Pharm Res (2018) 35: 144 https://doi.org/10.1007/s11095-018-2425-2 RESEARCH PAPER Iron-Based Metal-OrganicFrameworksasaTheranostic Carrier for Local Tuberculosis Therapy 1 2 3 3 Gabriela Wyszogrodzka & Przemysław Dorożyński & Barbara Gil & Wieslaw J. Roth & 3 3 4 5 Maciej Strzempek & Bartosz Marszałek & Władysław P. Węglarz & Elżbieta Menaszek & 3 6 Weronika Strzempek & Piotr Kulinowski Received: 14 February 2018 /Accepted: 3 May 2018 /Published online: 18 May 2018 # The Author(s) 2018 ABSTRACT system can also serve as the MRI contrast agent. The drug Purpose The purpose of the study was initial evaluation of dissolution showed extended release of isoniazid. MOF parti- applicability of metal organic framework (MOF) Fe-MIL-101- cles accumulated in the L929 fibroblast cytoplasmic area, sug- NH as a theranostic carrier of antituberculous drug in terms gesting MOF release the drug inside the cells. The cytotoxicity of its functionality, i.e. drug loading, drug dissolution, mag- confirmed safety of MOF system. netic resonance imaging (MRI) contrast and cytotoxic safety. Conclusions The application of MOF for extended release Methods Fe-MIL-101-NH was characterized using X-ray inhalable system proposes the novel strategy for delivery of powder diffraction, FTIR spectrometry and scanning electron standard antimycobacterial agents combined with monitoring microscopy. The particle size analysis was determined using of their distribution within the lung tissue. laser diffraction. Magnetic resonance relaxometry and MRI were carried out on phantoms of the MOF system suspended KEY WORDS iron metal-organic framework (MOF) MRI . . . in polymer solution. Drug dissolution studies were conducted contrast agent theranostic system tuberculosis treatment using Franz cells. For MOF cytotoxicity, commercially avail- inhaled dosage forms able fibroblasts L929 were cultured in Eagle’s Minimum Essential Medium supplemented with 10% fetal bovine serum. ABBREVIATIONS Results MOF particles were loaded with 12% of isoniazid. CPMG Carr Purcell Meiboom Gill Theparticlesize(3.37–6.45 μm) depended on the DDS Drug delivery systems micronization method used. The proposed drug delivery DMF Dimethylformamide EDS Element Energy Dispersive Spectroscopy FOV Field of view * Przemysław Dorożyński FTIR Fourier-transform infrared spectroscopy firstname.lastname@example.org HPMC Hydroxypropylmethylcellulose INH Isoniazid Faculty of Pharmacy, Department of Pharmacobiology Jagiellonian IR Infrared University Medical College, Medyczna 9 30-068 Kraków, Poland MCT Mercury-Cadmium-Telluride Pharmaceutical Research Institute, Rydygiera 8 MIC Minimum inhibitory concentration 01-793 Warszawa, Poland MOF Metal organic framework Faculty of Chemistry Jagiellonian University in Kraków, Gronostajowa 2 MRI Magnetic resonance imaging 30-387 Kraków, Poland MSME Multi-Slice Multi-Echo Department of Magnetic Resonance Imaging, Institute of Nuclear Physics MTB Mycobacterium tuberculosis Polish Academy of Sciences, Radzikowskiego 152 NE Number of echoes 31-342 Kraków, Poland 5 NP Nanoparticles Faculty of Pharmacy, Department of Cytobiology Jagiellonian University PBS Phosphate-buffered saline Medical College, Medyczna 9 30-068 Kraków, Poland 6 PDT Photodynamic therapy Faculty of Mathematics, Physics and Technical Science, Institute of SD Standard deviation Technology Pedagogical University of Cracow, Podchorążych 2 30-084 Kraków, Poland SEM Scanning Electron Microscopy 144 Page 2 of 11 Pharm Res (2018) 35: 144 TB Tuberculosis Therefore, it is necessary to evaluate their nanosafety regard- TE Inter-echo time ing particular application and involved cell type. Wuttke et al. TR Repetition time (11) discuss the effects of MIL-101(Cr) and MIL-101(Fe) on WHO World Health Organization human endothelial and mouse lung cells, a first line of defense XRD X-ray powder diffraction upon systemic blood-mediated and local lung-specific appli- cations of nanoparticles. Magnetic resonance imaging (MRI) has become a powerful INTRODUCTION tool in medicine for non-invasive imaging of the internal struc- ture and functions of living organisms as well as local proper- The use of nanotechnology for medical applications is rapidly ties of tissues (17). Magnetic nanoparticles may be applied as growing and is very promising in various branches of applied contrasting agents providing either negative (T -weighted) or science (1–4). Nanoparticles (NPs) are used as diagnostic im- positive imaging contrast (T -weighted) (18). A potential MRI aging agents or as drug delivery platforms, providing targeted contrasting agent has to fulfill several requirements related to or tissue-selective therapy, which may increase efficiency and tolerance, safety, toxicity, stability, osmolarity, viscosity, decrease the side effects of drugs. It is also possible to combine biodistribution, elimination, and metabolism (19). 3+ 2+ 3+ these two functions in one particle by the design and prepa- Embedding paramagnetic cations (Gd ,Mn ,Fe )in ration of dual-purpose nanomaterials, functioning as both di- MOF structure make them possible to be used as MRI con- agnostic medical devices and drug delivery systems (5,6). This trast agents. Among them, iron is the best option from a tox- concept of fusing diagnostics and therapy has been proposed icological point of view. Nanoscale iron MOFs (MIL-53, in 2002 and called theranostics (7). Theranostic agents have MIL-88A, MIL-88Bt, MIL-89, MIL-100 and MIL- been defined as Bintegrated nanotherapeutic systems, which 101_NH ), with engineered cores and surfaces, have been can diagnose, deliver targeted therapy, and monitor the re- shown to be able to serve as drug carrier and magnetic reso- sponse to the therapy^ (6). This integrated approach offers nance contrast agent according to good ability for modifica- great opportunities in the development of personalized med- tion of relaxivities (20). icine It allows for monitoring the drug release, its Polymeric surface allows for the MOF structures modifica- biodistribution and accumulation at the target site, dose ad- tion to implement properties such as increased chemical and justment to individual patients and finally, monitoring the colloidal stability which enhance the cellular uptake, or dye- course of a disease (5,8,9). labeling, which