Detection and Analysis of Targeted Biological Cells by Electrophoretic

Article Cite This: Anal. Chem. XXXX, XXX, XXX-XXX

pubs.acs.org/ac

Detection and Analysis of Targeted Biological Cells by Electrophoretic Coulter Method Yoshikata Nakajima,* Tomofumi Ukai, Toshiaki Shimizu, Kazuhei Ogata, Seiki Iwai, Naohiro Takahashi, Atsushi Aki, Toru Mizuki, Toru Maekawa, and Tatsuro Hanajiri Bio-Nano Electronics Research Center, Toyo University 2100, Kujirai, Kawagoe, Saitama 350-8585, Japan S Supporting Information *

ABSTRACT: Combining the electrophoresis and conventional Coulter methods, we previously proposed the electrophoretic Coulter method (ECM), enabling simultaneous analysis of the size, number, and zeta potential of individual specimens. We validated the ECM experimentally using standard polystyrene particles and red blood cells (RBCs) from sheep; the latter was the first ECM application to biological particles in biotechnology research. However, specimens are prevented from passing through the ECM module aperture, which prevents accurate determination of the zeta potential of each specimen. This problem is caused by electro-osmotic flow (EOF) due to the high zeta potential at the ECM microchannel surfaces. To significantly improve ECM feasibility for biomedicine, we here propose a method to estimate the zeta potential at the ECM microchannel surfaces separate from the zeta potential of each specimen, by investigating the electric-field dependence of the specimen’s experimental electrophoretic velocity. We minimize the zeta potential at the microchannel surfaces by applying an organic-molecule coating, and we suppress the surface zeta potential and its resultant EOF by optimizing the microchannel geometry. We demonstrate that the ECM can distinguish between different biological cells using the differences in zeta potential values and/or sizes. We also demonstrate that the ECM can determine the number of biomolecules attached to individual cells and identify whether the average cell state in an analyzed vial is alive or dead. The high-performance ECM can detect cellular morphology alterations, improve immunologic test sensitivity, and identify cell states (living, dying, and dead); this information is clinically useful for early diagnosis and its followup. small flow paths and reactive and mix chambers on glass or plastic plates, enabling quick separation and precise analysis of microparticles, DNA, and biological cells for even a tiny sample of a chemical solution or blood.8,9 The operating principle of μTAS is based on the fact that the surface conditions of microparticles and biological cells affect their properties, such as their adhesive interaction,3,10−14 size,15,16 and the zeta potential ζ at the particle and cell surfaces.17,18 In particular, ζ is utilized to estimate the particle surface condition and dispersion stability.19,20 This parameter is evaluated from the electrophoretic mobility during electrophoresis using the Smoluchowski21 or Hückel equations,22 with the velocity being measured using laser Doppler velocimetry or moving image analysis.23,24 Furthermore, in the past 10 years, novel analytical procedures for the detection of various targeted components have been proposed, which do not damage cells or require molecular labeling with enzymes, fluorescent dyes, or radioisotopes.25,26 Previously, we proposed the electrophoretic Coulter method (ECM) by combining the electrophoresis and Coulter methods.20 Further, we demonstrated that the ECM enables

B

lood contains many types of cells, including red blood cells (RBCs), leukocytes, platelets, and plasma, which perform a range of different functions, from oxygen transport to antibody production. As the blood properties and constituents are sensitive to changes in the body, such as the occurrence of disease states, blood analysis provides a valuable source of information for clinical analytical procedures.1,2 Blood is pumped by the heart through a network of arteries, veins, and capillaries to the various parts of the body, and it carries a variety of cells as well as nutrients. For instance, circulating tumor cells (CTCs), which are detached from the original solid tumor, are found in the blood of most cancer patients; these CTCs spread through the blood and lymph nodes to secondary tissue.1,3 The infected RBCs found in the blood of malaria patients are another example.4 The merozoites become enlarged in the infected cells, diffuse into the bloodstream, and infect immature RBCs. Thus, the separation and analysis of targeted components from the blood can provide valuable information; however, such separation and analysis constitutes a challenging problem from the engineering and medical perspectives.5−7 Recently, micrototal analysis systems (μ-TAS) have been developed for the efficient analysis of biological fluids. μ-TAS use semiconductor microfabrication technology to integrate © XXXX American Chemical Society

Received: August 30, 2017 Accepted: October 24, 2017 Published: October 24, 2017 A

