Biomaterials 35 (2014) 836e845

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Cancer stem cell therapy using doxorubicin conjugated to gold nanoparticles via hydrazone bonds Tian-Meng Sun 1, Yu-Cai Wang 1, Feng Wang, Jin-Zhi Du, Cheng-Qiong Mao, Chun-Yang Sun, Rui-Zhi Tang, Yang Liu, Jing Zhu, Yan-Hua Zhu, Xian-Zhu Yang, Jun Wang* Hefei National Laboratory for Physical Sciences at the Microscale and School of Life Sciences, University of Science and Technology of China, Hefei, Anhui 230027, PR China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 July 2013 Accepted 2 October 2013 Available online 18 October 2013

Nanoparticle-mediated delivery of chemotherapies has demonstrated enhanced anti-cancer efficacy, mainly through the mechanisms of both passive and active targeting. Herein, we report other than these well-elucidated mechanisms, rationally designed nanoparticles can efficiently deliver drugs to cancer stem cells (CSCs), which in turn contributes significantly to the improved anti-cancer efficacy. We demonstrate that doxorubicin-tethered gold nanoparticles via a poly(ethylene glycol) spacer and an acidlabile hydrazone bond mediate potent doxorubicin delivery to breast CSCs, which reduces their mammosphere formation capacity and their cancer initiation activity, eliciting marked enhancement in tumor growth inhibition in murine models. The drug delivery mediated by the nanoparticles also markedly attenuates tumor growth during off-therapy stage by reducing breast CSCs in tumors, while the therapy with doxorubicin alone conversely evokes an enrichment of breast CSCs. Our findings suggest that with well-designed drug delivery system, the conventional chemotherapeutic agents are promising for cancer stem cell therapy. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Breast cancer Cancer stem cells Drug delivery Drug resistance Gold nanoparticles

1. Introduction Nanoparticle-mediated delivery of chemotherapeutic agents has demonstrated enhanced anti-cancer efficacy, while simultaneously reducing side effects in cancer prevention and treatment [1e4]. In the past two decades a number of therapeutics based on nanoparticles are under clinical trials or have been approved for clinical use [5e7]. The key rationale for the success of nanoparticlemediated delivery lies in its ability to overcome the biological and physiological barriers, known to remove therapeutics from the body or prevent them from reaching tumor tissues and cells [8e10]. These barriers include the degradation of therapeutics in blood, rapid clearance by the immune system, glomerular filtration and excretion, and cellular barriers [9,10]. In particular, nanoparticles with the proper structures, sizes, and surface properties would readily accumulate and retain in solid tumors because of the enhanced permeability and retention (EPR) effect, which is also known as a passive targeting mechanism [11,12]. In addition, active

* Corresponding author. E-mail address: [email protected] (J. Wang). 1 Equal contribution first authors. 0142-9612/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biomaterials.2013.10.011

targeting by endowing nanoparticles with ligand can promote enhanced cell binding, further increasing the drug delivery efficacy [13e16]. The discovery of cancer stem cells (CSCs) has changed the view of chemotherapy. CSCs in tumor evade the anti-cancer effects of standard chemotherapy, emerging as an underestimated biological barrier to the success of systemic chemotherapy [17e19]. CSCs are thought to be responsible for the origin, growth, recurrence, and metastasis of cancer [17,18,20e24]. Increasing evidences support the idea that, instead of merely targeting the bulk non-CSCs, successful cancer curing may require both undifferentiated CSCs and differentiated non-CSCs to be efficiently eliminated [18,24e26]. To this end, several promising strategies have been developed for CSCtargeted therapy. These include direct inhibiting CSCs by blocking their essential self-renewal signaling, forcing CSCs to differentiate into bulk tumor cells that are susceptible to standard therapies, as well as screening and identification of drug candidates that can specifically kill CSCs [24e28]. However, CSCs are probably genetically unstable, and their surface marker phenotypes vary from patient to patient and constantly change as the disease progresses, potentially affecting the effect of such therapies [29]. Moreover, there are evidences that non-CSCs in tumors seem to spontaneously and stochastically turn into CSCs de novo [30,31].

T.-M. Sun et al. / Biomaterials 35 (2014) 836e845

Nevertheless, the standard chemotherapy towards CSCs using conventional chemotherapeutic agents is unfortunately ineffective as it leads to the enrichment of CSCs and further results in tumor relapse [29,32e35]. Therefore, we explored to establish a method that can affect both CSCs cells and their differentiated daughter cells. We have previously demonstrated that DOX-Hyd@AuNPs (Fig. 1a), a rationally designed gold nanoparticle-based drug delivery system tethering with doxorubicin on the surface with a poly(ethylene glycol) spacer via an acid-labile linkage, facilitates intracellular drug delivery and overcomes multidrug resistant in MCF-7/ADR cancer cells [36]. In this study, we address this challenge using DOX-Hyd@AuNPs that can overcome the inherent therapeutic resistance of CSCs and indiscriminately deliver chemotherapeutics to both cell populations (i.e., CSCs and non-CSCs cells). We proposed that with efficient deliver of drug to CSCs, DOX-Hyd@AuNPs can elicit marked enhancement in tumor growth inhibition and efficiently attenuated tumor growth by inhibiting the tumor initiating ability of breast CSCs (Fig. 1b).

