pubs.acs.org/Langmuir © 2010 American Chemical Society
)
)
Enantioselective Recognition between Polydiacetylene Nucleolipid Monolayers and Complementary Oligonucleotides† Ina Sigal-Batikoff,‡ Oleg Konovalov, Amarjeet Singh, and Amir Berman*,‡,§ Department of Biotechnology Engineering, and §National Institute of Biotechnology in the Negev and the Ilse Katz Institute of Nanoscale Science and Technology, Ben-Gurion University, Beer-Sheva, Israel, and ESRF, Grenoble, France )
‡
Received May 27, 2010. Revised Manuscript Received October 5, 2010 A two-dimensional bio/synthetic hybrid system at the air-solution interface made of a polymerized diacetylene Langmuir film with nucleobase modified headgroups is presented. The polymerized film presents a crystalline array of nucleobases, capable of specific binding of complementary mononucleoside or oligonucleotide sequences. Mixed monolayers of the linear polyconjugated polydiacetylene (PDA) films derivatized with cytosine (10,12-pentacosadiynecytidyl, PDC) monomers and alcohol-terminated diacetylene lipid (10,12-pentacosadiynol, PDOH) at a 3:1 ratio (PDC 75%) were compressed and polymerized at the air-water interface with circular polarized light (CPL) or nonpolarized UV light. Here we report a grazing incidence X-ray diffraction (GIXD) investigation of PDC films polymerized to different chirality and hybridized with complementary ssDNA strands. We have demonstrated enantioselective interactions on synthetic structured interfaces produced by Langmuir surface compression followed by polymerization with circular polarized UV light (CPL). The left- and right-CPL polymerized light exhibit the same well-defined crystalline structure. The observed difference between left- and right-CPL polymerized PDC 75% Langmuir films compressed over the complementary mononucleotide guanosine or hybridized with fully complementary ssG12T5 oligonucleotide in the subphase suggests that they are indeed enantiomeric structures, capable of enantioselective binding of their natural ligand, guanosine, solely as a result of surface induced asymmetry in “left” but not in “right” form. This observation may also be related to the intriguing question of chiral selection during the early period of “Origin of Life”. We show that achiral compounds, as a result of irradiation with circular polarized light, can organize in chiral surface structures capable of amplification of biopolymer binding of particular handedness.
1. Introduction The study of DNA has taken a significant turn toward its usage as an agent in preparation of new materials in the past decade. DNA high fidelity specific recognition properties which enable the hereditary mechanism are exploited for the purpose of programmable interconnections for the synthetic production of nanostructures with preconceived architectural parameters and properties.1-3 Such materials have led to the development of new biological detection schemes,4 novel nanostructures,5 construction of nanoelectronic devices,6 and “DNA computers”.7 Efforts also have been focused on the construction of DNA-polymer hybrid materials which have been explored for their potential in biodiagnostics and cellular uptake studies.8-10 Here we report a strategy that uses a DNA-driven recognition process to regulate the two-dimensional arrangements at the † Part of the Molecular Surface Chemistry and Its Applications special issue. *To whom correspondence should be addressed. E-mail:
[email protected].
(1) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382, 607–609. (2) Storhoff, J. J.; Mirkin, C. A. Chem. Rev. 1999, 99, 1849–1862. (3) Alivisatos, A. P. Nature 1996, 382, 609–611. (4) Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. J. Am. Chem. Soc. 2003, 125, 896– 900. (5) Seeman, N. C. Nature 2003, 421, 427–431. (6) Keren, K.; Berman, R. S.; Buchstab, E.; Sivan, U.; Braun, E. Science 2003, 302, 1380–1382. (7) Benenson, Y.; Paz-Elizur, T.; Adar, R.; Keinan, E.; Livneh, Z.; Shapiro, E. Nature 2001, 414, 430–434. (8) Korri-Youssoufi, H.; Garnier, F.; Srivastava, P.; Godillot, P.; Yassar, A. J. Am. Chem. Soc. 1997, 119, 7388–7389. (9) Thompson, L. A.; Kowalik, J.; Josowicz, M.; Janata, J. J. Am. Chem. Soc. 2003, 125, 324–325. (10) Jeong, J. H.; Park, T. G. Bioconjugate Chem. 2001, 12, 917–923.
16424 DOI: 10.1021/la102166k
air-solution interface. Specifically, monolayers of the linear polyconjugated polydiacetylene (PDA) films derivatized with cytosine (10,12-pentacosadiyne-cytidyl, PDC) monomers were compressed and polymerized at the air-water interface. We have demonstrated the preparation of an ordered assembly of polymerizable mixed Langmuir monolayers of amphiphilic diacetylene lipids of cytosine and alcohol headgroups (PDC and pentacosadiynol (PDOH), respectively) with a stoichiometric ratio of 3:1 PDC/PDOH, hence PDC 75%. This monolayer is capable of complementary basepair formation.11 It is therefore an attractive template for hybridization of nucleotide oligomers and construction of synthetic polyconjugated polymer-nucleic acid structured hybrid materials that are expected to have both conductance properties due to the PDA component and also carry local information from the DNA. Such hybrid conjugates will possibly find way into combined biosensor/molecular electronics devices. Considering a covalent yne-ene conjugated system periodicity of ∼5 A˚ in polydiacetylene, the average distance between adjacent cytidyl moieties is 5 A˚/0.75 = 6.67 A˚, in good agreement with the approximate separation between adjacent nucleotides on linear ssDNA of 5-7 A˚. The stacking periodicity of nucleotide base pairs in dsDNA is 3.4 A˚, which corresponds to the van der Waals distance between aromatic compounds (π-stack). In cases of approximate lattice match, the base pairs’ tendency to π-stack is expected to produce stacking periodicity with inclination tilt, τ, corresponding approximately to sin(τ) = (π-stack distance)/ (periodic in-plane distance). Lattice mismatch between the polydiacetylene backbone periodicity and that of extended ssDNA (11) Chen, J.; Berman, A. Nanotechnology 2004, 15, S303–S315.
