in Consecutively Adsorbed Polyelectrolyte Multilayers on Texturized

solutions on both unmodified and unidirectionally scratched silicon substrates. ... of NaClO4 were found to be significantly aligned along the scratch...
0 downloads 0 Views 178KB Size
Download PDF
Biomacromolecules 2001, 2, 262-269

262

Orientation of r-Helical Poly(L-lysine) in Consecutively Adsorbed Polyelectrolyte Multilayers on Texturized Silicon Substrates M. Mu ¨ ller† Institute of Polymer Research Dresden, Hohe Strasse 6, D-01069 Dresden, Germany Received October 18, 2000; Revised Manuscript Received December 13, 2000

Alternate polyelectrolyte multilayers composed of poly(L-lysine) (PLL) in the R-helical state and poly(maleic acid-co-R-methylstyrene) (PMA-MS) were deposited by consecutively adsorbing from aqueous solutions on both unmodified and unidirectionally scratched silicon substrates. Thereby, the R-helical PLL rods formed in the presence of NaClO4 were found to be significantly aligned along the scratching direction of the substrate. From the dichroic ratios of the amide I and amide II band, obtained by ATR-FTIR spectroscopy, order parameters with respect to uniaxial orientation were determined using a cone distribution model and the known angles of the transition dipole moments relative to the helix axis. They ranged from S ) 0.5-0.7 for the PLL rods oriented on the texturized Si substrate and S ) 0.0-0.2 (S ) 1 for perfect ordering) for the unoriented PLL rods in the dry and wet states, respectively. Additionally, multilayers composed of PLL and poly(vinyl sulfate) showed similar uniaxial alignment on texturized Si substrates (S ) 0.6). Generally, drying the multilayers caused a certain loss of order. Surfaces of anisotropically oriented R-helical polypeptides could be interesting for biomimetic purposes. Introduction Adsorbed layers of stiff R-helical polypeptides at surfaces seem to be an interesting surface modification concept to create biomimetic or biorecognizable surface structures, since the right-handed R-helix is a prominent and characteristic secondary structure for many proteins. Recently, films of poly(γ-methyl glutamate-co-γ-octadecyl glutamate) in the R-helical state were reported to be oriented on unidirectionally patterned silicon substrates, which was proven by dichroic ATR-FTIR spectroscopy using polarized light.1,2 In analogy to these hydrophobically modified polypeptides, we were now interested in the orientation behavior of hydrophilic charged polypeptides, like poly(L-lysine) (PLL), whose ability to adopt various conformations in both solution and at the surface by change of external parameters like pH, ionic strength, and solvent is well-known.3,4 In particular, PLL adopts an R-helical conformation in the presence of certain low molecular mineral anions,5 like ClO4-, which are situated in the far right limit of the well-known Hofmeister series (Cl- > Br- > ... > SCN- > ClO4-), featuring in increasing order the ability to create a chaotic water structure in the neighborhood of dissolved polar (charged) molecules. The induction of the R-helical PLL conformation by NaClO4 has not been fully understood up to now. Ebert5 claimed an exterior left-handed helix of ClO4anions as a stabilizing supramolecular template for the righthanded R-helical PLL conformation. In that respect, PLL is an interesting exception from conventional synthetic polyelectrolytes, since there is experimental and theoretical evidence that polyelectrolytes in the presence of salt form †

Fax: 0351/4658284. E-mail: [email protected].

coiled structures due to the screening of the monomer charges, which diminishes the electrostatically driven charge repulsion between charged monomers (e.g., refs 6 and 7). In the present paper, we report on the deposition and arrangement of the charged model polypeptide PLL from both saltless and NaClO4-containing solutions at unmodified as well as unidirectionally scratched silicon surfaces. Generally, adsorption of polycations like PLL from single component solutions leads to laterally incomplete coverage (in the nanometer to micrometer scale) at silica surfaces due to the charge repulsion between the equally charged PLL molecules, and only thin modifying layers not exceeding the polymer size can be deposited. Therefore, we used the multilayer concept introduced by Decher et al.,8 whereby PLL was consecutively adsorbed alternating with the polyanion poly(maleic acid-co-R-methylstyrene) (PMA-MS), enabling complete surface coverage and tunable layer thickness. Here, we used dichroic in situ ATR-FTIR spectroscopy,9 interested in the following two points: if PLL could preserve its R-helical solution conformation at the surface (i) and if these R-helical rods can be arranged in anisotropic distributions at the surface (ii) in the wet and dry state. Experimental Section Surface. Plasma-cleaned (plasma chamber PDC-32G, Harrick (distributed by Starna, Pfungstadt), 1 Torr, 2 min, 100W) trapezoidal internal reflection elements (IRE) of Si (n1 ) 3.5, 20 mm wide × 2 mm thick, upper and lower lengths 48 and 52 mm) with a 45° aperture were used as model surfaces for the polyelectrolyte multilayer deposition. The texturized IREs were prepared by carefully polishing in the direction of the short (20 mm) edge using