837

guidelines outlined in the Guide for the Care and Use of Laboratory Animals. The procedures were approved by the University of Science and Technology of China Animal Care and Use Committee. 2.3. Mammosphere culture Cells (1000 cells/ml) were cultured in suspension in serum-free DMEM-F12, supplemented with B27 (1:50, Invitrogen), 20 ng/ml hEGF (BD Biosciences), 0.4% low-endotoxin bovine serum albumin (BSA, Sangon Biotech), and 5 mg/ml insulin (SigmaeAldrich). To propagate mammospheres in vitro, mammospheres were collected by gentle centrifugation, dissociated into single cells as previously described [37], and then cultured to generate mammospheres of the next generation. 2.4. ABCB1 efflux assay The mammosphere cells were seeded into ultra low attachment 6-well plate and treated at 37
C with equivalent DOX, at a final concentration of 1 mg/ml for 2 h. After incubation, the mammosphere cells were collected and washed twice with PBS. One third of the treated cells were kept on ice to provide a control for maximal loading. Another one third of the cells was warmed at 37
C, and was used as the positive control for ABCB1-mediated DOX efflux. The final third of the cells were warmed to 37
C in the presence of vinblastine (Millipore), an inhibitor of ABCB1, at a concentration of 22 mM. All of the cells were incubated for 8 h. The MFI of the mammosphere cells were measured and analyzed by flow cytometry.

2. Materials and methods

2.5. Distribution of DOX-Hyd@AuNPs in CSCs

2.1. Materials

MDA-MB-231, BT-474 and MCF-7 mammosphere cells were treated at 37
C with equivalent DOX, at a final concentration of 1 mg/ml for 2 h. An anti-ABCB1 antibody (biotin-labeled UIC2, 1 mg/ml, Millipore) was added and the cells were cultured for another 15 min. After removal of the media, cells were washed twice with cold PBS, and treated with fixative solution, containing 1% formaldehyde and 0.25% Triton X-100 in TBS, for 5 min. The cells were then incubated with FITC labeled avidin (Santa Cruz) for 30 min following DAPI staining for 5 min at a concentration of 0.1 mg/ml. The cells were directly observed under a Zeiss LSM 710 confocal microscope using a 63 objective.

DOX-Hyd@AuNPs with a diameter of 30 nm and DOX conjugation were synthesized according to the previous report [36]. mPEG@AuNPs with a diameter of 30 nm but without DOX conjugation was synthesized similarly. 2.2. Animals Female NOD/SCID mice were obtained from Beijing HFK Bioscience Co., Ltd. and used at 4e6 weeks of age. All animals received care in compliance with the

Fig. 1. Schematic illustration of (a) endosomal pH-responsive DOX-Hyd@AuNPs (left) and a gold nanoparticle conjugated with poly(ethylene glycol) monomethyl ether (mPEG@AuNPs) (right); (b) relapses of solid tumors after initial remission by free DOX and DOX-Hyd@AuNP, respectively.

838

T.-M. Sun et al. / Biomaterials 35 (2014) 836e845

2.6. Mammosphere formation assay MDA-MB-231, BT-474, and MCF-7 mammosphere cells were treated with equivalent DOX doses of 1 mg/ml for 2 h at 37
C. After incubation, cells were washed twice with cold PBS. One thousand cells were plated per well in an ultra low attachment 24-well plate (Corning) and cultured for 20 days prior to being counted. The mammospheres were also digested into single cells and counted. 2.7. In vitro and in vivo limiting dilution assays For in vivo limiting dilution analysis (LDA), mammosphere cells were cultured with DOX at a final concentration of 1 mg/ml for 2 h at 37
C. Mammosphere cell suspensions were serially (1:2) diluted from 128 to 1 cell per well in the ultra low attachment 96-well plate in a growth medium. Cells from each group were seeded in 8 wells. After 10 days, the colonies larger than 100 mm were counted under the microscope. Limiting dilutions were calculated using the Extreme Limiting Dilution Analysis software [38]. For in vivo LDA, MDA-MB-231 mammosphere cells were treated under the same conditions as described above and then transplanted into the mammary fat pads of female NOD/SCID mice. Each mouse received 10,000 to 100 treated mammosphere cells. Tumors were monitored every 3 days by observation and palpation for up to 60 days. Cancer stem cell frequency was calculated using the L-calc program (Stem-Cell Technologies Inc.). To evaluate the effect of DOX-Hyd@AuNPs on CSCs in vivo, the MDA-MB-231, BT474 or MCF-7 mammosphere cells (2  105) diluted in matrigel (1:1, BD Biosciences) were subcutaneous injected into the mammary fat pad of the NOD/SCID mice. One day after injection, mice were treated with either PBS, mPEG@AuNPs, free DOX, DOX þ mPEG@AuNPs, or DOX-Hyd@AuNPs by i.v. injection every day at equivalent DOX doses of 0.3 mg/kg (32 injections for MDA-MB-231 group, 34 injections for BT474 group, and 38 injections for MCF-7 group). Tumor growth was monitored by measuring the perpendicular diameter by caliper every other day. The estimated volume was calculated based on the following equation: tumor volume ¼ 1/ 2  length  width2. 2.8. Accumulation of DOX in tumors and CSCs DOX-Hyd@AuNP and other controls were injected via i.v. into NOD/SCID mice with an MDA-MB-231 tumor. The tumor volume was approximately 100 mm3. The DOX dose was equivalent to 1 mg/kg body weight. Twenty-four hours later, the mice were sacrificed and tumor tissues were collected. The tissues were weighted and homogenized, and the DOX distributed in the tumor tissues was extracted and analyzed using HPLC analyses. For FACS analyses, the tumor tissues were transferred to a dish and cut into small pieces. The fragments were suspended with 20 ml of serum-free DMEM-F12 medium and collected by centrifugation for 5 min at 600 rpm. The pelleted materials were resuspended in 10 ml of tumor cell digestion solution (1 mg/ml collagenase Type I in PBS, Invitrogen) and incubated at 37
C for 2 h with persistent agitation. The tumor cells were collected by centrifuging at 1200 rpm for 6 min at room temperature and then washed twice with PBS containing 1% FBS. The tumour cells were filtered by a 200-mesh sieve, stained with ALDH substrate, CD24-PE and CD44APC as described above, and analyzed by a FACSCalibur flow cytometer. 2.9. Orthotopic xenograft model and tumor suppression study The xenograft tumor model was generated by the subcutaneous injection of MDA-MB-231, BT-474 or MCF-7 mammosphere cells (2  105) diluted in matrigel (1:1, BD Biosciences) into the mammary fat pad of the mice. For the MCF-7 tumor model, mice were supplemented with estrogen pellets (60-day release pellets containing 0.72 mg of 17 b-estradiol, Innovative Research of America). When the tumor volume was approximately 80 mm3, the mice were randomly divided into 5 groups. Animals were treated with either PBS, mPEG@AuNPs, free DOX, DOX þ mPEG@AuNPs, or DOX-Hyd@AuNPs by i.v. injection every day. DOX dose per injection was equivalent to 0.3 mg/kg body weight, equivalently. The treatment durations were about three weeks (22, 20, and 24 days for MDA-MB-231, BT-474, and MCF-7 groups, respectively). Tumor growth was monitored by measuring the perpendicular diameter by caliper every other day. The estimated volume was calculated based on the following equation: tumor volume ¼ 1/2  length  width2. 2.10. Analysis of the ratio of CSCs in tumor One day after the tumor suppression study, animals were sacrificed and tumor tissues were excised. After digestion into single cells, the cells were stained with ALDH substrate, anti-CD24-PE and anti-CD44-APC as described above, and analyzed by a FACSCalibur flow cytometer. 2.11. Prolonged tumor suppression study Animals bearing an MDA-MB-231 xenograft were treated using the same protocol as described in the tumor suppression study. The animals were received the first injection and last injection on day 10 and 32 post tumor xenografting, respectively. Afterwards, the animals were routinely maintained and the tumor volume was measured for an additional 14 days. The relative slopes were

