Programmable Display of DNA–Protein Chimeras for Controlling Cell

Article pubs.acs.org/Biomac

Programmable Display of DNA−Protein Chimeras for Controlling Cell−Hydrogel Interactions via Reversible Intermolecular Hybridization Zhaoyang Zhang,†,‡ Shihui Li,§ Niancao Chen,§ Cheng Yang,§ and Yong Wang*,‡,§,∥ Departments of ‡Chemical and Biomolecular Engineering and §Biomedical Engineering, University of Connecticut, Storrs, Connecticut 06269, United States ABSTRACT: Extensive studies have been recently carried out to achieve dynamic control of cell−material interactions primarily through physicochemical stimulation. The purpose of this study was to apply reversible intermolecular hybridization to program cell−hydrogel interactions in physiological conditions based on DNA−antibody chimeras and complementary oligonucleotides. The results showed that DNA oligonucleotides could be captured to and released from the immobilizing DNA-functionalized hydrogels with high specificity via DNA hybridization. Accordingly, DNA−antibody chimeras were captured to the hydrogels, successfully inducing specific cell attachment. The cell attachment to the hydrogels reached the plateau at approximately half an hour after the functionalized hydrogels and the cells were incubated together. The attached cells were rapidly released from the bound hydrogels when triggering complementary oligonucleotides were introduced to the system. However, the capability of the triggering complementary oligonucleotides in releasing cells was affected by the length of intermolecular hybridization. The length needed to be at least more than 20 base pairs in the current experimental setting. Notably, because the procedure of intermolecular hybridization did not involve any harsh condition, the released cells maintained the same viability as that of the cultured cells. The functionalized hydrogels also exhibited the potential to catch and release cells repeatedly. Therefore, this study demonstrates that it is promising to regulate cell-material interactions dynamically through the DNA-programmed display of DNA−protein chimeras.

1. INTRODUCTION Dynamic control of specific cell−material interactions is important in not only basic research, but also practical applications such as tissue engineering and cell separation.1−6 Specific cell−material interactions can be achieved through the functionalization of materials with a number of biomolecules such as antibodies, peptides, and growth factors, because many of them have been disclosed to interact with cell receptors specifically.7−9 However, nature does not provide a mechanism that allows for the active dissociation of these biomolecules from the receptors in physiological conditions. As a result, once cells bind to a functionalized material, it is challenging to induce cell release from the material in a specific and nondestructive manner. Therefore, great efforts have been recently made to explore the application of physicochemical stimulation to regulate cell−material interactions dynamically.10,11 Numerous physicochemical stimuli including light, electricity, temperature, mechanical stress, and ions have been studied to modulate the properties of materials and the interactions between cells and materials.10,12−18 For instance, light and electricity have been used to trigger the dissociation of affinity peptides from responsive materials to modulate the cell binding states.13,14 Indeed, these physicochemical stimuli hold great potential for the effective regulation of the interactions between © 2013 American Chemical Society

cells and materials. However, the generation of these stimuli often relies on large instruments or requires a significant shift of physiological conditions (e.g., temperature variation or ion exchange). Therefore, it would be desirable to develop materials whose properties could be modulated by biocompatible molecular triggers. Nucleic acid oligonucleotides have been recently used as a biomolecular trigger to control the properties of aptamerfunctionalized hydrogels because they can form complementary base pairs with aptamers.19−21 Resultantly, aptamers lose their binding functionalities through intermolecular hybridization and release the bound proteins or cells in physiological conditions.20−23 These studies have demonstrated the promise of using aptamers and triggering oligonucleotides to regulate cell−hydrogel interactions dynamically. However, aptamerbased studies often require special expertise. Moreover, the availability of the aptamers that bind cells with high affinity and specificity is currently limited. In contrast, a large number of antibodies, peptides, and growth factors are available. The techniques of producing and modifying these biomolecules are Received: January 20, 2013 Revised: February 15, 2013 Published: March 4, 2013 1174

dx.doi.org/10.1021/bm400096z | Biomacromolecules 2013, 14, 1174−1180

Biomacromolecules

Article

also well established.24−26 Therefore, we intended to couple nucleic acids and proteins to develop DNA−protein chimeras for the functionalization of hydrogels. The chimeras would not only possess the ability to recognize target cells, but also have the function to undergo DNA-mediated reversible hybridization. The overall experimental design of this study is shown in Figure 1. The DNA-functionalized hydrogel was coated on the

