Beta-Lapachone

Self-Amplified Apoptosis Targeting Nanoplatform for Synergistic Magnetic–Thermal/Chemo Therapy In Vivo

Wei Liu, Li Chen, Ming Chen, Wu Wang,* Xiaoling Li, Hong Yang, Shiping Yang, and Zhiguo Zhou*

Abstract

The low efficiency homing of nanomaterials in tumors remains a major challenge in nanomedicine. Inspired by the apoptosis targeting properties of phosphatidylserine (PS), a self-amplified apoptosis targeting nanoplatform (MNPs-ZnDPA/𝜷-Lap) is fabricated combining [email protected] Fe2O4 nanoparticles (MNPs) with an excellent magnetic hyperthermia effect, a chemotherapeutic drug of 𝜷-lapachone (𝜷-Lap) with the promotion of cell apoptosis, and the good apoptosis targeting moiety of Zn(II)-bis(dipicolylamine) (bis-ZnDPA) for PS. In an apoptotic 4T1 xenograft model, MNPs-ZnDPA/𝜷-Lap can first accumulate in tumors by the EPR effect. The released 𝜷-Lap triggers the apoptosis of cancer cells in the tumor and increases the apoptotic target, which results in amplifying their apoptosis targeting properties. This self-amplified apoptosis targeting efficiency of MNPs-ZnDPA/𝜷-Lap almost inhibits the growth of tumors with the synergistic magnetic–thermal/chemo therapy, which can offer a significant promise for targeting cancer theranostics.

1. Introduction

therapy strategy, which refers that more targets were dynamically produced in the process of targeting and therapy, has been proposed.[3b,4] For example, Lee and co- workers developed the CQRPPR peptide conjugated liposomes containing dox- orubicin for the apoptosis-targeted drug therapy with in situ dose amplification.[4a] Zhang et al. developed fibrin-targeting peptide modified carbon nanotubes for amplifying near infrared -driven photother- mal therapy of tumors.[4b] Recently, our group also fabricated Zn(II)-dipicolylamine (ZnDPA) conjugated Fe@Fe3O4 as an apoptosis-homing nanoplatform for an amplified apoptosis-homing pho- tothermal therapy.[4c] However, the weak apoptotic targeting efficiency of ZnDPA limited their amplified apoptosis- homing ability. The penetration depth and damage to normal tissues of pho- tothermal therapy hampered its clinical application.[5] To address the two problems, considering
In the past few decades, nanomaterial delivery systems for cancer therapy have attracted increasing attention.[1] How- ever, the unsatisfactory therapeutic efficiency was still an important factor in limiting their development. Alternatively, re- ceptors mediated active tumor-targeting therapy is a complemen- tary strategy to augment therapeutic efficiency.[2] For receptors mediated active tumor-targeting therapy, the insufficient recep- tor density of cancer cells limits the active uptake of nanopar- ticles by tumor cells and the efficiency of cancer therapy.[3] To address this problem, a promising self-amplified tumor-homing the advantages of magnetic hyperthermia with the deep tissue penetration and lower damage to normal tissue,[6] herein, we developed Zn(II)-bis(dipicolylamine) (bis-ZnDPA) conjugated and a chemotherapeutic drug (𝛽-Lap) loaded [email protected] (MNPs) nanoplatform (MNPs-ZnDPA/𝛽-Lap)with better-amplified apoptosis targeting ability,[7] the high transverse relaxation, and specific loss powers value.[8] MNPs-ZnDPA/𝛽-Lap could become an ideal choice for MRI-guided amplified apoptosis-homing magnetic hyperther- mia therapy in vivo (Scheme 1). The nanoplatform addressed the fundamental challenge of targeting cancer theranostics to some extent and opened up the potential clinical application.