Published on Web 10/12/2010
Langmuir 2010, 26(21), 16424–16433
Sigal-Batikoff et al.
Figure 1. Illustration of the formation of the two optical isomeric structures upon photopolymerization of PCDA with left- and right-CPL. Two alternative (equivalent) rectangular cells are depicted for the monomer (hexagonal cell at center; black and gray). These are “precursors” to the two chiral structures depicted for the polymerized PDA “blue” phase (oblique cells; light blue and pink, left and right, respectively). Adapted with permission from ref 19. Copyright 2009 American Chemical Society. Large and small triangles: tilted alkyl chains in the [11] direction, above and below the conjugated backbone, respectively, Curved arrows: CPL direction. Dashed arrows indicate the shifted position as a result of the corresponding CPL illumination. m: Mirror plane.
may induce deformation or twisting of the conjugated backbone, in a similar way to native DNA double helix formation. The polydiacetylene polymer backbone is a rigid, yne-ene polyconjugated, linear structure that tends to organize in parallel and form two- and three-dimensional (2-D and 3-D) large crystallites on surfaces.12 The long-range order and rigidity of the polymer film provide a good scaffold for ordered assemblies and foster the template properties of PDA. The polyconjugated backbone absorbs light in the visible range and provides the PDA with photoconduction properties.13 External physical triggers such as increased temperature,14 pH, charge,15 mechanical stress,16 or specific binding that sterically strains the film17,18 induce a blueshifted chromatic transition of PDA. 2-D crystallographic structures of the 10,12-pentacosadiynoic acid (PCDA) monomer and the “blue” and “red” polymer phases formed on water subphase were recently solved using in situ synchrotron grazing incidence X-ray diffraction (GIXD) and ex situ transmission electron microscopy and diffraction.19 The compressed PCDA Langmuir monolayer undergoes organized collapse into a stable trilayer structure. Within each layer, the headgroup and methylterminated chains are considered as sublayers located at the same in-plane lattice positions, but adopting different out-of-plane orientations. The monomer phase structure is a 2-D hexagonal lattice in which the headgroup chain sublayer is untilted, but the methyl-terminated alkyl chain sublayer is highly tilted. Structural rearrangement during polymerization involves sliding of the monomers along the direction of the tilted chains on the (11) plane. The transition from hexagonal to oblique cell necessarily breaks the hexagonal symmetry and produces chiral surface arrangement (Figure 1). This phenomenon was overlooked because enantiomeric structures cannot be resolved using X-ray diffraction. The Iwamoto group has reported that indeed PDA films are chiral surfaces, and that PDA surfaces with selected handedness can be produced by polymerization with circular polarized UV (12) Lieser, G.; Tieke, B.; Wegner, G. Thin Solid Films 1980, 68, 77–90. (13) Fisher, N. E. J. Phys.: Condens. Matter 1994, 6, 2047–2058. (14) Kaneko, F.; Shibata, M.; Kobayashi, S. Thin Solid Films 1992, 210, 548–550. (15) Cheng, Q.; Stevens, R. C. Langmuir 1998, 14, 1974–1976. (16) Carpick, R. W.; Sasaki, D. Y.; Burns, A. R. Langmuir 2000, 16, 1270–1278. (17) Charych, D. H.; Nagy, J. O.; Spevak, W.; Bednarski, M. D. Science 1993, 261, 585–588. (18) Charych, D.; Cheng, Q.; Reichert, A.; Kuziemko, G.; Stroh, M.; Nagy, J. O.; Spevak, W.; Stevens, R. C. Chem. Biol. 1996, 3, 113–120. (19) Lifshitz, Y.; Golan, Y.; Konovalov, O.; Berman, A. Langmuir 2009, 25, 4469–4477.
Langmuir 2010, 26(21), 16424–16433
Article
light (CPL).20,21 PDA films were found to have the chirality that corresponded to the direction of the CPL used for their photopolymerization.20 Organized monolayers at the air-water interface provide a convenient environment for 2-D molecular interactions and consequently for molecular recognition investigations, since the functional bonding sites are densely organized on the structured interface. Ordered molecular architectures can be preferentially formed on the basis of complementary hydrogen bonds because of their directionality and specific interactions.22 Numerous well-designed interfacial recognition systems formed on the basis of complementary hydrogen bonding have been reported.23-26 Despite the progress in conceiving many different supramolecular systems, knowledge of structural details of most 2-D specific host-guest monolayers is limited. To date, several studies on molecular recognition of cytosine and guanine base pairing at the interface have been reported25,27-29 primarily for cytosine-functionalized nucleolipids. Among the nucleobases, guanine possesses a unique ability to self-organize via cyclic interactions to form quartets of a regular, ordered structure by hydrogen bonds between the sites N1 and N2 as H-donors and O6 and N7 as H-acceptors.30,31 Synchrotron GIXD is a powerful tool to study the packing of amphiphiles at the air-water interface.32 It allows resolving the 2-D crystalline structure lattice parameters, molecular tilt, and its projected azimuth of Langmuir monolayers at the air-water interface. Degree of crystallinity and disorder can be evaluated from the diffraction Bragg rod broadening along the scattering vector in the sample plane, qxy, and the angular distribution of the diffracted intensity in the direction perpendicular to the sample surface, qz (Figure 2). Misfit of the projected areas of the headgroups and alkyl chains may induce disorder that may not be compensated by the tilt of the alkyl chains.33 In such cases, the intensity is distributed along a characteristic arc of equal diffraction vector length qtotal = (qxy2 þ qz2)1/2. Here we report the 2-D crystallographic structure of the PDC 75% film and enantioselective binding of left- and right-CPL polymerized PDC 75% Langmuir films with the complementary mono- and oligonucleotides of guanosine in the subphase.
