Eduardo Guisasola†, Alejandro Baeza†, Marina Talelli†, Daniel Arcos†, María Moros‡, Jesús M. de la Fuente§, and María Vallet-Regí†* †Depto. Química Inorgánica y Bioinorgánica. UCM. Instituto de Investigación Sanitaria Hospital 12 de Octubre i+12, Centro de Investigación Biomédica en Red de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Madrid, Spain. ‡Instituto de Nanociencia de Aragon, Universidad de Zaragoza, C/Mariano Esquillor s/n, Zaragoza, Spain. §Instituto de Ciencia de Materiales de Aragón, CSIC-Universidad de Zaragoza, C/ Pedro Cerbuna 12, Zaragoza, Spain. KEYWORDS hot spot, stimuli-responsive, mesoporous silica nanoparticles, drug delivery, magnetically triggered. ABSTRACT: Magnetically triggered drug delivery nanodevices have attracted great attention in nanomedicine, as they can feature as smart carriers releasing their payload at clinician’s will. The key principle of these devices is based on the properties of magnetic cores to generate thermal energy in the presence of an alternating magnetic field. Then, the tem- perature increase triggers the drug release. Despite this potential, the rapid heat dissipation in living tissues is a serious hindrance for their clinical application. It is hypothesized that magnetic cores could act as hot spots, this is, produce enough heat to trigger the release without the necessity to increase the global temperature. Herein, a nanocarrier has been designed to respond when the temperature reaches the 43ºC. This material has been able to release its payload un- der an alternating magnetic field without the need of increasing the global temperature of the environment, proving the efficacy of the hot spot mechanism in magnetic-responsive drug delivery devices. INDRODUCTION 1 The design of new stimuli-responsive nanodevices is an 2 ever-evolving research field for the development of effi-3 cient drug delivery systems, mainly for those addressed to 4 the treatment of malignant diseases1. In the case of porous 5 inorganic materials, their performance relays on the 6 open-close response of nanogates commonly assembled 7 at the pore entrance, which act as gatekeepers. In this 8 sense, silica mesoporous nanoparticles have widely 9 demonstrated to be an excellent support for stimuli-10 responsive purposes.2,3 Moreover, there is a range of sizes 11 in which they are able to extend the blood circulation 12 time. The consequence is an enhanced accumulation in 13 tumor areas, through the enhanced permeation and re-14 tention effect4 (EPR), allowing for more efficient and 15 accurate treatments with fewer side effects.5 After accu-16 mulation in the desired area, a stimuli responsive release 17 is often desirable. For that purpose many stimuli have 18 been used to trigger the cargo release such as pH6,7, light8, 19 redox9 and others10,11. Among them, significant efforts 20 have been made to develop magnetic-responsive 21 nanodevices, which respond to non-invasive and highly 22 penetrating alternating magnetic fields (AMF), inducing 23 local heating by means of the magnetic losses of the na-24 noparticles.12 Promising candidates in this field are mag-25 netic mesoporous silica nanoparticles (MMSNPs) coated 26 with thermosensitive polymers as a stimuli-responsive 27 gating system.13–15 However, magnetically induced heating 28 systems undergo the serious inconvenient of their ther-29 mal dissipation by the blood stream. Actually, their 3 0 stimuli-responsive behavior strongly depends on the envi-3 1 ronmental heating, which is difficult to maintain while 3 2 the blood irrigation increases.16,17 3 3 Recently, the local heating developed by magnet-3 4 ic nanoparticles have been tested using a DNA thermal 3 5 probe18 or an azo-functionalized polymer by thermal de-3 6 composition19. These works have demonstrated that the 3 7 temperature reached by the magnetic nanoparticles is 3 8 significantly higher than that measured in the surround-3 9 ing media. Recently, Zink and co-workers20 have also 40 demonstrated that it is possible to reach an effective heat-41 ing in silica nanoparticles surface, in response to an alter-42 nating magnetic field due to the embedded magnetite 43 nanoparticles. These findings have opened the possibility 44 for designing new stimuli-responsive devices that works 45 on the base of the hot spot effect. In other words, these 46 systems would magnetically trigger the drug release, 47 without the obligation of supplying high magnetic ener-48 gies commonly required for heating the environment, 1 thus overcoming the blood-mediated thermal dissipation. 