Binding of norharmane with RNA reveals two thermodynamically different binding modes with opposing heat capacity changes

Journal of Colloid and Interface Science 538 (2019) 587–596

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis

Regular Article

Binding of norharmane with RNA reveals two thermodynamically different binding modes with opposing heat capacity changes Bijan K. Paul a,⇑, Narayani Ghosh b, Saptarshi Mukherjee c,⇑ a

Department of Chemistry, Mahadevananda Mahavidyalaya, Barrackpore, Kolkata 700120, India Basic Science and Humanities Department, University of Engineering and Management, University Area, Newtown, Kolkata 700156, India c Department of Chemistry, Indian Institute of Science Education and Research Bhopal, Bhopal Bypass Road, Bhauri, Bhopal 426066, Madhya Pradesh, India b

g r a p h i c a l a b s t r a c t

a r t i c l e

i n f o

Article history: Received 3 September 2018 Revised 26 November 2018 Accepted 3 December 2018 Available online 4 December 2018 Keywords: Anti-cancer photosensitizer Norharmane RNA Thermodynamically different binding modes Opposing heat capacity changes Spectroscopy versus calorimetry

a b s t r a c t The binding interaction of a prospective anti-cancer photosensitizer, norharmane (NHM, 9H-pyrido[3,4b]indole) with double stranded RNA reveals a primarily intercalative mode of binding. Steady-state and time-resolved fluorescence spectroscopic results demonstrate the occurrence of drug-RNA binding interaction as manifested through environment-sensitive prototropic equilibrium of NHM. However, the key finding of the present study lies in unraveling the complexities in the NHM-RNA binding thermodynamics. Isothermal Titration Calorimetry (ITC) results reveal the presence of two thermodynamically different binding modes for NHM. An extensive temperature-dependence investigation shows that the formation of Complex I is enthalpically (DHI < 0) as well as entropically (TDSI > 0) favored with the enthalpic (entropic) contribution being increasingly predominant in the higher (lower) temperature regime. On the contrary, the formation of Complex II reveals a predominantly enthalpy-driven signature (DHI < 0) along with unfavorable entropy change (TDSI < 0) with gradually decreasing enthalpic contribution with temperature. Such differential dependences of DHI and DHII on temperature subsequently lead to opposing heat capacity changes underlying the formation of Complex I and II (DC Ip < 0 and DC IIp > 0). A negative DCp underpins the pivotal role of ‘hydrophobic effect’ (release of ordered water molecules) for the formation of Complex I, while a positive DCp marks the thermodynamic hallmark for ‘hydrophobic hydration’ (solvation of hydrophobic (or nonpolar) molecular surfaces in aqueous medium) for formation of Complex II. A detailed investigation of the effect of ionic strength enables a component analysis of the total free energy change (DG). Ó 2018 Elsevier Inc. All rights reserved.

⇑ Corresponding authors. E-mail addresses: [email protected] (B.K. Paul), [email protected] (S. Mukherjee). https://doi.org/10.1016/j.jcis.2018.12.011 0021-9797/Ó 2018 Elsevier Inc. All rights reserved.

588

B.K. Paul et al. / Journal of Colloid and Interface Science 538 (2019) 587–596

1. Introduction In addition to deoxyribonucleic acid (DNA), ribonucleic acid (RNA) has also played critically important roles underlying the evolution of life on this planet. In particular, what is important is the unique intermediary position of RNA between DNA controlling the stable storage of genetic information, and proteins in which the functional form of this information is expressed [1]. However, despite the plethora of important functionalities, RNA was long contemplated as the static carrier of genetic information only. Of late, the instrumental roles of RNAs in governing various cellular events/processes have revealed the multitude of functionalities of RNAs [1–5]. Recent research endeavors including microRNAs, small nucleolar and small nuclear RNAs, small interfering RNAs have unveiled their critical ability of controlling transcription and translation processes. Naturally, with such vital control of the genome, research in the field promises to push new frontiers of therapeutic intervention with RNAs as the potential target [1,2,5–9]. In this context, the role of double stranded RNA (dsRNA) has captured particular attention given their ability of sequence-specific degradation of mRNA, a process popularly designated as RNA interference [10,11]. Furthermore, specific degradation of RNAs in many animals, plants, and yeasts has been argued to proceed through similar mechanistic pathways involving dsRNA. It is also known that some viruses, including the lethal pathogenic viruses like HIV or hepatitis C, use RNA as the primary genetic material [1,8,9,12,13]; even for the organisms employing DNA the conversion of the genetic information into RNA is essential for it to become functional (or accessible) [1,8,9,12,13]. Thus, the contemporary global research has naturally witnessed a burgeoning thrust in RNA as a potential cellular target for drugs [6–9]. However, the important issue of recognition of RNA by small drug molecules still remains far from being meticulously explored, particularly in comparison to the extensive research efforts devoted almost exclusively to DNA binding studies [14–19]. A systematic evaluation to this end might also carry the potential to provide the rationale toward development of RNA-targeted anticancer therapeutics [3,11–13]. Various intra- and inter-molecular RNA interactions during gene expression are believed to be responsible for the production of dsRNA within cells [1]. Typically, intramolecular dsRNA elements are found within the local hairpin structures and may also emanate from binding of antisense RNAs to specific targets [5,20]. In many viruses, dsRNA is produced at some stage during the course of their replication. These perspectives also augment and necessitate the importance and relevance of recognition of dsRNA by small drug molecules for a host of cellular processes and thereby shedding light to the concerned mechanistic pathway. However, the binding affinity, mode, and the accompanying thermodynamics of interaction of small molecules with biopolymers are governed by a vista of structural and energetic factors [1,2,5,14–19]. The present work demonstrates a spectroscopic and calorimetric investigation on the interaction of norharmane (NHM, Scheme 1) with dsRNA extracted from torula yeast.

Scheme 1. Simplified iIllustration of the cation  neutral prototropic equilibrium of NHM.

