Vertically Oriented α-Helical Peptides in Langmuir-Blodgett Films∗

To obtain vertically-oriented helical-peptides in the lipid Langmuir-Blodgett (LB) films, three types of helixforming “amphiphilic peptides” (APs) with different length of the hydrophobic domain were designed, and the orientation of the helices and the surface morphology of the AP/lipid mixed LB films were investigated. The structural models, which were consistent with the results obtained by different types of measurements, were revealed that a phase separation between the APs and lipids occurred in the mixed film and the helix orientation significantly depended on the length of the AP’s hydrophobic domain. When the hydrophilic domain of AP consisted of three lysine residues, the hydrophobic domain consisting of 15 residues resulted in a uniform orientation of helical AP, whose helical axis tilted 27◦ from the film normal, but the longer or shorter hydrophobic domain than the 15 residues resulted in a mixture of vertically and horizontally oriented APs. Thus, the balance between the hydrophobic and hydrophilic domains in the AP is important to control the orientation of the helical AP in the AP/lipid mixed LB film. [DOI: 10.1380/ejssnt.2012.379]


I. INTRODUCTION
The large polypeptides fold into sophisticated structures in the cell membranes to form the transmembrane (TM) proteins [1].The TM proteins have hydrophilic and hydrophobic domains, and their folded structures determine their bio-activities such as an ion channel or a receptor in the membrane [2].The folding of the polypeptides into the TM proteins is governed by a self-organization ability of the polypeptides and is one of the key subjects in biology [3].Furthermore, an understanding of the mechanism of the self-organization ability would provide tremendous opportunities to engineer biosensors, artificial cells, and molecular devices using polypeptides [4][5][6][7].
To introduce such self-organization ability of the polypeptides into the Langmuir (L) and Langmuir-Blodgett (LB) films, various kinds of "amphiphilic peptides" (APs) have been introduced [8].Most peptidebased L and LB films were fabricated by helical peptides designed to orient their helical axes parallel to the film plane [8], modified peptides with an alkyl chain [9][10][11] and an ionic group [12], peptides composed of the same sequences of the amino acids (AAs) to the TM domains of existing proteins [13,14], or the artificial peptides con-sisting of non-natural AAs [15].
Our AP has two distinct domains in its AA sequence; [16] one domain mainly consists of hydrophobic AAs (average hydropathy [17] is scaled to be hydrophobic) and the other domain consists of hydrophilic AAs (average hydropathy is scaled to be hydrophilic), and the AAs used can be encoded by genes.Our final goal is to determine the primary structure of the AP that forms helix and orients its helical axis vertical to the plane of the L and LB films.To mimic the TM proteins in the cell membrane, the vertically oriented helical APs must be the building blocks of the artificial channels and receptors.Furthermore, the vertically oriented helices to the film plane result in a vertical alignment of the electric dipole moments of hydrogen bonds among the helices that would provide new electro-optical devices [12].
In the previous work [16], three APs with different average hydropathies from 3 to 1 were examined.These APs had the same hydrophilic domain consisting of three lysine (K) residues in their N-terminus, and their hydrophobic domains were different, resulting in a difference in the average hydropathy.These APs did not show a uniform structure in the LB film, and their secondary structures were a mixture of α-helix, β-strand, and random coil.The higher average hydropathy resulted in an increase in a randomness of the structure.Therefore, in the present study, we have redesigned the hydrophobic domain of the AP to obtain a LB film consisting of α-helical APs.The basic primary structure of hydrophobic domain was referred to the hydrophobic peptides designed to investigate the interaction between the peptides and lipid bilayers [18].By  changing the number of AA of the hydrophobic domain, three APs, whose average hydropathies were from 1 to 1.5, have been proposed (Table I), and we have examined how the difference in the length of the hydrophobic domain as well as the preparation temperature affect the structure of the L and LB films and the orientation of the helical axis of the APs in the films.The surface pressure-area (π-A) isotherms of the AP/lipid mixed L films were observed to obtain a rough impression of the molecular orientation in the monolayer.The circular dichroism (CD) spectra of the AP/lipid mixed LB multilayered films were observed to analyze the orientation of the helical axis of the APs.The topography images of the AP/lipid mixed LB monolayers obtained by the atomic force microscope (AFM) were used to discuss the phase separation between the APs and lipids and estimate the assembled structures of the mixed monolayers.Three types of molecular assemble models were examined to explain the results obtained by the different methodologies.

