Research Article www.acsami.org
Cite This: ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Using Magnetic Ions to Probe and Induce Magnetism of Pyrophosphates, Bacteria, and Mammalian Cells Jia-Lin Wei and Yu-Chie Chen* Department of Applied Chemistry, National Chiao Tung University, Hsinchu 300, Taiwan
ACS Appl. Mater. Interfaces Downloaded from pubs.acs.org by UNIV OF SOUTH DAKOTA on 08/28/18. For personal use only.
S Supporting Information *
ABSTRACT: Magnetic isolation using magnetic nanoparticles (MNPs) as trapping probes have been widely used in sample pretreatment to shorten analysis time. Nevertheless, to generate MNPs is time-consuming. Furthermore, the generated MNPs have to be further functionalized to gain the capability of recognizing their target species. Thus, an alternative approach that can impose magnetism to nonmagnetic species by simply using magnetic ions as the probes is developed in this study. That is, we employ magnetic ions (Fe3+, Co2+, and Ni2+) that can interact with nonmagnetic species containing oxygen-containing functional groups as the probes. Pyrophosphate (PPi), bacteria, and mammalian cells were selected as the model samples. Our results show that the as-prepared magnetic ion−PPi conjugates gain sufficient magnetism and can be readily aggregated by applying an external magnetic field. Moreover, the magnetic trapping is reversible. The PPi-containing conjugates can lose their magnetic property simply using ethylenediaminetetraacetic acid or aluminum ions as competing agents to remove or to replace, respectively, the conjugated magnetic ions. In addition, bacteria and mammalian cells that possess abundant oxygen-containing functional groups on their cell surfaces can be selectively probed by magnetic ions and gain sufficient magnetism for magnetic isolation from complex serum samples. KEYWORDS: magnetic ions, Fe3+, Co2+, Ni2+, magnetism, pyrophosphates (PPi), bacteria
■
INTRODUCTION Magnetic nanoparticles (MNPs), such as iron oxide (Fe3O4) MNPs, have attracted considerable attention because of their superparamagnetic properties, small sizes, and ease of separation.1,2 Thus, MNPs have been extensively used in different research fields1−7 such as analytical chemistry.7 Taking advantage of the magnetic property, MNPs are often used as affinity probes in the development of analytical methods for selectively trapping target species of interest. Generally, MNPs are fabricated by using bottom-up methods, which require several hours for generation of MNPs through liquid reactions.8 Moreover, further functionalization of the asgenerated MNPs is usually required. MNP fabrication and functionalization are relatively time-consuming. Thus, we aimed to explore a method that can simply implement sufficient magnetism to nonmagnetic species of interest for magnetic separation by simply using magnetic ions as affinity probes. Magnetic ions, including Fe3+ (1s22s22p63s23p63d5), Co2+ (1s22s22p63s2 3p63d7), and Ni2+ (1s22s22p63s23p63d8), contain five, seven, and eight d electrons in their outer shell orbitals, respectively. When nonmagnetic species are attached by a high density of magnetic ions to form magnetic ion complex conjugates, the density of unpaired electron spins in the conjugates can be raised to a great extent, thereby increasing the magnetic susceptibility.9−13 For example, 1-butyronitrile-3© XXXX American Chemical Society
methylimidazolium tetrachloroferrate is a magnetic ionic liquid, and its magnetic susceptibility was estimated to be approximately 4.19 × 10−5 emu g−1.10 In addition, the formation constant logarithm between Fe3+ and phosphate ions is ∼3.5.14 Thus, Fe3+ can chelate with phosphatecontaining molecules with high affinity.15,16 The strong interaction can also be predicated from the hard and soft acids and bases (HSAB) theory.17 Fe3+ is a hard acid, whereas phosphate is a hard base. In addition, two other magnetic ions, namely, Co2+ and Ni2+,18 are intermediate acids, which can also interact with either hard or soft bases.17 Hence, Co2+ and Ni2+ should also display high affinity toward phosphate groups.17 In addition, the surface of bacteria and mammalian cells contain abundant functional groups, such as phosphates and carboxylates.19−22 Thus, for proof of concept, pyrophosphate (PPi), bacteria, and mammalian cells were selected as model samples to interact with magnetic ions. The feasibility of using magnetic ions (Fe3+, Co2+, and Ni2+) as probes to trap and to induce magnetic properties toward PPi, bacteria, and mammalian cells was investigated in this study. Received: June 1, 2018 Accepted: August 15, 2018 Published: August 15, 2018 A
