Fibrinogen Layer

The squeeze-out of fibrinogen evidently removed a pronounced amount of DPPC from the interface, as judged from the corresponding νa-CH2 intensity and...
0 downloads 0 Views 128KB Size
Download PDF
Langmuir 2006, 22, 6629-6634

6629

Infrared Spectroscopy Analysis of Mixed DPPC/Fibrinogen Layer Behavior at the Air/Liquid Interface under a Continuous Compression-Expansion Condition Chia-Lin Yin and Chien-Hsiang Chang* Department of Chemical Engineering, National Cheng Kung UniVersity, Tainan, Taiwan 70101, ROC ReceiVed April 4, 2006. In Final Form: May 9, 2006 The mixed layer behavior of dipalmitoyl phosphatidylcholine (DPPC) with fibrinogen at continuously compressedexpanded air/liquid interfaces was analyzed in situ by infrared reflection-absorption spectroscopy (IRRAS). The reflectance-absorbance (RA) intensities and/or wavenumbers of νa-CH2 and amide I bands for a mixed DPPC/ fibrinogen layer at the interface were obtained directly by an infrared spectrometer with a monolayer/grazing angle accessory and a removable Langmuir trough. The νa-CH2 RA intensity-area hysteresis curves of a DPPC monolayer indicate a significant loss of free DPPC molecules at the interface during the first compression stage, which is also supported by the corresponding νa-CH2 wavenumber-area hysteresis curves. For a mixed DPPC/fibrinogen layer at the interface, the amide I RA intensity-area hysteresis curves suggest that the fibrinogen molecules were expelled from the interface upon compression, apparently because of the presence of insoluble DPPC molecules. The squeezeout of fibrinogen evidently removed a pronounced amount of DPPC from the interface, as judged from the corresponding νa-CH2 intensity and wavenumber data. Moreover, significant adsorption of fibrinogen was found during the subsequent interface expansion stage. With the in situ IRRAS analysis of the mixed layer behavior at the interface, the induced loss of DPPC by fibrinogen expulsion from the compressed interface and the dominant adsorption of fibrinogen to the expanded interface were clearly demonstrated.

Introduction Lung surfactants are the required materials for stabilizing the lung alveoli in breathing. The specific interfacial properties of lung surfactants at the air/liquid interface of alveolar lining layers include the abilities to reduce dynamic surface tension effectively, to spread readily at the expanded interface, and to adsorb quickly to the interface.1 Among the components of lung surfactants, dipalmitoyl phosphatidylcholine (DPPC) is the major lipid compound, and its interfacial behavior plays a key role in the specific functions of lung surfactants.2 It has been found that plasma proteins, such as albumin, γ-globulins, and fibrinogen, have strong inhibitory effects on the dynamic surface activity of lung surfactant systems, which may occur by the introduction of plasma proteins to alveolar space as a result of lung injury.3,4 For mixed spread DPPC/ albumin monolayers at interfaces, a considerable change in the monolayer composition might result under dynamic interface compression-expansion conditions, probably caused by proteinresidue rejection. When the mixed monolayers were compressed to reach very high surface pressures, DPPC collapsed and left the interface irreversibly.5 It has been reported that plasma proteins seemed to inhibit lung surfactant functions by competing with the space at the interface and thus preventing the phospholipids from adsorbing onto the interface.6-8 Moreover, plasma proteins * To whom correspondence should be addressed. E-mail: changch@ mail.ncku.edu.tw. Tel: (+) 886-6-275-7575 ext. 62671. Fax: (+) 886-6234-4496. (1) Notter, R. H.; Finkelstein, J. N. J. Appl. Physiol.: Respir. EnViron. Exercise Physiol. 1984, 57, 1613. (2) Keough, K. H. W. In Pulmonary Surfactant: From Molecular Biology to Clinical Practice; Robertson, B., van Glode, L. M. G., Batenburg, J. J., Eds.; Elsevier: New York, 1992. (3) Colacicco, G.; Basu, M. K. Respir. Physiol. 1978, 32, 265. (4) Gunther, A.; Seeger, W. In Surfactant Therapy for Lung Disease; Robertson, B., Taeusch, H. W., Eds.; Marcel Dekker: New York, 1995. (5) Taneva, S.; Panaiotov, I.; Ter-Minassian-Saraga, L. Colloids Surf. 1984, 10, 101.