10.1021/bm005628g CCC: $20.00 © 2001 American Chemical Society Published on Web 01/31/2001

Orientation of R-Helical Poly(L-lysine)

Biomacromolecules, Vol. 2, No. 1, 2001 263 led to absolute errors ∆S ) 0.1-0.2 for the order parameters S and of ∆γ ) 4-8° variations for the opening angles γ according to the cone model (computation see below). CD Spectroscopy. A JASCO spectral polarimeter (J-810, JASCO, Gross-Umstadt, Germany) was used. CD spectra in the range from 180 to 250 nm were recorded on D2O solutions of PLL, which were previously probed by transmission FTIR sperctroscopy (BRUKER IFS 55, Equinox). The CD spectra were qualitatively analyzed due to the appearance of the negative 222 nm/208 nm dublett (n-π*, π-π* transitions, respectively) indicative for the R-helical conformation.

Results and Discussion

Figure 1. SEM picture of the unidirectionally texturized surface of the Si-IRE.

a 0.2 µm diamond paste (Markon, Lotzwil, Switzerland) as described earlier.1 This resulted in parallel textures on the SiO surface. Scanning electron microscopy (SEM) showed that the width of the created surface grooves was about 50 nm, which is small compared to mid-IR wavelengths of 2.5-10 µm. A micrograph of the Si-IRE surface is shown in Figure 1. Polyelectrolyte Multilayers. The polycation poly(L-lysine hydrobromide (PLL, Sigma, Mw ) 205 000 g/mol) and the polyanions poly(maleic acid-co-R-methylstyrene) (PMA-MS, Leuna, Mw ) 23 000 g/mol), poly(vinyl sulfate) (PVS, Serva, Mw ) 162 000 g/mol), respectively, were used without further preparation and were dissolved in deionized water (Millipore, 18.2 MΩ) to a concentration of 5 mmol/L. The multilayers of oppositely charged polyelectrolytes were fabricated by consecutive depositing/rinsing cycles on the Si-IRE surfaces in the sample compartment of the ATRIR sorption cell (IPF Dresden) according to the stream coating procedure described therein (e.g., ref 10). Between every polyelectrolyte addition, the sorption cell was carefully rinsed with water. The multilayers were dried by a gentle N2 stream above the SiIRE in the sample compartment. ATR-FTIR Spectroscopy. The in situ ATR-FTIR apparatus for sorption measurements (U. P. Fringeli, University of Vienna, OPTISPEC, Zu¨rich),9 consisting of a special mirror setup and the in situ sorption cell (IPF Dresden), was used on a commercial rapid scan FTIR-spectrometer (IFS 28, BRUKER) equipped with globar source and MCT detector, as described elsewhere. ATR-FTIR absorbance spectra were recorded by the SBSR (single beam-sample reference) method,9 whereby single channel spectra IS,R were recorded of both the upper (S) and lower (R) half of the Si-IRE (50 × 20 × 2 mm3) by one IR beam. Two liquid chambers (S, R), which are filled with polyelectrolyte solution (S-chamber) and with water (R-chamber), are mounted above the sample and reference half, respectively. Ratioing of the single channel spectra according to ASBSR ) -log(IS/IR) resulted in absorbance spectra (ASBSR), with proper compensation of the background absorptions due to the SiOx layer, the solvent (water), the water vapor (spectrometer), and ice on the MCT detector window. For the determination of the integrated areas and peak intensities of the ν(CH) (3000-2800 cm-1), the amide A (3450-3250 cm-1), the amide I (1680-1610 cm-1), and the amide II (1590-1480 cm-1), tight baselines and the integration limits given in the brackets, respectively, were used (OPUS software package, BRUKER). Relative errors of 5-10% of the determined areas and peaks