DOI: 10.1021/acsami.8b09100 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
■
RESULTS AND DISCUSSION To demonstrate the feasibility of using magnetic ions to implement sufficient magnetism to nonmagnetic species, we used PPi, that is, a diphosphate, as a model sample to chelate with magnetic ions (Fe3+, Co2+, and Ni2+). We assumed that magnetic conjugates can be formed by using PPi to bridge a number of magnetic ions through phosphate−metal ion chelation in a limited space. For example, when using Fe3+ as an example, the conjugates resulting from Fe3+ and PPi may be formed after mixing (cartoon on the left-hand side in Scheme 1). PPi is added to the bottle containing Fe3+ (left), Scheme 1. Cartoon Illustration of Magnetic Manipulation of the Fe3+−PPi Complexes
Figure 1. Hysteresis curves of the samples containing (A) Fe3+−PPi, (B) Co2+−PPi, and (C) Ni2+−PPi complexes obtained at 10 K (black) and 300 K (red). (Inset) Photographs of the bottles containing the complexes without (left) and with (right) applying an external neodymium magnet (∼4000 gauss). The complexes were prepared by vortex-mixing aqueous metal chloride (0.1 M, 0.1 mL) and aqueous PPi (0.04 M, 0.1 mL) for 30 min followed by standing next to an external neodymium magnet for 1 h.
and the polymeric structure is formed by complexing a number of Fe3+ using PPi (cartoon on the center in Scheme 1). Without applying an external magnetic field, the magnetic dipole moment of Fe3+ on the polymeric structure randomly orients (cartoon on the center in Scheme 1). When applying an external magnetic field, the magnetic dipole moment of Fe3+ can be aligned in the same direction (cartoon on the righthand side in Scheme 1). Owing to the increased magnetic moment, the magnetism of the polymeric structure may become evident. For proof of the concept, we initially mixed aqueous FeCl3 (0.1 M, 0.1 mL) with PPi (0.04 M, 0.1 mL) under stirring. White-yellowish precipitates appeared after mixing (the bottle on the left-hand side in the photograph of Figure 1A), which resulted from the formation of heavy Fe3+−PPi complexes with poor water solubility. The bottle on the right-hand side in the photograph of Figure 1A shows that the precipitates resulting from the Fe3+−PPi complexes can be magnetically aggregated on the wall of the bottle adjacent to an external neodymium magnet (∼4000 gauss). That is, the complexes possessed magnetic property. The magnetic property of Fe3+−PPi complex-derived precipitates were further examined by superconducting quantum interference device (SQUID). Figure 1A presents the hysteresis curves of the precipitates resulting from the Fe3+−PPi complexes obtained at 300 K (red) and 10 K (black). When the external magnetic field was increased, the magnetization of the sample was also increased. Furthermore, the magnetic property became evident when analysis was conducted at 10 K (black). According to the Curie’s law,23 the magnetization of a paramagnetic material increases with decreased temperatures. The magnetic susceptibility of the
Fe3+−PPi complex was 4.56 × 10−5 emu g−1 at 300 K according to the slope of the hysteresis curve shown in Figure 1A. This value is slightly higher than that obtained from a previously explored magnetic ion liquid, that is, 4.19 × 10−5 emu g−1.10 Moreover, the magnetism of Fe3+−PPi complexes can be observed only when precipitates are formed. We estimated the solubility product constant (ksp) of Fe4PPi3 is ∼3 × 10−12 based on the observation of the disappearance when the sample containing aqueous FeCl3 (0.1 M) with PPi (3.13 mM) (Figure S1A). As the concentration of PPi in the mixture was raised to 6.25 mM, precipitates started to appear. We also mixed two other magnetic ions, that is, Co2+ and Ni2+, individually with PPi. The precipitates could be easily obtained when mixing aqueous CoCl2 with PPi. However, the precipitates disappeared when the sample containing CoCl2 (0.1 M) and PPi (0.39 mM). As the concentration of PPi in the sample was increased to 0.78 mM, precipitates were observed (Figure S1B). Accordingly, we estimated ksp of Co2PPi to be ∼4 × 10−6. In addition, the precipitates resulting from the Ni2PPi disappeared when the sample containing NiCl2 (0.1 M) and PPi (24.4 μM) (Figure S1C). However, as the concentration of PPi was increased to 48.7 μM, precipitates appeared (Figure S1C). Thus, the ksp of Ni2PPi was estimated to be ∼2 × 10−7. Figure 1B,C illustrates the hysteresis curves obtained from the Co2+−PPi and Ni2+−PPi complexes, respectively. The photographs next to these two figures show the corresponding results obtained without (left) and with (right) applying a magnet adjacent to the wall of the bottles. B