DOI: 10.1021/acs.analchem.7b03533 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

flow is stopped by controlling the solution volume in each insert hole. Specimen Analysis Using ECM. Figure 2 shows the shape of the microchannel and its equivalent circuit. Table 1 presents

simultaneous measurement of the size, ζ, and number of the same type of particles and biological cells using a simplified measurement system with a microchannel.20 However, we had not yet verified the accuracy of the ECM for identifying different types of biological cells in the same solution based on the size and ζ differences, or its precision in detecting changes on the surfaces of biological cells using the ζ changes due to the molecular reactions. In this study, we verify the validity of our proposed ECM for the separation and analysis of targeted blood components. First, we propose a method of estimating the ζ at the surfaces of the ECM microchannels separately from the ζ of each specimen. We also minimize the ζ at the microchannel surfaces and its resultant electroosmotic flow (EOF) by coating the microchannels with organic molecules and designing their geometries to improve the accuracy and precision of our ECM. In addition, we investigate whether an individual component of interest can be identified using the size and ζ differences of the various blood components. Finally, we investigate whether our ECM can determine whether the components of interest are present, and whether a targeted cell is alive or dead, using the differences in the size or ζ distributions.

■

Figure 2. Detailed structure of microchannels in ECM system and its equivalent circuit.

EXPERIMENTAL SECTION Measurement Setup for ECM. ECM System Operation. Figure 1 is a schematic diagram of the experimental system

Table 1. Detailed Dimensions of Microchannels for RBCs and IM-9 Cells length (μm)

width (μm)

height (μm)

la

lb

lc

wa

wc

h

microchannel for RBCs

45

100

3300

12

200

microchannel for IM-9 cells

45

100

5000

18

200

7.5, 10, 11.3 18

the microchannel dimensions for RBCs and IM-9 cells. The microchannel is configured to have five regions, and the total resistance of the solution in the microchannel Rtotal is given by the sum of the electrical resistances Ra, Rb, and Rc for the five regions: R total = Ra + 2R b + 2Rc

Figure 1. Schematic diagram of experimental system for our proposed electrophoretic Coulter method (ECM). Two Pt electrodes are set at both the inlet and outlet of a microchannel. Specimens are moved along the microchannel under application of a DC voltage. The narrow part at the center of the microchannel is the aperture, and the electrical resistance of a solution in the aperture increases as each specimen passes through the aperture.

(1)

To induce electrophoresis, a DC voltage is applied to the solution in the microchannel, and the resultant ion current is continuously measured. As each specimen passes through the aperture, the total resistance of the solution in the microchannel increases from Rtotal to Rtotal′, according to R total′ = Ra′ + 2R b + 2Rc

(2)

Each time a specimen encounters the aperture, the ion current through the aperture decreases and a pulse is generated. The ion current modulation is therefore described by20,27

used in this study. Platinum electrodes are placed at the holes, and a constant direct-current (DC) voltage is applied to the solution in the microchannel to prompt specimen electrophoresis. The applied voltage is 15 V for RBCs and 20 V for human B lymphoblast (IM-9) cells. The resultant ion current is continuously measured using a precision semiconductor parameter analyzer (4156C; Agilent Technologies, U.S.A.) under voltage application. The specimen movement is observed using an inverted microscope and a charge-coupled device camera. The specimen movement caused by pressure-driven

R ΔI I − I′ = = 1 − total I I R total′ =1−

la σhwa

+

w 2lb ln wc σh(wc − wa) a

2arctan(( d 2 ) / ( hwa π ) − ( d 2 )2 ) σπ ( hwa π ) − ( d 2 )2

+

la − d σhwa

+

+

2lc σhwc

w 2lb ln wc σh(wc − wa) a

+

2lc σhwc

(3) B


Article


Figure 3. Fabrication procedure for microchannel in ECM system. (a−c) a resist micropattern is formed on a Si substrate using conventional soft ultraviolet (UV) lithography. (d) PDMS liquid is then poured into the micropatterned resist on the Si substrate. (e) the micropattern-transferred PDMS substrate is attached to a PDMS plate to form microchannels, and the surfaces of the microchannels are coated with MPC molecules to prevent adhesion of the specimens on the microchannel surface.