might leave the interface and carry along surfactant lipids with them upon compression of the interface.9 For mixed layers of DPPC with γ-globulins or fibrinogen at air/liquid interfaces, it has been suggested from the surface pressure-area hysteresis isotherms that plasma protein molecules were expelled from the interfaces during the compression stage, inducing the loss of free DPPC molecules at interfaces.10,11 However, in most of the previous studies regarding the mechanisms involved in the surface activity inhibition of lung surfactant systems by plasma proteins, equilibrium/dynamic surface tensions and/or surface pressure-area monolayer isotherms of the mixtures were mainly measured, and the possible mechanisms were then proposed. Since infrared reflectionabsorption spectroscopy (IRRAS) was developed, it has become a unique technique for studying the mixed lipid/protein monolayer behavior at the air/liquid interface in situ, especially for the lung surfactant systems.12-16 Recently, the IRRAS technique has been successfully applied to probe the competitive adsorption behavior of lipids and plasma proteins directly at the air/liquid interface.17-21 It has been (6) Holm, B. A.; Enhorning, G.; Notter, R. H. Chem. Phys. Lipids 1988, 49, 49. (7) Cheng, C.-C.; Chang, C.-H. Langmuir 2000, 16, 437. (8) Liu, Y.-L.; Chang, C.-H. Colloid Polym. Sci. 2002, 280, 683. (9) Keough, K. M. W.; Parson, C. S.; Phang, P. T.; Tweeddale, M. G. Can. J. Physiol. Pharmacol. 1988, 66, 1166. (10) Chang, C.-H.; Yu, S.-D.; Chuang, T.-K.; Liang, C.-N. J. Colloid Interface Sci. 2000, 227, 461. (11) Kuo, R.-R.; Chang, C.-H.; Yang, Y.-M.; Maa, J.-R. J. Colloid Interface Sci. 2003, 257, 108. (12) Dluhy, R. A.; Cornell, D. G. J. Phys. Chem. 1985, 89, 3195. (13) Dluhy, R. A. J. Phys. Chem. 1986, 90, 1373. (14) Bi, X.; Flach, C. R.; Pe´rez-Gil, J.; Plasencia, I.; Andreu, D.; Oliveira, E.; Mendelsohn, R. Biochemistry 2002, 41, 8385. (15) Flach, C. R.; Cai, P.; Dieudonne, D.; Brauner, J. W.; Keough, K. H. W.; Stewart, J.; Mendelsohn, R. Biophys. J. 2003, 85, 340. (16) Shanmukh, S.; Biswas, N.; Waring, A. J.; Walther, F. J.; Wang, Z. D.; Chang, Y.; Notter, R. H.; Dluhy, R. A. Biophys. Chem. 2005, 113, 233. (17) Wen, X.; Franses, E. I. Colloids Surf., A 2001, 190, 319. (18) Phang, T.-L.; Franses, E. I. J. Colloid Interface Sci. 2004, 275, 477.

10.1021/la060895e CCC: $33.50 © 2006 American Chemical Society Published on Web 06/23/2006