II. MATERIALS AND METHODS
Pure water (> 18 MΩcm) was prepared in a Milli-Q system (Elix Advantage 3).Three APs (AP3-11, AP3-15, and AP3-19, Table I) were synthesized by Takara Bio Inc.Their average hydropathies were calculated using the values in the literature [17].AP3-11 was dissolved in chloroform/methanol (5:2 v/v).AP3-15 and AP3-19 were dissolved in dichloromethane/methanol (5:2 v/v).The concentration of AP3-11, AP3-15, and AP3-19 were 0.068, 0.059, and 0.057 mM, respectively.The lipid, dimethyl distearyl ammonium bromide (DSAB, Tokyo Kasei Kogyo Co., Ltd.), was dissolved in each solvent at the concentration of 0.2 mM.Each DSAB solution was mixed with the AP solution, whose solvent was the same to the DSAB solution, and these AP/DSAB mixed solutions were used as the spreading solutions for the AP/DSAB mixed L or LB films.5 mM poly (vinyl sulfate) potassium salt (ave.Mw ∼170,000, Sigma-Aldrich Co. LLC.), whose pH was adjusted to 7 by adding a condensed NaOH solution, was used as the subphase, and the subphase temperature was kept at 21.5 ± 0.5 • C or 12.0 ± 0.5 • C.
For the π-A isotherm measurements, the AP/DSAB mixed solution at the molar fraction of AP represented by M F AP = M AP /(M AP + M DSA ), where M AP and M DSA are the concentrations of AP and DSAB in mol/L, respectively, was spread on the subphase.The monolayer was compressed at a barrier speed of 5-11 mm/min and the π was measured by the Wilhelmy plate method.We observed the isotherms of the L films at M F AP = 0, 0.091, and 1.
For the LB depositions, the mixed L film (M F AP = 0.091) was compressed at the given deposition surface pressure (π dep ) and was transferred onto a substrate by the vertical dipping method [19].The AP3-11/DSAB mixed L films were deposited at π dep = 35 and 24 mN/m at the subphase temperature of 21.5 • C and 12.0 • C, respectively.The AP3-15/DSAB and AP3-19/DSAB mixed L films were deposited at π dep = 27 and 25 mN/m, respectively, at the subphase temperature of 21.5 • C. Hydrophobized fused silica substrates and freshly cleaved mica substrates were used to deposit eighteen layers of Y-type multilayers [19] and a single layer of Z-type films [19], respectively.The former and latter films were used for the CD and AFM observations, respectively.
The CD spectra were measured at room temperature using Jasco J-820 CD spectropolarimeter.The LB film was set in perpendicular to the optical axis of the spectropolarimeter.The solubilized APs were also prepared for the CD measurements.The fused silica substrate and the solvents in a fused silica cuvette were measured as the reference spectra for the AP/DSAB mixed LB films and the solubilized APs, respectively.The procedure for solubilizing the APs in the aqueous phase is described as follows: The AP solution prepared by the organic solvent was placed in a glass vial and the solvent was evaporated by a stream of nitrogen gas, followed by evacuation using a diaphragm pump to form an AP film inside the vial.The 10 mM Tris-HCl buffer (pH 7) with and without 20 mM sodium dodecyl sulfate (SDS, Nacalai Tesque, Inc.) were then added to be the AP concentration of 20 µM and the vial was put into a sonicator for few minutes.To estimate the content of secondary structures, the CD spectra of the peptide solutions were analyzed by the method based on the literature [20].
The surface morphology of the LB monolayers was observed by AFM (Dimension Icon, Bruker AXS GmbH) in a tapping mode.The resonant frequency of the cantilevers used (NCHV-10V) was in the ranged of 361 to 400 kHz, and the topography images were collected at 512 × 512 pixel resolution.

