Host–guest inclusion complexes of RNA nucleosides inside aqueous cyclodextrins explored by physicochemical and spectroscopic methods

Abstract

Stable host–guest inclusion complexes have been formed with guest RNA nucleosides inside aqueous α- and β-cyclodextrins. α-Cyclodextrin has been found to have favorable structural features for inclusion with uridine and cytidine, whereas β-cyclodextrin has that with all the four nucleosides, e.g., adenosine, guanosine, uridine and cytidine. The formation and nature of the inclusion complexes have been characterized using surface tension study, Job's method by ultraviolet spectroscopy and pH measurements. The limiting apparent molar volume, viscosity B-coefficient, limiting apparent molar adiabatic compressibility and limiting molar refraction data have been used to characterize the interaction between nucleosides and cyclodextrins in the experimental ternary solution systems. The inclusion phenomenon has been confirmed by proton NMR study. Association constants and thermodynamic parameters have been evaluated for the formed inclusion complexes by ultraviolet spectroscopy.

---

1. Introduction

Cyclodextrins are cyclic oligosaccharides containing six (α-CD), seven (β-CD) and eight (γ-CD) glucopyranose units, which are bound by α-(1–4) linkages forming a truncated conical structure with a hydrophobic interior and hydrophilic rims having primary and secondary –OH groups (Scheme 1).¹ Because of their unique structure, they can build up stable host–guest arrangements, i.e., a cyclodextrin can accommodate the hydrophobic moiety of a guest molecule into its hydrophobic cavity and the polar rims can stabilize the polar part of the guest, if any.² This is why CDs are of particular interest among various approaches to enhance the apparent solubility of biomolecules.³ CDs can act as molecular receptors (hosts) for a wide variety of organic and inorganic, as well as biological and pharmaceutical guest molecules, forming host–guest inclusion complexes or supramolecular assemblies.^4,5 The use of a CD for solubility, bioavailability, safety, stability and as a carrier may be achieved by the formation of inclusion complexes with biologically active molecules.^6,7 In fact, the use of CDs already has a long history in pharmaceuticals, pesticides, foodstuffs, toiletry, textile processing, supramolecular host–guest chemistry, models for studying enzyme activity, molecular recognition, molecular encapsulation, studying intermolecular interactions and chemical stabilization.^8–10 RNA nucleosides are very important biomolecules (Scheme 1) having enormous applications in the field of modern biological sciences, e.g., RNA-based information technologies, RNA cloning, recombinant RNA technology and other genetic engineering processes.¹¹ Xiang et al. demonstrated the formation of inclusion complexes of purine nucleosides with β-CD, as well as their stability and carrying capacity by solubility, circular dichroism, ultraviolet spectrophotometry and NMR techniques.¹² Formoso illustrated the binding of nucleic acid monomer units as well as dinucleoside phosphates with β-CD by circular dichroism studies and calculated the binding constants and thermodynamic parameters, which show significant interactions between β-CD and the nucleotide moieties.^13,14 In the present study, the inclusions of all four RNA nucleosides into aqueous α- and β-CD have been explored, especially towards their formation, stabilization, carrying and controlled release without chemical modification by different reliable methods (e.g., ¹H NMR study), focusing mainly on the encapsulation of the RNA nucleosides into the cavity of α- and β-CD. Associated thermodynamic parameters have also been evaluated to communicate a quantitative idea about encapsulation of the above RNA nucleosides while complexed with cyclodextrins. The nature of formation of inclusion complexes of the chosen adenosine (A), guanosine (G), uridine (U) and cytidine (C) have been studied in 0.001, 0.0025, and 0.004 molar aqueous α- and β-CD solutions by various physicochemical techniques.

Scheme 1 (a) Molecular structure, (b) stereo-chemical configuration (n = 6 for α-CD and n = 7 for β-CD), (c) truncated conical structure of α and β-cyclodextrins, (d) molecular structure of RNA nucleosides indicating the aromatic protons.

2. Experimental

2.1. Materials

The RNA nucleosides and cyclodextrins of puriss grade were bought from Sigma-Aldrich, Germany and used as purchased. Purity of adenosine, guanosine, uridine, cytidine, α-cyclodextrin and β-cyclodextrin was ≥0.99, 0.98, 0.99, 0.99, 0.98 and 0.98, respectively.