where I, I′, and ΔI are the ion currents without and with the specimens in the aperture and the pulse amplitude, respectively; l, h, and w are the length, height, and width of each region in the microchannel, respectively; σ is the electric conductivity of the solution; and d is the specimen diameter. Thus, the number of pulses for the ion current corresponds to the number of specimens passing through the aperture, and the ΔI of each pulse corresponds to the size of each specimen. More importantly, the width of each pulse corresponds to the residence time Δt of each specimen in the aperture. The specimen electrophoretic velocity v is then given by v=

la Δt

current pulses caused by the specimens passing through the aperture in the microchannel. Details of the procedure for estimation of the specimen size, ζ, and number are summarized in ref 20. Microchannel Fabrication Procedure. Our ECM module is fabricated using the procedure shown in Figure 3. First, a resist micropattern (SU-8 2010; MicroChem, MA) is formed on a Si substrate using conventional soft ultraviolet (UV) lithography.28 Polydimethylsiloxane (PDMS) liquid (TSE3450; Momentive Performance Materials Inc., Japan) is then poured into the micropatterned resist on the Si substrate. The PDMS liquid is vacuumed for 2 h to defoam a large amount of its constituent bubbles. Then, it is left at rest for 12 h at room temperature to solidify. The solidified PDMS substrate is next peeled from the micropatterned resist on the Si substrate. The micropattern-transferred PDMS substrate is then attached to a PDMS plate, and insert holes with a 2 mm diameter are formed at the ends of the microchannel on the plate to allow injection of specimen-containing solutions. The surfaces of the microchannels are then coated with 2methacryloyloxyethyl phosphorylcholine (MPC) molecules. Lipidure CM5206E (NOF, Japan) dissolved in ethanol is first diluted with deionized water at 40 wt %. In the ECM module, this MPC solution is then injected into the microchannel formed by the PDMS substrate and PDMS plate, and the ECM module is incubated for 12 h at 40 °C. This prevents adhesion of the specimens on the microchannel surface.27,29,30 Characterization of Zeta Potential Causing EOF of Microchannel Wall. In this study, we characterized the zeta potential (ζEO) causing EOF with velocity vEO near the PDMS channel surface and minimized ζEO by redesigning the microchannel geometries at the aperture; hence, the ECM accuracy and precision were improved. As shown in Figure 4, when the specimen moves to the electrode, the total specimen velocity (vtotal) is expressed as

(4)

Meanwhile, the strength of the electric field in the aperture Ea is given by Ea = −

V ( Ra R total ) la

V =− · la

la σhwa la σhwa

+

2lb w ln wc σh(wc − wa) a

+

2lc σhwc

(5)

Here, the electrophoretic mobility μ of the specimens is defined as v μ= Ea (6) ,and the ζ of the specimens is estimated using the Smoluchowski equation,21 according to η μ ζ= ε0εr (7) ,where η is the solution viscosity, ε0 is the vacuum permittivity, and εr is the relative permittivity. By substituting eqs 4 and 5 into eq 6, μ can be expressed as μ=−

la2

1 · V ( Ra R total ) Δt l

2lb w ln wc σh(wc − wa) a

a la2 σhwa + =− · V

+

2lc σhwc

la σhwa

·

1 Δt

(8)

Further, by substituting eq 8 into eq 7, ζ is described as l

a + 2 η la σhwa · · ζ=− ε0εr V

2lb w ln wc σh(wc − wa) a la σhwa

+

2lc σhwc

·

1 Δt

(9)

Thus, we can characterize the number, size, and ζ of the specimens by measuring the number, ΔI, and width (Δt) of the

Figure 4. Total specimen velocity (vtotal) in microchannel during specimen electrophoresis. C


Article

Analytical Chemistry vtotal = vEO + vEP + vD + vPS

(10)

PBS, with all other components such as leukocytes, platelets, and plasma being removed. The RBCs in the PBS were injected into the ECM microchannel through the inset hole using a micropipette. Attachment of IgG Molecules to RBC Surface. The RBCIgGs used in this experiment were prepared as follows: Antisheep RBC antibodies (Rockland, U.S.A.) distributed in a 100-μL PBS solution were added to the sheep RBCs (5 × 107 cells) and mixed together at 4 °C for 12 h. Different IgG concentrations of 50, 100, 200, and 500 μg/mL were used for the ECM measurements; these values were determined on the basis of results yielded by the enzyme-linked immunosorbent assay (ELISA) method (see Supporting Information S1). The solution was centrifuged, and its supernatant solution was replaced by a PBS solution to remove the excess primary IgG molecules. Cultivation of IM-9 Cells and Induction of Their Apoptosis. The apoptotic cells used in this experiment were prepared as follows: IM-9 cells were cultured to a density of 1.0 × 106 cells/ ml in a Roswell Park Memorial Institute (RPMI)-1640 medium (Sigma-Aldrich, U.S.A.), supplemented with 10% fetal bovine serum (FBS), 1% antibiotics (penicillin/streptomycin), 1% GlutaMAX (Thermo Fisher Scientific, U.S.A.), and 1% sodium pyruvate at 37 °C and in a 5% CO2 atmosphere. Further, a 10-μL apoptosis inducer known as camptothecin (Enzo Life Sciences, U.S.A.) was added to IM-9 cells distributed in a 10-ml medium solution. For the ECM measurement, the IM-9 cell/medium solution was centrifuged at 160 G for 5 min. The supernatant liquid was then removed from the suspension, and the same amount of 10-mg/mL bovine serum albumin (BSA) in PBS solution was added to the suspension. The BSA suppressed the release of intracellular protein into the PBS solution. The above procedure was repeated three times in order to completely replace the camptothecin/medium solution with the BSA/PBS solution.