2. Experimental Section 2.1. Materials. 10,12-Pentacosadiynoic acid (PCDA, 1) and cytosine were purchased from Sigma. The nucleolipid amphiphiles (20) Manaka, T.; Kon, H.; Ohshima, Y.; Zou, G.; Iwamoto, M. Chem. Lett. 2006, 35, 1028. (21) Kohn, H.; Ohshima, Y.; Manaka, T.; Iwamoto, M. Jpn. J. Appl. Phys 2008, 47, 1359–1362. (22) Cantor, C. R.; Schimmel, P. R. Biophysical Chemistry. Part 1: The Conformation of Biological Macromolecules; W.H. Freeman and Company: New York, 1980. (23) Kunitake, T. Supramol. Sci. 1996, 3, 45–51. (24) Ariga, K.; Kunitake, T. Acc. Chem. Res. 1998, 31, 371–378. (25) Shimomura, M.; Nakamura, F.; Ijiro, K.; Taketsuna, H.; Tanaka, M.; Nakamura, H.; Hasebe, K. J. Am. Chem. Soc. 1997, 119, 2341–2342. (26) Shimomura, M.; Nakamura, F.; Ijiro, K.; Taketsuna, H.; Tanaka, M.; Nakamura, H.; Hasebe, K. Thin Solid Films 1996, 285, 691–693. (27) Miao, W.; Du, X.; Liang, Y. Langmuir 2003, 19, 5389–5396. (28) Miao, W.; Du, X.; Liang, Y. J. Phys. Chem. B 2003, 107, 13636–13642. (29) Wang, Y. C.; Du, X. Z.; Miao, W. G.; Liang, Y. Q. J. Phys. Chem. B 2006, 110, 4914–4923. (30) Pinnavaia, T. J.; Miles, H. T.; Becker, E. D. J. Am. Chem. Soc. 1975, 97, 7198–7200. (31) Pinnavaia, T. J.; Marshall, C. L.; Mettler, C. M.; Fisk, C. L.; Miles, H. T.; Becker, E. D. J. Am. Chem. Soc. 1978, 100, 3625–3627. (32) Als-Nielsen, J.; Jacquemain, D.; Kjaer, K.; Leveiller, F.; Lahav, M.; Leiserowitz, L. Phys. Rep. 1994, 246, 251–313. (33) Weidemann, G.; Brezesinski, G.; Vollhardt, D.; M€ohwald, H. Langmuir 1998, 14, 6485–6492.
DOI: 10.1021/la102166k
16425
Article
Figure 2. Schematic drawing of surface scattering geometry: ki, ks, incident and scattered beams, respectively; q, scattering vector; qxy, qz scattering vector projections on the plane and plane normal directions, respectively; Ri, incident angle; δ, γ, in-plane and out-ofplane components of diffraction angle, respectively.
Sigal-Batikoff et al. monolayer was formed. Alternatively, the spreading solution was applied onto buffer solution without surface compression until the desired surface pressure (12 mN m-1 > π > 25 mN m-1) was reached and 15 min was allowed for solvent evaporation. Langmuir films were polymerized at the air-solution interface using either of the following UV sources: a hand-held 254 nm UV lamp (Pen-ray, 4 W, UVP). The intensity of this lamp was approximately 30 mW cm-1 from 1 cm above the surface. The irradiation time was 20 s from 20 cm above the surface. Alternatively, photopolymerization was carried out with circular polarized UV light (CPL, 315 nm), generated by a mercury-xenon lamp (HAMAMATSU; LC-8), coupled to a Babinet-Soleil compensator (Karl Lambrecht Corp. Chicago, IL) via a quartz optical fiber, used in order to induce surface chirality that corresponds to the CPL direction. The intensity of left- and right-CPL was approximately 10 mW cm-1 from 1 cm above the surface. Irradiation time was 2 min. A total of 500 μL of oligonucleotide 10 μM aliquouts were injected into the subphase and were allowed 30 min for hybridization. A schematic drawing of the polymerization and specific binding of ssDNA to the cytosine modified polydiacetylene at the interface is presented in Figure 4. 2.3. Brewster Angle Microscopy (BAM). Langmuir films of PDC 75% compressed on complementary and noncomplementary oligonucleotides were imaged with BAM at the airsolution interface. A Brewster angle microscope, EP3-SW-BAM NanoFilm Technology, Germany, was operated during the compression. The Nanofilm EP3 has a lateral resolution of 1 μm (20 objective) or 2 μm (10 objective) and a field of view of ∼0.22 mm or 0.44 mm, respectively. The light source was a 50 mW, 532 nm laser.
2.4. Grazing Incidence X-ray Diffraction (GIXD) Measurements. GIXD experiments were performed at the ID-10B
2.2. Langmuir Film Preparation and Pressure/Area Isotherms. Monolayers were prepared on a conventional Teflon
(Troika-II) beamline at the ESRF (Grenoble, France), a multipurpose setup providing horizontal scattering geometry for the study of liquid interfaces.34 Measurements were performed using a beam energy of 22 keV (λ = 0.56 A˚) radiation at an incident angle of R = 0.047°, or 8.04 keV (λ = 1.54 A˚) at an incident angle of R = 0.11°. At incidence angles smaller than 90% of the critical angle, an evanescent wave propagating parallel to the surface is formed with a penetration depth of 50-100 A˚. Hence, it probes the near surface structure of the film and effectively eliminates scattering from the subphase.35 Grazing incidence diffraction data were collected with a 150 mm position sensitive detector (PSD) oriented perpendicular to the film plane, combined with a 150 mm 15 mm Soller collimator providing an angular resolution of 1.4 mrad. The active detector area could be scanned both horizontally (angle δ) and vertically (angle γ) with respect to the direct beam. The scattering angle 2θ is given by cos(2θ) = cos γ cos δ.36 For the purpose of GIXD experiments, lipid monolayers are usually considered as ideal 2-D powders, in which the lipid crystalline domains are oriented in random orientations in the monolayer plane.37 The scattering vector q can be decomposed into two orthogonal components (qxy and qz) and is given by q2 = qxy2 þ qz2 (Figure 2). The in-plane vector qxy = (4π/λ) sin(δ/2) is parallel to the water surface, while the vector qz = (2π/λ)(sin γ þ sin Ri) is normal to the water surface. The in-plane diffraction vector, qxy component, gives information on the periodic structure of the monolayer parallel to the aqueous subphase, whereas qz, the out-of-plane component, together with qxy gives information on the tilt angle and tilt direction of the hydrophobic side chains
Langmuir trough (model 611A, Nima Technology Ltd. Coventry, U.K.) of 630 cm2 surface area equipped with a Wilhelmy plate and electronic balance. The error in the surface pressure measurements was estimated as (0.5 mN m-1. Spreading solution was applied dropwise onto the liquid surface, and 15 min was allowed for solvent evaporation. Compression was performed at rate of 1.5 A˚2 molecule-1 min-1 until the desired surface pressure (20 mN m-1 > π > 30 mN m-1) was reached and a uniform condensed
(34) Smilgies, D.-M.; Boudet, N.; Struth, B.; Konovalov, O. J. Synchrotron Radiat. 2005, 12, 329–339. (35) Als-Nielsen, J.; McMorrow, D. Elements of Modern X-ray Physics; John Wiley and Sons: New York, 2001. (36) Breiby, D. W.; Samuelsen, E. J.; Konovalov, O.; Struth, B. Langmuir 2004, 20, 4116–4123. (37) Tanaka, M.; Schneider, M., F.; Brezesinski, G. ChemPhysChem 2003, 4, 1316–1322.
Figure 3. Chemical structures of diacetylene lipids: 10,12-pentacosadiynoic acid (PCDA, 1), pentacosadiyne-cytosinyl derivative (PDC, 2), and alcohol derivative (PDOH, 3). The pure components, having identical diacetylene alkyl chains, differ in their headgroups. pentacosadiyne-cytosinyl derivative (PDC, 2) and alcohol derivative (PDOH, 3) were prepared from PCDA 1 as was previously reported11 (Figure 3). PDC and PDOH powders were dissolved in CHCl3 (Sigma) to a final concentration of 2 mM. A mixed 3:1 ratio of PDC/PDOH gave an optimal Langmuir film.11 ssDNA of C12T5 (CCCCCCCCCCCCTTTTT), G12T5 (GGGGGGGGGGGGTTTTT), and 8GT (GTGTGTGTGTGTGTGT) were purchased from Rhenium Ltd., Israel. Oligonucleotide solutions (10 μM), calculated with respect to oligonucleotide amount, were prepared by dissolving the powders in freshly prepared Trizma buffer. The prepared solutions of oligonucleotides were kept in -4 °C until use. Guanosine (G) (0.5 mM; Sigma) solutions were prepared by dissolving the powder in freshly prepared Trizma buffer, heating for 20 min (40 °C), and filtering with 0.22 μm GP express plus membrane filter. Trizma buffer 0.05 M subphase solution was prepared from Trizma base (Sigma) using ultrapure water (18.2 MΩ cm-1, 20 °C; Millipore); pH was adjusted to 7.5 with dilute HCl and NaOH solutions.
16426 DOI: 10.1021/la102166k
Langmuir 2010, 26(21), 16424–16433
Sigal-Batikoff et al.
Article
Figure 4. Schematic drawing of the PDC polymerization and specific binding of guanine-containing ssDNA to the cytosine modified polydiacetylene at the interface.
Figure 5. (A) π-A isotherms of PDC 75% film compressed on Trizma buffer (dashed black line), on C12T5 buffered solution (solid black line), and on G12T5 buffered solution (solid gray line), The arrows indicate the “shoulders”, at surface pressures where nucleation of LC domains take place. (B) BAM images of PDC 75% film compressed on C12T5 buffered solution (a,b) captured at surface pressures of 19 mN/m (a) and 22 mN/m (b) and of PDC 75% film compressed on G12T5 buffered solution (c,d) captured at surface pressures of 14 mN/m (c) and 20 mN/m (d) during PDC 75% film compression. The image size is 485 μm for (a). BAM images (a) and (c) were captured immediately after surface crystallization took place. of the lipids.38 The theoretical resolution for the in-plane transfer wave vector qxy is of the order of 0.01 A˚-1 FWHM (full width at half-maximum).39 The intensity value of each raw data point was corrected by multiplication of its corresponding qxy value in order to account for the background intensity decay, followed by curve smoothing.
3. Results and Discussion 3.1. π-A Isotherms. PDC 75% monolayers were examined for their specific binding capacity with complementary and noncomplementary oligonucleotides in the subphase, by measuring (38) Konovalov, O.; Myagkov, I.; Struth, B.; Lohner, K. Eur. Biophys. J. 2002, 31, 428–437. (39) Saint-Jalmes, A.; Graner, F.; Gallet, F.; Nassoy, P.; Goldmann, M. Chem. Phys. Lett. 1995, 240, 234–238.
