cells 6 h after administration of ND-PEG-DOX or free DOX to cells and could confirm that the drug was retained to a significant extent when compared to free drug (Supplementary Figure S2).

Figure 4.

In vitro uptake and killing. Cellular uptake of free DOX and DOX delivered by ND-PEG-DOX in (A) 6741 and (B) PANC-1 cell lines. DOX uptake was determined by analyzing the fluorescence signal of DOX. Cells were counterstained with DAPI. Scale bar: 100 μm. (C) Cytotoxicity profile of DOX and ND-PEG-DOX in PANC-1 and 6741 cell lines. Cells were treated for 72 h and cell viability was determined by using the MTS assay.

In vitro cytotoxicity of ND-PEG-DOX

We then evaluated the in vitro cytotoxicity of ND-PEG-DOX in the PANC-1 and 6741 cell lines and compared with free DOX. Contrary to our expectations, free DOX showed higher cytotoxicity compared to ND-PEG-DOX in both cell lines (Figure 4C). Similarly, ND-PEG-DOX did not provide any appreciable advantage when compared to free DOX when administered to the BxPC3 cell line at equivalent concentrations of DOX (Supplementary Figure S3A). Tumor spheroids offer the advantage of providing a more representative model of therapeutic responses compared with conventional 2D cultures.²⁸ We therefore cultured the BxPC3 cell line as 3D spheroids as described before,²³ and evaluated the response to ND-PEG-DOX versus free drug. No difference was observed when cells were monitored at 24 h after administration of ND-PEG-DOX or free DOX (Supplementary Figure S3B). However, as shown in Figure 5A, ND-PEG-DOX was clearly superior to free DOX (at equivalent concentrations of DOX) when evaluated after 96 h. Notably, ND-PEG alone was non-toxic to PDAC cells (Figure 5B). Thus, not only was DOX still capable of triggering cell death in PDAC cells following its delivery using PEGylated NDs, but we could also demonstrate a clear advantage of ND-PEG-DOX in a 3D model of PDAC. The increased efficacy that we observed after a prolonged period of cell culture is in line with the observation that the drug is retained to a greater extent in cells after ND delivery; NDs may thus act as a slow-release depot for the drug. Using TEM, we could detect NDs as clusters within cells in the outermost cell layers of BxPC3 tumor spheroids; several representative TEM images are shown in Figure 5C.

Figure 5.

In vitro killing in 3D spheroids. (A) BxPC3 PDAC cells were grown as 3D spheroids and exposed to free DOX versus ND-PEG-DOX for 96 h. The concentrations shown are the concentrations of DOX and the concentration of NDs. Cell killing was determined by using the CellTiter-Glo® 3D cell viability assay. Data shown are mean values ± S.D. of three independent experiments. ***p<0.001. (B) Cells were also exposed to ND-PEG and ND-COOH alone without DOX and cell viability was evaluated as described above. (C) TEM imaging shows cellular uptake of ND-PEG (without drug) in tumor spheroids. The spheroids were exposed for 24 h. The panel to the left (scale bar: 1 μm) illustrates internalization by endocytosis, the middle panel (scale bar: 1 μm) shows small clusters of NDs in the cytoplasm of the cells, and the panel to the right shows the presence of ND clusters in cells at higher magnification (scale bar: 500 nm). N, nucleus.

In vivo biodistribution of ND-PEG-DOX

Next, we analyzed the in vivo biodistribution of ND-PEG-DOX in NSG mice bearing orthotopic 6741 xenografts and compared with free DOX. DOX showed distribution throughout the body 6 h after i.v. administration of both DOX and ND-PEG-DOX. However, only ND-PEG-DOX showed tumor accumulation of DOX 24 h after the treatment and the accumulation persisted even after 48 h (Figure 6A). The ex vivo biodistribution analysis of the organs collected after 48 h showed significant accumulation of DOX in ND-PEG-DOX treated animals whereas no significant accumulation was seen in DOX-treated mice (Figure 6B). This observation was corroborated with fluorescent imaging of the sections obtained from the tumors (Figure 6C). ND-PEG-DOX treated tumors showed presence of DOX throughout the section whereas DOX-treated tumors did not. This suggests excellent tumor uptake of ND-PEG-DOX.

