Thermodynamics and binding mode of novel structurally related 1,2,4-thiadiazole derivatives with native and modified cyclodextrins

Chemical Physics Letters 671 (2017) 28–36

Contents lists available at ScienceDirect

Chemical Physics Letters journal homepage: www.elsevier.com/locate/cplett

Research paper

Thermodynamics and binding mode of novel structurally related 1,2,4-thiadiazole derivatives with native and modified cyclodextrins Irina V. Terekhova a,⇑, Mikhail V. Chislov a, Maria A. Brusnikina a, Ekaterina S. Chibunova a, Tatyana V. Volkova a, Irina A. Zvereva b, Alexey N. Proshin c a b c

G.A. Krestov Institute of Solution Chemistry of Russian Academy of Sciences, 1 Akademicheskaya str., 153045 Ivanovo, Russia Saint Petersburg State University, 7/9 Universitetskaya nab., St. Petersburg 199034, Russia Institute of Physiologically Active Compounds of Russian Academy of Sciences, 1 Severniy pr., 142432 Chernogolovka, Russia

a r t i c l e

i n f o

Article history: Received 22 November 2016 Revised 4 January 2017 In final form 6 January 2017 Available online 7 January 2017 Keywords: 1,2,4-Thiadiazole derivatives Cyclodextrins Complex formation Thermodynamics Spectroscopy Solubility Alzheimer’s disease

a b s t r a c t Study of complex formation of cyclodextrins with 1,2,4-thiadiazole derivatives intended for Alzheimer’s disease treatment was carried out using 1H NMR, ITC and phase solubility methods. Structure of cyclodextrins and thiadiazoles affects the binding mode and thermodynamics of complexation. The larger cavity of b- and c-cyclodextrins is more appropriate for deeper insertion of 1,2,4-thiadiazole derivatives which is accompanied by intensive dehydration and solvent reorganization. Benzene ring of the thiadiazoles is located inside macrocyclic cavity while piperidine ring is placed outside the cavity and can form H-bonds with cyclodextrin exterior. Complexation with cyclodextrins induces the enhancement of aqueous solubility of 1,2,4-thiadiazole derivatives. Ó 2017 Elsevier B.V. All rights reserved.

1. Introduction Solubility of novel drug candidates in biological fluids is an important subject in pharmaceutical science. It is well known that majority of newly discovered pharmaceutically active compounds have low aqueous solubility and, consequently, poor bioavailability. This is the main reason causing the delay in preclinical testing and drug development. In this connection, selection of proper technologies and effective solubilizers could resolve this problem and rapidly increase the progress of pharmaceutical industry. Besides the technological advancements, there are numerous excipients available to overcome poor solubility and bioavailability of drugs. A wide range of high functional excipients for multifunction applications in solid and/or liquid dosage formulation has been used. Water-soluble polymers, oligosaccharides, surfactants, alcohols, etc are the most popular solubilizers at the time [1]. Selection of an effective excipient and its combination with drug is the subject of continued interest to design drugs with improved properties [1,2].

⇑ Corresponding author. E-mail address: [email protected] (I.V. Terekhova). http://dx.doi.org/10.1016/j.cplett.2017.01.010 0009-2614/Ó 2017 Elsevier B.V. All rights reserved.

Our attention was focused on novel 1,2,4-thiadiazole derivatives which were recently synthesized and proposed for the treatment of Alzheimer’s disease [3,4]. It has been found on the basis of the biological tests that these compounds influence the glutamateinduced calcium ions uptake into synaptosomes of rat brain cortex. In spite of satisfactory biological activity, 1,2,4-thiadiazole derivatives are poorly soluble in the aqueous media. Indeed, solubility of these compounds is not high and lies in the range of 10
3–10
4 M. In this work, cyclodextrins (CDs) are proposed as solubilizers for 1,2,4-thiadiazole derivatives. CDs are cyclic oligosaccharides consisting of 6–8 glucopyranose units linked by a-(1,4)-glycosidic bonds. Owing to the sufficient solubilizing capacity of CDs they are widely used in pharmacy for preparation of different dosage forms of poorly soluble compounds [1,2,5,6]. Solubilizing effect of CDs is based on their ability to inclusion complex formation. Hydrophobic cavity of CDs can accommodate drug molecule, and hydrophilic exterior of CDs provides the dissolution of inclusion complexes in water. Therefore, investigation of complex formation of CDs with drugs is of practical importance since it allows to obtain new water-soluble formulations. The purpose of our work was to reveal the binding affinity of native and modified CDs to inclusion complex formation with novel 1,2,4-thiadiazole derivatives, the structures of which are

29

I.V. Terekhova et al. / Chemical Physics Letters 671 (2017) 28–36

depicted in Fig. 1. These compounds differ by the nature of side group –R located in para-position of the benzene ring. Compounds I-III have halogens of different van der Waals volume, electronegativity and other characteristics. Compound IV has methyl side group. Moreover, CDs under consideration differ by the size of macrocyclic cavity and the structure of substituents (e.g. methyl, carboxymethyl, hydroxyethyl, etc.). Therefore, the influence of the reagents structure on the binding mode and thermodynamics of complex formation is analyzed in the present study. Furthermore, the solubilizing effect of CDs towards 1,2,4-thiadiazole derivatives caused by the inclusion complex formation is also examined.

