Thermodynamics of Interactions Between Charged Surfactants and Ionic Poly(amino acids) by Isothermal Titration Calorimetry.

Interactions between charges play a role in protein stability and contribute to the energetics of binding between various charged ligands. Ionic surfactants are charged molecules, whose interactions with proteins are still rather poorly understood despite their wide applications. Here, we show by isothermal titration calorimetry that cationic alkylammonium surfactants bind to negatively charged polyaspartate and polyglutamate homopolymers stoichiometrically, i.e., one surfactant molecule per charged amino acid. Similarly, negatively charged alkyl sulfates (e.g., sodium dodecyl sulfate) and alkane sulfonates bind stoichiometrically to positively charged polylysine, polyornithine, and polyarginine homopolymers. In these reactions, the interacting counterparts form ion pairs and the resulting electrostatically neutral complex coprecipitates from solution. The enthalpies and heat capacities are determined for various pairs of ionic surfactants and charged amino acid homopolymers. These results show the energetic contributions of ionic headgroups and the CH2 group to surfactant interactions with proteins.

Introduction

Interactions between surfactants and proteins are governed by ionic, hydrophobic, and van der Waals forces, but it is difficult to dissect their energetic contributions in protein–ligand binding. The interacting system may be simplified by instead performing the binding reaction with amino acid homopolymers. The homopolymers made of charged amino acids are highly soluble and may resemble the role of this amino acid in real proteins and dissect the contributions of ionic and hydrophobic interactions.

Mixtures of charged polymer and oppositely charged monomer surfactant molecules have broad technological applications.[1−4] Interactions between polymers and ionic surfactants have been studied by various techniques. For example, nonionic water-soluble polymer interactions with ionic surfactants in solution have been investigated by means of conductance, surface tension, dye solubilization, viscosity, and dialysis equilibrium.[5−9] Protein precipitation by detergents[10−13] and more general studies of cationic polymer interactions with anionic detergents[14] were performed exploiting similar techniques to the ones used for nonionic polymers. The circular dichroism technique was used to investigate the detergent-initiated structural rearrangements of poly(amino acid)s from a random coil to an α-helix or β-sheet.[15−21] Light-scattering studies revealed that, for example, cetyltrimethylammonium bromide binding to poly(acrylic acid) is governed by both electrostatic attraction and hydrophobic interactions.[22]

The knowledge about the thermodynamics of polymer–surfactant interactions increased with the development of microcalorimeters, which directly determine the enthalpy of interaction in aqueous solution. Early microcalorimetry studies examined the binding of ionic surfactants, such as sodium dodecyl sulfate (SDS), to various globular proteins.[23−25] Later, isothermal titration calorimetry (ITC) was used to study interactions between SDS and electrically neutral polymers, such as poly(ethylene glycol)[26] and poly(propylene glycol),[27] and other polymer–surfactant systems.[28,29] Wang et al.[30] used ITC to study the adsorption of nonionic, cationic, and anionic surfactants onto polymer lattices with a negative surface charge density.

Li et al.[31] described the role of electrostatic forces by investigating SDS and tetradecyltrimethylammonium bromide interactions with polymers that possess negatively charged groups. Later Wang and Tam[32] have investigated cationic detergent (dodecyltrimethylammonium bromide) binding to anionic polymers, such as neutralized poly(acrylic acid) and methacrylic acid/ethyl acrylate copolymers, using ITC. The authors concluded that electrostatic binding of ionic headgroups to the charged sites on the polymer is enthalpy-opposed and driven by entropy.

Li and Wagner[33] have reported a semiempirical relationship for the cooperative binding in oppositely charged, salt-free polyelectrolyte–alkyl surfactant mixtures for a broad range of systems. The authors have summarized that the cooperative binding affinity of surfactants to oppositely charged polymers is determined by two main factors—hydrophobicity of the surfactant and the charge density of the polymer. In a comprehensive review on such interactions, Khan and Brettmann[34] concluded that the binding of charged polymers to ionic surfactants is driven by both electrostatic and hydrophobic interactions. The authors also reviewed many experimental factors that affect the strength of interaction, but the effects of temperature were not discussed.

Despite numerous applied techniques and various investigated surfactant–polymer systems, full assignment of the energies involved in the recognition between these molecules to molecular functional groups is uncertain. The results here dissect the energetic contributions of binding of several ionic surfactants to polyamino acids and show the additivity of ionic and hydrophobic interactions.