2.2. Apparatus and procedure

Prior to the start of the experimental study, solubility of the chosen cyclodextrins in triply distilled and degassed water (with a specific conductance of 1 × 10⁻⁶ S cm⁻¹) and solubility of the nucleosides in aqueous cyclodextrins were precisely checked, and it was observed that the selected nucleosides were freely soluble in all proportions of aqueous cyclodextrins. First, the cyclodextrin solutions of required molarity were prepared, and then, these solutions were used to make all the nucleoside stock solutions. All the stock solutions of the nucleosides were prepared by mass (weighed by Mettler Toledo AG-285 with uncertainty 0.0003 g), and then, the working solutions were obtained by mass dilution at 298.15 K. The conversions of molarity into molality were done using density values.¹⁵ Adequate precautions were made to reduce evaporation losses during mixing.

The surface tension experiments were carried out by a platinum ring detachment method using a Tensiometer (K9, KRŰSS; Germany) at the experimental temperature. The accuracy of the measurement was within ±0.1 mN m⁻¹. Temperature of the system was maintained by circulating auto-thermostated water through a double-wall glass vessel containing the solution.

UV-Visible spectra were obtained by a JASCO V-530 UV-VIS Spectrophotometer, with an uncertainty of wavelength resolution of ±2 nm. The measuring temperature was held constant by a thermostat.

The pH values of the experimental solutions were measured by a Mettler Toledo Seven Multi pH meter with uncertainty of 0.009. The measurements were made in a thermostated water bath maintaining the temperature at 298.15 K. The uncertainty in temperature was 0.01 K.

The densities (ρ) of the solvents were measured by means of a vibrating U-tube Anton Paar digital density meter (DMA 4500 M) with a precision of ±0.00005 g cm⁻³ maintained at ±0.01 K of the desired temperature. It was calibrated by passing triply distilled, degassed water and dry air.

The viscosities (η) were measured using a Brookfield DV-III Ultra Programmable Rheometer with fitted spindle size-42. The detailed description was described earlier.¹⁶

Refractive index was measured with the help of a Mettler Toledo Digital Refractometer. The light source was a LED, λ = 589.3 nm. The refractometer was calibrated twice using distilled water and calibration was checked after every few measurements. The uncertainty of refractive index measurement was ±0.0002 units.

Ultrasonic speed (u) was measured by a multi-frequency ultrasonic interferometer (Model M-81) from Mittal Enterprises, India. The interferometer working at 5 MHz was used based on the same principle as that described in the study of Ekka et al.¹⁶ The uncertainty in the speed is ±0.2 m s⁻¹ and the temperature was controlled within ±0.01 K using a Lauda thermostat during measurement.

NMR spectra were obtained in D₂O unless otherwise stated. ¹H NMR spectra were obtained at 400 MHz and 500 MHz using a Bruker AVANCE 400 MHz and Bruker AVANCE 500 MHz instruments at 298.15 K. Signals are quoted as δ values in ppm using residual protonated solvent signals as internal standard (D₂O: δ 4.79 ppm). Data are reported as chemical shifts.

3. Result and discussion

3.1. Surface tension study reveals the inclusion and also the stoichiometric ratio of the inclusion complexes

Surface tension (γ) was measured at 298.15 K for aqueous α- and β-cyclodextrins, which was found to be almost constant with increasing molarity (Fig. 1 and S1, Table S1, ESI†).¹⁷ Surface tensions (γ) were observed with corresponding concentrations of four nucleosides in different molarities of α- and β-cyclodextrins (Fig. 1 and S1†), in which each curve shows a similar increasing trend to that of pure nucleoside in the γ vs. concentration plot. This trend is observed due to the presence of hydrophilic phenolic –OH, –NH₂ and 〉C=O groups in nucleoside molecules, which form extensive H-bonding amongst themselves, resulting in an increase in γ with increasing concentration.

Fig. 1 Plot of surface tension with increasing concentration of adenosine in different molarities (w) of (a) α-cyclodextrin and (b) β-cyclodextrin at 298.15 K.

The plausibility of formation of an inclusion complex can be predicted from the surface tension (γ) study,⁸ wherein the formation of an inclusion complex has been confirmed from the break point in the curve of surface tension vs. concentration. The 1 : 1 and 1 : 2 stoichiometry of the host : guest inclusion complexes has been confirmed from the appearance of single and double by the break point in the γ vs. conc. curve.¹⁷ The value of γ and the corresponding concentration of nucleosides at the break point have been determined from the two intersecting straight lines, which shows a lowering in value with increasing molarity of cyclodextrins (Table 1), indicating the feasibility of inclusion with increasing amount of cyclodextrins in solution.^18–20

Table 1 Values of surface tension (γ) at the break point with corresponding concentration of RNA nucleosides in different molarities of aqueous α- and β-cyclodextrin, respectively, at 298.15 K^a