b-Carbolines have long been central to biological and biophysical research activities owing to their enormous medicinal properties including their interactions with neuromodulators and neurotransmitters of the Central Nervous System, functionality as monoamine-oxidase enzyme inhibitors and so forth [21]. More importantly, the ability of b-carbolines to produce singlet oxygen under UV radiation strives for potential medicinal prospects in photodynamic therapy toward development of targeted degradation of malignant and cancerous tissues [21]. Earlier we have reported the strength, mode, base-pair specificity, dynamics, kinetics, and thermodynamics of interaction of NHM and related drugs belonging to the b-carboline family with dsDNA [14,15]. The major focus of the study rests on delineating the complex thermodynamic landscape of the interaction profile which reveals two different thermodynamic binding modes of NHM to dsRNA. Such complex thermodynamic profile of the binding interaction clearly implies the presence of two distinct dug: RNA complexes [19]. This is further corroborated from spectroscopic results which categorically establish a principally intercalative mode of binding of NHM to dsRNA along with a significant contribution from electrostatic forces. Cumulatively, a complex interplay of more than one type of binding forces is found to govern the overall interaction. Additionally, our results also demonstrate that conformational modification of dsRNA following binding with the drug, which could provide a possible interpretation of the complex binding thermodynamics [19,22,23], is not operative in the present case. 2. Experimental All the materials, namely, NHM (Scheme 1), RNA, phosphate buffer, and sodium chloride (NaCl) were used as received from Sigma Chemical Co., USA. The solution of 10 mM phosphate buffer (pH 7.40) was prepared in deionized triply distilled Milli pore water. All spectroscopic measurements were carried out with a low concentration (ca. 2.0 mM) of NHM to ensure minimization of inner-filter and reabsorption effects. The absorption and fluorescence spectra were acquired on Cary 500 UV–vis spectrophotometer and Fluorolog 3–111 fluorometer, respectively. The timeresolved fluorescence decays were obtained by the method of Time-Correlated Single Photon Counting (TCSPC) [14,15,24,25]. The details of the experimental setup are described elsewhere [14,15,24,25]. Circular dichroism (CD) spectra were obtained on a JASCO J-815 spectropolarimeter. The ITC measurements were performed on a Nano ITC, TA Instrument in phosphate buffer (pH 7.40). An elaborate description of the experimental methods and protocols is given in the Supporting information. 3. Results and discussion 3.1. NHM-RNA binding interaction: spectroscopic deciphering Fig. 1a shows the absorption profile of NHM in aqueous buffer which is characterized by two distinct bands at kabs
348 nm and
370 nm, attributable to the neutral and cationic species of NHM (Scheme 1), respectively [14,15,24,26–29]. Incremental addition of RNA to the solution of NHM in aqueous buffer is found to result in a gradual decrease of absorbance of the neutral species of NHM (kabs
348 nm) with concomitant increase of absorbance of the cationic counterpart (kabs
370 nm). This observation can be argued to reflect a progressive stabilization of the cationic form of the drug following interaction with RNA containing negatively charged polyphosphate backbone [14,15]. However, it is imperative to note that the modulation of the absorption profile of NHM upon interaction with RNA does not proceed through an isosbestic

B.K. Paul et al. / Journal of Colloid and Interface Science 538 (2019) 587–596

589

Fig. 1. (a) Absorption profile of NHM with added RNA in aqueous buffer (pH 7.40). Curves (i) ? (ix) denote [RNA] = 0, 0.08, 0.16, 0.24, 0.32, 0.56, 0.81, 1.13, 1.29 mM. (b) Fluorescence profile (kex = 348 nm) of NHM with added NHM in aqueous buffer (pH 7.40). Curves (i) ? (xxi) denote [RNA] = 0, 0.08, 0.16, 0.24, 0.32, 0.40, 0.48, 0.56, 0.64, 0.72, 0.81, 0.89, 0.97, 1.13, 1.29, 1.45, 1.61, 1.93, 2.25, 2.42, 2.89 mM. (c) Variation of fluorescence anisotropy of NHM (kex = 348 nm, kem = 450 nm) with incremental RNA concentration. Each data point represents an average of 12 measurements. The dotted line is only a visual guide. (d) Fluorescence decay transients of NHM (kex = 340 nm, kem = 450 nm) with added RNA in aqueous buffer (pH 7.40). Curves (i) ? (xi) denote [RNA] = 0, 0.08, 0.16, 0.32, 0.56, 0.81, 1.29, 1.93, 2.25, 2.42, 2.89 mM. The symbols represent the raw data and the solid black lines are the best fit lines. The sharp black profile on the extreme left designates the Instrument Response Function (IRF).

point (Fig. 1a) even though the prototropic equilibrium of NHM is limited within the neutral and cationic species. This bears reasonable similarity to our previous reports on the modulation of absorption profile of NHM and other b-carboline drugs upon interaction with DNA [14,15], and is argued to reflect the operation of more than one type of interaction forces underlying the binding process. In this context it is relevant to state that a plot of the absorption spectra highlighting the region of wavelength for the prototropic transformation of NHM between the neutral and cationic species (Fig. S1a of the Supplementary Material) clearly shows the lack of an isosbestic point. It has been further confirmed by a plot of absorbance values surrounding the concerned wavelength regime as depicted in Fig. S1b of the Supplementary Material. The variation of the excitation profile of NHM following interaction with RNA (Fig. S1c of the Supplementary Material) also supports the observation of preferential stabilization of the cationic species of NHM over the neutral form in the presence of RNA. Fig. 1b presents the variation of the steady-state fluorescence profile of NHM following interaction with RNA. In aqueous buffer NHM shows a broad unstructured band at kem
450 nm, characteristic of the cationic species [14,15,24,26–29]. Incremental addition of RNA to the solution of NHM in aqueous buffer is found to accompany prominent quenching of the cationic fluorescence band (kem
450 nm) of NHM with no discernible shift of fluorescence wavelength, Fig. 1b. In analogy to literature reports [14,15,24,26– 29], a reduced polarity at the site of interaction of NHM within the RNA scaffolds may be invoked to account for the observed quenching of fluorescence of NHM (Fig. 1b). In tune with the absorption spectral results it appears reasonable to think that a progressive decrease of absorbance of the neutral species of NHM (kabs
348 nm) along with increase of absorbance of the cationic

form (kabs
370 nm) with added RNA may reflect a preferential interaction of the cationic species of the drug with RNA (and not of the neutral form). This is turn is in parity with the observation of a prominent fluorescence band corresponding to the cationic species of NHM only, while a characteristic fluorescence band (at
380 nm) for the neutral species of NHM is not observed. The variation of steady-state fluorescence anisotropy of NHM following interaction with RNA further substantiates the occurrence of the binding interaction because the fluorescence anisotropy of the drug is significantly enhanced within the RNA-bound state, Fig. 1c. A sharp rise of anisotropy of NHM with added RNA demonstrates the impartation of considerable motional constraints on the drug molecules upon binding to RNA [14,15,24,25,28,29]. The interaction of NHM with RNA is further corroborated from time-resolved fluorescence decay studies. Fig. 1d displays the modulation of fluorescence decay transients of NHM following interaction with RNA, the relevant decay parameters being assembled in Table 1 (the residuals of the fitted functions to the actual data plotted in Fig. S2 of Supplementary Material). In bulk aqueous buffer, NHM exhibits an exponential decay having a characteristic lifetime of the cationic species (s1 = 21.56 ns) [14,15,24,26–29]. The fluorescence decay behavior of RNA-bound NHM, however, requires a biexponential function for an adequate description comprising of a significantly slower component (s1 in Table 1) revealing the characteristic lifetime of the cationic species along with a relatively faster component (s2 in Table 1). The occurrence of the NHM-RNA interaction is categorically established from the progressive reduction of the population (relative amplitude, a1) of the cationic species of NHM with increasing concentration of RNA (Table 1). This bears striking resemblance to the steady-state spectroscopic results stated earlier in the sense that binding interaction with

590

B.K. Paul et al. / Journal of Colloid and Interface Science 538 (2019) 587–596

Table 1 Fluorescence decay parameters of NHM with incremental addition of RNA.

a

[RNA] (mM)

s1a (ns)

s2a (ns)

a1

a2

v2

0 0.08 0.16 0.32 0.56 0.81 1.29 1.93 2.25 2.42 2.89

21.56 21.15 21.11 20.74 20.58 20.23 19.42 18.90 18.41 18.03 16.91

– 1.96 1.93 1.93 1.64 1.54 1.53 1.52 1.45 1.44 1.40

1.00 0.97 0.97 0.96 0.95 0.94 0.92 0.88 0.82 0.78 0.68

– 0.03 0.03 0.04 0.05 0.06 0.08 0.12 0.18 0.22 0.32

1.01 1.01 1.03 1.04 1.06 1.08 1.09 1.1 1.08 1.09 1.08

±4%.

