- Email: [email protected]
Practical cell labeling with magnetite cationic liposomes for cell manipulation
Journal of Bioscience and Bioengineering VOL. 110 No. 1, 124 – 129, 2010 www.elsevier.com/locate/jbiosc
Practical cell labeling with magnetite cationic liposomes for cell manipulation Hiroshi Ito,1 Yurika Nonogaki,1 Ryuji Kato,1 and Hiroyuki Honda1,2,⁎ Department of Biotechnology, Graduate School of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan 1 and MEXT Innovative Research Center for Preventive Medical Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan 2 Received 27 November 2009; accepted 7 January 2010 Available online 4 February 2010
Personalization of the cell culture process for cell therapy is an ideal strategy to obtain maximum treatment effects. In a previous report, we proposed a strategy using a magnetic manipulation device that combined a palm-top size device and a cell-labeling method using magnetite cationic liposomes (MCLs) to enable feasible personalized cell processing. In the present study, we focused on optimizing the MCL-labeling technique with respect to cell manipulation in small devices. From detailed analysis with different cell types, 4 pg/cell of MCL-label was found to be obtained immediately after mixing with MCLs, which was sufficient for magnetic cell manipulation. The amount of label increased within 24 h depending on cell type, although in all cases it decreased along with cell doubling, indicating that the labeling potential of MCLs was limited. The role of free MCLs not involved in labeling was also investigated; MCLs' role was found to be a supportive one that maximized the manipulation performance up to 100%. We also determined optimum conditions to manipulate adherent cells by MCL labeling using the MCL dispersed in trypsin solution. Considering labeling feasibility and practical performance with 103–105 cells for personalized cell processing, we determined that 10 μg/ml of label without incubation time (0 h incubation) was the universal MCL-labeling condition. We propose the optimum specifications for a device to be combined with this method. © 2010, The Society for Biotechnology, Japan. All rights reserved. [Key words: Cell manipulation; Magnetite cationic liposome (MCL); Magnetic cell labeling; Small-scale cell handling; Magnetic force]
Cell-based therapy, as a part of regenerative medicine, is an innovative therapy to treat refractory defects of patients using their own cells (1–4). Although cell-based therapy is a promising approach (5, 6), its biggest drawback is the variability of clinically obtainable cells. Since cells from individual patients have significant individual differences, it is difficult to prepare optimal culture conditions for every patient's cells. To gain maximum effects with cell therapy, cell processing technology for personalized cell processing is required to culture patient's cells under most suitable conditions. For ideal personalized cell processing, it is desirable to obtain the optimal culture conditions for each patient's cells prior to mass production of cells for treatment. However, in order to obtain optimal cell culture conditions, many parameters, such as cytokine concentrations, medium components, and their inoculation density, should be examined (7–11), in spite of the limited number of cells obtained clinically from patients. Particularly in stem cell processing (12–14), the optimization of culture conditions is very difficult due to the limitation of acquiring adequate cell numbers relative to the large variations of culture parameters. Therefore, an effective technology to efficiently handle the limited number of cells and examine various culture conditions is required for personalized cell processing. Currently, the multi-well plate format along with automated cell assays is most frequently used to examine various culture conditions. ⁎ Corresponding author. Department of Biotechnology, Graduate School of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan. Tel.: + 81 52 789 3215; fax: +81 52 789 3214. E-mail address: [email protected] (H. Honda).
However, the multi-well plate format is accompanied by significant manipulation errors during the cell–liquid separation by centrifugation, especially with a small-volume sample (e.g., several hundred μl). To overcome this difficulty, we previously proposed a palm-top size device system, magnetic manipulation device (MMD), which is capable of multi-condition and micro-scale cell-based assays with a minimum loss of target cells in a closed environment (15). In the previous study, we described effective cell manipulation performance using magnetite cationic liposomes (MCLs) in combination with the multi-well plate format within a closed device that included a small magnet for cell manipulation. MCLs consist of 10 nm average-sized magnetite nanoparticles (Fe3O4) encapsulated in liposomes dioleoyl phosphatidylethanolamine (DOPE), dilauroyl phosphatidylcholine (DLPC), and N-(α-trimethylammonioacetyl)didodecyl-D-glutamate chloride (TMAG) (16). Since the primary binding force of an MCL label is the electrostatic force of cationic lipids and not dependent on antibodies, MCLs are able to label any cell type without prior knowledge of surface markers. Our previous study of MMD also indicated that the practical number of cells (104–106 cells) expected to be examined for one sample in cell-based assays could be labeled almost immediately and could be handled with high collection efficiency (N90% collection). This cell manipulation efficacy was far better than the commercial antibody-conjugated MACS (Magnetic Cell Sorting) beads system with a simple cylindrical magnet, and was considered to be more advantageous for use within small enclosed devices to enable personalized cell processing. Although MCLs have been confirmed as safe for animal adherent cells by our various researches (17–25), such as human umbilical vein
1389-1723/$ - see front matter © 2010, The Society for Biotechnology, Japan. All rights reserved. doi:10.1016/j.jbiosc.2010.01.014
VOL. 110, 2010
MCL LABELING CONDITIONS FOR CELL MANIPULATION
endothelial cells and normal human dermal fibroblast (21), and human mesenchymal stem cells, which could differentiate even if labeled with MCLs (19), detailed MCL-based cell-labeling mechanism and procedure with respect to cell manipulation require further investigation before practical use of MCLs for cell processing. Moreover, the application of MCL labeling to suspension type cells and tumor cells has not been examined. In this report, we examined the MCL-labeling performance for various cell types (normal and tumor adherent cells, normal and cancer suspension cells, and stem cells) and determined the manipulation capabilities that were not examined in previous studies. From detailed analysis of MCL-labeling conditions (capacity, time, and cell recovery rate), we determined universal MCL-labeling conditions for any cell type. We propose guidelines for using MCLs in palm-top size devices for personalized cell processing. MATERIALS AND METHODS Cells and culture condition Jurkat (human T-lymphoma cell line), RPMI-1788 (human B-lymphocyte cell line), and HT1080 (human sarcoma cell line) were obtained from the American Type Culture Collection (Manassas, VA, USA). Human mesenchymal stem cells (MSCs) were obtained from Lonza (Basel, Switzerland). Primary normal human fibroblasts (Fb) were collected from healthy volunteers with permission from the ethical committee of Nagoya University. Suspension type cells were cultured in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen), 100 U/ml penicillin (Invitrogen), 100 μg/ml streptomycin (Invitrogen), HEPES (2-[4-(2-hydroxyethyl)-1-piperazinyl] ethanesulfonic acid, Dojindo Laboratories, Kumamoto, Japan), and sodium bicarbonate (Wako Pure Chemical Industries, Osaka, Japan). HT1080 and Fb were cultured in Dulbecco's Modified Eagle Medium (Invitrogen) with 10% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin. MSCs were cultured in MSCGM (Mesenchymal Stem. Cell Growth Medium, Lonza). All cells were incubated at 37 °C in 5% CO2 and 95% air conditions. Preparation of MCLs Magnetite nanoparticles (Fe3O4; average particle size 10 nm; Toda Kogyo, Hiroshima, Japan) and 3 types of lipids (1:2:2 molar ratio), cationic TMAG (Sogo Pharmaceutical, Tokyo, Japan), DLPC (Sigma-Aldrich, St. Louis, MO, USA), and DOPE (Avanti Polar Lipids, Alabaster, AL, USA), were used (16). MCLs were mixed with a lipid and dried by evaporation for more than 30 min to form a lipid film. The film was hydrated into colloidal magnetite nanoparticles and vortexed. This resulted in selforganization of liposomes. All MCL concentrations were expressed as the net magnetite concentration by measuring the magnetite concentration using a potassium thiocyanate method (26). Determination of MCL-labeling performance with different cell types Jurkat and RPMI-1788 cells were seeded into 6-well cell culture plates (6 × 105 cells/well) with culture medium containing MCLs (net magnetite concentration, 100 pg/cell). Cell pellets were collected after 2 rounds of centrifugation and washed with phosphate buffered saline (PBS). Cell numbers were determined by trypan blue cell counting and their labeled magnetite amount was determined by the potassium thiocyanate method described previously. Fb, HT1080, and MSCs were first suspended in each culture medium containing MCLs (100 pg/cell) for labeling and then seeded into 6-well cell culture plates (2 × 104 cells/well). During cell incubation, MCLs remained in the medium. The cells were periodically collected by trypsin treatment and their numbers and labeled magnetite amount were determined using above described methods. Magnetite concentration for each cell type determined immediately after mixing with MCL was defined as the basal amount of “attached MCLs” on cells. Magnetite concentration obtained from the later incubation time was considered as the labeled amount of MCLs, including “attached MCLs” and “taken up MCLs.” Determination of MCL-labeling performance for cell manipulation Jurkat cells were used as a model to determine the relationship between MCL concentration 5 for labeling and the labeled amount. The cells were suspended (1 × 10 cells/ml, 2 ml) and labeled with varying MCL concentrations (10, 20, 40, 60, 80, and 100 pg/cell) without incubation time (0 h incubation), followed by measurement of the labeled amount. Labeled cells (1 × 104 cells, 100 μl) were re-suspended in siliconized 1.5-ml microcentrifuge tubes (Fisher Thermo Fisher Scientific, Waltham, MA, USA) for the following cell collection assay. A cylindrical magnet (shape-type, cylindrical; size, ϕ10.95 × 15 mm; surface magnetic flux density, 520 mT; magnetic attractive force, 4.0 kg; Niroku seisakusho, Kobe, Japan) was placed on the side of a tube for 5 min to collect MCL-labeled cells. Collected cells were re-suspended in fresh culture medium without the magnet and then transferred into a 96-well cell culture plate (Greiner bioone, Kremsmünster, Austria) for the determination of the collected cell number by the CellTiter-Blue Assay (Promega, Madison, WI, USA). For the positive control, cells (1 × 105 cells/ml) without MCL addition were collected as pellets by centrifugation and measured by the CellTiter-Blue Assay. In addition to Jurkat cells, as the sample with high labeled amount, HT1080 cells (1 × 104 cells; MCL concentration, 60 pg/cell) were also examined using the same protocol in order to determine the relationship between the labeled amount and manipulation performance.
