Porous metal organic framework nanoscale carriers as a potential platform for drug delivery and imaging

Chia sẻ: Thần Linh | Ngày: | Loại File: PDF | Số trang:7

Thêm vào BST

Báo xấu

75
lượt xem 1
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

Porous metal organic framework nanoscale carriers as a potential platform for drug delivery and imaging, In the domain of health, one important challenge is the efficient delivery of drugs in the body using non-toxic nanocarriers. Most of the existing carrier materials show poor drug loading and or rapid release of the proportion of the drug that is simply adsorbed at the external surface of the nanocarrier. In this context, porous hybrid solids, with the ability to tune their structures and porosities for better drug interactions and high loadings, are well suited to serve as nanocarriers for delivery and imaging applications.

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Porous metal organic framework nanoscale carriers as a potential platform for drug delivery and imaging

ARTICLES PUBLISHED ONLINE: 13 DECEMBER 2009 | DOI: 10.1038/NMAT2608 Porous metal–organic-framework nanoscale carriers as a potential platform for drug delivery and imaging Patricia Horcajada1 *, Tamim Chalati2 , Christian Serre1 , Brigitte Gillet3 , Catherine Sebrie3 , Tarek Baati1 , Jarrod F. Eubank1 , Daniela Heurtaux1 , Pascal Clayette4 , Christine Kreuz4 , Jong-San Chang5 , Young Kyu Hwang5 , Veronique Marsaud2 , Phuong-Nhi Bories6 , Luc Cynober6 , Sophie Gil7 , Gérard Férey1 , Patrick Couvreur2 and Ruxandra Gref2 * In the domain of health, one important challenge is the efficient delivery of drugs in the body using non-toxic nanocarriers. Most of the existing carrier materials show poor drug loading (usually less than 5 wt% of the transported drug versus the carrier material) and/or rapid release of the proportion of the drug that is simply adsorbed (or anchored) at the external surface of the nanocarrier. In this context, porous hybrid solids, with the ability to tune their structures and porosities for better drug interactions and high loadings, are well suited to serve as nanocarriers for delivery and imaging applications. Here we show that specific non-toxic porous iron(III)-based metal–organic frameworks with engineered cores and surfaces, as well as imaging properties, function as superior nanocarriers for efficient controlled delivery of challenging antitumoural and retroviral drugs (that is, busulfan, azidothymidine triphosphate, doxorubicin or cidofovir) against cancer and AIDS. In addition to their high loadings, they also potentially associate therapeutics and diagnostics, thus opening the way for theranostics, or personalized patient treatments. F or nanocarriers, the requirements for ensuring an efficient administration. To circumvent these problems, the strategy of the therapy are to (1) efficiently entrap drugs with high payloads, present paper (Fig. 1) was to take advantage of the character and (2) control the release and avoid the ‘burst effect’ (important performance of suitable iron(iii) carboxylate MOFs. Their non- release within the first minutes), (3) control matrix degradation, toxic nature and potential for nanoparticle synthesis (nanoMOFs), (4) offer the possibility to easily engineer its surface to control coupled with unusually large loadings of different drugs and in vivo fate and (5) be detectable by imaging techniques. Moreover, imaging properties, make them ideal candidates for a new valuable entering a new stage of molecular medicine requires the association solution in the field of drug-delivery nanocarriers. of therapeutics and diagnostics to make personalized patient MOFs result from the assembly, exclusively by strong bonds, of treatment a reality. A step forward aims at conceiving a nanocarrier inorganic clusters and easily tunable organic linkers (carboxylates, that could serve both as drug carrier and as diagnostic agent (satisfy imidazolates or phosphonates14 ). This huge family presents high criteria (4) and (5)), to evaluate drug distribution and treatment and regular porosities (φ up to 4.7 nm; pore volume up to efficiency (theranostics). 