6630 Langmuir, Vol. 22, No. 15, 2006

demonstrated from IRRAS analysis that for a DPPC/albumin mixture the fast adsorption of albumin could interfere with the adsorption of DPPC, thus inhibiting DPPC surface activity.17 A similar inhibitory effect has been found for a DPPC/fibrinogen mixture.21 However, these IRRAS studies were performed only under a constant-area condition. To further elucidate the mixed lipid/plasma protein layer behavior at interfaces under a condition simulating the movement of a lung alveolus to some extent, an in situ IRRAS analysis of the mixed DPPC/fibrinogen layer behavior at the cyclic air/ solution interface is performed in this study. From the intensity and wavenumber of characteristic peaks for DPPC and fibrinogen in IRRAS measurements under a cyclic-area condition, the composition variations of the mixed DPPC/fibrinogen layer at the cyclic interface could be successfully followed, which has not been demonstrated before. The specific objective of this study is to examine the possible inhibition mechanism of DPPC dynamic surface activity by fibrinogen at the air/solution interface under a continuous compression-expansion condition with the IRRAS technique. Experimental Section Bovine fibrinogen and L-R-dipalmitoyl phosphatidylcholine (DPPC) (>99%) were supplied by Sigma Chemical Co. and were used without further purification. The fibrinogen powder contains 70 wt % fibrinogen, 20 wt % sodium chloride, and 10 wt % sodium citrate. Ethanol (∼99.5%) was purchased from Seoul Chemical Industry Co., Ltd. (Korea) and n-hexane (>99%) was obtained from Tedia Co. An ethanol/n-hexane (1:9 v/v) mixture was used as the spreading solvent for DPPC at air/liquid interfaces. The water used in all experiments was purified by means of a Milli-Q plus water purification system (Millipore) with a resistivity of 18.2 MΩ‚cm. A buffer solution of NaH2PO4/Na2HPO4 with a pH value of 7.0 was used to prepare the aqueous subphases for IRRAS experiments. An IRRAS analysis was conducted using a Perkin Elmer FTIR spectrometer (model Spectrum GX) with a liquid-nitrogen-cooled mercury cadmium telluride (MCT) detector. A monolayer/grazing angle accessory (P/N 19650 series, Specac Inc.) with a removable Teflon Langmuir trough was used with the FTIR spectrometer to obtain the IR spectrum of a monolayer at the air/liquid interface. Measurements were performed at room temperature. The IR spectra were taken using unpolarized light with an angle of incidence of 40°, as measured perpendicular to the liquid surface. The sample chamber of the FTIR spectrometer was continuously purged with dry air from a purge gas generator. Spectra were collected at 8-cm-1 resolution with a scan number of 128. All of the spectra were reported by subtracting the buffer spectrum from the measured spectra. When the IRRAS analysis of a DPPC monolayer at a continuously compressed-expanded air/liquid interface was conducted, the removable Langmuir trough of the monolayer/grazing angle accessory was filled with 8 mL of buffer solution with a pH value of 7.0, and a 6.1-µL sample containing 500 ppm DPPC was spread at the interface by using a 10-µL microsyringe (Hamilton Co.). After a period of 20 min was allowed for solvent evaporation, the spread monolayer was continuously compressed and then expanded by a barrier at a rate of 0.00875 nm2/molecule‚min. The minimum relative area for the compression stage was controlled at 52.5%. During the interface compression-expansion cycles, IRRAS measurements of the monolayer were continuously performed with an acquisition time of 72-75 s/spectrum. Within each IRRAS spectrum acquisition period, there is no indication of significant signal interference caused by the monolayer composition or surface concentration variation under the interface compression or expansion condition. The compression or expansion rate is much slower than that under (19) Phang, T.-L.; McClellan, S. J.; Franses, E. I. Langmuir 2005, 21, 10140. (20) Kim, S. H.; Franses, E. I. Colloids Surf., B 2005, 43, 256. (21) Kim, S. H.; Franses, E. I. J. Colloid Interface Sci. 2006, 295, 84.

Yin and Chang

Figure 1. IRRAS spectra of (a) the amide I peak at ∼1650 cm-1 for the preadsorbed fibrinogen layer formed from a 700 ppm fibrinogen solution with an adsorption time of 5 h and (b) the antisymmetric methylene stretching vibration (νa-CH2) at ∼2920 cm-1 for DPPC in the mixed DPPC/fibrinogen layer on a 700 ppm fibrinogen subphase. physiological conditions. However, a higher rate may result in unavoidable signal interference due to interface perturbation and monolayer composition or surface concentration variation during the spectrum acquisition. To perform the IRRAS analysis for a mixed DPPC/fibrinogen layer, a 6.1-µL sample containing 500 ppm DPPC was spread at the air/fibrinogen solution interface, which had been equilibrated for 5 h. For the experiments involving DPPC, an amount of DPPC was spread at the interface with an initial area per molecule of 0.704 nm2. After the solvent evaporated, IR spectra of the mixed layer at the interface were taken under a continuous interface compressionexpansion condition. The IRRAS spectra of the amide I peak at ∼1650 cm-1 for the preadsorbed fibrinogen layer and the antisymmetric methylene stretching vibration (νa-CH2) at ∼2920 cm-1 for DPPC in the mixed layer acquired at the beginning of the first compression stage are demonstrated in Figure 1a and b, respectively. For the amide I peak, it appears that the signal noise caused by water vapor became significant with increasing experimental time.