WI). FITC-conjugated streptavidin and 3-(trimethoxysilyl) propyl methacrylate were purchased from Sigma-Aldrich (Louis, MO). Dulbecco’s phosphate buffered saline (DPBS), bovine serum albumin (BSA), Live/Dead staining kits, and biotinylated goat antihuman IgM antibody were purchased from Invitrogen (Carlsbad, CA). Biotinylated isotype control antibody was purchased from SouthernBiotech (Birmingham, AL). All oligonucleotides (Table 1) were synthesized by Integrated DNA Technologies (Coralville, IA). 2.2. Preparation of Silanized Glass Surface. Microscope glass slides were cut into small squares with a dimension of approximately 4 × 4 mm2. The glass squares were treated with 1 M NaOH for 10 min and then thoroughly washed with deionized water. After drying in the oven, the squares were incubated in a silanization solution for 5 min. The silanization solution was prepared by diluting 0.5 mL of 3-(trimethoxysilyl) propyl methacrylate in 50 mL of ethanol supplemented by 1.5 mL of dilute acetic acid (10% v/v). The silanized glass squares were washed with pure ethanol to terminate the reaction and clean the surface. 2.3. Synthesis of Hydrogels on Glass Surface. A thin layer of polyacrylamide hydrogel was synthesized on the silanized glass surface. The reaction solution was prepared by mixing 1 μL of 10% acrylamide solution containing sequence A (50 μM), 0.15 μL of APS (10% w/v), and 0.15 μL of TEMED (5% v/v). The reaction solution was transferred to a large piece of clean glass and covered by the silanized glass squares. A total of 1 h after the polymerization, the glass squares were carefully lifted and thoroughly rinsed with the PBS solution. 2.4. Fluorescence Imaging of Hydrogel. The hydrogel conjugated to the glass substrate was incubated in the solution of 10 μM TAMRA-labeled sequence B at 37 °C for 1 h. After thoroughly washed with DPBS, the hydrogel was imaged under an inverted fluorescence microscope (Axiovert 40CFL, Carl Zeiss). To illustrate hybridization-mediated dissociation of the AB complexes, the hydrogel was further incubated in the solution of 5 μM sequence C at 37 °C for 0.5 h and imaged with the inverted fluorescence microscope. 2.5. Cell Culture. Ramos cells (CRL-1596, human B lymphocyte cell line) and CCRF-CEM cells (CCL-119, human T lymphocytic leukemia cell line) were obtained from ATCC (Manassas, VA). Ramos cells were cultured in RPMI 1640 medium (ATCC, Manassas, VA) supplemented with 10% deactivated fetal bovine serum (ATCC, Manassas, VA) and 100 IU/mL penicillin−streptomycin (Mediatech, Manassas, VA) at 37 °C in a humidified incubator containing 5% CO2. CCRFCEM cells were cultured in the same conditions as that of Ramos cells except that 10% normal fetal bovine serum was used.

Figure 1. Schematic representation of the overall experimental design. (A) The synthesis of DNA-functionalized hydrogel on a glass substrate for the display of antibody−DNA chimeras. (B) Reversible nucleic acid hybridization for programmed cell catch and release.

glass surface via free radical polymerization. The hydrogel coating was treated with the DNA−antibody chimeras for cell catch and complementary sequences for cell release. Based on this design, numerous assays were used to prove the concept. Specifically, reversible intermolecular hybridization was examined using both gel electrophoresis and fluorescence imaging. The cell binding functionality of the antibodies was characterized with flow cytometry. Microscopy and cell counting were further used to determine the kinetics of cell attachment and release. A live/dead staining assay was also carried out to study the viability of released cells. Because the entire cell catch and release procedure was presumably nondestructive, the final experiment was pursued to verify the feasibility of regenerating the hydrogels for cell catch and release.

