Langmuir 2004, 20, 5891-5896
5891
Formation of a DNA Layer on Langmuir-Blodgett Films and Its Enzymatic Digestion Anusarka Bhaumik,† Madhugiri Ramakanth,† Loveleen Kaur Brar,‡ Arup K. Raychaudhuri,‡ Francis Rondelez,§ and Dipankar Chatterji*,† Department of Physics and Molecular Biophysics Unit, Indian Institute of Science, Bangalore-560012, India, and Laboratoire Physico Chimie Curie (PCC), UMR CNRS 168, Institut Curie-Section de Recherche, 11, rue Pierre et Marie Curie, 75231 Paris Cedex 05, France Received March 8, 2004 Here, we report a system we have developed where long double-stranded DNAs (dsDNAs) are immobilized on a monolayer of Zn-arachidate. We have applied the Langmuir-Blodgett technique to form the monolayer of Zn-arachidate where Zn(II) is bound to arachidic acid through charge neutralization. Because tetrahedral Zn(II) participates in DNA recognition through coordination, we have been able to layer DNA over the Zn-arachidate monolayer. The DNA layer shows a typical compression and expansion cycle in a concentration-dependent fashion. Interestingly, the DNA monolayer is available for enzymatic degradation by DNaseI. The detection of DNA and its accessibility towards biological reaction is demonstrated by imaging through fluorescence microscopy. The conformation of the DNA, immobilized on the monolayer, was studied with the help of atomic force microscopy (AFM). We observed that the dsDNAs were aligned in a stretched manner on the surface. To investigate further, we also demonstrate here that the small single-stranded DNA (ssDNA) immobilized on the air-water interface can act as a target molecule for the complementary ssDNA present in the subphase. The study of DNA hybridization done with the help of fluorescence spectroscopy clearly supports the AFM characterization.
Introduction DNA immobilized on a solid support has profound biological and technological advantages starting from DNA chips to single-molecule biology. It is an emerging field with a wide range of medical, pharmaceutical, and diagnostic applications.1,2 It is also important in a wide range of research areas including work on nanoparticles,3,4 DNA computing,5-7 and DNA chip technologies.4-6 Some of the common supports for the immobilization of DNA are latex particles,8-13 the glass surface, the silanized mica surface,14 gold nanoparticles,15 polystyrene microspheres,9 * To whom correspondence should be addressed. Tel.: +91-80-2293-2836. Fax: +91-80-2360-0535. E-mail: dipankar@ mbu.iisc.ernet.in. † Molecular Biophysics Unit, Indian Institute of Science. ‡ Department of Physics, Indian Institute of Science. § Institut Curie-Section de Recherche. (1) Southern, E.; Mir, K.; Shchepinov, M. Nat. Genet. 1999, 21, 5. (2) Fodor, S. P. A. Science 1997, 277, 393. (3) Storhoff, J. J.; Mirkin, C. A. Chem. Rev. 1999, 99, 1849. (4) Fotin, A. V.; Drobyshev, A. L.; Proudnikov, D. Y.; Perov, A. N.; Mirzabekov, A. D. Nucleic Acids Res. 1998, 26, 1515. (5) Wang, L. M.; Liu, Q. H.; Corn, R. M.; Condon, A. E.; Smith, L. M. J. Am. Chem. Soc. 2000, 122, 7435-7440. (6) Liu, Q.; Wang, L.; Frutos, A. G.; Condon, A. E.; Corn, R. M.; Smith, L. M. Nature 2000, 403, 175-179. (7) Vo-Dinh, T.; Cullum, B. Fresenius’ J. Anal. Chem. 2000, 366, 540-551. (8) Kawaguchi, H.; Asai, A.; Ohtsuka, Y.; Watanabe, H.; Wada, T.; Handa, H. Nucleic Acids Res. 1989, 17, 6229. (9) Ghosh, D.; Faure, N.; Kundu, S.; Rondelez, F.; Chatterji, D.; Langmuir 2003, 19, 5830-5837. (10) Arshady, R. Biomaterials 1993, 14, 5. (11) Kremesky, J. N.; Wooters, J. L.; Dougherty, J. P.; Meyers, R. E.; Collins, M.; Brown, E. L. Nuleic Acids Res. 1987, 15, 2891. (12) Miller, C. A.; Patterson, W. L.; Johnson, P. K.; Swartzell, C. T.; Wogoman, F.; Albarella, J. P.; Carrico, R. J. Clin. Microbiol. 1988, 26, 1271. (13) Elaissari, A.; Chevalier, Y.; Ganachaud, F.; Delair, T.; Pichot, C. Langmuir 2000, 16, 1261-1269. (14) Sasou, M.; Sugiyama, S.; Yoshino, T.; Ohtani, T.; Langmuir 2003, 19, 9845-9849.
