Sustained release profile of quatro stimuli nanocontainers as a multi sensitive vehicle exploiting cancer characteristics

Colloids and Surfaces B: Biointerfaces 148 (2016) 95–103

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Sustained release profile of quatro stimuli nanocontainers as a multi sensitive vehicle exploiting cancer characteristics Christos Tapeinos a,b,1 , Eleni K. Efthimiadou a,∗,1 , Nikos Boukos a , George Kordas a,∗ a b

Sol-Gel laboratory, Institute of Nanoscience and Nanotechnology, NCSR “Demokritos”, 153 10 Aghia Paraskevi Attikis, Greece Materials Science Department, School of Natural Sciences, University of Patras, 26 500 Patras, Greece

a r t i c l e

i n f o

Article history: Received 3 March 2016 Received in revised form 12 August 2016 Accepted 16 August 2016 Available online 20 August 2016 Keywords: Stimuli responsive Hollow nanocontainers pH-sensitive Thermo-sensitive Redox-sensitive Magnetic nanoparticles Quatro-stimuli Radiolabelling 99 m Tc in vivo imaging Biodistribution

a b s t r a c t A versatile drug delivery carrier that responds to external stimuli was synthesized via the emulsion polymerization process. This simple two-step process was carried out by using Poly (Methyl Methacrylate) as a soft template and a series of monomers, with desired properties, as coating monomers. It is noteworthy that during shell fabrication (2nd step) an inner cavity is created inside the nanocontainers that can be used as a host for small drug molecules. The thermo-, pH- and redox sensitive monomers used in the coating procedure were Dimethyl Amino Ethyl Methacrylate (DMAEMA), Acrylic Acid (AA) and N,N -(disulfanediylbis(ethane-2,1-diyl))bis(2-methylacrylamide) (Disulfide or DS), respectively. It has to be noted that DMAEMA is also pH- sensitive and acts synergistically with AA. The surface of the multistimuli nanocontainers was functionalized with magnetite nanoparticles in order to induce an alternating magnetic field (AMF) sensitivity. By using AMF in various strenghts and frequencies, the temperature of the final multi-stimuli nanocontainers (Q-NCs) can be increased in a controlled manner resulting in the Hyperthermia phenomenon. Loading and release studies were carried out using the anthracycline drug, Doxorubicin, aiming at the confirmation of the release mechanism. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Everyday new smart Drug Delivery Systems (DDS) are fabricated seeking to improve cancer therapy as well as the patients’ life. The development of these systems is substantial in order to achieve more effective treatment and fewer side effects [1–12]. Conventional treatments, like chemotherapy and radiotherapy, have the disadvantage of killing normal cells and destroying neighboring tissues, causing toxicity. The role of DDS is to avoid these side effects by treating only cancer cells. Some of the characteristics that can be used for fabricating an intelligent DDS are: 1) temperature difference between cancer cells and their surrounding area, 2) vulnerability of cancer cells at 42–43 ◦ C in contrast to 45–46 ◦ C for normal cells 3) difference between intra- and extra-cellular pH with values for cancer cells around 6.0–4.5 and 6.8, respectively and 6.5 and 7.4, respectively for normal cells, and finally 4)

∗ Corresponding authors at: Patriarxou Grigoriou E & Neapoleos, Aghia Paraskevi, Athens, 15341, Greece. E-mail addresses: [email protected], [email protected] (E.K. Efthimiadou). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.colsurfb.2016.08.019 0927-7765/© 2016 Elsevier B.V. All rights reserved.

enzyme copiousness that creates reductive and oxidizing (redox) conditions. By combining all these factors [25], a smart nanocarrier can be fabricated aiming at specific targeting and drug release in a controlled manner. The idea of the aforementioned nanocarrier has been studied for the past few years and lots of ideas came to forefront. The thermo-, pH-, and redox sensitivities were used as a single property, or by combining two of the properties, for the fabrication of polymer nanocarriers [13–25]. These nanocarriers can be in the form of nanospheres, micelles, nanorods, nanoparticles, nanocontainers, etc. and are synthesized by monomers that have specific properties taking advantage of some unique characteristics of cancer cells. One of the many properties that are currently investigated and used for the fabrication of nanocarriers is thermo-sensitivity [26–29]. Hydroxy Propyl Methacrylamide (HPMA) and Dimethyl Amino Ethyl Methacrylate (DMAEMA) are monomers that exhibit thermo-sensitivity. Similar to thermosensitive monomers are, pH-sensitive monomers [30–32] like Acrylic Acid (AA) and redox-sensitive monomers [33–35] like N,N -(disulfanediylbis(ethane-2,1-diyl))bis(2-methylacrylamide). An ideal combination of monomers can result to a co-polymer that integrates thermo- and pH- sensitivity, or thermo- and redoxsensitivity, or any other combination of the above properties. A

96

C. Tapeinos et al. / Colloids and Surfaces B: Biointerfaces 148 (2016) 95–103

Scheme 1. The possible interacted groups between DOX and Q-NCs.