Results and Discussion

For the study of binding between oppositely charged ionic surfactants and ionic amino acid homopolymers by ITC, two surfactant–polymer systems were chosen:

Anionic alkyl sulfates (sodium octyl sulfate, decyl sulfate, undecyl sulfate, dodecyl sulfate, octyl sulfonate, nonyl sulfonate, and decyl sulfonate) with cationic poly(amino acid)s (polyarginine, polylysine, and polyornithine hydrochlorides) (Figure A);

Figure 1

Surfactant and poly(amino acid) systems used in this work. (A) Anionic alkyl sulfates and cationic poly(amino acid)s (polyarginine, polylysine, and polyornithine); (B) cationic alkylamines and anionic poly(amino acid)s (polyaspartate and polyglutamate).

Cationic alkylammonium chlorides (decylammonium, undecylammonium, dodecylammonium, and tridecylammonium) and anionic poly(amino acid)s (polyaspartate and polyglutamate) (Figure B).

Titration of poly(amino acid)s bearing charged side chains with oppositely charged surfactants shows that the binding occurs only until the charge neutralization point is reached, at approximately one surfactant molecule added per amino acid of the polymer, meaning that one surfactant molecule is bound to each amino acid. The interaction strength depended on both the electrostatic attraction between charged groups of the surfactant and polymer, and the hydrophobic interaction between aliphatic tails of surfactants. ITC experiments at different temperatures revealed that constant-pressure heat capacity (ΔCp) values were negative for these interactions. The increase of a surfactant’s aliphatic tail by one CH2 group gives a constant negative contribution to the interaction enthalpy.

Figure shows the raw ITC data (left panels) and isotherms (right panels) of (A,B) SDS binding to poly(Arg+) and (C,D) dodecylammonium binding to poly(Glu–) at T = 25 °C. At this temperature, alkyl sulfate binding to poly(Arg+) is exothermic, whereas dodecylammonium binding to poly(Glu–) shows an endothermic profile. The isotherm of poly(Arg+) binding to SDS contains a distinct dip in the enthalpy at the charge neutralization point (when one surfactant molecule is bound to one amino acid monomer). A similar isotherm shape of SDS titration into polyethyleneimines was observed by Wang et al.[35] The first injection datapoints in Figure B,D are of lower accuracy because of the time needed for thermal equilibration and partial diffusion of the surfactant from the syringe to the cell prior to the first injection. Here, we prefer not to remove the first datapoints, but it should be kept in mind that their accuracy is lower than that of the later datapoints.

Figure 2

Raw ITC data and isotherms of SDS binding to poly(Arg+) (A, B) and dodecylammonium binding to poly(Glu–) (C, D) at neutral pH. The syringe contained 5 mM surfactant solution that was titrated into 0.5 mM (expressed per amino acid) solution of poly(amino acid) at 25 °C.

Stoichiometry and Dependence on Ionic Strength

ITC data analysis revealed that in all reactions between the surfactant and poly(amino acid), the heat is absorbed or released until the charge neutralization point is reached. Further titration produces only dilution heat. This leads to the conclusion that the binding stoichiometry is 1:1; i.e., one molecule of the surfactant binds to a single amino acid moiety of the polymer. We observed the same binding stoichiometry in the majority of the investigated pairs of the surfactant and oppositely charged poly(amino acid), similar to those in Figure , shown in later figures.

The reactions between surfactants and poly(amino acid)s were investigated at various ionic strengths by changing the NaCl concentration. The data in Figure show that 50 mM and 200 mM concentrations of sodium chloride diminished the binding stoichiometry, possibly by blocking access to the charged groups of poly(amino acid). NaCl concentrations of 1 M and higher completely terminated the binding of SDS to poly(Arg+) (see Figure S1 in the Supplementary material). Dodecylamine binding to poly(Glu–) was almost completely stopped by a 25 mM concentration of NaCl (Figure S2 in the Supplementary material). Such differences in the concentration of NaCl that prevent poly(Glu–) and poly(Arg+) interactions with the surfactant can be explained by the high affinity of the guanidinium group to sulfates.[36] Electrostatic interactions between a surfactant and a poly(amino acid) were diminished at an increased ionic strength of solution. High concentration of added salt lowers the critical micelle concentration of ionic surfactants, and micelles may become a dominating state of the surfactant in solution, thus affecting the thermodynamics of binding.