3.33 (1H, m, CH2CHMe), 7.40 and 7.52 (4H, d and d, J = 8.8 Hz, Harom). 2-[5-(4-Chlorophenylamino-[1,2,4]thiadiazol-3-yl]-1-methylethyl}(2,2,6,6-tetramethylpiperidine-4-yl)-amine (III). White powder. Yield 70%. Calc., %: C 58.88, H 7.41, N 17.16 C20H30ClN5S. Found, %: C 58.80, H 7.40, N 17.34. 1H NMR [200 MHz] d: 0.86 (2H, m, C

2. Experimental section

8.52, N 18.09. 1H NMR [200 MHz] d: 0.71 (2H, m, C(3)HH, C(5)

(3)HH, C(5)HH), 1.14 and 1.20 (12H, s and s, 4xCH3), 1.27 (3H, d, J = 6.65 Hz, CH2CHCH3), 1.87 (2H, d, J = 12.5 Hz, C(3)HH, C(5)HH), 2.84 (2H, d, J = 6.5 Hz, CH2CHMe), 3.03 (1H, m, C(4)H), 3.45 (1H, m, CH2CHMe), 7.40 and 7.50 (4H, d and d, J = 8.4 Hz, Harom). [1-Methyl-2-(5-p-tolylamino-[1,2,4]thiadiazol-3-yl)-ethyl]-(2,2,6, 6-tetramethylpiperidine-4-yl)-amine (IV). Light crystals. Yield 69%. Calc., %: C 65.08, H 8.58, N 18.07. C21H33N5S. Found, %: C 65.01, H HH), 1.02, 1.05 (6H, s + s, 2xCH3), 1.07 (3H, d, J = 6.5 Hz,

2.1. Materials

CH2CHCH3), 1.12, 1.14 (6H, s + s, 2xCH3), 1.75 (2H, dd, J = 3.1,

a-CD (P98%), b-CD (P99%), hydroxypropyl-b-cyclodextrin

12.5 Hz, C(3)HH, C(5)HH), 2.73 (2H, dd, J = 2.3, 6.5 Hz, CH2CHMe),

(HP-b-CD, P98%), hydroxyethyl-b-cyclodextrin (HEt-b-CD, P98%), methyl-b-cyclodextrin (M-b-CD, P98%) and c-CD (P90%) were obtained from Sigma Aldrich and were used as received. Car boxymethyl-b-cyclodextrin sodium salt (CM-b-CD, P95%) was purchased from Cyclolab (Hungry). Average substitution degree per glucose unit was 0.6, 0.7, 1.6 and 0.5 for HP-b-CD, HEt-b-CD, M-b-CD and CM-b-CD, respectively. CDs were stable crystallohydrates, the water content in which determined by thermogravimetry (thermogravimetric analyzer TG Netzsch 209 F1 Libra) was equal to 1.2, 12.0, 2.0, 0.1, 0.2, 2.8 and 2.3% for a-CD, b-CD, HP-bCD, HEt-b-CD, M-b-CD, CM-b-CD and c-CD, respectively. 1,2,4-Thiadiazole derivatives were synthesized by standard and accessible methods in the Institute of Physiologically Active Compounds of the Russian Academy of Sciences (Chernogolovka) [7]. The purity of these compounds was P 98%. 2-[5-(4-Fluorophenylamino-[1,2,4]thiadiazol-3-yl]-1-methylethyl}(2,2,6,6-tetramethylpiperidine-4-yl)-amine (I). Light crystals. Yield 69%. Calc.,%: C 61.35, H 7.72, N 17.89 C20H30FN5S. Found, %: C 61.23, H 7.87, N 17.92. 1H NMR [200 MHz] d: 0.71 (2H, m, C(3)

2.91 (1H, m, C(4)H), 3.32 (1H, m, CH2CHMe), 7.09 and 7.38 (4H, d + d, J = 8.6 Hz, Harom), 10.56 (1H, br.s, ArNH). Phosphate buffer (pH 7.4) was prepared on the basis of KH2PO4 and Na2HPO412H2O of analytical grade (P99%). The pH of solutions was controlled by means of Mettler Toledo S220 SevenCompact pH-meter. Bidistilled water was used throughout the work.

HH, C(5)HH), 1.02, 1.04 (12H, s, 4xCH3), 1.07 (3H, d, J = 6.6 Hz,

Stability constants were derived from the concentration dependences of Dd using nonlinear least squares curve-fitting procedure.