Results and Discussion Hysteresis Behavior of a Spread DPPC Monolayer. The hysteresis behavior of a spread DPPC monolayer at the cyclic air/water interface was analyzed by IRRAS and is illustrated as a plot of reflectance-absorbance (RA) intensity versus relative area (Figure 2). The RA intensity is equal to -log (R/Ro), where R and Ro are the reflectivities of the monolayer-covered and pure buffer surfaces, respectively. The original IRRAS data were plotted as RA intensity versus wavenumber, and the RA intensity data reported in Figure 2 were obtained with an uncertainty of 0.0002 from a major IR peak of DPPC, the antisymmetric methylene stretching vibration (νa-CH2), at ∼2920 cm-1.13,22 The absorption bands of monolayers at air/liquid interfaces are negative, and the basis for the negative absorbances in the monolayer spectra have been explored.12,13 The initial absolute RA value of the νa-CH2 band for a DPPC monolayer at 0.704 nm2/molecule was 0.0028. Upon interface compression, the absolute RA value increased gradually and reached 0.0049 at ∼0.37 nm2/molecule. This is expected because a higher surface concentration of molecules always results in a (22) Mendelsohn, R.; Brauner, J. W.; Gericke, A. Annu. ReV. Phys. Chem. 1995, 46, 305.

IR Analysis of Mixed DPPC/Fibrinogen Layer BehaVior

Figure 2. Consecutive νa-CH2 RA intensity-relative area hysteresis curves of a spread DPPC monolayer with an initial area per molecule of 0.704 nm2. The relative area represents the ratio of the actual interfacial area during the compression-expansion cycle to the initial interfacial area.

stronger RA response. During the first interface expansion stage, the absolute RA value kept decreasing and dropped to 0.0017 under the fully interface expanded condition. The decrease of the absolute RA value from 0.0028 to 0.0017 after the first compression-expansion cycle implies a loss of free DPPC molecules during the cycle. This is consistent with what would be expected from the poor respreading ability of DPPC after monolayer collapse.10,11,23 At the end of the second compression stage, an absolute RA value of 0.0039 was detected. This value is lower than the value detected at the end of the first compression stage (0.0049), indicating fewer free DPPC molecules at the interface. Thus, the loss of free DPPC molecules during the first compressionexpansion cycle was further confirmed. Moreover, similar absolute RA values at the beginning and at the end of the second or third cycles suggested that no more significant loss of free DPPC molecules occurred during the two cycles. It appears that at the end of the first compression stage excess DPPC molecules were removed because of monolayer collapse and the rest of DPPC molecules formed a stable monolayer at the compressed interface. Because of the poor respreading ability of DPPC, no more excess DPPC molecules could be found during the second and third cycles. That is, a pronounced irreversible loss of the initially spread free DPPC molecules occurred only upon the first interface compression under the experimental conditions. In a previous study performed under slightly different experimental conditions, it has also been shown that a significant loss of free DPPC molecules was found in the first cycle, which is apparently due to monolayer collapse.11 It is noted that the hysteresis behavior of a monolayer, which is usually demonstrated by a conventional surface pressure-area hysteresis isotherm (23) Notter, R. H. In Pulmonary Surfactant; Robertson, B., van Golde, L. M. G., Batenburg, J. J., Eds.; Elsevier: Amsterdam, 1984.

Langmuir, Vol. 22, No. 15, 2006 6631

Figure 3. Consecutive νa-CH2 wavenumber-relative area hysteresis curves of a spread DPPC monolayer with an initial area per molecule of 0.704 nm2.