Figure 6.

Tumor uptake of ND-DOX. (A) IVIS images showing in vivo biodistribution of free DOX and DOX delivered by using ND-PEG at the indicated time-points after i.v. administration into mice bearing orthotopic pancreatic tumors. An untreated mouse was included for fluorescence background correction. (B) Tumors and various organs were harvested 48 h after i.v. administration of DOX and ND-PEG-DOX and imaged ex vivo by IVIS imaging. ND-PEG-DOX treatment led to significantly higher tumor accumulation in comparison with free DOX. (C) Fluorescent images of sections from tumors. The arrows show the presence of DOX. Cell nuclei were counterstained with DAPI. Scale bar: 100 μm.

In vivo antitumor activity of ND-PEG-DOX

Finally, we sought to assess the in vivo therapeutic efficacy of the ND-PEG-DOX. To this end, approximately 1.0×10⁶ PANC-1 cells were orthotopically implanted in the pancreas of 6–8 week old NSG mice. After three weeks, mice were randomized into four groups before the initiation of treatment. Luciferase in vivo imaging of the animals before the start of treatment demonstrated no significant difference in tumor burden among the groups (data not shown) The experiment was terminated after three weeks of treatment. Notably, there were no discernible abnormalities in major organs such as liver, kidney, and spleen following administration of ND-PEG-DOX (Supplementary Figure S4). The tumor weight and tumor volume, respectively, of the treatment groups after 3 weeks of treatment are shown in Figure 7A and 7B. Although the free DOX treated group showed some inhibition of tumor growth compared to control or ND-PEG treated groups, ND-PEG-DOX was considerably more efficacious in inhibiting tumor growth. Quantification of Ki67-positive cell nuclei in tumor sections for all the treatment groups corroborates these results (Figure 7C). Representative H&E and Ki67 staining of the respective tumor sections are shown in Figure 7D. The results are suggestive of a higher antiproliferative effect of ND-PEG-DOX. These data demonstrate the superior antitumor activity of ND-PEG-DOX in an orthotopic PDAC xenograft model when compared to free DOX. We also assessed tumor lysates by western blot analysis and found an increase in expression of p53 and a concomitant decrease in anti-apoptotic Bcl-2 compared to other treatments groups (Supplementary Figure S5A). We also noted a decrease in the expression of pro-survival Akt, along with a decrease in expression of the ABC transporter, ABCG2 after ND-PEG-DOX treatment (Supplementary Figure S5B).

Figure 7.

In vivo antitumor activity of ND-PEG-DOX. (A) Tumor weight and (B) tumor volume of orthotopic xenografts generated in NSG mice after treatment twice per week for three weeks with vehicle alone, ND-PEG, DOX (10 mg/kg) and ND-PEG-DOX (10 mg/kg DOX equivalent). (C) Quantification of cell proliferation (Ki67 staining) in tumors. * and ** denote p<0.05 and p<0.01, respectively. (D) H&E staining (upper panel) and Ki67 staining (lower panel) of tumor tissues from each treatment group. Scale bar: 200 μm.

Discussion

The present study provides first evidence of tumor selective uptake of drug-ND complexes in an orthotopic patient-derived xenograft model of pancreatic cancer and a superior antitumor activity when compared to free drug. Importantly, the NDs themselves displayed excellent biocompatibility when evaluated using macrophage-like cells and no signs of toxicity were noted in major organs such as liver, kidney, and spleen in mice following the administration of ND-PEG-DOX. These findings are in accordance with a previous study of the long-term stability and biocompatibility of fluorescent NDs (FNDs).²⁸ In the latter study, histopathological analysis of various organs indicated that FNDs were non-toxic even when high doses, up to 75 mg/kg body weight, of the particles were administered i.p. to mice. For comparison, the 10 mg/kg DOX equivalent dose utilized in the present study corresponds to a ND-PEG dose of 66 mg/kg body weight. Other studies in which NDs were administered via i.v. injection or intratracheal instillation also revealed low toxicity.^29,30 Taken together, NDs appear to be well-tolerated in various animal models.²¹ In addition to the obvious importance of biocompatibility, one of the most important hurdles to clinical translation of nanomedicines relates to the precise control of the synthesis with the purpose of producing nanoformulations with high batch-to-batch reproducibility and feasibility of industrial scale-up.³¹ Furthermore, regulatory guidance is required with respect to the characterization of nanomedicines and standardized protocols for testing are needed.^32,33