2. MATERIALS AND METHODS 2.1. Materials. Phosphate buffered saline (PBS), a solution of acrylamide and bis-acrylamide (40% w/v; 29:1), ammonium persulfate (APS), N,N,N′,N′-tetramethylenediamine (TEMED), sodium hydroxide, and microscope glass slide were purchased from Fisher Scientific (Suwanee, GA). Streptavidin was purchased from Promega Inc. (Madison, Table 1. List of Oligonucleotide Sequences name A B Bs C15 C20 C25 C25S C30

sequence 5′-/acrydite/-ATATTGTTTGTTACACGGGATCCCGATTTT-3′ 5′- /biotin/-TAACATAGGTGGATAATTTGAAACGAAAATCGGGATCCCGTGTAA-3′ 5′-/TAMRA/-TAACATAGGTGGATAATTTGAAACGAAAATCGGGATCCCGTGTAA-3′ 5′-TCGAAGACGAGAAGCGAAGTTATAGTTTATAGCGAAATACGCTTA-3′ 5′-TTACACGGGATCCCG-3′ 5′-TTACACGGGATCCCGATTTT-3′ 5′-TTACACGGGATCCCGATTTTCGTTT-3′ 5′-ACACCTTGTCTTATTGTTCGCGGTA-3′ 5′-TTACACGGGATCCCGATTTTCGTTTCAAAT-3′ 1175

dx.doi.org/10.1021/bm400096z | Biomacromolecules 2013, 14, 1174−1180

Biomacromolecules

Article

2.6. Flow Cytometry. A total of 5 × 105 cells were incubated in 10 nM biotinylated antihuman IgM antibody or isotype control antibody solution at 4 °C for 0.5 h. After washing, the cells were further incubated in a FITC-conjugated streptavidin (2.5 ng/μL) solution at 4 °C for 0.5 h. The cells were washed again and subsequently analyzed using a BD FACSCaliburTM flow cytometer (San Jose, CA). The buffer used for washing cells was cold DPBS containing 4.5 g/L glucose and 10 mM MgCl2. 2.7. Cell Catch and Release. The buffer used for cell catch and release was DPBS containing 4.5 g/L glucose, 10 mM MgCl2, and 0.1% (w/v) BSA. The glass squares with hydrogel coatings were sequentially incubated in the sequence B solution (5 μM) for 1 h, the streptavidin solution (10 μM) for 0.5 h, and the biotinylated antibody solution (1 μM) for 0.5 h. Each incubation was followed by thorough washing. After the immobilization of the DNA−antibody chimeras, the glass squares were transferred to a 24-well plate and incubated in 800 μL of cell suspension (5× 105 cells/well) at 37 °C for 0.5 h. The unbound cells were gently removed from the hydrogel surface by shaking at 90 rpm for 2 min. To examine cell release, the glass squares were incubated in the solution containing 5 μM of C25 or control sequences at 37 °C for 0.5 h. The glass squares were imaged using an inverted microscope (Axiovert 40CFL, Carl Zeiss). The images were analyzed to count cell numbers using Image J. 2.8. Live/Dead Cell Staining. The released cells were stained with a mixture of calcein AM (1 μM) and ethidium homodimer-1 (1 μM) using the Live/Dead staining kit according to the protocol provided by Invitrogen. Normal cells harvested directly from the cell culture flask were also stained using the same protocol for comparison. The staining buffer was DPBS containing 4.5 g/L glucose and 10 mM MgCl2. The stained cells were imaged under the inverted fluorescence microscope.

Figure 2. Examination of reversible intermolecular hybridization. (A) Digital image of hybridized nucleic acids run on a polyacrylamide gel. Sequences A, B, and C25 are the immobilizing sequence used to functionalize the hydrogel and to capture sequence B, the singlestranded oligonucleotide used to functionalize the antibody, and the biomolecular trigger used to dissociate the AB complex. The subscript 25 indicates the length of sequence C and the number of base pairs of B and C25; the subscript S indicates that C25S is a scrambled sequence C25. (B) Fluorescence imaging of DNA-functionalized hydrogels. Sequence B was labeled with TAMRA for fluorescence imaging. In the lower panel, the hydrogels functionalized with the immobilized AB complex were treated with C25 and C25S.

