Simultaneous Enhancement of Cell Proliferation and Thermally

May 22, 2012 - The development of large-scale suspension cell cultures using microcarriers has long been a focus of attention in the fields of pharmac...
0 downloads 5 Views 2MB Size
Download PDF
Article pubs.acs.org/Biomac

Simultaneous Enhancement of Cell Proliferation and Thermally Induced Harvest Efficiency Based on Temperature-Responsive Cationic Copolymer-Grafted Microcarriers Atsushi Tamura,† Masanori Nishi,†,‡ Jun Kobayashi,† Kenichi Nagase,† Hirofumi Yajima,‡ Masayuki Yamato,† and Teruo Okano*,† †

Institute of Advanced Biomedical Engineering and Science, Tokyo Women’s Medical University (TWIns), and Global Center of Excellence (COE), 8-1 Kawadacho, Shinjuku, Tokyo 162-8666, Japan ‡ Department of Applied Chemistry, Tokyo University of Science, 12-1 Funagawara-cho, Ichigaya, Shinjuku, Tokyo 162-0826, Japan S Supporting Information *

ABSTRACT: The development of large-scale suspension cell cultures using microcarriers has long been a focus of attention in the fields of pharmacy and biotechnology. Previously, we developed cell-detachable microcarriers based on temperatureresponsive poly(N-isopropylacrylamide) (PIPAAm)-grafted beads, on which adhering cells can be noninvasively harvested by only reducing the temperature without the need for proteolytic enzyme treatment. In this study, to improve the cell harvest efficiency from bead surfaces while maintaining cell adhesion and proliferation properties, we prepared temperatureresponsive cationic copolymer-grafted beads bearing a copolymer brush consisting of IPAAm, positively charged quaternary amine monomer (3-acrylamidopropyl trimethylammonium chloride; APTAC), and hydrophobic monomer (N-tertbutylacrylamide; tBAAm). The incorporation of positively charged APTAC into the grafted copolymer brush facilitated bead dispersibility in a cell culture system containing Chinese hamster ovary (CHO-K1) cells and consequently allowed for enhanced cell proliferation in the system compared to that of unmodified CMPS and conventional PIPAAm homopolymer-grafted beads. Additionally, P(IPAAm-co-APTAC-co-tBAAm) terpolymer-grafted beads exhibited the most rapid and efficient cell detachment behavior after the temperature was reduced to 20 °C, presumably because the highly hydrated APTAC promoted the overall hydration of the P(IPAAm-co-APTAC-co-tBAAm) chains. Therefore, P(IPAAm-co-APTAC-co-tBAAm) terpolymer-grafted microcarriers are effective in facilitating both cell proliferation and thermally induced cell detachment in a suspension culture system.

1. INTRODUCTION The development of large-scale mammalian cell cultivation systems has long been a focus of attention in the fields of pharmacy and biotechnology because mammalian cells are important hosts for the industrial production of pharmaceutical recombinant proteins, including vaccines and antibodies.1,2 Notably, cell cultivation on a synthetic microsphere surface in a stirred suspension, often referred to as a microcarrier culture, represents a crucial technology in the large-scale culturing of anchorage-dependent cells because the surface of the microcarriers provides a large surface area-to-volume ratio compared to a two-dimensional planar surface, thereby maximizing the achievable cell density.3−5 This type of microcarrier culture has recently been utilized for the large-scale expansion of cells used in regenerative medicine including adult somatic cells5,6 and multipotent bone marrow-derived mesenchymal stem cells7,8 to therapeutically useful numbers (107 to 109 cells).9 Thus, the development of novel synthetic microcarriers is beneficial in © 2012 American Chemical Society

advancing both industrial-scale cell cultures and regenerative medicine. To date, various synthetic beads consisting of dextran, glass, poly(styrene), and poly(D,L-lactide) with diameters ranging from 50 to 300 μm have been developed.4−6 However, during the process of passage culture, most conventional beads require repeated trypsinization to harvest adhered cells from bead surfaces; however, this process leads to the degradation of plasma membrane proteins, in turn leading to a reduction in both cell viability and reattachment efficiency.10−12 To solve this problem, our laboratory has developed a facile and noninvasive cell harvest technique based on temperatureresponsive microcarriers composed of chloromethylated poly(styrene) (CMPS) beads bearing dense poly(N-isopropylacryReceived: February 17, 2012 Revised: May 1, 2012 Published: May 22, 2012 1765

dx.doi.org/10.1021/bm300256e | Biomacromolecules 2012, 13, 1765−1773

Biomacromolecules

Article

microcarrier surface by plasma treatment and the covalent attachment of low-molecular-weight amines or cationic polymers often results in enhanced cell adhesion and proliferation due to electrostatic interaction between the surfaces and serum proteins, including an extracellular matrix and/or cellular membrane.19−21 In this study, temperatureresponsive cationic copolymer-grafted beads bearing copolymer brushes composed of IPAAm, positively charged quaternary amine monomer (3-acrylamidopropyl trimethylammonium chloride; APTAC), and hydrophobic monomer (N-tertbutylacrylamide; tBAAm) were constructed on a CMPS bead surface by surface-initiated atom transfer radical polymerization (ATRP). Among the various types of amino groups that could have been used, highly hydrated and positively charged quaternary amino groups were employed in this study because they possess a permanent positive charge and hydrate polymers regardless of the temperature, ionic strength (salt concentration), and pH conditions.22,23 Additionally, the effect of positively charged amine-bearing temperature-responsive polymer surfaces on cell adhesion, proliferation, and thermally induced harvest were also investigated in a suspension culture.

lamide) (PIPAAm) brushes on their outermost surface (Figure 1).13 Because PIPAAm exhibits temperature-dependent hydra-

Figure 1. Schematic illustration of suspension cell culture on the surface of temperature-responsive copolymer-grafted beads.