2 For this goal, the mesoporous silica surface was coated 3 with an engineered thermosensitive random copolymer 4 which exhibits a linear (soluble) to globular (insoluble) 5 transition at 40-43ºC. Therefore in physiological tempera-6 ture, the polymer chains are extended and soluble, block-7 ing the release, while above the transition temperature 8 the polymer shrinks, allowing the cargo release. The arch 9 stone for this performance is the hot spots formed by the 10 superparamagnetic iron oxide nanoparticles (SPION) that 11 are embedded in the mesoporous silica matrix, in re-12 sponse to an alternate magnetic field. This hot spot effect 13 is shown to be capable to reach the polymer coating that 14 blocks the mesopores, thus provoking the shrinkage of 15 the polymer network and the release of its cargo. 16 The colloidal stability of nanocarriers is crucial 17 for biomedical applications, especially in the case of in-18 travenous administration21. Mesoporous silica nanoparti-19 cles have been widely studied as drug delivery vehicles 20 but they often suffer from aggregation in the blood-21 stream, that could be solved by polymer attachment on 22 the silica surface22. Coating of nanoparticles with 23 poly(ethylene glycol) or other hydrophilic polymers is 24 known to improve their colloidal stability in physiological 25 conditions and increase their circulation times23. There-26 fore, we have developed a synthetic procedure of magnet-27 ic mesoporous silica nanoparticles based on the following 28 key steps (Scheme 1). 29 3 0 Scheme 1. Synthesis path of 3 1 MMSNP@P(NIPAM/NHMA) device. 3 2 EXPERIMENTAL SECTION 3 3 Reagents. All chemicals were used without further pu-3 4 rification. Tetraethyl orthosilicate (TEOS, 98 %), n-3 5 Cetyltrimethylammonium bromide (CTAB, 99%), 3-3 6 [Tris(trimethylsiloxy)silyl]propyl methacrylate (TMSPMA, 3 7 98 %), N-Isopropylacrylamide (NIPAM, ≥99 %), N-3 8 (Hydroxymethyl)acrylamide solution (NHMA, 48 wt. % in 3 9 H2O), N,N′-Methylenebis(acrylamide) (MBA, 99 %) Oleic 40 acid (OA, ≥99 %), 4,4′-Azobis(4-cyanovaleric acid) 41 (ABCVA, ≥98.0%), fluorescein sodium salt, iron (II) chlo-42 ride tetrahydrate (FeCl2·4H2O, 99 %) and iron (III) chlo-43 ride hexahydrate (FeCl3·6H2O, >99%) were obtained from 44 Sigma Aldrich. [Hydroxy(polyethyleneoxy)propyl] tri-45 ethoxysilane, (PEG-Si, MW = 575-750 g/mol, 50 % in eth-46 anol) was purchased from Gelest. Ammonium nitrate 47 (NH4NO3, 99.9%), ammonium hydroxide (NH4OH, 28-30 48 wt % as NH3), chloroform (CHCl3, 99.8%), sodium hy-49 droxide (NaOH, ≥98 %), absolute ethanol were purchased 5 0 from Panreac. Ultrapure water was generated using a 5 1 Millipore Milli-Q system with a Milli-pak filter of 0.22 μm 5 2 pore size and used for all the preparation of aqueous solu-5 3 tions. 5 4 Characterization Techniques. Fourier transform in-5 5 frared spectroscopy (FTIR) in a Thermo Nicolet nexus 5 6 equipped with a Goldengate attenuated total reflectance 5 7 device. The textural properties of the materials were de-5 8 termined by nitrogen sorption porosimetry by using a 5 9 Micromeritics ASAP 2020. To perform the N2 measure-6 0 ments, the samples were previously degassed under vacu-6 1 um for 24 h at room temperature. Thermogravimetry 6 2 analysis (TGA) were performed in a Perkin Elmer Pyris 6 3 Diamond TG/DTA analyzer, with 5 °C min −1 heating 6 4 ramps, from room temperature to 600 °C. The hydrody-6 5 namic size of mesoporous and oleic acid iron oxide nano-6 6 particles were measured by means of a Zetasizer Nano ZS 6 7 (Malvern Instruments) equipped with a 633 nm “red” 6 8 laser. Transmission electron microscopy (TEM) was car-6 9 ried out with a JEOL JEM 2100 instruments operated at 7 0 200 kV, equipped with a CCD camera (KeenView Cam-7 1 era). Sample preparation was performed by dispersing in 7 2 distilled water and subsequent deposition onto carbon-7 3 coated copper grids. A solution of 1% of phosphotungstic 7 4 acid (PTA) pH 7.0 was employed as staining agent in 7 5 order to visualize the polymer coating attached on the 7 6 mesoporous surface. Scanning electron microscopy (SEM) 7 7 analyses were made on a JEOL 7600-LINK AN10000 mi-7 8 croscope (Electron Microscopy Centre, UCM) using a 7 9 graphite sample holder without any treatment. Liquid 1H-8 0 NMR experiments were made in a Bruker AV 250MHz. 8 1 UV-Vis spectrometry was used to determine the LCST of 8 2 linear polymers by means of a Biotek Synergy 4 device. 8 3 Iron quantification for specific absorption rate (SAR) 8 4 measurements at 480 nm absorbance was carried out on a 8 5 Thermo Scientific Multiskan GO UV/Vis microplate spec-8 6 trophotometer. A calibration curve was performed follow-8 7 ing the same procedure using iron standard solution 8 8 (Acros Organics) as reference. DC magnetic field: mag-8 9 netic parameters were determined by means of a vibrat-9 0 ing sample magnetometer (VSM, Instituto de Sistemas 9 1 Optoelectrónicos y Microtecnología, Universidad Politéc-9 2 nica de Madrid, Spain). Measurements were carried out at 9 3 room temperature and applying a maximum DC field of 9 4 5000G. The AMF assays were performed on a DM100 9 5 system (nanoScale Biomagnetics) in the frequency range 9 6 from 424 kHz to 838 kHz and magnetic fields of 20.05 to 9 7 23.87 kA·m-1. 