determined by measuring the slope of tumor growth curves in each group during the off-therapy stage from day 32 to day 46 post tumor xenografting (calculated from the exponential trendlines in Microsoft Excel) compared to PBS treatment. 2.12. Statistical analysis Statistical analyses were performed using ANOVA test to measure statistical differences among groups. Data with P < 0.05 were considered to be statistically significant.

3. Results 3.1. Effectively doxorubicine delivery into brease CSCs DOX-Hyd@AuNPs are gold nanoparticles tethered with doxorubicin via a poly (ethylene glycol) spacer and an acid-labile hydrazone bond (Fig. 1a). DOX-Hyd@AuNPs showed efficient doxorubicin (DOX) transportation to cancer cells as well as responsive intracellular drug release [36]. To investigate the role of nanoparticles in the delivery of chemotherapy to CSCs, we enriched CSCs by non-adherent mammosphere cell cultures of MDA-MB231, BT-474 cells, and MCF-7 cells in serum-free medium [33,39] and characterized by the surface markers of CD44, CD24, and ALDH. Breast cancer cells with CD44þ/CD24 and ALDHhi phenotypes have been demonstrated to have stem cells-like tumorinitiating and invasive features [31,39]. For instance, flow cytometry analysis of MDA-MB-231 mammosphere cells demonstrated a 16.7fold increase in the CD44þ/CD24/ALDHhi population compared with adherent cells (Fig. S1; CD44þ/CD24/ALDHhi cell ratios were 20.0% and 1.2% in mammospheres cells and adhere cells, respectively). The MDA-MB-231 mammosphere cells showed superior resistance to free DOX treatment, exhibiting 18.4-fold increases in the IC50, compared with the corresponding adherent cells (Fig. S2). To assess the DOX delivery property of nanoparticles to CSCs, we incubated the three mammosphere cells with DOX-Hyd@AuNPs. Cells incubated with PBS or mPEG@AuNPs (PEGylated gold nanoparticles with similar physico-chemical property to DOXHyd@AuNPs but lacking doxorubicin, Fig. 1a) were used as controls. As shown in Fig. 2a, after 2 h, the majority of breast CSCs (75.9% for CD44þ/CD24/ALDHhi MDA-MB-231, 77.1% for CD44þ/ CD24/ALDHhi BT-474, and 96.8% for CD44þ/CD24 MCF-7 mammosphere cells, respectively.) were positive in terms of DOX fluorescence following FACS analyses when cultured with DOXHyd@AuNPs, while incubation with free DOX only led to 6.1e 18.3% postive breast CSCs. In addition, the presence of mPEG@AuNPs did not influence the uptake of free DOX by breast CSCs (DOX þ mPEG@AuNPs). Conversely, the differences in uptake between DOX-Hyd@AuNPs and free DOX were not observed when nanoparticles or DOX were incubated with adherent cells. The similar phenomenon was observed when we analyzed the intracellular fluorescence of DOX. DOX-Hyd@AuNPs dramatically enhanced the intracellular DOX in breast CSCs but not in breast adherent cells compared with free DOX or DOX þ mPEG@AuNPs (Fig. S3). These results indicated that DOX-Hyd@AuNPs could more effectively deliver DOX to breast CSCs in comparison to the culturing with free DOX. We next corroborated these results by confocal microscopy. To visualize CSCs, cells were labeled with the ALDEFLUORÔ, an intracellular metabolized fluorescent substrate for aldehyde dehydrogenase. The successful labeling of CSCs with the ALDEFLUORÔ was demonstrated by the disappearance of green fluorescence upon blocking aldehyde dehydrogenase (ALDH) with diethylaminobenzaldehyde (DEAB), an inhibitor of ALDH. After a 2-h incubation with DOX-Hyd@AuNPs, extensive DOX fluorescence was observed in all the MDA-MB-231 mammosphere cells including ALDEFLUOR labeled (green) CSCs, whereas the DOX fluorescence was only observed in ALDEFLUOR-negative non-CSCs