DOI: 10.1021/acsami.8b09100 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 2. Photographs of the aqueous samples containing (A) mixture of PPi (0.04 M, 0.1 mL) and FeCl3 (0.1 M, 0.1 mL) that were added with EDTA (0.1 M) with the volumes of (B) 0.1 and (C) 0.2 mL and aqueous aluminum chloride (0.1 mL) with the concentrations of (D) 1 and (E) 2 M. Photographs of the aqueous samples containing (F) mixture of PPi (0.04 M, 0.1 mL) and CoCl2 (0.1 M, 0.1 mL) that were added with EDTA (0.1 M) with the volumes of (G) 0.1 and (H) 0.2 mL and aqueous aluminum chloride (0.1 mL) with the concentrations of (I) 1 and (J) 2 M. Photographs of the aqueous samples containing (K) mixture of PPi (0.04 M, 0.1 mL) and NiCl2 (0.1 M, 0.1 mL) that were added with EDTA (0.1 M) with the volumes of (L) 0.1 mL and (M) 0.2 mL and aqueous aluminum chloride (0.1 mL) with the concentrations of (N) 1 and (O) 2 M. A neodymium magnet was placed adjacent to each bottle on the left-hand side.
Presumably, the Fe3+−PPi conjugates possessed better water solubility than the conjugates resulting from Co2+−PPi and Ni2+−PPi. Thus, the Fe3+−PPi conjugates were relatively difficult to be magnetically aggregated. Moreover, we also investigated the relative ratios of magnetic ions to PPi in the resultant precipitates by examining the UV−vis absorption spectra of the supernatants of the samples obtained before (black band) and after (red band) the magnetic ions conjugated with PPi (Figure S3). On the basis of the results in Figure S3 and also considering ksp of each magnetic ion−PPi conjugate, the relative ratios of Fe3+ to PPi, Co2+ to PPi, and Ni2+ to PPi in the precipitates were estimated to be ∼2.4, ∼2.5, and ∼2.4, respectively. Namely, ∼5 magnetic ions bind to 2 PPi units in the precipitates. Presumably, a network structure with this magnetic ion−PPi combination as shown in Scheme 1 was formed, leading to the formation of precipitates. To further verify if the binding interaction between these magnetic ions and PPi in the complexes was reversible, ethylenediaminetetraacetic acid (EDTA) and Al3+ were used as competing agents. EDTA is a ligand that possesses good chelating capability with transition metals,19 and Al3+ is a hard acid with high affinity toward phosphates on PPi according to HSAB theory.17 When EDTA (0.1 M, 0.1 mL) was added to the sample containing the Fe3+−PPi complexes as shown in Figure 2A, the solution became transparent, and most of the precipitates disappeared (Figure 2B). Nevertheless, trace white-yellowish precipitates could still be attracted by the external magnetic field. When the amount of EDTA was further increased, the sample became relatively transparent (Figure 2C). This phenomenon attributed to that EDTA can replace PPi to bind with Fe3+, which resulted in the disappearance of magnetic precipitates resulting from Fe3+− PPi complexes. Owing to the good water solubility of the Fe3+−EDTA complexes, no apparent precipitates were observed. Thus, no visible aggregates adjacent to the magnet
The aqueous Co2+ and Ni2+ solutions looked pinkish and greenish, respectively. Their SQUID results were similar to that in Figure 1A. Evident paramagnetic properties were also observed in the precipitates resulting from these two complexes, which can be magnetically aggregated by an external magnet (photographs in Figure 1B,C). In addition, the magnetic susceptibility of the Co2+−PPi complexes was ∼6.16 × 10−5 emu g−1 at 300 K, which was higher than that obtained from the Fe3+−PPi complexes. The magnetic susceptibility of the Ni2+−PPi was ∼2.28 × 10−5 emu g−1 at 300 K, which was the lowest among these three magnetic ion− PPi complexes. These results indicated that we have successfully used a simple method to direct nonmagnetic species, PPi, to quickly gain observable magnetism by complexing with magnetic ions. The magnetism is high enough to be observed by the naked eye. In addition, we also examined the separation efficiency of these three magnetic ion−PPi conjugates in aqueous solution. In the beginning (0 min) of placing a magnet