and Langmuir monolayers.16-19 Critical factors of DNA chips are the density of the surface-immobilized DNA and its accessibility toward biological reactions. Hence, the requirement for reproducible stable surfaces has created increasing interest in designing different modified surfaces to immobilize DNA. Immobilizing DNA on Langmuir monolayers is wellknown, but successful transfer of the monolayer along with the DNA onto a solid substrate, however, to the best of our knowledge is still not achieved. The proximate aim of this work is to immobilize a long DNA on a solid surface and then follow its orientation and conformation in such a manner that, we believe, would have profound implications with respect to its biological function, protein recognition. We demonstrate here the immobilization of long double-stranded DNA (dsDNA) using simple the Langmuir-Blodgett (LB) technique and the transfer of the monolayer along with the DNA onto a solid substrate. The conformation of the DNAs immobilized on the monolayer has an important role in defining an ideal DNA chip. The surface morphology of the films has been studied by intermittent contact (I-C) mode AFM (atomic force microscopy). AFM shows that the DNAs immobilized in this method are properly aligned on the surface preferentially in a stretched way. The successful hybridization experiment of DNA at the air-water interface also gives the quite relevant idea that the DNAs present in the subphase intend to immobilize in a parallel and stretched alignment with the monolayer. We also report the (15) Kumar, A.; Pattarkine, M.; Bhadbade, M.; Mandale, B. A.; Ganesh, N. K.; Datar, S. S.; Dharmadhikari, V. C.; Sastry, M. Adv. Mater. 2001, 13, 341-344. (16) Okahata, O.; Kobayashi, T.; Tanaka, K. Langmuir 1996, 12, 1326. (17) Ijiro, K.; Shimonura, M.; Tanaka, M.; Nakamura, H.; Hasebe, K. Thin Solid Films 1996, 284-285, 780. (18) Shimonura, M.; Nakamura, F.; Ijiro, K.; Takestuna, H.; Tanaka, M.; Nakamura, H.; Hasebe, K. J. Am. Chem Soc. 1997, 119, 2341. (19) Ebara, Y.; Mizutani, K.; Okahata, Y. Langmuir 2000, 16, 2416.
10.1021/la049400g CCC: $27.50 © 2004 American Chemical Society Published on Web 05/20/2004
5892
Langmuir, Vol. 20, No. 14, 2004
experimental evidence in support of the accessibility of the DNA toward biological reactions, which is one of the most important requirements for the DNA chip. Arachidic acid, a long-chain fatty acid, can be used to form a stable monolayer by the Langmuir technique in the presence of a metal ion in the subphase.20,21 We used ZnSO4 solution containing dsDNA as the subphase on which the arachidic acid monolayer was formed. At higher pH, the repulsion between the adjacent ionized acid groups is reduced as a result of the presence of divalent Zn(II) ions in the subphase. Eventually, DNA was grafted on the monolayer through simple coordination interaction with the Zn ion. This monolayer was then transferred by a dipping technique22,23 onto a glass slide which was made hydrophobic by silanization to avoid simple hydrophilichydrophilic interaction