where vEP is the electrophoresis flow, vD is the diffusion flow, and vPS is the pressure gradient flow.31−34 Note that the first and second terms of eq 10, vEO and vEP, depend on the electrical field, whereas the third and fourth terms, vD and vPS, are independent of the electrical field. By substituting eqs 6 and 7 into eq 10, vtotal is obtained as εε vtotal = (ζEP + ζEO) · 0 r ·Ea + vD + vPS η (11) where ζEP is the specimen zeta potential during electrophoresis. Here, vD and vPS can be adjusted to be negligible by controlling the amount of solution in the inset holes. Hence, the observed vtotal should be zero in the absence of the applied electric field.35 Then, vtotal can be expressed as εε vtotal = (ζEP + ζEO) · 0 r ·Ea· η (12) By changing Ea, that is, by adjusting the DC voltage to promote specimen electrophoresis, we can obtain the electricalfield dependence of vtotal, as shown in Figure 5. As a result, we

Figure 5. Electric-field dependence of vtotal in microchannel.

■

can estimate ζEO, if we obtain ζEP through other means. In this study, we obtained ζEP via the laser Doppler microelectrophoresis method (Zetasizer Nano-ZS; Malvern Instruments, Ltd., U.K.). Cell Preparation. In this study, we demonstrated the size and ζ differences between sheep and rabbit RBCs, between pristine sheep RBCs and sheep RBCs with adsorbed immunoglobulin G (RBC-IgGs), and also between IM-9 cells with different cell states caused by an apoptosis inducer. First, we measured the sizes and ζ values of sheep and rabbit RBCs dispersed in the same solution to determine whether individual RBCs were derived from sheep or rabbits. Second, we characterized the differences in size and ζ between the RBCs and RBC-IgGs; hence, we could determine whether the solution contained sheep RBCs only, or whether it also contained RBC-IgGs. Third, we obtained the size and ζ distributions of the IM-9 cells with different states to identify whether an analyzed cell was alive or dead, for the case of cells of different states dispersed in the same solution. Separation of RBC in Whole Blood. The sheep and rabbit RBCs were prepared using the following procedure: Whole sheep blood and whole rabbit blood (Nippon Bio-Supp. Center, Japan) specimens were suspended in phosphatebuffered saline (PBS) of pH 7.4, and the blood/PBS solution was centrifuged at 450 G for 5 min. The supernatant liquid was then removed from the suspension, and the same amount of PBS was added to the suspension. The above procedure was repeated four times in order to suspend only the RBCs in the

RESULTS AND DISCUSSION Effect of EOF on Experimental Specimen Flow in Microchannel. We estimated the ζEO from the electrical-field dependence of vtotal, as shown in Figure 5. Then, we characterized the ζEO causing the vEO near the PDMS channel surface by changing h. The detailed dimensions of the microchannels are provided in Table 1 and Figure 2. Figure 6

Figure 6. Characterization of zeta potential (ζEO) causing electroosmotic flow (EOF) using microchannels with different heights (h).

shows their ζEO for different h. As h was increased, ζEO, i.e., the effect of the EOF on the specimen flow, was suppressed more effectively; this is possibly because the specimen was moving away from the channel walls. This behavior suggests that the EOF influencing a specimen should be minimized using an optimally designed channel; the D


Article


Figure 7. (a) Time (t) dependence of ion current (I) pulses when red blood cells (RBCs) pass through aperture. (b) Histograms of size and zeta potential (ζ) distributions in phosphate-buffered saline (PBS) solutions containing sheep or rabbit RBCs. (c) Histograms of size and ζ distributions in mixed solutions containing both sheep and rabbit RBCs.

