Hydration, Dehydrative Counter-ion Binding and Helix Formation of Charged Poly(a-amino acid)s in Aqueous Alcohol as Revealed by a Preferential Binding Study BY TOSHIO MORI, JIRO KOMIYAMA" AND TOSHIRO IIJIMA Department of Polymer Science, Tokyo Institute of Technology, Ookayama, Meguro-ku, Tokyo 152, Japan Received 8th September, 1977 The preferential binding behaviour of one of the components of aqueous alcohol solutions to sodium poly(or-L-glutamate) (PLGNa) and poly(a-L-lysine HBr) (PLLHBr) at 25°C has been studied by differential refractometry. Theaqueous mixtures have 1-propanol(1 -PrOH), 2-propanol(2-PrOH), t-butanol (t-BuOH) and 1,2ethanediol (EG) as their second component. The electrical conductance and the or-helix content of these polymers were also measured to interpret the preferential binding behaviour.Aqueous monohydric alcohols show a characteristic dependence on the alcohol concentration : the initial trend of the increasingly negative binding of the alcohols is reversed at certain alcohol concentrations, giving clear breaks. The results are interpreted invoking an operational model of these polymers, in which polar and less polar portions undergo solvation separately. PLGNa residue is hydrated by 20 water molecules, of which 12 are taken up by the polar portion and the rest by the less polar portion. PLLHBr residue is hydrated by 27 water molecules, of which 15 are taken up by the polar portion. With increase in the alcohol concentration, the hydration of the polar portions of these polymers is lost upon counter-ion binding, which is followed by the helix formation of the polymers.5 water molecules are released from around the less polar portions upon formation of the helix. In a previous paper, we reported the preferential binding behaviour of one of the components of aqueous organic solvents or organic solvent binary mixtures to PLLHBr.l The variation in the preferential binding parameters for the nine binary solvent mixtures was examined over different composition ranges and was interpreted by invoking an operational model of PLLHBr, in which polar and less polar portions of the polymer undergo solvation separately. Definition of these portions is rather diffuse, but we assume that the charge conveying group of the polymer and the counter-ion are included in the polar portion, and the main chain groups and the methylene groups of the side chain are included in the less polar portion.Of particular interest among the results is the preferential binding of 2-PrOH from the aqueous mixtures : the binding is increasingly negative with concentration until this trend is reversed from a break at around a mole fraction of 0.30 of the alcohol in the solvent mixture. We have shown that the negative preferential binding before the break is explained if the polymer residue is exclusively hydrated by 27 water molecules, against the composition change in the solvent mixture. We have speculated that the reverse trend may be associated with dehydration due to the counter-ion binding as well as to the a-helix formation of the polymer. Such extensive hydration of PLLHBr should be established firmly.Extended research on the preferential binding behaviour to another charged poly(a-amino acid) from aqueous alcohol solutions may give some information about the hydration as influenced by the changes in the states of the relevant polymers. In the present study, the preferential binding 1-82 25832584 PREFERENTIAL BINDING TO CHARGED POLY (AMINO ACID)S measurements for PLLHBr were supplemented for the aqueous t-BuOH system and the binding to PLGNa from aqueous I-PrOH, 2-PrOH and t-BuOH was investigated. For comparison, measurements for PLGNa in aqueous EG and dimethyl sulphoxide (DMSO) were included in the study. EXPERIMENTAL MATERIALS PLGNa was obtained by saponification of poly(y-methyl-a-L-glutamate) which was prepared according to Oya et aL3 The polymer, after being washed with