Scheme 2. The Disulfide segment degradation.

way to enhance the efficacy of the nanocarriers is to use magnetic nanoparticles either on the surface or in the interior of the nanocarriers. Magnetic nanoparticles, like Fe3 O4 (magnetite), have been used extensively in medicine for cancer therapy [36–40], as Magnetic Resonance Imaging (MRI) contrast agents or for local Hyperthermia. According to Habash et al.[41], hyperthermia is the state where the temperature of the body (general) or of the tissue (local or regional) is abnormally increased above 37 ◦ C for a certain period of time, aiming at the destruction of cancer cells. The influence of hyperthermia for human cancer cells has been stimulated by the consistent evidence that low pH exerts a major effect in sensitizing cultured cells to heat [24]. By using an alternating magnetic field (AMF) the temperature of the nanocarriers with the magnetic nanoparticles can be increased in a control manner to a desired temperature, adding this way one more capability to the smart DDS. Great progress has been also achieved in the design of nanocarriers, that are able to selectively carry radionuclides, in order to improve the outcome of cancer diagnosis and/or treatment [42,43]. In this report we quote the synthesis and characterization of a versatile DDS with pH-, thermo- and redox sensitivity that can be used in hyperthermia treatment. We loaded the drug Doxorubicin, as model drug, in the hollow Q-NCs and studied the release

behavior under external stimuli, such as pH, redox, temperature and hyperthermia (Schemes 1 and 2).

2. Materials and methods 2.1. Materials Acrylic Acid (AA) and Dimethyl Amino Ethyl Methacrylate (DMAEMA) were purchased from Sigma Aldrich and distilled before their use. Divinyl Benzene (DVB) and Poly(ethylene glycol) methacrylate (average Mn = 360) (PEG-360) were also purchased from Aldrich and used as received. Methyl Methacrylate (MMA) which was purchased from Merck was freshly distilled before its use and Potassium persulfate (KPS) was purchased from Panreac and used as received. Ethylene Glycol (EG) provided by Merck, Iron (II) Chloride tetrahydrate (FeCl2 × 4H2 O) provided by Riedel-de Haën, Potassium Nitrate (KNO3 ) provided by Acros and Hexamethylenetetramine (HETM) provided by Alfa Aesar were used as received. 95◦ ethanol was used as received. N,N -(disulfanediylbis(ethane-2,1-diyl))bis(2methylacrylamide) (Disulfide) was synthesized in our lab (See the synthetixc approach in Supl. material file). Doxorubicin·HCl was provided by Pharmacia & Upjohn and used as received. Phosphate

C. Tapeinos et al. / Colloids and Surfaces B: Biointerfaces 148 (2016) 95–103

97

Fig. 1. A) Transmission Electron Microscopy image of Q-NCs, B) bright field micrograph used for sulfur map and C) EF-TEM sulfur elemental map.

Buffer Saline (PBS) was used as a buffer solution for loading ant release study.

2.2. Equipment Scanning electron microscopy (SEM) and Transmission Electron Microscopy (TEM) images were obtained on an FEI Inspect microscope operating at 25 kV and a FEI CM20 microscope operating at 200 kV, equipped with a Gatan GIF200 Energy Filter utilized for EFTEM elemental mapping respectively. Fourier transform infrared (FT-IR) spectra were obtained on Perkin Elmer Spectrum 100 Spectrometer; the spectra were scanned over the range 4000–400 cm−1 . (FT-IR characterization is presented in Fig. S2, on Supl. file and the peaks are summarized in Table S3). Raman Spectra were obtained with a Renishaw, inVia Raman Microscope. Nuclear Magnetic Resonance (NMR) spectra were obtained by a Bruker Advance DRX-500 instrument at 500 MHz. X-Ray Diffraction data were obtained with a powder diffractometer (SIEMENS D-500 equipped with a CuKa lamp with wavelength 1.5418 A◦ , Siemens AG, Munich, Germany). Dynamic light scattering (DLS) measurements were performed on a Malvern Instruments Zetasizer Nano Series, with a multipurpose titrator. In the data presented in this study, each measurement represents the average value of 3 measurements, with 11–15 runs for each measurement. UV–vis absorption spectra in the wavelength range of 200–800 nm were obtained on a Jasco V-650 spectrometer. An ultrasonic bath was used for sonication (Elma Sonic, S 30H). A Vibrating Sample Magnetometry (VSM) model 155 with a Bell 640 Gaussmeter, source: Danfysik System 8000 (−2 to 2 T) was used for the magnetic measurements. Magnetic fluid hyperthermia was performed by a second magnetic-induction hyperthermia equipment, CELES MP 6 kW (Fives Celes, FR) operating at a frequency of 173 kHz and field strength of 23–27 kA/m with a coil