Multilayer Buildup. Multilayers of poly(L-lysine) (PLL) and poly(maleic acid-co-R-methylstyrene) (PMA-MS) were deposited by alternate adsorption on unmodified and uniaxially scratched silicon IRE in the in situ ATR-FTIR sorption cell, which is anlogous to the dip/rinse protocol of silicon wafers or glass slides initially given by Decher.8 Generally, the electrostatic self-assembly (ESA) of oppositely charged polyelectrolytes is dependent on a variety of parameters. As parameters we have concentrated here on the type of salt and the substrate surface morphology. In the following sections results on three different multilayer deposition series (A, B, C) are shown, whereby first, multilayer system PLL/ PMA-MS (MLS) was generated by consecutive adsorption from salt-free solutions of both PLL and PMA-MS (A) onto a Si-IRE. Second, a PLL/PMA-MS MLS was deposited in the presence of 1 m NaClO4 in both polyelectrolyte solutions (B). Third, we built up a PLL/PMA-MS MLS in the presence of 1 m NaClO4 onto a Si-IRE, which was modified by a scratching technique (C). Thereby, parallel aligned surface grooves were generated, having depthts and widths of about 50 nm, as is shown in Figure 1. This technique was published earlier,1 where hairy rod molecules of the PMOLG (poly(γ-methyl L-glutamate-co-γ-octadecyl L-glutamate)) type were shown to align in a distinct direction in the plane of the substrate by solution casting and swelling in CHCl3. The consecutive deposition of these three MLS (A, B, C) was followed in situ by ATR-FTIR spectroscopy, as is principally described therein.10-12 The ATR-FTIR spectra were recorded after each adsorption step and are shown in Figure 2, parts A-C, and their prominent IR bands are summarized in Table 1. The integrated areas of the diagnostic amide I band due to the polycation PLL as well as of the negative ν(OH) band due to the desorbed water at the interface10 are shown in Figure 3, parts A-C, respectively. In general, the integrated areas of the polyanion and polycation bands reflect the relative adsorbed amount at the surface. The absolute determination of adsorbed polyelectrolyte amounts based on the integrated areas of its characteristic IR bands is possible, knowing their integrated molar absorption coefficients , which are shown for adsorbed proteins therein.13 Qualitatively, there is a low consecutive adsorption level for the salt-free deposited MLS, whereas we observed an elevated adsorption level for the PLL/PMAMS MLS adsorbed in the presence of NaClO4 for the unmodified as well as for the unidirectionally polished SiIRE. In the case of conventional polyelectrolytes this could

264

Biomacromolecules, Vol. 2, No. 1, 2001

Figure 2. In situ ATR-FTIR spectra monitoring the consecutive multilayer deposition of PLL/PMA-MS at the SiO surface (A): (A) PLL/PMA-MS deposition without salt; (B) PLL/PMA-MS deposition in the presence of 1 m NaClO4 on texturized Si-IRE. The ATRFTIR spectra were recorded after 20 min of polyelectrolyte adsorption, respectively. (Note the different ordinate scale for parts A and B). Table 1. Assignment of Prominent IR Bands of the MLS PLL/ PMA-MS (H2O)

a

position/[cm-1]a

assignment

component

3700-3000 3000-2800 1710 (w) 1643-1648 (vs) 1580 (m) 1550-1540 (s)

ν(OH) ν(CH) ν(COOH) amide I νa(COO-) amide II

water PLL/PMA-MS PMA-MS PLL PMA-MS PLL

Key: vs, very strong; s, strong; m, medium; w, weak.

be explained by two effects. On one hand, low molecular salt screens the charges within a macromolecule causing higher coiling; on the other hand, low molecular salt screens the intermolecular repulsion causing also higher coverage. However, in the case of PLL the addition of NaClO4 has a further effect on the conformation of the peptide backbone, which is described in the following section. Conformations of Multilayer Incorporated PLL. Multilayers of PLL/PMA-MS deposited in the presence of NaClO4 show different amide I positions for the PLL component compared to those deposited in the absence of salt, which could be realized by visual inspection of Figure 2. It is well-known, that position and shape of the amide bands are correlated to peptide conformation (e.g., refs 1419). Being aware that the FTIR conformation determination is somehow limited to the amide bands, whose line shapes are not infinetely resolvable due to the natural line width of their secondary structure components (which might be circumvented by band narrowing techniques16,17), infrared spectra on PLL conformations are quite well understood, since their assignment can additionally be validated by circular dichroism (CD) data.20 In Figure 4 transmission spectra (A) of PLL solutions in D2O without salt (bottom) and in the presence of NaClO4 (0.1 m, (middle), 1 m, (top)) are compared to CD spectra (B) of the very same solutions. On the basis of that and on the manifold literature data, an amide I′ (due to deuterated amide units) position of 1637 cm-1 for PLL in the presence of 1 m NaClO4 in D2O can be unambiguously assigned to the R-helical state.18,19 This assignment is further supported by the appearance of the negative 222/208 nm dublett (n-π*, π-π* transitions,