acetone and dried, was dissolved in water and recovered by precipitation in 2-PrOH at 5°C.The impurities and low molecular weight fraction of the polymer were removed by electrodialysis of the aqueous solution for 50 h, followed by ultrafiltration through a Diafilter G 10 T (Bio- engineering) which is impermeable to polymers of molecular weight >lo4. The polymer was recovered by freeze drying and stored in a desiccator until use. The molecular weight as determined by the viscosity measurements was 1.1 x lo5. The preparation and purifica- tion of PLLHBr was described in the previous paper. The organic solvents (analytical grade) obtained from Wako Pure Chemicals were dried by anhydrous Na2S04 and fractionally distilled using an efficient column under the appropriate pressures.The solvents were stored in desiccators ; their densities agreed within & 5 x g ~ m - ~ with the literature values in table 1, in which other physical constants of the solvents are listed. Laboratory supply water, which was distilled after deionization, was used ; its conductance was (1-2) x Q-' cm-l. TABLE 1 .-PROPERTIES OF SOLVENTS HzO 1-PrOH 2-PrOH t-BuOH EG dielectric constant, (25°C) 78.4 20.3 19.9 12.5 37.7 dipole moment (25°C) 1.85b 3.09 1.66 1.66 2.28 (20°C) (30°C) (3OoC) (2OoC) refractive index, (25°C) 1.3325 1.3837 1.3752 1.3851 1.4306 density/g ~ m - - ~ (25°C) 0.9970 0.7998 0.7813 0.7812 1.1100 a Taken from ref. (5) unless specified; b ref. (6); C ref.(7). DMSO 46.7 3.89 1.4773 1.0958 APPARATUS AND METHODS The apparatus and procedure of equilibrium dialysis have been described e1sewhere.l The PLGNa concentration in the dialysis equilibrium was determined by adding an aliquot of 1 /200 residue mol dm-3 aqueous glycol chitosan solution, followed by colloidal titration with 1 I400 residue mol dm-3 aqueous potassium poly(vinylsulphonate), using toluidine blue as indicator.8 Permeation of PLGNa through the dialysis bag (Visking) was found to be <1 % of the total polymer. Circular dichroism measurements were performed using a Jasco J 20 automatic spectro- polarimeter at 250°C. The instruments were calibrated with (+ )-10-camphorsulphonic acid. The helix content, xH, of PLGNa and PLLHBr in the aqueous alcohol solutions was determined by measuring the variation of the residue ellipticity of the 1 .2 0 ~ residue mol dm-3 solutions at 222 nm, [6]222/deg cm2 dmol-l and using the r e l a t i ~ n , ~ Conductance measurements were made with a Yokogawa Universal Bridge BV-Z-103B at 1 kHz at 25.00"C. The cell constant was determined before each set of experiments using standard aqueous potassium chloride solution. The specific conductances were determined for solutions of both polymers whose concentration was fixed at 1.65 x residue mol dm-3, and therefore the equivalent conductances, A, were calculated on the residue mole bases.T. MORI, J . KOMIYAMA AND T. IIJIMA 2585 Corrections due to the solvent conductances were applied to all the conductance measure- ments and these amounted to <10 % of the conductances of the solutions in most cases.The viscosities of aqueous EG,l0s l1 l-PrOH,lO* l1 2-PrOH l2 and t-BuOH l3 were taken from 1iterat~re.l~ PREFERENTIAL BINDING PARAMETER The preferential binding parameter (in mole alcohol per residue mol polymer) of an alcohol from the aqueous mixture to polymer at 25°C was determined according to eqn (2) l5 Here subscripts 1, 2 and 3 refer to water, polymer and alcohol, respectively. n is the refractive index, rn is the molarity of the respective component in water and C the concentra- tion of solution in g ~ r n - ~ . Vis the partial specific volume, ,u is the chemical potential, A4 is the molecular weight (per residue for the polymers) and the superscript indicates infinite dilution of the polymer.Partial specific volumes were calculated graphically from specific volume against concentration plots. The specific volume was measured by an Anton Parr precision density meter