Characterization of aqueous dispersions of Fe3O4 nanoparticles and their biomedical applications

ARTICLE IN PRESS

Biomaterials 26 (2005) 729–738

Technical Note

Characterization of aqueous dispersions of Fe3O4 nanoparticles and their biomedical applications Fong-Yu Chenga, Chia-Hao Sua, Yu-Sheng Yanga, Chen-Sheng Yeha,*, Chiau-Yuang Tsaib, Chao-Liang Wub,*, Ming-Ting Wuc,d,e,f, Dar-Bin Shiehc,d,e,f,* a Department of Chemistry, National Cheng Kung University, Tainan 701, Taiwan, ROC Department of Biochemistry, National Cheng Kung University, Tainan 701, Taiwan, ROC c Department of Radiology, Kaohsiung Veterans General Hospital, Kaohsiung 813, Taiwan, ROC d School of Medicine, National Yang Ming University, Taipei, Taiwan, ROC e Department of Dentistry, National Cheng Kung University, Tainan 701, Taiwan, ROC f Center for Biosciences and Biotechology, National Cheng Kung University, Tainan 701, Taiwan, ROC b

Received 10 September 2003; accepted 13 March 2004

Abstract A newly developed non-polymer coated Fe3O4 nanoparticles showing well-dispersion were synthesized using Fe(II) and Fe(III) salt chemical coprecipitation with tetramethylammonium hydroxide (N(CH3)4OH) in an aqueous solution. Transmission electron microscopy (TEM), X-ray diffraction (XRD), Fourier transform infrared spectrometer (FT-IR), X-ray photoelectron spectrometer (XPS) and superconducting quantum interference measurement device (SQUID) measurements were employed to investigate the iron oxide properties. The resulting iron oxide particles were manipulated to be as small as 9 nm diameter in size. Based on FT-IR and X-ray photoelectron spectrometer results, it is suggested that the surfaces of the magnetite (Fe3O4) particles are covered with hydroxide (–OH) groups incorporated with (CH3)4N+ through electrostatic interaction. The in vitro cytotoxicity test revealed that the magnetite particles exhibited excellent biocompatibility, suggesting that they may be further explored for biomedical applications. NMR measurements revealed significantly reduced water proton relaxation times T1 and T2: The MR images of the nanoparticles in water, serum, and whole blood were investigated using a 1.5 T clinical MR imager. Significant reduction of the background medium signal was achieved in the T2-weighted and the T2 -weighted sequence especially in the serum and whole blood. Combining the advantage of MRI signal contrast, the non-polymer-coated surface chemistry for distinct bioconjugation and the homogenous nanometer size for better controlled biodistribution, these preliminary experiments demonstrated the potential of the as-synthesized magnetite material in functional molecular imaging for biomedical research and clinical diagnosis. r 2004 Elsevier Ltd. All rights reserved. Keywords: MRI; Nanoparticle; Cytotoxicity

1. Introduction Studies of nanoscale materials have captured significant scientific and industrial interest in recent years. Magnetite (Fe3O4) nanoparticles have been extensively exploited as ferrofluids. Various approaches were developed to synthesize iron oxides. [1–12]. In recent advanced nanoparticles synthesis, Sun and co-workers manipulated monodispersive Fe3O4 nanoparticles [2]. *Corresponding authors. E-mail addresses: [email protected] (C.-S. Yeh), [email protected] (M.-T. Wu), [email protected] (D.-B. Shieh). 0142-9612/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2004.03.016

Following high-temperature decomposition of an iron precursor in an organic solution phase, they were able to produce magnetite with several controllable particle diameters. Conversely, well-dispersed aqueous Fe3O4 colloids fabrication has met with very limited success. Although the Fe(II) acetate sonication method has been reported by Gedanken et al., most of the common protocols involve ferrous and ferric ions coprecipitation with aqueous NH4OH or NaOH [3–10]. In any case, severe particles aggregation accompanied with facile precipitation are the inevitable outcomes. The synthesis of biocompatible superparamagnetic materials has long been of interest in biomedical applications including magnetic resonance imaging for

