5398
Langmuir 2006, 22, 5398-5402
Semiempirical Quantum Mechanical Calculations of Dipolar Interaction between Dipyridamole and Dipalmitoyl Phosphatidyl Choline in Langmuir Monolayers K. Thirumoorthy and N. Nandi* Chemistry Department, Birla Institute of Technology and Science, Pilani, Rajasthan 333031, India
D. Vollhardt* Max Planck Institute of Colloids and Interfaces, D-14424 Potsdam/Golm, Germany
O. N. Oliveira, Jr. Instituto de Fisica de Sa˜ o Carlos, USP, Cx. Postal 780, 13560-970 Sa˜ o Carlos, SP, Brazil ReceiVed January 25, 2006. In Final Form: March 30, 2006 Recent studies have shown that dipalmitoyl phosphatidyl choline (DPPC) monolayers respond cooperatively to the presence of dipyridamole (DIP) guest molecules even at small concentrations, which is a signature of molecular recognition. Using semiempirical quantum mechanical calculations for the DIP-DPPC system, we show that the incorporation of DIP causes large changes in the vertical dipole moment of the DIP-DPPC system, which can explain why measurable changes in surface potential are observed experimentally even at very low DIP concentrations. The calculations are also consistent with the anomalous concentration dependence of the surface pressure and surface potential isotherms for DIP-DPPC monolayers. Rather than saturation or a continuous increase in the effects caused by the incorporation of increasing amounts of DIP, the experimentally observed inversion in the behavior of the surface potential as the DIP concentration reaches 0.5 mol % would be caused by a change in DIP conformation, from a vertical arrangement for the DIP rings to a horizontal or intermediate arrangement. The strong dipolar interactions indicated in the calculations may also be the origin of the drastic changes in monolayer morphology seen in fluorescence microscopy images, with triskellion-shaped domains being formed for condensed DIP-DPPC monolayers.
1. Introduction Phospholipid Langmuir monolayers have been widely used as biomimetic model systems to investigate molecular-level interactions of biologically relevant molecules such as peptides and pharmaceutical drugs with cell membranes.1 Despite their simplicity compared to the biological systems, Langmuir monolayers may provide important insight into molecular interactions because the distance between molecules can be easily controlled by compression compared to the 3D systems. In recent works, an interesting feature emerged that is associated with the cooperative response of phospholipid monolayers to the presence of guest molecules, where a small concentration of the drug or peptide was already sufficient to cause measurable changes in * Corresponding authors. (N.N.) E-mail:
[email protected]. Fax: 91-1596-244183. (D.V.) E-mail:
[email protected]. Fax: 49-331567-9202. (1) (a) Fitzgerald, G. A. N. Engl. J. Med. 1987, 316, 1247-1257. (b) Steimberg, D. Circulation 1992, 85, 2338-2343. (c) Iuliano, L.; Violi, F.; Ghiselli, A.; Alessandri, C. A.; Balsano, D. Lipids 1989, 24, 430-433. (d) Iuliano, L.; Pratico, D.; Ghiselli, A.; Bonavita, M.; Violi, F. Lipids 1992, 27, 349-353. (e) Iuliano, L.; Colavita, A. R.; Pratico, D.; Violi, F. Free Radical Biol. Med. 1997, 22, 999-1006. (f) Nepomuceno, M.; Alonso, A.; Pereira-da-Silva, L.; Tabak, M. Free Radical Biol. Med. 1997, 23, 1046-1054. (g) Nepomuceno, M. F.; Mamede, M. E. O.; Macedo, D. V.; Alves, A. A.; Pereira-da-Silva, L.; Tabak, M. Biochim. Biophys. Acta 1999, 1418, 285-294. (h) Borissevitch, I. E.; Borges, C. P. F.; Yushmanov, V. E.; Tabak, M. Biochim. Biophys. Acta 1995, 1238, 57-62. (i) Borissevitch, I. E.; Borges, C. P. F.; Borissevitch, G. P.; Yushmanov, V. E.; Louro, S. R.; Tabak, M. Z. Naturforsh., C 1996, 51, 578-590. (j) Nassar, P. M.; Almeida, L. E.; Tabak, M. Langmuir 1998, 14, 6811-6817. (k) Nassar, P. M.; Almeida, L. E.; Tabak, M. Biochim. Biophys. Acta 1997, 1328, 140-150. (l) Davies, L. P.; Huston, V. Eur. J. Pharmacol. 1981, 73, 209-211. (m) Chen, H.; Bamberger, U.; Heckel, A.; Guo, X.; Cheng, Y. Cancer Res. 1993, 53, 19741977. (n) Ramu, N.; Ramu, A. Int. J. Cancer 1989, 43, 487-491.