RNA accompanies prominent quenching of the cationic fluorescence of the drug. It could be pertinent to state in this context that a direct comparison of the characteristic fluorescence lifetime values of NHM in solvents of varying polarity reveals that in bulk aqueous medium the steady-state fluorescence profile of NHM is characterized by a broad profile having fluorescence maxima at
450 nm, typical of the cationic species of NHM, and the characteristic lifetime of the cationic species of NHM in the medium is found to be 21.56 ns, whereas in a less polar solvent, namely, 1,4-dioxane the steady-state fluorescence profile of NHM is characterized by a broad profile having fluorescence maxima at
380 nm designating the neutral species of NHM, and the characteristic lifetime of the neutral species of NHM in the medium is found to be 2.28 ns (Fig. S3 of the Supplementary Material) [15,24,27,28]. This direct comparison clearly shows that the fluorescence lifetime of the probe tends to decrease as it is exposed to an environment of less polarity, which is commensurate with the observation of a progressive decrease of fluorescence decay time constants with increasing RNA concentration. The reduced polarity of the medium with progressive addition of RNA may result in a gradual decrease of the fluorescence lifetime of NHM, however, a precise mechanistic interpretation in this context is difficult with a view to the complex microheterogeneous environment in the presence of RNA. Also, a precise mechanistic model underlying the origin of the relatively faster decay component (s2 in Table 1) following interaction with RNA is not tenably rationalized till now. Given the complex micro-heterogeneous environment within the RNA scaffolds coupled with the asymmetric charge distribution over the molecular framework of NHM, it is likely that the drug molecules might undergo complex and unequal hydration within the RNA-bound state; also this observation is in good agreement with the fluorescence decay behavior of NHM in a variety of micro-heterogeneous assemblies/aggregates [14,15,24,28–30].

Fig. 2. Effect of increasing ionic strength on the fluorescence profile of RNA-bound NHM (kex = 348 nm). Curves (i) ? (xi) denote [NaCl] = 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 mM.

fluorescence intensity of RNA-bound drug with incremental addition of NaCl;
57% increase of fluorescence intensity due to addition of 100 mM NaCl. This conforms to a substantial weakening of the NHM-RNA binding strength with increased ionic strength of the medium and hence substantiates the key role of electrostatic interaction underlying the binding mechanism. This is further corroborated from thermodynamic results based on calorimetric measurements as discussed below. 3.3. Mode of binding: circular dichroism (CD) spectroscopy The modulation of the CD spectral profile of RNA with added NHM yields persuasive evidence for intercalative mode of binding of the drug to the RNA duplex structure, Fig. 3 (path-

3.2. Effect of ionic strength Herein, we have been motivated to probe into the effect of ionic strength on the NHM-RNA binding interaction with an eye to the presence of counter charges on the binding partners, namely, negative polyphosphate backbone of dsRNA and cationic drug. To this end, the modulation of fluorescence profile of RNA-bound NHM is explored in the presence of a strong electrolyte NaCl [14,15]. An enhanced ionic strength of the medium with added NaCl may lead to weakening of the electrostatic interactions between NHM (cationic drug) and RNA. It is usually argued that increased ionic strength of the medium may result in reduction of the repulsive electrostatic forces between the sequential phosphate groups on the RNA duplex whereby leading to lowering of the unwinding tendency of the RNA duplex and thus assisting the helix to shrink [1,14,15,31]. Fig. 2 shows a significant enhancement of the cationic

Fig. 3. Effect of addition of the drug (NHM) on the intrinsic circular dichroic profile of RNA (70 mM). Curves (i) ? (ix) denote [NHM] = 0, 2, 4, 8, 12, 15, 20, 24, 30 mM.

591

B.K. Paul et al. / Journal of Colloid and Interface Science 538 (2019) 587–596

length = 0.1 cm). The far-UV CD profile of the RNA duplex structure is comprised of two peaks at
272 nm (positive peak) and
242 nm (negative peak), characteristic of the native conformation of dsRNA [32]. Typically, the stacking interactions between the base pairs and the helical structure of the polynucleotide backbone of dsRNA give rise to the intrinsic CD signal of dsRNA. Thus, the aforesaid modulation in the intrinsic CD signal of RNA in the presence of the drug (Fig. 3) confirms the perturbation of the stacking orientation between the base pairs in the dsRNA conformation due to intercalation of NHM as required during optimization of the binding interaction following accommodation of the intercalated drug [14,15,17,31–34]. On the contrary, only groove or electrostatic binding of small drug molecules to RNA duplex structure would have led to no significant change of the intrinsic CD spectrum of RNA [14,15,17,31–34]. It is pertinent to note in this context that the intrinsic CD profile of RNA exhibits no apparent shift of the peak wavelengths with added NHM (Fig. 3) which suggests no significant modification of the native conformation of RNA following binding with the drug [14,15,17,31–34]. Further evidence for the intercalative mode of binding of NHM to RNA has been obtained from the induced CD spectra of NHM in the presence of RNA. The drug molecule (NHM) is not chiral (and hence produces no CD signal), however, with added RNA induced CD signal is found to develop in the wavelength region of absorption of NHM (at
362 nm which is close to the characteristic absorption band feature of the cationic species of NHM), Fig. S4 of the Supplementary Material. As there is no prominent absorption of RNA in this wavelength regime, the induced CD spectra may be argued to originate exclusively from the interaction (intercalation) of the drug molecules with RNA (path-length = 0.1 cm) [35].