Aq. solvent mixture	Adenosine		Guanosine		Uridine		Cytidine
	Conc/mM	γ^a/mN m^−1	Conc/mM	γ^a/mN m^−1	Conc/mM	γ^a/mN m^−1	Conc/mM	γ^a/mN m^−1
a Standard uncertainties (u); temperature: u(T) = ±0.01 K, surface tension: u(γ) = ±0.1 mN m^−1.b w1 and w2 are the molarity of α- and β-CD in aqueous mixture, respectively.
w1 = 0.001^b	—	—	—	—	5.97	77.25	5.96	77.22
w1 = 0.0025^b	—	—	—	—	5.72	76.13	5.35	75.81
w1 = 0.004^b	—	—	—	—	5.21	75.17	4.64	74.76
w2 = 0.001^b	6.00	77.16	5.90	76.95	6.28	77.42	5.97	77.39
w2 = 0.0025^b	5.27	75.75	5.20	75.68	5.82	76.15	5.63	76.03
w2 = 0.004^b	4.49	74.62	3.89	74.14	5.42	75.37	4.93	74.89

---

In the surface tension curves for adenosine and guanosine with different molarities of α-CD, steady increase in surface tension was found against nucleoside concentration, i.e., there is no break point (Fig. 1 and S1†), suggesting that there may not be any possibility of formation of an inclusion complex. However, the similar plots for uridine and cytidine with α-CD clearly indicate single break point (Fig. S1†) in each curve at specific concentrations, i.e., after a certain point surface tension becomes relatively steady with increasing concentration of the nucleosides. This is the indication of formation of an inclusion complex, which occurs by the insertion of a relatively hydrophobic pyrimidine moiety into the hydrophobic cavity of α-CD. The pyrimidine moiety of uridine and cytidine enter the hydrophobic cavity of α-CD from the wider rim, which is geometrically allowed and also makes more fitting contact with the cyclodextrin cavity. The hydrophilic pentose sugar part remains outside and is stabilized with H-bonds with the hydrophilic groups of the wider rim of cyclodextrin and also the surrounding water molecules (scheme 2a). The value of the corresponding concentration of nucleosides at the break point was found to be lowered with increasing molarity of α-CD (Table 1), indicating the feasibility of inclusion with increasing amount of α-CD in solution.

Scheme 2 (a) Schematic of formation of inclusion complexes of RNA nucleosides with α-cyclodextrin. (b) Schematic of formation of inclusion complexes of RNA nucleosides with β-cyclodextrin.

In the surface tension (γ) study for the aforesaid four nucleosides with β-CD, there is a single break point (Fig. 1 and S1†) in each γ vs. conc. curve, which clearly indicates that β-CD can form 1 : 1 inclusion complexes with both the pyrimidine and purine moiety. Herein also inclusion occurs from the wider rim of β-CD, as that from the other rim would be geometrically and energetically unfavorable; the hydrophilic sugar moiety remains hydrated at the outside of the cyclodextrin cavity and stabilized with H-bonds with the secondary hydrophilic groups of cyclodextrin (Scheme 2b). At each break point the corresponding concentrations of the nucleosides were found to be lower with increasing molarity of β-CD (Table 1), indicating the easiness of inclusion in the presence of a greater amount of β-cyclodextrin.

3.2. Job plot confirms the stoichiometry of host–guest inclusion complexes

One of the first methods used for determination of the stoichiometry of inclusion complexes was Job's method, known as the continuous variation method.²¹ Herein, the solutions of each nucleoside and cyclodextrin were mixed at different molar ratios, R = [nucleoside]/([nucleoside] + [CD]) keeping the total concentration [nucleoside] + [CD] constant, and the mole fraction of the nucleoside varies in the range of 0–1 (Tables S2–S7†). The stoichiometry for each complex was determined by plotting ΔA × R against R (where ΔA is the absorbance difference of the nucleoside without and with cyclodextrin), (Fig. 2 and S2†).²² Absorbance values were measured at 261 nm for uridine, 270 nm for cytidine, 259 nm for adenosine and 253 nm for guanosine at 298.15 K for a series of solutions. The maximum deviation of R gives the stoichiometry of the inclusion complex (R = 0.5 for 1 : 1 complexes; R = 0.33 for 1 : 2 complexes; R = 0.66 for 2 : 1 complexes). In this study, the plots of U + α-CD, C + α-CD, A + β-CD, G + β-CD, U + β-CD and C + β-CD show maxima at R = 0.5, suggesting each of these forms an inclusion complex having a 1 : 1 molar ratio.