between the glass slide and the DNA present in the subphase. The presence of DNA on the slide was checked under a fluorescence microscope. The biological activity was checked by DNaseI treatment of the slide. Materials and Methods Preparation of LB Layers. The LB films were prepared on a LB trough purchased from Nima Technology (U.K.) and placed under a perplex box. Arachidic acid (purity 99%) and ZnSO4 (purity 99%) were purchased from Sigma-Aldrich for use. The monolayers were formed by spreading the arachidic acid solution (1 mg/mL) in chloroform (HPLC grade, Merck) at the surface of the water subphase (water purified with a Milli-Q system, resisitivity 18.2 ΜΩ, autoclaved to make DNase free before using), containing 10-4 M of ZnSO4 and DNA of concentration 0.25 µg/ mL. T7 DNA, prepared in the laboratory, was used for this purpose. Langmuir films were formed and transferred onto hydrophobic glass slides (24 × 60 mm) purchased from Marienfeld (Germany), at a constant surface pressure of 30 mN/m by vertical dipping. The rate of compression was 10 cm2/min, and the dipper speed while transferring the layers was kept at 1 mm/min. The pH of the subphase solution was kept at 7.4, adjusted by using N-(2hydroxyethyl)piperazine-N′- (2-ethanesulfonic acid) (Sigma) buffer (1 mM). The temperature was maintained at 20 ( 0.1 °C. In another approach, DNA was stained with a fluorescent dye and injected in the subphase prior to transfer, as described elsewhere.24 The other control experiments involved transferring Zn-arachidate or only arachidic acid (absence of any metal ion in the subphase) monolayers. Preparation of T7 DNA. For grafting to the monolayer, we have taken a long double-stranded bacteriophage DNA, T7 DNA (39 336 base pairs). T7 DNA was prepared using the protocol adapted from Nierman and Chamberlin.25 The molecular weight of this DNA is 25.962 × 106 Da. Preparation of Glass Slides. The glass slides (24 × 60 × 0.1 mm) were carefully cleaned in a 3:7 (v/v) mixture of 30% H2O2 and 95% H2SO4 for about 2 h. Then it was rinsed with Milli-Q water thoroughly and stored under water. It was then dried and kept vertical in a container with a few drops of hexamethyl disilazane (purchased from Sigma) at 70 °C overnight. After this treatment, as a result of vapor deposition, the slides would turn hydrophobic automatically. Fluorescence Microscopy. The LB films, which were transferred on the slides, were stained with 1X Sybr Green (Sigma-Aldrich) and then observed under a fluorescence microscope (Carl Zeiss, Axioskop 2, 100× oil immersion objective) to (20) Mahnke, J.; Vollhardt, D.; Stockelhuber, W. K.; Meine, K.; Schulze, J. H. Langmuir 1999, 15, 8220-8224. (21) Cha, J.; Park, Y.; Lee, B. K.; Chang, T. Langmuir 1999, 15, 1383-1387. (22) Blodgett, K. B. J. Am. Chem. Soc. 1935, 57, 1007. (23) Blodgett, K. B.; Langmuir, I. Phys. Rev. 1937, 51, 964. (24) Sastry, M.; Ramakrishnan, V.; Pattarkine, M.; Gole, A.; Ganesh, N. K. Langmuir 2000, 16, 9142-9146. (25) Nierman, W. C.; Chamberlin, M. J. J. Biol. Chem. 1979, 254, 7921.