(Figure 1A). Because the glass surface was silanized to carry methacrylate groups, the sequence A-functionalized hydrogel was chemically conjugated to the glass surface during the polymerization. Thus, the hydrogel was stably immobilized and it would not fall off from the glass during any washing step. The hydrogel was treated with the fluorophore-labeled sequence B and then washed thoroughly. The sequence B-treated hydrogel exhibited much stronger fluorescence than the original sequence A-functionalized hydrogel (Figure 2B). These differences showed that sequence A was successfully incorporated into the polyacrylamide hydrogel during the free radical polymerization, and that sequences A and B could stably hybridize in the hydrogel. In addition to the examination of the AB hybridization, we studied whether the AB complex immobilized in the hydrogel was responsive to the stimulation of the triggering complementary oligonucleotide (i.e., sequence C25). Consistent with the result of gel electrophoresis, the treatment of the hydrogel with C25 led to the disappearance of fluorophore from the hydrogel, whereas the control group maintained strong fluorescence. As sequence B was labeled with the fluorophore, the result showed that C25 triggered the dissociation of sequence B from the hydrogel. Because sequence B was used to functionalize the antibody and to form the DNA−antibody chimera, this result indicated that C25 would have the capability to trigger the dissociation of the chimera from the hydrogel for cell release. 3.2. Examination of Chimera-Mediated Cell Attachment. The antibody model used in this study was an anti-IgM antibody and the cell model was Ramos (human B lymphocyte cell line). Consistent with previous studies showing that Ramos expressed a large amount of IgM on its surface,27,28 the flow cytometry analysis demonstrated that the antibody treatment induced a significant shift of the histogram (Figure 3A). In contrast, CCRF-CEM had minimal expression of IgM on its surface (Figure 3A). It showed that CCRF-CEM was a good control cell line to evaluate the specificity of cell binding.

3. RESULTS AND DISCUSSION 3.1. Examination of DNA-Mediated Reversible Intermolecular Hybridization. The material system used in this study involved a total of three oligonucleotides including sequences A, B, and C. Sequence A played the role of functionalizing the hydrogel and immobilizing sequence B; sequence B was used to functionalize the antibody to form the DNA−antibody chimera; sequence C was used as a biomolecular trigger to compete against sequence A from the AB complex and to form a new complex BC with sequence B. The intermolecular hybridization was first studied in aqueous solutions and characterized by gel electrophoresis (Figure 2A). The result showed that sequences A and B formed a stable AB complex, whereas sequences A and BS (i.e., scrambled sequence B) did not. A similar result was observed in the group of B and C. It demonstrates that intermolecular hybridization is sequence-specific. More importantly, the addition of C25 into the solution of A and B led to the formation of a BC25 band and the disappearance of the AB band. This result showed that C25 was able to trigger the dissociation of the AB complex and to form the BC25 complex. After the demonstration of intermolecular hybridization in aqueous solutions, we further studied intermolecular hybridization in the hydrogels. Sequence A was initially functionalized with acrydite at its 5′ end (Table 1). Thus, sequence A, acrylamide, and bis-acrylamide formed the DNA-functionalized hydrogel via free radical polymerization on the glass substrate 1176