enables for example, the investigation of nano- Freund et al. has proposed the term Batom economy^ particle uptake into tumor cells by fluorescence microscopy which focuses on the design of highly active materials (21). It has been proved, that coated iron MOFs can retain possessing many functionalities that work together to serve a their MRI contrast properties. MOF NPs are frequently coat- specific purpose (10). Metal-Organic Frameworks (MOFs) are ed in order to prevent leakage of the drug before they reach excellent example to illustrate this concept and have the po- the target (e.g. exosome-coated MIL-88A, liposome coated tential to emerge as next-generation drug delivery systems MIL-88A) (21,22). MOFs, including iron MOFs, can be (DDS). MOFs may be of interest as carriers for theranostics, (multi)functionalized (example can be found in the work of being porous structures built from inorganic nodes, which are Roeder et al. 2017 (21)) – molecular units can be anchored single ions or clusters of ions, joined together by organic on the outer surface of MOFs. linkers. Such design allow to gain control over the framework MOFs can be used as stimuliresponsive DDS for cancer architecture and, even more importantly, the pore chemistry, therapy (targeted chemotherapy, gene therapy). Recent stud- enabling targeted functionalization for nanomedical applica- ies also shown that it is possible to develop MOF based deliv- tions (11). Thanks to their porous structure, MOFs seem to be ery systems for photodynamic therapy (PDT) of cancer (23). promising drug vehicles with potential high drug loading While theranostics has been intended mainly for cancer (12–14). The control of guest release profiles can be gained treatments, there are numerous other therapeutic targets for by the choice of the type of functional group of the linker and which the effectiveness of therapy could be increased by local tuning of the pore size (15). drug delivery and monitoring of its distribution. Among them In the case of MOF application for drug delivery, they tuberculosis (TB) seems to be especially important. should exhibit stability under physiological condition, mini- TB is one of the top three infectious diseases – together mal toxicity, biodegradability and as biocompatibility for both with HIV and malaria – causing morbidity and death world- metal and bridging linker ligand. MOFs as components of wide. According to the World Health Organization (WHO) drug delivery system are discussed in the context of other estimations, about 30% of the world’s population is infected nanoparticulate carriers (mesoporous silica, dendrimers) in with Mycobacterium tuberculosis (MTB). Every year, about 10 the work byWuttkeetal. (16). Safety of MOF as drug delivery million of new cases are registered and about 1.5 to 2 million nanomaterials varies strongly with effector cell type. of deaths are caused by TB, according to the report by the Pharm Res (2018) 35: 144 Page 3 of 11 144 WHO agency (24). TB infections frequently become multi- MATERIALS AND METHODS drug-resistant, since the conventional treatment protocol is based on an antibiotic therapy carried out over periods of 6 MOF Synthesis and Drug Incorporation to 9 months (25). Isoniazid (Isonicotinic Acid Hydrazide, INH) is particularly Fe-MIL-101-NH was synthesized according to the procedure suitable for use as a model drug in theranostic drug delivery reported by Bauer et al. (37). The MOF powder samples were because it is an antibiotic used as a first-line agent in the comminuted by milling in an agate ball mill for 24 h and then prevention and treatment of both latent and active tuberculo- sonicated for 5 min. Using a CP 130 PB (130 W, 20 kHz) sis. It is effective against mycobacteria, especially ultrasonic processor (Cole-Parmer, USA), at 70% of maxi- Mycobacterium tuberculosis since it inhibits biosynthesis of the mum amplitude. After drying MOF was activated under vac- mycolic acid. Oral administration of INH and long-term uum at 100°C for 30 min. therapy causes several serious side effects, which could even Isoniazid (Sigma-Aldrich, Germany) was incorporated into force treatment discontinuation. INH is known to cause the MOF matrix by mixing 1.5 mL of saturated solution of hepatitis (26), hepatic injury and neuropsychiatric INH in DMF with 300 mg of micronized Fe-MIL-101-NH disturbances (27) including, among others, uncontrollable The slurry was mixed for 12 h at room temperature and the seizures (28)orpolyneuropathy(29). The conventional oral product was separated by centrifugation and washed with route of INH administration causes periodic decrease of its ethanol. concentration below the effective minimum inhibitory con- centration (MIC), allowing MTB bacilli to develop resis- MOF Characterisation tance (30). The distribution of the antituberculous drug within the infected tissue is equally important. The studies The characterization by X-ray powder diffraction was carried of Kjellsson et al. (31) on a TB infected animal model have out using a Bruker AXS D8 Advance (Bruker, Germany) dif- shown that after systemic administration the distribution of fractometer in the range 1–30° 2Θ using CuKα (λ = isoniazid, rifampicin and pyrazinamide in the lung tissue is 0.154178 nm) radiation. uneven. The concentrations of drugs in pulmonary lesions Infrared (IR) spectra were measured in transmission mode where the pathogen is located have been markedly lower using Tensor 27 FTIR spectrometer (Bruker, Germany) than in the surrounding lung tissue. equipped with an MCT (Mercury-Cadmium-Telluride) de- −1 Hickey et al. (32) indicated the problem of dramatic in- tector at spectral resolution of 2 cm . Before measurement crease in extremely drug-resistant TB around the world a sample was deposited on an IR-transparent silicon wafer and highlighted that pulmonary administration offers the (pure for electronic purposes) by placing its ethanol solution ability to deliver drug directly to the infected macrophages directly on the disc and evaporating the solvent. The wafer in the deepest part of lungs. Comparison of various dry- was placed in an IR cell with KBr windows and slowly heated −3 powders containing anti-TB drugs for inhalation presented under constant pumping (10 Torr) to 50°C (with the