monolayer properties.2-4 An example was the interaction of dipyridamole (2,6-bis(diethanolamine)-4,8-dipiperidinopyrimido[5,4-d] pyrimidine) (DIP), known as a coronary vasodilator, antioxidant, and coactivator of antitumor activity and other drug effects,1 with L-R-1,2-dipalmitoyl-sn-3-glycero-phosphatidyl choline (DPPC).2,3 The properties of DPPC monolayers have been well documented in the literature,5-11 whereas the incorporation of DIP in DPPC monolayers was investigated using surface pressure-area isotherms, grazing incidence X-ray diffraction (GIXD),2 surface potential measurements, fluorescence microscopy, and Fourier transform infrared reflection absorption spectroscopy (FT-IRRAS).3 These experimental studies revealed that the presence of DIP markedly changes the properties of a pure DPPC monolayer because of the above-mentioned cooperative response. Moreover, the effects from DIP depend strongly on its relative concentration, cDIP, when co-spread with DPPC. This is the case of the surface pressure at which the liquid(2) Haas, H.; Caetano, W.; Borissevitch, G. P.; Tabak, M.; Mosquera Sanchez, M. I.; Oliveira, O. N., Jr.; Scalas, E.; Goldman, M. Chem. Phys. Lett. 2001, 335, 510-516. (3) Caetano, W.; Ferreira, M.; Tabak, M.; Mosquera Sanchez, M. I.; Oliveira, O. N., Jr.; Kru¨ger, P.; Schalke, M.; Lo¨sche, M. Biophys. Chem. 2001, 91, 21-35. (4) Oliveira, O. N., Jr.; Riul, A., Jr.; Leite, V. B. P. Braz. J. Phys. 2004, 34, 73-83. (5) Mo¨hwald, H. Annu. ReV. Phys. Chem. 1990, 41, 441-476. (6) McConnell, H. M. Annu. ReV. Phys. Chem. 1991, 42, 171-195. (7) Nandi, N.; Vollhardt, D. Chem. ReV. 2003, 103, 4033-4075. (8) Nandi, N.; Vollhardt, D. J. Phys. Chem. B 2002, 106, 10144-10149. (9) Kru¨ger, P.; Lo¨sche, M. Phys. ReV. E 2000, 62, 7031-7043. (10) Dahmen-Levison, U.; Brezesinski, G.; Mo¨hwald, H. Thin Solid Films 1998, 327, 616-620. (11) Brezesinski, G.; Dietrich, A.; Struth, B.; Bo¨hm, C.; Bouwman, W. G.; Kjaer, K.; Mo¨hwald, H. Chem. Phys. Lipids 1995, 76,145-157.
