SEARCH

SEARCH BY CITATION

Keywords:

  • superparamagnetic iron oxide (SPIO);
  • microcapsule;
  • hyperthermia;
  • theranostics;
  • magnetic resonance imaging (MRI);
  • chemotherapy;
  • cancer stem cells

Abstract

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Results and Discussion
  5. 3 Conclusion
  6. 4 Experimental Section
  7. Acknowledgements

Multifunctional magnetic microcapsules (MMCs) for the combined cancer cells hyperthermia and chemotherapy in addition to MR imaging are successfully developed. A classical layer-by-layer technique of oppositely charged polyelectrolytes (poly(allylamine hydrochloride) (PAH) and poly(4-styrene sulfonate sodium) (PSS)) is used as it affords great controllability over the preparation together with enhanced loading of the chemotherapeutic drug (doxorubicin, DOX) in the microcapsules. Superparamagnetic iron oxide (SPIOs) nanoparticles are layered in the system to afford MMC1 (one SPIOs layer) and MMC2 (two SPIOs layers). Most interestingly, MMC1 and MMC2 show efficient hyperthermia cell death and controlled DOX release although their magnetic saturation value falls below 2.5 emu g−1, which is lower than the 7–22 emu g−1 reported to be the minimum value needed for biomedical applications. Moreover, MMCs are pH responsive where a pH 5.5 (often reported for cancer cells) combined with hyperthermia increases DOX release predictably. Both systems prove viable when used as T2 contrast agents for MR imaging in HeLa cells with high biocompatibility. Thus, MMCs hold a great promise to be used commercially as a theranostic platform as they are controllably prepared, reproducibly enhanced, and serve as drug delivery, hyperthermia, and MRI contrast agents at the same time.

1 Introduction

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Results and Discussion
  5. 3 Conclusion
  6. 4 Experimental Section
  7. Acknowledgements

Most tumors are heterogeneous and many cancers contain a small population of highly tumorigenic and intrinsically drug-resistant cancer stem cells (CSCs). These cells have normal stem-cell-like properties and are thought to be responsible for tumor drug resistance and relapse. Therapies that can effectively eliminate CSCs will, therefore, likely inhibit tumor recurrence.[1] CSCs in various cancers including breast, leukemia, gastrointestinal, and brain are characterized by a relatively low level of reactive oxygen species (ROS), which combined to higher DNA repair results in radioresistance of CSCs.[2-6] The lower levels of ROS are attributed to the higher content of antioxidant ROS-scavenging compounds.[5, 7] CSCs are also often localized in the hypoxic areas of the tumors. Thus, it is suggested that by enhancing tumor oxygenation it is possible to make CSCs more susceptible to current treatments. Hyperthermia has been used to promote reoxygenation, radiosensitization, and heat shock in cancer cells.[8-11] Several studies have focused on killing CSCs by using magnetic nanoparticles (NPs) upon applying alternating magnetic field.[12-14] However, current clinical implementations of hyperthermia have been limited due to the nonspecific heating to the normal tissues and consequent treatment-limiting side effects.[15]

Together with therapy, diagnosis of CSCs is vital to guarantee successful results. Magnetic resonance imaging (MRI) is one of the most powerful medical diagnosis tools because MRI provides images with excellent details based on the soft tissue contrast and functional information in non-invasive and real-time monitoring manner.[16-25] The sensitivity of MRI can be greatly improved by the use of contrast agents that enhance the contrast of the region of interest from background. Thus, designing a theranostic system that can image and treat CMCs with accuracy and no to little side effects is highly desirable.