Mu¨ller

respectively) in the CD spectrum of the very same sample.20 On the other hand PLL in the random coil state causes a broader amide I’ band with a maximum at about 1643 cm-1 in the FTIR spectrum19 and a negative CD ellipticity band at 197 nm in the respective CD spectrum, whose assignment is confirmed therein.20 Furthermore, a β-sheet conformation of PLL would cause two amide I components at 1620 and 1690 cm-1 shown therein14-19,4 and a negative CD ellipticity band at 218 nm,20 both of which were not observed in our data. Orientation at the Uniaxially Scratched Surface. Since we have identified PLL to form R-helices upon addition of NaClO4, we could expect certain arrangements of this R-helical rods grown in multilayers in contact to a planar surface. To stimulate such an in-plane self-assembly process, we decided to create pretexturized surfaces as a command layer (Figure 1). Previously, we have published a concept to create texturized surfaces by a scratching technique using diamond paste, whereby R-helical rods of poly(γ-methyl glutamate-co-γ-octadecyl glutamate) (PMOLG) were solution-cast from CHCl3 solution and were oriented along the scratching direction.1 Here we used the similar concept aiming at the anisotropic orientation of water-soluble stiff polypeptides. In principal, there was no change in the deposition level of MLS-PLL/PMA-MS in the presence of 1 m NaClO4 for the untexturized Si-IRE (surface I) and for the texturized IRE (surface II) (data not shown). However, performing dichroic ATR-FTIR spectroscopy, whereby parallel (pp-) and vertically (vp-) polarized light is incided with respect to the ATR-plate normal, we were able to check whether there is a certain molecular alignment of R-helical PLL molecules on surface I compared to II. In Figure 5A, ppand vp-polarized ATR-FTIR spectra of PLL/PMA-MS multilayers recorded after 10 deposition cycles (MLS-10) on the unscratched surface (I) are shown, whereas Figure 5B shows the respective spectra of the PLL/PMA-MS MLS10 on the texturized surface (II). Both spectra sets (pp and vp) were recorded in the presence of NaClO4 (wet state). In general, the diagnostic value of orientation studies using polarized light, both in ATR- and transmission mode, is the dichroic ratio R, which is the ratio between IR band areas measured by pp- and vp-polarized light. Evidently, there was a difference in the dichroic ratio of the amide I and amide II band of PLL for surface I compared to surface II giving a qualitative hint to a certain PLL orientation at the texturized surface (II). Hence, for a further quantification, we applied an orientation analysis of ATR-FTIR dichroic data, which was initially described therein21 and further used for the PMOLG system previously.1 According to ref 21, at first we transferred all our dichroic ratios measured in the ATR mode, i.e., RyATR, to those in transmission mode, i.e., RT. For systems with in-plane (x/y plane) orientation the following relation holds:21 RyATR ) (Ex2 + Ez2)/Ey2RT

(1)

Thereby, Ex,y,z are the relative electrical field components in all three space fixed directions x, y, z of the evanescent wave

Orientation of R-Helical Poly(L-lysine)

Biomacromolecules, Vol. 2, No. 1, 2001 265

Figure 3. Plots of the integrated areas of the diagnostic IR bands ν(OH) and amide I vs the adsorption step. Three deposition series of PLL/ PMA-MS at Si-IRE, i.e., without salt on untexturized surface (A), in the presence of 1 m NaClO4 on untexturized surface (B), and in the presence of 1 m NaClO4 on texturized surface (C), are shown.

Figure 4. Transmission FTIR (A) and CD spectra (B) recorded on the same preparations of PLL solutions. FTIR and CD spectra on the top respresent PLL in the presence of 1 m NaClO4 and on the bottom those recorded in the absence of salt. Assignment of the amide I band and the CD spectra of PLL to R-helical (R) and unordered (random) (R) conformations. D2O was used as the solvent.

Figure 5. pp- and vp-polarized ATR-FTIR spectra of PLL/PMAMS multilayers recorded after 10 deposition cycles on untexturized (A) and on the uniaxially scratched Si IRE (B) in the presence of 1 m NaClO4 (wet state) in the range 4000-1200 cm-1.

at the crystal surface. These electrical components have constant values and can be determined knowing the refractive indices of all considered media (n1 ) 3.5 (Si), n2 ) 1.45 (multilayer), n3 ) 1 (air) or n3 ) 1.33 (water) and the

thickness of the multilayer dMLS. Since dMLS was determined by ellipsometry to a value of about 50 nm, which is much smaller than the depth of penetration (dp ) 0.5 µm), we assumed the “thin film case” according to Harrick.22 This has the consequence that in the formulas of the electric field components only n2 ) 1 for air (or n2 ) 1.33 for thin films in contact to water) has to be considered. For intermediate thicknesses between “thin” and “bulk”, an empirical interpolation formula9 might be valid, which is shown in the Appendix. The dichroic ratio RT itself is related to a so-called orientation parameter S’ in the following way:21 RT ) (sin2 θ + S′)/(2 cos2 θ + S′)

(2)

Thereby, θ denotes the angle of the amide I or amide II transition moments relative to the helical axis, which were θamide I ) 38° and θamide II ) 73° according to values determined recently therein.2 Furthermore, from the orienta-

266

Biomacromolecules, Vol. 2, No. 1, 2001

Mu¨ller

following relation: γ ) arcsin[(2S′)1/2/(2 + 3S′)1/2]

Figure 6. pp- and vp-polarized ATR-FTIR spectra of PLL/PMSAR-MeSty multilayers recorded after 10 deposition cycles on the untexturized (A) and on the uniaxially scratched Si IRE (B) in the presence of 1 m NaClO4, after rinsing (Millipore water) and drying (dry state) in the range 4000-1200 cm-1.

