- Email: [email protected]
In vitro study of magnetic particle seeding for implant assisted-magnetic drug targeting
ARTICLE IN PRESS Journal of Magnetism and Magnetic Materials 320 (2008) 2640– 2646
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm
In vitro study of magnetic particle seeding for implant assisted-magnetic drug targeting Misael O. Avile´s, Armin D. Ebner, James A. Ritter Department of Chemical Engineering, Swearingen Engineering Center, University of South Carolina, Columbia, SC 29208, USA
a r t i c l e in f o
a b s t r a c t
Article history: Received 8 February 2008 Received in revised form 12 May 2008 Available online 22 May 2008
The concept of using magnetic particles (seeds) as the implant for implant assisted-magnetic drug targeting (IA-MDT) was analyzed in vitro. Since this MDT system is being explored for use in capillaries, a highly porous (e70%), highly tortuous, cylindrical, polyethylene polymer was prepared to mimic capillary tissue, and the seeds (magnetite nanoparticles) were already fixed within. The well-dispersed seeds were used to enhance the capture of 0.87 mm diameter magnetic drug carrier particles (MDCPs) (polydivinylbenzene embedded with 24.8 wt% magnetite) under flow conditions typically found in capillary networks. The effects of the fluid velocity (0.015–0.15 cm/s), magnetic field strength (0.0–250 mT), porous polymer magnetite content (0–7 wt%) and MDCP concentration (C ¼ 5 and 50 mg/L) on the capture efficiency (CE) of the MDCPs were studied. In all cases, when the magnetic field was applied, compared to when it was not, large increases in CE resulted; the CE increased even further when the magnetite seeds were present. The CE increased with increases in the magnetic field strength, porous polymer magnetite content and MDCP concentration. It decreased only with increases in the fluid velocity. Large magnetic field strengths were not necessary to induce MDCP capture by the seeds. A few hundred mT was sufficient. Overall, this first in vitro study of the magnetic seeding concept for IAMDT was very encouraging, because it proved that magnetic particle seeds could serve as an effective implant for MDT systems, especially under conditions found in capillaries. & 2008 Elsevier B.V. All rights reserved.
Keywords: Magnetic drug targeting MDT Drug delivery High gradient magnetic separation HGMS Ferromagnetic seed Magnetic drug carrier particle MDCPs In vitro
1. Introduction Magnetic drug targeting (MDT) [1–3] is based on the use of an external magnetic field to attract and retain specially designed magnetic drug carrier particles (MDCPs) to a specific site in the body. Because the MDCPs accumulate at the desired site, the drug or active agent would necessarily be high in concentration at this site. This should improve the effectiveness of the treatment, while also reducing the dose needed to reach the therapeutic benefit, exposure of healthy tissue to the treatment regiment, and side effects. Although the concept of MDT has been around for over three decades [4,5], and despite the fact that it has shown some promise in both in vivo [6–10] and clinical [11–13] studies, wide spread acceptance of the technique is still looming. This is because traditional MDT has some inherent limitations. Typically, the magnetic force is not very strong, and it is also very short ranged. Since the magnetic force must overcome rather large hydrodynamic forces in the body, MDT applications have been limited to sites located close to the surface of the skin. Even in this most favorable situation, studies have shown that it is difficult to collect appreciable amounts of the MDCPs at the target site [10,11].
Corresponding author. Tel.: +1803 777 3590; fax: +1 803 777 8265.
E-mail address: [email protected] (J.A. Ritter). 0304-8853/$ - see front matter & 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jmmm.2008.05.022
To overcome these limitations of the traditional MDT approach, Ritter and co-workers [14–20] and others [21–24] recently touted the use of a magnetic implant placed strategically at the target site to help attract and retain the MDCPs there. They refer to this new MDT approach as implant assisted (IA) MDT. This new approach borrows notions from high gradient magnetic separation (HGMS) principles [25], wherein the magnetic force imparted on a magnetic particle (e.g., a MDCP) by a magnetic field is proportional to both the magnitude and the gradient of the magnetic field. Under the influence of an external magnetic field the implant becomes magnetized, and not only creates its own magnetic field around itself that can be greater in magnitude than the external magnetic field, but it also distorts the external magnetic field creating large gradients. Both of these changes in the local magnetic field around the implant increase the force exerted on the MDCPs that are traveling in the vicinity of the implant, which in turn increases their attraction to and retention at the site. Several theoretical and experimental studies [14–24] have shown the IA-MDT approach to be far superior to the traditional MDT approach when using implants such as wires [14,15,21,22], stents [16–18,23], and even seeds [19,24]. For a wire or stent to serve as the implant, it would be put in place most likely using a catheter. For seeds, i.e., tiny magnetic particles that serve as the implant, they would be put in place simply by injecting them transdermally in or near the target organ or tissue. In this way, the
ARTICLE IN PRESS M.O. Avile´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646
2641
sodium n-dodecyl sulfate (Alfa Aesar, 99%) were used as received. Magnetic microspheres, 0.87 mm in diameter, composed of carboxylmodified polydivinylbenzene, and containing 24.8 wt% magnetite, were obtained in a suspension from Bangs Laboratories, Inc. For atomic absorption spectroscopy measurements, concentrated nitric acid (Sigma-Aldrich, 69–70%) and 30% hydrogen peroxide (SigmaAldrich) were used as received. DI water was obtained from a MilliQ purification system.
V
2.2. Magnetic porous polymer preparation and characterization
seeds
MDCPs MDCPs B
Fig. 1. Schematic diagram depicting the magnetic particle seeding concept for implant assisted-magnetic drug targeting (IA-MDT), wherein magnetic seeds, such as magnetite particles, are magnetically collected at a target zone by the applied magnetic field B and then used to capture magnetic drug carrier particles (MDCPs) in this zone.
use of seeds is very advantageous over a wire or stent, as they would involve a less invasive technique for targeting drugs to a desired location in the body. Fig. 1 shows a schematic of this magnetic seeding concept for IA-MDT. In the presence of an externally applied magnetic field, the seeds would be injected first and collected in the capillary tissue, as described above. Due to them necessarily having a larger magnetization than the MDCPs, they would be easily retained in the tissue by the magnetic field. Once the seeds are in place and with the magnetic field still applied, the MDCPs would be injected essentially at the same location as the seeds. Any of the MDCPs that happen to travel close to these highly dispersed and magnetized seeds would be attracted to and retained by them in the same way a wire or stent would attract and retain the MDCPs. A recent theoretical study showed the feasibility of using magnetic seeds for IA-MDT [19]; however, an in vitro study has yet to be done. Therefore, the objective of this work is to provide the first experimental evidence of the feasibility of using seeds as an implant for IA-MDT. An in vitro study was carried using a porous polyethylene (PE) polymer as surrogate capillary tissue, magnetite nanoparticles as seeds, and polydivinylbenzene particles imbedded with magnetite as MDCPs. With the intent being to show that the seeds could indeed be used to collect the MDCPs over a wide range of physiologically relevant conditions, the seeds were a priori fixed in the polymer matrix in this initial study. The effects of the fluid velocity, magnetic field strength, magnetite content in the porous polymer matrix and MDCP concentration on the performance of this IA-MDT system were analyzed in terms of the capture efficiency (CE) of the MDCPs.