To this end, NPs including superparamagnetic iron oxide (SPIO),[26] metal NPs (Au,[27] Ag,[28] and quantum dot[29] have attracted plenty of interest due to their potential biomedical applications in drug delivery and cell imaging.[30-32] As one of the most versatile and biocompatible NPs in medicine,[33] SPIO can be used for drug delivery, cell separation and tracking, MRI, and hyperthermia.[34-37] Drug delivery system associated with SPIO can be controlled remotely and its cell internalization can be improved using external magnetic field.[38-40] SPIO can increase the sensitivity of MRI for cancer detection and therapeutic response as it induces a shorter T2 relaxation, improving contrast on a T2-weighted image.[20, 41] In addition, SPIO can induce cell apoptosis and cell death through magnetic hyperthermia.[42] Many reports have been published on the use of SPIOs as magnetic triggers; however, some major hurdles for these systems included overcoming SPIOs tendency to aggregate, improving SPIOs biocompatibility, as well as ensuring controllability and reproducibility of the preparation method.[33-35] Coating SPIOs is often needed to guarantee the general stability of the system.[43-45] Moreover, SPIOs as T2 contrast agents are hampered by several disadvantages mainly due to their usually high magnetization, which induces perturbation of the local magnetic field, causing the so-called “blooming effect.”[46] This effect exaggerates the size of the labeled area and blurs the image.

Here, low magnetization magnetic microcapsules (MMCs) are prepared via a facile layer-by-layer (LbL) technique that prove to have synergistic chemotherapy and hyperthermia effect together with T2 contrast applicability for MR imaging without perturbing the local magnetic field. LbL, first reported and developed by Decher et al.[47, 48] for assembling thin films on solid surface, has attracted lots of attention for developing polyelectrolyte multilayers microcapsule (MC). This unique architecture of MC has enormous applications in the fields of drug delivery,[49] gene carrier,[50] biosensors,[51, 52] and cellular imaging.[53] Most importantly, LbL assembly has the advantage of high controllability, reproducibility, and high loading capacity. Moreover, MMCs are superior to magnetic NPs due the their fast magnetic response and low loss rate during the treatment process.[45] As a proof of concept, poly(4-styrene sulfonate sodium) (PSS) and poly(allylamine hydrochloride) (PAH), one of the most classical polyelectrolyte pairs used to prepare LbL MCs, are used to construct our MCCs.[54] Another advantage of this system is pH responsiveness (at pH 5.5), as it was shown that the polyelectrolyte layers are highly influenced by acidic pH. This stimulated distortion in the polymeric layers causes the delivery or release of cargo from MCs. However, if pH is used as the only trigger then very low values are needed (pH 2) to guarantee sufficient release.[55] Different techniques and systems have been reported on magnetically controlled rapture of drug delivery systems but, to the best of our knowledge, none were used for combined hyperthermia and chemotherapy in addition to serving as T2 contrast agents at low magnetization value.[56-60]

Two kinds of magnetic microcapsules (MMCs) are prepared by LbL namely MMC1 (with one layer of SPIOs) and MMC2 (with 2 layers of SPIOs). MCs with no SPIO loading are also prepared using the same method and used as a control. A chemotherapeutic drug, doxorubicin (DOX), is effectively loaded into the MMCs, due to the electrostatic interaction between negative MCs center and positively charged DOX at ambient pH. Both MMC1 and MMC2 are responsive to hyperthermia excitation where DOX is also controllably released in spite of showing a very low magnetic saturation value. Lowering pH value caused further drug release, which is favorable in cancer tissues where pH is mainly acidic. MR imaging of HeLa cells is also successfully demonstrated using both systems where low magnetization should reduce the possibility of the blooming effect.

2 Results and Discussion

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Results and Discussion
  5. 3 Conclusion
  6. 4 Experimental Section
  7. Acknowledgements

2.1 Synthesis and Characterization of MMCs

Monodispersed spherical PSS-doped CaCO3 particles with a diameter of 4–5 μm are employed as a sacrificial template (Figure S1, Supporting Information). Hydrophobic SPIO NPs (Figure 1a) are synthesized by a thermal decomposition method and are further used to prepare hydrophilic SPIOs (Figure 1b) through a ligand-exchange reaction.[61] The particles are monodispersed with small size distribution and reveal a good ferromagnetic property (Figure 1c). The hydrophilic SPIO aqueous dispersion has a Zeta potential of −13.9 ± 4.87 mV. PSS, PAH, and SPIO are then used to prepare MCs by the LbL method (refer to experimental section).[62] After deposition of the desired layers, CaCO3 particles are etched by EDTA (ethylenediaminetetraacetic acid) to yield the hollow center for DOX loading (Scheme 1).

image

Figure 1. TEM images of a) hydrophobic and b) hydrophilic SPIO NPs. c) Field-dependent magnetization loop for the SPIO NPs in hexane at 300 K.