9 8 Calculation Procedures : The surface area was deter-9 9 mined using the Brunauer-Emmett-Teller (BET) method 10 0 and the pore volume, Vpore (cm3 g−1), was estimated from 10 1 the amount of N2 adsorbed at a relative pressure around 10 2 0.99. The pore size distribution between 0.5 nm and 40 10 3 nm was calculated from the desorption branch of the 10 4 isotherm by means of the Barrett-Joyner-Halenda (BJH) 1 method. The mesopore size, Øpore (nm), was determined 2 from the maximum of the pore size distribution curve. 3 The SAR calculations were performed by DM100 system 4 software (nanoScale Biomagnetics). The mole percentage 5 of NHMA (fNHMA) in the synthesized random copolymers 6 was determined using 1H NMR analysis using equation 1. 7 I4.66 ppm and I3.87 ppm are the integrals of the protons at 4.66 8 ppm and 3.87 ppm respectively. 9 𝑓𝑁𝐻𝑀𝐴(%) = 𝐼4.66 𝑝𝑝𝑚 2⁄ 𝐼3.87 𝑝𝑝𝑚+𝐼4.66 𝑝𝑝𝑚 2⁄ × 100 (1) 10 Preparation of Hydrophobic Magnetite (Fe3O4) 11 NPs. Hydrophobic magnetite NPs were synthesized by 12 one-pot chemical coprecipitation method. Deionized 13 water was purged with nitrogen gas for 10 min. Then, 4.80 14 g of FeCl3•6H2O, 2.00 g FeCl2•4H2O, and 0.85 mL oleic 15 acid were added to 30 mL of deionized water under nitro-16 gen atmosphere with vigorous stirring. The mixture solu-17 tion was heated to 90 ºC. Then, 20 mL of ammonium 18 hydroxide (14 wt %) was added rapidly to the solution, 19 and it immediately turned black. The reaction was kept at 20 90 ºC for 2.5 h and then allowed to cool to room tempera-21 ture. The black precipitate was collected by magnetic 22 decantation and resuspended in chloroform with an end 23 concentration of 32.8 mg·mL-1 oleic acid-capped Fe3O4. 24 Preparation of Mesoporous Magnetic Silica Nano-25 particles (MMSNPs). MMSNPs were prepared in a 50 mL 26 round-bottom flask, adding 582 mg of CTAB as a phase 27 transfer agent and structure-directing agent for silica 28 condensation that were dissolved in 10 mL of H2O (mQ). 29 Then, the mixture was mechanically stirred in an ultra-3 0 sound bath during the addition of 26.4 mg OA-Fe3O4 in 3 1 CHCl3 (0.04 mL·min-1 rate) until the complete removal of 3 2 the organic solvent. The aqueous suspension was added 3 3 through a 0.2 µm cellulose filter to an 86 mL NaOH 3 4 (0.016M) solution and stirred at 600rpm. When the sus-3 5 pension was stabilized at 45 ºC, a mixture of 1.2 mL of 3 6 EtOH and 1 mL of TEOS was added dropwise (0.25 3 7 mL·min-1 rate). Once the TEOS addition was finished, 3 8 130µL of PEG-Si were added, and the suspension was 3 9 stirred for 2h. The reaction mixture was washed three 40 times by centrifugation with 50 mL of H2O, and then two 41 more times with 50 mL of EtOH. The brown solid ob-42 tained was suspended in 200 mL of EtOH (99.5%). Then, 43 0.5 mL of MPS were added dropwise and the mix was kept 44 stirring at 40 ºC during 16h. Before the washing step with 45 absolute EtOH, the surfactant template was removed by 46 ion exchange using 175mL of 10 g·L-1 NH4NO3 in EtOH 47 (95%) extracting solution at 65ºC overnight. The brown 48 suspension was then centrifuged (15000 rpm, 30 min) and 49 washed three times with 50 mL of EtOH to be dried under 5 0 vacuum overnight. 5 1 Linear Polymers Synthesis. In a typical synthesis for a 5 2 polymer with a 90:10 NIPAM to NHMA ratio, 200mg (1.76 5 3 mmol) of NIPAM and 44.8 µL (0.19 mmol) of NHMA were 5 4 placed in vial A and 12.7 mg of initiator 4,4′-Azobis(4-5 5 cyanovaleric acid) (ABCVA) were placed in vial B. The 5 6 monomer mixture and the initiator were purged by nitro-5 7 gen flow prior to the addition of 2.5 mL and 1 mL of dry 5 8 DMF respectively to each vial. The solutions were bub-5 9 bled with N2 for 15 min. Vial A was stirred in a heated 6 0 oilbath at 80 ºC for 2 min before the fast addition of 100µL 6 1 of vial B solution (1.2 mg, 400:1 monomer to initiator ra-6 2 tio) and allowed to stir for 16 h. A maximum 2mL of the 6 3 reaction mixture were added dropwise to 45 mL of EtO2 6 4 in a centrifugation tube obtaining a white precipitate. The 6 5 precipitate was washed three times with 45 mL of diethy-6 6 lether and dried at ambient temperature. The white solid 6 7 was dissolved in H2O and lyophilized. 6 8 MMSNPs Polymer Coating. In a 100 mL three-neck 6 9 round-bottom flask, 150.9 mg (1.33 mmol) of NIPAM, 12 7 0 mg of MBA (0.078mmol), 49.4 µL of NHMA (0.148 mmol), 7 1 3.6 mg of CTAB and 5 mg of Na2CO3 were added to 45 mL 7 2 of water (mQ). The solution was stirred under N2 bub-7 3 bling at 70 ºC for 30 min to remove oxygen. Then, the 7 4 solution was kept under N2 and 50 mg of MMSNPs redis-7 5 persed in 5 mL of EtOH (99.5%) were added to the mon-7 6 omer solution and strirred for 15 min more. To initiate the 7 7 monomer polymerization 0.2 mL of a 25 mg·ml-1 APS 7 8 solution in H2O (mQ) previously deoxygenated were 7 9 added to the reaction mixture. 