Bhaumik et al. check for the presence of DNA on the slide. Images were captured using an Axiocam (color) camera connected to a computer. AFM Imaging. AFM images were acquired using the thermomicroscope CP-R, with a 100-µm closed-loop scanner in I-C mode. The spring constant of the cantilever was 2N/m, and the scan speed varied from 1 to 5 µm/s. Images were acquired for various scan ranges and for different numbers of LB layers transferred on the glass slides. The final result given here is for four layers transferred onto a slide and imaged. DNA Hybridization. Primers of sequences 5′-GGATGACCGCCCACTGACAGTGGGGATCC-3′ (ssDNA1), 5′-GGATCCCCACTGTCAGTGGGCGGTCATCC-3′ (ssDNA2), and 5′-GGCCGAGGACCAAGCTTACGGCGCAAGCG-3′ (ssDNA3; Microsynth, Switzerland) were used for this purpose. SsDNA1 and ssDNA2 are exactly complementary sequences, whereas ssDNA3 is noncomplemetary to ssDNA1. An arachidic acid monolayer was formed in a manner similar to that mentioned earlier on the subphase containing a 10-4 M ZnSO4 solution and 10-8 M ethidium bromide. ssDNA1 was injected to the subphase, and compression/expansion cycles were run until the isotherms indicated a constant area/molecule value. This condition stabilized 10 h after the spreading, and it is indicative of the completion/equilibration of DNA-Zn-arachidate monolayer complexation. Injection of ssDNA2 and ssDNA3 to the subphase was carried out after this equilibration step. Upon 10 h of further equilibration, 19 monolayers were transferred onto a quartz plate (Applied Optics, Inc., India) in each case. As a control, 19 monolayers of ssDNA1-Zn-arachidate alone were transferred onto a quartz plate without having ssDNA2 or ssDNA3 in the subphase. Fluorescence spectra were then recorded for each case with the help of a spetrofluorimeter (FluoroMax-3, Jobin Yvon Horiba) at 25 °C, with slit widths of 5 nm both for excitation (280 nm) and emission monochromators. Quartz plates containing the DNA samples were kept in the light path, replacing the normal cuvette. DNA Digestion. The glass slides with Zn-arachidate-T7 DNA layers were treated with DNaseI (Bangalore Genei, India) for different time periods and further checked under a fluorescence microscope. The area used for layer transfer was divided into three parts using thin wax lining. The first part was kept untreated, the second part was treated with buffer (50% glycerol, 20 mM Tris-HCl, 1 mM MgCl2, pH 7.4), and the third part was treated with buffer and enzyme (DNaseI). All the three parts were stained with Sybr green separately before the enzymatic digestion and imaged under a microscope. Subsequently, one of the grids was digested with DNaseI and imaged again. Hysteresis Loop. The T7 DNA concentration in the subphase solution was increased up to 1 mg/500 mL, and the compression/ expansion cycles were run maintaining the surface pressure at 30 mN/m and barrier speed at 10 cm2/min.
Results and Discussion Figure 1 demonstrates the π-A isotherm of the Znarachidate monolayer formed on the subphase without DNA (curve a) and also on the subphase containing DNA (curve b), recorded after solvent (chloroform) evaporation within 30 min of spreading. The recorded profile and the resulting pressure in both cases were found to be almost the same as those reported earlier for the arachidic acid monolayer.26 π-A isotherms for the Langmuir layers, when DNA of varying concentrations was added in the subphase, were recorded separately, having the time interval after each spreading as another variable parameter. Figure 2 represents such curves when the DNA concentration was varied within the range of 0.25-2 µg/mL. It can be noticed that the area/molecule decreases as a function of the DNA concentration, which indicates the increasing compactness of the monolayer with the increase in concentration of DNA in the subphase. In Figure 3, the change in area/ molecule was recorded up to 11 h after spreading at a (26) Blodgett, K. B. Phys. Rev. 1939, 55, 391.
DNA on LB Films
Langmuir, Vol. 20, No. 14, 2004 5893
Figure 1. π-A isotherm of Zn-arachidate monolayer recorded 30 min after spreading at pH 7.4 and temperature 20 ( 0.1 °C. Curves a and b are the isotherms of the Zn-arachidate monolayer in the absence and in the presence of DNA in the subphase, respectively. The concentration of DNA in the subphase is 0.25 µg/mL.
Figure 2. Compression/expansion cycles of the Zn-arachidate monolayer recorded at different concentrations of DNA in the subphase. Curve a indicates the π-A isotherm when the subphase DNA concentration was 0.25 µg/mL, and curve b indicates the isotherm when the DNA concentration was 2 µg/ mL.
Figure 3. π-A isotherms at a 1-h time gap after spreading until 11 h. The DNA concentration in all the above cases was 0.25 µg/mL, the temperature 20 ( 0.1 °C, pH 7.4, and barrier speed 10 cm2/min.
regular time interval of 1 h. It was observed that the isotherm cycles after 10 or 11 h remained invariant. The area/molecule at this point was calculated to be 26 Å2 and did not change thereafter. This value when compared with that of the area/molecule value of 20 Å2 for only the Znarachidate monolayer (Figure 1) suggests interaction
Figure 4. (a) Zn-arachidate monolayer transferred on the glass slide and stained. (b) Arachidic acid-DNA monolayer.
between Zn(II) and DNA on the monolayer. It probably indicates the overall steric repulsion between the adjacent surface-bound Zn(II)-DNA complexes on the monolayer resulting in an increase in the area/molecule value. We would also like to point out here that we have observed very thin compression/expansion cycles or a hysteresis loop in the presence of DNA, indicative of the stability of the films (Figure 3). Figure 4a shows the fluorescence image of a Znarachidate monolayer transferred as a LB film on a glass slide. In this case, because there was no DNA present in the solution it shows very few patches and no detectable image of the dye-bound DNA. These patches are perhaps due to the nonspecific attachment of dye on the glass slide. As a control, we also took the image of the LB film containing only arachidic acid without Zn(II) but in the presence of DNA in the subphase (Figure 4b). However, only when Zn(II) was present in the monolayer with arachidate did compression result in the uptake of DNA from the subphase as observed under the fluorescent microscope with Sybr green staining (Figure 5a). It should be kept in mind that the slide was treated with the dye after the transfer of the monolayer. However, upon transferring DNA on the Zn-arachidate monolayer when dye was present along with the DNA in the subphase, a similar profile was seen (Figure 5b) but the background
5894
Langmuir, Vol. 20, No. 14, 2004
Figure 5. (a) Zn-arachidate-DNA monolayer transferred onto the slide and stained. (b) Stained DNA transferred along with the Zn-arachidate monolayer.
was not very clean, expectedly. All these images point to the necessity of the presence of Zn(II) ions in the subphase for the grafting of DNA on the monolayer and, thus, support the interaction between the Zn ions and DNA. Such interaction, however, is well-known in the literature.27 These experiments were repeated several times to check for the reproducibility until satisfaction. Figure 6a shows the AFM images obtained for a typical scan of size 5 µm × 5 µm. The threadlike structure of DNA is clearly visible in all the images (these were not observed in case of films transferred without the DNA). The typical width of these strands from the image analysis appears to be 90 nm, and the length is 3.2 µm. The width observed is much more than expected for a single DNA (2 nm), so it is assumed that on the surface of the LB film many strands overlap together to form a bundle. The noticeable decrease in the length of the DNA from the ideal value (13.6 µm) suggests partial coiling up or self-folding of the DNA after the transfer. The small range scans (Figure 6b) reveal that the roughness of the surrounding Znarachidate film is mirrored in the DNA bundles. The roughness analysis on the DNA bundle along its length (27) Han, W.; Dlakic, M.; Zhu, J. Y.; Lindsay, M. S.; Harrington, E. R. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 10565-10570.
Bhaumik et al.
Figure 6. (a) AFM image (5 µm × 5 µm) of aligned DNA bundles on the LB film of Zn-Arachidate. (b) Small range scan (2 µm × 2 µm) showing the roughness of the LB film reflected in the DNA bundle.
and on the surrounding film surface yields values of 1.1 nm and 2.4 nm, respectively. This implies that in the present case the observed height of the DNA bundles does not have any special significance and is only a reflection of the roughness of the LB film. Two important features of these films, which came out in AFM imaging, were the partial straightening of the DNA strands and their alignment along the pull direction of the glass slide. This can be explained by two important factors, (a) parallel alignment of the DNAs with the monolayer at the airwater interface and (b) an effect created as a result of the receding meniscus force during the transfer process. Further confirmation on the orientation of the DNA layers over LB films came from fluorescence hybridization experiments. Fluorescence spectra of three different systems, namely, ssDNA1, ssDNA1-ssDNA3 hybrid, and ssDNA1-ssDNA2 hybrid over Zn-arachidate monolayers are shown in Figure 7 as curves 1-3, respectively. The basis of this experiment relies on the fact that ethidium bromide was added in the subphase in each case (see Materials and Methods), and it fluoresces distinctly at different wavelengths when intercalated to a dsDNA.28 Curve 3 shows a clear peak at 628 nm, indicating (28) LePecq, J. B.; Paoletti, C. J. Mol. Biol. 1967, 27, 87.