Langmuir 2010, 26(21), 16424–16433
the surface pressure-area (π-A) isotherms. The PDC 75% compression isotherm yields a stable monolayer with a liquid condensed phase (LC) limiting area (AL) of 58 ( 3 A˚2/ molecule, estimated from the π-A curve by extrapolating a tangent to the isotherm to zero surface pressure (Figure 5A). With compression on the C12T5 subphase, the nucleolipids yielded a stable monolayer with a limiting area (AL) of 60 ( 3 A˚2/molecule (Figure 5A). The PDC 75% monolayer compressed on G12T5 manifests a nearly continuous slope with a shallow “shoulder” at π = 14 mN/m, compared with the more pronounced shoulder observed in the presence or absence of the cytosine oligonucleotide at π = 20 mN/m. These shoulders are associated with the nucleation of the LC domains (Figure 5B). Moreover, the π-A curve is shifted to a larger molecular area (67 ( 3 A˚2/molecule), indicating the presence DOI: 10.1021/la102166k
16427
Article
of larger effective headgroups, possibly due to specific base pairs at the monolayer interface. Langmuir films of PDC 75% in the presence of G12T5 complementary or C12T5 noncomplementary oligonucleotide in the subphase were imaged with BAM at the air-solution interface during surface compression (Figure 5B). The PDC 75%/C12T5 domain morphology is linear primary shafts (Figure 5B, a,b). The primary shafts are the first to nucleate, and then further growth (secondary branches) takes place. The nucleation takes place at the surface pressure of 19 mN/m for PDC 75% compressed on C12T5 (Figure 5B, a) or in its absence (image not shown), compared to PDC 75% /G12T5 nucleation at a surface pressure of π = 14 mN/m (Figure 5B, c). The LC domains in the presence or absence of C12T5 appear linear over tens of micrometers. This long order without apparent perturbations indicates that the domain organization is strain free. The early LC nucleation in the presence of the complementary oligonucleotide indicates that indeed preferential interaction took place between the film and the DNA. PDC75%/G12T5 domains are more curved with higher surface density (Figure 5B, c and d). The domains are smaller and made up of several branches that emerge from a common nucleation center. Smaller crystalline domains with higher surface density were formed, possibly due to the decreased nucleation energy.40,41 The LC domain primary shafts, which with addition of noncomplementary sequence appear straight (Figure 5B, a and b), are curved in the presence of complementary sequences (Figure 5B, c and d), suggesting that the complementary binding caused surface deformation. Similar observations were made also for complementary mononucleosides.11 3.2. Grazing Incidence X-ray Diffraction. 3.2.1. PDC 75% UV Irradiated Film Structure. GIXD reciprocal maps obtained from PDC films that were polymerized with CPL UV light (Figure 6B) are compared to blue phase PDA (Figure 6A), to unirradiated PDC 75% film (monomer) (Figure 6C), and to PDC 75% films polymerized with nonpolarized light (Figure 6D). The reciprocal map depicted in Figure 6B was used to solve the PDC 75% film structures polymerized with CPL (the PDC 75%_ left and PDC 75%_right reciprocal maps are identical). Four reflections, detailed in Table 1, are observed at three distinct qxy positions. Considering the recently solved carboxylate-terminated headgroup PDA structure19 as a benchmark, we have indexed the reflections and deduced the structure of the CPL polymerized PDC films (Table 2). We note that left-and right-CPL produce indistinguishable 2-D structures. Nevertheless, they are of opposite handedness as was shown by circular dichroism (CD) measurements for PCDA (1)20 and for PDC 75% (manuscript in preparation). The three observed in-plane reflections centered at qxy = 1.266, 1.378, and 1.4825 A˚-1 correspond to interplanar distances of 4.963, 4.562, and 4.238 A˚, respectively. A 2-D oblique unit cell (n = 2) with the dimensions a = 4.917 A˚, b = 9.97 A˚, and γ = 84.7° was uniquely generated from these distances (Figure 7A). The main structural features in “blue phase PDA” are the rows of curved, highly inclined alkyl chains in the [11] direction, which slide against each other during the structural shift that is associated with PDA polymerization (Figure 1). This feature is manifested in the (11) reflection with qz = 0 (Figure 6A). In contrast, for the PDC structure emanating from the GIXD diffraction (40) Helm, C. A.; Laxhuber, L.; L€osche, M.; M€ohwald, H. Colloid Polym. Sci. 1986, 264, 46–55. (41) Helm, C. A.; M€ohwald, H. J. Phys. Chem. 1988, 92, 1262–1266.
16428 DOI: 10.1021/la102166k
Sigal-Batikoff et al.
Figure 6. 2-D reciprocal space maps obtained from PDA film compressed to π = 25 mN/m, UV irradiated with nonpolarized UV light19 (A); PDC 75% film compressed to π = 30 mN/m, UV irradiated with left- or right-CPL (B); unirradiated PDC 75% film spread to π = 25 mN/m (monomer) (C); and PDC 75% film spread to π = 12 mN/m irradiated with nonpolarized UV light (D). Arcs are qtot = 1.9, 1.60, and 1.51 A˚-1, which corresponds to d = 3.31, 3.93, and 4.16 A˚, respectively. The calculated positions for PDA and PDC, based on the crystallographic model, are marked by triangles in A and B, respectively. Langmuir 2010, 26(21), 16424–16433
Sigal-Batikoff et al.
Article Table 1. Summary of the GIXD Reflections in Figure 6Ba
Miller indices
qxy (A˚ )
(02)
1.266
-1
qz (A˚-1)
qtot (A˚-1)
τ (deg)
dxy (A˚)
ψ (azimuth from b*)
t (calculated tilt)
1.01 1.62 36.51 4.963 0.0° 38.58° 0.98 (obs) 1.60 0.0° 37.46 (obs) 1.378 0.41 1.44 16.55 4.562 68° 38.59° (11)A 1.4825 0.61 1.61 22.26 4.238 59.5° 38.56° (11) 1.378 0 1.38 0 4.562 68° (n.a.) (11)B a qxy and qz are the peak coordinates in reciprocal space; qtot = (qxy2 þ qz2)1/2 is the reciprocal interplanar distance, τ is the tilt angle of the diffraction vectors with respect to the normal; ψ is the azimuth of the alkyl chain, projected on the x-y plane with respect to the b* direction; t is the calculated tilt of the alkyl chains using tan(t) = qz/(qxy cos ψ).33
Figure 7. Schematic representation of PDC 75% cell calculated from the 2-D reciprocal map presented in Figure 6B. (A) 2-D cell dimensions are based on the observed three reciprocal spacings d(02), d(11), and d(11). Triangles represent uniformly tilted alkyl chains in the b* direction (the a axis was assigned as the polymer conjugated direction), calculated from reflections in Figure 6B. Tilt angle calculated from the qz values of these reflections is 38° from the normal. (B) 4 2 array of the unit cells in A. The long and short rectangles correspond to the PDC and PDOH headgroups, respectively. This suggested headgroup arrangement can give rise to the reflection at qxy = 1.375 A˚-1; qz = 0.0 A˚-1. (C) Schematic depiction of the film’s molecular components. PDC headgroup projection is larger than the projected area of the alkyl chain. This is compensated by the smaller headgroup of PDOH. The random distribution of PDOH in the film introduces disorder which is manifested in a single reflection from the headgroup sublayer.
