communications Magnetic nanoparticles DOI: 10.1002/smll.200600664
Double-Stranded RNA Polyinosinic– Polycytidylic Acid Immobilized onto g-Fe2O3 Nanoparticles by Using a Multifunctional Polymeric Linker** Mohammed Ibrahim Shukoor, Filipe Natalio, Vadim Ksenofontov, Muhammad Nawaz Tahir, Marc Eberhardt, Patrick Theato, Heinz C. Schrçder, Werner E. G. M'ller, and Wolfgang Tremel*
When the immune system goes to war against invading pathogens or the internal attack of cancer cells, it can deploy an arsenal of weapons. Regardless of the individual target, successful immunization results in the activation of adaptive immunity, which may be accomplished, in part, through stimulation of the so-called Toll-like receptors (TLRs), a family of pattern-recognition receptors that recognize structural components shared by many bacteria, viruses, and fungi.[1] One example of such a component is viral single- or double-stranded RNA, recognized by the TLR3 receptor.[2] Polyinosinic–polycytidylic acid (poly(IC)) is a synthetic double-stranded RNA (dsRNA). dsRNAs (except dsRNA stretches in RNA stem loops) are not normally found in mammalian cells but they are present in cells infected by some viruses and they can mimic viral infections. Among the most potent dsRNAs is the synthetic dsRNA poly(IC), which consists of a pair of strands of polyinosinic and polycytidylic acids. Poly(IC) shows antitumor and antiviral activity and recently entered into phase II clinical trials for patients with malignant gliomas.[3] The immobilization of biomolecules such as poly(IC) onto insoluble supports or nanoparticles is an important tool for the fabrication of functional materials or devices.[4] Superparamagnetic nanoparticles can contribute to precise [*] M. I. Shukoor, Dr. V. Ksenofontov, Dr. M. N. Tahir, Prof. W. Tremel Institut f r Anorganische Chemie und Analytische Chemie Johannes Gutenberg-Universit+t Mainz Duesbergweg 10–14, 55099 Mainz (Germany) Fax: (+ 49) 6131-3925605 E-mail:
[email protected] F. Natalio, Prof. H. C. Schrçder, Prof. W. E. G. M ller Institut f r Physiologische Chemie Johannes Gutenberg-Universit+t Mainz Duesbergweg 6, 55099 Mainz (Germany) Dr. M. Eberhardt, Dr. P. Theato Institut f r Organische Chemie Johannes Gutenberg-Universit+t Mainz Duesbergweg 10–14, 55099 Mainz (Germany) [**] We are grateful to the Materials Science Center (MWFZ) for financial support and to Prof. H. J. Schild (Institut f r Immunologie, Universit+t Mainz) for polyinosinic–polycytidylic acid samples. Supporting information for this article is available on the WWW under http://www.small-journal.com or from the author.
1374
delivery to an exact target site through the application of external magnetic fields. Their small size allows intravenous administration, without the risk of embolization, and passage through capillary vessels[5] and mucosae,[6] while affording special properties of large surface area, better solubility, and better tissue adhesion. This leads not only to applications as therapeutic drugs in noninvasive cancer treatment[7] but also to uses in clinical diagnostics,[8] for example, in the diagnosis of tumors and cardiovascular diseases,[9] or as contrast agents for magnetic resonance imaging.[10] Owing to their biocompatibility, iron oxides are the materials of choice. A variety of anchor groups, such as carboxylates,[11] phosphonates,[12] and sulphonates,[13] as well as the chelating hydroxamates[14] and catecholates,[15] have been utilized for the surface binding of iron oxides by cation complexation and for attaching functional molecules or biological entities such as nucleic acids, proteins, or cells. In this Communication we describe the binding of a double-stranded RNA (poly(IC)) to amino-functionalized g-Fe2O3. Poly(IC) contains a 5’-end phosphate group, which makes it amenable to functionalization by using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) in the presence of a primary amine by making use of phosphoramidate chemistry.[16] In order to provide functionality and stability to the nanoparticles, a multifunctional statistical copolymer containing two different functionalities[17] was used. The functionalities were 1) dopamine, serving as a robust anchor group capable of binding to many metal oxides, and 2) free amine groups for covalent binding of the phosphate groups present in the nucleic acid polymer. Ferrimagnetic g-Fe2O3 nanoparticles were synthesized by co-precipitation of ferrous and ferric ions in sodium hydroxide solution, as reported elsewhere.