Biomacromolecules 2005, 6, 2927-2929
2927
DNA-Responsive Hydrogels That Can Shrink or Swell Yoshihiko Murakami† and Mizuo Maeda* Bioengineering Laboratory, RIKEN (The Institute of Physical and Chemical Research), 2-1 Hirosawa, Wako-Shi, Saitama 351-0198, Japan Received June 22, 2005; Revised Manuscript Received September 29, 2005
Molecule-responsive hydrogels are reputed to be smart materials because of their unique properties. We recently reported that hydrogels containing directly grafted single-stranded (ss) DNA or ssDNApolyacrylamide conjugate in a semi-interpenetrating network (semi-IPN) manner that “only shrunk” by the addition of ssDNA samples. To date, however, no DNA-responsive hydrogels have been reported capable of “swelling” in response to specific DNAs. Smart materials capable of both shrinking and swelling in response to specific DNAs would be very useful in biochemical and biomedical applications. Here, we show a novel “shrinking or swelling” DNA-responsive mechanism. Novel hybrid hydrogels containing rationally designed ssDNA as the cross-linker were capable of shrinking or swelling in response to ssDNA samples and recognizing a single base difference in the samples. On the basis of the results presented in this paper, it is envisioned that these novel hybrid hydrogels could function and have potential in applications such as DNA-sensing devices and DNA-triggered actuators. A variety of molecule-responsive hydrogels have been reported to date.1-5 For example, a hydrogel containing N,Ndiethylaminoethyl methacrylate, 2-hydroxypropyl methacrylate, and polyacrylamide responds to the pH change induced by gluconic acid produced during a glucose oxidasecatalyzed enzymatic reaction.1 A thermosensitive poly(Nisopropyl acrylamide) hydrogel containing an antibody Fab′ fragment derived from a monoclonal anti-fluorescein BDC1 antibody (IgG2a) responds to the change in hydrophobicity that is induced by the addition of an antigen (for example, free fluorescein).2 Furthermore, the binding between an antigen and an antibody can introduce irreversible3 or reversible4 cross-links in the hydrogel. Since the competitive binding of the free antigen with the immobilized antibody results in a decrease in the number of cross-links, the addition of the free antigen induces a change in the swelling ratio of a hydrogel containing antigen-antibody bindings. We recently developed DNA-responsive biomaterials, i.e., polyacrylamide (polyAAm) hydrogels containing directly grafted single-stranded (ss) DNA or ssDNA-polyAAm conjugate in a semi-interpenetrating network (semi-IPN) manner that only “shrunk” by the addition of ssDNA samples.6 However, to date, no DNA-responsive hydrogels have been reported capable of “swelling” in response to specific DNAs. This is presumably due to the difficulty in incorporating a DNA-responsive mechanism into the hydrogels. Smart materials capable of both shrinking and swelling in response to specific DNAs would be very useful in biochemical and biomedical applications. * To whom correspondence should be addressed. Phone: +81 48-4679312. Fax: +81 48-462-4658. E-mail:
[email protected]. † Present address: Yokoyama Project, Kanagawa Academy of Science and Technology (KAST), KSP East 404, Sakado 3-2-1, Takatsu, Kawasaki, Kanagawa 213-0012, Japan.
Figure 1. The response of novel hybrid hydrogels containing ssDNA as a cross-linker to ssDNA. Hydrogel containing the ssDNA with a stem-loop structure like a “molecular beacon” molecule is expected to swell (top), whereas that containing the ssDNA without an intramolecular base pair is expected to shrink (bottom) on the basis of the binding with its complementary ssDNA samples.