Fig. 2 Job plot of (a) cytidine-α-cyclodextrin system at λ_max = 270 nm, (b) adenosine-β-cyclodextrin system at λ_max = 259 nm at 298.15 K. R = [nucleoside]/([nucleoside] + [CD]), ΔA = absorbance difference of the nucleoside without and with cyclodextrin.

3.3. pH study indicates inclusion phenomenon

Study of pH data at 298.15 K for the four nucleosides provides important clues about inclusion.²³ The nucleosides contain organic basic groups in their hydrophobic aromatic moiety. Because of this, there should be a slow increase in the pH value with increasing concentration of nucleosides, which is reflected in the adenosine + α-CD and guanosine + α-CD systems (Fig. 3 and S3, Table S1†), possibly indicating that no inclusion has been occurred. However, uridine + α-CD and cytidine + α-CD systems show typical inclusion natures, i.e., pH values initially increase with increasing concentration of nucleosides but at higher concentration range they become almost steady (Fig. 3 and S3†), which probably indicates that the hydrophobic moiety containing the basic group is encapsulated in the cyclodextrin cavity (Scheme 2a).

Fig. 3 Plot of pH with increasing concentration of nucleosides: (a) molarity (w₁) of α-cyclodextrine = 0.001 and (b) molarity (w₂) of β-cyclodextrine = 0.001 at 298.15 K.

In the case of nucleoside + β-CD systems, all the pH versus concentration curves show a similar trend, i.e., slow increase of pH at the beginning and approximately constant value at higher concentration (Fig. 3 and S3†), which is again an indication of encapsulation of the hydrophobic-basic residue of the nucleosides in the hydrophobic cavity of β-CD (Scheme 2b).

3.4. Apparent molar volume, viscosity B-coefficient and solvation number

The characteristic behavior of interactions (here, inclusion) of solute can also be obtained from the apparent molar volume and viscosity B-coefficient. Both the limiting molar volume (ϕ_v^o) and viscosity B-coefficient signify the solute–solvent interactions in the RNA nucleoside + aq. cyclodextrin ternary solution systems. The limiting molar volume (ϕ_v^o) and viscosity B-coefficient have been obtained from the appropriate equations using the experimental values of density (ρ) and viscosity (η), respectively, and are presented in Tables S1, S8 and S9† and Fig. 4 and 5. The inspection of Fig. 4 and 5 shows that the limiting molar volume (ϕ_v^o) and viscosity B-coefficient both increases regularly from adenosine to cytidine of RNA nucleosides in all concentrations (w) of aqueous α-cyclodextrin and decreases with increasing temperature. However, the values show a different nature in aqueous β-cyclodextrin, wherein both were higher for adenosine and guanosine than for uridine and cytidine. The observed trends are as follows:

A < G ≪ U < C in aq. α-cyclodextrin

Fig. 4 (a) Limiting apparent molar volume (ϕ_v^o) and (b) viscosity B-coefficient of nucleosides in different molarities (w₁) of aqueous α-CD at different absolute temperatures (T).

Fig. 5 (a) Limiting apparent molar volume (ϕ_v^o) and (b) viscosity B-coefficient of nucleosides in different molarities (w₂) of aqueous β-CD at different absolute temperatures (T).

and

U < C ≪ A < G in aq. β-cyclodextrin

The trend signals that interactions (solute–solvent interactions) are enhanced from adenosine to guanosine to uridine to cytidine in aqueous α-cyclodextrin solution. This is due to the fact that α-CD shows a more favorable interaction with pyrimidine based nucleosides than with the purine based nucleosides. However, the trend is reversed in the case of β-CD; the solute–solvent interaction is higher for adenosine and guanosine compared to uridine and cytidine, which shows that in these cases the purine based nucleosides have favorable fitting for inclusion due to the size of the cavity of β-CD.¹⁸ These data and results support the observed data in surface tension and pH, which were discussed earlier.

Solvation numbers (S_n) are evaluated from the apparent molar volume and viscosity B-coefficient. S_n expresses the solvation of the nucleosides by the cyclodextrin molecule, i.e., the interaction between the polar groups of the guest and the –OH groups at the primary and secondary rims of cyclodextrin.^23,24 In Table 2, the solvation numbers are listed, which show that after the formation of an inclusion complex, the nucleosides form H-bonds with the –OH groups at the primary and secondary rims of cyclodextrin that further stabilize the inclusion complex.