measured under continuous interface compression-expansion conditions, can also be characterized with an RA intensity-area hysteresis curve via real-time IRRAS analysis. The corresponding νa-CH2 wavenumber-relative area hysteresis curves of a DPPC monolayer are shown in Figure 3. At the beginning of the first interface compression-expansion cycle, the maximum in the νa-CH2 band was detected at a wavenumber of 2923 cm-1. The wavenumber kept shifting to lower values upon interface compression, and a value of 2918 cm-1 was found at the end of the first compression stage. According to the surface pressure-area isotherms of a DPPC monolayer reported before, the monolayer was compressed from a liquid-expanded state to a solid/collapse state during the first compression stage.10,11 It has been reported that the wavenumber corresponding to the maximum in the νa-CH2 band is sensitive to the packing and conformational change in the molecular hydrocarbon chains in the monolayer, and a lower wavenumber is a characteristic of the highly ordered conformation of the acyl chains.22,24-26 As shown in Figure 3, the νa-CH2 wavenumber shifted from an initial value of 2923 cm-1 to a lower value of 2918 cm-1 during the first compression stage, indicating that the DPPC molecules became close-packed or well-oriented upon interface compression. It is interesting that at the end of the first compressionexpansion cycle the maximum in the νa-CH2 band was shifted to 2925 cm-1, as compared with the initial value of 2923 cm-1. The change in the νa-CH2 wavenumber is probably related to less order in the free DPPC molecules at the end of the first cycle. Because a significant loss of free DPPC molecules occurred during the first cycle, the remaining free DPPC molecules at the interface became more disordered with a lower surface density, resulting in a higher νa-CH2 wavenumber. However, no pro(24) Mitchell, M. L.; Dluhy, R. A. J. Am. Chem. Soc. 1988, 110, 712. (25) Gericke, A.; Hu¨hnerfuss, H. J. Phys. Chem. 1993, 97, 12899. (26) Wen, X.; Lauterbach, J.; Franses, E. I. Langmuir 2000, 16, 6987.

6632 Langmuir, Vol. 22, No. 15, 2006

Yin and Chang

Figure 4. Dynamic amide I RA intensity behavior at the interface of an air/aqueous phase containing 700 ppm fibrinogen.

nounced change in the wavenumber corresponding to the maximum in the νa-CH2 band was found at the beginning and at the end of the second and third cycles. This result again indicated that a pronounced loss of free DPPC molecules occurred only during the first cycle but not during the second or third cycle. Hysteresis Behavior of an Adsorbed Fibrinogen Layer. For a 700 ppm fibrinogen solution, the dynamic adsorption behavior of fibrinogen molecules at the air/solution interface was monitored directly by IRRAS, and a plot of amide I RA intensity versus time is shown in Figure 4. The reported RA data with an uncertainty of 0.001 were obtained from the amide I peak at ∼1650 cm-1, which corresponds to the peptide CdO stretching mode.27 The adsorption of fibrinogen molecules at the interface led to the rise in the absolute RA value. It seems that the variation of the absolute RA value with time was still significant after 5 h, and an equilibrium adsorption layer might not be assumed in this case. However, a longer waiting period for equilibrium adsorption would be associated with a lower level of the subphase surface due to water evaporation and would cause a problem in the following measurements. Thus, an adsorption time of 5 h was allowed to form an adsorbed fibrinogen layer at the interface, and then the IRRAS analysis was performed under a continuous interface compression-expansion condition. The amide I RA intensity-relative area hysteresis curves of an adsorbed fibrinogen layer formed from a 700 ppm fibrinogen solution with an adsorption time of 5 h are shown in Figure 5. During the compression stage of the first cycle, the absolute RA value of the amide I band changed from an initial value of 0.011 to a maximum value of 0.014 for the minimum relative area of 52.5%. An increase in the RA intensity of the amide I band implies an increase in protein concentration at the interface.28 Apparently, the surface concentration of fibrinogen at the interface was increased upon interface compression. It is noted that the absolute RA values obtained during the interface expansion stage were always higher than those obtained during the interface compression stage. As shown in Figure 4, the initial adsorbed fibrinogen layer at the interface did not reach equilibrium adsorption, and the fibrinogen surface concentration would be lower than the equilibrium value. Thus, this extraordinary behavior was probably caused by the continuous adsorption of fibrinogen during the interface expansion stage with a tendency to reach equilibrium adsorption. During the second interface compression-expansion cycle, no regular pattern was found for the variations of RA intensity with relative area, which may reflect the complicated adsorption/desorption and/or spreading behavior of fibrinogen at the interface. It seems that a steady state for the adsorbed fibrinogen layer at the interface was reached, and the hysteresis behavior became insignificant in the third cycle. Hysteresis Behavior of a Mixed DPPC/Fibrinogen Layer. When DPPC molecules were spread with an initial area per (27) Herna´ndez, E. M.; Franses, E. I. Colloids Surf., A 2003, 214, 249. (28) Martin, A. H.; Cohen Stuart, M. A.; Bos, M. A.; van Vliet, T. Langmuir 2005, 21, 4083.

Figure 5. Consecutive amide I RA intensity-relative area hysteresis curves of an adsorbed fibrinogen layer formed from a 700 ppm fibrinogen solution with an adsorption time of 5 h.