CH2CHCH3), 1.72 (2H, dd, J = 3.0, 12.6 Hz, C(3)HH, C(5)HH), 2.72 (2H, dd, J = 2.3, 6.6 Hz, CH2CHMe), 2.91 (1H, m, C(4)H), 3.34 (1H, m, CH2CHMe), 7.04 (2H, t, J = 8.6 Hz, Harom), 7.58 (2H, dd, J = 4.7, 9.1 Hz, Harom). 2-[5-(4-Bromophenylamino-[1,2,4]thiadiazol-3-yl]-1-methylethyl}(2,2,6,6-tetramethylpiperidine-4-yl)-amine (II). Yellow powder. Yield 72%. Calc., %: C 53.09, H 6.68, N 15.48 C20H30BrN5S. Found, %: C 53.28, H 6.56, N 15.64. 1H NMR [200 MHz] d: 0.73 (2H, m, C (3)HH, C(5)HH), 1.02 and 1.05 (12H, c and c, 4xCH3), 1.08 (3H, d, J = 6.5 Hz, CH2CHCH3), 1.72 (2H, dd, J = 3.1, 12.5 Hz, C(3)HH, C(5) HH), 2.75 (2H, dd, J = 2.3, 6.5 Hz, CH2CHMe), 2.91 (1H, m, C(4)H),

2.2. 1H NMR spectroscopy The 1H NMR spectra were recorded with a Bruker-AV-500 spectrometer operating at 500 MHz and temperature of 298.15 K. Samples were prepared in D2O (99.9% isotopic purity). Cyclohexane was applied as an external reference. For evaluation of stability constants of the complexes, the 1H NMR spectra of 1,2,4-thiadiazole derivatives were recorded in the presence of variable concentrations of CDs. Chemical shift changes Dd were calculated as follows:

Dd ¼ dcomplexed
dfree

ð1Þ

2.3. Spectroscopic determination of stoichiometry of the complexes Stoichiometry of the complexes was determined using Job’s method [8]. Stock solutions of CD (610
5 mol/kg) and 1,2,4thiadiazole derivative (610
5 mol/kg) were prepared and mixed in such a way that the sum of total concentration of host and guest was constant. The mole fraction of 1,2,4-thiadiazole derivative   R ¼ X X thiadiazole was varied from 0.1 to 1.0. The UV–vis spectra þX CD thiadiazole

of solutions were recorded on UV–vis spectrometer (Shimadzu 1800, Japan). The corrected absorbance ðDA  X thiadiazole ) at fixed wavelength was plotted against R. 2.4. Isothermal titration calorimetry

Fig. 1. 1,2,4-Thiadiazole derivatives under study.

Thermodynamic parameters of complex formation were determined using an isothermal titration calorimeter TA Nano (TA Instruments, New Castle, USA). Calorimetric experiments were carried out in phosphate buffer (pH 7.4) at 298.15 K. The syringe was filled by 20 mM of CD solution or 15 mM in case of b-CD and the measurement cell contained 1 ml of the solution of 1,2,4thiadiazole derivative (0.2 mM). All solutions were degassed for 10 min before the titration experiments. The 9.99 lL of CD solution were injected stepwise. The stirring speed was set to 300 rpm, the equilibration period between the injections was 1400 s. The results

30

I.V. Terekhova et al. / Chemical Physics Letters 671 (2017) 28–36

of titration experiment were corrected by subtraction of the heats of dilution. Data analysis was performed by NanoAnalyse software program (TA Instruments, USA) using One-Site Binding model. 2.5. Phase solubility study Phase solubility study was carried out according to method proposed by Higuchi and Connors [9]. An excess amounts of 1,2,4thiadiazole derivatives were dissolved in buffer solution containing increasing concentrations of CDs (0–0.01 mol/kg). After equilibration for 72 h solutions were centrifuged (Biofuge pico, Thermo Electron LED GmbH, Germany) at 2000 rpm for 20 min at 298.15 K. Concentration of thiadiazole in solution was determined spectrophotometrically (Shimadzu 1800, Japan). The experiments were carried out in triplicate and average solubility value was calculated. 3. Results and discussion Complex formation of CDs with 1,2,4-thiadiazole derivatives (guest) is described by the following equation:

CD þ guest ¼ CD  guest

ð2Þ

at R = 0.5 indicating the 1:1 stoichiometric ratio, which was also confirmed during the nonlinear regression fit of the data points obtained by the titration calorimetry. The best-fit parameters were derived when a one-site binding model was applied in the fitting procedure (Fig. 3).

ΔΑR

ΔΑR

The 1:1 stoichiometry of the complexes formed was revealed by Job’s method [8]. As an example, Job’s plots for complex formation of CDs with II and IV are shown in Fig. 2. For all systems under study, the dependences were symmetric and have an extremum

Fig. 3. Isothermal titration calorimetry results for complex formation of IV with HP-b-CD in phosphate buffer (pH 7.4) at 298.15 K. One-site binding model gives the best fit of the heat of binding curve (Q). The initial concentrations were 20 mM for CD and 0.2 mM for 1,2,4-thiadiazole derivative.