A. CD spectra of the AP solutions
Figure 1 shows the CD spectra of the three APs in the 10 mM Tris-HCl buffer (pH 7) with and without 20 mM SDS.All the APs in the buffer with SDS exhibited a positive peak at 193 nm and two negative peaks at 208 and 222 nm, which are typically observed for the α-helix [21].While, in the buffer without SDS, AP3-11 and AP3-15 http://www.sssj.org/ejssnt(J-Stage: http://www.jstage.jst.go.jp/browse/ejssnt/) e-Journal of Surface Science and Nanotechnology  showed a single negative peak at 199 nm, and AP3-19 showed a positive peak at 190 nm and two negative peaks at 204 and 222 nm.The analysis of the secondary structure was performed on these spectra, and the results summarized in Table II clearly show that all the APs formed α-helix in the buffer with SDS and almost no helix was formed in the buffer without SDS.The amplitude of the CD spectra of the APs in the buffer with SDS decreased as the size of the hydrophobic domain increased, indicating that the long APs were not solubilized completely.Although the SDS did not solubilize the APs completely, the hydrophobic domain of the APs should be surrounded by the hydrophobic tails of the SDS surfactants.Therefore, it was concluded that the three APs shown in Table I form uniform α-helix in a hydrophobic circumstance.

B. π-A isotherms of the AP/DSAB mixed L films
Because all the APs were expected to form the α-helix, the dimensions of the helical structure of the APs are summarized in Table III.The helix can be treated as a cylindrical shape, and the diameter of the cylinder is ca.1.1 nm and the length of the cylinder per one AA residue is 0.15 nm [8,14].By using these values, the molecular area of the helix oriented perpendicular to the air/water interface (A perp ) and the areas of the helix oriented parallel to the interface (A para ) were estimated as shown in Table III.
In the case of AP3-11 at the subphase temperature of 21.5 • C (Fig. 2(a)), the isotherm of AP3-11 (M F AP = 1) exhibited a large molecular area at a low π, indicating that the helical peptides were laid on the subphase surface at low π (see A para in Table III).The isotherm at M F AP = 0.091 obeyed the linear combination of the isotherms of M F AP = 0 and 1 up to 8 mN/m, and above 8 mN/m, the molecular area of the observed isotherm of the mixed film (A mix ) was smaller than that of the linearly combined isotherm (A LC ) (A mix = 0.38 nm 2 and A LC = 0.44 nm 2 at π dep = 35 mN/m, see Fig. 2(a)).Because it is known that the DSAB forms a stable monolayer at the air/water interface [22,23], we doubt the stability of the APs.It is assumed that, in the mixed L film, the DSAB components remain as in the DSAB L film, while the AP components suffer all the influence of the difference between the isotherm of the mixed L film and the linearly combined isotherm.Thus, the apparent area per molecule of the AP in the mixed L film (A mix-AP ) of M F AP = x at any π can be estimated by following equation, where A mix and A DSA are the molecular area of the mixed L film and the molecular area of the DSAB L film at the same π, respectively.The calculated A mix-AP value at M F AP = 0.091 and π dep = 35 mN/m is extremely smaller than the value of A perp that would be the minimum molecular area of the helix (see Table III).This suggests that the AP molecules apparently disappeared from the air/water interface, and we assumed that AP3-11 was squeezed out from the air/water interface to the interface between the subphase and the DSAB monolayer as observed for the AP that has a low average hydropathy in our previous work [16].
In the case of AP3-11 at the subphase temperature of 12.0 • C (Fig. 2(b)), the isotherm of M F AP = 1 was observed up to 22 mN/m, and the isotherm of M F AP = 0.091 obeyed the linear combination of the isotherms up to 22 mN/m (Fig. 2(b)) that is higher than the case of the isotherm of M F AP = 0.091 at 21.5 • C (Fig. 2(a)), indicating that the mixed film was stabilized at the lower temperature.Because the isotherm of M F AP = 1 could not be observed above 22 mN/m due to the limitation of our Langmuir trough, the A mix-AP at M F AP = 0.091 and π dep = 24 mN/m was calculated using Eq. ( 1) and was comparable to the value of A perp (see Table III).Therefore, it was expected that the AP3-11 molecules remained at the air/water interface due to the lower thermal fluctuation in the mixed film on the subphase at 12.0 • C than 21.5 • C and oriented their helical axes nearly vertical to the film plane.
In the case of AP3-15 at the subphase temperature of 21.5 • C (Fig. 3), the isotherm of M F AP = 1 exhibited a better stability and reproducibility up to higher π than the isotherm of AP3-11.Therefore, the larger hydrophobic domain of the AP resulted in a more stable L film.The molecular area of ca. 3 nm 2 , where the π started to increase upon compression, was close to the value of the A para , indicating that the helix was laid on the subphase at low π.The isotherm of M F AP = 0.091 obeyed the linear combination of the isotherms of M F AP = 0 and 1 up to 20 mN/m.Above 20 mN/m, A mix deviated to lower value than A LC , and A mix = 0.57 nm 2 and A LC = 0.63 nm 2 at π dep = 27 mN/m.However, the apparent A mix-AP at π dep = 27 mN/m calculated using Eq.(1) was comparable to the A perp (see Table III).Therefore, in the mixed film, it was suggested that AP3-15 remained at the air/water interface due to its stronger hydrophobicity than AP3-11 and oriented their helical axes nearly vertical to the film plane.