A porous polymer of cylindrical shape was prepared by a compression melt molding technique, followed by a salt leaching step [26]. In short, the PE powder and NaCl particles were mixed in a ratio of 1.00 g of PE to 5.37 g of NaCl to obtain a material with about 70% porosity once all the salt was removed. To this mixture, magnetite nanopowder was added to obtain a material with up to 9.0 wt% magnetite. The materials were mixed for at least 2 h, and then added to a specially designed Teflon mold with a 1 cm diameter by 5 cm long cylindrical opening. A 1 cm long by 1 cm diameter Teflon plug was placed on top of the sample inside the mold opening to form the cylindrical shape. A 1 ton manual arbor press was used to compress the sample. The mold was then heated to 170 1C for 2 h, with recompression steps after 1 and 2 h. The mold was cooled to room temperature and the sample was removed. The sample, in the form of a solid rod, was immersed in water for at least 48 h to remove all the salt. The water was changed out every hour for the first 4 h, and then every 2–6 h. The resulting porous samples were cut into 1 cm long cylindrical pieces. The cut samples were dried for 48 h under vacuum at 60 1C. Four samples were prepared, each with different magnetite contents. They are referred to as S1, S2, S3 and S4, each initially targeted to contain 0.0, 1.0, 5.0 and 9.0 wt% magnetite, respectively. The morphologies of the porous polymers were imaged by examining cross-sections of each sample in a FEI Quanta 200 environmental scanning electron microscope (SEM) operated at 30 kV. Energy dispersive X-ray (EDX) analysis was used to map the locations of the magnetite particles in the samples. The porosity of the material was estimated from the apparent geometrical density (ra) and the skeletal density (rs). The apparent geometrical density was measured using the weight of a 1 cm long by 1 cm in diameter porous polymer sample. The skeletal density was measured using a Quantachrome Ultrapycnometer 1000 with helium as the carrier gas. The porosity (e) was calculated from e ¼ 1(ra/rs) A Perkin Elmer 3300 Flame Atomic Absorption (AA) Spectrometer was used to determine the amount of iron in the polymer samples. To prepare the samples for AA analysis, 50 mg of the porous polymer were digested inside tightly capped 50 mL Teflon vials at 165 1C for 2 h using a mixture of concentrated nitric acid (3 mL) and 30% hydrogen peroxide (2 mL). The digested samples were quantitatively diluted into 50 mL volumetric flasks and filtered prior to analysis. Subsequent dilutions were performed as needed. All measurements were done in duplicate. 2.3. Magnetic characterization
2. Materials and method 2.1. Materials Ultra-high molecular weight PE powder (Sigma-Aldrich, 180 mm), sodium chloride (Sigma-Aldrich, +80 mesh, 98+%), iron(II,III) oxide (magnetite) nanopowder (Sigma-Aldrich, 20–30 nm, 98+%) and
The magnetic susceptibilities of the MDCPs and the magnetite nanopowder were measured using a Quantum Design MPMS XL SQUID magnetometer. Either the magnetite nanopowder or the MDCPs was placed in a gelatin capsule. This capsule was then placed inside a plastic straw, which was subsequently placed inside the instrument. Magnetic field sweeps were recorded between +1 and 1 T at 300 K. The very small diamagnetic
ARTICLE IN PRESS 2642
´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646 M.O. Avile
Porous Polymer with Magnetite Seeds
NdFeB
XX:XX ml/m
NdFeB
Syringe Pump
Collection Vessel Dual Permanent Magnet Fig. 2. Experimental apparatus used in the in vitro study of the magnetic particle-seeding concept for implant assisted magnetic drug targeting (IA-MDT).
contribution of the gelatin capsule was considered to be negligible.
Table 1 Porosity and magnetite contents in the polyethylene porous polymer cylinders Sample
Porosity (e)
2.4. MDCP capture experimental setup The MDCP capture experimental setup is shown in Fig. 2. It consisted of the cylindrical porous polymer containing the magnetite seeds, a dual magnet system comprised of two NdFeB permanent magnets located adjacent to the porous polymer, a 50 mL syringe to supply the suspension of MDCPs, and a KDS200 syringe pump to control the flow through the polymer matrix. A calibrated Spectronic 20+ spectrophotometer set at 650 nm was used to determine the concentration of MDCPs in suspension as a function of the suspension transmittance. The dual magnet configuration consisted of two, 0.3 T (50–50–25 mm) rectangular magnets with their field oriented perpendicular to their 50 50 mm face and with the porous polymer sample located at the center of the gap between them. The magnitude of the homogeneous magnetic field created between the two permanent magnets was adjusted by varying the distance between them. The magnetic field was measured using a Model 4048 Bell Gauss/Tesla meter. The porous polymer was encapsulated inside the flow tube by placing the 1 cm diameter by 1 cm long cylinder inside a 3 cm long piece of shrink tube, along with 4 in pieces of 1/8 in Tygon tubing at each end. A heat gun was then used to shrink the tube while taking care not to overheat the polymer to avoid melting. Once the shrink tube was reduced in size, the porous polymer was encapsulated tightly inside, along with the two pieces of Tygon tubing at each end. 2.5. MDCP capture experimental procedure For a typical experiment, a suspension of the MDCPs at a concentration of either 5 or 50 mg/L was prepared in a 100 mM sodium n-dodecyl sulfate (SDS) aqueous solution. The initial concentration of the MDCPs in the suspension was measured using the spectrophotometer. The suspension was then charged into the 50 mL syringe. A magnetic field was focused on the porous polymer. The distance between the magnets was varied to modify the magnetic field (B) from 38 to 250 mT, while keeping the porous polymer centered between the two magnets. With the syringe filled and the magnets in place, the syringe pump was used to drive the MDCP suspension at a fixed flow rate through the tube containing the porous polymer initially solution free (dry). The effluent from the tube containing the porous polymer was collected in small vials. The concentration of the MDCPs in the first 10 mL of this effluent was measured and used to determine the CE. By comparing the concentration of the MDCPs in the initial suspension (Ci) to the concentration of the MDCPs in
S1 S2 S3 S4
0.732 0.765 0.712 0.740 a
Magnetite content (wt%) Expected
AASa
0.00 0.99 4.80 9.10
0.00 0.7970.04 2.8570.15 7.0670.91
Atomic Absorption Spectroscopy.
the effluent (Cf), the CE was calculated as CE ¼
Ci Cf 100 Ci
(1)
All experiments were done in duplicate, with the average CE reported around one standard deviation error bars.
3. Results and discussion To demonstrate the concept of ferromagnetic seeding for MDT using in vitro experiments, a highly porous structure was needed to represent capillary tissue. This surrogate capillary tissue had to effectively trap or retain the magnetite seed particles uniformly throughout its structure, just like real capillary tissue would. But, it could not behave as a gross, non-magnetic filter medium for the MDCPs. After some trial and error with a few different porous materials, the porous PE cylinders proved to have both of those properties. Table 1 lists the porosity and magnetite content of the polymer samples. The e ranged between 0.712 and 0.765. This range was close to the target value of 0.70. The measured (AAS) magnetite content in each sample was 0.0, 0.79, 2.85 and 7.6 wt% for samples S1, S2, S3 and S4, respectively. These values were close to the expected values of 0.00, 0.99, 4.80 and 9.10 wt%, based on the magnetite added to the polymer prior to the compression step. The difference was most likely due to the loss of magnetite during the salt leaching step. Nevertheless, a sufficient amount of magnetite was successfully incorporated into each polymer matrix. Fig. 3 shows SEM images of samples S1, S2, S3 and S4. These materials all had large, interconnected pores that ranged in size from 50 to 300 mm. Although these pores were much larger than the diameter of typical capillaries (5 mm), the intent was never to reproduce such small pores. On the contrary, the intent was to make a highly porous, large pore material with magnetite dispersed throughout that also had a tortuous path through the pores that would allow the MDCP suspension to flow through it
ARTICLE IN PRESS M.O. Avile´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646
2643
Fig. 3. Scanning electron micrographs of the porous polymer samples S1, S2, S3 and S4 (refer to Table 1 for properties).
without much hindrance. This was achieved with these porous PE cylinders. A qualitative depiction of the magnetite dispersed throughout each cylinder was obtained by mapping the location of the iron in each sample with EDX analysis. The results are shown in Fig. 4. As expected, the iron content increased in going from sample S2 to S4, with some agglomerated magnetite particles apparent. The level of magnetite dispersion achieved in these samples was considered sufficient for testing the magnetic seeding concept. Fig. 5 shows the normalized magnetization curves (M/Ms) for the MDCPs and the magnetite seeds. The saturation magnetization (Ms) was 22.4 and 385 kA/m for the MDCPs and magnetite seeds, respectively. Indicative of the magnetite in the MDCPs being superparamagnetic, they did not show any coercive fields. In contrast, the magnetite seeds showed a small, but essentially negligible, coercive field or residual magnetization, which resulted from its soft magnetic character. These results alluded to the notion that the seeds would be retained magnetically much more easily than the MDCPs in real capillary tissue because of them having a much higher Ms, and that they could then be used to collect the much less magnetic MDCPs. The results from the MDCP capture experiments discussed below provide evidence in support of the latter supposition. Evidence in support of the former supposition will be provided in future work, where the seeds will indeed be captured in the polymer matrix instead of being a priori anchored in it. The magnetic seeding concept was demonstrated by carrying out MDCP capture experiments to determine the amount of MDCPs collected inside the porous polymer as a suspension containing them was pumped through it in the presence of a magnetic field. The variables of interest included the fluid velocity (V ¼ 0.015–0.15 cm/s), the magnetic field strength (B ¼ 0.0–250 mT), the magnetite content in the polymer samples (0–7 wt%), and the concentration of the MDCPs in the suspension (C ¼ 5 and 50 mg/L). The performance of this IA-MDT system was evaluated in terms of the CE defined in Eq. (1).