[18] Phase identification of the naked iron oxide nanoparticles was carried out by using X-ray powder diffraction. Whereas hematite ACHTUNGRE(a-Fe2O3) and maghemite (g-Fe2O3) can be distinguished easily, a distinction between magnetite (Fe3O4) and maghemite is less straightforward, as both adopt an inverse spinel structure. The Mçssbauer spectrum of the naked nanoparticles at room temperature is shown in Figure 1 a. The spectrum is interpreted by applying a hyperfine field-distribution model.[19] The extracted hyperfine magnetic-field distribution (Figure 1 b) reveals one peak at a hyperfine field of 477(2) kOe and two broad peaks at 435(18) and 340(31) kOe with corresponding intensities of 26, 35, and 39 %, respectively. All subspectra have an isomer shift of 0.42(12) mms 1 and no significant quadrupole splitting. The value of the hyperfine magnetic field, Hhf = 477(2) kOe, the isomer shift, and the almost-zero quadrupole splitting of the first component are unambiguously compatible with maghemite with its typical ferrimagnetic ordering at room temperature.[20] The smaller-value hyperfine magnetic fields and relatively high intensities of the other two components stem from the magnetic layers close to the surface of the nanoparticles. A decrease in particle size leads to a decrease in the hyperfine field for maghemite (and other iron oxides).[21] Maghemite is superparamagnetic at ambient temperature for particle sizes smaller than 9 nm. The hysteresis loop in Figure 1 c indicates superparamagnetic behavior
E 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2007, 3, No. 8, 1374 – 1378
Figure 2. Schematic representation of the immobilization of poly(IC) to maghemite nanoparticles by using a reactive ester polymer carrying dopamine anchor ligands for surface binding and 1,3-propyldiACHTUNGREamine ligands for poly(IC) binding. The structure of the polymer is shown at the bottom.
Figure 1. a) Room-temperature Mçssbauer spectrum of maghemite nanoparticles used for surface functionalization with the block coACHTUNGREpolymer. The room-temperature data have been fitted with one sextet, as described in the text. b) Hyperfine field distribution. c) Hysteresis loop for maghemite nanoparticles obtained at 298 K.
without hysteresis at ambient temperature. The magnetization data in Figure 1 c indicate a magnetization of 65 emu g 1 for the g-Fe2O3 nanoparticles at 40 kOe. Subsequently, the g-Fe2O3 nanoparticles were functionalized by using a multidentate functional copolymer (Figure 2) carrying catecholate groups as surface-binding ligands for the iron oxide nanoparticles and free amino groups for the attachment of the poly(IC) ligands. Iron oxide nanoparticles (10 mg) were treated with the reactive polymer (50 mg) dissolved in N,N-dimethylformamide (DMF). The reaction was carried out with vigorous mechanical stirring at 40 8C for 12 h and was followed by cooling of small 2007, 3, No. 8, 1374 – 1378
the reaction system to room temperature. To remove unbound polymer, the coated magnetic particles in the solution were extracted by a magnetic particle concentrator (Dynal MPC1-50, Dynal Biotech, France) at room temperature. The isolated magnetic nanoparticles were washed with DMF to ensure removal of the unreacted polymer and were subsequently dispersed in methyl imidazole buffer (MeIm, 0.1 m, pH 7.5). A portion of the washed magnetic particles was freeze dried for subsequent characterization. The presence of primary amine groups on the surface-bound polymer ligand permits the attachment of biomolecules through phosphoramidate chemistry. The EDC-activated poly(IC) mixture was coupled to the amine-functionalized iron oxide nanoparticles. The aliquots were immediately frozen. All experiments were carried out in RNAse-free solutions and environments. The average crystallite size of the particles with and without the functional polymer coating was estimated with transmission electron microscopy (TEM) by using a Philips 420 instrument with an acceleration voltage of 120 kV. Figure 3 a shows the TEM image of the unfunctionalized gFe2O3 nanoparticles with an average particle size of 10– 12 nm. Figure 3 b shows an apparent size difference between the unfunctionalized and functionalized nanoparticles based on the TEM image; this result suggests a thin polymer coating. When the functional polymer, tagged with the fluorescent dye 4-chloro-7-nitrobenzylurazene (NBD) in order to visualize the surface functionalization, was introduced into the maghemite solution, the dopamine moieties attached to the polymer backbone strongly bound to the surface of the maghemite nanoparticles, because the catecholate ligand
E 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
1375
communications
Figure 3. TEM images of the g-Fe2O3 nanoparticles a) before and b) after polymer functionalization.
forms a very stable surface complex due to a strong affinity to Fe3 + and other hard and highly charged metal cations. The presence of the hydroxy groups on the iron oxide nanoparticle surface was confirmed by IR spectroscopy. Figure 4 (top) shows Fourier-transform infrared (FTIR) spectra of maghemite nanoparticles before and after coating with the polymer. The strong absorption bands at 567 and 579 cm 1, observed in the infrared spectra of the unfunctionalized and surface-coated iron oxide, respectively, indicated the main phase to be maghemite in both samples. The reduced intensity of the OH band (3400–3500 cm 1) for the functionalized nanoparticles compared to that of the unfunctionalized particles indicates the substitution of surface hydroxy groups on the iron oxide by the catecholate groups of the polymer. The presence of C H stretching (2950–2850 cm 1) and bending (1475–1375 cm 1) vibrations further confirms the presence of the polymer coating on the nanoparticles. The broad band at 3300–3500 cm 1 is assigned to the stretching vibrations whereas the strong absorption at 1630 cm 1 corresponds to bending vibrations of the amino groups of the polyfunctional polymer ligand. For comparison, the UV/Vis spectra of the polymeric ligand, as-synthesized Fe2O3 nanoparticles, and functionalized nanoparticles are presented in Figure 4 (bottom). The absorption bands of the polymeric ligand at 280 and around 480 nm are due to the dopamine groups and the NBD dye, respectively.[15c] The corresponding absorptions, along with a shoulder at around 373 nm (due to the g-Fe2O3 nanoparticles), can be seen in the UV/ Vis absorption spectrum of the g-Fe2O3 nanoparticles functionalized with the polymeric ligand. The binding of the polymer to the g-Fe2O3 nanoparticles can also be proved by fluorescence microscopy. The NBD dye attached to the
1376
www.small-journal.com
Figure 4. Top) FTIR spectra of unfunctionalized (top) and polymerfunctionalized (bottom) g-Fe2O3 nanoparticles. Bottom) UV/Vis spectra of the polymer with a) fluorescent NBD-dye, b) unfunctionalized g-Fe2O3 nanoparticles, and c) g-Fe2O3 nanoparticles with surfacebound polymer/dye.
polyACHTUNGREmer shows an intense green fluorescence at 488 nm excitation (see Figure S1 in the Supporting Information). To determine the experimental conditions and the efficiency of binding of poly(IC), we used polyacrylamide gel electrophoresis (PAGE) to analyze the stability and purity of the poly(IC)-functionalized nanoparticles (Figure 5). The samples were loaded onto a 1 % agarose gel (Figure 5, top). Lanes 1–4 contain poly(IC)-functionalized iron oxide nanoparticles with different precursor concentrations. Lane 5 shows the results of a control experiment (EDC + poly(IC)), which indicates a higher migration rate of the very negatively charged poly(IC), thereby suggesting binding between both components. Lane 6 is the crude extract still containing a large amount of unbound poly(IC). The results from lane 7 (g-Fe2O3/polymer-NH2 coated + poly(IC)) indicate that no binding occurs between poly(IC) and the polymer-coated magnetic particles in the absence of coupling agent. From lane 8 (g-Fe2O3/polymer-NH2 coated, no poly(IC)), it is evident that the nanoparticles do not run into the agarose matrix. A pure poly(IC) band is shown in lane 9 for comparison. Parallel experiments carried out by
E 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2007, 3, No. 8, 1374 – 1378
1) induction of interferons, 2) broad immune-enhancing effects, 3) activation of specific enzymes, especially oligoACHTUNGREadenylate synthetase (OAS) and the p68 protein kinase (PKR), and finally 4) broad gene-regulatory actions. Further studies on the application of poly(IC) coupled to iron nanoparticles are currently in progress in order to separate the enzymes involved in the innate immunity system.
Experimental Section
Figure 5. Top) Analysis by PAGE of functionalized maghemite nanoparticles with polymer-bound poly(IC). Lanes 1–4 contain poly(IC)functionalized nanoparticles with different precursor concentrations: Lane 1: 2 mg mL 1, lane 2: 1 mg mL 1, lane 3: 0.5 mg mL 1, and lane 4: 0.2 mg mL 1. Lane 5 shows the results of a control experiment comprising EDC and poly(IC). Lane 6 displays the raw product of the reaction, which still contains a large amount of unbound poly(IC). Lane 7 shows the results of an experiment comprising g-Fe2O3/polymer-NH2 and poly(IC) in the absence of coupling agent. Lane 8 contained only the polymer-functionalized nanoparticles. A pure poly(IC) band is shown in lane 9 for comparison. Bottom) Time-course experiment: Lanes 1–9: 0 min, 30 min, 1 h, 2 h, 3 h, 4 h, 5 h, 6 h, and 7 h, respectively. Lane 10: Poly(IC) and EDC, without g-Fe2O3/polymerNH2. Lane 11: Poly(IC) only. Lane 12: Poly(IC) and g-Fe2O3/polymerNH2 without a linker. The decreasing intensity of the fractions in lanes 1–9 and the increasing tailing indicates that the g-Fe2O3/polymer-NH2–poly(IC) complex is stable only within a small timeslot.
decreasing the percentage of agarose had no effect on the migration rate of the particles (results not shown). The efficiency and stability of binding between the amine-functionalized iron oxide nanoparticles and poly(IC) was studied by electrophoresis (Figure 5, bottom). A timecourse experiment was performed by deep freezing ( 20 8C) aliquots of the same volume from the mother solution at certain time intervals. The samples were loaded onto 1 % agarose gel. We observe an increase in the intensity of bands for fractions that had longer incubation times; this results from the decay of the phosphoramidate bond between the dsRNA and the amine-functionalized g-Fe2O3 complex. It is evident that the reaction proceeds in a sharply defined timeslot, which makes this system suitable for further biological applications and efficient for drug delivery. There are at least four interrelated clinical actions of poly(IC), any of which (alone or in combination) might be responsible for its antitumor and antiviral activity. These are small 2007, 3, No. 8, 1374 – 1378
Preparation of the polymer: The poly(active ester) poly(pentafluorophenylacrylate) (PFA) was prepared as reported earlier.[17] GPC analysis of the obtained polymer (tetrahydrofuran, lightscattering detection) gave the following values: number-average molecular mass Mn = 29.7 kg mol 1; weight-average molecular mass Mw = 58.5 kg mol 1. The number of repeating units (246) is based on the Mw value. For the synthesis of the multifunctional poly(acrylamides), poly(active ester) PFA (300 mg, 1.26 mmol repeating units) was dissolved in dry DMF (3 mL). A solution of 3-hydroxytyramine hydrochloride (24 mg) in DMF (1.5 mL) and triethylamine (0.1 mL) was added and the clear mixture was stirred for 3 h at 50 8C. After that, a solution of 1-tert-butoxycarbonyl-1,4-diaminobutane (240 mg) in DMF (4 mL) was added and the solution was stirred again for 5 h at 50 8C. The solution was concentrated to about 5 mL in volume under vacuum and the polymeric ligand was precipitated by cold ethyl ether. The precipitated polymer was centrifuged and the solvent was decanted. After drying, 286 mg of a white crystalline solid were obtained. Cleavage of the tert-butoxycarbonyl group: The polymer was dissolved in CH2Cl2 (30 mL). Trifluoroacetic acid (2.0 mL) was then added. The mixture was stirred at room temperature for 2 h. After that, 2 n HCl (40 mL) was added and the phases were separated. The organic