3.4. Thermodynamics of binding: isothermal titration calorimetry Fig. 4 displays the representative ITC profiles of NHM-RNA binding interaction with the relevant thermodynamic parameters being compiled in Table 2. ITC provides a direct experimental technique for an accurate evaluation of the thermodynamic landscape and binding parameters of an interaction process at a given temperature [36–39]. The ITC enthalpogram (Fig. 4b) clearly illustrates the unique nature of the thermodynamics of the concerned interaction which conforms to two different thermodynamic binding modes of NHM with RNA. The data collected in Table 2 show that the affinity constant (KI) for formation of Complex I is significantly higher in magnitude compared to that for formation of Complex II (KII). The complicated thermodynamic landscape of NHM-RNA binding show that the affinity constant values (KI and KII) for both

Complex I and Complex II are progressively reduced with increasing temperature (Table 2). The enthalpy change for formation of Complex I (DHI) is found to be slightly positive at a low temperature (T = 293 K) and becomes increasingly negative (indicating the exothermicity of the binding process, DHI < 0) with rise of temperature, Table 2. The process is, however, found to be characterized by a favorable entropic contribution (TDSI > 0) with the absolute magnitude of the entropic component (TDSI) being gradually lowered with increasing temperature, Table 2. Cumulatively, these results indicate that formation of Complex I is an entropy-dominated process in the lower temperature regime while rising temperature leads to predominant enthalpic contribution to the overall free energy change (DGI) of the process [15,37–42]. On the contrary, the formation of Complex II is found to be characterized by an unfavorable entropic contribution (TDSII < 0) coupled with a favorable enthalpic contribution (DHII < 0, exothermic process). The absolute magnitudes of both DHII and TDSII are found to decrease with rise of temperature, though they still remain negative (Table 2). Thus, the formation of Complex II points to an overall enthalpy-dominated process throughout the domain of temperatures being investigated. A negative free energy change of interaction (DGI < 0 and DGII < 0) in both the cases manifests the thermodynamic feasibility of the NHM-RNA binding process [15,37–43]. Evaluation of DCp. Herein, the variation of enthalpy change (both DHI and DHII) of NHM-RNA binding process with temperature has been processed to evaluate the constant pressure heat capacity change (DCp) according to the standard thermodynamic relationship:

DC p ¼ dðDHÞ=dT

ð1Þ

The variations of DHI and DHII as a function of temperature are plotted in Fig. 5. The sign and magnitude of DCp can yield critical insights into the molecular level interpretation of the nature of binding forces underlying the interaction. The relationship between DCp and change in accessible surface area due to complex formation may be crucial to the understanding of modulation of structural properties following complexation in terms of experimentally determined thermodynamic quantities 1

[15,38–43]. A negative value of DCp (DC Ip ¼ 1:12 kJ mol K1 , Table 2) reflects a typical thermodynamic signature corresponding to the release of ordered water molecules (often described as the water of hydrophobic hydration) and/or counterions from the hydrophobic surfaces on RNA or/and the drug (NHM) upon complex formation [15,38–43]. This in turn provides a reasonable

Fig. 4. (a) ITC profile for NHM-RNA interaction depicting the raw heat burst curves (appropriately corrected for heat of dilution) at 303 K in 10 mM phosphate buffer (pH 7.40). The power sign convention of the instrument designates an exothermic (endothermic) process by upward (downward) heat bursts. (b) The ITC enthalpograms for NHM-RNA interaction at various temperatures as specified in the figure legend (293 K: s, 298 K: j, 303 K: ▲, and 308 K: d). The solid lines represent the best fit lines according to multiple sites binding model.

592

B.K. Paul et al. / Journal of Colloid and Interface Science 538 (2019) 587–596

Table 2 Summary of ITC parameters for NHM-RNA interaction.a

a

T (K)

KI (104)

nI

DH I

TDS I

DG I

DC Ip

293 298 303 308

61.40 ± 7.4 32.82 ± 3.9 4.24 ± 0.51 1.52 ± 0.18

8.56 ± 0.43 7.02 ± 0.35 6.34 ± 0.32 5.67 ± 0.28

1.29 ± 0.24 4.74 ± 0.85 8.20 ± 1.48 16.30 ± 2.93

33.69 ± 1.7 26.29 ± 1.3 18.63 ± 0.93 8.37 ± 0.42

32.41 31.04 26.83 24.67

1.12

T (K)

KII (103)

nII

DHII

TDSII

DGII

DC II p

293 298 303 308

8.73 ± 0.44 5.61 ± 0.28 2.49 ± 0.12 1.75 ± 0.08

5.74 ± 0.28 5.88 ± 0.29 6.71 ± 0.33 7.54 ± 0.38

143.8 ± 10 133.9 ± 10 124.0 ± 8 108.3 ± 8

121.59 ± 6.1 113.09 ± 5.6 104.23 ± 5.2 89.14 ± 4.4

22.21 20.81 19.77 19.16

2.33

K values in M1, DH, TDS and DG values in kJ mol1 and DCp values in kJ mol1 K1.

Fig. 5. Plot of variation of DH (in kJ mol1) as a function of temperature for the processes of formation of Complex I (DHI: s) and Complex II (DHII: d). The error bars are within the symbols if not apparent.

rationale to account for the favorable entropic contribution to the binding interaction (TDSI > 0, Table 2) [15,38–43]. The release of structured water emanates from transfer of nonpolar surfaces into the hydrophobic core of the biomolecule accompanying a decrease in the heat capacity following complex formation and thus leading to a negative DCp [15,38–43]. This result conclusively indicates that the process of formation of Complex I in NHM-RNA binding comprises several hydration contributions which justifies a pivotal role of hydrophobic interaction force underlying the binding process (that is, formation of Complex I) [15,38–43]. On the other hand, it is intriguing to note that the process of formation of Complex II is characterized by a positive change of heat 1

capacity, that is, DC IIp ¼ 2:33 kJ mol K1 (Table 2), which is, indeed sporadically reported in the literature. A positive heat capacity change (DC IIp > 0) can be correlated with the notion of burial of polar interfaces following the formation of Complex II [40,41,44–48]. Usually, the genesis of hydrophobic interaction is connected to the large drop of Gibbs free energy accompanying exposure of hydrophobic (or nonpolar) surfaces to aqueous phase [40,41,44–48]. Thus, a positive change of heat capacity (DC IIp > 0) is believed to reflect the thermodynamic hallmark for ‘hydrophobic hydration’, that is, solvation of hydrophobic (or nonpolar) molecular surfaces in aqueous medium [40,41,44–48], which is in accord with the unfavorable entropic contribution accompanying the process (TDSII < 0, Table 2). In summary, it can be concluded that shielding of hydrophobic (or nonpolar) molecular surfaces from aqueous phase due to complexation furnishes the key negative contribution to heat capacity change (formation of Complex I, DC Ip < 0), while in contrast, removal of polar molecular surfaces from aqueous phase leads to positive contribution to DCp (formation of Complex II, DC IIp > 0) [40,41,44–48].