drug by Pham et al. (33) indicates that the strategy of inhaled present) or 100°C (pure MOF). therapy is the main and the most promising alternative to Scanning Electron Microscopy (SEM) analysis was per- traditional approach and is necessary to achieve global TB formed using Nova Nano SEM 200 (FEI Europe B.V.) control. cooperating with the Element Energy Dispersive The MOF-based theranostics may be an interesting alter- Spectroscopy (EDS) analyzer (EDAX Inc., U.S.A.) using native for the standard therapy of tuberculosis. Pulmonary secondary electrons in low vacuum conditions (60 Pa). route should ensure high local drug concentration (avoiding Samples of MOF without treatment, MOF after milling side effects of the systemic drug action). Moreover, possibility and MOF after milling and sonication were analyzed. to generate contrast in magnetic resonance images should The particle size distribution of the powder samples was allow deposition and lung clearance monitoring (34–36) – measured in terms of particle diameter at 50% in the cumu- the first tests on animals were performed in a clinical system lative distribution (Dx (38)) using laser diffraction particle size with a clinical nebulization setup and a low inhaled dose (34). analyzer Mastersizer 3000 (Malvern Instruments Ltd., United There are two reasons for deposition monitoring: optimiza- Kingdom). tion of the dosage form at formulation stage and optimization of therapy. Drug Release The aim of the study was initial evaluation of applicabil- ity of MOF Fe-MIL-101-NH as a theranostic carrier of The drug release study was performed in Franz cells (39). The antituberculous drug in terms of its functionality, i.e. drug donor compartment was separated from the acceptor com- loading, drug dissolution, MRI contrast and cytotoxic partment by a cellulose acetate filter with pore size of safety. 0.8 μm (Sartorius, Germany) covered with 200 μLof 1% 144 Page 4 of 11 Pharm Res (2018) 35: 144 mucin solution simulating the mucus layer deposited on lung maintained at 37°C in a humidified incubator (ThermoSci, epithelium (40–43). The 5 mg of composite powder was Germany) with 5% CO until 70–80% confluence was ob- placed in the donor compartment. The acceptor compart- tained. At passage 3 the cells were detached using 0.5% tryp- ment was filled with 5 mL of phosphate-buffered saline sin-EDTA, centrifuged and suspended in the fresh growth (PBS)atpH7.4 (44,45). The experiment was performed at medium. Next, the cells were seeded at density 0.5 · 10 37°C under continuous stirring at 140 rpm. Samples of cells/200 μL in 96-well culture plates (Nunc, Denmark) and 0.4 mL were taken from the acceptor compartment at 0.5, allowed to adhere. After 24 h, suspensions of milled or milled 1, 1.5, 2, 3, 4, 5, 6, 8, 10, 24 and 48 h and were replaced with and sonicated particles of Fe-MIL-101-NH were added to 0.4 mL of a fresh medium. A blank was carried out by evalu- the culture at two concentrations: 0.625 and 1.25 mg/mL. ation of released crystalline isoniazid in quantity correspond- Viability of L929 fibroblasts cultured in contact with MOF ing to the content of drug in MOF. The isoniazid concentra- suspension for 24 or 72 h was determined with resazurin- tion was analyzed in an aqueous solution using a UV-Vis based reagent PrestoBlue™ (Invitrogen, USA). Fluorescent spectrophotometer UV 1800 (Shimadzu, Japan) at a wave- product of the reaction was detected using POLARstar length λ = 262.0 nm. Omega microplate reader (BMG Labtech, Germany). Obtained results are presented as the mean ± standard devi- Magnetic Resonance Imaging and Relaxometry ation (SD) of five samples. Significant effects (p <0.05) were determined using Student’s t-test. Cells morphology was ob- Nuclear magnetic resonance imaging and relaxometry served under contrast phase inverted microscope CKX53 were performed using 9.4 T MRI research system (Bruker (Olympus, Japan). Biospin, Germany) and TopSpin 2.0 software (Bruker Biospin,Germany).Forthis purpose0.3,1.7,3.3,6.7, 13.3, 26.7 and 53.3 mg/mL MOF suspensions in 2% RESULTS AND DISCUSSION hydroxypropylmethylcellulose (HPMC; Metolose, 90SH, 400 cP - Shin-Etsu, Japan) water solution were prepared. Synthesis and Characterization of MOFs. T and T relaxation times were measured using Inversion 1 2 Recovery and Carr Purcell Meiboom Gill (CPMG) pulse se- Fe-MIL-101-NH of good quality was synthesied, as con- quences, respectively – number of accumulations (NA) = 8 firmed by the diffractogram (XRD pattern) (Fig. 1), which and dwell time (DW) = 10 μs. For T assessment separate showed a very good agreement with the calculated MIL-101 measurements with 16 logarithmically spaced inversion time patterns published in the literature (46). The intensities of the values were performed in order to sample T recovery. For reflections increased considerably after washing with ethanol CPMG sequence the signal was acquired after 16 logarithmi- of both the as-synthesized sample, containing free DMF mol- cally spaced n� TE time intervals (where TE = 0.2 ms). The ecules, and for the MOF sample with incorporated INH. This data were fitted assuming monoexponential T decay (T re- is most likely due to the common effect observed when micro- 2 1 covery). Linear regression of R (= 1/T )and R (= 1/T )vs. pores are emptied from the occluded guest molecules. 1 1 2 2 MOF concentration in suspension was performed to obtain r Iron-containing MOF compound Fe-MIL-101-NH was 1 2 and r relaxivities with intercept set at R and R values of chosen for two reasons. First, it is one of the most stable Fe- 2 1 2 pure polymer solution. based Metal-Organic Frameworks which has been already MR imaging was performed using Multi-Slice Multi-Echo reported as a carrier for bioactive (47,48)ormagnetic (MSME) imaging sequence for pure 2% HPMC solution as (49,50) compounds. Second, the presence of the –NH func- well as for 26.7 and 53.3 mg/mL MOF suspensions in 2% tional groups in the linker allows easy determination whether HPMC solution. Two sets of images at two different repetition guest molecules are located inside the MOF cavities, time (TR), i.e. 0.7 s and 3 s were obtained with following interacting with them and changing their properties, or are parameters: field of view (FOV) = 28 × 28 mm , slice thick- only