125
Determination of cell-labeling performance using trypsin-MCL The celllabeling performance of MCL was examined for Fb, HT1080, and MSCs using the MCL dispersed in trypsin solution (designated as trypsin-MCL). After the 24 h culture, cells seeded into 6-well cell culture plates (2 × 104 cells/well) were treated for 3 min with trypsin-MCL (0.05% trypsin, 100 pg/cell MCL). After inactivating trypsin by the addition of culture media to cell suspensions, the cells were collected by centrifugation and their MCL-labeled amount was determined as described above. Cell manipulation enhancement by excess MCL addition Jurkat cells were used to determine the enhancing effect of excess MCL addition into the target well. To examine labeling conditions, cells (1 × 104 cells/tube, 100 μl) were labeled with varying MCL concentrations (10, 20, 40, 60, 80, and 100 pg/cell) without incubation time and separated into two groups for comparisons: [1] washed cells which were subjected to 2 rounds of centrifugation and washed with PBS as described above and [2] unwashed cells with excess MCLs. In previous methods, labeling was described for the MCL concentration in pg/cell. However, 100 pg/cells for 1 × 104 in 100 μl cells was determined to be the universal labeling condition (10 μg/ml, without incubation time). To examine collection performance, different cell numbers (1 × 103, 1 × 104, and 1 × 105 cells) were labeled using the universal MCL-labeling condition. To determine collection speed, 1 × 104 cells were labeled using the universal labeling condition. Determination of expected device specifications based on MCL-labeled cell manipulation performance The manipulation distance of MCL-labeled cells was determined by collecting labeled cells (4 × 104 cells labeled using universal labeling conditions) from different distances (11, 12, 13, 14, and 16 mm), as described previously (15).
RESULTS AND DISCUSSION MCL-labeling performance for different cell types For optimized personalized cell processing, several days of cell culture are essential. Therefore, profiles of MCL-labeled amounts and their changes throughout the culture period were examined with different cell types. The incorporation of MCL during cell labeling was approximately equivalent with different cell types without incubation time (0 h incubation), and was considered to be not critical for further investigation. However, the labeled amount changed for some cells during the 24 h incubation (Fig. 1A). The average immediate labeled amount of all cells was 4 pg/cell, which could be regarded as the universal labeled amount for most cells after a short incubation time. Since the main driving force of MCL labeling during mixing is the electrostatic force of cationic lipids, it was convincing to find no difference among cell types. These results also indicated that the labeled amount of tumor cells (HT1080) was equivalent to that of normal cells (Fb), whereas that of suspension cancer cells (Jurkat) was much lower than that of normal cells (RPMI-1788). Since RPMI-1788 cells proliferated forming large aggregation, MCLs might be involved with aggregations and the cells got contact with MCLs more easily than Jurkat cells. Moreover, although lipid membrane fluidity of tumor cells is generally greater than that of normal cells (27), no difference in MCL labeling was observed. After 3 days of cell culture with excess MCLs in the medium, the MCL-labeled amount of all cell types decreased in exactly the same manner as the idealized amount of magnetite remaining after cell doubling (Fig. 1B). Along with the results indicated in Fig. 1A, the above results indicated that MCL labeling was mainly due to cellular responses to interactions with liposomes, and once MCLs were incorporated in the cytosol, they were strongly retained within cells and only decreased due to cell doubling. Interestingly, excess MCLs in the medium did not contribute to additional cell labeling after 24 h. Moreover, the excess MCLs showed no toxicity on growth of all cell types with 3 days culture (data not shown). Especially, the effects on a suspension type normal cell line were first examined and shown to be non-toxic. It was important that such normal suspension type cells could be treated with MCLs without toxicity, since these suspension type cells essentially require an effective solid-liquid separation for medium exchange compared to the adhesion type cells. Particle size distribution analysis indicated that large particles rapidly appeared in the cell culture medium but not in PBS or ultrapure water (data not shown). Therefore, it was deduced that charged protein molecules in the medium triggered MCL aggregation that negatively
126
ITO ET AL.
J. BIOSCI. BIOENG.,
FIG. 1. MCL-labeling profiles of different cell types. (A) The labeled MCL amount with rapid labeling conditions (without incubation time: 0 h) and 24 h incubation. (B) The changes of labeled MCL of magnetite per cell. Solid circles, measured MCL-labeled amount; open circles, estimated amount of MCLs remaining after cell doubling. This estimate was determined by dividing the first day magnetite amount by growth rate for each day. (C) Labeled MCL with trypsin-MCL treatment for 3 min. Results are expressed as means ± SE (n = 3). All labeling conditions, including the trypsin-MCL treatment, were set to an MCL concentration of 100 pg/cell of magnetite, and cell number was set for adherent (2 × 104 cells) and suspension cells (6 × 105 cells). Fb, fibroblasts; HT1080, fibrosarcoma; MSC, mesenchymal stem cell; Jurkat, T-lymphoma cell line; RPMI-1788, B-lymphocyte cell line.
affected cell-MCL interactions at later incubation times. Taken together, it was clear that only freshly dispersed MCLs were effective in cell labeling within 24 h, and that labeled cells could be sufficiently manipulated based on the magnetite remaining in the cytosol which was calculated from their 24 h labeled amount and doubling time. Compared to suspension type cells, adhesion type cells were considerably more difficult to label for long times throughout their passage culture. Thus, there are few palm-top size devices that can magnetically manipulate adhesion type cells during their passage culture. Therefore, we examined if MCL labels could be retained in adherent cells by introducing MCLs with trypsin treatment for applications in manipulating them throughout their passage culture
in a palm-top size device. Cells could be successfully labeled at 6 pg/ cells by treatment with trypsin-MCL for 3 min (Fig. 1C). This MCLlabeled amount was equivalent to the universal amount of cell labeling in media without incubation time. Therefore, it was concluded that any cell type could be labeled rapidly by adding fresh MCL or by trypsin-MCL treatment at their passages (designated as a rapid labeling protocol in the following sections). Relationship between MCL labeling and manipulation performance We investigated the relationship between the MCL concentration for labeling and the actual MCL-labeled amount (Fig. 2A). An increase in MCL concentration in order to increase the amount of labeled MCL per cell was effective up to 60 pg/cell.