ARTICLE IN PRESS 730

F.-Y. Cheng et al. / Biomaterials 26 (2005) 729–738

clinical diagnosis, magnetic drug targeting, hyperthermia anti-cancer strategy, and enzyme immobilization [11–17]. The efficacy of many medical applications may strongly rely upon generating narrow size distribution and well-dispersed nanoparticles in an aqueous solution. Iron-oxide of nanometer size presents superparamagnetic property and is ideal for MR contrast enhancement by alterations of proton relaxation in the tissue microenvironment [18,19]. However, magnetic nanoparticles without polymer coating often suffered from the aggregation problem in water or tissue fluid, which may limit in vitro magnetic-based isolation and detection strategies, as well as clinical MRI applications. Polymercoated iron-oxide particles (SPIO and USPIO) reduced aggregation problems and has been developed for various fields of clinical MR imaging [19–23]. In fact, all SPIO or USPIO MR contrast agents already approved for clinical usage nowadays, as well as most of the currently developing contrast agents, were stabilized by dextran or its derivatives. The polymer coating significantly increases their overall size and therefore may limit their tissue distribution, penetration, and metabolic clearance. Polymer-coated particles are often uptaken rapidly by the reticuloendothelial system, such as Kupffer cells of the liver. In general, the biodistribution of these polymer-based nanoparticles was mainly influenced by their size and surface chemistry [24,25]. It has been shown in the kinetic studies of the liver MR contrast agents that the particle’s hydrodynamic size may play an important role [20,26]. Larger polymer-coated SPIO particles (about 50 nm) were mainly trapped in the liver (e.g. Ferridexw, Berlex Lab., USA), while smaller sizes (about 30 nm) normally were useful for imaging the lymph node systems (e.g. Combidexw, Advanced Magnetics, Cambridge, MA; Sineremw, Laboratoire Guerbet, F). In the surface chemistry aspect, hydrophobic surface may enhance the uptake of the nanoparticles by the liver, while hydrophilic coating and surface charges may influence their retention period in the circulation and the chance to penetrate into interstitial spaces [20–27]. Although these polymer coatings are generally considered to be biocompatible, adverse reactions have also been reported [28–32]. On the other hand, only limited recent reports have investigated non-polymer coated superparamagnetic nanoparticles in MR imaging [33–36]. These particles were stabilized by citrate monomer and presented adequate performance in coronary MR angiography. In this study, we have prepared another type of non-polymer dispersive superparamagnetic ironoxide nanoparticle and demonstrated their potential as a new class of MR contrast agent for the future development. In the current study, a modified Fe(II) and Fe(III) salt coprecipitation synthesis using tetramethylammonium hydroxide (N(CH3)4OH) was used to produce well-

dispersed Fe3O4 colloidal solutions. The physical properties and the crystalline structure of the newly formed magnetite nanoparticles were characterized by Transmission electron microscopy (TEM) and X-ray diffraction (XRD). Fourier transform infrared spectrometer (FT-IR) and X-ray photoelectron spectrometer (XPS) were performed to analyze the surface characteristics of the nanoparticles. The magnetization was determined from superconducting quantum interference measurement device (SQUID) measurements. Furthermore, the in vitro cytotoxicity test and in vitro hemolysis assay were performed to evaluate the biocompatibility of the prepared Fe3O4 nanoparticles in vitro. To further investigate the potential usage of the nanoparticles in MR imaging, the T1 longitudinal and T2 transverse relaxation times were measured using the NMR spectrometer at 9.4 T. Although the ability of a given contrast agent to enhance the longitudinal and transverse relaxation rates is usually specified in terms of the in vitro dipolar relaxivity values (r1 and r2) of the agent, the observed T1 and T2 effects in MR imaging may not be directly deduced from these values alone since r1 and r2 only describe the agent’s ability to enhance the respective relaxation rates in a perfectly homogeneous medium, typically water without interactions with their local micro-environment. More complicated interactions may exist between the contrast agents and the body fluids, typically serum or blood, which may in fact dominate the final effect of the observed image property. As a result, arrays of the concentration gradient of the nanoparticles in water, serum, and human whole blood were analyzed for MR imaging to obtain a quantitative measure of the T1 and T2 under 1.5 T in a clinical MR imager (Signa, CVi, GE Medical System, Milwaukee, USA). The signal intensity alterations influenced by interactions between nanoparticles and water, or under a more physical environment of serum and blood, were analyzed using different imaging sequences.

2. Experimental section 2.1. Fe3O4 nanoparticles preparation For magnetite nanoparticles synthesis, 1 m ferric chloride hexahydrate (FeCl3
6H2O>99%) and 2 m ferrous chloride tetrahydrate (FeCl2
4H2O>99%) were prepared by dissolving iron salts in 2 m HCl solutions, respectively. In a typical experimental procedure, 10 ml of 1 m FeCl3 solution was mixed with 2.5 ml of 2 m FeCl2 solution in a flask. This solution was stirred, followed by the slow addition of 21 ml 25% (w/w) N(CH3)4OH, until a pH of 13 was reached. Vigorous stirring was continued for 20 min. The solution color could be seen to alter from orange to black, leading to a black precipitate. The

ARTICLE IN PRESS F.-Y. Cheng et al. / Biomaterials 26 (2005) 729–738

precipitated powder was isolated by applying a permanent magnet. The supernatant was discarded by decantation. Deionized water was then added to wash the precipitates. This procedure was repeated 3 times to remove excess ions and tetramethylammonium salt in the suspension. Finally, the washed precipitates were dispersed in deionized water for further investigation. 2.2. Characterization Electron micrographs of the magnetite dispersions were carried out using a drop of the sample onto a copper mesh coated with an amorphous carbon film. This mesh was then dried in a vacuum desiccator. TEM images were performed using a JEOL-JEM-1200-EX. XRD data were collected on a Shimadzu XD-D1 X-ray diffractometer employing Cu-Ka radiation ( at 30 kV and 30 mA. The surface (l ¼ 1:54056 A) properties were characterized using FT-IR and XPS studies. The iron oxide precipitates were dried and ground with KBr for IR measurements by a JASCO 200E FT-IR spectrophotometer with a resolution of 2 cm1. The XPS spectra were recorded using an Omicron ultra high vacuum system, ESCA 2000-125. All XP spectra were performed using a Mg-Ka source (12 kV and 10 mA). The binding energy scale was calibrated to 284.6 eV for the main C (1s) peak. The magnetites magnetization was carried out at room temperature using a Quantum Design MPMS-7 SQUID magnetometer. The Fe3O4 concentrations were measured by treating the iron oxides with nitric acid until complete dissolution. The solution samples were analyzed using an atomic absorption spectrometer (UNICAM Solaar M6 series). Once the cationic iron concentrations were determined, the Fe3O4 nanoparticle concentrations were calculated on the basis of the average diameter of the iron oxide particles. 2.3. The in vitro cytotoxicity evaluation To evaluate whether the Fe3O4 nanoparticles can be used as a biocompatible material suitable for biomedical applications, an MTT assay was performed. Cos-7 monkey kidney cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 2 mm l-glutamine and 50 mg/ml gentamicin at 3  104 cells/well in 24-well plates at 37 C under a humidified atmosphere of 95% air and 5% CO2. After 24 h, serial 5-fold dilutions of the Fe3O4 nanoparticles were added to cells in triplicate with final concentrations ranging from 0.92 to 23.05 mm (iron cations concentration) for 4 h. The cells were washed once with PBS and replenished with fresh medium, followed by incubation for a further 48 h. Cell viability was determined by using the 5-dimethylthiazol-2-yl-2,5diphenyl tetrazolium bromide (MTT) assay.