Table 2 Solvation number (S_n) of nucleosides in different molarities of aqueous α- and β-cyclodextrin mixtures at different temperatures

Temp^b/K	Sn^a
	Adenosine	Guanosine	Uridine	Cytidine	Adenosine	Guanosine	Uridine	Cytidine
a Mean error in solvation number = ±0.02.b Standard uncertainties in temperature u are: u(T) = ±.01 K.c w1 and w2 are the molarity of α- and β-CD in aqueous mixture, respectively.
	w1 = 0.001^c				w2 = 0.001^c
298.15	9.24	10.47	11.30	12.37	13.40	14.28	11.49	12.47
303.15	9.16	9.70	10.44	10.92	11.61	12.32	10.27	10.97
308.15	9.35	10.24	11.10	11.91	12.70	13.01	10.80	11.81
	w1 = 0.0025^c				w2 = 0.0025^c
298.15	10.33	10.55	11.48	12.28	13.53	14.47	12.28	12.76
303.15	9.54	10.17	10.59	10.98	11.84	12.21	11.05	11.38
308.15	9.69	9.03	11.24	11.78	12.66	12.97	11.30	11.93
	w1 = 0.004^c				w2 = 0.004^c
298.15	11.74	12.63	13.27	13.99	14.24	14.68	13.07	13.36
303.15	10.85	12.18	12.49	12.87	12.57	13.30	11.68	11.97
308.15	11.55	12.58	12.76	12.94	13.24	13.86	12.39	12.95

---

3.5. Ultrasonic speed: interactions between nucleosides and cyclodextrins

Ultrasonic speed in the present ternary system provides information about interactions among nucleosides and cyclodextrins (Tables S1 and S10†). The values of apparent molar adiabatic compressibility (ϕ_K) and limiting apparent molar adiabatic compressibility (ϕ_K^o) have been evaluated at 298.15 K from the measured data (Table S11†). ϕ_K^o is an important parameter that gives information about the extent of interaction between nucleosides and cyclodextrins,²⁵ which are shown in Fig. 6a against the molarity of α and β-CD. The observed ϕ_K^o values follow the order

C + α-CD > U + α-CD ≫ G + α-CD > A + α-CD in the case of α-CD

and

G + β-CD > A + β-CD ≫ C + β-CD > U + β-CD in the case of β-CD,

which is in good agreement with the data found from density and viscosity measurements. The results suggest that the pyrimidine based nucleosides are more feasible for interaction with α-CD. On the other hand, both the purine and pyrimidine based nucleosides have great tendency for interaction with β-CD, but from Fig. 6a, it is observed that purine based nucleosides have greater efficiency for interaction with β-CD than the pyrimidine based nucleosides. Fig. 6a also shows the increasing value of ϕ_K^o with increasing molarity of both CDs, suggesting that the interactions become stronger with increasing molarity of CDs in solution.

Fig. 6 (a) Limiting partial molar adiabatic compressibilities (ϕ_K^o) and (b) limiting molar refraction (R_M^o) of nucleosides at different molarities of α- and β-cyclodextrin at 298.15 K.

3.6. Refractive index shows compactness of inclusion complexes

Refractive index (RI) measurement also provides important information about molecular interactions in solution. Higher value of RI (Tables S1 and S10†), molar refraction (R_M) and consequently the limiting molar refraction (R_M^o) (Table S11†) suggest that the solute–solvent interaction is higher in that solution system. As a result, the medium becomes denser or more compact.¹⁸ In the present ternary solution system, the interactions occurring between the four different nucleosides and two different cyclodextrins are explored. The R_M^o values of different nucleoside-CD systems for different molarities of α- and β-CD at 298.15 K are shown in Fig. 6b, which represents a clear comparison among them. It is evident from Fig. 6b that α-CD interacts more strongly with cytidine and uridine than with adenosine and guanosine; moreover, the interactions strengthen with increasing molarity o -CD. The order of interaction is C + α-CD > U + α-CD ≫ G + α-CD > A + α-CD, which is probably due to favorable inclusion of uridine and cytidine into the cavity of α-CD, and unfavorable inclusion of adenosine and guanosine, having purine moieties.

The R_M^o values for nucleoside + β-CD systems also show significant variation, and the order of which is G + β-CD > A + β-CD ≫ C + β-CD > U + β-CD (Fig. 6b). This is also understandable from the point of view of host–guest size. Guanosine and adenosine contain the larger purine moiety, which can be better fitted into the hydrophobic cavity of β-CD, making a compact structure, which is reflected in their high R_M^o values, whereas fitting of the smaller pyrimidine moiety of uridine and cytidine into the hydrophobic cavity of β-CD is not so good, resulting in a less dense structure and lower R_M^o values. This data also supports the findings from density, viscosity and acoustic measurements.