T.-M. Sun et al. / Biomaterials 35 (2014) 836e845

839

Fig. 2. DOX-Hyd@AuNPs significantly enhance doxorubicin (DOX) accumulation in CSCs in vitro. (a) The percentage of DOX positive cells in CD44þ/CD24/ALDHhi MDA-MB-231 and BT-474 cells and CD44þ/CD24 MCF-7 mammosphere cells, or MDA-MB-231, BT-474 and MCF-7 adherent cells treated with different formulations for 2 h. The dose of DOX or its equivalent was 1 mg/ml. Data are shown as means  s.e.m. (n ¼ 3). (b) Confocal laser scanning microscopy images of MDA-MB-231 mammosphere cells incubated with different formulations for 2 h. The dose of DOX or its equivalent was 1 mg/ml. The CSCs were labeled with ALDEFLUORÔ (green). Scale bar ¼ 10 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

when the cells were treated with free DOX or DOX þ mPEG@AuNPs for the same period of time (Fig. 2b). To determine whether such enhanced DOX uptake by nanoparticles is related to their potency in overcoming the intrinsic resistance of CSCs, we performed inhibition studies in which the drug resistance mechanism of CSCs were suppressed. The drug resistance of CSCs was reported to be correlated to the high level expression of the ATP-binding cassette drug transporters (ABC transporter) in CSCs [17,40,41]. We incubated the mammosphere cells with DOX-Hyd@AuNPs or a control at 37
C for 2 h, and the culture media was replaced with fresh medium. The treated cells were then incubated for 10 h at either 37
C or on ice, as culturing cells on ice is known to suppress the ATP-dependent efflux mediated by P-gp [42]. In comparison with the culture at 4
C, the intracellular fluorescence intensity of DOX in the cells incubated at

37
C with the vinblastine, an inhibitor of P-gp, did not show significantly difference with the otherwise identical treatment (Fig. 3a). Conversely, incubating cells treated with free DOX or DOX þ mPEG@AuNPs at 37
C without vinblastine caused a significant decrease in the intracellular fluorescence, due to the efflux effect of DOX by P-gp. However, this phenomenon was not seen when the cells were treated with DOX-Hyd@AuNPs, demonstrating that DOX-Hyd@AuNPs is insensitive to P-gp transporter-based efflux, while the retention of drug in the CSCs was enhanced (Fig. 3a). We further observed that the expression of ATP-binding cassette transporter subfamily B member 1 (ABCB1), one of the most important ABC transporters [43], was significantly up-regulated in mammosphere cells when compared to adherent cells cultured in serum-containing medium. This was shown to be the case for both

840

T.-M. Sun et al. / Biomaterials 35 (2014) 836e845

Fig. 3. DOX-Hyd@AuNPs overcome drug resistance of CSCs. (a) DOX-Hyd@AuNPs overcome the DOX efflux ability of P-gp of CSCs. Data are shown as means  s.e.m. (n ¼ 3). *P < 0.005. (b) Expression of ABCB1 in adherent cells and mammosphere cells determined by quantitative real-time PCR (mRNA, left) and western blot (protein, right). Data are shown as means  s.e.m. (n ¼ 6). (c) Confocal laser scanning microscopy images of the localization of DOX [13] and activated P-gp (green) in MDA-MB-231, BT-474, and MCF-7 CSCs after incubating with PBS, mPEG@AuNPs, free DOX, DOX þ mPEG@AuNPs, or DOX-Hyd@AuNPs for 2 h. The activated P-gp was marked with P-gp antibody (UIC2) (green). Cell nuclei were stained with DAPI (blue). Scale bar ¼ 4 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

mRNA and the corresponding encoded P-gp protein levels (Fig. 3b), which is consistent with the intrinsic drug resistance nature of CSCs. The activated P-gp, which is over-expressed on the membrane of mammosphere cells, was visualized using UIC2 antibody (green) labeling (Fig. 3c). DOX was found to bind to P-gp when the CSC cells were treated with free DOX or DOX þ mPEG@AuNPs for 2 h at 37
C. This was eventually exocytosed from the CSC cells, meaning that the fluorescence of DOX was merely observed in the cytoplasm. In contrast, although partial DOX bound to the activated P-gp on the cell membrane of CSCs, a remarkably increased amount of DOX fluorescence was observed in their cytoplasm (Fig. 3c) when the CSCs were treated with DOX-Hyd@AuNPs under identical conditions, indicating that DOX-Hyd@AuNPs can bypass the efflux of P-gp. 3.2. Inhibition of mammosphere formation and cancer initiation activity We next examined whether the enhanced delivery of DOX into CSCs by DOX-Hyd@AuNPs would reduce the stemness in breast mammosphere cells. We treated the three mammosphere cells with the various formulations for 2 h, and then tested their ability

to regenerate mammosphere as well as the cell number in each regenerated mammosphere on the 20th day after treatments. There was no significant difference in the regeneration of mammosphere or the cell number per regenerated mammosphere between PBS, mPEG@AuNPs, free DOX, and DOX þ mPEG@AuNPs treatments. Conversely, treatment with DOX-Hyd@AuNPs led to a reduction of regenerated mammosphere by 40%e60% (Fig. 4a). In addition, the cell number per regenerated mammosphere was reduced to approximately 35%e60% compared with the control (Fig. 4b). These results suggest that DOX-Hyd@AuNPs treatment of CSCs reduced the quantity of cells capable of mammosphere regeneration which reflects the self-renewal ability of CSCs in vitro [33,37,44,45]. For confirmation, we determined the minimal frequency of CSCs within the treated three mammosphere cell population by in vitro LDA [33]. The number of MDA-MB-231 mammosphere cells capable of regenerating one mammosphere per well after 10 days culture was calculated to be 10 for PBS and 14 for both free DOX and DOX þ mPEG@AuNPs, respectively. However, DOX-Hyd@AuNPs treatment increased the number of mammosphere cells to 28 per well to regenerate one mammosphere (Fig. S4). Similar results were observed from the treated BT-474 and MCF-7 mammosphere cells (Fig. S4), indicating that DOX-Hyd@AuNPs significantly decreases