2. Results and Discussion

2.1. Synthesis and MRI Property of MNPs-ZnDPA

[email protected] nanoparticles (MNPs) were prepared via thermal decomposition of metal acetylacetonate precursor.[9] X-ray diffraction pattern implied that MNPs were successfully synthesized (Figure S1, Supporting Information). A TEM image showed MNPs were uniform with an average diameter of around 13 nm (Figure S2, Supporting Information). We performed the water solubility and function of MNPs, PEG-functionalized MNPs (MNPs-PEG) and ZnDPA- functionalized MNPs (MNPs-ZnDPA) by modifying with PEG-phospholipid and the mixture of PEG-phospholipid and bis-ZnDPA functionalized phospholipid PEG (w/w, 5:5), re- spectively. As shown by TEM (Figure 1a; Figure S3, Supporting Information) and dynamic light scattering data (Figure S4a, Supporting Information), the average diameter and hydrody- namic diameter (HD) of MNPs-ZnDPA were 13.1 ± 1.1 and ∼53.7 nm, respectively, while these of MNPs-PEG were 13.2 ± 1.2 nm (Figure S5, Supporting Information) and ∼57.8 nm (Figure S4b, Supporting Information), respectively. The lattice spacing was ∼0.525 nm (Figure 1a). Whereas MNPs-PEG ex- hibited negative 𝜁-potential of −25.8 mV, MNPs-ZnDPA had a positive surface charge (∼16.7 mV) (Figure S4d, Supporting Information), which confirmed the successful modification of positive Zn(II) complexes on the surface of MNPs (Figure S4c, Supporting Information). Furthermore, they exhibited the outstanding colloidal stability within at least one week in RPMI-1640 confirmed by no obvious change of HD, which was a benefit for the following biomedical application (Figure 1b; Figure S6, Supporting Information).
Considering the better magnetic properties of MNPs, the MRI imaging was quantitatively evaluated with MNPs-PEG and MNPs-ZnDPA, respectively. The transversal relaxivity (r2) is 213.04 mM−1 s−1 (Figure 1c) and the corresponding ratio of r2/r1 is 7.30 for MNPs-ZnDPA, while MNPs-PEG exhibited a simi- lar r2 (209.02 mM−1 s−1) and r2/r1 (6.52) (Figure S7, Supporting Information), which suggested they are potential as T2-positive contrast agents. To explore the possibility of tumor therapy, we further investigated their magnetothermal efficiency. As shown in Figure 1d, the temperature elevated along with the increase of concentration of MNPs-ZnDPA. The temperature increased from 25 to 47.3 °C with 1000 µg mL−1 under AMF (598 kHz, 300 G) for 15 min, which suggested that the desired tempera- ture (40–42 °C) in magnetic hyperthermia therapy for cell dam- age could be attained under AMF.[10] In order to further study in vitro magnetic hyperthermia, we measured the magnetic heat- ing rate in the cell suspension by incubating with different concentrations of MNPs-ZnDPA (Figure S8, Supporting Infor- mation). When the incubation concentration of MNPs-ZnDPA was 1000 µg mL−1, the cell temperature increased from 36 to 42 °C, which was enough to damage tumor cells. No obvious temperature decrease was detected for six cycles, indicating the good magnetothermal stability (Figure S9, Supporting Informa- tion). Therefore, those unique properties MNPs offered a chance for MRI and magnetic hyperthermia therapy for cancer cells and tumors in vivo.

2.2. Apoptosis Targeting Property In Vitro

To assess the apoptosis targeting ability of MNPs-ZnDPA, we first constructed an apoptotic cell model by injecting with 𝛽- Lap loaded MNPs-PEG (MNPs-PEG/𝛽-Lap). On the one hand, 𝛽-Lap could selectively kill cancer cells instead of normal cells within a certain concentration range. On the other hand, MNPs-PEG/𝛽-Lap could be accumulated in tumor through EPR effect and 𝛽-Lap was slowly released in tumor microenvironment. First, MNPs-PEG/𝛽-Lap was synthesized.[11] The hydrodynamic diameter and 𝜁-potentials were determined to be ∼59.0 nm and −27.2 mV for MNPs-PEG/𝛽-Lap (Figures S10b,d, Supporting In- formation). The loading capacity of 𝛽-Lap was estimated to be ∼77.4 µg mg−1 (Figures S11 and S12, Supporting Information). The drug release ratio of MNPs-PEG/𝛽-Lap was calculated to be 55.1 ± 3.3% at pH 5.2 and only 15.6 ± 3.5% in neutral solution (Figure S13b, Supporting Information) within 48 h. Second, cel- lular viability was determined. The cellular viability was 97.0 ± 0.5% for HUVEC, while it was 67.0 ± 1.3% for 4T1 cells after the incubation of MNPs-PEG/𝛽-Lap for 0.5 h (Figure S14, Sup- porting Information). With an increase of the incubation time, the viability of HUVEC dropped significantly, which was per- haps due to the increased 𝛽-Lap. Therefore, the incubation time of 30 min was selected for the construction of the apoptotic cell model. On the apoptotic cell model, MNPs-ZnDPA and MNPs- PEG were incubated for different time. T2-weighted MR images and the ΔT2/T2 value were determined for different groups. The T2-weighted MR images of MNPs-ZnDPA were darker than that of MNPs-PEG at 2 h. At the same time, the ΔT2/T2 value of MNPs-ZnDPA (−67.4 ± 3.5%) was much lower than MNPs-PEG (−31.3 ± 3.9%) (Figure S15, Supporting Information). As shown in Figure 2a, in apoptotic 4T1 cells, the ΔT2/T2 value and the uptaken Fe of MNPs-ZnDPA were −68.3 ± 2.2% and 627.7 ± 39.1 pg/cell, respectively. In 4T1 cells, the higher ΔT2/T2 value (−50.6 ± 5.7%) and lower the uptaken Fe (243.6 ± 47.8 pg/cell) of MNPs-ZnDPA were obtained. In apoptotic 4T1 cells, MNPs- ZnDPA also showed lower ΔT2/T2 value and higher Fe concen- tration than MNPs-PEG (27.5 ± 4.7% and 251.0 ± 28.0 pg/cell). Meanwhile, the ΔT2/T2 value of in the inhibition group increased to −38.7 ± 3.6% and the uptaken Fe decreased to 368.7 ± 16.3 pg/cell. The decreased MR signal and the increased iron uptake in apoptotic cells are ascribed to the typical interaction between PS of the cell membranes and ZnDPA on the surface of MNPs and.[12]