DMA 02C, at 25.00"C. Though the preferential binding parameters have been expressed by the above quantity in this report in conformity with the previous one, the values were found to be negative irrespective of the alcohol species over the concentration ranges investigated. The following relation may be used to obtain (aml/dmz)~,,p3, the preferential hydration parameter, from the values reported here l5 The refractive index difference was measured at 25.00"C by a Carl Zeiss Jena laboratory interferometer. (dn/aC2)p1,f13 and (an/aC,),, were measured for the polymer solutions in the concentration range (1-5) x g cm-3 when the alcohol volume of the solvent mixture was <60 %.Greater than this alcohol content, the polymers would not dissolve in the solvent mixture, giving the solution a slightly opaque appearance. The solution was prepared therefore initially by dissolving the polymer in water and then by adding alcohol gravimetrically to the solution with care to attain sufficient accuracy. Measurements of the refractive index increments were made for the duplicate polymer solutions in the concentration range (0.5-2) x g ~m--~. RESULTS AND DISCUSSION the systems PLGNa in aqueous 1-PrOH, 2-PrOH, t-BuOH, EG and DMSO are shown in table 2. The values for the system PLLHBr in aqueous t-BuOH are listed in table 3. The preferential binding parameters for PLGNa in aqueous organic solvents are plotted against x3, the mole fraction of the organic solvent component in the solvent mixture, in fig. 1.The plots of the parameters for PLLHBr in aqueous t-BuOH are shown in fig. 2, together with those in aqueous 2-PrOH reported previously. Fig. 1 shows that the preferential binding parameter of EG to PLGNa increases with EG concentration, and after passing a small maximum at x3 N 0.1, it is inverted by a change to preferential hydration at x3 N 0.25. The preferential binding of DMSO to PLGNa is found to be increasingly negative with the content, though the extent is moderate. In this case, the polymer precipitated if x3 exceeded 0.28. Fig. 1 shows the remarkably strong negative dependence of the preferential binding parameters of the three monohydric alcohols to PLGNa.Before the breaks which appear at x3 N 0.27, the points for each alcohol are found close together giving a decreasing trend, while after the breaks, they give different increasing trends. The values of C3, J!3, (an/ac2)fl,,p3, ( a n / a c 2 ) m 3 , (anlacs)m, and (am,lam2),",,,3 for2586 PREFERENTIAL BINDING TO CHARGED POLY (AMINO AC1D)S TABLE 2.PREFERENTIA.L BINDING OF ORGANIC SOLVENTS TO PLGNa FROM AQUEOUS MIXTURES solvent (3) vol. % x3 10 20 30 40 60 70 75 85 10 20 30 40 50 60 70 75 10 20 30 50 60 70 75 80 10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 0.026 0.057 0.093 0.138 0.265 0.359 0.41 9 0.575 0.025 0.055 0.091 0.135 0.190 0.260 0.354 0.41 6 0.020 0.043 0.066 0.158 0.219 0.304 0.359 0.429 0.034 0.074 0.121 0.176 0.241 0.326 0.423 0.561 0.739 0.028 0.060 0.097 0.143 0.198 0.272 C3/g cm-3 0.08 16 0.1626 0.2446 0.3266 0.4894 0.5696 0.6095 0.6878 0.0789 0.1585 0.2407 0.3224 0.4030 0.4764 0.561 1 0.6015 0.0772 0.1535 0.2152 0.3977 0.4775 0.5559 0.5949 0.6343 0.1095 0.2223 0.3335 0.4461 0.5587 0.6733 0.7789 0.8934 1.0014 0.1 105 0.2221 0.33 1 1 0.4439 0.5535 0.6703 (A) H,O(l)+ 1-PrOH(3) 1.1211 0.174 1.1477 0.143 1.1740 0.133 1.2078 0.111 1.2172 0.089 1.2288 0.103 1.2321 0.107 1.2370 0.1 19 (B) H20(l)+ 2-PrOH(3) 1.1140 1.1340 1.1597 1.1920 1.2098 1.2370 1.2522 1.2687 0.167 0.140 0.130 0.109 0.103 0.097 0.117 0.119 (C) HzO(l)+ t-BuOH(3) 1.1305 1.1511 1.1539 1.2285 1.2444 1.2596 1.2596 1.2668 0.171 0.142 0.127 0.084 0.082 0.090 0.100 0.103 0.8640 0.8682 0.8731 0.8787 0.8871 0.8906 0.8441 0.8982 0.8997 0.182 0.176 0.166 0.150 0.138 0.126 0.120 0.112 0.loo 0.191 0.181 0.171 0.157 0.141 0.136 0.131 0.128 0.185 0.174 0.166 0.151 0.143 0.135 0.129 0.127 0.191 0.176 0.168 0.138 0.132 0.128 0.124 0.123 0.172 0.161 0.152 0.142 