VOL. 110, 2010
MCL LABELING CONDITIONS FOR CELL MANIPULATION
127
considered that not charged proteins but the magnetic force formed condensation of MCLs in the solution that incorporated cells and assisted the magnetic cell collection. Therefore, the formerly determined rapid labeling condition was considered to be practically useful with in any type of medium or solution if there were MCLs remaining after cell labeling. Next, we examined the capacity of this excess MCL-labeling condition. To assay cells in a conventional palm-top size device system, solution volume in the device is usually limited to a volume (i.e., b400 μl into 96-well plates) that makes centrifugation difficult. With such solution volume, reasonable cell culture could be managed
FIG. 2. Relationship between labeled MCL and manipulation performance. (A) The relationship between MCL concentration for labeling and MCL-labeled amount (actual labeled amount). Jurkat cells were labeled with varying MCL concentrations. (B) The relationship between the labeled MCL and cell manipulation performance. Jurkat and HT1080 cells (1 × 104) were used after labeling with varying MCL concentrations under the rapid or 1 day labeling conditions. All cell collection processes were performed by placing a magnet (size, ϕ10.95 × 15 mm; surface magnetic flux density, 520 mT) on the side of a siliconized tube for 5 min. Solid circles, Jurkat; open circles, HT1080 which represented sample with high labeled amount. Results are expressed as mean ± SE (n = 3).
The cell manipulation performance was presumed to reach 100% with a labeled amount of 10 pg/cell (Fig. 2B). Although, in addition to Jurkat, HT1080 was used as the sample with high labeled amount, it was considered that not cell type but labeled amount influenced the manipulation performance dominantly. This suggested that Fb, HT1080, MSC, and RPMI-1788 could be adequately manipulated for 2 days (Fig. 1B). Contribution of excess MCLs for cell manipulation performance From these results, it appeared as if the rapid labeling protocol described in previous report (15) was insufficient (b60%) for cell manipulation. However, during practical MCL labeling in closed devices, excess MCLs were not washed away. Therefore, even if rapid MCL labeling is not adequate for 100% manipulation with labeled MCLs (4–6 pg/cells), if the excess free MCLs contribute as support molecules, the manipulation performance should be high enough for practical use. To investigate the supporting effect of excess MCLs for cell manipulation performance, we compared the cell manipulation performance of cells labeled with the same MCL concentration followed by a process of washing or no washing (Fig. 3A). The manipulation performance reached a plateau at a 60% collection rate when remaining MCLs were washed away. This result was consistent with that the labeled amount of Jurkat cells reached a plateau at about 6 pg/cell (Fig. 2). However, it was observed that, even with the same MCL concentration for labeling (6 μg/ml), the cell manipulation rate reached 100% when excess MCLs were not eliminated. By microscopic observation, pellet-like condensation including all free MCLs was observed. Consequently, since it was found that MCLs do not accumulate on cells after 24 h (Fig. 1B), we
FIG. 3. Cell manipulation performance with excess MCLs. (A) The cell manipulation performance with different cell-labeling concentrations (collection time, 5 min). Jurkat cells (1 × 104 cells) were labeled with varying MCL concentrations under the rapid labeling condition. Solid circles, unwashed MCL-labeled cells in the presence of excess MCLs in the surrounding medium; open circles, thoroughly washed MCL-labeled cells without any excess MCLs in the surrounding medium. (B) The manipulation performance with different cell densities (collection time, 5 min) labeled with different units. White bars, 100 pg/cell; gray bars, 10 μg/ml. (C) The manipulation response with MCL-labeled cells (1 × 104 cells). All cell collection processes were performed by placing a magnet (size, ϕ10.95 × 15 mm; surface magnetic flux density, 520 mT) on microcentrifuge tubes. Results are expressed as mean ± SE (n = 3).
128
ITO ET AL.
only for the scale of 1000–100,000 cells/100 μl (equivalent to 104–106 cells/ml). Within this scale of cells, 10 μg/ml MCLs for cell labeling was found to be universally effective with any cell density (Fig. 3B). Moreover, to make clear the role of the excess MCL, we examined the cell collection rate using cells labeled by [1] 100 pg/cell and [2] 10 μg/ml. When 10,000 cells were examined in 10 μg/ml of MCL, the added amount of MCL corresponded to 100 pg/cell. In the case of 1000 cells, the excess MCLs in case 2 (10 μg/ml) were more than that of case 1 (100 pg/cell) and the cell collection rate of cells labeled by case 1 was significantly lower than that of case 2 (10 μg/ml), in which the MCL amount corresponds to 1000 pg/cell. On the other hand, 100,000 cells, the excess MCLs in case 2 (10 μg/ml) were lower than that of case 1 (100 pg/cell). However, the cell collection rate of cells labeled by case 1 (100 pg/cell) was similar to that of case 2 (10 μg/ml), in which the MCL amount corresponds to 10 pg/cell. This means that the excess free MCL in the solution has a good effect on cell manipulation even if MCL-labeled amount was low (in the case of 10 pg/cell). Since MCLs could not interact frequently with cells in the case of low cell concentration, the collection rate with 1000 cells was still slightly lower. However, the high collection rate of MCL-labeled cells by μg/ml compared to pg/cell also indicated that the more the excess MCLs exist, the more the manipulation performance increases with same cell number. To label cells without excess MCLs, the unit for adjusting the MCL concentration for labeling was first set to pg/cell. However, since the excess free MCL was also found to be effective for cell manipulation, we decided to unify the MCL unit to μg/ml. Moreover, by unifying the unit to μg/ml, MCLs could be added into the target well without considering the balance cell concentration or existing cell number into the well. The unit “μg/ml” makes the users more feasible to prepare and use MCLs in practical usage, since cell number determination in every step with rare cells is an inconvenient process. Consequently, we determined 10 μg/ml MCLs (100 pg/cell for 1 × 105 cells/ml) as the universal MCL-labeling condition. After this determination, the manipulation response of MCLlabeled cells with excess MCLs was evaluated (Fig. 3C). From the assay solution (100 μl volume), 104 cells could be completely collected within 3 min, which was much faster than by centrifugation. A positive effect of excess MCLs was also observed in cases with trypsinMCL treatments (data not shown). When there were excess MCLs, the cell manipulation performance after trypsin-MCL treatment was equivalent to centrifugation. Moreover, the viability of passaged cells after trypsin-MCL treatment was no different to viability of conventional trypsin treated cells (data not shown). Therefore, it was again confirmed that adhesion type cells could be sufficiently labeled and manipulated during cell passage. Speculated specifications of micro-scale cell assay devices for maximum MCL-labeled cell manipulation performance The proposed MCL cell-labeling technique is effective for different types of cell assays involving handling small cell volume. Moreover, the technique is believed to be most effective when it is combined with a palm-top size device system for performing personalized cell processing. Magnetic cell manipulation which also enables handling of small cell numbers is urgently required for cell therapy, as it promises completely sanitized cell assays with limited number of patient cells. These advantages of magnetic cell manipulation are suitable with such palm-top size device systems. Therefore, we examined the manipulation performance of the determined universal cell-labeling condition using small devices to provide guidelines for the design of cell manipulation devices for MCL cell labeling. With cells labeled using the universal labeling condition, 4 × 104 cells could be stably pulled up from a 400 μl solution from a depth of 11 mm, which was equal to the depth of the 96-well plates, using a small magnet (surface magnetic flux density, 520 mT) (Fig. 4). Therefore, our determined condition showed promise for maximum
J. BIOSCI. BIOENG.,
FIG. 4. Specification estimation for designing palm-top size device systems that combine MCL cell manipulation. Distance and MCL-labeled cell collection rate were examined. All cell collection processes were performed by placing a magnet (size, ϕ10.95 × 15 mm; surface magnetic flux density, 520 mT) for 3 min using the 96-well plate as described previously. Cell suspensions (4 × 104 cells in 400 μl) were labeled using the universal labeling condition. Results are expressed as mean ± SE (n = 3).