Figure 7. Schematic idealized illustration of the cone model, showing exemplarily three grooves unidirectionally scratched into a Si-IRE, in which the R-helical PLL rods, deposited alternately with PMAMS, are aligned with different uniaxial order (order parameter S ) 0.2, 0.8, 1). (The individual rods need not have point symmetry, as is idealized and indicated here).

tion parameter S′, an order parameter S may be calculated by S ) 1/(1 + 3/2S′)

(3)

with the extreme cases S′ ) 0, S ) 1: for perfect order S′ ) ∞, S ) 0: for no order A very roughly approximating model, which may serve as an illustration of the determined order parameters, is based on a (double) cone model21 and is shown in Figure 7. This model takes into account that the more or less aligned stiff PLL R-helices, which are deposited alternately with PMAMS, have inclination from a main axis resulting in bundles of PLL rods lying within two cones. The opening angle γ gives a measure of the distribution of the rods, so that the greater it is, the less axial order prevails. Rearrangement of formulas given therein21 enables the computation of these γ values from the determined order parameter S via S′ by the

(4)

Thereby, the limiting value is the magic angle γ ) 54.7°, corresponding to S ) 0, meaning no uniaxial order. The presentation of the experimental and computed orientation data is organized as follows: Tables 3 and 4 summarize orientation data (R, S, γ) based on Figures 5 and 6, whereby the orientation of the MLS-10 on the texturized substrate (I) was compared to the orientation on the untexturized one (II). Generally, we performed measurements on the MLS still being in contact to 1 m NaClO4 solution, i.e., in the wet state (i) as well as the dry state (ii), whereby the MLS sample was rinsed with Millipore water followed by drying under a N2 stream. In Table 5, additional orientation data are presented for the MLS-3 in the wet state in comparison to the dry state, which corresponds to Figure 8. Finally, orientation data on multilayer assemblies consisting of R-helical PLL and poly(vinyl sulfate) (PVS) are presented in Figure 9 and summarized in Table 6, which could be compared to the PLL/PMA-MS system. Orientation in the Wet State. At first, dichroic ATRFTIR measurements were performed on the MLS in the wet state (presence of NaClO4), whose pp- and vp-polarized IR spectra are presented in Figure 5, parts A and B. In these spectra, the ν(OH) (3700-3000 cm-1) and the δ(OH) bands (around 1640 cm-1) could not be compensated sufficiently, i.e., there might be too much signal disturbance, especially of the amide I band (1700-1600 cm-1), which could cause misleading values (too high or too low) for the integrated amide I band areas. The amide II line shape is not influenced by improper compensation of the δ(OH); however, we have to take into account that the amide II band may be partly overlapped with the ν(COO-) band (1580 cm-1). We checked for this and found out that the contribution of the ν(COO-) band to the amide II integrated area (1590-1480 cm-1) is negligible and we could consider the amide II for orientation analysis in the wet state. Significantly, there is an increase of the molecular order parameter when the substrate is changed from the untexturized (SIamide II ) 0.2) to the texturized case (SIIamide II ) 0.5) (Table 3), giving experimental proof that the PLL rods tend to self-assemble into the parallel aligned surface grooves of the texturized substrate (Figure 1). For comparison, in ref 1 we observed values of about S ) 0.8 for PMOLG molecules aligned along a similarly texturized Si-IRE. These findings may be additionally supported by the approximate rod length of R-helical PLL of about 150 nm, which is higher compared to the groove width (50 nm). The approximation is based on the known height of 0.15 nm per amide residue for the left- and right-handed R-helix23 and the degree of polymerization of about 1000 (Mw ) 205 000 g/mol). Orientation in the Dry State. In Figure 6, the polarized ATR-FTIR spectra of the MLS-10 sample after rinsing with water and subsequent drying are given. These spectra generally offer the advantage, that the amide I can be analyzed, since its line shape is not disturbed by any incompensation of the δ(OH) band. The dichroic ratios and the calculated order parameters are summarized in Table 4.