Here, we show a novel “shrinking or swelling” DNAresponsive mechanism. Smart hydrogels that contain rationally designed ssDNA as the cross-linker can shrink or swell in response to ssDNA samples. We adopted a novel strategy that covalently incorporated a rationally designed ssDNA probe (complementary to the ssDNA sample) into the hydrogels as a cross-linker. This enabled these DNAconjugated hydrogels to “shrink or swell” in response to ssDNA samples (Figure 1). Three types of ssDNA samples were used for testing the ssDNA responsiveness of the hydrogels containing P1, P2, and P3 as cross-linkers: (i) samples that were complementary to the ssDNA probe in the hydrogel (S1, S4, S5, and S8), (ii) samples that had a single base mismatch with the ssDNA probe in the hydrogel (S2 and S6), and (iii) samples that were uncomplementary
10.1021/bm0504330 CCC: $30.25 © 2005 American Chemical Society Published on Web 10/22/2005
2928
Biomacromolecules, Vol. 6, No. 6, 2005
Communications
Table 1: ssDNA Probes and Samplesa
a
entry
base
sequence
P1 P2 P3 S1 S2 S3 S4 S5 S6 S7 S8
29 29 29 18 18 18 29 18 18 18 29
NH2-(CH2)6-5′-CTTGTGCCACCAGCTCCAACTACCACAAG-3′ -CH2C(CH2OH)-(CH2)4-NH2 NH2-(CH2)6-5′-AAAAAAAAAAAAAAAAAAAAAAAAAAAAA-3′ -CH2C(CH2OH)-(CH2)4-NH2 NH2-(CH2)6-5′-TTTTTTTTTTTTTTTTTTTTTTTTTTTTT-3′ -CH2C(CH2OH)-(CH2)4-NH2 5′-GTAGTTGGAGCTGGTGGC-3′ 5′-GTAGTTGGAGCTGATGGC-3′ 5′-AGGATGGCTGGTGTCTGG-3′ 5′-CTTGTGGTAGTTGGAGCTGGTGGCACAAG-3′ 5′-TTTTTTTTTTTTTTTTTT-3 5′-TTTTTTTTTTTTTCTTTT-3′ 5′-CCCCCCCCCCCCCCCCCC-3′ 5′-TTTTTTTTTTTTTTTTTTTTTTTTTTTTT-3
Italics indicate one-base mismatch.
to the ssDNA probe in the hydrogel (S3 and S7). P1 and P2 were designed using a computational algorithm RNAstructure7 that has been widely used to design “molecular beacon” molecules. The hydrogel containing P3 was incapable of forming a duplex with any of the samples and was therefore used in the control experiments. The sequences of the ssDNA samples (S1-S8) and probes (P1-P3) are summarized in Table 1. The hydrogels can be synthesized by standard radical polymerization which has been widely used for polyAAm hydrogels.3,4 Methacryloyloxy succinimide was synthesized from methacrylic acid and N-hydroxy succinimide, and subsequently purified. 5′- and 3′-Methacryloyl-modified ssDNA was obtained by mixing methacryloyloxy succinimide (32.5 µmol) and commercially available 5′- and 3′aminoalkyl-modified ssDNAs (P1, P2, or P3 in Table 1, 0.65 µmol, SIGMA) in 50 mM Na2CO3/NaHCO3 buffer solution at pH 9.0 for 12 h. A hydrogel was synthesized by polymerizing acrylamide (1.40 M) with 5′- and 3′-methacryloyl-modified ssDNA (2.80 mM) as the cross-linker in Tris-HCl buffer solution (20 mM, pH 7.4) containing APS (8 mM) and TEMED (80 mM). The synthesis was carried out in a Durham tube (i.d. ) 3 mm) at 25 °C for 24 h. The hydrogels were immersed in water (the dialysate was changed every 12 h) until no residual chemicals and unreacted monomers were detected by high-performance liquid chromatography (HPLC). Furthermore, during the immersion of hydrogels, no ssDNA probes were detected in the dialysate water using HPLC. All cross-linkable ssDNA probes were therefore expected to be immobilized in the hydrogels. The hydrogels were sliced into sections of length as 0.5 mm. The response of the hydrogels to ssDNA was determined by the change in their shrinking or swelling ratio induced by adding a ssDNA sample (the concentration of this was 1 or 10 equiv of the ssDNA probe in the hydrogel) to the 100 mM MgCl2/10 mM Tris-HCl buffer solution (pH 7.4) in which each hydrogel was immersed. MgCl2 was used because it was known to stabilize the DNA duplex.8 The hydrogel was immersed at least 12 h before adding the ssDNA sample to confirm that its volume was constant. All experiments were performed in triplicate at 20 °C. The digital images of hydrogels obtained using an optical microscope were quantitatively analyzed by graphic software. The values of d and d0 were accurately determined by magnifying the images with 4 significant figures. The swelling ratio of the hydrogels,
Figure 2. Typical photograph of response of hydrogel obtained using an optical microscope (a) before and (b) after adding S1 to the hydrogel containing P1 as a cross-linker.