cell manipulation in any type of device smaller than the 96-well plate. From these results, the following equation was proposed for obtaining the optimum amount of MCLs to be prepared for labeling cells using the universal labeling condition: X½ng = 10½ng = μl × 11½mm × S mm2 (S, culture surface area; X, MCL amount to be prepared for labeling). In this study, the detailed analysis of MCL-labeled cell manipulation performance supported the potential of our previously proposed MMDs (15) and 10 μg/ml MCLs (100 pg/cell for 1 × 105 cells/ml) was determined as the universal MCL-labeling condition for various cell densities. Moreover, since this condition was also compatible with conventional evaluation techniques, such as microscopic observation and flow cytometric analysis (Supplementary Fig. 1 and Supplementary Table 1), it is expected that the concept of our MMD can be feasibly introduced into many medical systems. For the design of such devices or additional applications to single cell analysis methodologies based on magnetic cell transport, such as magnetic force-based lab-on-a-chip (28–30), our present results could provide guidelines for the use of MCLs. We believe our detailed analysis on basic cell manipulation technology will contribute to the investigation of personalized cell processing technology for cell therapy. APPENDIX A. SUPPLEMENTARY DATA Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jbiosc.2010.01.014. References 1. Chien, K.: Regenerative medicine and human models of human disease, Nature, 453, 302–305 (2008). 2. Tang, Q., Henriksen, K., Bi, M., Finger, E., Szot, G., Ye, J., Masteller, E., McDevitt, H., Bonyhadi, M., and Bluestone, J.: In vitro-expanded antigen-specific regulatory T cells suppress autoimmune diabetes, J. Exp. Med., 199, 1455–1465 (2004). 3. June, C.: Adoptive T cell therapy for cancer in the clinic, J. Clin. Invest., 117, 1466–1476 (2007). 4. Zakrzewski, J., Suh, D., Markley, J., Smith, O., King, C., Goldberg, G., Jenq, R., Holland, A., Grubin, J., and Cabrera-Perez, J., et al.: Tumor immunotherapy across MHC barriers using allogeneic T-cell precursors, Nat. Biotechnol., 26, 453–461 (2008). 5. Petit-Zeman, S.: Regenerative medicine, Nat. Biotechnol., 19, 201–206 (2001). 6. Dove, A.: Cell-based therapies go live, Nat. Biotechnol., 20, 339–343 (2002). 7. Daley, J., Dadey, B., Wysocki, M., Caligiuri, M., and Biddle, W.: Ex vivo expansion of human hematopoietic progenitor cells in serum-free StemPro™-34 medium, Focus, 18, 62–67 (1996).
VOL. 110, 2010 8. Liu, C., Chu, I., and Hwang, S.: Factorial designs combined with the steepest ascent method to optimize serum-free media for CHO cells, Enzyme Microb. Technol., 28, 314–321 (2001). 9. Yao, C., Liu, C., Chu, I., Hsieh, T., and Hwang, S.: Factorial designs combined with the steepest ascent method to optimize serum-free media for ex vivo expansion of human hematopoietic progenitor cells, Enzyme Microb. Technol., 33, 343–352 (2003). 10. Fan, X., Liu, T., Liu, Y., Ma, X., and Cui, Z.: Optimization of primary culture condition for mesenchymal stem cells derived from umbilical cord blood with factorial design, Biotechnol. Prog., 25, 499–507 (2009). 11. Perez-Ilzarbe, M., Diez-Campelo, M., Aranda, P., Tabera, S., Lopez, T., del Canizo, C., Merino, J., Moreno, C., Andreu, E., Prosper, F., and Perez-Simon, J.: Comparison of ex vivo expansion culture conditions of mesenchymal stem cells for human cell therapy, Transfusion, 49, 1901–1910 (2009). 12. Appelbaum, F.: Haematopoietic cell transplantation as immunotherapy, Nature, 411, 385–389 (2001). 13. Barry, F. and Murphy, J.: Mesenchymal stem cells: clinical applications and biological characterization, Int. J. Biochem. Cell Biol., 36, 568–584 (2004). 14. Lu, S., Feng, Q., Park, J., Vida, L., Lee, B., Strausbauch, M., Wettstein, P., Honig, G., and Lanza, R.: Biologic properties and enucleation of red blood cells from human embryonic stem cells, Blood, 112, 4475–4484 (2008). 15. Ito, H., Ryuji, K., Ino, K. and Honda, H., H., Ryuji, K., Ino, K., and Honda, H.: Magnetic manipulation device for the optimization of cell processing conditions, J. Biosci. Bioeng., 109, 182–188 (2010). 16. Shinkai, M., Yanase, M., Honda, H., Wakabayashi, T., Yoshida, J., and Kobayashi, T.: Intracellular hyperthermia for cancer using magnetite cationic liposomes: in vitro study, Jpn. J. Cancer Res., 87, 1179–1183 (1996). 17. Ito, A., Hayashida, M., Honda, H., Hata, K., Kagami, H., Ueda, M., and Kobayashi, T.: Construction and harvest of multilayered keratinocyte sheets using magnetite nanoparticles and magnetic force, Tissue Eng., 10, 873–880 (2004). 18. Ito, A., Ino, K., Hayashida, M., Kobayashi, T., Matsunuma, H., Kagami, H., Ueda, M., and Honda, H.: Novel methodology for fabrication of tissue-engineered tubular constructs using magnetite nanoparticles and magnetic force, Tissue Eng., 11, 1553–1561 (2005). 19. Shimizu, K., Ito, A., Yoshida, T., Yamada, Y., Ueda, M., and Honda, H.: Bone tissue
MCL LABELING CONDITIONS FOR CELL MANIPULATION
20.
21.
22. 23.
24.
25.
26. 27. 28.
29.
30.
129
engineering with human mesenchymal stem cell sheets constructed using magnetite nanoparticles and magnetic force, J. Biomed. Mater. Res., Part B, 82B, 471–480 (2007). Shimizu, K., Ito, A., Lee, J., Yoshida, T., Miwa, K., Ishiguro, H., Numaguchi, Y., Murohara, T., Kodama, I., and Honda, H.: Construction of multi-layered cardiomyocyte sheets using magnetite nanoparticles and magnetic force, Biotechnol. Bioeng., 96, 803–809 (2007). Ino, K., Ito, A., Kumazawa, H., Kagami, H., Ueda, M., and Honda, H.: Incorporation of capillary-like structures into dermal cell sheets constructed by magnetic forcebased tissue engineering, J. Chem. Eng. Jpn., 40, 51–58 (2007). Ino, K., Ito, A., and Honda, H.: Cell patterning using magnetite nanoparticles and magnetic force, Biotechnol. Bioeng., 97, 1309–1317 (2007). Ino, K., Okochi, M., Konishi, N., Nakatochi, M., Imai, R., Shikida, M., Ito, A., and Honda, H.: Cell culture arrays using magnetic force-based cell patterning for dynamic single cell analysis, Lab Chip, 8, 134–142 (2008). Ino, K., Okochi, M., and Honda, H.: Application of magnetic force-based cell patterning for controlling cell-cell interactions in angiogenesis, Biotechnol. Bioeng., 102, 882–890 (2009). Koide, H., Isaji, Y., Okochi, M., and Honda, H.: Preparation of L929 cell array by magnetic pin holder device for single cell function analysis, J. Chem. Eng. Jpn., 42, 290–297 (2009). Owen, C. and Sykes, N.: Magnetic labeling and cell sorting, J. Immunol. Methods, 73, 41–48 (1984). Inbar, M.: Fluidity of membrane lipids; a single cell analysis of mouse normal lymphocytes and malignant lymphoma-cells, FEBS Lett., 67, 180–185 (1976). Ito, H., Takayanagi, K., Okochi, M., Shikida, M., Sato, K., and Honda, H.: Proposal of pumpless, valveless, and flowless miniaturized reactor using magnetic beads for the portable analysis device, J. Chem. Eng. Jpn., 39, 1296–1299 (2006). Tsuchiya, H., Okochi, M., Nagao, N., Shikida, M., and Honda, H.: On-chip polymerase chain reaction microdevice employing. A magnetic droplet-manipulation system, Sens. Actuator B-Chem., 130, 583–588 (2008). Okochi, M., Tsuchiya, H., Kumazawa, F., Shikida, M., and Honda, H.: Dropletbased gene expression analysis using a device with magnetic force-based-droplethandling system, J. Biosci. Bioeng., 109, 193–197 (2010).