next to the sample vials containing different magnetic ion−PPi conjugates, the sample solutions looked quite blurred (Figure S2). Fe3+− PPi conjugates slowly aggregated (Figure S2A). Within 5 min, only a very small amount of the conjugates (indicated by red arrows) aggregated on the wall of the vial next to a magnet. The precipitates preferred to aggregate on the wall of the vial next to the two edges (bottom and top) of the magnet because the two edges possessed higher magnetic strength than the center of the magnet. After 1 h, the magnetic aggregates became apparent (indicated by red arrows). However, the solution was still blurred. On the other hand, the Co2+−PPi (Figure S2B) and Ni2+−PPi conjugates (Figure S2C) quickly aggregated after placing a magnet next to the sample vials containing these conjugates within 5 min. After 30 min, clear solution was found in each sample vial. Namely, the conjugates resulting from Co2+−PPi and Ni2+−PPi conjugates can be magnetically aggregated faster than the Fe3+−PPi conjugates. C
DOI: 10.1021/acsami.8b09100 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces
Figure 3. Photographs of the samples containing (A) aqueous FeCl3 (0.1 M) alone, S. aureus suspension (2 mg mL−1) prepared in Tris buffer (pH 7.4) containing Fe3+ (0.1 M) (B) without and (C) with applying an external magnet, E. coli O157:H7 suspension (2 mg mL−1) prepared in Tris buffer (pH 7.4) containing Fe3+ (0.1 M) (D) without and (E) with applying an external magnet. (F) Aqueous CoCl2 (0.1 M), S. aureus suspension (2 mg mL−1) prepared in Tris buffer (pH 7.4) containing Co2+ (0.1 M) (G) without and (H) with applying an external magnet, E. coli O157:H7 suspension (2 mg mL−1) prepared in Tris buffer (pH 7.4) containing Co2+ (0.1 M) (I) without and (J) with applying an external magnet. (K) Aqueous NiCl2, S. aureus suspension (2 mg mL−1) prepared in Tris buffer (pH 7.4) containing Ni2+ (0.1 M) (L) without and (M) with applying an external magnet, E. coli O157:H7 suspension (2 mg mL−1) prepared in Tris buffer (pH 7.4) containing (0.1 M) (N) without and (O) with applying an external magnet. After addition of metal ions, the bacterial suspensions were heated in a microwave oven (power: 180 W) for three cycles (90 s cycle−1).
used as the substrate to chelate with a number of magnetic ions. The conjugates of E. coli and magnetic ions may also gain observable magnetism. In addition, Staphylococcus aureus, a Gram-positive bacterium, possesses a cell wall that is composed of peptidoglycans and abundant phosphatecontaining species, including teichoic acid and lipoteichoic.20 Consequently, S. aureus may also gain noticeable magnetism after chelation with a sufficient number of magnetic ions, such as Fe3+, Co2+, and Ni2+. We mixed bacteria with magnetic ions in aqueous solution under microwave heating (power, 180 W) for 1.5 min (three cycles) to shorten the sample preparation time. The magnetism of the resultant samples was investigated by applying an external magnet. Figure 3A shows the photograph of a bottle that contained aqueous Fe3+. An external neodymium magnet was placed adjacent to the bottle on the left-hand side. No apparent aggregates were observed near the wall adjacent to the magnet. Figure 3B,C shows the photographs of the samples containing the mixture of Fe3+ and S. aureus suspension without and with application of an external magnet, respectively. Evidently, S. aureus can be magnetically aggregated after mixing with Fe3+. The resultant conjugates can be aggregated on the wall of the bottle adjacent to an external magnet (Figure 3C). Figure 3D,E shows the photographs of the samples containing the mixture of E. coli O157:H7 and Fe3+, without and with application of an external magnet, respectively. The conjugates of E. coli and Fe3+ were apparently aggregated to the wall of the bottle adjacent to the external magnet on the left-hand side (Figure 3E). We further used Co2+ and Ni2+ to bind with S. aureus and E. coli O157:H7. Figure 3F−O shows the corresponding photographs. The bacteria can be aggregated to the wall of the bottles adjacent to the magnet with addition of Co2+ (Figure 3H,J) and Ni2+ (Figure 3M,O). These magnetic ions can quickly bind to the surface of these bacteria, thereby resulting in the feasibility of magnetic separation of bacteria by an external magnet. In addition, we also noticed that the survival rate of bacteria is related to the concentration of spiked Fe3+ in the samples.