Figure 3

Raw ITC data and isotherms of SDS binding to poly(Arg+) at various concentrations of added NaCl: (A, B) 0 mM, (C, D) 50 mM, and (E, F) 200 mM. Curves in (B, D, F) were obtained using the 1:1-binding model that yielded the stoichiometry parameter, n.

Binding Enthalpy as a Function of Aliphatic Chain Length

To address the role of aliphatic chain length in the binding process, we conducted a series of experiments with various aliphatic tail lengths of surfactants.

The beginning of the titration of poly(Lys+) and poly(Orn+) with alkyl sulfates exhibited endothermic peaks at room temperature if the alkyl chain of the surfactant was up to 10 carbon atoms (Figure E). Continuation of the titration followed one of two possible scenarios: either the reaction remained endothermic, e.g., C8H17SO4––poly(Lys+), or the enthalpy began to decrease as the titration approached the charge neutralization point and the reaction enthalpy became exothermic, e.g., C10H21SO4––poly(Lys+). This phenomenon of endothermic-to-exothermic reaction enthalpy change (and hence a change in the sign of enthalpy) was absent in the poly(Lys+) and poly(Orn+) systems if the surfactants had 11 or more carbon atoms in their aliphatic chain and also in all of the investigated RSO4––poly(Arg+) systems (Figure C). However, the sign of enthalpy is dependent on temperature due to heat capacity change upon binding as shown later.

Figure 4

Enthalpies of surfactant–poly(amino acid) interactions as a function of the aliphatic chain length at T = 25 °C. Linear alkyl sulfates and alkylamines were used to measure their binding enthalpies to (A) positively and (B) negatively charged poly(amino acid)s, respectively. Panels (C) and (E) show isotherms of RSO4– binding to poly(Arg+) and poly(Lys+), and (D) and (F) show that of RNH3+ binding to poly(Glu–) and poly(Asp–), respectively.

In most cases, the titration of the negatively charged poly(Glu–) and poly(Asp–) with alkylammonium also followed one of the two above-mentioned reaction enthalpy scenarios (Figure D,F). A distinct pair in the negatively charged poly(amino acid)s was C13H27NH3+–poly(Asp–) – its entire titration had exothermic peaks.

In panels A and B of Figure , the measured ΔH values of surfactant–poly(amino acid) interactions are plotted as a function of the total number of carbon atoms in the aliphatic chain, m. Surfactants with longer hydrophobic tail had a more negative contribution to the interaction enthalpy. These dependencies showed a linear behavior with the enthalpic contribution of the CH2 group for various surfactant–poly(amino acid) systems varying from −2.7 kJ mol–1 to −0.92 kJ mol–1 per CH2 group at 25 °C.

The experimental ITC data are summarized and presented in Tables and 2.

Table 1

Changes in Enthalpy and Constant-Pressure Heat Capacity upon Alkyl Sulfate and Alkane Sulfonate Binding to Positively Charged Poly(amino acid)s Measured by ITC

	poly(Arg+)			poly(Lys+)		poly(Orn+)
	T (°C)	ΔH (kJ mol–1)	ΔCp (kJ mol–1 K–1)	T (°C)	ΔH(kJ mol–1)	T (°C)	ΔH (kJ mol–1)
C12H25SO4–	13	–10.2 ± 0.2
	25	–15.4 ± 1.1		25	–6.9 ± 1.0	25	–4.2 ± 0.8
	37	–21.0 ± 0.8	–0.41 ± 0.01
	49	–26.8 ± 0.9
	61	–28.7 ± 0.5
C11H23SO4–	13	–8.8 ± 0.2
	25	–13.6 ± 1.1		25	–4.4 ± 0.6	25	–2.7 ± 0.1
	37	–19.9	–0.43 ± 0.01
	49	–24.1 ± 2.0
	61	–28.5 ± 2.4
C10H21SO4–	13	–7.0 ± 0.4
	25	–12.3 ± 0.6		25	–1.7 ± 0.4	25	–0.2 ± 0.6
	37	–17.0 ± 2.1	–0.40 ± 0.01
	49	–21.6 ± 1.6
	61	–25.2 ± 2.4
C8H17SO4–	13	–2.9 ± 0.1
	25	–5.8 ± 0.5		25	1.6 ± 0.1	25	–0.8 ± 0.2
	37	–7.0 ± 0.9	–0.18 ± 0.04
	49	–9.3 ± 0.4
	61	–11.6 ± 0.1
C10H21SO3–	25	–8.6 ± 0.6		25	2.9 ± 0.3	25	2.2 ± 0.2
C9H19SO3–	25	–4.2 ± 1.0		25	2.4 ± 0.1	25	3.0 ± 0.5
C8H17SO3–	25	–3.0 ± 1.2		25	2.1 ± 0.3	25	2.0 ± 0.2