The effect of the fluid velocity on the CE is shown in Fig. 6. In this study, samples S1 and S2, a MDCP concentration of 50 mg/L, and an applied magnetic field strength of 250 mT were used. To mimic conditions found in real capillaries, the velocities were varied from 0.015 to 0.150 cm/s. A set of control experiments under the same conditions and no magnetic field were also carried out for comparison. The CEs obtained with the control experiments never exceeded 10%, showing that physical filtration was minimal. However, when the magnetic field was applied, a significant increase in the CE was observed in both samples, which revealed the strong role of the magnetic field in capturing the MDCPs. More importantly, the CE for the sample containing magnetite (S2) was statistically higher in all cases, which clearly demonstrated the positive role of the magnetite seeds. The results in Fig. 6 also showed that the CE decreased with increasing velocity, as expected. The larger the fluid velocity, the larger the hydrodynamic force over the MDCPs, the more this force dominated the fixed magnetic force. However, the important role of the magnetic seeds in overcoming the hydrodynamic force became apparent with increasing velocity, as the detrimental effects of velocity were much more pronounced for the case when the seeds were not present compared to when they were present. It is noteworthy that the significant CEs achieved by sample S1 in the absence of magnetite seeds indicated that other magnetic mechanisms were taking place and causing collection of the MDCPs. There are two plausible explanations that might explain this phenomenon. First, slight magnetic gradients created by the permanent magnets perhaps exerted an attractive force upon the MDCPs. Although the dual magnet configuration was designed to create a very uniform magnetic field in the vicinity of the polymer cylinder, some magnetic gradients were most likely present. Second, magnetic interparticle interactions between the MDCPs most likely existed. Like the magnetite seeds, the MDCPs also became magnetized in the presence of the uniform magnetic field. This most likely induced an attraction between neighboring MDCPs that lead to them agglomerating into small clusters that
ARTICLE IN PRESS 2644
´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646 M.O. Avile
1.00 0.75
Magnetite Seeds MDCP Surrogates
0.50 M/Ms
0.25 0.00 -0.25 -0.50 -0.75 -1.00 -0.25 -0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 B (T) Fig. 5. Normalized magnetization curves for the MDCPs and the magnetite seeds.
100
S1 - B = 250.0 mT S2 - B = 250.0 mT S1 - B = 0.00 mT S2 - B = 0.00 mT
90 80
Dual Magnet
CE %
70 60 50 40 30 20 10 0 0.015
0.030
0.060 0.090 Velocity (cm/s)
0.150
Fig. 6. Effect of fluid velocity on the capture efficiency (CE) of MDCPs for samples S1 (0 wt% magnetite) and S2 (0.79 wt% magnetite) at a MDCP concentration (C) of 50 mg/L, and applied magnetic field strengths (B) of 0 (control experiments) and 250 mT.
Fig. 4. Energy dispersive X-ray analysis of iron in samples: (a) S2 (0.79 wt% magnetite), (b) S3 (2.85 wt% magnetite) and (c) S4 (7.06 wt% magnetite).
became more easily captured by the slight magnetic field gradients created by the permanent magnets alone. This agglomeration between the MDCPs might have also lead to self-seeding,
in which clusters of the already retained MDCPs acted as seeds to further attract incoming MDCPs. The effect of the applied magnetic field strength (B) on the CE is shown in Fig. 7. In this study, samples S1 and S2, a MDCP concentration of 50 mg/L, a velocity of 0.06 cm/s, and magnetic field strengths of 250, 125, 70, 38 and 0 mT (control experiments) were used. A very positive effect of the magnetic field strength on the CE was observed, whether the magnetite seeds were present or not. However, the effect was more pronounced in all cases when the magnetite seeds were encased in the polymer matrix. The pronounced effect observed in Fig. 7 at low magnetic field strengths was perhaps due to the magnetic response of both the magnetite seeds and the MDCPs. As shown in Fig. 5, at low magnetic field strengths, both materials had a steep response to the applied magnetic field. Thus, even at the low magnetic field strengths, they were nearly magnetically saturated, which meant that they were imparting their greatest magnetic forces upon each other. On the other hand, it was interesting that at a magnetic field strength of 125 mT or lower, there were no significant differences in the CE for the samples without magnetite seeds (S1) as long as a magnetic field was applied. This behavior was due to having to increase the magnet-to-magnet separation to decrease
ARTICLE IN PRESS M.O. Avile´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646
100
100
S1 - v = 0.06 cm/s S2 - v = 0.06 cm/s
90
80
70
70
60
60
50 Dual Magnet
40
B = 0.0 mT - v = 0.06 cm/s B = 38 mT - v = 0.06 cm/s B = 250 mT - v = 0.06 cm/s
90
CE %
CE %
80
2645
Dual Magnet
50 40
30
30
20
20
10
10 0
0 125.0 70.0 38.0 Magnetic Field (mT)
Fig. 7. Effect of applied magnetic field strength (B) on the capture efficiency (CE) of the MDCPs for samples S1 (0 wt% magnetite) and S2 (0.79 wt% magnetite) at a MDCP concentration (C) of 50 mg/L, and a velocity of 0.06 cm/s. The control experiments were done at B of 0.0 mT.
the magnetic field strength. As the distance between the magnets increased, it caused a reduction in the magnetic field gradients created from the magnets alone. However, the CE in this case was still significant, even with a considerable decrease in the applied magnetic field strength, possibly due to magnetic interparticle interactions between the MDCPs, as discussed above. These results collectively confirmed that large magnetic field strengths were not necessary to induce MDCP capture by the seeds. They also showed that at the low magnetic field strengths, where the magnetic field gradients created by the magnets alone were minimal, the seeds were the major contributor to MDCP capture. The effect of the magnetite content inside the porous polymer on the CE is shown in Fig. 8. In this study, samples S1, S2, S3 and S4 (refer Table 1), a MDCP surrogate concentration of 50 mg/L, a velocity of 0.06 cm/s, and applied magnetic fields of 0 (control experiments), 38 and 250 mT were used. For the control experiments there were essentially no differences between the different samples, and the CE never exceeded 5%. In contrast, when the magnetic field was applied, the CE increased significantly with increasing magnetite content inside the porous polymer sample. This result was expected and indicated that the probability of capturing a MDCP increased when more seeds were present. The effect of the concentration of the MDCPs in the suspension on the CE is shown in Fig. 9. In this study, samples S1 and S2, MDCP concentrations of 50 and 5 mg/L, applied magnetic field strengths of 0 (control experiments) and 250 mT, and a velocity of 0.06 cm/s were used. The control experiments without the magnetic field showed essentially no difference between the CEs at both concentrations and between both samples. In contrast, when the magnetic field was applied, strong effects of concentration were observed, with higher CE always resulting when the seeds were present. This effect of the MDCP concentration in the suspension was probably due to magnetic interparticle interactions that induced magnetic agglomeration between them and self-seeding.
4. Conclusions The use of magnetite particles as seeds for implant assistedmagnetic drug targeting (IA-MDT) in, for example, capillary tissue, was studied in vitro. To mimic flow through actual capillary tissue, a highly porous (e70%), highly tortuous, cylindrical, PE polymer was prepared with the magnetite particles already fixed within.