Cumulatively, it can be argued that in addition to the apparent electrostatic interaction force underlying the NHM-RNA interaction, the hydrophobic hydration accompanying a substantial modulation of the hydration structure of both NHM and RNA also plays a governing role. Thus, it appears logical to state that the overall interaction process is governed by a complicated interplay of multiple interaction forces. The origin of the ‘‘hydrophobic effect” of hydrophobic solutes in aqueous medium is primarily related to the modulation in clustering of the adjacent water molecules rather than direct water-solute interactions. The ‘classical hydrophobic effect’ is conventionally interpreted based on the concept of organization of water molecules surrounding nonpolar moieties [15,38–43,49,50]. The organized (relatively more ordered) water molecules being typically characterized by a low entropy and high heat capacity in comparison to bulk water, the hydrophobic effect is classically rationalized with regard to positive entropy change (TDS > 0) coupled with a decrease of heat capacity (DCp < 0) [15,38–43,49–51]. Incorporation of hydrophobic substances in aqueous medium usually accompanies a tendency of the hydrophobic molecules to aggregate leading to minimization of the surface of contact with water and hence the associated surface energy. Small hydrophobic molecules, however, dissolve in water (though the solubility is low enough as indicated by increase of the chemical potential of the solute) which is usually attributed to the reorientation of water molecules surrounding the small molecules without the cleavage of hydrogen bond interactions, and the possible interactions of small hydrophobic molecules with water molecules via multiple van der Waals interactions (the feasibility of which is argued on the basis of small size of water molecules and the flexibility of their spatial configuration) [52]. Hydrophobic hydration (burial of polar interfaces), on the other hand, accompanies a negative enthalpy change (DH < 0), and a negative (unfavorable) entropy change (TDS < 0, enhanced ordering of the water molecules surrounding the hydrophobic surface), and positive heat capacity change (DCp > 0) [53,54]. It has been experimentally established that accommodation of small hydrophobic molecules in water is associated with enhanced ordering (constrained mobility) of the neighboring water molecules [55–57]. However, endeavors to rationalize such experimental data have led to contrasting views. The classical view centers on the formation of a semi-ordered, transient structure (often referred to as the ‘iceberg’ or ‘clathrate-like’ structure) of water molecules surrounding the hydrophobic solute. The formation of the socalled clathrate-like structure is ascribed to an enhanced hydrogen bonding interaction of water molecules [55,56] (enhancement in terms of strengthening [58] or/and increase in the number of intermolecular hydrogen bonds within the water network [59]). The classical interpretation explains the thermodynamic properties typical of ‘hydrophobic hydration’, namely, DCp > 0 and DH < 0, DS < 0. The reduced mobility of the water molecules surrounding

593

B.K. Paul et al. / Journal of Colloid and Interface Science 538 (2019) 587–596

the hydrophobic solute has been confirmed by NMR studies [57]. However, results based on neutron diffraction [60,61] and extended X-ray absorption fine structure [62] experiments conclude that the water molecules in the vicinity of small hydrophobic surfaces are not significantly different from those in bulk water (pure liquid). The dynamic view of the phenomenon, on the contrary, argues that the dynamics of the neighboring water molecules is considerably slowed down near the surface of the hydrophobic solutes (due to constrained rotational relaxation) whereas the strength of hydrogen bonds, and water-structure remain essentially unaffected [63,64]. An alternative explanation of the hydrophobic hydration as suggested by Lee [65] takes into account the difficulty (and hence the accompanying energetic cost) of accommodating a hydrophobic solute into the cavity of water network given the small size of water molecules. The structuring (ordering) of water molecules in the hydration shell of hydrophobic solutes has also been argued by Baldwin in recent time [66] to be emanating from attractive van der Waals forces. Very recently, Avbelj et al. [67] have provided experimental evidence in support of the classical view of the ‘hydrophobic hydration’ in terms of strengthening of hydrogen bonds between water molecules in the vicinity of (pure) hydrophobic solutes (resembling those in clathrate-like structures or ice). They have also shown that the water molecules involved in strengthened hydrogen bonding interaction exhibit extensive ordering in structural organization. According to the results of Avbelj et al. [67] the lack of intercalating water (which causes electrostatic screening of H-bonds in pure bulk of liquid water) plays an instrumental role underlying the strengthening of H-bonding interaction in the neighborhood of a hydrophobic solute, thus endorsing the classical interpretation of hydrophobic hydration.

3.5. Effect of ionic strength on the thermodynamics of interaction Fig. 6 displays the ITC profile of NHM-RNA interaction in the presence of varying concentrations of NaCl and the corresponding data are assembled in Table 3. RNA has a natural tendency of condensing counter-cations around the polyanionic phosphodiester backbone, which in turn fosters the possibility of competition with cationic ligands/drugs during binding in terms of expulsion of the counter-cations. Thus, a significant role of electrostatic interaction underlying the binding of the cationic drug (NHM) to RNA can be reasonably argued. Table 3 clearly shows a progressive lowering of the affinity constant values (KI as well as KII) with increasing NaCl concentrations. This result, in harmony with the spectroscopic results, corroborates to our inference of weakening of the NHM-RNA binding strength with increasing ionic strength of the medium. It could be important to note in this context that the thermodynamics of the interaction in the absence of NaCl sensibly differ from those in the presence of NaCl (as evident from comparison of the data presented in Tables 2 and 3 at 293 K). Usually increase of ionic strength of the medium accompanies lowering of the electrostatic repulsion between the sequential phosphate groups of the helix of a nucleic acid resulting in a reduced tendency of the double helix of the nucleic acid to unwind, that is, assisting the helix to shrink [68]. Such compaction of nucleic acid duplex with enhanced ionic strength of the medium has been previously reported in the literature [68,69]. This is, however, not unlikely that such compaction (reduced unwinding tendency) of the helix would influence the interaction of small drug molecules with RNA. Naturally, a precise molecular level interpretation of the exact mechanism influencing the NHM-RNA interaction in the presence of NaCl becomes difficult as it is governed by a complex interplay

Fig. 6. (a) ITC profile for NHM-RNA interaction in the presence of 20 mM NaCl depicting the raw heat burst curves (appropriately corrected for heat of dilution) at 293 K in 10 mM phosphate buffer (pH 7.40). The power sign convention of the instrument designates an exothermic (endothermic) process by upward (downward) heat bursts. (b) The ITC enthalpograms for NHM-RNA interaction in the presence of various NaCl concentrations as specified in the figure legend (20 mM NaCl: ▲, 30 mM NaCl: d, and 40 mM NaCl: s). The solid lines represent the best fit lines according to multiple sites binding model.

Table 3 Dependence of the thermodynamic parameters of NHM-RNA interaction on varying NaCl concentrations as derived from ITC.a

a

[NaCl] (mM)

KI (104)

nI

DH I

TDSI

DG I

DGpe

DGnpe

20 30 40 50

9.78 ± 0.8 8.84 ± 0.7 7.89 ± 0.6 6.95 ± 0.5

3.80 ± 0.11 4.53 ± 0.22 5.25 ± 0.21 3.02 ± 0.12

1.03 ± 0.18 3.26 ± 0.58 5.50 ± 0.95 7.70 ± 1.3

29.01 ± 1.2 30.98 ± 1.2 32.96 ± 1.3 34.9 ± 1.4

27.98 27.72 27.46 27.2

3.48 3.12 2.87 2.68

24.49 24.59 24.59 24.53

[NaCl] (mM) 20 30 40 50

KII (103) 9.79 ± 0.5 9.00 ± 0.4 8.22 ± 0.4 7.44 ± 0.4

nII 2.18 ± 0.11 3.48 ± 0.17 4.79 ± 0.24 3.95 ± 0.19

DHII 112.2 ± 8 156.1 ± 11 199.9 ± 13 244.4 ± 15

TDSII 90.16 ± 5.4 133.78 ± 8 177.41 ± 11 220.0 ± 13

DGII 22.04 22.27 22.49 24.4

DGpe 2.38 2.13 1.96 1.82

DGnpe 19.66 20.12 20.53 22.58

K values in M1, DH, TDS and DG (including DGpe and DGnpe) values in kJ mol1.