tion/dehydration or a hydrophilic/hydrophobic transition across its lower critical solution temperature (LCST) at approximately 32 °C, the cell attachment/detachment behavior can be regulated simply by altering the temperature without requiring the use of proteolytic enzymes.10−15 Indeed, PIPAAm-grafted beads allow cells to be cultured on their surface at 37 °C and allow cells to detach themselves when the temperature is reduced to 20 °C (Figure 1). Previous studies have reported that the physicochemical characteristics of PIPAAm-grafted beads, such as the grafted amount of PIPAAm and the bead diameter largely influence their cell adhesion, proliferation, and thermally induced harvest characteristics. Notably, PIPAAm-grafted beads with a small diameter (e.g., 56 μm) exhibit high initial cell adhesion and the most rapid proliferation due to the large number of contacts made by the beads with the cells. However, the thermally induced cell detachment efficiency decreases with decreasing bead diameter because the entire surface of small beads is easily covered with expanded cells that are connected through strong cell−cell junctions.13 Although the cell detachment efficiency can be increased by increasing the amount of grafted PIPAAm or bead diameter, these methods are accompanied by a reduction in initial cell adhesion and proliferation efficiency.13 To overcome this disadvantage, an alternative method using temperatureresponsive PIPAAm-grafted beads is demanded for facilitating the cell harvest efficiency while maintaining cell adhesiveness. To accelerate cell detachment from a PIPAAm-grafted substrate surface, highly hydrated anionic monomer (e.g., 2carboxyisopropylacrylamide and acrylic acid) and IPAAm were copolymerized;16 in this method, the hydrated monomer plays a critical role in facilitating the rapid overall hydration of PIPAAm-based copolymer chains.16 However, the creation of anionic surfaces often results in a reduction in cell adhesion due to electrostatic repulsion between the surfaces and negatively charged plasma membranes.17,18 One possible strategy to overcome this limitation is the copolymerization of IPAAm with positively charged amino group-containing monomers. The incorporation of amino groups onto a planar or

2. MATERIALS AND METHODS 2.1. Materials. m,p-Chloromethylated poly(styrene) (CMPS) beads (200−400 mesh, chlorine content: 2.2 mmol/g) were obtained from Tokyo Chemical Industry (TCI, Tokyo, Japan). The average diameter and specific surface area of the CMPS beads were determined to be 55.5 ± 12.5 μm and 1030 cm2/g using phase contrast microscopic images (n = 295) and a Brunauer−Emmett−Teller (BET) Kr gas adsorption isotherm, respectively. N-Isopropylacrylamide (IPAAm) was kindly provided by Kohjin (Tokyo, Japan) and purified by recrystallization from n-hexane twice. 3-Acrylamidopropyl trimethylammonium chloride (APTAC) was also kindly provided by Kohjin and purified by passing through an inhibitor removal column (Aldrich, Milwaukee, WI). N-tert-Butylacrylamide was obtained from Wako Pure Chemical Industries (Osaka, Japan) and purified by recrystallization from acetone. Tris[(2-dimethylamino)ethyl]amine (Me6TREN) was synthesized according to a previously described procedure and purified by distillation under reduced pressure.24 Copper(I) chloride (CuCl), copper(II) dichloride (CuCl2), and anhydrous 2-propanol were obtained from Wako Pure Chemical Industries and used as received. Ethylenediamine-N,N,N′,N′-tetraacetic acid disodium salt (EDTA), calcein-AM, and propidium iodide (PI) were obtained from Dojindo Laboratories (Kumamoto, Japan) and used as received. The Milli-Q water used in this study was prepared using an ultrapure water purification system (synthesis A10) (Millipore, Billerica, MA). 2.2. Synthesis of Temperature-Responsive CopolymerGrafted Beads by Surface-Initiated Atom Transfer Radical Polymerization (SI-ATRP). Three types of copolymer-grafted beads with different monomer compositions were prepared by SI-ATRP using CMPS beads (Table 1).13,25 The typical procedure for the preparation of P(IPAAm-co-APTAC-co-tBAAm) terpolymer-grafted beads was as follows: IPAAm (3.16 g, 27.9 mmol), APTAC (889 mg,

Table 1. Polymerization Conditions and the Amount of Copolymers Grafted onto Bead Surfaces feed molar ratio bead codea CMPS PIPAAm PIA PIAT

feed total monomer concentration (mmol/L) 500 500 500

IPAAm 100 90 65

APTAC 0 10 10

tBAAm 0 0 25

nitrogen content (wt %) 0.47 7.61 8.97 9.97

± ± ± ±

0.04 0.22 0.02 0.07

× × × ×

grafted amount of copolymer (μg/cm2)

−3

10 10−3 10−3 10−3

0.56 ± 0.02 0.65 ± 0.01 0.76 ± 0.01

a

CMPS indicates unmodified chloromethylated poly(styrene) (CMPS) beads; PIPAAm, PIA, and PIAT indicate PIPAAm homopolymer-, poly(IPAAm-co-APTAC) bipolymer-, and poly(IPAAm-co-APTAC-co-tBAAm) terpolymer-grafted CMPS beads, respectively. 1766