S2 S3 Porous Polymer Sample
S1
0.0
S4
Fig. 8. Effect of the magnetite content in the porous polymer sample on the capture efficiency (CE) of the MDCPs at a velocity of 0.06 cm/s, a MDCP concentration (C) of 50 mg/L, and applied magnetic field strengths (B) of 0 (control experiments) and 250 mT.
100
B = 0.0 mT - C = 5 mg/L - v = 0.06 cm/s B = 0.0 mT - C = 50 mg/L - v = 0.06 cm/s B = 250 mT - C = 5 mg/L - v = 0.06 cm/s B = 250 mT - C = 50 mg/L - v = 0.06 cm/s
90 80 70 CE %
250.0
Dual Magnet
60 50 40 30 20 10 0 S1
S2 Porous Polymer Sample
Fig. 9. Effect of the MDCP concentration (C) in the suspension on the capture efficiency (CE) of the MDCPs for samples S1 (0 wt% magnetite) and S2 (0.79 wt% magnetite), at a velocity of 0.06 cm/s, and applied magnetic field strengths (B) of 0 (control experiments) and 250 mT.
The seeds were used to enhance the capture of 0.87 mm diameter MDCPs comprised of polydivinylbenzene embedded with 24.8 wt% magnetite. An experiment consisted of pumping a known volume of the MDCP suspension through a tube containing the porous polymer that was positioned between two permanent magnets. The effluent was collected and its MDCP concentration was compared to that in the initial sample to compute a CE. The effects of the fluid velocity (0.015–0.15 cm/s), magnetic field strength (0.0–250 mT), porous polymer magnetite content (0–7 wt%), and MDCP concentration (C ¼ 5 and 50 mg/L) on the CE of the MDCPs were studied. Control experiments, done in the absence of an applied magnetic field, demonstrated minimal physical filtration effects and that all MDCP capture was due to magnetic effects. In all cases, when the magnetic field was applied there were tangible increases in the CE, whether the porous polymer contained magnetite seeds or not. However, the CEs were always statistically greater when the magnetite seeds were present, indicating the
ARTICLE IN PRESS 2646
´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646 M.O. Avile
positive influence of the seeds on the capture of MDCPs. The CE increased with increases in the magnetic field strength, porous polymer magnetite content, and MDCP concentration; it decreased only with increases in the fluid velocity. In terms of the magnetic field effects, the results demonstrated that large magnetic field strengths were not necessary to induce MDCP capture by the seeds. This very positive outcome was due to both the magnetite seeds and the MDCPs becoming magnetically saturated at relatively low magnetic field strengths. This allowed them to impart their maximum attractive forces upon each other at magnetic field strengths of only a few hundred mT. It also allowed the MDCPs to magnetically agglomerate with each other making capture easier and possibly giving rise to self-seeding. This became evident from the large CEs obtained from the samples that did not contain any magnetite seeds, but that were exposed to a magnetic field. Overall, the results from this first in vitro study of the magnetic seeding concept for IA-MDT were very encouraging. They proved that magnetic seeds, such as magnetite particles, could serve as a very effective implant for MDT systems, especially under conditions found in capillaries. Future studies will further develop this magnetic seeding concept by demonstrating in vitro that the magnetic seeds can be magnetically captured first and then used to collect the MDCPs.
Acknowledgments The research support provided by the following agencies and institutions was greatly appreciated: NSF under Grant Nos. CTS-0314157 and CTS-0508391, the Sloan Foundation Graduate Research Fellowship provided to MOA, the Ford Foundation Graduate Research Fellowship provided to MOA, and the USC NanoCenter. The authors also thank Franklyn Rainsford for his assistance with carrying out some of the experiments in the laboratory.
References [1] U.O. Ha¨feli, Int. J. Pharm. 277 (2004) 19. [2] Q.A. Pankhurst, J. Connolly, S.K. Jones, J. Dobson, J. Phys. D: Appl. Phys. 36 (2003) R167. [3] T. Neuberger, B. Schopf, H. Hofmann, M. Hofmann, B. von Rechenberg, J. Magn. Magn. Mater. 293 (2005) 483. [4] K.J. Widder, A.E. Senyei, D.G. Scarpelli, Proc. Soc. Exp. Biol. Med. 58 (1978) 141. [5] A. Senyei, K. Widder, C. Czerlinski, J. Appl. Phys. 49 (1978) 3578. [6] Ch. Alexiou, W. Arnold, R.J. Klein, F.G. Parak, P. Hulin, C. Bergemann, W. Erhardt, S. Wagenpfeil, A.S. Lu¨bbe, Cancer Res. 60 (2000) 6641. [7] Ch. Alexiou, R.J. Schmid, R. Jurgons, M. Kremer, G. Wanner, C. Bergemann, E. Huenges, T. Nawroth, W. Arnold, F.G. Parak, Eur. Biophys. J. 35 (2006) 446. [8] S. Goodwin, C. Peterson, C. Hoh, C. Bittner, J. Magn. Magn. Mater. 194 (1999) 132. [9] K. Schulze, A. Koch, B. Schopf, A. Petri, B. Steitz, M. Chastellain, M. Hofmann, H. Hofmann, B. von Rechenberg, J. Magn. Magn. Mater. 293 (2005) 419. [10] H. Xu, T. Song, X. Bao, L. Hu, J. Magn. Magn. Mater. 293 (2005) 514. [11] A.S. Lu¨bbe, C. Bergemann, J. Brock, D.G. McClure, J. Magn. Magn. Mater. 194 (1999) 149. [12] A.J. Lemke, M.-I. Senfft von Pilsach, A. Lu¨bbe, C. Bergermann, H. Riess, R. Felix, Eur. Radiol. 14 (2004) 1949. [13] A.S. Lu¨bbe, C. Alexiou, C. Bergemann, J. Surg. Res. 95 (2001) 200. [14] J.A. Ritter, A.D. Ebner, K.D. Daniel, K.L. Stewart, J. Magn. Magn. Mater. 280 (2004) 184. [15] M.O. Avile´s, A.D. Ebner, H. Chen, A.J. Rosengart, M.D. Kaminski, J.A. Ritter, J. Magn. Magn. Mater. 293 (2005) 605. [16] H. Chen, A.D. Ebner, A.J. Rosengart, M.D. Kaminski, J.A. Ritter, J. Magn. Magn. Mater. 284 (2004) 181. [17] H. Chen, A.D. Ebner, A.J. Rosengart, M.D. Kaminski, J.A. Ritter, J. Magn. Magn. Mater. 293 (2005) 616. [18] M.O. Avile´s, A.D. Ebner, J.A. Ritter, J. Magn. Magn. Mater. 311 (2007) 306. [19] M.O. Avile´s, A.D. Ebner, J.A. Ritter, J. Magn. Magn. Mater. 310 (2007) 131. [20] A.J. Rosengart, M.D. Kaminski, H. Chen, P.L. Caviness, A.D. Ebner, J.A. Ritter, J. Magn. Magn. Mater. 293 (2005) 633. [21] G. Iacob, O. Rotariu, H. Chiriac, J. Optoelec. Adv. Mater. 6 (2004) 713. [22] O. Rotariu, N.J.C. Strachan, J. Magn. Magn. Mater. 293 (2005) 639. [23] B.B. Yellen, Z.G. Forbes, D.S. Halverson, G. Fridman, K.A. Barbee, M. Chorny, R. Levy, G. Friedman, J. Magn. Magn. Mater. 293 (2005) 647. [24] Z.G. Forbes, B.B. Yellen, K.A. Barbee, G. Friedman, IEEE Trans. Magn. 39 (2003) 3372. [25] R. Gerber, in: R. Gerber, C.D. Wright, G. Asti (Eds.), Applied Magnetism, NATO ASI Series, Series E: Applied Sciences, vol. 253, Kluwer Academic Publishers, Dordrecht, 1994. [26] R.C. Thomson, M.J. Yaszemski, J.M. Powers, A.G. Mikos, Fabrication of biodegradable polymer scaffolds to engineer trabecular bone, J. Biomater. Sci. Polymer Edn. 7 (1995) 23.