reciprocal map (Figure 6B), the alkyl chains were found to be tilted by 38.6° toward the b* direction, that is, perpendicular to the conjugated backbone (Figure 7A). As is the case for all PDA films, the conjugated backbone is a rigid structural scaffold onto which the methyl-terminated (hydrophobic) and headgroupterminated (hydrophilic) alkyl chains are attached (Figure 4). This results in two sublayers with identical 2-D (in-plane) order, but possibly different out-of-plane organization, which is manifested in two or more reflections at the same qxy positions. In contrast, the non-CPL polymerized PDC film exhibits poor crystallinity, judged from its weak intensity reflections, packed in a different structure (Figure 6D). The three reflections observed for non-CPL PDC have similar interplanar tilted spacings of approximately 4.16 A˚-1. Their projections onto the plane correspond to (11) and (11) reflections of the PDA-blue phase. The third reflection cannot be unequivocally related to the PDA-blue (02) reflection, so although this structural variation cannot be unambiguously solved at this stage, it appears more disordered. 3.2.2. Interactions of CPL-Polymerized PDC 75% Films with Complementary Guanosine Mononucleotides. PDC 75% Langmuir films were compressed over a subphase containing the complementary mononucleoside, guanosine. The resultant film is organized in a structure that is related to that of PDC monomer (Figure 6C), though with several important differences (Figure 8A). The prominent observations are the appearance of a series of reflections at qxy = 1.31 A˚-1 and qz = 0.0, 0.4, 0.5, and 0.7 A˚-1 and a shift of the reflection to higher qxy values, indicating a smaller, more compact unit cell as the result of base-pair formation. It is plausible to index these reflections as follows qxy = 1.31 (02); qxy = 1.42 - (11); qxy = 1.50 - (11), according to their Langmuir 2010, 26(21), 16424–16433
relative order in PDC 75% CPL irradiated structure (Figure 6B). The absence of the (02) reflection in PDC 75% monomer (Figure 6C) and its appearance in the presence of G indicate the increase in the film’s order, induced by base-pairing. Polymerization of the PDC film compressed on the G subphase with non-CPL UV light for 20 s results in distorted reflections along equal qtotal arcs (Figure 8B). The massive reflection spot centered at qtotal = 1.9 A˚-1 corresponds to d = 3.3 A˚, the typical π-stacking distance of DNA base pairs. Yet, its broad nature suggest highly disordered organization. PDC 75% polymerization on the G containing subphase with right-CPL UV light results in a slightly different structure (Figure 8C), compared with the PDC film in the absence of G (Figure 6B). The (11) and the (02) reflections appear at the same positions and tilt as in CPL polymerized PDC 75% in the absence of G. However, the (11) reflection has shifted toward a higher qxy value (from qxy =1.48 to qxy =1.51 A˚-1), indicating a more condensed packing in this direction. As a result, this reflection is intensified and appears at qz = 0, indicating that these planes are not tilted outside the [11] diection. The significant shift in position and reflected intensity suggest that the (11) reflecting planes are directly affected by the subphase; hence, it may be associated with the planar cytidyl groups. Very distinct reflections centered on the qtotal = 1.9 A˚-1 arc were observed when the PDC 75% film was polymerized with left-CPL UV light (Figure 8D). The two observed reflections projection onto the qxy axis are at qxy = 1.43 A˚-1 (very strong) and qxy = 1.375 A˚-1 (weak). The observed difference between left- and right-CPL suggests that they are indeed enantiomeric structures, capable of enantioselective binding of their DOI: 10.1021/la102166k
16429
Article
Figure 8. 2-D reciprocal space maps obtained from PDC 75% films spread to π = 12 mN/m on guanosine mononucleoside buffered solution: (A) unirradiated (monomer) film, (B) irradiation with nonpolarized UV light, (C) irradiation with a right-CPL and (D) left-CPL UV light. Arcs are qtot = 1.9, 1.60, and 1.51 A˚-1, which corresponds to d = 3.31, 3.93, and 4.16 A˚, respectively. 16430 DOI: 10.1021/la102166k
Sigal-Batikoff et al.