Table 2. Average Size and ζ Values for Reaction-Free Sheep RBCs and Those of RBCs with Different Concentrations of IgG Molecules in the Range of 50 to 500 μg/mL specimen

w/o IgG

50 (μg/mL)

100 (μg/mL)

200 (μg/mL)

500 (μg/mL)

diameter d (μm) zeta potential ζ (mV)

4.62 ± 0.30 −13.38 ± 0.85

4.59 ± 0.29 −12.75 ± 1.11

4.43 ± 0.32 −10.39 ± 0.51

4.47 ± 0.30 −9.13 ± 0.73

4.54 ± 0.33 −8.39 ± 0.74

Table 2 lists the average size and ζ values for the reactionfree sheep RBCs and those for the solutions of sheep RBCs with different IgG molecule concentrations, and it shows differences between the ζ values for the sheep RBCs and RBCIgGs, even though the corresponding size values are overlapped. Additionally, the actual molecular mass of a primary IgG molecule was characterized using sodium dodecyl sulfate− poly acrylamide gel electrophoresis (SDS-PAGE),36 yielding a mass of 150 kDa (see Supporting Information S2). Here, under the assumption that all the IgG molecules were attached to the surfaces of the 5 × 107 RBCs, the number of IgG molecules on each individual RBC was estimated to be within the range of 4 × 105−4 × 106 IgG/cell. Figure 8 shows the correlation between the ζ values of the RBCs and the number of IgG molecules attached to the RBC

resultant ζ of the specimen should then be characterized accurately in our ECM systems. Separation of Targeted Components Using Differences in Size and Zeta Potential. As the presence or absence of an individual component of interest can be determined using the size and ζ differences, we measured the sizes and ζ values of individual RBCs in PBS solutions containing sheep or rabbit RBCs; the results were then compared with those for a mixed solution containing sheep and rabbit RBCs. The ECM measurements were conducted using the MPC-coated and redesigned microchannel. Figure 7a shows the time dependence of the ion current pulses generated when the sheep and rabbit RBCs were passing through a microchannel aperture. The ΔI and Δt values in the current pulses correspond to the sizes and ζ values of the RBCs,20 respectively. Figure 7b shows the histograms of the size and ζ distributions in the PBS solutions containing sheep or rabbit RBCs. The average size values were 4.40 ± 0.28 and 6.07 ± 0.38 μm for the sheep and rabbit RBCs, respectively, while the average ζ values were −12.41 ± 1.20 and −6.65 ± 1.08 mV for the sheep and rabbit RBCs, respectively. Figure 7c shows the histograms of the size and ζ distributions in the mixed solutions containing both sheep and rabbit RBCs. The mixed-solution histograms clearly demonstrate that the size and ζ distributions are divided into two groups, corresponding to the values for the sheep and rabbit RBCs derived from the individual solutions. This finding implies that the ECM is useful for the identification and analysis of targeted components in the blood, when obvious differences in size and ζ are present. Detection of Targeted Components Using Differences in Size or Zeta Potential. To employ the ECM to determine whether a solution contained targeted components even when the individual components could not be identified because of their overlapping size or ζ distributions, we characterized a solution of sheep RBCs with no reactions and solutions comprised of sheep RBCs (5 × 107 cells) reacting with different IgG molecule concentrations (i.e., RBC-IgGs) in the range of 50 to 500 μg/mL. Recall that these values were determined on the basis of the results of the ELISA method (see Supporting Information S1).

Figure 8. Correlation between ζ values of RBCs and number of IgG molecules attached to RBC surfaces.

surfaces. Table 2 and Figure 8 demonstrate the following: (1) the ζ values in the RBC-IgG case were shifted lower compared with those of the RBCs; and (2) the ζ values decreased as the number of IgG molecules attached to the RBC surface increased. The correlation between the ζ values and the antigen−antibody reaction on the RBC surface can be explained as follows. The attachment of the IgG molecules to the cell surfaces changes the surface charges slightly, varying the electrophoretic mobility and resultant ζ, as described in refs 25 and 29. Detection of the ζ values of the RBCs enables E


Article


in three categories (normal, early apoptotic, and late apoptotic/ dead), as follows: (1) green fluorescence is emitted from the annexin V-FITC binding to exposed phosphatidylserine on the surfaces of the early apoptotic cells; (2) red fluorescence is emitted from the PI binding to the nucleic acids in late apoptotic or dead cells; and (3) zero fluorescence indicates normal cells. From Figure 10b, it is apparent that the degree of green and red fluorescence increased with the time lapse after dosing. In other words, the number of early and late apoptotic cells or dead cells increased. In addition, Figure 11 shows the viability of the IM-9 cells at different times, which was determined using trypan blue stain.