(d/d0)3, was determined from their diameter ratio, d/d0, where d and d0 were the diameters of hydrogels with and without the ssDNA sample, respectively. The ssDNA responsibility of the hydrogel had reproducibility within 10% error. It was confirmed that 8% and 28% of ssDNA probes in the hydrogel were collapsed and detected in the dialysate water during the immersion in water at 80 and 90 °C for a day, respectively. It was presumably because the hydrolysis of the acrylamide chain was enhanced. However, we can conclude that the hydrogel is stable in its practical use, since it is structurally stable below 70 °C. Typical response of the hydrogel to ssDNA is shown in the photograph in Figure 2. The hydrogel containing P1 as the cross-linker gradually swelled after the addition of S1 until equilibrium was reached. Figure 3 shows the response of hydrogels to ssDNA when the final concentration of the ssDNA sample was 10-fold equivalent to that of the ssDNA probes in the hydrogel. Because of the change in osmotic pressure outside the hydrogel, the control hydrogel containing P3 instead of P1 shrunk slightly when the ssDNA sample was added. The same phenomenon was previously reported when a target antigen was added to a solution in which a polyAAm hydrogel was immersed.4,5 As shown in Figure 3, the hydrogel containing the ssDNA with a stem-loop structure similar to that of a “molecular beacon” molecule (P1, i.e., a short oligonucleotide sequence that forms an intermolecular base pair at the 3′ and 5′ ends of the loop structure) swelled. On the other hand, the hydrogel containing the ssDNA without an intermolecular base pair (P2) shrunk, because the chain length of double-stranded (ds) DNA is generally shorter than that of ssDNA.9 The shape of the hydrogels did not affect their responsiveness (data not shown).
Communications
Figure 3. Response of hydrogels containing (a) P1 and (b) P2 as cross-linkers to ssDNAs. The ssDNA samples used were (a) S1 (filled circle), S2 (filled triangles), S3 (filled squares), and S4 (filled reverse triangles) and (b) S5 (filled circles), S6 (filled triangles), S7 (filled square), and S8 (filled reverse triangles). The open circles indicate the case wherein ssDNA ((a) S1 or (b) S5) was added to hydrogels containing P3 as a cross-linker.