0.137 0.132 0.130 0.126 0.121 (E) H20( 1)+ DMSO(3) 0.8592 0.157 0.180 0.8591 0.131 0.170 0.8628 0.120 0.161 0.8634 0.092 0.151 0.8749 0.073 0.141 0.8859 0.045 0.130 /cm3 g-1 0.091 0.084 0.064 0.056 0.048 0.043 0.040 0.036 0.097 0.081 0.073 0.062 0.046 0.036 0.032 0.032 0.111 0.098 0.074 0.057 0.047 0.040 0.038 0.035 0.090 0.090 0.089 0.088 0.087 0.084 0.083 0.082 0.081 0.118 0.133 0.133 0.148 0.152 0.157 (am ,/am 2); 1 ,ps /mol mol-1 - 0.52 - 1.40 - 2.09 - 3.41 - 6.74 - 6.43 - 6.06 - 4.21 - 0.51 - 1.29 - 1.72 - 2.77 - 4.27 - 5.78 - 3.43 - 2.65 - 0.40 - 0.91 - 1.50 - 3.78 - 5.35 - 6.42 - 5.14 - 5.93 0.30 0.50 0.54 0.36 0.06 - 0.43 - 0.97 - 2.25 - 6.37 - 0.42 - 0.70 - 0.84 - 1.25 - 1.68 - 2.58T .MORI, J . ROMIYAMA AND T . IIJIMA 2587 TABLE 3.-PREFERENTIAL BINDING OF t-BuOH TO PLLHBr FROM AQUEOUS MIXTURES ( ~ a ~ 2 ) ~ ~ , ~ ~ (an/acz),, ( a n / a ~ ) , , , ~ (8m318m2)E~,~~ /cmJ g-1 /cm3 g-1 /mol mol-1 - solvent (3) vol. % x3 C3/g cm-3 Y3/cm3 g-1 /cm3 g-1 (F) H20(1)+ t-BuOH(3) 10 0.020 0.0772 1.1305 0.160 0.178 0.111 - 0.50 20 0.043 0.1535 1.1511 0.123 0.160 0.098 - 1.29 40 0.110 0.3166 1,1990 0.086 0.140 0.064 - 3.83 50 0.158 0.3977 1.2285 0.077 0.134 0.057 - 5.52 60 0.219 0.4775 1.2444 0.080 0.130 0.047 - 7.40 70 0.304 0.5559 1.2561 0.085 0.126 0.040 - 9.58 80 0.429 0.6343 1.2668 0.111 0.124 0.033 - 5.33 Such characteristic preferential binding behaviour has been reported for PLLHBr in aqueous 2-Pr0H.I Fig.2 shows that the preferential binding of t-BuOH to PLLHBr from the aqueous solution also conforms to this behaviour pattern. Summarizing these results, the charged poly(a-amino acid)s in aqueous monohydric alcohol solutions 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 x3 FIG. 1 .-Dependence on solvent composition of preferential binding of organic solvent to PLGNa ; in aqueous 1-PrOH (O), ZPrOH (A), t-BuOH (0), EG (0) and DMSO (V). The lines for aqueous EG and DMSO were drawn through the experimental points. The line for aqueous monohydric alcohol solutions was drawn assuming a hydration number of 20 for PLGNa. See text for details.Tanford l6 and Inoue and Timasheff have related the preferential binding parameter to the actual solvation number as (d1nJ/am2);1,pj = n’3-(x3ixl> (4) j j where n{ and n’J are the solvation numbers ofjth portion of the polymer by component 1 and 3, respectively. In terms of this equation, the linear relation shown in fig. 3 means that En4 = 0, i.e., the polymer is not solvated by the alcohols and the slope gives the hydration number which is constant with the change in solvent composition. Linear regression analysis gave the slopes as 19, 20 and 21 for aqueous 2-PrOH, t-BuOH and 1-PrOH, respectively. However, the shortage of points for each series and the errors inherent in these measurements exclude attachment of any significance to these differences ; hence we conclude that PLGNa is hydrated by20 water molecules2588 PREFERENTIAL BINDING TO CHARGED POLY (AMINO ACID)S in aqueous monohydric alcohol solutions.The hydration number, 27, of PLLHBr in aqueous 2-PrOH,' was confirmed for aqueous t-BuOH as shown by the plot in fig. 4. The persistence of these hydrations indicates that the hydrated layers of polymer including the counter-ions are not perturbed unless the alcohol content of 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 x3 FIG. 2.-Dependence on solvent composition of preferential binding of organic solvent to PLLHBr ; in aqueous 2-PrOH (A), and t-BuOH (0). The line was drawn assuming a hydration number of 27 for PLLHBr. See text for details. the solvent mixtures exceeds certain thresholds. It is also recalled that analysis using a definite solvation number was successfully applied to the preferential binding behaviour of PLLHBr in aqueous organic and organic solvent binary mixtures in the previous rep0rt.l The bulk dielectric constants of the solvent mixtures investigated in this and the previous study never fell below 40, except for the aqueous monohydric 0 r( 4 I 8 -2.0 8 CI -- -4.0 3 0; n (0 $' -6.0 c$ -8.0 \ m 0 A 0 0 A 00 0 I I I I I I 1 ' <L 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 1.3 1.4 x3 /XI xni = 20.FIG. 3.