2.3. Self-Amplified Apoptosis Targeting Property In Vitro

In order to further perform a self-amplified apoptosis target- ing process, MNPs-ZnDPA/𝛽-Lap was synthesized.[13] The HD and 𝜁-potentials of MNPs-ZnDPA/𝛽-Lap were determined to be ∼50.8 nm and +12.6 mV (Figure S10a,b, Supporting Informa- tion). The 𝛽-Lap loading of MNPs-ZnDPA/𝛽-Lap was calculated to be ∼73.3 µg mg−1 (Figure S16, Supporting Information). At pH = 5.2, when the temperatures increased from 25 to 50 °C, the released 𝛽-lap in the MNPs-PEG/𝛽-Lap NPs raised from 28.5 ± 1.4% to 87.3 ± 2.5%. Under AMF (15 min) for eight times, 77.1 ± 1.6% and 62.3 ± 2.5% of 𝛽-Lap was released from MNPs- ZnDPA/𝛽-Lap at pH 5.2 and 6.0 for 48 h, which was nearly 2 fold of that at pH 7.4 (33.0 ± 1.1%) under AMF. For comparison, with- out AMF, 48.1 ± 2.3% and 39.8 ± 1.5% of 𝛽-Lap was released at pH 5.2 and pH = 6.0, respectively. Only 10.6 ± 1.4% was released in a neutral buffer (Figure S13a, Supporting Information). Obvi- ously, AMF and weak acidic tumor environment both accelerated the release of 𝛽-Lap, which might be resulted from that AMF trigged the generation of heat for further facilitation to weaken the hydrophobic interactions between 𝛽-Lap and lipids, and acid- triggered drug release was possibly due to partially increase the water solubility of 𝛽-Lap.[14]
The changes of the ΔT2/T2 value, confocal fluorescence imag- ing and quantitative analysis of the uptaken Fe proved an am- plified apoptosis targeting character of MNPs-ZnDPA/𝛽-Lap (Figure 3). In Figure 3a, in apoptotic 4T1 cells, the ΔT2/T2 value and uptaken Fe of MNPs-PEG were −25.7 ± 4.7% and 236.8 ± 22.3 pg/cell, respectively. Similarly, those of MNPs-PEG/𝛽-Lap in apoptotic cells were −27.4 ± 6.5% and 255.2 ± 24.3 pg/cell, respectively. The similar ΔT2/T2 value and the uptaken Fe indi- cated that no obvious endocytosis behavior of MNPs by 𝛽-Lap was observed. By strong contrast, the ΔT2/T2 value of MNPs- ZnDPA and MNPs-ZnDPA/𝛽-Lap in apoptotic cells were −67.8 ± 8.2% and −98.6 ± 2.5%, respectively. The uptaken Fe of MNPs- ZnDPA and MNPs-ZnDPA/𝛽-Lap in apoptotic cells were 641.7 ± 60.8 pg/cell and 975.0 ± 82.3 pg/cell, respectively. The obvi- ous decrease of ΔT2/T2 value and the increased Fe accumula- tion in apoptotic cells were caused by that the released 𝛽-Lap from MNPs-ZnDPA/𝛽-Lap promoted the apoptosis of 4T1 cells and resulted in producing more PS in the cellular surface with the strong affinity with ZnDPA. We defined this increased tar- geting property of MNPs-ZnDPA/𝛽-Lap as their self-amplified apoptosis-targeting property.
In order to intuitively observe the self-amplified apoptosis targeting property of MNPs-ZnDPA/𝛽-Lap, MNPs were la- beled by FITC for laser confocal fluorescence imaging. The diameters of MNPs-ZnDPA/FITC and MNPs-PEG/FITC were ≈56.8 nm and ≈59.6 nm, and the zeta potentials were +22.0 mV and −21.7 mV, respectively (Figure S17, Supporting Informa- tion). MNPs-ZnDPA/FITC and MNPs-PEG/FITC had similar excitation and emission spectra (Figure S18, Supporting Infor- mation). Confocal laser scanning microscopy (CLSM) was used to investigate the self-amplified apoptosis targeting ability of MNPs-ZnDPA/FITC/𝛽-Lap. As shown in Figure 3c,d, compared with apoptosis 4T1 group, in apoptosis 4T1 cells, the increased fluorescence intensity of MNPs-PEG/FITC and MNPs- PEG/FITC/𝛽-Lap were 15.0 ± 2.2% and 15.5 ± 3.7%, respectively, suggesting no obvious increased endocytosis behavior of MNPs by 𝛽-Lap. However, the strong fluorescence intensity of MNPs- ZnDPA/FITC (54.4± 5.8%) was three times higher than that of MNPs-PEG/FITC (15.0 ± 2.2%). The increased fluorescence intensity further confirmed the apoptosis targeting property. In MNPs-ZnDPA/FITC/𝛽-Lap group, the fluorescence inten- sity achieved 73.3 ± 1.5%, which was much higher than that treated with MNPs-ZnDPA/FITC and demonstrated the effective self-amplified apoptosis character.