adsorbed in the intercrystalline voids. ness = 1 mm (axial cross section), image matrix size of 256 × MIL-101 has a rigid zeotype (MTN) crystal structure (50) 256, number of echoes (NE) = 256, inter-echo time (TE) = with two types of cages. Its medium size cavities with the 3.5 ms, NA = 2. diameter of 2.9 nm are accessible through pentagonal win- dows with the opening of 1.2 nm, while large 3.4 nm cavities In Vitro Cytotoxicity have hexagonal windows with the diameter of 1.6 nm (51). The 3D molecular size for INH was determined (52)using For MOF cytotoxicity study, commercially available fibro- Chem Office Software 2008 as equal to 1.05 × 0.72 × blasts L929 (Sigma-Aldrich, Germany) were cultured in 0.31 nm. The spacious MOF cavities together with relatively Eagle’s Minimum Essential Medium supplemented with large apertures make carrier suitable for incorporation of 10% fetal bovine serum (ATCC, USA). Cells were bulky drug molecules. Pharm Res (2018) 35: 144 Page 5 of 11 144 aggregates resulting in very homogeneously looking SEM im- ages of single crystals, as shown in Fig. 2c. The measurement of particle size distribution was per- formed using laser diffraction method. The Dx (38) parameter for MOF samples without treatment was 6.45 μm (SD: ± 0.20). Dx (38) decreased as a result of milling to 5.51 μm (SD: ± 0.21) and to 3.37 μm (SD: ± 0.03) after subsequent sonication. It indicates the possibility to adjust Fe-MIL-101- NH size by applying the appropriate micronization method to obtain the desired size. INH Incorporation Fig. 1 XRD patterns of Fe-MIL-101-NH : as synthesized, washed with ethanol, after introduction of isoniazid and again washed with ethanol after To evaluate the content of the drug inside the pores the XRD isoniazid introduction (from bottom to top) study was conducted, and no free isoniazid crystals were evi- dent in the XRD pattern (Fig. 1). The intensities of all XRD Particle Size reflections decreased markedly for the isoniazid-containing MOF, almost to the same level as observed for the as- Regarding particle size, according Hirschle et al. (53), MOF synthesized samples filled with the solvent (DMF). Washing NPs intended to use as drug delivery systems shoud be char- the drug-MOF composite with ethanol caused partial extrac- acterized using multiple techniques – as powder and also in tion of the encapsulated isoniazid, which was indicated by dispersion. They also discuss the appropriate method for increased intensities of the XRD reflections. This suggests that obtaining the nanoparticle size that is meaningful in the con- INH molecules were located inside the pores of the MOF text of the desired application. In our work we evaluated the material. impact of milling and subsequent ultrasonication on Fe-MIL- The mode of INH incorporation was further investigated 101-NH crystal size and shape. Three kinds of samples were 2 by FTIR spectroscopy (Fig. 3). After INH introduction the examined: MOF without treatment (Fig. 2a), MOF after mill- spectrum was not a simple superposition of the spectra ing (Fig. 2b) and MOF after milling and ultrasonication (Fig. chearchterstic of the pure component of MOF and INH, 2c). again suggesting that the drug molecules were located inside For dry powders SEM imaging and measurement of par- the pores of Fe-MIL-101-NH . In the pure MOF material the ticle size distribution were carried out. –NH groups of the structure-forming linker were character- −1 Milling the samples did not change the overall shape and ized by two IR maxima at 3505 and 3390 cm characteristic size of crystals, which appeared to be undamaged in SEM of ν and ν N-H vibrations (38,54), each split into two as sym images (Fig. 2b). It was also not able to break crystal aggre- components – one at the higher and one at the lower frequen- gates; thus this method cannot be used by itself for cy. The lower frequency components (bands at 3485 and −1 micronization of the samples. For this reason, crystals after 3295 cm ) could be due to hydrogen-bonding of the –NH milling were ultrasonicated, which caused breakage of the moieties. This would suggest that –NH functionalities of the Dx(50) = 6.45µm (SD: ± 0.20) Dx(50) = 5.51µm (SD: ± 0.21) Dx(50) = 3.37µm (SD: ± 0.03) a. Fe-MIL-101-NH b. Fe-MIL-101-NH -M c. Fe-MIL-101-NH -MU 2 2 2 Fig. 2 SEM images of Fe-MIL-101-NH samples: (a) without treatment, (b) after milling (-M), (c) after milling and ultrasound treatment (-MU) 2 144 Page 6 of 11 Pharm Res (2018) 35: 144 Fig. 3 Drug loading: IR spectra in the N-H stretching vibration region (left) and framework vibra- tions (right) in transmission mode of: pure isoniazid (a), pure Fe-MIL- 101-NH (b), physical mixture of Fe-MIL-101-NH with isoniazid (c), and isoniazid incorporated into Fe- MIL-101-NH (d). Arrows show the red-shift of some of the IR maxima linker may be present both as free and the intramolecular was observed (Fig. 4). These results allowed the conclusion hydrogen-bonded species even in the absence of DMF mole- that MOF can act as isoniazid carrier for extended release. cules. After INH introduction, the IR maxima characteristic Moreover, it has been also shown, that varying NH to of free –NH disappeared due to formation of new hydrogen C H linker ratio for MIL-101(Fe) it is possible to obtain con- 2 4 4 bonds with the isoniazid molecules. The INH molecules were tinuum of guest molecules binding energy states (representing concluded to interact via its carbonyl groups because this par- specific interaction with guest molecules) and in consequence ticular maximum was much wider in the spectrum of the to tune release profile. (15) composite than in the pure, crystalline isoniazid (Fig. 3b, band INH dissolution requires more extensive discussion. To −1 at 1664 cm ) and of lower frequency than expected from date there are no pharmacopoeial dissolution method for vibrations of the free C=O bond. The spectral characteristics inhalatory formulations and no single in vitro method has of isoniazid also changed considerably. The most important emerged as the ideal choice for performing dissolution tests changes were these in the ring breathing vibrations (Fig. 3b, and to estimate in vivo solubility in the lung fluids (56,57). The −1 −1 band at 995 cm ) – this band red-shifted (by 30 cm ) and its reason