Orientation of R-Helical Poly(L-lysine)

Biomacromolecules, Vol. 2, No. 1, 2001 267

Table 2. Positions of the Amide I and Amide II Bands of PLL Samples in H2O and D2O, Determined by Transmission FTIR Spectroscopy PLL, no salt PLL, 1 m NaClO4

amide I/[cm-1] (H2O)

amide I′/[cm-1] (D2O)

amide II/[cm-1] (H2O)

assignment

1648-1646 (broad) 1647-1646 (narrow)

1644-1642 (broad) 1637-1635 (narrow)

1540-1535 1548

random R-helix

Table 3. Experimental Dichroic Ratios Ry, Computed Order Parameters S, and Cone Opening Angles γ (Eqs 1-4) of R-Helical PLL Incorporated in MLS-10 of PLL and PMA-MS in the Presence of NaClO4 on Untexturized and Texturized Si-IRE in the Wet Statea MLS-10, wet state PLL/PMA-MS

MLS-10, untexturized IRE (II) amide Ib

Ry S ( 0.1 γ/deg

amide II

MLS-10, texturized IRE (II) amide Ib

amide II

1.94 (2.04) 0.2 (0.2) 47 (47) ( 4

2.54 (2.82) 0.4 (0.5) 39 (35) ( 4

a The thin film approximation22 was used (n ) 3.5 (Si), n ) 1.33 1 2 (water)). Generally, values based on the integrated areas of the amide bands are considered, whereby in parentheses the respective peak heights are given (see Experimental Section) b The amide I band areas are not considered here because of different compensation of the δ(OH) band at about 1640 cm-1 for the pp- and vp-spectrum (Figure 5).

Table 4. Experimental Dichroic Ratios Ry, Computed Order Parameters S, and Cone Opening Angles γ (Eqs 1-4) of R-Helical PLL Incorporated in MLS-10 of PLL and PMA-MS in the Presence of NaClO4 on Untexturized and Texturized Si-IRE in the Dry Statea MLS-10, dry state PLL/PMA-MS

Ry S ( 0.1 γ/deg

MLS-10, untexturized IRE (II) amide I

amide II

MLS-10, texturized IRE (II) amide Ib

amide II

1.22 (1.19) 1.46 (1.43) 0.79 (0.78) 1.77 (2.00) 0.0 (0.0) 0.2 (0.2) 0.3 (0.3) 0.4 (0.5) 55 (55) ( 4b 47 (47) ( 4 43 (43) ( 4 39 (35) ( 4

a The thin film approximation22 was used (n ) 3.5 (Si), n ) 1 (air)). 1 2 Generally, values based on the integrated areas of the amide bands are considered, whereby in brackets the respective peak heights are given (see Experimental Section) b Magic angle (54.74).

Table 5. Experimental Dichroic Ratios and Computed Order Parameters (Eqs 1-4) of R-Helical PLL Incorporated in the MLS-3 of PLL and PMA-MS in the Presence of NaClO4 on Texturized Si-IRE in the Wet and Dry Statesa MLS-3 PLL/PMA-MS

Ry S ( 0.1 γ/deg

wet amide Ib

dry

amide II

amide I

amide II

3.37 (3.91) 0.6 (0.7) 31 (27) ( 4

0.91 (0.79) 0.2 (0.3) 47 (43) ( 4

1.62 (1.89) 0.3 (0.4) 43 (39) ( 4

a The thin film approximation22 was used (n ) 3.5 (Si), n ) 1 (air)). 1 2 Generally, values based on the integrated areas of the amide bands are considered, whereby in brackets the respective peak heights are given (see Experimental Section) b Not determined because of interference between the amide I and uncompensated δ(OH) band.

Similar to the wet state data, the molecular order parameter had increased, when the substrate was changed from the untexturized (SIamide I ) 0.0, SIamide II ) 0.2) to the texturized case (SIIamide I ) 0.3, SIIamide II ) 0.5). In general, for the MLS10 on the texturized substrates, the calculated order parameters for the dry state (S ) 0.3-0.5) were comparable to those of the wet state (S ) 0.4-0.5) (compare to MLS-3 below). Comparison to MLS-3. Since we were interested in the influence of the layer number on rod alignment, we performed dichroic measurements on the MLS-3 additionally.