2.4. Self-Amplified Apoptosis Targeting MR Imaging In Vivo

An apoptotic tumor model was established by the intravenous injection of MNPs-PEG/𝛽-Lap (60 mg kg−1) twice within 4 days and exposed under AMF for 30 min after each injection, which was confirmed by TUNEL staining. The apoptosis rates for tu- mor group and apoptotic tumor group were calculated to be 8.9 ± 3.9% and 28.3 ± 5.0%, respectively (Figure S19, Supporting Information). First, the apoptosis targeting efficiency of MNPs- ZnDPA in tumor was affirmed by MRI. The T2 images were acquired after the tail vein injection of MNPs-ZnDPA and MNPs- PEG with the same dose of 60 mg kg−1, respectively. The signal in tumor after the tail vein injection of MNPs-ZnDPA got darker than that of preinjection, and reached the darkest at 180 min (Figure 4b). The ratio of the signal intensity of tumor to that of water (abbreviated as Ttumor/Twater) decreased from 0.66 ± 0.06 to 0.24 ± 0.07 from 0 to 180 min after the injection of MNPs- ZnDPA (Figure 4e). However, for the MNPs-PEG group, the MRI signal of tumor showed no significant difference after the injection at 180 min. The Ttumor/Twater of MNPs-PEG only de- creased from 0.63 ± 0.04 to 0.45 ± 0.06 (Figure 4a,e). Ana- lyzed by ICP-MS, while the maximum accumulated Mn value in tumor was 4.1 ± 1.2 ID g−1 at 180 min for the MNPs- PEG group, it was 4.8 ± 0.8 ID g−1 for the MNPs-ZnDPA group at the same injection point (Figure S20, Supporting Infor- mation). The apparent difference indicated that MNPs-ZnDPA could achieve a rapid ZnDPA-mediated apoptosis targeting effect.
Second, self-amplified apoptosis targeting effectiveness of MNPs-ZnDPA/𝛽-Lap was further examined by in vivo MRI. The T2-weighted MR images in an apoptotic tumor model were recorded after the injection of MNPs-PEG/𝛽-Lap and MNPs- ZnDPA/𝛽-Lap, respectively. For MNPs-ZnDPA/𝛽-Lap group, the Ttumor/Twater decreased from 0.68 ± 0.07 to 0.13 ± 0.07 from 0 to 180 min, while for MNPs-PEG/𝛽-Lap group, the signal de- creased slowly from 0.62 ± 0.1 to 0.49 ± 0.1 (Figure 4c). The Ttumor/Twater showed a greater decrease for the MNPs-ZnDPA/𝛽- Lap group at 180 min compared to that of the MNPs-ZnDPA group, indicating that MNPs-ZnDPA/𝛽-Lap could achieve a self- amplification apoptosis targeting property. The maximum accu- mulated Mn value in tumor (5.4 ± 1.0 ID g−1) for the MNPs- ZnDPA/𝛽-Lap group was more than that for the MNPs-ZnDPA group (4.0 ± 0.6 ID g−1) (Figure S20, Supporting Informa- tion). The self-amplification apoptosis targeting property was attributed to the following process. The released 𝛽-Lap gen- erated more PS. The generated PS could be combined with more ZnDPA-conjugated MNPs and amplified the MRI signal in tumor. As expected, no obvious difference of the Ttumor/Twater between MNPs-PEG/𝛽-Lap group and MNPs-PEG group was observed.