IV + β -CD

-0.12

IV + γ-CD

-0.15

-0.14 -0.20

-0.16 -0.25

-0.18

-0.30

-0.20 0.4

R

0.6

0.8

0.2

ΔΑR

ΔΑR

0.2

II + α-CD -0.16

0.4

0.6

R

0.8

II + HP-β -CD

-0.16

-0.20

-0.20

-0.24

-0.24 -0.28

0.2

0.4

R

0.6

0.8

-0.28 0.2

0.4

R

0.6

Fig. 2. Job’s plots for complex formation of 1,2,4-thiadiazole derivatives (II and IV) with CDs in phosphate buffer (pH 7.4) at 298.15 K.

0.8

31

I.V. Terekhova et al. / Chemical Physics Letters 671 (2017) 28–36

H-3 H-5

II/α-CD H-3 H-5

I/α-CD

H-6 H-3

H-2

H-4

H-5

4.00

3.90

3.80

3.70

3.60

3.50

α-CD

H-3 H-5

I/β-CD H-3 H-5

β-CD 3.90

3.80

3.70

3.60

3.50

H-3 H-5

I/γ-CD

H-6

H-3 H-2

H-5

H-4

γ-CD 3.90

3.80

3.70

3.60

3.50

Fig. 4. Partial 1H NMR spectra of CDs (0.003 mol/kg) with and without 1,2,4-thiadiazole derivatives (0.003 mol/kg) in D2O (pD 7.4) at 298.15 K.

H18',H20'

I+α-CD 0.09

H17',H21'

Δδ (ppm)

I.V. Terekhova et al. / Chemical Physics Letters 671 (2017) 28–36

Δδ (ppm)

32

II+α-CD

H17',H21'

0.3

0.06

0.2

0.03

0.1

H18',H20'

0.00

Δδ (ppm)

0.00

0.01

0.02

0.03

0.04

mα-CD (mol/kg)

I+β-CD H6' H17',H21' H18',H20' H22' H23'-H26'

0.05

0.00

H10'

-0.05

0.0

H9'

0.000

0.005

0.010

0.015

0.000

H9' 0.003

0.006

0.009

0.012

mβ-CD (mol/kg)

0.020

H22' H6' H9' H14' H23'-H26' H10'

I+γ-CD

0.00

H18',H20'

-0.03

H14' -0.10

H14' H23'-H26' H6' H22' H10'

mα-CD (mol/kg)

Δδ (ppm)

H23'-H26' H14' H22' H10' H9'

H17',H21'

-0.06 0.00

0.01

0.02

0.03

mγ-CD (mol/kg)

Fig. 5. Dependences of chemical shift changes of protons of I (0.003 mol/kg) on CD concentration in D2O (pD 7.4) at 298.15 K. (signal of H70 thiadiazole proton was overlapped with the signals CD protons).

Thermodynamic parameters of 1:1 complex formation of CDs with 1,2,4-thiadiazole derivatives are listed in Table 1. The values of thermodynamic parameters are determined by the contributions from dehydration and various kinds of noncovalent interactions that may take place in complexation process [10,11]. These are van der Waals forces and hydrogen bonding (DcH < 0 and TDcS < 0) as well as the hydrophobic interactions (DcH > 0 and TDcS > 0). As follows from data presented in Table 1, complex formation of a-CD with the majority of the guests under study is weaker in comparison with b-CD and c-CD. Binding of a-CD with 1,2,4thiadiazole derivatives is characterized by the negative enthalpy and entropy changes, whereas negative DcH and positive TDcS were obtained for binding with b-CD and c-CD. Thus, the cavity size affects the thermodynamics of complex formation. The influence of cavity dimensions on complex formation should be discussed taking into account the binding modes. The binding mode of CDs with 1,2,4-thiadiazole derivatives was derived from 1H NMR experiments. Fig. 4 shows the fragments of 1 H NMR spectra of CDs recorded in the presence of some 1,2,4thiadiazoles under study (I and II). In the CD structure, protons H-1, H-2, H-4 and H-6 are placed on the external surface of CD molecule and protons H-3 and H-5 are located inside the macrocyclic cavity. Namely, H-3 and H-5 protons are sensitive to the inclusion of the guest molecule. Additionally, the 1H NMR spectra of thiadiazoles were recorded in the presence of variable amounts of various CDs. Dependences of the chemical shift changes of thiadiazole protons on CD concentration are depicted in Fig. 5. As follows from the 1H NMR spectra of a-CD depicted in Fig. 4, the upfield shifts of the signals of H-3 and H-5 protons observed in the presence of I are not high. Fig. 5 shows the dependences of