-Plot of preferential binding parameter against x3/x1 for PLGNa in aqueous 1-PrOH (o), 2-PrOH (A) and t-BuOH (0). The line was drawn according to eqn (2) assuming 2n3 = 0 andT. MORI, J . KOMIYAMA AND T. IIJIMA 2589 alcohol solutions at their higher alcohol contents.From the electrostatic point of view, the lowering of the dielectric constant is primarily concerned with the binding of the counter-ions by the polymer charges. If the binding is such that dehydration of the pairing ions occurs, the preferential binding behaviour must reflect this, and we discuss below the preferential binding behaviour after the breaks in this context. Therefore no dehydrative counter-ion binding takes place for these charged poly(a- amino acid)s when the bulk dielectric constant of the solvent mixture is above a threshold, ~ 4 0 . This argument has been added here because dehydrative counter- ion binding has so often been referred to in discussions of the electrolytic behaviour of poly(carboxy1ic acid)s, 8-21 including PLGNa, in aqueous solution.I 0 - 2.0 I 2 -4.0 - M \ 8 g -10.0 m -6.0 5 2 -8.0 0s (D \ -12 .o 0 A A I I I I I I I ' c 4 c L 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80.9 1.3 x3 1x1 FIG. 4.-Plot of preferential binding parameter against x3/x1 for PLLHBr in aqueous 2-PrOH (A) and t-BuOH (0). The line was drawn according to eqn (2) assuming En3 = 0 and Cnl = 27. The exceptionally large hydration numbers assigned to PLGNa and PLLHBr in the aqueous monohydric alcohol solutions need some comment. Kauzmann and Kuntz 22 pointed out that the hydration numbers of polypeptides depend critically on the method of investigation and the state of the polymer being measured. Kuntz 23 gave 4.5 and 7.5, respectively for PLGK and PLLHCl residue hydrations, by n.m.r. measurements. Breuer and Kennerly 24 and recently Rochester and Westerman 25 calculated these numbers as 3 and 4 for PLGNa and PLLHBr from the water sorption isotherms of these polymers.From the buoyant density measurements of PLGCs and PLLHBr in aqueous CsCI, Ifft and coworkers 26 have reported values of 6.5 and 16.5 for the residue hydrations. These instances show that at least several water molecules are " strongly " bound to the charged groups and the counterions of these poly(a-amino acid)s. In addition, the hydration of the peptide bond has often been referred to in the literature. Though the large hydration numbers found for PLGNa and PLLHBr in the present study in principle include the excess water molecules in unspecified regions around the polymers, they are probably located in the immediate neighbourhood of the polymer entities, by thermodynamic preference, not necessarily by strong interaction with the charges or the polar groups of the polymers.Their differences from the hydration numbers cited above may be explained by this fact. It is questionable whether the hydrations found in this study persist in aqueous solution.2590 PREFERENTIAL BINDING TO CHARGED POLY (AMINO AC1D)S TABLE 4.-ELECTRICAL CONDUCTANCE OF PLGNa AND PLLHBr IN AQUEOUS ALCOHOL SYSTEMS alcohol A* VOI. % x3 /cmz Ja-1 equiv.-1 q/mW (A) PLGNa (1) H20+ 1-PrOH 0 10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 70 75 80 85 90 10 20 30 40 50 60 70 75 80 85 10 20 30 40 50 60 70 80 90 O.OO0 0.026 0.057 0.093 0.138 0.192 0.265 0.359 0.491 0.681 43.8 31.9 23.2 17.2 13.2 9.24 6.61 2.72 1.12 0.26 (2) H20+ 2-PrOH 0.025 0 055 0.091 0.135 0,190 0.260 0.354 0.416 0.484 0.571 0.679 34.3 22.7 15.8 10.8 7.40 5.15 2.30 1.27 0.77 0.45 0.19 (3) H20+ t-BuOH 0.020 0.043 0.066 0.110 0.158 0.219 0.304 0.359 0.429 0.519 0.034 0.074 0.121 0.176 0.241 0.326 0.423 0.561 0.739 31.0 19.5 13.4 8.96 5.40 2.66 1.16 0.82 0.30 0.10 (4) HzO+EG 34.4 25.5 19.5 15.6 11.3 7.81 5.75 4.21 2.80 8.91 12.8 16.3 21.0 24.4 26.2 27.1 26.9 25.0 22.9 12.9 18.1 23.2 27.8 30.1 31.0 30.1 28.6 27.1 25.1 23.2 13.0 20.0 25.8 34.4 40.4 45.1 47.8 48.2 47.8 46.6 11.5 15.0 20.5 26.6 34.1 48.0 68.0 104.0 160.0T.MORI, J . KOMIYAMA AND T. IIJIMA 259 1 TABLE 4.