2.3 cm3 g−1 ) enabling, for instance, the entrapment of large Currently, for delivery, some materials are being used (for amounts of greenhouse gases18 . They can show simultaneously example, liposomes, nanoemulsions, nanoparticles or micelles; hydrophilic and hydrophobic entities, as well as tunable pore size refs 1–5) but are, for the most part, unsatisfactory; better routes and connectivities, which can be adapted to the physico-chemical are therefore necessary to address the limitations. Very recently, properties of each drug and its medical application19,20 . Moreover, our group6,7 (ibuprofen storage/long time release) and those the high structural flexibility of some MOFs (refs 21, 22) enables the of R. Morris8,9 (gas delivery of NO for antithrombosis and adaptation of their porosity to the shape of the hosted molecule. vasodilatation) and Lin10–13 (imaging) introduced a new pathway We have synthesized, in biologically and environmentally by using hybrid porous solids14 (or metal–organic frameworks favourable aqueous or ethanolic medium, some non-toxic iron(iii) (MOFs)) for this purpose. However, most of the materials described carboxylate MOFs (MIL-53, MIL-88A, MIL-88Bt, MIL-89, MIL- in these publications (that is, Co-, Ni- and Cr-based MOFs) were 100 and MIL-101_NH2 ; MIL = Materials of Institut Lavoisier; not compatible with biomedical and pharmaceutical applications, refs 23–27) and have adapted the synthesis conditions to obtain and, with few exceptions10–13,15–17 , they were not engineered as these materials as nanoparticles (see Methods and Supplementary nanoparticles to enable controlled drug release by intravenous Sections S1 and S7; Figs S1–S5 and S11–S12), which were 1 Institut Lavoisier (CNRS 8180) & Institut universitaire de France, Université de Versailles, 78035 Versailles Cedex, France, 2 Faculté de Pharmacie (CNRS 8612), Université Paris-Sud, 92296 Châtenay-Malabry, France, 3 CNRS 2301, 91190 Gif-sur-Yvette France and CNRS8081, Université PARIS-Sud 91405 Orsay, France, 4 Laboratoire de Neurovirologie, SPI-BIO, CEA, 92260 Fontenay aux Roses Cedex, France, 5 Catalysis Center for Molecular Engineering, Korea Research Institute of Chemical Technology (KRICT), PO Box 107, Yusung, Daejeon 305-600, Korea, 6 Laboratoire de Biochimie—Hôpital Hôtel-Dieu—AP-HP 75004 Paris, France, 7 EA 2706, Faculté de Pharmacie, Université Paris-Sud, 92296 Châtenay-Malabry, France. *e-mail: horcajada@chimie.uvsq.fr; ruxandra.gref@u-psud.fr. 172 NATURE MATERIALS | VOL 9 | FEBRUARY 2010 | www.nature.com/naturematerials © 2010 Macmillan Publishers Limited. All rights reserved. NATURE MATERIALS DOI: 10.1038/NMAT2608 ARTICLES CORONA Biodistribution ~ 200 nm Targeting CORE Biodegradable porous iron carboxylates MIL-53 MIL-88 MIL-100 MIL-101 8Å 6–11 Å 24–29 Å 29–34 Å Controlled release of challenging drugs Busulfan Azidothimidine Doxorubicin Cidofovir triphosphate Imaging Figure 1 | Scheme of engineered core–corona porous iron carboxylates for drug delivery and imaging. 200 nm 100 nm 200 nm MIL-100 MIL-88A MIL-88A-PEG Figure 2 | Scanning electron micrographs of MIL-100 (left), MIL-88A (centre) and PEGylated MIL-88A nanoparticles (right). characterized in terms of biocompatibility, degradability and case of MIL-88A (fumarate) and MIL-100 (trimesate), a major imaging properties (Figs 1 and 2). Their efficiency as drug carriers degradation occurred after seven days of incubation at 37 ◦ C. The was tested with four challenging anticancer or antiviral drugs nanoparticles lose their crystallinity and release large quantities (busulfan (Bu), azidothymidine triphosphate (AZT-TP), cidofovir of their ligands (72 and 58 wt% of the fumaric and trimesic (CDV) and doxorubicin (doxo)), which, except the latter, could acids, respectively), indicating a reasonable in vitro degradability not be successfully entrapped using existing nanocarriers (Table 1). of the MOF nanoparticles. Interestingly, in the case of MIL- Some cosmetic molecules, such as caffeine (liporeductor), urea 88A, the degradation products, iron and fumaric acid, are (hydrating agent), benzophenone 3 and benzophenone 4 (UVA endogenous (see Supplementrary Section S7), and show low toxicity and UVB filters) were also tested. For biological applications, the values (LD50 (Fe) = 30 g kg−1 , LD50 (fumaric acid) = 10.7 g kg−1 ; nanoMOF surfaces were engineered by coating with several relevant LD50 (trimesic acid) = 8.4 g kg−1 ) and LD50 (terephthalic acid) polymers28 (see Methods); this treatment prevented aggregation > 6.4 g kg−1 (refs 29–32). of the nanoparticles but did not improve the results. Finally, the The nanoMOF cytotoxicity, studied in vitro (MTT assay; ref 33) potential of these nanoMOFs as contrast agents is reported. on mouse macrophages (see Supplementary Section S8), was low The first step of the study was to evaluate the performances (57 ± 11 µg ml−1 for MIL-88A) and comparable with that of the of the pure nanosized iron carboxylates in terms of degradability currently available nanoparticulate systems34 . Acute in vivo toxicity and cytotoxicity. Their in vitro degradation under physiological experiments were then carried out after intravenous administration conditions (see Supplementary Fig. S10) shows that, in the of nanoMOFs in Wistar female rats (see Supplementary Section S7). NATURE MATERIALS | VOL 9 | FEBRUARY 2010 | www.nature.com/naturematerials 173 © 2010 Macmillan Publishers