estimation of the average number of IgG molecules on the individual RBCs along with determination of the presence/ absence of RBC-IgGs in the RBC solution even though an individual cell cannot be identified as an RBC or RBC-IgG. This result implies that our ECM can detect the targeted cells separately from the another type of cells in a solution as long as the difference in their average ζ values exist and can determine the surface states of cell membranes from modification of the morphologic alteration cells ζ away from the original cell ζ values. Identification of Living or Dead Cells Using Differences in Size and/or Zeta Potential. To determine if our ECM can identify whether an individual component of interest is alive or dead, we characterized the size and ζ distributions in solutions of IM-9 cells before and after dosing with the apoptosis inducer. Initially, the effect of the apoptosis inducer on the IM-9 was confirmed as follows: (1) the surface morphologies of the IM-9 were studied using a high-resolution scanning electron microscope (HRSEM; SU6600; Hitachi, Japan; see Supporting Information S3); (2) the IM-9 cell states were analyzed with a confocal microscope (A1; Nikon, Japan), an apoptosis detection kit including annexin V-fluorescein isothyocyanate (FITC), and propidium iodide (PI; Abcam, U.K.); and (3) the viability in IM-9 was measured with an automated cell counter (Countess II; Thermo Fisher Scientific, U.S.A.) and a 0.4% solution of trypan blue stain (Thermo Fisher Scientific, U.S.A.). Figure 9 shows the SEM images. The image taken before dosing with the apoptosis inducer shows cell pseudopodia on

Figure 11. Viability of IM-9 cells as function of time lapse after dosing with apoptosis inducer.

With an increased time lapse after dosing, the viability of the IM-9 cells decreased. These results confirm that the apoptosis inducer used in this study acts to promote apoptotic cell death. Using ECM measurements, we characterized the size and ζ distributions in solutions for different IM-9 cell states; that is, for a solution without the apoptosis inducer and for dosed solutions with time lapses of 3, 24, and 71 h after dosing. Table 1 shows the microchannel dimensions for the IM-9 cells. Figure 12 shows the histograms of the size and ζ distributions in the IM-9 cells for different elapsed times after dosing. These histograms demonstrate that the size and ζ distributions shifted to lower values as the time lapse after dosing with the apoptosis inducer increased. Further, Figure 13 shows the changes in the (a) ζ and (b) sizes of the IM-9 cells, with the time after dosing with the apoptosis inducer being replaced by the viability, using the relationship shown in Figure 11. Figure 13 demonstrates that the absolute ζ and size values of the IM-9 cells decreased as the cell viability decreased. Therefore, our ECM can identify which cell states (normal, early, or late apoptotic, dead) are dominant for IM-9 after dosing with the apoptosis inducer.

Figure 9. Scanning electron microscopy images of IM-9 cells (a) before and (b) after dosing with apoptosis inducer.

the substrate (Figure 9a), whereas the image taken after dosing indicates that the cells lose the ability to form pseudopodia and produce apoptotic bodies (Figure 9b). Figure 10 shows confocal images of the IM-9 cells (a) before dosing with the apoptosis inducer, and the IM-9 cells at a (b) 30 min and (c) 24-h time lapse after dosing. Here, the fluorescent agents/colors can be used to classify the IM-9 cells

Figure 10. Confocal images of IM-9 cells (a) before dosing with apoptosis inducer and at time lapses of (b) 30 min and (c) 24 h after dosing. F


Article


for mixed specimens, we characterized the ζEO causing EOF near the PDMS channel surface by analyzing the electrical-field dependence of vtotal in the ECM module. In addition, we minimized ζEO by applying an organic-molecule coating to the microchannels and redesigning their geometries at the aperture. Hence, we confirmed that our ECM devices with optimized microchannels have sufficient performance to separate and analyze targeted blood components while excluding normal blood cells. The above conclusion is based on the following experimental results. First, we demonstrated that the ECM can classify cell categories with differences in the size and ζ distributions in solutions of sheep or rabbit RBCs, and their mixed solution. Furthermore, we demonstrated that the ECM can clearly estimate the number of attached IgGs on RBC surfaces if the ζ distributions change, even when the size distributions are overlapped. This experiment involved analysis of a reaction-free solution of sheep RBCs and solutions of sheep RBCs reacting with different IgG molecule concentrations. Finally, we demonstrated that the ECM can identify whether the present cell state is alive or dead for solutions of IM-9 cells, both before and after dosing using an apoptosis inducer. This approach can be applied to other biological cells and biomaterials, for cases in which their size and/or ζ distributions are influenced by disease. Therefore, the ECM is promising for realization of quick point-of-care diagnostics, such as blood glucose monitoring for diabetic patients and early stage detection of various diseases in remote healthcare.