were observed. Additionally, the magnetic precipitates gradually disappeared with increased amount of Al3+ (Figure 2D,E). It was because that Al3+ had replaced a number of metal centers in the complexes, leading to the loss of the magnetism. Owing to poor water solubility, the precipitates resulting from Fe3+−Al3+−PPi complexes were still observed. However, the precipitates could not be tightly aggregated on the wall adjacent to the external magnet. The precipitates completely disappeared when 1 mL of Al3+ (2 M) was added to the Fe3+− PPi complex precipitates (results not shown). We further added EDTA and Al3+ to the samples containing the complexes of Co2+−PPi (Figure 2F−J) and Ni2+−PPi (Figure 2K−O) to examine if similar results could be obtained. Upon addition of EDTA and Al3+ to these precipitates resulting from the Co2+−PPi (Figure 2F) and Ni2+−PPi complexes (Figure 2K), the precipitates gradually disappeared. The loss of the magnetism was owing to the replacement of ligands and magnetic ions from the complexes. These results suggested that imposing magnetic property to nonmagnetic species by simply using magnetic ions is possible. Furthermore, when magnetic ion-containing complexes displayed poor water solubility, the magnetism of the resultant precipitates can be readily observed by the naked eye. It was because that the generated precipitates composed of magnetic ion−PPi complexes can carry a sufficient number of magnetic ions. As a consequence, the gained magnetism can be readily observed by the naked eye. On the basis of these results, we believed that bacteria can be used as an ideal model to further demonstrate the feasibility of imposing magnetism on nonmagnetic species by simple mixing with magnetic ions. Bacteria can suspend in aqueous solution and gradually precipitated because of their large sizes in micrometer scales. Bacterial cell surfaces also contain abundant functional groups, such as phosphates and carboxylates.19−21 For example, Escherichia coli, a Gramnegative bacterium, possesses an outer membrane, which contains abundant phosphate units.20,21 Thus, E. coli can be D
DOI: 10.1021/acsami.8b09100 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces Figure S4A−J shows the resultant photographs obtained after incubating E. coli and S. aureus with Fe3+ at different concentrations, respectively. When the concentration of Fe3+ in the bacteria samples was 0.1 M, only few bacterial colonies were found on the resultant agar plates (Figure S4B,G). As the concentration of Fe3+ was decreased, the survival rate of bacteria was increased. As the concentration of Fe3+ was reduced to 1 mM, the bacterial growth conditions of E. coli (Figure S4D) and S. aureus (Figure S4I) were similar to those obtained without being treated by Fe3+ (Figure S4A,F). In addition, one may wonder that heating bacterial samples in a microwave oven (power: 180 W, 90 s cycle−1 × 3) that we used to prepare the conjugates of magnetic ion−bacterium may be the reason to cause the death of the bacteria. Thus, we examined how the temperature changed in the samples (0.2 mL) containing bacteria (2 mg mL−1 (=OD600 = ∼3.12) and magnetic ions (0.1 M) after microwave heating treatment. The starting temperature was ∼25 °C. Table S1 shows the three replicates of the temperature measurement in the samples containing S. aureus and Fe3+ (0.1 M) that had been heated under a microwave oven (power: 180 W) for three cycles (90 s cycle−1). The resultant temperature was ∼30 °C after microwave heating for three cycles. It is unlikely that such a temperature can kill bacteria. To further confirm the results, we prepared the bacterial samples (0.3 mL) containing S. aureus (OD600 = ∼1) in the presence and absence of Fe3+ (0.1 M) with and without under microwave heating (power: 180 W) for three heating cycles (90 s cycle−1) followed by overnightculture on agar plates. Figure S5 shows the photographs of the resultant agar plates. Apparently, bacteria grew well in the absence of Fe3+ without (Figure S5A) and with undergoing microwave heating (Figure S5C). On the other hand, only few colonies were found in the agar plates in the presence of Fe3+ (0.1 M) without (Figure S5B) and with (Figure S5D) undergoing microwave heating. The results indicated that the presence of Fe3+ in the bacterial sample affected the survival rate of bacteria, whereas microwave heating (power: 180 W; 90 s cycle−1 × 3) did not cause an apparent adverse effect on the growth of bacteria. To further verify whether the bacterial cells still remained intact in the presence of magnetic ions on the surface of the bacterial cells, we used scanning electron microscopy (SEM) and the equipped energy dispersive X-ray (EDX) spectroscopy to examine the morphology and the elemental compositions of the Co2+−bacterium conjugates, respectively. Figures S6A and S7A show the