Figure 6. Consecutive νa-CH2 RA intensity-relative area hysteresis curves of a mixed DPPC/fibrinogen layer on a 700 ppm fibrinogen subphase.

molecule of 0.704 nm2 at the interface with a preadsorbed fibrinogen layer formed from a 700 ppm fibrinogen solution, an initial absolute νa-CH2 RA value of 0.0036 was detected (Figure 6). In comparison with the initial absolute RA value of 0.0028 detected for a pure DPPC monolayer (Figure 2), the higher initial absolute RA value was evidently attributed to the close packing of spread DPPC molecules due to the presence of preadsorbed

IR Analysis of Mixed DPPC/Fibrinogen Layer BehaVior

Langmuir, Vol. 22, No. 15, 2006 6633

Figure 7. Consecutive νa-CH2 wavenumber-relative area hysteresis curves of a mixed DPPC/fibrinogen layer on a 700 ppm fibrinogen subphase.

Figure 8. Consecutive amide I RA intensity-relative area hysteresis curves of a mixed DPPC/fibrinogen layer on a 700 ppm fibrinogen subphase.

fibrinogen molecules at the interface.29,30 The corresponding νaCH2 wavenumber-relative area data are shown in Figure 7. The close packing of DPPC molecules in the mixed layer at the interface was also suggested by the lower νa-CH2 wavenumber of 2920 cm-1 detected at the beginning of the first compressionexpansion cycle. A similar experiment has been performed by Kim and Franses21 with DPPC molecules spread at a higher surface density, and the preadsorbed molecules on the surface of a 750 ppm fibrinogen solution were almost expelled from the interface. In this study, DPPC molecules were spread at a lower surface density, and thus some fibrinogen molecules were still available at the interface, resulting in the closely packed DPPC molecules. The presence of fibrinogen molecules at the interface after DPPC molecules were spread could be confirmed from the measurable amide I RA intensity shown in Figure 8. For a few surfactant/protein or lipid/protein systems, it has been shown that the components might not be completely miscible and phase separation might be observed at the interface. However, it appears that the separate phases were homogeneously distributed in the mixed layers.31-34 Moreover, no assumption regarding DPPC and fibrinogen being miscible at the molecular level at the interface was required for IRRAS analysis. Therefore, the following IRRAS results were not affected by the phase separation, if there is any, in the mixed DPPC/fibrinogen layer. For the mixed DPPC/fibrinogen layer, the absolute νa-CH2 RA value increased from 0.0036 to 0.0044 during the first compression stage (Figure 6). As compared with the absolute

RA value of 0.0049 observed at the end of the first compression stage for a pure DPPC monolayer (Figure 2), the lower value of 0.0044 reported in Figure 6 implies that extra DPPC molecules were removed from the interface upon compression as a result of the presence of fibrinogen, resulting in a smaller number of free DPPC molecules at the interface. The corresponding consecutive amide I RA intensity-relative area hysteresis curves for the mixed DPPC/fibrinogen layer are shown in Figure 8. During the first compression stage, the absolute amide I RA value for the fibrinogen molecules at the interface always decreased, suggesting that a significant expulsion of fibrinogen molecules from the interface occurred upon interface compression. As compared with that observed in Figure 5, the expulsion of fibrinogen molecules was apparently due to the presence of insoluble DPPC molecules at the interface. Upon compression, the available interface space for DPPC and fibrinogen became limited, and thus the soluble fibrinogen molecules were forced to desorb. However, if one considers the absolute amide I RA value of 0.002 detected at the end of the compression stage, it can be concluded that some fibrinogen molecules still existed at the interface even though the rest of free DPPC molecules were close-packed. In addition, because the remaining fibrinogen molecules would occupy certain interface space at the end of the first compression stage, the number of free DPPC molecules at the compressed interface was even fewer than would be expected for a pure DPPC monolayer with an absolute νa-CH2 RA value of 0.0044. During the following interface expansion stage, the absolute νa-CH2 RA value decreased gradually from 0.0044 to 0.0018 (Figure 6). Though this value is very close to the value observed for a pure DPPC monolayer (0.0017, Figure 2), a smaller number of free DPPC molecules was expected if one takes fibrinogen adsorption into account. It has been proposed that fibrinogen molecules would adsorb at the interface during the interface expansion stage.11 The fibrinogen adsorption is clearly demon-