3.7. NMR Study: confirmation of inclusion

Insertion of a guest molecule into the hydrophobic cavity of cyclodextrin results in the chemical shift of the guest as well as of the cyclodextrin molecule in the NMR spectra, which is due to the interaction of the host with the guest molecule.²⁶ In the case of aromatic compounds, the spectral changes that can be observed upon inclusion is the diamagnetic shielding of the aromatic guest with the interacting atoms of the host molecule.²⁷ In the structure of cyclodextrin, the H3 and H5 hydrogens are situated inside the conical cavity; in particular, the H3 are placed near the wider rim, whereas H5 are placed near the narrower rim of the cyclodextrin molecule. The other, H1, H2 and H4, hydrogens are located at the exterior of the cyclodextrin molecule (Scheme 1).²⁸

The molecular interactions have been studied by ¹H NMR spectra. Upon inclusion the signals of H3 and H5 of cyclodextrin as well as the interacting aromatic protons of the nucleosides show considerable upfield shift (Δδ). In this study, the guest nucleosides also contain the sugar residue, which contain the hydrophilic –CH(OH)– groups and signals of which merge with that of the H3 and H5 of cyclodextrin. Therefore, it is better to investigate the chemical shifts (Δδ) of the aromatic protons here rather than that of the H3 and H5 of cyclodextrin (though upfield chemical shift is observable for H3 and H5, the values are not clear), (Scheme 1). Table 3 lists the chemical shift values (Δδ) of the aromatic protons of different nucleosides. The shifts of adenosine and guanosine when they interact with α-CD are negligible (Fig. S4 and S5†), whereas that of the uridine and cytidine with α-CD are considerably high (Fig. S6 and S7†). Therefore, it is clear that adenosine and guanosine do not form inclusion complexes with α-CD, whereas uridine and cytidine form inclusion complexes with α-CD. In the case of β-CD, considerable chemical shifts are observed for all the four nucleosides, confirming the formation of inclusion complexes between each of the four different RNA nucleosides with β-CD (Fig. S8–S11†).

Table 3 Change in chemical shifts (ppm) of the protons at the aromatic residue of the nucleosides when mixed with α- and β-cyclodextrins in 1 : 1 molar ratio in D₂O at 298.15 K^a

Δδ
Adenosine		Guanosine	Uridine		Cytidine
H2	H8	H8	H5	H6	H5	H6
a Standard uncertainties in temperature u are: u(T) = ±0.01 K.
α-Cyclodextrin
0.004	0.002	0.001	0.152	0.105	0.098	0.136
β-Cyclodextrin
0.128	0.122	0.161	0.138	0.121	0.080	0.108

---

3.8. Ultraviolet spectroscopy: association constants and thermodynamic parameters

The association constants K_a for different nucleoside-cyclodextrin systems have been evaluated by spectroscopic methods on the basis of changes of molar absorptivity of the nucleosides when complexed with the cyclodextrin molecules. This is due to the changes in the polarity of the environment of the chromophore of the nucleoside when it goes from the polar aqueous environment to the apolar cavity of cyclodextrin. Changes in absorption intensity of uridine (261 nm), cytidine (270 nm), adenosine (259 nm) and guanosine (253 nm) were studied as a function of concentration of cyclodextrin to determine the value of K_a (Tables S12–S17†). On the basis of the reliable Benesi–Hildebrand method for a 1 : 1 host–guest complex, the double reciprocal plots have been drawn using the equation as follows (Fig. S12†):^29,30

The values of the association constants for each of the systems were evaluated by dividing the intercept by the slope of the straight line of the double reciprocal plot (Table 4).^31,32

Table 4 Association constant (K_a) and thermodynamic parameters ΔH^o, ΔS^o and ΔG^o of different nucleoside-cyclodextrin inclusion complexes

	Temp^a/K	Ka^b (×10^−3)/M^−1	ΔH^o^b/kJ mol^−1	ΔS^o^b/J mol^−1 K^−1	ΔG^o^b (298.15 K)/kJ mol^−1
a Standard uncertainties in temperature u are: u(T) = ±0.01 K.b Mean errors in Ka = ±0.02 × 10^−3 M^−1; ΔH^o = ±0.01 kJ mol^−1; ΔS^o = ±0.01 J mol^−1 K^−1; ΔG^o = ±0.01 kJ mol^−1.
U + α-CD	298.15	1.40	−9.44	−5.52	−7.80
	303.15	1.22	—	—	—
	308.15	1.05	—	—	—
C + α-CD	298.15	1.49	−9.69	−6.09	−7.87
	303.15	1.31	—	—	—
	308.15	1.11	—	—	—
A + β-CD	298.15	1.58	−9.00	−3.54	−7.94
	303.15	1.43	—	—	—
	308.15	1.20	—	—	—
G + β-CD	298.15	1.70	−8.99	−3.30	−8.01
	303.15	1.47	—	—	—
	308.15	1.30	—	—	—
U + β-CD	298.15	1.04	−9.07	−5.34	−7.48
	303.15	0.92	—	—	—
	308.15	0.79	—	—	—
C + β-CD	298.15	1.10	−8.71	−3.87	−7.55
	303.15	0.99	—	—	—
	308.15	0.85	—	—	—