-contd. alcohol A*/ vol. % x3 /an2 a-1 equiv.-1 q/mPt (B) PLLHBr (1) H20+ 2-PrOH 0 O.Oo0 54.9 8.91 20 0.055 28.1 18.1 40 0.135 14.3 27.8 60 0.260 8.31 31.0 70 0.354 5.78 30.1 80 0.484 3.92 27.1 90 0.679 1.95 23.2 (2) HzO+ t-BuOH 10 20 30 40 50 60 70 75 80 90 0,020 0.043 0.066 0.110 0.158 0.219 0.304 0.359 0.429 0.632 37.2 25.8 18.8 13.1 7.84 4.95 3.20 1.94 1.10 0.37 13.2 20.0 25.8 34.4 40.4 45.1 47.8 48.2 47.8 45.3 * Measured for the solutions of PLGNa and PLLHBr at 1.65 x residue mol dm-3.Visco- sities were taken from ref. (14). Recently, Inoue and Izumi 27 reported negative preferential binding of lY4-dioxan to poly[N5-(a-hydr~xypr~py1)-~-glutamine], a neutral poly(a-amino acid), from aqueous mixtures. Replotting their results in terms of eqn (4) indicated that 12 or more water molecules are in excess around the polymer residue in 0 to 70 weight % dioxan solutions. It seems that the large excess of water molecules are not specific to the charged poly(a-amino acid)s in aqueous monohydric alcohol solutions.The reversal of the decreasing trends in preferential binding in aqueous mono- hydric alcohol solutions will now be discussed. Fig. 1 and 2 show that breaks appear at x3 - 0.27 for PLGNa and at x3 - 0.29 for PLLHBr, irrespective of the alcohol species. The dielectric constants of the solvent mixtures at these points are 40, 40 and 29 for aqueous 1-PrOH, 2-PrOH and t-BuOH at x3 = 0.27, and are 38 and 27 for aqueous 2-PrOH and t-BuOH at x3 = 0.29, re~pective1y.l~ After these points, the preferential binding of the alcohol begins to increase, or the preferential hydration to decrease, with the alcohol content. The decreasing trend differs according to the alcohol species ; it is much sharper for PLGNa in aqueous 2-PrOH than in the other aqueous alcohol solutions and is also sharper for PLLHBr in aqueous t-BuOH than in aqueous 2-PrOH.If we compare this behaviour with the helix contents of the polymers shown in fig. 7 and 8, in some cases, the helix formation starts at a signifi- cantly higher value of x3 than that where the breaks are located and this parallels the less drastic decrease in preferential hydration. These results suggest two possible reasons for the reversed trends : (i) by dehydration accompanied by counter-ion binding and (ii) by the helix formation of the po1ymers.l. 2 8 In terms of the solvation model proposed in the previous paper, water molecules bound to the polar portion of the relevant polymer are released upon counter-ion binding and a fraction of those bound to the less polar portion are released upon helix formation.According to2592 PREFERENTIAL BINDING TO CHARGED POLY (AMINO ACID)S eqn (4), the remaining hydration numbers at each stage are estimated from the slopes of the straight lines connecting the relevant points to the origin in fig. 3 and 4 and are shown in fig. 7 and 8. The difference from the full hydration number gives the hydration loss, which can be resolved into two contributions if the degree of the F - 8 3.0 b 0 2.0 1.0 I I I I I I I 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 x3 FIG. 5.-Dependence on solvent composition of Walden products of PLGNa in aqueous 1-PrOH (O), 2-PrOH (A), t-BuOH (0) and EG (0). * Represents the value in water.The broken line indicates 1/10 of the constant value at lower alcohol concentrations. counter-ion binding is known. The electrical conductance of the polymers in aqueous alcohol solutions was measured for this purpose. The results are given in table 4 and are shown in fig. 5 and 6 as the Walden products. In aqueous EG, whose dielectric constant is relatively high at over 40, the products for PLGNa are essentially constant. In aqueous monohydric alcohol solutions, the products for both polymers x3 FIG. 6.