DNA on LB Films
Langmuir, Vol. 20, No. 14, 2004 5895
Figure 7. Fluorescence spectra of 19 layers of the ssDNA1Zn-arachidate layer (curve 1), ssDNA1-ssDNA3-Zn-arachidate (curve 2), and ssDNA1-ssDNA2-Zn-arachidate (curve 3). The excitation wavelength was kept at 280 nm.
hybridization between complementary ssDNA1 and ssDNA2. The emission maximum is red-shifted compared to that of the free ethidium bromide around 580 nm.29 The earlier report24 reasoned that the differences in polarity of the fluorescent probe when bound to DNA compared to when free in solution could cause this shift. Thus, the successful hybridization experiment at the airwater interface supports the fact that the ssDNA1s are quite freely available and act as a target molecule for the complementary ssDNA2 in the subphase. There are two possible conformations of the immobilized DNA at the air-water interface: (a) DNAs (both dsDNA and ssDNA) are end-grafted to the monolayer, keeping the rest of the molecule free or hanging, and (b) each single DNA molecule is grafted to the monolayer at different points throughout its length, thus, giving an alignment parallel to the monolayer. AFM images in the case of long dsDNA supports the latter assumption quite adequately. However, it should be emphasized here that contrary to the AFM experiments, we used a short synthetic DNA for hybridization, because long DNA for such hybridization is difficult to handle because of entanglement. Nonhybridizable oligonucleotide pairs such as ssDNA1-ssDNA3 or ssDNA1 and their resulting fluorescence with ethidium bromide acts as a control. To check for the accessibility of the DNA toward biological activity, we selected a simple biological reaction, digestion with DNaseI. The control experiments, that is, the treatment of the different parts of the same slide with buffer without DNaseI, were also carried out. It was important to have the control and the test on the same slide primarily to avoid different amounts of DNA transfer in different experiments. This, in a way, also helped us (a) to fix a scan area and (b) to avoid focusing on different DNA molecules before and after the reaction. Figure 8a shows the fluorescence image of the DNA that was not treated with the enzyme. It shows the presence of DNA whose length is around 1.6-1.7 µm. It should be mentioned at this point that the length of extended T7 DNA would be 13.6 µm. Thus, it appears that the DNA is sufficiently coiled up as also observed from our AFM experiments. Figure 8b shows the same image of the DNA when treated with the buffer in the absence of DNaseI. However, Figure 8c shows the digested DNA after the treatment with enzyme for 5 min. Clearly, DNA was shortened to a great extent to about 0.6-0.7 µm. Several such fragments were also noticeable, and the net fluorescence was found to be quenched. (29) Kumar, A.; Pattarkine, M.; Bhadbhade, M.; Mandale, B. A.; Ganesh, N. K.; Datar, S. S.; Dharmadhikari, V. C.; Sastry, M. Adv. Mater. 2001, 13, 5.
Figure 8. Panels a-c show the parts of the slides that were untreated, treated with buffer alone, or treated with DNaseI, respectively.
Conclusion (a) Through the experiments presented in this manuscript, we demonstrated the immobilization of long dsDNA on a LB film with the help of Zn coordination. To our
5896
Langmuir, Vol. 20, No. 14, 2004
knowledge, no such attempts have been made earlier. We recently showed the immobilization of T7 DNA on latex beads,9 but as a result of the nature of the beads, we could not transfer them onto a monolayer. We believe that the method employed here is a better way to immobilize long DNA, which in turn may act as an affinity surface for protein recognition. (b) Both AFM and fluorescence images point out the occurrence of significant coiling up of DNA, but still, such DNAs are good substrates for digestion reaction. Despite being held on the monolayer, these DNAs can act as a target molecule for hybridization through Watson-Crick base pairing. However, we are not sure how they would behave as templates and currently we are trying to follow
Bhaumik et al.
the transcription reaction by Escherichia coli RNA polymerase because T7 DNA has a strong promoter for the same. We believe that this configuration of the monolayer of DNA may also have some other applications. Acknowledgment. The authors wish to thank Department of Biotechnology for financial support, IndoFrench centre for advanced scientific research (IFCPAR) for the Langmuir trough, University Grants Commission, Government of India, for the scholarship to L.K.B., and Ms. Suma, Department of Biochemistry, Indian Institute of Science, for fluorescence microscopy. LA049400G