In the case of AP3-19 at 21.5 • C (Fig. 4), the isotherm of M F AP = 1 exhibited a kink at 28 mN/m, suggesting that the reorientation of the AP3-15 helix might occur.Because such kink in the isotherm did not observed in the cases of AP3-11 and AP3-15, the way of reorientation of the AP3-19 helix during the compression is different from those of the AP3-11 or AP3-15 helices.The isotherm of the mixed L film of M F AP = 0.091 obeyed the linear combination of the isotherms of M F AP = 0 and 1 up to 32.5 mN/m, above which the isotherm of M F AP = 1 could not observed due to the limitation of the trough.Therefore, the A mix-AP at M F AP = 0.091 and π dep = 25 mN/m was calculated using Eq. ( 1) and the value was in between the values of A perp and A para (see Table III cases of AP3-11 and AP3-19, the averaged transfer ratios of the depositions in the downstroke and the upstroke of the substrate were 0.75 and 0.9, respectively.However, in the case of AP3-15, the ratios in the downstroke and the upstroke were 0.35 and 0.8, respectively, indicating that the structure of the L film deposited in the downstroke was not maintained in the LB film. When all the peptides in the LB film formed α-helix, an elementary analysis on the helix orientation can be made by the CD spectrum of the film.The shape of the CD spectrum is different when the incident axis is parallel or perpendicular to the helical axis of α-helix, and the spectral shape as a function of wavelength (λ) in the former geometry (G 1 (λ)) and that in the latter geometry (G 2 (λ)) were given in Ref. [24].When the helices are "uniformly" oriented at a tilt angle of the helical axis (θ) from the film normal (see Fig. 5(a), model a), the CD spectrum of the film, whose plane is set to be perpendicular to the optical axis of the spectropolarimeter, is given by where C is the amplitude coefficient.Although the distribution of θ is not taken into account, this equation can be widely adopted as a method for an elementary analysis on the helical axis orientation in the film.However, as discussed on the π-A isotherm shown in Fig. 2(a), there was a possibility that a certain portion of APs in the mixed L film was located at the DSAB/subphase interface at the π dep .For this type of film, another model, which consists of a mixture of parallel and perpendicular orientations, shown in Fig. 5(b) (model b) is proposed.For simplicity, it is assumed that the helices remained at the air/water interface were oriented perpendicular to the film plane and those located at the DSAB/subphase interface were oriented parallel to the film plane.When the fractions of the former and the latter geometries are written by R ⊥ and R // , respectively, the CD spectrum can be given by This equation can be also adapted to the model c (Fig.

5(c))
, where the parallel orientated APs stay at the air/water interface.In this model, the horizontally oriented APs might form not only a monolayer but also a bilayer.
As shown in Fig. 6, the obtained CD spectra of the mixed LB films were fitted using Eqs.though the fitting parameters R ⊥ and R // in Eq. ( 3) were set to be independent, the same best fits were obtained by using Eqs.( 2) and ( 3), i.e., the ratio of cos 2 θ : sin 2 θ was the same as the ratio of R ⊥ : R // for each sample.
The obtained tilt angle θ and the ratio of R ⊥ : R // were listed in Table IV.All the samples exhibited θ < 45  IV), and these values are compared with A mix−AP calculated from the π-A isotherms (see Table III).
In the case of AP3-11 at 21.5 • C, A AP−b (0.76 nm 2 ) is the closest to A mix−AP (0.013 nm 2 ) as we expected, because, in the model b, only the vertically oriented AP component contributes to the molecular area of AP.Thus, the model b is the candidate for modeling the film structure of AP3-11 prepared at 21.5 • C.
In the case of AP3-11 at 12.0 • C, A AP−c (1.04 nm 2 ) exhibits good agreement with A mix−AP (0.90 nm 2 ).This A AP−c was calculated under the assumption that all the horizontally oriented helical-APs formed a bilayer.From the value of A mix−AP , the APs were expected to be oriented vertical, but the CD analysis indicated that the model c is the candidate for modeling the structure.