chemical shift changes of I protons on a-CD concentration. As one can see, the signals of the protons belonging to the benzene ring (H170 , H180 , H200 and H210 ) are downfield shifted upon complex formation of I with a-CD, while the Dd values for all other protons (H60 , H90 , H100 , H140 , H220 , H230 -H260 ) are insignificant (Fig. 5). Taking into account the low stability constant for I/a-CD complexes (Table 1) one can assume that inclusion of I into macrocyclic cavity is absent and complex formation is governed by the surface interactions. To clarify the binding mode of I with a-CD the 2D ROESY analysis was performed (Fig. 6). The strong crosspeaks are observed between the protons of I benzene ring (H170 , H180 , H200 and H210 ) and H-3 protons of a-CD. At the same time, cross-peaks between H-5 protons of a-CD and H180 and H200 protons of I are visible. According to these data, benzene ring inters the a-CD cavity. Owing to the accommodation of aromatic guest inside the cavity the upfield shifts of CD protons (Fig. 4) and downfield shifts of the guest protons are observed (Fig. 5). These shift changes are attributed to the ring current effect of the aromatic fragment of guest molecule in the macrocyclic cavity [12]. It should be pointed out that all other protons of I (H60 , H90 , H100 , H140 , H220 , H230 -H260 ) have cross-peak only with the a-CD protons located outside the cavity. This indicates the possibility of interactions of thiadiazole and piperidine rings of I with an outer a-CD surface formed by OH-groups. It is interesting to examine the complexation in the system II +a-CD where more stable complexes are detected (Table 1). Compared with I, more noticeable shifting of the signals of the protons of benzene ring of compound II was observed in the presence of a-CD (Fig. 5). In addition, the shifts of the signals of H-3 and H-5 protons of a-CD induced by complex formation with II are also significant (Fig. 4). It should be stressed that the signals of H-3

33

I.V. Terekhova et al. / Chemical Physics Letters 671 (2017) 28–36 Table 1 Thermodynamic parameters of 1:1 complex formation of 1,2,4-thiadiazole derivatives with cyclodextrins in phosphate buffer at 298.15 K.a Complex

K

Dc G

I/a-CD

w.b.b 21.1 ± 0.7c 427 ± 24 624 ± 27c 855 ± 190 746 ± 29d 327 ± 128 217 ± 69 527 ± 34 81 ± 30 132 ± 3c 659 ± 25 732 ± 27c 855 ± 17 966 ± 25 826 ± 47d 505 ± 13 1493 ± 20 2500 ± 14 616 ± 7 146 ± 2 559 ± 8 827 ± 50 737 ± 40d 340 ± 20 506 ± 20 1411 ± 10 521 ± 23 w.b. 458 ± 19 725 ± 22 668 ± 38d 476 ± 29 –e 336 ± 16 734 ± 129

w.b.
7.6 ± 0.3c
15.0 ± 0.8
16.0 ± 0.7
16.7 ± 3.7
16.4 ± 0.7d
14.4 ± 5.6
13.3 ± 4.2
15.5 ± 1.0
10.9 ± 4.0
12.1 ± 0.3c
16.1 ± 0.6
16.3 ± 0.6c
16.7 ± 0.3
17.0 ± 0.4
16.6 ± 0.9d
15.4 ± 0.4
18.1 ± 0.2
19.4 ± 0.1
15.9 ± 0.2
12.4 ± 0.2
15.9 ± 0.2
16.7 ± 1.0
16.4 ± 0.9d
14.5 ± 0.9
15.4 ± 0.6
18.0 ± 0.1
15.5 ± 0.7 w.b.
15.2 ± 0.6
16.3 ± 0.5
16.1 ± 0.6d
15.3 ± 0.9 –e
14.4 ± 0.7
16.4 ± 2.9

Dc H

TDcS

w.b. –
3.6 ± 0.3

0.7 ± 0.05

1.1 ± 0.5 1.5 ± 0.1
10.5 ± 2.9
0.8 ± 0.4

17.0 ± 0.2

11.8 ± 0.1
3.9 ± 0.1

4.9 ± 0.1 0.2 ± 0.1
16.2 ± 0.1
11.0 ± 0.1
17.0 ± 0.2
8.2 ± 0.1
1.7 ± 0.1

2.8 ± 0.1 2.2 ± 0.1
12.5 ± 0.1
2.9 ± 0.3 w.b.
11.2 ± 0.2
4.1 ± 0.1

5.5 ± 0.2 –e
7.6 ± 0.2
2.2 ± 0.15

w.b. – 11.4 ± 0.9
16.0 ± 3.7
13.3 ± 5.6 11.8 ± 4.2 5.0 ± 3.1 10.1 ± 4.0

0.9 ± 0.6
4.9 ± 0.3 13.1 ± 0.4
10.5 ± 0.4 18.3 ± 0.2 3.2 ± 0.1 4.9 ± 0.2
4.6 ± 0.3 7.7 ± 0.2 15.0 ± 0.1
11.7 ± 0.9 13.2 ± 0.6 5.5 ± 0.1 13.0 ± 0.7 w.b. 4.0 ± 0.6 12.2 ± 0.5
9.8 ± 0.9 –e 6.8 ± 0.7 14.2 ± 2.9

(kJ/mol)