natural ligand, guanosine, solely as a result surface induced asymmetry. 3.2.3. Interactions of CPL-Polymerized PDC 75% Films with Complementary ssDNA GT Oligonucleotides. PDC 75% monolayers were compressed on buffer and polymerized with non-CPL, left- or right-CPL UV light. After polymerization, 16- or 17-mer oligonucleotides, fully or partially complementary to the PDC film, were carefully injected under the film and given 30 min to hybridize. Figure 9A depicts the 2-D diffraction map obtained when fully complementary ssG12T5 oligonucleotide was injected under a film polymerized with non-CPL light. The diffraction pattern resembles that of the same film without the oligonucleotide (Figures 9A and 6D, respectively) and reflects a poorly crystalline assembly. Injecting the partially complementary, alternate (GT)8 16-mer nucleotide, severe deformation of the film is observed (Figure 9B): the well localized diffraction spot transformed into an arc of equal qtotal value. This can be interpreted as the result of a planar film undergoing out-of-plane deformation, possibly due to the significant mismatch that cannot be accommodated between the regular presentation of cytidyl moieties in PDC film, and the doubly spaced, partial complementary nucleotide. The fully complementary oligonucleotide ssG12T5 did not induce any observable change also when hybridized with the right-CPL light (Figure 9C). The 2-D diffraction map resembles that of similarly polymerized PDC 75% film without oligonucleotide (Figure 6B). A very different effect results when using left-CPL. A series of high qz Bragg rods centered at qtotal = 1.9 A˚-1 (d = 3.31 A˚) are observed (Figure 9D). Notably, the PDC film reflections appear at the same positions as those for CPL polymerized bare PDC. This observation suggests that complementary base pairs are formed between the cytidyl headgroups of the film and the complementary G12T5 oligomers at the subphase, without significantly disrupting or deforming the film order. The qxy positions of these reflection correspond to their in-plane spacings, between 1.17 A˚-1 < qxy < 1.50 A˚-1, corresponding to in-plane spacing of 5.37-4.19 A˚ and an inclination of 45 ( 7°. The observation of these reflections may suggest nonuniform orientation with respect to the PDC unit cell. It follows that base-pair π-stacking may take place between neighboring cytidyl moieties, not necessarily on the same polydiacetylene backbone. 3.2.4. Interactions of PDC 75% Films with Noncomplementary ssDNA C12T5. Interaction of non-Watson-Crick complementary oligomers C12T5 resulted in a similar effect to what was observed for the complementary oligomer: a series of intense reflections with typical π-stack spacing, tilted at about 45° ( 7° with respect to the surface plane (Figure 10). Surprisingly, this nonstandard hybridization took place on right-CPL polymerized PDC, but not on the left-CPL, as is the case for hybridization of the mononucletide G and the fully complementary oligonucleotides. We cannot account at this stage for the exact stereochemical details of these structured enantioselective surface interactions. The somewhat disturbing observation of hybridization of cytosine-rich oligonucleotides with the cytidyl headgroups in an ordered manner can possibly be explained on the basis of observations on cytosine-cytosine base pairing42 or cytosine quartets.43,44 Similarly, it may well be that the observed hybridi(42) Gehring, K.; Leroy, J. L.; Gueron, M. Nature 1993, 363, 561–565. (43) Spackova, N.; Berger, I.; Sponer, J. J. Am. Chem. Soc. 2001, 123, 3295– 3307. (44) Cai, L.; Chen, L.; Raghavan, S.; Ratliff, R.; Moyzis, R.; Rich, A. Nucleic Acids Res. 1998, 26, 4696.
Langmuir 2010, 26(21), 16424–16433
Sigal-Batikoff et al.
Article
Figure 10. 2-D reciprocal space map obtained from PDC 75% film
compressed to π = 20 mN/m, UV irradiated with a right-CPL and incubated with C12T5-ssDNA. Arcs are qtot = 1.9, 1.60, and 1.51 A˚-1, which corresponds to d = 3.31, 3.93, and 4.16 A˚, respectively. Dotted circles correspond to the reflections observed for CPL polymerized film in the absence of oligonucleotide in Figure 6B. These reflections are partially masked due to the intense reflections.
Figure 9. 2-D reciprocal space map obtained from PDC 75% films (A) compressed to π = 22 mN/m, incubated with complementary oligonucleotide G12T5-ssDNA followed by UV irradiation with nonpolarized UV light; (B) spread to π = 12 mN/m, incubated with partially complementary oligonucleotide (GT)8ssDNA; (C, D) the same film as in (A) spread to π = 12 mN/m and polymerized with right-CPL (C) or compressed to π = 20 mN/m and polymerized with left-CPL (D). Arcs are qtot = 1.9, 1.60, and 1.51 A˚-1, which corresponds to d = 3.31, 3.93, and 4.16 A˚, respectively. Langmuir 2010, 26(21), 16424–16433
zation of the G-rich oligonucleotides is related to non-WatsonCrick base pairing such as Hoogsteen base pair or G/C quartet formation.45 3.2.5. Bragg Rod Analysis. Analysis of the intensity profile along qz was performed in order to gain information regarding the layer thickness. The film thickness is inversely proportional to the intensity distribution width, calculated as thickness (A˚) = 0.9π/ FWHM, where FWHM is the full width at half-maximum of the fitted Gaussian curves in reciprocal length units (A˚-1). Figure 11 and Table 2 present the analyses of selected Bragg rods of leftCPL-polymerized PDC 75% in the absence and presence of hybridized mono- and oligonucleotides. In polydiacetylene lipid monolayers, two structurally different sublayers are tethered at the same lattice positions, dictated by the conjugated polymer backbone.19 According to the experimental evidence and structural interpretation presented in Figures 6B and 7, the PDC 75% methyl-terminated sublayer (“air-side”) is the 12-C alkyl chain, tilted approximately 38°. This yields a layer of thickness 12C 1.2 A˚/ (methylene bond) cos(38°) ∼ 11.5 A˚, in agreement with the observed 13.3 ( 4.7 A˚ for the Bragg-rods at qxy = 1.38 A˚-1; qz = 0.4 A˚-1. Similarly, the headgroup chains bearing the cytidyl groups give rise to the reflection at qxy = 1.38 A˚-1; qz = 0.0 A˚-1. Similar estimation of the “water-side” sublayer yields 11 bonds (alkyl chain þ pyrimidine ring), corresponding to ∼13.2 A˚ (no tilt), also in agreement with the measured 13.5 ( 5.4 A˚. Broad Bragg rod reflections that are associated with the basepair stacking have the typical diffraction vector of qtotal = 1.9 A˚-1, corresponding to d-spacing of 3.3 A˚. Few of these reflections, located at qxy = 1.38 and 1.44 A˚-1, are presented and analyzed. The average calculated thickness for the “base-pairs layer” reflections is 4.1 ( 0.34 and 4.0 ( 0.37 A˚, for qxy = 1.38 and 1.44 A˚-1, respectively. As a reference, we consider that the diameter of compact B-DNA double helix is 20 A˚, about half of which is occupied by the base pairs (excluding the DNA sugarphosphate backbones). Further, taking into consideration the ca. 45° tilt, the expected thickness would be 10 A˚ cos(45) ∼ 7 A˚. (45) Lavery, R.; Zakrzewska, K.; Sun, J. S.; Harvey, S. C. Nucleic Acids Res. 1992, 20, 5011.