SEM images of the conjugates resulting from Co2+−S. aureus and Co2+−E. coli O157:H7, respectively. Apparently, the bacterial cells remained intact although the shape did not look very perfect. It was because that we had boiled the bacterial samples at 100 °C for 1 h to inactive bacteria prior to conducting the experiments. Nevertheless, S. aureus still remained in their spherical shapes, whereas E. coli O157:H7 still possessed rod shapes. Figures S6B and S7B show the corresponding EDX spectra obtained from Co2+−S. aureus and Co2+−E. coli O157:H7, respectively. The results indicated that the Co2+−S. aureus and Co2+−E. coli O157:H7 are composed of ∼6 and ∼8% cobalt, respectively. In addition, we also used fivefold diluted fetal bovine serum (FBS) spiked with S. aureus as the sample to examine if magnetic ions could selectively trap bacteria from complex serum samples. We used Co2+ as the affinity probes. Figure 4A,B shows the photographs of the samples containing FBS alone and FBS spiked with S. aureus, respectively, in which
Figure 4. Photographs of the fivefold diluted FSB samples containing Co2+ (0.1 M) and (A) without and (B) with S. aureus (2 mg mL−1) prepared in Tris buffer (2 mM, pH 7.4). The samples were heated in a microwave oven (power: 180 W) for 3 cycles (heating cycles: 90 s cycle−1) followed by standing next to a magnet for 6 h before taking photographs. (C) Optical microscopic image of the sample collected from the magnetically isolated conjugates obtained from panel (B). The supernatant in panel (B) was removed, and the remained conjugates next to the magnet were taken by using deionized water (30 μL) to suspend a small amount of conjugates from the wall. The suspension (8 μL) was deposited on a glass slide. After drying, the remaining sample was deposited with methylene blue (2 μL, 0.2% (w/ v)). After standing for 2 min, the excess methylene blue was rinsed by deionized water. The glass slide was covered and sealed with a cover glass slide.
both samples were added with Co2+ (0.1 M). The samples were incubated in a microwave oven (power: 180 W) for three cycles (heating cycle: 90 s cycle−1) followed by standing next to a magnet. Apparently, precipitates attached next to the wall of the sample vial were observed only in the FBS sample containing S. aureus mixed with Co2+ (Figure 4B). Figure 4C shows the image of the magnetic isolates of the Co2+−S. aureus conjugates from the FBS sample. Methylene blue that can stain bacterial cells was used to stain the magnetic isolated bacteria. The image showed clearly the conjugates containing bacterial cells with spherical shapes. The results indicated that magnetic ions still have the capability to interact with bacteria in complex serum samples. One may wonder if the developed method is suitable for separation of mammalian cells. The cell membrane of mammalian cells generally contains abundant phospholipids,22 in which phosphate can chelate with magnetic ions. Figure 5A,B shows the photographs of the samples containing breast cancer cells T47D in the presence of Fe3+ (0.1 M) taken at the time points 0 and 30 min, respectively, after mixing followed by magnetic separation. It is apparent that the cells were aggregated on the wall of the sample vial after magnetic separation for 30 min (Figure 5B). That is, our method can also be suitable for magnetic separation of mammalian cells. To examine whether the cells could survive in the presence of Fe3+ (0.1 M), trypan blue, which can penetrate cell membrane of dead cells easily, was used for staining the cells after the cells were treated by Fe3+ (0.1 M). Figure 5C shows the resultant E
DOI: 10.1021/acsami.8b09100 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
ACS Applied Materials & Interfaces
■
Research Article
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.8b09100. Additional experimental details including reagents and materials, instrumentation, preparation of bacterial samples, and preparation of SQUID samples; additional table including the measurement of the temperature change after microwave heating; and additional figures including estimation of solubility product constant (ksp), examination of separation efficiency, examination of the ratio of magnetic ions to PPi in magnetic ion−PPi conjugates, Fe3+ tolerance of bacteria, microwave heating effects against bacteria, SEM image and EDX of Co2+−S. aureus conjugates, and SEM image and EDX of Co2+−E. coli O157:H7 conjugates (PDF)
Figure 5. Examination of magnetic separation of mammalian cells and Fe3+ effects. Photographs of the samples (0.2 mL) containing breast cancer cells (T47D) (7.5 × 106 cells mL−1) and Fe3+ (0.1 M) taken at the time points (A) 0 and (B) 30 min after mixing for 30 min followed by magnetic separation. (C) Optical image of T47D cells treated by Fe3+ (0.1 M) followed by stained with trypan blue (50%).