731

2.4. Histochemistry analysis For Prussian blue staining, using for the presence of iron, cells were fixed with 3% formaldehyde and washed with PBS, followed by the incubation with 2% potassium ferrocyanide in 6% hydrochloric acid for 30 min. After wash, they were counterstained with Giemsa solution. The specimen were then mounted and examined under a light microscope. 2.5. In vitro hemolysis assay Hemolysis assay was performed using human whole blood from a healthy donor with permission, following guidelines for studies using human specimens. The Fe3O4 nanoparticles were added to the 1 ml human whole blood stored in the vacutainer (BD Inc., USA) containing 120IU sodium heparin to a final concentration of 0, 0.001, 0.005, 0.01, 0.05 and 0.1 m (iron concentration). The tubes were gently mixed in a rotary shaker, and then incubated for 4 h at 37 C. The specimens were then centrifuged under 1200g for 10 min to collect the serum. The serum was further centrifuged under 13000g to remove the magnetites and the supernatant was analyzed for the presence of the hemoglobin by specific 454 and 528 nm spectrophotometric absorption (Coulter Gen-S, Beckman Inc., USA). 2.6. Characterization of the r1 and r2 relaxivities All NMR experiments were performed in the field of 9.4 T. The Larmor frequency for 1H is 400.13 MHz. The T1 and T2 measurements were carried out using the inversion recovery and solid echo various time delay pulse sequences, respectively. Fe3O4 samples were dissolved into deionized water and human whole blood in a volume ratio of 1:4. The 1H chemical shifts were externally referenced to d6-benzene at 7.15 ppm. To further explore the r1 and r2 relaxivity of the nanoparticles in water in 1.5 T field strength, the magnetite nanoparticles dissolved in 20 ml of deionized water to a final concentration of 4.61 mm (iron concentration) were also analyzed in a clinical MR imager (General Electronics Medical System, Milwaukee, WI) for subsequent determination of the r1 and r2 properties. 2.7. Magnetic resonance imaging To evaluate the potential of the as-synthesized Fe3O4 nanoparticles in clinical MR imaging serial 10-fold dilutions of the Fe3O4 nanoparticles were prepared in the deionized water and fetal bovine serum to the final concentrations, ranging from 101 to 1015 m. The same concentration gradients were also prepared for two commercially available MRI contrast agents for

ARTICLE IN PRESS 732

F.-Y. Cheng et al. / Biomaterials 26 (2005) 729–738

comparative observation. 1.5 ml of each preparation was loaded into an array of 2 ml Eppendorf tubes in a plastic rack. The array was embedded in a phantom which consisted of tanks of waters to allow appropriate image acquisition. The image was taken using the designed sequences by 3 acquisitions in the matrix size of 256  128 with the field of view being 260  260 mm2. The sequence parameter for (1), T1 weighed image was: fast gradient echo with repetition time (TR), 9 ms; echo time (TE), 4.2 ms; (2), T2 weighed sequence was: fast spin echo with TR/TE=2500 ms/79.4 ms, echo train length (ET)=10 ms; (3), T2 weighted image was: gradient echo with TR/TE/flip angle=150 ms/4.2 ms/ 30 . The signal intensity of each sample was determined by standard region-of-interest measurements of cross sectional image of each individual tube using the provided image quantification tool. The T1- and T2weighted imaging measurements were then further evaluated for the whole blood samples containing iron concentration gradients ranging from 0.1 to 105 m. The sequence parameter for (1), T1-weighted image was: TR/TE was 650 ms/13.9 ms, and for (2), T2-weighted image was: TR/TE/ET was 3500 ms/106.1 ms/15 ms. The image was taken in the matrix size of 256  224, with the field of view being 120  120 mm2.

(a)

Dav = 9.1 ± 2.1 nm

(b)

Dav = 6.2 ± 1.4 nm

Fig. 1. TEM imaging showing Fe3O4 nanoparticles (a) from the assynthesized colloidal suspensions and (b) after following centrifugation at 10,000 rpm.