---

Thermodynamic parameters can easily be derived from the association constants found by the abovementioned method with the help of the van 't Hoff equation (eqn (2)) as follows:³³

There is a linear relationship between ln K_a and 1/T in the abovementioned equation (eqn (2)) (Fig. S13†). Based on eqn (2), the thermodynamic parameters ΔH^o, ΔS^o and ΔG^o for the formation of the inclusion complex can be obtained (Table S18†).

The value of ΔG^o was found to be negative, which suggests that the inclusion process proceeds spontaneously. ΔH^o and ΔS^o were also found to be negative, suggesting that the inclusion process is exothermic and entropy controlled, not entropy driven (Table 4). This is expected, as while the inclusion complex is formed between cyclodextrin and any guest molecule a molecular association occurs, resulting in a drop of entropy, which is unfavorable for the spontaneity of the inclusion complex formation. However, this effect is conquered by the higher negative value of ΔH^o, making the overall inclusion process thermodynamically favorable.

3.9. Structural influence of cyclodextrins

Formation of host–guest inclusion complex also depends on the size of the incoming guest molecules and the cavity diameter of the host molecule.³⁴ As the cavity diameters of the cyclodextrins are fixed, it is more convenient to discuss the size effect of the chosen nucleoside molecules. In this respect, it is very difficult for a purine moiety to be encapsulated by α-CD, whereas it can encapsulate the pyrimidine based nucleosides, which is in good agreement with the findings of this study. Moreover, β-CD has a relatively larger cavity, which can encapsulate both types of nucleosides and form stable inclusion complexes. The main driving force is the slightly apolar cyclodextrin cavity being occupied by water molecules,¹⁷ which is energetically unfavoured (polar–apolar interaction) and therefore can readily be substituted by the incoming hydrophobic residue of a nucleoside to attain an apolar–apolar association and decrease cyclodextrin ring strain, resulting in a more stable and lower energy state. The trapped water molecules are liberated in the bulk, increasing the entropy of the system (Scheme 2a and b).⁹ Solvation numbers (S_n), (Table 2) also show that after inclusion, the nucleosides form H-bonds with the –OH groups at the primary and secondary rims of cyclodextrin, which further stabilizes the inclusion complex. Thus, the efficiency of inclusion among different nucleoside-cyclodextrin systems on the basis of association constants may be illustrated in Scheme 3, wherein α-CD can accommodate only pyrimidine type nucleosides, but β-CD can accommodate all the four nucleosides under experimental conditions.

Scheme 3 The order of efficiency of inclusion among different nucleoside-cyclodextrin systems.

4. Conclusion

The present study reveals a unique behavior of the aqueous cyclodextrin-nucleoside system. It establishes the possibility of formation of host–guest inclusion complexes between cyclodextrin and RNA nucleosides by physicochemical as well as spectroscopic methods. Surface tension measurement and pH study support that α-cyclodextrin forms inclusion complexes with only pyrimidine based nucleosides, whereas β-cyclodextrin forms complexes with both purine and pyrimidine based nucleosides. In addition, the ratio of host : guest was found to be 1 : 1 by Job's method. The measured parameters, e.g., density, viscosity, acoustic data, and refractive index data, support the order of interaction among different nucleosides and cyclodextrin systems, whereas NMR data confirm the inclusion phenomenon. The determination of association constants and various thermodynamic parameters quantitatively explain the significance of the study. Therefore, this exclusive study has diverse applications in the broad field of biology and chemistry.

Acknowledgements

The authors are grateful to the Special Assistance Scheme, Department of Chemistry, NBU under the University Grants Commission, New Delhi (No. 540/27/DRS/2007, SAP-1) for financial sustenance and instrumental conveniences to carry on this study. Prof. M. N. Roy is also highly obliged to University Grants Commission, New Delhi, Government of India for being awarded one time Grant under Basic Scientific Research via the Grant-in-Aid No. F.4-10/2010 (BSR) concerning his dynamic service for augmenting of research facilities to expedite the advance research work.