Limited. All rights reserved. ARTICLES NATURE MATERIALS DOI: 10.1038/NMAT2608 Table 1 | Structure description, particle size, drug loading (wt%) and entrapment efficiency (below the drug loading values in parentheses, wt%) in several porous iron(III) carboxylate nanoparticles. MIL-89 MIL-88A MIL-100 MIL-101 _NH2 MIL-53 Trimesic Amino Muconic Fumaric acid terephthalic Terephthalic acid O acid HO O acid NH acid Organic linker HO HO OH O O OH HO OH O 2 O O O O HO OH O HO OH O Crystalline structure Flexibility Yes Yes No No Yes Pore size (Å) 11 6 25 (5.6) 29 (12) 8.6 29 (8.6) 34 (16) Particle size (nm) 50–100 150* 200 120 350* Bu loading (efficiency) (%) 13.4 × 3.5 9.8 8.0 25.5 14.3 O O - s O O s amphiphilic (4.2) (3.3) (31.9) (17.9) O O AZT-TP loading (efficiency) (%) O OH OH P OH O O P O OH 11.9 × 9.1 0.60 21.2 42.0 0.24 HO P - O O OH O hydrophilic (6.4) (85.5) (90.4) (2.8) N N N N NH2 CDV loading (efficiency) (%) NH2 10.8 × 7.7 14 2.6 16.1 41.9 N - O N O hydrophilic (81) (12) (46.2) (68.1) HO P O HO OH Doxorubicin loading (efficiency)(%) O OH O OH OH 15.3 × 11.9 9.1 - - - - hydrophobic (11.2) MeO O OH O O NH2 HO Ibuprofen loading (efficiency) (%) 10 × 5 33 22 - - - O hydrophobic (11.0) (7.3) OH Caffeine loading (efficiency) (%) 6.1 × 7.6 24.2 23.1 CH3 - - - N N O amphiphilic (16.5) (15.7) N N CH3 CH3 O Urea loading (efficiency) 4.1 × 3.1 69.2 63.5 (%) O - - - hydrophilic (2.1) (1.9) C H2N NH2 Benzophenone 4 loading (efficiency) (%) O OH 12.0 × 7.2 15.2 5 - - - C hydrophilic (22.8) (7.5) O CH3 SO3H Benzophenone 3 loading (efficiency)(%) 12.1 × 5.6 1.5 OH O - - - - hydrophobic (74.0) O *Bimodal distribution of sizes, with micrometric particles. 174 NATURE MATERIALS | VOL 9 | FEBRUARY 2010 | www.nature.com/naturematerials © 2010 Macmillan Publishers Limited. All rights reserved. NATURE MATERIALS DOI: 10.1038/NMAT2608 ARTICLES 100 as liposomes or polymeric nanoparticles, is not satisfactory because AZT-TP Doxo CDV loading never exceeds 5–6 wt% (ref. 38), rendering our search of efficient nanocarriers an attractive challenge. Bu was loaded in the preformed nanoMOFs by soaking in Released drug (%) saturated drug solutions (Supplementary Table S2, Fig. S18). Table 1 shows the maximum amounts of drug adsorbed in several 50 porous iron carboxylates. The Bu loading in the rigid mesoporous MIL-100 may be considered as exceptionally high (25 wt%). This result is five times higher than the best system of polymer nanoparticles (5–6 wt%; ref. 38) and 60 times higher than with liposomes (0.4 wt%; refs 37, 39). Owing to their lower pore volumes, Bu entrapment in microporous flexible structures (MIL- 0 88A, MIL-53, MIL-89) is lower than for MIL-100, but significantly 0 1 2 3 4 5 11 12 13 14 larger than for the existing materials. Consequently, the use of Time (days) porous iron carboxylates as nanocarriers could represent important progress for Bu therapy, especially because smaller amounts of Figure 3 | CDV (black), doxo (red) and AZT-TP (green) delivery under solids would be required to deliver the needed dose of this drug. simulated physiological conditions (PBS, 37 ◦ C) from MIL-100 Indeed, considering the actual intravenous dosage of Bu (Busilvex, nanoparticles. All experiments were carried out in quadruplicate. Bu solution in N ,N 0 -dimethylacetamide; ref. 40), the total amount of MIL-88A or MIL-100 to be administered would be around Three different loaded porous iron(iii) carboxylate nanoparticles 100 and 20 mg kg−1 d−1 , respectively, for four days. Moreover, were used. They were built up either from a hydrophilic aliphatic Bu-loaded nanoMOFs could avoid the use of toxic organic solvents linker, fumarate (MIL-88A), from a hydrophilic aromatic linker, (N ,N 0 -dimethylacetamide) during administration and reduce the trimesate (MIL-100), or from a hydrophobic aromatic linker, liver toxicity mentioned above (hepatic veno-occlusive disease37,39 ) tetramethylterephthalate (MIL-88Bt) (see Methods). Doses up owing to the entrapment of Bu in its molecular form within to the highest possible injectable amounts were administrated the pores. We have verified on cell culture experiments that the (220 mg kg−1 for MIL-88A and MIL-100, and 110 mg kg−1 for nanoMOFs were able to release Bu in its active form. Studies on MIL-88Bt). Different indicators (the animal behaviour, body human leukaemia and human multiple myeloma cells in culture and organ weights and serum parameters) were evaluated up have shown that Bu has the same activity whether it is in its to three months after injection (see Supplementray Section S7; free form or entrapped in the nanoMOFs (see Supplementary Figs S13 and S14). Their comparison with control groups did Section S9; Fig. S19). In the same way, we have confirmed the not show significant differences between them, except a slight total absence of cytotoxicity of the empty MIL-100 nanoparticles increase in the spleen and liver weights, attributed to the fast in the same cell lines. sequestration by the reticuloendothelial organs of the nanoMOFs In addition to alkylating agents such as Bu, nucleoside analogues not protected by a PEG (polyethylene glycol) coating. As all are also of major importance in the treatment of cancer and the body organ weights were back to normality one to three viral infections. They include the monophosphorylated form of months after injection (see Supplementary Figs S13 and S14), the antiviral phosphonate cidofovir, and the triphosphorylated the phenomenon was fully reversible. The absence of immune or form of azidothymidine, which are the active forms of these inflammatory reactions after nanoparticle administration supports anti-cytomegalovirus and anti-HIV compounds, and doxorubicin, their lack of toxicity. Moreover, the absence of activation of one of the most effective agents in the treatment of breast cancer. cytochrome P-450 suggests a direct excretion of the polyacids, However, the clinical use of nucleoside analogues is limited by their in agreement with their high polarity. Finally, in vivo subacute poor stability in biological media, often resulting in short half-lives toxicity assays were carried out by injecting up to 150 mg of and low bioavailabilities37 , as well as sometimes partial resistance MIL-88A kg−1 d−1 during four consecutive days. No significant to the drug41 . The important hydrophilic character of nucleoside toxic effects were observed up to ten days after administration (see analogues also strongly limits their intracellular penetration owing Supplementary Figs S15–S17). to their low membrane permeability42,43 . Some nanocarriers were The non-toxicity of the iron nanoMOFs, proved above, led previously developed to circumvent these inconveniences, but show us to investigate their ability to entrap anticancer and antiviral poor efficiencies together with ‘burst effects’44 . drugs. Chemotherapy indeed plays a key role in the treatment The performance of iron carboxylates therefore indicated major of cancer in children. Thanks to its efficiency, three out of four promise for the entrapment of all the above important drugs children can now be cured. Nevertheless, 25% of paediatric cancer (Table 1). In the case of AZT-TP and CDV, this was achieved patients go uncured, and chemotherapy-induced long-term side by simply soaking the preformed dried nanoMOFs in aqueous effects justify the continued development of new strategies to solutions of the drugs. Even if the concentration of the drug in fight childhood cancer. Research in paediatric oncology is now the solution was low, the active molecules could be loaded with encouraged and supported by European legislation (Paediatric high efficiency (in most cases, higher than 80%); the nanoMOFs Use Marketing Authorization, PUMA) and new international act as remarkable molecular ‘sponges’. For instance, MIL-100 organizations, such as the consortium Innovative Therapies for nanoparticles load up to 25, 21, 16 and 29 wt% of Bu, AZT-TP, CDV Children with Cancer (ITCC). and doxo, respectively. An unprecedented capacity of 42 wt% can be In this context, the amphiphilic antitumoural drug busulfan achieved for AZT-TP and CDV with MIL-101_NH2 nanoparticles (Bu) is widely used in combination high-dose chemotherapy (Table 1; Supplementary Section S10 and S11; Table S3–S5), regimes for leukaemias, especially in paediatrics, because it rep- compared with 1 wt% values reported in the literature for these resents a good alternative to total-body irradiation35,36 . However, drugs in usual nanocarriers41 . Bu possesses a poor stability in aqueous solution and an impor- A progressive release of the three active molecules (AZT-TP, tant hepatic toxicity due to its microcrystallization in the hepatic CDV and doxo) is observed using MIL-100 nanoparticles (Fig. 3), microvenous system (hepatic veno-occlusive diseases37 ). Moreover, with no ‘burst effect’. The comparison between kinetics of drug the current encapsulation of Bu in known drug nanocarriers, such delivery and the degradation profiles suggests that the delivery NATURE MATERIALS | VOL 9 | FEBRUARY 2010 | www.nature.com/naturematerials 175 © 2010 Macmillan Publishers Limited. All rights reserved. ARTICLES NATURE MATERIALS DOI: 10.1038/NMAT2608 a d CONTROL 220 mg kg¬1 MILL-88A_nano dm st li b e dm st li c f dm k s Figure 4 | Magnetic resonance images. The images were acquired with gradient echo (a, c, d, f) or spin echo (b, e) sequence of control rats (left; a–c) and rats injected with 220 mg kg−1 MIL-88A (right; d–f), in liver (a, b, d, e) and spleen (c, f) regions. 30 min after injection, product effect is observable on the liver and spleen. (dm, dorsal muscle; k. kidney; li, liver; s, spleen; st, stomach.) process is governed mainly by diffusion from the pores and/or Finally, we have investigated the potential of the nanoMOFs drug–matrix interactions and not by the MOF degradation. Indeed, as contrast agents. We first proved by Mössbauer spectroscopy the total delivery of AZT-TP occurred within 3 days, when only that the MOFs themselves (and not eventual iron oxide and/or approximately 10% of MIL-100 was degraded. Moreover, tests hydroxide degradation products) act as contrast agents. Magnetic carried out in nanoparticles with smaller pore size than the drug resonance imaging measurements have been made on Wistar female dimensions have shown very low drug capacities and ‘burst’ release rats 30 min after injection of 220, 44 and 22 mg kg−1 suspensions kinetics. This suggests that, in this last case, the drug was adsorbed of MIL-88A nanoparticles (Fig. 4 and Supplementary Section S14). only onto the external surface and not within the pores (see Both gradient echo and spin echo sequences show that the treated Supplementary Information, Fig. S20). organs are darker than the normal ones (Fig. 4d–f versus Fig. 4a–c.). The promising data obtained with AZT-TP in MIL-100 nanopar- The resulting aspects of the liver and the spleen are indeed different ticles incited us to evaluate, in vitro in human peripheral blood between control and treated rats (Supplementary Figs S21 and S22). mononuclear cells infected by HIV-1-LAI (see Supplementary Sec- Also, three months after injection, the liver and spleen returned to tion S10), the anti-HIV activity of AZT-TP. A significant anti-HIV a similar appearance to that of the untreated animals (results not activity was observed only for (AZT-TP)-charged nanoparticles shown). This is in accordance with the temporary accumulation of (about 90% inhibition of HIV replication) for a concentration of the nanoparticles in these organs, as discussed previously. 200 nM in AZT or AZT-TP. In parallel, the empty nanoparticles The favourable in vivo detection of the iron carboxylate MOF demonstrated no cytotoxic effects, even at the highest tested dose nanoparticles makes them interesting candidates for contrast (10 µg ml−1 of nanoparticles). agents, and, to the best of our knowledge, this represents the From the above results, it is clear that porous iron(iii) first example for iron-based MOFs. However, some examples of carboxylates currently represent the best nanocarriers for the drug MOFs based on Gd (ref. 12) or Mn (ref. 14) as potential contrast release of important drugs. Their unprecedented encapsulation agents have been recently reported. The efficiency of our iron- capacities apply to a large number of challenging drugs, not based nanoMOFs is directly related to their relaxivity, in other only hydrophilic (AZT-TP, CDV, urea and benzophenone 4) but words their capacity to modify the relaxation times of the water also hydrophobic (doxorubicin, ibuprofen and benzophenone 3) protons in the surrounding medium when a magnetic field is and amphiphilic (busulfan and caffeine) molecules (Table 1; see applied. The higher the quantity and the mobility of the metal Supplementary Section S13; Table S6). The adaptive internal coordinated water in the first and second coordination spheres, microenvironment (for example, amphiphilic polar metal and the higher the relaxivity. In this sense, our MOF nanoparticles non-polar linker) of the pores of this family of solids could probably possess not only paramagnetic iron atoms in their matrix, but also explain the exceptional qualities of these porous materials. an interconnected porous network filled with metal coordinated 176 NATURE MATERIALS | VOL 9 | FEBRUARY 2010 | www.nature.com/naturematerials © 2010 Macmillan Publishers Limited. All rights reserved. NATURE MATERIALS DOI: 10.1038/NMAT2608 ARTICLES In most cases, the nanoparticles’ mean diameter, determined by both scanning Table 2 | Transversal (r2) relaxivities of MIL-88A and electron microscopy and quasi-elastic light scattering investigations, was lower than MIL-100 nanoparticles, PEGylated or not, measured at 9.4 T. 200 nm, compatible with the intravenous route of administration (see Table 1). The nanoparticle size distribution of MIL-53 and MIL-88A was bimodal, probably Fe (mmol l−1 ) PEG (wt%) r2 (s−1 mM−1 ) owing to the competition between nucleation and growth during the crystallization process and to an aggregation of the particles. MIL-88 A 0.428 0 56 To control crystal growth, PEG chains with only one terminal reactive group (amino or carboxyl) were added during the course of the synthesis process (see MIL-88 A + PEG 0.364 13.6 95 Supplementary Section S2). Thus, PEG led to the formation of a superficial PEG MIL-100 0.187 0 73 ‘brush’ sterically protecting the nanoparticles from aggregation. Zeta-potential MIL-100 + PEG 0.15 13.3 92 measurements clearly indicated that neutral PEG chains were located at the surface of the nanoparticles. Zeta-potential values of uncoated MIL-100 (−14 mV) were shifted to almost neutral values (−2 mV) in the case of PEGylated MIL-100, and and/or free water molecules. Table 2 shows the relaxivity of the iron from 17 to 2 mV in the case of PEGylated MIL-88A. This is in accordance with previously reported data on PEG-coated nanoparticles2 . fumarate MOF (MIL-88A) nanoparticles under a 9.4 T magnetic Bound PEG could be removed only after particle degradation under acidic field. Relaxivity values r1 could not be measured, but r2 of MIL- conditions, supporting the fact that it was firmly bound to the nanoparticles 88A nanoparticles are of the order of 50 s−1 mM−1 , which can through coordination of its amino or carboxyl end-group with the metal centres. be considered as sufficient for in vivo use45 . The relaxivity values Indeed, when PEG with two non-reactive monomethoxy end-groups was added are related not only to the iron content, but also to the size of to the reaction mixture, a negligible surface modification occurred. Thus, PEG was successfully bound to the nanoparticles’ surface, and PEG contents up to 17 wt% the nanoparticles. The PEGylated nanoparticles showed slightly were obtained, of the same order of magnitude as those described as being sufficient higher r2 values than the non-PEGylated ones. The PEG coating to ensure ‘stealth’ properties (see Supplementary Section S2). may modify the nanoparticle relaxivities in two opposite ways46 : increasing the size of individual nanoparticles and decreasing Received 16 December 2008; accepted 11 November 2009; their aggregation. These results show that the iron-based core is published online 13 December 2009 responsible for the favourable relaxivities and imaging properties of the MOF nanoparticles. Their framework contains (1) water References 1. Peer, D. et al. Nanocarriers as an emerging platform for cancer therapy. molecules strongly coordinated to the Lewis acid metal sites, as well Nature Nanotech. 2, 751–760 (2007). as (2) free water molecules, probably in exchange with these bound 2. Couvreur, P., Gref, R., Andrieux, K. & Malvy, C. Nanotechnology for drug water molecules, diffusing through the interconnected pores. The delivery: Applications to cancer and autoimmune diseases. Prog. Solid State presence of this last type of water molecule should induce an effect Chem. 34, 231–235 (2006). on the relaxation times of the water protons, resulting in good 3. Gref, R. et al. Biodegradable long-circulating polymeric nanospheres. Science 263, 1600–1603 (1994). imaging properties. 4. Gabizon, A. Stealth liposomes and tumor targeting: One step further in the In conclusion, our porous iron carboxylate nanoMOFs have quest for the magic bullet. Clin. Cancer Res. 7, 223–225 (2001). many advantages when used as non-toxic and biocompatible 5. Sheikh Hasan, A. et al. Effect of the microencapsulation of nanoparticles on the drug nanocarriers. In terms of synthesis, they are obtained in reduction of burst release. Int. J. Pharm. 344, 53–61 (2007). aqueous or ethanolic solutions, instead of using organic solvents, 6. Horcajada, P. et al. Metal–organic frameworks as efficient materials for drug delivery. Angew. Chem. Int. Ed. 45, 5974–5978 (2006). and provide an example of what ‘green’ technology can afford 7. Horcajada, P. et al. Flexible porous metal–organic frameworks for a controlled for biomedical applications. In the biomedical sense, they act as drug delivery. J. Am. Chem. Soc. 130, 6774–6780 (2008). molecular sponges, encapsulating drugs with different polarities, 8. Hinks, N. J., McKinlay, A. C., Xiao, B., Wheatley, P. S. & Morris, R. E. sizes and functional groups by immersion in corresponding Metal organic frameworks as NO delivery materials for biological applications. solutions. This simple entrapment method has been applied to Microporous Mesoporous Mater. doi:10.1016/j.micromeso.2009.04.031 (2009) (in the press). previously challenging antitumoural and antiviral drugs, as well as 9. Xiao, B. et al. Chemically blockable transformation and ultraselective cosmetic agents. Progressive release was obtained under simulated low-pressure gas adsorption in a non-porous metal organic framework. physiological conditions. Moreover, anti-HIV activity of AZT-TP Nature Chem. 1, 289–294 (2009). loaded nanoMOFs has been proven. 