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm
In vitro study of magnetic particle seeding for implant assisted-magnetic drug targeting Misael O. Avile´s, Armin D. Ebner, James A. Ritter Department of Chemical Engineering, Swearingen Engineering Center, University of South Carolina, Columbia, SC 29208, USA
a r t i c l e in f o
a b s t r a c t
Article history: Received 8 February 2008 Received in revised form 12 May 2008 Available online 22 May 2008
The concept of using magnetic particles (seeds) as the implant for implant assisted-magnetic drug targeting (IA-MDT) was analyzed in vitro. Since this MDT system is being explored for use in capillaries, a highly porous (e70%), highly tortuous, cylindrical, polyethylene polymer was prepared to mimic capillary tissue, and the seeds (magnetite nanoparticles) were already fixed within. The well-dispersed seeds were used to enhance the capture of 0.87 mm diameter magnetic drug carrier particles (MDCPs) (polydivinylbenzene embedded with 24.8 wt% magnetite) under flow conditions typically found in capillary networks. The effects of the fluid velocity (0.015–0.15 cm/s), magnetic field strength (0.0–250 mT), porous polymer magnetite content (0–7 wt%) and MDCP concentration (C ¼ 5 and 50 mg/L) on the capture efficiency (CE) of the MDCPs were studied. In all cases, when the magnetic field was applied, compared to when it was not, large increases in CE resulted; the CE increased even further when the magnetite seeds were present. The CE increased with increases in the magnetic field strength, porous polymer magnetite content and MDCP concentration. It decreased only with increases in the fluid velocity. Large magnetic field strengths were not necessary to induce MDCP capture by the seeds. A few hundred mT was sufficient. Overall, this first in vitro study of the magnetic seeding concept for IAMDT was very encouraging, because it proved that magnetic particle seeds could serve as an effective implant for MDT systems, especially under conditions found in capillaries. & 2008 Elsevier B.V. All rights reserved.
Keywords: Magnetic drug targeting MDT Drug delivery High gradient magnetic separation HGMS Ferromagnetic seed Magnetic drug carrier particle MDCPs In vitro
1. Introduction Magnetic drug targeting (MDT) [1–3] is based on the use of an external magnetic field to attract and retain specially designed magnetic drug carrier particles (MDCPs) to a specific site in the body. Because the MDCPs accumulate at the desired site, the drug or active agent would necessarily be high in concentration at this site. This should improve the effectiveness of the treatment, while also reducing the dose needed to reach the therapeutic benefit, exposure of healthy tissue to the treatment regiment, and side effects. Although the concept of MDT has been around for over three decades [4,5], and despite the fact that it has shown some promise in both in vivo [6–10] and clinical [11–13] studies, wide spread acceptance of the technique is still looming. This is because traditional MDT has some inherent limitations. Typically, the magnetic force is not very strong, and it is also very short ranged. Since the magnetic force must overcome rather large hydrodynamic forces in the body, MDT applications have been limited to sites located close to the surface of the skin. Even in this most favorable situation, studies have shown that it is difficult to collect appreciable amounts of the MDCPs at the target site [10,11].
Corresponding author. Tel.: +1803 777 3590; fax: +1 803 777 8265.
E-mail address: [email protected] (J.A. Ritter). 0304-8853/$ - see front matter & 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jmmm.2008.05.022
To overcome these limitations of the traditional MDT approach, Ritter and co-workers [14–20] and others [21–24] recently touted the use of a magnetic implant placed strategically at the target site to help attract and retain the MDCPs there. They refer to this new MDT approach as implant assisted (IA) MDT. This new approach borrows notions from high gradient magnetic separation (HGMS) principles [25], wherein the magnetic force imparted on a magnetic particle (e.g., a MDCP) by a magnetic field is proportional to both the magnitude and the gradient of the magnetic field. Under the influence of an external magnetic field the implant becomes magnetized, and not only creates its own magnetic field around itself that can be greater in magnitude than the external magnetic field, but it also distorts the external magnetic field creating large gradients. Both of these changes in the local magnetic field around the implant increase the force exerted on the MDCPs that are traveling in the vicinity of the implant, which in turn increases their attraction to and retention at the site. Several theoretical and experimental studies [14–24] have shown the IA-MDT approach to be far superior to the traditional MDT approach when using implants such as wires [14,15,21,22], stents [16–18,23], and even seeds [19,24]. For a wire or stent to serve as the implant, it would be put in place most likely using a catheter. For seeds, i.e., tiny magnetic particles that serve as the implant, they would be put in place simply by injecting them transdermally in or near the target organ or tissue. In this way, the
ARTICLE IN PRESS M.O. Avile´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646
2641
sodium n-dodecyl sulfate (Alfa Aesar, 99%) were used as received. Magnetic microspheres, 0.87 mm in diameter, composed of carboxylmodified polydivinylbenzene, and containing 24.8 wt% magnetite, were obtained in a suspension from Bangs Laboratories, Inc. For atomic absorption spectroscopy measurements, concentrated nitric acid (Sigma-Aldrich, 69–70%) and 30% hydrogen peroxide (SigmaAldrich) were used as received. DI water was obtained from a MilliQ purification system.
V
2.2. Magnetic porous polymer preparation and characterization
seeds
MDCPs MDCPs B
Fig. 1. Schematic diagram depicting the magnetic particle seeding concept for implant assisted-magnetic drug targeting (IA-MDT), wherein magnetic seeds, such as magnetite particles, are magnetically collected at a target zone by the applied magnetic field B and then used to capture magnetic drug carrier particles (MDCPs) in this zone.
use of seeds is very advantageous over a wire or stent, as they would involve a less invasive technique for targeting drugs to a desired location in the body. Fig. 1 shows a schematic of this magnetic seeding concept for IA-MDT. In the presence of an externally applied magnetic field, the seeds would be injected first and collected in the capillary tissue, as described above. Due to them necessarily having a larger magnetization than the MDCPs, they would be easily retained in the tissue by the magnetic field. Once the seeds are in place and with the magnetic field still applied, the MDCPs would be injected essentially at the same location as the seeds. Any of the MDCPs that happen to travel close to these highly dispersed and magnetized seeds would be attracted to and retained by them in the same way a wire or stent would attract and retain the MDCPs. A recent theoretical study showed the feasibility of using magnetic seeds for IA-MDT [19]; however, an in vitro study has yet to be done. Therefore, the objective of this work is to provide the first experimental evidence of the feasibility of using seeds as an implant for IA-MDT. An in vitro study was carried using a porous polyethylene (PE) polymer as surrogate capillary tissue, magnetite nanoparticles as seeds, and polydivinylbenzene particles imbedded with magnetite as MDCPs. With the intent being to show that the seeds could indeed be used to collect the MDCPs over a wide range of physiologically relevant conditions, the seeds were a priori fixed in the polymer matrix in this initial study. The effects of the fluid velocity, magnetic field strength, magnetite content in the porous polymer matrix and MDCP concentration on the performance of this IA-MDT system were analyzed in terms of the capture efficiency (CE) of the MDCPs.