References

1. M. L. Bender and M. Komiyama, Cyclodextrin Chemistry, Springer-Verlag, New York, 1978.
2. J. Szejtli, Inclusion of guest molecules, selectivity and molecular recognition by cyclodextrins, incomprehensive supramolecular chemistry, ed. J. Szejtli and T. Osa, Elsevier, Oxford, 1996, vol. 3, pp. 189–203.
3. E. M. M. D. Valle, Process Biochem., 2004, 39, 1033–1046.
4. B. G. Mathapa and V. N. Paunov, Phys. Chem. Chem. Phys., 2013, 15, 17903–17914.
5. K. A. Connors, Chem. Rev., 1997, 97, 1325–1357.
6. Y. Kang, K. Guo, B. J. Li and S. Zhang, Chem. Commun., 2014, 50, 11083–11092.
7. S. M. N. Simoes, A. R. Rico, A. Concheiro and C. A. Lorenzo, Chem. Commun., 2015, 51, 6275–6289.
8. Y. Gao, X. Zhao, B. Dong, L. Zheng, N. Li and S. Zhang, J. Phys. Chem. B, 2006, 110, 8576–8581.
9. J. Szejtli, Chem. Rev., 1998, 98, 1743–1753.
10. W. Saenger, Angew. Chem., Int. Ed., 1980, 19, 344–362.
11. P. C. Bevilacqua and J. SantaLucia Jr., ACS Chem. Biol., 2007, 2, 440–444.
12. T. Xiang and B. D. Anderson, Int. J. Appl. Pharm., 1990, 59, 45–55.
13. C. Formoso, Biochem. Biophys. Res. Commun., 1973, 50, 999–1005.
14. C. Formoso, Biopolymers, 1974, 13, 909–917.
15. D. P. Shoemaker and C. W. Garland, Experiments in Physical Chemistry, McGraw-Hill, New York, 1967, pp. 131–138.
16. D. Ekka and M. N. Roy, RSC Adv., 2014, 4, 19831–19845.
17. A. Pineiro, X. Banquy, S. Perez-Casas, E. Tovar, A. Garcia, A. Villa, A. Amigo, E. A. Mark and M. Costas, J. Phys. Chem. B, 2007, 111, 4383–4392.
18. M. N. Roy, D. Ekka, S. Saha and M. C. Roy, RSC Adv., 2014, 4, 42383–42390.
19. S. Saha, T. Ray, S. Basak and M. N. Roy, New J. Chem., 2016, 40, 651–661.
20. M. N. Roy, M. C. Roy and K. Roy, RSC Adv., 2015, 5, 56717–56723.
21. P. Job, Ann. Chim., 1928, 9, 113–203.
22. J. S. Renny, L. L. Tomasevich, E. H. Tallmadge and D. B. Collum, Angew. Chem., Int. Ed., 2013, 52, 11998–12013.
23. K. A. Connors and J. M. Lipari, J. Pharm. Sci., 1976, 65, 379–383.
24. T. Ogawa, M. Yasuda and K. Mizutani, Bull. Chem. Soc. Jpn., 1984, 57, 662–666.
25. D. Ekka and M. N. Roy, Amino Acids, 2013, 45, 755–777.
26. J. V. Caso, L. Russo, M. Palmieri, G. Malgieri, S. Galdiero, A. Falanga, C. Isernia and R. Iacovino, Amino Acids, 2015, 47, 2215–2227.
27. T. Wang, M. D. Wang, C. Ding and J. Fu, Chem. Commun., 2014, 50, 12469–12472.
28. K. P. Sambasevam, S. Mohamad, N. M. Sarih and N. A. Ismail, Int. J. Mol. Sci., 2013, 14, 3671–3682.
29. H. A. Benesi and J. H. Hildebrand, J. Chem. Soc., 1949, 71, 2703–2707.
30. F. Cramer, W. Saenger and H. C. Spatz, J. Am. Chem. Soc., 1967, 89, 14–20.
31. Y. Dotsikas, E. Kontopanou, C. Allagiannis and Y. L. Loukas, J. Pharm. Biomed. Anal., 2000, 23, 997–1003.
32. S. Das, K. G. Thomas, M. V. George and P. V. Kamat, J. Chem. Soc., Faraday Trans., 1992, 88, 3419–3422.
33. Y. He and X. Shen, J. Photochem. Photobiol., A, 2008, 197, 253–259.
34. M. A. Spies and R. L. Schowen, J. Am. Chem. Soc., 2002, 124, 14049–14053.

---

Footnote

† Electronic supplementary information (ESI) available: The theory, tables (Tables S1–S18) and figures (Fig. S1–S13) have been provided. See DOI: 10.1039/c5ra24102b

---