T.-M. Sun et al. / Biomaterials 35 (2014) 836e845

841

Fig. 4. DOX-Hyd@AuNPs significantly affect the stemness of CSCs. (a) Quantification of the ability to regenerate mammosphere and (b) the relative cell number per regenerated mammosphere of MDA-MB-231, BT-474 or MCF-7 mammosphere cells on day 20 after incubation with PBS, free DOX, DOX þ mPEG@AuNPs, and DOX-Hyd@AuNPs for 2 h. Data are shown as means  s.e.m. (n ¼ 3). *P < 0.005. (c) Incidence of tumors of MDA-MB-231 mammosphere cells serially transplanted into NOD/SCID mice. The cells were pre-treated with PBS, free DOX, DOX þ mPEG@AuNPs, and DOX-Hyd@AuNPs. Tumors were monitored every 3 days by observation and palpation for up to 60 days.

the stemness of CSCs in comparison with free DOX and DOX þ mPEG@AuNPs. We further determined the number of cells required to generate tumors and CSC frequency in vivo. MDA-MB-231 mammosphere cells were treated with various formulations for 2 h. Equal cell numbers from each of the treatment groups were injected into 10 non-obese diabetic/severe combined immunodeficient (NOD/SCID) mice at four different cell numbers, and tumor formation was assessed after by 60 days. Treatment with DOX-Hyd@AuNPs significantly reduced the tumor formation ability of the cells, with the CSC frequency 32efold lower than the PBS control group and approximately 18efold lower than free DOX and DOX þ mPEG@AuNPs (Fig. 4c). To understand whether the effect of DOX-Hyd@AuNPs on decreasing the stemness of CSCs is due to the enhanced DNA damage induced by DOX that is released intracellularly, we incubated MDA-MB-231 mammosphere cells for 72 h with DOXHyd@AuNPs at an equivalent DOX dose of 1.3 mg/ml. This is equivalent to the IC50 of free DOX treatment in MDA-MB-231 mammosphere cells, and this test was used to determined the DNA double-strand breakage (DSB) marked by the serine-139phosphorylated histone H2AX (g-H2AX) [46]. It was observed that much more pronounced DNA damage (higher MFI of g-H2AX) occurred when MDA-MB-231 mammosphere cells were treated with DOX-Hyd@AuNPs in comparison with other controls (Fig. S5), implying that DOX-Hyd@AuNPs may induce DNA DSB in breast CSCs following intracellular DOX release from the nanoparticles.

The effect of DOX delivery to breast CSCs by nanoparticles was further determined in vivo. MDA-MB-231, BT-474, and MCF-7 breast mammosphere cells (2  105) were implanted in the mammary fat pads of 6 NOD/SCID mice per group. The mice were then given continuous therapy (1 injection per day at equivalent DOX doses of 0.3 mg/kg) via tail vein injection, starting from the first day after mammosphere cells implantation. It is demonstrated that sustained administration with DOX-Hyd@AuNPs resulted in a delay in the onset of tumor formation as well as a decrease in tumor size (Fig. 5). After treatment with DOX-Hyd@AuNPs, tumor formation was observed in 3 out of 6 mice received MDA-MB-231 mammosphere cells implantation, while none of tumor was observed when the mice received BT-474 or MCF-7 mammosphere cells implantation. On the contrary, there was no significant delay in the tumor formation between PBS, mPEG@AuNPs, free DOX, and DOX þ mPEG@AuNPs treatments. These results suggested that DOX-Hyd@AuNPs treatment can directly affect breast CSCs in vivo, resulting in the prolonged tumor suppression. 3.3. Tumor suppression and prolonged tumor suppression We next investigated whether DOX-Hyd@AuNPs could enhance the delivery of DOX into the engrafted breast tumors in vivo. We treated MDA-MB-231 tumor-bearing mice with different formulations via intravenous injection at equivalent DOX doses of 1 mg/kg, and analyzed the amount of DOX in tumors at 24 h post injection.

842

T.-M. Sun et al. / Biomaterials 35 (2014) 836e845

Fig. 5. Inhibition of tumor formation of MDA-MB-231, BT-474 or MCF-7 mammosphere cells by DOX-Hyd@AuNPs in comparison with other formulations. Data are shown as means  s.e.m. (n ¼ 6).

Fig. 6a shows that the DOX level in the tumor tissue was 955.1  64.8 ng/g protein for mice received a single injection of DOX-Hyd@AuNPs, whilst the corresponding levels for mice treated with DOX alone and DOX þ mPEG@AuNPs were only 89.7  9.8 ng/ g and 69.6  6.7 ng/g, respectively. This demonstrates that DOXHyd@AuNPs can enhance the accumulation of DOX in breast tumors, likely due to the enhanced permeability and retention effect of nanoparticles. We further determined whether DOX-Hyd@AuNPs delivered more DOX into the CSCs in tumors. The intracellular DOX fluorescence of CSCs defined by CD44þCD24ALDHhi was determined by flow cytometric analysis. Fig. 6b shows that the relative mean fluorescence intensity of DOX in CSCs after treatment with a single injection of DOX-Hyd@AuNPs was 2.01  0.38 when compared with PBS treatment (set as 1.00). However, negligible levels of DOX were detected in the CSCs following systemic administration of DOX (1.04  0.02) or DOX þ mPEG@AuNPs (1.08  0.05). To assess whether DOX-Hyd@AuNPs can inhibit breast tumor growth following systemic administration, we treated MDA-MB231 orthotopic xenografts-bearing mice with DOX-Hyd@AuNPs or various other formulations through i.v. injection every day from the 10th day after xenograft implantation. As indicated in Fig. 7a, DOXHyd@AuNPs at an equivalent DOX dose of 0.3 mg/kg per injection significantly reduced the tumor growth, whereas treatments of free DOX or DOX þ mPEG@AuNPs at the same doses and frequency only moderately inhibited tumor growth. Similar patterns were