dx.doi.org/10.1021/bm400096z | Biomacromolecules 2013, 14, 1174−1180

Biomacromolecules

Article

DNA−antibody chimera mediated specific cell attachment to the functionalized hydrogel. After the illustration of the chimera-mediated specific cell attachment, the effect of time on cell attachment was studied to determine the kinetics of cell catch on the hydrogel. Figure 3D shows a typical biphasic profile of cell attachment versus time. The cell density on the hydrogel rapidly increased within half an hour and gradually reached a plateau in the next half an hour. This result demonstrated that the first half an hour was critical to specific cell attachment. 3.3. Determination of Triggering DNA-Mediated Cell Release. Antibodies have been widely used to functionalize various materials to achieve specific cell recognition through antibody−antigen interactions.30−32 Although antibody-based molecular recognition is beneficial for initial cell catch, the strong antibody−antigen binding is difficult to break for subsequent cell release. Particularly, when antibodies are immobilized on the surface of a substrate, it is highly possible that multiple antibodies will simultaneously interact with multiple cell receptors. Thus, the cell−material interactions would be much stronger than individual antibody−receptor interactions. Harsh conditions such as high shear stress or protease treatment may be applied to break these strong interactions and to facilitate cell release from antibodyfunctionalized materials.33−35 However, these conditions may result in significant decrease of cell viability.36 In addition, it is difficult to use these approaches in certain conditions. For instance, shear stress plays an important role on the surface but it may fail to produce the same effect inside the hydrogels. Thus, the cell binding state inside the hydrogels may not be affected by shear stress. In contrast, two stably hybridized nucleic acid sequences can be easily dissociated by introducing a third complementary oligonucleotide without involving any harsh factor (Figure.2). Thus, antibody−DNA chimeras possess two major advantages of antibodies and nucleic acids: strong cell binding capability and DNA-mediated reversibility. After the reversible binding, multivalent cell−material interactions become individual antibody−receptor interactions, which will facilitate the dissociation of antibodies from the cell surface. In addition, the triggering oligonucleotide can virtually reach any location of a hydrogel to induce reversible cell release from the hydrogel. Antibody−DNA chimeras have been used in the areas of nanotechnology and biosensing.37 For instance, single- or double-stranded DNA molecules have also been conjugated to antibodies to develop ultrasensitive immune-PCR for detecting proteins.38 However, no study has been carried out to demonstrate the feasibility of using the DNA−antibody chimeras to functionalize hydrogels for reversible cell catch and release. This study was aimed at coupling DNAfunctionalized hydrogels and DNA-functionalized antibodies to develop a DNA-responsive system for programmable display of DNA−antibody chimeras, which would be useful for dynamic regulation of cell catch and release. A total of four triggering oligonucleotides were studied to trigger cell release. The gel electrophoresis image (Figure 4A) clearly exhibited the bands of BC15, BC20, BC25, and BC30, showing that all of these oligonucleotides could form stable complementary base pairs with sequence B. However, the image also showed that the AB band barely changed in the presence of sequences C15 or C20, indicating that the AB complex was stable after the addition of sequences C15 or C20 into the system. In contrast, in the group of C25 or C30, the

Figure 3. Determination of chimera-mediated cell attachment. (A) Flow cytometry analysis of labeled Ramos and CCRF-CEM cells: Ab, biotinylated goat antihuman IgM antibody; C-Ab, biotinylated isotype antibody; FITC-SA, FITC-conjugated streptavidin. (B) Images of Ramos cell attachment on hydrogels including the native hydrogel (a), the sequence A-functionalized hydrogel (b), the chimera-treated native hydrogel (c), the control chimera-functionalized hydrogel (d), and the chimera-functionalized hydrogel (e). The attachment of CCRF-CEM cells to the chimera-functionalized hydrogel (f) was used as a control. Scale bar = 20 μm. (C) Density of attached cells on different hydrogels. The cell numbers were quantified using ImageJ. (D) The kinetics of the chimera-mediated Ramos cell attachment.

The antibody−DNA chimeras were prepared by using streptavidin to link a biotinylated antibody and a biotinylated single-stranded nucleic acid sequence (i.e., sequence B). The chimeras were immobilized to the hydrogel through the hybridization of sequences A and B. After the immobilization of the chimeras, the hydrogels were thoroughly washed to remove free antibodies. Because the binding dissociation constant of streptavidin and biotin is approximately 10−14 M,29 the dissociation of the chimeras within a short period of time is negligible. Thus, the cells interacted with the immobilized chimeras when they were incubated with the hydrogels. To demonstrate the specificity of chimera-mediated cell binding, we prepared a total of five hydrogel samples including the native hydrogel (group a), the sequence A-functionalized hydrogel (group b), the chimera-treated native hydrogel (group c), the control chimera-functionalized hydrogel (group d), and the chimera-functionalized hydrogel (group e). After these hydrogels were incubated in the cell suspension for half an hour, they were gently washed to remove unbound cells. The results of microscopy imaging showed that the functional chimera was able to induce the attachment of Ramos cells to the hydrogel (Figure 3B,C). In contrast, a very small number of Ramos cells were observed on the control hydrogels (Figure 3B,C). The specificity of cell binding was also studied by using the control cell line, CCRF-CEM. A very small number of CCRF-CEM cells were observed on the chimera-functionalized hydrogel (Figure 3B,C). These results demonstrated that the 1177

dx.doi.org/10.1021/bm400096z | Biomacromolecules 2013, 14, 1174−1180

Biomacromolecules

Article

Figure 5. (A) Comparison of C25 and C25S in releasing the attached cells. Both cell catch and release lasted 0.5 h. Sequence C25S is a scrambled sequence of C25. Scale bar = 20 μm. (B) Live/Dead cell staining. Live and dead cells are indicated by green and red, respectively. The dead cells are also pointed by the red arrows. Cultured cells: cells directly harvested from the flask. Scale bar = 20 μm. (C) The kinetics of the triggering DNA-mediated cell release.