Figure 12. Histograms of size and ζ distributions for IM-9 cells for different time lapses after dosing.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.7b03533. Details for measurement of number of IgGs attached to cell surface using ELISA method, estimation of actual molecular mass of primary IgG using SDS-PAGE, sample preparation for SEM observation (PDF)

Figure 13. Changes in (a) ζ and (b) size values of IM-9 cells at different viabilities after dosing with apoptosis inducer.

In principle, the ECM can be applied to various biological cells or biomaterials provided their sizes and/or ζ values change. We propose three approaches to ECM application: (1) for classification of various types of cells in the same vials; (2) for estimation of the number of molecules or nanomaterials attached to cell surfaces; and (3) for cell-state identification. In this study, we performed experimental verification of the ECM for each approach, and ECM will be useful for (1) the detection of RBCs infected with malaria; (2) the counting of blood cells exposed to an allergen that causes hay fever (or other allergens) in the bloodstream; and (3) the analysis of changes in cell states induced by drugs introduced to the cells. In addition, the ECM constitutes a promising analytical technique that can be widely used in analytical chemistry, medical science, and biotechnology. Finally, the ECM has another advantage: it is favorable for use in a compact system, as it has a simple structure and uses electrical signals. Consequently, the ECM module will serve as portable devices for point-of-care applications, which will be useful for early home diagnoses of diseases including malaria, various allergies, and various cancers.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +81 49 239 1375. Fax: +81 49 234 2502. ORCID

Yoshikata Nakajima: 0000-0002-0302-7386 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The present study was partially supported by a Grant for the Program for the Strategic Research Foundation at Private Universities (S1101017) through the Ministry of Education, Culture, Sports, Science, and Technology (MEXT), Japan, to March 2016. Y. Nakajima has also been supported by a Grantin-Aid for Young Scientists (B) KAKENHI 15K18056 from MEXT, Japan, since April 2015. T. Hanajiri has also been supported by a Grant-in-Aid for Scientific Research (C) KAKENHI 17K01394 from MEXT, Japan, since April 2017.

■

CONCLUSIONS We previously verified that the ECM enables simultaneous measurement of the size, ζ, and number of particles and biological cells, when the same type of particles or cells are considered. Here, to improve the ECM accuracy and precision G


Article


■

(32) Henriquez, R. R.; Ito, T.; Sun, L.; Crooks, R. M. Analyst 2004, 129, 478−482. (33) Tallarek, U.; Rapp, E.; Scheenen, T.; Bayer, E.; Van As, H. Anal. Chem. 2000, 72, 2292−2301. (34) Buszewski, B.; Dziubakiewicz, E., Szumski, M. Principles of Electromigration Techniques; Springer: Berlin, 2013; pp 1324−1327. (35) Shimizu, T.; Yoshihara, Y., Nakajima, Y., Hanajiri, T. Presented in part at the International Semiconductor Device Research Symposium, Bethesda, Maryland, December 12, 2013. (36) Johnstone, A.; Thorpe, R. Immunochemistry in Practice, 3 ed; Blackwell Science: Cambridge, 1996; pp 182−193.