3. Results and discussion 3.1. Preparation and characterization of iron oxides Fig. 1 shows TEM images of the iron oxide suspensions. In addition to the dispersed and wellseparated features, the prepared colloids also exhibited some degree of aggregated morphology, as shown in Fig. 1a. The particle diameter was calculated as 9.172.1 nm. Following centrifugation at 10,000 rpm for 10 min, the particles in suspensions exhibited mostly the well-dispersed appearance. The average size decreased to 6.271.4 nm, as seen in Fig. 1b. Powder XRD was carried out to identify the nanocrystalline structure of the iron oxide species. Fig. 2 displays the XRD patterns for the as-precipitated powders corresponding to Fig. 1a without a thorough centrifugation procedure. The Fe3O4 colloids with B9 nm diameters were used for the various measurements, i.e. FT-IR, XPS, SQUID, cytotoxicity, and T1 relaxation time. The XRD indicates that magnetite (Fe3O4) was the resulting material. The pH dependence of the magnetite dispersions was investigated from pH 5 to pH 13, as well. The dispersions exhibited good stability without precipitation of between pH 7 and pH 13. The ferrofluids became unstable, resulting in faster sedimentation as the pH decreased. For example, magnetite settled out of solution within 1 day at pH 5.

Fig. 2. X-ray powder diffraction patterns of the Fe3O4 nanoparticles.

FT-IR analysis was performed to characterize the surface nature of the resulting magnetite nanoparticles, as depicted in Fig. 3. An intense and broad band appeared in the region 3200–3600 cm1 region, corresponding to the O–H stretching vibration. Note that the iron oxide surfaces are readily covered with hydroxyl groups in an aqueous environment [37]. Another

ARTICLE IN PRESS F.-Y. Cheng et al. / Biomaterials 26 (2005) 729–738

733

sample holder background. In addition to the 398.8 eV band, pure (CH3)4NOH has an additional peak at 401.7 eV. Therefore, it could be proposed that the electrostatic interaction of the (CH3)4N+ with

Fe − O δ − | H contributes to the higher binding energy 402 eV for the amine adsorption mechanism, as shown in Scheme 1.

Fig. 3. FT-IR spectrum of the Fe3O4 nanoparticles.

vibrational feature at 1634 cm1 is assigned to the C–N stretching motion. The IR absorption peak of pure N(CH3)4OH gave the C–N stretching mode at 1649 cm1. This characteristic peak could be derived from tetramethylammonium hydroxide, N(CH3)4OH. This might indicate the adsorption of N(CH3)4OH on Fe3O4. To further analyze the magnetite properties, the XPS analysis was carried out. The iron band appearing at 709.5 eV was assigned to the Fe (2P3/2). The observed position is consistent with the iron assignments of the magnetites [37,38]. Fig. 4a displays the typical XPS spectrum of the O1s region. The oxygen 1s peak was deconvoluted into three spectral bands at 528.5, 529.7, and 530.9 eV. The most intense peak at 528.5 eV is attributed to the lattice oxygen (O 2 ) in the metal oxide. The 529.7 eV of binding energy is due to the hydroxide (OH) in the surface hydroxyl and the relatively small peak at 530.9 represents physically adsorbed H2O. These assignments for the O1 s components agree with previous studies on iron oxide surfaces [38, 39]. It is interesting that the XPS of the N1s (Fig. 4b) detected two speaks at 398.8 and 402 eV. This has been usually interpreted as the formation of nitrogen coordinated metal complexes [40,41]. However, tetramethylammonium hydroxide is readily dissolved to form (CH3)4N+ and OH ion species in an aqueous environment. It is unreasonable to expect an iron complex from quaternary amine cation ((CH3)4N+) chemisorption on the surface iron. We very carefully characterized the characteristic peak at 398.8 eV by measuring the sample holder used for XPS inspection and the pure (CH3)4NOH. The sample holder was filter paper made of a nylon material. The nylon component contains many amide groups. It was found that both the filter paper alone and (CH3)4NOH with filter paper measurements showed the appearance of the N1s at B398.6 eV. Hence, the nitrogen 1s peak at 398.8 eV arose from the

Fig. 4. XPS spectra of the Fe3O4 nanoparticles: (a) oxygen 1s peak; spectral curve fitting was performed with the SPECTRA presenter software (OMICRON; SPECTRA Presenter, version 7.0). The oxygen 1s peak was deconvoluted by the Gaussian function with the FWHM of 2.0 ev after Shirley background subtraction and (b) nitrogen 1s peak.

Scheme 1.

ARTICLE IN PRESS 734

F.-Y. Cheng et al. / Biomaterials 26 (2005) 729–738

One might argue the possibility of (CH3)4N+ acting on the lattice oxygen in Fe–O–Fe. From the bonding configuration point of view, normally the Fe–O has ( [42] and the bond distance at approximately 2 A structure of the magnetite displays +FeOFe ¼ 127 [37]. With the tetramethylammonium geometry, (CH3)4N+ is likely to expel toward the lattice oxygen, due to the steric effect. As aforementioned in the text, magnetite solutions began to precipitate below pH 7 by addition of HCl. Based on Scheme 1 model, it is proposed that Cl plays a role to neutralize with (CH3)4N+ leaving Fe–OH exposed. Without (CH3)4N+ coverage, ferrofluid turned out to coagulate readily. Fig. 5 shows the magnetization loop of the Fe3O4 nanoparticles measured at room temperature. The assynthesized magnetites indicate a superparamagnetic behavior, as evidenced by zero coercivity and remanance on the magnetization loop. A saturation magnetization of B40 emu/g was determined for the fine Fe3O4 particles.