phase was extracted with 2 n HCl (2 B 20 mL). The combined aqueous phases were evaporated under vacuum and dialyzed against water until the pH value of the outer solution was 7. Finally, the product was dried under vacuum. The reaction yielded 200 mg of the polymer. Binding of poly(IC) to the functionalized g-Fe2O3 nanoparticles: The presence of primary amine groups on the polymer ligand on the surface of the iron oxide nanoparticles permitted the attachment of biomolecules through phosphoramidate chemistry. Poly(IC), obtained commercially and prepared to a final concentration of 2 mg mL 1, was bound to the maghemite nanoparticles through the amine side groups. The reaction between a primary amine and a phosphate group by using EDC in the presence of MeIm buffer (0.1 m, pH 7.5) is described elsewhere in detail.[16] In a typical experiment, EDC (38 mL, 0.013 m stock solution in MeIm buffer) was mixed with poly(IC) (10 mL, 1:10 dilution in MeIm buffer). The activated poly(IC) mixture was then coupled with the amine-functionalized iron oxide nanoparticles in solution (2 mL). The aliquots were immediately frozen. All experiments were carried out in RNAse-free solutions and environments. Physical characterization: FTIR spectra of the particles were recorded on a Mattson Instruments 2030 Galaxy spectrometer. The dsRNA immobilization on the particles was studied by using a 1 % agarose gel prepared in tris-borate EDTA (TBE) buffer. The
E 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
1377
communications run was performed at 60 V and the gel was stained with ethidium bromide (0.02 %) for 30 min. Magnetic-susceptibility measurements were performed by using a Quantum Design MPMS-XL SQUID magnetometer. 57Fe Mçssbauer spectra were measured with a constant-accelerationtype Mçssbauer spectrometer operated in a multichannel scaling mode. A 10 mCi 57Co/Rh source was employed and maintained at ambient temperature. The isomer shifts reported here are relative to metallic iron. The Mçssbauer spectra were analyzed with the computer program EFFINO[22] by using Lorentzian line shapes.
[9] [10]
[11] [12] [13]
Keywords: immobilization · iron · magnetic materials · nanoparticles · RNA
[14]
[15] [1] K. Takeda, T. Kaisho, S. Akira, Annu. Rev. Immunol. 2003, 21, 335 – 376. [2] O. Schulz, S. S. Diebold, M. Chen, T. I. NHslund, M. A. Nolte, L. Alexopolou, Y.-T. Azuma, R. A. Flavell, P. Liljestrçm, C. Reise Sousa, Nature 2005, 433, 887 – 892. [3] http://www.clinicaltrials.gov/ct/show/NCT00262730. [4] a) C. M. Niemeyer, Angew. Chem. 2001, 113, 4254 – 4287; Angew. Chem. Int. Ed. 2001, 40, 4128 – 4158; b) G. M. Whitesides, Nat. Biotechnol. 2003, 21, 1161 – 1165. [5] Anticancer Agents: Frontiers in Cancer Chemotherapy (Eds.: I. Ojima, G. D. Vite, K. H. Altmann), American Chemical Society, Washington, 2001. [6] A. T. Florence, N. Hussain, Adv. Drug Delivery Rev. 2001, 50 (Suppl. 1), S69 – 89. [7] a) R. K. Gilchrist, R. Medal, W. D. Shorey, R. C. Hanselman, J. C. Parrott, C. B. Taylor, Ann. Surg. 1957, 146, 596 – 606; b) Y. Matsumura, H. Maeda, Cancer Res. 1986, 46, 6387 – 6392; c) C. Streffer, P. Vaupel, G. M. Hahn, Biological Basis of Oncologic Thermo-therapy, Springer, Berlin, 1990; d) A. Jordan, R. Scholz, P. Wust, H. FHhling, R. Felix, J. Magn. Magn. Mater. 1999, 201, 413 – 419; e) T. Neuberger, B. Schopf, H. Hofmann, M. Hofmann, B. Von Rechenberg, J. Magn. Magn. Mater. 2005, 293, 483 – 496; f) M. Ferrari, Nat. Rev. Cancer 2005, 5, 161 – 171. [8] a) H. Gu, P.-L. Ho, K. W. T. Tsang, L. Wang, B. Xu, J. Am. Chem. Soc. 2003, 125, 15 702 – 15 703; b) R. Weissleder, K. Kelly, E. Y.