dx.doi.org/10.1021/bm300256e | Biomacromolecules 2012, 13, 1765−1773

Biomacromolecules

Article

where H (μm) and R (μm) represent the immersion depth of the beads into the liquid phase and the radius of the beads, respectively. The data are expressed as the mean ± standard deviation of 20 beads. 2.4. Cell Adhesion, Proliferation, and Viability on the Copolymer-Grafted Bead Surfaces. Chinese hamster ovary (CHO-K1) cells were obtained from Health Science Research Resources Bank (HSRRB, Tokyo, Japan) and grown in Ham F12 medium (Gibco BRL, Grand Island, NY) containing 10% fetal bovine serum (FBS; Japan Bioserum, Hiroshima, Japan), 100 units/mL penicillin, and 100 μg/mL streptomycin (Gibco BRL) in a humidified 5% CO2 atmosphere at 37 °C. The copolymer-grafted beads were sterilized by immersion in 70% ethanol, followed by UV exposure before the experiments. CHO-K1 cells were seeded at a density of 2.5 × 104 cells/cm2 and added to each well of a non-cell adhesive 12-well plate (HydroCell; CellSeed, Tokyo, Japan). For each well, 2 mg of copolymer-grafted beads (surface area: 2 cm2) was added, and the final volume of the medium was adjusted to 1 mL/well. After 24 h of incubation, 900 μL of medium was carefully exchanged for fresh medium. This procedure was repeated three times to remove nonadhered cells. The cells were further incubated for 168 h, and the medium was exchanged once every 2 days. Both the beads and the cells were observed under a phase contrast microscope (Eclipse TE2000-U; Nikon) for a prescribed time period. To determine the number of cells adhering to the bead surfaces, the beads were washed twice with phosphate buffered saline (PBS; SigmaAldrich, St. Louis, MO), and treated with 0.25% trypsin containing 0.26 mM EDTA (Sigma-Aldrich) for 15 min to harvest the cells. The cell number was counted using a hemocytometer. The number of adhered cells was then determined as follows: [number of cells adhered to bead surface (cells/cm2)] = [number of cells in the medium (cells)]/[amount of beads in each well (g)]/[specific surface area of bead (cm2/g)]. The morphology of the cells adhered to the bead surfaces was observed using field-emission scanning electron microscopy (SEM). The cells adhered to the bead surfaces were carefully washed with PBS three times and fixed in 4% paraformaldehyde for 15 min at 37 °C. The cells were then dehydrated by immersion in a graded ethanol series (50%, 70%, 80%, 90%, and 100%) for 10 min at each concentration and then twice in tert-butyl alcohol. The cells were then freeze-dried in tert-butyl alcohol. Finally, each sample was treated with osmium tetroxide in a Neoc-ST osmium coater (Meiwafosis, Osaka, Japan) for 15 s. SEM observations were performed using an S-4300 (Hitachi, Tokyo, Japan) instrument at an accelerating voltage of 2 kV. To evaluate the viability of the cells adhered to the surface of each bead, the cells were stained with calcein-AM and PI to visualize viable and dead cells, respectively. The viable and dead cells were observed using a fluorescent microscope (Eclipse TE2000-U; Nikon) with the appropriate filter sets. 2.5. Thermally Induced Cell Detachment from the Copolymer-Grafted Bead Surfaces. Two milligrams of copolymer-grafted CMPS beads (surface area: 2 cm2) was added to each well of a HydroCell 12-well plate. In each well, CHO-K1 cells were seeded at a density of 1 × 105 cells/cm2, and the final volume of the medium was adjusted to 1 mL/well. After incubation for a prescribed time period, 900 μL of medium was carefully exchanged for fresh medium. This procedure was repeated three times to remove nonadherent cells. To harvest the cells from the bead surfaces, the cells were incubated at 20 °C for 120 min and then pipetted several times. The detached cell number was counted using a hemocytometer. Likewise, the cells were completely detached by treatment with 0.25% trypsin containing 0.26 mM EDTA at 37 °C for 15 min to determine the number of cells adhered to the bead surfaces. Then, the thermally induced cell detachment efficiency was determined as follows: [cell detachment efficiency (%)] = [number of detached cells by low temperature treatment (cells)]/[number of cells detached by trypsin-EDTA treatment (number of cells adhered to bead surfaces) (cells)] × 100. 2.6. Cell Reattachment Assay. Two milligrams of copolymergrafted beads (surface area: 2 cm2) was added to each well of a HydroCell 12-well plate. For each well, CHO-K1 cells were seeded at a density of 1 × 105 cells/cm2, and the final volume of the medium was

4.3 mmol, 10 mol % with respect to total monomer), and tBAAm (1.37 g, 10.8 mmol, 25 mol % with respect to total monomer) were loaded into a round-bottom flask equipped with a three-way stopcock and dissolved in 43 mL of dehydrated 2-propanol (total monomer concentration: 500 mmol/L), followed by bubbling with nitrogen for 30 min to deoxygenate the reaction mixture. CuCl (84.9 mg, 858 μmol), CuCl2 (11.5 mg, 86 μmol), and Me6TREN (217 mg, 948 μmol) were successively added to the reaction mixture, and the solution was stirred for 15 min to allow a CuCl/CuCl2/Me6TREN catalytic complex to form. Then, the monomer solution and CMPS beads (1.0 g; 2.2 mmol of chlorine) held in a 50 mL glass vessel were placed into a glovebag, the atmosphere of which was then replaced with dry nitrogen by running three cycles of repeated vacuum and nitrogen purging. The monomer solution was then poured into a glass vessel containing CMPS beads, and the reaction mixture was agitated on a shaker (SN-M40S; Nissin, Tokyo, Japan) for 16 h at room temperature. After polymerization, the copolymer-grafted CMPS beads were collected by centrifugation and washed by ultrasonication in methanol. This purification process was repeated three times to remove the catalysts and unreacted monomers. Further purification of the copolymer-grafted beads was carried out by agitating in 50 mmol/ L EDTA for 1 day to remove the catalysts, followed by agitating in 1 mol/L NaCl for 1 day to remove electrostatically adsorbed EDTA. The copolymer-grafted beads were then extensively washed with MilliQ water and collected by filtration. Finally, the obtained copolymergrafted beads were washed with acetone and dried in vacuo at 50 °C for 5 h. 2.3. Surface Characterization of Copolymer-Grafted Beads. The surface of copolymer-grafted beads was analyzed using X-ray photoelectron spectroscopy (XPS) on a K-alpha (ThermoFisher Scientific, East Grinstead, UK) instrument equipped with a monochromatic Al Kα X-ray source (1486.6 eV). The copolymergrafted beads were mounted on carbon tape, and high resolution analyses of the C1s and N1s spectra were performed at a takeoff angle of 90°. The obtained spectra were referenced to an unfunctionalized aliphatic carbon at 285.0 eV, and the obtained spectra were deconvoluted using Origin v. 7.0 (Microcal Software, Northampton, MA). The amount of copolymer grafted onto the CMPS bead surfaces was calculated based on the increment in nitrogen content of the beads derived from the amide bond of the copolymer-grafted beads. The nitrogen content of the bead was measured using a total nitrogen analyzer TN-110 (Mitsubishi Chemical Analytech, Kanagawa, Japan), and the amount of copolymer grafted onto the bead surfaces was calculated as follows: amount of grafted copolymer(μg/cm 2) Np% − N0% = × 106 (Np,theor% − N0%)S where Np% and N0% represent the nitrogen content of the copolymergrafted and unmodified beads, respectively, Np,theor% represents the theoretical nitrogen content of the copolymer (feed monomer composition), and S represents the specific surface area of the CMPS bead (1030 cm2/g). The wettability of the copolymer-grafted bead surfaces was determined using a parallel-plate method at various temperatures.26 A 10 μL of water droplet was formed on a glass coverslip; then, a small number of copolymer-grafted CMPS beads was carefully mounted around the droplet. The droplet was spread by pressing with another coverslip, whereby the beads were collected at the air/water interface. The beads at the boundary were photographed using a phase contrast microscope (Eclipse TE300; Nikon, Tokyo, Japan) at 25.0 and 37.0 °C. The immersion depth of the beads into the liquid phase was measured from the obtained image using the software program Image J (National Institutes of Health, Bethesda, MD). The contact angle of the beads was determined as follows:

cos θ =

H−R R 1767

dx.doi.org/10.1021/bm300256e | Biomacromolecules 2012, 13, 1765−1773

Biomacromolecules

Article

Scheme 1. (A) Synthesis of PIPAAm-Based Cationic Copolymer-Grafted Beads by Surface-Initiated ATRP. (B) SEM Image and (C) Size Distribution of the CMPS Beads Used in This Study

Figure 2. X-ray photoelectron spectroscopy (XPS) C1s and N1s high resolution spectra of unmodified CMPS (i), PIPAAm homopolymer- (ii), PIA bipolymer- (iii), and PIAT terpolymer-grafted beads (iv). adjusted to 1 mL/well. After incubation for 24 h at 37 °C, 900 μL of medium was carefully exchanged for fresh medium. This procedure was repeated three times to remove nonadherent cells. To harvest the cells from the bead surfaces, the cells were incubated at 20 °C for 120 min and then pipetted several times. Separately, the cells were harvested by trypsin-EDTA treatment at 37 °C for 15 min. Both the thermally and the enzymatically detached cells were again seeded on a 35-mm tissue culture polystyrene (TCPS) dish (BD Falcon, Franklin Lakes, NJ) at a density of 1 × 104 cells/cm2. After incubation for a prescribed time period at 37 °C, the cells were imaged by a phase

contrast microscopy (Eclipse TE2000-U; Nikon), and the number of adhering cells was counted from the microscopic image. 2.7. Statistical Analysis. Statistical analysis was performed using a two-tail Student's t-test. A p-value of less than 0.05 was considered to indicate statistical significance. All values are expressed as the mean ± standard deviation of triplicate experiments.

3. RESULTS AND DISCUSSION 3.1. Synthesis and Characterization of TemperatureResponsive Cationic Copolymer-Grafted Beads. The 1768

dx.doi.org/10.1021/bm300256e | Biomacromolecules 2012, 13, 1765−1773

Biomacromolecules

Article

surface modification of chloromethylated poly(styrene) (CMPS) beads with a temperature-responsive copolymer brush was carried out through the surface-initiated atom transfer radical copolymerization (SI-ATRP) of N-isopropylacrylamide (IPAAm), positively charged 3-acrylamidopropyl triethylammonium chloride (APTAC), and hydrophobic Ntert-butylacrylamide (tBAAm) in 2-propanol (Scheme 1).13,25 In this study, three types of copolymer brushes consisting of PIPAAm homopolymer, P(IPAAm-co-APTAC) bipolymer (PIA), and P(IPAAm-co-APTAC-co-tBAAm) terpolymer (PIAT) were grafted onto the CMPS bead surfaces to evaluate the effect of positively charged quaternary amines on cell adhesion, proliferation, and thermally induced harvest in a suspension culture system (Table 1). The conditions used to prepare the copolymer-grafted bead are summarized in Table 1; the total feed monomer concentration was adjusted to an equal concentration for each bead (500 mmol/L). A previous study showed that approximately 0.5 μg/cm2 of grafted PIPAAm was sufficient to achieve both cell adhesion and thermally induced detachment for a CMPS bead with 56 μm in diameter; thus, this feed monomer concentration was also employed in this study.13 The detailed chemical composition of the obtained copolymer-grafted bead surfaces was analyzed by X-ray photoelectron spectroscopy (XPS) (Figure 2). In the C1s and N1s spectra, new peaks at 288 eV (C1s) and 399.8 eV (N1s), which were assignable to CO and N−H bonds derived from the amide linkage of the copolymer, were confirmed for all copolymer-grafted beads compared to the unmodified CMPS bead surface.27 This result indicated that the successful construction of temperature-responsive copolymer brush layer on the bead surfaces. Additionally, in the N1s spectra of the PIA and PIAT beads, signals at 402.5 eV, which were assignable to quaternary amines (N+) derived from APTAC moieties, were also observed, clearly indicating the incorporation of APTAC moieties into the grafted polymer. The molar fraction of APTAC in the copolymer brush, which was calculated by comparing the peak areas between the N+ peak (402.5 eV) and amide linkage peak (399.8 eV), was slightly higher than the feed monomer ratio (15.8 mol % for PIA and 14.0 mol % for PIAT); this result was presumably due to a difference in monomer reactivity.28 The amount of copolymer grafted onto the bead surfaces was determined by comparing the nitrogen content before and after surface modification (Table 1). The amount of copolymer grafted onto the bead surfaces was between 0.56 and 0.76 μg/cm2. The surface wettability of a substrate is of great concern when constructing a cell culture substrate surface because this property is known to be related to protein adsorption and cell adhesion.29 Thus, the water contact angle of the copolymergrafted CMPS beads was determined using a parallel-plate method at the temperature of both the cell culture (37 °C) and at that of near-cell-harvest conditions (25 °C; Figure 3). The contact angle of the PIPAAm-grafted beads was significantly smaller than that of unmodified CMPS beads at 37 °C, similar to our previous study.13 The contact angle of PIPAAm-grafted beads was found to be further reduced by reducing the temperature from 37 to 25 °C, whereas unmodified CMPS beads showed a negligible change in contact angle. This change was due to the temperature dependence of the hydration/ dehydration transition of the PIPAAm-grafted chains. The contact angle of the positively charged PIA beads was found to be slightly lower than that of the PIPAAm-grafted beads at 37