diameter of ∼100 mm and 3 loops, being continuously water cooled and a 2.4 kW Easyheat 0224 system (Easyheat® , Ameritherm Inc) equipped with a copper coil (diameter 40 mm, 3 loops). Temperature was recorded every second with an RF-immune fiber optic probe. 2.3. Synthesis 2.3.1. Synthesis of hollow PMMA@P(MMA-co-DMAEMA-co-DVB-co-PEG360-co-AA-co-DS) NCs (HNCs) The synthetic procedure of hollow nanospheres (Q-NCs) also depends on the emulsion polymerization described in the supplementary material. PMMA nanospheres were added with a mixture of water and ethanol in a 50 ml round flask. The mixture was sonicated in an ultrasonic bath for 30 min at 55 ◦ C and then it was agitated on a magnetic stirrer at 55 ◦ C, under nitrogen atmosphere. After 3.5 h of agitation, MMA, DMAEMA, DVB and PEG360 were slowly added and the mixture was agitated again for 1 h. Next, the temperature was raised at 80 ◦ C and 2 ml water solution of the initiator KPS was added. After 10 min, 1 ml ethanol solution of Disulfide was added dropwise. The reaction ended after 18 h and the product was collected by centrifugation (3 × 7000 rpm for 5 min each). In between centrifugations the polymeric Q-NCs were washed and dispersed in distilled water. The exact amount of all the reagents used is summarized in Table S2. (The core synthesis is summarized in Table S1). 2.3.2. Synthesis of hollow magnetic NCs (Q-NCs) The previously synthesized HNCs were dispersed in a mixture of ethylene glycol and distilled water, under stirring, for 5 min. The resulting mixture/dispersion was brought under nitrogen atmo-

98

C. Tapeinos et al. / Colloids and Surfaces B: Biointerfaces 148 (2016) 95–103

Fig. 2. Raman Spectra of a) PMMA@P(MMA-co-DMAEMA-co-DVB-co-PEG360-co-AA-co-DS) Q-NCs and b) Q-NCs.

sphere to be agitated for another 40 min, followed by the addition of 1 ml from each aqueous solution of FeCl2 ·4H2 O and of HETM. The reaction ended after 3 h and the product was collected by centrifugation (1 × 6000 rpm for 5 min and 2 × 5000 rpm for 5 min). Table S3 summarized the reagents amount.

2.3.3. Loading and release A typical procedure for the loading of the drug Doxorubicin in the fabricated Q-NCs is described below. Equal amounts of the Q-NCs and drug were treated under neutral conditions (pH = 7.4, PBS buffer solution) and the mixture was gently stirred for three days at room temperature in dark conditions. After that, the mixture was centrifuged three times (10000 × 5 min) and the isolated material was washed three times with water. The unloaded drug was estimated in the supernatant according to a DOX free standard curve in PBS [25]. The release study was performed in two different pH conditions, pH = 4.5 and pH = 7.4, and in combination with glutathione and hyperthermia treatment. In more detail, standard amount of drug loaded Q-NCs was re-suspended in the different conditions. Afterwards, 0.5 ml of the solution was removed in different time intervals and then centrifuged, in order to calculate the released drug in the supernatant. The DOX concentration was calculated according to our previously described work [43,44]. The release percentage experiment was carried out three times and the released percentage expressed as the average release ± SD.

3. Results 3.1. Electron microscopy (SEM & TEM) Scanning Electron Microscopy was used for the morphological characterization of PMMA@ P(MMA-co-DMAEMA-co-DVBco-PEG360-co-AA-co-DS) (Fig. S1A) and PMMA@ P(MMA-coDMAEMA-co-DVB-co-PEG360-co-AA-co-DS)@Fe3 O4 (Q-NCs) (See Supl. Info. Fig. S1). The size of the PMMA@P(MMA-co-DMAEMA-coDVB-co-PEG360-co-AA-co-DS) Q-NCs is in the range 380–450 nm. As can be seen in Fig. S1A, the surface roughness of some Q-NCs, indicated by the red arrows, is higher than the average, while small cavities are formed on the surface of other Q-NCs, indicated by the blue arrows. The high roughness can be attributed to uneven deposition of coating monomers resulting in a bigger and uneven shell. The small cavities on the surface are an indication of the void created during the coating procedure [25]. Fig. S1B depicts the Q-NCs after functionalization with the magnetic nanoparticles. The morphology of the Q-NCs is similar to Fig. S1B and the roughness is indicated again with the red arrows. Fig. 1A shows a bright field image of the Q-NCs. It is evident that they are spherical, hollow and that the magnetic nanoparticles are on the surface as well as the interior. Fig. 1C is the EF-TEM sulfur elemental map of the area imaged in the bright field micrograph of Fig. 1B. Since bright contrast corresponds to S, it is evident that there is a rather homogeneous distribution of the Disulfide on the surface of the Q-NCs.