Practical cell labeling with magnetite cationic liposomes for cell manipulation Hiroshi Ito,1 Yurika Nonogaki,1 Ryuji Kato,1 and Hiroyuki Honda1,2,⁎ Department of Biotechnology, Graduate School of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan 1 and MEXT Innovative Research Center for Preventive Medical Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan 2 Received 27 November 2009; accepted 7 January 2010 Available online 4 February 2010
Personalization of the cell culture process for cell therapy is an ideal strategy to obtain maximum treatment effects. In a previous report, we proposed a strategy using a magnetic manipulation device that combined a palm-top size device and a cell-labeling method using magnetite cationic liposomes (MCLs) to enable feasible personalized cell processing. In the present study, we focused on optimizing the MCL-labeling technique with respect to cell manipulation in small devices. From detailed analysis with different cell types, 4 pg/cell of MCL-label was found to be obtained immediately after mixing with MCLs, which was sufficient for magnetic cell manipulation. The amount of label increased within 24 h depending on cell type, although in all cases it decreased along with cell doubling, indicating that the labeling potential of MCLs was limited. The role of free MCLs not involved in labeling was also investigated; MCLs' role was found to be a supportive one that maximized the manipulation performance up to 100%. We also determined optimum conditions to manipulate adherent cells by MCL labeling using the MCL dispersed in trypsin solution. Considering labeling feasibility and practical performance with 103–105 cells for personalized cell processing, we determined that 10 μg/ml of label without incubation time (0 h incubation) was the universal MCL-labeling condition. We propose the optimum specifications for a device to be combined with this method. © 2010, The Society for Biotechnology, Japan. All rights reserved. [Key words: Cell manipulation; Magnetite cationic liposome (MCL); Magnetic cell labeling; Small-scale cell handling; Magnetic force]
Cell-based therapy, as a part of regenerative medicine, is an innovative therapy to treat refractory defects of patients using their own cells (1–4). Although cell-based therapy is a promising approach (5, 6), its biggest drawback is the variability of clinically obtainable cells. Since cells from individual patients have significant individual differences, it is difficult to prepare optimal culture conditions for every patient's cells. To gain maximum effects with cell therapy, cell processing technology for personalized cell processing is required to culture patient's cells under most suitable conditions. For ideal personalized cell processing, it is desirable to obtain the optimal culture conditions for each patient's cells prior to mass production of cells for treatment. However, in order to obtain optimal cell culture conditions, many parameters, such as cytokine concentrations, medium components, and their inoculation density, should be examined (7–11), in spite of the limited number of cells obtained clinically from patients. Particularly in stem cell processing (12–14), the optimization of culture conditions is very difficult due to the limitation of acquiring adequate cell numbers relative to the large variations of culture parameters. Therefore, an effective technology to efficiently handle the limited number of cells and examine various culture conditions is required for personalized cell processing. Currently, the multi-well plate format along with automated cell assays is most frequently used to examine various culture conditions. ⁎ Corresponding author. Department of Biotechnology, Graduate School of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan. Tel.: + 81 52 789 3215; fax: +81 52 789 3214. E-mail address: [email protected] (H. Honda).
However, the multi-well plate format is accompanied by significant manipulation errors during the cell–liquid separation by centrifugation, especially with a small-volume sample (e.g., several hundred μl). To overcome this difficulty, we previously proposed a palm-top size device system, magnetic manipulation device (MMD), which is capable of multi-condition and micro-scale cell-based assays with a minimum loss of target cells in a closed environment (15). In the previous study, we described effective cell manipulation performance using magnetite cationic liposomes (MCLs) in combination with the multi-well plate format within a closed device that included a small magnet for cell manipulation. MCLs consist of 10 nm average-sized magnetite nanoparticles (Fe3O4) encapsulated in liposomes dioleoyl phosphatidylethanolamine (DOPE), dilauroyl phosphatidylcholine (DLPC), and N-(α-trimethylammonioacetyl)didodecyl-D-glutamate chloride (TMAG) (16). Since the primary binding force of an MCL label is the electrostatic force of cationic lipids and not dependent on antibodies, MCLs are able to label any cell type without prior knowledge of surface markers. Our previous study of MMD also indicated that the practical number of cells (104–106 cells) expected to be examined for one sample in cell-based assays could be labeled almost immediately and could be handled with high collection efficiency (N90% collection). This cell manipulation efficacy was far better than the commercial antibody-conjugated MACS (Magnetic Cell Sorting) beads system with a simple cylindrical magnet, and was considered to be more advantageous for use within small enclosed devices to enable personalized cell processing. Although MCLs have been confirmed as safe for animal adherent cells by our various researches (17–25), such as human umbilical vein
1389-1723/$ - see front matter © 2010, The Society for Biotechnology, Japan. All rights reserved. doi:10.1016/j.jbiosc.2010.01.014
VOL. 110, 2010
MCL LABELING CONDITIONS FOR CELL MANIPULATION
endothelial cells and normal human dermal fibroblast (21), and human mesenchymal stem cells, which could differentiate even if labeled with MCLs (19), detailed MCL-based cell-labeling mechanism and procedure with respect to cell manipulation require further investigation before practical use of MCLs for cell processing. Moreover, the application of MCL labeling to suspension type cells and tumor cells has not been examined. In this report, we examined the MCL-labeling performance for various cell types (normal and tumor adherent cells, normal and cancer suspension cells, and stem cells) and determined the manipulation capabilities that were not examined in previous studies. From detailed analysis of MCL-labeling conditions (capacity, time, and cell recovery rate), we determined universal MCL-labeling conditions for any cell type. We propose guidelines for using MCLs in palm-top size devices for personalized cell processing. MATERIALS AND METHODS Cells and culture condition Jurkat (human T-lymphoma cell line), RPMI-1788 (human B-lymphocyte cell line), and HT1080 (human sarcoma cell line) were obtained from the American Type Culture Collection (Manassas, VA, USA). Human mesenchymal stem cells (MSCs) were obtained from Lonza (Basel, Switzerland). Primary normal human fibroblasts (Fb) were collected from healthy volunteers with permission from the ethical committee of Nagoya University. Suspension type cells were cultured in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen), 100 U/ml penicillin (Invitrogen), 100 μg/ml streptomycin (Invitrogen), HEPES (2-[4-(2-hydroxyethyl)-1-piperazinyl] ethanesulfonic acid, Dojindo Laboratories, Kumamoto, Japan), and sodium bicarbonate (Wako Pure Chemical Industries, Osaka, Japan). HT1080 and Fb were cultured in Dulbecco's Modified Eagle Medium (Invitrogen) with 10% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin. MSCs were cultured in MSCGM (Mesenchymal Stem. Cell Growth Medium, Lonza). All cells were incubated at 37 °C in 5% CO2 and 95% air conditions. Preparation of MCLs Magnetite nanoparticles (Fe3O4; average particle size 10 nm; Toda Kogyo, Hiroshima, Japan) and 3 types of lipids (1:2:2 molar ratio), cationic TMAG (Sogo Pharmaceutical, Tokyo, Japan), DLPC (Sigma-Aldrich, St. Louis, MO, USA), and DOPE (Avanti Polar Lipids, Alabaster, AL, USA), were used (16). MCLs were mixed with a lipid and dried by evaporation for more than 30 min to form a lipid film. The film was hydrated into colloidal magnetite nanoparticles and vortexed. This resulted in selforganization of liposomes. All MCL concentrations were expressed as the net magnetite concentration by measuring the magnetite concentration using a potassium thiocyanate method (26). Determination of MCL-labeling performance with different cell types Jurkat and RPMI-1788 cells were seeded into 6-well cell culture plates (6 × 105 cells/well) with culture medium containing MCLs (net magnetite concentration, 100 pg/cell). Cell pellets were collected after 2 rounds of centrifugation and washed with phosphate buffered saline (PBS). Cell numbers were determined by trypan blue cell counting and their labeled magnetite amount was determined by the potassium thiocyanate method described previously. Fb, HT1080, and MSCs were first suspended in each culture medium containing MCLs (100 pg/cell) for labeling and then seeded into 6-well cell culture plates (2 × 104 cells/well). During cell incubation, MCLs remained in the medium. The cells were periodically collected by trypsin treatment and their numbers and labeled magnetite amount were determined using above described methods. Magnetite concentration for each cell type determined immediately after mixing with MCL was defined as the basal amount of “attached MCLs” on cells. Magnetite concentration obtained from the later incubation time was considered as the labeled amount of MCLs, including “attached MCLs” and “taken up MCLs.” Determination of MCL-labeling performance for cell manipulation Jurkat cells were used as a model to determine the relationship between MCL concentration 5 for labeling and the labeled amount. The cells were suspended (1 × 10 cells/ml, 2 ml) and labeled with varying MCL concentrations (10, 20, 40, 60, 80, and 100 pg/cell) without incubation time (0 h incubation), followed by measurement of the labeled amount. Labeled cells (1 × 104 cells, 100 μl) were re-suspended in siliconized 1.5-ml microcentrifuge tubes (Fisher Thermo Fisher Scientific, Waltham, MA, USA) for the following cell collection assay. A cylindrical magnet (shape-type, cylindrical; size, ϕ10.95 × 15 mm; surface magnetic flux density, 520 mT; magnetic attractive force, 4.0 kg; Niroku seisakusho, Kobe, Japan) was placed on the side of a tube for 5 min to collect MCL-labeled cells. Collected cells were re-suspended in fresh culture medium without the magnet and then transferred into a 96-well cell culture plate (Greiner bioone, Kremsmünster, Austria) for the determination of the collected cell number by the CellTiter-Blue Assay (Promega, Madison, WI, USA). For the positive control, cells (1 × 105 cells/ml) without MCL addition were collected as pellets by centrifugation and measured by the CellTiter-Blue Assay. In addition to Jurkat cells, as the sample with high labeled amount, HT1080 cells (1 × 104 cells; MCL concentration, 60 pg/cell) were also examined using the same protocol in order to determine the relationship between the labeled amount and manipulation performance.