Table 2

Changes in Enthalpy and Constant-Pressure Heat Capacity upon Alkylammonium Binding to Negatively Charged Poly(amino acid)s Measured by ITC

	poly(Asp–)			poly(Glu–)
	T (°C)	ΔH (kJ mol–1)	ΔCp (kJ mol–1 K–1)	T (°C)	ΔH (kJ mol–1)	ΔCp (kJ mol–1 K–1)
C13H27NH3+	25	–2.2 ± 0.6		25	0.55 ± 0.5
	37	–5.7 ± 0.4	–0.15 ± 0.05	37	–5.1 ± 0.6	–0.12 ± 0.02
	49	–10.3 ± 1.0		49	–9.9 ± 1.3
	60	–16.3 ± 0.8		60	–12.7 ± 1.7
C12H25NH3+	25	–0.5		25	2.0 ± 0.2
	37	–3.7 ± 0.3	–0.20 ± 0.04	37	–2.3 ± 0.2	–0.20 ± 0.12
	49	–8.1 ± 0.9		49	–6.2 ± 1.1
	60	–8.9 ± 1.3		60	–9.7 ± 0.3
C11H23NH3+	25	–2.3 ± 0.2		25	2.4 ± 0.7
	37	–2.1 ± 1.1	–0.26 ± 0.03	37	0.2 ± 0.1	–0.33 ± 0.02
	49	–3.1 ± 0.1		49	–3.0 ± 0.5
	60	–5.9 ± 0.9		60	–4.5 ± 0.1
C10H21NH3+	25	3.3 ± 0.3		25	3.4 ± 0.3
	37	0.9 ± 0.2	–0.39 ± 0.05	37	1.44	–0.43 ± 0.03
	49	–1.4 ± 0.4		49	0.0 ± 0.1
	60	–1.9 ± 0.5		60	–1.3 ± 0.1

Temperature Dependence of Interaction Enthalpy

Charged surfactant binding to a poly(amino acid) of an opposite charge was increasingly more exothermic at higher temperatures. The decrease in interaction enthalpy (increase in its absolute value) was observed in all of the tested systems (Figures –7). The data show that at higher temperatures the above-described positive enthalpies at the beginning of titrations disappeared. For example, in a RNH3+–poly(Glu–) system at 37 °C, the endothermic reaction profiles were observed only for aliphatic chains of 10 and 11 carbon atoms, whereas in a RNH3+–poly(Asp–) system at 37 °C, they were seen only for decylammonium. The reactions were already fully exothermic in all of the tested systems at 49 °C.

Figure 5

Enthalpies of poly(Glu–) interactions with alkylammonium surfactants at different temperatures. Panel (A) shows the effect of temperature on the binding enthalpies. The lines fitted to the data yielded the values of ΔCp. Panel (B) shows the same data plotted as enthalpy versus the length of the surfactant aliphatic chain at different temperatures yielding the methyl group contribution to the interaction enthalpy. Panels (C)–(F) show integrated ITC curves for binding of various alkylammonium surfactants at different temperatures.

Figure 7

Enthalpies of poly(Arg+) interactions with alkyl sulfates at different temperatures. Panel (A) shows the effect of temperature on the binding enthalpies. The lines fitted to the data yielded the values of ΔCp. Panel (B) shows the same data plotted as enthalpy versus the length of the surfactant aliphatic chain at different temperatures yielding the methyl group contribution to the interaction enthalpy. Panels (C)–(F) show integrated ITC curves for binding of various alkylammonium surfactants at different temperatures.