C. Tapeinos et al. / Colloids and Surfaces B: Biointerfaces 148 (2016) 95–103

99

Fig. 3. A. Size and Zeta potential measurements of PMMA@P(MMA-co-DMAEMA-co-DVB-co-PEG360-co-AA-co-DS) Q-NCs vs. concentration, B. Size and Zeta potential measurements of PMMA@P(MMA-co-DMAEMA-co-DVB-co-PEG360-co-AA-co-DS) QNCs (0.025 mg/ml) vs. time, C) Size and Zeta potential measurements of PMMA@P(MMA-co-DMAEMA-co-DVB-co-PEG360-co-AA-co-DS) NCs (0.025 mg/ml) vs. temperature, D) Size and Zeta potential measurements of PMMA@P(MMA-coDMAEMA-co-DVB-co-PEG360-co-AA-co-DS) Q-NCs (0.025 mg/ml) vs. temperature (The experiment repeated three times and the size expressed as Average Size ± SD).

Electron Dispersive X-Ray spectorscopy (EDS) was used for the elemental analysis of the synthesized Q-NCs (See Sup. Info. Fig. S3). The stoichiometry of the Q-NCs is shown in Table S4. The existence of Nitrogen (N), Sulfur (S) and Iron (Fe) is an additional proof that the coating procedure and the functionalization with Magnetic Nanoparticles (MNPs), were successful. 3.2. Raman spectroscopy Raman spectra of PMMA@P(MMA-co-DMAEMA-co-DVB-coPEG360-co-AA-co-DS) before and after the surface modification with iron oxide particles can be observed in Fig. 2. The two spectra shown, are similar enough, except from the peaks at 664 and 677 cm−1 . These two bands can be attributed to A1 g (Fe O) symmetric stretching of Fe3 O4 [45–47] and to the C-S aliphatic vibration [50] of Disulfide respectively. In comparison with the PMMA@P(MMA-co-DMAEMA-co-DVB-co-PEG360-co-AAco-DS) spectrum, there is a small shift of the C-S vibration (640 cm−1 ) and this can be attributed to the influence of the Fe-O bond. 3.3. Dynamic light scattering (DLS) The hydrodynamic diameter and the colloidal stability of the Q-NCs were studied by using the dynamic light scattering technique. Distilled water and 5 mM Phosphate Buffer Saline (PBS) were used as dispersant mediums. The first diagram (Fig. 3A) depicts the relation between the hydrodynamic size and the zeta potential, in different sample concentrations in distilled water (pH ∼ 5.5). First of all, we can see that the minimum hydrodynamic size is

greater compared to the diameter measured with SEM and TEM and this is expected considering that our Q-NCs deviate from the perfect sphere model. However, the measured values are close to the values of SEM and TEM. As concentration increases the hydrodynamic size increases also, whereas the zeta potential remains almost stable. The increment in the hydrodynamic radius is related to the increment of the concentration and to zeta potential as well. At the specific range of zeta the colloidal system is unstable and tends to create flocculation and sometimes aggregates. Flocculation occurs because low value in zeta potential means lower repulsive forces. Hence, as concentration increases, the chances are higher for flocculation or aggregation and leading to higher measured diameter. The chosen concentration for the following measurements was 25 mg/L, based on Fig. 3A and on the low conductivity values that affect zeta potential measurements. Concentration measurements were followed by time dependence measurements. In Fig. 3B, the hydrodynamic diameter (Dh ) varies from 620 to 660 nm. These variations are regular due to the fact that zeta potential varies from −28 to −21 mV. These zeta values mean that the repulsive forces in the solution, normally competing with Van der Waals forces are not so strong, leading to a partial flocculation and deflocculation of the sample. In continuation, in order to study the thermo-sensitivity of our sample, a temperature trend measurement was performed. From 20 ◦ C to 35 ◦ C, Dh of the sample shows an increment from 600 nm to 650 nm that can be attributed to a small flocculation, but after 35 ◦ C exhibits a decrement to 530 nm at 50 ◦ C. This decrement can be ascribed to the thermo-sensitivity of DMAEMA which has an LCST (Lower critical solution temperature) in water, at 38 ◦ C to 40 ◦ C [49–51]. Above this temperature, the side chains turn