Download figure to PowerPoint

image

Scheme 1. Schematic illustration of layer-by-layer process and release of DOX from MMCs.

Download figure to PowerPoint

The LbL assembly progress of polyelectrolyte layers and SPIO is monitored by ζ-potential measurements (Figure S2, Supporting Information). The MCs with one layer of SPIO NPs (MMC1) are recorded as (PSS/PAH)(SPIO/PAH)(PSS/PAH)2; and the MCs with two layers of SPIO NPs (MMC2) are recorded as (PSS/PAH)(SPIO/PAH)2(PSS/PAH). While MCs without SPIO NPs layers (MC) are denoted as (PSS/PAH)5 and used as a control. The morphology of MCs at the air-dried state can be seen in Figure 2. The visible collapse caused by the evaporation of the aqueous content in the MCs during the drying process verifies their hollow nature. Optical microscope (Figure 3) and transmission electron microscopy (TEM) images (Figure 4a–d) provide ample evidence of the hollow structures. The magnified TEM images indicate the rather uniform and continuous distribution of SPIO NPs in the wall of MCs and the thickness of SPIO layer in MMC1 (27.14 ± 6.29 nm, Figure 4b) is thinner than that in MMC2 (56.0 ± 9.47 nm, Figure 4d).

image

Figure 2. SEM images of a) MC, b) MMC1, and c) MMC2.

Download figure to PowerPoint

image

Figure 3. Optical microscopy images of a) MC, b) MMC1, and c) MMC2. The MC embedded ultrathin sections of 0.5% toluidine blue for optical microscope analysis.

Download figure to PowerPoint

image

Figure 4. TEM images of a,b) MMC1 and c,d) MMC2; (b) is the amplified image of (a) and (d) is the amplified image of (c). e) Field-dependent magnetization loops for lyophilized MC, MMC1, and MMC2, inset: photos of MC, MMC1, and MMC2 solutions in the absence and presence of a magnet. T2-weighted MRI images of f) MMC1 and g) MMC2.

Download figure to PowerPoint

From the inductively coupled plasma (ICP-OES) test, the iron ion concentration in 1 × 107 MMC1 and MMC2 are 5.66 ppm and 11.78 ppm, respectively. Thus, the average iron concentration of each MC can be calculated. The consistent results of SPIO thickness and Fe concentrations demonstrate consecutive coatings by PAH and SPIO are carried through successfully.

Owing to the presence of SPIO in the wall of MCs, MMC1 and MMC2 show superparamagnetism, as revealed by the field-dependent magnetization measurement, while no magnetism is shown from the magnetization loop of MC (Figure 4e). Furthermore, the ferromagnetic property of MCs is shown by visual photos (Figure 4e, inset photos). The obtained magnetization values for MMC1 and MMC2 are 1.2 and 2.2 emu g−1, respectively, which are very low compared to published reports where a magnetization range of 7–22 emu g−1 is usually adopted for bioapplications.[45, 63-66] The reason behind this huge drop in magnetization compared to bare SPIOs is well documented in Literature as the spin-canting effect.[46] This is explained simply by the presence of polymer coating that can disorder the spins of SPIOs causing a reduction in the overall magnetization value. Oleic acid is usually used to stabilize SPIOs and thus magnetization values are corrected in terms of oleic acid content.[63] In this case, LbL polyelectrolytes are not conjugated to SPIOs and thus the magnetization value is controlled by the amount of SPIO present in the layers. Interestingly, both MMC1 and MMC2 can act as T2 contrast agents for MR imaging (Figure 4f,g). Concentration-dependent darkening effect is observed when measuring R2 (1/T2), which is the transverse relaxation rate with a known concentration of contrast agent, and was through T2-weighted MRI images. r2 of MMC1 (Figure 4f) and MMC2 (Figure 4g) can be calculated as, r2, MMC1 = 56.54 ± 3.59 mm−1 s−1, r2, MMC2 = 29.59 ± 4.71 mm−1 s−1.