Enthalpies of poly(Asp–) interactions with alkylammonium surfactants at different temperatures. Panel (A) shows the effect of temperature on the binding enthalpies. The lines fitted to the data yielded the values of ΔCp. Panel (B) shows the same data plotted as enthalpy versus the length of the surfactant aliphatic chain at different temperatures yielding the methyl group contribution to the interaction enthalpy. Panels (C)–(F) show integrated ITC curves for binding of various alkylammonium surfactants at different temperatures.

The plots of enthalpy as a function of temperature were used to calculate the constant-pressure reaction heat capacities (ΔCp) by applying linear fits and assuming that the heat capacity is temperature-independent. In the poly(Glu–) and poly(Asp–) systems, the ΔCp values were increasingly negative for the alkylamines of longer aliphatic chains (Table ). Slightly different tendencies were observed in the RSO4––poly(Arg) system. We determined ΔCp = (−0.18 ± 0.04) kJ mol–1 K–1 for C8H17SO4––poly(Arg+). The increased chain length resulted in an increase in the absolute value of ΔCp and, within an error range, it remained similar for other longer chain alkyl sulfate series (Table ).

Sulfonate and Sulfonic Acid Binding to Poly(amino acid)s

The enthalpies of alkyl sulfate binding to poly(Arg+) were more negative than those of sulfonic acid bearing aliphatic chains of the same length. The data show that the interaction enthalpy between poly(Arg+) and alkyl sulfonic acid containing m carbon atoms in the aliphatic chain is approximately equal to the interaction enthalpy between poly(Arg+) and alkyl sulfate containing m – 1 carbon atoms (Figure ). This suggests that the oxygen atom between the sulfur and carbon atoms of the sulfate plays a similar role to a CH2 group in the aliphatic chain.

Figure 8

Comparison between enthalpies of alkyl sulfate (filled symbols, continuous lines) and alkane sulfonate (open symbols, dashed lines) binding to poly(Arg+), poly(Lys+), and poly(Orn+). The enthalpy as a function of chain length plots yielded the CH2 group contribution to the interaction enthalpy. The slopes were quite similar except for the short-chain alkane sulfonate binding, where the enthalpy values were positive and small, thus difficult to determine accurately.

The slope was not confirmed in experiments of sulfonate binding to other cationic poly(amino acid)s, where the enthalpies of binding to either poly(Lys+) or poly(Orn+) seemed to increase with longer alkyl chains, contrary to all of the other systems studied in this paper. It appears that the slopes may still be similar if longer chain surfactants had been tested. In these experiments, enthalpies were close to zero and, therefore, were difficult to determine accurately by ITC.

Heat Capacity of Surfactant Binding to Poly(amino acid)s

The temperature dependencies of the enthalpy changes upon binding allowed estimating the constant-pressure heat capacities (ΔCp) of such binding reactions.

Figure shows the heat capacities obtained by applying linear fits to the enthalpy dependencies on temperature under the assumption that the heat capacity is independent of temperature and thus the slope remains constant within the studied 13–61 °C temperature range. All measured heat capacities were negative in sign and increased in absolute value upon increasing the surfactant chain length. With a possible exception of poly(Arg+), the dependencies of the heat capacity on chain length were linear, yielding the ΔCp of the CH2 group equal to (−0.092 ± 0.018) kJ mol–1 K–1 (Figure ). In our opinion, the model is consistent with a case when the slopes are identical, but additional data may be necessary to confirm this conclusion.

Figure 9

ΔCp values of varied-length alkyl sulfate surfactant binding to poly(Arg+) and alkylammonium surfactant binding to poly(Asp–) and poly(Glu–). The dependencies on alkyl chain length were approximated as linear yielding the values of the ΔCp of the CH2 group equal to −0.061 kJ mol–1 K–1 for poly(Arg+), −0.075 kJ mol–1 K–1 for poly(Asp–), and −0.10 kJ mol–1 K–1 for poly(Glu–). The average was (−0.092 ± 0.018) kJ mol–1 K–1.

Figure shows an alternative way of obtaining the heat capacity by plotting the temperature dependencies for the enthalpies of CH2 group contribution to the binding to poly(amino acid). The obtained average value of the ΔCp of the CH2 group was similar to the average value from Figure and was equal to (−0.064 ± 0.006) kJ mol–1 K–1.