100

C. Tapeinos et al. / Colloids and Surfaces B: Biointerfaces 148 (2016) 95–103

from hydrophilic to hydrophobic leading to size reduction. When a thermo-sensitive polymer is copolymerized with hydrophobic monomers its LCST is decreased. In our case, the highest percentage of our monomers are hydrophobic and we can assume that LCST drops to 35 ◦ C. Therefore, the decrease in size can be attributed to the thermo-sensitive DMAEMA. According to Van’t Hoff equation when temperature increases, pH decreases. A pH reduction can lead to a zeta potential reduction and that’s the reason why zeta values become less negative as temperature increases. Furthermore, another fact that contributes to the zeta potential decrement is that the Q-NCs become more hydrophobic above 35 ◦ C. Finally, the increase in size above 50 ◦ C can be due to flocculation because zeta potential is around −18 mV and this means that the Van der Waals forces override the repulsive forces. Size and zeta measurements versus temperature were also performed in 5 mM PBS (Fig. 3D) in order to study the behavior of the Q-NCs in simulated conditions of the human body. The behavior of the Q-NCs in PBS solution was not the same as in water. To begin with, at the starting temperature of 20 ◦ C, Dh is 1400 nm and varies from 1350 nm to 1500 nm until 33.5 ◦ C. These variations, depicted in Fig. 8, can be attributed to flocculation and deflocculation of the sample, as a result of the reduction in zeta potential values, which stem from the addition of PBS. What should be emphasized is the fact that above 33.5 ◦ C there is a steep decrease in size to 850 nm. This decrease is close enough to the previously discussed diagram (Fig. 3C) strengthening our assumption that LCST of our copolymer is at around 35 ◦ C. The aforementioned decrease can be explained, as DMAEMA chains become less hydrophilic they tend to de-swell suspending any flocculation that has been created. Zeta potential plot doesn’t follow the same pattern as in Fig. 3C. At small temperatures zeta potential measurements have low values, as a result of the PBS addition. PBS increases the conductivity and ionic strength, but reduces the electric double layer of the nanocontainers. The thickness of the double layer depends upon the concentration of ions in solution and can be calculated from the ionic strength of the medium. The higher the ionic strength, the more compressed the double layer becomes, leading to less negative zeta potential values. As temperature increases zeta potential values become more negative and this phenomenon can be attributed to the increase of electrophoretic mobility. According to Henry equation electrophoretic mobility is proportional to Zeta potential, so as temperature increases, ions move faster, electrophoretic mobility increases and subsequently zeta potential’s negative values are increasing. The pH analysis of the Q-NCs at three different temperatures was carried out (Fig. 4) in order to study the behavior of them in slightly basic and acidic conditions, but also to study the behavior of two combining factors. Before pH titration, samples were regulated at pH = 8.5, so as for the repulsive forces to be strong enough in order to avoid flocculation. Zeta potential shows an anticipated behavior in all temperatures, by starting with negative values (−36 mV at 25 ◦ C, −34 mV at 37 ◦ C and −32 mV at 42 ◦ C) at pH = 8.5 and shifting to less negative as the carboxyl groups of Poly (Acrylic Acid) starts to protonate. As pH continues to become more acidic, zeta potential gets positive values. The effect of temperature is obvious even at high pH. As temperature increases, zeta potential decreases and this result can be attributed to the pH variations (slight reductions), according to Van’t Hoff. Isoelectric point (IP) is also affected by temperature changes and while temperature increases IP shifts to higher pH (3.3 for 25 ◦ C and 3.8 for 37 ◦ C and 42 ◦ C). Size dependence is also obvious in the figure below (Fig. 4). At the temperature of 25 ◦ C and at pH = 8.5, hydrodynamic diameter is 670 nm. Dh remains relatively steady until pH ∼ 5, and then increases. Knowing that pKa of Poly (acrylic acid) is at about 4.5 and pKa of Poly (DMAEMA) is at 7.5 it can be assumed that the

Fig. 4. Size and Zeta potential measurements of PMMA@P(MMA-co-DMAEMA-coDVB-co-PEG360-co-AA-co-DS) Q-NCs (0.025 mg/ml) for three different temperatures vs. pH (The experiment repeated three times and the size expressed as Average Size ± SD).

pKa of our Q-NCs is at pH ∼ 5. Below this pH value, the protonation percentage increases leading to lower zeta potential values and to higher hydrodynamic diameters. The high values of the Q-NC’s size (>800 nm) can be attributed to flocculants that are created as pH decreases. The phenomenon is more intense at higher temperatures and this leads to the inference that pH and temperature act synergistic. 3.4. Vibrating sample magnetometry The hysteresis loop in Fig. 5 is characteristic for a ferromagnetic material. The low coercive field, the size of the magnetic nanoparticles (MNPs) as calculated from XRD pattern (See Supl. Info. Fig. S5) and the saturation magnetization (Ms ) value are an indication of super-paramagnetism. As it is shown, Ms is lower than the bulk magnetite but this can be attributed to the low amount of magnetite nanoparticles compared to the total polymeric amount. A small negative slope between 0.5 and 2 T can be attributed to the diamagnetic content of the polymer. These results were also supported by TGA and XRD analysis where the Fe3 O4 percentage (%) was calculated (See Supl. Info. Fig. S4 and S5). 3.5. Loading and release studies In order to study the loading efficiency of the multi-responsive Q-NCs as well as their therapeutic efficacy, loading and release studies have been performed as described below. The loading capacity (LC%) and encapsulation efficiency (EE%) of the Q-NCs were calculated according to the standard curve. The exact mass of the drug in Q-NCs was found to be 486 ± 2 ␮g/mg of the polymer, the LC% = 94.7 and EE% = 94.7. Amongst the cancer research scientific community, it has been well established that tumor cells present acidic pH, increased temperature and high redox microenvironment in contrary to healthy and normal cells, due to rapid proliferation. The aforesaid differences can be stimulated in order to study the release efficacy of the Q-NCs. [23–25,44,45,52] Generally the drug release behavior depends on a number of factors that include pH, thermo and redox conditions as well as the applied alternate magnetic field. The investigation of the thermo depended drug release behavior was performed at two different temperatures of 25 ◦ C and 42 ◦ C, in combination with the pH depended drug release behavior study under slightly basic (pH = 7.4) and acidic (pH = 4.5) conditions. From Fig. 6A it is observed that the ideal combination of temperature and

C. Tapeinos et al. / Colloids and Surfaces B: Biointerfaces 148 (2016) 95–103

101

Fig. 5. Hysteresis Loop of Q-NCs derived from Vibrating Sample Magnetometry.