-Dependence on solvent composition of Walden products of PLLHBr in aqueous 2-PrOH (A) and t-BuOH (0). * Represents the value in water. The broken line indicates 1 /lo of the constant value at lower alcohol concentrations.T . ICIORI, J . KOMIYAMA A N D T.IIJIMA 2593 deviate downward from x3 - 0.07, become about a half of the constant value as x3 is increased to 0.18-0.24, and finally fall to < 1/10 at the highest alcohol contents. In fig. 7 and 8, the numbers in the symbols indicate the points which were measured when the corresponding Walden products became <1/10 of the constant value at lower alcohol contents. The small values suggest that the dehydrative counter-ion binding is close to completion. If this is assumed, the points denotecl by *1 and *2 in fig. 7 should give the hydration number of PLGNa, which has essentially full counter-ion binding and also maximum helical conformation. Subtraction of this 2- 9'' f 1 I I I 4' I I I I ; ; I 1 ; 0 0.1 0.2 0.3 0.4 0.5 0.6 x3 FIG. 7.-Dependence on solvent composition of hydration number and helix content of PLGNa in aqueous 1-PrOH (o), 2-PrOH (A) and t-BuOH (0).number, 3, from 20 gives the cumulative dehydration number by these processes as 17. Points indicated by *3 and "4 in fig. 7 give 8 as the hydration number of PLGNa with the same counter-ion binding but with only a small fraction of helical content ; hence the dehydration number due to the counter-ion binding is 12. The difference between these numbers gives 5 as the dehydration number upon the helix formation of PLGNa. In terms of the solvation model, the total hydration number, 20, of PLGNa, is divided into 12 for the polar portion and 8 for the less polar portion, of which 5 water molecules are released upon helix formation. The same calculation can be applied to PLLHBr data.The points denoted by *6 in fig. 8 should give the2594 PREFERENTIAL BINDING TO CHARGED POLY (AMINO ACID)S hydration number of helical PLLHBr with full counter-ion binding. This number, 7, gives the dehydration number due to the two processes as 20, which may be divided into 15, previously given as the hydration number for the polar portion and 5, as the dehydration number due to helix formation. The last number is comparable with 6 , the desolvation number for PLLHBr upon helix formation in dimethyl- formamide + DMSO mixtures.l The comparison of hydration, conductance and helix content, sheds further light on the influence of the polymer charge on helix formation. Fig. 7 shows that, in aqueous 1-PrOH and t-BuOH, the helix formation of PLGNa takes place after the C I I I Q r I 0 0.1 0.2 0.3 0.4 0.5 0.6 i o o 80 6 0 4 0 2 0 x o ' = 3 8 2 - .Y .-( x3 FIG.8.-Dependence on solvent composition of hydration number and helix content of PLLHBr in aqueous 2-PrOH (A) and t-BuOH (0). dehydrative counter-ion binding is completed ; while in 2-PrOH, the formation precedes completion and fig. 8 shows that this is the case for PLLHBr in both aqueous alcohol solutions. For the last three cases, the hydration numbers at the 5 % helix level were read from the figures as 10 for PLGNa in aqueous 2-PrOH, 18 for PLLHBr in aqueous t-BuOH and 19 for PLLHBr in aqueous 2-PrOH. The differences between these numbers and the total solvation numbers of the respective polymers give the dehydration numbers, from which the degree of counter-ion binding is obtained as 80 % for PLGNa in aqueous 2-PrOH, 60 and 50 % for PLLHBr in aqueous t-BuOH and 2-PrOH.In all cases, the dehydrative counter-ion binding proceeds in preference to the helix formation, implying that fractions of the chargesT. MORI, J . KOMIYAMA AND T. IIJIMA 2595 on the polymer side chains are rendered dormant by the binding, whereby the electro- static repulsion amongst the charges is reduced. In a given aqueous alcohol, PLLHBr begins to form a helix under a higher charge density than PLGNa: a