In the case of AP3-15, all the values calculated from the three models (see Table IV) resemble to A mix−AP (0.97 nm 2 ).Because the isotherm of AP3-15 indicated A mix < A LC (Fig. 3 and Table III) as in the case of AP3-11 at 21.5 • C, the model b cannot be eliminated for modeling the AP3-15/DSAB mixed film.Therefore, we cannot identify the model for explaining the molecular area of AP3-15.
In the case of AP3-19, A AP−c (1.91 nm 2 ) is the closest value to A mix−AP (2.24 nm 2 ) and shows good agreement.This A AP−c was calculated under the assumption that all the horizontally oriented helical-APs formed a monolayer to maximize the value of A AP−c .III.

D. AFM images of the AP/DSAB mixed LB monolayers
In order to confirm whether the phase separation between the APs and DSAB occurred or not, the topography images of the mixed LB monolayers were observed by the AFM and the images are shown in Fig. 7.Here we note that the smooth and flat surface of the DSAB monolayer was confirmed by the AFM (data not shown).
The images of Figs.7(a) and (b) are the topologies of the AP3-11/DSAB mixed monolayers prepared at 21.5 • C and 12.0 • C, respectively.Both images show that there are small holes in the flat surface and their depth was around 1 nm, but the distributions of diameter were different.In the mixed monolayer prepared at 21.5 • C, the diameter of the holes was distributed from 50 to 300 nm, and the majority of the diameter were 100∼150 nm.The holes, whose diameters are larger than 100 nm, are not circular shape and seem to be aggregates of the small holes with the diameter of less than 100 nm.In the mixed monolayer prepared at 12.0 • C, much smaller holes were observed.The diameter was distributed from 30 to 90 nm, and its majority were 30∼50 nm.Again, the larger holes seem to be the assemblies of the smaller ones.The lower preparation temperature, the smaller hole shaped domains.This temperature dependence implies a phase separation phenomenon, i.e., a nucleation rate increases and a speed of domain growth decreases at a lower temperature, resulting in fine domains.http://www.sssj.org/ejssnt(J-Stage: http://www.jstage.jst.go.jp/browse/ejssnt/) e-Journal of Surface Science and Nanotechnology Volume 10 (2012) The AFM image of the AP3-15/DSAB mixed monolayer is shown in Fig. 7(c).In this case, holes and winding trenches with ca. 1 nm depth were observed.The number of the holes was less than that in the AP3-11/DSAB mixed monolayer.The major diameters of the holes were around 50 nm, and the holes, whose diameter was larger than 100 nm, seemed to be the aggregates of small holes.The widths of the trenches were from 40 to 70 nm.The DSAB molecule has a more hydrophobic tail compared to the APs.If the phase separation between APs and DSAB occurs, a line tension between the AP and DSAB phases is expected to decrease as the average hydropathy of the AP increases, resulting in a change in the domain shape from circular to noncircular shape to increase a contacting length between two phases.Thus, the difference in the average hydropathy between AP3-11 and AP3-15 (see Table I) might reflect the difference of the domain shape between Figs. 7(a) and (c).
The image of the AP3-19/DSAB mixed monolayer is shown in Fig. 7(d).There were almost no circular domains and rectangular-like domains with the depth of around 1 nm were observed.The major lengths of the short axes of the rectangular domains were distributed from 40 to 70 nm.The shape of these domains implies that the crystalline phase was formed.Thus, the change in the length of AP induced the significant morphology changes.
Besides the difference in the domain shape, all the images seem to be composed by two different height regions.To obtain the area ratio of the two regions, the histograms of the relative height in the topography images were analyzed.In all the cases, the height distribution exhibited two peaks, and the two peaks were decomposed by two Gaussian functions, and the area ratio between higher and lower height regions was calculated.The area ratios of the higher and lower height regions obtained from several AFM images at different positions were averaged and listed in Table V.When these values obtained by the AFM were compared with the area ratios between AP and DSAB at π dep calculated from the π-A isotherms using Eq. ( 1), it was found that, in all the cases, the values of higher and lower height regions correspond well to the values of DSAB and AP, respectively.Thus, it was shown that the phase separation between the AP and DSAB molecules occurred at the air/water interface.