I/b-CD I/HP-b-CD I/HEt-b-CD I/M-b-CD I/CM-b-CD I/c-CD II/a-CD II/b-CD II/HP-b-CD II/HEt-b-CD II/M-b-CD II/CM-b-CD II/c-CD III/a-CD III/b-CD III/HP-b-CD III/HEt-b-CD III/M-b-CD III/CM-b-CD III/c-CD IV/a-CD IV/b-CD IV/HP-b-CD IV/HEt-b-CD IV/M-b-CD IV/CM-b-CD IV/c-CD a b c d e

Standard uncertainty u is: u(T) = 0.01 K; standard deviation with a 95% confidence interval is given in parentheses. Weak binding (binding isotherms were nearly linear and calculation of K was not possible). Obtained from 1H NMR measurements. Obtained from phase solubility study. Calculation of K and DcH was not possible due to zero heat effects.

protons are upfield shifted, while the downfield shift of signals of H-5 protons was displayed. The same trend of chemical shift changes of H-3 and H-5 protons was demonstrated for complex formation of a-CD with benzoic acid [13] and substituted benzoic acid [14,15]. As it has been detected in our previous works [13,14], ACOOH group of the aromatic carboxylic acids is placed inside the a-CD cavity near to H-5 protons, whereas a part of the benzene ring is positioned near to H-3 protons. Relying on these data we can conclude that benzene ring with Br-substituent is inserted into a-CD cavity and atom Br is in close contact with H-5 protons. Compared to –Cl and –F, the –Br substituent having the largest van der Waals volume displays higher binding affinity to a-CD cavity [16–18]. Therefore, its location inside the a-CD cavity is favorable and results in formation of stable complexes. In the 1H NMR spectra of b-CD, the noticeable upfield shifts of the signals of H-3 and H-5 protons were observed in the presence of I (Fig. 4). Usually, the upfield shifts of the inner CD protons are explained by an anisotropic effect of the aromatic ring entering the cavity [19]. Thus, this result suggests the deep inclusion of 1,2,4-thiadiazole and location of the aromatic ring of the guest molecule inside b-CD cavity. The intermolecular cross-peaks observed between the inner b-CD protons and protons of benzene ring (H170 , H180 , H200 and H210 ) of I molecule (Fig. 6) evidence the proposed binding mode. It should be noted that all other protons of I do not have cross-peaks with H-3 and H-5 protons of b-CD (Fig. 6)

and, therefore, they do not penetrate the cavity. Their possible interactions with b-CD outer surface are confirmed by measurable upfield shifts for the protons H90 , H100 and H140 (Fig. 5). In case of c-CD, the chemical shift changes of the signals of inner protons H-3 and H-5 are less pronounced in comparison with those values for b-CD protons (Fig. 4). Fig. 5 also shows that the magnitudes of chemical shift changes of I protons are not high in the presence of c-CD. It is attributed to lower binding affinity of 1,2,4-thiadiazole derivative to the c-CD cavity (Table 1), which has the largest size of macrocyclic cavity. Probably, van der Waals interactions are weaker and thiadiazole poorly retained in the c-CD cavity due to the absence of close contact between the guest and host molecules. Nevertheless, the appreciable upfield shifts of the signals of protons H170 , H180 , H200 and H210 indicate the location of I benzene ring inside c-CD cavity (Fig. 5). Cross-peaks detected between inner protons of c-CD and protons of benzene ring of I (Fig. 6) prove this binding mode. Thus, 1H NMR results point out that inclusion of benzene ring into CD cavity is more preferable in comparison with piperidine ring. Similar results were obtained by Maheshwari et al. [20] during study on complex formation of roxatidine acetate hydrochloride with b-CD. Madi et al. [21] investigated the complexation of N-nitroso-N0 -(2-chloroethyl)-N0 -sulfamide piperidine with b-CD using standard B3LYP and MPW1PW91 Density Functional Theory calculations as well as mass and 15N NMR spectroscopy methods. Two orientations of the guest molecule (tail – piperidine ring is

34

I.V. Terekhova et al. / Chemical Physics Letters 671 (2017) 28–36

Fig. 6. Expanded regions of the ROESY spectrum of I/CD complexes (a – a-CD; b – b-CD; c – c-CD).

solubility (mol/kg)

I.V. Terekhova et al. / Chemical Physics Letters 671 (2017) 28–36

I 0.020

IV

0.015

III II

0.010

0.005

0.000

0.003

0.006

0.009

0.012

mHP-β-CD (mol/kg) Fig. 7. Solubility diagrams of 1,2,4-thiadiazoles in the presence of HP-b-CD (phosphate buffer pH 7.4, T = 298.15 K).