2.5. Self-Amplified Apoptosis Targeting Therapy In Vivo

Encouraged by the high efficiency of magnetic hyperthermia un- der AMF, the accumulation in tumor and in vivo therapeutic effect were evaluated in the 4T1 apoptosis tumor mice model. MNPs-PEG/𝛽-Lap (60 mg kg−1) was injected on the 2nd and 4th day through a tail vein to construct an apoptotic tumor model. The apoptotic 4T1 tumor mice were randomly for eight groups. Mice were injected MNPs on the 5th and 7th day and then treated under AMF on 5th, 6th, 7th, and 8th day for 30 min (Figure 5a). During the process of therapy, the relative tumor volumes were evaluated every day. As presented in Figure 5b,d, the V/V0 value and the tumor inhibition rate (TIR) in AMF and MNPs-ZnDPA were 23.6 ± 1.2, 21.9 ± 1.0 and 3.6 ± 11.5% and 10.6 ± 6.6%, respectively, which were similar to that of the saline group (V/V0 value 26.6 ± 1.5), implying only AMF and MNPs could not suppress tumor growth. Additionally, for MNPs- ZnDPA/𝛽-Lap group, tumor growth was inhibited to some extent with the V/V0 of 18.1 ± 1.0 and TIR of 18.9 ± 8.8%, which was as- cribed to the chemotherapy effect of 𝛽-Lap. Under AMF, while the V/V0 and TIR values in MNPs-PEG plus AMF group were 14.3 ± 0.7 and 43.3 ± 10.7%, the V/V0 and TIR values in MNPs-ZnDPA plus AMF group were 10.9 ± 1.7 and 58.4 ± 15.3%, demonstrat- ing the apoptosis targeting and magnetic hyperthermia property of MNPs-ZnDPA. More obviously, the V/V0 in MNPs-ZnDPA/𝛽- Lap plus AMF group was 2.0 ± 0.2 smaller than that in MNPs- PEG/𝛽-Lap plus AMF group (9.7 ± 1.0). Meanwhile, TIR val- ues in MNPs-ZnDPA/𝛽-Lap plus AMF group was 86.8 ± 7.4%, which was much higher than that in MNPs-PEG/𝛽-Lap plus AMF group (63.7 ± 8.8%). MNPs-ZnDPA/𝛽-Lap plus AMF group demonstrated the best tumor-inhibiting effect among all groups, which strongly supported the self-amplified apoptosis targeting magnetic hyperthermia therapy combined with chemotherapy of MNPs-ZnDPA/𝛽-Lap in vivo. The photos of mice on the 1st, 4th, 8th, 12th, and 16th day were recorded and except that MNPs- ZnDPA/𝛽-Lap plus AMF group showed no tumor growth, all other groups had the different degree of growth (Figure S21, Sup- porting Information). After the treatment for 16 days, all mice were euthanized, and the tumors were taken photos and weighed (Figure 5c; Figure S22, Supporting Information). In comparison, the tumor growth in MNPs-ZnDPA/𝛽-Lap plus AMF group was greatly inhibited. Meanwhile, body weight of mice in all experi- mental groups showed no significant change (Figure S23, Sup- porting Information).
To further verify self-amplified apoptosis targeting magnetic hyperthermia therapy combined with chemotherapy in vivo, resected tumor tissues were analyzed by H&E and TUNEL stain- ing after 16 days of treatment. Apoptotic tumor mice treated with MNPs-ZnDPA/𝛽-Lap plus AMF exhibited remarkable cell death in tumor tissue compared to other groups (Figure 5f,g), which was in accordance with the therapeutic outcome of afore- mentioned in vivo magnetic hyperthermia therapy. The apopto- sis rate of MNPs-ZnDPA/𝛽-Lap plus AMF group (72.7 ± 4.7%) was much higher than those of MNPs-PEG/𝛽-Lap plus AMF group (45.1 ± 3.6%), MNPs-ZnDPA plus AMF group (39.9 ± 2.0%), MNPs-PEG plus AMF group (28.3 ± 1.1%), and MNPs- ZnDPA/𝛽-Lap group (21.4 ± 5.0%) (Figure 5e), also demonstrat- ing a self-amplified apoptosis targeting magnetic hyperthermia therapy for in vivo tumor. After i.v. injection of MNPs-ZnDPA (60 mg kg−1) 16 days, the main indexes of blood biochemistry and hematology exhibited no obvious abnormality compared with the indexes of healthy mice (Figure S24, Supporting Informa- tion). Moreover, the histological examination of major organs in MNPs-ZnDPA-treated mice measured revealed no apparent histopathological abnormalities or lesions (Figure S25, Support- ing Information).

3. Conclusion

In conclusion, a self-amplified apoptosis targeting nanoplat- form of MNPs-ZnDPA/𝛽-Lap has been successfully es- tablished by combining ZnDPA-conjugated magnetic [email protected] and 𝛽-Lap, in which bis-Zn-DPA performed the self-amplified apoptosis tar- geting moiety, 𝛽-Lap triggered the cell apoptosis, and [email protected] behaved the magnetic hyperthermia function. As authenticated by in vitro exper- iments, in vivo MR imaging and anti-tumor experiments, MNPs-ZnDPA/𝛽-Lap plus AMF could achieve the amplified apoptosis targeting and significantly suppress the tumor growth in 4T1 xenograft mice model. Altogether, our nanoplatform with the strong tumor-targeting capacity and powerful therapeutic effect will have a great potential for the clinical transformation.