3.2. Cytotoxicity and histochemical analysis Fig. 6a–d show that Prussian blue staining of the labeled Cos-7 cells contained abundant Fe3O4 nanoparticles in the cytoplasm after 4 h of incubation. The viability of cells evaluated by MTT assay was apparently unaltered upon exposure to various concentrations of Fe3O4 nanoparticles for 4 h, ranging from 0.92 to 23.05 mm of iron concentrations (Fig. 6e). According to Prussian blue staining and cell viability assay, Fe3O4 nanoparticles are generally considered to be biocompatible.

Fig. 6. Prussian blue staining of labeled Cos-7 revealed uptake in the cytoplasm for 4 h with FBS (a) no Fe3O4 nanoparticles (b) with Fe3O4 nanoparticles; without FBS (c) no Fe3O4 nanoparticles (d) with Fe3O4 nanoparticles. (e) Viability of Cos-7 monkey kidney cells exposed to the ferrofluids at various iron concentrations. Cell viability is expressed as the mean 7 S.E. of the percentage of absorbance of controls where 100% equals viability of untreated control cells.

3.3. In vitro hemolysis analysis The in vitro hemolysis test, by detecting free hemoglobin in the serum after incubation with various concentrations of the nanoparticles, indicated that significant hemolysis (0.5 g/dL) could only be detected in 0.1 m iron concentration of the nanoparticle. Other samples presented undetectable hemoglobin concentration to the instrument’s detection limit. 0.1 m concentration was below the dose required for MR contrast enhancement in whole blood. 3.4. Relaxivity (r1 and r2) measurements

Fig. 5. Magnetization curve for the Fe3O4 nanoparticles at room temperature.

Superparamagnetic nano-ferrofluids have been recognized to hold great potential in clinical diagnostic applications as the magnetic resonance (MR) imaging

ARTICLE IN PRESS

3.5. MRI measurement of the nanoparticle dispersions The signal contrast enhancement performance of the synthesized materials evaluated in clinical MR imager is illustrated in Fig. 8. Fig. 8(a) shows a typical array image of the nanoparticles and the two commercial contrast agents in a concentration gradient of serum solution taken by the T2 weighed MR sequence (Fig. 8(a)). The image was further converted into signal intensity by the provided image analysis tool for quantitative measurement (Figs. 9 and 10). The Fe3O4 nanoparticles enhanced the MR signal for water in the T2 and T2 sequences, while the T1 effect was not significant. Reduction of T2 MRI intensity was more obvious in both synthesized nanoparticles and the Resovist, compared to the Magnevist, as expected. A

735

10000

y = 1759.4e-0.0029x R2 = 0.9907 1000

100 0

100

200

400

500

600

700

1000

100

y = 971.64e-0.0829x R2 =0.9996 10 0

(B)

300

TI (msec)

(A) Signal (logarithmic scale; arbitrary unit)

contrast agents that could exhibit the ability to alter the proton relaxivity of water. Water proton relaxation for the as-synthesized Fe3O4 was performed using standard and solid state NMR spectrometer at 9.4 T. The magnetic nanoparticles strongly reduce both relaxation times (T1 and T2) with T2 significantly lower than T1: At a concentration of 0.86 mM magnetite particles (4.61 mm of iron concentration), the longitudinal relaxation time (T1) was reduced from 3000 ms for pure water to 22.88 ms, while the T2 relaxation time was reduced from 212.8 to 0.36 ms. In conclusion, the calculated r1 and r2 relaxivities of the newly synthesized nanoparticles were 9.4 s1 mm 1 and 605.5 s1 mm 1, respectively. In whole blood samples, however, the as-synthesized Fe3O4 nanoparticles have 0.03 and 2.06 s1 mm 1 for r1 and r2; respectively. The commercial ferrofluid Resovists (Schering AG, Germany) with 0.5 m of iron concentration were also analyzed for a comparative study. Resovists consists of Fe3O4 and Fe2O3 superparamagnetic iron oxides (SPIO) and is a MR contrast agent for liver imaging. With the same iron concentration (4.61 mm), those SPIO particles have 1.2 and 280.7 s1 mm 1 for r1 and r2 in water in the NMR measurements under 9.4 T field strength, respectively. The T1 and T2 property was then evaluated in a 1.5 T clinical MR imager. Fig. 7 shows the Signal intensity fitting curve for T1 and T2 measurement of the magnetite in water measured in 1.5 T clinical MR imager. T1 was calculated to be 345 ms and T2 was 12 ms. The r1 and r2 was r1 was 0.63 and 17.8 mm 1 s1, respectively. The r1 and r2 relaxivities of Resovist were shown to be 7.2 and 82 s1 mm 1, respectively, at 1.5 T in the manufacturer’s product summary. Both r1 and r2 of the synthesized nanoparticles were much shorter than Resovist. These results present the evidence that the newly prepared Fe3O4 nanoparticles strongly reduce both T1 and T2 relaxation times. Therefore, the magnetite material presented here may have great potential for clinical MR imaging as a contrast agent.