Fig. 6. A. Release study of Q-NCs under pH & conventional heating conditions, B. under pH & Hyperthermia conditions (n = 3, Released percentage expressed as DOX (%) ± SD).

pH, for the maximum drug release, is at 42 ◦ C in acidic pH (blue line), where notably the absolute drug release increase, from 8 h to 38 h, was about 20%. In more detail, the release percentage was about 55% for the first 8 h when it started reaching a plateau at 75% and remained there until 38 h later. Moreover, at the same pH but at 25 ◦ C (black line), the loaded Q-NCs present the same profile of a sustained release in acidic pH as a function of time, but at lower percentages, because for the first 8 h 35% of the drug was released and then the amount of the released drug reaches a plateau at about 50% for 50 h. On the contrary, at pH = 7.4 the drug release, at either 25 ◦ C or 42 ◦ C, was significantly lower. Briefly, only the 8% and 15%

of the loaded drug is released after 48 h treatment at 25 ◦ C and 45 ◦ C at pH = 7.4, respectively These results definitely support the pH and temperature sensitivity of the fabricated Q-NCs. The above mentioned release behavior can be attributed to the acidic acid (AA) and thermo/pH (DMAEMA) sensitive segments of the shell copolymer. Furthermore, the release of Doxorubicin from the Q-NCs was found to be significantly enhanced after heating either conventionally in a water bath or by using AMF- in both acidic and slightly basic pH environment, compared with control experiments that were carried out at 25 ◦ C (Hyperthermia measurements can be found in the Supl. Info. Figs. S7 & S8).

102

C. Tapeinos et al. / Colloids and Surfaces B: Biointerfaces 148 (2016) 95–103

Scheme 3. The developing equilibriums between the protonated and deprotonated form of Q-NCs.

After 1 h hyperthermia treatment at 42 ◦ C, in both pH = 4.5 and 7.4, the drug release percentage was 8% and 10% higher than the control experiment at 25 ◦ C, respectively. Nonetheless the release percentage remains also low (Fig. 6B). AMF heating induces doxorubicin release in good agreement with the one observed under conventional heating in the same time and at the same pH values. Moreover, considering the fact that, in the presence of GSH the disulfide segment of the copolymer can break, according to Scheme 3, where it can be seen that the nucleophilic attack of the thiol group of GSH- leads to the Q-NCs’ degradation, which results in an increase of the drug’s release rate. In order to investigate the drug release behavior, at different concentrations of GSH, a preliminary experiment was performed. The release percentage of the doxorubicin loaded Q-NCs was monitored, after treatment with various GSH concentrations (5 mM, 10 mM 15 mM and 20 mM), for 24 h at 42 ◦ C (Supl. Info. Fig. S6), and the ideal concentration was found to be 20 mM. Subsequently, the drug release percentage was evaluated at pH = 7.4, in the presence and absence of glutathione (GSH), after heating with either AMF and or conventional heating (water bath). The enhancement in drug release, in the presence of a magnetic field, is a result of the magnetic nanoparticle’s heating capacity, which are doped to the Q-NCs in combination with the presence of GSH. The Doxorubicin release amount reached at 65% in contrast with conventional heating in which the percentage reached 20%, under the above mentioned conditions, as seen in Fig. 7. What is more, the effect of each factor and their ideal combination of them can also be observed. From the results, it is obvious that the highest drug release percentage occurs when at least two factors act. The functional carboxyl groups of Q-NCs can form strong ionic, electrostatic and hydrogen bonding interactions, with drug candidates, due to their ability to work as Lewis acids (hydrogen donors). Based on the ability of the Q-NCs to be negatively charged, it is possible to bind to the cationic drugs formatting a drug-polymer complex. Carboxylic acid groups of the PAA segment of Q-NCs have a pKa close to the range from 4 to 4.5, so the carboxyl groups are negatively charged above the pH 4.5. Conversely the primary NH2 group of the DOX is positively charged, because of its pKa = 8.4 (pHpzc DOX = 8.4) at 25 ◦ C. As a result, the drug at pH = 7.4, where the loading takes place, remains positively charged in contrast with carboxylic groups which are negative charged, allowing this way maximum electrostatic interactions. At pH = 4.5, the drug

Fig. 7. Release study of Q-NCs combining GSH, conventional heating & Hyperthermia conditions at stable pH (=7.4) (n = 3, Released percentage expressed as DOX (%) ± SD).