Figure 10

Enthalpy contributions of the CH2 group for alkyl sulfate surfactant binding to poly(Arg+) and alkylammonium binding to poly(Asp–) and poly(Glu–), at different temperatures. The values were obtained from the slopes of linear fits shown in panels (B) of Figures –7. The slopes yielded approximate ΔCp values for the CH2 group equal to −0.056 kJ mol–1 K–1 for poly(Arg+), −0.062 kJ mol–1 K–1 for poly(Asp–), and −0.071 kJ mol–1 K–1 for poly(Glu–). The average was (−0.064 ± 0.006) kJ mol–1 K–1.

In this paper, we described ITC data obtained for two systems of oppositely charged molecules: (1) cationic poly(amino acid)s–anionic surfactants and (2) anionic poly(amino acid)s–cationic surfactants. The ITC data showed that surfactants bind stoichiometrically to the oppositely charged moieties of poly(amino acid)s with a charge ratio of 1:1. The small discrepancies from this stoichiometry ratio can be attributed to the deviations in the concentration of both the poly(amino acid) and surfactant solutions. Previous elemental analysis of the poly(amino acid) showed that up to 20% of water could be present in these batches,[37] which could also affect the binding stoichiometry.

The majority of surfactant–poly(amino acid) isotherms had a dip (decrease) in enthalpy at the charge neutralization point. This additional enthalpy might be a result of aggregate formation or structural transitions of the polymer. McCord et al.[38] have determined that SDS induces the formation of an ordered polyarginine with an α-helix structure at the charge neutralization point. The values for enthalpy of an α-helix formation were in the range of −2 to −0.9 kcal mol–1 per residue,[39−41] which is in agreement with our results.

The CH2 group contributed almost two times less enthalpy to the systems of poly(Orn+) and poly(Glu–) if compared to all of the other surfactant–poly(amino acid) systems at 25 °C. According to the literature,[17,18,21,38] poly(Orn+), poly(Glu–), and poly(Arg+) undergo a coil-to-helix transition after the neutralization point is reached. Poly(Lys+) undergoes a coil-to-β-sheet transition with all of the tested surfactants except octyl sulfate, whereas poly(Asp–) does not undergo a transition to the more structured state. There seems to be no clear correlation between the CH2 group contributions to the enthalpy and the transitions in the secondary structure. Some of these transitions may also be too slow to be observed in an ITC experiment.[42]

We also considered the counterion condensation effect in these systems. As discussed previously,[37] the charge density parameter in the poly(amino acid) systems was less than a critical value, which marks the initial point of counterion condensation. Thus, the counterion condensation effect was negligible and did not interfere with the binding in the investigated surfactant–poly(amino acid) systems.

The negative ΔCp values in this study were similar to those of hydrocarbon transfer from water to an organic solvent,[43] suggesting that the hydrophobic tail interaction between bound surfactants was a plausible explanation for the observed enthalpies. A linear increase in the absolute value of the binding enthalpy with the chain length of a surfactant means that longer chain surfactants are more prone to binding. This increase in enthalpy was present in all surfactant–poly(amino acid) pairs, even those with endothermic interaction profiles. However, the CH2 group contributions to the interaction enthalpy were similar in various systems. The highest increased-tail surfactant contribution was in the RSO4––poly(Arg+) system. The strong affinity between sulfate and guanidinium groups[36] is also accompanied by a more strict geometry of the salt bridge, which seems to be also related to the higher surfactant’s tail contributions to the binding enthalpy.

Our previous studies have shown that the enthalpy of aliphatic chain interaction between themselves can indicate whether the chains aggregate to a liquid phase or a solid phase. The difference in the enthalpies of binding can match the enthalpy of fusion.[44−47] The enthalpy of the CH2 group binding into a liquid form yielded the value of −1.25 kJ mol–1, whereas upon binding into the solid form, the absolute value was greater and was equal to −5.2 kJ mol–1 per CH2 group.[46] Here in this study, the value was in the range of −2.7 to −0.92 kJ mol–1 per CH2 group, and the absolute magnitude indicates that the phase of the poly(amino acid)–surfactant complex may be liquid or solid, but only part of the contacts have formed between the aliphatic chains in the solid because poly(amino acid)s are quite bulky and would not enable a conformation where surfactant aliphatic chains fully bind to each other. Therefore, it is most likely that the chains only partially interact with each other and the remaining length of the chains interacts with the amino acids of the polymer. Furthermore, all of these interactions occur in aqueous solution, and thus are highly dependent on hydration-related effects. Charged ionic groups are strongly hydrated, whereas the aliphatic chains are poorly hydrated resulting in opposite effects upon binding.