Figure 4. Determination of DNA-programmed cell release. (A) Digital image of hybridized nucleic acids run on a polyacrylamide gel. Sequences A, B, and C are the immobilizing sequence, the oligonucleotide used to functionalize the antibody, and the biomolecular trigger. The subscript indicates the length of sequence C and the number of base pairs of sequences B and C. (B) Images of Ramos cells attached to the hydrogels before and after the treatment of sequence C. Scale bar = 20 μm. (C) Quantification of the density of the cells before and after cell release.

ethidium homodimer-1 was only ∼1% (Figure 5B). It was the same as the staining of the cultured cells. This result showed that the triggering DNA-mediated release was not only specific, but also nondestructive. We also carried out an experiment to study the effect of treatment time on cell release. The kinetics of cell release showed that more than 95% of the captured cells were released within 10 min (Figure 5C). Therefore, these data showed that the attached Ramos cells could be nondestructively and rapidly released from the chimera-functionalized hydrogel through reversible intermolecular hybridization in physiological conditions. The future work will test cell release in a microfluidic system with which real-time cell release can be examined. 3.4. Examination of the Feasibility of Repeating Cell Catch and Release. Because the whole procedure of cell catch and release was carried out in physiological conditions without involving any destructive factor, the DNA-functionalized hydrogel was expected to recover the capability to catch cells after cell release. To evaluate this potential, a two-round experiment of cell catch and release was carried out. The entire procedure of hydrogel functionalization as well as cell catch and release in each round maintained the same. The used triggering oligonucleotide was sequence C25. As shown in the microscopic images (Figure 6), the density of attached cells was virtually the same in the two rounds of cell catch. In addition, the treatment of the hydrogels with sequence C25 led to the successful release of a majority of the attached cells. However, the result also showed that the efficiency of cell release exhibited a difference in these two rounds of cell release. The percentage of the first round cell release (∼97%) was higher than that of the second round cell release (∼86%). The exact reason for this difference is not clear. One possible reason is that the whole procedure for the functionalization of the hydrogel involved multiple steps because of the usage of streptavidin as the linker to synthesize the DNA−antibody chimera. Thus, future work will test the chimeras that are synthesized by the direct conjugation of DNA

presence of sequences C25 or C30 led to the disappearance of the AB band and the occurrence of the BC25 or BC30 band. This result indicated that both C25 and C30 were able to trigger the dissociation of the stable AB complex. The difference among these four sequences in triggering the dissociation of the AB complex showed that the triggering oligonucleotide needed to have a critical length. It is reasonable since the stability of DNA duplexes is a function of the number of base pairs. The AB complex had 20 base pairs. The numbers of base pairs formed between B and C15, C20, C25, and C30 were 15, 20, 25, and 30 base pairs, respectively. The latter two triggering oligonucleotides formed more base pairs with sequence B than the immobilizing sequence A. Therefore, these two sequences were capable of competing against sequence A, whereas the former two were not. Consistent with the data of gel electrophoresis, the attached cells were successfully released in the presence of sequences C25 and C30 (Figure 4B). In contrast, the cells were barely released when C15 and C20 were used to treat the hydrogels (Figure 4B). Therefore, these observations showed that the triggering oligonucleotides needed to form more base pairs with the chimeras than the immobilizing DNA sequences to release the captured cells effectively. We also pursued another three sets of experiments to further examine triggering DNA-mediated cell release. In one experiment, the scrambled sequence C25 (i.e., C25S) was applied to treat the hydrogel after the cell attachment. In contrast to C25, C25S did not induce significant cell release from the chimerafunctionalized hydrogel though both C25 and C25S had the same length and composition (Figure 5A). This difference showed that the DNA-mediated cell release was sequence-specific. In another experiment, the released cells were stained with the mixture of calcein AM and ethidium homodimer-1 to determine the numbers of live and dead cells. The microscopy imaging showed that the percentage of dead cells stained by 1178

dx.doi.org/10.1021/bm400096z | Biomacromolecules 2013, 14, 1174−1180

Biomacromolecules

Article ∥

Department of Bioengineering, Pennsylvania State University, University Park, PA 16802−6804, United States. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The U.S. National Science Foundation (DMR-1241461/ 1322332) and the CT Department of Public Health (09SCAUCON02) are greatly acknowledged.