2.2 Synergistic Hyperthermia and Chemotherapy Effects

After verifying the basic optical and magnetic properties of the synthesized MMCs, they are used as cancer therapy platforms. The cellular internalization of MMCs is investigated by incubating MMCs with HeLa cells for 12 h at 37 °C. The nuclei and cell membrane are stained by Hoechst 33342 and CellMask deep red separately, while MMCs are labeled with FITC. The co-localization of MMC1 (Figure 5a–e) and MMC2 (Figure 5a′–e′) is then determined using confocal laser scanning microscopy (CLSM). Z-stack was performed to identify the MMCs intracellular location, showing the orthogonal view of the three planes (x/y/z) of the internalized MMC1 (Figure 5e) and MMC2 (Figure 5e′). The detailed cell internalization of MCs was also studied using scanning electron microscopy (SEM) and TEM. There is a closer adherence of MMC1 (Figure 6a) and MMC2 (Figure 6d) onto cells after incubation and the filopodia of the cells surround MMCs to stabilize them at the cell membrane.[67] This kind of interaction makes the following internalization possible. From the TEM images, the intracellular MMC1 (Figure 6b) and MMC2 (Figure 6e) can be seen clearly. The embedded SPIO in the walls of MMCs can be clearly identified in the magnified TEM images (Figure 6c,f). CLSM (Figure S4, Supporting Information), SEM (Figure S5a, Supporting Information), and TEM (Figure S5b, Supporting Information) images of cells incubated with plain MC are shown in the Supporting Information as reference.

image

Figure 5. CLSM images of HeLa cells after incubation with a–e) MMC1 and a′–e′) MMC2.

Download figure to PowerPoint

image

Figure 6. SEM images of cells after incubation with a) MMC1 and d) MMC2 for 12 h. TEM images of cells with intracellular b,c) MMC1 and e,f) MMC2. ((c) is the amplified image of (b) and (f) is the amplified image of (e)).

Download figure to PowerPoint

MMCs are quantified using a hemacytometer under a microscope. DOX was employed as a model drug, and mixed with MMCs at 37 °C overnight. Positively charged DOX can diffuse into the MCs and bind with excess negatively charged PSS in the center of MCs due to electrostatic interaction. DOX will not bind to the layered PSS as the charge is occupied by PAH. According to the calibration curve of DOX (Figure S6, Supporting Information), the average load efficiency of DOX in MMC1, MMC2 is calculated to be 53.1% and 58.6% (Figure S7, Supporting Information), respectively from Equation S1 (Supporting Information), which shows a decent loading potential. The cumulative release amount of DOX from MMC1 (Figure 7a) and MMC2 (Figure 7b) with or without magnetic hyperthermia stimulation (MHS, 7 kW, output power of 80%, 375Hz) is then investigated by testing the time-dependent UV–vis spectra of supernant after dispersing the DOX-loaded MMCs in PBS (pH 5.5 or 7.4). It is obvious that the release rate of both samples at pH 5.5 is much larger than that at pH 7 confirming that MMCs are pH sensitive, which is a known fact for MCs composed of weak polyelectrolyte.[68, 69] This type of MCs is permeable in an acidic environment because of weakening interaction between the polyelectrolyte pairs whereas it is impermeable in an alkaline environment.[70] This provides an extra benefit of the designed system as cancer cells are known for their mildly acidic environment. Moreover, the SPIOs in the wall of MCs vibrate and consequently heat their surroundings inducing the relaxation of polymer chains and dissociation of polyelectrolyte pairs when the temperature is raised beyond their Tg. Moreover, external alternating magnetic fields can induce mechanical vibration and motion of SPIO, which can increase the stress in the wall of MCs.[71] All these factors synergistically contribute to the MCs increased permeability and their eventual rapture. After magnetic stimulation for 2 h, the cumulative release of DOX from MMC1 and MMC2 is around 60% and 85%, respectively. For MC (Figure S8, Supporting Information), it shows a minor pH-responsive release (pH 5.5) with a self-heating effect caused by MHS. MCs with magnetic stimulation do not show much release due to the low content of SPIOs, which gives further stability and controllability to the system as a synergistic effect between low pH and MHS is needed for the proper function of the system.

image

Figure 7. Release behavior of DOX from a) MMC1 and b) MMC2 with or without MHS at different pH for 2 h. c) Relative cell death after treatment with MMCs (MMC1, DOX-MMC1, MMC2, and DOX–MMC2) with or without MHS for 30 min. d) Cell viability after treatment with MC, MMC1, and MMC2.