observed in the other two breast cancer orthotopic models (Fig. 7b and c). Subsequently, we analyzed the percentage of CSCs marked with CD44þCD24ALDHhi in the MDA-MB-231 and BT-474 tumor models and CSCs marked with CD44þCD24 in the MCF-7 tumor model at the end of therapy. In the MDA-MB-231 orthotopic xenograft model, DOX-Hyd@AuNPs treatment resulted in a decrease of percentage of CSC population in the residual tumor cells, which was 27.9  27.1% of that with PBS treatment (set as 100%) (Fig. 7d). In contrast, free DOX or DOX þ mPEG@AuNPs treatment increased this population by more than seventeen-fold (1804.1  35.5% and 1706.3  49.4%, respectively, P < 0.001). Similar trends were also observed in the BT-474 and MCF-7 tumor models after treatment. Considering that the enrichment of CSCs in the tumor after treatment might lead to rapid tumor growth after treatment [28], the prolonged tumor suppression during the off-therapy stage (defined as the time after the drug treatment was ceased) was examined by measuring the slope of tumor growth curves in each group from day 32 to day 46 (Fig. 8). As indicated, the reduction of CSCs in MDA-MB-231 tumors caused by DOX-Hyd@AuNPs treatments led to significant tumor growth inhibition after drug cessation, showing a relative growth slope of 0.45 to that of PBS treatment from the tumor growth curve. This was much smaller than that following treatments with free DOX or DOX þ mPEG@AuNPs treatments (the relative growth slopes were

T.-M. Sun et al. / Biomaterials 35 (2014) 836e845

843

Fig. 8. Off-therapy study of MDA-MB-231 xenograft tumor groups after different treatments. Data are shown as means  s.e.m. (n ¼ 6).

1.58 and 1.65 to that of PBS treatment, respectively). These results indicate that DOX-Hyd@AuNPs can efficiently attenuate tumor growth after drug cessation, suggesting their therapeutic potential in terms of killing breast CSCs. Fig. 6. Accumulation of DOX in the (a) tumor tissue and (b) CSCs in the tumors of MDA-MB-231 xenograft-bearing mice at 24 h post i.v. injection of the various formulations. Data are shown as means  s.e.m. (n ¼ 3). *P < 0.005.

4. Discussion Although debating on the CSC theory over the past few years, recent investigations have traced the contribution of individual

Fig. 7. (aec) Inhibition of MDA-MB-231 (a), BT-474 (b), and MCF-7 (c) xenograft tumor growth by DOX-Hyd@AuNPs in comparison with other formulations. Data are shown as means  s.e.m. (n ¼ 6). *P < 0.0001. (d) Analyses of the CSC ratios within MDA-MB-231, BT-474, and MCF-7 tumors after the tumor suppression study. Data are shown as means  s.e.m. (n ¼ 3). *P < 0.005.

844

T.-M. Sun et al. / Biomaterials 35 (2014) 836e845

tumor cells to tumor formation, providing evidences that CSCs do exist and arise de novo during tumor formation in intact organs [47e49]. More recently, it has been found that tumorigenic glioma stem cells (GSCs) can generate vascular pericytes, demonstrating that GSCs have the capacity to remodulate their microenvironment [50]. CSCs are believed to have crucial roles in cancer treatment failure and be responsible for quick relapses of solid tumors after initial remission [20,24,51,52]. It is assumed that the CSCs within a tumor must be eradicated to achieve cures, which inspires many new approaches for anti-cancer drug discovery with regards to CSC therapy. To date, CSC-specific therapeutic strategies, including forcing their differentiation and targeting their self-renewal pathways and microenvironments, have been underexplored [24,53e 56]. However, CSCs are probably genetically unstable, and a committed progenitor could restore renewal activity [24,32,33], and evidences show that non-CSCs in tumors seem to spontaneously and stochastically turn into CSCs de novo [30,31]. As a result, the CSC-specific treatment approaches may not result in complete regression of tumors [24,27,51]. On the other hand, recent studies have shown that the treatment of mice bearing human breast cancer xenografts with conventional chemotherapeutic agents, such as doxorubicin and paclitaxel, can lead to the enrichment of CSCs and further result in tumor relapse [28,33]. It was also found that chemotherapy treatment increased the percentage of CSC-like CD44þCD24e/low tumor cells in patients with breast cancer [33]. It therefore appears that the presence of CSCs in tumor is an unavoidable hindrance of conventional chemotherapeutic agents, although it has been recognized that it could be therapeutically advantageous to combine agents that target CSCs with conventional agents that reduce the bulk of the tumor [27,51]. Clinical results have suggested that nanoparticle-mediated drug delivery can show enhanced efficacy in cancer therapy, while simultaneously reducing side-effects, as a result of properties including targeted localization in tumors and active cellular uptake, but cancer therapy towards CSCs by nanoparticle-base drug delivery systems is unfortunately poorly understood. In this study, it has been demonstrated that unlike free drugs, drug-conjugated nanoparticles based on inert gold nanoparticles, can eradicate CSCs in certain breast tumor xenografts while killing bulk cancer cells. This mechanism was shown to occur by overcoming the intrinsic resistance of breast CSCs mediated by ABCB1 and also by potentially inducing DNA damage. The results indicate that, with rational design, nanocarriers can contribute to approaches towards breast CSCs with conventional chemotherapeutic agents and to another mechanism that may explain its improved anti-cancer efficacy, therefore, therapies towards CSCs are not limited by the discovery of agents exclusively targeting CSCs, and this can further emerge as a potential platform for CSCs targeted therapy in solid tumors. 5. Conclusion We have reported that DOX-Hyd@AuNPs nanoparticles can deliver more DOX into the breast CSCs by overcoming the intrinsic resistance through the evasion of the efflux of P-gp. Consequently breast tumor growth both during and after the drug treatment period is inhibited without causing the enrichment of CSCs in the tumors. The treatment with these rationally designed nanoparticles results in the removal of all tumor cell sub-populations and avoid the possible repopulation of the tumor mass by CSCs. The results suggest that rationally designed nanoparticles as drug carrier can overcome the CSC hindrance for improved chemotherapy and nanoparticle-based delivery system of chemotherapeutic agents can be an anti-cancer strategy towards CSC therapy.