10.1021/la0602416 CCC: $33.50 © 2006 American Chemical Society Published on Web 05/12/2006
Dipolar Interaction between DIP and DPPC
Figure 1. Normalized surface potential relative to the pure DPPC monolayer vs the concentration of DIP, with the surface potential being taken in the condensed phase. Adapted from ref 3.
expanded (LE) to liquid-condensed (LC) phase transition starts. The ratio between the onset pressure for the mixed and pure DPPC, πrel ) πc(DIP-DPPC)/π(DPPC), gradually decreases from 1 to 0.75 for cDIP in the range from 0 to 0.25 mol % and then increases back to unity. Similarly, the relative molecular area, ∆Arel ) ∆A(DIP-DPPC)/∆A(DPPC), shows anomalous variation as a function of surface pressure for all cDIP values ranging from 1 to 15 mol %. ∆Arel increases to a maximum value of 1.08 for a surface pressure of 10 mN/m and then gradually decreases to 1.0.2,3 The same applies to the monolayer surface potential: taking the surface potential at the onset of the LE/LC phase transition, ∆Vrel ) ∆V (DIP-DPPC)/∆V (DPPC) increases to 1.075 for cDIP ) 0.5 mol % and sharply decreases to approximately 0.925 and retains this steady low value for higher cDIP, as is illustrated in Figure 2b3 (displayed in Figure 1). Another piece of evidence coming from the FTIR data was that DIP interacts preferentially with the phosphate group of the zwitterion in the DPPC molecule, which was expected because DIP is protonated at the pH of the experiments.3 It is clear that some sort of recognition process of DIP by DPPC on the molecular length scale is responsible for the observation of these macroscopic anomalies. DIP effects are also induced on the mesoscopic length scales, as demonstrated with fluorescence microscopy images indicating that the domains in the DIP-DPPC system grow a larger number of arms from the nucleation center compared to the number in the pure DPPC monolayer. In fact, at high pressures triskelion-shaped domains are observed. The LE/LC boundary lines are more elongated in the DIP-DPPC monolayer than in the pure DPPC monolayer, which is related to the development of spikes with the increase in surface pressure. In summary, the incorporation of DIP induces changes in the monolayer properties at all length scales, from macro- to mesoscopic dimensions. The molecular recognition of the DIP and DPPC systems raises questions about the concentration dependence of the surface pressure and surface potential isotherms of the mixed DIP-DPPC monolayers and the appearance of spikes in the domains. In this article, we use semiempirical quantum mechanical calculations for the DIP-DPPC system to understand the molecular recognition process and address these issues on the basis of the dipolar interaction. The details of the theoretical calculations of dipolar interactions are presented in the next section, and the results are analyzed in section 3, followed by concluding remarks.
2. Theoretical Calculations The conformations of DPPC and DIP molecules are energy minimized at the semiempirical PM3 level using MOPAC in
Langmuir, Vol. 22, No. 12, 2006 5399
CHEM3D software.12,13 Charges were obtained using the Mulliken population analysis (MPA) technique.14 Initially, five conformers of uncharged, singly charged, and doubly charged DIP molecule are generated. We compared the results for the uncharged and protonated DIP with the literature,15 including the heat of formation of neutral and protonated species and the isosurface of HOMO for the neutral state. The heat of formation of the five DIP conformers varied from -140.92 to -146.24 kcal/mol, which is consistent with the quoted heat of formation of -145.05 kcal/mol.15 For singly protonated DIP conformers, the heat of formation ranged from -6.76 to -11.10 kcal/mol, to be compared with -11.42 kcal/mol from the literature.15 The heat of formation of five doubly protonated DIP conformers ranged from 202.11 to 216.38 kcal/mol. A total of ninety (90) energy-minimized conformers of DPPC were studied, all of which had an all-trans configuration for the alkyl tails but with different conformations in the headgroup region. Because there is evidence from the literature that DIP is incorporated in the headgroup region rather than in the tail region,2,3 in this work the structure of DIP-DPPC systems was generated with DIP in the proximity of the headgroup. Ideally, one should consider a number of possible geometric arrangements for DIP in relation to the DPPC headgroup, but the computational cost would be prohibitive. One can, however, systematically study how the DIP-DPPC monolayer differs from a pure DPPC monolayer by considering three limiting cases, as follows. In the first arrangement, the plane containing the rings of the DIP molecule is collinear with the DPPC alkyl chains (Figure 2a), which is referred to here as a “vertical” arrangement. Second, the plane containing the rings of the DIP molecule is perpendicular to the DPPC alkyl chains (Figure 2c), which is denoted as a “lateral” arrangement. Finally, the plane containing the rings of the DIP molecule could be between the vertical and lateral arrangements with the DPPC alkyl chains (Figure 2b), referred to as an “intermediate” arrangement. Obviously, this classification for the mutual arrangements between DIP and DPPC is arbitrary, but the final energy-minimized structures are unbiased by the classification, and the conclusions reached in the present article are independent of the classification. In addition to calculating the heat of formation, which was used as a benchmark to validate the calculations with the semiempirical method because data are available in the literature, we calculated the total dipole moment and its vertical component of the 90 energy-minimized conformations for DIP-DPPC. Obtaining the vertical component, µ⊥, was important because the measured surface potential, ∆V, is given as16
∆V )
µ⊥ 0A
where A is the area per molecule and 0 is the vacuum permittivity.