Download figure to PowerPoint

Determining the relative cell death is then accomplished by testing the lactate dehydrogenase (LDH) level. Different MMCs (106) (MMC1, MMC2, and DOX-loaded MMC1, MMC2) are incubated with HeLa cells and placed in or out of MHS for 30 min. The relative cell death is then measured by LDH assay (Figure 7c). The unloaded MMCs showed the lowest death rates of cells without MHS treatment (25%–30%); while cell death is in the range of 35%–40% with MHS treatment. Thus, hyperthermia therapy can induce extra cell death.

The cytotoxicity of MMCs is evaluated by incubating HeLa cells with eight different concentrations of MCs (MMC1, MMC2, and MC) for 24 h followed by standard MTT assay. In previous studies,[72, 73] at a concentration of 100 capsules per cell, cells cultured with 1–2 μm PSS/PAH MCs have >80% viability, while the number is less than 25% when PSS/PAH MCs had a diameter of 8–10 μm. Thus, the cytotoxicity of PSS/PAH could be size dependent. In our work, the diameter of MCs is 4–6 μm and its cytotoxicity is consistent with literature. For MMC1 and MMC2, with reducing the ratio to cells, there is a continuous increase of cell viability and it proves to be more viable than plain MCs and even other LbL systems already reported (Figure 7d).[72, 73]

2.3 Magnetic Resonance Imaging

Magnetic resonance imaging of HeLa cells using MMCs as the contrast agents is shown in Figure 8. HeLa cells are incubated with MCs (MMC1, and MMC2) for 24 h and washed by DPBS (Dulbecco's phosphate buffered saline) three times to remove the excess MCs, followed with fixation. Three vials are prepared for each kind of MC: cells incubated with MCs and embedded in 2% agarose gel (sample), cells in 2% agarose gel (control), and 2% agarose gel (background).

image

Figure 8. MRI images of cells with intracellular a) MC, b) MMC1, and c) MMC2. Three vials were prepared containing 2% agarose: (background) blank gels; (control) HeLa cells in 2% agarose gels, and (sample) HeLa cells after incubation with microcapsules for 24 h.

Download figure to PowerPoint

These three vials are then put in a bigger vial together and embedded in 2% agarose gel in order to simulate the biological tissues before and after injection of MRI contrast agents. Owing to the relatively strong dipolar interaction between SPIO in MMCs and protons in the intracellular fluid, HeLa cells incubated with MMC1 (Figure 8b) and MMC2 (Figure 8c) are displayed as black dots. However, no black contrast is seen in cells incubated with MC (Figure 8a). This test confirms that MMCs can act as effective MRI contrast agents and the T2-weighted images of cells can be obtained after cell internalization. Moreover, recent studies have connected the “dipole blooming effect,” which exaggerates the size of the labeled area and blurs the image, to the high magnetization of T2 agents usually employed in MRI.[46, 74] Our system has the advantage of low magnetic saturation value that can overcome this obstacle for MR imaging. Further investigations, for the application of this system as T1 imaging agent is currently underway.

3 Conclusion

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Results and Discussion
  5. 3 Conclusion
  6. 4 Experimental Section
  7. Acknowledgements

A theranostic platform based on MCCs is prepared by a simple and feasible LbL technique. MCCs show high capability to load anticancer drugs such as DOX, automatically. Other than the controllable preparation and high loading, these MCs have the synergistic effect of hyperthermia and chemotherapy on HeLa cells. Stimuli-responsive release of DOX is successfully demonstrated, with a combined effect of magnetic hyperthermia stimulus and low pH (5.5) giving the best release values. Cancer cells killing is also achieved by using hyperthermia alone through cell heating. MR imaging was successfully demonstrated using MCCs as T2 contrast agents. The low magnetization values of MCCs affords extra biocompatibility and superior controllability over drug release, in addition to overcoming possible drawbacks using high-magnetization contrast agents. Current efforts are directed towards improving the biocompatibility of the system to increase its chances of being employed as an effective theranostic agent for treating and imaging of CSCs.