is that lungs have unique features that are difficult to intensity decreased. Such changes may be assigned to so- replicate in vitro, such as extremely small amount of aqueous called confinement effect, resulting in constrained breathing fluid and the presence of lung mucus and surfactant (56–59). of the aromatic ring inside the MOF pores. Similar changes, In the current study, similar setup was used as it has been −1 also in the 1800–1300 cm region, were observed upon in- previously presented, for example, by Kim et al. (39). Due to corporation of INH into the montmorillonite and saponite the physiology of the lung and the relatively low water content clays (55). From the IR results it can be deduced that INH in the respiratory tract, the 5 mL Franz cells allow in vitro molecules were located inside the pores of Fe-MIL-101-NH , approach in comparing the drug release profiles of inhalation strongly interacting by hydrogen bonding via carbonyl groups dry powder formulations (60,61). Highly viscous mucus is a with the –NH functionalities of the organic linker. Such major obstacle for particles to reach the respiratory airway strong interaction may result in the slower release of the INH. The study of total INH content by mixing in water during 120.00 12 h revealed that INH constituted 12% (SD: ± 0.8) of the 100.00 composite mass. This amount was considered as 100% of the isoniazid content in the release study. 80.00 60.00 INH Dissolution 40.00 After first 6 h of dissolution inside the Franz cell 55.0% (SD: ± 20.00 5.2) of the isoniazid content was released from the composite 0.00 powder, it reached 89.3% (SD: ± 1.2) after 24 h and 94.2% 0.00 10.00 20.00 30.00 40.00 50.00 60.00 (SD: ± 4.3) after 48 h. No burst effect was observed. % of released INH (MOF) % of released crystalline INH Dissolution of crystalline isoniazid, not incorporated in MOF structure, was much faster and after 3 h (94.6%; SD: Fig. 4 Isoniazid dissolution: Comparative in vitro dissolution (drug re- lease) profile of INH from Fe-MIL-101-NH (blue) and crystalline INH (red) ± 6.6) no significant increase in the amount of released drug % of released INH Pharm Res (2018) 35: 144 Page 7 of 11 144 (41) thus, similarly to Terzano et al. (40) in our study the echo (echo time of 21 ms) is presented. In this case, image mucin solution imitating mucus barrier was used. intensity for samples No. 2 and No. 3 was lower than for Observation made by May et al. (62) revealed that dissolu- reference (sample No. 1 – pure polymer solution) and it de- tion profiles obtained in Franz cell never reached more than creased with increasing MOF concentration. 90% of recovery rate which might be caused by not homog- Only small number of ex vivo and one in vivo study showed enous contact area to dissolution medium under membrane that regional distribution/deposition of aerosol, containing due to small air bubbles or wrinkles in the membrane. MRI contrast agents (iron oxide, Gd-DOTA) in rat lungs Similarly, in our study incomplete dissolution (~95%) can also can be successfully monitored using MRI (34,35). The results be observed for both crystalline isoniazid powder and INH of these studies suggest that the approach to combine drug incorporated in MOF. delivery with contrast agent (theranostic) is promising for such demanding application. Nuclear Magnetic Resonance Relaxometry In Vitro Viability/Cytotoxicity and Magnetic Resonance Imaging The viability of fibroblasts cultured for 24 h with both con- The spin-lattice relaxation rates (1/T or R ) and spin-spin 1 1 centration of Fe-MIL-101-NH (0.625 and 1.25 mg/mL) did relaxation rates (1/T or R ) versus concentration of MOF in 2 2 not differ between samples and between samples and the con- the suspensions are presented in Fig. 5a. The relaxation rates trol (Fig. 6). No cytotoxic effect of MOF was observed for this R and R increased linearly with the concentrations of 1 2 series. After 72 h the viability of cells cultured with MOF 2 2 suspended material (R = 0.9926 and R = 0.9948, respec- samples was significantly (p < 0.05) lower in comparison to tively). The relationships between relaxation rates and con- the control, but the number of cells was still higher than ob- centration of Fe-MIL-101-NH in the suspensions were found served after 24 h, which means that the addition of MOF did to be equal to: not inhibit their proliferation. It was also shown that micronization of MOF crystals did not influence their R ¼ r C þ i ð1Þ 1;2 1;2 1;2 cytotoxicity. Where C is the concentration of Fe-MIL-101-NH in the The morphology of L929 fibroblasts after 24 h culture with suspension, expressed in mg/mL. The values of relaxivity the addition of two concentrations of milled or milled and −1 −1 −1 r =2.4 (mg/mL) s and intercept i =0.4 s for T re- sonicated particles of Fe-MIL-101-NH is presented in Fig. 1 1 1 2 −1 −1 laxation while r = 22.6 (mg/ml) s and intercept i = 6b. In the case of 1.25 mg/mL MOF concentration, the cells 2 2 −1 0.64 s for T relaxation were found, respectively. These were eclipsed with the particles. MOF particles in lower con- results suggested that Fe-MIL-101-NH could be used as an centration (0.625 mg/mL) were mostly phagocytosed and were effective contrast agent. clearly visible inside the cells. It proved that Fe-MIL-101-NH To demonstrate the possibility to use Fe-MIL-101-NH particles could release drug inside cells. Tubercle bacilli after as an effective contrast agent, MR images were obtained reaching the alveoli are phagocytosed and accumulate in alve- using multi-echo pulse sequence with inter-echo time of olar macrophages to form tubercles. It implies that delivery of 3.5 ms and with two different repetition times, i.e. 0.7 and isoniazid directly to the cell increase the effectiveness of thera- 3 s. Therefore, two sets of images were obtained. Images py. In our study, no damage in L929 cells’ morphology (shape demonstrating positive and negative contrasts were chosen and appearance) after treatment was observed. Majority of cells from these two image sets, and are presented in Fig. 5b. had elongated shape, characteristic for fibroblast that proved Sample No. 1 was a 2% HPMC solution. Two other sam- the cells’ viability. Cytotoxicity results described above are the ples were 26.7 and 53.3 mg/ml MOF