Thereby, the experimental data of Figure 8 are summarized in Table 5. Again, only data for the amide II band are presented because of the obvious incompensation of the δ(OH) band at 1700-1600 cm-1. First, for the wet state the order parameters S based on the amide II dichroism for MLS-3 (S ) 0.6-0.7) showed a slight increase compared to the MLS-10 (S ) 0.2-0.5), whose data are shown in Table 3. This might be due to growing imperfections during the subsequent layer deposition or to the fact, that later (outer) deposited layers could not recognize the texturized surface any more. Unsimilar to the MLS-10, the MLS-3 showed a certain loss of unidirectional order, after the multilayer assembly had been dried (S ) 0.2-0.4), which was not observed after drying the MLS-10 (S ) 0.3-0.5) (Table 4). Oriented Multilayers Composed of PLL/PVS (MLS10). For comparison to the oriented PLL/PMA-MS multilayers, we changed the polyanion component and used poly(vinyl sulfate) (PVS) instead of PMA-MS. Analogous to PLL/PMA-MS, we consecutively adsorbed PLL and PVS in the presence of NaClO4. The concerning pp- and vppolarized ATR-FTIR spectra are given in Figure 9, parts A and B, whereby Figure 9A represents the data on the wet state and Figure 9B shows the spectra after drying the multilayer assemblies. Again for the wet state we have certain problems with the accurate δ(OH) compensation in the amide I region, so that these integrated areas should be handled with caution. However the amide II band areas were not disturbed and could be analyzed reliably. For the dry state both amide bands could be considered and analyzed. In the case of PVS we have an additional spectroscopic advantage, that the amide II band is not interfered with any contribution of the νa(COO-) band. The data for the PLL/PVS multilayers are presented in Table 6. In general, we could also achieve high molecular order for the MLS-10 of PLL/PVS at uniaxially scratched substrates in the wet state (S ) 0.6), whereby again drying caused a certain loss of order (S ) 0.3-0.5). Furthermore, the dichroic ratio of the ν(SO2) of the sulfate groups of PVS was Ry ) 1.46 (dry state). Qualitatively, from this value a certain arrangement of the O-S-O plane of the sulfate groups lying more or less parallel to the substrate surface may be concluded, which presumably could be supported by further modeling studies. Conclusion We could show that PLL undergoes conformational changes from random to R-helix when it is dissolved in 1 m NaClO4 solution. We used this to build up polyelectrolyte multilayers by consecutive adsorption of poly(L-lysine) (PLL) in the R-helical state and poly(maleic acid-co-R-methylstyrene) (PMA-MS) on unmodified and uniaxially scratched silicon substrates. Dichroic ATR-FTIR measurements on these MLS, which were performed in both the wet and the dry state, showed a significant dichroism of the amide I and

268

Biomacromolecules, Vol. 2, No. 1, 2001

Mu¨ller

Figure 8. pp- and vp-polarized ATR-FTIR spectra of PLL/PMSA-R-MeSty multilayers (presence of 1 m NaClO4) recorded after 3, deposition cycles (MLS-3) on the uniaxially scratched Si IRE in the wet state (A) and after rinsing (Millipore water) and drying (B) (dry state) in the range 1720-1480 cm-1. ((A) and (B) represent different but representative samples.)

Figure 9. pp- and vp-polarized ATR-FTIR spectra of PLL/PVS multilayers in the presence of 1 m NaClO4 recorded after 10 deposition cycles (MLS-10) on the uniaxially scratched Si IRE in the wet state (A) and after rinsing (Millipore water) and drying (B) (dry state) in the range 40001000 cm-1. Table 6. Experimental Dichroic Ratios and Computed Order Parameters (Eqs 1-4) of R-Helical PLL Incorporated in the MLS-10 of PLL and PVS in the Presence of NaClO4 on Texturized Si-IRE in the Wet and Dry Statesa MLS-10 PLL/PVS

Ry S(area) ( 0.1 γ(area)/deg

wet amide Ib

dry

amide II

amide I

amide II

3.69 (3.38) 0.6 (0.6) 31 (31) ( 4

0.79 (0.73) 0.3 (0.4) 43 (39) ( 4

1.79 (2.10) 0.4 (0.5) 39 (35) ( 4

a The thin film approximation22 was used (n ) 3.5 (Si), n ) 1 (air)). 1 2 Generally, values based on the integrated areas of the amide bands are considered, whereby in brackets the respective peak heights are given (see Experimental Section) b Not considered due to δ(OH) incompensation.

amide II bands. On the basis of the known angles of the transition dipole moments of the amide I and amide II bands relative to the helical axis, molecular order parameters (S, γ), representing the degree of unidirectional alignment, were

determined using a methodology described earlier.1,21 The determined molecular order parameters ranged from S ) 0.5-0.7 for the aligned PLL rods and gave evidence, that the R-helical rods of PLL were found to be more or less perfectly aligned along the scratching direction of the substrate. At the untexturized surface, we obtained minor order parameters between S ) 0.0-0.2 for the PLL molecules. For the uniaxially scratched surfaces, the following general tendencies were obtained: (i) The maximum order parameters, i.e., Smax (based on different MLS and amide bands) for the wet state (Smax ) 0.7) appeared to be significantly higher compared to the dry state (Smax ) 0.5) for all investigated multilayer systems. (ii) For the wet state, the order parameters of the MLS-3 of PLL/PMA-MS (Smax ) 0.7) were higher compared to the respective MLS-10 ones (Smax ) 0.5). (iii) For the wet state, the MLS-10 of PLL/PVS (Smax )

Orientation of R-Helical Poly(L-lysine)