125
Determination of cell-labeling performance using trypsin-MCL The celllabeling performance of MCL was examined for Fb, HT1080, and MSCs using the MCL dispersed in trypsin solution (designated as trypsin-MCL). After the 24 h culture, cells seeded into 6-well cell culture plates (2 × 104 cells/well) were treated for 3 min with trypsin-MCL (0.05% trypsin, 100 pg/cell MCL). After inactivating trypsin by the addition of culture media to cell suspensions, the cells were collected by centrifugation and their MCL-labeled amount was determined as described above. Cell manipulation enhancement by excess MCL addition Jurkat cells were used to determine the enhancing effect of excess MCL addition into the target well. To examine labeling conditions, cells (1 × 104 cells/tube, 100 μl) were labeled with varying MCL concentrations (10, 20, 40, 60, 80, and 100 pg/cell) without incubation time and separated into two groups for comparisons: [1] washed cells which were subjected to 2 rounds of centrifugation and washed with PBS as described above and [2] unwashed cells with excess MCLs. In previous methods, labeling was described for the MCL concentration in pg/cell. However, 100 pg/cells for 1 × 104 in 100 μl cells was determined to be the universal labeling condition (10 μg/ml, without incubation time). To examine collection performance, different cell numbers (1 × 103, 1 × 104, and 1 × 105 cells) were labeled using the universal MCL-labeling condition. To determine collection speed, 1 × 104 cells were labeled using the universal labeling condition. Determination of expected device specifications based on MCL-labeled cell manipulation performance The manipulation distance of MCL-labeled cells was determined by collecting labeled cells (4 × 104 cells labeled using universal labeling conditions) from different distances (11, 12, 13, 14, and 16 mm), as described previously (15).
RESULTS AND DISCUSSION MCL-labeling performance for different cell types For optimized personalized cell processing, several days of cell culture are essential. Therefore, profiles of MCL-labeled amounts and their changes throughout the culture period were examined with different cell types. The incorporation of MCL during cell labeling was approximately equivalent with different cell types without incubation time (0 h incubation), and was considered to be not critical for further investigation. However, the labeled amount changed for some cells during the 24 h incubation (Fig. 1A). The average immediate labeled amount of all cells was 4 pg/cell, which could be regarded as the universal labeled amount for most cells after a short incubation time. Since the main driving force of MCL labeling during mixing is the electrostatic force of cationic lipids, it was convincing to find no difference among cell types. These results also indicated that the labeled amount of tumor cells (HT1080) was equivalent to that of normal cells (Fb), whereas that of suspension cancer cells (Jurkat) was much lower than that of normal cells (RPMI-1788). Since RPMI-1788 cells proliferated forming large aggregation, MCLs might be involved with aggregations and the cells got contact with MCLs more easily than Jurkat cells. Moreover, although lipid membrane fluidity of tumor cells is generally greater than that of normal cells (27), no difference in MCL labeling was observed. After 3 days of cell culture with excess MCLs in the medium, the MCL-labeled amount of all cell types decreased in exactly the same manner as the idealized amount of magnetite remaining after cell doubling (Fig. 1B). Along with the results indicated in Fig. 1A, the above results indicated that MCL labeling was mainly due to cellular responses to interactions with liposomes, and once MCLs were incorporated in the cytosol, they were strongly retained within cells and only decreased due to cell doubling. Interestingly, excess MCLs in the medium did not contribute to additional cell labeling after 24 h. Moreover, the excess MCLs showed no toxicity on growth of all cell types with 3 days culture (data not shown). Especially, the effects on a suspension type normal cell line were first examined and shown to be non-toxic. It was important that such normal suspension type cells could be treated with MCLs without toxicity, since these suspension type cells essentially require an effective solid-liquid separation for medium exchange compared to the adhesion type cells. Particle size distribution analysis indicated that large particles rapidly appeared in the cell culture medium but not in PBS or ultrapure water (data not shown). Therefore, it was deduced that charged protein molecules in the medium triggered MCL aggregation that negatively
126
ITO ET AL.
J. BIOSCI. BIOENG.,
FIG. 1. MCL-labeling profiles of different cell types. (A) The labeled MCL amount with rapid labeling conditions (without incubation time: 0 h) and 24 h incubation. (B) The changes of labeled MCL of magnetite per cell. Solid circles, measured MCL-labeled amount; open circles, estimated amount of MCLs remaining after cell doubling. This estimate was determined by dividing the first day magnetite amount by growth rate for each day. (C) Labeled MCL with trypsin-MCL treatment for 3 min. Results are expressed as means ± SE (n = 3). All labeling conditions, including the trypsin-MCL treatment, were set to an MCL concentration of 100 pg/cell of magnetite, and cell number was set for adherent (2 × 104 cells) and suspension cells (6 × 105 cells). Fb, fibroblasts; HT1080, fibrosarcoma; MSC, mesenchymal stem cell; Jurkat, T-lymphoma cell line; RPMI-1788, B-lymphocyte cell line.
affected cell-MCL interactions at later incubation times. Taken together, it was clear that only freshly dispersed MCLs were effective in cell labeling within 24 h, and that labeled cells could be sufficiently manipulated based on the magnetite remaining in the cytosol which was calculated from their 24 h labeled amount and doubling time. Compared to suspension type cells, adhesion type cells were considerably more difficult to label for long times throughout their passage culture. Thus, there are few palm-top size devices that can magnetically manipulate adhesion type cells during their passage culture. Therefore, we examined if MCL labels could be retained in adherent cells by introducing MCLs with trypsin treatment for applications in manipulating them throughout their passage culture
in a palm-top size device. Cells could be successfully labeled at 6 pg/ cells by treatment with trypsin-MCL for 3 min (Fig. 1C). This MCLlabeled amount was equivalent to the universal amount of cell labeling in media without incubation time. Therefore, it was concluded that any cell type could be labeled rapidly by adding fresh MCL or by trypsin-MCL treatment at their passages (designated as a rapid labeling protocol in the following sections). Relationship between MCL labeling and manipulation performance We investigated the relationship between the MCL concentration for labeling and the actual MCL-labeled amount (Fig. 2A). An increase in MCL concentration in order to increase the amount of labeled MCL per cell was effective up to 60 pg/cell.