594

B.K. Paul et al. / Journal of Colloid and Interface Science 538 (2019) 587–596

of electrostatic interaction forces as well as subtle structural modifications of the RNA duplex. These data are further analyzed in association with van’t Hoff relationship leading to parsing of the Gibbs free energy of interaction and thus enabling a more critical insight into the molecular level interpretation of the binding forces raising deeper questions. The affinity constant value (Ki) can be related to the Na+ ion concentration in the medium through the following van’t Hoff relationship [17,48]:

 
dlnK i ¼ hu  dln Naþ

ð2Þ

where h represents the apparent charge on the bound drug and u denotes an estimate of the fraction of Na+ ions bound per phosphate group. The dependence of the affinity constant on the Na+ ion concentration eventually leads to partitioning of the total Gibbs free energy (DG) of the interaction into polyelectrolytic (DGpe) and nonpolyelectrolytic (DGnpe) components through the following relationships [17,48]:

  DGpe ¼ hu  RTln Naþ

ð3Þ

DG ¼ DGpe þ DGnpe

ð4Þ

Here, the polyelectrolytic component (DGpe) represents the contribution of electrostatic force to the total Gibbs free energy of interaction. Analysis of the ITC results based on the above equations (Eqs. (2)–(4)) shows that at 50 mM concentration of Na+, the contribution of electrostatic interaction (DGpe) amounts to 2.68 kJ mol1 (which is 9.8% of total DGI) and 1.82 kJ mol1 (which is 7.5% of total DGII), respectively for the formation of Complex I and Complex II (Table 3). Our calculations also reveal that with increasing concentration of NaCl, the formation of Complex I accompanies a gradual decrease of the DGpe value with a simultaneous reduction in the affinity constant (KI), Table 3. However, the contribution of the nonpolyelectrolytic component (DGnpe) to the total Gibbs free energy remains almost invariant to incremental addition of NaCl (Table 3). Furthermore, it is imperative to note in this context that the process of formation of Complex I in the presence of NaCl is characterized by an unfavorable enthalpic contribution (DHI > 0, Table 3) coupled with a favorable entropic contribution (TDSI > 0), and with added NaCl the process becomes increasingly entropy-dominant (TDSI gradually increases with addition of NaCl, Table 3). Collectively, these results point out a predominant contribution of hydrophobic interaction underlying the formation of Complex I as against electrostatic interaction. This is further corroborated from the aforementioned results on the effect of temperature on the binding thermodynamics (Table 2). As far as the formation of Complex II is considered, the process is found to be associated with a progressively decreasing magnitude of DGpe with added NaCl (Table 3), whereas the contribution of the nonpolyelectrolytic component (DGnpe) to the total Gibbs free energy of interaction exhibits a steady increase with incremental addition of NaCl (Table 3). The data summarized in Table 3 also suggest that formation of Complex II is an enthalpydominated process (DHII < 0, along with an unfavorable entropic contribution, that is, TDSII < 0), and with increasing NaCl concentration the process becomes increasingly enthalpy-dominant (gradually increasing contribution of DHII to the total free energy change (DGII) as against an increasingly unfavorable (TDSII < 0) contribution from the entropic component, Table 3). Our results based on the effect of temperature on the binding thermodynamics of NHM-RNA interaction, (Table 2) indicate that the process of formation of Complex II is described by the thermodynamic signature of hydrophobic hydration (solvation of hydrophobic (or nonpolar) molecular surfaces in aqueous medium [40,41,44–48]) rationaliz-

ing an unfavorable entropic contribution (TDSII < 0 in Table 2). The increasingly negative contribution from the entropic component (TDSII < 0, Table 3) with increasing ionic strength of the medium (addition of NaCl) thus further substantiates the governing role of hydrophobic hydration underlying the formation of Complex II in course of NHM-RNA binding. 4. Conclusions The salient finding of the present work is the revelation of the intrinsic complexities underlying the binding interaction of the potential anti-cancer photosensitizer NHM with RNA. Of particular interest in this context is the presence of two thermodynamically different RNA binding modes of NHM having substantially different affinity constants and thermodynamic parameters accompanying the formation of Complex I and II. All our data reveal the remarkably complex modulation of the hydration structure accompanying the NHM-RNA interaction process, e.g., the key role of ‘hydrophobic effect’ (release of ordered water molecules) underlying the formation of Complex I versus ‘hydrophobic hydration’ (solvation of hydrophobic (or nonpolar) molecular surfaces in aqueous medium) for formation of Complex II [40,41,44–48]. Furthermore, parsing of DG as elucidated from the effect of medium ionic strength on the binding thermodynamics reveals differential contributions of polyelectrolytic (DGpe) and nonpolyelectrolytic (DGnpe) components to the total free energy change (DG) for the two different thermodynamic modes of binding [17,48]. Cumulatively, it can be stated that the overall binding phenomenon is governed by a complex interplay of various molecular forces; the central role of hydrophobic effect/hydrophobic hydration coupled with the importance of electrostatic interaction. With a view to the extensive pharmacological applications of NHM and other b-carboline drugs, an in-depth understanding of their interaction with relevant biological targets appears pertinent. Furthermore, b-carboline drugs have found promising applications in photodynamic therapy due to their ability to produce singlet oxygen; however, under uncontrolled use these drugs might also induce DNA damage when exposed to UV radiation [70]. Naturally, a precise and comprehensive characterization of the strength and mode of interaction of the drugs with suitable receptors and their dependence on various extrinsic parameters (like temperature, medium ionic strength) is important, and we are optimistic that the as-employed experimental protocols and methodologies of the present study can be extended to other congeners of bcarboline toward their safe-engineered applications. It could also be important to note that the spectroscopic results do not provide a complete visualization of the interaction phenomenon, whereas the thermodynamic results based on ITC measurements can reveal the more complex nature of the interaction scenario. Acknowledgments We express our sincere thanks to the Central Instrumentation Facility (CIF) of our Institute for access to ITC and CD measurements. Appendix A. Supplementary material Supplementary data to this article can be found online at https://doi.org/10.1016/j.jcis.2018.12.011. References: [1] J.M. Berg, J.L. Tymoczko, L. Stryer, Biochemistry, fifth ed., W. H. Freeman and Company, New York, 2002.