The only reason for the shrinking or swelling of the hydrogels was the structural changes in the cross-linker probes P1 and P2, because no ssDNA-responsive structure was incorporated into the two hydrogels with the exception of these cross-linkers. Furthermore, these hydrogels showed the following interesting ssDNA-sensitive properties (Figure 3): (i) their shrinking or swelling ratio significantly changed in response to the ssDNA samples that were complementary to the ssDNA probe in the hydrogels and (ii) these hydrogels recognized the single base difference in the ssDNA sample and did not respond to the samples that were uncomplementary to the ssDNA probe in the hydrogels. These results indicate that the douplex formation between the ssDNA sample and the ssDNA probe was a dominant factor governing the change in the shrinking or swelling ratio of the hydrogels. Furthermore, the response time was reduced with longer ssDNA samples (S4 and S8), presumably because the formation of a longer duplex was effective in changing the probe structures in the hydrogels. Since DNA duplex formation leads to the uptake of counterions into the hydrogels, they may swell in response to ssDNA samples. On the other hand, as previously reported by us, the formation of a DNA duplex inside the hydrogels containing directly grafted single-stranded (ss) DNA or ssDNA-polyAAm conjugate in a semi-IPN manner leads only to its shrinkage (not swelling). This is presumably because the dsDNA was dehydrated following the addition of the ssDNA target, and consequently, water diffused from the inside to the outside of the hydrogels.6 Interestingly, however, the novel hydrogels proposed in the present paper can shrink or swell in response to ssDNA targets (Figure 3), as intended by rational designing of the probes. These results strongly suggest that the structural change on cross-
Biomacromolecules, Vol. 6, No. 6, 2005 2929
linking primarily governed the shrinking or swelling behavior of the hydrogels, even though there were many other factors that led to their shrinking or swelling. In comparison with stimuli-responsive hydrogels, the response of molecule-sensitive hydrogels in reaching their equilibrium is generally slower, for example, 1-10 h2-5 (in this study, it was approximately 4 h when longer samples, S4 and S8, were used). This is presumably due to the difficulty in recognizing and subsequently changing the structure in the hydrogels when the probes are tightly immobilized as cross-linkers. The change in the shrinking or swelling ratio of both the hybrid hydrogels at equilibrium was approximately 20% as shown in Figure 3. This was of the same (or higher) order of magnitude as the changes previously reported in the case of molecule-responsive hydrogels (7-30%).2-5 These properties show the usefulness of our hydrogels, similar to other molecular-responsive hydrogels. The response rate of the hydrogels reduced to one-third of the original rate when the concentration of the ssDNA sample was decreased to onefold equivalent that of the ssDNA probes in the hydrogel (data not shown). A possible strategy for further enhancing the responsiveness of the hydrogels is as follows: (i) utilizing a porous hydrogel structure developed by freeze-drying10 in order to enhance the diffusion of the ssDNA sample, (ii) decreasing the number of cross-links in the hydrogel to achieve high flexibility in its conformation, or (iii) increasing the length of probes to sensitively detect specific ssDNAs. The novel hydrogels presented in this study possess several interesting features. Unlike stimuli-responsive hydrogels, these hybrid hydrogels retain the advantage of using cross-linkable ssDNAs, the conformational properties of which are well-characterized. Since only the target-specific probes form cross-links, we can rationally design hydrogels that show the desired responsiveness. Furthermore, two or more of these cross-linkable DNAs could be used simultaneously, resulting in hydrogels capable of individual stepwise transitions. The transitions are triggered at a different temperature, which is difficult to achieve using polyAAm hydrogels alone. Therefore, the results presented in this paper show that these DNA-conjugated hydrogels are biomaterials that could have potential in applications such as DNAsensing devices and DNA-triggered actuators. References and Notes (1) Ishihara, K.; Kobayashi, M.; Ishimaru, N.; Shinohara, I. Polymer J. 1984, 16, 625-631. (2) Lu, Z.-R.; Kopeckova, P.; Kopecek, J. Macromol. Biosci. 2003, 3, 296-300. (3) Miyata, T.; Asami, N.; Uragami, T. Macromolecules 1999, 32, 20822084. (4) Miyata, T.; Asami, N.; Uragami, T. Nature (London) 1999, 399, 766769. (5) Miyata, T.; Jikihara, A.; Nakamae, K.; Hoffmann, A. S. Macromol. Chem. Phys. 1996, 197, 1135-1146. (6) Murakami, Y.; Maeda, M. Macromolecules 2005, 38, 1535-1537. (7) Mathews, D. H.; Sabina, J.; Zuker, M.; Turner, D. H. J. Mol. Biol. 1999, 288, 911-940. (8) Krakauer, H.; Biochemistry 1974, 13, 2579-2589. (9) Manning, G. S.; Q. ReV. Biophys. 1978, 2, 179-246. (10) Kato, N.; Sakai, Y.; Shibata, S. Macromolecules 2003, 36, 961-963.
BM0504330