1860 - signal (logarithmic scale; arbitrary unit)

F.-Y. Cheng et al. / Biomaterials 26 (2005) 729–738

10

20

30

40

50

60

TE (msec)

Fig. 7. r1; r2 measurement of Fe3O4 nanoparticle in the water by a 1.5 T clinical MR imager. (A) inversion-recovery prepared fast gradient echo (TR=7.7 ms, TE=3.3 ms, flip angle=90 ) with a variety of inversion time from 40 to 2500 ms. The calculated T1 was 345 ms and r1 was 0.63 mm1 s1. (B) spin echo (TR=1500 ms) with a variety of echo time from TE=10–50 ms. The calculated T2 was 12 ms and r2 was 17.8 mm 1 s1.

significant decline of T2 signal was observed when serum iron concentration of the nanoparticles exceeded 1 mm. Resovist showed a significant reduction of the T2 signal intensity when the serum iron concentration exceeded 0.1 mm. Both agents performed better in the serum environment (Fig. 9a) than in water alone (Fig. 10a), which may be caused by interactions with the serum proteins. In the T2 sequence, the signal intensity was significantly reduced in the iron concentration above 0.1 mm for Resovist and 1 mm for the newly synthesized Fe3O4 nanoparticles in both serum and water (Figs. 9b and 10b). The T2 and T2 contrast enhancement by the nanoparticles in the whole blood was further evaluated in the concentration range showing significant signal alterations in the above study (0.1–105 m iron concentration). The results are presented in Fig. 8b. The signal intensity gradually dropped in the iron concentration above 103 m iron concentration in the T1 weighed sequence and above 104 m in the T2 weighed sequence. These results indicate that the

ARTICLE IN PRESS F.-Y. Cheng et al. / Biomaterials 26 (2005) 729–738

736

Fig. 8. (a) A representative image acquired from T2 weighted MR image of the synthesized Fe3O4 nanoparticle dispersions and two contrast agents in an array of concentration gradient in serum ranged from 101 to 1015 m iron concentration. (b) The MR images of the nanoparticle dispersion in an array of concentration gradient in whole blood within the range of 0.1 to 105 m iron concentration. The image was acquired in the T1 and T2 weighted image sequences as described previously.

Fe3O4 ResovistT MagnevistTM

1500

Signal intensity (T2)

Signal intensity

150

100

0

10-14

10-12

10-10

10-8

10-6

10-4

0

10-14

10-12

10-10

10-8

10-6

10-4

10-2

Sample ion concentrations (M) 800

Signal intensity (T2*)

Signal intensity (T2*)

500

(a)

80

60

40

20 0

0

(b)

1000

0

10-2

Sample ion concentrations (M)

(a)

Fe3O4 ResovistT MagnevistTM

10-14

10-12

10-10

10-8

10-6

10-4

Fig. 9. The T2 (a) and T2 (b) MR signal intensity affected by the ion concentrations of the three contrast agents in the serum measured by a 1.5 T clinical MR imager.

400

200 0

10-2

Sample ion concentrations (M)

600

0

(b)

10-14

10-12

10-10

10-8

10-6

10-4

10-2

Sample ion concentrations (M)

Fig. 10. The T2 (a) and T2 (b) MR signal intensity affected by the ion concentrations of the three contrast agents in water measured by a 1.5 T clinical MR imager.

ARTICLE IN PRESS F.-Y. Cheng et al. / Biomaterials 26 (2005) 729–738

as-synthesized nanoparticles may be potentially further developed for contrast enhanced MR imaging. Since the introduction of particular contrast agents in MRI application in 1987, most of the SPIO and Ultrasmall super paramagnetic iron oxide (USPIO) agents have been fabricated with dextran or other types of polymer coatings to achieve their dispersion status and selectivity taken up by macrophage or endothelial cells [43–45]. Their clinical applications were mostly limited to imaging of the reticuloendothelial system (RES) such as hepatic or lymph node tumors, inflammatory or other types of lesions where macrophage is accumulated. In addition, polymer coatings significantly increase the particle size, which may affect their penetration and metabolic clearance rate in the body. Only a few recent papers have reported non-polymer coated nanoparticles dispersions in MR imaging applications [33]. The nanoparticles synthesized in this study were fabricated in the water phase while maintaining their monodispersion without polymer or surfactant protection. This provided several additional advantages over those synthesized in the organic phase or those which required additional polymer protection for stabilization. First, the water environment during nanoparticle fabrication provides a more physiological and compatible reaction background than in the organic phase. Biomolecules such as antibodies, functional peptides or DNA may denature or precipitate in the organic phase and lose their activity in the extreme organic environment. Second, the nanoparticle without polymer coating could have a much smaller size for distinct biodistribution and metabolic clearance profile than the conventional polymer coated particles. Third, the non-polymer coated nanoparticles have a better orientation control of the tagged functionalized biomolecules for specific application development, compared to those polymer coated particles. Nevertheless, the in vivo bio-distribution, metabolic dynamics, and MR imaging effect of the native nanoparticles, and the bio-conjugation strategy may require further investigation for the future development of potential clinical applications.

4. Conclusion Newly formed Fe3O4 nanoparticles of 9 nm diameter were developed using ferrous and ferric ions with N(CH3)4OH. The resulting superparamagnetic magnetite exhibited a well-dispersed property. From the FT-IR and XPS analysis, the surface nature of the iron oxides was viewed as the electrostatic interaction between quaternary (CH3)4N+ cations and the surface hydroxyl groups. The Cos-7 monkey kidney cells were used for estimating the biological effect of the superparamagnetic fluids on cell viability and proliferation. No apparent cytotoxic effects were observed at various Fe3O4 doses.