■
optical microscopic images. All of the cells had been stained by trypan blue, indicating that they were dead. Namely, this approach can be used for separation of mammalian cells, but the cells are damaged after mixing with Fe3+ (0.1 M).
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Phone: +886-3-5131527.
■
ORCID
Yu-Chie Chen: 0000-0003-2253-4049
CONCLUSIONS MNPs have been extensively used as trapping probes for species of interest owing to their unique magnetic property and ease of isolation. However, to generate MNPs and to impose targeting capability on the MNPs are time-consuming. We herein explore a new and straightforward method to direct nonmagnetic specie to gain magnetic properties by simply mixing the model samples including PPi and bacteria with magnetic ions. The model samples attached by magnetic ions can be simply isolated by applying an external magnetic field. To the best of our knowledge, the feasibility of using magnetic ions to yield the magnetic property of nonmagnetic species such as bacteria for magnetic isolation is demonstrated for the first time. Our results demonstrate that if certain species can chelate with magnetic ions at a sufficient number, imposing magnetic property to such nonmagnetic species of interest is possible. These three magnetic ions have high affinity to those functional groups containing nitrogen and oxygen. Thus, this approach can be further extended to nonmagnetic nanoprobes, in which their surfaces are functionalized with functional groups containing nitrogen and oxygen. The conjugates of nonmagnetic nanoprobes with additional magnetic ions and their target species can be readily isolated by magnetic aggregation. Therefore, the time spent on isolation of target species can be shortened to a great extent by implementing this simple approach. In addition, we also believe that it may be potentially possible to employ the current approach to water treatment for elimination of bacteria by simply adding magnetic ions such as iron ions in waste water. The bacteria attached by magnetic ions can be easily removed through magnetic isolation. In addition, mammalian cells can also be magnetically isolated using the developed method. However, the presence of Fe3+ with the concentration of 0.1 M has adverse effects on the growth of bacteria and mammalian cells. Most of bacteria and mammalian cells cannot survive after being treated by Fe3+ (0.1 M). Nevertheless, the magnetically isolated bacteria/cells can be further identified by suitable analytical methods. Namely, the developed method should be potentially useful in sample preparation when handling bacteria/cell containing samples.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank the Ministry of Science and Technology of Taiwan (MOST 105-2113-M-009-022-MY3) for the financial support of this research.
■
REFERENCES
(1) Hao, R.; Xing, R.; Xu, Z.; Hou, Y.; Gao, S.; Sun, S. Synthesis, Functionalization, and Biomedical Applications of Multifunctional Magnetic Nanoparticles. Adv. Mater. 2010, 22, 2729−2742. (2) Wei, Y.; Han, B.; Hu, X.; Lin, Y.; Wang, X.; Deng, X. Synthesis of Fe3O4 Nanoparticles and their Magnetic Properties. Procedia Eng. 2012, 27, 632−637. (3) Pan, Y.; Du, X.; Zhao, F.; Xu, B. Magnetic Nanoparticles for the Manipulation of Proteins and Cells. Chem. Soc. Rev. 2012, 41, 2912− 2942. (4) Gawande, M. B.; Branco, P. S.; Varma, R. S. Nano-magnetite (Fe3O4) as a support for recyclable catalysts in the development of sustainable methodologies. Chem. Soc. Rev. 2013, 42, 3371−3393. (5) Kolhatkar, A.; Jamison, A.; Litvinov, D.; Willson, R.; Lee, T. Tuning the Magnetic Properties of Nanoparticles. Int. J. Mol. Sci. 2013, 14, 15977−16009. (6) Fang, J.; Chen, Y.-C. Nanomaterials for Photohyperthermia: A Review. Curr. Pharm. Des. 2013, 19, 6622−6634. (7) Beveridge, J. S.; Stephens, J. R.; Williams, M. E. The Use of Magnetic Nanoparticles in Analytical Chemistry. Annu. Rev. Anal. Chem. 2011, 4, 251−273. (8) Laurent, S.; Forge, D.; Port, M.; Roch, A.; Robic, C.; Vander Elst, L.; Muller, R. N. Magnetic Iron Oxide Nanoparticles: Synthesis, Stabilization, Vectorization, Physicochemical Characterizations, and Biological Applications. Chem. Rev. 2008, 108, 2064−2110. (9) Hayashi, S.; Hamaguchi, H.-o. Discovery of a Magnetic Ionic Liquid [bmim]FeCl4. Chem. Lett. 2004, 33, 1590−1591. (10) Hayashi, S.; Saha, S.; Hamaguchi, H. A new class of magnetic fluids: bmim[FeCl/sub 4/] and nbmim[FeCl/sub 4/] ionic liquids. IEEE Trans. Magn. 2006, 42, 12−14. (11) Del Sesto, R. E.; McCleskey, T. M.; Burrell, A. K.; Baker, G. A.; Thompson, J. D.; Scott, B. L.; Wilkes, J. S.; Williams, P. Structure and Magnetic Behavior of Transition Metal Based Ionic Liquids. Chem. Commun. 2008, 447−449.