1378
www.small-journal.com
[16]
[17] [18] [19] [20]
[21] [22]
Sun, T. Shatland, L. Josephson, Nat. Biotechnol. 2005, 23, 1418 – 1423; c) E. Y. Sun, L. Josephson, K. A. Kelly, R. Weissleder, Bioconjugate Chem. 2006, 17, 109 – 113. F. A. Jaffer, R. Weissleder, Circ. Res. 2004, 94, 433 – 445. a) R. Weissleder, G. Elizondo, J. Wittenberg, C. A. Rabito, H. H. Bengele, L. Josephson, Radiology 1990, 175, 489 – 493; b) R. Lawaczeck, M. Menzel, H. Pietsch, Appl. Organomet. Chem. 2004, 18, 506 – 513. S. H. Tolbert, P. Sieger, G. D. Stucky, S. M. J. Aubin, C. C. Wu, D. N. Hendrickson, J. Am. Chem. Soc. 1997, 119, 8652 – 8661. M. Nagtegaal, R. Seshadri, J. Ensling, P. GMtlich, W. Tremel, J. Mater. Chem. 1999, 9, 1115 – 1120. M. Nagtegaal, P. Stroeve, J. Ensling, P. GMtlich, M. Schurrer, H. Voit, J. Flath, J. KHshammer, W. Knoll, W. Tremel, Chem. Eur. J. 1999, 5, 1331 – 1337. a) R. M. Cornell, U. Schwertmann, The Iron Oxides, 2nd ed., Wiley-VCH, Weinheim, Germany, 2003; b) Y.-S. Lu, J. A. Ramsay, World J. Microbiol. Biotechnol. 2003, 19, 215 – 225. a) C. Xu, K. Xu, H. Gu, R. Zheng, H. Liu, X. Zhang, Z. Guo, B. Xu, J. Am. Chem. Soc. 2004, 126, 9938 – 9939; b) H. Gu, K. Xu, Z. Yang, C. K. Chang, B. Xu, Chem. Commun. 2005, 4270 – 4272; c) M. N. Tahir, M. Eberhardt, P. Theato, S. Faiß, A. Janshoff, T. Gorelik, U. Kolb, W. Tremel, Angew. Chem. 2006, 118, 922 – 926; Angew. Chem. Int. Ed. 2006, 45, 908 – 912; d) M. N. Tahir, T. Gorelik, U. Kolb, H. C. Schrçder, W. E. G. MMller, W. Tremel, Angew. Chem. 2006, 118, 4921 – 4927; Angew. Chem. Int. Ed. 2006, 45, 4803 – 4809. a) B. C. F. Chu, G. M. Wahl, L. E. Orgel, Nucleic Acids Res. 1983, 11, 6513 – 6529; b) V. Lund, R. Schmid, D. Rickword, E. Hornes, Nucleic Acids Res. 1988, 16, 10 861 – 10 880. M. Eberhardt, R. Mruk, P. Theato, R. Zentel, Eur. Polym. J. 2005, 41, 1569 – 1575. Y. S. Kang, S. Risbud, J. F. Rabolt, P. Stroeve, Chem. Mater. 1996, 8, 2209 – 2211. D. G. Rancourt, J. Y. Ping, Nucl. Instrum. Methods. Phys. Res., Sect. B 1991, 58, 85 – 97. a) N. N. Greenwood, T. G. Gibb, Mçssbauer Spectroscopy, Chapman and Hall, London, 1971; b) R. M. Taylor, U. Schwertmann, Clay Miner. 1974, 10, 299 – 310. K. Haneda, A. H. Morrish, Phys. Lett. A 1977, 64, 259 – 262. H. Spiering, L. DePk, L. BottyPn, Hyperfine Interact. 2000, 125, 197 – 204. Received: November 24, 2006 Revised: April 10, 2007 Published online on June 21, 2007
E 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2007, 3, No. 8, 1374 – 1378