From the model a, b, and c, we calculated the area ra-  III and IV), it is concluded that the structure of the AP3-11/DSAB mixed monolayer prepared at 21.5 • C is represented by the model b at R ⊥ : R // = 3.9 : 1.Thus, the circular domains observed in the AFM image are formed by the vertically oriented APs, and all the horizontally oriented APs are distributed beneath the DSAB phase.
In the case of the AP3-11/DSAB mixed monolayer prepared at 12.0 • C, the ratio calculated from the model c corresponds well to the ratios indicated in Table V.Because the model c also gave the closest value of the apparent molecular area (compare Tables III and IV), it is concluded that the structure of the AP3-11/DSAB mixed monolayer prepared at 12.0 • C is represented by the model c, where R ⊥ : R // = 1.4 : 1 and all the APs oriented parallel to the film plane formed a bilayer.Thus, the small circular domains observed in the AFM image were formed by the mixture of the vertically oriented APs and the horizontally oriented and bilayered APs.
In the case of the AP3-15/DSAB mixed monolayer, the ratios calculated from the models a and c correspond well to the ratios indicated in Table V.Although we cannot determine the candidate of the structural model by the discussion made on the apparent molecular area (compare Tables III and IV), we can safely eliminate the model b as the candidate.
In the case of the AP3-19/DSAB mixed monolayer, the ratio calculated from the model c corresponds well to the ratios indicated in Table V.Because the model c also gave the closest value of the apparent molecular area (compare Tables III and IV), it is concluded that the structure of the AP3-19/DSAB mixed monolayer is represented by the model c, where R ⊥ : R // = 1.8 : 1 and all the APs oriented parallel to the film plane form a monolayer (no overlapping among the APs).Thus, the rectangular-like domains observed in the AFM image are formed by the mixture of the APs oriented vertical and parallel to the film plane without molecular overlapping.
The distance between the two peaks of the height dis-tribution is the average height difference between the AP and DSAB regions in the monolayer observed.The height differences (∆h AFM ) obtained from the several AFM images at different positions were averaged and compared with the height differences between the AP and DSAB phases calculated from the models a, b and c (see ∆h in Fig. 5).The heights of the DSAB phase in the model b and the AP phase in the model c, whose surfaces are expected not to be flat (see Figs. 5(b) and (c)), are calculated by dividing the total volume of the phase by the total area occupied by the phase.
In the case of the AP3-11/DSAB mixed monolayer prepared at 21.5 • C, ∆h AFM = 1.2 nm, and the model b gives ∆h = 0.7 nm.This difference is not fatal.In the case of the AP3-11/DSAB mixed monolayer prepared at 12.0 • C, the AFM observation and the model c show the same value (∆h AFM = ∆h = 0.7 nm).
In the case of the AP3-15/DSAB mixed monolayer, ∆h AFM = 0.7 nm.The closest value to this ∆h AFM was obtained from the model a and ∆h = 0.3 nm.Therefore, the most possible model for the AP3-15/DSAB mixed monolayer is the model a at tilt angle of 27 • , and the trenches observed in the AFM image (Fig. 7(c)) is formed by the uniformly tilted APs at θ = 27 • (Fig. 5(a)).In the case of the AP3-19/DSAB mixed monolayer, again the AFM observation and the model c give the same value (∆h AFM = ∆h = 1.1 nm).
Overall, good agreements between the experimental values and the calculated values based on each model are obtained, suggesting that the proper models for the four types of the AP/DSAB mixed films are determined.

IV. CONCLUSIONS
The three types of helix-forming APs (Table I), whose hydrophilic domains are the same (three K residues) and hydrophobic domains are different, were used to realize the AP/lipid mixed L and LB films consisting of the vertically aligned APs as the TM proteins exhibit in the cell membrane.We investigated how the length of the hydrophobic domain as well as the subphase temperature affected the film structure.From the three types of structural models of the mixed monolayer shown in Fig. 5, the model that was consistent with the experimental results obtained by the different methodologies was determined for each sample.
To investigate a thermal effect on the film structure, the L and LB films using AP3-11 were prepared at the different subphase temperatures (21.5 and 12.0 • C).The lowering the subphase temperature stabilized the AP3-11 monolayer, e.g., the squeezing out the AP3-11 molecules from the air/subphase interface into the DSAB/subphase interface was observed at 21.5 • C but not at 12.0 • C.However, the fraction of the vertically aligned helical-APs in the film did not increase at the lower temperature.