Figure 3. Contact angle of unmodified CMPS, PIPAAm homopolymer-, PIA bipolymer-, and PIAT terpolymer-grafted beads. The measurements were performed at 25 °C (gray bars) and 37 °C (black bars). The data are expressed as the mean ± SD of 20 particles (*p < 0.01, **p < 0.001, NS indicates not significant).

°C, presumably due to the highly hydrated nature of the quaternary amine moieties. However, the PIA beads showed a negligible change in contact angle with temperature. This result was most likely due to the drastic shift in LCST to high temperature by copolymerization with the highly hydrated quaternary amines.28 However, the LCST of PIPAAm-based copolymers containing ionic monomer units is known to be reduced by the copolymerization of an adequate amount of hydrophobic monomer.30,31 Indeed, the copolymerization of both APTAC and hydrophobic tBAAm in the copolymer brush resulted in an increase in the contact angle of the PIAT beads compared to that of the PIA beads at 37 °C. As a result, the contact angle of the PIAT beads exhibited a significant change in contact angle with a decrease in temperature from 37 to 25 °C. Consequently, the temperature-responsive PIPAAm and PIAT-grafted beads had the possibility of regulating cell adhesion and detachment by temperature alteration. 3.2. Cell Culture on the Surface of TemperatureResponsive Cationic Copolymer-Grafted Beads. To investigate cell adhesion and proliferation on the surfaces of the copolymer-grafted beads, CHO-K1 cells were selected as a model cell because they have been widely utilized to produce human recombinant proteins in microcarrier cultures.2 Herein, we performed these experiments in a static rather than stirred culture to avoid the effect of shear forces. Figures 4A−D and E−H show the phase contrast images of unmodified CMPS and copolymer-grafted beads incubated with CHO-K1 cells (2.5 × 104 cells/cm2) for 24 h and 72 h, respectively. The black dots observed in these images represent the beads. The unmodified CMPS and the PIPAAm-grafted beads were found to have aggregated with cells over the cell culture period (Figure 4A, B, E, and F). In the absence of cells, no aggregation of the unmodified CMPS and the PIPAAm-grafted beads was observed (Supporting Information, Figure S1), indicating that the observed aggregation was induced by the cells. In sharp contrast, the quaternary amine-introduced PIA- and PIATgrafted beads were dispersed as single beads even after 72 h of culture (Figure 4C, D, G, and H). This result clearly indicates that the incorporation of positively charged quaternary amine moieties facilitated the dispersibility of the beads even in the absence of shear forces. The aggregation of the beads reduced 1769

dx.doi.org/10.1021/bm300256e | Biomacromolecules 2012, 13, 1765−1773

Biomacromolecules

Article

Figure 4. Phase contrast microscopic images of unmodified CMPS (A, E), PIPAAm homopolymer- (B, F), PIA bipolymer- (C, G), and PIAT terpolymer-grafted beads (D, H) after incubation with CHO-K1 cells for 24 h (A−D) and 72 h (E−H) (scale bars: 200 μm). SEM images of CHOK1 cells adhered on the surface of unmodified CMPS (I, M), PIPAAm homopolymer- (J, N), PIA bipolymer- (K, O), and PIAT terpolymer-grafted beads (L, P) after incubation for 24 h (I−L) and 120 h (M−P) (scale bars: 20 μm). The cell seeding density was 2.5 × 104 cells/cm2 of bead surface.

albumin (BSA) adsorbed onto the surface of the PIA- and PIAT-grafted beads was significantly higher than that adsorbed onto the surface of PIPAAm-grafted beads (Supporting Information, Figure S2), indicating that the surface of the PIA- and PIAT-grafted beads was positively charged.25 Consequently, although the PIA- and PIAT-grafted beads had highly hydrated quaternary amines, the incorporation of APTAC into the PIPAAm-grafted layer scarcely affected the number of adhering cells on the bead surface. Additionally, the difference in the dispersibility of the beads did not significantly affect the number of initially attached cells. To evaluate the cell proliferation efficacy on the bead surface, the number of expanded cells on the bead surface was observed for 168 h (Figure 5B). The proliferation efficiency of CHO-K1 cells on the surface of quaternary amine-bearing PIA- and PIAT-grafted beads was significantly higher than that of CHOK1 cells on unmodified CMPS and PIPAAm-grafted beads and reached confluence within 168 h of culture. At that time, the density of CHO-K1 cells increased ca. 15-fold on the PIA- and PIAT-grafted beads, whereas the density of CHO-K1 cells increased ca. 9-fold on the unmodified and PIPAAm-grafted beads with respect to the initial cell seeding density (2.5 × 104 cells/cm2). These results suggest that cell proliferation on unmodified CMPS and PIPAAm-grafted bead surfaces might be restricted by the bead aggregation, resulting in the reduction of the surface area available for cell expansion. In sharp contrast, the high dispersibility of the quaternary amine-bearing PIA- and PIAT-grafted beads was speculated to enable cell expansion over the entire surface of beads, leading to the enhanced cell proliferation efficacy observed in the suspension culture. Because most cationic polymers such as poly(ethylenimine), poly(L-lysine), and poly(amidoamine) dendrimer are known to exhibit cytotoxicity,33,34 the viabilities of CHO-K1 cells adhered to the quaternary amine-bearing bead surfaces were investigated by LIVE/DEAD staining (Figure 6). The results show that almost all of the cells were detected as green signals derived