B.K. Paul et al. / Journal of Colloid and Interface Science 538 (2019) 587–596 [2] T. Hermann, Rational ligand design for RNA: the role of static structure and conformational fexibility in target recognition, Biochimie 84 (2002) 869–875. [3] M.J. Fedor, J.R. Williamson, The catalytic diversity of RNAs, Nat. Rev. Mol. Cell. Biol. 6 (2005) 399–412. [4] M. Jovanovic, M.O. Hengartner, miRNAs and apoptosis: RNAs to die for, Oncogene 25 (2006) 6176–6187. [5] J.R. Thomas, P.J. Hergenrother, Targeting RNA with small molecules, Chem. Rev. 108 (2008) 1171–1224. [6] A.C. Cheng, V. Calabro, A.D. Frankel, Design of RNA-binding proteins and ligands, Curr. Opin. Struct. Biol. 11 (2001) 478–484. [7] Q. Vicens, E. Westhof, RNA As a Drug Target: the case of aminoglycosides, Chem. Bio. Chem. 4 (2003) 1018–1023. [8] Y. Tor, Targeting RNA with small molecules, Chem. Bio. Chem. 4 (2003) 998– 1007. [9] A. Sall, Z. Liu, H.M. Zhang, J. Yuan, T. Lim, Y. Su, D. Yang, microRNAs-based therapeutic strategy for virally induced diseases, Curr. Drug Discov. Technol. 5 (2008) 49–58. [10] E.H. Bayne, R.C. Allshire, RNA-directed transcriptional gene silencing in mammals, Trends Genet. 21 (2005) 370–373. [11] S.M. Hammond, A.A. Caudy, G.J. Hannon, Post-transcriptional gene silencing by double stranded RNA, Nat. Rev. Genet. 2 (2001) 110–119. [12] J. Gallego, G. Varani, Targeting RNA with small-molecule drugs: therapeutic promise and chemical challenges, Acc. Chem. Res. 34 (2001) 836–843. [13] V. Ambros, The functions of animal microRNAs, Nature 431 (2004) 350–355. [14] B.K. Paul, N. Guchhait, Exploring the strength, mode, dynamics, and kinetics of binding interaction of a cationic biological photosensitizer with DNA: implication on dissociation of the drug-DNA complex via detergent sequestration, J. Phys. Chem. B 115 (2011) 11938–11949. [15] B.K. Paul, N. Ghosh, S. Mukherjee, Interaction of an anti-cancer photosensitizer with a genomic DNA: from base pair specificity and thermodynamic landscape to tuning the rate of detergent-sequestered dissociation, J. Colloid Interface Sci. 470 (2016) 211–220. [16] E.N. Lorenzón, K.A. Riske, G.F. Troiano, G.C.A. Da Hora, T.A. Soares, E.M. Cilli, Effect of dimerization on the mechanism of action of aurein 1.2, Biochim. Biophys. Acta – Biomem. 1858 (2016) 1129–1138. [17] A. Kabir, G.S. Kumar, Targeting double-stranded RNA with spermine, 1naphthylacetyl spermine and spermidine: a comparative biophysical investigation, J. Phys. Chem. B 118 (2014) 11050–11064. [18] S. Nafisi, M. Bonsaii, P. Maali, M.A. Khalilzadeh, F. Manouchehri, b-carboline alkaloids bind DNA, J. Photochem. Photobiol. B 100 (2010) 84–91. [19] N. Foloppe, N. Matassova, F. Aboul-ela, Towards the discovery of drug-like RNA ligands?, Drug Discov Today 11 (2006) 1019–1027. [20] L. Manche, S.R. Green, C. Schmedt, M.B. Mathews, Interactions between double-stranded RNA regulators and the protein kinase DAI, Mol. Cell. Biol. 12 (1992) 5238–5248. [21] H. Bloom, J. Barchas, M. Sandler, E. Usdin, Progress in Clinical and Biological Research, b-Carbolines and Tetrahydroisoquinolines, vol. 90, Liss, A. R. D., Inc., New York, 1982. [22] M.W. Freyer, R. Buscaglia, A. Hollingsworth, J. Ramos, M. Blynn, R. Pratt, W.D. Wilson, E.A. Lewis, Break in the heat capacity change at 303 K for complex binding of netropsin to AATT containing hairpin DNA constructs, Biophys. J. 92 (2007) 2516–2522. [23] N.G. Anthony, D. Breen, G. Donoghue, A.I. Khalaf, S.P. Mackay, J.A. Parkinson, C. J. Suckling, A new synthesis of alkene-containing minor-groove binders and essential hydro bonding in binding to DNA and in antibacterial activity, Org. Biomol. Chem. 7 (2009) 1843–1850. [24] B.K. Paul, N. Ghosh, S. Mukherjee, Prototropic transformation and rotationalrelaxation dynamics of a biological photosensitizer norharmane inside nonionic micellar aggregates, J. Phys. Chem. B 118 (2014) 11209–11219. [25] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, third ed., Plenum, New York, 2006. [26] D. Reyman, M.H. Vinas, G. Tardajos, E. Mazario, The impact of dihydrogen phosphate anions on the excited-state proton transfer of harmane. Effect of bcyclodextrin on these photoreactions, J. Phys. Chem. A 116 (2012) 207–214. [27] A. Dias, A.P. Varela, M.G. Miguel, A.L. Macanita, R.S. Becker, b-Carboline photosensitizers. 1. Photophysics, kinetics and excited-state equilibria in organic solvents, and theoretical calculations, J. Phys. Chem. 96 (1992) 10290– 10296. [28] B.K. Paul, N. Ghosh, R. Mondal, S. Mukherjee, Contrasting effects of salt and temperature on niosome-bound norharmane: direct evidence for positive heat capacity change in niosome: b-cyclodextrin interaction, J. Phys. Chem. B 120 (2016) 4091–4101. [29] M. Balon, M.A. Munoz, C. Carmona, P. Guardado, M. Galan, A fluorescence study of the molecular interactions of harmane with the nucleobases, their nucleosides and mononucleotides, Biophys. Chem. 80 (1999) 41–52. [30] B.K. Paul, N. Guchhait, Differential interactions of a biological photosensitizer with liposome membranes having varying surface charges, Photochem. Photobiol. Sci. 11 (2012) 661–673. [31] R. Acher, Proteins, Nucleic Acids, Comprehensive Biochemistry, vol. 8, Elsevier Publishing Company, New York, 1963. [32] H.T. Steely Jr., D.M. Gray, R.L. Ratliff, CD of Homopolymer DNA-RNA hybrid duplexes and triplexes containing A-T or A-U base pairs, Nucl. Acids Res. 14 (1986) 10071–10090. [33] T.R. Pearce, E. Kokkoli, DNA Nanotubes and helical nanotapes via self-assembly of ssDNA-amphiphiles, Soft Matter 11 (2015) 109–117.