0.6) revealed a slightly higher order parameter compared to the MLS-10 of PLL/PMA-MS (Smax ) 0.5). These studies were done in correlation to those with the R-helical peptides of the PMOLG type.1 Since R-helices exhibit strong dipole moments (D ) 1.5 D per helix turn), it has further to be elaborated how proteins of high R-helix content, which are adsorbed from solution, might interact with the dipole field of these highly aligned R-helical rods. It could be interesting to compare the recognition of such proteins by surfaces exhibiting aligned R-helical peptides which are either hydrophilic (PLL) or hydrophobic (PMOLG) due to their charged ((CH2)4NH4+) or hydrocarbon ((CH2)17CH3) side chains, respectively. Furthermore, dichroic ATR-FTIR spectroscopy might be used to check for polyelectrolyte stiffness by the quality of uniaxial alignment. Acknowledgment. This work was performed within the framework of the Sonderforschungsbereich “Reactive Polymers at Inhomogeneous Systems, in Melts and at Interfaces” of the DFG (SFB 287). B. Kessler is thanked for skillfully performing the FTIR measurements. The SEM picture on the texturized Si IRE was kindly measured by T. Kratzmu¨ller, IPF Dresden. Appendix Even if an intermediate thickness has to be assumed, according to Fringeli9 the following empirical relation, which weights the deviation between the electric field components for the thin film and the thick film case by an 1 exp(-d/dp) function may be used: Ex,y,z (inter) ) (Ex,y,zbulk - Ex,y,zthin) * [1 - exp (-d/dp)] (5)

Biomacromolecules, Vol. 2, No. 1, 2001 269

References and Notes (1) Schmitt, F. J.; Mu¨ller, M. Thin Solid Films 1997, 310, 138-147. (2) Marsh, D.; Mu¨ller, M.; Schmitt, F. J. Biophys. J. 2000, 78, 24992510. (3) Susi, H.; Timasheff, S. N.; Stevens, L. J. Biol. Chem. 1967, 242, 5460. (4) Mu¨ller, M.; Buchet, R.; Fringeli, U. P. J. Phys. Chem. 1996, 100, 10810-25. (5) Sugai, S.; Ebert, G. AdV. Colloid Interface Sci. 1986, 24, 247-282. (6) Odijk, T. J. Polym. Sci. (Polym. Phys. Ed.) 1977, 15, 477. (7) Skolnick, J.; Fixman, M. Macromolecules 1977, 10, 944. (8) Decher, G.; Hong, J. D.; Schmitt, J. Thin Solid Films 1992, 210/ 211, 831. (9) Fringeli, U. P. In Encyclopedia of Spectroscopy and Spectrometry; Lindon, J. C., Tranter, G. E., Holmes, J. L., Eds.; Academic Press: San Diego, CA, 2000. Fringeli, U. P. Chimia 1992, 46, 200-214. (10) Mu¨ller, M.; Rieser, T.; Lunkwitz, K.; Berwald, S.; Meier-Haack, J.; Jehnichen, D. Macromol. Rapid Commun. 1998, 19, 333. (11) Mu¨ller, M.; Brisˇsˇova´, M.; Rieser, T.; Powers, A. C.; Lunkwitz, K. Mater. Sci. Eng. C 1999, 8-9, 167-173. (12) Mu¨ller, M.; Rieser, T.; Lunkwitz, K.; Meier Haack, J. Macromol. Rapid Commun 1999, 20, 607-611. (13) Mu¨ller, M.; Werner, C.; Grundke, K.; Eichhorn, K.-J.; Jacobasch, H. J. Microchim. Acta 1997, 14, 671-674. (14) Elliott, A.; Ambrose, E. J. Nature 1950, 165, 921. (15) Miyazawa, T. J. Chem. Phys. 1960, 32, 1647. (16) Byler, M.; Susi, H. Biopolymers 1986, 25, 469. (17) Surewics, W. K.; Mantsch, H. H. Biochim. Biophys. Acta 1988, 952, 115. (18) Chirgadze, Y. N.; Brazhnikov, E. V. Biopolymers 1974, 13, 17011712. (19) Jackson, M.; Haris, P. I.; Chapman, D. Biochim. Biophys. Acta 1989, 998, 75-79. (20) Greenfield, N.; Fasman, G. D. Biochemistry 1969, 8, 4108-4116. (21) Fringeli, U. P.; Schadt, M.; Rihak, P.; Gu¨nthard, Hs. H. Z. Naturforsch. 1976, 31a, 1098-1107. (22) Harrick, N. J. Internal Reflection Spectroscopy; John Wiley & Sons: New York, 1967. (23) Schultz, G. E.; Schirmer, R. H. In Principles of Protein Structure; Cantor, C. R., Ed., Springer: New York 1979.

BM005628G