– which remains positively charged- is no longer attracted from the Q-NCs, due to neutrally charged carboxylic groups ( COOH). At pH 4.5 a burst release of DOX was observed, for the first 8 h (32% at 25 ◦ C and 52% at 42 ◦ C) followed by a slow release (50% at 25 ◦ C and 72% at 42 ◦ C) over three days. The hyperthermia effect on the drug release percentage can be attributed to two primary mechanisms arising during nanoparticle heating in an external magnetic field. The first is Neel relaxation, where the energy is dissipated by reorientation of the magnetization and the second is Brownian relaxation, where the energy is dissipated due to frictional losses as a result of the rotation of the particles in the liquid. However, the exact mechanism about the hyperthermia effect in drug release is hitherto unclear, besides the observation that the AMF increases the drug release percentage. Additionally, the release percentage was further enhanced under treatment with GSH of different concentrations at pH = 7.4 (Fig. 8). The combination of thermo or hyperthermia in the presence of appropriate GSH concentration led to 65% drug release after 30 min treatment. As mentioned earlier, tumors exhibit a substantially lower extracellular pH than normal tissues, whereas the intracellular pH of both tissues is similar while present higher temperature

C. Tapeinos et al. / Colloids and Surfaces B: Biointerfaces 148 (2016) 95–103 [3] [4] [5] [6] [7] [8] [9] [10]

[11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] Fig. 8. Release study of Q-NCs under GSH & conventional heating conditions at pH = 7.4. (n = 3, Released percentage expressed as DOX (%) ± SD).

and redox conditions in relation to normal tissues. [24] It is well known also, that in cancer cells the glutathione concentration is ten-fold higher than the concentration in normal cells. Based on the above results, it was observed that when the sample was treated with GSH under neutral conditions the drug was released due to disulfide bond breakage. From Fig. 8. It can be seen that the drug release percentage increases when also the temperature increases. Consequently, the loaded Q-NCs will release the drug when at least two factors act together. After the drug release experiment the Q-NCs were characterized by SEM to determine the structure affection by glutathione due to the fact that disulfide bonds break leading to the structure collapsing (See Supl. Info. Figs. S5 & S6). 4. Conclusions This work demonstrates the Q-NC’s use as smart drug delivery vehicles in order to fight cancer. Here we present the fabrication and full characterization of a versatile multifunctional polymeric NC which is triggered when at least two factors, such as pH, redox, thermo and alternate magnetic field, act together. The in vitro specific behavior of the loaded Q-NCs with the anticancer drug Doxorubicin, was also evaluated. From the experimental results in total, it was determined that under the desired stimuli the drug can be released from the Q-NCs. It is noteworthy, that each stimulus separately causes a small amount of drug to be released, but only when at least two stimuli act together the drug release can be significantly enhanced. This synergistic result indicates the presence of an ON/OFF switch system in the tumor local area.

[22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46]

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.colsurfb.2016.08. 019.

[47] [49] [50] [51]

References [1] A. Kumari, S.K. Yadav, S.C. Yadav, Colloids Surf. B 75 (2010) 1–18. [2] M. Shen, Y. Huang, L. Han, J. Qin, X. Fang, J. Wang, V.C. Yang, J. Controlled Release 161 (2012) 884–892.

[52]