Conclusions

The stoichiometry of charged surfactants binding to oppositely charged linear poly(amino acid)s is one molecule of surfactant per amino acid moiety of the polymer. The high ionic strength of the solutions diminishes the electrostatic interactions between the surfactants and the poly(amino acid)s. The increased length of surfactant’s aliphatic chain results in a more favorable interaction enthalpy. The observed enthalpy gain per CH2 group in longer aliphatic chains is similar in various surfactant–poly(amino acid) systems.

Experimental Section

Chemicals

Cationic surfactants used in this study—octylamine, nonylamine, decylamine, undecylamine, dodecylamine, and tridecylamine—were purchased from Sigma-Aldrich and Acros Organics, either as hydrochloride salts or as amines. The following anionic surfactants were purchased from Sigma-Aldrich and Acros Organics as sodium salts: 1-decanesulfonic acid, 1-nonanesulfonic acid, 1-octanesulfonic acid, dodecyl sulfate, undecyl sulfate, decyl sulfate, and octyl sulfate. Poly(l-glutamic acid) sodium salt, poly(l-aspartic acid) sodium salt, poly(l-arginine) hydrochloride, poly(l-lysine) hydrochloride, and poly(-ornithine) hydrochloride were purchased from Sigma-Aldrich and Alamanda Polymers. Polyamino acids were of variable length, ranging from 50 to 700 amino acids per polymer molecule. Solutions were prepared by dissolving substances in distilled Milli-Q water. Solutions from pure amines were prepared by dissolving them in Milli-Q water and titrating with HCl until the pH value of 5 was reached and amines were visibly dissolved. Polyamino acid stock solutions were prepared as 50 mM, their concentration expressed per amino acid. Concentrations of the surfactant stock solutions ranged from 5 to 20 mM. Polyamino acid solutions were stored at 4 °C, whereas surfactants and salts at 20 °C. The ITC experiments were performed a large number of times (e.g., poly(Arg+)with SDS over 10 times) using the surfactants of different producers to avoid their purity-related effects.

ITC Experiments

Concentrations of poly(amino acid)s were determined by weighing and were expressed per amino acid monomer in the ITC experiments. The purity of the poly(amino acid)s and their elemental analysis were described previously.[37] The majority of experiments were performed with a Nano ITC calorimeter (TA Instruments) with cell and syringe volumes of 1.00 mL and 250 μL, respectively. Some experiments were repeated using a Microcal (Northampton, MA) Micro-Calorimetry System calorimeter with cell and syringe volumes of 1.4 mL and 250 μL, respectively. Calorimeter measurements were validated by performing the measurement of electrical pulses of known applied power and also by titrating Tris base with nitric acid.[48]

The calorimeter cell was usually loaded with a (0.5–2) mM solution of the poly(amino acid), expressed per amino acid, and the syringe was loaded with a (3.75–20) mM solution of the oppositely charged surfactant. The majority of experiments were performed with a 1 mM polymer solution (per amino acid) in the cell and a 10 mM surfactant solution in the syringe. This setup is assumed as a default in this article unless stated otherwise. The distribution of poly(amino acid) size varied from 50 to 700 amino acids per polymer molecule.

The usual titration at a constant temperature was carried out in 25 injections of 10 μL at time intervals of at least 4 min and with a syringe stirring speed of 200 rpm. Data for at least 3 min were collected prior to the first injection to ensure the stability of the baseline. Before an experiment, the calorimeter cell was washed with Milli-Q water and pre-rinsed with the solution for the cell. Experiments were performed in the temperature range from 13 °C to 61 °C.

Calculation of Reaction Enthalpies

One binding site model, which is commonly used in the ITC data analysis, was not suitable to accurately analyze most of the experimental data. Therefore, we calculated enthalpies of surfactant binding to poly(amino acid)s by integrating areas under the isotherm curve. Isotherms were obtained using NITPIC software,[49] which generated a noise-filtered baseline and integrated the experimental peaks according to the new baseline. Reaction enthalpies were determined by calculating the area under the isotherm using a numerical trapz function from the SciPy package for Python 3.[50] The dilution heats of the titrants were essentially negligible as compared to the poly(amino acid)–surfactant reaction heats, and they were subtracted from the isotherms prior to integration.