suspensions in 2% first results for Fe-MIL-101-NH – to the best of our knowledge HPMC solution, i.e. samples No. 2 and 3 respectively. such study has not been published previously. Moreover, stud- When working with short repetition time of 0.7 s at the 1st ies on MOF materials toxicity are scarce. The benefits of MOF echo (3.5 ms), positive contrast was obtained (Fig. 5b, left miniaturization, apart from their proven effectiveness in image). In this case, the image intensity for samples No. 2 cellurar uptake, defined their in vivo fate and consequently, their and No. 3 was higher than for sample No. 1 and it increased toxicity/activity (63). with MOF concentration. An image obtained at the 2nd echo Wuttke et al. conducted cytotoxicity study for MIL-100(Fe) (echo time of 7 ms) of this image set also demonstrated positive and MIL-101(Cr) nanoparticles with and without lipid (1,2- contrast compared to the reference sample, however the dif- dioleoyl-sn-glycero-3-phosphocholine) layer on human endo- ference in image intensity between samples No. 2 and No. 3 thelial cells (HUVEC and HMEC), alveolar epithelial cells was negligible (data not presented). When working with the (MLE12) and mouse alveolar macrophages (MH-S). Results long repetition time of 3 s (Fig. 5b, right image) negative con- revealed that both MOFs are well tolerated by endothelial trast was achieved. As an example, the image obtained at 6th cells whereas the MIL-100(Fe) with a lipid layer caused some 144 Page 8 of 11 Pharm Res (2018) 35: 144 Fig. 5 Nuclear Magnetic Resonance Relaxometry and Magnetic Resonance Imaging: (a) The quantitative linear correlation between relaxation rates (R ,R )and concentrationof Fe- 1 2 y = 2.4265 x + 0.4 MIL-101-NH in a suspension. (b) 2 R² = 0.9926 MR images of FeMIL-101-NH showing the examples of positive 80 (left image) and negative (right image) contrast due to differences in imaging sequence parameters (TR = 0.7 s, TE =3.5 ms andTR =3s,TE = 21 ms respectively) 0 102030405060 y = 22.598 x + 0.6389 R² = 0.9948 0 10 20 30 40 50 c, mg/ml apoptotic cell death. Alveolar epithelial cells tolerate only pulmonary, ingestion or intravenous exposure modes were lipid-coated MOF at lower doses of up to 50–100 μgmL-1. not toxic to the investigated cell lines. The examples above, Alveolar macrophages appear to be particularly sensitive to reveal that the tested MOF show differential toxicity and iron MOF, which cause pronounced induction of a cellular bioresponse in different effector cells tested, which indicate stress response. their differential suitability for specific medical purposes (11). Grall et al. (64) have recently investigated in vitro cell tox- In the study by Baati et al. (65), it has been demonstrated, that icity of Fe-MIL-100 nanoparticles and their Cr and Al ana- high doses (220 mg/kg) of Fe-MOFs (MIL-100, MIL-88A and logue nanoparticles on lung (A549 and Calu-3) and hepatic MIL-88B-4CH ) have shown no severe in vivo toxicity when (HepG2 and Hep3B) cell lines. Authors proved that administered intravenously to rats (65). R₂, 1/s R₁, 1/s Pharm Res (2018) 35: 144 Page 9 of 11 144 Fig. 6 MOF cytotoxicity: (a) Dependence of fluorescence (in relative fluorescence units) on the 3000 24h 72h concentration of Fe-MIL-101-NH contacted with L929 fibroblasts for 24 and 72 h. M – milled and MU – milled and ultrasonicated samples. C- control. (b) Photomicrographs of morphology of L929 fibroblasts after 24 h culture with addition of 0.625 mg/ml or 1.25 mg/ml of Fe- MIL-101-NH MOF. M – milled 2 0 and MU – milled and ultrasonicated M MU M MU C sample 0.625 mg/ml 1.25 mg/ml Scale Bar = 100µm Scale Bar = 100µm Fe-MIL-101-NH -M 0.625 mg/ml Fe-MIL-101-NH -M 1.25 mg/ml 2 2 Scale Bar = 100µm Scale Bar = 100µm Fe-MIL-101-NH -MU 0.625 mg/ml Fe-MIL-101-NH -MU 1.25 mg/ml Similarly to previous studies performed on various mem- properties, combined in one carrier allow to classify Fe-MIL- bers of the MIL family, the results presented in the current 101-NH as theranostic agent. study revealed the low cytotoxicity of investigated Fe-MIL- According to performed in vitro cytotoxicity study the ma- 101-NH material. It proves safety of Fe-MIL-101-NH as a terial was found to be safe. It has been observed that Fe-MIL- 2 2 potential drug carrier. 101-NH particles were accumulated in the cell cytoplasmic area and were able to release drug inside cells, which makes them promising drug delivery system for local TB therapy. It can be expected that local drug action accomplished this way CONCLUSIONS should increase therapy effectiveness, due to direct drug deliv- ery to the MTB bacilli locations and diminish clinically assessed This work shows that Fe-MIL-101-NH Metal-Organic side effects of traditional systemic drug administration. Framework can be an effective carrier for first-line anti-tuber- Presented results are the first step in development of culosis antibiotic – isoniazid. The developed material assured inhalable drug delivery system based on iron-MOF. The ob- sustained drug release in opposite to fast dissolution of crystal- tained results suggest that it will be possible to optimize flow line isoniazid powder. Additionally, magnetic resonance im- properties of the system to assure drug loaded MOF particles aging and relaxometry on phantoms of the MOF system to reach alveoli level. Incorporation of other anti-TB drugs suspended in HPMC solution proved that proposed drug de- into MOF structure seems to be promising to ensure multi- livery system based on iron-MOF can also serve as the MRI therapy and in consequence, prevent the development of contrast agent. These two features: drug delivery and imaging Relative Fluorescence Units 144 Page 10 of 11 Pharm Res (2018) 35: 144 11. Wuttke S, Zimpel A, Bein T, Braig S, Stoiber K, Vollmar A, et al. MTB bacilli resistance. Illes et al. (22,66) indicated the appli- Validating metal-organic framework nanoparticles for their cability of MOF in multi-drug therapy which has proven to be Nanosafety in diverse biomedical applications. Adv Funct Mater. more effective than single-drug therapies in cancer treatment 2017;6(2). https://doi.org/10.1002/adhm.201600818. and in tuberculosis therapy is even obligatory. 