3. Results and Discussion The plot of the ratio between the normal components for the DIP-DPPC and DPPC molecules is shown in Figure 3 for the three types of arrangements, with DIP being assumed to be singly protonated according to the experimental results.2,3 A considerable (12) CHEM3D Ultra, version 8.0; Cambridge Soft Corporation, Cambridge, MA, 2003. (13) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209-220. (14) Jensen, F. Introduction to Computational Chemistry; Wiley: Chichester, England, 1999; pp 81-99. (15) Alves, C. N.; Castilho, M.; Mazo, L. H.; Tabak, M.; da Silva, A. B. F. Chem. Phys. Lett. 2001, 349, 146-152. (16) Dynarowicz-Latka, P.; Cavali, A.; Silva Filho, D. A.; dos Santos, M. C.; Oliveira, O. N., Jr. Chem. Phys. Lett. 2000, 326, 39-44.
5400 Langmuir, Vol. 22, No. 12, 2006
Thirumoorthy et al.
Figure 2. Schematic representation of possible types of mutual packing arrangements of the DIP-DPPC system. Three arrangements are considered: (a) vertical, (b) intermediate, and (c) lateral. In each case, the image on the left corresponds to a pair of DIP-DPPC molecules with tails directed normal to the air/water interface. The types of arrangements (e.g., vertical, intermediate, and lateral) refer to the relative arrangement of the plane containing the rings of the DIP molecule relative to the normal direction of tails of the DPPC molecule. The middle image corresponds to the arrangement of the DIP-DPPC pair when the tails of DPPC are tilted per the experimentally observed tilt at air-water interface, keeping the relative orientation of DIP the same as that in the left image. The right image corresponds to an aggregate composed of three DIP-DPPC pairs. All calculations of dipole moments are based on a pair of DPPC and DIP.
Dipolar Interaction between DIP and DPPC
Figure 3. Ratio of the normal component of the dipole moment between DIP-DPPC and pure DPPC for (a) vertical-type, (b) intermediate-type, and (c) lateral-type conformers.