20 min before the initiator 8 0 addition appears a brown solid precipitate and the reac-8 1 tion mixture was allowed to cool down to room tempera-8 2 ture and kept at that temperature for 6 h. The mixture 8 3 was centrifuged and washed three times with H2O to 8 4 remove the unreacted monomers and be dried under 8 5 vacuum overnight. 8 6 Fluorescein Release Experiments. Hybrid 8 7 MMSN@P(NIPAM/NHMA) nanoparticles were loaded 8 8 with a 20 mg·mL-1 fluorescein sodium salt solution in PBS 8 9 (1x) at 50 ºC for 16 h. Then, the hybrid material was 9 0 washed with PBS (1x) until no fluorescence was observed. 9 1 Fluorescein release experiments were carried out with 10 9 2 mg of fluorescein loaded MMSN@P(NIPAM/NHMA) 9 3 dispersed in 1 mL of PBS (1x) and divided in two 0.5 mL 9 4 aliquots. To determine the polymer VPTT attached to the 9 5 silica surface one aliquot was submitted to a heating ramp 9 6 in an incubator to the three target temperatures (37 ºC, 40 9 7 ºC and 43 ºC). The second aliquot was carried to AC mag-9 8 netic inductor to apply an 838 KHz and 20.05 kA·m-1 al-9 9 ternating magnetic field. The temperature increase devel-10 0 oped by the nanoparticles was recorded with a fiber-optic 10 1 probe. After the dispersion reached the target tempera-10 2 ture the samples were cooled down, collected by centrifu-10 3 gation and the fluorescence of supernatants was meas-10 4 ured. Isotherm fluorescein release experiment was made 10 5 keeping one of the aliquots placed in the AC magnetic 10 6 inductor under isotherm conditions at 37ºC with a water 10 7 recirculating system. The other aliquot was kept at 37 ºC 10 8 in an incubator as control. Alternating magnetic field was 10 9 applied. After 45 min the samples were cooled to 4 ºC, 110 collected by centrifugation and the fluorescence of super-111 natants was measured. 112 RESULTS AND DISCUSSION 113 Ultrastable hydrophobic oleic acid iron oxide nanocrys-114 tals (OA-Fe3O4) were obtained by the co-precipitation 115 method previously described by Haynes and colleagues24, 116 obtaining a colloidal suspension in organic solvent of 1 nanoparticles that present a diameter distribution cen-2 tered on 6 nm, confirmed by TEM images (Figure S1). 3 Then, these magnetic cores were transferred to an aque-4 ous phase using n-cetyltrimethylammonium bromide 5 (CTAB) solution, which acts as a phase transfer agent and 6 as porous structure-directing agent at the same time. 7 Next, a mesoporous silica matrix was formed by the slow 8 addition of tetraethyl orthosilicate (TEOS) in the pres-9 ence of NaOH as a catalyst. After the addition of TEOS, a 10 solution of sililated poly (ethyleneglycol) (Hy-11 droxy(polyethyleneoxy)propyl] tri-ethoxysilane, PEG-Si, 12 MW = 575-750 g/mol) was incorporated in the reaction 13 mixture in order to avoid self-aggregation in the reaction 14 media. Finally, 3-[Tris(trimethylsiloxy)silyl]propyl meth-15 acrylate (MPS) was added, to provide polymerizable 16 groups on the particle surface. Passivation with PEG helps 17 to maintain the nanoparticle dispersity and long-term 18 stability in various biological media and phosphate buff-19 ered saline (PBS).25 At this point, a distribution of 78 nm 20 magnetic mesoporous nanoparticles was obtained by 21 dynamic light scattering that was also confirmed by SEM 22 and TEM images (Figure S2). The organic content of the 23 particles was determined by thermogravimetric analysis 24 (TGA) having a 10% weight loss (Figure S3), correspond-25 ing to PEG and methacrylate groups on the silica surface. 26 This was also confirmed by the FTIR spectra of the mate-27 rial that exhibited the characteristic νas (OC=O) ester 28 band of the methacrylate groups at 1700 cm-1 and σ (CH) 29 at 1472 cm-1 from both functionalizing agents (Figure S4). 3 0 Textural parameters were evaluated by N2 porosimetry 3 1 showing a profile of mesoporous adsorption-desorption 3 2 isotherms and surface area of 1391 m2·g-1 (Figure S5 and 3 3 table S1) which confers high loading capacity for drug 3 4 delivery purposes26. 3 5 Thermosensitive polymers based on poly-N-3 6 isopropylacrylamide (pNIPAm) are widely used for drug 3 7 delivery applications because it can undergo a linear to 3 8 globular phase transition when heated above its lower 3 9 critical solution temperature (LCST), which is 32 oC27 and 40 it presents good biocompatibility.28 The stimuli-41 responsive behavior of pNIPAm based polymers arises 42 from the entropic gain produced when water molecules 43 attached by hydrogen-bonding to the polymer chains are 44 discharged to the aqueous phase above the lower critical 45 solution temperature (LCST) or volume phase transition 46 temperature (VPTT) when crosslinkers are used. This 47 entropy change is responsible for the transition from 48 hydrophilic to hydrophobic state. However, the LCST of 49 pNIPAm is not suitable for drug delivery applications, as 5 0 it is below the body temperature. For the purpose of this 5 1 work, a thermosensitive polymer with an LCST of 42-45 5 2 oC was desired, in order to be in its swelling state (and 5 3 pore-blocking) at the body temperature (37 oC) and col-5 4 lapsed (and pore opening) when heated in temperatures 5 5 between 42 ºC to 45 oC. Therefore, we introduced hydro-5 6 philic co-monomers in the pNIPAm chain to increase the 5 7 polymer transition temperature, specifically N-5 8 hydroxymethyl acrylamide (NHMA), which increases the 5 9 polymer-water interactions resulting in a LCST or VPTT 6 0 enhancement. In order to synthesize a suitable polymer 6 1 which suffers a polymer transition in the 41 ºC to 43ºC 6 2 temperature range, we first synthesized a library of linear 6 3 polymers by radical polymerization in liquid phase using 6 4 different NIPAM/NHMA monomer ratios (Figure 1). 6 5 6 6 Figure 1. Lower critical solution temperature changes with 6 7 the increase of hydrophilic monomer NHMA for linear pol-6 8 ymers measured by turbidimetry (a) and polymerization data 6 9 obtained using 1/400 monomer-to-initiator ratio (b). 7 0 The transition temperature of the synthesized polymers 7 1 was determined by UV-VIS spectrometry keeping the pH 7 2 and the ionic strength fixed at 7.4 and 0.0134 M following 7 3 a procedure described in the supporting information. As 7 4 expected, the LCST of the polymers augmented as the 7 5 hydrophilic monomer ratio increased, while the LCST 7 6 behavior was not detectable when 20% of monomer was 7 7 employed. Once we screened our prepared co-polymers 7 8 library, we chose for further in situ polymer attachment, a 7 9 feed monomer ratio of 90:10 NIPAM/NHMA, which re-8 0 sulted in a polymer with and LCST of 42 ºC. Polymeric 8 1 layer was carried out by a slight modification of the 8 2 method developed by Yang et al29. Therefore, this 8 3 NIPAM/NHMA ratio was used to coat the MMSNPs, in 8 4 the presence of methylene bisacrylamide (MBA) for cross-8 5 linking purposes. After the polymeric shell incorporation, 8 6 the MMSNPs average size distribution raised from 78nm 8 7 to 100nm as measured by dynamic light scattering and in 8 8 accordance with SEM and TEM images. Besides, the mes-8 9 oporous structure was retained and the polymeric coating 9 0 became clearly visible upon staining the nanoparticles 9 1 with uranyl acetate (UA) and TEM observation following 9 2 a similar method reported elsewere30 (Figure 2). The 9 3 amount of grafted polymer was 26 % as determined by 9 4 TGA (Figure S3) and FTIR demonstrated a characteristic 9 5 νas (NC=O) amide band at 1650 cm-1 showing the grafting 1 of the polymeric shell (Figure S4). Stability studies carried 2 out in PBS by DLS demonstrated that the colloidal sus-3 pension was maintained for 8 h, thanks to the polymeric 4 coating (Figure S7). 5 In order to assess the magnetically responsive behavior, 6 the polymer coated mesoporous nanoparticles were char-7 acterized under DC (direct current) and AC (alternating 8 current) magnetic field conditions. DC experiments 9 showed loops with coercive fields close to zero and satu-10 ration magnetization values of 2.80 emu·g-1, evidencing 11 the superparamagnetic behavior of the SPIONS contained 12 within the silica matrix (Figure S8). Different conditions 13 were considered for AC experiments using different fre-14 quencies and field amplitudes listed in table S2. A maxi-15 mum SAR of 178.53 W·g-1 was registered for AC fields of 16 20.05 kA·m-1 and 838 kHz. Therefore these AC field pa-17 rameters were selected for further experiments. 18 19 Figure 2. DLS measurements of MMSNPs and 20 MMSNP@P(NIPAM/NHMA) (a). SEM micrographs (b) and 21 TEM images (c) of the polymer coated nanoparticles. Poly-22 meric coating detail (d). 23 The addition of MBA as a cross-linker to obtain a dense 24 polymer network, besides the surface coating process, 25 could affect the polymer transition temperature of the 26 nanodevice. Additionally, the attachment of a polymer 27 onto a nanoparticle could also affect its temperature-28 responsive behavior. However, LCST measurements in 29 the presence of MMSNP were not possible, due to the UV 3 0 light scattering of the nanoparticles. In order to confirm 3 1 that the transition temperature of the polymer coating is 3 2 maintained after immobilization on the surface, a colloi-3 3 dal suspension of 10 mg/mL fluorescein loaded 3 4 MMSNP@P(NIPAM/NHMA) nanoparticles were placed in 3 5 an incubator. The target temperatures were chosen to be 3 6 37, 40 and 43 ºC. After reaching the target temperatures 3 7 the samples were collected and the fluorescence of the 3 8 supernatants (released fluorescein) was measured. The 3 9 amounts of released fluorophore were similar at 37 ºC and 40 40 ºC, but once the global temperature reached 43 ºC, an 41 increase in fluorescein release was clearly observed (Fig-42 ure 3). This fact indicates that the thermosensible poly-43 mer coating has its VPTT near LCST linear polymer be-44 havior and transition temperature, being able to retain its 45 cargo at physiological temperature and release it once the 46 temperature overcomes the VPTT value, even with a high-47 ly-soluble model drug as the fluorescein sodium salt. 48 Time evolution of the fluorescein release was also tested 49 at 37 ºC and 50 ºC, showing a low premature release of 20 5 0 % of the cargo within the first 24 h (Figure S9). This result 5 1 indicates that this device would release the most of its 5 2 cargo once the target tissue was reached and AMF will be 5 3 applied. 5 4 Therefore, the same experiment was repeated by expos-5 5 ing the fluorescein loaded particles to an AMF at a fre-5 6 quency of 838 kHz and a field of 20.05 kA·m-1.The tem-5 7 perature was monitored using a fiberoptic temperature 5 8 probe and the temperature ramp was replicated to three 5 9 different global temperatures of 37 ºC, 40 ºC and 43 ºC 6 0 similarly to that obtained with the incubator (Figure S10). 6 1 The samples submitted to an AMF presented a similar 6 2 fluorescein release profile at the three target temperatures 6 3 (Figure 3). 6 4 6 5 Figure 3. Fluorescein release of polymer coated nanoparti-6 6 cles at different target temperatures in an incubator (below) 6 7 and under AMF application (above). 6 8 Below the polymer’s VPTT, the fluorescein release was 6 9 low compared to the sample that reached 43 ºC, revealing 7 0 that it is possible to provoke the polymer transition under 7 1 an alternating magnetic field, as we have previously de-7 2 scribed13. Moreover, there was a significant difference in 7 3 the released cargo between the samples exposed to ther-7 4 mal and magnetic heating after reaching the VPTT. As it 7 5 is clearly shown in figure 3, the sample exposed to mag-7 6 netic field was able to release a significantly higher 7 7 amount of fluorescein in the same period of time. This 7 8 variance in fluorescein release could be explained by dif-7 9 ferent mechanisms: (i) the polymer transition tempera-8 0 ture is reached faster in the particle surroundings when 8 1 AMF is applied, which is consistent with previously cited 8 2 works,18,19 (ii) the localized temperature at the nanoparti-8 3 cle proximities is higher than the global temperature, 8 4 which causes a faster diffusion of the loaded molecules 8 5 once the pore is opened and (iii) the vibrations of the 8 6 SPIONs when subjected to AMF can also induce a higher 8 7 release as compared to simple heating in the incubator. 8 8 These mechanisms could act cooperatively resulting in 1 the enhanced fluorescein release. 2 The response to heat dissipation by physiological envi-3 ronment is a critical issue still unclear. There are doubts 4 whether these magnetic devices are capable to overcome 5 the heat dissipation that is observed in living tissues. To 6 resolve this matter of concern, we developed a second 7 experiment by applying AMF under isotherm conditions, 8 keeping constant the temperature at 37ºC with a water 9 recirculating system while the AMF is applied. Then, the 10 fluorescein release experiment was performed using this 11 isothermal set-up, placing one aliquot in the AC magnetic 12 inductor at 37ºC, while another aliquot was placed at 37 13 ºC in an incubator as control. The alternating magnetic 14 field was applied for 45 min. Subsequently, the samples 15 were rapidly cooled to 4 ºC to avoid fluorescein leaking, 16 collected by centrifugation and the fluorescence of super-17 natants was measured. Results showed a 2-fold fluoresce-18 in release upon an AMF application with regard to the 19 sample placed in an incubator (Figure 4). 20 21 Figure 4. Fluorescein release from 22 MMSNP@P(NIPAM/NHMA) under AMF in isothermal con-23 ditions. 24 CONCLUSIONS 25 This fact proves that thermal energy dissipated by SPI-26 ONs under a magnetic field is capable to effectively reach 27 the surface of the nanoparticles through the silica matrix, 28 provoking the polymer hydrophilic to hydrophobic transi-29 tion and the subsequent drug release when under iso-3 0 thermal conditions at 37ºC, which mimics the physiologi-3 1 cal conditions. This phenomenon can be explained by the 3 2 so called hot spot effect, where the existence of a hot 3 3 source in the interior of the device accomplishes high 3 4 local temperatures but no global heating is observed. 3 5 This work shows that the application of an AMF to our 3 6 system provokes the release of the cargo without the 3 7 necessity to achieve a global temperature increase. The 3 8 hot spots generated by magnetite nanoparticles are capa-3 9 ble to provoke the shrinkage of the polymer network on 40 the silica matrix, achieving the effective triggered release 41 of the cargo. It must be highlighted that the aim of this 42 work is not demonstrating the hot spot effect, but its 43 capability for releasing the cargo in a stimuli-responsive 44 way, although the environment remained at physiological 45 temperature. Heating the environment requires much 46 magnetic energy that, very often, is non-compatible with 47 the clinical/technical requirements, which is a very seri-48 ous limitation for the transference to the biomedical 49 field.31 These results pave the way to the application of 5 0 magnetically triggered carriers for the treatment of com-5 1 plex diseases due to the biocompatibility and high pene-5 2 tration capacity of magnetic fields. Moreover, the pres-5 3 ence of magnetic cores within the silica matrix opens the 5 4 possibility to combine the stimuli-responsive capacity 5 5 with imaging applications in the same carrier, converting 5 6 this device in an excellent prototype in the theranostics 5 7 field. Further work is ongoing in order to test the capacity 5 8 of this device to transport and release cytotoxic drugs 5 9 using in vivo models. 6 0 6 1 Supporting Information. Characterization techniques as 6 2 transmission and scanning electron microscopy, dynamic 6 3 light scattering, nitrogen sorption isotherms, IR and NMR 6 4 spectroscopy, release experiments, thermogravimetric ana-6 5 lisys, vibrating sample magnetometer and SAR measure-6 6 ments, Figures S1−S10 and Tables S1 and S2 are showed in this 6 7 associated content, as well as materials and methods used for 6 8 the preparation of the nanoparticles. This material is availa-6 9 ble free of charge via the Internet at http://pubs.acs.org. 7 0 7 1 7 2 *E-mail: vallet@ucm.es, Phone 34 91 3941870. Fax 34 91 7 3 3941786. 7 4 7 5 The manuscript was written through contributions of all 7 6 authors. All authors have given approval to the final version 7 7 of the manuscript. 7 8 7 9 This work was supported by the Ministerio de Economía y 8 0 Competitividad, through projects MAT2012-35556, MAT2013-8 1 43299R and CSO2010-11384-E (Agening Network of Excel-8 2 lence), ERC-Starting Grant 239931-NANOPUZZLE project, 8 3 Fondo Social Europeo (FSE; Gobierno de Aragón), CIBER-8 4 BBN is an initiative funded by the VI National R&D&i Plan 8 5 2008-2011, Iniciativa Ingenio 2010, Consolider Program, 8 6 CIBER Actions and financed by the Instituto de Salud Carlos 8 7 III with assistance from the European Regional Development 8 8 Fund.. We also thank the X-ray Diffraction C.A.I., NMR 8 9 C.A.I. and the National Electron Microscopy Center, UCM. E. 9 0 G. thanks CEI Campus Moncloa for the PICATA fellowship. 9 1 9 2 (1) Jain, R. K.; Stylianopoulos, T. Delivering Nanomedicine to 9 3 Solid Tumors. Nat. Rev. Clin. Oncol. 2010, 7, 653–664. 9 4 (2) Baeza, A.; Colilla, M.; Vallet-Regí, M. Advances in 9 5 Mesoporous Silica Nanoparticles for Targeted Stimuli-9 6 Responsive Drug Delivery. Expert Opin. Drug Deliv. 2015, 12, 9 7 319–337. 9 8 mailto:vallet@ucm.es (3) Vallet-Regi, M.; Rámila, a.; Del Real, R. P.; Pérez-Pariente, J. 1 A New Property of MCM-41:  Drug Delivery System. Chem. 2 Mater. 2001, 13, 308–311. 3 (4) Maeda, H.; Nakamura, H.; Fang, J. The EPR Effect for 4 Macromolecular Drug Delivery to Solid Tumors: 5 Improvement of Tumor Uptake, Lowering of Systemic 6 Toxicity, and Distinct Tumor Imaging in Vivo. Adv. Drug 7 Deliv. Rev. 2013, 65, 71–79. 8 (5) Wicki, A.; Witzigmann, D.; Balasubramanian, V.; Huwyler, 9 J. Nanomedicine in Cancer Therapy: Challenges, 10 Opportunities, and Clinical Applications. J. Control. Release 11 2015, 200, 138–157. 12 (6) Angelos, S.; Khashab, N. M.; Yang, Y. W.; Trabolsi, A.; 13 Khatib, H. a.; Stoddart, J. F.; Zink, J. I. pH Clock-Operated 14 Mechanized Nanoparticles. J. Am. Chem. Soc. 2009, 131, 15 12912–12914. 16 (7) Chen, L.; Li, L.; Zhang, L.; Xing, S.; Wang, T.; Wang, Y. A.; 17 Wang, C.; Su, Z. Designed Fabrication of Unique Eccentric 18 Mesoporous Silica Nanocluster-Based Core-Shell 19 Nanostructures for pH-Responsive Drug Delivery. ACS 20 Appl. Mater. Interfaces 2013, 5, 7282–7290. 21 (8) Mackowiak, S. A.; Schmidt, A.; Weiss, V.; Argyo, C.; Von, C. 22 Targeted Drug Delivery in Cancer Cells with Red Light 23 PhotoactivatedMesoporous Silica Nanoparticles_SI. Nano 24 Lett. 2013, 13, 1–19. 25 (9) Zhang, J.; Yuan, Z. F.; Wang, Y.; Chen, W. H.; Luo, G. F.; 26 Cheng, S. X.; Zhuo, R. X.; Zhang, X. Z. Multifunctional 27 Envelope-Type Mesoporous Silica Nanoparticles for Tumor-28 Triggered Targeting Drug Delivery. J. Am. Chem. Soc. 2013, 29 135, 5068–5073. 3 0 (10) Slowing, I. I.; Trewyn, B. G.; Giri, S.; Lin, V. S. Y. 3 1 Mesoporous Silica Nanoparticles for Drug Delivery and 3 2 Biosensing Applications. Adv. Funct. Mater. 2007, 17, 1225–3 3 1236. 3 4 (11) Tarn, D.; Ashley, C. E.; Xue, M.; Carnes, E. C.; Zink, J. I.; 3 5 Brinker, C. J. Mesoporous Silica Nanoparticle Nanocarriers: 3 6 Biofunctionality and Biocompatibility. Acc. Chem. Res. 3 7 2013, 46, 792–801. 3 8 (12) Ruiz-Hernández, E.; Baeza, A.; Vallet-Regí, M. Smart Drug 3 9 Delivery through DNA/Magnetic Nanoparticle Gates. ACS 40 Nano 2011, 5, 1259–1266. 41 (13) Baeza, A.; Guisasola, E.; Ruiz-Hernández, E.; Vallet-Regí, M. 42 Magnetically Triggered Multidrug Release by Hybrid 43 Mesoporous Silica Nanoparticles. Chem. Mater. 2012, 24, 44 517–524. 45 (14) Chang, B.; Sha, X.; Guo, J.; Jiao, Y.; Wang, C.; Yang, W. 46 Thermo and pH Dual Responsive, Polymer Shell Coated, 47 Magnetic Mesoporous Silica Nanoparticles for Controlled 48 Drug Release. J. Mater. Chem. 2011, 21, 9239. 49 (15) Liu, C.; Guo, J.; Yang, W.; Hu, J.; Wang, C.; Fu, S. Magnetic 5 0 Mesoporous Silica Microspheres with Thermo-Sensitive 5 1 Polymer Shell for Controlled Drug Release. J. Mater. Chem. 5 2 2009, 19, 4764. 5 3 (16) Wust, P.; Hildebrandt, B.; Sreenivasa, G.; Rau, B.; 5 4 Gellermann, J.; Riess, H.; Felix, R.; Schlag, P. M. 5 5 Hyperthermia in Combined Treatment of Cancer. Lancet 5 6 Oncol. 2002, 3, 487–497. 5 7 (17) Dutz, S.; Hergt, R. Magnetic Nanoparticle Heating and 5 8 Heat Transfer on a Microscale: Basic Principles, Realities 5 9 and Physical Limitations of Hyperthermia for Tumour 6 0 Therapy. Int. J. Hyperthermia 2013, 29, 790–800. 6 1 (18) Dias, J. T.; Moros, M.; Del Pino, P.; Rivera, S.; Grazú, V.; de 6 2 la Fuente, J. M. DNA as a Molecular Local Thermal Probe 6 3 for the Analysis of Magnetic Hyperthermia. Angew. Chem. 6 4 Int. Ed. Engl. 2013, 52, 11526–11529. 6 5 (19) Riedinger, A.; Guardia, P.; Curcio, A.; Garcia, M. A.; 6 6 Cingolani, R.; Manna, L.; Pellegrino, T. Subnanometer Local 6 7 Temperature Probing and Remotely Controlled Drug 6 8 Release Based on Azo-Functionalized Iron Oxide 6 9 Nanoparticles. Nano Lett. 2013, 13, 2399–2406. 7 0 (20) Dong, J.; Zink, J. I. Taking the Temperature of the Interiors 7 1 of Magnetically Heated Nanoparticles. ACS Nano 2014, 8, 7 2 5199–5207. 7 3 (21) Kim, J.; Kim, H. S.; Lee, N.; Kim, T.; Yu, T.; Song, I. C.; 7 4 Moon, W. K.; Hyeon, T. Multifunctional Uniform 7 5 Nanoparticles Composed of a Magnetite Nanocrystal Core 7 6 and a Mesoporous Silica Shell for Magnetic Resonance and 7 7 Fluorescence Imaging and for Drug Delivery. Angew. 7 8 Chemie 2008, 120, 8566–8569. 7 9 (22) Lin, Y.-S.; Hurley, K. R.; Haynes, C. L. Critical 8 0 Considerations in the Biomedical Use of Mesoporous Silica 8 1 Nanoparticles. J. Phys. Chem. Lett. 2012, 3, 364–374. 8 2 (23) Meng, H.; Xue, M.; Xia, T.; Ji, Z.; Tarn, D. Y.; Zink, J. I.; Nel, 8 3 A. E. Use of Size and a Copolymer Design Feature to 8 4 Improve the Biodistribution and the Enhanced 8 5 Permeability and Retention Effect of Doxorubicin-Loaded 8 6 Mesoporous Silica Nanoparticles in a Murine Xenograft 8 7 Tumor Model. ACS Nano 2011, 5, 4131–4144. 8 8 (24) Lin, Y.-S. S.; Haynes, C. L. Synthesis and Characterization of 8 9 Biocompatible and Size-Tunable Multifunctional Porous 9 0 Silica Nanoparticles. Chem. Mater. 2009, 21, 3979–3986. 9 1 (25) Lin, Y.-S.; Abadeer, N.; Haynes, C. L. Stability of Small 9 2 Mesoporous Silica Nanoparticles in Biological Media. 9 3 Chem. Commun. 2011, 47, 532–534. 9 4 (26) Vallet-Regí, M.; Balas, F.; Arcos, D. Mesoporous Materials 9 5 for Drug Delivery. Angew. Chemie - Int. Ed. 2007, 46, 7548–9 6 7558. 9 7 (27) Schild, H. G. Poly ( N-Isopropylacrylamide ): Experiment , 9 8 Theory and Application. Prog. Polym. Sci. 1992, 17, 163–249. 9 9 (28) Wadajkar, A. S.; Koppolu, B.; Rahimi, M.; Nguyen, K. T. 10 0 Cytotoxic Evaluation of N-Isopropylacrylamide Monomers 10 1 and Temperature-Sensitive poly(N-Isopropylacrylamide) 10 2 Nanoparticles. J. Nanoparticle Res. 2009, 11, 1375–1382. 10 3 (29) Chang, B.; Chen, D.; Wang, Y.; Chen, Y.; Jiao, Y.; Sha, X.; 10 4 Yang, W. Bioresponsive Controlled Drug Release Based on 10 5 Mesoporous Silica Nanoparticles Coated with Reductively 1 Sheddable Polymer Shell. Chem. Mater. 2013, 25, 574–585. 2 (30) Baeza, A.; Guisasola, E.; Torres-Pardo, A.; González-Calbet, 3 J. M.; Melen, G. J.; Ramirez, M.; Vallet-Regí, M. Hybrid 4 Enzyme-Polymeric Capsules/mesoporous Silica Nanodevice 5 for in Situ Cytotoxic Agent Generation. Adv. Funct. Mater. 6 2014, 24, 4625–4633. 7 (31) Brezovich, I. A.; Meredith, R. F. Practical Aspects of 8 Ferromagnetic Thermoseed Hyperthermia. Radiol. Clin. 9 North Am. 1989, 27, 589—602. 10 11 12 SYNOPSIS TOC The key of magnetic triggered stimuli responsive devices is to generate enough thermal energy to trigger the drug release. We proved for the first time the possibility to produce enough local heat to trigger the cargo release without the need of increasing the global temperature in a stimuli-responsive device. This verifies that the so called “hot spot” effect can overcome heat dissipation at physiological temperature.