REFERENCES

(1) Madigan, M. T.; Martinko, J. M., Parker, J. Brock Biology of Microorganisms, 9th ed.; Prentice Hall, Upper Saddle River, NJ, 2000; pp 966−969. (2) Li, P. C. H.; Microfluidic Lab-on-a-Chip for Chemical and Biological Analysis and Discovery; CRC Press: Boca Raton, 2006; pp 251−292. (3) Alberts, B.; Johnson, A., Lewis, J., Raff, M., Roberts, K., Walter, P., 4th ed.; Molecular Biology of the Cell; Garland Science: New York, 2002; pp 1324−1327. (4) Hoffman, R.; Benz, E. J., Jr.; Silberstein, L. E., Heslop, H., Weitz, J., Anastasi, J., 6th ed.; Hematology: Basic Principles and Practice; Churchill Livingstone: Philadelphia, 2012; pp 2208−2233. (5) Hou, S.; Zhao, H.; Zhao, L.; Shen, Q.; Wei, K. S.; Suh, D. Y.; Nakao, A.; Garcia, M. A.; Song, M.; Lee, T.; Xiong, B.; Luo, S.-C.; Tseng, H.-R.; Yu, H.-H. Adv. Mater. 2013, 25, 1547−1551. (6) Warkiani, M. E.; Khoo, B. L.; Tan, D. S.-W.; Bhagat, A. A. S.; Lim, W.-T.; Yap, Y. S.; Lee, S. C.; Soo, R. A.; Han, J.; Lim, C. T. Analyst 2014, 139, 3245−3255. (7) Asghar, W.; Wan, Y.; Ilyas, A.; Bachoo, R.; Kim, Y.-T.; Iqbal, S. M. Lab Chip 2012, 12, 2345−2352. (8) King, T. L.; Gatimu, E. N.; Bohn, P. W. Biomicrofluidics 2009, 3, 012004. (9) Tulock, J. J.; Shannon, M. A.; Bohn, P. W.; Sweedler, J. V. Anal. Chem. 2004, 76, 6419−6425. (10) Pysher, M. D.; Hayes, M. A. Anal. Chem. 2007, 79, 4552−4557. (11) Mehrishi, J.; Bauer, J. Electrophoresis 2002, 23, 1984−1994. (12) Nagrath, S.; Sequist, L. V.; Maheswaran, S.; Bell, D. W.; Irimia, D.; Ulkus, L.; Smith, M. R.; Kwak, E. L.; Digumarthy, S.; Muzikansky, A.; Ryan, P.; Balis, U. J.; Tompkins, R. G.; Haber, D. A.; Toner, M. Nature 2007, 450, 1235−1239. (13) Wang, S.; Liu, K.; Liu, J.; Yu, Z. T.-F.; Xu, X.; Zhao, L.; Lee, T.; Lee, E. K.; Reiss, J.; Lee, Y.-K.; Chung, L. W. K.; Huang, J.; Rettig, M.; Seligson, D.; Duraiswamy, K. N.; Shen, C. K.-F.; Tseng, H.-R. Angew. Chem., Int. Ed. 2011, 50, 3084−3088. (14) Stott, S. L.; Hsu, C.; Tsukrov, D. I.; Yu, M.; Miyamoto, D. T.; Waltman, B. A.; Rothenberg, S. M.; Shah, A. M.; Smas, M. E.; Korir, G. K.; Floyd, F. P.; Gilman, A. J.; Lord, J. B.; Winokur, D.; Springer, S.; Irimia, D.; Nagrath, S.; Sequist, L. V.; Lee, R. J.; Isselbacher, K. J.; Maheswaran, S.; Haber, D. A.; Toner, M. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 18392−18397. (15) Kalluri, R.; Weinberg, R. A. J. Clin. Invest. 2009, 119, 1420− 1428. (16) Sieben, S.; Bergemann, C.; Lübbe, A.; Brockmann, B.; Rescheleit, D. J. Magn. Magn. Mater. 2001, 225, 175−179. (17) Walter, H.; Widen, K. E. Biochim. Biophys. Acta, Biomembr. 1995, 1234, 184−190. (18) Schultz, N.; Metreveli, G.; Franzreb, M.; Frimmel, F. H.; Syldatk, C. Colloids Surf., B 2008, 66, 39−44. (19) Tokumasu, F.; Ostera, G. R.; Amaratunga, C.; Fairhurst, R. M. Exp. Parasitol. 2012, 131, 245−251. (20) Takahashi, N.; Aki, A.; Ukai, T.; Nakajima, Y.; Maekawa, T.; Hanajiri, T. Sens. Actuators, B 2011, 151, 410−415. (21) Smoluchowski, M. V. Z. Phys. Chem. 1917, 92, 129−168. (22) Hückel, E. Phys. Z. 1924, 25, 204−210. (23) Wilson, W. W.; Wade, M. M.; Holman, S. C.; Champlin, F. R. J. Microbiol. Methods 2001, 43, 153−164. (24) Uzgiris, E. E.; Kaplan, J. H. Anal. Biochem. 1974, 60, 455−461. (25) Aki, A.; Nihei, Y.; Asai, H.; Ukai, T.; Morimoto, H.; Nakajima, Y.; Hanajiri, T.; Maekawa, T. Sens. Actuators, B 2008, 131, 285−289. (26) Akagi, T.; Ichiki, T. Anal. Bioanal. Chem. 2008, 391, 2433−2441. (27) Koch, M.; Evans, A. G. R.; Brunnschweiler, A. J. Micromech. Microeng. 1999, 9, 159−161. (28) Whitesides, G. M.; Stroock, A. D. Phys. Today 2001, 54, 42−48. (29) Aki, A.; Baiju, G. N.; Morimoto, H.; Kumar, D. S.; Maekawa, T. PLoS One 2010, 5, e15641. (30) Schrott, W.; Slouka, Z.; Č ervenka, P.; Ston, J.; Nebyla, M.; Přibyl, M.; Šnita, D. Biomicrofluidics 2009, 3, 044101. (31) Sun, L.; Crooks, R. M. J. Am. Chem. Soc. 2000, 122, 12340− 12345. H


Detection and Analysis of Targeted Biological Cells by Electrophoretic

Recommend Documents