A porous polymer of cylindrical shape was prepared by a compression melt molding technique, followed by a salt leaching step [26]. In short, the PE powder and NaCl particles were mixed in a ratio of 1.00 g of PE to 5.37 g of NaCl to obtain a material with about 70% porosity once all the salt was removed. To this mixture, magnetite nanopowder was added to obtain a material with up to 9.0 wt% magnetite. The materials were mixed for at least 2 h, and then added to a specially designed Teflon mold with a 1 cm diameter by 5 cm long cylindrical opening. A 1 cm long by 1 cm diameter Teflon plug was placed on top of the sample inside the mold opening to form the cylindrical shape. A 1 ton manual arbor press was used to compress the sample. The mold was then heated to 170 1C for 2 h, with recompression steps after 1 and 2 h. The mold was cooled to room temperature and the sample was removed. The sample, in the form of a solid rod, was immersed in water for at least 48 h to remove all the salt. The water was changed out every hour for the first 4 h, and then every 2–6 h. The resulting porous samples were cut into 1 cm long cylindrical pieces. The cut samples were dried for 48 h under vacuum at 60 1C. Four samples were prepared, each with different magnetite contents. They are referred to as S1, S2, S3 and S4, each initially targeted to contain 0.0, 1.0, 5.0 and 9.0 wt% magnetite, respectively. The morphologies of the porous polymers were imaged by examining cross-sections of each sample in a FEI Quanta 200 environmental scanning electron microscope (SEM) operated at 30 kV. Energy dispersive X-ray (EDX) analysis was used to map the locations of the magnetite particles in the samples. The porosity of the material was estimated from the apparent geometrical density (ra) and the skeletal density (rs). The apparent geometrical density was measured using the weight of a 1 cm long by 1 cm in diameter porous polymer sample. The skeletal density was measured using a Quantachrome Ultrapycnometer 1000 with helium as the carrier gas. The porosity (e) was calculated from e ¼ 1(ra/rs) A Perkin Elmer 3300 Flame Atomic Absorption (AA) Spectrometer was used to determine the amount of iron in the polymer samples. To prepare the samples for AA analysis, 50 mg of the porous polymer were digested inside tightly capped 50 mL Teflon vials at 165 1C for 2 h using a mixture of concentrated nitric acid (3 mL) and 30% hydrogen peroxide (2 mL). The digested samples were quantitatively diluted into 50 mL volumetric flasks and filtered prior to analysis. Subsequent dilutions were performed as needed. All measurements were done in duplicate. 2.3. Magnetic characterization
2. Materials and method 2.1. Materials Ultra-high molecular weight PE powder (Sigma-Aldrich, 180 mm), sodium chloride (Sigma-Aldrich, +80 mesh, 98+%), iron(II,III) oxide (magnetite) nanopowder (Sigma-Aldrich, 20–30 nm, 98+%) and
The magnetic susceptibilities of the MDCPs and the magnetite nanopowder were measured using a Quantum Design MPMS XL SQUID magnetometer. Either the magnetite nanopowder or the MDCPs was placed in a gelatin capsule. This capsule was then placed inside a plastic straw, which was subsequently placed inside the instrument. Magnetic field sweeps were recorded between +1 and 1 T at 300 K. The very small diamagnetic
ARTICLE IN PRESS 2642
´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646 M.O. Avile
Porous Polymer with Magnetite Seeds
NdFeB
XX:XX ml/m
NdFeB
Syringe Pump
Collection Vessel Dual Permanent Magnet Fig. 2. Experimental apparatus used in the in vitro study of the magnetic particle-seeding concept for implant assisted magnetic drug targeting (IA-MDT).
contribution of the gelatin capsule was considered to be negligible.
Table 1 Porosity and magnetite contents in the polyethylene porous polymer cylinders Sample
Porosity (e)
2.4. MDCP capture experimental setup The MDCP capture experimental setup is shown in Fig. 2. It consisted of the cylindrical porous polymer containing the magnetite seeds, a dual magnet system comprised of two NdFeB permanent magnets located adjacent to the porous polymer, a 50 mL syringe to supply the suspension of MDCPs, and a KDS200 syringe pump to control the flow through the polymer matrix. A calibrated Spectronic 20+ spectrophotometer set at 650 nm was used to determine the concentration of MDCPs in suspension as a function of the suspension transmittance. The dual magnet configuration consisted of two, 0.3 T (50–50–25 mm) rectangular magnets with their field oriented perpendicular to their 50 50 mm face and with the porous polymer sample located at the center of the gap between them. The magnitude of the homogeneous magnetic field created between the two permanent magnets was adjusted by varying the distance between them. The magnetic field was measured using a Model 4048 Bell Gauss/Tesla meter. The porous polymer was encapsulated inside the flow tube by placing the 1 cm diameter by 1 cm long cylinder inside a 3 cm long piece of shrink tube, along with 4 in pieces of 1/8 in Tygon tubing at each end. A heat gun was then used to shrink the tube while taking care not to overheat the polymer to avoid melting. Once the shrink tube was reduced in size, the porous polymer was encapsulated tightly inside, along with the two pieces of Tygon tubing at each end. 2.5. MDCP capture experimental procedure For a typical experiment, a suspension of the MDCPs at a concentration of either 5 or 50 mg/L was prepared in a 100 mM sodium n-dodecyl sulfate (SDS) aqueous solution. The initial concentration of the MDCPs in the suspension was measured using the spectrophotometer. The suspension was then charged into the 50 mL syringe. A magnetic field was focused on the porous polymer. The distance between the magnets was varied to modify the magnetic field (B) from 38 to 250 mT, while keeping the porous polymer centered between the two magnets. With the syringe filled and the magnets in place, the syringe pump was used to drive the MDCP suspension at a fixed flow rate through the tube containing the porous polymer initially solution free (dry). The effluent from the tube containing the porous polymer was collected in small vials. The concentration of the MDCPs in the first 10 mL of this effluent was measured and used to determine the CE. By comparing the concentration of the MDCPs in the initial suspension (Ci) to the concentration of the MDCPs in
S1 S2 S3 S4
0.732 0.765 0.712 0.740 a
Magnetite content (wt%) Expected
AASa
0.00 0.99 4.80 9.10
0.00 0.7970.04 2.8570.15 7.0670.91
Atomic Absorption Spectroscopy.
the effluent (Cf), the CE was calculated as CE ¼
Ci Cf 100 Ci
(1)
All experiments were done in duplicate, with the average CE reported around one standard deviation error bars.
3. Results and discussion To demonstrate the concept of ferromagnetic seeding for MDT using in vitro experiments, a highly porous structure was needed to represent capillary tissue. This surrogate capillary tissue had to effectively trap or retain the magnetite seed particles uniformly throughout its structure, just like real capillary tissue would. But, it could not behave as a gross, non-magnetic filter medium for the MDCPs. After some trial and error with a few different porous materials, the porous PE cylinders proved to have both of those properties. Table 1 lists the porosity and magnetite content of the polymer samples. The e ranged between 0.712 and 0.765. This range was close to the target value of 0.70. The measured (AAS) magnetite content in each sample was 0.0, 0.79, 2.85 and 7.6 wt% for samples S1, S2, S3 and S4, respectively. These values were close to the expected values of 0.00, 0.99, 4.80 and 9.10 wt%, based on the magnetite added to the polymer prior to the compression step. The difference was most likely due to the loss of magnetite during the salt leaching step. Nevertheless, a sufficient amount of magnetite was successfully incorporated into each polymer matrix. Fig. 3 shows SEM images of samples S1, S2, S3 and S4. These materials all had large, interconnected pores that ranged in size from 50 to 300 mm. Although these pores were much larger than the diameter of typical capillaries (5 mm), the intent was never to reproduce such small pores. On the contrary, the intent was to make a highly porous, large pore material with magnetite dispersed throughout that also had a tortuous path through the pores that would allow the MDCP suspension to flow through it
ARTICLE IN PRESS M.O. Avile´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646
2643
Fig. 3. Scanning electron micrographs of the porous polymer samples S1, S2, S3 and S4 (refer to Table 1 for properties).
without much hindrance. This was achieved with these porous PE cylinders. A qualitative depiction of the magnetite dispersed throughout each cylinder was obtained by mapping the location of the iron in each sample with EDX analysis. The results are shown in Fig. 4. As expected, the iron content increased in going from sample S2 to S4, with some agglomerated magnetite particles apparent. The level of magnetite dispersion achieved in these samples was considered sufficient for testing the magnetic seeding concept. Fig. 5 shows the normalized magnetization curves (M/Ms) for the MDCPs and the magnetite seeds. The saturation magnetization (Ms) was 22.4 and 385 kA/m for the MDCPs and magnetite seeds, respectively. Indicative of the magnetite in the MDCPs being superparamagnetic, they did not show any coercive fields. In contrast, the magnetite seeds showed a small, but essentially negligible, coercive field or residual magnetization, which resulted from its soft magnetic character. These results alluded to the notion that the seeds would be retained magnetically much more easily than the MDCPs in real capillary tissue because of them having a much higher Ms, and that they could then be used to collect the much less magnetic MDCPs. The results from the MDCP capture experiments discussed below provide evidence in support of the latter supposition. Evidence in support of the former supposition will be provided in future work, where the seeds will indeed be captured in the polymer matrix instead of being a priori anchored in it. The magnetic seeding concept was demonstrated by carrying out MDCP capture experiments to determine the amount of MDCPs collected inside the porous polymer as a suspension containing them was pumped through it in the presence of a magnetic field. The variables of interest included the fluid velocity (V ¼ 0.015–0.15 cm/s), the magnetic field strength (B ¼ 0.0–250 mT), the magnetite content in the polymer samples (0–7 wt%), and the concentration of the MDCPs in the suspension (C ¼ 5 and 50 mg/L). The performance of this IA-MDT system was evaluated in terms of the CE defined in Eq. (1).