■

Figure 6. Examination of two-round cell catch and release. (A) Representative images of the hydrogels before and after cell release. Scale bar = 20 μm. (B) Quantitative analysis of cell density using Image J. Sequence C25 was used to trigger the cell release. The whole procedure of hydrogel functionalization, cell seeding, and C 25 treatment in each round was the same.

to the antibody and test the concept with multiple rounds of cell catch and release. The capability to repeat the catch and release of the DNA− antibody chimeras and the cells would potentially benefit numerous applications. For instance, it is promising to integrate this biomolecular system into microfluidic channels for the development of regenerable and reusable biomedical devices in separating circulating rare tumor cells from a human sample. Because of the potential of reusing those devices, it is possible to develop a platform that can automatically analyze a large number of samples without the need to change the microfluidic component manually. Meanwhile, it is important to realize that the immobilizing DNA may be gradually degraded due to the multiple contact with biological fluids and cells. Such a potential problem may be solved by using chemically modified nucleic acids or nucleic acid analogues that can resist nuclease degradation. It is believed that the use of robust nucleic acids or nucleic acid analogues will not only benefit in in vitro applications (e.g., microfluidic cell separation), but also in vivo applications (e.g., regenerative medicine).

4. CONCLUSIONS The DNA−antibody chimera was applied to functionalize hydrogels and to investigate cell−hydrogel interactions in the presence of triggering complementary oligonucleotides. The results have successfully demonstrated that the DNA−antibody chimera is capable of inducing specific cell attachment to the hydrogel, and that the triggering oligonucleotide can trigger the rapid release of most attached cells from the hydrogel through reversible intermolecular hybridization. In addition, the results indicate that it is promising to repeat the procedure of cell catch and release using the same hydrogel. Because reversible DNA hybridization is performed in physiological conditions without involving destructive factors, the released cells can maintain high viability. Therefore, the programmable display of DNA− protein chimeras holds great potential for the dynamic regulation of cell−hydrogel interactions.