inside cavity; head – piperidine ring is outside cavity) were considered. According to the theoretical and experimental results, tail orientation was found to be more favorable than the head one. It means that peperidine ring of the guest is not included into macrocyclic cavity. Taking into account the binding modes proposed on the basis of 1 H NMR results we can discuss in details the thermodynamic parameters of complex formation of CDs with 1,2,4-thiadiazole derivatives. As can be seen from Table 1, highly exothermic complex formation of thiadiazoles II and III with a-CD is accompanied by the negative entropy changes. The enthalpy-driven binding with a-CD is governed by van der Waals interactions and possible hydrogen bonding which can be realized between the polar groups of the guest molecule and hydroxyls surrounding the host cavity. Deep penetration of 1,2,4-thiadiazole derivatives into cavity of b-CD and c-CD is characterized by small negative DcH (except complexation with CM-b-CD) and positive TDcS. The enthalpyentropy favorable binding of thiadiazoles with b-CD and c-CD suggests that hydrophobic interactions and dehydration play the major role in complex formation. Comparative analysis of thermodynamic parameters of complex formation of 1,2,4-thiadiazole derivatives with parent and modified b-CDs shows the effect of substituents surrounding the macrocyclic cavity. As can be seen from Table 1, substitution of AOH by hydroxypropyl-groups favors complex formation. Complex formation with HP-b-CD is characterized by higher DcH and TDcS values compared with those for b-CD complexation. This can be caused by intensive dehydration of bulky hydroxyropyl-groups and their probable hydrophobic interactions with thiadiazoles. Increase of DcH and TDcS with the rise of CD hydrophobicity was revealed. In particular, binding of 1,2,4-thiadiazole derivatives with methylated b-CD, which is most hydrophobic among all CDs studied, is mainly entropy driven. Enthalpy changes obtained for M-b-CD complexation are small negative or positive (Table 1). This fact confirms the significant role of hydrophobic interactions in complex formation of b-CDs having nonpolar substituents. On the contrary, presence of polar substituents such as carboxymethyl-groups results in significant increase of exothermicity of complex formation. As follows from Table 1, binding of 1,2,4-thiadiazole derivatives with CM-b-CD is characterized by high negative values of DcH caused by participation of ionized carboxymethyl-groups in hydrogen bonding and electrostatic interactions with a part of thiadiazole molecule which is placed outside the macrocyclic cavity. However, positive values of TDcS indicate that insertion of nonpolar part of thiadiazole molecule

35

into CD cavity is accompanied by intensive dehydration and hydrophobic interactions. Thus, inclusion of the aromatic ring as well as the surface interactions between carboxymethyl-groups and the remaining part of I molecule positioned outside the cavity determine the thermodynamics of CM-b-CD binding with 1,2,4thiadiazole derivatives under study. It is interesting to analyze the influence of the guest structure on complex formation process. Inspection of the data given in Table 1 shows that the structure of substituent disposed in the para-position of the benzene ring (Fig. 1) affects the thermodynamics of complex formation. More stable inclusion complexes are formed with compound II having –Br substituent. On the contrary, the binding of CDs with compound I having –F side group is weakest. Selectivity of binding of CDs with halogen ions has been documented. It has been shown that in comparison with Cl
and F
the anions Br
form more stable inclusion complexes with CDs [16–18]. Weak binding affinity of CDs to F
and Cl
can be caused by small size and high electronegativity of these anions as well as their hydration in aqueous solution. It is well known that among halogens fluorine has a small size and high electron density. Therefore, it attracts more water molecules and becomes strongly hydrated. It is evident from the values of the enthalpy of hydration of halogen-ions reported by Smith [22]: DhydrH(F
) =
515 kJ/mol, DhydrH(Br
) =
347 kJ/mol and DhydrH(Cl
) =
381 kJ/mol. Thus, high electronegativity and hydration prevent the complexation and result in low exothermicity of binding of I with CDs. As one can see from Table 1, complex formation of CDs with I is characterized by small negative DcH values. On the contrary, complex formation of CDs with II is more exothermic compared with I, III and IV (Table 1). The –Br substituent is bulky and less hydrated [22], therefore, it is included into CD cavity and retained by van der Waals forces. Thus, CD complex formation is more favorable with less hydrated thiadiazoles. Inclusion complex formation with CD is mainly used in pharmacy to increase the aqueous solubility of poorly soluble drugs [1,2,23]. In this connection, we examined the solubilizing action of HP-b-CD towards the 1,2,4-thiadiazoles under study. Solubility diagrams are shown in Fig. 7. As one can see, solubility of I, II, III and IV linearly increases 2, 3, 3 and 2-fold, respectively, at HP-bCD concentration equal to 0.01 mol/kg. The linear solubility diagrams correspond to AL-type, which defines the formation of water soluble complexes stability constants of which were calculated from the slope of solubility diagrams [9]:

K¼

slope S0 ð1
slopeÞ

ð3Þ

where S0 is solubility of 1,2,4-thiadiazole derivative in buffer without CD. Values of stability constants are reported in Table 1. A good agreement between K obtained by different experimental methods was observed. As one can see, formation of more stable inclusion complexes results in more considerable increase of aqueous solubility of thiadiazoles under study. 4. Conclusions Complex formation of different CDs with 1,2,4-thiadiazole derivatives in phosphate buffer (pH 7.4) at 298.15 K was studied by isothermal titration calorimetry, 1H NMR and phase solubility methods. Calorimetric study provided a set of thermodynamic data for complex formation of native and modified CDs with 1,2,4thiadiazole derivatives under consideration. It was found that more stable inclusion complexes are formed with b-CD and c-CD and they are typically enthalpy-entropy stabilized. The formation of the inclusion complexes of b-CD and c-CD is governed by hydrophobic interactions and release of water molecules from

36

I.V. Terekhova et al. / Chemical Physics Letters 671 (2017) 28–36

hydration shells of the solutes and CD cavity. Contrariwise, van der Walls interactions and hydrogen bonding determine the negative values of DcH and TDcS obtained for complex formation of 1,2,4thiadiazole derivatives with a-CD. Thermodynamic parameters of complex formation are sensitive to the electronegativity, size and hydration state of side group located in para-position of the benzene ring of the guest molecule. More favorable binding takes place with thiadiazole having –Br substituent which is less hydrated in aqueous solution. According to the proposed binding mode, benzene ring of the guest molecule is included into host cavity while piperidine ring is placed outside and can form H-bonds with hydroxyls or other polar substituents of the CD external. As a result, enhancement of aqueous solubility of 1,2,4-thiadiazole derivatives in CD solutions is observed up to CD concentration of 0.01 mol/kg. Acknowledgements This work was supported by Russian Science Foundation (grant № 15-13-10017). Authors thank the ‘The upper Volga region Centre of physicochemical research’ (Ivanovo, Russia) and St. Petersburg State University Center ‘Thermogravimetric and Calorimetric Research’ for the provision of scientific equipment. References [1] A. Katdare, M.V. Chabal (Eds.), Excipient Development for Pharmaceutical, Biotechnology, and Drug Delivery Systems, CRC Press, New York, 2006.

[2] D. Douroumis, A. Fahr (Eds.), Drug delivery Strategies for Poorly Water-soluble Drugs, J. Wiley & Sons, 2012. [3] A.N. Proshin, L.N. Petrova, S.O. Bachurin, Bull. Exp. Biol. Med. 142 (2006) 43– 46. [4] A. Castro, T. Castaño, A. Encinas, W. Porcal, C. Gil, Bioorg. Med. Chem. 14 (2006) 1644–1652. [5] G. Tiwari, R. Tiwari, A.K. Rai, J. Pharm. Bioallied. Sci. 2 (2010) 72–79. [6] A. Vyas, S. Saraf, S. Saraf, J. Incl. Phenom. Macrocycl. Chem. 62 (2008) 23–42. [7] A.N. Proshin, I.V. Serkov, L.N. Petrova, S.O. Bachurin, Russ. Chem. Bull. 63 (2014) 1148–1152. [8] P. Job, Ann. Chim. Appl. 9 (1928) 113–203. [9] T. Higuchi, K.A Connons, in: C.N. Reilly (Ed.), Advances in Analytical Chemistry and Instrumentation, 4, Wiley-Interscience, New York, 1965, pp. 117–212. [10] L. Liu, Q.-X. Guo, J. Incl. Phenom. Macrocycl. Chem. 42 (2002) 1–14. [11] K.A. Connors, Chem. Rev. 97 (1997) 1325–1357. [12] S.M. Ali, S. Shamim, JEAS 4 (2014) 63–70. [13] I.V. Terekhova, M. Koz´biał, R.S. Kumeev, P. Gierycz, Chem. Phys. Lett. 514 (2011) 341–346. [14] I.V. Terekhova, R.S. Kumeev, G.A. Alper, J. Incl. Phenom. Macrocycl. Chem. 59 (2007) 301–306. [15] S. Simova, H.-J. Schneider, Chem. Soc. Perkin Trans. 2 (2000) 1717–1722. [16] I.V. Terekhova, E.S. Chibunova, R.S. Kumeev, G.A. Alper, Carbohydr. Polym. 110 (2014) 472–479. [17] I.V. Terekhova, E.S. Chibunova, R.S. Kumeev, G.A. Alper, Chem. Phys. Lett. 557 (2013) 134–139. [18] Y. Matsiu, M. Ono, Sh. Tokunaga, Bull. Chem. Soc. Jpn. 70 (1997) 535–541. [19] P.V. Demarco, A.L. Thakkar, J. Chem. Soc. D 1 (1970) 2–4. [20] A. Maheshwari, M. Sharma, D. Sharma, Carbohydr. Res. 346 (2011) 1809–1813. [21] F. Madi, D. Khatmi, N. Dhaoui, A. Bouzitouna, M. Abdaoui, A. Boucekkine, C. R. Chimie 12 (2009) 1305–1312. [22] D.W. Smith, J. Chem. Edu. 54 (1977) 540–542. [23] T. Loftsson, M.E. Brewster, J. Pharm. Sci. 101 (2012) 3019–3332.