rise is observed for almost all conformers, especially considering the absolute values of the dipole moment. This is due to the zwitterion in DPPC, where small changes in charge separation may lead to large changes in the dipole moment. We now compare these theoretical predictions with the experimental data from ref 3, displayed in Figure 1. In the latter, the surface potential (and therefore µ⊥) increases for low DIP concentrations and then decreases at larger concentrations. This change in behavior was attributed to the self-aggregation of DIP molecules as its relative concentration increased.2,3 A consistent
Langmuir, Vol. 22, No. 12, 2006 5401
description of the conformation of the DIP-DPPC system suggests that DIP positioning could change with concentration, with DIP adopting a lateral or intermediate arrangement at higher concentrations. Because in all three arrangements the effect of DIP is to increase the dipole moment for most conformers, the increased surface potential at very low DIP concentrations is readily explained. From the data in Figure 3, one may conclude that a vertical arrangement should be preferred at low DIP concentrations, though the intermediate or lateral arrangements cannot be discarded. The area per molecule is expected to increase because of the enhanced electrostatic repulsion in comparison with DPPC, as is experimentally observed. However, at high DIP concentrations a vertical arrangement is unlikely because of the difficulties in accommodating the large vertical dipole moment component of DIP, and DIP appears to flip to a lateral (or intermediate) arrangement, probably in one of the conformers with a reduced normal component of the dipole moment. Having demonstrated that dipolar interactions between DIP and DPPC may lead to large changes in the total vertical component of the dipole moment (and hence in the measured surface potential), we now analyze possible implications from these interactions to the film morphology. It is well established that the domain morphology is essentially determined by three major factors: (1) line tension at the domain boundary, (2) dipolar repulsion between molecules, and (c) molecular chirality.7 The incorporation of DIP is not expected to affect the chirality-induced features of DPPC monolayer such as the curvature of domain arms and their handedness. Indeed, such effects are not observed in the experimental data.3 Changes in the line tension due to the presence of DIP are difficult to analyze because we lack detailed information at the molecular level about the domain boundaries. With regard to the second factor, viz., dipolar repulsion, the theoretical calculations presented here point to an increased dipolar repulsion caused by the incorporation of DIP, which is consistent with the development of spikes from the end of triskellions because the growth of more elongated domains is a signature of enhanced electrostatic repulsion. One should keep in mind the simplifications made in this work. First, DPPC and DIP were made to interact in a vacuum, whereas in practice these molecules are surrounded by water molecules whose structure may vary considerably depending on whether the water molecules are in the hydration shell or not. This also means that in estimating the changes in surface potential caused by changes in dipole moment one needs to consider polarization effects, which was not carried out here. However, the effect of water in such a case could be only to screen the strength of the dipolar interaction between DIP and DPPC with the solvent dielectric constant at the interface as a scale factor. Furthermore, it was not possible to directly compare the stability or the dipolar magnitudes of the DIP-DPPC aggregates at the same concentrations studied in the experiments. The latter are carried out at rather low DIP concentrations, which is equivalent to a large number of DPPC molecules per DIP molecule. Calculations of thermodynamic parameters or dipole moments for a large number of DPPC molecules are computationally very expensive using quantum chemical methods. In future work, molecular dynamics simulations could be used for this purpose. However, because of the long-range nature of dipolar interactions, it is expected that the conclusions made on the factors mentioned previously, namely, the effect of molecular orientations in the recognition process and the changes in dipole moment, which are based on pair calculations, may remain valid for infinite systems.
5402 Langmuir, Vol. 22, No. 12, 2006
4. Conclusions The large changes in the vertical dipole moment caused by the incorporation of DIP in the DIP-DPPC system, predicted by the semiempirical quantum calculations (PM3 level) presented here, may explain why measurable changes in surface potential are observed experimentally even at very low DIP concentrations (0.2-0.5 mol %).3 The theoretical results also point to a dependence on the concentration of DIP. Rather than saturation or a continuous increase in the effects caused by the incorporation of increasing amounts of DIP, the inversion in behavior as the DIP concentration exceeds 0.5 mol %, observed experimentally, would be caused by a change in DIP conformation, where DIP
Thirumoorthy et al.
molecules are expected to adopt a lateral or intermediate arrangement. Finally, the strong dipolar interactions indicated in the calculations may be the origin of the drastic changes in monolayer morphology as seen in fluorescence microscopy images. Acknowledgment. This work is supported by a grant from CSIR (India) to N.N. K.T thankfully acknowledges CSIR (India) for a Senior Research Fellowship, and O.N.O. is supported by Fapesp and CNPq (Brazil). Partial support from the Alexander von Humboldt foundation is thankfully acknowledged. LA0602416