VOL. 110, 2010
MCL LABELING CONDITIONS FOR CELL MANIPULATION
127
considered that not charged proteins but the magnetic force formed condensation of MCLs in the solution that incorporated cells and assisted the magnetic cell collection. Therefore, the formerly determined rapid labeling condition was considered to be practically useful with in any type of medium or solution if there were MCLs remaining after cell labeling. Next, we examined the capacity of this excess MCL-labeling condition. To assay cells in a conventional palm-top size device system, solution volume in the device is usually limited to a volume (i.e., b400 μl into 96-well plates) that makes centrifugation difficult. With such solution volume, reasonable cell culture could be managed
FIG. 2. Relationship between labeled MCL and manipulation performance. (A) The relationship between MCL concentration for labeling and MCL-labeled amount (actual labeled amount). Jurkat cells were labeled with varying MCL concentrations. (B) The relationship between the labeled MCL and cell manipulation performance. Jurkat and HT1080 cells (1 × 104) were used after labeling with varying MCL concentrations under the rapid or 1 day labeling conditions. All cell collection processes were performed by placing a magnet (size, ϕ10.95 × 15 mm; surface magnetic flux density, 520 mT) on the side of a siliconized tube for 5 min. Solid circles, Jurkat; open circles, HT1080 which represented sample with high labeled amount. Results are expressed as mean ± SE (n = 3).
The cell manipulation performance was presumed to reach 100% with a labeled amount of 10 pg/cell (Fig. 2B). Although, in addition to Jurkat, HT1080 was used as the sample with high labeled amount, it was considered that not cell type but labeled amount influenced the manipulation performance dominantly. This suggested that Fb, HT1080, MSC, and RPMI-1788 could be adequately manipulated for 2 days (Fig. 1B). Contribution of excess MCLs for cell manipulation performance From these results, it appeared as if the rapid labeling protocol described in previous report (15) was insufficient (b60%) for cell manipulation. However, during practical MCL labeling in closed devices, excess MCLs were not washed away. Therefore, even if rapid MCL labeling is not adequate for 100% manipulation with labeled MCLs (4–6 pg/cells), if the excess free MCLs contribute as support molecules, the manipulation performance should be high enough for practical use. To investigate the supporting effect of excess MCLs for cell manipulation performance, we compared the cell manipulation performance of cells labeled with the same MCL concentration followed by a process of washing or no washing (Fig. 3A). The manipulation performance reached a plateau at a 60% collection rate when remaining MCLs were washed away. This result was consistent with that the labeled amount of Jurkat cells reached a plateau at about 6 pg/cell (Fig. 2). However, it was observed that, even with the same MCL concentration for labeling (6 μg/ml), the cell manipulation rate reached 100% when excess MCLs were not eliminated. By microscopic observation, pellet-like condensation including all free MCLs was observed. Consequently, since it was found that MCLs do not accumulate on cells after 24 h (Fig. 1B), we
FIG. 3. Cell manipulation performance with excess MCLs. (A) The cell manipulation performance with different cell-labeling concentrations (collection time, 5 min). Jurkat cells (1 × 104 cells) were labeled with varying MCL concentrations under the rapid labeling condition. Solid circles, unwashed MCL-labeled cells in the presence of excess MCLs in the surrounding medium; open circles, thoroughly washed MCL-labeled cells without any excess MCLs in the surrounding medium. (B) The manipulation performance with different cell densities (collection time, 5 min) labeled with different units. White bars, 100 pg/cell; gray bars, 10 μg/ml. (C) The manipulation response with MCL-labeled cells (1 × 104 cells). All cell collection processes were performed by placing a magnet (size, ϕ10.95 × 15 mm; surface magnetic flux density, 520 mT) on microcentrifuge tubes. Results are expressed as mean ± SE (n = 3).
128
ITO ET AL.
only for the scale of 1000–100,000 cells/100 μl (equivalent to 104–106 cells/ml). Within this scale of cells, 10 μg/ml MCLs for cell labeling was found to be universally effective with any cell density (Fig. 3B). Moreover, to make clear the role of the excess MCL, we examined the cell collection rate using cells labeled by [1] 100 pg/cell and [2] 10 μg/ml. When 10,000 cells were examined in 10 μg/ml of MCL, the added amount of MCL corresponded to 100 pg/cell. In the case of 1000 cells, the excess MCLs in case 2 (10 μg/ml) were more than that of case 1 (100 pg/cell) and the cell collection rate of cells labeled by case 1 was significantly lower than that of case 2 (10 μg/ml), in which the MCL amount corresponds to 1000 pg/cell. On the other hand, 100,000 cells, the excess MCLs in case 2 (10 μg/ml) were lower than that of case 1 (100 pg/cell). However, the cell collection rate of cells labeled by case 1 (100 pg/cell) was similar to that of case 2 (10 μg/ml), in which the MCL amount corresponds to 10 pg/cell. This means that the excess free MCL in the solution has a good effect on cell manipulation even if MCL-labeled amount was low (in the case of 10 pg/cell). Since MCLs could not interact frequently with cells in the case of low cell concentration, the collection rate with 1000 cells was still slightly lower. However, the high collection rate of MCL-labeled cells by μg/ml compared to pg/cell also indicated that the more the excess MCLs exist, the more the manipulation performance increases with same cell number. To label cells without excess MCLs, the unit for adjusting the MCL concentration for labeling was first set to pg/cell. However, since the excess free MCL was also found to be effective for cell manipulation, we decided to unify the MCL unit to μg/ml. Moreover, by unifying the unit to μg/ml, MCLs could be added into the target well without considering the balance cell concentration or existing cell number into the well. The unit “μg/ml” makes the users more feasible to prepare and use MCLs in practical usage, since cell number determination in every step with rare cells is an inconvenient process. Consequently, we determined 10 μg/ml MCLs (100 pg/cell for 1 × 105 cells/ml) as the universal MCL-labeling condition. After this determination, the manipulation response of MCLlabeled cells with excess MCLs was evaluated (Fig. 3C). From the assay solution (100 μl volume), 104 cells could be completely collected within 3 min, which was much faster than by centrifugation. A positive effect of excess MCLs was also observed in cases with trypsinMCL treatments (data not shown). When there were excess MCLs, the cell manipulation performance after trypsin-MCL treatment was equivalent to centrifugation. Moreover, the viability of passaged cells after trypsin-MCL treatment was no different to viability of conventional trypsin treated cells (data not shown). Therefore, it was again confirmed that adhesion type cells could be sufficiently labeled and manipulated during cell passage. Speculated specifications of micro-scale cell assay devices for maximum MCL-labeled cell manipulation performance The proposed MCL cell-labeling technique is effective for different types of cell assays involving handling small cell volume. Moreover, the technique is believed to be most effective when it is combined with a palm-top size device system for performing personalized cell processing. Magnetic cell manipulation which also enables handling of small cell numbers is urgently required for cell therapy, as it promises completely sanitized cell assays with limited number of patient cells. These advantages of magnetic cell manipulation are suitable with such palm-top size device systems. Therefore, we examined the manipulation performance of the determined universal cell-labeling condition using small devices to provide guidelines for the design of cell manipulation devices for MCL cell labeling. With cells labeled using the universal labeling condition, 4 × 104 cells could be stably pulled up from a 400 μl solution from a depth of 11 mm, which was equal to the depth of the 96-well plates, using a small magnet (surface magnetic flux density, 520 mT) (Fig. 4). Therefore, our determined condition showed promise for maximum
J. BIOSCI. BIOENG.,
FIG. 4. Specification estimation for designing palm-top size device systems that combine MCL cell manipulation. Distance and MCL-labeled cell collection rate were examined. All cell collection processes were performed by placing a magnet (size, ϕ10.95 × 15 mm; surface magnetic flux density, 520 mT) for 3 min using the 96-well plate as described previously. Cell suspensions (4 × 104 cells in 400 μl) were labeled using the universal labeling condition. Results are expressed as mean ± SE (n = 3).