DOI: 10.1021/la102166k
16431
Article
Sigal-Batikoff et al. Table 2. Selected Bragg Rod Analysisa qxy = 1.38
a b d f
PDC 75 R/L PDC 75/G L PDC 75/G12T5 L PDC 75/C12T5 R
qxy = 1.44
Δqz 0.0 (HWHM)
thickness (A˚)
Δqz 0.4 (FWHM)
thickness (A˚)
Δqz 1.4 (FWHM)
thickness (A˚)
0.0738 0.1043 0.093 0.228
19.16 13.55 15.20 6.20
0.1535 0.2334 0.1843 0.3875
18.42 12.11 15.34 7.30
0.744 0.726 0.619 0.697
3.80 3.89 4.57 4.06
average 13.53 std error 5.42 a a-g correspond to the same in Figure 11.
13.29 4.75
4.08 0.34
c e g
Δqz 1.4 (FWHM)
thickness (A˚)
0.732 0.75 0.635
3.86 3.77 4.45 4.03 0.37
Figure 11. Selected Bragg rods intensity profiles. (a) PDC 75% left- or right-CPL (Figure 6B). (b,c) PDC 75% compressed on guanosine mononucleoside and polymerized with left-CPL (Figure 8D). (d,e) PDC 75% incubated with complementary oligonucleotide G12T5 and polymerized with left-CPL (Figure 9D). (f,g) PDC 75% incubated with noncomplementary oligonucleotide C12T5 and polymerized with right-CPL (Figure 10). The qxy positions are 1.38 A˚-1 for a, b, d, and f; qxy = 1.44 ( 0.015 A˚-1 for c, d, and g. The intensity profiles were background subtracted and fitted with Lorentzian profiles, used for film thickness calculation (presented in Table 2).
The measured base-pair thickness of approximately 4 A˚ is about half the expected thickness. This discrepancy may be the results of a nonuniform or partially disordered layer.
4. Conclusions In this work, we have demonstrated enantioselective molecular recognition interactions on synthetic hybrid structured interfaces produced by Langmuir surface compression followed by polymerization with circular polarized UV light (CPL). The left- and right-CPL polymerized films exhibit a well-defined crystalline structure. The main difference between PDC and blue-PDA is that PDC is organized in a monolayer while PDA tends to orderly collapse into a stable trilayer. A considerable difference in the limiting area between PDA and PDC (PDA: 27 A˚2/molecule vs ∼60 A˚2/molecule for PDC) likely stems from the domain 16432 DOI: 10.1021/la102166k
morphology of the two films. Another important difference is the alkyl chain tilt direction. In blue-phase PDA, the main structural feature is the highly inclined alkyl chains in the [11] direction, manifested in the (11) reflection with qz = 0, while in PDC the tilt direction is perpendicular to the conjugated backbone. This difference in the alkyl chain packing may be the reason for the different chromatic response of the two films: PDA is attracting attention for its facile “blue to red” transition in response to many external triggers.12,14-16,18 In our earlier GIXD study on PDA, we claim that the transition to the “red-phase” involves sliding of the alkyl chains in the [11] direction, followed by shortening the distance between the conjugated backbones, and simultaneous shift of the curved and inclined chains to upright position. In contrast, the alkyl chains of CPL-polymerized PDC, shown here to be inclined perpendicular to the Langmuir 2010, 26(21), 16424–16433
Sigal-Batikoff et al.
conjugated backbone, likely restrict the restructuring associated with the “chromatic phase transition”. Possibly, this is the reason for negligible “chromatic response” observed in this system.11 The non-CPL polymerized PDC film exhibits poor crystallinity, judging from its weak intensity reflections, and probably represents a racemic structure. Since the monomer is achiral, this notion is of particular interest and indicates that the chiral surface structures originate during UV irradiation by deformation of symmetric packing. The observed difference between left- and right-CPL polymerized PDC 75% Langmuir films compressed over the complementary mononucleotide G or hybridized with fully complementary ssG12T5 oligonucleotide in the subphase suggests that they are indeed enantiomeric structures, capable of enantioselective binding of their natural ligand, guanosine, solely as a result surface induced asymmetry in the “left” but not in the “right” form. (46) Hazen, R. M.; Sholl, D. S. Nat. Mater. 2003, 2, 367–374.
Langmuir 2010, 26(21), 16424–16433
Article
Our finding may also be related to the intriguing question of chiral selection during the early period of “Origin of Life”.46 We show that chiral compounds, as a result of irradiation with circular polarized light, or otherwise stabilized at the interface, can organize in chiral surface structures capable of selective amplification of biopolymer binding of particular handedness. Acknowledgment. We thank G. Lemcoff and Y. Dayagi for help in organic synthesis. The German-Israel Science Foundation Grant #G-791-133.10/2003 and Ministry of Science Joint IsraelKorea research Grant #3-4619 are gratefully acknowledged for partially funding this research. Note Added after ASAP Publication. This article was published ASAP on October 12, 2010 with an incorrect version of the authors’ first and last names. The correct version was reposted October 14, 2010.
DOI: 10.1021/la102166k
16433