The effect of the fluid velocity on the CE is shown in Fig. 6. In this study, samples S1 and S2, a MDCP concentration of 50 mg/L, and an applied magnetic field strength of 250 mT were used. To mimic conditions found in real capillaries, the velocities were varied from 0.015 to 0.150 cm/s. A set of control experiments under the same conditions and no magnetic field were also carried out for comparison. The CEs obtained with the control experiments never exceeded 10%, showing that physical filtration was minimal. However, when the magnetic field was applied, a significant increase in the CE was observed in both samples, which revealed the strong role of the magnetic field in capturing the MDCPs. More importantly, the CE for the sample containing magnetite (S2) was statistically higher in all cases, which clearly demonstrated the positive role of the magnetite seeds. The results in Fig. 6 also showed that the CE decreased with increasing velocity, as expected. The larger the fluid velocity, the larger the hydrodynamic force over the MDCPs, the more this force dominated the fixed magnetic force. However, the important role of the magnetic seeds in overcoming the hydrodynamic force became apparent with increasing velocity, as the detrimental effects of velocity were much more pronounced for the case when the seeds were not present compared to when they were present. It is noteworthy that the significant CEs achieved by sample S1 in the absence of magnetite seeds indicated that other magnetic mechanisms were taking place and causing collection of the MDCPs. There are two plausible explanations that might explain this phenomenon. First, slight magnetic gradients created by the permanent magnets perhaps exerted an attractive force upon the MDCPs. Although the dual magnet configuration was designed to create a very uniform magnetic field in the vicinity of the polymer cylinder, some magnetic gradients were most likely present. Second, magnetic interparticle interactions between the MDCPs most likely existed. Like the magnetite seeds, the MDCPs also became magnetized in the presence of the uniform magnetic field. This most likely induced an attraction between neighboring MDCPs that lead to them agglomerating into small clusters that
ARTICLE IN PRESS 2644
´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646 M.O. Avile
1.00 0.75
Magnetite Seeds MDCP Surrogates
0.50 M/Ms
0.25 0.00 -0.25 -0.50 -0.75 -1.00 -0.25 -0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 B (T) Fig. 5. Normalized magnetization curves for the MDCPs and the magnetite seeds.
100
S1 - B = 250.0 mT S2 - B = 250.0 mT S1 - B = 0.00 mT S2 - B = 0.00 mT
90 80
Dual Magnet
CE %
70 60 50 40 30 20 10 0 0.015
0.030
0.060 0.090 Velocity (cm/s)
0.150
Fig. 6. Effect of fluid velocity on the capture efficiency (CE) of MDCPs for samples S1 (0 wt% magnetite) and S2 (0.79 wt% magnetite) at a MDCP concentration (C) of 50 mg/L, and applied magnetic field strengths (B) of 0 (control experiments) and 250 mT.
Fig. 4. Energy dispersive X-ray analysis of iron in samples: (a) S2 (0.79 wt% magnetite), (b) S3 (2.85 wt% magnetite) and (c) S4 (7.06 wt% magnetite).
became more easily captured by the slight magnetic field gradients created by the permanent magnets alone. This agglomeration between the MDCPs might have also lead to self-seeding,
in which clusters of the already retained MDCPs acted as seeds to further attract incoming MDCPs. The effect of the applied magnetic field strength (B) on the CE is shown in Fig. 7. In this study, samples S1 and S2, a MDCP concentration of 50 mg/L, a velocity of 0.06 cm/s, and magnetic field strengths of 250, 125, 70, 38 and 0 mT (control experiments) were used. A very positive effect of the magnetic field strength on the CE was observed, whether the magnetite seeds were present or not. However, the effect was more pronounced in all cases when the magnetite seeds were encased in the polymer matrix. The pronounced effect observed in Fig. 7 at low magnetic field strengths was perhaps due to the magnetic response of both the magnetite seeds and the MDCPs. As shown in Fig. 5, at low magnetic field strengths, both materials had a steep response to the applied magnetic field. Thus, even at the low magnetic field strengths, they were nearly magnetically saturated, which meant that they were imparting their greatest magnetic forces upon each other. On the other hand, it was interesting that at a magnetic field strength of 125 mT or lower, there were no significant differences in the CE for the samples without magnetite seeds (S1) as long as a magnetic field was applied. This behavior was due to having to increase the magnet-to-magnet separation to decrease
ARTICLE IN PRESS M.O. Avile´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646
100
100
S1 - v = 0.06 cm/s S2 - v = 0.06 cm/s
90
80
70
70
60
60
50 Dual Magnet
40
B = 0.0 mT - v = 0.06 cm/s B = 38 mT - v = 0.06 cm/s B = 250 mT - v = 0.06 cm/s
90
CE %
CE %
80
2645
Dual Magnet
50 40
30
30
20
20
10
10 0
0 125.0 70.0 38.0 Magnetic Field (mT)
Fig. 7. Effect of applied magnetic field strength (B) on the capture efficiency (CE) of the MDCPs for samples S1 (0 wt% magnetite) and S2 (0.79 wt% magnetite) at a MDCP concentration (C) of 50 mg/L, and a velocity of 0.06 cm/s. The control experiments were done at B of 0.0 mT.
the magnetic field strength. As the distance between the magnets increased, it caused a reduction in the magnetic field gradients created from the magnets alone. However, the CE in this case was still significant, even with a considerable decrease in the applied magnetic field strength, possibly due to magnetic interparticle interactions between the MDCPs, as discussed above. These results collectively confirmed that large magnetic field strengths were not necessary to induce MDCP capture by the seeds. They also showed that at the low magnetic field strengths, where the magnetic field gradients created by the magnets alone were minimal, the seeds were the major contributor to MDCP capture. The effect of the magnetite content inside the porous polymer on the CE is shown in Fig. 8. In this study, samples S1, S2, S3 and S4 (refer Table 1), a MDCP surrogate concentration of 50 mg/L, a velocity of 0.06 cm/s, and applied magnetic fields of 0 (control experiments), 38 and 250 mT were used. For the control experiments there were essentially no differences between the different samples, and the CE never exceeded 5%. In contrast, when the magnetic field was applied, the CE increased significantly with increasing magnetite content inside the porous polymer sample. This result was expected and indicated that the probability of capturing a MDCP increased when more seeds were present. The effect of the concentration of the MDCPs in the suspension on the CE is shown in Fig. 9. In this study, samples S1 and S2, MDCP concentrations of 50 and 5 mg/L, applied magnetic field strengths of 0 (control experiments) and 250 mT, and a velocity of 0.06 cm/s were used. The control experiments without the magnetic field showed essentially no difference between the CEs at both concentrations and between both samples. In contrast, when the magnetic field was applied, strong effects of concentration were observed, with higher CE always resulting when the seeds were present. This effect of the MDCP concentration in the suspension was probably due to magnetic interparticle interactions that induced magnetic agglomeration between them and self-seeding.
4. Conclusions The use of magnetite particles as seeds for implant assistedmagnetic drug targeting (IA-MDT) in, for example, capillary tissue, was studied in vitro. To mimic flow through actual capillary tissue, a highly porous (e70%), highly tortuous, cylindrical, PE polymer was prepared with the magnetite particles already fixed within.