Orthotopic xenograft mouse models are emerging as the preferred model for cancer research due to the increased clinical relevance over subcutaneous (heterotopic) models.³⁴ We noted a superior antitumor activity in our in vivo model, and a superior cytotoxic effect was also recorded when we compared ND-PEG-DOX to free DOX in a 3D tumor spheroid model of PDAC, but this was not evidenced when PDAC cells (BxPC3 or PANC-1) were maintained as conventional 2D cultures. These observations underscore the importance of using relevant preclinical models when evaluating the antitumor effects of novel drugs. Interestingly, we observed tumor-selective uptake of DOX in NSG mice bearing orthotopic 6741 xenografts following drug delivery using ND-PEG as compared with free DOX, despite the absence of specific targeting moieties on the surface of the NDs. This is likely due, in part, to the enhanced permeability and retention (EPR) effect,³⁵ though a recent study suggested that NDs may also promote vascular leakiness allowing more drug to reach the tumor.³⁶ Further studies are warranted in order to explore whether tumor uptake can be enhanced by adding targeting ligands to the NDs. Studies on the potential role of the bio-corona are also of interest, as the adsorption of biomolecules may influence tumor targeting of nanoparticles.³⁷ Few, if any, studies have focused specifically on the ND bio-corona. Notably, using 3D tumor spheroids, we observed penetration of the NDs into the outer cell layers of the spheroids and TEM imaging analysis disclosed that the NDs were present as small clusters within the cells. TEM imaging also provided morphological evidence of endocytosis of NDs leading us to believe that the NDs were exposed to an acidic environment in the cells. Importantly, we could demonstrate that drug release was pH-dependent with an increased rate of release of DOX from ND-PEG at low pH. Taken together, we surmise that the NDs are internalized into tumor cells and may act as a slow-release depot thereby promoting the antitumor effect of DOX. It is noteworthy that the superior effect of ND-PEG-DOX versus free DOX in 3D spheroids was evident following several days (96 h) of treatment, but was not seen after short-term (24 h) exposure.

Previous studies by Chow et al.¹⁷ have suggested that NDs may overcome drug resistance in murine liver and breast cancer models by circumventing drug efflux by drug transport proteins such as MDR1, ABCG2, and MRP1. Similarly, we noted a greater retention of DOX in PDAC cells following delivery of DOX using ND-PEG when compared to free DOX. Prolonged drug retention in orthotopic tumors was also observed in vivo when compared to free drug. Interestingly, we noted a decrease in ABCG2 in tumor sections following in vivo treatment with ND-PEG-DOX. ABCG2 is important for imparting drug resistance by expunging drugs from cells by means of efflux.³⁸ Consequently, a reduction in ABCG2 level signifies less efflux and, consequently, a greater drug response. DOX was used in the present study as it can be conveniently monitored in vitro and in vivo due to its intrinsic fluorescence, and because it is known to reversibly adsorb to the multifaceted surfaces of NDs.^17,19,39,40 Further studies are needed to address NDs complexed with other clinically relevant drugs used for the treatment of PDAC. Indeed, while translation of these findings to patients remains to be explored, these studies, using a combination of in vitro and in vivo models of PDAC, along with other studies using mouse models of liver and breast cancer,^17,18 suggest that functionalized NDs are a promising nanomedicine platform for cancer treatment.

Abbreviations:

DOX

doxorubicin

FBS

fetal bovine serum

FNDs

fluorescent NDs

NDs

nanodiamonds

PDAC

pancreatic ductal adenocarcinoma

PEG

poly(ethylene glycol)