reasonable result considering that PLLHBr has the longer side chain by two methylenes. In the past, many investigations have been made on the helix formation of charged poly(a-amino acid)s in aqueous monohydric alcohol solutions.Barteri and Pispisa studied the helix formation of PLLHBr in aqueous 2-PrOH. Though they interpret this as " fully charged helix formation " and this is the current view of the helix in aqueous alcohol the term seems to be misleading. Their speculation that extensive dehydration from around the charged sites on the polymer induces the helix formation is valid in the sense that such a dehydration is accompanied by counter-ion binding and so is indirectly concerned with the helix formation. Among the three aqueous monohydric alcohol solutions, aqueous t-BuOH has the lowest dielectric constant at a given composition. However, as found in fig. 7, the dehydrative counter-ion binding for PLGNa in aqueous 2-PrOH takes place at lower alcohol contents than in aqueous t-BuOH, indicating the ease of dehydration in the former mixture.The dehydration from around the less polar portion of the polymer should also be connected with the energetics of the helix formation. Fig. 7 shows that the helix formation of PLGNa in aqueous 2-PrOH takes place at lower alcohol contents than in aqueous 1-PrOH. Since the dielectric constants of these aqueous alcohol solutions are the same at any composition investigated, it is suggested that this dehydration in aqueous 2-PrOH is favoured at lower alcohol contents than in aqueous 1 -PrOH. This research was supported in part by the Scientific Research Fund of the Japanese Ministry of Education. The authors are indebted to Dr. Hisashi Uedaira of the Research Institute for Polymers and Textiles for the use of a Carl Zeiss inter- ferometer.J. Komiyama, T. Mori, K. Yamamoto and T. Iijima, J.C.S. Faraday I, 1977, 73,203. Y. Koiwa and Y. Fujimoto, 22nd Disc. Meeting SOC. ofPolymer Sci (Tokyo, 1973). Y. Iwakura, K. Uno and M. Oya, J. Polymer Sci., 1967,5,2867. R. J. Hawkins and A. Holtzer, Macromolecules, 1972,5,294. J. A. Riddick and W. Bunger, Techniques of Chemistry, Vol. 11, Organic Solvents, ed. A. Weissberger (Wiley-Interscience, New York, 1970). Handbook of Chemistry and Physics, ed. R. C. Weast (Chemical Rubber Co., Cleveland, Ohio, 48th edn, 1968). C. W. N. Cumper, J. F. Read and A. I. Vogel, J. Chem. SOC., 1965, 5323. G. Holzwarth and P. Doty, J. Amer. Chem. Soc., 1965,87,218. * H. Terayama, J. Polymer Sci., 1952, 8,243. lo A. E. Dunstan, 2. phys. Chem. A, 1905,51,732. l1 A. E. Dunstan, J. Chem. Soc., 1905,87,11. l2 A. L. Olsen and E. R. Washburn, J. Phys. Chem., 1938,42,275. l3 E. P. Irany, J. Amer. Chem. Soc., 1943,65,1396. l The Physico-chemical Constants of Binary Systems in Concentrated Solutions, ed. J. Timmermans (Interscience, New York, 1960), vol. N. l5 H. Inoue and S. N. Timasheff, J. Amer. Chem. SOC., 1968,90,1890. l6 C. Tanford, J. Mol. Biol., 1969,39, 539. l7 H. Inoue and S. N. Timasheff, Biopolymers, 1972, 11,737. l 8 G. S. Manning and A. Holtzer, J. Phys. Chem., 1973,77,2206. l 9 A. Ikegami, Biopolymers, 1968,6,431. 'O C . Tondre and R. Zana, J. Phys. Chem., 1972,76,3451. 21 J. Komiyama, M. Ando, Y . Takeda and T. Iijima, Europ. Polymer J., 1976, 12,201. '' I. D. Kuntz and W. Kauunann, Adv. Protein Chem., 1974,28,239. 23 I. D. Kuntz, J. Amer. Chem. Soc., 1971,93,514. 24 M. M. Breuer and M. G. Kennerley, J. Colloid Interface Sci., 1971,37, 124.2596 PREFERENTIAL BINDING TO CHARGED POLY (AMINO ACID)S 2 5 C. H. Rochester and A. V. Westerman, J.C.S. Faraday I, 1976, 72, 2753. 26 R. Almassy, J. S. V. Zill, L. G. Lum and J. B. Ifft, Biopolymers, 1973, 12, 2713. 27 H. Inoue and T. Izumi, BiopoZymers, 1976, 15,797. 28 G. Nemethy and H. A. Sheraga, J. Chem. Phys., 1962,36, 3401. 29 M. Barteri and B. Pispisa, BiopoZymers, 1973, 12,2309. 30 R. H. Liem, D. Poland and H. Scheraga, J. Amer. Chem. Soc., 1970,92,5717. (PAPER 7/1599)