103

M.J. Lawrence, G.D. Rees, Adv. Drug Deliv. Rev. 64 (2012) 175–193. K. Kataoka, A. Harada, Y. Nagasaki, Adv. Drug Deliv. Rev. 64 (2012) 37–48. L. Brannon-Peppas, J.O. Blanchette, Adv. Drug Deliv. Rev. 64 (2012) 206–212. J. Panyam, V. Labhasetwar, Adv. Drug Deliv. Rev. 64 (2012) 61–71. P. Parhi, C. Mohanty, S.K. Sahoo, Drug Discov. Today 17 (2016) 1044–1052. I. Brigger, C. Dubernet, P. Couvreur, Adv. Drug Deliv. Rev. 64 (2012) 24–36. C. Wang, L. Cheng, Z. Liu, Biomaterials 32 (2011) 1110–1120. C. Wong, T. Stylianopoulos, J. Cui, J. Martin, V.P. Chauhan, W. Jiang, Z. Popovic, R.K. Jain, M.G. Bawendi, D. Fukumura, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 2426–2431. O. Veiseh, J.W. Gunn, M. Zhang, Adv. Drug Deliv. Rev. 62 (2010) 284–304. J. Shi, A.R. Votruba, O.C. Farokhzad, R. Langer, Nano Lett. 10 (2010) 3223–3230. J.L. Zhang, R.S. Srivastava, R.D. Misra, Langmuir 23 (2007) 6342–6351. V.P. Torchilin, AAPS J. 9 (2007) E128–E147. A. Chan, R.P. Orme, R.A. Fricker, P. Roach, Adv. Drug Deliv. Rev. 65 (2012) 497–514. L. Jabr-Milane, L. van Vlerken, H. Devalapally, D. Shenoy, S. Komareddy, M. Bhavsar, M. Amiji, J. Control. Release 130 (2008) 121–128. P. Bawa, V. Pillay, Y.E. Choonara, L.C. du Toit, Biomed. Mater. 4 (2009) 1–15. Q. Yuan, R. Venkatasubramanian, S. Hein, R.D.K. Misra, Acta Biomater. 4 (2008) 1024–1037. E.J. Kim, S.H. Cho, S.H. Yuk, Biomaterials 22 (2001) 2495–2499. Y. Morishima, Angew. Chem. Int. Ed. 46 (2007) 1370–1372. M. Delcea, H. Möhwald, A.G. Skirtach, Adv. Drug Deliv. Rev. 63 (2011) 730–747. K.S. Soppimath, T.M. Aminabhavi, A.M. Dave, S.G. Kumbar, W.E. Rudzinski, Drug Dev. Ind. Pharm. 28 (2002) 957–974. C. Lu, B. Mu, P. Liu, Colloids Surf. B 83 (2011) 254–259. C. Tapeinos, E.K. Efthimiadou, N. Boukos, C.A. Charitidis, M. Koklioti, G. Kordas, J. Mater. Chem. B 1 (2013) 194–203. I.F. Tannock, D. Rotin, Cancer Res. 49 (1989) 4373–4384. S. Ganta, H. Devalapally, A. Shahiwala, M. Amiji, J. Controlled Release 126 (2008) 187–204. Z. Feng, H. Guihua, D. Shaojun, L. Rong, W. Changchun, Colloid. Polym. Sci. 290 (2012) 1341–1346. F. Xu, T.-T. Yan, Y.-L. Luo, J. Bioact. Compat. Polym. 28 (2013) 66–85. B. Trzebicka, B. Robak, R. Trzcinska, D. Szweda, P. Suder, J. Silberring, A. Dworak, J. Eur. Polym. 49 (2013) 499–509. H. Hu, H. Wang, Q. Du, Soft Matter 8 (2012) 6816–6822. C. Peniche, M. Fernández, A. Gallardo, A. López-Bravo, J.S. Román, Macromol. Biosci. 3 (2003) 540–545. X. Yang, L. Chen, B. Huang, F. Bai, X. Yang, Polymer 50 (2009) 3556–3563. H. Cho, J. Bae, V.K. Garripelli, J.M. Anderson, H.-W. Jun, S. Jo, Chem. Commun. 48 (2012) 6043–6045. R.L. McCarley, Annu. Rev. Anal. Chem. 5 (2012) 391–411. X. Sui, X. Feng, M. Hempenius, G.J. Vancso, J. Mater. Chem. B. 1 (2013) 1658–1672. O. Salata, J. Nanobiotechnol. 2 (2004) 3. E. Duguet, S. Vasseur, S. Mornet, J.-M. Devoisselle, Nanomedicine (Lond) 1 (2006) 157–168. Z. Zhao, D. Huang, Z. Yin, X. Chi, X. Wang, J. Gao, J. Mater. Chem. 22 (2012) 15717–15725. A.P. Khandhar, R.M. Ferguson, K.M. Krishnan, Jpn. J. Appl. Phys. 109 (2011), 7B103–7B310. R.M. Ferguson, K.R. Minard, A.P. Khandhar, K.M. Krishnan, Med. Phys. 38 (2011) 1619–1626. R.W.Y. Habash, R. Bansal, D. Krewski, H.T. Alhafid, Crit. Rev. Biomed. Eng. 34 (2006) 491–542. M. Hamoudeh, M.A. Kamleh, R. Diab, H. Fessi, Adv. Drug Deliv. Rev. 60 (2008) 1329–1346. A.-F. Metaxa, E.K. Efthimiadou, N. Boukos, G. Kordas, J. Colloid Interface Sci. 384 (2012) 198–206. E.K. Efthimiadou, L.-A. Tziveleka, P. Bilalis, G. Kordas, Int. J. Pharm. 428 (2012) 134–142. D.L.A. de Faria, S. Venâncio Silva, M.T. de Oliveira, J. Raman Spectrosc. 28 (1997) 873–878. G. Socrates, Infrared and Raman Characteristic Group Frequencies: Tables and Charts, third edition, Wiley, 2001. G. Burillo, E. Bucio, E. Arenas, G.P. Lopez, Macromol. Mater. Eng. 292 (2007) 214–219. L.H. Reddy, R.K. Sharma, K. Chuttani, A.K. Mishra, R.R. Murthy, AAPS J. 6 (2004) 55–64. R. Madru, P. Kjellman, F. Olsson, K. Wingardh, C. Ingvar, F. Stahlberg, J. Olsrud, J. Latt, S. Fredriksson, L. Knutsson, S.E. Strand, J. Nucl. Med. 53 (2012) 459–463. Y.-F. Wang, C.-M. Fu, M.-H. Chuang, T.-M. Cham, M.-I. Chung, J. Nanomater. 2011 (2011) 1–5. H. Thakkar, R.K. Sharma, A.K. Mishra, K. Chuttani, R.S. Murthy, J. Pharm. Pharmacol. 56 (2004) 1091–1099.