■

REFERENCES

(1) Holmes, T. C. Trends Biotechnol. 2002, 20, 16−21. (2) Shin, H.; Jo, S.; Mikos, A. G. Biomaterials 2003, 24, 4353−64. (3) Bacáková, L.; Filová, E.; Rypácek, F.; Svorcík, V.; Starý, V. Physiol. Res. 2004, 53 (Suppl 1), S35−45. (4) Khademhosseini, A.; Langer, R.; Borenstein, J.; Vacanti, J. P. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 2480−7. (5) Farokhzad, O. C.; Jon, S.; Khademhosseini, A.; Tran, T.-N. T.; Lavan, D. A.; Langer, R. Cancer Res. 2004, 64, 7668−72. (6) Safarík, I.; Safaríková, M. J. Chromatogr., B: Biomed. Sci. Appl. 1999, 722, 33−53. (7) Ito, Y. Soft Matter 2008, 4, 46−56. (8) Hersel, U.; Dahmen, C.; Kessler, H. Biomaterials 2003, 24, 4385− 415. (9) Turková, J. J. Chromatogr., B: Biomed. Sci. Appl. 1999, 722, 11− 31. (10) Cole, M. A.; Voelcker, N. H.; Thissen, H.; Griesser, H. J. Biomaterials 2009, 30, 1827−50. (11) Edahiro, J.-I.; Sumaru, K.; Tada, Y.; Ohi, K.; Takagi, T.; Kameda, M.; Shinbo, T.; Kanamori, T.; Yoshimi, Y. Biomacromolecules 2005, 6, 970−4. (12) Luo, Y.; Shoichet, M. S. Nat. Mater. 2004, 3, 249−53. (13) Kloxin, A. M.; Kasko, A. M.; Salinas, C. N.; Anseth, K. S. Science 2009, 324, 59−63. (14) Jiang, X.; Ferrigno, R.; Mrksich, M.; Whitesides, G. M. J. Am. Chem. Soc. 2003, 125, 2366−7. (15) Auernheimer, J.; Dahmen, C.; Hersel, U.; Bausch, A.; Kessler, H. J. Am. Chem. Soc. 2005, 127, 16107−10. (16) Persson, K. M.; Karlsson, R.; Svennersten, K.; Löffler, S.; Jager, E. W. H.; Richter-Dahlfors, A.; Konradsson, P.; Berggren, M. Adv. Mater. 2011, 23, 4403−8. (17) Da Silva, R. M. P.; Mano, J. F.; Reis, R. L. Trends Biotechnol. 2007, 25, 577−83. (18) Dainiak, M. B.; Kumar, A.; Galaev, I. Y.; Mattiasson, B. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 849−54. (19) Beyer, S.; Simmel, F. C. Nucleic Acids Res. 2006, 34, 1581−7. (20) Soontornworajit, B.; Zhou, J.; Snipes, M. P.; Battig, M. R.; Wang, Y. Biomaterials 2011, 32, 6839−49. (21) Battig, M. R.; Soontornworajit, B.; Wang, Y. J. Am. Chem. Soc. 2012, 134, 12410−3. (22) Zhang, Z.; Chen, N.; Li, S.; Battig, M. R.; Wang, Y. J. Am. Chem. Soc. 2012, 134, 15716−9. (23) Wan, Y.; Liu, Y.; Allen, P. B.; Asghar, W.; Mahmood, M. A. I.; Tan, J.; Duhon, H.; Kim, Y.-T.; Ellington, A. D.; Iqbal, S. M. Lab Chip 2012, 12, 4693−701. (24) Birch, J. R.; Racher, A. J. Adv. Drug Delivery Rev. 2006, 58, 671− 85. (25) Wurm, F. M. Nat. Biotechnol. 2004, 22, 1393−8. (26) Bray, B. L. Nat. Rev. Drug Discov. 2003, 2, 587−93. (27) Cragg, M. S.; Zhang, L.; French, R. R.; Glennie, M. J. Br. J. Cancer 1999, 79, 850−7. (28) Ollila, J.; Vihinen, M. Genes Immun. 2003, 4, 343−50. (29) Green, N. M. Adv. Protein Chem. 1975, 29, 85−133. (30) McDevitt, M. R.; Chattopadhyay, D.; Kappel, B. J.; Jaggi, J. S.; Schiffman, S. R.; Antczak, C.; Njardarson, J. T.; Brentjens, R.; Scheinberg, D. A. J. Nucl. Med. 2007, 48, 1180−9. (31) De la Rica, R.; Matsui, H. Chem. Soc. Rev. 2010, 39, 3499−509.

AUTHOR INFORMATION

Corresponding Author

*Tel.: 814-865-6867. Fax: 814-863-0490. E-mail: yxwbio@engr. psu.edu. Present Addresses †

National Institutes For Food and Drug Control, Beijing 100050, China. 1179

dx.doi.org/10.1021/bm400096z | Biomacromolecules 2013, 14, 1174−1180

Biomacromolecules

Article

(32) Arruebo, M.; Valladares, M.; González-Fernández, Á . J. Nanomater. 2009, 2009, 1−24. (33) Kawahara, D.; Matsuda, T. J. Biomed. Mater. Res., Part B 2012, 100, 1218−28. (34) Murthy, S. K.; Sin, A.; Tompkins, R. G.; Toner, M. Langmuir 2004, 20, 11649−55. (35) Huttenlocher, A.; Palecek, S. P.; Lu, Q.; Zhang, W.; Mellgren, R. L.; Lauffenburger, D. A; Ginsberg, M. H.; Horwitz, A. F. J. Biol. Chem. 1997, 272, 32719−22. (36) Fujioka, N.; Morimoto, Y.; Takeuchi, K.; Yoshioka, M.; Kikuchi, M. Appl. Spectrosc. 2003, 57, 241−3. (37) Niemeyer, C. M.; Adler, M.; Pignataro, B.; Lenhert, S.; Gao, S.; Chi, L.; Fuchs, H.; Blohm, D. Nucleic Acids Res. 1999, 27, 4553−61. (38) Sano, T.; Smith, C. L.; Cantor, C. R. Science 1992, 258, 120−2.

1180

dx.doi.org/10.1021/bm400096z | Biomacromolecules 2013, 14, 1174−1180