(29) Ivanova, M.; Panaiotov, I.; Trifonova, T.; Echkenazi, M.; Konstantinov, G.; Ivanova, R. Colloids Surf. 1984, 10, 269. (30) Cho, D.; Narsimhan, G.; Franses, E. I. Langmuir 1997, 13, 4710. (31) Mackie, A. R.; Gunning, A. P.; Wilde, P. J.; Morris, V. J. J. Colloid Interface Sci. 1999, 210, 157. (32) Rodriguez Patino, J. M.; Rodriguez Nino, M. R.; Sanchez, C. C.; Fernandez, M. C. J. Colloid Interface Sci. 2001, 240, 113. (33) Gunning, A. P.; Mackie, A. R.; Kirby, A. R.; Morris, V. J. Langmuir 2001, 17, 2013. (34) Rodriguez Patino, J. M.; Rodriguez Nino, M. R.; Sanchez, C. C.; Fernandez, M. C. Langmuir 2001, 17, 7545.

6634 Langmuir, Vol. 22, No. 15, 2006

strated in Figure 8, as judged from the increased absolute amide I RA value upon interface expansion. In addition, the wavenumber corresponding to the maximum in the νa-CH2 band for DPPC remained essentially unchanged at ∼2920 cm-1 during the interface expansion stage (Figure 7), implying that the DPPC molecules were always in a highly ordered state upon interface expansion. Because it is well known that DPPC has poor respreading ability, the close packing of the DPPC molecules was apparently related to the significant adsorption of fibrinogen molecules to the interface upon expansion. Under the circumstances, more fibrinogen molecules were available at the interface and occupied part of the space at the interface, enhancing the conformational order of the hydrocarbon chains of the remaining free DPPC molecules. During the second and third cycles, the absolute νa-CH2 RA values at the ends of the compression stages were 0.0043 and 0.0040, respectively. These lower RA values in comparison with the value detected in the first cycle (0.0044) suggest that some of the remaining free DPPC molecules after the first compression stage might be removed with fibrinogen during the two cycles. This is also supported by the lower absolute RA value detected at the ends of the cycles (0.0016), as compared with the value obtained at the end of the first cycle (0.0018). The intensity change from 0.0018 to 0.0016 seems insignificant, but the variation in the surface density of free DPPC molecules might be pronounced if one considers that the fibrinogen molecules adsorb at the interface and occupy certain interface space upon expansion. The significant fibrinogen adsorption during the interface expansion stage can be realized from Figure 8, in which the absolute amide I RA value of fibrinogen changed from 0.008 detected at the end of the first cycle to 0.010 detected at the end of the third cycle, indicating that more fibrinogen molecules were available at the interface.

Yin and Chang

Conclusions Mixed DPPC/fibrinogen layer behavior at a cyclic air/solution interface was probed directly by an IRRAS technique. For a pure DPPC monolayer, a significant loss of free DPPC molecules at the interface during the first compression-expansion cycle was demonstrated by the νa-CH2 RA intensity-area and wavenumber-area hysteresis curves. For a mixed DPPC/fibrinogen layer at the interface of the air/aqueous phase containing 700 ppm fibrinogen, the amide I RA intensity-area hysteresis curves imply that the fibrinogen molecules were expelled from the interface upon compression, apparently because of the presence of insoluble DPPC molecules. Moreover, the exclusion of fibrinogen molecules was accompanied by the loss of free DPPC molecules at the interface, as judged from the variations of the νa-CH2 RA intensity and wavenumber with the interfacial area. However, during the subsequent interface expansion stage, significant adsorption of fibrinogen molecules to the interface occurred. For a mixed DPPC/fibrinogen layer at the cyclic interface, fibrinogen expulsion upon interface compression, associated with the removal of DPPC, and fibrinogen adsorption at the interface upon interface expansion were detected and analyzed by the in situ IRRAS technique. The induced loss of DPPC by fibrinogen expulsion from the interface and the dominant adsorption of fibrinogen to the interface in a dynamic interface compressionexpansion process would result in pronounced depletion and thus surface-activity inhibition of DPPC at the interface. Acknowledgment. This work was supported by the National Science Council of the Republic of China through grants NSC912214-E006-023 and NSC92-2214-E006-010. LA060895E