10. Rieter, W. J., Pott, K. M., Taylor, K. M. L. & Lin, W. Nanoscale coordination polymers for platinum-based anticancer drug delivery. J. Am. Chem. Soc. 130, These results open new perspectives for improved treatment 11584–11585 (2008). with anticancer and antiviral drugs and for the development 11. Rieter, W. J., Taylor, K. M. L., An, H. & Lin, W. Nanoscale metal–organic of adapted formulations in paediatrics (using Bu nanoMOFs). frameworks as potential multimodal contrast enhancing agents. J. Am. Finally, the iron-based cores are endowed with good relaxivities, Chem. Soc. 128, 9024–9025 (2006). which makes these nanoparticles candidates for magnetic resonance 12. Taylor, K. M. L., Jin, A. & Lin, W. Surfactant-assisted synthesis of nanoscale gadolinium metal–organic-framework for potential multimodal imaging. imaging (contrast) agents. These complementary properties might Angew. Chem. Int. Ed. 47, 7722–7725 (2008). open new opportunities to use nanoMOFs for the eventual goal 13. Taylor, K. M. L., Rieter, W. J. & Lin, W. Manganese-based nanoscale as theranostic agents. metal–organic frameworks for magnetic resonance imaging. J. Am. Chem. Soc. 130, 14358–14359 (2008). Methods 14. Férey, G. Hybrid porous solids: Past, present, future. Chem. Soc. Rev. 37, The syntheses of nanoscale (Fig. 2) porous iron(iii) carboxylates (labelled MIL-n) 191–241 (2008). with different topologies and compositions (iron trans,trans-muconate (MIL-89; 15. Jhung, S. H. et al. Microwave synthesis of chromium terephthalate MIL-101 refs 21–24)), fumarate (MIL-88A; refs 21–24), tetramethylterephthalate (MIL-88Bt; and its benzene sorption ability. Adv. Mater. 19, 121–124 (2007). refs 21–24), terephthalate (MIL-53; ref. 25), trimesate (MIL-100; ref. 26) and 16. Hermes, S. et al. Selective nucleation and growth of metal–organic aminoterephthalate (MIL-101 _NH2 ; ref. 27) were optimized by an appropriate open framework thin films on patterned COOH/CF3 -terminated choice of the reaction conditions (conventional solvothermal or microwave self-assembled monolayers on Au(111). J. Am. Chem. Soc. 127, synthesis, solvent, additives, iron source, concentrations, energy, temperature and 13744–13745 (2005). time) (see supplementary Section S1). These porous iron(iii) carboxylates are built 17. Scherb, C. & Bein, T. Directing the structure of metal–organic frameworks by up from the assembly of either oxo-centred trimers of iron octahedra (MIL-88, oriented surface growth on an organic monolayer. Angew. Chem. Int. Ed. 47, MIL-89, MIL-100, MIL-101 _NH2 ) or chains of corner sharing octahedra (MIL-53) 5777–5779 (2008). and di- or tri-carboxylate linkers, leading either to microporous flexible solids 18. Llewellyn, P. L. et al. High uptakes of CO2 and CH4 in mesoporous (MIL-88, MIL-89, MIL-53) or mesoporous rigid frameworks (MIL-100, MIL-101 metal–organic-frameworks MIL-100 and MIL-101. Langmuir 24, _NH2 ) (Fig. 1). The structure and composition of the resulting nanoparticles were 7245–7250 (2008). analysed using X-ray powder diffraction, thermogravimetric analysis and infrared 19. Horcajada, P., Serre, C., Gref, R., Férey, G. & Couvreur, P. Nanoparticules spectroscopy. In the case of MIL-53, MIL-88A, MIL-89, MIL-100 and MIL-101 hybrides organiques inorganiques à base de carboxylates de fer, PCT _NH2 , the synthesis could be carried out in water or ethanol. applications PCT/FR2008/001366, 01 October 2008. NATURE MATERIALS | VOL 9 | FEBRUARY 2010 | www.nature.com/naturematerials 177 © 2010 Macmillan Publishers Limited. All rights reserved. ARTICLES NATURE MATERIALS DOI: 10.1038/NMAT2608 20. Horcajada, P., Serre, C., Gref, R., Férey, G. & Couvreur, P. Solides 41. Thierry, A. R. et al. Modulation of doxorubicin resistance in multidrug-resistant hybrides organique–inorganique à surface modifiée, PCT applications cells by liposomes. FASEB J. 7, 572–579 (1993). PCT/FR2008/001367, 01 October 2008. 42. Loke, S. L. et al. Characterization of oligonucleotide transport into living cells. 21. Férey, G. & Serre, C. Large breathing effects in three-dimensional porous Proc. Natl Acad. Sci. USA 86, 3474–3478 (1989). hybrid matter: Facts, analyses, rules and consequences. Chem. Soc. Rev. 38, 43. Kukhanova, M. et al. Design of anti-HIV compounds: From nucleoside 1380–1399 (2009). to nucleoside 50 -triphosphate analogs. Problems and perspectives. 22. Serre, C. et al. Role of solvent-host interactions that lead to very large swelling Curr. Pharm. Des. 6, 585–598 (2000). of hybrid frameworks. Science 315, 1828–w1831 (2007). 44. Hillaireau, H. et al. Encapsulation of antiviral nucleotide analogues 23. Serre, C. et al. A route to the synthesis of trivalent transition-metal porous azidothymidine-triphosphate and cidofovir in poly(iso-butylcyanoacrylate) carboxylates with trimeric secondary building units. Angew. Chem. Int. Ed. 43, nanocapsules. Int. J. Pharm. 324, 37–42 (2006). 6285–6289 (2004). 45. Roch, A. et al. Theory of proton relaxation induced by superparamagnetic 24. Surblé, S. et al. A new isoreticular class of metal–organic-frameworks with the particles. J. Chem. Phys. 110, 5403–5411 (1999). MIL-88 topology. Chem. Commun. 