595

[34] N.C. Garbett, P.A. Ragazzon, J.B. Chaires, Circular dichroism to determine binding mode and affinity of ligand-DNA interactions, Nat. Proto. 2 (2007) 3166–3172. [35] G. Barone, G. Gennaro, A.M. Giuliani, M. Giustini, Interaction of Cd(II) and Ni(II) terpyridine complexes with model polynucleotides: a multidisciplinary approach, RSC Adv. 6 (2016) 4936–4945. [36] T. Sakamoto, E. Ennifar, Y. Nakamura, Thermodynamic study of aptamers binding to their target proteins, Biochimie 145 (2018) 91–97. [37] H. Kettiger, G. Québatte, B. Perrone, J. Huwyler, Interactions between silica nanoparticles and phospholipid membranes, Biochim. Biophys. Acta – Biomem. 2016 (1858) 2163–2170. [38] B.K. Paul, N. Ghosh, S. Mukherjee, Interaction of bile salts with b-cyclodextrins reveals nonclassical hydrophobic effect and enthalpy-entropy compensation, J. Phys. Chem. B 120 (2016) 3963–3968. [39] P. Sadatmousavi, E. Kovalenko, P. Chen, Thermodynamic characterization of the interaction between a peptide-drug complex and serum proteins, Langmuir 30 (2014) 11122–11130. [40] P.R. Connelly, J.A. Thomson, Heat capacity changes and hydrophobic interactions in the binding of FK506 and rapamycin to the FK506 binding protein, Proc. Natl. Acad. Sci. USA 89 (1992) 4781–4785. [41] S.W. Homans, Dynamics and thermodynamics of ligand-protein interactions, Top. Curr. Chem. 272 (2007) 51–82. [42] D. Ondo, M. Costas, Complexation thermodynamics of a-cyclodextrin with ionic surfactants in water, J. Colloid Interface Sci. 505 (2017) 445–453. [43] U. Anand, M. Mukherjee, Exploring the self-assembly of a short aromatic Ab (16–24) peptide, Langmuir 29 (2013) 2713–2721. [44] M.F.M. Sciacca, C. Tempra, F. Scollo, D. Milardi, C. La Rosa, Amyloid growth and membrane damage: current themes and emerging perspectives from theory and experiments on Ab and hIAPP, Biochim. Biophys. Acta – Biomem. 2018 (1860) 1625–1638. [45] K.P. Murphy, P.L. Privalov, S.J. Gill, Common features of protein unfolding and dissolution of hydrophobic compounds, Science 247 (1990) 559–561. [46] E. Kerek, M. Hassanin, W. Zhang, E.J. Prenner, Preferential binding of Inorganic Mercury to specific lipid classes and its competition with Cadmium, Biochim. Biophys. Acta – Biomem. 1859 (2017) 1211–1221. [47] R.S. Spolar, J.R. Livingstone, M.T. Record Jr, Use of liquid hand amide transfer data to estimate contributions to thermodynamic functions of protein folding from the removal of nonpolar and polar surface from water, Biochemistry 31 (1992) 3947–3955. [48] R.A. Fideles, G.M.D. Ferreira, F.S. Teodoro, O.F.H. Adarme, L.H.M. da Silva, L.F. Gil, L.V.A. Gurgel, Trimellitated sugarcane bagasse: a versatile adsorbent for removal of cationic dyes from aqueous solution. Part I: batch adsorption in a monocomponent system, J. Colloid Interface Sci. 515 (2018) 172–188. [49] C. Tanford, The Hydrophobic Effect: Formation of Micelles and Biological Membranes, Wiley, NewYork, 1980. [50] J.M. Sturtevant, Heat capacity and entropy changes in processes involving proteins, Proc. Natl. Acad. Sci. USA 74 (1977) 2236–2240. [51] B.K. Paul, N. Ghosh, S. Mukherjee, Interplay of multiple interaction forces: binding of norfloxacin to human serum albumin, J. Phys. Chem. B 119 (2015) 13093–13102. [52] E. Fisicaro, C. Compari, A. Braibanti, Hydrophobic hydration processes. General thermodynamic model by thermal equivalent dilution determinations, Biophys. Chem. 151 (2010) 119–138. [53] P.W. Snyder, M.R. Lockett, D.T. Moustakes, G.M. Whitesides, Is it the shape of the cavity, or the shape of the water in the cavity?, Eur Phys. J. 223 (2014) 853–889. [54] V.V. Yaminsky, E.A. Vogler, Hydrophobic hydration, Curr. Opin. Colloid Interface Sci. 6 (2001) 342–349. [55] H.S. Frank, M.W. Evans, Free volume and entropy in condensed systems III. Entropy in binary liquid mixtures; partial molal entropy in dilute solutions; structure and thermodynamics in aqueous electrolytes, J. Chem. Phys. 13 (1945) 507–532. [56] W. Kauzmann, Some factors in the interpretation of protein denaturation, Adv. Protein Chem. 14 (1959) 1–63. [57] R. Haselmeier, M. Holz, W. Marbach, H. Weingartner, Water dynamics near a dissolved noble-gas – first direct experimental-evidence for a retardation effect, J. Phys. Chem.-Us 99 (1995) 2243–2246. [58] N. Muller, Search for a realistic view of hydrophobic effects, Acc. Chem. Res. 23 (1990) 23–28. [59] G. Nemethy, H.A. Scheraga, Structure of water and hydrophobic bonding in proteins. II. Model for the thermodynamic properties of aqueous solutions of hydrocarbons, J. Chem. Phys. 36 (1962) 3401–3417. [60] C.A. Koh, R.P. Wisbey, X.P. Wu, R.E. Westacott, A.K. Soper, Water ordering around methane during hydrate formation, J. Chem. Phys. 113 (2000) 6390– 6397. [61] P. Buchanan, N. Aldiwan, A.K. Soper, J.L. Creek, C.A. Koh, Decreased structure on dissolving methane in water, Chem. Phys. Lett. 415 (2005) 89–93. [62] D.T. Bowron, A. Filipponi, C. Lobban, J.L. Finney, Temperature-induced disordering of the hydrophobic hydration shell of Kr and Xe, Chem. Phys. Lett. 293 (1998) 33–37. [63] G. Stirnemann, J.T. Hynes, D. Laage, Water hydrogen bond dynamics in aqueous solutions of amphiphiles, J. Phys. Chem. B 114 (2010) 3052–3059. [64] D. Laage, G. Stirnemann, F. Sterpone, R. Rey, J.T. Hynes, Reorientation and allied dynamics in water and aqueous solutions, Annu. Rev. Phys. Chem. 62 (2011) 395–416.

596

B.K. Paul et al. / Journal of Colloid and Interface Science 538 (2019) 587–596

[65] B. Lee, The physical origin of the low solubility of nonpolar solutes in water, Biopolymers 24 (1985) 813–823. [66] R.L. Baldwin, Dynamic hydration shell restores Kauzmann’s 1959 explanation of how the hydrophobic factor drives protein folding, Proc. Natl. Acad. Sci. USA 111 (2014) 13052–13056. [67] J. Grdadolnik, F. Merzel, F. Avbelj, Origin of hydrophobicity and enhanced water hydrogen bond strength near purely hydrophobic solutes, Proc. Natl. Acad. Sci. USA 114 (2017) 332–1327.

[68] I. Manzano, A.L. Zydney, Quantitative study of RNA transmission through ultrafiltration membranes, J. Membr. Sci. 544 (2017) 272–277. [69] D.R. Latulippe, A.L. Zydney, Salt-induced changes in plasmid DNA transmission through ultrafiltration membranes, Biotechnol. Bioeng. 99 (2008) 390–398. [70] M. Vignoni, F.A. Rasse-Suriani, K. Butzbach, R. Erra-Balsells, B. Epe, F.M. Cabrerizo, Mechanisms of DNA damage by photoexcited 9-methyl-bcarbolines, Org. Biomol. Chem. 11 (2013) 5300–5309.