the surface area available for cell culture, possibly resulting in a reduction in cell expansion efficiency. Although the detailed mechanism govering this behavior has not yet been fully clarified, the adhesive forces between the cells and the positively charged amine-bearing bead surfaces may be sufficiently strong that cells preferentially remained adhered to these surfaces when they came into contact with cells on other beads. On the contrary, the cells adhered to unmodified CMPS and PIPAAmgrafted beads might easily undergo bead-to-bead transfer, leading to aggregation between the cells and the beads.32 The cells adhered to the bead surfaces were observed using SEM to examine their morphology in detail (Figure 4I−P). After 24 h of culture, the adhesion and spreading of cells were clearly observed for all bead surfaces, and a negligible difference in cell morphology was confirmed among individual beads. After further incubation for 120 h, the number of spread cells clearly increased and consequently covered the entire surface of the copolymer-grafted beads; this result indicated that the cells that were adhered to the bead surfaces were able to expand during the culture period, regardless of the dispersibility of the beads themselves. The number of cells adhered to the bead surfaces after 24 h of incubation is shown in Figure 5A. In our previous study, the incorporation of charged monomer (e.g., acrylic acid) into the PIPAAm-grafted planar surface resulted in a reduction in the number of initially attached cells due to the highly hydrophilic nature of the surface.16 However, although the PIA- and PIATgrafted beads had highly hydrated quaternary amines, a negligible difference in the number of initially attached cells was observed, regardless of the cell seeding density, compared to those of the unmodified CMPS and PIPAAm-grafted beads after 24 h of culture. This result was most likely due to the positively charged nature of APTAC, which enhanced the electrostatic interaction between the bead surface and serum proteins, including those found in the ECM and/or plasma membrane. Indeed, the amount of anionic bovine serum 1770

dx.doi.org/10.1021/bm300256e | Biomacromolecules 2012, 13, 1765−1773

Biomacromolecules

Article

from their highly hydrated nature.22,35 Thus, the low cytotoxicity of PIA- and PIAT-grafted beads was most likely due to the low membrane disruption of quaternary amines. 3.3. Thermally Induced Cell Detachment from the Surface of Temperature-Responsive Cationic Copolymer-Grafted Beads. The effect of highly hydrated and positively charged APTAC groups in temperature-responsive copolymer brush surfaces on thermally induced cell detachment behavior was investigated. Figure 7A−D shows the phase

Figure 7. Phase contrast microscopic images of CHO-K1 cells cultured on the surface of unmodified CMPS (A, E), PIPAAm homopolymer- (B, F), PIA bipolymer- (C, G), and PIAT terpolymergrafted beads (D, H) after 24 h of incubation at 37 °C (A−D), and following incubation for 120 min at 20 °C (E−H) (scale bar: 100 μm). The cell seeding density was 1 × 105 cells/cm2 of bead surface.

contrast images of beads after 24 h of incubation at 37 °C with densely seeded CHO-K1 cells (1 × 105 cells/cm2). In these images, the white objects at the periphery of the beads represent adhered CHO-K1 cells; the cells were found to be adhered to the entire surface of the beads. Figure 7E−H shows the phase contrast images of beads and cells after a subsequent 120 min of incubation at 20 °C. In these images, white spherical objects corresponding to detached CHO-K1 cells were observed for PIPAAm- and PIAT-grafted bead surfaces (Figure 7F,H), whereas the cells adhered on the surface of the unmodified CMPS and PIA-grafted beads were retained (Figure 7E,G). These results clearly indicate that the temperature-dependent hydrophilic/hydrophobic transition and the hydration/dehydration transition of the grafted copolymer chains played a pivotal role in regulating cell detachment from the bead surfaces.10−16 Figure 8A shows the time course of the detached cell percentage at 20 °C. The detached cell ratio increased gradually on both the PIPAAmand PIAT-grafted beads with increasing incubation time at 20 °C, and almost all cells were detached from the bead surfaces after 120 min of low-temperature treatment. Notably, the temperature-responsive PIAT-grafted beads exhibited more rapid cell detachment than the PIPAAm-grafted beads, despite the abundance of positively charged amines. This difference was presumably because the highly hydrated nature of APTAC promotes the overall hydration of the copolymer chains relative to that of the PIPAAm homopolymer chains, leading to efficient cell detachment within a short incubation time at 20 °C.16 In comparison, the cell detachment ratio of the PIAgrafted bead surfaces was only 31.2%, even after 120 min of low-temperature treatment. Because the PIA-grafted bead surface exhibited a negligible change in contact angle at the temperature ranging from 25 to 37 °C (Figure 3), the grafted copolymers was hardly hydrated by temperature reduction. Thus, the temperature-dependent change in the properties of the PIA-grafted bead surface is suggested to be insufficient for the detachment of cells. Note that the viability of detached cells

Figure 5. (A) Number of CHO-K1 cells adhered to the surface of unmodified CMPS, PIPAAm homopolymer-, PIA bipolymer-, and PIAT terpolymer-grafted beads after 24 h of incubation. The gray and black bars represent initial cell seeding densities of 2.5 × 104 and 1 × 105 cells/cm2, respectively. (B) Proliferation curves of CHO-K1 cells cultured on the surface of unmodified CMPS (open squares), PIPAAm homopolymer- (closed circles), PIA bipolymer- (open triangles), and PIAT terpolymer-grafted beads (closed diamonds). The initial cell seeding density was 2.5 × 104 cells/cm2. The data are expressed as the mean ± SD (n = 3) (*p < 0.05, NS indicates not significant).