12. Huxford RC, Della Rocca J, Lin W. Metal-organic frameworks as potential drug carriers. Curr Opin Chem Biol. 2010;14(2):262–8. In the advent of feasible translation of inhalable, pulmo- 13. Wyszogrodzka G, Marszalek B, Gil B, Dorozynski P. Metal-organic nary deposition monitoring to human (34), the application of frameworks: mechanisms of antibacterial action and potential ap- MOF for extendedrelease inhalable system proposes the novel plications. Drug Discov Today. 2016;21(6):1009–18. strategy for delivery of standard antimycobacterial agents 14. Furukawa S, Reboul J, Diring S, Sumida K, Kitagawa S. combined with the monitoring of their distribution within Structuring of metal-organic frameworks at the mesoscopic/ macroscopic scale. Chem Soc Rev. 2014;43(16):5700–34. the lung tissue. 15. Dong ZY, Sun YZS, Chu J, Zhang XZ, Deng HX. Multivariate metal-organic frameworks for dialing-in the binding and program- ACKNOWLEDGMENTS AND DISCLOSURES ming the release of drug molecules. J Am Chem Soc. 2017;139(40): 14209–16. 16. Wuttke S, Lismont M, Escudero A, Rungtaweevoranit B, Parak This work was supported by the National Science Center WJ. Positioning metal-organic framework nanoparticles within the Poland [grant numbers 2014/15/B/ST5/04498 and context of drug delivery - a comparison with mesoporous silica 2016/21/N/NZ7/02663]; The IR measurements were car- nanoparticles and dendrimers. Biomaterials. 2017;123:172–83. ried out with the equipment purchased thanks to the financial 17. Brown RW, Cheng YCN, Haacke EM, Thompson MR, Venkatesan R. Magnetic resonance imaging: physical principles support of the European Regional Development Fund in the and sequence design. Hoboken: John Wiley & Sons; 2014. framework of the Polish Innovation Economy Operational 18. Laurent S, Vander Elst L, Muller RN. Iron oxide nanoparticles as Program [contract number POIG.02.01.00–12-023/08]. molecular MRI probes In: Merbach A, Helm L, Tóth E, editors. The chemistry of contrast agents in medical magnetic resonance imaging. Chinchester: John Wiley & Sons; 2013. p. 442–443. Open Access This article is distributed under the terms of the 19. Aime S, Barge A, Gianolio E, Pagliarin R, Silengo L, Tei L. High Creative Commons Attribution 4.0 International License Relaxivity contrast agents for MRI and molecular imaging. In: Bogdanov AA, Jr., Licha K, editors. Molecular Imaging. Ernst (http://creativecommons.org/licenses/by/4.0/), which per- Schering Research Foundation Workshop. 49: Springer Berlin mits unrestricted use, distribution, and reproduction in any Heidelberg; 2005. p. 99–121. medium, provided you give appropriate credit to the original 20. Horcajada P, Chalati T, Serre C, Gillet B, Sebrie C, Baati T, et al. author(s) and the source, provide a link to the Creative Porous metal-organic-framework nanoscale carriers as a potential Commons license, and indicate if changes were made. platform for drug delivery and imaging. Nat Mater. 2010;9(2):172– 21. Zimpel A, Preiss T, Roder R, Engelke H, Ingrisch M, Peller M, et al. Imparting functionality to MOF nanoparticles by external REFERENCES surface selective covalent attachment of polymers. Chem Mater. 2016;28(10):3318–26. 22. Illes B, Hirschle P, Baenert S, Cauda V, Wuttke S, Engelke H. 1. Moghimi SM, Hunter AC, Murray JC. Nanomedicine: current Exosome-coated metal-organic framework nanoparticles: an effi- status and future prospects. FASEB J. 2005;19(3):311–30. cient drug delivery platform. Chem Mater. 2017;29(19):8042–6. 2. Kim J, Piao Y, Hyeon T. Multifunctional nanostructured materials 23. Lismont M, Dreesen L, Wuttke S. Metal-organic framework nano- for multimodal imaging, and simultaneous imaging and therapy. particles in photodynamic therapy: current status and perspectives. Chem Soc Rev. 2009;38(2):372–90. Adv Funct Mater. 2017;27(14):1606314. 3. Parveen S, Misra R, Sahoo SK. Nanoparticles: a boon to drug 24. [Available from: http://www.who.int/tb/publications/global_ delivery, therapeutics, diagnostics and imaging. Nanomedicine. report/en/. 2012;8(2):147–66. 25. Gagandeep GT, Malik B, Rath G, Development GAK. 4. Formoso P, Muzzalupo R, Tavano L, De Filpo G, Nicoletta FP. Characterization of nano-fiber patch for the treatment of glauco- Nanotechnology for the environment and medicine. Mini-Rev ma. Eur J Pharm Sci. 2014;53(1):10–6. Med Chem. 2016;16(8):668–75. 26. Ormerod LP, Horsfield N. Frequency and type of reactions to 5. Ahmed N, Fessi H, Elaissari A. Theranostic applications of nano- antituberculosis drugs: observations in routine treatment. Tuber particles in cancer. Drug Discov Today. 2012;17(17–18):928–34. Lung Dis. 1996;77(1):37–42. 6. Sumer B, Gao J. Theranostic nanomedicine for cancer. Nano. 27. Gulbay BE, Gurkan OU, Yildiz OA, Onen ZP, Erkekol FO, 2008;3(2):137–40. Baccioglu A, et al. Side effects due to primary antituberculosis 7. Funkhouser J. Reinventing pharma: the theranostic revolution. drugs during the initial phase of therapy in 1149 hospitalized Curr Drug Discov. 2002;2:17–9. patient's for tuberculosis. Respir Med. 2006;100(10):1834–42. 8. Terreno E, Uggeri F, Aime S. Image guided therapy: the advent of 28. Bray PF. Isoniazid-induced acute toxic encephalopathy. theranostic agents. J Control Release. 2012;161(2):328–37. Neurology. 1984;34(5):703. 9. Picard FJ, Bergeron MG. Rapid molecular theranostics in infec- 29. Stettner M, Steinberger D, Hartmann CJ, Pabst T, Konta L, tious diseases. Drug Discov Today. 2002;7(21):1092–101. Hartung HP, et al. Isoniazid-induced polyneuropathy in a tuber- 10. Freund R, Lachelt U, Gruber T, Ruhle B, Wuttke S. culosis patient - implication for individual risk stratification with Multifunctional efficiency: extending the concept of atom economy genotyping? Brain Behav. 2015;5(8):n/a. to functional nanomaterials. ACS Nano. 2018;12(3):2094–105. Pharm Res (2018) 35: 144 Page 11 of 11 144 30. Kaur M, Garg T, Narang RK. A Review of emerging trends in the 49. Jin Z, Luan Y, Yang M, Tang J, Wang J, Gao H, et al. Imparting magnetic functionality to iron-based MIL-101 via facile Fe3O4 treatment of tuberculosis. Artif Cells Nanomed Biotechnol. 