cell manipulation in any type of device smaller than the 96-well plate. From these results, the following equation was proposed for obtaining the optimum amount of MCLs to be prepared for labeling cells using the universal labeling condition: X½ng = 10½ng = μl × 11½mm × S mm2 (S, culture surface area; X, MCL amount to be prepared for labeling). In this study, the detailed analysis of MCL-labeled cell manipulation performance supported the potential of our previously proposed MMDs (15) and 10 μg/ml MCLs (100 pg/cell for 1 × 105 cells/ml) was determined as the universal MCL-labeling condition for various cell densities. Moreover, since this condition was also compatible with conventional evaluation techniques, such as microscopic observation and flow cytometric analysis (Supplementary Fig. 1 and Supplementary Table 1), it is expected that the concept of our MMD can be feasibly introduced into many medical systems. For the design of such devices or additional applications to single cell analysis methodologies based on magnetic cell transport, such as magnetic force-based lab-on-a-chip (28–30), our present results could provide guidelines for the use of MCLs. We believe our detailed analysis on basic cell manipulation technology will contribute to the investigation of personalized cell processing technology for cell therapy. APPENDIX A. SUPPLEMENTARY DATA Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jbiosc.2010.01.014. References 1. Chien, K.: Regenerative medicine and human models of human disease, Nature, 453, 302–305 (2008). 2. Tang, Q., Henriksen, K., Bi, M., Finger, E., Szot, G., Ye, J., Masteller, E., McDevitt, H., Bonyhadi, M., and Bluestone, J.: In vitro-expanded antigen-specific regulatory T cells suppress autoimmune diabetes, J. Exp. Med., 199, 1455–1465 (2004). 3. June, C.: Adoptive T cell therapy for cancer in the clinic, J. Clin. Invest., 117, 1466–1476 (2007). 4. Zakrzewski, J., Suh, D., Markley, J., Smith, O., King, C., Goldberg, G., Jenq, R., Holland, A., Grubin, J., and Cabrera-Perez, J., et al.: Tumor immunotherapy across MHC barriers using allogeneic T-cell precursors, Nat. Biotechnol., 26, 453–461 (2008). 5. Petit-Zeman, S.: Regenerative medicine, Nat. Biotechnol., 19, 201–206 (2001). 6. Dove, A.: Cell-based therapies go live, Nat. Biotechnol., 20, 339–343 (2002). 7. Daley, J., Dadey, B., Wysocki, M., Caligiuri, M., and Biddle, W.: Ex vivo expansion of human hematopoietic progenitor cells in serum-free StemPro™-34 medium, Focus, 18, 62–67 (1996).
VOL. 110, 2010 8. Liu, C., Chu, I., and Hwang, S.: Factorial designs combined with the steepest ascent method to optimize serum-free media for CHO cells, Enzyme Microb. Technol., 28, 314–321 (2001). 9. Yao, C., Liu, C., Chu, I., Hsieh, T., and Hwang, S.: Factorial designs combined with the steepest ascent method to optimize serum-free media for ex vivo expansion of human hematopoietic progenitor cells, Enzyme Microb. Technol., 33, 343–352 (2003). 10. Fan, X., Liu, T., Liu, Y., Ma, X., and Cui, Z.: Optimization of primary culture condition for mesenchymal stem cells derived from umbilical cord blood with factorial design, Biotechnol. Prog., 25, 499–507 (2009). 11. Perez-Ilzarbe, M., Diez-Campelo, M., Aranda, P., Tabera, S., Lopez, T., del Canizo, C., Merino, J., Moreno, C., Andreu, E., Prosper, F., and Perez-Simon, J.: Comparison of ex vivo expansion culture conditions of mesenchymal stem cells for human cell therapy, Transfusion, 49, 1901–1910 (2009). 12. Appelbaum, F.: Haematopoietic cell transplantation as immunotherapy, Nature, 411, 385–389 (2001). 13. Barry, F. and Murphy, J.: Mesenchymal stem cells: clinical applications and biological characterization, Int. J. Biochem. Cell Biol., 36, 568–584 (2004). 14. Lu, S., Feng, Q., Park, J., Vida, L., Lee, B., Strausbauch, M., Wettstein, P., Honig, G., and Lanza, R.: Biologic properties and enucleation of red blood cells from human embryonic stem cells, Blood, 112, 4475–4484 (2008). 15. Ito, H., Ryuji, K., Ino, K. and Honda, H., H., Ryuji, K., Ino, K., and Honda, H.: Magnetic manipulation device for the optimization of cell processing conditions, J. Biosci. Bioeng., 109, 182–188 (2010). 16. Shinkai, M., Yanase, M., Honda, H., Wakabayashi, T., Yoshida, J., and Kobayashi, T.: Intracellular hyperthermia for cancer using magnetite cationic liposomes: in vitro study, Jpn. J. Cancer Res., 87, 1179–1183 (1996). 17. Ito, A., Hayashida, M., Honda, H., Hata, K., Kagami, H., Ueda, M., and Kobayashi, T.: Construction and harvest of multilayered keratinocyte sheets using magnetite nanoparticles and magnetic force, Tissue Eng., 10, 873–880 (2004). 18. Ito, A., Ino, K., Hayashida, M., Kobayashi, T., Matsunuma, H., Kagami, H., Ueda, M., and Honda, H.: Novel methodology for fabrication of tissue-engineered tubular constructs using magnetite nanoparticles and magnetic force, Tissue Eng., 11, 1553–1561 (2005). 19. Shimizu, K., Ito, A., Yoshida, T., Yamada, Y., Ueda, M., and Honda, H.: Bone tissue
MCL LABELING CONDITIONS FOR CELL MANIPULATION
20.
21.
22. 23.
24.
25.
26. 27. 28.
29.
30.
129
engineering with human mesenchymal stem cell sheets constructed using magnetite nanoparticles and magnetic force, J. Biomed. Mater. Res., Part B, 82B, 471–480 (2007). Shimizu, K., Ito, A., Lee, J., Yoshida, T., Miwa, K., Ishiguro, H., Numaguchi, Y., Murohara, T., Kodama, I., and Honda, H.: Construction of multi-layered cardiomyocyte sheets using magnetite nanoparticles and magnetic force, Biotechnol. Bioeng., 96, 803–809 (2007). Ino, K., Ito, A., Kumazawa, H., Kagami, H., Ueda, M., and Honda, H.: Incorporation of capillary-like structures into dermal cell sheets constructed by magnetic forcebased tissue engineering, J. Chem. Eng. Jpn., 40, 51–58 (2007). Ino, K., Ito, A., and Honda, H.: Cell patterning using magnetite nanoparticles and magnetic force, Biotechnol. Bioeng., 97, 1309–1317 (2007). Ino, K., Okochi, M., Konishi, N., Nakatochi, M., Imai, R., Shikida, M., Ito, A., and Honda, H.: Cell culture arrays using magnetic force-based cell patterning for dynamic single cell analysis, Lab Chip, 8, 134–142 (2008). Ino, K., Okochi, M., and Honda, H.: Application of magnetic force-based cell patterning for controlling cell-cell interactions in angiogenesis, Biotechnol. Bioeng., 102, 882–890 (2009). Koide, H., Isaji, Y., Okochi, M., and Honda, H.: Preparation of L929 cell array by magnetic pin holder device for single cell function analysis, J. Chem. Eng. Jpn., 42, 290–297 (2009). Owen, C. and Sykes, N.: Magnetic labeling and cell sorting, J. Immunol. Methods, 73, 41–48 (1984). Inbar, M.: Fluidity of membrane lipids; a single cell analysis of mouse normal lymphocytes and malignant lymphoma-cells, FEBS Lett., 67, 180–185 (1976). Ito, H., Takayanagi, K., Okochi, M., Shikida, M., Sato, K., and Honda, H.: Proposal of pumpless, valveless, and flowless miniaturized reactor using magnetic beads for the portable analysis device, J. Chem. Eng. Jpn., 39, 1296–1299 (2006). Tsuchiya, H., Okochi, M., Nagao, N., Shikida, M., and Honda, H.: On-chip polymerase chain reaction microdevice employing. A magnetic droplet-manipulation system, Sens. Actuator B-Chem., 130, 583–588 (2008). Okochi, M., Tsuchiya, H., Kumazawa, F., Shikida, M., and Honda, H.: Dropletbased gene expression analysis using a device with magnetic force-based-droplethandling system, J. Biosci. Bioeng., 109, 193–197 (2010).