Acknowledgments This work was supported by the National Basic Research Program of China (973 Programs, 2010CB934001 and 2012CB932500), the National Natural Science Foundation of China (51125012), and the Open Project of State Key Laboratory of Supramolecular Structure and Materials (SKLSSM201301). Appendix A. Supplementary data Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.biomaterials.2013.10.011. References [1] Cabral H, Matsumoto Y, Mizuno K, Chen Q, Murakami M, Kimura M, et al. Accumulation of sub-100 nm polymeric micelles in poorly permeable tumours depends on size. Nat Nanotechnol 2011;6:815e23. [2] Davis ME, Chen ZG, Shin DM. Nanoparticle therapeutics: an emerging treatment modality for cancer. Nat Rev Drug Discov 2008;7:771e82. [3] Peer D, Karp JM, Hong S, Farokhzad OC, Margalit R, Langer R. Nanocarriers as an emerging platform for cancer therapy. Nat Nanotechnol 2007;2:751e60. [4] Petros RA, DeSimone JM. Strategies in the design of nanoparticles for therapeutic applications. Nat Rev Drug Discov 2010;9:615e27. [5] Heidel JD, Davis ME. Clinical developments in nanotechnology for cancer therapy. Pharm Res 2011;28:187e99. [6] Lammers T, Kiessling F, Hennink WE, Storm G. Drug targeting to tumors: principles, pitfalls and (pre-) clinical progress. J Control Release 2012;161: 175e87. [7] Wang AZ, Langer R, Farokhzad OC. Nanoparticle delivery of cancer drugs. Annu Rev Med 2012;63:185e98. [8] Chrastina A, Massey KA, Schnitzer JE. Overcoming in vivo barriers to targeted nanodelivery. Wiley Interdiscip Rev Nanomed Nanobiotechnol 2011;3:421e37. [9] Ferrari M. Frontiers in cancer nanomedicine: directing mass transport through biological barriers. Trends Biotechnol 2010;28:181e8. [10] Florence AT. “Targeting” nanoparticles: the constraints of physical laws and physical barriers. J Control Release 2012;164:115e24. [11] Fang J, Nakamura H, Maeda H. The EPR effect: unique features of tumor blood vessels for drug delivery, factors involved, and limitations and augmentation of the effect. Adv Drug Deliv Rev 2011;63:136e51. [12] Kim J, Kim PH, Kim SW, Yun CO. Enhancing the therapeutic efficacy of adenovirus in combination with biomaterials. Biomaterials 2012;33:1838e50. [13] Davis ME, Zuckerman JE, Choi CHJ, Seligson D, Tolcher A, Alabi CA, et al. Evidence of RNAi in humans from systemically administered siRNA via targeted nanoparticles. Nature 2010;464:1067e70. [14] Hrkach J, Von Hoff D, Ali MM, Andrianova E, Auer J, Campbell T, et al. Preclinical development and clinical translation of a PSMA-targeted docetaxel nanoparticle with a differentiated pharmacological profile. Sci Transl Med 2012;4. 128ra39. [15] Liu Q, Li RT, Qian HQ, Wei J, Xie L, Shen J, et al. Targeted delivery of miR-200c/ DOC to inhibit cancer stem cells and cancer cells by the gelatinases-stimuli nanoparticles. Biomaterials 2013;34:7191e203. [16] Cai LL, Qiu N, Li X, Luo KL, Chen X, Yang L, et al. A novel truncated basic fibroblast growth factor fragment-conjugated poly (ethylene glycol)cholesterol amphiphilic polymeric drug delivery system for targeting to the FGFR-overexpressing tumor cells. Biomaterials 2011;408:173e82. [17] Clevers H. The cancer stem cell: premises, promises and challenges. Nat Med 2011;17:313e9. [18] Nguyen LV, Vanner R, Dirks P, Eaves CJ. Cancer stem cells: an evolving concept. Nat Rev Cancer 2012;12:133e43. [19] Visvader JE, Lindeman GJ. Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nat Rev Cancer 2008;8:755e68. [20] Dalerba P, Cho RW, Clarke MF. Cancer stem cells: models and concepts. Annu Rev Med 2007;58:267e84. [21] Hermann PC, Huber SL, Herrler T, Aicher A, Ellwart JW, Guba M, et al. Distinct populations of cancer stem cells determine tumor growth and metastatic activity in human pancreatic cancer. Cell Stem Cell 2007;1:313e23. [22] Zhang L, Yao HJ, Yu Y, Zhang Y, Li RJ, Ju RJ, et al. Mitochondrial targeting liposomes incorporating daunorubicin and quinacrine for treatment of relapsed breast cancer arising from cancer stem cells. Biomaterials 2012;33:565e82. [23] Ma X, Zhou J, Zhang CX, Li XY, Li N, Ju RJ, et al. Modulation of drug-resistant membrane and apoptosis proteins of breast cancer stem cells by targeting berberine liposomes. Biomaterials 2013;34:4452e65. [24] Zhou BB, Zhang H, Damelin M, Geles KG, Grindley JC, Dirks PB. Tumourinitiating cells: challenges and opportunities for anticancer drug discovery. Nat Rev Drug Discov 2009;8:806e23. [25] Sachlos E, Risueno RM, Laronde S, Shapovalova Z, Lee JH, Russell J, et al. Identification of drugs including a dopamine receptor antagonist that selectively target cancer stem cells. Cell 2012;149:1284e97.