3, 284–286 (2006). 46. Muller, R. N. et al. in The Chemistry of Contrast Agents in Medical Magnetic 25. Whitfield, T. R., Wang, X., Liu, L. & Jacobson, A. J. Metal-organic frameworks Resonance Imaging (eds Merbach, A. E. & Tóth, É.) 417–435 (Wiley, 2001). based on iron oxide octahedral chains connected by benzenedicarboxylate dianions. Solid State Sci. 7, 1096–1103 (2005). Acknowledgements 26. Horcajada, P. et al. Synthesis and catalytic properties of MIL-100(Fe), an We acknowledge E. Legenre, M. Belle, F. Kani, C. Bellanger and E. Jubeli for their iron(iii) carboxylate with large pores. Chem. Commun. 27, 2820–2822 (2007). help with the experiments. We are grateful to J-M. Greneche, H. Chacun, M. Apple, 27. Bauer, S. et al. High-throughput assisted rationalization or the formation of C. Bories, H.Hillarieu, and O. David for their collaboration. We thank K. Storck, metal–organic framework in the iron(iii) aminoterephthalate solvothermal V. Huyot and R. Yousfi for their technical assistance with the AZT-TP experiments. system. Inorg. Chem. 47, 7568–7576 (2008). This work was partially supported by the CNRS, Université Paris Sud, Université de 28. Gref, R. et al. The controlled intravenous delivery of drugs using PEG-coated Versailles Saint-Quentin, EU funding through the ERC-2007-209241-BioMOFs, ERC sterically stabilized nanospheres. Adv. Drug Deliv. Rev. 16, 215–233 (1995). and KOCI through the Institutional Research Program of KRICT. KRICT’s authors 29. Sheftel, V. O. Indirect Food Additives and Polymers: Migration and Toxicology thank You-Kyong Seo for his experimental assistance. 148–154 (Lewis Publishers, 2000). 30. (2008). 31. (2008). P.H., nanoMOF synthesis, surface modification of nanoparticles, drug and cosmetic 32. (2008). encapsulation tests, toxicity assays, degradation tests,in vivo magnetic resonance imaging; 33. Mosmann, T. Rapid colourimetric assay for cellular growth and survival: C. Serre, nanoMOF synthesis, surface modification of nanoparticles, drug and cosmetic Application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, encapsulation tests, degradation tests, imaging applications; T.C., nanoMOF synthesis, 55–63 (1983). PEG modification, drug encapsulation and delivery, in vitro toxicity assays, degradation 34. Soma, C. E. et al. Drug delivery to resistance tumors: The potential of poly(alkyl tests, in vitro magnetic resonance imaging; B.G. and C. Sebrie, imaging applications; cyanoacrylate) nanoparticles. J. Control. Release 68, 283–289 (2000). T.B., in vivo toxicity assays, nanoMOF degradation tests, doxorubicin encapsulation and 35. Vassal, G. et al. Is 600 mg m−2 the appropriate dosage of busulfan in children delivery; J.F.E., nanoMOF degradation tests; D.H., synthesis of nanoparticles of MIL-101 undergoing bone marrow transplantation? Blood 79, 2475–2479 (1992). _NH2 ; P. Clayette and C.K., anti-HIV activity of MIL-100 nanoparticles; J.-S.C. and 36. Vassal, G. et al. Pharmacokinetics of high-dose busulfan in children. Y.K.H., synthesis of nanoparticles of MIL-100 and MIL-53 in water; V.M., busulfan Cancer Chemother. Pharmacol. 24, 386–390 (1989). activity tests; P.-N.B. and L.C., liver function evaluation in the in vivo toxicity assays; 37. Slattery, T. et al. Graft-rejection and toxicity following bone S.G., activity of Cyp-450 in the in vivo toxicity assays; G.F., nanoMOF synthesis, surface marrow transplantation in relation to busulfan pharmacokinetics. modification of nanoparticles; P. Couvreur, drug encapsulation and delivery, toxicity Bone Marrow Transplant. 16, 31–42 (1995). assays, surface modification of nanoparticles; R.G., drug encapsulation and delivery, 38. Layre, A., Gref, R., Richard, J., Requier, D. & Couvreur, P. Nanoparticules toxicity assays, surface modification of nanoparticles, imaging applications. polymériques composites, FR 04 07569, 7 July 2004. 39. Hassan, Z., Nilsson, C. & Hassan, M. Liposomal busulphan: Bioavailability and Additional information effect on bone marrow in mice. Bone Marrow Transplant. 22, 913–918 (1998). The authors declare no competing financial interests. Supplementary information 40. Madden, T. et al. Pharmacokinetics of once-daily IV busulfan as part of accompanies this paper on www.nature.com/naturematerials. Reprints and permissions pretransplantation preparative regimens: A comparison with an every 6-hour information is available online at http://npg.nature.com/reprintsandpermissions. dosing schedule. Biol. Blood Marrow Transplant. 13, 56–64 (2007). Correspondence and requests for materials should be addressed to P.H. or R.G. 178 NATURE MATERIALS | VOL 9 | FEBRUARY 2010 | www.nature.com/naturematerials © 2010 Macmillan Publishers Limited. All rights reserved.