4 Experimental Section

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Results and Discussion
  5. 3 Conclusion
  6. 4 Experimental Section
  7. Acknowledgements

Synthesis of Hydrophilic SPIO NPs: According to the literature,[61] Fe(acac)3 (2 mmol), 1,2-hexadecanediol (10 mmol), oleic acid (6 mmol), and oleylamine (6 mmol) were dissolved into benzyl ether (20 mL) under an nitrogen flow. In the ambient nitrogen, the mixture was heated to 200 °C for 2 h, then up to reflux for another 1 h. After the mixture was cooled to room temperature, crude particles were precipitated and collected by adding ethanol (40 mL), and then dispersed in a hexane solution (15 mL) consisting of oleic acid (0.05 mL) and oleylamine (0.05 mL). After removing undispersed substances, the hydrophobic SPIO NPs were obtained after being re-precipitated with ethanol and re-dispersed in hexane.

The hexane solution of hydrophobic SPIO NPs was added into 2 mL of dichloromethane suspension of tetramethylammonium 11-aminoundecanoate (20 mg). After shaking for 20 min, the particles were collected by using magnet and then washed with dichloromethane. The hydrophilic Fe3O4 particles were dried under nitrogen and dispersed in deionized water.

The morphology of SPIO NPs was observed by TEM (an FEI Tecnai T12 microscope) and the magnetic property of SPIO NPs was studied by a commercial Quantum Design superconducting quantum interference device (SQUID) magnetometer.

Preparation and Characterization of MMCs: In brief,[62] 200 mL of calcium nitrate solution (Ca(NO3)2·4H2O, 0.025 m) consisting of 200 mg PSS, was rapidly poured into an equal volume of sodium carbonate (Na2CO3, 0.025 m) under magnetic stirring. After standing for 20 min, the precipitated PSS-doped CaCO3 particles were collected and washed with deionized water. The morphology of CaCO3 particles was observed by SEM (FEI Quanta 200).

Two types of MMCs with different Fe concentrations were prepared. In a typical preparation process, the PSS-doped CaCO3 particles (1% w/v in water) were incubated in PAH solution (2 mg mL−1 in 0.5 m NaCl) for 20 min under continuous shaking. The coated microparticles were collected through centrifugation and washed three times with water. Then they were incubated in SPIO NPs solution (0.1 mg mL−1) using the same procedure. The multilayer coatings with different configurations were formed by alternate deposition of corresponding materials. The hollow MCs were obtained after removing the CaCO3 templates by using EDTA solution (0.1 m, pH 7.0). The MC with one layer of SPIO NPs (MMC1) was recorded as (PSS/PAH)(SPIO/PAH)(PSS/PAH)2; and the MC with two layers of SPIO NPs (MMC2) was recorded as (PSS/PAH)(SPIO/PAH)2(PSS/PAH). While (PSS/PAH)5 without SPIO NPs layers (MC) was similarly prepared for comparison.

The LbL process was tracked by a Malvern NanoZS Zetasizer. The morphology of MCs was observed by SEM. In order to characterize the structure of MCs, they were first pre-embedded in argarose and dehydrated using a graded ethanol series, then replaced by acetone followed by epoxy resin. Finally, the samples were curried at 60 °C over night, and cut into ultrathin sections. TEM was used to further verify the hollow structure and the distribution of SPIO NPs in the wall of MMCs. The Fe content in MMC1 and MMC2 was measured by inductively coupled plasma optical emission spectrometry (ICP-OES). MMC1, MMC2 (2 mL) suspension was centrifuged and the solvent was removed, then hydrochloric acid was added to make ferric ion in the solution. The magnetic properties of MC, MMC1, MMC2 were measured by a SQUID magnetometer as described above.

Doxorubicin Loading and Release: The MC number was quantified using a hemacytometer under a microscope. MCs (2.7 × 107) were mixed with 500 μL DOX solution (1 mg mL−1) and incubated at 37 °C over night. The DOX-loaded MCs were washed with PBS (phosphate buffered saline, pH 7.4), and the supernatant was collected to calculate the loading efficiency after centrifugation. The calibration curve of DOX was shown as Figure S6 (Supporting Information).