737

NMR measurements of the T1 and T2 relaxation time showed the feasibility for MRI applications. The MR imaging in the water, serum, and whole blood confirmed their contrast enhancement effect in both T2 and T2 weighted sequences, especially in the presence of serum or blood. These results indicate the great potential future applications of the newly synthesized nanoparticles in functional molecular MR imaging.

Acknowledgements This work was supported by the National Science Council of the Republic of China. We thank Ms. S.Y. Hsu for the TEM and Ms. R.R. Wu for the NMR measurements at the Tainan Regional Instrument Center, National Cheng Kung University. We also gratefully acknowledge Dr. Shu-Hua Chien for her kindness in allowing us to measure XPS at the Institute of Chemistry, Academia Sinica, NanKang, Taipei, Taiwan.

References [1] Shen L, Laibinis PE, Hatton TA. Bilayer surfactant stabilized magnetic fluids: synthesis and interactions at interfaces. Langmuir 1999;15:447–53. [2] Sun S, Zeng H. Size-controlled synthesis of magnetite nanoparticles. J Am Chem Soc 2002;124:8204–5. [3] Vijayakumar R, Koltypin Y, Felner I, Gedanken A. Sonochemical synthesis and characterization of pure nanometer-sized Fe3O4 particles. Mater Sci Eng A 2000;286:101–5. [4] Knag YS, Risbud S, Rabolt JF, Stroeve P. Synthesis and characterizations of nanometer-size Fe3O4 and Fe2O3 particles. Chem Mater 1996;8:2209–11. [5] Mann S, Hannington JP. Formation of iron oxides in unilamellar vesicles. J Colloid Interface Sci 1988;122:326–35. [6] Qu S, Yang H, Ren D, Kan S, Zou G, Li D, Li M. Magnetite nanoparticles prepared by precipitation from partially reduced ferric chloride aqueous solutions. J Colloid Interface Sci 1999;215:190–2. [7] Kim DK, Zhang Y, Voit W, Rao KV, Muhanned M. Synthesis and characterization of surfactant-coated superparamagnetic monodispersed iron oxide nanoparticles. J Magn Magn Mater 2001;225:30–6. [8] Fu L, Dravid VP, Johnson DL. Self-assembled(SA) bilayer molecular coating on magnetic nanoparticles. Appl Surf Sci 2001;181:173–8. [9] Zhang Y, Kohler N, Zhang M. Surface modification of superparamagnetic magnetite nanoparticles and their intracellular uptake. Biomaterials 2002;23:1553–61. [10] Harris LA, Goff JD, Carmichael AY, Riffle JS, Harburn JJ, PTGS, Saunders M. Magnetite nanoparticle dispersions stabilized with triblock copolymers. Chem Mater 2003;15:1367–77. [11] Kim DK, Zhang Y, Kehr J, Klason T, Bjelke B, Muhammed M. Characterization and MRI study of surfactant-coated superparamagnetic nanoparticles administered into the rat brain. J Magn Magn Mater 2001;225:256–61. [12] Reynolds CH, Annan N, Beshah K, Huber JH, Shaber SH, Lenkinski RE, Wortman JA. Gadolinium-loaded nanoparticles:

ARTICLE IN PRESS 738

[13]

[14]

[15]

[16]

[17]

[18] [19] [20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

F.-Y. Cheng et al. / Biomaterials 26 (2005) 729–738 new contrast agents for magnetic resonance imaging. J Am Chem Soc 2000;122:8940–5. Lubbe AS, Bergemann C, Riess H. Clinical experiences with magnetic drug targeting: a phase I study with 40 -epidoxorubicin in 14 patients with advanced solid tumors. Cancer Res 1996;56:4686–93. Bergemann C, Muller-Schulte D, Oster J, Brassard L, Lubbe AS. Magnetic ion-exchange nano- and macroparticles for medical, biochemical and molecular biological applications. J Magn Magn Mater 1999;194:45–52. Chan DCF, Kirpotin DB, Bunn P. Synthesis and evaluation of colloidal magnetic iron oxides for the site-specific radiofrequencyinduced hyperthermia of cancer. J Magn Magn Mater 1993; 122:374–8. Jordan A, Scholz R, Wust P, Schirra H, Schiestel T, Schmidt H, Felix R. Endocytosis of dextran and silan-coated magnetite nanoparticles and the effect of intracellular hyperthermia on human mammary carcinoma cells in vitro. J Magn Magn Mater 1999;194:185–96. Dyal A, Loos K, Noto M, Chang SW, Spagnoli C, Shafi KVPM, Ulman A, Cowman M, Gross RA. Activity of candida rugosalipase immobilized on Fe2O3 magnetic nanoparticles. J Am Chem Soc 2003;125:1684–5. Bean CP, Livingston JD. Superparamagnetism. J Appl Phys 1959;30:120–9. Koenig SH, Brown III RD, Relaxometry of magnetic resonance imaging contrast agents. Magn Reson Annu 1987:263–86. Lind K, Kresse M, Debus NP, Muller RH. A novel formulation for superparamagnetic iron oxide (SPIO) particles enhancing MR lymphography: comparison of physicochemical properties and the in vivo behaviour. J Drug Target 2002;10:221–30. Bellin MF, Beigelman C, Precetti-Morel S. Iron oxide-enhanced MR lymphography: initial experience. Eur J Radiol 2000;34: 257–64. Bonnemain B. Superparamagnetic agents in magnetic resonance imaging: physicochemical characteristics and clinical applications. A review. J Drug Target 1998;6:167–74. Muller RN, Vallet P, Maton F, Roch A, Goudemant JF, Vander Elst L, Gillis P, Peto S, Moiny F, Van Haverbeke Y. Recent developments in design, characterization, and understanding of MRI and MRS contrast media. Invest Radiol 1990;25(Suppl 1):S34–6. Weissleder R, Elizondo G, Wittenberg J, Rabito CA, Bengele HH, Josephson L. Ultrasmall superparamagnetic iron oxide: characterization of a new class of contrast agents for MR imaging. Radiology 1990;175:489–93. Thode K, Luck M, Schroder W, Semmler W, Blunk T, Muller RH, Kresse M. The influence of the sample preparation on plasma protein adsorption patterns on polysaccharide-stabilized iron oxide particles and N-terminal microsequencing of unknown proteins. J Drug Target 1997;5:35–43. Reimer P, Kwong KK, Weisskoff R, Cohen MS, Brady TJ, Weissleder R. Dynamic signal intensity changes in liver with superparamagnetic MR contrast agents. J Magn Reson Imaging 1992;2:177–81. Chouly C, Pouliquen D, Lucet I, Jeune JJ, Jallet P. Development of superparamagnetic nanoparticles for MRI: effect of particle size, charge and surface nature on biodistribution. J Microencapsul 1996;13:245–55. Reimer P, Balzer T. Ferucarbotran (Resovist): a new clinically approved RES-specific contrast agent for contrast-enhanced MRI