4. Experimental Section

Synthesis of MNPs-ZnDPA and MNPs-ZnDPA/FITC: [email protected] nanoparticles (MNPs) were syn- thesized according to the literature.[9] 1 mL of MNPs (10 mg mL−1) was mixed with 5 mL chloroform solution of DSPE-PEG2000 and DSPE- PEG2000-ZnDPA chloroform solution (1 mg mL−1) in a glass vial. After ultrasound for 30 min, chloroform was removed. Then, 5 mL secondary water was added and shaken at 70 °C for 2 min. The excess lipids were removed by dialysis (8000–14 000). MNPs-ZnDPA was obtained after dialysis and 0.1 µm syringe filter. The synthesis of MNPs-ZnDPA/FITC was similar to that of MNPs-ZnDPA except for 5 mL chloroform solu- tion with a concentration of 1 mg mL−1 DSPE-PEG2000, 0.5 mg mL−1 DSPE-PEG2000-ZnDPA, and 0.5 mg mL−1 DSPE-PEG2000-FITC were adopted.
Synthesis of MNPs-ZnDPA/𝛽-Lap: 1 mL of MNPs (10 mg mL−1 in chlo- roform) was mixed with 5 mL with a concentration of 2 mg mL−1 DSPE- PEG2000 and 1 mg mL−1 DSPE-PEG2000-ZnDPA in a glass vial. Then, 𝛽- lapachone hydrochloride (𝛽-Lap) (5 mg) was added and shaken for 24h. The solvent was removed and 5 mL secondary water was added. Af- terward, MNPs-ZnDPA/𝛽-Lap was obtained by centrifugation and washed with deionized water.
MRI in an Aqueous Solution: MNPs-PEG and MNPs-ZnDPA were dis- persed in the deionized water in a 1.5 mL microcentrifuge tube. MR imag- ing in an aqueous solution was carried out with low-field nuclear magnetic resonance (0.5 T). T1/T2 relaxation time of MNPs-PEG and MNPs-ZnDPA was measured by changing the Fe concentration (0.0017 × 10−3, 0.0045 × 10−3, 0.0089 × 10−3, 0.018 × 10−3, 0.044 × 10−3, 0.089 × 10−3, 0.133 × 10−3, and 0.178 × 10−3 M, respectively).
In Vitro Release of 𝛽-Lap: The drug-loaded MNPs were dispersed into PBS solution (pH 5.2, 6.0, and 7.4, respectively) with or without alternat- ing magnetic field (598 kHz, 280 G) for 15 min at 37 °C. At different pre- determined time intervals, nanoparticles were centrifugated, followed by replacing the supernatant with 2 mL PBS. The amount of released 𝛽-Lap was analyzed with UV−vis spectra located at 432 nm. In addition, the same was done at different temperatures (25, 37, and 50 °C) at pH 5.2 without AMF. The 𝛽-Lap load capacity and encapsulation rate were counted accord- ing to the following formula, respectively. The loading content and loading efficiency were calculated to be 7.7% and 38.7% for MNPs-ZnDPA/𝛽-La. The load capacity and encapsulation rate were calculated to be 8.4% and 42.0% for MNPs-PEG/𝛽-La, respectively
In Vitro MRI: 4T1 cells were cultured in a cell plate (1 × 107 cells/plate) for 24 h. First, 1000 µg mL−1 of MNPs-PEG/𝛽-Lap (𝛽-Lap: 77.4 µg mg−1) was used to induce cells to apoptosis. Second, MNPs-PEG and MNPs- ZnDPA were incubated with apoptotic cells for 0, 0.5, 1, 2, 3, and 5 h, respectively. Finally, the apoptotic cells were washed, collected, and analyzed by 0.5 T MR imaging. The self-amplified apoptosis targeting property of MNPs-ZnDPA/𝛽-Lap was then investigated by designing two groups with seven subgroups. One group was 4T1 cells group including two subgroups: 4T1 cells subgroup incubated with DMEM and MNPs- ZnDPA subgroup incubated with MNPs-ZnDPA. The other group was apoptotic 4T1 cells group with six subgroups: apoptotic group incu- bated with DMEM, MNPs-PEG group incubated with MNPs-PEG, inhi- bition group incubated with 5 × 10−6 M ZnDPA for 30 min first, and then cultured with MNPs-ZnDPA, MNPs-ZnDPA group incubated with MNPs-ZnDPA, MNPs-PEG/𝛽-Lap group incubated with MNPs-PEG/𝛽- Lap, and MNPs-ZnDPA/𝛽-Lap group incubated with MNPs-ZnDPA/𝛽- Lap. All groups were incubated with a concentration of 1000 µg mL−1 for 2 h. Afterward, the cells were treated and washed with PBS, trypsinized, centrifuged, and resuspended in a glass vial with PBS including 0.1% agarose. Transverse relaxation time (T2) was surveyed by 0.5 T NM120- Analyst MR System. ΔT2/T2 value = (grouped cell suspension T2 value – control cell suspension T2 value)/control cell suspension T2 value × 100% (n = 3).