To investigate the effect of the length of the hydrophobic domain on the film structure, the L and LB films were fabricated using the three types of APs at the subphase temperature of 21.5 • C. The AP with the longer hydrophobic domain provided the more stable monolayer due to increase in hydrophobicity, e.g., the squeezing out the AP3-11 molecules from the air/subphase interface into the DSAB/subphase interface was not observed in the cases of AP3-15 and AP3-19.The AFM observations revealed that the phase separation occurred between the AP and DSAB molecules for all the APs.The shape of the AP domain was significantly depended on the length of the hydrophobic domain; AP3-11 exhibited the circular domains, AP3-15 the curved trench domains (liquid crystalline-like phase), and AP3-19 exhibited the rectangular-like domains (crystalline-like phase).The film structure and the orientation of the helical APs also depended on the length of the hydrophobic domain.When the hydrophobic domain consisted of 11 AAs (AP3-11), the helices did not show a uniform orientation but the mixture of parallel and vertical orientations as shown in the model b (Fig. 5 Although the ambiguity in the modeling of the A3-15/DSAB mixed monolayer cannot be eliminated completely, e.g., the transfer ratio on the LB deposition was low, the orientation of the AP3-15 helices was the most similar structure to what we expected, i.e., the uniformly oriented AP helices with their tilt angle of 27 • from the film normal, and the other APs did not exhibit a uniform orientation.Thus, we conclude that the balance between the lengths of the hydrophilic and hydrophobic domains in the AP is more important factor to align the APs vertical in the AP/lipid mixed monolayer than the subphase temperature.
FIG. 5: Models of the helix orientations in the mixed monolayer.(a) Model a; uniform orientation model.(b) Model b; mixture of parallel and perpendicular orientations.The parallel orientated APs are located at the DSAB/subphase interface.(c) Model c; mixture of parallel and perpendicular orientations.All the APs are located at the air/subphase interface.
(2) and(3).Alhttp://www.sssj.org/ejssnt(J-Stage: http://www.jstage.jst.go.jp/browse/ejssnt/) Volume 10 (2012) Togashi, et al.TABLE IV: Obtained θ in model a and R ⊥ : R // in model b or c, and calculated molecular areas of AP (AAP−a, A AP−b , and AAP−c) according to the models a, b, and c, respectively.Peptide θ in model a R ⊥ : R // in model b or c the model c, the horizontally orientated APs are assumed to form a bilayer.b) In the model c, the horizontally orientated APs are assumed to form a monolayer.
• and R ⊥ > R // .Accidentally, the mixed film of AP3-11 at 21.5 • C and that of AP3-15 exhibited the same values; θ = 27 • and R ⊥ : R // = 3.9 : 1.These values indicate that the APs are almost vertically oriented in the model a or the 4/5 of APs are vertically oriented in the model b or c.According to the values of θ and R ⊥ : R // , the molecular areas of the AP based on each model (A AP−a , A AP−b , and A AP−c ) were calculated (see Table
(b)), and the APs oriented parallel to the film plane were located beneath the DSAB phase.When the 15 AAs formed the hydrophobic domain (AP3-15), the APs exhibited a uniform orientation as shown in the model a (Fig. 5(b), θ = 27 • ).Further elongation of the hydrophobic domain (AP3-19) resulted in the mixture of parallel and vertical orientations of helical APs as shown in the model c (Fig. 5(c)), and the fraction of the AP parallel to the film plane was larger than the case of AP3-11.

TABLE I :
Primary structures and average hydropathies of the amphiphilic peptides (APs).The underlined part corresponds to the hydrophobic domain of the AP.

TABLE II :
Fraction of the secondary structures of the APs in the 10 mM Tris-HCl buffer with and without 20 mM SDS.

TABLE V :
Area ratios obtained from the AFM images and the π-A isotherms.

TABLE VI :
Area ratios calculated from the model a, b and c using the values of θ and R ⊥ : R // listed in TableIV.between the AP and DSAB phases using the values of θ and R ⊥ : R // in TableIV, and the results are shown in Table VI.In the case of the AP3-11/DSAB mixed monolayer prepared at 21.5 • C, the ratio calculated from the model b corresponds well to the experimentally obtained ratios in Table V.Because the model b also gave the closest value of the apparent molecular area (compare Tables tios