S2 S3 Porous Polymer Sample
S1
0.0
S4
Fig. 8. Effect of the magnetite content in the porous polymer sample on the capture efficiency (CE) of the MDCPs at a velocity of 0.06 cm/s, a MDCP concentration (C) of 50 mg/L, and applied magnetic field strengths (B) of 0 (control experiments) and 250 mT.
100
B = 0.0 mT - C = 5 mg/L - v = 0.06 cm/s B = 0.0 mT - C = 50 mg/L - v = 0.06 cm/s B = 250 mT - C = 5 mg/L - v = 0.06 cm/s B = 250 mT - C = 50 mg/L - v = 0.06 cm/s
90 80 70 CE %
250.0
Dual Magnet
60 50 40 30 20 10 0 S1
S2 Porous Polymer Sample
Fig. 9. Effect of the MDCP concentration (C) in the suspension on the capture efficiency (CE) of the MDCPs for samples S1 (0 wt% magnetite) and S2 (0.79 wt% magnetite), at a velocity of 0.06 cm/s, and applied magnetic field strengths (B) of 0 (control experiments) and 250 mT.
The seeds were used to enhance the capture of 0.87 mm diameter MDCPs comprised of polydivinylbenzene embedded with 24.8 wt% magnetite. An experiment consisted of pumping a known volume of the MDCP suspension through a tube containing the porous polymer that was positioned between two permanent magnets. The effluent was collected and its MDCP concentration was compared to that in the initial sample to compute a CE. The effects of the fluid velocity (0.015–0.15 cm/s), magnetic field strength (0.0–250 mT), porous polymer magnetite content (0–7 wt%), and MDCP concentration (C ¼ 5 and 50 mg/L) on the CE of the MDCPs were studied. Control experiments, done in the absence of an applied magnetic field, demonstrated minimal physical filtration effects and that all MDCP capture was due to magnetic effects. In all cases, when the magnetic field was applied there were tangible increases in the CE, whether the porous polymer contained magnetite seeds or not. However, the CEs were always statistically greater when the magnetite seeds were present, indicating the
ARTICLE IN PRESS 2646
´s et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 2640–2646 M.O. Avile
positive influence of the seeds on the capture of MDCPs. The CE increased with increases in the magnetic field strength, porous polymer magnetite content, and MDCP concentration; it decreased only with increases in the fluid velocity. In terms of the magnetic field effects, the results demonstrated that large magnetic field strengths were not necessary to induce MDCP capture by the seeds. This very positive outcome was due to both the magnetite seeds and the MDCPs becoming magnetically saturated at relatively low magnetic field strengths. This allowed them to impart their maximum attractive forces upon each other at magnetic field strengths of only a few hundred mT. It also allowed the MDCPs to magnetically agglomerate with each other making capture easier and possibly giving rise to self-seeding. This became evident from the large CEs obtained from the samples that did not contain any magnetite seeds, but that were exposed to a magnetic field. Overall, the results from this first in vitro study of the magnetic seeding concept for IA-MDT were very encouraging. They proved that magnetic seeds, such as magnetite particles, could serve as a very effective implant for MDT systems, especially under conditions found in capillaries. Future studies will further develop this magnetic seeding concept by demonstrating in vitro that the magnetic seeds can be magnetically captured first and then used to collect the MDCPs.
Acknowledgments The research support provided by the following agencies and institutions was greatly appreciated: NSF under Grant Nos. CTS-0314157 and CTS-0508391, the Sloan Foundation Graduate Research Fellowship provided to MOA, the Ford Foundation Graduate Research Fellowship provided to MOA, and the USC NanoCenter. The authors also thank Franklyn Rainsford for his assistance with carrying out some of the experiments in the laboratory.
References [1] U.O. Ha¨feli, Int. J. Pharm. 277 (2004) 19. [2] Q.A. Pankhurst, J. Connolly, S.K. Jones, J. Dobson, J. Phys. D: Appl. Phys. 36 (2003) R167. [3] T. Neuberger, B. Schopf, H. Hofmann, M. Hofmann, B. von Rechenberg, J. Magn. Magn. Mater. 293 (2005) 483. [4] K.J. Widder, A.E. Senyei, D.G. Scarpelli, Proc. Soc. Exp. Biol. Med. 58 (1978) 141. [5] A. Senyei, K. Widder, C. Czerlinski, J. Appl. Phys. 49 (1978) 3578. [6] Ch. Alexiou, W. Arnold, R.J. Klein, F.G. Parak, P. Hulin, C. Bergemann, W. Erhardt, S. Wagenpfeil, A.S. Lu¨bbe, Cancer Res. 60 (2000) 6641. [7] Ch. Alexiou, R.J. Schmid, R. Jurgons, M. Kremer, G. Wanner, C. Bergemann, E. Huenges, T. Nawroth, W. Arnold, F.G. Parak, Eur. Biophys. J. 35 (2006) 446. [8] S. Goodwin, C. Peterson, C. Hoh, C. Bittner, J. Magn. Magn. Mater. 194 (1999) 132. [9] K. Schulze, A. Koch, B. Schopf, A. Petri, B. Steitz, M. Chastellain, M. Hofmann, H. Hofmann, B. von Rechenberg, J. Magn. Magn. Mater. 293 (2005) 419. [10] H. Xu, T. Song, X. Bao, L. Hu, J. Magn. Magn. Mater. 293 (2005) 514. [11] A.S. Lu¨bbe, C. Bergemann, J. Brock, D.G. McClure, J. Magn. Magn. Mater. 194 (1999) 149. [12] A.J. Lemke, M.-I. Senfft von Pilsach, A. Lu¨bbe, C. Bergermann, H. Riess, R. Felix, Eur. Radiol. 14 (2004) 1949. [13] A.S. Lu¨bbe, C. Alexiou, C. Bergemann, J. Surg. Res. 95 (2001) 200. [14] J.A. Ritter, A.D. Ebner, K.D. Daniel, K.L. Stewart, J. Magn. Magn. Mater. 280 (2004) 184. [15] M.O. Avile´s, A.D. Ebner, H. Chen, A.J. Rosengart, M.D. Kaminski, J.A. Ritter, J. Magn. Magn. Mater. 293 (2005) 605. [16] H. Chen, A.D. Ebner, A.J. Rosengart, M.D. Kaminski, J.A. Ritter, J. Magn. Magn. Mater. 284 (2004) 181. [17] H. Chen, A.D. Ebner, A.J. Rosengart, M.D. Kaminski, J.A. Ritter, J. Magn. Magn. Mater. 293 (2005) 616. [18] M.O. Avile´s, A.D. Ebner, J.A. Ritter, J. Magn. Magn. Mater. 311 (2007) 306. [19] M.O. Avile´s, A.D. Ebner, J.A. Ritter, J. Magn. Magn. Mater. 310 (2007) 131. [20] A.J. Rosengart, M.D. Kaminski, H. Chen, P.L. Caviness, A.D. Ebner, J.A. Ritter, J. Magn. Magn. Mater. 293 (2005) 633. [21] G. Iacob, O. Rotariu, H. Chiriac, J. Optoelec. Adv. Mater. 6 (2004) 713. [22] O. Rotariu, N.J.C. Strachan, J. Magn. Magn. Mater. 293 (2005) 639. [23] B.B. Yellen, Z.G. Forbes, D.S. Halverson, G. Fridman, K.A. Barbee, M. Chorny, R. Levy, G. Friedman, J. Magn. Magn. Mater. 293 (2005) 647. [24] Z.G. Forbes, B.B. Yellen, K.A. Barbee, G. Friedman, IEEE Trans. Magn. 39 (2003) 3372. [25] R. Gerber, in: R. Gerber, C.D. Wright, G. Asti (Eds.), Applied Magnetism, NATO ASI Series, Series E: Applied Sciences, vol. 253, Kluwer Academic Publishers, Dordrecht, 1994. [26] R.C. Thomson, M.J. Yaszemski, J.M. Powers, A.G. Mikos, Fabrication of biodegradable polymer scaffolds to engineer trabecular bone, J. Biomater. Sci. Polymer Edn. 7 (1995) 23.