F
DOI: 10.1021/acsami.8b09100 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
Research Article
ACS Applied Materials & Interfaces (12) Santos, E.; Albo, J.; Rosatella, A.; Afonso, C. A. M.; Irabien, Á . Synthesis and characterization of Magnetic Ionic Liquids (MILs) for CO2separation. J. Chem. Technol. Biotechnol. 2014, 89, 866−871. (13) Branco, A.; Branco, L. C.; Pina, F. Electrochromic and Magnetic Ionic Liquids. Chem. Commun. 2011, 47, 2300−2302. (14) Wilhelmy, R. B.; Patel, R. C.; Matijevic, E. Thermodynamics and Kinetics of Aqueous Ferric Phosphate Complex Formation. Inorg. Chem. 1985, 24, 3290−3297. (15) Wu, S.-P.; Chen, Y.-P.; Sung, Y.-M. Colorimetric detection of Fe3+ ions using pyrophosphate functionalized gold nanoparticles. Analyst 2011, 136, 1887−1891. (16) Li, P.-H.; Lin, J.-Y.; Chen, C.-T.; Ciou, W.-R.; Chan, P.-H.; Luo, L.; Hsu, H.-Y.; Diau, E. W.-G.; Chen, Y.-C. Using Gold Nanoclusters as Selective Luminescent Probes for PhosphateContaining Metabolites. Anal. Chem. 2012, 84, 5484−5488. (17) Pearson, R. G. Hard and Soft Acids and Bases. J. Am. Chem. Soc. 1963, 85, 3533−3539. (18) Jiles, D. C. Introduction to Magnetism and Magnetic Materials, 3rd ed.; CRC press, 2015. (19) Yu, T.-J.; Li, P.-H.; Tseng, T.-W.; Chen, Y.-C. Multifunctional Fe3O4/alumina core/shell MNPs as photothermal agents for targeted hyperthermia of nosocomial and antibiotic-resistant bacteria. Nanomedicine 2011, 6, 1353−1363. (20) Brown, L.; Wolf, J. M.; Prados-Rosales, R.; Casadevall, A. Through the Wall: Extracellular Vesicles in Gram-positive Bacteria, Mycobacteria and Fungi. Nat. Rev. Microbiol. 2015, 13, 620−630. (21) Reddy, P. M.; Chang, K.-C.; Liu, Z.-J.; Chen, C.-T.; Ho, Y.-P. Functionalized Magnetic Iron Oxide (Fe3O4) Nanoparticles for Capturing Gram-positive and Gram-negative Bacteria. J. Biomed. Nanotechnol. 2014, 10, 1429−1439. (22) Alberts, B.; Alexander, J.; Lewis, J.; Raff, M.; Roberts, K.; Walter, P. Molecular Biology of the Cell, 4th ed.; Garland Science: New York, 2002. (23) Shiraishi, Y.; Nishimura, G.; Hirai, T.; Komasawa, I. Separation of Transition Metals using Inorganic Adsorbents Modified with Chelating Ligands. Ind. Eng. Chem. Res. 2002, 41, 5065−5070.
G
DOI: 10.1021/acsami.8b09100 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX
