New Insights into the Structure and Reactivity of Uracil Derivatives in Different Solvents-A Computational Study.

Ab initio calculations were carried out to understand the reactivity and stability of some uracil derivatives, cytosine, 1-methyl cytosine, and cytidine in solvents, water, dimethyl sulfoxide (DMSO), n-octanol, and chloroform. Geometries were fully optimized at MP2 and B3LYP using the 6-31+G(d,p) basis set by applying the Solvation Model on Density (SMD) in solvent systems. The syn conformer of cytidine (cytidine II) is the most stable conformer in the gas phase, while the anticonformer (cytidine IV) is most stable in all of the solvents. Solvation free energy and polarizability values in different solvents decrease in the order water > DMSO > n-octanol > chloroform, while dipole moment, first-order hyperpolarizability, and HOMO-LUMO energy gap values follow the order of polar protic solvent (water and n-octanol) > polar aprotic solvent (DMSO) > nonpolar solvent (chloroform). The solvation free energy, dipole moment, polarizability, and first-order hyperpolarizability values also follow the order of cytosine > 1-methyl cytosine > cytidine. To illustrate that the molecular properties correlate well with the reactivity of the molecules, ab initio calculations were carried out for the reaction of uracil derivatives with Br2 in the gas phase, water, DMSO, n-octanol, and chloroform. All ground and transition state geometries were fully optimized at B3LYP/6-31+G(d,p), and energies were also calculated at G3MP2 for cytosine and 1-methyl cytosine. For cytosine and 1-methyl cytosine, Gibbs energies of activation decrease with the polarity of the solvent that is chloroform > n-octanol > DMSO > water, while the Gibbs energies of activation for the reaction with cytidine decrease in the order of water > DMSO > n-octanol > chloroform. These results suggest that solvent polarity is very important for the stability and reactivity of uracil derivatives. Hydrogen bonding may also play an important role mainly for cytidine. Free energies of activation decrease with the size of the molecule, i.e., cytosine > 1-methyl cytosine > cytidine.

Introduction

Uracil derivatives such as the cytosine, 1-methyl cytosine, and cytidine are very important molecules as they are building blocks of deoxyribonucleic acid (DNA) and ribonucleic acid (RNA). They are pyrimidine derivatives containing a heterocyclic aromatic ring and two substituents consisting of an amine group and a keto group, as shown in Figure . The simplest of these uracil derivatives is cytosine, one of the main bases found in DNA and RNA. The other three bases are adenine, guanine, and thymine in DNA and adenine, guanine, and uracil in RNA. Cytosine forms three hydrogen bonds with guanine forming the DNA double helix. Cytosine plays an important role in the evolution of mammalian isochors[1] and DNA repairing.[2] 1-Methyl cytosine is the methylated form of cytosine and it is one of the essential components of the Hachimoji DNA,[3] which is recently synthesized in vitro. Hachimoji DNA consists of four synthetic nucleotide bases, where one of those bases, 1-methyl cytosine, binds with another base isoguanine. Since these nucleotide bases are more hydrophobic than natural bases, they always form a standard double helix.[4] Cytidine, which is an integral part of RNA, is formed when cytosine is attached to a five-membered ring ribofuranose via a glycosidic bond. This ribose ring contains a 2′-hydroxyl group, which is susceptible to hydrolysis causing the RNA to be single-stranded.

Figure 1

Molecular models of (a) cytosine, (b) 1-methyl cytosine, and two conformers of cytidine, (c) cytidine II (gas phase) and (d) cytidine IV (in solvents). Atom numbers are also shown on the molecules.

The stability of double-stranded DNA greatly depends on its ambient conditions.[5−8] Cui et al. studied[5] the change in DNA structure when dragged from water to solvents such as octane, diethylbenzene, and 1-propanol using atomic force microscope (AFM)-based single-molecule force spectroscopy (SMFS) as well as molecular dynamics (MD) simulations. They have found that the double-stranded DNA dissociates into single strands, and it is caused due to the low polarity of the solvents. On the contrary, DNA–lipid complex is confirmed to form double-helical strands even in organic solvents such as chloroform and ethanol.[8] This indicates that a double-stranded DNA can maintain its structure in the low polarity of solvents if the structure of the DNA were modified. We are interested to understand if the chemical modification of DNA bases could retain the double-helical strands in different types of solvents such as the nonpolar solvents, polar aprotic solvents, and polar protic solvents. In this regard, it is important to understand first how the structure and interaction of different natural and modified DNA bases change in different solvent environments.

Cytosine can also exist as a cofactor to enzymes where it is involved in the transfer of phosphate to convert adenosine diphosphate (ADP) to adenosine triphosphate (ATP), which provides energy to drive many processes in living cells. This reaction occurs in the physiological condition in the presence of water. Recently, cytosine is also synthesized by NASA scientists in spacelike laboratory conditions, i.e., in vacuum from pyrimidine.[9] Although cytosine is not available in meteorites, pyrimidine is found in meteorites suggesting cytosine may have formed from pyrimidine but it most likely reacted in vacuum to produce uracil via a deamination reaction. This indicates that the stability and reactivity of cytosine, 1-methyl cytosine, and cytidine are most likely affected by the change of phase and solvent. In fact, the solubility of cytosine is very low in nonpolar solvents, but it is soluble in a polar protic solvent such as the water.[10]

DNA mutations happen due to environmental conditions as well as chemical reactions where the nucleic acid bases react with various chemical compounds.[11−13] DNA mutations may cause fatal human diseases such as blindness, deafness, dementia, movement disorders, cardiac failure, diabetes, renal dysfunction, and liver disease. Depurination and deamination are the most frequent spontaneous chemical reactions that are known to develop serious DNA damages in cells. Depurination can release guanine, as well as adenine, from DNA. The major type of deamination reaction converts cytosine to an altered DNA base, uracil, due to reaction with HNO2[14] and NO.[15] Aqueous solutions of cytosine have also shown to undergo interesting oxo ⇌ hydroxy tautomerism.[10,16−19] Cytosine may undergo tautomerization more easily than adenine, uracil, or thymine leading to transition-type point mutations in DNA, which may cause diseases such as cancer, neurofibromatosis, sickle-cell anemia, Tay–Sachs disease, and color blindness.[20] Different cytosine tautomers are also favored in different solvents and gas phase.[21] Cytosine tautomer that is most stable in polar solvent has also found to have the larger dipole moment.[22] Acidic and proton affinity of cytosine and 1-methyl cytosine have been studied in the gas phase and in a nonpolar environment using both the ab initio and experimental bracketing and Cooks kinetic methods.[23] The proton affinities (PA = ΔH) of cytosine and 1-methyl cytosine were observed 954 ± 13 and 962 ± 13 kJ mol–1, respectively, suggesting 1-methyl cytosine is stronger base compared to cytosine. However, cytosine is the most alkaline in the aqueous solution (pKa = 4.6) among the primary nucleic acid bases adenine, cytosine, guanine, thymine, and uracil, which plays an important role in many biochemical processes.[24]

Cytosine, its nucleosides and nucleotides, and many of its derivatives have been extensively studied experimentally,[19,22,25−28] as well as computationally, in both the gas phase[19,29−31] and the aqueous phase.[29,32−35] Fewer studies[21,22,34] have been reported to understand the chemical reactivity of uracil derivatives in nonpolar and polar aprotic solvents. In this study, we have undertaken a comprehensive study about the effects of different solvents on molecular properties such as geometry, solvation free energy, dipole moment, molecular electrostatic potential (MESP), charge distribution, polarizability, hyperpolarizability, highest occupied molecular orbital (HOMO), and lowest occupied molecular orbital (LOMO) energies for cytosine, 1-methyl cytosine, and cytidine in different solvent systems such as chloroform, n-octanol, DMSO, and water using ab initio methods with an implicit solvent model, Solvation Model on Density (SMD). Ab initio calculations could provide greater insights into the molecular characteristics and interactions of these molecules in various solvents.[36] Most of the computational calculations have been performed with Møller–Plesset (MP) perturbation theory, MP2, and density functional theory (DFT) using the hybrid meta functional, B3LYP, which are able to describe both stacking and hydrogen-bonding interactions that are present in these molecules. Finally, bromination reactions of cytosine, 1-methyl cytosine, and cytidine have been investigated to understand the reactivity of these molecules in different solvents. Br2 is a nonpolar molecule. However, Br–Br bond breaks during the reaction leading to the formation of charged species involving atomic Br. In fact, the bromination reaction of ethylene forms an ion pair of bromonium ion (C–Br+–C) and Br– ion. Different solvents may stabilize these charged specifies differently affecting the reaction barriers and energies. Note that no computational studies have been reported for the potential energy surface for the bromination of cytosine, 1-methyl cytosine, and cytidine. The chemical reactivity of these uracil derivatives in various solvents could provide a greater understanding of these systems, which would be potentially helpful for the development of pharmaceutical and (bio)chemical products.

Methods

Gaussian16 software package[37] was used for all of the electronic structure calculations. Geometries of all of the structures, reactants, transition states, and products were fully optimized at the B3LYP and the second-order Møller–Plesset (MP2) levels of theory using the 6-31+G(d,p) basis set. Since B3LYP and MP2 correspond to the density functional and wave function theories, respectively, both were used to cross-validate our work. For calculations involving reactions, G3MP2 calculations were also carried out due to its reliability.[38,39] All solvent calculations involving water, DMSO, n-octanol, and chloroform were accomplished with the Solvation Model on Density (SMD), where the model was well parametrized to provide reliable results.[40,41] The SMD model is built upon the original polarizable continuum model, where the united atom topological model is applied to each atomic radius. Frequencies were calculated for all optimized structures to make sure no imaginary frequency for the ground states and only one imaginary frequency for the transition states. From the optimized structures of cytosine, 1-methyl cytosine, and cytidine, solvation free energy, charges, polarizability, first-order hyperpolarizability, dipole moment, and HOMO–LUMO energies have been calculated. The complete pathway of the reaction mechanism of cytosine, 1-methyl cytosine, and cytidine reacting with Br2 has been confirmed with the intrinsic reaction coordinate (IRC) that connects transition state to both the reactants and products.

The polarizability (α) of the molecules were calculated with the equationwhere α, α, and α are known as principal values of polarizability tensor.

The first-order hyperpolarizability measures how easily a dipole is induced by an electric field in a molecule. The hyperpolarizability (βtot) was calculated using the following equationwhereGaussian16[37] output file provides all of these matrix components, β, β, β, β, β, β, β, β, β, in its output file.

Results and Discussion

Conformational Analysis

All of the level of theory and basis set provided one optimized structure for cytosine and 1-methyl cytosine; however, conformational analysis of cytidine yielded five conformers, namely cytidine I, cytidine II, cytidine III, cytidine IV, and cytidine V both in the gas phase and solvents (Figure ). The relative internal energy, enthalpy, and Gibbs energy of five conformations of cytidine are listed in Table .

Figure 2

Optimized structure of different conformers of cytidine (I–V) obtained from B3LYP/6-31+G(d,p).

Table 1

Relative Internal Energy, ΔE (kJ mol–1), Enthalpy, ΔH (kJ mol–1), and Gibbs Energy, ΔG (kJ mol–1), of Five Different Conformations of Cytidine in the Gas Phase and in Different Solvents at MP2 and B3LYP Using 6-31+G(d,p)a,b

	MP2/6-31+G(d,p)			B3LYP/6-31+G(d,p)
	ΔE	ΔH	ΔG	ΔE	ΔH	ΔG
gas phase
cytidine I	0.4	0.4	–0.3	0.3	0.6	0.0
cytidine II	–4.6	–3.9	–6.0	–7.0	–6.2	–7.5
cytidine III	14.2	14.1	11.3	11.1	11.0	8.5
cytidine V	0.4	0.4	–0.3	0.3	0.7	0.0
water
cytidine I	–	–	–	6.9	7.0	6.5
cytidine II	14.6	14.6	13.0	11.0	11.1	11.4
cytidine III	10.5	10.5	12.1	7.4	7.4	6.7
cytidine V	17.2	17.2	16.0	11.3	11.6	9.7
DMSO
cytidine I	1.6	2.1	1.6	1.8	2.3	1.5
cytidine II	7.1	7.9	7.2	4.9	5.6	2.2
cytidine III	17.7	18.2	16.9	14.8	14.9	12.0
cytidine V	1.6	2.1	1.5	3.1	4.2	1.3
n-octanol
cytidine I	7.5	7.8	7.4	0.8	8.0	7.6
cytidine II	10.9	11.0	8.6	0.1	8.6	7.6
cytidine III	12.3	12.6	14.2	1.6	10.6	12.1
cytidine V	7.5	7.8	7.4	2.8	12.3	11.2
chloroform
cytidine I	5.1	5.2	5.1	5.3	5.0	2.3
cytidine II	5.9	6.7	6.5	3.9	4.2	1.4
cytidine III	16.1	16.3	12.8	13.5	13.5	12.1
cytidine V	14.2	14.3	11.0	8.0	8.1	5.4

Relative stability is calculated based on the most stable cytidine, cytidine IV, found in all solvents, i.e., ΔX = X(molecule) – X(cytidine IV), where X is E, H, or G.

– represents no optimized geometry is obtained.

Cytidine II, the syn conformer, is found to be the most stable conformer in the gas phase; however, cytidine IV, the anticonformer, is the most stable conformer in the solvent system. The syn conformer is more stable than anticonformer in the gas phase by a Gibbs energy difference (ΔG) of 6.0 and 7.5 kJ mol–1 at MP2/6-31+G+G(d,p) and B3LYP/6-31+G(d,p), respectively. Both MP2 and B3LYP provided very similar free energies. Anticonformer is more stable than the syn conformer in water, DMSO, n-octanol, and chloroform by ΔG of 13.0, 7.2, 8.6, and 6.5 kJ mol–1, respectively, at MP2/6-31++G(d,p) and by ΔG of 11.4, 2.2, 7.1, and 1.4 kJ mol–1, respectively, at B3LYP/6-31+G(d,p).

Some of the bond lengths, angles, and dihedral angles of cytosine, 1-methyl cytosine, and cytidine calculated at B3LYP/6-31+G(d,p) in the gas phase and in different solvents are found to change significantly (see the Supporting Information Tables S1–S4). The larger differences due to substituents have been observed for C3–N8, N8–H10 (C13 and C15 for 1-methyl cytosine and cytidine, respectively), and C1–C2 bonds by at least 0.01 Angstrom. Bond angles C1–C2–N8, N8–C3–N9, and C2–N8–C3 and dihedral angles H6–C2–N8–H10 (C13 and C15 for 1-methyl cytosine and cytidine, respectively), O7–C3–N8–C10, and C1–C4–N10–H12 (N11 and H12 for cytosine) also show substantial changes with different substituents. Solvent polarity also affects the geometrical parameter for all three compounds (Tables S2–S4). Polar protic solvents, water and n-octanol, affect the geometry more than the nonpolar solvent. The change in bond lengths, bond angles, and dihedral angles due to variation in solvent polarity range from 0.005 to 0.029 Å, 0.631 to 3.263°, and 2.2 to 16.8°, respectively, in cytosine, 0.009 to 0.028 Å, 0.682 to 3.594°, and 5.1 to 18.6°, respectively, in 1-methyl cytosine, and 0.003 to 0.025 Å, 0.552 to 3.420°, and 1.3 to 17.4°, respectively, in cytidine.

Solvation Free Energy

The effect of solvent on the chemical system, solute, is very accurately quantified with the solvation free energy. Since cytidine IV is most stable in a solvent, this structure is used to calculate molecular properties in all solvents and it will be denoted as cytidine from now. Table represents the solvation free energy of cytosine and its derivatives calculated with the SMD model.[40,41]

Table 2

Solvation Free Energy (kJ mol–1) in Different Solvents with SMD at B3LYP/6-31+G(d,p)

medium (dielectric constant)	cytosine	1-methyl cytosine	cytidine
chloroform (4.7)	–60.52	–59.50	–85.65
n-octanol (9.9)	–70.50	–67.10	–112.79
DMSO (46.8)	–74.73	–71.36	–101.54
water (78.3)	–82.75	–74.21	–130.08

The solvation free energies of cytosine negatively increase with the polarity of the solvents, i.e., with the dielectric constants of the solvents. The solvation free energies are found −60.52, −70.50, −74.73, and −82.75 kJ mol–1 in chloroform, n-octanol, DMSO, and water, respectively. A similar trend is observed for 1-methyl cytosine (Table ). However, the solvation free energies of cytidine negatively increase more in the polar protic solvents, water and n-octanol, than the polar aprotic solvent DMSO. The free energy of solvation for cytidine in chloroform, n-octanol, DMSO, and water are −85.65, −112.79, −101.54, and −130.08 kJ mol–1, respectively. Unlike cytosine and 1-methyl cytosine, cytidine contains a deoxyribose ring with three −OH groups attached. These −OH groups can form a H-bond with polar protic solvents such as the water and n-octanol. Therefore, it is apparent that a systematic trend, i.e., increasing solvation free energy with increasing polarity of the solvent is found in the case of cytosine and 1-methyl cytosine, while hydrogen bond plays a significant role for cytidine. Paglieri et al.[21] also showed that solvation energies of cytosine tautomers affected more in the aqueous phase. Interestingly, the dipole moments of all three compounds, cytosine, 1-methyl cytosine, cytidine, as shown in Table , decrease in the order of water > n-octanol > DMSO > chloroform, i.e., dipole moment is higher in a polar protic solvent, decrease in a polar aprotic solvent, decrease further in a nonpolar solvent, and it is the lowest in the gas phase. This is in accordance with the study conducted by Dreyfus et al.[22] who also observed that cytosine tautomer in a polar solvent to have the larger dipole moment.

Table 3

Dipole Moment (Debye, D) in the Gas Phase and in Different Solvents Using SMD at B3LYP/6-31++G(d,p)

medium (dielectric constant)	cytosine	1-methyl cytosine	cytidine
gas	6.3	5.9	8.7
chloroform (4.7)	8.1	7.7	10.7
n-octanol (9.9)	9.0	8.7	11.8
DMSO (46.8)	8.5	8.3	11.5
water (78.3)	9.8	9.5	12.7

Possible Reaction Sites in Cytosine Derivatives

Possible reaction sites in the cytosine derivatives are determined from the molecular electrostatic potential (MESP). MESP is the net electrostatic effect on a molecule as it arises from the total static charge distribution. In general, MESP should agree well with the partial charges, dipole moments, electronegativity, and chemical reactivity of the molecule.[42−44] MESP contour map of a cytosine derivative identifies regions in the molecule that are susceptible to electrophilic and nucleophilic attacks. Figure represents the MESP contour maps of cytosine and its derivatives calculated at B3LYP/6-31+G(d,p).

Figure 3

Molecular electrostatic potential (MESP) contour of (a) cytosine, (b) 1-methyl cytosine, and (c) cytidine. The red and blue color regions in the MESP surface represent the regions susceptible to the attack of electrophiles and nucleophiles, respectively.

The MESP contour map reveals that the most possible sites for an electrophilic attack are O7 and N9 (also O22 and O14 in cytidine) and nucleophilic attack are C1, C2, and N8. Although MESP is generated by the electric field of internal charge distribution,[45] the MESP and the atomic charges obtained from Mulliken population analysis (MPA) at B3LYP/6-31+G(d,p), as shown in Table S5, are found consistent with each other. The oxygen atom (O7) in cytosine, 1-methyl cytosine, and cytidine has an atomic charge of −0.516, −0.525, and −0.535, respectively, suggesting O7 could be susceptible for an electrophilic attack in these molecules. Similarly, N9 is also negatively charged for all three compounds. The presence of N8 and N9 imposes large positive charges on C3 and C4 atoms, with charges of 0.682, 0.698, and 0.712 for C3 and 0.504, 0.500, 0.501 for C4 in cytosine, 1-methyl cytosine, and cytidine, respectively. O14, O22, and O29 attached to the deoxyribose ring also contain large negative charges, −0.541, −0.548, −0.557, respectively, in cytidine IV, suggesting more possible sites for an electrophilic attack in cytidine.

Then, polarizability and first-order hyperpolarizability of cytosine, 1-methyl cytosine, and cytidine have been studied to understand the distortion of these molecules in an electric field. The calculated polarizability and hyperpolarizability of cytosine and its derivatives in different solvents are presented in Table .

Table 4

Effect of Solvent Polarity on Polarizability (a.u.) and First-Order Hyperpolarizability (a.u.) at B3LYP/6-31+G(d,p)

	gas phase	chloroform (4.7)	n-octanol (9.9)	DMSO (46.8)	water (78.3)
cytosine
αtot	61.37	76.02	80.81	81.78	84.84
βtot	47.42	64.47	73.32	68.53	80.23
1-methyl cytosine
αtot	73.91	92.10	97.71	99.56	102.65
βtot	49.74	68.83	77.27	74.25	83.76
cytidine
αtot	128.31	156.80	167.29	167.16	176.13
βtot	160.76	192.94	212.37	209.04	229.69

Polarizability gradually increases from lower to higher dielectric constants of the solvents, i.e., with the solvent polarity. The hyperpolarizability of cytosine, 1-methyl cytosine, and cytidine IV in different solvents decrease in the order of water > n-octanol > DMSO > chloroform. Both polarizability and first-order hyperpolarizability in all solvents decrease in the order of cytidine > 1-methyl cytosine > cytosine.

Frontier Molecular Orbitals

Polarity of the solvent as well as different solute–solvent interactions can influence different molecular properties and the reactivity due to their variable interactions with the frontier molecular orbitals.[46,47] The frontier molecular orbital diagrams of the highest occupied molecular orbital (HOMO) and the lowest occupied molecular orbital (LUMO) structures are shown in Figure .

Figure 4

Frontier molecular orbitals of (a) cytosine, (b) 1-methyl cytosine, and (c) cytidine at B3LYP/6-31+G(d,p).

The HOMO is confined over O7, C1, and C2 atoms suggesting that these atoms will act as nucleophilic or electron donating, while the LUMO orbitals are confined over N10 (N11 for cytosine), C3 and C4 atoms suggesting that the atoms are electrophilic or electron accepting. This is in agreement with the MESP contour map. The values of HOMO–LUMO energies of cytosine and its derivatives in various solvents at B3LYP/6-31+G(d,p) are presented in Table .

Table 5

Molecular Orbital Energy (eV) in Different Solvents with SMD at B3LYP/6-31+G(d,p)

	cytosine			1-methyl cytosine			cytidine
medium (dielectric constant)	HOMO	LUMO	ΔE	HOMO	LUMO	ΔE	HOMO	LUMO	ΔE
gas phase	–6.17	–0.83	5.34	–6.04	–0.72	5.32	–5.85	–0.49	5.36
chloroform (4.7)	–6.12	–0.62	5.50	–5.99	–0.54	5.45	–5.98	–0.53	5.45
n-octanol (9.9)	–6.24	–0.66	5.58	–6.10	–0.59	5.51	–6.13	–0.61	5.52
DMSO (46.8)	–6.05	–0.53	5.52	–5.93	–0.47	5.46	–5.98	–0.52	5.46
water (78.3)	–6.30	–0.67	5.63	–6.17	–0.61	5.55	–6.22	–0.67	5.55

HOMO energies are found highest in DMSO with values of −6.05, −5.93, and −5.98 eV for cytosine, 1-methyl cytosine, and cytidine, respectively, suggesting that the molecules are best electron donor in DMSO. LUMO energies are lowest in water, with the energies of −0.67, −0.61, −0.67 eV, respectively, for cytosine, 1-methyl cytosine, and cytidine, which implies that the molecules are best electron acceptors in solvent water. The difference between the HOMO and LUMO energy characterizes the chemical stability of a molecule. In the case of conjugated pi orbital systems, such as cytosine, 1-methyl cytosine, or cytidine, the smaller HOMO–LUMO energy gap corresponds to greater mobility of the π electrons throughout the molecule, which stabilizes it. The trend in the HOMO–LUMO energy gap in different solvents is graphically shown in Figure . The HOMO–LUMO energy gap of cytosine and its derivatives in different solvents decreases in the order polar protic solvent > polar aprotic solvent > nonpolar solvent. This suggests that all three cytosine derivatives may become more stable in nonpolar solvents than the polar aprotic and polar protic solvents.

Figure 5

Effect of solvent polarity on the HOMO–LUMO energy gap.

Reaction of Cytosine Derivatives with Br2

To better understand the reactivity of cytosine, 1-methyl cytosine, and cytidine, reactions of these uracil derivatives with Br2 have been studied with B3LYP/6-31+G(d,p) and G3MP2. We have extensive experience with the kinetics and thermodynamic studies of the bromination reactions of organic molecules such as ethene, propene, isobutene, flouroethene, chloroethene, (E)-1,2-difluoroethene, (E)-1,2-dichloroethene, and adamantylideneadamantane.[48−51] Note that no computational studies have been reported for the potential energy surface for the bromination of cytosine, 1-methyl cytosine, and cytidine. Taguchi et al.[52] have shown that the reaction of cytosine and 1-methyl cytosine with Br2 in aqueous and methanolic solutions produces multiple intermediates via several intermediate steps, ultimately producing the final product of 5-bromocytosine and 5-bromo-1-methyl cytosine, respectively. In this study, we investigated a pathway that leads to the formation of trans-dibromo adduct via a one-step pathway, which would provide important insights as to how the uracil derivatives in different solvents may behave during a reaction. The results for the reaction of cytosine, 1-methyl cytosine, and cytidine with Br2 in the gas phase and in chloroform, n-octanol, DMSO, and water are given in Tables and 7. The structures and relative energies of reactant, transition state, and product are shown in Figure . The transition state (TS) structure shows that one Br atom of Br2 is attacking the C1 = C2 bond of the uracil derivate forming an ion pair of the bromonium ion (−C1–Br+–C2−) and the Br– ion. This agrees with the predictions from MESP and frontier molecular orbitals, which predicted C1 and C2 atoms likely to be nucleophilic or electron donating. Thus, the positively charged Br+ attacks these carbon atoms during the reaction. In the reactant complex R, the Br–Br bond distances for Br2 attacking the C1=C2 bond in cytosine, 1-methyl cytosine, and cytidine are 2.39, 2.37, and 2.32 Å, respectively, at the B3LYP/6-31+G(d,p) level of theory in the gas phase, while the distances increase to 3.07, 3.06, 2.98 Å, respectively, in TS. The Br atom that is attacking the C1=C2 bond is closer to one of the C atoms, C1, with the C1–Br bond distance of 2.42, 2.45, and 2.5 Å in cytosine, 1-methyl cytosine, and cytidine, respectively, while the C2–Br bond is 2.95, 2.99, and 3.12 Å, respectively. The Gibbs energy of activation for cytosine, 1-methyl cytosine, and cytidine are 190.4, 168.4, and 124.2 91.6 kJ mol–1, respectively, at B3LYP/6-31G(d,p), which are lower than the values obtained from G3MP2 by 17.9 and 30.7 kJ mol–1 in cytosine and 1-methyl cytosine, respectively. This indicates that the Gibbs energies of activation values are slightly underestimated with B3LYP/6-31+G(d,p). A similar mechanism is also obtained in all of the solvents. The solvent model used in this study shows that the Gibbs energy of activation for these reactions decreases in the order chloroform > n-octanol > DMSO > water for cytosine and 1-methyl cytosine (Table ), which suggests more reactivity of cytosine and 1-methyl cytosine in polar solvents. This is in accordance with the HOMO–LUMO energy gap values, as shown in Table , which predicted that the cytosine derivatives may become more stable in nonpolar solvents than the polar aprotic and polar protic solvents, which are supported by less reactivity and higher reaction barrier in nonpolar solvents. However, Gibbs energy of activation for the reaction with cytidine decrease in the order of water > DMSO > n-octanol > chloroform. Polar solvents may provide more stability to the ribose ring in cytidine as a consequence it becomes harder to get the molecule to react with Br2. The change in reaction barrier between the gas phase and the most polar solvent, water, is about 16.3 kJ mol–1 for cytosine, 9.5 kJ mol–1 for 1-methyl cytosine, and 13.3 kJ mol–1 for cytidine. These differences are reasonable as the uracil derivates are reacting with a nonpolar molecule, Br2. Since the Br atom is not strongly electronegative, the transition state structure along the reaction pathway may not be affected by solvent too much. The Gibbs energy of activation for all reactions decreases in the order cytosine > 1-methyl cytosine > cytidine in both the gas phase and solvents. This suggests that cytosine is less reactive than 1-methyl cytosine, which is again less reactive than cytidine. This is in accordance with molecular properties such as the polarizability and first-order hyperpolarizability values, which decrease in the order of cytidine > 1-methyl cytosine > cytosine (Table ). Due to the higher polarizability, cytidine will be more reactive compared to 1-methyl cytosine, which is again more reactive than cytosine. The HOMO–LUMO energy gap values (Table ) for all three systems are very similar, so HOMO–LUMO energies may not be able to explain reactivity for these systems.

Table 6

Free Energies of Activation (kJ mol–1) at 298.15 K for the Reaction of Cytosine, 1-Methyl Cytosine, and Cytidine IV with Br2 in Gas Phase, Chloroform, DMSO, n-Octanol, and Water at B3LYP/6-31+G(d,p)a,b,c

phase (ε)	cytosine	1-methyl cytosine	cytidine
gas	190.4	168.4	124.2
	(208.3)e	(199.1)e
chloroform (4.7)d	177.9	161.0	110.9
n-octanol (9.9)d	175.8	160.4	115.0
DMSO (46.8)d	174.2	159.1	118.6
water (78.3)d	174.1	158.9	119.1

Barriers as defined in Figure .

The products are all in trans conformation.

The Solvation Model on Density (SMD) was used for optimized structures. In all cases, ΔG = ΔΔG (thermal correction) + ΔGsolv.

Represent dielectric constant.

Obtained from G3MP2; G3MP2 calculations were not possible to perform for cytidine.

Table 7

Free Energies of Reaction (kJ mol–1) at 298.15 K for the Reaction of Cytosine, 1-Methyl Cytosine, and Cytidine with Br2 in Gas Phase, Chloroform, DMSO, n-Octanol, and Water at B3LYP/6-31G+(d,p)a,b

phase (ε)	cytosine	1-methyl cytosine	cytidine
gas	1.4	–0.5	–8.5
	(−17.0)d	(−20.1)d
chloroform (4.7)c	12.6	13.6	9.3
n-octanol (9.9)c	13.0	15.1	10.6
DMSO (46.8)c	9.0	13.3	7.1
water (78.3)c	8.2	7.8	8.6

The products are all in trans conformation.

The Solvation Model on Density (SMD) was used for optimized structures. In all cases, ΔG = ΔΔG (thermal correction) + ΔGsolv.

Represent dielectric constant.

Obtained from G3MP2.

Figure 6

Reaction pathway and mechanism for the reaction of cytosine, 1-methyl cytosine, and cytidine with Br2 at the B3LYP/6-31+G(d,p) level of theory.

The thermodynamic properties for the bromination reactions investigated are listed in Table . All of the reactions are exergonic in the gas phase both at B3LYP/6-31+G(d,p) and G3MP2, except for the reaction with cytosine where free energy of reaction (ΔG) is found 1.4 kJ mol–1 at B3LYP/6-31+G(d,p). On the contrary, ΔGs are found endergonic in all of the solvents (Table ). The ΔGs decrease in the order n-octanol > chloroform > DMSO > water for cytosine and 1-methyl cytosine, while the ΔGs decrease in the order n-octanol > chloroform > water > DMSO for the reaction with cytidine (Table ). No experimental or theoretical enthalpy of formation (ΔHf) has been reported for trans-dibromocytosine. From the enthalpies of reaction (ΔH) of the reactions of cytosine with Br2 calculated at G3MP2, it is possible to calculate ΔHf for trans-dibromocytocine as experimental ΔHf for cytosine (−59.00 kJ mol–1)[53] and Br2 (−30.01 kJ mol–1)[54] are known. Enthalpies of reaction are exothermic in the gas phase with ΔH values of −43.4 (−62.7 kJ mol–1 at G3MP2), −45.3 (−65.5 kJ mol–1 at G3MP2), and −55.4 kJ mol–1 for cytosine, 1-methyl cytosine, and cytidine, respectively, at B3LYP/6-31+G(d,p). We report that the ΔHf for trans-dibromocytosine is −90.82 kJ mol–1.

Conclusions

In this study, ab initio calculations such as MP2 and B3LYP level of theory and 6-31+G(d,p) basis set have been used to calculate geometry, solvation free energy, dipole moment, molecular electrostatic potential (MESP), Mulliken charge distribution, polarizability, hyperpolarizability, and HOMO–LUMO energy gap values for cytosine, 1-methyl cytosine, and cytidine to understand how these properties change with different solvents. Solvation free energy and polarizability values decrease in the order water > DMSO > n-octanol > chloroform, while dipole moment, first-order hyperpolarizability, and HOMO–LUMO energy gap values in different solvents follow the order of polar protic solvent (water and n-octanol) > polar aprotic solvent (DMSO) > nonpolar solvent (chloroform), suggesting that these molecular properties are affected by both solvent polarity and hydrogen-bonding interactions. In addition to polarity, hydrogen-bonding interactions may play a major role in the stability of the larger uracil derivate, cytidine, which contains deoxyribose ring with three −OH groups that can form H-bond with polar protic solvents such as the water and n-octanol. It is harder for a molecule to undergo a chemical reaction, if it is stable in a particular solvent. To check if this is really the case, we studied bromination reaction of cytosine, 1-methyl cytosine, and cytidine in different solvents. The free energies of activation decrease with the polarity of the solvent that is chloroform > n-octanol > DMSO > water for cytosine and 1-methyl cytosine. Thus, we conclude that solvent polarity is important for the reactivity of these uracil derivatives. On the other hand, Gibbs energy of activation for the reaction of cytidine with Br2 decreases in the order of water > DMSO > n-octanol > chloroform. Thus, for a larger and complex system such as the cytidine, it may be harder to predict if the molecule is stable or reactive in a particular solvent based on only molecular properties such as solvation free energy, dipole moment, polarizability, first-order hyperpolarizability, and HOMO–LUMO energy gap values as they either show an opposite trend or remains almost the same in different solvents. The Gibbs energy of activation for all reactions decreases in the order cytosine > 1-methyl cytosine > cytidine in both the gas phase and solvents. The calculated results also reveal that solvation plays an important role in the stabilization of different conformers of cytidine. The syn conformer (cytidine II) is the most stable conformer in the gas phase, while the anticonformer (cytidine IV) is most stable in all solvents. Solvent polarity also greatly affects the geometry of uracil derivatives. This study also reports that the heat of formation for trans-dibromocytosine is −90.82 kJ mol–1.