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37] [38]

[39]

[40] [41]

[42] [43]

[44] [45]

of the liver: properties, clinical development, and applications. Eur Radiol 2003;13:1266–76. Atkinson TP, Smith TF, Hunter RL. In vitro release of histamine from murine mast cells by block co-polymers composed of polyoxyethylene and polyoxypropylene. J Immunol 1988;141:1302–6. Ennis M, Lorenz W, Gerland W. Modulation of histamine release from rat peritoneal mast cells by non-cytotoxic concentrations of the detergents Cremophor El (oxethylated castor oil) and Triton X100. A possible explanation for unexpected adverse drug reactions Agents Actions 1986;18:235–8. McLachlan SJ, Morris MR, Lucas MA, Fisco RA, Eakins MN, Fowler DR, Scheetz RB, Olukotun AY. Phase I clinical evaluation of a new iron oxide MR contrast agent. J Magn Reson Imaging 1994;4:301–7. Nolte H, Carstensen H, Hertz H. VM-26 (teniposide)-induced hypersensitivity and degranulation of basophils in children. Am J Pediatr Hematol Oncol 1988;10:308–12. Taupitz M, Schnorr J, Abramjuk C, Wagner S, Pilgrimm H, Hunigen H, Hamm B. New generation of monomer-stabilized very small superparamagnetic iron oxide particles (VSOP) as contrast medium for MR angiography: preclinical results in rats and rabbits. J Magn Reson Imaging 2000;12:905–11. Schnorr J, Wagner S, Pilgrimm H, Hamm B, Taupitz M. Preclinical characterization of monomer-stabilized very small superparamagnetic iron oxide particles (VSOP) as a blood pool contrast medium for MR angiography. Acad Radiol 2002; 9(Suppl 2):S307–9. Wagner S, Schnorr J, Pilgrimm H, Hamm B, Taupitz M. Monomer-coated very small superparamagnetic iron oxide particles as contrast medium for magnetic resonance imaging: preclinical in vivo characterization. Invest Radiol 2002;37:167–77. Taupitz M, Schnorr J, Wagner S, Abramjuk C, Pilgrimm H, Kivelitz D, Schink T, Hansel J, Laub G, Hunigen H, Hamm B. Coronary MR angiography: experimental results with a monomer-stabilized blood pool contrast medium. Radiology 2002; 222:120–6. Cornell RM, Schwertmann U. The Iron Oxide: Structure, Properties, Reactions, Occurrence and Uses. VCH:1996. Jung CW. Surface properties of superparamagnetic iron oxide MR contrast agents: ferumoxides, ferumoxtran, ferumoxsil. Magn Reson Imaging 1995;13:675–91. Harvey DT, Linton RW. Chemical characterization of hydrous ferric oxides by X-ray photoelectron spectroscopy. Anal Chem 1981;53:1684–8. Finn P, Jolly WL. The nitrogen 1s binding energies of transition metal nitrosyls. Inorg Chem 1972;11:893–5. Incorvia MJ, Contarini S. X-ray photoelectron spectroscopic studies of metal/inhibitor systems: structure and bonding at the iron/amine interface. J Electrochem Soc 1989;136:2493–8. Closmann C, Krvittle E, Bridges F. An XAFS study of the crystal chemistry of Fe in orthopyroxene. Am Mineral 1996;81:1321–31. Stark DD, Weissleder R, Elizondo G, Hahn PF, Saini S, Todd LE, Wittenberg J, Ferrucci JT. Superparamagnetic iron oxide: clinical application as a contrast agent for MR imaging of the liver. Radiology 1988;168:297–301. Saini S, Ferrucci JT. Ferrite particles as an MR contrast agent. Radiology 1988;169:656. Hahn PF, Stark DD, Weissleder R, Elizondo G, Saini S, Ferrucci JT. Clinical application of superparamagnetic iron oxide to MR imaging of tissue perfusion in vascular liver tumors. Radiology 1990;174:361–6.