T.-M. Sun et al. / Biomaterials 35 (2014) 836e845 [26] Takeishi S, Matsumoto A, Onoyama I, Naka K, Hirao A, Nakayama KI. Ablation of fbxw7 eliminates leukemia-initiating cells by preventing quiescence. Cancer Cell 2013;23:347e61. [27] Gupta PB, Onder TT, Jiang G, Tao K, Kuperwasser C, Weinberg RA, et al. Identification of selective inhibitors of cancer stem cells by high-throughput screening. Cell 2009;138:645e59. [28] Lang JY, Hsu JL, Meric-Bernstam F, Chang CJ, Wang Q, Bao Y, et al. BikDD eliminates breast cancer initiating cells and synergizes with lapatinib for breast cancer treatment. Cancer Cell 2011;20:341e56. [29] Dylla SJ, Beviglia L, Park IK, Chartier C, Raval J, Ngan L, et al. Colorectal cancer stem cells are enriched in xenogeneic tumors following chemotherapy. PLoS One 2008;3:e2428. [30] Chaffer CL, Brueckmann I, Scheel C, Kaestli AJ, Wiggins PA, Rodrigues LO, et al. Normal and neoplastic nonstem cells can spontaneously convert to a stemlike state. Proc Natl Acad Sci U S A 2011;108:7950e5. [31] Gupta PB, Fillmore CM, Jiang G, Shapira SD, Tao K, Kuperwasser C, et al. Stochastic state transitions give rise to phenotypic equilibrium in populations of cancer cells. Cell 2011;146:633e44. [32] Du Z, Qin R, Wei C, Wang M, Shi C, Tian R, et al. Pancreatic cancer cells resistant to chemoradiotherapy rich in “stem-cell-like” tumor cells. Dig Dis Sci 2011;56:741e50. [33] Yu F, Yao H, Zhu P, Zhang X, Pan Q, Gong C, et al. let-7 regulates self renewal and tumorigenicity of breast cancer cells. Cell 2007;131:1109e23. [34] Li X, Lewis MT, Huang J, Gutierrez C, Osborne CK, Wu MF, et al. Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. J Natl Cancer Inst 2008;100:672e9. [35] Massard C, Deutsch E, Soria JC. Tumour stem cell-targeted treatment: elimination or differentiation. Ann Oncol 2006;17:1620e4. [36] Wang F, Wang YC, Dou S, Xiong MH, Sun TM, Wang J. Doxorubicin-tethered responsive gold nanoparticles facilitate intracellular drug delivery for overcoming multidrug resistance in cancer cells. ACS Nano 2011;5:3679e92. [37] Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ, et al. In vitro propagation and transcriptional profiling of human mammary stem/ progenitor cells. Genes Dev 2003;17:1253e70. [38] Hu Y, Smyth GK. ELDA: extreme limiting dilution analysis for comparing depleted and enriched populations in stem cell and other assays. J Immunol Methods 2009;347:70e8. [39] Ginestier C, Hur MH, Charafe-Jauffret E, Monville F, Dutcher J, Brown M, et al. ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 2007;1:555e67. [40] Dean M, Fojo T, Bates S. Tumour stem cells and drug resistance. Nat Rev Cancer 2005;5:275e84.

845

[41] Soltanian S, Matin MM. Cancer stem cells and cancer therapy. Tumor Biol 2011;32:425e40. [42] Ruetz S, Raymond M, Gros P. Functional expression of P-glycoprotein encoded by the mouse mdr3 gene in yeast cells. Proc Natl Acad Sci U S A 1993;90: 11588e92. [43] Fletcher JI, Haber M, Henderson MJ, Norris MD. ABC transporters in cancer: more than just drug efflux pumps. Nat Rev Cancer 2010;10:147e56. [44] Ponti D, Costa A, Zaffaroni N, Pratesi G, Petrangolini G, Coradini D, et al. Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer Res 2005;65:5506e11. [45] Zucchi I, Sanzone S, Astigiano S, Pelucchi P, Scotti M, Valsecchi V, et al. The properties of a mammary gland cancer stem cell. Proc Natl Acad Sci U S A 2007;104:10476e81. [46] Nussenzweig A, Paull T. DNA repair: tails of histones lost. Nature 2006;439: 406e7. [47] Chen J, Li Y, Yu TS, McKay RM, Burns DK, Kernie SG, et al. A restricted cell population propagates glioblastoma growth after chemotherapy. Nature 2012;488:522e6. [48] Driessens G, Beck B, Caauwe A, Simons BD, Blanpain C. Defining the mode of tumour growth by clonal analysis. Nature 2012;488:527e30. [49] Schepers AG, Snippert HJ, Stange DE, van den Born M, van Es JH, van de Wetering M, et al. Lineage tracing reveals Lgr5þ stem cell activity in mouse intestinal adenomas. Science 2012;337:730e5. [50] Cheng L, Huang Z, Zhou WC, Wu QL, Donnola S, Liu JK, et al. Glioblastoma stem cells generate vascular pericytes to support vessel function and tumor growth. Cell 2013;153:139e52. [51] Liu S, Wicha MS. Targeting breast cancer stem cells. J Clin Oncol 2010;28: 4006e12. [52] Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem cells, cancer, and cancer stem cells. Nature 2001;414:105e11. [53] Jin L, Hope KJ, Zhai Q, Smadja-Joffe F, Dick JE. Targeting of CD44 eradicates human acute myeloid leukemic stem cells. Nat Med 2006;12:1167e74. [54] Lonardo E, Hermann PC, Mueller MT, Huber S, Balic A, Miranda-Lorenzo I, et al. Nodal/Activin signaling drives self-renewal and tumorigenicity of pancreatic cancer stem cells and provides a target for combined drug therapy. Cell Stem Cell 2011;9:433e46. [55] McAuliffe SM, Morgan SL, Wyant GA, Tran LT, Muto KW, Chen YS, et al. Targeting notch, a key pathway for ovarian cancer stem cells, sensitizes tumors to platinum therapy. Proc Natl Acad Sci U S A 2012;109:E2939e48. [56] Piccirillo SG, Reynolds BA, Zanetti N, Lamorte G, Binda E, Broggi G, et al. Bone morphogenetic proteins inhibit the tumorigenic potential of human brain tumour-initiating cells. Nature 2006;444:761e5.