Figure 6. LIVE/DEAD fluorescent staining microscopic images of CHO-K1 cells cultured on the surface of unmodified CMPS (A, E), PIPAAm homopolymer- (B, F), PIA bipolymer- (C, G), and PIAT terpolymer-grafted beads (D, H) after incubation for 24 h (A−D) and 72 h (E−H). The live and the dead cells were stained with calcein-AM (green) and PI (red), respectively (scale bars: 100 μm). The cell seeding density was 2.5 × 104 cells/cm2 of bead surface.

from viable cells, and negligible red signals derived from dead cells were observed after 24 and 72 h of culture. Therefore, the quaternary amine-bearing PIA- and PIAT-grafted beads were confirmed to have negligible cytotoxicity. According to numerous studies on cationic polymers utilized in the field of gene delivery, quaternary amino groups tend to show less toxicity than primary, secondary, and tertiary amino groups, mainly due to their low membrane disruption activity arising 1771

dx.doi.org/10.1021/bm300256e | Biomacromolecules 2012, 13, 1765−1773

Biomacromolecules

Article

this structure is disrupted in enzymatically detached cells due to the proteolytic enzyme reaction. Additionally, we reported that temperature-induced detachment is less invasive for deposited ECMs, and cells can be detached by retaining ECM on their surface.12 Therefore, thermally detached cells exhibited a higher efficiency of reattachment to another surface than that of enzymatically detached cells.36 This result is advantageous for large-scale cell culture because repeated passaging is required to scale up culture volume. Accordingly, the reattachment efficiency of detached cells is important in improving the efficiency of large-scale microcarrier culture.2,4,5 In this regard, the reattachment efficiency of cells detached from PIAT-grafted bead surfaces by low-temperature and trypsin-EDTA treatment was investigated (Figure 9). The thermally detached cells

Figure 8. (A) Time course of the detached cell ratio from unmodified CMPS (open squares), PIPAAm homopolymer- (closed circles), PIA bipolymer- (closed triangles), and PIAT terpolymer-grafted bead surfaces (closed diamonds) on incubation at 20 °C (initial cell density: 1 × 105 cells/cm2). (B) Time course of the detached cell ratio from PIPAAm homopolymer- (closed circles) and PIAT terpolymer-grafted bead surfaces (closed diamonds) after long-term culture at 37 °C (initial cell density: 2.5 × 104 cells/cm2). The cells were detached by incubation for 120 min at 20 °C. The data are expressed as the mean ± SD (n = 3) (*p < 0.05, **p < 0.01).

Figure 9. Time course of the number of reattached cells on a TCPS surface after harvest from PIAT bead surfaces by trypsin-EDTA treatment for 15 min (open diamonds) and low-temperature treatment for 120 min (closed diamonds). The data are expressed as the mean ± SD (n = 3) (*p < 0.05, **p < 0.01).

exhibited a significantly higher initial attachment efficiency on a planar TCPS surface compared with enzymatically detached cells. This result was most likely due to the effect of ECMs remaining on the surface of detached cells, improving the interaction between the cell and the substrate surface.12 Thus, thermally induced cell detachment from microcarrier surfaces is beneficial for improving initial cell reattachment efficiency.

from both the PIPAAm- and the PIAT-grafted bead surfaces after 120 min of incubation at 20 °C was greater than 90%, as determined by a trypan blue exclusion assay. This result is consistent with the results of our previous study.13 Thus, the incorporation of positively charged APTAC into temperatureresponsive copolymer brushes did not affect the viability of thermally detached cells. When cells are expanded across the entire surface of the beads, temperature-dependent cell harvest might be suppressed by the formation of strong cell−cell junctions. Hence, the effect of cell culture time on thermally induced cell harvest from bead surfaces was investigated (Figure 8B). The harvested cell ratio decreased gradually with increasing cell culture time, presumably due to the formation of cell−cell junctions among proliferated cells. However, the PIAT-grafted beads exhibited a persistent and high cell detachment efficiency compared to the PIPAAm-grafted beads during 264 h of culture. Thus, the incorporation of highly hydrated APTAC was beneficial in both accelerating cell detachment and improving cell harvest efficiency after long-term culture. In our previous study, we confirmed that the surface morphology of thermally and enzymatically detached cells from PIPAAm-grafted beads by SEM observation.13 The thermally detached cells retain the complex microvilli, whereas

4. CONCLUSION Temperature-responsive microcarriers bearing PIPAAm-based cationic copolymer brush comprising P(IPAAm-co-APTAC) bipolymer and P(IPAAm-co-APTAC-co-tBAAm) terpolymer were developed to facilitate a cell proliferation and thermally induced harvest efficiency in suspension culture. When CHOK1 cells were seeded on the surface of copolymer-grafted beads, the aggregation of the beads with cells was observed for unmodified CMPS and PIPAAm-grafted beads. In sharp contrast, the quaternary amine-bearing copolymer brushes facilitated the dispersibility of the beads and eventually enhanced cell proliferation compared to that of PIPAAmgrafted beads. Additionally, thermally induced cell detachment was observed for PIPAAm homopolymer- and P(IPAAm-coAPTAC-co-tBAAm) terpolymer-grafted beads with a reduction in temperature, whereas unmodified CMPS and P(IPAAm-coAPTAC) bipolymer-grafted beads showed negligible cell detachment. Notably, the P(IPAAm-co-APTAC-co-tBAAm) 1772