The release experiment was performed in two different PBS buffers (pH 7.4, pH 5.5) at room temperature. Two comparative behaviors of release, with or without treatment of magnetic hyperthermia system (MHS), were observed. The release process was carried out as follows: DOX-loaded MCs were mixed with 1 mL of PBS, 600 μL of supernatant was taken for absorbance measurment as original point, and then the supernatant was poured back into the tube. Samples treated or not treated with MHS were centrifuged at the time intervals of 15 min, and 600 μL of supernatant was taken to test the time-resolved absorbance.

Cells Viability of HeLa Cells Treated with MC, MMC1, and MMC2: The cytotoxicity of MC, MMC1, and MMC2 incubated with HeLa cells was evaluated using the MTT assay. Cells were seeded at a density of 5 × 103 cells per well in 96-well flat bottom plates and incubated with EMEM medium containing 10% FBS and 0.1% penicillin–streptomycin at 37 °C in a humidified 5% CO2 atmosphere for 12 h. After cell attachment, they were washed with DPBS and incubated with different proportion concentration (50:1, 10:1, 5:1, 1:1, 0.5:1, 0.1:1, 0.05:1, 0.01:1) of MCs solutions in EMEM media for 24 h. Cell viability was evaluated by the MTT colorimetric procedure.

Intracellular Localization and Internalization of MC, MMC1, and MMC2: HeLa cells were seeded on glass cover slides or in petri dish, and cultured in EMEM medium containing 10% FBS, and 0.1% penicillin–streptomycin at 37 °C in a humidified 5% CO2 atmosphere. After cell attachment, the medium was replaced by medium containing MCs, followed by incubation for 12 h. Cells on cover slides were washed twice with DPBS, then fixed with 4% paraformaldehyde for 1 h and washed three times with DPBS. Cells were then dehydrated using a graded ethanol series, and then dried by critical point CO2. The cover slides were coated by carbon prior to SEM analysis. Cells in petri dish were washed twice with DPBS following a series of fixation and washed three times with DPBS, then dehydrated using a graded ethanol series, and replaced by acetone, followed by epoxy resin. The samples were curried at 60 °C overnight, cut into ultrathin sections, and loaded onto copper TEM grids to TEM analysis.

HeLa cells were seeded in CLSM dish and cultured in EMEM medium containing 10% FBS and 0.1% penicillin–streptomycin at 37 °C in a humidified 5% CO2 atmosphere. After cell attachment, nuclei were stained with Hoechst 33342 for 10 min and then washed for three times with DPBS. FITC-labeled MCs were added and incubated for 12 h. Cell membrane was stained with CellMask deep red plasma membrane dye for 3 min and then washed for three times with DPBS. Finally, cells were fixed by 4% paraformaldehyde for CLSM (Zeiss LSM 710 upright confocal microscope).

LDH Assay: Cells (1 × 106 ) were suspended in DPBS buffer consisting of 1% of FBS and mixed with 1 × 106 MCs (MMC1, DOX-MMC1, MMC2, DOX-MMC2). Then they were placed under MHS (7 kW, output power of 80%, 375 Hz) or kept at room temperature for 30 min. The cells were centrifuged and the amount LDH in supernatant was analyzed by LDH kit. The DPBS buffer with 1% FBS was used as background and the LDH released from equal number of cells incubated with 2% Triton X-100 was used to calculate 100% cell death.

Magnetic Resonance Imagining: Different concentrations of MCs (Figure 4f,g) were dispersed in 2% agarose (w/v) gels. Blank gel acted as reference, and MRI data were acquired on a Bruker 500SWB spectrometer with a superwide bore 11.7 T magnet resonating at 500 MHz. The average T2 relaxation times were calculated using a Carr–Purcell–Meiboom–Gill (CPMG) spin echo imaging pulse sequence.[75]

MRI images of cells were carried on the same Bruker 500SWB spectrometer. First, HeLa cells were incubated with MCs (MMC1, MMC2) for 24 h. Then, the free MCs were removed by washing three times with DPBS, followed by fixation by 4% paraformaldehyde. The cells were collected and redispersed in 2% agarose gels for MRI.

Acknowledgements

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Results and Discussion
  5. 3 Conclusion
  6. 4 Experimental Section
  7. Acknowledgements

This work was supported by King Abdullah University of Science and Technology (KAUST), the Natural Science Foundation of China (NSFC-21006072), and Natural Science Foundation of Tianjin (No.11JCYBJC04400).