2014;44(2):478–84. nanoparticle encapsulation: an efficient and recoverable catalyst 31. Kjellsson MC, Via LE, Goh A, Weiner D, Low KM, Kern S, et al. for aerobic oxidation. RSC Adv. 2015;5(96):78962–70. Pharmacokinetic evaluation of the penetration of Antituberculosis 50. Ke F, Yuan Y-P, Qiu L-G, Shen Y-H, Xie A-J, Zhu J-F, et al. Facile agents in rabbit pulmonary lesions. Antimicrob Agents Chemother. fabrication of magnetic metal-organic framework nanocomposites 2012;56(1):446–57. for potential targeted drug delivery. J Mater Chem. 2011;21(11): 32. Hickey AJ, Durham PG, Dharmadhikari A, Nardell EA. Inhaled 3843–8. drug treatment for tuberculosis: past progress and future prospects. 51. Lebedev OI, Millange F, Serre C, Van Tendeloo G, Ferey G. First J Control Release. 2016;240:127–34. direct imaging of giant pores of the metal-organic framework MIL- 33. Pham DD, Fattal E, Tsapis N. Pulmonary drug delivery systems for 101. Chem Mater. 2005;17(26):6525–7. tuberculosis treatment. Int J Pharm. 2015;478(2):517–29. 52. Saifullah B, El Zowalaty ME, Arulselvan P, Fakurazi S, Webster 34. Wang HC, Sebrie C, Ruaud JP, Guillot G, Bouazizi-Verdier K, TJ, Geilich BM, et al. Synthesis, characterization, and efficacy of Willoquet G, et al. Aerosol deposition in the lungs of spontaneously antituberculosis isoniazid zinc aluminum-layered double hydroxide breathing rats using Gd-DOTA-based contrast agents and ultra- based nanocomposites. Int J Nanomedicine. 2016;11:3225–37. short Echo time MRI at 1.5 tesla. Magn Reson Med. 2016;75(2): 53. Hirschle P, Preiss T, Auras F, Pick A, Volkner J, Valdeperez D, 594–605. et al. Exploration of MOF nanoparticle sizes using various physical 35. Oakes JM, Scadeng M, Breen EC, Prisk GK, Darquenne C. characterization methods - is what you measure what you get? Regional distribution of aerosol deposition in rat lungs using mag- CrystEngComm. 2016;18(23):4359–68. netic resonance imaging. Ann Biomed Eng. 2013;41(5):967–78. 54. Wu B, Lin X, Ge L, Wu L, Xu T. A novel route for preparing 36. Oakes JM, Breen EC, Scadeng M, Tchantchou GS, Darquenne C. highly proton conductive membrane materials with metal-organic MRI-based measurements of aerosol deposition in the lung of frameworks. Chem Commun. 2013;49(2):143–5. healthy and elastase-treated rats. J Appl Physiol. 2014;116(12): 55. Akyuz S, Akyuz T. FT-IR and FT-Raman spectroscopic studies of 1561–8. adsorption of isoniazid by montmorillonite and saponite. Vib 37. Bauer S, Serre C, Devic T, Horcajada P, Marrot J, Ferey G, et al. Spectrosc. 2008;48(2):229–32. High-throughput assisted rationalization of the formation of metal 56. Shaji J, Shaikh M. Current development in the evaluation methods organic frameworks in the iron(III) aminoterephthalate of pulmonary drug delivery system. Indian. J Pharm Sci. solvothermal system. Inorg Chem. 2008;47(17):7568–76. 2016;78(3):294–306. 38. Hartmann M, Fischer M. Amino-functionalized basic catalysts with 57. Azarmi S, Roa W, Lobenberg R. Current perspectives in dissolu- MIL-101 structure. Micropor Mesopor Mater. 2012;164:38–43. tion testing of conventional and novel dosage forms. Int J Pharm. 39. Kim SY, Naskar D, Kundu SC, Bishop DP, Doble PA, Boddy AV, 2007;328(1):12–21. et al. Formulation of biologically-inspired silk-based drug carriers 58. Marques MRC, Loebenberg R, Almukainzi M. Simulated biolog- for pulmonary delivery targeted for lung Cancer. Sci Rep-UK. ical fluids with possible application in dissolution testing. Dissolut 2015;5:11878–91. Technol. 2011;18(3):15–28. 40. Terzano C, Allegra L, Alhaique F, Marianecci C, Carafa M. Non- 59. McConville JT, Patel N, Ditchburn N, Tobyn MJ, Staniforth JN, phospholipid vesicles for pulmonary glucocorticold delivery. Eur J Woodcock P. Use of a novel modified TSI for the evaluation of Pharm Biopharm. 2005;59(1):57–62. controlled-release aerosol formulations. I. Drug Dev Ind Pharm. 41. Carvalho CDS, Daum N, Lehr CM. Carrier interactions with the 2000;26(11):1191–8. biological barriers of the lung: advanced in vitro models and chal- 60. Adi H, Young PM, Chan HK, Salama R, Traini D. Controlled lenges for pulmonary drug delivery. Adv Drug Deliv Rev. 2014;75: release antibiotics for dry powder lung delivery. Drug Dev Ind 129–40. Pharm. 2010;36(1):119–26. 42. Duncan GA, Jung J, Hanes J, Suk JS. The mucus barrier to inhaled 61. Haghi M, Traini D, Bebawy M, Young PM. Deposition, diffusion gene therapy. Mol Ther. 2016;24(12):2043–53. and transport mechanism of dry powder microparticulate 43. Rogers DF. Physiology of airway mucus secretion and pathophys- salbutamol, at the respiratory epithelia. Mol Pharm. 2012;9(6): iology of hypersecretion. Respir Care. 2007;52(9):1134–49. 1717–26. 44. Pelfrene A, Cave MR, Wragg J, Douay F. In vitro investigations of 62. May S, Jensen B, Wolkenhauer M, Schneider M, Lehr CM. human bioaccessibility from reference materials using simulated Dissolution techniques for in vitro testing of dry powders for inha- lung fluids. Int J Environ Res Public Health. 2017;14(2):112–27. lation. Pharm Res. 2012;29(8):2157–66. 45. Son YJ, Horng M, Copley M, McConville JT. Optimization of an 63. Gimenez-Marques M, Hidalgo T, Serre C, Horcajada P. in vitro dissolution test method for inhalation formulations. Dissolut Nanostructured metal-organic frameworks and their bio-related Technol. 2010;17(2):6–13. applications. Coord Chem Rev. 2016;307:342–60. 46. Ferey G, Mellot-Draznieks C, Serre C, Millange F, Dutour J, 64. Grall R, Hidalgo T, Delic J, Garcia-Marquez A, Chevillard S, Surble S, et al. A chromium terephthalate-based solid with unusu- Horcajada P. In vitro biocompatibility of mesoporous metal (III; ally large pore volumes and surface area. Science. 2005;309(5743): Fe, Al, Cr) trimesate MOF nanocarriers. J Mater Chem B. 2040–2. 2015;3(42):8279–92. 47. Horcajada P, Serre C, Vallet-Regi M, Sebban M, Taulelle F, Ferey 65. Baati T, Njim L, Neffati F, Kerkeni A, Bouttemi M, Gref R, et al. G. Metal-organic frameworks as efficient materials for drug deliv- In depth analysis of the in vivo toxicity of nanoparticles of porous ery. Angew Chem Int Ed. 2006;45(36):5974–8. iron(III) metal-organic frameworks. Chem Sci. 2013;4(4):1597– 48. Taylor-Pashow KML, Della Rocca J, Xie Z, Tran S, Lin W. 607. Postsynthetic modifications of Iron-carboxylate nanoscale metal- 66. Illes B, Wuttke S, Engelke H. Liposome-coated Iron fumarate organic frameworks for imaging and drug delivery. J Am Chem metal-organic framework nanoparticles for combination therapy. Soc. 2009;131(40):14261–3. Nano. 2017;7(11):172–8.
– Springer Journals
Published: May 18, 2018