dx.doi.org/10.1021/bm300256e | Biomacromolecules 2012, 13, 1765−1773

Biomacromolecules

Article

(19) Gigout, A.; Levasseur, S.; Girard-Lauriault, P.-L.; Buschmann, M. D.; Wertheimer, M. R.; Jolicoeur, M. Macromol. Biosci. 2009, 9, 979−988. (20) Cer, E.; Gürpınar, Ö . A.; Onur, M. A.; Tuncel, A. J. Biomed. Mater. Res., Part B 2007, 80B, 406−414. (21) Kato, D.; Takeuchi, M.; Sakurai, T.; Furukawa, S.; Mizokami, H.; Sakata, M.; Hirayama, C.; Kunitake, M. Biomaterials 2003, 24, 4253−4264. (22) Tamura, A.; Oishi, M.; Nagasaki, Y. J. Controlled Release 2010, 146, 378−387. (23) Arigita, C.; Zuidam, N. J.; Crommelin, D. J. A.; Hennink, W. E. Pharm. Res. 1999, 16, 1534−1541. (24) Queffelec, J.; Gaynor, S. G.; Matyjaszewski, K. Macromolecules 2000, 33, 8629−8639. (25) Nagase, K.; Kobayashi, J.; Kikuchi, A.; Akiyama, Y.; Kanazawa, H.; Okano, T. Biomaterials 2011, 32, 619−627. (26) Hórvölgyi, Z.; Németh, S.; Fendler, J. H. Langmuir 1996, 12, 997−1004. (27) Pan, Y. V.; Wesley, R. A.; Luginbuhl, R.; Denton, D. D.; Ratner, B. D. Biomacromolecules 2001, 2, 32−36. (28) Utsel, S.; Malmström, E. E.; Carlmark, A.; Wågberg, L. Soft Matter 2010, 6, 342−352. (29) van Wachem, P. B.; Beugeling, T.; Feijen, J.; Bantjes, A.; Detmers, J. P.; van Aken, W. G. Biomaterials 1985, 6, 403−408. (30) Shao, D.; Ni, C. J. Appl. Polym. Sci. 2007, 105, 2299−2305. (31) Yokoyama, M.; Okano, T. J. Biomater. Sci., Polym. Ed. 2001, 12, 769−782. (32) Ohlson, S.; Branscomb, J.; Nilsson, K. Cytotechnology 1994, 14, 67−80. (33) Fischer, D.; Li, Y.; Ahlemeyer, B.; Krieglstein, J.; Kissel, T. Biomaterials 2003, 24, 1121−1131. (34) Moreau, E.; Domurado, M.; Chapon, P.; Vert, M.; Domurado, D. J. Drug Targeting 2002, 10, 161−173. (35) Wang, J.; Gao, S. J.; Zhang, P. C.; Wang, S.; Mao, H. Q.; Leong, K. W. Gene Ther. 2004, 11, 1001−1010. (36) Okano, T.; Yamada, N.; Sakai, H.; Sakurai, Y. J. Biomed. Mater. Res. 1993, 27, 1243−51.

terpolymer-grafted beads exhibited a rapid and persistent cell detachment compared to the PIPAAm-grafted beads. Overall, these fundamental evaluations indicate that P(IPAAm-coAPTAC-co-tBAAm) terpolymer-grafted beads are a promising molecular design for achieving an efficient, scalable suspension culture that allows for the noninvasive harvest of anchoragedependent cells by simple temperature reduction.

■

ASSOCIATED CONTENT

S Supporting Information *

Phase contrast microscopic images of copolymer-grafted beads without cells and adsorption of FITC-BSA on copolymergrafted bead surfaces. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +81-3-5367-6645 (ext. 30233). Fax: +81-3-3359-6046. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Global COE Program, the Multidisciplinary Education and Research Center for Regenerative Medicine (MERCREM) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan, and the Formation of Innovation Center for Fusion of Advanced Technologies in the Special Coordination Funds for Promoting Science and Technology from the MEXT of Japan. We are grateful to Dr. Norio Ueno for English editing.

■

REFERENCES

(1) Wurm, F. M. Nat. Biotechnol. 2004, 22, 1393−1398. (2) Chu, L.; Robinson, D. K. Curr. Opin. Biotechnol. 2001, 12, 180− 187. (3) van Wezel, A. Nature 1967, 216, 64−65. (4) Griffiths, B. In Animal cell culture, 3rd ed.; Masters, J. R. W., Eds.; Oxford University Press: New York, 2000, pp 19−67. (5) Malda, J.; Frondoza, C. G. Trends Biotechnol. 2006, 24, 299−304. (6) Martin, Y.; Eldardiri, M.; Lawrence-Watt, D. J.; Sharpe, J. R. Tissue Eng., Part B 2011, 17, 71−80. (7) Yang, Y.; Rossi, F. M. V.; Putnins, E. E. Biomaterials 2007, 28, 3110−3120. (8) Schop, D.; Janssen, F. W.; Borgart, E.; de Bruijn, J. D.; van Dijkhuizen-Radersma, R. J. Tissue Eng. Regen. Med. 2008, 2, 126−135. (9) Mason, C.; Dunnill, P. Regen. Med. 2009, 4, 153−157. (10) Yamada, N.; Okano, T.; Sakai, H.; Karikusa, F.; Sawasaki, Y.; Sakurai, Y. Makromol. Chem., Rapid Commun. 1990, 11, 571−576. (11) Kushida, A.; Yamato, M.; Konno, C.; Kikuchi, A.; Sakurai, Y.; Okano, T. J. Biomed. Mater. Res., Part B 2000, 51, 216−223. (12) Yamato, M.; Utsumi, M.; Kushida, A.; Konno, C.; Kikuchi, A.; Okano, T. Tissue Eng. 2001, 7, 473−480. (13) Tamura, A.; Kobayashi, J.; Yamato, M.; Okano, T. Biomaterials 2012, 33, 3803−3812. (14) Yamato, M.; Akiyama, Y.; Kobayashi, J.; Yang, J.; Kikuchi, A.; Okano, T. Prog. Polym. Sci. 2007, 32, 1123−1133. (15) Cooperstein, M. A.; Canavan, H. E. Langmuir 2010, 26, 7695− 7707. (16) Ebara, M.; Yamato, M.; Hirose, M.; Aoyagi, T.; Kikuchi, A.; Sakai, K.; Okano, T. Biomacromolecules 2003, 4, 344−349. (17) Lee, J. H.; Lee, J. W.; Khangt, G.; Lee, H. B. Biomaterials 1997, 18, 351−358. (18) Lee, J. H.; Jung, H. W.; Kang, I. K.; Lee, H. B. Biomaterials 1994, 15, 705−711. 1773

dx.doi.org/10.1021/bm300256e | Biomacromolecules 2012, 13, 1765−1773