Laser Confocal Scanning Microscopy Studies: 4T1 cells (1.0 × 106 cells mL−1) were seeded in a plate and incubated overnight. All the groups were incubated with the concentration of 1000 µg mL−1 of MNPs-PEG/𝛽- Lap for 0.5 h (𝛽-Lap: 77.4 µg mg−1) to induce apoptosis, and washed with PBS. Then, apoptotic cells were randomly for six similar groups: apoptotic group, MNPs-PEG/FITC group, MNPs-PEG/FITC/𝛽-Lap group, MNPs-ZnDPA/FITC group, inhibition group, and MNPs-ZnDPA/FITC/𝛽- Lap group. In all groups, the incubated concentration and time of MNPs were 200 µg mL−1 and 2 h, respectively. Afterward, the cells were washed by PBS and fixed with 3% paraformaldehyde. The fluorescence images were collected from 560 to 600 nm on a Leica Microsystem. The average flu- orescence intensity was determined by the software of CLSM. The rela- tive fluorescence intensity = (experimental group fluorescence intensity – control cell fluorescence intensity)/control fluorescence intensity × 100%. In Vivo MRI Imaging: All animals were lawfully acquired and their re- tention and use were in every case in compliance with federal, state and local laws and regulations, and in accordance with the Institutional Ani- mal Care and Use Committee of SLAC (IACUC) Guide for Care and Use of Laboratory Animals. After the tumor size was about 200 mm3, the tu- mor bearing mice were imaged on a 3.0 T MRI scanner with two groups: tumor group (mice injected with 200 µL saline); apoptosis tumor group (mice injected with MNPs-PEG/𝛽-Lap (60 mg kg−1) twice in four days to construct apoptosis model). To investigate the self-amplified apoptosis targeting property of MNPs-ZnDPA/𝛽-Lap in tumor, apoptotic mice were divided into four subgroups for MRI in vivo on a Magnetom Trio medical MR system (3.0 T, Siemens) as follows. MNPs-PEG group (mice with i.v. injection of MNPs-PEG), MNPs-ZnDPA group (mice with i.v. in- jection of MNPs-ZnDPA), MNPs-PEG/𝛽-Lap group (mice with i.v. injec- tion of MNPs-ZnDPA/𝛽-Lap twice in 4 days, and then exposed under AMF for 30 min. MRI was immediately obtained after the injection of MNPs- PEG/𝛽-Lap into apoptotic mice in the 5th day), and MNPs-ZnDPA/𝛽-Lap group (mice with i.v. injection of MNPs-ZnDPA/𝛽-Lap twice in four days, and then exposed under AMF for 30 min. MRI was immediately obtained after the injection of MNPs- ZnDPA/𝛽-Lap into apoptotic mice in the 5th day). The dose was 60 mg kg−1 in all experiments. The parameters for MRI in vivo were shown as follows: slice thickness = 1 mm; TR = 800 ms; TE = 10.25 ms; FOV = 4 cm × 6 cm; and point resolution = 320 mm × 256 mm. The data were analyzed by Image J software. The ratio of tumor to water T2 (T/W) = T2 value per scan of the tumor/T2 value of water.
In Vivo Therapy: Apoptotic mice were first constructed by injecting with MNPs-PEG/𝛽-Lap (60 mg kg−1) on the 2nd and 4th day until the tumor volumes reached about 100 mm3, and then apoptotic mice were randomly separated into eight groups. Saline group (mice injected with 200 µL saline); AMF group (mice under AMF for 30 min); MNPs-ZnDPA group (mice with the injection of MNPs-ZnDPA through the tail vein); MNPs-ZnDPA/𝛽-Lap group (mice with the injection of MNPs-ZnDPA/𝛽- Lap through the tail vein); MNPs-PEG+AMF group (mice with the injec- tion of MNPs-PEG through the tail vein, and then radiated by AMF twice for 30 min at the 5th hour and 12th hour, respectively.) MNPs-PEG/𝛽- Lap+AMF group (mice with the injection of MNPs-PEG/𝛽-Lap through the tail vein, and then radiated by AMF twice for 30 min at the 5th hour and 12th hour, respectively). MNPs-ZnDPA+AMF group (mice with the in- jection of MNPs-ZnDPA, and then radiated by AMF twice for 30 min at the 3rd hour and 12th hour, respectively). MNPs-ZnDPA/𝛽-Lap+AMF group (mice with the injection of MNPs-ZnDPA/𝛽-Lap through the tail vein, and then radiated by AMF twice for 30 min at the 3rd hour and 12th hour, re- spectively). Every Beta-Lapachone group had five mice. The dose was 60 mg kg−1 in all experiments. The parameters of AMF were 256 kHz and 300 G, respec- tively.

References

[1] a) Z. Li, J. Han, L. Yu, X. Qian, H. Xing, H. Lin, M. Wu, T. Yang, Y. Chen, Adv. Funct. Mater. 2018, 28, 1800145; b) S. Y. Yin, G. Song, Y. Yang, Y. Zhao, P. Wang, L. M. Zhu, X. Yin, X. B. Zhang, Adv. Funct. Mater. 2019, 29, 1901417; c) M. Wang, K. Deng, W. Lu, X. Deng, K. Li, Y. Shi, B. Ding, Z. Cheng, B. Xing, G. Han, Z. Hou, J. Lin, Adv. Mater. 2018, 30, 1706747.
[2] a) D. Ni, C. A. Ferreira, T. E. Barnhart, V. Quach, B. Yu, D. Jiang, W. Wei, H. Liu, J. W. Engle, P. Hu, W. Cai, J. Am. Chem. Soc. 2018, 140, 14971; b) Z. Li, C. Di, S. Li, X. Yang, G. Nie, Acc. Chem. Res. 2019, 52, 2703.
[3] a) D. B. Cheng, D. Wang, Y. J. Gao, L. Wang, Z. Y. Qiao, H. Wang, J. Am. Chem. Soc. 2019, 141, 4406; b) Z. Q. Zhang, Y. M. Kim, S. C. Song, ACS Appl. Mater. Interfaces 2019, 11, 34634.
[4] a) K. Wang, M. H. Na, A. S. Hoffman, G. Shim, S. E. Han, Y. K. Oh, I. C. Kwon, I. S. Kim, B. H. Lee, J. Controlled Release 2011, 154, 214; b) B. Zhang, H. Wang, S. Shen, X. She, W. Shi, J. Chen, Q. Zhang, Y. Hu, Z. Pang, X. Jiang, Biomaterials 2016, 79, 46; c) H. Zhao, P. Zhou, K. Huang, G. Deng, Z. G. Zhou, J. Wang, M. W. Wang, Y. J. Zhang, H. Yang, S. P. Yang, Adv. Healthcare Mater. 2018, 7, 1800296.
[5] a) B. Sanz, M. P. Calatayud, T. E. Torres, M. L. Fanarraga, M. R. Ibarra, G. F. Goya, Biomaterials 2017, 114, 62; b) N. A. Stocke, P. Sethi, A. Jyoti, R. Chan, S. M. Arnold, J. Z. Hilt, M. Upreti, Biomaterials 2017, 120, 115; c) S. Moise, J. M. Byrne, A. J. El Haj, N. D. Telling, Nanoscale 2018, 10, 20519.
[6] a) X. Liu, J. Zheng, W. Sun, X. Zhao, Y. Li, N. Gong, Y. Wang, X. Ma, T. Zhang, L. Y. Zhao, Y. Hou, Z. Wu, Y. Du, H. Fan, J. Tian, X.J. Liang, ACS Nano 2019, 13, 8811; b) A. Espinosa, R. Di Corato, J. Kolosnjaj-Tabi, P. Flaud, T. Pellegrino, C. Wilhelm, ACS Nano 2016, 10, 2436.
[7] a) K. H. Min, Y. H. Kim, Z. Wang, J. Kim, J. S. Kim, S. H. Kim, K. Kim, I.C. Kwon, D. O. Kiesewetter, X. Chen, Theranostics 2017, 7, 4240; b) J. J. Lee, K. J. Jeong, M. Hashimoto, A. H. Kwon, A. Rwei, S. A. Shankarappa, J. H. Tsui, D. S. Kohane, Nano Lett. 2014, 14, 1; c) J. M. Kwong, C. Hoang, R. T. Dukes, R. W. Yee, B. D. Gray, K. Y. Pak, J. Caprioli, Invest. Ophthalmol. Visual Sci. 2014, 55, 4913; d) K. Y. Choi, O. F. Silvestre, X. Huang, N. Hida, G. Liu, D. N. Ho, S. Lee, S. W. Lee, J. I. Hong, X. Chen, Nat. Protoc. 2014, 9, 1900.
[8] J. Lee, H., J. T. Jang, J. S. Choi, S. H. Moon, J. Kim, C. Kyu, S. H. Noh, J. W. Kim, J. G. Kim, I. S. Kim, K. I. Park, J. Cheon, Nat. Nanotechnol. 2011, 6, 418.
[9] S. Sun, H. Z., D. B. Robinson, S. Raoux, P. M. Rice, a. G. L. Shan, X. Wang, J. Am. Chem. Soc. 2004, 126, 273.
[10] H. A. Albarqi, L. H. Wong, C. Schumann, F. Y. Sabei, T. Korzun, X. Li, M. N. Hansen, P. Dhagat, A. S. Moses, O. Taratula, O. Taratula, ACS Nano 2019, 13, 6383.
[11] W. Chen, S. Zhou, L. Ge, W. Wu, X. Jiang, Biomacromolecules 2018, 19, 1732.
[12] G. A. F. v. Tilborg, W. J. M. Mulder, P. T. K. Chin, G. Storm, C. P. Reutel- ingsperger, K. Nicolay, G. J. Strijkers, Bioconjugate Chem. 2006, 17, 865.
[13] M. Z. Ye, Y. X. Han, J. B. Tang, Y. Piao, X. R. Liu, Z. X. Zhou, J. Q. Gao, J. H. Rao, Y. Q. Shen, Adv. Mater. 2017, 29, 1702342.
[14] a) X. T. Zheng, C. M. Li, Mol. Pharmaceutics 2012, 3, 615; b) K. Hayashi, K. Ono, H. Suzuki, M. Sawada, M. Moriya, W. Sakamoto, T. Yogo, ACS. Appl. Mater. Interfaces 2010, 2, 1903.